|
|
||||||||
Articles |
1
Academic Medical Center, University of Amsterdam, Emma Childrens Hospital, and the Department of Clinical Chemistry, PO Box 22700, 1100 DE Amsterdam, The Netherlands.
2
Department of Pediatrics, Nagoya City University Medical
School, Kawasumi 1, Mizuho-cho, Mizuho-ku, Nagoya 467-8601, Japan.
a Author for correspondence. Fax 31-20-6962596;
| Abstract |
|---|
|
|
|---|
Methods: We used urine or urine-soaked filter-paper strips as samples and measured thymine, uracil, and their degradation products dihydrothymine, dihydrouracil, N-carbamyl-ß-aminoisobutyric acid, and N-carbamyl-ß-alanine. Reversed-phase HPLC was combined with electrospray ionization tandem mass spectrometry, and detection was performed by multiple-reaction monitoring. Stable-isotope-labeled reference compounds were used as internal standards.
Results: All pyrimidine degradation products could be measured in one analytical run of 15 min. Detection limits were 0.44 µmol/L. The intraassay imprecision (CV) of urine samples with added compounds was 1.312% for liquid urines and 1.010% for filter-paper extracts of the urines. The interassay imprecision (CV) was 311% (100200 µmol/L). Recoveries were 8999% at 100200 µmol/L and 95106% at 1 mmol/L in liquid urines, and 93103% at 100200 µmol/L and 100106% at 1 mmol/L in filter-paper samples. Correct identifications of deficiencies of the pyrimidine-degrading enzymes were readily made with urine samples from patients with known defects.
Conclusions: HPLC with electrospray ionization tandem mass spectrometry allows rapid testing for disorders of the pyrimidine degradation pathway, and filter-paper samples allow easy collection, transport, and storage of urine samples.
| Introduction |
|---|
|
|
|---|
The widely used anticancer drug 5-fluorouracil (5FU) is degraded by the
same pyrimidine degradation enzymes to dihydrofluorouracil,
-fluoro-ß-ureidopropionic acid, and
-fluoro-ß-alanine. A
complete or partial DPD deficiency can cause severe and, in some cases,
lethal 5FU-related toxicity (1)(2)(3). DHP-deficient patients
also may have a high risk of 5FU-toxicity (4).
Defects of the enzymes DPD and DHP are well known (5)(6)(7),
and recently, a case of ß-UP deficiency has been described
(8). On the basis of population analysis of the DPD
activity, it has been estimated that
3% of the general population
could be heterozygous, and
0.1% homozygous, for mutant DPD alleles
(9)(10). The prevalence of dihydropyrimidinuria
was estimated to be
1 per 10 000 births in Japan (4).
Several methods for the urinary screening of disorders of pyrimidine catabolism have been described (11)(12)(13)(14). These methods cannot detect all compounds in one analytical run. The increasing demand for measurement of pyrimidine breakdown products prompted us to develop a method that can measure all components in one analytical run and is more rapid than the methods that are currently available. We describe a method that screens for pyrimidine degradation defects, using urine samples or urine-soaked filter-paper strips, is simple and rapid, and uses reversed-phase HPLC coupled to electrospray ionization (ESI) tandem mass spectrometry (MS/MS).
| Materials and Methods |
|---|
|
|
|---|
The internal standards used were
,
,
,6-2H4-thymine;
1,3-15N2-uracil;
5,6,6-2H3-Me-2H3-DHT;
13C4,15N2-DHU;
2H6-N-C-ß-AIB; and
15N2,13C4-N-C-ß-Ala.
Stable-isotope-labeled thymine, uracil, DHT, and DHU were obtained from
Cambridge Isotope Laboratories, and we made stable-isotope-labeled
N-C-ß-AIB and N-C-ß-Ala by alkaline hydrolysis of the corresponding
stable-isotope-labeled dihydropyrimidines as described above.
internal standard mixture
We made stock solutions of the stable-isotope-labeled compounds,
and then made an aqueous mixture containing 1 mmol/L each of
stable-isotope-labeled thymine, uracil, DHT, N-C-ß-AIB, and
N-C-ß-Ala and 2 mmol/L DHU. This mixture was used as the internal
standard (IS) mixture.
liquid urine samples
Fresh urine samples were stored at 4 °C for analysis within 1
week, and others were stored at -20 °C until analysis. We
centrifuged the urines at 10 000g for 5 min to remove
debris and added 200 µL of clear urine to 20 µL of IS. Of this
urine, we injected 20 µL into the HPLC-MS/MS system. Urinary
creatinine concentrations were determined by the conventional
alkaline-creatinine-picrate method (16).
urine-soaked filter-paper strips
We dipped filter-paper strips (12 x 40 mm) completely into
urine and then removed the excess urine by wiping it off along the wall
of the test tube. The strips were dried completely at room temperature.
IS (20 µL) was pipetted onto the center of each strip, and after the
strips were completely dry, they were cut into small pieces and put
into a 2-mL Eppendorf tube. The pieces were extracted twice by
sonification for 10 min with 750 µL of
methanol-H2O mixture (75:25 by volume). Both
extracts were combined and dried at 40 °C under a stream of
nitrogen. The dried sample was dissolved in 200 µL of HPLC eluant A
(0.05 mol/L acetic acid, adjusted to pH 4.0 with 13 mol/L
NH4OH), and after centrifugation at
10 000g for 5 min, 20 µL of the clear extract was
injected into the HPLC-MS/MS. The remaining extract was used to measure
creatinine as previously described.
hplc-esi ms/ms
The HPLC system consisted of an HP 1100 series binary gradient
pump, a vacuum degasser, and a column temperature controller (all from
Hewlett Packard) and was connected to a Gilson 231 XL autosampler
(Gilson). The Phenomenex Aqua analytical column (250 x 4.6 mm;
particle size, 5 µm; Phenomenex) was protected by a guard column
(SecurityGuard C18 ODS; 4 x 3.0 mm; Phenomenex).
The column temperature was maintained at 23 °C. The mobile phases
were as follows: 0.05 mol/L acetic acid, adjusted to pH 4.0 with 13
mol/L NH4OH (eluant A); and 0.05 mol/L acetic acid (pH
4):methanol (1:1 by volume; eluant B). The elution gradient was as
follows (flow rate, 1 mL/min): 07 min, 100% A to 30% A; 77.1 min,
30% A to 0% A; 7.18 min, 100% B; 88.1 min, 100% B to 100% A;
and 8.115 min, equilibration with 100% A. All gradient steps were
linear, and the total analysis time, including equilibration, was 15
min. A splitter between the HPLC column and the mass spectrometer was
used, and 1020 µL/min of eluant was introduced into the mass
spectrometer. An electrically operated valve was used so that only the
eluant from 4.1 to 9 min was introduced into the mass spectrometer
(preventing early-eluting salts and late-eluting peaks from
contaminating the mass spectrometer).
A Quattro II tandem mass spectrometer (Micromass) was used in the
positive ESI mode. Nitrogen was used as the nebulizing gas. The
collision gas was argon, and the cell pressure was 0.25 Pa. The source
temperature was set at 80 °C, and the capillary voltage was
maintained at 3.5 kV. The detector was used in MS/MS mode using
multiple-reaction monitoring to detect a specific transition of
precursor ion to fragment for each analyte. The transition, cone
voltage, and collision energy established for each compound are listed
in Table 1
.
|
validation
The mass spectrometer settings were optimized using a 100 µmol/L
solution of each analyte prepared in 0.05 mol/L ammonium acetate buffer
(adjusted to pH 5 with glacial acetic acid). The solution was
introduced at 10 µL/min directly into the mass spectrometer.
The linearity and detection limits for each compound were established by injection of calibration mixtures with different concentrations. The stable-isotope-labeled compound of each analyte was used as IS. Analyte concentrations were determined using the slope and intercept of the calibration curve, which were obtained from a linear least-squares regression for the analyte/IS peak-area ratio vs the concentration of the calibrator.
The efficiency of the filter-paper extraction was obtained by pipetting 200 µL of a calibration mixture (100200 µmol/L) onto the filter paper and treating it as previously described for the urine-soaked filter paper. The concentrations in this filter-paper extract were calculated by the external standard method. The extraction efficiency of creatinine was obtained by comparing the creatinine concentrations in urine samples and in the filter-paper extracts of the same urines (10 different urines with creatinine concentrations of 0.817.1 mmol/L were used).
The intraassay (within-day) variation (CV) of the method was established by measuring 10 times a urine and a urine enriched with synthetic compounds at low (510 µmol/L), medium (100200 µmol/L), and high (1 mmol/L) concentrations. The interassay (between-day) variation was established by measuring blank urines and urines enriched with synthetic compounds (100200 µmol/L) during 5 separate weeks. The recovery of the method was established by measuring 13 different urines (creatinine concentrations of 2.314.6 mmol/L) before and after enrichment with known concentrations of synthetic compounds (510 µmol/L, 100200 µmol/L, and 1 mmol/L). The interassay variation and recovery of the filter-paper method were established with the same blank and enriched urines used for direct injection into the HPLC-MS/MS system.
IS mixtures were added to compensate for losses in the preparation of samples and to compensate for losses in sensitivity because of quenching of the signal by coeluting components. In all experiments, no special precautions (e.g., cleaning of the high-voltage lens and sample cone) were taken to optimize the detection limit of the mass spectrometer system for this method.
Urines from patients with DPD, DHP, or ß-UP deficiencies were analyzed to establish the usefulness of this method.
| Results |
|---|
|
|
|---|
|
When urine samples with a higher pH (>7) were used for direct injection, the peak of N-C-ß-AIB (and stable-isotope-labeled N-C-ß-AIB) was sometimes broad or split into two peaks. This phenomenon was a result of the low buffering capacity of the eluant and was overcome by acidification of the urine (to pH 4) before injection. In practice, the pH of each urine is checked and acidified if necessary.
Several urines that gave interference because of medication when analyzed by the conventional amino acid procedure used in our laboratory (13) showed no interfering peaks in the HPLC-ESI MS/MS method.
linearity and detection limits
The calibration curves for each compound were linear up to at
least 1 mmol/L (r2 >0.996). Above
this concentration, a deviation from linearity was observed for most
compounds. When the sample amount was not limited and concentrations
were above the linear part of the calibration curve, samples were
measured again after dilution to concentrations within the linear part
of the curve. When the sample amount was limited, concentrations were
calculated by interpolation on a quadratic curve (concentrations up to
5 mmol/L).
The detection limits were established and defined as the lowest signals with a signal-to-noise ratio of 3. The detection limits were as follows: thymine, 0.4 µmol/L; uracil, 0.8 µmol/L; DHT, 0.6 µmol/L; DHU, 4 µmol/L; N-C-ß-AIB, 0.5; and N-C-ß-Ala, 4 µmol/L. It should be noted that the detection limit was dependent on the status of the mass spectrometer (e.g., contamination of the high-voltage lens and cone of the mass spectrometer may have influenced the detection limit and noise) and on the presence of coeluting compounds in urine samples quenching the signal. In several measurement series, we observed a variation in detection limit and noise resulting in detection limits ranging from 0.5 to 2 times the previously presented values.
recovery and accuracy
The extraction efficiencies for the various compounds of the
calibration mixture from filter-paper strips were 80107% with CVs of
412% (n = 4). The extraction efficiency for creatinine was 87%
with a CV of 6% (n = 10).
The intraassay (within-day) results are summarized in Table 2
. The intraassay CV was 3.218% for nonenriched urines,
3.612% for urines enriched with low concentrations, 2.111% for
urines enriched with medium concentrations, and 1.36.3% for urines
enriched with high concentrations. In filter-paper extracts, the
intraassay CV was 2.528% for extracts of nonenriched urines,
2.17.9% for extracts of urines enriched with low concentrations,
1.010% for extracts of urines enriched with medium concentrations,
and 2.27.8% for extracts of urines enriched with high
concentrations.
|
The interassay (between-day) results are summarized in Table 3
. The interassay CV was 9.6123% for nonenriched urines and
2.711% for enriched urines.
|
The recovery data are summarized in Table 4
. The recoveries were 8637% with a CV of 5.875% for urines
enriched with 510 µmol/L and 80119% (CV, 5.133%) for
filter-paper extracts of these low-enriched urines. In the
medium-enriched (100200 µmol/L) and high-enriched (1 mmol/L) urine
samples and the corresponding filter-paper extracts, the recoveries and
CVs were much better.
|
Chromatograms of urines of confirmed DPD-, DHP-, and ß-UP-deficient
patients are shown in Fig. 2
, A, B, and C, respectively. The profiles of a control urine are
given for comparison (notice the difference in scaling for patients and
control). The high concentrations of pyrimidine degradation compounds
in patients make diagnosis very easy.
|
| Discussion |
|---|
|
|
|---|
Reversed-phase HPLC with ultraviolet detection (11) and gas chromatographymass spectrometry (12) are currently the most widely used methods to measure the pyrimidine bases in various body fluids. For general screening of high-risk patients for purine and pyrimidine disorders, an HPLC-ESI MS/MS method has already been developed in our laboratory (17). Using this method, we can measure thymine and uracil very well; however, the dihydropyrimidines and the N-carbamyl compounds cannot be measured without affecting the measurement of other important metabolites of the purine and pyrimidine pathway. Therefore, until now, dihydropyrimidines and N-carbamyl compounds have been measured in our laboratory by amino acid analysis after prefractionation of urines to separate the dihydropyrimidines from the N-carbamyl amino acids and subsequent hydrolysis of the components in these fractions to their corresponding amino acids (13).
The method used to measure DHT, DHU, N-C-ß-AIB, and N-C-ß-Ala is time-consuming and not entirely specific because dihydrouridine is hydrolyzed to ß-alanine as well. This prompted us to develop a single method that is rapid and selective for all compounds of the pyrimidine degradation pathway.
The use of HPLC coupled with ESI MS/MS enables analysis of all compounds in one run with a very short analysis time compared with conventional methods. The Phenomenex Aqua HPLC column was chosen because polar compounds are retained better on this column compared with standard C18 columns. This is especially true for N-C-ß-Ala, which elutes near the salt fraction on standard C18 columns. The use of stable-isotope-labeled internal standards with identical chromatographic behavior enables correction for quenching of the signal by coeluting compounds.
The reproducibility and accuracy of the new method are adequate for the
screening of disorders of pyrimidine degradation. The highest CVs were
obtained for concentrations at or slightly above the detection limits
of the method and were minimally relevant because, diagnostically,
enzyme deficiencies produce very large increases of the relevant
metabolites. The detection limits can vary, depending on the status of
the mass spectrometer. This led to a large variation in the interassay
measurements for concentrations at or near the detection limits
mentioned. The concentrations below the limit of quantification
(signal-to-noise ratio of 10) were nevertheless assessed to get an
estimation of the precision at the low concentrations observed in
normal urines. At higher concentrations (100200 µmol/L), the
interassay variation (Table 3
) was low and comparable to the intraassay
variation (Table 2
). With respect to some the high CV values obtained
for the accuracy measurements in urines enriched with low
concentrations, it should be noted that sometimes a low concentration
(510 µmol/L) was added to a very high endogenous concentration of
the same compound. The precision data tended to be better for the
filter-paper extracts compared with undiluted urine, most likely
because matrix effects were reduced by the filter-paper extraction of
urine.
Urine samples are very suitable for the detection of pyrimidine breakdown defects because the metabolites accumulate in this body fluid. Collection of urine sometimes may be a problem, especially from neonates. The filter strips we used offer the advantage of easy collection because they can be placed in the babys diaper (18)(19). In addition, transportation of dried filter paper is very easy and can be done at a low cost. Our results with urine-soaked filter paper are comparable to those determined with liquid urine.
The high concentrations of metabolites that accumulate in patients with an enzyme defect of pyrimidine degradation compared with controls make identification of such patients easy. This was clearly illustrated by the application of the method to urine samples from patients with confirmed pyrimidine degradation defects.
In conclusion, the HPLC-ESI MS/MS analysis of urine allows rapid screening for disorders of the pyrimidine degradation pathway. Liquid urine or urine-soaked filter-paper strips can be used for this purpose. Filter-paper strips offer the advantage of easy collection, transport, and storage of urine samples.
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
The following articles in journals at HighWire Press have cited this article:
![]() |
A. B.P. van Kuilenburg, H. van Lenthe, L. Zoetekouw, and W. Kulik HPLC-Electrospray Tandem Mass Spectrometry for Rapid Determination of Dihydropyrimidine Dehydrogenase Activity Clin. Chem., March 1, 2007; 53(3): 528 - 530. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Hartmann, J. G. Okun, C. Schmidt, C.-D. Langhans, S. F. Garbade, P. Burgard, D. Haas, J. O. Sass, W. L. Nyhan, and G. F. Hoffmann Comprehensive Detection of Disorders of Purine and Pyrimidine Metabolism by HPLC with Electrospray Ionization Tandem Mass Spectrometry Clin. Chem., June 1, 2006; 52(6): 1127 - 1137. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. B.P. van Kuilenburg, R. Meinsma, E. Beke, B. Assmann, A. Ribes, I. Lorente, R. Busch, E. Mayatepek, N. G.G.M. Abeling, A. van Cruchten, et al. {beta}-Ureidopropionase deficiency: an inborn error of pyrimidine degradation associated with neurological abnormalities Hum. Mol. Genet., November 15, 2004; 13(22): 2793 - 2801. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. B.P. van Kuilenburg, H. van Lenthe, A. van Cruchten, and W. Kulik Quantification of 5,6-Dihydrouracil by HPLC-Electrospray Tandem Mass Spectrometry Clin. Chem., January 1, 2004; 50(1): 236 - 238. [Full Text] [PDF] |
||||
![]() |
A. B. P. van Kuilenburg, R. Meinsma, B. A. Zonnenberg, L. Zoetekouw, F. Baas, K. Matsuda, N. Tamaki, and A. H. van Gennip Dihydropyrimidinase Deficiency and Severe 5-Fluorouracil Toxicity Clin. Cancer Res., October 1, 2003; 9(12): 4363 - 4367. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. J. Pitt, M. Eggington, and S. G. Kahler Comprehensive Screening of Urine Samples for Inborn Errors of Metabolism by Electrospray Tandem Mass Spectrometry Clin. Chem., November 1, 2002; 48(11): 1970 - 1980. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |