OUP user menu

Abnormal cardiac and skeletal muscle mitochondrial function in pacing-induced cardiac failure

José Marín-García, Michael J. Goldenthal, Gordon W. Moe
DOI: http://dx.doi.org/10.1016/S0008-6363(01)00368-6 103-110 First published online: 1 October 2001


Background: Previous studies have shown that marked changes in myocardial mitochondrial structure and function occur in human cardiac failure. To further understand the cellular events and to clarify their role in the pathology of cardiac failure, we have examined mitochondrial enzymatic function and peptide content, and mitochondrial DNA (mtDNA) integrity in a canine model of pacing-induced cardiac failure. Methods: Myocardium and skeletal muscle tissues were evaluated for levels of respiratory complex I–V and citrate synthase activities, large-scale mtDNA deletions as well as peptide content of specific mitochondrial enzyme subunits. Levels of circulating and cardiac tumor necrosis factor-alpha (TNF-α), and of total aldehyde content in left ventricle were also assessed. Results: Specific activity levels of complex III and V were significantly lower in both myocardial and skeletal muscle tissues of paced animals compared to controls. In contrast, activity levels of complex I, II, IV and citrate synthase were unchanged, as was the peptide content of specific mitochondrial enzyme subunits. Large-scale mtDNA deletions were found to be more likely present in myocardial tissue of paced as compared to control animals, albeit at a relatively low proportion of mtDNA molecules (<0.01% of wild-type). In addition, the reduction in complex III and V activities was correlated with elevated plasma and cardiac TNF-α levels. Significant increases in left ventricle aldehyde levels were also found. Conclusions: Our data show reductions in specific mitochondrial respiratory enzyme activities in pacing-induced heart failure which is not likely due to overall decreases in mitochondrial number, or necrosis. Our findings suggest a role for mitochondrial dysfunction in the pathogenesis of cardiac failure and may indicate a commonality in the signaling for pacing-induced mitochondrial dysfunction in myocardial and skeletal muscle. Increased levels of TNF-α and oxidative stress appear to play a contributory role.

  • Energy metabolism
  • Heart failure
  • Mitochondria

Time for primary review 31 days.

1 Introduction

Heart failure is a chronic disorder characterized by a relentless progressive clinical course often resulting in repeated hospital admissions which imposes a burden on the health care delivery system [1]. Accordingly, further understanding of the mechanisms mediating the progression of heart failure is of major importance. Although hemodynamic stress frequently encountered in heart failure may play a role in the disease progression [2], it is currently believed that many other mechanisms such as the activation of neurohormones and the pro-inflammatory cytokines may be more important in mediating the progression of heart failure [3,4]. Recent studies with isolated cells including cardiomyocytes have implicated the effects of pro-inflammatory cytokines tumor necrosis factor-alpha (TNF-α) in the generation of reactive oxygen species (ROS) in mitochondria, altered mitochondrial membrane permeability and in mitochondrial enzymatic dysfunction as both early events and critical to the physiological mechanism of TNF-α action [5–7].

A potentially important and yet poorly studied mechanism for the progression of chronic heart failure (CHF) is perturbed myocardial energetics which have been reported in models of heart failure or cardiac hypertrophy [8]. Indeed, discrete mitochondrial oxidative phosphorylation (OXPHOS) defects have been documented in human cardiomyopathies. Both dilated and hypertrophic cardiomyopathies are frequently accompanied by changes in OXPHOS/respiratory enzyme activities in cardiac tissues [9–11]. Also, large-scale deletions in heart mitochondrial DNA (mtDNA) have been reported in patients with cardiomyopathy [12–14]. Furthermore, changes in bioenergetics and in both mitochondrial structure and respiratory enzyme function have been described in skeletal muscle in CHF [15–17]. However, most of these studies have involved patients with heart failure with diverse etiologies, which would invariably confound the interpretation of the energetic data. Accordingly, to study changes in mitochondrial function in heart failure per se without the confounding effects of such as myocardial ischemia, would require a relatively pure animal model of heart failure. Canine pacing-induced cardiomyopathy is a model of heart failure in which rapid ventricular pacing leads to cardiac myopathy consisting of an increase in chamber dimension, mural thinning, elevation in ventricular wall stress, and congestive heart failure, mimicking dilated cardiomyopathy in humans [18–21].

In our previous studies, myocardial tissues from paced dogs were found to harbor markedly reduced activities of key ATP-utilizing and ATP generating enzymes including mitochondrial ATP synthase [19]. However, the pathophysiologic importance of mitochondrial enzyme activity reductions in mediating the progression of heart failure remains undetermined. Bioenergetic and mitochondrial enzyme dysfunction could play an integral part in the primary mechanism of cardiac dysfunction or may represent a common downstream event in the pathways leading to the heart failure phenotype. Data gauging the extent to which mitochondrial abnormalities are either a primary defect or are secondary to other myocardial changes (e.g. cardiac hypertrophy and remodeling) that contribute to the pathophysiology of cardiac cell function are needed. To assess the extent and tissue-specificity of mitochondrial dysfunction in heart failure, we therefore conducted a comprehensive analysis of mitochondrial respiratory enzymatic activities and of mtDNA integrity in both the myocardial and skeletal muscle of dogs with severe pacing-induced heart failure as compared to tissues from normal animals. To evaluate the potential effects of cytokine activation and oxidative stress, levels of both circulating and cardiac-localized TNF-α and of overall aldehyde content were evaluated.

2 Methods

The study population consisted of two groups of dogs. A total of ten normal dogs served as controls. A second group of 11 dogs underwent continuous rapid right ventricular pacing for 3 weeks and was designated as the group with heart failure. Approval was obtained from the institutional animal research committee before study commencement, in accordance with the guidelines on the Care and Use of Experimental Animals issued by the Canadian Council on Animal Care, Ottawa, Canada.

2.1 In vivo studies

The method of induction of heart failure and the clinical and in vivo hemodynamic assessment have been described in detail in previous reports [18–20]. In brief, under general anesthesia, a bipolar lead was placed in the right ventricle and a programmable pulse generator (SX-5985, Medtronic, Missisauga, Ont., Canada) was placed in a cervical pocket. A chronic indwelling cannula was placed in the carotid artery for serial blood sampling. Dogs were recovered from anesthesia for 1 week before the first in vivo study. After baseline radiographic and hemodynamic assessment to ensure healthy baseline parameters, the pacemaker was programmed to deliver impulses at 250 beats/min. The animals were examined weekly for clinical and radiographic signs of heart failure. Arterial blood sampling was performed weekly to monitor changes in plasma cytokine levels. All in vivo measurements were obtained with the conditioned animal in a conscious state. In paced dogs, data were acquired after sinus rhythm was resumed temporarily for 15 min to eliminate the artifacts produced by pacing. Hemodynamic measurements were obtained using a micromanometer-tipped catheter introduced via the femoral artery and a thermodilution catheter positioned in the pulmonary artery introduced via the femoral vein. Measurements were obtained at baseline and after 3 weeks of pacing. For the terminal hemodynamic study, the data were acquired after the pacer was programmed off for 20 min. After the study, the dogs were euthanized and tissue from the lateral wall of the left ventricle and skeletal muscle from the vastus lateralis, ∼500 mg of each, were obtained and stored at −80°C until analysis.

2.2 Enzymology

Tissues were analyzed for mitochondrial enzyme activities using tissue homogenates prepared in MTE buffer (0.25 M mannitol, 20 mM Tris–S04 and 1 mM EDTA, pH 7.4) with a glass homogenizer. Complex I activity measured by the oxidation of NADH by ubiquinone-1 at 340 nm, complex II activity measured by the oxidation of succinate by ubiquinone-2 at 600 nm, complex III assayed by reduction of cytochrome c by ubiquinol-2 at 550 nm, complex IV assayed by oxidation of dithionite-reduced cytochrome c at 550 nm, complex V assayed by NADH oxidation using the coupled enzyme assay with pyruvate kinase and lactate dehydrogenase at 340 nm, and citrate synthase assayed at 412 nm were all performed spectrophotometrically at room temperature [9,10]. In all cases duplicate assays were performed. Protein determination was performed according to Bradford [22].

2.3 Immunoblot analysis of specific mitochondrial proteins

Samples of crude homogenate protein (20–100 μg) were denatured in buffer containing 2% sodium dodecyl sulfate (SDS), 62.5 mM Tris–HCl, pH 6.8, and 10% glycerol, and incubated before electrophoresis for 5 min at 100°C with β-mercaptoethanol. SDS polyacrylamide gel electrophoresis was carried out at 100 V using 4–20% acrylamide (Tris–glycine) gradient gels. The proteins were then transferred electrophoretically to Immobilon P membranes (Millipore) in Tris–glycine buffer containing 10% methanol for 1 h at 100 V. After blocking the membrane with phosphate buffered saline with 0.05% Tween-20 and 3% (w/v) instant Carnation dry milk for 60 min, the blots were incubated for 1 h at room temperature with antibody against either cytochrome c (PharmGen), heat shock protein (HSP-60) (StressGen) or subunits α and γ of human ATP synthase kindly provided by M.F. Marusich (University of Oregon, Eugene). A detection system employing alkaline phosphatase conjugated to an appropriate second antibody, and colorimetric substrate (NBT-BCIP) was used (Roche). Quantitation of the signals was performed on scanned blots using Image 1.33 software (NIH).

2.4 PCR amplification

DNA was extracted from cardiac homogenates using QIAamp Tissue kit (Qiagen) and amplified by PCR using primers of specific canine mtDNA sequences [23] designed using Right Primer software (BioDisk). PCR amplification was performed in 100 μl total volume with 200 μM of each dNTP, 0.4 μM of each primer, 1 U of Taq polymerase (Promega), 2.0 mM MgCl2, 10 mM Tris–HCl pH 8, with 50 mM KCl and 100 ng of DNA template. After heating at 94°C for 3 min to ensure DNA denaturation, each cycle of PCR consisted of denaturation at 94°C for 60 s, annealing at 54°C for 60 s and extension at 72°C for 60 s. After 35 cycles of the PCR reaction and an additional extension step at 72°C for 8 min, 15 μl of the reaction mixture was electrophoresed in a 2.5% agarose gel and stained with ethidium bromide.

2.5 Tumor necrosis factor-α

Quantitative analysis of serum and tissue tumor necrosis factor-α (TNF-α) was performed with ‘sandwich’ enzyme linked immunoadsorbant assay (ELISA) using a commercially available Mouse TNF-α DuoSet kit (Genezyme Diagnostics, Cambridge, MA). The antibodies cross-react with human, dog, and rat TNF-α. Results represented the mean value of three separate measurements performed in duplicate. The buffer used for solubilizing the membrane bound protein was used as negative control to confirm that homogenizing buffer did not interfere with the ELISA result.

2.6 Oxidative stress

Total aldehydes (expressed as pmol/100 mg tissue) were measured in left ventricular tissues using gas chromatography–mass spectrometry [24].

2.7 Statistics

Data concerning enzyme activity, hemodynamic measurements, plasma and cardiac TNF-α levels, aldehyde and peptide content, were subjected to an unpaired t-test. Differences were considered significant when the calculated probability value was less than 0.05.

3 Results

Development of heart failure in paced dogs was confirmed by clinical, radiographic and hemodynamic assessments. In the paced group, all dogs developed at least two of the three signs of apathy, anorexia and ascites, and all displayed severe pulmonary congestion on the chest radiographs. The hemodynamic data of control normal dogs and paced dogs are summarized in Table 1. In paced dogs, heart rate, mean pulmonary arterial, pulmonary capillary wedge, left ventricular end diastolic and right atrial pressures were all markedly elevated compared to normal controls. On the other hand, cardiac index and left ventricular dP/dt were significantly depressed. Mean arterial pressure was unaffected in paced dogs. Data for plasma TNF-α level are shown in Fig. 1. Compared to control dogs, paced dogs had markedly elevated serum TNF-α levels suggesting an activation of the pro-inflammatory cytokines. Levels of cardiac TNF-α were also significantly elevated in the paced animals (171.7±12.3 pg/mg tissue) compared to the controls (68.4±12.3 pg/mg tissue).

Fig. 1

TNF-α serum levels in paced-induced CHF and control animals. The data are expressed as mean value±S.E.M. error bars.

View this table:
Table 1

Hemodynamic data

Control dogsPaced dogs
LV end diastolic pressure (mmHg)10.9±0.6737.6±2.15*
Mean pulmonary artery pressure (mmHg)16.3±1.0239.5±2.4*
Right atrial pressure (mmHg)7.4±0.5312.7±0.88*
Heart rate (beats/min)75.0±9.07137.4±6.6*
Cardiac index (ml/kg)151.7±11.5594.22±8.22*
Mean arterial pressure (mmHg)107.6±3.74107±2.18
Pulmonary capillary wedge pressure (mmHg)9.0±0.629.1±2.01*
Left ventricular dP/dT (mmHg/s)2069.5±121.801158.5±36.77*
  • Values shown are mean±S.E.M.

  • * P<0.05 versus control.

Left ventricular samples from each of 21 dog hearts (11 paced, ten non-paced controls) and skeletal muscle biopsies were analyzed for respiratory enzymatic activity levels (complex I–V) and for levels of citrate synthase. As shown in Table 2, specific activity levels of myocardial complex III and V were significantly lower in paced animals (greater than 70% reduced compared to mean values from non-paced animals, P<0.05). Furthermore, complex III and V activities correlated inversely with serum TNF-α levels (r=−0.6631 and −0.5498, respectively, both P<0.05)., and with cardiac TNF-α (r=−0.7500 and −0.7838, respectively, both P<0.05). Levels of other mitochondrial respiratory complex activities including complex I, II and IV as well as citrate synthase activity, an enzyme often used as a gauge of overall mitochondrial content, were not significantly diminished in the paced compared to untreated animals. If activities were normalized with respect to citrate synthase activity, significant reduction was noted in the activity ratios for complex III/citrate synthase and complex V/citrate synthase (greater than 65% reduced compared to control activity ratios), while activity ratios for complex I, II and IV were not affected. Similar findings were obtained in skeletal muscle activity levels and normalized activity ratios for complex III and V (Table 3).

View this table:
Table 2

Mitochondrial enzyme activities and activity ratios in myocardial tissues

Control dogsPaced dogs
Complex Ia25.4±3.219.8±3.3
Complex I/CS0.047±0.0030.047±0.006
Complex IIa24.7±3.4922.7±3.18
Complex II/CS0.04±0.010.03±0.01
Complex IIIa11.2±2.062.09±0.73*
Complex III/CS0.022±0.0040.004±0.002*
Complex IVa207.20±15.19187.45±11.98
Complex IV/CS0.39±0.030.33±0.02
Complex Va192.60±24.5556.82±8.05*
Complex V/CS0.34±0.030.10±0.01*
Citrate synthasea555.90±49.65583.55±40.95
  • Values shown are mean±S.E.M. with n=10 for each group. CS, citrate synthase.

  • a Results shown correspond to the calculated mean values of specific activity±S.E.M. All units are nmol substrate used/min per mg protein.

  • * P<0.05 versus control.

View this table:
Table 3

Mitochondrial enzyme activities and activity ratios in skeletal muscle

Control dogsPaced dogs
Complex Ia16.6±1.116.1±1.8
Complex I/CS0.047±0.0030.047±0.006
Complex IIa16.4±3.3716.7±2.96
Complex II/CS0.06±0.010.07±0.02
Complex IIIa10.50±1.510.90±0.46*
Complex III/CS0.04±0.010.004±0.002*
Complex IVa142.69±13.19138.10±14.43
Complex IV/CS0.50±0.050.51±0.06
Complex Va151.20±14.4183.80±5.39*
Complex V/CS0.51±0.050.31±0.03*
Citrate synthasea314.90±38.04286.51±27.03
  • Values shown are mean±S.E.M. with n=10 for each group. CS, citrate synthase.

  • a Results shown correspond to the calculated mean values of specific activity±S.E.M. All units are nmol substrate used/min per mg protein.

  • * P<0.05 versus control.

In humans, mtDNA damage can be evaluated by gauging the presence and abundance of specific mtDNA deletions which frequently increase in overall abundance in cardiac and skeletal muscle tissues in several cardiac and neurological diseases as well as during the course of aging [25,26]. Using primers designed from the known canine mtDNA sequence [23] roughly analogous to sequences used to evaluate the common 7.4-kb deletion in humans [13], specific large-scale mtDNA deletions were found to be present in ∼50% of paced dog cardiac tissues examined. These deletions were not observed in the control myocardial tissues as depicted in Fig. 2. The mtDNA deletions in the paced dog myocardium (estimated to be ∼7.55 kb in size) are somewhat larger (100–200 bp) than those reported in human mtDNA but delete the same relative regions of the mitochondrial genome as found in the common 7.4-kb deletion in humans. They are present at extremely low abundance (less than 0.01%) compared to wild-type genomes (results not shown). These deletions were not present in skeletal muscle from either paced or non-paced dogs.

Fig. 2

Large-scale deletions in mtDNA in CHF. Cardiac mtDNA was amplified using a set of dog-specific mtDNA primers analogous to human sequences previously used to assess the presence of the common 7.4-kb mtDNA deletion [10]. While the expected amplification product with this primer set is greater than 7.8 kb (and would not likely be amplified nor visualized under these conditions), smaller amplified fragments (sized ∼475 and 280 bp) indicative of large-scale mtDNA deletions are shown on this representative agarose gel (2.5%) with the paced (P) and not with the control (C) DNA templates. Lanes (M) containing size markers of 271, 449, 683 and 885 bp are also shown.

Levels of peptide content for specific mitochondrial proteins including cytochrome c, two subunits of respiratory complex V (ATPase γ and α subunits) and HSP-60 were not significantly different in cardiac tissues of paced as compared to control animals (Fig. 3).

Fig. 3

Peptide content of specific mitochondrial enzyme subunits in paced and control tissues. A representative Western blot probed with antibodies directed against cytochrome c and ATP synthase subunit α (as described in Methods) is shown. The M (marker) lane contains pre-stained molecular weight markers for SDS–PAGE (InVitrogen).

As an indication of increased oxidative stress in CHF, levels of cardiac tissue total aldehydes were markedly increased in paced (22284±1690 pmol/100 mg tissue) compared to control values (8398±477 pmol/100 mg tissue). Increases were also observed in several species of unsaturated aldehydes including 4-OH-alkenals and malondialdehyde.

4 Discussion

Abnormalities in the number, structure and function of myocardial mitochondria have been described in a considerable number of patients with cardiomyopathy, although such ultrastructural changes are often less specific and diagnostic than demonstrated enzymatic activity changes [17]. Myocardium with an increased number of smaller mitochondria has been reported in animals with severe heart failure including dogs with CHF produced by sequential intracoronary microembolizations [27]. Evidence has also been presented describing myocyte loss, and altered myocyte geometry indicating extensive myocyte re-modelling and triggering of apoptotic pathway(s) in the canine model of pacing-induced heart failure used in our studies [28,29]. However, although there have been reports of mitochondrial dysfunction in cases of human cardiomyopathy, there is at present little information available concerning the comprehensive assessment of mitochondrial respiratory enzymatic function in both myocardial and skeletal muscle tissues in an animal model of heart failure. Our principal findings in this study indicate that marked reductions in respiratory complex III and V activities are present in both myocardial and skeletal muscle tissues after pacing-induced cardiac failure. These changes do not appear to be accompanied by changes in peptide content of specific mitochondrial proteins.

Our previous studies using this model have reported decreases in myocardial tissues of paced animals of the major mitochondrial ATP-generating enzyme (complex V or mitochondrial ATP synthase) as well as downregulation of several ATP utilizing enzyme activities [19,20] associated with a possibly compensatory increase in fatty acid oxidation and Krebs cycle activities. The reduction in the levels of myocardial mitochondrial ATP synthase activity has been further identified as both an early and a persistent event during the development of heart failure [20]. Those studies however were limited since other mitochondrial enzymes involved in oxidative phosphorylation and evidence of general mitochondrial damage were not assessed. In this study, the data confirm the marked reduction of myocardial complex V as well as identifying a reduction in complex III activities in the paced animals. Our results also show no significant change, or evidence of compensatory increase, in the activity levels of citrate synthase, the mitochondrial-matrix located Krebs cycle enzyme. Importantly, the reduced levels of complex III and V activities do not appear to be due to generalized mitochondrial damage, necrosis or overall decreased levels of mitochondria as gauged by unchanged levels of respiratory complex I, II, IV and citrate synthase activities relative to controls, levels of mitochondrial located peptides including cytochrome c, HSP-60 subunits α and γ of complex V and are further corroborated by data showing unchanged levels of mtDNA in paced dog myocardium compared to control (data not shown).

The reason for the pronounced selective reduction in these specific enzyme activities in heart failure at present remains undetermined. Peptide levels for two of the nuclear-encoded subunits of ATP synthase (complex V) were unchanged. However, reduced levels of other complex V subunits could potentially account for the reduction in complex V activity level; other subunit peptide levels have not been examined and should be assessed when specific antibody probes to canine peptides become available. Moreover, future studies aimed at determining both the specific levels of mitochondrial peptides in complex V (e.g. ATPase 6 and ATPase 8) and complex III (e.g. cyt b) as well as the import and assembly of the nuclear-encoded subunits into functioning mitochondrial respiratory complexes, could prove informative in understanding the pathophysiology of cardiac failure. In this regard, recent studies of complex IV (cytochrome c oxidase) deficiency in Leigh disease have demonstrated defective assembly of complex IV subunits in several tissues [30]. Our results also suggest that a common element in either the synthesis, import, and/or assembly, or regulation of both complex V and III (but not complex I or IV) subunits may be defective, in both cardiac and skeletal muscle in canine pacing-induced heart failure. Defective regulation of mitochondrial transcription appears to be the least likely mechanism in explaining the reduction of these specific enzyme activity levels. Differential mitochondrial transcriptional regulation is also unlikely since a common promoter with co-ordinated transcription of mtDNA-encoded complex I, III, IV and V subunits is operative. Also, unchanged levels of specific mitochondrial transcripts (e.g. steady state level of ATPase 6 mRNA) were noted in paced animals (data not shown). Studies directed to assess the post-transcriptional regulation of mitochondrial enzymes in cardiac failure including translational, import and assembly levels for these enzyme subunits are therefore needed.

Increased levels of proinflammatory cytokines including TNF-α, which have been demonstrated to play a role in the pathogenesis of CHF [4], may be pivotally involved in the mitochondrial dysfunction observed in the paced dogs. Elevated levels of circulating TNF-α and its soluble receptors have been reported in patients with CHF. The rapid production of cardiac TNF-α in response to a variety of physiological insults has supported its assignment as a ‘cardiac stress response’ gene [31]. In transgenic mice with cardiac-restricted overexpression of TNF-α, a heart failure phenotype resulted characterized by left ventricular dysfunction, elevated levels of TNF-α in the peripheral circulation from cardiac spillover and biochemical markers indicating high levels of oxidative stress [32]. Moreover, mitochondrial dysfunction has been shown to be a key element in the mechanism of TNF-α action in an assortment of cell types including skeletal muscle and cardiomyocytes [5–7]. Changes in mitochondrial membrane permeability as well as the increased generation of reactive oxygen species (ROS) primarily produced by mitochondrial electron transport complexes have been proposed to be direct results of TNF-α action [5,6]. In addition, mitochondrial respiratory activities are reduced in cardiomyocytes treated with TNF-α [7]. Studies with cells treated with TNF-α have recently shown that the mitochondrial cytochromes are critical targets of TNF-α action; cytochrome c is released to the cytosol through the mitochondrial membrane pore, cytochrome b (a major component of complex III) is reduced as are cytochromes cc1 and aa3 [6]. Ceramide, another mediator of TNF-α function, has been reported to selectively inhibit complex III activity in isolated cardiac mitochondria [33]. The TNF-α induced changes not only fundamentally modulate mitochondrial respiration but also allow mitochondria to play a key role in signalling redox and apoptotic changes to the cell. Cytokines have also been shown to impact on mitochondrial function during physiological stress conditions such as myocardial injury. Recent studies have shown that cytokine-mediated increases in inducible nitric oxide synthase (iNOS) directly inhibit mitochondrial respiration in cardiomyocytes [34,35]. The role of TNF-α in signalling mitochondrial dysfunction in both cardiac and skeletal muscle offers an attractive hypothesis which needs to be further critically examined.

Our findings show an inverse correlation of TNF-α levels with activity levels of complex III and V in paced animals consistent with a potential contributory role of TNF-α in the observed mitochondrial dysfunction. In addition, the demonstration of markedly increased left ventricular tissue aldehyde level in paced dogs indicates increased levels of myocardial oxidative stress, further evidence of free radical-induced damage in this model. The role of NO-dependent pathways (another potential molecular mechanism for TNF-α to mediate mitochondrial changes) appears to be less operative in this model since paced dogs have recently been shown to have neither increases in iNOS and eNOS activities nor in transcript levels in left ventricular tissues [36]. In order to further probe the potential relationship of TNF-α to specific mitochondrial enzyme dysfunction during CHF, future experiments to determine the precise timing of TNF-α elevation and ROS changes relative to the onset of mitochondrial enzymatic changes, as well as gauging the effects on mitochondrial function by selective blocking of TNF-α binding and function (at the level of its receptors), will be useful.

Another indication of mitochondrial dysfunction, myocardial mtDNA damage, was detectable in over 50% of the paced animals as compared to controls. However, given the low level of the mtDNA deletions found in this study, mtDNA deletions are unlikely to be a causal factor for either the reduced enzyme activity levels of complex III and V, or a key mediator for progression of heart failure. Previously, it have been suggested that increased levels of mtDNA deletions are a result of increased oxygen free radical-mediated damage to mtDNA with the generation of free radicals largely from hypoxic mitochondria and from respiratory dysfunction [37]. This contention is further supported by increased levels of specific mtDNA deletions noted in canine cardiac tissues in response to myocardial ischemia generated by ameroid constriction [38].

In children with cardiomyopathy, the use of skeletal muscle biopsies (instead of endomyocardial biopsies) has been recommended to assess respiratory enzyme function in the diagnostic evaluation of mitochondrial-mediated cardiomyopathy [17]. Pronounced deficiencies in cytochrome c oxidase (complex IV) have been shown in both cardiac and skeletal muscle of patients with Kearns-Sayre syndrome and cardiomyopathy [39]. The findings of mitochondrial dysfunction in skeletal muscle as well as in cardiac tissue in this study suggest that there may exist a commonality in the signaling for mitochondrial dysfunction in both skeletal muscle and cardiac tissues in heart failure. While more evidence confirming these findings is warranted, our current findings of skeletal muscle mitochondrial dysfunction in paced dogs provides further support that changes in skeletal muscle may reflect changes in cardiac muscle.

Currently, there is a gap in information about the correlation between the physiological, biochemical and molecular events in heart failure. There is a need for better ways to quantify changes in mitochondrial bioenergetics, structure and function not only in heart but also in skeletal muscle. This would include an evaluation of the levels of mitochondrial OXPHOS and respiratory enzyme activities, mtDNA, ATP and NADH, as well as a comprehensive investigation of the timing of mitochondrial changes as a function of progression in the severity of cardiac failure. Our findings suggest that reduction in specific mitochondrial enzyme activity levels and changes in mtDNA integrity may have the potential of becoming sensitive markers of tissue damage and progression of heart failure.


This work was supported in part by a grant-in-aid from the Heart and Stroke Foundation of Ontario (Toronto, Ontario). The authors acknowledge the expert technical assistance provided by Andrea DiRaddo and Marina Romanova.


  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
  34. [34]
  35. [35]
  36. [36]
  37. [37]
  38. [38]
  39. [39]
View Abstract