Pharmacologic inhibition of small-conductance calcium-activated potassium (SK) channels by NS8593 reveals atrial antiarrhythmic potential in horses Maria Mathilde Haugaard, DVM, FHRS,* Eva Zander Hesselkilde, DVM,* Steen Pehrson, MD, PhD,† Helena Carstensen, DVM,* Mette Flethøj, DVM,* Kirstine Færgemand Præstegaard, DVM,* Ulrik Svane Sørensen, PhD,‡ Jonas Goldin Diness, PhD, FHRS,‡ Morten Grunnet, PhD, FHRS,‡ Rikke Buhl, DVM, PhD,* Thomas Jespersen, PhD, FHRS§ From the *Department of Large Animal Sciences, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark, †Department of Cardiology, The Heart Centre, Copenhagen University Hospital, Taastrup, Denmark, ‡Acesion Pharma, Copenhagen, Denmark, and §Danish National Foundation Research Centre in Arrhythmias (DARC) and Department of Biomedical Sciences, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark. BACKGROUND Small-conductance calcium-activated potassium (SK) channels have been found to play an important role in atrial repolarization and atrial fibrillation (AF). OBJECTIVE The purpose of this study was to investigate the existence and functional role of SK channels in the equine heart. METHODS Cardiac biopsies were analyzed to investigate the expression level of the most prominent cardiac ion channels, with special focus on SK channels, in the equine heart. Subcellular distribution of SK isoform 2 (SK2) was assessed by immunohistochemistry and confocal microscopy. The electrophysiologic and anti-AF effects of the relative selective SK channel inhibitor NS8593 (5 mg/kg IV) were evaluated in anesthetized horses, focusing on the potential of NS8593 to terminate acute pacing-induced AF, drug-induced changes in atrial effective refractory period, AF duration and vulnerability, and ventricular depolarization and repolarization times. RESULTS Analysis revealed equivalent mRNA transcript levels of the 3 SK channel isoforms in atria compared to ventricles. Immunohistochemistry and confocal microscopy displayed a widespread distribution of SK2 in both atrial and ventricular cardiomyocytes. NS8593 terminated all induced AF episodes (duration Z15 minutes), caused pronounced prolongation of atrial effective refractory period, and reduced AF duration and vulnerability. QRS duration and QTc interval were not affected by treatment.

The study was generously funded by The Danish Horse Levy Foundation, The University of Copenhagen, The Danish Council for Independent Research, and The Lundbeck Foundation. Drs. Sørensen, Diness, and Grunnet are employed by Acesion Pharma, which is the company providing the test drug NS8593. However, Acesion Pharma did not provide financial support for this study. Address reprint requests and correspondence: Dr. Maria M. Haugaard, Department of Large Animal Sciences, Faculty of Health and Medical Sciences, University of Copenhagen, Højbakke gaardallé 5, 2630 Taastrup, Denmark. E-mail address: [email protected].

1547-5271/$-see front matter B 2015 Heart Rhythm Society. All rights reserved.

CONCLUSION SK channels are widely distributed in atrial and ventricular cardiomyocytes and contribute to atrial repolarization. Inhibition by NS8593 terminates pacing-induced AF of short duration and decreases AF duration and vulnerability without affecting ventricular conduction and repolarization. Thus, inhibition by NS8593 demonstrates clear atrial antiarrhythmic properties in healthy horses. KEYWORDS Horse; Equine; NS8593; Atrial fibrillation; Pacing; Programmed electrical stimulation; Atrial electrophysiology; Antiarrhythmic drugs; Reverse transcription polymerase chain reaction ABBREVIATIONS aEGM ¼ intra-atrial electrogram; aERP ¼ atrial effective refractory period; AF ¼ atrial fibrillation; ANOVA ¼ analysis of variance; APD ¼ action potential duration; cDNA ¼ complementary DNA; dV/dtmax ¼ maximum upstroke velocity; HR ¼ heart rate; IV ¼ intravenous; RAA ¼ right atrial appendage; RTPCR ¼ reverse transcription polymerase chain reaction; RV ¼ right ventricular free wall; S-ECG ¼ surface electrocardiogram; SK ¼ small-conductance calcium-activated potassium; SK1 ¼ smallconductance calcium-activated potassium channel isoform 1; SK2 ¼ small-conductance calcium-activated potassium channel isoform 2; SK3 ¼ small-conductance calcium-activated potassium channel isoform 3; SR ¼ sinus rhythm; T½ ¼ plasma concentration half-life (Heart Rhythm 2015;12:825–835) I 2015 Heart Rhythm Society. All rights reserved.

Introduction Substantial amounts of data have provided evidence of an important functional role of small-conductance Ca2þ-activated Kþ channels (SK1–SK3 channels) in atrial cardiomyocytes. Studies have shown that inhibition of these channels displays atrial antiarrhythmic effects across a number of species.1–4 SK channels are voltage-insensitive Kþ channels activated exclusively by an increase in free http://dx.doi.org/10.1016/j.hrthm.2014.12.028

826 intracellular Ca2þ that interacts with calmodulin bound to the C-terminal region of the SK channels.5 Activation of the SK channels results in Kþ outflow, thereby contributing to atrial repolarization.1,6 Since atrial diastolic sarcoplasmic reticulum Ca2þ leak seems to be increased during atrial fibrillation (AF),7 this may lead to an enhanced activation of SK channels, accelerated repolarization, and consequently shortened action potential duration (APD), which could constitute a proarrhythmic substrate for the maintenance of AF. Genome-wide association studies have linked variants in the KCNN3 gene, which encodes the SK isoform 3 (SK3) channel, to an increased risk of AF in human patients.8 Overexpression of SK3 has been shown to be proarrhythmic,9 and studies of experimental AF in various animal models have reported that inhibition of SK channels has antiarrhythmic properties.1–4 Because SK channels have not been shown to make a considerable contribution to the repolarization of the ventricular action potential,4,6,10–12 except in failing hearts13 and acute myocardial infarction,14 pharmacologic inhibition of SK channels has been proposed as a potential atrialselective target for treatment of AF. SK channels have been described as functionally important in atrial repolarization in humans,6,10 dogs,1 rabbits,4 guinea pigs,4 rats,3,4 and mice,10 but the existence and possible role of SK channels in the equine heart have not yet been investigated. Horses have an extraordinary cardiac capacity, an impressive ability to increase cardiac output in response to increased requirements, and a relatively large heart compared to other mammals. Horses are under high parasympathetic tone at rest and have an impressive heart rate (HR) span that ranges from approximately 30 bpm at rest to 220 bpm during maximal performance. The heart of the horse is unique, but with respect to AF, the horse shares multiple features with humans. Most importantly, AF is a common arrhythmia in both species. Like humans, horses develop AF both with and without detectable underlying structural heart disease; therefore, the horse may constitute an interesting large-animal AF model.15 The compound NS8593 is a potent and relatively selective negative modulator of SK channels that has shown antiarrhythmic potential in multiple species.1–4,6,16 The negative modulatory effect of NS8593 is exerted via a concentration-dependent rightward shift of the 2þ concentration-response curve for Ca , making SK channels less sensitive to free intracellular Ca2þ.16,17 NS8593 blocks all 3 SK channel isoforms with equal potency.16 The present study was designed to investigate the presence and functional role of SK channels in the equine heart. Specifically, we focused on studying SK isoform mRNA expression and SK protein distribution as well as the electrophysiologic and anti-AF effects after treatment with NS8593 in anesthetized horses.

Heart Rhythm, Vol 12, No 4, April 2015 study and 7 were Standardbred horses included in the in vivo part.

Quantitative reverse transcription polymerase chain reaction In order to investigate the regional expression level of the most prominent calcium, sodium, and potassium ion channels, with special focus on the expression and distribution of SK channels in the equine heart, quantitative reverse transcription polymerase chain reaction (RT-PCR) was performed. Steadystate mRNA transcription from the following genes was investigated: CACNA1C (Cav1.2), SCN5A (Nav1.5), KCND3 (Kv4.3), KCNIP2 (KChiP2), KCNA5 (Kv1.5), KCNH2 (ERG1), KCNQ1 (Kv7.1), KCNJ2 (Kir2.1), KCNJ3 (GIRK1), KCNJ5 (GIRK4), KCNN1 (SK1), KCNN2 (SK2), and KCNN3 (SK3). Cardiac tissue (right atrial appendage [RAA], left atrial appendage, right ventricular free wall [RV], left ventricular endocardium, left ventricular mid-myocardium, and left ventricular epicardium) from 6 slaughtered horses (1 stallion and 5 mares) were included in the in vitro study population (age 9.6 ⫾ 2.3 years). Inclusion criteria were a history of absent cardiac-related diseases of any kind. Detailed information regarding the collection of tissue is provided in the Online Supplemental Appendix, including Table A1. The tissue samples were homogenized in Precellys24 (Bertin Technologies, Paris, France) and purified with Tri Reagent (Sigma-Aldrich, St. Louis, MO) according to the manufacturer’s instructions. Chloroform was added to separate the RNA fraction from DNA and proteins. RNA was extracted and washed several times with a 70% alcohol solution. Successful RNA extraction was evaluated on 1% agarose gels to verify the presence of the ribosomal RNA 28S and 18S. RNA concentration was manually estimated using the RNA ladder as reference (150 ng/mL). Furthermore, RNA was reverse transcribed and copied into a complementary DNA (cDNA) sequence, using random sequence oligonucleotide primers. cDNA was synthesized from 2 mg total RNA using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Naerum, Denmark). Quantitative RTPCR was performed using TaqMan Gene Expression (Life Technologies, Naerum, Denmark) with specific probes and primers targeting the cDNA sequences of interest (see Online Supplemental Appendix). The efficiency of each assay was evaluated as previously described.18 Each of the 13 genes of interest from the 6 described cardiac regions from 6 hearts was analyzed in triplets on the 7300 real-time PCR System (Applied Biosystems). All data were normalized to the highly conserved gene expression of ACTB (β-actin). In each experiment, 3 sample-free wells were included, testing the plate for DNA contamination. Relative gene expression was analyzed using the 2-ΔCt-method.18,19

Materials and methods

Immunohistochemistry and confocal microscopy

Animals

To confirm the expression and determine the subcellular distribution of SK2 protein, immunohistochemistry with confocal microscopy using thin slices of equine RAA and

A total of 13 horses were included in the study. Of these, 6 were slaughter horses included in the molecular part of the

Haugaard et al

SK Channel Inhibition and AF in Horses

RV tissue was performed (see Online Supplemental Appendix).

Equine in vivo AF model The study population comprised 7 Standardbred trotters (6 mares and 1 gelding, age 8 ⫾ 2 years, body weight 480 ⫾ 18 kg). Before enrollment in the present study, 6 horses participated in a standing electrophysiologic study investigating the effects of flecainide on equine atrial electrophysiology, in which 4 horses also participated in a standing control study.20 A washout period of 8 ⫾ 2 days (range 2–14 days) between flecainide treatment and the present study was incorporated in the study design. The horses showed no signs of cardiovascular disease based on history, clinical examination, 24-hour ECG, and routine echocardiographic examination.21 The studies were approved by The Danish Animal Experiments Inspectorate (license number 2012-15-293400198) and performed in accordance with the Danish guidelines for animal experiments according to the European Commission Directive 86/609/EEC. The procedures were conducted under general anesthesia. Before anesthesia induction, all horses were prepared and catheterized as previously described.20 Preparation, catheterization, and anesthesia details are provided in the Online Supplemental Appendix. Intravenous (IV) dobutamine (PharmaCoDane, Herlev, Denmark) treatment was momentarily initiated to obtain a stable arterial blood pressure 470 mm Hg throughout the procedure. A multipolar steerable electrode (Inquiry Steerable Diagnostic Catheter, 6Fr/110 cm, St. Jude Medical, Glostrup, Denmark) was introduced through an introducer sheath in the right jugular vein and advanced into the right atrium for recording of intra-atrial electrograms (aEGMs). In addition, a screw-in fixative electrode (Capsurefix Novus MRI Surescan 507685 cm, Medtronic Inc, Minneapolis, MN) was introduced through an additional introducer sheath and positioned in the right atrium for atrial pacing, guided by thorax fluoroscopy. Both electrodes were positioned so that aEGMs appeared in close association with the beginning of the P wave on the 2 simultaneously recorded surface electrocardiograms (S-ECG1 þ S-ECG2), and pacing at 60 bpm resulted in consistent atrial capture. ECG electrodes (Kruuse, Langeskov, Denmark) were positioned as shown in Online Supplemental Figure A1. S-ECG1 was optimized to express atrial activity, whereas S-ECG2 was a modified base–apex lead focusing on ventricular activity. Both aEGMs and surface ECGs were monitored during the experiments and stored for later analysis. An additional surface ECG (S-ECG3), identical to S-ECG2, was recorded using the Televet system (Kruuse). Atrial effective refractory period A pulse width of 2 ms was used throughout the experiment. Measurements of atrial effective refractory period (aERP) were conducted using twice the baseline atrial pacing threshold values. The aERP was determined before and after

827 NS8593 treatment, at atrial pacing rates of 60, 75, 120, and 182 bpm, respectively. The aERP was measured with 10 basic stimuli (S1) followed by an extrastimulus (S2) applied in 10-ms increments, and was defined as the longest S1–S2 interval that failed to elicit atrial capture. Every S1–S2 interval was applied multiple times (5–10 repetitions) at each pacing rate, and if intermittent atrial capture occurred, the longest S1–S2 interval with r50% captures was defined as the aERP. AF duration, AF vulnerability, and NS8593 treatment AF was induced by atrial tachypacing. A schematic outline of the AF induction protocol is shown in Figure 1. The duration of each tachypacing period was 2–4 seconds. If the duration of an induced AF episode was Z5 minutes at any given induction setting, tachypacing was repeated 5 times. If AF episodes were o5 minutes, 10–15 inductions were completed and the mean AF duration was calculated. Comparisons of AF duration (baseline vs posttreatment) were based on results obtained at the baseline induction setting capable of inducing a mean AF duration 41 minute. If the mean AF duration at a particular induction setting did not exceed 1 minute, the stimulation frequency was stepwise increased, and at the final step (burst pacing at 3000 bpm, 50 Hz) the current was subsequently increased (Figure 1). Each horse was ascribed an AF vulnerability score before and after treatment, based on the induction setting capable of inducing a single AF episode 41 minute in duration. The definition of vulnerability scores is presented in Online Supplemental Table A2. When a solitary AF episode Z15 minutes in duration occurred, AF was terminated by IV infusion of 5 mg/kg NS8593 (Acesion Pharma, Copenhagen, Denmark) over 10 minutes (vehicle; kleptose HP parenteral grade, Roquette, Lillerød, Denmark). If no AF episodes Z15 minutes occurred and spontaneous termination occurred repeatedly, horses were treated with NS8593 while in sinus rhythm (SR) (n ¼ 1). NS8593 is a prototype compound that is able to cross the blood–brain barrier and penetrate into the central nervous system. Inhibition of neuronal SK channels may induce tremors.22 Thus, because of ethical concerns, horses were anesthetized, and, if tremors were induced, the horses were treated with diazepam IV (Actavis, Gentofte, Denmark). AF inductions were repeated after NS8593 treatment, and the drug-induced effects on AF duration and AF vulnerability were assessed. Blood samples were collected at baseline and at specific time points after NS8593 administration to determine plasma concentrations and the pharmacokinetic profile of the compound in horses (see Online Supplemental Appendix). Horses were euthanized at the end of the procedure. ECG analysis QRS duration and QT interval were manually analyzed on SECG2. In addition, HR was calculated in the sections chosen for EGC analysis (see Online Supplemental Appendix).

828

Heart Rhythm, Vol 12, No 4, April 2015

Figure 1 Flow diagram showing the atrial fibrillation (AF) induction protocol. For each horse, AF duration and AF vulnerability were measured at the induction setting capable of inducing AF episodes 41 minute in duration (see text for details). The number of horses entering AF Z15 minutes at that particular induction setting before and after treatment with NS8593 is represented by npre and npost, respectively. aERP ¼ atrial effective refractory period; CV ¼ cardioversion; IV ¼ intravenous; SR ¼ sinus rhythm.

Data analysis All data are presented as mean ⫾ SEM. Analysis was formed using GraphPad Prism 5 software (GraphPad Software, San Diego, CA, USA), and P r .05 was considered significant. Differences in quantitative RT-PCR expression levels between RAA and RV (Figure 3A) were evaluated by Mann-Whitney U tests. Regional differences (Figure 3B) in SK1, SK2, and SK3 expression were analyzed by 1-way Kruskal-Wallis analysis of variance (ANOVA) test, followed by the Dunn multiple comparisons test. Atrial pacing stimulation threshold, AF duration, and AF vulnerability were analyzed by the Wilcoxon matched-pairs t test. aERP data were analyzed by 2way repeated-measures ANOVA, followed by the Bonferroni post hoc test for pairwise comparisons. Changes in QRS, QTc and HR were analyzed by 1-way repeated ANOVA, followed by the Dunnet post hoc test for multiple comparisons using time point T-1 as reference. Nonparametric tests were chosen whenever a gaussian distribution was not present or could not be evaluated because the sample size was too small.

Results Gene expression of major cardiac ion channels To obtain an overview of the expression and regional distribution of cardiac ion channel mRNA transcripts in horses, including SK channels, quantitative RT-PCR was performed on atrial and ventricular tissue from slaughtered horses in SR (Figure 2 and Online Supplemental Table A3). The expression level of the genes responsible for depolarizing currents SCNA5 (Nav1.5/INa) and CACNA1C (Cav1.2/ ICa,L) were similar between RAA and RV. When evaluating the genes encoding for repolarizing currents, KCNH2 (ERG1/IKr) and KCNJ2 (Kir2.1/IK1) were significantly

higher expressed in RV compared to RAA (P o .01), whereas KCNJ3 (Kir3.1/IKAch) and KCNJ5 (Kir3.4/GIRK4/ IKAch) expression were predominant in RAA (P o .05 and P o .01, respectively). The remaining genes—KCNQ1 (Kv7.1/IKs), KCND3 (Kv4.3/Ito), KCNIP2 (KChiP2/Ito), KCNA5 (Kv1.5/IKur), KCNN1 (SK1/ISK1), KCNN2 (SK2/ ISK2), and KCNN3 (SK3/ISK3)—each had equal relative expression in RAA and RV (Figure 2A). Additionally, no significant difference was observed in the expression level of the individual SK isoforms between the different cardiac regions (Figure 2B). SK3 had the highest expression level throughout the heart (Figure 2B). SK2 showed low expression levels in general; however, a noticeable but not significant predominance was observed in the left atrial appendage (Figure 2B).

Immunohistochemistry and confocal microscopy SK2 protein distribution was studied by immunohistochemistry and confocal microscopy. Staining revealed a widespread distribution in both RAA and RV cardiomyocytes (see Online Supplemental Figure A2). As previously demonstrated in mouse and human cardiomyocytes,6,10 SK2 channels were primary located in the periphery of the cells but were also found in a striated linear fluorescence pattern at the z-lines, suggesting that SK2 may be associated with the tubular network. Quantificational comparison of SK2 protein between atrial and ventricular cells could not be conducted because the settings used for analysis changed between cardiac locations.

In vivo results In order to investigate the in vivo anti-AF effects of NS8593, an electrical stimulation protocol was performed as indicated

Haugaard et al

SK Channel Inhibition and AF in Horses

829

Figure 2 Reverse transcription polymerase chain reaction results. A: Normalized expression levels of Naþ, Ca2þ, and Kþ channel genes in equine right atrial appendage (RAA) and right ventricle (RV), respectively. B: Regional relative distribution of SK1, SK2, and SK3 in RAA, left atrial appendage (LAA), RV, left ventricular endocardium (LV Endo), left ventricular mid-myocardium (LV Mid), and left ventricular epicardium (LV Epi). N ¼ 6 samples per region from 6 healthy equine hearts in sinus rhythm.

in Figure 3, which is based on representative aEGM and ECG recordings. Electrophysiologic findings: Atrial pacing threshold and aERP To study the role of SK channels on equine cardiac electrophysiology, the relatively selective SK channel inhibitor NS8593 was administered IV (5 mg/kg over 12.6 ⫾ 1.0 minutes) in anesthetized animals. All horses experienced drug-induced tremors to some extent, as also seen in other animal models (unpublished data from our group and Lallement et al22). All tremor episodes were initiated by intermittent bilateral nystagmus and evolved to include the musculature of the head, neck, and front legs in most animals. Tremor episodes were successfully controlled by

low-dose diazepam (0.04 ⫾ 0.02 mg/kg IV) in 4 horses, whereas tremors were mild and treatment was unnecessary in the remaining 3 horses. All 7 horses were momentarily treated with dobutamine IV during the procedure to ensure stable arterial blood pressures throughout the procedure. To assess the electrophysiologic effects of SK inhibition, the atrial pacing threshold and aERP were measured before and after NS8593 treatment. NS8593 did not change the atrial sensitivity toward electrical stimulations, whereas inhibition of SK channels resulted in pronounced prolongation of aERP at all measured atrial stimulation rates (Figure 4). To rule out a possible effect of anesthesia on aERP, animal-matched results from experiments in nonsedated standing animals20 were compared to results

830

Heart Rhythm, Vol 12, No 4, April 2015

Figure 3 Simplified figure showing the in vivo experimental protocol based on representative atrial electrogram (aEGM) and surface ECG (s-ECG1, s-ECG2) recordings. Stim ¼ atrial stimulation panel. To emphasize that the same protocol was used before and after NS8593, identical sequences are presented in B and E, and in C and F. A: Stabilization period and atrial pacing threshold determination. B: Atrial effective refractory period (aERP) measurements at atrial pacing rates of 60, 75, 120, and 182 bpm. C: Atrial fibrillation (AF) duration and AF vulnerability measurements. AF was induced, and the cardioversion time at different induction settings (see Figure 1) was recorded. D: If AF Z15 minutes occurred, horses were treated with NS8593 (5 mg/kg IV) and cardioversion times recorded. E: Posttreatment reassessment of atrial threshold and aERP at identical atrial pacing rates as before treatment. F: Posttreatment repeated AF duration and AF vulnerability assessments conducted as described in C. Arrows in B and E indicate S2 stimuli associated with atrial capture. Arrows in C and F indicate tachypacing periods leading to AF and subsequent spontaneous cardioversion.

Figure 4 NS8593- and anesthesia-induced changes in atrial effective refractory period (aERP) in healthy Standardbred trotters (n ¼ 5–6) at atrial stimulation rates of 60, 75, 120, and 182 bpm, respectively. NS8593 vs baseline anesthetized: **P o .01, ***P o .001.

Haugaard et al

SK Channel Inhibition and AF in Horses

obtained in the present study and revealed that baseline aERP in the absence of SK channel inhibition was unaffected by anesthesia (Figure 4). Clinical findings: AF duration, AF vulnerability, and drug-induced cardioversion time To evaluate the effect of SK inhibition on AF duration and AF vulnerability, AF was repeatedly induced before and after NS8593 treatment using programmed electrical stimulation. In general, horses were highly sensitive to AF. Five of 7 horses developed S2-induced AF episodes, with a mean duration 41 minute during baseline aERP recordings. However, in the presence of NS8593 this was reduced to 2 of 7 horses. AF duration and AF vulnerability were found to be significantly reduced by NS8593 (Figure 5). Representative aEGM and ECG recordings showing induction at 50 Hz before and after NS8593 are shown in Figure 6. Mean AF duration of the latter AF episode before initiation of NS8593 infusion was 22.1 ⫾ 1.9 minutes, all of which were terminated (n ¼ 6) by NS8593. The termination time was 10 ⫾ 2.0 minutes (range 5–19 minutes) from infusion start.

Pharmacokinetic data on NS8593 To calculate the plasma concentration half-life (T½) for NS8593 and to compare these with the pharmacologic effects observed, blood samples were taken between 5 and 90 minutes after the full dose of NS8593 was infused (Figure 7). After sample analysis (CiToxLAB Scantox, Lille Skensved Denmark) T½ of NS8593 was found to be 12 minutes, and the maximal plasma concentration was approximately 5000 ng/mL, corresponding to a total molar plasma concentration of 20 mM. The effect of the compound is exerted by the fraction not bound to plasma proteins. The amount of free unbound NS8593 was determined to be approximately 9% in equine plasma (see Online Supplemental Table A4), corresponding to a free plasma concentration of approximately 2 mM.

Figure 5 NS8593-induced decreases in atrial fibrillation (AF) duration (A) and AF vulnerability (B) in healthy Standardbred trotters (n ¼ 7). *P o .05.

831 ECG analysis: NS8593-induced changes in QRS duration, QTc interval, and HR In order to address the effects on ventricular conduction and repolarization, NS8593-induced effects on QRS duration and QT interval were assessed. Ventricular conduction and repolarization were unaffected by NS8593, as neither QRS nor QTc was significantly changed. HR also was unaffected by treatment (Figure 8).

Discussion SK channels have been shown to contribute to atrial repolarization,1,6,10 and inhibition of these channels demonstrates antiarrhythmic potential in multiple species.1–4 This study provides evidence that SK channels are also present and contribute to atrial repolarization in the equine heart. Treatment with NS8593 was associated with prolonged aERP, decreased AF duration, and AF vulnerability, thereby demonstrating clear antiarrhythmic properties in an in vivo horse model of acute pacing-induced AF of short duration.

SK channel expression and regional heterogeneity in the healthy equine heart in SR In 2003, Xu et al10 were the first to demonstrate the functional significance of SK2 channels in the heart and reported that the presence of ISK in mouse and human cardiomyocytes played a role in atrial action potential morphology. In addition, they identified a higher expression of SK2 mRNA in atria compared to ventricles in both humans and mice, thereby suggesting a possible novel atrial-selective pharmacologic target, which was subsequently supported by others.10,11,23 Tuteja et al11 reported both SK1 and SK2 had higher mRNA expression levels in mouse atria compared to ventricles and that SK3 had the lowest level of expression and showed similar expression in both atria and ventricles. Skibsbye et al6 reported SK2 to be the isoform showing the highest mRNA expression in human RAA. In canine left atrium, Qi et al1 recently reported low SK2 mRNA and protein expression compared to SK1 and SK3. In the present study, the general expression of SK2 was lower than those of SK1 and SK3, and none of the isoforms was more highly expressed in atria than in ventricle. SK3 was found to have the highest expression level of all 3 SK isoforms throughout the heart. These somewhat conflicting reports suggest that mRNA distribution of SK channel subtypes is dependent on species and cardiac region. At the protein level, data from the present study solely document the presence of SK2 in both RAA and RV of the equine heart and do not allow for quantitative comparison of expression between cardiac locations. It could be argued that staining for SK3, being the isoform with the highest expression level, would be more appropriate. However, the specificity of the available SK3 antibody was low (data not shown), as also seen in human tissue (unpublished data6). In addition, because cardiac SK isoforms form both homotetramers and

832

Figure 6 a horse.

Heart Rhythm, Vol 12, No 4, April 2015

Representative atrial electrogram (aEGM) and ECG recordings showing atrial fibrillation (AF) induction at 50 Hz before and after NS8593 in

heterotetramers and have been shown to overlap in their atrial distribution in mice,24 it is likely that all 3 SK isoforms have overlapping distribution in the equine heart.

Figure 7 Plasma concentration of NS8593 after intravenous infusion of 5 mg/kg over 12.6 ⫾ 1.0 minutes. Plasma samples were repeatedly collected and analyzed between 5 and 90 minutes after the end of IV infusion. The plasma half-life of NS8593 was 12 minutes, and the maximal plasma concentration (Cmax) was approximately 5000 ng/mL.

In vivo electrophysiologic findings The present study provides in vivo evidence for a role of SK channels in atrial repolarization and shows that NS8593 reduces AF duration and AF vulnerability and terminates acute pacing-induced AF of short duration in horses. The study also demonstrates that horses are reliable models to study potential antiarrhythmic drugs and that despite the size of the animal and larger requirements for compound amounts, they could constitute a useful supplement to large-animal models such as dogs, goats, and pigs. Our results are in agreement with those reported by others and support the current evidence that SK channels play a role in atrial electrophysiology and AF pathophysiology. Most recently, Qi et al1 reported substantial NS8593-induced aERP prolongations and suppressed AF vulnerability in an in vivo dog model of pacing-induced AF. In vivo open chest and burst pacing-induced AF models in rats found NS8593 to reduce AF duration,2,4 prolong aERP, and terminate AF of 15 minutes’ duration.2 In addition, NS8593 induced aERP prolongations and decreased AF duration in an in vivo model of aged rats with hypertensive-induced atrial remodeling.3 In the present study, treatment with NS8593 terminated

Haugaard et al

SK Channel Inhibition and AF in Horses

833

175

680

170

660 640 620

160

Qtc (ms)

QRS (ms)

165

155

600 580

150

560

145

540 520

140 -6

-4

-2

0

-6

-4

-2

0

2

4

6

8

10 15 20 25 30 35 40 45 50 55 60 65 Time (min)

-6

-4

-2

0

2

4

6

8

10 15 20 25 30 35 40 45 50 55 60 65 Time (min)

120

HR (bpm)

100

80

60

40

20 2

4

6

8

10 15 20 25 30 35 40 45 50 55 60 65 Time (min)

Figure 8 QRS duration (A), QTc interval (B), and heart rate (HR) (C) between 5 minutes before and 60 minutes after infusion of NS8593 (5 mg/kg IV). Time points (minutes) T-5, T1, T2, T4, T6, T8, T10, T15, T20, T25, T30, T45, and T60 were compared to the reference time point T-1 (just before infusion). NS8593 did not induce any significant changes in QRS (P ¼ .76), QTc (P ¼ .97), or HR (P ¼ .24) in healthy Standardbred trotters (n ¼ 6, 1 horse was excluded as an outlier because of extraordinary high HRs that deviated more than 3 the SEM value from the total mean at all but 1 time point).

pacing-induced AF episodes (Z15 minutes in duration) and promoted extensive increases in aERP at all measured cycle lengths. These findings suggest a functional correlation between prolonged aERP and the antiarrhythmic properties of NS8593 in horses. It is well recognized that the mechanisms of AF are multifactorial, most likely involving properties of reentry, enhanced automaticity, and/or triggered activity. Even though unexpected proarrhythmic effects of SK channel blockage-induced prolongation of the atrial APD have been reported in SK2 knockout mice as well as in an optical mapping study using the isolated canine left atrium,25,26 prolongation of aERP is widely accepted as a relevant antiarrhythmic property.27,28 However, because of the selectivity profile of NS8593, it should be emphasized that NS8593-induced block of other atrial ion channels than SK channels, particularly Naþ channels, may contribute in part to the observed aERP prolongation and antiarrhythmic effects. At increasing concentrations, we have reported NS8593 to inhibit ICa,L, INa, IKr, ICa,T, IK,Ach, IKs and IK1, but with IC50 values 5–50 times higher than those reported for ISK6 (selectivity profile in Online Supplemental Table A5), whereas others did not find any NS8593-induced effect on a broad range of native Kþ (Ito, IK1, IKs, IKr), Naþ, and Ca2þ currents.1 Atrial ion channel inhibition by NS8593, as well as with another SK channel-blocking compound ICAGEN, has been shown to depolarize the resting membrane potential and to reduce action potential amplitude and maximum upstroke velocity (dV/dtmax).6 In addition, studies

have shown limited NS8593-induced prolongations in atrial APD90 compared to the concomitant aERPprolongations,1,6,29 resulting in postrepolarization refractoriness. NS8593-induced resting membrane potential depolarization has been suggested to be the primary cause of the observed postrepolarization refractoriness through a mechanism of indirect Naþ channel block and hence a reduction in action potential amplitude and dV/dtmax.6 However, others suggest a direct NS8593-induced atrial selective Naþ channel block to be responsible for the observed postrepolarization refractoriness.29 Either way, it is most likely that, in addition to direct SK channel inhibition, the clear NS8593-induced antiAF effects observed in the present study is mediated in part by an indirect/direct Naþ channel block. For the compound to be effective during AF, it is essential that the induced prolongations of aERP occur during high atrial rates and not exclusively during low rates. In order to evaluate whether NS8593 was uniformly active at all measured pacing rates, we analyzed the data by 2-way ANOVA and confirmed a uniform effect of the compound at both low and high atrial pacing rates. Additionally, it is essential that antiarrhythmic compounds designed to elicit an effect in the atria do not have proarrhythmic effects in the ventricles. In the present study, although similar mRNA expression levels were found in atria and ventricles, QRS duration and QTc intervals were unaffected by NS8593 treatment. The lack of NS8593-induced changes in these parameters suggests no functional important changes in

834 ventricular conduction patterns and repolarization, as reported by others,1,4,6 suggesting neither block of ventricular SK channels nor Naþ channels have functional consequences. However, various cardiac diseases, such as acute myocardial infarction and heart failure, have been linked to a changed SK channel activation, and blocking of SK channels under these conditions has been reported to be both antiarrhythmic and proarrhythmic in the ventricles.14,30,31 How inhibition by NS8593 exerts its atrial-selective effects without ventricular adverse effects is uncertain, although a reduced functional role of SK channels and different Ca2þ dynamics in ventricles may play a role. Possibly, SK contribution is relatively more important in atria because other Kþ channels are less prone to act as a repolarization reserve in atria than in ventricles when SK channels are blocked. Anesthesia is a limitation to the present study because it induces negative hemodynamic changes. A detailed discussion about premedication, anesthesia, and animal recumbency is provided in the Online Supplemental Appendix. In brief, to stabilize mean arterial blood pressure, all horses were intermittently treated with dobutamine IV, which directly affects the contractility of the heart and increases blood pressure in a dose-dependent manner.32 In horses, dobutamine only increases HR at high infusion rates.32 The electrophysiologic effects of dobutamine have not been described in horses, but other species have shown increased sinus node automaticity, decreased aERP and ventricular effective refractory period (ERP), and decreased conduction time through the atrioventricular node in healthy hearts.33 These effects may be characterized as proarrhythmic, potentially inducing up-regulated SK channel activity (due to increased [Ca2þ]i and a possible increase in HR) and therefore an increased effect of NS8593 on aERP in the present study. However, when comparing baseline aERP values obtained during the present study with animal-matched baseline aERP values obtained in standing unsedated horses, no significant differences were found. In addition to dobutamine, 4 of 7 horses were treated with diazepam IV during the procedures. Diazepam is a benzodiazepine and elicits its primary effects in the CNS. It has been reported to demonstrate no significant changes in aERP in humans, even at higher doses than those used in the present study.34 Hence, taking the low dose of diazepam into account, the use of diazepam is not suspected to influence the results of the present study. In the present study we showed that the cardiac ion channel distribution of the most prominent sodium, calcium, and potassium channels highly resembles that of human hearts.35 Based on these observations and the fact that previous in vivo studies investigating the functional role of SK channels all were conducted in animals that are smaller in both body and heart size and have higher heart rates than humans, we find horses to be interesting large-animal models to study cardiac electrophysiology and effects of compounds with potential antiarrhythmic properties.

Heart Rhythm, Vol 12, No 4, April 2015

Conclusion SK channels are present and contribute significantly to atrial repolarization in horses, whereas ventricular conduction and repolarization are unaffected by NS8593 treatment. NS8593 terminates pacing-induced AF of short duration and decreases AF duration and AF vulnerability, thereby demonstrating clear antiarrhythmic properties in anesthetized healthy horses. The antiarrhythmic effects of NS8593 probably are caused both by SK channel blockage and by either direct or indirect Naþ channel block.

Acknowledgments In vivo and in vitro studies were conducted at The Large Animal Teaching Hospital, Department of Large Animal Sciences, and at the Department of Biomedical Sciences, Faculty of Health and Medical Sciences, University of Copenhagen, respectively. We gratefully acknowledge all staff members at The Large Animal Teaching Hospital, Department of Large Animal Sciences, University of Copenhagen, Denmark, involved in the study, especially Peter Urban, Stine Post, Bent Hansen, Henrik Kildeberg, Kristina Købsted, and Tina Olesen, for invaluable and professional assistance with monitoring anesthesia and guiding the fluoroscopy during the procedures.

Appendix Supplementary data Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.hrthm. 2014.12.028.

References 1. Qi XY, Diness JG, Brundel BJ, Zhou XB, Naud P, Wu CT, Huang H, Harada M, Aflaki M, Dobrev D, Grunnet M, Nattel S. Role of small-conductance calciumactivated potassium channels in atrial electrophysiology and fibrillation in the dog. Circulation 2014;129:430–440. 2. Skibsbye L, Diness JG, Sorensen US, Hansen RS, Grunnet M. The duration of pacing-induced atrial fibrillation is reduced in vivo by inhibition of small conductance Ca(2þ)-activated K(þ) channels. J Cardiovasc Pharmacol 2011;57: 672–681. 3. Diness JG, Skibsbye L, Jespersen T, Bartels ED, Sorensen US, Hansen RS, Grunnet M. Effects on atrial fibrillation in aged hypertensive rats by Ca(2þ)activated K(þ) channel inhibition. Hypertension 2011;57:1129–1135. 4. Diness JG, Sorensen US, Nissen JD, Al-Shahib B, Jespersen T, Grunnet M, Hansen RS. Inhibition of small-conductance Ca2þ-activated Kþ channels terminates and protects against atrial fibrillation. Circulation. Arrhythmia and electrophysiology 2010;3:380–390. 5. Xia XM, Fakler B, Rivard A, Wayman G, Johnson-Pais T, Keen JE, Ishii T, Hirschberg B, Bond CT, Lutsenko S, Maylie J, Adelman JP. Mechanism of calcium gating in small-conductance calcium-activated potassium channels. Nature 1998;395:503–507. 6. Skibsbye L, Poulet C, Diness JG, Bentzen BH, Yuan L, Kappert U, Matschke K, Wettwer E, Ravens U, Grunnet M, Christ T, Jespersen T. Small-conductance calcium-activated potassium (SK) channels contribute to action potential repolarization in human atria. Cardiovasc Res 2014;103:156–167. 7. Voigt N, Heijman J, Wang Q, Chiang DY, Li N, Karck M, Wehrens XH, Nattel S, Dobrev D. Cellular and molecular mechanisms of atrial arrhythmogenesis in patients with paroxysmal atrial fibrillation. Circulation 2014;129:145–156. 8. Ellinor PT, Lunetta KL, Glazer NL, Pfeufer A, Alonso A, Chung MK, Sinner MF, de Bakker PI, Mueller M, Lubitz SA, Fox E, Darbar D, Smith NL, Smith JD, Schnabel RB, Soliman EZ, Rice KM, Van Wagoner DR, Beckmann BM, van Noord C, Wang K, Ehret GB, Rotter JI, Hazen SL, Steinbeck G, Smith AV, Launer LJ, Harris TB, Makino S, Nelis M, Milan DJ, Perz S, Esko T, Kottgen A,

Haugaard et al

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

SK Channel Inhibition and AF in Horses

Moebus K, Newton-Cheh C, Li M, Mohlenkamp S, Wang TJ, Kao WH, Vasan RS, Nothen MM, MacRae CA, Stricker BH, Hofman A, Uitterlinden AG, Levy D, Boerwinkle E, Metspalu A, Topol EJ, Chakravarti A, Gudnason V, Psaty BM, Roden DM, Meitinger T, Wichmann HE, Witteman JC, Barnard J, Arking DE, Benjamin EJ, Heckbert SR, Kaab S. Common variants in kcnn3 are associated with lone atrial fibrillation. Nat Genet 2010;42:240–244. Zhang XD, Timofeyev V, Li N, Myers RE, Zhang DM, Singapuri A, Lau VC, Bond CT, Adelman J, Lieu DK, Chiamvimonvat N. Critical roles of a small conductance Ca(2)(þ)-activated K(þ) channel (SK3) in the repolarization process of atrial myocytes. Cardiovasc Res 2014;101:317–325. Xu Y, Tuteja D, Zhang Z, Xu D, Zhang Y, Rodriguez J, Nie L, Tuxson HR, Young JN, Glatter KA, Vazquez AE, Yamoah EN, Chiamvimonvat N. Molecular identification and functional roles of a Ca(2þ)-activated Kþ channel in human and mouse hearts. J Biol Chem 2003;278:49085–49094. Tuteja D, Xu D, Timofeyev V, Lu L, Sharma D, Zhang Z, Xu Y, Nie L, Vazquez AE, Young JN, Glatter KA, Chiamvimonvat N. Differential expression of smallconductance Ca2þ-activated Kþ channels SK1, SK2, and SK3 in mouse atrial and ventricular myocytes. Am J Physiol Heart Circ Physiol 2005;289:H2714–H2723. Nagy N, Szuts V, Horvath Z, Seprenyi G, Farkas AS, Acsai K, Prorok J, Bitay M, Kun A, Pataricza J, Papp JG, Nanasi PP, Varro A, Toth A. Does smallconductance calcium-activated potassium channel contribute to cardiac repolarization? J Mol Cell Cardiol 2009;47:656–663. Chang PC, Hsieh YC, Hsueh CH, Weiss JN, Lin SF, Chen PS. 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. Gui L, Bao Z, Jia Y, Qin X, Cheng ZJ, Zhu J, Chen QH. Ventricular tachyarrhythmias in rats with acute myocardial infarction involves activation of small-conductance Ca2þ-activated Kþ channels. Am J Physiol Heart Circ Physiol 2013;304:H118–H130. Reef VB, Bonagura J, Buhl R, McGurrin MK, Schwarzwald CC, van Loon G, Young LE. Recommendations for management of equine athletes with cardiovascular abnormalities. J Vet Intern Med 2014;28:749–612. Strobaek D, Hougaard C, Johansen TH, Sorensen US, Nielsen EO, Nielsen KS, Taylor RD, Pedarzani P, Christophersen P. Inhibitory gating modulation of small conductance Ca2þ-activated Kþ channels by the synthetic compound (r)-n(benzimidazol-2-yl)-1,2,3,4-tetrahydro-1-naphtylamine (NS8593) reduces afterhyperpolarizing current in hippocampal ca1 neurons. Mol Pharmacol 2006;70: 1771–1782. Sorensen US, Strobaek D, Christophersen P, Hougaard C, Jensen ML, Nielsen EO, Peters D, Teuber L. Synthesis and structure-activity relationship studies of 2(n-substituted)-aminobenzimidazoles as potent negative gating modulators ofsmall conductance Ca2þ-activated Kþ channels. J Med Chem 2008;51: 7625–7634. Soltysinska E, Olesen SP, Christ T, Wettwer E, Varro A, Grunnet M, Jespersen T. Transmural expression of ion channels and transporters in human nondiseased and end-stage failing hearts. Pflugers Arch 2009;459:11–23. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using realtime quantitative pcr and the 2(-delta delta c(t)) method. Methods 2001;25: 402–408. Haugaard MM, Pehrson S, Carstensen H, Flethøj M, Hesselkilde E, Præstegaard K, Diness JG, Grunnet M, Jespersen T, Buhl R. Antiarrhythmic and electrophysiological effects of flecainide on acutely induced atrial fibrillation in healthy horses. J Vet Intern Med 2014. [Epub ahead of print].

835 21. Buhl R, Ersboll AK, Larsen NH, Eriksen L, Koch J. The effects of detomidine, romifidine or acepromazine on echocardiographic measurements and cardiac function in normal horses. Vet Anaesth Analg 2007;34:1–8. 22. Lallement G, Fosbraey P, Baille-Le-Crom V, Tattersall JE, Blanchet G, Wetherell JR, Rice P, Passingham SL, Sentenac-Roumanou H. Compared toxicity of the potassium channel blockers, apamin and dendrotoxin. Toxicology 1995;104:47–52. 23. Grunnet M, Bentzen B, Sørensen U, Diness J. Cardiac ion channels and mechanisms for protection against atrial fibrillation. In: Nilius B, Amara SG, Gudermann T, Jahn R, Lill R, Offermanns S, Petersen OH, eds. Reviews of Physiology, Biochemistry and Pharmacology. Berlin, Heidelberg: Springer; 2012:1–58. 24. Tuteja D, Rafizadeh S, Timofeyev V, Wang S, Zhang Z, Li N, Mateo RK, Singapuri A, Young JN, Knowlton AA, Chiamvimonvat N. Cardiac small conductance Ca2þ-activated Kþ channel subunits form heteromultimers via the coiled-coil domains in the c termini of the channels. Circ Res 2010;107: 851–859. 25. Li N, Timofeyev V, Tuteja D, Xu DY, Lu L, Zhang Q, Zhang Z, Singapuri A, Albert TR, Rajagopal AV, Bond CT, Periasamy M, Adelman J, Chiamvimonvat N. Ablation of a Ca2þ-activated Kþ channel (SK2 channel) results in action potential prolongation in atrial myocytes and atrial fibrillation. J Physiol Lond 2009;587:1087–1100. 26. Hsueh C-H, Chang P-C, Hsieh Y-C, Reher T, Chen P-S, Lin S-F. Proarrhythmic effect of blocking the small conductance calcium activated potassium channel in isolated canine left atrium. Heart Rhythm 2013;10:891–898. 27. Nattel S. New ideas about atrial fibrillation 50 years on. Nature 2002;415: 219–226. 28. Nattel S. Calcium-activated potassium current: a novel ion channel candidate in atrial fibrillation. J Physiol 2009;587:1385–1386. 29. Burashnikov A, Dang S, Grunnet M, Antzelevitch C. Small conductance calcium-activated potassium channel (SK) inhibitor NS8593 prevents induction of atrial fibrillation by producing atrial-selective prolongation of effective refractory period secondary to atrial-selective sodium channel block (abstract). Heart Rhythm 2012;9:S256. 30. Chua SK, Chang PC, Maruyama M, Turker I, Shinohara T, Shen MJ, Chen Z, Shen C, Rubart-von der Lohe M, Lopshire JC, Ogawa M, Weiss JN, Lin SF, Ai T, Chen PS. Small-conductance calcium-activated potassium channel and recurrent ventricular fibrillation in failing rabbit ventricles. Circ Res 2011;108: 971–979. 31. Hsieh YC, Chang PC, Hsueh CH, Lee YS, Shen C, Weiss JN, Chen Z, Ai T, Lin SF, Chen PS. Apamin-sensitive potassium current modulates action potential duration restitution and arrhythmogenesis of failing rabbit ventricles. Circ Arrhythmia Electrophysiol 2013;6:410–418. 32. Vitale V, Sgorbini M, Briganti A, Corazza M, Breghi G, Staffieri F. Evaluation of echocardiographic parameters during increasing infusion rates of dobutamine in isoflurane-anesthetized horses. J Equine Vet Sci 2013;33:1110–1114. 33. Tisdale JE, Patel R, Webb CR, Borzak S, Zarowitz BJ. Electrophysiologic and proarrhythmic effects of intravenous inotropic agents. Prog Cardiovasc Dis 1995;38:167–180. 34. Kumagai K, Yamanouchi Y, Matsuo K, Tashiro N, Hiroki T, Arakawa K. Antiarrhythmic and proarrhythmic properties of diazepam demonstrated by electrophysiological study in humans. Clin Cardiol 1991;14:397–401. 35. Gaborit N, Le Bouter S, Szuts V, Varro A, Escande D, Nattel S, Demolombe S. Regional and tissue specific transcript signatures of ion channel genes in the nondiseased human heart. J Physiol 2007;582:675–693.