vol. ,268. no. 34. issue december 5, pp. … of acetyl-coa and citrate. however, using affinity...

THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1993 by The American Society for Biochemistry and Molecular Biology, Inc

VOl. ,268 . No. 34. Issue of December 5, pp. 2583625845.1993 Printed in U. S. A.

Acetyl-coA Carboxylase Regulation of Fatty Acid Oxidation in the Heart*

(Received for publication, April 8, 1993, and in revised form, August 2, 1993)

Maruf SaddikSj, James GambleSll, Lee A. Wittersll, and Gary D. LopaschukS** From the $Cardiovascular Disease Research Group and Lipid and Lipoprotein Research Group, Departments of Pediatrics and Pharmacology, University of Alberta, Edmonton T6G 2S2, Canada and the IIEndocrine-Metabolism Division, Departments of Medicine and Biochemistry, Dartmouth Medical School, Hanover, New Hampshire, 03755-3833

The role of acetyl-coenzyme A carboxylase (ACC) in regulating fatty acid oxidation was investigated in isolated fatty acid perfused working rat hearts. Over- all fatty acid oxidation rates were determined by addition of 1.2 mM [‘Hlpalmitate to the perfusate of hearts in which the endogenous triglyceride pool was prelabeled with [14C]palmitate. Rates of both exogenous and endogenous fatty acid oxidation were measured by simultaneous measurement of ‘HzO and 14COz production, respectively. A second series of hearts were perfused under similar conditions except that [U-”C]glucose was present in the perfusate for measurement of glucose oxidation rates. Addition of dichloroacetate (DCA, 1 mM) to the perfusate resulted in a dramatic stimulation of glucose oxidation (a 41 1% increase), with a parallel decrease in fatty acid oxidation (from 305 f 51 to 206 f 40 nmol/g dry weight *min-unit work). DCA treatment increased the contribution of glucose oxidation to ATP production from 7.1 to 30.6%, while decreasing the contribution of overall fatty acid oxidation from 92.9 to 69.4%. Tissue levels of malonyl-CoA in hearts treated with DCA were higher compared to controls (14.0 f 0.6 and 10.0 f 0.7 nmol/g dry weight, respectively) and were negatively correlated ( r = -0.85) with overall fatty acid oxidation rates. Acetyl-coA levels were also significantly higher in DCA-treated hearts, and a positive correlation ( r = 0.88) was seen between myocardial acetyl-coA and malonyl-CoA levels. This suggests that DCA treatment increased the supply of acetyl-coA for ACC. Western blots revealed the presence of both the 280-kDa (ACC-280) and the 265-kDa (ACC-265) isoforms of ACC in cardiac tissue, with a predominance of ACC-280. The activity of ACC extracted from hearts was similar in both groups when assayed under optimal conditions of acetyl-coA and citrate. However, using affinity purified ACC, it was demonstrated that heart ACC (predominantly ACC-280) had a higher K, for

*This work was funded in part by a grant from the Medical Research Council of Canada (to G. D. L.) and by a grant from the Muscular Dystrophy Association and National Institutes of Health Grant DK 35712 to (L. A. W.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisemnt” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

3 Graduate student trainee of the Alberta Heritage Foundation for Medical Research and a post-doctoral fellow of the Medical Research Council of Canada.

1 Graduate student trainee of the Muttart Diabetes Research and Training Center and Heart and Stroke Foundation of Canada.

** Scholar of the Alberta Heritage Foundation for Medical Re- search. To whom reprint requests should be addressed 423 Heritage Medical Research Bldg., The University of Alberta, Edmonton, Al- berta, T6G 2S2 Canada. Tel.: 403-492-2170; Fax: 403-492-3383.

acetyl-coA than ACC isolated from white adipose tissue (predominantly ACC-265). We conclude that ACC is an important regulator of fatty acid oxidation in the heart and that acetyl-coA supply is a key determinant of heart ACC-280 activity. As acetyl-coA levels increase, ACC-280 is activated resulting in an increase in malonyl-CoA inhibition of fatty acid oxidation.

Acetyl-coenzyme A carboxylase (ACC)’ is a key regulatory enzyme in the fatty acid synthetic pathway. It is a biotin- containing enzyme which catalyzes the carboxylation of acetyl-coA to form malonyl-CoA, a substrate for fatty acid synthase (1, 2). In addition, malonyl-CoA is also a potent inhibitor of carnitine palmitoyltransferase 1 (CPT 1) (3-6), which catalyzes the conversion of acyl-CoA to acylcarnitine. Since CPT 1 is a key regulatory enzyme in the mitochondrial uptake of fatty acids, ACC has also been suggested to play a pivotal role in regulating fatty acid oxidation (7). This is supported by studies which show that ACC is widely distrib- uted in mammalian tissues, including those where fatty acid synthesis is not prominent, e.g. heart and skeletal muscle (7, 8-10). Two isoforms of this enzyme have been identified to date, a 265- and a 280-kDa isoform (7, 8), although more isoforms are thought to exist (11,12). The majority of studies have characterized the role of the 265-kDa isoform (ACC- 265), which predominates in highly lipogenic tissues such as liver and white adipose tissue. While less is known about the 280-kDa isoform ,of ACC (ACC-280), the presence of this isoform in tissues with a high fatty acid oxidative capacity suggests a role of this enzyme in regulating fatty acid oxidation (7, 8).

Fatty acid oxidation provides 60-90% of the ATP production by the heart, depending to a large extent on circulating fatty acid concentrations. As with liver, a key regulatory enzyme involved in fatty acid oxidation is CPT 1, which is highly sensitive to malonyl-CoA inhibition (5). Cook et al. (13) have recently suggested that, unlike the liver, it is probably the malonyl-CoA supply, as opposed to the sensitivity of CPT 1 to malonyl-CoA inhibition, that is the key factor regulating myocardial fatty acid oxidation. However, the relationship between malonyl-CoA levels and fatty acid oxidation rates in the heart has not been determined.

Stimulation of the pyruvate dehydrogenase complex (PDC) will increase glucose oxidation rates in the isolated fatty acid perfused rat heart, resulting in a parallel decrease in fatty

The abbreviations used are: ACC, acetyl-coA carboxylase; CPT I, carnitine palmitoyltransferase I; DCA, dichloroacetate, PDC, pyruvate dehydrogenase complex; HPLC, high performance liquid chromatography.

25836

Acetyl-coA Carbox

acid oxidation rates (14). We hypothesize that acetyl-coA derived from PDC can be transported from the mitochondria to the cytosol where it can then serve as a substrate for ACC, increasing malonyl-CoA production and decreasing fatty acid oxidation. This transport could occur via a carnitine acetyltransferase and carnitine acetyltranslocase pathway which is active in the heart and transports acetyl groups from intramitochondrial acetyl-coA to cytosolic CoA (15-17). Recently, we demonstrated that stimulating this pathway with carnitine increases glucose oxidation rates in the heart (probably oc- curring secondary to a decrease in the intramitochondrial acetyl-CoA/CoA ratio), with a parallel decrease in fatty acid oxidation (18). Further support to this hypothesis comes from observations by Lysiak et al. (16) that most of acetyl-coA generated from pyruvate by PDC is readily accessible to carnitine acetyltransferase while that generated from 8-oxidation is more available to the tricarboxylic acid cycle.

In the intact heart, fatty acids destined for /3-oxidation can be derived from exogenous sources, or from endogenous myocardial triglyceride stores (19). Using a “pulse-chase’’ technique in the isolated perfused working heart, we can measure the contribution of both exogenous and endogenous fatty acids to @oxidation (19). This technique involves prelabeling the triglyceride pool with [“C]palmitate and chasing hearts with [3H]palmitate. This study uses this procedure to determine if ACC is involved in the short term regulation of fatty acid oxidation in the heart. To modify fatty acid oxidation rates, hearts were perfused with dichloroacetate (DCA), a stimulator of PDC. This results in a marked stimulation of glucose oxidation with a parallel decrease in the contribution of fatty acid oxidation to overall ATP production (14). Using this approach we demonstrate a strong correlation between increased acetyl-coA production from PDC and an increase in myocardial malonyl-CoA production. Increases in malonyl- CoA production also correlated with a reduction in overall myocardial fatty acid oxidation rates, suggesting that ACC is an important regulator of fatty acid oxidation in the heart.

EXPERIMENTAL PROCEDURES

Materials ~-[“C(U)]glucose, [S,lO-’H]palmitic acid, and [1-“C]palmitic acid

were obtained from New England Nuclear. Bovine serum albumin (BSA fraction V) was obtained from Boehringer Mannheim. Hyamine hydroxide (1 M in methanol solution) was obtained from NEN Re- search Products (Boston, MA). ACS Aqueous Counting Scintillant was obtained from Amersham Canada Ltd (Oakville, Ontario). Tri- glyceride assay kits were obtained from Wako Pure Chemical Indus- tries LM (Osaka, Japan). For HPLC analysis of malonyl-CoA, a precolumn cartridge C18, size 3 cm, 7 pM was purchased from Pierce Chemical Co., and a Microsorb short-one column type C18, particle size 3 pm, size 4.6 X 100 mm was purchased from Rainin Instruments Company (Emeryville, CA). ECL Western blotting detection reagents were purchased from Amersham International (Amersham, United Kingdom). IgGl monoclonal antibodies (7AD3) to ACC-280 were chosen from a panel of monoclonal antibodies raised against avidin- Sepharose-purified fasted/refed rat liver enzyme (see Ref. 7, for more details). A polyclonal antibody to ACC-265 was raised in rabbits against fasted/refed rat liver ACC where the 265 kDa band was eluted from an SDS gel for use as an immunogen as described previously (7). Secondary antibodies (peroxidase-conjugated goat anti-mouse I& (H+L) and peroxidase conjugated goat anti-rabbit IgG (H+L) were purchased from Jackson Immunoresearch Laboratories Inc./ Bio/Can Scientific (Mississauga, Ontario). Peroxidase-labeled streptavidin was purchased through Mandel Scientific from Kirkegaard & Perry Laboratories Inc. (Gaithersburg, MD). For streptavidin and immunoblots, Trans-Blot Transfer Medium (pore nitrocellulose membrane 0.45 pm) was obtained from Bio-Rad. X-ray films (X- OMAT AR Film) were purchased from Kodak. All other chemicals were obtained from Sigma.

ylase in the Heart 25837

Heart Perfusions

Adult male Sprague-Dawley rats (250-300 g) were anesthetized with sodium pentobarbital (60 mg/kg) intraperitoneally. Hearts were quickly excised and placed in ice-cold Krebs-Henseleit buffer. The aorta was rapidly cannulated, and a retrograde perfusion using Krebs- Henseleit buffer was initiated as described previously (19). During this initial perfusion, each heart was trimmed of excess tissue, the pulmonary artery was cut, and the opening to the left atrium was cannulated. Following a 10-min equilibration period, hearts were switched to the working heart mode, and perfused at an 11.5-mm Hg left atrial preload and an 80-mm Hg aortic afterload. Spontaneously beating hearts were used throughout the studies, with heart rate and peak systolic pressure being measured by a Gould P21 pressure transducer in the aortic outflow line. Cardiac output, aortic flow, and coronary flow were measured using Transonic in-line ultrasonic flow probes connected to a T 101 ultrasonic blood flow meter.

In all hearts, mechanical function was monitored throughout the entire perfusion. Heart work was expressed as the product of peak systolic pressure X cardiac output.

Perfusion Protocols

All hearts were perfused with Krebs-Henseleit buffer containing

Ca2+. 1.2 mM palmitate, 3% albumin, 11 mM glucose, and 1.25 mM free

Overall Fatty Acid Oxidation Measurements-Table I shows the protocol used to measure both exogenous and endogenous fatty acid oxidation in the heart. In this series of perfusions, hearts were initially perfused for a 60-min period with recirculated Krebs-Henseleit buffer containing 1.2 mM [1-“Clpalmitate to label the endogenous lipid pools (pulse). During this labeling period, exogenous steady state fatty acid oxidation was measured by quantitative collection of myocardial “CO, production as described previously (19). At the end of the “pulse” period, hearts were switched to a retrograde Langendorff drip-out perfusion with Krebs-Henseleit buffer containing 11 mM glucose. A group of hearts were frozen at the end of this washout perfusion (with Wollenberger tongs cooled to the temperature of liquid Nz). In the remainder of hearts, the buffer containing [“C] palmitate was removed from the system during this 10-min period and replaced with buffer containing 11 mM glucose and 1.2 mM [9,10- 3H]palmitate 1 mM DCA. Hearts were then switched back to the working mode, and perfused for a subsequent 60-min period with the new buffer described above.

During the pulse, steady state exogenous palmitate oxidation rates were determined by quantitatively measuring “Con production by the hearts, as described in detail previously (19). Hearts were perfused in a closed system that allowed collection of both perfusate and gaseous “COZ. During the “chase,” “CO, production was used as a measure of endogenous fatty acid oxidation, while ’H20 production from [’HI palmitate was used as a measure of exogenous fatty acid oxidation rates (19). Perfusate and gaseous samples were collected at 10,20,40, and 60 min during the chase. ‘H20 was separated from [’Hlpalmitate as described in detail previously (19). Steady state palmitate oxidation rates during the chase were expressed as nanomoles of labeled palmitate oxidized/gram dry weight. minute. unit work.

Glucose Oxidation Measurements-In this series of heart perfusions, glucose oxidation rates were measured during the chase perfusion. The same perfusion protocol and perfusion substrates described above were used, except that the palmitate was not labeled during either the pulse or the chase period. Instead, during the chase period perfusate contained 11 mM [U-“C]glucose and 1.2 mM palmitate in the presence or absence of 1 mM DCA. Glucose oxidation was determined by quantitative measurement of 14C02 production (“CO, is liberated at the level of PDC and in the tricarboxylic acid cycle). “CO, production was determined using the same methods described above for palmitate oxidation. Glucose oxidation rates were expressed as nanomoles of glucose oxidized/gram dry weight. minute. unit work.

An additional series of hearts was also perfused as described above except that glucose was omitted from the perfusate during the chase perfusion. In these hearts only exogenous and endogenous fatty acid oxidation were measured during the chase perfusion.

Tissue Workup

At the end of the perfusions, heart ventricles were quickly frozen with Wollenberger clamps cooled to the temperature of liquid Ne. The frozen ventricular tissue was then weighed and powdered in a mortar and pestle cooled to the temperature of liquid N2. A portion

25838 Acetyl-coA Carboxylase in the Heart

TABLE I Perfusion protocol for measuring overall fatty acid oxidation (exogenous and endogenous) and glucose oxidation in isolated working hearts

perfused in the absence or presence of dichloroacetate Pulse period + Washout period + Chase period

Perfusion conditions Fatty acid oxidation” 60-min prelabeling with 11 mM glucose 10-min aerobic retrograde 60-min perfusion with 11 mM glucose,

and 1.2 mM [“C]palmitate perfusion 1.2 mM [3H]palmitate with No addition 1 mM dichloroacetate

or

Glucose oxidationb 60-min preperfusion with 11 mM glucose 10-min aerobic retrograde 60-min perfusion with 11 mM [“CI- and 1.2 mM palmitate perfusion glucose, 1.2-mM palmitate with

No addition 1 mM dichloroacetate

Procedure performed Measure 14COz production in [“Clpalmi- Change Perfusate Measure 3Hz0 and/or “Con produc- tate hearts tion

3Hz0 and WO, production during the chase was a measure of exogenous and endogenous fatty acid oxidation, respectively. WO, production during the chase was a measure of glucose oxidation.

of the powdered tissue was used to determine the dry-to-wet ratio of the ventricles. The atrial tissue remaining on the cannula was removed, dried in an oven for 12 h at 100 “C, and weighed. The dried atrial weight, frozen ventricular weight, and ventricular dry-to-wet ratio were then used to determine total dry weight of the heart.

Measurement of Lipid Metabolic Intermediates-Tissue lipids from frozen ventricular tissue were extracted as described previously (19). Neutral lipids were separated from phospholipids using the method described by Bowyer and King (20). [l‘C]palmitate and [3H]palmitate incorporation into neutral lipids was measured using a double radioisotope counting technique described previously (19). Label content of neutral lipids was expressed as micromoles of palmitate incorporated into this pool/gram dry weight. Absolute myocardial triglyceride content (micromoles of fatty acid equivalents/gram dry weight) was determined using Wako enzymatic colorimetric assay kits.

Acetyl-coA Carboxyluse Assay-Approximately 200 mg of frozen tissue was homogenized with a buffer containing Tris-HC1 (50 mM, pH 7.5 at 4 “C), 100 mM NaF, 2 mM EDTA, 0.25 M sucrose, 70 pl of ~-mercaptoethanol/lOO ml, and a mixture of seven protease inhibitors (21). Samples were then ultracentrifuged at 180,000 X g for 60 min. The supernatant was then dialyzed overnight a t 4 “C with a buffer containing 50 mM Tris-HC1 (pH 7.5 at 4 “C), 100 mM NaF, 2 mM EDTA, 10 mM 8-mercaptoethanol, and 10% (v/v) glycerol. Dialysate protein content was measured using Bradford’s method (22). To measure ACC activity, 25 pl of the dialysate was added to a reaction mixture (final volume 190 pl) containing Tris acetate (11.5 mM, pH 7.5), BSA (2.9 p ~ ) , 8-mercaptoethanol (1.5 pM), ATP (0.41 mM), acetyl-coA (0.21 mM), magnesium acetate (0.97 mM), NaHC03 (3.5 mM), and 10 mM magnesium citrate. Samples were incubated at 37 “c for either 0, 1,2,3, or 4 min, and the reaction stopped by adding 10% perchloric acid. Samples were then spun for 10 min and the malonyl- CoA concentration in the supernatant measured using an HPLC procedure described below. In many studies ACC activity was (7, 8), and is still, determined using [“Clbicarbonate fixation to acid-soluble products. We employed an HPLC method to measure the product of the ACC assay, malonyl-CoA, rather than the [“Clbicarbonate fixation method. This assay provides a direct estimate of the enzyme activity through measurement of product levels (malonyl-CoA) and has been shown to be accurate (23). The presence, at high concentrations, of other carboxylases (e.g. pyruvate carboxylase and propionyl- CoA carboxylase) in the heart could potentially make the [“C] bicarbonate fixation method less specific, since the incorporation of “COZ into non-malonyl-CoA products could occur. Comparison studies of both assays (not shown) did not, however, suggest this to be the case, under the assay conditions used in this study.

In order to determine whether cardiac ACC is citrate dependent, another set of reactions was conducted at citrate concentrations ranging from 0-10 mM for 4 min at 37 “c. In all these assay experi- ments, ACC activity was expressed as the amount of malonyl-CoA produced/gram dry weight. minute.

To determine the acetyl-coA kinetics of both ACC-280 and ACC- 265, ACC was purified from rat heart and skeletal muscle (predominantly ACC-280) and white adipose tissue (predominantly ACC-265)

by monomeric avidin-Sepharose chromatography as described previously (24). ACC activity was then measured in the presence of varying concentrations of acetyl-coA. Lineweaver-Burke plots were then generated, and K, values of acetyl-coA for purified preparations of ACC were determined.

Determination of CoA Esters-CoA esters were extracted as described previously (25). The 6% perchloric acid extract was maintained at a pH of 2-3. The CoA esters were measured using a modified HPLC procedure described by King et al. (23). Separation was performed on a Beckman System Gold with a UV detector 167. Each sample (100 p1 each) was run through a precolumn cartridge (C18, size 3 cm, 7 pm) and a Microsorb short-one column (type C18, particle size 3 pm, size 4.6 X 100 mm). Absorbance was set at 254 nm and flow rate a t 1 ml/min. A gradient was initiated using two buffers: buffer A consisted of 0.2 M NaHzP04 (pH 5.0) and buffer B was a mixture of 0.25 M NaH2P04 and acetonitrile (pH 5.0) in a ratio of 8020 (v/v). Buffers were filtered using filter pure, Nylon-66 filter membrane (Pierce Chemical Co.). Initial conditions (97% A, 3% B) were maintained for 2.5 min and were changed thereafter to 18% B over 5 min using Beckman’s curve 3. At 15 min the gradient was changed linearly to 37% B over 3 min and subsequently to 90% B over 17 min. At 42 min the composition was returned linearly back to 3% B over 0.5 min, and at 50 min column equilibration was complete. Peaks were integrated by Beckman System Gold software package. Chromatographic separation of standard CoA esters and CoA esters obtained from cardiac tissue extract are shown in Fig. 1.

Western Blot Analysis of ACC-Dialysate samples were subjected to SDS-polyacrylamide gel electrophoresis using the method of Laem- mli (26). Following the gel electrophoresis, the protein bands were transferred to nitrocellulose. Membranes were then probed with either streptavidin, a monoclonal antibody (7AD3) specific to ACC- 280, or a polyclonal antibody against the NHn-terminal region of ACC-265. Secondary peroxidase-conjugated goat anti-mouse IgG (against the monoclonal antibody) and goat anti-rabbit IgG (against the polyclonal antibody) were used to visualize ACC. Chemilumines- cent detection was performed on the membranes using an ECL Western blotting detection kit.

Measurement of Mitochondrial CPT IActivity-Mitochondria from a fresh adult rat heart were isolated as described previously (27) using ice-cold MSE buffer containing 225 mM mannitol, 75 mM sucrose, and 1 mM EGTA (pH 7.5). The mitochondrial preparation was used to determine CPT I sensitivity to inhibition by malonyl-CoA by measuring CPT I activity in the presence of varying concentrations of malonyl-CoA. CPT I activity was determined as described previously by Bremer (28). Briefly, an incubation medium containing 75 mM KCl, 50 mM mannitol, 25 mM HEPES (pH 7.3), 2 mM NaCN, 0.2 mM EGTA, 1 mM dithiothreitol, and 1% fat free albumin was used. The mitochondrial preparation (35-40 pg) was preincubated with this buffer a t 30 “C in the presence of 75 pM palmitoyl-CoA (dissolved in 25 mM KH2P04, pH 5.3) and varying concentrations of malonyl-CoA ranging from 0-2.5 p ~ . 1 pCi of ~-[methyl-~H]carnitine was then added to a final L-carnitine concentration of 200 pM, and the incubation was continued for a further 6 min. Reactions were

25839

A) Standards

O.OT!

u

s" 3 9 0.010

0.001

O.Oo0

Acetyl-coA Carboxylase in the Heart

0 ) Hcnrl tissue

2

0.030 -

0.021-

I 1 ,,.. ......

1 1 0 2 0 1 0 4 0

Timc (Inin)

FIG. 1. High performance liquid chromatography profiles of short chain coenzyme A esters from known Standards (A) and from a perchloric acid extract obtained from an isolated perfused heart ( B ) . CoA esters were extracted from heart tissue, separated, and assayed as described under "Experimental Procedures." 1, malonyl-CoA; 2, glutathione-CoA; 3, CoASH; 4, methylmalonyl-CoA; 5, succinyl-CoA; 6, acetyl-coA; 7 propionyl-CoA; 8, isobutyryl-CoA; 9, valeryl-CoA.

TABLE I1 Mechanical function of isolated working hearts perfused in the presence or absence of the pyruvate dehydrogenase complex stimulator,

dichloroacetate Data represent the mean f S.E. of at least six hearts in each group. Hearts were perfused as described under "Experimental Procedures."

HR, heh-rate; PSP, peak systolic pressure. Heart work was determined as the product of PSP X cardiac output. -

Condition Heart Peak systolic HR x PSP x rate pressure 10-8 Heart work

beatlmin mm Hg beat. mmHg/min mmHg.ml/min X IO-' During pulse perfusion 247 f 10 90.1 f 1.0 22.28 f 1.06 2.3 f 0.4 During chase perfusion

Control 217 f 15 97.9 f 4.2 21.4 * 1.5 Dichloroacetate (1 mM)

1.5 f 0.5 256 f 12 96.6 f 2.8 24.7 f 1.3 2.4 f 0.7

stopped by adding 100 pl of concentrated HCI. The [SH]palmitoylcarnitine formed was measured using n-butanol extraction as described previously (28) and then counted using standard radioisotope counting techniques.

Statistical Analysis

Data are presented as the mean f S.E. of the mean. The unpaired Students t test was used to determine statistical significance in groups containing two sample populations. A value of p < 0.05 was regarded as significant.

RESULTS

Effect of Dichloroacetate on Heart Function-Because the work performed by the heart is a key determinant of carbon substrate oxidation rates, mechanical function was continu- ously monitored throughout the perfusion period. Table I1 shows heart rate, peak systolic pressure, and heart work in control and DCA-treated hearts during both the initial pulse period, and during the chase period. In control hearts, no deterioration of mechanical function was seen throughout the perfusion period. Addition of DCA following the pulse period did not have major effects on heart function. A small increase in both heart rate and heart work was seen in the DCA-

treated hearts, although this increase was statistically not significant. However, to rule out any potential effects of this small increase in function on oxidative metabolism, all subsequent oxidative rate measurements were corrected for difference in heart work.

Effects of Dichloroacetate on Palmitate and Glucose Oxida- tion Rates-Previous studies from our laboratory have shown that endogenous myocardial triglycerides are an important source of fatty acids for mitochondrial oxidation (19). There- fore, to determine the role of ACC in regulating fatty acid oxidation, both exogenous and endogenous fatty acid oxidation rates were measured. To achieve this, we prelabeled the myocardial triglyceride pool with ['4C]palmitate (see Table I). This resulted in 23.4 k 2.6 pmol/g dry weight of ['4C]palmitate being incorporated into myocardial neutral lipids. During the 60-min chase perfusion, [14C]palmitate was not present in the perfusate, and hearts were perfused in the presence of 1.2 mM [3H]palmitate. As a result, oxidation of ["Clpalmitate during the chase originated from endogenous triglycerides (19). Dur- ing the 60-min chase period, rates of both endogenous and exogenous fatty acid oxidation were linear (data not shown, see Ref. 19).

25840 Acetyl-CoA Curb&ase in the Heurt

Addition of DCA to the perfusate had a dramatic effect on myocardial glucose oxidation. An absolute increase in glucose oxidation from 110 f 15 to 722 f 69 nmol/g dry weight .min occurred. Absolu~ rates of palmitate oxidation oxidation also increased in the DCA-treated hearts (exogenous palmitate oxidation rates were 401 f 56 nmol/g dry weight .min in control hearts uersus 502 f 64 in the DCA-treated hearts, while endogenous palmitate oxidation rates were 24.6 f 6.9 and 27 f 8.8 nmol/g dry weight .min, respectively). This resulted in an overall increase in acetyl-coA production in the DCA-treated hearts. However, DCA treatment also resulted in an improvement in heart function (Table 11). To account for this, oxidative rates were normalized for differences in heart function (Table 111). Even when corrected for functional differences, DCA treatment resulted in a marked stimulation of glucose oxidation (Table IIIA) (secondary to a stimulation of PDC activity). This was accompanied by a parallel decrease in both exogenous and endogenous fatty acid oxidation rates. Table IIIB, shows the percent contribution of these pathways to overall myocardial ATP production. In control hearts perfused with 1.2 mM palmitate, glucose oxidation provided 7.1% of the overall ATP production. DCA treatment resulted in an increase in the contribution of glucose oxidation to 30.1%. The 23% increase in the contribution of glucose oxidation to ATP production was accompanied by a parallel 24% decrease in the contribution of overall fatty acid oxidation to ATP production. This demonstrates that DCA treatment resulted in a significant shift away from fatty acid oxidation toward glucose oxidation as a source of ATP production.

TABLE I11 Oxidation rates of glucose and palmitate (A) and contribution to ATP production (B) in isolated working rat hearts perfused in the presence

or absence of d ~ ~ ~ e t a t e Data are the mean f S.E. of at least six hearts in each group.

Hearts were perfused as described in Table I, with oxidation rates measured during the chase perfusion. Cont~bution to ATP production was calculated from the absolute oxidative rates, using a value of 129 mol of ATP produced/mol of palmitate oxidized, and 38 mol of ATP produced/mol of glucose oxidized.

Condition Glucose oxidation

Exogenous Endogenous palmitate palmitate oxidation oxidation

A) Steady state rates nmol,'g dry tot, min. unit work

Control 72.9 4 10.0 304.9 4 51.0 20.1 4 5.0 Dichloroacetate 300.0 4 28.6" 205.5 f 40.1" 14.4 f 3.9 (1 mM)

B) Contribution to ATP production % contribution

Control 7.1 87.5 5.4 Dichloroacetate 30.6 65.5 3.9 (1 mM)

Significantly different from control hearts.

M y o c a ~ ~ l - ~ b e k d Neutral Lipid Content and Triacylgyc- erol Content-Table IV shows the label content of ['*C]palmitate and I3HIpalmitate in hearts frozen at the end of the 60-min chase period. In both control and DCA-treated hearts, the amount of ["C]palmitate remaining in the heart was similar. In contrast, a s i ~ l c a n t l y greater amount of ['HI palmitate was incorporated into neutral lipids (almost 7 pmol/ g dry weight more) during the chase period in DCA-treated hearts compared with control. This combined with the lower fatty acid oxidation rates (almost 6 pmol/g dry weight .60 min.unit function less), suggests that fatty acids were shunted away from fatty acid oxidation and toward triacylglycerol synthesis.

Myocardial Leuels of CoA Esters-Table V shows levels of malonyl-CoA and acetyl-coA measured in hearts frozen at the end of each perfusion. ~alonyl-CoA levels were significantly higher in DCA-treated hearts compared with control hearts. Acetyl-coA levels were also significantly greater in the DCA-treated hearts. This indicates that both substrate (acetyl-coA) and product (malonyl-CoA) levels increased in DCA-treated hearts.

A significant correlation was seen between myocardial acetyl-coA levels and malonyl-CoA levels (Fig. 2 4 ) . This suggests that acetyl CoA supply to ACC may be an important regulator of malonyl CoA production. Fig. 2B shows the correlation between myocardial levels of malonyl-CoA and overall fatty acid oxidation rates in the heart. A significant negative correlation was also seen between malonyl-CoA levels and palmitate oxidation rates, suggesting that malonyl- CoA levels are an important determinant of myocardial fatty acid oxidation rates. To further support this suggestion, another group of hearts were perfused using the same perfusion protocol as shown in Table I except that glucose was absent from the perfusate during the chase perfusion. As shown in Table VI, absolute rates of exogenous ['HHfpalmitate oxidation significantly increased in hearts perfused in the absence of glucose. This was accompanied by a dramatic decrease in malonyl-CoA levels. Overall, these data suggest that an increase in myocardial acetyl-coA levels from DCA stimulation of PDC increases malonyl-CoA p ~ u c t i o n , resulting in an inhibition of fatty acid oxidation,

Cardiac CPT I Sensitivity to Malonyl-CoA-The inhibition of fatty acid oxidation by malonyl-CoA would occur at the level of CPT I. Fig. 3 depicts the exquisite sensitivity of CPT I in cardiac m i t ~ h o n ~ i a preparation to malonyl-CoA. A 50% inhibition of CPT 1 activity was seen in the presence of as little as 50 nM malonyl-CoA/mg mitochondrial protein. In- creasing the concentration of malonyl-CoA to 0.5 pM/mg mitochondrial protein almost completely abolished the enzyme activity.

Characterization of Acetyl-coA Carboxylase in Perfused Hearts-In order to explore the potential explanations for the DCA-induced alterations in cardiac malonyl-CoA levels, the activity, content, and isozyme distribution of ACC were de-

TABLE IV ["CIPalmitate and PHIpalmitate content in neutral lipids and triocylglycerol content in isolated working rat hearts perfused in the presence

or absence of dichloroacetate Data represent the mean f S.E. of at least six hearts in each group. Hearts were perfused as described under "Experimental Procedures."

Condition as neutral lipids ["CIPalmitate

as neutral lipids [SH]Palmitate Triacylglycerol content

amollg dry wt pmoi fatty acidlg dry wt Control 14.49 f 2.33 15.80 f 1.65 22.54 f 1.26 Dichloroacetate (1 mM) 14.33 t 1.87 22.43 f 1.25" 27.29 f 1.87

a Significantly different from control.

Acetyl-CoA Carboxylase in the Heart 25841

TABLE V Levels of malonyl-CoA and acetyl-coA in isolated working rat har t s

perfused in the presence or absence of 1 mM dichloroacetate Data represent the mean f S.E. of a t least seven hearts in each

group. Hearts were perfused as described under "Experimental Pro- cedures."

Condition Malonyl-CoA Acetyl-coA nmol/g dry wt

Control 10.0 f 0.7 32.8 f 2.9 Dichloroacetate (1 mM) 14.0 f 0.6" 90.8 f 6.0"

Significantly different from control hearts.

20 - r = 0.88

16 -

12 -

8 -

4 1 O ! I I

0 40 80 120

Acetyl CoA (nmoVp dry wt)

"1 r = - 0.85

100 6 8 10 12 14 16 18

Malonyl-CoA ( n m d g dry wt) FIG. 2. Correlation between myocardial total acetyl-coA

and malonyl-CoA levels ( A ) and between malonyl-CoA levels and total palmitate oxidation rates ( B ) in isolated hearts perfused in the presence or absence of dichloroacetate. Hearts were perfused as described in Table I. Palmitate oxidation rates shown are the sum of exogenous and endogenous palmitate oxidation. CoA esters were extracted from hearts using 6% perchloric acid and separated by HPLC as as described under "Experimental Proce- dures." Triangles, DCA-treated hearts. Squares, control hearts.

termined in extracts of the perfused hearts. ACC activity, measured in extracts of control and DCA-perfused hearts under V,, conditions with respect to acetyl-coA and in the presence of 10 mM citrate, showed no difference between

TABLE VI Effect of remving glucose from the perfusate on malonyl-CoA levels and overall fatty acid oxidation rates in isolated working rat hearts Data are the mean f S.E. of six hearts perfused with 11 mM

glucose, 1.2 mM palmitate, and four hearts perfused with 1.2 mM palmitate alone. Hearts were perfused as described in Table I, with overall fatty acid oxidation rates being measured during the chase perfusion.

Perfusate condition palmitate Exogenous Endogenous

oxidation palmitate oxidation

n m l / g dry wt nmollg dry wt . min

11 mM glucose 10.0 f 0.7 401.3 f 56.1 24.6 2 6.9 1.2 mM palmitate

1.2 mM ualmitate 1.1 f 0.5" 749.5 f 179" 69.3 f 14.6" a Significantly different from hearts perfused with 11 mM glucose

and 1.2 mM palmitate.

these two groups (89.6 f 15 nmol of malonyl-CoA produced/ g dry weight/min uerus 84.4 2 14 in control and DCA-treated hearts, respectively (means f S.E. of the mean of n = 4 determinations in each group). The content of the ACC isozymes in these extracts was also invariant between control and DCA-treated groups (Fig, 4). As shown in this protein plot analysis of ACC extracted from three control and three DCA-treated hearts, both ACC-280 and ACC-265 are present in these extracts, as determined with isoform-specific antibodies (Fig. 4, A and B ) , although the former is the predominant ACC isozyme, as determined by protein blotting with streptavidin-peroxidase, which recognizes the biotin moiety of ACC and the other biotin-containing cardiac enzymes (Fig. 4C).

In liver and adipose tissue, ACC activity (predominantly due to ACC-265) can be markedly stimulated by the allosteric activator, citrate. Furthermore, this isozyme is also highly regulated by variable enzyme phosphorylation (29-31). Changes in citrate reactivity can predict changes in ACC phosphorylation; for example, highly phosphorylated ACC shows little activity in the absence of citrate and an increased citrate K, (29, 30). However, as shown in Fig. 5, cardiac ACC shows little citrate dependence. This lack of citrate dependence persisted if ACC was prepared from fresh hearts or if hearts were either slowly frozen or frozen and rethawed (which increases ACC phosphorylation post-mortem (31) (data not shown)). Furthermore, perfusion of hearts with glucagon, which, in liver, leads to ACC phosphorylation and inactivation (2,32,33), did not lead to alterations in measured enzyme velocity or observable citrate dependence (data not shown). Taken together, these data suggest that, unlike liver and adipose tissue ACC (ACC-265 predominant), heart ACC (ACC-280 predominant) is not significantly regulated either by the allosteric activator, citrate, or by variable enzyme phosphorylation.

Another major distinguishing feature of these two ACC isozymes is their affinity for the substrate, acetyl-coA. ACC was purified from white adipose tissue (mainly ACC-265) and from both heart and skeletal muscle (mainly ACC-280) by monomeric avidin-Sepharose chromatography and acetyl- CoA kinetics determined (Fig. 6). This Lineweaver-Burke transformation plot illustrates the marked difference in acetyl-coA K,,, for these isolated isozymes; in this representative parallel isolation of ACC from each tissue, ACC-265 (adipose tissue) displays a K , of 67 pM, while ACC-280 (heart and skeletal muscle) has a K , of 117 and 109 p ~ , respectively. Such lower affinity of cardiac ACC for acetyl-coA suggests

25842 Acetyl-CoA Carboxylase in the Heart

FIG. 3. Sensitivity of cardiac mitochondrial carnitine palmitoyltransferase I to inhibition by malonyl-CoA. Mitochondria were isolated from a rat heart as described under “Ex- perimental Procedures.” CPT I activity in the mitochondrial preparation was determined by measuring the formation of [~H]palmitoylcarnitine from [3H]carni- tme, as described under “Experimental Procedures.”

that substrate regulation of ACC activity could play an important role in this organ.

DISCUSSION

The role of ACC-265 as the rate-limiting enzyme of fatty acid biosynthesis in liver and adipose tissue has been well documented. A 280-kDa isoform of ACC has also recently been identified and is immunologically distinct from the 265- kDa isoform. The observation that ACC-280 is present in the heart (8, Fig. 4), and that malony-CoA is a potent inhibitor of mitochondrial CPT 1 (3-5) has led to the suggestion that ACC-280 may have an important role in regulating fatty acid oxidation in the heart. However, direct evidence for this role of ACC-280 has not previously been provided. Our data not only demonstrates that ACC is important in regulating fatty acid oxidation in the heart, but also that the acute regulation of ACC-280 activity differs from ACC-265, i.e. ACC-280 is regulated primarily by acetyl-coA supply to the enzyme as opposed to regulation by phosphorylation or by the allosteric activator, citrate. To address the role of ACC in the heart, we used isolated working rat hearts in which overall fatty acid oxidation was directly measured. If hearts are perfused with 1.2 mM palmitate, 93% of ATP production originates from fatty acid oxidation (exogenous and endogenous). Stimulation of glucose oxidation with DCA decreased the contribution of palmitate oxidation to 65% of myocardial ATP production. This decrease in fatty acid oxidation was closely correlated with an increase in cardiac malonyl-CoA levels. The increase in malonyl-CoA levels in the heart was also closely correlated

0.0 OS 1 .o 1.5 2.0 2.5

Malonyl CoA (pM)

with a DCA-induced increase in cardiac acetyl-coA levels, suggesting that an increase in acetyl-coA supply from PDC stimulates ACC activity. In contrast, if PDC activity is decreased by removing glucose from the perfusate a marked decrease in malonyl-CoA levels occur (Table VI). This is accompanied by an increase in fatty acid oxidation rates in the heart.

Despite the increase in malonyl-CoA levels in DCA-treated hearts, ACC activity measured in extracts obtained from frozen cardiac tissue was similar in both the DCA-treated and control hearts. Previous methods for measuring ACC have used COz fixation as a measure of ACC activity. This involves using [14C]bicarbonate and measuring its incorporation into malonyl-CoA. However, the strepavidin-probed Western blot shown in Fig. 4C demonstrates the high concentration of other carboxylases in the heart relative to ACC. We therefore developed an ACC assay based on measuring the end product of ACC, malonyl-CoA (see “Experimental Procedures”). This assay provides an efficient and accurate measure of ACC activity. It should be pointed out, however, that under the assay conditions used, no difference between the HPLC assay and the measurement of [“Clacid-soluble products derived from H14C0, were observed. This suggests that some of the controversy surrounding the H”CO3 assay method is unwar- ranted.

This lack of difference in ACC activity is unlikely to be due to nonspecific phosphorylation during isolation, since isolation conditions were chosen which are known to preserve the phosphorylation state of the enzyme (34). Furthermore, car-

Acetyl-coA Carboxylase in the Heart 25843

ACC 280 Ab

ACC 265 Ab

STREPTAWDIN

DCA CONTROL TREATED

205,

116.

, 205-

116-

,ACC 280 LACC 265

” 205 -

116- am“ .L- “ - - -. - FIG. 4. Immunoblot and streptavidin-peroxidase analysis of

acetyl-coA carboxylase in tissue extracts from isolated perfused hearts. Immunoblots were performed using a monoclonal antibody (Ab) to the 280-kDa isoform of acetyl-coA carboxylase (ACC-280) ( A ) and a polyclonal antibody to the 265-kDa isoform of acetyl-coA carboxylase (ACC-265) ( B ) . Streptavidin-peroxidase (C) was used to determine the relative content of ACC-280 and ACC-265 (streptavidin recognizes the biotin containinggroups of carboxylases). Following perfusion, control and DCA-treated hearts were quickly frozen and ACC was isolated as described under “Experimental Procedures.” SDS-polyacrylamide gel electrophoresis was performed followed by transfer to nitrocellulose membrane. Immunoblot was then performed as described under “Experimental Procedures.” PC, pyruvate carboxylase.

diac ACC activity did not show any significant citrate dependence, which occurs when the 265-kDa isoform of ACC is inhibited by phosphorylation (34). The lack of effect of DCA on ACC activity in uitro also speaks against a covalent mod- ification of ACC as the mechanism behind the increase in malonyl-CoA levels in the heart.

ACC-265 has been shown to be regulated over two different time frames. A rapid regulation (minutes) involves changes in covalent phosphorylation, allosteric regulation, and polym- erization (7,35). A long term (hours-days) regulation involves changes in enzyme mass (36) together with changes in enzyme activity caused by the aforementioned rapid regulation (30, 34). While liver ACC was reported to be dependent on citrate for activation (29, 37), Thampy et al. (34) found that it is the phosphorylated form of the enzyme that is citrate dependent; quickly frozen livers yielded a more active, citrate-independ-

loo 1

1 0 : I I I I I 1 0 2 4 6 8 10 12

Citrate (mM)

FIG. 5. Citrate dependence of acetyl-coA carboxylase activity isolated from control and dichloroacetate-treated hearts. Data represent the mean f S.E. of at least six hearts. ACC was extracted from control and DCA-treated hearts frozen at the end of the chase perfusion. ACC activity was measured by HPLC analysis of the amount of malonyl-CoA produced/minute/g dry weight of the tissue, as described under “Experimental Procedures.”

.10 -10 0 10 20 30 40 SO 60

l/[Acetyl-CoA] (mM) FIG. 6. Acetyl-coA kinetics of ACC enzyme isolates. ACC

was isolated and purified in parallel from rat white adipose tissue ( WAT, open circles), heart (closed circles), and skeletal muscle (SM, quadriceps and soleus, squares) as described under “Experimental Procedures.” The graph depicts the Lineweaver-Burke transformation of activity measured at variable [acetyl-CoA] (10-500 p ~ ) . The inset displays the immunoblot of the enzyme isolates, as probed simultaneously with anti-ACC-265 and anti-ACC-280 antibodies. These data obtained from a single set of parallel isolates are representative of three such simultaneous isolations.

ent ACC with lower phosphate content. Davies et al. (38) recently reported that the higher phosphorylated state of ACC was seen when freeze-clamping was not used and that this higher phosphorylation correlated with a large increase in AMP and decrease in ATP (presumably caused by hypoxia during removal of the liver) and with increased activity of AMP-activated protein kinase. In this study, where hearts were quickly frozen to the temperature of liquid N2 during the isolated heart perfusion, no citrate dependence was seen in either control or DCA-treated hearts (Fig. 5). If freezing of heart tissue was delayed, we did not see any increase in citrate dependence of the enzyme (data not shown). Similarly, if hearts are perfused with glucagon (which can increase ACC phosphorylation in liver) ACC activity, myocardial malonyl-

25844 Acetyl-coA Carboxylase in the Heart

CoA levels and rates of fatty acid oxidation are not altered (data not shown). Thus, the ACC-280 which predominates in the heart may not be regulated by phosphorylation to the same degree as ACC-265. Rather, our data suggest that the increase in malonyl-CoA was caused by the increase in the amount of acetyl-coA available to the enzyme, i.e. cardiac ACC-280 is a substrate-driven enzyme. As depicted in Fig. 2A, a significant correlation was seen between acetyl-coA and malonyl-CoA levels in both DCA-treated and untreated hearts frozen at the end of perfusion. Furthermore, ACC purified from rat hearts (in which ACC-280 predominates) exhibits lower affinity for acetyl CoA (K. = 117 pM) compared to (KO = 67 p ~ ) in white adipose tissue (in which ACC-265 predominates). Cytosolic levels of total CoA in the heart are in the range of 15-50 p~ (25). Although cytosolic content of acetyl- CoA has not been accurately determined, it is obviously lower than overall CoA levels. As a result, cytosolic acetyl-coA availability may be an important factor modulating cardiac ACC activity. In contrast, overall cytosolic CoA levels in the liver (39) are much greater than in the heart, and the affinity of ACC-265 for acetyl-coA is higher (7) suggesting that acetyl- CoA supply is probably less important as a regulator of ACC in this organ.

Although most of the acetyl-coA in the heart is intramitochondrial, a mechanism exists in the heart for the transport of acetyl groups out of the mitochondria via a carnitine acetyltransferase, and carnitine acetyltranslocase (15-17). As shown in Fig. 7, we propose that as mitochondrial acetyl-coA levels increase (such as by a stimulation of PDC), acetyl groups are transferred to acetylcarnitine via carnitine acetyltransferase, where they are subsequently transported into the cytosol. As a result, an increase in intramitochondrial acetyl-

CYTOSOL MITOCHONDRIAL MATRIX

FAlTY ACIDS V A l T Y ACYL CO

MALONYL COA 1, I I t FAlTY ACYLCARNITINE ‘ FArrY ACYlfARNITlNE

CARNITINE

FAlTY ACYL COA C O p

FIG. 7. Proposed relationship between pyruvate dehydrogenase, acetyl-coA carboxylase, and oxidative metabolism of fatty acids and glucose in the heart. 1, carnitine palmitoyltransferase 1; 2, carnitine-acylcarnitine translocase; 3, pyruvate dehydrogenase complex; 4, @-oxidation; 5, carnitine acetyltransferase; 6, carnitine-acetylcarnitine translocase; 7, acetyl-coA carboxylase. In- creasing pyruvate dehydrogenase complex activity (3) will increase the supply of acetyl-coA for carnitine acetyltransferase (5) and the short chain carnitine carrier system ( 6 ) . As a result, cytosolic acetyl- CoA levels increase, resulting in an increase in acetyl-coA carboxylase activity (7). Increased malonyl-CoA production will then inhibit carnitine palmitoyltransferase 1 activity (1 ), resulting in a decrease in fatty acid oxidation ( 4 ) .

CoA production due to DCA stimulation of the PDC can result in an increase in cytosolic levels of acetyl-coA. Unfor- tunately, technical limitations do not allow for an accurate measure of cytosolic versus intramitochondrial acetyl-coA levels in our perfused hearts (see Ref. 25 for discussion of limitations). The export of acetyl groups from mitochondrial to cytosolic CoA presents an attractive hypothesis for the feedback regulation of fatty acid oxidation. Under conditions of low work, a decrease in acetyl-coA demand by the tricarboxylic acid cycle could result in a shuttling of these groups into the cytosol where acetyl-coA activates ACC resulting in an inhibition of fatty acid oxidation. In contrast, under conditions of high work, an increase in acetyl-coA demand would result in a decrease in ACC activity and therefore, an increase in fatty acid oxidation rates.

It cannot be discounted that an increase in acetyl-coA production from PDC may directly inhibit &oxidation of fatty acids via an inhibition of ketoacyl-CoA thiolase. This, however, does not explain the increase in malonyl-CoA levels we observed in DCA-treated hearts. Further support for the export of acetyl-coA to cytosol resulting in ACC activation comes from our recent work with carnitine-supplemented hearts. Increasing intracellular carnitine levels was shown to increase glucose oxidation in fatty acid-perfused hearts (18), which presumably occurred due to a stimulation of carnitine acetyltransferase, and a lowering of intramitochondrial acetyl-coA (relieving inhibition of PDC). This stimulation of glucose oxidation was accompanied by a parallel decrease in fatty acid oxidation, which cannot be explained by an inhibition of thiolase (i.e. intramitochondrial acetyl-coA levels decrease under these conditions). We hypothesized that carnitine effects can be explained by a stimulation of carnitine acetyltransferase, resulting in an increased transport of acetyl groups from the mitochondria to the cytosol. Increased cytosolic acetyl-coA levels could then stimulate ACC and malonyl-CoA production, resulting in a decrease in fatty acid oxidation at the level of CPT 1 (Fig. 7). To date, however, we have not determined the effects of carnitine supplementation on myocardial malonyl-CoA levels.

Accompanying the decrease in fatty acid oxidation in DCA- treated hearts was an increased incorporation of fatty acids in neutral lipids. This is consistent with a malonyl-CoA induced decrease in CPT 1 activity, since decreasing fatty acid oxidation can redirect fatty acids in the form of acyl- CoA toward complex lipid synthesis.

Cardiac levels of malonyl-CoA reported in this study are comparable to previously reported values of 4-5 nmol/g wet weight (16-20 nmol/g dry weight) (40, 41). Although the increase in malonyl-CoA levels in DCA-treated hearts was significant (compared to untreated perfused hearts), this increase was only around 40%. However, and as shown in Fig. 3, cardiac CPT 1 is highly sensitive to inhibition by malonyl- CoA. It is possible, therefore, that even with a slight change in malonyl-CoA concentration in the heart, a dramatic change in fatty acid oxidation could be expected. In addition, these data suggest that most of the malonyl-CoA present in the heart is inaccessible to CPT I. The malonyl CoA content we measured in these hearts is similar to levels previously reported (40,41). conversion of these values to a concentration would result in cytosolic levels of at least 5 pM. However, as McGarry et al. (41) have previously shown, and we confirm in Fig. 3, heart CPT 1 is extremely sensitive to malonyl CoA inhibition (the ICso is between 50 and 100 nM). As a result, if all of the malonyl-CoA present in the heart were accessible to CPT 1, fatty acid oxidation would always be completely inhibited. This of course is not the case, suggesting that most

Acetyl-coA Carboxylase in the Heart 25845

of the malonyl-CoA produced in the heart is inaccessible to CPT 1.

Until recently, the malonyl-CoA that is found in the heart was thought to be formed in the mitochondria by propionyl- CoA carboxylase which is abundant in the heart and has some affinity toward acetyl-coA as well as to its natural substrate propionyl-CoA (42). Scholte et al. (43), however, have detected a cytosolic Cop-fixing activity that was dependent on citrate in rat hearts suggestive of ACC activity in the heart. Recently Thampy (8) has shown that ACC is the enzyme involved in the synthesis of malonyl-CoA in the heart. Therefore, the increase in cardiac malonyl-CoA levels seen in DCA-treated hearts in this study is explained by an increase in ACC activity resulting from an increase in cytosolic acetyl-coA concentrations.

In summary, this study demonstrates, for the first time, a direct link between malonyl-CoA production by ACC and a decrease in myocardial fatty acid oxidation rates. Both the ACC-280 and ACC-265 are present in the heart, although ACC-280 predominates. Our data suggest that short term regulation of this isoform of ACC is regulated by acetyl-coA supply to the enzyme.

REFERENCES

1. Wakil, S. J., Titchener, E. B., and Gibson, D. M. (1958) Biochim. Biophys. Acta 29.225-226

2. Msbrouk, G. M., Helmy, I. M., Thampy, K. G., and Wakil, S. J. (1990) J.

4. Cook, G. A., and Gamble, M. S. (1987) J. Biol. Chem. 262,2050-2055 3. Cook, G. A. (1984) J. Bwl. Chem. 269,12030-12033

5. Paulson, D. J., Ward, K. M., and Shug, A. L. (1984) FEBS Lett. 176,381-

6. McCarry, J. D., Woeltje, K. F., Kuwajima, M., and Foster, D. W. (1989)

7. Bianchi, A., Evans, J. L., Iverson, A. J., Nordlund, A. C. , Watts, T. D., and

8. Thampy, K. G. (1989) J. BbL Chem. 264,17631-17634 9. Iverson, A. J., Bianchi, A., Nordlund, A. C., and Witters, L. A. (1990)

. . . . - - - , " " .

Bwl. Chem. 266,63304338

384

Diabetes/Met&lism Reus. 6 (31, 271-284

Witters, L. A. (1990) J. Biol. Chem. 266,1502-1509

Bwchem. J. 269,365-371

10.

11. 12.

13. 14.

15.

16.

17. 18.

20. 19.

21.

22. 23.

24.

25.

26. 27.

28.

30. 29.

31. 32.

33. 34. 35. 36.

37.

38.

39. 40.

41.

42. 43.

Trumble, G. E., Smith, M. A,, and Winder, W. W. (1991) Life Sci. 49,39-

Lopez-Casillas, F., and Kim, K.-H. (1989) J. Biol. Chem. 264,7176-7184 Kong, 1.-S., Lopez-Casillas, F., and Kim, K.-H. (1990) J. Biol. Chem. 266,

Cook, G. A., and Lappi, M. D. (1992) Mol. Cell. Biochem. 116,39-45 Lopaschuk, G. D., and Saddik, M. (1992) Mol. Cell. Biochem. 116, 111-

43

13695-13701

Lysiak, W., Lilly, K., DiLisa, F., Toth, P. P., and Bieber, L. L. (1988) J.

Lysiak, W., Toth, P. P., Suelter, C . H., and Bieber, L. L. (1986) J. Biol.

Pearson, D. J., and Tubbs, P. K. (1967) Biochem. J. 106,953-963 Broderick, T. L., Quinney, H. A., and Lopaschuk, G. D. (1992) J. Biol.

Saddik, M., and Lo aschuk, G. D. (1991) J. Biol. Chem. 266,8162-8170 Bowyer D. E. andkin J P (1977) J. Chromatogr. 143,743-790 Witters: L. A,: Watts, 4: D., Daniels, D. L., and Evans, J. L. (1988) Proc.

Bradford, M. M. (1976) Anal. Biochem. 72,248-254 King, M. T., Reiss, P. D., and Cornell, N. W. (1988) Methods Enzymol.

116

Bwl. Chem. 263,1151-1156

Chem. 261,13698-13703

Chem. 267,3758-3763

Natl. Acad. Sci. U. S. A. 86,5473-5477

1aa 7n-70 Tipper, J. P., and Witters, L. A. (1982) Biochim. Biophys. Acta 716, 162-

Idell-Wenger, J. A,, Grotyohann, L. W., and Neely, J. R. (1978) J. Biol.

Laemmli, U. K. (1970) Nature 227,680-685 Idell-Wenger, J. A,, Grotyohann, L. W., and Neely, J. R. (1982) Anal.

Bremer, J. (1981) Biochim. Biophys. Acta 666,628-631 Thamp K G , and Wakil, S. J. (1985) J. Biol. Chem. 260,6318-6323 Jamil, k , and'hladsen, N. B. (1987) J. Biol. Chem. 262,630-637 Jamil, H., and Madsen, N. B. (1987) J. Biol. Chem. 262,638-642 Holland, R., Witters, L. A., and Hardie, D. G. (1984) Eur. J. Biochem. 140,

"-, ." ." 169

Chem. 263,4310-4318

Biochern. 126,269-276

295-222 "" -"- Swensen, T. L., and Porter, J. W. (1985) J. Biol. Chem. 260,3791-3797 Thampy, K. G., and Wakil, S. J. (1988) J. Biol. Chem. 263,6454-6458 Witters, L. A. (1986) Adu. Protein Phosphutases 3, 201-224 Paw. M. E.. LoDez-Casillas. F.. and Kim. K.-H. (1988) Arch. Biochem. , ,

B&phys. 267,'104-109

Bwchem. 7, 121-141

J. Biochem. 203,615-623

. , Lane, M. D., Watkins, P. A,, and Meredith, M. J. (1979) CRC Crit. Rev.

Davies, S. P., Carling, D., Munday, M. R., and Hardie, D. G. (1992) Eur.

Williamson, J., and Corkey, B. (1979) Methods Enzymol. 66,200-222 Singh, B., Stakkestad, J. A., Bremer, J., and Borrebaek, B. (1984) Anal.

McGarrv. J. D.. Mills. S. E.. Lone. C . S.. and Foster. D. W. (1983) Biochem. Biochem. 138,107-111

J. 214, 21-28 . . -. . . ,

Hulsmann, W. C. (1966) Biochim. Biophys. Acta 178,137-144 Scholte, H. R., Luyt-Houwen, I. E. M., Dubelaar, M-L., and Hulsmann, W.

C. (1986) FEBS Lett. 198,47-50

vol. ,268. no. 34. issue december 5, pp. … of acetyl-coa and citrate. however, using affinity...

Documents