Gradient of sodium current across the left ventricular wall of adult rat hearts

Abstract

1. Gradients of ion channels across the left ventricular (LV) wall have been well characterized and it has been shown that disruption of such gradients leads to altered rates of repolarization across the wall, which is associated with the generation of arrhythmias.

2. We have hypothesized that a transmural gradient of I_Na is present and have directly measured this current in adult rat myocytes isolated from both the epicardial and endocardial layers of the left ventricle. Currents were also recorded in right ventricular (RV) myocytes for comparison.

3. Peak inward I_Na currents, at −30 mV, were −49.7 ± 2.5 pA pF⁻¹ (n = 22), −32.9 ± 3.2 pA pF⁻¹ (n = 16) and −49.7 ± 3.7 pA pF⁻¹ (n = 24) for RV, LV epicardial and LV endocardial myocytes, respectively. No differences in the voltage dependence of inactivation, the voltage dependence of steady-state inactivation, or reactivation were reported.

4. Our results demonstrate that a gradient of sodium current density is present across the LV wall of adult rat hearts.

Gradients of ion channels across the LV wall are known to play key roles in normal cardiac function (Antzelevitch et al. 1999) as well as having the potential to generate arrhythmias (Kimura et al. 1990; Geller et al. 1993; Wolk et al. 1999; Yu et al. 1999). It has been well established in several species that the Ca²⁺-independent transient outward K⁺ current (I_to) is greater in the LV sub-epicardial layer compared to the endocardium (Litovsky & Antzelevitch, 1988, 1989; Liu et al. 1993). Additionally, delayed rectifier K⁺ currents (I_K) are also reported to be heterogeneous with significantly less current in myocytes of the mid-myocardium (Liu et al. 1993). Together, these differences combine to establish a heterogeneous pattern of repolarization across the LV wall forming the basis for an upright T-wave (Shimizu & Antzelevitch, 1997). Several disease states such as myocardial ischaemia and reperfusion as well as long-QT syndromes (LQTS) are known to disrupt the relative rates of repolarization, thus increasing the incidence of life-threatening arrhythmias (Vincent, 2000).

Although there are several K⁺ channel mutations associated with LQTS, novel long-QT mutations (LQT3) in cardiac sodium channels have also been identified (Kambouris et al. 1998). Such mutations induce persistent inward sodium current (I_Na,P) and delay repolarization (Kambouris et al. 1998). Interestingly, recent indirect evidence suggests that a transmural gradient of I_Na may exist (Sicouri & Antzelevitch, 1995; Sicouri et al. 1996; Cook et al. 1997; Sakmann et al. 2000; Zygmunt et al. 2001), providing further insight into an additional means by which arrhythmias may be initiated. Such a gradient may also be of importance to ischaemia-reperfusion injury as we have previously shown that hydrogen peroxide, endogenously generated during reperfusion, selectively alters the rate of I_Na inactivation (similar to known sodium channel gene mutations underlying LQT3), dramatically prolonging action potential duration (Ward & Giles, 1997). Together, these studies indicate that in addition to K⁺ currents, I_Na may also contribute to heterogeneous rates of repolarization across the LV wall. In the present study, we characterized a transmural gradient of I_Na in myocytes isolated from the epicardial and endocardial regions of adult rat left ventricle.

METHODS

Cell isolation

LV epicardial and endocardial as well as RV myocytes were isolated from adult Sprague-Dawley rats (200–225 g) using the methods of Ward & Giles (1997) with modifications as indicated. Rats were decapitated (under sodium pentobarbital anaesthesia; 50 mg kg⁻¹i.p.) using protocols approved by the Queen's University Animal Care Committee in accordance with Canadian Council on Animal Care guidelines. For isolation of epicardial and endocardial cell populations the LV free wall was removed following enzymatic digestion and, by fine dissection, either the extreme endocardial or epicardial layer was removed and minced in 10 ml of Tyrode solution (mm: NaCl 140; KCl 5.4; Na₂HPO₄ 1; Hepes 5; glucose 10; MgCl₂ 1; pH adjusted to 7.4 with 1 m NaOH and continuously gassed with 100 % O₂) containing collagenase (0.5 mg ml⁻¹; Yakult, Japan), protease (0.1 mg ml⁻¹; Sigma), bovine serum albumin (3 mg ml⁻¹, Sigma) and CaCl₂ (50 μm). The tissue was gently agitated in a shaking water bath for 10–30 min at 36 °C until single dissociated cells were observed. Once myocytes were present, aliquots were removed at 5 min intervals over 30 min and stored in 4 ml of modified KB solution containing (mm): potassium glutamate 100; potassium aspartate 10; KCl 25; glucose 20; KH₂PO₄ 10; Hepes 5; MgSO₄ 2; taurine 20; creatine 5; and EGTA 0.5; with 0.1 % BSA, pH adjusted to 7.2 with 1 m KOH. RV myocyte cell suspensions were considered to be a homogeneous cell population as we have been unable to demonstrate differences in the ionic currents between the epicardial and endocardial layers of this tissue.

Electrophysiological methods

Pipette resistances ranged from 1.5 to 2.5 MΩ when filled with internal solution. Junction potentials were corrected for. Whole-cell currents and action potentials were recorded with an Axopatch-1D amplifier (Axon Instruments) using pCLAMP 8.0 data acquisition software (Axon Instruments). Membrane potential and whole-cell currents were filtered at 2 kHz and sampled between 5 and 20 kHz depending on the experiment type.

Action potential waveforms were recorded using a standard internal pipette solution containing (mm): KCl 20; potassium aspartate 110; EGTA 10; Hepes 10; MgCl₂ 1; K₂ATP 5; CaCl₂ 1; NaCl 10; pH adjusted to 7.2 using 1 m KOH. Tyrode solution with 1 mm CaCl₂ added was used for superfusion. Action potentials were elicited by 5 ms, 800 pA current injection at a stimulation frequency of 1 Hz.

I_to was recorded using solutions identical to those for action potential recordings. From a holding potential of −80 mV, whole-cell currents were elicited by 500 ms square-wave voltage steps to command potentials between −120 and +60 mV, in 10 mV increments, at a rate of 0.5 Hz. This protocol was repeated with a 100 ms prepulse to −40 mV immediately prior to the test potential to voltage-inactivate I_to. The currents derived by subtracting these protocols represent the isolated I_to.

Whole-cell I_Na was recorded using a pipette solution containing (mm): CsF 120; Hepes 10; MgCl₂ 2; EGTA 10; CaCl₂ 1; NaCl 10; pH adjusted to 7.4 with 1 m KOH. To further isolate I_Na the superfusion solution contained (mm): NaCl 10; Hepes 10; KCl 5; MgCl₂ 1; CaCl₂ 1; CoCl₂ 5; CsCl 5; glucose 10; choline chloride 110; pH adjusted to 7.4 with 1 m NaOH. To ensure adequate voltage control a minimum of 80 % series resistance compensation was applied. I_Na was elicited using 50 ms rectangular voltage steps from a holding potential of −90 mV to command potentials ranging from −100 to 20 mV in 5 mV steps applied at a frequency of 1 Hz.

For some experiments the steady-state voltage dependence of I_Na was examined by applying a 10 s conditioning step to membrane potentials between −140 and −20 mV, in 5 mV increments, followed by a test pulse to −30 mV to activate I_Na. The time dependence of I_Na reactivation was also examined using a typical two-pulse protocol. The conditioning pulse (P1) consisted of a 20 ms square-wave depolarization to −30 mV from a holding potential of −100 mV. The test pulse (P2) was identical and was applied between 3 and 150 ms, in 3 ms increments, following the conditioning pulse. Reactivation data were expressed as the ratio of P2/P1 and fitted to the Boltzmann equation.

Whole-cell currents were statistically analysed using Student's unpaired t test or analysis of variance where appropriate. Differences were considered statistically significant when P < 0.05.

RESULTS

To demonstrate the successful isolation of LV epicardial and endocardial myocytes, I_to was measured in the two cell populations, since it is well documented that I_to is significantly greater in LV epicardial myocytes (Litovsky & Antzelevitch, 1988, 1989; Liu et al. 1993). At +50 mV we report peak I_to as 24.7 ± 3.1 pA pF⁻¹ (n = 10) and 2.8 ± 0.8 pA pF⁻¹ (n = 10) for epicardial (Fig. 1D) and endocardial (Fig. 1F) myocytes, respectively. I_to was also recorded from RV cells (Fig. 1B) for comparison and was found to be 32.3 ± 4.6 pA pF⁻¹ (n = 8). Representative action potential waveforms of LV epicardial, LV endocardial and RV myocytes are also shown in Fig. 1. As expected, LV epicardial and RV-derived myocytes characteristically had shorter action potential durations compared to LV endocardial myocytes. At 90 % repolarization these values were 20 ± 2 ms (n = 7), 22 ± 5 ms (n = 5) and 73 ± 12 ms (n = 8), respectively.

Figure 1. Representative action potential waveforms and Ca²⁺-independent transient outward K⁺ current (I_to) traces in cardiac myocytes.

Action potentials from RV, LV epicardial and LV endocardial myocytes were elicited as outlined in Methods and are shown in A, C and E, respectively. Representative I_to traces for RV (B), LV epicardial (D) and LV endocardial myocytes (F) were recorded at command potentials of −40, 0 and +40 mV. Currents were normalized for cell capacitance.

I_Na was measured using voltage clamp to determine whether there are differences in I_Na across the LV wall and compared to RV myocytes. RV peak inward (at −30 mV) I_Na density was −49.7 ± 2.5 pA pF⁻¹ (n = 22; Fig. 2A). A similar I_Na density of −49.7 ± 3.7 pA pF⁻¹ (n = 24; Fig. 2B) was recorded in LV endocardial myocytes. In contrast, I_Na was significantly lower in LV epicardial myocytes compared to endocardial myocytes, with a peak inward current of −32.9 ± 3.2 pA pF⁻¹ at −30 mV (n = 16, P < 0.05; Fig. 2C). I_Na current-voltage relationships for all three cell populations are illustrated in Fig. 2D. The voltage dependence of the inactivation kinetics was assessed by fitting the data to a single exponential equation and is illustrated in Fig. 3. No differences in calculated time constant (tau) values were found between the three populations of myocytes over the voltage ranges examined.

Figure 2. I_Na densities recorded from myocardial cell populations.

Representative families of RV (A), LV endocardial (B) and LV epicardial (C) sodium currents were elicited using 50 ms rectangular voltage steps from a holding potential of −90 mV to command potentials ranging from −100 to +20 mV in 5 mV steps. D, current-voltage relationship of I_Na density recorded from LV epicardial (▪; n = 16), LV endocardial (□; n = 22) and RV (•; n = 24) myocytes. Values are all normalized to cell capacitance and expressed as means ±s.e.m.* Significantly different (P < 0.05) compared to endocardial myocytes at the same membrane potential.

Figure 3. Voltage dependence of inactivation kinetics for RV (•), LV epicardial (▪) and LV endocardial (□) myocytes.

Tau values were calculated by fitting I_Na traces to a single exponential decay equation and are plotted against command potential. Data points represent the mean ±s.e.m. for n = 10 myocytes.

Steady-state inactivation was assessed to confirm that the LV transmural differences in current density were not due to an altered voltage dependence of inactivation (Fig. 4A). For each individual myocyte, steady-state voltage-dependent inactivation data were fitted to the Boltzmann equation to determine the membrane potential resulting in 50 % inactivation (V_11/2sol;2). LV endocardial and epicardial cells revealed V_11/2sol;2 values of −68 ± 1 mV (n = 10) and −68 ± 1 mV (n = 7), respectively, indicating that the voltage dependence of inactivation of these channels was identical. A V_11/2sol;2 value of −70 ± 1 mV (n = 15) was determined for RV myocytes.

Figure 4. Voltage dependence of sodium current steady-state inactivation and recovery from fast inactivation.

A, plot of steady-state inactivation of I_Na is recorded as voltage versus relative peak current from LV epicardial (▪; n = 7), LV endocardial (□; n = 10) and RV (•; n = 15) myocytes. Data were normalized to peak inward current and expressed as the mean ±s.e.m.B, reactivation data were recorded using a paired-pulse protocol and expressed as the P2/P1 ratio for LV epicardial (▪; n = 13), LV endocardial (□; n = 11) and RV (•; n = 12) myocytes (means ±s.e.m.).

The time course of recovery from fast inactivation was assessed to demonstrate that this parameter was similar for each of the three cell populations (Fig. 4B). P2/P1 data were acquired for individual myocytes at incrementing time intervals and were fitted to the Boltzmann equation to determine the time at which 50 % of I_Na had recovered from fast inactivation (RT₅₀). LV endocardial and epicardial myocytes had derived RT₅₀ values of 6.0 ± 0.3 ms (n = 8) and 5.8 ± 0.2 ms (n = 9), respectively. For RV myocytes, this value was 6.2 ± 0.2 ms (n = 9). Statistical analysis determined that these data were not significantly different from each other.

DISCUSSION

In the present study we have demonstrated that Na⁺ current density is significantly lower in LV epicardial myocytes compared to those isolated from either the LV endocardium or RV and that the activation, inactivation and re-activation kinetics of these currents are similar for the three cell populations. These findings are of significance as they identify a novel LV transmural current gradient and suggest that this gradient is not attributed to differential distribution of Na⁺ channel isoforms.

While there have been no reports on a transmural I_Na gradient, previous studies have examined the possibility of a transmural gradient of I_Na,P. Sakmann et al. (2000) have reported that although I_Na,P is detectable in guinea-pig LV epicardial and endocardial layers there are no differences in current density. Upon closer examination of the macroscopic recordings our results are in agreement in that we also did not detect any difference of I_Na,P across the ventricular wall. However, this may be attributed to the fact that we did not observe I_Na,P in any of our recordings, suggesting either that this current is not present in rat ventricular myocytes or that differences in recording conditions have masked this current. The former is unlikely since we have previously demonstrated that in RV myocytes I_Na,P can be enhanced/activated by exposing the myocyte to the reactive oxygen species hydrogen peroxide, yet this current was not present under control conditions (Ward & Giles, 1997). Clearly, additional studies are required to further explore these discrepancies.

The presence of a I_Na gradient is of potential importance to several disease states. Specifically, LQT3 is known to be associated with mutations in the SCN5A gene encoding I_Na (Bennett et al. 1995; Wang et al. 1996; Abriel et al. 2000). LQT3 mutations prolong the QT interval by inducing a small persistent inward current caused by the reopening of Na⁺ channels (Bennett et al. 1995; Dumaine et al. 1996). Our previous studies have reported similar single channel events relevant to ischaemic heart disease. We have shown that hydrogen peroxide, a reactive oxygen species known to be produced during ischaemia and subsequent reperfusion, also induces a small persistent inward current that we now attribute to I_Na,P (Ward & Giles, 1997). What remains unknown is whether I_Na,P is a distinct ionic current or whether it is a modified conductance state of I_Na. The latter is more likely as it has been shown that phosphorylation of the inactivation gate of I_Na results in a I_Na,P-like current (Qu et al. 1996; Franceschetti et al. 2000) and several drugs and toxins such as BDF9148 and anthopleura toxin also modulate the inactivation kinetics of I_Na to induce I_Na,P (Ward & Giles, 1997; Yuill et al. 2000).

Modulation of repolarization across the LV wall is well described to be an initiator of arrhythmic events (Antzelevitch et al. 1999). The direct evidence presented in this study of a transmural difference of I_Na density between LV epicardial and endocardial myocytes suggests that I_Na is also important with respect to arrhythmia initiation. However, additional studies are required to further reveal the intricate relationship between I_Na and I_Na,P and the role of these currents as modulators of repolarization across the LV wall.