Synthesis of dicarboxylic acylcarnitines

Article abstract of DOI:10.1016/j.chemphyslip.2004.01.001

Syntheses of malonyl, methylmalonyl, succinyl, glutaryl, methylglutaryl, dodecanedioyl and hexadecanedioyl carnitines are described. The dicarboxylic acylcarnitines were prepared from eight equivalents of cyclic anhydride or isopropylidene ester of the di

Full text of DOI:10.1016/j.chemphyslip.2004.01.001

Chemistry and Physics of Lipids 129 (2004) 161–171
Synthesis of dicarboxylic acylcarnitines
David W. Johnson∗
Department of Chemical Pathology, Women’s and Children’s Hospital, 72 King William Road, North Adelaide, South Australia 5006, Australia
Received 19 November 2003; received in revised form 6 January 2004; accepted 7 January 2004
Abstract
Syntheses of malonyl, methylmalonyl, succinyl, glutaryl, methylglutaryl, dodecanedioyl and hexadecanedioyl carnitines are
described. The dicarboxylic acylcarnitines were prepared from eight equivalents of cyclic anhydride or isopropylidene ester
of the dicarboxylic acid and carnitine chloride in trifluoroacetic acid solution. Long chain dicarboxylic acylcarnitines were
additionally purified by partitioning between water and n-butanol. Stable isotope labeled analogs, containing 3, 6 or 9 deuterium
atoms, were also prepared. They are for use as standards in the electrospray ionization tandem mass spectrometric analysis of
dicarboxylic acylcarnitines in samples from patients with inherited disorders of fatty acid oxidation.
© 2004 Elsevier Ireland Ltd. All rights reserved.
Keywords: Dicarboxylic acid; Acylcarnitine; Glutarylcarnitine; Deuterium labeling; Tandem mass spectrometry; Newborn screening
1. Introduction
nitines and isotope labeled analogs. It results in poor
accuracy and precision in the measurement of dicar-
boxylic acylcarnitines, especially glutarylcarnitine,
which are typically found at lower concentrations
than most monocarboxylic acylcarnitines.
Dicarboxylic acylcarnitine standards would also be
of value in the differential diagnosis of dicarboxylic
aciduria in urine samples (Shimizu et al., 1994) col-
lected for organic aciduria screening.
The synthesis of succinylcarnitine (C4DC, conven-
tional nomenclature for an acyl carnitine of a four
carbon dicarboxylic acid) from succinyl chloride has
previously been shown (Bohmer and Bremer, 1968)
to afford a product with only 70% purity. The major
problem in coupling a dicarboxylic acid to carnitine is
that one carboxyl group must be protected during the
esterification to avoid a binary derivative. Subsequent
deprotection reactions result in solvolysis of the acyl-
carnitine ester. A synthetic strategy that effectively un-
masks a carboxylic acid group during the esterification
Newborn screening programs, utilizing tandem
mass spectrometry, now aid in the diagnosis of many
inherited diseases involving defects in fatty acid ox-
idation (Rashed, 2001; Chace et al., 2003). These
diseases are characterized by abnormally high levels
of acylcarnitines of fatty acids. Monocarboxylic acyl-
carnitines are measured using isotope labeled internal
standards. The standards are synthesized from acyl
chlorides and carnitine chloride (Ziegler et al., 1966;
Bohmer and Bremer, 1968). Dicarboxylic acylcar-
nitines, which are diagnostic metabolites for the in-
herited diseases summarized in Table 1, are measured
relative to the labeled monocarboxylic acylcarnitine
with the closest mass. This is a consequence of the
lack of commercially available dicarboxylic acylcar-
∗
Fax: +61-8-81617100.
E-mail address: david.johnson@adelaide.edu.au (D.W. Johnson).
0009-3084/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.chemphyslip.2004.01.001

162
D.W. Johnson / Chemistry and Physics of Lipids 129 (2004) 161–171
Table 1
Dicarboxylic acylcarnitines and the inherited diseases for which they are diagnostic metabolites
Acylcarnitine
C3DC
Inherited disease
Reference
Malonic aciduria
Malonyl CoA decarboxylase deficiency
Methylmalonic aciduria
Matalon et al. (1993)
Santer et al. (2003)
Oberholzer et al. (1967)
Baric et al. (1998)
C4brDC
C5DC
Glutaric aciduria type I
C6brDC
C12DC
C12DC
C12DC
C14DC
C16DC
HMG-CoA lyase deficiency
Reye’s syndrome
Roe et al. (1986)
Tracey et al. (1988)
Fontaine et al. (1998)
Rocchicioli et al. (1996)
Rocchicioli et al. (1996)
Rizzo et al. (2003)
Roschinger et al. (2000)
Rizzo et al. (2003)
Roschinger et al. (2000)
Carnitine palmitoyltransferase type II
Peroxisomal biogenesis defect
Peroxisomal biogenesis defect
Peroxisomal biogenesis defect
Carnitine-acylcarnitine translocase deficiency
Peroxisomal biogenesis defect
Carnitine-acylcarnitine translocase deficiency
C18DC
br indicates a branched dicarboxylic acid.
of an alcohol, is the use of an anhydride. There are no
previous reports of the use of anhydrides to prepare
carnitine esters.
spectrometer (Concord, Ontario, Canada) fitted with
an Ionspray assembly and operated in positive ion
mode. Samples for analysis were dissolved in ace-
tonitrile/water/formic acid (50:50:0.025, v/v/v) and
infused, at 10 ␮l/min, with a Harvard syringe pump
(Cambridge, MA, USA).
The synthesis of a comprehensive selection of di-
carboxylic acylcarnitines, outlined in Scheme 1, was
accomplished using this strategy. In addition, analogs
of dicarboxylic acylcarnitines that are deuterium la-
beled in either the acyl group or the carnitine were
prepared. They will be used for the quantification of
dicarboxylic acylcarnitines by electrospray ionization
tandem mass spectrometry (ESI-MS/MS).
2.3. NMR analysis
NMR spectra were obtained on a Varian Inova 600
NMR spectrometer (Palo Alto CA, USA) operated by
Mr. Philip Clements in the Adelaide University De-
partment of Chemistry.
2. Experimental
2.4. Preparation of dl-glutarylcarnitine chloride
C5DC (3)
2.1. Chemicals
(Methyl-2H3) l-carnitine chloride (enrichment
dl-Carnitine chloride (200 mg), glutaric anhy-
dride (900 mg) and trifluoroacetic acid (0.35 ml) were
heated at 75 ◦C, in a sealed tube, for 18 h. Acetone
(4 ml) was added to the cooled mixture. Diethyl ether
(5 ml) was then added dropwise over 15 min. The mix-
ture was centrifuged (4000 rpm), the supernatant dis-
carded, and the residue was washed twice with diethyl
ether. The residue was dissolved in methanol (0.3 ml),
again precipitated from acetone/diethyl ether, washed
twice with diethyl ether and dried thoroughly in a
stream of nitrogen. dl-Glutarylcarnitine chloride 3
(300 mg, 95% yield) was obtained as a colourless gum.
It resisted all attempts at crystallization. 1H NMR
(D2O, 600MHz): δ 1.88 (m, 2H, CH2CH2CH2), 2.41
2
>99% H3), (methyl-2H9)-dl-carnitine chloride (en-
richment 99.8% 2H9) and 2H6-pentanedioic acid
(enrichment 98.3% 2H6) were purchased from C/D/N
Isotopes (Pointe-Claire, Quebec, Canada).
Glutaric anhydride (97% purity) was purchased
from Lancaster (Eastgate, Lancashire, UK). All
other chemicals and solvents were purchased from
Sigma-Aldrich (Castle Hill, NSW, Australia).
2.2. Mass spectrometric analysis
Mass spectrometric analysis was performed on an
Applied Biosystems/SCIEX API365 tandem mass

D.W. Johnson / Chemistry and Physics of Lipids 129 (2004) 161–171
163
Scheme 1. Synthesis of dicarboxylic acylcarnitines from carnitine chloride and anhydrides and isopropylidene esters of dicarboxylic acids.
(m, 2H, CH2CH2CO2H), 2.50 (m, OCOCH2CH2),
2.78 (m, 2H, CHCH2CO2H), 3.16 (m, 9H, N(CH3)3),
3.76 (m, 2H, NCH2CH), 5.65 (m, H, CH2CHCH2).
Purity was determined as 96% from trimethylamine
protons. ESI mass spectrum (see Fig. 1A): m/z 276
(M) +, 217 (M − N(CH3)3)+, 199 (M − N(CH3)3,
H2O)+, 144 (M − CH2(CH2CO2H)2)+. ESI mass
spectrum of dibutyl ester: m/z 388 (M)+.
crystals, m.p. 60 ◦C, was reacted with dl-carnitine
chloride (150 mg) in the same manner described above
to afford 2H6-dl-glutarylcarnitine (175 mg, 73%
yield) as a colourless gum. 1H NMR (D2O, 600 MHz):
δ 2.78 (m, 2H, CHCH2CO2H), 3.16 (m, 9H, N(CH3)3),
1
1
3.76 (m, 2H, NCH2CH), 5.65 (m, H, CH2CHCH2).
Purity was determined as 92% from trimethylamine
protons. ESI mass spectrum (see Fig. 1B): m/z 282
(M)+, 223(M − N(CH3)3)+, 205 (M − N(CH3)3,
H2O)+, 144 (M − C2H2(C2H2CO2H)2)+. ESI mass
spectrum of dibutyl ester: m/z 394 (M)+.
2
2.5. Preparation of H6-dl-glutarylcarnitine
chloride d6C5DC
2H6-Pentanedioic acid (800 mg) and acetic an-
hydride (10 ml) were heated at 75 ◦C for 18 h. The
mixture was evaporated in a stream of nitrogen. Two
(5 ml) portions of dichloromethane were added and
evaporated to remove the last traces of acetic acid. The
2H6-glutaric anhydride, which was obtained as white
2.6. Preparation of dl-methylglutarylcarnitine
chloride C6brDC (4)
dl-Methylglutarylcarnitine chloride 4 was pre-
pared, as a colourless gum, in an identical manner
(91% yield) to that described for dl-glutarylcarnitine

164
D.W. Johnson / Chemistry and Physics of Lipids 129 (2004) 161–171
Fig. 1. ESI mass spectra of (A) C5DC and (B) 2H6-C5DC.

D.W. Johnson / Chemistry and Physics of Lipids 129 (2004) 161–171
165
chloride by substituting 3-methylglutaric anhydride
for glutaric anhydride. Purity was determined as 90%
by mass spectrometry. ESI mass spectrum: m/z 290
(M)+, 231 (M − N(CH3)3)+, 213 (M − N(CH3)3,
H2O)+, 144 (M − CH(CH3)(CH2CO2H)2)+. ESI
mass spectrum of dibutyl ester: m/z 402 (M)+.
forded dl-malonylcarnitine chloride 1 (150 mg, 88%
yield) as a yellow, viscous oil. Purity was determined
as 90% by mass spectrometry. ESI mass spectrum:
+
+
= =
m/z 248 (M) , 162 (M-O C CHCO2H) , 144 (M-
CH2(CO2H)2)+. ESI mass spectrum of dibutyl ester:
m/z 360 (M)+.
2
2.7. Preparation of H3-l-methylglutarylcarnitine
chloride d3C6BrDC
2.10. Preparation of dl-methylmalonylcarnitine
chloride C4brDC (2)
2H3-l-Methylglutarylcarnitine was prepared, in
90% purity, from 3-methylglutaric anhydride and
(methyl-2H3) l-carnitine chloride. ESI mass spec-
trum: m/z 293 (M)+, 231 (M − N(CH3)2(C2H3))+,
213 (M − N(CH3)2(C2H3), H2O)+, 147 (M −
CH(CH3)(CH2CO2H)2)+. ESI mass spectrum of
dibutyl ester: m/z 405 (M)+.
dl-Methylmalonylcarnitine chloride 2 was pre-
pared in an identical manner as described for dl-
Glutarylcarnitine chloride by substituting 2,2,5-
trimethyl-1,3-dioxane-4,6-dione for glutaric an-
hydride. It was obtained as a colourless gum
(61% yield). Purity was determined as 90% by
NMR. 1H NMR (D2O, 600 MHz): δ 1.38 (s, 3H,
CHCH3), 2.84 (m, 2H, CHCH2CO2H), 3.19 (m, 9H,
N(CH3)3), 3.81 (m, 2H, NCH2CH), 5.70 (m, 1H,
CH2CHCH2). ESI mass spectrum: m/z 262 (M)+, 203
(M − N(CH3)3)+, 185 (M − N(CH3)3, H2O)+, 144
(M − CH(CH3)(CH2CO2H)2)+. ESI mass spectrum
of dibutyl ester: m/z 374 (M)+.
2.8. Preparation of dl-succinylcarnitine chloride
C4DC (5)
A
mixture of powdered succinic anhydride
(790 mg) and dl-carnitine chloride (200 mg) was
heated in a sealed container, under a nitrogen at-
mosphere, at 130 ◦C for 16 h. The cooled mixture
was washed repeatedly with acetone. The remaining
gum was dissolved in methanol and twice precipi-
tated from acetone (4 ml) and diethyl ether (6 ml).
dl-Succinylcarnitine chloride 5 (200 mg, 66% yield)
was obtained as a pale brown, viscous oil of approx-
imately 85% purity by mass spectrometry. ESI mass
spectrum: m/z 262 (M)+, 203 (M − N(CH3)3)+, 144
(M − (CH2CO2H)2)+. ESI mass spectrum of dibutyl
ester: m/z 374 (M)+.
2.11. Preparation of dl-octan-1-ylsuccinylcarnitine
chloride C12brDC (6)
2-Octen-1-ylsuccinic anhydride (5.0 g), dl-
carnitine chloride (600 mg) and trifluoroacetic acid
(1.0 ml) were heated at 65 ◦C for 18 h. The cooled
mixture was diluted with dichloromethane (5 ml)
and evaporated in a stream of nitrogen to remove
most of the trifluoroacetic acid. The residue was dis-
solved in diethyl ether (5 ml) and extracted with water
(2 × 2 ml). The aqueous solution was washed with
diethyl ether (2 × 2 ml) and then extracted with n-
butanol (3×3 ml, saturated with water). The combined
n-butanol solution was washed with water (2 × 1 ml).
The combined aqueous layers were again extracted
with n-butanol (3 × 3 ml, saturated with water) and
washed with water (2 × 1 ml). Evaporation of the
combined n-butanol solutions afforded dl-2-octen-1-
ylsuccinylcarnitine chloride (103 mg) as a colourless
oil. The dl-2-octen-1-ylsuccinylcarnitine chloride
was dissolved in ethanol (3 ml) and palladium on
carbon catalyst (20 mg, 10% w/w) was added. The
mixture was stirred under an atmosphere of hydro-
gen for 4 h, filtered through Celite, and evaporated
2.9. Preparation of dl-malonylcarnitine chloride
C3DC (1)
2,2-Dimethyl-1,3-dioxane-4,6-dione (500 mg), dl-
carnitine chloride (100 mg) and trifluoroacetic acid
(0.2 ml) were heated at 45 ◦C for 18 h. Acetone (2 ml)
and then diethyl ether (8 ml) were added. The mix-
ture was centrifuged (4000 rpm) and the supernatant
discarded. The red oil was washed twice with di-
ethyl ether and dissolved in methanol (1 ml). The
methanol solution was passed through a column
(4 mm diameter × 40 mm) of Florisil in a Pasteur
pipette to remove some colouration. Evaporation af-

166
D.W. Johnson / Chemistry and Physics of Lipids 129 (2004) 161–171
to dryness. dl-Octan-1-ylsuccinylcarnitine chloride 6
was obtained as a colourless gum (90 mg, 7% over-
all yield). Better yields were subsequently obtained
on a smaller scale preparation with the same solvent
purification volumes. It contained approximately 2%
free carnitine. ESI mass spectrum: m/z 374 (M)+,
315 (M − N(CH3)3)+. ESI mass spectrum of dibutyl
ester: m/z 486 (M)+.
obtained as a gum despite numerous attempts at crys-
tallization. It is both hygroscopic and absorptive of sol-
vent of crystallization. Attempts to remove the small
amount of unreacted carnitine were counterproductive
since the polar solvents, like methanol and water, nec-
essary to dissolve the acylcarnitine, slowly solvolysed
it. For the preparation of a deuterium labeled analog of
C5DC, 2H6-pentanedioic acid was dehydrated to 2H6-
glutaric anhydride and reacted with carnitine chloride.
2
2.12. Preparation of dl-dodecan-1-
The ESI mass spectra of C5DC and H6-C5DC are
ylsuccinylcarnitine chloride C16brDC (7)
shown, for comparison, in Fig. 1. Unreacted carnitine
can be seen as the small ion with m/z 162.
The above reaction was repeated using 2-dodecen-
1-ylsuccinic anhydride. dl-Dodecan-1-ylsuccinylcar-
nitine chloride 7 was obtained as a colourless gum
in 7% overall yield. It contained approximately 2%
free carnitine. ESI mass spectrum: m/z 430 (M)+, 371
(M−N(CH3)3)+. ESI mass spectrum of dibutyl ester:
m/z 542 (M)+.
ESI-MS/MS analysis, with a precursor ion (85 Da)
scan, of an equimolar mixture of the butyl esters of
2H3-decanoylcarnitine and 2H6-C5DC afforded ions at
m/z 375 and 394, in the ratio 2.7:1. This confirms that if
2H3-decanoylcarnitine (the closest in mass to C5DC)
was used as a standard to measure C5DC, without
correction, the true amount of C5DC would be 2.7
times higher. This is in agreement with comparisons of
C5DC and 2H3-octanoylcarnitine (Chace et al., 2003).
Methylglutarylcarnitine (C6brDC, where br refers
to a branched dicarboxylic acid) was similarly pre-
pared from 3-methylglutaric anhydride. Since no
equivalent deuterium labeled methylpentanedioic
2
2.13. Preparation of H9-dl-octan-1-
2
ylsuccinylcarnitine d9C12brDC and H9-dl-
dodecan-1-ylsuccinylcarnitine d9C16brDC chlorides
These two deuterium labeled dicarboxylic acylcar-
nitines were prepared from (methyl-2H9) dl-carnitine
chloride and 2-octen-1-ylsuccinic and 2-dodecen-1-
ylsuccinic anhydrides. Overall yields of 42 and 35%
2
acid was available, H3-C6brDC was prepared from
2H3-carnitine chloride.
Succinylcarnitine (C4DC) required a modified ex-
perimental procedure. Because succinic anhydride is
almost insoluble in trifluoroacetic acid, a powdered
mixture of succinic anhydride and carnitine chloride
was heated at just above its melting point. An un-
avoidable minor contamination (approximately 10%)
was crotonic acid betaine (dehydrated carnitine) from
thermal decomposition of the carnitine ester. Crotonic
acid betaine was difficult to initially identify by ESI-
MS/MS analysis because it has the same mass as a
fragmentation ion (with m/z 144) of C4DC. It afforded
a more prominent ion, with m/z 200 (203 for 2H3 ana-
log), in the ESI mass spectrum of the product after
butylation.
The successful syntheses of C6brDC and C4DC
prompted a modified strategy for preparing larger
dicarboxylic acylcarnitines. Cyclic anhydrides with
a ring size >6 atoms are not commercially available.
2-Substituted succinic anhydrides with various length
alkenyl side chains, however, can be purchased. Di-
carboxylic acylcarnitines prepared from them would
2
respectively, were obtained using 100 mg of H9-dl-
carnitine chloride. They contained <2% free carnitine.
ESI mass spectra: m/z 383 (M)+ and 439 (M)+, re-
spectively. ESI mass spectra of dibutyl esters: m/z 495
(M)+ and 551 (M)+, respectively.
3. Results and discussion
Eight equivalents of glutaric anhydride were re-
acted with carnitine chloride, in trifluoroacetic acid
at 65 ◦C. Thin layer chromatography on silica with
chloroform/methanol/concentrated ammonia solution
(25:15:4, v/v/v) and staining with iodine revealed only
the C5DC (rf 0.4) and a small amount of carnitine
(rf 0.05). An ESI mass spectrum showed C5DC and
1
carnitine in the ratio 34:1. H NMR analysis showed
the C5DC to be 92% pure (from the ratio of trimethy-
lamine proton peaks). With a reaction temperature of
75 ◦C the purity improved to 96%. C5DC was always

D.W. Johnson / Chemistry and Physics of Lipids 129 (2004) 161–171
167
Fig. 2. ESI-MS/MS analysis by a product ion scan of the dibutyl ester of (A) C16DC and (B) C16brDC.

168
D.W. Johnson / Chemistry and Physics of Lipids 129 (2004) 161–171
Fig. 3. ESI-MS/MS analysis by a product ion scan of the dibutyl ester of (A) C4DC and (B) C4brDC.

D.W. Johnson / Chemistry and Physics of Lipids 129 (2004) 161–171
169
Fig. 4. ESI-MS/MS analysis by a precursor ion scan (85 Da) of an equimolar mixture of butyl esters of synthesised dicarboxylic acylcarnitines
and deuterium labeled analogs.
be isomeric with those formed biologically (from
straight chain dicarboxylic acids). Dicarboxylic C12
(dodecanedioyl), C16 (hexadecanedioyl) and C18
(octadecanedioyl) carnitine esters were prepared by
this strategy. Eight equivalents of 2-substituted C8:1,
C12:1 and C14:1 succinic anhydrides esterified ap-
proximately 60%, 30% and <10%, respectively of
the carnitine at 65 ◦C. Increasing the temperature
above 75 ◦C improved the esterification but afforded
crotonic acid betaine (dehydrated carnitine) from
thermolysis of the carnitine ester. Consequently, the
partially esterified mixture was purified by extraction
from aqueous solution into n-butanol, as used in the
extraction of acylcarnitines from urine (Morrow and
Rose, 1992). The double bond was removed by cat-
alytic hydrogenation. The ESI mass spectrum showed
that the resultant dicarboxylic acylcarnitines con-
tained less than 2% carnitine. Insufficient C18brDC
was obtained for it to be completely characterised.
2H9-labeled analogs of C12brDC and C16brDC were
dioic acid, via its acid chloride, in the manner used
for monocarboxylic acylcarnitines. Although this was
impure a clean ESI-MS/MS product ion spectrum
could be generated from the molecular cation of its
butyl ester. Comparison of the product ion spectrum
with that of the butyl ester of C16brDC is shown
in Fig. 2. The C16brDC showed fewer higher mass
product ions relative to the product ion with 85 Da.
This affords an estimated 10% greater response for
ESI-MS/MS quantification of C16brDC using an
85 Da precursor ion scan.
Preparation of malonylcarnitine (C3DC) involved
another variation of the strategy. Eight equivalents of
Meldrum’s acid (the isopropylidene ester of malonic
acid) were used, instead of an anhydride, in the reac-
tion with carnitine chloride. Meldrum’s acid is a re-
active substrate and rapidly turns a deep red colour in
trifluoroacetic acid. The C3DC, even after chromatog-
raphy from Florisil, is an oily substance that retains
a yellow colour. ESI-MS analysis reveals a purity of
approximately 90%.
2
prepared from H9-carnitine chloride.
A small amount of n-hexadecanedioylcarnitine
(C16DC) chloride was prepared from hexadecane-
The preparation of methylmalonylcarnitine (C4br
DC) required a higher temperature (75 ◦C) but af-

170
D.W. Johnson / Chemistry and Physics of Lipids 129 (2004) 161–171
Reaction temperatures above 75 ◦C, especially when
trifluoroacetic acid is used as solvent, result in ther-
molysis of the carnitine ester. The solubility of larger
dicarboxylic acylcarnitines in n-butanol, allowing their
separation from unreacted carnitine in water, compen-
sates for their poorer conversion. In all cases the di-
carboxylic acylcarnitines are obtained as viscous oils
or foamy solids after exhaustive vacuum drying as ob-
served with acetoacetylcarnitine (Bohmer and Bremer,
1968). They are hygroscopic and should be stored in
cold (−20 ◦C), dry conditions to avoid solvolysis of
the carnitine ester. Additionally, in methanol solution,
the carboxylic acid groups are slowly methylated.
Stable isotope labeled analogs of dicarboxylic acyl-
carnitines, suitable for their quantification by ESI-
MS/MS, can be prepared containing isotope labels in
either the dicarboxylic acid or the carnitine. The use of
a stable isotope labeled standard of an acylcarnitine of
a branched dicarboxylic acid, such as C12brDC and
C16brDC, results in a small (typically 10%) under-
estimation of the acylcarnitine of its isomeric straight
chain carboxylic acid. This was concluded from an
examination of the product ion spectra of the butyl
esters of a number of isomeric acylcarnitines (both
mono- and dicarboxylic). The product ion spectra of
branched acylcarnitines show greater fragmentation to
the product ion with 85 Da, relative to higher mass
product ions, than straight chain acylcarnitines.
forded a colourless product. The 1H NMR indicated a
minor contaminant (6%) with trimethylamine protons
in addition to unreacted carnitine (4%). The ESI mass
spectrum of C4brDC after butylation showed no unac-
counted for ions. The ESI mass spectrum of underiva-
tized C4brDC, however, contained an ion at m/z 218,
which corresponds to loss of CO2 from C4brDC. The
identity of this contaminant has not been established.
The ESI-MS/MS product ion spectra of the molecular
cations of the butyl esters of C4DC and C4brDC are
shown in Fig. 3. The two isomers can be differentiated
by the relative intensities of ions at m/z 119 and 185.
This observation may be of potential use in differenti-
ating a patient with methylmalonic aciduria or methyl-
CoA-decarboxylase deficiency from one with gener-
alized dicarboxylic aciduria. In principle, it should be
possible to separately measure C4DC and C4brDC in
a mixture using the strategy developed for the butyl
esters of methylmalonic and succinic acids (Kushnir
et al., 2001). Likewise, differences in the product ion
spectra of C6DC and C6brDC could also be exploited.
An equimolar mixture of C3DC, C4DC, C5DC,
C6brDC, C12brDC and C16brDC and a deuterium
labeled analog of each was butylated. The butyla-
tion with 3N hydrogen chloride in n-butanol was per-
formed at 75 ◦C (rather than the normal 65 ◦C), for
15 min, to ensure complete esterification of the larger
dicarboxylic acylcarnitines. The mixture was analysed
by ESI-MS/MS with a precursor ion (85 Da) scan.
Fig. 4 graphically illustrates the range of dicarboxylic
acylcarnitines synthesized. The increase in ion inten-
sity with size of the dicarboxylic acylcarnitine reflects
the greater proportion of product ion with 85 Da rela-
tive to total product ions (cf. Figs. 2 and 3).
It is expected that these synthesized isotope labeled
dicarboxylic acylcarnitines will provide improved ac-
curacy and precision in the measurement of the im-
portant metabolites C3DC and C5DC for tandem MS
screening programs. The larger dicarboxylic acylcar-
nitines will be used as qualitative and quantitative tools
for the differentiation of fatty acid oxidation disorders.
References
4. Conclusions
Baric, I., Zschocke, J., Christensen, E., Duran, M., Goodman, S.I.,
Leonard, J.V., Muller, E., Morton, D.H., Superti-Purga, A.,
Hoffmann, G.F., 1998. Diagnosis and management of glutaric
aciduria type I. J. Inher. Metab. Dis. 21, 326–340.
Bohmer, T., Bremer, J., 1968. Propionylcarnitine physiological
variations in vivo. Biochim. Biophys. Acta 152, 559–567.
Chace, D.H., Kalas, T.A., Naylor, E.W., 2003. Use of tandem
mass spectrometry for multianalyte screening of dried blood
specimens from newborns. Clin. Chem. 49, 1797–1817.
Fontaine, M., Briand, G., Largilliere, C., Degand, P., Divry, P.,
Vianey-Saban, C., Mousson, B., Vamecq, J., 1998. Metabolic
Dicarboxylic acylcarnitines (chloride salts) can be
prepared with a purity of 85–95% by a condensation
reaction between carnitine chloride and cyclic anhy-
drides or isopropylidene esters of dicarboxylic acids.
These cyclic compounds are less reactive than acid
chlorides and higher reaction temperatures are neces-
sary. Consequently there is a practical limitation to the
preparation of dicarboxylic acylcarnitines containing
a dicarboxylic acid with more than 16 carbon atoms.

D.W. Johnson / Chemistry and Physics of Lipids 129 (2004) 161–171
171
studies in a patient with severe carnitine palmitoyltransferase
type II deficiency. Clin. Chim. Acta. 273, 161–170.
Kushnir, M.M., Komaromy-Hiller, G., Shushan, B., Urry, F.M.,
Roberts, W.L., 2001. Analysis of dicarboxylic acids by tandem
mass spectrometry. High-throughput quantitative measurement
of methylmalonic acid in serum, plasma, and urine. Clin. Chem.
47, 1993–2002.
Matalon, R., Michaels, K., Kaul, R., Whitman, V., Rodriguez-
Novo, J., Goodman, S., Thornburn, D., 1993. Malonic aciduria
and cardiomyopathy. J. Inher. Metab. Dis. 16, 571–573.
Morrow, R.J., Rose, M.E., 1992. Isolation of acylcarnitines from
urine: a comparison of methods and application to long-chain
acyl-CoA dehydrogenase deficiency. Clin. Chim. Acta 211,
73–81.
Rocchicioli, F., Aubourg, P., Bougneres, P.F., 1996. Medium- and
long-chain dicarboxylic aciduria in patients with Zellweger
syndrome and neonatal adrenoleukodystrophy. Pediatr. Res. 20,
62–66.
Roe, C.R., Millington, D.S., Maltby, D.A., 1986. Identification
of 3-methylglutarylcarnitine. A new diagnostic metabolite of
3-hydroxy-3-methylglutaryl-coenzyme A lyase deficiency. J.
Clin. Invest. 77, 1391–1394.
Roschinger, W., Muntau, A.C., Duran, M., Dorland, L., Ijist, L.,
Wanders, R.J., Roscher, A.A., 2000. Carnitine-acylcarnitine
translocase deficiency: metabolic consequences of an impaired
mitochondrial carnitine cycle. Clin. Chim. Acta 298, 55–68.
Santer, R., Fingerhut, R., Lassker, U., Wightman, P.J., Fitzpatrick,
D.R., Olgemuller, B., Roscher, A.A., 2003. Tandem mass
spectrometric determination of malonylcarnitine: diagnosis and
neonatal screening of malonyl-CoA decarboxylase deficiency.
Clin. Chem. 49, 660–662.
Oberholzer, V.G., Levin, B., Burgess, E.A., Young, W.F., 1967.
Methylmalonic aciduria. An inborn error of metabolism leading
to chronic metabolic acidosis. Arch. Dis. Child. 42, 492–
504.
Shimizu, N., Yamaguchi, S., Orii, T., 1994.
A study of
Rashed, M.S., 2001. Clinical applications of tandem mass
spectrometry: ten years of diagnosis and screening for inherited
metabolic diseases. J. Chromatogr. B 758, 27–48.
Rizzo, C., Boenzi, S., Wanders, R.J.A., Duran, M., Caruso, U.,
Dionisi-Vici, C., 2003. Characteristic acylcarnitine profiles in
inherited defects of peroxisome biogenesis: a novel tool for
screening diagnosis using tandem mass spectrometry. Pediatr.
Res. 53, 1013–1018.
urinary metabolites in patients with dicarboxylic aciduria for
differential diagnosis. Acta Pediatr. Jpn. 36, 139–145.
Tracey, B.M., Cheng, K.N., Rosankiewicz, J., Stacey, T.E.,
Chalmers, R.A., 1988. Urinary C6–C12 dicarboxylic
acylcarnitines in Reye’s syndrome. Clin. Chim. Acta. 175, 79–
87.
Ziegler, H.J., Bruckner, P., Binon, F., 1966. O-Acylation of dl-
carnitine chloride. J. Org. Chem. 32, 3989–3991.