INTRODUCTION

TOP ABSTRACT INTRODUCTION RESEARCH DESIGN AND METHODS RESULTS DISCUSSION REFERENCES

EPIDEMIOLOGICAL STUDIES have demonstrated that regular exercise is associated with a reduction in the long-term risk of myocardial events (19, 20, 27, 32). People who exercise regularly not only have a lower baseline risk of myocardial infarction but also they have a lower risk that an infarction will be triggered by heavy physical exertion (26). The overall risk of death from coronary heart disease decreased about twofold in individuals who were physically active compared with less active individuals (20). On the other hand, heavy physical exercise has been recognized as a factor that triggers severe myocardial events such as impaired ventricular performance (29), cardiac ischemia (33, 34), myocardial infarction, and cardiac arrest (37, 40, 42). In symptom-free men, the risk of fatal and nonfatal heart attacks has been shown to be 4-56 times higher during physical exercise than when seated at home reading a book (37, 47). Because 1,500,000 myocardial infarctions occur annually in the United States (2), 75,000 infarctions that lead to 25,000 deaths may occur soon after exercise (26, 41). However, no direct experimental data are available to demonstrate that forced exercise can damage the heart. Studies are limited by suitable animal models, as well as by cardiac-specific markers that may be released into the circulation after injury.

Cardiac troponin T (cTnT) and cardiac troponin I (cTnI) have unique amino acid sequences that differentiate them from their respective skeletal muscle isoforms (7, 48). These differences have allowed for the development of specific monoclonal antibodies that have been incorporated into immunoassays for detection of myocardial injury. Serum cTnT and cTnI have been used as specific markers for myocardial infarction (1, 3, 9, 22, 23) and ischemic injury (1, 9, 18, 49). Serum cTnT and cTnI measurements are specific in the assessment of cardiac injury in the presence of skeletal muscle damage (1). Increased concentrations of cTnT and cTnI in serum have been related to the extent of myocardial damage also (1, 9, 22, 23). Decreased myocardial cTnT concentrations also indicate heart damage, as demonstrated in dogs after coronary artery occlusion (34).

The purpose of the present study was threefold. First, we determined whether the stress of prolonged intense exercise would damage the heart. Second, we determined whether increased exercise volume would induce an increase in heart damage. Third, we determined whether training before prolonged intense exercise would have a protective effect on heart damage. Measurement of cTnT concentrations in the serum and myocardium of swimming rats was used as the marker for myocardial injury.

	RESEARCH DESIGN AND METHODS

TOP ABSTRACT INTRODUCTION RESEARCH DESIGN AND METHODS RESULTS DISCUSSION REFERENCES

Animals and animal care. Male Sprague-Dawley rats (n = 143), weighing 285-335 g (body weight on the day before swimming), were used in these experiments. The rats were housed in cages in rooms regulated for temperature (23 ± 1°C), humidity (45-55%), and light cycle (0600-1800) and provided laboratory rat chow and water ad libitum. The research involving rodents in this study conforms with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and was approved by the Institutional Animal Care and Use Committee.

Experimental groups and exercise protocol. To attempt to minimize the general stress response of forced swimming, all rats were familiarized with swimming for 20 min (water temperature 35°C), without body weight attached to the rats, 2 days before any swimming protocols. The rats were then divided into three protocols. The first protocol consisted of 32 rats: 6 rats were used as sedentary controls and killed at rest, and 26 rats swam for 3.5 h (3.5S) in a swimming tank with 8% body weight (workload) attached to the tail. One rat was removed from this protocol when it was unable to finish the required forced swimming. After 3.5 h of swimming, the rats were towel dried, weighed, and killed immediately (0 h PS, n = 6), 3 h (3 h PS, n = 7), 24 h (24 h PS, n = 6), and 48 h after swimming (48 h PS, n = 6). The second protocol consisted of 63 rats: 12 rats were used as sedentary controls, and 51 rats swam for 5 h (5S) in a swimming tank with 8% body weight attached to the tail (25 or 26 rats swam together at each time). Three rats were removed from this protocol when they were unable to finish the required forced swimming. After the forced swimming, the rats were weighed and killed at 0 h PS (n = 12), 3 h PS (n = 12), 24 h PS (n = 12), and 48 h PS (n = 12). The third protocol consisted of 48 rats. The rats were pretrained for 8 days before a 5-h swim (T5S). Pretraining included 25 min of swimming on day 1 followed by an increase to 60 min on day 6 with no weight on the tail. The rats swam with a 5% tail weight for 20 min on day 7 and for 30 min on day 8. A 2-day rest period followed the training. Twelve rats were used as the pretrained control group; the other 36 rats swam for 5 h with 8% body weight attached to the tail (each time, 18 rats swam together). After the forced swimming, the rats were weighed and killed at 0 h PS (n = 12), 3 h PS (n = 12), and 48 h PS (n = 12). No rats were killed at 24 h PS in the T5S protocol. In all protocols, the 8% body weight load attached to the tail determined exercise intensity. The water temperature was 35°C, and the water depth was 26 cm.

Animal death and tissue sampling. Rats were anesthetized with pentobarbital sodium (Nembutal, 50 mg/kg ip). The abdominal cavity was quickly opened, and 10 ml of blood were drawn out from the abdominal aorta. Serum samples were collected and stored at 80°C for cTnT measurements. The chest cavity was then quickly opened, and the heart was excised, frozen on dry ice, and stored at 80°C until biochemical analysis for cTnT by Western blot. Within each group, rats were killed within 30 min.

Sample preparation. Frozen myocardial samples were cut into small pieces on dry ice-cooled sample plates and then added to 2 ml of ice-cold buffer [200 mM potassium phosphate (pH 7.4), 5.0 mM EGTA, 5.0 mM -mercaptoethanol, and 10% (vol/vol) glycerol] (46). The samples were homogenized three times in an ice bath for 10 s each at high speed with a Polytron tissue homogenizer (Brinkman Instruments, Westbury, NY) and then centrifuged at 3,000 g for 30 min at 4°C. The procedure yielded >98% recovery of cytosolic and myofibril proteins (46). The supernatants were used for Western blotting. Protein concentrations were determined using a modified Lowry method (21) with BSA as a standard. Serum cTnT measurements were obtained using the Roche second-generation cTnT immunoassay, which is 100% specific for cTnT (5). No cross-reactivity was observed in the cTnT assay at high concentrations of rat skeletal muscle troponin T (TnT; data not shown). Precision (percent coefficient of variation) at the upper reference limit for cTnT (0.1 µg/l) was <10%. The cTnT assay has been shown to be linear from 0.01 ng/ml (lower limit of detection) to 50.0 ng/ml (upper limit of linearity). Furthermore, this assay has been previously validated for use in several laboratory animals, including rats (30).

Histological examination. Left ventricular tissue was removed and immediately placed in 10% buffered Formalin. The apex of the heart was also removed and immediately frozen at 70°C after the animal was killed. The apex heart section remained frozen until preparation for histological examination, at which time it was placed in 10% buffered Formalin and permitted to thaw. One or two hearts from each protocol, including controls, were chosen randomly from the different time periods. From the randomly selected hearts, the entire apical segment and left ventricular tissues were embedded in paraffin, sectioned at 5-µm intervals, and stained with hematoxylin-eosin and/or trichrome stains.

Western blot analysis. Tissue homogenates were size fractionated on 12% SDS-polyacrylamide gels (43) and subsequently transferred to Hybond nitrocellulose membranes (Amersham, Arlington Heights, IL). Nonspecific binding sites were blocked by incubating the membranes in a blocking buffer [5.0% nonfat dry milk in Tris-buffered saline (TBS): 20 mM Tris · HCl (pH 7.6), 137 mM NaCl] for 1 h. A primary antibody specific for human cTnT (a gift from Fortron Bio-Scientific, Morrisville, NC) was diluted (1:1,000) in antibody buffer (1.0% nonfat dry milk in TBS) and incubated with the membranes for 2 h on a rotating cylinder. The membranes were washed three times with Tween-TBS for 30 min. Horseradish peroxidase-labeled secondary antibody (sheep anti-mouse IgG) was then incubated with the membranes for 1 h at a dilution of 1:3,000. The membranes were again washed three times in Tween-TBS buffer before a 1-min incubation with enhanced chemiluminescent substrate (Amersham). Light emission was detected by exposure to Fuji RX autoradiography film in the presence of Cronex intensifying screens. Signal intensities within the linear range were quantitated using laser densitometry (Molecular Dynamics, Sunnyvale, CA). Linearity was established by analysis of a standard curve generated with known amounts of proteins (with use of myocardial muscle homogenate) by Western blot. Myocardial cTnT was linear from 0.5- to 6-µg sample loading amounts of total protein. An excellent correlation (r = 0.99) between loading amounts and signal intensities of cTnT was obtained. To ensure that proteins on different sides of the gel were equally transferred, blotted, and detected, a control sample (pooled tissue samples) was loaded onto all the gels in at least two lanes as internal control (I) lanes, allowing for blotting quality control. Sample loading amount for cTnT was 2 µg/lane. Although there is a small amount of cross-reactivity against skeletal TnT with the Fortron antibody, rat skeletal TnT isoforms migrate distinctly to lower molecular weights and do not interfere in the determination of cTnT when the Fortron antibody is used (data not shown). However, no skeletal TnT isoforms were observed in the rat heart samples studied.

Statistical analysis. Data from all groups were analyzed using a two-way ANOVA to determine main effects across time after swimming. Tukey's post hoc test for multiple comparisons was used to determine significant differences from preexercise values when significant main effects were found. Significance was set at P < 0.05. Values are means ± SD unless stated otherwise. The values of Western blot are presented as percentage of the control group.

	RESULTS

TOP ABSTRACT INTRODUCTION RESEARCH DESIGN AND METHODS RESULTS DISCUSSION REFERENCES

Figure 1 shows that body weight was maximally decreased (P < 0.01) at 0 h PS for all three protocols: 7.4% after the 3.5S protocol, 9.4% after the 5S protocol, and 9.9% after the T5S protocol. Body weights remained decreased (P < 0.01) up to 24 h PS in the 3.5S protocol, up to 48 h PS in the 5S protocol, and up to 24 h PS in the T5S protocol.

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Body weight alterations of rats after 3.5 h of swimming (3.5S), 5 h of swimming (5S), and 8 days of training before a 5-h swim (T5S). * P < 0.05; ** P < 0.01 compared with before swimming. + P < 0.05; ++ P < 0.01 between groups. Data were obtained from rats killed 48 h after swimming (PS) from each protocol and expressed as percentages of body weight of controls before swimming.

Figure 2 shows that serum cTnT concentrations increased significantly (P < 0.01) above control concentrations (0.01 µg/l) after all three protocols at 0 and 3 h PS. The largest increases occurred in the 5S protocol, with a peak cTnT concentration of 0.15 ± 0.07 (range 0.06-0.26) µg/l at 0 h PS. This increase was significantly (P < 0.01) greater than any other cTnT increase in the 3.5S (mean 0.04 µg/l, range 0.03-0.06 µg/l) and T5S (mean 0.06 µg/l, range 0.03-0.12 µg/l) groups. The increases in the 3.5S and T5S groups at 0 and 3 h PS were not significantly different.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Serum cardiac troponin T (cTnT) concentrations of rats in 3.5S, 5S, and T5S protocols. ** P < 0.01 compared with controls. ++ P < 0.01 between groups.

Figure 3 shows a representative Western blot analysis of cTnT in the 0, 3, 24, and 48 h PS groups after the 5S protocol. One cTnT band migrated to a molecular weight position corresponding to ~40 kDa. An internal control sample, I, was used to confirm equivalence of protein transfer across each gel. Skeletal TnT isoforms were not detected.

View larger version (5K): [in this window] [in a new window]   Fig. 3.   Representative Western blot analysis of myocardial cTnT from 0, 3, 24, and 48 h PS groups after 5S protocol. One cTnT band migrated to a position corresponding to ~40 kDa. I, internal control (pooled tissue samples) used for determining equivalence of protein transfer across each gel.

Figure 4 shows the changes in cTnT determined by Western blot analysis in left ventricles after the 3.5S, 5S, and T5S protocols. cTnT did not significantly decrease after the 3.5S and T5S protocols. However, after the 5S protocol, cTnT decreased 14% (arbitrary scanning units 3,729 ± 153, P < 0.01) and 10% (arbitrary scanning units 3,909 ± 156, P < 0.05) at 0 and 3 h PS, respectively, compared with controls (arbitrary scanning units 4,349 ± 174). Furthermore, cTnT in the 5S protocol was significantly decreased 10% (P < 0.05) compared with the same PS groups in the 3.5S and T5S protocols. By 24 h PS, cTnT concentrations returned to preexercise control values.

View larger version (27K): [in this window] [in a new window]   Fig. 4.   cTnT alterations determined by Western blot analysis in left ventricles after 3.5S, 5S, and T5S protocols compared with nonswimming controls. * P < 0.05; ** P < 0.01 compared with controls. +P < 0.05; ++ P < 0.01 between groups.

As shown in Fig. 5, within the wall of the left ventricular myocardium, there are foci of myocyte injury with an interstitial inflammatory infiltrate consisting of neutrophils, lymphocytes, and histiocytes, as well as vesicular nuclei-enlarged chromatin patterns. These foci of myocyte injury in the left ventricles were observed in only four rats in the 5S group at 24 and 48 h PS. Furthermore, no histological evidence of recent or remote ischemic damage was evident in any of the apical sections examined.

View larger version (159K): [in this window] [in a new window]   Fig. 5.   Representative hematoxylin-eosin-stained cryosection of a left ventricle specimen obtained from a rat subjected to 5S protocol at 48 h PS. Damaged myocardial fibers and damaged intracellular areas are shown by foci of inflammatory infiltrates consisting of neutrophils, lymphocytes, and histiocytes (a) and vesicular nuclei-enlarged chromatin patterns (b).

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESEARCH DESIGN AND METHODS RESULTS DISCUSSION REFERENCES

The present study demonstrates that cTnT is released into the circulation from rat myocardium after 3.5 h of forced swimming, indicative of myocardial injury. Furthermore, an increase in exercise volume from 3.5 to 5 h of forced swimming resulted in a larger increase of circulating cTnT, indicative of a greater amount of myocardial injury compared with the 3.5-h model. However, an 8-day period of conditioning or pretraining before the 5-h forced swim resulted in a significant decrease in the release of cTnT into the circulation (Fig. 2) compared with the 5-h swimming protocol without pretraining, indicative of less myocardial injury than the 5-h model. Concordant with these findings, myocardial (left ventricle) cTnT concentrations were significantly decreased at 0 and 3 h PS in the 5S protocol, and histological evidence of myocyte damage in the 5S group at 24 and 48 h PS was also observed (Fig. 5). However, no measurable myocardial tissue loss of cTnT or histological evidence of damage was noted in the 3.5S or T5S protocol at any time period. These studies appear to be unique in studying the alteration of cTnT in myocardial tissue and serum after a stressful, forced swimming exercise protocol. Our findings are in agreement with recent findings in humans, in which ultraendurance exercise (Hawaii Ironman Triathlon) was demonstrated to cause myocardial damage, as indicated by increased serum cTnI and cTnT concentrations and alterations in echocardiography in 6 of 23 athletes immediately after the event (36). The cellular nature of damage in this study was not examined.

Our finding raises several questions. First, were the increased serum concentrations of cTnT in the 3.5S and T5S protocols, which do not show loss of myocardial cTnT, due to reversible or irreversible ischemic injury to the heart? Second, do the increased serum concentrations of cTnT in the 5S protocol, which does correspond to myocardial tissue loss of cTnT, represent a necrotic, irreversible injury to the heart? Serum cTnT is a highly sensitive and specific marker for myocardial injury in humans and animals (23, 28, 34, 46, 49). Serum cTnT is measured in the present study by the second-generation cTnT immunoassay from Roche, which is 100% specific for the heart, with no cross-reactivity against human skeletal muscle TnT (5, 35) or rat skeletal muscle TnT (data not shown). It is interesting to observe that serum cTnT concentrations in the 3.5S, 5S, and T5S protocols peaked ~5-6 h after the initiation of exercise. Thus the exercise volume and the release kinetics of cTnT from myocardial tissue into the circulation are likely responsible for the increased serum cTnT. The cTnT release patterns we observed in rats appear similar to those observed after injury to the myocardium after myocardial infarction in humans, in that 4-8 h are required for cTnT to become increased in the serum after the myocardial insult (23, 44, 49). Furthermore, serum cTnT concentrations in the three protocols returned to baseline concentrations by 24 h. We hypothesize that the rapid return to baseline suggests that release of cTnT from the exercise-induced myocardial injury is likely from localized minor, irreversible myocyte degeneration, since large irreversible injury to myocardial tissue after myocardial infarction in humans and in animal models remains increased for 5-10 days (22, 34, 49). The rapid increase and normalization of serum cTnT in our exercise model likely represent the loss of the 2-8% cytoplasmic compartment of cTnT, which has a half-life of <6 h (16, 46). The rapid normalization of cTnT to preexercise control values by 24 h of the decreased myocardium at 0 h PS in the 5S protocol likely is due to a reequilibration between the 95% myofibril compartment and the depleted 5% cytoplasmic cTnT compartment.

Histological evidence of myocyte damage in the left ventricles from four rats in the 5S group at 24 and 48 h PS demonstrated by inflammatory infiltrates of neutrophils, lymphocytes, and histiocytes and vesicular nuclei-enlarged chromatin patterns supports the hypothesis that loss of cTnT into the circulation is due to myocardial damage. No ischemic changes were identified histologically in any of the apical heart specimens for any of the time intervals after exercise, regardless of the duration of exercise. It is possible that the apex of the heart is not the best area for histopathological observations. However, the apex specimens were frozen before histological examination, which may have interfered with interpretation. Initial signs of ischemia, myocyte hypereosinophilia, and prominent contraction bands are not evident until 4-6 h after the onset of ischemia. Therefore, ischemic changes were not expected histologically in rats killed 0 or 3 h after exercise. The foci of myocyte injury, as shown in Fig. 5, are likely irreversible, suggestive of myocarditis, and may not always be identifiable by limited light-microscopic examination. No electron-microscopic examination was performed in this study. Mechanisms responsible for the myocyte injury were not studied.

Studies have shown that exercise of short duration induced regional myocardial dysfunction in dogs with severe coronary stenosis or ventricular hypertrophy (13-15). Repetitive episodes of exercise-induced ischemia resulted in cumulative postexercise regional myocardial dysfunction that lasted >2 h after the last episodes of exercise (14). Unfortunately, no serum biochemical markers for myocardial injury were tested in the above studies. Epidemiological studies have confirmed that heavy physical exertion can trigger myocardial infarction (26, 38). Studies also have shown that prolonged intense exercise can induce impaired left ventricular function and silent myocardial ischemia in some athletes (8, 17, 29, 36). However, most studies failed to show elevation of serum cTnT in symptomless athletes after marathon or ultramarathon running (1, 24, 39).

Our findings regarding alterations of the cardiac contractile protein cTnT after the prolonged single bout of forced swimming in rats are likely due to the development of our high-intensity (imposed tail weight) model, as well as the effect of severe stress placed on the rats. Our unpublished data indicated that rats swimming with 8% body weight attached to the tail generate a blood lactate (a marker for exercise intensity) concentration of 7.2 mM (vs. nonswimming control concentration of 3.7 mM), whereas rats run on a treadmill at 25 m/min fail to show any significant increase in blood lactate. A blood lactate level of 7 mM was reported in rats running up a 16° incline at a speed of 50 m/min with O₂ uptake of ~110 ml · kg¹ · min¹ (4). Furthermore, a maximal O₂ uptake of ~80 ml · kg¹ · min¹ was obtained from the swimming rats with 2% body weight attached to the tail (25). In comparison, the exercise intensity in our swimming model was four times higher than that with 2% body weight attached to the tail. Our study shows that a 40% increase of total exercise volume from 3.5 to 5 h of swimming resulted in an increase in serum cTnT from 0.04 µg/l to 0.15 ng/l (3.5 times above the cTnT concentration in the 3.5S group) immediately after swimming. The mechanism for this finding is not totally clear. It may relate to the different metabolic processes between the early and late stages of prolonged exercise, including depletion of glycogen stores (11, 12), alterations in glucose and fatty acid utilization (12), calcium alteration (6), free radical-mediated injury (45), and prolonged elevation of circulating catecholamines. Furthermore, no studies were undertaken to study whether there was evidence of impairment of cardiac function. However, numerous studies in the literature now clearly demonstrate the important prognostic role of increased serum cTnT levels in predicting short and long cardiac events in patients presenting with ischemic chest pain (10, 31). Furthermore, in an acute myocardial infarction study involving a dog model, increases in serum cTnT correlated with loss of myocardial cTnT (34). Thus, in humans, there appears to be an important physiological message for which medical and pharmacological management decisions are now influenced by increased serum cTnT values.

We also demonstrated a significant protective effect on prolonged exercise-induced myocardial damage as a consequence of the pretraining. In this study, the rats were pretrained for 8 days. Immediately after swimming, serum cTnT concentration in pretrained rats was 60% lower than in the untrained rats [0.06 µg/l in trained rats (T5S) vs. 0.15 µg/l in untrained rats (5S)]. Pretraining also blunted any decrease in myocardial cTnT after 5 h of swimming. These data suggest that even a short-term, low exercise volume of physical training could significantly protect the heart from prolonged, intense exercise-induced myocardial damage. Comparatively, in humans, the effect of regular physical exercise (preconditioning) may be very important. Our data may help in understanding the effect of regular physical exercise in reducing the long-term risk of myocardial events and the fact that people who are physically active have a lower risk of myocardial events triggered by heavy physical exertion (26).

In conclusion, our study shows that a single bout of stressful, prolonged, intense exercise induced damage as monitored by cTnT in the heart and serum in the myocardium of rats. Furthermore, an increase in exercise volume induced more myocardial damage, and training before stressful, prolonged, intense exercise provided protection from myocardial damage.