Heart Failure Reviews

, Volume 18, Issue 5, pp 645–656

Alterations in mitochondrial function in cardiac hypertrophy and heart failure

Authors

  • Moritz Osterholt
    • Department of Cardiothoracic Surgery, Jena University HospitalFriedrich Schiller University of Jena
  • T. Dung Nguyen
    • Department of Cardiothoracic Surgery, Jena University HospitalFriedrich Schiller University of Jena
  • Michael Schwarzer
    • Department of Cardiothoracic Surgery, Jena University HospitalFriedrich Schiller University of Jena
    • Department of Cardiothoracic Surgery, Jena University HospitalFriedrich Schiller University of Jena
Article

DOI: 10.1007/s10741-012-9346-7

Cite this article as:
Osterholt, M., Nguyen, T.D., Schwarzer, M. et al. Heart Fail Rev (2013) 18: 645. doi:10.1007/s10741-012-9346-7

Abstract

Normal cardiac function requires high and continuous supply with ATP. As mitochondria are the major source of ATP production, it is apparent that mitochondrial function and cardiac function need to be closely related to each other. When subjected to overload, the heart hypertrophies. Initially, the development of hypertrophy is a compensatory mechanism, and contractile function is maintained. However, when the heart is excessively and/or persistently stressed, cardiac function may deteriorate, leading to the onset of heart failure. There is considerable evidence that alterations in mitochondrial function are involved in the decompensation of cardiac hypertrophy. Here, we review metabolic changes occurring at the mitochondrial level during the development of cardiac hypertrophy and the transition to heart failure. We will focus on changes in mitochondrial substrate metabolism, the electron transport chain and the role of oxidative stress. We will demonstrate that, with respect to mitochondrial adaptations, a clear distinction between hypertrophy and heart failure cannot be made because most of the findings present in overt heart failure can already be found in the various stages of hypertrophy.

Keywords

MitochondriaCardiac hypertrophyHeart failureFatty acid oxidationGlucose oxidationOxidative stress

Introduction

Despite considerable progress in cardiovascular research that has translated into improved therapies, heart failure (HF) is still among the most common causes for hospitalization and remains a major public health burden [1, 2]. With a continuously aging population, it is expected that the prevalence of heart failure will further rise [3]. A better understanding of the pathophysiology of heart failure will help to counter this trend by extending targeted therapeutic options.

Heart failure is a multifactorial clinical syndrome [4]. It is suggested that progressive development of HF follows an index event that acutely or chronically impairs the heart’s pumping capacity (e.g., myocardial infarction, pressure overload due to hypertension or aortic stenosis, volume overload due to aortic and mitral regurgitation, etc.) [5]. In pressure overload, elevated left-ventricular pressure leads to an increase in ventricular wall stress during systole. This increase in wall stress triggers left-ventricular concentric hypertrophy, which then, according to the Laplace law, would normalize ventricular wall stress again [6, 7]. When subjected to volume overload, increased wall stress during diastole leads to eccentric hypertrophy of the heart [6]. Hypertrophy is considered to be an initially adaptive response (compensated hypertrophy), counteracting the increased wall tension and helping to maintain cardiac output. However, if the heart is persistently exposed to increased load, hypertrophy may become maladaptive. Anatomical changes during ventricular hypertrophy include perivascular and myocardial fibrosis as well as thickening of intramyocardial coronary arteries. Hypertrophy and fibrosis impede cardiac microcirculation, resulting in tissue hypoxia and following loss of cardiac myocytes [8, 9]. Therefore, cardiac function may progressively decline as hypertrophy gradually transitions into HF. Hypertrophy has been suggested to occur in two distinct types: physiological hypertrophy and pathological hypertrophy. We have reviewed mitochondrial alterations in both entities before [10]. Briefly, different patterns of mitochondrial adaptation are accompanied by the activation of divergent signaling pathways. For example, during physiological hypertrophy, mitochondrial adaptations are mediated at least in part by phosphatidylinositol-3-kinase (PI3K), independent of Akt [11]. On the other hand, activation of NFAT is linked to the development of pathological hypertrophy [12]. Nevertheless, there is also considerable overlap rendering some findings highly contradictory. Exercise, for example, has been related to the development of physiological hypertrophy [13, 14]. However, we showed that repetitive exercise (with similar protocols) can also induce mitochondrial and contractile dysfunction, although these changes appear to be temporary [15]. With these inconsistencies in the literature, we still lack the ability to clearly distinguish physiological from pathological hypertrophy, except by the phenotype: While cardiac function is maintained (or even improved) in physiological hypertrophy, pathological hypertrophy will eventually lead to HF.

Clinically, HF may further be divided into diastolic and systolic HF [16]. While the exact reasons and therefore also potential differences between the two entities are not fully elucidated, it might well be that common mechanisms are the basis for all types of HF [10, 17]. Diastolic dysfunction can be caused by several mechanisms. One of them is a reduction in the ATP/ADP ratio as relaxation is an energy-requiring process [18]. Reduced availability of ATP will therefore impede myocardial relaxation and increase left-ventricular stiffness. At the same time, sufficient ATP levels are needed for optimal function of ion pumps (e.g., calcium transporters). Low ATP content may impair myocardial calcium homeostasis, thereby reducing contractility and causing systolic dysfunction [19]. It is therefore important to address the basic principles of how contractile function is generated and supported under conditions of increased loading conditions.

While “hemodynamic stress” (i.e., pressure or volume overload) is the initial event that induces hypertrophy, various mechanisms have been implicated in the development of hypertrophy and the progression to HF, mitochondrial dysfunction being one of them [20]. Mitochondria are multi-functional organelles that are manifoldly involved in the development of HF (e.g., myocardial cell death by apoptosis, oxidative stress, abnormal intracellular calcium handling, etc.) [2124]. To maintain cardiac function and to meet the constantly present workload, the heart requires a continuous supply of ATP. About 90 % of ATP is synthesized inside the mitochondria [25]. Hence, perturbed mitochondrial function results in a mismatch between ATP demand and production, leading to impaired contractile function.

Here, we review mitochondrial changes that occur during cardiac hypertrophy and that lead to heart failure, in particular mitochondrial substrate metabolism, respiratory chain activity, and proteomic remodeling as well as oxidative stress. We will illustrate that mitochondrial alterations involved in hypertrophy cannot really be distinguished from those involved in HF (cf. Table 1). Thus, it is necessary to address both conditions when discussing mitochondrial alterations in hypertrophy.
Table 1

Alterations in mitochondrial function in left-ventricular hypertrophy and heart failure and the models used for investigation (compared to control)

 

Compensated left-ventricular hypertrophy (preserved contractile function)

Heart failure (reduced contractile function)

Fatty acid oxidation

Decreased

 Pressure overload in rats [33, 34]

 Dahl salt-sensitive rats [38]

 Clinical study [67]

Unchanged

 Pressure overload in rats [39]

Decreased

 Pressure overload in rats [33]

 Rapid pacing in dogs [46, 56]

 Clinical study [68]

 Coronary artery ligation in rats [41]

mRNA of fatty acid oxidation genes

Decreased

 Spontaneous hypertension in rats [55]

Unchanged

 Pressure overload in rats [39]

Decreased

 Rapid pacing in dogs [49]

 Myocardial infarction in mice [50]

 Spontaneous hypertension in rats [55]

 Clinical study [55]

Glucose oxidation

Increased

 Spontaneous hypertension in rats [45]

Unchanged

 Pressure overload in rats [33, 34]

Increased

 Rapid pacing in dogs [46, 56]

Decreased

 Pressure overload in rats [33]

ETC complex activity

Decreased

 Spontaneous hypertension in rats [79]

Unchanged

 Pressure overload in rats [73]

Decreased

 Pressure overload in rats [73]

 Rapid pacing in dogs [7678]

 Spontaneous hypertension in rats [79]

 Coronary artery ligation in rats [74]

 Coronary artery ligation in mice [87]

Unchanged

 Coronary artery ligation in rats [75]

ROS production

Decreased

 Pressure overload in guinea pigs [108]

Increased

 Pressure overload in guinea pigs [107]

Unchanged

 Pressure overload in rats [105]

Increased

 Rapid pacing in dogs [78]

 Clinical study [102]

 Ang-II administration in mice [104]

 Pressure overload in rats [105]

 Pressure overload in guinea pigs [108]

 Pressure overload in mice [113]

 Coronary artery ligation in mice [87]

ETC electron transport chain, ROS reactive oxygen species, Ang-II angiotensin II

Mitochondria and cardiac substrate metabolism

When beating at a pace of 70 bpm, the heart consumes approximately 30 kg of ATP each day [26], which is primarily regenerated in mitochondria by oxidative phosphorylation. In the normal heart, 60–90 % of cardiac ATP demand can be satisfied by fatty acid oxidation. Oxidizing pyruvate, which is formed via glycolysis or from lactate, provides the remaining 10–40 % of ATP [27]. Both pathways, pyruvate and fatty acid oxidation, are situated in the mitochondrial matrix. Due to its intrinsic metabolic flexibility, the heart is able to vary the proportion of ATP produced by the metabolism of individual substrates. This flexibility allows for optimized adaptation to changes in substrate availability, cardiac workload, and oxygen supply [28]. However, shifted substrate preference may be associated with altered ATP yield as the oxidation of fatty acids yields more ATP per mole substrate than the oxidation of glucose [29].

A possible link between disturbed mitochondrial substrate metabolism and mechanical dysfunction of the heart was established by the theory of the “energy-starved heart” [30, 31]. In theory, an increase in ATP demand should be adequately matched by an increase in the activity of the ATP-producing machinery and thus augmented ATP supply. According to the energy-starvation concept, increased load and hypertrophy elevate cardiac ATP demand, which is, however, not met by a sufficiently increased ATP supply. The consequence would be impaired cardiac mechanical function. The most striking indirect but clinically relevant evidence for this concept is the inverse correlation of the myocardial phosphocreatine/ATP ratio and mortality in patients with heart failure [32]. Several factors may contribute to impaired ATP availability. Below, we review changes in the mitochondrial utilization of fatty acids and glucose, which may limit ATP production rates in the setting of cardiac hypertrophy and failure.

Numerous studies have reported reduced fatty acid oxidation rates in hypertrophy and HF (Table 1). Using ex vivo heart perfusions, we [33] and others [3437] have assessed the rates of cardiac glucose and fatty acid oxidation in a model of pressure overload-induced hypertrophy that ultimately leads to HF. By assessing both substrate oxidation and cardiac function at different points in time, we could demonstrate that fatty acid oxidation is not only reduced in HF, but already impaired at the stage of compensated hypertrophy and precedes the onset of HF [33]. In a comparable model of pressure overload, Akki et al. [34] reported reduced palmitate oxidation during compensated hypertrophy. Similar data are found by dual-tracer autoradiography in Dahl salt-sensitive rats, where fatty acid oxidation is reduced in compensated hypertrophy, although cellular fatty acid uptake is increased [38]. Other studies found a reduction in fatty acid utilization only in HF and normal rates in compensated hypertrophy [39]. In a model of infarction-induced hypertrophy and left-ventricular dysfunction, fatty acid oxidation was mildly reduced 8 weeks after coronary artery ligation, however, this reduction did not reach statistical significance [40]. In a similar model, this reduction was more pronounced and significant when palmitate oxidation rates were determined 6 months after coronary artery ligation and correlated with the extent of left-ventricular dysfunction [41]. By using simultaneous infusion of labeled substrates in a study of microembolization-induced hypertrophy, fatty acid oxidation was found to be normal during the stage of mild HF [42]. Preserved fatty acid oxidation was also calculated in a study in mild HF at basal workload [43], but fatty acid oxidation could not be triggered by beta-adrenergic stimulation. Under these conditions, it was significantly reduced compared to control animals.

The decline in fatty acid oxidation is often accompanied by a shift to glucose utilization. This shift is often referred to as a “substrate switch” [44]. However, the term can be misleading, as it does not necessarily imply an increase in glucose oxidation. As it is defined by the ratio of fatty acid oxidation to glucose oxidation, an isolated decrease in fatty acid oxidation would mark a substrate switch even in the absence of absolute changes in glucose oxidation. In cardiac hypertrophy and failure, glucose oxidation has been found to be increased [40, 45, 46], unchanged [3436], or even reduced [33]. Of particular interest are reports of increased uncoupling of glycolysis and glucose oxidation in cardiac hypertrophy [35, 36, 47]. Although basal glucose uptake and glycolysis are increased, glucose oxidation is not stimulated to the same extent [37]. This raises the question, how Krebs cycle flux can be maintained under conditions of reduced oxidation of fatty acids and glucose, resulting in limited acetyl-coA availability. A possible mechanism may be anaplerosis [48], by which carbon is introduced at other sites of the Krebs cycle. Sorokina et al. [35] suggested that glucose oxidation by pyruvate dehydrogenase is decreased in favor of increased anaplerotic flux. Thereby, Krebs cycle flux can be maintained at a relatively normal level. However, as this would be a less efficient mode to utilize glucose for ATP synthesis, it may feed into the energy-starvation hypothesis and actually contribute to the transition from compensated hypertrophy to heart failure.

Several mechanisms may be responsible for reduced fatty acid oxidation. Interestingly, despite somewhat opposite changes in the utilization of fatty acids and glucose, a global transcriptional down-regulation was found in hypertrophy and failure, affecting fatty acid and glucose oxidation alike [49, 50]. The concurrent increase in glucose utilization can be explained by the mutual regulation of glucose and fatty acid oxidation, as initially proposed by Randle [51]. According to this, glucose oxidation is repressed by fatty acids mainly at the level of the pyruvate dehydrogenase complex [52]. Reciprocally, fatty acid oxidation can be reduced by glucose via inhibition of CPT-1, the rate-limiting enzyme of fatty acid oxidation [53].

Transcriptionally, fatty acid oxidation is regulated by the peroxisome proliferator-activated receptor-α (PPARα) and its coactivator peroxisome proliferator-activated receptor-γ coactivator 1α (PGC-1α). Increased expression and activity of PPARα leads to increased transcription and translation of genes involved in fatty acid oxidation (for further details see [54]). However, gene expression of an enzyme does not necessarily depict its activity. In an elegant study, Sack et al. [55] assessed mRNA and protein levels as well as activity of MCAD (medium chain acyl-coA dehydrogenase) in rats with left-ventricular hypertrophy and HF. During hypertrophy, MCAD mRNA was significantly reduced, while MCAD protein level and activity were unchanged compared to control rats. Only at the stage of HF, a coordinated down-regulation of RNA levels, protein levels, and MCAD activity was observed. Reduced MCAD activity was also reported in a dog model of pacing-induced heart failure [56]. Besides defects of enzymes directly involved in beta-oxidation, various abnormalities in cellular fatty acid handling may impair fatty acid oxidation as well [57]. Reduced expression of sarcolemmal fatty acid transporters may contribute to reduced cellular availability of fatty acids, thereby limiting fatty acid oxidation [58, 59]. Further, carnitine deficiency in cardiac hypertrophy may limit mitochondrial uptake of long chain fatty acids and lead to reduced fatty acid oxidation [57]. Besides a lack of carnitine, reduced CPT-1 activity was associated with impaired fatty acid oxidation [56].

Hypertrophy does not only raise cardiac ATP demand but also increases diffusion distances for oxygen and substrates [6, 60, 61]. In a model of microembolization-induced heart failure, reduced capillary density and increased diffusion distances are found in regions with severe interstitial fibrosis [62]. Under these circumstances, cardiomyocytes may become hypoxic. In theory, glucose oxidation is favorable under these conditions, as ATP production from glucose by glycolysis and glucose oxidation requires less oxygen (6 mol oxygen per mole glucose) than ATP production from fatty acid oxidation (23 mol oxygen per mole palmitic acid) [63, 64]. However, eventually hypoxia will impair mitochondrial respiration, impeding global aerobic substrate metabolism. This response is in part mediated by the hypoxia-inducible factor 1-alpha (HIF-1α) [65]. While hypoxia limits the substrate availability for complex IV, a coordinated transcriptional response leads to increased expression of genes encoding enzymes involved in glycolysis and concomitant reduced expression of mRNA encoding proteins of the respiratory chain, additionally favoring anaerobic glucose utilization [66]. Therefore, ATP generation is shifted from aerobic pathways with high ATP yield (i.e., oxidative phosphorylation) to anaerobic pathways at the substrate level in the cytosol with substantially lower ATP yield (i.e., anaerobic glycolysis).

Above-cited studies in various animal models of HF have been complemented by clinical studies addressing substrate metabolism in human left-ventricular hypertrophy and HF. In agreement with our results in compensated hypertrophy in the rat heart [33], cardiac fatty acid oxidation is already impaired in patients with left-ventricular hypertrophy with preserved ejection fraction [67]. During HF with reduced systolic function, fatty acid oxidation impairment persists [68, 69]. Interestingly, when left-ventricular function is severely impaired and patients show signs of insulin resistance, fatty acid oxidation rates tend to increase again [69]. Recapitulating these findings, it is conclusive that dysregulation of the fatty acid oxidation system begins already in compensated hypertrophy.

While the energy-starvation concept has become widely accepted, other mechanisms may as well contribute to the development of mechanical dysfunction in HF. Less efficient use of ATP may result in impaired contractile function. If ATP deficiency alone would cause contractile dysfunction, one might expect that the ratio of ATP production to cardiac power would not change during the development of HF: Decreased ATP production (leading to decreased ATP levels) would result in a concerted decline of cardiac power. However, data from our lab indicate that the amount of ATP needed to generate the same amount of power is increased with the onset of decompensated hypertrophy [33]. These changes do not necessarily controvert the impact of reduced ATP production following reduced substrate oxidation rates. Nevertheless, further studies will be helpful to assess other mechanisms that may play a role in the development of contractile dysfunction in HF.

From the current knowledge on mitochondrial substrate metabolism, we conclude the following: (1) Early in the development of hypertrophy and HF, defects in the mitochondrial metabolic machinery begin to evolve, as shown by impaired expression of genes involved in both fatty acid and glucose oxidation. The extent of disruptions in the oxidation of fatty acids is expanding during the progression of hypertrophy and HF. (2) Although the decrease in fatty acid oxidation is often paralleled by an increase in glucose utilization, the latter cannot compensate for the reduced ATP production arising from lower fatty acid oxidation rates. (3) As an addition to metabolic changes as primary events in the development of pressure overload-induced hypertrophy leading to HF, we suggest that a less efficient use of ATP (i.e., a reduction in the ratio of cardiac power to ATP production) plays an important role in the development of mechanical dysfunction.

As mentioned above, substrate oxidation produces reducing equivalents that are fed into the respiratory chain to enable oxidative phosphorylation (Fig. 1). As the accumulation of reducing equivalents inhibits substrate oxidation, defects in the respiratory chain that limit oxidation of NADH and FADH2 may have adverse effects on substrate oxidation rates as well. In the following paragraph, we will therefore review changes in the level of the electron transport chain that are found in HF.
https://static-content.springer.com/image/art%3A10.1007%2Fs10741-012-9346-7/MediaObjects/10741_2012_9346_Fig1_HTML.gif
Fig. 1

Mitochondrial centrality in cellular metabolism. After initial metabolization in the cytosol, fatty acyl-coA and pyruvate are imported into the mitochondria. Here, further degradation results in the formation of acetyl-coA, which is fed into the Krebs cycle. Alternatively, Krebs cycle intermediates may be replenished by anaplerotic reactions (e.g., carboxylation of pyruvate). In the respiratory chain, reducing equivalents (NADH/H+) are oxidized as protons are pumped across the inner mitochondrial membrane (IMM). Utilizing the proton gradient, complex V (CV) regenerates ATP from ADP. Reactive oxygen species (as \( {\text{O}}_{2}^{ \cdot - } \)) are formed in particular at complex I (CI) and complex III (CIII) of the electron transport chain, increasing oxidative stress. Further, NADPH oxidase 4 (Nox4) contributes to the development of oxidative stress. Whether Nox4 is localized in the inner or outer mitochondrial membrane still needs to be elucidated. FACS Fatty acyl-coA synthase. CPT1/CPT2 Carnitine palmitoyltransferase 1/2. CT Carnitine acyltranslocase. OMM Outer mitochondrial membrane. MPC Mitochondrial pyruvate carrier. PDC Pyruvate dehydrogenase complex. PC Pyruvate carboxylase. CIV Complex IV

Altered activity of the electron transport chain

The respiratory chain constitutes the final common pathway of aerobic ATP production. The enzymes that are involved in oxidative phosphorylation (e.g., complexes I-IV and ATP synthase, also referred to as complex V) are situated in the inner mitochondrial membrane (Fig. 1) [70, 71]. In principle, NADH and FADH2 feed electrons into the electron transport chain (ETC), which are then transported along the complexes to eventually reduce oxygen. This electron transport is directly linked to the expulsion of protons from the mitochondrial matrix across the inner mitochondrial membrane into the inter-membrane space. Consequently, a chemical and electrical gradient across the inner membrane is generated. The energy conserved in this gradient is then utilized by the ATP synthase through rechanneling protons to the mitochondrial matrix, regenerating ATP out of ADP [71, 72]. A defect of enzymes involved in mitochondrial respiration would have two consequences: (1) A decline in the proton gradient across the inner mitochondrial membrane with reduced ATP production by complex V and (2) Accumulation of reducing equivalents leading to inhibition of substrate oxidation pathways. A number of defects in the electron transport chain have been described in various models of HF, while most studies describe unchanged complex activities in compensated hypertrophy. Data from our lab suggest that impaired ATP production in pressure overload-induced hypertrophy leading to HF is due to a defect in the electron transport chain, specifically at complex I. State 3 respiration—a marker of mitochondrial function—is impaired in HF for a variety of substrates that feed complex I [73]. Only when using the complex II-substrate succinate, maximal respiratory capacity was unchanged [73]. Notably, systolic dysfunction paralleled this decline in mitochondrial respiratory capacity, while diastolic dysfunction occurred already at earlier stages, when state 3 respiration was increased or unchanged [33]. In a study of coronary artery ligation-induced heart failure in rats, Heather et al. [74] report reduced mitochondrial respiration associated with reduced complex III activity and cytochrome c content 2 weeks after ligation. In a comparable model, however, Rennison et al. [75] found no changes in mitochondrial complex activities or respiration rates 8 weeks after coronary artery ligation.

Pacing-induced hypertrophy at the stage of overt HF has been associated with reduced activity of complexes III and V [76], complexes I, III, and V [77], or complex I [78]. In hypertensive rats, HF was accompanied by a defect of complex IV [79]. Various defects of electron chain complexes have been described during different stages of HF in patients [8083].

Mitochondria contain their own, extra-nuclear DNA (mtDNA), which encodes 13 subunits of the respiratory chain complexes, with the exception of complex II that is encoded by nuclear DNA only [84]. Mitochondrial dysfunction in terms of reduced activity of any of these enzymes may therefore be caused by mtDNA mutations or depletion. In fact, several defects of mtDNA have been linked to cardiomyopathies [85]. Studies in human and experimental HF reported mtDNA mutations associated with mitochondrial dysfunction [76, 86]. However, the authors doubt the relevance of these findings, as only a small proportion of mtDNA molecules displayed mutations. Nevertheless, there is evidence for the impact of mtDNA depletion on mitochondrial function: In a model of infarction-induced heart failure, increased oxidative stress was associated with reduced mtDNA copy number and a parallel decrease in mtDNA-encoded gene transcripts [87]. This was associated with reduced activity of mitochondrial complexes I, III, and IV, which all contain subunits that are encoded by mtDNA, whereas complex II activity was unchanged. Further, mice lacking the mitochondrial transcription factor TFAM present a decline in mtDNA that was paralleled by a drastic reduction in respiratory chain enzyme activities [88]. Consequently, ATP synthesis rates dramatically dropped. In human HF, defects in replication lead to a depletion of mtDNA, combined with a significant reduction in mtDNA-encoded proteins and impaired mitochondrial biogenesis [89].

Apart from enzyme activities, changes in mitochondrial protein levels have been evaluated by numerous studies. We have reported reduced expression of proteins involved in fatty acid oxidation as well as diverse changes in the expression of proteins of the electron transport chain during pressure overload-induced hypertrophy and HF [90]. A reduction in state 3 respiration was accompanied by reduced ex vivo oxidation rates and a decrease in abundance of approximately 50 % of ETC proteins, while 25 % of all ETC proteins were increased in abundance. In particular, reduced expression of complex I subunits corresponds with a simultaneous reduction in complex I activity in a different study using the same animal model [73]. Although other subunits were increased in their abundance, this likely adaptive mechanism fails to normalize its function. Similar results of altered mitochondrial protein expression come from studies in pacing-induced HF [91], hypertensive rats with hypertrophy resulting in HF [92], as well as in patients with inflammatory dilated cardiomyopathy that showed signs of HF [93]. Defects of mitochondrial complexes are more frequently described in overt heart failure than in compensated hypertrophy (Table 1). This may suggest that the reduction in ETC enzyme activities could be involved in the transition of cardiac hypertrophy into HF. However, the great variability of complex defects associated with HF may also indicate a rather epiphenomenal role. In this case, adverse circumstances in overt HF (e.g., oxidative stress) may be responsible for the impairment of mitochondrial function. Nevertheless, deterioration of ATP production and ultimately cardiac function would progress.

Concluding, pressure overload-induced hypertrophy and HF are associated with defects in the respiratory chain. As these defects are more frequently found in overt HF, they may be implicated in the decompensation of cardiac hypertrophy. Another aspect in which the electron transport chain may contribute to the progression of HF is its potential to generate reactive oxygen species.

The role of oxidative stress

Oxidative stress, characterized by the accumulation of reactive oxygen species (ROS), is implicated in the pathogenesis of many diseases, including HF [9496]. Although ROS are essential for a variety of physiological processes [97], excessive concentrations of ROS may damage cellular components, compromising their function and thereby triggering a variety of adverse changes that are found in HF (e.g., hypertrophy, cardiac fibrosis, impaired contraction, cell death via necrosis, and apoptosis) [98].

The mitochondrial electron transport chain (ETC) is a known source of ROS. As mitochondria account for approximately 30 % of the cardiomyocytes’ mass [99], it has been suggested to be one of the most significant sources of ROS in the heart [100, 101]. The implication of oxidative stress in HF has been postulated based on studies in animals [78, 87] and humans [102, 103]. Frequently, oxidative stress is accompanied by impaired endogenous protection against increased ROS production [103]. The relevance of mitochondrial ROS production in the development of cardiac hypertrophy and failure was supported by a recent study by the Rabinovitch group [104]. In a model of angiotensin-II-induced hypertrophy, they report that overexpression of mitochondria-targeted catalase, but not wild-type catalase, was able to reduce oxidative damage and attenuated left-ventricular hypertrophy and diastolic dysfunction [104]. However, we were unable to establish such a link in a non-genetically modified rat model, where hypertrophy and heart failure were induced by constriction of the aorta. After inducing pressure overload, ROS production was normal during compensated hypertrophy. It increased with the onset of diastolic dysfunction and remained high during the following progression of HF [105]. Interestingly, while the application of an antioxidant significantly reduced ROS production during decompensated hypertrophy, it neither had an effect on the further decline of cardiac function nor on survival [106].

While there is considerable evidence that oxidative stress occurs in HF, less is known about ROS production rates in early, compensated hypertrophy. It is beyond doubt that increased ROS production can damage mitochondrial proteins. However, it is still unclear whether the emergence of oxidative stress triggers the transition from compensated hypertrophy to failure. This uncertainty is mainly due to the scarcity of longitudinal studies assessing cardiac function and ROS production at different stages of the development of hypertrophy and HF. Li et al. [107] suggested a role for oxidative stress in the progression of hypertrophy and the development of HF. However, their study lacks data showing beneficial effects of inhibited oxidative stress. In pressure overload-induced hypertrophy, a beneficial effect of long-time vitamin E therapy was reported: Continuous supplementation with vitamin E following aortic banding reduced oxidative stress during hypertrophy and prevented the development of HF [108]. In clinical studies, effects of antioxidant supplementation were less promising [109]. In double-blind, randomized clinical studies, application of the antioxidant vitamin E in high-risk patients did not alter rates of hospitalization for heart failure, nor did it improve cardiac function or prognosis in patients with advanced heart failure [110, 111].

Recently, other sources in- and outside the mitochondrion have aroused interest of scientists and raised doubt about the principal importance of ROS arising from the ETC. Besides the ETC, NADPH oxidase has been implicated in the occurrence of oxidative stress in the heart. In patients with end-stage HF, NADPH oxidase activity is increased and a likely contributor to oxidative stress [102]. NADPH oxidase 4 (Nox4) is localized in mitochondria, and there is evidence that it is a significant source of O2 in cardiac myocytes [112, 113]. In mice with cardiac-specific Nox4-deletion, 80 % reduction in Nox4 protein amount in the heart attenuates the development of hypertrophy, interstitial fibrosis, and mitochondrial dysfunction in response to pressure overload and is associated with reduced ROS production [113]. Improved mitochondrial function in mice with cardiac Nox4-deletion is likely due to reduced oxidative damage to mitochondrial proteins.

While oxidative stress is found in various models of HF, its role in the development of HF may still be model dependent. With clinical studies failing to prove a beneficial effect of antioxidant therapy and supporting results from studies in different animal models of HF, it remains a matter of debate whether increased ROS production is a primary or secondary event in the progression of cardiac failure. Therefore, despite a large number of studies assessing oxidative stress in the heart, we still lack a comprehensive understanding of its implication in the development and progression of HF.

Implication of new regulatory mechanisms

A plethora of transcriptional and translational changes are described in cardiac hypertrophy and failure. In the past years, numerous studies have extended our knowledge about how these changes contribute to disturbed energy homeostasis and the progression of HF. Our current knowledge, however, is now challenged by the advent of studies assessing the role of posttranslational modifications of mitochondrial enzymes, such as protein phosphorylation [114], nitrosylation [115, 116], or deacetylation [117].

One of the first mitochondrial enzymes found to be regulated via phosphorylation was the pyruvate dehydrogenase complex [118]. Today, a number of mitochondrial proteins in various tissues are known to be regulated by phosphorylation, including various complexes of the ETC [119121]. While it is conceivable that these modifications do have regulatory effects on enzyme function, further studies are needed to elucidate the pathogenetic role of these changes in cardiac disease.

Besides phosphorylation, protein deacetylation has been recognized as a posttranslational regulatory mechanism. In mammals, deacetylation of lysine residues is facilitated by a class of enzymes named histone deacetylases (HDACs). Sirtuins constitute a subgroup of HDACs that target histones as well as several non-histone proteins. To date, seven individual sirtuins (SIRT1-7) are identified, with partially differing cellular localizations and functions. Their activity is implicated in the regulation of various cellular functions, including metabolism and mitochondrial homeostasis [122, 123]. The most-studied sirtuins in mammals are SIRT1 and SIRT3. The former is mainly localized in the nucleus, while SIRT 3 is primarily found in mitochondria [124]. Effects of SIRT1 activation on mitochondrial function were suggested to be due to increased mitochondrial biogenesis (via activation of PGC-1α) as well as improved resilience against oxidative stress (via increased expression of ROS-scavenging enzymes) [124]. Interestingly, while moderate overexpression of SIRT1 was found to be cardioprotective [125], high-level overexpression of SIRT1 caused dilated cardiomyopathy [126]. More recently, SIRT 3 has attracted interest of scientists. Although only a limited number of studies exist on the role of SIRT3 in the development of heart disease, it is reported to reduce oxidative stress and increase the activity of electron transport chain complexes as well as enzymes of the Krebs cycle, thereby improving mitochondrial metabolism and exerting cardioprotection [127, 128].

With increased interest in posttranslational modifications modulating mitochondrial function, further studies will not only clarify their importance in cardiac hypertrophy and failure but also help us to re-evaluate our current concepts how exactly mitochondrial function is affected in hypertrophy and HF.

Conclusions

Alterations in mitochondrial function are an established part of cardiac hypertrophy and heart failure alike (Table 1). When the heart faces increased loading conditions, ATP demand increases and changes in both the contractile apparatus and the ATP-producing apparatus occur. It is a widely accepted notion that initial changes are adaptive. During hypertrophy and heart failure, mitochondrial changes consist mainly of: (1) Altered substrate oxidation, with a decreased oxidation of fatty acids and increased reliance on glucose, (2) Reduced activity of individual complexes of the electron transport chain, impairing mitochondrial respiration, and (3) Increased production of reactive oxygen species with a higher risk of oxidative damage to mitochondrial proteins. Despite intensive research in this area, it is still unclear whether these mechanisms (and if so which) are responsible for the transition of compensated hypertrophy to heart failure. Current notions include remodeling of the mitochondrial proteome with defects in individual components of the respiratory chain, increased oxidative stress resulting in substantial mitochondrial damage, and severely impaired substrate oxidation, that limit ATP production. Owing to the large heterogeneity of mitochondrial changes found in hypertrophy and heart failure, we suggest that both entities are not associated with a distinct pattern of mitochondrial alterations. With increasing knowledge on newly identified regulatory processes on a posttranslational level, we will gain further insights into the molecular distinction of compensated and decompensated hypertrophy. This will help us to better understand the events that trigger hypertrophy and also the transition to heart failure.

Copyright information

© Springer Science+Business Media, LLC 2012