ANALYTICALBIOCHEMISTRY
185,274-278
(1990)
Radioisotopic Assays of CoASH and Carnitine and Their Acetylated Forms in Human Skeletal Muscle G. Cederblad,* J. I. Carlin, D. Constantin-Teodosiu,
P. Harper,? and E. Hultman
Departments of Clinical Chemistry* I and II, Karolinska Institute, Huddinge Hospital, S-141 86 Huddinge, and tSankt Gb?anHospital, S-112 82 Stockholm, Sweden
Received
June
26,1989
Radioisotopic assays for the determination of acetylCoA, CoASH, and acetylcarnitine have been modified for application to the amount of human muscle tissue that can be obtained by needle biopsy. In the last step common to all three methods, acetyl-CoA is condensed with [14C]oxaloacetate by citrate synthase to give [14C]citrate. For determination of CoASH, CoASH is reacted with acetylphosphate in a reaction catalyzed by phosphotransacetylase to yield acetyl-CoA. In the assay for acetylcarnitine, acetylcarnitine is reacted with CoASH in a reaction catalyzed by carnitine acetyltransferase to form acetyl-CoA. Inclusion of new simple steps in the acetylcarnitine assay and conditions affecting the reliability of all three methods are also described. Acetylcarnitine and free carnitine levels in human rectus abdominis muscle were 3.0 f 1.6 (SD) and 13.5 f 4.0 rmol/g dry wt, respectively. Values for acetyl-CoA and CoASH were about 500-fold lower, 6.7 + 1.8 and 21 f 8.9 nmol/g dry wt, respectively. A strong correlation between acetylcarnitine (y) and short-chain acylcarnitine (LX), determined as the difference between total and free carnitine, was found in biopsies from the vastus lateralis muscle obtained during intense muscular effort, y = 1.0~ + 0.6; P’ = 0.976. ~~~~~Aca~~emic~ress,~nc.
through the formation of acetylcarnitine in the matrix of mitochondria. The acetylcarnitine efflux from rat heart mitochondria was shown to be proportional over a wide range of carnitine concentrations, while in liver mitochondria it reached a plateau (2). In human muscle tissue, an increase in short-chain acylcarnitine occurs during endurance exercise (3). During intense muscular effort, acetylcarnitine is a major metabolite formed (4), apparently functioning as an acceptor for acetyl groups buffering excess formation from pyruvate decarboxylation and @-oxidation. Knowledge of the concentrations of acetyl-CoA and CoASH would be important for understanding the regulation of pyruvate utilization and mitochondrial oxidation. The aim of the present paper was to adapt the existing methodology for the measurement of these components to be suitable for the amount of muscle tissue obtained by the needle biopsy technique. The radioisotopic assays of acetyl-CoA (5) and CoASH (6), originally applied to liver tissue, have been adapted for human muscle tissue, and conditions affecting reliability are described. The sensitive radioisotopic method for acetylcarnitine determination described by Pande and Caramancion (5) and modified by Cooper et al. (7) has been further modified. MATERIALS
The most established role of carnitine is the transport of long-chain fatty acids into the mitochondrial matrix. As interest in the function of carnitine increased, it became apparent that carnitine had both a direct and an indirect role in intermediary metabolism. One involves conjugation of acyl residues to the B-hydroxyl group with subsequent translocation from one cellular compartment to another (1). This process affects both the availability of activated acyl residues and the availability of CoASH. Considerable evidence supports the conclusion that carnitine buffers the CoASH/acetyl-CoA ratio
AND METHODS
Chemicals. Sources of enzymes and chemicals were as follows: carnitine acetyltransferase, EC 2.3.1.7 (pigeon breast muscle, 5 mg/ml, 80 U/mg), and citrate synthase, EC 4.1.3.7 (pig heart, 2 mg/ml, 110 U/mg), were from Boehringer-Mannheim, (Mannheim, West Germany). Aspartate aminotransferase, EC 2.6.1.1 (porcine heart, 9 mg/ml, 240 U/mg), and phosphotransacetylase, EC 2.3.1.8 (Clostridium kluyveri, 3000 U/mg protein), were from Sigma Chemical Co. (St. Louis, MO). Acetylphosphate (lithium-potassium salt, 90% pure), free CoASH (sodium salt, 95% pure), acetyl-CoA (sodium
274 All
0003~2697/90 $3.00 Copyright 0 1990 by Academic Press, Inc. rights of reproduction in any form reserved.
RADIOISOTOPIC
ASSAYS
OF
CoASH
AND
salt, 95% pure), acetyl-DL-carnitine hydrochloride, Nethylmaleimide (NEM),l cation-exchange resin Dowex (5OW-X8, H’ form, 100-200 mesh), and dithiothreitol were from Sigma Chemical Co. L-[U-*4C]Aspartic acid (225 mCi/mmol) was from Amersham (Buckinghamshire, UK). All other chemicals were of Analar or equivalent grade. Reagents. Phosphotransacetylase was suspended in 1 ml of 3.2 mM ammonium sulfate, pH 8.0, and stored refrigerated in suitable portions. Before assay it was freshly diluted 1:lO with distilled water. Aspartate aminotransferase was diluted 1:20 and citrate synthase and carnitine acetyltransferase were diluted 1:lO with distilled water. These enzymes were portioned into suitable aliquots and kept frozen. Interference due to endogenous oxaloacetate and citrate was eliminated by using potassium acetate and CuS04 as described by Pande and Caramancion (5). A mixture of 1 mM &SO,, 400 mM K+ acetate, and water (3+3+2, by volume) was prepared. Other reagents were prepared as described by Cooper et al. (7) for acetylcarnitine and acetyl-CoA and by Pande and Caramancion (5) for CoASH. Preparation of muscle extracts. Muscle samples were initially obtained from (a) the rectus abdominis muscle during surgery on 10 patients suffering mainly from uncomplicated gallstone disease or (b) the vastus lateralis muscle by needle biopsy (8) of two healthy subjects during different degrees of aerobic exercise. The muscle tissue was rapidly frozen in liquid nitrogen and stored at -70°C. The samples were handled as described by Harris et al. (9). Briefly, they were freeze-dried, dissected free of connective tissue and blood, and then powdered. Usually about 40-60 mg of wet muscle tissue, i.e., 8-12 mg of freeze-dried dissected powder, was obtained. Five to ten milligrams of the powder was extracted with 0.5 mM perchloric acid in 1 mM NazEDTA and neutralized with 2.1 M KHC03. The volume of perchloric acid was adjusted to the weight of muscle powder, so that about 1 ml of neutralized supernatant per 10 mg dry wt was obtained. All samples were run in duplicate and the following volumes (in ~1) were taken for acetyl-CoA, CoASH, acetylcarnitine, and free carnitine, respectively; 200,100,25 (diluted 1:lO to 1:lOO with water), and 10 (cf. below). The total volume of muscle perchloric acid supernatant utilized was approximately 650 ~1. Acetyl-CoA assay. The complete isotope assay for acetyl-CoA determination is composed of two enzymatic reactions: (a) L- [ 14C]aspartate + 2-oxoglutarate + [ 14C] oxaloacetate + L-glutamate, catalyzed by aspartate aminotransferase, and (b) [ 14C]oxaloacetate + acetylCoA = [14C]citrate + CoASH, catalyzed by citrate 1 Abbreviation
used: NEM,
N-ethylmaleimide.
CARNITINE
IN
HUMAN
SKELETAL
MUSCLE
275
synthase. [14C]Oxaloacetate was freshly prepared according to Cooper et al. (7). The following solutions were added to test tubes: 200 yl of standard solutions of acetyl-CoA (O-125 pmol) or muscle perchloric acid supernatant, 2 ~1 of 0.1 M dithiothreitol, and 20 ~1 of the CuS04K+ acetate mixture. After incubation for 30 min at room temperature, the tubes were placed at 4°C for 10 min. EDTA (20 ~1,60 mM) was added, and the tubes were left for 5 min at room temperature. NEM (30 ~1,30 mM) was then added and, after 5 min at room temperature, 20 ~1 of [ 14C]oxaloacetate was added. The reaction was started with 10 ~1 of citrate synthase followed by an incubation for 20 min. Unreacted [14C]oxaloacetate was then transaminated to aspartate, and [14C]citrate was separated as described by Cooper et al. (7). Radioactivity was determined in a 0.5-ml aliquot of the supernatant, to which was added 7 ml of scintillation fluid.
CoASH assay. CoASH was determined by use of the following enzymatic reaction: CoASH + acetylphosphate = acetyl-CoA + phosphate, catalyzed byphosphotransacetylase, and acetyl-CoA was determined as reported above. The procedure of Pande and Caramancion (5) was used for CoASH determination. Briefly, 100 ~1 of standard solutions (O-125 pmol) or perchloric acid supernatant, 40 ~1 of 440 mM Tris buffer-HCl, pH 7.4,2 ~1 of 0.1 mM dithiothreitol, and 50 ~1 of 25 mM acetylphosphate were mixed in glass tubes and incubated for 20 min after the addition of 10 ~1 of phosphotransacetylase. The tubes were then heated for 3 min at 95°C and cooled in an ice bath for 5 min. Twenty microliters of the CuS04K+ acetate mixture was then added, and the procedure described above for acetyl-CoA was followed. The CoASH stock solution was standardized by the spectrophotometric method of Michal and Bergmeyer (10). Acetykarnitine assay. The muscle perchloric acid supernatant was diluted with water (from 1:lO to 1:lOO). Into glass tubes, 10 X 55 cm, 25 ~1 of standard solutions, containing O-125 pmol of acetylcarnitine (corresponding to the L-form), or muscle perchloric acid supernatant, 125 ~1 of HzO, 50 ~1 of 0.5 mM Hepes, 20 ~1 of 11 mM EDTA, 30 ~1 of freshly reduced 0.2 mM CoA (7), and 10 ~1 of carnitine acetyltransferase were added to start the reaction. After incubation for 30 min, the tubes were heated for 3 min at 95°C and then transferred to an ice bath. Thirty microliters of 30 mM NEM was added, and the procedure followed was thereafter the same as that described for the acetyl-CoA assay. The acetylcarnitine values were not corrected for endogenous acetyl-CoA present in tissue samples due to the great difference in concentration in muscle tissue of the two compounds. Carnitine and free carnitine assay. Carnitine was assayed by an enzymatic radioisotope method (ll), modified as described previously (12). For total carnitine, 10 ~1 of the supernatant was mixed with 100 ~1 of 0.1 M
276
CEDERBLAD
ET
TABLE Effect
of Varying
Added” Carnitine b-nol)
Proportions
of CoASH,
Measured
Acetylcarnitine (pmol)
75 60 45 30 15 0
Acetylcarnitine (pmol)
0 15 30 45 60 75
a pmol/assay. * CoASH is calculated
the values
measured
Carnitine,
AND
Methodological
3.2 2.9 3.5 2.0 3.2
Note. To one volume
95 91 100 92 98
on the
Assays
Added
Calculated*
Acetyl-CoA bmol)
CoASH (pmol)
CoASH (pm4
100 95 90 80 60 40 20 0
0 5 10 20 40 60 80 100
0 6 11 20 40 63 82 104
100 80 60 40 20 0
0 20 40 60 80 100
1 24 40 64 76 98
in CoASH
assay minus
TABLE
Recovery of added acetyl-CoA” (%I
Forms
Acetyl-CoA bnol)
The same preparation of [14C]oxaloacetate was used for assays of acetyl-CoA, CoASH, and acetylcarnitine. The standard amount used contained 0 to 125 pmol in the final step of all assays and the highest standard resulted in a radioactivity of about 18,000 cpm per assay. The radioactivity in the zero standard was used as a blank in each assay and was about 1500,1900, and 3000 cpm for acetyl-CoA, CoASH, and carnitine, respectively. The low acetyl-CoA concentration in muscle tissue re-
Basal acetyl-CoA bmol)
Acetylated
Acetyl-CoA bmol)
Consideration
of Components
Their
CoASH bmol)
DISCUSSION
Recovery
and
Measured
KOH, digested for 2 h at 5O”C, and neutralized with 20 ~1 of 0.5 M HCl. For the free carnitine determination, 120 ~1 of water was added to 10 ~1 of supernatant. Shortchain acylcarnitine was calculated as the difference between total carnitine, obtained after alkaline hydrolysis, and free carnitine. RESULTS
1
Added
0 16 31 46 61 75
from
AL.
Added Basal CoASH (pmol) 3.4 3.9 2.1 2.9 5.4
to Perchloric
the values
by acetyl-CoA
assay.
quired acetyl-CoA standards containing only O-50 pmol (O-O.25 pmol/liter). Acetyl-CoA and CoA Assays Acetyl-CoA could be correctly determined in the presence of CoASH with the acetyl-CoA assay (Table 1). The CoASH assay gives the sum of acetyl-CoA and CoASH. CoASH was obtained as the difference between these two measurements (Table 1). Standard curves for both assays were linear up to 125 pmol. Increasing volumes of muscle perchloric acid supernatants showed a linear response in concentrations when tested up to 200 ~1 in the acetyl-CoA assay and up to 100 ~1 in the CoASH assay, i.e., to the volumes routinely taken. Recoveries of acetyl-CoA and CoASH added to muscle perchloric acid supernatants were over 90% (Table 2). Elimination of Endogenous Citrate and Oxaloacetate To eliminate the interference of endogenous oxaloacetate and citrate, the muscle perchloric acid supernatants
2 Acid
Supernatants
Recovery of added CoASH’ (%) 90 93 99 95 90
of Muscle
Tissue
Basal acetylcarnitine” (pmol)
Recovery of added acetylcarnitine’ (%o)
45 36 62 39
of muscle perchloric acid supernatant was added one volume of the added compound or water (= basal a 20 pm01 in 100 pl of water. b 25 pmol in 50 ~1 of water. c The muscle supernatant was diluted 1:lO before the addition of water or 37.5 pmol acetylcarnitine. 25 ~1 was taken.
93 90 91 99
value).
RADIOISOTOPIC
0
1
ASSAYS
2
4
3
Acetylcarnitine,
OF
CoASH
5
6
AND
CARNITINE
pmolll
FIG. 1. Standard curves for acetylcarnitine. Cooper et al. (7) (0) and the modified procedure
Run (w).
as described
by
and the standard solutions were treated with CuS04 at pH 5 (5). The addition of CuS04 at the concentration indicated had no effect on the standard curve, but higher concentrations of CuSO, and EDTA decreased the measured acetyl-CoA. Inclusion of 10 nmol citrate and 1.0 nmol oxaloacetate to the standard solutions did not influence the intercept and slope of the standard curve of acetyl-CoA.
Acetylcarnitine
Assay
The effect of destroying the carnitine acetyltransferase. Standard solutions of acetylcarnitine, containing 0 to 125 pmol, were assayed according to Cooper et al. (7). The standard curve was nonlinear, which probably was due to a reversal of the carnitine acetyltransferase reaction. This was prevented by destroying the enzyme with 3 min of heating before the addition of citrate synthase (Fig. 1). Despite this modification, a series of muscle perchloric acid supernatant dilutions did not produce a linear concentration response (Fig. 2). A linear response was obtained when reduced glutathione so-
IN
HUMAN
SKELETAL
MUSCLE
277
FIG. 3. Comparison of acetylcarnitine assayed by the direct acetylcarnitine assay and short-chain acylcarnitine values calculated as the difference between total and free carnitine determination. The same muscle perchloric acid supernatants were used in all three assays and obtained from 10 muscle biopsies during intense exercise. The regression line was y = 1.0~ + 0.5; r = 0.976.
lution was omitted (Fig. 2). This modification did not affect the standard curve (not shown). Influence of added carnitine and recovery experiments. The acetylcarnitine assay was unaffected by varying amounts of carnitine (Table 1). The recovery of added acetylcarnitine to muscle perchloric acid supernatant was between 90 and 100% (Table 2). Figure 3 shows the very good correlation between the present acetylcarnitine assay and the short-chain acylcarnitine values obtained by calculating the difference between total carnitine and free carnitine obtained by the radioenzymatic assay (11,12). The same muscle perchloric acid supernatants were used in all three assays. The extent to which propionylcarnitine and propionyl-CoA could interfere in the assay has been determined by Pande and Caramancion (5). Interference from propionyl esters did result but only when their amount was several-fold greater than that of the corresponding acetyl esters. Under most conditions, propionylcarnitine constitutes only a fraction of the amount of acetylcarnitine present in tissues, with the exception of that present in children with inborn errors of propionate metabolism.
Precision Data The coefficients of variation from two aliquots of perchloric supernatants of muscle tissue (n = 10-12) were 3.8,2.6, and 3.6% for acetyl-CoA, CoASH, and acetylcarnitine, respectively. Aliquots of the same muscle perchloric acid supernatant were run in consecutive series, and the coefficients of variation (100 X SD/mean) were in the above order 13.6,12.3, and 1.9%, n = 4-6. 0
2b
4b
6b
6b
100’
% FIG. 2. Effect of dilutions of muscle perchloric acid supernatants (100% is the first 1:lO dilution) on concentration assayed as described by Cooper et al. (7) (0) and according to the present modification (w).
Muscle Analysis in Human
Subjects
The results from studies of muscle biopsies taken during surgery are shown in Table 3. The acetyl-CoA and CoASH concentrations were about 500-fold lower than the acetylcarnitine and free carnitine concentrations. In
278
CEDERBLAD TABLE Muscle
No.
Acetyl-CoA (nmolk)”
1 2 3 4 5 6 7 8 9 10
11.0 6.8 6.4 7.1 7.5 5.6 4.2 5.5 6.2 6.7
Mean SD
6.7 1.8
Note. Muscle
CoASH bmol/d
Levels
*
of Acetyl-CoA,
CoASH,
ET
AL.
3 Acetylcarnitine,
and
Free carnitine (wol/d
Acetylcarnitine (mol/g)
Carnitine
Acetyl-CoA/CoASH ratio
Acetylcarnitinelfree carnitine ratio
29 12 17 37 14 23 25 11 15 29
5.5 2.4 5.3 1.2 3.6 1.9 1.3 3.7 2.9 2.6
12.6 8.9 17.9 14.5 8.5 15.7 18.1 12.2 8.4 18.9
0.38 0.58 0.39 0.19 0.53 0.24 0.17 0.51 0.42 0.23
0.44 0.27 0.30 0.08 0.43 0.12 0.07 0.30 0.34 0.15
21 8.9
3.0 1.5
13.5 4.0
0.36 0.15
0.25 0.14
tissue was obtained from rectus a g dry weight. * Calculated as the difference of the measured
abdominis CoASH
muscle
during
surgery.
described
under
Material
these patients, under the present conditions about a third was acetyl-CoA and two-thirds was free CoASH in abdominal muscle tissue. Previously reported values in human skeletal muscle tissue were 42 + 12 (SD) pmol/g noncollagen protein (sum of CoASH + acetyl-CoA), which is equivalent to approximately 60 pmol/g dry wt (13). This is higher than the sum of CoASH and acetylCoA found in abdominal muscle. In a preliminary study using the described method, the corresponding value was 54.1 f 7.9 pmol/g dry wt in the vastus lateralis muscle (unpublished). As the type of muscle used is not given in (13), a strict comparison cannot be made. The carnitine values were in the same range as those previously reported in rectus abdominis muscle, with a median value for total carnitine of 13.7 pmol/g dry wt., range 7.7-32.5 (14). Biopsies were also taken from the vastus lateralis muscle during aerobic exercise in order to obtain high acetylcarnitine concentrations (4). The increases in acetylcarnitine and acid-soluble short-chain acetylcarnitine were parallel under these experimental conditions (Fig. 3), implying that acetylcarnitine is the dominant short-chain carnitine ester formed during intense muscular effort.
ACKNOWLEDGMENTS The work was supported by grants from the Swedish Medical Research Council (7136 and 02647), the Karolinska Institute, the Swedish Sports Research Council, and the Swedish Work Environment
and Methods
and acetyl-CoA.
Fund (Grant 81/0173). Dr James I. Carlin is a recipient of a John F. Fogarty International Fellowship from the National Institutes of Health, Bethesda, Maryland. The expert technical assistance of Agneta Laveskog is acknowledged.
REFERENCES 1. Bremer, 2. Lysiak,
J. (1983)
Physiol. Reu. 63,1420-1480.
W., Lilly, K., DiLisa, F., Toth, P., and Bieber, J. Biol. Chem. 263,1151-1156. 3. Carlin, J. I., Reddan, W. G., Sanjak, M., and Hodach, Appl. Physiol. 61,1275-1278. 4. Harris,
R. C., Foster,
C. V. L., and H&man,
L. L. (1988) R. (1986)
J.
J. Appl.
E. (1987)
Physiol. 63,440-442. 5. Pande, 30-38.
S. V., and Caramancion,
M. N. (1981)
Anal. Biochem. 112,
6. Rabier, D., Briand, P., Petit, F. K., Kamoun, P., and Cathelineau, L. (1983) And. Biochem. 134,325-329. 7. Cooper, M. B., Forster, C. A., and Jones, D. A. (1986) Clin. Chim.
Acta 159,291-299. 8. Bergstriim, J. (1972) Stand. J. Clin. Lab. R. C., Hultman, E., and Nordesjo, 9. Harris,
Znuest. 14, Suppl. L. 0. (1974)
68. Scar&
J.
Clin. Lab. Invest. 33,109-120. G., and Bergmeyer, H. U. (1983) in Methods of Enzymatic 10. Michal, Analysis (Bergmeyer, H. U., Ed.), Vol. 7,3rd ed., pp. 166-169, Verlag Chemie, Weinheim. 11. Cederblad, 243. 12. Cederblad,
G., and Lindstedt, G., Finnstrom,
S. (1972)
Clin. Chin. Acta 37,235-
O., and Mirtensson,
J. (1982)
Biochem.
J. 27,260-265. 13. Carrol, J. E., Brooke, M. H., Villadiego, Trefz, J. I. (1986) Arch. Neural. 43,128131. 14. Cederblad, G., Lundbolm, K., and Lindstedt, Acta 53,311-321.
A., Norris,
B. J., and
S. (1974)
Clin. Chim.
185,274-278
(1990)
Radioisotopic Assays of CoASH and Carnitine and Their Acetylated Forms in Human Skeletal Muscle G. Cederblad,* J. I. Carlin, D. Constantin-Teodosiu,
P. Harper,? and E. Hultman
Departments of Clinical Chemistry* I and II, Karolinska Institute, Huddinge Hospital, S-141 86 Huddinge, and tSankt Gb?anHospital, S-112 82 Stockholm, Sweden
Received
June
26,1989
Radioisotopic assays for the determination of acetylCoA, CoASH, and acetylcarnitine have been modified for application to the amount of human muscle tissue that can be obtained by needle biopsy. In the last step common to all three methods, acetyl-CoA is condensed with [14C]oxaloacetate by citrate synthase to give [14C]citrate. For determination of CoASH, CoASH is reacted with acetylphosphate in a reaction catalyzed by phosphotransacetylase to yield acetyl-CoA. In the assay for acetylcarnitine, acetylcarnitine is reacted with CoASH in a reaction catalyzed by carnitine acetyltransferase to form acetyl-CoA. Inclusion of new simple steps in the acetylcarnitine assay and conditions affecting the reliability of all three methods are also described. Acetylcarnitine and free carnitine levels in human rectus abdominis muscle were 3.0 f 1.6 (SD) and 13.5 f 4.0 rmol/g dry wt, respectively. Values for acetyl-CoA and CoASH were about 500-fold lower, 6.7 + 1.8 and 21 f 8.9 nmol/g dry wt, respectively. A strong correlation between acetylcarnitine (y) and short-chain acylcarnitine (LX), determined as the difference between total and free carnitine, was found in biopsies from the vastus lateralis muscle obtained during intense muscular effort, y = 1.0~ + 0.6; P’ = 0.976. ~~~~~Aca~~emic~ress,~nc.
through the formation of acetylcarnitine in the matrix of mitochondria. The acetylcarnitine efflux from rat heart mitochondria was shown to be proportional over a wide range of carnitine concentrations, while in liver mitochondria it reached a plateau (2). In human muscle tissue, an increase in short-chain acylcarnitine occurs during endurance exercise (3). During intense muscular effort, acetylcarnitine is a major metabolite formed (4), apparently functioning as an acceptor for acetyl groups buffering excess formation from pyruvate decarboxylation and @-oxidation. Knowledge of the concentrations of acetyl-CoA and CoASH would be important for understanding the regulation of pyruvate utilization and mitochondrial oxidation. The aim of the present paper was to adapt the existing methodology for the measurement of these components to be suitable for the amount of muscle tissue obtained by the needle biopsy technique. The radioisotopic assays of acetyl-CoA (5) and CoASH (6), originally applied to liver tissue, have been adapted for human muscle tissue, and conditions affecting reliability are described. The sensitive radioisotopic method for acetylcarnitine determination described by Pande and Caramancion (5) and modified by Cooper et al. (7) has been further modified. MATERIALS
The most established role of carnitine is the transport of long-chain fatty acids into the mitochondrial matrix. As interest in the function of carnitine increased, it became apparent that carnitine had both a direct and an indirect role in intermediary metabolism. One involves conjugation of acyl residues to the B-hydroxyl group with subsequent translocation from one cellular compartment to another (1). This process affects both the availability of activated acyl residues and the availability of CoASH. Considerable evidence supports the conclusion that carnitine buffers the CoASH/acetyl-CoA ratio
AND METHODS
Chemicals. Sources of enzymes and chemicals were as follows: carnitine acetyltransferase, EC 2.3.1.7 (pigeon breast muscle, 5 mg/ml, 80 U/mg), and citrate synthase, EC 4.1.3.7 (pig heart, 2 mg/ml, 110 U/mg), were from Boehringer-Mannheim, (Mannheim, West Germany). Aspartate aminotransferase, EC 2.6.1.1 (porcine heart, 9 mg/ml, 240 U/mg), and phosphotransacetylase, EC 2.3.1.8 (Clostridium kluyveri, 3000 U/mg protein), were from Sigma Chemical Co. (St. Louis, MO). Acetylphosphate (lithium-potassium salt, 90% pure), free CoASH (sodium salt, 95% pure), acetyl-CoA (sodium
274 All
0003~2697/90 $3.00 Copyright 0 1990 by Academic Press, Inc. rights of reproduction in any form reserved.
RADIOISOTOPIC
ASSAYS
OF
CoASH
AND
salt, 95% pure), acetyl-DL-carnitine hydrochloride, Nethylmaleimide (NEM),l cation-exchange resin Dowex (5OW-X8, H’ form, 100-200 mesh), and dithiothreitol were from Sigma Chemical Co. L-[U-*4C]Aspartic acid (225 mCi/mmol) was from Amersham (Buckinghamshire, UK). All other chemicals were of Analar or equivalent grade. Reagents. Phosphotransacetylase was suspended in 1 ml of 3.2 mM ammonium sulfate, pH 8.0, and stored refrigerated in suitable portions. Before assay it was freshly diluted 1:lO with distilled water. Aspartate aminotransferase was diluted 1:20 and citrate synthase and carnitine acetyltransferase were diluted 1:lO with distilled water. These enzymes were portioned into suitable aliquots and kept frozen. Interference due to endogenous oxaloacetate and citrate was eliminated by using potassium acetate and CuS04 as described by Pande and Caramancion (5). A mixture of 1 mM &SO,, 400 mM K+ acetate, and water (3+3+2, by volume) was prepared. Other reagents were prepared as described by Cooper et al. (7) for acetylcarnitine and acetyl-CoA and by Pande and Caramancion (5) for CoASH. Preparation of muscle extracts. Muscle samples were initially obtained from (a) the rectus abdominis muscle during surgery on 10 patients suffering mainly from uncomplicated gallstone disease or (b) the vastus lateralis muscle by needle biopsy (8) of two healthy subjects during different degrees of aerobic exercise. The muscle tissue was rapidly frozen in liquid nitrogen and stored at -70°C. The samples were handled as described by Harris et al. (9). Briefly, they were freeze-dried, dissected free of connective tissue and blood, and then powdered. Usually about 40-60 mg of wet muscle tissue, i.e., 8-12 mg of freeze-dried dissected powder, was obtained. Five to ten milligrams of the powder was extracted with 0.5 mM perchloric acid in 1 mM NazEDTA and neutralized with 2.1 M KHC03. The volume of perchloric acid was adjusted to the weight of muscle powder, so that about 1 ml of neutralized supernatant per 10 mg dry wt was obtained. All samples were run in duplicate and the following volumes (in ~1) were taken for acetyl-CoA, CoASH, acetylcarnitine, and free carnitine, respectively; 200,100,25 (diluted 1:lO to 1:lOO with water), and 10 (cf. below). The total volume of muscle perchloric acid supernatant utilized was approximately 650 ~1. Acetyl-CoA assay. The complete isotope assay for acetyl-CoA determination is composed of two enzymatic reactions: (a) L- [ 14C]aspartate + 2-oxoglutarate + [ 14C] oxaloacetate + L-glutamate, catalyzed by aspartate aminotransferase, and (b) [ 14C]oxaloacetate + acetylCoA = [14C]citrate + CoASH, catalyzed by citrate 1 Abbreviation
used: NEM,
N-ethylmaleimide.
CARNITINE
IN
HUMAN
SKELETAL
MUSCLE
275
synthase. [14C]Oxaloacetate was freshly prepared according to Cooper et al. (7). The following solutions were added to test tubes: 200 yl of standard solutions of acetyl-CoA (O-125 pmol) or muscle perchloric acid supernatant, 2 ~1 of 0.1 M dithiothreitol, and 20 ~1 of the CuS04K+ acetate mixture. After incubation for 30 min at room temperature, the tubes were placed at 4°C for 10 min. EDTA (20 ~1,60 mM) was added, and the tubes were left for 5 min at room temperature. NEM (30 ~1,30 mM) was then added and, after 5 min at room temperature, 20 ~1 of [ 14C]oxaloacetate was added. The reaction was started with 10 ~1 of citrate synthase followed by an incubation for 20 min. Unreacted [14C]oxaloacetate was then transaminated to aspartate, and [14C]citrate was separated as described by Cooper et al. (7). Radioactivity was determined in a 0.5-ml aliquot of the supernatant, to which was added 7 ml of scintillation fluid.
CoASH assay. CoASH was determined by use of the following enzymatic reaction: CoASH + acetylphosphate = acetyl-CoA + phosphate, catalyzed byphosphotransacetylase, and acetyl-CoA was determined as reported above. The procedure of Pande and Caramancion (5) was used for CoASH determination. Briefly, 100 ~1 of standard solutions (O-125 pmol) or perchloric acid supernatant, 40 ~1 of 440 mM Tris buffer-HCl, pH 7.4,2 ~1 of 0.1 mM dithiothreitol, and 50 ~1 of 25 mM acetylphosphate were mixed in glass tubes and incubated for 20 min after the addition of 10 ~1 of phosphotransacetylase. The tubes were then heated for 3 min at 95°C and cooled in an ice bath for 5 min. Twenty microliters of the CuS04K+ acetate mixture was then added, and the procedure described above for acetyl-CoA was followed. The CoASH stock solution was standardized by the spectrophotometric method of Michal and Bergmeyer (10). Acetykarnitine assay. The muscle perchloric acid supernatant was diluted with water (from 1:lO to 1:lOO). Into glass tubes, 10 X 55 cm, 25 ~1 of standard solutions, containing O-125 pmol of acetylcarnitine (corresponding to the L-form), or muscle perchloric acid supernatant, 125 ~1 of HzO, 50 ~1 of 0.5 mM Hepes, 20 ~1 of 11 mM EDTA, 30 ~1 of freshly reduced 0.2 mM CoA (7), and 10 ~1 of carnitine acetyltransferase were added to start the reaction. After incubation for 30 min, the tubes were heated for 3 min at 95°C and then transferred to an ice bath. Thirty microliters of 30 mM NEM was added, and the procedure followed was thereafter the same as that described for the acetyl-CoA assay. The acetylcarnitine values were not corrected for endogenous acetyl-CoA present in tissue samples due to the great difference in concentration in muscle tissue of the two compounds. Carnitine and free carnitine assay. Carnitine was assayed by an enzymatic radioisotope method (ll), modified as described previously (12). For total carnitine, 10 ~1 of the supernatant was mixed with 100 ~1 of 0.1 M
276
CEDERBLAD
ET
TABLE Effect
of Varying
Added” Carnitine b-nol)
Proportions
of CoASH,
Measured
Acetylcarnitine (pmol)
75 60 45 30 15 0
Acetylcarnitine (pmol)
0 15 30 45 60 75
a pmol/assay. * CoASH is calculated
the values
measured
Carnitine,
AND
Methodological
3.2 2.9 3.5 2.0 3.2
Note. To one volume
95 91 100 92 98
on the
Assays
Added
Calculated*
Acetyl-CoA bmol)
CoASH (pmol)
CoASH (pm4
100 95 90 80 60 40 20 0
0 5 10 20 40 60 80 100
0 6 11 20 40 63 82 104
100 80 60 40 20 0
0 20 40 60 80 100
1 24 40 64 76 98
in CoASH
assay minus
TABLE
Recovery of added acetyl-CoA” (%I
Forms
Acetyl-CoA bnol)
The same preparation of [14C]oxaloacetate was used for assays of acetyl-CoA, CoASH, and acetylcarnitine. The standard amount used contained 0 to 125 pmol in the final step of all assays and the highest standard resulted in a radioactivity of about 18,000 cpm per assay. The radioactivity in the zero standard was used as a blank in each assay and was about 1500,1900, and 3000 cpm for acetyl-CoA, CoASH, and carnitine, respectively. The low acetyl-CoA concentration in muscle tissue re-
Basal acetyl-CoA bmol)
Acetylated
Acetyl-CoA bmol)
Consideration
of Components
Their
CoASH bmol)
DISCUSSION
Recovery
and
Measured
KOH, digested for 2 h at 5O”C, and neutralized with 20 ~1 of 0.5 M HCl. For the free carnitine determination, 120 ~1 of water was added to 10 ~1 of supernatant. Shortchain acylcarnitine was calculated as the difference between total carnitine, obtained after alkaline hydrolysis, and free carnitine. RESULTS
1
Added
0 16 31 46 61 75
from
AL.
Added Basal CoASH (pmol) 3.4 3.9 2.1 2.9 5.4
to Perchloric
the values
by acetyl-CoA
assay.
quired acetyl-CoA standards containing only O-50 pmol (O-O.25 pmol/liter). Acetyl-CoA and CoA Assays Acetyl-CoA could be correctly determined in the presence of CoASH with the acetyl-CoA assay (Table 1). The CoASH assay gives the sum of acetyl-CoA and CoASH. CoASH was obtained as the difference between these two measurements (Table 1). Standard curves for both assays were linear up to 125 pmol. Increasing volumes of muscle perchloric acid supernatants showed a linear response in concentrations when tested up to 200 ~1 in the acetyl-CoA assay and up to 100 ~1 in the CoASH assay, i.e., to the volumes routinely taken. Recoveries of acetyl-CoA and CoASH added to muscle perchloric acid supernatants were over 90% (Table 2). Elimination of Endogenous Citrate and Oxaloacetate To eliminate the interference of endogenous oxaloacetate and citrate, the muscle perchloric acid supernatants
2 Acid
Supernatants
Recovery of added CoASH’ (%) 90 93 99 95 90
of Muscle
Tissue
Basal acetylcarnitine” (pmol)
Recovery of added acetylcarnitine’ (%o)
45 36 62 39
of muscle perchloric acid supernatant was added one volume of the added compound or water (= basal a 20 pm01 in 100 pl of water. b 25 pmol in 50 ~1 of water. c The muscle supernatant was diluted 1:lO before the addition of water or 37.5 pmol acetylcarnitine. 25 ~1 was taken.
93 90 91 99
value).
RADIOISOTOPIC
0
1
ASSAYS
2
4
3
Acetylcarnitine,
OF
CoASH
5
6
AND
CARNITINE
pmolll
FIG. 1. Standard curves for acetylcarnitine. Cooper et al. (7) (0) and the modified procedure
Run (w).
as described
by
and the standard solutions were treated with CuS04 at pH 5 (5). The addition of CuS04 at the concentration indicated had no effect on the standard curve, but higher concentrations of CuSO, and EDTA decreased the measured acetyl-CoA. Inclusion of 10 nmol citrate and 1.0 nmol oxaloacetate to the standard solutions did not influence the intercept and slope of the standard curve of acetyl-CoA.
Acetylcarnitine
Assay
The effect of destroying the carnitine acetyltransferase. Standard solutions of acetylcarnitine, containing 0 to 125 pmol, were assayed according to Cooper et al. (7). The standard curve was nonlinear, which probably was due to a reversal of the carnitine acetyltransferase reaction. This was prevented by destroying the enzyme with 3 min of heating before the addition of citrate synthase (Fig. 1). Despite this modification, a series of muscle perchloric acid supernatant dilutions did not produce a linear concentration response (Fig. 2). A linear response was obtained when reduced glutathione so-
IN
HUMAN
SKELETAL
MUSCLE
277
FIG. 3. Comparison of acetylcarnitine assayed by the direct acetylcarnitine assay and short-chain acylcarnitine values calculated as the difference between total and free carnitine determination. The same muscle perchloric acid supernatants were used in all three assays and obtained from 10 muscle biopsies during intense exercise. The regression line was y = 1.0~ + 0.5; r = 0.976.
lution was omitted (Fig. 2). This modification did not affect the standard curve (not shown). Influence of added carnitine and recovery experiments. The acetylcarnitine assay was unaffected by varying amounts of carnitine (Table 1). The recovery of added acetylcarnitine to muscle perchloric acid supernatant was between 90 and 100% (Table 2). Figure 3 shows the very good correlation between the present acetylcarnitine assay and the short-chain acylcarnitine values obtained by calculating the difference between total carnitine and free carnitine obtained by the radioenzymatic assay (11,12). The same muscle perchloric acid supernatants were used in all three assays. The extent to which propionylcarnitine and propionyl-CoA could interfere in the assay has been determined by Pande and Caramancion (5). Interference from propionyl esters did result but only when their amount was several-fold greater than that of the corresponding acetyl esters. Under most conditions, propionylcarnitine constitutes only a fraction of the amount of acetylcarnitine present in tissues, with the exception of that present in children with inborn errors of propionate metabolism.
Precision Data The coefficients of variation from two aliquots of perchloric supernatants of muscle tissue (n = 10-12) were 3.8,2.6, and 3.6% for acetyl-CoA, CoASH, and acetylcarnitine, respectively. Aliquots of the same muscle perchloric acid supernatant were run in consecutive series, and the coefficients of variation (100 X SD/mean) were in the above order 13.6,12.3, and 1.9%, n = 4-6. 0
2b
4b
6b
6b
100’
% FIG. 2. Effect of dilutions of muscle perchloric acid supernatants (100% is the first 1:lO dilution) on concentration assayed as described by Cooper et al. (7) (0) and according to the present modification (w).
Muscle Analysis in Human
Subjects
The results from studies of muscle biopsies taken during surgery are shown in Table 3. The acetyl-CoA and CoASH concentrations were about 500-fold lower than the acetylcarnitine and free carnitine concentrations. In
278
CEDERBLAD TABLE Muscle
No.
Acetyl-CoA (nmolk)”
1 2 3 4 5 6 7 8 9 10
11.0 6.8 6.4 7.1 7.5 5.6 4.2 5.5 6.2 6.7
Mean SD
6.7 1.8
Note. Muscle
CoASH bmol/d
Levels
*
of Acetyl-CoA,
CoASH,
ET
AL.
3 Acetylcarnitine,
and
Free carnitine (wol/d
Acetylcarnitine (mol/g)
Carnitine
Acetyl-CoA/CoASH ratio
Acetylcarnitinelfree carnitine ratio
29 12 17 37 14 23 25 11 15 29
5.5 2.4 5.3 1.2 3.6 1.9 1.3 3.7 2.9 2.6
12.6 8.9 17.9 14.5 8.5 15.7 18.1 12.2 8.4 18.9
0.38 0.58 0.39 0.19 0.53 0.24 0.17 0.51 0.42 0.23
0.44 0.27 0.30 0.08 0.43 0.12 0.07 0.30 0.34 0.15
21 8.9
3.0 1.5
13.5 4.0
0.36 0.15
0.25 0.14
tissue was obtained from rectus a g dry weight. * Calculated as the difference of the measured
abdominis CoASH
muscle
during
surgery.
described
under
Material
these patients, under the present conditions about a third was acetyl-CoA and two-thirds was free CoASH in abdominal muscle tissue. Previously reported values in human skeletal muscle tissue were 42 + 12 (SD) pmol/g noncollagen protein (sum of CoASH + acetyl-CoA), which is equivalent to approximately 60 pmol/g dry wt (13). This is higher than the sum of CoASH and acetylCoA found in abdominal muscle. In a preliminary study using the described method, the corresponding value was 54.1 f 7.9 pmol/g dry wt in the vastus lateralis muscle (unpublished). As the type of muscle used is not given in (13), a strict comparison cannot be made. The carnitine values were in the same range as those previously reported in rectus abdominis muscle, with a median value for total carnitine of 13.7 pmol/g dry wt., range 7.7-32.5 (14). Biopsies were also taken from the vastus lateralis muscle during aerobic exercise in order to obtain high acetylcarnitine concentrations (4). The increases in acetylcarnitine and acid-soluble short-chain acetylcarnitine were parallel under these experimental conditions (Fig. 3), implying that acetylcarnitine is the dominant short-chain carnitine ester formed during intense muscular effort.
ACKNOWLEDGMENTS The work was supported by grants from the Swedish Medical Research Council (7136 and 02647), the Karolinska Institute, the Swedish Sports Research Council, and the Swedish Work Environment
and Methods
and acetyl-CoA.
Fund (Grant 81/0173). Dr James I. Carlin is a recipient of a John F. Fogarty International Fellowship from the National Institutes of Health, Bethesda, Maryland. The expert technical assistance of Agneta Laveskog is acknowledged.
REFERENCES 1. Bremer, 2. Lysiak,
J. (1983)
Physiol. Reu. 63,1420-1480.
W., Lilly, K., DiLisa, F., Toth, P., and Bieber, J. Biol. Chem. 263,1151-1156. 3. Carlin, J. I., Reddan, W. G., Sanjak, M., and Hodach, Appl. Physiol. 61,1275-1278. 4. Harris,
R. C., Foster,
C. V. L., and H&man,
L. L. (1988) R. (1986)
J.
J. Appl.
E. (1987)
Physiol. 63,440-442. 5. Pande, 30-38.
S. V., and Caramancion,
M. N. (1981)
Anal. Biochem. 112,
6. Rabier, D., Briand, P., Petit, F. K., Kamoun, P., and Cathelineau, L. (1983) And. Biochem. 134,325-329. 7. Cooper, M. B., Forster, C. A., and Jones, D. A. (1986) Clin. Chim.
Acta 159,291-299. 8. Bergstriim, J. (1972) Stand. J. Clin. Lab. R. C., Hultman, E., and Nordesjo, 9. Harris,
Znuest. 14, Suppl. L. 0. (1974)
68. Scar&
J.
Clin. Lab. Invest. 33,109-120. G., and Bergmeyer, H. U. (1983) in Methods of Enzymatic 10. Michal, Analysis (Bergmeyer, H. U., Ed.), Vol. 7,3rd ed., pp. 166-169, Verlag Chemie, Weinheim. 11. Cederblad, 243. 12. Cederblad,
G., and Lindstedt, G., Finnstrom,
S. (1972)
Clin. Chin. Acta 37,235-
O., and Mirtensson,
J. (1982)
Biochem.
J. 27,260-265. 13. Carrol, J. E., Brooke, M. H., Villadiego, Trefz, J. I. (1986) Arch. Neural. 43,128131. 14. Cederblad, G., Lundbolm, K., and Lindstedt, Acta 53,311-321.
A., Norris,
B. J., and
S. (1974)
Clin. Chim.
