Synthesis and purification of [2-13C]-5-fluorouracil

Article abstract of DOI:10.1002/jlcr.1872

[2-¹³C]-5-Fluoropyrimidine-2,4(1H,3H)-dione ([2- ¹³C]-5-fluorouracil or [2-¹³C]-5-FU) is a potential diagnostic agent for measuring 5-FU-induced toxicity in cancer patients. It was prepared and purified with isotopic and chemical purity of>99% on a multigram scale in a two-step synthesis from [¹³C]-urea. Preparative separation of [2-¹³C]-FU and [2-¹³C]-uracil was carried out by automated medium pressure silica gel column chromatography. The method is applicable to a broader range of 5-FU isotopic analogs derived from labeled uracil. Copyright 2011 John Wiley & Sons, Ltd. [2- ¹³C]-5-Fluoropyrimidine-2,4(1H,3H)-dione ([2-¹³C]-5- fluorouracil or [2-¹³C]-5-FU) is a potential diagnostic agent for measuring 5-FU-induced toxicity in cancer patients. It was prepared and purified with isotopic and chemical purity of >99% on a multigram scale in a two-step synthesis from [¹³C]-urea. Preparative separation of [2- ¹³C]-FU and [2-¹³C]-urea was carried out by medium pressure automated silica gel column chromatography. The method is applicable to a broader range of 5-FU isotopic analogs derived from labeled uracil. Copyright

Full text of DOI:10.1002/jlcr.1872

Note
Received 28 October 2010,
Revised 1 November 2010,
Accepted 4 November 2010
Published online 17 February 2011 in Wiley Online Library
(
wileyonlinelibrary.com) DOI: 10.1002/jlcr.1872
Synthesis and purification
of [2- C]-5-fluorouracil
1
3
a
Huzaifa S. Rangwala, John W. Giraldes, and Vadim J. Gurvich
a
a,bÃ
1
3
13
13
2- C]-5-Fluoropyrimidine-2,4(1H,3H)-dione ([2- C]-5-fluorouracil or [2- C]-5-FU) is a potential diagnostic agent for
[
measuring 5-FU-induced toxicity in cancer patients. It was prepared and purified with isotopic and chemical purity of499%
1
3
13
13
on a multigram scale in a two-step synthesis from [ C]-urea. Preparative separation of [2- C]-FU and [2- C]-uracil was
carried out by automated medium pressure silica gel column chromatography. The method is applicable to a broader range
of 5-FU isotopic analogs derived from labeled uracil.
1
Keywords: 5-FU; [2- C]-5-fluorouracil; carbon-13; stable isotope; 5-FU/uracil separation
3
2
hypofluorite, , selective F-Cl exchange, or a lengthy multistep
1
22
Introduction
3
,23
approach.
More recently, a thermal conversion of series of
5
-Fluoropyrimidine-2,4(1H,3H)-dione (5-fluorouracil or 5-FU) has O-alkyl and O-acyl substituted 5-fluoro-6-hydroxy-5,6-dihydro-
2
4
been used as an effective anti-tumor agent for several pyrimidinediones to 5-FU was reported. All these strategies
involve using dangerous reagents that also represent a
1
decades. The original strategy for its synthesis from pseudo-
,2
s
urea and a-fluoro-b-keto ester enolates was reported by challenge in scaling up. In this work, Selectfluor was chosen
3
–5
Duschinsky over 50 years ago.
patients are treated with 5-FU every year of which a certain
Hundreds of thousands of as the effective and most safe agent for direct fluorination of
uracil. The use of this reagent for the preparation of unlabeled
percent of patient population exhibit severe toxicity and even 5-FU was reported in 1995 by Lal et al.25
death. That, in part, can be attributed to the dihydropyrimidine
dehydrogenase (DPD) deficiency, a result of decreased activity
of DPD, that plays a key role in the uracil catabolic pathway
Experimental
6
–9
13
(
Scheme 1).
a 5-FU that contains a nonradioactive isotope, such as carbon- Andover, MA. All other reagents and supplies were purchased
3, in position 2, can be used as a diagnostic agent. As a result of from Sigma Aldrich and Fisher Scientific. Nuclear magnetic
catabolic activation in patients, it will produce a highly stable resonance spectra were recorded on a 400 MHz Bruker Avance
To assess this deficiency in cancer patients, [ C]-urea was obtained from Cambridge Isotope Laboratories,
1
1
3
CO
breath test.
To enable this diagnostic test, a multigram quantity of a high- internal reference for fluorine NMR. Purity and isotopic
2
that can be detected and quantitatively measured in a spectrophotometer. Chemical shifts are reported as d values
1
0
with reference to TMS. Fluorotrichloromethane was used as an
1
3
13
quality [2- C]-5-FU (1) is required. It can be prepared from [ C]- enrichment of the final product were determined using an
urea (2) which is commercially available. The use of the labeled Alliance 2695 HPLC System equipped with PDA detector and
urea for introducing an isotope into position-2 of the uracil ring Shimadzu LCMS 2010 MS Detector. Chromatographic separation
1
1
was first reported in 1952 by Mander and Brown who reacted was performed on Teledyne ISCO Automated Chromatography
s
1
4
[
approach pioneered by Davidson and Baudusch. The same
C]-urea with malic acid in fuming sulfuric acid following the system Combiflash XL using Teledyne ISCO flash columns.
1
2
1
4
13
year, Bennett reported a condensation of [ C]-thiourea with [2- C]-Pyrimidine-2,4(1H,3H)-dione (4)
1
diethoxypropionate to produce [2- C]-thiouracil that was
4
1
4
13
Polyphosphoric acid (PPA; 405 g) was placed into a 1 L round
bottom flask and heated at 1101C under stirring for 20 min. Then
subsequently converted to [2- C]-uracil. A similar approach
was reintroduced in 2001 by a Pfizer group that reported the
1
4
use of [ C]-thiourea for introduction of an isotope into position-
a
2
followed by hydrolysis with chloroacetic acid to produce
1
Institute for Therapeutics Discovery and Development, College of Pharmacy,
4
14
11
[
2- C]-uracil. The use of labeled [ C]-phosgene was reported University of Minnesota, Minneapolis, MN, USA
for the introduction of the C-11 moiety into the 2-position of
b
Department of Medicinal Chemistry, College of Pharmacy, University of
-FU for use as an agent for the positron emission tomographic Minnesota, Minneapolis, MN, USA
5
imaging-based diagnostic Strauss test.
1
5
*
Correspondence to: Vadim J. Gurvich, Institute for Therapeutics Discovery and
Most of the previously reported pathways either include
1
Development, College of Pharmacy, University of Minnesota, 717 Delaware St.
SE, Minneapolis, MN 55414, USA.
6
direct fluorination of uracil with hydrogen fluoride HF,
1
7–19
16,20
fluorine,
trifluoromethyl hypofluorite CF OF,
3
acetyl E-mail: vadimg@umn.edu
J. Label Compd. Radiopharm 2011, 54 340–343
Copyright r 2011 John Wiley & Sons, Ltd.

H. S. Rangwala et al.
Ã
13
= C.
Scheme 1. (a) Dihydropyrimidine dehydrogenase (PDP); (b) Dihydropyrimidinase (DHP); and (c) b-Ureidopropionase (BUP);
s
O; (c) Selectfluor , Ph
À
1
Ã
13
= C.
Scheme 2. (a) PPA; (b) NaOH/H
2
4
B Na ; and (d) 220–230 1C;
1
[
3
C]-urea (2, 15 g, 246 mmol) was added. After the reaction possible. If it was greater than 5%, the material was dissolved in
mixture became completely homogenic, propiolic acid (3, methanol, stirred with silica gel and the solvent was evaporated
6.6 mL; 270 mmol) was slowly added in approx 4 mL portions. to dryness under reduced pressure. The residue was then loaded
1
The heating bath temperature was lowered to 951C and the on the loading column of Teledyne ISCO Automated Chromato-
reaction mixture stirred for 4 h. Water (600 mL) was slowly added graphy system Combiflash XL and purified on a silica gel column
s
followed by charcoal (3 g) and the mixture was stirred for approx with eluent acetyl acetate–isopropanol using the following
1
min. Charcoal was filtered out and the filtrate was placed into a gradient: 0–5 min – 100% ethyl acetate, 5–15min – 99.5% ethyl
refrigerator and left overnight. The precipitated solid was then acetate, 15–30min – 99% ethyl acetate, 35min – 95% ethyl
collected by vacuum filtration and re-crystallized from boiling acetate. The fractions were analyzed by HPLC and those
water (600 mL); charcoal (3 g) was added to the boiling solution containing the pure product were combined and the solvents
and immediately filtered off. The solution was allowed to cool were removed under reduced pressure. After re-crystallization
down, the precipitated solid was collected and dried in vacuo to from ethanol, the material was dried in vacuo to produce 13.4g
1
produce 17.6 g (64% yield) of the desired product 4: H NMR (41% yield) of the desired product 1 with499% purity as
(
DMSO-d ) d 11.0 (s, 1H), 10.8 (s, 1H), 7.38 (m, 1H), 5.44 (dd, determined by HPLC and499.9% isotopic enrichment as
6
1
= 1.85, 1H). C NMR (DMSO-d
3
1
J
(
1
= 7.56, J
2
6
) d 100, 142, 152 determined by LCMS: H NMR (DMSO-d
6
) d 11.5 (bs, 1H), 10.7
) d 158, 151 (high-
1
high-intensity signal), 164. MS m/z 112 [M -1].
13
(bs, 1H), 7.75 (m, 1H). C NMR (DMSO-d
6
1
9
intensity signal), 141, 139, 126. F NMR (DMSO-d ) d 171.4.
6
1
MS m/z 130 [M -1].
1
2- C]-5-Fluoropyrimidine-2,4(1H,3H)-dione (1)
3
[
1
2- C]-Pyrimidine-2,4(1H,3H)-dione 4 (28.4 g, 251 mmol) was
3
[
placed into distilled water (850 mL). Selectfluor (88.9 g,
Results and discussion
s
2
51 mmol) was added to the suspension and the mixture was The primary goal of this study was to develop a reliable and
1
3
stirred at 1051C for 16 h. Sodium tetraphenylborate (192 g, scalable methodology for the preparation of [2- C]-5-FU (1) on
61 mmol) was added. The heat source was removed from the a multigram scale with the chemical and isotopic purity
5
flask and the mixture was stirred on an ice bath for 30 min. of499%. This approach is designed for the preparation of this
The precipitated solid was then filtered out and washed with material under FDA-regulated cGMP conditions that makes it
water. The filtrate and washing water were combined and suitable for use as an investigational new agent in clinical trials.
1
3
concentrated under reduced pressure and the residue was dried
For the preparation of [2- C]-5-FU (1), a three-step chemical
1
3
overnight in a vacuum drying oven at 951C. The residue was process represented in Scheme 2 has been chosen. [2- C]-Uracil
transferred into the sublimating apparatus base and flushed (4) was prepared using a modified procedure of the Iida
13
with nitrogen. The sublimating apparatus was heated in a silicon application of the Harada and Suzuki condensation of [ C]-
oil bath (bath temperature 195–2051C) under vacuum (0.5 mm urea (2) with propiolic acid (3). The resulting [2- C]-uracil (4)
Hg) while the temperature of the cold finger was maintained at was converted to [2- C]-5-FU (1) by direct fluorination with
the range of 10–501C. The sublimation proceeded until the Selectfluor . Our initial experimental results following Lal’s
2
6
27
1
3
1
3
s
2
5
2
5
residue on the bottom of the apparatus was completely approach revealed that the actual yield and purity of the final
blackened. Then the sublimed material was removed, the product were far lower than expected. It was unclear whether
sublimation apparatus was rinsed with water and acetone, and the reported 82% yield was referring to 5-FU or its immediate
dried. The sublimation then proceeded with a silicon oil bath precursor, 5-fluoro-6-hydroxydihydropyrimidine-2,4(1H,3H)-dione,
heated to a temperature of 220–2301C until the residue on the that is formed as a result of the direct fluorination of uracil with a
bottom of the apparatus was completely blackened. The reported yield of 82%. The purity of the final product was not
sublimed material was removed from the apparatus and its reported.
1
composition was determined by H NMR. If the content of the
We attempted to optimize the fluorination step by varying
s
1
3
unreacted [2- C]-pyrimidine-2,4(1H,3H)-dione 4 was less than the amount of Selectluor from 0.9 to 1.5 eq of uracil. The
%, further purification by recrystallization from ethanol is conversion of uracil was only 75–80% at 0.9 eq and an
5
J. Label Compd. Radiopharm 2011, 54 340–343
Copyright r 2011 John Wiley & Sons, Ltd.
www.jlcr.org

H. S. Rangwala et al.
unidentified impurity was increasingly fsormed when we followed by final re-crystallization from ethanol allowing us to
1
3
progressed from 1.1 to 1.5 eq of Selectfluor . While the use of achieve499% purity of [2- C]-5-FU (1); Figure 1. Crystallization
.0 eq did not result in full conversion of the starting material, from ethanol without a chromatography step allowed us to
the formation of the impurity was minimal. Our experiments achieve a high level of purity but would not remove the residual
1
1
3
revealed that an unidentified impurity formed at the fluorination [2- C]-uracil (4).
step, was not being removed by sublimation. We also found this
specific impurity to be difficult to remove by both chromato- Conclusion
graphy and recrystallization. However, it was determined that
1
2- C]-5-FU (1) was successfully synthesized and purified by
3
[
the impurity is sublimed at a lower temperature range than the
target material 1. To resolve this problem, a fractionated
sublimation process was developed. At the first step, the
sublimation apparatus was heated to 195–2051C in which some
preparative chromatography on silica gel followed by recrys-
tallization. The material was prepared on a multigram scale that
can be further scaled up for obtaining larger quantities. The
material is characterized by high chemical and isotopic purities
and it is suitable for further clinical development as a diagnostic
agent. This approach is applicable to any conversion of uracil to
5-FU on a larger scale.
1
3
[
second step was carried out at a higher temperature range
2- C]-5-FU (1) and the entire impurity were sublimed. The
1
3
of 220–2301C that allowed the isolation of [2- C]-5-FU (1)
contaminated only with 5–10% of [2- C]-uracil (4).
1
3
The composition of the fluorohydrin 5 products, as determined
8
2
Acknowledgement
by NMR, was consistent with that described by Visser. Our
attempts to substitute the sublimation procedure with certain
other types of hydrolysis and elimination previously described in We are grateful to the Mayo Clinic and our collaborator, Robert
2
9–31
B. Diasio, M.D., for providing financial support for this study.
literature,
as well as using microwave irradiation, did not
result in sufficient conversion of the fluorohydrin 5 to the target
compound 1 or any advantages for purification.
1
The sublimated material contained up to 10% of [2- C]-uracil
3
References
(4) and our attempts to remove it by re-crystallizations were not
successful. Chromatographic separation of 5-FU and uracil is
considered a challenge that reportedly can only be achieved
[
1] G. C. Daher, B. E. Harris, R. B. Diasio, Pharmacol. Ther. 1990, 48,
189–222.
2] J. L. Grem, Invest. New Drugs 2000, 18, 299–313.
3] R. Duschinsky, E. Pleven, C. Heidelberger, J. Am. Chem. Soc. 1957,
79, 4559–4560.
1
9,32–35
by preparative HPLC.
We applied the normal phase
[
[
HPLC approach developed by Ka zˇ oka for detection of uracil in
3
4,36
5-FU
to a multigram preparative separation by Teledyne
s
ISCO Automated Chromatography system Combiflash XL. The
separation was performed on a silica gel column using an acetyl
acetate–isopropanol gradient. The chromatography step was
[4] C. Heidelberger, R. Duschinsky, US 2802005, 1957.
[
5] C. Heidelberger, N. K. Chaudhuri, P. Danneberg, D. Mooren,
L. Griesbach, R. Duschinsky, R. J. Schnitzer, E. Pleven, J. Scheiner,
Nature 1957, 179, 663–666.
[
[
6] A. Lee, H. Ezzeldin, J. Fourie, R. Diasio, Clin. Adv. Hematol. Oncol.
2004, 2, 527–532.
7] H. R. Thomas, H. H. Ezzeldin, V. Guarcello, L. K. Mattison,
B. L. Fridley, R. B. Diasio, Pharmacogenetics and Genomics 2007,
1
8] H. R. Thomas, H. H. Ezzeldin, V. Guarcello, L. K. Mattison, B. L. Fridley,
7, 973–987.
[
R. B. Diasio, Pharmacogenetics and Genomics 2008, 18, 25–35.
9] H. H. Ezzeldin, R. B. Diasio, J. Clin. Oncol. 2008, 26, 2080–2082.
[
10] L. K. Mattison, H. Ezzeldin, M. Carpenter, A. Modak, M. R. Johnson,
R. B. Diasio, Clin. Cancer Res. 2004, 10, 2652–2658.
[
[11] H. G. Mandel, C. L. Brown, J. Am. Chem. Soc. 1952, 74, 2439–2440.
[12] D. Davidson, O. Baudisch, J. Am. Chem. Soc. 1926, 48, 2379–2383.
[13] L. L. Bennett, J. Am. Chem. Soc. 1952, 74, 2432–2433.
[14] K. S. Zandi, S. A. Miller, Synth. Appl. Isot. Label. Compd. Proc. Int.
Symp. 7th 2001, 268–271.
[
[
[
15] K.-i. Seki, K.-i. Nishijima, Y. Kuge, N. Tamaki, L. I. Wiebe, K. Ohkura,
J. Pharm. Pharm. Sci. 2007, 10, 212–216.
16] R. H. Hesse, D. H. R. Barton, H. T. Toh, M. M. Pechet, J. Org. Chem.
1
972, 37, 329–330.
17] E. N. Vine, D. Young, W. H. Vine, W. Wolf, Int. J. Appl. Radiat. Isot.
979, 30, 401–405.
1
[
[
18] M. Diksic, P. Di Raddo, Tetrahedron Lett. 1984, 25, 4885–4888.
19] F. Oberdorfer, E. Hofmann, W. Maier-Borst, J. Label. Compd.
Radiopharm. 1989, 27, 137–145.
[20] O. Miyashita, K. Matsumura, H. Shimadzu, N. Hashimoto, Chem.
Pharm. Bull. 1981, 29, 3181–3190.
[21] G. W. M. Visser, R. E. Herder, F. J. J. De Kanter, J. D. M. Herscheid,
J. Chem. Soc. Perkin Trans. 1: Org. Bio-Org. Chem. (1972–1999)
1988, 1203–1207.
[
[
22] B. Baasner, E. Klauke, J. Fluorine Chem. 1989, 45, 417–430.
23] G. Q. Shi, Q. Wang, M. Schlosser, Tetrahedron 1996, 52,
4
403–4410.
1
3
Figure 1. Chromatograms of analytical HPLC of (a) mixture of [2- C]-5-FU (1) and
[
chromatographic purification.
[
[
24] S. G. Semenov, B. N. Maximov, Fluorine Notes 2006, 45.
25] G. S. Lal, W. Pastore, R. Pesaresi, J. Org. Chem. 1995, 60,
13
13
2- C]-uracil (4) before chromatographic purification and (b) [2- C]-5-FU (1) after
7
340–7342.
www.jlcr.org
Copyright r 2011 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2011, 54 340–343

H. S. Rangwala et al.
[
26] K. Iida, T. Chiyoda, R. Hirasawa, A. Iwata, M. Kajiwara, J. Label.
Compd. Radiopharm. 1997, 39, 69–77.
[31] G. W. M. Visser, R. Wedzinga, R. P. Klok, J. D. M. Herscheid,
J. Chem. Soc. Perkin Trans. 2 1994, 231–236.
[
[
27] K. Harada, S. Suzuki, Tetrahedron Lett. 1976, 2321–2322.
28] G. W. M. Visser, S. Boele, B. W. Vonhalteren, G. Knops,
J. D. M. Herscheid, G. A. Brinkman, A. Hoekstra, J. Org. Chem.
[32] A. A. Miller, J. A. Benvenuto, T. L. Loo, J. Chromatogr. B: Biomed.
Sci. Appl. 1982, 228, 165–176.
[33] F. P. LaCreta, W. M. Williams, J. Chromatogr. B: Biomed. Sci. Appl.
1987, 414, 197–201.
1
986, 51, 1466–1471.
[
29] H. A. Lozeron, M. P. Gordon, T. Gabriel, W. Tautz, R. Duschinsky,
Biochemistry 1964, 3, 1844–1850.
30] R. Duschinsky, T. Gabriel, W. Tautz, A. Nussbaum, M. Hoffer,
E. Grunberg, J. H. Burchenal, J. J. Fox, J. Med. Chem. 1967, 10, 47–58.
[34] H. Ka zˇ oka, Khimiko-Farmatsevticheskii Zh. 1995, 29, 49–51.
[35] G. J. Peters, I. Kraal, E. Laurensse, A. Leyva, H. M. Pinedo,
J. Chromatogr. 1984, 307, 464–468.
[
[36] H. Ka zˇ oka, J. Chromatogr. A 2003, 994, 221–225.
J. Label Compd. Radiopharm 2011, 54 340–343
Copyright r 2011 John Wiley & Sons, Ltd.
www.jlcr.org