Impaired Myocardial Fatty Acid Oxidation and Reduced Protein Expression of Retinoid X Receptor-α in Pacing-Induced Heart Failure
Background— The nuclear receptors peroxisome proliferator-activated receptor-α (PPARα) and retinoid X receptor α (RXRα) stimulate the expression of key enzymes of free fatty acid (FFA) oxidation. We tested the hypothesis that the altered metabolic phenotype of the failing heart involves changes in the protein expression of PPARα and RXRα.
Methods and Results— Cardiac substrate uptake and oxidation were measured in 8 conscious, chronically instrumented dogs with decompensated pacing-induced heart failure and in 8 normal dogs by infusing 3 isotopically labeled substrates: 3H-oleate, 14C-glucose, and 13C-lactate. Although myocardial O2 consumption was not different between the 2 groups, the rate of oxidation of FFA was lower (2.8±0.6 versus 4.7±0.3 μmol · min−1 · 100g−1) and of glucose was higher (4.6±1.0 versus 1.8±0.5 μmol · min−1 · 100g−1) in failing compared with normal hearts (P<0.05). The rates of lactate uptake and lactate output were not significantly different between the 2 groups. In left ventricular tissue from failing hearts, the activity of 2 key enzymes of FFA oxidation was significantly reduced: carnitine palmitoyl transferase-I (0.54±0.04 versus 0.66±0.04 μmol · min−1 · g−1) and medium chain acyl-coenzyme A dehydrogenase (MCAD; 1.8±0.1 versus 2.9±0.3 μmol · min−1 · g−1). Consistently, the protein expression of MCAD and of RXRα were significantly reduced by 38% in failing hearts, but the expression of PPARα was not different. Moreover, there were significant correlations between the expression of RXRα and the expression and activity of MCAD.
Conclusions— Our results provide the first evidence for a link between the reduced expression of RXRα and the switch in metabolic phenotype in severe heart failure.
Received March 20, 2002; revision received May 7, 2002; accepted May 7, 2002.
The healthy, aerobically perfused heart uses free fatty acids (FFA) as the primary source of energy, although cardiac substrate metabolism can be profoundly altered in some pathological conditions.1 It has been suggested since the late 1950s that during severe heart failure, the myocardium shifts to preferential utilization of carbohydrates,2,3⇓ and we confirmed these findings in canine pacing-induced heart failure.4 This topic remains controversial, because other studies indicate that the failing heart preferentially uses FFA,5–7⇓⇓ and recent data suggest that pharmacological inhibition of myocardial fatty acid oxidation is a potential therapy for heart failure.8 Moreover, the molecular mechanisms responsible for alterations in the metabolic phenotype of the failing heart are only understood in part. Seminal studies on myocardial tissue from animal models and humans demonstrate that the failing heart reverts to a fetal metabolic phenotype, with reduced expression of key enzymes of FFA oxidation.9,10⇓ The mechanisms involved in this enzyme downregulation are not yet known, but new hints can be derived from the growing number of investigations on the role of peroxisome proliferator-activated receptor-α (PPARα) and retinoid X receptors (RXR) in the control of cellular intermediate metabolism in various tissues, including myocardium. PPARα is a member of the superfamily of nuclear receptors that promotes the expression of key enzymes of the fatty acid oxidative pathway.11,12⇓ Activated PPARα heterodimerize in the nucleus with RXR,13 and then the dimer binds to specific DNA sequences and activates the expression of enzymes of lipid metabolism.11,14,15⇓⇓
In the light of these findings, we formulated the hypothesis that the altered metabolic phenotype of the failing heart involves changes in myocardial protein expression of PPARα and/or of RXRα, one of the best described retinoid receptors. Therefore, in the present study, we first quantified changes in cardiac substrate uptake and oxidation by infusing 3 isotopic tracers (3H-oleate, 14C-glucose and 13C-lactate) in conscious dogs with normal and failing hearts. This is the most accurate method of measurement available and was adopted to avoid the inaccuracies of previous results based on indirect indices of substrate use.2,4–7⇓⇓⇓⇓ At the end of the experiments in vivo, myocardial biopsies were freeze-clamped to determine PPARα and RXRα protein expression and the activity of key enzymes of the FFA oxidative pathway, specifically carnitine palmitoyl transferase-I (CPT-I), acetyl-coenzyme A (CoA) carboxylase (ACC) and malonyl-CoA decarboxylase (MCD), which regulate FFA transfer into mitochondria, and the activity and expression of medium-chain acyl-CoA dehydrogenase (MCAD), one of the mitochondrial enzymes of β-oxidation.
A total of 17 male mongrel dogs (aged 12 to 18 months; weight, 23 to 27 kg; Friedensburg, Pa) were sedated with acepromazine maleate (1 mg/kg IM), anesthetized with sodium pentobarbital (25 mg/kg IV), ventilated with room air, and instrumented as previously described.4,16,17⇓⇓ After 7 to 10 days of recovery from surgery, dogs were trained to lie quietly on the laboratory table. The protocol was approved by the Institutional Animal Care and Use Committee of the New York Medical College and conformed to the guiding principles for the care and use of laboratory animals published by the National Institutes of Health.
Hemodynamics Recordings and Calculated Parameters
Left ventricular and aortic pressure, blood flow in the left circumflex coronary artery, and left ventricular diameter were measured and acquired, and dP/dtmax and percent shortening of the left ventricular diameter were calculated as previously described.4,16,17⇓⇓
Total and Labeled Metabolite Measurements
Oxygen content and total cardiac substrate concentrations were measured in arterial and coronary sinus blood samples. Three isotopically labeled substrates were infused into the dogs: [9,10-3H]-oleate, [U-14C]-glucose, and L-[1-13C]-lactate.17 The concentration of oxygen, total and labeled substrate, and catabolites in arterial and coronary blood samples and mean coronary blood flow were used to calculate the rates of FFA, lactate, and glucose uptake, as well as FFA and glucose oxidation.17 Tracer-measured lactate output was calculated as the difference between tracer-measured lactate uptake, and net lactate uptake was calculated as arterial minus coronary sinus difference of total lactate times mean coronary blood flow.17 Lactate output quantifies the rate of nonoxidative glycolysis of endogenous and exogenous glucose. Myocardial oxygen consumption (MV̇O2) and rates of substrate consumption were normalized by cardiac weight and expressed as μmol · min−1 · 100 g of tissue−1.
Enzyme Activities and Metabolic Products in Cardiac Tissue
The activities of MCAD, CPT-I, citrate synthase, ACC, and MCD were measured in powdered left ventricular tissue as previously described.17–19⇓⇓ Free CoA, short-chain, and long-chain CoA ester concentration was also determined.17 Left ventricular tissue triglyceride18 and glycogen20 contents were measured using enzymatic spectrophotometric assays.
Western Immunoblot Analysis
Protein was extracted from frozen tissue as previously described.21 Fifty micrograms of total protein were separated by electrophoresis and transferred onto a polyvinylidene fluoride membrane. Membranes were incubated with a specific antibody to PPARα, RXRα (1:500 dilution; both from Santa Cruz Biotechnology, Inc), and MCAD (1:2000 dilution; Cayman Chemical). After conjugation with the secondary antibody, the membranes were developed in a chemiluminescence substrate solution (Pierce Supersignal Chemiluminescents Substrate). Successively, they were reprobed for β-actin (1:2000, Chemicon) to verify the uniformity of protein loading. Bands were visualized by autoradiography and quantified using commercially available software. Results are expressed as percentage of the density of a standard sample loaded on all membranes in triplicate.
Heart failure was induced in 9 dogs by pacing the left ventricle at 210 bpm for 3 weeks; then the pacing rate was increased to 240 bpm. The experiments were performed when left ventricular end-diastolic pressure reached ≈25 mm Hg and clinical signs of severe decompensation were observed.4 Because it was necessary to harvest large cardiac biopsies at the end of each experiment, a separate group of 8 similarly instrumented, healthy dogs was used as a control. Experiments were conducted in conscious dogs placed on the laboratory table after overnight fasting. Hemodynamics were recorded, and the isotopic tracers [9,10-3H]-oleate, [U-14C]-glucose, and L-[1-13C]-lactate were continuously infused for the duration of the experiment through a peripheral vein.17
In dogs with heart failure, the experiments were performed at spontaneous heart rates with the pacemaker turned off. In control dogs, the heart was paced throughout the experiment to match the spontaneous rate of the dogs with heart failure. After 40 minutes of tracer infusion, paired blood samples were withdrawn from the aorta and coronary sinus. In one dog with heart failure, isotopes were not infused due to the occlusion of the coronary sinus catheter and the consequent inability to withdraw paired blood samples. At the end of this procedure, the dogs were anesthetized with 30 mg/kg sodium pentobarbital IV and then intubated and ventilated. The fifth intercostal space was rapidly opened to harvest a large transmural biopsy (≈10 g) from the left ventricular anterior free wall while the heart was still beating. The harvested tissue was immediately freeze-clamped with tongs precooled in liquid nitrogen. This approach was previously used by us and others.17,22,23⇓⇓ The heart was then removed and weighed.
Data are presented as mean±SEM. Statistical analysis was performed by employing commercially available software (Sigma Stat 2.01). Differences between dogs with heart failure and control dogs were tested using the t test. Correlations between groups of values were evaluated calculating the best fit, based on least-squares regression analysis. The regression lines were then represented and the coefficient of correlation (R) was indicated. For all the statistical analyses, significance was accepted at P<0.05.
The experiments in dogs with heart failure were performed after 29.0±1.6 days of cardiac pacing, the time when left ventricular end-diastolic pressure reached a value of 25.8±1.9 mm Hg. At that stage of failure, the dogs presented with severe clinical signs of decompensation such as dyspnea, ascites, pale mucosae, and lethargy. Table 1 shows the hemodynamic changes that occurred in dogs with heart failure compared with control dogs. Left ventricular and aortic pressure, dP/dtmax, and percent shortening of the left ventricular diameter decreased significantly, whereas left ventricular end-diastolic diameter increased by ≈17% and mean flow in the left circumflex coronary artery did not change significantly.
In dogs with heart failure, the arterial concentration of FFA was significantly lower and the concentration of glucose significantly higher when compared with normal dogs (Table 2).
As shown in Figure 1, FFA oxidation was ≈40% lower in failing compared with normal hearts, despite the absence of a significant difference in MV̇O2. Conversely, glucose oxidation was ≈150% higher in failing hearts. Total lactate uptake did not change significantly. Changes in substrate uptake reflected those relative to substrate oxidation (Figure 1), although the increase in glucose uptake was not significantly different due to a higher degree of variability that is often associated with this method of measurement in the glucose arterial/coronary sinus difference. Lactate output did not change significantly, indicating no difference in the rate of nonoxidative glycolysis.
Enzyme Activities and Metabolic Products in Cardiac Tissue
The activities of CPT-I and MCAD, shown in Figure 2, were, respectively, 19% and 36% lower in failing compared with normal hearts (Table 3; P<0.05). The activities of citrate synthase, a marker of mitochondrial metabolic activity, and of ACC and MCD, the 2 enzymes that control malonyl-CoA cytosolic concentration, were not different between groups. Consistently, the tissue content of malonyl-CoA, the only known endogenous inhibitor of CPT-I, was not different in failing compared with normal hearts. Also, the tissue content of free CoA, long chain-CoA esters, and glycogen were not different between groups, but triglyceride concentration, an index of FFA storage, was significantly elevated in failing hearts (Table 3).
MCAD, PPARα, and RXRα Protein Expression
Results from Western blot analysis are shown in Figure 3. Both MCAD and RXRα protein expressions were ≈38% lower in failing hearts (P<0.05), but PPARα protein level was not different between groups. A significant correlation was found between the protein expression of RXRα and MCAD (Figure 4). A second significant correlation was found between the protein expression and enzyme activity of MCAD; in addition, RXRα protein expression consistently correlated significantly with MCAD enzyme activity (Figure 4). Points were distributed without solution of continuity, and their positions well reflected the differences between the 2 groups.
This study provides the first evidence for a link between reduced protein expression of RXRα and altered myocardial metabolic phenotype in severe heart failure. Conscious dogs in end-stage heart failure presented a dramatic alteration in substrate metabolism, with a 40% reduction in FFA oxidation and a 150% increase in glucose oxidation but no changes in nonoxidative glycolysis, as measured for the first time with multiple isotopic tracers. The impaired FFA oxidation was associated with concomitant decreases in the activity of CPT-I and in the activity and protein expression of MCAD. Myocardial levels of PPARα were not significantly different from control; however, the protein expression of RXRα, an obligate cofactor of PPARα, fell by ≈40% and was significantly correlated with MCAD expression and activity.
Sack et al9 formulated the stimulating hypothesis that the metabolic alterations occurring during heart failure represent a recapitulation of fetal metabolism, as characterized by a downregulation of FFA oxidation enzymes and an increased oxidation of glucose as the primary myocardial substrate. Razeghi et al10 widely explored this topic and found that the gene expression of key enzymes responsible for FFA oxidation and for pyruvate dehydrogenase inhibition was similarly lower in failing and fetal hearts compared with nonfailing, adult hearts. In the present study, we measured the activity and protein expression of MCAD, a marker of mitochondrial capacity to oxidize fat. These values fell during heart failure in the absence of a significant change in the activity of the mitochondrial marker enzyme citrate synthase. Therefore, on the basis of prior studies, we expected a decreased protein expression of PPARα, the nuclear receptor that controls the transcription of CPT-I and some enzymes of β-oxidation, including MCAD.12,24⇓
We first focused on this receptor, which has captured the interest of numerous investigators in the field of cardiovascular pathophysiology over the past 10 years.12,15,21,24–27⇓⇓⇓⇓⇓⇓ Surprisingly, we detected a slight, although not significant, increase in PPARα protein concentration in failing hearts. In the light of our findings, a reasonable explanation for the impaired FFA oxidative capacity is that the PPARα control on enzymes expression was in part deactivated by the decreased concentration of circulating FFA, important PPARα activators, and/or by the reduced availability of its obligate cofactor RXRα.14 This interpretation is supported by one previous study in vitro showing that, even in the presence of normal levels of PPARα, some key enzymes of lipid metabolism are downregulated when the levels of RXRα are low.15 Further support comes from the significant correlation between protein expression of RXRα and the expression and activity of MCAD that we found in cardiac biopsies. Because tissue samples were freeze-clamped at the end of the experiments in vivo, our data are the first to indicate that RXRα plays a key role in the control of cardiac FFA oxidation in the beating heart.
Little information is presently available about the intracellular signaling regulating retinoid receptors in myocytes, and we can only formulate hypotheses about possible mechanisms for the reduced expression of RXRα in decompensated hearts. Hypoxia, for instance, causes RXRα downregulation in cultured myocytes15 and, although we did not find significant differences in mean coronary blood flow and in the rate of nonoxidative glycolysis between normal and failing hearts, we cannot exclude the possibility that repetitive episodes of regional myocardial perfusion abnormalities, similar to those observed with positron emission tomography in idiopathic dilated cardiomyopathy,28 could have caused oxygen supply/demand mismatch and led progressively to alterations in metabolic phenotype. However, it is very intriguing that a single gene knockout of the RXRα gene in mice results in decreased cardiac muscle mass during organ development, with consequent embryonic heart failure and lethality at ≈2 weeks of embryonic life.29 Hearts from RXRα−/− embryos display, among numerous alterations, reduced gene expression of MCAD and long-chain CoA dehydrogenase and energy deprivation.30 These findings in knockouts suggest the possibility that the reduced expression of RXRα and consequent metabolic alterations could play an important role in the structural and functional progressive deterioration of cardiac muscle during chronic heart failure. This hypothesis might be tested by blocking RXRα in vivo but, unfortunately, specific pharmacological inhibitors are not available.
CPT-I activity determines the rate of FFA transport into mitochondria. Cytosolic malonyl-CoA, whose concentration is regulated by the opposing activities of ACC and MCD, inhibits CPT-I activity and, to date, this is the only known step of the FFA oxidative pathway regulated by a negative feedback control.31,32⇓ Prior studies have not explored alterations of this rate-limiting step of FFA oxidation in failing hearts. We did not find significant differences in the activities of ACC and MCD or in malonyl-CoA concentrations between the 2 groups.
The concept that the failing myocardium switches toward greater glucose oxidation and less fatty acid oxidation is still controversial and based on a limited number of animal studies2–4,6⇓⇓⇓ whose results are, in part, inconsistent with the few published data obtained from clinical studies.5–7⇓⇓ Discordances between clinical and experimental studies, including ours, could be in part due to indirect measurements of substrate metabolism in humans, as well as to the severity of heart failure and to the absence of any pharmacological treatment in animals, whereas patients enrolled in some clinical investigations were quite compensated and regularly treated. We previously found that changes in the rate of FFA and glucose uptake and an increase in cardiac respiratory quotient occurred only toward the end of the fourth week of pacing, when hemodynamic alterations were associated with reduced PO2 and clinical signs of overt heart failure.4 The study by Sack et al9 supports the hypothesis that a frank alteration in cardiac metabolism occurs only during decompensation, because they showed in rats that MCAD expression and activity were significantly decreased only in decompensated failure. Their companion findings in humans are consistent with those by Razeghi et al,10 and in fact both groups studied myocardial tissue obtained from cardiac transplant or ventricular assist device recipients in end-stage heart failure.
Several limitations should be pointed out. First, the increased glucose oxidation in failing hearts was likely consequent to the depressed use of the competitive substrate FFA; however, it is possible that glucose transport and the glycolytic pathway were potentiated by additional and independent mechanisms. Given the difficulty to explore, in a single study, all of the enzymes of different metabolic pathways, we chose to start with selected, key enzymes of FFA metabolism. Future studies will assess the enzymes of glucose transport and oxidation. For the same reason, we focused only on 2 nuclear receptors, PPARα and RXRα, whose interaction in the gene regulatory process is well documented in vitro, but we cannot exclude a role for other PPARs and RXRs. Finally, it should be considered that, although PPARα expression was not altered in our model of dilated cardiomyopathy, the downregulation of this receptor likely constitutes a critical pathophysiological component of other models, such as pressure overload–induced cardiac hypertrophy.27,33⇓
In conclusion, the present study provides the first evidence for a link between reduced RXRα and altered myocardial metabolic phenotype in severe heart failure. The 40% decrease in RXRα protein expression and the significant correlation with MCAD protein expression and activity indicate that RXRα downregulation could be responsible for the impairment of the FFA oxidative pathway in the failing heart.
This study was supported by the National Heart, Lung and Blood Institute grant R01-HL62573 (to F.A.R.) and in part by: R01-58653 and 64848 (to W.C.S.), P01 HL-43023 (to T.H.H.), and by a grant from the Canadian Institutes of Health (to G.D.L.). Dr Linke received a postdoctoral fellowship (Li 946/1-1) from the German Research Foundation. Dr Diep received a postdoctoral fellowship from the Canadian Institutes of Health Research.
- ↵Recchia FA, McConnell PI, Bernstein RD, et al. Reduced nitric oxide production and altered myocardial metabolism during the decompensation of pacing-induced heart failure in the conscious dog. Circ Res. 1998; 83: 969–979.
- ↵Taylor M, Wallhaus TR, Degrado TR, et al. An evaluation of myocardial fatty acid and glucose uptake using PET with [F-18]fluoro-6-thia-heptadecanoic acid and [F-18]FDG in patients with congestive heart failure. J Nucl Med. 2001; 42: 55–62.
- ↵Sack MN, Rader TA, Park S, et al. Fatty acid oxidation enzyme gene expression is downregulated in the failing heart. Circulation. 1996; 94: 2837–2842.
- ↵Razeghi P, Young ME, Alcorn JL, et al. Metabolic gene expression in fetal and failing human heart. Circulation. 2001; 104: 2923–2931.
- ↵Watanabe K, Fujii H, Takahashi T, et al. Constitutive regulation of cardiac fatty acid metabolism through peroxisome proliferator-activated receptor alpha associated with age-dependent cardiac toxicity. J Biol Chem. 2000; 275: 22293–22299.
- ↵Keller H, Dreyer C, Medin J, et al. Fatty acids and retinoids control lipid metabolism through activation of peroxisome proliferator-activated receptor-retinoid X receptor heterodimers. Proc Natl Acad Sci U S A. 1993; 90: 2160–2164.
- ↵Huss JM, Levy FH, Kelly DP. Hypoxia inhibits the peroxisome proliferator-activated receptor alpha/retinoid X receptor gene regulatory pathway in cardiac myocytes: a mechanism for O2-dependent modulation of mitochondrial fatty acid oxidation. J Biol Chem. 2001; 276: 27605–27612.
- ↵Recchia FA, McConnell PI, Loke KE, et al. Nitric oxide controls cardiac substrate utilization in the conscious dog. Cardiovas Res. 1999; 44: 325–332.
- ↵Recchia FA, Osorio JC, Chandler MP, et al. Reduced synthesis of NO causes marked alterations in myocardial substrate metabolism in conscious dogs. Am J Physiol. 2002; 282: E197–E206.
- ↵Diep QN, Schiffrin EL. Increased expression of peroxisome proliferator-activated receptor-alpha and -gamma in blood vessels of spontaneously hypertensive rats. Hypertension. 2001; 38: 249–254.
- ↵Knight RJ, Kofoed KF, Schelbert HR, et al. Inhibition of glyceraldehyde-3-phosphate dehydrogenase in post-ischemic myocardium. Cardiovasc Res. 1996; 32: 1016–1023.
- ↵Brandt JM, Djouadi F, Kelly DP. Fatty acids activate transcription of the muscle carnitine palmitoyltransferase I gene in cardiac myocytes via the peroxisome proliferator-activated receptor alpha. J Biol Chem. 1998; 273: 23786–23792.
- ↵Takano H, Nagai T, Asakawa M, et al. Peroxisome proliferator-activated receptor activators inhibit lipopolysaccharide-induced tumor necrosis factor-alpha expression in neonatal rat cardiac myocytes. Circ Res. 2000; 87: 596–602.
- ↵Young ME, Laws FA, Goodwin GW, et al. Reactivation of peroxisome proliferator-activated receptor alpha is associated with contractile dysfunction in hypertrophied rat heart. J Biol Chem. 2001; 276: 44390–44395.
- ↵Neglia D, Parodi O, Gallopin M, et al. Myocardial blood flow response to pacing tachycardia and to dipyridamole infusion in patients with dilated cardiomyopathy without overt heart failure. A quantitative assessment by positron emission tomography. Circulation. 1995; 92: 796–804.
- ↵Dyson E, Sucov HM, Kubalak SW, et al. Atrial-like phenotype is associated with embryonic ventricular failure in retinoid X receptor alpha −/− mice. Proc Natl Acad Sci U S A. 1995; 92: 7386–7390.
- ↵Ruiz-Lozano P, Smith SM, Perkins G, et al. Energy deprivation and a deficiency in downstream metabolic target genes during the onset of embryonic heart failure in RXRalpha−/− embryos. Development. 1998; 125: 533–544.
- ↵Awan AA, Saggerson ED. Malonyl-CoA metabolism in cardiac myocytes and its relevance to the control of fatty acid oxidation. Biochem J. 1993; 295: 61–66.
- ↵Saddik M, Gamble J, Witters LA, et al. Acetyl CoA carboxylase regulation of fatty acid oxidation in the heart. J Biol Chem. 1993; 286: 25836–25845.
- ↵Sack MN, Disch DL, Rockman HA, et al. A role for Sp and nuclear receptor transcription factors in a cardiac hypertrophic growth program. Proc Natl Acad Sci U S A. 1997; 94: 6438–43.