HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 21, 18469-18476, May 24, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/21/18469    most recent M200792200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Joubert, F. Articles by Hoerter, J. A. Search for Related Content PubMed PubMed Citation Articles by Joubert, F. Articles by Hoerter, J. A. Social Bookmarking                      What's this?

³¹P NMR Detection of Subcellular Creatine Kinase Fluxes in the Perfused Rat Heart

CONTRACTILITY MODIFIES ENERGY TRANSFER PATHWAYS^*

Frederic Joubert, Jean-Luc Mazet, Philippe Mateo, and Jacqueline A. Hoerter^§

From INSERM U-446, Cardiologie Cellulaire et Moléculaire, Université Paris-Sud, Faculté de Pharmacie, 92296 Chatenay Malabry, France

Received for publication, January 24, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The subcellular fluxes of exchange of ATP and phosphocreatine (PCr) between mitochondria, cytosol, and ATPases were assessed by ³¹P NMR spectroscopy to investigate the pathways of energy transfer in a steady state beating heart. Using a combined analysis of four protocols of inversion magnetization transfer associated with biochemical data, three different creatine kinase (CK) activities were resolved in the rat heart perfused in isovolumic control conditions: (i) a cytosolic CK functioning at equilibrium (forward, F_f = PCr  ATP, and reverse flux, F_r = ATP  PCr = 3.3 mM·s¹), (ii) a CK localized in the vicinity of ATPases (MM-CK bound isoform) favoring ATP synthesis (F_f = 1.7 × F_r), and (iii) a mitochondrial CK displaced toward PCr synthesis (F_f = 0.3 and F_r = 2.6 mM·s¹). This study thus provides the first experimental evidence that the energy is carried from mitochondria to ATPases by PCr (i.e. CK shuttle) in the whole heart. In contrast, a single CK functioning at equilibrium was sufficient to describe the data when ATP synthesis was partly inhibited by cyanide (0.15 mM). In this case, ATP was directly transferred from mitochondria to cytosol suggesting that cardiac activity modified energy transfer pathways. Bioenergetic implications of the localization and activity of enzymes within myocardial cells are discussed.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Two major roles have been attributed to the creatine kinase (CK,¹ adenosine triphosphate creatine phosphotransferase, E.C. 2.7.3.2) reaction, which catalyzes the reversible exchange of high energy phosphate:

(Eq. 1)

Due to its equilibrium constant favoring ATP synthesis, CK acts as a spatial and temporal buffer for ATP. In addition, due to specific intracellular localization, CK has been suggested to play a key role in energy transfer from sites of ATP production to ATPases by a phosphocreatine (PCr)-creatine (Cr)-CK shuttle. In the myocardial cell, half of the total CK activity is present in the cytosol (cyto-CK), and the other half is accounted for by MM and mitochondrial (mito-CK) isoforms located in the vicinity of other enzymes: a particulate MM-bound CK located in close vicinity to the ATPases of myofibrils, sarcoplasmic reticulum (SR), and sarcolemma (SL) and a mito-CK isoform that is close to the adenine nucleotide translocase (ANT) in the intermembrane space (i.m.s.) of mitochondria. The vicinity of CK with other enzymes was shown to influence enzyme kinetics both in vitro (1, 2) and in skinned fiber preparations (3, 4). Mito-CK is the most obvious example of an exclusive localization in the restricted i.m.s. of the mitochondria where mito-CK octamers and their vicinity to ANT and porin have been proposed to efficiently channel energy from the mitochondria to cytosol (5, 6). The functional consequence of this localization was first evidenced in vitro where the activity of mito-CK changes the apparent affinity of respiration for externally added ADP in isolated mitochondria and saponin-skinned fibers (7-9). In other words there is a restriction of the outer mitochondrial membrane permeability to ADP. Although there is no physical membrane to restrict MM-CK isozyme in myofibrils, SR, or SL, the localization and activity of MM-CK modifies the kinetics of the myofibrillar ATPases (4), the SR Ca-ATPase (10), the Na,K-ATPase (11), and SL ionic channels (12). Thus, in the highly organized structure of an adult myocardial cell, the localization of enzymes at distinct subcellular sites may be regulated in different ways, play distinct roles, and fulfill specialized functions. This localization may contribute to cardiac efficiency by playing a crucial role in energy transfer (for review, see Refs. 13 and 14). The complex interplay of subcellular energy exchanges emphasizes the need for specific information on the subcellular energetic exchanges in the whole heart. Several mathematical models have been proposed to evaluate the subcellular CK fluxes in the whole organ (15-18). However, such theoretical approaches rely on several assumptions (the pre-knowledge of the structure of the exchange scheme, the extrapolation of the CK kinetics from in vitro data, a restriction in ADP diffusion at the outer mitochondrial membrane, and so on). Our aim was to develop an experimental approach that does not rely on these assumptions.

Up to now intracellular compartmentation has been mostly neglected in NMR analysis of CK kinetics in the whole organ. As already pointed out by Wallimann (19), the understanding of CK function might be greatly limited when considering the cell as a homogenous system where enzymes and metabolites have uniform distributions and concentrations. Although the necessity of considering metabolic compartmentation has been previously questioned in NMR analysis of the skeletal muscle (20), its importance was earlier proposed in the heart (21, 22) but hardly experimentally explored. Neglecting the presence of mitochondrial compartments in the NMR analysis indeed results in errors on the determination of CK fluxes (23). A major challenge in the interpretation of NMR data in the whole organ is thus the choice of a kinetic model of analysis. When a single magnetization transfer protocol is performed only simple schemes can be used in the analysis, and the complex cellular organization is beyond the potentiality of NMR. To overcome this limitation, we recently described a new experimental approach that allows the demonstration of the different kinetics of CK localized in subcellular compartments (24). Our strategy was to find the best scheme of energetic pathways describing the subcellular exchanges by comparing the time evolution of ATP and PCr magnetization under four NMR inversion protocols with their theoretical evolution predicted from the solution of Bloch equations in various structures of exchange between compartments. We used magnetization transfer of the phosphate moiety as a tracer of myocardial energetic fluxes. Such a combined NMR analysis allowed us to estimate the mitochondrial metabolite compartments and to identify subcellular CK fluxes in a heart perfused in control conditions (24). However, due to the high number of variables in complex schemes of exchange, the NMR data per se were still insufficient to resolve all unidirectional subcellular fluxes and pools. In this study, we propose to increase the amount of information by using biochemical data as additional constraints. These include the size of the mitochondrial ATP and PCr compartments measured by subcellular fractionation in non-polar solvent and the mitochondrial flux of ATP synthesis estimated from oxygen consumption.

This approach was applied to the perfused rat heart contracting in isovolumic conditions to characterize the pathways of energy transfer. This allowed the experimental validation of a major role of the PCr-CK-Cr shuttle in the energy transfer of a myocardium working under medium load. Interestingly, under partial inhibition of mitochondrial ATP synthesis, which impairs contractility, we found a shift in the pathways of energy transfer toward a direct transfer of ATP suggesting that the pathways of energy transfer are modulated by cardiac activity.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Physiology

This investigation conforms with INSERM guidelines defined by the guiding principles of the European Community in the care and use of animals and by the French decree No. 87/84, 1987. Authorization to perform animal experiments was obtained from the French "Ministère de l'Agriculture de la Pêche et de l'Alimentation" (No. 7473, 1997). Hearts of Wistar male rats (350-450 g) were perfused by the Langendorff technique at a constant flow. A latex balloon inserted in the left ventricle was inflated to isovolumic conditions of work, allowing the recording of contractile parameters. The HEPES-buffered perfusate contained sodium acetate (10 mM) as oxidative substrate to minimize the activity of glycolysis. Contractility was characterized by the mean coronary pressure, left ventricular systolic pressure (LVP), end diastolic pressure (EDP), heart spontaneous frequency (HR), and rate pressure product (RPP = HR × LVP in 10⁴ mm Hg·beats·min¹). For each heart the oxygen consumption (QO₂) was inferred from the relationship between contractility and QO₂ as described previously (25). Besides the control group (n = 20), we designed a steady state condition of partial decrease in ATP synthesis by a low cyanide concentration (CN group, n = 17). A stock solution of sodium cyanide dissolved in HEPES buffer (pH 7.35) was prepared just before the experiment and infused at a final concentration of 0.15 mM in the perfusate.

At the end of the NMR experiments all hearts were freeze-clamped. Part of the frozen hearts were used to measure ATP, PCr, and creatine contents (in nmol·mg of protein¹) to calculate the metabolite concentrations during magnetization transfer. All data are expressed in mmol·liter¹ of intracellular water assuming 2.72 µl of H₂O·mg of protein¹ and 160 mg of protein·g wet weight¹.

NMR

NMR Protocols-- ³¹P NMR spectra were acquired at 161.93 MHz on an INOVA Varian wide bore magnet in a 20-mm-diameter tube as described previously (23). Control spectra were obtained with 80° pulse angle, 4 K data points acquisition, a spectral width of 10,000 Hz, and a line broadening of 20 Hz. Fully relaxed spectra (repetition time, 10 s) were acquired before and after each inversion experiment. Selective inversion of either PCr (inv-PCr) or ATP (inv-ATP) was achieved by a sinc pulse of 15 ms followed by a variable mixing time (13 mixing times ranging from 0 to 10 s) before the sampling pulse and a 10-s delay for complete relaxation. Acquisition required 24 scans (four scans cycling six times through the whole protocol to minimize eventual metabolic changes). To mask the contribution of ATP-P_i exchange, inv-PCr and inv-ATP protocols were additionally performed with a continuous saturation of P_i resonance by a selective pulse. For each inversion protocol four to five hearts in similar metabolic and contractile steady state conditions were used. For each heart the magnetization of ATP and PCr was followed over time.

Data Analysis-- The method of a combined analysis of several protocols of inversion transfer (inv-) was recently described in detail (24). Fig. 1 summarizes the strategy of analysis used to select the minimal kinetic scheme(s) of exchange best describing the NMR data.

View larger version (26K): [in this window] [in a new window]   Fig. 1.   Strategy of analysis of the NMR data. The strategy of analysis is based on the comparison of NMR data (combination of four different inversion protocols) with the theoretical time evolution of magnetization in a specific scheme of subcellular exchanges. Additional biochemical data (size of the mitochondrial metabolite compartment and of the mitochondrial ATP synthesis flux) used as constraints allowed the identification of both forward and reverse fluxes in subcellular compartments. In each experimental condition, a minimal exchange scheme best describing the NMR data was chosen from the best data fit (lowest min-2 value).

Briefly, the averaged (±S.D.) time evolutions of -ATP and PCr magnetization measured in each of the four inversion protocols were combined in a single set of data. Several schemes of exchange of the phosphorus species describing the pathways of energy synthesis and utilization were investigated. There was no a priori selection of exchange schemes, but only those conferring an organization of compartments compatible with the current knowledge were investigated. Then the theoretical responses of magnetization perturbations induced by the four inversion protocols were computed from the specific Bloch equations describing each exchange scheme. Experimental and computed data were compared. The parameters (unidirectional fluxes, times of relaxation T₁ of ATP and PCr relaxation, and size of MM-bound compartment, when considered) were adjusted to best fit the measured responses of -ATP and PCr of the four inversion protocols. The quality of the fit was evaluated by the ² function given by:

(Eq. 2)

where t_i, ATP_i, and PCr_i are the averaged experimental data points for the various mixing times (ranging from 1 to N) with their own standard deviation _i, and the functions f and g are the calculated values of the ATP_i and PCr_i magnetizations. For each scheme we considered parameters as optimal when ² reached its lowest value (min-²) and the confidence intervals were small. The best minimal scheme(s) of exchange was selected by comparison of the min-² values.

We imposed as constraint to the system the global concentrations of metabolites PCr, ATP, and P_i as assessed by NMR and the equality of the global forward and reverse fluxes (steady state conditions of all metabolites) as expected in a steady state organ as described previously (24). To determine new parameters of interest, in particular both forward and reverse fluxes of the various CKs, the additional constraints were (i) the value of mitochondrial ATP synthesis flux as estimated from the oxygen consumption measured in parallel experiments outside the magnet (25), and (ii) the size of the mitochondrial ATP and PCr compartments previously measured in the same physiological conditions both by NMR and by subcellular fractionation in non-aqueous medium. Mitochondrial ATP amounted to 23 and 16% of total metabolite content in control and CN group, respectively, and mitochondrial PCr amounted to 14 and 7% of total, respectively (26).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Characteristics of the Hearts-- All experiments started after a 10-min equilibration in isovolumic working conditions. Fig. 2, a and b, shows typical control and cyanide experiments. The metabolite concentrations measured by NMR as well as the contractile characteristics of the hearts remained constant and similar to control throughout the duration of the experiment in the control heart. Both experimental groups were considered in steady state since the changes in metabolite concentrations occurring during the inversion transfer in CN were at least 3 orders of magnitude slower than the kinetics of the fastest reaction (CK) resolved by the NMR protocols. In each physiological condition, contractile and metabolic characteristics were similar in the four inversion protocols (inv-PCr, inv-ATP, inv-PCr-satP_i, and inv-ATP-satP_i), thus data were pooled (Table I). Partial inhibition of ATP synthesis by low CN concentration resulted in about 60% decrease in contractility and in a moderate rise in EDP. As expected PCr dropped, P_i rose, and ATP decreased by 20%. Intracellular pH remained similar in both groups due to the absence of glucose in the perfusate. For both groups, metabolite contents were similar to those previously observed in hearts used for the determination of mitochondrial fractions (26).

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Time course of the experiment: contractility and metabolic contents in a representative control and cyanide heart. Top panels, original recordings of contractility estimated by RPP (in 104·mm Hg·beats·min1). Bottom panels, evolution of PCr, ATP, and Pi concentrations measured in steady state spectra. Inversion transf. corresponds to the period of flux measurement by inversion. a, control group; b, CN group. Hatched bar, continuous infusion of sodium cyanide at a final concentration of 0.15 mM.

View this table: [in this window] [in a new window]   Table I Contractile and metabolic characteristics of hearts Metabolic concentrations (PCr, ATP, and Pi) expressed in mmol·liter1 of H2O, LVP, coronary pressure (CP), and rise in EDP in mm Hg (at the beginning of the experiment EDP = 5 mm Hg) are shown. Heart rate is in beats·min1, the rate pressure product (RPP = LVP × heart rate) is in 104·mm Hg·beats·min1, and QO2 is in µmol of O2·g wet weight1·min1. ATP synthesis calculated from QO2 was 2.3 ± 0.1 mmol·liter1·s1 for the control hearts and 1.1 ± 0.1 mmol·liter1·s1 for CN (assuming a phosphate/oxygen ratio of 3 and an intracellular volume of 435 µl of H2O·g wet weight1).

Analysis of the NMR Experimental Data and Selection of the Best Minimal Exchange Scheme-- A series of spectra obtained during a protocol of inv-ATP and the time evolutions of PCr and -ATP magnetization are shown for a representative control heart (Fig. 3, a and b, respectively). The averaged time evolution of PCr and ATP magnetization (mean ± S.D. of four to six perfused hearts for each inversion protocol) was queued in a single set of data for global analysis (Fig. 3c for the control and Fig. 3d for the CN group).

View larger version (32K): [in this window] [in a new window]   Fig. 3.   Inversion transfer experimental and analysis protocols. a, stacked plot of NMR spectra corresponding to an inversion of -ATP protocol in a control heart. Mixing time increasing from 0 to 10 s is shown. b, analysis of the -ATP inversion shown in panel a. Magnetization intensity of PCr and -ATP in arbitrary units (A.U.) as a function of the mixing time is shown. c and d, protocol of the simultaneous analysis of the four inversion transfer protocols. All protocols of inversion are queued in time for the global analysis. The sequence of inversions is shown on top: i-P is the inversion of PCr, i-A is the inversion of ATP, and then i-P and i-A were performed under continuous saturation of Pi resonance. The measured time evolution of magnetization intensities is shown by the symbols. For each protocol, four to five hearts were measured: the average (±S.D.) magnetization intensities measured for -ATP are represented by  and for PCr by . The lines show the fitted values of PCr, ATP, and Pi magnetization intensities resulting from the analysis in a specific scheme of exchange. c, control group (exchange scheme 4); d, cyanide group (exchange scheme 2). Saturation of Pi decreased ATP magnetizations in both groups. Notice, however, that saturation of Pi hardly affected PCr magnetization in CN.

Then the theoretical response of PCr and ATP magnetization was computed for the various schemes of metabolite exchange increasing in complexity from three to six subcellular metabolite compartments (Fig. 4). The cell was first considered to behave as a well mixed solution without compartmentation of metabolites: scheme 1 describes the energetic exchange as a global CK flux and an exchange of ATP with P_i. Then a mitochondrial ATP compartment (ATP₂) was considered either exchanging directly with cytosolic ATP, ATP₁ (scheme 2), or through mitochondrial CK (scheme 3). Scheme 4 considers three specific CK fluxes (MM-bound, mito-, and cyto-CK) with three ATP compartments (including ATP₃, the ATP compartment at the level of the consuming sites: myofibrils, SR, and SL).

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Main schemes of kinetic exchange studied. Numbers in subscript refer to kinetic compartments: 1 = cytosolic; 2 = mitochondrial, 3 = ATP-consuming sites. Scheme 1, the cell is homogenous with one ATP compartment and one CK flux; scheme 2, two compartments of ATP (cytosolic, ATP1; and mitochondrial, ATP2) and a unique CK flux, mitochondrial ATP is transported to cytosol by ATP2  ATP1; scheme 3, same as scheme 2 but two CK fluxes (cyto- and mito-CK); scheme 4, three ATP compartments (cytosolic, ATP1; mitochondrial, ATP2; and close to ATPases, ATP3) and three CK (cyto-CK, mito-CK, and MM-bound CK) fluxes. PCr2, mitochondrial PCr compartment.

Control Hearts-- The comparison of the min-² values of each exchange scheme, shown in Fig. 5a, allowed us to propose a minimal structure of exchange best describing the NMR data. In the simplest scheme 1, a homogenous cell exchanging phosphorus between ATP, PCr, and P_i, the min-² value was 324 (confidence interval, c.i. = 4). The introduction of a mitochondrial ATP compartment, ATP₂, exchanging directly with cytosolic ATP (scheme 2) clearly degraded the fit (min-² value of 724, c.i. = 5). In contrast the fit significantly improved (min-² = 237, c.i. = 7) when ATP₂ exchanged with PCr via the mito-CK reaction (scheme 3). Scheme 4, which considered the presence of three ATP compartments (ATP₁, cytosolic ATP; ATP₂, mitochondrial ATP; and ATP₃, ATP bound close to myofibrils, SR, and SL) and their exchange via three CK reactions, provided the best data fit (min-² = 149, c.i. = 10). Thus, in our control conditions, the consideration of three kinetically distinct CK reactions (scheme 4) provided the best kinetic scheme describing the NMR experimental data.

View larger version (12K): [in this window] [in a new window]   Fig. 5.   Comparison of the data fit in the various scheme of kinetic exchange. The min-2 values (and their confidence intervals) for each exchange scheme described in Fig. 4 are shown. a, control; b, cyanide.

Table II shows the dynamic parameters obtained for the control hearts in the various exchange scheme. The global CK flux was in the range previously measured by conventional NMR protocols at similar contractile performance (27, 28). Its value had a tendency to increase (from 7 to 9 mmol·liter¹·s¹) when CK subcellular localization was considered in the analysis. Relaxation parameters of ATP and PCr were hardly affected by the scheme structure. Cytosolic T₁ values, which ranged from 3.6 to 5.4 s for PCr and 0.7 to 0.8 s for ATP, were also in agreement with published data. The most striking result was the clear difference in subcellular CK kinetics depending on their localization. When mito-CK was considered (schemes 3 and 4), no significant forward mitochondrial CK flux (ATP synthesis) could be evidenced. In scheme 4, selected from the ² analysis as our best minimal exchange scheme, the flux F_f of ATP synthesis was negligible (F_f = 0.3 mmol·liter¹·s¹, c.i. = 0.6) compared with the flux of PCr synthesis (F_r = 2.6 mmol·liter¹·s¹), which accounted for about 27% of the total reverse CK flux. On the opposite, at the level of MM-bound CK, both forward and reverse fluxes were detected. The reaction slightly favored ATP synthesis (F_f was 1.7-fold F_r). The ATP compartment (ATP₃) exchanging via MM-bound CK amounted to 15% (c.i. = 11) of total cellular ATP. Cytosolic CK was at equilibrium and accounted for about one-third of the total CK flux.

View this table: [in this window] [in a new window]   Table II Dynamic parameters obtained in each model tested for the control condition Ff refers to the flux PCr  ATPx and Fr to ATPx  PCr of each compartment. Global = total forward or reverse CK flux. For each scheme of exchange the parameters of the best fit are shown as the value and confidence interval in parentheses. Scheme numbers refer to Fig. 4. The ATP-Pi exchange was considered in all cases. The constraints imposed were: global metabolite ATP, PCr, and Pi concentrations (from Table I); global steady state for each metabolite (*); equality of global Ff and Fr CK flux; flux Pi ATP2 = 2.3 mmol·liter1·s1 (estimated from oxygen consumption); and size of mitochondrial compartment: ATP2 = 23% and PCr2 = 14% (in percentage of total cellular content) in schemes 2-4. In scheme 4, which provided the best data fit (in bold), the ATP3 compartment amounted to 15% of total ATP (c.i. = 11). PCr1, cytosolic PCr compartment; PCr2, mitochondrial PCr compartment.

Partial Inhibition of ATP Synthesis (by Low Cyanide)-- Fig. 5b shows the min-² values of the CN group. As for control conditions, scheme 1, which assimilates the cell to a homogenous compartment, was not the best scheme of analysis (min-² = 261, c.i. = 4). The best quality of fit was observed for a direct exchange of mitochondrial to cytosolic ATP (scheme 2, min-² = 145, c.i. = 5). In contrast with the control hearts, neither the introduction of a mito-CK flux (scheme 3) nor that of an additional MM-bound CK (scheme 4) improved the fit. Indeed the analysis did not allow the separation of the contribution of MM-bound CK and mito-CK to the global CK flux: the confidence interval was 3-5-fold higher than their flux value (not shown). Thus the description of energetic phosphorus exchanges in the cyanide group did not require the distinction of kinetically different CKs.

Despite the fact that we choose a non-glycolytic substrate, acetate, to minimize the contribution of glycolytic ATP synthesis, partial inhibition of respiration could activate glycogenolytic ATP synthesis. Neglecting this additional ATP synthesis flux would result in an underestimation of the flux of ATP hydrolysis. To evaluate the influence of this potential pitfall we imposed an additional cytosolic ATP synthesis (P_i  ATP₁ flux) to the exchange scheme 2. Increasing the global ATP synthesis from 20 to 80% by the addition of P_i  ATP₁ flux resulted in a progressive degradation of the fit (the min-² progressively rose from a value of 145, in the absence of this exchange, to 190). This suggests that in our condition of moderate mitochondrial ATP synthesis inhibition by low CN concentration glycogenolytic contribution to ATP synthesis was negligible.

In the exchange scheme 2, best describing the experimental condition of partial inhibition of respiration, the relaxation parameters T₁ (3.8 s, c.i. = 1.7 and 0.4 s, c.i. = 0.1 for PCr₁ and ATP₁, respectively, and 6.3 s, c.i. = 4 for PCr₂) were similar to those observed in the control condition except for an increase in T_1ATP2 (3.5 s, c.i. = 2). Forward and reverse CK fluxes equaled 6.7 mmol·liter¹·s¹ (c.i. = 0.8). Although we did not previously observe any relation between CK flux and contractility, the global CK flux, measured here in CN, had a tendency to be lower than in the control when CK compartmentation was taken into account in the latter. The CN CK flux value was, however, similar to the global forward CK flux previously measured in the same CN condition by time-dependent saturation transfer technique (27).

Cardiac Activity Modifies the Pathway of Energy Transfer-- The kinetic compartmentation of creatine kinases observed in control hearts and the disappearance of mito-CK activity in CN hearts suggest, as a consequence, that the pathway of energy transfer between mitochondria and sites of ATP utilization differs with cardiac activity. Fig. 6 proposes a schematic representation of energy transfer in both experimental conditions.

View larger version (22K): [in this window] [in a new window]   Fig. 6.   Pathways of subcellular exchange change with cardiac function. Schematic representation based on the minimal kinetic models selected for control (scheme 4) and during partial inhibition of respiration (scheme 2). Single line arrows, unidirectional fluxes; double line arrows, net diffusion fluxes. a, in control, three kinetic compartments corresponding to mitochondria, cytosol, and ATP-consuming sites are shown. The ATP extruded from matrix in i.m.s. by ANT displaced the mito-CK reaction in the direction of PCr synthesis. PCr ensured the transfer of energy to the ATP-consuming sites. b, in cyanide a unique CK at equilibrium, ATPim, is directly exported to a global compartment corresponding to cytosol and ATP-consuming sites. cyt, cytosol; im, inner membrane; m, matrix; myof., myofibrils.

In control heart, the displacement of mito-CK from equilibrium (Table II) resulted in a net flux of PCr synthesis, and PCr was the phosphorus species freely diffusing from mitochondria to cytosol and ATP-consuming sites (Fig. 6a). Notice, however, that, although an exclusive diffusion flux of ATP between mitochondria and cytosol in the absence of mito-CK activity was clearly rejected as the worse scheme of exchange (as shown in Fig. 4, scheme 2), this does not exclude the possibility that both PCr and ATP efflux from mitochondria could participate in myocardial energy transfer. Because it was impossible to simultaneously resolve both pathways due to the increase in unknown parameters, we attempted to evaluate this hypothesis by computation. In scheme 4, we imposed as a constraint a proportion of direct ATP transport from mitochondria to cytosol at the expense of PCr extrusion and followed the influence of this constraint on the quality of the fit. A transfer of ATP₂  ATP₁, increasing from 0 to 100% of mitochondrial ATP synthesis, did not significantly improve min-². Other possible exchanges of ATP between compartments were also considered. Min-² was unaffected by a direct exchange of ATP₂  ATP₃ (mitochondrial to ATPase compartment) or by a simultaneous activity of ATP₂  ATP₁ and ATP₁  ATP₃ (mitochondria to cytosol and cytosol to ATPases, respectively), whereas an exclusive ATP₁  ATP₃ transport degraded the quality of the adjustment. Although this computation approach did not exclude the possibility of a direct ATP diffusion from mitochondria to cytosol, its negligible influence on the adjustment of the data was in line with a major role of mito-CK in the transfer of energy in the control heart.

When cardiac function was impaired by partial inhibition of ATP synthesis, a unique CK functioning at equilibrium was sufficient to describe the NMR data. As an interesting consequence, mitochondrial ATP extruded by the adenine nucleotide translocase in i.m.s. was directly exported to cytosol (Fig. 6b) by a flux, ATP₂  ATP₁, equal to the ANT flux. This shift from a transfer of energy by phosphocreatine in control isovolumic condition of work to the direct extrusion of ATP in CN suggests that the pathways of energy transfer depend on the cardiac activity: high cardiac activity might recruit the PCr-Cr-CK shuttle.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Our results emphasize that NMR spectroscopy has the potency to assess metabolic dynamic compartments and quantify subcellular exchanges in the whole organ if the strategy of analysis is optimized. In control hearts, we confirmed our previous suggestion that subcellular CKs have different kinetic behavior depending on their localization (24) and showed the consequence of the vicinity of CK with other enzymes on their activity in the whole organ. We also observed that kinetic compartmentation depends on cardiac function. To our knowledge, this is the first experimental evidence supporting the role of the PCr-Cr-CK shuttle in the beating heart and suggesting that the pathways transferring energy from mitochondria to sites of energy utilization may depend on cardiac activity.

Here our purpose was to investigate this kinetic compartmentation in the whole beating heart. The interest of our new experimental strategy, which combines analysis of NMR and biochemical data, is to allow for the first time the quantification of all unidirectional fluxes of the subcellular CKs in the whole heart. In a control heart performing a medium level of work, cytosolic CK functioned at equilibrium, and mito-CK was purely dedicated to PCr synthesis (Fig. 6a). Such kinetic behavior of mito-CK in the perfused heart is consistent with the early observation of Jacobus and Saks (8) showing in the isolated mitochondria that an active mitochondrial respiration displaced mito-CK from its equilibrium. The consequence of the negligible forward mito-CK flux observed here is a net flux of PCr synthesis. This is in line with the suggestion that a complex of ANT, mito-CK octamers, and porin might channel energy from the mitochondrial matrix to cytosol (13). The transport of energy from the mitochondria to ATP-consuming sites occurs via PCr diffusion: that is the PCr-Cr-CK shuttle as earlier proposed (29, 30).

Interestingly alteration of heart function modified the pathway of energy transport toward a direct extrusion of mitochondrial ATP to cytosol (Fig. 6b). The kinetic compartmentation of CKs, evidenced in control conditions, disappeared with the partial inhibition of ATP mitochondrial synthesis. Two possible hypotheses can account for this observation: either mito-CK was not functional (the ATP produced by mitochondria cannot be used by mito-CK and is directly transfer to cytosol) or mito-CK cannot be kinetically separated from the cytosolic CK. The fact that mito-CK appeared as uncoupled from the translocase activity could directly result from an inhibition of mito-CK, for example, by reactive oxygen species production (CK activity is highly sensitive to reactive oxygen species (31)), or by a change in its degree of octamerization as observed in ischemia (32). Besides, alteration in mito-CK flux might merely result from a change in kinetic factors. The concentration of substrates and products in the vicinity of mito-CK could indeed be altered by the decrease in translocase activity, by a change in the i.m.s. volume resulting from mitochondrial matrix contraction as recently observed in isolated mitochondria in the presence of CN (33), or by an alteration of the outer membrane permeability for adenine nucleotides as observed in models of ischemic reperfusion (34). Our data suggest that the control strength exerted by the mito-CK and by the ANT on the energy transfer depends on contractile activity. The origin of this shift in metabolic control requires further study.

The complexity of cell structure and the specific behavior of mito- and MM-bound CK has been underestimated in the field of NMR spectroscopy where the cell had mostly been considered as a unique compartment of CK at equilibrium (20, 28). As a consequence of this assumption, the free ADP concentration in vivo is estimated from the total concentrations of ATP, PCr, and Cr and the global CK equilibrium constant. The knowledge of free [ADP] is indeed essential to the understanding of cellular energetics in as much as its micromolar concentration is in the range activating respiration and inhibiting ATPases. Besides, ADP is a major determinant of the affinity for ATP hydrolysis (A_ATP). However, since the early observation of Chance in isolated mitochondria (46), attempts to reveal a control of respiration by free ADP in various muscles in vivo has led to contradictory reports. Both in vivo and in the isolated muscle, NMR investigations suggested that respiratory control by ADP with the operation of CK at equilibrium was sufficient to explain the regulation of respiration in the fast glycolytic skeletal (35, 36) but not in slow oxidative muscles (37). Neglecting CK compartmentation would indeed have a minor functional consequence in a fast muscle where 95% of total CK is cytosolic (and mito-CK is at most 2%) but might result in increasing errors in slow muscle as the contribution of mito-CK to total CK activity rises. The extreme case is indeed the myocardium where mito-CK amounts to 25% of total CK, and half of CK total activity is co-localized with enzymes. Thus, assuming that free [ADP] in the i.m.s. equals that calculated from the global metabolite concentrations can be a major drawback in the understanding of cardiac respiratory control. Indeed in vivo wide changes in cardiac work occur without alteration in the total P_i, ATP, and PCr (and thus of free ADP) concentrations (38-40). This led to the proposal that, in vivo, ADP could not be a signal responsible for the continuous coordination of myofibrillar and mitochondrial cardiac activities and promoted the search for other signals. Besides, in a normoxic perfused heart, experimental manipulation of PCr and ATP contents leading to a rise in the global [ADP] (calculated in the hypothesis of CK equilibrium) and a drop in A_ATP induces no major functional alteration (41, 42). This can indeed be understood if the concentration of [ADP] in the vicinity of ANT and ATPases differs from that in the bulk cytosol. The fact that, in control hearts, we could experimentally separate the kinetics of cytosolic, mitochondrial, and MM-bound CK strongly suggests that the concentrations of CK substrates and products indeed differ according to CK localization. The free ADP concentrations in i.m.s. and ATPase compartments cannot, however, be directly calculated because the mito-CK flux was displaced from equilibrium by ANT activity, and MM-bound CK forward and reverse fluxes were unequal. However, it should be possible to model the metabolite concentrations in the different compartments using our NMR analysis, the known kinetic characteristics of CK, ANT, and ATPases, and the volume distribution of the intracellular compartments. This is a promising approach for a new assessment of cardiac respiratory control in the whole heart.

Optimal energy transfer is a fundamental component of the continuous adequation of ATP synthesis and utilization characteristic of the healthy adult myocardium. Indeed most human and animal cardiac pathologies (dilated hypertrophy, diabetes, heart failure, and genetic cardiomyopathy) are associated with alterations of mito-CK expression or activity (43-45). We believe that our strategy of analysis will prove to be valuable to explore the pathways of energy transfer in these models and could contribute to a better understanding of the implication of bioenergetics in the development of cardiac failure.

	ACKNOWLEDGEMENTS

The NMR experiments were performed in collaboration with B. Gillet and J.-C. Beloeil (Résonance Magnétique Nucléaire biologique, Institut de Chimie des Substances Naturelles, CNRS, Gif/Yvette, France). We thank R. Fischmeister for continuous support and help in revising the manuscript as well as R. Ventura-Clapier.

	FOOTNOTES

^* The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

This work is dedicated to W. E. Jacobus, who initiated the interest (of J. A. H.) in CK and NMR.

Recipient of a grant from the French Ministère de la Recherche.

^§ To whom correspondence should be addressed: U-446 INSERM, Cardiologie Cellulaire et Moléculaire, Faculté de Pharmacie, 5 rue J.B. Clément, 92296 Chatenay-Malabry, France. Tel.: 33-1-46835759; Fax: 33-1-46835475; E-mail: jacqueline.hoerter@cep.u-psud.fr.

Published, JBC Papers in Press, March 8, 2002, DOI 10.1074/jbc.M200792200

	ABBREVIATIONS

The abbreviations used are: CK, creatine kinase; ANT, adenine nucleotide translocator; ATP₁, cytosolic ATP; ATP₂, mitochondrial ATP; ATP₃, ATP in the vicinity of the ATP-consuming sites; c.i., confidence interval; Cr, creatine; cyto-CK, cytosolic CK; EDP, end diastolic pressure; F_f, forward CK flux, unidirectional flux of ATP synthesis by CK; F_r, reverse flux, unidirectional flux of PCr synthesis by CK; i.m.s., intermembrane space between the outer and inner mitochondrial membrane; inv- (inv-PCr or inv-ATP), inversion (of PCr or of -ATP), protocol of magnetization transfer experiment; LVP, left ventricular systolic pressure; min-², minimal ², estimation of the quality of the adjustment; mito-CK, mitochondrial CK isoform; MM-bound CK, MM creatine kinase isoform localized in the vicinity of ATP-consuming sites; PCr, phosphocreatine; QO₂, cardiac oxygen consumption; RPP, rate pressure product (estimation of contractility); satP_i, continuous saturation of inorganic phosphate resonance; SL, sarcolemma; SR, sarcoplasmic reticulum; T_1x, spin lattice relaxation time of species x.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				C. J. Barclay Getting energy to where it is required is a problem in the failing heart J. Physiol., November 1, 2008; 586(21): 5037 - 5038. [Full Text] [PDF]

				K. Carvajal, E. Zarrinpashneh, O. Szarszoi, F. Joubert, Y. Athea, P. Mateo, B. Gillet, S. Vaulont, B. Viollet, X. Bigard, et al. Dual cardiac contractile effects of the {alpha}2-AMPK deletion in low-flow ischemia and reperfusion Am J Physiol Heart Circ Physiol, June 1, 2007; 292(6): H3136 - H3147. [Abstract] [Full Text] [PDF]

				V. A. Selivanov, S. Krause, J. Roca, and M. Cascante Modeling of Spatial Metabolite Distributions in the Cardiac Sarcomere Biophys. J., May 15, 2007; 92(10): 3492 - 3500. [Abstract] [Full Text] [PDF]

				M. S Parihar and G. J. Brewer Mitoenergetic failure in Alzheimer disease Am J Physiol Cell Physiol, January 1, 2007; 292(1): C8 - C23. [Abstract] [Full Text] [PDF]