Exercise Training Attenuates Acute Doxorubicin-Induced Cardiac Dysfunction

Abstract

The use of doxorubicin, a highly effective antitumor antibiotic, is limited by a dose-dependent cardiotoxicity. The purpose of this study was to determine whether chronic exercise training (ET) prior to doxorubicin treatment would preserve cardiac function and reduce myocardial oxidative stress following treatment. Rats were exercise trained on a motorized treadmill or confined to sedentary cage activity for 12 weeks, then administered an intraperitoneal injection of doxorubicin (15 mg/kg) or 0.9% saline. Five days following the injections, hearts were isolated and Langendorf perfused to assess cardiac function and then processed for biochemical analyses. Doxorubicin treatment induced significant inotropic, lusitropic, and chronotropic cardiac dysfunction, reduced coronary flow, and increased cardiac lipid peroxidation in the sedentary animals. Doxorubicin treatment was also associated with a decrease in cardiac manganese superoxide dismutase protein expression and an increase in heat shock protein-72 (Hsp72) compared with saline-treated animals. Exercise training attenuated doxorubicin-induced cardiac dysfunction, and lipid peroxidation, and led to a greater cardiac expression of Hsp72 compared with the sedentary animals. The results of this study demonstrate for the first time that chronic exercise training before doxorubicin treatment protects against cardiac dysfunction following treatment, and provide evidence for a sustained increase in myocardial Hsp72 following exercise training and doxorubicin treatment in vivo.

Doxorubicin (DOX) is an antineoplastic antibiotic widely used in the management of a variety of cancers. Unfortunately, the clinical use of this highly effective anticancer drug is limited due to a severe, dose-dependent cardiotoxicity characterized by acute cardiac injury that may progress to irreversible cardiomyopathy and congestive heart failure months to years following treatment.^1,2 Early signs of cardiac dysfunction following DOX treatment have been found to be highly predictive of future heart failure in human² and animal studies.³ Substantial effort has been devoted to developing strategies for avoiding acute DOX cardiotoxicity (DCT) while maintaining the drug's antineoplastic effectiveness. Of the several molecular mechanisms of DCT that have been postulated (see⁴ for review), a DOX-mediated generation of reactive oxygen species (ROS) is widely believed to play a central role.^3,5,6 Prophylactic treatment prior to and during DOX treatment has proven to be only partially effective in the clinical setting, and the therapeutic value of DOX remains limited by its cardiotoxicity.

Exercise training (ET) has been shown to confer protection against a variety of acute and chronic myocardial insults, particularly ischemia and reperfusion (I/R).⁷ ET has been repeatedly shown to evoke increases in myocardial superoxide dismutase (SOD) and heat shock protein-72 (Hsp72) protein.^8,9 These have been proposed to mediate cardioprotection by providing resistance against the superoxide radical and preserving protein function during states of cellular stress, respectively. Indeed, animals with elevated myocardial levels of SOD or Hsp72 have exhibited cardioprotection against I/R^10,11 and DOX treatment.^6,12 However, exercise-induced cardioprotection (EIC) has been observed in the absence of any increase in SOD⁹ or Hsp72,^13,14 indicating that other protective mechanisms must be involved.

To our knowledge, outside of our laboratory, only five studies have been conducted that specifically address the effects of exercise on DCT.^15-19 Ji et al¹⁵ reported that administration of DOX after a single bout of acute exhaustive exercise reduced the exercise-induced increase in cardiac mitochondrial respiration. In an early report by Combs and co-workers,¹⁶ the researchers hypothesized that an acute bout of exhaustive exercise after DOX treatment (18 mg/kg-23 mg/kg) would exacerbate the drug-induced cardiotoxicity as evidenced by a decrease in survival rate. Results indicated that the survival rate was actually increased in those animals that had been subjected to exhaustive exercise after DOX treatment. In a study by Heon et al¹⁷ young rats were treated with 3 mg/kg DOX and subjected to 2 weeks of swim training, with results showing that training decreased the expression of proapoptotic markers in young female rats. In two recently published reports, Ascensao et al^18,19 demonstrated that ET conducted prior to DOX exposure protected the hearts of rats and mice against DCT. The authors showed that DOX-induced alterations in protein carbonyl formation, serum levels of cardiac proteins, and cardiomyocyte mitochondrial respiration were significantly attenuated by ET. However, none of these previous studies^15-19 reported any measures of cardiac function.

Recent evidence from our laboratory indicates that isolated hearts from physically active rats exhibit resistance against the left ventricular dysfunction and lipid peroxidation induced by DOX exposure in vitro.²⁰ However, the effect of ET on cardiac function following in vivo DOX treatment has not been previously examined. Therefore, the purpose of this study was to determine the effect of chronic endurance ET prior to DOX treatment on intrinsic cardiac function and lipid peroxidation following treatment. Given their putative role in mediating EIC, the effects of ET and DOX treatment on myocardial SOD isoforms and Hsp72 protein expression were also examined.

METHODS

Animal Subjects and Experimental Design

Forty-two male Sprague-Dawley rats (175 g-200 g) were obtained from Harlan Sprague Dawley, Inc. (Indianapolis, IN). All rats were housed two per cage in a temperature- and humidity-controlled animal facility, maintained on a 12:12 hour light:dark cycle, and provided standard rat chow and water ad libitum. Animals were randomly selected to remain sedentary (SED) or be exercise trained on a motorized treadmill (ET) for 12 weeks. Following the ET or sedentary period, rats were randomized into groups receiving a bolus i.p. injection of DOX (SED + DOX and ET + DOX groups) or saline (SED + SAL and ET + SAL groups). Five days following the injections, animals were killed for cardiac functional and biochemical studies described below. All experimental procedures were approved by the University of Northern Colorado Institutional Animal Care and Use Committee.

Exercise Training Protocol

One week after arrival, all rats were acclimated to the motorized rodent treadmill (Exer 3/6 model; Columbus Instruments, OH) by walking at 10 to 15 m·min⁻¹ for 5 to 10 min·d⁻¹ for 5 days. Following this acclimation period, animals were randomly assigned into a sedentary or ET group. Animals in the ET group were trained on the treadmill for 12 weeks according to the running protocol described in Table 1. This regimen is similar to those reported by others to induce increases in myocardial antioxidant enzyme protein concentration and activity,²¹ with slight modifications based upon pilot studies with rats on equipment in our laboratory. Animals were trained 5 days·wk⁻¹ (Monday-Friday) during their active dark cycle to promote compliance and reduce the stress associated with the forced ET protocol. When necessary, rats were motivated to run by manual prodding and tapping on the equipment, but no electric shock was used. Animals consistently requiring more than 20 prods per session were excluded from the study. Rats in the sedentary group were confined to sedentary cage activity in the same room as the trained animals. Sedentary rats were handled daily and cages were placed on the treadmill table for one ET session each day to expose these animals to the same noise, vibration, and handling stress experienced by the ET animals.

TABLE 1: Treadmill Exercise Training Protocol

Doxorubicin Treatment

Twenty-four hours following completion of the 12-week ET or sedentary period, animals were randomly selected to receive either DOX or SAL treatment. A single 15 mg·kg⁻¹ i.p. injection of doxorubicin·HCl (Bedford Labs, Bedford, OH) was administered to exercise-trained (ET + DOX; n = 15) and sedentary (SED + DOX; n = 15) rats. This dose has been shown in our pilot experiments and by others to significantly increase myocardial oxidative stress and depress cardiac function 1 to 5 days following treatment.^22,23 Rats in the ET + SAL (n = 6) and SED + SAL (n = 6) groups received a 1 mL bolus of 0.9% sterile saline in place of DOX. The number of rats in the SED + DOX and ET + DOX groups was greater to accommodate the potential mortality following DOX treatment. All rats were confined to sedentary cage activity for 5 days following the injections to allow time for the development of overt DOX-induced cardiac dysfunction before being killed for cardiac functional and biochemical experiments.

Assessment of Cardiac Function

Cardiac function was determined in isolated hearts using a constant pressure Langendorf perfusion preparation. All rats were anesthetized with 50 mg·kg⁻¹ sodium pentobarbital (Sigma Chemical, St. Louis, MO) administered i.p. along with 500 U sodium heparin serving as an anticoagulant. Following confirmation of anesthesia by the absence of the tail-pinch reflex, the heart was rapidly excised, and the ascending aorta was cannulated for retrograde perfusion with a Krebs bicarbonate buffer consisting of: 120 mM NaCl, 5.9 mM KCl, 2.5 mM CaCl₂, 1.2 mM MgCl, 25 mM NaHCO₃, 17 mM glucose, and 0.5 mM EDTA. All chemicals were ACS reagent grade obtained from Sigma-Aldrich (St. Louis, MO). Buffer was saturated with 95% O₂/5% CO₂ and maintained at 37^oC in a water-jacked reservoir suspended above the isolated heart, providing constant pressure of approximately 80 mm Hg. To assess left ventricular function, a microtip catheter pressure transducer (SPR-671; 1.4F; Millar Instruments Inc., Houston, TX) was inserted into the left ventricular (LV) cavity via the mitral valve for determination of LV pressures. Intrinsic (non-paced) heart rate (HR) was recorded following a 15-minute equilibration period, after which time the hearts were paced at 300 bpm via the atria using a stimulus isolator in conjunction with a PowerLab/8e system (ADInstruments, Colorado Springs, CO). After an additional 5-minute stabilization period, baseline values of end diastolic pressure (EDP), LV developed pressure (LVDP), and its maximal and minimal first derivatives (dP/dt_max and dP/dt_min, respectively) were calculated and recorded under paced conditions using the PowerLab data acquisition system and software. In addition, coronary flow (CF) was quantified by measuring the rate of perfusate flow from the resivoir to the aortic cannula (Transonic, Ithaca, NY). Immediately following the isolated perfused heart protocol, the heart was removed and the left ventricle isolated. Left ventricular tissue was frozen in liquid nitrogen and stored for subsequent biochemical analyses.

Myocardial Lipid Peroxidation

To examine the effect of DOX treatment on myocardial lipid peroxidation, a common manifestation of cellular injury mediated by ROS, malondialdehyde and 4-hydroxy-alkenals (MDA + 4-HAE) were analyzed in LV of all hearts following the isolated heart functional experiments described above using a spectrophotometric assay kit (Calbiochem, San Diego, CA) according to the manufacturer's instructions. Frozen LV tissue was homogenized in ice-cold 20 mM Tris-HCl, pH 7.4, containing 5 mM butylated hydroxytoluene (to prevent lipid peroxidation during homogenization) at a dilution of 1:5 (w/v) and centrifuged at 3000g for 10 minutes at 4°C. The supernatant was collected and used for spectrophotometric analyses. MDA + 4-HAE is expressed relative to the protein content of sample supernatant determined by the method of Bradford²⁴ using bovine serum albumin as the standard.

Left Ventricular SOD and Hsp72 Content

To examine the effect of ET on classic candidates of exercise-induced cardioprotection, Hsp72, CuZnSOD, and MnSOD contents were determined in LV samples using Western immunoblotting methods previously described.²⁴ Briefly, LV samples were homogenized in ice-cold homogenizing buffer containing (in mM) 100 KCl, 50 MOPS, 5 MgCl₂, 1 ATP, 1 EGTA (pH 7.4), and protease inhibitor cocktail (Roche Diagnostics, Mannheim, Germany). Homogenates were then diluted 1:3 (vol/vol) in lysis buffer containing 20 mM HEPES, 150 mM NaCl, 1 mM EDTA, 1% NP-40, 0.1% SDS (pH 7.4), allowed to incubate at room temperature for 10 minutes, then centrifuged at 3000g for 5 minutes. The supernatant was then assayed for total protein,⁵ and a volume of supernatant containing 50 μg (Hsp72) or 3 μg (SOD isoforms) of LV protein was electrophoresed on polyacrylamide gels for separation of proteins by molecular weight. Proteins were then transferred to membranes (Amersham, Piscataway, NJ) and incubated with an alkaline phosphatase-conjugated polyclonal antibody specific for rat Hsp72, or polyclonal antibodies specific for CuZnSOD or MnSOD (Stressgen, Victoria, BC, Canada). Hsp72 blots were then reacted with 5-bromo-4-chloro-3-indolyl phosphate-nitro blue tetrazolium substrate (Sigma Chemical, St. Louis, MO) (Hsp72) and scanned for band density analysis. SOD isoform blots were incubated with a secondary horseradish peroxidase-conjugated goat anti-rabbit antibody (Santa Cruz Biotechnology, Santa Cruz, CA) followed by a chemiluminescent substrate (Western Lighting, Perkin-Elmer Life Sciences, Boston, MA) and developed on film. Quantification of relative Hsp72 and SOD isoforms band densities were performed on computerized scans of immunoblots using ImageJ densitometry software (NIH, Bethesda, MD). Band densities are expressed as a percentage of the mean SED + SAL band density from the same membrane.

Statistical Analyses

All data are presented as means ± SE. Data collected prior to drug treatment (starting and injection body weights) were analyzed by one-way analyses of variance (ANOVA) with Tukey HSD tests post hoc. Two-way ANOVAs were used to determine significant differences in end-point data due to the main effects (ET and DOX treatment) and the interaction of these two factors. When two group means were compared separately, independent sample t tests were conducted with the Bonferroni adjustment for multiple comparisons. A significance level of P ≤ 0.05 was established for all statistical analyses.

RESULTS

General Observations

A summary of animal characteristics before and after the experimental treatments is provided in Table 2. There were no significant group differences in mean body weight at the beginning of the 12-week ET or sedentary period. Three of the 30 rats in the ET groups were excluded from the study during the course of the ET period due to persistent non-compliance with the treadmill training protocol. ET significantly attenuated weight gain during the 12-week ET period, indicated by significantly lower body weights in the ET versus SED animals at the time of drug treatment (P < 0.05). DOX induced significant weight loss in both the ET and SED animals during the 5-day period following treatment (DOX effect P < 0.05), with no interaction observed between ET and DOX treatment. DOX and ET both led to significant increases in heart-to-body weight ratio with no interaction observed between these two factors. However, this result should be interpreted with caution, as it is likely attributable to the marked depressive effect of these two factors on body weight. Indeed, DOX actually induced a reduction in absolute heart weight (effect of DOX P < 0.05), with no effect of ET or any interaction of these two factors.

TABLE 2: Effect of Exercise Training and Doxorubicin Treatment on Animal Characteristics

Mortality within the 5 days following DOX injections was 46% in the SED + DOX group compared with 33% in the ET + DOX group. Surviving DOX-treated animals all exhibited a significant accumulation of fluid in the abdominal cavity at the time of sacrifice (ascites), which has been previously reported following a 15 mg·kg⁻¹ i.p. DOX injection.²⁵ ET + DOX rats exhibited slightly lower mean volume of ascites than SED + DOX rats, though this difference was not statistically significant (P = 0.07).

Effect of Doxorubicin and Exercise Training on Cardiac Function

The effects of ET and DOX treatment on cardiac functional parameters are presented in Table 3. Two-way ANOVA analyses indicated that DOX treatment significantly suppressed all measures of cardiac function except EDP (effect of DOX P < 0.05). ET had no independent effect on cardiac function, but significantly attenuated the adverse effects of DOX treatment, indicated by a significant ET × DOX interaction on HR, LVDP, dP/dt_max, and dP/dt_min (P < 0.05). HR was 19% lower in SED + DOX hearts compared with the ET + DOX. LVDP and dP/dt_max, measures of LV systolic function, were 46% and 39% lower in the SED + DOX animals than in ET + DOX rat hearts, respectively. The rate of LV relaxation (dP/dt_min) was 46% slower in SED + DOX versus ET + DOX hearts. DOX treatment significantly impaired CF (ml·g⁻¹·min⁻¹; effect of DOX P < 0.05), resulting in a 24% reduction in SED + DOX versus SED + SAL hearts. There was a trend toward a protective effect of ET (interaction effect P = 0.076), with no independent effect of ET alone.

TABLE 3: Effect of Exercise Training and Doxorubicin Treatment on Cardiac Function

Myocardial Lipid Peroxidation

Left ventricular homogenates obtained from rats 5 days after saline or DOX injection were analyzed for MDA + 4-HAE, products of lipid peroxidation, to provide an index of oxidative stress in the myocardium (Fig. 1). DOX treatment significantly increased MDA + 4-HAE levels (DOX effect P < 0.05), indicating that DOX treatment induced significant lipid peroxidation in the heart. ET had no independent effect on MDA + 4-HAE levels, but completely abolished the DOX-induced increase (interaction effect P = 0.05), indicating that ET protected against DOX-induced lipid peroxidation in the heart.

FIGURE 1: Lipid peroxidation in left ventricular homogenates. Data are means ± SE. Two-way ANOVA (Exercise × DOX) analyses indicated that DOX treatment significantly increased MDA + 4-HAE levels (DOX effect P < 0.05), indicating that DOX treatment induced significant lipid peroxidation in the heart. ET had no independent effect, but completely abolished the DOX-induced increase (interaction effect P = 0.05), indicating that ET protected against DOX-induced lipid peroxidation in the heart.

Left Ventricular SOD and Hsp72 Content

Figure 2 summarizes LV SOD protein expression. There was no effect of DOX or ET on LV CuZnSOD protein expression. MnSOD protein expression was significantly depressed by DOX treatment (DOX effect P < 0.05), resulting in levels 10% to 12% lower in SED + DOX group than in saline-treated groups. ET had no independent effect on MnSOD protein, and no interaction with DOX was indicated by (P = 0.26). Myocardial Hsp72 protein (Fig. 3) was significantly increased by ET (effect P < 0.05) and DOX treatment (effect P < 0.05), resulting in 45% and 28% greater levels in the ET + SAL and SED + DOX groups compared with SED + SAL, respectively. No ET × DOX interaction effect was detected (P = 0.24), indicating that the greater Hsp72 expression in the ET + DOX rats was due simply to the additive effects of ET and DOX exposure.

FIGURE 2: MnSOD (A) and CuZnSOD (B) content in left ventricular homogenates, with representative sample blots and mean band densities ± SE for each group relative to SED + SAL levels. MnSOD protein expression was significantly depressed by DOX treatment (DOX effect P < 0.05), with no significant effect of ET or any interaction of these two factors.

FIGURE 3: Hsp72 protein content in left ventricular homogenates. Representative sample blots and mean band densities for each group densities ± SE for each group relative to SED + SAL levels. Myocardial Hsp72 protein was significantly increased by ET (effect P < 0.05) and DOX treatment (effect P < 0.05). No ET × DOX interaction effect was detected (P = 0.24).

DISCUSSION

The results of this study demonstrate for the first time that endurance ET prior to DOX treatment in vivo protects against DOX-induced cardiac dysfunction. In addition, we have corroborated previous evidence¹⁹ that ET prevents DOX-induced myocardial lipid peroxidation and reduces mortality following treatment. These exercise-induced improvements were associated with an increased myocardial protein expression of Hsp72, but not SOD, in the ET rats compared with their sedentary counterparts.

Effect of Doxorubicin and Exercise Training on Animal Characteristics

Doxorubicin treatment resulted in a marked loss of body weight in the 5 days following treatment in the SED + DOX and ET + DOX rats. Weight loss has been consistently reported following DOX treatment and can be most likely attributed to a reduction in food intake following treatment and an inhibition of protein synthesis that results from the drug's antineoplastic effect. The significant effect of DOX treatment observed on heart-to-body weight ratio appears to be due to the marked DOX-mediated loss in body weight rather than any increase in absolute cardiac mass.

Exercise trained animals gained significantly less weight than sedentary animals after the 12-week ET or sedentary period. This may be explained by prior evidence that male rats do not compensate for increased physical activity by increasing their caloric intake, resulting in a lower mean weight gain over time compared with sedentary age-matched animals.²⁶ ET led to significant cardiac hypertrophy indicated by greater absolute and relative heart weights in the trained versus sedentary animals, which is consistent with previously reported data from male rats treadmill-trained for 12 weeks.²⁷

Doxorubicin-Induced Cardiac Dysfunction

Doxorubicin treatment significantly impaired chronotropic, lusitropic, and inotropic cardiac function in the sedentary DOX animals. Intrinsic HR was significantly lower in the sedentary DOX rats than in all other groups, which corroborates other reports of a decreased HR in rat²⁸ and mouse²⁹ hearts perfused with DOX in vitro. The mechanism of this effect is unclear, but may involve injury or loss of the pacemaker cells in the SA node and/or other cells throughout the cardiac conduction system.

LVDP and dP/dt_max in the hearts of DOX-treated animals were markedly lower than in SED + SAL rats, indicating significant impairment of systolic LV function following DOX treatment. DOX-induced systolic dysfunction is the hallmark of clinical DCT, and has been reported in humans as a decrease in LV ejection fraction and in rats as a decrease in LV fractional shortening in vivo following 15 mg·kg⁻¹ of DOX treatment.³⁰ Several mechanisms have been proposed to explain the depression of LV systolic function that occurs within days following DOX treatment including cardiomyocyte apoptosis,³¹ myofibrillar damage,³¹ impaired Ca²⁺ handling,³² and interference with cardiomyocyte energy metabolism.³³ An enhanced generation of reactive oxygen species (ROS) by DOX has been proposed to mediate many of these processes, and may therefore be a key upstream event in the cellular pathophysiology of acute DCT.^34,35

Diastolic dysfunction was also evident in the sedentary DOX-treated animals in this study, which has been previously reported in isolated rat hearts³⁶ and in humans.² Optimal diastolic function relies on efficient removal of calcium ions (Ca²⁺) from the cytoplasm, which is principally achieved by the ATP-dependent pumping of Ca²⁺ into the sarco(endo)plasmic reticulum by the Ca²⁺-ATPase (SERCA). DOX has been reported to reduce Ca²⁺ uptake by SERCA in isolated cardiomyocytes^37,38 and induce cytoplasmic Ca²⁺ overload in vivo.³⁹ Whereas the precise mechanisms by which DOX impairs cardiomyocyte Ca²⁺ uptake are not fully understood, DOX-induced production of ROS has been shown to mediate oxidative inhibition of the SERCA protein,³⁷ decrease SERCA2 gene transcription,⁴⁰ and interfere with mitochondrial ATP production.³³

Several investigators have examined myocardial lipid peroxidation products (e.g., MDA) as an indicator of DOX-induced oxidative injury in the heart. In the current study, MDA + 4-HAE levels were significantly higher in the LV tissue isolated from the SED + DOX animals compared with SED + SAL, which is in agreement with previous reports of similar elevations in cardiac MDA levels 1 to 7 days following a single 15 mg·kg⁻¹ i.p. DOX injection in rats^23,41 and mice.⁴² Our finding that DOX treatment resulted in a significant depression of LV MnSOD, while minimal (∼10%), suggests that this may have contributed to the increase in lipid peroxidation observed in the hearts of DOX-treated animals. Given the important role of ROS-mediated injury in acute DCT, it is reasonable to attribute the cardiac dysfunction observed in DOX-treated rats in this study, at least in part, to an increase in myocardial oxidative stress.

Doxorubicin treatment also resulted in an increase in myocardial Hsp72 protein expression, indicated by levels that were 128% of SED + SAL. While increased Hsp72 expression has been reported in neonatal mouse myocardial cells incubated with DOX in vitro,⁴³ this study is the first to report a sustained increase in Hsp72 protein in the hearts of rats treated with DOX in vivo. This result is not surprising given substantial evidence that oxidant stress and ROS are primary stimuli for Hsp72 gene transcription and protein expression in cardiomyocytes,^44,45 but may provide further insight into the subcellular effects of DOX in the heart.

Coronary Flow

In the current study, DOX treatment led to a significant reduction in CF per gram of cardiac tissue in the sedentary animals. There have been reports of reduced CF in hearts perfused with DOX,²⁹ and in vitro studies indicate that DOX can induce vascular endothelial cell apoptosis,⁴⁶ inhibit endothelial nitric oxide synthase activity,^47,48 and interfere with endothelium-dependent vasorelaxation.⁴⁹ However, a recent study reported no significant effect of supraclinical (up to 50 μM) DOX concentrations on endothelium-dependent vasorelaxation or nitric oxide production in the isolated rat aorta.⁵⁰ While it is plausible that a DOX-mediated increase in ROS production in the myocardium may interfere with coronary vascular endothelial function, the reduction in CF in DOX-treated animals may be secondary to changes in the lusitropic state of the myocardium. Considering that most CF occurs during the diastolic phase of the cardiac cycle, impaired LV relaxation may have restricted coronary perfusion and reduced CF.

Cardioprotective Effects of Exercise Training

Exercise training did not significantly alter LVDP, dP/dt_max, dP/dt_min, EDP, or CF from sedentary levels in the current study, which corroborates previous reports that ET does not significantly alter baseline parameters of LV function in the isolated perfused rat heart.^27,51 However, ET did ameliorate the cardiac dysfunction associated with DOX treatment. The precise mechanism of this cardioprotective effect cannot be determined from the experiments in the current study; however our evidence that myocardial oxidative stress was attenuated in trained animals exposed to DOX suggests that ET may have attenuated the ROS-mediated mechanisms of DOX-induced cardiac dysfunction discussed above.

Several mechanisms may be responsible for the prevention of oxidative injury in the ET + DOX hearts, including an exercise-induced elevation in antioxidant enzymes and/or other cardioprotective stress proteins previously reported to mediate EIC against oxidative stress. In this study, no significant effect of ET on the protein expression of SOD isoforms was detected in the heart at the time of sacrifice (6 days following cessation of training). Biochemical analyses were not conducted in hearts immediately prior to DOX treatment, but previous investigations have demonstrated elevations in the activity and protein content of SOD isoforms in the heart 24 to 48 hours following ET.^9,52 Therefore, it is plausible that ET-induced increases in myocardial SOD may have been present at the time of DOX treatment, but declined to sedentary levels during the 6-day sedentary period following completion of the ET protocol.

Exercise training led to a significant increase in myocardial Hsp72 content in the heart compared with sedentary animals. Myocardial Hsp72 is an inducible stress protein believed to exert cardioprotective effects by stabilizing and refolding damaged proteins and contributing to antioxidant defenses during states of cellular stress.^53,54 Induction of Hsp72 is well known to occur in myocardial tissue following ET, and has been associated with the exercise-induced preservation of cardiac function during states of oxidative stress such as I/R^8,52 and H₂O₂ exposure.⁵¹ While exercise-induced induction of myocardial Hsp72 in the heart has been well established, our finding is particularly interesting because it indicates that the exercise-induced increase in Hsp72 is sustainable for at least 6 days after cessation of ET. Lennon et al¹⁴ recently reported that myocardial Hsp72 content was elevated at 1 and 3 days after short-term ET, but declined to sedentary control values by 9 days after training. Whereas Hsp72 content was not determined in hearts at the time of DOX treatment in the current study, prior evidence indicates that Hsp72 induction occurs in the heart within 24 hours of ET,⁹ which suggests that the ET-induced elevations in Hsp72 may have been in place at the time of DOX treatment.

There are several mechanisms by which an increase in myocardial Hsp72 could have contributed to the preservation of cardiac function observed in the ET + DOX hearts. Heat shock proteins of the 70-kDa family have been reported to protect against apoptosis,⁵⁵ prevent lipid peroxidation,⁵⁶ preserve Ca²⁺ handling,⁵⁷ preserve mitochondrial integrity,^58,59 and increase ATP and phosphocreatine content⁶⁰ in cardiac tissue exposed to a variety of oxidative stressors. These effects of Hsp72, directly or indirectly, oppose several proposed mechanisms of DCT discussed above, and may therefore have contributed to the preservation of cardiac function observed in the ET + DOX animals in the present study.

SUMMARY AND CONCLUSION

This study demonstrates for the first time that ET prior to DOX treatment in vivo preserves intrinsic cardiac function following treatment. This cardioprotective effect was associated with an exercise-induced increase in myocardial Hsp72 content, a prevention of DOX-induced myocardial lipid peroxidation, and reduced mortality following DOX treatment. These data suggest that endurance ET may reduce the risk of the cardiotoxic manifestations of DOX treatment, which may serve as the basis for future studies exploring the therapeutic value of ET in cancer patients undergoing long-term DOX chemotherapy.

LIMITATIONS

Based upon the results of this investigation, the following limitations and recommendations for future study are provided. The present study employed a cancer-free rat model of acute DCT to avoid the potential confounding variability that would be introduced by using a cancer model. While it is reasonable to assume that the cardioprotective effects of ET would also occur in the presence of malignant cancer, future study is needed to confirm this hypothesis. In addition, it is important to determine if ET has any effect on the therapeutic index (ie, the antineoplastic effect) of DOX in cancer models. Finally, whereas the results of this study indicate a cardioprotective effect of ET on DCT observed 5 days following treatment, future study is needed to determine the effect of exercise on chronic DCT that may develop weeks to months following treatment.