Sinus node dysfunction following targeted disruption of the murine cardiac sodium channel gene Scn5a
1 University Laboratory of Physiology, University of Oxford, Oxford, OX1 3PT UK
2 Cardiovascular Group, Departments of Biochemistry and Physiology, University of Cambridge, Tennis Court Road, Cambridge CB2 1QW UK
3 INSERM U533, Laboratoire de Physiopathologie et Pharmacologie Cellulaires et Moléculaires, Faculté de Médecine, Nantes, France
4 Union Hospital, Huazhong University of Science and Technology, Wuhan, The People's Republic of China
, http://www.100md.com
5 First Hospital, Xi'an Medical School, Xi'an Jiaotong University, The People's Republic of China
6 Biological Physics Group, Physics Department, UMIST, Manchester, M60 1QD UK
7 Electrophysiology and Biophysics Program, Victor Chang Cardiac Research Institute, 384 Victoria Street, Darlinghurst, NSW2010, Australia
Abstract
We have examined sino-atrial node (SAN) function in hearts from adult mice with heterozygous targeted disruption of the Scn5a gene to clarify the role of Scn5a-encoded cardiac Na+ channels in normal SAN function and the mechanism(s) by which reduced Na+ channel function might cause sinus node dysfunction. Scn5a+/– mice showed depressed heart rates and occasional sino-atrial (SA) block. Their isolated peripheral SAN pacemaker cells showed a reduced Na+ channel expression and slowed intrinsic pacemaker rates. Wild-type (WT) and Scn5a+/– SAN preparations exhibited similar activation patterns but with significantly slower SA conduction and frequent sino-atrial conduction block in Scn5a+/– SAN preparations. Furthermore, isolated WT and Scn5a+/– SAN cells demonstrated differing correlations between cycle length, maximum upstroke velocity and action potential amplitude, and cell size. Small myocytes showed similar, but large myocytes reduced pacemaker rates, implicating the larger peripheral SAN cells in the reduced pacemaker rate that was observed in Scn5a+/– myocytes. These findings were successfully reproduced in a model that implicated iNa directly in action potential propagation through the SAN and from SAN to atria, and in modifying heart rate through a coupling of SAN and atrial cells. Functional alterations in the SAN following heterozygous-targeted disruption of Scn5a thus closely resemble those observed in clinical sinus node dysfunction. The findings accordingly provide a basis for understanding of the role of cardiac-type Na+ channels in normal SAN function and the pathophysiology of sinus node dysfunction and suggest new potential targets for its clinical management.
, 百拇医药
Introduction
The sino-atrial node (SAN) initiates normal rhythmic cardiac excitation. SAN automaticity in turn requires a spontaneous diastolic depolarization within its contained pacemaker cells that is the result of complex reciprocal interactions between decays in outward delayed rectifier K+ currents (iK) on the one hand and activation of inward Na+-dependent background (ib,Na), hyperpolarization-activated inward (if), T-type (iCa,T) and L-type Ca2+ (iCa,L) and sustained inward currents (ist) on the other (Irisawa et al. 1993; Boyett et al. 2000; Mitsuiye et al. 2000). Na+–K+ pump (iNaK) and Na+–Ca2+ exchanger currents (iNaCa) and intracellular Ca2+ handling also contribute (Irisawa et al. 1993; Boyett et al. 2000; Bogdanov et al. 2001). Inward voltage-dependent Na+ current, iNa, has been recorded from SAN cells (Nathan, 1986; Irisawa et al. 1993; Honjo et al. 1996; Mangoni & Nargeot, 2001; Cho et al. 2003). However possible contributions from voltage-dependent Na+ currents, iNa, to SAN pacemaker function remain uncertain: iNa occurs only in some pacemaker cells and in any case may inactivate at relatively positive potentials (Nathan, 1986; Irisawa et al. 1993; Honjo et al. 1996).
, http://www.100md.com
There has nevertheless been significant recent progress in our molecular understanding of voltage-gated Na+ channels. These contain both pore-forming -(Nav) and auxiliary -subunits (Catterall, 2000; Goldin, 2001). Thus far, 10 -subunits, with tissue-specific distributions and distinct pharmacological properties, and four -subunits have been identified (Goldin, 2001; Yu & Catterall, 2003). The tetrodotoxin (TTX)-resistant Nav1.5 isoform accounts for most of the cardiac Na+ channels in the heart; however, recent work has also demonstrated TTX-sensitive neuronal (Nav1.1, Nav1.3 and Nav1.6) isoforms (Malhotra et al. 2001; Maier et al. 2002, 2003). The SAN expresses both TTX-sensitive neuronal Na+ channels, whose block slows heart rate (Maier et al. 2003), and TTX-resistant cardiac Na+ channels (Lei et al. 2004). However, these channels may be differentially expressed in different SAN regions compatible with their having contrasting contributions to rhythm generation and/or propagation (Lei et al. 2004).
, 百拇医药
Loss-of-function mutations in the SCN5A encoding cardiac Nav1.5 Na+ channels were initially described in patients with the Brugada and Lev-Lenègre syndromes (Tan et al. 2001; Antzelevitch et al. 2003) but have now been reported in kindreds with familial sinus node dysfunction (SND) (Benson et al. 2003). Specific mechanisms underlying the observed sinus bradycardia, sinus arrest and/or sino-atrial block with which they are associated, whether involving abnormal automaticity, exit block, or impaired intra-atrial conduction and excitability, remain uncertain. Nevertheless, some patients with SCN5A mutations associated with long QT syndrome type 3 (LQT3) also show evidence of SND (Veldkamp et al. 2003) compatible with roles for cardiac-type Na+ channels in normal SAN function. Furthermore, heterologous expression of recombinant Na+ channels predicted from clinical studies in non-excitable cells suggest that Na+ conductance is reduced by SCN5A mutations causing SND (Benson et al. 2003).
, 百拇医药
The work described in this paper explores possible roles for Na+ channels in the normal function of the SAN and mechanisms by which Na+ channel defects might cause SND in pacemaker myocytes using a genetic model system. We have recently established mice with a null mutation in Scn5a whose homozygous disruption causes intrauterine lethality; heterozygotes (Scn5a+/–) are viable but show defects suggesting altered Na+ channel function including impaired atrioventricular conduction, delayed intramyocardial conduction, and ventricular tachycardia with characteristics of re-entrant excitation (Papadatos et al. 2002).
, 百拇医药
Methods
Scn5a-deficient and normal (WT) mice
Mice with targeted disruption of Scn5a were generated as previously described (Papadatos et al. 2002). The experiments used adult mice aged between 2.5 and 5 months for long-term ECG recordings and 3–10 months for single cell and intact SA node studies. Mice were inbred onto a 129/Sv genetic background. Comparisons were made unless otherwise stated between equal numbers of male and female Scn5a+/– mutants and WT littermates. All animal procedures conformed with the United Kingdom Animals (Scientific Procedures) Act 1986.
, 百拇医药
Electrocardiographic (ECG) recordings
Numerical recordings of lead I–III surface ECGs (EMKA technologies, Paris, France) used 25-gauge subcutaneous electrodes and ECG channels bandpass-filtered between 0.5 and 250 Hz in adult anaesthetized (15 mg kg–1 intraperitoneal etomidate) mice. The long-term ECG recordings introduced subcutaneous telemetric transmitters (TA10EA-F20, Data Sciences International, St Paul, MN, USA), with paired wire bipolar electrodes placed over the thorax, into a posterior incision made under anaesthesia (intraperitoneal etomidate, 8 mg kg–1; thiopental, 40 mg kg–1). Experiments began at least 4 days after surgical recovery with mice housed in individual cages in thermostatically controlled rooms subject to 12 h light–dark cycles (light: 7: 00 am to 7: 00 pm) with free access to food and water. ECG signals were recorded telemetrically following analog-to-digital conversion (Data Sciences International).
, 百拇医药
Electrophysiological recordings from single cells
Individual SAN cells isolated from hearts of adult mice (either sex) (Lei et al. 2004) were initially perfused with normal Tyrode solution (mM: 140 NaCl; 5.4 KCl; 1.8 CaCl2; 1 MgCl2; 10 glucose; 5 Hepes; pH 7.4 with NaOH) as described earlier (Lei et al. 2004). The SAN cells were identified by their morphology and spontaneous activity. The studies of iNa reduced NaCl to 67 mM (NaCl replaced by CsCl) and added 300 nM extracellular nisoldipine to block iCa,L. The patch pipette solutions contained (mM): 10 NaCl; 140 KCl; 2 EGTA; 1 MgCl2; 5 MgATP; 10 Hepes (pH 7.3 with KOH for voltage clamping experiments) and 140 KCl; 1.8 MgSO4; 5 Hepes; 0.1 EGTA; titrated to pH 7.3 with KOH with amphotericin (200 μg ml–1) added before the beginning of the current clamping experiments.
, 百拇医药
Cells were superfused at 37°C at 2 ml min–1. Current and voltage clamping used an Axopatch-200B patch clamp amplifier (Axon Instruments, Foster City, CA, USA); whole cell recording was performed as previously described (Lei et al. 2004). Residual uncompensated series resistance prior to compensation was < 5 M. Capacitative current transients were electronically subtracted with series resistance compensation typically at 80%. Current signals were filtered at 10 kHz and sampled at 20 kHz. All current traces were leak-subtracted offline following acquisition and analysis using pCLAMP8 software (Axon Instruments). Cell capacitance (Cm) was obtained from the capacity compensation control of the amplifier following correlation of the whole cell capacity current in response to 5 ms pulses to –70 mV from a holding potential of –60 mV repeated at a frequency of 100 Hz. A previous study has checked the accuracy of this method by integrating the area of the uncompensated capacity current and fitting an exponential function to the decay of the uncompensated capacity current (Honjo et al. 1996). We did not observe significant rundown in either groups of cells over the 20 min duration of our experiments, which were carried out under identical conditions in both groups: in the course of our procedures, the magnitudes of the peak currents at the beginning and end of the protocols were routinely checked.
, 百拇医药
Current–voltage relationships obtained from cells subject to activating steps also yielded normalized activation curves for the peak Na+ conductance, gNa. The corresponding iNa inactivation curves were obtained using standard double pulse protocols repeated at 1 s intervals between series in which conditioning pulses of duration 100 ms were made to potentials between –120 and –20 mV from a –120 mV holding potential; these were then followed by test pulses of 10 ms duration to a fixed membrane potential of –30 mV. Fitting of both the activation and inactivation curves to both the sum of two as well as a single Boltzmann function was attempted using a Levenberg-Marquadt procedure (Press et al. 1986) that performed successive least squares minimizations for the values of each of the parameters of a generalized non-linear function simultaneously over all the experimentally obtained mean values obtained at each test voltage. Successive iterations sought to minimize the values of 2 in the weighted fit using the means and standard errors of the data points. The number of degrees of freedom incorporated not only the number of data points but also the number of variables. The curve fits used a weighting factor for each point that was determined by the inverse of its variance in turn normalized to the average of all such weighting factors. This precaution yielded the maximum likelihood that the fitted function would represent the distribution of the parent data. However, whereas the fits to a single Boltzmann function converged to give half-activation voltages, V1/2, steepness factors, k, and maximum values, those using the sum of two rather than one Boltzmann function did not converge.
, 百拇医药
Electrical mapping of intact SAN
Intact SAN preparations dissected as previously described (Lei et al. 2001, 2002) were superfused with Tyrode solution at 37°C at a rate of 5–6 ml min–1 via a heat exchanger. Intracellular action potentials were recorded using conventional microelectrodes (resistance 30 M); activation times were mapped from extracellular potentials recorded using modified bipolar electrodes placed on 90–100 sites using a calibrated x,y,z-micromanipulator with 0.1 mm precision (Yamamoto et al. 1998). This adopted the output from similar, reference, bipolar electrodes placed on the atrial muscle near the crista terminalis as reference signal. Electrical signals were digitized at 5 kHz by a DigiData 1322A analog to digital converter (Axon Instruments Inc.) and stored on a computer for later analysis. Activation times were given by the time interval separating the initial negative deflections of electrograms from the exploring and reference electrodes with the site showing the earliest activation taken to be the leading pacemaker site. Such electrodes were also used to map the dimensions of the preparation at the outset of each experiment by recording the coordinates of various anatomical landmarks. This permitted a coordinate mapping for the excitation spread from the leading pacemaker site using SigmaPlot 9.0 (Systat Software Inc., Richmond, CA, USA), drawing isochrones at 5 ms intervals.
, 百拇医药
Immunohistochemistry
Immunohistochemistry was performed on tissue sections from adult mice (either sex) as described earlier (Lei et al. 2004). Sections were incubated with rabbit polyclonal anti-Nav1.5 and anti-Nav1.1 (diluted 1: 100; Alomone Laboratories, Israel). For control sections, primary antibodies were pre-incubated with their antigenic peptides, or no primary antibody was used. Image J (NIH) was used for relative fluorescence intensity quantitative analysis. Immunolabelled sections, mounted in Citifluor mounting medium (Agar Scientific, UK), were examined on a Leica DMIRM fluorescence microscope and a TCS-SP2 confocal laser-scanning microscope (Leica Microsystems, Germany) using 488 nm excitation and 500–535 nm emission for Alexa488 labelling, Single optical slices were recorded under consistent pinhole, laser intensity and detector settings (Pawley, 1990).
, 百拇医药
Computer simulations
The possibility that iNa could participate in both the initiation and conduction of pacemaker activity was assessed in computer simulations that were based upon a gradient model of the intact SAN and atrium. This was developed from single cell models of electrical activity of rabbit SAN (Zhang et al. 2000) and atrial cells (Oxsoft HEART, version 4.8. Oxsoft Ltd, Oxford) and used the one-dimensional partial differential equation previously described (Zhang et al. 2000). Though developed for the rabbit heart, the model shares common features with the mouse SAN, in particular the gradient of iNa distribution across the tissue. The modelling assumed a SAN centre of radius 1.5 mm that contained a uniform density of TTX-sensitive neuronal Na+ channels, and began by assuming that these would give a maximum current of 10 pA pF–1. The surrounding SAN periphery was assumed to have an annular radius of 1.5 mm and to contain both neuronal and cardiac Na+ channels at a density that increased towards the periphery giving maximum currents from 10 pA pF–1 to the 60 pA pF–1 towards the periphery as reported below. The remaining ionic current densities were as described experimentally for the rabbit SAN (Honjo et al. 1996, 1999; Lei et al. 2000, 2001). The numerical simulations then investigated the effect of progressively reducing iNa density selectively both in the SAN centre and in the SAN periphery adjacent to the atrium. The pacemaker cycle length (CL) was measured as the time interval between two successive APs. The sinoatrial conduction time (SACT) was measured as the time taken for the excitation wave fronts to travel between two recording sites s1, within the SAN 2 mm from the SAN-atrial border and s2, within the atrium, 9 mm from the SAN-atrial border.
, http://www.100md.com
Data are expressed as means ±S.E.M. (number of animals or preparations or cells). Differences were evaluated by Student's unpaired t test with a P value < 0.05 considered significant.
Results
The mouse SA nodes, located in the intercaval region parallel to the crista terminalis (CT) (Verheijck et al. 2001; Liu et al. 2004), resembled those previously reported in rabbit and other small mammals (Bleeker et al. 1980; Opthof et al. 1987; Boyett et al. 2000). Thus they showed a greater nuclear density and lighter cytoplasm than the surrounding atrial tissue reflecting the smaller size of SA node cells. Such cells nevertheless increased in size from the centre to the periphery. It was thus possible to identify the (300 μm x 150 μm) primary pacemaker area thought typically to contain 450 primary pacemaker cells. The central cells interweave in a circular pattern around the sinus artery and are orientated perpendicular to the CT; cells in the SAN periphery and atrial muscle of the CT are mainly orientated parallel to the CT. The nodal tissue is separated from atrial myocardium at its atrial free wall and septal sides by connective tissue (Liu et al. 2004).
, http://www.100md.com
Expression of Na+ channels in Scn5a+/– SAN cells
Figure 1A–C compare typical tissue sections (A and B), and relative fluorescence intensity plots (C) through the thick CT and thin intercaval region, which contains the SAN, from male WT (n= 5 for Nav1.5, n= 4 for Nav1.1) and male Scn5a+/– (n= 5 for Nav1.5, n= 4 for Nav1.1) hearts labelled using a specific antibody against Nav1.5 or Nav1.1 channels. Immunofluorescence signal from both SAN and surrounding regions from WT and Scn5a+/– SAN tissue sections were compared under identical imaging conditions (Methods) at low and high magnifications using fluorescence confocal microscopy. There was much lower labelling in Scn5a+/– cells consistent with a reduced expression of Nav1.5 following heterozygous disruption of Scn5a (Fig. 1A) exemplified by the results of the semi-quantitative analysis illustrated in Fig. 1C. Scn5a+/– tissue demonstrated considerably lower overall Nav1.5 fluorescence signal in both atrial (63 ± 4% that of WT) and SAN (75 ± 9% that of WT) compared with corresponding regions in WT hearts. Furthermore, the Nav1.5 channel was selectively expressed in peripheral SAN and the surrounding atrial muscle: cells in the centre of the SAN surrounding the SAN artery showed little labelling in either case.
, 百拇医药
A, labelling of Nav1.5 in WT (top) and Scn5a+/– (bottom) SANs. B, labelling of Nav1.1 in WT (left) and Scn5a+/– (right) SANs. C, regional comparison of relative fluorescence intensity in and around the SAN in Nav1.5 in WT and Scn5a+/– SANs. C, centre of the SAN; P, periphery of the SAN; endo, endocardium; epi, epicardium; CT, crista terminalis; SEP, atrial septum. Relative fluorescence intensity was calculated by Image J 1.29x (NIH, USA). Measurements of grey scale pixel intensity are based on the selected area of interests. The pixel intensity is plotted against a fixed distance within the region of interest along the line from CT, peripheral SA node, centre SAN and septum. Scale bar, 100 μm.
, 百拇医药
In contrast, Nav1.1 signal was detected throughout the tissue including the centre and the periphery of the SAN with the SAN and surrounding atrial muscle showing similar expression patterns in the WT and Scn5a+/– hearts (Fig. 1B). It was not possible fully to quantify fluorescence intensity of Nav1.1 in the SAN regions in the WT and Scn5a+/– hearts owing to the presence of non-specific nuclear staining that has been previously reported with use of some Na+ channel antibodies (see Kucera et al. 2002). No labelling above background was obtained when the primary antibodies (anti-Nav1.5 or anti-Nav1.1) or secondary antibody were omitted or were pre-incubated with their antigenic peptides (data not shown). SAN cells nevertheless increased in size from centre to the periphery.
, 百拇医药
Na+ channel properties in Scn5a+/– SAN cells
Isolated SAN cells from two male WT and two male Scn5a+/– mice were identified through their spindle-shaped morphology (Fig. 2A), spontaneous activity and the presence of Na+ currents (iNa) in response to patch clamp voltage steps of 20 ms duration from a –120 mV holding potential to test potentials between –90 and +30 mV. Both WT and Scn5a+/– SAN cells displayed inward Na+ currents with typical activating and inactivating kinetics (Fig. 2B and C). The corresponding current–voltage relationships derived from the maximum iNa (Fig. 2D) demonstrated similar thresholds for iNa activation close to a –70 mV test potential and maximum peak currents at similar voltages (–20 mV) at 37°C. However, Scn5a+/– cells showed significantly reduced peak current densities (Fig. 2D and Table 1) (36 ± 1 pA pF–1, n= 6, Cm= 38–55 pF, 47 ± 2.4 pF) compared with WT myocytes (57 ± 4 pA pF–1, n= 6, Cm= 36–58 pF, 48 ± 3.0 pF, P < 0.01). Moreover, there was a significant correlation between the density of iNa and cell membrane capacitances in WT cells (n= 6, r2= 0.71, P < 0.05) but not in Scn5a+/– cells (n= 6, r2= 0.21, P > 0.05).
, 百拇医药
A, an example of typical mouse SAN pacemaker cell for patch clamping. Scale bar, 20 μm. B and C, recordings of total iNa in a WT and Scn5a+/– cell. D, mean (±S.E.M.) current–voltage relationships for total iNa from WT (n= 6) and Scn5a+/– cells (n= 6). E and F, mean (±S.E.M.; n= 6 for each group) activation and inactivation curves for iNa.
The current–voltage relationships also yielded normalized activation curves (Fig. 2E) for the peak Na+ conductance, gNa. Fitting of the activation (Fig. 2E) and inactivation curves (Fig. 2F) to a Boltzmann equation (Table 1) gave respective values of half-activation voltage, V1/2, and steepness factor, k, in Scn5a+/– myocytes that were statistically indistinguishable from the corresponding values for the WT. These findings suggest that SAN cells from Scn5a+/– mice have reduced Na+ channel densities but similar Na+ channel gating characteristics.
, 百拇医药
Heart rate and SAN function in Scn5a+/– mice
We next studied heart rates and rhythm in ambulant mice (5 male WT, 6 male Scn5a+/–) and electrical activity in isolated SAN preparations (from 4 male and 4 female WT, 4 male and 3 female Scn5a+/– mice) to assess the physiological effects of reduced Na+ channel conductance on SAN pacemaker function. Telemetered, long-term ECG recordings showed that Scn5a+/– (546 ± 24 beats min–1, n= 6) had significantly lower mean heart rates, as recorded over 60 h, compared with WT mice (592 ± 17 beats min–1, n= 5; P < 0.01). Standard ECG studies under anaesthesia showed the occasional Scn5a+/– mouse (2 out of 67 studied) but none of the WT mice (out of 57) showed a persistent sino-atrial (SA) block, which might reflect an extreme manifestation of the sinus node anomalies shown in the Scn5a+/– mice. Figure 3B exemplifies recordings from a (24 week old) Scn5a+/– mouse with completely absent P waves showing a junctional escape rhythm, and Fig. 3C illustrates a trace, recorded at 60 weeks, in which P waves were present only before a minority of the ECG waveforms. These findings might reflect either SA block or sinus arrest. Both these findings contrast with the normal ECG patterns that were shown by the WT (Fig. 3A).
, 百拇医药
Standard lead I surface ECG recordings obtained in anaesthetized WT (A) and Scn5a+/– mice showing either complete (absence of P wave; first recording at 24 weeks old) (B) or partial (second recording at 60 weeks old) SA block with the presence of a few conducted atrial (P) waves (arrows) (C).
Secondly, pacemaker activity and activation patterns were determined from extracellular potential (ECP) recordings that were made from 200 recording sites at a resolution of 0.2–0.3 mm over the SAN and surrounding atrial muscle (Fig. 4A). Spontaneous pacemaker activity began in the SAN centre (the leading pacemaker site). It then propagated to the crista terminalis and interatrial septum. The Scn5a+/– (n= 8) and WT (n= 7) SAN preparations showed similar overall patterns of activation. However, the Scn5a+/– preparations showed a significantly slower conduction towards the SAN periphery and around the SAN–atrial junction indicating a peripheral exit block reflecting slowed conduction (Fig. 4B and C).
, 百拇医药
A, an example of a SAN preparation used for the electrical mapping. Scale bar, 200 μm. B and C, activation sequence in WT and a Scn5a+/– SAN. D and E, action potentials recorded from the sites near the centre of the SAN in A and B are shown from WT (left) and Scn5a+/– (right) SAN preparations. F and G, SAN conduction in WT and Scn5a+/– preparations. Extracellular potentials from sites a (leading pacemaker site in the centre of the SAN; see B and C) and c (atrial muscle, AM) are shown. Vertical dashed lines indicate the time of initiation of the AP at the leading pacemaker site (left) and the arrival of the AP in the atrial muscle (right). H, simultaneous SAN and atrial muscle recordings showing sino-atrial conduction block in Scn5a+/– SAN that was never observed in WT. SEP, septum; SVC, superior vena cava; IVC, inferior vena cava; CT, crista terminalis; and RA, right atrial appendage.
, 百拇医药
Thirdly, intact SAN preparations from WT mice showed regular spontaneous activity with a 170 ± 7 ms (n= 7) cycle length (CL). The Scn5a+/– SAN showed longer CLs (223 ± 14 ms, n= 8, P < 0.05 compared with WT SAN) indicating lower pacemaker rates (Fig. 4D and E). Sino-atrial conduction times (SACTs) obtained from simultaneous ECP recordings (Fig. 4F and G) from the SAN centre (site a: Fig. 4B and C) and neighbouring atrial muscle (site c: Fig. 4B and C) were significantly longer in Scn5a+/– (16 ± 2 ms, n= 8) compared with WT (12 ± 1 ms, n= 7, P < 0.05). Additionally, 3 out of 8 Scn5a+/– SANs but none of the WT preparations also showed evidence of a 3: 2 or 2: 1 SA conduction block (Fig. 4H).
, http://www.100md.com
Finally, those features of spontaneous pacemaker activity that related to cell size, type and current densities (Honjo et al. 1996; Lei et al. 2001) were compared in isolated SAN cells (from 3 male and 1 female WT, 3 male and 1 female Scn5a+/– mouse, respectively). Intracellular recordings from small cells, previously defined as having cell membrane capacitances Cm 30 pF in rabbit SAN cells, presumably from the SAN centre (Honjo et al. 1996, 1999; Lei & Boyett, 1998, Lei et al. 2001; Musa et al. 2002) (Fig. 5A and B), demonstrated similar pacemaker rates in Scn5a+/–(CL = 164 ± 3 ms; Cm, 30 ± 1 pF, n= 3) compared with WT myocytes (CL = 177 ± 2 ms; Cm, 30 ± 3 pF, n= 2). In contrast, larger cells, defined as having Cm > 35 pF, presumably from the SAN periphery (Honjo et al. 1996, 1999; Lei & Boyett, 1998; Lei et al. 2001; Musa et al. 2002) (Fig. 5C and D), demonstrated generally slower pacemaker rates and lower AP peaks in Scn5a+/–(CL = 186 ± 7 ms; Cm, 45 ± 4 pF, n= 4) than in WT myocytes (CL = 161 ± 3 ms; Cm, 46 ± 4 pF, n= 5). However, the average firing rate of SAN cells was not significantly different (P > 0.05, WT versus Scn5a+/–, unpaired t test) between WT (CL = 166 ± 4 ms, n= 7) and Scn5a+/–(CL = 177 ± 6 ms, n= 7). Figure 5E–I plots a range of cellular characteristics that bear on spontaneous pacemaker activity against cell size as reflected in Cm. It confirms that the values of the independent variate Cm explored through the two data sets were not significantly different (P > 5%; unpaired t test). However, WT and Scn5a+/– gave significantly different fitted slopes to a P < 0.1% significance level in the dependence of cycle length (CL), maximum upstroke velocity (MUP), and action potential amplitude (APA), but not maximum diastolic potential (MDP) or the diastolic depolarization rate (DDR), upon Cm consistent with earlier reports in which tetrodotoxin (TTX)-mediated Na+ channel block did not influence electrical activity in small ball preparations from the centre, but influenced spontaneous activity (by 50%) in preparations from rabbit SAN periphery without changing MDP (Kodama et al. 1997). There were thus significant contrasts between the two groups not simply attributable to quantitative anatomical differences between SAN cells in WT and Scn5a+/– mice.
, http://www.100md.com
A and B, spontaneous APs recorded in WT (A and C) and Scn5a+/– cells (B and D). E–I, relationship between maximum diastolic potential (MDP), cycle length (CL), diastolic depolarization rate (DDR), maximum upstroke velocity (dv/dt) and AP amplitude (APA), and cell size as reflected in cell capacitance Cm from WT (n= 7) and Scn5a+/– (n= 7) SAN cells.
INa facilitates SAN function in driving atrial activation
The present findings therefore directly implicate the Nav1.5 channel in action potential conduction through the SAN and from the SAN to the surrounding atrial muscle. Preliminary modelling studies then suggested a possible mechanism that might explain the effect of Nav1.5 channels upon the normal SAN pacemaker rate. Loss-of-function mutations in the Na+ channel alone might then form the entire physiological basis for SND. The possible feasibility for such a scheme was explored numerically in a modified rabbit SAN model for pacemaker function comprising successively coupled cells from the SAN centre, which initiates spontaneous APs, to its periphery, that propagate such APs to reach atrial muscle, which is described in the Methods (Zhang et al. 2000). This predicted that a 35% block of iNa in a 2 mm length of the peripheral part of the SAN, would result in a 41 ms increase in CL (316 ms in control versus 357 ms in the condition of a 35% block of peripheral iNa) implying a significantly reduced pacemaker rate, and an 18.4 ms increase in SACT and reduction of conduction velocity (0.059 m s–1 in control versus 0.047 m s–1 in the condition of a 35% block of peripheral iNa) corresponding to a compromised sino-atrial conduction. Figure 6A–E illustrates modelled AP waveforms in successive myocytes under control conditions (A), following a 100% reduction of iNa confined to the SAN centre (B), and successive 35% (C), 70% (D) and 80% (E) reductions of iNa selectively in the SAN periphery. These decrease pacemaker rate, as reflected in the time interval between successive APs (B), and decrease conduction velocity (C) to result in a conduction exit block in which APs originating from the SAN totally fail to conduct into the atrium (D) and finally a complete termination of SAN pacemaker activity (E).
, http://www.100md.com
A, APs displayed along the length of SAN and atrial string. The AP is first initiated in the SAN centre then propagates via its periphery towards atrium. B, effect of selectively blocking iNa in the SAN centre by 100% increases CL by 11 ms compared with the control. C–E, APs with progressive selective reductions of iNa in peripheral SAN by 35% (C), 70% (D) and 80% (E). AM, atrial muscle.
Figure 7 illustrates an experimental fulfilment of such predictions. Thus, Fig. 7A shows extracellular potentials recorded from the centre of the SA node and from the neighbouring atrial muscle from a WT mouse. Such simultaneous recordings permitted SA node conduction time (time taken for conduction from points a to c in Fig. 4B and C) to be assessed. In such experiments, applications of 10 (n= 5) and 100 (n= 5) nM TTX caused no significant change of SA node conduction. However, increases in TTX concentration from 1 to 5 μM prolonged SA node conduction by 22 ± 7 and 39 ± 4% (P < 0.01, n= 5) in two male and three female WT mice and conduction by 30 ± 5 and 56 ± 9% (P < 0.01, n= 5) in two male and three female Scn5a+/– mice. TTX at 12.5 μM completely blocked SA node conduction.
, 百拇医药
A, effect of TTX on SA node pacemaking. Extracellular potentials recorded from centre of the SAN and periphery of the SAN. B, effect of TTX on sinoatrial conduction. Vertical dashed lines indicate the time of initiation of the action potential at the leading pacemaker site (left) and the arrival of the action potential in the periphery or atrial muscle (right).
Discussion
The present experiments explored electrophysiological roles for cardiac Scn5a-encoded Na+ channels in SAN function using mice with a targeted genetic disruption. They provide the first direct evidence implicating Nav1.5 channels in action potential conduction through the SAN and from the SAN to surrounding atrial muscle and additionally suggest indirect roles for Nav1.5 channels in maintenance of the normal SAN pacemaker rate. They thus corroborate indirect clinical evidence (Tan et al. 2001; Benson et al. 2003; Veldkamp et al. 2003) associating human genetic defects in SCN5A with sinus node disorders (SND). Analysis of mutant channels in stable mammalian cells demonstrated that 2 of 6 mutations associated with human congenital SND result in non-functional Na+ channels with the remaining resulting in mild to severe channel dysfunction (Tan et al. 2001; Benson et al. 2003). Furthermore, the 1795insD, gain of function, mutation has been associated with sinus bradycardia and sinus pauses, in an inherited human long-QT syndrome (LQT3) that also shows phenotypic characteristics of SND (Veldkamp et al. 2003). Thus LQT3 Na+ channel mutations causing persistent inward currents could result not only in QT prolongation but also sinus bradycardia and sinus pauses (Veldkamp et al. 2003). Both loss (Benson et al. 2003) and gain of function mutations (Veldkamp et al. 2003) in Nav1.5 (Scn5a) might thus be implicated in sinus node dysfunction.
, http://www.100md.com
Our experiments first directly demonstrated the clinically observed sinus bradycardia, slow SA conduction and sino-atrial exit block following targeted genetic disruption of cardiac Scn5a-encoded Na+ channels (Asseman et al. 1983), and went on to clarify possible roles for the Scn5a Na+ channel in both conduction and pacing in the mouse SAN system for the first time. Thus both intact Scn5a+/– mice and isolated Scn5a+/– hearts showed depressed mean rates and persistent SA block. Both immunochemical and biophysical analyses demonstrated a reduced expression of SAN Na+ channels: patch clamp studies of Scn5a+/– SAN cells observed similar steady-state activation and inactivation properties but reduced maximum Na+ currents (30%) compared with findings from WT mice. Recent reports describe larger (50%) reductions in iNa in ventricular Scn5a+/– myocytes (Papadatos et al. 2002). However, in contrast to ventricular myocytes, SAN cells express neuronal (Nav1.1, Nav1.3) (Maier et al. 2003) in addition to the adult cardiac Scn5a-encoded Nav1.5 channel isoforms at relative densities that may vary with cell size (Lei et al. 2004). Mouse SAN cells accordingly demonstrate separable, TTX-sensitive (neuronal) and TTX-resistant (cardiac) Na+ currents and this could reconcile the apparent differences between the two cell types (Boyett et al. 2000; Maier et al. 2003; Lei et al. 2004). Although this presence of Nav1.1 in the SAN has been reported in several species (Maier et al. 2003; Lei et al. 2004; Haufe et al. 2005), there is no report concerning its expression in human SAN.
, 百拇医药
Secondly, comparisons of mouse SAN preparations suggested specific roles for Scn5a-encoded Na+ channels in SAN conduction and pacemaking. The presence or otherwise of Nav1.5 channels strongly influenced both upstroke velocity and APA. Correspondingly, although multiple extracellular recordings revealed similar overall activation patterns over the SAN and surrounding atrial muscle, Scn5a+/– hearts showed a slower conduction, both towards the SAN periphery and between SAN and atrium suggesting an SA exit block, giving rise to longer overall conduction times from the leading pacemaker site in the SAN centre to the surrounding atrial muscle, than in WT. Furthermore some Scn5a+/– but no WT SANs showed SA conduction block.
, 百拇医药
Thirdly, the present findings permitted a hypothesis in which cells containing cardiac Scn5a-encoded Na+ channels influenced pacemaker rate through their electrical coupling to primary pacemaker cells in the SAN centre. Thus, Scn5a+/– hearts showed a slowed SAN pacemaker activity. This implicates iNa in the bradycardia observed in both experimental and clinical conditions. Microelectrode AP recordings from sites near the centre of intact Scn5a+/– SAN preparations showed longer cycle lengths than did WT. Furthermore, 3 out of 8 Scn5a+/–, but none of the WT, preparations showed a 3: 2 or 2: 1 SA conduction block. Furthermore, isolated WT and Scn5a+/– SAN cells showed contrasting correlations between cycle length, maximum upstroke velocity and AP amplitude, all characteristics related to spontaneous pacemaker activity, and cell size.
, http://www.100md.com
In contrast, disruption of Scn5a-encoded cardiac Na+ channels did not affect MDP (Fig. 5), findings that parallel previous reports (Kodama et al. 1997; Lei et al. 2004) that implicate Nav1.5 (TTX-resistant) currents in pacemaker action potentials, but not MDP in single SAN cells: block of both TTX-sensitive and -resistant iNa by 30 μM TTX reduced the slope of the pacemaker potential and the threshold potential and increased cycle length without changing MDP (Lei et al. 2004). TTX similarly spared electrical activity in small ball preparations from the centre but decreased take-off potential and upstroke velocity and slowed spontaneous activity (by 50%;) in balls from the periphery of rabbit SAN without changing MDP (Kodama et al. 1997), findings consistent with a higher iNa expression in peripheral node cells (Honjo et al. 1996; Lei et al. 2004). Our results also correlated Nav1.5 current densities with cell capacitance in WT but not Scn5a+/– cells localizing the decreased Nav1.5 expression to peripheral SAN cells in Scn5a+/– mice. Finally, pacemaker rates in Scn5a+/– SAN resembled those of WT in small myocytes but were slower than those of WT in large myocytes.
, http://www.100md.com
Previous studies reported that cells from the centre of the rabbit SAN had more positive take-off potentials, slower upstrokes, longer durations in their action potentials, less negative maximum diastolic potentials, and slower intrinsic pacemaker activity than cells from the periphery (Bleeker et al. 1980; Kodama & Boyett, 1985; Kodama et al. 1997; Boyett et al. 1999. We have suggested that these regional differences in electrical activity can be attributed to regional differences in the intrinsic properties of their contained cells. Thus: (1) single SAN cells from the rabbit show parallel heterogeneities in action potential characteristics (Honjo et al. 1996). Furthermore (2) cell size increases from the SAN centre to its periphery (Bleeker et al. 1980; Masson-Pévet et al. 1984; Oosthoek et al. 1993; Boyett et al. 2000), and (3) the smaller cells show electrical activity that resembles that in the SAN centre whereas electrical activity in (4) the larger cells closely parallels that observed in the SAN periphery (Honjo et al. 1996). (5) We have recently suggested (Liu et al. 2004) that the mouse SAN closely resembles the SAN in other small mammals in structure (Bleeker et al. 1980; Opthof et al. 1987; Boyett et al. 2000; Dobrzynski et al. 2005). It differs histologically from the surrounding myocardium in its greater nuclear density and a lighter cytoplasmic staining and shows a gradient in cell size, arrangement, orientation and connective tissue content from centre to periphery. In particular (6) cell size increases from centre to periphery as expected from the gradient model of SAN organization (Boyett et al. 2000). These studies relate differences in ionic currents densities with cell size to the regional differences in electrical activity within the mouse SAN in the present experiments (Honjo et al. 1996, 1999; Lei et al. 2000).
, 百拇医药
Earlier work had not considered possible roles for iNa in normal or abnormal SAN pacemaker function: micromolar levels of TTX abolishes activity in the SAN periphery, but not its centre (Lei et al. 2004) suggesting that the AP upstroke in the SAN periphery only depends on iNa whereas the AP upstroke at its centre depends instead on iCa,L (Kodama et al. 1997). Nevertheless, some recent reports suggest an involvement of cardiac iNa. Thus, both intact Langendorff-perfused mouse Scn5a+/– hearts (Papadatos et al. 2002) and patients with SCN5A-linked hereditary Lenègre disease (Probst et al. 2003) are bradycardic. Furthermore, increasing cell-cell electrotonic coupling progressively converts independent pacemaker behaviour in rabbit SAN cells to a situation in which SAN and atrial AP frequencies and waveforms became entrained (Verheijck et al. 1998). Besides altering AP conduction in the SAN, decreased AP upstroke velocities might then both affect SAN–atrial coupling as well as depress rhythmic activity in the central pacemaker cells themselves. This hypothesis was tested by a numerical modelling of the initiation and propagation of SAN activity (Zhang et al. 2000) which assumed a constant density of TTX-sensitive Na+ channels in central SAN cells and a positive gradient of TTX-resistant channel densities from the SAN periphery to its boundary with atrial muscle. This analysis predicted that reducing iNa density either in the SAN centre or its periphery significantly decreased pacemaker rate. Reductions in iNa restricted to the SAN periphery, progressively resulted in decreased SAN conduction velocity, an exit block of AP conduction from SAN totally to atrium, and finally a complete termination of SAN pacemaker activity. Such findings corroborate recent reports that block of neuronal Na+ channels by TTX concentrations (100 nM) sufficient to inhibit at least 90% of the Nav1.1 channels (KD, 1–10 nM) reduced heart rate by 65% as assessed in intact heart (Maier et al. 2003), 25% in intact SANs (Lei et al. 2004) and 15% in isolated SAN cells (Lei et al. 2004).
, 百拇医药
Our findings following targeted heterozygous disruption of Scn5a thus directly implicate Nav1.5 channel currents in the conduction of APs through the SAN and from the SAN to atrial muscle and they suggest that these also have indirect effects in the maintenance of normal SAN pacemaker rhythm and rate. Disruption of Scn5a channels appears, firstly, to reduce pacemaker potentials through a mechanism that involves the coupling of electrical events between pacemaker cells and cells surrounding them within the SAN: the resulting impairment of impulse initiation accounts for the observed sinus bradycardia. It also affects impulse propagation through the SAN and between the SAN and atria that may result in conduction block. Thus our findings from the Scn5a+/– mouse SAN model provide a background for further explorations seeking to clarify the role of cardiac-type Na+ channels in normal SAN function in other mammalian (including human) species, for which an understanding of the pathophysiology of SND, the most common indication for pacemaker prescription, might offer potential alternative targets for its clinical management.
, 百拇医药
References
Antzelevitch C, Brugada P, Brugada J, Brugada R, Towbin JA & Nademanee K (2003). Brugada syndrome: 1992–2002. A historical perspective. J Am Coll Cardiol 41, 1665–1671.
Asseman P, Berzin B, Desry D, Vilarem D, Durand P, Delmotte C, Sarkis EH, Lekieffre J & Thery C (1983). Persistent sinus nodal electrograms during abnormally prolonged postpacing atrial pauses in sick sinus syndrome in humans: sinoatrial block vs overdrive suppression. Circ Res 68, 33–41.
, 百拇医药
Benson DW, Wang DW, Dyment M, Knilans TK, Fish FA, Strieper MJ, Rhodes TH & George AL Jr (2003). Congenital sick sinus syndrome caused by recessive mutations in the cardiac sodium channel gene (SCN5A). J Clin Invest 112, 1019–1028.
Bleeker WK, Mackaay AJ, Masson Pevet M, Bouman LN & Becker AE (1980). Functional and morphological organization of the rabbit sinus node. Circ Res 46, 11–22.
Bogdanov KY, Vinogradova TM & Lakatta EG (2001). Sinoatrial nodal cell ryanodine receptor and Na+-Ca2+ exchanger: molecular partners in pacemaker regulation. Circ Res 88, 1254–1258.
, http://www.100md.com
Boyett MR, Honjo H & Kodama I (2000). The sinoatrial node, a heterogeneous pacemaker structure. Cardiovasc Res 47, 658–687.
Boyett MR, Honjo H, Yamamoto M, Nikmaram MR, Niwa R & Kodama I (1999). Downward gradient in action potential duration along conduction path in and around the sinoatrial node. Am J Physiol 276, H686–H698.
Catterall WA (2000). From ionic currents to molecular mechanisms: the structure and function of voltage-gated sodium channels. Neuron 26, 13–25.
, 百拇医药
Cho HS, Takano M & Noma A (2003). The electrophysiological properties of spontaneously beating pacemaker cells isolated from mouse sinoatrial node. J Physiol 550, 169–180.
Dobrzynski H, Li J, Tellez J, Greener ID, Nikolski VP, Wright SE et al. (2005). Computer three-dimensional reconstruction of the sinoatrial node. Circulation 111, 846–854.
Goldin AL (2001). Resurgence of sodium channel research. Annu Rev Physiol 63, 871–894.
, 百拇医药
Haufe V, Cordeiro JM, Zimmer T, Wu YS, Schiccitano S, Benndorf K & Dumaine R (2005). Contribution of neuronal sodium channels to the cardiac fast sodium current INa is greater in dog heart Purkinje fibers than in ventricles. Cardiovas Res 65, 117–127.
Honjo H, Boyett MR, Kodama I & Toyama J (1996). Correlation between electrical activity and the size of rabbit sino-atrial node cells. J Physiol 496, 795–808.
Honjo H, Lei M, Boyett MR & Kodama I (1999). Heterogeneity of 4-aminopyridine-sensitive current in rabbit sinoatrial node cells. Am J Physiol 45, H1295–H1304.
, 百拇医药
Irisawa H, Brown HF & Giles W (1993). Cardiac pacemaking in the sinoatrial node. Physiol Rev 73, 197–227.
Kodama I & Boyett MR (1985). Regional differences in the electrical activity of the rabbit sinus node. Pflugers Arch 404, 214–226.
Kodama I, Nikmaram MR, Boyett MR, Suzuki R, Honjo H & Owen JM (1997). Regional differences in the role of the Ca2+ and Na+ currents in pacemaker activity in the sinoatrial node. Am J Physiol 272, H2793–H2806.
, http://www.100md.com
Kucera JP, Rohr S & Rudy Y (2002). Localization of sodium channels in intercalated disks modulates cardiac conduction. Circ Res 91, 1176–1182.
Lei M & Boyett MR (1998). Inhibition of transient outward current, it(to), by flecainide and quinidine in rabbit isolated sinoatrial node cells. J Physiol 511P, 78–79P.
Lei M, Cooper P, Camelliti P & Kohl P (2002). Contribution of the fast sodium inward current, iNa, to murine sino-atrial node pacemaking. Biophys J 82, 605a.
, http://www.100md.com
Lei M, Honjo H, Kodama I & Boyett MR (2000). Characterisation of the transient outward K+ current in rabbit sinoatrial node pacemaker cells. Cardiovasc Res 46, 433–441.
Lei M, Honjo H, Kodama I & Boyett MR (2001). Heterogeneous expression of expression of the delayed rectifier K+ currents iK,r and iK,s in rabbit sinoatrial node cells. J Physiol 535, 703–714.
Lei M, Jones SA, Liu J, Lancaster MK, Fung SS-M, Dobrzynski H, Camelliti P, Maier SKG, Noble D & Boyett MR (2004). Requirement of neuronal- and cardiac-type sodium channels for murine sinoatrial node pacemaking. J Physiol 559, 835–848.
, http://www.100md.com
Liu J, Dobrzynski H, Boyett MR & Lei M (2004). Sophisticated structural organization is required for normal functioning of murine sinoatrial node. Biophys J 86, 297a.
Maier SK, Westenbroek RE, Schenkman KA, Feigl EO, Scheuer T & Catterall WA (2002). An unexpected role for brain-type sodium channels in coupling of cell surface depolarization to contraction in the heart. Proc Natl Acad Sci U S A 99, 4073–4078.
Maier SK, Westenbroek RE, Yamanushi TT, Dobrzynski H, Boyett MR, Catterall WA & Scheuer T (2003). An unexpected requirement for brain-type sodium channels for control of heart rate in the mouse sinoatrial node. Proc Natl Acad Sci U S A 100, 3507–3512.
, 百拇医药
Malhotra JD, Chen C, Rivolta I, Abriel H, Malhotra R, Mattei LN, Brosius FC, Kass RS & Isom LL (2001). Characterization of sodium channel - and -subunits in rat and mouse cardiac myocytes. Circulation 103, 1303–1310.
Mangoni ME & Nargeot J (2001). Properties of the hyperpolarization-activated current (If) in isolated mouse sino-atrial cells. Cardiovasc Res 52, 51–64.
Masson-Pevet MA, Bleeker WK, Besselsen E, Treytel BW, Jongsma HJ & Bouman LN (1984). Pacemaker cell types in the rabbit sinus node: a correlative ultrastructural and electrophysiological study. J Mol Cell Cardiol 16, 53–63.
, 百拇医药
Mitsuiye T, Shinagawa Y & Noma A (2000). Sustained inward current during pacemaker depolarization in mammalian sinoatrial node cells. Circ Res 87, 88–91.
Musa H, Lei M, Honjo H, Kodama I & Boyett M (2002). Heterogeneous expression calcium handling proteins in the pacemaker cells of the rabbit. J Histochem Cytochem 2002, 311–324.
Nathan RD (1986). Two electrophysiologically distinct types of cultured pacemaker cells from rabbit sinoatrial node. Am J Physiol 250, H325–H329.
, http://www.100md.com
Oosthoek PW, Viragh S, Mayen AE, Van Kempen MJ, Lamers WH & Moorman AF (1993). Immunohistochemical delineation of the conduction system. I: The sinoatrial node. Circ Res 73, 473–481.
Opthof T, De Jonge B, Jongsma HJ & Bouman LN (1987). Functional morphology of the mammalian sinuatrial node. Eur Heart J 8, 1249–1259.
Papadatos GA, Wallerstein PMR, Head CEG, Ratcliff R, Brady PA, Benndorf K et al. (2002). Slowed conduction and ventricular tachycardia after targeted disruption of the cardiac sodium channel gene Scn5a. Proc Natl Acad Sci U S A 99, 6210–6215.
, 百拇医药
Pawley JB (1990). Handbook of biological confocal microscopy. Plenum Press, NY.
Press WH, Flanney BP, Teukolsky SA & Vetterling WT (1986). Numercial recipes. In The Art of Scientific Computing, pp. 523–528. Cambridge University Press, Cambridge.
Probst V, Kyndt F, Potet F, Trochu JN, Mialet G, Demolombe S, Schott JJ, Baro I, Escande D & Le Marec H (2003). Haploinsufficiency in combination with aging causes SCN5A-linked hereditary Lenegre disease. J Am Coll Cardiol 41, 643–652.
, http://www.100md.com
Tan HL, Bink Boelkens MT, Bezzina CR, Viswanathan PC, Beaufort Krol GC, Van Tintelen PJ, Van Den Berg MP, Wilde AA & Balser JR (2001). A sodium-channel mutation causes isolated cardiac conduction disease. Nature 409, 1043–1047.
Veldkamp MW, Wilders R, Baartscheer A, Zegers JG, Bezzina CR & Wilde AA (2003). Contribution of sodium channel mutations to bradycardia and sinus node dysfunction in LQT3 families. Circ Res 92, 976–983.
, 百拇医药
Verheijck EE, Van Kempen MJA, Mike Veereschild M, Lurvink J, Jongsma HJ & Bouman LN (2001). Electrophysiological features of the mouse sinoatrial node in relation to connexin distribution. Cardiovasc Res 52, 40–50.
Verheijck EE, Wilders R, Joyner RW, Golod DA, Kumar R, Jongsma HJ, Bouman LN & Van Ginneken AC (1998). Pacemaker synchronization of electrically coupled rabbit sinoatrial node cells. J General Physiol 111, 95–112.
, http://www.100md.com
Yamamoto M, Honjo H, Niwa R & Kodama I (1998). Low-frequency extracellular potentials recorded from the sinoatrial node. Cardiovasc Res 39, 360–372.
Yu FH & Catterall WA (2003). Overview of the voltage-gated sodium channel family. Genome Biol 4, 207.
Zhang H, Holden AV, Kodama I, Honjo H, Lei M, Varghese T & Boyett MR (2000). Mathematical models of action potentials in the periphery and center of the rabbit sinoatrial node. Am J Physiol 279, H397–H421., 百拇医药(Ming Lei, Catharine Godda)
2 Cardiovascular Group, Departments of Biochemistry and Physiology, University of Cambridge, Tennis Court Road, Cambridge CB2 1QW UK
3 INSERM U533, Laboratoire de Physiopathologie et Pharmacologie Cellulaires et Moléculaires, Faculté de Médecine, Nantes, France
4 Union Hospital, Huazhong University of Science and Technology, Wuhan, The People's Republic of China
, http://www.100md.com
5 First Hospital, Xi'an Medical School, Xi'an Jiaotong University, The People's Republic of China
6 Biological Physics Group, Physics Department, UMIST, Manchester, M60 1QD UK
7 Electrophysiology and Biophysics Program, Victor Chang Cardiac Research Institute, 384 Victoria Street, Darlinghurst, NSW2010, Australia
Abstract
We have examined sino-atrial node (SAN) function in hearts from adult mice with heterozygous targeted disruption of the Scn5a gene to clarify the role of Scn5a-encoded cardiac Na+ channels in normal SAN function and the mechanism(s) by which reduced Na+ channel function might cause sinus node dysfunction. Scn5a+/– mice showed depressed heart rates and occasional sino-atrial (SA) block. Their isolated peripheral SAN pacemaker cells showed a reduced Na+ channel expression and slowed intrinsic pacemaker rates. Wild-type (WT) and Scn5a+/– SAN preparations exhibited similar activation patterns but with significantly slower SA conduction and frequent sino-atrial conduction block in Scn5a+/– SAN preparations. Furthermore, isolated WT and Scn5a+/– SAN cells demonstrated differing correlations between cycle length, maximum upstroke velocity and action potential amplitude, and cell size. Small myocytes showed similar, but large myocytes reduced pacemaker rates, implicating the larger peripheral SAN cells in the reduced pacemaker rate that was observed in Scn5a+/– myocytes. These findings were successfully reproduced in a model that implicated iNa directly in action potential propagation through the SAN and from SAN to atria, and in modifying heart rate through a coupling of SAN and atrial cells. Functional alterations in the SAN following heterozygous-targeted disruption of Scn5a thus closely resemble those observed in clinical sinus node dysfunction. The findings accordingly provide a basis for understanding of the role of cardiac-type Na+ channels in normal SAN function and the pathophysiology of sinus node dysfunction and suggest new potential targets for its clinical management.
, 百拇医药
Introduction
The sino-atrial node (SAN) initiates normal rhythmic cardiac excitation. SAN automaticity in turn requires a spontaneous diastolic depolarization within its contained pacemaker cells that is the result of complex reciprocal interactions between decays in outward delayed rectifier K+ currents (iK) on the one hand and activation of inward Na+-dependent background (ib,Na), hyperpolarization-activated inward (if), T-type (iCa,T) and L-type Ca2+ (iCa,L) and sustained inward currents (ist) on the other (Irisawa et al. 1993; Boyett et al. 2000; Mitsuiye et al. 2000). Na+–K+ pump (iNaK) and Na+–Ca2+ exchanger currents (iNaCa) and intracellular Ca2+ handling also contribute (Irisawa et al. 1993; Boyett et al. 2000; Bogdanov et al. 2001). Inward voltage-dependent Na+ current, iNa, has been recorded from SAN cells (Nathan, 1986; Irisawa et al. 1993; Honjo et al. 1996; Mangoni & Nargeot, 2001; Cho et al. 2003). However possible contributions from voltage-dependent Na+ currents, iNa, to SAN pacemaker function remain uncertain: iNa occurs only in some pacemaker cells and in any case may inactivate at relatively positive potentials (Nathan, 1986; Irisawa et al. 1993; Honjo et al. 1996).
, http://www.100md.com
There has nevertheless been significant recent progress in our molecular understanding of voltage-gated Na+ channels. These contain both pore-forming -(Nav) and auxiliary -subunits (Catterall, 2000; Goldin, 2001). Thus far, 10 -subunits, with tissue-specific distributions and distinct pharmacological properties, and four -subunits have been identified (Goldin, 2001; Yu & Catterall, 2003). The tetrodotoxin (TTX)-resistant Nav1.5 isoform accounts for most of the cardiac Na+ channels in the heart; however, recent work has also demonstrated TTX-sensitive neuronal (Nav1.1, Nav1.3 and Nav1.6) isoforms (Malhotra et al. 2001; Maier et al. 2002, 2003). The SAN expresses both TTX-sensitive neuronal Na+ channels, whose block slows heart rate (Maier et al. 2003), and TTX-resistant cardiac Na+ channels (Lei et al. 2004). However, these channels may be differentially expressed in different SAN regions compatible with their having contrasting contributions to rhythm generation and/or propagation (Lei et al. 2004).
, 百拇医药
Loss-of-function mutations in the SCN5A encoding cardiac Nav1.5 Na+ channels were initially described in patients with the Brugada and Lev-Lenègre syndromes (Tan et al. 2001; Antzelevitch et al. 2003) but have now been reported in kindreds with familial sinus node dysfunction (SND) (Benson et al. 2003). Specific mechanisms underlying the observed sinus bradycardia, sinus arrest and/or sino-atrial block with which they are associated, whether involving abnormal automaticity, exit block, or impaired intra-atrial conduction and excitability, remain uncertain. Nevertheless, some patients with SCN5A mutations associated with long QT syndrome type 3 (LQT3) also show evidence of SND (Veldkamp et al. 2003) compatible with roles for cardiac-type Na+ channels in normal SAN function. Furthermore, heterologous expression of recombinant Na+ channels predicted from clinical studies in non-excitable cells suggest that Na+ conductance is reduced by SCN5A mutations causing SND (Benson et al. 2003).
, 百拇医药
The work described in this paper explores possible roles for Na+ channels in the normal function of the SAN and mechanisms by which Na+ channel defects might cause SND in pacemaker myocytes using a genetic model system. We have recently established mice with a null mutation in Scn5a whose homozygous disruption causes intrauterine lethality; heterozygotes (Scn5a+/–) are viable but show defects suggesting altered Na+ channel function including impaired atrioventricular conduction, delayed intramyocardial conduction, and ventricular tachycardia with characteristics of re-entrant excitation (Papadatos et al. 2002).
, 百拇医药
Methods
Scn5a-deficient and normal (WT) mice
Mice with targeted disruption of Scn5a were generated as previously described (Papadatos et al. 2002). The experiments used adult mice aged between 2.5 and 5 months for long-term ECG recordings and 3–10 months for single cell and intact SA node studies. Mice were inbred onto a 129/Sv genetic background. Comparisons were made unless otherwise stated between equal numbers of male and female Scn5a+/– mutants and WT littermates. All animal procedures conformed with the United Kingdom Animals (Scientific Procedures) Act 1986.
, 百拇医药
Electrocardiographic (ECG) recordings
Numerical recordings of lead I–III surface ECGs (EMKA technologies, Paris, France) used 25-gauge subcutaneous electrodes and ECG channels bandpass-filtered between 0.5 and 250 Hz in adult anaesthetized (15 mg kg–1 intraperitoneal etomidate) mice. The long-term ECG recordings introduced subcutaneous telemetric transmitters (TA10EA-F20, Data Sciences International, St Paul, MN, USA), with paired wire bipolar electrodes placed over the thorax, into a posterior incision made under anaesthesia (intraperitoneal etomidate, 8 mg kg–1; thiopental, 40 mg kg–1). Experiments began at least 4 days after surgical recovery with mice housed in individual cages in thermostatically controlled rooms subject to 12 h light–dark cycles (light: 7: 00 am to 7: 00 pm) with free access to food and water. ECG signals were recorded telemetrically following analog-to-digital conversion (Data Sciences International).
, 百拇医药
Electrophysiological recordings from single cells
Individual SAN cells isolated from hearts of adult mice (either sex) (Lei et al. 2004) were initially perfused with normal Tyrode solution (mM: 140 NaCl; 5.4 KCl; 1.8 CaCl2; 1 MgCl2; 10 glucose; 5 Hepes; pH 7.4 with NaOH) as described earlier (Lei et al. 2004). The SAN cells were identified by their morphology and spontaneous activity. The studies of iNa reduced NaCl to 67 mM (NaCl replaced by CsCl) and added 300 nM extracellular nisoldipine to block iCa,L. The patch pipette solutions contained (mM): 10 NaCl; 140 KCl; 2 EGTA; 1 MgCl2; 5 MgATP; 10 Hepes (pH 7.3 with KOH for voltage clamping experiments) and 140 KCl; 1.8 MgSO4; 5 Hepes; 0.1 EGTA; titrated to pH 7.3 with KOH with amphotericin (200 μg ml–1) added before the beginning of the current clamping experiments.
, 百拇医药
Cells were superfused at 37°C at 2 ml min–1. Current and voltage clamping used an Axopatch-200B patch clamp amplifier (Axon Instruments, Foster City, CA, USA); whole cell recording was performed as previously described (Lei et al. 2004). Residual uncompensated series resistance prior to compensation was < 5 M. Capacitative current transients were electronically subtracted with series resistance compensation typically at 80%. Current signals were filtered at 10 kHz and sampled at 20 kHz. All current traces were leak-subtracted offline following acquisition and analysis using pCLAMP8 software (Axon Instruments). Cell capacitance (Cm) was obtained from the capacity compensation control of the amplifier following correlation of the whole cell capacity current in response to 5 ms pulses to –70 mV from a holding potential of –60 mV repeated at a frequency of 100 Hz. A previous study has checked the accuracy of this method by integrating the area of the uncompensated capacity current and fitting an exponential function to the decay of the uncompensated capacity current (Honjo et al. 1996). We did not observe significant rundown in either groups of cells over the 20 min duration of our experiments, which were carried out under identical conditions in both groups: in the course of our procedures, the magnitudes of the peak currents at the beginning and end of the protocols were routinely checked.
, 百拇医药
Current–voltage relationships obtained from cells subject to activating steps also yielded normalized activation curves for the peak Na+ conductance, gNa. The corresponding iNa inactivation curves were obtained using standard double pulse protocols repeated at 1 s intervals between series in which conditioning pulses of duration 100 ms were made to potentials between –120 and –20 mV from a –120 mV holding potential; these were then followed by test pulses of 10 ms duration to a fixed membrane potential of –30 mV. Fitting of both the activation and inactivation curves to both the sum of two as well as a single Boltzmann function was attempted using a Levenberg-Marquadt procedure (Press et al. 1986) that performed successive least squares minimizations for the values of each of the parameters of a generalized non-linear function simultaneously over all the experimentally obtained mean values obtained at each test voltage. Successive iterations sought to minimize the values of 2 in the weighted fit using the means and standard errors of the data points. The number of degrees of freedom incorporated not only the number of data points but also the number of variables. The curve fits used a weighting factor for each point that was determined by the inverse of its variance in turn normalized to the average of all such weighting factors. This precaution yielded the maximum likelihood that the fitted function would represent the distribution of the parent data. However, whereas the fits to a single Boltzmann function converged to give half-activation voltages, V1/2, steepness factors, k, and maximum values, those using the sum of two rather than one Boltzmann function did not converge.
, 百拇医药
Electrical mapping of intact SAN
Intact SAN preparations dissected as previously described (Lei et al. 2001, 2002) were superfused with Tyrode solution at 37°C at a rate of 5–6 ml min–1 via a heat exchanger. Intracellular action potentials were recorded using conventional microelectrodes (resistance 30 M); activation times were mapped from extracellular potentials recorded using modified bipolar electrodes placed on 90–100 sites using a calibrated x,y,z-micromanipulator with 0.1 mm precision (Yamamoto et al. 1998). This adopted the output from similar, reference, bipolar electrodes placed on the atrial muscle near the crista terminalis as reference signal. Electrical signals were digitized at 5 kHz by a DigiData 1322A analog to digital converter (Axon Instruments Inc.) and stored on a computer for later analysis. Activation times were given by the time interval separating the initial negative deflections of electrograms from the exploring and reference electrodes with the site showing the earliest activation taken to be the leading pacemaker site. Such electrodes were also used to map the dimensions of the preparation at the outset of each experiment by recording the coordinates of various anatomical landmarks. This permitted a coordinate mapping for the excitation spread from the leading pacemaker site using SigmaPlot 9.0 (Systat Software Inc., Richmond, CA, USA), drawing isochrones at 5 ms intervals.
, 百拇医药
Immunohistochemistry
Immunohistochemistry was performed on tissue sections from adult mice (either sex) as described earlier (Lei et al. 2004). Sections were incubated with rabbit polyclonal anti-Nav1.5 and anti-Nav1.1 (diluted 1: 100; Alomone Laboratories, Israel). For control sections, primary antibodies were pre-incubated with their antigenic peptides, or no primary antibody was used. Image J (NIH) was used for relative fluorescence intensity quantitative analysis. Immunolabelled sections, mounted in Citifluor mounting medium (Agar Scientific, UK), were examined on a Leica DMIRM fluorescence microscope and a TCS-SP2 confocal laser-scanning microscope (Leica Microsystems, Germany) using 488 nm excitation and 500–535 nm emission for Alexa488 labelling, Single optical slices were recorded under consistent pinhole, laser intensity and detector settings (Pawley, 1990).
, 百拇医药
Computer simulations
The possibility that iNa could participate in both the initiation and conduction of pacemaker activity was assessed in computer simulations that were based upon a gradient model of the intact SAN and atrium. This was developed from single cell models of electrical activity of rabbit SAN (Zhang et al. 2000) and atrial cells (Oxsoft HEART, version 4.8. Oxsoft Ltd, Oxford) and used the one-dimensional partial differential equation previously described (Zhang et al. 2000). Though developed for the rabbit heart, the model shares common features with the mouse SAN, in particular the gradient of iNa distribution across the tissue. The modelling assumed a SAN centre of radius 1.5 mm that contained a uniform density of TTX-sensitive neuronal Na+ channels, and began by assuming that these would give a maximum current of 10 pA pF–1. The surrounding SAN periphery was assumed to have an annular radius of 1.5 mm and to contain both neuronal and cardiac Na+ channels at a density that increased towards the periphery giving maximum currents from 10 pA pF–1 to the 60 pA pF–1 towards the periphery as reported below. The remaining ionic current densities were as described experimentally for the rabbit SAN (Honjo et al. 1996, 1999; Lei et al. 2000, 2001). The numerical simulations then investigated the effect of progressively reducing iNa density selectively both in the SAN centre and in the SAN periphery adjacent to the atrium. The pacemaker cycle length (CL) was measured as the time interval between two successive APs. The sinoatrial conduction time (SACT) was measured as the time taken for the excitation wave fronts to travel between two recording sites s1, within the SAN 2 mm from the SAN-atrial border and s2, within the atrium, 9 mm from the SAN-atrial border.
, http://www.100md.com
Data are expressed as means ±S.E.M. (number of animals or preparations or cells). Differences were evaluated by Student's unpaired t test with a P value < 0.05 considered significant.
Results
The mouse SA nodes, located in the intercaval region parallel to the crista terminalis (CT) (Verheijck et al. 2001; Liu et al. 2004), resembled those previously reported in rabbit and other small mammals (Bleeker et al. 1980; Opthof et al. 1987; Boyett et al. 2000). Thus they showed a greater nuclear density and lighter cytoplasm than the surrounding atrial tissue reflecting the smaller size of SA node cells. Such cells nevertheless increased in size from the centre to the periphery. It was thus possible to identify the (300 μm x 150 μm) primary pacemaker area thought typically to contain 450 primary pacemaker cells. The central cells interweave in a circular pattern around the sinus artery and are orientated perpendicular to the CT; cells in the SAN periphery and atrial muscle of the CT are mainly orientated parallel to the CT. The nodal tissue is separated from atrial myocardium at its atrial free wall and septal sides by connective tissue (Liu et al. 2004).
, http://www.100md.com
Expression of Na+ channels in Scn5a+/– SAN cells
Figure 1A–C compare typical tissue sections (A and B), and relative fluorescence intensity plots (C) through the thick CT and thin intercaval region, which contains the SAN, from male WT (n= 5 for Nav1.5, n= 4 for Nav1.1) and male Scn5a+/– (n= 5 for Nav1.5, n= 4 for Nav1.1) hearts labelled using a specific antibody against Nav1.5 or Nav1.1 channels. Immunofluorescence signal from both SAN and surrounding regions from WT and Scn5a+/– SAN tissue sections were compared under identical imaging conditions (Methods) at low and high magnifications using fluorescence confocal microscopy. There was much lower labelling in Scn5a+/– cells consistent with a reduced expression of Nav1.5 following heterozygous disruption of Scn5a (Fig. 1A) exemplified by the results of the semi-quantitative analysis illustrated in Fig. 1C. Scn5a+/– tissue demonstrated considerably lower overall Nav1.5 fluorescence signal in both atrial (63 ± 4% that of WT) and SAN (75 ± 9% that of WT) compared with corresponding regions in WT hearts. Furthermore, the Nav1.5 channel was selectively expressed in peripheral SAN and the surrounding atrial muscle: cells in the centre of the SAN surrounding the SAN artery showed little labelling in either case.
, 百拇医药
A, labelling of Nav1.5 in WT (top) and Scn5a+/– (bottom) SANs. B, labelling of Nav1.1 in WT (left) and Scn5a+/– (right) SANs. C, regional comparison of relative fluorescence intensity in and around the SAN in Nav1.5 in WT and Scn5a+/– SANs. C, centre of the SAN; P, periphery of the SAN; endo, endocardium; epi, epicardium; CT, crista terminalis; SEP, atrial septum. Relative fluorescence intensity was calculated by Image J 1.29x (NIH, USA). Measurements of grey scale pixel intensity are based on the selected area of interests. The pixel intensity is plotted against a fixed distance within the region of interest along the line from CT, peripheral SA node, centre SAN and septum. Scale bar, 100 μm.
, 百拇医药
In contrast, Nav1.1 signal was detected throughout the tissue including the centre and the periphery of the SAN with the SAN and surrounding atrial muscle showing similar expression patterns in the WT and Scn5a+/– hearts (Fig. 1B). It was not possible fully to quantify fluorescence intensity of Nav1.1 in the SAN regions in the WT and Scn5a+/– hearts owing to the presence of non-specific nuclear staining that has been previously reported with use of some Na+ channel antibodies (see Kucera et al. 2002). No labelling above background was obtained when the primary antibodies (anti-Nav1.5 or anti-Nav1.1) or secondary antibody were omitted or were pre-incubated with their antigenic peptides (data not shown). SAN cells nevertheless increased in size from centre to the periphery.
, 百拇医药
Na+ channel properties in Scn5a+/– SAN cells
Isolated SAN cells from two male WT and two male Scn5a+/– mice were identified through their spindle-shaped morphology (Fig. 2A), spontaneous activity and the presence of Na+ currents (iNa) in response to patch clamp voltage steps of 20 ms duration from a –120 mV holding potential to test potentials between –90 and +30 mV. Both WT and Scn5a+/– SAN cells displayed inward Na+ currents with typical activating and inactivating kinetics (Fig. 2B and C). The corresponding current–voltage relationships derived from the maximum iNa (Fig. 2D) demonstrated similar thresholds for iNa activation close to a –70 mV test potential and maximum peak currents at similar voltages (–20 mV) at 37°C. However, Scn5a+/– cells showed significantly reduced peak current densities (Fig. 2D and Table 1) (36 ± 1 pA pF–1, n= 6, Cm= 38–55 pF, 47 ± 2.4 pF) compared with WT myocytes (57 ± 4 pA pF–1, n= 6, Cm= 36–58 pF, 48 ± 3.0 pF, P < 0.01). Moreover, there was a significant correlation between the density of iNa and cell membrane capacitances in WT cells (n= 6, r2= 0.71, P < 0.05) but not in Scn5a+/– cells (n= 6, r2= 0.21, P > 0.05).
, 百拇医药
A, an example of typical mouse SAN pacemaker cell for patch clamping. Scale bar, 20 μm. B and C, recordings of total iNa in a WT and Scn5a+/– cell. D, mean (±S.E.M.) current–voltage relationships for total iNa from WT (n= 6) and Scn5a+/– cells (n= 6). E and F, mean (±S.E.M.; n= 6 for each group) activation and inactivation curves for iNa.
The current–voltage relationships also yielded normalized activation curves (Fig. 2E) for the peak Na+ conductance, gNa. Fitting of the activation (Fig. 2E) and inactivation curves (Fig. 2F) to a Boltzmann equation (Table 1) gave respective values of half-activation voltage, V1/2, and steepness factor, k, in Scn5a+/– myocytes that were statistically indistinguishable from the corresponding values for the WT. These findings suggest that SAN cells from Scn5a+/– mice have reduced Na+ channel densities but similar Na+ channel gating characteristics.
, 百拇医药
Heart rate and SAN function in Scn5a+/– mice
We next studied heart rates and rhythm in ambulant mice (5 male WT, 6 male Scn5a+/–) and electrical activity in isolated SAN preparations (from 4 male and 4 female WT, 4 male and 3 female Scn5a+/– mice) to assess the physiological effects of reduced Na+ channel conductance on SAN pacemaker function. Telemetered, long-term ECG recordings showed that Scn5a+/– (546 ± 24 beats min–1, n= 6) had significantly lower mean heart rates, as recorded over 60 h, compared with WT mice (592 ± 17 beats min–1, n= 5; P < 0.01). Standard ECG studies under anaesthesia showed the occasional Scn5a+/– mouse (2 out of 67 studied) but none of the WT mice (out of 57) showed a persistent sino-atrial (SA) block, which might reflect an extreme manifestation of the sinus node anomalies shown in the Scn5a+/– mice. Figure 3B exemplifies recordings from a (24 week old) Scn5a+/– mouse with completely absent P waves showing a junctional escape rhythm, and Fig. 3C illustrates a trace, recorded at 60 weeks, in which P waves were present only before a minority of the ECG waveforms. These findings might reflect either SA block or sinus arrest. Both these findings contrast with the normal ECG patterns that were shown by the WT (Fig. 3A).
, 百拇医药
Standard lead I surface ECG recordings obtained in anaesthetized WT (A) and Scn5a+/– mice showing either complete (absence of P wave; first recording at 24 weeks old) (B) or partial (second recording at 60 weeks old) SA block with the presence of a few conducted atrial (P) waves (arrows) (C).
Secondly, pacemaker activity and activation patterns were determined from extracellular potential (ECP) recordings that were made from 200 recording sites at a resolution of 0.2–0.3 mm over the SAN and surrounding atrial muscle (Fig. 4A). Spontaneous pacemaker activity began in the SAN centre (the leading pacemaker site). It then propagated to the crista terminalis and interatrial septum. The Scn5a+/– (n= 8) and WT (n= 7) SAN preparations showed similar overall patterns of activation. However, the Scn5a+/– preparations showed a significantly slower conduction towards the SAN periphery and around the SAN–atrial junction indicating a peripheral exit block reflecting slowed conduction (Fig. 4B and C).
, 百拇医药
A, an example of a SAN preparation used for the electrical mapping. Scale bar, 200 μm. B and C, activation sequence in WT and a Scn5a+/– SAN. D and E, action potentials recorded from the sites near the centre of the SAN in A and B are shown from WT (left) and Scn5a+/– (right) SAN preparations. F and G, SAN conduction in WT and Scn5a+/– preparations. Extracellular potentials from sites a (leading pacemaker site in the centre of the SAN; see B and C) and c (atrial muscle, AM) are shown. Vertical dashed lines indicate the time of initiation of the AP at the leading pacemaker site (left) and the arrival of the AP in the atrial muscle (right). H, simultaneous SAN and atrial muscle recordings showing sino-atrial conduction block in Scn5a+/– SAN that was never observed in WT. SEP, septum; SVC, superior vena cava; IVC, inferior vena cava; CT, crista terminalis; and RA, right atrial appendage.
, 百拇医药
Thirdly, intact SAN preparations from WT mice showed regular spontaneous activity with a 170 ± 7 ms (n= 7) cycle length (CL). The Scn5a+/– SAN showed longer CLs (223 ± 14 ms, n= 8, P < 0.05 compared with WT SAN) indicating lower pacemaker rates (Fig. 4D and E). Sino-atrial conduction times (SACTs) obtained from simultaneous ECP recordings (Fig. 4F and G) from the SAN centre (site a: Fig. 4B and C) and neighbouring atrial muscle (site c: Fig. 4B and C) were significantly longer in Scn5a+/– (16 ± 2 ms, n= 8) compared with WT (12 ± 1 ms, n= 7, P < 0.05). Additionally, 3 out of 8 Scn5a+/– SANs but none of the WT preparations also showed evidence of a 3: 2 or 2: 1 SA conduction block (Fig. 4H).
, http://www.100md.com
Finally, those features of spontaneous pacemaker activity that related to cell size, type and current densities (Honjo et al. 1996; Lei et al. 2001) were compared in isolated SAN cells (from 3 male and 1 female WT, 3 male and 1 female Scn5a+/– mouse, respectively). Intracellular recordings from small cells, previously defined as having cell membrane capacitances Cm 30 pF in rabbit SAN cells, presumably from the SAN centre (Honjo et al. 1996, 1999; Lei & Boyett, 1998, Lei et al. 2001; Musa et al. 2002) (Fig. 5A and B), demonstrated similar pacemaker rates in Scn5a+/–(CL = 164 ± 3 ms; Cm, 30 ± 1 pF, n= 3) compared with WT myocytes (CL = 177 ± 2 ms; Cm, 30 ± 3 pF, n= 2). In contrast, larger cells, defined as having Cm > 35 pF, presumably from the SAN periphery (Honjo et al. 1996, 1999; Lei & Boyett, 1998; Lei et al. 2001; Musa et al. 2002) (Fig. 5C and D), demonstrated generally slower pacemaker rates and lower AP peaks in Scn5a+/–(CL = 186 ± 7 ms; Cm, 45 ± 4 pF, n= 4) than in WT myocytes (CL = 161 ± 3 ms; Cm, 46 ± 4 pF, n= 5). However, the average firing rate of SAN cells was not significantly different (P > 0.05, WT versus Scn5a+/–, unpaired t test) between WT (CL = 166 ± 4 ms, n= 7) and Scn5a+/–(CL = 177 ± 6 ms, n= 7). Figure 5E–I plots a range of cellular characteristics that bear on spontaneous pacemaker activity against cell size as reflected in Cm. It confirms that the values of the independent variate Cm explored through the two data sets were not significantly different (P > 5%; unpaired t test). However, WT and Scn5a+/– gave significantly different fitted slopes to a P < 0.1% significance level in the dependence of cycle length (CL), maximum upstroke velocity (MUP), and action potential amplitude (APA), but not maximum diastolic potential (MDP) or the diastolic depolarization rate (DDR), upon Cm consistent with earlier reports in which tetrodotoxin (TTX)-mediated Na+ channel block did not influence electrical activity in small ball preparations from the centre, but influenced spontaneous activity (by 50%) in preparations from rabbit SAN periphery without changing MDP (Kodama et al. 1997). There were thus significant contrasts between the two groups not simply attributable to quantitative anatomical differences between SAN cells in WT and Scn5a+/– mice.
, http://www.100md.com
A and B, spontaneous APs recorded in WT (A and C) and Scn5a+/– cells (B and D). E–I, relationship between maximum diastolic potential (MDP), cycle length (CL), diastolic depolarization rate (DDR), maximum upstroke velocity (dv/dt) and AP amplitude (APA), and cell size as reflected in cell capacitance Cm from WT (n= 7) and Scn5a+/– (n= 7) SAN cells.
INa facilitates SAN function in driving atrial activation
The present findings therefore directly implicate the Nav1.5 channel in action potential conduction through the SAN and from the SAN to the surrounding atrial muscle. Preliminary modelling studies then suggested a possible mechanism that might explain the effect of Nav1.5 channels upon the normal SAN pacemaker rate. Loss-of-function mutations in the Na+ channel alone might then form the entire physiological basis for SND. The possible feasibility for such a scheme was explored numerically in a modified rabbit SAN model for pacemaker function comprising successively coupled cells from the SAN centre, which initiates spontaneous APs, to its periphery, that propagate such APs to reach atrial muscle, which is described in the Methods (Zhang et al. 2000). This predicted that a 35% block of iNa in a 2 mm length of the peripheral part of the SAN, would result in a 41 ms increase in CL (316 ms in control versus 357 ms in the condition of a 35% block of peripheral iNa) implying a significantly reduced pacemaker rate, and an 18.4 ms increase in SACT and reduction of conduction velocity (0.059 m s–1 in control versus 0.047 m s–1 in the condition of a 35% block of peripheral iNa) corresponding to a compromised sino-atrial conduction. Figure 6A–E illustrates modelled AP waveforms in successive myocytes under control conditions (A), following a 100% reduction of iNa confined to the SAN centre (B), and successive 35% (C), 70% (D) and 80% (E) reductions of iNa selectively in the SAN periphery. These decrease pacemaker rate, as reflected in the time interval between successive APs (B), and decrease conduction velocity (C) to result in a conduction exit block in which APs originating from the SAN totally fail to conduct into the atrium (D) and finally a complete termination of SAN pacemaker activity (E).
, http://www.100md.com
A, APs displayed along the length of SAN and atrial string. The AP is first initiated in the SAN centre then propagates via its periphery towards atrium. B, effect of selectively blocking iNa in the SAN centre by 100% increases CL by 11 ms compared with the control. C–E, APs with progressive selective reductions of iNa in peripheral SAN by 35% (C), 70% (D) and 80% (E). AM, atrial muscle.
Figure 7 illustrates an experimental fulfilment of such predictions. Thus, Fig. 7A shows extracellular potentials recorded from the centre of the SA node and from the neighbouring atrial muscle from a WT mouse. Such simultaneous recordings permitted SA node conduction time (time taken for conduction from points a to c in Fig. 4B and C) to be assessed. In such experiments, applications of 10 (n= 5) and 100 (n= 5) nM TTX caused no significant change of SA node conduction. However, increases in TTX concentration from 1 to 5 μM prolonged SA node conduction by 22 ± 7 and 39 ± 4% (P < 0.01, n= 5) in two male and three female WT mice and conduction by 30 ± 5 and 56 ± 9% (P < 0.01, n= 5) in two male and three female Scn5a+/– mice. TTX at 12.5 μM completely blocked SA node conduction.
, 百拇医药
A, effect of TTX on SA node pacemaking. Extracellular potentials recorded from centre of the SAN and periphery of the SAN. B, effect of TTX on sinoatrial conduction. Vertical dashed lines indicate the time of initiation of the action potential at the leading pacemaker site (left) and the arrival of the action potential in the periphery or atrial muscle (right).
Discussion
The present experiments explored electrophysiological roles for cardiac Scn5a-encoded Na+ channels in SAN function using mice with a targeted genetic disruption. They provide the first direct evidence implicating Nav1.5 channels in action potential conduction through the SAN and from the SAN to surrounding atrial muscle and additionally suggest indirect roles for Nav1.5 channels in maintenance of the normal SAN pacemaker rate. They thus corroborate indirect clinical evidence (Tan et al. 2001; Benson et al. 2003; Veldkamp et al. 2003) associating human genetic defects in SCN5A with sinus node disorders (SND). Analysis of mutant channels in stable mammalian cells demonstrated that 2 of 6 mutations associated with human congenital SND result in non-functional Na+ channels with the remaining resulting in mild to severe channel dysfunction (Tan et al. 2001; Benson et al. 2003). Furthermore, the 1795insD, gain of function, mutation has been associated with sinus bradycardia and sinus pauses, in an inherited human long-QT syndrome (LQT3) that also shows phenotypic characteristics of SND (Veldkamp et al. 2003). Thus LQT3 Na+ channel mutations causing persistent inward currents could result not only in QT prolongation but also sinus bradycardia and sinus pauses (Veldkamp et al. 2003). Both loss (Benson et al. 2003) and gain of function mutations (Veldkamp et al. 2003) in Nav1.5 (Scn5a) might thus be implicated in sinus node dysfunction.
, http://www.100md.com
Our experiments first directly demonstrated the clinically observed sinus bradycardia, slow SA conduction and sino-atrial exit block following targeted genetic disruption of cardiac Scn5a-encoded Na+ channels (Asseman et al. 1983), and went on to clarify possible roles for the Scn5a Na+ channel in both conduction and pacing in the mouse SAN system for the first time. Thus both intact Scn5a+/– mice and isolated Scn5a+/– hearts showed depressed mean rates and persistent SA block. Both immunochemical and biophysical analyses demonstrated a reduced expression of SAN Na+ channels: patch clamp studies of Scn5a+/– SAN cells observed similar steady-state activation and inactivation properties but reduced maximum Na+ currents (30%) compared with findings from WT mice. Recent reports describe larger (50%) reductions in iNa in ventricular Scn5a+/– myocytes (Papadatos et al. 2002). However, in contrast to ventricular myocytes, SAN cells express neuronal (Nav1.1, Nav1.3) (Maier et al. 2003) in addition to the adult cardiac Scn5a-encoded Nav1.5 channel isoforms at relative densities that may vary with cell size (Lei et al. 2004). Mouse SAN cells accordingly demonstrate separable, TTX-sensitive (neuronal) and TTX-resistant (cardiac) Na+ currents and this could reconcile the apparent differences between the two cell types (Boyett et al. 2000; Maier et al. 2003; Lei et al. 2004). Although this presence of Nav1.1 in the SAN has been reported in several species (Maier et al. 2003; Lei et al. 2004; Haufe et al. 2005), there is no report concerning its expression in human SAN.
, 百拇医药
Secondly, comparisons of mouse SAN preparations suggested specific roles for Scn5a-encoded Na+ channels in SAN conduction and pacemaking. The presence or otherwise of Nav1.5 channels strongly influenced both upstroke velocity and APA. Correspondingly, although multiple extracellular recordings revealed similar overall activation patterns over the SAN and surrounding atrial muscle, Scn5a+/– hearts showed a slower conduction, both towards the SAN periphery and between SAN and atrium suggesting an SA exit block, giving rise to longer overall conduction times from the leading pacemaker site in the SAN centre to the surrounding atrial muscle, than in WT. Furthermore some Scn5a+/– but no WT SANs showed SA conduction block.
, 百拇医药
Thirdly, the present findings permitted a hypothesis in which cells containing cardiac Scn5a-encoded Na+ channels influenced pacemaker rate through their electrical coupling to primary pacemaker cells in the SAN centre. Thus, Scn5a+/– hearts showed a slowed SAN pacemaker activity. This implicates iNa in the bradycardia observed in both experimental and clinical conditions. Microelectrode AP recordings from sites near the centre of intact Scn5a+/– SAN preparations showed longer cycle lengths than did WT. Furthermore, 3 out of 8 Scn5a+/–, but none of the WT, preparations showed a 3: 2 or 2: 1 SA conduction block. Furthermore, isolated WT and Scn5a+/– SAN cells showed contrasting correlations between cycle length, maximum upstroke velocity and AP amplitude, all characteristics related to spontaneous pacemaker activity, and cell size.
, http://www.100md.com
In contrast, disruption of Scn5a-encoded cardiac Na+ channels did not affect MDP (Fig. 5), findings that parallel previous reports (Kodama et al. 1997; Lei et al. 2004) that implicate Nav1.5 (TTX-resistant) currents in pacemaker action potentials, but not MDP in single SAN cells: block of both TTX-sensitive and -resistant iNa by 30 μM TTX reduced the slope of the pacemaker potential and the threshold potential and increased cycle length without changing MDP (Lei et al. 2004). TTX similarly spared electrical activity in small ball preparations from the centre but decreased take-off potential and upstroke velocity and slowed spontaneous activity (by 50%;) in balls from the periphery of rabbit SAN without changing MDP (Kodama et al. 1997), findings consistent with a higher iNa expression in peripheral node cells (Honjo et al. 1996; Lei et al. 2004). Our results also correlated Nav1.5 current densities with cell capacitance in WT but not Scn5a+/– cells localizing the decreased Nav1.5 expression to peripheral SAN cells in Scn5a+/– mice. Finally, pacemaker rates in Scn5a+/– SAN resembled those of WT in small myocytes but were slower than those of WT in large myocytes.
, http://www.100md.com
Previous studies reported that cells from the centre of the rabbit SAN had more positive take-off potentials, slower upstrokes, longer durations in their action potentials, less negative maximum diastolic potentials, and slower intrinsic pacemaker activity than cells from the periphery (Bleeker et al. 1980; Kodama & Boyett, 1985; Kodama et al. 1997; Boyett et al. 1999. We have suggested that these regional differences in electrical activity can be attributed to regional differences in the intrinsic properties of their contained cells. Thus: (1) single SAN cells from the rabbit show parallel heterogeneities in action potential characteristics (Honjo et al. 1996). Furthermore (2) cell size increases from the SAN centre to its periphery (Bleeker et al. 1980; Masson-Pévet et al. 1984; Oosthoek et al. 1993; Boyett et al. 2000), and (3) the smaller cells show electrical activity that resembles that in the SAN centre whereas electrical activity in (4) the larger cells closely parallels that observed in the SAN periphery (Honjo et al. 1996). (5) We have recently suggested (Liu et al. 2004) that the mouse SAN closely resembles the SAN in other small mammals in structure (Bleeker et al. 1980; Opthof et al. 1987; Boyett et al. 2000; Dobrzynski et al. 2005). It differs histologically from the surrounding myocardium in its greater nuclear density and a lighter cytoplasmic staining and shows a gradient in cell size, arrangement, orientation and connective tissue content from centre to periphery. In particular (6) cell size increases from centre to periphery as expected from the gradient model of SAN organization (Boyett et al. 2000). These studies relate differences in ionic currents densities with cell size to the regional differences in electrical activity within the mouse SAN in the present experiments (Honjo et al. 1996, 1999; Lei et al. 2000).
, 百拇医药
Earlier work had not considered possible roles for iNa in normal or abnormal SAN pacemaker function: micromolar levels of TTX abolishes activity in the SAN periphery, but not its centre (Lei et al. 2004) suggesting that the AP upstroke in the SAN periphery only depends on iNa whereas the AP upstroke at its centre depends instead on iCa,L (Kodama et al. 1997). Nevertheless, some recent reports suggest an involvement of cardiac iNa. Thus, both intact Langendorff-perfused mouse Scn5a+/– hearts (Papadatos et al. 2002) and patients with SCN5A-linked hereditary Lenègre disease (Probst et al. 2003) are bradycardic. Furthermore, increasing cell-cell electrotonic coupling progressively converts independent pacemaker behaviour in rabbit SAN cells to a situation in which SAN and atrial AP frequencies and waveforms became entrained (Verheijck et al. 1998). Besides altering AP conduction in the SAN, decreased AP upstroke velocities might then both affect SAN–atrial coupling as well as depress rhythmic activity in the central pacemaker cells themselves. This hypothesis was tested by a numerical modelling of the initiation and propagation of SAN activity (Zhang et al. 2000) which assumed a constant density of TTX-sensitive Na+ channels in central SAN cells and a positive gradient of TTX-resistant channel densities from the SAN periphery to its boundary with atrial muscle. This analysis predicted that reducing iNa density either in the SAN centre or its periphery significantly decreased pacemaker rate. Reductions in iNa restricted to the SAN periphery, progressively resulted in decreased SAN conduction velocity, an exit block of AP conduction from SAN totally to atrium, and finally a complete termination of SAN pacemaker activity. Such findings corroborate recent reports that block of neuronal Na+ channels by TTX concentrations (100 nM) sufficient to inhibit at least 90% of the Nav1.1 channels (KD, 1–10 nM) reduced heart rate by 65% as assessed in intact heart (Maier et al. 2003), 25% in intact SANs (Lei et al. 2004) and 15% in isolated SAN cells (Lei et al. 2004).
, 百拇医药
Our findings following targeted heterozygous disruption of Scn5a thus directly implicate Nav1.5 channel currents in the conduction of APs through the SAN and from the SAN to atrial muscle and they suggest that these also have indirect effects in the maintenance of normal SAN pacemaker rhythm and rate. Disruption of Scn5a channels appears, firstly, to reduce pacemaker potentials through a mechanism that involves the coupling of electrical events between pacemaker cells and cells surrounding them within the SAN: the resulting impairment of impulse initiation accounts for the observed sinus bradycardia. It also affects impulse propagation through the SAN and between the SAN and atria that may result in conduction block. Thus our findings from the Scn5a+/– mouse SAN model provide a background for further explorations seeking to clarify the role of cardiac-type Na+ channels in normal SAN function in other mammalian (including human) species, for which an understanding of the pathophysiology of SND, the most common indication for pacemaker prescription, might offer potential alternative targets for its clinical management.
, 百拇医药
References
Antzelevitch C, Brugada P, Brugada J, Brugada R, Towbin JA & Nademanee K (2003). Brugada syndrome: 1992–2002. A historical perspective. J Am Coll Cardiol 41, 1665–1671.
Asseman P, Berzin B, Desry D, Vilarem D, Durand P, Delmotte C, Sarkis EH, Lekieffre J & Thery C (1983). Persistent sinus nodal electrograms during abnormally prolonged postpacing atrial pauses in sick sinus syndrome in humans: sinoatrial block vs overdrive suppression. Circ Res 68, 33–41.
, 百拇医药
Benson DW, Wang DW, Dyment M, Knilans TK, Fish FA, Strieper MJ, Rhodes TH & George AL Jr (2003). Congenital sick sinus syndrome caused by recessive mutations in the cardiac sodium channel gene (SCN5A). J Clin Invest 112, 1019–1028.
Bleeker WK, Mackaay AJ, Masson Pevet M, Bouman LN & Becker AE (1980). Functional and morphological organization of the rabbit sinus node. Circ Res 46, 11–22.
Bogdanov KY, Vinogradova TM & Lakatta EG (2001). Sinoatrial nodal cell ryanodine receptor and Na+-Ca2+ exchanger: molecular partners in pacemaker regulation. Circ Res 88, 1254–1258.
, http://www.100md.com
Boyett MR, Honjo H & Kodama I (2000). The sinoatrial node, a heterogeneous pacemaker structure. Cardiovasc Res 47, 658–687.
Boyett MR, Honjo H, Yamamoto M, Nikmaram MR, Niwa R & Kodama I (1999). Downward gradient in action potential duration along conduction path in and around the sinoatrial node. Am J Physiol 276, H686–H698.
Catterall WA (2000). From ionic currents to molecular mechanisms: the structure and function of voltage-gated sodium channels. Neuron 26, 13–25.
, 百拇医药
Cho HS, Takano M & Noma A (2003). The electrophysiological properties of spontaneously beating pacemaker cells isolated from mouse sinoatrial node. J Physiol 550, 169–180.
Dobrzynski H, Li J, Tellez J, Greener ID, Nikolski VP, Wright SE et al. (2005). Computer three-dimensional reconstruction of the sinoatrial node. Circulation 111, 846–854.
Goldin AL (2001). Resurgence of sodium channel research. Annu Rev Physiol 63, 871–894.
, 百拇医药
Haufe V, Cordeiro JM, Zimmer T, Wu YS, Schiccitano S, Benndorf K & Dumaine R (2005). Contribution of neuronal sodium channels to the cardiac fast sodium current INa is greater in dog heart Purkinje fibers than in ventricles. Cardiovas Res 65, 117–127.
Honjo H, Boyett MR, Kodama I & Toyama J (1996). Correlation between electrical activity and the size of rabbit sino-atrial node cells. J Physiol 496, 795–808.
Honjo H, Lei M, Boyett MR & Kodama I (1999). Heterogeneity of 4-aminopyridine-sensitive current in rabbit sinoatrial node cells. Am J Physiol 45, H1295–H1304.
, 百拇医药
Irisawa H, Brown HF & Giles W (1993). Cardiac pacemaking in the sinoatrial node. Physiol Rev 73, 197–227.
Kodama I & Boyett MR (1985). Regional differences in the electrical activity of the rabbit sinus node. Pflugers Arch 404, 214–226.
Kodama I, Nikmaram MR, Boyett MR, Suzuki R, Honjo H & Owen JM (1997). Regional differences in the role of the Ca2+ and Na+ currents in pacemaker activity in the sinoatrial node. Am J Physiol 272, H2793–H2806.
, http://www.100md.com
Kucera JP, Rohr S & Rudy Y (2002). Localization of sodium channels in intercalated disks modulates cardiac conduction. Circ Res 91, 1176–1182.
Lei M & Boyett MR (1998). Inhibition of transient outward current, it(to), by flecainide and quinidine in rabbit isolated sinoatrial node cells. J Physiol 511P, 78–79P.
Lei M, Cooper P, Camelliti P & Kohl P (2002). Contribution of the fast sodium inward current, iNa, to murine sino-atrial node pacemaking. Biophys J 82, 605a.
, http://www.100md.com
Lei M, Honjo H, Kodama I & Boyett MR (2000). Characterisation of the transient outward K+ current in rabbit sinoatrial node pacemaker cells. Cardiovasc Res 46, 433–441.
Lei M, Honjo H, Kodama I & Boyett MR (2001). Heterogeneous expression of expression of the delayed rectifier K+ currents iK,r and iK,s in rabbit sinoatrial node cells. J Physiol 535, 703–714.
Lei M, Jones SA, Liu J, Lancaster MK, Fung SS-M, Dobrzynski H, Camelliti P, Maier SKG, Noble D & Boyett MR (2004). Requirement of neuronal- and cardiac-type sodium channels for murine sinoatrial node pacemaking. J Physiol 559, 835–848.
, http://www.100md.com
Liu J, Dobrzynski H, Boyett MR & Lei M (2004). Sophisticated structural organization is required for normal functioning of murine sinoatrial node. Biophys J 86, 297a.
Maier SK, Westenbroek RE, Schenkman KA, Feigl EO, Scheuer T & Catterall WA (2002). An unexpected role for brain-type sodium channels in coupling of cell surface depolarization to contraction in the heart. Proc Natl Acad Sci U S A 99, 4073–4078.
Maier SK, Westenbroek RE, Yamanushi TT, Dobrzynski H, Boyett MR, Catterall WA & Scheuer T (2003). An unexpected requirement for brain-type sodium channels for control of heart rate in the mouse sinoatrial node. Proc Natl Acad Sci U S A 100, 3507–3512.
, 百拇医药
Malhotra JD, Chen C, Rivolta I, Abriel H, Malhotra R, Mattei LN, Brosius FC, Kass RS & Isom LL (2001). Characterization of sodium channel - and -subunits in rat and mouse cardiac myocytes. Circulation 103, 1303–1310.
Mangoni ME & Nargeot J (2001). Properties of the hyperpolarization-activated current (If) in isolated mouse sino-atrial cells. Cardiovasc Res 52, 51–64.
Masson-Pevet MA, Bleeker WK, Besselsen E, Treytel BW, Jongsma HJ & Bouman LN (1984). Pacemaker cell types in the rabbit sinus node: a correlative ultrastructural and electrophysiological study. J Mol Cell Cardiol 16, 53–63.
, 百拇医药
Mitsuiye T, Shinagawa Y & Noma A (2000). Sustained inward current during pacemaker depolarization in mammalian sinoatrial node cells. Circ Res 87, 88–91.
Musa H, Lei M, Honjo H, Kodama I & Boyett M (2002). Heterogeneous expression calcium handling proteins in the pacemaker cells of the rabbit. J Histochem Cytochem 2002, 311–324.
Nathan RD (1986). Two electrophysiologically distinct types of cultured pacemaker cells from rabbit sinoatrial node. Am J Physiol 250, H325–H329.
, http://www.100md.com
Oosthoek PW, Viragh S, Mayen AE, Van Kempen MJ, Lamers WH & Moorman AF (1993). Immunohistochemical delineation of the conduction system. I: The sinoatrial node. Circ Res 73, 473–481.
Opthof T, De Jonge B, Jongsma HJ & Bouman LN (1987). Functional morphology of the mammalian sinuatrial node. Eur Heart J 8, 1249–1259.
Papadatos GA, Wallerstein PMR, Head CEG, Ratcliff R, Brady PA, Benndorf K et al. (2002). Slowed conduction and ventricular tachycardia after targeted disruption of the cardiac sodium channel gene Scn5a. Proc Natl Acad Sci U S A 99, 6210–6215.
, 百拇医药
Pawley JB (1990). Handbook of biological confocal microscopy. Plenum Press, NY.
Press WH, Flanney BP, Teukolsky SA & Vetterling WT (1986). Numercial recipes. In The Art of Scientific Computing, pp. 523–528. Cambridge University Press, Cambridge.
Probst V, Kyndt F, Potet F, Trochu JN, Mialet G, Demolombe S, Schott JJ, Baro I, Escande D & Le Marec H (2003). Haploinsufficiency in combination with aging causes SCN5A-linked hereditary Lenegre disease. J Am Coll Cardiol 41, 643–652.
, http://www.100md.com
Tan HL, Bink Boelkens MT, Bezzina CR, Viswanathan PC, Beaufort Krol GC, Van Tintelen PJ, Van Den Berg MP, Wilde AA & Balser JR (2001). A sodium-channel mutation causes isolated cardiac conduction disease. Nature 409, 1043–1047.
Veldkamp MW, Wilders R, Baartscheer A, Zegers JG, Bezzina CR & Wilde AA (2003). Contribution of sodium channel mutations to bradycardia and sinus node dysfunction in LQT3 families. Circ Res 92, 976–983.
, 百拇医药
Verheijck EE, Van Kempen MJA, Mike Veereschild M, Lurvink J, Jongsma HJ & Bouman LN (2001). Electrophysiological features of the mouse sinoatrial node in relation to connexin distribution. Cardiovasc Res 52, 40–50.
Verheijck EE, Wilders R, Joyner RW, Golod DA, Kumar R, Jongsma HJ, Bouman LN & Van Ginneken AC (1998). Pacemaker synchronization of electrically coupled rabbit sinoatrial node cells. J General Physiol 111, 95–112.
, http://www.100md.com
Yamamoto M, Honjo H, Niwa R & Kodama I (1998). Low-frequency extracellular potentials recorded from the sinoatrial node. Cardiovasc Res 39, 360–372.
Yu FH & Catterall WA (2003). Overview of the voltage-gated sodium channel family. Genome Biol 4, 207.
Zhang H, Holden AV, Kodama I, Honjo H, Lei M, Varghese T & Boyett MR (2000). Mathematical models of action potentials in the periphery and center of the rabbit sinoatrial node. Am J Physiol 279, H397–H421., 百拇医药(Ming Lei, Catharine Godda)