Development of radioiodinated nucleoside analogs for imaging tissue proliferation: comparisons of six 5-iodonucleosides.

Abstract

The aim of this study was to determine the most suitable iodonucleoside analogs for use in tissue proliferation imaging by means of single photon emission tomography (SPECT). In this study, 5-[(125)I]iodo-(2-deoxy-2-fluoro-4-thio-beta-D-arabinofuranosyl)uracil ([(125)I]FITAU, 1E) and 5-[(125)I]iodo-1-methyl-(2-deoxy-2-bromo-beta-D-arabinofuranosyl)uracil ([(125)I]IMBAU, 1F) were synthesized and their biological data were compared with previously published results regarding 4'-thio nucleoside analogs and the reference compound 5-[(125)I]iodo-(2-deoxy-2-fluoro-beta-D-arabinofuranosyl)uracil ([(125)I]FIAU, 1D). 5-Iodo-(2-deoxy-2-fluoro-beta-D-arabinofuranosyl)uracil (FIAU, 2D), 5-iodo-(2-deoxy-2-fluoro-4-thio-beta-D-arabinofuranosyl)uracil (FITAU, 2E), and 5-iodo-1-methyl-(2-deoxy-2-bromo-beta-D-arabinofuranosyl)uracil (IMBAU, 2F) were successfully labeled with (125)I and their in vitro cytosolic thymidine kinase (TK(1)) phosphorylation, recombinant thymidine phosphorylase enzymatic catabolism, TK(1)-dependent cell uptake, and in vivo biodistribution in normal mice were evaluated. Five compounds (1B, 1C, 1D, 1E, and 1F) were stable against C-N glycoside degradation induced by recombinant thymidine phosphorylase. However, 5-[(125)I]iodo-2'-deoxyuridine ([(125)I]IUdR, 1A) was not shown to be stable against such degradation. The TK(1) assay showed that [(125)I]FIAU (1D) expressed 16% of the phosphorylation potential of [(125)I]IUdR (1A). Furthermore, [(125)I]FITAU (1E) was shown to have reduced phosphorylation potential, in comparison with that of [(125)I]IUdR (1A) (<0.01). [(125)I]IMBAU (1F) did not show any phosphorylation. In vitro cell uptake and in vivo proliferation-selective uptake of each nucleoside was largely dependent on its potential as a TK(1) substrate. Neither [(125)I]FITAU (1E) nor [(125)I]IMBAU (1F) were shown to have distinct TK(1)-dependent cell uptake and retention in the proliferating tissues. From these results, we concluded that [(125)I]FITAU (1E) and [(125)I]IMBAU (1F) are not effective as imaging agents of cell proliferation. The biological data obtained with these nucleosides were compared, and requirements for the design of pharmaceutically useful radioiodinated nucleoside analogs were also considered.

Full text

Page 1

Development of radioiodinated nucleoside analogs for imaging tissue
proliferation: comparisons of six 5-iodonucleosides
Jun Toyoharaa,c, Akio Hayashia, Mikiko Satoa, Akie Gogamia, Hiromichi Tanakab,
Kazuhiro Haraguchib, Yuichi Yoshimurab, Hiroki Kumamotob, Yoshiharu Yonekurac,
Yasuhisa Fujibayashic,*
aResearch Center, Research and Development Division, Nihon Medi-Physics, Co., Ltd., Chiba, Japan
bSchool of Pharmaceutical Science, Showa University, Tokyo, Japan
cBiomedical Imaging Research Center, Fukui Medical University, Fukui, Japan
Abstract
The aim of this study was to determine the most suitable iodonucleoside analogs for use in tissue proliferation imaging by means of single
photon emission tomography (SPECT). In this study, 5-[125I]iodo-(2-deoxy-2-fluoro-4-thio-?-D-arabinofuranosyl)uracil ([125I]FITAU, 1E)
and 5-[125I]iodo-1-methyl-(2-deoxy-2-bromo-?-D-arabinofuranosyl)uracil ([125I]IMBAU, 1F) were synthesized and their biological data
were compared with previously published results regarding 4?-thio nucleoside analogs and the reference compound 5-[125I]iodo-(2-deoxy-
2-fluoro-?-D-arabinofuranosyl)uracil ([125I]FIAU, 1D). 5-Iodo-(2-deoxy-2-fluoro-?-D-arabinofuranosyl)uracil (FIAU, 2D), 5-iodo-(2-de-
oxy-2-fluoro-4-thio-?-D-arabinofuranosyl)uracil (FITAU, 2E), and 5-iodo-1-methyl-(2-deoxy-2-bromo-?-D-arabinofuranosyl)uracil (IM-
BAU, 2F) were successfully labeled with125I and their in vitro cytosolic thymidine kinase (TK1) phosphorylation, recombinant thymidine
phosphorylase enzymatic catabolism, TK1-dependent cell uptake, and in vivo biodistribution in normal mice were evaluated. Five
compounds (1B, 1C, 1D, 1E, and 1F) were stable against C-N glycoside degradation induced by recombinant thymidine phosphorylase.
However, 5-[125I]iodo-2?-deoxyuridine ([125I]IUdR, 1A) was not shown to be stable against such degradation. The TK1assay showed that
[125I]FIAU (1D) expressed 16% of the phosphorylation potential of [125I]IUdR (1A). Furthermore, [125I]FITAU (1E) was shown to have
reduced phosphorylation potential, in comparison with that of [125I]IUdR (1A) (?0.01). [125I]IMBAU (1F) did not show any phosphor-
ylation. In vitro cell uptake and in vivo proliferation-selective uptake of each nucleoside was largely dependent on its potential as a TK1
substrate. Neither [125I]FITAU (1E) nor [125I]IMBAU (1F) were shown to have distinct TK1-dependent cell uptake and retention in the
proliferating tissues. From these results, we concluded that [125I]FITAU (1E) and [125I]IMBAU (1F) are not effective as imaging agents of
cell proliferation. The biological data obtained with these nucleosides were compared, and requirements for the design of pharmaceutically
useful radioiodinated nucleoside analogs were also considered. © 2003 Elsevier Inc. All rights reserved.
Keywords: [125I]IUdR; [125I]ITdU; [125I]ITAU; [125I]FIAU; [125I]FITAU; [125I]IMBAU
1. Introduction
The design of new and effective agents for use in tumor
imaging is an important goal of radiopharmaceutical chem-
ists working in the field of nuclear medicine. Radiolabeled
nucleosides are currently thought to be promising candi-
dates for in vivo imaging of tumor cell proliferation. We
recently demonstrated that the radioiodinated 4?-thio deriv-
ative 5-[125I]iodo-4?-thio-2?-deoxyuridine ([125I]ITdU, 1B)
(Fig. 1) is a potentially useful imaging agent of cell prolif-
eration [18]. This compound was exhibited to be highly
resistant to the C-N glycosidic bond cleavage reaction pro-
voked by thymidine phosphorylase. In spite of the substi-
tution of sulfur for oxygen in the furanose, [125I]ITdU (1B)
maintains an affinity for TK1as a substrate. The [125I]ITdU
(1B) compound was shown to have higher in vivo stability
and more rapid DNA incorporation than does [125I]IUdR
(1A) (Fig. 1); in addition, [125I]ITdU (1B) was shown to be
selectively and significantly taken up in rapidly proliferating
tissues. These findings suggest that a “4?-thio strategy”
could potentially be useful in the design of imaging agents
of cell proliferation. However, the specific and significant
uptake of [125I]ITdU (1B) by proliferating tissues requires a
greater degree of washout and clearance of the exchange-
* Corresponding author. Tel.: 81-776-61-8430; fax: 81-776-61-8170
E-mail address: yfuji@fmsrsa.fukui-med.ac.jp (Y. Fujibayashi).
Nuclear Medicine and Biology 30 (2003) 687–696 www.elsevier.com/locate/nucmedbio
0969-8051/03/$ – see front matter © 2003 Elsevier Inc. All rights reserved.
doi:10.1016/S0969-8051(03)00081-7

Page 2

able fraction of background radioactivity. The metabolites
of [125I]ITdU (1B) would influence specific imaging results
early in the time course of metabolism. Therefore, as re-
gards the development of nucleoside-based imaging agents,
the modification of the current strategy, or the introduction
of a new design strategy, remains necessary.
The 2-arabino fluorination of the sugar moiety of nucleo-
sides is a well-known approach for the effective use of
anti-virus chemotherapy agents as well as anti-tumor ther-
apeutic agents [5,20,21]. It is also well known that 2-arabino
fluorination will stabilize the C-N glycosidic bond of
nucleosides [12]. Moreover, numerous structure-activity re-
lationship studies regarding the phosphorylation of nucleo-
side analogs to monophosphates by nucleoside kinases have
suggestedthat5-methyl-(2-deoxy-2-fluoro-?-D-arabino-
furanosyl)uracil (FMAU) and 5-iodo-(2-deoxy-2-fluoro-?-
D-arabinofuranosyl)uracil (FIAU, 2D) (Fig. 2) are both good
substrates for TK1as well as for TK2[7]. Therefore, there
have been several investigations focusing on radiolabeled
2-arabino fluorinated nucleoside analogs for use in in vivo
imaging agents of cell proliferation. Recently, 5-[11C]
methyl-(2-deoxy-2-fluoro-?-D-arabinofuranosyl)uracil ([11C]
FMAU) and 5-methyl-(2-deoxy-2-[18F]fluoro-?-D-arabino-
furanosyl)uracil ([18F]FMAU) have been investigated as im-
aging agents of DNA synthesis [4,16]. Moreover, the 2?-sub-
stituted 5-bromo-2?-deoxyuridine analog 5-[76Br]bromo-2?-
fluoro-2?-deoxyuridine ([76Br]BFU) was also proven to be a
stable in vivo tracer of cell proliferation [3,11]. However, as
pointed out in the literature, the naming of this compound and
the description of its structure were initially incorrect, and the
compound’s 2?-fluoro substitution is actually in the arabino-
configuration (up) [76Br]FBAU [8]. At any rate, the 2-arabino
fluorination of nucleosides is also thought to be a promising
approach for obtaining an imaging agent of cell proliferation.
Consequently, 2-arabino fluorination combined with the 4?-
thio nucleosides appears to overcome the problems associated
with [125I]ITdU (1B). Furthermore, this approach allows for
the introduction of a radiohalogen in the base-5 position, and,
alternatively, for fluorine-18 in the sugar 2-arabino positions.
Therefore, we evaluated 5-[125I]iodo-(2-deoxy-2-fluoro-4-
thio-?-D-arabinofuranosyl)uracil ([125I]FITAU, 1E) (Fig. 1) as
a2-arabinofluorinated4?-thioderivativeforuseintheimaging
of cell proliferation. We also tested 5-[125I]iodo-(2-deoxy-2-
fluoro-?-D-arabinofuranosyl)uracil ([125I]FIAU, 1D) (Fig. 1)
as a reference compound.
The abovementioned 2-arabino fluorinated and 4?-thio
approaches would increase the stability of the C-N glyco-
sidic bond. In a similar project, we introduced a new ap-
proach for obtaining a metabolically stable in vivo imaging
agents of cell proliferation. The cleavage of the C-N glyco-
sidic bond of nucleosides has been considered to be an SN1
reaction as well as an SN2 reaction [13,14]. The SN2 reac-
tion requires that a nucleophilic attack by phosphate ions
take place on the anomeric carbon. Therefore, it is reason-
able to test the 1? -substituted derivative in terms of its
participation in an SN2 reaction. Assuming that the 1?-
methyl fits into the active site of the phosphorylase, the
addition of a 1?-methyl would prohibit an SN2 reaction, but
would enhance an SN1 reaction. However, we could not
prepare the 5-iodo-1?-methyl-2?-deoxyuridine due to the
rapid degradation of the C-N glycosidic bond. A compound
with a stable C-N glycosidic bond was obtained when an
electron-withdrawing substitutive group, bromine, was in-
troduced at the 2-arabino position. Thus, we tested the
1?-methylatedand2-arabino-brominatedcompound
Fig. 1. Chemical structures of 5-[125I]iodonucleosides. 5-[125I]iodo-2?-
deoxyuridine ([125I]IUdR,
1A),
([125I]ITdU,
1B), 5-[125I]iodo-(4-thio-2-?-D-arabinofuranosyl)uracil
([125I]ITAU, 1C), 5-[125I]iodo-(2-deoxy-2-fluoro-?-D-arabinofuranosyl)u-
racil ([125I]FIAU,
1D), 5-[125I]iodo-(2-deoxy-2-fluoro-4-thio-?-D-
arabinofuranosyl)uracil([125I]FITAU, 1E), 5-[125I]iodo-1-methyl-(2-de-
oxy-2-bromo-?-D-arabinofuranosyl)uracil ([125I]IMBAU), 1F).
5-[125I]iodo-4?-thio-2?-deoxyuridine
Fig. 2. Synthetic pathway for the preparation of radiolabeled FIAU (2D), FITAU (2E), and IMBAU (2F) by destannylation.
688
J. Toyohara et al. / Nuclear Medicine and Biology 30 (2003) 687–696

Page 3

5-[125I]iodo-1-methyl-(2-deoxy-2-bromo-?-D-arabino-
furanosyl)uracil ([125I]IMBAU, 1F) (Fig. 1).
The goal of this investigation was to test the potential of
these new approaches; accordingly, we compared the bio-
logical data with results from previously published studies
of 4?-thio nucleoside analogs and from the reference com-
pound [125I]FIAU (1D) (Fig. 1). The present data provide
new information regarding the most suitable approaches for
developing radioiodinated nucleoside analogs.
2. Materials and methods
2.1. Radiochemicals
Carrier-free sodium [125I]iodide and 5-[125I]iodo-2?-de-
oxyuridine ([125I]IUdR, 1A, 74 TBq/mmol) were purchased
from Amersham Biosciences (Buckinghamshire, UK). The
specific activity of 5-[125I]iodo-2?-deoxyuridine (1A) was
adjusted to 37 MBq/?mol with cold 5-iodo-2?-deoxyuridine
purchased from Aldrich (Milwaukee, WI). 5-[125I]iodo-4?-
thio-2?-deoxyuridine ([125I]ITdU, 1B) and 5-[125I]iodo-(4-
thio-2-?-D-arabinofuranosyl)uracil ([125I]ITAU, 1C) (Fig.
1) were synthesized as reported previously [18].
2.2. Synthesis of 5-trimethylstannyl-(2-deoxy-2-fluoro-?-D-
arabinofuranosyl)uracil (3D)
The method of synthesis of 5-trimethylstannyl-(2-deoxy-
2-fluoro-?-D-arabinofuranosyl)uracil (3D) (Fig. 2) was fun-
damentally identical to that which was previously reported,
with the exception that FIAU (2D) was used as the starting
material to produce 5-trimethylstannyl-(2-deoxy-2-fluoro-
?-D-arabinofuranosyl)uracil (3D) [19]. 5-Iodo-(2-deoxy-2-
fluoro-?-D-arabinofuranosyl)uracil (FIAU, 2D) was synthe-
sized according to a previously described method [20]. The
FIAU (2D, 5.0 mg, 0.013 mmol), bis(trimethyltin) (20.5
mg,0.063 mmol)and
dium(II)dichloride (6.2 mg) were dissolved in anhydrous
1,4-dioxane (3 mL) in an argon atmosphere. The mixture
was heated under conditions of reflux for 2 h, and then was
concentrated under reduced pressure. The residue was pu-
rified by silica gel thin layer chromatography (CHCl3/
MeOH ? 6/1) to produce 5-trimethylstannyl-(2-deoxy-2-
fluoro-?-D-arabinofuranosyl)uracil (3D, 3.6 mg, 66%).1H
bis(triphenylphosphine)palla-
NMR (500MHz, CD3OD), d 0.25 (s, 9H, (CH3)3Sn), 3.72
(dd, J?5.0, 12.0 Hz, 1H, H-5?), 3.79-3.91 (m, H-4?), 4.33
(ddd, J?3.0, 5.0, 18.5 Hz, 1H, H-3?), 5.02 (td, J?4.0, 53.0
Hz, 1H, H-2?), 6.25 (dd, J?4.5, 16.0 Hz, 1H, H-1?), 7.56 (s,
1H, H-5).
2.3. Synthesis of 5-trimethylstannyl-(2-deoxy-2-fluoro-4-
thio-?-D-arabinofuranosyl)uracil (3E)
5-Iodo-(2-deoxy-2-fluoro-4-thio-?-D-arabinofuranosy-
l)uracil (FITAU, 2E) (Fig. 2) was synthesized according to
a previously described method [22]. FITAU (2E) was then
used as the starting material to give 5-trimethylstannyl-(2-
deoxy-2-fluoro-4-thio-?-D-arabinofuranosyl)uracil (3E) by
the following procedure. FITAU (3E, 5.0 mg, 0.013 mmol),
bis(trimethyltin) (16.9 mg, 0.052 mmol) and bis(triph-
enylphosphine)palladium(II)dichloride (6.0 mg) were dis-
solved in anhydrous 1,4-dioxane (3 mL) under an argon
atmosphere. After the mixture was heated at reflux for 3.5 h
it was concentrated under reduced pressure. The residue was
purified by silica gel thin layer chromatography (CHCl3/
MeOH ? 6/1) to produce 5-trimethylstannyl-(2-deoxy-2-
fluoro-4-thio-?-D-arabinofuranosyl)uracil
35%).
(CH3)3Sn), 3.61-3.68 (m, 1H, H-5?), 3.80-3.81 (m, H-4?),
4.37 (td, J?6.0, 12.0 Hz, 1H, H-3?), 4.97 (td, J?5.5, 49.0
Hz, 1H, H-2?), 6.46 (dd, J?5.5, 11.5 Hz, 1H H-1?), 7.99 (s,
1H, H-5).
(3E,
1.9 mg,
1H NMR (500MHz, CD3OD), d 0.26 (s, 9H,
2.4. Synthesis of 5-iodo-1-methyl-(2-deoxy-2-bromo-?-D-
arabinofuranosyl)uracil (2F)
1-[2-bromo-3,5-bis-O-(tert-butyldimethylsilyl)-2-deoxy-
1-C-methyl-?-D-arabinofuranosyl]uracil (4) (Fig. 3) was
prepared according to a previously described method [6].
Bu4NF (1M solution in THF, 6 mL, 6.0 mmol) was added
to a solution of 4 (1.96 g, 4.17 mmol) in THF (20 mL). After
stirring for 1.5 h at room temperature, the mixture was
reacted with Ac2O (1.2 mL, 12.7 mmol) for 1 h at room
temperature. The reaction mixture was evaporated and then
partitioned between saturated aqueous NaHCO3and CHCl3.
Silica gel column chromatography (CHCl3/MeOH ? 50/1)
of the organic layer gave 5 (1.29 g, 76%). Iodination of 5
was carried out according to a previously described method
[1]. A mixture of 5 (357 mg, 0.88 mmol), iodine (134 mg),
Fig. 3. Synthetic pathway for the preparation of 5-iodo-1-methyl-(2-deoxy-2-bromo-?-D-arabinofuranosyl)uracil (IMBAU, 2F).
689
J. Toyohara et al. / Nuclear Medicine and Biology 30 (2003) 687–696

Page 4

and (NH4)2Ce(NO3)6(240 mg) in CH3CN (10 mL) was
refluxed for 1.5 h. The solvent was evaporated and the
residue was partitioned between EtOAc and 5% NaHSO3in
H2O. Silica gel column chromatography (hexane/EtOAc ?
2/1) of the organic layer gave 6 (345 mg, 74%). Compound
6 (345 mg) was treated with NH3/MeOH overnight at 0 °C.
The reaction mixture was evaporated and then purified by
silica gel column chromatography (3% MeOH in CHCl3).
This reaction sequence gave 2F (220 mg, 75%).1H NMR
(DMSO-d6,after addition of D2O) ? 1.83 (3H, s, 1?-CH3),
3.58-3.65 (2H, m, H-5?), 3.92-3.96 (1H, m, H-4?), 4.34 (1H,
d, J? 3.1 Hz, H-3?), 4.61 (1H, s, H-2?), 8.23 (1H, s, H-6);
FAB-MS m/z 447 and 449 (M??H). Anal. Calcd for
C10H12BrIN2O5: C, 26.87; H, 2.71; N, 6.27. Found: C,
27.15; H, 2.92; N, 6.34.
2.5. Synthesis of 5-trimethytin-1-methyl-(2-deoxy-2-bromo-
?-D-arabinofuranosyl)uracil (3F)
IMBAU (2F) was used as starting material to produce
5-trimethytin-1-methyl-(2-deoxy-2-bromo-?-D-arabino-
furanosyl)uracil (3F) by the following procedure. IMBAU
(2F, 4.9 mg, 0.011 mmol), bis(trimethyltin) (16.0 mg, 0.049
mmol) and bis(triphenylphosphin)palladium(II)dichloride
(5 mg) were dissolved in anhydrous 1,4-dioxane (3 mL) in
an argon atmosphere. After heating the mixture at reflux for
2.5 h, it was concentrated under reduced pressure. The
residue was purified by silica gel thin layer chromatography
(CHCl3/MeOH ? 6/1) to produce 5-trimethylstannyl-1-
methyl-(2-deoxy-2-bromo-?-D-arabinofuranosyl)uracil (3F,
3.8 mg, 72%).1H NMR (500MHz, CD3OD) d 0.25 (s, 9H,
(CH3)3Sn), 1.95 (s, 3H, 1?-CH3), 3.61-3.70 (m, 2H, H-5?),
4.08-4.11 (m, 1H, H-4?), 4.53 (d, J?3.0 Hz, 1H, H-5?), 4.79
(s, 1H, H-2?), 7.75 (s, 1H, H-5).
2.6. Synthesis of 5-radioiodinated nucleoside analogs
Radiolabeling of a trimethylstannyl-precursor was per-
formed using fundamentally the same method as that which
had previously been reported (Fig. 2) [18]. The 4?-thio
derivatives, including [125I]ITdU (1B), [125I]ITAU (1C),
and [125I]FITAU (1E) compounds, possess a sulfide that can
be oxidized by a strong oxidant. Preliminarily, we observed
the formation of an unknown byproduct (possibly S-oxide)
under strong oxidant conditions for the de-stannylation re-
action of 4?-thio derivatives. In order to eliminate this by-
product, we selected iodine as a mild oxidant. Regrettably,
this reaction dramatically lowered the specific activity of the
final radiolabeled compound due to the resultant isotopic
dilution. However, in this screening study, the specific ac-
tivity sufficed for the evaluation. Briefly, [125I]NaI and
iodine (in the same molar concentration as that of the
trimetylstannyl-precursor) were added to a reaction vial
containing MeOH. The trimethylstannyl-precursor solution
was added to the vial and allowed to stand for 2 h at room
temperature. The labeling reaction was terminated by the
addition of sodium metabisulfite, and the MeOH was evap-
orated. The labeled compound was purified by means of
high-performance liquid chromatography (HPLC) on an
ODS-80™ column (4.5?250 mm; Toyo-soda, Co., Ltd.,
Tokyo, Japan). A MeOH/H2O/TFA ? 30/70/0.1 solution
was used as the mobile phase, under a flow rate of 0.8
mL/min. In the case of [125I]IMBAU (1F), MeOH/H2O/
TFA ? 40/60/0.1 solution was also used as the mobile
phase. The radioactive fraction, eluted with a retention time
corresponding to that of the authentic sample was collected.
The HPLC solvent of the collected fraction was then evap-
orated by heating, and the purified radioiodinated com-
pound was dissolved using sterile saline. In the case of
[125I]IMBAU (1F), the HPLC solvent of the collected frac-
tion was evaporated at room temperature ([125I]IMBAU
(1F) was decomposed by heating). For calculations of spe-
cific activity, standard solutions of cold compounds were
used to compare the area under ultraviolet (UV) absorption.
The labeling yield of [125I]FIAU (1D) was 12% (9.5 MBq),
with a radiochemical purity of ?99%. The specific activity
of [125I]FIAU (1D) was calculated as approximately 39
MBq/?mol. The labeling yield of [125I]FITAU (1E) was
7.8% (3.5 MBq), with a radiochemical purity of ?99%. The
specific activity of [125I]FITAU (1E) was calculated as
approximately 702 MBq/?mol. The labeling yield of
[125I]IMBAU (1F) was 13% (8.3 MBq), with a radiochem-
ical purity of ?99%. The specific activity of [125I]IMBAU
(1F) was calculated as approximately 239 MBq/?mol.
2.7. Cytosolic thymidine kinase (TK1) assay
Cytosolic thymidine kinase (TK1) activity was assayed
according to a previously reported method [18]. In order to
eliminate the mitochondria, the source of TK2,and deox-
yguanosine kinase (dGK), we used a 105,000 g supernatant
of Lewis lung carcinoma (LL/2) cell lysate as the enzyme
source of cytosolic thymidine kinase (TK1). The 105,000 g
supernatant also included deoxycytidine kinase (dCK).
Since all of the compounds tested in this article possessed
5-iodouracil as a base and a deoxyribose analog as a sugar,
dCK would be not be expected to have any effect on the
phosphorylation of these compounds.
2.8. Recombinant thymidine phosphorylase assay
The experimental procedure used for the recombinant
thymidine phosphorylase assay was almost the same as a
previously reported method, with a slight modification [18].
Briefly, the reaction mixture for the assay of recombinant
thymidine phosphorylase contained 168 mM potassium
phosphate buffer (pH 7.6), 2 nmol of nucleoside, and sev-
eral units (from 0.9 munits to 9 units) of recombinant
thymidine phosphorylase (Sigma Chemical Co., St. Louis,
MO). The nucleoside degradation reaction catalyzed by
thymidine phosphorylase was carried out at 25 °C for 30
690
J. Toyohara et al. / Nuclear Medicine and Biology 30 (2003) 687–696

Page 5

min and then was terminated by heating the mixture at 100
°C for 3 min. As mentioned above, [125I]IMBAU (1F) was
unstable when heat-treated; hence, the reaction was stopped
by ice treatment. The reaction mixture was then subjected to
centrifugal separation, and the 1-?L supernatant of the re-
action mixture was applied to a reversed phase silica gel
plate (RP-18 F254, Merck, Schuchardt, Germany) contain-
ing standards for the nucleoside. The resultant metabolite
5-[125I]iodouracil ([125I]IU) was then separated by TLC
using a solvent system consisting of MeOH/H2O ? 3/7. In
the case of [125I]IUdR (1A), [125I]ITdU (1B), and
[125I]ITAU (1C), a silica gel plate (Sil60 F254, Merck,
Schuchardt, Germany) was used, and the plate was devel-
oped using a mixture of CHCl3/isopropyl alcohol ? 3/1.
The spot of 5-iodouracil (IU) was identified under UV
irradiation by comparison with the standards. The radioac-
tivity on the plate was measured and quantified by means of
a bio-imaging analyzer (BAS-1500, Fuji Photo Film Co.,
Tokyo, Japan). Data were expressed as pmol of formed IU
per unit per 30 min.
2.9. Cell uptake study with Lewis lung carcinoma (LL/2),
L-M, and L-M(TK?) cell lines
In order to determine the TK1-dependent incorporation
of the nucleoside, we compared the cell uptake of nucleo-
sides between the L-M parent cell line and its TK mutant
cell line, L-M(TK?) [9]. L-M(TK?) cells lack TK1,but
contain a genetically distinct mitochondrial isozyme (TK2)
[2]. Therefore, in this assay, the TK1-specific compound
might have accumulated to a large extent in the L-M cells
rather than in the L-M(TK?) cells. The assay conditions
were identical to those of a previously described protocol
[18].
2.10. Biodistribution in normal mice
The experimental procedure used to test for biodistribu-
tion in normal mice was identical to a previously reported
procedure [18].
3. Results
3.1. Relative phosphorylation of iodonucleosides by TK1
The phosphorylation potentials of [125I]FIAU (1D),
[125I]FITAU (1E), and [125I]IMBAU (1F), as well as the
summarized results regarding the six compounds are shown
in Table 1. The in vitro TK1affinity of [125I]IUdR (1A),
[125I]ITdU (1B), and [125I]ITAU (1C) was reported with our
previously published data [18]. The TK1assay showed that
the 2-arabino fluorinated analog, [125I]FIAU (1D), pos-
sessed 16% of the phosphorylation potential of [125I]IUdR
(1A). The combination of 2-arabino-fluorine and the 4?-
thio-substituted compound [125I]FITAU (1E) showed re-
duced phosphorylation potential (?0.01). Therefore, this
combination approach did not produce a suitable substrate
for TK1.In addition, 1?-methyl and 2-arabino brominated
[125I]IMBAU (1F) did not show any phosphorylation, and
thus was also unsuitable as a substrate for TK1. Only 2?-
deoxy 4?-thio derivatives [125I]ITdU (1B) showed effective
phosphorylation potential (i.e., half of the [125I]IUdR (1A)
activity).
3.2. Susceptibility of iodonucleosides to glycosidic bond
cleavage
The catabolism of [125I]FIAU (1D), [125I]FITAU (1E),
and [125I]IMBAU (1F) due to recombinant thymidine phos-
phoryase, as well as the summarized results from the six
compounds, are shown in Table 2. The in vitro thymidine
phosphorylasecatabolism
[125I]IUdR (1A), [125I]ITdU (1B), and [125I]ITAU (1C)
were obtained from our previously published data [18]. The
recombinant thymidine phosphorylase assay showed that all
iodonucleosides, with the exception of [125I]IUdR (1A),
were highly resistant to the C-N glycosidic bond cleavage
reaction by recombinant thymidine phosphorylase. The
2-arabino fluorine-containing derivatives ([125I]FIAU (1D)
and [125I]FIATU (1E)) were more highly resistant to the
C-N glycosidic bond cleavage reaction than were the 2?-
deoxy 4?-thio derivatives ([125I]ITdU, 1B) and the 2-arabino
hydroxy 4?-thio derivatives ([125I]ITAU, 1C). The combi-
reactiondata regarding
Table 1
Relative phosphorylation of iodonucleosides by TK1
Compd.R1
R2
R3
pmol phosphorylated* Relative potency
IUdR
ITdU
ITAU
FIAU
FITAU
IMBAU
H
H
H
H
H
CH3
H
H
OH
F
F
Br
O
S
S
0
S
0
1620.8 ? 123.4
802.4 ? 67.9
13.3 ? 0.6
270.0 ? 25.2
0.3 ? 0.0
0.0
1.00
0.50
?0.01
0.16
?0.01
0.0
* pmol phosphorylated: pmol / mg protein / h
The data of IUdR, ITdU and ITAU were from our previous report [18].
The relative phosphorylation potency, which is normalized to IUdR as 1.0.
691
J. Toyohara et al. / Nuclear Medicine and Biology 30 (2003) 687–696

Page 6

nations of the 2-arabino bromine and the 1?-methyl-substi-
tuted compound, [125I]IMBAU (1F), were entirely resistant
to the thymidine phosphorylase reaction.
3.3. Cell uptake study with lewis lung carcinoma (LL/2),
L-M, and L-M(TK?) cell lines
The comparative cell uptake of [125I]FIAU (1D), [125I]F-
ITAU (1E), and [125I]IMBAU (1F), as well as the summa-
rized results from the six compounds, are shown in Table 3.
The comparative cell uptake studies of [125I]IUdR (1A),
[125I]ITdU (1B), and [125I]ITAU (1C) were taken from our
previously published data [18]. The %S-phase fractions of
each cell line were the same as those described in our
previous report [18]. A comparison between L-M and L-
M(TK?) uptake data showed that the 2-arabino fluorine–
containing compounds, [125I]FIAU (1D) and [125I]FITAU
(1E), showed significant but slight TK1-dependent cell up-
take ([125I]FIAU (1D) and [125I]FITAU (1E): p?0.01; Stu-
dent’s t-test). The [125I]IMBAU (1F) did not show any
significant TK1-dependent cell uptake, as was expected
from the results of the TK1assay. Both [125I]IUdR (1A) and
[125I]ITdU (1B) exhibited significant TK1-dependent cell
uptake (p?0.0005; Student’s t-test). The relative uptake
ratio of the Lewis lung carcinoma (LL/2) cell line compared
with that of the L-M cell line showed that [125I]FIAU (1D)
was incorporated at a 1.6-fold greater rate into the tumor
cells. However, [125I]ITdU (1E) was incorporated most dra-
matically into rapidly proliferating tumor cells, with a ratio
of 20.8.
3.4. Biodistribution in normal mice
The biodistributions of [125I]IUdR (1A), [125I]ITdU
(1B), [125I]ITAU (1C), [125I]FIAU (1D), [125I]FITAU (1E),
and [125I]IMBAU (1F) are shown in Figs. 4A–F . In order
to compare the target-specific uptakes of [125I]IUdR (1A),
[125I]ITdU (1B), and [125I]FIAU (1D), the target-to-blood
ratios at 18 h post-injection were also calculated (Table 4).
In contrast to our previous study, a thyroid block was
performed by 1.0-mL intravenous injections of a 0.9% NaI
solution (30 mg/kg NaI) administered daily for 3 consecu-
tive days before the beginning of the experiment. The
[125I]FIAU (1D) displayed a significantly higher and more
retained uptake in the rapidly proliferating tissues (i.e., the
thymus and small intestine) than in the non-proliferating
tissues (i.e., the lung and liver) (Fig. 4D). However, [125I]F-
ITAU (1E) and [125I]IMBAU (1F) did not show any sig-
nificant proliferating tissue uptake, as was to be expected,
based on the in vitro TK1assay (Figs. 4E, 4F). As shown in
Table 2
Susceptibility of iodonucleosides to glycosidic bond cleavage
Compd.R1
R2
R3
pmol IU formed* Relative IU formation
IUdR
ITdU
ITAU
FIAU
FITAU
IMBAU
H
H
H
H
H
CH3
H
H
OH
F
F
Br
O
S
S
O
S
O
138606.2 ? 14902.3
3778.7 ? 692.0
514.8 ? 367.0
0.3 ? 0.3
0.5 ? 0.1
0.0
1.00
0.03
?0.01
?0.01
?0.01
0.0
* pmol IU formed: pmol / unit / 30 min
The data of IUdR, ITdU and ITAU were from our previous report [18].
The relative phosphorylation potency, which is normalized to IUdR as 1.0.
Table 3
Cell uptake study with Lewis lung carcinoma (LL/2), L-M and L-M(TK?) cell lines
Compd.R1
R2
R3
Tracer uptake* (Relative uptake ratio)
LL/2 L-ML-M(TK?)
IUdR
ITdU
ITAU
FIAU
FIATU
IMBAU
H
H
H
H
H
CH3
H
H
OH
F
F
Br
O
S
S
O
S
O
539.7 ? 81.7 (6.9)
226.7 ? 13.1 (20.8)
1.1 ? 0.0 (0.6)
17.4 ? 0.3 (1.6)
0.2 ? 0.0 (0.7)
0.3 ? 0.0 (0.5)
77.8 ? 7.4†(2.8)
10.9 ? 1.5†(2.8)
1.7 ? 0.3§(1.4)
11.0 ? 1.8‡(1.5)
0.3 ? 0.1‡(1.6)
0.6 ? 0.1 (1.2)
27.9 ? 2.9
3.9 ? 0.6
1.2 ? 0.2
7.3 ? 0.4
0.2 ? 0.1
0.5 ? 0.1
* Tracer uptake: pmol / mg protein /h
†p?0.0005 compared to L-M(TK?).
‡p?0.01 compared to L-M(TK?)
§p?0.05 compared to L-M(TK?) (Student’s t-test).
Data in parenthesis in LL/2 row and L-M row indicate relative uptake ratios of LL/2 compared with those of L-M cells and relative uptake ratios of L-M
compared with those of L-M (TK?), respectively.
The data of IUdR, ITdU and ITAU were from our previous report [18].
692
J. Toyohara et al. / Nuclear Medicine and Biology 30 (2003) 687–696

Page 7

Fig. 4. Clearance of iodonucleosides from proliferating and non-proliferating tissues in normal mice. A: [125I]IUdR (1A). B: [125I]ITdU (1B), C: [125I]ITAU
(1C). D: [125I]FIAU (1D). E: [125I]FITAU (1E). F: [125I]IMBAU (1F). Data represent mean percentage injected dose per gram ? s.d. (n?3).
693
J. Toyohara et al. / Nuclear Medicine and Biology 30 (2003) 687–696

Page 8

Table 4, [125I]ITdU (1B) tended to display a preferable
target (i.e., the femur and small intestine)-to-blood ratio
than did [125I]IUdR (1A) and [125I]FIAU (1D) at 18 h
post-injection.
4. Discussion
In this study, we evaluated the fundamental biological
characteristics of [125I]FITAU (1E) and [125I]IMBAU (1F).
We compared our results with those obtained using a ref-
erence compound, [125I]FIAU (1D), and also with previ-
ously published results obtained with [125I]IUdR (1A),
[125I]ITdU (1B), and [125I]ITAU (1C). The present results
demonstrate that all of the sugar modifications of these
iodonucleosides successfully introduced C-N glycosidic
bond stability. However, these approaches also reduced the
affinity to cytosolic thymidine kinase (TK1), compared to its
non-modified counterpart, [125I]IUdR (1A). The data in this
investigation revealed that relatively small structural
changes in 5-iodo-2?-deoxyuridine derivatives will pro-
foundly affect their biochemical activity.
The C-N glycosidic bond of nucleosides is more easily
cleaved than the C-N glycosidic bond of any other organic
compound, because the generated cyclic oxocarbenium cat-
ion is stabilized by the electron-donating resonance effect of
the 4?-oxygen of the furanose. Therefore, an electron-with-
drawing group substitution at the 2?-position, which is close
to the oxocarbenium cation, will render the cyclic oxocar-
benium cation unstable. Hence, the introduction of an elec-
tron-withdrawing group would be expected to enhance the
stability of the C-N glycosidic bond of nucleosides [17].
Moreover, the fluorine at the 2-arabino position of nucleo-
sides is thought to lock the sugar-base bond into the “anti”
conformation, which is furthermore thought to increase re-
sistance to the cleavage of the glycosyl bond [8,15]. The
conformation analysis of 4?-thio-2?-deoxythymidine re-
vealed that this compound did not indicate a locked bond,
although it did indicate a predominantly anti-conformation
of the thymine base [10]. In contrast to the mechanism of
the 2-arabino derivatives, the mechanism of the stabilization
effect of 4?-thio substitution is thought to be due to the
lesser stability of its cyclic thiocarbenium ion, as compared
to that of the corresponding oxocarbenium cation [18].
Therefore, we hypothesized that the introduction of 2-ar-
abino-fluorination into the 4?-thio derivatives would en-
hance C-N glycosidic bond stability. In support of this
consideration, [125I]FITAU (1E) showed great stability
against the catabolic reaction provoked by recombinant thy-
midine phosphorylase. [125I]FITAU (1E) showed signifi-
cantly slower catabolic kinetics than its 2?-deoxy counter-
part, [125I]ITdU (1B) (p?0.001; Student’s t-test). Moreover,
the kinetics of [125I]FITAU (1E) were slower than those of
[125I]ITAU (1C), which has a larger hydroxyl group in the
2-arabino position; however, this difference was not statis-
tically significant (p?0.07; Student’s t-test). Based on these
results, we concluded that 2-arabino fluorination would be
more effective at enhancing C-N glycosidic bond stability
than would be achieved by the 2?-deoxy 4?-thionucleosides.
Nonetheless, the catabolic kinetics of [125I]FIATU (1E) in
reaction with thymidine phosphorylase did not significantly
differ from those of its 4?-oxy counterpart, [125I]FIAU (1D).
We also demonstrated that 1?-methyl and 2-arabino-bromi-
nated [125I]IMBAU (1F) did not undergo a degradation
reaction. However, the catabolic kinetics of [125I]IMBAU
(1F) in reaction with thymidine phosphorylase were not
significantly different from those of [125I]FIAU (1D). These
results suggest that 2-arabino fluorination is sufficient
enough for obtaining a metabolically stable compound.
The structural elements of IMBAU (2F) would provide
an extremely stable intermediate under “chemical” SN1 con-
ditions, the cation being both tertiary and “oxo”-stabilized.
These same elements provide considerable hindrance to a
“chemical” SN2 nucleophilic attack. Therefore, the meta-
bolic stability of IMBAU (2F) against thymidine phosphor-
ylase might suggest that there is no SN1 character to the
mechanism of the glycosidic bond cleavage and the subse-
quent phosphorylation. However, the effect of the 1?-methyl
group alone (i.e., without 2?-bromo) could not be ascer-
tained under the present experimental conditions, and there
are no data regarding whether or not the C-1 methyl group
has an effect on the interaction between the substrate and
the enzyme. Therefore, we were unable to conclude that
there was no SN1 character to the glycosidic bond cleavage
and phosphorylation mechanisms in these cases.
In spite of their extensive resistance to catabolic reac-
tions, [125I]FITAU (1E) and [125I]IMBAU (1F) did not
show any phosphorylation by TK1. The in vivo biodistribu-
tion data also showed that these compounds did not accu-
mulate in the rapidly proliferating tissues. Therefore, these
compounds would not be suitable as in vivo imaging agents
of cell proliferation. However, if these compounds were to
be phosphorylated by virus-encoded thymidine kinase, they
could then be rendered useful as a reporter probe for the
noninvasive imaging of transgene expression. In particular,
[125I]FITAU (1E) has proven to be a strong antiviral agent
against herpes simplex virus type 1 strain [22]. The effective
dose of FITAU (2E) is 10 times lower than that of acyclovir,
which is a clinically approved anti-viral agent. An advan-
Table 4
Tissue to blood ratio of iodonucleosides at 18 h post injection (n?3)
IUdR FIAUITdU
Femur
Spleen
Small intestine
Thymus
Muscle
Liver
Lung
19.4 ? 4.0
20.5 ? 9.6
61.7 ? 16.9
78.2 ? 18.3
1.1 ? 0.5
1.4 ? 0.3
3.2 ? 1.2
8.2 ? 3.6
5.5 ? 2.4
113.7 ? 34.1
137.2 ? 97.2
1.5 ? 0.1
1.5 ? 0.6
2.4 ? 1.1
24.4 ? 2.2
26.0 ? 12.3
201.9 ? 89.2
137.9 ? 101.7
1.6 ? 0.6
2.9 ? 1.1
3.4 ? 0.6
Data are expressed as mean ?SD.
694
J. Toyohara et al. / Nuclear Medicine and Biology 30 (2003) 687–696

Page 9

tage of [125I]FITAU (1E) as a reporter probe lies in is its
low abdominal background activity and its small amount of
uptake in proliferating tissues. Therefore, we are currently
evaluating the potential of [125I]FITAU (1E) as a reporter
probe for herpes simplex virus thymidine kinase gene ex-
pression.
The in vivo biodistribution of each nucleoside also cor-
responded closely to its in vitro biological data. The corre-
lation between thymus uptake at 18 h and the phosphory-
lation potential of each nucleoside was significant (r?0.95,
p?0.005; Pearson’s correlation coefficient). The correlation
between the uptake in the small intestine at 18 h and the
phosphorylation potential of each nucleoside was higher
than that of uptake in the thymus (r?0.98, p?0.001; Pear-
son’s correlation coefficient). Regardless of their relatively
higherstabilities,the target-to-background
[125I]FIAU (1D) was not greater than that of [125I]ITdU
(1B). This finding might be accounted for by the TK1
phosphorylation potential of [125I]FIAU (1D), which was
lower thanthat of[125I]ITdU
[125I]ITdU (1B) showed a better proliferating tissue-to-
background ratio than did [125I]IUdR (1A) and [125I]FIAU
(1D) at 18 h post-injection. Based on the above results, the
in vivo proliferation-imaging potential of nucleosides might
be estimated by their in vitro affinity for TK1and their C-N
glycosidic bond stability. However, in the present study, we
did not examine the nucleoside transport activity of each
iodonucleoside, a step that would also be important in the
metabolic pathway of pyrimidine. Nucleoside transport has
a large impact on cellular kinetics and biodistribution; there-
fore, further investigation of this issue will be needed in the
future.
ratioof
(1B). Consequently,
5. Conclusion
This study determined the fundamental biological char-
acteristics of 5-[125I]iodo-(2-deoxy-2-fluoro-4-thio-?-D-ara-
binofuranosyl)uracil ([125I]FITAU, 1E) and 5-[125I]iodo-1-
methyl-(2-deoxy-2-bromo-?-D-arabinofuranosyl)uracil ([125I]
IMBAU, 1F). Based on our results, we concluded that [125I]
FITAU (1E) and [125I]IMBAU (1F) would not be effective as
imaging agents of cell proliferation. The present findings also
suggested that the 4?-thio-2?-deoxy derivative, 5-[125I]iodo-4?-
thio-2?-deoxyuridine ([125I]ITdU, 1B), is a more sensitive and
specific tracer for imaging tissue proliferation than is 4?-
oxo-2-arabino-fluorinated 5-[125I]iodo-(2-fluoro-2-?-D-
arabinofuranosyl)uracil ([125I]FIAU, 1D). These biolog-
ical data may provide useful information for the
determination of suitable approaches for developing ra-
dioiodinated nucleoside analogs.
References
[1] Asakura J, Robins MJ. Cerium(IV)-medicated halogenation at C-5 of
uracil derivatives. J Org Chem 1990;55:4928–33.
[2] Berk AJ, Clayton DA. A genetically distinct thymidine kinase in
mammalian mitochondria. Exclusive labeling of mitochondrial de-
oxyribonucleic acid. J Biol Chem 1973;248:2722–9.
[3] Borbath I, Vre ´goire V, Bergstro ¨m M, Laryea D, Långstro ¨m B, Pau-
wels S. Use of 5-[76Br]bromo-2?-fluoro-2?-deoxyuridine as a ligand
for tumor proliferation: validation in an animal tumor model. Eur
J Nucl Med 2000;29:19–27.
[4] Conti PS, Bading JR, Alauddin MM, Fissekis JD, Berenji B, Hayles
S, Jadvar H. Tumor imaging with 2?-fluoro-5-11C-methyl-1-?-D-ar-
abinofuranosyluracil (11C-FMAU): Initial clinical studies. J Nucl
Med 2002;43(suppl 5):26.
[5] Wertel LW, Boder GB, Kroin JS, Rinzel SM, Poore GA, Todd GC,
Grindey GB. Evaluation of the antitumor activity of gemcitabine
(2?,2?-difluoro-2?-deoxycytidine). Cancer Res 1990;50:4417–22.
[6] Itoh Y, Haraguchi Y, Tanaka H, Gen E, Miyasaka T. Divergent
and stereocontrolled approach to the synthesis of uracil nucleo-
sides branched at the anomeric position. J Org Chem 1995;60:
656–62.
[7] Johansson NG, Eriksson S. Structure-activity relationships for phos-
phorylation of nucleoside analogs to monophosphates by nucleoside
kinases. Acta Biochemica Polonica 1996;43:143–60.
[8] Kao C-HK, Waki A, Sassaman MB, Jagoda EM, Szajek LP, Ravasi
L, Shimoji K, Eckelman WC. Evaluation of [76Br]FBAU 3?,5?-diben-
zonate as a lipopholic prodrug for brain imaging. Nucl Med Biol
2002;29:527–35.
[9] Kit S, Dubbes DR, Piekarski LJ, Hsu TC. Deletion of thymidine
kinase activity from L cells resistant to bromodeoxyuridine. Exp Cell
Res 1963;31:297–312.
[10] Koole LH, Plavec J, Liu H, Vincent BR, Dyson MR, Coe PL, Walker
RT, Hardy GW, Rahim SG, Chattopadhyaya J. Conformation of two
4?-thio-2?-deoxynucleoside analogs studied by 500-MHz1H NMR
spectroscopy and x-ray crystallography. J Am Chem Soc 1992;114:
9936–43.
[11] Lu L, Bergstro ¨m M, Fasth K-J, Långstro ¨m B. Synthesis of [76Br]bro-
mofluorouridine and its validation with regard to uptake, DNA incor-
poration, and excretion modulation in rats. J Nucl Med 2000;41:
1746–52.
[12] Philips FS, Feinberg A, Chou TC, Vidal PM, Su TL, Watanabe KA,
Fox JJ. Distribution, metabolism, and excretion of 1-(2-fluoro-2-
deoxy-?-D-arabinofuranosyl)thymine and 1-(2-fluoro-2-deoxy-?-D-
arabinofuranosyl)-5-iodocytosine. Cancer Res 1983;43:3619–27.
[13] Pugmire MJ, Ealick SE. The crystal structure of pyrimidine nucleo-
side phosphorylase in a closed conformation. Structure 1998;6:1467–
79.
[14] Rick SW, Abashkin YG, Hilderbrandt RL, Burt SK. Computational
studies of the domain movement and the catalytic mechanism of
thymidine phosphorylase. Prot Struct Func Gen 1999;37:242–52.
[15] Sapse AM, Snyder G. Ab initio studies of the antiviral drug 1-(2-
fluoro-2-deoxy-?-D-arabinofuranosyl)thymine. Cancer Invest 1985;2:
115–21.
[16] Sun H, Mangner TJ, Muzik O, Collins JM, Douglas KA, Shields AF.
Imaging DNA synthesis in tumor patients with18F-FMAU and PET.
J Nucl Med 2002;43(suppl 5):26.
[17] Tanaka H, Miyasaka T. Nucleosides and related compounds which
act as inhibitors against replication of HIV-1. Tanpakushitsu Kakusan
Koso 1995;40:1250–60 Japanese.
[18] Toyohara J, Hayashi A, Sato M, Tanaka H, Haraguchi K, Yoshimura
Y, Yonekura Y, Fujibayashi Y. Rationale of 5-125I-iodo-4?-thio-2?-
deoxyuridine as a potential iodinated proliferation marker. J Nucl
Med 2002;43:1218–26.
[19] Vaidyanathan G, Zalutsky MR. Preparation of 5-[131I]iodo- and
5-[211At]astato-1-(2-deoxy-2-fluoro-?-D-arabinofuranosyl) uracil by
a halodestannylation reaction. Nucl Med Biol 1998;25:487–96.
[20] Watanabe KA, Reichman U, Hirota K, Lopez C, Fox JJ. Nucleosides
695
J. Toyohara et al. / Nuclear Medicine and Biology 30 (2003) 687–696

Page 10

110: synthesis and antiherpes virus activity of some 2?-fluoro-2?-
deoxyarabinofuranosylpyrimidine nucleosides. J Med Chem 1979;29:
21–4.
[21] Watanabe KA, Su TL, Klein RS, Chu CK, Matsuda A, Chun MW,
Lopez C, Fox JJ. Nucleosides. 123. Synthesis of antiviral nucleosides:
5-substituted 1-(2-deoxy-2-halogeno-?-D-arabinofuranosyl)cytosines
and -uracils. Some structure-activity relationships. J Med Chem 1983;
26:152–6.
[22] Yoshimura Y, Kitano K, Yamada K, Sakata S, Miura S, Ashida N,
Machida H. Synthesis and biological activities of 2?-deoxy-2?-fluoro-
4?-thioarabinofuranosylpyrimidine and -purine nucleosides. Bioorg
Med Chem 2000;8:1545–58.
696
J. Toyohara et al. / Nuclear Medicine and Biology 30 (2003) 687–696