INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

IN THE HEART, ischemia and reperfusion markedly change the coupling of excitation to contraction [excitation-contraction (E-C) coupling]. Ischemia causes abbreviation of action potentials and a decline in resting membrane potential (RMP) (13, 23). In addition, contractile activity is markedly decreased (26). With reperfusion, electrical activity may recover fully, although a transient period of arrhythmias may occur early in reperfusion (34). However, contractile activity may show a prolonged period of depression that lasts well beyond electrical recovery (19; for a review, see Ref. 11). This prolonged contractile dysfunction is called myocardial stunning (6).

Most studies of myocardial stunning have been conducted in in situ hearts or in isolated perfused hearts. These investigations largely focused on factors that reduce or exacerbate myocardial stunning (for a review, see Ref. 5). Less is known about the cellular changes that underlie myocardial stunning. However, some studies have described changes in expression and function of proteins involved in E-C coupling. Decreases in the amounts of different proteins that are important in sequestration and release of Ca²⁺ from the sarcoplasmic reticulum (SR) also have been reported. For example, SR Ca²⁺-ATPase, which is responsible for SR uptake of Ca²⁺, has been reported to be decreased in stunned hearts (32, 36). The density of SR Ca²⁺ release channels (ryanodine receptors) also has been reported to be reduced in stunning (21, 37, 39, cf 38). In addition, phosphorylation of both SR Ca²⁺-ATPase and ryanodine receptors by Ca²⁺/calmodulin-dependent protein kinase may be decreased in stunning (32). Other studies indicate that decreased myofilament Ca²⁺ sensitivity plays an important role in contractile depression in stunning (17, 20, 31). This decrease in Ca²⁺ sensitivity may reflect proteolysis of troponin I in stunning (18). Together, these observations suggest that changes in SR Ca²⁺ uptake, SR Ca²⁺ release, and myofilament Ca²⁺ sensitivity might all contribute to contractile depression in stunning. Furthermore, decreased density of L-type Ca²⁺ channels and reduced capacity of the sarcolemmal Na⁺/Ca²⁺ exchanger have been observed in ischemia (3, 10, 33, 35). These findings suggest that transsarcolemmal Ca²⁺ fluxes also may be modified by ischemia and could contribute to stunning in reperfusion.

Although changes in many proteins that are central to E-C coupling have been reported, the consequences of these changes to stunning at the cellular level are not clear. The overall goals of the present study were to establish a single cell model of stunning in ventricular myocytes and to evaluate changes in E-C coupling that contribute to stunning in this model. In the present study, we adapted an isolated myocyte model of simulated ischemia and reperfusion, developed previously by us (8, 9), to investigate cellular changes in stunning. This model permits simultaneous measurements of transmembrane voltage, ion currents, and cell shortening. In addition, cytosolic Ca²⁺ levels and transients also can be measured in this model with Ca²⁺-sensitive fluorescent dyes.

The specific objectives of this study were as follows: 1) to determine whether stunning can be elicited in our isolated myocyte model of ischemia and reperfusion; 2) to determine the possible roles of RMP, action potential configuration, and L-type Ca²⁺ current (I_Ca,L) in stunning in this model; 3) to compare changes in the efficacy of different mechanisms for E-C coupling in stunning; and 4) to determine whether stunning is related to changes in the magnitude of Ca²⁺ transients or Ca²⁺ release from SR stores in this model.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Cell isolation. Experiments were conducted on isolated ventricular myocytes from guinea pigs and performed in accordance with the guidelines published by the Canadian Council on Animal Care. Male guinea pigs (325-375 g, Charles River) were anesthetized with pentobarbital sodium (160 mg/kg with 3.3 IU/g ip heparin). The aorta was cannulated in situ, and the heart was removed and perfused for 7-8 min at 10-12 ml/min with oxygenated (100% O₂, 37°C) Ca²⁺-free solution of the following composition (in mM): 120 NaCl, 3.8 KCl, 1.2 KH₂PO₄, 1.2 MgSO₄, 10 HEPES, and 11 glucose (pH 7.4 with NaOH). Collagenase A (25 mg/50 ml buffer, Boehringer-Mannheim) and protease (4.8 mg/50 ml, Sigma type XIV) were then included in the perfusate for 5 min. The ventricles were then minced in a buffer with the following composition (in mM): 80 KOH, 50 glutamic acid, 30 KCl, 30 KH₂PO₄, 20 taurine, 10 HEPES, 10 glucose, 3 MgSO₄, and 0.5 EGTA (pH 7.4 with KOH). Myocytes were placed in a 0.75-ml chamber on the stage of an inverted microscope. After 5-10 min, they were superfused at 3 ml/min at 37°C with Tyrode solution [containing (in mM) 129 NaCl, 20 NaHCO₃, 0.9 NaH₂PO₄, 4 KCl, 0.5 MgSO₄, 2.5 CaCl₂, and 5.5 glucose (pH 7.4)] gassed with 95% O₂-5% CO₂. Experiments were performed only on cells that were free of membrane blebs and had a maximum diastolic potential more negative than 85 mV.

General methods. Myocytes were superfused for 10 min with Tyrode solution and then for 30 min with a solution mimicking the specific conditions of myocardial ischemia, including hypoxia, hypercapnia, hyperkalemia, acidosis, lactate accumulation, and substrate deprivation (14, 16). This "ischemic solution" had the following composition (in mM): 123 NaCl, 6 NaHCO₃, 0.9 NaH₂PO₄, 8 KCl, 0.5 MgSO₄, 2.5 CaCl₂, and 20 Na-lactate, gassed with 90% N₂-10% CO₂ (pH 6.8). A 90% N₂-10% CO₂ gas phase was layered over the experimental chamber during simulated ischemia. Reperfusion was simulated by return to Tyrode solution and removal of the gas phase. Cells were exposed to only one cycle of ischemia and reperfusion.

Measurement of action potentials and ionic currents. Cells were impaled with high-resistance electrodes (18-25 M) to minimize dialysis and avoid buffering intracellular Ca²⁺ levels. Electrodes were filled with 2.7 M KCl, and a 2.7 M KCl-agar bridge was used as a bath ground. Action potentials were recorded with conventional microelectrode techniques. Currents were recorded with discontinuous single electrode voltage-clamp techniques (sample rate 8-10 kHz). Recordings were made with an Axoclamp 2B amplifier (Axon Instruments).

Fluorescence measurements. Intracellular Ca²⁺ was measured by whole cell photometry (DeltaRam, Photon Technology International). Myocytes were loaded with fura 2 by incubating them with the acetoxymethyl ester (0.1 µM) for 20 min at room temperature. The ratio of emission at 510 nm, during alternate excitation at 340 nm and 380 nm, was used to determine intracellular Ca²⁺ concentrations ([Ca²⁺]_i). The fluorescence ratio was converted to [Ca²⁺] with a calibration curve determined in vitro experimentally at pH 7.2. The calibration curve was determined in the same experimental chamber and with the same optical path as used for data collection. Because intracellular pH has been reported to recover to preischemic levels rapidly upon reperfusion (22, 24), the calibration curve determined at pH 7.2 is appropriate for measurement of intracellular Ca²⁺ in reperfusion, which is the focus of this study. However, it should be noted that use of this calibration curve may have resulted in a slight underestimate in the measurements of intracellular Ca²⁺ made during the ischemic period. This may have occurred because the pH of the ischemic solution was 6.8, which has been shown experimentally to change the dissociation constant for fura 2 from ~0.14 to 0.18 µM (30). Fluorescence was recorded and measured with Felix software (version 1.4, Photon Technology International). The Ca²⁺ transients were measured relative to diastolic [Ca²⁺] (background subtracted).

Estimation of SR Ca²+ stores. Amplitudes of caffeine-elicited Ca²⁺ transients were used as a measure of SR Ca²⁺ content during ischemia and reperfusion. In these experiments, cells were impaled and exposed to control conditions for 10 min, ischemia for 30 min, and reperfusion for 40 min. A second group of cells were impaled, but not exposed to ischemic conditions, and served as time controls. In both groups, cells were stimulated with current pulses delivered through the microelectrode at 2 Hz. At 10, 40, 55, 70, and 80 min during the experiment, stimulation was briefly interrupted, and 10 mM caffeine was applied to cells for 1 s at 37°C with a rapid solution changer. Changes in solution with this device were triggered by the voltage-clamp protocol and were complete within 300 ms (27).

Ca²+ transients and cell length in field-stimulated myocytes. In some experiments, myocytes were field stimulated continuously at 2 Hz through a pair of platinum electrodes. Myocytes were exposed to 30 min of simulated ischemia and 40 min of reperfusion. Parallel time controls were performed. Changes in cell length and fura 2 [Ca²⁺]_i were measured in separate experiments. Diastolic cell length and [Ca²⁺]_i were measured at a point immediately before responses. Systolic cell length and [Ca²⁺]_i were measured at peak values during responses. Cell shortening and Ca²⁺ transients were calculated as the difference between diastolic and systolic measurements. Cell length and [Ca²⁺]_i were recorded at 5-min intervals throughout the experiment except for the first 5 min of reperfusion, when recordings were made every minute. Three responses were averaged for each recording period.

Data analyses. Differences between experimental groups and time controls were tested for statistical significance with a two-way repeated-measures ANOVA. All other data were analyzed relative to preischemic values using a one-way repeated-measures ANOVA. Post hoc comparisons were made with a Bonferroni test. Statistical analyses were performed with SigmaStat (version 2.0, Jandel). Data are presented as means ± SE. The value of n represents the number of myocytes sampled. In most cases, each myocyte came from a different heart. Occasionally more than one myocyte from the same heart was utilized, therefore the numbers of myocytes and hearts are indicated in the figures for each data set. Because most myocytes were from different hearts, the use of multiple samples was not incorporated in the statistical analysis.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Effects of ischemia and reperfusion on action potentials and contractions. Figure 1 shows representative recordings of action potentials and contractions at selected times during an experiment. Action potentials abbreviated and cell membranes depolarized during ischemia (Fig. 1A). Ischemia also caused a marked reduction in the amplitude of contraction (Fig. 1B). Action potential configuration recovered rapidly upon reperfusion (Fig. 1A). Amplitudes of contractions increased in early reperfusion but decreased with continued reperfusion (Fig. 1B).

View larger version (10K): [in this window] [in a new window]   Fig. 1.   Changes in membrane potential and cell shortening in cells exposed to simulated ischemia and reperfusion. Sample recordings are shown for a myocyte in control, ischemia, early (5 min) reperfusion, and late (20 min) reperfusion. Action potentials (A) and associated contractions (B, top) were elicited by current pulses (B, bottom). During ischemia, action potentials shortened, cell membranes depolarized, and contractions were reduced in magnitude. During reperfusion, action potentials gradually recovered to control levels, although recovery in this example was not quite complete. In contrast, contractions recovered in early reperfusion but decreased again in late reperfusion.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Action potentials were normal, but myocytes exhibited contractile depression (stunning) in late reperfusion. Ischemia was associated with a gradual decrease in action potential duration (APD; A) and elevation in resting membrane potential (RMP; B). Mean contraction amplitude also was decreased during ischemia (C). RMP and APD both recovered to preischemic levels by late reperfusion, when significant contractile depression (stunning) was observed. * Significant difference from control, P < 0.05; n = 21 myocytes from 21 hearts.

Effects of ischemia and reperfusion in voltage-clamped myocytes. Similar experiments were conducted in voltage-clamped myocytes to eliminate effects mediated by changes in action potential configuration. The voltage-clamp protocol used in these experiments is shown in Fig. 3, inset. Representative currents and contractions, recorded at selected times during ischemia and reperfusion, are shown in Fig. 3, A and B. Peak inward current declined in ischemia and showed little recovery in reperfusion (Fig. 3A). Contractions also were depressed by ischemic conditions but, in contrast to current, exhibited rapid recovery early in reperfusion, followed by sustained depression later in reperfusion (Fig. 3B).

View larger version (10K): [in this window] [in a new window]   Fig. 3.   Changes in membrane currents and contractions in voltage-clamped myocytes exposed to simulated ischemia and reperfusion. The voltage-clamp protocol (inset) was used to elicit L-type Ca2+ current (ICa,L; A) and Ca2+-induced Ca2+ release (CICR) contractions (B). Representative recordings show that peak ICa,L was reduced during ischemia and did not recover in reperfusion. CICR contractions were decreased in ischemia, recovered in early reperfusion, and were reduced in late reperfusion.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Stunned myocytes exhibited reduced ICa,L and recovery of steady-state current (ISS). Mean ICa,L amplitude was gradually reduced during ischemia and did not recover in reperfusion (A). However, outward ISS increased during ischemia and recovered to control levels in late reperfusion (B). As was observed for contractions elicited by action potentials, contractions elicited by CICR (C) were reduced during ischemia and recovered rapidly in early reperfusion. In late reperfusion, myocytes exhibited significant stunning. * Significant difference from control, P < 0.05; n = 21 myocytes from 21 hearts.

View larger version (14K): [in this window] [in a new window]   Fig. 5.   Contractions initiated by the voltage-sensitive release mechanism (VSRM) and CICR were similarly depressed in late reperfusion. The two-step voltage protocol (A) was used to separately elicit VSRM and CICR contractions. VSRM contractions were elicited by a voltage step to 40 mV and were associated with a small inward current. ICa,L and CICR contractions were observed in response to the voltage step to 0 mV. VSRM and ICa,L contraction amplitudes were similarly affected by ischemia and reperfusion (B and C, respectively). Contractions initiated by both mechanisms were depressed compared with control levels during ischemia and late reperfusion. * Significant difference from control, P < 0.05; n = 21 myocytes from 21 hearts.

Effects of ischemia and reperfusion on the magnitude of SR Ca²+ stores. Figure 6A shows Ca²⁺ transients recorded from a myocyte stimulated at 2 Hz in normal Tyrode solution. Each stimulus elicited a rapid transient rise in [Ca²⁺]_i. Stimulation was interrupted briefly and the superfusate was rapidly switched for 1 s to one containing 10 mM caffeine. Caffeine application elicited a Ca²⁺ transient, which was taken as a measure of SR Ca²⁺ stores. After caffeine application, Ca²⁺ transients elicited by stimulation were temporarily reduced in magnitude. Caffeine-elicited Ca²⁺ transients were measured in cells exposed to ischemia and reperfusion and in cells that were not exposed to ischemia and therefore served as time controls (Fig. 6B). Surprisingly, the magnitudes of caffeine-induced transients did not change significantly in response to ischemia and reperfusion compared with time controls (Fig. 6B). These data indicate that the decrease in contraction observed late in reperfusion (Figs. 2, 4, and 5) cannot be attributed to a decline in SR stores of Ca²⁺.

View larger version (20K): [in this window] [in a new window]   Fig. 6.   Caffeine-elicited Ca2+ transients were normal in stunned myocytes. A: representative recording of intracellular Ca2+ concentration ([Ca2+]i) in a myocyte exposed to 10 mM caffeine. A 1-s caffeine application was interpolated in a 2-Hz stimulus train as indicated. The magnitude of caffeine-elicited Ca2+ transients was not altered from values at 10 min in time control experiments or during ischemia and reperfusion (B). Data are presented as a percentage of preischemic values (ischemia-reperfusion, n = 13 myocytes from 12 hearts; time controls, n = 6 myocytes from 6 hearts).

Relation between cytosolic Ca²+ concentration and cell length in myocytes exposed to ischemia and reperfusion. To determine whether changes in cell length in ischemia and reperfusion are caused by corresponding changes in cytosolic Ca²⁺, we compared Ca²⁺ levels and cell shortening in field-stimulated myocytes. Figure 7 shows representative recordings of Ca²⁺ concentration and cell length in time controls in which myocytes were not exposed to ischemia. Diastolic Ca²⁺ concentration and Ca²⁺ transients were well maintained (Fig. 7A). Contractions also were stable (Fig. 7B). Figure 8A shows mean data for diastolic and systolic Ca²⁺ concentrations in time controls. Mean data demonstrate a slight rundown in the magnitude of Ca²⁺ transients (shaded region, Fig. 8A) and a gradual increase in diastolic Ca²⁺. By 60 min, diastolic [Ca²⁺]_i was significantly different from values at 10 min. Mean data for cell length are shown in Fig. 8B. Myocytes exhibited a slight reduction in both diastolic length and contraction magnitude (shaded region) with time. These data show that relatively stable recordings of Ca²⁺ concentrations and cell shortening could be measured for up to 80 min in field-stimulated time controls.

View larger version (18K): [in this window] [in a new window]   Fig. 7.   Ca2+ transients and contractions in time controls. Field-stimulated time controls exhibited only slight reductions in Ca2+ transients (A) and contractions (B) with time. Diastolic [Ca2+]i appeared to increase slightly during these experiments. There was no visible deterioration in the quality of recordings during the duration of the experiment.

View larger version (21K): [in this window] [in a new window]   Fig. 8.   Ca2+ transients and contractions were relatively unchanged during time control experiments. Mean measurements of [Ca2+]i (A) show that during 80 min of recording, myocytes exhibited a small decrease in the magnitude of Ca2+ transients (shaded region) and a gradual increase in diastolic [Ca2+]i. B: magnitude of contractions (shaded region) was also slightly decreased with time, although mean diastolic length was relatively unchanged. * Significant difference from mean preischemic values, P < 0.05; [Ca2+]i measurements, n = 5 myocytes from 4 hearts; length measurements, n = 12 myocytes from 10 hearts.

View larger version (18K): [in this window] [in a new window]   Fig. 9.   Ca2+ transients and contractions in field-stimulated myocytes exposed to simulated ischemia and reperfusion. Representative recordings of Ca2+ transients and contractions are shown in A and B, respectively. During ischemia, diastolic [Ca2+]i increased, although the magnitude of Ca2+ transients was unchanged from control levels. Contractions were depressed during ischemia. In early reperfusion, myocytes exhibited recovery and overshoot of Ca2+ transients. Contraction also recovered in early reperfusion, and myocytes exhibited a marked reduction in diastolic cell length. Aftercontractions were occasionally observed in early reperfusion (arrow). In late reperfusion, Ca2+ transients were only slightly reduced from preischemic levels, whereas contractions were markedly depressed. Diastolic cell length remained below control levels in late reperfusion.

View larger version (24K): [in this window] [in a new window]   Fig. 10.   Stunning was not associated with obvious changes in Ca2+ transients. A: mean measurements of [Ca2+]i. Ischemia was associated with increased diastolic [Ca2+]i and no obvious change in the magnitude of Ca2+ transients (shaded region). In early reperfusion, mean diastolic [Ca2+]i partially recovered but increased gradually with continued reperfusion. Ca2+ transients were slightly reduced in late reperfusion. B: mean measurements of cell length. During ischemia, myocytes exhibited a marked reduction in the magnitude of cell shortening (shaded region) that was associated with a significant decrease in systolic length. With reperfusion, both diastolic and systolic cell length were decreased, and the magnitude of contractions was briefly increased. Contractions were reduced in late reperfusion, and diastolic cell length remained below control levels. * Significant difference from mean preischemic values, P < 0.05; n = 15 myocytes from 11 hearts for Ca2+ transients; n = 15 myocytes from 10 hearts for cell length.

View larger version (18K): [in this window] [in a new window]   Fig. 11.   In late reperfusion, diastolic [Ca2+]i was normal, but resting cell length was depressed. Mean measurements of diastolic [Ca2+]i and cell length are shown in A and B, respectively, for time controls and myocytes exposed to simulated ischemia and reperfusion. Data are normalized to mean preischemic values. Diastolic [Ca2+]i was significantly elevated during ischemia but was unchanged from time control values in reperfusion. Diastolic length was slightly increased during ischemia and dramatically reduced in early reperfusion. With continued reperfusion, diastolic length partially recovered and was not significantly different from time controls after 60 min. * Significant difference from time controls, P < 0.05. For the ischemia-reperfusion data, n = 15 myocytes from 11 hearts for Ca2+ transients; n = 15 myocytes from 10 hearts for cell length. For the time controls, n = 5 myocytes from 4 hearts for Ca2+ transients; n = 12 myocytes from 10 hearts for cell length.

View larger version (17K): [in this window] [in a new window]   Fig. 12.   Ca2+ transients were normal in stunning. A: measurements of Ca2+ transients as a percentage of preischemic values and compared in time controls and in cells exposed to simulated ischemia and reperfusion. B: normalized contraction measurements. During ischemia, Ca2+ transients were unchanged from time controls, but contractions were significantly depressed. In early reperfusion, myocytes exhibited a significant overshoot in the magnitudes of both Ca2+ transients and contractions. In late reperfusion, myocytes exhibited stunning as contractions were reduced to 60% of time controls. This reduction in contraction amplitude was correlated with normal Ca2+ transient magnitude. * Significant difference from time controls, P < 0.05. For the ischemia-reperfusion data, n = 15 myocytes from 11 hearts for Ca2+ transients; n = 15 myocytes from 10 hearts for cell length. For the time controls, n = 5 myocytes from 4 hearts for Ca2+ transients; n = 12 myocytes from 10 hearts for cell length.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The goals of the present study were to establish a model of stunning in isolated guinea pig ventricular myocytes and to evaluate changes in E-C coupling that contribute to stunning in this model. Our experiments demonstrate that, although action potential configuration and membrane potential recovered fully, myocytes in this model exhibited stunning. Furthermore, stunning persisted even when changes in action potentials in ischemia and early reperfusion were eliminated with voltage clamp. Contractile depression in reperfusion was evident regardless of whether contractions were initiated by CICR or the VSRM and occurred without changes in SR Ca²⁺ stores. In this model, reperfusion resulted in an initial brief rebound in cell contraction before stunning developed. This brief rebound in amplitude of contractions was accompanied by a parallel rebound in Ca²⁺ transients in early reperfusion. In contrast, stunning was not accompanied by a parallel change in Ca²⁺ transients in late reperfusion.

To investigate stunning in ventricular myocytes, we adapted a previously developed model of simulated ischemia and reperfusion (8). As shown by the time controls in the present study, this model exhibits stable Ca²⁺ transients and contractions up to 80 min of recording time. This stability allowed sufficient time for a complete cycle of ischemia and reperfusion. Our study clearly demonstrates that postischemic contractile depression occurs in this model and provides a model in which stunning can be correlated with possible changes in electrophysiology and Ca²⁺ transients. An additional advantage of this model is that contractile responses and Ca²⁺ transients can be elicited by action potentials or by voltage-clamp protocols. With the use of this model, we were able to investigate possible roles of RMP, action potential configuration, and I_Ca,L in stunning.

Ischemia and reperfusion caused marked changes in electrical activity. During ischemia, we observed decreased RMPs and abbreviation of action potentials, as reported in previous studies in single and multicellular preparations (8, 13, 23). In voltage-clamp experiments, these changes were correlated with an increase in steady-state outward current and a decline in I_Ca,L. This decrease in I_Ca,L may reflect reduction of the density of L-type Ca²⁺ channels reported by others in ligand binding studies (3, 35). Changes in resting potentials, action potential configuration, and steady-state outward current recovered in early reperfusion, and therefore likely did not play a role in stunning. In contrast, depression of I_Ca,L persisted into late reperfusion. This reduction in magnitude of I_Ca,L might be expected to contribute to stunning, because Ca²⁺ released through CICR is proportional to the magnitude of I_Ca,L (4). In addition, I_Ca,L is believed to contribute to maintenance of SR stores of Ca²⁺ (12). Thus inhibition of I_Ca,L in late reperfusion might inhibit CICR both by decreasing this trigger for Ca²⁺ release and by decreasing SR Ca²⁺ stores available for release.

Because I_Ca,L was observed to be depressed, and because previous studies have reported changes in proteins important for sequestration and release of SR Ca²⁺ in stunned tissues (21, 32, 36, 37, 39), we assessed SR Ca²⁺ stores in the present study. Interestingly, Ca²⁺ stores remained constant through the cycle of ischemia and reperfusion. Thus stunning could not be attributed to a decrease in SR Ca²⁺ stores in the present study. Furthermore, we did not observe depression of Ca²⁺ transients in field-stimulated cells in reperfusion, which indicates that stunning could not be attributed to decreased Ca²⁺ release. Normal Ca²⁺ transients also have been reported in intact stunned myocardium (7, 24, 29). These results indicate that changes in Ca²⁺-ATPase, ryanodine receptor density, or phosphorylation did not alter either SR Ca²⁺ stores or Ca²⁺ transients in the present model. Nevertheless, it is possible that compensatory changes in SR load and release may have obscured such changes (12). It also is possible that these protein structures are not normally limiting to SR function in unloaded myocytes or that changes in these proteins are not required for stunning to occur.

It is not clear how normal Ca²⁺ transients are maintained during depression of I_Ca,L in stunning. There are several possibilities. For example, 1) the gain of CICR (SR Ca²⁺ released/I_Ca,L) might be increased, and/or 2) elevation of intracellular Na⁺ levels in ischemia and reperfusion might shift the equilibrium of Na⁺/Ca²⁺ exchange and thereby reduce loss of released Ca²⁺ through the sarcolemma. The latter mechanism might increase the recirculating fraction of SR Ca²⁺. Evidence for one or both of these possibilities will require further investigation. It also is possible that CICR might be depressed in stunning, but this is compensated by continued operation of the VSRM. This possibility was tested by comparing effects of stunning on CICR and the VSRM (15). Contractions initiated separately by these two mechanisms showed parallel decreases in amplitude during ischemia and late reperfusion. Thus stunning was not mediated by selective depression of either mechanism of E-C coupling.

Our investigations with caffeine-induced contractures indicated that ischemia and late reperfusion were not accompanied by significant changes in SR Ca²⁺ stores. Interestingly, a marked rebound in the amplitudes of Ca²⁺ transients and contractions was observed in early reperfusion. Thus in early reperfusion the brief overshoot in contraction is likely caused by an increase in the magnitude of the Ca²⁺ transients. The mechanism for this rebound is not clear. Because Ca²⁺ overload is believed to be an important event in ischemia and early reperfusion (5), one may postulate that SR stores might be elevated temporarily during early reperfusion. Because it was not practical to assess SR Ca²⁺ stores with caffeine at very short intervals, we could not determine whether SR stores were altered during this period. Regardless of whether SR stores are briefly elevated, the increase in amplitude of Ca²⁺ transients must reflect enhanced Ca²⁺ release.

Our study also demonstrated that ischemia and reperfusion caused distinct changes in diastolic free Ca²⁺ and diastolic cell length. Diastolic Ca²⁺ was significantly elevated in ischemia, as reported in other models (7, 29), and recovered rapidly upon reperfusion. Although diastolic Ca²⁺ was elevated in ischemia, this was not reflected in cell length, which actually increased slightly. Experiments on perfused hearts also reported no elevation in diastolic pressure during ischemia (17). This dissociation between Ca²⁺ levels and cell length is consistent with a decrease in myofilament sensitivity to Ca²⁺ during ischemia (17, 20, 31). The presence of decreased myofilament sensitivity during ischemia was even more dramatically illustrated by marked depression of stimulated contractions, despite the presence of normal Ca²⁺ transients. Dissociation between the magnitudes of contractions and Ca²⁺ transients also was observed in reperfusion when stunning developed. Contractions remained depressed despite the persistence of Ca²⁺ transients with normal amplitudes. This observation closely resembles changes seen in intact hearts in stunning, and thus our results support the idea that contractile depression in stunned myocytes is largely a function of depressed myofilament Ca²⁺ sensitivity (17, 20, 31). Interestingly, diastolic cell length shortened greatly in early reperfusion and did not recover fully within the duration of our experiments. Corresponding changes in diastolic pressure have been reported for perfused hearts, where diastolic pressure remained elevated in early reperfusion and later in stunning (17). In the present experiments, incomplete relaxation of the myocytes was not accompanied by elevation of diastolic Ca²⁺, which returned rapidly to control levels during reperfusion. Extrinsic effects such as coronary vascular dilatation also can be excluded in this isolated cell model; however, it is possible that changes in intrinsic factors such as myofilaments might contribute. At the present time, the mechanism for this diastolic dysfunction is not clear.

A previous study (28) examined the effects of exposure of isolated rat ventricular myocytes for 90 min to acidosis, substrate deprivation, and hypoxia. Contraction also was depressed by this combination of conditions. However, unlike the present study, Ca²⁺ transients were depressed during ischemia, and no overshoot in amplitudes of contractions or transients were observed in early reperfusion. Although this study did not address stunning, it did report a prolonged decrease in amplitude of contraction with return to control conditions (28). Whether this defect in contraction is analogous to stunning in the present model is not clear.

In the present study, we utilized an isolated guinea pig ventricular myocyte model of ischemia and reperfusion, in which cells exposed to hypoxia, hypercapnia, hyperkalemia, acidosis, lactate accumulation, and substrate deprivation exhibit stunning. Our previous studies with this model have shown that myocytes exposed to these conditions show contractile and electrophysiological changes that mimic those observed in ischemic myocardium in vivo (8, 9). The present study demonstrates that this model also mimics changes in contractions and Ca²⁺ transients reported to accompany stunning in intact hearts (7, 24, 29). Thus our study provides a reproducible model that can be useful in studies of cellular changes occurring in stunning.