This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sipido, K. R. Articles by Van de Werf, F. Search for Related Content PubMed PubMed Citation Articles by Sipido, K. R. Articles by Van de Werf, F.

(Circulation Research. 1997;81:1034-1044.)
© 1997 American Heart Association, Inc.

Articles

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

Karin R. Sipido, Micheline Maes, Frans Van de Werf

From the Laboratory of Experimental Cardiology (K.R.S., F. Van de W.), University of Leuven (Belgium), and the Laboratory of Electrobiology (M.M.), University of Antwerp (Belgium).

Correspondence to Karin R. Sipido, MD, PhD, Laboratory of Experimental Cardiology, K.U.L., Campus Gasthuisberg O/N 7th Floor, Herestraat 49, B-3000 Leuven, Belgium. E-mail Karin.Sipido{at}med.kuleuven.ac.be.

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Abstract It has been proposed that Ca²⁺ entry through the Na⁺-Ca²⁺ exchanger can contribute significantly to the trigger for Ca²⁺ release from the sarcoplasmic reticulum (SR). We have compared the characteristics of Ca²⁺ release triggered by reverse-mode Na⁺-Ca²⁺ exchange and by L-type Ca²⁺ current (I_CaL) during depolarizing steps in single guinea pig ventricular myocytes (whole-cell voltage clamp, fluo 3 and fura-red as [Ca²⁺]_i indicators, 36±1°C, K⁺-based pipette solution with 20 mmol/L [Na⁺]). Conditioning pulses to +60 mV ensured comparable Ca²⁺ loading of the SR. In the presence of I_CaL, [Ca²⁺]_i transients typically have an early and rapid rising phase reflecting Ca²⁺ release, which has a bell-shaped voltage dependence with a peak at +10 mV. With Ca²⁺ entry through Na⁺-Ca²⁺ exchange only (20 µmol/L nisoldipine), Ca²⁺ release flux from the SR is decreased and directly related to the amplitude of the depolarizing step. Ca²⁺ release is preceded by a significant delay (81±21 ms at +20 mV, 24±4 ms at +70 mV) related to Ca²⁺ entry through the exchanger. Triggered release interrupts Ca²⁺ entry, as evidenced by reversal of the exchanger current. At potentials positive to +40 mV, Ca²⁺ influx through Na⁺-Ca²⁺ exchange, calculated from the outward exchange current, reaches magnitudes comparable to I_CaL, but Ca²⁺ release due to reverse-mode Na⁺-Ca²⁺ exchange still has a significant delay. We calculated trigger efficiency as the ratio between the maximal rate of Ca²⁺ release and the Ca²⁺ influx preceding this release; efficiency of reverse-mode Na⁺-Ca²⁺ exchange is approximately four times less than that of I_CaL. With both I_CaL and reverse-mode Na⁺-Ca²⁺ exchange present, Ca²⁺ release is triggered by I_CaL, and a contribution of reverse-mode Na⁺-Ca²⁺ exchange to the trigger could not be detected at potentials below +60 mV. These characteristics of reverse-mode Na⁺-Ca²⁺ exchange predict that its role as a trigger for Ca²⁺ release during the action potential is likely to be negligible.

Key Words: Na⁺-Ca²⁺ exchange • sarcoplasmic reticulum • Ca²⁺ channel • cardiomyocyte

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
The Ca²⁺ release channel of the cardiac SR opens in response to an increase in [Ca²⁺] at the cytoplasmic side of the channel.¹ In cells from which the sarcolemma had been removed, it was shown that the rate of Ca²⁺ release was dependent on the rate of change in [Ca²⁺] near the SR, being larger for rapid increases in [Ca²⁺] near the SR,² and that Ca²⁺ entry through the L-type Ca²⁺ channel would have all the characteristics necessary for optimal triggering of the Ca²⁺ release channel.³ Studies that found a close relationship between the voltage dependence of the L-type Ca²⁺ current and the voltage dependence of contraction and of Ca²⁺ release in intact isolated single cardiac cells have supported this hypothesis.^4–8 The colocalization of L-type Ca²⁺ channels and Ca²⁺ release channels in the junctional complex of T tubules would facilitate direct interaction.^9,10 Theoretical modeling has indicated that control of Ca²⁺ release by local interactions between individual L-type Ca²⁺ channels and Ca²⁺ release channels of the SR could indeed reproduce the characteristics of Ca²⁺ release in the intact cell.^11,12 Recently, [Ca²⁺]_i transients resulting from the activation of small release units have been observed during confocal microscopy as Ca²⁺ sparks or local Ca²⁺ transients. These results indicate that Ca²⁺ entry through one L-type Ca²⁺ channel controls activation of a small cluster of Ca²⁺ release channels.^13,14

Although the important role of the L-type Ca²⁺ current is well established, it has been proposed that it may not be the only trigger for Ca²⁺ release (reviewed in Reference 15¹⁵ ). Studies of force development in multicellular preparations and of cell shortening in single cells have shown significant force and/or shortening at positive potentials, where Ca²⁺ entry through the L-type Ca²⁺ channel was unlikely. Evidence was presented supporting the hypothesis that at these positive potentials Ca²⁺ entry through the Na⁺-Ca²⁺ exchanger (reverse-mode Na⁺-Ca²⁺ exchange) provided the trigger for Ca²⁺ release from the SR.^16–18 It was also proposed that Na⁺ accumulation associated with the Na⁺ current induced reverse-mode Na⁺-Ca²⁺ exchange and sufficient Ca²⁺ entry to trigger Ca²⁺ release,^19,20 but conflicting results have been reported.^21–24 A different line of evidence for a role of reverse-mode Na⁺-Ca²⁺ exchange as a trigger for release comes from experiments showing that a considerable amount of cell shortening or a significant [Ca²⁺]_i transient still remains in the presence of Ca²⁺ channel blockers applied shortly before the test pulse by means of a rapid solution switcher.^25–28 In these studies, it was proposed that Na⁺-Ca²⁺ exchange would contribute significantly to the triggering of Ca²⁺ release in the voltage range around 0 mV (where typically Ca²⁺ current is maximally activated) and during the action potential.

The extent of the contribution of reverse-mode Na⁺-Ca²⁺ exchange to the triggering of Ca²⁺ release during the action potential remains controversial. In contrast to numerous studies on the Ca²⁺ current–related release, many characteristics of Na⁺-Ca²⁺ exchange as a trigger for release are unknown. Although in the data presented by Bridge, Levi, and colleagues^27,28 the time course of Ca²⁺ release triggered by the exchanger is the same as the one triggered by the Ca²⁺ current, differences in the time course of shortening at positive potentials suggest that there are important kinetic differences between the two trigger mechanisms.^16,17 Thermodynamics of the exchanger predict a decrease and/or reversal of the exchange current during Ca²⁺ release. However, the outward Na⁺-Ca²⁺ exchange current and its predicted decrease or reversal during Ca²⁺ release have not yet been not documented, except in one study involving Ca²⁺ influx during action potential recording.²⁹ In contrast to I_CaL-related Ca²⁺ release, the efficiency of reverse-mode Na⁺-Ca²⁺ exchange, ie, the relation between the amplitude of Ca²⁺ entry through the exchanger and the amplitude of Ca²⁺ release, has not been studied. Therefore, in the present study we have investigated the characteristics of Ca²⁺ entry through the exchanger as a trigger for Ca²⁺ release and compared those characteristics with those of Ca²⁺ entry through the L-type Ca²⁺ channel. Following previously published reports that have emphasized the importance of experimental conditions, our experiments were done at 36°C and with K⁺-containing pipette solutions. To increase Ca²⁺ entry through the exchanger, we included 20 mmol/L Na⁺ in the pipette solution. Our findings indicate that Na⁺-Ca²⁺ exchange can trigger Ca²⁺ release from the SR but that this trigger is less efficient than L-type Ca²⁺ current and that its contribution during the action potential is likely to be negligible.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Cell Isolation
Single guinea pig ventricular myocytes were isolated enzymatically as previously described.²³ Cell isolation routinely yielded 60% to 70% of viable rod-shaped cells.

Voltage-Clamp and [Ca²⁺]_i Measurements
We used the whole-cell ruptured patch-clamp technique.³⁰ Membrane currents were recorded with an Axopatch 1D amplifier, filtered at 1 kHz, and sampled and digitized at 4 kHz (Fastlb45, Indec Systems).

[Ca²⁺]_i was monitored with fluo 3 (60 µmol/L) or a combination of fluo 3 (30 µmol/L) and fura-red (70 µmol/L). We chose to use fluo 3 because of the brightness of the signal and fast response time, which made it easy to study the rising phase of the [Ca²⁺]_i transient. In combination with fura-red, motion artifacts could be minimized, and [Ca²⁺]_i could be estimated after calibration of the ratio signal. Excitation wavelength was 485±8 nm. The dichroic mirror under the objective was centered at 510 nm. The emission light was split by a second dichroic mirror centered at 585 nm. Fluo 3 fluorescence was sampled at 535±15 nm; the fura-red–dependent emission was recorded by a second red-sensitive photomultiplier with a bandpass filter of 615±25 nm in front of it. The microscope was also equipped with a transmitted light source at 700 nm; a CCD camera made it possible to follow cell shortening visually on a TV monitor.

With fluo 3 alone, fluorescence values are normalized for baseline fluorescence. For the combination of fluo 3 and fura-red, we used calibration parameters that were obtained partly during in vitro calibration (ßxK_d) and partly during in vivo calibration (maximum and minimum fluorescence ratios).^31,32 Such calibration can yield a best estimate of [Ca²⁺]_i but is less reliable than calibration of the ratiometric dyes, such as fura 2 or indo 1, because of possible inhomogeneous equilibration of the two dyes. Despite these limitations, calibrated [Ca²⁺]_i values between cells were well reproducible for a given experimental protocol.

Ca²⁺ release from the SR is indicated by a rapid increase of [Ca²⁺]_i, and it has been shown previously that the Ca²⁺ release flux is proportional to the rate of rise or derivative of [Ca²⁺]_i.^12,33 Quantitative calculations of the release flux were not possible in the presence of K⁺ currents and Na⁺-Ca²⁺ exchange currents. We have therefore used an indirect approach: we have measured and compared the amplitude of the derivative of [Ca²⁺]_i in different conditions as an indirect estimate of the amplitude of the SR Ca²⁺ release flux.³⁴ To analyze the rate of rise of [Ca²⁺]_i, we have used the derivative of the fluo 3 signal. With the derivative of the calibrated [Ca²⁺]_i signal, results were qualitatively comparable, but the signal-to-noise ratio was less favorable.

Solutions and Experimental Protocols
The pipette solution contained (mmol/L) potassium aspartate 120, KCl 20, potassium HEPES 10, MgATP 5, MgCl₂ 0.5, NaCl 20, and fluo 3 0.06 (or fluo 3 0.03 and fura-red 0.07), pH 7.20. Alternatively, potassium aspartate and KCl were replaced with CsCl (130 mmol/L). For some experiments, a pipette solution without NaCl was used. The presence of high [Na⁺]_i with 20 mmol/L [Na⁺]_pip compared with 0 mmol/L [Na⁺]_pip was confirmed by the presence of 4-fold larger dihydro-ouabain–sensitive Na⁺-K⁺ pump currents (K.R. Sipido and F. Verdonck, unpublished data, 1997). The external solution contained (mmol/L) NaCl 130, KCl 5.4, sodium HEPES 11.8, MgCl₂ 0.5, CaCl₂ 1.8, and glucose 6, pH 7.35. For some experiments, KCl was omitted and replaced with 10 mmol/L CsCl. All experiments were done at 36±1°C. External solution exchange was done by a rapid perfusion system positioned close to the cell. From the block of the inward rectifier K⁺ current at a holding potential of -50 mV by application of CsCl (10 mmol/L) or BaCl₂ (0.5 mmol/L), we measured a solution exchange time of 2 s. To block L-type Ca²⁺ channels, we used 20 µmol/L nisoldipine (Bayer), prepared as a 20 mmol/L stock in dimethyl sulfoxide. Ca²⁺ release from the SR was disabled with 10 µmol/L ryanodine (Sigma Chemical Co) prepared as a 10 mmol/L stock in water. In the presence of nisoldipine, NiCl₂ (2 or 5 mmol/L) was used to block Na⁺-Ca²⁺ exchange.

The characteristics of Ca²⁺ release were studied during 225-ms test depolarizations from -40 or -45 mV to -30 up to +70 mV. These depolarizations were preceded by two conditioning steps (300 ms) to +60 mV at 1-s intervals; the test depolarization followed the last conditioning step with an interval of 2 s. The pulse protocol was repeated every 20 or 30 s, and the holding voltage between pulse trains was -70 mV.

With 20 mmol/L NaCl in the pipette, the SR was not depleted at rest.³⁵ With two conditioning depolarizing steps to +60 mV, large and reproducible loading of the SR could be obtained. We tested and confirmed that with this loading protocol, steady-state block of the L-type Ca²⁺ channel did not significantly affect the extent of loading. This is illustrated in Fig 1. After the two conditioning steps to +60 mV, the Ca²⁺ content of the SR was estimated from the amplitude of Ca²⁺ release induced by a brief application of caffeine.^36,37 The protocol was then repeated in the presence of 20 µmol/L nisoldipine (n=7). Caffeine application induced more than one [Ca²⁺]_i transient in five of seven cells, probably related to the high loading state of the SR, since this was not observed with lower [Na⁺]_pip. Therefore, we compared both the peak fluorescence of the first release signal and the integrated transient inward current during Ca²⁺ release, which mainly results from Ca²⁺ extrusion through Na⁺-Ca²⁺ exchange. Neither parameter was significantly affected by Ca²⁺ channel block, indicating that loading of the SR in the present conditions occurs primarily through Ca²⁺ entry through Na⁺-Ca²⁺ exchange.

View larger version (26K): [in this window] [in a new window]   Figure 1. Ca2+ content of the SR after two conditioning pulses to +60 mV is estimated from Ca2+ release induced by rapid application of caffeine, as shown in the insert at the top. A, Membrane current (I) and fluorescence (F) records of the last conditioning pulse and during caffeine (10 mmol/L) application in control conditions (left) and after Ca2+ channel block (right). a.u. indicates arbitrary units. B, Pooled data of seven cells (mean±SEM). On the left is the integrated transient inward current (C) during caffeine application; on the right, peak value of first fluorescence transient (F) during caffeine application. Experiments were performed in potassium aspartate–based pipette solution; [Na+]pip was 20 mmol/L.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Voltage Dependence and Time Course of [Ca²⁺]_i Transients With Both I_CaL and Na⁺-Ca²⁺ Exchange
In Fig 2, membrane currents (top traces) and [Ca²⁺]_i transients (bottom traces) during test depolarizations between -30 and +70 mV are illustrated by a representative example. The early inward current during the depolarizing step is I_CaL (nisoldipine sensitive; see below). On repolarization, a time-dependent inward current is seen, reflecting Ca²⁺ efflux through the Na⁺-Ca²⁺ exchanger ([Ca²⁺]_i dependent and blocked by NiCl₂). At -30 mV, there is only a very small increase in [Ca²⁺]_i, but as the test potential becomes more positive, the amplitude and rate of rise of [Ca²⁺]_i increase. At +10 mV, the rate of rise of [Ca²⁺]_i is maximal, and peak [Ca²⁺]_i is reached 20 ms after the onset of the depolarizing step. At potentials negative to +30, [Ca²⁺]_i declines after this early peak and before the membrane is repolarized. As the depolarizing pulse becomes more positive, a second component of slow increase of [Ca²⁺]_i can be seen after the fast initial rise. Peak [Ca²⁺]_i coincides with the end of the depolarization; decline of [Ca²⁺]_i starts on repolarization.

View larger version (19K): [in this window] [in a new window]   Figure 2. Membrane currents (I) (top traces) and [Ca2+]i transients (bottom traces) during 225-ms depolarizing steps to the indicated potentials from a holding potential of -45 mV. Test depolarization was preceded by two conditioning steps to +60 mV, as illustrated in the insert at the top. Experiments were performed in potassium aspartate–based pipette solution; [Na+]pip was 20 mmol/L.

Because of this second slow component, a plot of the maximal values of [Ca²⁺]_i reached during the depolarizing pulse is directly related to the test membrane potential, as shown in Fig 3B (solid circles, mean±SEM of nine cells). The presence of a second slow component at positive potentials is also evident from the plot of the amplitude of the time-dependent inward current on repolarization, reflecting Ca²⁺ efflux via the Na⁺-Ca²⁺ exchanger (Fig 3B, solid squares). This plot illustrates the pronounced increase in this current at positive potentials. Although the plot of peak [Ca²⁺]_i is a monotonic function of membrane voltage, there are clear differences in the time course of the [Ca²⁺]_i transient at different potentials. These are illustrated by the superimposed traces obtained at +10 and at +70 mV (Fig 3A). The rate of rise of the [Ca²⁺]_i transient at +70 mV appears to be delayed compared with the [Ca²⁺]_i transient at +10 mV. Therefore, if we plot [Ca²⁺]_i at 20 ms after the depolarizing step as a function of the test membrane potential, a different voltage dependence is seen (Fig 3B, open circles). The bell-shaped voltage dependence of the early increase in [Ca²⁺]_i suggests a close relation to I_CaL; however, the voltage dependence of the peak value of [Ca²⁺]_i suggests that a second process with more linear voltage dependence also contributes to the [Ca²⁺]_i transient. This process was absent in cells studied with a pipette solution without NaCl, as illustrated in Fig 3C and 3D. Taken together, these data indicate the presence of a contribution of reverse-mode Na⁺-Ca²⁺ exchange to the [Ca²⁺]_i transients of cells studied with 20 mmol/L [Na⁺]_pip. Therefore, we further examined the [Ca²⁺]_i transient after block of I_CaL in cells with 20 mmol/L [Na⁺]_pip.

View larger version (24K): [in this window] [in a new window]   Figure 3. A, Membrane currents (I) and [Ca2+]i transients at +10 mV and at +70 mV on expanded time scale (same records as in Fig 1). Note the delay before the rapid upstroke of the [Ca2+]i transient at +70 mV and the presence of a second slow component, resulting in a high [Ca2+]i at the end of the depolarizing step and a large inward current on repolarization. B, Pooled data of eight cells studied with 20 mmol/L [Na+]pip (mean±SEM) for maximal [Ca2+]i during the depolarizing step (), [Ca2+]i at 20 ms after the depolarizing step (), and the inward current 10 ms after repolarization (). C, Membrane currents and [Ca2+]i transients at +10 mV and at +70 mV in a cell studied with 0 mmol/L [Na+]pip. At +70 mV, the [Ca2+]i transient is very small. D, Pooled data from six cells studied with 0 mmol/L [Na+]pip (mean±SEM) for maximal [Ca2+]i during the depolarizing step (), [Ca2+]i at 20 ms after the depolarizing step (), and the inward current 10 ms after repolarization (Iinw on rep, ). All parameters have a similar bell-shaped voltage dependence.

Voltage Dependence and Time Course of [Ca²⁺]_i Transients After Block of I_CaL
Nisoldipine (20 µmol/L) was applied and Ca²⁺ release was examined after full solution exchange, during steady-state perfusion. With conditioning pulses to +60 mV and with 20 mmol/L [Na⁺]_pip, the block of I_CaL had no significant effect on the loading of the SR (see "Materials and Methods" and Fig 1). Fig 4 shows the [Ca²⁺]_i transients before and after block of I_CaL. The rapidly rising [Ca²⁺]_i transients at potentials below +30 mV are blocked; however, the [Ca²⁺]_i transient at the most positive potential (+70 mV) is only slightly changed. This is illustrated in more detail in Fig 5, where membrane currents and [Ca²⁺]_i transients at +10, +40 and +70 mV are shown on an expanded time scale. The membrane current traces clearly show the block of I_CaL by nisoldipine. The rapid rise of the [Ca²⁺]_i transient at +10 mV is also blocked, and instead, a delayed, more slowly rising [Ca²⁺]_i transient is seen. At +70 mV the [Ca²⁺]_i transient has changed very little, whereas at +40 mV an intermediate picture is seen. The changes in the [Ca²⁺]_i transients at +10 and at +40 mV can be described by two parameters: (1) a decrease in the rate of rise of [Ca²⁺]_i, corresponding to a decrease in the Ca²⁺ release flux,^12,33,34 and (2) a delay between the onset of depolarization and the onset of a more rapid rise in [Ca²⁺]_i. These parameters were quantified in a total of six cells by measuring the maximal value of the derivative of the fluo 3 signal and of the time to this maximal value in control conditions and after the block of I_CaL by nisoldipine, illustrated in Fig 6. The plot of the maximal rate of rise in control conditions (Fig 6A, solid squares) is bell-shaped, with a maximum at +10 mV. With nisoldipine (open squares), the maximal rate of rise is decreased significantly at potentials between -30 and +40 mV. At more positive potentials, little change or even a slight increase is seen. However, even at these potentials the rate of rise is still less than the maximal rate of rise observed with I_CaL at +10 mV. The delay to the onset of the rapid increase in [Ca²⁺]_i is shown in Fig 6B as the time to the maximal dF/dt. In control conditions, the maximal rate of rise is reached within 10 to 15 ms after the onset of the depolarizing pulse, with the minimal delay seen around +10 mV. After block of I_CaL, this delay is increased significantly and exceeds 30 ms even at the most positive voltages; the time to peak dF/dt is now a monotonic decreasing function of the amplitude of the depolarizing step.

View larger version (23K): [in this window] [in a new window]   Figure 4. [Ca2+]i transients during depolarizations to the indicated potentials in control conditions (A) and after block of L-type Ca2+ current by 20 µmol/L nisoldipine (B). Cell was different from that in Figs 2 and 3. Experiments were performed in potassium aspartate–based pipette solution; [Na+]pip was 20 mmol/L.

View larger version (17K): [in this window] [in a new window]   Figure 5. Membrane currents (I) (top traces) and [Ca2+]i transients (bottom traces) during depolarization to +10, +40, and +70 mV before and after block of ICaL by 20 µmol/L nisoldipine (+nis). This is the same cell as in Fig 4.

View larger version (17K): [in this window] [in a new window]   Figure 6. A, The rate of rise of [Ca2+]i was quantified as the maximal value of the first derivative of the fluo 3 signal; for each cell, these values were normalized to the maximal value of that cell. Values are mean±SEM (pooled data from six cells, potassium aspartate–based pipette solution, [Na+]pip=20 mmol/L). V indicates voltage. B, The delay to the rapid rise of [Ca2+]i was quantified as the time to the maximal value of the first derivative.

The monotonic and nearly linear relation between membrane voltage and the amplitude and rate of rise of the [Ca²⁺]_i transients after block of I_CaL strongly suggest that these [Ca²⁺]_i transients are related to reverse-mode Na⁺-Ca²⁺ exchange. The presence of a phase of rapid increase in [Ca²⁺]_i also suggests that these [Ca²⁺]_i transients do not merely result from Ca²⁺ entry through the exchanger but that Ca²⁺ is released from the SR. We hypothesized that the presence of a delay to the more rapid rise in Ca,²⁺ a delay that is inversely related to the membrane potential, reflected the time between Ca²⁺ entry and Ca²⁺ release. To investigate this hypothesis, we compared the [Ca²⁺]_i transients in the presence of nisoldipine with the [Ca²⁺]_i transients after block of Ca²⁺ release by ryanodine.

Ca²⁺ Entry via Na⁺-Ca²⁺ Exchange Precedes Triggered Ca²⁺ Release
Fig 7 shows a representative example of superimposed membrane currents and [Ca²⁺]_i transients before and after block of Ca²⁺ release with 10 µmol/L ryanodine (all in the presence of nisoldipine). Similar results were obtained in five other cells. In the presence of ryanodine, the rapidly rising part of the [Ca²⁺]_i transient is blocked, and [Ca²⁺]_i transients are slow, increasing throughout the depolarizing pulse, as reported previously.³⁸ The [Ca²⁺]_i transient before application of ryanodine can be superimposed on the one in the presence of ryanodine up to the time when a sudden increase in the rate of rise occurs. These findings support the idea that the rapidly rising phase represents Ca²⁺ release, occurring with a delay during which Ca²⁺ entry occurs. This is also reflected in the time course of the membrane currents. Ryanodine induced a small linear leak current, but the most important change during the depolarizing steps is the disappearance of the inward shift of the membrane current coinciding with Ca²⁺ release. Comparing [Ca²⁺]_i transients and the time course of membrane currents in the presence and absence of ryanodine (as shown in Fig 7), we have measured the delay for Ca²⁺ release triggered by Na⁺-Ca²⁺ exchange as the time from onset of depolarization to the time of rapid increase in [Ca²⁺]_i, with its associated inward shift of the membrane current. At +10 mV, this delay was 93±20 ms; at +40 mV, 55±13 ms; and at +70 mV, 24±4 ms (mean±SEM, n=6, same group of cells as in Fig 6). These data reflect the delay to the onset of Ca²⁺ release and confirm that this delay is related to Ca²⁺ entry.

View larger version (22K): [in this window] [in a new window]   Figure 7. Membrane currents (I) (top) and [Ca2+]i transients (bottom) in the presence of 20 µmol/L nisoldipine only (nis) and in the presence of both 20 µmol/L nisoldipine and 10 µmol/L ryanodine (nis+rya) at the indicated potentials. For each pair of records, the first dotted line marks the onset of the depolarization, and the second one marks the onset of Ca2+ release from the SR, ie, the rapid increase in [Ca2+]i that is accompanied by an inward shift of the membrane current, both of which are blocked by ryanodine. The delay to Ca2+ release in this example was 50 ms at +10 mV, 30 ms at +40 mV, and 24 ms at +70 mV. This is the same cell as in Figs 4 and 5.

Our results so far indicate that after block of I_CaL, Na⁺-Ca²⁺ exchange can trigger Ca²⁺ release from the SR, but this Ca²⁺ release differs in two important aspects from Ca²⁺ release triggered by I_CaL. First, the same maximal rate of rise of [Ca²⁺]_i and therefore the same Ca²⁺ release flux are not attained in the range of membrane potentials studied. Second, Ca²⁺ release is delayed significantly. One possible explanation for these differences is that the rate of Ca²⁺ entry through the exchanger is much less than the rate of Ca²⁺ entry through the L-type Ca²⁺ channel. We therefore have quantified this rate of entry by measuring I_CaL and the Na⁺-Ca²⁺ exchange current.

Amplitude and Time Course of the Na⁺-Ca²⁺ Exchange Current During Ca²⁺ Entry and Ca²⁺ Release
Because of previous reports that substitution of internal K⁺ with Cs⁺ affects Ca²⁺ release from the SR,^18,39,40 the experiments shown so far were done with a K⁺-containing pipette solution. However, the membrane currents are then always mixed with time-dependent K⁺ currents, complicating the interpretation. Therefore, to measure the Na⁺-Ca²⁺ exchange current, we have repeated our experiments with a Cs⁺-containing pipette solution and with external KCl replaced with CsCl (n=12). An example of such an experiment is shown in Fig 8. We recorded membrane currents and [Ca²⁺]_i transients first in control conditions, then in the presence of nisoldipine to block I_CaL, and last in a solution with 2 mmol/L NiCl₂ added to the nisoldipine-containing solution to block Na⁺-Ca²⁺ exchange. The Na⁺-Ca²⁺ exchange current is the difference between the membrane current in the presence of nisoldipine and the current in the presence of NiCl₂⁴¹ and is shown in the middle panels. It is clear that the initial Na⁺-Ca²⁺ exchange current is outward at the onset of the depolarizing pulse and decreases during the pulse as [Ca²⁺]_i increases to become a net inward current during Ca²⁺ release. Furthermore, the Na⁺-Ca²⁺ exchange outward current increases in amplitude with increasing membrane potential. The amplitude of the outward Na⁺-Ca²⁺ exchange current was measured 10 ms into the depolarizing pulse; I_CaL was measured as the nisoldipine-sensitive peak inward current at 5 to 10 ms after depolarization. The average values of membrane current density for five cells are shown in Fig 10A. These values are in the same order of magnitude as reported by others in whole-cell experiments.^38,41,42

View larger version (26K): [in this window] [in a new window]   Figure 8. Membrane currents (I) (top) and [Ca2+]i transients (normalized fluorescence records of fluo 3 signal [F]) (bottom) at the indicated potentials with internal potassium aspartate and KCl replaced with 130 mmol/L CsCl ([Na+]pip=20 mmol/L) and with 10 mmol/L CsCl added to the external solution. a.u. indicates arbitrary units; con, records in control conditions; nis, records in the presence of 20 µmol/L nisoldipine; and Ni, membrane currents recorded with both 20 µmol/L nisoldipine and 2 mmol/L NiCl2 (dashed line). Note the absence of a transient inward current on repolarization; [Ca2+]i transients in the presence of NiCl2 are flat. The difference currents between nis and Ni are the Na+-Ca2+ exchange currents (middle). The increase in peak [Ca2+]i after block of ICaL is probably related to higher loading of the SR during Ca2+ entry via the exchanger preceding release.54

View larger version (14K): [in this window] [in a new window]   Figure 10. Amplitude (I) and voltage (V) dependence of ICaL measured as the peak of the nisoldipine-sensitive current () and of Na+-Ca2+ exchange current measured as the difference between the membrane current recorded in the presence of 20 µmol/L nisoldipine and the current recorded after the addition of 2 to 5 mmol/L NiCl2 in the presence of 20 µmol/L nisoldipine (). Results are mean±SEM (pooled data from 10 cells). A, Data obtained with a Cs+-based pipette solution (130 mmol/L CsCl) (n=5 cells). B, Data obtained with K+-based pipette solution (120 mmol/L potassium aspartate and 20 mmol/L KCl) (n=5 cells). The data in panel B are shifted to the right by 10 mV because of an offset potential with the low [Cl-] of the K+-based pipette solution. [Na+]pip was 20 mmol/L.

Experiments with Cs⁺-containing pipette solutions were complicated by the fact that in the majority of cells (8 of 12) after block of I_CaL the remaining [Ca²⁺]_i transient no longer had a rapidly rising phase indicative of triggered Ca²⁺ release (versus in only 1 of 15 cells tested with K⁺-containing pipette solution). This is in line with previous reports.^18,39,40 One possible explanation is that the Cs⁺ pipette solution somehow decreased Na⁺-Ca²⁺ exchange. We therefore repeated the protocol shown in Fig 8 with K⁺-containing solutions. To block K⁺ currents as much as possible, we added 1 µmol/L almokalant (to block the rapidly activating component of the delayed rectifier K⁺ current) and 0.25 or 0.5 mmol/L BaCl₂. Records of such an experiment are shown in Fig 9. As with CsCl in the pipette, the Na⁺-Ca²⁺ exchange current is outward at the onset of depolarization and shows a large inward shift during Ca²⁺ release. In a manner similar to that for the experiments performed with Cs⁺ in the pipette, we have pooled the results of five cells and have plotted the membrane current density of the Ni²⁺-sensitive initial outward current and of the nisoldipine-sensitive peak inward current (Fig 10B). The maximal I_CaL appears to be shifted by +10 mV compared with the data obtained with CsCl, which is compatible with a 10-mV offset potential with low Cl^- pipette solutions. With this taken into account, the values for Na⁺-Ca²⁺ exchange current or for I_CaL are not significantly different between cells studied with a K⁺-containing or with a Cs⁺-containing pipette solution.

View larger version (27K): [in this window] [in a new window]   Figure 9. Membrane currents (I) (top) and [Ca2+]i transients (normalized fluorescence records of fluo-3 signal [F]) (bottom) at the indicated potentials with a K+-based pipette solution ([Na+]pip=20 mmol/L) and with 1 µmol/L almokalant and 0.25 mmol/L BaCl2 added to the external solution. a.u. indicates arbitrary units; con, records in control conditions; nis, records in the presence of 20 µmol/L nisoldipine; and Ni, membrane currents recorded with both 20 µmol/L nisoldipine and 2 mmol/L NiCl2 (dashed line). The difference currents between nis and Ni are the Na+-Ca2+ exchange currents (middle). The increase in peak [Ca2+]i after block of ICaL is probably related to higher loading of the SR during Ca2+ entry via the exchanger preceding release.54

Relation Between Ca²⁺ Entry and Release: Trigger Efficiency
The data shown in Fig 10 indicate that high rates of Ca²⁺ entry can be achieved through reverse-mode Na⁺-Ca²⁺ exchange. To evaluate the efficiency of Ca²⁺ entry both through the exchanger and I_CaL as a trigger for Ca²⁺ release, we have measured in this same group of cells the maximal rate of rise as an indicator of the Ca²⁺ release flux and related this to the Ca²⁺ entry (with potassium aspartate–based pipette solution). These data are shown in Fig 11. Solid symbols indicate data obtained in control conditions; open symbols, after block of I_CaL. The top left panel shows the rate of Ca²⁺ entry (in µmol · s^-1 · F^-1), obtained by dividing current densities by valence and Faraday's constant. At +70 mV, Ca²⁺ entry in the cell is almost exclusively through the exchanger and nearly as large as Ca²⁺ entry through the Ca²⁺ channel at 0 mV, whereas at 0 mV, Ca²⁺ entry through the exchanger is small. If we then look at the Ca²⁺ release flux at +70 mV after block of I_CaL (solid symbols in top middle panel), it is also nearly as large as at 0 mV with I_CaL (open symbols). However, a major difference is the time to this release, which is 53±8 ms with the exchanger at +70 mV but only 11±3 ms with I_CaL at 0 mV (top right panel). These data imply that for the exchanger a larger amount of Ca²⁺ entry is required before Ca²⁺ release is triggered. We have further quantified trigger efficiency as the relation between the peak rate of Ca²⁺ release and the preceding Ca²⁺ influx, obtained by integrating Ca²⁺ entry through I_CaL or through the exchanger, up to the time of peak dF/dt. This is shown in the bottom panel of Fig 11; solid symbols indicate I_CaL (measured at -10, 0, and +10 mV), and open symbols indicate the Na⁺-Ca²⁺ exchange current (measured at +50, +60, and +70 mV after block of I_CaL). Because the data are pooled for different voltages, we have not curve-fitted the data, but it can be seen that the relation is much steeper for I_CaL than for Na⁺-Ca²⁺ exchange. For five cells, the ratio between peak dF/dt and the preceding Ca²⁺ entry was 54±5 (arbitrary units) for I_CaL at 0 mV versus 15±4 for Na⁺-Ca²⁺ exchange at +70 mV. These data indicate that the trigger efficiency of Ca²⁺ entry through the exchanger is lower than of Ca²⁺ entry through the L-type Ca²⁺ channel.

View larger version (21K): [in this window] [in a new window]   Figure 11. Relation between Ca2+ entry and Ca2+ release in control conditions () and after block of the L-type Ca2+ channel (). Values are mean±SEM (pooled data from five cells). Top left, Ca2+ entry was calculated by dividing the Na+-Ca2+ exchange and ICaL densities by the valence and Faraday's constant; data are the same as in Fig 10A. Top middle, Ca2+ release was measured as the maximal value of the first derivative of the fluo 3 signal; for each cell these values were normalized to the maximal value of that cell. Top right, The delay to Ca2+ release was quantified as the time to the maximal value of the first derivative. Bottom, Relation between Ca2+ release, peak dF/dt, and preceding Ca2+ entry, measured as the integrated Ca2+ entry up to the time of peak dF/dt. indicates ICaL, measured at -10, 0, and +10 mV in six cells; , reverse-mode Na+-Ca2+ exchange, measured at +50, +60, and +70 mV after block of ICaL; and a.u., arbitrary units. Experiments were performed in potassium aspartate–based pipette solution; [Na+]pip was 20 mmol/L.

Contribution of Reverse-Mode Na⁺-Ca²⁺ Exchange as Trigger for Ca²⁺ Release in the Presence of I_CaL
Even if the efficiency of Ca²⁺ entry through the exchanger as trigger for release is low, it could still act as an additional trigger to I_CaL in normal conditions. We have examined this issue by comparing Ca²⁺ release with 20 mmol/L [Na⁺]_pip and with 0 mmol/L [Na⁺]_pip in the presence of I_CaL. If reverse-mode Na⁺-Ca²⁺ exchange contributes to the trigger, then release should not decline, or at least decline less, at positive voltages with 20 mmol/L [Na⁺]_pip. The plot of peak dF/dt with 20 mmol/L [Na⁺]_pip and I_CaL present (solid symbols in Fig 6A) already showed that the voltage dependence of release follows the voltage dependence of I_CaL, except at +70 mV. Comparing Ca²⁺ release observed with 20 mmol/L [Na⁺]_pip with Ca²⁺ release observed with 0 mmol/L [Na⁺]_pip, panels B and D of Fig 3 showed that [Ca²⁺]_i at 20 ms, as a parameter for Ca²⁺ release in this time period, declines with voltage whether or not reverse-mode Na⁺-Ca²⁺ exchange is present. This is further illustrated by the superimposed plots of I_CaL and [Ca²⁺]_i for these experiments, shown in Fig 12A and 12B. Although [Ca²⁺]_i is higher at all voltages with 20 mmol/L [Na⁺]_pip, compatible with a higher loading state of the SR in the presence of increased [Na⁺]_i,⁴³ the voltage dependence still follows the voltage dependence of I_CaL. Only at +70 mV is a deviation observed, with release seen with 20 mmol/L [Na⁺]_pip but not with 0 mmol/L [Na⁺]_pip.

View larger version (23K): [in this window] [in a new window]   Figure 12. Contribution of reverse-mode Na+-Ca2+ exchange to triggering of Ca2+ release in the presence of ICaL. A, Superimposed plots of the increase in [Ca2+]i at 20 ms after the depolarizing step () and of ICaL (; ICaL has been scaled, with an inverse sign) with 0 mmol/L [Na+]pip (pooled data from six cells). The curves are superimposed throughout the full voltage range. B, Same data for six cells studied with 20 mmol/L [Na+]pip. indicates the increase in [Ca2+]i at 20 ms after the depolarizing step; , ICaL (ICaL has been scaled, with an inverse sign). Only at +70 mV do the curves have a different voltage dependence with more Ca2+ release. C, Apparent efficiency of ICaL as trigger for release calculated as the ratio between peak dF/dt as measure for Ca2+ release and the preceding Ca2+ entry, ie, the integrated nisoldipine-sensitive current up to the time of peak dF/dt. Each data point is the mean of six cells. indicates 20 mmol/L [Na+]pip; , 0 mmol/L [Na+]pip. Even with 20 mmol/L [Na+]pip, the apparent trigger efficiency of ICaL decreases with more positive potentials, indicating that reverse mode does not contribute to the trigger.

We also calculated the relation between peak dF/dt and the preceding Ca²⁺ influx through I_CaL as a measure of the apparent efficiency of I_CaL for 0 and 20 mmol/L [Na⁺]_pip at -10 mV, +20 mV, and +50 mV. If reverse-mode Na⁺-Ca²⁺ exchange adds substantially to the trigger, the apparent efficiency of I_CaL should increase with voltage with 20 mmol/L [Na⁺]_pip, as the Ca²⁺ influx through the exchanger increases >5-fold in this range of voltages (Fig 10), or at least decrease less, compared with conditions with 0 mmol/L [Na⁺]_pip. The results are shown in Fig 12C. Whether or not the pipette contains 20 mmol/L Na⁺, the efficiency of I_CaL decreases with voltage in a similar way. This similar voltage dependence of the apparent efficiency of I_CaL in the absence or presence of [Na⁺]_pip indicates that reverse-mode Na⁺-Ca²⁺ exchange does not contribute substantially to the trigger for release in the presence of I_CaL for voltages below +60 mV.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
The Ca²⁺ release channel of the SR opens in response to an increase in Ca²⁺ near the channel. Our data show that important differences exist between Ca²⁺ entry through the L-type Ca²⁺ channel and Ca²⁺ entry through the Na⁺-Ca²⁺ exchanger as a trigger for Ca²⁺ release. Reverse-mode Na⁺-Ca²⁺ exchange as a trigger is characterized by (1) a low Ca²⁺ entry rate except at very positive potentials and (2) a low efficiency with prolonged Ca²⁺ entry before release is triggered. Furthermore, we have shown that the contribution of reverse-mode Na⁺-Ca²⁺ exchange to the triggering of Ca²⁺ release in the presence of I_CaL can be detected only at potentials +60 mV. These characteristics predict that the role of reverse-mode Na⁺-Ca²⁺ exchange as a trigger for Ca²⁺ release during the action potential will be negligible.

Comparison With Previous Studies
Several early studies (reviewed in Reference 15¹⁵ ) had indicated that mechanical activity, ie, force development or shortening, at positive potentials could be related to Na⁺-Ca²⁺ exchange. However, in a series of studies on single cells, no Ca²⁺ release was observed at these positive potentials.^4–8,22 One possible explanation was a temperature effect, since most of these studies were done at room temperature. A first study to document Ca²⁺ release triggered by Ca²⁺ entry through the exchanger was done at 35°C¹⁶; in a comparative study, Vornanen et al¹⁷ demonstrated the importance of working at a physiological temperature. In the study of Vornanen et al and in the study of Wasserstrom and Vites,¹⁸ the voltage dependence of peak shortening was sigmoidal. Contractions at positive potentials were linked to Ca²⁺ release triggered by Na⁺-Ca²⁺ exchange; they were observed only at 35°C to 37°C.^17,18 In the study of Vornanen et al, it was noted that the time to peak shortening at very positive potentials was larger than that at +10 mV and that application of Ca²⁺ channel block increased the time to peak shortening.¹⁷ A delay for Na⁺-Ca²⁺ exchange–triggered release after Ca²⁺ channel block was also observed by Nuss and Houser,¹⁶ and it was suggested that reverse-mode Na⁺-Ca²⁺ exchange was less effective as a trigger than was I_CaL.^16,17,21 Our data confirm that Ca²⁺ entry through Na⁺-Ca²⁺ exchange can trigger release, whereas our analysis of the kinetics of the [Ca²⁺]_i transients and of the relation between Ca²⁺ release and Ca²⁺ entry establishes the low intrinsic efficiency of this mechanism.

Our experimental findings differ substantially from the results published by Bridge, Levi, and colleagues,^25–28 who used a different approach to study the contribution of Ca²⁺ entry via the exchanger. These investigators used a rapid switch device to block I_CaL just before the action potential or depolarizing test pulse to minimize changes in loading of the SR. With this approach, they found that the amplitude of cell shortening and of the [Ca²⁺]_i transient during a test pulse to +10 mV or during the action potential was decreased, but only to a moderate extent and without the slowing in the rate of upstroke that we have shown in the present study. The major difference between their approach and our approach is that we used a steady-state block of I_CaL. The advantage of our approach is that one can obtain a more complete block of I_CaL, which can otherwise pose problems because of the voltage dependence of block. A possible drawback is that loading of the SR might have been significantly decreased and that this decrease in the SR content would be responsible for the observed changes in the [Ca²⁺]_i transient. In our conditions, Ca²⁺ loading was not severely reduced and cannot account for our findings. With 20 mmol/L Na⁺ in the pipette, significant loading of the SR occurs via the Na⁺-Ca²⁺ exchanger during the conditioning prepulses, and this was not affected by I_CaL block, as demonstrated during caffeine-induced Ca²⁺ release (Fig 1). In addition, if the Ca²⁺ content had been reduced, [Ca²⁺]_i transients at all potentials would have been reduced, but in contrast, [Ca²⁺]_i transients at very positive potentials were only marginally affected (see Fig 4). Last, we and others have studied [Ca²⁺]_i transients for different degrees of Ca²⁺ loading of the SR with I_CaL as the trigger for release.^33,37,39,44 With decreased Ca²⁺ loading, the amplitude and rate of rise of the [Ca²⁺]_i transient decreased, but a delay as seen in the present study was not observed.

It has also been reported previously that intracellular Cs⁺ would reduce Ca²⁺ release from the SR.^18,39,40 Such a difference must be quantitative, since many investigators have observed Ca²⁺ release from the SR with Cs⁺-containing pipette solution, and it has been proposed that especially Na⁺-Ca²⁺ exchange–dependent release would be sensitive to inhibition by Cs⁺. Our data seem to support this view, although at present it is unclear what causes this difference. In species such as the rat, with large transient outward K⁺ currents, the use of Cs⁺ would improve voltage control and possibly reduce spurious activation of Ca²⁺ channels during large depolarization, thereby reducing Ca²⁺ release. However, this does not apply to the guinea pig myocytes. Ni²⁺-sensitive currents with either K⁺- or Cs⁺-based pipette solutions were comparable, which would seem to exclude the possibility that reduced Ca²⁺ release may be related to an inhibition of Na⁺-Ca²⁺ exchange itself. A remaining possibility is that intracellular Cs⁺ reduces Ca²⁺ release itself; block of K⁺ channels in the SR membrane⁴⁵ may affect the rate of release or decrease Ca²⁺ uptake and affect the Ca²⁺ load of the SR. The higher Cl^- concentration with CsCl-based pipette solution could also affect Cl^- transport across the SR membrane.⁴⁶

Ca²⁺ Entry Through the Na⁺-Ca²⁺ Exchanger as Trigger for Ca²⁺ Release
Ca²⁺ entry through the Na⁺-Ca²⁺ exchanger is capable of triggering Ca²⁺ release from the SR but differs substantially from the Ca²⁺ current. A striking characteristic of Ca²⁺ release triggered by Na⁺-Ca²⁺ exchange is the delay between the onset of depolarization and the actual release, a delay that is inversely related to the rate of Ca²⁺ entry through the exchanger. A similar delay is never observed with Ca²⁺ release triggered by I_CaL and is most likely related to fundamental differences between Ca²⁺ entry through both pathways. We have shown that with 20 mmol/L [Na⁺]_pip and large depolarizations, Ca²⁺ entry rate through the Na⁺-Ca²⁺ exchanger on the whole-cell level can reach values comparable to the rate of Ca²⁺ entry through I_CaL. However, Ca²⁺ entry through the exchanger appears to be intrinsically less efficient, since a larger amount of Ca²⁺ entry is required to obtain comparable Ca²⁺ release. This lower efficiency is most likely related to differences in the Ca²⁺ entry at the molecular level. As shown experimentally, the Ca²⁺ flux through a single L-type Ca²⁺ channel versus the rate of Ca²⁺ entry through a single Na⁺-Ca²⁺ exchange molecule is 2 to 3 orders of magnitude larger, ie, 0.3 pA for the L-type Ca²⁺ channel at 0 mV⁴⁷ and 0.6 to 1.3 fA for the outward exchange current (current evoked by switch from 0 to 40 mmol/L [Na⁺] in the cytosolic solution in a giant excised patch at 37°C⁴⁸). This parameter may be critical to obtain high [Ca²⁺] near the Ca²⁺ release channel. In addition to the amplitude of the single-channel Ca²⁺ influx, the localization in relation to the release channel will determine the local increase in Ca²⁺ near the SR. In studies with labeled antibodies, it was shown that Na⁺-Ca²⁺ exchange proteins could be found over the entire sarcolemma, although the density in T tubules seemed to be higher than in the external sarcolemma.^49,50 Even with Na⁺-Ca²⁺ exchange proteins localized in the vicinity of Ca²⁺ release channels, the lower molecular Ca²⁺ entry rate may still be a limiting factor. In their modeling of [Ca²⁺] in the junctional cleft between sarcolemma and SR feet, Langer and Peskoff⁵¹ propose that 11 L-type Ca²⁺ channel and 100 Na⁺-Ca²⁺ exchange proteins have access to this restricted space. With Ca²⁺ influx via the L-type Ca²⁺ channel, [Ca²⁺] at the Ca²⁺ release channels in the cleft at 1 ms after the depolarizing step is 3 orders of magnitude higher than the concentration reached by reverse-mode Na⁺-Ca²⁺ exchange after 10 ms (assuming 16 mmol/L [Na⁺]_i); this model predicts that Ca²⁺ release triggered by Na⁺-Ca²⁺ exchange would be delayed, as we have observed. In the model, simultaneous influx through the Ca²⁺ channel further reduces Ca²⁺ entry through the exchanger. However, this may be less pronounced near Ca²⁺ channels with a longer first latency.

Contribution of Reverse-Mode Na⁺-Ca²⁺ Exchange to Ca²⁺ Release From the SR
Whereas the data show that reverse-mode Na⁺-Ca²⁺ exchange can trigger release in specific experimental conditions, this trigger cannot simply be assumed to be additive to I_CaL in normal conditions. To facilitate the study of Na⁺-Ca²⁺ exchange as a trigger for release, we worked with high [Na⁺]_pip, block of I_CaL, and a wide range of voltages. These experimental conditions will lead to an overestimation of the contribution of the exchanger to triggered Ca²⁺ release during the action potential. First, [Na⁺]_i in guinea pig ventricular myocytes in normal conditions is <10 mmol/L.⁵² Second, large depolarizations, as required for the exchanger to trigger Ca²⁺ release with a delay of <50 ms, are reached only very briefly during the normal action potential. Last, in the presence of I_CaL, early activation of Ca²⁺ release is likely to interrupt Ca²⁺ entry through the Na⁺-Ca²⁺ exchanger very early. Even I_CaL itself may lessen the influx through the exchanger.⁵¹ Grantham and Cannell²⁹ have recorded significant outward Na⁺-Ca²⁺ exchange currents during action potential clamp recording after block of I_CaL. However, these authors have also pointed out that in the presence of I_CaL and I_CaL-triggered release, Ca²⁺ entry through the exchanger will be reduced. Aside from these considerations, direct evaluation of Ca²⁺ release triggered by I_CaL with 20 mmol/L versus 0 mmol/L [Na⁺]_pip failed to reveal a significant contribution of reverse-mode Na⁺-Ca²⁺ exchange to the triggering of Ca²⁺ release in the presence of I_CaL (Fig 12).

Although we have presented strong evidence that the contribution of reverse-mode Na⁺-Ca²⁺ exchange to the triggering of Ca²⁺ release is likely to be minor during a normal action potential, it is clear that (reverse-mode) Na⁺-Ca²⁺ exchange will significantly influence excitation-contraction coupling. Earlier observations on contractions^16,17,40 and the data shown in Fig 3 demonstrate the important differences between contractions and [Ca²⁺]_i transients with low or high [Na⁺]_i. The plot of peak [Ca²⁺]_i shown in Fig 3B corresponds to the plots of peak shortening published previously. With high [Na⁺]_i, [Ca²⁺]_i transients are larger and will be prolonged during maintained depolarizations. This can be related to a higher Ca²⁺ content of the SR,⁴³ with a larger fractional release,^37,39,53 and to decreased Ca²⁺ efflux or net Ca²⁺ influx during maintained depolarizations. For very strong depolarizations at +60 mV, as can only be obtained during voltage clamp, reverse-mode Na⁺-Ca²⁺ exchange will also trigger release. At these potentials, influx via the exchanger may increase the Ca²⁺ load of the SR before Ca²⁺ release is triggered and further enhance Ca²⁺ release.⁵⁴

In conclusion, our data support the view that in normal conditions, I_CaL will be the major trigger for Ca²⁺ release from the SR. Na⁺-Ca²⁺ exchange will modulate Ca²⁺ release from the SR as both reverse and forward mode contribute to the net Ca²⁺ flux in and out of the cell and thereby determine the Ca²⁺ content of the SR. Reverse-mode Na⁺-Ca²⁺ exchange may possibly become more important as a trigger in certain conditions, eg, when [Na⁺]_i is increased (such as during block of the Na⁺-K⁺ pump), in the presence of a decreased Ca²⁺ current, or possibly with an increased expression of the Na⁺-Ca²⁺ exchanger (such as has been described during cardiac hypertrophy or failure).⁵⁵ Even in these conditions, it is expected to be less efficient than I_CaL, as was also recently reported for transgenic mice overexpressing a mutant of the Na⁺-Ca²⁺ exchanger with high activity.⁵⁶ Lower efficiency of reverse-mode Na⁺-Ca²⁺ exchange as a trigger for Ca²⁺ release is most likely related to lower molecular Ca²⁺ entry rates and spatial organization.

	Selected Abbreviations and Acronyms

 

cyt (as subscript) = cytosolic solution ICaL = L-type Ca2+ current pip (as subscript) = pipette solution SR = sarcoplasmic reticulum

	Acknowledgments

 
The study was supported by a grant from the Fund for Scientific Research (FWO) and the Bekales Foundation. Dr Sipido is a postdoctoral researcher of the FWO, Belgium. The stay of Ms Maes was supported by the Born Bunge Foundation. We thank Drs Carmeliet, Mubagwa, and Callewaert for critical comments on the manuscript and Dr Verdonck for helpful discussions on [Na⁺]_i regulation and measurements of Na⁺-K⁺ pump currents.

Received March 13, 1997; accepted September 2, 1997.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Anderson K, Lai FA, Liu QY, Rousseau E, Erickson HP, Meissner G. Structural and functional characterization of the purified cardiac ryanodine receptor-Ca²⁺ release channel complex. J Biol Chem. 1989;264:1329–1335.[Abstract/Free Full Text]
2. Fabiato A. Time and calcium dependence of activation and inactivation of calcium-induced release of calcium from the sarcoplasmic reticulum of a skinned canine cardiac Purkinje cell. J Gen Physiol. 1985;85:247–289.[Abstract/Free Full Text]
3. 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.[Abstract/Free Full Text]
4. London B, Krueger JW. Contraction in voltage-clamped, internally perfused single heart cells. J Gen Physiol. 1986;88:475–505.[Abstract/Free Full Text]
5. Cannell MB, Berlin JR, Lederer WJ. Effect of membrane potential changes on the calcium transient of single rat cardiac muscle cells. Science. 1987;238:1419–1423.[Abstract/Free Full Text]
6. Beuckelmann DJ, Wier WG. Mechanism of release of calcium from sarcoplasmic reticulum of guinea-pig cardiac cells. J Physiol (Lond). 1988;405:233–255.[Abstract/Free Full Text]
7. Nabauer M, Callewaert G, Cleemann L, Morad M. Regulation of calcium release is gated by calcium current, not gating charge, in cardiac myocytes. Science. 1989;244:800–803.[Abstract/Free Full Text]
8. duBell WH, Houser SR. Voltage and beat dependence of Ca²⁺ transient in feline ventricular myocytes. Am J Physiol. 1989;257:H746–H759.[Abstract/Free Full Text]
9. Wibo M, Bravo G, Godfraind T. Postnatal maturation of excitation–contraction coupling in rat ventricle in relation to the subcellular localization and surface density of 1,4-dihydropyridine and ryanodine receptors. Circ Res. 1991;68:662–673.[Abstract/Free Full Text]
10. Carl SL, Felix K, Caswell AH, Brandt NR, Ball WJJ, Vaghy PL, Meissner G, Ferguson DG. Immunolocalization of sarcolemmal dihydropyridine receptor and sarcoplasmic reticular triadin and ryanodine receptor in rabbit ventricle and atrium. J Cell Biol. 1995;129:672–682.
11. Stern MD. Theory of excitation-contraction coupling in cardiac muscle. Biophys J. 1992;63:497–517.[Abstract/Free Full Text]
12. Wier WG, Egan TM, Lopez-Lopez JR, Balke CW. Local control of excitation-contraction coupling in rat heart cells. J Physiol (Lond). 1994;474:463–471.[Abstract/Free Full Text]
13. Lopez-Lopez JR, Shacklock PS, Balke CW, Wier WG. Local calcium transients triggered by single L-type calcium channel currents in cardiac cells. Science. 1995;268:1042–1045.[Abstract/Free Full Text]
14. Cannell MB, Cheng H, Lederer WJ. The control of calcium release in heart muscle. Science. 1995;268:1045–1049.[Abstract/Free Full Text]
15. Levi AJ, Brooksby P, Hancox JC. 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;27:1743–1757.[Free Full Text]
16. Nuss HB, Houser SR. Sodium-calcium exchange-mediated contractions in feline ventricular myocytes. Am J Physiol. 1992;263:H1161–H1169.[Abstract/Free Full Text]
17. Vornanen M, Shepherd N, Isenberg G. Tension-voltage relations of single myocytes reflect Ca release triggered by Na/Ca exchange at 35 degrees C but not 23 degrees C. Am J Physiol. 1994;267:C623–C632.[Abstract/Free Full Text]
18. Wasserstrom JA, Vites A-M. The role of Na⁺-Ca²⁺ exchange in activation of excitation-contraction coupling in rat ventricular myocytes. J Physiol (Lond). 1997;493:529–542.[Medline] [Order article via Infotrieve]
19. Leblanc N, Hume JR. Sodium current-induced release of calcium from cardiac sarcoplasmic reticulum. Science. 1990;248:372–376.[Abstract/Free Full Text]
20. Lipp P, Niggli E. Sodium current-induced calcium signals in isolated guinea-pig ventricular myocytes. J Physiol (Lond). 1994;474:439–446.[Abstract/Free Full Text]
21. Sham JSK, Cleemann L, Morad M. Gating of the cardiac Ca²⁺ release channel: the role of Na⁺ current and Na⁺-Ca²⁺ exchange. Science. 1992;255:850–853.[Abstract/Free Full Text]
22. Bouchard RA, Clark RB, Giles WR. Role of sodium-calcium exchange in activation of contraction in rat ventricle. J Physiol (Lond). 1993;472:391–413.[Abstract/Free Full Text]
23. Sipido KR, Carmeliet E, Pappano AJ. Na⁺ current and Ca²⁺ release from the sarcoplasmic reticulum during action potentials in guinea-pig ventricular myocytes. J Physiol (Lond). 1995;489:1–17.[Medline] [Order article via Infotrieve]
24. Cannell MB, Grantham CJ, Main MJ, Evans AM. The roles of the sodium and calcium current in triggering calcium release from the sarcoplasmic reticulum. Ann N Y Acad Sci. 1996;779:443–450.[Medline] [Order article via Infotrieve]
25. Levi AJ, Spitzer KW, Kohmoto O, Bridge JHB. Depolarization-induced Ca entry via Na-Ca exchange triggers SR release in guinea pig cardiac myocytes. Am J Physiol. 1994;266:H1422–H1433.