SUMMARY

1. This review is presented with the intent of illustrating therepresentative studies of functional and myocardial energeticconsequences of hearts with postinfarction left ventricular (LV)remodelling or with concentric hypertrophy and diastolic LVdysfunction in porcine models.

2. Both eccentric and concentric cardiac hypertrophy areassociated with the abnormal myocardial energetics that aremost severe in hearts with congestive heart failure (CHF).Presently, these abnormalities cannot be satisfactorily explainedto be the cause(s) of the dysfunction of failing hearts or causethe progress from compensated cardiac hypertrophy to CHF.

3. Mechanisms governing abnormal myocardial high-energyphosphate (HEP) metabolism in hearts with cardiac hyper-trophy and CHF are unclear. Myocardial energy metabolismstudies use both kinetic and thermodynamic models. Thethermodynamic studies examine the myocardial steady state levels of high- and low-energy phosphate, which indicate myocardial energy state or phosphorylation potential that isdefined by the ratio of [ATP]/([ADP][Pi]). The kinetics studiesexamine the reaction velocity that is regulated by: (i) quantityand activity of the key enzymes; (ii) the concentrations of all the substrates and products; and (iii) the Michaelis–Menten constants of each substrate of the reaction.

4. Significant alterations in myocardial concentrations ofphosphocreatine (PCr), ATP and ADP, myocardial oxidativephosphorylation (OXPHOS) protein expression and substratepreference are found in hearts with postinfarction LV remodel-ling and CHF. However, to define a causal relationship is a different matter.

5. Future studies of animal models of LV hypertrophy orheart failure using gene manipulation may provide additionalinsights to answer the persisting question of whether limitationsof ATP synthetic or transport capacities contribute to the patho-genesis of LV remodelling or failure.

Key words: creatine kinase, heart failure, hypertrophy, phos-phates, spectrscopy.

INTRODUCTION

Concentric or eccentric cardiac hypertrophy occurs as a compen-satory response to pressure overload, volume overload or post-infarction left ventricular (LV) remodelling, to restore the LV wallstresses. After a prolonged period of compensatory adaptation of cardiac hypertrophy with a restored wall stress, myocardial exhaus-tion occurs with congestive heart failure (CHF). The underlyingmechanisms of this change from compensated hypertrophy to CHFare not known. It is hypothesized that energy starvation may contribute to this disease evolution.1,2 In recent years, the cardiacapplication of magnetic resonance spectroscopy (MRS) has resultedin numerous new findings into the understanding of myocardial energetics. Alterations in myocardial high-energy phosphate (HEP)metabolism were identified in hearts with LV hypertrophy (LVH)or CHF.3–8 A decrease of the myocardial phosphocreatine (PCr)/ATPratio, which indicates an increase of [ADP] and an alteration ofmyocardial oxidative phosphorylation (OXPHOS) regulation, wasfound in patients with LVH or heart failure,3,4 as well as in large or small experimental animal models of LVH secondary to pressureoverload,5,6 volume overload7 and postinfarction LV remodelling.8–13

Despite these important findings, whether these abnormalitiescausally contribute to the evolution from compensated hypertrophyto CHF is not known. It is also not clear whether these alterationsconstrain the LV performance at basal or increased cardiac work-states. This review is presented with the intent of illustrating the representative studies of myocardial energetics in heart failure usingporcine models of cardiac hypertrophy. The discussion is focusedon the most recent data from the author’s laboratory.

MYOCARDIAL OXIDATIVE PHOSPHORYLATION REGULATION IN

NORMAL HEARTS

A major unsettled question in cardiovascular physiology is whetherthe reactions involved in ATP generation or utilization can limit themaximal performance of either normal or diseased heart and con-tribute to dysfunction of the failing heart.2–17 In the normal heart,energy demands for the contractile work are primarily met by oxida-tive synthesis of ATP. Myocardial energy demand and, consequently,the oxygen consumption rate (MVO2), increases in response toinotropic and/or chronotropic interventions or, in the intact animal,by the transition from rest to exercise. The mechanisms by whichthis MVO2 change is accomplished have been the focus of numer-ous investigations, not only because the problem is of fundamental

Experimental Biology 2001 Symposium Mitochondria and Energy Metabolism inHeart Failure, Hypertrophy and Remodeling

MYOCARDIAL ENERGETICS IN CARDIAC HYPERTROPHY

Jianyi Zhang

Department of Medicine, University of Minnesota, Minneapolis, Minnesota, USA

Correspondence: Dr Jianyi Zhang, University of MN Health ScienceCenter, Mayo Mail Code 508, 420 Delaware Street SE, Minneapolis, MN55455, USA. Email: zhang047@tc.umn.edu

Presented at the Experimental Biology Symposium Mitochondria andEnergy Metabolism in Heart Failure, Hypertrophy and Remodeling, Orlando,Florida, USA, March/April 2001.

Received 3 October 2001; accepted 26 October 2001.

Clinical and Experimental Pharmacology and Physiology (2002) 29, 351–359

352 J Zhang

importance to our understanding of aerobic metabolism in general,but also because dysfunction at the level of mitochondrial OXPHOSmay play a pathogenic role in cardiac failure.

Previous studies in this area have mainly focused on the question of ‘mitochondrial respiratory control’ in the myocardium,examining parameters such as concentrations of intracellular ADP, intracellular Pi and mitochondrial NADH, which may determine the rate of mitochondrial oxygen consumption duringalterations in cardiac workload.14–19 While this is an important ques-tion in itself, it does not necessarily address the issue of respiratory control in the whole cell or, ultimately, for the organ in vivo. In theintact system, the rate-limiting step(s) and, therefore, control ofmyocardial respiration may reside in pathways that supply the primary substrates for oxidative phosphorylation. Such rate-limiting steps can include blood flow (hence the O2 and/or carbonsubstrate delivery) to the myocardium or subsequent metabolic pathways, such as the reactions involved in mitochondrial NADHgeneration.

MYOCYTE OXYGENATION

Understanding of myocyte oxygenation is extremely important inexamining myocardial oxidative phosphorylation regulation. Mostprevious studies of cellular oxygenation in the in vivo myocardiumare based on optical techniques like infrared or ultraviolet (UV) detec-tion of myoglobin (Mb) oxygenation.20,21 Information obtained usingthese techniques suffers from the fact that the Mb signal suffers inteference from other oxygen-binding molecules, such as cyto-chrome and haemoglobin, and the sample volume is poorly defined.Measurements obtained with microelectrodes can estimate tissue oxygen level,22 but are limited by the destructive nature of the tech-nique. Although both thermodynamic and kinetic models emphasizethe regulatory roles of ADP, Pi, ATP/ADP, NADH and O2 in oxida-tive phosphorylation,19,23 without knowledge of the intracellular O2 level it is very difficult to define precisely the specific role of oxygen in the regulation of cellular energy metabolism and MVO2.In principle, deoxyMb could be used to assess mitochondrial oxygen availability.24 Gayeski and Honig have reported a minimalPO2 gradient between Mb and mitochondria.25 If this is correct, then it is reasonable to use Mb saturation as an indicator of peri-mitochondrial PO2. Jue and Kreutzer26 reported that the saturation ofMb can be assessed in the perfused rat heart using [1H]-MRS techniques. This MRS approach can sample the entire myocardialwall and evaluate the myocyte PO2 as reflected by Mb saturation,thereby providing an important tool for investigating questions relatedto oxygen limitation of the ATP synthetic process. For example, usinga decrease of MVO2 (10%) as an indicator, these investigators foundan intracellular critical PO2 of approximately 4 mmHg,26 which is much higher than the apparent Km for cytochrome oxidase (< 0.2 mmHg).27 Although this MRS approach is superior to opticaltechniques, the findings of this study are limited by the facts that: (i) the experimental system was maintained at 25°C; and (ii) the heartswere perfused with haemoglobin-free oxygenated buffer. In addition,this study used glucose as the carbon substrate, which is known to result in a lower phosphorylation potential compared with the pyruvate-perfused heart or in vivo conditions. Finally, the surgicaltrauma in the perfused heart preparation may alter contractile function and energy demand. These limitations prevent direct extra-polation of the findings to in vivo conditions.

We recently developed a [1H]-MRS technique for detecting theN-d proton signal of the proximal histidine in deoxyMb (Mb-�) inthe in vivo canine heart at 4.7 Tesla.28,29 We initially examined nineanaesthetized (pentobarbital), mechanically ventilated open-chestdogs. Left ventricular and aortic pressures and transmural myocardialblood flows were measured. The left anterior descending coronaryartery was used for partial and complete occlusions set by moni-toring intracoronary pressure distal to the occluder. A surface coil(25 mm diameter) was used for excitation and reception. A single-pulse collection sequence with a frequency selective gauss excitationpulse (1 msec) was used for selectively exciting the Mb-� resonance.This provided enough water suppression due to the large chemicalshift difference between water and Mb (> 14 kHz) and other tech-niques, such as chemical shift selective (CHESS) and inversionrecovery pulse, did not significantly improve water suppression. The nuclear magnetic resonance (NMR) signal was optimized byadjusting radio frequency (RF) pulse power using the water signalas a reference. A short repetition time (TR = 25 msec) was used dueto short spin–lattice relaxation time (T1) of Mb-�. Each spectrumwas acquired in approximately 5 min (10 000 free induction decay(FID)). Although the short T1 of Mb-� and fast acquisition preventgating to the heart beat, the signal loss from motion was negligibledue to the inherently broad line width of the Mb-� peak. Figure 1demonstrates the Mb-� proton NMR signals (around 70–71 p.p.m.from water resonance) during baseline (Fig. 1a), coronary stenosis(Fig. 1b; perfusion pressure 45 mmHg) and complete coronary occlusion (Fig. 1c). The corresponding spatially localized [31P]-NMRspectra from the same heart are illustrated in Fig. 2. The results ofthis study demonstrate that the severity of flow reduction duringischaemia is linearly related to the fraction of Mb oxygenation (r2 = 0.86).28 Although Mb-� is a temperature-sensitive NMR signal, the results show that the chemical shift of the Mb-� signal,which appeared at 71–72 p.p.m. relative to water, remained constantthroughout the study protocol. No other resonances were seen withina 10 p.p.m. region.

MYOCARDIAL HEP METABOLISM IN HEARTSWITH POSTINFARCTION LV REMODELLING

Following transmural myocardial infarction, left ventricular remodelling (LVR), including chamber dilation and hypertrophy,

Fig. 1 Typical [1H]-magnetic resonance spectra showing myocardialdeoxyMb (Mb-�) resonances (a) under baseline conditions, (b) during coro-nary stenosis, which reduced distal coronary pressure to 40 mmHg, and (c)during total left descending coronary artery occlusion. A prominent Mb-�resonance accrued at 71 p.p.m. (relative to water at 0 p.p.m) during coronary stenosis. The intensity of the Mb-� resonance is linearly relatedto the decrease of myocardial blood flow measured by radioactive micro-spheres. 2,3-DPG, 2,3-diphosphoglycerate; PCr, phosphocreatine.

Myocardial energetics in cardiac hypertrophy 353

occurs to compensate for loss of contracting myocardium. After aperiod of stable remodelling, progressive myocardial dysfunctioncan develop and may ultimately lead to overt CHF. Recently, we have described a new porcine model of postinfarction LVR.10–12

In this model, acute coronary occlusion is followed by remarkableremodelling of the non-infarcted myocardium with LV chamber dilation, reduced systolic performance, increased systolic and diastolic wall stresses10,11 and significant bioenergetic abnormalities.All these alterations are correlated with the size of the initiating

infarct. The severity of the abnormalities varies from compensatedventricular dysfunction to frank heart failure, which develops in 20–30% of animals. In normals, disaggregated myocyte length,volume and cross-sectional area were 133 � 21 �m, 21 917 �

955 �m3 and 146.3 � 4.6 �m2 (Fig. 3a), respectively; increased to 205.0 � 17.4 �m, 35 396 � 2733 �m3 and 173.9 � 5.8 �m2

in compensated remodelled hearts (Fig. 3b) and changed to 271.6 � 18.4 �m (P < 0.05 vs LVR), 42 146 � 1923 �m3 and 140.0 � 6.2 �m2 in hearts with CHF (Fig. 3c; P < 0.05 vs LVR).11

Fig. 3 Disaggregated myocytes from (a) a normal pig heart, (b) a heart with postinfarction left ventricular (LV) remodelling and (c) a heart with congestiveheart failure as evidenced by ascites of approximately 1000 mL. The LV specimens were taken 2 months after the left circumflex coronary artery (LCX)ligature in areas remote from the LV scar. The myocyte hypertrophy of LV remodelling heart is evident.

Fig. 2 Interleaved [31P]-magnetic resonance spectra obtained from the same heart as in Fig. 1, showing myocardial high-energy phosphate (HEP) and Piresonances (a) under baseline conditions, (b) during coronary stenosis, which reduced distal coronary pressure to 40 mmHg, and (c) during total left descend-ing coronary artery occlusion.

354 J Zhang

Thus, the remodelled LV demonstrated severely decreased systolicfunction, increased myocyte length and volume. Transmurally localized [31P]-MRS measurements were performed to determinePCr/ATP ratios under basal workstate conditions and biopsies were obtained to determine total creatine and ATP levels. The PCr/ATP ratio was decreased in remodelled hearts (LVR) relative to normals (subendocardical layers P < 0.05 vs normal) and further reduced when CHF was present, attaining significancein all layers (Fig. 4). Both myocardial ATP and total creatine levels were significantly decreased and calculated myocardial free ADP levels were significantly increased in LVR and CHFhearts.10,11 Coronary reserve was not significantly different betweennormal, LVR and CHF hearts when maximum myocardial perfu-sion was induced by adenosine infusion.10 Adenosine infusion did not affect basal HEP levels in any group; however, pyruvate infusion markedly improved the depressed subendocardial PCr/ATP ratio in the LVR, but not in CHF, group.10 These data document the haemodynamic, functional and basal myocardialbioenergetic perturbations present in the remodelled left ventricle.In addition, these abnormalities are related to the severity of LVdysfunction, which, in turn, is dependent upon the severity of the initiating myocardial infarction (as evidenced by scar size). These changes in myocardial HEP levels are not caused by a persistent myocardial ischaemia because hyperaemia did not correct the abnormalities. The data support the concept that thebioenergetic abnormalities may be partially explained by abnor-malities of carbohydrate metabolism, although whether these abnor-malities are associated with limitation of ATP synthetic capacity is unknown.

ROLE OF CREATINE KINASE IN MYOCARDIAL ENERGY METABOLISM

OF REMODELLED HEART

In the heart, ATP used for contractile activity is synthesized in themitochondria through oxidative phosphorylation and transported tothe contractile apparatus, where it is consumed by myosin ATPaseto generate power. The creatine kinase (CK) system plays an impor-tant role in myocardial energy metabolism by maintaining ADPlevels high at the mitochondria, where ATP is generated,30,31 and low at the contractile apparatus, where ATP is utilized.30,31 This function is thought to facilitate ATP production, transportation andutilization. However, the precise role of the CK system in myocardialenergy metabolism remains controversial. Using [31P]-NMR magnet-ization transfer techniques with a Langendorff perfused heart prepar-ation, it was found that, in the normal heart, catecholaminestimulation induced an increase in the CK flux rate parallel to theincrease of the workstate, while myocardial HEP levels did notchange.32 It was therefore concluded that the CK flux rate matchescardiac performance, but not the myocardial HEP content. However,this increase of CK flux rate in proportion to the increase of cardiacperformance has not been observed in in vivo studies of large animal models.33,34

Creatine kinase exists as a dimmer of two subunits, M and B,resulting in three isozymes: MM, MB and BB. The other two iso-forms of CK are ubiquitous and sarcomeric mitochondrial isozymesof CK (CK-Mt).2,30,31 The CK-Mt is located in the inner mito-chondrial membrane, where it is functionally coupled with adenylnucleotide translocase ANT.35 The CK-MM isoenzyme predominatesin mature myocardium and is localized at the thick filaments of the

Fig. 4 [31P]-Nuclear magnetic resonance spectra from five transmural layers across the left ventricular wall of (a) a normal heart, (b) a heart with compensated left ventricular (LV) remodelling and (c) a heart with congestive heart failure under basal conditions. Spectra were scaled to optimize visualization of the resonances, so that only the phosphocreatine (PCr)/ATP ratio can be compared. The PCr/ATP ratio was reduced in the endocardium in LV remodelled hearts and in all transmural layers in the heart with congestive heart failure. CP, creatine phosphate; Pi, inorganic phosphate.

Myocardial energetics in cardiac hypertrophy 355

contractile apparatus, where it appears functionally coupled tomyosin ATPase;2,36 CK-MB and CK-BB exist mainly in fetalmyocardium during developmental transition period. The fractionof myocardial CK-Mt in the rat is approximately 28% compared withonly 10% in large animal or human hearts.2 A ‘PCr shuttle’ hypo-thesis was proposed that emphasizes the higher diffusion capacitybetween mitochondria and contractile apparatus of creatine (Cr)compared with ADP.30,31 If the ‘PCr shuttle’ hypothesis is correct,then alterations of CK-Mt could be the ‘weak link’ of the shuttlebecause of its low fraction. In a recent study using this animal modelof LVR, we found that the mRNA and protein levels of CK-Mtdecreased by 46 and 53%, respectively (each P < 0.05 compared withthe control group).37 A decrease in mitochondrial CK would requirehigher cytosolic ADP levels to support a given rate of ATP synthesisand may limit the maximal rate of ATP synthesis. In hearts with compensated postinfarction LVR, the creatine kinase forward rateconstant (kf) measured with magnetization transfer (0.38 � 0.04 /s)was not different from normal (0.41 � 0.03 /s). However, in animals

with overt CHF, the creatine kinase forward rate constant (0.21 �

0.03 /s) was decreased by approximately 50%.12 Chemically deter-mined ATP was normal in hearts with compensated remodelling, but was decreased in the presence of CHF. The PCr/ATP ratios were significantly decreased in the setting of CHF and were alsodecreased in the subendocardium of hearts with compensated post-infarction remodelling. The CK forward flux rate (fluxf) was 20.3 � 2.4 �mol/g per s in normal hearts; this rate was decreasedto 14.3 � 2.1 �mol/g per s in hearts with compensated postinfarctremodelling and further decreased to 6.4 � 2.3 �mol/g per s in thesetting of CHF. Hearts with compensated postinfarction remodellingresponded normally to inotropic stimulation with dobutamine, indicating that the decreased CK forward flux rate did not limit thecontractile response to inotropic stimulation. In the setting of end-stage CHF, dobutamine produced essentially no inotropic response.12

In normal myocardium, the rate of phosphoryl exchange betweenPCr and ATP is more than an order of magnitude higher than theATP utilization rate, so that the shuttle hypothesis does not requirean increase in the CK flux rate to accommodate physiologicalincreases in myocardial workload. The ratio of the CK flux rate to the ATP synthesis rate (assuming P:0 = 3.0) was calculated toexamine whether the rate of ATP utilization would approach the rateof flux through the CK reaction during catecholamine stimulation.12

The ratios of CK flux to the ATP synthesis rate (from MVO2) underbaseline conditions were 10.9 � 1.2, 8.03 � 0.9 and 3.86 � 0.5 fornormal, LVR and CHF hearts, respectively (each P < 0.05).12

During infusion of dobutamine, this ratio decreased to 3.7 and 2.6in the normal and compensated remodelled groups, respectively.Interestingly, during high workstates in both normal and LVR hearts,this ratio decreased (as a consequence of the unchanged CK fluxrate but an increased ATP utilization rate) to a level similar to thatof the CHF hearts during baseline conditions (2.8; Fig. 5).12 It is possible that this ratio of the CK flux rate to the rate of ATP syn-thesis reflects a minimum value to maintain optimal cross-bridgefunction.36 In previous studies, decreased contractile reserve wasobserved in hearts in which CK activity was suppressed bysulfhydryl inhibition,38 by guanidino substrate replacement39 and byCK-M subunit gene knock-out.40 These data support the concept thata decrease of the CK flux rate may contribute to decreased contractilereserve in the failing heart.

MYOCARDIAL ENERGETICS AT HIGH CARDIAC WORKSTATES

In hearts with cardiac hypertrophy, at high workstates induced by catecholamine stimulation, further HEP loss occurs,17 which

Fig. 6 Representative western blotsshow the �-subunit F1-ATPase proteinmigrated to the position correspondingto approximately 50 kDa on a 12%sodium dodecyl sulfate–polyacry-lamide gel. The numbers 1, 2, 3 and 4 represent individual animals in therespective groups. The densitometry ofautoradiograms showed that the proteinlevels in failing hearts decreased significantly in failing hearts. LVR, left ventricular remodelling; CHF, congestive heart failure.

Fig. 5 The ratio of creatine kinase (CK) flux : ATP utilization from the threegroups under baseline conditions and during dobutamine stimulation. *P < 0.05 compared with normal (�); †P < 0.05 compared with left ven-tricular remodelling (LVR; ); ‡P < 0.05 compared with baseline. (�), con-gestive heart failure (CHF). In normal hearts and hearts with LVR, theincreased oxygen consumption during dobutamine infusion with no changein phosphocreatine (PCr) flux rate caused a decrease in the ratio of PCr flux : ATP synthesis to approximately 1 : 3. In hearts with CHF, this ratiowas approximately one-third of normal hearts at baseline and did not changesignificantly in response to dobutamine stimulation. HWL, high cardiacworkload.

356 J Zhang

suggests a ‘demand-induced ischaemia’. Applying this [1H]-MRStechnique and using the swine model, we tested the hypothesis that inadequate myocyte oxygen availability is the basis for theseHEP abnormalities.11 Myocardial infarction was produced by leftcircumflex coronary artery ligation in domestic swine. Studies wereperformed in 20 normal animals, 14 animals with compensated LVRand nine animals with LVR that developed CHF. Phosphocreatine(PCr)/ATP ratios were determined with [31P]-NMR spectroscopy anddeoxyMb (Mb-�) was determined with [1H]-NMR spectroscopy inmyocardium remote from the infarct. Basal PCr/ATP ratios tendedto be lower in postinfarct hearts than in normal hearts and this difference was significant in animals that developed CHF. Infusionof dobutamine (20 �g/kg/min, i.v.) caused doubling of the rate–pressure product (RPP) in both normal and LVR hearts and resultedin comparable significant decreases of PCr/ATP in both groups. This decrease in PCr/ATP was not associated with detectable Mb-�.11 In CHF hearts, RPP increased only 40% (RPP reached only 15 000 mmHg/min) in response to dobutamine; this attenuatedresponse was also not associated with detectable Mb-�. The dataindicate that the decrease of PCr/ATP in normal and LVR hearts during dobutamine infusion is not the result of insufficient myo-cardial oxygen availability. Furthermore, in CHF hearts, the lowbasal PCr/ATP and the attenuated response to dobutamine occurredin the absence of myocardial hypoxia, indicating that the HEP andcontractile abnormalities were not the result of insufficient oxygenavailability.11 However, because the decompensated remodelledhearts were intolerant of rapid pacing (a quick drop of mean aorticpressure to below 45 mmHg) and have a markedly blunted con-tractile response to catecholamine infusion, the hypothesis that thefailing hearts exposed to myocardial ischaemia at increased cardiacworkstates was not examined in this study.11 This question wasexamined more recently as described in the next experiment.

ALTERED MYOCARDIAL OXIDATIVE PHOSPHORYLATION REGULATION AND

MITOCHONDRIAL ATPASE PROTEINEXPRESSION IN FAILING HEARTS

In a more resent study of this animal model of postinfarction LVR,we examined the relationships between altered myocardial OXPHOSand mitochondrial ATPase gene expression in hearts with compen-sated postinfarction LVR and hearts with CHF. Seventeen pigs werestudied 8 weeks after myocardial infarction produced by left circumflex coronary artery ligation (nine with LVR and eight with CHF) and the results were compared to those obtained in sevennormal pigs (N).41 The myocardial HEP levels ([31P]-MRS) weremeasured under basal conditions (B) and during moderate increasedcardiac workstates (ICW). The ICW was aimed to reach a level ofRPP = 20 000 mmHg/min by dobutamine and dopamine infusion(2.5–10 �g/kg per min, i.v.) and rapid pacing (to 200 b.p.m. asneeded). The mitochondrial protein levels of F0F1-ATPase subunitswere examined by western blots. Under basal conditions, PCr/ATPratios for N, LVR and CHF groups, were 2.23 � 0.09, 1.94 � 0.08and 1.43 � 0.07 (P < 0.05 vs N or LVR), respectively.41 The myo-cardial Pi was undetectable in each group at B. During ICW, myo-cardial HEP did not change significantly in normal hearts and heartswith LVR. However, in hearts with CHF, PCr/ATP decreased to 1.17 � 0.07 (P < 0.01 vs baseline) and Pi/PCr increased to 0.32 �

0.04 (P < 0.01). In hearts with CHF, the mitochondrial �-subunit

F1-ATPase decreased by 36% (P < 0.05), the �-F1-ATPase decreasedby 16% (P < 0.05) and the oligomycin sensitivity conferring protein(OSCP) subunit decreased by 40% (P < 0.01) (Figs 6,7). In heartswith compensated LVR, the subunits of F0F1-ATPase did not changesignificantly. The experimental intervention induced a moderateincrease of RPP to approximately 20 000 mmHg·b.p.m. At this levelof cardiac workstates, it is well known that normal hearts maintainconstant myocardial HEP levels.14–16 Indeed, in this study, the normal hearts and hearts with compensated LVR maintained normal myocardial HEP and Pi levels. However, in animals withend-stage CHF, a significant further decrease of myocardial PCr/ATPratios and increase of free ADP were observed.41 These changesoccurred in the presence of a normal increase of myocardial bloodflow and unchanged myocardial oxygenation level (from Mb-� data).Thus, at a moderate increase of cardiac workstates, the occurrenceof these ‘metabolic ischaemic markers’, observed in the failinghearts, is not caused by myocardial ischaemia. This finding is inagreement with the previous observations that changes in myocardialOXPHOS at high cardiac workstates are independent from myo-cardial ischaemia in hearts with compensated cardiac hyper-trophy.11,42 Taken together, these data indicate that the alterations inmyocardial OXPHOS in the failing hearts are independent frommyocardial ischaemia.

HIGH ENERGY PHOSPHATE METABOLISM IN HEARTS WITH SEVERE CONCENTRIC

HYPERTROPHY AND DIASTOLIC DYSFUNCTION

A concentric LVH secondary to severe pressure overload oftenresults in left ventricular diastolic dysfunction and failure, whichremains difficult to manage clinically. Recently, a new study wascompleted that aimed to create a pig model of diastolic heart failuresecondary to severe aortic stenosis and to examine the relationshipsbetween LV function, alterations in myocardial HEP metabolism and protein expression of CK isoforms.6 Sixteen pigs with LVH

Fig. 7 Densitometric intensities for protein bands from western blots of�-, �-, oligomycin sensitivity conferring protein (OSCP)- and inhibition factor 1 (IF1)-ATPase subunits normalized to �-actin. Values are themean�SEM. (�), normal (n = 7); ( ), left ventricular remodelling (LVR;n = 8); (�), congestive heart failure (CHF; n = 8). *P < 0.05, **P < 0.01 compared with normal. The levels of �-, �-, OSCP- and IF1-mtATPase subunits are decreased significantly only in hearts with end-stage CHF.

Myocardial energetics in cardiac hypertrophy 357

secondary to ascending aortic banding that produced a pressure gradient of 70 mmHg across the narrowing and 10 normal pigs (N)were studied. Myocardial protein levels of CK isoforms (westernblot), HEP levels and CK kinetics ([31P]-MRS) were measured under asal conditions. Nine of 16 animals with LVH developed CHF, asevidenced by ascites (100–2000 mL). The LV weight/bodyweight(g/kg) ratio was 2.18 � 0.15 in normal, 3.04 � 0.14 in hearts with LVH (P < 0.01 vs normal) and 4.23 � 0.36 in hearts with CHF(P < 0.01 vs LVH). Right ventricle weight/bodyweight ratio and LVend-diastolic pressure were significantly higher in hearts with CHF(each P < 0.01). Myocardial PCr/ATP ratios and the CK forward flux rates were decreased in LVH hearts, which was most severe inhearts with CHF. The CK-M/�-actin ratios were 2.21 � 12 (normal),1.69 � 0.15 (LVH) and 1.39 � 0.27 (CHF; P < 0.05 vs normal). TheCK-mitochondria (CK-Mt)/�-actin ratios were 1.40 � 0.09 (normal),1.24 � 0.09 (LVH) and 1.02 � 0.08 (CHF; P < 0.05 vs normal orLVH). The severity of the reduction of CK flux rate was linearlyrelated to the severity of the decrease of CK-Mt/�-actin (r = 0.68;P < 0.01).6 Thus, in this new model of diastolic heart failure, theabnormal myocardial HEP metabolism is related to the decreasedCK-Mt protein level, which, in turn, is related to the severity of thehypertrophy.

Although myocardial HEP metabolism is abnormal in failinghearts (CHF), it is unclear whether this is associated with impairedoxidative capacity. A new study was performed to test this hypo-thesis using this severe concentric LVH secondary to aortic banding. The mitochondrial uncoupling agent 2,4-dinitrophenol(DNP), which accelerates intramitochondrial metabolism proximalto ATP synthase, was administered to hypertrophied and failinghearts, while the relationship between HEP levels and MVO2 wasused to characterize the characteristics of OXPHOS regulation.Measurements were obtained during basal conditions (B), duringcombined dobutamine and dopamine infusion (each 20 �g/kg permin, i.v.) and during continuing catecholamine infusion with theaddition of four graded doses of DNP (2–8 mg/kg, i.v.). The myo-cardial PCr/ATP ratio was significantly decreased in LVH hearts,which was most severe in hearts with CHF. In CHF hearts, myo-cardial concentrations of ATP and total creatine were significantlydecreased and the calculated free ADP concentration significantlyincreased. Interestingly, the myocardial oxygen consumption rate,which reflects the OXPHOS capacity, increased progressively andsimilarly in response to DNP in all groups. Thus, both hypertrophiedand failing ventricles required higher ADP levels to maintain ATPsynthesis, but oxidative capacity was not reduced in failing hearts.

THE INCREASE OF MYOCARDIAL FREE ADP AND REDUCTION OF ATP CONTENTS

The increase of myocardial free ADP would initiate adenylate kinase(myokinase) activation, which catalyses the transfer of a phosphorylgroup between two ADP to form one AMP and one ATP.43,44 Theincreased AMP would induce the conversion of AMP to adenosine.45

Adenosine can cross the cell membrane to get into the interstitialspace, where it is further degraded to inosine, hypoxanthine andleaves the heart with the myocardial blood flow.46 This loss of thetotal adenine nucleotide pool (TAN), results in the reduction of ATPas the resynthesis of adenine nucleotide is a slow and energy costlyprocess through de novo synthesis, where inosine monophosphate(IMP) is produced from ribose-5-phosphate that uses six HEPbonds.47 It has been reported that in the ‘stunned’ myocardium the

loss of ATP during acute myocardial ischaemia requires several days to recover, which is caused by the slow de novo resynthesis ofadenine nucleotide.47,48 Taken together, these data indicate that thefailing hearts are associated with the alterations in mitochondrialOXPHOS with a manifestation of increase of cytosolic free ADP.This abnormality could initiate reactions that result in the reductionof myocardial purines and ATP concentration.

ALTERATIONS IN MYOCARDIALBIOENERGETICS AND LV DYSFUNCTION

How this reduction of myocardial ATP concentration contributes tothe contractile performance of the failing hearts is not known. It isnot likely, however, that this approximate 30% decrease of myo-cardial ATP concentration would cause the dysfunction of failinghearts. The fact that myocardial ATP concentration is many foldgreater than the Km values of either the ATPase at the contractileapparatus or sarcoplasmic reticulum calcium ATPase (SERCA)49

makes it unlikely that this reduction would directly depress the contractile performance of the left ventricle.

As a result of the decrease of the myocardial ATP/ADP ratio,myocardial free energy release per unit ATP hydrolysis (�G) wouldbe significantly reduced.2,38 A reduced �G has been shown pre-viously to be related to the decreased LV contractile performance.38

Perhaps, it is the decreased energy state (as indicated by a decreaseof ATP/ADP ratio) that contributes to the opening of mitochondrialpermeable transition pore (mtPTP) and initiates apoptosis,50,51 whichthen contributes to the LV dysfunction. Recently, there is an increas-ing volume of evidence indicating that alterations in mitochondrialfunction may play a central role in triggering apoptosis in failinghearts.50–52 A reduction of mtATPase can also induce the accumu-lation of reactive oxygen species (ROS), which opens the mtPTPand triggers myocardial apoptosis.50 The structure of mtPTP is onlypartially defined. It is believed to include the adenine nucleotidetranslocator (ANT) in the inner mitochondrial membrane. The ANTis coupled with mitochondrial CK (CK-mt) and mtATPase. In thepig model of postinfarction LVR, mtATPase, ANT and CK-mt wereall found to be decreased significantly.37,41,53 Whether the alterationsof these three terminal enzymes of the mitochondrial OXPHOSresult in apoptosis warrant future studies. It should be noted thatthis understanding of the signalling pathways that link the mito-chondria dysfunction to programmed cell death is based on culturedneonatal myocyte experiments. This remains to be examined in in vivo hearts. The data of recent studies demonstrate that failinghearts are associated with decreases in the steady state level of mito-chondrial ATPase protein subunits. The abnormalities are associatedwith an increased myocardial free ADP level, which increase thedriving force of mitochondrial OXPHOS. This increase of myo-cardial free ADP level is more severe during increased cardiac workstates. These changes are not caused by myocardial ischaemia.We speculate that the increased myocardial free ADP level may alsoactivate adenylate kinase and initiate reactions that ultimately lead to the loss of purine pools and, therefore, the reduction ofmyocardial ATP concentration.54

ACKNOWLEDGEMENTS

This work was supported by US Public Health Service GrantsHL50470, HL61353 and an Established Investigator Award from theAmerican Heart Association.

358 J Zhang

REFERENCES

1. Katz AM. Cardiomyopathy of overload. A major determinant of prognosis in congestive heart failure. N. Engl. J. Med. 1989; 322:100–10.

2. Ingwall JS. Is cardiac failure a consequence of decreased energy reserve?Circulation 1993; 87 (Suppl. VII): VII-58–62.

3. Lamb HJ, Beyerbacht HP, van der Laarse A et al. Diastolic dysfunctionin hypertensive heart disease is associated with altered myocardialmetabolism. Circulation 1999; 99: 2261–7.

4. Neubauer S, Horn M, Godde M, Ertl G, Ingwall JS. The myocardialphosphocreatine/ATP ratio is a predictor of mortality in patients withdilated cardiomyopathy. Circulation 1997; 96: 2190–6.

5. Zhang J, Merkle H, From AHL, Ugurbil K, Bache RJ. Bioenergeticabnormalities associated with severe left ventricular hypertrophy. J. Clin.Invest. 1993; 92: 993–1003.

6. Ye Y, Gong G, Ochiai K, Liu J, Zhang J. High-energy phosphatemetabolism and creatine kinase in failing hearts. Circulation 2001; 103:1570–6.

7. Zhang J, Toher C, Zhang Y, Ugurbil K, Bache RJ, Homans D.Bioenergetic consequences of volume overloaded hypertrophy.Circulation 1997; 96: 334–43.

8. Zhang J, McDonald K. Bioenergetic consequence of left ventricularremodeling secondary to discrete myocardial infarction. Circulation1995; 92: 1011–19.

9. Neubauer S, Horn M, Naumann A et al. Impairment of energymetabolism in intact residual myocardium of rat hearts with chronicmyocardial infarction. J. Clin. Invest. 1995; 95: 1092–100.

10. Zhang J, Wilke N, Wang Y et al. Functional and bioenergetic con-sequences of post-infarction left ventricular remodeling in a new porcinemodel: An MRI 31P MRS study. Circulation 1996; 94: 1089–99.

11. Murakami Y, Zhang Y, Cho YK et al. Myocardial oxygenation duringhigh workstates in hearts with post-infarction remodeling. Circulation1999; 99: 942–8.

12. Murakami Y, Zhang J, Eijgelshoven MHJ et al. Myocardial creatinekinase kinetics in hearts with postinfarction left ventricular remodeling.Am. J. Physiol. Heart Circ. Physiol. 1999; 276: H892–900.

13. Zhang J, Murakami Y, Zhang Y et al. Oxygen delivery does not limitcardiac performance during high work states. Am. J. Physiol. Heart Circ.Physiol. 1999; 276: H50–7.

14. Balaban RS, Kantor HL, Katz LA, Briggs RW. Relation between workand phosphate metabolites in the in vivo paced mammalian heart.Science 1986; 232: 1121–3.

15. Zhang J, Duncker DJ, Xu Y et al. Bioenergetic responses of normalmyocardium at very high workstates: An in vivo transmural 31P NMRstudy. Am. J. Physiol. 1995; 268: H1891–905.

16. Massie BM, Schwartz GG, Carcia J, Wisneski JA, Weiner MW, TwymanO. Myocardial metabolism during increased work states in the porcineleft ventricle in vivo. Circ. Res. 1994; 74: 64–73.

17. Massie BM, Schaefer S, Garcia J et al. Myocardial high-energy phos-phate and substrate metabolism in swine with moderate left ventricularhypertrophy. Circulation 1995; 91: 1814–23.

18. Jacobus WE, Moreadith RW, Vandegaer KM. Mitochondrial respiratorycontrol. J. Biol. Chem. 1982; 257: 2397–402.

19. From AHL, Zimmer SD, Michurski SP et al. Regulation of the oxid-ative phosphorylation rate in the intact cell. Biochemistry 1990; 29:3731–43.

20. Heineman FW, Balaban RS. Effects of afterload and heart rate onNAD(P)H redox state in isolated rabbit heart. Am. J. Physiol. 1993; 264:H433–40.

21. Parsons WJ, Rembert JC, Bauman RP, Greenfield JC, Plantadosi CA.Dynamic mechanisms of cardiac oxygenation during brief ischemia andreperfusion. Am. J. Physiol. 1990; 259: H1477–85.

22. Whalen WJ. Intracellular PO2 in heart and skeletal muscle. Physiologist1971; 14: 69–82.

23. Chance B, Leigh JS, Kent J et al. Multiple controls of oxidativemetabolism in living tissues as studied by phosphorus magnetic reso-nance. Proc. Natl Acad. Sci. USA 1986; 83: 9458–62.

24. Antonini E, Brunori M. Hemoglobin and myoglobin in their reactionswith ligands. North-Holland Publications, Amsterdam. 1971.

25. Gayeski TEJ, Honig CR. Intracellular PO2 in individual cardiomyocytesin dogs, cats, rabbits, ferrets and rats. Am. J. Physiol. 1991; 260:H522–31.

26. Kreutzer U, Jue T. 1H-Nuclear magnetic resonance deoxymyoglobin signal as indicator of intracellular oxygenation in myocardium. Am. J. Physiol. 1991; 261: H2091–7.

27. Oshino N, Jamieson D, Sugano T, Chance B. Mitochondrial functionunder hypoxic conditions: The steady states of cytochrome aa3 and theirrelation to mitochondrial energy states. Biochem. Biophys. Acta 1974;368: 298–310.

28. Chen W, Zhang J, Eljgelshoven MHJ et al. Determination of deoxymyo-globin changes during graded myocardial ischemia: An in vivo 1H NMRspectroscopy study. Magn. Reson. Med. 1997; 38: 193–7.

29. Zhang J, Ugurbil K, From AHL, Bache RJ. Myocardial oxygenationand high-energy phosphate levels during graded coronary hypoper-fusion. Am. J. Physiol. Heart Circ. Physiol. 2001; 280: H318–26.

30. Wyss M, Smeitink J, Wevers RA, Wallimann T. Mitochondrial creatinekinase: A key enzyme of aerobic energy metabolism. Biochem.Biophysica Acta 1992; 1102: 119–66.

31. Bessman SP, Carpenter CL. The creatine–creatine phosphate energyshuttle. Annu. Rev. Biochim. 1985; 54: 831–62.

32. Bittle J, Balschi J, Ingwall JS. Effects of norepinephrine infusion onmyocardial high-energy phosphate content and turnover in the livingrat. J. Clin. Invest. 1987; 79: 1852–9.

33. Portman MA, Ning XH. Maturational changes in respiratory controlthrough creatine kinase in heart in vivo. Am. J. Physiol. 1992; 263:C453–60.

34. Katz LA, Swain JA, Portman MA, Balaban BS. Relation between phosphate metabolites and oxygen consumption of heart in vivo. Am. J. Physiol. 1989; 256: H256–74.

35. Khuchua ZA, Qin W, Boero J et al. Octamer formation and coupling of cardiac sarcomeric mitochondrial creatine kinase are mediated by charged N-terminal residues. J. Biol. Chem. 1998; 273: 22 990–6.

36. Sata M, Sugiura S, Yamashita H, Momomura S-I, Serizawa T. Coupling between myosin ATPase cycle and creatine kinase cycle facilitates cardiac actomyosin sliding in vitro. Circulation 1996; 93:310–17.

37. Hoang CD, Zhang J, Payne RM, Apple F. Postinfarction left ventric-ular remodeling induces changes in creatine kinase mRNA and protein subunit levels in porcine myocardium. Am. J. Pathol. 1997; 151:257–64.

38. Tian R, Ingwall JS. Energetic basis for reduced contractile reserve inisolated rat hearts. Am. J. Physiol. Heart Circ. Physiol. 1996; 270:H1207–16.

39. Kapelko VI, Kupriyanov VV, Novikova NA et al. The cardiac contractilefailure induced by chronic creatine and phosphocreatine deficiency. J. Mol. Cell. Cardiol. 1988; 20: 465–79.

40. Saupe KW, Spindler M, Tian R, Ingwall JS. Impaired cardiac energeticsin mice lacking muscle-specific isoenzymes of creatine kinase. Circ.Res. 1998; 82: 898–907.

41. Liu J, Wang C, Murakami Y et al. The mitochondrial ATPase and high energy phosphates in failing hearts. Am. J. Physiol. 2001; 281:H1319–26.

42. Bache RJ, Zhang J, Murakami Y et al. Myocardial oxygenation at highworkstates in hearts with left ventricular hypertrophy. Cardiovasc. Res.1999; 42: 616–26.

43. Kroll K, Kinzie DJ, Gustafson LA. Open-system kinetics of myocardialphosphoenergetics during coronary underperfusion. Am. J. Physiol.1997; 272: H2563–7.

44. Gustafson LA, Kroll K. Downregulation of 5�-nucleotidase in rabbitheart during coronary underperfusion. Am. J. Physiol. 1998; 274:H529–38.

45. Skladanowski AC, Newby AC. Partial purification and properties of an AMP-specific soluble 5�-nucleotidase from pigeon heart. Biochem.J. 1990; 268: 117–22.

Myocardial energetics in cardiac hypertrophy 359

46. Imai S, Riley AL, Berne RM. Effect of ischemia on adenine nucleotidesin cardiac and skeletal muscle. Circ. Res. 1964; 15: 443–50.

47. Manfredi JP, Holmes EW. Purine salvage pathways in myocardium.Annu. Rev. Physiol. 1985; 47: 691–705.

48. Zhang J, Ugurbil K, From AHL, Bache RJ. High energy phosphatemetabolism in normal and abnormal myocardium: Evaluation with in vivo magnetic resonance spectroscopy. J. Cardiovasc. Magn. Reson.2000; 2: 23–32.

49. Davies CH, Harding SE, Poole-Wilson PA. Cellular mechanisms of contractile dysfunction in human heart failure. Eur. Heart J. 1996; 17:189–98.

50. Green DR, Reed JC. Mitochondria and apoptosis. Science 1998; 281:1309–12.

51. Liu X, Kim CN, Yang J, Jemmerson R, Wang X. Induction of apoptoticprogram in cell-free extracts: Requirement for dATP and cytochromec. Cell 1996; 86: 147–57.

52. Wallace DC. Mitochondrial diseases in man and mouse. Science 1999;283: 1482–8.

53. Ning X-H, Zhang J, Liu J et al. Signaling and expression for mito-chondrial membrane proteins during left ventricular remodeling and contractile failure after myocardial infarction. J. Am. Coll. Cardiol. 2000; 36: 282–7.

54. Shen W, Asai K, Uechi M et al. Progressive loss of myocardial ATPdue to a loss of total purines during the development of heart failure indogs. Circulation 1999; 100: 2113–28.