Distribution of a Persistent Sodium Current Across the Ventricular Wall in Guinea Pigs
Abstract—A tetrodotoxin-sensitive persistent sodium current, IpNa, was found in guinea pig ventricular myocytes by whole-cell patch clamping. This current was characterized in cells derived from the basal left ventricular subendocardium, midmyocardium, and subepicardium. Midmyocardial cells show a statistically significant (P<0.05) smaller IpNa than subendocardial and subepicardial myocytes. There was no significant difference in IpNa current density between subepicardial and subendocardial cells. Computer modeling studies support a role of this current in the dispersion of action potential duration across the ventricular wall.
First indications for a slowly inactivating, persistent sodium current in cardiac cells came from studies on Purkinje fibers of dogs and rabbits.1 2 This was followed by the discovery of slowly inactivating sodium channels in ventricular myocytes of rats.3
Subsequently, a small late sodium current was described in guinea pig ventricular myocytes using single-channel and whole-cell patch clamping.4 This study also suggested a significant effect of the late sodium current on action potential duration in ventricular cells: application of 60 μmol/L tetrodotoxin (TTX) reversibly shortened action potential duration at 95% repolarization (ADP95) by about 10%. Kiyosue et al4 speculated that this was because of block of slowly inactivating sodium channels.
Additional studies investigated the changes in the I-V relation of a slowly inactivating, TTX-sensitive sodium current (IpNa) in rat ventricular myocytes in the absence and presence of hypoxia.5 6 This persistent sodium current increased during hypoxia, and, thus, it was suggested that it could be involved in the development of early after depolarizations (EADs) and arrhythmias during hypoxic states. Most recently, Maltsev et al7 showed that a persistent sodium current (which they called the late sodium current) is present in ventricular myocytes from human midmyocardium in normal donor hearts and heart failure patients. Interestingly, they found a similar (15% to 20%) reduction in action potential duration (APD) after application of 1.5 μmol/L TTX, as previously described4 in guinea pigs. They also showed that TTX abolished EADs in myocytes isolated from heart failure patients. Computer modeling studies fit these data surprisingly well. It has been demonstrated8 that EADs can be induced in a cell model by increasing IpNa; this mechanism also requires that the inactivation curve of the fast sodium current is shifted in the depolarizing direction, as has been observed in an SCN5A missense mutation.9 An increase in late sodium current is also present in some well-known genetic predispositions to arrhythmias, for example, one type of the long-QT syndrome.10
IpNa is distinct from the sodium background current IbNa, which is TTX-insensitive.11 Furthermore, IbNa has a linear voltage dependence, whereas IpNa has an I-V relation similar to that of the fast sodium current. IpNa is also different from the sodium window current,12 which is attributed to the overlap of the activation and inactivation curves of the fast sodium current in this region. The window current is TTX-sensitive but present only at much more negative voltages.
Computer modeling studies suggest that the magnitude of IpNa in guinea pig ventricular myocytes can have very large effects on the action potential plateau duration.13 Therefore, it is important to determine the exact magnitude of IpNa in such cells.
It is well-known that APD varies in myocytes from across the ventricular wall. The slow component of the delayed rectifier current Iks is at least in part responsible for the differences in APD. However, another possible mechanism could be a differential distribution in the transmural levels of IpNa, because it significantly prolongs APD in computer models, a finding supported by the available experimental evidence described above.
The aim of this study was, firstly, to determine the size and current-voltage relation of the persistent sodium current in guinea pig ventricular myocytes and, secondly, to test the hypothesis that differences in transmural distribution of persistent sodium current may contribute to the variation in APD across the ventricular wall.
Materials and Methods
Myocytes from subendocardium, midmyocardium, and subepicardium were prepared as previously described.14
The following solutions were used for all experiments, unless stated otherwise.
External solution (mmol/L): NaCl 130, KCl 5.4, CaCl2 1.8, MgCl2 1, glucose 10, HEPES 30, BaCl2 2, glibenclamide 20 μmol/L, and nifedipine 10 μmol/L; pH 7.4 with NaOH.
Internal solution (mmol/L): NaF 10, CsF 109, MgCl2 1, BAPTA 10, TrisATP 5, TEA-Cl 20, TrisCrP 5, and HEPES 5; pH 7.4 with CsOH.
For the Li-replacement experiments, 130 mmol/L LiCl was used to replace 130 mmol/L NaCl in the external solution, and 10 mmol/L LiCl replaced the internal NaF.
For the cadmium (Cd)-block experiments, 100 μmol/L CdCl was added to the external solution.
The standard Tyrode solution contained (mmol/L): NaCl 140, KCl 5.4, CaCl2 1.8, MgCl2 1.0, Na2PO4 0.33, glucose 5, and HEPES 5; adjusted to pH 7.4 with NaOH; osmolarity: 310 mmol/L.
Myocytes were allowed to settle for ≈5 minutes onto the cover slip of a perfusion chamber situated on the stage of an inverted microscope (Diaphot, Nikon). Initially, the cells were superfused with standard Tyrode solution (flow rate ≈1 to 2 mL/min) at 36°C.
The cells were then voltage clamped in the whole-cell configuration using 1-mm-square borosilicate glass electrodes of 2 to 5 MΩ resistance filled with internal solution. Superfusion with the external solution specified above was started. The currents were recorded using Axon Instruments equipment (Axopatch 200B and Digidata 1200) and software (Clampex, sampling rate 5 kHz, signal Bessel filtered at 1 kHz).
The size of the transient sodium current was monitored every 10 seconds with a 100-ms clamp step from a holding potential of −100 to 0 mV.
After ≈2 minutes, the voltage step-protocol shown in Figure 1⇓ (inset) was applied. TTX 63 μmol/L was allowed to wash on for about 3 minutes, and the protocol was repeated.
For the calculation of membrane capacitance, a voltage ramp (from −40 to −50 mV, slope: 5.5 V/s) was applied, and the capacitance was calculated offline by dividing dQ/dt, which corresponds to the maximal current measured (I0) by dV/dt.
Currents were leak subtracted offline using pClamp 6. The difference currents (control and TTX) were calculated. The size of the difference current was measured as the average between 300 and 350 ms after the onset of the depolarizing pulse. Data collected from the same region of the ventricle were averaged, and SEM was calculated. Student’s t test was performed (using Sigma Plot 5.0) on the averages of the 3 different regions. P<0.05 was taken as statistically significant.
Isolated guinea pig ventricular myocytes from subendocardium, midmyocardium, and subepicardium were voltage-clamped using the ruptured whole-cell patch-clamp technique described above.
Examples of original traces, recorded using the protocol described above, are shown under control conditions and after application of TTX (Figure 1⇑).
Figure 2⇓ shows the magnitudes of the difference currents (before and after application of 63 μmol/L TTX) determined as the average between 300 and 350 ms after the onset of depolarization. The size of the difference currents did not change significantly when measured as the average between 700 and 900 ms. However, the difference currents were larger when measured at 150 ms after the onset of depolarization (data not shown).
The I-V relations indicate that this late TTX-sensitive current (IpNa) is about equally large in subepicardial (n=13) and subendocardial (n=18) cells and smaller in midmyocardial (n=20) cells.
The differences are statistically significant for the midmyocardium and endocardium and midmyocardium and epicardium means between −50 and 0 mV (P<0.01 for the region −50 to −20 mV and P<0.05 for −10 and 0 mV). Subepicardial IpNa is indistinguishable from subendocardial IpNa. The maximal current at membrane potentials of −30 to −20 mV is subendocardium −0.36±0.03 pA/pF, midmyocardium −0.23±0.02 pA/pF, and subepicardium −0.40±0.04 pA/pF.
There was no significant difference between the membrane capacitances of subendocardial (149±8 pF, n=18), midmyocardial (153±8 pF, n=20), and subepicardial (152±10 pF, n=13) cells.
To exclude the influence of the Na+-Ca2+ exchanger on IpNa measurement in our conditions, NaCl was replaced by LiCl. Sodium channels are highly permeable to lithium, which, on the other hand, has been reported to block the Na+-Ca2+ exchanger.15
Ventricular myocytes isolated from bulk ventricular tissue were tested after LiCl was used to replace NaCl in the external and internal solutions. The experiments described above were repeated, and the resultant I-V relation (Figure 3⇓) was found to be very similar to the I-V relation for midmyocardial cells shown in Figure 2⇑ (n=8).
To test the hypothesis that the TTX-sensitive current observed is attributable to influx through TTX-sensitive calcium channels that has recently been described,16 100 μmol/L Cd was added to the external solution. Cd at this concentration has been reported to produce a near-complete block of calcium currents.17 However, the IpNa I-V relationship in the presence of 100 μmol/L Cd is not significantly altered (n=3; average current density at −20 mV with 100 μmol/L Cd: −0.18±0.02 pA/pF versus −0.23±0.02 pA/pF in midmyocardial cells under control conditions).
Additional evidence that IpNa is not carried by calcium ions was obtained under experimental conditions with reduced (0.2 mmol/L) calcium in the external solution. The I-V relation of IpNa obtained under these conditions is qualitatively similar to those obtained in 1.8 mmol/L calcium (n=4; average current density at −20 mV with 0.2 mmol/L calcium: −0.21±0.08 pA/pF versus −0.23±0.02 pA/pF in midmyocardial cells under control conditions).
Previous work5 7 suggested a block of the persistent sodium current by small (0.1 to 1.5 μmol/L) concentrations of TTX in rat and human myocytes. Thus, we determined the sensitivity of IpNa to block by TTX in guinea pig ventricular myocytes (Figure 4⇓). TTX 1.5 μmol/L was sufficient to block IpNa; however, application of 0.5 μmol/L TTX failed to produce a block.
Computer Modeling Analysis
The experimentally obtained I-V relations were fitted by the least-squares method to the general function: I=G(V−E)/(1+exp[V′−V]/k), where I is the current, V is the voltage, G is the maximum IpNa conductance, E is the sodium equilibrium potential, V′ is the voltage at half-maximal activation, and k is the slope factor. For the purposes of the calculation, a reversal potential of +50 mV was assumed to avoid outward sodium current at potentials positive to the experimentally observed reversal potentials (which are around +33 mV). Incorporation of these I-V relationships into the standard guinea pig ventricular cell model in OXSOFT Heart 4.8 produced an increase in APD (Figure 5⇓). The extent of the increase is related to the amplitude of the persistent sodium current. Note that the different IpNa formulations were applied to the same basic cell model, ie, there was no difference in the models used for the 3 regions, except for IpNa. Obviously, this is an oversimplification to illustrate specifically the particular contribution of IpNa to APD differences in subendocardium, midmyocardium, and subepicardium.
The computed change in APD is most likely an underestimation, because in the computation it was assumed that IpNa has no time dependence and was taken as the experimentally observed average current between 300 and 350 ms after depolarization. IpNa is larger if defined as the TTX-sensitive current present earlier on during depolarization. Nevertheless, even relatively small amounts of IpNa would significantly prolong APD (Figure 5⇑). Differences in IpNa across the ventricle could thus contribute to APD variations.
The results presented here establish the presence and relative magnitude of IpNa in guinea pig myocytes from the basal free left ventricle. There is a clear difference in the size of IpNa transmurally. Perhaps surprisingly, IpNa is smallest in the midmyocardium and of about equal size in the subepicardium and subendocardium.
It is interesting to compare our results with those on rat myocytes (B.N. Honen and D.A. Saint, unpublished data, May 2000). There is agreement in both species on the experimental side in showing that there are no large gradients of persistent sodium current between the epicardial and endocardial surfaces. However, when computations are done on the very different action potential shapes in the 2 species, the results are quite different. As shown here in the guinea pig (and presumably also in humans and in other species in which the action potential plateau is high and long lasting), the contribution of IpNa is significantly large. In the rat, however, with a much faster early repolarization phase and a low plateau, computations suggest virtually no contribution to determining action potential duration (P. Noble, unpublished data, May 2000).
IpNa is smallest in midmyocardial tissue, where it has been assumed to be largest, particularly in M cells.18 This discrepancy can be explained in several ways. IpNa could be smallest in M cells despite their long APD, assuming that differences in other currents, such as Ik, are responsible for their long APD. Alternatively, M cells could be a small subset of all midmyocardial cells and, therefore, only occasionally observed in midmyocardial single-cell preparations. In agreement with this, several studies 19 20 21 failed to show significant evidence of M cells in guinea pig, rat, and pig hearts.
IpNa is more TTX-sensitive than INa but has similarly shaped I-V relation. It is identified as a current carried by sodium ions. This is on the basis of the following evidence: IpNa is reversibly blocked by TTX and is still present when sodium is replaced by lithium, a blocker of the Na+-Ca2+ exchanger15 ; it is still present when 100 μmol/L Cd is added, a blocker of Ca2+ channels17 ; and lowering the external calcium concentration from 1.8 to 0.2 mmol/L does not greatly affect the I-V relationship.
Main et al14 observed that 0.1 μmol/L TTX shortens APD90 in guinea pigs by about 36% in subendocardial myocytes, 16% in midmyocardial myocytes, and 20% in subepicardial myocytes. This result only partly matches the computer modeling predictions if one assumes that the principal effect by which TTX shortened APD90 was through block of IpNa. On the basis of the model, one would expect TTX to shorten subendocardial and subepicardial APD about equally and midmyocardial APD the least (Figure 5⇑). One explanation for this difference could be that these authors used apical subepicardial cells and basal subendocardial and midmyocardial cells, whereas the experiments described above were performed solely on cells from the base of the guinea pig heart. If there was a base-to-apex gradient of IpNa, subepicardial cells from the apex would not necessarily have the same amount of IpNa as subepicardial cells from the base. Nevertheless, the effect of TTX on APD seems to be of comparable magnitude to the effect of IpNa on APD in the computer model. In the model, APD90 is shortened by 15% through IpNa block in midmyocardium, 25% in subepicardium, and 21% in subendocardium.
The differences in APD shown in Figure 5⇑ are relatively small, and one could argue that these differences will be masked by the variability in APD observed in single-cell studies, which is often much larger. On a single-cell level, this view is undoubtedly correct. However, it has been shown theoretically22 that when myocytes communicate electrically via gap junctions, the variability in APD is smoothed out and becomes relatively equal for all connected cells. When there are systematic differences between distant but connected regions, APD will vary in a systematic manner between those regions. Thus, even though one could expect the differences computed above to be nonsignificant in the face of random variation in APD in individual isolated myocytes, on a tissue level they are likely to be relevant. This is also the reason that we could not measure the effects of low-dose TTX on APD. The variation between individual myocytes from the same region is much greater than any systematic variations between regions.
In this study, we also noticed a discrepancy between the expected and the observed reversal potentials. The experimentally observed average reversal potential of IpNa lies at +33.5 mV. The calculated reversal potential for sodium at 36°C with the solutions used is about +69 mV. The full reasons for this discrepancy are unclear. Sodium channels are known to show a small (≈10%) permeability to K+ ions, which would account for ≈20 mV of the difference. This would give a reversal potential ≈+50 mV, as assumed in the calculations. An additional factor in the present work is that at positive potentials, IpNa forms a very small fraction of the total current, which is strongly outward in this potential range. Small errors attributable to inadequacy of voltage clamping would make a large difference to the measured reversal potential, particularly if the clamp errors differ in the presence and absence of TTX.
It is still not clear which Na+ channel type is responsible for IpNa. Two types of late channel openings described as burst and background were identified in guinea pig ventricular myocytes and rabbit Purkinje and ventricular myocytes.4 23 These discrete channel openings could represent different modes of function of the same Na+ channel. The burst sodium channel activity decreased with time, and inactivation was complete; however, this was not the case with background channels,3 because these only partly inactivated. Thus, it is possible that fast sodium channels in the background mode are the source of IpNa. Alternatively, it has been proposed that a naturally occurring isoform of the fast sodium current could account for the persistent sodium current.7 The techniques used in the present paper, which enabled IpNa to be recorded in close to normal physiological conditions, do not allow accurate recording of the fast sodium current, so these questions must be left to future studies.
In conclusion, this study establishes the presence of IpNa in guinea pig ventricular myocytes and demonstrates that midmyocardial myocytes have a lower density of IpNa than subendocardial and subepicardial cells. These differences could be important in contributing to the variation in APD across the ventricle.
The work in this laboratory is supported by the British Heart Foundation and the Medical Research Council. We would like to thank our colleagues of the Oxford Cardiovascular Electrophysiology Group and, in particular, Patricia Cooper, Victor Twist, and Dr Peter Kohl for his helpful comments on the manuscript.
- Received May 10, 2000.
- Revision received October 4, 2000.
- Accepted October 4, 2000.
- © 2000 American Heart Association, Inc.
Gintant GA, Datyner NB, Cohen IS. Slow inactivation of tetrodotoxin-sensitive current in canine cardiac Purkinje fibres. Biophys J.
Carmeliet E. Slow inactivation of the sodium current in rabbit cardiac Purkinje fibres. Pflügers Arch. 1987;408:18–26.
Patlak JB, Ortiz M. Slow currents through single sodium channels of the adult rat heart. J Gen Physiol. 1985;86:89–104.
Kiyosue T, Arita M. Late sodium current and its contribution to action potential configuration in guinea pig ventricular myocytes. Circ Res. 1989;64:389–397.
Maltsev VA, Sabbah HN, Higgins RSD, Silverman N, Lesch M, Undrovinas AI. Novel, ultraslow inactivating sodium current in human ventricular myocytes. Circulation. 1998;98:2545–2552.
Noble D, Noble P. Reconstruction of a cellular mechanism of a human genetically based arrhythmia. J Physiol (Lond). 1999;518:P2–P3.
Chen Q, Kirsch GE, Zhang D, Brugada R, Brugada J, Brugada P, Potenza D, Moya A, Borggrefe M, Breithardt G, Ortiz-Lopez R, Wang Z, Anzelevitch C, O’Brien RE, Schulze-Bahr E, Keating MT, Towbin JA, Wang Q. Genetic basis and molecular mechanism for idiopathic ventricular fibrillation. Nature. 1998;392:293–296.
Attwell D, Cohen I, Eisner D, Ohba M, Ojeda C. The steady state TTX sensitive (“window”) sodium current in cardiac Purkinje fibres. Pflügers Arch. 1979;379:137–142.
Sheu SS, Sharma VK, Uglesity A. Na+/Ca2+ exchange contributes to increase cytosolic Ca2+ concentration during depolarization in heart muscle. Am J Physiol. 1986;250:C651–660.
Cole WC, Chartier D, Martin M, Leblanc N. Ca2+ permeation through Na+ channels in guinea pig ventricular myocytes. Am J Physiol. 1997;273:H128–137.
Hobai IA, Bates JA, Howarth FC, Levi AJ. Inhibition by external Cd2+ of Na/Ca exchange and L-type Ca channel in rabbit ventricular myocytes. Am J Physiol. 1997;272:H2164–2172.
Sicouri S, Antzelevitch C. A subpopulation of cells with unique electrophysiological properties in the deep subepicardium of canine ventricle: the M cell. Circ Res. 1991;68:1729–1741.
Shipsey SJ, Bryant SM, Hart G. Effects of hypertrophy on regional action potential characteristics in the rat left ventricle: a cellular basis for T-wave inversion? Circulation. 1997;96:2061–2068.
Bryant SM, Wan X, Shipsey SJ, Hart G. Regional differences in the delayed rectifier current (IKr and IKs) contribute to the differences in action potential duration in basal left ventricular myocytes in guinea-pig. Cardiovasc Res. 1998;40:322–331.
Winslow R, Kimball A, Varghese A, Noble D. Simulating cardiac sinus and atrial network dynamics on the Connection Machine. Physica D.1993;64:281–298.