It is now appreciated that mitochondrial creatine kinase (CK,) may play an important role in heart high- energy phosphate metabolism and that this isozyme is.
THEJOURNAL OF BIOLOGICAL CHEMISTRY 198s hy The American Society of Biologxal Chemists, Inc.
Vol. 260. No. 1. Issue of Januaty 10, pp. 208-214.1985 Prtnted in U.S.A.
Creatine Kinaseof Heart Mitochondria T H E PROGRESSIVE LOSS OF ENZYMEACTIVITYDURING IN VIVO ISCHEMIA AND ITS CORRELATIONTODEPRESSED MYOCARDIAL FUNCTION* (Received for publication, August 8, 1984)
John A. BittlSg, Myron L. WeisfeldtS, and William E. JacobusSTII From The Peter Belfer Laboratory for Myocardial Research, $Department of Medicine and the VDepartment of Biological Chemistry, The Johns Hopkins Uniuersity, School of Medicine, Baltimore, Maryland 21205
It is now appreciated that mitochondrial creatine tionhas focused on identifyingmitochondrial respiratory kinase (CK,) may play an important role in heart high- defects arising during ischemia that limit the rate of highenergy phosphate metabolism and that thisisozyme is energy phosphate synthesis. Many laboratories have demonsolubilized in vitro by dilute solutions of Pi. Since an strated that mitochondria isolated from ischemic cells have increase in cellular Pi is known to occur with even decreased rates of electron transport andoxidative phosphobrief periods of myocardial ischemia, we investigated rylation, as well as reduced activities of specific components the relationshipbetween CK, activity andmyocardial of the cytochrome chain (4-12). Other studieshave illustrated performance in rabbit heartssubjected to total global ischemia. CK, activity is expressed as a ratio tomito- mitochondrial structural abnormalities (13-17). It has been chondrialmalatedehydrogenase (MDH,), a stable shown in liver that several of these defects induced during marker enzyme. A significant decline in this ratio wasischemia are potentiallyreversible (18, 19). It is less well appreciated that postischemic dysfunction observed after only 10 min of ischemia, a timeprior to of changes in total homogenate creatine kinase activity. might also involve changes in the intracellular integration energy After 60 min of ischemia, the CK,,JMDHm ratio was the high-energy phosphates.Suchintegration,or depressed by more than 70%. Since there was no res- transport, involving phosphocreatine hasbeen investigated in heart’s toration of activity following 30 min of reperfusion, a number of laboratories. Early studies correlating the content of the high-energy phosphates to contractility sugwe correlated changes in enzyme activity to contractile dysfunctionfollowing variableperiods of total is- gested the existence of two intracellular pools of ATP which chemia. The data showed a close correlation between were not in directcommunication. These pools were thought the decline in the CKJMDH, ratio and the reduction to be localized within the mitochondrial matrix and near the in performance, measured as left ventricular devel- myofibrils (20, 21). The shuttling of energy by phosphocreaoped pressure. No correlation was observed between tine between these loci presumably involved creatine kinase State 3 respiratory rates and performance.Using KC1 bound at specific sites (21). Recent 31-phosphorus nuclear arrest at27 “C or hyperthermic ischemia at 40 “C, the magnetic resonance saturation transfer studies (22, 23) have CK,/MDH, ratio consistently correlated to the degree of postischemic functional depression, independent of confirmed the nonequilibrium nature of the heart creatine the duration of ischemia. Isoenzyme electrophoresis kinase reaction, and have shown that the flux rates of heart CK, activity in the postischemic creatine kinase can be influenced by changes incardiac funcfailed to detect soluble in uiuo supernatant. Therefore, CK, activity appears to be tion (24, 25). Together,theseNMRresultslend altered rapidly and irreversibly by ischemia. The im- support to theenergy transport hypothesis. A critical enzyme involved in this pathwayis the mitochonplications of these observations on the integration of myocardial high-energy phosphate metabolism are dis- drial isozyme of creatine kinase (26) which accounts for 3040% of the creatine kinase activity in heart. It isnow estabcussed. lished that the enzyme resides on the exteriorsurface of the inner mitochondrial membrane. At this location, CK,’ appears to be functionally coupled to the adenine nucleotide The depression of cardiac performance afterischemia may translocase both for ATP supply (27-29) and ADP removal be related to persistent alterations in either the production or utilization of ATP. Since heart ATP generation is derived (30, 31). Our most recent kinetic studies have underscored almost exclusively from aerobic metabolism (3), much atten- the importance of this compartmentation for the effective coupling of phosphocreatine synthesis tooxidative phospho* This work was supported by United States Public Health Service rylation (32),as well as for the general control of heart oxygen Grants HL-20658 and P50 HL-17655 for the Specialized Center of consumption (33).Changes in thelocalization of this enzyme Research for Ischemic Heart Disease. Preliminary reports of a portion could potentially alter the integration of heart high-energy of this work have been published (1, 2). The costs of publication of phosphate metabolism by disrupting this coupling, and thus this article were defrayed in part by the payment of page charges. have a secondary impact onperformance. This articlemusttherefore be hereby marked“aduertisement” in Mukherjee et al. (34) have shown in uitro that conditions accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Current address: Cardiovascular Division, The Brigham and prevailing during ischemia can induce the release of CK, from Womans Hospital, Harvard Medical School, Boston, MA 02115. 11 Established Investigator of the American Heart Association during the time when this work was conducted, and author to whom reprint requests should be addressed at: 538 Carnegie Building, Division of Cardiology, The Johns Hopkins Hospital, 600 North Wolfe Street, Baltimore, MD 21205.
The abbreviations used are: CK,, mitochondrial creatine kinase; MDH,, mitochondrial malate dehydrogenase; LVDP, left ventricular developed pressure; LV dP/dt, the first derivative of the left ventricular pressure curve; EGTA, ethylene glycol bis(S-aminoethyl ether)N,N,N’,N’-tetraacetic acid.
208
Ischemic Changes in Heart ~ i t o c Creatine ~Kinase ~ r ~ ~
209
lation medium, 210 mM mannitol, 70 mM sucrose, 0.1 mM EGTA, 10 mM Tris-Cl (pH 7.4). Approximately 1.0 g of left ventricular tissue was minced using a Delepine tissue press with 1.0-mm holes. The mitochondrial fraction was isolated by the Nagarse method of Pande and Blanchaer (37), substituting EGTA for EDTA (26). The yield was 10-15 mg of protein/g, wet weight, of tissue. Mitochondria were resuspended to a protein concentration of 5.0 mg/ml. Stability of Marker Enzymes-An aliquot of mitochondria (50 pl) containing 0.25 mg of protein was added to either 0.5 ml of 115 mM KC1 + 5 mM KH2P04 (pH6.10) for a final pH of 6.12, or to 0.5 ml of 115 mM KC1 + 5 mM K2HP04 (pH10.1) for a final pH of 8.72. A t the start of the 30 "C incubation, samples wereremovedfor the analysis of total activity. After 40 min, the samples were centrifuged a t maximum speed for 1 min in a Beckman Microfuge. Supernatant solutions were decanted, and thepellets resuspended for the analysis of t.otal activity and recovery. EXPERIMENTAL PROCEDURES Enzyme Assays-The assay for creatine kinase was the spectroMaterials-Phosphocreatine, nucleotides, oxalacetic acid, glucose, photometric method of Eppenberger et al. (38), determined in the malic acid, hexokinase, and glucose-6-phosphate dehydrogenase were reverse direction. The rate of ATP formation was measured by a purchased from Sigma. Agarosewas obtained from Calbiochem- hexokinase/glucose-6-phosphatedehydrogenase method, and activity Behring. Nagarse was bought from the Enzyme Development Corp., calculated from the increase in absorbance accompanying the forNew York, and mannitol was the product of J. T. Baker. All other mation of NADPH, recorded a t 340 nm in a Gilford Model2400 chemicals wereof the highest purity commercially available. The spectrophotometer. The 3.0-ml assay medium contained 100 mM Trissolutions were prepared using deionized water. C1 (pH 7.4), 0.15 mM NADPf, 3.3 mM MgClz, 3.3 mM glucose, 0.5 Perfusion of Rabbit Hearts-Albino New Zealand rabbits were mM ADP, 8.3 mM phosphocreatine, 5 pug of hexokinase (140 IU/mg), anesthetized with 100 mg/kg pentobarbital injected intraperitoneally. 5 pgof glucose-6-phosphate dehydrogenase (140 IU/mg), and 1.33 Hearts were perfused within 45 s after removal using a modified mM AMP to inhibit adenylate kinase. The assay was conducted at Langendorff technique (35). The perfusate was Krebs-Henseleit bi- 30 "C with the reaction initiated by the addition of 12 pg of mitocarbonate buffer (pH 7.4) containing 118 mM NaCl. 4.7 mM KC1,1.3 chondrial protein or 15 pg of homogenate protein. Malate dehydromM CaC12,1.2 mM MgSO,, 1.2 mM KH2P04,25 mM NaHCOa, and genase activity was measured by the method of Ochoa (39), while 16.7 mM glucose, vigorously oxygenated with 95% 02, 5% CO2. The cytochrome oxidase activity was assayed according to Schnaitman aorta was cannulated and connected to a column with an overflow and Greenawalt, i40). Protein was quantitated by a biuret method 120 cm above the aortic valve. Extensive water-jacketing of the (41) with crystallized, nitrogen-standardized, bovine serum albumin apparatus ensured a constant temperature of 37 "C. The pulmonary as theprimary standard. artery was vented and 18-gauge AngioCath (1.5 cm) was pierced Isoenzyme Electrophoresis-Thin-layer agarose gels wereprepared through the apex into the left ventricle to permit drainage of non- 24 h prior to use by spreading 10.0 ml of 1% agarose in 60 mM barbital coronary flow. Hearts were paced a t 180 beats/min with a Grass SD9 buffer (pH 8.4) on 8 X 12-cm glass plates. The gels were electrophostimulator at a voltage 10% above threshold. Hearts were immersed resed 90 min at 4 "C (12 mA/plate), using 60 mM barbital buffer (pH ina heated reservoir containing perfusate at 37 "C. Total global 8.4). Plates were stained for malate dehydrogenase activity by overischemia was induced by clamping the perfusion line. When indicated laying with the nitro blue tetrazolium-phenazine methosulfate soluin Fig. 1, 1.0-g left ventricle samples were homogenized in 10 ml of tion described by Goldberg (42). The plates were incubated a t 37 "C 210 mM mannitol, 70 mM sucrose, 0.1 mM EGTA, 1.0 mM cysteine, for 1 h; theestimated sensitivity was1.0mIU. Creatine kinase and 10 mM Tris-C1 (pH 7.4) using loose and tight fitting Teflon isozymes (43) were detected by overla.ving the plates with 7.5 ml of homogenizers (26); these hearts were not used for measurements of 100 mM Tris-C1 (pH 7.4), 5.0 mMMgCIZ, 5.0 mM glucose, 1.0 mM performance. The homogenate protein concentration, creatinekinase ADP, 15.0 mM phosphocreatine, 3.0 mM AMP, 10.0 mM cysteine/ and malate dehydrogenase activities were determined i m m ~ i a ~ l y . HCl, 10 pg of hexokinase (140 IU/mg), 10 pg of glucose-6-phospha~ Hearts employed for initial m i t ~ h o n ~ istudies al were reperfused for dehydrogenase (140 IU/mg), 0.3 mM NADP, and incubated for 1h at 5 min prior to mitochondrial isolation. 37 "C.Enzyme activity was located by fluorescence and thesensitivity To measure cardiac performance, a 0.8-cm diameter, fluid-filled, was 1.0 mIU. latex balloon attached to a plastic cannula was secured in the left Statistics-Regression analysis and Student's ttest were conducted ventricle and connected to a Statham P23Db pressure transducer. according to Snedecor and Cochran (44). Comparison of regression Left ventricular isovolumic developed pressure-end diastolic volume relationships were obtained by increasing balloon volume in 0.1-ml analyses were performed according to Duncan (45). Standard statistical programs were used for the one-way analysis of variance. All crements from 0.0 to 0.7 ml, and recording LVDP, systolic pressurediasoolic pressure, and dP/dt at these stepwise changes. Recordings data reported in the figures, tables, and text, are given as the mean &
intact heartmitochondria. Thus, thegoal ofthis research was to determine the effects of in uiuo ischemia on CK, activity and evaluate the possible relationships between the activity of CK, and thepostischemic depression of maximal performance (16). The results show an almost immediate loss of CK, activity and a direct correlation to contractile abnormalities. These data arediscussed from the view that changes in CK, activity, resulting in an alteration of energy transport, may be one of the earliest mitochondrial defects induced by ischemia. They also suggest that a number of important biochemical events transpireprogressively, not suddenly, during the presumed "reversible" stages of ischemic cell damage.
were made on a Honeywell 1858 CRT Visicorder with an Accudata 143 bridge amplifier and 132 differentiator. Control performance was determined after a 15-min stabilization period. The mean control peak LVDP for the hearts was 106 + 3 mm Hg (n = 40, &S.E.). The perfusion line was then clamped to induce total global ischemia. During ischemia, the balloon was deflated, the bath maintained at 37 "C, and pacing continued. In the hyperthermic studies, the bath was warmed during ischemia to 40 "C, and pacing increased to 240 beats/min. KC1 arrest was induced by flushing the ischemic coronary arteries with 5.0 ml of ice-cold solution containing 90 mM NaCI, 37 mM KCI, 24 mM NaHC03,and 35 mM glucose at pH 7.4 (36). Immediately following arrest, the balloon was deflated and thehearts cooled to 27 "C. At designated intervals, the clamp was released and reflow occurred. Since 25% of the hearts fibrillated during reperfusion, a defibrillator set at 2 W-s was used to restore normal pacing rhythm. Aft.er30 min of reperfusion at 37 "C, peak LVDP, maximum dP/dt, and rates of flow were again measured in all hearts. I t should be noted that in a separate series of hearts subjected to 30 min of ischemia, ventricular performance recovered and stabilized after only 10 min of reflow, and remained constant during an additional 50 min of perfusion. After 30 min of reperfusion, postischemic hearts were used for mitochondrial isolations. ~ - Hsubmerged e a r t s in ice-cold isoIsolation of ~ ~ ~ o c ~ o ~ ~ r ~ were
S.E.
RESULTS
Selection of ~ i t o c ~ ~ dMarker r ~ a ZEnzymes-There is some normal variability in the specific activity of creatine kinase in isolated mitochondrial fractions (26,46). These differences probably result from nonmitochondrial protein contamination. Since postischemic preparations could be more severely contaminated, we sought a stable mitochondrial enzyme for the normalization of CK, activity. The activity ratio of CK, to other mitochondrial enzymes is relatively constant (46). Table I presents data for the stability of cytochrome oxidase and MDH,, classic marker enzymes, under incubation conditions known to solubilize CK, (26, 34, 46). The results confirm that MDH, and cytochrome oxidase remained associated with t.he mitochondrial pellet when 50-70% of the CK, activity wasreleased. A slight decrease in the recoveryof cytochrome oxidase activity was observed at pH 6.1, from 0.90 to 0.78 IU/mg of protein. In contrast, only 2% of the MDH, activity was detected in the supernatant and the recovery was
210
Ischemic Changes in Heart Mitochondrial Creatine Kinase
TABLE I Distribution of creatine kinase, malate dehydrogenase, and cytochrome oxidase in phosphate-incubatedrabbit heart mitochondria Intact rabbit heart mitochondria were incubated in KCl-KPi medium under conditions detailed under “Experimental Procedures.” The values are the mean of 4 experiments f S.E. Enzyme activities are expressed in international units. Control, total
40-min incubation Supernatant Pellet
pH 6.12 Creatine kinase Malate dehydrogenase Cytochrome oxidase
0.34 f 0.05 0.07 f 0.04 0.09 & 0.03 2.45 f 0.07 2.46 f 0.04 0.05 & 0.05 0.90 f 0.11 0.73 f 0.03 0.05 f 0.04
pH 8.72 Creatine kinase Malate dehydrogenase Cytochrome oxidase
0.34 f 0.05 0.05 f 0.04 0.14 f 0.03 2.43 & 0.15 2.42 f 0.18 0.06 f 0.04 0.89 f 0.04 0.77 f 0.84 0.05 f 0.04
- 1
v
a W
I
v)
a
z
20-
W
zl-
a W
consistently near 100%. These data suggested that MDH, was more stable, and thus thepreferred enzyme for normalization. In all mitochondrial studies CK, activity was divided by MDH, activity, and presented as a CKJMDH, ratio. Isoenzyme Contaminations-In control experiments, malate dehydrogenase (42) andcreatine kinase (43) isozyme electrophoresis was performed on mitochondria isolated from normal and ischemic hearts. The purpose was to assess the contamination of thesepreparations by nonmitochondrial isozymes. Mitochondria were diluted in 100 mM K2HP04 (pH 8.0) and 10 mM cysteine to enzyme activities up to 20 mIU/ pl. The samples were applied to plates as 2-4 aliquots, electrophoresed, and stained. Only mitochondrial forms of the isozymes were detected in either normal or ischemic mitochondrial fractions. Since the sensitivity of both methods was 1 mIU, the contamination by other isozymes was apparently less than 2.5%. Stability of Total Tissue HomogenateCreatine Kinase Activity during Ischemia-The release of creatine kinase from the myocardium has been used as anindex of irreversible ischemic cell damage. We therefore determined the activity of creatine kinase in heart homogenates during global ischemia (Fig. 1). Fig. L4 shows that thespecific activity of creatine kinase has a rather large variance, with average values ranging from 3.0 to 4.5 IU/mg of protein. Even after 120 min of global ischemia, there was no statistically significant decrease in homogenate activity. This time course was similar to that reported by Braasch et al. (47). When homogenate creatine kinase activity was normalized to total malate dehydrogenase activity (Fig. lB), the variance at each time point was reduced. However, little decrease was noted in the ratio of these enzymes, and only at 120 min was a significant reduction detectable from control (p < 0.02). Loss of Mitochondrial Creatine Kinase during Total Global Ischemia-To determine if the activity of CK, was altered during ischemia, a group of 24 hearts was subjected to periods of global ischemia ranging from 0 to 60 min followed by 5 min of reperfusion. Mitochondrial fractions were subsequently isolated, and the CK, and MDH, activities determined, Fig. 2. CK, specific activity (Fig. 2 A ) declined steadily during 1 h of ischemia, with significant differences ( p < 0.005) seen at 45 and 60 min. MDH, activity (Fig. 2B) did not significantly change. When the enzyme data were normalized on a preparation to preparation basis (Fig. 2C), the CKJMDH, ratio showed a significant decrease at 10 min of ischemia (p < 0.02). As a result of the normalization process, we now were able to detect very early reductions in CK, activity. Brief reperfusion after ischemia could reduce CK, because
[L
V -I
F
P
o
0
30
90
60
DURATION OF ISCHEMIA
I20
(min)
B.
(r
o
‘
0
”
’
30
~
”
60
’
’
90
120
DURATION OF ISCHEMIA (mln)
FIG. 1. Measurement of total creatine kinase activity in the ischemic rabbit heart. Five isolated, perfused rabbit hearts were subjected to total global ischemia at 37 “C for 120 min by crossclamping the aortic input line. Biopsy samples (0.5 g) were taken at 0, 15, 30, 60, and 120 min of ischemia. The tissue samples were prepared as described under “Experimental Procedures,” with protein and enzyme activities measured immediatek. A, total homogenate creatine kinase activity; €3,homogenate creatine kinase activity normalized to total homogenate malate dehydrogenase activity.
of enzyme washout (48). In control experiments, hearts were subjected to 30 min of global ischemia, then reperfused for 30 min. In these hearts, the total homogenate creatine kinase activity was 3.9 IU/mg of protein, a value similar to control. In 4 other heartssubjected to 30 min of total global ischemia and reflow, we collected the first 5.0 ml of coronary effluent. The samples contained only 40 mIU of creatine kinase activof the heart’s total ity, which represented less than 1 X activity. Transmembrane leakage cannot account for the observed changes in CK, activity. Irreversible Loss ofCK, Activity during Ischemia and Reperfusion-To test the reversibility of CK, loss in vivo, control hearts were perfused for 75 min at normal flow rates. In mitochondria isolated from thesehearts, the CK,/MDH, ratio was 0.136f 0.001, Table 11, quite close to thepreischemic control value reported in Fig. 2C. Other hearts weremade ischemic for 30 min with no reflow.In these, the CK,/MDH, ratio was significantly depressed ( p < 0.005) by 30% to a value of 0.096 f 0.003. Finally, 4 other hearts were subjected
Ischemic Changes Heart in
CKm
r-A
1.8
m
.
1s
I
Mitochondrial Creatine Kinase .I8
f
c_
CKm/MDHm RATIO
.IS
0
0
211
.I2
-
.09
-
.06
-
0
-
0.3
*
.03
r=0.84 J
0
20
40
60
I
0
20
60
40
TIME (mid FIG. 2. Mitochondrial enzyme activities during total global ischemia. Twenty-four isolated, perfused rabbit hearts were subjected to global ischemia for 0, 10, 20, 30, 45, and 60 min. Creatine kinase and malate dehydrogenase activities were measured in mitochondrial fractions isolated from each heart. A , mitochondrial creatine kinase activity, IU per mgof mitochondrial protein; B , mitochondrial malate dehydrogenase activity, international units per mgof mitochondrial protein; C , the ratio of mitochondrial creatine kinase activity normalized to the malate dehydrogenase activity from the same mitochondrial preparation. In all panels the lines were drawn by linear regression, and the r values are at thelower left corner.
TABLE I1 Effects of ischemia and reperfusion on the CKJMDH, ratio Control hearts were normally perfused for 75 min. Experimental hearts were stabilized for 15 min prior to totalglobal ischemia. Reflow was initiated by releasing the clamp. Hearts were removed from the apparatus, cooled, and themitochondrial fractions isolated from each heart. The activities of creatine kinase and malate dehydrogenase were immediately determined as described under "Experimental Procedures." Conditions
1. Perfusion (75 min) 2. ischemia (30 min) with no reperfusion 3. Ischemia (30 min) with reperfusion (30 min) a NS, not significant.
n
4 4 4
CKJMDHm
pvalue
+
W [r
3 v) v)
120
0 W
100
a
9 Y
0.136 0.002 0.096 + 0.003 0.005 (1us 2)
W
0.091 + 0.007 NS" (2 us 3)
J 3
to 30 min of ischemia, followed by 30 min of reperfusion prior to mitochondrial isolation. Under these latterconditions, CK, activity was not restored to normal, but remained reduced at 0.091 -t 0.007. The data show that the loss of CK, activity was irrevrsible in nature during brief periods of reflow. Contractility Measurements in Isovolurnic Rabbit HeartsThe major goal of this study was to determine the relationships between postischemic reductions in energy demand, expressed as contractile performance (16), and alterations in the phosphocreatine energy transport pathway, estimated by the loss ofCK,. To achieve this it was necessary to assess maximum contractile performance. Other researchers (49) have shown a correlation between peak developed pressure and maximum O2consumption in Langendorff paced hearts. Therefore, we monitored contractility by determining performance at thepeak of the developed pressure-end diastolic volume curve (Fig. 3). It is well knownthat ischemia depresses myocardial contractility, which for a limited period of reperfusion appears to be a persistent defect. Fig. 3 shows that peak cardiac performance deteriorated by approximately 14% after only 10 min of total global ischemia. After 60 min of ischemia there was an 88% depression in cardiac performance, with intermediate values at other times. Since ventricular performance was 95 f 3% after 90 min of perfusion at normal flow rates, the functional alterations observed in Fig. 3 rep-
DURATION OF ISCHEMIA
g a
n
-
80 -
s!W
40 20 -
t
-
(L
a
9
a
> W
J
60
20 MIN
30 MIN
,* 45 MIN 60
00
0.2
04
06
0.8
VENTRICULAR VOLUME ( m l )
FIG. 3. Pressure-volume curves of control and postischemic peak hearts. Left ventricular developed pressure was measured at the of the pressure-volume curve. The control curves were generated during the preischemic period as outlined under "Experimental Procedures." Hearts were then made totally ischemic at 37 "C for 0, 10, 20, 30, 45, and 60 min by clamping the perfusion line. Following a 30-min period of reperfusion, postischemic pressure-volume curves were again determined (mean f S.E., n = 4 at each time).
resent damage induced by the ischemic protocol. Relationships between CK, Activity and Cardiac Performances-Since 30 min of reflow did not restore CK, activity (Table 11), the correlations between CK, and performance were evaluated (Fig. 4). In Fig. 4A the CKJMDH, ratio was plotted versus function, calculated as per cent control peak LVDP. The data show a veryclose correlation ( r = 0.97) between these two parameters. The reduction of maximum dP/dt, expressed as per cent control dP/dt, also closely correlated ( r = 0.95) to theloss of CK, activity (Fig. 4B).A more common index of ischemic mitochondrial damage has been the depression of State 3 respiration. In data notshown, there was no relationship between the rates of State 3 and contractile depression. Overall, the data inFigs. 2 and 4 suggest
TABLE I11 Mitochondriul creatine kinase activity and ventricular function after interventions Isolated, perfused rabbit hearts were stabilized for 15 min, and LVDP measured at thepeak of a pressure-volume curve. Hearts were arrested with cold KC1 cardioplegic solution, and maintained at 27 "C during the arrest period. These hearts were not paced. Another group of hearts was ischemic at 40 "C and paced at 240 beats/min. Peak LVDP was measured during the reperfusion period, and calculated as a per centof the control performance of each heart.Four hearts were tested per group, and the values are expressed as mean +. S.E. Conditions
020
040
060 080 I00 CK,/MDH, RATIO
I20
140
1. Control 2. Ischemia (10 min) a t 37 "C 3. Ischemia (30 min) a t 27 "C with KC1 arrest 4. Ischemia (10 min) at 40 "C 5. Ischemia (30 min) a t 37 "C 6. Ischemia (120 min) at 27 "C with KC1 arrest
CK,/MDH,
+ + 0.101 + 0.003
0.136 0.002 0.119 + 0,007 0.118 0.003
0.091 + 0.007 0.086 + 0.003
9; of control 96 peak ratio
LVDP
100
88 87
100 86 + 5 89 + 2
74 74 55 67 63
+3 +3 59 + 7
."
ORIGIN I
I
6.
FIG. 5. The fate of mitochondrial creatine kinase in the ischemic rabbit heart. Creatine kinase isoenzyme electrophoresis FIG.4. The relationshipbetween mitochondrial creatine kiperformed on soluble fractions of whole heart homogenates from nase activity and ventricular function in rabbit hearts after was and 30-min ischemic hearts. All homogenates were prepared 37 "C total global ischemia. Isolated, perfused rabbit hearts were control in 10 volumes of 210 mM mannitol, 70 mM sucrose, 1.0 mM cysteine,
stabilized for 15 min; pressure-volume curves and maximum dP/dt were measured. Total global ischemia was induced at 37 "C for 0, 10, 20,30, 45, and 60 min. Peak LVDP and maximum dP/dt were again measured following 30 min of reperfusion, and expressed as a per cent of control peak developed pressure ( A ) ,or control maximum dP/ dt ( B ) for each heart. After 30 min of reperfusion, the mitochondrial fractions were isolated and assayed for creatine kinase (CK,) and malate dehydrogenase (MDH,) activities. A, regression for per cent control peak LVDP uersuS the CK,/MDH, ratio: per cent peak LVDP = (930 X CK,/MDH,) - 28; r = 0.97. I3, regression for per cent maximum dP/dt uersm the C K ~ / M D Hratio: ~ per cent maximum dP/dt = (948 X CK~/MDH,) - 28; r = 0.95: n = 22.
that themeasurement of CK, activity in mitochondrial fractions may provide important bioenergetic information about myocardial ischemic deterioration. Other Ischemic Interventions-The data presented in the previous section report enzyme changes as a function of the duration of ischemia. It is also possible to change the "severity" of global ischemia, and thusmodify ischemiccell damage. Such approaches could include 40 "C hyperthermic ischemia to increase the rateof damage, or KC1 arrest at 27 "C to retard these effects (50). When hearts were subjected to 40 "C hyperthermic ischemia, the magnitude of enzyme loss was greater than that observed for comparable times at 37 "C, Table 111. On the other hand, 30 min of KC1 arrest a t 27 "C
0.1 mM EGTA, and 10 m M Tris-C1 (pH 7.4). One-half of the control homogenate was incubated in 100 mM KlHPOd (pH 8.0), 0.5%deoxycholate, and 1.0 mM cysteine at 20 "C for 8 min. Soluble fractions of all homogenates were prepared by centrifugation at 37,000 X g for 20 min and diluted in homogenizing mediumto final enzyme activities of 1.0 and 10.0 IU/ml. Aliquots, 2 pl, were plated in the agarose gel wells, electrophoresed, and stained as previously described (46). Slot 1:phosphate-deoxycholate control supernatant fraction, 1IU/ml; slot 2 phosphate-deoxycholate control supernatant fraction, 10 IU/ml; slot 3 control heart supernatant fraction, 1 IU/ml; slot 4 control heart supernatantfraction, 10 IU/ml; slot 5: ischemic (30 min) heart s u p e r n a ~ n t1, IU/ml; slot 6 ischemic (30 min) heart supernatant, 10 IUlml.
preserved CK, activity relative to normothermic (37 "C) ischemia. In fact, 120 min of KC1 arrest was required to detect enzyme and functional changes similar to those of normothermic 30-min ischemia. Under these conditions, data not shown, the correlation between CK, loss and contractile depression was quite similar to thatseen in Fig. 4.Therefore, the CK,/MDH, ratio is more than a measure of ischemic duration since it consistently monitors the extent of damage under variable time condition. ~ ~ Kinase ~ r in~ the~ I s e Fute of M ~ t o Creatine Heart-While the previous data documented changes in CK,
~
~
~
Ischemic Changesin Heart Mitochondrial Creatine Kinase
213
activity, the mechanismof enzyme loss was undefined. Since transport, and be expressed as decreased performance. Howbe taken with some caution. dilute solutions of inorganic phosphate (5-20 mM) solubilize ever, such an interpretation must CK, (26, 34, 46), we investigated for CK, release into the For example, we have not ruled cut thepossibility that ischesarcoplasmicfraction. Isoenzyme electrophoresis was per- mic induced changesinthe regulation of the actomyosin formed on the 37,000 x g supernatant fractions from control ATPase may occur, although there are reports that such is and 30-minischemic hearts. It was assumed thatif CK, were not the case(57). Likewise, we now appreciate that there are solubilized and remained active, it would be detected in the persistent changes in the ATP content of the myocardium supernatant fraction. In control experiments, heart tissue was thatcorrelatewith regional contractile depressionwhen homogenized in 100 mM potassium phosphate, pH 8.0, con- hearts aresubjected to short-termischemia (58,59). Itis clear taining 0.5% deoxycholate and 1.0 mM cysteine. This resulted that a number of potentially damaging events transpire during in the release of active CK,, Fig. 5, slots 1 and 2. Slots 3 and ischemia to summate as contractile depression. It would there4 show that noCK, activity was detectable in the supernatantfore be inappropriate to overinterpret our results consider and fraction from normal homogenates. In postischemic hearts, any single event “responsible”forpostischemic contractile we were not able to detect enzyme activity in the soluble failure. fraction, slot 5, even when the gel was overloaded, slot 6. The As a final point, for over two decades investigators have changes in CK, activity documented in this article must be successfully defined a sequenceof events occurring during the explained by mechanisms other than simple release. Such progression to irreversible myocardial damage.Cell death and (51), spontaneous necrosis aretheendresult. possibilitiesincludeenzymeaggregation Because the morphological denaturatim, or degradationby proteases (52-54). changes canquickly disappear upon reperfusion, pathological studies suggested that up tosome transition point, ischemicDISCUSSION induceddamage may be rapidly and completelyreversible (16). The extension of this work has led to the concept of a T o appreciate the integration occurring between high-en“point of no return” transition between reversible and irreergy phosphate production and contractile utilization, one versible damage (7), implying that such a change is abrupt. must consider the general metabolic restrictions placed on the However, in spite of the fact that cell morphology may be normal heart. The heart is rather unique in that it utilizes normal, myocardial performance deterioratesin away directly more than 90% of the metabolically generated ATP for a related to the duration of ischemia (compare Fig. 3 uersus single, well-defined purpose. This is toenergize the reactions Figs. 1and 2 in Ref. 16). More recently it has beenrecognized involved in the excitation-contraction-relaxation cycle (3). that there areregions in a previously ischemic zone that have While ATP is produced by glycolysis and mitochondrial oxinormal morphology but depressed function. This has led to dative phosphorylation, glycolysis usually accounts for less the concept of an ischemic induced “stunned” myocardium than 10% of the net ATP, with 90% coupled to oxidative (60). Since both our biochemical (Fig. 2) and functional (Fig. phosphorylation. Kobayashi and Neely (55) have most care3) dataprogressively degenerate, the results presented in this fully documented this balance by measuring the ratesof ATP article suggest that “reversible ischemia”may be characterized production under varying conditions of work stress. At normal by a definable sequenceof events occurring in a time-dependwork loads, the flux of ATP from glycolysis was 4 pmol/min/ ent manner. In other words, the transition to cell death is g of tissue, while during maximum work this increased to12 gradual, and not an allor none event. This importantrecogpmol/min/g, the same asobserved during anoxia. For oxidanition now opensthe doorformore detailed biochemical tive phosphorylation, the rates of ATP synthesis were norstudies of the subtle changes induced during the early, and mally 60 pmol/min/g, which peaked at 120 pmol/min/g. From the standpointof our results, itis important torealize potentially correctablereversible phase.The sensitivityof the CKJMDH, ratio to the duration and severity of ischemia that during veryhigh work conditionstherates of ATP suggests that this approach may provide useful information productioncanapproachthetheoretical metaboliclimits. about cell damage in normally appearing tissue. Glycolytic production is restrictedby a fixed enzyme content, with hexokinase having the lowest specific activity, 7 IU/g Acknowledgments-We acknowledge Dr.John T. Flaherty for tissue. Since 2 ATPs are synthesized per mol of phosphorylmany helpful suggestions during thedesign of these experiments. We ated glucose, the upper limitwould be 14 pmol of ATP/min/ thank Edward Grunwaldfor expert technical assistance. g, quite close to the measured value of 12. With respect to oxidative phosphorylation, Scarpa and Graziotti (56) reported REFERENCES that the mitochondrial contentof the rat heart was 81 mg/g 1. Bittl, J., and Jacobus, W. E. (1979) J. Mol. Cell. Curdiol. l l b , of tissue. Isolated rat heart mitochondriarespire in State3 at Suppl. 1 , 8 a rate of 500 ng atomsof oxygen/min/mg. Assuming an ADP/ 2. Bittl, J., and Jacobus, W. E. (1979) Clin. Res. 27,154a 0 ratio of 3.0, ATP production would be 1.5 pmol/min/mg, 3. Neely, J. R., and Morgan, H. E. (1974) Annu. Reu. Physiol. 36, or 121.5/g of heart.This value isstrikingly close to the 413-459 4. Jennings, R. B., Herdson, P. B., and Sommers,H. M. (1969) Lab. observed rate of 120 pmol/min. In other words, during stress Invest. 20,548-557 the rateof ATP utilization almost reaches the maximum rate 5. Schwartz, A., Wood, J. M., Allen, J. C., Bornet, E. P., Entman, of production. For these reasons, investigators recognized that M. L., Goldstein, M. A., Sordahl, L. A., and Suzuki, M. (1973) derangements in aerobic energy metabolism could limit ATP Am. J. Curdiol. 32,46-61 production andpossibly contractile performance in the postis-6. Ziegelhoffer, A. Kostolaysky, S., and Fedelesova, M. (1974) J . chemic state. This is also why we assessed myocardialfunction Mol. Cell. Cardiol. 6,567-574 7. Trump, B. F., Mergner, W. J., Kahng, M. W., and Saladino, A. under maximal conditions. Under normal conditions,modest J. (1976) Circulation 53,Suppl. I, 17-26 derangements in mitochondrial metabolism might not be de8. Jennings, R. B., and Ganote, C. E. (1976) Circ. Res. 38,Suppl. I, tectable since there is an apparent2-fold metabolic reserve. 80-91 The data presented in the report are consistent with the 9. Wood, J. M., Handey, H. G., Entman, M. L., Hartley, C. J., idea thatchangesin localization of a criticalintegrative Swain, J. A., Busch, U., Chang, C., Lewis, R. M., Morgan, W. enzymemay affect the fidelity of phosphocreatine energy J., and Schwartz, A. (1979) Circ. Res. 44,52-61
214
Ischemic Changes
Heart in
Mitochondrial Creatine Kinase
10. Mergner, W. J., Chang, S. H., Morzella, L., Kahng, M.W., and Trump, B. F. (1979) Lab. Inuest. 4 0 , 686-694 11. Toleikis, A., Dzeja, P., Praskevicius, A., and Jasaitis, A. (1979) J . Mol. Cell. Cardiol. 1 1 , 57-76 12. Rouslin, W., and Millard, R. W. (1980) J. Mol. Cell. Cardiol. 1 2 , 639-645 13. Jennings, R.B., and Ganote, C. E. (1974) Circ. Res. 3 4 , Suppl 111, 156-172 14. Jennings, R. B., Ganote, C. E., and Reimer, K. A. (1975) Am. J. Pathol. 8 1 , 179-198 15. Hagler, H. K., Sherwin, L., and Buja, L.M. (1979) Lab. Invest. 40,529-544 16. Schaper, J., Mulch, J., Winkler, B., and Schaper, W. (1979) J. Mol. Cell. Cardiol. 1 1 , 521-541 17. McAllister, L. P., Trapukdi, S., and Neely, J. R. (1979) J. Mol. Cell. Cardiol. 11,619-630 18. Mittnacht, S., Jr., Sherman, S. C., and Farber, J. L. (1979) J. Biol. Chem. 254,9871-9878 19. Mittnacht, S., Jr., and Farber, J. L. (1981) J. Biol. Chem. 2 5 6 , 3199-3206 20. Pool, P. E., Covell, J. W., Chidsey, C.A., and Braunwald, E. (1966) Circ. Res. 19, 221-229 21. Gudbjarnason, S., Mathes, P., and Ravens, K. G. (1970) J . Mol. Cell. Cardiol. 1 , 325-339 22. Brown, T. R., Gadian, D. G., Garlick, P. B., Radda, G. K., Seeley, P. J., and Styles, P. (1978) in Frontiers of Biochemical Energetics (Dutton, P. L., Leigh, J. S., and Scarpa, A., eds) Vol. 11, pp. 1341-1349, Academic Press, New York 23. Nunnally, R. L., and Hollis, D. P. (1979) Biochemistry 18,36423646 24. Miceli, M. V., Hoerter, J. A., and Jacobus, W. E. (1983) Biophys. J. 41,249a 25. Ingwall, J. S., Kobayashi, K., and Bittl, J. A. (1983) Biophys. J. 41, la 26. Jacobus, W. E., and Lehninger, A.L. (1973) J. Biol. Chem. 2 4 8 , 4803-4810 27. Saks, V. A., Lipina, N. V., Smirnov, V.N., and Chazov, E. I. (1976) Arch. Biochem. Biophys. 1 7 3 , 34-41 28. Yang, W. C. T., Geiger, P. J., Bessman, S. P., and Borrebaek, B. (1977) Biochem. Biophys. Res. Commun. 7 6 , 882-887 29. Saks, V. A., Kupriyanov, V. V., Elizarova, G. V., and Jacobus, W. E. (1980) J. Biol. Chem. 2 5 5 , 755-763 30. Moreadith, R. W., and Jacobus, W. E. (1982) J. Biol. Chem.2 5 7 , 899-905 31. Gellerich, F., and Saks, V.A. (1982) Biochem.Biophys. Res. Commun. 1 0 5 , 1473-1481 32. Jacobus, W. E.,and Saks, V.A. (1982) Arch. Biochem. Biophys. 219,167-178 33. Jacobus, W. E., Vandegaer, K. M., and Moreadith, R. W. (1984) in Proceedings of the Second International Congress on Myocardial and Cellular BioenergeticsandCompartmentation
(Brautbar, N., ed) Plenum Publishing Corp., New York, in press 34. Mukherjee, A., Wong, T. M., Templeton, G., Buja, L.M., and Willerson, J. T. (1979) Am. J. Physiol. 2 3 7 , H224-H238 35. Neely, J. R., and Rovetto, J. J. (1975) Methods Enzymol. 3 9 , 289-331 36. Schaff, H. V., Dombroff, R., Flaherty, J. T., Bulkley, B.H., Hutchins, M., G. Goldman, A., R. andL.Gott, V. (1978) Circulation 58,240-249 37. Pande, S. V., and Blanchaer, C. M. (1971) J. Biol. Chem. 2 4 6 , 402-411 38. Eppenberger, M., H. Dawson, M., D. and Kaplan, N. 0.(1967) J. Biol. Chem. 2 4 2 , 204-209 39. Ochoa, S. (1955) Methods Enzymol. 1 , 735-739 40. Schnaitman, C., and Greenawalt, J. W. (1968) J. Cell. Biol. 3 8 , 158-175 41. Gornall, A. G., Bardawill, C. J., and David, M. M. (1949) J. Biol. Chem. 177,751-766 42. Goldberg, E. (1963) Science (Wash. D.C.) 139,602-603 43. Jacobus, W. E. (1975) J. Mol. Cell. Cardiol. 7 , 783-791 44. Snedecor, G. W., and Cochran, W. G. (1967) Statistical Methods, Iowa State University Press, Ames, IA 45. Duncan, A. J. (1952) Quality Control and Industrial Statistics, Irwin Press, Homewood, IL 46. Saks, V. A., Chernousova, G. B., Voronkov, I. I., Smirnov, V. N., and Chazov, E. I. (1974) Circ. Res. 3 5 , Suppl. 111,139-149 47. Braasch, W., Gudbjarnason, S., Puri, P. S., Ravens, K. G., and Bing, R. J. (1968) Circ. Res. 23,429-438 48. Jennings, R. B. (1975) A m . J. Pathol. 80,419-450 49. Neely, J. R., Liebermeister, H., Battersby, E. J., and Morgan, H. E. (1967) Am. J. Physiol. 212,804-818 50. Flaherty, J. T., Weisfeldt, M.L., Hollis, D. P., Schaft, H. V., Gott, V. L. and Jacobus, W. E. (1980) in Advances in Myocardiology (Tajuddin, M., Bhatia, B., Siddiqui, H. H., and Rona, G., eds) Vol. 2, pp. 487-499, University Park Press, Baltimore 51. Hall, N.,Addis, P., and DeLuca, M. (1979) Biochemistry 1 8 , 1745-1751 52. Haas, R., Nagasawa, T., and Heinrich, P. C. (1977) Biochem. Biophys. Res. Commurz. 7 4 , 1060-1065 53. Aoki, Y. (1978) J. Biol. Chem. 253,2026-2032 54. Hare, J. F. (1978) Biochem. Biophys. Res. Commun. 8 3 , 12061215 55. Kobayashi, K., and Neely, J. R. (1979) Circ. Res. 4 4 , 166-175 56. Scarpa, A., and Graziotti, P. (1973) J. Gen. Physiol. 6 2 , 756-772 57. Krause, S., and Hess, M. L. (1982) Fed. Proc. 4 1 , 1525 58. DeBoer, V.,W. L. Ingwall, J. S., Kloner, A.,R. and Braunwald, E. (1980) Proc. Natl. Acud. Sci. U. S. A. 77, 5471-5475 59. Nicklas, J. M., Becker, L. C., and Bulkley, B. H. (1982) Clin. Res. 30,209A 60. Braunwald, E., and Kloner, R.A. (1982) Circulation 66, 11461149
