| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
CARDIOVASCULAR |
1 Physiological Laboratory, University of Cambridge, Downing Street, Cambridge CB2 3EG, UK
2 Section of Cardiovascular Biology, Departments of Biochemistry, University of Cambridge, Tennis Court Road, Cambridge CB2 1QW, UK
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
murine hearts modelling the long-QT(3) syndrome.
(Received 22 January 2007;
accepted after revision 8 February 2007;
first published online 15 February 2007)
Corresponding author C. L.-H. Huang, Physiological Laboratory, University of Cambridge, Downing Street, Cambridge CB2 3EG, UK. Email: clh11{at}cam.ac.uk
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
-subunit of the cardiac Na+ channel result in a ‘loss-of-function’ in BrS and a ‘gain-of-function’ in LQT3 (Brugada & Brugada, 1992; Wang et al. 1995; Gussak et al. 1999). The two syndromes show contrasting responses to pharmacological intervention. The antiarrhythmic class Ic agent flecainide used to suppress atrial fibrillation (AF) (Hopson et al. 1996), shortens the electrocardiographic QT interval and normalizes ventricular repolarization in LQT3 (Moss et al. 2005), but produces re-entrant effects in tissues whose conduction velocity is compromised by the BrS (Brugada et al. 2000). Indeed, intravenous flecainide unmasks the arrhythmogenic phenotype in otherwise asymptomatic BrS patients (Balser, 1999; Brugada, 2000; Priori et al. 2000; Viswanathan et al. 2001; Gasparini et al. 2003; Wolpert et al. 2005). Flecainide also produces surrogate arrhythmogenic effects and increases the incidence of electrocardiographic patterns characteristic of the BrS in genetically normal hearts (Yasuda et al. 2001; Hudson et al. 2004). Ventricular tachyarrhythmias likewise occur in AF patients receiving flecainide therapy (Falk, 1989). Moreover, proarrhythmic effects relating to flecainide therapy were implicated in the prospective, randomized, Cardiac Arrhythmia Suppression Trial (CAST, 1989); long-term suppression of ventricular premature complexes following acute myocardial infarction was accompanied by a 3.6-fold increased mortality rate (Epstein et al. 1993; Anderson et al. 1994).
In contrast, the class Ia agent quinidine prevents BrS-related arrhythmias and reverses the associated electrocardiographic abnormalities by inhibiting K+ currents including the transient outward current Ito (Alings et al. 2001; Belhassen et al. 2004; Hermida et al. 2004; Mok et al. 2004). However, quinidine causes prolongation of the electrocardiographic QT interval (Roden et al. 1986a,, b) and exacerbates the repolarization abnormalities typically associated with LQT3 (Wang et al. 2004). Quinidine-induced tachyarrhythmias have also been observed in patients receiving treatment for AF (Bauman et al. 1984; Roden & Anderson, 2006).
The present study was prompted by these contrasts between BrS and LQT3. It directly follows from a recent paper that extended characterizations of the biophysical effects of flecainide and quinidine at the cellular and isolated tissue level (Anno & Hondeghem, 1990; Starmer et al. 1991; Krishnan & Antzelevitch, 1993; Yan & Antzelevitch, 1999; Liu et al. 2003; Ramos & O'Leary, 2004) to an examination of arrhythmogenic effects at the whole-heart level in genetically modified (Scn5a+/) ‘gain-of-function’ mice modelling LQT3 (Stokoe et al. 2007). It also extends previous electrophysiological reports of alterations in Na+ channel characteristics in isolated Scn5a+/– myocytes (Papadatos et al. 2002). The experiments described here thus examine the arrhythmogenic properties of Langendorff-perfused, ‘loss-of-function’
Scn5a+/– murine hearts, the effects upon these of flecainide and quinidine and the extent to which such whole-heart models recapitulate the human BrS.
The experiments first assessed arrhythmic tendency using a programmed electrical stimulation (PES) technique (Saumarez & Grace, 2000; Balasubramaniam et al. 2003; Saumarez et al. 2003, 2006; Turner et al. 2005). This demonstrated that Scn5a+/– hearts have a tendency to arrhythmia in common with Scn5a+/ hearts (Head et al. 2005; Stokoe et al. 2007). Secondly, paced electrogram fractionation analysis (PEFA) yielded conduction curves (Balasubramaniam et al. 2003; Head et al. 2005; Stokoe et al. 2007) that associated increases in ventricular electrogram duration (EGD) with arrhythmogenesis whether due to genetic modification or pharmacological intervention in Scn5a+/– hearts, as also observed in Scn5a+/ hearts (Stokoe et al. 2007). Such abnormal patterns of myocardial activation or changes in EGD reflecting the distribution (‘fractionation’) of conduction velocities have been attributed to re-entrant substrates on earlier occasions (Saumarez et al. 2003, 2006; Saumarez & Grace, 2000; Turner et al. 2005). In contrast, prolonged ventricular effective refractory periods (VERPs) were associated with antiarrhythmic effects in the presence of pharmacological agents in Scn5a+/– hearts in common with Scn5a+/ hearts (Stokoe et al. 2007).
Thirdly, flecainide and quinidine exerted sharply contrasting effects on arrhythmogenesis in Scn5a+/– and Scn5a+/ hearts (Stokoe et al. 2007), recapitulating the clinical findings relating to BrS and LQT3, respectively (Fujiki et al. 1999; Liu et al. 2003; Wilde & Tan, 2003; Belhassen et al. 2004). Thus, flecainide increased EGDs and exerted proarrhythmic effects in Scn5a+/– and wild-type (WT) hearts, in contrast to the reduced EGDs and arrhythmic tendency previously reported in Scn5a+/ hearts (Stokoe et al. 2007). Conversely, quinidine reduced EGDs, prolonged VERPs and exerted antiarrhythmic effects in Scn5a+/– hearts, in contrast to the increased EGDs and proarrhythmic effects reported in Scn5a+/ hearts (Stokoe et al. 2007).
Fourthly, the arrhythmogenic properties of Scn5a+/– hearts accompanied consistently positive gradients of action potential (AP) duration derived from epicardial and endocardial monophasic AP recordings. This contrasted with the negative gradients of repolarization previously demonstrated in Scn5a+/ hearts (Stokoe et al. 2007). The findings of the present paper consequently attribute the arrhythmogenic properties associated with the Scn5a+/– mutation and flecainide treatment to re-entrant substrates resulting from delayed epicardial activation in contrast to the transmural heterogeneities reported in Scn5a+/ hearts modelling LQT3 (Stokoe et al. 2007).
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Breeding pairs of heterozygote Scn5a+/– (Papadatos et al. 2002) and wild-type (WT) mice with an inbred 129/Sv genetic background (supplied initially by Harlan, U.K) were set up and their offspring weaned accordingly. Mice were kept in plastic cages at room temperature in an animal facility, given free access to sterile rodent chow and water and subjected to 12 h light/dark cycles. All procedures conformed to the U.K. Animals (Scientific Procedures) Act 1986.
Experimental preparations
Full experimental protocols for the preparation and study of murine hearts have been previously described (Balasubramaniam et al. 2003; Stokoe et al. 2007). Male and female mice (age 6–8 months) were randomly selected and killed by cervical dislocation (Schedule 1: Animals (Scientific Procedures) Act 1986). Hearts were excised following bilateral sternotomy and submerged in ice-cold bicarbonate Krebs–Henseleit buffer solution containing (mM): 119 NaCl, 25 NaHCO3, 4.0 KCl, 1.2 KH2PO4, 1.0 MgCl2, 1.8 CaCl2, 10 glucose and 2.0 sodium pyruvate (pH 7.4). After removal of excess lung tissue, the severed end of the aorta was fed over a 21-gauge tailor-made cannula consisting of a blunted needle and small length of plastic tubing prefilled with buffer solution, to which the heart was fixed securely with fine thread suture. The preparation was then mounted onto a Langendorff system and perfused retrogradely for not less than 5 min prior to electrophysiological testing at a constant flow rate (2–2.5 ml min–1) (Watson-Marlow Bredel Peristaltic pumps, model 505S, Falmouth, Cornwall, UK) with oxygenated (95% O2, 5% CO2) Krebs–Henseleit buffer solution. The perfusate was filtered through two, 200 µm and 5 µm, membranes (Millipore UK, Watford, UK), and warmed to 37°C by a water-jacketed heat-exchange coil (Model C-58A, Techne, Cambridge, UK) before entering the coronary arterial network. The aortic valve was shut by the pressure of the perfusate that ultimately drained through the vena cava. Viable hearts regained a pink appearance and spontaneous rhythmic contractions upon warming and filling of the left anterior descending artery. Ischaemic hearts were discarded.
Bipolar electrogram recording
The electrophysiological properties of the resulting preparations were examined prior and subsequent to the administration of pharmacological agents using an established, programmed electrical stimulation (PES), technique (Saumarez & Grace, 2000; Balasubramaniam et al. 2003; Saumarez et al. 2003, 2006; Head et al. 2005; Turner et al. 2005; Stokoe et al. 2007). Electrograms were recorded from the basal epicardial surface of the left ventricle using paired platinum recording electrodes (1 mm interpole spacing). Paired platinum stimulating electrodes connected to a Grass S48 stimulator (Grass-Telefactor, UK, Slough, UK) were used to pace the heart high on the right ventricular septum at 1.5 to 3 times excitation threshold (average stimulus amplitude 4.18 ± 0.22 V, n = 60) (Tyers et al. 1997; Balasubramaniam et al. 2003, 2004, 2005; Head et al. 2005; Gammage et al. 2006; Stokoe et al. 2007) with square-wave stimuli of 2 ms duration. Stimulation of this strength ensured a consistent 1: 1 capture. Custom-made magnetic grips fixed to a metallic platform were used to anchor the electrodes and maintain stable contact pressure. The position of the stimulation and recording electrodes was fixed upon gaining a clear signal, and the distance between them maintained at 1 cm throughout. The heart was paced at a frequency of 10 Hz for 20 min to reach a physiological steady state.
Cycles of a decremental, paced electrogram fractionation sequence comprising an eight-beat stimulus (S1) drive train followed by an extra-stimulus (S2) were then applied at 8 Hz and 10 Hz for 
120 s and 100 s, respectively. The S1–S2 interval was reduced by 1 ms between successive drive trains until the preparation became refractory. Observations of either non-sustained ventricular tachycardia (VT) or sustained VT exceeding 8 cycles of stimulation in duration (see Figs 1 and 2), were confirmed in further procedures in which S1 pacing was discontinued following the initiation of an S2-induced arrhythmia. The resulting bipolar electrograms (BEGs) were amplified, band-pass filtered (30 Hz to 1 kHz) using a Gould 2400S amplifier (Gould-Nicolet Technologies, Ilford, Essex, UK) and digitized using an analog to digital converter at a sampling frequency of 5 kHz (CED 1401plus, Cambridge Electronic Design, Cambridge, UK). Spike2 software (Cambridge Electronic Design) was used to capture and analyse data at a corresponding sampling frequency of 0.2 ms; electrogram duration (EGD), ventricular effective refractory period (VERP) and conduction latency values were obtained from the resulting ventricular conduction curves to one additional decimal place.
|
|
|
Monophasic action potentials (MAPs) were recorded in addition to BEGs using an established contact-electrode technique described in detail previously (Franz, 1999; Knollmann et al. 2001; 2002,; Stokoe et al. 2007). Epicardial MAPs were recorded from the basal surface of the left ventricular epicardium using a miniaturized MAP electrode tip designed specifically for use on rodent hearts (Linton Instruments, Harvard Apparatus, UK). This was mounted on a perpendicular, spring-loaded electrode holder to ensure stable contact and constant contact pressure. Endocardial MAPs were recorded using electrodes constructed from galvanically chlorided, Teflon-coated 0.25 mm diameter silver wire (99.99% purity). Introduction to the left ventricular cavity was enabled through a small access window created in the interventricular septum. The tip of the electrode was rotated until it came to rest against the free wall. As above, paired platinum stimulating electrodes were used to pace the heart high on the right ventricular septum. MAP signals were amplified and band-pass filtered between 0.1 Hz and 300 Hz using a Gould 2400S amplifier (Gould-Nicolet Technologies), and then digitized using a 1401plus interface (Cambridge Electronic Design).
The above decremental PES protocol was applied to precipitate arrhythmia in the presence and absence of pharmacological intervention. Most epicardial and endocardial MAPs were recorded from the same heart, in which case waveforms were similar. Analysis of MAP waveforms was performed using Spike2 software (Cambridge Electronic Design). MAPs were derived from recordings that satisfied previously established criteria including a rapid upstroke phase with consistent amplitude, a smooth contoured repolarization phase and a stable baseline (Fabritz et al. 2003). The point of maximum positive deflection for each MAP was considered the point of 0% repolarization, and the point of full return to baseline that of 100% repolarization. The intervening waveform was described in terms of action potential duration (APDx) measurements at x
= 90% (APD90), 70% (APD70) and 50% (APD50) repolarization. The effect of heterogeneous repolarization on the difference between epicardial and endocardial AP recovery was expressed empirically as the difference (APDx) between endocardial APDx and epicardial APDx values, i.e. endocardial APDx
– epicardial APDx.
Statistical procedures
A one-way analysis of variance (ANOVA) for independent samples was used to compare data sets from separate experimental groups (WT and Scn5a+/–), whereas results from individual hearts acquired during pharmacological intervention were compared to their respective untreated controls using a one-way ANOVA for correlated samples (SPSS software). The APDx values calculated were also compared to a zero APDx. In addition, values obtained from hearts treated with flecainide or quinidine were compared to values obtained from untreated hearts under comparable conditions, and APDx values from untreated Scn5a+/– hearts were compared to APDx values from untreated WT hearts. All procedures assumed a significance level of P < 0.05. Results are expressed as mean ±
S.E.M. values.
Reagents
The reagents flecainide and quinidine (Sigma-Aldrich, Poole, UK) were dissolved in doubly distilled water to make 1.0 mM stock solutions. Final drug concentrations were achieved by dilution with buffer solution which was perfused for 15 min prior to the imposition of PES and maintained throughout. The concentrations used matched known clinical therapeutic levels (flecainide: 0.2–0.9 mg l–1, i.e. 0.5–2.0 µM; quinidine: 2.0–5.0 mg l–1, i.e. 6.0–20 µM) (Birkett, 1997). Flecainide was stored at 4°C between experiments, and quinidine at room temperature in darkness to prevent light degradation, in a tightly sealed container.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Murine Scn5a+/– hearts recapitulate the clinical arrhythmogenic phenotype
Table 1 compares the incidence of VT following extra-stimuli (S2) imposed during PES prior to and in the course of treatment with flecainide and quinidine. The BEG records (i) demonstrate that untreated Scn5a+/– murine hearts have a tendency to arrhythmia in common with BrS patients (Alings & Wilde, 1999), and (ii) recapitulate a number of the major clinical effects of flecainide and quinidine. All hearts showed a regular rhythm upon (S1) pacing with no evidence of spontaneous arrhythmogenesis in both the presence and absence of flecainide and/or quinidine prior to the introduction of PES. However, BEGs acquired in the absence of pharmacological intervention revealed a greater propensity to arrhythmia in Scn5a+/– as opposed to WT hearts, in close parallel with the clinical reports (Brugada & Brugada, 1992; Gussak et al. 1999). Furthermore, the arrhythmogenic population of Scn5a+/– hearts demonstrated either non-sustained VT (nsVT) (lasting 0.45 ± 0.10 s, n = 5), sustained VT (sVT) exceeding eight stimulation cycles in duration (26.70 s, n = 1), or sVT (10.31 ± 2.44 s, n = 10) in addition to one or more episodes of nsVT (0.38 ± 0.058 s, n = 10); PES induced VT in Scn5a+/– hearts isolated from both male (eight out of 19) and female (eight out of 10) mice. Figure 1A and B shows PES sequences recorded from an untreated arrhythmogenic Scn5a+/– heart. The single vertical markers at the base of each trace indicate the timing of S1 stimuli (stim1) and the double vertical markers S2 extra-stimuli (stim2); Fig. 1A illustrates nsVT initiated by an extra-stimulus (S2), and Fig. 1B illustrates sVT that persisted whether or not S1 pacing was continued. In all pacing sequences, the VT that occurred followed those extra-stimuli (S2) that had been delivered at an S1–S2 interval close to the VERP. In contrast, PES failed to induce VT in any of the WT control hearts.
|
The Na+ channel-blocking agent flecainide exerted marked arrhythmogenic effects on isolated, perfused WT hearts. Thus, Table 1 indicates that (10 µM) flecainide increased arrhythmic tendency in WT hearts; it also induced nsVT (0.73 ± 0.13 s, n = 6), sVT (21.50 s, n = 1), and sVT (6.72 ± 3.42 s, n = 4) in addition to one or more episodes of nsVT (0.61 ± 0.087 s, n = 4). Flecainide at a concentration of 5.0 µM was without effect (n = 5) (cf. Krishnan & Antzelevitch, 1993). However, Scn5a+/– hearts were more sensitive to lower concentrations of flecainide than WT hearts in parallel with the sensitivity to flecainide of asymptomatic BrS patients; levels otherwise ineffective in normal human subjects induce ventricular arrhythmias and unmask the concealed BrS (Brugada, 2000; Viswanathan et al. 2001). Thus, in the presence of only 1.0 µM flecainide, a similar proportion of initially non-arrhythmogenic Scn5a+/– hearts became arrhythmogenic; two hearts demonstrated nsVT (0.50 ± 0.25 s, n = 2), and seven hearts demonstrated sVT (3.73 ± 0.98 s, n = 7) in addition to one or more episodes of nsVT (0.45 ± 0.066 s, n = 7). Moreover, the higher (10 µM) flecainide concentration caused either stimulus capture or electrogram generation to fail in Scn5a+/– hearts. Finally, quinidine reduced arrhythmic tendency in flecainide-treated WT and arrhythmogenic Scn5a+/– hearts in parallel with its effects on the symptomatic BrS (Belhassen et al. 2004; Hermida et al. 2004; Mok et al. 2004). Table 1 accordingly demonstrates that (10 µM) quinidine reduced the incidence of VT in arrhythmogenic Scn5a+/– hearts and (10 µM) flecainide-treated WT hearts; minor arrhythmogenic effects were exerted by quinidine in WT hearts.
Figures 1 and 2 illustrate the effects of pharmacological intervention on arrhythmogenesis in Scn5a+/– hearts. Traces were obtained from arrhythmogenic (Fig. 1) and non-arrhythmogenic Scn5a+/– hearts (Fig. 2) paced at 8 Hz prior to (Figs 1, A and B; Fig. 2A) and during treatment with quinidine (10 µM: Fig. 1C) and/or flecainide (1.0 µM: Fig. 2B). Figure 1C goes on to show that quinidine abolished VT close to the VERP and returned the arrhythmogenic Scn5a+/– heart to a regular rhythm. Figure 2A shows a persistent, regular rhythm in an untreated non-arrhythmogenic Scn5a+/– heart recorded during PES; sVT was induced by an (S2) extra-stimulus (double vertical markers) applied in the presence of flecainide (Fig. 2B). Three cycles of PES are shown to illustrate the time period over which the VT was sustained before reversion.
Conduction curve representations of BEG data
PEFA is used clinically to stratify arrhythmogenic risk in terms of alterations in EGD, VERP and conduction latency (Saumarez et al. 2003, 2006; Saumarez & Grace, 2000; Turner et al. 2005). These parameters are derived from families of ventricular conduction curves which reflect the presence or otherwise of arrhythmogenic re-entrant pathways. PEFA has been applied experimentally on earlier occasions to assess arrhythmic tendency and its physiological basis in KCNE1–/– (Balasubramaniam et al. 2003) and Scn5a+/ mice (Head et al. 2005; Stokoe et al. 2007). Its application in the present study likewise implicated re-entrant substrates in arrhythmogenesis.
Figure 3 illustrates typical BEG data (inset) recorded during PES, expressed in conduction curve form. The response latencies (ms) of each BEG deflection, defined as the time differences between the extra-stimulus (S2) and the peaks and troughs of the resulting BEG (inset), are plotted against the corresponding S1–S2 interval (ms). Figure 4 goes on to illustrate typical conduction curves from WT (Fig. 4A), non-arrhythmogenic Scn5a+/– (Fig. 4B) and arrhythmogenic Scn5a+/– hearts (Fig. 4C) paced at 8 Hz before (+) and during perfusion (
) with 10 µM (Fig. 4A) and 1.0 µM flecainide (Fig. 4B), and 10 µM quinidine (Fig. 4C), respectively. The time differences (ms) between extra-stimuli (S2) imposed during PES and the successive peaks and troughs of the resulting BEGs are plotted against the corresponding S1–S2 interval (ms). Immediate inspection of the conduction curves reveals that in WT hearts, flecainide (10 µM, ) markedly increased VERP, conduction latency and EGD, hence the vertical spread of data points, particularly those representing the peaks reflecting the slowest conduction velocities obtained at short S1–S2 intervals, compared to drug-free controls (Fig. 4A, +). The conduction curves also indicate that non-arrhythmogenic Scn5a+/– hearts (Fig. 4B, +) showed increased VERPs compared to WT hearts (Fig. 4A, +) in the absence of flecainide, and in this respect resembled flecainide-treated WT hearts. However, flecainide treatment (1.0 µM) then produced further increases in VERP and EGD as evident in the increased vertical spread of data points at the more rapid conduction velocities (Fig. 4B, ). Finally, VERP was reduced in untreated arrhythmogenic Scn5a+/– hearts (Fig. 4C, +) relative to values from non-arrhythmogenic Scn5a+/– hearts (compare Fig. 4B and C). Furthermore, the inhibition of arrhythmogenesis that followed treatment with quinidine (10 µM) was accompanied by sharp increases in conduction latency consistent with slowed AP conduction, but also a prolongation of VERP (Fig. 4C, ).
|
A closer comparison of the parameters derived from conduction curves was enabled using a one-way analysis of variance (ANOVA) for independent samples. This was applied to the PEFA data obtained from separate populations of WT and Scn5a+/– hearts prior to pharmacological intervention. Figure 5 compares mean values of EGD ratio (defined in Methods) (Fig. 5A), VERP (Fig. 5B) and conduction latency (Fig. 5C) (± S.E.M.) from hearts paced at 8 Hz and 10 Hz. The arrhythmogenic population of Scn5a+/– hearts had significantly greater EGD ratios than the WT hearts (Fig. 5A), suggestive of a re-entrant substrate, but similar VERPs (Fig. 5B). In contrast, non-arrhythmogenic Scn5a+/– hearts had similar EGD ratios to WT hearts but significantly longer VERPs. Furthermore, Fig. 5C agrees with earlier work in demonstrating that Scn5a+/– hearts had increased electrogram latencies consistent with reduced conduction velocities relative to WT hearts (cf. Papadatos et al. 2002). Thus, the entire population of Scn5a+/– hearts together had conduction latencies of 20.97 ± 1.24 ms (n = 36, P = 0.001) and 23.49 ± 1.59 ms (n = 36, P = 0.0008) at 8 Hz and 10 Hz, respectively, with less clear-cut differences in EGD ratio (1.67 ± 0.12, n = 30, P = 0.02 and 1.44 ± 0.077, n = 27, P = 0.2) and VERP (50.42 ± 2.54, n = 31, P = 0.2 and 54.14 ± 2.38, n = 29, P = 0.01) when compared to WT.
|
Table 2 summarizes the effects of flecainide and quinidine on values of EGD ratio (A), VERP (B) and conduction latency (C) in WT and Scn5a+/– hearts (means ± S.E.M.) paced at 8 and 10 Hz. The effects of the pharmacological agents on individual hearts were assessed to significance levels of P < 0.05(*), P < 0.005(**) and P < 0.0005(***), respectively, in experimental studies where such a full quantitative analysis was possible; stimulus capture or electrogram generation failed in arrhythmogenic Scn5a+/– hearts treated with flecainide. These results confirmed an association linking increased EGD ratios with an arrhythmogenic phenotype, and increased VERPs with a non-arrhythmogenic phenotype.
|
Monophasic action potential studies
These empirical observations of the murine ‘loss-of-function’ arrhythmogenic Scn5a+/– heart and the flecainide-treated initially non-arrhythmogenic Scn5a+/– and WT hearts thus agree with previous studies of the ‘gain-of-function’
Scn5a+/ heart (Stokoe et al. 2007). All of these systems showed a marked propensity to arrhythmia and evidence of re-entrant substrates reflected by similar alterations in EGD ratio and VERP (Head et al. 2005; Stokoe et al. 2007). However, the present findings demonstrate that Scn5a+/– hearts contrast with Scn5a+/ hearts in terms of the proarrhythmic effects of flecainide and the antiarrhythmic effects of quinidine. The previous study had attributed the arrhythmogenic properties of Scn5a+/ hearts and the effects upon these of flecainide and quinidine to observed differences in AP waveforms and repolarization kinetics detected by recording MAPs (Stokoe et al. 2007). This prompted a parallel investigation of AP waveforms in Scn5a+/– hearts. Thus, epicardial and endocardial MAPs were recorded at 8 Hz and 10 Hz pacing frequencies in WT and Scn5a+/– hearts prior to and during treatment with flecainide (1.0 µM: Scn5a+/–; 10 µM: WT) and quinidine (10 µM) using similar pulse protocols.
Figure 6 illustrates MAP waveforms recorded from the epicardium of intrinsically (Fig. 6A) and extrinsically paced (Fig. 6B–F) initially arrhythmogenic (Fig. 6A, B, E and F) and non-arrhythmogenic Scn5a+/– hearts (Fig. 6C and D). Experiments were performed in the absence (Fig. 6A) and presence of electrical stimulation prior to (Fig. 6B) and during PES (Fig. 6C–F), before (Fig. 6A–C and E) and subsequent to perfusion with (1.0 µM) flecainide (Fig. 6D) and (10 µM) quinidine (Fig. 6F). The single vertical markers represent S1 stimuli and the double vertical markers S2 extra-stimuli. The MAPs recorded from intrinsically paced Scn5a+/– hearts (Fig. 6A) demonstrated a marked bradycardia (78 ± 6.51 beats min–1, n = 42) compared to WT (96 ± 2.66 beats min–1, n = 12), in agreement with earlier reports (Lei et al. 2005). Triggering events followed by extra-systolic AP excitation (Fig. 6A) and spontaneous VT (Fig. 6B) were observed in only one out of 18 arrhythmogenic Scn5a+/– hearts studied before (Fig. 6A) and during regular S1 pacing at 8 Hz (Fig. 6B). This agrees with the previous BEG records from untreated arrhythmogenic Scn5a+/– hearts in which no such events were observed. Furthermore, 14 untreated arrhythmogenic Scn5a+/– hearts demonstrated nsVT (lasting 0.62 ± 0.056 s, n = 14); an additional four hearts demonstrated sVT exceeding eight stimulation cycles in duration (4.19 ± 1.87 s, n = 4).
|
Effects of flecainide and quinidine on MAP waveforms
Figures 7 and 8 illustrate results from experiments investigating MAP waveforms recorded from both the epicardium and endocardium of typical WT (Fig. 7), non-arrhythmogenic (Fig. 8A–C) and arrhythmogenic (Fig. 8D–F) Scn5a+/– hearts paced at 8 Hz prior to (Fig. 8A and D) and during treatment with flecainide (10 µM: Fig. 7B, C; 1.0 µM: Fig. 8B and C) and quinidine (10 µM: Figs 7E and F, and 8E and F). The traces are displayed in a format that permits direct comparison with the recently reported findings in Scn5a+/ hearts (Stokoe et al. 2007). The top panels in both Figs 7 and 8 overlay epicardial and endocardial MAPs to emphasize AP waveform differences in untreated hearts. The lower two panels illustrate the effects of flecainide and quinidine on MAP waveform by superimposing records obtained from the epicardium (‘epi’: panels B and E) and endocardium (‘endo’: panels C and F) before (designated ‘epi’ or ‘endo’) and during treatment (‘+ flecainide’ or ‘+ quinidine’). Untreated WT hearts had longer endocardial repolarization phases than their corresponding epicardial MAPs (Fig. 7A and D). However, in the presence of flecainide, both the epicardial (Fig. 7B) and endocardial (Fig. 7C) MAPs returned to baseline more rapidly than their respective controls, whereas in the presence of quinidine the epicardial (Fig. 7E) as opposed to endocardial (Fig. 7F) MAPs had longer repolarization phases than their respective controls.
|
|
hearts previously described (Stokoe et al. 2007). Thus, epicardial MAPs from untreated non-arrhythmogenic (Fig. 8A) and arrhythmogenic (Fig. 8D) Scn5a+/– hearts resembled WT hearts; repolarization phases were shorter than their corresponding endocardial MAPs. The presence of flecainide resulted in a more rapid return to baseline of the epicardial (Fig. 8B) and endocardial MAPs obtained from non-arrhythmogenic Scn5a+/– hearts (Fig. 8C). In contrast, in the presence of quinidine, the epicardial (Fig. 8E) and endocardial MAPs obtained from arrhythmogenic Scn5a+/– hearts (Fig. 8F) had longer repolarization phases than their respective controls. Quantitative representation of epicardial and endocardial action potential time courses
Figures 9–11 quantify the characteristics of epicardial and endocardial MAP waveforms and the effects upon these of flecainide and quinidine as recorded from WT (Fig. 9), non-arrhythmogenic Scn5a+/– (Fig. 10) and arrhythmogenic Scn5a+/– hearts (Fig. 11) paced at 8 Hz and 10 Hz. The parameter of 0% repolarization was assigned to the point of maximum positive deflection for each MAP waveform, and 100% repolarization to the point of full return to baseline. The intervening waveform was described in terms of AP duration (APDx) measurements at x
= 90% (APD90), 70% (APD70) and 50% (APD50) repolarization (mean ±
S.E.M.), respectively. The effect of heterogeneous repolarization on the difference between epicardial and endocardial AP recovery was expressed empirically as the difference (APDx) between endocardial APDx and epicardial APDx values. Letters a–f (Figs 10 and 11) provide statistical comparisons between corresponding APDx and APDx values from untreated non-arrhythmogenic Scn5a+/– (Fig. 10), arrhythmogenic Scn5a+/– (Fig. 11) and WT hearts (Fig. 9).
|
|
|
APD90 values remained positive throughout. Finally, whereas untreated non-arrhythmogenic Scn5a+/– hearts showed no clear-cut differences (Fig. 10), arrhythmogenic Scn5a+/– hearts showed reduced APD90 values (Fig. 11) relative to WT hearts (Fig. 9). However, flecainide significantly reduced the epicardial and endocardial APD90 values in non-arrhythmogenic Scn5a+/– hearts paced at 8 Hz and 10 Hz. In contrast, quinidine significantly prolonged epicardial APD90 values at 8 Hz and 10 Hz, but reduced endocardial APD90 values at 8 Hz. Nevertheless, APD90 values remained positive throughout in sharp contrast to the Scn5a+/ hearts reported earlier (Stokoe et al. 2007). |
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
-subunit of the cardiac Na+ channel (Gussak et al. 1999), syncope and sudden cardiac death (SCD) (Antzelevtich et al. 2002; Grant et al. 2002). However, BrS has been attributed to ‘loss-of-function’, whereas LQT3 has been attributed to ‘gain-of-function’ mutations causing incomplete Na+ channel inactivation (Brugada & Brugada, 1992; Wang et al. 1995; Gussak et al. 1999). Furthermore, BrS and LQT3 show contrasting responses to pharmacological intervention with flecainide and quinidine. The class Ic cardiotropic agent flecainide exerts antiarrhythmic effects in LQT3 patients by normalizing ventricular repolarization and shortening the electrocardiographic QT interval (Moss et al. 2005). It also maintains normal sinus rhythm in patients treated with electrical cardioversion for atrial fibrillation (AF) (Hopson et al. 1996). Conversely, flecainide accentuates arrhythmogenic features and unmasks the concealed BrS in otherwise asymptomatic patients (Balser, 1999; Brugada, 2000; Priori et al. 2000; Brugada et al. 2000; Gasparini et al. 2003; Wolpert et al. 2005). It increases the incidence of surrogate arrhythmogenic effects and reproduces the BrS phenotype in normal hearts (Alings & Wilde, 1999; Fujiki et al. 1999; Gussak et al. 1999; Yasuda et al. 2001; Hudson et al. 2004), as well as sharply increasing thresholds for ventricular capture (Antonelli et al. 2001; Fornieles-Perez et al. 2002). Furthermore, flecainide has been known to produce ventricular tachycardia (VT) and ventricular fibrillation in AF patients receiving flecainide (Falk, 1989). Indeed, flecainide therapy was associated with life-threatening arrhythmias in the Cardiac Arrhythmia Suppression Trial (CAST, 1989) (Anderson et al. 1994; Epstein et al. 1993).
The class Ia cardiotropic agent quinidine slows ventricular repolarization and prolongs the QT interval (Roden et al. 1986a,, b), which might increase arrhythmic tendency in LQT3 (Wang et al. 2004). In contrast, quinidine suppresses BrS-related arrhythmias and reduces the associated clinical risks of SCD (Alings et al. 2001; Belhassen et al. 2004; Hermida et al. 2004; Mok et al. 2004). It is also used to reduce the risk of recurrent AF after conversion to sinus rhythm, although torsades de pointes is a major side-effect (Bauman et al. 1984; Dell'Orfano et al. 1998; Roden & Anderson, 2006).
This study was prompted by the above contrasting arrhythmogenic properties associated with BrS and LQT3, and the availability of analogous murine models. Genetically modified murine hearts are increasingly used as models of human cardiac disease (London, 2001). However, the translation of results from isolated murine hearts to the clinical setting requires consideration of the differences between species as well as the experimental protocols employed. Thus, despite similar anatomy and fibre organization, murine and human hearts show marked differences in basal heart rate, heart size and ion channel expression (Vaidya et al. 1999). Moreover, differences in the repolarizing currents that constitute the AP waveform cause the murine ventricular AP to overshoot and fall sharply in contrast to the pronounced plateau phase of the human ventricular AP. This produces a shorter APD in the mouse (30–80 ms) relative to the human (150–400 ms) (Danik et al. 2002). Nevertheless, murine and human ventricular APs share a steep upstroke (phase 0) reflecting rapid Na+ channel gating (Guo et al. 1999). The kinetics of the subsequent transient outward repolarizing K+ current (Ito) strongly influence AP recovery and APD, resulting in gradients of repolarization common to both. Admittedly, such parallels may not extend to situations in which species differences are large, namely reproduction and endocrine function, as may apply to the male predominance of overt clinical symptoms in the human BrS (Di Diego et al. 2002; Antzelevitch, 2003; Shimizu, 2004). Nonetheless, at the very least, murine hearts provide sufficient resemblances with human hearts to permit a basic investigation of arrhythmogenic mechanisms.
A previous study had examined Langendorff-perfused, ‘gain-of-function’
Scn5a+/ hearts modelling LQT3 in the presence and absence of flecainide and quinidine (Stokoe et al. 2007). The present experiments proceeded to examine the arrhythmogenic properties of genetically modified (Scn5a+/–) ‘loss-of-function’ murine hearts (Papadatos et al. 2002) compared to WT hearts under similar conditions, and the extent to which they replicated features of the human BrS. Arrhythmic tendency was first assessed using a PES technique adapted from clinical practice that applied extra-stimuli (S2) following trains of pacing stimuli (S1) at progressively reduced S1–S2 intervals (Saumarez et al. 2003; Saumarez & Grace, 2000). This demonstrated that whereas untreated WT hearts consistently yielded stable rhythms, untreated Scn5a+/– hearts had a tendency to arrhythmia in response to (S2) extra-stimulation in common with the Scn5a+/ hearts described earlier (Stokoe et al. 2007).
Secondly, the electrophysiological effects of flecainide treatment on WT and Scn5a+/– hearts agreed with clinical findings. Flecainide (10 µM) increased arrhythmic tendency in WT hearts in parallel with the BrS-like manifestations reported in flecainide-treated human subjects (Hudson et al. 2004; Yasuda et al. 2001). PES also induced VT in flecainide-treated initially non-arrhythmogenic Scn5a+/– hearts at reduced concentrations (1.0 µM), reflecting the sensitivity to flecainide of otherwise asymptomatic BrS patients (Priori et al. 2000; Wilde & Tan, 2003). Higher (10 µM) concentrations caused loss of electrogram capture in parallel with human observations (Antonelli et al. 2001; Fornieles-Perez et al. 2002). These results sharply contrast with the antiarrhythmic effects of flecainide reported in Scn5a+/ hearts (Stokoe et al. 2007), and confirm that Scn5a+/– hearts recapitulate the pharmacological phenotype of and provide suitable experimental models for the BrS.
Thirdly, quinidine relieved arrhythmogenic tendency in Scn5a+/– and flecainide-treated WT hearts, similarly recapitulating its clinical antiarrhythmic effects (Belhassen et al. 2004). In contrast, quinidine exerted proarrhythmic effects in the Scn5a+/ hearts previously described (Stokoe et al. 2007). Quinidine also exerted minor arrhythmogenic effects in the WT hearts studied here, in agreement with clinical reports (Roden & Anderson, 2006; Tong et al. 2001).
Fourthly, PEFA (Saumarez & Grace, 2000; Saumarez et al. 2003; 2006,; Turner et al. 2005) yielded conduction curves from which EGD, VERP and conduction latency values could be derived following progressive shortening of the S1–S2 interval. Alterations of EGD reflecting abnormal patterns of myocardial activation or the dispersion of myocardial conduction velocities, demonstrated re-entrant arrhythmic substrates in Scn5a+/– hearts in common with the Scn5a+/ hearts (Balasubramaniam et al. 2003; Head et al. 2005; Stokoe et al. 2007). Arrhythmogenesis, whether due to the respective mutations or pharmacological intervention, hence correlated with prolonged EGDs, expressed as EGD ratios normalized to their corresponding values obtained at long S1–S2 intervals; reversal of arrhythmogenesis correlated with reduced EGD ratios and prolonged VERPs. Thus, untreated arrhythmogenic Scn5a+/– hearts showed prolonged EGD ratios compared to WT controls, whereas untreated non-arrhythmogenic Scn5a+/– hearts showed prolonged VERPs. Furthermore, flecainide increased EGD ratios in WT and non-arrhythmogenic Scn5a+/– hearts at 10 µM and 1.0 µM concentrations, respectively, whereas quinidine reduced EGD ratios and prolonged VERPs in arrhythmogenic Scn5a+/– and WT hearts in parallel with its clinical effects. This contrasts with earlier reports in which flecainide reduced EGD ratios in arrhythmogenic Scn5a+/ hearts, and quinidine increased EGD ratios in non-arrhythmogenic Scn5a+/ hearts (Stokoe et al. 2007).
Fifthly, comparisons of epicardial and endocardial MAP recordings excluded an arrhythmogenic mechanism involving heterogeneities of ventricular repolarization (Antzelevitch & Fish, 2001) in Scn5a+/– hearts. This also contrasts with the findings reported in Scn5a+/ hearts and the effects upon these of flecainide and quinidine (Stokoe et al. 2007). Thus, endocardial APD exceeded epicardial APD in the Scn5a+/– hearts, resulting in a positive difference, APDx, in common with untreated WT hearts. Flecainide treatment further shortened epicardial APDx at 90% repolarization and similarly generated positive APD90 values in arrhythmogenic Scn5a+/– and WT hearts despite its proarrhythmic effects. In contrast, quinidine lengthened epicardial APD90 in arrhythmogenic Scn5a+/– hearts, while nevertheless preserving APD90.
Taken together, the present results, reported for the first time in Langendorff-perfused Scn5a+/– murine hearts, demonstrate features of re-entrant arrhythmogenesis that contrast with those recently described in Scn5a+/ hearts (Stokoe et al. 2007). These observations in whole hearts thus remain consistent with earlier reports at the cellular level that both the Scn5a+/– deletion mutation (Antzelevitch et al. 2002; Grant, 2005) and flecainide treatment reduce cardiac Nav1.5 channel function (Ramos & O'Leary, 2004) to a greater extent than they reduce K+ channel function (Slawsky & Castle, 1994; Lei et al. 2000; Akar et al. 2004). The Scn5a+/– mutation has already been shown to impair epicardial AP propagation and decrease total Na+ conductance, generating conduction block and re-entrant arrhythmias (Papadatos et al. 2002; Zhang et al. 2007). Flecainide is known to have a high affinity for the pore region of activated mutant Na+ channels, and also slows AP conduction (Anno & Hondeghem, 1990; Liu et al. 2003). Loss of Na+ channel function shifts the balance between inward (INa and ICa) and outward currents (Ito), causing heterogeneous early repolarization, preferentially of the epicardial as opposed to the endocardial AP (Subramaniam et al. 1991; Zhou et al. 1998; Yan & Antzelevitch, 1999), owing to the greater density of Ito in the epicardium (Yan & Antzelevitch, 1996; Tachibana et al. 1999; Antzelevitch et al. 2002). These re-entrant arrhythmogenic effects have been demonstrated in epicardial sheets of canine ventricle (Krishnan & Antzelevitch, 1993), and in modelling analyses exploring Na+ channel insufficiency (Starmer et al. 1991). Furthermore, intracoronary flecainide perfusion through intact canine hearts generates re-entrant electrical instabilities (Tachibana et al. 1999). Pharmacological management of the BrS seeks to rebalance the currents active in the early phases of the ventricular epicardial AP (Antzelevitch et al. 2002); quinidine is a potent non-specific K+- a
P < 0.05, 
P < 0.05, 
