Clinical Science (1988) 75,437-440
437
Free and total carnitine and acylcarnitine content of plasma,
urine, liver and muscle of alcoholics
C . DE SOUSA, N. W. Y. LEUNG, R. A. CHALMERS
AND
T. J. PETERS
Section of F'erinatal and Child Health and Division of Clinical Cell Biology, MRC Clinical Research Centre, Harrow, Middlesex, U.K.
(Received 14 July 1987/26 January 1988; accepted 29 February 1988)
SUMMARY
1. Plasma and urine free and total carnitine and acylcarnitine levels were assayed in 12 control subjects and
20 chronic alcoholics with fatty liver. Although the
alcoholics had a wider range of values than the controls,
there was no significant difference between the two
groups.
2. Hepatic free and total carnithe and long- and shortchain acylcarnitines were assayed by a radioenzymatic
method in samples from seven control subjects and seven
alcoholics. No significant differences in any of the indices
were noted between the patient and control groups and it
was concluded that carnitine deficiency did not contribute
to alcoholic fatty liver in patients without cirrhosis.
3. Skeletal muscle free and total carnitine and longand short-chain acylcarnitines were assayed in eight
alcoholics and seven control subjects. The alcoholics had
significantly higher total and free carnitine levels. It is
suggested that this reflects a selective enrichment of the
biopsy sample with type I carnitine-rich fibres due to the
type II fibre atrophy found in approximately half the
patients.
Key words: alcoholic myopathy, carnitine, fatty acid oxidation, fatty liver.
Abbreviation: Hepes, 4-(2-hydroxyethy1)- 1-piperazineethanesulphonic acid.
INTRODUCTION
Excessive consumption of alcohol commonly leads to
fatty liver, and a lipid storage myopathy may also occur
more rarely [ 11. Decreased fatty acid oxidation offers the
most likely explanation for the deposition of fat in the
liver [2], although the precise mechanism underlying the
decrease is unclear. Carnitine (B-hydroxy-y-trimethylCorrespondence: Dr C. de Sousa, Queen Mary's Hospital for
Children, Carshalton, Surrey SM5 4NR, U.K.
aminobutyric acid) is essential for the mitochondria1 Boxidation of fatty acids [3]. Human carnitine deficiency,
whether due to a primary defect in carnitine metabolism
[4, 51, or secondary to other disorders [6], may lead to a
lipid myopathy and to excessive fat deposition at other
sites including the liver. Carnitine deficiency could therefore also contribute to excessive fat accumulation in liver
and skeletal muscle of alcoholics.
Previous studies of plasma carnitine in both alcoholics
and experimental animals have given rise to conflicting
results [7-121 and tissue carnitines have not been
measured in alcoholic patients. This study set out to
measure plasma, urinary, liver and muscle carnitines in a
series of patients with early alcoholic liver disease.
METHODS
Subjects and protocol
Twenty alcoholic subjects were studied: 16 male and
four female patients with a history of alcohol intake in
excess of 100 g per day for more than 3 years. Patients
were not off alcohol at the time of study. The mean age of
the group was 45 years (range 24-62 years). The liver
biopsies showed variable degrees of fatty change. None of
the patients showed clinical or biopsy evidence of cirrhosis or clinical or biochemical evidence of malnutrition.
Blood samples were obtained after an overnight fast,
and 24 h collections of urine were also obtained from all
20 patients. Liver tissue was obtained from seven patients
(six males and one female, mean age 48 years, range
38-62 years) who underwent percutaneous needle
biopsy, and muscle tissue was obtained from eight
patients (seven males and one female, mean age 46 years,
range 40-62 years) who underwent needle biopsy of the
quadriceps muscle. Portions of all liver and muscle biopsies were also examined histologically.
The control group for plasma and urine samples consisted of 12 healthy adult volunteers with no history of
alcohol abuse (although most were light or moderate con-
C. de Sousa et al.
438
sumers of alcohol). There were seven males and five
females and the mean age of the group was 34 years (range
26-52 years). The control group from whom liver
samples were obtained consisted of seven patients with no
history of alcohol abuse (five male, two female, mean age
62 years, range 47-70 years) who were undergoing.liver
biopsy for diagnostic purposes. The comparison group
from whom muscle samples were obtained consisted of
seven patients with no history of alcohol abuse (four male,
three female, mean age 58 years, range 38-73 years) who
were either undergoing muscle biopsy for diagnostic purposes or had muscle removed at the time of surgery. The
liver and muscle biopsies from the control subjects were
histologically normal.
This project was approved by the Ethical Committee of
Harrow Health Authority and patients gave informed
consent to liver biopsy and muscle biopsy.
tion in 0.6 mol/l perchloric acid, in the acid-soluble phase
(‘free’carnitine), in alkali-hydrolysed samples (‘total acidsoluble’ carnitine) and in the acid-insoluble extract. The
difference between the values for total acid-soluble and
free carnitine was short-chain acylcarnitine. Long-chain
acylcarnitines were measured in the acid-insoluble
extract. The assay used a radioenzymatic method in which
the sample was incubated with [ l-14C]acetyl-CoA
together with carnitine acetyltransferase, Hepes buffer and
N-ethylmaleimide as a trap for free CoASH. The [1-I4C]acetylcarnitine formed in the reaction mixture was separated using a Dowex 2-X8 anion-exchange column,
counted and related to standard curves of L-carnitine and
acetylcarnitine. Liver and muscle carnitine was also
related to the protein content of the tissue as determined
by the method of Lowry et al. [14], with bovine serum
albumin as standard.
Reagents
Statistical analysis
Reagents were obtained from the following sources:
[ l-14C]acetyl-CoAfrom Amersham International, acetylCoA, Hepes [4-(2-hydroxyethyl)-l-piperazine-ethanesulphonic acid], N-ethylmaleimide, L-carnitine and carnitine
acetyltransferase from Sigma Chemical Company, acetylcarnitine from P-L Biochemicals, and Dowex 2-X8
200-400 mesh in chloride form from Bio Rad Laboratories.
Normally distributed data were compared by Student’s
t-test. Logarithmic transformation was required for some
variables.
Analytical methods
L-Carnitine was measured in plasma (separated and
frozen within an hour of collection), in urine (stored
frozen) and in liver and muscle biopsy tissue (collected
immediately into liquid nitrogen and stored at - 70°C) by
a modification of the method of Cederblad & Lindstedt
[ 131. Plasma and urine L-carnitine were assayed directly
(to measure ‘free’ carnitine) and in alkali-hydrolysed
samples (to measure ‘total’ carnitine). The difference
between these two values represented acylcarnitine. Liver
and muscle L-carnitine were assayed, after homogeniza-
RESULTS
In 20 alcoholics the concentration of total and free carnitine and acylcarnitines in fasting plasma and urine, and
the 2 4 h excretion of carnitine and acylcarnitines did not
differ significantly from normal subjects (Table 1),
although the results from the alcoholics spanned a much
wider range than those from the normal individuals. The
results of measurements of carnitine in percutaneous liver
biopsies obtained from seven alcoholics did not differ
significantly in total or free carnitine and long-chain or
short-chain acylcarnitines from the non-alcoholic group
(Table 2). There was no correlation between liver carnitine measurements and the degree of fat infiltration in the
biopsies.
The eight alcoholics in whom muscle carnitine was
measured had significantly greater levels of total and free
Table 1. Plasma and urine carnitines in alcoholics and normal subjects
Results are means k SD.
Normal subjects ( n= 12) Alcoholics ( n= 20)
Plasma concn. (pmol/l)
Total carnitine
Free carnitine
Ac ylcarnitines
41.7 f 5.9
35.6 f 5.4
6.1 f 3 . 2
48.1 f 11.9
39.5 f 1 1.3
8.6 f4.5
Urine carnitine/creatinine ratio (mmol/mol of creatinine)
Total carnitine
Free carnitine
Acylcarnitines
25.5 f 8.0
10.6 f 6 . 2
15.1 f 2 . 4
34.2 f 27.3
19.5 rt 21.3
14.7f6.9
9.8 f6.0
4.3 f 3.8
5.5 f 2.3
9.4 8.0
5.5 k 6.2
3.9f2.1
Urinary excretion (pmo1/24 h)
Total carnitine
Free carnitine
Acylcarnitines
*
Carnitine status in alcoholics
439
Table 2. Liver and muscle carnitines in alcoholics and normal subjects
Results are means iz SD. Statistical significance: * P < 0.02.
~ _ _ _ _ _
Liver
Normal subjectst ( n= 7)
Alcoholics ( n= 7)
Muscle
Normal subjects ( n= 7)
Alcoholics ( n= 8)
~~
Total
carnitine
(nmol/mg of protein)
Free
carnitine
(nmol/mg of protein)
Long-chain
acylcarnitine
(nmol/mgof protein)
Short-chain
acylcamitine
(nmol/mgof protein)
6.1 f 2.8
4.3 & 1.3
3.5 f 2.1
2.3 f 0.7
0.8 f 0.3
0.7 k 0.2
1.9f 1.0
1.3 f 0.8
13.3& 5.2
24.3 f 9:3*
1.0f0.5
1.4f0.8
2.1 & 1.8
4.7 f 5.0
16.5 f 5.5 ’
30.4f 11.7*
t Liver biopsies were obtained for diagnostic purposes from patients with the following disorders: diabetes (two subjects),carcinoma
of the stomach, suspected primary biliary cirrhosis, hyperlipidaemia, unexplained abdominal pain and hepatomegaly. Histology was
normal in each case.
carnitine than the control group. They also showed
increased levels of long-chain and short-chain acylcarnitines but these did not reach statistical significance. This
group included four alcoholics who had either plasma
total carnitine or urinary total carnitine or both above the
95% confidence limits for normal subjects. These four
also had the highest measurements of muscle total carnitine. None of the muscle biopsies from the alcoholics
showed fat infiltration on histological examination. However, four of the biopsies showed a degree of type I1 fibre
atrophy. There was no correlation between muscle carnitine measurements and the degree of fibre atrophy.
DISCUSSION
This study showed that a group of alcoholics with fatty
liver without cirrhosis had normal levels of plasma, urine
and hepatic total and free carnitine and acylcarnitine.
There are a number of studies of carnitine in alcoholics
which have given variable results. Bohmer et al. [7]
measured plasma free carnitine in patients with liver
cirrhosis (two of whom had primary biliary cirrhosis).The
results from this group of patients could not be
distinguished from those of the control group. In a study
in which serum total carnitine was surveyed in a large
group of hospitalized patients, hypocarnitinaemia was
found only in 60 patients with advanced alcoholic cirrhosis [lo]. All of these patients had evidence of protein
calorie malnutrition. Post-mortem examination of tissue
from eight cases revealed low levels of muscle, liver,
kidney and brain carnitine and increased muscle and liver
triacylglycerols. The authors concluded that hypocarnitinaemia was due to dietary insufficiency of precursors for endogenous carnitine synthesis as well as
hepatocellular disease which blocks the normal pathway
for the synthesis of carnitine. Another study of alcoholic
liver cirrhosis had quite different findings [ 111. Plasma
carnitines were elevated in these patients, primarily due to
high levels of acylcarnitines. A further study of a group of
alcoholics without liver disease found them also to have
significantly raised plasma carnitines [ 151.
In addition to the studies of ethanol and carnitine in
man, there have been a number of animal studies. These
have concentrated on three main areas: the effects of
acute and chronic ethanol administration on carnitine
content; the effect of supplemental carnitine on triacylglycerol accumulation; and the influence of carnitine on
some of the neurotoxic effects of ethanol. Kondrup &
Grunnet [81 found that acute administration of ethanol
increased hepatic free carnitines and acylcarnitines but
reduced hepatic long-chain acylcarnitines. Prolonged
ethanol treatment was thought not to cause any additional
changes. Bode et al. [9]found that the carnitine content of
rat livers was unaffected by chronic ethanol administration alone. In rats receiving a protein-restricted diet, however, there was a rise in hepatic acylcarnitines; this rise
was more marked if these rats also received ethanol.
Sachan et al. [ 121 found that prolonged administration of
ethanol to rats reduced plasma total and free carnitine.
The protective effects of carnitine on liver lipid
metabolism after ethanol administration were investigated
by Hosein & Bexton [16]. They found that carnitine
lowered serum triacylglycerols in all expenmental animals
and in addition reduced hepatic triacylglycerols in rats
receiving ethanol. Sachan et al. [12] found that carnitine
abolished the rise in plasma and hepatic triacylglycerols
that occurred in animals receiving prolonged ethanol
treatment. A study of the neurotoxic effects of ethanol in
rats [ 171found a reduced incidence of post-ethanol withdrawal seizures in animals who also received carnitine.
Ethanol causes a significant decrease in hepatic fatty
acid oxidation and this could be due to carnitine deficiency at the hepatocyte mitochondria1level or to inhibition of carnitine acyltransferase by ethanol [ 181.
However, the recent observations that ketone body formation from radiolabelled palmitate is normal in patients
with fatty liver [2] does not suppo? these suggestions.
Reduced fatty acid oxidation appears to be due to
impaired tricarboxylic acid cycle activity, possibly secondary to ethanol induced redox changes [19-221 as well as
loss of cycle intermediates [2].
The results of the present study do not suggest a primary role for carnitine in the pathogenesis of alcoholic
fatty liver. In particular, liver carnitines were normal in
these patients. However, the high levels of plasma carnitines that were observed in some patients may have been a
440
C. de Sousa et al.
response to an impaired ability to oxidize fatty acids by
one of the mechanisms outlined above. Carnitine effectively modulates the intramitochondrial acetyl-CoA/CoA
ratio [23, 241. If a reduction in tricarboxylic acid cycle
activity was the cause of impaired fatty acid oxidation this
would be expected to lead to a rise in intramitochondrial
acetyl-CoA, and a decrease in the availability of free
CoA. Such a rise would be buffered by carnitine. The
result would be an increase in acylcarnitines. On the other
hand, if a decrease in the activity of carnitine acyltransferases was the cause of impaired fatty acid oxidation in
these patients a rise in the extramitochondrial pool of free
carnitine would be expected. However, the alcoholics in
whom plasma carnitines were elevated had high levels of
free and acylcarnitines. Liver carnitines were normal in
these patients and the foregoing account cannot fully
explain this finding.
The high carnitine content of muscle in the alcoholics
was an unexpected finding. Chronic alcoholism is associated with muscle fibre atrophy, particularly of type 11
fibres [l].Half of the biopsies obtained from alcoholics in
the present study showed evidence of fibre atrophy. It is
likely that type I1 muscle fibres have a lower carnitine
content than type I fibres, which use free fatty acids as a
fuel. Evidence for this is provided by the observation that
muscle carnitine deficiency is associated with atrophy and
lipid vacuolation in type I fibres [4]. Muscle fibre atrophy
in alcoholics would therefore be expected to lead to a preponderance of carnithe-richer type I fibres, and a relative
rise in the carnitine content of muscle.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
REFERENCES
1. Martin, F., Ward, K., Slavin, G., Levi, A.J. & Peters, T.J.
(1985) Alcoholic skeletal myopathy; a clinical study. Quarterly Journal of Medicine, 55, 233-25 1.
2. Leung, N.W.Y. & Peters, T.J. ( 1985) Palmitic acid oxidation
and incorporation into triglyceride by needle liver biopsy
specimens from control subjects and patients with alcoholic
fatty liver disease. Clinical Science, 71,253-260.
3. Fritz, I.B. (1955) The effects of muscle extracts on the
oxidation of palmitic acid by liver slices and homogenates.
Acta Physiologica Scandinavica, 34,367-385.
4. Engel, A.G. & Angelini, C. (1973) Carnitine deficiency of
human skeletal muscle with associated lipid storage
myopathy: a new syndrome. Science, 179,899-902.
5. Karpati, G., Carpenter, S., Engel, A.G., Watters, G., Allen,
J., Rothman, S., Klassen, G. & Mamer, O.A. (1975) The
syndrome of systemic carnitine deficiency. Neurology, 25,
16-24.
4 . DiDonato, S., Rimoldi, M., Garvaglia, B. & Uziel, G. (1984)
Propionylcarnitine excretion in propionic and methyl-
19.
20.
21.
22.
23.
24.
malonic acidurias: a cause of carnitine deficiency. Clinica
Chimica Acta, 139, 13-21.
Bohmer, T., Rydning, A. & Solberg, H.E. (1974) Carnitine
levels in human serum in health and disease. Clinica
Chimica Acta, 57,55-61.
Kondrup, J. & Grunnet, N. (1973) The effect of acute and
prolonged ethanol treatment on the contents of coenzyme
A. carnitine and their derivatives in rat liver. Biochemical
Journal, 132, 373-379.
Bode. C.. Kono. H. & Bode. J.C. 11978) Effect of chronic
ethanol administration on hepatic content of coenzyme A,
carnitine, and their acyl esters in rats fed a standard diet or a
diet with low protein content. Hoppe-Seyler’s Zeitschriftfuer
Physiologische Chemie, 359, 140 1- 1406.
Rudman, D., Ansley, J.D. & Sewell, C.W. (1980) Carnitine
deficiency in cirrhosis. In: Carnitine Biosynthesis, Metabolism and Functions, pp. 307-319. Ed. Frenkel, R.A. &
McGarry, J.D. Academic Press, New York.
Fuller, R.K. & Hoppel, C.L. (1983) Elevated plasma
carnitine in hepatic cirrhosis. Hepatology, 3,554-558.
Sachan, D.S., Rhen, T.H. & Ruark, R.A. ( 1 984) Ameliorating effect of camitine and its precursors on alcohol-induced
fatty liver. American Journal of Clinical Nutrition, 39,
738-744.
Cederblad, G. & Lindstedt, S. (1972) A method for the
determination of carnitine in the picomole range. Clinica
Chimica Acta, 37,235-243.
Lowry, O.H., Rosenborough, N.J., Farr, A.L. & Randall, R.J.
(1951) Protein measurement with the Folin-phenol reagent.
Journal of Biological Chemistry, 193,265-267.
Fuller, R.K. & Hoppel, C.L. ( 1985) Plasma carnitine in alcoholism and alcoholic cirrhosis. Alcoholism: Clinical and
Experimental Research, 2,200.
Hosein, E.A. & Bexton, B. (1975) Protective action of
carnitine on liver lipid metabolism after ethanol administration to rats. Biochemical Pharmacology, 24, 1859- 1863.
Corbett, R. & Leonard, B.E. ( 1984) Effects of carnitine on
changes caused by chronic administration of alcohol.
Neuropharmacology, 23,269-27 1.
Parker, S.L.,Thompson, J.A. & Reitz, R.C. (1974)Effects of
chronic ethanol ingestion upon acyl-CoA: carnitine acyltransferase in liver and heart. Lipids, 9,520-525.
Lindros, K.O. (1970) Suppression by ethanol of acetate and
hexanoate oxidation studied by non-recirculating liver perfusion. Biochemical and Biophysical Research Communications, 41,635-641.
Blomstrand, R., Kager, L. & Lantto, 0. (1973) Studies on
the ethanol-induced decrease of fatty acid oxidation in rats
and human liver. Life Sciences, 13, 1 131- 1141.
Ontko, J.A. (1973) Effects of ethanol on the metabolism of
free fatty acids in isolated liver cells. Journal of Lipid
Research, 14, 78-86.
Lieber, C.S. (1985) Alcohol and the liver: metabolism of
ethanol, metabolic effects and pathogenesis o f injury. Acta
Medica Scandinavica, Suppl. 703, 11-55.
Pearson, D.J. & Tubbs, P.K. (1967) Carnitine and derivatives in rat tissue. Biochemical Journal, 105,953-963.
Roe, C.R., Hoppel, C.L., Stacey, T.E., Chalmers, R.A. &
Tracey, B.M. (1983) Metabolic response to carnitine in
methylmalonic aciduria. Archives of Disease in Childhood,
58,9 16-920.
I
,