| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of 1Biochemistry and 2Nutrition, Université de Montréal, and 3Centre de Recherche, Centre Hospitalier de l'Université de Montréal, Hôpital Notre-Dame, Montréal, Québec, Canada H2L 4M1
Submitted 25 July 2003 ; accepted in final form 14 August 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
20%) of cardiac output and efficiency with unaltered O2 consumption. At the metabolic level, beyond the expected effects of high fatty acid levels on the contribution of pyruvate decarboxylation (reduced >3-fold) and 
-oxidation (enhanced 3-fold) to citrate synthesis, there was also a 2.4-fold lowering of anaplerotic pyruvate carboxylation. Despite the dual inhibitory effect of high fatty acids on pyruvate decarboxylation and carboxylation, tissue citrate levels were twofold higher, but citrate release rates remained unchanged at 11–14 nmol/min, representing <0.5% of citric acid cycle flux. A similar trend was observed for most metabolic parameters after oleate or octanoate addition. Together, these results emphasize a differential modulation of anaplerotic pyruvate carboxylation and citrate release in the heart by fatty acids. We interpret the lack of effects of high fatty acid concentrations on citrate release rates as suggesting that, under physiological conditions, this process is maximal, probably limited by the activity of its mitochondrial or plasma membrane transporter. Limited citrate release at high fatty acid concentrations may have important consequences for the heart's fuel metabolism and function.
citric acid cycle; isotopomer analysis; cardiac energy metabolism
-oxidation after its conversion to malonyl-CoA, an inhibitor of carnitine palmitoyltransferase-I (CPT-I) (13, 32, 35). The regulatory role of cytosolic citrate in the heart has, however, been controversial because of the low activity of its mitochondrial citrate transporter (9, 39), the participation of which is necessary for citrate transfer from the mitochondria to the cytosol. A better understanding of factors regulating cardiac citrate release might explain why patients suffering from cardiac diseases show a modified substrate oxidation profile associated with greater citrate and lactate release rates (47, 48). Furthermore, it may suggest interventions that favor the conversion of cytosolic citrate to malonyl-CoA and, hence, inhibit LCFA -oxidation. A number of studies have underlined the benefit of shifting the heart's preference from fatty acid to carbohydrate use for improved cardiac efficiency in the setting of ischemia (37, 40).
Over the past years, we conducted a series of studies using 13C-labeled substrates and isotopomer analysis by gas chromatography-mass spectrometry (GC-MS) in which we assessed substrate fluxes through metabolic pathways related to citrate synthesis, utilization, and release. These were achieved in various study models, namely, ex vivo Langendorff-perfused rat hearts (11, 12, 32, 49), in situ infused pig hearts (29, 30), and ex vivo work-performing rat hearts (50). Results that appear to support a potential metabolic link between anaplerotic pyruvate carboxylation, mitochondrial citrate efflux, and cytosolic cleavage of citrate to malonyl-CoA can be summarized as follows. Rat hearts perfused ex vivo, in the Langendorff or working mode with 13C-labeled substrates, constantly released small quantities of citrate (nmol) with a 13C-labeling pattern that is similar to mitochondrial citrate. This finding suggested that citrate release reflects its mitochondrial efflux (11, 12, 49, 50). The specificity of the process was supported by the lack of correlation between citrate release rates and indexes of O2 deprivation or cellular membrane damage, as well as the effect of the specific inhibitor of the citrate transporter 1,2,3-benzenetricarboxylic acid (49, 50). Consistent with the activity of the cardiac citrate transporter (9, 39), citrate release rates in the various study models and conditions ranged from 5 to 20 nmol·g–1·min–1 (12, 29, 30, 32, 49, 50), which are similar to rates reported in humans (47). In the Langendorff-perfused rat heart, citrate release rates were modulated by the heart's energy demand, O2 supply, and nature and concentration of substrates supplied for citrate synthesis, namely, anaplerotic oxaloacetate (OAA) and acetyl-CoA. The magnitude and direction of the changes observed were in agreement with a regulatory role for cytosolic citrate in fuel partitioning (49). They were also sufficient to support the measured increase in tissue malonyl-CoA levels induced by changes in substrate supply (
0.7 nmol/min) (32). Citrate release rates were augmented in work-performing spontaneously hypertensive rat hearts, an animal model of cardiac hypertrophy, compared with Wistar-Kyoto control hearts (50). In pig hearts infused in situ, however, citrate release rates were not enhanced by increased substrate supply (30). They were, however, reduced under low-flow ischemic conditions mimicking hibernation concomitantly with flux through anaplerotic pyruvate carboxylation and the CAC (29), while citrate tissue levels remained constant. These results suggested a fine interregulation between cardiac citrate efflux and anaplerosis in maintaining citrate pool size.
This study was undertaken to expand our understanding of the aforementioned potential interregulation among pathways involved in cardiac citrate synthesis and release. Furthermore, we aimed to specifically explain why citrate release was modulated by substrate supply in the Langendorff-perfused rat heart (49) but not in the pig heart in situ (30). For this purpose, ex vivo work-performing hearts were perfused under normoxia with semirecirculating buffer containing physiological concentrations of unlabeled (5.5 mM glucose) and 13C-labeled (1 mM lactate, 0.2 mM pyruvate, and 0.4 mM LCFA-oleate) substrates. We tested the impact of increasing fatty acid concentrations from 0.4 to 1 mM, mimicking fed and fasting levels, respectively (28), with the LCFA oleate or the medium-chain fatty acid (MCFA) octanoate. In contrast to oleate, octanoate -oxidation is not regulated by CPT-I (38). In parallel to the continuous monitoring of indexes of cardiac performance and cellular integrity, the following metabolic parameters related to citrate synthesis and release were documented: 1) the contribution of pyruvate decarboxylation and anaplerotic carboxylation, as well as fatty acid -oxidation, to citrate synthesis; 2) citrate release rates and CAC intermediate tissue levels; and 3) maximal tissue activities of mitochondrial enzymes catalyzing citrate synthesis and utilization, namely, citrate synthase (CS) and aconitase.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The sources of chemicals, biological products, and 13C-labeled substrates have been identified previously (11, 12, 20, 49). The albumin solution [BSA, fraction V, fatty acid-free (Intergen): 300 g of BSA in 1.5 liters of modified Krebs-Henseleit bicarbonate buffer without glucose but with 0.1 mM EDTA] was dialyzed in membranes (mol wt cut-off = 6,000–8,000) at 4°C against 25 liters of the same buffer for 28 h to reduce the background citrate concentration to the low micromolar range (1.2 ± 0.1 µM).
Heart Perfusions in Semirecirculating Working Mode
Animal experiments were approved by the local animal care committee in compliance with the guidelines of the Canadian Council on Animal Care. Male Wistar rats (300–330 g; Charles River) were provided with food and water ad libitum. The following procedure for heart isolation and its ex vivo perfusion in the working mode has been described previously in detail (50). After anesthesia by injection of pentobarbital sodium (65 mg/kg ip), the hearts were cannulated rapidly and perfused retrogradely through the aorta at a constant pressure of 70 mmHg with semirecirculating modified Krebs-Henseleit bicarbonate buffer (pH 7.4) containing 119 mM NaCl, 4.8 mM KCl, 2.5 mM CaCl2, 1.2 mM KH2PO4, 1.2 mM MgSO4, 25 mM NaHCO3, 5.5 mM glucose, 8 nM insulin, 50 µM carnitine, 1 mM lactate, and 0.2 mM pyruvate. A 10-min perfusion period was allowed for insertion of a polyethylene (PE-50) catheter through the pulmonary vein in the left ventricle. The polyethylene tubing, pulled through the ventricular wall and anchored in the apex of the heart by a fluted end, was connected to a pressure transducer for continuous monitoring of left ventricular functions. The left atrium was then cannulated through the pulmonary vein. Spontaneously beating hearts were switched to the anterograde work-performing mode and perfused for 30 min through the left atrial cannula with semirecirculating modified Krebs buffer (800 ml); they were then freeze clamped with aluminum tongs chilled in liquid N2. Aortic afterload pressure was set at 80 mmHg. Atrial preload was monitored continuously by a pressure transducer (Digi-Med Blood Pressure Analyzer, Micro-Med) and maintained at 11.5 mmHg. The volume of air in the compliance chamber was fixed at 1.5 cm3, which represents 30% of its total volume. The following functional parameters were monitored continuously during the perfusion experiments: 1) atrial inflow and aortic outflow, with calibrated electromagnetic flow probes (model FM501, Carolina Medical Electronics); 2) temperature, with a thermocouple; and 3) left ventricular contractile functions, namely, heart rate (HR), maximum left ventricular pressure (LVPmax), left ventricular enddiastolic pressure (LVEDP), and maximum value for the first derivative of LVP (+dP/dtmax), with a pressure transducer (Digi-Med Heart Performance Analyzer, Micro-Med). Free Ca2+, PO2, and PCO2 were measured in influent and coronary effluent perfusates collected on ice at 20 min with a blood gas, electrolyte, and pH analyzer (ABL 77 series, Radiometer).
Perfusion Protocols
Hearts from Wistar rats were perfused under normoxia with semirecirculating Krebs-Henseleit bicarbonate buffer containing 5.5 mM glucose, 8 nM insulin, 1 mM [U-13C3]lactate, 0.2 mM [U-13C3]-pyruvate, 50 µM carnitine, 3% albumin, 0.1 mM EDTA, and 1) 0.4 mM [1-13C]oleate (n = 8, group I), 2) 1.0 mM [1-13C]oleate (n = 8, group II), or 3) 0.4 mM oleate + 0.6 mM [1-13C]octanoate (n = 6, group III). We also conducted other sets (n = 5 in each group) of perfusion experiments with 1) [1-13C]oleate or [U-13C18]oleate (groups I and II) or 2) [1-13C]octanoate (group III) as the only 13C substrate for methodological purposes (see Flux parameters). All 13C-labeled substrates were supplied at 99% enrichment except for [U-13C18]oleate, which was supplied at 15% enrichment. The ionized Ca2+ concentration of the albumin-containing buffer was determined to be 1.8 mM. The substrate concentrations in the perfusion buffer mimic those in plasma. LCFA concentrations of 0.4 and 1 mM mimic the fed and fasting states, respectively (28). Lactate and pyruvate were added in a physiological ratio to minimize perturbations of the cytosolic redox state that occur when lactate is supplied alone. The addition of carnitine compensates for its potential loss during heart isolation (34). We previously verified that a near-isotopic steady state is reached after 20–25 min of heart perfusion (50). Under all conditions, citrate release rates (25–30 min), lactate and pyruvate uptake and efflux rates (25–30 min), and lactate dehydrogenase (LDH) release rates (5, 15, and 25 min) were quantified from influent and coronary effluent perfusate samples collected on ice at the indicated times. 13C enrichment, concentrations of the CAC intermediates [citrate, isocitrate, 
-ketoglutarate (-KG), succinate, fumarate, and malate], CAC enzyme activities (CS and aconitase), and flux ratios relevant to substrate selection for citrate synthesis, namely, pyruvate and fatty acids from 13C enrichment of the acetyl (C4+5) and OAA (C1+2+3+6) moiety of citrate, were measured in freeze-clamped powdered tissues.
Analytic Procedures
Citrate release rates were quantitated by isotope-dilution GC-MS and flow rate measurements (49). Tissue levels of CAC intermediates were determined in 100-mg tissue samples spiked with 10 nmol of [1,5-13C2]citrate, 25 nmol of [1,4-13C2]succinate, and 5 nmol of [U-13C4]fumarate, and quantification was achieved with standard curves for isocitrate, -KG, and malate. 13C mass isotopomer distribution (MID) of CAC intermediates and related metabolites in heart tissue samples (citrate and its OAA moiety, -KG, succinate, malate, and pyruvate) was determined by GC-MS assays (11, 12, 20, 49). Perfusate lactate concentrations, tissue CS and aconitase activities, and LDH release were quantified by enzymatic assay with a spectrophotometer (Cobas Fara, Roche) (3, 10, 20, 49). Protein contents were measured with a Bio-Rad kit, and BSA served as the standard. Enzyme activities are expressed as units per gram of total protein, where 1 unit of enzyme activity is defined as the amount catalyzing the conversion of 1 µmol of substrate per minute at 37°C. The 13C MIDs of lactate and pyruvate in influent and coronary effluent perfusates were determined after treatment with 1 M NaB2H4 (see below).
Myocardial O2 consumption. Myocardial O2 consumption (M
O2, µmol/min) was calculated from the product of O2 concentration (mM) differences between influent and effluent perfusates from the oxygenator and pulmonary artery, respectively, and the coronary flow rate (ml/min), 1.06 mM, was taken as the concentration of dissolved O2 at 100% saturation (42).
Functional status. The rate-pressure product (RPP) was calculated from left ventricular developed pressure [LVDP (mmHg) = LVPmax (mmHg) – LVEDP (mmHg)] and HR [RPP (mmHg·beats·min–1) = LVDP (mmHg)·HR (beats/min)]. Cardiac power [CP (mW) = cardiac output (m3/s)·LVDPmax (Pa)] and cardiac efficiency [CE (mW·µmol–1·min–1) = CP (mW)/MO2 (µmol/min)] were calculated with a conversion factor of 133.32 Pa/mmHg. The factor for conversion of measured gram wet weight to gram dry weight was determined to be 8.85 ± 0.16 (n = 29).
Flux parameters. GC-MS data are expressed as molar percent enrichment (MPE), as defined previously (11, 12, 20, 34, 49). Briefly, mass isotopomers of metabolites containing 1 to n 13C atoms were identified as Mi with i = 1, 2,..., n, and the absolute MPE of individual 13C-labeled mass isotopomers (Mi) of a given metabolite was calculated as follows
 | (1) |
To extrapolate the MPE M1 and M2 of ACCit from the measured MID of citrate and OAACit, we used different mathematical approaches that depended on the choice of 13C-labeled substrate(s) and/or perfusion conditions. For hearts perfused with [U-13C3](lactate + pyruvate) and [1-13 C]oleate or [1-13C]octanoate (groups I–III), we applied Eq. 2
 | (2) |
In Eq. 2, the MPE M1 of ACCit resulting from oleate oxidation was measured in separate experiments in which [1-13C]oleate was the only 13C-labeled substrate. These equations were used because the MPE values of ACCit calculated with Eqs. 18 and 19 of Ref. 11 were imprecise for perfusions under conditions where pyruvate carboxylation is substantial (group I). This is explained by the fact that tissue citrate was highly enriched in isotopomers of higher masses (M4–M6), whereas Eqs. 18 and 19 in Ref. 11 are based solely on the MPE M1 and M2 of citrate and OAACit. For hearts perfused with [1-13C]oleate, [U-13C18]oleate, or [1-13C]octanoate as the sole 13C-labeled substrate, we employed Eqs. 18 and 19 of Ref. 11. For groups II and III, flux ratios calculated with Eqs. 18 and 19 in perfusions with the mix of 13C-labeled substrates, lactate, pyruvate, and a fatty acid did not differ from those calculated using Eq. 2, which necessitated additional perfusions with the 13C-labeled fatty acid as the sole 13C-labeled substrate. For groups I and II, because similar oleate oxidation flux ratios were obtained with 0.4 or 1 mM [1-13C]oleate or [U-13C18]oleate, the results with these two 13C-labeled substrates were averaged.
Lactate and pyruvate uptake and efflux. Lactate and pyruvate uptake and efflux rates were determined by an adaptation of the nuclear magnetic resonance approach described recently (50). Briefly, in hearts perfused with a nonrecirculating buffer containing unlabeled glucose and [U-13C3](lactate + pyruvate), lactate produced by glycolysis from exogenous glucose and endogenous glycogen is unlabeled (M) and can be distinguished by GC-MS from [U-13C3]lactate (M3) added to the buffer. The uptake of [13C3]lactate is quantified from the difference between its influent and effluent perfusate concentration. In practice, the MID of perfusate lactate and pyruvate was determined by GC-MS in samples treated with 1 M NaB2H4. This treatment reduces pyruvate to lactate, and the four mass isotopomers of lactate obtained can be distinguished by GC-MS: [12C]lactate 
[12C]lactate (M), [13C]lactate [13C]lactate (M3), [12C]pyruvate [12C]lactate deuterated (M1), and [13C]pyruvate [13C]lactate deuterated (M4). The concentrations of unlabeled (M and M1) and 13C-labeled (M3 and M4) lactate and pyruvate were calculated from their corresponding MIDs and enzymatically determined perfusate lactate concentrations, which included unlabeled (M) and 13C-labeled lactate (M3). Efflux and uptake rates of lactate and pyruvate (µmol/min) were obtained by multiplying their perfusate concentrations in unlabeled (M and M1) and 13C-labeled (M3 and M4) (µmol/ml) isotopomers, respectively, by the coronary flow rate (ml/min).
Absolute CAC flux rate. The CAC flux rate was calculated from MO2 and the stoichiometric relations between O2 consumption and citrate formation from carbohydrates and fats. The equation of Panchal et al. (Eq. 3 of Ref. 30) was corrected and modified (W. C. Stanley, personal communication) to take into account 1) the specific contribution of exogenous oleate or octanoate to citrate formation, as assessed from the flux ratios OLE/CS and OCT/CS; 2) the contribution of other sources as assessed from the flux ratio OS/CS, which we assumed to be endogenous triglyceride stores consisting of equal proportions of oleate and palmitate (groups I and II), or exogenously supplied oleate (group III); and 3) the fact that glucose forms 0.333 µmol citrate/µmol O2 consumed instead of 0.6 µmol citrate/µmol O2 consumed, as assumed previously by Panchal et al. That is, for groups I and II, we used Eq. 3a, which considers that 1 µmol of consumed O2 results in the formation of 0.333, 0.353, and 0.348 µmol of citrate from carbohydrates (glucose, lactate, and pyruvate), oleate, and palmitate, respectively. For group III, where hearts were perfused with oleate + octanoate, we used Eq. 3b, which considers that octanoate forms 0.364 µmol citrate/µmol O2 consumed
 | (3a) |
 | (3b) |
Statistical Analysis
Values are means ± SE of 5–15 heart perfusions and were compared with those obtained for group I, which we refer to as the control condition. One-way ANOVA followed by Bonferroni's selected-comparison test was applied for statistical evaluation of the data. P < 0.05 was considered to be significant.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The values of the functional and physiological parameters measured throughout the working heart perfusion experiments in the presence of various substrate mixtures are presented in Table 1. Under all conditions examined, these parameters remained constant for the entire 30-min perfusion period. HR was stable without external pacing. LDH release rates, an index of cellular necrosis, were low (< 0.2 U/min) compared with values observed after an ischemic insult (5). Compared with the control condition, in which hearts were perfused in the presence of 0.4 mM oleate (group I), hearts perfused with 1 mM fatty acids, either oleate (group II) or oleate + octanoate (group III), showed 1) significantly lower values for cardiac output, aortic flow, and cardiac efficiency; 2) slightly lower, although not significantly different, values for RPP, LVDP, and coronary flow rates; and 3) similar values for HR and MO2. As a whole, a similar trend was observed for all parameters whether the hearts were perfused with 1 mM oleate alone or with oleate + octanoate. However, some small differences did reach significance for only one of these conditions. Intracellular pH was significantly lower for hearts perfused with 1 mM oleate, and cardiac power was significantly lower for hearts perfused with oleate + octanoate.
|
13C Enrichment and Flux Data Relevant to Citrate Synthesis
Table 2 reports the MIDs of CAC intermediates isolated from Wistar rat hearts perfused with [U-13C3](lactate + pyruvate) and [1-13C]oleate or [1-13C]octanoate. From these 13C enrichment data, we emphasize the following points. First, for hearts perfused with 0.4 mM oleate (group I), tissue CAC intermediates were highly enriched in isotopomers of higher masses (M4, M5, and M6), which is typical for hearts perfused with [U-13C3](lactate + pyruvate) under conditions of substantial pyruvate decarboxylation. Second, for hearts perfused with 1 mM fatty acids (groups II and III), tissue CAC intermediates showed their highest 13C enrichments in M1 isotopomers, resulting from increased [1-13C]fatty acid oxidation. Finally, the MID of malate was similar to that of fumarate (data not shown), reflecting rapid equilibration by fumarase (12).
|
Table 3 reports the MPE values for tissue pyruvate and for the acetyl and OAA moieties of tissue citrate, and Fig. 1 depicts the various flux parameters calculated from these MPE values for all perfusion groups. From the MPE of tissue pyruvate in M3 isotopomers, we conclude that, under all conditions examined, 70% of tissue pyruvate arose from exogenously supplied pyruvate and/or lactate, and the remaining 30% arose from exogenously supplied glucose and/or endogenous glycogen. These data concur with those of Lloyd et al. (23) in emphasizing the importance of exogenous lactate and pyruvate as energy substrates. Tissue pyruvate was not enriched in M1 or M2 isotopomers (data not shown), indicating negligible decarboxylation of malate to pyruvate through the malic enzyme reaction (12, 30, 49, 50). There were significant changes in the MPEs M1 and M2 of ACCit between hearts perfused with 0.4 mM oleate and 1 mM fatty acids, clearly indicating that the sources of acetyl units for citrate formation differed greatly between these perfusion groups.
|
|
From the flux parameters (Fig. 1), we conclude that pyruvate decarboxylation (PDC/CS) represented the predominant source of acetyl-CoA for citrate synthesis under the control condition (Fig. 1A; 0.66 ± 0.06 for PDC/CS vs. 0.22 ± 0.07 for OLE/CS). Raising fatty acid concentrations from 0.4 to 1 mM by the addition of 0.6 mM oleate or 0.6 mM octanoate significantly reduced by >70% the contribution of pyruvate decarboxylation to citrate synthesis, whereas the contribution of exogenous fatty acid -oxidation was increased more than threefold. In the latter condition, octanoate represented by far the major fatty acid source of acetyl-CoA, a finding that concurs with the notion that its -oxidation is not regulated at CPT-I. As evaluated from the flux ratio OS/CS, the percentage of acetyl-CoA that could arise from the -oxidation of exogenous oleate in hearts perfused with oleate + octanoate would be 8 ± 1%. The inhibition of exogenous oleate -oxidation by octanoate concurs with its capacity to increase tissue levels of the CPT-I inhibitor malonyl-CoA (24, 32). The relative contribution of other substrates to acetyl-CoA formation was low and did not differ significantly between perfusion groups (OS/CS = 0.08–0.16).
Figure 1 also reports the rates of pyruvate carboxylation, an anaplerotic reaction, expressed relative to that of the CS reaction (PC/CS; Fig. 1B) or the pyruvate decarboxylation reaction (PC/PDC; Fig. 1C). From the PC/CS flux ratio, we conclude that this reaction generated 10% of OAA for citrate synthesis when the hearts were perfused with 0.4 mM oleate (group I), but its contribution was reduced significantly by 60% in hearts perfused with 1 mM fatty acids consisting of oleate (group II) or oleate + octanoate (group III). The carboxylation of pyruvate represented 20% of its decarboxylation in hearts perfused with 0.4 mM oleate. Pyruvate partitioning between carboxylation and decarboxylation was not affected by the addition of 0.6 mM oleate, but the PC/PDC flux ratio was increased nearly twofold after the addition of 0.6 mM octanoate. There appears to be little, if any, entry of unlabeled carbon through anaplerosis at site(s) other than pyruvate carboxylation, as suggested by the dilution factors calculated [Eq. 10 of Ref. 11: 1.17 ± 0.08 (group I), 0.98 ± 0.05 (group II), and 1.01 ± 0.04 (group III); not significant].
Absolute flux rates for CAC, PDC, and PC were calculated from MO2 values using Eq. 3a (groups I and II) and Eq. 3b (group III). This calculation assumes complete oxidation of the substrates. The CAC flux rate, 3 µmol/min, did not differ significantly between perfusion groups (3.2 ± 0.3, 2.9 ± 0.1, and 3.6 ± 0.2 µmol/min for groups I, II, and III, respectively). The calculated absolute flux rates for fatty acid -oxidation [0.080 ± 0.005, 0.203 ± 0.009, and 0.709 ± 0.033 µmol fatty acid oxidized/min for group I (oleate), group II (oleate), and group III (octanoate), respectively] are within the range of reported values for similarly perfused working rat hearts (24). The calculated absolute flux rates for pyruvate decarboxylation (2.2 ± 0.3, 0.6 ± 0.1, and 0.4 ± 0.1 µmol/min for groups I, II, and III, respectively; P < 0.001, groups II and III vs. group I) and carboxylation (0.31 ± 0.04, 0.12 ± 0.04, and 0.14 ± 0.01 µmol/min for groups I, II, and III, respectively; P < 0.01, groups II and III vs. group I), as well as their reduction by high fatty acid concentrations, were in agreement with the measured rates of lactate + pyruvate uptake. These rates, which were assessed from the concentrations of unlabeled and 13C-labeled lactate and pyruvate in influent and effluent perfusates, and coronary flow rate measurements are shown in Fig. 2. As reported previously (50), perfused rat hearts also constantly released unlabeled lactate, presumably reflecting glycolysis from exogenous glucose or endogenous glycogen, which is depicted in Fig. 2 as lactate efflux rates. Increasing the fatty acid concentration from 0.4 to 1 mM with oleate or octanoate decreased lactate and pyruvate uptake (13C) and efflux (12C) rates, although most of these changes did not reach significance. Although lactate and pyruvate were supplied at the physiological ratio of 5, lactate and pyruvate uptake rates differed by a factor of 2.5, indicating preferential pyruvate uptake, in agreement with the data of Lloyd et al. (23). This was most apparent when the hearts were perfused with 1 mM oleate, where there was little, if any, lactate uptake. The ratio of lactate to pyruvate efflux, an index of the cytosolic redox state, was also greater at 1 mM than at 0.4 mM fatty acids, the greatest increase being observed with 1 mM oleate (5.8 ± 0.6, 10.8 ± 0.8, and 8.5 ± 0.5 for groups I, II, and III, respectively; P < 0.01, groups II and III vs. group I).
|
CAC Intermediate Concentrations, Enzyme Activities, and Citrate Release Rates
Figure 3 depicts the tissue concentrations of citrate and other CAC intermediates as well as the total pool size of CAC intermediates for all perfusion groups. Compared with the control condition (group I), the addition of 0.6 mM oleate (group II) or 0.6 mM octanoate (group III) significantly increased tissue citrate and isocitrate levels (Fig. 3A) as well as the total pool size of CAC intermediates (Fig. 3B). With the addition of 0.6 mM octanoate, not only were the observed changes of greater magnitude, but the tissue levels of -KG, fumarate, and malate were also elevated. Under all conditions examined, the tissue levels of isocitrate represented 1% of citrate, which was lower than that found at equilibrium for the aconitase reaction in vitro (4–6%) (27). The tissue concentrations of succinate, an index of O2 deprivation, were also similar. The activities of enzymes responsible for citrate synthesis and utilization, namely, CS and aconitase, did not differ significantly between the perfusion groups. The activity of CS averaged 2.0 ± 0.1 U/mg protein, whereas that of aconitase was >10-fold lower, averaging 0.15 ± 0.01 U/mg protein. Citrate release rates were similar for all perfusion groups and ranged from 11 to 14 nmol/min (Fig. 4A), which represented 0.3–0.5% of CAC flux rates. These rates represented <17% of those through the pyruvate carboxylation reaction (Fig. 4B), indicating that cardiac citrate efflux was well compensated by anaplerosis. In hearts perfused with 1 mM oleate (group II), however, citrate release rates represented a significantly greater percentage of anaplerosis than in the other conditions (groups I and III).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-oxidation (enhanced >3-fold) and pyruvate decarboxylation (reduced 3-fold) to the formation of acetyl-CoA for citrate synthesis. These effects of fatty acids concur with previous reports (25, 33, 40). The inhibitory effects of fatty acids on pyruvate oxidation were also associated with a lowering of lactate and pyruvate uptake and efflux rates, as reported previously by Chatham et al. (7). The impact of fatty acids on the anaplerotic flux of pyruvate carboxylation has not been examined previously under physiologically relevant conditions. Unexpectedly, our results clearly demonstrated a 2.4-fold lowering of the contribution of pyruvate carboxylation for citrate synthesis when the exogenous fatty acid concentration was increased from 0.4 to 1 mM. This finding may explain why in studies of the pig heart in situ, where plasma fatty acid concentrations were set at >0.6 mM by infusing Intralipid alone or with octanoate, the PC/CS flux was only 4.7 ± 0.3%, and pyruvate addition did not elicit a rise in relative flux through pyruvate carboxylation under normoxia (30) or under ischemia mimicking hibernation (29). Clearly, more work is necessary to assess the significance of anaplerotic intervention with pyruvate at high fatty acid concentrations, as well as to identify the mechanism by which high fatty acid concentrations inhibit flux through pyruvate carboxylation. This will require designing an isotopic protocol to evaluate the relative contributions of pyruvate carboxylase and NADP+-linked malic enzyme, because the two enzymes are differentially regulated by several metabolites (acetyl-CoA, malonyl-CoA, pyruvate, malate, glutamate, fumarate, and succinate) (14, 16, 19, 29, 31, 36, 41, 45, 46). From our literature search, the only plausible candidate to explain our data appears to be acetyl-CoA. Although pyruvate carboxylase activity is stimulated at low concentrations of acetyl-CoA (20–200 µM), its activity is inhibited at >200 µM, which could prevail at high fatty acid concentrations (14). We excluded the inhibition of pyruvate carboxylase by glutamate (19), because the tissue levels of glutamate were similar under all perfusion conditions, averaging 0.85 ± 0.01 mmol/g dry wt (data not shown).
Let us now consider the effects of fatty acid concentrations on the heart's citrate tissue levels and release rates. Despite a twofold decrease in flux through anaplerotic pyruvate carboxylation, citrate tissue levels, as well as total CAC pool size, were increased severalfold at 1 mM fatty acids. The magnitude and type of changes depended on the nature of the fatty acid added: the changes were of greater magnitude, and more CAC intermediates showed a significant increase in their tissue levels with octanoate (group III), consistent with previous reports (10, 30). The greater tissue citrate accumulation in hearts perfused with 1 mM fatty acids, combined with the high concentration ratios of citrate to isocitrate (100-fold) and to -KG (5-fold), suggests a greater imbalance or mismatch between citrate synthesis and utilization rates. This imbalance cannot be explained by changes in the maximal tissue activities of enzymes involved in these reactions, CS and aconitase, or by changes in absolute CAC flux rates, which were similar under all conditions examined. The rise in tissue citrate levels of hearts perfused with 1 mM fatty acid was not associated with a parallel increase in the citrate release rates. These results concur with those obtained in the pig heart in situ (30). Although they appear to argue against our previous data obtained in Langendorff-perfused rat hearts, which showed modulation of citrate release by substrate supply (49), this apparent discrepancy can be explained by major differences in perfusion protocols between these two studies. In our previous investigation, except for one condition, the hearts were perfused in the absence of albumin and LCFA, and citrate release rates were modulated in response to drastic changes in substrate supply, such as the removal of lactate + pyruvate or the decrease of octanoate concentration from 0.2 to 0.02 mM. In contrast, in this study, the hearts were perfused with physiological concentrations of glucose, lactate, pyruvate, and LCFA as our control condition, and fatty acid concentration was raised from 0.4 to 1 mM by the addition of oleate or octanoate. Irrespective of the mode of perfusion (Langendorff or working mode), citrate release rates reached similar maximal values (14–20 nmol/min) when hearts were supplied with a mixture of substrates mimicking the in situ milieu.
We interpret our results on citrate release in the perfused working rat heart, which concur with those in the pig heart in situ, as suggesting that, under physiological conditions, citrate release rates are maximal. The rates could be limited by the activity of the mitochondrial and/or plasma membrane citrate transporter. We cannot distinguish between these possibilities, because measured rates may underestimate mitochondrial citrate efflux because of citrate metabolism in the cytosol by ATP-citrate lyase. In the heart, mitochondrial citrate transport is unidirectional (8, 9). Thus cytosolic citrate has no known metabolic fate other than cleavage or efflux at the level of the plasma membrane. ATP-citrate lyase shows an activity of 0.2 µmol·min–1·g wet wt–1 in heart tissue (1), which is <10-fold greater than that of the mitochondrial citrate transporter. Its participation in the generation of acetyl-CoA for malonyl-CoA synthesis is supported by the results of our recent study (32). Little is known about the regulation of myocardial ATP-citrate lyase. Potentially, its activity could be restricted by the low level of CoA resulting from LCFA activation (9). Hence, proportionally more cytosolic citrate should be cleaved to OAA and acetyl-CoA in hearts perfused with 0.4 mM oleate + 0.6 mM octanoate than with 1 mM oleate. In support of this possibility are the differential changes observed under these two perfusion conditions in tissue: 1) malate levels (i.e., increased with 1 mM oleate/octanoate) and 2) intracellular pH (i.e., decreased with 1 mM oleate), as well as 3) the ratio of lactate to pyruvate released or taken up by the heart, an index of the cytosolic NADH-to-NAD+ ratio (1 mM oleate >1 mM oleate/octanoate). These effects would concur with the potential metabolic benefit of greater cleavage of cytosolic citrate to OAA followed by reduction to malate, namely, the oxidation of cytosolic NADH combined with proton (H+) consumption. An additional benefit would be the recycling of anaplerotic malate via the mitochondrial citrate transporter, which could compensate for decreased flux through pyruvate carboxylation. Additional studies are, however, necessary to substantiate these explanations. The differential effects could be of clinical relevance, because substitution of LCFA with short-chain fatty acids or MCFA in the diet prevents the development of cardiac hypertrophy in spontaneously hypertensive rats (17).
What is the significance of restricted cardiac citrate release? Limited release would appear advantageous under conditions of limited substrate supply, in view of the crucial role of citrate in energy metabolism. Could there be, however, some detrimental consequences linked to limited cardiac citrate release under conditions of substrate abundance? In our previous study, we found that blocking mitochondrial citrate efflux with 1,2,3-benzenetricarboxylic acid resulted in increased LDH release, indicating a potential adverse effect on membrane integrity (49). Could restricted citrate release contribute to tissue citrate accumulation? Interestingly, the difference in tissue citrate levels between hearts perfused for 30 min with 0.4 and 1 mM fatty acids amounts to 1.1 and 2.4 µmol/g dry wt. This is equivalent to accumulation rates of 5 and 15 nmol citrate/min, which is in the range of measured values for citrate release rates. That citrate accumulates, at least in part, in the cytosol and has the expected inhibitory effect on phosphofructokinase is suggested by measured indirect estimates of glycolytic flux, namely, the release rate of unlabeled lactate and tissue MPE M3 pyruvate (1 – MPE M3 pyruvate = glucose contribution to tissue pyruvate). Both values showed a similar trend, supporting a lower glycolytic rate at 1 mM than at 0.4 mM fatty acids, although the difference did not reach significance. Further work should, however, substantiate these findings, with the use of [U-13C6]glucose to provide a direct and more precise quantitation of glycolytic flux. Nevertheless, these considerations raise the following question: Could the enhanced citrate release by diseased hearts (50) be part of an adaptive mechanism to prevent citrate accumulation and, hence, enable the heart to maintain a higher glycolytic rate that is crucial for pump function (51)? This appears to be especially relevant in cardiac patients when LCFA concentrations are chronically elevated (1 mM). Under these conditions, restriction of cytosolic citrate cleavage may result in sustained accumulation of cytosolic citrate and/or additional effects such as a reduction of the redox state (i.e., higher NADH/NAD+) and/or pH. The detrimental consequences of H+ accumulation in association with high LCFA concentrations on the heart's performance, especially after ischemia, include inhibition of contractile proteins and Ca2+ overload (see Ref. 37 for a review). Interestingly, the participation of cytosolic ATP-citrate lyase and malate dehydrogenase in a redox shuttle has been proposed in -cells, where it would be implicated in glucose-induced insulin secretion (15).
With regard to the potential consequences of high fatty acid concentrations on cardiac performance, an interesting peripheral finding of this study was that high fatty acids lowered cardiac output and efficiency with unaltered O2 consumption. This finding concurs with the notion that the heart's contractile performance at a given MO2, referred to as cardiac efficiency, is worse when it is oxidizing more fatty acids in lieu of glucose and lactate (for a review, see Ref. 40). However, in most (6, 18, 21, 26) but not all (4) studies, MO2 was increased at high fatty acids. Thus the effects of fatty acids on cardiac efficiency could be attributed to the theoretical 11% lower ATP yield per O2 consumed during oxidation of fatty acids than during oxidation of carbohydrates (43) or to an uncoupling or ATP-wasting effect of fatty acids (4, 18). Further work appears warranted to explain how high fatty acids can lower cardiac output and efficiency without change in MO2 and also to clarify the potential link between the metabolic and functional effects of fatty acids. A direct effect of fatty acids on the hemodynamics of the isolated, denervated heart perfused ex vivo appears not unlikely given the capacity of fatty acids to modulate sarcolemmal ATP-sensitive K+ channels (22) or to impair endothelium-dependent dilation via nitric oxide (43). In vivo, the situation differs, because any direct actions of fatty acids on the heart would be counterbalanced by the response of the sympathetic and parasympathetic systems and may explain the increase in MO2 by fatty acids (2, 18, 21, 44).
In conclusion, this study, conducted in the working rat heart perfused with a substrate mixture mimicking the in situ milieu, expands on our knowledge of the functional and metabolic effects of high fatty acid concentrations in the heart. At the functional level, increasing the fatty acid concentration from 0.4 to 1 mM resulted in a lowering of cardiac output and efficiency with unaltered O2 consumption. At the metabolic level, beyond the well-documented effects of high fatty acid levels on pyruvate oxidation and -oxidation, our results demonstrate a 2.4-fold lowering of the contribution of anaplerotic pyruvate carboxylation to citrate synthesis. Despite the dual inhibitory effect of high fatty acids on pyruvate decarboxylation and carboxylation, tissue citrate levels were nearly two-fold higher, but citrate release rates remained unchanged, probably limited by the activity of its mitochondrial or plasma membrane transporter. The latter results emphasize a differential modulation by fatty acids of anaplerotic pyruvate carboxylation and citrate release in the heart. Additional work is, however, needed to identify the mechanism by which fatty acids inhibit anaplerotic pyruvate carboxylation and to dissect out the various factors that confound the interpretation of measured cardiac citrate release rates. Of specific interest in this regard is the activity of cytosolic ATP-citrate lyase, for which our results suggest a differential modulation by LCFA vs. MCFA and a potential impact on the cytosolic redox state and pH. The proposed lines of investigation appear relevant to the evaluation of the clinical role of anaplerotic interventions with pyruvate or of an MCFA diet in cardiac patients.
|
ACKNOWLEDGMENTS |
|---|
Part of this work was presented at the XVII World Heart Congress of the International Society for Heart Research held in Winnipeg and at its satellite meeting held in Banff in July 2001.
GRANTS
This study was supported by the Canadian Institutes of Health Research Grants 9575 and 10920 (to C. Des Rosiers) and a studentship (to G. Vincent).
|
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.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-oxidation of fatty acids on the kinetics of mitochondrial citrate transporter. Can J Biochem 54: 171–177, 1976.[Web of Science][Medline]
-cells: possible implication of a pyruvate/citrate shuttle in insulin secretion. Diabetes 49: 718–726, 2000.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  Y.-G. Niu and R. D. Evans Myocardial metabolism of triacylglycerol-rich lipoproteins in type 2 diabetes J. Physiol., July 1, 2009; 587(13): 3301 - 3315. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Gelinas, F. Labarthe, B. Bouchard, J. Mc Duff, G. Charron, M. E. Young, and C. Des Rosiers Alterations in carbohydrate metabolism and its regulation in PPAR{alpha} null mouse hearts Am J Physiol Heart Circ Physiol, April 1, 2008; 294(4): H1571 - H1580. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. Durgan, M. W. S. Moore, N. P. Ha, O. Egbejimi, A. Fields, U. Mbawuike, A. Egbejimi, C. A. Shaw, M. S. Bray, V. Nannegari, et al. Circadian rhythms in myocardial metabolism and contractile function: influence of workload and oleate Am J Physiol Heart Circ Physiol, October 1, 2007; 293(4): H2385 - H2393. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. An and B. Rodrigues Role of changes in cardiac metabolism in development of diabetic cardiomyopathy Am J Physiol Heart Circ Physiol, October 1, 2006; 291(4): H1489 - H1506. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. C. Stanley, F. A. Recchia, and G. D. Lopaschuk Myocardial Substrate Metabolism in the Normal and Failing Heart Physiol Rev, July 1, 2005; 85(3): 1093 - 1129. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Labarthe, M. Khairallah, B. Bouchard, W. C. Stanley, and C. Des Rosiers Fatty acid oxidation and its impact on response of spontaneously hypertensive rat hearts to an adrenergic stress: benefits of a medium-chain fatty acid Am J Physiol Heart Circ Physiol, March 1, 2005; 288(3): H1425 - H1436. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Sharma, I. C. Okere, D. Z. Brunengraber, T. A. McElfresh, K. L. King, J. P. Sterk, H. Huang, M. P. Chandler, and W. C. Stanley Regulation of pyruvate dehydrogenase activity and citric acid cycle intermediates during high cardiac power generation J. Physiol., January 15, 2005; 562(2): 593 - 603. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Khairallah, F. Labarthe, B. Bouchard, G. Danialou, B. J. Petrof, and C. Des Rosiers Profiling substrate fluxes in the isolated working mouse heart using 13C-labeled substrates: focusing on the origin and fate of pyruvate and citrate carbons Am J Physiol Heart Circ Physiol, April 1, 2004; 286(4): H1461 - H1470. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
