Clinical Chemistry 50: 1391-1395, 2004.
First published May 27, 2004; 10.1373/clinchem.2004.033399
(Clinical Chemistry. 2004;50:1391-1395.)
© 2004 American Association for Clinical Chemistry, Inc.
|
Endocrinology and Metabolism |
Measurement of Urinary D- and L-2-Hydroxyglutarate Enantiomers by Stable-Isotope-Dilution Liquid Chromatography–Tandem Mass Spectrometry after Derivatization with Diacetyl-L-Tartaric Anhydride
Eduard A. Struysa,
Erwin E.W. Jansen,
Nanda M. Verhoeven and
Cornelis Jakobs
1 Metabolic Unit, Department of Clinical Chemistry, VU Medical Center, Amsterdam, The Netherlands
aAddress correspondence to this author at: Metabolic Unit, Department of Clinical Chemistry, VU Medical Center, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands. E-mail e.struys{at}vumc.nl.
 |
Abstract
|
|---|
Background: The differential diagnosis of D-2-hydroxyglutaric aciduria (D-2-HGA), L-2-hydroxyglutaric aciduria (L-2-HGA), and the combined D/L-2-hydroxyglutaric aciduria (D/L-2-HGA) can be accomplished only by the measurement of the corresponding 2-hydroxyglutarate (2-HG). Available methods for the determination of D- and L-2-HG in urine are either time-consuming and expensive or have not been extensively validated. We aimed to develop a method for their rapid and sensitive measurement.
Methods: We used liquid chromatography–tandem mass spectrometry (LC-MS/MS) for the determination of D- and L-2-HG with stable-isotope-labeled internal standards. Urine samples of 20 µL were mixed with 250 µL of methanol containing the internal standards and subsequently dried under nitrogen. The analytes were derivatized by use of diacetyl-L-tartaric anhydride (DATAN) to obtain diastereomers, which were separated on an achiral C18 HPLC column and detected by MS/MS in multiple-reaction-monitoring mode.
Results: The use of DATAN as chiral derivatization reagent provided very well separated peaks of the formed diastereomers of D- and L-2-HG, with a total runtime of 5 min. The inter- and intraassay CVs for D- and L-2-HG ranged from 3.4% to 6.2%. Mean recoveries of D- and L-2-HG, evaluated on two concentrations, were 94%. Detection limit of the presented method was 20 pmol for a sample volume of 20 µL. Method comparison of the LC-MS/MS method with a gas chromatography–mass spectrometry method, in which D- and L-2-HG were derivatized with R-(–)-butanol, showed good agreement between the two methods.
Conclusions: Urinary D- and L-2-HG can be analyzed by MS/MS after derivatization with DATAN. The presented method may be suitable for the differential diagnosis of 2-HGA.
|
Introduction
|
|---|
The enantiomeric determination of D-2-hydroxyglutarate (D-2-HG)1
and L-2-hydroxyglutarate (L-2-HG) is the only reliable test for the differential diagnosis of D-2-hydroxyglutaric aciduria (D-2-HGA), L-2-hydroxyglutaric aciduria (L-2-HGA), and the combined D/L-2-hydroxyglutaric aciduria (D/L-2-HGA). In many cases, urinary organic acid analysis reveals increased excretion of 2-HG. Both D- and L-2-HGA were described in 1980 (1)(2), whereas the combined D/L-2-HGA was described in 2000 (3). At present, >75 patients with L-2-HGA and >40 patients with D-2-HGA have been diagnosed, and the number of patients is still growing. Little is known about the underlying metabolic and genetic defects that lead to these organic acidurias. Recently, using mass isotopomer analyses of labeled products formed by human lymphoblasts grown on media supplemented with labeled substrates, we found that there is a strong metabolic relationship between D-2-HG and the Krebs cycle intermediate 2-ketoglutarate (4).
The number of published methods for the enantiomeric chromatographic determination of urinary D- and L-2-HG is limited. In our laboratory, we use the stable-isotope-dilution (SID) method described by Gibson et al. (5), in which D- and L-2-HG are converted to their corresponding di-R-butyl-O-acetyl derivatives followed by analysis by gas chromatography–mass spectrometry (GC-MS). The method described by Kim et al. (6), in which the analytes were converted to O-trifluoroacetyl-(–)-menthyl derivatives followed by GC, is a second example of the use of a chiral derivatization reagent to achieve enantiomeric separation. In other methods, a chiral GC column connected to MS is used to separate the formed achiral derivatives of D- and L-2-HG (7)(8)(9). Recently, a liquid chromatography–tandem MS (LC-MS/MS) method was described in which nonderivatized D- and L-2-HG were separated on a chiral LC column (10).
With the exception of the method of Gibson et al. (5), the above-mentioned methods have not been extensively validated and/or the applicability was shown only in patient urine samples containing massive amounts of the corresponding 2-HG.
The method currently used in our laboratory (5) is accurate, but it has the disadvantage of being time-consuming and, because of the use of the reagent R-(–)-2-butanol, expensive. We therefore developed a reliable, robust, and simple SID LC-MS/MS method for the enantiomeric analysis of D- and L-2-HG in urine.
|
Materials and Methods
|
|---|
materials
D-2- and L-2-HG were obtained from Sigma. Diacetyl-L-tartaric anhydride (DATAN) was from Aldrich. [3,3,4,4-2H4]-D/L-2-HG (isotopic purity >98%) was prepared by chemical reduction of 
-keto[3,3,4,4-2H4]glutarate as described previously (5). All other reagents and solvents were of analytical grade.
sample preparation
A 20-µL urine aliquot was pipetted into a glass GC vial, and 250 µL of methanol containing 0.004 mmol/L 2H4-D/L-2-HG was added as internal standard (IS). The vial contents were evaporated to dryness at 50 °C by a gentle flow of nitrogen. The di-acetyltartaryl derivative was prepared by treating the dry residue with 50 µL of freshly made 50 g/L DATAN in dichloromethane–acetic acid (4:1 by volume). The vial was capped and heated at 75 °C for 30 min. After the vial was cooled to room temperature, the mixture was evaporated to dryness by a nitrogen stream at room temperature, the residue was redissolved in 500 µL of distilled water, and 10 µL of the aqueous solution was injected on the LC column. We also processed 20-µL aliquots of the aqueous calibrators (included in each batch of samples) containing 0, 50, 100, 200, 500, and 1000 pmol of the individual enantiomers as described above. The concentration of the analyte in urine was calculated by interpolation of the observed analyte/IS peak-area ratio based on the linear regression line for the calibration curve, which was obtained by plotting peak-area ratios vs analyte concentration. The outcome was multiplied by 50 (20-µL urine volume used for sample preparation) and divided by the creatinine concentration of the urine sample to obtain outcomes in mmol/mol creatinine.
methods
All analyses were performed on an API 3000 triple-quadrupole tandem mass spectrometer (Applied Biosystems). Other instrumentation consisted of a Perkin-Elmer Series 200 HPLC pump, a Perkin-Elmer Series 200 autosampler, and a Harvard Apparatus Pump 11 infusion pump. LC was performed on an Xterra C18 analytical column [150 x 3.9 mm (i.d.); 5-µm bead size] with water–acetonitrile (96.5:3.5 by volume) containing 125 mg/L ammonium formate (pH adjusted to 3.6 by addition of formic acid) as mobile phase. The flow of 0.75 mL/min was split postcolumn at a ratio of 4:1, giving an inlet flow of 150 µL/min into the mass spectrometer, which operated in the negative multiple-reaction-monitoring mode. The MS/MS system was optimized by continuous infusion of a 10 µmol/L calibrator at a flow rate of 20 µL/min. The temperature of the electrospray was set at 500 °C; turbo ion gas (nitrogen) was used at a flow of 8 L/min; and the ion spray voltage was set at –4500 V. Other settings were as follows: collision energy, –7.5 V; declustering potential, –10 V; nebulizer gas flow; 10 L/min; curtain gas flow, 8 L/min. Data were acquired and processed by Analyst software for Windows (Ver. 1.3.1).
|
Results
|
|---|
derivatization and mass spectra
The reaction scheme for the formation of the DATAN derivatives of D- and L-2-HG is shown in Fig. 1
. This derivatization procedure produced a set of D- and L-2-HG diastereomers, which allowed separation by a nonchiral LC column. The derivatization was complete in 30 min (data not shown). Initial scans on the first quadrupole of the MS showed intensive signals at m/z –363.2 and m/z –367.2 for 2-HG and the IS, respectively, which correspond to the deprotonated derivatives. Collision-activated decomposition MS/MS spectra of the formed derivatives of nonlabeled D/L-2-HG and labeled IS are shown in Fig. 2
. At a collision energy of –7.5 V, exclusive and highly efficient fragments of m/z –147.1 and m/z –151.1 were generated in the collision cell, originating from the nonlabeled and labeled 2-HGs, respectively. These fragments corresponded to the HG backbone of the derivative and contained all of the deuterium labeling present in the IS. For the quantitative measurement of D- and L-2-HG, the following multiple-reaction-monitoring transitions were used: endogenous 2-HG, m/z –363.2
m/z –147.1; IS, m/z –367.2m/z –151.1.
View larger version (7K):
[in this window]
[in a new window]
|
Figure 1. Derivatization scheme for the formation of DATAN derivatives of D- and L-2-HG.
* indicate asymmetric carbon atoms. HAc, acetic acid.
|
|
View larger version (17K):
[in this window]
[in a new window]
|
Figure 2. Collision-activated decomposition MS/MS spectra of the formed derivatives of nonlabeled D/L-2-HG (top) and labeled 2H4-D/L-2-HG (bottom) with a collision energy of –7.5 V.
|
|
chromatography
Separation of the DATAN derivatives of D- and L-2-HG strongly depended on the pH of the mobile phase. Use of pH 3.6 enabled us to obtain baseline separation with a total LC runtime of 5 min. The elution order was determined by use of enantiomeric pure standards: L-2-HG eluted at 3.5 min, and D-2-HG eluted at 4.3 min. The separation factor () was 1.33, as calculated by:
 | (1) |
The resolution of the two peaks (Rs) was calculated by the equation:
 | (2) |
Where 
t represents the time difference between the apex of the two peaks; wb1 represents the peak width at baseline of peak 1; and wb2 represents the peak width at baseline of peak 2. The Rs was 1.8, indicating baseline separation. Representative mass fragmentograms are shown in Fig. 3
. In the mass fragmentogram on the left, a distinct signal can be seen eluting just after D-2-HG. Use of proper standards enabled us to identify this signal as 3-HG, an important metabolite accumulating in glutaric aciduria type I and short-chain 3-hydroxyacyl-CoA dehydrogenase deficiency (11).
View larger version (17K):
[in this window]
[in a new window]
|
Figure 3. Mass fragmentograms of D- and L-2-HG.
Shown are a control urine sample (top left panel), 100-fold diluted urine sample from a patient with D-2-HGA (top middle panel), and a 100-fold diluted urine sample from a patient with L-2-HGA (top right panel). L-2-HG elutes at 3.5 min, D-2-HG elutes at 4.3 min, and 3-hydroxyglutarate (3-OHG) elutes at 4.8 min. Top panels represent the endogenous 2-HGs; bottom panels represent the IS.
|
|
linearity
The calibration curves were linear over the range of 0–1000 pmol for D- and L-2-HG. In all cases, the coefficient of linear regression (r2) was >0.995 for both calibration curves.
limits of detection, precision, and recovery
The minimum detectable concentration for both D- and L-2-HG (signal-to-noise ratio >5) was 20 pmol, which corresponds to a concentration of 1 µmol/L for a sample volume of 20 µL. The limit of detection was estimated by verifying the peak height of the analyte and the noise in the chromatographic region of the analyte. The formed derivatives of D- and L-2-HG were stable for >7 days when stored at 4 °C.
The validation data for the presented method are listed in Tables 1
and 2
. All validation experiments were performed with a pooled urine sample with a creatinine concentration of 4.1 mmol/L. The intra- and interassay CVs (n = 8) for both D- and L-2-HG ranged from 3.4% to 6.2%. For recovery experiments, we added 0.2 nmol and 0.4 nmol of the individual enantiomers to 20-µL urine sample. The recoveries (n = 8) were 91–97%, with CVs of 2.6–6.4%.
lc-ms/ms vs gc-ms method comparison
We validated the performance of the presented LC-MS/MS method by comparing it with our current SID GC-MS method for the determination of D- and L-2-HG in urine. We established control values by analyzing urine samples from 12 healthy children (age range, 4–17 years), who donated urine after parental consent. The mean (SD) concentrations were 5.1 (2.2) mmol/mol creatinine (range, 3.3–10.8 mmol/mol creatinine) and 4.1 (1.8) mmol/mol creatinine (range, 2.2–9.3 mmol/mol creatinine) for L-2-HG and D-2-HG, respectively. These values were in agreement with the reference values currently used for the GC-MS method: 6.0 (5.4) mmol/mol creatinine (range, 1.3–18.9 mmol/mol creatinine) and 6.0 (3.6) mmol/mol creatinine (range, 2.8–17.0 mmol/mol creatinine) for L-2-HG and D-2-HG, respectively. Urine samples from patients with confirmed HGA were reanalyzed by the LC-MS/MS method. This group of patients consisted of 10 with L-2-HGA, 12 with D-2-HGA, and 7 patients with combined increased concentrations of D- and L-2-HG. Before addition of the IS, the urine samples were properly diluted to obtain peak-area ratios in the range of the calibration curve. The results are displayed in Fig. 4
, in which the outcomes obtained by the GC-MS method are displayed on the y axis, and the results obtained by the LC-MS/MS method are displayed on the x axis. Linear data fitting of the values gave the equations: y = 0.93x – 8.1 mmol/mol creatinine (r2 = 0.98) for L-2-HG; and y = 0.99x – 7.0 mmol/mol creatinine (r2 = 0.97) for D-2-HG. The mean absolute differences, i.e., without the sign of the found difference, between the two methods were 12.3 (8.0)% and 13.4 (8.7)% for L-2-HG and D-2-HG, respectively. The Bland–Altman method comparison, which does take into account the sign of the found difference, gave differences between the LC-MS/MS vs GC/MS outcomes of +2.6% and –2.5% for L-2-HG and D-2-HG, respectively.
View larger version (15K):
[in this window]
[in a new window]
|
Figure 4. xy plot for method comparison of LC-MS/MS and GC/MS analyses of urine.
LC-MS/MS results are displayed on the x axis, and the corresponding GC-MS results are displayed on the y axis.
|
|
|
Discussion
|
|---|
We have developed an analytical tool for the differential diagnosis of 2-HGAs in urine. In this method we took advantage of the specificity and selectivity of LC-MS/MS, which requires little sample clean up. Our current GC-MS method is accurate but is laborious and expensive because it requires the use of the reagent R-(–)-2-butanol (US $10.00 per sample). The sample preparation time in the LC-MS/MS method is considerably shorter: only 20 µL if urine needs to be pipetted into a vial, followed by a drying step, 30 min for derivatization, another drying step, and redissolving of the residue. The derivatization reagent DATAN is not expensive, has great separation potential, and is a good leaving group in the collision cell, allowing the use of low collision energy and thereby increasing the selectivity of the measurement. In spite of these unique characteristics, the reagent has rarely been mentioned in the literature (12)(13)(14)(15). Other biochemically important hydroxyacids, i.e., D/L-lactate, D/L-glycerate, and D/L-3hydroxybutyrate, can also be separated enantiomerically by use of DATAN as the derivatizing reagent, without the need to change chromatographic conditions (data not shown). The two-step derivatization procedure in the GC-MS method (butylation followed by acetylation) induced 5–7% racemization. Racemization is <1% for the proposed LC-MS/MS method, ensuring accurate results in cases in which there is a pronounced difference in the concentrations of D- and L-2-HG.
As shown by the validation data for the LC-MS/MS method (Tables 1
and 2
), the method is well suited for analysis of urinary D- and L-2-HG. The limit of detection (0.02 nmol in 20 µL of urine) enabled the quantification of D- and L-2-HG in urine samples from healthy children. The results of D- and L-2-HG by LC-MS/MS measurements were compared with the results obtained by GC-MS, and there was good agreement between the two analytical procedures with no systematic difference.
In conclusion, the presented SID LC-MS/MS method for the determination of D- and L-2-HG in urine is a good alternative for our current GC-MS method. In the future, we plan to adapt this procedure to other body fluids, such as amniotic fluid, plasma, and cerebrospinal fluid.
|
Footnotes
|
|---|
1 Nonstandard abbreviations: D-2-HG and L-2-HG, D-2- and L-2-hydroxyglutarate, respectively; D-2-HGA and L-2-HGA, D-2- and L-2-hydroxyglutaric aciduria, respectively; SID, stable-isotope dilution; GC-MS, gas chromatography–mass spectrometry; LC-MS/MS, liquid chromatography–tandem mass spectrometry; DATAN, diacetyl-L-tartaric anhydride; and IS, internal standard. 
|
References
|
|---|
- Chalmers RA, Lawson AM, Watts RWE, Tavill AS, Kamerling JP, et al. D-2-Hydroxyglutaric aciduria: case report and biochemical studies. J Inherit Metab Dis 1980;3:11-15.[Medline]
[Order article via Infotrieve]
- Duran M, Kamerling JP, Bakker HD, van Gennip AH, Wadman SK. L-2-Hydroxyglutaric aciduria: an inborn error of metabolism. J Inherit Metab Dis 1980;3:109-112.[CrossRef][Medline]
[Order article via Infotrieve]
- Muntau AC, Röschinger W, Merkernschlager A, van der Knaap MS, Jakobs C, Duran M, et al. Combined D-2- and L-2-hydroxyglutaric aciduria with neonatal onset encephalopathy: a third biochemical variant of 2-hydroxyglutaric aciduria. Neuropediatrics 2000;31:137-140.[Medline]
[Order article via Infotrieve]
- Struys EA, Verhoeven NM, Brunegraber H, Jakobs C. Investigations by mass isotopomer analysis of the formation of D-2-hydroxyglutarate by cultured lymphoblasts from two patients with D-2-hydroxyglutaric aciduria. FEBS Lett 2004;557:115-120.[Medline]
[Order article via Infotrieve]
- Gibson KM, ten Brink HJ, Schor DS, Kok RM, Bootsma AH, Hoffmann GF, et al. Stable-isotope dilution analysis of D- and L-2-hydroxyglutaric acid: application to the detection and prenatal diagnosis of D- and L-2-hydroxyglutaric acidemias. Pediatr Res 1993;34:277-280.[Web of Science][Medline]
[Order article via Infotrieve]
- Kim KR, Lee J, Ha D, Jeon J, Park HG, Kim JH. Enantiomeric separation and discrimination of 2-hydroxy acids as O-trifluoroacetylated (S)-(+)-3-methyl-2-butyl esters by achiral dual-capillary column gas chromatography. J Chromatogr A 2000;874:91-100.[Medline]
[Order article via Infotrieve]
- Muth A, Mosandl A, Wanders RJ, Nowaczyk MJ, Baric I, Bohles H, et al. Stereoselective analysis of 2-hydroxysebacic acid in urine of patients with Zellweger syndrome and of premature infants fed with medium-chain triglycerides. J Inherit Metab Dis 2003;26:583-592.[CrossRef]
- dasNeves HJC, Noronha JP, Rufino H. New method for the chiral HRGC assay of L-2-hydroxyglutaric aciduria in urine. J High Res Chromatogr 1996;19:161-164.
- Kaunzinger A, Rechner A, Beck T, Mosandl A, Sewell AC, Bohles H. Chiral compounds as indicators of inherited metabolic disease. Simultaneous stereodifferentiation of lactic-, 2-hydroxyglutaric- and glyceric acid by enantioselective cGC. Enantiomer 1996;1:177-82.[Medline]
[Order article via Infotrieve]
- Rashed MS, AlAmoudi M, Aboul-Enein HY. Chiral liquid chromatography tandem mass spectrometry in the determination of the configuration of 2-hydroxyglutaric acid in urine. Biomed Chromatogr 2000;14:317-320.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
- Molven A, Matre GE, Duran M, Wanders RJ, Rishaug U, Njolstad PR, et al. Familial hyperinsulinemic hypoglycemia caused by a defect in the SCHAD enzyme of mitochondrial fatty acid oxidation. Diabetes 2004;53:221-227.[Abstract/Free Full Text]
- Brocks DR, Dennis MJ, Schaefer WH. A liquid chromatographic assay for the stereospecific quantitative analysis of halofantrine in human plasma. J Pharm Biomed Appl 1995;13:911-918.[CrossRef]
- Nicklassan M, Brjorkman S, Roth B, Jonsson M, Hoglund P. Stereoselective metabolism of pentoxifylline in vitro and vivo in humans. Chirality 2002;14:643-652.[Medline]
[Order article via Infotrieve]
- Brocks DR, Pasutto FM, Jamali F. Analytical and semi-preparative high-performance liquid chromatographic separation and assay of hydroxychloroquine enantiomers. J Chromatogr B 1992;581:83-92.[CrossRef]
- Lindner W, Rath M, Stoschitzky K, Uray G. Enantioselective drug monitoring of (R)-propanolol and (S)-propanolol in human plasma via derivatization with optically-active (R,R)-O,O-diacetyltartaric acid anhydride. J Chromatogr B 1989;487:375-383.[CrossRef]
Top
Abstract
Introduction
