INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

IT IS WIDELY RECOGNIZED that a variety of processes can contribute to regulation of intracellular Ca²⁺ concentration ([Ca²⁺]_i) in cardiac myocytes, including Ca²⁺ uptake by the sarcoplasmic reticulum (SR) Ca²⁺-adenosinetriphosphatase (ATPase) and mitochondria, Ca²⁺ extrusion by the sarcolemmal Ca²⁺-ATPase and Na⁺/Ca²⁺ exchange, and Ca²⁺ binding by intracellular proteins (6). Experiments performed by Bassani et al. (5) as well as work from our laboratory (3) have indicated that Ca²⁺ extrusion from the cell via the sarcolemmal Ca²⁺-ATPase is approximately an order of magnitude slower than that caused by the action of the SR Ca²⁺-ATPase or the Na⁺/Ca²⁺ exchanger. The sarcolemmal Ca²⁺ pump is, therefore, unlikely to contribute significantly to the rapid decline in [Ca²⁺]_i that produces relaxation. Likewise, Ca²⁺ uptake by mitochondria is a relatively slow process that does not appear to contribute significantly to rapid relaxation (5). Therefore, uptake of Ca²⁺ by the SR Ca²⁺-ATPase and Ca²⁺ extrusion via the Na⁺/Ca²⁺ exchanger are thought to be the major contributors to the rapid decline of [Ca²⁺]_i and relaxation (7, 10).

Ca²⁺ extrusion via Na⁺/Ca²⁺ exchange has been estimated to account for 20-40% of total relaxation in ventricular myocytes, depending on the species studied and the experimental conditions (4, 20, 25). However, the relative importance of these processes at different phases of relaxation has not been established. We previously observed two phases of relaxation and decline in the [Ca²⁺]_i transient in cultured chick embryo ventricular myocytes: an initial, rapid phase with a time constant of the decay curve () of 30 ms, and a second, slower phase with a  of ~ 200 ms (2). The uptake rate of Ca²⁺ by the SR increases with increasing [Ca²⁺]_i values (Michaelis-Menten constant = ~300-500 nM) (16, 30), and the SR has a higher transport capacity than the Na⁺/Ca²⁺ exchanger (11). In addition, the Na⁺/Ca²⁺ exchanger is inhibited at depolarized membrane potentials (10). We therefore suggested that the initial phase of decline in [Ca²⁺]_i, when [Ca²⁺]_i is still relatively high and the membrane has not yet repolarized, represents primarily Ca²⁺ uptake by the SR, and the second, slower phase represents Ca²⁺ extrusion by Na⁺/Ca²⁺ exchange (2). However, recent work by Bers and Berlin (8) has clearly demonstrated that the  for [Ca²⁺]_i is dependent on the peak [Ca²⁺]_i, which complicates the assumption that a slower  for [Ca²⁺]_i decline reflects the action of a less rapid Ca²⁺ transport system. Also, the degree of Ca²⁺ loading of the SR can affect the rate of decline in [Ca²⁺]_i and relaxation (15, 21, 24).

To examine the relative importance of the SR and Na⁺/Ca²⁺ exchanger during various phases of relaxation in intact adult ventricular myocytes, we have used a rapid switcher device (SW) to trigger a contraction and abruptly disable Na⁺/Ca²⁺ exchange in isolated rabbit ventricular myocytes. Relaxation and the decline of [Ca²⁺]_i transient can then be compared in beats with and without Na⁺/Ca²⁺ exchange, under conditions in which [Ca²⁺]_i and SR Ca²⁺ loading are unchanged. Our results indicate that the effects of Ca²⁺ extrusion by Na⁺/Ca²⁺ exchange on the decline of the [Ca²⁺]_i transient are most apparent during the terminal phase of relaxation.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Dissociation of ventricular myocytes. Adult rabbit myocyte isolation was performed by a modification of a previously reported method (17). Hearts were removed from New Zealand White rabbits (2-2.5 kg) anesthetized with pentobarbital sodium (65 mg/kg iv). The heart was first immediately attached to an aortic cannula, and continuous retrograde coronary arterial perfusion at 37°C by pump (Masterflex; Cole-Parmer, Chicago, IL) was initiated at a coronary perfusion pressure of 60 mmHg. The heart was first perfused with nominally Ca²⁺-free modified Krebs-Ringer bicarbonate buffer solution for 5 min, immediately followed by 20 min of recirculating perfusion with the same solution containing 0.28 mg/ml collagenase (class II; Worthington Biochemicals, Freehold, NJ), 0.4 mg/ml hyaluronidase (type I-S; Sigma Chemical, St. Louis, MO), and 50 µM CaCl₂. Both cell isolation solutions contained (in mM) 91.7 NaCl, 30 KCl, 1.2 MgSO₄, 19 NaHCO₃, 1.2 NaHPO₄, 15 glucose, 20 taurine, and 0.5 adenosine and were gassed with 5% CO₂-95% O₂ (pH 7.40). The heart was then detached from the cannula, and the left ventricle was cut into small pieces in the same solution containing 50 µM CaCl₂, 1% albumin, and 1 mg/dl insulin (Sigma). These pieces were incubated in another solution including 50 µM Ca²⁺, 0.28 mg/ml collagenase, and 0.0025% trypsin for 10 min. The cell suspensions were mixed with the same amount of inhibitor solution including 50 µM Ca²⁺, 0.0025% trypsin inhibitor (Sigma), and 12% fetal calf serum (FCS) and centrifuged at 250 revolutions/min for 5 min. The supernatant was discarded, and the cells were resuspended in minimum essential medium (GIBCO) containing 0.1% penicillin-streptomycin.

Solutions. The control superfusate for myocytes during physiological studies was a N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES)-buffered normal Tyrode solution containing (in mM) 118 NaCl, 3.7 KCl, 0.5 MgCl₂, 1.08 CaCl₂, 5.6 glucose, and 24 HEPES (free acid) titrated to pH 7.40 with NaOH. For rapid SW experiments (see Solution switching), a Na⁺-free, Ca²⁺-free (0 Na⁺-0 Ca²⁺) solution was made by substituting KCl for NaCl and KOH for NaOH, eliminating CaCl₂, and adding 2 mM K-ethylene glycol-bis(-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA).

Measurement of [Ca²⁺]_i. The time course of changes in [Ca²⁺]_i was estimated with the Ca²⁺-sensitive fluorescent dye fluo 3. To prepare the fluo 3 loading solution, we mixed 10 ml of FCS with 234 µl of 25% Pluronic F127 (BASF; wt/wt in dimethyl sulfoxide) and sonicated the mixture. We then added 1,000 µl of fresh 1 mM fluo 3-acetoxymethyl ester (fluo 3-AM; Molecular Probes) in dimethyl sulfoxide to 9 ml of FCS-Pluronic mixture, sonicated to mix, and aliquoted into 50 200-µl samples that were stored frozen at 70°C in light-proof containers. This "loading stock" of 100 µM fluo 3-AM was diluted in physiological buffer to give 1 µM fluo 3-AM for cell loading. Myocytes on laminin-coated glass coverslips were exposed to 1 µM fluo 3-AM in physiological solution for 30 min at 30°C, washed in dye-free solution for an additional 30 min, and then perfused at 5 ml/min in a heated chamber (37°C) on the stage of an inverted epifluorescence microscope (Nikon Diaphot) for 5-10 min before experiments were begun. Probenecid (0.5 mM) was added to loading, wash, and perfusion solutions to inhibit the anion transporter and reduce the loss of dye from the cells.

Measurement of contraction. The cells were illuminated with light from the standard microscope light source passed through a 700-nm band-pass filter. The magnified image of the contracting and relaxing myocyte was transmitted by a Pulnix video camera to a television monitor. Changes in cell length were measured with a two-edge video-based motion detector (Crescent Electronics, Salt Lake City, UT). The signal from the motion detector was calibrated to indicate changes in cell length in micrometers. Relaxation was quantitated by measuring the time for the cells to return halfway from their contracted to their resting lengths and was expressed as relaxation half-time (t_1/2). Simultaneous motion and fluo 3 [Ca²⁺]_i transients could not be routinely measured with this system because of "spillover" of illumination light into the 530-nm fluo 3 fluorescence PMT. Therefore, studies of motion and [Ca²⁺]_i were made using the same protocol in different cells. In a few instances, an intensified charge-coupled device camera (Gen II+, Stanford Photonics), which was more sensitive, was used to permit imaging of the cell at 700 nm, a low brightness that does not interfere with the 530-nm fluo 3 fluorescence measurements.

Electrophysiology measurements. Membrane potential was measured with suction pipettes using an Axoclamp 2B amplifier system (Axon Instruments, Foster City, CA) in bridge mode. Pipettes were made from borosilicate capillary tubing (Corning 7052) and had resistances of 2-4 M when filled with a solution containing (in mM) 113 KCl, 5 or 10 NaCl, 0.5 MgCl₂, 5 K₂ATP, 10 HEPES, 5 dextrose, and 11 KOH (pH 7.1). Action potentials were elicited by intracellular injection of threshold constant current pulses (3-ms duration).

Solution switching. A solenoid-driven rapid solution SW was used to make fast changes of the extracellular solution surrounding a single myocyte (28). Two adjacent microstreams flowed simultaneously from neighboring square glass tubes (each 200 µm wide) separated by a 70-µm glass septum. A myocyte was positioned in the control stream so that activation of the solenoid moved the two streams laterally and suddenly exposed the cell to the test solution stream. Activation of the SW was controlled by a stimulation pulse 2 s in duration and delivered 4 s after the last of a train of eight 0.25-Hz electrically stimulated (ES) contractions.

Data analysis. Motion and Ca²⁺ data were digitized and analyzed using a Digidata 1200A analog-to-digital converter and Axoscope 1.0 (Axon Instruments) and Micro Origin 4.0 (Microcal) software running on a PC Innovation (West Valley City, UT) computer. Curve fitting was carried out using the Origin program. The statistical significance of differences in motion and [Ca²⁺]_i transient parameters were assessed using Student's t-test or analysis of variance as appropriate.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Rapidity of SW-induced changes in perisarcolemmal K⁺ and Ca²⁺ concentrations. To define the role of the Na⁺/Ca²⁺ exchanger in the decline of [Ca²⁺]_i, we sought to rapidly inactivate the "forward" Na⁺/Ca²⁺ exchange (Na⁺ in, Ca²⁺ out) by rapid removal of extracellular Na⁺ and to prevent Ca²⁺ influx by "reverse" exchange (Ca²⁺ in, Na⁺ out). We therefore measured the rate at which changes in external concentrations of Na⁺ and Ca²⁺ could be accomplished with a rapid SW. With this device, the bulk solution surrounding a myocyte can be exchanged within ~2 ms. However, an intact cardiac myocyte has unstirred layers, which include an extensive T-tubule system. To estimate the time required for exchange of the univalent cation concentration adjacent to the sarcolemma, we performed the experiments illustrated in Fig. 1. In Fig. 1A, an isolated myocyte was abruptly exposed to an elevated KCl concentration (8.8 mM) while membrane potential was measured with a patch electrode. Exposure to high K⁺ caused the expected depolarization of membrane potential, which was virtually complete within 100 ms. In Fig. 1B, a cell was exposed to a 0 Na⁺-0 Ca²⁺ solution with a higher KCl concentration. For comparison, Fig. 1B shows an action potential evoked by an ES in normal external solution in the same cell. Exposure to high-K⁺ solution (SW activated) caused an action potential that was followed by a sustained depolarization to 0 mV. Compared with the electrically activated action potential, the SW-activated action potential had a slightly lower peak amplitude and a reduced plateau, probably secondary to the reduced influx of Na⁺ and Ca²⁺. On the basis of these results, we estimate that the univalent cation concentration adjacent to the sarcolemma can be changed abruptly with the SW device within 100 ms. Furthermore, it is apparent that exposure to a high-K⁺ solution by means of the SW can induce a rapid enough depolarization to elicit an action potential.

View larger version (13K): [in this window] [in a new window]   Fig. 1.   A: a myocyte was abruptly exposed to an elevated external K+ concentration ([K+]o; 8.8 mM) by means of rapid switcher device, and time course of resulting membrane depolarization (bottom) was measured with a patch electrode. Top record shows switcher solenoid. B: membrane potential change evoked by sudden exposure (switcher ON) to a higher [K+]o (130 mM) in presence of 0 Na+-0 Ca2+. High-K+, 0 Na+-0 Ca2+ solution caused an abbreviated action potential, followed by sustained membrane depolarization. Similar results were obtained in 3 myocytes. An action potential elicited by an electrical stimulation from same cell before switcher pulse is also shown for comparison. See text for discussion.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Estimated time required to remove external Ca2+ from unstirred layer of a ventricular myocyte. Cell was field stimulated to elicit contractions. Cell shortening was measured with a video motion detector. Cell was initially exposed to microstream containing 1.08 mM Ca2+. A: contraction was completely blocked by exposing cell to 0 Ca2+ (no added Ca2+, 2 mM ethylene glycol-bis(-aminoethyl ether)-N,N,N',N'-tetraacetic acid) 500 ms before stimulus (arrowhead). B: removal of Ca2+ 100 ms before stimulus induced a small contraction. C: summary of results suggests that Ca2+ is washed from unstirred layer in ~200 ms. [Ca], Ca2+ concentration.

Comparison of rate of relaxation of SW-induced contractions with ES contractions. In view of results described in Rapidity of SW-induced changes in perisarcolemmal K⁺ and Ca²⁺ concentrations, it seemed likely that abrupt exposure to a high-K⁺, Ca²⁺-free EGTA solution would induce depolarization of the cell with resultant opening of the voltage-sensitive Ca²⁺ channels within <50 ms when the extracellular Ca²⁺ concentration was still sufficient to provide adequate influx of Ca²⁺ via the L-type Ca²⁺ channel to trigger a contraction. However, the subsequent relaxation of the cell would occur in the effective absence of extracellular Na⁺ and Ca²⁺, thus inhibiting the function of the Na⁺/Ca²⁺ exchanger and preventing Ca²⁺ influx during the declining phase of the [Ca²⁺]_i transient. To test this possibility, we examined the effects of abrupt exposure to 140 mM K⁺ (0 Na⁺-0 Ca²⁺ in the presence of 2 mM EGTA) on cell motion. An example is shown in Fig. 3A. In this protocol, the cell was paced at a rate of 0.25 Hz for 8 consecutive stimuli. This resulted in a stable level of contractile amplitude. After the eighth ES, the cell was activated by abrupt exposure to the 0 Na⁺-0 Ca²⁺ 140 mM K⁺ solution (SW) and the contraction and relaxation were measured. The amplitude of the contraction induced by the SW was virtually identical to that induced by the preceding ES. The t_1/2 of relaxation was slightly longer for the SW beats, but this increase did not reach statistical significance, as shown in the average results in Fig. 3B for 10 separate cells.

View larger version (11K): [in this window] [in a new window]   Fig. 3.   A: traces (control) show motion signals induced by last of a train of electrical stimuli (ES) delivered at 0.25 Hz and a subsequent contraction induced by abrupt exposure to high-K+, 0 Na+-0 Ca2+ solution (SW). B: average half-time of relaxation (t1/2) for ES and SW beats. Values are means ± SE; n = 10 myocytes.

View larger version (17K): [in this window] [in a new window]   Fig. 4.   Simultaneous recording of motion and fluo 3 fluorescence ([Ca2+]i transient) in an adult rabbit ventricular myocyte imaged with an intensified charge-coupled device camera and stimulated to contract with extracellular electrodes. Fluo 3 fluorescence peaked well before maximal shortening, and decline in fluorescence was nearly complete before mechanical relaxation occurred. Relaxation of unloaded myocyte was much more abrupt than decline of fluo 3 fluorescence transient.

View larger version (13K): [in this window] [in a new window]   Fig. 5.   Example of a fluo 3 fluorescence transient induced by an ES vs. SW stimulus. Protocol was same as in Fig. 3. Amplitude and initial decline of fluo 3 transient were similar in ES and SW beats, but duration of peak was shortened and terminal phase of relaxation was prolonged in SW transient.

Effects of impaired SR function and Ca²⁺ binding on relaxation. We investigated the difference between the rate of relaxation in ES beats and SW beats in myocytes in which the SR function was impaired. We exposed adult rabbit myocytes to 500 nM thapsigargin for 15 min to inhibit SR Ca²⁺-ATPase (18), and the effects are shown in Fig. 6. Exposure to thapsigargin caused a reduced rate of shortening and a prolongation in the rate of relaxation. However, when an SW contraction was initiated, essentially no relaxation was detected until the extracellular Na⁺ concentration was restored to normal at the end of the SW pulse. This indicates, as expected, that the Na⁺/Ca²⁺ exchanger is effectively disabled by the SW protocol and that Ca²⁺ transport processes other than Na⁺/Ca²⁺ exchange and SR function do not produce rapid relaxation in these cells.

View larger version (12K): [in this window] [in a new window]   Fig. 6.   A: effects of inhibition of sarcoplasmic reticulum Ca2+-adenosinetriphosphatase with 500 µM thapsigargin on contraction and relaxation of ES and SW beats. Protocol was same as in Fig. 3. B: abrupt elimination of Na+/Ca2+ exchange produced a marked impairment of relaxation in SW beats.

View larger version (11K): [in this window] [in a new window]   Fig. 7.   A: effects of Ca2+ buffering with indo 1 on relaxation of ES and SW contractions. Cells were loaded with 10 µM indo 1-acetoxymethyl ester for 15 min. Protocol was as described in Fig. 3. B: Ca2+ buffering slowed onset of contraction and relaxation, but abrupt inhibition of Na+/Ca2+ exchange did not affect t1/2.

Importance of Na⁺/Ca²⁺ exchange during terminal phase of [Ca²⁺]_i decline. To examine the relative contributions of SR and Na⁺/Ca²⁺ exchange to the decline in [Ca²⁺]_i during different phases of relaxation, we measured end-diastolic (239 ± 30 nM) and peak systolic (700 ± 87 nM) calibrated [Ca²⁺]_i values in 14 cells studied under identical conditions (37°C, pacing at 0.25 Hz). These values are similar to those we have previously measured (17) in these cells with indo 1 under comparable conditions (37°C, 0.5-Hz stimulation). We used these average end-diastolic and peak systolic [Ca²⁺]_i values to convert the fluo 3 fluorescence traces of ES and SW signals averaged from 12 cells to [Ca²⁺]_i traces. The averaged [Ca²⁺]_i traces for ES and SW beats during the declining phase of the [Ca²⁺]_i transient, beginning at 90% of the peak of the ES transient, are displayed in Fig. 8A. The initial decline in the [Ca²⁺]_i is very similar for ES and SW beats, but there is a slowing of the terminal decline in [Ca²⁺]_i in the SW beats, in which the Na⁺/Ca²⁺ exchanger is abruptly inhibited. In addition, the diastolic level to which the [Ca²⁺]_i declined was higher in SW than ES beats. The difference in the ES and SW [Ca²⁺]_i transient declines (difference transient), which represents the component of [Ca²⁺]_i decline due to Na⁺/Ca²⁺ exchange, is also apparent in Fig. 8A. We plotted the first derivative of [Ca²⁺]_i with respect to time against [Ca²⁺]_i for SR function (SW transient) and for Na⁺/Ca²⁺ exchanger function (difference transient) (Fig. 8C). There is a measurable contribution of Na⁺/Ca²⁺ exchange to the decline in the [Ca²⁺]_i transient. This contribution first becomes apparent ~50 ms after the Ca²⁺ has declined to 90% of its peak value. The SR and Na⁺/Ca²⁺ exchanger curves shown in Fig. 8C were extrapolated to peak systolic (700 nM) and end-diastolic (239 nM) [Ca²⁺]_i values. Subsequent integration of the areas under these curves revealed that the Na⁺/Ca²⁺ exchanger contributes ~13% to the first 50% decline and 45% to the last 50% of the decline of the total [Ca²⁺]_i transient.

View larger version (14K): [in this window] [in a new window]   View larger version (14K): [in this window] [in a new window]   View larger version (11K): [in this window] [in a new window]   Fig. 8.   A: composite ES and SW [Ca2+]i transients averaged from 12 records and plotted from 90% of peak value during [Ca2+]i decline. Difference transient shows difference between SW and ES traces, indicating component due to Na+/Ca2+ exchange. B: solid lines overlying same transients in A indicate exponential fit of raw data. C: from curve-fit lines in B, absolute derivative of [Ca2+]i with respect to time (d[Ca2+]i/dt) was plotted for SW transient, indicating sarcoplasmic reticulum function, and for difference transient, indicating Na+/Ca2+ exchanger function. See text for discussion.

Forward and reverse Na⁺/Ca²⁺ exchange during excitation-contraction coupling and relaxation in rabbit ventricular myocytes. Abrupt elimination of Na⁺/Ca²⁺ exchange by the described SW protocol could decrease reverse as well as forward exchange during the plateau of the action potential and the peak of the [Ca²⁺]_i transient. Thus the relatively small effect of elimination of Na⁺/Ca²⁺ exchanger function on the initial decline in [Ca²⁺]_i could reflect a reduction in Ca²⁺ influx by reverse exchange as well as indicating that Ca²⁺ extrusion by forward exchange is relatively unimportant during this phase of [Ca²⁺]_i decline. To assess this possibility, we estimated the reversal potential for the Na⁺/Ca²⁺ exchanger (E_Na/Ca) during a [Ca²⁺]_i transient and action potential (Fig. 9). This analysis was performed for an intracellular Na⁺ concentration ([Na⁺]_i) of 5 mM, which we have recently measured with sodium green and sodium-binding benzofuran isophthalate in rabbit ventricular myocytes under these experimental conditions, and for an [Na⁺]_i of 10 mM, allowing for a possible doubling of subsarcolemmal [Na⁺]_i relative to cytosolic [Na⁺] due to "fuzzy space" (12). Figure 9 shows that membrane potential (E_m) > E_Na/Ca, indicating reverse exchange, only during the initial upstroke of the action potential and the early part of the [Ca²⁺]_i transient. Thereafter, E_m < E_Na/Ca, resulting in forward exchange. The difference between E_Na/Ca and E_m is much larger in the late phase of decline of [Ca²⁺]_i than during the initial phase, suggesting that the forward function of the Na⁺/Ca²⁺ exchanger would be most apparent during this period. This would predict that the Na⁺/Ca²⁺ exchanger contributes principally to the terminal decline of the [Ca²⁺]_i transient and is thus consistent with our results.

View larger version (18K): [in this window] [in a new window]   Fig. 9.   Membrane potential (Em) was measured with a suction pipette during pacing at 0.25 Hz, and a fluo 3 [Ca2+]i transient was measured simultaneously. Reversal potential for Na+/Ca2+ exchange (ENa/Ca) was calculated from average end-diastolic [Ca2+]i (239 nM), peak systolic [Ca2+]i (700 nM), extracellular [Ca2+] (1.08 mM), extracellular Na+ concentration (140 mM), and intracellular Na+ concentration ([Na+]i; 5 and 10 mM). Na+/Ca2+ exchanger will operate in "reverse" mode when Em > ENa/Ca. From this analysis, Na+/Ca2+ exchanger functions almost exclusively in forward mode during decline of [Ca2+]i transient, and difference between ENa/Ca and Em, which is driving force for Na+/Ca2+ exchanger, is greatest during terminal phase of [Ca2+]i decline. Pipette [Na+]i was 5 mM and temperature was 32°C in this action potential recording.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

Use of a rapid SW to induce a contraction and disable Na⁺/Ca²⁺ exchanger. Our results clearly indicate that abrupt exposure to 0 Na⁺-0 Ca²⁺ 140 mM K⁺ solution by means of a rapid SW can induce contractions and [Ca²⁺]_i transients that are identical in amplitude to those caused by a preceding ES delivered after an identical diastolic interval. We believe this is because the rapid exposure to 140 mM K⁺ induces depolarization and opening of the L-type Ca²⁺ channel within a time interval (<50 ms) during which the [Ca²⁺] adjacent to the Ca²⁺ channel has not yet declined significantly because of the slower diffusion of Ca²⁺ compared with K⁺, demonstrated in Figs. 1 and 2.

Relative importance of SR and Na⁺/Ca²⁺ exchanger function during initial and terminal relaxation. Bassani et al. (4) reported that inhibition of the Na⁺/Ca²⁺ exchanger prolongs the overall time constant of the decline in the [Ca²⁺]_i transient measured with indo 1 by as much as 45% in normal rabbit myocytes. In their experiments, the cells were first exposed to a 0 Na⁺-0 Ca²⁺ solution for 5-7 min, and then Ca²⁺ was restored in the presence of 0 Na before ES. Our experiments, however, have shown only a small and statistically insignificant delay in the initial rate of relaxation and decline in [Ca²⁺]_i when the Na⁺/Ca²⁺ exchanger is abruptly disabled in adult rabbit ventricular myocytes. We believe this difference in results may be attributed to several factors. First, in our experiments we used a low concentration of the Ca²⁺ indicator fluo 3. This indicator has a higher K_d for Ca²⁺ than indo 1 (864 vs. 260 nM) and thus would be expected to buffer Ca²⁺ less markedly than indo 1, making it possibly more suitable for examining the rapid kinetics of initial Ca²⁺ decline. Our experiments with fluo 3 do indicate clearly that Na⁺/Ca²⁺ exchange does contribute importantly to the rate of decline in [Ca²⁺]_i from the midportion of the [Ca²⁺]_i transient and appears to be necessary to reach end-diastolic [Ca²⁺]_i levels. In the experiments of Bassani et al. (4), resting [Ca²⁺]_i was also somewhat increased in the 0 Na⁺-0 Ca²⁺ solution, although this increase did not reach statistical significance. Second, in the experiments of Bassani et al. (4), simultaneous exposure to 0 Na⁺-0 Ca²⁺ solution could have resulted initially in some reversal of the Na⁺/Ca²⁺ exchange due to a difference in rate of removal of perisarcolemmal Ca²⁺ and Na⁺, with increased Ca²⁺ loading of the SR. Increased Ca²⁺ influx and loading of the SR can slow relaxation (24), and thus an increase in the degree of SR Ca²⁺ loading could have caused a slight slowing of relaxation in these experiments. The amplitude of contraction and the amplitude of the [Ca²⁺]_i transient was somewhat greater in the experiments of Bassani et al. (4) in myocytes stimulated in the presence of 0 Na⁺ and normal Ca²⁺ compared with that in myocytes stimulated under control conditions. This was attributed by Bassani et al. (4) to extrusion of Ca²⁺ by Na⁺/Ca²⁺ exchange during the peak of the [Ca²⁺]_i transient in a normal beat (see also Ref. 20). However, in our experiments, the amplitude of the [Ca²⁺]_i transient and the amplitude of the contraction were identical in ES beats and in beats in which the Na⁺/Ca²⁺ exchanger was abruptly disabled by the use of the SW. Because the extent of SR loading during ES and SW beats should be identical in our experiments, our findings suggest that Ca²⁺ extrusion by Na⁺/Ca²⁺ exchange does not significantly reduce the peak magnitude of the [Ca²⁺]_i transient during a normal beat. This conclusion is also supported by data shown in Fig. 9 and by the findings of Bouchard et al. (9) and is predicated by the dynamic model of the cardiac ventricular action potential formulated by Luo and Rudy (22).

Potential value of SW method for assessment of SR function in intact myocytes. As previously demonstrated (3, 4, 10) in myocytes in which SR function has been pharmacologically impaired by exposure to caffeine or thapsigargin, the dependence of relaxation on Na⁺/Ca²⁺ exchanger function is much more marked. In these situations, abrupt disabling of the Na⁺/Ca²⁺ exchanger produces a dramatic prolongation of relaxation, as shown in Figs. 3 and 6. Our results also confirm previous reports by Noble and Powell (26) that Ca²⁺ indicators such as indo 1 can slow relaxation by buffering Ca²⁺. In rabbit myocytes well loaded with indo 1, a prolongation of relaxation comparable to that achieved with thapsigargin was achieved (t_1/2 = 350 ms). However, in myocytes in which relaxation was prolonged because of abnormal Ca²⁺ buffering within the cell, abrupt disabling of the Na⁺/Ca²⁺ exchanger had a very minor effect on relaxation, as shown in Fig. 7. Thus the magnitude of the effect of abruptly disabling the Na⁺/Ca²⁺ exchanger on relaxation appears to depend on the cause of the slow relaxation.