Printer Friendly

Simultaneous analysis of plasma free fatty acids and their 3-hydroxy analogs in fatty acid [beta]-oxidation disorders.

Mitochondrial [beta]-oxidation of fatty acids is a fundamental energy-producing pathway, especially during periods of fasting or when energy demands are increased (e.g., fever and vigorous exercise) [1]. In this regard, fatty acids as a primary fuel source account for >80% of the total body energy needs in normal infants and children after 12-24 h of fasting [2].

During the late stages of fasting, fatty acids are mobilized from adipose tissue stores. Typical endogenous fatty acids are long-chain compounds containing 16-18 carbons, such as palmitate, stearate, and oleate [3]. Long-chain fatty acids are converted to CoA esters in the outer mitochondrial membrane by an acyl-CoA thioesterase. However, CoA esters cannot readily transverse the inner mitochondrial membrane; instead, they are shuttled into the mitochondrial matrix by a carnitine-dependent mechanism that involves the sequential action of carnitine palmitoyl transferase I and II (CPTI and CPTII)(4), together with a carnitine-acylcamitine translocase. Within the mitochondrial matrix, acyl-CoA esters can now undergo the [beta]-oxidation cycle.

In recent years, a growing number of genetically distinct inborn errors of mitochondrial fatty acid [beta]-oxidation have been described [4]. Nearly all of these defects present in early infancy with acute life-threatening episodes of hypoketotic hypoglycemia and coma induced by fasting [3]. However, recognition of these defects is often difficult, in part because the flux in [beta]-oxidation is negligible under nonfasting conditions. Thus, defects in this pathway may be clinically silent until metabolic decompensation is induced by prolonged fasting or intercurrent infections.

The primary diagnosis of mitochondrial fatty acid [beta]-oxidation defects requires a high level of suspicion in the appropriate clinical settings and a selective screening. Analysis of urine organic acids [5], blood or plasma acylcarnitines [6-8], in vitro screening of overall [beta]-oxidation [9-11], in vitro probing of mitochondrial [beta]-oxidation [12, 13], in vitro measurement of individual [beta]-oxidation enzymes [14, 15], and DNA testing [16, 17] are some of the diagnostic tools available, and each has its own discriminating power. The analysis of plasma free fatty acids (FFA) and their 3-hydroxyfatty acid (3OHFA) analogs is an alternative approach for the diagnosis of fatty acid oxidation disorders [18, 19]. However, simultaneous analysis of 3OHFAs and FFAs by gas chromatography is a cumbersome task. The use of a gas chromatography-mass spectrometry (GC-MS) procedure in the selected ion monitoring mode at m/z 233, a common ion of all bis-trimethylsilylated 3OHFAs, has been proposed by Hagenfeldt et al. [19]. However, the coelution of some 3OHFAs with commonly found FFAs [18], which are in concentrations 3-100 times higher, can interfere with the fragmentation of target compounds and thus create serious problems for an accurate quantitation.

We present a new derivatization procedure for the concomitant analysis of FFAs and 3OHFAs, which was accomplished by the simultaneous trifluoroacetylation (TFA) of hydroxyl groups and tert-butyldimethylsilylation (tBDMS) of the carboxyl groups. As such, simultaneous analysis of both FFAs and 3OHFAs in plasma has become possible, using a very sensitive gas chromatographic-mass fragmentography approach.

Materials and Methods


All of the FFAs, including the internal standards non-anonic acid ([C.sub.9)] and heptadecanoic acid ([C.sub.17)], were purchased from Fluka. 3OHFAs were previously synthesized in our laboratory as described by Dorland et al. [18]. 3-Hydroxyoctadecanoic acid 3OH[C.sub.18] was synthesized following the same procedure. The internal standard 2-hydroxyoctanoic acid (2OH[C.sub.8)] was purchased from Fluka. Ethyl acetate, methanol, and chloroform were analytical grade and were obtained from Rathburn. The derivatization reagents N-methyl-N-(tert-butyldimethylsilyl)trifluoroacetamide (MTBSTFA), methyl-bis(trifluoroacetamide) (MBTFA), and pyridine were purchased from Pierce.


Standard solutions from the silver salts of 3-hydroxyhexanoic acid (3OH[C.sub.7)] and 3-hydroxyoctanoic acid (3OH[C.sub.8)] were prepared by dissolving 12.5 [micro]mol in 50 mL of acidified (drops of 1 mol/L HN[O.sub.3] added until pH 4-5) water. The other standard solutions for 3OHFAs ([C.sub.10] to [C.sub.18)] were prepared by dissolving 12.5 [micro]mol in 50 mL of methanol. For medium-chain fatty acids ([C.sub.6] to [C.sub.12)], standard solutions were prepared by dissolving 12.5 [micro]mol in 50 mL of chloroform; for long-chain fatty acids ([C.sub.14] to [C.sub.18)], 25 [micro]mol was dissolved in 50 mL of chloroform.


Patients. We studied 20 plasma samples from 15 patients with various mitochondrial fatty acid [beta]-oxidation defects. Seven plasma samples from five children with long-chain 3-hydroxyacyl-CoA dehydrogenase (LCHAD) deficiency were analyzed. Diagnosis was based on the measurement of LCHAD activity in cultured fibroblasts or leukocytes [14]. Plasma samples were collected either at presentation or in the asymptomatic period. One child with a mitochondrial trifunctional protein (MTP) defect was studied. Diagnosis was proven by enzyme assay. Two children had very-long-chain acyl-CoA dehydrogenase (vLCAD) deficiency that had been proven by measurement of enzyme activity in fibroblasts (by C. Vianey-Saban, Hopital Debrousse, Lyon) [15]. Three plasma samples were analyzed; among these, one was collected postmortem. One child had medium-chain acyl-CoA dehydrogenase (MCAD) deficiency; he was homozygous for the common A985G mutation [16]. Four patients with multiple acyl-CoA dehydrogenase deficiency (MADD) were studied. One patient had the neonatal form, whereas the other three presented with the mild form. Diagnosis was proven in two patients by measurement of electron transfer flavoprotein in fibroblasts (by C. Vianey-Saban). Finally, two patients had a defect in the so-called carnitine cycle, one at the step involving CPTI [20], and the other one at the step involving CPTII [21].

Normal controls. We studied plasma FFA and 3OHFA from two control groups. Group 1 was composed of 10 selected pediatric patients whose ages ranged from a few days to 9 years. Group 2 comprised 9 adults whose ages ranged between 19 and 50 years. In both groups, a fatty acid [beta]-oxidation defect was excluded by our usual protocol of plasma and urinary organic acid analysis, as well as by plasma acylcarnitine profiling [22]. None of the controls received a special diet. Plasma samples were collected after an overnight fast.


To 500 [micro]L of heparinized plasma or serum, 6.2 nmol of 2OH[C.sub.8], 12.5 nmol of [C.sub.9] and 25 nmol of [C.sub.17] were added as internal standards. After acidification to pH 1-2 with 2 mol/L HCl, 0.5 mL of a saturated sodium chloride solution was added. The samples were subsequently extracted twice with 2 mL of ethyl acetate by vigorously shaking for 1 min. After centrifugation (15 000 g for 5 min), the organic layer was collected, and anhydrous sodium sulfate was used to dry the combined organic layers. The solvent was removed by rotary evaporation at room temperature under reduced pressure.


Simultaneous tBDMS and TFA were carried out by the addition of 50 [micro]L of a mixture of MTBSTFA, MBTFA, and pyridine (40:50:10 by vol). This mixture was allowed to react at 60[degrees]C for 60 min, and 1 [micro]L was used for the GC-MS analysis.


GC-MS analyses were carried out with an Automass series I (ATIUnicam) quadrupole mass spectrometer interfaced with a 610 ATIUnicam GC. Samples were introduced via a split/splitless injector (10:1) at 300[degrees]C. GC separation was achieved on a WCOT fused silica capillary column [25 m x 0.25 mm (i.d.)] coated with CPSil 19 CB, film thickness 0.25 [micro]m (Chrompack BV). The initial oven temperature was kept at 90[degrees]C for 1 min and then programmed to rise 5[degrees]C/min to 280[degrees]C. The final temperature was maintained for 20 min. Helium was used as carrier gas at a pressure of 60 kPa. The column was inserted directly into the ion source with an interface temperature of 280[degrees]C. For routine measurements, we operated the instrument in the electron impact mode, with an electron energy of 70 eV and a source temperature of 250[degrees]C. Detection of target compounds was performed by multiple selected ion recording of the fragment ions at m/z (M - 57)[sup.+] for the FFAs and at m/z (M - 117)[sup.+] for the 3OHFAs, with a dwell time of 50 ms.


Standard curves were set up over a concentration range of 0.5-50 [micro]mol/L for 3OHFA and medium-chain fatty acids in water. For the long-chain fatty acids, the concentrations ranged from 1 to 100 [micro]mol/L. After evaporation of the methanolic and chloroformic standard solutions, the residue was dissolved by the addition of the aqueous standard mixture, and the volume was brought to 500 [micro]L with water. After the addition of internal standards, each sample was extracted and derivatized in the same manner as the plasma samples. The peak-area ratios [standard (for individual FFA or 3OHFA) over internal standard of M--57 ions for FFAs and M--117 ions for 3OHFAs] were calculated and submitted to a linear regression analysis that was subsequently used to calculate the concentrations of plasma FFAs and 3OHFAs. Calculations were based on the assumption of an equal detector response for the M--57 (FFAs) or M--117 (3OHFAs) for both saturated and unsaturated compounds; therefore, the quantitation of the latter was performed on the basis of the calibration curve for the corresponding saturated compound.


Extraction recoveries of 14 FFAs and 3OHFAs from plasma were determined by comparison of the peak-area ratios (standard /internal standard) of a plasma pool with the ones obtained for the standard solution. Different concentrations of FFAs and 3OHFAs were added to four 500-[micro]L aliquots of a plasma pool. Long-chain fatty acids were added to a plasma pool at a concentration of 25 [micro]mol/L, and medium-chain fatty acids and 3OHFAs were added at 12.5 [micro]mol/L. Samples were extracted and analyzed by GC-MS as described above.

Intraassay reproducibility, or within-day variability, was assessed by quantitation of FFAs and 3OHFAs in the same plasma samples prepared for the recovery studies.

Interassay reproducibility, or day-to-day variability, was assessed by quantifying FFAs and 3OHFAs in the same patient sample extracted and analyzed on three different days.


Data were compared by using the Student's Mest procedure [23], and calculations were made by using Quattro Pro for Windows (Ver. 5.0).


Only a single derivative was observed for each fatty acid or 3OHFA when a mixture of MTBSTFA and MBTFA was allowed to react. Despite the great reactivity of MTBSTFA, no double tBDMS derivatives were formed because MBTFA reacts almost instantaneously with the hydroxyl groups [24, 25].

The summed ion intensities obtained by selected ion recording from a standard solution of FFAs and 3OHFAs with chain lengths from [C.sub.6] to [C.sub.18] are shown in Fig. 1. This derivatization procedure permitted baseline separation between all of the FFAs and 3OHFAs analyzed. Moreover, a single chromatographic peak with almost no tailing was obtained for each compound, with the exception of 3OH[C.sub.18], the derivatization of which gave rise to the formation of octadecenoic acid ([C.sub.18:1]) through the loss of a water molecule. The amount of [C.sub.18:1] formed was estimated to be 20% of the 3OH[C.sub.18]. The percentage of the unsaturated compound formed was found to be independent of the concentration of 3OH[C.sub.18].


The summed ion intensities obtained by selected ion recording from a pool plasma and a patient with MCAD deficiency are depicted in Fig. 2.

The mass spectra of the tBDMS esters of FFAs are characterized by very intense M-57 ions corresponding to the loss of a tert-butyl group (M-[C.sub.4][H.sub.9])[sup.+] [26-28]. Fig. 3 shows the mass spectra of the mixed tBDMS esters and trifluoracetyl ethers of 3OHFAs from [C.sub.6] to [C.sub.18], which are characterized by very intense M-117 ions corresponding to the loss of trifluoroacetic acid accompanied by loss of the tert-butyl group from tBDMS esters. Thus, although saturated FFAs gave rise to fragment ions with the same mass as the corresponding unsaturated FFAs, different retention times allowed us to readily quantify both compounds.



The derivatizations of the FFAs and 3OHFAs were quantitative. All of the calibration curves generated from both FFAs and 3OHFAs were linear over the range of concentrations used, and correlation coefficients (r) were 0.998.

These derivatives proved to be stable for at least 12 months when stored at -20[degrees]C. Reliability of the overall analytical procedure was assessed by the recovery studies of standards from a plasma pool as well as by the intraassay and interassay reproducibility (Table 1). Therefore, sensitivity and reliability of the method for the quantitation of plasma FFAs and 3OHFAs seemed suitable for the determination of reference values in biological samples.

Normal values for all of the plasma FFAs and 3OHFAs were determined in two distinct groups of adults and children composed of selected subjects without fatty acid [beta]-oxidation defect. Using the Student's t-test, we did not find any significant difference between the adult and pediatric population (P <0.001) for any of the FFAs or 3OHFAs studied. Hence, data from both adults and children were combined to establish the reference values (Tables 2 and 3).

The diagnostic suitability of our method was readily demonstrated by the concentrations of FFAs and 3OHFAs from well-characterized patients with the most common fatty acid [beta]-oxidation defects.

The plasma 3OHFA profiles of LCHAD- and MTP-deficient patients were characterized by a clear increase of all 3OHFAs, especially the long-chain acids. As expected, the increase was more pronounced for the untreated patients (LCHAD1, LCHAD2, LCHAD-3 sample 1, and MTP; Tables 2 and 3). On the other hand, the patients whose metabolic defect did not give rise to the accumulation of 3OHFAs (MCAD, vLCAD, MADD, CPTI, and CPTII; Tables 2 and 3) presented long-chain 3OHFA concentrations within the normal range, although medium-chain 3OHFA concentrations were frequently increased during crisis.

The FFA profiles of the LCHAD- and MTP-deficient patients showed a generalized but nonspecific increase of all long-chain fatty acids. However, plasma FFA profiles are of particular importance in the diagnosis of MCAD- and vLCAD-deficiencies, as well as MADD. The MCAD-deficient patient presented a FFA pattern characterized by a strong increase of mainly [C.sub.8], [C.sub.10:1], and [C.sub.10] fatty acids. In the vLCAD-deficient patients, the plasma FFA profile was characterized by the increase of long-chain fatty acids, mainly the unusual [C.sub.14:1], [C.sub.14:2], and [C.sub.16:2]. The MADD patients, both the neonatal and the mild forms, showed a variable increase of the pathognomonic metabolites found for either MCAD- or vLCAD-deficient patients. The neonatal form presented a much more severe profile, characterized by a strong increase of the whole range of fatty acids from [C.sub.6] to [C.sub.18].

The patients with a defect in the carnitine transport system (CPTI and CPTII) presented a very similar plasma FFA pattern, characterized by an increase of long-chain fatty acids, mainly the [C.sub.16] and [C.sub.18], species, whereas the medium-chain fatty acids were within the normal range.


Several methods have been reported for the individual analysis of FFAs and 3OHFAs; however, the concomitant analysis of both analytes still has some problems of coelution [19, 29], low derivatization yields [26], or instability of the derivatives [19, 30], which have prevented their widespread use. Our main goal was to develop a GC-MS procedure combining simplicity and sensitivity with baseline separation between all of the target compounds from [C.sub.6] to [C.sub.18]. Both FFAs and 3OHFAs could be quantitatively extracted with ethyl acetate. The electron impact GC-MS analysis was based on a novel derivatization procedure that simultaneously converted the carboxyl groups to tBDMS esters and the hydroxyl groups to TFA ethers. This derivatization for both functional groups was crucial for the GC separation of all target compounds. When both groups were modified with the same derivatization reagents, there was always coelution of some 3OHFAs with the FFAs (data not shown). The simultaneous derivatization of both functional groups with different reagents was possible because MTBSTFA has a much higher activity toward the carboxylic groups than does MBTFA, whereas the latter has great affinity toward the hydroxyl groups. Moreover, the formation of the double tBDMS derivatives of 3OHFAs was prevented by the immediate reaction of the carboxyl group with MTBSTFA, which resulted in steric hindrance of the hydroxyl group in the [beta]-position [31]. Thus, after careful study of the optimum reaction parameters (solvent, concentrations, temperature, and reaction time), we could obtain single peaks with baseline separation for all compounds. A unique exception was found for 3OH[C.sub.18], whose derivatization gave rise to the formation of a second peak that eluted 10 s later and corresponded to an unsaturated compound formed by the release of trifluoroacetic acid from the derivatized product. This phenomenon could not be avoided, even when the derivatization reaction was performed at room temperature. The possible formation of this byproduct in the heated injection port of the GC was ruled out by a cold on-column injection, which could not overcome the problem. As the formation of the unsaturated compound equalled a constant fraction of the parent 3OH[C.sub.18], it constituted no problem for quantitation.

Our results concerning linearity, extraction recoveries, and reproducibility of the overall analytical procedure were fully acceptable for a multicomponent analysis. The reliability of this quantitative plasma FFA- and OHFA-profiling system for the diagnosis and monitoring of some fatty acid [beta]-oxidation defects was illustrated by the data obtained for controls and patients with different defects of this pathway.

As expected, the chain-length distribution of plasma FFAs in controls was characterized by a preponderance of long-chain fatty acids with 16 and 18 carbon atoms, whereas the concentrations of the shorter chain fatty acids ([C.sub.6] to [C.sub.14)] were 10-100 times lower. This profile mainly reflects the catabolism of fat stores that are predominantly composed of [C.sub.16] and [C.sub.18] fatty acids, both saturated and unsaturated [3, 32]. During periods of fasting, these fatty acids are released into the bloodstream and rapidly metabolized through mitochondrial [beta]-oxidation. This explains the low concentration of medium-chain fatty acids in plasma compared with their parent long-chain equivalents. The 3OHFAs were detected in normal plasma in very low concentrations, always <1 [micro]mol/L, which is in agreement with the values reported by Hagenfeldt et al. [19, 33]. Their chain-length distribution appears to be opposite the one of their corresponding FFAs. Short-chain 3OHFAs were preponderant, and the long-chain concentrations were very low.

A characteristic profile could be achieved for all the intramitochondrial deficiencies studied. In both LCHAD- and MTP-deficient patients, the plasma 3OHFA profile was characterized by an increase of all 3OHFAs with a chain length >10 carbons. Their concentrations varied with the metabolic condition of the patient and were clearly exacerbated during crisis. However, even when the patients were in remission or on a high carbohydrate diet, the concentrations of long-chain 3OHFAs were still clearly above the normal range. In contrast to the values reported by Hagenfeldt et al. [33], we found a concentration of 3-hydroxytetradecenoic acid lower than the one of 3-hydroxytetradecanoic acid, which is in accordance with the concentration of tetradecanoic acid ([C.sub.14]) and tetradecenoic acid ([C.sub.14:1]), their immediate precursors in plasma. We conclude that plasma 3OHFAs can play an important role in the diagnosis of LCHAD and MTP deficiencies in patients, especially during asymptomatic periods. In our experience, confirming that reported by others [33], plasma 3OHFAs are always increased in these patients. In contrast to patients with LCHAD or MTP deficiencies, for MCAD-deficient, vLCAD-deficient, and MADD patients, the plasma FFA profile was the most relevant one. The MCAD-deficient patient accumulated mainly octanoic ([C.sub.8]), decenoic ([C.sub.10:1]), and decanoic ([C.sub.10]) FFAs (Table 3). Octanoate was the predominant acid, which is in accordance with the data reported by Duran et al. [34] and Onkenhout et al. [35]. Decenoic acid, in controls invariably <0.6 [micro]mol/L, reaches concentrations several orders of magnitude higher in MCAD-deficient patients, rendering its quantitation an important biochemical tool for the diagnosis and monitoring of MCAD deficiency. Free octanoate and decanoate, although informative, are less specific because other conditions, such as medium-chain triglyceride supplementation, lead to increased concentrations of these fatty acids [36].

In vLCAD-deficient patients, the unusual [C.sub.14:1], [C.sub.14:2], and [C.sub.16:2] acids were of particular importance in terms of diagnosis. These fatty acids presumably correspond to [C.sub.14:1[omega]9], [C.sub.14:2[omega]6], and [C.sub.16:2[omega]6], the first being formed from oleate and the latter two from linoleate, as proposed by Onkenhout et al. [35].

MADD patients showed the whole range of unusual fatty acids observed in both MCAD and vLCAD deficiencies, reflecting the generalized defect of all the mitochondrial acyl-CoA dehydrogenases. In the neonatal form, there was a strong increase of all these metabolites, whereas in the mild forms of MADD, the increase of [C.sub.10:1] and [C.sub.12:1] was pathognomonic, and the increase of [C.sub.14:2]. [C.sub.14:1], and [C.sub.16:2] was not so evident. Onkenhout et al. [35] reported the presence of these metabolites, mainly in the esterified form, in mild cases of MADD, which may explain our slightly different results. However, when a fasting test was performed (MADD-m3; Table 3), a generalized increase of all the unsaturated fatty acids was observed.

CPTI and CPTII, shown here as negative controls for intramitochondrial defects of fatty acid [beta]-oxidation, present a FFA profile clearly distinct from the other defects. Although the increase of the concentrations of long-chain fatty acids was consistent with a defect in this pathway, none of the pathognomonic metabolites found in the above-mentioned defects could be detected in any of the patients with a defective carnitine cycle.

It was also interesting to note that medium-chain 3OHFAs, mainly 3OH[C.sub.6] and 3OH[C.sub.18], were consistently increased during metabolic derangement, independent of the site of the block. This suggests an origin of these substances that is not related to a genetic disturbance of mitochondrial fatty acid [beta]-oxidation. They may even be considered nonspecific markers of metabolic decompensation. A possible accumulation of these metabolites could also be expected in situations of increased [beta]-oxidation flux, such as during ketotic states (fasting, vomiting, or diabetic ketoacidosis). In this respect, more studies are needed to investigate whether metabolic derangements with different etiologies other then a genetic defect in mitochondrial [beta]-oxidation can be distinguished by this method. Moreover, it would be worthwhile to investigate the sensitivity of short-chain 3-hydroxy-acyl-CoA dehydrogenase to various potentially intoxicating products.

In conclusion, the present method is selective, sensitive, and accurate, and thus is the method of choice for the simultaneous quantitation of FFAs and 3OHFAs. One to 2 h for sample preparation and 40 min for the GC-MS analysis are all the time required. This method can be very helpful for less-equipped metabolic laboratories that depend on electron impact mass spectrometry. It is an attractive alternative to the powerful tandem-MS techniques [7, 8], which require very costly instrumentation. The cost of the present analysis will be comparable with that of urine organic acid analysis. In daily practice, it can be used in all cases that show inconclusive urine organic acid profiles.

This work was made possible in part by grant BD 2589/93-ID (to C.G.C.) of "Junta Nacional de Investigacao Cientifica e Tecnologica-Programa Praxis XXI" (Lisbon, Portugal).


[1.] Stanley CA. Disorders of fatty acid oxidation. In: Fernandes J, Saudubray JM, van den Berghe G, eds. Inborn metabolic diseases diagnosis and treatment, 2nd ed. Berlin: Springer-Verlag, 1995: 133-43.

[2.] Stanley CA, Hale DE. Genetic disorders of mitochondrial fatty acid oxidation. Curr Opin Pediatr 1994;6:476-81.

[3.] Hale DE, Bennett MJ. Fatty acid oxidation disorders: a new class of metabolic diseases. J Pediatr 1992;121:1-11.

[4.] Roe CR, Coates PM. Mitochondrial fatty acid oxidation disorders. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The metabolic and molecular bases of inherited disease, 7th ed. New York: McGraw-Hill, 1995:1501-33.

[5.] Rabier D, Bardet J, Parvy P, Poggi F, Brivet M, Saudubray JM, Kamoun P. Do criteria exist from urinary organic acids to distinguish [beta]-oxidation defects? J Inherited Metab Dis 1995;18:257-60.

[6.] Costa CG, Struys EA, Bootsma A, ten Brink HJ, Dorland L, Tavares de Almeida I, et al. Quantitative profiling of plasma acylcarnitines using gas chromatography chemical ionization mass fragmentography. J Lipid Res 1997;38:173-82.

[7.] Millington DS, Norwood DL, Kodo N, Roe CR, Inove F. Application of fast atom bombardment with tandem mass spectrometry and liquid chromatography/mass spectrometry to the analysis of acylcarnitines in human urine, blood and tissue. Anal Biochem 1989;180:331-9.

[8.] Millington DS, Kodo N, Norwood DL, Roe CR. Tandem mass spectrometry: a new method for acylcarnitine profiling with potential for neonatal screening for inborn errors of metabolism. J Inherited Metab Dis 1990;13:321-4.

[9.] Brivet M, Slama A, Saudubray JM, Legrand A, Lemonnier A. Rapid diagnosis of long chain and medium chain fatty acid oxidation disorders in lymphocytes. Ann Clin Biochem 1995;32:154-9.

[10.] Wanders RJA, Ulst L, van Elk E, de Klerk JBC, Przyrembel H. Octanoate and palmitate R-oxidation in human leukocytes: implications for the rapid diagnosis of fatty acid R-oxidation disorders. J Inherited Metab Dis 1991;14:317-20.

[11.] Moon A, Rhead WJ. Complementary analysis of fatty acid oxidation disorders. J Clin Invest 1987;79:59-64.

[12.] Nada MA, Rhead WJ, Sprecher H, Schulz H, Roe CR. Evidence for intermediate channeling in mitochondrial beta-oxidation. J Biol Chem 1995;270:530-5.

[13.] Schaefer J, Pourfarzam M, Bartlett K, Jackson S, Turnbull DM. Fatty acid oxidation in peripheral blood cells: characterization and use for the diagnosis of defects of fatty acid oxidation. Pediatr Res 1995;37:354-60.

[14.] IJIst L, Wanders RJA. A simple, straightforward assay for long-chain 3-hydroxyacyl-CoA dehydrogenase based on the use of 4ethylmaleimide: potential for pre and postnatal diagnosis. J Inherited Metab Dis 1993;16:568-70.

[15.] Bertrand C, Largilliere C, Zabot MT, Mathieu M, Vianey-Saban C. Very long-chain acyl-CoA dehydrogenase deficiency: identification of a new inborn error of mitochondrial fatty acid oxidation in fibroblasts. Biochim Biophys Acta 1993;1180:327-9.

[16.] Yokoda I, Indo Y, Coates PM, Tanaka K. Molecular basis of medium-chain acyl-coenzyme A dehydrogenase deficiency. J Clin Invest 1990;86:1000-3.

[17.] IJIst L, Ruiter JPN, Hoovers JMN, Jakobs ME, Wanders RJA. Common missense mutation G1528C in long-chain 3-hydroxyacylCoA dehydrogenase deficiency. Characterization and expression of mutant protein, mutation analysis on genomic DNA and chromosomal localization of the mitochondrial trifunctional protein a subunit gene. J Clin Invest 1996;98:1028-33.

[18.] Dorland L, Ketting D, Bruinvis L, Duran M. Medium and long-chain 3-hydroxymonocarboxylic acids: analysis by gas chromatography combined with mass spectrometry. Biomed Chromatogr 1991;5: 161-4.

[19.] Hagenfeldt L, von Dobeln U, Holme E, Alm J, Brandberg G, Enocksson E, Lindeberg L. 3-Hydroxydicarboxylic aciduria--a fatty acid oxidation defect with severe prognosis. J Pediatr 1990;116: 387-92.

[20.] Bergman AJ, Donckerwolcke RA, Duran M, Smeitink JA, Mousson B, Vianey-Saban C, Poll-The BT. Rate-dependent distal renal tubular acidosis and carnitine palmitoyltransferase I deficiency. Pediatr Res 1994;36:582-8.

[21.] Smeitink JAM, Semmekrot BA, Raaijmakers BCMS, Scholte HR, van de Heuvel LPWJ, Costa C, et al. Treatment and molecular analysis of neonatal carnitine palmitoyltransferase II deficiency. Pediatr Res (in press).

[22.] Dorland L, Horn ME, Struys EA, Costa CG, Ketting D, Duran M, et al. Novel FAB-MS of plasma acylcarnitines using single mass spectrometry. 33rd SSIEM Annual Symposium, Toledo, Spain, 1995:P-137.

[23.] Altman DG, ed. Practical statistics for medical research. London: Chapman and Hall, 1991;178-228.

[24.] Donike M. Acylation with bis (acylamides). N-Methyl-bis (trifluoroacetamide) and bis (trifluoroacetamide) two new reagents for trifluoroacetylation. J Chromatogr 1973;78:273-9.

[25.] Sullivan JE, Schewe LR. Preparation and gas chromatography of highly volatile trifluoroacetylated carbohydrates using 4methyl bis (trifluoroacetamide). J Chromatogr Sci 1977;15:196-7.

[26.] Woollard PM. Selective silylation using tertbutyldimethylsilyl reagents. Their use in the quantification of fatty acids. Biomed Mass Spectrom 1983;10:143-54.

[27.] Kim YR, Hahn MK, Zlatkis A, Horning EC, Middleditch BS. Simultaneous gas chromatography of volatile and non-volatile carboxylic acids as tertbutyldimethylsilyl derivatives. J Chromatogr 1989; 468:289-301.

[28.] Schwenk WF, Berg PJ, Beaufrere B, Miles JM, Haymond MR. Use of t-butyldimethylsilylation in the gas chromatographic/mass spectrometric analysis of physiologic compounds found in plasma using electron impact ionization. Anal Biochem 1984;141:101-9.

[29.] Laine RA, Young ND, Gerber JN, Sweeley CC. Identification of 2-hydroxy fatty acids in complex mixtures of fatty acid methyl esters by mass chromatography. Biomed Mass Spectrom 1974;1:10-4.

[30.] Drozo J. Chemical derivatization in gas chromatography. J Chromatogr 1975;113:303-56.

[31.] Mallet AI, Barr RM, Newton JA. Analysis of dihydroxyeicosatetraenoic acids by gas chromatography-mass spectrometry. J Chromatogr Biomed Appl 1986;378:194-200.

[32.] Vockley J. The changing face of disorders of fatty acid oxidation. Mayo Clin Proc 1994;69:249-57.

[33.] Hagenfeldt L, Venizelos N, von Dobeln U. Clinical and biochemical presentation of long-chain 3-hydroxyacyl-CoA dehydrogenase deficiency. J Inherited Metab Dis 1995;18:245-8.

[34.] Duran M, Bruinvis L, Ketting D, de Klerk JBC, Wadman SK. cis-4-Decenoic acid in plasma: a characteristic metabolite in medium-chain acyl-CoA dehydrogenase deficiency. Clin Chem 1988;34:548-51.

[35.] Onkenhout W, Venizelos V, van de Poel PFH, van de Heuvel MPM, Poorthuis BJHM. Identification and quantitation of intermediates of unsaturated fatty acid metabolism in plasma of patients with fatty acid oxidation disorders. Clin Chem 1995;41:1467-74.

[36.] Haidukewych D, Forsythe WI, Sills M. Monitoring octanoic and decanoic acids in plasma from children with intractable epilepsy treated with medium-chain triglyceride diet. Clin Chem 1982;28: 642-5.


(1) Centro de Metabolismos e Genetica, University of Lisboa, Lisbon 1699, Portugal.

(2) Free University Hospital, Department of Clinical Chemistry (Metabolic Unit), Amsterdam, The Netherlands.

(3) Wilhelmina Kinderziekenhuis, University of Utrecht, Utrecht, The Netherlands.

(4) Nonstandard abbreviations: CPT, carnitine palmitoyltransferase; FFA, free fatty acid; 3OHFA, 3-hydroxyfatty acid; GC-MS, gas chromatography mass spectrometry; TFA, trifluoroacetylation; tBDMS, tert-butyldimethylsilylation; MTBSTFA, N-methyl-N-(tert-butyldimethylsilyl)trifluoroacetamide; MBTFA, methyl-bis(trifluoroacetamide); LCHAD, 3-hydroxyacyl-CoA dehydrogenase; MTP, mitochondrial trifunctional protein; vLCAD, very-long-chain acyl-CoA dehydrogenase; MCAD, medium-chain acyl-CoA dehydrogenase; and MADD, multiple acyl-CoA dehydrogenase deficiency.

* Address correspondence to this author at: University Children's Hospital, "Het Wilhelmina Kinderziekenhuis", Nieuwe Gracht 137, 3512 LK Utrecht, The Netherlands. Fax 31-30-232 0793; e-mail

Received August 11, 1997; revision accepted December 10, 1997.
Table 1. Extraction recoveries from plasma and Intraassay and
Interassay reproducibility for 14 FFA and their respective 3OHFA

 Intraassay reproducibility (n = 4)

 Recovery, % Avg
Fatty acid (n = 4) ([micro]mol/L) CV,%

Hexanoic 105 15.55 2.06
3-OH-hexanoic 84 10.97 4.96
Octanoic 92 11.7 1.8
3-OH-octanoic 92 11.7 3.56
Decanoic 101 13.14 1.63
3-OH-decanoic 100 12.6 2.31
Dodecanoic 99 13.02 1.8
3-OH-dodecanoic 100 12.96 2.06

Tetradecanoic 95 24.72 3.71
3-OH-tetradecanoic 99 12.36 6.08
Hexadecanoic 102 40.44 2.1
3-OH-hexadecanoic 86 10.85 1.91
Octadecanoic 96 44.59 1.25
3-OH-octadecanoic 98 12.33 4.39

 Interassay reproducibility (n = 3)

Fatty acid ([micro]mol/L CV,%

Hexanoic 1.77 11.77
3-OH-hexanoic 6.57 8.29
Octanoic 5.57 9.17
3-OH-octanoic 1.20 10.64
Decanoic 9.28 4.56
3-OH-decanoic 2.15 10.74
Dodecanoic 24.77 9.36
3-OH-dodecanoic 1.61 10.50
Tetradecanoic 93.40 10.62
3-OH-tetradecanoic 2.23 10.81
Hexadecanoic 25.62 6.68
3-OH-hexadecanoic 15.52 7.54
Octadecanoic 110.31 9.21
3-OH-octadecanoic 2.97 11.6

Table 2. Quantitative ([micro]mol/L) profiling of plasma 3OHFA
in controls and patients with known defects of mitochondrial
fatty acid [beta]-oxidation.

 6:0 8:0 10:0 12:0 14:2
Patient Sample CS (171) (199) (227) (255) (279)

LCHAD1 1 C 3.6 4.4 5.9 5.5 0.33
LCHAD2 1 C 1.7 1.9 2.1 1.2 0.18
LCHAD3 1 C 0.79 0.91 0.51 1.1 0.16
 2 A 0.19 0.09 0.1 0.14
 3 A 0.19 0.11 0.12 0.15
LCHAD4 1 A 0.52 0.1 0.09 0.1
LCHAD5 1 A 0.48 0.2 0.19 0.13
MTP 1 C 6.9 1.1 2.4 1.8 0.13
MCAD1 1 C 5.8 4.4 0.13 0.08
vLCAD1 1 C 8.0 2.0 0.01
 2 PM 3.1 0.12 0.41
vLCAD2 1 C 1.4 0.87 1.4 0.49
MADD-N (c) 1 C 15 2.2 0.27 0.03
MADD-m1 (d) 1 A 0.24 0.11 0.21 0.01
MADD-m2 1 A 1.3 1.1 0.33 0.13
MADD-m3 1 A 0.92 1.3 0.35 0.06
 2 (a) A 3.0 2.3 0.63 0.22
 3 (b) A 2.1 2.1 0.68 0.21
CPTI 1 A 0.43 0.96 0.18 0.13
CPTII 1 C 1.1 0.3 0.22 0.04
Controls AVG 0.31 0.2 0.15 0.07
(n = 19) Range 0.05-1 0.005-1 0.02-0.27 <0.152

 14:1 14:0 16:2 16:1 16:0 18:2
Patient (281) (283) (307) (309) (311) (335)

LCHAD1 2.0 4.1 0.10 15 0.27
LCHAD2 1.1 1.3 0.25 0.08 5.3 0.46
LCHAD3 0.53 0.77 0.16 0.29 4.1 0.91
 0.09 0.18 0.004 0.06 0.58 2.3
 0.05 0.1 0.02 0.11 1.2 2.6
LCHAD4 0.13 0.09 2.1 8.8
LCHAD5 0.25 0.81 0.12 0.25 1.8 4.5
MTP 0.45 2.0 0.03 0.53 4.4 3.6
MCAD1 0.11
vLCAD1 0.03

vLCAD2 0.38
MADD-N (c) 0.02 0.10 0.10 0.08 0.60 0.03
MADD-m1 (d) 0.08 0.08 0.22 0.01
MADD-m2 0.21 0.15
MADD-m3 0.08 0.25
 0.49 0.52
 0.29 0.35
CPTI 0.01 0.04
CPTII 0.11 0.08 0.58
Controls 0.01 0.02 0.004 0.18 0.03
(n = 19) <0.09 <0.123 <0.017 0.08-0.29 <0.104

 18:1 18:0
Patient (337) (339)

LCHAD1 2.1 3.2
LCHAD2 0.83 2.8
LCHAD3 3.7 4.1
 0.68 1.3
 0.69 1.4
LCHAD4 1.6 4.6
LCHAD5 2.6 2.4
MTP 1.2 2.6

MADD-N (c) 0.05
MADD-m1 (d) 0.03

Controls 0.04 0.02
(n = 19) <0.130 <0.199

In parentheses below each 30HFA Is the m/z the respective
fragment ion used for quantitation purposes. CS, clinical
status; C, crisis; A, asumptomatic, PM.
postmortem; AVG, average concentration.

(a) 12-h fast. (b) 20-h fast. (c) Neonatal. (d) Mild.

Table 3. Quantitative ([micro]mol/L) profiling of plasma FFA in
controls and patients with known defects of mitochondrial
fatty acid [beta]-oxidation.

 6:0 8:0 10:1 10:0
Patient Sample CS (173) (201) (227) (229)

LCHAD1 1 C 14 4.2 0.78 4.7
LCHAD2 1 C 5.0 4.1 0.85 5.0
LCHAD3 1 C 3.1 5.0 0.38 8.1
 2 A 1.4 2.3 0.02 2.5
 3 A 0.75 3.3 0.14 9.7
LCHAD4 1 A 1.6 6.8 0.02 11
LCHAD5 1 A 4.6 2.7 2.0 4.2
MTP 1 C 2 21 0.10 8.9
MCAD1 1 C 61 408 103 20
vLCAD1 1 C 6.9 8.3 2.8 17
 2 PM 2.0 1.6 0.13 20
vLCAD2 1 C 3.2 2.4 2.6 5.4
MADD-N (c) 1 C 154 110 36 71
MADD-m1 (d) 1 A 1.2 3.3 0.70 4.7
MADD-m2 1 A 6.7 13 3.1 10
MADD-m3 1 A 2.2 9.7 1.3 5.9
 2 (a) A 5.5 13 5.0 17
 3 (b) A 4.5 14 6.3 19
CPTI 1 A 8.1 10 0.35 4.5
CPTII 1 C 3.9 10 0.40 3.6
Controls AVG 2.4 2.1 0.26 1.8
(n = 19) Range 0.50-4.2 0.10-10 0.04-0.60 0.6-4.0

 12:1 12:0 14:2 14:1 14:0 16:2
Patient (255) (257) (281) (283) (285) (309)

LCHAD1 0.56 9.4 1.6 7.6 139 0.05
LCHAD2 0.31 2.6 0.53 1.5 46 0.20
LCHAD3 0.28 4.8 0.85 1.5 15 1.6
 0.04 1.0 0.89 0.99 10 0.64
 0.23 4.5 1.2 2.4 37 0.72
LCHAD4 0.02 0.25 4.5 0.66
LCHAD5 0.64 40 3.9 5.8 121 0.97
MTP 0.23 27 1.3 2.8 103 2.7
MCAD1 6.0 33 0.24 0.72 57 0.68
vLCAD1 0.60 195 8.5 55 172 3.0
 20 1.7 16 49 0.47
vLCAD2 1.2 23 21 191 51 1.2
MADD-N (c) 2.9 80 35 82 123 7.5
MADD-m1 (d) 0.68 7.3 0.89 7.0 23 0.20
MADD-m2 1.6 13 0.25 0.8 14 0.21
MADD-m3 0.88 20 0.06 0.18 6.7 0.02
 1.6 21 6.7 24 255 0.20
 2.7 47 0.18 3.7 63 0.07
CPTI 0.09 14 37
CPTII 0.16 8.6 0.29 0.74 84 0.34
Controls 0.06 3.7 0.03 0.09 10 0.04
(n = 19) <0.2 0.96-12 <0.191 <0.38 3.23-23 <0.14

 16:1 16:0 18:2 18:1 18:0
Patient (311) (313) (337) (339) (341)

LCHAD1 107 523 32 234 91
LCHAD2 16 240 74 181 50
LCHAD3 18 112 57 99 57
 3.2 61 16 27 7
 32 237 45 99 46
LCHAD4 6.2 131 115 357 158
LCHAD5 80 545 218 275 93
MTP 70 347 128 314 76
MCAD1 39 398 585 1573 448
vLCAD1 120 1300 371 1127 126
 24 223 107 217 29
vLCAD2 21 260 93 271 110
MADD-N (c) 87 1128 317 441 67
MADD-m1 (d) 10 193 15 60 30
MADD-m2 16 175 119 311 235
MADD-m3 6.6 64 36 82 111
 145 897 197 830 292
 84 626 379 1612 361
CPTI 70 473 196 170 121
CPTII 148 1349 59 277 67
Controls 8.3 109 45 89 47
(n = 19) 1.8-19 58-221 13-167 37-264 17-75

In parentheses below each FFA is the m/z of the respective
fragment ion used for quantitation purposes. CS, clinical
status; C, crisis; A, asymptomatic; PM, postmortem; AVG,
average concentration.

(a) 12-h fast.

(b) 20-h fast.

(c) Neonatal.

(d) Mild.
COPYRIGHT 1998 American Association for Clinical Chemistry, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1998 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Molecular Pathology and Genetics
Author:Costa, Catarina G.; Dorland, Lambertus; Holwerda, Ulbe; De Almeida, Isabel Tavares; Poll-The, Bwee-T
Publication:Clinical Chemistry
Date:Mar 1, 1998
Previous Article:Assessment of sequence-based p53 gene analysis in human breast cancer: messenger RNA in comparison with genomic DNA targets.
Next Article:Expression of prostate-specific antigen and prostate-specific membrane antigen transcripts in blood cells: implications for the detection of...

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |