Antiarrhythmic Mechanisms of SK Channel Inhibition in the Rat Atrium.

ORIGINAL ARTICLE

Antiarrhythmic Mechanisms of SK Channel Inhibition in the Rat Atrium Lasse Skibsbye, PhD,* Xiaodong Wang, MD,† Lene Nygaard Axelsen, PhD,* Sofia Hammami Bomholtz, PhD,‡ Morten Schak Nielsen, PhD,* Morten Grunnet, DrSc,‡ Bo Hjorth Bentzen, PhD,*‡ and Thomas Jespersen, DMSc*

Introduction: SK channels have functional importance in the cardiac atrium of many species, including humans. Pharmacological blockage of SK channels has been reported to be antiarrhythmic in animal models of atrial fibrillation; however, the exact antiarrhythmic mechanism of SK channel inhibition remains unclear.

Objectives: We speculated that together with a direct inhibition of repolarizing SK current, the previously observed depolarization of the atrial resting membrane potential (RMP) after SK channel inhibition reduces sodium channel availability, thereby prolonging the effective refractory period and slowing the conduction velocity (CV). We therefore aimed at elucidating these properties of SK channel inhibition and the underlying antiarrhythmic mechanisms using microelectrode action potential (AP) recordings and CV measurements in isolated rat atrium. Automated patch clamping and two-electrode voltage clamp were used to access INa and IK,ACh, respectively. Results: The SK channel inhibitor N-(pyridin-2-yl)-4-(pyridin-2-yl) thiazol-2-amine (ICA) exhibited antiarrhythmic effects. ICA prevented electrically induced runs of atrial fibrillation in the isolated right atrium and induced atrial postrepolarization refractoriness and depolarized RMP. Moreover, ICA (1–10 mM) was found to slow CV; however, because of a marked prolongation of effective refractory period, the calculated wavelength was increased. Furthermore, at increased pacing frequencies, SK channel inhibition by ICA (10–30 mM) demonstrated prominent depression of other sodium channel–dependent parameters. ICA did not inhibit IK,ACh, but at concentrations above 10 mM, ICA use dependently inhibited INa. Conclusions: SK channel inhibition modulates multiple parameters of AP. It prolongs the AP duration and shifts the RMP towards more depolarized potentials through direct ISK block. This indirectly leads to sodium channel inhibition through accumulation of state dependently inactivated channels, which ultimately slows conduction Received for publication November 8, 2014; accepted March 20, 2015. From the *Danish National Research Foundation Centre for Cardiac Arrhythmia and Department of Biomedical Sciences, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark; †Shanghai East Hospital, Tongji University School of Medicine, Shanghai, China; and ‡Acesion Pharma, Copenhagen, Denmark. The authors report no conflicts of interest. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.jcvp.org). Reprints: Thomas Jespersen, DMSc, Department of Biomedical Sciences, Faculty of Health and Medical Sciences, University of Copenhagen, Blegdamsvej 3, DK-2200 Copenhagen, Denmark (e-mail: [email protected]). Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

J Cardiovasc Pharmacol Volume 66, Number 2, August 2015

and decreases excitability. However, a contribution from a direct sodium channel inhibition cannot be ruled. We here propose that the primary antiarrhythmic mechanism of SK channel inhibition is through direct potassium channel block and through indirect sodium channel inhibition. Key Words: atrial fibrillation, Small-conductance Ca2+-activated K+ channels, SK channels, Antiarrhythmic mechanisms, Electrophysiology, atrial specific, ICA, frequency dependence (J Cardiovasc Pharmacol 2015;66:165–176)

INTRODUCTION Small-conductance Ca2+-activated K+ (SK) channels are expressed in mammals in various tissues, including nervous system, vasculature, skeletal muscle, smooth muscle, and cardiac tissue.1–4 The SK channel proteins SK1, SK2, and SK3 are encoded by KCNN1, KCNN2, and KCNN3 genes, respectively.5 SK channels are characterized by their small single-channel conductance, activation by submicromolar [Ca2+]i, voltage insensitivity, and selective blockade by the bee venom toxin apamin.6–8 In recent years, an increasing amount of evidence has described the functional expression of SK channels in cardiac tissue.9 SK channels have prominent impact on atrial electrophysiology, and some reports also describe functional ventricular importance under pathophysiological conditions, such as heart failure.10–14 Moreover, genomewide association studies have provided evidence for common variants in the KCNN3 gene being associated with risk of atrial fibrillation (AF).15–17 We and others have previously shown SK channel inhibition to be antiarrhythmic in various ex vivo and in vivo animal models of experimental AF.18–21 Furthermore, we recently described the functional importance of SK current in human atria but not in the ventricles.22 The aim of this study was to address the mechanisms underlying the antiarrhythmic effects of SK channel inhibition, with particular focus on the impact of changes in atrial membrane potential during diastole as a consequence of SK channel inhibition. Experiments were performed by monitoring action potential (AP) parameters in the isolated rat right atrium subjected to refined pacing protocols, as well as by measuring conduction velocity (CV) and contractility in atrial muscle strips. The small molecule N-(pyridin-2-yl)-4-(pyridin-2-yl) thiazol-2-amine (ICA)23 was used in this study, as this compound previously has been found to be a rather selective inhibitor of the ISK current. www.jcvp.org |

165

Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.


Skibsbye et al

MATERIALS AND METHODS Animal Preparation Male Sprague–Dawley rats (350–450 g) were anesthetized subcutaneously by Hypnorm/midazolam (5 mg/mL, containing sterile water 4 mL, hypnorm 2 mL, midazolam 2 mL) 0.2 mL/kg. After anesthetization, the trachea was cannulated to ventilate the rat during the open-chest procedure. The thorax was opened, the heart was explanted, and the right atrium was dissected. The study protocol was approved by the Danish Research Animal Committee, and all procedures were performed in accordance with the Danish legislation of animal use for scientific procedures as described in the “Animal Testing Act.”

Ex Vivo Electrophysiology: Microelectrode Measurement The right atrium was placed in a Steiert papillary muscle bath with microelectrode amplifier (Hugo Sachs Harvard Apparatus GmbH) and superfused with modified Tyrode’s solution at 378C. The superfusion buffer consisted of (in mM) NaCl 118.2, KCl 4.0, CaCl2$2H2O 1.8, MaCl2$6H2O 1.0, NaH2PO3$6H2O 1.8, NaHCO3 25, and C6H12O6$H2O (D-glucose monohydrate) 11, pH 7.2. The buffer was aerated with 95% O2 and 5% CO2. Pacing was induced through a bipolar coaxial electrode using pulses of 2-millisecond width. Pacing voltage was set to 2 · threshold. Threshold was adjusted every time before effective refractory period (ERP) measurement. The ERP was determined by applying an extra stimulus (S2) by increments of 1 millisecond after every 10th regular stimulus (S1) until initiation of an extra AP. Intracellular recordings were achieved by impaling the muscle with a sharp filamented glass microelectrode filled with 3 M KCl. The electrode had a tip resistance of 20–30 MV and was connected to the headstage amplifier through an Ag–AgCl electrode holder. Signals were amplified with a bipotential amplifier (Hugo Sachs Harvard Apparatus GmbH). The rat right atrium beats at an intrinsic rate between 3.5 and 4.5 Hz; therefore, all experiments were performed at a 5 Hz baseline pacing frequency. After microelectrode impalement and satisfactory AP recordings, the tissue was continuously paced at 5 Hz for 40 minutes to obtain stable AP recordings. After the stabilization period, ERP was measured and the pacing protocol (5–9, 9–11, 11–9, and 9–5 Hz) as depicted in Supplemental Digital Contents 2 and 3 (see Figs. 2 and 3, http://links.lww.com/JCVP/A187 and http://links.lww.com/ JCVP/A188) was initiated. The compound ICA was applied to the perfusion buffer in 3 different concentrations in a cumulative fashion acquiring (10, 20, and 30 mM). ERP measurements and pacing protocols were initiated every 30 minutes after application of ICA. Action potential duration at 90% repolarization (APD90), ERP, and resting membrane potential (RMP) were continuously monitored during the whole experiment and measured at the end of each perfusion period. Timematched controls (TMC) (n = 6) were conducted applying the solvent dimethyl sulfoxide (DMSO) to the perfusion buffer to ensure stability during the experimental protocol. None of the measured parameters were significantly changed during the time course of the TMC experiments, as exemplified in the

166

| www.jcvp.org

TMC recordings shown in Supplemental Digital Content 3 (see Fig. 3, http://links.lww.com/JCVP/A188). Data acquisition was performed continually throughout the experimental procedures using a 16-channel PowerLab system (ADInstruments) with a sample rate of 100 kHz and analyzed using Chart 7.3cht73 Pro software (ADInstruments).

Conduction Velocity A tissue strip from the right atrium was dissected and used for CV measurements as previously described.24 In short, the tissue strip was superfused with oxygenated Tyrode’s buffer at a rate of 2 mL/min, and the temperature was maintained at 378C. The tissue was paced at 5 Hz with an impulse duration of 0.5 milliseconds, and voltage was set to 2 · threshold. CV was measured with 2 microelectrodes (Platinum/ Iridium [PI20030.5B10], Micro Probe, Inc, Gaithersburg) placed on the longitudinal axis of the tissue strip. The distance between the 2 microelectrodes was measured with the calibrated ocular grid in the microscope. The distance between the 2 microelectrodes was 1.2–2.2 mm, and the distance between the stimulation electrode and the first microelectrode was 1.7–3.4 mm. The extracellular field potentials and force were band-pass filtered at 300–10,000 Hz and sampled at 10 kHz (Digidata 1322A; Axon Instruments, Union City). Time of local activation under the first and second microelectrodes was determined as the time of minimum dU/dt by the custom-written MATLAB routine. CV was calculated as the interelectrode distance divided by the interelectrode delay. Simultaneously, contractility was monitored using a force transducer connected to one end of the tissue strip with the other end fixed to a hook using fine suture. Contractile force was measured both as diastolic and systolic (developed) forces. Tissue strips were allowed to stabilize for 20 minutes in all experiments. Subsequently, CV was measured for 40 minutes before application of increasing concentrations of either ICA (1, 3, 10 mM) or vehicle (DMSO) in 30-minute intervals.

Cell Cultures Chinese hamster ovary (CHO) cells stably expressing rNaV1.5 were grown at 378C in 5% CO2 in Dulbecco’s modified Eagle’s medium, supplemented with 10% fetal bovine serum (Th. Geyer, Denmark) and GlutaMAX (Substrate Department, Panum Institute, Copenhagen, Denmark). On the day of experiments, the cells were rinsed with phosphate-buffered saline and detached from T175 bottles with 5 mL Detachin (Th. Geyer, Denmark). Finally, the cells were resuspended in serum-free medium supplemented with 100 U/mL penicillin/ streptomycin, 0.04 mg/mL soybean trypsin inhibitor, and 25 mM 4-(2-Hydroxyethyl)piperazine-1-ethanesulfonic acid, N-(2-Hydroxyethyl)piperazine-N0 -(2-ethanesulfonic acid) (HEPES) (Sigma-Aldrich, Denmark) at a density of 5 · 106 cells per milliliter. Cells were transferred to the QPatch stirring station and allowed to recover for 15 minutes before initiating the experiment.

QPatch Recordings The whole-cell recordings were performed using an automated patch-clamp system (QPatch 16 HT; Biolin Scientific, Sophion, Denmark) with disposable single-hole Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.



QPlates (Biolin Scientific, Sophion, Denmark) and CHO cells stably expressing rNaV1.5. In brief, this system automatically allows for giga sealing, whole-cell formation, liquid application, access resistance compensation and recording. NaV1.5 currents were recorded with an extracellular solution containing the following (in mM): 2 CaCl2, 1 MgCl2, 10 HEPES, 4 KCl, 145 NaCl, 10 glucose, pH 7.2, 310 mOsm (adjusted with sucrose). The intracellular solution contained (in mM) 135 CsF, 1/5 EGTA/CsOH, 10 HEPES, 10 NaCl, 4 Na-ATP, pH 7.3, 300 mOsm. A dual-frequency protocol was used to address possible use- and state-dependent block of NaV1.5. Initially, NaV1.5 currents were elicited every 1000 milliseconds (1 Hz) by depolarizing the membrane potential to 220 mV for 50 milliseconds from a holding potential of 2120 mV (a total of 80 pulses). This was followed by a step to 2120 mV lasting for 5000 milliseconds in order to release all channels from inactivation. Hereafter, the cells were stimulated with 130 pulses at 5 to 220 mV for 50 milliseconds from a holding potential of 275 mV. The full protocol lasted 120 seconds and was repeated during each application. The application protocol used was baseline recordings in standard extracellular solution followed by application of drug (ICA 10 or 30 mM), or extracellular solution for TMC experiments, and finally 10 mM flecainide. Data were sampled at 25 kHz, eighth-order Bessel filter, cutoff frequency 3 kHz, and 80% Rs compensation. Please note that during the 1 Hz stimulation, data were only acquired for every second pulse to reduce sample size. Only experiments with a whole-cell seal of .500 MV were used.

Oocyte Experiments Two-electrode voltage-clamp recordings were performed in Xenopus laevis oocytes at room temperature under superfusion with Kulori solution consisting of (in mM) 4 KCl, 90 NaCl, 1 MgCl2, 1 CaCl2, 5 HEPES, pH 7.4. Xenopus laevis surgery and oocyte treatment were done as previously described.25 Fifty nanoliters of complementary RNA [Kir3.1 + Kir3.4; 0.5 + 0.5 ng (0.04 + 0.04 mg/mL)] was injected using a Nanoject microinjector (Drummond Scientific, Broomall, PA). The oocytes were kept at 19BC for 2 to 3 days before the measurements were performed. Data were recorded using a Dagan CA-1B amplifier (Minneapolis, MN), a HEKA EPC-9 interface, and HEKA Pulse v8.54 software (HEKA Elektronik, Lambrecht/Pfalz, Germany). Electrodes, pulled from borosilicate glass capillaries using a Sutter Instrument Co Model P-97 Puller, were filled with 2.5 M KCl and had a resistance of 0.3– 0.6 MV. The currents were analyzed by applying a ramp protocol from 2120 to +40 mV (over 1000 milliseconds) from a holding potential of 280 mV. Only oocytes with membrane potential more negative than 250 mV were used. After a 10minute stabilization period under Kulori solution superfusion, oocytes were superfused with ICA 30 mM for at least 5 minutes, before a washout with Kulori solution. Currents were recorded continuously during the experimental protocol, and the maximal inward and outward current levels were quantified at the end of each superfusion period.

Drugs and Chemicals Unless otherwise mentioned, all chemicals used were of analytical grade and obtained from Sigma-Aldrich (Steinheim, Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

Antiarrhythmic Mechanisms of SK Channel Inhibition

Germany). The selective and potent SK channel blocker used in this study N-(pyridin-2-yl)-4-(pyridin-2-yl)thiazol-2-amine (ICA)23 was synthesized at Acesion Pharma.

Data Analysis

All data are presented as mean 6 SEM. Statistical analysis was performed using the following: paired Student’s t test (Fig. 2); 1-way analysis of variance (ANOVA), repeated measurements with Bonferroni post hoc test (Figs. 3 and 8); two-way ANOVA, repeated measurements with Bonferroni post hoc test (Fig. 6); and 1-way ANOVA, TMC versus drug with Tukey’s multiple comparison post hoc test (Fig. 7). P values ,0.05 were considered statistically significant. GraphPad Prism 5 (GraphPad Software) was used for statistical analysis. In figures, statistical significance is denoted by *P , 0.05, **P , 0.01, and ***P , 0.001.

RESULTS Arrhythmias in Rat Right Atrium In acutely isolated right atrium, 3 types of atrial arrhythmias were observed in the absence of drug: (A) periods of intrinsically initiated AF, lasting for seconds after S1 stimulation only (n = 2 of 12), (B) S2 stimuli–induced sustainable AF, lasting for more than 10 minutes (n = 1 of 12), and (C) short runs of AF induced by S2 stimuli (n = 5 of 12), lasting from a few seconds to minutes (Fig. 1). A higher magnification of the inset in Figure 1B depicting the initiation of an arrhythmic event is found in Supplemental Digital Content 1 (see Fig. 1, http://links.lww.com/JCVP/A186). In the presence of SK channel inhibitor ICA (10, 20, and 30 mM applied at 30, 60, and 90 minutes, respectively), none of the previously described arrhythmias were observed and AF could not be induced by S2 stimuli (Fig. 1D). During the arrhythmias, analyzed as a mean of 5 AF episodes from 5 individual experiments, we observed a profound depolarization of RMP (SR: 273.7 6 0.9 to AF: 260.2 6 1.7 mV), which was accompanied by a significant reduction in action potential amplitude (APA) and maximal upstroke velocity (dV/dtmax). The frequency during AF varied from z15 to 30 Hz (Fig. 2). Interestingly, atrial arrhythmias were initially chaotic and irregular, with a fast frequency. However, with time, this evolved into a more regular arrhythmia, a graduate slowing of the frequency and eventually spontaneous reversion to SR (Fig. 2D). As the arrhythmic event reverted, APA, RMP, and dV/dtmax returned back to baseline values.

Effects of SK Channel Block on APD and ERP The small-molecule SK channel blocker N-(pyridin-2yl)-4-(pyridin-2-yl)thiazol-2-amine, recently found to show selective efficacy for SK channels among relevant cardiac ion channels,22 was tested in 3 concentrations (10, 20, and 30 mM) (Fig. 3). ICA prolonged APD and ERP in a concentration-dependent manner at 5 Hz. The ERP was substantially longer after application of 20 and 30 mM ICA as compared with APD90, giving rise to postrepolarization refractoriness (PRR). In addition, SK channel block also resulted in a concentration-dependent depolarization of RMP, with maximal www.jcvp.org |

167


Skibsbye et al


FIGURE 1. Representative recordings of AF arrhythmias developed in the isolated right atrium during control recordings. Three types of atrial arrhythmias were observed in the absence of drug: (A) Intrinsically initiated AF; (B) S2 stimuli–induced AF, lasting for more than 10 minutes; and (C) Short runs of AF induced by S2 stimuli, lasting for seconds to minutes. D, SK channel inhibition by 10 mM ICA rendered the atrium nonsusceptible to S2-induced arrhythmias. None of the described types of arrhythmia were observed after application of ICA at any given concentration. In all cases described, the pacing frequency was 5 Hz. A higher magnification at the initiation of an arrhythmic event shown by the insert in (B) is found in Supplemental Digital Content 1 (see Fig. 1, http://links.lww.com/JCVP/A186).

effect at 30 mM (5 Hz; from 277.9 6 1.9 to 275.7 6 2.4 mV). In the TMC group, there was no tendency of a change in RMP throughout the experimental procedure (data not shown).

Frequency-dependent Effects on Atrial AP Parameters To investigate frequency-dependent effects, the atria were paced at increasing frequencies from 5 to 9 Hz and from 9 to 11 Hz followed by a decrease back to 9 and 5 Hz, with each frequency lasting approximately 200 stimulations. This was done in the absence and presence of ICA (10, 20, and 30 mM). Representative AP recordings are shown in Figure 4. A diary plot measuring RMP, APA, dV/dtmax, and APD90 as the mean of all individual experiments (n = 6) is presented in Figure 5, and data analysis on concentration dependency of ICA, analyzed at steady-state conditions, at the end of each frequency are summarized in Figure 6A–D. Online representative recordings of the stimulation protocol and drug effect are depicted in Supplemental Digital Content 2 (see Fig. 2, http://links.lww. com/JCVP/A187) and TMC in Supplemental Digital Content 3 (see Fig. 3, http://links.lww.com/JCVP/A188).

Frequency Dependence of SK Channel Block

addition, application of the SK channel blocker resulted in a concentration-dependent depolarization of the RMP and a depression of APA and upstroke velocity (dV/dtmax) at all frequencies. The rate-dependent depression of sodium channel–mediated parameters all reverted upon returning to slower pacing rates, both in the absence and presence of ICA. To address whether the effects on RMP, APA, dV/dtmax, and APD90 caused by ICA were more pronounced at higher frequencies, we analyzed the percentage changes normalized to control values for the frequencies 5, 9, and 11 Hz in Figure 6E–H. In summary, the effects on APA showed frequencydependent increase in effect at all tree ICA concentrations, while dV/dtmax only showed a significant difference between 5 and 11 Hz at 30 mM ICA. For RMP, there was only a significant increase in effect between 5 and 11 Hz at 10 mM, and for APD90, there was no difference in the magnitude of effect between frequencies at any concentration applied. A presentation of the frequency dependence without normalization to control values is given in Supplemental Digital Content 4 (see Fig. 4, http://links.lww.com/JCVP/A189).

ICA Does not Change the Adaptation of AP Parameters to Changes in Pacing Frequency

To demonstrate frequency effects on the atrial AP, superimposed APs from each pacing regime are depicted in Figure 4. As expected, increasing the pacing rate shortened the APD under control conditions. Moreover, depression of sodium channel–mediated parameters such as APA, RMP, and dV/dtmax was also observed with increasing pacing rates (Figs. 5 and 6). Application of the SK channel blocker ICA resulted in prolongation of APD90 at all pacing rates. In

During the time course of the pacing regime both in the presence and absence of drug, the action potential RMP, APA, and upstroke velocity (dV/dtmax) were depressed by increasing pacing frequency (Fig. 5). The depression going from 5 to 9 Hz and from 9 to 11 Hz is similar and recovers almost completely to baseline levels going back from 11 to 9 Hz and 9 to 5 Hz. Both depression and recovery exhibit an initial fast adaptation within the first 10–20 beats followed by slow adaptation reaching a plateau within 200 beats.

168

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

| www.jcvp.org




FIGURE 2. Diary plots of AP parameters recorded at SR and during an episode of AF, as shown in Figure 1B, elicited by a single S2 stimulus. The arrhythmia spontaneously reverts back to SR after approximately 2000 APs. Bar graphs depicts the average parameter in n = 5 individual experiments measured as mean 6 SEM during SR and during the entire sequence of AF. (A) RMP = resting membrane potential, (B) APA = action potential amplitude, (C) dV/dtmax = upstroke velocity, (D) frequency is measured as the mean Hz during the AF episode and is fixed at 5 Hz in SR.

Although application of ICA (10, 20, and 30 mM) depresses these parameters, the adaptation of depression and recovery was not affected as compared with control (Fig. 5).

Frequency and Use-dependent Block of INa To address whether changes in sodium channel–associated parameters could be mediated through direct block of NaV1.5 by ICA, automated patch-clamp experiments were performed. When stimulated at 1 Hz from a holding potential of 2120 mV, ICA (10 or 30 mM) did not cause any significant effect on steady-state current levels. In comparison, 10 mM flecainide significantly inhibited the current amplitude by 15.5% 6 1.7%. However, when applying conditions that resemble the diastolic membrane potential (275 mV) observed from intracellular recordings in rat atria stimulated at 5 Hz, the inhibition by flecainide was significantly augmented (76.0% 6 2.5% inhibition as compared with baseline) (Fig. 7). This is in line with previous findings that flecainide primarily blocks NaV1.5 channels in a use- and statedependent manner, favoring binding to the open and inactivated state of the channel.26 Likewise, at 5 Hz, 30 mM ICA Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

significantly inhibited NaV1.5 by (45.9% 6 2.9%), whereas no significant inhibition was observed at 10 mM ICA (26.9% 6 4.4%) as compared with TMC (24.6% 6 1.4%). The above reported current measurements at 5 Hz were quantified at the 130th stimulation pulse. It can be observed from Figure 7 that stimulating at 5 Hz from a holding potential of 275 mV results in a decrease in peak current amplitude even before drugs are applied (comparing the first pulse and 130th pulse during the 5 Hz stimulation, Fig. 7A). This can be explained by the buildup of inactivated channels, as a result of insufficient time for the sodium channels to be released from inactivation. Moreover, the rapid stimulation of the channel also resulted in a rundown of current, as can be seen on the TMC data (a 24.6% 6 1.4% reduction in current amplitude as compared with baseline recordings). Overall, at high stimulation frequencies and high ICA concentrations (30 mm), some degree of direct usedependent inhibition of the sodium current is achieved.

ICA Does not Inhibit IK,ACh In a series of experiments using two-electrode voltageclamp recordings in Xenopus oocytes expressing Kir3.1 and www.jcvp.org |

169


Skibsbye et al


milliseconds (n = 6) (Fig. 8B). Despite a slowing in CV, the calculated wavelength (CV · ERP) was increased by 74% from 18.3 mm at baseline values to 31.8 mm at 10 mM ICA. Effects on neither CV nor ERP could be washed out (Fig. 8A, B). This is in line with the fact that high concentrations of ICA are needed for producing functional effects and that the temporal onset of effect is slow, indicating low diffusion rates of ICA into the tissue. The contractile force was measured simultaneously. ICA had no effects on inotropy (systolic developed force analyzed at baseline and 10 mM: 196.3 6 44.5 to 185.2 6 49.2 mg/mm2, n = 6) compared with DMSO-treated group (187.0 6 38.4 to 144.5 6 39.3 mg/mm2, n = 5) (Fig. 8C).

DISCUSSION SK Channel Block Prevents Arrhythmias in the Isolated Rat Atrium

FIGURE 3. Representative recordings superimposed depicting the concentration dependency of SK channel inhibition (n = 6). At 5 Hz, ICA (10, 20, and 30 mM) prolongs APD90 and depolarizes RMP in a concentration-dependent manner (A). ERP augmentation exceeded APD90 prolongation, giving rise to PRR (B).

Kir3.4, 30 mM ICA exhibited no effect on the maximal IK,ACh current levels (see Fig. 5, Supplemental Digital Content 5, http://links.lww.com/JCVP/A190).

SK Block by ICA Slows CV It was observed that application of ICA was associated with a delay in conduction time (measured as the time delay from stimulation artifact to maximal upstroke velocity) at higher frequencies (Fig. 4D). To address whether this SK blocker–mediated delay in conduction time correlates to a slowing in CV, we performed a series of experiments in rat right atrial muscle strips measuring CV. The atrial preparations were stimulated at 5 Hz during the whole experimental procedure, and ERP measurements were performed with an S1–S2 pacing protocol after every perfusion period. With this methodological approach, lower concentrations of ICA (1–10 mM) were applied to avoid conduction block, which was observed in a couple of experiments after application of 30 mM. ICA significantly slowed CV as compared with time-matched vehicle controls (DMSO) in a concentration-dependent manner from 3 mM with a maximal decrease by 16% from 0.61 6 0.12 m/s to 0.51 6 0.08 m/s (n = 6) at 10 mM compared with baseline (Fig. 8A). Likewise, ERP was significantly prolonged compared with baseline at 3 and 10 mM with a maximal effect at 10 mM from 30.0 6 5.5 milliseconds to 62.5 6 7.5

170

| www.jcvp.org

The arrhythmias observed in the isolated rat atria, which were either spontaneously occurring or induced by extra stimuli, are very similar in shape and frequency to those observed in previous studies measured either in vivo or in Langendorff hearts.18,19 This is the first study to describe atrial arrhythmias recorded intracellularly in the isolated rat atrium. All atrial preparations were protected against S2 stimuli– induced arrhythmias in the presence of SK channel blocker. Work done by Antzelevitch et al previously reported intracellular recorded episodes of AF in canine isolated right atrial preparations.27 However, these arrhythmias were reported to be more atrial flutter-like, which might be a consequence of the longer ERP in the canine atrium, compared with the rat, not generating substrate for fibrillating rotors to perpetuate AF. During the arrhythmic event, RMP was consistently depolarized to around 260 mV. This would be expected to have dramatic effects on sodium channel availability, as also seen by the immediate depression of APA and dV/dtmax. However, still at this potential, the membranes were excitable enough to maintain electrical propagation of arrhythmic impulses. It is, however, possible that at high frequencies and drug concentrations, ICA exhibits some degree of use-dependent sodium channel inhibition, which would also depress APA and dV/dtmax.

SK Channel Block Affects APD, ERP, and RMP In this study, we observed multiple effects of SK channel inhibition by ICA, including prolonged APD and ERP, as well as depolarization of RMP. Although the human atrial AP is different from the rat atrial AP in terms of both morphology and duration, we recently described quite similar AP effects in trabeculae from human atrial appendages after SK channel inhibition.22 In this study, we sought to investigate the consequences of SK channel–mediated changes in RMP. The SK current has distinct rectifying kinetics different from the IK1 current, which is known to be an important player in setting cardiac RMP. Furthermore, the SK current does not rectify to the same degree as IK1, leaving a larger outward potassium conductance (ISK)25 at more depolarized potentials, such as during phase 3 of the AP. SK channel opening is obligatedependent on rises in intracellular calcium, which makes it questionable whether SK channels are activated throughout Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.




FIGURE 4. Frequency-dependent effects of AP parameters at 5, 9, and 11 Hz in control situation (A) and in the presence of SK channel blocker applied in 3 different concentrations (ICA: 10 (B), 20 (C), and 30 (D) mM). RMP was depolarized with increasing frequency, and this depolarization was augmented with increasing concentrations of ICA. With increasing frequencies, APD90 was abbreviated; however, this abbreviation was compensated by a concentration-dependent prolongation by ICA, rendering the cells unable to reach full repolarization at high frequencies. Conduction time (D time between pacing artifact to maximal upstroke velocity) was only prolonged by ICA at 11 Hz in a concentration-dependent manner. In all panels, 200-millisecond recording durations are shown.

the diastolic interval. The exact association and dissociation rates of Ca2+ to calmodulin have not been studied in cardiac cells. However, in CA1 pyramidal neurons, SK channel current mediates intermediate afterhyperpolarization that lasts hundreds of milliseconds.2 This notion thereby opens up for a spatial Ca2+ regulation that will not be directly correlated to global Ca2+ dynamics, which could explain diastolic active SK channels (see Limitations section). SK channel block by ICA depolarizes RMP by 2–3 mV, which will influence the voltage-dependent release from

inactivation and the voltage-dependent steady-state inactivation of voltage-gated sodium channels. Increasing the stimulation frequency will further push the sodium channel into the inactivated state, as less time is available for releasing Na channels from inactivation. Under control conditions, we found, as expected, that increasing pacing rates abbreviates the APD. However, in the presence of ICA, the APD is not abbreviated to the same degree as in the absence of drug (due to the direct effect of blocking repolarizing SK currents), rendering the cells unable to reach full repolarization at high

FIGURE 5. Diary plot of analyzed AP parameters depicting the frequency dependence in effect obtained by ICA. Summarized values of all experiments (n = 6). Each point representing a mean of 6 APs from 6 different experiments summarized at the same time point. ICA produced an increasing frequency-dependent depression on APA (B) and dV/dtmax (C). However, for RMP (A) and APD90 (D), the effect produced by ICA was not significantly different at higher frequencies compared with 5 Hz. For RMP, APA, and dV/dtmax, an almost instantaneous frequency adaptation was observed followed by slow adaptation to a plateau phase before the next frequency was applied. For APD90, the frequency adaptation was slow and steady-state levels were not reached after 200 beats before frequency changes. Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

www.jcvp.org |

171


Skibsbye et al


FIGURE 6. A–D, Bar graphs depicting the concentration dependency analyzed from the summarized data in Figure 5 in mean values taking at the plateau phase (at the end) of each frequency. It was observed that the all parameters analyzed followed a concentrationdependent effect by ICA 10–30 mM. E–H, Bar graphs depicting the frequency dependence of drug effect analyzed from the data in (A–D). This was analyzed as the percent change from baseline, normalized to the respective control value at the given frequency for each concentration (10, 20, and 30 mM). Summarized, the effects on APA at 9 and 11 Hz were significantly greater in the presence of 20 and 30 mM ICA, when normalized to control (absence of drug), stating the concentration dependency of drug effect on sodium channel–dependent parameters. At 10 uM ICA there were only difference between 5 and 11 Hz for RMP and APA, and only at 30 mM ICA for dV/dtmax. For APD90, there were no differences in the magnitude of drug effect among frequency groups.

172

| www.jcvp.org





FIGURE 7. State- and use-dependent effects of ICA on rNaV1.5 channels. A, Representative current recording of rNaV1.5 elicited by 80 depolarization pulses to 220 mV from a holding potential of 2120 (1 Hz) followed by 130 depolarization pulses to 220 mV from a holding potential of 275 mV (5 Hz). Recordings are superimposed as before drug (black), after application of 30 mM ICA (dark gray) and 10 mM flecainide (light gray). Vertical arrows indicate the last pulse at 1 Hz and the 130th pulse at 5 Hz, respectively, used for data analysis. B, Zoom view of the 130th pulse during 5 Hz stimulation for TMC experiments (left) and application of ICA (30 mM) (Right). The horizontal arrows indicate current amplitude at baseline, saline and flecainide (TMC, 5 Hz) and baseline, ICA 30 mM and flecainide (ICA, 5 Hz). C, Percentage of current amplitude inhibition as compared with baseline recordings, measured at stimulation rates of 1 and 5 Hz (final pulse, indicated by the vertical arrows) of TMC, ICA 10, 30 mM and flecainide 10 mM (n = 5–7).

frequencies (Fig. 4D). In combination with the effects on RMP by SK channel block, this depolarized potential produces a clear frequency-dependent depression of the recorded sodium channel–mediated parameters, APA, dV/dtmax, and CV. In addition, a combined effect of direct sodium channel inhibition by ICA at high (30 mM) concentration and pacing frequency might also contribute. As ICA affect in 10 mM is not found to block sodium channel activity directly, our results suggest that SK channel Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

inhibition, in addition to prolonging APD and ERP by delaying repolarization, might also cause a reduced availability of sodium channels, an effect that is potentiated at higher frequencies. The combined effects of use-dependent indirect sodium channel block and prolonged APD/ERP will be expected to act in concert, producing PRR leading to reduced excitability. This combined effect would thereby constitute the antiarrhythmic mechanism of SK channel inhibition. However, as ICA in higher concentrations was found to www.jcvp.org |

173


Skibsbye et al


directly inhibit INa, a contribution from a direct INa block cannot be ruled out. It is of general understanding that traditional antiarrhythmic drugs, associated to class IA and IC sodium channel blockers, produce negative inotropic effects secondary to their antiarrhythmic effect.29,30 In this study, the indirect effect on sodium channel functionality was not associated with changes in systolic contractility.

Slowing of CV To address other parameters that are strictly dependent on sodium channel availability, we conducted a series of experiments testing the effects of ICA on CV in tissue strips from the rat right atrium. Similar to the effects on dV/dtmax, SK channel inhibition by ICA slows CV in a concentrationdependent manner, at concentrations not found to directly inhibit sodium channels. This slowing of CV was accompanied with a prolongation of the ERP and speaks in favor of an indirect block of sodium channels mediated through the block of SK channel activity in the diastolic phase. Although we know today that the genesis of AF is highly complex, the wavelength theory (wavelength = conduction velocity · refractory period) tells us that a slowing of the conducting impulse will shorten the length of the excitation wave, thus increasing self-sustainability of a circulating wave ultimately leading to reentry arrhythmias.32 In accordance, drugs preventing CV slowing by preservation of gap junctional coupling are antiarrhythmic.33,34 Also, CV slowing combined with heterogeneity in cellular communication and conduction has been shown to be involved in the genesis of spiral reentry arrhythmias.35 However, according to the spiral wave theory, reduced sodium current and conduction slowing could be antiarrhythmic through several mechanisms—among others: (1) enlargement of the center of rotation and (2) reduction in the number of secondary wavelets that could provide new primary rotors.36 In this study, we observed a decrease in CV, a prolonged ERP, resulting in an overall increased wavelength that may partly explain the antiarrhythmic properties of SK inhibition exhibited by ICA. Furthermore, the combined effects of increased wavelength associated with a depolarized RMP and thereby decreased excitability, could result in block of abnormal AP initiation and propagation at high frequencies and become antiarrhythmic by terminating multiple reentry arrhythmias.

Could a Single-channel Targeting Drug be Antiarrhythmic? Whether single-channel targeting is sufficient for antiarrhythmic efficacy in humans is still an open question. Here, we demonstrate how a reported selective SK channel inhibitor in addition to blocking potassium channels seemingly leads to indirect effects on sodium channel availability

FIGURE 8. CV measured in rat right atrial muscle strips in vehicle controls (n = 6) and ICA (n = 6) at (baseline) and after application of 1, 3, and 10 mM ICA tested at 5 Hz frequency. Normalized CV is given in (A) showing that ICA significantly decreases CV in a concentration-dependent irreversible manner (z10% at 3 mM and z16% at 10 mM). Concomitant to the decrease in CV, the ERP (B) was concentration-dependent and irreversibly prolonged at 3 and 10 mM. The calculated

wavelength was, however, markedly increased. No effect was observed in time-matched DMSO controls neither on CV nor on ERP. Developed contractile force (C) was measured simultaneously, without any effect of ICA throughout the experiment compared with DMSO controls.

174


| www.jcvp.org



by depolarizing the diastolic potential. Such dual effects of blocking potassium channels could perhaps increase antiarrhythmic efficacy and help to explain the antiarrhythmic mechanisms of SK channel inhibition. Although caution needs to be taken into account as direct sodium channel inhibition was observed at high concentration of ICA. SK channels seem to be more abundantly expressed in the atrium compared with the ventricles, and they are functionally more important in human atria compared with ventricles.5,22 We have recently reported that inhibition of SK channels is antiarrhythmic in different animal models of experimental AF.18–21 SK channel inhibition modulates multiple parameters of AP, both through direct blockage of repolarizing SK current and shifting RMP through blockage of diastolic active SK current, which leads to an indirect sodium block and ultimately conduction slowing. The antiarrhythmic effects could therefore be more comparable with the effects seen from multiple channel blockers.37 However, future studies addressing possible antiarrhythmic efficacy of depolarizing the diastolic potential through compounds inhibiting potassium channels believed to be important for setting the diastolic potential will help to clarify whether this concept constitutes a new antiarrhythmic principle. Moreover, clinical evidence is needed to verify translatability of experimental antiarrhythmic effects into therapeutic efficacy in human AF.

Limitations As we have not addressed subcellular calcium concentration changes, we can only speculate on how Ca2+ dynamics are affected by the applied pacing protocols and how this interplays with SK channel Ca2+ activation and regulation by SK inhibition. Also, association–dissociation kinetic studies of the cardiac Ca2+/calmodulin interaction were not addressed in this study but would be valuable assets. ICA was in the CV experiments applied in lower concentrations compared with whole atrial AP experiments, due to conduction block at 30 mM and higher concentration, which was not observed in other experiments. We assume that the reason for this inconsistency must be related to drug availability between experimental methods. Small muscle strips that are sutured in both ends and stretched out in the organ bath have larger availability (less perfusion boundary) for lipophilic compounds to reach their target, compared with an intact isolated atrium. Second, CV measurements were recorded from the endocardial side of the atrial tissue strip, while AP measurements were recorded from the epicardial side of the atrium. From automated patch-clamp experiments, we observed a significant state- and use-dependent inhibition of NaV1.5 current amplitude at 30 mM ICA at 5 Hz stimulation frequency. Thus, in a physiological context, at high stimulation frequencies and high ICA concentrations, a direct use-dependent inhibition of the sodium current cannot be ruled out. It should however be noted that electrophysiological experiments performed at room temperature will decelerate channel kinetics to an extent leading to potentiation of use-dependent block of voltage-dependent sodium channels. This fact also prevented us from conducting NaV1.5 electrophysiology at 11 Hz frequencies (resembling the functional studies conducted Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.


in an intact rat atrium) due to the build-up of inactivated channels and a dramatic depression of current amplitude. As ICA is reported to block hKv4.3 channels with an IC50 value of z21 mM, we cannot exclude that some of the effects observed partly can be attributed to the inhibition of Ito current.

CONCLUSIONS In this study, we aimed at elucidating the direct and indirect antiarrhythmic mechanisms of SK channel inhibition. ICA induces PRR and exhibits antiarrhythmic effects by protecting against the induction of AF. We observed no effect on systolic contractile force as a consequence of ICA application. SK channel inhibition modulates multiple parameters of the AP, both through direct blockage of the repolarizing SK current and by shifting RMP. This most likely leads to an indirect sodium block through accumulation of state dependently inactivated channels and ultimately conduction slowing. The effects by SK channel block on sodium channel–associated parameters showed frequency dependence. The antiarrhythmic effect obtained by SK channel inhibition is thereby multifaceted, potentially constituting a new antiarrhythmic principle.

ACKNOWLEDGMENTS The authors thank Acesion Pharma A/S, Copenhagen, Denmark, for synthesizing and providing ICA. The Danish National Research Foundation, The Lundbeck Foundation, The Novo Nordisk Foundation, Innovation Fund Denmark, and The Danish Council for Independent Research are thanked for financial support. REFERENCES 1. Chen MX, Gorman SA, Benson B, et al. Small and intermediate conductance Ca(2+)-activated K+ channels confer distinctive patterns of distribution in human tissues and differential cellular localisation in the colon and corpus cavernosum. Naunyn Schmiedebergs Arch Pharmacol. 2004;369:602–615. 2. Stocker M. Ca(2+)-activated K+ channels: molecular determinants and function of the SK family. Nat Rev Neurosci. 2004;5:758–770. 3. Berkefeld H, Fakler B, Schulte U. Ca2+-activated K+ channels: from protein complexes to function. Physiol Rev. 2010;90:1437–1459. 4. Kuiper EF, Nelemans A, Luiten P, et al. K(Ca)2 and k(ca)3 channels in learning and memory processes, and neurodegeneration. Front Pharmacol. 2012;3:107. 5. Tuteja D, Rafizadeh S, Timofeyev V, et al. Cardiac small conductance Ca2+-activated K+ channel subunits form heteromultimers via the coiled-coil domains in the C termini of the channels. Circ Res. 2010; 107:851–859. 6. Park YB. Ion selectivity and gating of small conductance Ca(2+)-activated K+ channels in cultured rat adrenal chromaffin cells. J Physiol. 1994;481(pt 3):555–570. 7. Lang DG, Ritchie AK. Tetraethylammonium ion sensitivity of a 35-pS CA2(+)-activated K+ channel in GH3 cells that is activated by thyrotropin-releasing hormone. Pflugers Arch. 1990;416:704–709. 8. Pedarzani P, Stocker M. Molecular and cellular basis of small—and intermediate-conductance, calcium-activated potassium channel function in the brain. Cell Mol Life Sci. 2008;65:3196–3217. 9. Mahida S. Expanding role of SK channels in cardiac electrophysiology. Heart Rhythm. 2014;11:1233–1238. 10. Terentyev D, Rochira JA, Terentyeva R, et al. Sarcoplasmic reticulum Ca (2)(+) release is both necessary and sufficient for SK channel activation in ventricular myocytes. Am J Physiol Heart Circ Physiol. 2014;306: H738–H746.

www.jcvp.org |

175


Skibsbye et al

11. Chua SK, Chang PC, Maruyama M, et al. Small-conductance calciumactivated potassium channel and recurrent ventricular fibrillation in failing rabbit ventricles. Circ Res. 2011;108:971–979. 12. Ni Y, Wang T, Zhuo X, et al. Bisoprolol reversed small conductance calcium-activated potassium channel (SK) remodeling in a volumeoverload rat model. Mol Cell Biochem. 2013;384:95–103. 13. Chang PC, Turker I, Lopshire JC, et al. Heterogeneous upregulation of apamin-sensitive potassium currents in failing human ventricles. J Am Heart Assoc. 2013;2:e004713. 14. Chang PC, Hsieh YC, Hsueh CH, et al. Apamin induces early afterdepolarizations and torsades de pointes ventricular arrhythmia from failing rabbit ventricles exhibiting secondary rises in intracellular calcium. Heart Rhythm. 2013;10:1516–1524. 15. Ellinor PT, Lunetta KL, Glazer NL, et al. Common variants in KCNN3 are associated with lone atrial fibrillation. Nat Genet. 2010;42:240–244. 16. Olesen MS, Jabbari J, Holst AG, et al. Screening of KCNN3 in patients with early-onset lone atrial fibrillation. Europace. 2011;13:963–967. 17. Luo Z, Yan C, Zhang W, et al. Association between SNP rs13376333 and rs1131820 in the KCNN3 gene and atrial fibrillation in the Chinese Han population. Clin Chem Lab Med. 2014. 18. Diness JG, Sorensen US, Nissen JD, et al. Inhibition of smallconductance Ca2+-activated K+ channels terminates and protects against atrial fibrillation. Circ Arrhythm Electrophysiol. 2010;3:380–390. 19. Skibsbye L, Diness JG, Sorensen US, et al. The duration of pacinginduced atrial fibrillation is reduced in vivo by inhibition of small conductance Ca(2+)-activated K(+) channels. J Cardiovasc Pharmacol. 2011;57:672–681. 20. Diness JG, Skibsbye L, Jespersen T, et al. Effects on atrial fibrillation in aged hypertensive rats by Ca(2+)-activated K(+) channel inhibition. Hypertension. 2011;57:1129–1135. 21. Qi XY, Diness JG, Brundel BJ, et al. Role of small-conductance calciumactivated potassium channels in atrial electrophysiology and fibrillation in the dog. Circulation. 2014;129:430–440. 22. Skibsbye L, Poulet C, Diness JG, et al. Small-conductance calciumactivated potassium (SK) channels contribute to action potential repolarization in human atria. Cardiovasc Res. 2014. 23. Gentles RG, Grant-Young K, Hu S, et al. Initial SAR studies on apamin-displacing 2-aminothiazole blockers of calcium-activated small

176

| www.jcvp.org


24. 25. 26. 27. 28. 29. 30.

31. 32. 33. 34. 35. 36. 37.

conductance potassium channels. Bioorg Med Chem Lett. 2008;18: 5316–5319. Olsen KB, Braunstein TH, Sorensen CM, et al. Angiotensin II does not acutely regulate conduction velocity in rat atrial tissue. Scand J Clin Lab Invest. 2011;71:492–499. Grunnet M, Jensen BS, Olesen SP, et al. Apamin interacts with all subtypes of cloned small-conductance Ca2+-activated K+ channels. Pflugers Arch. 2001;441:544–550. Wang GK, Russell C, Wang SY. State-dependent block of wild-type and inactivation-deficient Na+ channels by flecainide. J Gen Physiol. 2003; 122:365–374. Burashnikov A, Di Diego JM, Barajas-Martinez H, et al. Ranolazine effectively suppresses atrial fibrillation in the setting of heart failure. Circ Heart Fail. 2014;7:627–633. Deleted in proof. Kim SY, Benowitz NL. Poisoning due to class IA antiarrhythmic drugs. Quinidine, procainamide and disopyramide. Drug Saf. 1990;5:393–420. Kihara Y, Inoko M, Hatakeyama N, et al. Mechanisms of negative inotropic effects of class Ic antiarrhythmic agents: comparative study of the effects of flecainide and pilsicainide on intracellular calcium handling in dog ventricular myocardium. J Cardiovasc Pharmacol. 1996;27:42–51. Deleted in proof. Lammers WJ, Allessie MA. Pathophysiology of atrial fibrillation: current aspects. Herz. 1993;18:1–8. Xing D, Kjolbye AL, Nielsen MS, et al. ZP123 increases gap junctional conductance and prevents reentrant ventricular tachycardia during myocardial ischemia in open chest dogs. J Cardiovasc Electrophysiol. 2003;14:510–520. Haugan K, Olsen KB, Hartvig L, et al. The antiarrhythmic peptide analog ZP123 prevents atrial conduction slowing during metabolic stress. J Cardiovasc Electrophysiol. 2005;16:537–545. Nakagami T, Tanaka H, Dai P, et al. Generation of reentrant arrhythmias by dominant-negative inhibition of connexin43 in rat cultured myocyte monolayers. Cardiovasc Res. 2008;79:70–79. Kneller J, Kalifa J, Zou R, et al. Mechanisms of atrial fibrillation termination by pure sodium channel blockade in an ionically-realistic mathematical model. Circ Res. 2005;96:e35–e47. Ehrlich JR, Nattel S. Novel approaches for pharmacological management of atrial fibrillation. Drugs. 2009;69:757–774.



Pharmacologic inhibition of small-conductance calcium-activated potassium (SK) channels by NS8593 reveals atrial antiarrhythmic potential in horses.

Relaxin binding in the rat heart atrium.

Upregulation of the rat cardiac sodium channel by in vivo treatment with a class I antiarrhythmic drug.

Mechanisms of glycolytic inhibition in ischemic rat hearts.

Mechanisms of neurotensin-induced inhibition in rat ileal smooth muscle.

Synergistic antiarrhythmic effect of combining inhibition of Ca²⁺-activated K⁺ (SK) channels and voltage-gated Na⁺ channels in an isolated heart model of atrial fibrillation.

Quinacainol, a new antiarrhythmic with class I antiarrhythmic actions in the rat.

Inhibition of corneal neovascularization in the rat by SK&F 86002, a dual inhibitor of arachidonic acid metabolism.

Erratum to: Short- and long-term inhibition of cardiac inward-rectifier potassium channel current by an antiarrhythmic drug bepridil.

Properties of the muscarinic cholinergic receptors in rat atrium.

Modes of the Na channel blocking action of pilsicainide, a new antiarrhythmic agent, in cardiac cells.

Antiarrhythmic drugs impair hepatic uptake and secretory function by different mechanisms in the isolated perfused rat liver.

Inhibition of lipid deposition in the hypercholesterolemic rat by clentiazem, a calcium channel blocker.

Sympathoexcitation in ANG II-salt hypertension involves reduced SK channel function in the hypothalamic paraventricular nucleus.

Mitosis-arresting effect of the calcium channel inhibitor SK&F 96365 on human leukemia cells.

PITX2 Modulates Atrial Membrane Potential and the Antiarrhythmic Effects of Sodium-Channel Blockers.

Molecular basis of hERG potassium channel blockade by the class Ic antiarrhythmic flecainide.

Dual mechanisms of inhibition of the immune response by enterocytes isolated from the rat small intestine.

Ionic mechanisms responsible for the antiarrhythmic action of dehydroevodiamine in guinea-pig isolated cardiomyocytes.

Inhibition of acid-sensing ion channel currents by propofol in rat dorsal root ganglion neurons.

Inhibition of acid sensing ion channel by ligustrazine on angina model in rat.

Distinct mechanisms underlying cholesterol protection against alcohol-induced BK channel inhibition and resulting vasoconstriction.

SK channel blockade reverses cognitive and motor deficits induced by nigrostriatal dopamine lesions in rats.

Involvement of uptake2 in the termination of activity of neurogenic noradrenaline in the rat isolated atrium.