Regulation of Glycolytic Flux in Ischemic PreconditioningA STUDY EMPLOYING METABOLIC CONTROL ANALYSIS*

Received for publication, February 4, 2002, and in revised form, May 1, 2002Published, JBC Papers in Press, May 2, 2002, DOI 10.1074/jbc.M201138200

Achim M. Vogt‡§, Mark Poolman¶, Cordula Ackermann‡, Murat Yildiz‡, Wolfgang Schoels‡,David A. Fell¶�, and Wolfgang Kubler‡

From the ‡Medizinische Universitatsklinik (Ludolf-Krehl-Klinik), Abteilung Innere Medizin III (Schwerpunkt Kardiologie,Angiologie und Pulmologie), Bergheimer Straße 58, D-69115 Heidelberg, Germany and the ¶School of Molecular andBiological Sciences, Oxford Brookes University, Headington, Oxford OX3 0BP, United Kingdom

Exact adjustment of the Embden-Meyerhof pathway(EMP) is an important issue in ischemic preconditioning(IP) because an attenuated ischemic lactate accumula-tion contributes to myocardial protection. However,precise mechanisms of glycolytic flux and its regulationin IP remain to be elucidated. In open chest pigs, IP wasachieved by two cycles of 10-min coronary artery occlu-sion and 30-min reperfusion prior to a 45-min indexischemia and 120-min reperfusion. Myocardial contentsin glycolytic intermediates were assessed by high per-formance liquid chromatographic analysis of serialmyocardial biopsies under control conditions and IP.Detailed time courses of metabolite contents allow anin-depth description of EMP regulation during indexischemia using metabolic control analysis. IP reducedmyocardial infarct size (control, 90.0 � 3.1 versus 5.05 �2.1%; p < 0.001) and attenuated myocardial lactate accu-mulation (end-ischemic contents, 31.9 � 4.47 versus10.3 � 1.26 �mol/wet weight, p < 0.0001), whereby adecrease in anaerobic glycolytic flux by at least 70%could constantly be observed throughout index ische-mia. By calculation of flux:metabolite co-responses, themechanisms of glycolytic regulation were investigated.The continuous deceleration of EMP flux in control myo-cadium could neither be explained on the basis of sub-strate availability nor be attributed to regulatory “keyenzymes,” as multisite regulation was employed for fluxadjustment. In myocardium subjected to IP, an evenpronounced deceleration of EMP flux during index is-chemia was observed. Again, the adjustment of EMP fluxwas because of multisite modulation without any evi-dence for flux limitation by substrate availability or akey enzyme. However, IP changed the regulatory prop-erties of most EMP enzymes, and some of these patternscould not be explained on the basis of substrate kinetics.Instead, other regulatory mechanisms, which have pre-viously not yet been described for EMP enzymes, mustbe considered. These altered biochemical properties ofthe EMP enzymes have not yet been described.

Myocardial survival in states of supply/demand imbalancecritically depends on cellular energy status (1). To limit energydeficit in conditions of energy shortage, e.g. hypoxia and ische-mia, myocardial energy production switches from the preferen-tial use of fatty acids to carbohydrates, thereby allowing main-tenance of adequate ATP synthesis when decreased oxygenavailability becomes the limiting factor (2–5). However, in zero-flow ischemia, experimental analysis has shown increased gly-colysis to be a double-edged sword, as the accumulation ofglycolytic end products (6, 7) outweighs the potential benefits ofincreased ATP synthesis.

In accordance with this paradigm, myocardial protection byprior exposure to ischemic preconditioning (IP)1 employs alimitation in myocardial energy deficit (8, 9). Because IP tre-mendously reduces ischemic myocardial energy demands (8),energy deficit is largely decreased at even reduced rates ofanaerobic glycolytic ATP formation in zero-flow ischemia.There is good evidence that this attenuation in ischemic lactateaccumulation represents an important mechanism whereby IPmyocardium better withstands the challenges of sustained is-chemia. Although a decrease in anaerobic glycolytic flux is aconsistent finding in myocardium protected by ischemic pre-conditioning (10–12), the precise mechanism used to adjustanaerobic glycolytic flux still remains unclear. The elucidationof this adaptive mechanism, which is not merely of theoreticalinterest, was the aim of our study.

To analyze the regulation of metabolic pathways, traditionalconcepts mostly imply that control over a pathway is achievedby action on a single pathway enzyme, which is often assumedto be a nonequilibrium step near the beginning of the pathwaysubjected to feedback inhibition (13, 14). However, because therise of metabolic control analysis (MCA) (15) there is growingevidence that these comfortable, time honored concepts maymislead more than enlighten. Hence, the tools of MCA wereused to obtain a state-of-the-art analysis of glycolytic regula-tion in ischemic myocardium protected by preceding ischemicpreconditioning.

MATERIALS AND METHODS

The experimental protocol described in this study was approved bythe Bioethical Committee of the District of Karlsruhe, Germany. Allanimals in this study were handled in accordance with the guidingprinciples in care and use of animals as approved by the American

* This work was supported in part by a grant from the DeutscheForschungsgemeinschaft, Bonn, Germany, within the Sonderforschun-gsbereich 320, “Herzfunktion und ihre Regulation,” and the Universityof Heidelberg (Teilprojekt C14), Germany (to A. V.). The costs of pub-lication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked “advertisement”in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

§ To whom correspondence should be addressed. Tel.: 49-6221-56-8611; Fax: 49-6221-56-5515; E-mail: Achim_Vogt@med.uni-heidelberg.de.

� Supported by Wellcome Trust, UK, Showcase awards 048728 and056275.

1 The abbreviations used are: IP, ischemic preconditioning; MCA,metabolic control analysis; LAD, left anterior descending coronary ar-tery; 2-PG, 2-phosphoglycerate; 3-PG, phosphoglycerate; DHAP, dihy-droxyacetone phosphate; PGM, phosphoglyceromutase; GAPDH, glu-ceraldehyde-3-phosphate dehydrogenase; ANOVA, analysis of variance;PFK-1, phosphofructokinase-1; LDH, lactate dehydrogenase.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 277, No. 27, Issue of July 5, pp. 24411–24419, 2002© 2002 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

This paper is available on line at http://www.jbc.org 24411

at NY

U School of M

edicine Library on February 6, 2015

http://ww

w.jbc.org/

Dow

nloaded from

Physiological Society and the investigation conformed with the Guidefor Care and Use of Laboratory Animals, United States National Insti-tutes of Health.

Animal Preparation—12 castrated German domestic pigs with bodyweights between 25.0 and 32.5 kg (27.8 � 0.32 kg) were used in thisstudy employing an established model of coronary occlusion and reper-fusion as described in detail previously (16).

Experimental Groups—The experimental protocol is shown in Fig. 1.In all animals, myocardial ischemia causing infarction was induced bya 45-min LAD occlusion (index ischemia). The animals of the controlgroup (C, n � 6) were only subjected to index ischemia followed by 120min reperfusion. In the IP group, the animals were subjected to twoischemic episodes (10 min each, with 30 min reperfusion between)before the onset of the 45-min index ischemia (IP long; n � 6).

Myocardial drill biopsies were taken at the end of the stabilizationperiod, i.e. before any experimental intervention, from virgin, non-ischemic myocardium, and directly before occluding the LAD at theonset of index ischemia. During the 45-min LAD occlusion, biopsieswere taken from the center of the ischemic area after 2, 5, 10, 20, and45 min.

Determination of the Infarcted Area and Quantification of Myocar-dial Protection—Determination of myocardial infarct size was deter-mined according to standard methods by identifying ischemic myocar-dium using fluorescent microspheres and detecting infarctedmyocardium following incubation in triphenyltetrazolium chloride.Quantification was performed by normalizing infarcted myocardialmass (infarcted region) to left ventricular myocardium or to the mass ofischemic myocardium (risk region). These ratios are given in percent.

Biopsies and Metabolite Analysis—Left ventricular drill biopsies(�20 mg each) were taken at various time points (Fig. 1). The biopsieswere immediately frozen (within 5 s) and kept in liquid nitrogen untilfurther use. Homogenization and deproteinization were performed inice-cold 60% acetonitrile (v/v) using a Branson sonifier. Using twodifferent high performance liquid chromatography protocols, the myo-cardial contents in ATP, ADP, and AMP, and the intermediary metab-olites and end products of glycolysis were determined (17, 18). Theinjury caused by the biopsies did not interfere with the triphenyltetra-zolium chloride stainings. Myocardial glycogen contents was assessedaccording to Ref. 19.

Data Analysis—Dihydroxyacetone phosphate (DHAP) and 1,3-bisphosphoglycerate could not be detected. The DHAP content was,therefore, calculated from the glyceraldehyde 3-phosphate content as-suming the equilibrium state of the triose-phosphate isomerase reac-tion (20). Two- and 3-phosphoglycerate (2-PG and 3-PG, respectively)were not always clearly separated in the chromatograms, although itscommon peak could be easily detected. The individual proportions werecalculated according to the equilibrium of the phosphoglyceromutase(PGM) reaction as shown.

0.154 ��2-PG][3-PG]

(Eq. 1)

The equilibrium of isomerase and phosphoglyceromutase are knownto be unaffected by ischemia (13). Free myocardial ADP content wascalculated from the myocardial contents of ATP and AMP, assuming anequilibrium state of the adenylate kinase reaction (21). Glycolytic flux(J) (in micromole of C6-units/g wet weight/min) during ischemia was

calculated from the accumulation of lactate and pyruvate and glyceral-dehyde 3-phosphate (GA3P) (22).

J �C6 � units metabolized

�t�

0.5��lactate�

�t(Eq. 2)

Our experimental animal model employs total myocardial ischemia.As a consequence, the use of extracellular glucose, which is taken up bythe cell via glucose transporters and enters glycolysis after phosphoryl-ation to glucose 6-phosphate (Glu-6-P) by hexokinase can be quantita-tively neglected. In total myocardial ischemia, fuel for glycolysis is onlyprovided by the glycosyl units resulting from glycogen breakdown thatenter glycolysis via glucose 1-phosphate (Glu-1-P), and to a lesserextent, free glucose.

As 1,3-bisphosphoglycerate could not be detected, glyceraldehydedehydrogenase and phosphoglycerate kinase were considered as a sin-gle glycolytic step. As 2- and 3-PG, as well as of DHAP, contents werecalculated the given values are not mathematically independent. There-fore, also GAPDH and 3-PGK as well as PGM and enolase were also puttogether.

On basis of these calculations, anaerobic glycolytic regulation wasinvestigated using flux:metabolite co-responses as described in detailpreviously (23).

Oflux,metabolite �

�ln flux�ln metabolite

��ln flux

�ln metabolite�

�flux/flux� metabolite/metabolite

(Eq. 3)

Note that it follows from the definition of a co-response OJ,S that J �cSO, where c is a constant. Where an enzyme converts metabolite S tometabolite P the co-responses OJ,S and OJ,P involve the same flux J, soOJ,S/OJ,P � �lnP/�lnS, and if the enzyme had no change in its mass:action ratio (P/S) when �lnP � �lnS then the co-responses will be equal.If OJ,S is greater than OJ,P then there has been a relatively largerincrease in product concentration, and the reaction has moved closer toequilibrium, whereas if OJ,S is smaller than OJ,P, the reaction hasmoved away from equilibrium. Where only a single enzyme in a path-way is altered, flux:substrate co-response coefficients are likely to beless than 1 (or less than the Hill coefficient for cooperative enzymes)(23). Under several known physiological conditions (e.g. hypoxia) (23),observed co-response coefficients are larger than this; exact mechanisticexplanations of this are not known, but in theory, parallel activation ofseveral steps (the “universal method” (24), multisite modulation (25,26), or proportional activation (27)) can account for this. Enzymes thatare extremely close to equilibrium (25) can show large co-responsesduring passive adaptation to a change in metabolic flux, but this wouldimply virtually no detectable change in the displacement from equilib-rium, and hence �ln (P/S) � 0, with virtually identical substrate andproduct co-responses as described above. A negative flux:substrate co-response corresponds to a classic crossover effect, where activation of anenzyme has induced a decrease in its substrate. Thus the co-responsecoefficients provide insight into the events associated with a change influx.

Assumptions—Our study is based on two major assumptions. (i)Essentially maintained tissue integrity during index ischemia and (ii)constant levels and activities in the enzymes of anaerobic glycolysis,which are both supported by a vast body of evidence. (i) Lethal myocar-dial injury is initialized and determined during ischemia, whereas thedemarcation of necrosis happens during reperfusion. In this context, itis one of the seminal findings of myocardial pathophysiology that “myo-cytes have suffered severe injury before reperfusion, and that the dra-matic consequences of reperfusion are simply postmortem manifesta-tions of lethal injury made possible by the sudden availability of largevolumes of plasma water, calcium, or both. In other words, reperfusionaccelerates the undertaking events associated with cell death” (28). (ii)Because protein synthesis and degradation are inhibited in myocardialischemia (29), and because of the short time course of index ischemia,constant levels and unchanged activities in the enzymes of the Embden-Meyerhof pathway can be assumed.

In accordance to these basic findings, we feel justified to perform ouranalysis of anaerobic glycolytic regulation in myocardium with an un-changed enzymatic composition and an essentially maintained tissueintegrity prior to reperfusion. Thus, lactate accumulation could be usedas a measure of anaerobic glycolytic flux.

Statistical Analysis—Comparisons of metabolite levels betweengroups were performed using ANOVA, for comparisons within groupsrepeated measures of ANOVA (Scheffe-test) was used. Analysis of vari-

FIG. 1. Control animals (C), following a stabilization period of2 h, were subjected to 45 min of index ischemia and 120 min ofreperfusion. In the IP group, a preconditioning cycle of brief LADocclusion (10 min) and reperfusion (30 min) was performed twice. LADocclusions are indexed by thick-lined boxes, reperfusion periods bywhite bars. The numbers indicate duration of these periods in minutes.The arrows mark the time points when myocardial biopsies were taken(before as well as at 2, 5, 10, 20, and 45 min after onset of indexischemia).

Glycolysis in Ischemic Preconditioning24412

at NY

U School of M

edicine Library on February 6, 2015

http://ww

w.jbc.org/

Dow

nloaded from

ance was also employed to calculate regression differences. The ob-tained correlations were compared using the Spearman’s rank correla-tion coefficient. Statistical differences were assumed at p values smallerthan 0.05.

RESULTS

Myocardial Protection—At constant areas at risk (C, 17.2 �0.12%; IP, 15.8 � 0.72%), IP limits infarct size (C, 90.0 � 3.1versus 5.05 � 2.1%; p � 0.001).

Anaerobic Glycolytic Flux—As ischemia in control animalsproceed, anaerobic glycolytic flux is slowed down. In precondi-tioned myocardium, glycolytic flux is decreased at the onset ofindex ischemia, showing an even more pronounced decelerationduring the subsequent phase (Fig. 2). Glycogen content at theonset of index ischemia did not differ between normal and IPmyocardium (control, 99.0 � 8.09 �mol of C6-units/g wetweight; IP, 96.2 � 16.7).

Flux:Metabolite Co-responses—Myocardial metabolite con-tents are given in Table I. Flux:metabolite co-responses areillustrated in Figs. 3 and 4, whereby the co-responses of me-tabolites within glycolysis itself are depicted in Fig. 3, and theco-responses of non-glycolytic metabolites are shown in Fig. 4.The corresponding regression data are given in Table II, sig-nificant differences could be attributed to every alteration inthe flux:metabolite co-response between control and IP myo-cardium. As kinetic properties of an enzyme may depend onboth substrate as well as product levels, the following descrip-tion of glycolytic regulation during acute myocardial ischemiaand its modulation by preceding ischemic preconditioning fo-cuses on the enzymes of the Embden-Meyerhof pathway.

Glycogen Phosphorylase/Synthase—Glycogen is the sourceof glucose 6-phosphate during ischemia, so the reducing glyco-lytic flux corresponds to a reduced net flux through this poten-tial substrate cycle.

In control myocardium, Glu-1-P levels decrease as the netflux through the enzymes breaking down glycogen to Glu-1-Pincreases. The co-response (in the range �0.5 to �1) couldeasily be consistent with a slowing of this step by productinhibition of the glycogen phosphorylase and substrate activa-tion of glycogen synthesizing reactions (UDP-glucose phospha-tase/synthase). In IP myocardium, the decrease in flux for agiven Glu-1-P level is significantly more pronounced. Becausethe graph for IP does not represent an extension of the graphfor controls the altered flux:metabolite co-responses indicatechanged biochemical properties of these enzymes.

The myocardial contents in its co-substrate Pi and allostericactivator AMP only showed poor correlations (not significant)

with glycolytic flux under control conditions. Thus, it is un-likely that they contributed to flux regulation at this level.Although there was a better formal correlation between Pi (p �0.05) and AMP (not significant) following IP, decreases in Pi

and AMP with increasing flux still do not seem sufficient toaccount for the change in enzyme activity observed.

Phosphoglucomutase—In control animals, the decrease inglycolytic flux during index ischemia is associated with anincrease in the substrate (G1P) content of PGM. This occurs atbasically stable levels in the product of PGM, Glu-6-P.

These findings indicate that the observed inhibition of gly-colytic flux during acute myocardial ischemia in control myo-cardium is not because of limited G1P availability. Substrateaccumulation at increasing net flux, corresponding to a classi-cal crossover effect, suggests a regulatory step downstreamfrom the Embden-Meyerhof pathway, causing a passive in-crease in Glu-1-P. However, as changes in net flux were inde-pendent of Glu-6-P levels, product inhibition of PGM may notaccount for this alteration. Hence, other regulatory mecha-nisms must be considered. Moreover, the observation of de-creasing substrate contents at stable product levels indicatesthat the displacement from equilibrium must be decreasingdetectably during ischemia, even though this reaction is com-monly viewed as close to equilibrium.

Following IP, comparable increases in Glu-1-P content dur-ing ischemia were associated with even more inhibition of fluxthrough PGM. Correspondingly, the correlation between fluxinhibition and substrate accumulation shows a significantlysteeper slope in IP myocardium. Moreover, the graph for pre-conditioned myocardium does not represent an extension of thecorrelation, which has been calculated for control myocardium.This altered correlation between PGM and changing substrateconcentrations indicates a different kinetic behavior of thisenzyme. Moreover, as substrate content changed independ-ently of product levels, a shift in the PGM mass:action ratioalso occurred in this case, although these shifts were in anopposite direction as observed in control myocardium.

Also following IP, the decrease in glycolytic flux throughoutindex ischemia occurred at basically stable Glu-6-P levels.However, there was a significantly negative flux-Glu-6-P co-response. So although there was potentially an increase inproduct inhibition by Glu-6-P, it is doubtful whether it couldcounteract the larger increase in Glu-1-P.

As a consequence, the kinetic response of this enzyme doesnot seem accounted for by the changes in its substrate andproduct according to classical substrate and product kinetics.Together with the fact that the response following IP is clearlydifferent from that of controls, suggests that preconditioninghas had an effect on this step by an unknown mechanism,reinforced by the fact that for a given Glu-1-P concentration theflux is much lower in IP hearts than in controls.

Glucose-phosphate Isomerase—In control myocardium, thedecrease in flux through glucose-phosphate isomerase is onlyvery weakly correlated to substrate content, and no signifi-cance can be attached to the co-response coefficient with glu-cose 6-phosphate. Hence, the behavior of this enzyme cannot beexplained by substrate kinetics. However, as Fru-6-P contentwas increased at reduced flux levels, product inhibition maycontribute to enzymatic regulation. As substrate levels changeindependently of product concentration, a changing mass:ac-tion ratio (closer to equilibrium) must be assumed for glucose-phosphate isomerase.

In IP myocardium, the changes in substrate and productco-response coefficients are indicative of a shift of the mass:action ratio away from equilibrium as ischemia progresses.However, because of the large negative co-responses, mecha-

FIG. 2. Anaerobic glycolytic flux in control (filled) and IP myo-cardium (unfilled).

Glycolysis in Ischemic Preconditioning 24413

at NY

U School of M

edicine Library on February 6, 2015

http://ww

w.jbc.org/

Dow

nloaded from

nisms distinct from classical substrate or product-dependentenzyme kinetics are apparently operative in IP.

Phosphofructokinase-1—In control myocardium, the de-crease in glycolytic flux is accompanied with an increase in thesubstrate Fru-6-P for PFK-1. Fru-1,6-P2 increases with in-creasing flux, showing a positive flux co-response and a changein mass:action ratio away from the equilibrium. Strictly, theADP:ATP ratio should be included in this calculation, but as itchanges in the same sense as the Fur-1,6-P2:Fru-6-P ratio(from the relative sizes of the ATP and ADP co-response coef-ficients), it reinforces rather than counteracts the change.

These observations imply that under control conditions aninhibition of glycolytic flux was not because of limited Fru-6-Pavailability. Moreover, Fru-6-P accumulation suggests a regu-latory step in downstream glycolysis, causing a passive in-crease in Fru-6-P. This finding is consistent with an inhibitoryrole of PFK-1 under these conditions.

AMP, ADP, citrate, and especially Fru-2,6-P2 are potentactivators of PFK-1. Hence, the finding of decreased contents inthese metabolites at decreased net fluxes through PFK-1 mayaccount for the behavior of this enzyme, assuming that lowconcentrations of these stimulatory factors have more impacton regulation of PFK-1 than its inhibitor ATP, for which adecreased content was found. However, control myocardiumshows a reduced glycolytic rate with increased Fru-6-P, and

little change in allosteric effector AMP or co-substrate ATP (atleast relative to Km for ATP).

In IP myocardium, the decrease in glycolytic flux is associ-ated with a steep increase in Fru-6-P content, whereas thedecrease in flux through PFK-1 occurs independent of Fru-1,6-P2 levels. For this enzyme under IP conditions, substrateand product contents are indicative of a change in mass:actionratio away from equilibrium.

According to the altered flux/substrate relationship, themechanism involved in downstream regulation has apparentlychanged following IP. Under this condition, PFK-1 regulationmay not be explained on the basis of co-response, as largepositive co-responses of PFK flux to Fru-6-P are not consistentwith an effect of substrate kinetics. Moreover, the change influx through PFK-1 relative to controls occurred at largelyunchanged contents in citrate, AMP, and Fru-2,6-P2. Again,mechanisms distinct and independent from substrate levelsand contents of effector metabolites are apparently operative.

Phosphofructokinase 2/Fructose-2,6-bisphosphatase—As PFK1/Fru-2,6-P2ase represents a dead-end side branch of the Embden-Meyerhof pathway, flux through this enzyme differs from net gly-colytic flux. However, Fru-2,6-P2 has been shown to be importantfor regulating the biochemical characteristics of PFK-1.

As PFK-1/Fru-2,6-P2ase is known to be inhibited by Fru-6-P,its decreased content as glycolytic flux increases may well

TABLE IMyocardial metabolite content, given as mean � S.E. in control myocardium and myocardium subjected to ischemic preconditioning

In addition, values for anaerobic glycolytic flux are also provided (in �mol/g/min). Italic numbers indicate significant differences between controland IP myocardium.

Metabolite Group

Time

�80 0 2 5 10 20 45

Mean S.E. Mean S.E. Mean S.E. Mean S.E. Mean S.E. Mean S.E. Mean S.E.

Glu-1-P Control 0.048 0.004 0.045 0.004 0.143 0.044 0.285 0.060 0.550 0.108 1.048 0.125 1.382 0.198IP 0.056 0.022 0.048 0.017 0.229 0.058 0.374 0.089 0.595 0.094 0.870 0.098 1.192 0.165

Glu-6-P Control 0.310 0.060 0.274 0.056 0.471 0.054 0.388 0.051 0.546 0.078 0.502 0.114 0.402 0.068IP 0.287 0.040 0.440 0.075 0.440 0.056 0.427 0.080 0.463 0.061 0.604 0.096 0.549 0.098

Fru-6-P Control 0.037 0.005 0.023 0.005 0.095 0.014 0.156 0.016 0.215 0.028 0.355 0.076 0.546 0.128IP 0.035 0.006 0.037 0.005 0.062 0.009 0.084 0.013 0.085 0.017 0.129 0.026 0.093 0.031

F1.6P2 Control 0.022 0.012 0.040 0.003 0.067 0.004 0.051 0.010 0.089 0.030 0.040 0.013 0.018 0.007IP 0.017 0.004 0.042 0.012 0.035 0.006 0.057 0.029 0.072 0.032 0.048 0.014 0.028 0.013

F2.6P2 Control 0.063 0.023 0.082 0.019 0.236 0.082 0.218 0.082 0.224 0.109 0.168 0.123 0.023 0.006IP 0.057 0.020 0.139 0.038 0.110 0.053 0.135 0.064 0.156 0.037 0.104 0.038 0.090 0.046

Glyceraldehyde3-phosphate

Control 0.010 0.002 0.014 0.009 0.014 0.002 0.010 0.003 0.019 0.004 0.019 0.003 0.007 0.002

IP 0.009 0.001 0.028 0.002 0.038 0.011 0.044 0.012 0.030 0.006 0.022 0.003 0.024 0.009DHAP Control 0.239 0.041 0.336 0.207 0.337 0.037 0.250 0.060 0.449 0.087 0.462 0.068 0.165 0.049

IP 0.213 0.036 0.661 0.055 0.913 0.273 1.064 0.287 0.717 0.150 0.522 0.072 0.564 0.2143-PG Control 0.041 0.015 0.047 0.017 0.074 0.033 0.049 0.026 0.107 0.035 0.089 0.032 0.045 0.017

IP 0.039 0.014 0.066 0.025 0.067 0.031 0.063 0.019 0.070 0.024 0.053 0.020 0.053 0.0112-PG Control 0.006 0.002 0.007 0.003 0.011 0.005 0.007 0.004 0.016 0.005 0.014 0.005 0.007 0.003

IP 0.006 0.002 0.010 0.004 0.010 0.005 0.010 0.003 0.011 0.004 0.008 0.003 0.008 0.002PEP Control 0.009 0.002 0.012 0.003 0.019 0.005 0.014 0.004 0.018 0.004 0.013 0.002 0.012 0.003

IP 0.009 0.002 0.019 0.003 0.012 0.003 0.014 0.003 0.014 0.004 0.015 0.003 0.012 0.003Pyr Control 0.051 0.010 0.041 0.013 0.066 0.017 0.067 0.021 0.070 0.020 0.068 0.028 0.095 0.022

IP 0.047 0.015 0.046 0.015 0.062 0.017 0.107 0.041 0.079 0.024 0.096 0.029 0.093 0.037Lac Control 2.573 0.498 2.926 0.704 7.246 1.548 12.949 2.067 19.770 2.827 31.862 4.473 49.120 8.227

IP 2.688 0.506 4.092 1.002 6.384 0.765 7.233 0.836 9.402 1.227 10.344 1.263 15.868 3.770ATP Control 3.142 0.314 3.357 0.313 2.365 0.236 2.183 0.273 1.559 0.153 1.094 0.267 0.147 0.028

IP 2.995 0.277 2.427 0.362 2.009 0.220 0.987 0.107 0.994 0.120 0.954 0.156 0.599 0.100ADP Control 0.352 0.128 0.254 0.025 1.029 0.057 1.056 0.107 1.066 0.157 0.536 0.115 0.310 0.039

IP 0.330 0.136 0.220 0.038 0.831 0.122 0.761 0.035 0.753 0.073 0.664 0.038 0.513 0.082AMP Control 0.091 0.032 0.030 0.008 0.597 0.078 0.675 0.069 0.748 0.098 0.262 0.061 0.555 0.155

IP 0.088 0.036 0.063 0.029 0.400 0.110 0.651 0.049 0.557 0.086 0.500 0.100 0.461 0.072Free ADP Control 0.351 0.102 0.296 0.039 1.125 0.074 1.179 0.127 1.027 0.068 0.501 0.113 0.249 0.026

IP 0.330 0.095 0.342 0.107 0.825 0.156 0.774 0.049 0.716 0.077 0.621 0.045 0.484 0.052Pi Control 3.970 0.493 4.700 0.143 10.158 0.744 10.283 1.011 13.452 1.427 12.388 1.958 11.758 1.375

IP 4.187 0.443 6.624 1.091 10.225 1.420 9.976 0.975 10.602 0.722 11.567 1.555 10.862 1.636Citrate Control 0.139 0.029 0.175 0.053 0.426 0.050 0.336 0.034 0.349 0.027 0.235 0.045 0.077 0.015

IP 0.128 0.033 0.304 0.065 0.352 0.055 0.312 0.064 0.325 0.061 0.391 0.085 0.186 0.036Glycolytic flux Control 1.080 0.383 0.951 0.228 0.682 0.162 0.605 0.172 0.345 0.127

IP 0.573 0.150 0.142 0.051 0.217 0.099 0.047 0.014 0.110 0.061

Glycolysis in Ischemic Preconditioning24414

at NY

U School of M

edicine Library on February 6, 2015

http://ww

w.jbc.org/

Dow

nloaded from

account for the increased contents in Fru-2,6-P2 in controlmyocardium. In addition to the pattern of regulation found forPFK-1, this finding is in good agreement with a regulatory rolefor PFK-1 under control conditions. Thus, an increased contentin the positive regulator of Fru-2,6-P2 may be involved inadjusting glycolytic flux.

In IP myocardium, a significant correlation between Fru-2,6-P2 content and glycolytic flux was not observed. This find-ing fits well to the altered pattern for PFK-1 (see above) withrespect to flux adjustment following IP, which appears to beindependent of its “traditional” regulators.

Aldolase—In control myocardium, a decrease in aldolase fluxwas accompanied by a decrease in Fru-1,6-P2 content. Hence,for aldolase, the positive co-response observed indicates thatchanges in substrate content may well account for changes inaldolase flux. At aldolase, the courses of substrate and productcontent indicate a shift of the mass:action ratio toward equi-librium conditions.

In contrast to the control myocardium, no significant corre-lations between substrate or product content and net flux couldbe observed following IP. This finding is not in favor of aflux-regulatory role for aldolase under these conditions.

GAPDH and 3-PGK—In control myocardium, the observedreductions in flux occurs with poor correlations with glyceral-dehyde 3-phosphate and 3-PG content. To the extent that theco-responses with respect to substrate and product are compa-rable, the pattern observed indicates a similar inhibition ofsubstrate supply and product demand, implying that theseenzymes apparently do not play major roles in flux regulation.No relevant shift in mass:action ratio could be seen.

In IP myocardium, again no significant correlation betweensubstrate content and net flux was observed. For 3-PG content,a significant correlation was calculated. However, the largevalue for the co-response might indicate that in addition toproduct concentration, additional modifications of the enzymeactivities themselves may be responsible for the altered behav-

FIG. 3. Flux:metabolite graphs for glycolytic metabolites (mean � S.E.). Filled, control; unfilled, ischemic preconditioning.

Glycolysis in Ischemic Preconditioning 24415

at NY

U School of M

edicine Library on February 6, 2015

http://ww

w.jbc.org/

Dow

nloaded from

ior of these glycolytic steps in adjusting flux in IP myocardium.Mass:action ratio shifted away from equilibrium.

PGM/Enolase—In control myocardium, the change in fluxwas poorly correlated with the substrate and product levels. Thispattern does not suggest a regulating role for these enzymes, andthe mass:action ratios remained largely unchanged.

In IP myocardium, the biochemical properties of these en-zymes have changed as different correlations between enzymefluxes and substrate content were seen in the co-response val-

ues. The values are large and positive for the substrate but notsignificant for the product. Because changes in flux occur atalmost unchanged substrate and product contents, the behav-iors of these enzymes may show that they are very close toequilibrium.

Pyruvate Kinase—In control myocardium, the decrease inglycolytic flux was not accompanied by statistically significantchanges in substrate content. Although for its product pyruvatea significant co-response could be observed, the value was too

FIG. 4. Flux:metabolite graphs for non-glycolytic metabolites (mean � S.E.). Filled, control; unfilled, ischemic preconditioning.

TABLE IIRegression data (power, y � a � xb) and correlation coefficients (r2 values) for the regressions displayed in Figs. 3 and 4. p Values for statistical

significances of the observed correlations were calculated using the Spearman’s rank correlation coefficients. The regressions between controland IP myocardium were compared using ANOVA (p values given).

Control Ischemic preconditioning Cversus

IPa b r2 p a b r2 p

Glu-1-P 0.502 �0.444 0.858 �0.01 0.181 �1.094 0.615 �0.005 0.037Glu-6-P 0.937 0.410 0.018 0.2 0.005 �4.824 0.623 �0.05 0.022Fru-6-P 0.270 �0.627 0.923 �0.001 0.000 �3.417 0.951 �0.001 0.002Fru-1.6-P2 4.126 0.589 0.620 �0.05 0.088 �0.186 0.006 0.2 0.017Fru-2.6-P2 1.522 0.401 0.800 �0.01 1.912 1.167 0.077 0.2 0.015Glyceraldehyde

3-phosphate5.158 0.466 0.202 0.1 163.041 1.989 0.438 0.05 0.012

DHAP 1.174 0.466 0.202 0.1 0.293 1.989 0.438 0.05 0.0123-PG 1.593 0.318 0.071 0.2 1.142 5.390 0.626 �0.05 0.0232-PG 2.889 0.318 0.071 0.2 1.354E 10 5.390 0.626 �0.05 0.023PEP 252.901 1.409 0.424 0.05 2.963E � 14 �6.789 0.382 0.1 0.028Pyr 0.001 �2.650 0.814 �0.01 2.24E � 5 �3.572 0.657 �0.02 0.022Lac 3.718 �0.571 0.913 �0.001 4.889 �1.542 0.355 0.1 0.004

ATP 0.668 0.374 0.911 �0.001 0.150 1.522 0.516 �0.05 0.007AMP 0.808 0.274 0.065 0.2 0.045 �1.833 0.136 0.2 0.019ADPfree 0.849 0.622 0.884 �0.02 0.452 2.685 0.386 0.05 0.009Pi 95.469 �2.021 0.299 0.1 1.555E 11 �11.690 0.535 �0.05 0.007Citrate 1.616 0.617 0.887 �0.001 1.379E � 7 0.235 0.005 0.2 0.010

Glycolysis in Ischemic Preconditioning24416

at NY

U School of M

edicine Library on February 6, 2015

http://ww

w.jbc.org/

Dow

nloaded from

large to explain the behavior of this enzyme on the basis ofproduct inhibition, so other mechanisms must be considered.

In IP myocardium, flux through PK again changes at basi-cally stable levels for substrate content. For its product pyru-vate, again significant co-responses could be observed, al-though this correlation did differ with controls. For bothexperimental conditions, there was little evidence for a changein the mass:action ratio of the reaction, because the co-re-sponses to PEP and pyruvate are comparable within experi-mental error. These observations imply that this enzyme takespart in flux adjustment under both experimental conditions,although its role could not be attributed to substrate or productkinetics and distinct mechanisms must be considered.

Lactate dehydrogenase—As flux through LDH decreases,pyruvate content remains basically stable, whereas lactate con-tent increases. These data not only indicate a shift in mass:action ratio away from equilibrium but suggest that productinhibition of LDH may be a contributory factor.

In IP myocardium, the substrate for LDH again remained atbasically stable levels, but the correlation between lactate con-tent and glycolytic flux has lost its statistical significance. Theslope of the regression graph was much lower than �1.0, indi-cating that regulatory mechanisms distinct from classical me-tabolite-flux kinetics must occur.

Response of Glycolytic Flux to Changes in Non-glycolyticMetabolites—As changes in glycolytic flux may occur in re-sponse to alterations in contents of non-glycolytic metabolitesacting as modulators of certain glycolytic enzymes, their pos-sible contribution to flux adjustment was also investigated.

In control myocardium, a decrease in glycolytic flux withdecreasing ATP content could be seen. Although there was acomparable response of glycolytic flux at high ATP concentra-tions in IP myocardium, the deceleration in glycolytic flux withdecreasing ATP content was even more pronounced. A compa-rable pattern was found for myocardial- free ADP content,although the flux:metabolite co-responses for free ADP werenot significant in IP myocardium. For AMP content, no signif-icant correlation to glycolytic flux was observed under controland IP conditions. In control myocardium, inorganic phosphatecontent did not significantly correlate to glycolytic flux. Al-though this correlation formally improved in IP myocardium,

flux increased as Pi remained stable, indicating an independ-ence of flux adjustment from Pi levels. In control myocardium,anaerobic flux decelerated with decreasing citrate contents,whereas for IP myocardium, no significant correlation could beobserved.

Summarizing Considerations—Under both experimentalconditions, glycolytic flux slows down as index ischemia pre-cedes. However, this inhibition is more pronounced in IP myo-cardium and employs distinct mechanisms of flux adjustment(Table III).

In control myocardium, decreasing substrate concentrationsaccount for decreased fluxes at PFK-1 and aldolase levels. Theregulating properties of glycogen phosphorylase, glucose-phos-phate isomerase, and lactate dehydrogenase could be attrib-uted to product inhibition. For phosphoglucomutase, GAPDH/3-PGK, PGM/enolase, and PK, the mechanisms employed forflux adjustment were found to be independent of traditionalsubstrate kinetics. Of these enzymes, a regulating role could befound for phosphoglucomutase and PK, whereas the block con-sisting of GAPDH/3-PGK and PGM/enolase showed a ratherpassive behavior.

In IP myocardium, the pattern of flux regulation haschanged. Decreasing substrate concentrations do not accountfor decreased fluxes through PFK-1 and aldolase. Here, thecharacteristics of these enzymes have obviously changed, andfor both enzymes, mechanisms distinct to traditional sub-strate kinetics, such as covalent modification, must be con-sidered. For adjustment of flux by glycogen phosphorylase/synthase and LDH, product inhibition could also be shown inIP myocardium. However, the sensitivities of these enzymeswere modulated by IP, and for LDH, an additional mecha-nism appeared. Aldolase has lost its importance for fluxadjustment following IP, whereas formerly insignificant en-zymes (block from GAPDH to enolase) play active roles inmodulating flux in IP myocardium.

DISCUSSION

Metabolic Control Analysis—For almost every known en-zyme, numerous data describing their biochemical character-istics are readily available (30). On the basis of these kineticand thermodynamic properties, various concepts were devel-

TABLE IIISummary of enzymatic properties of the glycolytic enzymes in control and IP myocardium

Enzyme

Control Ischemic preconditioning

Change inequilibriumduring index

ischemia?

Mechanism offlux adjustment

Role for fluxadjustment

Change inequilibriumduring index

ischemia?

Mechanism offlux adjustment

Role for fluxadjustment

Glycogenphosphorylase/synthase

NDa Productinhibition

Active NDa Productinhibition

Active

Phosphoglucomutase Away fromequilibrium

Other Active Towardsequilibrium

Other Active

Glucose-phosphateisomerase

Towardsequilibrium

Productinhibition

Active Away fromequilibrium

Other Active

PFK-1 Away fromequilibrium

Substratekinetics/allosteric

Active Away fromequilibrium

Other Active

Aldolase Towardsequilibrium

Substratekinetics

Active No change Other Passive

GAPDH/3-PGK No change Other Passive Away fromequilibrium

Substratekineticsand other

Active

PGM/enolase No change Other Passive Towardsequilibrium

Other Active

PK Towardsequilibrium

Other Active Towardsequilibrium

Other Active

LDH Away fromequilibrium

Productinhibition

Active Away fromequilibrium

Productinhibitionand other

Active

a ND, not determined.

Glycolysis in Ischemic Preconditioning 24417

at NY

U School of M

edicine Library on February 6, 2015

http://ww

w.jbc.org/

Dow

nloaded from

oped to explain metabolic regulation (13, 14). These traditionalconcepts mostly convey that control over a pathway is achievedby action on a single pathway enzyme, which is often assumedto be a non-equilibrium step near the beginning of the pathwaybeing subjected to feedback inhibition. However, starting withthe rise of MCA, there is growing evidence that these comfort-able, traditional concepts may mislead more than enlighten(23). Therefore, MCA was first applied to obtain a state-of-the-art analysis of glycolytic flux, and its regulation in myocardiumwas subjected to zero flow ischemia with and without priorischemic preconditioning.

Regulation of Anaerobic Glycolytic Flux during Ischemia inControl Myocardium—In control myocardium, a continuousdecrease in anaerobic glycolytic flux was observed. As regula-tory roles for most pathway enzymes could be shown, our datastrongly contradict concepts implying glycolytic flux to be ad-justed by only one or two regulatory sites (“key enzymes”).Rather, as the tools for altering glycolytic flux were distributedalong the entire Embden-Meyerhof pathway, the findings ofour study are in good agreement with the concept of multisitemodulation, enabling rapid and tight adjustments of glycolyticflux to changing demands (23). This cannot be achieved if onlyone regulatory site is operative. Moreover, as Glu-1-P contentdid not decrease but increased with decelerating glycolytic flux,the concept of substrate availability as regulator of fluxthrough glycolysis in acute zero flow myocardial ischemia (31)is not supported by our findings.

Of the steps of the glycolytic chain, only the downstreamblock from GAPDH to PK was shown not to play dominatingroles in flux adjustment. For the remaining enzymes, the flux-substrate (product) analyses revealed an active participation,whereas the mechanism used for regulating enzymatic proper-ties were different. Product inhibition accounted for the behav-iors of glycogen phosphorylase/synthase, glucose-phosphateisomerase, and LDH, whereas for PFK-1 a regulatory patternconsistent with traditional substrate kinetics could be ob-served. Moreover, the in vivo properties of PFK-1 for fluxadjustment as documented by MCA show a good correlation tothe known biochemical in vitro properties of this enzyme, suchas its modulation by changing concentrations in adenine nu-cleotides, Fru-2,6-P2, and citrate. For phosphoglucomutase,however, MCA suggests that this step may no longer be seenonly passively linking glycogen phosphorylase to glucose-phos-phate isomerase. Moreover, the mechanism of the regulation ofthis enzyme could not be explained on the basis of substratekinetics, so that other regulatory mechanisms must be takeninto account.

According to traditional concepts of glycolytic regulation, itwas assumed that decreasing concentrations in ATP and in-creasing levels in ADP, AMP, and citrate, indicating energyshortage, will increase flux through glycolysis. In our analysis,this concept was supported by considering the co-responses ofglycolytic flux to citrate and ADP, whereas for AMP, regressionanalysis could not prove a dependence of net flux from thismetabolite. Moreover, glycolytic flux decreased as ATP levelsdeclined, contradicting traditional models. However, as lactateaccumulation appears to be a stronger noxious stimulus forischemic myocardium than ATP depletion under the conditionsof zero flow ischemia (32), this finding might indicate a usefuladaptive myocardial mechanism.

Regulation of Anaerobic Glycolytic Flux during Ischemia inMyocardium Protected by Ischemic Preconditioning—Followingischemic preconditioning, the deceleration of anaerobic glyco-lytic flux was even more marked. As it was shown for controlmyocardium, the adjustment of anaerobic glycolytic flux couldneither be attributed to substrate availability (myocardial gly-

cogen content was unchanged in control and IP myocardium)nor to one certain flux regulating the key enzyme. Again,multisite modulation could be observed, whereas the patternobserved along the glycolytic chain has been altered by thepreceding proconditioning protocol.

Although the regulation at the glycogen phosphorylase/syn-these level was also characterized by product inhibition, thesensitivity of this enzyme to react to changing product concen-trations has been changed, indicating a modification of thisenzyme. As it was seen in control myocardium, the propertiesof PGM could not be explained on the basis of classical sub-strate kinetics. However, the much better correlation observedbetween flux and substrate content indicates an alteration inthe biochemical property of the enzyme. Also for glucose-phos-phate isomerase, a change in enzyme characteristics could bedocumented, as the regulatory mechanism following IP couldno longer be explained by product inhibition. Comparable find-ings were obtained for PFK-1 and aldolase, whereby the latterenzyme has lost its active role for flux regulation in IP myo-cardium. As their biochemical properties had changed, theformerly rather passive enzymes downstream from aldolaseare now actively involved in flux adjustment. For LDH, anincrease in its degree of product inhibition could be shown,whereas additional regulatory mechanisms must be consid-ered. Unlike in control myocardium, the adjustment of glyco-lytic flux following IP was independent of the contents incitrate.

As the biochemical behaviors and the regulatory propertiesof most glycolytic enzymes have changed following IP, thedescription of the mechanisms responsible for these partiallyunexpected alterations is an important issue. In this context,the method used for analyzing glycolytic regulation validlyallows to differentiate whether an enzyme employs substrate/product kinetics in classical terms or uses distinct regulatorymechanisms. Although the latter mechanisms cannot be spe-cifically characterized using the co-response approach, a differ-entiation whether or not a change in the regulating mechanismoccurs can be performed.

Anaerobic glycolysis was always assumed to be a metabolicpathway, which is exclusively regulated on the basis of sub-strate kinetics or allosteric mechanisms (13, 33–35). Our dataindicate that in addition to these traditional mechanisms, otherprinciples must also be considered to explain flux adjustmentby glycolytic enzymes. For glycogen phosphorylase, its regula-tion by enzyme phosphorylation is firmly established. There-fore, this mechanism must also be considered by analyzing thealterations in enzyme behavior following IP. Interestingly,there is increasing evidence that also the kinetic behavior ofglycolytic enzymes may be influenced by enzyme modification,which must not only be because of phosphorylation (36–44). Asit could already be seen in control myocardium, glycolytic fluxdecreased as ATP levels declined, although the sensitivity ofchanging glycolytic flux to alterations in ATP content was evenincreased, supporting the view that decelerating lactate accu-mulation rather than ATP depletion indicates a useful adaptivemechanism to severe myocardial ischemia.

Summary and Possible Implications—The deceleration ofglycolytic flux in myocardium protected by ischemic precondi-tioning is achieved by multisite modulation, whereby the reg-ulatory mechanism of the glycolytic enzymes could only partlybe explained by classical enzyme kinetics. Moreover, previouslyunknown patterns of flux adjustment were observed. Becausethe attenuation of lactate accumulation represents a majorprotective mechanism of IP, the analysis of these regulatorymechanisms might have therapeutic implications.

Glycolysis in Ischemic Preconditioning24418

at NY

U School of M

edicine Library on February 6, 2015

http://ww

w.jbc.org/

Dow

nloaded from

REFERENCES

1. Jennings, R., and Reimer, K. (1991) Annu. Rev. Med. 42, 225–2462. Bing, R. (1954–1955) in Harvey Lecture Series, pp. 27–70, Academic Press,

New York3. Bing, R. (1964) in Circulation of the Blood: Men and Ideals (Fishman, A., and

Richards, D., eds) Oxford University Press, New York4. Vary, T., Reibel, D., and Neely, J. (1981) Annu. Rev. Physiol. 43, 419–4305. Neely, J., and Morgan, H. (1974) Annu. Rev. Physiol. 36, 413–4596. Rovetto, M., Lamberton, W., and Neely, J. (1975) Circ. Res. 37, 742–7517. Neely, J., and Grotyohann, L. (1984) Circ. Res. 55, 816–8248. Reimer, K. (1996) Ann. N. Y. Acad. Sci. 793, 13–269. Vogt, A., Ackermann, C., Noe, T., and Kubler, W. (1999) J. Am. Coll. Cardiol.

33, 387A (abstr.)10. Jennings, R., Murry, C., and Reimer, K. (1991) J. Mol. Cell. Cardiol. 23,

1449–145811. Murry, C., Richard, V., Reimer, K., and Jennings, R. (1990) Circ. Res. 66,

913–93112. Wikstrom, B., Ronquist, G., and Waldenstrom, A. (1995) Eur. Heart J. 16,

563–56913. Kubler, W., and Spieckermann, P. (1970) J. Mol. Cell. Cardiol. 1, 351–37714. Gordon, E., and Morgan, H. (1992) in The Heart and Cardiovascular System:

Scientific Foundations (Fozzard, H., Haber, E., Jennings, R., Katz, A., andMorgan, H., eds) Second Edition, pp. 151–61, Raven Press, New York

15. Fell, D. (1997) Understanding the Control of Metabolism, Portland Press,London

16. Vogt, A., Ando, H., Arras, M., and Elsasser, A. (1998) J. Am. Coll. Cardiol. 31,1134–1141

17. Vogt, A., Htun, P., Arras, M., Podzuweit, T., and Schaper, W. (1996) Basic Res.Cardiol. 91, 389–400

18. Sellevold, O., Jynge, P., and Aarstad, K. (1986) J. Mol. Cell. Cardiol. 18,517–527

19. Weinbrenner, C., Wang, P., and Downey, J. (1996) Basic Res. Cardiol. 91,374–381

20. Buddecke, E. (1984) Grundriss der Biochemie, 7 Ed., Walter de Gruyter & Co.,Berlin

21. Connett, R. (1988) Am. J. Physiol. 254, R949–R95922. Spriet, L. (1990) Pflugers Arch. 417, 278–28423. Thomas, S., and Fell, D. (1996) J. Theor. Biol. 182, 285–29824. Kacser, H., and Acerenza, L. (1993) Eur. J. Biochem. 216, 361–36725. Fell, D., and Thomas, S. (1995) Biochem. J. 311, 35–3926. Thomas, S., and Fell, D. (1998) Adv. Enzyme Regul. 38, 65–8527. Korzeniewski, B., Harper, M., and Brand, M. (1995) Biochim. Biophys. Acta

1229, 315–32228. Reimer, K., and Jennings, R. (1992) in The Heart and Cardiovascular System:

Scientific Foundations (Fozzard, H., Haber, E., Jennings, R., Katz, A., andMorgan, H., eds) Second edition, pp. 1875–1973, Raven Press, New York

29. Morgan, H., Chua, B., and Russo, L. (1992) in The Heart and CardiovascularSystem: Scientific Foundations (Fozzard, H., Haber, E., Jennings, R., Katz,A., and Morgan, H., eds) Second edition, pp. 1505–1524, Raven Press, NewYork

30. Schomburg, D. (1997) Enzyme Handbook, Springer-Verlag, Berlin31. Schaefer, S., and Ramasamy, R. (1997) Cardiovasc. Res. 35, 90–9832. Vogt, A., Ackermann, C., Yildiz, M., Schoels, W., and Kubler, W. (2002)

Biochim. Biophys. Acta 1586, 219–22633. Depre, C., Vanoverschelde, J.-L., and Taegtmeyer, H. (1999) Circulation 99,

578–58834. van Schaftingen, E. (1993) Diabetologia 36, 581–58835. Stanley, W., Lopaschuk, G., Hall, J., and McCormack, J. (1997) Cardiovasc.

Res. 33, 243–25736. Hers, H. (1983) Philos. Trans. R. Soc. Lond. B Biol. Sci. 302, 27–3237. Sale, E., White, M., and Kahn, C. (1987) J. Cell. Biochem. 33, 15–2638. Yasykova, M. Y., Petukhov, S. P., and Muronetz, V. I. (2000) Biochemistry

(Mosc.) 65, 1192–119639. Cowan, K. J., and Storey, K. B. (1999) Mol. Cell. Biochem. 195, 173–18140. Cai, G. Z., Callaci, T. P., Luther, M. A., and Lee, J. C. (1997) Biophys. Chem.

64, 199–20941. Meyerhof, O. (1994) Experientia (Basel) 50, 382–38942. Clayton, C. E., and Fox, J. A. (1989) Mol. Biochem. Parasitol. 33, 73–7943. Nettelblad, F. A., and Engstrom, L. (1987) FEBS Lett. 214, 249–25244. Cooper, J. A., Reiss, N. A., Schwartz, R. J., and Hunter, T. (1983) Nature 302,

218–223

Glycolysis in Ischemic Preconditioning 24419

at NY

U School of M

edicine Library on February 6, 2015

http://ww

w.jbc.org/

Dow

nloaded from

David A. Fell and Wolfgang KüblerAckermann, Murat Yildiz, Wolfgang Schoels, Achim M. Vogt, Mark Poolman, Cordula  METABOLIC CONTROL ANALYSISPreconditioning: A STUDY EMPLOYING Regulation of Glycolytic Flux in IschemicMETABOLISM AND BIOENERGETICS:

doi: 10.1074/jbc.M201138200 originally published online May 2, 20022002, 277:24411-24419.J. Biol. Chem. 

  10.1074/jbc.M201138200Access the most updated version of this article at doi:

  .JBC Affinity SitesFind articles, minireviews, Reflections and Classics on similar topics on the

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/277/27/24411.full.html#ref-list-1

This article cites 36 references, 8 of which can be accessed free at

at NY

U School of M

edicine Library on February 6, 2015

http://ww

w.jbc.org/

Dow

nloaded from