Pacemaker shift in the rabbit sinoatrial node in response to vagal nerve stimulation

Transcription

1 Pacemaker shift in the rabbit sinoatrial node in response to vagal nerve stimulation Nitaro Shibata *, Shin Inada, Kazuyuki Mitsui, Haruo Honjo, Mitsuru Yamamoto, Ryoko Niwa, Mark R. Boyett and Itsuo Kodama Departments of Circulation and Humoral Regulation, Research Institute of Environmental Medicine, Nagoya University, Nagoya , Japan, *Department of Cardiology, The Heart Institute of Japan, Tokyo Women s Medical College, Tokyo , Japan, School of Engineering, Tokyo Denki University, Tokyo , Japan and School of Biomedical Sciences, University of Leeds, Leeds LS2 9JT, UK (Manuscript received 13 July 2000; accepted 25 January 2001) Effects of brief postganglionic vagal nerve stimulation on the activation sequence of the rabbit sinoatrial (SA) node were investigated. Activation sequences in a small area (7 mm w 7 mm) on the epicardial surface were measured in a beat-tobeat manner using an extracellular potential mapping system composed of 64 modified bipolar electrodes with high-gain and low-frequency band-pass filtering. The leading pacemaker site was recognised clearly from both the activation sequence and the characteristic morphology of the potentials. Vagal stimulation resulted in a short-lasting initial slowing of spontaneous rate followed by a long-lasting secondary slowing; a brief period of relative or absolute acceleration was interposed between the two slowing phases. During these changes of spontaneous rate, the leading pacemaker site shifted in a complex beat-to-beat manner by 1 6 mm alongside the crista terminalis in the superior or inferior direction. For the first spontaneous excitation following stimulation, the greater the slowing, the larger the distance of the pacemaker shift. There was no such linear relationship between the extent of slowing and the distance of pacemaker shift for the subsequent beats. These changes in the leading pacemaker site in response to vagal stimulation may be the result of the functional and morphological heterogeneity of the mammalian SA node in terms of innervation, receptor distribution and ion channel densities. Experimental Physiology (2001) 86.2, The origin of the spontaneous action potential in the mammalian sinoatrial (SA) node is not static: it is dynamic and changing according to the prevailing conditions (Schuessler et al. 1996). In humans, changes in P wave morphology occur under a variety of physiological and pathological situations (e.g. Irisawa & Seyama, 1966; Grossman & Delman, 1969; Boineau et al. 1988). This probably results from changes in the sequence of atrial muscle activation secondary to a change in the exit site from the SA node as a result of a change in the position of the leading pacemaker site in the SA node. The phenomenon of pacemaker shift has mainly been studied in the rabbit, although it has also been observed in other species such as human and dog (Opthof, 1988). In the rabbit, pacemaker shift has been observed with autonomic agonists, premature stimulation, overdrive, changes in temperature, changes in extracellular ions and ion channel block (Bouman et al. 1978; Jalife & Moe, 1979; Mackaay et al. 1980; Boyett et al. 2000). Because of the morphological and functional heterogeneity of the SA node, many interventions are considered to affect different regions of the node differentially and as a consequence the site showing the earliest excitation is altered. Pacemaker shift in response to vagal nerve stimulation has been seen in rabbit and dog (Bouman et al. 1968; Spear et al. 1979; Mackaay et al. 1980; Boineau et al. 1983). These previous studies, in which a single exploring electrode was used, were unable to determine beat-to-beat changes of the activation sequence of the SA node. Brown & Eccles (1934) first showed in the cat that brief vagal stimulation results in a complex change in heart rate: an initial decrease in rate followed by a relative or absolute acceleration and, finally, a secondary decrease in rate. This observation has been confirmed many times thereafter in other species including human (Bouman et al. 1968; Levy et al. 1981; Pappana, 1991; Kodama et al. 1996). This complex heart rate response could be the result of pacemaker shift. The present study was designed to shed light on this issue. We mapped extracellular potentials in a small area (7 mm w 7 mm) on the epicardial surface of rabbit right atria, including the whole SA node, using a newly designed 64-channel modified bipolar electrode Publication of The Physiological Society Corresponding author: ikodama@riem.nagoya-u.ac.jp 2100

2 178 N. Shibata and others Exp. Physiol (MBE) system with high-gain amplification and lowfrequency band-pass filtering. With the aid of this system, we measured the beat-to-beat trajectory of the leading pacemaker site in the SA node in response to brief postganglionic vagal nerve stimulation. METHODS This investigation conforms with the guide for the care and use of laboratory animals published by the US National Institute of Health (NIH Publication No , revised 1996). Preparations Rabbits weighing kg were anaesthetised with sodium pentobarbital (30 40 mg kg _1, I.V.) and the heart was rapidly excised. The right atrium was separated from the rest of the heart and opened by a longitudinal incision in the free wall to expose the endocardial surface. The atrium was then trimmed to leave a preparation (~15 mm w 15 mm), which included the whole SA node and some of the surrounding atrial muscle. The preparation was fixed endocardial surface up in a tissue bath equipped with a 64-electrode grid on the bottom, and superfused with Krebs-Ringer solution gassed with 95 % O 2 5 % CO 2 at 34 ºC. The composition of the solution was as follows (mm): NaCl 120.3, KCl 4.0, CaCl 2 1.2, MgSO 4 1.3, NaH 2 PO 4 1.2, NaHCO , glucose 11.0; ph 7.4. Recording Extracellular potentials were recorded from 64 sites on the epicardial surface with MBEs. Sixty-four pairs of MBEs were arranged as an 8 w 8 array with an interelectrode distance of 1 mm to cover a 7 mm w 7 mm square. The configuration of each pair of MBEs was the same as reported previously (Yamamoto et al. 1998). In brief, MBEs were made of two 100 µm stainlesssteel wires insulated to the tip. The indifferent electrode of each pair was located 1 mm below the point at which the active electrode contacted the epicardium. The potential difference between the two electrodes was recorded, and the signal was amplified (gain, 80 db) and filtered (band-pass filter, Hz). The 64 signals were acquired and monitored using a computer. Activation time at the recording site was identified by the initial negative deflection in each electrogram, which was recognised by an electronic differentiation. The site showing earliest activation was taken to be the leading pacemaker site. At the start of each experiment, one pair of exploring MBEs was positioned on the endocardial surface using a calibrated X,Y, Z- micromanipulator with 0.1 mm precision to measure the dimensions of the preparation by recording the co-ordinates of various anatomical landmarks and the location of the epicardial electrode grid recognised through the very thin and transparent tissue). In this way the drawing and the recording sites shared common co-ordinates. Maps of the excitation spread from the leading pacemaker site were constructed with 5 ms isochrones by off-line analysis. Figure 1 Spread of spontaneous excitation and the morphology of extracellular potentials. Left, activation pattern of entire preparation (endocardial view) and recording sites of extracellular potentials (0; Ch., channel number). CT, crista terminalis; SVC, superior vena cava; IVC, inferior vena cava; SEP, interatrial septum; RA,right atrial appendage; RSARB,right branch of the sinoatrial ring bundle; LSARB, left branch of the sinoatrial ring bundle; * position of the leading pacemaker site; large filled circles, position of a pair of silver-wire electrodes to stimulate the vagal nerve terminals. Right, extracellular electrograms recorded from 64 sites. The electrograms are labelled with channel numbers.

3 Exp. Physiol Vagal stimulation-induced pacemaker shift 179 Vagal stimulation The method for stimulation of the vagal nerve terminals was essentially the same as employed in our previous studies (Boyett et al. 1995; Kodama et al. 1996; Yamamoto et al. 1998). In brief, a pair of silver wire (1 mm in diameter) electrodes (insulated to the tip) with an interelectrode distance of 4 mm was placed on the endocardial surface of the preparation. One electrode was sited on the atrial muscle and the other on the SA node close to the leading pacemaker site. We found in pilot experiments that this electrode arrangement is the most efficient to induce the negative chronotropic response to postganglionic vagal nerve stimulation. A train of 5 80 pulses (monophasic pulses of 100 µs in duration at 200 Hz) was applied during the spontaneous action potential at the leading pacemaker site (the train was triggered by the signal from a pair of MBEs placed on the endocardial surface of the atrial muscle opposite the leading pacemaker site). The intensity of the stimuli was adjusted (10 20 V) to a level subthreshold for excitation of atrial or nodal cells, but sufficient to stimulate postganglionic nerve terminals. Stimulation of b-receptors by noradrenaline released from sympathetic nerves was prevented by the presence of propranolol (1 µm) in the superfusate. Data analysis Data are presented as mean ± S.E.M. unless otherwise specified. Analysis of variance was used to test differences and a difference was considered significant at P <0.05. RESULTS Under control conditions, the activation sequence of spontaneous excitation on the epicardial surface showed a similar pattern in five preparations studied. Representative results are shown in Fig. 1. Cycle length of spontaneous excitation of this preparation was constant at ms. The site of earliest activation (leading pacemaker site) was located in the intercaval region 1 mm away from the medial Figure 2 Spread of spontaneous excitation and morphology of extracellular potentials before and after vagal nerve stimulation with 80 pulses at 200 Hz. Seven panels at the top are the activation sequences (endocardial view) of the spontaneous beat immediately before vagal stimulation (VS0) and the first (VS1), second (VS2), third (VS3), fourth (VS4), eighth (VS8) and ninth (VS9) beats following the stimulation. Abbreviations are the same as in Fig. 1 and the crista terminalis is shown shaded. * Position of the leading pacemaker site. A beat-to-beat trajectory of the leading pacemaker site after the vagal stimulation is illustrated in the right lower panel. The trace at the bottom is the electrogram recorded by Ch.12 closest to the original leading pacemaker site.

4 180 N. Shibata and others Exp. Physiol Figure 3 Changes in extracellular potentials following vagal stimulation with 40 pulses before (top trace) and after (bottom trace) the application of atropine (2 µm). Electrograms recorded from atrial muscle in the crista terminalis are shown. Note the change in the configuration of the electrogram after vagal stimulation before but not after the application of atropine; this is indicative of pacemaker shift before but not after the application of atropine. border of the crista terminalis. From here, the excitation wave was propagated preferentially in an oblique cranial direction towards the crista terminalis. The spread of excitation towards the interatrial septum was very slow or almost blocked. Extracellular potentials recorded from the epicardial surface had a variety of morphologies. In a small central area at the leading pacemaker site (Fig. 1, Ch.20, asterisk), a slow primary negative deflection was preceded by a gradual increase of the negativity, and was followed by a second slow and smaller negative wave. At the periphery of the SA node superior and medial to the leading pacemaker site, slow positive/negative waves were recorded (Fig. 1, Ch.29, 36, 44). On the septal side of the SA node with slow conduction, long, slow positive waves were recorded (Fig. 1, Ch.18, 26, 50). In the atrial muscle surrounding the SA node, the extracellular potentials showed a sharp positive wave followed by a short negative wave. These characteristics of the activation sequence and morphology of epicardial extracellular potentials in and around the SA node are the same as those recorded from the endocardial surface, which we reported previously (Yamamoto et al. 1998). Figure 2 shows an experiment with vagal stimulation. Brief vagal stimulation (80 pulses) resulted in an initial increase in cycle length (from 437 ms in control to 1029 ms at Ch.12), which was followed by a decrease in cycle length (beats VS1 VS2) and then a secondary increase of cycle length Figure 4 Effects of the number of vagal stimulation pulses on the prolongation of spontaneous cycle length and the distance of pacemaker shift for the first spontaneous beat (VS1) following the stimulation. A, increase of cycle length plotted against the pulse number. B,distance of pacemaker shift from the original site plotted against the pulse number. Values are mean ± S.E.M. (the number of observations is shown in parentheses).

5 Exp. Physiol Vagal stimulation-induced pacemaker shift 181 (VS6 VS10). The seven maps illustrate the activation patterns of spontaneous beats before (VS0) and after (VS1 VS4, VS8, VS9) vagal stimulation. In the control (VS0), the leading pacemaker was located at Ch.12 in the intercaval region ~1 mm medial to the crista terminalis with the usual activation pattern (preferential spread towards the crista terminalis in an oblique cranial direction). In the next two beats immediately after vagal stimulation (VS1, VS2), the earliest excitation appeared at Ch.60 (~6 mm superior to the original leading pacemaker), and the excitation spread towards the crista terminalis in an oblique caudal direction. A more complex activation pattern suggesting dual leading pacemaker sites (close to the new and original sites) was observed in VS3. Activation patterns in the subsequent five beats (VS4 VS8) were essentially similar to those in VS1 and VS2. At the ninth beat after vagal stimulation (VS9), the leading pacemaker site returned suddenly to the original site. The morphology of the extracellular potentials also returned to the control configurations (Fig. 2, bottom trace). All the effects of vagal stimulation could be inhibited by 2 µm atropine (Fig. 3). In all five preparations tested, vagal stimulation always resulted in a short-lasting slowing of spontaneous excitation followed by a long-lasting secondary slowing; a brief period of relative or absolute acceleration was interposed between the two slowing phases (Fig. 2, bottom trace). There was some fluctuation of cycle length at this period (Fig. 2, VS2 VS4). The larger the number of stimulation pulses, the greater the slowing (Fig. 4A; P<0.0001). For the first beat after vagal stimulation (VS1), the shift of the leading pacemaker site tended to be greater with a larger number of stimulation pulses (Fig. 4B; P=0.133). Figure 5 summarises the pacemaker shift of VS1 in the five preparations. The shift was plotted using orthogonal axes crossing at the original leading pacemaker site; the superior to inferior direction is along the ordinate and the lateral to medial direction (towards the right atrial appendage and interatrial septum, respectively) is along the abscissa. A total of 31 vagal stimulation tests were divided into three groups based on the change of the first cycle length ( SCL); 9 with SCL < 250 ms, 12 with SCL > 250 ms but < 500 ms, and 10 with SCL > 500 ms. The pacemaker shift occurred along the superior inferior axis, and preferentially towards the superior vena cava. The larger the SCL, the greater the distance of the pacemaker shift; the average distance was 0.5 ± 0.2 mm with short SCL, 2.0 ± 0.5 mm with medium SCL, and 4.2 ± 0.6 mm with long SCL (P< vs. short 9SCL; P<0.005 vs. medium 9SCL). For the subsequent beats after VS1, however, there was no simple relationship between the change of cycle length and the distance of pacemaker shift. Figure 6 shows beat-to-beat changes of cycle length and distance of pacemaker shift plotted against beat number after vagal stimulation. Figure 6A shows an example obtained from a preparation following stimulation with 80 pulses, and average data from five preparations following stimulation with 5 and 80 pulses are presented in Fig. 6B. In contrast to the complex change in cycle length (initial short-lasting, large amplitude slowing followed by an acceleration and then a long-lasting, moderate slowing), the pacemaker shift was simpler: the distance was at an almost constant level from VS1 to VS8. There may be different mechanisms involved in the pacemaker shift in VS1 and subsequent beats (see Discussion). Figure 5 Pacemaker shift of the first spontaneous beat (VS1) following vagal stimulation with 5 80 pulses. The shift is plotted with orthogonal axes crossing at the original leading pacemaker site; the superior to inferior direction is along the ordinate and the lateral to medial direction is along the abscissa. Abbreviations are the same as in Fig. 1. A total of 31 vagal stimulation tests were divided into three groups based on the change of the first post-stimulation cycle length ( SCL). A, 9 with SCL < 250 ms; B, 12 with SCL > 250 ms but < 500 ms; C, 9 with SCL > 500 ms.

6 182 N. Shibata and others Exp. Physiol DISCUSSION In the present study we used a custom-made extracellular potential mapping system: a 64-channel modified bipolar electrode system. This system is an extension of our singlechannel modified bipolar electrode system employed for sequential endocardial SA node potential mapping (Yamamoto et al. 1998). In the SA node region, the amount of volume conductor is small and the electrical coupling between cells is weak compared to those in atrial muscle (Anumonwo et al. 1992; Coppen et al. 1999). Accordingly, extracellular potential changes localised in (?) the SA node are easily interrupted or masked by relatively large far-field potentials from the atrial muscle mass surrounding the SA node (Hoffman, 1998). Our modified bipolar electrodes were designed to minimise such far-field potentials by a high level of common mode rejection with a good preservation of local potentials (Yamamato et al. 1998). Validation of technique The variety of morphologies of extracellular potentials recorded from the epicardial surface, and the activation sequence under control conditions, are similar to those obtained by endocardial surface mapping using a pair of roving modified bipolar electrodes (Yamamato et al. 1998). For instance, the epicardial electrograms recorded from the leading pacemaker site, like endocardial electrograms from the leading pacemaker site, showed characteristic slow dual negative deflections normally preceded by a gradual increase of negativity. The rabbit SA node is a superficial thin structure with dimensions of ~6 mm w 2 mm just beneath the endocardium, and occupies the entire thickness ( mm) between endocardium and epicardium within the intercaval region (Bleeker et al. 1980; Opthof, 1988; Coppen et al. 1999; Boyett et al. 2000). The extracellular potentials recorded from the endocardial and epicardial surfaces may therefore reflect almost identical two-dimensional local current distribution during the spread of spontaneous excitation. Comparison with previous studies In response to brief vagal stimulation, the spontaneous rate showed triphasic changes as reported previously (Brown & Eccles, 1934; Bouman et al. 1968; Levy et al. 1981; Pappano, 1991; Kodama et al. 1996): a short-lasting initial slowing was followed by a relative or absolute acceleration, and finally a long-lasting secondary slowing. During these changes in spontaneous rate, the leading pacemaker site shifted by 1 6 mm alongside the crista terminalis mostly towards the superior vena cava, but sometimes towards the inferior vena cava, with a complex beat-to-beat trajectory. In the rabbit and dog, vagal stimulation has been reported to cause a shift of the leading pacemaker site in the SA node (Bouman et al. 1968; Spear et al. 1979; Boineau et al. 1983). Spear et al. (1979) showed a minimal shift ( mm) in the superior direction Figure 6 Beat-to-beat changes of the spontaneous cycle length and the distance of pacemaker shift following vagal stimulation. A, change of spontaneous cycle length and distance of pacemaker shift in a preparation following vagal stimulation with 80 pulses plotted against the beat number after the stimulation (the data were obtained from the same experiment as shown in Fig. 2). B, average percentage changes of spontaneous cycle length and the distance of pacemaker shift from the original leading pacemaker site in the nine beats following vagal stimulation with 5 (0) and 80 (ª) pulses. Values are mean ± S.E.M. (n=5).

7 Exp. Physiol Vagal stimulation-induced pacemaker shift 183 in the rabbit, whereas Boineau et al. (1983) demonstrated a shift in the inferior direction (~10 mm) in the dog. However, no further topological information (exact distance of the shift and beat-to-beat trajectory of the new leading pacemaker site) was provided by these previous investigators because of technical limitations. Mechanism of pacemaker shift The present data have revealed that the greater the initial slowing, the larger the shift of the leading pacemaker site for the first post-stimulation beat (VS1) (Figs 4 and 5). However, there was no such linear relationship between the extent of slowing and the distance of pacemaker shift for the several subsequent beats; as shown in Figs 2 and 6A, the leading pacemaker site remained in a similar position during the acceleration and the secondary slowing phases. Different mechanisms may therefore be involved in the pacemaker shift for the first beat and the subsequent beats. The leading pacemaker is the site showing the fastest spontaneous activity. In response to an intervention that decreases spontaneous activity, the leading pacemaker is expected to shift to the site with intrinsic spontaneous activity that is least suppressed. Acetylcholine (ACh) released from parasympathetic vagal nerve terminals affects the heart by binding to the M 2 muscarinic receptor, resulting in an activation of ACh-activated K + current (I K,ACh ), a hyperpolarising shift of the activation curve of the hyperpolarisation-activated inward current (I f ), and an inhibition of L-type Ca 2+ current (I Ca,L ) (Boyett et al. 1995, 2000; DiFrancesco et al. 1995). On the basis of experimental studies and computer modelling, Boyett et al. (1995, 2000) have suggested that the slowing of the spontaneous rate in the SA node by ACh is primarily the result of the activation of I K,ACh, and that the reductions in I f and I Ca,L play a subsidiary role. The pacemaker shift could be the result of a regional difference in the vagal innervation, M 2 receptor density or density of I K,ACh. It is not known whether there is a regional difference in the density of I K,ACh. However, the density of vagal innervation of the SA node is greater than that of atrial muscle (Löffelholz & Pappano, 1985). In the rabbit SA node, fine nerve processes form a basket or nest around the pacemaker cells at the normal leading pacemaker site, but there are few or no visible fibres in the periphery of the SA node (Roberts et al. 1989). The density of the M 2 receptor measured in the dog SA node was ~18 and ~29% higher at the normal leading pacemaker site than in adjacent superior and inferior regions, respectively (Beau et al. 1995). The pacemaker shift could be the result of the regional difference in either vagal innervation or M 2 receptor density. The mechanism of pacemaker shift for the subsequent beats can only be speculated upon. Following brief vagal nerve stimulation, the shortening of the action potential is longer lasting in the peripheral, superior and inferior regions of the SA node than in the SA node centre (Kodama et al. 1996). This may enhance the intrinsic pacemaker activity in these regions thus overcoming the direct inhibitory effect of vagal stimulation. A sudden prolongation of cycle length and hyperpolarisation may cause a secondary increase in I f not fully offset by the hyperpolarising shift of the I f activation curve by M 2 receptor stimulation (Boyett et al. 1995). This may preferentially facilitate the pacemaker activity in the periphery, since activation of I f is considered to play a more important role in the spontaneous excitation of peripheral SA node cells than that of central cells (Honjo et al. 1996; Nikmaram et al. 1997; Boyett et al. 2000). Alterations of intercellular coupling induced by vagal stimulation (Duivenvoorden et al. 1992) may also contribute to the pacemaker shift through a modulation of mutual entrainment among different types of pacemaker cells in the SA node (Anumonwo et al. 1992). In the present study, the extracellular potentials were recorded from the epicardial surface. In the rabbit, the periphery of the SA node close to the crista terminalis is separated from the epicardium by atrial muscle (Coppen et al. 1999). Accordingly, the apparent shift of the leading pacemaker site within the SA node could, theoretically, reflect an alteration of exit site from the SA node tissue to the atrial muscle as reported in larger animal species. Further experimental studies will be required to clarify this point. In conclusion, the pacemaker shift within the SA node in response to vagal nerve stimulation may be the result of the functional and morphological heterogeneity of the pacemaker tissue and different mechanisms may be involved in the initial and later slowing phases of the spontaneous rate. ANUMONWO, J. M. B., WANG, H.-Z., TRABKA-JANIK, E., DUNHAM, B., VEENSTRA, R. D., DELMAR, M. & JALIFE, J. (1992). Gap junctional channels in adult mammalian sinus nodal cells. Immunolocalization and electrophysiology. Circulation Research 71, BEAU, S. L., HAND, D. E., SCHUESSLER, R. B., BROMBERG, B. I., KWON, B., BOINEAU, J. P. & SAFFITZ, J. E. (1995). Relative densities of muscarinic cholinergic and b-adrenergic receptors in the canine sinoatrial node and their relation to sites of pacemaker activity. Circulation Research 77, BLEEKER, W. K., MACKAAY, A. J. C., MASSON-PEVET, M., BOUMAN, L. N. & BECKER, A. E. (1980). Functional and morphological organization of the rabbit sinus node. Circulation Research 46, BOINEAU, J. P., CANAVAN, T. E., SCHUESSLER, R. B., CAIN, M. E., CORR, P. B. & COX, J. L. (1988). Demonstration of a widely distributed atrial pacemaker complex in the human heart. Circulation 77, BOINEAU, J. P., SCHUESSLER, R. B., ROESKE, W. R., AUTRY, L. J., MILLER, C. B. & WYLDS, A. C. (1983). Quantitative relation between sites of atrial impulse origin and cycle length. American Journal of Physiology 245, H BOUMAN, L. N., GERLINGS, E. D., BIERSTEKER, P. A. & BONKE, F. I. (1968). Pacemaker shift in the sino-atrial node during vagal stimulation. Pflügers Archiv 302, BOUMAN, L. N., MACKAAY, A. J. C., BLEEKER, W. K. & BECKER, A. E. (1978). Pacemaker shifts in the sinus node. Effects of vagal stimulation, temperature and reduction of extracellular calcium. In The Sinus Node. Structure, Function and Clinical Relevance, ed. BONKE, F. I. M., pp Martinus Nijhoff, The Hague, Boston, London.

8 184 N. Shibata and others Exp. Physiol BOYETT, M. R., HONJO, H. & KODAMA, I. (2000). The sinoatrial node, a heterogeneous pacemaker structure. Cardiovascular Research 47, BOYETT, M. R., KODAMA, I., HONJO, H., ARAI, A., SUZUKI, R. & TOYAMA, J. (1995). Ionic basis of the chronotropic effect of acetylcholine on the rabbit sinoatrial node. Cardiovascular Research 29, BROWN, G. L. & ECCLES, J. C. (1934). The action of a single vagal volley on the rhythm of the heart beat. Journal of Physiology 82, COPPEN, S. R., KODAMA, I., BOYETT, M. R., DOBRZYNSKI, H., TAKAGISHI, Y., HONJO, H., YEH, H.-I. & SEVERS, N. J. (1999). Connexin45, a major connexin of the rabbit sinoatrial node, is coexpressed with connexin43 in a restricted zone at the nodal-crista terminalis border. Journal of Histochemistry and Cytochemistry 47, DIFRANCESCO, D., MANGONI, M. & MACCAFERRI, G. (1995). The pacemaker current in cardiac cells. In Cardiac Electrophysiology. From Cell to Bedside, 2nd edn, ed. ZIPES, D. P. & JALIFE, J., pp W. B. Saunders, Philadelphia, London, Toronto, Montreal, Sydney, Tokyo. DUIVENVOORDEN, J. J., BOUMAN, L. N., OPTHOF, T., BUKAUSKAS, F. F. & JONGSMA, H. J. (1992). Effect of transmural vagal stimulation on electrotonic current spread in the rabbit sinoatrial node. Cardiovascular Research 26, PAPPANO, A. J. (1991). Vagal stimulation of the heartbeat. Muscarinic receptor hypothesis. Journal of Cardiovascular Electrophysiology 2, ROBERTS, L. A., SLOCUM, G. R. & RILEY, D. A. (1989). Morphological study of the innervation pattern of the rabbit sinoatrial node. American Journal of Anatomy 185, SCHUESSLER, R. B., BOINEAU, J. P. & BROMBERG, B. I. (1996). Origin of the sinus impulse. Journal of Cardiovascular Electrophysiology 7, SPEAR, J. F., KRONHAUS, K. D., MOORE, E. N. & KLINE, R. P. (1979). The effect of brief vagal stimulation on the isolated rabbit sinus node. Circulation Research 44, YAMAMOTO, M., HONJO, H., NIWA, R. & KODAMA, I. (1998). Lowfrequency extracellular potentials recorded from the sinoatrial node. Cardiovascular Research 39, Acknowledgements We thank Dr Tobias Opthof (University Medical Center Utrecht, The Netherlands) for his helpful comments on the manuscript. GROSSMAN, J. I. & DELMAN, A. J. (1969). Serial P wave changes in acute myocardial infarction. American Heart Journal 77, HOFFMAN, B. F. (1998). The sinus node electrogram rediscovered. Cardiovascular Research 39, HONJO, H., BOYETT, M. R., KODAMA, I. & TOYAMA, J. (1996). Correlation between electrical activity and the size of rabbit sinoatrial node cells. Journal of Physiology 496, IRISAWA, H. & SEYAMA, I. (1966). The configuration of the P wave during mild exercise. American Heart Journal 71, JALIFE, J. & MOE, G. K. (1979). Phasic effects of vagal stimulation on pacemaker activity of the isolated sinus node of the young cat. Circulation Research 45, KODAMA, I., BOYETT, M. R., SUZUKI, R., HONJO, H. & TOYAMA, J. (1996). Regional differences in the response of the isolated sinoatrial node of the rabbit to vagal stimulation. Journal of Physiology 495, LEVY, M. N., MARTIN, P. J. & STESSE, S. L. (1981). Neural regulation of the heart beat. Annual Reviews in Physiology 43, LÖFFELHOLZ, K. & PAPPANO, A. J. (1985). The parasympathetic neuroeffector junction of the heart. Pharmacological Reviews 37, MACKAAY, A. J., OPTHOF, T., BLEEKER, W. K., JONGSMA, H. J. & BOUMAN, L. N. (1980). Interaction of adrenaline and acetylcholine on cardiac pacemaker function. Functional inhomogeneity of the rabbit sinus node. Journal of Pharmacology and Experimental Therapeutics 214, NIKMARAM, M. R., BOYETT, M. R., KODAMA, I., SUZUKI, R. & HONJO, H. (1997). Variation in the effects of Cs +, UL-SF-49 and ZD-7288 within the sinoatrial node. American Journal of Physiology 271, H OPTHOF, T. (1988). The mammalian sinoatrial node. Cardiovascular Drugs and Therapy 1,