Mitochondrial DNA Damage and Dysfunction Associated With Oxidative Stress in Failing Hearts After Myocardial Infarction
Abstract—Mitochondria are one of the enzymatic sources of reactive oxygen species (ROS) and could also be a major target for ROS-mediated damage. We hypothesized that ROS may induce mitochondrial DNA (mtDNA) damage, which leads to defects of mtDNA-encoded gene expression and respiratory chain complex enzymes and thus may contribute to the progression of left ventricular (LV) remodeling and failure after myocardial infarction (MI). In a murine model of MI and remodeling created by the left anterior descending coronary artery ligation for 4 weeks, the LV was dilated and contractility was diminished. Hydroxyl radicals, which originated from the superoxide anion, and lipid peroxide formation in the mitochondria were both increased in the noninfarcted LV from MI mice. The mtDNA copy number relative to the nuclear gene (18S rRNA) preferentially decreased by 44% in MI by a Southern blot analysis, associated with a parallel decrease (30% to 50% of sham) in the mtDNA-encoded gene transcripts, including the subunits of complex I (ND1, 2, 3, 4, 4L, and 5), complex III (cytochrome b), complex IV (cytochrome c oxidase), and rRNA (12S and 16S). Consistent with these molecular changes, the enzymatic activity of complexes I, III, and IV decreased in MI, whereas, in contrast, complex II and citrate synthase, encoded only by nuclear DNA, both remained at normal levels. An intimate link among ROS, mtDNA damage, and defects in the electron transport function, which may lead to an additional generation of ROS, might play an important role in the development and progression of LV remodeling and failure.
Reactive oxygen species (ROS) induce the functional and structural damage of cardiac myocytes and may play an important role in the pathophysiology of heart failure (HF). Recent evidence has suggested an intimate link between an excessive generation of ROS and the development of myocardial remodeling and failure.1 2 3 We have demonstrated by electron spin resonance (ESR) spectroscopy that the amount of ROS increases in the noninfarcted left ventricle (LV) in a murine model of myocardial infarction (MI) with HF.4 Furthermore, we also demonstrated that mitochondrial electron transport is a possible site of superoxide anion (O2−·) generation in failing hearts,1 and H2O2, generated via O2−· dismutation by superoxide dismutase (SOD), reacted in the presence of transition metals to yield more reactive hydroxyl radicals (OH·).5
ROS can damage various cellular components, such as proteins, lipids, and DNAs. They can damage mitochondrial macromolecules either at or near the site of their formation. Therefore, in addition to the role of mitochondria as a source of ROS, the mitochondria themselves can be damaged by ROS.6 7 Mitochondrial injury is reflected by mitochondrial DNA (mtDNA) damage as well as by a decline in the mitochondrial RNA (mtRNA) transcripts, protein synthesis, and mitochondrial function. mtDNA is more susceptible to oxidative attack than nuclear DNA, possibly because of its proximity to the respiratory chain in the mitochondrial inner membrane, the lack of protective histone-like proteins, and its poor repair activity against damage.8 mtDNA mutations may prevent its replication or expression. Therefore, mitochondrial ROS may result in the progressive destruction of the mtDNA, and such mtDNA damage can lead to a decline of mtRNA transcription and a loss of function.9 Recent in vitro studies have clearly shown that ROS mediate mtDNA damage, alterations of gene expression, and mitochondrial dysfunction in cultured vascular endothelial and smooth muscle cells.10
Mitochondria contain closed circular, double-stranded DNA of ≈16.5 kb. Both strands of the mtDNA are transcribed. The mitochondrial genome encodes 13 polypeptides involved in oxidative phosphorylation, including 7 subunits (ND1, ND2, ND3, ND4, ND4L, ND5, and ND6) of rotenone-sensitive NADH-ubiquinone oxidoreductase (complex I), 1 subunit (cytochrome b) of ubiquinol-cytochrome c oxidoreductase (complex III), 3 subunits (COI, COII, and COIII) of cytochrome c oxidase (complex IV), and 2 subunits (ATPases 6 and 8) of complex V along with 22 tRNA and 2 rRNA (12S and 16S) subunits. The polypeptides are translated by mitochondrial ribosomes and consist of components of the electron transport chain.8 As opposed to nuclear-encoded genes, mitochondrial-encoded gene expression is largely regulated by the copy number of mtDNA.11 A decrease in the mtDNA copy number results in a corresponding decrease in mtRNA and proteins, finally leading to mitochondrial dysfunction.
We thus hypothesized that mtDNA, mtRNA, and proteins all decrease in failing mitochondria, in which higher levels of ROS are produced. Therefore, the objective of the present study was to measure the amount of ROS using ESR techniques and to examine the changes in the mtDNA copy number and the steady-state content of mtRNA transcripts as well as the activity of the electron transport chain complex enzymes in failing hearts. In the present study, we used a murine model of myocardial infarction (MI) after a coronary artery ligation for 4 weeks, in which the LV cavity was dilated and contractility was reduced.4
Materials and Methods
The study was approved by our institutional animal research committee and conformed to the animal care guidelines of the American Physiological Society. MI was created in male CD-1 mice by ligating the left anterior descending coronary artery.4
On the day the study was terminated, 4 weeks after surgery, one subset of investigators (S.H. and N.S.), who were not informed of the experimental groups, performed echocardiographic studies.4
Myocardial Tissue Preparation
The myocardial tissue specimens with MI were carefully dissected into two parts, one consisting of the infarcted LV with the peri-infarct rim (a 0.5- to 1-mm rim of normal-appearing tissue), and the remaining consisting of noninfarcted (remote) LV. In all subsequent assays, the comparison was made between noninfarcted LV myocardium from MI animals and control LV myocardium from sham-operated animals.
ROS were quantified in the LV by using ESR spectroscopy with 4-hydroxy-2,2,6,6-tetramethyl-piperidine-N-oxyl (hydroxy-TEMPO).5
The formation of lipid peroxides was measured in the mitochondrial fraction isolated from LV myocardium through a biochemical assay of TBARS.1
Myocardial Biochemical Measurements
The myocardial content of DNA, RNA, and protein was measured and expressed as mg/g wet weight.15 To determine whether quantitative changes in specific contractile protein content occur with MI, an electrophoretic separation of myocardial protein homogenates was performed.
DNA Isolation and Southern Blot Analysis of mtDNA
DNA was extracted from LV, and a Southern blot analysis was performed to measure the mtDNA copy number.12 13 Primers for the mtDNA probe corresponded to nucleotides 2424–3605 of the mouse mitochondrial genome, and those for the nuclear-encoded mouse 18S rRNA probe corresponded to nucleotides 435–1951 of the human 18S rRNA genome. The mtDNA levels were normalized to the abundance of the 18S rRNA gene run on the same gel. The protein levels of transcription factor A (Tfam), a nucleus-coded protein with the capacity to recognize and bind to specific mtDNA sequences, were analyzed by Western blot analysis.13
RNA Isolation and Northern Blot Analysis
Total RNA was isolated from frozen LV by the guanidinium method, and a Northern hybridization analysis was performed.14 Probes for mtRNA analysis were prepared by amplification of nucleotides 1209–2606 (probe 1), nucleotides 3351–7570 (probe 2), nucleotides 8861–14549 (probe 3), and nucleotides 14729–15837 (probe 4) of mtDNA from mouse genomic DNA (Figure 1⇓). In addition, the mRNA levels for the contractile proteins, including cardiac α-actin and myosin heavy chain, as well as mitochondrial proteins, such as manganese superoxide dismutase (MnSOD), all of which are coded by nuclear DNA, were also measured in the LVs from both groups.
Mitochondrial Enzyme Activity
The specific activity of mitochondrial complex enzymes, including complexes I, II, III, and IV, was measured in the LV. The activity of citrate synthase, one of nuclear DNA-coded mitochondrial enzymes, was also measured.
The coronal sections from mid-LV were fixed in 6% formaldehyde, and 5-μm-thick paraffin-embedded sections were stained with Masson’s trichrome. The mitochondrial ultrastructure was assessed by electron microscopy.16 17
An expanded Materials and Methods section can be found in an online data supplement available at http://www.circresaha.org.
Thirteen sham-operated and 10 MI mice were used for the present studies. Two-dimensional and M-mode echocardiography demonstrated LV dilatation and contractile impairment in the MI mouse. The summarized data of the echocardiographic measurements obtained from mice 4 weeks after the operation are presented in online Table 1 (available at http://www.circresaha.org.). In comparison with sham-operated animals, MI animals showed a significant increase in LV end-diastolic diameter (3.4±0.1 versus 6.0±0.3 mm; P<0.01) and a decrease in percent fractional shortening (42.3±1.2 versus 17.3±2.6%; P<0.01).
Oxidative Stress in the Mitochondria
ESR signals of hydroxy-TEMPO decreased more rapidly in the presence of myocardial homogenates from MI hearts compared with sham (Figure 2A⇓). There was a linear relation in the semilogarithmic plot of peak signal intensity versus time, thus indicating the first order kinetics of the signal decay (Figure 2B⇓). The rate constant of signal decay, calculated from the slope of this line, was significantly larger in MI than that in sham mice (Figure 2C⇓). DMTU (50 mmol/L) added to the reaction mixture completely abolished an increase of signal decay in MI, indicating that OH· contributed to this increase (Figure 2C⇓). SOD (50 U/mL) plus catalase (50 U/mL) also attenuated an increased signal decay rate in MI, which implies the contribution of O2−· to the production of OH·. These results confirmed our recent results that O2−· and OH· increased in the LV from MI mice.4
MI mice had a 4-fold increase in TBARS formation in the mitochondrial fraction compared with sham-operated animals (2.7±0.9 versus 11.4±2.4 nmol/mg mitochondrial protein, P<0.01), thus indicating enhanced lipid peroxidation.
Myocardial Biochemical Composition
The biochemical composition of LV from the sham and MI groups is shown in online Table 2. In the MI mice, no significant change was observed in the myocardial DNA, RNA, or protein content compared with sham mice. The quantification of actin revealed no significant change in the LV after MI (Figure 3⇓ and online Table 2).
mtDNA Copy Number and mtRNA Expression
The copy number of mtDNA, expressed as the ratio of mtDNA to nuclear DNA (18S rRNA), as measured by a Southern blot analysis, decreased by 44% (P<0.05) in MI compared with sham mice (1.03±0.02 versus 0.58±0.13; Figure 3⇑). Parallel to the mtDNA copy number, the Tfam protein level also decreased by 35% in MI animals (148±9 versus 96±6, P<0.05).
To determine the effects of a decreased mtDNA copy number on mtRNA, mtRNA transcript levels were measured by Northern blot analysis. MI showed a significant decrease in ND1+ND2, ND4, ND4L, ND5, cytochrome b, COI, COII, and COIII transcripts as well as 16S rRNA (Figure 4⇓). In contrast, the steady-state mRNA levels for cardiac α-actin were virtually identical between the sham and MI mice (63.4±5.9 versus 51.4±5.6, P=NS; Figure 4⇓). Similarly, myosin heavy chain mRNA levels were comparable between the groups (125.2±13.5 versus 102.0±19,3 P=NS). Furthermore, the MnSOD mRNA level was also comparable (68.7±3.3 versus 71.4±3.4, P=NS). As a result, all mitochondrial mRNAs tested were downregulated by 30% to 50% in MI (Figure 5⇓), whereas the mRNA levels for the nucleus-coded mRNA levels remained unchanged.
Having demonstrated that the mtRNA levels are downregulated in MI, the enzymatic activity of the mitochondria was quantified (Figure 6⇓). The enzymatic activities of complexes I, III, and IV all decreased in MI. In contrast, there was no decrease in the enzymatic activity of complex II and citrate synthase in MI, both of which were exclusively encoded by nuclear DNA.
A light microscopic analysis of myocardial tissue sections has indeed shown increased interstitial fibrosis in MI. However, ≈90% to 95% of all cross sections were occupied by cardiac myocytes in either the sham or MI mice (Figure 7⇓).
We detected no evidence of myocyte injury in MI at an ultrastructural level (Figure 7⇑). Myofibrillar organization was maintained, and the membrane structure was preserved at the sarcolemma. In both the sham and MI myocytes, the mitochondria were present throughout the cytoplasm in a characteristic organized pattern around the Z line. The sarcomere (sm) length was comparable between sham and MI myocytes (1.59±0.03 versus 1.65±0.03 μm; P=NS). However, the overall number of interfibrillar mitochondria increased in MI compared with sham mice (117±14 versus 149±9/100 sm2; P<0.05), and the overall average size of the mitochondria decreased in MI (0.59±0.04 versus 0.45±0.03 μm2; P<0.05).
The present study demonstrated that the increased generation of ROS was associated with mitochondrial damage and a dysfunction in the post-MI failing hearts, which were characterized by an increased lipid peroxidation in the mitochondria, decreased mtDNA copy number, decrease in the number of mtRNA transcripts, and reduced oxidative capacity attributable to low complex enzyme activities. These results suggest that mtDNA defects may thus play an important role, at least in part, in the development and progression of myocardial remodeling and failure after MI.
Mitochondrial Oxidative Stress
The production of ROS increased in the noninfarcted LV, which was originated from O2−· (Figure 2⇑). An increased generation of ROS was associated with a concomitant increase in the oxidation of lipids in the mitochondria. These results are consistent with our previous studies in a canine model of HF attributable to rapid ventricular pacing.1 5
The present results showed that the mtDNA copy number is decreased in the post-MI failing myocardium, whereas nuclear DNA is less affected (Figure 3⇑). Even though we could not rule out the possibility of dilutional effects by the proliferation of nonmyocytes, including fibroblasts and endothelial cells, on our measures of mtDNA, mtRNA, and enzyme activities, a relative loss of myocytes to nonmyocytes is not sufficient to explain the magnitude of the decline in mtDNA and mtRNA in MI for several reasons. First, the combination of a biochemical analysis of myocardial composition (online Table 2, available at http://www.circresaha.org) and a morphological analysis of the myocardial structure (Figure 7⇑) indicated that myocyte loss was not prominent in the noninfarcted LV. Second, the amount of Tfam protein that binds to mtDNA concurrently decreased. Third, the mRNA levels for contractile proteins did not change in MI. Fourth, nucleus-coded mitochondrial proteins, such as MnSOD and citrate synthase, did not decrease in MI. Fifth, when separating noninfarcted and infarcted tissues, we took special care not to include the peri-infarct tissue in the sample.
Furthermore, the decline in mtDNA copy number was not attributable to the defects of mitochondrial biogenesis, because an electron micrographic analysis showed that the number of mitochondria increased, and did not decrease, in the MI hearts (Figure 7⇑). Therefore, the number of mitochondrial genomes per mitochondria must be even lower. A decreased mtDNA copy number is responsible for mitochondrial dysfunction, because the maintenance of mtDNA is essential for maintaining the expression of proteins. In fact, a decrease in the mtDNA copy number has been implicated in the pathogenesis of mitochondrial diseases and zidovudine-induced myopathy.18 19
Even though the mechanisms controlling the mtDNA copy number are still poorly understood, mtDNA depletion may result from a ROS-induced mutation at the origins of replication. In addition, a mutation outside the origins of replication is capable of interfering with replication.18 It is conceivable to speculate that the mtDNA could be a major target for ROS-mediated damage for several reasons. First, mitochondria do not have a complex chromatin organization consisting of histone proteins, which may serve as a protective barrier against ROS. Second, mtDNA has a limited repair activity against DNA damage. Third, a large part of O2−·, which is formed inside the mitochondria, cannot pass through the membranes and, hence, ROS damage may be contained largely within the mitochondria. In fact, mtDNA accumulates significantly higher levels of DNA oxidation product, 8-hydroxydeoxyguanosine, than nuclear DNA does.20
Parallel to the decrease in the mtDNA copy number; ie, mitochondrial template availability, the mtRNA transcript levels also decreased in MI (Figures 4⇑ and 5⇑). The significance of these findings is additionally amplified in the absence of any changes in the expression of nuclear-derived transcripts (Figure 4⇑). Interestingly, a Northern blot analysis showed a similar extent of decrease in all of the mtRNAs examined (Figure 5⇑). In agreement with previous studies,11 these results indicate that the decrease in mitochondrial transcription could be explained by the decrease in the mtDNA copy number, and thus no changes in either the transcriptional or translational efficiencies are required.
Consistent with alterations in the mtDNA and transcript levels, the oxidative capacities decreased in MI (Figure 6⇑). Importantly, the enzymatic activity of complex II and citrate synthase was not affected, because it is entirely encoded by nuclear DNA, whereas the activity of complexes I, III, and IV, which are partially encoded by the mtDNA genes, was significantly lower in the post-MI hearts (Figure 6⇑). Our data thus imply a relationship between a decreased mtDNA copy number and an impaired mitochondrial function, because the magnitude of the mtDNA defects is parallel to quantitative deficiencies in the mtRNA transcripts and electron transport chain function.
Previous studies have suggested an intimate link between mtDNA damage, increased lipid peroxidation, and a decrease in complex enzyme activities.9 In fact, Tfam knockout mice, which had a 50% reduction in their transcript and protein levels, exerted a 34% reduction in the mtDNA copy number, 22% reduction in the mitochondrial transcript levels, and partial reduction in the cytochrome c oxidase levels in the heart.13 Furthermore, recent studies using cultured vascular endothelial and smooth muscle cells have clearly demonstrated that ROS can cause mtDNA damage, altered mitochondrial gene expression, and dysfunction.10 Nevertheless, the present observations do not allow us to establish a meaningful mechanistic link between ROS and the mtDNA defects in MI. To establish a cause-and-effect relationship between these cellular events, time-dependent changes in ROS, mtDNA, and mitochondrial function should be clarified during the development of postinfarct HF. Additionally, it would be significant to determine whether the direct exposure of cardiac myocytes to ROS results in mtDNA depletion.
Even though the mitochondrial respiration data were not available in our animals, previous studies have shown that the enzymological abnormalities of the respiratory chain are consistent with the functional defects.21 The decreased mitochondrial transcript levels and enzymatic activities are intriguing in their implication that the decline of mtDNA is associated with oxidative stress and can lead to mitochondrial dysfunction. Presumably, mitochondrial template depletion may result in its functional decline, which additionally exacerbates the defects of electron transport and produces ROS (Figure 8⇓).5 In addition, ROS can directly damage complex enzymes, because they consist of iron-sulfur clusters that can be inactivated by O2−·.
Several pathogenic mtDNA base substitution mutations, such as missense mutations and mtDNA rearrangement mutations (deletions and insertions), have been identified in patients with mitochondrial diseases.9 An accumulation of the deleted forms of mtDNA in the myocardium frequently results in either cardiac hypertrophy, conduction block, or HF.22 Furthermore, there is now a consensus view that mutations in mtDNA and abnormalities in mitochondrial function are associated with common forms of cardiac diseases, such as ischemic heart disease23 and dilated cardiomyopathy.24 In these conditions, however, the strict causal relationships between abnormalities in mtDNA and cardiac dysfunction have yet to be fully elucidated.25 Even though the mechanisms by which mtDNA damage arises in these conditions have not been clarified, ROS have been proposed to be the primary contributing factor. In the present study, we provided direct evidence that mtDNA defects occur not only in a limited small subset of mitochondrial diseases but also in a more common HF phenotype occurring after MI in mice. In addition, these alterations in mtDNA and mitochondrial function may also result from oxidative stress. The present findings are also supported by the studies on mice lacking MnSOD, which show an accumulation of oxidative damage of mtDNAs and electron transport complexes21 in association with the development of dilated cardiomyopathy.26
The remodeling of noninfarcted LV is characterized by myocyte hypertrophy and interstitial fibrosis, both of which are also recognized in pressure-overloaded cardiac hypertrophy. Therefore, we could not rule out the possibility that the similar alterations in mtDNA and electron transport complexes could be observed in hypertrophied hearts in general, especially because increased ROS generation has been demonstrated in an animal model of cardiac hypertrophy.3
The present study demonstrates that chronic increases in ROS production are associated with mitochondrial damage and dysfunction, which thus lead to a catastrophic cycle of mitochondrial functional decline, additional ROS generation, and cellular injury. Therefore, these cellular events might be involved in myocardial remodeling and failure.
This study was supported in part by grants from the Ministry of Education, Science and Culture (Nos. 09670724 and 12670676). We thank Brian Quinn for critically reading the manuscript.
Original received August 21, 2000; resubmission received October 31, 2000; revised resubmission received January 12, 2001; accepted January 21, 2001.
↵1 Both authors contributed equally to this study.
- © 2001 American Heart Association, Inc.
Ide T, Tsutsui H, Kinugawa S, Utsumi H, Kang D, Hattori N, Uchida K, Arimura K, Egashira K, Takeshita A. Mitochondrial electron transport complex I is a potential source of oxygen free radicals in the failing myocardium. Circ Res. 1999;85:357–363.
Siwik DA, Tzortzis JD, Pimental DR, Chang DL, Pagano PJ, Singh K, Sawyer DB, Colucci WS. Inhibition of copper-zinc superoxide dismutase induces cell growth, hypertrophic phenotype, and apoptosis in neonatal rat cardiac myocytes in vitro. Circ Res. 1999;85:147–153.
Kinugawa S, Tsutsui H, Hayashidani S, Ide T, Suematsu N, Satoh S, Utsumi H, Takeshita A. Treatment with dimethylthiourea prevents left ventricular remodeling and failure after experimental myocardial infarction in mice. Circ Res. 2000;87:392–398.
Ide T, Tsutsui H, Kinugawa S, Suematsu N, Hayashidani S, Ichikawa K, Utsumi H, Machida Y, Egashira K, Takeshita A. Direct evidence for increased hydroxyl radicals originating from superoxide in the failing myocardium. Circ Res. 2000;86:152–157.
Williams RS. Canaries in the coal mine: mitochondrial DNA and vascular injury from reactive oxygen species. Circ Res. 2000;86:915–916.
Sawyer DB, Colucci WS. Mitochondrial oxidative stress in heart failure: “oxygen wastage” revisited. Circ Res. 2000;86:119–120.
Wallace DC. Mitochondrial diseases in man and mouse. Science. 1999;283:1482–1488.
Ballinger SW, Patterson C, Yan CN, Doan R, Burow DL, Young CG, Yakes FM, Van Houten B, Ballinger CA, Freeman BA, Runge MS. Hydrogen peroxide- and peroxynitrite-induced mitochondrial DNA damage and dysfunction in vascular endothelial and smooth muscle cells. Circ Res. 2000;86:960–966.
Williams RS. Mitochondrial gene expression in mammalian striated muscle. Evidence that variation in gene dosage is the major regulatory event. J Biol Chem. 1986;261:12390–12394.
Igarashi-Saito K, Tsutsui H, Yamamoto S, Takahashi M, Kinugawa S, Tagawa H, Usui M, Yamamoto M, Egashira K, Takeshita A. Role of SR Ca2+-ATPase in contractile dysfunction of myocytes in tachycardia-induced heart failure. Am J Physiol. 1998;275:H31–H40.
Eble DM, Spinale FG. Contractile and cytoskeletal content, structure, and mRNA levels with tachycardia-induced cardiomyopathy. Am J Physiol. 1995;268:H2426–H2439.
Lewis W, Gonzalez B, Chomyn A, Papoian T. Zidovudine induces molecular, biochemical, and ultrastructural changes in rat skeletal muscle mitochondria. J Clin Invest. 1992;89:1354–1360.
Williams MD, Van Remmen H, Conrad CC, Huang TT, Epstein CJ, Richardson A. Increased oxidative damage is correlated to altered mitochondrial function in heterozygous manganese superoxide dismutase knockout mice. J Biol Chem. 1998;273:28510–28515.
Anan R, Nakagawa M, Miyata M, Higuchi I, Nakao S, Suehara M, Osame M, Tanaka H. Cardiac involvement in mitochondrial diseases: a study on 17 patients with documented mitochondrial DNA defects. Circulation. 1995;91:955–961.
Arbustini E, Diegoli M, Fasani R, Grasso M, Morbini P, Banchieri N, Bellini O, Dal Bello B, Pilotto A, Magrini G, Campana C, Fortina P, Gavazzi A, Narula J, Vigano M. Mitochondrial DNA mutations and mitochondrial abnormalities in dilated cardiomyopathy. Am J Pathol. 1998;153:1501–1510.
Clayton D, Williams R, Liang I. Meeting highlights: mitochondrial DNA mutations and cardiomyopathy, heart failure, and ischemic heart disease. Circulation. 1995;92:2022–2023.