INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

FAMILIAL HYPERTROPHIC CARDIOMYOPATHY (HCM) is an autosomal dominant disease characterized by varying degrees of ventricular hypertrophy and myofibrillar disarray. The vast majority of mutations in HCM thus far identified occur in genes encoding sarcomeric proteins, including -myosin heavy chain (MyHC), myosin light chains, -cardiac actin, -tropomyosin, cardiac troponin T (cTnT), cTnI, and cardiac myosin-binding protein C (41, 42, 45). In addition to being genetically diverse, the disease is also clinically heterogeneous. For example, mutations in the MyHC gene result in generally uniform hypertrophy, hemodynamic dysfunction, and variable sudden death. On the contrary, mutations in cTnT display moderate or no hypertrophy and a high incidence of sudden death (45). Thus, although it is clear that mutations in cardiac contractile proteins result in HCM, the underlying mechanisms responsible for the wide variability in phenotype have not been determined.

Undoubtedly, in addition to the primary mutation, many secondary and tertiary responses contribute to the disease phenotypes of HCM. One candidate might be cardiac mitochondria. Mitochondria, which occupy a large portion of the myocyte volume, play a critical role in the constant supply of energy to the heart. In fact, increasing evidence suggests that mitochondria could play an important role in the pathogenesis of heart disease (18, 33, 37, 44). Ide et al. (18) reported that mitochondria exhibit respiratory dysfunction in a rapid pacing model of heart failure. Abnormalities in mitochondrial respiration in patients with heart failure have been reported and attributed to decreases in the rates of NADH-ubiquinol oxidoreductase (complex I) and cytochrome c oxidase (complex IV) activities of the electron transport chain (1, 2, 46).

The majority of research investigating the mechanisms responsible for mitochondrial dysfunction in heart failure has focussed on electron transport chain components. However, it is widely believed that the supply of NADH to the electron transport chain is the rate-limiting step of NADH-linked mitochondrial respiration (3, 5, 16, 10, 29). Alterations in the rate of NADH synthesis and delivery to the electron transport chain would therefore likely have profound effects on respiratory activity. Inhibition of -ketoglutarate dehydrogenase (-KGDH), a tricarboxylic acid cycle intermediate, has been reported to have detrimental effects on mitochondrial respiration in various diseases such as cardiac ischemia-reperfusion injury, Alzheimer's disease, and Parkinson's disease (19, 21, 28, 25). In addition, unlike mitochondrial electron transport chain components, inactivation of -KGDH has been shown to directly correlate with a decrease in state 3 (ADP dependent) respiration (17). Therefore, inhibition of -KGDH as a likely mediator of mitochondrial dysfunction and contributor to the pathogenesis of HCM should be investigated.

Transgenic mouse models of HCM caused by cardiac-specific mutations in MyHC and cTnT have been made (13, 38, 40, 43). Each model exhibits several features of HCM in humans, but they differ substantially relative to each other. At 4 mo of age, mutant MyHC mice of both genders exhibit ventricular hypertrophy but normal hemodynamic function. At 8 mo of age, females continue to exhibit significant hypertrophy, but mutant MyHC male mice exhibit left ventricular hypertrophy, chamber dilation, systolic and diastolic dysfunction, exercise intolerance, and altered adrenergic responsiveness (13). These mice express ~10% of their MyHC as the mutant, with the total amount of MyHC remaining unchanged (43). Hearts with mutations in cTnT (either a COOH-terminal truncation or R92Q missense mutation) result in significantly decreased heart mass (38, 40). In addition, R92Q cTnT transgenic mouse hearts are hypercontractile and exhibit diastolic dysfunction. The three R92Q lines of mice express 33%, 67%, and 94% of their TnT as mutant, with the mutant protein replacing the endogenous TnT, leaving total TnT levels unchanged (40). In both of these models, transgenic expression of wild-type -MyHC or wild-type cTnT did not result in any detectable abnormalities (38, 39). Thus these models exhibit distinct sets of phenotypes, and there is little known about the underlying mechanisms of dysfunction. The assessment of mitochondrial function has not been reported in these or in any of the other published transgenic models of HCM.

Most genetic models of HCM are characterized by some measure of hemodynamic dysfunction, and, in the case of the mutant MyHC model, the -adrenergic pathway is altered similarly to that seen in human heart failure (13). That is, -adrenergic receptor kinase (ARK) activity is elevated and catecholamine responsiveness is blunted. In contrast, several manipulations in the adrenergic pathway have been shown to result in enhanced hemodynamic function (4, 8, 14, 35). For example, overexpression of the ₂-adrenergic receptor, expression of an inhibitor of ARK, and ablation of phospholamban all result in enhanced cardiovascular function (20, 26, 35). Therefore, a crossbreeding strategy was implemented to investigate the effects of manipulations in the adrenergic pathway on the phenotypes of mutant MyHC mice. COOH-terminal ARK (ARKct)/MyHC doubly transgenic mice have normal hemodynamic function and none of the cardiomyopathic phenotypes (14).

In this paper, we demonstrate that hearts isolated from mutant male MyHC transgenic mice exhibit ultrastructural alterations in mitochondrial organization similar to those reported in the R92Q cTnT mouse (40) and an increase in mitochondrial size. Cardiac mitochondrial respiration is significantly decreased in mutant male MyHC mice but not in R92Q cTnT mice. Mitochondrial dysfunction is observed as early as 4 mo of age and, significantly, it precedes hemodynamic dysfunction. The decrease in mitochondrial respiration is due, at least in part, to inactivation of -KGDH. Although a decrease in both complex I and IV activity was observed in hearts isolated from mutant MyHC mice, this was not likely of sufficient magnitude to cause declines in mitochondrial respiratory activity. These decreases, however, may play an important role in providing insight into mechanisms contributing to the decease in oxidative phosphorylation. Defects in the -adrenergic pathway may contribute to oxidative phosphorylation defects because inhibition of ARK in mutant MyHC mice prevents mitochondrial respiratory dysfunction.

This study suggests that altered mitochondria function may contribute to the hemodynamic dysfunction seen in HCM and offers a plausible mechanism underlying some of the heterogeneity of this disease. These experiments expand approaches that can be applied to the characterization of the pathogenesis of HCM at the cellular and molecular level.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Experimental animals. Mutant MyHC, R92Q cTnT, and ARKct/MyHC mice were developed as previously described (13, 14, 38, 39, 40, 43). Briefly, the mouse -MyHC HCM transgene was developed by inducing a point mutation, G1445A, resulting in Arg403Gln and a deletion of amino acids 468-527 bridged by the addition of nine nonmyosin amino acids. R92Q cTnT mice contain a missence mutation of mouse cTnT R92Q, and hearts of these animals express 67% of the total cTnT as the transgene, without a change in the total amount of cTnT. To investigate ARK inhibition on HCM, mutant MyHC mice were crossbred with mice expressing ARKct. The phenotypes of these mice have been described by Freeman et al. (14). Only males were used for this study, and age-matched littermate nontransgenic (NTG) mice were used as controls.

Electromicroscopic analysis. Ventricles from NTG and mutant MyHC, cTnT, ARKct, and ARKct/MyHC mouse hearts were dissected into 1-mm³ sections in a fixative consisting of 2% glutaraldehyde in 0.16 M sodium phosphate, pH 7.2, for 1 h. Sections were additionally fixed in 1% osmium tetroxide buffer for 1 h. The fixed tissue was then dehydrated in a graded ethanol series, followed by propylene oxide and embedding in Epon-Arldite. Sections (60-70 mm thick) were stained with 2% uranyl acetate and lead citrate and viewed with a Phillips CM-10 electron microscope operating at 80 kV. Quantitative analysis of mitochondrial density and size were carried out at ×11,500 using NIH Image software.

Mitochondrial isolation. Subsarcolemmal mitochondria were isolated from cardiac tissue as previously described (21). Briefly, male NTG and mutant MyHC cTnT, ARKct, and ARKct/MyHC mice (8 mo of age) were euthanized by cervical dislocation. Hearts were rapidly excised and then immersed and rinsed in ice-cold buffer containing 180 mM KCl, 5.0 mM MOPS, and 2.0 mM EGTA at pH 7.25 (I-buffer). Ventricles were isolated, dry blotted, weighed, and then minced with scissors, followed by homogenization in 20 ml I-buffer/g tissue with a Polytron homogenizer (low setting, 3 s). Elapsed time between the removal of hearts and completion of weighing was typically 30-60 s. We previously determined that no declines in the quality of the mitochondria are evident (as judged by the rates of state 3 and state 4 respiration and by the respiratory control ratio) if the time from heart extraction through weighing was extended up to 5 min. The homogenate was centrifuged at 500 g for 7.5 min at 4°C to remove nuclear and myofibrillar material. The supernatant was filtered through cheesecloth and centrifuged at 5,000 g for 10 min at 4°C. The resulting mitochondrial pellet was washed two times (ice-cold I-buffer) and resuspended in 100 µl I-buffer. A small aliquot was taken from the final suspension and used for protein determination using a Bio-Rad protein assay kit. Mitochondria kept at 4°C exhibited no change in state 3 or state 4 respiratory rates for up to 3.0 h. A portion of each mitochondrial preparation was frozen and stored for later enzymatic analyses.

Evaluation of mitochondrial O₂ consumption. ADP-independent (state 4) and -dependent (state 3) respiration were measured using a Clark-type oxygen electrode (Instech) as described previously (21). Mitochondria were diluted to a protein concentration of 0.25 mg/ml in respiration buffer (120 mM KCl, 5.0 mM KH₂PO₄, 5.0 mM MOPS, and 1.0 mM EGTA at pH 7.25). State 2 respiration was initiated by the addition of glutamate (15 mM) for the supply of reducing equivalents (NADH) to the electron transport chain. After a 2.0-min equilibration period, state 3 respiration was initiated by addition of ADP (0.25 mM). Upon depletion of ADP, state 4 respiration was monitored.

Assay of -KGDH activity. Mitochondria were diluted to 0.025 mg/ml in 25.0 mM KH₂PO₄, pH 7.2, containing 0.05% Triton X-100 and placed in a sonicating water bath for 30 s (Fisher Scientific). -KGDH activity was measured spectrophotometrically (Hitachi U-2000 spectrophotometer) by the reduction of NAD⁺ at 340 nm on the addition of 5.0 mM MgCl₂, 40.0 µM rotenone, 2.0 mM -ketoglutarate, 40 µM acetyl coenzyme A, 0.2 mM thymine pyrophosphate, and 0.5 mM NAD⁺ to sonicated mitochondria. Assays were performed at room temperature.

Electron transport chain component assays. To measure the activities of complexes I and III, mitochondria were diluted to 0.025 mg/ml of protein in buffer containing 35.0 mM KH₂PO₄, 5.0 mM MgCl₂, and 2.0 mM NaCN (to block complex IV) at pH 7.25. After three freeze-thaw cycles, enzyme activity was evaluated as previously described with minor modifications (22). Complex I activity was assayed as the rate of NADH consumption (340 nm,  = 6,200 M¹ · cm¹) on addition of 2 µg antimycin A (to inhibit complex III), 50 µM ubiquinone-1, and 75 µM NADH to a 1.0-ml vol of mitochondria (0.025 mg/ml mitochondrial protein). Complex III activity was measured as the initial rate of the reduction of cytochrome c at 550 nm ( = 18,500 M¹ · cm¹) on addition of 40 µM reduced decylubiquinone and 50 µM cytochrome c to 2.5 µg/ml mitochondrial protein. The addition of 2 µg antimycin A completely inhibited the reduction of cytochrome c. For the assay of complex IV, mitochondria were diluted in hypotonic buffer (20 mM KH₂PO₄ at pH 7.25) followed by sonication for 10 s in a sonicator bath. Enzyme activity was measured as the rate of oxygen consumption on addition of 5.0 mM ascorbate, 250 µM N,N,N',N'-tetramethyl-p-phenyldiamine, and 10 µM cytochrome c. All assays were performed at room temperature.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Ultrastructural analysis. Mitochondria have been observed to undergo significant ultrastructural alterations in various cardiomyopathies (1, 2, 40). Therefore, we decided to examine the morphology and structure of mitochondria in 8-mo-old male mutant MyHC transgenic mice. For this study, we focused on male mice because of the significant changes in ventricular geometry and function that occur progressively from 4 to 8 mo of age in male but not female transgenic mice (13). Compared with NTG mouse hearts (Fig. 1, a and b), mitochondria from mutant male MyHC mice exhibited gross disorganization and appeared to be larger in size (Fig. 1, c and d). Micrographs were analyzed for the percentage of the field occupied by mitochondria, the number of mitochondria in the field, and the cross-sectional area of individual mitochondrial. The percentage of the field occupied by mitochondria was the same in NTG (43.3%) and mutant MyHC (44.2%) mice. However, there were significantly fewer mitochondria per field in mutant MyHC hearts (27.8 vs. 38.6 mitochondria/field in MyHC vs. NTG hearts, P < 0.02). Finally, mutant MyHC mitochondria were significantly larger (1.5-fold, P < 0.0001) than those in the NTG hearts (Fig. 1e). The mitochondrial ultrastructural alterations observed in the mutant MyHC mice were similar to those previously described in the (5 wk old) mutant R92Q cTnT transgenic mouse model of HCM (40).

View larger version (118K): [in this window] [in a new window]   Fig. 1.   Electron micrographs of sections from nontransgenic (NTG; a and b) and myosin heavy chain (MyHC) mutant (c and d) 8-mo-old male mouse hearts. Magnifications are as follows: ×6,600 (a and b) and ×11,600 (b and d). e: histogram showing the distribution of individual mitochondrial size in NTG and mutant MyHC analyzed from ×11,600 electron micrographs. *P < 0.0001 (unpaired t-test).

Mitochondrial respiration in transgenic mouse hearts. To determine whether the rate of mitochondrial respiration is affected in these two transgenic models of HCM, cardiac mitochondria were isolated and oxygen consumption was measured (Table 1). Mitochondria isolated from 4-mo-old mutant MyHC transgenic mouse hearts exhibited a 29.25% decrease (P  0.001) in NADH-linked state 3 (ADP dependent) respiration. State 3 respiratory rates were 187.84 ± 24.61 and 132.90 ± 19.20 nmol O₂ · min¹ · mg mitochondrial protein¹ in NTG and mutant MyHC mice, respectively. Mitochondria isolated from 8-mo-old mutant MyHC mice demonstrated a similar 22.64% decrease in NADH-linked respiration compared with age-matched littermate NTG controls (P  0.01). Rates were 181.90 ± 28.07 and 140.73 ± 14.01 nmol O₂ · min¹ · mg mitochondrial protein¹ in NTG and mutant MyHC mice, respectively (Table 1). In contrast to the mitochondria isolated from mutant MyHC transgenic mouse hearts, mitochondria isolated from 8-mo-old R92Q cTnT transgenic mouse hearts did not exhibit any significant decrease in state 3 respiration compared with NTG controls (Table 1). The state 3 respiratory rate was 180.96 ± 11.52 nmol O₂ · min¹ · mg mitochondrial protein¹. It should be noted that mitochondrial respiration was unchanged in liver mitochondria isolated from age-matched mutant MyHC and R92Q cTnT mice, demonstrating that the dysfunction in mutant MyHC mice was specific to cardiac mitochondria (data not shown).

Dehydrogenase activities. It is widely accepted that the supply of NADH to the electron transport chain is the rate-controlling step of NADH-linked mitochondrial respiration (3, 5, 10, 17, 29). Inactivation of dehydrogenases responsible for the synthesis of NADH would likely have adverse effects on mitochondrial respiratory function. Therefore, -KGDH was assayed because its activity is required for the synthesis of NADH from glutamate, the respiratory substrate used in the respiration assay. Glutamate was chosen as an NADH-linked substrate based on comparisons of rates of oxygen consumption utilizing glutamate-malate, pyruvate-malate, and palmitoylcarnitine. Our mitochondrial preparations exhibited the greatest rate of NADH-linked ADP-dependent (state 3) respiration with glutamate as the substrate. In our hands, the addition of malate did not enhance glutamate-supported respiration further. This is in contrast to liver or brain mitochondria, neither of which utilize glutamate efficiently in the absence of malate. As shown in Table 2, -KGDH activity was significantly decreased by 27.78% (P  0.002) in cardiac mitochondria isolated from 8-mo-old mutant MyHC mice. Rates were 111.20 ± 14.22 and 80.30 ± 14.42 nmol NADH · min¹ · mg protein¹ in NTG and mutant MyHC mice, respectively. It is interesting to note that the decrease in -KGDH directly paralleled the loss in state 3 respiration observed in the mutant MyHC transgenic mice. -KGDH activity measured from cardiac mitochondria isolated from R92Q cTnT mutant mice was unchanged compared with NTG controls. The rate was 106.08 ± 11.58 nmol NADH · min¹ · mg mitochondrial protein¹ for the R92Q cTnT mutant mice. Glutamate dehydrogenase (GDH) activity was unaffected in any of the transgenic mice studied. These results identify -KGDH as a potential mediator of mitochondrial dysfunction in hearts isolated from mutant MyHC mice.

View this table: [in this window] [in a new window]   Table 2.   -KGDH and GDH activities

Complex activities. Abnormalities in electron transport chain activities have been observed in various models of heart failure (1, 2, 46). Therefore, NADH-linked electron transport chain complexes I, III, and IV were measured in mitochondria isolated from 8-mo-old mutant MyHC and R92Q cTnT mouse hearts. Measurement of complex III revealed no significant decrease in activity of mitochondria isolated from either transgenic group (Table 3). A significant decrease in complex I activity was observed in mutant MyHC cardiac mitochondria. The degree of inactivation was 18.90% with rates of 146.44 ± 15.65 and 118.77 ± 11.88 nmol NADH · min¹ · mg mitochondrial protein¹ in NTG and mutant MyHC mice, respectively (P  0.004; Table 3). Complex IV also exhibited a significant decrease in activity with rates of 779.04 ± 33.64 and 699.95 ± 56.10 nmol NADH · min¹ · mg mitochondrial protein¹ (P  0.004). Complex I and IV activities of R92Q cTnT mutant mice were unchanged compared with NTG mice.

ARKct/MyHC crossbreeding. To begin to dissect the contribution of different pathways to the mitochondrial respiratory dysfunction seen in mutant MyHC mice, we examined the effect of manipulating -adrenergic function. A ARKct/MyHC double-transgenic cross was originally developed to determine the role of ARK inhibition on the pathogenesis of disease in mutant MyHC mice. These doubly transgenic mice exhibit a wild-type phenotype in terms of hemodynamic function, the lack of expression of hypertrophic markers, and exercise (14). Therefore, mitochondrial morphology, respiratory rates, and enzymatic activities were measured in these doubly transgenic mice to determine the effect of ARK inhibition on mitochondrial dysfunction of mutant MyHC hearts. ARKct singly transgenic mice had no indication of mitochondrial ultrastructural abnormalities (Fig. 2c) compared with NTG mice (Fig. 2a). ARKct/MyHC crossbred mice exhibited disorganization of mitochondria in some limited regions of the heart, whereas, in other regions, the mitochondria appeared to be normal (Fig. 2d). Both the ARKct and ARKct/MyHC doubly transgenic mice exhibited state 3 respiratory rates that were not significantly different from NTG mice at 8 mo of age (Table 4). State 3 respiration was 191.54 ± 30.51 and 181.49 ± 16.65 nmol O₂ · min¹ · mg mitochondrial protein¹ for ARKct and ARKct/MyHC transgenic mice, respectively. In addition, ARKct and ARKct/MyHC mice did not exhibit any significant difference in -KGDH or GDH activity (Table 4). Comparatively, ARKct and ARKct/MyHC mutant mice did not exhibit any significant difference in any NADH-linked electron transport chain complex activities compared with NTG littermate controls.

View larger version (164K): [in this window] [in a new window]   Fig. 2.   Electron micrographs of sections from hearts of 8-mo-old NTG (a), MyHC mutant (b), COOH-terminal -adrenergic receptor kinase (ARKct; c), and ARKct/MyHC male mice (d). Magnifications are ×6,600.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Although the basis for the diverse disease heterogeneity of HCM is unknown, it seems likely that different downstream cellular mechanisms resulting from the specific mutation contribute to the overall phenotype. In this study, we tested the hypothesis that alterations in mitochondrial structure and function may contribute to the diversity of phenotypes in HCM. The conclusions of this paper are fourfold. First, overall mitochondrial respiratory dysfunction was demonstrated in the mutant MyHC but not the R92Q cTnT transgenic mouse model. Second, a decrease in a specific mitochondrial enzyme, -KGDH, was identified, which most likely contributes to the decrease in state 3 respiration. Third, mitochondrial dysfunction precedes hemodynamic dysfunction, consistent with a causal role. Finally, alterations in the -adrenergic pathway prevent mitochondrial dysfunction in mutant MyHC mice. Thus this study suggests that mitochondrial function may play a role in the pathogenesis of and may contribute to heterogeneity of HCM.

Mitochondria isolated from hearts of 4-mo-old mutant MyHC transgenic mice displayed a 29.3% decrease in activity that remained constant through 8 mo of age. Because hemodynamic dysfunction is not present at 4 mo of age, this result is consistent with the idea that the observed decrease in mitochondrial respiratory activity is not an effect of cardiac dysfunction but may play a causal or contributory role. Mitochondria isolated from R92Q cTnT transgenic mouse hearts did not demonstrate a significant decrease in mitochondrial respiration, despite some evidence of ultrastructural alterations (40). In addition, all NADH-linked electron transport chain components and dehydrogenases measured were unchanged in R92Q cTnT transgenic mice. Therefore, although myocyte disorganization caused by mutations in contractile proteins may result in changes in mitochondrial morphology, the alterations do not always result in respiratory dysfunction. Also, because mitochondrial respiration was not affected in this model and was significantly decreased in the mutant MyHC model, this offers a plausible mechanism for at least some of the heterogeneity of the distinct phenotypes exhibited by the two models.

Of the NADH-linked electron transport chain components measured, complexes I and IV exhibited decreases in mutant MyHC transgenic mice, whereas no significant differences were observed in R92Q cTnT mutant mice. In addition, cardiac mitochondria from MyHC but not R92Q cTnT mutants exhibited declines in NADH-linked state 3 respiration and inhibition of the tricarboxylic acid cycle enzyme -KGDH. It has previously been demonstrated that when cardiac mitochondria were titrated with varying concentrations of the specific complex I inhibitor rotenone, declines in mitochondrial respiratory rates were evident only when complex I had been inhibited by >50% (22). In contrast, when cardiac mitochondria were treated with a specific inhibitor of -KGDH, a 1:1 relationship between the degree to which NADH-linked state 3 respiration was inhibited and the magnitude of -KGDH inactivation was observed (16, 17). In the current study, the magnitude to which complex I was inhibited in cardiac mitochondria from mutant MyHC mice is <50% (18.90%). This defect is therefore unlikely to contribute significantly to observed declines in NADH-linked state 3 respiration. In contrast, the magnitude of -KGDH inactivation (27.78%) was closely paralleled by the degree to which the state 3 respiratory rate declined (22.64%). Therefore, with glutamate as the respiratory substrate, -KGDH inhibition appears responsible in a large part for inhibition of state 3 respiration. Although the observed decrease in complex I and IV activities may not be directly responsible for the loss in mitochondrial respiration, this process may play an alternative role(s) in the decrease in mitochondrial respiratory activity in the MyHC transgene. The mechanisms responsible for decreases in complex I and IV activities and their potential role in mitochondrial dysfunction in the pathogenesis of HCM need to be further investigated.

The downstream functional consequences of mutations in sarcomeric proteins linked to HCM in humans are not well understood. However, several hypotheses on the mechanisms involved in the development of the diverse phenotypes are being investigated. One area of intense investigation is the role that Ca²⁺ plays in the pathogenesis of heart disease. In fact, it has long been known that chronic elevation of intracellular Ca²⁺ is associated with the development of heart failure (7, 15, 27, 36). Myocytes isolated from mutant MyHC mouse myocardium exhibit prolonged Ca²⁺ cycling (M. C. Olsson, B. M. Palmer, L. A. Leinwand, and R. L. Moore, unpublished observations). In contrast, it has been recently reported that myofibrils from R92Q cTnT mutant mice have increased Ca²⁺ sensitivity of tension development, resulting in hypercontractility (6).

Mitochondria can sequester large amounts of Ca²⁺ from the cytosol, which can be deleterious in pathological states (12). The prolongation of Ca²⁺ transients in mutant MyHC transgenic mice may therefore result in an increase in mitochondrial Ca²⁺ concentrations. In vitro studies have shown that increased accumulation of mitochondrial Ca²⁺ can stimulate calcium-dependent proteases, nucleases, and phospholipases, such as phospholipase 2 (PLA₂) (23, 31). Activation of PLA₂ results in an increase in the concentration of acrachidonic acid. In vitro studies have reported that complex I is inhibited by arachidonic acid (12, 30). Arachidonic acid has been reported to inhibit complex I, uncouple mitochondrial respiration, and induce the permeability transition. Complex I inhibition may be due to tight associations between arachidonic acid and this multisubunit enzyme. Uncoupling, as manifested by high rates of state 4 respiration, may result from interactions between free arachidonic acid and mitochondrial membranes, thereby enabling collapse of the proton gradient independent of ATP synthesis. In this study, the mitochondria are not uncoupled as reflected by relatively low state 4 respiratory rates (and respiratory control ratio values >6). This is likely because during mitochondrial preparation, free arachidonic acid is not retained within the mitochondria and the effects of free arachidonic acid (e.g., uncoupled respiration and activation of the permeability transition) would not be expected to be evident. Importantly, tight associations between complex I and arachidonic acid may be preserved and could contribute to the observed inhibition of complex I.

Consistent with the notion that Ca²⁺ plays a role in the observed mitochondrial dysfunction is the fact that the -adrenergic pathway is responsible for the strength of cardiac contraction by regulation of Ca²⁺ cycling. Activation of cardiac -adrenergic receptors increases cardiac contraction by a number of different mechanisms. However, continual adrenergic activation, which is observed in clinical cardiac failure and in the mutant MyHC transgenic mouse, results in an upregulation of ARK1, an uncoupler of the -adrenergic pathway. As a result, Ca²⁺ cycling is altered. Expression of a ARKct peptide inhibitor has been shown to provide beneficial effects in several models of cardiomyopathy (14, 35). ARKct/MyHC doubly transgenic mice were not hypertrophied and were similar to the wild type with respect to hemodynamic function and exercise tolerance. The observed prevention of the cardiomyopathy seems likely due to the prevention of altered Ca²⁺ cycling observed in the mutant MyHC mice, and it is possible that the beneficial effects of ARK1 inhibition are due to other biological effects in cardiac myocytes than increased cardiac contractility. For example, another possibility is that increased Ca²⁺ cycling in ARKct/MyHC mice may prevent Ca²⁺ sequestration by the mitochondria. Therefore, the potential role of Ca²⁺ mediated mechanisms of mitochondrial damage in mutant MyHC mice deserves attention.

The majority of investigations into potential sites responsible for degeneration of mitochondrial function during the development of heart failure have focused on alterations in the activities and composition of various electron transport chain components. The findings presented in this study indicate that other mechanisms can be responsible for the development of cardiac mitochondrial dysfunction. It is well established that the supply of reducing equivalents (i.e., NADH) controls the rate of NADH-linked mitochondrial respiration and ATP synthesis (3, 5, 10, 17, 29). Thus, in contrast to electron transport, any alterations in concentrations of NADH would likely be reflected in the rate of NADH-linked mitochondrial respiration. -KGDH, a multienzyme complex, has previously been reported to be responsible for mitochondrial dysfunction in cardiac ischemia-reperfusion injury, Alzheimer's disease, and Parkinson's disease (21, 19, 25, 28). In vitro studies have revealed that selective inhibition of -KGDH leads to a loss in NADH-linked respiration, which directly correlates with the degree of -KGDH inactivation (16, 26). In addition, -KGDH is highly sensitive to free radical-mediated inactivation (16). In the current study, the decrease in -KGDH activity in mutant MyHC transgenic mice reflected declines in mitochondrial respiration. In contrast, GDH activity was unaffected in any of the transgenic mice, indicating the specificity of the affect to -KGDH. This study has identified specific sites of enzymatic dysfunction responsible for the decrease in mitochondrial respiratory activity that play a role in the pathogenesis of HCM.

In summary, this study suggests a role of mitochondrial dysfunction in the heterogeneity of HCM. Furthermore, inhibition of ARK1 in the mutant MyHC transgenic mouse prevents mitochondrial dysfunction and offers plausible mechanisms responsible for the pathogenesis of HCM. Increasing knowledge in the relationship between mitochondrial calcium concentrations and oxidative damage to specific mitochondrial proteins responsible for mitochondrial dysfunction will likely result in the development of potential therapeutic agents for the treatment of HCM.