Effect of superoxide dismutase and catalase on regional dysfunction after exercise-induced ischemia DAVID C. HOMANS, PAUL LINDSTROM, Department Minneapolis,

RICHARD DOUGLAS

ASINGER, TODD PAVEK, MELANIE CRAMPTON, PETERSON, AND ROBERT J. BACHE

of Medicine, Cardiovascular Minnesota 55405

Division,

Homans, David C., Richard Asinger, Todd Pavek, Melanie Crampton, Paul Lindstrom, Douglas Peterson, and Robert J. Bathe. Effect of superoxide dismutase and catalase on regional dysfunction after exercise-induced ischemia. Am. J. Physiol. 263 (Heart Circ. Physiol. 32): H392-H398, 1992.-This study was designed to test the hypothesis that the oxygen free radical scavengers superoxide dismutase (SOD) and catalase may reduce myocardial “stunning” after exercise-induced ischemia. To test this hypothesis, 8 mongrel dogs performed treadmill exercise for IO min in the presence of a flowlimiting coronary artery stenosis. Regional left ventricular function was measured with ultrasonic microcrystals implanted to measure regional wall thickening. Regional myocardial perfusion was measured with radioactive microspheres. The combination of SOD (5 mg/kg iv) and catalase (5 mg/kg iv) did not affect heart rate, blood pressure, coronary artery flow, or regional myocardial blood flow at rest, during exercise, or in the postexercise period. SOD and catalase had no effect on regional wall thickening at rest before exercise. During exercise in the absence of a coronary artery stenosis, thickening was slightly lower during SOD and catalase infusion (27 t 11.0 vs. 30.8 t 11.5%, SOD vs. control P = 0.05). During exercise in the presence of a coronary artery stenosis, there was no difference in thickening. Infusion of SOD and catalase affected neither the transient rebound function occurring early after exercise nor the prolonged period of stunning. These results indicate that the myocardial stunning that follows exercise-induced ischemia is unlikely to be mediated by oxygen free radicals. myocardial

ischemia;

free radicals;

exercise

PERIODS of myocardial ischemia insufficient in duration to cause myocardial necrosis may result in prolonged periods of reversible regional dysfunction termed myocardial “stunning” (3). Although the mechanisms responsible for myocardial stunning have not been completely elucidated, increasing attention has focused on the role of oxygen free radicals (1, 2, 21). Several investigators have demonstrated that superoxide dismutase (SOD) and catalase enhance recovery of function after 15min coronary artery occlusions in anesthetized, open-chest dogs (6,12,13,14). Preliminary observations also suggest a similar benefit in conscious dogs after 15min coronary artery occlusions (20). Thus evidence currently favors a significant role for oxygen-derived free radicals in myocardial stunning after relatively brief coronary artery occlusions in resting animals, particularly in the open-chest, acutely anesthetized state. Transient reversible regional myocardial dysfunction may also follow episodes of exercise-induced ischemia (8, 9, 16). There are significant differences between exercise-induced ischemia and ischemia resulting from coronary artery occlusions at rest, however, which may have implications regarding mechanisms of stunning and the effectiveness of interventions designed to reduce myocardial stunning. During exercise-induced ischemia, BRIEF

University

of Minnesota

Medical

School,

oxygen consumption is higher than during coronary artery occlusions at rest, and the degree of ,&adrenergic stimulation is higher. The duration of ischemia is limited by exercise tolerance and is generally shorter than for coronary artery occlusions at rest. There is substantial residual myocardial perfusion, particularly to the mid and epicardial layers, which permits continued oxygen delivery and allows for removal of products of anerobic metabolism. Finally, the mechanical responses to ischemia differ, with substantially less wall thinning and regional dilatation during exercise-induced ischemia. The impact of these differences on mechanisms of myocardial stunning is not known. Recovery of function after release of coronary occlusions at rest is a triphasic process characterized by an immediate “rebound” in function (sometimes to levels very close to control), subsequent deterioration, and a final slow, gradual improvement over a period of hours to possibly days (19). When a coronary artery stenosis is released immediately after a lo-min episode of exerciseinduced ischemia, the same pattern is observed. The transient rebound is temporally associated with maximal reactive hyperemia in both exercise and resting models of ischemia; when reactive hyperemia is prevented the rebound does not occur (9). The mechanisms responsible for the transient rebound and subsequent deterioration have not yet been established, but the concept that oxygen free radicals generated during the period of reactive hyperemia might be responsible for the subsequent deterioration in regional function is an attractive possibility. The objective of this study was, therefore, to test the hypothesis that the oxygen free radical scavengers SOD and catalase reduce myocardial stunning after exercise-induced ischemia. To test this hypothesis, eight chronically instrumented dogs were studied during and after exercise-induced ischemia. Studies were performed under control conditions and during the infusion of SOD and catalase in doses previously demonstrated to reduce myocardial stunning after coronary artery occlusions in resting animals (6, 13, 14). METHODS Studies were performed on eight adult mongrel dogs (23 & 2 kg) previously trained to run on a motor-driven treadmill. The animals were anesthetized with pentobarbital sodium (25-30 mg/kg iv) and ventilated with a Harvard respirator. A left thoracotomy was performed in the fourth intercostal space. A polyvinyl chloride catheter (3.0 mm OD) filled with heparin saline solution (200 U/ml) was inserted into the root of the aorta via the internal thoracic artery for pressure monitoring and blood withdrawal. The pericardium was then incised and the heart supported in a pericardial sling. The proximal 1.5 cm of the circumflex artery was dissected free and instrumented with an

H392 0363-6135/92 $2.00 Copyright 0 1992 the American Physiological Society Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.

SOD AND

CATALASE

IN POSTEXERCISE

electromagnetic flow probe and hydraulic occluder. A chronically implantable micromanometer (model P-5, Konigsberg Instruments, Pasadena, CA) and a fluid-filled polyvinyl catheter were then inserted into the left ventricle through the apical dimple. Two pairs of ultrasonic microcrystals were inserted for measurement of regional myocardial wall thickness. The endocardial crystal was inserted through a small epicardial stab wound and advanced obliquely to the subendocardium. The epicardial crystal was positioned so as to minimize the distance between the two crystals and sutured in place. One crystal pair was placed within the distribution of the circumflex coronary artery, and one crystal pair was placed within the distribution of the left anterior descending artery. A fluid-filled polyvinyl chloride catheter was inserted into the left atria1 cavity via the left atria1 appendage and secured with a purse-string suture. The pericardium was loosely closed, and all catheters and electronic cables were tunneled dorsally to the base of the neck where they were exteriorized. The thoracotomy was then repaired and animals allowed to recover from surgery. Experiments were performed IO- 14 days postoperatively when the animals were in good physical condition. Catheters were protected with a nylon vest (Alice King Chatham, Los Angeles, CA). Measurements of aortic and left ventricular pressures were obtained using fluid-filled catheters connected to a Statham pressure transducer. Measurements of wall thickness were obtained by activating the implanted piezoelectric crystals with a Triton model 120 ultrasonic dimension system (San Diego, CA), modified so as not to interfere with the electromagnetic flowmeter function. All data were recorded on an eightchannel direct-writing oscillograph (Hewlett-Packard model 8800, Palo Alto, CA). In a subset of five dogs, myocardial blood flow was measured with injections of microspheres, 15 pm in diameter, labeled with one of the following gamma-emitting nuclides: 1251, 51Cr, 85Sr, g5Nb, l13Sn, or 46Sc. Microspheres were injected at rest, before exercise, during control exercise without stenosis, during SOD and catalase exercise without stenosis, and during both exercise periods with stenosis. Before injection, the microspheres were agitated for at least 15 min in an ultrasonic bath. During each intervention, 2 x IO6 microspheres were injected into the left atria1 catheter over a 15-s interval, and the atria1 catheter was flushed with 10 ml of isotonic saline. Beginning 5 s before each microsphere injection and continuing for 90 s, a reference sample of blood was drawn from the aortic catheter at a constant flow rate of 15.0 ml/min. Experimental protocol. Postoperatively the dogs were reexercised on a motor-driven treadmill to ensure that they could run at a speed of 6.4 km/h at a 14% grade for at least IO min and to refamiliarize them with the experimental setting. On the first day of study each dog performed a brief (3-4 min) warm-up run. They were then transferred to a sling and allowed to rest for 30 min in a darkened laboratory. Baseline values of heart rate, circumflex artery flow, arterial pressure, left ventricular pressure, left ventricular dP/dt, and ultrasonic dimension data were then recorded. In a subset of five dogs, microspheres were injected into the left atrium for determination of resting regional myocardial blood flow. One of two exercise protocols was then performed. All dogs underwent both protocols (separated by 24-48 h) and the order of protocols was randomized for each dog. For the control exercise protocol, the treadmill was advanced to a speed and grade that resulted in -100% increase in circumflex coronary blood flow. Two minutes after stabilization of hemodynamic and ultrasonic dimension measurements, radioactive microspheres were injected for determination of regional myocardial blood flow. Coronary artery blood flow was then reduced to the resting control value, and dogs continued exercising with the coronary artery occluder partially inflated. Ra-

STUNNING

H393

dioactive microspheres were again injected 2 min after the stenosis had been applied. After 10 min of exercise in the presence of a coronary stenosis, exercise was discontinued, and the occluder was completely deflated to permit unimpeded hyperemia in the postexercise period. After exercise, the dogs were transferred to a sling to maintain upright posture and were observed for 2 h. Hemodynamic measurements were recorded 5, IO, 15, 30, 45, 60, 90, and 120 min after exercise. For the drug intervention exercise, a warm-up run was performed identical to that previously described for the control exercise protocol. Twenty minutes after the warm-up exercise, animals received an infusion of bovine serum SOD [5 mg/kg (2,900 U/mg, Sigma)] and catalase [5 mg/kg (11,000 U/mg, Sigma)] via the left atria1 catheter. The infusion was begun 10 min before exercise, continued throughout the exercise period, and discontinued 5 min after exercise (total duration 30 min). Identical treadmill speeds and grades were used for both control and SOD and catalase runs, and coronary blood flow was reduced to the same value during each run. After exercise the hydraulic occluder was again completely released and the dogs were observed for 2 h in a manner identical to that previously described for the control exercise. After the study had been completed, the dogs were killed with a lethal dose of pentobarbital sodium. A left thoracotomy was performed, and the heart and both kidneys were excised. The circumflex coronary artery was cannulated at the site of the hydraulic occluder, and Evans blue dye was injected to identify the area of left ventricle supplied by the circumflex artery distal to the site of stenosis. The heart was then weighed and fixed in 10% buffered Formalin. After fixation, the atria, right ventricle, and aorta were removed and the left ventricle was weighed. The ultrasonic microcrystals were then inspected to ensure that all pairs of ischemic crystals were located at least 1 cm within the perfusion boundary of the circumflex coronary artery, as identified by the blue-stained myocardium, and all nonischemic pairs were at least I cm outside the perfusion boundary of the circumflex coronary artery. Full thickness blocks of tissue, at least 1 cm in diameter and containing the pairs of ultrasonic microcrystals, were then removed. Each block of tissue was divided into four layers of approximately equal thickness from epicardium to endocardium. Myocardial samples were then weighed on an analytical balance and placed in vials for determination of radioactivity. Myocardial and blood reference specimens were counted in a gamma counting system (model 5912, Packard) with a multichannel analyzer at window settings corresponding to the peak energies of each radionuclide. The activity in each energy window was corrected for background and for overlapping counts contributed by accompanying isotopes (5) Blood flow to each myocardial specimen was computed as &m = Q, x Cm/C,, where Qm is myocardial blood flow (ml/min), QT is the reference blood flow rate (ml/min), and C, is the counts per minute of the myocardial specimen. Blood flow was divided by the sample weight and expressed as milliliter per minute per gram of myocardium. Mean transmural flow was calculated as the combined total flow to all four layers, divided by the combined weight of all four samples. The ratio of subendocardial to subepicardial flow was obtained by dividing flow to the innermost layer by the corresponding flow to the subepicardial layer. For ultrasonic microcrystal determination of wall thickening, the end-diastolic wall thickness was measured at the initiation of the upstroke of the left ventricular pressure tracing recorded by the micromanometer, and the end-systolic thickness was measured 20 ms before peak negative dP/dt on the differentiated left ventricular pressure tracing (7). The values for 6-10 beats corresponding to one respiratory cycle were measured by

Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.

H394

SOD AND

CATALASE

IN POSTEXERCISE

hand and averaged. Percent wall thickening was defined as endsystolic thickness minus end-diastolic thickness divided by enddiastolic thickness times 100. The volume of circumflex coronary artery flow was determined by electrical integration of the electromagnetic flowmeter tracing. Hemodynamic, myocardial blood flow, and ultrasonic dimension data were compared using analysis of variance for repeated measures. When an overall difference was found, individual comparisons were made using the paired t test with the Bonferroni correction for multiple simultaneous comparisons. A P value < 0.05 was considered significant. Unless otherwise specified, values are reported as means t SD. This study was conducted in accordance with the guiding principles of the American Physiological Society for the Use of Animals in Scientific Research and approved by the Committee on the Use of Animals in Biomedical Research at the University of Minnesota. RESULTS

Hemodynamic data are summarized in Table 1. Heart rate at rest before control exercise was 125 t 22 beats/ min. Heart rate increased to 203 t 14 beats/min during exercise in the absence of circumflex stenosis and to 216 t 12 beats/min during exercise in the presence of a stenosis. There were no differences in heart rate between the control and drug exercise periods. Aortic systolic pressure at rest was 132 t 12 mmHg, which increased to 154 t 13 mmHg during exercise in the absence of a stenosis, and fell slightly to 145 t 14 mmHg during exercise in the presence of a coronary artery stenosis. Left ventricular end-diastolic pressure was 10.4 t 2.4 mmHg at rest and increased slightly to 12.3 t 3.5 mmHg during exercise in the absence of a stenosis. This pressure subsequently increased significantly to 21.1 t 7.9 mmHg during exercise in the presence of stenosis. There were no significant differences in heart rate, left ventricular systolic pressure, or left ventricular diastolic pressure between control and drug treatment exercise periods. Similarly, there were no differences in the rate-pressure product between control exercise and drug treatment exercise periods (3 1,400 t 4,200 vs. 30,600 t 4,700). There were no changes in +dP/dt between the control and drug exercise periods (3,600 t 3,lO vs. 3,350 t 820 mmHg/s). Circumflex coronary flow at rest prior to exercise was 36 t 8 ml/min. During exercise in the absence of a stenosis this flow increased to 68 t I2 ml/min and fell to 33 t 8 ml/min during exercise in the presence of a coronary artery stenoTable 1. Hemodynamic

HR Ao Sys Ao Dias HR x BP LV Dias +dP/dt -dP/dt Flow

measurements

125t22 132tl2 8629 16.4k2.7 10.4t2.4 24.7t5.3 22.3t3.7 3628

122kl6 122tlO 83tlO 14.9t2.4 11.3k8.8 20.6k2.8 18.5k3.7 38tl3

during

STUNNING

sis. One minute after exercise, after the coronary artery stenosis had been released, circumflex artery flow increased to 115 t 18 ml/min and subsequently returned to control values by 10 min postexercise (39 t 9 ml/min). There were no differences in circumflex coronary artery flow between control and drug exercise runs at any time. Myocardial blood pow. Myocardial blood flow at rest before exercise, during control exercise, and during exercise with SOD and catalase is summarized in Fig. 1. Myocardial blood flow is presented by each individual layer, mean transmural flow, and the ratio of subendocardial to subepicardial flow. Mean transmural myocardial blood flow at rest before exercise was 1.29 t 0.30 ml mine1 g-l myocardium. This increased to 2.41 t 0.69 ml min. g-l during exercise in the absence of a coronary artery stenosis and fell to 1.28 t 0.35 mlmin-l lg-l during control exercise in the presence of a coronary artery stenosis. At rest there was a transmural gradient of myocardial perfusion that favored the subendocardium with a subendocardial-to-subepicardial flow ratio of 1.78 t 0.41. This perfusion gradient was not changed during exercise in absence of coronary artery stenosis (1.79 t 0.51) but fell substantially to 0.52 t 0.36 during exercise in the presence of a coronary artery stenosis. This was accompanied by a significant fall in subendocardial perfusion from a baseline value of 1.46 t 0.37 to 0.65 t 0.16 ml min-l g-l during exercise in the presence of a coronary artery stenosis. Again, there were no significant differences in mean transmural blood flow, the ratio of endocardial to epicardial myocardial blood flow, or myocardial blood flow to any layer of myocardium between the control and the drug intervention exercise periods. Systolic wall thickening. Systolic wall thickening in the distribution of the circumflex coronary artery is summarized in Fig. 2. Systolic wall thickening at rest before control exercise was 24.1 t 10.6% and increased modestly to 30.8 t 11.5% (P < 0.05) during exercise in the absence of a coronary artery stenosis. Infusion of SOD and catalase did not significantly alter systolic wall thickening at rest before exercise (23.2 t 11.3 before drug infusion vs. 22.4 t 9.3% after drug infusion). During exercise in the absence of a coronary artery stenosis, systolic wall thickening was slightly but statistically lower during the drug infusion (27.0 t 11.9 vs. 30.8 t 11.5%; P < 0.02). After application of a coronary artery stenosis during exercise, l

l

l

l

l

exercise

117tl9 11825 79t6 lZ.lk4.9 6.8t3.9 19.123.4 19.9t2.8 41214

203tl4 154tl3 76t33 31.2t3.0 12.3t3.5 38.9t2.9 29.8t5.5 68t1.2

207t20 216tl2 217~~16 142t9 145tl4 14ltl5 83tl2 83t9 83tl6 29.3t3.9 31.4t4.2 30.6t4.7 13.9t7.3 21.1t7.9 17.3t7.7 34.4t5.1 36.0t3.1 33.5k8.2 25.6t7.9 29.4k5.7 24.9t7.3 80t31 33t8 38tl3 Values are means t SE. Predrug, resting value before infusion of superoxide dismutase (SOD) and catalase; S + C, SOD and catalase infusion; HR, heart rate (beats/min); Ao Sys and Ao Dias, aortic systolic and diastolic pressures, respectively (mmHg); HR x BP, peak heart rate x peak systolic blood pressure x 1,000; LV dias, left ventricular end-diastolic pressure (mmHg); +dP/dt, first derivative of LV pressure (mmHg x 100); -dP/dt, maximal rate of relaxation (mmHg x 100); flow, circumflex coronary artery flow (ml/min).

Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.

SOD AND CATALASE Exercise

+ Stenosis v

with

Control

SOD

Exercise l

IN POSTEXERCISE

Baseline

2.5 f-

(Resting) 2.0

+ i

1.5

1.0

0.5

Epi I 1

I

I

2

3

Transmural

Endo I

I

4

0.0

I

Mean

Ratio

Endo:Epi

Layer

Fig. 1. Regional myocardial blood flow to circumflex coronary arteryperfused myocardium at rest, during control exercise in the presence of a coronary stenosis, and during drug treatment exercise in presence of a coronary stenosis. Flow to each of 4 transmural layers of myocardium is represented, as well as mean transmural blood flow and ratio of subendocardial to subepicardial flow (endo:epi). Layer 1, subepicardial layer; layer 4, subendocardial layer. Exercise in presence of a stenosis was characterized by altered transmural pattern of myocardial perfusion with reduction in subendocardial flow and endo:epi. No differences observed between control exercise and drug treatment exercise.

-I* Y 0

,,,,,,,,,,,,,,,,,,,,,,,,,,,l BL

( Drug

1 5

15

30

Minutes

H395

thermore, the slope of the regression line is very close to 1.0, and X- and y-intercepts are close to 0. A similar relationship is noted for function in the postexercise period (Fig. 3B). Thus these data do not suggest that treatment with SOD and catalase may be more effective in subgroups with more or less severe dysfunction. Systolic wall thickening in the distribution of the left anterior descending coronary artery is summarized in Table 2. There were no significant differences between the control exercise period and the drug intervention period at any time.

& Catalase + Stenosis

STUNNING

60

Following

120

Exercise

Fig. 2. Regional systolic wall thickening at rest, during exercise in absence of a stenosis (Ex NS), exercise in the presence of a stenosis (Ex S+), and for 120 min after exercise. A small but statistically significant difference was observed in systolic wall thickening during exercise in absence of a coronary artery stenosis (P < 0.05) and 120 min after exercise (P < 0.05) between control and superoxide dismutase (SOD) + catalase-treated animals. No other differences observed between control and drug treatment exercise runs. Values are means & SE.

systolic wall thickening fell to 9.1 t 9.1% during the control exercise period and 8.1 t 8.9% during the drug intervention exercise. Immediately after exercise there was a transient rebound in systolic wall thickening to 20.6 t 7.2% for the control period and 21.1 t 6.5% for the drug intervention period. There was a subsequent decrease in systolic wall thickening to values substanti .ally below baseline with gradual return toward control thereafter. Although function was generally similar following control and drug treatment runs, there was a statistically significant difference in systolic wall thickening 120 min after exercise (18.6 t 7.0 for the control exercise period vs. 14.9 t 7.2% for the drug intervention period; P < 0.05). Figure 3A depicts ischemic zone thickening during control exercise plotted against thickening for the exercise period during SOD and catalase runs. There is a close linear relationship between the two exercise periods. Fur-

DISCUSSION

This study demonstrates that in chronically instrumented, awake dogs, SOD and catalase in the doses administered have no effect on regional myocardial function during or after a lo-min period of exercise-induced ischemia. Free radical scavengers affected neither the transient rebound of regional function early after exercise nor the subsequent deterioration and prolonged reduction in regional function that persisted at least 2 h in this study. In addition, the free radical scavengers SOD and catalase had no effect on reactive hyperemic flow in the first 10 min following exercise-induced ischemia. These data therefore fail to support the hypothesis that oxygen free radicals play an important role in the myocardial stunning that occurs after exercise-induced ischemia. Furthermore, these findings underscore the model dependence of myocardial stunning and suggest that significant differences exist between stunning produced by total coronary artery occlusions at rest and stunning that follows exercise-induced ischemia. Numerous investigators have consistently found that SOD and catalase reduce myocardial stunning after 15min coronary artery occlusions in resting animals (6, 10, 12-14). Although the majority of these studies have been conducted in acutely anesthetized, open-chest animals (6, 10,12-14), preliminary results suggest a similar benefit in chronically instrumented conscious animals (20). Buchwald et al. (4), although unable to demonstrate a benefit of SOD and catalase on the prolonged period of dysfunction following 8- and 12-min coronary artery occlusions in acutely anesthetized, open-chest pigs, did demonstrate an improvement in the early rebound of function noted immediately after reperfusion. That observation emphasizes the potential importance of interspecies differences in myocardial stunning. These findings suggest important differences between different models of stunning, which may have mechanistic and therapeutic implications. There are several possible explanations for the lack of a protective effect of SOD and catalase on myocardial stunning after exercise-induced ischemia. First, exerciseinduced ischemia may be less likely to be associated with free radical production. Sources of free radicals in the setting of coronary artery occlusions in resting animals have been hypothesized to include xanthine oxidase activity (in those animal species possessing significant myocardial xanthine oxidase activity), local accumulation

Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.

H396

SOD AND CATALASE Q)

IN POSTEXERCISE

25-A

a,

35-

20-

iii 2

30

z 3

:

-

15

r =

0.97

SEE

= 2.4

z

+

g

5-

2 .4 F:

o-

-0.4

= 0.98

25-

SEE 20

z . .

15-

.dF G

lo-

$ .d

G 2

_ y = 0.98x

=2.8

-

n

lo-

cn ..

B

r

+

STUNNING

5-

-55

be

o-

R -10

I -5

-10

I 0

% Thickening:

I 5

I 10

Control

I 15

I 20

I 25

Exercise

-5

III,

-5

0

% Thickening:

1,

5

10

15

Control

20

1,

25

30

35

Exercise

Fig. 3. A: scatter plot of systolic wall thickening during control exercise plotted against systolic wall thickening for same animals during exercise with drug treatment (1 data point/animal). Very close linear relationship exists between control and exercise runs, with slope of 0.94 and y-intercept of -0.43. B: plot of systolic wall thickening 5 min after control exercise vs. systolic wall thickening for same animals 5 min after drug treatment exercise. Again, a close linear relationship exists between the 2 exercise periods. SEE, standard error of the estimate.

Table 2. Percent systolic wall thickening region during

in control

and after exercise Control

Run

SOD

+ Catalase

BL 23.5t9.2 21.1t10.8 BL + D 22.2k12.0 EXNS 30.7t13.6 29.7t12.6 EXS+ 32.9513.6 30.9t13.8 1 min 26.9tl5.1 23.2t10.8 5 min 23.0k7.2 22.2tll.8 10 min 25.3t10.4 29.2t25.5 15 min 24.8t 14.6 22.8t10.8 30 min 24.8k 10.9 23.4t12.4 60 min 23.7k10.7 21.3t9.2 120 min 22.1k11.4 20.0t9.4 Values are means +: SD. BL, Resting baseline; BL + D, baseline + SOD + catalase; EXNS, exercise, no stenosis; EXS+, exercise with stenosis.

of polymorphonuclear leukocytes (during prolonged coronary artery occlusion), and autooxidation of catecholamines. Exercise-induced ischemia may differ significantly from coronary artery occlusions with regard to the extent of conversion of xanthine dehydrogenase to xanthine oxidase, or catecholamine metabolism. Second, the ischemic insult is probably less severe during exercise-induced ischemia than coronary artery occlusions at rest; the ischemic period is shorter in duration, myocardial perfusion is higher during exerciseinduced ischemia, and the magnitude of postischemic dysfunction is generally less than after coronary artery occlusions at rest. It is noteworthy in this regard that Murry et al. (12) failed to demonstrate a protective effect of SOD and catalase after coronary artery occlusions of only 5 min in duration but did note protection following a single, 15min coronary artery occlusion. Ischemic insults of limited duration or severity may therefore fall below the “threshold” for activation of free radical mechanisms, and thus interventions designed to reduce free radical generation may have no demonstrable effect. This latter possibility suggests that alternative mechanisms may be operative after exercise-induced ischemia or after shorter periods of coronary artery occlusion in resting animals. The likelihood that several different mecha-

nisms contribute to the phenomenon of myocardial stunning is further supported by the observation that even in studies demonstrating a benefit from SOD and catalase, stunning still occurs in the treated animals. Free radical scavengers therefore reduce but do not eliminate stunning, suggesting the presence of additional factors unrelated to free radicals. If several mechanisms are involved in the genesis of postischemic dysfunction, the relative importance of each mechanism may vary depending on the precise model under investigation. With an ischemic insult produced by total coronary occlusion of limited duration and followed by an intense reactive hyperemia, generation of oxygen free radicals may be maximal. Bolli et al. (2), using electron paramagnetic resonance, have shown that the beneficial effects of SOD and catalase were more readily demonstrable in regions of more severe ischemia. Furthermore, Kobayashi et al. (ll), using electron spin resonance spectrometry, demonstrated higher levels of free radicals after 40 min than 20 min of ischemia. These data suggest that short-lived, modest ischemic insults may fail to fully activate free radical-generating systems. Because free radicals are produced mainly during the early period of reactive hyperemia after release of coronary occlusions in resting animals (22), it is possible that differences in the reactive hyperemia after exercise-induced ischemia could occur, which could contribute to the lack of effect of free radical scavengers noted in this study. A brisk reactive hyperemia is usually observed after release of the coronary stenosis at the end of exercise in this model, however (9), which is quite comparable to the reactive hyperemias noted after release of transient coronary occlusions in resting animals. It is therefore unlikely that differences in reactive hyperemia contribute significantly to the results observed in this study. An additional difference between this model and previously reported studies demonstrating a benefit of oxygen free radical scavengers is the fact that this study was conducted in closed-chest, conscious animals free from the acute effects of surgical anesthesia. The possibility remains that some of the previously demonstrated beneficial effects of oxygen free radical scavengers may have

Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.

SOD AND

CATALASE

IN POSTEXERCISE

been related to acute surgical trauma or effects of anesthesia. There are several limitations to this study. SOD and catalase levels were not measured, and a range of doses was not tested. However, the doses of SOD and catalase were identical to those reported in three previous studies demonstrating a benefit from SOD and catalase on stunning after total coronary artery occlusions at rest (6, 13, 14). The total duration of infusion in the current study (30 min) was shorter than previous studies, however, and the infusion was discontinued 5 min after discontinuation of exercise. Although different rates of SOD and catalase metabolism cannot be entirely excluded in exercising animals, this infusion schedule should have produced peak concentrations of SOD and catalase that were as high or higher than those noted in previous studies and that should have been maximal during late exercise and in the early postischemic period. After release of total coronary occlusion, oxygen free radical generation has been demonstrated to be maximal during this early postischemic reactive hyperemia (2, 22). Furthermore, it is in the first 5 min after exercise that regional myocardial function deteriorates after the transient rebound observed at 1 min. It is highly probable, therefore, that sufficient concentrations of SOD and catalase were present during this early period when postischemic stunning first becomes evident. Although it is possible that free radical mechanisms may have been important in the late postexercise period, when the SOD and catalase levels had fallen, this is unlikely based on the observation that most free radical generation occurs early after reperfusion (22). A second limitation is that observations of regional myocardial function were not systematically made later than 2 h postexercise. The current data do not suggest any protective effect as late as 2 h after exercise, however. Finally, regional myocardial blood flow was not measured in all animals in this study. However, in the subset of animals in which myocardial blood flow was measured, there was no evidence of a protective effect. Measurement of myocardial perfusion is most important when comparing separate groups of dogs undergoing coronary artery occlusions, due to the high degree of variability in collateral flow among dogs. In this study, each dog served as its own control, and studies were performed on alternate days. Neither coronary artery flows nor regional myocardial blood flows differed between control and drug treatment runs. Since microsphere measurements of regional myocardial blood flow measure both coronary artery and collateral vessel contribution to myocardial blood flow, it is doubtful that collateral flow varied substantially within animals between the treatment and control exercise periods. Furthermore, functional data demonstrate a high degree of reproducibility between the control and treatment exercise periods. This is consistent with reproducible collateral flows and does not suggest a selective benefit of SOD and catalase in subgroups with either more or less severe ischemia and dysfunction. In summary, these data in chronically instrumented exercising dogs fail to provide evidence for a protective effect of oxvgen free radical scavengers SOD and catalase

STUNNING

H397

on myocardial stunning after exercise-induced ischemia. These data therefore also imply that several mechanisms may contribute to myocardial stunning and that the relative importance of these mechanisms may vary in different models. The authors gratefully acknowledge the assistance of Dr. Thomas Rector with statistical analysis and the secretarial assistance of Dana Souther. This study was supported in part by National Heart, Lung, and Blood Institute Grants HL-01162, HL-34701, and HL-20598. Address for reprint requests: D. C. Homans, University of Minnesota, Cardiovascular Div., Box 375 UMHC, Minneapolis, MN 55455. Received 27 September

1991; accepted in final form 24 February

1992.

REFERENCES 1. Bolli, R. Mechanism of myocardial “stunning”. CircuZation 82: 723-738, 1990. 2. Bolli, R., M. 0. Jeroudi, B. S. Patel, C. M. DuBose, E. K. Lai, R. Roberts, and P. B. McCay. Direct evidence that oxygen-derived free radicals contribute to postischemic myocardial dysfunction in the intact dog. Proc. NutZ. Acad. Sci. USA 86: 4695-4699, 1989. 3. Braunwald, E., and R. A. Kloner. The stunned myocardium: prolonged, postischemic ventricular dysfunction. Circulation 66: 1146-1149, 1982. 4. Buchwald, A., H. H. Klein, S. Lindert, S. Pith, K. Nebendahl, V. Wiegand, and H. Kreuzer. Effect of intracoronary superoxide dismutase on regional function in stunned myocardium. J. Cardiovasc. Pharmacol. 13: 258-264, 1989. 5. Domenech, R. J., J. I. E. Hoffman, M. I. M. Noble, K. B. Saunders, J. R. Henson, and S. Subijanto. Total and regional coronary blood flow measured by radioactive microspheres in conscious and anesthetized dogs. Circ. Res. 25: 581-596, 1989. 6. Gross, G. J., N. E. Farber, H. F. Hardman, and D. C. Warltier. Beneficial actions of superoxide dismutase and catalase in stunned myocardium of dogs. Am. J. Physiol. 250 (Heart Circ. Physiol. 19): H372-H377, 1986. 7. Heyndrickx, G. R., H. Baig, P. Nellen, I. Leusen, M. C. Fishbein, and S. F. Vatner. Depression of regional blood flow and wall thickening after brief coronary occlusions. Am. J. Physiol. 234 (Heart Circ. Physiol. 3): H653-H659, 1978. 8. Homans, D. C., D. D. Laxson, E. Sublett, P. Lindstrom, and R. J. Bathe. Cumulative deterioration of myocardial function after repeated episodes of exercise-induced ischemia. Am. J. Physiol. 256 (Heart Circ. Physiol. 25): Hl462-Hl471, 1989. 9. Homans, D. C., E. Sublett, X. Dai, and R. J. Bathe. Persistence of regional left ventricular dysfunction after exercise-induced myocardial ischemia. J. Clin. Invest. 77: 66-73, 1986. 10. Jeroudi, M. O., F. J. Triana, S. Patel, and R. Bolli. Effect of superoxide dismutase and catalase, given separately, on myocardial “stunning”. Am. J. Physiol. 259 (Heart Circ. Physiol. 28): H889-H901, 1990. 11. Kobayashi, A., H. Watanabe, K. Ozawa, H. Hayashi, and N. Yamazaki. Oxygen-derived free radicals related injury in the heart during ischemia and reperfusion. Jpn. Circ. J. 3: 1122-1131, 1989. 12. Murry, C. E., V. J. Richard, R. B. Jennings, and K. A. Reimer. Free radicals do not cause myocardial stunning after four 5 minute coronary occlusions (Abstract). Circukztion 80, Suppl. II: 11-296, 1989. 13. Myers, M. L., R. Bolli, R. F. Lekich, C. J. Hartley, and R. Roberts. Enhancement of recovery of myocardial function by oxygen free-radical scavengers after reversible regional ischemia. Circulation 72: 915-921, 1985. 14. Przyklenk, K., and R. A. Kloner. Superoxide dismutase plus catalase improve contractile function in the canine model of the “stunned myocardium”. Circ. Res. 58: 148-156, 1986. 15. Richard, V. J., C. E. Murry, R. B. Jennings, and K. A. Reimer. Therapy to reduce free radicals during early reperfusion does not limit the size of myocardial infarcts cause by 90 minutes of ischemia in dogs. Circulation 78: 473-480, 1988.

Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.

H398 16. Robertson,

SOD AND CATALASE

IN POSTEXERCISE

W. S., H. Feigenbaum, W. F. Armstrong, J. C. Dillon, J. O’Donnell, and P. W. McHenry. Exercise echocardiography: a clinically practical addition in the evaluation of coronary artery disease. J. Am. CoZZ.CardioZ. 2: 1085-1091, 1983. 17. Tamura, Y., L. Chi, E. M. Driscoll, P. T. Hoff, B. A. Freeman, K. P. Gallagher, and B. R. Lucchesi. Superoxide dismutase conjugated to polyethylene glycol provides sustained protection against myocardial ischemia/reperfusion injury in canine heart. Circ. Res. 63: 944-959, 1988. 18. Tanaka, M., R. C. Stoler, G. P. FitzHarris, R. B. Jennings, and K. A. Reimer. Evidence against the “early protection-delayed death” hypothesis of superoxide dismutase therapy in experimental myocardial infarction. Circ. Res. 67: 636-644, 1990.

STUNNING

19. Theroux,

P., D. Franklin, J. Ross, and W. S. Kemper. Regional myocardial function during acute coronary artery occlusion and its modification by pharmacologic agents in the dog. Circ. Res. 45: 896-908, 1974. 20. Triana, J. F., A. Unisa, and R. Bolli. Antioxidant enzymes attenuate myocardial “stunning” in the conscious dog (Abstract). FASEB J. 7: A622, 1990. 21. Werns, S. W., M. J. Shea, E. M. Driscoll, C. Cohen, G. D. Abrams, B. Pitt, and B. R. Lucchesi. The independent effects of oxygen radical scavengers on canine infarct size. Circ. Res. 56: 895-898,

1985.

22. Zweir, J. L. Measurement of superoxide-derived free radicals in the reperfused heart. J. Biol. Chem. 263: 1353-1357, 1988.

Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.