The role of L-type Ca2+ current and Na+ current-stimulated Na/Ca exchange in triggering SR calcium release in guinea-pig cardiac ventricular myocytes (original) (raw)

Abstract

Objective: This study examines the relative ability of sodium current (_I_Na)-stimulated reverse mode Na/Ca exchange and the L-type calcium current (_I_Ca) to trigger calcium-induced calcium release (CICR) in guinea-pig ventricular myocytes. Methods: Cytosolic Ca2+ transients were recorded from enzymatically dissociated guinea-pig ventricular mycocytes using Indo-1. Macroscopic membrane currents were simultaneously recorded using the whole-cell patch-clamp technique. Results: At room temperature (22–25°C) Ca2+ transients were associated with the activation of _I_Na, _I_Ca or _I_Na plus _I_Ca in combination. However, after _I_Ca was blocked by verapamil (10 μM), no Ca2+ transient could be evoked by the activation of _I_Na alone at either −40 or +5 mV. Similar results were obtained with 5 and 8 mM intracellular sodium, and when the temperature of the bathing solution was raised to 35°C and cAMP (10 μM) added to the pipette solution. Conclusions: From consideration of the relative magnitudes of the Ca2+ influx via _I_Ca and Na/Ca exchange and thermodynamic considerations, we suggest that _I_Ca is the major source of ‘trigger’ calcium for CICR (and cardiac contraction) under normal conditions. Although the Na/Ca exchanger was incapable of triggering CICR under the conditions of these experiments, we suggest that it may become more important when cytosolic Ca2+ is elevated, a condition which will also lead to decrease the amplitude of _I_Ca.

Time for primary review 35 days.

1 Introduction

Calcium-induced calcium release (CICR) from the sarcoplasmic reticulum (SR) underpins excitation–contraction coupling in cardiac muscle [1–3]and it has been shown that a simulated calcium current (_I_Ca) may supply sufficient calcium to ‘trigger’ CICR in a skinned cell preparation [4]. The obligatory role of _I_Ca in triggering CICR has, however, been questioned, in light of evidence supporting a role for the sodium–calcium (Na/Ca) exchanger (for review, see Ref. [5]).

The Na/Ca exchanger plays a major role in cardiac calcium homeostasis by extruding calcium from the cytoplasm at rest [6–8]. However, Na/Ca exchange is voltage-dependent, with a reversal potential determined by the sodium (Na) and calcium electro-chemical gradients [6, 7, 9–12]and depolarisation of the cell during the action potential should reverse the direction of exchange to produce calcium influx (see [13]for calculations). Direct support for this idea has come from experiments examining the effects of Na/Ca exchanger modulation on action potential time course [14]and ‘slow inward currents’ [15]. When the exchanger is operating in such a ‘reverse’ mode, the resulting calcium influx may trigger CICR [16–18].

Na/Ca exchange-triggered SR calcium release was first demonstrated in a ‘calcium-overloaded’ cardiac preparation [16]. Exchanger-triggered SR release has also been demonstrated during depolarisation to +100 mV [19, 20], and at less positive potentials in myocytes perfused with a high (20 mM) internal Na+ solution [21]. Under more physiological conditions, SR calcium release has been evoked by reverse-mode Na/Ca exchange when _I_Ca was blocked [5, 18, 22, 23]. It has also been proposed that the Na/Ca exchange may supply a larger fraction of the trigger calcium for CICR than _I_Ca[5, 18]. Additional support for this proposal has come from observations of sodium-current (_I_Na)-triggered CICR, which was explained by _I_Na increasing the [Na] at the cytoplasmic surface of the exchanger which, in turn, accelerated calcium influx via the exchanger to a point where CICR was activated [17, 22, 24].

Although the above evidence supports the idea that the exchanger can trigger CICR under some conditions, a major role for this mechanism has been questioned [19, 25, 26]. In view of the potential importance of _I_Na-stimulated reverse-mode Na/Ca exchange in triggering SR calcium release we have repeated some recent experiments [22]to re-examine the ability of _I_Na-stimulated Na/Ca exchange to trigger CICR at low (5–8 mM) internal Na+ levels. A preliminary account of some of these experiments has been reported previously [27].

2 Methods

2.1 Cell dissociation

The investigation conforms with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23 1985). The isolation procedure was adapted from that reported elsewhere [28]. Isolated hearts from adult guinea-pigs were initially perfused for 5 min (at 37°C) with a nominally calcium-free solution (composition in mM): NaCl, 120; KCl, 5.4; HEPES, 10; pyruvate, 5; glucose, 20; taurine, 20 (pH 7.05 with NaOH). This basic solution was then switched to one containing 4 U·ml−1 protease (Sigma type XXIV) and 100 μM CaCl2 for 3 min, after which the protease in the perfusate was replaced with collagenase (Worthington type II, 1 mg·ml−1) for a further 5–10 min. On completion of the enzyme treatment, the ventricles were cut free and placed in a Petri dish filled with warmed (35°C) enzyme-free isolation solution containing 100 μM CaCl2. The tissue was chopped into small pieces and triturated using a wide-bore pipette. The resulting cell suspension was filtered through gauze, centrifuged briefly and the pellet resuspended. The cells were stored at room temperature (22–25°C) until used.

2.2 Electrophysiology

Cells were transferred to the experimental chamber (200 μl volume), mounted on the stage of a Nikon Diaphot microscope (Nikon Instruments, Japan), and superfused (2 ml/min) with a physiological solution (composition in mM/l): NaCl, 135; KCl, 5.4; MgCl2, 1; CaCl2, 2.0; Glucose, 10; Na-HEPES, 10; NaH2PO4, Na pyruvate, 1; CsCl, 20 (pH 7.4 with NaOH). Experiments were performed at room temperature (22–25°C), or at 35°C as indicated. _I_Na and _I_Ca were recorded in the whole-cell voltage-clamp configuration [29]using an Axopatch 1B patch-clamp amplifier (Axon Instruments Inc., USA). Patch pipettes (1–2.5 MΩ) were pulled from fibre-filled borosilicate glass. The pipette filling solution was (composition in mM/l): CsOH, 100; aspartate, 100; CsCl, 30; NaCl, 5; HEPES, 10; Mg-ATP, 5 (pH 7.2 with CsOH). Contaminating potassium currents were minimised by the presence of Cs+ in both pipette and bath solutions.

Voltage-clamp command potentials were generated via pCLAMP data acquisition software (Axon instruments Inc., USA), driving a TL-1 analogue to digital converter (Axon Instruments Inc., USA). All electrical records were digitised and recorded on videotape, or collected and stored in a computer using the pCLAMP data acquisition software. A dual sampling rate was used to increase the sampling frequency at the beginning of the record and allow accurate measurement of the peak _I_Na, _I_Ca and the rising phase of the calcium transient.

2.3 Measurement of intracellular calcium

The Ca2+ indicator Indo-1 (25 μM) was added to the pipette solution, and injected into the cell by brief pulses of positive pressure. A low concentration of the dye was used to minimise calcium buffering [30]at the expense of some reduction in the signal-to-noise ratio. An epifluorescence illumination wavelength of 360 nm was used, derived from a xenon arc lamp. Emitted light was recorded at 410 and 510 nm wavelengths. Cell autofluorescence was measured after seal formation (before entering the whole-cell configuration) and subtracted from all data. Changes in fluorescence are presented as the 410 nm/510 nm ratio.

2.4 Conditioning protocols

In this investigation we used two different conditioning protocols in order to maintain a defined calcium-loaded state of the sarcoplasmic reticulum:

(1) When _I_Ca was active, 10 conditioning pulses were applied (250 ms, 0.75 Hz) from a holding potential (VH) of −90 to +5 mV to load the SR. After these conditioning pulses, a 10 s rest period was allowed at post-conditioning potential of −90 mV. Test pulses of −40 mV and +5 mV were then used to elicit _I_Na alone or a combination of _I_Na and _I_Ca, respectively. _I_Ca was ‘selectively’ activated by inactivating _I_Na with a post-conditioning potential of −40 mV and then applying a test pulse to +5 mV.

(2) When _I_Ca was blocked, reverse-mode Na/Ca exchange was used to maintain the SR Ca2+ load, by applying depolarising steps from −40 to +60 mV (250 ms, 0.75 Hz). After these pulses, a 10 s rest was allowed at −40 mV. Since the efficacy of verapamil block of _I_Ca was reduced at potentials negative to −40 mV, a brief (50 ms) pre-pulse to −90 mV had to be given to remove _I_Na inactivation without affecting the block of _I_Ca.

2.5 Chemicals

Indo-1 was obtained from Molecular Probes Inc. (Eugene, USA). Verapamil, c-AMP and all other chemicals were obtained from SIGMA Chemical Co. Ltd. (Poole, Dorset, England).

3 Results

3.1 INa- and ICa-induced calcium transients

Fig. 1A shows the activation of calcium current (_I_Ca, upper panel) at +5 mV from a post-conditioning potential of −40 mV at room temperature (22–25°C). The peak amplitude of _I_Ca was 7.51 pA/pf and was followed, upon repolarisation, by an inward tail current typical of that produced by ‘forward mode’ Na/Ca exchange [31]. The lower panel shows the concomitant Ca2+ transient triggered by _I_Ca, which was characterised by a rapid rising phase, a sustained plateau and an exponential decline after repolarisation. A test pulse to −40 mV from a post-conditioning potential of −90 mV resulted in the rapid activation and inactivation of a large ‘Na current’ (_I_Na) (Fig. 1B). The lower panel shows the _I_Na-evoked Ca2+ transient. It is notable that this Ca2+ transient was smaller than that activated by _I_Ca and declined during the test pulse. Fig. 1C shows the combined activation of _I_Na and _I_Ca (upper panel) at +5 mV from a post-conditioning potential of −90 mV. The peak amplitude of the inward current was twice that obtained by activation of _I_Na at −40 mV, but the evoked calcium transient (lower panel) was of comparable amplitude and time course to that evoked by _I_Ca alone (panel A). Similar results were obtained in 14 other cells. These results suggest that SR calcium release is principally activated by _I_Ca and that _I_Na-stimulated reverse-mode Na/Ca exchange does not further increase SR calcium release.

Fig. 1

Activation of _I_Ca, _I_Na and combined activation of _I_Na and _I_Ca evoke a Ca2+ transient in ventricular myocytes. In order to provide a defined Ca2+-loaded state of the SR, a conditioning protocol was applied prior to the test pulse in A, B and C (see Section 2, protocol 1). (A) Activation of _I_Ca alone (upper panel) by a test pulse to +5 mV after a 10 s rest at a post-conditioning potential of −40 mV (to inactivate _I_Na). Lower panel shows the _I_Ca-evoked Ca2+ transient. (B) activation of _I_Na alone (upper panel) by a test pulse to −40 mV after a 10 s rest at a post-conditioning potential of −90 mV. Lower panel shows the _I_Na-evoked Ca2+ transient. (C) Combined activation of _I_Na and _I_Ca (upper panel) by a test pulse to +5 mV after a 10 s rest at a post-conditioning potential of −90 mV. Lower panel shows the Ca2+ transient evoked by the combination of _I_Na and _I_Ca. The transients in this and all subsequent figures are shown as the 410/510 nm fluorescence ratio of Indo-1. Panels A, B and C were recorded from the same cell at room temperature (22–25°C).

3.2 INa- and ICa-induced transients are verapamil-sensitive

In the presence of verapamil (10 μM), _I_Ca was blocked by repeated depolarization using the conditioning protocol (see Section 2). Fig. 2A shows that with _I_Ca blocked in this way: (1) a test pulse from −40 mV to +5 mV failed to activate any discernable inward current; (2) there was no Na/Ca tail current upon repolarisation (upper panel); (3) no measurable calcium transient was evoked (lower panel). It is therefore apparent that verapamil completely abolished _I_Ca-induced SR calcium release. Fig. 2B shows that with _I_Ca blocked, ‘selective’ activation of _I_Na by stepping from −90 to −40 mV did not activate SR calcium release (lower panel). As might be expected, no calcium transient was observed when the membrane potential was stepped from −90 to +5 mV (Fig. 2C). Similar results were obtained in every cell examined (n = 14), and suggest that the small Ca2+ transient evoked by ‘selective’ activation of _I_Na (Fig. 1B), could be explained by threshold activation of _I_Ca rather than by _I_Na per se. These findings differ from those of the previous study [22]since the _I_Na-evoked Ca2+ transient was verapamil-sensitive and smaller than the _I_Ca-evoked transient. We therefore examined the extent to which these contrasting results might be accounted for by some inevitable loss in voltage control during _I_Na[19, 26].

Fig. 2

Verapamil sensitivity of _I_Ca- and _I_Na-evoked Ca2+ transients. With _I_Ca blocked by addition of verapamil (10 μM) to the bath solution, the SR was loaded by repeated activation of reverse mode Na/Ca exchange (see Section 2, protocol 2). (A) No inward current (upper record) or Ca2+ transient (lower record) was evoked by a test pulse to +5 mV after a 10 s rest at a post-conditioning potential of −40 mV. (B) Activation of _I_Na (recovered by a 50 ms prepulse to −90 mV) by a test pulse to +5 mV after a 10 s rest at a post-conditioning potential of −40 mV (upper panel). The lower panel shows that _I_Na failed to evoke a Ca2+ transient when _I_Ca was blocked. (C) Stepping from −90 to +5 mV also fails to evoke a calcium transient in the presence of verapamil. All panels were recorded at room temperature (22–25°C) from the same cell as in Fig. 1.

When _I_Na is activated at −40 mV, voltage escape could cause the membrane potential to overshoot the desired command potential and thereby enter the activation range of _I_Ca. Fig. 3A shows that, in the absence of verapamil, activation of _I_Na at −50 mV (upper panel) can produce an (apparent) inward current of 216 pA/pF, which is of comparable magnitude to that reported in other studies [22, 24]. However, in the same cell, when the series resistance (_R_s) was reduced (from 5.8 to 1.8 MΩ by application of positive pressure pulses to the pipette and electronic compensation), the time course of _I_Na was quite different (n = 3). As shown in Fig. 3B, the peak current was reduced to 11 pA/pf and inactivated (non-exponentially) with a half-time of 33 ms, in reasonable agreement with results obtained using an oil gap voltage-clamp method [32]. (Mitsuiye and Noma [32]reported that _I_Na inactivated with two exponential time constants of 10 and 55 ms at −40 mV. At −50 mV slightly slower time constants would be expected—as observed here). These results suggest that, at least under some conditions, voltage escape may allow sufficient activation of _I_Ca to cause SR Ca2+ release (and contraction). In connection with this point, it is notable that the small Ca2+ transient observed before _R_s compensation was abolished when _R_s was reduced (lower panels). In any case, the inability to clamp _I_Na (as shown by the very rapid inactivation of _I_Na at −40 mV) would always be problematic for such experiments.

Fig. 3

Voltage escape and the _I_Na-evoked Ca2+ transient. At a holding potential of −90 mV the SR was loaded by activation of _I_Ca as described in Fig. 1. (A) Activation of _I_Na at −50 mV after a 10s rest at a post-conditioning potential of −90 mV (upper panel), and the _I_Na evoked Ca2+ transient (lower panel). The series resistance was 5.81 MΩ, and was uncompensated. (B) Activation of _I_Na at −50 mV from the same cell as in A (upper panel), after the series resistance was reduced by applying pulses of positive pressure to the pipette, and the remainder reduced by series resistance compensation (leaving 1.8 MΩ uncompensated). The magnitude of _I_Na was substantially reduced under these conditions and failed to evoke a Ca2+ transient (lower panel). Records were obtained at room temperature (22–25°C).

3.3 Effects of increasing the SR Ca2+ content and temperature

Although conditioning protocols were employed to ensure a defined SR Ca2+ load, it is possible that increasing the SR Ca2+ content might enable a smaller trigger influx (via Na/Ca) exchange to activate SR Ca2+ release [2, 4, 24, 33–35]. In addition, Ca2+ transport by the Na/Ca exchanger is very temperature-sensitive (_Q_10=3–4 [36]; see also [19]). We therefore examined the ability of _I_Na to evoke SR Ca2+ release in the presence of 10 μM cAMP (which stimulates SR Ca2+ uptake) and at 35°C.

The addition of cAMP to the pipette solution at an experimental temperature of 35°C produced somewhat different results. Fig. 4 shows Ca2+ transients (lower panel) induced by activation of _I_Ca alone at +5 mV (panel A) _I_Na alone at −40 mV (panel B) and combined activation of _I_Na and _I_Ca at +5 mV (panel C). The Ca2+ transient evoked by _I_Ca (panel A) rose rapidly to a peak and then declined to a plateau level during the test pulse. In contrast to results shown above, the activation of _I_Na alone at −40 mV evoked a Ca2+ transient of equivalent amplitude to that evoked by _I_Ca, but which decayed almost to baseline levels within the duration of the test pulse (Fig. 4B). Furthermore, combined activation of _I_Na and _I_Ca evoked a much larger Ca2+ transient (Fig. 4C) than did _I_Ca (Fig. 4A) although, in each case, the Ca2+ transients declined to a similar plateau level during the test pulse. In fact, the Ca2+ transient evoked by combined activation of _I_Na and _I_Ca (Fig. 4C) appeared to be the sum of the Ca2+ transients evoked by the selective activation of _I_Ca and _I_Na. Apart from the different experimental conditions, these findings are in general agreement with those reported previously [22], except that the Ca2+ transient evoked by the combination of _I_Na and _I_Ca decayed rapidly during the test pulse, and there was no detectable difference in the kinetics of the rising phase of the Ca2+ transient evoked by either protocol (Fig. 4D).

Fig. 4

At a holding potential of −90 mV, the SR was loaded by repeated activation of _I_Ca, as described in Fig. 1. (A) Activation of _I_Ca alone (upper panel) by a test pulse to +5 mV after a 10 s rest at a post-conditioning potential of −40 mV (to inactivate _I_Na). Lower panel shows the _I_Ca-evoked Ca2+ transient. (B) Activation of _I_Na alone (upper panel) by a test pulse to −40 mV after a 10 s rest at a post-conditioning potential of −90 mV. Lower panel shows the _I_Na-evoked Ca2+ transient. (C) Combined activation of _I_Na and _I_Ca (upper panel) by a test pulse to +5 mV after a 10 s rest at a post-conditioning potential of −90 mV. Lower panel shows the Ca2+ transient evoked by the combined activation of _I_Na and _I_Ca. (D) Comparison of the rise time kinetics of the Ca2+ transients evoked by activation of (1) _I_Ca, (2) _I_Na and a (3) combination of _I_Na and _I_Ca. The records in panels A, B, C and D were recorded from the same cell at 35°C. The pipette solution included cAMP (10 μM).

The greater magnitude of the Ca2+ transient evoked by the combined activation of _I_Na and _I_Ca could be due to several factors: (1) the magnitude of _I_Ca activated from −90 mV is greater than when activated from −40 mV [15, 37]; (2) voltage escape during the combined activation of _I_Na and _I_Ca would lead to a more rapid depolarisation of the membrane and therefore a greater rate of Ca2+ channel activation; (3) the larger Ca2+ transient could be the product of two different Ca2+ influx pathways, one being _I_Ca and the other being activated by _I_Na independently of _I_Ca (i.e., reverse-mode Na/Ca exchange).

Fig. 5 shows that when _I_Ca was inhibited by verapamil (10 μM), there was a profound reduction in the amplitude of the Ca2+ transient. Under these conditions, a test pulse from −40 to +5 mV activated a small residual _I_Ca which evoked a small Ca2+ transient (Fig. 5A). Although a small Ca2+ transient was also observed in response to the activation of both _I_Na and _I_Ca by a test pulse from −90 to +5 mV (Fig. 5C), it was of comparable amplitude to that observed with the selective activation of the residual _I_Ca (Fig. 5A). Since there was no transient observed during activation of _I_Na alone at −40 mV (Fig. 5B), the _I_Na-evoked Ca2+ release was verapamil-sensitive and explainable by a small unblocked _I_Ca, rather than being due to _I_Na per se. This result is in direct conflict with the results of a previous study [22], although it is in general agreement with the conclusions of Sipido et al. [25].

Fig. 5

With a holding potential of −40 mV, _I_Ca was blocked by addition of verapamil (10 μM) to the bath solution at 35°C; the pipette solution contained 10 μM cAMP. The SR was loaded by repeated activation of reverse-mode Na/Ca exchange as described in Fig. 2. (A) A small residual inward current (upper panel) and a small, concomitant Ca2+ transient (lower panel) were evoked by a test pulse to +5 mV after 10 s rest at a post-conditioning potential of −40 mV. (B) Activation of _I_Na (recovered during a 50 ms prepulse to −90 mV, after a 10 s rest at a post-conditioning potential of −40 mV) by a test pulse to −40 mV (upper panel). Lower panel shows that when activated at −40 mV, _I_Na failed to evoke a Ca2+ transient when _I_Ca was blocked. (C) Activation of _I_Na (recovered by a 50 ms prepulse to −90 mV) by a test pulse to +5 mV after a 10s rest at a post-conditioning potential of −40 mV (upper panel). Lower panel shows that, when _I_Ca was blocked, activation of _I_Na at +5 mV failed to evoke a Ca2+ transient greater than that evoked in the absence of _I_Na (A, lower panel).

4 Discussion

In the experiments reported here, we have failed to demonstrate that _I_Na-stimulated reverse-mode Na/Ca exchange can trigger SR Ca2+ release, despite reproducing the experimental conditions used in a previous study [22]which suggested that _I_Na activation at −40 mV could evoke a larger SR Ca2+ release than _I_Ca. Our failure to observe Na/Ca exchange-triggered SR release immediately suggests that the Na/Ca exchanger is less able to trigger SR calcium release than _I_Ca at 5 mM internal [Na+].

It is possible that the method used to assess SR Ca2+ release may have some bearing on our failure to detect any _I_Na-stimulated SR release. In the previous study [22], a confocal microscope was used to measure local [Ca2+]i changes whereas we used conventional wide-field microscopy. Since the local light levels are very high in confocal microscopy, it is possible that there may be some local relief of nifedipine block (as nifedipine is light-sensitive) which would not be detected in the whole cell current record. However, this complication by itself does not explain why Li substitution blocked the _I_Na-evoked transient. Simple interpretation of the Li-substitution experiments may be complicated by a Li-induced increase in resting [Ca2+]i (e.g., [6, 7]) which may inhibit _I_Ca, and thereby decrease spurious calcium channel activation during the voltage escape that accompanies activation of _I_Na.

4.1 INa escape

A major problem in studying the role of _I_Na in activating contraction is the difficulty of obtaining adequate voltage control during the activation of the large and fast _I_Na. Indeed, Bouchard et al. [26]showed that, without series resistance compensation, a series resistance of 6.7 MΩ results in a serious loss of voltage control and suggested that many of the previous reports of _I_Na-activated Ca2+ transients could be explained by the membrane potential escaping to a point where _I_Ca was activated. This view is supported by our findings (Fig. 3) and those of others [19, 25, 26]. Adequate voltage control over _I_Na will always be problematic for these types of experiments and the current records that we (and others) have obtained at −40 mV do not reflect the true time course of _I_Na. Nevertheless, a lack of voltage control does not, by itself, explain all the observations of Lipp and Niggli [22]or Levesque et al. [24]since a loss of voltage control should have occurred during lithium exposure also.

4.2 Thermodynamics of Na/Ca exchange during E-C coupling

Calcium entry on the exchanger is determined by the membrane potential (_E_m) and the sodium (_E_Na) and calcium (_E_Ca) electrochemical gradients (e.g., [6, 9, 10]). As an equation:

formula

where (_E_Na/Ca) is the exchanger equilibrium potential. It has been suggested that normal E-C coupling results from a 100–250-fold increase in [Ca2+]i in the space between the SR and T-tubule membranes (the diadic cleft) [38, 39]which implies that for normal resting levels of [Ca2+]i, the local [Ca2+]i that activates SR calcium release must be ∼6–15 μM.

Fig. 6A shows the relationship between _E_Na/Ca and [Na]i for the exchanger to achieve a local [Ca2+]i of 6–15 μM (shaded region). At 10 mM [Na]i, the membrane potential would have to be >+60 mV for the exchanger to produce such a trigger [Ca2+]i level while at 20 mM [Na]i, the trigger [Ca2+]i could be reached at membrane potentials >+10 mV—predictions in agreement with the results of Sham et al. [19]and Nuss and Houser [21], respectively. Between −50 and −40 mV (the test potentials used by Lipp and Niggli [22]and in this study) [Na]i would have to be >35 mM to achieve a 6–15 μM trigger [Ca2+]i. Therefore, if [Na]i is clamped to ∼5 mM by the patch pipette, thermodynamics show that it is impossible to achieve a normal trigger [Ca2+]i level via the Na/Ca exchanger. However, it has been suggested that _I_Na may locally increase [Na]i to enable Na/Ca exchange to achieve such trigger [Ca2+]i levels [40].

Fig. 6

(A) Relationship between the reversal potential for Na/Ca exchange and internal Na to achieve various trigger calcium levels. To meet or exceed a trigger level between 6 and 15 μM at −40 mV the exchanger would have to operate in the shaded area of the graph showing that internal Na would have to be >35 mM. If the trigger calcium level is only 1 μM, internal Na would have to be >18 mM. (B) Calculations of the Na levels that will occur across the diad during the activation of a 4 pA Na channel situated in the centre of the diadic space (modelled as a circular region 150 nm in diameter and 15 nm high) at steady state. Curves correspond to diffusion coefficients that are 50, 21, 14, 7 and 5% of those in free solution (corresponding to diffusion coefficients of 7×10−6, 3×10−6, and 2×10−6, 1×10−6 and 7×10−7 cm2·−1, respectively) are shown (note that both axes have logarithmic scales). The levels of intracellular Na that would be required to achieve trigger levels of 1 and 6 μM at −40 mV are shown. Note that such trigger levels can only be achieved by the exchanger if the diffusion of Na is restricted and if the exchanger is very close to the Na channel (i.e., within about 10 nm).

To examine this point, the diadic cleft was modeled as a thin disk, 15 nm thick (the space between the SR and T-tubule membranes) and 150 nm in diameter with a single Na channel in its center (only one channel was included because the Na current density suggests that there are only ∼5 Na channels per μm2). The Na diffusion coefficients were reduced to 50, 21, 14, 7 and 5% of those observed in free solution to allow for the possibility of restricted diffusion. As shown in Fig. 6B, activation of a Na channel with a 4 pA single channel current would result in an appreciable increase in the local [Na]i. However, the required local increase in [Na]i (38 mM) is only achieved in the immediate vicinity (i.e., within 4 nm) of the Na channel even with highly restricted diffusion. In other words, the Na/Ca exchangers would have to be: (1) tightly packed around Na channels for the local accumulation of [Na]i to be sufficient to achieve a trigger [Ca2+]i level of 6 μM as well as (2) Na diffusion being highly restricted. In addition, these increases in [Na]i only persist during the open time of the sodium channel, so the exchanger would also have to be kinetically capable of achieving thermodynamic equilibrium very rapidly (since 10 nm from the channel [Na]i takes only 8 μs to fall to 6.5 mM after channel closure). These considerations suggest that it is unlikely that the exchanger could achieve a trigger [Ca2+]i level of 6 μM at −50 mV.

Nevertheless, Lipp and Niggli [22]were able to evoke SR calcium release, so we are forced to conclude that some other factors must have increased the ability of the exchanger to achieve the required trigger [Ca2+]i level and/or the trigger [Ca2+]i level must have been reduced. Under conditions of calcium overload, propagating waves of SR calcium release occur (e.g., Refs. [33, 41]), implying that the required trigger calcium level could be reduced to ∼1 μM in calcium overload. To achieve this level of trigger [Ca2+]i at −40 mV with the exchanger, internal Na would have to be about 18 mM (see Fig. 6A), which is more compatible with the levels of which are likely to occur during Na channel activation (cf. Fig. 6B). Thus the results of Lipp and Niggli [22]may be partly due to SR calcium release being much more sensitive to the local trigger calcium level than in the experiments reported here. In addition, voltage escape during _I_Na (see above) would make the trigger calcium level more attainable. However, our inability to record an _I_Na-evoked [Ca2+]i transient when SR load was increased (with cAMP) suggests that further factors are involved (see below).

4.3 Rates of calcium influx via the exchanger and _I_Ca

The maximum _I_Ca we have observed is 25 pA/pF, which is equivalent to an outward exchanger current of 12.5 pA/pf. This is 5 times larger than the exchanger current density reported by Kimura et al. [11]in whose experiments exchanger activity was increased by using 30 mM [Na]i at 35°C and a test potential of +10 mV. However, if the exchanger is deregulated by proteolysis, the exchanger current can be increased to ∼30 pA/pF [42]. Therefore the exchanger can provide a calcium influx comparable to that of _I_Ca, albeit under experimental conditions that increase outward exchanger current by deregulation and/or at very high [Na]i. Such high levels of exchanger activity should not normally occur during depolarisation in the physiological range, even when accompanied by the activation of _I_Na (see above). It may be argued that at the peak of the action potential (+40 mV) calcium influx via _I_Ca will be reduced while the Na/Ca exchange-mediated influx will increase. However, the voltage dependence of the exchanger is quite shallow (changing e-fold in ∼70 mV) and the stimulation of reverse-mode Na/Ca exchange by voltage should be more than offset by the concomitant decrease _I_Na (which will change e-fold in ∼27 mV) leading to a reduction in local Na accumulation. This view is supported by some recent estimates of exchanger current density during the action potential; at the time of peak _I_Ca, the exchanger current density was less than 0.55 pA/pF whereas _I_Ca was 3.9 pA/pF [44].

It should also be noted that any residual unblocked _I_Ca can give rise to a calcium transient (e.g., Fig. 5) and that it may not always be possible to detect such a small current against larger background currents (in addition, the residual _I_Ca is not readily seen in the low-gain current record of Fig. 5B). This problem will always be present at some level, since no channel blocker can produce complete block except at infinite concentration (from mass action). Therefore, it would be necessary to show that the amplitude of the unblocked _I_Ca is not sufficient to explain the observed calcium transient before one could be sure of the importance (or existence) of an _I_Ca-independent release pathway.

4.4 The role of SR Ca2+ content

Levesque et al. [24]showed that the ability of _I_Na to induce SR Ca2+ release appeared to depend on the amount of Ca2+ in the SR. The relationship between SR Ca2+ release and L-type Ca2+ current amplitude has been noted to be variable among studies [43], and such differences could be partly explained by alterations in SR calcium content affecting the sensitivity of CICR [35]. Additional support for this idea comes from the observation that when the SR load is increased by a conditioning train, the Ca2+ release evoked by a submaximal _I_Ca trigger (2 ms depolarisation to 0 mV) increases from 25 to 54% of the maximum [34]. It is therefore possible that the increase in SR Ca2+ content during conditioning trains increases the sensitivity of CICR to the point where even a small Ca2+ influx via the Na/Ca exchanger can produce calcium release. Unfortunately, there is no simple way to set the exact level of SR Ca2+ loading in different experiments. Nevertheless, this possibility provides an additional explanation for the different results obtained in this study compared to those described earlier [17, 22, 24].

From all of the above considerations, we suggest that the exchanger-mediated calcium influx is unlikely to be as large as that due to _I_Ca under normal physiological conditions [44]. Our failure to demonstrate _I_Na-induced calcium release should not be taken to imply that the exchanger cannot evoke SR calcium release, but rather that the relative magnitude of _I_Na-induced calcium release depends heavily on many other factors such as the SR calcium content, temperature and internal [Na] levels as well as the accumulation of Na during _I_Na and the extent of exchanger proteolysis produced by the experiment. Since these factors are not necessarily well controlled during experiments, one may expect wide variations in the reported ability of the exchanger to trigger SR calcium release. In contrast, _I_Ca-induced SR calcium release is always observed, even under depotentiated conditions (e.g., [34, 44]). Nevertheless, the exchanger could become more important in triggering SR calcium release under conditions that lead to a large increase in [Ca2+]i as there will be: (1) a decrease in the magnitude of _I_Ca (due to calcium-dependent inactivation); (2) an increase in the SR calcium content which may decrease the amplitude of the calcium influx needed to trigger release; and (3) an increase in internal [Na] due to the exchanger increasing calcium extrusion at rest which will promote calcium entry during the action potential [1, 16, 33, 34]. Under such conditions, it is also possible that calcium-dependent proteases may lead to exchanger deregulation, which will increase the rate of calcium influx via the exchanger. Such a shift in the dependence of E–C coupling from the calcium current to the exchanger could provide a mechanism to help maintain myocardial contraction under pathological conditions. In connection with this point, it is notable that the first observation of exchanger-mediated calcium release was obtained in a calcium-overloaded preparation [16].

Acknowledgements

This study was supported by the British Heart Foundation.

References

Calcium-induced release of calcium from the sarcoplasmic reticulum

Am J Physiol

1983

245

C14

Time and calcium dependence of activation and inactivation of calcium-induced calcium release of calcium from the sarcoplasmic reticulum of a skinned cardiac Purkinje cell

J Gen Physiol

1985

247

290

Calcium-induced calcium release activates contraction in intact cardiac cells

Pflügers Arch

1989

413

676

678

Simulated calcium current can both cause calcium loading in and trigger the calcium release from the sarcoplasmic reticulum of a skinned cardiac cell

J Gen Physiol

1985

291

320

One hump or two? The triggering of calcium release from the sarcoplasmic reticulum and the voltage dependence of contraction in mammalian cardiac muscle

Cardiovasc Res

1993

1743

1757

Kinetics, stoichiometry and role of the Na–Ca exchange in isolated cardiac myocytes

Nature

1990

345

618

621

Contribution of sodium–calcium exchange to calcium regulation in cardiac muscle

Ann NY Acad Sci

1991

639

428

443

The relationship between charge movements associated with _I_Ca and _I_Na–Ca in cardiac myocytes

Science

1990

248

376

378

The stoichiometry of the cardiac sodium–calcium exchange system

J Biol Chem

1984

259

7733

7739

Na–Ca exchange: stoichiometry and electrogenicity

Am J Physiol

1985

248

C189

C202

Na–Ca exchange current in mammalian heart cells

Nature

1986

319

596

599

Sodium–calcium exchange in guinea-pig cardiac cells: exchange current and changes in intracellular calcium

J Physiol

1989

414

499

520

Rat vs. rabbit ventricle: Ca flux and intracellular Na assessed by ion-selective microelectrodes

Am J Physiol

1989

256

C813

C822

Sodium–calcium exchange during the action potential in guinea-pig ventricular cells

J Physiol

1989

411

639

661

Inward current related to contraction in guinea-pig ventricular myocytes

J Physiol

1987

385

565

589

Regulation of twitch tension in sheep cardiac Purkinje fibres during calcium overload

Am J Physiol

1987

253

H1540

H1547

Sodium current-induced release of calcium from cardiac sarcoplasmic reticulum

Science

1990

248

372

376

A role for depolarisation induced calcium entry on the Na–Ca exchange in triggering intracellular calcium release and contraction in rat ventricular myocytes

Cardiovasc Res

1993

1677

1690

Gating of the cardiac Ca2+ release channel: the role of the Na+ current and the Na+–Ca2+ exchange

Science

1992

255

850

853

Tension–voltage relations of single myocytes triggered by Na/Ca exchange at 35 degrees C but not 23 degrees C

Am J Physiol

1994

267

C623

C632

Sodium–calcium exchange-mediated contractions in feline ventricular myocytes

Am J Physiol

1992

263

H1161

H1169

Sodium current-induced calcium signals in isolated guinea-pig ventricular myocytes

J Physiol

1994

474

439

446

Relation between reverse sodium–calcium exchange and sarcoplasmic reticulum calcium release in guinea pig ventricular cells

Circ Res

1994

550

554

Release of calcium from guinea-pig cardiac sarcoplasmic reticulum induced by sodium–calcium exchange

Cardiovasc Res

1994

370

378

Na+ current and Ca2+ release from the sarcoplasmic reticulum during action potentials in guinea-pig ventricular myocytes

J Physiol

1995

489

Role of sodium–calcium exchange in activation of contraction in rat ventricle

J Physiol

1993

472

391

413

The roles of the calcium current and Na/Ca exchange in triggering calcium release in cardiac myocytes

Ann NY Acad Sci

1996

779

443

450

Species dependence of contraction velocity in single isolated cardiac myocytes

Cardioscience

1990

Improved patch clamp techniques for high resolution current recording from cells and cell-free membrane patches

Pflügers Arch

1981

391

100

Ca2+ transients in cardiac myocytes measured with high and low affinity indicators

Biophys J

1993

1632

1647

Sodium–calcium exchange in the heart: Membrane currents and changes in [Ca2+]i

Science

1987

238

1720

1722

Inactivation of the cardiac Na+ channels in guinea-pig ventricular cells through the open state

J Physiol

1995

485

581

594

Calcium sparks: Elementary events underlying excitation–contraction coupling in heart muscle

Science

1993

262

740

744

Ca2+ load of guinea-pig ventricular myocytes determines efficacy of brief Ca2+ currents as trigger for Ca2+ release

J Physiol

1994

480

411

421

Fractional SR Ca release is regulated by trigger Ca and SR Ca content in cardiac myocytes

Am J Physiol

1995

268

C1313

C1319

Turnover rate of the cardiac Na–Ca exchanger in guinea-pig ventricular myocytes

J Physiol

1993

472

Interconversion between distinct gating pathways of the high threshold calcium channel in rat ventricular myocytes

J Physiol

1993

462

197

228

The control of excitation–contraction coupling in heart

Science

1995

268

1045

1049

Spatial non-uniformities in [Ca2+]i during excitation–contraction coupling in cardiac myocytes

Biophys J

1994

1942

1956

Sodium–calcium exchange in excitable cells: Fuzzy space

Science

1990

248

283

Cellular origins of the transient inward current in cardiac myocytes

Circ Res

1989

115

126

Charge movement during Na+ translocation by native and cloned Na+/Ca2+ exchanger

Nature

1991

352

715

718

Callewaert G. Excitation–contraction coupling in mammalian heart cells. Cardiovasc Res 1992;26:923–932,415–417.

Ca2+ influx during the cardiac action potential in guinea pig ventricular myocytes

Circ Res

1996

194

200

Author notes

Present address: University Department of Pharmacology, Mansfield Road, Oxford, OX1 3QT, UK.