INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

CARDIAC MYOCYTES SURVIVING significant myocardial infarction (MI) typically undergo cellular hypertrophy. This is manifested by increases in cell length but little to no change in cell width (8, 20, 29), with proportional increases in cell surface area as determined by whole cell capacitance measurements (36, 39). Despite the compensatory hypertrophic response, recent studies indicate that excitation-contraction coupling is abnormal in surviving myocytes of the infarct heart. For example, alterations in intracellular Ca²⁺ concentration ([Ca²⁺]_i) transients during a twitch (8, 21, 35) and contraction abnormalities (8, 19-22, 37) have been observed in post-MI myocytes. By focusing on individual steps involved in excitation-contraction coupling, others (11, 36) have reported a decrease in L-type Ca²⁺ channel density, as assessed by dihydropyridine binding in post-MI myocytes. Whole cell Ca²⁺ current density, however, remained unchanged in rat myocytes 3 wk after MI (36) or, at most, decreased 33% in rabbit myocytes 8 wk after MI (20). Na⁺/Ca²⁺ exchange activity measured in sarcolemmal (SL) vesicles (12) and whole cell Na⁺/Ca²⁺ exchange current (I_NaCa) density in rat cardiac myocytes (38, 39) significantly decreased after MI, although Litwin and Bridge (20) recently reported a 32% increase in I_NaCa in rabbit myocytes 8 wk after MI. Despite a shift of myosin heavy chain isozyme distribution from fast to slow forms (27, 38), myofilament Ca²⁺ sensitivity did not appreciably change in post-MI myocytes (2, 21).

Ca²⁺ transport systems in sarcoplasmic reticulum (SR) occupy key roles in cardiac excitation-contraction coupling. According to the theory of Ca²⁺-induced Ca²⁺ release (33), influx of extracellular Ca²⁺ via L-type Ca²⁺ channels and reverse Na⁺/Ca²⁺ exchange (28) triggers SR Ca²⁺ release, with a resultant increase in [Ca²⁺]_i and myofilament activation. During diastole, most of the myoplasmic Ca²⁺ is resequestered in the SR by sarco(endo)plasmic reticulum Ca²⁺-ATPase (SERCA2), with a small amount of Ca²⁺ extruded by forward Na⁺/Ca²⁺ exchange and, to a minor extent, by SL Ca²⁺-ATPase (4, 5, 34). We previously showed significant decreases in SR Ca²⁺ content in post-MI rat myocytes (39). Decreased SR Ca²⁺ content would not only reduce the maximal amount of Ca²⁺ available for release but would also lower the "gain" (ratio of trigger Ca²⁺ to Ca²⁺ released from the SR) of SR Ca²⁺ release channels (33). In addition, in the absence of differences in I_Ca between sham and post-MI rat myocytes (36) and with no filling of the SR Ca²⁺ pool by reverse I_NaCa, since pipette Na⁺ concentration was set to zero (39), the lower SR Ca²⁺ content that we observed in post-MI myocytes was most likely due to decreased SERCA2 amounts and/or activities and/or increased SR Ca²⁺ leak. Because most of the previous studies on SERCA2 activities in diseased hearts were performed in SR vesicles (1, 15, 23-25) or tissue homogenates (31), the present study was undertaken to systemically evaluate SR function in situ in intact post-MI rat myocytes. In addition, SR Ca²⁺ leak in rat myocytes was evaluated, for the first time, by a novel electrophysiological technique that did not require assigning empirical constants for intracellular Ca²⁺ buffering (6).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal preparation. Male Sprague-Dawley rats (~300 g) were anesthetized (3% halothane-97% O₂), intubated, and ventilated. The left main coronary artery was ligated 3-5 mm distal to its origin from the ascending aorta. In survivors, typically 36 ± 3% of myocardium was infarcted according to our previous histological measurements (26). At 3 wk after MI, survivors (445 ± 12 g, n = 22) and sham-operated rats (449 ± 8 g, n = 24) were anesthetized with pentobarbital sodium (35 mg/kg body wt ip), and their hearts were excised for myocyte isolation. Our previous studies indicate that, 3 wk after MI, myocyte adaptations include cellular hypertrophy, as reflected by an ~10% increase in cell length (8, 37) and a 13-15% increase in whole cell capacitance (36, 39), altered [Ca²⁺]_i transients (8, 35), abnormal contractile behavior (8, 37), reduced dihydropyridine binding sites (36), decreased I_NaCa density and SR Ca²⁺ contents (39), slowed relaxation from caffeine-induced contracture (37), and attenuated response to -adrenergic agonists (35, 36).

Left ventricular myocyte isolation. Myocytes were isolated from the septum and left ventricular (LV) free wall, as previously described (9); the infarct scars in MI hearts were excised before the final enzymatic digestion step (8). Myocytes were allowed to adhere for 2 h to laminin-coated coverslips in 2 ml of medium 199 (pH 7.4, 95% air-5% CO₂, 37°C) before experiments (10).

[Ca²⁺]_i transient measurements in paced LV myocytes. Freshly isolated myocytes were exposed to 2 µM fura 2-AM for 15 min at 37°C. We (10) previously demonstrated that this brief fura 2-AM loading at low concentrations did not result in significant organellar compartmentation or accumulation of Ca²⁺-insensitive fura 2 forms. Fura 2-loaded myocytes mounted in a Dvorak-Stotler chamber were placed on the heated stage (37°C) of a Zeiss IM 35 microscope, superfused with medium 199 {extracellular Ca²⁺ concentration ([Ca²⁺]_o) = 1.8 or 5.0 mM}, buffered with 20 mM HEPES, and field stimulated to contract (1 Hz) between platinum wire electrodes, as previously described (8, 35, 37). Excitation light (360 and 380 nm, ±10 nm band pass; Ionoptix, Milton, MA) was directed to individual myocytes only during data acquisition to minimize photobleaching. Epifluorescence (510 ± 10 nm) collected by a Nikon ultraviolet Fluor ×100/1.3 NA oil objective was passed through a pinhole (1.6 mm) and captured by a photomultiplier (model R928-07, Hamamatsu). Approximately 8-10% of a myocyte x-y area was visible through the pinhole. Photomultiplier output was routed through an amplifier/discriminator (model C609, Thorn EMI, Middlesex, UK) before arrival at a counter/timer board (model C660, Thorn EMI) situated in a PC-compatible computer.

SERCA2 immunoblotting. Myocytes were suspended in 10 mM Tris-1 mM EDTA buffer with 0.5 mM phenylmethylsulfonyl fluoride and sonicated three times for 15 s each. Cell homogenates (100 µg/lane) in SDS sample buffer were subjected to electrophoresis in a 7.5% polyacrylamide gel. Proteins from SDS-PAGE were transferred onto Trans-Blot membranes (Bio-Rad, Hercules, CA). A monoclonal antibody (1:500) against rat SERCA2 (MA3-919, Affinity Bioreagents, Golden, CO) was used. Bound primary antibodies were detected with rabbit anti-mouse IgG, and then with ¹²⁵I-protein A (ICN, Costa Mesa, CA). The blots were subjected to autoradiography for the purpose of displaying the images. Subsequently, band signal intensities were quantitated by scanning the blots with a PhosphorImager (Molecular Dynamics, Sunnyvale, CA).

Measurement of SR Ca²⁺ leak in rat myocytes. In contrast to rabbit and feline myocytes, rat myocytes displayed the negative treppe phenomenon (17); therefore, after a period of rest, the first stimulated beat would have a larger amplitude than the steady-state beats. This "rest potentiation" effectively precludes the "rest decay" method successfully applied to investigate SR Ca²⁺ leak in hypertensive feline myocytes (3). We have developed a novel electrophysiological technique to measure SR Ca²⁺ leak in rat myocytes. The principle is based on measurement of decline in SR Ca²⁺ content at various time points after thapsigargin treatment. SR-releasable Ca²⁺ content was estimated by measuring the time integral of Na⁺-dependent transient inward current induced by caffeine exposure (38, 39). Caffeine not only releases Ca²⁺ from the SR but also impairs Ca²⁺ reuptake into the SR. The Ca²⁺ released is predominantly extruded by forward Na⁺/Ca²⁺ exchange (3 Na⁺ in/1 Ca²⁺ out) (4-6), and thus a transient inward current is generated.

Statistics. Values are means ± SE. Means were compared by paired or unpaired Student's t-tests, as appropriate. P  0.05 was taken to be statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Effects of caffeine on [Ca²⁺]_i transient decline in sham and MI myocytes. After 11 paced beats (1 Hz) to ensure steady-state SR Ca²⁺ content, application of caffeine to simultaneously release SR Ca²⁺ and inhibit SR Ca²⁺ reuptake (39) resulted in a large [Ca²⁺]_i transient that slowly declined with time (Fig. 1A). The t_1/2 of [Ca²⁺]_i decline of caffeine-induced [Ca²⁺]_i transient in sham myocytes was 2.666 ± 0.198 s (n = 14) and was significantly (P < 0.001) longer than the t_1/2 of the twitch immediately preceding the caffeine pulse (0.252 ± 0.016 s). In the presence of caffeine, lowering of the [Ca²⁺]_i transient was effected primarily through forward Na⁺/Ca²⁺ exchange, with minor contributions from mitochondrial Ca²⁺ uniporter and SL Ca²⁺-ATPase (4, 5). The >10-fold prolongation of t_1/2 of [Ca²⁺]_i decline by caffeine suggests that the early decline of the [Ca²⁺]_i transient during a twitch in rat myocytes was primarily effected by SR Ca²⁺ uptake.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Effects of caffeine on half-time (t1/2) of decline in cytosolic Ca2+ concentration ([Ca2+]i) in rat cardiac myocytes. Fura 2-loaded myocytes incubated at extracellular Ca2+ concentration ([Ca2+]o) of 5.0 mM and 37°C were paced to contract at 1 Hz. A: after 11 paced beats to ensure steady-state sarcoplasmic reticulum (SR) Ca2+ load, 5 mM caffeine was puffed onto a sham myocyte to release SR Ca2+ and inhibit SR Ca2+ reuptake. Note large caffeine-induced [Ca2+]i transient and much more gradual [Ca2+]i decline, indicating that removal of cytosolic Ca2+ by Na+/Ca2+ exchange, mitochondrial Ca2+ uniporter, and sarcolemmal Ca2+-ATPase was much slower than that by SR Ca2+-ATPase. B: same experiment performed on a myocyte after myocardial infarction (MI). Decline in t1/2 of [Ca2+]i during caffeine pulse is longer in MI than in sham myocyte. This is consistent with our previous finding that Na+/Ca2+ exchange activity is depressed in MI myocytes (39).

In situ SR Ca²⁺ uptake in sham and MI myocytes. Results depicted in Fig. 1 suggest that the initial decline of the [Ca²⁺]_i transient during a paced beat was mediated predominantly by SR Ca²⁺-ATPase. Thus the initial rate of [Ca²⁺]_i decline or t_1/2 could be taken as a measure of in situ SR Ca²⁺-ATPase function in isolated rat cardiac myocytes. Figure 2, A and C, shows paced twitches (1 Hz) of sham and MI myocytes, respectively, in medium 199 with [Ca²⁺]_o of 5.0 mM, whereas Fig. 2, B and D, shows time-expanded views of steady-state twitches from which systolic and diastolic [Ca²⁺]_i and t_1/2 of [Ca²⁺]_i decline were calculated for sham and MI myocytes, respectively. The composite results are shown in Table 1. At [Ca²⁺]_o of 1.8 mM, neither [Ca²⁺]_i nor t_1/2 of [Ca²⁺]_i decline was different between sham and MI myocytes (Table 1, top). At [Ca²⁺]_o of 5 mM, systolic [Ca²⁺]_i was significantly lower and t_1/2 of [Ca²⁺]_i decline was significantly longer in MI myocytes (Table 1, middle). Because Bers and Berlin (7) reported that t_1/2 for [Ca²⁺]_i decline decreased as the amplitude of systolic [Ca²⁺]_i increased, it may be argued that the significantly longer t_1/2 for the [Ca²⁺]_i decline in MI myocytes was due to the lower systolic [Ca²⁺]_i in MI myocytes. In our present experiments, scatterplots for t_1/2 vs. systolic [Ca²⁺]_i did not demonstrate an apparent inverse relationship in sham or MI myocytes (data not shown). In addition, when systolic [Ca²⁺]_i values were "matched" between sham and MI myocytes by arbitrary examination of all systolic [Ca²⁺]_i values between 200 and 450 nM for sham and MI myocytes (Table 1, bottom), t_1/2 of [Ca²⁺]_i decline was still significantly (P = 0.0019) longer in MI than in sham myocytes.

View larger version (13K): [in this window] [in a new window]   Fig. 2.   In situ SR Ca2+ uptake activity in sham and MI myocytes. Fura 2-loaded myocytes were paced (1 Hz) to contract at 37°C and [Ca2+]o of 5 mM. A and C: steady-state paced [Ca2+]i transients of sham and 3-wk MI myocytes, respectively. B and D: time-expanded views of steady-state [Ca2+]i transients from which t1/2 values of [Ca2+]i decline were measured for sham and MI myocytes, respectively. Note more gradual decline of [Ca2+]i and longer t1/2 in MI myocyte. Composite results are summarized in Table 1.

Effects of MI on SERCA2 abundance in cardiac myocytes. To further elucidate the mechanisms by which SR Ca²⁺-ATPase activity was reduced in myocytes isolated from 3-wk MI hearts, we performed immunoblots of isolated cardiac myocyte homogenates probed with antibody to SERCA2 (Fig. 3). We detected a band of 9-10 ×10⁴ Da apparent molecular weight, which is similar to that of the SERCA2 cloned from rat hearts (18). Significantly less (P < 0.02) SR Ca²⁺-ATPase was present in myocytes isolated from 3-wk MI hearts (95,811 ± 2,950 arbitrary units, n = 5) than in myocytes isolated from sham hearts (125,072 ± 6,466 arbitrary units, n = 5).

View larger version (94K): [in this window] [in a new window]   Fig. 3.   Immunoblots of sarco(endo)plasmic reticulum Ca2+-ATPase (SERCA2) protein. Proteins in myocyte suspension homogenates (100 µg/lane) were separated by gel electrophoresis and transferred to Trans-Blot membranes, and SERCA2 was identified by immunoblotting. Polypeptide bands of 9.7-10 × 104 Da were quantitated by scanning blots with a PhosphorImager. At 3 wk after MI, myocytes lost 23% of SERCA2 proteins.

Effects of MI on SR Ca²⁺ content. SR-releasable Ca²⁺ content was estimated by measuring the time integral of Na⁺-dependent transient inward current induced by caffeine exposure (Fig. 4, B-D, left) (39). At [Ca²⁺]_o of 1.8 mM, the I_NaCa-time integral was 83.9 ± 10.9 pC/cell or 0.870 ± 0.113 fmol/cell in MI myocytes (n = 15) and tended (P = 0.0755) to be less than that in sham myocytes (119.6 ± 15.7 pC/cell or 1.240 ± 0.163 fmol/cell, n = 16). These results are similar to our previous report that the normalized SR Ca²⁺ content per unit cell surface area (fmol/pF), but not SR Ca²⁺ content per cell (fmol/cell), was significantly (P < 0.02) lower in MI myocytes (39).

View larger version (17K): [in this window] [in a new window]   Fig. 4.   Electrophysiological measurement of SR Ca2+ leak in rat myocytes. Myocytes were incubated at [Ca2+]o of 1.8 mM and 29°C and voltage clamped at 70 mV. A: 200 ms after 10 conditioning pulses (70 to 0 mV, 300 ms, 1 Hz), 5 mM caffeine was puffed onto myocyte for 2.4 s to release SR Ca2+ and inhibit SR Ca2+ reuptake. This resulted in a large inward current (B-D, left) carried by Na+ due to forward Na+/Ca2+ exchange (3 Na+ in/1 Ca2+ out). Time integral of this inward current (INaCa) provided an estimate of steady-state SR Ca2+ content of that myocyte (39). After caffeine-induced SR Ca2+ release was complete, medium was changed within 60 s to remove caffeine. A 2nd train of 10 conditioning pulses was applied to reload SR with Ca2+. Thapsigargin (1 µM) was then added to inhibit SR Ca2+-ATPase, and various time intervals (0-5 min) were allowed to elapse before application of 2nd caffeine pulse to empty SR of remaining Ca2+ (A). B: right, 2nd caffeine-induced INaCa-time integral without thapsigargin and time delay. After medium change to remove caffeine of 1st pulse, 2nd train of 10 conditioning pulses was effective in restoring 90% of initial SR Ca2+ stores (compare INaCa-time integrals in B, left and right). With progressive time lapse after thapsigargin treatment, INaCa-time integrals became increasingly smaller [60 s (C, right) and 300 s (D, right)] than their initial values (C and D, left).

SR Ca²⁺ leak in sham and MI myocytes. One plausible mechanism to account for the observed reduction in SR Ca²⁺ content in MI myocytes (39) was increased SR Ca²⁺ leak. Figure 4A illustrates our approach to measure SR Ca²⁺ leak in rat cardiac myocytes. We first established that, after discharge of SR Ca²⁺ by caffeine, the SR could be reloaded to >90% of its initial Ca²⁺ value by caffeine removal and 10 additional conditioning pulses (Fig. 4B; see METHODS). This allowed us to establish a "control" value for SR Ca²⁺ content for each individual myocyte studied. Addition of thapsigargin to inhibit SR Ca²⁺ reuptake resulted in progressive loss of SR Ca²⁺ (Fig. 4, C and D) due to SR Ca²⁺ leak. Figure 5 shows that SR Ca²⁺ leak rate was not different between sham and MI myocytes. The t_1/2 of SR Ca²⁺ leak, as determined from the slope of the linear regression-fit line, was 241 s in sham and 230 s in MI myocytes. With the assumption of monoexponential SR Ca²⁺ loss, the time constant () of SR Ca²⁺ leak was 290 s in sham and 268 s in MI myocytes.

View larger version (10K): [in this window] [in a new window]   Fig. 5.   SR Ca2+ leak is similar between sham and MI myocytes. SR Ca2+ remaining at time intervals after thapsigargin treatment (Fig. 4, B-D, right) is plotted as percentage of control SR Ca2+ content for that same myocyte (Fig. 4, B-D, left). Values are means ± SE of 3-5 myocytes. , Sham myocytes; , 3-wk MI myocytes. Dashed and solid curves, monoexponential fit for sham and 3-wk MI myocytes, respectively. For sham myocytes, percent SR Ca2+ remaining at time t was 100.23et/289.6, where t is time in seconds after thapsigargin (r = 0.9653). For MI myocytes, percent SR Ca2+ remaining at time t was 101.07et/268.0 (r = 0.9707). If data are fitted as straight lines by linear regression, then for sham myocytes, percent SR Ca2+ remaining at time t was 93.17  0.1979t (t1/2 = 241 s, r2 = 0.9931). For MI myocytes, percent SR Ca2+ remaining at time t was 93.22  0.2082t (t1/2 = 230 s, r2 = 0.9945). For clarity of data presentation, only monoexponential fits are shown.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Previous studies on cardiac SR function (1, 15, 18, 24, 25, 31, 32) have predominantly been performed with SR membrane vesicles. One major goal of our present study is to devise an estimate of in situ SR Ca²⁺ homeostatic function. Results from experiments depicted in Fig. 1 indicate that the early phase of [Ca²⁺]_i transient decline was largely mediated by SR Ca²⁺ uptake in sham and post-MI rat cardiac myocytes. Our conclusions are similar to those of Bassani et al. (4), in that in rat cardiac myocytes, inhibition of forward Na⁺/Ca²⁺ exchange resulted in a statistically insignificant increase in  of the [Ca²⁺]_i transient decline (20%) (4). On the other hand, inhibition of SR Ca²⁺ uptake by caffeine was associated with a significant prolongation of  (8-fold) (4) or t_1/2 (10-fold; present study) of the [Ca²⁺]_i transient decline. Even in rabbit ventricular myocytes in which Na⁺/Ca²⁺ exchange played a more dominant role in [Ca²⁺]_i transient decline than in rat myocytes (4), recent studies indicated that forward Na⁺/Ca²⁺ exchange produced only 13% of the first half of the decline in [Ca²⁺]_i (34). Contributions by mitochondrial Ca²⁺ uniporter and SL Ca²⁺-ATPase to the early decline of the [Ca²⁺]_i transient were also very minor (5). The observations by Bassani et al. (4, 5) and Yao et al. (34), as well as our present results, indicate that the early decline of the [Ca²⁺]_i transient during a twitch in rat cardiac myocytes was primarily mediated by SR Ca²⁺-ATPase and that the t_1/2 of [Ca²⁺]_i decline may be taken as an in situ measure of SR Ca²⁺ uptake.

At [Ca²⁺]_o of 1.8 mM, systolic [Ca²⁺]_i, diastolic [Ca²⁺]_i, and t_1/2 of [Ca²⁺]_i decline were similar between sham and MI myocytes (Table 1, top). This result is consistent with our previous observations (8) that neither the [Ca²⁺]_i transient (amplitude and time course) nor cell shortening (amplitude and dynamics) was different between sham and 3-wk MI myocytes incubated at [Ca²⁺]_o of 1.1-1.9 mM and paced at 0.2 Hz. In a more recent study (37), we also did not detect any significant differences in twitch amplitudes between sham and 3-wk MI myocytes incubated at [Ca²⁺]_o of 1.8 mM and paced at 0.1-5 Hz. In rabbit myocytes isolated 8 wk after MI, Litwin and Bridge (20) did not observe differences in cell shortening amplitude, shortening, and relaxation velocities at [Ca²⁺]_o of 2.7 mM and pacing at 1 Hz. These observations suggest that, in single myocytes isolated from MI hearts and studied in vitro, functional consequences of altered [Ca²⁺]_i homeostatic mechanisms may not be manifest under "normal" [Ca²⁺]_o conditions. It should be emphasized that, in our rat MI model and the rabbit MI model of Litwin and Bridge, overt LV failure has not developed at the time of myocyte isolation. This is because there were no differences in heart rate and cardiac output between sham and MI rats (26) and in heart rate and LV systolic pressure between control and MI rabbits (20). The relative preservation of LV function at the time the animals were killed may relate to the "moderate" LV infarct sizes: 36 ± 3% in our rat MI model (26) and 25% in the rabbit MI model (20). By contrast, in rats that have suffered "large" myocardial infarction and developed LV failure, as evidenced by elevated heart rate and decreased cardiac output, LV myocytes isolated 1 wk after MI demonstrated a significant reduction in the extent of cell shortening and velocities of myocyte shortening and relengthening at [Ca²⁺]_o of 1.2 mM and pacing at 1 Hz (22).

Altered [Ca²⁺]_i homeostastic mechanisms and their functional consequences could be unmasked at higher [Ca²⁺]_o in our rat MI model. At [Ca²⁺]_o of 5.0 mM, we previously demonstrated lower systolic [Ca²⁺]_i and decreased extent of cell shortening in MI myocytes (8, 37). Our present results confirm that, at [Ca²⁺]_o of 5.0 mM, systolic [Ca²⁺]_i was significantly lower in MI myocytes (Table 1, middle). Most germane to our present discussion is the finding that t_1/2 of [Ca²⁺]_i decline was significantly prolonged in MI myocytes (Table 1, middle). It may be argued that the longer t_1/2 of [Ca²⁺]_i decline could be explained by the lower systolic [Ca²⁺]_i in MI myocytes incubated at [Ca²⁺]_o of 5 mM (7). However, it should be pointed out that, when comparisons of t_1/2 values of [Ca²⁺]_i decline were restricted to [Ca²⁺]_i transients of similar peak and amplitude (Table 1, bottom), t_1/2 of [Ca²⁺]_i decline was still significantly longer in MI myocytes. Our data strongly suggest that SR Ca²⁺ uptake in situ was less in MI than in sham myocytes exposed to high [Ca²⁺]_o. Our finding that 3-wk MI myocytes exhibited abnormal Ca²⁺ homeostasis (lower systolic [Ca²⁺]_i and prolonged t_1/2 of [Ca²⁺]_i decline) and contractile responses (8, 37) at high [Ca²⁺]_o has functional significance when viewed in the context that, in vivo, myocytes are normally influenced by Ca²⁺-mediated hormonal and neural factors, especially in failing hearts. Thus the altered Ca²⁺ homeostasis unmasked at high [Ca²⁺]_o in the absence of neurohumoral influences may suggest that myocytes will also respond abnormally in vivo to such adrenergic and nervous stimuli.

Decreased SR Ca²⁺ uptake activity was correlated with a 23% decrease in SERCA2 expression in MI hearts (Fig. 3, see RESULTS). Our observations that SR Ca²⁺ uptake activity and SERCA2 expression were decreased in MI myocytes are consistent with depressed ATP-dependent Ca²⁺ uptake activity (decreased maximal velocity but no change in affinity for Ca²⁺) in SR membrane fractions isolated from rat LV 4, 8, and 16 wk after MI (1). In dogs in which chronic LV failure was produced by multiple sequential intracoronary embolization with polystyrene latex microspheres that led to progressive loss of viable myocardium, SR membrane fractions prepared from failing LV also demonstrated a significant decrease in maximal velocity but no change in the affinity of Ca²⁺-ATPase activity (15). By contrast, in failing human LV secondary to idiopathic dilated cardiomyopathy, despite reduced SR Ca²⁺-ATPase mRNA levels (23), protein levels of SERCA2 and SR Ca²⁺ uptake activity measured in isolated SR membrane fractions were unchanged (24, 25, 32). In a more recent study in which tissue homogenates were obtained from idiopathic dilated cardiomyopathic and ischemic cardiomyopathic patients, SR Ca²⁺-ATPase activity was depressed in cardiomyopathic but not normal patients (31). These conflicting observations may indicate fundamental differences in SR Ca²⁺-ATPase expression and function in different disease processes (e.g., ischemic vs. idiopathic dilated cardiomyopathy) or species-dependent alterations.

Previous studies to measure SR Ca²⁺ leak in indo 1-loaded rat myocytes (6) involved pretreating in zero extracellular Na⁺ and [Ca²⁺]_o medium containing thapsigargin, applying caffeine at various intervals, and monitoring decay of the caffeine-induced indo 1 fluorescence transient. To ensure that SR Ca²⁺ content did not appreciably decline during development of thapsigargin inhibition of SR Ca²⁺-ATPase, myocytes were exposed to medium with zero extracellular Na⁺ and [Ca²⁺]_o of zero for 3-10 min (6). Exposure to medium with [Ca²⁺]_o of zero has been associated with increased ionic membrane permeabilities, lower high-energy phosphate levels, and depressed metabolism in rat cardiac myocytes (9). Because we wished to avoid the deleterious effects of [Ca²⁺]_o removal on cardiac myocytes, we adopted the protocol depicted in Fig. 4A to measure SR Ca²⁺ leak. The major potential error with our protocol is that SR Ca²⁺-ATPase inhibition by thapsigargin was not instantaneous but may require up to 2 min after thapsigargin exposure for complete blockade to develop (6). Two observations suggest that the error introduced would be small. First, the time course of SR Ca²⁺ leak (Fig. 5) was not biphasic, as would be expected for a significant "lag" due to slow development of SR Ca²⁺-ATPase blockade, but was rather nicely fitted by a straight line or monoexponential. This suggests that the SR Ca²⁺ leak rate was significantly slower than the rate of development of SR Ca²⁺-ATPase inhibition by thapsigargin. Second, our t_1/2 for SR Ca²⁺ leak (240 s) agreed very well with that reported for rat myocytes (t_1/2 = 246 s) by Bassani and Bers (6), who measured SR Ca²⁺ leak after SR Ca²⁺-ATPase blockade by thapsigargin was complete. The excellent agreement between our SR Ca²⁺ leak rates and those reported by Bassani and Bers again suggests that error introduced by the assumption of instantaneous inhibition of SR Ca²⁺-ATPase by thapsigargin would likely be small.

With our SR-releasable Ca²⁺ content of 120 pC/cell or 1.24 fmol/cell in sham myocytes (see RESULTS) and average rat myocyte volume of 25 pl (13), the instantaneous [Ca²⁺]_i attained after caffeine-induced SR Ca²⁺ release (Fig. 4) would be 76 µmol/l nonmitochondrial cell water [with the assumption that 65% of cell volume is nonmitochondrial (30)]. This value is similar to 114 µmol/l nonmitochondrial cell water estimated for rat myocytes by Bassani and Bers (6) on the basis of indo 1 fluorescence transient measurements, various assumptions for empirical constants for intracellular Ca²⁺ buffering, and "estimated" intracellular indo 1 concentrations. With the assumption that SR Ca²⁺ leak is monoexponential, with  of 290 s (Fig. 5), SR Ca²⁺ leak rate can be estimated to be 0.24 µM/s, which is similar to 0.32 µM/s reported for rat myocytes (6). Our value for Ca²⁺ SR leak rate is also comparable to that estimated from elementary Ca²⁺ release events (Ca²⁺ sparks), i.e., 0.2-0.8 µM/s (6). Unlike the method utilizing indo 1 fluorescence, our method of estimating SR Ca²⁺ leak does not require values assigned to empirical constants for cellular Ca²⁺ buffering or estimates for Ca²⁺ indicator concentrations (generally unknown in single myocytes) and its affinity for Ca²⁺. This offers a significant advantage, since the effects of MI on the empirical constants for intracellular Ca²⁺ buffering are unknown.

We previously demonstrated that SR Ca²⁺ content was significantly lower in rat MI myocytes (39). Our first major finding was that SR Ca²⁺ uptake activity and SERCA2 expression were reduced in MI myocytes. Our second major finding was that SR Ca²⁺ leak rates were not different between sham and MI myocytes (Fig. 5, see RESULTS). Our results indicate that, independent of effects on SR Ca²⁺ filling due to decreased Ca²⁺ influx, decreases in SR Ca²⁺ uptake, but not increased SR Ca²⁺ leak, contributed to the lower SR Ca²⁺ content in MI myocytes.

In summary, we have demonstrated that the early decline of [Ca²⁺]_i transient in sham and 3-wk MI rat myocytes was mediated primarily by SR Ca²⁺-ATPase. Using t_1/2 of the [Ca²⁺]_i transient decline as an in situ measure of SR Ca²⁺ uptake activity, we found a 19% decrease in SR Ca²⁺ uptake in 3-wk MI myocytes. This was associated with a 23% decrease in SERCA2 expression. Finally, SR Ca²⁺ leak was not different between sham and MI myocytes. This suggests that, independent of decreased SR filling by Ca²⁺ influx, the lower SR Ca²⁺ content in MI myocytes was due to subnormal SR Ca²⁺ reuptake, but not to enhanced SR Ca²⁺ leak.