was determined by gas-liquid chromatography of trimethylsilylated derivatives of the ... tions of free octanoate, cis-4-decenoate,and decanoate in their plasma.
CLIN. CHEM.
34/3, 548-551 (1988)
Cis-4-Decenoic Acid in Plasma: a CharacteristicMetabolitein Medium-ChainAcyI-C0A DehydrogenaseDeficiency M. Duran, L Bruinvls, D. Kettlng, J. B. C. de Klerk,
andS. K. Wadman
The profile of organic acids in plasma of patients with a deficiency of medium-chain acyl-CoA dehydrogenase (EC 1.3.99.3) was determined by gas-liquid chromatography of trimethylsilylated derivatives of the acids isolated by ethyl acetate extraction. All 13 patients had increased concentrations of free octanoate, cis-4-decenoate,and decanoate in their plasma. Cis-4-decenoate, an intermediary metabolite of Iinoleic acid, is pathognomonic of medium-chain acyl-C0A dehydrogenase deficiency. This metabolite does not accumulate in plasma after oral loading with medium-chain tnglycerides, in contrast to octanoate and decanoate. Two postmortem plasma samples from victims of infant sudden-death syndrome had detectable octanoate and decanoate, but cis4-decenoate could not be detected. The identification of cis4-decenoate in plasma may be an aid in the diagnosis of an inherited defect in oxidation of medium-chain fatty acids.
AddItIonalKeyphrases: heritable disorders dicarboxylic aciduria hypoglycemia
.
fatty acid oxidation infant sudden-deathsyndrome
Medium-chain acyl-coenzyme A dehydrogenase (MCAI)) deficiency has been recognized as a potentially lethal inborn error of metabolism in childhood.’ Affected children usually have attacks of a Reye-like syndrome after a minor illness. The main biochemical hallmark is hypoketotic hypoglycemia after prolonged glucose deprivation (1). Hence the first suspicion of such a defect arises when a negative or only slightly positive reaction for urinary acetone is found during a hypoglycemic attack. In several patients the fatal metabolic decompensation progressed so quickly that these patients were originally misclassified as victims of the infant sudden-death syndrome. A direct clue to the diagnosis is obtained by analyzing for organic acids in urine. The abnormal profile is highly specific, comprising excessive amounts of the medium-chain dicarboxylic acidsadipate, suberate, and sebacate. In addition, MCAD-deficient patients excrete considerable amounts of conjugates of medium-chain fatty acidssuch as hexanoylglycine, suberylglycine, octanoylcarrntine (2), and octanoylglucuronide (3). The lattertwo metabolites are also excreted by subjects who receive foods containing medium-chain triglycerides (MCD (4), which diminishes the value of their detection for the diagnosis of MCAD deficiency. Analysis for organic acids in plasma for the diagnosis of disorders of fatty acid beta-oxidation has received little attention so far, although we have shown that free octanoate and decanoate are always increased in plasma in MCAD deficiency (2). University Children’s Hospital ‘Het Wilhehnina Kinderziekenhuis,’ Nieuwe Gracht 137, 3512 LK Utrecht, The Netherlands. ‘Nonstandard abbreviations: MCAD, medium-chain acyl-coenzyme A dehydrogenase (EC 1.3.99.3); MCT, medium-chain triglycerides;FFA, free fatty acid(s);3-HB, 3-hydroxybutyrate. ReceivedSeptember 16, 1987; acceptedDecember 28, 1987. 548
CLINICALCHEMISTRY, Vol. 34, No. 3, 1988
Here we report that the presence of ci.s-4-decenoic acid, a metabolite of linoleic acid, in plasma is characteristic of MCAD deficiency and is not produced as a consequence of feeding with MCT-containing formulas. We also describe in detail the analysis for free fatty acids in plasma by capillary gas-liquid chromatography. This technique can be applied to the differential diagnosis of the infantsudden-death syndrome, in which a post-mortem sample of plasma may be the only material left for diagnostic studies.
Materials and Methods Patients Included in this study were eight clinically affected patients, ages six to 20 months, who were unable to oxidize medium-chain fatty acids. The metabolic defect in most of the patients was confirmed by assays in cultured fibroblasts and (or) liver tissue: either a direct assay of MCAD with the tritium release method (5) or an estimate of [‘4C]octanoate oxidation in whole fibroblasts (6). The clinical condition of the subjects under investigation varied: five patients recovered completely after an attack of presumably hypoketotic hypoglycemia, one remained decerebrate after two such attacks, and two died during hypoglycemic crises. In addition, five biochemically affected subjects were detected on screening family members. These included both adults and children who had never had any symptom related to disturbed energy homeostasis. A second group of patients included five children with an inability to oxidize hong-chain fatty acids: one child with systemic carnitinedeficiency, a child with glutaric acidemia type I with (secondary?) carnitine deficiency, and three children with an apparent inability to oxidize long-chain fatty acids, but in whom no clear diagnosis could be made. The latter four children died in early childhood. Finally, blood was sampled post mortem from eight infants who died with the syndrome, and analyzed. Various hospitalized children served as controls. Methods
The organic acids (includingketone bodies and free fatty acids) were analyzed by capillary gas-liquid chromatography of the corresponding trimethylsilyl esters, prepared as follows. Acidi1’ 0.5 mL of heparinized plasma with one drop of 2 moIJL HC1, then add 50 p.L of a 1 mg/mL aqueous solution of 3-phenylbutyric acid (internal standard). Transform the ketoacids into ethoximes by reacting them with 10 mg of ethoxyaniine HCI at room temperature for 2 h. Add 0.5 mL of a saturated NaC1 solution and extract twice with 2-mL portions of ethyl acetate, shaking vigorously for 1 miii each time. Centrifuge (2000 x g, 10 mm), pipette off the organic phase, and dry it over anhydrous Na2SO4. Remove the solvent by rotary evaporation at 35 #{176}C under reduced pressure. To trimethylsilylate the acids, add 50 L of a mixture of N,N-bis(triniethylsilyl)trifiuoroacetamide, pyridine, and trimethylchlorosilane, 5/1/0.05 (by vol), and react at 60#{176}C for 30 miii. .
OCTANOATE
DECENOATE
C 8:0
C 10:1
DECANOATE
C
10.0
#{163}
a
urn o IlL
La a AL a a aaAa
aaa
100
aL
a a6 a a a
a
a
aa a a
a
La
a A a6
6
aLL a6a
a a
a #{163}666
a La #{163}
a
LLL aaa
#{163}
LA LLL
10
a
a
a
a A LA A
#{163}
A
1Fig. 1. Concentrations offree octanoate, cis-4-decenoate, and decanoate in the plasma of 13 patients with medium-chain acyl-C0A
dehydrogenasedeficiency Nodifierentiation wasmaderegardingthe timing ofthebloodsamplingexceptin the MCT-badingtests(dataomittedfromthis graph) We analyzed 1 pL of the derivatized acids by capillary gas-liquid chromatography, using a 25 m x 0.25 mm (i.d.) column of “CPSil 19 CB,” film thickness 0.19 m (Chrompack BV, Middelburg, The Netherlands) and a 1:25 splitter. Nitrogen was the carrier gas, and detection was by flame ionization. The column temperature was initially kept isothermal at 75#{176}C for 5 miii and then programmed at a rate of 5 #{176}C/mun up to 280 #{176}C. This temperature was maintained for another 15mm. When necessary, we separated the acids by gas chromatography, using helium as the carrier gas, and identified them by electron-impact quadrupole mass spectrometry, comparing the mass spectra with those of the authentic reference compounds. The gas-chromatographic parameters were identical to those just mentioned. Ci8-4decenoic acid was a gift of Dr. N. Gregersen (Aarhus, Denmark). We quantified the various acids by calibration with the reference compounds taken through the whole procedure.
Results and DiscussIon Measurement of free fatty acids in plasma gives additional information for the differential diagnosis of hypoketotic
hypoglycemia. We have previously shown that patients with disorders in the beta-oxidation of fatty acids have increased ratios of free fatty acid (FFA)/3-hydroxybutyrate (3-HB) (7) and we could confirm this in the present group of MCADpatients. Blood samples were taken either in an attack of hypoketotic hypoglycemia or during a controlled fasting experiment. The concentrations of free fatty acids, calculated by addition of the individual concentrations of the longchain fatty acids, ranged from 158 to 3083 pmoIJL and exceeded those of 3-hydroxybutyrate in 40 of 45 samples. The ratio FFA/3-HB in healthy controls is always lower than 1, even after a prolonged fast, probably because the ketone bodies themselves act as feedback inhibitors of the mobilization of fatty acids. An apparently normal FFA/3-HB ratio in MCAD patients, such as we noticed in only five of 45 observations, may be the consequence of the timing of the blood sampling. Lipolysis is inhibited by insulin, a hormone that is released immediately after the intravenous administration of glucose. Thus the glucose causes the concentrations of free fatty acids to decline rapidly, while that of 3hydroxybutyrate remains unchanged somewhat longer, presumably reflecting the amount of fatty acids that was being oxidized at the moment of administration of glucose. The highest concentrations of 3-RB and of the free fatty acids in plasma were observed when the patients were in an acute metabolic crisis, which occurred after a period of accidental fasting for much longer than 24 h. Concentrations of free octanoate and decanoate in the plasma of patients with MCAD-deficiency varied widely (Figure 1). Nevertheless, these concentrations paralleled the extent to which beta-oxidation of fatty acids took place. This was demonstrated by controlled experiments on six fasting patients, which included serial measurements of blood glucosetogether with gas-chromatographic assays of 3-HB and medium- as well as long-chain fatty acids. Fasting was continued for a maximum of 24 h in children under five years of age. None of the patients showed clinical signs of hypoglycemia at the end of the fast; their blood glucose values did not decrease below 2.5 mmol/L. The concentrations of free octanoate in plasma were invariably higher than those of decanoate in MCAD-deficient patients (Figure 1); the concentrations of cis-4-decenoate were almost without exception between those of octanoate and decanoate. Whenever possible, we analyzed more than one blood sample from a patient. We do not consider it useful, therefore, to calculate mean concentrations, because of the wide variation in individual patients. Cis-4-decenoic acid was identified by comparison of its gas-chromatographic and mass-spectrometric behavior with that of the reference compound. Its triniethylsilyl ester was eluted just before that of decanoic acid, their retention times differing by 0.20 miii. Its mass spectrum was rather nondescript, with a weak molecular ion at mlz 242 and the base peak at niJz 117, representing the [COOTMS]-ion. Ci.s4decenoic acid can be regarded as a metabolite of linoleic acid, a normal constituent of most lipids. It is formed after four successiveturns of beta-oxidation of linoleic acid and, in addition, one isomerization step to convert cis-3,6-dodecadienoyl-CoA to trans-2,6-dodecadienoyl-CoA. Dommes and Kunau (8) have shown that cis-4-decenoyl-CoA is usually metabolized via MCAD, and not via an isomerase. Our finding of decenoate in all plasma samples of patients with MCAD-deficiency is fully in harmony with their findings. In contrast, none of the four patients with presumed disorders of the beta-oxidation of long-chain fatty acids accumulated CLINICAL CHEMISTRY, Vol. 34, No. 3, 1988 549
patient. However, as there have been reports on the potential hazard of MCT in MCAD-patients, we no longer recommend use of this test as a routine investigation. Having shown that the concentrations of cis4-decenoic acid did not react to MCT-loading, we decided to investigate the effect of fasting on plasma cis-4-decenoicacid in MCADdeficient patients. All six patients who were fasted showed an increase of its concentrations similar to that of decanoate. Clinically, fasting of moderate duration (15-20 h) entailed no problem. Because of the reportedly high incidence of deficiency of medium-chain acyl-CoA dehydrogenase and other defects of fatty acid oxidation among victims of the infant suddendeath syndrome (9, 10), we analyzed plasma samples obmined post mortem from eight such patients. Three of them had detectable concentrations of octanoate, between 18 and 26 pmolIL; in two of these we could also detect decanoate (7 and 8 molIL, respectively). However, none of the samples contained any detectable cis-4-decenoic acid, and we therefore conclude that none of these children died from a deficiency of medium-chain acyl-CoA dehydrogenase. For comparison: the two patients in this series who suddenly died from MCAD-deficiency at 17 and 19 months, respectively, had increased post-mortem concentrations of cis-4decenoate in plasma, 84 and 162 jmoI/L, respectively. Why do MCAD patients recover so slowly from a hypeketotic hypoglycemic coma? Almost immediately after glucose is given intravenously their blood glucose normalizes, yet they may remain
0
1
2
3hours
Fig. 2. Time course of concentrations of medium-chain fatty acids in plasma of a patient with MCAD deficiency after a loading test with
medium-chainttiglycerides (1 g/kg body wt.) Theaccompanytng increasein3-hydroxybutyratewasinadequate (seetext)
measurable amounts of cis-4-decenoic acid in plasma. Only trace amounts of octanoate were present in these samples. The endogenous formation of cis-4-decenoic acid was shown in a test in which one of the MCAD patients, after informed consent of the parents, was given an oral load of medium-chain triglycerides (1 g/kg body wt.; Liquigen; Scientific Hospital Supplies, Liverpool, U.K.). This formula was shown by gas chromatography/mass spectrometry not to contain unsaturated decanoic acid. After this load, the patient’s concentration of octanoate in plasma increased sharply, with a concomitant modest increase of decanoate (Figure 2). The peak concentration of octanoate in the plasma of a healthy control after a similar load was 400 1zznoIJL. There was no change in the concentration of cis4. decenoate in plasma. Additional findings included an increase in plasma hexanoate from 11 to 121 mol/L and of 7OH-octanoate from undetectable to an abnormally high value of 151 zmol/L at 3 h after the load. Ketone-body formation was inadequate, as could be concluded from the maximum 3-hydroxybutyrate concentration of only 0.41 mmol/L (age-matched control: 1.3 mmolIL). Reassuringly, the MCT-load did not have any adverse clinical effect on the 550
CLINICALCHEMISTRY, Vol. 34, No. 3, 1988
drowsy for a considerable
length
of
time. In this respect, the role of medium-chain fatty acids, especially of octanoate,has been discussed.Increased octanoate concentrations have been associated with encephalopathy (11). However, the real contribution of octanoate to the development of cerebral dysfunction has to be doubted seriously in view of the fact that somany children nowadays receive diets that contain medium-chain triglycerides, with no deleterious effect. Also, when we increased the plasma octanoate in one of our patients after a loading test with MCT, no clinical effects were observed. Octanoate is even in use as a preservative for transfusion plasma. If any of the medium-chain fatty acids plays a role in the pathogenesis of cerebral dysfunction, a more likely candidate would be cis-4decenoic acid. This substance has some structural resemblance to 4-pentenoic acid, a compound with proven hypeglycemic effects (12). A similar resemblance exists with ivalproate, a hepatotoxic metabolite of the anti-epileptic drug valproic acid (13). However, the latter two compounds both possessa terminal vinyl-group, which may be responsible for their actions and which possibly distinguishes them from ci.s-4-decenoicacid. If the cis-4-decenoate has some clinical effect, it would be tempting to treat patients in such a way as to lower their linoleic acid stores, e.g., by decreasing their intake of essential fatty acids. Thus far the importance of detecting cis-4-decenoic acid lies in the fact that it can help in the diagnosis of inherited medium-chain CoA dehydrogenase deficiency, even in serum samples obtained post mortem from victims of the infant sudden-death syndrome. References 1. Pollitt RJ. Inherited disorders of straight chain fatty acid oxidation. Arch Dis Child 1987;62:6-7. 2. Duran M, Mitchell G, de Klerk JBC, et al. Octanoic acidemia and octanoylcarnitine excretionwith dicarboxylic aciduriadue to
lefective oxidation of medium-chain fatty acids. J Pediatr L985;107:397-404. t. Duran M, Ketting D, van Vossen R et al. Octanoylglucuromde txcretionin patients with a defectiveoxidationof medium-chain htty acids.Clin Chim Acts 1985;152:253-60. I. Kuhara T, Matsuinoto I, Ohno M, Ohura T. Identification and iuantiflcation of octanoylglucuronidein the urine of childrenwho Lngestedmedium-chain triglycerides.BiomedEnviron Mass Specrom 1986;13:595-8. i. Aniendt BA, Rhead WJ. Catalytic defectof medium-chain acyloenzyme A dehydrogenasedeficiency.Lack ofboth cofactorre8pontivenessand biochemicalheterogeneityin eight patients. J Clin [nvest 1985;76:963-9. Saudubray JM, CoudeFX, Demaugre F, et al. Oxidation of fatty tcids in cultured fibroblasts: a model system for the detection and tudy of defects in oxidation.Pediatr Res 1982;16:877-81. T. Duran M, de Kl#{232}rk JBC, Wadman SK, Bruinvis L, Ketting D. [‘he differential diagnosis of dicarboxylic aciduria. J Inherit Metab .
Dis 1984;7,suppl 1:48-51.
8. DommesV, Kunau WH. Purification and properties of acyl-CoA dehydrogenase from bovine liver. Formation of 2-tnzns,4-cis-decadienoyl-CoA. J Biol Chem 1984;259:1789-97. 9. Anon. Suddeninfant death and inherited disorders of fat oxidation [Editorial]. Lancet 1986;ii:1073-4. 10. Harpey JP, Charpentier C, Coude M, Divry P, Paturneau4ouas M. Suddeninfant death syndrome and multiple acyl-coenzyme A dehydrogenasedeficiency,ethylnialonic-adipic aciduria, or systemic carnitine deficiency. J Pediatr 1987;110:881-4. 11. McCandless DW. Octanoic acid-induced coma and reticular formation energy metabolism. Brain Rae 1985;335:131-7. 12. Sheratt HSA, Osmundsen H. On the mechanism of some pharmacological actions of the hypoglycaemic toxins hypoglycin and pent-4-enoic acid: a way out of the present confusion. Biochem Pharmacol 1976;25:743. 13. Rettie AE, Rettenmeier AW, Howald WN, Baillie TA. Cytochrome P450-catalyzed formation of 4-VPA, a toxic metabolite of valproic acid. Science 1987;235:890-3.
CLINICALCHEMISTRY, Vol. 34, No. 3, 1988 551
