Simple and High radiochemical yield synthesis of 2'-Deoxy-2'-[18F]fluorouridine via a new nosylate precursor

Article abstract of DOI:10.1002/jlcr.1140

We synthesized 2'-deoxy-2'-[¹⁸F]fluorouridine (7) as a radiotracer for positron emission tomography from a new nosylate precursor (6). This new precursor was synthesized from uridine in four steps. The overall synthetic yield was 9.4 percent and we have high stability of >98 percent purity up to 6 months at 4 deg C. The optimal manual [¹⁸F]fluorination conditions were 30 mg of the precursor 6 in 500 μl of acetonitrile at 145 deg C for 15 min with 370 MBq of [¹⁸F]fluoride. The [¹⁸F]fluorination yield was 76.5 +/- 2.7 percent (n = 3). After hydrolysis of protecting groups with 1N HCl and purification by HPLC, the overall radiochemical yield and purity were 26.5 +/- 1.4 percent and 98.2 +/- 2.5 percent, respectively. The preparation time was 70.0 +/- 10.5 min (n = 3 for each result). We also developed an automated method with a radiochemical yield and purity of 24.0 +/- 2.8 and 98.0 +/- 1.5 percent (n = 10) using a GE TracerLab MX chemistry module. This new nosylate precursor for 2'-deoxy-2'-[¹⁸F]fluorouridine synthesis showed higher radiochemical yields and reproducibility than previous methods.

Full text of DOI:10.1002/jlcr.1140

JOURNAL OF LABELLED COMPOUNDS AND RADIOPHARMACEUTICALS
J Label Compd Radiopharm 2006; 49: 1237–1246.
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jlcr.1140
Research Article
Simple and High radiochemical yield synthesis of
20-Deoxy-20-[18F]fluorouridine via a new nosylate
precursor
Se Hun Kang2, Seung Jun Oh1,*, Mi Kyung Yoon1, Jin Sook Ryu1, Won Koo
Lee2, Sun Joo Choi3, Kyung Pae Park3 and Dae Hyuk Moon1
1 Department of Nuclear Medicine, Asan Medical Center, University of Ulsan College of
Medicine, 388-1 Pungnap-2-dong, Songpa-gu, Seoul 138-736, Korea
2 Department of Chemistry, Sogang University, Seoul 121-742, Korea
3 Division of Radioisotope Production and Application, Hanaro Center, Korea Atomic
Energy Research Institite, Yusongku, Taejeon, 305-600, Korea
Summary
We synthesized 2’-deoxy-2’-[18F]fluorouridine (7) as a radiotracer for positron
emission tomography from a new nosylate precursor (6). This new precursor was
synthesized from uridine in four steps. The overall synthetic yield was 9.4% and we
have high stability of >98% purity up to 6 months at 48C. The optimal manual
[18F]fluorination conditions were 30 mg of the precursor 6 in 500 ml of acetonitrile at
1458C for 15 min with 370 MBq of [18F]fluoride. The [18F]fluorination yield was
76.5 ꢀ 2.7% (n=3). After hydrolysis of protecting groups with 1 N HCl and
purification by HPLC, the overall radiochemical yield and purity were 26.5 ꢀ 1.4%
and 98.2 ꢀ 2.5%, respectively. The preparation time was 70.0 ꢀ 10.5 min (n=3 for
each result). We also developed an automated method with a radiochemical yield and
purity of 24.0 ꢀ 2.8 and 98.0 ꢀ 1.5% (n=10) using a GE TracerLab MX chemistry
module. This new nosylate precursor for 2’-deoxy-2’-[18F]fluorouridine synthesis
showed higher radiochemical yields and reproducibility than previous methods.
Copyright # 2006 John Wiley & Sons, Ltd.
Received 6 July 2006; Revised 17 September 2006; Accepted 18 September 2006
Key Words: 2’-[18F]fluorouridine; tumor cell proliferation; uridine; positron emission
tomography; [18F]fluoride; gene imaging
*Correspondence to: Seung Jun Oh, Department of Nuclear Medicine, Asan Medical Center, 388-1
Pungnap-dong, Songpa-gu, Seoul 138-736, South Korea. E-mail: sjoh@amc.seoul.kr
Contract/grant sponsor: Korea Ministry of Science and Techonology
Copyright # 2006 John Wiley & Sons, Ltd.

1238
S. H. KANG ET AL.
Introduction
A variety of nucleoside analogs, including [11C]thymidine,1 [76Br]bromodeox-
yuridine,2–4 131I]iododeoxyuridine,5–6 and [18F]fluorothymidine ([18F]FLT)7,8
[
have been developed for tumor imaging by single photon emission and
positron emission tomography. Another nucleoside analog developed as an
antitumor agent, 2’-deoxy-2’-fluorouridine (2’-FU) (5), has shown a high ratio
of DNA incorporation in rat and woodchucks.9,10 In the previous report,
2’-[18F]FU was prepared from an anhydro-ring type precursor and labeled
with 18Fꢁ so that it can be used to follow tumor cell proliferation.10 Although
the method for synthesizing 2-[18F]FU is rapid and simple, the radiochemical
yield is very low (51%), and the [18F]fluorination step is slow and requires
dioxane as a solvent.
In the previous study, we developed a more efficient method for producing
2’-[18F]FU. We first examined the use of a triflate precursor (Figure 1).11 We
obtained an 82.4 ꢀ 3.6% yield (n=3) of [18F]fluorination as determined by
thin-layer chromatography (TLC). This triflate precursor, however, was stable
for only 1 day even below 08C due to sensitivity to moisture. In addition, it
was difficult to purify the precursor and final product due to the presence of
breakdown products.
Recently, better results have been achieved using a new nosylate precursor
in place of an anhydrothymidine precursor for [18F]FLT synthesis.12–15
This new nosylate precursor showed 5- to 10-fold higher radiochemical yields
than the previous anhydrothymidine precursor, and it was stable at 48C for
several months. Therefore, we used the nosylate leaving group to synthesize
3’,5’-di-O-tetrahydropyranyl-2’-O-nosyl-1-b-d-arabinopentofuranosyluracil (6)
as a new precursor for [18F]2’-FU and 2’-FU. In this study, we optimized
[18F]fluorination conditions with this precursor and applied them to an
automatic production with disposable cassette type synthesis module.
O
NH
OTf
N
THPO
O
O
OTHP
Figure 1. Chemical structure of the triflate precursor, 3’,5’-di-O-tetrahydropyr-
anyl-2’-O-triflate-1-b-d-arabinopentofuranosyluracil. In the previous report, this
triflate leaving group procedure showed high [18F]fluorination yields but very low
chemical stability even below 08C11
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 1237–1246
DOI: 10.1002/jlcr

RADIOSYNTHESIS OF 20-[18F]FU
1239
Results and discussion
Synthesis of 2’-FU (5) and a nosylate precursor (6)
We synthesized 3’,5’-di-O-tetrahydropyranyl-2’-O-nosyl-1-b-d-arabinopento-
furanosyluracil (6) as a new precursor for the synthesis of 2’-[18F]FU from
uridine via four steps (Scheme 1), with an overall yield of 9.4%. Starting from
uridine (1), 2,2’-anhydro-1-b-d-arabinofuranosyluracil (2) was obtained to
protect the hydroxyl group at the 2’-position.16 The remaining hydroxyl
groups at the 3’- and 5’-positions were protected with tetrahydropyranyl
(THP) groups, which we previously showed to give higher [18F]fluorination
yields than acetyl-, benzyl-, and dimethoxytrityl-protected precursors.11
The anhydro-ring opening reaction of compound 3 was carried out in 1 N
NaOH. Unlabeled 2’-FU (5) was obtained using DAST ((diethylamino)sulfur
trifluoride) as a fluorinating agent in 9:1 (v/v) CH2Cl2/THF, followed by
O
N
O
N
O
N
N
N
NH
(PHO)2CO
DMF, 71%
3,4-Dihydropyran
HO
O
O
O
HO
THPO
O
O
O
p-TsOH, THF,73%
OH OH
1
OH
2
OTHP
3
O
O
NH
NH
1. DAST, MC:THF(9:1)
2. 1N HCl, 27%
1N NaOH
OH
N
O
THPO
HO
N
O
O
O
5
MeOH, 90%
OTHP
4
HO
F
p
-NosCl
pyridine
MC
20%
O
O
N
NH
NH
1. K222, K2CO3, 18F-, CH3CN
2. 1N HCl, 26.5 1.4%
ONos
N
O
HO
O
THPO
O
O
HO 18
F
OTHP
6
7
(b)
Scheme 1. Chemical and radiochemical synthesis of the precursor and 2’-
[18F]FU. The precursor, 3’,5’-di-O-tetrahydropyranyl-2’-O-nosyl-1-b-d-arabino-
pentofuranosyluracil, was synthesized from uridine in four steps, and [18F]2’-FU
was obtained from the precursor in two steps ([18F]fluorination and hydrolysis of
protecting groups)
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 1237–1246
DOI: 10.1002/jlcr

1240
S. H. KANG ET AL.
hydrolysis of THP-protecting groups of the ring-opened 3’,5’-di-O-tetra-
hydropyranyl-1-b-d-arabinofuranosyluracil (4) with 1N HCl. Nosylation of 4
was performed using p-nitrophenylsulfonyl chloride and pyridine in CH2Cl2.
This nosylated precursor 6 was isolated as a white solid after column
chromatography purification on silica gel. Compared with the previous triflate
precursor, this new nosylate precursor showed good stability (>95% after 6
1
months at 48C). H and 13C NMR and elemental analyses confirmed the
production of 2’-FU (5) and the nosylate precursor (6).
Radiochemistry
[18F]fluorination yields of the nosylate precursor (6) were examined by TLC
and autoradiography. As shown in Table 1, the yields depended on the
reaction temperature and the precursor concentrations. The optimal
[18F]fluorination conditions were 30 mg of precursor (6) at 1458C for 15 min
with 370 MBq of [18F]fluoride. After purification by HPLC, we obtained a
decay-corrected radiochemical yield of 26.5 ꢀ 1.4% and a radiochemical
purity of 98.2 ꢀ 2.5%. The total synthetic time including HPLC purification
was 70.0 ꢀ 10.5 min (n=3). The synthesis of 2’-[18F]FU (7) was confirmed by
co-injection of unlabeled 2’-FU standard (5) and [18F]FU. They showed same
retention time (14.4 min) on UV (254 nm) and radioactive spectrum because
two compounds of same chemical structure have same k-factor. The crude,
purified samples and co-injection HPLC spectrum was in Figure 2.
We selected 15 min as the reaction time for the [18F]fluorination because it
gave the maximum yield. Hydrolysis of the THP-protecting group was
performed in 0.5 ml of 1 N HCl at 858C for 10 min. This hydrolysis
temperature was lower than in previous reports to prevent decomposition of
the compound. Despite a lower temperature, hydrolysis was complete.
We found only two radioactive peaks on the radioTLC. We assumed that
the major peak corresponded to the target product after [18F]fluorination, but
the yield was low. In contrast to radioTLC, HPLC purification procedures
Table 1. Effect of reaction temperature and precursor concentration on radiochemical
synthesis of 2’-[18F]FU
Radiochemical yields (%) (decay-corrected, n=3 for each condition)
mg precursor per 500 ml of CH3CN
Temperature (8C) 10
20
30
105
125
145
1.9 ꢀ 3.1
8.5 ꢀ 4.7
10.3 ꢀ 1.4
2.1 ꢀ 0.9
13.0 ꢀ 1.1
17.2 ꢀ 2.1
2.3 ꢀ 0.1
17.4 ꢀ 1.2
26.5 ꢀ 1.4
Reactions were carried out for 15 min.
Decay-corrected radiochemical yields from 370 MGBq/0.5 ml of [18F]fluoride.
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 1237–1246
DOI: 10.1002/jlcr

RADIOSYNTHESIS OF 20-[18F]FU
1241
Figure 2. Purification of 2’-[18F]FU: (a) HPLC profile of 2’-[18F]FU (14.4 min)
after crude HPLC purification; and (b) HPLC chromatogram of HPLC-purified
2’-[18F]FU; and (c) HPLC chromatogram of coinjected purified 2’-FU and 2’-
[18F]FU. Unlike the TLC results, there were three radioactive peaks in the
HPLC profile of the crude sample (a). Upper chromatogram shows UV
detection, and lower chromatogram shows radioactivity. In (c) the retention times
of 2’-FU and 2’-[18F]FU were the same
separated the product into three distinct radioactive peaks, corresponding to
18Fꢁ, an unknown polar [18F]-labeled impurity product, and 2’-[18F]FU. The
unknown polar impurity was the major [18F]-labeled compound. Although we
tried to separate these three compounds on the radioTLC analysis with
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 1237–1246
DOI: 10.1002/jlcr

1242
S. H. KANG ET AL.
various radioTLC developing conditions, we could not effectively separated
them. Unlike radioTLC analysis, HPLC analysis showed same results from
two different columns of ODS C18 (Beckman) and Luna C18 (Phenomenex).
We also developed an automated method for preparing 2’-[18F]FU using a
cassette-type chemistry module. After HPLC purification, we obtained a
radiochemical yield and purity of 24.0 ꢀ 2.8% and 98.0 ꢀ 1.5% (n=10),
respectively. The total preparation time including HPLC purification was
65.0 ꢀ 2.7 min (n=10). We used the same sequence program which was used
for [18F]FLT14 synthesis in a previous report with minor modification of
temperature and time. Because the reaction procedures such as [18F]fluorina-
tion and followed hydrolysis steps are the same between [18F]FU and [18F]FLT
but only the difference is [18F]fluorination temperature and time. Although we
had a slightly higher [18F]fluorination temperature (1458C), the reaction vessel
did not rupture and the disposable cassette system did not leak.
Conclusion
We developed a new nosylate precursor (6) for the synthesis of 2’-[18F]FU.
This new precursor shows higher stability and [18F]fluorination yields than
other precursors. We also developed a simple automated method for
synthesizing this tracer with high radiochemical yields using a single
disposable cassette.
Experimental
General
TLC was performed using Merck Silica Gel 60 F254, and chemical spots were
checked by UV light or 5% H2SO4 charring. Flash column chromatography
1
was performed using silica gel Merck 60 (Art 9385; 230–400 mesh). H NMR
spectra were obtained on 300 MHz spectrometers. NMR spectra were
recorded in ppm (d) in relation to tetramethylsilane as an internal standard
unless stated otherwise and are reported as follows: chemical shift, multiplicity
(br=broad, s=singlet, t=triplet, q=quartet, m=multiplet), coupling con-
stant, and integration. Elemental analyses were performed using a Carlo Erba
EA 1108 analyzer. All reagents were purchased from Sigma-Aldrich Korea
(Seoul, Korea) and were of reagent grade.
2,2’-Anhydro-1-b-d-arabinofuranosyluracil (2)
To a stirred suspension of diphenyl carbonate (9.65 g; 45 mmol) and N,N-
dimethylformamide (11 ml) in a flask, uridine (10 g; 41 mmol) was added, and
the slurry was heated to 808C. Next, sodium bicarbonate was added (40 mg;
0.50 mmol), and the mixture was heated at 1108–1208C for 4 h. The reaction
mixture was allowed to cool to room temperature and the precipitate was
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 1237–1246
DOI: 10.1002/jlcr

RADIOSYNTHESIS OF 20-[18F]FU
1243
isolated by filtration through a Buchner funnel. The solid was washed with two
bed volumes of methanol and then suspended as a slurry in 15 ml of methanol
and stirred for 30 min. The product was isolated by filtration and dried under
vacuum to afford 2,2’-anhydro-1-b-d-arabinofuranosyluracil (2) as a white
solid (6.62 g; 71%). 1H NMR (400 MHz, D2O): d 7.85 (d, 1 H, J=7.2 Hz), 6.47
(d, 1 H, J=5.6 Hz), 6.13 (d, 1 H, J=7.2 Hz), 5.40 (d, 1 H, J=5.2 Hz), 4.60
(s, 1 H), 4.33 (s, 1 H), 3.50 (m, 2 H).
3’,5’-Di-O-tetrahydropyranyl-2,2’-anhydro-1-b-d-arabinofuranosyluracil (3)
To a suspension of (2) (4 g, 17.70 mmol) in THF (160 ml), p-toluenesulfonic
acid (6.70 g; 35.40 mmol) was added at 08C, after which 2,3-dihydropyran
(16 ml; 177 mmol) was added dropwise at 08C, and the reaction mixture was
stirred at room temperature for 2 h, resulting in a clear solution. The reaction
mixture was neutralized with triethylamine. The solvent was removed by
evaporation and the residue was diluted with CH2Cl2, washed with aqueous
NaHCO3 and water, and then dried with MgSO4. The solvent was evaporated,
and the crude product was purified by recrystallization in CH2Cl2-hexane to
give 3’-5’-di-O-tetrahydropyranyl-2,2’-anhydro-1-b-d-arabinofuranosyluracil
1
(3) as a white solid (four diastereomers; 3.80 g; 73%). H NMR (400 MHz,
CDCl3): d 7.34 (s, 1 H), 6.25 (s, 1 H), 5.93 (s, 1 H), 5.49 & 5.27 (s, 1 H), 4.76–
4.38 (m, 4 H), 3.86–3.82 (m, 6 H), 1.73–1.20 (m, 12 H).
3’,5’-Di-O-tetrahydropyranyl-1-b-d-arabinofuranosyluracil (4)
3’-5’-di-O-tetrahydropyranyl-2,2’-anhydro-1-b-d-arabinofuranosyluracil (3)
(3.80 g; 9.60 mmol) was dissolved in methanol (60 ml) and mixed with 20 ml
of 1 N NaOH. After stirring for 2 h at room temperature, the reaction was
neutralized with acetic acid and evaporated. The residue was concentrated to
give 3’-5’-di-O-tetrahydropyranyl-1-b-d-arabinofuranosyluracil (4) as a foamy
1
solid (four diastereomers; 3.60 g; 90%). H NMR (400 MHz, CDCl3): d 7.78
(ddd, 1 H, J=8.0, 2.0 Hz), 6.24 & 6.14 (dd & dd, 1 H, J=5.2, 4.0 Hz), 5.59
(m, 1 H), 4.90 & 4.73 (m & s, 2 H), 4.51 (s, 1 H), 4.25 (m, 1 H), 4.13 (m, 1 H),
4.09–3.55 (m, 6 H), 1.88–1.54 (m, 12 H).
2’-Deoxy-2’-fluorouridine (5)
3’,5’-di-O-tetrahydropyranyl-1-b-d-arabinofuranosyluracil (4) (0.50 g, 1.21 mmol)
was dissolved in 50 ml of 9:1 (v/v) CH2Cl2/THF. After the solution was cooled
to 08C, (diethylamino)sulfur trifluoride (0.4 ml; 3 mmol) was added, and the
mixture was stirred for 7 h at room temperature. Next, 21 ml of 1 N HCl was
added, and the reaction mixture was stirred at room temperature for 2 h. The
reaction mixture was neutralized with 1 M NaOH and the solvent was
removed by evaporation. The residue was purified by column chromatography
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 1237–1246
DOI: 10.1002/jlcr

1244
S. H. KANG ET AL.
on silica gel with 9:1 (v/v) CH2Cl2/methanol as a solvent to give 5 as a foamy
solid (80 mg; 27%). 1H NMR (400 MHz, methanol-d4): d 8.00 (d, 1 H,
J=8.0 Hz), 5.98 (dd, 1 H, J=1.6, 17.6 Hz), 5.69 (d, 1 H, J=7.6 Hz), 5.02 (dd &
dd, 1 H, J=53.1, 4.4 Hz), 4.32 (dd & dd, 1 H, J=20.3, 7.6 Hz), 4.01 (d, 1 H,
J=7.6 Hz), 3.94 (dd, 1 H, J=12.4, 2.4 Hz), 3.76 (dd, 1 H, J=12.6, 2.8 Hz). 13
C
NMR (100 MHz, methanol-d4): d 168.0, 155.4, 142.8, 103.1, 95.5
0
0
0
(J2 ,F=185.3 Hz), 90.4 (J1 ,F=34.7 Hz), 85.0, 69.8 (J3 ,F=16.4 Hz), 61.5.
Calculated for C9H11FN2O5: C, 43.91; H, 4.50; N, 11.38. Found: C, 43.84;
H, 4.72; N, 11.40.
3’,5’-Di-O-tetrahydropyranyl-2-O-p-nitrophenylsulfonyl-1-b-d-arabinofuranosy-
luracil (6)
To a solution of (4) (1.80 g; 4.36 mmol) in CH2Cl2 (20 ml) and pyridine
(2.3 ml), p-nitrophenylsulfonyl chloride (3.3 g; 14.8 mmol) was added dropwise
at ꢁ208C. The brown solution was stirred for 48 h at room temperature and
washed with water, 1 N HCl, and water. The organic layer was dried by
Na2SO4, filtered, and concentrated to dryness. The residue was purified by
column chromatography on silica gel with 1:1 (v/v) n-hexane/ethyl acetate to
give 3’-5’-di-O-tetrahydropyranyl-2-O-p-nitrophenylsulfonyl-1-b-d-arabino-
furanosyluracil (6) as a white solid (four diastereomers; 0.52 g; 20%). m.p.:
1478–1488C. 1H-NMR (500 MHz, CDCl3): 9.28 (d, 1 H, J=14.4 Hz, NH), 8.40
(dd, 2 H, J=2.2, 8.9 Hz, p-NO2-benzene H), 8.04 (dd, 2 H, J=2.5, 9.0 Hz, p-
NO2-benzene H), 7.61 (dd & dd, 1 H, J=8.2, 12.8 Hz & J=7.1, 8.2 Hz, H-6),
6.10 (d, 1 H, J=4.0 Hz, H-5), 5.71 (m, 1 H, H-2’), 5.34–5.11 (m, 1 H, H-1’),
4.81-4.27 (m, 3 H, THP H-1, H-3’ & THP H-1), 4.24 (m, 1 H, H-4’), 4.07–3.56
(m, 6 H, H-5’ & THP H-5), 1.80–1.59 (m, 12 H, THP H-2,3,4). 13C-NMR
(125 MHz, CDCl3): 163.0 & 162.9 (C-4), 151.3, 149.7 & 149.6 (C-2), 141.2,
141.0 & 140.8 (C-6), 129.5, 129.3, 124.83 & 124.76 (p-NO2-benzene ring), 101.8
(C-5), 99.8, 99.5, 99.38, 99.35, 99.2, 98.5 & 98.3 (THP C-1), 84.7, 84.5, 84.0 &
83.9 (C-2’), 83.1, 82.9, 82.6 & 82.5 (C-1’), 82.1, 81.9, 81.6 & 81.5 (C-4’), 79.3 &
79.2 (C-3’), 66.7 & 66.2 (C-5’), 63.0, 62.9, 62.8, 62.7, 62.6 & 62.5 (THP C-5),
30.7, 30.6, 30.5 & 30.3 (THP C-2), 25.49, 25.46, 25.3 & 25.2 (THP C-4), 19.6,
19.4, 19.3 & 19.2 (THP C-3). Calculated for C25H31N3O12S: C, 50.25; H, 5.23;
N, 7.03. Found: C, 50.46; H, 5.39; N, 6.97.
Radiosynthesis of 2’-[18F]FU (7)
[18F]Fluoride 370 MBq/0.5 ml was trapped on a QMA cartridge and eluted
with a mixture of 22 mg of Kryptofix222 in 300 ml of acetonitrile and 7 mg of
K2CO3 in 300 ml of water. The solution was dried completely by azeotropic
distillation with CH3CN (3 ꢂ 500 ml) under a gentle N2 gas stream at 1008C.
The final dried residue was dissolved in 500 ml CH3CN, and 10, 20, or 30 mg
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 1237–1246
DOI: 10.1002/jlcr

RADIOSYNTHESIS OF 20-[18F]FU
1245
(25.3, 50.6, or 75.9 mmol) of nosylate precursor (6) were added to a closed-cap
vial. The reaction mixture was stirred at 105, 125, or 1458C for 15 min and
developed as dark brown color. After [18F]fluorination, acetonitrile was
removed, and the THP group was deprotected by adding 500 ml of 1 N HCl for
10 min at 858C. After neutralization with 2 N NaOH, the reaction mixture was
diluted with 2 ml of distilled water and then purified by HPLC. The synthesis
of 2’-[18F]FU (7) was confirmed by its coelution with unlabeled 2’-FU (5) at
14.4 min (Figure 1).
HPLC system and purification conditions
Compounds were purified and analyzed with a TSP P4000/UV 3000 system
(Thermo Separation Product, USA) with an Econosil C18 column (Alltech;
10 mm; 10 ꢂ 250 mm) and a flow-type radioactive detector. Reaction mixtures
were manually injected into the HPLC loop, and filterable impurities were
removed with a 0.22- mm vented filter prior to injection into the HPLC.
Compounds were eluted with water for 10 min, followed by 5% ethanol in
water for 10 min.
Automated synthesis
We used a GE TracerLab MX [18F]FDG module and the same sequence
program14 for [18F]FLT synthesis with minor changes in the reaction
temperature and time based on the results of manual synthesis. All Sep-Pak
cartridges in the disposable cassette were removed, and manifolds 2 and 3 were
connected with silicone tubing as described previously.14 This cassette had
four reagent supply vials. Acetonitrile (7 ml) was placed in vial 1 (blue), the
precursor (30 mg) and 2 ml of CH3CN in vial 2 (red), 1.75 ml of 2 N NaOH
and 1.7 ml of citrate buffer in vial 3 (yellow), and 3 ml of 1 N HCl in vial 4
(green). Two 30-ml disposable syringes were installed.
2’-[18F]FU was prepared by fluorination and hydrolysis (Scheme 1).
[18F]fluoride (3.7 GBq) delivered from a cyclotron was trapped on a QMA
cartridge and eluted into the reactor with a mixture of 7 mg of K2CO3 in 300 ml
of H2O and 22 mg of K222 in 300 ml of CH3CN. Following complete drying
under vacuum and purging with nitrogen at 958C, 30 mg of the precursor in
2 ml of CH3CN was added. [18F]fluorination was allowed to proceed for 150 s
at 1458C, followed by 300 s at 858C. The solvent was removed at 858C for
2 min under vacuum pressure and nitrogen purging. Hydrolysis was performed
with 1 N HCl at 858C for 5 min. After hydrolysis, the reaction mixture was
diluted with 2 N NaOH and buffer solution from the yellow vial, and the
mixture was injected automatically by vacuum into the HPLC. Following
HPLC, the purified 2’-[18F]FU was passed through a 0.22-mm sterile filter and
collected into a sterile vial.
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 1237–1246
DOI: 10.1002/jlcr

1246
S. H. KANG ET AL.
Acknowledgements
We gratefully acknowledge a grant from the Korea Ministry of Science and
Technology for Real-time Molecular Imaging and Mid- and Long-term
Nuclear Research.
References
1. Vander Borght T, Pauwels S, Lambotte L, Labar D, De Maeght S, Stroobandt G,
Laterre C. J Nucl Med 1994; 35: 974–982.
2. Gudjonsson O, Bergstrom M, Kristjansson S, Wu F, Nyberg G, Fasth K-J,
¨
Langstrom B. Nucl Med Biol 2001; 28: 59–65.
3. Ryser JE, Blauenstein P, Remy N, Weinreich R, Hasler PH, Novak-Hofer I,
Schubiger PA. Nucl Med Biol 1999; 26: 673–679.
4. Bergstrom M, Lu L, Fasth K-J, Wu F, Bergstrom-Pettermann E, Tolmachev V,
Hedberg E, Cheng A, Langstrom B. J Nucl Med 1998; 39: 1273–1279.
5. Tjuvajev JG, Macapinlac HA, Daghighian F. J Nucl Med 1994; 35: 1407–1417.
6. Gratzner HG. Science 1982; 218: 474–475.
7. Shields AF, Grierson JR, Dohmen BM, Machulla HJ, Stayanoff JC, Lawhorn-
Crews JM, Obradovich JE, Muzik O, Mangner TJ. Nat Med 1998; 4: 1334–1336.
8. Major P, Egan E, Herrick D, Kufe DW. Cancer Res 1982; 42: 3005–3009.
9. Richardson FC, Zhang C, Lehrman SR, Koc H, Swenberg JA, Richardson KA,
Bendele RA. Chem Res Toxicol 2002; 15: 922–926.
10. Abrams DN, Mercer JR, Knaus EE, Wiebe LI. Int J App Radiat Isot 1985; 36:
233–238.
11. Yun MK, Oh SJ, Yeo JS, Ryu JS, Moon DH. J Nucl Med 2003; 44: 297P.
12. Martin SJ, Eisenbarth JA, Wagner-Utermann U, Mier W, Henze M, Pritzkow H.
Nucl Med Biol 2002; 29: 263–273.
13. Yun M, Oh SJ, Ha HJ, Ryu JS, Moon DH. Nucl Med Biol 2003; 30: 151–157.
14. Oh SJ, Mosdzianowski C, Chi DY, Kim JY, Kang SH, Ryu JS, Moon DH. Nucl
Med Biol 2004; 31: 803–809.
15. Machulla HJ, Blocher A, Kuntzsch M, Piert M, Wei R, Grierson JR. J Radioanal
Nucl Chem 2000; 243: 843–846.
16. McGee DP, Vaughn-Settle A, Vargeese C, Zhai Y. J Org Chem 1996; 61: 781–785.
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 1237–1246
DOI: 10.1002/jlcr