INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

POSTISCHEMIC CA²⁺ overload is a major contributor of myocardial ischemia-reperfusion injury and may originate from reversed Na⁺-Ca²⁺ exchange upon reperfusion (26). This exchange is triggered by ischemic intracellular Na⁺ (Na) accumulation (21, 27, 28) and additional Na⁺ influx upon reperfusion via Na⁺-H⁺ exchange (29). Rapid resumption of Na⁺-K⁺-ATPase activity upon reperfusion may be beneficial by allowing efflux of Na⁺ without concomitant influx of Ca²⁺. Recently, we have shown that Na decreases immediately upon reperfusion via Na⁺-K⁺-ATPase activity in hearts reperfused with both glucose and pyruvate as substrates despite simultaneous Na⁺ influx via Na⁺-H⁺ exchange (29).

The trigger for reactivation of Na⁺-K⁺-ATPase and the energy source involved are still a matter of debate. Even under aerobic conditions, energy derived from either oxidative phosphorylation or glycolysis may be coupled to specific tasks. Oxidative ATP has been suggested to preferentially support mechanical activity, whereas glycolytic ATP may preferentially fuel membrane functions (33). For example, glycolysis has been suggested to be required to maintain Ca²⁺ homeostasis under conditions of increased Ca²⁺ entry by -adrenergic stimulation (22), probably by preferentially fueling the sarcoreticular Ca²⁺-ATPase (36). Colocalization of glycolytic enzymes and sarcolemmal membrane proteins, e.g., ATP-sensitive K⁺ channels (34), further strengthens the existence of a coupling of glycolytic energy metabolism to membrane functions.

Myocardial recovery during postischemic reperfusion has been associated with glycolytic activity as well. Recovery of contractile function in isolated rabbit hearts after reperfusion with glucose was superior to either pyruvate or palmitate, although ATP contents during reperfusion were similar (12). Furthermore, recovery of Ca²⁺ homeostasis (11) and metabolic activity (10) were better in glucose-reperfused hearts compared with reperfusion with pyruvate in combination with iodoacetate (IAA, an irreversible inhibitor of glycolysis). Therefore, it would be interesting to evaluate whether resumption of Na⁺-K⁺-ATPase activity is inhibited in pyruvate-reperfused hearts. Because inhibition of the pyruvate dehydrogenase (PDH) enzyme complex has been demonstrated during early postischemic reperfusion (17), one inhibiting factor involved may be inadequate activity of this first step in the oxidative energy production from pyruvate. Such a low PDH activity during reperfusion may be overcome by the PDH-kinase inhibitor dichloroacetate (DA) (18, 25).

In the present study using ²³Na NMR spectroscopy, we evaluate the postischemic time course of Na in isolated rat hearts reperfused with either the glycolytic substrate glucose, the oxidative substrate pyruvate, or pyruvate in combination with DA. Our data demonstrate that activation of the Na⁺-K⁺-ATPase during early reperfusion is delayed in hearts fueled by oxidatively derived ATP compared with glycolytic ATP. Moreover, DA does not enhance postischemic Na⁺-K⁺-ATPase activity in pyruvate hearts, which indicates that poor activity of the PDH complex is not a limiting factor in this regard. Restoration of Na to nearly baseline levels after prolonged reperfusion was not affected by substrate choice in our model.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Heart preparation. Fifty-four male Wistar rats (weighing 300-400 g) were anesthetized with diethylether. After hearts were anticoagulated with heparin (250 IU iv), they were excised and cooled in ice-cold perfusate. After aortic cannulation, hearts were perfused according to Langendorff by constant-pressure (76 mmHg) perfusion with a modified Krebs-Henseleit buffer (pH 7.35 ± 0.05) at 37°C. The standard buffer contained (in mmol/l) 148.0 Na⁺, 4.7 K⁺, 1.3 Ca²⁺, 1.0 Mg²⁺, 128.3 Cl, 24.0 HCO, and 11.0 glucose. A drain was inserted through the apex of the left ventricle to remove any thebesian flow, and a latex balloon (Sachs, March, Germany) connected to a Statham P23dB pressure transducer (Gould, Cleveland, OH) was positioned in the left ventricle to assess contractility. The balloon volume was adjusted to obtain a diastolic pressure of 7.5 mmHg; thereafter, the volume was kept constant during the remainder of the experimental protocol. Radio frequency-filtered copper-wire electrodes were attached to the outflow tract of the right ventricle, and hearts were paced at 5 Hz throughout the protocols with a Grass S48 pulse stimulator (Grass Instruments, Quincy, MA). The rate-pressure product (RPP; heart rate × developed pressure) was used as an index of cardiac function. Coronary flow was measured continuously with electromagnetic flow probes (Skalar, Delft, The Netherlands). These functional parameters were registered on-line with a personal computer-based data-acquisition program (Erasmus University, Rotterdam, The Netherlands) and evaluated afterward.

5

Experimental protocols. Hearts studied using ²³Na NMR spectroscopy were divided into four groups (n = 6 in all groups). The first group of hearts was subjected to 5 min of preischemic glucose (11 mmol/l) perfusion, 20 min of substrate-free perfusion with 500 µg/l glucagon (Novo Nordisk A/S) to exhaust intracellular glycogen stores (16), and 1 min of pyruvate perfusion (5 mmol/l). After preischemic perfusion, the group was subjected to 20 min of normothermal global ischemia and 30 min of reperfusion using glucose as the only substrate (group G; Fig. 1). The second group of hearts underwent the same perfusion protocol albeit without preischemic glycogen depletion and was reperfused with glucose as well (group G-no; Fig. 1). The last two groups of hearts were reperfused after preischemic glycogen depletion and 20 min of ischemia using either pyruvate as the sole substrate (group P; Fig. 1) or pyruvate in combination with DA (both 5 mmol/l, group PDA; Fig. 1). The 1-min period of preischemic pyruvate perfusion (or pyruvate + DA in DA-reperfused hearts) was employed to attenuate unfavorable side effects of the substrate-free glucagon treatment. In all groups, reperfusion was performed by constant-flow perfusion with ~70% of the preischemic flow (see DISCUSSION).

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Schematic representation of experimental protocols and simultaneous 23Na NMR data acquisition. G, glucose; P, pyruvate; D, pyruvate + dichloroacetate (DA). Solid boxes denote time points at which freeze-clamping was performed in separate groups of hearts.

NMR methods. ²³Na NMR spectra were recorded at 52.9 MHz on a Bruker MSL 200 spectrometer. The spectrometer was equipped with a 4.7-T vertical 150-mm bore magnet and a multinuclear 20-mm NMR probe. Magnetic field homogeneity was optimized using the ²³Na NMR free-induction decay (FID). We obtained 1-min and 5-s ²³Na NMR spectra (Figs. 1 and 2) by the accumulation of 256 and 24 consecutive FIDs, respectively, using 90° pulses and a 207-ms interpulse delay. Spectra were recorded with a 5-kHz spectral width and a time domain of 2,048 data points. The resonance of a standard solution in a glass capillary (containing a fixed amount of Na⁺ shifted downfield by 5 mmol/l of TmDOTP⁵) was used for calculations, and an intracellular volume of 2.45 ml/g dry weight (2) was assumed. Spectra were quantified by a time-domain-fitting routine (AMARES) (31) after the extracellular resonance was filtered, and prior knowledge was employed on the biexponential transverse relaxation of the Na resonance (30). NMR visibility of all Na⁺ signals was assumed to be 1.0 (27).

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Representative 23Na NMR spectra of an isolated perfused rat heart during preischemic glucose perfusion. A: 1-min spectrum. B: 5-s spectrum. Numbered Na+ peaks are as follows: 1) reference signal, 2) residual extracardiac, 3) extracellular, and 4) intracellular. ppm, Parts per million.

Glycogen assay. Glycogen content was determined by enzymatic analysis (13). Briefly, hearts were homogenized in ice-cold perchloric acid and incubated with amyloglucosidase (all enzymes were obtained from Sigma, St. Louis, MO) for 2 h at 40°C. Thereafter, using glucose-6-phosphate dehydrogenase and hexokinase, glucose liberated from glycogen was determined spectrophotometrically (Unicam 8630, Cambridge, UK) at 339 nm from the formation of NADPH.

Statistics. Results are presented as means ± SE. Data were analyzed by one-way ANOVA or repeated measures ANOVA. If significant differences were observed, groups were compared at relevant time points according to Tukey's procedure. Comparisons within groups were performed using a similar approach.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Glycogen levels. Glycogen content at 5 min of preischemic glucose perfusion was in accordance with values reported in the literature (1, 32) and amounted to 67.1 ± 4.4 µmol/g dry weight. Glucagon infusion, which stimulates glycogen breakdown, significantly reduced glycogen levels (0.0 ± 3.4 vs. 61.3 ± 6.1 µmol/g dry weight in groups G and G-no, respectively; P < 0.001); however, at the end of ischemia, the values were not significantly different between the groups (1.4 ± 2.9 vs. 10.9 ± 5.1 µmol/g dry weight in groups G and G-no, respectively). These values are also similar to published values for periods after glycogen depletion (7, 19) and ischemia (1).

Preischemic perfusion. Na (Fig. 3) and RPP [Fig. 4; overall mean (18.9 ± 0.8) × 10³ mmHg/min] did not differ between the groups during preischemic glucose perfusion. Upon infusion of glucagon, a significant but transient rise of RPP was observed in glucagon-treated groups (Fig. 4). However, after 20 min of glucagon infusion, RPP was lower than the respective control values in all glucagon-treated groups, and left ventricular end-diastolic pressure (EDP) had increased (32 ± 4, 19 ± 5, and 28 ± 3 mmHg in group G, P, and PDA hearts, respectively, vs. 7 ± 2 mmHg in group G-no hearts; P < 0.01, group G-no vs. groups G and PDA). Na was not altered significantly during glucagon infusion, although values tended to increase (Fig. 3). The 1 min of pyruvate (or pyruvate + DA in PDA hearts) perfusion immediately before ischemia effectively attenuated the unfavorable effects of the substrate-free glucagon perfusion, which resulted in a decrease of EDP (23 ± 4, 16 ± 5, and 18 ± 3 mmHg in group G, P, and PDA hearts, respectively, vs. 11 ± 3 mmHg in group G-no hearts; P = not significant) and Na (Fig. 3) and a complete recovery of phosphocreatine (as observed by ³¹P NMR in pilot experiments; data not shown).

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Intracellular sodium (Na) during preischemic perfusion, ischemia, and reperfusion in glucose-reperfused hearts without preischemic depletion of glycogen () and in glycogen-depleted glucose (), pyruvate (), and pyruvate + DA ()-reperfused hearts. For reasons of clarity, from the 5-s spectra obtained during the last 5 min of ischemia and the first 10 min of reperfusion only, every 12th data point is shown and error bars are omitted. Hatched box indicates glucagon infusion in glycogen-depleted hearts; solid box shows 1-min preischemic pyruvate perfusion period.

View larger version (26K): [in this window] [in a new window]   Fig. 4.   Rate-pressure product during preischemic perfusion, ischemia, and reperfusion in glucose-reperfused hearts without preischemic depletion of glycogen () and in glycogen-depleted glucose (), pyruvate (), and pyruvate + DA ()-reperfused hearts. Error bars indicate SE. Hatched box indicates glucagon infusion in glycogen-depleted hearts; solid box shows 1-min preischemic pyruvate-perfusion period.

Ischemia. In glucagon-treated hearts, the time to onset of ischemic contracture was reduced by ~60% (2.2 ± 0.2, 2.1 ± 0.1, and 2.0 ± 0.2 min in group G, P, and PDA hearts, respectively, vs. 4.9 ± 0.2 min in group G-no hearts; P < 0.01), and the amplitude of maximal contracture was exacerbated (130 ± 2, 108 ± 9, and 124 ± 4 mmHg in group G, P, and PDA hearts, respectively, vs. 66 ± 12 mmHg in group G-no hearts; P < 0.01). However, end-ischemic EDP did not differ between groups (overall mean value 54.4 ± 2.9 mmHg). The ischemic increase of Na (Fig. 3) did not differ significantly between groups, and end-ischemic Na levels amounted to 29.2 ± 2.2, 26.4 ± 1.6, 27.6 ± 1.7, and 26.0 ± 1.6 mmol/l in group G, G-no, P, and PDA hearts, respectively.

Reperfusion. During reperfusion, Na declined in all groups (Fig. 3). In glucose-reperfused hearts both with or without preischemic glycogen depletion, Na started to decrease immediately upon reperfusion. The initial rate of decline (first five time points of reperfusion) amounted to 34.0 ± 2.5% and 21.5 ± 2.9% of the end-ischemic value per minute in group G and G-no hearts, respectively, as determined by linear regression (P = not significant; Fig. 5). In pyruvate hearts, however, the decline of Na was delayed by ~25 s, resulting in an initial rate of decline of 2.2 ± 4.4% (P < 0.05 vs. groups G and G-no; Fig. 5). Addition of the PDH activator DA did not enhance this initial rate of decline of Na in pyruvate hearts, which amounted to 3.1 ± 5.5% (P < 0.01 vs. groups G and G-no; Fig. 5). However, irrespective of the substrate used, after 30 min of reperfusion Na had reached almost baseline levels in all groups (Fig. 3), yet functional recovery (RPP) was significantly improved (P < 0.05) in DA-treated hearts compared with glucose-reperfused hearts (66 ± 7, 63 ± 5, 78 ± 5, and 97 ± 11% of the control value in group G, G-no, P, and PDA hearts, respectively). Furthermore, EDP after 30 min of reperfusion was lowest in group PDA hearts as well (46 ± 8, 41 ± 8, 34 ± 10, and 33 ± 5 mmHg in group G, G-no, P, and PDA hearts, respectively; P = not significant).

View larger version (26K): [in this window] [in a new window]   Fig. 5.   Na during the first 30 s of reperfusion in glucose-reperfused hearts without preischemic depletion of glycogen () and in glycogen-depleted glucose (), pyruvate (), and pyruvate + DA ()-reperfused hearts. Error bars indicate SE.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

In the present study, we show that Na decreases immediately upon postischemic reperfusion in glucose-reperfused rat hearts. However, activation of Na⁺-K⁺-ATPase during early reperfusion is delayed in pyruvate hearts fueled only by oxidatively derived ATP. DA does not enhance postischemic Na⁺-K⁺-ATPase activity in pyruvate hearts, which indicates that poor activity of the PDH complex is not a limiting factor in this regard. The decline of Na to nearly baseline levels after prolonged reperfusion is not affected by the type of substrate.

Na and postischemic Na⁺-K⁺-ATPase activity. Na accumulation is a well-known consequence of myocardial ischemia (21, 27, 28) and may develop due to decreased efflux of Na⁺ by Na⁺-K⁺-ATPase and due to (continuing) influx via the Na⁺ channel (28), Na⁺-H⁺ exchange (24), and other routes. Na⁺-K⁺-ATPase activity is inhibited during ischemia (3), most likely because of a lack of ATP, a rise in ADP and P_i levels, and development of intracellular acidosis (4). Interestingly, although it was not the main topic of this investigation, the ischemic increase of Na in the glycogen-depleted hearts (~200% after 20 min) of the present study is similar to the ischemic increase in hearts without prior depletion of glycogen (group G-no; Fig. 3); however, the time course differs. In the glycogen-depleted hearts, the increase of Na is initially faster and becomes slower later on, whereas the increase in the hearts without prior depletion of glycogen is more linear. A likely explanation for this is that the hearts in the latter group are able to fuel the Na⁺-K⁺-ATPase during ischemia for a considerably longer time via anaerobic glycolysis, at least until contracture develops (14), which limits Na accumulation during the first 5-10 min of ischemia. However, because ischemic acidosis will inevitably develop as a consequence of continued anaerobic glycolysis, this will in turn lead to activation of Na⁺-H⁺ exchange and consequently to an increased rate of Na⁺ influx during the remainder of ischemia.

Na+-K⁺-ATPase activity and glycolysis. As we have shown previously (29), the decline of Na upon postischemic reperfusion is ouabain sensitive and can be attributed to a rapid resumption of Na⁺-K⁺-ATPase activity despite concomitant Na⁺ influx via Na⁺-H⁺ exchange. In the present study, the initial decline of Na was significantly faster in the two glucose-reperfused groups of hearts compared with pyruvate-reperfused hearts (Fig. 5). This indicates that recovery of Na⁺-K⁺-ATPase activity is delayed in the absence of glycolytic flux and suggests some degree of functional coupling between Na⁺-K⁺-ATPase activity and glycolytic flux during postischemic reperfusion, especially because endogenous contributions to postischemic glycolysis could be prevented by preischemic glycogen depletion.

Glycolysis and postischemic functional recovery. In evaluating the importance of glycolysis for postischemic contractile recovery, the effects of (continuing) glycolysis during ischemia must be separated from the effects of glycolytic activity during reperfusion. Earlier studies have revealed a salutary effect of preischemic glycogen reduction on postischemic functional recovery (1), and, accordingly, accumulation of glycolytic waste products (lactate, H⁺) was found to adversely affect recovery (23). Paradoxically, other studies found the rate of glycolysis during (low-flow) ischemia to be inversely proportional to the severity of ischemic injury, and maintenance of glycolytic ATP production improved postischemic contractile recovery (8, 9, 32). Cross and colleagues (6) hypothesized that glycolysis during ischemia is only beneficial until contracture occurs. If the heart is reperfused after development of contracture, the degree of accumulation of metabolic waste products before contracture is inversely correlated to postischemic recovery (6). In the present study, within 5 min all hearts went into ischemic contracture, and preischemic glycogen reduction (compare groups G and G-no; Fig. 4) did not improve postischemic contractile recovery. This indicates that protection afforded by diminished accumulation of glycolytic waste products in the glycogen-depleted hearts of the present study was apparently similar to the beneficiary effects of a prolonged glycolytic ATP production in the nondepleted hearts.

Conclusions. During postischemic reperfusion some degree of functional coupling of glycolytic ATP and Na⁺-K⁺-ATPase activity exists, although glycolysis is not essential for recovery of Na homeostasis and contractility after prolonged reperfusion. In addition, the delayed postischemic activation of the Na⁺-K⁺-ATPase in pyruvate-reperfused hearts is not a consequence of inhibition of PDH upon reperfusion.