- Email: [email protected]
Metabolic Recovery of Isolated Adult Rat Cardiomyocytes after Energy Depletion: Existence of an ATP Threshold?
J Mol Cell Cardiol 30, 2111–2119 (1998) Article No. mc980805
Metabolic Recovery of Isolated Adult Rat Cardiomyocytes after Energy Depletion: Existence of an ATP Threshold? A. Bonz1, B. Siegmund2, Y. Ladilov2, C. F. Vahl1 and H. M. Piper2 1
Chirurgische Universita¨tsklinik, Abt. Herzchirurgie, Universita¨t Heidelberg, Im Neuenheimer Feld 110, D-69120 Heidelberg; 2Physiologisches Institut, Universita¨t Giessen, Aulweg 129, D-35392 Giessen (Received 19 February 1998, accepted in revised form 18 August 1998) A. B, B. S, Y. L, C. F. V H. M. P. Metabolic Recovery of Isolated Adult Rat Cardiomyocytes after Energy Depletion: Existence of an ATP Threshold? Journal of Molecular and Cellular Cardiology (1998) 30, 2111–2119. The question was investigated whether cardiomyocytes can be resuscitated after extreme energy depletion, i.e. after loss of ATP >70%. Isolated ventricular cardiomyocytes of the adult rat were exposed to metabolic inhibition with dinitrophenol and cyanide (DNP 0.2 m; KCN 2 m). After rapid energy depletion, cells were “reoxygenated” by wash-out of DNP and KCN. Intracellular calcium, cell length, ATP and creatine phosphate (CrP) of the cardiomyocytes were monitored. Metabolic inhibition resulted in a depletion of the stores of ATP and CrP by more than 95% of the normoxic values and caused a cytosolic Ca2+ overload. Parameters of metabolic recovery were: (i) resynthesis of CrP; (ii) recovery of a normal cytosolic Ca2+ control; and (iii) the elicitation of energy-dependent hypercontracture. “Reoxygenation”, i.e. wash-out of metabolic inhibitors, reactivated oxidative phosphorylation. Consecutively, CrP levels recovered to 76.0±7.3%, ATP levels recovered to 10.4±2.3% (means±.., n=10) of the initial normoxic values, a normoxic intracellular calcium level was re-established and hypercontracture was elicited. Prolongation of metabolic inhibition with KCN (2 m) or inhibition of the Na+/K+ pump with ouabain (0.5 m) disabled the cardiomyocytes to recover from cytosolic Ca2+ overload and prevented hypercontracture. It is concluded that even after extensive energy depletion metabolic resuscitation of the myocardial cell remains possible and a critical range of ATP for recovery, i.e. a “threshold” 1998 Academic Press of a 70% loss of ATP, does not exist. K W: ATP depletion; Metabolic inhibition; Cardiomyocytes; Irreversible cell injury; Anoxia/reoxygenation.
Introduction In ischemic or anoxic myocardium aerobic energy production ceases, and anaerobic processes of energy production are not sufficient to balance the demand. Consequently, cellular stores of high-energy phosphates are depleted. Eventually energydepleted myocardial cells develop irreversible cell injury. Systematic studies on energy metabolism and reversibility of ischemic myocardial injury have been started around 1970. Results of these studies
suggested a close relationship between the development of lethal injury in the ischemic myocardium and a considerable depletion of energy stores (Jennings et al., 1978). A concept has been developed, that a critical loss of high-energy phosphates initiates processes which finally result in irreversible cell damage. However, the causal relationship between energy depletion and manifestation of irreversibility of cell injury is not clear, because ischemic depletion of ATP is not necessarily related causally to irreversible cell damage. If the pathogenesis of consecutive events from energy
Please address all correspondence to: B. Siegmund, Physiologisches Institut, Justus-Liebig-Universita¨t, Aulweg 129, D-35292 Giessen, Germany.
0022–2828/98/102111+09 $30.00/0
1998 Academic Press
2112
A. Bonz et al.
depletion to ischemic cell death is interrupted, breakdown of high-energy phosphates even to a very low level need not end up with the lethal phase, i.e. with sarcolemmal disruption and necrosis (Reimer and Jennings, 1986). Attempts to determine a critical “threshold” of high-energy phosphate contents have indicated that ischemic-reperfused myocardium may fail to recover metabolically and functionally once the ATP contents have dropped below a critical threshold of approximately one third and creatine phosphate contents below approximately one tenth of the preischemic level (Spieckermann et al., 1969; Ku¨bler and Spieckermann, 1970). However, progression of ATP loss beyond two thirds of the normoxic value has not yet been disclosed as a sufficient cause of irreversible injury. Recent studies using isolated cardiomyocytes supported doubts that cellular metabolic recovery would be limited by such an “ATP threshold” of about 30% of the initial cellular ATP contents. It was shown, for example, that isolated myocardial cells retain metabolic competence and the ability to regain control over their cytosolic ion homeostasis in spite of a very prolonged time of energy deprivation (Saks et al., 1989; Siegmund et al., 1990, 1992). The present study was undertaken to explicitly address the question whether metabolic recovery and reactivation of energy-dependent processes are possible in myocardial cells after they had been depleted by more than 95% of their high-energy phosphate reserves. This degree of ATP loss exceeds by far the alleged critical “ATP threshold”. In the present study the capability of isolated cardiomyocytes to recover from extreme energy depletion was investigated. Cardiomyocytes were submitted to a very rapid and extensive depletion of creatine phosphate and ATP provoked by a respiratory uncoupler. Then, cardiomyocytes developed a Ca2+ overload. Finally, conditions for reactivation of oxidative energy metabolism were established and the synthesis of high-energy phosphates contents, the return to a normal cytosolic Ca2+ control, and activation of myofibrils were investigated.
Material and Methods Isolation of cardiomyocytes Ventricular heart muscle cells were isolated from adult male Wistar rats (200–250 g) as described previously (Piper et al., 1982), plated in medium
199 with 4% fetal calf serum on glass cover-slips or plastic culture dishes, which had been preincubated with this medium overnight. Four hours after plating the cover-slips were washed with medium 199. As a result of the wash, broken cells were removed, leaving a homogenous population of rod-shaped quiescent cardiomyocytes (>95%) attached to the cover-slips and culture dishes.
Fura-2 loading of the cells To measure the cytosolic Ca2+ concentration, cardiomyocytes were loaded with the fluorescent dye fura-2 at 37°C. The cells, attached to the coverslips, were incubated for 20 min in medium 199 with the acetoxymethylester of fura-2 (2.5 l). After loading, the cells were washed twice with medium 199 without fura-2. This washing step was followed by a 30-min post-incubation period in medium 199 to allow hydrolysis of the acetoxymethyl ester within the cell. The fluorescence from dye-loaded cells was at least 20 times higher than background fluorescence, i.e. fluorescence from cells not loaded with the dye. This loading protocol used was selected from several variations, because it provided the highest yield in fluorescence and minimal dye compartmentation (Ladilov et al., 1995).
Calcium and length measurements The cover-slip with the fura-2 loaded cells was introduced into an air-tight, temperature controlled (37°C), transparent perfusion chamber positioned in the light beam of an inverted microscope (Diaphot TDM, Nikon, Du¨sseldorf, Germany). Alternating excitation of the fluorescent dye at wavelengths of 340 nm and 380 nm was performed with an ARCation Measurement System adapted to the microscope (ISA, Grasbrunn, Germany). The emitted light (500–520 nm) from a 10×10 lm area within the fluorescent cell was collected by a photomultiplier of the system. Data were collected in the respective channels using an IBM PC/AT based data analysis system (model DM3000CM, ISA, Grasbrunn, Germany). Autofluorescence of the cells was measured prior to the loading procedure and was subtracted from the fura-2 signal. Since in the presented experiments the fura-2 signal saturates at a ratio value of 7, the measured maximum of the fura-2 signal indicates severe Ca2+ overload ([10−5 Ca2+). Simultaneous to the measurement of the cell’s
Recovery after ATP Depletion
fluorescence, the microscopic image was recorded with a video camera and stored on tape. Changes of the cell length were determined from these recordings. In the case of hypercontracted cells, the previous longitudinal axis was determined.
Experimental protocol The perfusion chamber (1 ml filling volume) placed on the microscope was perfused at a flow rate of 0.5 ml/min with modified, glucose-free Tyrode’s solution at 37°C containing (in m): 125.0 NaCl, 2.6 KCl, 1.2 KH2PO4, 1.2 MgSO4, 1.0 CaCl2, 25.0 HEPES; pH was 7.4 or 6.8, respectively. All experiments were started with an equilibration period by superfusing the cells with modified, glucose-free Tyrode’s solution (step 1). Then 50 m 2,3-butanedione monoxime (BDM) was added to the perfusion medium for 3 min (step 2) to prevent rigor shortening during the subsequent metabolic inhibition. This was followed by a period of 7 min of energy depletion (step 3). The Tyrode’s solution was supplemented with 50 m BDM, 0.2 m dinitrophenol (DNP), 4 m bis-(aminoethyl)-glycolether-N, N, N′, N′-tetraacetic acid (EGTA) and 20 m MgCl2. DNP was added in order to achieve a very rapid ATP depletion. EGTA, MgCl2 and BDM were added to reduce rigor shortening. In the next step (step 4), lasting 8 min, DNP was washed out with 1% (w/v) albumin and replaced by 2 m potassium cyanide (KCN). DNP was replaced by KCN in order to keep oxidative phosphorylation inhibited, but to make a fast metabolic recovery by wash-out of KCN possible. The following step (step 5) was varied in protocols A–E. In protocol A (control), the cells were allowed in step 5 to recover in Tyrode’s solution (pH 7.4) containing no metabolic inhibitors. Protocol B (acidosis) was as protocol A, but the pH of the medium was decreased to 6.8. In protocol C (continued metabolic inhibition), 2 m KCN remained present to prevent energetic recovery. In protocol D (continued metabolic inhibition with glucose), 2 m KCN remained present and 15 m glucose was added to allow glycolytic energy production. In protocol E (ouabain), 0.5 m ouabain was added to inhibit Na+/ K+-ATPase.
Determination of high-energy phosphates In separate experiments the state of energy of the cells was determined during chemical anoxia (0.2 m DNP + 2 m KCN in modified Tyrode’s
2113
solution) using cardiomyocytes plated on culture. The experiments were terminated by adding HCIO4 to the culture dishes (0.6 HCIO4 final concentration). Protein contents of the samples were determined in the acid precipitates according to the method of Lowry et al. (1951) using bovine serum albumin as standard. After neutralization, perchloric acid extracts were analysed for creatine phosphate (CrP) (Bergmeyer, 1974) and adenosine triphosphate (ATP) (Ju¨ngling and Kammermeier, 1980).
Statistical analysis All data are given as a representative example of n experiments or as mean values±.. of n independent culture preparations. Data were analysed according to Student’s t-test for non-paired samples. Differences with P<0.05 are regarded to indicate statistical significance.
Materials Falcon tissue culture dishes were obtained from Becton Dickinson (Heidelberg, Germany), medium 199 from Boehringer (Mannheim, Germany), fetal calf serum from GIBCO (Berlin, Germany), crude collagenase from Biochrom (Berlin, Germany), fura2/AM from Paesel and Lorey (Frankfurt/M., Germany) and 2,3-butanedione monoxime from Sigma (Deisenhofen, Germany). All other chemicals were purchased from Merck (Darmstadt, Germany) and of highest analytical grade.
Results State of energy during metabolic inhibition and metabolic reactivation Protocol A contained five steps. In the initial control situation cells were incubated in a normal Ca2+ containing medium (step 1) and a short preincubation with 30 m BDM (step 2). Then energy depletion was induced by the uncoupler of mitochondrial respiration DNP. DNP was applied in a Ca2+-free incubation medium supplemented with BDM and high Mg2+ to reduce rigor shortening (step 3). In the subsequent step 4, Ca2+ was repleted and DNP was eluted, with help of albumin, but oxidative energy production remained inhibited by the presence of KCN. In the last step 5, the initial
2114
A. Bonz et al.
Figure 1 Cellular contents of adenosine triphosphate (ATP) and creatine phosphate (CrP) during metabolic inhibition and recovery. Steps 1 and 2 of the protocol were without metabolic inhibition, steps 3 and 4 with metabolic inhibition, step 5 represents the recovery period. Details were as specified in Material and Methods. Data are means±.., n=10, where the 100% values correspond to 22.2±4.4 nmol/mg protein ATP and 23.0±4.9 nmol/mg protein CrP, respectively.
control conditions were re-established (“reoxygenation”) and metabolic recovery could take place. Under initial control conditions cardiomyocytes contained high levels of high-energy phosphates: ATP 22.2±4.4 nmol and CrP 33.0±4.9 nmol per mg protein (n=10). As shown in Figure 1, highenergy phosphates were rapidly degraded in presence of the uncoupler DNP (step 3). After only 1 min ATP content had declined to 14% and CrP below its detection limit (Ζ2% of control). After 3 min ATP was below 5%. Cells were left in that state of complete exhaustion of energy stores until end of step 3, i.e. for 15 min in total. When all inhibitors were removed in step 5 of the protocol, CrP contents recovered, reaching 44% of the initial control value within 10 min. Only a minor recovery of ATP contents was observed. To investigate whether metabolic recovery takes place under acidotic conditions or in the presence of ouabain CrP levels were measured (Fig. 2). CrP decreased below the detection limit during metabolic inhibition (MI) and recovered to about 55% of the initial normoxic control value within 30 min after wash-out of the inhibitor (protocol A). Resynthesis of CrP under acidotic conditions (pH 6.8)
Figure 2 Cellular contents of creatine phosphate (CrP) during metabolic inhibition and subsequent 30 min of reoxygenation. Ctr, control value before metabolic inhibition (i.e. after step 1); MI, metabolic inhibition (i.e. after step 4); A, 30 min recovery after wash-out of KCN (protocol A of step 5); B, 30 min recovery after wash-out of KCN under acidotic conditions of pHe 6.8 (protocol B of step 5); E, 30 min recovery after wash-out of KCN in presence of ouabain (protocol E of step 5). Details were as specified in Material and Methods. Data are means±.., n=10 (Ctr) and 5, where the 100% value corresponds to 33.0±4.9 nmol/mg protein CrP. Difference from A: ∗ P<0.0001.
was not different to normal pH conditions (protocol B) but was only minimal in the presence of ouabain (protocol E).
Experiments with metabolic recovery In each cardiomyocyte the changes in cell length and in the fura-2 fluorescence ratio, indicating changes in cytosolic Ca2+ concentrations, were monitored. Figure 3 shows a typical example. Starting from the initial slack length, the cell slightly shortened in step 3 of protocol A, i.e. during rapid energy depletion. Upon wash-out of metabolic inhibitors (step 5) the cardiomyocyte hypercontracted to about 25% of its initial length. The fura-2 ratio started to rise during step 4, i.e. in the energydepleted state where extracellular Ca2+ was again present. Upon removal of the inhibitor KCN (step 5) the fura-2 ratio gradually declined to the initial control value with an intermittent phase of oscillations. Protocol B differs from protocol A by a lower pH (6.8 instead of 7.4) in all media. Under these acidotic conditions the changes in fura-2 ratio and cell
Recovery after ATP Depletion
Figure 3 Protocol A (control). Time course of fura-2 ratio (original recording, arbitrary units) and cell length (O, percentage of resting length) in a single cardiomyocyte during metabolic inhibition and recovery. Steps 1 and 2 of the protocol were without metabolic inhibition, steps 3 and 4 with metabolic inhibition, step 5 represents the recovery period. Representative example of six separate experiments.
2115
Figure 5 Cell length of cardiomyocytes prior to metabolic inhibition (prior MI) and at end of protocols A–E. Data are means±.., n=6. Differences from MI: ∗ P<0.0001.
Figure 6 Fura-2 ratio of cardiomyocytes prior to metabolic inhibition (prior MI) and at end of protocols A–E. Data are means±.., n=6. Differences from MI: ∗ P<0.0001. Figure 4 Protocol B (acidosis). Time course of fura-2 ratio (original recording, arbitrary units) and cell length (O, percentage of resting length) in a single cardiomyocyte during metabolic inhibition and recovery. Steps 1 and 2 of the protocol were without metabolic inhibition, steps 3 and 4 with metabolic inhibition, step 5 represents the recovery period. The pH of the medium in step 5 was 6.8. Representative example of six separate experiments.
length in steps 3 and 4 of the protocol, i.e. during energy depletion, closely resembled the changes seen under protocol A (Fig. 4). When metabolic inhibition was removed (step 5) a marked difference compared to protocol A became apparent. In media with pH 6.8 cells did not develop hypercontracture. The recovery of the fura-2 ratio towards the control level proceeded more slowly than in media with pH 7.4. The intermittent oscillations seen in the recovery phase of protocol A remained absent.
Average data for all cells of cell length and fura-2 ratio at the end of step 5 are shown in Figures 5 and 6.
Experiments without metabolic reactivation In protocol C cardiomyocytes were submitted to the same conditions as in protocol A, except that in step 5 metabolic inhibition by KCN was not removed (Fig. 7). Under these conditions cells did not hypercontract nor did the fura-2 ratio recover. Comparison with protocol A indicates that the development of hypercontracture and recovery of the fura-2 ratio seen in step 5 of protocol A are a consequence of removing the preceding metabolic inhibition.
2116
A. Bonz et al.
Figure 7 Protocol C (continued metabolic inhibition). Time course of fura-2 ratio (original recording, arbitrary units) and cell length (O, percentage of resting length) in a single cardiomyocyte during metabolic inhibition and recovery. Steps 1 and 2 of the protocol were without metabolic inhibition, steps 3 and 4 with metabolic inhibition. In step 5 KCN (2 m) remained present. Representative example of six separate experiments.
Figure 9 Protocol E (ouabain). Time course of fura-2 ratio (original recording, arbitrary units) and cell length (O, percentage of resting length) in a single cardiomyocyte during metabolic inhibition and recovery. Steps 1 and 2 of the protocol were without metabolic inhibition, steps 3 and 4 with metabolic inhibition. In step 5 inhibitors of energy metabolism were absent, but ouabain (0.5 m) was added. Representative example of six separate experiments.
ouabain (0.5 m) was added in step 5 after washout of the respiratory uncoupler DNP in order to prevent the re-establishment of a large transsarcolemmal Na+ gradient which mediates the driving force of the Na+/K+ ATPase for extrusion of Ca2+ via the Na+/Ca2+ exchanger. Similarly to protocol D, the fura-2 ratio did not recover although the metabolic inhibitor KCN was not present (Fig. 9).
Discussion Figure 8 Protocol D (continued metabolic inhibition with glucose). Time course of fura-2 ratio (original recording, arbitrary units) and cell length (O, percentage of resting length) in a single cardiomyocyte during metabolic inhibition and recovery. Steps 1 and 2 of the protocol were without metabolic inhibition, steps 3 and 4 with metabolic inhibition. In step 5 KCN (2 m) remained present and glucose (15 m) was added. Representative example of six separate experiments.
It was tested in protocol D whether administration of a high concentration of the glycolytic substrate glucose (15 m) in step 5 could compensate in energy production for the continuation of mitochondrial inhibition by KCN (Fig. 8). But development of hypercontracture and recovery of the fura-2 ratio in step 5 remained absent in presence of glucose. In protocol E the inhibitor of Na+/K+ ATPase
The major finding of the present study is that metabolic resuscitation of myocardial cells remains possible after a [95% depletion of ATP and CrP reserves. Parameters of metabolic recovery were the resynthesis of CrP, the recovery of a normal cytosolic Ca2+ control and the elicitation of energydependent hypercontracture. Experiments were performed using the model of isolated cardiomyocytes. We have analysed in several previous studies the consequences of energy depletion and metabolic reactivation in a similar model by exposing the cells to periods of anoxia and reoxygenation (Siegmund et al., 1990, 1992; Ladilov et al., 1995). In the present study the uncoupler DNP was used to very rapidly exhaust cellular energy stores. Uncouplers activate mitochondrial ATP hydrolysis by dissipating the electrochemical gradient of the inner mitochondrial
Recovery after ATP Depletion
membrane. In presence of DNP more than 95% of high-energy phosphates was lost within 3 min. Loss of high-energy phosphates in isolated cardiomyocytes is normally accompanied by rigor shortening, leading to a 35% reduction of cell length. If that rigor shortening lasts for more than a few minutes it becomes irreversible, probably due to rearrangements in cytoskeletal structures (Stern et al., 1985). In the present paper BDM was used to reduce rigor shortening, because this substance is known to prevent ischemic contracture when applied in high concentrations (Armstrong and Ganote, 1991; Vanoverschelde et al., 1994; Koyama et al., 1996), possibly by affecting the binding constant of MgATP (Zhao and Kawai, 1994) or by interfering directly with myofibrillar ATPase as a non-competitive inhibitor (Herrmann et al., 1992). BDM may act as a chemical phosphatase (Green and Saville, 1956; Fryer et al., 1988), but dephosphorylation of Ca2+ regulating proteins which may cause some lasting effects of increased osmotic fragility (Armstrong and Ganote, 1991) or stunning (Koyama et al., 1996) was tolerated in the present experiments. In the presence of the contractile inhibitory agents BDM and Mg2+ (step 2) rigor shortening was greatly attenuated by performing rapid energy depletion in Ca2+-free media. After a virtually complete loss of ATP and CrP had been achieved, the uncoupler was eluted from the cell preparation while mitochondrial energy production remained inhibited by KCN, a blocker of cytochrome C oxidase. During this step (4) of the experimental protocol of metabolic inhibition the extracellular media contained Ca2+ and consequently the cytosolic Ca2+ level started to rise. It has been demonstrated before that Ca2+ accumulates in energydepleted cardiomyocytes by entering the cells from outside (Siegmund et al., 1992). When the metabolic inhibitor KCN was removed from the cell preparation (step 5), mitochondrial energy production was reactivated. This was demonstrated by the recovery of CrP levels which were restored to nearly 50% of control values within 10 min after removal of KCN. The absolute contents of ATP remained low since energy-depleted cardiomyocytes lose most of the purine contents rapidly and de novo synthesis of adenine nucleotides is very slow (Zimmer et al., 1973; Isselhard et al., 1980; Vial et al., 1982). During the phase of metabolic recovery cytosolic Ca2+ accumulation was reversed. Eventually, a normal cytosolic Ca2+ control was re-established. That this recovery of Ca2+ control is due to active metabolism was demonstrated by experiments in which
2117
the metabolic inhibitor KCN had not been removed. In these latter experiments recovery from Ca2+ overload did not occur. Recovery from Ca2+ overload was also impossible when the Na+ pump of the sarcolemma was inhibited by ouabain. Impairment of Na+ extrusion prevents the re-establishment of a normal Na+ gradient across the sarcolemma and consequently prevents Ca2+ extrusion via Na+/Ca2+ exchange. The Na+/Ca2+ exchange mechanism is an important route for Ca2+ extrusion after anoxic Ca2+ overload (Siegmund et al., 1994). Presence of ouabain also prevented recovery of CrP. The reason why CrP level did not renormalize remains unclear. A possible explanation might be that mitochondria are forced to take up Ca2+ on the expense of the mitochondrial electrochemical gradient, and therefore produce less ATP (Lehninger, 1970), when Ca2+ remains high during reoxygenation due to inhibition of Na+/Ca2+ exchange by ouabain. This indication of an impaired energy metabolism may then explain why cardiomyocytes did not hypercontract in absence of sufficient metabolic energy, i.e. when the contractile machinery is not activated by ATP. Cardiomyocytes with a manifest Ca2+ overload in the cytosol developed hypercontracture upon metabolic reactivation. This phenomenon has been analysed in several studies before (e.g. Siegmund et al., 1992; Ladilov et al., 1995). It has been shown that hypercontracture develops because of sustained myofibrillar activation when mitochondrial energy production is initiated in presence of a massive Ca2+ overload. It can be prevented by pharmacological or acidotic inhibition of the contractile apparatus (Fabiato and Fabiato, 1978; Kitakaze et al., 1988; Siegmund et al., 1991; Ladilov et al., 1995). Hypercontracture was also prevented in the present study when experiments were carried out in acidotic media (pH 6.8). In the course of this type of experiment the same extent of Ca2+ overload as in experiments with non-acidotic media developed. Under metabolic reactivation the cells could recover energetically and could also recover from Ca2+ overload, but did not develop hypercontracture. Prevention of hypercontracture in this and other experiments can be attributed to the Ca2+ desensitizing effect of acidosis on the contractile machinery (Bing et al., 1973; Ventura-Clapier and Veksler, 1994). Media used for energy depletion of the cells were normally free of glucose in order to prevent glycolytic energy production. In one set of experiments in which the metabolic inhibitor KCN remained present for a prolonged period it was investigated
2118
A. Bonz et al.
whether administration of glucose would be sufficient to initiate signs of metabolic recovery (protocol D). In these experiments, however, neither a recovery of cytosolic Ca2+ control nor the development of energy-dependent hypercontracture was observed. This suggests that substrate chain phosphorylation is not sufficient for metabolic recovery. It also indirectly shows that the results of metabolic reactivation seen in experiments, where Ca2+ overload is reversed and hypercontracture develops, is due to the reactivation of oxidative energy production. In conclusion, the results of the present study demonstrate that myocardial cells depleted by more than 95% of their contents of high-energy phosphates are still able to reactivate oxidative energy production and use that energy for recovery from severe Ca2+ overload and reactivation of the contractile machinery. It was shown by experiments carried out in acidotic media (this study; Ladilov et al., 1995) that the unwanted consequence of metabolic reactivation in presence of cytosolic Ca2+ overload, i.e. hypercontracture, can be prevented by acidosis and ouabain. Thus, the results of the present study are consistent with the hypothesis stated by Reimer and Jennings (1986) that ischemic depletion of ATP is not necessarily related causally to irreversible cell damage, if cellular disruption can be avoided. Then, metabolic and structural survival can be achieved on the cellular level even though the high-energy phosphate contents had fallen far below the alleged “ATP threshold” of about 30% of the initial value. Consequently, the latter does not represent a true limit of reversibility. Restoration of high-energy phosphates and Ca2+ homeostasis do support this conclusion. If oxidative metabolism is restored and sarcolemmal rupture is avoided, severe energy depletion does not destroy the capacity of the Na+ pump to reestablish a normal sarcolemmal Na+ gradient and the Na+/Ca2+ exchanger to restore normal sarcoplasmic Ca2+ concentrations. It has not been tested if cells that have undergone a depletion of energy so severe would readily resume a normal mechanical function. In the isolated cell system this question cannot be addressed adequately. It is probably likely that these cells are markedly stunned, i.e. their mechanical performance in vivo is (reversibly) impaired for considerable time. It is also obvious that cell functions depending on high rates of energy consumption cannot be satisfactorily performed as long as purine reserves are deeply depressed (Spieckermann et al., 1980).
Acknowledgments This study was supported by the Deutsche Forschungsgemeinschaft (grant Si 618/1). The authors wish to thank Daniela Schreiber and Hermann Holztra¨ger for perfect technical assistance. A. Bonz is the recipient of a research fellowship (Gemetron Preis) of the Deutsche Gesellschaft fu¨r Thorax-, Herz- und Gefa¨ßchirurgie.
References A SC, G CE, 1991. Effects of 2,3-butanedione monoxime (BDM) on contracture and injury of isolated rat myocytes following metabolic inhibition and ischemia. J Mol Cell Cardiol 23: 1001–1014. B HU, 1974. Methods of Enzymatic Analysis. New York: Academic Press. B OH, B WW, M JV, 1973. Heart muscle viability following hypoxia: protective effect of acidosis. Science 180: 1297–1298. F A, F F, 1978. Effects of pH on the myofilaments and the sarcoplasmic reticulum of skinned cells from cardiac skeletal muscle. J Physiol (Lond) 276: 233–255. F MW, G PW, N IR, D AF, L GD, 1988. Paralysis of skeletal muscle by monoxime, a chemical phosphatase. Pflu¨gers Arch 411: 76–79. G AL, S B, 1956. The reaction of oxime with isopropyl methylphosphophonofluoridate (sarin). Chem Soc 756: 3887–3892. H C, W J, T F, B T, 1992. Effect of 2,3-butanedione monoxime on myosin and myofibrillar ATPases. An example of an uncompetitive inhibitor. Biochemistry 31: 12227–12232. I W, E¨ J, M W, D A, R H, C A, S J, H G, 1980. Increase in adenine nucleotides induced by adenosine: dosage, mode of application and duration, species differences. J Mol Cell Cardiol 12: 619–634. J RB, H HK, L JE, H ML, K S, R KA, 1978. Relation between high energy phosphate and lethal injury in myocardial ischemia in the dog. Am J Pathol 92: 187–214. J¨ E, K H, 1980. Rapid assay of adenine nucleotides or creatine compounds in extract of cardiac tissue by paired-ion reverse phase high performance liquid chromatography. Anal Biochem 102: 358–361. K M, W ML, M E, 1988. Acidosis during early reperfusion prevents myocardial stunning in reperfused ferret hearts. J Clin Invest 82: 920–927. K T, B D, I H, B WH, 1996. Survival of metabolically inhibited ventricular myocytes is enhanced by inhibition of rigor and SR Ca2+ cycling. Am J Physiol 271: H643–H650. K¨ W, S PG, 1970. Regulation of glycolysis in the ischemic and the anoxic myocardium. J Mol Cell Cardiol 1: 351–377. L YV, S B, P HM, 1995. Protection of reoxygenated cardiomyocytes against hypercontracture by inhibition of Na+/H+ exchange. Am J Physiol 268: H1531–H1539.
Recovery after ATP Depletion L AL, 1970. Mitochondria and calcium transport. Biochem J 119: 129–138. L OH, R NJ, F AL, R RJ, 1951. Protein measurements with the folin phenol reagent. J Biol Chem 193: 265–275. P HM, P I, H¨ JF, S PG, 1982. Culturing of calcium stable adult cardiac myocytes. J Mol Cell Cardiol 14: 397–412. R KA, J RB, 1986. Myocardial ischemia, hypoxia and infarction. In: Fozzard HA et al. (ed.) The Heart and Cardiovascular System. New York, 1133–1201. S VA, K VI, K VV, K AV, L VL, V VI, S VG, J SA, S EK, K C, 1989. Quantitative evaluation of relationship between cardiac energy metabolism and post ischemic recovery of contractile function. J Mol Cell Cardiol 21 (Suppl. I): 67–78. S B, K A, K T, S P, P HM, 1990. Sarcolemmal integrity and metabolic competence of cardiomyocytes under anoxia-reoxygenation. Am J Physiol 258: H285–H291. S B, K T, S P, P HM, 1991. Temporary contractile blockade prevents hypercontracture in anoxic-reoxygenated cardiomyocytes. Am J Physiol 260: H426–H435. S B, Z R, P HM, 1992. Recovery of anoxic-reoxygenated cardiomyocytes from severe Ca2+ overload. Am J Physiol 263: H1262–H1269. S B, L YV, P HM, 1994. Importance of sodium for the recovery of calcium control in reoxygenated cardiomyocytes. Am J Physiol 267: H506– H513. S PG, B¨ J, K¨ W, L B, B-
2119
HJ, 1969. Pra¨ischa¨mische Belastung und Wiederbelebungszeiten des Herzens [Preischemic stress and resuscitation time of the heart]. Verh Dtsch Ges Kreislaufforsch 35: 358–364. S PG, G MM, G¨ GG, K H, M VA, P CJ, S M, 1980. Functional alterations of cardiac subcellular structures during energy deficiency in relation to the metabolic state of the heart muscle cell. Basic Res Cardiol 75: 97–104. S MD, C AM, C MC, P DJ, L E, 1985. Direct observation of the “oxygen paradox” in single rat ventricular myocytes. Circ Res 56: 899–903. V J-LJ, J MF, B SR, 1994. The relative importance of myocardial energy metabolism compared with ischemic contracture in the determination of ischemic injury in isolated perfused rabbit hearts. Circ Res 74: 817–828. V-C R, V V, 1994. Myocardial ischemic contracture: metabolites affect rigor tension development and stiffness. Circ Res 74: 920–929. V C, C B, G D, F B, 1982. Adenine nucleotide content and regional function during ischemia and reperfusion in canine ventricular myocardium. Basic Res Cardiol 77: 645–655. Z Y, K M, 1994. BDM affects nucleotide binding and force generation steps of the cross-bridge cycle in rabbit psoas muscle fibers. Am J Physiol (Cell Physiol) 266: C437–C447. Z HG, T G, K H, G E, 1973. De novo synthesis of myocardial adenine nucleotides in the rat. Circ Res 32: 635–642.
Metabolic Recovery of Isolated Adult Rat Cardiomyocytes after Energy Depletion: Existence of an ATP Threshold? A. Bonz1, B. Siegmund2, Y. Ladilov2, C. F. Vahl1 and H. M. Piper2 1
Chirurgische Universita¨tsklinik, Abt. Herzchirurgie, Universita¨t Heidelberg, Im Neuenheimer Feld 110, D-69120 Heidelberg; 2Physiologisches Institut, Universita¨t Giessen, Aulweg 129, D-35392 Giessen (Received 19 February 1998, accepted in revised form 18 August 1998) A. B, B. S, Y. L, C. F. V H. M. P. Metabolic Recovery of Isolated Adult Rat Cardiomyocytes after Energy Depletion: Existence of an ATP Threshold? Journal of Molecular and Cellular Cardiology (1998) 30, 2111–2119. The question was investigated whether cardiomyocytes can be resuscitated after extreme energy depletion, i.e. after loss of ATP >70%. Isolated ventricular cardiomyocytes of the adult rat were exposed to metabolic inhibition with dinitrophenol and cyanide (DNP 0.2 m; KCN 2 m). After rapid energy depletion, cells were “reoxygenated” by wash-out of DNP and KCN. Intracellular calcium, cell length, ATP and creatine phosphate (CrP) of the cardiomyocytes were monitored. Metabolic inhibition resulted in a depletion of the stores of ATP and CrP by more than 95% of the normoxic values and caused a cytosolic Ca2+ overload. Parameters of metabolic recovery were: (i) resynthesis of CrP; (ii) recovery of a normal cytosolic Ca2+ control; and (iii) the elicitation of energy-dependent hypercontracture. “Reoxygenation”, i.e. wash-out of metabolic inhibitors, reactivated oxidative phosphorylation. Consecutively, CrP levels recovered to 76.0±7.3%, ATP levels recovered to 10.4±2.3% (means±.., n=10) of the initial normoxic values, a normoxic intracellular calcium level was re-established and hypercontracture was elicited. Prolongation of metabolic inhibition with KCN (2 m) or inhibition of the Na+/K+ pump with ouabain (0.5 m) disabled the cardiomyocytes to recover from cytosolic Ca2+ overload and prevented hypercontracture. It is concluded that even after extensive energy depletion metabolic resuscitation of the myocardial cell remains possible and a critical range of ATP for recovery, i.e. a “threshold” 1998 Academic Press of a 70% loss of ATP, does not exist. K W: ATP depletion; Metabolic inhibition; Cardiomyocytes; Irreversible cell injury; Anoxia/reoxygenation.
Introduction In ischemic or anoxic myocardium aerobic energy production ceases, and anaerobic processes of energy production are not sufficient to balance the demand. Consequently, cellular stores of high-energy phosphates are depleted. Eventually energydepleted myocardial cells develop irreversible cell injury. Systematic studies on energy metabolism and reversibility of ischemic myocardial injury have been started around 1970. Results of these studies
suggested a close relationship between the development of lethal injury in the ischemic myocardium and a considerable depletion of energy stores (Jennings et al., 1978). A concept has been developed, that a critical loss of high-energy phosphates initiates processes which finally result in irreversible cell damage. However, the causal relationship between energy depletion and manifestation of irreversibility of cell injury is not clear, because ischemic depletion of ATP is not necessarily related causally to irreversible cell damage. If the pathogenesis of consecutive events from energy
Please address all correspondence to: B. Siegmund, Physiologisches Institut, Justus-Liebig-Universita¨t, Aulweg 129, D-35292 Giessen, Germany.
0022–2828/98/102111+09 $30.00/0
1998 Academic Press
2112
A. Bonz et al.
depletion to ischemic cell death is interrupted, breakdown of high-energy phosphates even to a very low level need not end up with the lethal phase, i.e. with sarcolemmal disruption and necrosis (Reimer and Jennings, 1986). Attempts to determine a critical “threshold” of high-energy phosphate contents have indicated that ischemic-reperfused myocardium may fail to recover metabolically and functionally once the ATP contents have dropped below a critical threshold of approximately one third and creatine phosphate contents below approximately one tenth of the preischemic level (Spieckermann et al., 1969; Ku¨bler and Spieckermann, 1970). However, progression of ATP loss beyond two thirds of the normoxic value has not yet been disclosed as a sufficient cause of irreversible injury. Recent studies using isolated cardiomyocytes supported doubts that cellular metabolic recovery would be limited by such an “ATP threshold” of about 30% of the initial cellular ATP contents. It was shown, for example, that isolated myocardial cells retain metabolic competence and the ability to regain control over their cytosolic ion homeostasis in spite of a very prolonged time of energy deprivation (Saks et al., 1989; Siegmund et al., 1990, 1992). The present study was undertaken to explicitly address the question whether metabolic recovery and reactivation of energy-dependent processes are possible in myocardial cells after they had been depleted by more than 95% of their high-energy phosphate reserves. This degree of ATP loss exceeds by far the alleged critical “ATP threshold”. In the present study the capability of isolated cardiomyocytes to recover from extreme energy depletion was investigated. Cardiomyocytes were submitted to a very rapid and extensive depletion of creatine phosphate and ATP provoked by a respiratory uncoupler. Then, cardiomyocytes developed a Ca2+ overload. Finally, conditions for reactivation of oxidative energy metabolism were established and the synthesis of high-energy phosphates contents, the return to a normal cytosolic Ca2+ control, and activation of myofibrils were investigated.
Material and Methods Isolation of cardiomyocytes Ventricular heart muscle cells were isolated from adult male Wistar rats (200–250 g) as described previously (Piper et al., 1982), plated in medium
199 with 4% fetal calf serum on glass cover-slips or plastic culture dishes, which had been preincubated with this medium overnight. Four hours after plating the cover-slips were washed with medium 199. As a result of the wash, broken cells were removed, leaving a homogenous population of rod-shaped quiescent cardiomyocytes (>95%) attached to the cover-slips and culture dishes.
Fura-2 loading of the cells To measure the cytosolic Ca2+ concentration, cardiomyocytes were loaded with the fluorescent dye fura-2 at 37°C. The cells, attached to the coverslips, were incubated for 20 min in medium 199 with the acetoxymethylester of fura-2 (2.5 l). After loading, the cells were washed twice with medium 199 without fura-2. This washing step was followed by a 30-min post-incubation period in medium 199 to allow hydrolysis of the acetoxymethyl ester within the cell. The fluorescence from dye-loaded cells was at least 20 times higher than background fluorescence, i.e. fluorescence from cells not loaded with the dye. This loading protocol used was selected from several variations, because it provided the highest yield in fluorescence and minimal dye compartmentation (Ladilov et al., 1995).
Calcium and length measurements The cover-slip with the fura-2 loaded cells was introduced into an air-tight, temperature controlled (37°C), transparent perfusion chamber positioned in the light beam of an inverted microscope (Diaphot TDM, Nikon, Du¨sseldorf, Germany). Alternating excitation of the fluorescent dye at wavelengths of 340 nm and 380 nm was performed with an ARCation Measurement System adapted to the microscope (ISA, Grasbrunn, Germany). The emitted light (500–520 nm) from a 10×10 lm area within the fluorescent cell was collected by a photomultiplier of the system. Data were collected in the respective channels using an IBM PC/AT based data analysis system (model DM3000CM, ISA, Grasbrunn, Germany). Autofluorescence of the cells was measured prior to the loading procedure and was subtracted from the fura-2 signal. Since in the presented experiments the fura-2 signal saturates at a ratio value of 7, the measured maximum of the fura-2 signal indicates severe Ca2+ overload ([10−5 Ca2+). Simultaneous to the measurement of the cell’s
Recovery after ATP Depletion
fluorescence, the microscopic image was recorded with a video camera and stored on tape. Changes of the cell length were determined from these recordings. In the case of hypercontracted cells, the previous longitudinal axis was determined.
Experimental protocol The perfusion chamber (1 ml filling volume) placed on the microscope was perfused at a flow rate of 0.5 ml/min with modified, glucose-free Tyrode’s solution at 37°C containing (in m): 125.0 NaCl, 2.6 KCl, 1.2 KH2PO4, 1.2 MgSO4, 1.0 CaCl2, 25.0 HEPES; pH was 7.4 or 6.8, respectively. All experiments were started with an equilibration period by superfusing the cells with modified, glucose-free Tyrode’s solution (step 1). Then 50 m 2,3-butanedione monoxime (BDM) was added to the perfusion medium for 3 min (step 2) to prevent rigor shortening during the subsequent metabolic inhibition. This was followed by a period of 7 min of energy depletion (step 3). The Tyrode’s solution was supplemented with 50 m BDM, 0.2 m dinitrophenol (DNP), 4 m bis-(aminoethyl)-glycolether-N, N, N′, N′-tetraacetic acid (EGTA) and 20 m MgCl2. DNP was added in order to achieve a very rapid ATP depletion. EGTA, MgCl2 and BDM were added to reduce rigor shortening. In the next step (step 4), lasting 8 min, DNP was washed out with 1% (w/v) albumin and replaced by 2 m potassium cyanide (KCN). DNP was replaced by KCN in order to keep oxidative phosphorylation inhibited, but to make a fast metabolic recovery by wash-out of KCN possible. The following step (step 5) was varied in protocols A–E. In protocol A (control), the cells were allowed in step 5 to recover in Tyrode’s solution (pH 7.4) containing no metabolic inhibitors. Protocol B (acidosis) was as protocol A, but the pH of the medium was decreased to 6.8. In protocol C (continued metabolic inhibition), 2 m KCN remained present to prevent energetic recovery. In protocol D (continued metabolic inhibition with glucose), 2 m KCN remained present and 15 m glucose was added to allow glycolytic energy production. In protocol E (ouabain), 0.5 m ouabain was added to inhibit Na+/ K+-ATPase.
Determination of high-energy phosphates In separate experiments the state of energy of the cells was determined during chemical anoxia (0.2 m DNP + 2 m KCN in modified Tyrode’s
2113
solution) using cardiomyocytes plated on culture. The experiments were terminated by adding HCIO4 to the culture dishes (0.6 HCIO4 final concentration). Protein contents of the samples were determined in the acid precipitates according to the method of Lowry et al. (1951) using bovine serum albumin as standard. After neutralization, perchloric acid extracts were analysed for creatine phosphate (CrP) (Bergmeyer, 1974) and adenosine triphosphate (ATP) (Ju¨ngling and Kammermeier, 1980).
Statistical analysis All data are given as a representative example of n experiments or as mean values±.. of n independent culture preparations. Data were analysed according to Student’s t-test for non-paired samples. Differences with P<0.05 are regarded to indicate statistical significance.
Materials Falcon tissue culture dishes were obtained from Becton Dickinson (Heidelberg, Germany), medium 199 from Boehringer (Mannheim, Germany), fetal calf serum from GIBCO (Berlin, Germany), crude collagenase from Biochrom (Berlin, Germany), fura2/AM from Paesel and Lorey (Frankfurt/M., Germany) and 2,3-butanedione monoxime from Sigma (Deisenhofen, Germany). All other chemicals were purchased from Merck (Darmstadt, Germany) and of highest analytical grade.
Results State of energy during metabolic inhibition and metabolic reactivation Protocol A contained five steps. In the initial control situation cells were incubated in a normal Ca2+ containing medium (step 1) and a short preincubation with 30 m BDM (step 2). Then energy depletion was induced by the uncoupler of mitochondrial respiration DNP. DNP was applied in a Ca2+-free incubation medium supplemented with BDM and high Mg2+ to reduce rigor shortening (step 3). In the subsequent step 4, Ca2+ was repleted and DNP was eluted, with help of albumin, but oxidative energy production remained inhibited by the presence of KCN. In the last step 5, the initial
2114
A. Bonz et al.
Figure 1 Cellular contents of adenosine triphosphate (ATP) and creatine phosphate (CrP) during metabolic inhibition and recovery. Steps 1 and 2 of the protocol were without metabolic inhibition, steps 3 and 4 with metabolic inhibition, step 5 represents the recovery period. Details were as specified in Material and Methods. Data are means±.., n=10, where the 100% values correspond to 22.2±4.4 nmol/mg protein ATP and 23.0±4.9 nmol/mg protein CrP, respectively.
control conditions were re-established (“reoxygenation”) and metabolic recovery could take place. Under initial control conditions cardiomyocytes contained high levels of high-energy phosphates: ATP 22.2±4.4 nmol and CrP 33.0±4.9 nmol per mg protein (n=10). As shown in Figure 1, highenergy phosphates were rapidly degraded in presence of the uncoupler DNP (step 3). After only 1 min ATP content had declined to 14% and CrP below its detection limit (Ζ2% of control). After 3 min ATP was below 5%. Cells were left in that state of complete exhaustion of energy stores until end of step 3, i.e. for 15 min in total. When all inhibitors were removed in step 5 of the protocol, CrP contents recovered, reaching 44% of the initial control value within 10 min. Only a minor recovery of ATP contents was observed. To investigate whether metabolic recovery takes place under acidotic conditions or in the presence of ouabain CrP levels were measured (Fig. 2). CrP decreased below the detection limit during metabolic inhibition (MI) and recovered to about 55% of the initial normoxic control value within 30 min after wash-out of the inhibitor (protocol A). Resynthesis of CrP under acidotic conditions (pH 6.8)
Figure 2 Cellular contents of creatine phosphate (CrP) during metabolic inhibition and subsequent 30 min of reoxygenation. Ctr, control value before metabolic inhibition (i.e. after step 1); MI, metabolic inhibition (i.e. after step 4); A, 30 min recovery after wash-out of KCN (protocol A of step 5); B, 30 min recovery after wash-out of KCN under acidotic conditions of pHe 6.8 (protocol B of step 5); E, 30 min recovery after wash-out of KCN in presence of ouabain (protocol E of step 5). Details were as specified in Material and Methods. Data are means±.., n=10 (Ctr) and 5, where the 100% value corresponds to 33.0±4.9 nmol/mg protein CrP. Difference from A: ∗ P<0.0001.
was not different to normal pH conditions (protocol B) but was only minimal in the presence of ouabain (protocol E).
Experiments with metabolic recovery In each cardiomyocyte the changes in cell length and in the fura-2 fluorescence ratio, indicating changes in cytosolic Ca2+ concentrations, were monitored. Figure 3 shows a typical example. Starting from the initial slack length, the cell slightly shortened in step 3 of protocol A, i.e. during rapid energy depletion. Upon wash-out of metabolic inhibitors (step 5) the cardiomyocyte hypercontracted to about 25% of its initial length. The fura-2 ratio started to rise during step 4, i.e. in the energydepleted state where extracellular Ca2+ was again present. Upon removal of the inhibitor KCN (step 5) the fura-2 ratio gradually declined to the initial control value with an intermittent phase of oscillations. Protocol B differs from protocol A by a lower pH (6.8 instead of 7.4) in all media. Under these acidotic conditions the changes in fura-2 ratio and cell
Recovery after ATP Depletion
Figure 3 Protocol A (control). Time course of fura-2 ratio (original recording, arbitrary units) and cell length (O, percentage of resting length) in a single cardiomyocyte during metabolic inhibition and recovery. Steps 1 and 2 of the protocol were without metabolic inhibition, steps 3 and 4 with metabolic inhibition, step 5 represents the recovery period. Representative example of six separate experiments.
2115
Figure 5 Cell length of cardiomyocytes prior to metabolic inhibition (prior MI) and at end of protocols A–E. Data are means±.., n=6. Differences from MI: ∗ P<0.0001.
Figure 6 Fura-2 ratio of cardiomyocytes prior to metabolic inhibition (prior MI) and at end of protocols A–E. Data are means±.., n=6. Differences from MI: ∗ P<0.0001. Figure 4 Protocol B (acidosis). Time course of fura-2 ratio (original recording, arbitrary units) and cell length (O, percentage of resting length) in a single cardiomyocyte during metabolic inhibition and recovery. Steps 1 and 2 of the protocol were without metabolic inhibition, steps 3 and 4 with metabolic inhibition, step 5 represents the recovery period. The pH of the medium in step 5 was 6.8. Representative example of six separate experiments.
length in steps 3 and 4 of the protocol, i.e. during energy depletion, closely resembled the changes seen under protocol A (Fig. 4). When metabolic inhibition was removed (step 5) a marked difference compared to protocol A became apparent. In media with pH 6.8 cells did not develop hypercontracture. The recovery of the fura-2 ratio towards the control level proceeded more slowly than in media with pH 7.4. The intermittent oscillations seen in the recovery phase of protocol A remained absent.
Average data for all cells of cell length and fura-2 ratio at the end of step 5 are shown in Figures 5 and 6.
Experiments without metabolic reactivation In protocol C cardiomyocytes were submitted to the same conditions as in protocol A, except that in step 5 metabolic inhibition by KCN was not removed (Fig. 7). Under these conditions cells did not hypercontract nor did the fura-2 ratio recover. Comparison with protocol A indicates that the development of hypercontracture and recovery of the fura-2 ratio seen in step 5 of protocol A are a consequence of removing the preceding metabolic inhibition.
2116
A. Bonz et al.
Figure 7 Protocol C (continued metabolic inhibition). Time course of fura-2 ratio (original recording, arbitrary units) and cell length (O, percentage of resting length) in a single cardiomyocyte during metabolic inhibition and recovery. Steps 1 and 2 of the protocol were without metabolic inhibition, steps 3 and 4 with metabolic inhibition. In step 5 KCN (2 m) remained present. Representative example of six separate experiments.
Figure 9 Protocol E (ouabain). Time course of fura-2 ratio (original recording, arbitrary units) and cell length (O, percentage of resting length) in a single cardiomyocyte during metabolic inhibition and recovery. Steps 1 and 2 of the protocol were without metabolic inhibition, steps 3 and 4 with metabolic inhibition. In step 5 inhibitors of energy metabolism were absent, but ouabain (0.5 m) was added. Representative example of six separate experiments.
ouabain (0.5 m) was added in step 5 after washout of the respiratory uncoupler DNP in order to prevent the re-establishment of a large transsarcolemmal Na+ gradient which mediates the driving force of the Na+/K+ ATPase for extrusion of Ca2+ via the Na+/Ca2+ exchanger. Similarly to protocol D, the fura-2 ratio did not recover although the metabolic inhibitor KCN was not present (Fig. 9).
Discussion Figure 8 Protocol D (continued metabolic inhibition with glucose). Time course of fura-2 ratio (original recording, arbitrary units) and cell length (O, percentage of resting length) in a single cardiomyocyte during metabolic inhibition and recovery. Steps 1 and 2 of the protocol were without metabolic inhibition, steps 3 and 4 with metabolic inhibition. In step 5 KCN (2 m) remained present and glucose (15 m) was added. Representative example of six separate experiments.
It was tested in protocol D whether administration of a high concentration of the glycolytic substrate glucose (15 m) in step 5 could compensate in energy production for the continuation of mitochondrial inhibition by KCN (Fig. 8). But development of hypercontracture and recovery of the fura-2 ratio in step 5 remained absent in presence of glucose. In protocol E the inhibitor of Na+/K+ ATPase
The major finding of the present study is that metabolic resuscitation of myocardial cells remains possible after a [95% depletion of ATP and CrP reserves. Parameters of metabolic recovery were the resynthesis of CrP, the recovery of a normal cytosolic Ca2+ control and the elicitation of energydependent hypercontracture. Experiments were performed using the model of isolated cardiomyocytes. We have analysed in several previous studies the consequences of energy depletion and metabolic reactivation in a similar model by exposing the cells to periods of anoxia and reoxygenation (Siegmund et al., 1990, 1992; Ladilov et al., 1995). In the present study the uncoupler DNP was used to very rapidly exhaust cellular energy stores. Uncouplers activate mitochondrial ATP hydrolysis by dissipating the electrochemical gradient of the inner mitochondrial
Recovery after ATP Depletion
membrane. In presence of DNP more than 95% of high-energy phosphates was lost within 3 min. Loss of high-energy phosphates in isolated cardiomyocytes is normally accompanied by rigor shortening, leading to a 35% reduction of cell length. If that rigor shortening lasts for more than a few minutes it becomes irreversible, probably due to rearrangements in cytoskeletal structures (Stern et al., 1985). In the present paper BDM was used to reduce rigor shortening, because this substance is known to prevent ischemic contracture when applied in high concentrations (Armstrong and Ganote, 1991; Vanoverschelde et al., 1994; Koyama et al., 1996), possibly by affecting the binding constant of MgATP (Zhao and Kawai, 1994) or by interfering directly with myofibrillar ATPase as a non-competitive inhibitor (Herrmann et al., 1992). BDM may act as a chemical phosphatase (Green and Saville, 1956; Fryer et al., 1988), but dephosphorylation of Ca2+ regulating proteins which may cause some lasting effects of increased osmotic fragility (Armstrong and Ganote, 1991) or stunning (Koyama et al., 1996) was tolerated in the present experiments. In the presence of the contractile inhibitory agents BDM and Mg2+ (step 2) rigor shortening was greatly attenuated by performing rapid energy depletion in Ca2+-free media. After a virtually complete loss of ATP and CrP had been achieved, the uncoupler was eluted from the cell preparation while mitochondrial energy production remained inhibited by KCN, a blocker of cytochrome C oxidase. During this step (4) of the experimental protocol of metabolic inhibition the extracellular media contained Ca2+ and consequently the cytosolic Ca2+ level started to rise. It has been demonstrated before that Ca2+ accumulates in energydepleted cardiomyocytes by entering the cells from outside (Siegmund et al., 1992). When the metabolic inhibitor KCN was removed from the cell preparation (step 5), mitochondrial energy production was reactivated. This was demonstrated by the recovery of CrP levels which were restored to nearly 50% of control values within 10 min after removal of KCN. The absolute contents of ATP remained low since energy-depleted cardiomyocytes lose most of the purine contents rapidly and de novo synthesis of adenine nucleotides is very slow (Zimmer et al., 1973; Isselhard et al., 1980; Vial et al., 1982). During the phase of metabolic recovery cytosolic Ca2+ accumulation was reversed. Eventually, a normal cytosolic Ca2+ control was re-established. That this recovery of Ca2+ control is due to active metabolism was demonstrated by experiments in which
2117
the metabolic inhibitor KCN had not been removed. In these latter experiments recovery from Ca2+ overload did not occur. Recovery from Ca2+ overload was also impossible when the Na+ pump of the sarcolemma was inhibited by ouabain. Impairment of Na+ extrusion prevents the re-establishment of a normal Na+ gradient across the sarcolemma and consequently prevents Ca2+ extrusion via Na+/Ca2+ exchange. The Na+/Ca2+ exchange mechanism is an important route for Ca2+ extrusion after anoxic Ca2+ overload (Siegmund et al., 1994). Presence of ouabain also prevented recovery of CrP. The reason why CrP level did not renormalize remains unclear. A possible explanation might be that mitochondria are forced to take up Ca2+ on the expense of the mitochondrial electrochemical gradient, and therefore produce less ATP (Lehninger, 1970), when Ca2+ remains high during reoxygenation due to inhibition of Na+/Ca2+ exchange by ouabain. This indication of an impaired energy metabolism may then explain why cardiomyocytes did not hypercontract in absence of sufficient metabolic energy, i.e. when the contractile machinery is not activated by ATP. Cardiomyocytes with a manifest Ca2+ overload in the cytosol developed hypercontracture upon metabolic reactivation. This phenomenon has been analysed in several studies before (e.g. Siegmund et al., 1992; Ladilov et al., 1995). It has been shown that hypercontracture develops because of sustained myofibrillar activation when mitochondrial energy production is initiated in presence of a massive Ca2+ overload. It can be prevented by pharmacological or acidotic inhibition of the contractile apparatus (Fabiato and Fabiato, 1978; Kitakaze et al., 1988; Siegmund et al., 1991; Ladilov et al., 1995). Hypercontracture was also prevented in the present study when experiments were carried out in acidotic media (pH 6.8). In the course of this type of experiment the same extent of Ca2+ overload as in experiments with non-acidotic media developed. Under metabolic reactivation the cells could recover energetically and could also recover from Ca2+ overload, but did not develop hypercontracture. Prevention of hypercontracture in this and other experiments can be attributed to the Ca2+ desensitizing effect of acidosis on the contractile machinery (Bing et al., 1973; Ventura-Clapier and Veksler, 1994). Media used for energy depletion of the cells were normally free of glucose in order to prevent glycolytic energy production. In one set of experiments in which the metabolic inhibitor KCN remained present for a prolonged period it was investigated
2118
A. Bonz et al.
whether administration of glucose would be sufficient to initiate signs of metabolic recovery (protocol D). In these experiments, however, neither a recovery of cytosolic Ca2+ control nor the development of energy-dependent hypercontracture was observed. This suggests that substrate chain phosphorylation is not sufficient for metabolic recovery. It also indirectly shows that the results of metabolic reactivation seen in experiments, where Ca2+ overload is reversed and hypercontracture develops, is due to the reactivation of oxidative energy production. In conclusion, the results of the present study demonstrate that myocardial cells depleted by more than 95% of their contents of high-energy phosphates are still able to reactivate oxidative energy production and use that energy for recovery from severe Ca2+ overload and reactivation of the contractile machinery. It was shown by experiments carried out in acidotic media (this study; Ladilov et al., 1995) that the unwanted consequence of metabolic reactivation in presence of cytosolic Ca2+ overload, i.e. hypercontracture, can be prevented by acidosis and ouabain. Thus, the results of the present study are consistent with the hypothesis stated by Reimer and Jennings (1986) that ischemic depletion of ATP is not necessarily related causally to irreversible cell damage, if cellular disruption can be avoided. Then, metabolic and structural survival can be achieved on the cellular level even though the high-energy phosphate contents had fallen far below the alleged “ATP threshold” of about 30% of the initial value. Consequently, the latter does not represent a true limit of reversibility. Restoration of high-energy phosphates and Ca2+ homeostasis do support this conclusion. If oxidative metabolism is restored and sarcolemmal rupture is avoided, severe energy depletion does not destroy the capacity of the Na+ pump to reestablish a normal sarcolemmal Na+ gradient and the Na+/Ca2+ exchanger to restore normal sarcoplasmic Ca2+ concentrations. It has not been tested if cells that have undergone a depletion of energy so severe would readily resume a normal mechanical function. In the isolated cell system this question cannot be addressed adequately. It is probably likely that these cells are markedly stunned, i.e. their mechanical performance in vivo is (reversibly) impaired for considerable time. It is also obvious that cell functions depending on high rates of energy consumption cannot be satisfactorily performed as long as purine reserves are deeply depressed (Spieckermann et al., 1980).
Acknowledgments This study was supported by the Deutsche Forschungsgemeinschaft (grant Si 618/1). The authors wish to thank Daniela Schreiber and Hermann Holztra¨ger for perfect technical assistance. A. Bonz is the recipient of a research fellowship (Gemetron Preis) of the Deutsche Gesellschaft fu¨r Thorax-, Herz- und Gefa¨ßchirurgie.
References A SC, G CE, 1991. Effects of 2,3-butanedione monoxime (BDM) on contracture and injury of isolated rat myocytes following metabolic inhibition and ischemia. J Mol Cell Cardiol 23: 1001–1014. B HU, 1974. Methods of Enzymatic Analysis. New York: Academic Press. B OH, B WW, M JV, 1973. Heart muscle viability following hypoxia: protective effect of acidosis. Science 180: 1297–1298. F A, F F, 1978. Effects of pH on the myofilaments and the sarcoplasmic reticulum of skinned cells from cardiac skeletal muscle. J Physiol (Lond) 276: 233–255. F MW, G PW, N IR, D AF, L GD, 1988. Paralysis of skeletal muscle by monoxime, a chemical phosphatase. Pflu¨gers Arch 411: 76–79. G AL, S B, 1956. The reaction of oxime with isopropyl methylphosphophonofluoridate (sarin). Chem Soc 756: 3887–3892. H C, W J, T F, B T, 1992. Effect of 2,3-butanedione monoxime on myosin and myofibrillar ATPases. An example of an uncompetitive inhibitor. Biochemistry 31: 12227–12232. I W, E¨ J, M W, D A, R H, C A, S J, H G, 1980. Increase in adenine nucleotides induced by adenosine: dosage, mode of application and duration, species differences. J Mol Cell Cardiol 12: 619–634. J RB, H HK, L JE, H ML, K S, R KA, 1978. Relation between high energy phosphate and lethal injury in myocardial ischemia in the dog. Am J Pathol 92: 187–214. J¨ E, K H, 1980. Rapid assay of adenine nucleotides or creatine compounds in extract of cardiac tissue by paired-ion reverse phase high performance liquid chromatography. Anal Biochem 102: 358–361. K M, W ML, M E, 1988. Acidosis during early reperfusion prevents myocardial stunning in reperfused ferret hearts. J Clin Invest 82: 920–927. K T, B D, I H, B WH, 1996. Survival of metabolically inhibited ventricular myocytes is enhanced by inhibition of rigor and SR Ca2+ cycling. Am J Physiol 271: H643–H650. K¨ W, S PG, 1970. Regulation of glycolysis in the ischemic and the anoxic myocardium. J Mol Cell Cardiol 1: 351–377. L YV, S B, P HM, 1995. Protection of reoxygenated cardiomyocytes against hypercontracture by inhibition of Na+/H+ exchange. Am J Physiol 268: H1531–H1539.
Recovery after ATP Depletion L AL, 1970. Mitochondria and calcium transport. Biochem J 119: 129–138. L OH, R NJ, F AL, R RJ, 1951. Protein measurements with the folin phenol reagent. J Biol Chem 193: 265–275. P HM, P I, H¨ JF, S PG, 1982. Culturing of calcium stable adult cardiac myocytes. J Mol Cell Cardiol 14: 397–412. R KA, J RB, 1986. Myocardial ischemia, hypoxia and infarction. In: Fozzard HA et al. (ed.) The Heart and Cardiovascular System. New York, 1133–1201. S VA, K VI, K VV, K AV, L VL, V VI, S VG, J SA, S EK, K C, 1989. Quantitative evaluation of relationship between cardiac energy metabolism and post ischemic recovery of contractile function. J Mol Cell Cardiol 21 (Suppl. I): 67–78. S B, K A, K T, S P, P HM, 1990. Sarcolemmal integrity and metabolic competence of cardiomyocytes under anoxia-reoxygenation. Am J Physiol 258: H285–H291. S B, K T, S P, P HM, 1991. Temporary contractile blockade prevents hypercontracture in anoxic-reoxygenated cardiomyocytes. Am J Physiol 260: H426–H435. S B, Z R, P HM, 1992. Recovery of anoxic-reoxygenated cardiomyocytes from severe Ca2+ overload. Am J Physiol 263: H1262–H1269. S B, L YV, P HM, 1994. Importance of sodium for the recovery of calcium control in reoxygenated cardiomyocytes. Am J Physiol 267: H506– H513. S PG, B¨ J, K¨ W, L B, B-
2119
HJ, 1969. Pra¨ischa¨mische Belastung und Wiederbelebungszeiten des Herzens [Preischemic stress and resuscitation time of the heart]. Verh Dtsch Ges Kreislaufforsch 35: 358–364. S PG, G MM, G¨ GG, K H, M VA, P CJ, S M, 1980. Functional alterations of cardiac subcellular structures during energy deficiency in relation to the metabolic state of the heart muscle cell. Basic Res Cardiol 75: 97–104. S MD, C AM, C MC, P DJ, L E, 1985. Direct observation of the “oxygen paradox” in single rat ventricular myocytes. Circ Res 56: 899–903. V J-LJ, J MF, B SR, 1994. The relative importance of myocardial energy metabolism compared with ischemic contracture in the determination of ischemic injury in isolated perfused rabbit hearts. Circ Res 74: 817–828. V-C R, V V, 1994. Myocardial ischemic contracture: metabolites affect rigor tension development and stiffness. Circ Res 74: 920–929. V C, C B, G D, F B, 1982. Adenine nucleotide content and regional function during ischemia and reperfusion in canine ventricular myocardium. Basic Res Cardiol 77: 645–655. Z Y, K M, 1994. BDM affects nucleotide binding and force generation steps of the cross-bridge cycle in rabbit psoas muscle fibers. Am J Physiol (Cell Physiol) 266: C437–C447. Z HG, T G, K H, G E, 1973. De novo synthesis of myocardial adenine nucleotides in the rat. Circ Res 32: 635–642.