No Apparent Requirement for Neuronal Sodium Channels in Excitation-Contraction Coupling in Rat Ventricular Myocytes
The majority of Na channels in the heart are composed of the tetrodotoxin (TTX)-resistant (KD, 2 to 6 μmol/L) “cardiac” NaV1.5 isoform; however, TTX-sensitive (KD, 1 to 25 nmol/L) “neuronal” Na channel isoforms have recently been detected in several cardiac preparations. In the present study, we determined the functional subcellular localization of Na channel isoforms (according to their TTX sensitivity) in rat ventricular myocytes by recording INa in control and detubulated myocytes. We found that TTX-sensitive INa (KD, &8.8 nmol/L) makes up 14±3% of total INa in control and ≤4% in detubulated myocytes and calculated that &80% of TTX-sensitive INa is located in the t-tubules, where it generates &1/3 of t-tubular INa. In contrast, TTX-resistant INa is located predominantly (&78%) at the surface membrane. We also investigated the possible contribution of TTX-sensitive INa to excitation-contraction coupling, using 200 nmol/L TTX to selectively block TTX-sensitive INa. TTX decreased the rate of depolarization of the action potential by 10% but did not delay the rise of systolic Ca2+ in the center of the cell (transverse confocal line scan), suggesting that TTX-sensitive INa does not play a role in synchronizing Ca2+ release at the t-tubules; the amplitude of the Ca2+ transient and contraction were also unchanged by 200 nmol/L TTX. The quantity of charge entering via ICa elicited by control or TTX action potential waveforms was similar, suggesting that the trigger for Ca2+ release is not altered by blocking TTX-sensitive INa. We conclude that neuronal INa is concentrated at the t-tubules, but there is no evidence of a requirement for these channels in normal excitation-contraction coupling in ventricular myocytes.
In mammalian cardiac muscle, release of Ca2+ from the sarcoplasmic reticulum (SR) is the key event linking membrane depolarization and mechanical activity during excitation-contraction (EC) coupling.1 Ca2+ influx via L-type Ca2+ channels is the major source of trigger Ca2+, which activates ryanodine receptors in the membrane of the adjacent SR by a process known as Ca2+-induced Ca2+ release.2 Ca2+-induced Ca2+ release may also be triggered by Ca2+ entry via reverse Na-Ca2+ exchange, albeit with lower efficacy.3
Voltage-gated Na channels play an important role in EC coupling by causing the rapid upstroke of the action potential and the propagation of excitation from cell to cell.1 Ten genes encoding the α (pore-forming) subunit of the Na channel have been cloned from different mammalian tissues.4 Until recently, it appeared that the main pore-forming subunit in mammalian cardiac tissue was the NaV1.5 isoform.1 However, although NaV1.5 is highly expressed at the mRNA level in the mammalian heart, other α subunit isoforms have also been detected (NaV1.1, NaV1.2, NaV1.3, NaV1.4, NaV1.6, collectively called neuronal in this study).5–8 These isoforms are preferentially located in the central nervous system and adult skeletal muscle (see Goldin4 for review) and are inhibited by nanomolar concentrations of tetrodotoxin (TTX) (KD, 1 to 25 nmol/L),4 in contrast to NaV1.5 (KD, 2 to 6 μmol/L).4 Therefore, the cardiac isoform is classically called TTX-resistant INa, and neuronal isoforms are called TTX-sensitive INa. Recent immunocytochemistry data show that these neuronal Na channel isoforms are present in murine5,9–13 and canine6 cardiac myocytes. This is supported by electrophysiological recordings showing that TTX-sensitive Na current (INa) represents between 5% and 30% of total INa in diverse cardiac preparations.5,6,10,13
In ventricular myocytes, the subcellular localization of Na channel isoforms is unclear. Some immunocytochemistry data show the neuronal Na channel isoforms located only in the transverse tubules (t-tubules) and NaV1.5 present exclusively on the surface membrane,12,13 whereas other studies show NaV1.5 present in the t-tubules14,15 and the neuronal isoforms at the cell surface.5,6,9 Furthermore, these data do not provide quantitative or functional information regarding the distribution of INa. Because the t-tubules are the main site of EC coupling and underlie synchronous Ca2+ release (see Brette and Orchard 16 for review), this subcellular localization of neuronal Na channels raises the possibility that these channels contribute to t-tubule excitability and hence EC coupling. Indeed, it has been shown recently that block of these neuronal Na channels by a low concentration of TTX (100 and 200 nmol/L) decreases left ventricular contractility in a whole guinea pig heart preparation.13 This unexpected observation has been interpreted as suggesting a distinct role for the neuronal Na channels localized at the t-tubules, in linking depolarization of the surface sarcolemma with EC coupling in ventricular myocytes.13 This hypothesis has important functional implications. However, direct evidence at the single cell level is lacking.
In the present study, we used acute detubulation, which enables us to determine the functional localization of currents,16 in conjunction with the patch clamp technique, to determine the functional localization of the different INa isoforms in rat ventricular myocytes. We found that TTX-sensitive INa is concentrated at the t-tubules. We therefore explored a possible functional role of TTX-sensitive INa by investigating the effect of low concentrations of TTX on key steps of EC coupling in isolated rat ventricular myocytes.
Materials and Methods
Isolation and Detubulation of Rat Ventricular Myocytes
Myocytes were isolated from the ventricles of Wistar rat hearts using a standard enzymatic dissociation protocol. Detubulation was induced by osmotic shock as described previously.17 All experiments were performed at room temperature (&23°C).
Membrane potential and currents were recorded using the whole-cell configuration of the patch clamp technique; settings and properties were as described previously.18 INa was measured in low extracellular sodium ([Na]o) solution (20 mmol/L) to reduce INa and improve voltage control. Low-resistance (1.6±0.1 MΩ, n=49) patch pipettes were used and cell capacitance and series resistance (2.9±0.2 MΩ, n=49) compensated by 85% to 90%. Activation curves were fitted using the Boltzmann equation, a=1/(1+exp((Vm−V1/2)/k)), where a is the activation variable, Vm is the membrane potential, V1/2 is the potential at the current is half activated and k is the slope factor. Action potentials were recorded as previously described.19 The rate of rise of the action potential (dV/dt) was calculated by differentiation using Origin software. In some experiments, the action potential and cell shortening were recorded simultaneously. ICa was measured using Na- and K-free external and internal solutions to avoid contamination by overlapping ionic currents and to allow us to use a physiological holding potential.18 ICa was elicited by action potential waveforms. Ca2+ influx measurements were converted to changes in [Ca2+] as previously described.19
Ca2+ imaging was performed using a laser scanning inverted confocal microscope; settings and properties were as described previously.20 Line-scan images (8-bit) are presented as the original signal. Traces showing the time course of fluorescence are presented as a ratio of fluorescence/background fluorescence (F/F0). The rate of rise of Ca2+ transients recorded from the subsarcolemmal space (SS) and cell center (CC) was calculated by differentiation of F/F0 using Origin software.
Cell Shortening Recording
Cell length was monitored using an edge detection system as previously described,17 and the change in cell length during stimulation was used as an index of contractility.
Data are presented as mean±SEM. The estimate of standard error (ESE) and 95% confidence interval (CI) are given for parameters derived from Hill fitting (using Origin software). Paired or unpaired t tests (2-tailed) were performed as appropriate. P<0.05 was taken as significant.
An expanded Material and Methods section is available in the online data supplement at http://circres.ahajournals.org.
Presence of TTX-Resistant and TTX-Sensitive INa in Rat Ventricular Myocytes
To establish the presence of TTX-sensitive INa, we determined the concentration dependence of TTX block of INa at test potentials of −60 mV and −10 mV from a holding potential of −120 mV. At −60 mV, TTX-resistant Na channels (NaV1.5) are activated, whereas TTX-sensitive Na channels are closed: they are activated at more depolarized potentials.4,5,10,21 At −10 mV, all channels are fully activated. Thus if cells contain TTX-sensitive and TTX-resistant Na channels, the dose-response curves should be different at the 2 test potentials with a deviation at nanomolar concentrations of TTX.5,10 Original recordings of control INa, and block by TTX, at the 2 potentials are shown in Figure 1A. Figure 1B shows the dose-response curves; at TTX concentrations of 10 and 100 nmol/L, a significantly larger fraction of current was blocked at −10 mV, compared with −60 mV. This differential effect of TTX is not attributable to voltage-dependent block of NaV1.5 because heterologously expressed NaV1.5 exhibits the same block at both potentials.5 These results suggest that 2 components of INa, TTX sensitive and TTX resistant, are present in rat ventricular myocytes. The dose-response curve at −60 mV is well fitted with a Hill curve (with a Hill coefficient of 1), yielding a KD of 628 nmol/L (black line, Figure 1B). This value is lower than most reports of the TTX sensitivity of NaV1.5, which are generally in the micromolar range (2 to 6 μmol/L).4 However, it is known that TTX affinity depends on [Na]o.22,23 Our KD is similar to previous reports using low [Na]o in cardiac ventricular myocytes5,13 and after correction for the effect of [Na]o22 is in agreement with the micromolar range reported previously. The dose-response curve at −10 mV is well fitted with a double Hill curve (gray line, Figure 1B): if it is assumed that the KD of TTX-resistant current is 628 nmol/L (and the Hill coefficients of both TTX-resistant and TTX-sensitive currents are 1), this gives a KD for the TTX-sensitive current of 8.8 nmol/L (ESE, 6.4; CI, 27) and the fraction of TTX-resistant and TTX-sensitive currents as 0.86 and 0.14 (ESE, 0.03; CI, 0.11), respectively. The KD of the TTX-sensitive current is consistent with published data (1 to 25 nmol/L),4 even after correction for the effect of [Na]o (which has a similar effect on cardiac and neuronal Na channels).22 The fraction of TTX-sensitive INa (14±3%) is in the range of other cardiac preparations (5% to 30%).5,6,10,13
Characterization of TTX-Resistant and TTX-Sensitive INa in Rat Ventricular Myocytes
INa subtypes were characterized by their differential TTX sensitivity: 100 nmol/L TTX is expected to block &92% of TTX-sensitive INa, with little effect on TTX-resistant INa (&14% block, assuming the KD given above). Figure 2A shows INa recorded at different potentials under control conditions, in the presence of 100 nmol/L TTX and the current blocked by 100 nmol/L TTX (obtained by subtraction, TTX-sensitive INa). Figure 2 shows current-voltage relationships for 8 myocytes (Figure 2B) and activation curves constructed and fitted using a Boltzmann equation (Figure 2C) (see Materials and Methods). TTX-sensitive INa was activated at more positive potentials than TTX-resistant INa: the V1/2 and k values are −43±1 and 8.5±1.7 mV and −55±2 and 7.0±1.2 mV, respectively (P<0.05; Figure 2C). Kinetic analysis showed that the time to peak and inactivation of INa accelerate with increasing amplitude of the test pulse, with TTX-sensitive INa activating and inactivating more rapidly than TTX-resistant INa (Figure 2D and 2E). Thus, 2 distinct currents with different activation and inactivation can be isolated by their sensitivity to TTX. Such differences in activation and inactivation properties are hallmarks of TTX-sensitive INa.21 These data therefore support the idea that TTX-sensitive INa and TTX-resistant INa are carried by different Na channel isoforms in rat ventricular myocytes.
Presence of TTX-Resistant and TTX-Sensitive INa at the Cell Surface of Rat Ventricular Myocytes
We next investigated the presence of TTX-sensitive INa at the cell surface only by recording INa in detubulated myocytes. Cell capacitance was significantly decreased following detubulation (by 31%; Table). However, INa density was not significantly different between control and detubulated myocyte (Table).
We determined the block of INa by TTX in detubulated myocytes using a similar approach as for control myocytes (Figure 3A). Figure 3B shows that the dose-response curve at −60 mV is well fitted by a Hill curve with a KD of 614 nmol/L (black line), a value similar to that obtained for control myocytes. However, assuming a KD of 614 nmol/L for TTX-resistant INa, the dose-response curve at −10 mV is not well fitted with a double Hill curve, probably because a smaller fraction of TTX-sensitive INa was present, and hence blocked at TTX concentrations of 10 and 100 nmol/L, compared with control myocytes. This occurs when TTX-sensitive INa represents ≤5% in the preparation.13 However, if we assume a KD of 614 nmol/L for TTX-resistant and 8.8 nmol/L for TTX-sensitive INa (the values for control myocytes), the data are well fitted by a double Hill yielding a fraction of TTX-resistant and sensitive currents of 0.96 and 0.04, respectively (r2=0.9954; Figure 3B). Thus, TTX-sensitive INa represents a maximum of 4% of total INa in detubulated myocytes compared with 14±3% in control myocytes (see below). This suggests that TTX-sensitive INa is located predominantly in the t-tubules rather than on the surface sarcolemma.
Distribution of INa Isoforms in the Surface Sarcolemma and T-Tubules
Using these data, we calculated the distribution of currents (Table) as previously described.24 We used total INa density at −10 mV, where both TTX-sensitive and TTX-resistant INa are fully activated (Figure 2C). TTX-sensitive INa density was calculated assuming that it represents up to 14% and 4% of total INa at the total sarcolemma and cell surface membrane, respectively (ie, data from Figures 1B and 3⇑B). INa densities in the t-tubules were calculated as the difference in whole-cell current between control and detubulated myocytes. These currents were divided by the difference in membrane capacitance between control and detubulated cells to derive the current density in the t-tubules. In rat ventricular myocytes, 70% of total INa is present at the surface sarcolemma and 30% in the t-tubules, so that total INa density is not different at the surface sarcolemma and the t-tubules (Table). However, 78% of the TTX-resistant INa is located at the surface membrane, where it is &1.6 times more concentrated than in the t-tubules (Table). In contrast, 80% of the TTX-sensitive INa is located in the t-tubules, where it is &9 times mores concentrated than at the surface membrane (Table). To derive the proportion of current generated by specific isoforms at each subcellular location, we corrected the current density to account for the partial block of INa by Co (by &30% for TTX-resistant and &15% for TTX-sensitive INa; see online data supplement). Assuming similar block by Co at the t-tubule and surface membranes, this correction does not alter quantification of subcellular localization (above). However, the corrected values indicate that TTX-sensitive INa accounts for &11% of total INa in rat ventricular myocytes. At the surface membrane, TTX-resistant INa generates nearly all INa (&97%). In contrast, at the t-tubules TTX-resistant INa generates &69% and TTX-sensitive INa represents &31% of INa. This suggests that TTX-sensitive INa may play a role in EC coupling, which occurs predominantly at the t-tubules in rat myocytes. To explore this possibility, we investigated the effect of a low concentration of TTX (200 nmol/L as in other recent work on cardiac preparations6,10,11,13) on key steps in EC coupling. At a concentration of 200 nmol/L, TTX is expected to block &92% of TTX-sensitive INa, with little effect on TTX-resistant INa (&14% block), after accounting for the use of physiological [Na]o, which modifies TTX affinity.22,23
Effect of a Low Concentration of TTX on the Action Potential
We first determined the effect of a low concentration of TTX on the action potential, which initiates EC coupling.1 Action potentials recorded from a representative myocyte before, during, and after application of 200 nmol/L TTX are shown in Figure 4A. As summarized in Figure 4B, the dV/dt (inset Figure 4A) and action potential amplitude, which are indicators of Na channel activity,1 are significantly decreased (by −17±5 V/sec and −3±1 mV, respectively, n=12, P<0.05) in the presence of 200 nmol/L TTX (Figure 4B and 4C). In contrast, action potential duration was not altered (Figure 4D). These data suggest that TTX-sensitive INa influences the shape of action potential. We next investigated whether these changes in the action potential alter synchronization of Ca2+ release at the t-tubules.
Ca2+ Transients Are Not Altered by a Low Concentration of TTX
To investigate the possible role of TTX-sensitive INa in SR Ca2+ release, we recorded transverse confocal line-scan images of Ca2+ transients in myocytes loaded with fluo-3 and field stimulated at 0.5 Hz. If TTX-sensitive INa plays a role in t-tubule excitability and function, SR Ca2+ release at the t-tubules should be altered by 200 nmol/L TTX. Thus, we investigated the characteristics of Ca2+ transients in the SS and CC regions in the absence and presence of TTX. Myocytes bathed in Tyrode solution exhibited temporally and spatially uniform Ca2+ release (Figure 5A, left); the amplitude and rate of rise of the Ca2+ transient (d[Ca2+]i/dt) was the same in the SS and CC (Figure 5B and 5C). This spatial and temporal uniformity of SR Ca2+ release is attributable to the t-tubules.20,25 Surprisingly, myocytes bathed in 200 nmol/L TTX also had uniform SR Ca2+ release (Figure 5A, right); the amplitude and d[Ca2+]i/dt of the Ca2+ transient in SS and CC were similar to those observed in the absence of TTX (Figure 5B and 5C). As a result, the peak systolic Ca2+ (average fluorescence along the line scan) and d[Ca2+]i/dt of the global Ca2+ transient were the same in the absence and presence of 200 nmol/L TTX (Figure 5B and 5C). These results suggest that despite block of TTX-sensitive INa, SR Ca2+ release at the t-tubules is unaltered. To investigate this finding further, we recorded cell shortening, as an index of cell contraction.
Cell Shortening Is Not Altered by Low Concentration of TTX
Cell shortening was first recorded during current-clamp experiments. Application of 10 μmol/L TTX abolished cell contraction (Figure 6A). In contrast, application of 200 nmol/L TTX had no significant effect on cell contraction (Figure 6B, n=5). Because cell shortening was small during these experiments (presumably because of the presence of 0.5 mmol/L EGTA in the pipette solution), we also recorded shortening in field-stimulated cells. Myocytes bathed in Tyrode and Tyrode plus 200 nmol/L TTX had similar amplitude and time to peak of cell shortening (Figure 6C, n=11 and 12, respectively). These results support the idea that blocking TTX-sensitive INa does not alter SR Ca2+ release and, hence, contraction. We next examined the consequences of the changes in the action potential, induced by blocking TTX-sensitive INa, on ICa (the major trigger for SR Ca2+ release; see the introduction) by recording ICa during action potential command waveforms.
ICa Is Not Altered by Change in Action Potential Following Blocking of TTX-Sensitive INa
Figure 7A shows representative recordings of ICa elicited by a control action potential waveform (left) and by an action potential waveform recorded in the presence of 200 nmol/L TTX (right). The action potential waveforms were from the myocyte shown in Figure 4. The time to peak ICa was not significantly different when using the 2 voltage waveforms (Figure 7B, n=8), although a small but significant increase in the amplitude of ICa was observed when using the waveform recorded in TTX (Figure 7C). However, the amount of Ca2+ entering via ICa during the time to peak, which can be doubled to give the effective Ca2+ trigger for SR Ca2+ release,26 and the amount of Ca2+ entering during the action potential, were not significantly different between the 2 voltage waveforms (Figure 7D). These data suggest that the change in action potential characteristics following block of TTX-sensitive INa does not alter ICa, which triggers SR Ca2+ release.
The objective of the present study was to determine the functional subcellular localization of Na channel isoforms and whether neuronal Na channel isoforms participate in EC coupling of ventricular cardiac myocytes at the single cell level.
The method used to detubulate rat ventricular myocytes has been described and validated previously.17,20 This method enables direct determination of the distribution of ion channel function between the surface and t-tubule membranes. Notably, this procedure has no effect on ionic currents measured in atrial myocytes, which lack t-tubules.20 Using this technique, we have shown previously that ICa, Na-Ca2+ exchange, and Na-K pump currents are located predominantly in the t-tubules.18,24 In the present study, we found that total INa is evenly distributed between surface sarcolemma and the t-tubules, in accordance with our previous work.27 Recording INa is prone to voltage clamp errors. To minimize these, we decreased the amplitude of INa by using low [Na]o and worked at room temperature. It remains possible that small voltage errors occurred during our experiments. However, the kinetics of total INa (time to peak and T50 inactivation), which are prone to change because of voltage drop, were independent of series resistance in the range used (≤4.2 MΩ before compensation by 85% to 90%, data not shown), suggesting good voltage control in our experiments.
Subcellular Localization of Neuronal Na Channels in Ventricular Myocytes
We determined the fraction of TTX-sensitive and TTX-resistant INa by fitting with double Hill curves. To date, this method has successfully shown the presence of TTX-sensitive INa in 2 cardiac preparations: sino-atrial and ventricular myocytes from mouse.6,10 Our data show that TTX-sensitive INa accounts for &11% of total INa (after correction for Co block) in rat ventricular myocytes, a value within the range described for other cardiac preparations (5% to 30%) using electrophysiological techniques.5,6,10,13 Our data also provide clarification of recent conflicting immunocytochemistry results regarding the localization of the different Na channel isoforms in ventricular myocytes. Interpretation and quantification of immunocytochemistry data are difficult because of a number of potential problems.16 Furthermore, immunocytochemistry reports protein distribution rather than distribution of function. Our data show that TTX-resistant INa contributes &97% of total INa at the surface membrane, compatible with immunocytochemistry data showing intense labeling of NaV1.5 at the intercalated disk.9,12–14 It remains unclear, however, whether this intense labeling is caused by high sodium channel protein density or represents constant density that, because of membrane folding, gives the appearance of an increased concentration of channel protein. In contrast to some immunocytochemistry studies,11,12 we found that TTX-resistant INa contributes &69% of total INa at the t-tubules, in agreement with the data of others.5,6,14,15 Thus, TTX-resistant INa is the major Na channel isoform generating INa in ventricular myocytes, independent of subcellular location. Our results showed the reverse distribution for TTX-sensitive INa (80% in the t-tubules), in agreement with previous immunocytochemistry data.12,13 However, it remains possible that up to 4% of TTX-sensitive INa (because of the fit in detubulated myocytes; Figure 3B), is present at the surface sarcolemma. These uncertainties will quantitatively slightly alter our conclusion about the distribution of TTX-sensitive INa (ie, up to 100 instead of 80% at the t-tubules) but not our results investigating EC coupling.
Functional Role of Neuronal Na Channels in Ventricular Myocytes
The present study shows marked differential localization of INa isoforms in rat ventricular myocytes. This may, as hypothesized by Maier et al,13 suggest specific physiological roles for TTX-resistant INa (excitability and conduction between cells) and TTX-sensitive INa (t-tubule excitability). However, our experiments show no change in the amplitude or time course of the Ca2+ transient and contraction in isolated ventricular myocytes (Figures 5 and 6⇑) during application of 200 nmol/L TTX. We found that d[Ca2+]i/dt near the cell surface (SS) and at the t-tubules (CC) is similar in control and TTX-treated cells. Computer simulation has shown that a decrease in EC coupling efficiency is expected to alter d[Ca2+]i/dt28; our data therefore suggest no difference in the SR Ca2+ release process at the 2 sites (Figure 5) and that despite inhibition of INa by &31% in the t-tubules, the remaining INa is able to depolarize the t-tubules sufficiently to activate ICa to trigger SR Ca2+ release. We also found that the changes in action potential characteristics induced by blocking TTX-sensitive INa (mainly a small decrease in dV/dt; Figure 4) did not alter the characteristics of ICa (Figure 7). The amount of Ca2+ entering the cell via ICa during the time to peak, which can be doubled to give the effective Ca2+ trigger for SR Ca2+ release26 was the same when using the 2 waveforms. This estimate gives a value very similar to the integral of the first 20 ms of ICa, which was suggested by Fabiato29 to provide the trigger for SR Ca2+ release. Thus, despite the change in the action potential, it seems likely that the trigger for SR Ca2+ release provided by ICa is unaltered, consistent with studies showing that the initial part of action potential repolarization (phase 1) is crucial in the efficiency of ICa to trigger SR Ca2+ release.30 Our results therefore contrast with those of Maier et al,13 which show that low [TTX] decreases ventricular contractility in a whole heart preparation; it is possible that this effect was caused by block of TTX-sensitive INa in other regions of the heart, rather than ventricular myocytes. Indeed, it has been shown recently that the contribution of TTX-sensitive INa is greater in Purkinje fibers than in ventricle (&22%).6
To summarize, the present study shows that TTX-sensitive INa accounts for a small fraction of total INa in rat ventricular myocytes and is concentrated in the t-tubules. However, our data show that at the single cell level, low concentrations of TTX have no effect on EC coupling, providing no evidence of a requirement for neuronal Na channels in EC coupling in ventricular myocytes. It remains possible that neuronal Na channels present in ventricular myocytes act as a safety mechanism, which can be recruited under pathological conditions when the resting membrane potential is depolarized, because of the electrophysiological properties of these currents.
This work was supported by the Wellcome Trust. F.B. is a Wellcome Trust Fellow.
Original received December 12, 2005; revision received January 18, 2006; accepted February 2, 2006.
Bers DM. Excitation-Contraction Coupling and Cardiac Contractile Force. 2nd ed. Dordrecht, The Netherlands: Kluwer Academic; 2001.
Sipido KR, Maes M, Van De WF. Low efficiency of Ca2+ entry through the Na(+)-Ca2+ exchanger as trigger for Ca2+ release from the sarcoplasmic reticulum. A comparison between L-type Ca2+ current and reverse-mode Na(+)-Ca2+ exchange. Circ Res. 1997; 81: 1034–1044.
Haufe V, Cordeiro JM, Zimmer T, Wu YS, Schiccitano S, Benndorf K, Dumaine R. Contribution of neuronal sodium channels to the cardiac fast sodium current INa is greater in dog heart Purkinje fibers than in ventricles. Cardiovasc Res. 2005; 65: 117–127.
Rogart RB, Cribbs LL, Muglia LK, Kephart DD, Kaiser MW. Molecular cloning of a putative tetrodotoxin-resistant rat heart Na+ channel isoform. Proc Natl Acad Sci U S A. 1989; 86: 8170–8174.
Schaller KL, Krzemien DM, McKenna NM, Caldwell JH. Alternatively spliced sodium channel transcripts in brain and muscle. J Neurosci. 1992; 12: 1370–1381.
Dhar MJ, Chen C, Rivolta I, Abriel H, Malhotra R, Mattei LN, Brosius FC, Kass RS, Isom LL. Characterization of sodium channel alpha- and beta-subunits in rat and mouse cardiac myocytes. Circulation. 2001; 103: 1303–1310.
Maier SK, Westenbroek RE, Yamanushi TT, Dobrzynski H, Boyett MR, Catterall WA, Scheuer T. 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. 2003; 100: 3507–3512.
Maier SK, Westenbroek RE, McCormick KA, Curtis R, Scheuer T, Catterall WA. Distinct subcellular localization of different sodium channel alpha and beta subunits in single ventricular myocytes from mouse heart. Circulation. 2004; 109: 1421–1427.
Maier SK, Westenbroek RE, Schenkman KA, Feigl EO, Scheuer T, Catterall WA. 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. 2002; 99: 4073–4078.
Cohen SA. Immunocytochemical localization of rH1 sodium channel in adult rat heart atria and ventricle. Presence in terminal intercalated disks. Circulation. 1996; 94: 3083–3086.
Brette F, Orchard CH. T-tubule function in mammalian cardiac myocytes. Circ Res. 2003; 92: 1182–1192.
Brette F, Salle L, Orchard CH. Differential modulation of L-type Ca2+ current by SR Ca2+ release at the T-tubules and surface membrane of rat ventricular myocytes. Circ Res. 2004; 95: e1–e7.
Brette F, Komukai K, Orchard CH. Validation of formamide as a detubulation agent in isolated rat cardiac cells. Am J Physiol. 2002; 283: H1720–H1728.
Fozzard HA, Hanck DA. Structure and function of voltage-dependent sodium channels: comparison of brain II and cardiac isoforms. Physiol Rev. 1996; 76: 887–926.
Doyle DD, Guo Y, Lustig SL, Satin J, Rogart RB, Fozzard HA. Divalent cation competition with [3H]saxitoxin binding to tetrodotoxin-resistant and -sensitive sodium channels. A two-site structural model of ion/toxin interaction. J Gen Physiol. 1993; 101: 153–182.
Yang Z, Pascarel C, Steele DS, Komukai K, Brette F, Orchard CH. Na+-Ca2+ exchange activity is localized in the T-tubules of rat ventricular myocytes. Circ Res. 2002; 91: 315–322.
Fabiato A. Simulated calcium current can both cause calcium loading in and trigger calcium release from the sarcoplasmic reticulum of a skinned canine cardiac Purkinje cell. J Gen Physiol. 1985; 85: 291–320.
Sah R, Ramirez RJ, Backx PH. Modulation of Ca(2+) release in cardiac myocytes by changes in repolarization rate: role of phase-1 action potential repolarization in excitation-contraction coupling. Circ Res. 2002; 90: 165–173.