- Email: [email protected]
In vitro analysis of hepatic carnitine biosynthesis in human systemic carnitine deficiency
295
Clinica Chimica Acta, 106 (1980) 295-300 0 Elsevier/North-Holland Biomedical Press
CCA
1483
IN VITRO ANALYSIS OF HEPATIC CARNITINE HUMAN SYSTEMIC CARNITINE DEFICIENCY
CHARLES
J. REBOUCHE
* and ANDREW
February
26th,
IN
G. ENGEL
Neuromuscular Disease Research Center and Department Mayo Foundation, Rochester, MN 55901 (U.S.A.) (Received
BIOSYNTHESIS
of Neurology,
Mayo
Clinic and
1980)
Summary The syndrome of systemic carnitine deficiency (progressive muscle weakness, recurrent metabolic encephalopathy, low liver and muscle and fluctuating serum carnitine levels) has been attributed to a defect of carnitine biosynthesis. We determined activities in liver of the four enzymes which convert e-N&methyl-L-lysine to L-carnitine in three patients with systemic carnitine deficiency and in 12 control subjects. In the three patients all enzyme activities were within the normal range except one, which was slightly below the normal range. We conclude that in systemic carnitine deficiency no enzymatic defect exists in the conversion of e-N-trimethyl-L-lysine to carnitine.
Introduction Human systemic carnitine deficiency is a highly fatal disease characterized episodes of metabolic encephalopathy, by progressive muscle weakness, decreased muscle and liver and fluctuating serum carnitine levels and tissue triglyceride excess at some stages of the disease [l-4]. The etiology of this disease is uncertain. Several hypotheses have emerged, including (a) a defect in carnitine biosynthesis, (b) excessive catabolism of camitine, or (c) excessive renal excretion of camitine [4]. Carnitine is derived ultimately from lysine and methionine [5]. Lysine residues in a wide variety of proteins [6] are methylated by protein methylase III [7] and e-N-trimethyl-L-lysine residues are released by protein hydrolysis. e-N-Trimethyl-L-lysine is hydroxylated at the P-position [8] and cleaved between carbons 2 and 3 yielding glycine [9] and y-trimethylaminobutyraldehyde. The latter compound is oxidized to form y-butyrobetaine [lo] which
* To
whom
correspondence
should
be addressed.
296
is hydroxylated at the P-position to yield L-carnitine [ 111. The enzymes catalyzing these reactions, termed e-N-trimethyl-L-lysine /3-hydroxylase, /3-hydroxy-e-N-trimethyl-L-lysine aldolase, y-trimethylaminobutyraldehyde dehydrogenase, and y-butyrobetaine hydroxylase (EC 1.14.11.1), were measured in liver specimens of control subjects and patients with systemic camitine deficiency. Materials
and methods
Control subjects Liver specimens from 12 subjects were obtained at autopsy. Age and cause of death of each subject are listed in Table I. There was no clinical or microscopic evidence of liver disease in 11 of 12 cases. In one infant’s liver centrolobular necrosis affecting not more than 10% of the tissue was noted. Tissues were collected 2-18 h after death, frozen in isopentane chilled by liquid nitrogen to -15O”C, and stored in liquid nitrogen until analyzed.
Patients Three children with typical systemic carnitine deficiency were studied. Patient one was an ll-year-old male, patient two a 5-year-old female, and patient three a 4-year-old male. Clinical features of the disease observed in patients one and two are described by Karpati et al. [l] and Glasgow et al. [ 31, respectively. Liver specimens were obtained by open biopsy, frozen in isopentane at -150°C and stored in liquid nitrogen.
Tissue preparation
and enzyme
activity
determination
Methods for homogenization of tissues, preparation of subcellular fractions and determination of enzyme activities have been previously described [ 121. All enzymes were assayed under conditions of linearity with respect to time of incubation and protein concentration (cf. Figs. 2 and 3 [12]).
TABLE I DATA ON CONTROL
SUBJECTS
Subjects
Age at death/sex
Cause of death
Adults
15 yrs (F)
automobile accident motorcycle accident carbon monoxide poisoning automobile accident automobile accident carbon monoxide poisoning train accident coronary artery disease head injuries at delivery abdominal wall infection crib death died during surgery to correct congenital heart malformation
16 22 23 25 31 34 Infants
Child
ws yrs yrs yrs yrs yrs
(M) (M) (F) (M) (M) (M)
61 YTS(M) 59 hrs (F) * 24 days (F) 3 mths (M) 2.5 YI‘S (M)
* Centrolobular necrosis affecting less than 10% of the liver was noted at autopsy. No microscopic abnormalities were found in the other specimens.
297
Results
and discussion
Enzyme activities measured in control subjects and patients with systemic carnitine deficiency are summarized in Table II and shown graphically in Fig. 1. c-N-Trimethyl-L-lysine &hydroxylase activity is associated with the mitochondrial fraction of the liver cell [8,12]. We employed washed mitochondrial fractions to measure this activity in human liver. In patients one and two, activity of this enzyme was within the normal range, but in patient three the enzyme activity was slightly lower than normal. Activities of P-hydroxyc-N-trimethyl-L-lysine aldolase, y-trimethylaminobutyraldehyde dehydrogenase and y-butyrobetaine hydroxylase were measured in soluble protein fractions. Activities of these enzymes in all three patients with systemic carnitine deficiency fell within or above the normal range (Table II). It is of interest that liver y-butyrobetaine hydroxylase activity in infants and children is significantly lower than in adults. We have shown previously [12] that liver y-butyrobetaine hydroxylase activity shows age dependence which is described by the equation y = a3tb, where y is the enzyme activity and x is the age of the subject (a = 0.139, b = 0.287, r = 0.95, p < 0.001). /3-Hydroxy-c-N-trimethyl-L-lysine aldolase activity varied considerably in control subjects (5.8-143 pmol product formed per min per mg protein). To exclude the possibility that the wide variation was an artifact of tissue preparation or enzyme assay, we added (at the time of homogenization or enzyme determination) 0.1 mmol/l phenylmethylsulfonyl fluoride (PMSF), a protease inhibitor, or 0.01-l mmol/l pyridoxal phosphate, a cofactor for most aldolasetype enzymes. Enzyme activity was not increased by these additions. Thus, if our results had been affected by an artifact, then this artifact must have originated between the time of death and freezing of the tissue. However, no correlation was found between enzyme activity and interval between death and autopsy. In this regard it is significant to note than in the patients with systemic carnitine deficiency fl-hydroxy-c-N-trimethyl-L-lysine aldolase activity was relatively constant and in the upper range of control values. Rudman et al. [ 131 found tissue and serum carnitine deficiency in cachecticcirrhotic patients. They fed patients with advanced cirrhosis and normal subjects a camitine-free diet rich in lysine and methionine. After 12 days for the cirrhotic patients the mean serum carnitine level and the mean urinary carnitine excretion were 25% and 15% respectively, of the control means. These results indicate impaired carnitine biosynthesis in patients with liver disease. In our study, the control subjects had no clinical or microscopical evidence for liver disease. (In one infant centrolobular necrosis was noted at autopsy, but this involved less than 10% of the tissue.) We studied one male subject (not included in the control group in Tables I and II) who died of aleukemic leukemia and had severe anemia for several months before death. His liver showed no necrosis or fat infiltration. Hepatic enzyme levels in this subject were : c-N-trimethyl-I.,-lysine P-hydroxylase, 29.6; y-trimethylaminobutyraldehyde dehydrogenase, 38 300; and y-butyrobetaine hydroxylase, 11 pmol . aldolase activity was not min-l * mg-1 protein. /3-Hydroxy-c-N-trimethyl-L-lysine measured. The marked decrease in y-butyrobetaine hydroxylase activity in this patient indicates that severe anemia can lower the activity of at least one of the
II
OF CARNITINE
pm01 formed
* Activity Activity
**
was measured was measured
-
38.7 k 48.0 (6) 5.8-143
111 74.9 76.3
in liver washed mitochondrial fractions [ 121. in 105 000 X g supernatant fractions of human liver [121.
191
34.2 zt 7.8 (3) 26.2-44.8
Infant controls mean * SEM (N) range
Child control
45.6 + 15.5 (7) 28.6-77.9
43.9 62.2 21.2
LIVER
P-Hydroxy-e-N-trimethylL-lysine aldolase * *
protein
IN HUMAN
min/mg
ENZYMES
E-A’-trimethyl-lysine fl-hydroxylase *
Activity,
BIOSYNTHETIC
mean f SEM (N) range
Adult controls
Patient one Patient two Patient three
Subject
ACTIVITY
TABLE
63 900
25 700 (1)
46 800 + 36 500-66
34 500 71 000 67 800 9 200 (7) 300
y-trimethylaminobutyraldehyde dehydrogenase **
102
hydroxylase
2.6 (3) 47
f 95 (8) -488 43.3 * 41 -
369 209
157 303 243
y-butyrobetaine
**
299
Fig. 1. Comparison
of activities
of carnitine
biosynthetic
enzymes
in patients
with systemic
carnitine
defi-
ciency and in control subjects. Panel A, f-N-trimethyl-L-lysine P-hydroxylase; panel B. P-hydroxy-E-N-trimethyl-L-lysine aldolase; panel C, y-trimethylaminobutyraldehyde dehydrogenase; and panel D, ?-butyrobetaine hydroxylase. Activities are designated by circles for adults. squares for children and triangles for infants.
hepatic enzymes which subserve camitine biosynthesis. The results of this study do not exclude a defect in the conversion of lysine to e-N-trimethyl-L-lysine in systemic camitine deficiency. Neurosporu crassa can methylate free lysine [ 141 but this reaction has not been demonstrated in mammalian systems. Therefore, it is generally believed that in mammals e-N-trimethyl-L-lysine destined for camitine biosynthesis is derived from e-N-trimethyl-L-lysine residues in proteins [ 15-171. The significance of methylated amino acids in proteins is presently unclear; presumably they play a role in the structural, chemical, or functional integrity of the protein molecule. Because c-N-trimethyl-L-lysine residues occur in a wide variety of proteins [6], it is unlikely that the absence of this residue from a single protein would impair carnitine biosynthesis. Absence of e-N-trimethyl-L-lysine from all proteins in which it normally occurs would very likely lead to multiple clinical manifestations due to a variety of structurally altered proteins, rather than just carnitine deficiency. The results of this study are consistent with the conclusion that human systemic carnitine deficiency is not the result of a defect in camitine biosynthesis. We are currently investigating possible causes of this disease, including abnormal degradation, excretion or cellular transport of camitine. Acknowledgement This work was supported by a Research Center Dystrophy Association and NIH Grant NS 6277.
Grant from
the Muscular
300
References 1 Karpati, G., Carpenter, S., Engel, A.G.. Watters. G.. Allen, J., Rothman. S., Klassen. G. and Mamer. O.A. (1975) The syndrome of systemic carnitine deficiency. Neural. (Minneap.) 25, 16-24 2 Scholte. H.R.. Meijer, A.E.F.H., Van Wijngaarden, G.K. and Leaders, K.L. (1979) Familial carnitine deficiency. A fatal case and subclinical state in a sister. J. Neural. Sci. 42, 87-101 3 Glasgow, A.M.. Eng, G. and Engel. A.G. (1980) Systemic carnitine deficiency simulating recurrent Reye’s syndrome. J. Pediatr. 98. 889-891 4 Engel, A.G. (1980) Possible causes and effects of carnitine deficiency in man. In Biosynthesis, Metabolism and Functions of Carnitine (Frenkel, R.A. and McGarry, J.D.. eds.). Academic Press, New York, PP. 271-284 5 Tanphaichitr. V. and Broquist, H.P. (1973) Role of lysine and c-N-trimethyllysine in carnitinc biosynthesis. II. Studies in rat. J. Biol. Chem. 248. 2176-2181 6 Paik. W.K. and Kim, S. (1975) Protein methylation: chemical, eruymological and biological significance. Adv. Enzymol. 42, 227-286 7 Paik. W.K. and Kim, S. (1970) Solubilization and partial purification of protein methylase III from calf thymus nuclei. J. Biol. Chem. 245. 6010-6015 8 Hulse. J.D., Ellis, S.R. and Henderson, L.M. (1978) Carnitine biosynthesis. P-Hydroxylation of trimethyllysine by a a-ketoglutarate-dependent mitochondrial dioxygenase. J. Biol. Chem. 253, 16541659 9 Hochalter, J.B. and Henderson, L.M. (1976) Carnitine biosynthesis. The formation of glycine from carbons 1 and 2 of 6-N-trimethyl-L-lysine. Biochem. Biophys. Res. Commun. 70, 364-366 10 Hulse. J.D. and Henderson, L.M. (1980) Carnitine biosynthesis. Purification of 4-N-trimethylaminobutyraldehyde dehydrogenase from beef liver. J. Biol. Chem. 255, 1146-1151 11 Lindstedt. G. (1967) Hydroxylation of y-butyrobetaine to carnitine in rat liver. Biochem. 6. 12711282 12 Rebouche, C.J. and Engel. A.G. (1980) Tissue distribution of carnitine biosynthetic enzymes in man. Biochim. Biophys. Acta 630, 22-29 13 Rudman. D., Swewell, C.W. and Ansley, J.D. (1977) Deficiency of carnitine in cachectic cirrhotic patients. J. Clin. Invest. 60. 716-723 14 Rebouche, C.J. and Broquist, H.P. (1976) Carnitine biosynthesis in Neurospora CPCISSU: enzymatic conversion of lysine to e-N-trimethyllysine. J. Bacterial. 126. 1207-1214 15 Cox, R.A. and Hoppel, C.L. (1973) Biosynthesis of carnitine and 4-trimethylaminobutyrate from 6-N-trimethyllysine. Biochem. J. 136, 1083-1090 16 LaBadie. J., Dunn, W.A. and Aronson, N.N.. Jr. (1976) Hepatic synthesis of carnitine from proteinbound trimethyl-lysine. Lysosomal digestion of methyl-lysine-labeled asialo-fetuin. Biochem. J. 160, 85-95 17 Paik, W.K.. Nochumson, S. and Kim. S. (1977) Carnitine biosynthesis via protein methylation. Trends Biochem. Sci. 2. 159-161
Clinica Chimica Acta, 106 (1980) 295-300 0 Elsevier/North-Holland Biomedical Press
CCA
1483
IN VITRO ANALYSIS OF HEPATIC CARNITINE HUMAN SYSTEMIC CARNITINE DEFICIENCY
CHARLES
J. REBOUCHE
* and ANDREW
February
26th,
IN
G. ENGEL
Neuromuscular Disease Research Center and Department Mayo Foundation, Rochester, MN 55901 (U.S.A.) (Received
BIOSYNTHESIS
of Neurology,
Mayo
Clinic and
1980)
Summary The syndrome of systemic carnitine deficiency (progressive muscle weakness, recurrent metabolic encephalopathy, low liver and muscle and fluctuating serum carnitine levels) has been attributed to a defect of carnitine biosynthesis. We determined activities in liver of the four enzymes which convert e-N&methyl-L-lysine to L-carnitine in three patients with systemic carnitine deficiency and in 12 control subjects. In the three patients all enzyme activities were within the normal range except one, which was slightly below the normal range. We conclude that in systemic carnitine deficiency no enzymatic defect exists in the conversion of e-N-trimethyl-L-lysine to carnitine.
Introduction Human systemic carnitine deficiency is a highly fatal disease characterized episodes of metabolic encephalopathy, by progressive muscle weakness, decreased muscle and liver and fluctuating serum carnitine levels and tissue triglyceride excess at some stages of the disease [l-4]. The etiology of this disease is uncertain. Several hypotheses have emerged, including (a) a defect in carnitine biosynthesis, (b) excessive catabolism of camitine, or (c) excessive renal excretion of camitine [4]. Carnitine is derived ultimately from lysine and methionine [5]. Lysine residues in a wide variety of proteins [6] are methylated by protein methylase III [7] and e-N-trimethyl-L-lysine residues are released by protein hydrolysis. e-N-Trimethyl-L-lysine is hydroxylated at the P-position [8] and cleaved between carbons 2 and 3 yielding glycine [9] and y-trimethylaminobutyraldehyde. The latter compound is oxidized to form y-butyrobetaine [lo] which
* To
whom
correspondence
should
be addressed.
296
is hydroxylated at the P-position to yield L-carnitine [ 111. The enzymes catalyzing these reactions, termed e-N-trimethyl-L-lysine /3-hydroxylase, /3-hydroxy-e-N-trimethyl-L-lysine aldolase, y-trimethylaminobutyraldehyde dehydrogenase, and y-butyrobetaine hydroxylase (EC 1.14.11.1), were measured in liver specimens of control subjects and patients with systemic camitine deficiency. Materials
and methods
Control subjects Liver specimens from 12 subjects were obtained at autopsy. Age and cause of death of each subject are listed in Table I. There was no clinical or microscopic evidence of liver disease in 11 of 12 cases. In one infant’s liver centrolobular necrosis affecting not more than 10% of the tissue was noted. Tissues were collected 2-18 h after death, frozen in isopentane chilled by liquid nitrogen to -15O”C, and stored in liquid nitrogen until analyzed.
Patients Three children with typical systemic carnitine deficiency were studied. Patient one was an ll-year-old male, patient two a 5-year-old female, and patient three a 4-year-old male. Clinical features of the disease observed in patients one and two are described by Karpati et al. [l] and Glasgow et al. [ 31, respectively. Liver specimens were obtained by open biopsy, frozen in isopentane at -150°C and stored in liquid nitrogen.
Tissue preparation
and enzyme
activity
determination
Methods for homogenization of tissues, preparation of subcellular fractions and determination of enzyme activities have been previously described [ 121. All enzymes were assayed under conditions of linearity with respect to time of incubation and protein concentration (cf. Figs. 2 and 3 [12]).
TABLE I DATA ON CONTROL
SUBJECTS
Subjects
Age at death/sex
Cause of death
Adults
15 yrs (F)
automobile accident motorcycle accident carbon monoxide poisoning automobile accident automobile accident carbon monoxide poisoning train accident coronary artery disease head injuries at delivery abdominal wall infection crib death died during surgery to correct congenital heart malformation
16 22 23 25 31 34 Infants
Child
ws yrs yrs yrs yrs yrs
(M) (M) (F) (M) (M) (M)
61 YTS(M) 59 hrs (F) * 24 days (F) 3 mths (M) 2.5 YI‘S (M)
* Centrolobular necrosis affecting less than 10% of the liver was noted at autopsy. No microscopic abnormalities were found in the other specimens.
297
Results
and discussion
Enzyme activities measured in control subjects and patients with systemic carnitine deficiency are summarized in Table II and shown graphically in Fig. 1. c-N-Trimethyl-L-lysine &hydroxylase activity is associated with the mitochondrial fraction of the liver cell [8,12]. We employed washed mitochondrial fractions to measure this activity in human liver. In patients one and two, activity of this enzyme was within the normal range, but in patient three the enzyme activity was slightly lower than normal. Activities of P-hydroxyc-N-trimethyl-L-lysine aldolase, y-trimethylaminobutyraldehyde dehydrogenase and y-butyrobetaine hydroxylase were measured in soluble protein fractions. Activities of these enzymes in all three patients with systemic carnitine deficiency fell within or above the normal range (Table II). It is of interest that liver y-butyrobetaine hydroxylase activity in infants and children is significantly lower than in adults. We have shown previously [12] that liver y-butyrobetaine hydroxylase activity shows age dependence which is described by the equation y = a3tb, where y is the enzyme activity and x is the age of the subject (a = 0.139, b = 0.287, r = 0.95, p < 0.001). /3-Hydroxy-c-N-trimethyl-L-lysine aldolase activity varied considerably in control subjects (5.8-143 pmol product formed per min per mg protein). To exclude the possibility that the wide variation was an artifact of tissue preparation or enzyme assay, we added (at the time of homogenization or enzyme determination) 0.1 mmol/l phenylmethylsulfonyl fluoride (PMSF), a protease inhibitor, or 0.01-l mmol/l pyridoxal phosphate, a cofactor for most aldolasetype enzymes. Enzyme activity was not increased by these additions. Thus, if our results had been affected by an artifact, then this artifact must have originated between the time of death and freezing of the tissue. However, no correlation was found between enzyme activity and interval between death and autopsy. In this regard it is significant to note than in the patients with systemic carnitine deficiency fl-hydroxy-c-N-trimethyl-L-lysine aldolase activity was relatively constant and in the upper range of control values. Rudman et al. [ 131 found tissue and serum carnitine deficiency in cachecticcirrhotic patients. They fed patients with advanced cirrhosis and normal subjects a camitine-free diet rich in lysine and methionine. After 12 days for the cirrhotic patients the mean serum carnitine level and the mean urinary carnitine excretion were 25% and 15% respectively, of the control means. These results indicate impaired carnitine biosynthesis in patients with liver disease. In our study, the control subjects had no clinical or microscopical evidence for liver disease. (In one infant centrolobular necrosis was noted at autopsy, but this involved less than 10% of the tissue.) We studied one male subject (not included in the control group in Tables I and II) who died of aleukemic leukemia and had severe anemia for several months before death. His liver showed no necrosis or fat infiltration. Hepatic enzyme levels in this subject were : c-N-trimethyl-I.,-lysine P-hydroxylase, 29.6; y-trimethylaminobutyraldehyde dehydrogenase, 38 300; and y-butyrobetaine hydroxylase, 11 pmol . aldolase activity was not min-l * mg-1 protein. /3-Hydroxy-c-N-trimethyl-L-lysine measured. The marked decrease in y-butyrobetaine hydroxylase activity in this patient indicates that severe anemia can lower the activity of at least one of the
II
OF CARNITINE
pm01 formed
* Activity Activity
**
was measured was measured
-
38.7 k 48.0 (6) 5.8-143
111 74.9 76.3
in liver washed mitochondrial fractions [ 121. in 105 000 X g supernatant fractions of human liver [121.
191
34.2 zt 7.8 (3) 26.2-44.8
Infant controls mean * SEM (N) range
Child control
45.6 + 15.5 (7) 28.6-77.9
43.9 62.2 21.2
LIVER
P-Hydroxy-e-N-trimethylL-lysine aldolase * *
protein
IN HUMAN
min/mg
ENZYMES
E-A’-trimethyl-lysine fl-hydroxylase *
Activity,
BIOSYNTHETIC
mean f SEM (N) range
Adult controls
Patient one Patient two Patient three
Subject
ACTIVITY
TABLE
63 900
25 700 (1)
46 800 + 36 500-66
34 500 71 000 67 800 9 200 (7) 300
y-trimethylaminobutyraldehyde dehydrogenase **
102
hydroxylase
2.6 (3) 47
f 95 (8) -488 43.3 * 41 -
369 209
157 303 243
y-butyrobetaine
**
299
Fig. 1. Comparison
of activities
of carnitine
biosynthetic
enzymes
in patients
with systemic
carnitine
defi-
ciency and in control subjects. Panel A, f-N-trimethyl-L-lysine P-hydroxylase; panel B. P-hydroxy-E-N-trimethyl-L-lysine aldolase; panel C, y-trimethylaminobutyraldehyde dehydrogenase; and panel D, ?-butyrobetaine hydroxylase. Activities are designated by circles for adults. squares for children and triangles for infants.
hepatic enzymes which subserve camitine biosynthesis. The results of this study do not exclude a defect in the conversion of lysine to e-N-trimethyl-L-lysine in systemic camitine deficiency. Neurosporu crassa can methylate free lysine [ 141 but this reaction has not been demonstrated in mammalian systems. Therefore, it is generally believed that in mammals e-N-trimethyl-L-lysine destined for camitine biosynthesis is derived from e-N-trimethyl-L-lysine residues in proteins [ 15-171. The significance of methylated amino acids in proteins is presently unclear; presumably they play a role in the structural, chemical, or functional integrity of the protein molecule. Because c-N-trimethyl-L-lysine residues occur in a wide variety of proteins [6], it is unlikely that the absence of this residue from a single protein would impair carnitine biosynthesis. Absence of e-N-trimethyl-L-lysine from all proteins in which it normally occurs would very likely lead to multiple clinical manifestations due to a variety of structurally altered proteins, rather than just carnitine deficiency. The results of this study are consistent with the conclusion that human systemic carnitine deficiency is not the result of a defect in camitine biosynthesis. We are currently investigating possible causes of this disease, including abnormal degradation, excretion or cellular transport of camitine. Acknowledgement This work was supported by a Research Center Dystrophy Association and NIH Grant NS 6277.
Grant from
the Muscular
300
References 1 Karpati, G., Carpenter, S., Engel, A.G.. Watters. G.. Allen, J., Rothman. S., Klassen. G. and Mamer. O.A. (1975) The syndrome of systemic carnitine deficiency. Neural. (Minneap.) 25, 16-24 2 Scholte. H.R.. Meijer, A.E.F.H., Van Wijngaarden, G.K. and Leaders, K.L. (1979) Familial carnitine deficiency. A fatal case and subclinical state in a sister. J. Neural. Sci. 42, 87-101 3 Glasgow, A.M.. Eng, G. and Engel. A.G. (1980) Systemic carnitine deficiency simulating recurrent Reye’s syndrome. J. Pediatr. 98. 889-891 4 Engel, A.G. (1980) Possible causes and effects of carnitine deficiency in man. In Biosynthesis, Metabolism and Functions of Carnitine (Frenkel, R.A. and McGarry, J.D.. eds.). Academic Press, New York, PP. 271-284 5 Tanphaichitr. V. and Broquist, H.P. (1973) Role of lysine and c-N-trimethyllysine in carnitinc biosynthesis. II. Studies in rat. J. Biol. Chem. 248. 2176-2181 6 Paik. W.K. and Kim, S. (1975) Protein methylation: chemical, eruymological and biological significance. Adv. Enzymol. 42, 227-286 7 Paik. W.K. and Kim, S. (1970) Solubilization and partial purification of protein methylase III from calf thymus nuclei. J. Biol. Chem. 245. 6010-6015 8 Hulse. J.D., Ellis, S.R. and Henderson, L.M. (1978) Carnitine biosynthesis. P-Hydroxylation of trimethyllysine by a a-ketoglutarate-dependent mitochondrial dioxygenase. J. Biol. Chem. 253, 16541659 9 Hochalter, J.B. and Henderson, L.M. (1976) Carnitine biosynthesis. The formation of glycine from carbons 1 and 2 of 6-N-trimethyl-L-lysine. Biochem. Biophys. Res. Commun. 70, 364-366 10 Hulse. J.D. and Henderson, L.M. (1980) Carnitine biosynthesis. Purification of 4-N-trimethylaminobutyraldehyde dehydrogenase from beef liver. J. Biol. Chem. 255, 1146-1151 11 Lindstedt. G. (1967) Hydroxylation of y-butyrobetaine to carnitine in rat liver. Biochem. 6. 12711282 12 Rebouche, C.J. and Engel. A.G. (1980) Tissue distribution of carnitine biosynthetic enzymes in man. Biochim. Biophys. Acta 630, 22-29 13 Rudman. D., Swewell, C.W. and Ansley, J.D. (1977) Deficiency of carnitine in cachectic cirrhotic patients. J. Clin. Invest. 60. 716-723 14 Rebouche, C.J. and Broquist, H.P. (1976) Carnitine biosynthesis in Neurospora CPCISSU: enzymatic conversion of lysine to e-N-trimethyllysine. J. Bacterial. 126. 1207-1214 15 Cox, R.A. and Hoppel, C.L. (1973) Biosynthesis of carnitine and 4-trimethylaminobutyrate from 6-N-trimethyllysine. Biochem. J. 136, 1083-1090 16 LaBadie. J., Dunn, W.A. and Aronson, N.N.. Jr. (1976) Hepatic synthesis of carnitine from proteinbound trimethyl-lysine. Lysosomal digestion of methyl-lysine-labeled asialo-fetuin. Biochem. J. 160, 85-95 17 Paik, W.K.. Nochumson, S. and Kim. S. (1977) Carnitine biosynthesis via protein methylation. Trends Biochem. Sci. 2. 159-161