The objective of this work was to test the hypothesis that endurance training may be protective against in vivo doxorubicin (DOX)-induced cardiomyopathy through mitochondria-mediated mechanisms. Forty adult (6–8 wk old) male Wistar rats were randomly divided into four groups (n = 10/group): nontrained, nontrained + DOX treatment (20 mg/kg), trained (14 wk of endurance treadmill running, 60–90 min/day), and trained + DOX treatment. Mitochondrial respiration, calcium tolerance, oxidative damage, heat shock proteins (HSPs), antioxidant enzyme activity, and apoptosis markers were evaluated. DOX induces mitochondrial respiratory dysfunction, oxidative damage, and histopathological lesions and triggers apoptosis (P < 0.05, n = 10). However, training limited the decrease in state 3 respiration, respiratory control ratio (RCR), uncoupled respiration, aconitase activity, and protein sulfhydryl content caused by DOX treatment and prevented the increased sensitivity to calcium in nontrained + DOX-treated rats (P < 0.05, n = 10). Moreover, training inhibited the DOX-induced increase in mitochondrial protein carbonyl groups, malondialdehyde, Bax, Bax-to-Bcl-2 ratio, and tissue caspase-3 activity (P < 0.05, n = 10). Training also increased by ∼2-fold the expression of mitochondrial HSP-60 and tissue HSP-70 (P < 0.05, n = 10) and by ∼1.5-fold the activity of mitochondrial and cytosolic forms of SOD (P < 0.05, n = 10). We conclude that endurance training protects heart mitochondrial respiratory function from the toxic effects of DOX, probably by improving mitochondrial and cell defense systems and reducing cell oxidative stress. In addition, endurance training limited the DOX-triggered apoptosis.
- heart mitochondria
- oxidative phosphorylation
alterations in heart physiology have been related to mitochondrial dysfunction, including depressed mitochondrial respiration, which is frequently associated with several cardiomyopathies, particularly those linked with increased oxidative stress (17, 55). The heart is particularly sensitive to oxidative damage because of its relatively low levels of some antioxidant enzymes, its large density/volume of mitochondria, and its elevated rate of oxygen consumption (19). Moreover, increasing evidence indicates that, under certain conditions, cardiac mitochondria are major production sites and primary targets for reactive oxygen species (ROS) through the so-called electron leakage from the electron transport complexes (ETC) (28). For instance, an acute and severe bout of physical exercise increases oxygen consumption around sixfold and also enhances mitochondrial oxygen flux, with subsequent additional ROS production (see Ref. 3 for references). Nevertheless, moderate and systematic exercise generally is an excellent tool to provide enhanced parallel resistance to the cardiac muscle (3, 4). It is possible that mitochondrial structural and biochemical adaptations induced by training, such as increased volume density, heat shock protein (HSP) expression, and upregulation of antioxidant enzyme activity, could be related to training-induced enhanced cardiac resistance (18, 25, 44). This phenomenon, usually referred to as cross-tolerance, has been demonstrated by several studies in which endurance training induced cardioprotection after acute stress stimuli (13, 43).
Most of the training-related cross-tolerance cardiac studies used ischemia-reperfusion (I/R) as a model to test cardiac susceptibility to oxidative damage and dysfunction. However, in addition to I/R, other stimuli associated with distinct known mechanisms of cellular injury, such as in vivo treatment with the antitumor antibiotic doxorubicin (DOX, or adriamycin), should also be considered in an analysis of the beneficial effects of training cross-tolerance. DOX induces a dose-related and potentially lethal cardiotoxicity, which may be due, at least partially, to free radical production (12, 15). Moreover, there is considerable evidence that mitochondria are principal targets in the development of DOX-induced cardiomyopathy (59) and that the onset and severity of DOX toxicity correlate with disturbances in heart mitochondrial function and bioenergetics (55). Therefore, ultrastructural data from rats exposed to acute and cumulative doses of DOX revealed important mitochondrial morphological changes, namely, significant swelling and loss of cristae, simultaneously with abnormal mitochondrial respiration and calcium-loading capacity (42, 48).
Recent data from our studies of heart tissue homogenate clearly demonstrated that endurance training induces cross-tolerance against DOX cardiotoxicity (2). However, the protective effect of endurance training on altered mitochondrial function of DOX-treated animals has not been studied. For the above-mentioned reasons, the main purpose and the novelty of this study consist of assessing the effect of endurance training on DOX-induced mitochondriopathy (48, 57) and relating these findings to mitochondrial function and markers of oxidative stress. Because gene transfection-mediated overexpression of cardiac HSPs have been extensively reported to result in enhanced myocardial mitochondrial tolerance (54) and because endurance training is known to upregulate cardiac antioxidant enzymes (40, 43), we also investigated whether endurance training-induced increases in the levels of HSP and in antioxidant enzyme activity could be related to protection of mitochondrial function. Furthermore, because it is known that mitochondrial dysfunction has an important role in DOX-induced apoptosis (9), another purpose of the present study was to analyze the effect of training on DOX-induced apoptosis.
Forty male Wistar rats (6–8 wk old, 200 g body wt at the beginning of the experiments) were housed in collective cages (2 per cage) in a room at normal atmosphere (21–22°C, ∼50–60% humidity) in a 12:12-h light-dark cycle; they received food and water ad libitum. The animals were randomly divided into two groups: trained and nontrained. Because of the female estrogen-protective effect on cardiac tissue reported elsewhere (8), only male animals were used. The Ethics Committee of the Scientific Board of the Faculty of Sport Sciences approved the experimental protocol, which followed the National Institutes of Health Guide for Care and Use of Laboratory Animals.
The trained animals were exercised 5 days/wk (Monday–Friday) for 14 wk on a motor-driven treadmill. The treadmill speed and grade were gradually increased over the course of the 14-wk training period (Table 1), including 5 days of habituation to the treadmill with 10 min of running at 30 m/min and 0% grade with daily increases of 10 min until a 50-min training period was achieved. Habituation was followed by 1 consecutive wk of continuous running (60 min/day) at 30 m/min and 0% grade. This protocol proved to be efficient in raising cardiac dimensions (hypertrophy) and in altering antioxidant biochemistry to protect cardiac tissue during in vivo I/R (43). The nontrained animals were not exercised but were placed on a nonmoving treadmill three times per week (10–30 min/session) with the purpose of normalizing the possible environmental stress induced by the treadmill without promoting physical training adaptations.
At 24 h after the end of the endurance-training program, animals from the nontrained and trained groups were again randomly separated into two subgroups: DOX and placebo treatment. Thus nontrained animals were distributed into nontrained + placebo (NT + P, n = 10) and nontrained + DOX (NT + DOX, n = 10) groups and trained animals into trained + placebo (T + P, n = 10) and trained + DOX (T + DOX, n = 10) groups. The placebo groups were injected with a saline solution (0.9% NaCl ip) and the DOX groups with a single dose of DOX (20 mg/kg ip) in solution according to previous studies (9, 38). Both treatments were carried out 24 h before the animals were killed.
Euthanasia and plasma, heart, and soleus extractions.
Animals were anesthetized with diethyl ether and placed in the supine position. The abdominal cavity was opened to expose the inferior vena cava, and a ∼2-ml blood sample was collected in a heparinized tube. The blood was immediately centrifuged (5 min at 5,000 g, 4°C), and an aliquot of plasma was obtained and stored at −80°C for biochemical determination of cardiac troponin I (cTnI). The chest cavity was quickly opened, and the hearts were rapidly excised, rinsed, carefully dried, and weighed. A portion (∼20–25 mg) of cardiac ventricle and one soleus muscle were separated and homogenized in homogenization buffer (0.05 M Tris, 0.03 M l-serine, and 0.06 M boric acid, pH 7.6; 100 mg of tissue/ml of buffer) using a Teflon pestle on a motor-driven Potter-Elvehjem glass homogenizer at 0–4°C (3–5 times for 5 s at low speed, with a final burst at a higher speed). Homogenates were centrifuged (2 min at 2,000 g, 4°C) to eliminate cellular debris, and the resulting supernatant was stored at −80°C for later determination of antioxidant enzyme activity, HSP-70 expression, caspase-3 activity (cardiac ventricle), and citrate synthase (CS) activity (soleus muscle). The protein content of the cardiac muscle homogenate and soleus muscle was assayed, with BSA used as standard according to Lowry et al. (33). Additional small sections were taken from the left ventricle papillary muscles for morphological qualitative evaluation.
Isolation of rat heart mitochondria.
Rat heart mitochondria were prepared using conventional methods of differential centrifugation as follows (5). Briefly, the animals were euthanized as stated above, and the hearts were immediately excised and finely minced in an ice-cold isolation medium containing 250 mM sucrose, 0.5 mM EGTA, 10 mM HEPES-KOH (pH 7.4), and 0.1% defatted BSA (catalog no. A 7030, Sigma). The minced blood-free tissue was resuspended in 40 ml of isolation medium containing protease subtilopeptidase A type VIII (catalog no. P 5380, Sigma; 1 mg/g tissue) and homogenized in a tight-fitting homogenizer (Teflon-glass pestle). The suspension was incubated for 1 min (4°C) and rehomogenized. The homogenate was centrifuged at 14,500 g for 10 min. The supernatant fluid was decanted, and the pellet, essentially devoid of protease, was gently resuspended in its original volume (40 ml) with a loose-fitting homogenizer. The suspension was centrifuged at 750 g for 10 min, and the resulting supernatant was centrifuged at 12,000 g for 10 min. The pellet was resuspended using a paint brush and repelleted at 12,000 g for 10 min. EGTA and defatted BSA were omitted from the final washing medium. Mitochondrial protein content was determined by the biuret method calibrated with BSA. All isolation procedures were performed at 0–4°C.
An aliquot of the mitochondrial suspension was taken after isolation for immediate measurement of aconitase activity. Additional mitochondrial aliquots were separated and frozen at −80°C for later determination of protein carbonyls, sulfhydryl groups, malondialdehyde (MDA), HSP-60, and pro- and antiapoptotic Bax and Bcl-2 proteins.
The remaining mitochondrial suspensions were used within 4 h for respiratory assays and were maintained on ice (0–4°C) throughout this period. Isolation procedures yielded well-coupled mitochondria: the respiratory control ratio (RCR) of isolated mitochondria varied from 7 to 10 (with glutamate + malate) or from 3 to 4 (with succinate + rotenone) for controls, as determined according to the method of Estabrook (16).
Mitochondrial oxygen consumption assays.
Mitochondrial respiratory function was measured polarographically, at 25°C, with a Biological Oxygen Monitor System (Hansatech Instruments) and a Clark-type oxygen electrode (model DW 1, Hansatech, Norfolk, UK). Reactions were conducted in a 0.75-ml closed, thermostated, and magnetically stirred glass chamber containing 0.5 mg of mitochondrial protein in a respiration buffer consisting of 65 mM KCl, 125 mM sucrose, 10 mM Tris, 20 μM EGTA, and 2.5 mM KH2PO4, pH 7.4. After a 1-min equilibration period, mitochondrial respiration was initiated by addition of glutamate and malate (10 and 5 mM, respectively, final concentration) or succinate and rotenone (10 mM and 4 μM, respectively, final concentration). State 3 respiration was determined after addition of 444 μM ADP; state 4 respiration was measured as the rate of oxygen consumption in the absence of ADP. RCR (state 3-to-state 4 respiration ratio) and ADP-to-O ratio (ADP/O; i.e., nmol of ADP phosphorylated/nmol of O consumed) were calculated according to Estabrook (16), with 474 natom O/ml used as the value for oxygen solubility at 25°C in doubly distilled water. State 4 respiration was also measured in the presence of 10 mM glutamate and 5 mM malate, where 1 mM ADP, 1.5 μg of oligomycin, and 2 μM carbonyl cyanide m-chlorophenylhydrazone (CCCP) were added to induce state 3 respiration, inhibition of state 3 respiration through inhibition of the ATP synthase, and uncoupled respiration, respectively. In an independent trial, succinate-energized mitochondria were treated with 300 μM CaCl2 for analysis of mitochondrial calcium-loading capacity. The ratio of respiratory rate obtained before to that obtained 1 min after addition of CaCl2 was determined.
Plasma cTnI and skeletal muscle CS.
cTnI concentration was quantitatively determined with an established immunoassay using a commercial kit (Abbott Laboratories). Soleus CS activity was measured using the method proposed by Coore et al. (10). The principle of the assay was to initiate the reaction of acetyl-CoA with oxaloacetate and link the release of CoA-SH to 5,5-dithiobis(2-nitrobenzoate) at 412 nm.
Immediately before measurement of aconitase activity, mitochondrial fractions were resuspended in 0.5 ml of buffer containing 50 mM Tris·HCl (pH 7.4) and 0.6 mM MnCl2 and sonicated for 2 s. Aconitase activity was immediately measured spectrophotometrically by monitoring the formation of cis-aconitate from isocitrate at 240 nm in 50 mM Tris·HCl (pH 7.4) containing 0.6 mM MnCl2 and 20 mM isocitrate at 25°C according to Krebs and Holzach (29). One unit was defined as the amount of enzyme necessary to produce 1 μmol of cis-aconitate per minute (molar extinction coefficient at 240 nm = 3.6 mM−1·cm−1).
Tissue antioxidant enzyme activities.
After the cardiac homogenate was subjected to five freeze-defrost cycles and 10 min of centrifugation at 2,000 g (4°C), the activities of total SOD (tSOD) and mitochondrial SOD (Mn-SOD), glutathione peroxidase (GPx), and glutathione reductase (GR) were determined.
Activity of tSOD was determined spectrophotometrically according to Marklund and Marklund (36) by monitoring the rate of autoxidation of pyrogallol at 420 nm. The reaction buffer (25°C) consisted of 1 mM N,N-bis(2-(bis(carboxymethyl)amino)ethyl)glycine and 50 mM Tris·HCl, pH 8.2, in a final volume of 1 ml. Pyrogallol (0.3 mM) was added to the cuvette to start the reaction, and the rate of absorbance increase was measured. Mn-SOD activity was assayed by inhibition of cytosolic SOD (Cu-Zn-SOD) with 1 mM KCN under the same experimental conditions. One unit of SOD activity was defined as the amount of sample required to inhibit the autoxidation of pyrogallol by 50%, and the activity is expressed as units per milligram of protein.
GPx and GR activities were measured spectrophotometrically at 340 nm with a RANSEL kit and a GR kit, respectively (Randox Laboratories, Crumlin, UK), according to the manufacturer's instructions.
Sulfhydryl protein groups.
The mitochondrial content of oxidative modified sulfhydryl protein groups was quantified by spectrophotometric measurement according to the method proposed by Hu (23). Briefly, colorimetric assay was performed after reaction of 50 μl of mitochondrial extract with 10 μl of 5,5′-dithio-bis(2-nitrobenzoic acid) (10 mM) in a medium containing 150 μl of Tris (0.25 M) and 790 μl of methanol at 414 nm against a blank test. Sulfhydryl content was expressed in nanomoles per gram of mitochondrial protein (molar extinction coefficient at 414 nm = 13.6 mM−1·cm−1).
Lipid peroxidation was measured by determining the levels of lipid peroxides as the amount of thiobarbituric acid-reactive substances (TBARS) formed according to Rohn et al. (46) with some modifications. Mitochondrial protein (0.5 mg) was incubated at 25°C in 500 μl of a medium consisting of 175 mM KCl and 10 mM Tris, pH 7.4. Samples of 50 μl were mixed with 450 μl of a TBARS reagent (1% thiobarbituric acid, 0.6 N HCl, and 0.0056% butylated hydroxytoluene). The mixture was heated at 80–90°C for 15 min and recooled in ice for 10 min before centrifugation in an Eppendorf centrifuge (1,500 g, 5 min). Lipid peroxidation was estimated by the appearance of TBARS spectrophotometrically quantified at 535 nm. The amount of TBARS formed was calculated using a molar extinction coefficient of 1.56 × 105 M−1·cm−1 and expressed as nanomoles of MDA per milligram of protein (7).
Analysis of protein carbonylation, HSP-60, HSP-70, Bcl-2, and Bax.
Equivalent amounts of proteins were electrophoresed on a 15% SDS-polyacrylamide gel as described by Laemmli (30) and then blotted onto a nitrocellulose membrane (Hybond-ECL, Amersham Pharmacia Biotech) according to Locke et al. (32). After the samples were blotted, nonspecific binding was blocked with 5% nonfat dry milk in TTBS (Tris-buffered saline + Tween 20), and the membrane was incubated with anti-Bcl-2 (1:500 dilution, mouse monoclonal IgG; catalog no. sc-7382, Santa Cruz Biotechnology), anti-Bax (1:500 dilution, rabbit polyclonal IgG; catalog no. sc-493, Santa Cruz Biotechnology), anti-HSP-60 (1:2,000 dilution, mouse monoclonal IgG; catalog no. 386028, Calbiochem), or anti-HSP-70 (1:2,000 dilution, monoclonal clone; catalog no. BRM-22, Sigma) antibody for 2 h at room temperature, washed, and incubated with secondary horseradish peroxidase-conjugated anti-mouse or anti-rabbit IgG antibody (1:1,000 dilution; Amersham Pharmacia Biotech) for 2 h.
Protein carbonyl derivatives were assayed according to Robinson et al. (45) with some modifications. Briefly, a cardiac mitochondrial volume containing 20 μg of protein was derivatized with dinitrophenylhydrazine. The sample was then mixed with 1 vol of 12% SDS + 2 vol of 20 mM dinitrophenylhydrazine and 10% trifluoroacetic acid and incubated in the dark for 30 min; then 1.5 vol of 2 M Tris-18.3% β-mercaptoethanol were added. A negative control was simultaneously prepared for each sample. After dilution of the derivatized proteins in Tris-buffered saline to obtain a final concentration of 0.001 μg/μl, a 100-ml volume was slot-blotted onto a Hybond-polyvinylidene difluoride membrane. Immunodetection of carbonyls was then performed using rabbit antidinitrophenol (DAKO) as the primary antibody (1:2,000 dilution) and anti-rabbit IgG peroxidase (Amersham Pharmacia Biotech) as the secondary antibody (1:2,000 dilution).
For the referred methods, the bands were visualized by treating the immunoblots with enhanced chemiluminescence reagents (Amersham Pharmacia Biotech) according to the supplier's instructions and then exposed to X-ray films (Kodak Biomax Light Film, Sigma). The films were analyzed with QuantityOne Software (Bio-Rad). Optical density results were expressed as percent variation of control values.
Caspase-3 activation levels.
For analysis of caspase-3, 100 μg of protein in muscle homogenate were used. After addition of reaction buffer (10% sucrose, 0.1% CHAPS, and 25 mM HEPES, pH 7.4) and 40 μM acetyl-Asp-Glu-Val-Asp-p-nitroaniline (catalog no. 235400, Calbiochem) substrate for caspase-3, the samples were incubated at 37°C for 2 h, as previously described (14) with some adaptations. Absorbance at 405 nm was read in a plate reader (Labsystem iEMS Reader MF). Percent increase in caspase activity was determined by comparison of these results with the absorbance at 405 nm in a simultaneously incubated control sample.
Small sections were cut from left ventricular papillary muscles and transferred to 2.5% glutaraldehyde in 0.2 M sodium cacodylate buffer over a 2-h period. The specimens were postfixed with 2% osmium tetroxide, dehydrated in graded alcohol, and embedded in Epon. Ultrathin sections for transmission electron microscopy (model EM10A, Zeiss, Jena, Germany) were stained for contrast with 0.2% lead citrate and 0.5% uranyl acetate for a qualitative ultrastructural analysis. The heart tissue was also semiquantitatively examined for histopathological evidence of cardiomyopathy, according to severity scores from 0 to 3, as previously described (42, 48). Severity of damage was scored using electron microscopy grids: grade 0, no change from normal; grade 1, a limited number of isolated cells (<5% of the total number of cells per block) exhibiting early myofibrillar loss and/or cytoplasmic vacuolization; grade 2, groups of cells (5–30% of the total number) exhibiting early myofibrillar loss and/or cytoplasmic vacuolization; and grade 3, diffuse cell damage (>30% of total number), with the majority of cells exhibiting marked loss of contractile elements, loss of organelles, and mitochondrial and nuclear degeneration. All slides were scored independently by an examiner who was blind to each tissue sample code.
Values are means ± SE. One-way ANOVA followed by Bonferroni's post hoc test was used to compare groups. A t-test for independent measures was used to test the HSP-70 and HSP-60 differences between NT + P and T + P groups. Statistical Package for the Social Sciences (version 10.0) was used for all analysis. The significance level was set at 5%.
Body weights, absolute and relative heart weights, and soleus muscle CS activity are shown in Table 2. Consistent with the well-described body mass and cardiac adaptations induced by endurance training (39), 14 wk of endurance running training decreased body weight and increased skeletal muscle CS activity, heart weight, and heart weight-to-body weight ratio (P < 0.05). Improved enzymatic activity in the soleus muscle indicates that endurance training is an efficient chronic stimulus to ameliorate muscle oxidative metabolism.
To elucidate the effects of endurance training on heart mitochondrial respiration of DOX-treated rats, we studied respiratory parameters using NADH- and FADH2-linked substrate oxidation through mitochondrial complexes I and II, respectively. DOX treatment induced a significant decrease in state 3 respiration (59%) as well as RCR (44%) of glutamate + malate-energized heart mitochondria isolated from nontrained animals (Table 3). However, no changes in state 4 respiration and ADP/O were detected in the same groups (NT + P vs. NT + DOX).
DOX-induced inhibition of respiration was completely prevented by endurance running training (NT + DOX vs. T + DOX), which means that DOX did not affect mitochondrial respiration in trained animals. Moreover, endurance training alone significantly increased RCR in nontreated animals (42%), mostly because of a small, but nonstatistically significant, decrease in state 4 respiration in trained animals.
When the FADH2-linked substrate succinate was used, state 3 respiration (47%) and RCR (31%) were significantly depressed by DOX administration, whereas no changes were detected in state 4 respiration or phosphorylation efficiency, expressed as ADP/O (NT + P vs. NT + DOX). As with complex I-related substrates, these respiratory parameters were similar to control in succinate-energized heart mitochondria isolated from trained rats treated with DOX. Mitochondria from trained animals did not show any significant difference in state 4 respiration, RCR, or ADP/O compared with mitochondria from nontrained groups (Table 4).
Neither DOX nor training altered respiratory rates in the presence of oligomycin, whereas uncoupled respiration (induced by CCCP) was significantly depressed (22%) by DOX treatment (NT + P vs. NT + DOX), reflecting an impairment in the rate of electron transfer (Fig. 1). However, training significantly attenuated the decrease in uncoupled respiration caused by DOX treatment.
We also analyzed the changes in mitochondrial oxygen consumption after addition of calcium. Figure 2 shows that calcium-stimulated respiration largely failed to return to normal state 2 values in mitochondria isolated from DOX-treated rats, and it demonstrates an accelerated and mostly uncoupled oxygen consumption (NT + P vs. NT + DOX), which was reversed by cyclosporin A (1 μM). In contrast, after calcium accumulation, heart mitochondria from trained animals were better able to recover from the increase in respiration rates, with and without DOX, which means that endurance training improved mitochondrial calcium control. These data can be confirmed by the increased (32% for NT + P vs. NT + DOX and 27% for NT + DOX vs. T + DOX) ratio of the respiratory rate of succinate-energized mitochondria after calcium to that before calcium. This ratio was significantly elevated in nontrained DOX-treated rats compared with all other groups (Fig. 2).
DOX induced a significant increase in plasma levels of cTnI (Table 5), which reflects enhanced cTnI release from the heart and indicates loss of cardiomyocyte membrane integrity. However, endurance training resulted in a significant reversal (P < 0.05) of DOX-induced cardiotoxicity (NT + DOX vs. T + DOX).
We measured the activity levels of aconitase, a citric acid cycle enzyme containing Fe-S clusters, which are susceptible to ROS-induced damage and, consequently, enzyme inactivation, as an indirect assessment of in vivo mitochondrial O2−· production and associated damage (37). Mitochondrial aconitase activity was significantly lower DOX-treated animals than in all other groups (Table 5).
The content of sulfhydryl groups was also significantly lower in heart mitochondria isolated from DOX-treated rats than from all other treatment groups (Table 5).
A significant increase in the levels of the stress-related markers, protein carbonyl groups and MDA, from heart mitochondrial extracts was found in the NT + DOX group (Fig. 3 and Table 5, respectively). However, carbonyl and MDA levels were lower in mitochondria isolated from T + DOX than NT + DOX animals.
Endurance training increased significantly the activities of tSOD, Mn-SOD and Cu-Zn-SOD in T + P and T + DOX groups compared with their nontrained counterparts, although no significant changes were observed in GPx and GR activities (Table 6). DOX treatment per se was not sufficient to significantly alter any of the measured enzyme activities.
Endurance running training raised significantly the expression levels of heart mitochondrial HSP-60 and the whole cardiac muscle homogenate HSP-70 (Fig. 4). Increases of 95% (HSP-60) and 112% (HSP-70) were found in trained hearts compared with controls.
We measured mitochondrial levels of Bax and Bcl-2 to determine whether there were any DOX-induced changes in these pro- and antiapoptotic makers and to analyze the effect of training on these changes. DOX administration resulted in significant elevation of Bax and Bax-to-Bcl-2 ratio (Fig. 5). However, training restored the increase in Bax and Bax-to-Bcl-2 ratio. No changes were observed in Bcl-2 protein expression between groups.
Caspase-3 activity revealed a significant increase in cardiac muscle from nontrained DOX-treated rats (NT + P vs. NT + DOX; Fig. 6). However, training significantly attenuated the increase in caspase-3 activity induced by DOX (NT + DOX vs. T + DOX).
Qualitatively examination of cardiomyocytes from DOX-treated animals by electron microscopy revealed marked myocardial damage, such as cytoplasmic vacuolization, myofibrillar disorganization, mitochondrial damage with extensive degeneration or even loss of cristae, intramitochondrial vacuoles and myelin figures, and mitochondrial swelling, in contrast to the normal appearance of the NT + P group (Fig. 7, A and B). Endurance training per se (NT + P vs. T + P) induced signs of mitochondrial biogenesis with an elevation in the number of encroached mitochondria. Mitochondrial division, mild and focal loss of crista density, and loss of organization within mitochondria, as well as scarce secondary lysosomes, were also observed in NT + P hearts. In hearts harvested from T + DOX animals, although ultrastructural alterations described for the T + P group were maintained, no major signs of DOX-induced cardiotoxicity were observed (Fig. 7, C and D). Semiquantitative analysis of these histopathological changes confirmed that the severe ultrastructural abnormalities induced by DOX treatment in sedentary hearts were significantly attenuated in the T + DOX group (Table 5).
Overview of the principal findings.
The present investigation provides new insights into the biochemical mechanisms by which endurance exercise training protects cardiac muscle tissue against the toxicity induced by DOX. Our data confirm that DOX induced mitochondrial O2−· production and oxidative damage, impaired mitochondrial respiratory function, and, ultimately, triggered the mitochondrial pathway to signal apoptotic cell death. Moreover, in the presence of calcium, heart mitochondria from DOX-treated animals have an increased sensitivity to permeability transition pore (PTP)-mediated uncoupled respiration, which was limited in the T + DOX group. However, 14 wk of endurance training limited the impact of DOX on heart mitochondrial O2−· production, oxidative damage, respiratory function, and apoptosis. This protection was corroborated by the histological data as well as by the levels of the specific systemic markers of cardiac damage found in the different groups. Although cardiovascular functional parameters were not determined in the present study, the consistency between the variations in cTnI and the functional response reported in previous studies (22, 24) might indicate that the adaptations induced by endurance training would possibly translate into hemodynamic benefits in trained animals.
Evidence of training-induced protection in heart mitochondrial respiration.
Previous studies have shown that endurance training attenuates DOX-induced cardiac damage, diminishing the increased biochemical and morphological signs of toxicity induced by in vivo DOX treatment (2, 26). However, no data are available concerning the cross-tolerance effect of moderate endurance training on DOX-induced mitochondrial malfunction. Our present results demonstrate that 14 wk of moderate endurance running training prevented the inhibition of mitochondrial respiration caused by DOX. The depressed activity of mitochondrial enzyme components of complexes I and II caused by DOX (48, 57) could partially justify the diminished electron transport through ETC in the NT + DOX group. Thus the unaltered state 3 respiration observed in the T + DOX group suggests that, among other possible effects, training probably prevented the inactivation of complex I and II components in DOX-treated heart mitochondria.
The above-mentioned data concerning NADH-linked substrates were confirmed by mitochondrial oxygen consumption using oligomycin and CCCP (Fig. 1). Because state 3 and CCCP-induced respiration decreased after DOX insult and because endurance training reversed these effects of DOX, one can argue that exercise training prevented the damage of the respiratory chain enzymes and/or the upstream supply of electrons to the respiratory components that were limited after DOX treatment.
The Fe-S centers of aconitase can be reversely inactivated by O2−· and related species, allowing the measurement of aconitase inactivity as an index of in vivo O2−· production in mitochondria (37). In the present study, enhanced mitochondrial O2−· generation in DOX-treated groups was indirectly demonstrated by significantly lower aconitase activity than in other groups (Table 5). This might lead to a decrease in the supply of reducing equivalents to the ETC through the oxidative damage inflicted by other citric acid cycle enzymes, namely, α-ketoglutarate dehydrogenase and succinate dehydrogenase (41). Furthermore, because mitochondrial complexes include enzymes that consist of polypeptides that also comprise Fe-S clusters, namely, those from complexes I and II, they can also become prone to oxidative deactivation by O2−· or other ROS, resulting in accumulating products of protein oxidation, such as carbonyl groups or oxidized thiol residues (Fig. 3). These alterations probably contributed to the decline of electron transfer through ETC, compromising state 3 respiration and RCR (34, 35). Considering that changes in markers of oxidative damage, such as MDA, carbonyl and sulfhydryl groups, and aconitase activity, were consistent with alterations in mitochondrial respiratory function of all groups, one could suggest that training counteracts the changes in redox homeostasis induced by DOX. It is possible that training-induced upregulation of mitochondrial and cytosolic defenses might contribute to the mitochondrial tolerance against DOX effects. Indeed, data collected in studies using other models of cardiac and mitochondrial dysfunction, such as I/R, revealed that alterations in mitochondrial respiration are prevented in myocardium expressing high levels of HSPs (47, 54). In the present study, the finding that HSP-60 and HSP-70 were overexpressed in trained animals, together with the observed improvement of mitochondrial respiratory function, strongly suggests that the upregulation of these molecular chaperones possibly contributed to the preservation of integrity and activity of the mitochondrial complexes. This may be accomplished through facilitation of nuclear-encoded protein importation and assembly in the mitochondrial matrix and through the improved assisted folding of proteins within mitochondria. Moreover, Suzuki et al. (54) described overexpression of Mn-SOD content and activity in HSP-70-transfected hearts, which was consistent with improved mitochondrial respiratory function. Indeed, another possible explanation for the increased respiratory function of mitochondria isolated from trained rats compared with those isolated from nontrained rats on exposure to DOX could be associated with the training-mediated effect in the upregulation of the cardiac antioxidant enzyme system. The enhanced activity of Cu-Zn-SOD and Mn-SOD in trained hearts found in the present study (Table 6), which was previously reported by others (40, 43, 52, 53), may be an important factor in counteraction of DOX-induced free radical-mediated mitochondriopathy, although we did not find any significant effect of DOX treatment per se on the activity of either SOD isoform. Similar findings were obtained in DOX-treated rats overexpressing Mn-SOD, in which impaired mitochondrial NADH- and FADH2-sustained state 3 respiration and RCR (57), as well as histological disarrangements (58), were prevented. Taken together, these findings suggest that protection of mitochondrial function exerted by training, through its ability to induce HSP overexpression and increased SOD activity, particularly Mn-SOD, could be an important factor in counteracting DOX cardiotoxicity.
Sensitivity of mitochondria to calcium-induced uncoupled respiration.
In addition to interfering with mitochondrial respiration, there is growing evidence that DOX disrupts the ability of cardiac mitochondria to accumulate calcium, diminishing calcium-loading capacity (42, 48, 59). Mitochondria, which act as major sources of ROS and cytosolic calcium buffers, can become severely dysfunctional as a result of induction of the PTP under the synergistic effects of oxidative stress and increased matrix calcium (11). Under these conditions, induction of the PTP uncouples oxidative phosphorylation and collapses transmembrane potential, leading to mitochondrial bioenergetic dysfunction (11, 50). In the present study, the complete block of this altered mitochondrial regulation of respiration by cyclosporin A, the specific inhibitor of the PTP (6), supports the notion that calcium-induced PTP can be responsible for the stimulation of mitochondrial respiration. After a pulse of 300 μM calcium, the respiratory rate of succinate-energized mitochondria from DOX-treated rats remained uncoupled. These results differed from the control assay (NT + P vs. NT + DOX), in which respiration was recovered after complete calcium accumulation by heart mitochondria. Moreover, this enhanced recovered respiratory rate after calcium addition was also accomplished by mitochondria from T + P and T + DOX groups. The differences in oxygen consumption of heart mitochondria after calcium accumulation allow us to suggest that the subpopulation of mitochondria with calcium-induced PTP was lower in T + DOX than in NT + DOX hearts. From the results, we suggest that endurance training also protects heart mitochondria of DOX-treated rats from the enhanced susceptibility to calcium-induced PTP. Because PTP has a marked oxidative etiology, one possible explanation for the enhanced tolerance of trained mitochondria to calcium could be related to the upregulation of cardiomyocyte and mitochondrial defenses caused by endurance training. Therefore, because of the elevated content of sulfhydryl groups in T + DOX compared with NT + DOX mitochondria (Table 5), we wonder about the enhanced levels of sulfhydryl donors, such as GSH (27), which also contributes to the augmented capacity of these mitochondria to accumulate calcium through inhibition of the PTP, in mitochondria from trained animals. Clearly, further studies are need to clarify the influence of training on this mechanism of mitochondrial calcium regulation.
Impact of endurance training on DOX-induced mitochondrial and cell apoptosis.
Programmed cell death is a widely conserved general phenomenon that occurs in many processes involving the reconstruction of multicellular organisms, as well as in the elimination of old or damaged cells (50, 51). Given that mitochondrial (dys)function during prooxidant redox changes is increasingly considered a key event in a variety of forms of cell death, including apoptosis, we analyzed the effects of in vivo DOX insult and training on expression of the pro- and antiapoptotic Bcl-2 family proteins (Bax and Bcl-2). The relative expression of these proteins in the mitochondrial outer membrane is thought to decide the fate of the cell by regulating membrane integrity (20). DOX has recently been shown to cause apoptosis in rat heart (1, 21), and this phenomenon was attributed, at least in part, to mitochondria-mediated pathways (9, 56). In our study, mitochondria from nontrained DOX-treated rats seem to be more susceptible to apoptotic cell death than control (NT + P) rats, as shown by a significant increase in Bax content and in the Bax-to-Bcl-2 ratio (Fig. 5). Our study also revealed lower Bax content and Bax-to-Bcl-2 ratio in heart mitochondria isolated from trained hearts, suggesting, as in mitochondrial respiration, a protective effect in rats treated with DOX. However, the lack of increase of Bcl-2 in the trained groups (NT + P vs. T + P) was unexpected because of earlier reports of its overexpression in trained hearts (31, 49).
Our data demonstrated that the activation of the caspase-3 apoptotic pathway by DOX was dramatically inhibited in trained hearts (Fig. 6) and followed the same trend as Bax content and Bax-to-Bcl-2 ratio, suggesting a mitochondria-mediated origin for caspase-3 activation. Training-induced increases in HSP expression and SOD activity could be among the protective mechanisms by which training limited apoptotic DOX side effects in heart mitochondria. The protective role of these important proteins was highlighted by the prevention of apoptotic cell death induced by I/R in HSP-70-transfected hearts (54). Although other possible mechanisms to explain this protection induced by endurance training cannot be excluded, our data suggest that it could be mediated by an improvement of mitochondrial and cell defense systems.
In conclusion, our results highlight the advantage of endurance training against DOX cardiotoxicity and provide evidence for a mitochondria-mediated mechanism, which supports our initial hypothesis.
We thank Celeste Resende and Serafim Pereira for technical assistance with animal care and training protocol. We are grateful to Pfizer Laboratories, Portugal, for providing doxorubicin.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2005 by the American Physiological Society