INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

UPREGULATION OF NA⁺/ca²⁺ exchange in the failing heart may be a compensatory adaptive mechanism to promote influx and efflux of cytostolic Ca²⁺, despite failure-associated defects in sarcoplasmic reticulum (SR) intracellular calcium (Ca_i²⁺) transport (10, 39, 43). Under basal conditions, Ca_i²⁺ homeostasis is maintained primarily via release and uptake by the SR and secondarily by flux across the sarcolemma dominated by Na⁺/Ca²⁺ exchange. The predominant mode of the Na⁺/Ca²⁺ exchanger supplements the SR Ca²⁺- ATPase to decrease cell Ca²⁺ and facilitate cardiac relaxation; whether the Na⁺/Ca²⁺ exchanger may provide a source of Ca²⁺ to trigger SR calcium release and support the inotropic capacity of the heart (26, 28, 47) remains controversial. Several studies (5, 17, 34, 39, 43) have demonstrated downregulation of SR function in the failing heart that may, in part, contribute to alterations in Ca_i²⁺, reduced tension development, and a blunted force-frequency relation in heart failure (8, 12, 18, 40). Gene expression of the cardiac Na⁺/Ca²⁺ exchanger, however, is enhanced in failing human hearts (10, 43), perhaps as a compensatory adaptation to assist impaired SR Ca_i²⁺ resequesteration during relaxation (10, 39, 43) or to boost inotropy by increasing Ca_i²⁺ influx during the action potential (10, 27, 31, 43). Data from a recent study in dogs (39) indicate that a greater fraction of Ca_i²⁺ removal is attributable to Na⁺/Ca²⁺ exchange than to SR uptake in failing myocytes. Na⁺/Ca²⁺ exchange current density and contraction amplitude were significantly increased in cells from infarcted rabbit hearts (27), suggesting that increased activity of the Na⁺/Ca²⁺ exchanger may promote Ca_i²⁺ entry and enhance SR Ca_i²⁺ loading and release in the damaged heart. Bers et al. (7) demonstrated that, in the absence of a functional SR, Na⁺/Ca²⁺ exchange might promote Ca_i²⁺ entry to activate contractility if intracellular sodium is sufficiently elevated. Because of concomitant alterations in genetic expression of other Ca²⁺ handling proteins (11, 16, 35, 39, 43), the functional significance of enhanced gene expression of the Na⁺/Ca²⁺ exchanger in mediating alterations in Ca_i²⁺ and myocardial function in the failing heart remains speculative. The availability of transgenic mice engineered to have specific enhanced gene expression of the cardiac Na⁺/Ca²⁺ exchanger (1) provides an opportunity to investigate the effects of increased abundance of the exchanger on Ca_i²⁺ release and uptake (45) in the absence of other pathological genetic perturbations. We recently developed a technique for simultaneously assessing Ca_i²⁺ and left ventricular (LV) function in the isovolumically contracting mouse heart to describe alterations in Ca_i²⁺ with alterations in SR protein expression (15) and changes in Ca_i²⁺ and myocardial function during myocardial ischemia (14). Accordingly, we applied this approach to test the hypothesis that transgenic overexpression of the Na⁺/Ca²⁺ exchanger may play a unique role in maintaining Ca_i²⁺ exchange and contractility during pathologies that are known to effect SR Ca_i²⁺ transport and excitation-contraction (E-C) coupling (2, 13, 21, 22, 29, 30, 32). We found that hearts that overexpress the Na⁺/Ca²⁺ exchanger are significantly better able to maintain Ca_i²⁺ homeostasis while boosting myocardial function during ischemia and hypoxia. These results point to the functional relevance of upregulation of Na⁺/Ca²⁺ exchange in pathological settings where SR calcium transport mechanisms are impaired.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Production of transgenic mice. Heterozygous transgenic mice were generated as previously described (1). Adult male and female animals were used in the present study. The expression levels of Na⁺/Ca²⁺ exchanger in transgenic hearts was shown to be ~200% of controls as measured by 45 Ca²⁺ uptake (1). We did not observe any grossly apparent altered phenotype in transgenic animals; the heart and body weights were not different between transgenic and wild-type animals, as has been previously reported (1).

Isolated heart perfusion. Hearts from equal numbers of male and female wild-type and transgenic animals were perfused in vitro as previously described (14, 15). Briefly, the aorta was slipped over a 20-gauge blunt-tipped stainless steel needle through which oxygenated (95% O₂-5% CO₂) buffer [containing (in mmol/l) 118.0 NaCl, 4.7 KCl, 1.2 KH₂PO₄, 1.5 CaCl₂, 1.2 MgCl₂, 23.0 NaHCO₃, 10.0 dextrose, and 0.3 EDTA; pH 7.4] was pumped at a rate of ~3 ml/min. An intraventricular balloon-catheter system specially designed for the mouse heart (14) was passed through the mitral annulus until seated in the LV, and the distal end of the balloon catheter was connected to a Statham P23B pressure transducer (Gould, Cleveland, OH) to record functional parameters. All in vitro studies were conducted at 30°C, and hearts were paced at ~6 Hz to minimize consumption of aequorin.

Measurement of Ca_i²⁺. Ca_i²⁺ activity was monitored using aequorin, as previously described (14, 15). Briefly, after modifying the perfusate to contain 0.5 mmol/l CaCl₂, 0.6 mmol/l MgCl₂, and 20 mmol/l dextrose, we injected 1-3 µl of aequorin with a glass micropipette within 90 s into a localized region of 2 mm² at the apex of the heart. The heart was positioned in an organ bath such that the aequorin-loaded region was ~2 mm from the bottom of the bath. The Ca²⁺ and Mg²⁺ concentrations of the perfusate were restored to 2.5 mmol/l Ca²⁺ and 1.2 mmol/l Mg²⁺ in a step-wise fashion over a period of 40 min. The entire isolated heart preparation was positioned in a light-tight box for collection of the aequorin light signal, as described previously (14, 15). Aequorin luminescence was detected by a photomultiplier tube and recorded as anodal current. For estimation of Ca_i²⁺, Triton X-100 was injected into the coronary perfusate to quickly permeabilize the myocardial cell membranes and expose the remaining active aequorin to saturating Ca²⁺. This procedure results in a burst of light, the integral of which approximates the maximum light (L_max), against which light signals of interest (L) provided the fractional luminescence (L/L_max). L/L_max was referred to a calibration equation to estimate Ca_i²⁺ (21).

Signal recording. After light was passed through a current discriminator, the light signal was filtered through an analog low-pass window (6 dB at 25 Hz; 8-pole Bessel, UAF-41, Burr Brown, Tucson, AZ). The light signal and LV isovolumic pressure were then simultaneously recorded on a strip recorder (model 35-V704-10, Gould) and a digital oscilloscope (model 4094A, Nicolet Instruments, Madison, WI). Digitized LV pressure and light signals were stored every 0.2 ms on a disk recorder (model XF-44, Nicolet) connected to the oscilloscope. LV pressure recordings were analyzed with regard to peak systolic pressure, end-diastolic pressure, and peak rates of pressure development and decline. The time constant of pressure decay, _P, was calculated using a nonlinear least-squares technique to solve the equation LVP(t) = (LVP_o  LVP_a)e^t/_P + LVP_a (36), where LVP_o is the measured LV pressure (LVP) corresponding to the peak negative pressure derivative with respect to time (dP/dt_min), t is time, and LVP_a is the pressure 10% above the measured minimum diastolic pressure. Aequorin light signals, which were converted to Ca_i²⁺ estimates, were analyzed with regard to diastolic and peak systolic Ca_i²⁺. The time constant of the decline of Ca_i²⁺ signal, _L90  10% (where 90 and 10 are the percent light signal), was calculated with a modified method of Bers and Berlin (6, 50). Briefly, the digitized data from 90 to 10% of the peak in the decline phase of the aequorin light signal were fitted with the formula Ca_i²⁺ = (C_o  C_a)e^t/_L9010% + C_a (50). We defined C_a as the minimum diastolic light and C_o as the initial light at the beginning of the decline. Values of _L90  10% were calculated using a nonlinear least-squares technique. Time constants of decline from 90 to 50% (_L90  50%) and from 50 to 10% (_L50  10%) of the Ca_i²⁺ signal were also calculated.

In vitro ischemia. After a 15-min equilibrium period, we recorded baseline conditions in 10 transgenic and 10 wild-type hearts. The hearts were then subjected to no-flow global ischemia by the abrupt termination of power to the pump. The organ bath was evacuated of oxygenated solution and refilled with nitrogen saturated-perfusate maintained at 30°C. Pacing at 6 Hz was maintained during ischemia. During ischemia, the aequorin light signals were time averaged each individual minute interval and for the first 5-min interval from the onset of ischemia.

In vitro hypoxia. After a 15-min equilibrium period, we recorded baseline conditions in 10 transgenic and 10 wild-type hearts. Hypoxia was induced by an abrupt change of buffer solution from one bubbled with a gaseous mixture of 95% O₂-5% CO₂ to one bubbled with 95% N₂-5% CO₂ (PO₂ < 15 mmHg; see Ref. 30). No other conditions were altered, including the temperature and the pH, during this exchange. After 10 min of hypoxia, we recorded the functional and Ca_i²⁺ indexes.

Cyanide perfusion. After a 15-min equilibrium period, we recorded baseline conditions in five transgenic and five wild-type hearts. Cyanide inhibition of oxidative phosphorylation was accomplished by changing the perfusate to one containing 0.5 mmol/l KCN (20) and 4.2 mmol/l K⁺. The buffer was not bubbled, and the temperature and buffer pH were maintained. After 10 min of perfusion with cyanide-containing perfusate, we recorded the functional indexes.

Statistics. Data are represented as the means ± SE. Comparisons between transgenic and wild-type hearts were performed using Student's t-test for unpaired observations. Comparisons between baseline and ischemia, hypoxia, or perfusion with cyanide were performed using Student's t-test for paired observations.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Baseline measurements of Ca_i²⁺ and functional parameters of isolated hearts from wild-type and transgenic mice are summarized in Table 1. Peak and diastolic Ca_i²⁺ were similar in the two groups of hearts. Differences in the decline of the Ca_i²⁺ transient were apparent when we compared _L90  50% (transgenic 29 ± 4 ms and wild type 44 ± 4 ms; P < 0.05), suggesting an enhanced effectiveness of Ca_i²⁺ extrusion by Na⁺/Ca²⁺ exchange at baseline (50). There was a trend for increased peak LV systolic (LVSP) and end-diastolic pressures (LVEDP) as well as an increased peak positive pressure derivative with respect to time (dP/dt_max) and dP/dt_min, although these differences were not significant. Ventricular relaxation, as reflected by _P, was faster (P < 0.05) in transgenic hearts (13 ± 1 ms) than in wild-type hearts (17 ± 1 ms).

Effects of ischemia. When coronary flow was interrupted, there was an abrupt fall in LV pressure in all hearts. During the first minute of ischemia, the decline in LVSP was significantly attenuated in transgenic hearts, whereas the decline in LVEDP was comparable between groups. Developed pressure fell to zero by 3 min in both groups of hearts. However, whereas LVSP was depressed to 25 ± 2% of baseline in wild-type hearts after 1 min of ischemia, transgenic hearts maintained 40 ± 3% of their ability to generate pressure through 1 min of no-flow ischemia (n = 10, P < 0.01). Figure 1 illustrates strip-chart recordings of LV pressure in a wild-type and transgenic heart during the early moments of no-flow ischemia. Superimposing the profiles of the decline in function illustrates the relative preservation of function in the transgenic hearts during early no-flow ischemia.

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Strip-chart recordings of time course of ischemic contractile failure. A: wild-type heart; B: transgenic heart. C: declines in left ventricular (LV) pressure are superimposed, illustrating the delayed decline in LV pressure in the transgenic hearts during no-flow ischemia.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   A: amplitude (difference between peak and minimum) of the aequorin light signal during no-flow ischemia in wild-type and transgenic hearts (n = 10 hearts in each group). The amplitudes represent the time averages of amplitudes recorded for 1-min intervals subsequent to interruption of perfusion of the heart, and the percentage of baseline is the comparison to the amplitude of the signal recorded just before interruption of perfusion of the heart. Symbols and error bars represent mean values ± SE. The amplitude of the light signal in wild-type compared with transgenic hearts was significantly depressed through 5 min of ischemia. *P < 0.05. After 5 min of ischemia, the amplitudes of the light signals were not quantitatively different between groups. B: aequorin light signals continuously recorded and averaged by the oscilloscope for 5 min in a wild-type and transgenic heart during no-flow ischemia. Whereas light signals remained distinct throughout several minutes of ischemia in the transgenic heart, they degenerated rapidly in the wild-type heart.

Effects of hypoxia. The perfusate and organ bath were switched to buffer bubbled with 95% N₂-5% CO₂ for 10 min. Figure 3 summarizes the percent decline from baseline of functional parameters after 10 min of hypoxia in wild-type hearts and hearts that overexpress the Na⁺/Ca²⁺ exchanger. In wild-type hearts, functional indexes were significantly decreased during hypoxia compared with baseline, resulting in nearly a one-third reduction in the ability of the hearts to generate pressure. In transgenic hearts, however, the functional impairment during hypoxia was less pronounced. For example, peak LVSP decreased by barely 10% during hypoxia in transgenic hearts, whereas in wild-type hearts the reduction exceeded 28% (P < 0.001). Ca_i²⁺ was also significantly less affected by hypoxia in transgenic hearts. Figure 3B summarizes the percent changes from prehypoxia conditions of the Ca_i²⁺ transient after 10 min of hypoxia. In transgenic hearts, neither peak Ca_i²⁺ (0.81 ± 0.04 µM) nor diastolic Ca_i²⁺ (0.35 ± 0.05 µM) changed significantly during hypoxia. In wild-type hearts, however, peak Ca_i²⁺ decreased from 0.82 ± 0.02 µM at baseline to 0.70 ± 0.02 µM (P < 0.01), and diastolic Ca_i²⁺ decreased from 0.35 ± 0.05 to 0.25 ± 0.04 µM (P < 0.01) after 10 min of hypoxia. Moreover, the time course of the decline of the Ca_i²⁺ transients tended to accelerate during hypoxia in transgenic hearts. In wild-type hearts, however, the decline of the Ca_i²⁺ signal was prolonged. Whereas _L90  10% was prolonged to 127 ± 13% of baseline in wild-type hearts, it was abbreviated to 83 ± 5% of baseline in transgenic hearts (P < 0.05) during hypoxia (Fig. 3). The most prominent difference was in the terminal phase of the decline of the Ca_i²⁺ transient, reflected by _50  10, which was significantly prolonged to 39 ± 8 ms in wild-type but shortened to 21 ± 2 ms in transgenic hearts (P < 0.05) during hypoxia. Figure 4 illustrates simultaneous recordings of aequorin light signals and LV pressure in a wild-type and transgenic heart at baseline and after 10 min of hypoxia. In the transgenic heart, the peak of the aequorin light signal, representing peak Ca_i²⁺, did not change significantly, and the decline of the transient was more rapid during hypoxia than at baseline. LV pressure did not change significantly. In the wild-type heart, the peak of the aequorin signal was significantly decreased, and the decline of the transient was slightly prolonged during hypoxia. The relative preservation of function with concomitant preservation of the Ca_i²⁺ signal in transgenic hearts compared with wild-type hearts suggest enhanced effectiveness of both Ca²⁺ influx and efflux by the increased number of Na⁺/Ca²⁺ exchangers, which become more relevant under conditions of hypoxic stress. In contrast, in wild-type hearts, functional and Ca_i²⁺ indexes are depressed during hypoxia.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   A: percent decline from baseline of functional parameters 10 min after imposing myocardial hypoxia. LV developed pressure (LVP), contractility [peak positive pressure derivative with respect to time (dP/dtmax)], and relaxation [peak negative pressure derivative with respect to time (dP/dtmin)] were significantly better maintained in transgenic hearts during hypoxia. Means ± SE from 10 hearts in each group are shown. B: percent change from baseline of peak intracellular Ca2+ (Cai2+), diastolic Cai2+, and the time constant of decline of the Cai2+ light signal from 90 to 10% (L90  10%) 10 min after imposing myocardial hypoxia. Neither peak nor diastolic Cai2+ changed significantly during hypoxia in transgenic hearts, whereas both peak and diastolic Cai2+ significantly decreased in wild-type hearts. L90  10% was relatively briefer during hypoxia in transgenic hearts but was significantly prolonged in wild-type hearts, most prominently during the terminal phase. Means ± SE from 8 hearts in each group are shown.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Aequorin light signals (top) and isovolumic LVP waveforms (bottom) from a wild-type (left) and transgenic heart (right) at baseline conditions (solid lines) and after 10 min of perfusion with 95% N2-5% CO2 (dashed lines). In the wild-type heart, hypoxia resulted in a decrease in the Cai2+ signal and a corresponding decrease in LVP. In the transgenic heart, neither Cai2+ nor LVP was reduced by hypoxia. Moreover, the decline of Cai2+ was accelerated during hypoxia in transgenic hearts.

Effects of cyanide perfusion. The perfusate and organ bath were switched to buffer containing cyanide, which inhibits oxidative phosphorylation in both the surface and core cells (30). In wild-type and transgenic hearts, cyanide perfusion depressed LV function to a greater extent than during hypoxia. Transgenic hearts, however, generated significantly higher systolic pressure than wild-type hearts (36 ± 3 vs. 22 ± 8 mmHg, P < 0.05) after 10 min of perfusion with cyanide. In wild-type hearts, peak LVSP decreased by 64 ± 5%, whereas in transgenic hearts, peak LVSP decreased by 22 ± 4% (P < 0.05). LVEDP was higher (P = 0.07) in transgenic hearts (32 ± 7 mmHg) than in wild-type hearts (15 ± 2 mmHg) after 10 min of cyanide perfusion. In three transgenic hearts, pressure alternans was observed during cyanide perfusion, whereas in the wild-type hearts, pressure alternans was not apparent.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Our results demonstrate that enhanced gene expression of the Na⁺/Ca²⁺ exchanger preserves Ca_i²⁺ homeostasis under conditions of ischemic and hypoxic stress, thereby attenuating contractile dysfunction. These findings, in transgenic mouse hearts engineered to have increased abundance of the Na⁺/Ca²⁺ exchanger, point to the functional relevance of upregulation of the Na⁺/Ca²⁺ exchanger in pathological settings where SR calcium exchange mechanisms are impaired. The present findings indicate that hearts with upregulation of the Na⁺/Ca²⁺ exchanger demonstrate preserved peak systolic Ca_i²⁺ during ischemia and hypoxia and more rapid removal of Ca_i²⁺ during hypoxia than wild-type hearts. The inotropic and lusitropic properties of the myocardium were better preserved during ischemia and hypoxia in hearts with increased expression of the Na⁺/Ca²⁺ exchanger. The benefits of enhanced gene expression of the Na⁺/Ca²⁺ exchanger in settings where SR calcium transport mechanisms are defective may include Ca_i²⁺ removal and entry to sustain myocardial function.

The Na⁺/Ca²⁺ exchanger is a major Ca_i²⁺ efflux mechanism during myocardial relaxation. Its capacity to function in the reverse mode, promoting sarcolemmal Ca_i²⁺ entry, has also been described but most conspicuously as a phenomenon during ischemia and reperfusion (23, 37, 44). Evidence is accumulating, however, that demonstrates that, under basal conditions, the Na⁺/Ca²⁺ exchanger may operate in the Ca_i²⁺ influx mode to increase SR Ca_i²⁺ content and contribute to contractility (7, 9, 26). The lack of consensus on this point may be partly ascribed to the different cell preparations and species studied. Indeed, a recent report by Terracciano et al. (45) describes how, during the latter part of the decline in the Ca_i²⁺ transient, the Na⁺/Ca²⁺ exchanger reverses and brings Ca²⁺ into mouse ventricular myocytes. Moreover, transgenic overexpression of the Na⁺/Ca²⁺ exchanger resulted in a significant increase in the SR Ca_i²⁺ content of mouse ventricular myocytes (45), which may have contributed to their faster rate of Ca_i²⁺ release and uptake (50). In view of reports of enhanced Na⁺/Ca²⁺ exchange in the failing myocardium (10, 39, 43), examination of Ca_i²⁺ dynamics and function in mice engineered to overexpress the exchanger provides an opportunity for understanding its functional importance in the damaged heart.

Under baseline conditions, we found no differences in peak systolic or diastolic Ca_i²⁺ levels between wild-type and transgenic hearts overexpressing the Na⁺/Ca²⁺ exchanger, which is consistent with previous studies using this model (1, 45, 50). We found that the rate of decline of Ca_i²⁺, especially in the terminal phase, is accelerated in transgenic hearts, suggesting an enhanced effectiveness of Ca_i²⁺ extrusion by Na⁺/Ca²⁺ exchange (50). Our measurement of _P in transgenic hearts (Table 1) suggests that faster extrusion of Ca_i²⁺ may have a functional correlate under baseline conditions. These results confirm previous findings in transgenic ventricular myocytes (1, 45, 50), in which faster relaxation was ascribed to the increased abundance of Na⁺/Ca²⁺ exchangers operating in the forward mode to extrude Ca_i²⁺. Molecular studies of this transgenic model (45, 50) have reported no differences in other SR calcium regulatory proteins.

The mechanisms of early ischemic contractile failure are numerous and include ionic, energetic, and mechanical pathways (4). With interruption of a glucose-rich perfusate to the heart, the vasculature collapses and a complex set of ionic and metabolic changes in the intracellular milieu of the cardiac myocyte ensues. It is generally agreed that the abrupt decline of LV pressure is largely a consequence of the loss of turgor of the perfused coronary vasculature (22). However, ATP decreases and hydrogen ions and inorganic phosphates increase immediately upon interruption of myocardial perfusion. H⁺ decreases SR Ca_i²⁺ release by alterations in the activity of the Ca_i²⁺ release channel. Acidic pH inhibits Ca_i²⁺ binding to myofibrillar troponin C, thereby reducing actomyosin ATPase activation and force generation. The evolving metabolic acidosis is believed to impair SR calcium transport during the early phase of myocardial ischemia, contributing to the decrease in force development (2). With prolonged ischemia, increasing acidosis as well as depressed levels of ATP impair the ability of the SR to resequester Ca_i²⁺, such that diastolic Ca_i²⁺ increases during ischemia (21, 32). During hypoxia as well, accumulation of intracellular inorganic phosphates and H⁺ may contribute to contractile depression (2, 13, 21). As in no-flow ischemia, alterations of the action potential, changes in Ca_i²⁺ transport by the SR, and a decrease in the responsiveness of the myofilaments to Ca_i²⁺ contribute to contractile dysfunction during hypoxia (13). When the internal pH of cardiac cells is lowered, as occurs during ischemia and hypoxia, Na⁺/H⁺ exchange becomes the major regulatory system by which cells recover from acidosis (25). Accumulation of H⁺ has been shown to mediate a rise in Na⁺ by activation of the Na⁺/H⁺ exchanger (3, 25). Inhibition of the Na⁺/K⁺-ATPase activity during hypoxia markedly increases Na⁺ accumulation (3). The resultant net increase in intracellular Na⁺ during ischemia and hypoxia is thought to provoke Ca_i²⁺ influx via reverse transport of the Na⁺/Ca²⁺ exchanger in efforts to alleviate Na⁺ loading (37, 38). In fact, it was recently demonstrated that the reverse mode of Na⁺/Ca²⁺ exchange accounts for the largest contribution to Ca_i²⁺ accumulation in anoxic cardiomyocytes (24).

Our measurements of Ca_i²⁺ and function during ischemia and hypoxia are consistent with the above described sequelae. Our results demonstrate that, during the first minute of ischemia, while LV pressure decreased 75% from baseline, there was a 20% decrease in peak Ca_i²⁺ accompanied by an increase in diastolic Ca_i²⁺ in wild-type hearts. During the second minute of ischemia, the diastolic Ca_i²⁺ returned to baseline, but peak Ca_i²⁺ remained depressed. Subsequent ischemia resulted in progressive increases in peak Ca_i²⁺ toward preischemic values and significant increases in diastolic Ca_i²⁺ above baseline. These observations are consistent with the notion that early contractile failure during ischemia may, in part, be a consequence of impaired SR Ca_i²⁺ release and that protracted ischemia leads to reduced SR Ca_i²⁺ uptake. Analyses of digitized pressure recordings indicated some evidence of arrhythmias, which may have contributed to the rise in diastolic calcium during ischemia. In transgenic hearts, LV function decreased 60% from baseline during the first minute of ischemia, whereas peak Ca_i²⁺ did not change. Peak Ca_i²⁺ decreased only slightly during the second minute of ischemia and remained significantly greater than peak Ca_i²⁺ in wild-type hearts throughout 5 min of ischemia (Fig. 2). In transgenic hearts, diastolic Ca_i²⁺ did not change during the first minute of ischemia and decreased during the second minute of ischemia. Yao et al. (50) demonstrated that the Na⁺ content was not different between wild-type and transgenic cardiomyocytes. Cross et al. (9), using ³¹P NMR in the same transgenic model, reported comparable decreases in myocardial ATP levels and pH during ischemia in wild-type and transgenic hearts. Accordingly, our data suggest that reduced diastolic Ca_i²⁺ and preserved peak Ca_i²⁺ during ischemia in transgenic hearts were a consequence of a greater abundance of Na⁺/Ca²⁺ exchangers operating in the forward mode and the reverse mode. A previous study (9) indicated that overexpression of the Na⁺/Ca²⁺ exchanger increased susceptibility to ischemia and reperfusion injury, more prominently in male hearts, perhaps due to sustained Ca_i²⁺ influx during ischemia or increased Ca_i²⁺ overload upon reperfusion. Interestingly, however, we found that, during early ischemia, the amplitudes of the Ca_i²⁺ signals were greatest in the female transgenic hearts (data not shown).

During hypoxia, peak Ca_i²⁺ decreased by 15%, and functional indexes decreased by ~25% in wild-type hearts. Diastolic Ca_i²⁺ significantly decreased in wild-type hearts, but LVEDP did not change. Moreover, _L90  10% was prolonged during hypoxia in wild-type hearts, but _P was unchanged. In transgenic hearts, neither peak nor diastolic Ca_i²⁺ decreased during hypoxia. During hypoxia, _L90  10% was significantly abbreviated in transgenic hearts, suggesting an enhanced effectiveness of Ca_i²⁺ extrusion by Na⁺/Ca²⁺ exchange (50). The preservation of Ca_i²⁺ in hearts overexpressing the Na⁺/Ca²⁺ exchanger was paralleled by better preservation of function compared with wild-type hearts, in which both Ca_i²⁺ and function were depressed during hypoxia (Fig. 3). Previous studies (3, 24, 38) have shown that intracellular Na⁺ loading during hypoxia is the driving force behind Ca²⁺ influx via Na⁺/Ca²⁺ exchange. Our current observations, then, suggest that, during hypoxia and as in ischemia, preservation of systolic and diastolic Ca_i²⁺ in transgenic hearts was a consequence of a greater number of Na⁺/Ca²⁺ exchangers operating in both the reverse and forward modes.

We (14) have previously shown that, in vitro, Ca_i²⁺ exchange during no-flow ischemia may be sustainable in the small mouse heart by oxygen that diffuses epicardially into the myocardium (14). Accordingly, despite efforts to minimize the partial pressure of oxygen in the buffer, hypoxia may still have promoted generation of ATP by aerobic metabolism. To exclude the effects of inhomogeneous oxygen tensions within hearts or between groups of hearts, we studied the effects of metabolic inhibition by cyanide, which inhibits oxidative phosphorylation in both surface and core cells. Chemical hypoxia with cyanide-containing perfusate resulted in a much more profound depression in cardiac function than did hypoxia induced by 95% N₂-5% CO₂-bubbled perfusate. Yet the extent of systolic dysfunction in transgenic hearts was significantly blunted compared with wild-type hearts during cyanide perfusion, suggesting that the increased abundance of Na⁺/Ca²⁺ exchangers in the transgenic hearts may contribute to their ability to generate pressure despite uniform inhibition of oxidative phosphorylation. The increases in diastolic pressure with cyanide were larger in transgenic hearts. Because more complete inhibition of oxidative phosphorylation would further impair SR calcium reuptake, the observations in transgenic hearts are consistent with a greater abundance of Na⁺/Ca²⁺ exchangers working in the reverse mode during cyanide perfusion. The occurrence of pressure alternans in transgenic hearts might indicate greater Ca_i²⁺ influx. More profound changes in membrane stability and/or ion homeostasis imposed by cyanide intoxication (19, 20) may also have contributed to the increased diastolic pressure in transgenic hearts.

Observations from hypoxic transgenic hearts indicate that, despite preservation of Ca_i²⁺ influx and efflux, there remains a significant (~10%) depression of function, suggesting that reduced responsiveness of the myofilaments to Ca_i²⁺ substantially contributes to contractile failure in the mouse heart. Our findings from this transgenic model provide additional insight into the relative contributions of reduced Ca_i²⁺ availability and decreased Ca_i²⁺ myofilament sensitivity to contractile failure in hypoxic and ischemic myocardium. Reasoning that hypoxia-induced decreases in pH and ATP levels were comparable in transgenic and wild-type hearts, as was demonstrated in an NMR study of the same transgenic model during ischemia (9), our data suggest that ~10% of the contractile failure is attributable to loss of Ca_i²⁺ myofilament responsiveness, and the additional ~15% of contractile failure observed in the wild-type hearts is attributable to the 15% reduction in peak Ca_i²⁺. Likewise, the differences in Ca_i²⁺ and function observed provide additional insight into the question of the relative contribution of changes in Ca_i²⁺ to early ischemic contractile failure. In wild-type hearts, LV pressure decreased 75% during the first minute of ischemia, whereas the peak of the Ca_i²⁺ signal decreased <20%. In transgenic hearts, LV pressure decreased 60% during 1 min of ischemia, whereas peak Ca_i²⁺ was unchanged. These data infer that, in the mouse heart, nearly 20% of the reduction of the pressure-generating capability during the first minute of ischemia is a consequence of a reduction in peak Ca_i²⁺. Moreover, assuming that during early no-flow ischemia the intracellular environment is similar to that occurring during hypoxia, we deduce that nearly 10% of the contractile failure during early ischemia may be attributable to reduced myofilament responsiveness to Ca_i²⁺. Therefore, we propose that, in the mouse heart, the rapid loss of LV function during the first minute of no-flow ischemia is attributable to ~70% loss of turgor of the coronary vasculature, 20% to reduced levels of Ca_i²⁺ availability, and 10% to decreased Ca_i²⁺ myofilament sensitivity. Because absolute anoxia may not occur in the ischemic heart (42, 51), these observations could be relevant to human coronary disease, where coronary occlusion is often not complete.

In vivo and in vitro studies have shown that inhibiting Na⁺/Ca²⁺ exchange by inhibiting Na⁺/H⁺ exchange reduces the occurrence of ischemia-induced arrhythmia (33, 41, 49), suggesting that sarcolemmal influx of Ca_i²⁺ during ischemia affects E-C coupling. Moreover, interventions to block Ca²⁺ influx during coronary artery occlusion have resulted in more severe ischemic dysfunction (46, 48), despite improvements after reperfusion. A corollary to these observations may be that enhanced Ca²⁺ influx and efflux during myocardial ischemia might attenuate ischemic dysfunction. To the best of our knowledge, this hypothesis has not been tested. In light of our findings, demonstrating better preservation of Ca_i²⁺ and function during hypoxia and ischemia in transgenic hearts, we propose that the ionic rearrangements that contribute to the sequelae of cardiac contractile failure may be less detrimental in the presence of enhanced gene expression of the Na⁺/Ca²⁺ exchanger.

In summary, the results demonstrate that hearts with enhanced gene expression of the Na⁺/Ca²⁺ exchanger are significantly better able to maintain Ca_i²⁺ homeostasis while boosting myocardial function under conditions of ischemic or hypoxic stress and metabolic inhibition by cyanide. The data indicate the approximate contributions of vascular collapse, Ca_i²⁺ availability, and myofilament sensitivity to early ischemic contractile failure. Our findings suggest that enhanced gene expression of the Na⁺/Ca²⁺ exchanger attenuates hypoxic and ischemic contractile failure by preserving Ca_i²⁺ homeostasis and point to the functional relevance of upregulation of Na⁺/Ca²⁺ exchange in pathological settings where SR transport mechanisms are impaired.