lable at ScienceDirect

Progress in Biophysics and Molecular Biology 98 (2008) 198–207

Contents lists avai

Progress in Biophysics and Molecular Biology

journal homepage: www.elsevier .com/locate/pbiomolbio

Original Research

Sinus node dysfunction in ATX-II-induced in-vitro murine model of long QT3syndrome and rescue effect of ranolazine

Jingjing Wu a,b, Longxian Cheng a, Wim J. Lammers c, Lin Wu d, Xin Wang e, John C. Shryock d,Luiz Belardinelli d, Ming Lei a,b,*

a Centre for Ion Channel Research and Department of Cardiovascular Diseases, Xiehe Hospital, Huazhong University of Sciences and Technology, Wuhan, Chinab Cardiovascular Group, School of Clinical and Laboratory Sciences, The University of Manchester, Grafton Street, Manchester, M13 9NT, UKc Department of Physiology, Faculty of Medicine and Health Sciences, UAE University, P.O. Box 17666, Al Ain, United Arab Emiratesd Pharmacological Sciences, CV Therapeutics, Inc., Palo Alto, CA, USAe Section II, Faculty of Life Sciences, The University of Manchester, Grafton Street, Manchester, M13 9NT, UK

a r t i c l e i n f o

Article history:Available online 25 January 2009

Keywords:ATX-IISCN5ASinus node dysfunctionLate Na currentINa,L

LQT3

Abbreviations: TdP, torsades de pointes; VT, ventanemone toxin; A-V, atrioventricular; MAPD, monophnode recovery time.

* Correspondence to: Ming Lei, Cardiovascular ResTel.: þ44 161 2751194; fax: þ44 161 2751183.

E-mail address: ming.lei@manchester.ac.uk (M. Le

0079-6107/$ – see front matter � 2009 Elsevier Ltd.doi:10.1016/j.pbiomolbio.2009.01.003

a b s t r a c t

The aim of this study was to characterize the role of the late Naþ current (INa,L) as a mechanism forinduction of both tachy and bradyarrhythmias in murine heart and sino-atrial node tissue. The seaanemone toxin ATX-II and ranolazine were used to increase and inhibit, respectively, INa,L. In sixteenhearts studied, exposure to 1–10 nM ATX-II caused a slowing of intrinsic heart rate and prolongations ofthe P–R and QT intervals, the duration of the monophasic action potential, and the sinus node recoverytime, accompanied by frequent occurrences of early afterdepolarisations, delayed afterdepolarisationsand rapid, repetitive ventricular tachy and sino-atrial bradyarrhythmias. ATX-II also slowed sinus nodepacemaking, and induced bradycardic arrhythmias in isolated sino-atrial preparations (n¼ 5). TheATX-II-induced alteration of electrophysiological properties and occurrence of arrhythmic events weresignificantly attenuated by 10 mM ranolazine in intact hearts (n¼ 11) and isolated sino-atrial preparations(n¼ 5). In conclusion, the INa,L enhancer ATX-II causes both tachy and bradyarrhythmias in the murineheart, and these arrhythmias are markedly attenuated by the INa,L blocker, ranolazine (10 mM). The resultssuggest that INa,L blockade may be the mechanism underlying the reductions of both brady and tachy-arrhythmias by ranolazine that were observed during the MERLIN-TIMI clinical outcomes trial.

� 2009 Elsevier Ltd. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1992. Materials and methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .199

2.1. Langendorff perfused hearts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1992.2. Monophasic action potential (MAP) and ECG measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1992.3. Programmed electrical stimulation (PES) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1992.4. Mapping electrical propagation in isolated sino-atrial preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2002.5. Determination of arrhythmic events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2012.6. Statistical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2012.7. Sources of drugs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201

3. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2013.1. Electrophysiological properties and arrhythmic events in ATX-II treated hearts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201

ricular tachycardia; APD, action potential duration; EAD, early afterdepolarisation; LQT3, long QT syndrome 3; ATX,asic action potential duration; MAP, monophasic action potential; PES, programmed electrical stimulation; SNRT, sinus

earch Group, School of Clinical and Laboratory Sciences, The University of Manchester, Manchester, M13 9NT, UK.

i).

All rights reserved.

J. Wu et al. / Progress in Biophysics and Molecular Biology 98 (2008) 198–207 199

3.2. Effect of ranolazine on electrophysiological properties and arrhythmic events in ATX-II treated hearts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2023.2.1. Attenuation by ranolazine of the effect of ATX-II on MAPD and QT interval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2023.2.2. Attenuation by ranolazine of ATX-II-induced EADs and VT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2023.2.3. Attenuation by ranolazine of ATX-II-induced sinus bradycardia and atrioventricular (AV) conduction block . . . . . . . . . . . . . . . . . . . . . . . 202

3.3. Mapping of electrical propagations in sino-atrial preparations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2024. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205

4.1. Summary of the key findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2054.2. The mechanism of ATX-II-induced sinus node dysfunction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2054.3. Attenuation by ranolazine of ATX-II-induced alteration of electrophysiological properties and occurrence of arrhythmias . . . . . . . . . . . . . . . . . . 206Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207

1. Introduction

Inherited mutations in SCN5A, the gene encoding the pore-forming subunit of the cardiac Naþ channel isoform NaV 1.5, resultin a spectrum of disease entities termed Naþ channelopathies thatinclude multiple arrhythmic syndromes. After the first report byBenson and colleagues (2003) of loss-of-function SCN5A mutationsin patients with familial sinus bradycardia syndromes, severalother groups (Benson et al., 2003; Groenewegen et al., 2003;Veldkamp et al., 2003; Smits et al., 2005; Zhang et al., 2008) havelinked the loss-of-function SCN5A mutations to familial sick sinussyndrome (SSS). Fourteen loss-of-function SSS associated SCN5Amutations have been identified to date (Lei et al., 2007). The resultsof our study of mice with targeted genetic disruption of thischannel caused by a null mutation in the Naþ channel gene, SCN5A(Papadatos et al., 2002), have further demonstrated that loss offunction of cardiac Naþ channels produced a sinus bradycardia,slowed sino-atrial conduction, and caused sino-atrial exit block,thereby replicating the major features of clinically observed SSS inhuman subjects. The results suggested that loss of function ofSCN5A-encoded Naþ channels leads to sinus node dysfunction andaltered sinus node pacemaking (Lei et al., 2007).

Interestingly, sinus node dysfunction also occurs in patientswith gain-of-function mutations of SCN5A that are associated withthe LQT3 syndrome (Wang et al., 1997). Because QT prolongation ismost pronounced at lower heart rates, and it is virtually absent athigh heart rates, sinus bradycardia represents an important indirectfactor in predisposition to lethal arrhythmias in LQT3 patients.Furthermore, sinus node dysfunction has been observed in patientswith acquired heart conditions such as heart failure and cardiacischemia (Zicha et al., 2005).

Recently, the role of a slowly-inactivating component of voltage-gated Naþ current, INa,L, in controlling cardiac action potential (AP)repolarisation, and its importance in arrhythmogenesis havegarnered great attention (Noble and Noble, 2006). An abnormalincrease of INa,L produced by ‘‘late openings’’ of the Naþ channelscauses prolongation of AP repolarisation that may lead to repolar-isation failure (early afterdepolarisations, EADs) and Naþ-inducedCa2þ overloading that triggers delayed afterdepolarisations (DADs),calcium oscillations, and rapid tachyarrhythmias such as ventric-ular tachycardia (VT) or fibrillation (VF).

Reduction of INa,L would therefore be expected to have thera-peutic benefits. Ranolazine, a relatively selective inhibitor of INa,L,has been shown to be effective to reduce angina and the incidenceof non-sustained VT in patients with ischemic heart disease (Haleet al., 2008).

In this study, we investigated the effects of the Anemonia sul-cata toxin ATX-II, an enhancer of INa,L, and ranolazine on arrhyth-mogenesis in murine isolated working hearts and sino-atrialpreparations. As demonstrated previously (Isenberg and Ravens,1984), ATX-II prevents full inactivation of the inward Naþ current

and therefore mimics in an exaggerated manner the effects ofmutations of the cardiac Naþ channel gene SCN5A that are impli-cated as the mechanism for the LQT3 syndrome (Wang et al.,1997).

2. Materials and methods

2.1. Langendorff perfused hearts

Adult male C57 mice (8–10 wks) were purchased from CharlesRiver (UK). All animal procedures conformed to the UnitedKingdom Animals (Scientific Procedures) Act of 1986. Animals wereinjected intraperitoneally with heparin (0.5 unit/g) and anes-thetized with Avertin (0.24 mg/g), and killed by cervical dislocation.Hearts were quickly excised and cannulated for Langendorffretrograde perfusion at a rate of 3.5–4 ml/min with modifiedTyrode’s solution (in mmol/L: NaCl 120, NaHCO3 25.2, NaH2PO4 1.2,MgCl2 1.3, glucose 5, KCl 4.0, CaCl2 1.8, gassed with 95% O2/5% CO2)at 37 �C.

2.2. Monophasic action potential (MAP) and ECG measurements

A pair of ECG electrodes (Hugo Sachs, Harvard Apparatus, UK)was placed onto the epicardial surfaces of the atrium and leftventricle for ECG recording. An MAP electrode (Hugo Sachs, Har-vard Apparatus, UK) was placed against the epicardial surface of theapex of the left ventricle for epicardial MAP recording. The elec-trode was adjusted during the experiment both to maintain stableMAP signals and to prevent damage to the ventricular surface. ECGand MAPs were amplified, filtered (0.5 Hz–1 kHz band-pass; Neu-rolog System, NL900DDigtimer Ltd, Welwyn Garden City, UK), anddigitised at a sampling frequency of 5 kHz (micro1401, CambridgeElectronic Design, Cambridge, UK). Analysis of ECG and MAP signalswas performed using Spike 2 (Cambridge Electronic Design, Cam-bridge, UK).

For experiments to study the conduction system, heartsshowing a stable and regular sinus rhythm with one to one atrio-ventricular (A-V) conduction were used. ECG and ventricular epi-cardiac MAP were continuously monitored and recorded. The heartwas first perfused with control Tyrode’s solution for 10–15 minuntil stable baseline values for all parameters were recorded. Thepreparation was then exposed to ATX-II with increasing concen-trations of drug in a cumulative manner, allowing 7–15 minbetween changes of concentration.

2.3. Programmed electrical stimulation (PES)

Two chlorinated silver wire stimulation electrodes were con-nected to a Digitimer DS2A triggered stimulus isolation unit(Digitimer Ltd, Welwyn Garden City, UK) with variable voltage

Fig. 1. Concentration-dependent effect of ATX-II on MAP and ECG of murine heart. A: representative recordings of ECG (upper panel) and epicardial MAP (lower panel) of isolatedheart preparation in absence and presence of 1–5 nM ATX-II. B–D: summary of effects of ATX-II on HR, MAPD90, and QT interval.

J. Wu et al. / Progress in Biophysics and Molecular Biology 98 (2008) 198–207200

output. A stimulus amplitude of 1.5� diastolic capture threshold ofvoltage was used, with a stimulus duration of 1 ms. Sinus noderecovery time (SNRT) was measured after a 30-s pacing train ata basic cycle length of 100 ms and was defined as the intervalbetween the last stimulus in the pacing train and the onset of thefirst sinus return beat. Inducibility of ventricular arrhythmias wastested by using a programmed stimulation protocol comprising aneight-beat stimulus (S1) drive train at 8 Hz followed by an extra-stimulus (S2). The S1–S2 interval was reduced by 1 ms betweensuccessive drive trains until the preparation became refractory. Thesignals were amplified, filtered (band-pass filter 30 Hz–1 kHz)

(NeuroLog System, NL900DDigtimer Ltd, Welwyn Garden City, UK)and digitised using an analogue-to-digital converter (CED 1401plus,Cambridge Electronic Design, Cambridge, UK).

2.4. Mapping electrical propagation in isolated sino-atrialpreparation

Preparation. The sino-atrial tissues were prepared as describedpreviously (Lei et al., 2004). In brief, after dissection of the sinusnode and surrounding atrial muscle, the preparation (endocar-dial surface up) was placed in a tissue bath. The preparation was

Table 1Summary of ECG, MAP parameters and arrhythmic events in murine hearts in the conditions of control and treatment with ATX-II alone and ATX-II plus ranolazine.

n HRbeats/min

P–R interval (ms) QRS interval (ms) QT(ms)

QTc(ms)

APD90

(ms)SNRT(ms)

Control 16 334� 14 42.5� 1.5 20.0� 1.0 81.1� 3.6 56.9� 2.5 67.2� 5.1 306.2� 18.71 nM ATX-II 11 270� 22 44.6� 2.2 20.3� 1.5 104.0� 7.4 68.6� 3.0 70.6� 8.2 412.0� 70.32 nM ATX-II 11 226� 23 46.6� 2.1 20.4� 1.1 110.0� 9.7 69.7� 3.3 75.6� 7.2 425.7� 58.85 nM ATX-II 12 207� 31 49.4� 2.7 22.7� 1.2 111.0� 7.3 80.3� 5.3 81.2� 12.0 515.0� 37.910 nM ATX-II 9 183� 34 50.0� 3.4 22.8� 1.4 143.0� 3.0 83.6� 4.6 110.7� 20.0 661.0� 57.210 mM Ranþ 5 nM ATX-II 6 243� 28 44.5� 3.0 21.3� 1.4 96.0� 6.5 66.7� 3.9 72.6� 5.6 501.0� 24.1P-value

(one-way ANOVA)(control vs. 1, 2, 5, 10 nM ATX-II;5 nM ATX-II vs. 5 nM ATX-II þRan)

<0.05 <0.05 NS <0.05 <0.05 <0.05 <0.05

n EAD/DAD (Yes:No) Non-and sustainedVT (Yes:No)

Sinus bradycardia(�30% above baselinereduction of HR)

Sinus pause orarrest (Yes:No)

II0 or higherAVB (Yes:No)

Control 16 0:16 1:15 0:16 0:16 0:161 nM ATX-II 11 1:10 2:9 3:8 0:11 0:112 nM ATX-II 11 2:9 3:8 6:5 2:9 1:105 nM ATX-II 12 9:3 6:6 8:4 5:7 5:710 nM ATX-II 9 9:0 8:1 9:0 6:3 3:610 mM Ranþ 5 nM ATX-II 6 4:2 2:4 3:3 3:3 2:4

J. Wu et al. / Progress in Biophysics and Molecular Biology 98 (2008) 198–207 201

superfused with Tyrode’s solution at a temperature of37� 0.5 �C at a rate of 4 w 5 ml/min.Electrode array. The custom-made electrode array consists of 64separate electrodes (Teflon-coated silver wires; 0.125 mmdiameter; Science Products) in an 8� 8 configuration with aninterelectrode distance of 0.55 mm. The dimensions of the totalarray are approximately 4� 4 mm.Data acquisition. Unipolar electrical recordings were performedwith a large reference electrode located close to the array, butnot touching the tissue, acting as the indifferent pole The 64recording electrodes were connected through shielded wires totwo 32-channel amplifiers (SCXI-1102C, National InstrumentsCorporation (U.K.) Ltd, Newbury, UK). Sampling frequencies foreach signal was set at 1 kHz. The signals were continuouslystored on disk and displayed on screen using a custom-devel-oped program, written in Labview 7.0.Propagation maps. For the off-line analysis, signals were dis-played on screen in sets of 8–16 electrograms (Fig. 9). Theactivation time was determined as the point of maximal nega-tive slope and marked with a cursor. After marking all signifi-cant waveforms in all leads, the activation times were thendisplayed in a grid representing the layout of the originalrecording array. All activation times, in milliseconds, wererelated to the timing of the first detected waveform (electrode 3in control panel Fig. 9). Isochrones were drawn manually aroundareas activated in steps of 5 m.

2.5. Determination of arrhythmic events

Ventricular tachycardia (VT) was defined as a sequence of threeor more ventricular depolarisations at a rate of �400 beats/min. VTthat terminated spontaneously was defined as transient or non-sustained VT. VT that did not terminate spontaneously unlessinterrupted by a treatment (e.g., ranolazine) was defined as sus-tained VT. Sinus bradycardia was defined as a 30% reduction ofbaseline heart rate (HR) from the control condition. Sinus pausewas defined as a spontaneous interruption in the regular sinusrhythm, by a pause lasting for a period that is not an exact multipleof the sinus cycle. Sinus arrest was defined as cessation of sinusnode pacemaker activity; the ventricles may continue to beatunder ectopic atrial, atrioventricular junctional, or idioventricularcontrol.

2.6. Statistical analysis

All data are reported as means� S.E.M. Repeated measure one-way analysis of variance was used to compare values of measure-ments obtained from the same heart before and after treatment.When analysis of variance revealed the existence of a significantdifference among values, Tukey’s test was applied to determine thesignificance of a difference between selected group means. Ap-value <0.05 was taken as an upper limit to indicate a significantdifference.

2.7. Sources of drugs

Ranolazine [(�)-N-(2,6-dimethylphenyl)-(4[2-hydroxy-3-(2-methoxyphenoxy)propyl]-1-piperazine] was synthesized at CVTherapeutics, Inc (Palo Alto, California, USA). ATX-II was purchasedfrom Alomone Labs (Jerusalem, Israel).

3. Results

3.1. Electrophysiological properties and arrhythmic events in ATX-IItreated hearts

ATX-II (1–10 nM) caused concentration-dependent reductionsof HR, increases in SNRT, and prolongations of the QT interval andMAPD (Fig. 1). Intrinsic HR was decreased approximately by 22% at1 nM (n¼ 11), 30% at 2 nM (n¼ 11), 60% at 5 nM (n¼ 11), and 70% at10 nM (n¼ 9) ATX-II. The effects of ATX-II on MAP and ECGparameters in 12 hearts are summarized in Table 1 and Fig. 1B–D.

Both tachycardia-related (EADs, DADs, ectopic beats, non-sus-tained and sustained episodes of VT and VF) and bradycardia-related (sinus bradycardia, sinus pause, sinus arrest, and A-Vconduction block) arrhythmic events were markedly increased byATX-II (5 and 10 nM), compared to control (Figs. 2–8 and Table 1).EAD/DAD events occurred in two of eleven hearts treated with2 nM ATX-II, nine of twelve hearts treated with 5 nM ATX-II, nine ofnine hearts treated with 10 nM ATX, but none of the control hearts.A similar effect of ATX-II was observed with other tachycardia-related and bradycardia-related arrhythmic events (Table 1). Non-sustained VT was observed in hearts infused with 5 nM ATX-II inboth the absence and presence of programmed S1S2 electricalstimulation (PES; Fig. 2). In the presence of 10 nM ATX-II, EADs,

Fig. 2. Ventricular arrhythmias in a murine heart exposed to ATX-II. A: representativeECG and MAP recordings of non-sustained spontaneous VT in an isolated heartpreparation in absence and presence of 5 nM ATX-II. B: representative ECG and MAPrecordings of non-sustained VT induced by a programmed S1S2 stimulation protocol inan isolated heart preparation in the presence of 5 nM ATX-II.

J. Wu et al. / Progress in Biophysics and Molecular Biology 98 (2008) 198–207202

DADs, ectopic beats, and TdP events were observed (Fig. 3). Inter-estingly, the EAD/DADs and TdP often occurred in the presence ofsevere sinus bradycardia or A-V conduction block (Fig. 3C).

3.2. Effect of ranolazine on electrophysiological properties andarrhythmic events in ATX-II treated hearts

3.2.1. Attenuation by ranolazine of the effect of ATX-II on MAPD andQT interval

In eleven hearts examined, ranolazine significantly attenuatedthe prolongations of MAP (MAPD90) duration and QT intervalcaused by ATX-II. A representative recording from one of six heartstreated with 10 mM ranolazine in the presence of 5–10 nM ATX-II isshown in Fig. 4. The action of 5 nM ATX-II to lengthen MAPD90 and

QT was markedly reduced by ranolazine. QT and MAPD90 increasedby 45�16% and 55�18% in hearts treated with 5 nM ATX-II alone(n¼ 12), but only by 26� 5% and 19� 5% in hearts treated with5 nM ATX-II in the presence of 10 mM ranolazine (n¼ 6) (Table 1).Ranolazine (10 mM) also attenuated sinus bradycardia caused by5 nM ATX-II (Fig. 4A and Table 1).

3.2.2. Attenuation by ranolazine of ATX-II-induced EADs and VTRanolazine greatly reduced the occurrence of EADs, premature

ventricular beats, and VT caused by ATX-II. In twelve hearts,perfusion with 5–10 nM ATX-II for 10 min significantly prolongedthe MAPD90 and QT interval and caused frequent EADs and DADsand ventricular extrasystolic beats that were followed by episodesof transient and sustained polymorphic VT (Figs. 3–5 and Table 1).Administration of 10 mM ranolazine in the continued presence ofATX-II led to the significant suppression of these events andrestoration of a regular sinus rhythm at all concentrations of ATX-IIexamined (1–10 nM). Fig. 5 shows representative ECG recordingsfrom a heart exposed to 5 nM ATX-II and subsequent administra-tion of 10 mM ranolazine in the continued presence of 5 nM ATX-II.Hearts exposed to 5 nM ATX-II developed frequent ventricularextrasystolic beats followed by sustained VT. Ranolazine (10 mM)abolished VT in some hearts treated with ATX-II (Fig. 5). As shownin Table 1, VT occurred in 50% of hearts (six of twelve) treated with5 nM ATX-II alone, and in 33% hearts (two of six) treated with 5 nMATX-II in the presence of 10 mM ranolazine.

3.2.3. Attenuation by ranolazine of ATX-II-induced sinusbradycardia and atrioventricular (AV) conduction block

Ranolazine significantly attenuated the occurrence of sinusbradycardia and AV conduction block caused by ATX-II. Perfusion ofhearts with 2–10 nM ATX-II for 10 min led to a significant sinusbradycardia, including sinus pause or arrest (Table 1). Administra-tion of 10 mM ranolazine markedly suppressed ATX-II-inducedbradyarrhythmias. Representative recordings from one of twelvehearts treated with 5–10 nM ATX-II are shown in Fig. 6. Ranolazine(10 mM) significantly attenuated ATX-II-induced sinus pauses. Theeffect of ranolazine on ATX-II-induced sinus node dysfunction wasfurther investigated by determining the SNRT. Perfusion of heartswith ATX-II led to significant prolongation of SNRT in a concentra-tion-dependent manner (increased by 80�12% at 5 nM ATX-II,n¼ 12) (Table 1 and Fig. 7). Administration of 10 mM ranolazine inthe continued presence of ATX-II led to a suppression of ATX-II-induced prolongation of SNRT (from 661�57 ms at 5 nM ATX-IIalone to 501�24 ms in the presence of 10 mM ranolazine and 5 nMATX-II) (Fig. 7; n¼ 6; p< 0.05). A representative recording from oneof twelve hearts of the effect of ranolazine on A-V conduction blockcaused by ATX-II is illustrated in Fig. 8. Perfusion of hearts with2 nM ATX-II led to prolongation of the P–R interval and occasionalAV conduction block. The degree of AV conduction block wasgreater at the higher ATX-II concentrations (5 and 10 nM). In thepresence of ATX, there were 20� 6 non-conducted beats perminute (n¼ 12 hearts). Administration of 10 mM ranolazine in thecontinued presence of 5–10 nM ATX-II led to significant reductionin the number of non-conducted atrial beats to 4�1 per minute(n¼ 6) (Fig. 8).

3.3. Mapping of electrical propagations in sino-atrial preparations

Finally, the effect of ranolazine to suppress the action of ATX-IIon pacemaker activity and activation patterns of the sinus node wasstudied in isolated sino-atrial preparations. A custom-made 64electrode array was used to record extracellular potentials (ECP)from 64 sites simultaneously from the sinus node and surroundingatrial muscle (Fig. 9). Representative recordings of extracellular

Fig. 3. Examples of EADs, DADs, ectopic beats, and TdP events in murine hearts exposed to 10 nM ATX-II. A: MAP recoding shows frequent occurrences of EADs and DADs (arrows).B: corresponding ECG shows frequent occurrence of ectopic beats. C: VT occurred in the presence of severe sinus bradycardia.

Fig. 4. Representative recordings of ECGs (left) and MAPs (right) from a murine heart treated with ATX-II (1–5 nM) in the absence and presence of 10 mM ranolazine. A–D: ECGs andMAPs recordings in the presence of ATX-II. E: ECGs and MAPs recordings in the presence of ATX-II and ranolazine.

J. Wu et al. / Progress in Biophysics and Molecular Biology 98 (2008) 198–207 203

Fig. 5. Representative ECG recordings from a murine heart treated with ATX-II in the absence and presence of 10 mM ranolazine. A: ECG recording in the absence of drug (control).Panels B–D: following exposure to 5 nM ATX-II, the heart developed frequent ventricular extrasystolic beats followed by sustained VT. Panel E: 10 mM ranolazine abolished VT in thepresence of 5 nM ATX-II.

Fig. 6. Representative ECG recordings from a murine heart treated with 10 nM ATX-II in the absence and presence of 10 mM ranolazine. A: recording in the absence of drug (control).B: recording in the presence of 10 nM ATX-II. A long sinus pause occurred. C: ranolazine significantly shortened the length of the 10 nM ATX-II-induced sinus pauses.

J. Wu et al. / Progress in Biophysics and Molecular Biology 98 (2008) 198–207204

Fig. 7. Attenuation by ranolazine of ATX-II-induced prolongation of SNRT in murine heart. Panel A: SNRT recording in the absence of drug (control). Panels B–D: SNRT recording inthe presence of 1–5 nM ATX-II. Panel E: SNRT recording during exposure of heart to 10 mM ranolazine in the presence of 5 nM ATX-II.

J. Wu et al. / Progress in Biophysics and Molecular Biology 98 (2008) 198–207 205

potentials from different regions of the preparation and activationmaps during control, after perfusion of 5 nM ATX-II for 10 min andafter administration of 10 mM ranolazine in the continued presenceof ATX-II are shown in Fig. 9. In the absence of drug (control), theintrinsic rate of the leading pacemaker site close to the crista ter-minalis was 268� 5 beats/min (n¼ 5). Propagation occurred to theright atrium appendage, the interatrial septum, and the atrioven-tricular junction regions. After perfusion of ATX-II for 10 min, theintrinsic rate decreased markedly to 125� 5 beats/min (n¼ 5) andcycle length prolonged, indicating that pacemaker activity wasdepressed by ATX-II. The site of pacemaking shifted slightly andconduction times were slightly increased compared to control.Administration of 10 mM ranolazine in the continued presence of5 nM ATX-II led to a significant reversal of these effects of ATX-II(e.g., the average intrinsic rate increased to 257�6 beats/min;n¼ 5, p< 0.05 vs. ATX-II treatment alone group). Similar resultswere seen in five isolated sinus node preparations mapped 15–20times per preparation.

4. Discussion

4.1. Summary of the key findings

ATX-II (1–10 nM) caused a slowing of HR, prolongations of theP–R and QT intervals, MAPD, and SNRT with accompanying

frequent occurrences of EADs, DADs, and rapid, repetitive ventric-ular tachyarrhythmias and sino-atrial bradyarrhythmias. TheATX-II-induced sinus bradyarrhythmias were also observed in iso-lated sino-atrial preparations. The ATX-II-induced alteration ofelectrophysiological properties and increased frequency ofarrhythmic events were attenuated by ranolazine (10 mM), a rela-tively selective inhibitor of INa,L (Antzelevitch et al., 2004).

4.2. The mechanism of ATX-II-induced sinus node dysfunction

ATX-II is known to increase the slowly-inactivating componentof Naþ current, INa,L via increasing in slow component and gener-ating a persistent current fraction. Both actions lead to QT prolon-gation. An increase of INa,L is reported to facilitate the formation ofEADs and DADs and to trigger ventricular tachycardias in animalmodels (Noble and Noble, 2006). An ATX-II-induced increasing inrecovery from inactivation (Chahine et al., 1996; Richard Benzingeret al., 1999) which could be responsible for triggering cardiacevents.

The actions of ATX-II resemble those caused by gain-of-functionmutations in the gene SCN5A (Wang et al., 1995) that lead to thehuman LQT3 syndrome. Murine hearts exposed to ATX-II in thepresent study mimicked the principle electrophysiological features(i.e., a long QT interval and increased susceptibility to ventriculartachyarrhythmias) observed in LQT3 patients and in a mouse model

Fig. 8. Attenuation by ranolazine of A-V conduction block caused by ATX-II in murineheart. Panel A: ECG recording in the absence of drug (control). Panels B–C: ECGrecording during exposure of the heart to 2 and 10 nM ATX-II. Panel D: ECG recordingof the same heart exposed to 10 mM ranolazine in the continued presence of 10 nMATX-II.

Fig. 9. Site of initiation and propagation of electrical activity in the isolated murinesino-atrial preparation. Top panel (control) shows one electrogram, recorded fora period of 10 s from electrode #3 (location of which is shown in the map). Left paneldisplays eight electrograms, recorded during 100 ms from locations 1–8. The black starindicates the site of earliest detected waveform. The right panel displays the corre-sponding propagation map with isochrones drawn every 5 ms after the first detectedwaveform (black star at site #3). Propagation occurred from electrode site #3 to thelower right corner in about 15–20 ms. The electrodes in the lower row and three in theright hand corner did not record significant signals and were left blank. Middle panelsshow the effects of 5 nM ATX-II and lower panels the effects of ATX-II in the presenceof 10 mM ranolazine.

J. Wu et al. / Progress in Biophysics and Molecular Biology 98 (2008) 198–207206

of LQT3 generated by a heterozygous knock-in KPQ-deletion ofSCN5A (SCN5AD/þ) (Nuyens et al., 2001). The phenotype of themurine heart exposed to ATX-II is also consistent with that of theguinea pig isolated heart exposed to ATX-II (Wu et al., 2004).Murine hearts exposed to ATX-II also showed a significantsuppression of sinus node function including prolongation of thesinus cycle length and SNRT, and frequent occurrences of sinuspauses and arrest (as shown in Figs. 5, 7, 8 and Table 1). A similarphenotype was also observed in isolated sino-atrial preparationsexposed to ATX-II (Fig. 9). ATX-II reduced the intrinsic rate of sino-atrial preparations, caused pacemaker site shift, and prolongedconduction time from the leading pacemaker site to thesurrounding atrium tissue (as shown in Fig. 9) The effects of ATX-IIin this study suggest that the ATX-II treated heart is a useful modelfor studying LQT3-associated bradyarrhythmias. Despite heart ratereduction is not general hallmark of LQT3, because QT prolongationis most pronounced at lower heart rates, whereas it is virtuallyabsent at high heart rates, bradycardia may represent an importantindirect factor in the predisposition to lethal arrhythmias in fami-lies with LQT3. Bradycardia may also be a direct cause of suddendeath (Lei et al., 2008).

Whereas the precise role of INa,L in sinus node function is yet tobe defined, INa,L is expected to result in sinus node dysfunctionthrough at least two known cellular mechanisms that lead toventricular tachyarrhythmias: (1) direct electrophysiological alter-ations and (2) dysregulation of cell Naþ and Ca2þ cycling (Noble andNoble, 2006). Thus, the INa,L-induced prolongation of APD results ina slowing of the diastolic depolarisation rate and a reduction of thesinus node pacemaker rate. In addition, INa,L can lead to cellular Naþ

loading, altered rates of Naþ/Ca2þ exchange (Bers et al., 1996), andintracellular Ca2þ overload. Cellular Ca2þ overload may directly orindirectly affect a range of pacemaker currents. Furthermore, theheart rate reduction could be also caused by a current reductiondue to the leftward shift of steady-state inactivation (Oliveira et al.,2004), but not mainly by an increased INa,L.

4.3. Attenuation by ranolazine of ATX-II-induced alteration ofelectrophysiological properties and occurrence of arrhythmias

The present study has demonstrated the beneficial effects ofranolazine to suppress ATX-II-induced pro-arrhythmic events inmurine isolated hearts and sino-atrial preparations. In agreementwith results of previous studies (Wu et al., 2004; Song et al., 2008)10 mM ranolazine markedly attenuated ATX-II (1–10 nM)-inducedprolongations of MAPD90 and QT interval, and reduced the

J. Wu et al. / Progress in Biophysics and Molecular Biology 98 (2008) 198–207 207

occurrence of EADs, premature ventricular beats and VT in murineisolated hearts. In a previous study of the guinea pig isolated heart(Wu et al., 2004), ranolazine (5–30 mM) significantly attenuated theincrease in MAPD90, suppressed EADs, and reduced episodes of VTcaused by ATX-II. The effect of ranolazine to suppress ATX-II-induced arrhythmias was also demonstrated at the single cell level.Furthermore, ranolazine (10 mM) significantly inhibited or abol-ished the effects of ATX-II to cause prolongation of the APD andincreased occurrences of EADs, DADs and sustained triggeredactivity of guinea pig isolated atrial myocytes (Song et al., 2008).The antiarrhythmic effects of ranolazine have also been demon-strated in various other animal models and cardiac preparations(Antzelevitch et al., 2004; Ma et al., 2008; Sicouri et al., 2008; Songet al., 2008; Wang et al., 2008).

The present results also demonstrate for the first time the noveleffect of ranolazine to antagonise ATX-II-induced bradyarrhythmiasand slowing of atrial and AV conduction. Administration of 10 mMranolazine produced a remarkable inhibition of ATX-II-inducedprolongations of R–R and P–R intervals and SNRT, ATX-II-inducedsinus bradycardia, sinus pause, and sinus arrest in isolated heartpreparations, and reduction of the intrinsic rate in isolated sino-atrial tissue preparations (Figs. 6–9).

Inhibition of pharmacologically or pathologically enhanced INa,L

by ranolazine has been proposed to explain its reported antiar-rhythmic effects in various cardiac preparations (Antzelevitch et al.,2004; Ma et al., 2008; Sicouri et al., 2008; Song et al., 2008; Wanget al., 2008; Antzelevitch and Belardinelli, 2006; Makielski andValdivia, 2006). We suggest that inhibition by ranolazine of INa,L

shortens the prolonged pacemaker APD and reverses the slowing ofdiastolic depolarisation of the sinus node caused by ATX-II. Byreduction of intracellular Naþ overload and [Na]i-dependent Ca2þ

overload caused by INa,L, ranolazine may contribute to an improvedintracellular Naþ and Ca2þ homeostasis and electrical stability incardiac pacemaker tissue.

In conclusion, ranolazine antagonizes ATX-II-induced electro-physiological alterations of ventricular repolarisation and sinusnode function, and suppresses both tachy and bradyarrhythmiascaused by ATX-II. We suggest that INa,L blockade is the mechanismby which ranolazine reduces bradyarrhythmias.

Acknowledgement

The work was supported by the Chinese Nature Science Foun-dation (Project No, 30470634, ML), Chinese Scholar ResearchCouncil (JW, LC), The Wellcome Trust (ML), The British HeartFoundation (ML) and a CVT Research Grant (ML).

References

Antzelevitch, C., Belardinelli, L., Zygmunt, A.C., Burashnikov, A., Di Diego, J.M., Fish, J.M.,Cordeiro, J.M., Thomas, G., 2004. Electrophysiological effects of ranolazine, a novelantianginal agent with antiarrhythmic properties. Circulation 110, 904–910.

Antzelevitch, C., Belardinelli, L., 2006. The role of sodium channel current inmodulating dispersion of repolarization and Arrhythmogenesis. J. Cardiovasc.Electrophysiol. 17, 79–85.

Benson, D.W., Wang, D.W., Dyment, M., Knilans, T.K., Fish, F.A., Strieper, M.J.,Rhodes, T.H., George Jr., A.L., 2003. Congenital sick sinus syndrome caused byrecessive mutations in the cardiac sodium channel gene (SCN5A). J. Clin. Invest.112, 1019–1028.

Bers, D.M., Bassani, J.W., Bassani, R.A., 1996. Na–Ca exchange and Ca fluxes duringcontraction and relaxation in mammalian ventricular muscle. Ann. N. Y. Acad.Sci. 15, 779430–779442.

Chahine, M., Plante, E., Kallen, R.G., 1996. Sea anemone toxin (ATX II) modulation ofheart and skeletal muscle sodium channel a-subunits expressed in tsA201 cells.J. Membr. Biol. 152, 39–48.

Groenewegen, W.A., Firouzi, M., Bezzina, C.R., Vliex, S., van Langen, I.M.,Sandkuijl, L., Smits, J.P.P., Hulsbeek, M., Rook, M.B., Jongsma, H.J., Wilde, A.A.M.,2003. A cardiac sodium channel mutation cosegregates with a rare connexin40genotype in familial atrial standstill. Circ. Res. 92, 14–22.

Hale, S.L., Shryock, J.C., Belardinelli, L., Sweeney, M., Kloner, R.A., 2008. Late sodiumcurrent inhibition as a new cardioprotective approach. J. Mol. Cell. Cardiol. 44,954–967.

Isenberg, G., Ravens, U., 1984. The effects of the anemonia sulcata toxin (ATX-II) onmembrane currents of isolated mammalian myocytes. J. Physiol. 357, 127–149.

Lei, M., Jones, S.A., Liu, J., Lancaster, M.K., Fung, S.S.-M., Dobrzynski, H., Camelliti, P.,Maier, S.K.G., Noble, D., Boyett, M.R., 2004. Requirement of neuronal- andcardiac-type sodium channels for murine sinoatrial node pacemaking.’’.J. Physiol. (London) 559, 835–848.

Lei, M., Zhang, H., Huang, C., Grace, A.A., 2007. SCN5A and sinoatrial node pace-maker function. Cardiovasc. Res. 74, 356–365.

Lei, M., Zhang, Y., Huang, C., 2008. Genetic Naþ channelopathies and sinus nodedysfunction. Prog. Biophys. Mol. Biol. 98 (2-3), 171–178.

Makielski, J.C., Valdivia, G.R., 2006. Ranolazine and late cardiac sodium current –a therapeutic target for angina, arrhythmia and more? Brit. J. Pharmacol. 148,4–6.

Ma, J., Song, Y., Hu, L., Yan, X., Zhang, P., Shryock, J.C., Belardinelli, L., 2008. Rano-lazine inhibits hypoxia-stimulated late sodium current. FASEB J. 22 971.3.

Noble, D., Noble, P.J., 2006. Late sodium current in the pathophysiology ofcardiovascular disease: consequences of sodium-calcium overload. Heart 92,iv1–iv5.

Nuyens, D., Stengl, M., Dugarmaa, S., Rossenbacker, T., Compernolle, V., Rudy, Y.,Smits, J.F., Flameng, W., Clancy, C.E., Moons, L., Vos, M.A., Dewerchin, M.,Benndorf, K., Collen, D., Carmeliet, E., Carmeliet, P., 2001. Abrupt rate acceler-ations or premature beats cause life-threatening arrhythmias in mice withlong-QT3 syndrome. Nat. Med. 7, 1021–1027.

Oliveira, J.S., Redaelli, E., Zaharenko, A.J., Cassulini, R.R., Konno, K., Pimenta, D.C.,Freitas, J.C., Clare, J.J., Wanke, E., 2004. Binding specificity of sea anemone toxinsto Nav 1.1–1.6 sodium channels: UNEXPECTED CONTRIBUTIONS FROMDIFFERENCES IN THE IV/S3-S4 OUTER LOOP. J. Biol. Chem. 279, 33323–33335.

Papadatos, G.A., Wallerstein, P.M.R., Head, C.E.G., Ratcliff, R., Brady, P.A.,Benndorf, K., Saumarez, R.C., Trezise, A.E.O., Huang, C.L.H., Vandenberg, J.I.,Colledge, W.H., Grace, A.A., 2002. Slowed conduction and ventricular tachy-cardia after targeted disruption of the cardiac sodium channel gene Scn5a. Proc.Natl. Acad. Sci. U.S.A. 99, 6210–6215.

Richard Benzinger, G., Tonkovich, G.S., Hanck, D.A., 1999. Augmentation of recoveryfrom inactivation by site-3 Na channel toxins. A single-channel and whole-cellstudy of persistent currents. J. Gen. Physiol. 113, 333–346.

Sicouri, S., Glass, A., Belardinelli, L., Antzelevitch, C., 2008. Antiarrhythmic effects ofranolazine in canine pulmonary vein sleeve preparations. Heart Rhythm 5,1019–1026.

Smits, J.P.P., Koopmann, T.T., Wilders, R., Veldkamp, M.W., Opthof, T., Bhuiyan, Z.A.,Mannens, M.M.A.M., Balser, J.R., Tan, H.L., Bezzina, C.R., Wilde, A.A.M., 2005. Amutation in the human cardiac sodium channel (E161K) contributes to sicksinus syndrome, conduction disease and Brugada syndrome in two families.J. Mol. Cell. Cardiol. 38, 969–981.

Song, Y., Shryock, J.C., Belardinelli, L., 2008. An increase of late sodium currentinduces delayed afterdepolarizations and sustained triggered activity in atrialmyocytes. Am. J. Physiol. Heart Circ. Physiol. 294, H2031–H2039.

Veldkamp, M.W., Wilders, R., Baartscheer, A., Zegers, J.G., Bezzina, C.R., Wilde, A.A.,2003. Contribution of sodium channel mutations to bradycardia and sinus nodedysfunction in LQT3 families. Circ. Res. 92, 976–983.

Wang, Q., Chen, Q., Li, H., Towbin, J.A., 1997. Molecular genetics of long QT syndromefrom genes to patients. (Review) (61 refs). Curr. Opin. Cardiol. 12, 310–320.

Wang, Q., Shen, J., Splawski, I., Atkinson, D., Li, Z., Robinson, J.L., Moss, A.J.,Towbin, J.A., Keating, M.T., 1995. SCN5A mutations associated with an inheritedcardiac arrhythmia, long QT syndrome. Cell 80, 805–811.

Wang, W.-Q., Robertson, C., Dhalla, A.K., Belardinelli, L., 2008. Antitorsadogeniceffects of ({þ/�})-N-(2,6-Dimethyl-phenyl)-(4[2-hydroxy-3-(2-methoxyphenoxy)propyl]-1-piperazine (ranolazine) in anesthetized rabbits. J. Pharmacol.Exp. Ther. 325, 875–881.

Wu, L., Shryock, J.C., Song, Y., Li, Y., Antzelevitch, C., Belardinelli, L., 2004. Antiar-rhythmic effects of ranolazine in a guinea pig in vitro model of long-QTsyndrome. J. Pharmacol. Exp. Ther. 310, 599–605.

Zhang, Y., Wang, T., Ma, A., Zhou, X., Gui, J., Wan, H., Shi, R., Huang, C., Grace, A.,Huang, C., Trump, D., Zhang, H., Zimmer, T., Lei, M., 2008. Correlations betweenclinical and physiological consequences of the novel mutation R878C ina highly conserved pore residue in the cardiac Naþ channel. Acta Physiologica9999.

Zicha, S., Fernandez-Velasco, M., Lonardo, G., L’Heureux, N., Nattel, S., 2005. Sinusnode dysfunction and hyperpolarization-activated (HCN) channel subunitremodeling in a canine heart failure model. Cardiovasc. Res. 66, 472.