Syntheses of labeled vitamers of folic acid to be used as internal standards in stable isotope dilution assays

Article abstract of DOI:10.1021/jf025571k

[²H₄]Folic acid was synthesized by deuterating p-aminobenzoic acid, which was then coupled to glutamic acid and 6-formylpterin, Using [²H₄]folic acid as starting component enabled the preparation of labeled vitamers tetrahydrofolate, 5-formyltetrahydrofolate, 5-methyltetrahydrofolate, and 10-formylfolate which were characterized by electrospray mass spectrometry and collision-induced dissociation. The mass spectrometric studies confirmed that the compounds could be used as internal standards in stable isotope dilution assays.

Full text of DOI:10.1021/jf025571k

4760 J. Agric. Food Chem. 2002, 50, 4760−4768
Syntheses of Labeled Vitamers of Folic Acid to Be Used as
Internal Standards in Stable Isotope Dilution Assays
ACHIM FREISLEBEN, PETER SCHIEBERLE, AND MICHAEL RYCHLIK*
Institut fu¨r Lebensmittelchemie der Technischen Universita¨t Mu¨nchen,
Lichtenbergstrasse 4, D-85748 Garching, Germany
[2H4]Folic acid was synthesized by deuterating p-aminobenzoic acid, which was then coupled to
glutamic acid and 6-formylpterin. Using [2H4]folic acid as starting component enabled the preparation
of labeled vitamers tetrahydrofolate, 5-formyltetrahydrofolate, 5-methyltetrahydrofolate, and 10-
formylfolate which were characterized by electrospray mass spectrometry and collision-induced
dissociation. The mass spectrometric studies confirmed that the compounds could be used as internal
standards in stable isotope dilution assays.
KEYWORDS: Electrospray mass spectrometry; folates; labeled folic acid; stable isotope dilution assay
INTRODUCTION
METHODS AND MATERIALS
The folate group comprises several derivatives of folic acid
differing by their one-carbon substituents, their oxidation states,
and by the number of glutamate residues attached. The major
problem in quantifying this group of B vitamins lies in the need
to distinguish between the single derivatives, called vitamers,
as they show different bioavailability and vitamin activity (1).
Up to now, only the recently developed high-performance liquid
chromatographic (HPLC) approaches (2, 3) enabled the quan-
titation of the different folates, although these methods still lack
accuracy. Because cleanup procedures include ion exchange
chromatography (4) as well as several enzyme treatments (5),
and because folates show different stabilities (6), recovery values
for the vitamers differ widely (2, 7).
In recent years we reported on the analysis of flavor
compounds (8), of the mycotoxin patulin (9), and of the vitamin
pantothenic acid (10), and demonstrated that stable isotope
dilution assays (SIDA) exhibit excellent sensitivity and reliability
and are accurate alternatives to other quantification methods.
Using labeled analogues of the vitamers as internal standards
enables losses of each analyte during cleanup to be precisely
corrected.
Until recently, the use of stable isotopically labeled folates
had been limited either to bioavailability studies (11) or to
analyses including cleavage of folates prior to mass spectrometry
(12, 13). Hence, the information about concentrations of single
vitamers was lost. The first study using a labeled folate and
detecting it without destruction was reported by Pawlosky and
co-workers who quantified folic acid in fortified foods (14) and
5-methyltetrahydrofolate in blood serum (15) by LC-MS
without analyzing other folates. To quantify the naturally
occurring vitamers, we decided to synthesize the isotopomers
of important derivatives besides labeled folic acid.
Chemicals. The following chemicals were obtained commercially
from the sources given in parentheses: deuterium oxide, dicyclohexyl-
carbodiimide, ethyl-3-(dimethylaminopropyl)carbodiimide, glutamic
acid dimethylester, platinum oxide, [2H6]dimethyl sulfoxide ([2H6]DMSO),
trifluoroacetic anhydride (Aldrich, Steinheim, Germany), tris(hydroxy-
methyl)-aminomethane (TRIS; Merck, Darmstadt, Germany), 2-amino-
3-cyano-5-chlormethylpyrazine-4-oxide, dimethylaminoborane, N,N-
dimethylaminonitrosobenzene, 1-hydroxybenzotriazole (Acros Organics,
Geel, Belgium), guanidine hydochloride, phosphorus trichloride, pyri-
dine (Merck, Darmstadt, Germany), Dowex 50-WX8 ion-exchange resin
(hydrogen form), folic acid (Fluka, Neu-Ulm, Germany), 4-amino-
benzoic acid (Sigma, Deisenhofen, Germany). Unlabeled folate vitamers
tetrahydrofolate (H4folate), 5-formylH4folate, 5-methylH4folate, and 10-
formylfolate were purchased from Dr. Schirks Laboratories, Jona,
Switzerland.
Liquid Chromatography/Mass Spectrometry (LC-MS). LC-MS
spectra were recorded by means of a LCQ Classic (Finnigan MAT,
Bremen, Germany) coupled to a spectra series high-performance liquid
chromatograph (Thermo Separation Products, San Jose, CA). To verify
the syntheses and to determine the degree of labeling, the LC-MS
system was equipped with a 250 × 4.6 mm i. d., 5 µm, Discovery C18
column (Supelco, Bellefonte, PA). Aliquots of 1 to 10 µL of the sample
solutions were chromatographed using gradient elution and a flow rate
of 0.8 mL/min. A linear gradient was programmed within 9 min, starting
with acetonitrile/1 mmol/L aqueous formic acid (7:93, v/v) to aceto-
nitrile/1 mmol/L aqueous formic acid (20:80, v/v). Then the column
was flushed with acetonitrile/1 mmol/L aqueous formic acid (80:20,
v/v) for 4 min.
For analysis of folate vitamer mixtures, the LC-MS system was
coupled to a 250 × 4.6 mm i. d.; 5 µm, Aqua C-18 reversed phase
column (Phenomenex, Aschaffenburg, Germany). When using the latter
column, the mobile phase consisted of variable mixtures of aqueous
formic acid (0.1%) and acetonitrile, at a flow of 0.8 mL/min. Gradient
elution started at 7% acetonitrile maintained for 9 min, followed by
raising the acetonitrile concentration linearly to 13% within 13 min,
and to 25% within further 4 min. Subsequently, the mobile phase was
programmed to 100% acetonitrile over 4 min before equilibrating the
column for 5 min with the initial mixture.
* To whom correspondence should be addressed. Phone +49-89-289 132
55. Fax +49-89-289 141 83. E-mail michael.rychlik@ch.tum.de.
10.1021/jf025571k CCC: $22.00 © 2002 American Chemical Society
Published on Web 07/23/2002

Syntheses of Labeled Folates
J. Agric. Food Chem., Vol. 50, No. 17, 2002 4761
acid (0.1 mol/L) to give a final pH of 8. Freeze-drying of the solution
afforded 5 (778 mg, 2.89 mmol, 76%).
Table 1. Base Peaks Produced by Collision-Induced Dissociation
(CID) of the Quasimolecular Ions of Folate Vitamers in Positive
Electrospray Mass Spectrometry
Pteridine Precursor. 2-Amino-3-cyano-5-chlormethylpyrazine (7).
To a solution of 2-amino-3-cyano-5-chlormethylpyrazine-4-oxide 6
(13.02 g, 70.56 mmol) in tetrahydrofuran (500 mL), phosphorus
trichloride (27 g, 146.27 mmol, 17.15 mL) was added dropwise with
ice-bath cooling. The solution was stirred for 45 min at room
temperature and then concentrated to about 60 mL. After adding ice
water (300 mL) the precipitating solid 7 (10.31 g, 61.16 mmol, 86.7%)
was collected by centrifugation and washed thoroughly with water.
1[(2-Amino-3-cyano-5-pyrazinyl)methyl]pyridinium Chloride (8).
Compound 7 (10.31 g) was added to a flask containing pyridine (110
mL). Some undissolved substances were removed by filtration, and
the solution was stirred for 17.5 h at room temperature. After adding
diethyl ether (800 mL) the salt 8 precipitated and was collected by
centrifugation. Washing with diethyl ether twice gave pure 8 (12.04 g,
48.65 mmol, 79.5%).
product ion
precursor ion
at m/z
vitamer
at m/z
(rel. collision energy)
folic acid
442
446
446
450
460
464
474
478
470
474
295 (26%)
299 (26%)
2
[ H ]folic acid
4
tetrahydrofolic acid
299 (23.8%)
303 (23.8%)
313 (23.8%)
317 (23.8%)
327 (23%)
331 (23%)
452 (24.5%)
456 (24.5%)
2
[ H ]tetrahydrofolate
4
5-methyltetrahydrofolate
2
[ H ]5-methyltetrahydrofolate
4
5-formyltetrahydrofolate
2
[ H ]5-formyltetrahydrofolate
4
10-formylfolate
2
[ H ]formylfolate
4
N-[p-(Dimethylamino)phenyl]-R-(2-amino-3-cyano-5-pyrazinyl)-
nitrone (9). The pyridinium salt 8 (12.04 g) was suspended in an
ethanolic solution of N,N-dimethylaminonitrosobenzene (7.30 g, 48.60
mmol, 250 mL), and an aqueous solution of potassium carbonate (40.3
g, 291.6 mmol, 150 mL) was added. The solution was stirred for 30
min at room temperature followed by ice bath cooling. The mixture
was centrifuged and the resulting nitrone 9 (10.42 g, 36.82 mmol,
75.7%) was washed twice with water, then washed with ethanol, and
finally washed with diethyl ether.
2-Amino-3-cyano-5-formylpyrazine (10). A solution of nitrone 9
(10.42 g, 36.82 mmol) in at least 650 mL of cold hydrochloric acid (6
mol/L) was extracted with ethyl acetate (500 mL). The organic layer
was separated, and the aqueous layer was twice extracted with ethyl
acetate. Sodium chloride (70 g) was added to the aqueous layer, which
was once again extracted with ethyl acetate. The organic layers were
collected, dried over anhydrous sodium sulfate, and evaporated to give
10 (3.279 g, 22.15 mmol, 60.2%).
2-Amino-3-cyano-5-formylpyrazine Dimethyl Acetal (11). To a
solution of formylpyrazine 10 (3.279 g, 22.15 mmol) in dry methanol
(70 mL) Dowex 50-WX8 ion-exchange resin (hydrogen form, 7 g) was
added. After the solution was stirred for 1 h, a few grams of 3 Å
molecular sieves were added and the mixture was shaken briefly. The
molecular sieves were filtered off and the solution was processed further
without isolating the product.
2,4-Diamino-6-formylpteridine Dimethyl Acetal (12). The methanolic
solution of 11 was mixed with a solution of guanidine hydochloride
(2.9 g, 30.36 mmol) in a methanolic solution of sodium methylate (100
mL, 0.2 mol/L) and refluxed for 18 h at 70 °C. After the solution was
cooled to room temperature, it was concentrated to a small volume
and cooled to -30 °C. The resulting product (4.52 g, 19.15 mmol,
86.5%) was collected by centrifugation.
6-Formylpterin Dimethyl Acetal (13). 2,4-Diamino-6-formylpterine
dimethyl acetal 12 (4.52 g, 19.15 mmol) was dissolved in aqueous
sodium hydroxide (180 mL, 5%) and heated gently under reflux for
10 min. The solution was filtered through a sintered glass funnel and
neutralized with glacial acetic acid. The yellow product precipitated
after storing the solution overnight at 4 °C and was collected by
centrifugation. Washing twice with water, ethanol, and diethyl ether
afforded the pure title product (4.183 g, 17.65 mmol, 92.1%).
6-Formylpterin (14). A mixture of 6-formylpterin dimethyl acetal
13 (4.1828 g, 17.649 mmol), formic acid (80 mL, 97-100%), and water
(8 mL) was stirred slowly for 30 min at room temperature, poured
into water (50 mL), and neutralized with 25% ammonium hydroxide.
The yellow precipitate was collected by centrifugation and washed twice
with water, ethanol, and diethyl ether to yield 14 (3.879 g, 17.70 mmol,
100%).
The mass spectrometer was operated in the positive electrospray
mode with a spray needle voltage of +3.55 kV and a spray current of
0.5 µA. The temperature of the capillary was 200 °C and the capillary
voltage was +29.39 V. The sheath and auxiliary gas nitrogen nebulized
the effluent with flows of 80 and 20 arbitrary units, respectively. The
ion trap was operated at a helium pressure of 10-3 Torr. Collision-
induced dissociation of folate vitamers was performed by applying to
their quasimolecular ions the relative collision energies detailed in
Table 1.
Nuclear Magnetic Resonance Spectroscopy (NMR). 1H NMR
spectra were recorded with an AMX-400 III (Bruker, Karlsruhe,
Germany) at a frequency of 400.13 MHz. Shifts are expressed in ppm
downfield from tetramethylsilane in [2H6]DMSO.
Anion Exchange Chromatography. Anion exchange cellulose DE-
52 (Whatman, 23 g) was suspended in an aqueous solution of
ammonium formate (0.15 mol/L, 150 mL, adjusted to pH 7) in a
chromatography column (2 × 25 cm). The sample to be chromato-
graphed was dissolved in an aqueous solution of ammonium formiate
(0.15 mol/L, 5 mL), adjusted to pH 7 by an aqueous solution of sodium
hydroxide (0.1 mol/L) and applied on the column. Elution was
performed by a 2-h gradient of ammonium formate starting with 0.15
mol/L to 2 mol/L at a flow of 1.5 mL/min. The eluate was monitored
by a Uvicord detector (LKB, Uppsala, Sweden) set at 280 nm.
Syntheses: Aminobenzoyl Glutamate Precursor. [2H4]4-Amino-
benzoic Acid (2). 4-Aminobenzoic acid 1 (potassium salt; 3.0503 g,
17.41 mmol) was suspended with palladium on activated charcoal (1.31
g, 393 mg Pd) in deuterium oxide (12 mL) and heated at 200 °C in an
autoclave for 2 h. The autoclave was cooled with cold water and the
suspension was filtered and washed repeatedly with redistilled water.
The solution was then lyophilized and the above process was repeated
twice to give pure 2 (1.812 g, 10.123 mmol, 58.1%).
Positive ESI-MS: m/z (%) ) 142 (100), 183 (42), 141 (12), 143
(7).
Trifluoroacetyl-[2H4]4-aminobenzoic Acid (3). The resulting [2H4]4-
aminobenzoic acid (1.812 g) was suspended in cold trifluoroacetic
anhydride (30 mL) and stirred at room temperature for 2 h. The mixture
was then dried in a stream of nitrogen and directly used for the next
synthetic step.
Trifluoroacetyl-[2H4]4-aminobenzoylglutamate Dimethylester (4). 3
was dissolved by stirring in anhydrous tetrahydrofuran (150 mL), and
after addition of dicyclohexylcarbodiimide (DCC, 5 g, 24.23 mmol)
and 1-hydroxybenzotriazole (3.27 g, 24.23 mmol) the mixture was
stirred at room temperature for 30 min. Then, glutamic acid dimethyl-
ester (4.9 g, 23.16 mmol) and diisopropylethylamine (2.99 g, 23.16
mmol) were added, and the solution was stirred for a further 4 h at
room temperature and then suction filtered. The filtrate was evaporated
to dryness and provided the title compound (3.23 g, 8.199 mmol, 80%).
[2H4]p-Aminobenzoylglutamic Acid (5). Protected [2H4]4-amino-
benzoylglutamic acid (3.81 mmol) was dissolved in methanol (20 mL)
followed by addition of aqueous sodium hydroxide (100 mL, 0.1 mol/
L). Then, the solution was stirred for precisely 1 h 45 min at room
temperature before the reaction was quenched by adding hydrochloric
1H NMR: ([2H6]DMSO) δ ) 8.16 (1H, s, NH), 9.08 (1H, s, CH),
9.98 (1H, s, CHO).
Positive ESI-MS: m/z (%) ) 190 (100), 382 (45), 381 (16), 147
(1).
N2-Acetyl-6-formylpterin (15). 6-Formylpterin (1.000 g, 5.24 mmol)
was suspended in acetic anhydride (1000 mL) and heated under stirring
for 10 h at 100 °C. The reaction mixture became homogeneous during
the process, except for some unsoluble components which were

4762 J. Agric. Food Chem., Vol. 50, No. 17, 2002
Freisleben et al.
separated by filtration. The acetic anydride was evaporated, and the
residual 15 (0.847 g, 3.62 mmol, 69%) was resuspended in water and
lyophilized.
1H NMR: ([2H6]DMSO) δ ) 2.21 (3H, s, COCH3), 5.87 (1H, t,
OH), 9.04 (1H, s, NH), 9.27 (1H, s, CH), 10.08 (1H, s, CHO).
RESULTS AND DISCUSSION
Consideration of the Most Suitable Labelings and Selec-
tion of Vitamers to be Synthesized. A literature survey
indicated that different forms of folic acid labeled by stable
isotopes have been prepared in the past. In a prior attempt Plante
et al. (16) reported the synthesis of [carbonyl-13C1]folic acid
before Gregory and Toth prepared [3′,5′-2H2]folic acid (17) as
well as pteroyl[2H4]glutamic acid (18), and Dueker et al. (19)
generated [2′,3′,5′,6′-2H4]folic acid. More recently, Rogers and
co-workers (20) synthesized [13C5]folic acid by coupling
[13C5]glutamic acid with trifluoroacetyl pteroic acid.
Positive ESI-MS: m/z (%) ) 252 (100), 485 (62), 234 (55), 467
(15).
[2H4]Folic Acid. N2-Acetyl[2H4]folic Acid (16). To a flask containing
glacial acetic acid (30 mL) N2-acetyl-6-formylpterin (559 mg, 2.4 mmol)
and [2H4]4-aminobenzoylglutamate (700 mg, 2.6 mmol) were added.
After maintaining the solution for 10 min, dimethylaminoborane (153.4
mg, 2.6 mmol) in glacial acetic acid (10 mL) was added, and the
solution was heated for 20 min at 60 °C in an oil bath. The resulting
solution was evaporated and the residue was redissolved in aqueous
ammonium formate (0.15 mol/L) at pH 7. The solution was purified
by anion exchange chromatography as described above. The fractions
absorbing at 280 nm were verified to contain 16 using HPLC-MS,
then pooled and lyophilized to yield 415 mg (0.85 mmol, 35%) of 16.
[2H4]Folic Acid (17). Compound 16 (415 mg, 0.85 mmol) was
dissolved in aqueous sodium hydroxide (50 mL, 0.1 mol/L) and stirred
for 6 h at room temperature. The reaction was quenched by adjusting
the pH to 3.5 by addition of hydrochloric acid (1 mol/L). [2H4]Folic
acid (303 mg, 0.68 mmol, 80%) precipitated quantitatively after standing
overnight at 4 °C and was lyophilized after the solution was centrifuged
and the supernatant was discarded.
1H NMR: ([2H6]DMSO): 1.91 (2 H, m, â-C-2H); 2.32 (2 H, m,
γ-C-2H); 4.30 (1 H, t, R-C-1H); 4.47 (2 H, d, C9-2 H); 6.89 (1 H, t,
C4-OH); 7.99 (1 H, d, N8′-H); 8.64 (1 H, s, C7-H).
Positive ESI-MS: m/z (%) ) 446 (100), 447 (23), 468 (17), 445
(7).
Deuterated Vitamers of Folic Acid. [2H4]Tetrahydrofolic Acid. A
reaction vessel with a three-way gas inlet and a flowmeter counter was
filled with glacial acetic acid (25 mL) and platinum oxide (100 mg).
The vessel was flushed for 1 min with nitrogen. Upon addition of
[2H4]folic acid (250 mg, 0.562 mmol), the mixture was flushed
repeatedly with hydrogen gas and finally stirred under a slight positive
pressure of hydrogen for 4 h. At the end of the reaction, the catalyst
was filtered off, and the solution was lyophilized to give [2H4]tetra-
hydrofolic acid diacetate (197 mg, 0.348 mmol, 62%).
The anticipated determination by LC-MS, however, renders
only a few labelings promising. As quantification would be
based on measuring the quasimolecular ions (M + 1)+ of folic
acid vitamers which all contain at least 19 carbon, 7 nitrogen,
and 6 oxygen atoms, interferences from naturally occurring
isotopes in the analytes have to be taken into consideration.
Due to natural abundance of the isotopes C-13, N-15, and O-18,
an intensity of nearly 4% can be calculated for the quasi-
molecular ions of doubly labeled folates. Therefore, using a
labeling of only two mass units would result in mass spectro-
metric overlap between the unlabeled analytes and the labeled
internal standards. For this reason we considered a label
producing at least three mass increments to overcome this
hindrance. Regarding the syntheses mentioned above, only
[2′,3′,5′,6′-2H4]folic acid, pteroyl[2H4]glutamic acid or pteroyl-
[13C5]glutamic acid would be suitable. The latter preparations,
however, both require the use of labeled glutamic acid, which
is too expensive to give sufficient folic acid to be used as starting
component for the syntheses of further vitamers. Therefore, we
decided to synthesize [2′,3′,5′,6′-2H4]folic acid by introducing
a 4-fold deuterium label in the benzene ring of p-aminobenzoic
acid, which enabled us to obtain higher yields of labeled folic
acid.
Recent analyses of folates by high-performance liquid chro-
matography (HPLC) revealed tetrahydrofolate, 5-methyltetra-
hydrofolate, 5-formyltetrahydrofolate, 10-formylfolate, and folic
acid to predominate in different foods (7, 21) and 5-methyl-
tetrahydrofolate to be the most important vitamer in blood (22).
Therefore we chose these five vitamers (Figure 1) to be
synthesized starting from [2H4]folic acid.
Positive ESI-MS: m/z (%) ) 450 (100), 451 (22), 449 (4), 452 (2).
[2H4]5-Formyltetrahydrofolic Acid. [2H4]Tetrahydrofolic acid di-
acetate (80 mg, 0.142 mmol) was dissolved in formic acid (960 µL,
98-100%) and aqueous dipotassium hydrogenphosphate (10 mL, 50
mM containing 0.5% mercaptoethanol), and the solution was adjusted
to pH 3.5 with aqueous ammonium hydroxide (0.1 mol/L). Then,
1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (32 mg, 0.168 mmol)
was added, and the mixture was stirred for 15 min under an argon
atmosphere. Lyophilization gave the crude product, which was purified
by anion exchange chromatography yielding [2H4]5-formyltetrahydro-
folate (14 mg, 0.029 mmol, 18%) with a purity exceeding 98% (HPLC).
Synthesis of Labeled Folic Acid. The intended synthetic
approach to labeled folic acid required the coupling of labeled
p-aminobenzoylglutamic acid and 6-formylpterin.
The first component can be generated either by labeling
p-aminobenzoic acid in a catalytic protium-deuterium exchange
followed by coupling to glutamic acid, or by reacting deuterated
toluene to nitrobenzoylglutamate and subsequent reduction of
the nitro group. In a first series of experiments the latter reaction
sequence, according to Dueker et al. (19), was pursued as it
proposed higher yields. The synthesis involved nitration of
[2H8]toluene, followed by oxidation of the methyl group and
formation of the chloride. The subsequent coupling with
glutamyl diethylester and reduction to p-aminobenzoylglutamic
acid gave a reaction mixture in which purification of the target
compound from a host of byproducts could not be accomplished.
Therefore, we chose the former alternative starting with deu-
teration of p-aminobenzoic acid as depicted in Figure 2.
Comparing the mass spectra of unlabeled and labeled p-
aminobenzoic acid (Figure 3) the incorporation of the four
deuterium atoms can be clearly seen from the mass shift of the
quasimolecular ion from m/z 138 to m/z 142. The isotopic peak
areas of p-aminobenzoic acid indicate a 95% incorporation of
Positive ESI-MS: m/z (%) ) 478 (100), 479 (20), 477 (4), 480 (3).
[2H4]5-Methyltetrahydrofolic Acid. [2H4]Tetrahydrofolic acid (80 mg,
0.142 mmol) was dissolved in an aqueous solution of TRIS at pH 8
(25 mL) and stirred for 15 min at room temperature upon the addition
of aqueous formaldehyde (150 µL, 37%). Sodium borohydride (800
mg, 21 mmol) was added and the solution was heated at 50 °C for 1
h. The excess of reducing agent was destroyed by adding acetic acid
(5 mol/L) dropwise. Purification by anion exchange chromatography
and lyophilization of the pooled eluates provided the title compound
(22 mg, 0.048 mmol, 34%).
Positive ESI-MS: m/z (%) ) 464 (100), 465 (19), 317 (7), 466 (3).
[2H4]10-Formylfolic Acid. [2H4]Folic acid (30 mg, 0.067 mmol) was
dissolved in 50 mL of concentrated formic acid and heated for 3 h at
60 °C. Formic acid was evaporated, and the residue was redissolved
in 1 mL of aqueous formic acid (50%). Purification by high-
performance liquid chromatography provided [2H4]10-formylfolic acid
with a purity exceeding 97% (5.1 mg, 0.0107 mmol, 16%).
Positive ESI-MS: m/z (%) ) 474 (100), 475 (27), 473 (10), 476
(9).

Syntheses of Labeled Folates
J. Agric. Food Chem., Vol. 50, No. 17, 2002 4763
Figure 1. Structures of labeled folates.
2
Figure 2. Reaction scheme leading to [ H ]4-aminobenzoyl glutamic acid (5); TFAA, trifluoroacetic anhydride; DCC, dicyclohexylcarbodiimide; HOBT,
4
1-hydroxybenzotriazole.
four deuterium atoms and a 5% incorporation of three deuterium
atoms (HPLC-MS chromatograms not shown).
phosphorus trichloride gave the pyrazine 7, which was coupled
with pyridine to 8 in a nucleophilic substitution of the side-
chain chloride. The pyridine moiety was subsequently substi-
tuted by N,N-dimethylaminonitrosobenzene to provide the
nitrone 9. Acidic hydrolysis of 9 gave rise to formylpyrazine
10 being then protected to acetal 11 by reaction with methanol.
Coupling 11 with guanidine then provided the second ring of
pterin, and the amino group at C-4 was substituted by a hydroxyl
in a reaction of 12 with aqueous sodium hydroxide. Acidic
Trifluoroacetylation of labeled 2 enabled the coupling with
glutamic acid dimethylester in a reaction similar to that reported
by Vinale et al. (23). Subsequent hydrolysis of the protecting
groups yielded labeled p-aminobenzoylglutamic acid 5.
The second substructure, 6-formylpterin, was generated in
an eight-step procedure (Figure 4) according to Taylor et al.
(24) starting from pyrazine oxide 6. Reduction of 6 by

4764 J. Agric. Food Chem., Vol. 50, No. 17, 2002
Freisleben et al.
2
Figure 3. ESI-mass spetra of unlabeled (above) and [ H ]4-aminobenzoic acid (below).
4
1
hydrolysis of the acetal moiety converted 13 to 6-formylpterin
14 in an overall yield of 25.0%. The H NMR spectrum of 14
To verify the positions of the deuterium label a H NMR
spectrum of 17 was recorded and compared to that of unlabeled
folic acid. As the spectrum of [2H4]folic acid lacks the four
signals at 6.63, 6.65, 7.63, and 7.66 ppm present in the spectrum
of unlabeled folic acid, and these signals represent the reso-
nances of the four benzene-ring protons, their absence is
consistent with the incorporation of the deuterium atoms therein.
Conversely to [2H4]folic acid synthesized in the present study,
the analoguous compound generated by Dueker et al. (19)
1
was identical with that reported by Taylor et al. (24).
6-Formylpterin 14 then had to be protected by acetylation,
and 15 could subsequently be reacted with labeled p-amino-
benzoylglutamic acid 5 to the intermediate imine which was
not isolated. The reduction of the imine to 16 was achieved
with dimethylaminoborane (DMAB) in a sequence according
to Maunder et al. (25). Using DMAB as reducing agent is
advantageous, as further reduction of the pterine ring system
does not occur. As detailed in Figure 5, subsequent removing
of the acetic function afforded [2H4]folic acid 17 in a total yield
of 26% starting from 5.
1
revealed in its H NMR spectrum discernible signals at the
positions of the benzene-ring protons. On this basis the authors
calculated a 17% residue of protium in these positions of their
labeled folic acid, thus indicating that a substantial protium-

Syntheses of Labeled Folates
J. Agric. Food Chem., Vol. 50, No. 17, 2002 4765
2
Figure 4. Reaction scheme leading to N -acetyl-6-formylpterin (15); DMANB, N,N-dimethylaminonitrosobenzene.
2
Figure 5. Reaction scheme leading to [ H ]folic acid; DMAB, dimethylaminoborane.
4
deuterium exchange had occurred. By contrast, the labeled folic
acid prepared in the present study proved to contain only 5%
protium in the benzene moiety. Therefore, this compound is
better distinguishable from the unlabeled isotopomer by mass
spectrometry and is a more suitable internal standard in stable
isotope dilution assays (SIDAs).
sodium borohydride to [2H4]5-methyltetrahydrofolate, which
was purified by anion exchange chromatography (26). The
intermediate 5,10-methylenetetrahydrofolic acid was not iso-
lated.
Synthesis of [2H4]5-Formyltetrahydrofolic Acid. Labeled
5-formyltetrahydrofolate was prepared according to a procedure
reported by Moran and Colman (27). By coupling formic acid
with [2H4]tetrahydrofolate in the presence of 1-ethyl-3-(di-
methylaminopropyl)carbodiimide (EDC), [2H4]5-formyltetra-
hydrofolate was formed and purified again by anion exchange
chromatography.
Synthesis of [2H4]Tetrahydrofolic and [2H4]5-Methyl-
tetrahydrofolic Acid. By hydrogenation of labeled folic acid
in the presence of platinum oxide, [2H4]tetrahydrofolate was
prepared according to a procedure reported by Scott (26).
[2H4]tetrahydrofolate was then reacted with formaldehyde and

4766 J. Agric. Food Chem., Vol. 50, No. 17, 2002
Freisleben et al.
2
Figure 6. ESI-mass spectra of undeuterated (above) and [ H ]folic acid (below).
4
Synthesis of [2H4]10-Formylfolic Acid. [2H4]Folic acid was
formylated by heating in concentrated formic acid according
to the procedure proposed by Huennekens et al. (28). As the
resolving efficiency of the anion exchange column was not
sufficient in this case, purification was performed by preparative
HPLC.
Mass Spectrometric Studies of Folate Vitamers. The
synthesized vitamers revealed upon positive electrospray ioniza-
tion an abundant quasimolecular ion (M + 1)+, being the base
peak in each spectrum. As electrospray ionization is very soft
compared to other methods (e.g., atmospheric pressure chemical
ionization), few fragmentations occurred. Those resulted mostly
from ejection of the glutamyl residue (M + 1 - 147)+.
Comparing the spectra with those of the unlabeled reference
compounds, a 4-fold labeling is evident for all vitamers, as the
representative spectra of labeled and unlabeled folic acid reveal
in Figure 6. The base peak at m/z 442 is shifted toward m/z
446 in [2H4]folic acid. On the other hand, the unlabeled
isotopomers show only negligible ion intensity from naturally
occurring isotopes falling at the m/z values for the respective
ions of the labeled forms. This behavior enables accurate mass
spectrometric differentiation of the respective isotopomers in
future studies. To enhance specificity of the detection, collision-
induced dissociation was applied to the quasimolecular ions and

Syntheses of Labeled Folates
J. Agric. Food Chem., Vol. 50, No. 17, 2002 4767
2
2
Figure 7. HPLC−MS−MS chromatogram of a standard mixture of [ H ]tetrahydrofolate (1), tetrahydrofolate (2), [ H ]5-methyl-tetrahydrofolate (3), 5-methyl-
4
4
2
2
2
tetrahydrofolate (4), [ H ]10-formyl-folate (5), 10-formyl-folate (6), [ H ]5-formyl-tetrahydrofolate (7), 5-formyl-tetrahydrofolate (8), [ H ]folic acid (9), and
4
4
4
folic acid (10) showing the mass ranges of labeled folate vitamers as well as those of the corresponding unlabeled analogues; SRM, selected reaction
monitoring.
LITERATURE CITED
gave the fragmentation shown in Table 1. The great majority
of fragments arise from expulsion of the glutamyl moiety, and
all fragments still contain the four deuterium labels.
High-Performance Liquid Chromatography-Double Stage
Mass Spectrometry. To prove that the synthesized folate
isotopomers could be used as internal standards (IS) in SIDAs,
we optimized the HPLC conditions to resolve all vitamers in a
single run. As is evident from the HPLC-MS-MS chromato-
gram shown in Figure 7, (1) separation of the folates is possible,
and (2) each isotopomer can be specifically detected in its MS-
MS trace.
High-performance liquid chromatography coupled to elec-
trospray mass spectrometry revealed the synthesized compounds
to be an excellent choice for use as IS. In the course of this
study we synthesized 5-30 mg of each vitamer. Because in
each SIDA approximately 1.5 µg of them serve as IS, the total
amount will be sufficient to perform 3000 to 20 000 assays.
On the other hand, Gregory et al. (11) stated that 200 µg of
folate are adequate to perform a bioavailability study. Thus,
the preparations likewise afforded enough amounts of the labeled
folates to apply them in examinations of the bioavailability of
the single vitamers.
(1) Tamura, T.; Stokstad, E. L. R. The availability of food folate in
man. Br. J. Haematol. 1973, 25, 513-532.
(2) Finglas, P. M.; Wigertz, K.; Vahteristo, L.; Witthoft, C.; Southon,
S.; deFroidmontGortz, I. Standardisation of HPLC techniques
for the determination of naturally occurring folates in food. Food
Chem. 1999, 64, 245-255.
(3) Bagley, P. J.; Selhub, J. Analysis of folate form distribution by
affinity followed by reversed-phase chromatography with elec-
trochemical detection. Clin. Chem. 2000, 46, 404-411.
(4) Gregory, J. F., III; Sartain, D. B.; Day, B. P. F. Fluorometric
determination of folacin in biological materials using high
performance liquid chromatography. J. Nutr. 1984, 114, 341-
353.
(5) Pfeiffer, C. M.; Rogers, L. M.; Gregory, J. F. Determination of
folate in cereal-grain food products using trienzyme extraction
and combined affinity and reversed-phase liquid chromatography.
J. Agric. Food Chem. 1997, 45, 407-413.
(6) O’Broin, J. D.; Temperley, I. J.; Brown, J. P.; Scott, J. M.
Nutritional stability of various naturally occurring mono-
glutamate derivatives of folic acid. Am. J. Clin. Nutr. 1975, 28,
438-444.

4768 J. Agric. Food Chem., Vol. 50, No. 17, 2002
Freisleben et al.
(7) Konings, E. J. M. A validated liquid chromatographic method
for determining folates in vegetables, milk powder, liver, and
flour. J. AOAC Int. 1999, 82, 119-127.
(8) Rychlik, M.; Warmke, R.; Grosch, W. Ripening of Emmental
cheese wrapped in foil with and without addition of lactobacillus
casei subsp. casei. III. Analysis of character impact flavour
compounds. Lebensm.-Wiss. Technol. 1997, 30, 471-478.
(9) Rychlik, M.; Schieberle, P. Quantification of the mycotoxin
patulin by a stable isotope dilution assay. J. Agric. Food Chem.
1999, 47, 3749-3755.
(10) Rychlik, M. Quantification of free and bound pantothenic acid
in foods and blood plasma by a stable isotope dilution assay. J.
Agric. Food Chem. 2000, 48, 1175-1181.
(11) Gregory, J. F., III; Bailey, L. B.; Toth, T. G. Stable-isotope
methods for assessment of folate bioavailability. Am. J. Clin.
Nutr. 1990, 51, 212-215.
(19) Dueker, S. R.; Jones, D. A.; Smith, G. M.; Clifford, A. J.
Preparation of [2′,3′,5′,6′-2H4]pteroylglutamic acid. J. Labelled
Compd. Radiopharm. 1995, 36, 981-991.
(20) Rogers, L. M.; Pfeiffer, C. M.; Bailey, L. B.; Gregory, J. F. A
dual-label stable-isotopic protocol is suitable for determination
of folate bioavailability in humans: Evaluation of urinary
excretion and plasma folate kinetics of intravenous and oral doses
of [13C(5)] and [2H(2)]folic acid. J. Nutr. 1997, 127, 2321-
2327.
(21) Konings, E. J. M. A validated liquid chromatographic method
for determining folates in vegetables, milk powder, liver, and
flour. J. AOAC Int. 1999, 82, 119-127.
(22) Wigertz, K.; Ja¨gerstad, M. Comparison of a HPLC and radio-
protein-binding assay for the determination of folates in milk
and blood samples. Food Chem. 1995, 54, 429-436.
(23) Vinale, F.; Fogliano, V.; Schieberle, P.; Hofmann, T. Develop-
ment of a stable isotope dilution assay for an accurate quantifica-
tion of protein-bound N-ꢀ-(1-deoxy-D-fructos-1-yl)-L-lysine using
a 13C-labeled internal standard. J. Agric. Food Chem. 1999, 47,
5084-5092.
(12) Toth, J. P.; Gregory, J. F. Analysis of folacin by GC/MS:
Structures and mass spectra of fluorinated derivatives of para-
aminobenzoyl glutamic acid. Biomed. EnViron. Mass Spectrom.
1988, 17, 73-79.
(13) Santosh-Kumar, C. R.; Deutsch, J. C.; Hassel, K. L.; Kolhouse,
N. M.; Kolhouse, J. F. Quantitation of red blood cell folates by
stable isotope dilution gas chromatography-mass spectrometry
utilizing a folate internal standard. Anal. Biochem. 1995, 225,
1-9.
(14) Pawlosky, R. J.; Flanagan, V. P. A quantitative stable-isotope
LC-MS method for the determination of folic acid in fortified
foods. J. Agric. Food Chem. 2001, 49, 1282-1286.
(15) Pawlosky, R. J.; Flanagan, V. P.; Pfeiffer, C. M. Determination
of 5-methyltetrahydrofolic acid in human serum by stable-isotope
dilution high-performance liquid chromatography-mass spec-
trometry. Anal. Biochem. 2001, 298, 299-305.
(16) Plante, L. T.; Williamson, K. L.; Pastore, E. J. Preparation of
folic acid specifically labeled with Carbon 13 in the benzoyl
carbonyl. Methods Enzymol. 1980, 66, 533-553.
(17) Gregory, J. F.; Toth, J. P. Preparation of folic acid specifically
labeled with deuterium at the 3′,5′-positions. J. Labelled Compd.
Radiopharm. 1988, 15, 1349-1359.
(24) Taylor, E. C.; Henrie, R. N.; Portnoy, R. C. Pteridines. 44. A
convenient synthesis of 6-formylpterin. J. Org. Chem. 1978, 43,
736-737.
(25) Maunder, P.; Finglas, P. M.; Mallet, A. I.; Mellon, F. A.;
Razzaque, M. A.; Ridge, B.; Vahteristo, L.; Wittho¨ft, C. The
synthesis of folic acid, multiply labelled with stable isotopes,
for bio-availability studies in human nutrition. J. Chem. Soc.,
Perkin Trans. 1 1999, 1311-1323.
(26) Scott, J. M.; Thin-layer chromatography of pteroylglutamates
and related compounds. Methods Enzymol. 1980, 66, 437-443.
(27) Moran, R. G.; Colman, P. D. A simple procedure for the synthesis
of high specific activity tritiated (6S)-5-formyltetrahydrofolate.
Anal. Biochem. 1982, 122, 70-78.
(28) Huennekens, F. M.; Ho, P. P. K.; Scrimgeour, K. G. Preparation
and properties of “active formaldehyde” and “active formate”.
Methods Enzymol. 1963, 6, 806-811.
Received for review April 1, 2002. Revised manuscript received June
17, 2002. Accepted June 18, 2002. This study was supported by a grant
from the Deutsche Forschungsgemeinschaft (Ry, 19/2-1).
(18) Gregory, J. F., III; Toth, T. G. Chemical synthesis of deuterated
folate monoglutamate and in vivo assessment of urinary excretion
of deuterated folates in man. Anal. Biochem. 1988, 170, 94-
104.
JF025571K