INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

NUMEROUS HUMAN epidemiological studies exist to support the notion that regular exercise is cardioprotective (20, 33, 40). Indeed, Morris et al. (33) have shown that the incidence of myocardial infarctions is reduced in physically active individuals. Furthermore, these investigators reported that the survival rate of heart attack victims is greater in active individuals compared with sedentary ones.

In contrast to the large number of epidemiological studies, experimental studies examining the effects of regular exercise on cardiac responses to ischemia and reperfusion (I/R) are relatively few. Furthermore, studies examining the effects of endurance exercise on myocardial responses to I/R have provided equivocal results, with some investigators reporting that training improves myocardial tolerance to I/R (5, 7, 8, 23, 30), whereas others report no improvement (15, 35) or even a decreased myocardial tolerance to I/R (28). The explanation for the divergent findings is unclear but may be linked to differences in exercise training programs and/or variations in the experimental protocols used to study myocardial responses to I/R. Clearly, additional research is required to clarify this issue.

Many of the aforementioned experimental studies have used rodent swim training as the exercise model. In this regard, Bowles et al. (7) have argued that the swimming model in rats has significant limitations. Their argument centers around two major points. First, swimming results in markedly different cardiovascular, sympathetic, and hormonal responses compared with treadmill exercise (13, 16, 43). Second, the addition of tail weights to increase the swimming workload, and/or large numbers of animals swimming in a single pool, results in animals being submerged periodically. This event can promote a diving reflex and result in intermittent periods of hypoxia (43). Hence, it appears that the swim training exercise model for rodents is not a good model to study human adaptation to exercise in a terrestrial environment.

To date, only four investigations have examined myocardial responses to I/R in treadmill-trained animals. All of these studies employed an in vitro model of global ischemia (7, 8, 30, 35). Although this type of experimental protocol has the advantage of studying the myocardium during carefully controlled preload and afterload conditions, this paradigm does not mimic in vivo conditions associated with myocardial I/R. Furthermore, these studies did not examine whether exercise training reduces I/R-induced myocardial damage (e.g., lipid peroxidation) or alters the incidence of cardiac arrhythmias during I/R. Given the clinical significance of I/R injury and the limited data available concerning the effects of treadmill exercise training on in vivo myocardial responses to I/R, there is a need for additional research in this area. Therefore, this study investigated the effects of 10 wk of rigorous treadmill training on myocardial antioxidant capacity and performance during in vivo myocardial I/R in the rat. The specific goals of this study were to determine 1) whether exercise training results in improved myocardial contractile performance during I/R, 2) whether training results in a reduced incidence of ventricular arrhythmias during I/R, 3) whether exercise training promotes alterations in heat shock protein (HSP) content and the antioxidant capacity of the myocardium, and 4) whether exercise training results in a reduction in myocardial lipid peroxidation following I/R.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Animals. This experimental protocol was approved by the University of Florida Animal Care and Use Committee and followed the guidelines established by the American Physiological Society for the use of animals in research. Fifty young adult female Sprague-Dawley rats (4 mo old) were randomly assigned to either a sedentary control group (n = 20) or to an exercise training group (n = 30). The exercise training group contained more animals than the control group to account for the fact that ~20-30% of exercise-trained animals do not complete a 10-wk training program because of limb injuries associated with rigorous treadmill exercise. During the experimental period, animals were maintained on a 12:12-h light-dark photoperiod and provided rat chow and water ad libitum.

Exercise training protocol. The control animals were not exercised but were placed on a nonmoving treadmill three times per week (10-30 min/session) for acclimatization. The exercise-trained animals were exercised 4 days/wk (Monday, Tuesday, Thursday, and Friday) for 10 wk on a motor-driven treadmill. Both the treadmill speed and grade were gradually increased over the course of the 10-wk training period to maintain the relative work rate at ~75-80% of maximal oxygen consumption throughout the training regimen (see Table 1). Electrical shocks were used sparingly to motivate animals to run. The decision to train animals for 10 wk at this work rate is based on work from our laboratory that demonstrates that this training protocol results in both cardiac hypertrophy and biochemical alterations (37).

In vivo protocol for studying I/R-induced myocardial damage. Ten control and 10 trained animals were studied during I/R. Trained animals were studied within 42-48 h after their last training session. Animals were anesthetized with 30 mg/kg pentobarbital sodium and ventilated with room air using a small animal ventilator (Harvard Apparatus model 661). Throughout the experiments body temperatures were monitored via a rectal thermistor probe. Body temperature was maintained at ~37 ± 1°C with a heated operating platform and appropriate heating lamps.

Validation of coronary occlusion and reperfusion. The aforementioned technique of coronary occlusion has been shown to be effective by others (4, 23). However, to validate that complete coronary occlusion was achieved in our hands, we performed preliminary experiments where small amounts of Evans blue dye was injected directly into the right ventricle cavity (i.e., upstream from the ligature). After injection of the dye, the heart was removed within 10 s and examined for evidence of dye in the ventricular mass supplied by the left anterior descending coronary artery. Failure to observe dye stain in this area of the ventricle was interpreted as achievement of coronary occlusion.

Measurement of arterial blood pressure. To monitor cardiovascular function during the I/R protocol, an arterial cannula was introduced into the carotid artery and advanced to the arch of the aorta. Peak systolic pressure was measured using a pressure transducer interfaced with a computerized heart performance analysis system (Digi-Med, Louisville, KY). A primary factor in our decision not to measure ventricular pressures directly was that we have observed that catheters placed in the left ventricles of rats results in a significant number of catheter-induced ventricular arrhythmias. This is meaningful because ventricular arrhythmias were considered to be an important dependent variable in this investigation. Furthermore, experiments in our laboratory have shown that peak systolic pressures in rats measured in the arch of the aorta do not differ significantly (P > 0.05) from pressures measured in the ventricle.

Sham operation. To determine the effects of endurance training on the biochemical properties of ventricular tissue not exposed to I/R, both control and trained animals were studied following a sham operation (i.e., thoracotomy performed) that did not include myocardial I/R. Animals were anesthetized with 30 mg/kg pentobarbital sodium and ventilated with room air using a small animal ventilator (Harvard Apparatus model 661). The hearts and plantaris muscles were rapidly removed and quickly frozen in liquid nitrogen for subsequent biochemical analysis.

Tissue preparation. Selected biochemical properties of the ventricular myocardium were studied in both untrained (n = 9) and trained (n = 10) animals following the I/R protocol and from untrained (n = 10) and trained (n = 11) animals who were not exposed to I/R (i.e., sham surgery). Furthermore, as an index of the training-induced increase in oxidative capacity in locomotor muscles, we measured the citrate synthase (CS) activity in the plantaris muscle of sham-operated animals.

Biochemical assays. To assess the effects of both training and I/R on the myocardium, we measured tissue thiol content, two separate measures of lipid peroxidation, and the activities of both bioenergetic and antioxidant enzymes. Before homogenization, the epicardium and endocardium were removed from each sample using a dissecting microscope. Bioenergetic enzyme activity and antioxidant enzyme activity were measured in tissues obtained from the sham-operated animals. The left ventricular myocardium and the plantaris muscle were individually minced and homogenized in cold 100 mM phosphate buffer with 0.05% bovine serum albumin (1:20 wt/vol; pH = 7.4). Homogenization included a 15-s treatment with a tissue homogenizer (Ultra-Turrax T25; IKA Works, Cincinnati, OH), followed by 10 passes of the homogenate in a tight-fitting Potter-Elvehjem homogenizer. Homogenates were then centrifuged (3°C) for 10 min at 400 g. The supernatant was decanted and assayed for CS (EC 4.1.3.7), SOD (EC 1.15.1.1), Cat (EC 1.11.1.6), PFK (EC 2.7.1.11), LDH (EC 1.1.1.27), and GPX (EC 1.11.1.9). SOD and GPX activities were determined in the left ventricle using a modification of the procedures described by Oyanagui (34) and Flohe and Gunzler (14), respectively. Training-induced changes in SOD isoforms were determined by KCN inhibition of the Cu-Zn SOD. CS activity was assayed using the procedure described by Srere (42). PFK, LDH, and Cat activities were assayed using techniques reported by Shonk and Boxer (41), Bergmeyer et al. (3), and Aebi (1), respectively. In our laboratory the coefficients of variation for SOD, GPX, CS, PFK, LDH, and Cat assays were ~3, 3, 3, 6, 3, and 5%, respectively. These and all other biochemical assays were performed in duplicate at 22°C, and samples from all experimental groups were assayed on the same day to avoid interassay variation.

Data analysis. Myocardial contractility measurements were analyzed using a two-way analysis of variance with repeated measures with Fisher's least-significant difference (LSD) test used post hoc. Myocardial biochemical parameters were analyzed using a two-way analysis of variance with Fisher's LSD test used post hoc.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Body weight and heart weights. Twenty-two animals successfully completed the exercise training program. Table 2 contains the body masses, heart masses, and heart-to-body mass ratios in both control and exercise-trained animals. Initial and final body weights did not differ between groups. In contrast, both heart weight and the heart-to-body weight ratio was greater (P < 0.05) in the exercise-trained animals compared with control. The training-induced increase in heart weight and the heart-to-body weight ratio suggests that cardiac hypertrophy occurred.

Myocardial performance during I/R. Successful I/R protocols were performed on 10 trained and 9 untrained animals. Figure 1 contains the mean (±SE) values for peak aortic systolic blood pressure (SBP) during preischemia, ischemia, and reperfusion in both untrained and trained animals. No differences existed (P > 0.05) between groups in SBP before the introduction of ischemia. However, trained animals maintained higher (P < 0.05) SBP throughout the ischemic period and during reperfusion.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Changes in peak systolic pressure during ischemia-reperfusion (I/R) protocol. Values are means ± SE; n = 8 animals in each group. * Group differences (P < 0.05).

I/R and cardiac arrhythmias. Table 4 summarizes the cardiac arrhythmias evoked by the I/R protocol. Note that one trained animal and one control animal were eliminated from this analysis because of ventricular arrhythmias that occurred during the surgical procedure before occlusion of the coronary artery.

Oxidative and antioxidant enzyme activities. Table 5 contains the bioenergetic and antioxidant enzyme activities in the left ventricle and CS activity in the plantaris muscle. Exercise training resulted in a large and significant increase (~80%) in CS activity in the plantaris muscle. Furthermore, exercise training resulted in a significant increase in left ventricular PFK and LDH activity. Also, training resulted in a significant increase in total SOD activity as well as an increase in the activities of both the Mn-SOD and Cu-Zn SOD isoforms. No differences in ventricular CS, Cat, or GPX activities existed between the control and trained animals.

Myocardial thiol content. Figure 2 contains the means (±SE) of all experimental groups for total thiols, protein thiols, and nonprotein thiols. Training resulted in an increase (P < 0.05) in nonprotein thiols in the left ventricle (sham animals); this increase in nonprotein thiols also resulted in an increase in total thiols in trained animals.

View larger version (23K): [in this window] [in a new window]   Fig. 2.   A: total thiol concentration in left ventricle of experimental groups. B: protein thiol concentration in left ventricle of experimental groups. C: nonprotein thiol levels in left ventricle of experimental animals. Values are means ± SE. Sample sizes from animals exposed to I/R surgery are as follows: untrained (UT), n = 9; trained (TR), n = 10. Sample sizes from sham animals from sham surgery are as follows: untrained, n = 10; trained, n = 11. a Different (P < 0.05) from sham (within same group). b Different (P < 0.05) from untrained sham group.

Myocardial HSP32, HSP72, and HSP73 content. To determine whether HSP32, HSP72, and HSP73 content was elevated in the hearts from trained animals, portions of the left ventricle from untrained and trained animals were analyzed for HSP isoform content using Western blotting. The results revealed that training did not alter ventricular HSP32 and HSP73 content. However, compared with control, HSP72 content was significantly greater in the left ventricle of trained animals (Fig. 3).

View larger version (24K): [in this window] [in a new window]   Fig. 3.   A: graphical representation of left ventricular heat shock protein (HSP) values (trained animals) obtained from densitometric scanning of Western blots reacted with antibodies for HSP32, HSP72, and HSP73. Values are means ± SE expressed as a percent of untrained group. Sample sizes from animals exposed to I/R surgery are as follows: untrained, n = 9; trained, n = 10. Sample sizes from sham animals from sham surgery are as follows: untrained, n = 10; trained, n = 11. * Different from untrained at P < 0.05. B: typical Western blots indicating the responses of HSP32, HSP72, and HSP73 in the left ventricle of exercise-trained and untrained animals.

Myocardial lipid peroxidation. Figure 4 contains mean (±SE) values for TBARS and lipid hydroperoxides in the left ventricle of all experimental groups. I/R resulted in an increase in TBARS concentration in the left ventricle in both untrained and trained animals; note, however, that the I/R-induced increase in TBARS was significantly lower in the trained animals compared with untrained.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   A: Levels of thiobarbituric acid-reactive substances (TBARS) in left ventricle of experimental groups. B: levels of cumene hydroperoxide equivalents (CHE) in left ventricle of experimental groups. Values are means ± SE. Sample sizes from animals exposed to I/R surgery are as follows: untrained, n = 9; trained, n = 10. Sample sizes from sham animals from the sham surgery are as follows: untrained, n = 10; trained n = 11. a Different (P < 0.05) from sham (within same group). b Different (P < 0.05) from trained I/R group.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

Overview of principal findings. To our knowledge, this is the first experiment to comprehensively examine the effects of treadmill exercise training on cardiac biochemistry, arrhythmias, and contractile performance during I/R in vivo. Our results indicate that endurance exercise training results in an improved myocardial performance during both ischemia and reperfusion. Also, compared with untrained animals, endurance-trained animals experienced lower levels of myocardial lipid peroxidation as a result of I/R. Training promoted an increase in nonprotein thiols, which may act as nonenzymatic antioxidants and thus contribute to the observed reduction in myocardial lipid peroxidation. Furthermore, exercise training was associated with an increase in SOD activity as well as the activities of both PFK and LDH. Finally, training resulted in a significant increase in HSP72 in the left ventricle. A brief discussion of each of these points follows.

Training improves myocardial contractile performance during ischemia. As evidenced by measures of SBP, the present data clearly demonstrate that rigorous treadmill training results in improved myocardial performance during ischemia. Furthermore, our in vivo data are in general agreement with studies that have examined in vitro myocardial performance in response to ischemia following both treadmill (7, 8) and swim exercise training in rats (5).

Training reduces I/R-induced myocardial injury and improves postischemic myocardial recovery. Our findings support the notion that endurance training improves in vivo myocardial functional recovery during reperfusion following 20 min of coronary ligation-induced ischemia. Furthermore, similar to what was found in work by Kihlstrom (26), training results in a reduction in reperfusion-induced myocardial injury as evidenced by lower levels of myocardial lipid peroxidation (Fig. 4) and higher levels of protein thiols following I/R (Fig. 2).

Exercise training and I/R arrhythmias. The specific mechanisms responsible for I/R arrhythmias are complex and vary depending on the duration of ischemia. Nonetheless, it is clear that myocardial I/R results in cardiac myocyte ionic disturbances due to a reduction in cellular energy levels and increased production of ROS (11). To date, there are no published reports concerning the effects of treadmill exercise training on the incidence of arrhythmias during in vivo I/R. Therefore, one of the objectives in these experiments was to determine whether endurance training alters the incidence of ventricular arrhythmias following I/R in the rat. Our results indicated that training provided only moderate protection against cardiac arrhythmias during this type of I/R protocol. This is evidenced by the fact that the number or severity of VPBs, salvos, and ventricular tachycardia did not differ between the sedentary and trained animals (see Table 4). Nonetheless, exercise training did reduce the incidence of VPBs during reperfusion. Indeed, the incidence of VPBs was greater in the untrained animals during reperfusion compared with the trained animals (80% vs. 10%).

Critique of experimental model and limitations of study. The adult female Sprague-Dawley rat was chosen as an experimental model because 1) the nature of these invasive experiments prevents the use of human subjects, 2) this rat is highly inbred and does not display large interanimal variation in coronary collateral circulation (17), and 3) the rat is a widely accepted model for the study of exercise-induced myocardial adaptations (2, 5, 23, 37).

Perspectives

Finally, although our data clearly indicate that exercise training is associated with a reduction in lipid peroxidation and an improvement in myocardial performance during moderate-duration I/R, whether or not exercise training would result in myocardial protection against long-duration ischemia (e.g., 120 min) cannot be determined by these experiments. Indeed, when comparing protective interventions against myocardial I/R injury, direct extrapolation from one duration of ischemia to another is tenuous and should be approached with caution.