Atrial fibrillation (AF) is the most common of all sustained

Transcription

1 Pacing-Induced Spontaneous Activity in Myocardial Sleeves of Pulmonary Veins After Treatment With Ryanodine Haruo Honjo, MD; Mark R. Boyett, PhD; Ryoko Niwa, MD; Shin Inada, MEng; Mitsuru Yamamoto, MD; Kazuyuki Mitsui, PhD; Toshiyuki Horiuchi, PhD; Nitaro Shibata, MD; Kaichiro Kamiya, MD; Itsuo Kodama, MD Background Recent clinical electrophysiology studies and successful results of radiofrequency catheter ablation therapy suggest that high-frequency focal activity in the pulmonary veins (PVs) plays important roles in the initiation and perpetuation of atrial fibrillation, but the mechanisms underlying the focal arrhythmogenic activity are not understood. Methods and Results Extracellular potential mapping of rabbit right atrial preparations showed that ryanodine (2 mol/l) caused a shift of the leading pacemaker from the sinoatrial node to an ectopic focus near the right PV-atrium junction. The transmembrane potential recorded from the isolated myocardial sleeve of the right PV showed typical atrial-type action potentials with a stable resting potential under control conditions. Treatment with ryanodine (0.5 to 2 mol/l) resulted in a depolarization of the resting potential and a development of pacemaker depolarization. These changes were enhanced transiently after an increase in the pacing rate: a self-terminating burst of spontaneous action potentials (duration, s; n 32) was induced by a train of rapid stimuli (3.3 Hz) applied after a brief rest period. The pacing-induced activity was attenuated by either depletion of the sarcoplasmic reticulum of Ca 2 or blockade of the sarcolemmal Na -Ca 2 exchanger or Cl channels and potentiated by -adrenergic stimulation. Conclusions PV myocardial sleeves have the potential to generate spontaneous activity, and such arrhythmogenic activity is uncovered by modulation of intracellular Ca 2 dynamics. (Circulation. 2003;107: ) Key Words: electrophysiology fibrillation veins action potentials Atrial fibrillation (AF) is the most common of all sustained tachyarrhythmias and is one of the major causes of stroke. The most widely accepted mechanism of AF is multiple reentrant wavelets. 1 However, recent clinical studies have shown that paroxysmal AF is initiated by bursts of premature excitations originating primarily in the pulmonary veins (PVs), and radiofrequency ablation or electrical isolation of these foci can eliminate AF. 2,3 More recently, Haïssaguerre et al 4 reported that in patients with drug-resistant chronic AF and structural heart disease, after electrical cardioversion, the PVs are also the dominant trigger reinitiating AF. Therefore, the PVs are important not only for the initiation of AF but also for its maintenance. 5 The myocardial fibers of the left atrium wrap around the PVs entering the left atrium to form myocardial sleeves (MSs), 6,7 and this structure is the origin of focal activity. 8 Previous studies have suggested that PVMSs show a variety of spontaneous activities such as sinoatrial (SA) node type automaticity, rapid spontaneous activities via early or delayed afterdepolarizations, 8 12 and microreentry based on a marked heterogeneous tissue structure. 12,13 However, the electrophysiological properties of PVMSs have not been fully characterized. In the present study, we investigated the electrophysiological properties of rabbit PVMSs. The results show that addition of 0.5 to 2 mol/l ryanodine to PVMSs uncovers pacing-induced spontaneous activity. Ryanodine at low concentrations locks the sarcoplasmic reticulum (SR) Ca 2 release channel, the ryanodine receptor (RyR), in a subconductance state, causing a Ca 2 -independent Ca 2 release from the SR. 14 Such Ca 2 leakage during diastole may alter intracellular Ca 2 and thus Ca 2 -dependent ionic currents, 15 and this may be the cause of the spontaneous activity. The results may be important clinically, because remodeling of Ca 2 handling proteins, including RyR, occurs in disease states. 16 Methods Rabbits weighing 1.5 to 2.0 kg (Chubu-Kagaku-Shizai, Nagoya, Japan) were anesthetized with pentobarbital (30 to 40 mg/kg IV), and the heart was quickly excised. The right atrium, including the SA node and the junction between the right superior PV and the atria, was isolated and opened to expose the endocardial surface (Figure 1, B and C). The preparation was fixed in a tissue bath equipped with an electrode array. Extracellular recordings were made from the epicardium (simultaneously from 120 sites) by use of the modified bipolar electrode array (interelectrode distance, 1 mm), 17 and in some experiments, membrane potential was recorded by a glass Received September 17, 2002; revision received December 31, 2002; accepted January 7, From RIEM, Nagoya University, Nagoya, Japan (H.H., R.N., M.Y., T.H., K.K., I.K.); School of Biomedical Sciences, University of Leeds, Leeds, UK (M.R.B.); School of Engineering, Tokyo Denki University, Tokyo, Japan (S.I., K.M.); and Tokyo Women s Medical University, Tokyo, Japan (N.S.). Correspondence to Dr Haruo Honjo, Research Institute of Environmental Medicine, Nagoya University, Nagoya , Japan. honjo@riem.nagoya-u.ac.jp 2003 American Heart Association, Inc. Circulation is available at DOI: /01.CIR BD 1937

2 1938 Circulation April 15, 2003 Figure 1. Ryanodine-induced pacemaker shift. A, Extracellular electrogram from CT and membrane potential from right superior PVMS. Traces speeded up at intervals to show time course of action potential. Ryanodine (2 mol/l) was applied at time shown. B and C, Activation maps at times B and C marked in A. Dots indicate recording sites; *, leading pacemaker site; thick line, block zone; RA, right atrial appendage; SVC and IVC, superior and inferior venae cavae; SAN, SA node; and IAS, interatrial septum. D, Expanded view of section marked by # in A. E, F, and G, Activation maps at times E, F, and G marked in D. microelectrode from the endocardium near the PV-atrium junction (Figure 1, A and D). Alternatively, the MS was isolated from the right superior PV (Figure 2A, box), and membrane potential was recorded during regular pacing and a train of rapid stimuli after a brief rest. Preparations were superfused with Krebs-Ringer solution containing (in mmol/l): NaCl, 4.0 KCl, 1.2 CaCl 2, 1.3 MgSO 4, 1.2 NaH 2 PO 4, 25.2 NaHCO 3, and 11.0 glucose (ph 7.4, equilibrated with 95% O 2 /5% CO 2 )at33 C. When Ni 2 was applied, HEPES-buffered solution gassed with 100% O 2 was used, with a composition of (in mmol/l): NaCl, 4.0 KCl, 1.2 CaCl 2, 1.3 MgCl 2, 5.0 HEPES, and 11.0 glucose (ph 7.4). Ryanodine (Sigma) was dissolved in water to make a stock solution (5 mmol/l). Data are presented as mean SEM. Student s t test and ANOVA were used to test differences, and a difference was considered significant at a value of P Results Ectopic Atrial Pacemaker Right atrial preparations showed regular spontaneous activity, and the center of the SA node was the leading pacemaker under control conditions (n 9). A representative activation map is shown in Figure 1B. For this and 2 other experiments, extracellular recordings were made from the crista terminalis (CT) and transmembrane recordings from the atrial muscle close to the PV-atrium junction (Figure 1A). Excitation propagated from the SA node to the CT, and the spread of excitation toward the interatrial septum was partially blocked (Figure 1B, thick line). The membrane potential recorded from the PV region showed typical atrial-type action potentials, with a stable resting potential (RP) ( mv, n 3) under control conditions. After ryanodine (2 mol/l) was applied, there was a depolarization of the RP (by mv, n 3) and an elevation of the action potential plateau (Figure 1A). In addition, after application of ryanodine, the upstroke of the PV action potential preceded the activation of the CT, whereas before application of ryanodine, it followed (Figure 1A). The activation map obtained at this time shows that the pacemaker was shifted to an ectopic site close to the PV (Figure 1C). Further treatment with

3 Honjo et al Pacing-Induced Spontaneous Activity in PVs 1939 Figure 2. Spontaneous activity recorded from PVMS. A, Photograph of posterior of atria. RA and LA, right and left atrial appendages; R-SVC, right superior vena cava; IVC, inferior vena cava; RS-PV, RI-PV, LS-PV, and LI-PV, right superior, right inferior, left superior, and left inferior PVs. MS was isolated from RS-PV (box). B, Effect of ryanodine (0.5 mol/l) on action potential. Top, Stimuli; bottom, membrane potential. C, Effect of rapid pacing (20 pulses at 3.3 Hz) on membrane potential of PVMS treated with ryanodine. Top, Stimuli; middle, slow time base recording of membrane potential of PVMS treated with 0.5 mol/l ryanodine. Preparation was stimulated at 2 Hz, rested for 1 minute, and then briefly stimulated at 3.3 Hz (20 pulses). Burst of spontaneous action potentials was induced by rapid pacing. Bottom, Superimposed recordings of selected action potentials during rapid pacing (number of stimuli shown). ryanodine caused complex changes in the membrane potential and the activation pattern (# in Figure 1A). Fast-time base recordings of the electrical signals during this period are shown in Figure 1D. The ectopic pacemaker close to the PV stopped beating, and this was associated with a sudden change in the extracellular potential at the CT. The activation map at this time (Figure 1E) shows that there was complete conduction block from the SA node toward the septum. During the quiescent period, the RP gradually hyperpolarized (by mv, n 3) (Figure 1D). The ectopic pacemaker started beating again, and the activation map of the first beat (Figure 1F) shows that the conduction from the SA node to the septum recovered and the ectopic pacemaker was driven by the SA node. The ectopic pacemaker started to depolarize gradually after the resumption of beating (Figure 1D). The extracellular potential of the CT changed suddenly at the 10th beat, and this was associated with a takeover of the ectopic focus as the leading pacemaker (Figure 1G). Such dynamic equilibrium between the SA node and the ectopic pacemaker was repeated continuously (# in Figure 1A). Ryanodine (0.5 to 2 mol/l) caused a similar intermittent shift of the leading pacemaker in 7 of 9 preparations. Histological examination (not shown) confirmed that the ectopic pacemaker corresponded to the wall of the right superior PV. Pacing-Induced Spontaneous Activity To characterize the electrical properties of the ectopic pacemaker without interference from the SA node, the right superior PVMS was isolated (Figure 2A) and membrane potential recorded. The majority of isolated PVMSs (39 of 41 preparations) were electrically quiescent without stimulation under control conditions, and the remaining 2 showed spontaneous activity. Figure 2B shows the membrane potential during regular pacing (2 Hz) before and after application of ryanodine in a quiescent preparation. Under control conditions, normal atrial-type action potentials were elicited from a stable RP in response to stimuli (Figure 2B). Treatment with ryanodine (0.5 mol/l) for 30 minutes resulted in an elevation of the action potential plateau, a depolarization of the RP, and a development of pacemaker depolarization (Figure 2B).

4 1940 Circulation April 15, 2003 Pacing-Induced Spontaneous Activity Changes in Pacing Rate, Hz PV Control: Incidence of Spontaneous Activity Depolarization of Resting Potential, mv Ryanodine, mol/l Incidence of Spontaneous Activity (Duration, s) (2.0) / /39 ( ) / /7 ( ) / /6 ( ) RAA, LAA (2.0) / /7 RAA and LAA indicate right and left atrial appendages. The pacemaker depolarization could be similar to that in the SA node, or it could be a pacing-induced delayed afterdepolarization such as that described in the coronary sinus. 18 These changes were more marked during rapid pacing. Figure 2C shows an example in which a train of rapid stimuli (20 pulses at 3.3 Hz) was applied after a 1-minute rest to a preparation treated with ryanodine (0.5 mol/l). Rest resulted in a large hyperpolarization of the RP (similar hyperpolarization was seen in preparations including the SA node when the ectopic pacemaker became quiescent after application of ryanodine; Figure 1D). The first action potential during the stimulation train after the rest was atrial-like with a stable RP, although the plateau of the action potential was low and short (Figure 2C). During rapid pacing, there was a broadening of the action potential plateau and a depolarization of the RP (Figure 2C). These changes were associated with a development of pacemaker depolarization, and by the end of pacing, the action potential had been converted from a typical atrial-type to an apparently SA node type pacemaker action potential (Figure 2C). As a result, spontaneous action potentials developed from the pacemaker depolarization when pacing was stopped (Figure 2C). The slope of the pacemaker depolarization gradually declined during the train of spontaneous action potentials, and the spontaneous activity eventually ceased. Such behavior is analogous to that seen in preparations including the SA node (Figure 1). Table 1 summarizes the pacing-induced spontaneous activity. A train of 3.3-Hz stimulation (after a 1-minute rest) caused self-terminating spontaneous activity in 82.1% of PVMS preparations after treatment with ryanodine (0.5 to 2 mol/l; Table 1). A brief rest period before rapid pacing was not essential for the spontaneous activity, and an abrupt increase in the pacing rate (from 0.5 or 1.0 to 3.3 Hz) could also induce the activity in PVMSs treated with ryanodine (Table 1). Spontaneous action potentials were never induced by pacing in the normal atrial muscle of the right and left atrial appendages treated with ryanodine (Table 1). Steady-State Action Potentials Development of pacemaker depolarization in PVMSs treated with ryanodine was accompanied by changes in the action potential configuration. We examined rate-dependent changes of the steady-state action potential in 6 PVMSs. Representative examples at 3 stimulation rates before and after application of ryanodine (2 mol/l) are shown in Figure 3A, and changes in action potential parameters are summarized in Figure 3B. Under control conditions, there was a small but significant depolarization of the RP at pacing rates 3.3 Hz (Figure 3B). After treatment with ryanodine (2 mol/l), there was a greater depolarization of the RP and a reduction of the action potential overshoot at high rates (Figure 3B). The ryanodine-induced depolarization of the RP at 2.0 Hz was significantly larger in PVMSs ( mv, n 8) than in the normal atrial muscle ( mv, n 6, P 0.01; Figure 3B). Ryanodine also caused biphasic changes in action potential duration: it was prolonged during pacing at 0.5 to 2 Hz, whereas it was shortened at faster rates (Figure 3B). Effects of Modulation of Intracellular Ca 2 Dynamics The effect of depletion of SR Ca 2 on the pacing-induced spontaneous activity was examined. Figure 4A shows that an Figure 3. Effect of ryanodine on rate dependence of action potential recorded from PVMS. A, Steady-state action potentials of PVMSs at 3 pacing rates before (C) and after treatment with 2 mol/l ryanodine (R). B, Rate-dependent changes of resting potential (RP), action potential overshoot (OS), and action potential duration at 60 mv (APD) before (open circle, n 6) and after treatment with 2 mol/l ryanodine (solid circle, n 6). RP and OS of atrial muscle (at 2.0 Hz) before (open triangle, n 6) and after 2 mol/l ryanodine (solid triangle, n 6) are also shown. *Significantly different from control; #significantly different from value at 0 Hz; significantly different from PVMS.

5 Honjo et al Pacing-Induced Spontaneous Activity in PVs 1941 There are various Ca 2 -regulated ionic currents, 15 and the possible contribution of Na -Ca 2 exchange current and Ca 2 -activated Cl current to the spontaneous activity was investigated. Figure 4B shows the effects of Ni 2 to inhibit the Na -Ca 2 exchanger. 20 Application of Ni 2 (5 mmol/l) abolished the development of the pacemaker depolarization during rapid pacing and spontaneous activity after pacing. It is unlikely that these effects of Ni 2 are the result of its blocking action on Ca 2 current (and a decrease of SR Ca 2 content), because Ni 2 abolished the pacing-induced spontaneous activity (duration, from to 0 s; n 5), whereas a partial block of Ca 2 current by nifedipine (3 mol/l), which caused a similar shortening of the action potential to that induced by Ni 2, had no appreciable effects on the duration of the activity (from to s, n 4). Figure 4C shows that niflumate, a Cl channel blocker, 15 attenuated the development of the pacemaker depolarization induced by rapid pacing, and there were no spontaneous action potentials after pacing. The duration of the pacinginduced spontaneous activity was decreased significantly by the Cl channel blockers, 50 mol/l niflumate or 100 mol/l DIDS from to s(n 7, P 0.01). These results suggest that the contribution of the Na -Ca 2 exchanger is major but that of the Ca 2 -activated Cl channel is also significant in the pacing-induced spontaneous activity. Figure 4. Effect of interventions that affect intracellular Ca 2 handling on pacing-induced spontaneous activity in ryanodinetreated PVMS. A, Stimulation protocol of Figure 2C repeated under control conditions (A1), after treatment with 2 mol/l ryanodine (A2), in presence of 50 mol/l CPA after wash-off of ryanodine (A3), and after wash-off of CPA (A4). B, Same protocol after treatment with 2 mol/l ryanodine (B1) and in presence of 5 mmol/l Ni 2 after wash-off of ryanodine (B2). C, Same protocol after treatment with 2 mol/l ryanodine (Ry) (C1) and in presence of 50 mol/l niflumate after wash-off of ryanodine (C2). Top, stimuli; bottom, membrane potential. inhibition of the SR Ca 2 pump by cyclopiazonic acid (CPA) 19 attenuated the pacing-induced activity. After treatment of a PVMS with ryanodine (2 mol/l) for 30 minutes, rapid pacing resulted in the development of pacemaker depolarization during pacing and repetitive spontaneous action potentials after pacing (Figure 4A2). These effects were not reversed after wash-off of ryanodine. After addition of CPA (50 mol/l) to the PVMS treated with ryanodine, the development of pacemaker depolarization was attenuated, and only a few nondriven action potentials were induced (Figure 4A3). These effects of CPA were partially reversed after its wash-off (Figure 4A4). In 4 PVMSs, CPA (50 mol/l) caused a significant decrease in the duration of the pacing-induced spontaneous activity from to s(P 0.01). In PVMSs pretreated for 30 minutes with thapsigargin (5 to 50 mol/l), another inhibitor of the SR Ca 2 pump, 19 rapid stimulation induced only a few spontaneous action potentials, even after treatment with ryanodine (duration, s; n 4). These results support the hypothesis that pacing-dependent changes in intracellular Ca 2 are involved in the spontaneous activity. Effects of -Adrenergic Stimulation Figure 5 shows that isoproterenol (0.1 mol/l) markedly increased the development of pacemaker depolarization during pacing and prolonged the spontaneous activity after rapid pacing in the PVMSs treated with ryanodine. The duration of the spontaneous activity was significantly increased after application of isoproterenol from to s (n 6, P 0.01). These findings support the hypothesis that the pacing-induced spontaneous activity is the result of changes in intracellular Ca 2 dynamics. Spontaneous action potentials were never induced by rapid pacing in the presence of isoproterenol (0.1 mol/l) in PVMSs without treatment with ryanodine (n 3). Effects of Strophanthidin The effects of strophanthidin were investigated on right atrial preparations including the SA node and isolated PVMS. In all the right atrial preparations tested (n 5), strophanthidin (0.1 mol/l) induced spontaneous activity in an ectopic site close to the PV-atrium junction (Figure 6A), as in the case of ryanodine. Figure 6B shows the effects of strophanthidin (0.1 mol/l) on the isolated PVMS. In the presence of strophanthidin, a train of rapid pacing at 3.3 Hz induced a short burst of spontaneous action potentials (duration, s; n 4), but the activity induced by strophanthidin was not associated with the development of pacemaker depolarization during diastole. Discussion In the present study, we have demonstrated that ryanodine, presumably by altering intracellular Ca 2 dynamics, causes pacing-induced spontaneous activity in the rabbit PVMS.

6 1942 Circulation April 15, 2003 Figure 5. Effect of -adrenergic stimulation on pacing-induced spontaneous activity in ryanodine-treated PVMS. Stimulation protocol of Figure 2C was repeated after treatment with 2 mol/l ryanodine (Ry) and in presence of 0.1 mol/l isoproterenol (ISP) after washoff of ryanodine. Top, stimuli; middle, membrane potential before and after ISP application; bottom, superimposed recordings of selected action potentials during period of rapid pacing (number of stimuli shown) and first spontaneous action potential (*) before and after ISP application. Spontaneous action potentials induced by rapid pacing were associated with pacemaker depolarization during diastole. Various types of spontaneous activities, including SA node like spontaneous action potentials, have been recorded from guinea pig and dog PVs in previous studies However, spontaneous activity was observed under control conditions in the previous studies, whereas in the rabbit, spontaneous activity under control conditions was rare, and interventions affecting intracellular Ca 2 (rapid pacing and ryanodine) were necessary to generate the activity. Electron microscopy of rat PVMSs demonstrated the presence of clear cells possessing the characteristic feature of SA node cells, 7 although such cells have not been found in other species and may be artifacts. 12,13 In addition, immunohistochemical studies of embryonic hearts using specific antibodies for the cardiac conduction system suggest that the PVMSs share a common embryonic origin with the SA node. 21,22 It is therefore possible that the electrophysiological characteristics of PVMSs are similar to those of the SA node. The fact that the spontaneous activity could be induced only in PVMSs and not in the normal atrial muscle (Table 1) suggests that PVMS cells have distinct electrophysiological characteristics. Chen et al 11 reported that canine PVMS cells with spontaneous activity have a significantly lower density of inward rectifier K current, I K1. It is well known that SA node cells lack I K1, and this is essential for their normal pacemaking. 23 The low density of I K1 may also be an essential feature for the ectopic pacemaker activity of the PVMSs. The present study suggests that Na -Ca 2 exchange current along with Ca 2 -dependent Cl current may be involved in the spontaneous activity. It is possible that during diastole, these inward currents are able to cause depolarization of the Figure 6. Triggered activity induced by strophanthidin in PVMS. Top, Activation maps before (control) and after treatment with strophanthidin (0.1 mol/l). Format of activation maps and abbreviations are same as those in Figure 1. *Leading pacemaker site. Bottom, Spontaneous activity induced by rapid pacing in presence of strophanthidin (0.1 mol/l). Preparation stimulated at 0.5 Hz and then at 3.3 Hz (20 pulses). Fast time base recording of spontaneous action potentials shown at right. Top, stimuli; middle, membrane potential.

7 Honjo et al Pacing-Induced Spontaneous Activity in PVs 1943 membrane (and thus spontaneous activity) because of the low density of I K1. The results obtained from rabbit PVMSs may help to explain why AF begets AF. 24 When there is already AF (by any mechanism), the high-frequency activity may provoke the arrhythmogenic activity of the PVMSs and thereby initiate another episode of AF. Clinical electrophysiology studies in patients with AF have demonstrated that rapid focal activity originating in one PV can trigger focal activity in another PV to maintain AF. 25 Atrial tachyarrhythmias, including AF, are more common under pathological conditions, such as hypertrophy, hyperthyroidism, and heart failure, and these conditions result in a remodeling of Ca 2 -handling proteins and thus a modification of intracellular Ca 2 dynamics. 16 For example, heart failure is associated with upregulation of the Na -Ca 2 exchanger and downregulation of SR Ca 2 -ATPase, causing an increase in the propensity for intracellular Ca 2 overload. 26 In addition, it has recently been reported that in failing hearts, RyR molecules are hyperphosphorylated. 27 Hyperphosphorylation of RyR dissociates its stabilizing subunit, FKBP12.6, resulting in altered channel function (increased open probability and subconductance states). 27 This is comparable to the action of low concentrations of ryanodine. It is reported that mutations in RyR result in catecholamine-sensitive tachyarrhythmias. 28 However, there is no evidence at present for remodeling of intracellular Ca 2 handling in PVMSs of AF patients, and therefore, it is uncertain whether the pacing-induced spontaneous activity observed in this study has any relationship to rapid the PV activity seen in AF patients. References 1. Moe GK, Abildskov JA. Atrial fibrillation as a self-sustaining arrhythmia independent of focal discharge. Am Heart J. 1959;58: Jaïs P, Haïssaguerre M, Shah DC, et al. A focal source of atrial fibrillation treated by discrete radiofrequency ablation. Circulation. 1997;95: Haïssaguerre M, Jaïs P, Shah DC, et al. Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. N Engl J Med. 1998;339: Haïssaguerre M, Jaïs P, Shah DC, et al. Catheter ablation of chronic atrial fibrillation targeting the reinitiating triggers. J Cardiovasc Electrophysiol. 2000;11: Chen P-S, Wu T-J, Hwang C, et al. Thoracic veins and the mechanisms of non-paroxysmal atrial fibrillation. Cardiovasc Res. 2002;54: Nathan H, Eliakim M. The junction between the left atrium and the pulmonary veins: an anatomic study of human hearts. Circulation. 1966; 34: Masani F. Node-like cells in the myocardial layer of the pulmonary vein of rats: an ultrastructural study. J Anat. 1986;145: Chen YJ, Chen SA, Chang MS, et al. Arrhythmogenic activity of cardiac muscle in pulmonary veins of the dog: implication for the genesis of atrial fibrillation. Cardiovasc Res. 2000;48: Cheung DW. Pulmonary vein as an ectopic focus in digitalis-induced arrhythmia. Nature. 1981;294: Cheung DW. Electrical activity of the pulmonary vein and its interaction with the right atrium in the guinea-pig. J Physiol. 1981;314: Chen Y-J, Chen S-A, Chen Y-C, et al. Effects of rapid atrial pacing on the arrhythmogenic activity of single cardiomyocytes from pulmonary veins: implication in initiation of atrial fibrillation. Circulation. 2001;104: de Bakker JMT, Ho SY, Hocini M. Basic and clinical electrophysiology of pulmonary vein ectopy. Cardiovasc Res. 2002;54: Hocini M, Ho SY, Kawara T, et al. Electrical conduction in canine pulmonary veins: electrophysiological and anatomical correlation. Circulation. 2001;105: Zucchi R, Ronca-Testoni S. The sarcoplasmic reticulum Ca 2 channel/ ryanodine receptor: modulation by endogenous effectors, drugs and disease states. Pharmacol Rev. 1997;49: Schlotthauser K, Bers DM. Sarcoplasmic reticulum Ca 2 release causes myocardial depolarization: underlying mechanism and threshold for triggered action potentials. Circ Res. 2000;87: Tomaselli GF, Marban E. Electrophysiological remodeling in hypertrophy and heart failure. Cardiovasc Res. 1999;42: Shibata N, Inada S, Mitsui K, et al. Pacemaker shift in the rabbit sinoatrial node in response to vagal nerve stimulation. Exp Physiol. 2001;86: Aronson RS, Cranefield PF, Wit AL. The effects of caffeine and ryanodine on the electrical activity of the canine coronary sinus. J Physiol. 1985;368: Baudet S, Shaoulian R, Bers D. Effects of thapsigargin and cyclopiazonic acid on twitch force and sarcoplasmic reticulum Ca 2 content of rabbit ventricular muscle. Circ Res. 1993;73: Sigekawa M, Iwamoto T. Cardiac Na -Ca 2 exchange: molecular and pharmacological aspects. Circ Res. 2001;88: Gorza L, Vitadello M. Distribution of conduction system fibers in the developing and adult rabbit heart revealed by an antineurofilament antibody. Circ Res. 1989;65: Blom NA, Gittenberger-de Groot AC, DeRuiter MC, et al. Development of the cardiac conduction tissue in human embryos using HNK-1 antigen expression: possible relevance for understanding of abnormal atrial automaticity. Circulation. 1999;99: Boyett MR, Honjo H, Kodama I. The sinoatrial node, a heterogeneous pacemaker structure. Cardiovasc Res. 2000;47: Wijffels MCEF, Kirchhof CJHJ, Dorland R, et al. Atrial fibrillation begets atrial fibrillation: a study in awake chronically instrumented goats. Circulation. 1995;92: Kumagai K, Yasuda T, Tojo H, et al. Role of rapid focal activation in the maintenance of atrial fibrillation originating from the pulmonary veins. Pacing Clin Electrophysiol. 2000;23: Pogwizd SM, Schlotthauer K, Li L, et al. Arrhythmogenesis and contractile dysfunction in heart failure: roles of sodium-calcium exchange, inward rectifier potassium current, and residual -adrenergic responsiveness. Circ Res. 2001;88: Marx SO, Reiken S, Hisamatsu Y, et al. PKA phosphorylation dissociates FKBP12.6 from the calcium release channel (ryanodine receptor): defective regulation in failing hearts. Cell. 2000;101: Prior SG, Napolitano C, Memmi M, et al. Clinical and molecular characterization of patients with catecholaminergic polymorphic ventricular tachycardia. Circulation. 2002;106:69 74.