Synthesis and biological evaluation of carbon-11-labeled acyclic and furo[2,3-d]pyrimidine derivatives of bicyclic nucleoside analogues (BCNAs) for structure-brain uptake relationship study of BCNA tracers

Article abstract of DOI:10.1002/jlcr.1494

We reported earlier on radiolabeled alkoxyphenyl bicyclic nucleoside analogues (BCNAs) as potential positron emission tomography (PET) reporter probes for imaging of varicella zoster virus thymidine kinase (VZV-tk) gene in vivo. Despite their favorable ph

Full text of DOI:10.1002/jlcr.1494

Research Article
Received 17 September 2007,
Revised 1 November 2007,
Accepted 30 November 2007
Published online 1 February 2008 in Wiley Interscience
(www.interscience.wiley.com) DOI: 10.1002/jlcr.1494
Synthesis and biological evaluation of carbon-
11-labeled acyclic and furo[2,3-d]pyrimidine
derivatives of bicyclic nucleoside analogues
(BCNAs) for structure–brain uptake
relationship study of BCNA tracers
Satish K. Chitneni,a Jan Balzarini,b Sofie Celen,a Natalia Dyubankova,c
Ã
Alfons M. Verbruggen,a and Guy M. Bormansa
We reported earlier on radiolabeled alkoxyphenyl bicyclic nucleoside analogues (BCNAs) as potential positron emission
tomography (PET) reporter probes for imaging of varicella zoster virus thymidine kinase (VZV-tk) gene in vivo. Despite their
favorable physicochemical properties, these tracers are not taken up in the brain in mice. In order to probe the role of the
deoxyribose sugar moiety in blood-brain barrier (BBB) penetration of these molecules, we have synthesized and evaluated
a carbon-11-labeled acyclic bicyclic nucleoside derivative ([11C]-10) where the 20-deoxyribose sugar is replaced with a (2-
hydroxyethoxy)methyl group and [11C]-12, which has no sugar moiety but a [11C]methyl group on the N-3 position of the
pyrimidine ring. Methylation was achieved on the phenol ([11C]-10) or the N-3 position ([11C]-12) using [11C]methyl triflate
(radiosynthesis). The (non-radioactive) acyclic O-methyl derivative 10 has rather poor affinity for the enzyme VZV-TK
in vitro (IC50: 430 lM), compared with the moderate affinity of the BCNA-base N-methyl derivative 12 (IC50: 79 lM). In
normal mice, none of the two tracers ([11C]-10 or [11C]-12) showed significant uptake in the brain, suggesting that
compounds containing a furo[2,3-d]pyrimidine system do not cross the BBB.
Keywords: BCNA; VZV-tk; carbon-11; PET; brain uptake
Bicyclic nucleoside analogues (BCNAs) are a novel class of
antiviral compounds active against varicella zoster virus (VZV).
Introduction
Non-invasive imaging of reporter gene systems using positron
emission tomography (PET) or single photon emission tomo-
graphy (SPECT) offers valuable information about the progress
of gene- or cell-based therapy in several diseases including
cancer, Parkinson’s disease, Alzheimer’s disease and stroke.1–4
Herpes simplex virus type 1 thymidine kinase (HSV1-tk) is the
most widely studied reporter gene for which at least 15 different
radiolabeled nucleoside analogues have been synthesized and
evaluated as substrates.5–8 These substrates are activated by the
protein product of HSV1-tk, which converts them to the mono-
and di-phosphate derivatives, and further conversion by cellular
kinases results in the formation of tri-phosphate derivatives.5–10
These phosphorylated metabolites are negatively charged at
physiological pH and, hence, can be supposed to remain
trapped inside the cells. This entrapment of the radiolabeled
substrate in HSV1-tk expressing cells (Note: tk refers to the gene
and TK refers to the protein) can be quantitatively detected
externally by PET or SPECT imaging, depending on the
radioisotope used. However, as none of these tracers for
HSV1-tk efficiently crosses the blood-brain barrier (BBB),
presumably due to their hydrophilicity, there is a need for
lipophilic BBB penetrating tracers, either for the existing reporter
genes or as new reporter systems.
They display a high lipophilicity with calculated log P (C log P)
values ranging between 0.4 and 4.6.11,12 Extensive structure–-
activity relationship (SAR) studies have been carried out with
respect to their antiviral activity since their discovery by
Mc Guigan et al.11 Of all the BCNAs synthesized and evaluated
so far, para-alkylphenyl BCNAs represent the most active series,
in particular, the p-pentyl analogue has a potency of about
10 000 times higher than that of the current anti-VZV drug of
choice (acyclovir).13–15 Because of their lipophilicity, BCNAs were
anticipated to efficiently pass through cellular membranes and
biological barriers such as the BBB.14 In addition to their high
(anti-VZV) potency, BCNAs also have high specificity for VZV-TK,
aLaboratory for Radiopharmacy, Faculty of Pharmaceutical Sciences, Katholieke
Universiteit Leuven, Leuven, Belgium
bLaboratory for Virology and Chemotherapy, Rega Institute for Medical
Research, Katholieke Universiteit Leuven, Leuven, Belgium
cLaboratory for Medicinal Chemistry, Rega Institute for Medical Research,
BioMacs, Katholieke Universiteit Leuven, Leuven, Belgium
*Correspondence to: Guy M. Bormans, Onderwijs & Navorsing 2, Herestraat 49,
bus 821, BE-3000 Leuven, Belgium.
E-mail: guy.bormans@pharm.kuleuven.be
J. Label Compd. Radiopharm 2008, 51 159–166
Copyright r 2008 John Wiley & Sons, Ltd.

S. K. Chitneni et al.
R
5
evaluated them in human embryonic kidney cell line HEK-
293T cells.18,19 In vitro enzymatic assays using non-radioactive
compounds and purified VZV-TK have revealed that the
affinity of meta-alkoxyphenyl BCNAs for VZV-TK decreases
as the alkyl chain length increases (see Table 1), which
was further evident in cell uptake experiments using [11C]-1
and [18F]-2.
4
6
O
R
OH
1
6
7
2
O
O
7a
5
N
N
1
5a
4
O
N
O
N
3
5'
HO
HO
O
O
4'
1'
In contrast to the meta-fluoroethoxyphenyl-BCNA ([18F]-2),
the para-compound ([18F]-5) has good affinity for the enzyme
(IC50 of 4.2 mM vs 53 mM for the meta-compound; Table 1).
Indeed, [18F]-5 showed good uptake in cell studies with uptake
ratios similar as those of [11C]-1, with about 54-fold
higher accumulation in VZV-tk expressing cells than in control
cells.19
3'
OH
2'
OH
123I]-3, R = 123
I
[
[
11C]-1, R = 11CH3
18F]-2, R = (CH2)218F
[
18
F
O
R
O
We also synthesized an iodine-123-labeled BCNA ([123I]-3) for
SPECT and evaluated its accumulation in VZV-tk expressing 293T
cells.20 In spite of its comparable affinity for VZV-TK (IC50: 4.2 mM;
Table 1) its uptake ratio (VZV-TK cells/control cells) was not
more than 1.74, which is about 30 times less than that of [11C]-1
and [18F]-5. This low accumulation is probably due to the higher
hydrophilicity of [123I]-3, which is evident from its lower log P
value (see Table 1). Similarly, none of the BCNA tracers
synthesized so far had significant brain uptake in biodistribution
studies in normal mice, although they fulfill the theoretical
requirements for passive diffusion of such molecules through
the BBB. Further, we synthesized the fluorine-18-labeled 5-
ethynyl-20-deoxyuridine derivative [18F]-6, which is a BCNA
precursor (non-cyclized BCNA) and is very lipophilic (log P: 2.4;
Table 1) compared with the intact BCNA tracers.21 However,
brain uptake was minimal for this tracer too. On the basis of the
results of these biodistribution studies of BCNA tracers, we
hypothesized that the presence of the polar sugar moiety in the
chemical structure might be detrimental for the BBB passage of
this class of molecules, irrespective of the overall lipophilicity. In
order to examine this hypothesis, we now synthesized a carbon-
11-labeled acyclic-BCNA tracer [11C]-10, which has a hydro-
xyethoxymethyl side chain instead of a 20-deoxy ribose sugar for
BCNAs, and also a carbon-11-labeled BCNA without a sugar
moiety ([11C]-12). We chose to synthesize acyclic BCNAs bearing
at the C-6 position of the furopyrimidine system a phenyl ring
instead of alkyl or alkyn-1-yl as a substituent22 as they can be
easily compared with the previously reported BCNA tracers18–20
and to use a phenol as a precursor for radiolabeling. This paper
reports the chemical and radiochemical synthesis of [11C]-10
and [11C]-12, their in vitro affinity data, and biodistribution
studies in normal mice.
O
O
N
N
HN
O
N
O
HO
HO
O
O
OH
OH
18F]-6
[
[
11C]-4, R = 11CH3
[
18F]-5, R = (CH2)218F
Figure 1. Chemical structures of the previously reported radiolabeled BCNAs and a
5-ethynyl-20-deoxyuridine derivative ([18F]-6).
Table 1. Affinity for VZV-TK and log P7.4 values for the
previously reported and currently synthesized compounds
Compound
IC50 (mM)a
Log Pb
Reference
[11C]-1
[18F]-2
123I]-3
[11C]-4
[18F]-5
[18F]-6
9
[11C]-10
11
[11C]-12
4.871.7
53712
4.270.6
1.570.2
4.270.6
4500
3778
430729
64
1.2770.04
1.2470.03
0.8370.01
1.3070.03
1.2170.02
2.4070.2
—
1.2070.01
—
1.1870.03
18
18
20
19
19
21
—
—
—
—
[
79
aData are expressed as mean7SD; n = 2; IC50 was determined
using different concentrations of the non-radioactive BCNA
analogue and defined as the 50% inhibitory concentration or
compound concentration in mM required to inhibit VZV-TK-
catalyzed phosphorylation of 1 mM [CH3-3H]dThd by 50%.
bData are expressed as mean7SD; n = 6.
Results and discussion
Chemistry
which converts them to the mono- and di-phosphate deriva-
tives. Interestingly, BCNAs do not have affinity for the closely Precursors 9 and 11 for radiolabeling were synthesized by a
related HSV1-TK or mammalian cytosolic or mitochondrial Sonogashira coupling of 1-[(2-hydroxyethoxy)methyl]-5-iodour-
TKs.16,17 The high specificity and the lipophilicity of the BCNAs acil (8) and 5-iodouracil (7), respectively, with 3-hydroxyphenyl
make them suitable candidates for evaluation as reporter acetylene, followed by in situ cyclization of the reaction
probes, which might cross BBB, for gene expression imaging products. First, compound 8 was synthesized by linking a 2-
in the brain.
With this aim, we evaluated the possibility of using this purpose,
hydroxyethoxymethyl chain to the N-1 of 5-iodouracil (7).23 For
was first protected by treatment with
7
radiolabeled BCNAs as PET reporter probes in combination with bis(trimethylsilyl)acetamide (BSA) to give 2,4-bis(trimethylsilyl)-
VZV-tk as reporter gene. We have synthesized fluorine-18- or 5-iodouracil followed by the addition of 1,3-dioxolane, chloro-
carbon-11-labeled meta- or para-alkoxyphenyl-BCNA tracers trimethylsilane, and potassium iodide for the generation of the
whose chemical structures are presented in Figure 1 and iodomethyl[(trimethylsilyl)oxy]ethyl ether in situ. Subsequent
www.jlcr.org
Copyright r 2008 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2008, 51 159–166

S. K. Chitneni et al.
O
O
O
O
N
OSi Me
N
3
BSA (2eq.)
I
I
I
HN
HN
O
N
Me3SiCl/KI
RT, 16 h
CH3CN
O
N
H
O
Me SiO
3
7
8 (52 %)
OH
Pd(PPh3)4, CuI, TEA,
DMF, 70 ˚C, 8 h
H
OH
O
R
OH
O
O
O
N
OH
N
O
N
MeI/[11C]MeOTf
HN
O
O
N
O
N
O
K2CO3/Cs2CO3,
DMF, 70 ˚C
O
OH
OH
OH
10,
R = CH3 (56 %)
9 (25 %)
[11C]-10, R = 11CH3 (28 %)
Figure 2. Synthesis of radiolabeling precursor 9 and its non-radioactive or radioactive O-methyl derivative 10 (BSA = N,O-bis(trimethylsilyl)acetamide).
reaction with 2,4-bis(trimethylsilyl)-5-iodouracil yielded com- reaction conditions. Further, 2D NMR (1H-13C) analysis was
pound 8 (Figure 2). Compound 8 was then coupled to 3- performed on the phenol precursor 11 and its methylated
hydroxyphenyl acetylene under co-catalysis of Pd and cuprous product 12 in heteronuclear single quantum correlation (HSQC)
ions followed by cyclization of the intermediate in a one-pot experiments and heteronuclear multiple bond correlation
synthesis via the procedure reported by Janeba et al.22 to obtain (HMBC) experiments. Based on HSQC analysis, C-4 (furopyrimi-
the fluorescent compound 9 in moderate yield (25%). Com- dine) and phenol carbons were undoubtedly assigned. Accord-
pound 11 was also synthesized following the coupling reaction ing to the HMBC spectrum of 11, the hydroxyl group of the
analogous to the synthesis of 9 but starting from 5-iodouracil (7) phenyl moiety at 9.73 has long range couplings with C-2, C-3
instead of compound 8, in 44% yield (Figure 3). The relatively and C-4 carbons (phenyl moiety). A similar interaction has been
lower yield of compound 9 was mainly due to the cleavage of observed for the proton at 9.73 ppm for compound 12
the N-3-hydroxyethoxymethyl chain during the reaction, result- also, indicating the presence of a free hydroxyl group on C-3
ing in the eventual formation of compound 11 as a side position of the phenyl ring. These data strongly suggest that
product.
the methylation indeed occurred at the N-3 position of the
Compounds 9 and 11 were used as precursors for stable non- pyrimidine ring, resulting in compound 12 as shown in Figure 4.
radioactive methoxy compounds 10 and 12, respectively. Reversed phase-high-performance liquid chromatography (RP-
Methylation of 9 and 11 was achieved by heating the respective HPLC) analysis of methylation reaction mixture of 11 has shown
precursors with methyl iodide (MeI) in the presence of K2CO3 in the presence of another mono-methyl isomer (as determined
DMF at 701C for 1 h,18 in 56 and 32% yield, respectively. In the 1H using mass spectrometry), presumably the O-methyl derivative,
NMR spectrum of the reaction product of the methylation of 9, which was not isolated and characterized due to its relatively
the absence of a signal (d 9.75) corresponding to the phenol low amount (about 4%).
group and the presence of a sharp singlet (d 3.83) from the
methyl ether support a successful O-methylation, yielding
compound 10. We anticipated that the methylation of 11
Radiochemistry
would also occur preferentially at the phenol group, due to the Radiolabeling of 9 was initially performed by heating 0.2 mg of
higher reactivity of a phenol in its deprotonated form in the the precursor with [11C]MeI in the presence of Cs2CO3 (about
presence of the base (K2CO3), than at the N-3 of the 1 mg) in DMF at 701C for 8 min.18 However, this reaction resulted
1
furopyrimidine system. However, the H NMR spectrum of the in a low radiochemical yield (about 5%). Further optimization of
reaction product after methylation of 11 revealed the absence reaction conditions was not tried as increase in reaction time
of the NH proton at 12.11 ppm and the presence of a signal and/or temperature may result in the cleavage of the (2-
corresponding to the phenol (d 9.73), suggesting that N- hydroxyethoxy)methyl chain. Instead, [11C]methyl triflate
methylation rather than the O-methylation occurred in these ([11C]MeOTf), which is a more powerful methylation reagent
J. Label Compd. Radiopharm 2008, 51 159–166
Copyright r 2008 John Wiley & Sons, Ltd.
www.jlcr.org

S. K. Chitneni et al.
OH
H
O
O
O
OH
I
OH
HN
HN
N
O
N
H
O
N
H
O
N
Pd(PPh3)4, CuI, TEA,
DMF, 70 ˚C, 8 h
H
11 (44 %)
7
MeI/[11C]MeOTf
K2CO3/Cs2CO3,
DMF, 70 ˚C
OH
O
N
O
N
R
12,
R = CH3 (32 %)
[11C]-12, R = 11CH3 (50 %)
Figure 3. Synthesis of radiolabeling precursor 11 and its non-radioactive or radioactive N-methyl derivative 12.
125
285
115
105
Radiometric signal
Radiometric signal
UV signal
185
85
UV signal
95
85
0
2
4
6
8
10
12
14
0
2
4
6
8
10
12
14
(a)
(b)
m
m
Figure 4. RP-HPLC chromatograms of [11C]-10 (A; Rt = 4.5 min) and [11C]-12 (B; Rt = 7.1 min) co-injected with their authentic non-radioactive analogues.
than the [11C]MeI, was used for radiolabeling and a methylation alter the lipophilicity of the BCNAs. Compound [11C]-12
yield of 28% was obtained. Radiolabeling of 11 was also appeared to be even less lipophilic than the acyclic-BCNA tracer
performed starting from [11C]MeOTf to provide [11C]-12 with a [11C]-10 as its elution from an RP-C18 HPLC column required an
yield of 50%. Both tracers were purified by semi-preparative RP- about 10% lower concentration of EtOH in the mobile phase for a
HPLC. The identity of the tracers was confirmed by co-injection similar retention time (Rt). However, this was not reflected in its
with their authentic non-radioactive analogues on an analytical partition coefficient as there was no significant difference between
RP-HPLC system (Figure 4). The radiochemical purity (RCP) after the log P values of [11C]-10 and [11C]-12. In general, these log P
isolation by HPLC was found to be 499% for both the tracers. values are similar or comparable to those of the previously
synthesized intact BCNA tracers and are within the optimal range
Lipophilicity of the tracers
(0.9–2.5) for passive diffusion of a neutral compound over the BBB.24
The lipophilicity of the RP-HPLC purified radiolabeled products
[11C]-10 and [11C]-12 was determined by partitioning between
Biodistribution and brain uptake of the tracers
1-octanol and 0.025 M phosphate buffer pH 7.4 (n = 6; Table 1). The biodistribution of [11C]-10 and [11C]-12 at 2 and 60 min post
For comparison, the log P values of previously synthesized injection (p.i.) was evaluated in normal male NMRI mice and the
BCNAs tracers are also presented in the same table. We results are presented in Tables 2 and 3 as percentage of injected
anticipated [11C]-10 and [11C]-12 to be more lipophilic than dose (% ID) and as standard uptake values (SUVs) for selected
the previously reported intact BCNA tracers (Figure 1), but the organs. SUVs were calculated as (radioactivity in cpm in organ/
log P value of 1.2 for both tracers indicates that the modifica- weight of the organ)/(total counts recovered/body weight). The
tions of the 20-deoxyribose sugar moiety do not significantly in vivo distribution and excretion of [11C]-10 are similar to that of
www.jlcr.org
Copyright r 2008 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2008, 51 159–166

S. K. Chitneni et al.
Table 2. Biodistribution of [11C]-10 in normal mice at 2
and 60 min p.i.
was succeeded by the liver with SUVs of 5.1 for [11C]-10 and 3.6
for [11C]-12. At 60 min p.i., these values were 1.4 or less for [11C]-
10, and 0.5 or less for [11C]-12. This indicates a good clearance of
the tracers from blood and major organs.
%IDa
SUVb
Organ
[11C]-10 and [11C]-12 were synthesized with the aim to
evaluate their brain uptake for a better understanding whether
the presence of a ribose/deoxyribose sugar is required for
affinity for the VZV-TK and how its absence influences BBB
penetration of such nucleoside analogues. It is generally
accepted that for a passive diffusion of a compound over the
BBB, its log P value should be between 0.9 and 2.5, its molecular
mass less than 650 Da, and the compound should be uncharged
at physiological pH.24–26 The log P of [11C]-10 and [11C]-12 is
about 1.2, both compounds are uncharged at physiological pH,
and their molecular mass is 316 and 242 Da, respectively.
Despite these favorable characteristics, none of these two
tracers showed significant brain uptake (0.04–0.1% ID at 2 min
p.i.), similar to the previously synthesized BCNA tracers.18–20 This
seems to indicate that a furopyrimidine base system is unable to
diffuse over the BBB, irrespective of the presence or the absence
of a sugar moiety. Lack of brain uptake for the radiolabeled
BCNA precursor [18F]-6 (log P: 2.4)21 further suggests the uracil
base as the entity responsible for the failure of such molecules
to cross the BBB. However, this does not exclude an active efflux
(from the endothelium of the BBB back to blood) as potential
underlying mechanism for the failed efficient transport of these
tracers across the BBB.27 Several multispecific transport proteins
have been identified as mediators of efflux in the BBB, including
P-glycoprotein (P-gp) and the multi-drug-resistance associated
protein (MRP) whose function determines cerebral uptake of
many drugs. Biodistribution of the tracers after pretreatment of
the animals with a specific substrate of the efflux protein can
confirm whether or not the test drug is a substrate for that efflux
protein.27 However, the negligible initial uptake (0.04–0.1% ID in
brain at 2 min p.i.)18–20 suggests that the active efflux mechan-
ism might not be the predominant mechanism involved in the
poor penetration of the BCNA tracers over the BBB.
2 min
60 min
2 min
—
60 min
—
Urine
Kidneys
Liver
Spleen1pan-
creas
Lungs
1.771.0 30.672.3
11.671.2
29.371.7
1.670.5
0.770.1 6.270.6 0.470.1
6.271.4 5.170.4 1.470.2
0.370.1 1.870.3 0.370.1
1.270.0
0.970.1
1.270.1
0.170.0 1.670.2 0.270.0
0.070.0 1.970.2 0.170.0
0.670.1
Heart
Stomach
Intestines
Brain
Blood
Carcass
—
—
—
—
15.272.3 55.471.4
0.170.0
7.970.4
27.072.1
0.170.0 0.270.0 0.170.0
1.070.1 1.170.1 0.170.0
5.370.4
—
—
Note: Data are expressed as mean7SD; n = 3 per time point.
aPercentage of injected dose calculated as cpm in organ/
total cpm recovered.
bStandard uptake values calculated as (radioactivity in cpm in
organ/weight of the organ in g)/(total counts recovered/
body weight in g).
Table 3. Biodistribution of [11C]-12 in normal mice at 2
and 60 min p.i.
%IDa
SUVb
Organ
2 min
60 min
2 min
—
60 min
—
Urine
Kidneys
Liver
Spleen1pan-
creas
Lungs
20.571.5 45.270.8
7.370.5
21.170.7
1.170.3
0.170.0 4.570.2 0.170.0
3.073.4 3.670.2 0.570.5
0.070.0 1.570.3 0.070.0
0.570.1
0.370.0
0.570.2
0.070.0 0.770.1 0.070.0
0.070.0 0.770.0 0.070.0
0.270.1
Heart
Affinity for the VZV-TK enzyme in vitro
Stomach
Intestines
Brain
Blood
Carcass
—
—
—
—
24.572.7 47.973.1
The affinity of the synthesized compounds (non-radioactive) for
the nucleoside kinase enzyme was examined using purified
recombinant VZV-TK in the presence of tritiated dThd (1 mM) as
the natural substrate, and the results are expressed as 50%
inhibitory concentration (IC50) values in Table 1. For comparison,
the IC50 values of the previously reported BCNA tracers are also
given in Table 1. The acyclic BCNA phenol precursor 9 showed
moderate affinity (IC50: 37 mM) for the enzyme, whereas its
O-methyl derivative 10 has rather poor affinity with an IC50 value
of 430 mM. The moderate affinity of compound 9 suggests the
potential usefulness of acyclic BCNAs as inhibitors of VZV-TK as
structural modification may further enhance the observed
0.070.0
3.771.0
19.973.7
0.070.0 0.170.0 0.070.0
0.170.0 0.570.1 0.070.0
3.071.1
—
—
Note: Data are expressed as mean7SD; n = 3 per time point.
aPercentage of injected dose calculated as cpm in organ/
total cpm recovered.
bStandard uptake values calculated as (radioactivity in cpm in
organ/weight of the organ in g)/(total counts recovered/
body weight in g).
the previously reported BCNA tracers18–20 with about 55% ID affinity for VZV-TK. To be a useful tracer for gene expression
excreted into the intestines and 31% ID into the urine at 60 min imaging in the brain, the BCNA tracer agent should have a
p.i. On the other hand, [11C]-12 was cleared almost equally via reasonable initial uptake in the brain and have a good affinity
the intestines (48% ID at 60 min p.i.) and kidneys (45% ID). The and be a substrate for phosphorylation by the VZV-TK enzyme.
clearance of both tracers from blood was rapid with p1.0% ID Therefore, radiolabeled acyclic BCNAs may not offer additional
remaining in blood at 60 min p.i. Except for liver (3–6% ID) and advantages over the previously reported intact BCNA tracers as
intestines (48–55% ID), the radioactivity in other major organs they too do not show passage over the BBB. Compounds 11 and
was negligible 60 min after injection of the tracers. At 2 min p.i., 12, i.e. the 6-(3-hydroxy)phenyl-furo[2,3-d]pyrimidin-2(3H)-one
the highest concentration of radioactivity was found in the precursor and its N-methyl derivative, respectively, were
kidneys with a SUV of 6.2 for [11C]-10 and 4.5 for [11C]-12. This anticipated not to have any affinity for the enzyme VZV-TK
J. Label Compd. Radiopharm 2008, 51 159–166
Copyright r 2008 John Wiley & Sons, Ltd.
www.jlcr.org

S. K. Chitneni et al.
due to the absence of a 50-hydroxyl group or deoxyribose sugar (1.7 g, 10 mmol) and chlorotrimethylsilane (1.38 mL, 10 mmol)
that is required for the phosphorylation of these molecules. were sequentially added, and the mixture was stirred at room
Hence, the moderate affinity values (IC50 11: 64 mM; IC50 12: temperature (RT) for 16 h. The reaction was quenched by the
79 mM) observed for these two compounds suggest that they addition of methanol (20 mL) followed by neutralization with
bind to the enzyme without being a substrate, thereby sodium bicarbonate (4 g). Solids were removed by filtration and
inhibiting the phosphorylation of [CH3-3H]deoxythymidine, the filtrate was concentrated under reduced pressure. The crude
which was used as a natural substrate for the VZV-TK in the product was purified using column chromatography on silica gel
competition experiments. Radiolabeled 12 is probably not a eluted with gradient mixtures of CH3OH in CH2Cl2 (up to 8%) to
useful tracer for VZV-tk detection in vivo because it cannot be give 1.63 g of compound 8 as a white solid (52% yield). 1H NMR
phosphorylated and hence will not accumulate inside the VZV- (300 MHz, DMSO-d6) d: 11.70 (brs, 1H), 8.24 (s, 1H), 5.08 (s, 2H),
3
tk expressing cells because of the presence of an ionized 4.68 (t, J = 5.1 Hz, 1H), 3.50–3.46 (m, 4H).
phosphate group.
6-(3-Hydroxy)phenyl-3-[(2-hydroxyethoxy)methyl]furo[2,3-d]pyrimi-
din-2(3H)-one 9
Experimental
To a stirred suspension of compound 8 (1.24 g, 3.99 mmol),
tetrakis(triphenylphosphine)palladium(0) (0.23 g, 0.2 mmol) and
Instruments, chemicals and general conditions
copper(I) iodide (0.76 g, 3.98 mmol) in DMF (80 mL) under N2 at
RT was added 3-hydroxyphenylacetylene (0.57 g, 4.78 mmol) and
dry triethylamine (4 mL). The reaction mixture was stirred at
701C for 8 h followed by the evaporation of solvents under
reduced pressure. The crude product was treated with toluene
(2 ꢁ 20 mL) and the volatiles were evaporated under reduced
pressure. The resulting residue was then extracted with hot
methanol, and the suspension was filtered and washed with
methanol. After concentration of the combined filtrate under
reduced pressure, the residue was adsorbed on silica gel and
purified using a silica gel column chromatography eluted with
gradient mixtures of CH3OH (up to 10%) in CH2Cl2. Appropriate
fractions were combined and evaporated to yield 0.3 g of a light-
yellow solid (25% yield). 1H NMR (300 MHz, DMSO-d6) d: 9.75
(s, 1H), 8.64 (s, 1H), 7.27 (m, 2H), 7.20 (s, 1H), 7.19 (m, 1H), 6.82
(m, 1H), 5.39 (s, 2H), 4.71 (s, 1H), 3.60 (m, 2H), 3.51 (m, 2H). HRMS
ESI1: Theor. mass [C15H14N2O51Na]1: 325.0800; found
325.0801.
A Cyclone 18/9 cyclotron (Ion Beam Applications, Louvain-la-
1
Neuve, Belgium) was used for the production of carbon-11. H
NMR spectra were recorded on a Gemini 300 MHz spectrometer
(Varian, Palo Alto, USA) using DMSO-d6 as a solvent. HSQC and
HMBC spectra for compounds 11 and 12 were recorded on a
Bruker Avance 600 MHz NMR spectrometer (Bruker BioSpin,
Rheinstetten, Germany). The 2D 1H/13C correlation via double
inept transfer (HSQC) was acquired in a phase-sensitive mode
with Echo-Antiecho gradient selection. The 2D 1H/13C correla-
tion via heteronuclear zero and double quantum coherence
(HMBC) was acquired using gradient pulses for selection. The 2D
NMR data were processed with Bruker-TopSpin software (Bruker
BioSpin). Accurate mass determination was performed on a
time-of-flight mass spectrometer (LCT, Micromass, Manchester,
UK) equipped with an orthogonal electrospray ionization (ESI)
interface, operated in positive mode (ES1) or negative mode
(ESꢀ). Samples were infused in acetonitrile/water using a
Harvard 22 syringe pump (Harvard instruments, MA, USA).
Acquisition and processing of data was done using MasslynxTM
software (version 3.5, Waters). HPLC purification was performed
using a Merck Hitachi L6200 intelligent pump (Hitachi, Tokyo,
Japan) or a Waters 600 pump (Waters, Milford, USA) connected
to a UV spectrometer (Waters 2487 Dual l absorbance detector)
set at 254 nm. For the purification and analysis of radiolabeled
compounds, the HPLC eluate was passed through a UV detector
followed by a radiometric detector connected to a single
channel analyzer (Medi-Lab Select, Mechelen, Belgium). Radio-
activity measurements during log P determination and biodis-
tribution studies were done using an automated gamma
counter (3-in NaI(Tl) well crystal) coupled to a multi-channel
analyzer (Wallac 1480 WizardTM 3’’, Wallac, Turku, Finland). The
results were corrected for background radiation and physical
decay during counting. All reagents and solvents were obtained
commercially from Acros Organics (Geel, Belgium) or Aldrich,
Fluka, Sigma (Sigma-Aldrich, Bornem, Belgium) or Fischer
Bioblock Scientific (Tournai, Belgium).
6-(3-Methoxy)phenyl-3-[(2-hydroxyethoxy)methyl]furo[2,3-d]pyri-
midin-2(3H)-one 10
To a stirred solution of 9 (50 mg, 0.165 mmol) in DMF (8 mL) was
added K2CO3 (23 mg, 0.165 mmol) and methyl iodide (10 mL,
0.165 mmol). The reaction mixture was stirred at 701C for 1 h and
purified by HPLC using
a C-18 semi-preparative column
(Econosphere, 10 mm ꢁ 250 mm; Alltech, Deerfield, USA) eluted
with water–acetonitrile (75:25, V/V) at a flow rate of 3 mL/min
(Rt = 15.2 min). After evaporation of the solvent, the methylated
product was obtained as a white solid (29 mg, 56%). 1H NMR
(300 MHz, DMSO-d6) d: 8.69 (s, 1H), 7.41 (m, 2H), 7.36 (m, 2H),
7.01 (m, 1H), 5.39 (s, 2H), 4.73 (s, 1H), 3.83 (s, 3H), 3.60 (m, 2H),
3.51 (m, 2H). HRMS ESI1: Theor. mass [C16H16N2O51Na]1:
339.0957; found 339.0950.
6-(3-Hydroxy)phenyl-furo[2,3-d]pyrimidin-2(3H)-one 11
Compound 11 was synthesized in the same manner as
compound 9 but starting from 5-iodouracil 7 (1.19 g; 5 mmol)
instead of compound 8. The crude product was purified using a
silica gel column eluted with gradient mixtures of CH3OH in
CH2Cl2 (up to 10%). The resulting residue was treated with hot
acetonitrile followed by filtration to yield the product as yellow
solid (0.5 g, 44%). NMR: dH (300 MHz, DMSO-d6) 12.11 (brs, 1H),
9.73 (s, 1H), 8.32 (s, 1H), 7.27 (m, 2H), 7.17 (s, 1H), 7.12 (s, 1H),
6.81 (m, 1H); dC (600 MHz, DMSO-d6) 172.30, 158.27, 156.34,
Chemistry
1-[(2-Hydroxyethoxy)methyl]-5-iodouracil 8
To
a suspension of 5-iodouracil (7, 2.38 g, 10 mmol) in
acetonitrile (15 mL) was added N,O-BSA (5.4 mL, 22 mmol) and
the mixture was stirred to obtain a clear solution. To this
solution, 1,3-dioxolane (0.7 mL, 10 mmol), potassium iodide
www.jlcr.org
Copyright r 2008 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2008, 51 159–166

S. K. Chitneni et al.
153.72, 140.91, 130.71, 130.06, 116.94, 115.89, 111.39, 106.94, V/V) at a flow rate of 1 mL/min (Rt = 7.1 min). The RCY of the tracer
99.82. HRMS ESI-: Theor. mass [C12H8N2O3 - H]-: 227.0457; found was 50% (relative to [11C]MeOTf radioactivity, NDC) and the specific
227.0464.
radioactivity was 68.9 GBq/mmol (1.86 Ci/mmol) at EOS (Figure 3).
6-(3-Methoxy)phenyl-furo[2,3-d]pyrimidin-2(3H)-one 12
Partition coefficient determination
Compound 12 was synthesized in the same manner as that of An aliquot of 25 mL of RP-HPLC purified radiolabeled product
10 but using compound 11 (50 mg, 0.22 mmol) as a starting containing approximately 550 kBq of [11C]-10 or [11C]-12
material. The reaction mixture was purified in a similar manner was added to a tube containing a mixture of 1-octanol and
using water–acetonitrile (85:15, V/V) at a flow rate of 2.5 mL/min 0.025 M phosphate buffer pH 7.4 (2 mL each). The tube was
(Rt = 19.6 min). After evaporation of the solvent, the methylated vortexed at RT for 2 min followed by centrifugation at 3000 rpm
product was obtained as a white solid (17 mg, 32%). NMR: dH (1837g) for 5 min (Eppendorf centrifuge 5810, Eppendorf,
(300 MHz, DMSO-d6) 9.73 (s, 1H), 8.64 (s, 1H), 7.30 (m, 2H), 7.20 (s, Westbury, NY). Aliquots of about 60 and 500 mL were drawn
1H), 7.18 (s, 1H), 6.82 (m, 1H), 3.53 (s, 3H); dC (600 MHz, DMSO-d6) from the 1-octanol and aqueous phases, respectively, taking
171.57, 158.27, 155.47, 153.72, 144.39, 130.71, 130.06, 116.93, care to avoid cross-contamination between the two phases, and
115.88, 111.38, 106.84, 99.68, 55.79. HRMS ESI-: Theor. mass weighed. The radioactivity in the aliquots was counted using
[C13H10N2O3 - H]-: 241.0613; found 241.0634.
a automated gamma counter. After correcting for density and
the mass difference between the two phases, the partition
coefficient (P) was calculated as (radioactivity (cpm/mL) in
1-octanol)/(radioactivity (cpm/mL) in phosphate buffer at
pH 7.4).
Radochemistry
Production of [11C]methyl triflate ([11C]MeOTf)
Carbon-11 was produced by a 14N(p,a)11C nuclear reaction.
The target gas (a mixture of 95% N2 and 5% H2) was irradiated
using 18-MeV protons from a Cyclone 18/9 cyclotron (Ion
Beam Applications, Louvain-la-Neuve, Belgium) at a beam
current of 25 mA for about 30 min to yield [11C]CH4. This was
then reacted with vaporous I2 at 6501C to convert it to
[11C]methyl iodide ([11C]MeI). The resulting volatile [11C]MeI
was passed through a glass column (3 mm ꢁ 150 mm) contain-
ing silvertriflate (AgOTf), which was heated at 275–2901C to
yield [11C]MeOTf.
Biodistribution in normal mice
Solutions of [11C]-10 or [11C]-12 obtained after RP-HPLC
purification were diluted using water for injection to obtain an
ethanol concentration o5%, and further to a concentration of
about 18.5 MBq/mL. The biodistribution of both tracers was
determined in male NMRI mice (body weight 30–38 g). The
animal studies were performed according to the Belgian code of
practice for the care and use of animals, after approval from the
university ethics committee for animals. A volume of 0.2 mL of
the diluted tracer solution was injected into each mouse via a
tail vein, under anesthesia (2.5% isoflurane in 100% O2 at a flow
rate of 1 L/min). The mice were sacrificed by decapitation at
2 min or 60 min after injection (n = 3 at each time point). Blood
and major organs were collected in a tared tube after dissection
and weighed. All organs and other body parts were measured
Synthesis of [11C]-10
The [11C]MeOTf resulting from the above reaction was trapped
in a reaction vial containing a mixture of 9 (0.2 mg) and Cs2CO3
(about 1 mg) in anhydrous DMF (0.2 mL). When maximum
radioactivity was trapped, the reaction mixture was heated at
701C for 8 min. The crude reaction mixture was diluted with
water (1.6 mL) and applied onto a semi-preparative HPLC
for their radioactivity using
a gamma counter. Results
are expressed as %ID (cpm in organ/total cpm recovered), or,
where possible, as SUVs. SUVs were calculated as (radioactivity
in cpm in organ/weight of the organ in g)/(total counts
recovered/body weight in g). For the calculation of total
radioactivity in blood, blood mass was assumed to be 7% of
the body mass.26
˚
column (HS HyperPrep RP C18, 100 A, 8 mm; 10 mm ꢁ 250 mm;
Alltech), which was eluted with 0.05 M sodium acetate buffer
(pH = 5.5)/EtOH (65:35, V/V) at
a flow rate of 3 mL/min
(Rt = 11.5 min). The identity and purity of the labeled product
was analyzed using RP-HPLC on an XTerraTM RP C18 column
(5 mm, 4.6 mm ꢁ 250 mm; Waters), eluted with 0.05 M sodium
Affinity of test compounds for nucleoside kinase enzyme
acetate buffer (pH = 5.5)/acetonitrile (65:35, V/V) at a flow rate VZV-TK in vitro
of 1 mL/min (Rt = 4.5 min). [11C]-10 was synthesized in 28%
The 50% inhibitory concentration (IC50) value of the synthesized
radiochemical yield (RCY; relative to [11C]MeOTf radioactivity,
non-decay corrected (NDC)) and the specific radioactivity was
175.6 GBq/mmol (4.74 Ci/mmol) at the end of synthesis (EOS)
(Figure 2).
compounds 9–12 against phosphorylation of [CH3-3H]dThd
(deoxythymidine) as the natural substrate for VZV-TK was
determined. Briefly, the activity of the purified recombinant
VZV-TK enzyme was assayed in a 50-mL reaction mixture
containing 50 mM Tris–HCl, pH 8.0, 2.5 mM MgCl2, 10 mM
dithiothreitol, 2.5 mM ATP, 1.0 mg/mL bovine serum albumin,
Synthesis of [11C]-12
[11C]-12 was also synthesized in the same manner as that of [11C]- 10 mM NaF, [CH3-3H]dThd (3.7 kBq in 5 mL; 1 mM final concentra-
10 but using compound 11 as a precursor, and purification of the tion) and 5 mL of a solution of recombinant enzyme (containing
product was done using 0.05 M sodium acetate buffer (pH = 5.5)/ 7.75 ng of VZV-TK protein). The samples were incubated at 371C
EtOH (75:25, V/V) as a mobile phase at a flow rate of 3 mL/min for 30 min in the presence or absence of different concentra-
(Rt = 13.3 min). The identity and purity analysis of the tracer was tions of the test compounds. During this time period, the
performed on analytical XTerraTM RP C18 column (Waters) eluted enzymatic reaction proceeded linearly. Aliquots of 45 mL of the
with 0.05 M sodium acetate buffer (pH = 5.5)/acetonitrile (75:25, reaction mixtures were spotted on Whatman DE-81 filter paper
J. Label Compd. Radiopharm 2008, 51 159–166
Copyright r 2008 John Wiley & Sons, Ltd.
www.jlcr.org

S. K. Chitneni et al.
disks (Whatman, Maidstone, UK). The filters were washed three
times for 5 min in 1 mM ammonium formate and once for 5 min
in ethanol. The radioactivity on the filters was determined by
liquid scintillation counting.
[6] M. M. Alauddin, P. S. Conti, Nucl. Med. Biol. 1998, 25, 175–180.
[7] J. G. Tjuvajev, M. Doubrovin, T. Akhurst, S. D. Cai, J. Balatoni, M. M.
Alauddin, R. Finn, W. Bornmann, H. Thaler, P. S. Conti, R. G.
Blasberg, J. Nucl. Med. 2002, 43, 1072–1083.
[8] M. M. Alauddin, A. Shahinian, R. Park, M. Tohme, J. D. Fissekis, P. S.
Conti, J. Nucl. Med. 2004, 45, 2063–2069.
Conclusion
[9] S. Yaghoubi, J. R. Barrio, M. Dahlbom, M. Iyer, M. Namavari,
R. Goldman, H. R. Herschman, M. E. Phelps, S. S. Gambhir, J. Nucl.
Med. 2001, 42, 1225–1234.
Carbon-11-labeled BCNAs without sugar moiety or with an
acyclic sugar have been synthesized and evaluated for their
affinity for VZV-TK in vitro and for their brain uptake in normal
mice. Neither of the synthesized tracers was taken up in the
brain. Both the acyclic and the ribose lacking derivative still
showed moderate affinity for the VZV-TK. The results from our
previous studies18–21 and this study strongly suggest that
bicyclic furopyrimidine system analogues (BCNAs) do not cross
BBB and hence can be used for in vivo gene expression imaging
outside of the brain only. This is an important insight into the
structure–brain uptake relationship of this class of compounds
as the p-pentylphenyl derivative is currently in the drug
development phase for use in shingles infection.28 Our
experience with radiolabeled BCNAs suggests that a potential
BCNA tracer for in vivo visualization of VZV-tk should have good
affinity for the enzyme VZV-TK (IC50 value ideally be o5 mM) and
should have reasonable lipophilicity (log P value ideally be over
1) for a good accumulation in VZV-tk expressing cells.
[10] S. S. Yaghoubi, M. A. Couto, C. C. Chen, L. Polavaram, G. G. Cui, L. Y.
Sen, S. S. Gambhir, J. Nucl. Med. 2006, 47, 706–715.
[11] C. McGuigan, C. J. Yarnold, G. Jones, S. Velazquez, H. Barucki,
A. Brancale, G. Andrei, R. Snoeck, E. De Clercq, J. Balzarini, J. Med.
Chem. 1999, 42, 4479–4484.
[12] C. McGuigan, H. Barucki, S. Blewett, A. Carangio, J. T. Erichsen,
G. Andrei, R. Snoeck, E. De Clercq, J. Balzarini, J. Med. Chem. 2000,
43, 4993–4997.
[13] C. McGuigan, A. Brancale, H. Barucki, S. Srinivasan, G. Jones,
R. Pathirana, A. Carangio, S. Blewett, G. Luoni, O. Bidet, A. Jukes,
C. Jarvis, G. Andrei, R. Snoeck, E. De Clercq, J. Balzarini, Antiviral
Chem. Chemother. 2001, 12, 77–89.
[14] J. Balzarini, C. McGuigan, Biochim. Biophys. Acta-Mol. Basis Dis.
2002, 1587, 287–295.
[15] J. Balzarini, C. McGuigan, J. Antimicrob. Chemother. 2002,
50, 5–9.
[16] R. Sienaert, L. Naesens, A. Brancale, E. De Clercq, C. McGuigan,
J. Balzarini, Mol. Pharmacol. 2002, 61, 249–254.
[17] J. Balzarini, R. Sienaert, S. Liekens, A. Van Kuilenburg, A. Carangio,
R. Esnouf, E. De Clercq, C. McGuigan, Mol. Pharmacol. 2002, 61,
1140–1145.
[18] S. K. Chitneni, C. M. Deroose, J. Balzarini, R. Gijsbers, S. J. L. Celen,
T. J. de Groot, Z. Debyser, L. Mortelmans, A. M. Verbruggen, G. M.
Bormans, J. Med. Chem. 2007, 50, 1041–1049.
Acknowledgement
[19] S. K. Chitneni, C. M. Deroose, J. Balzarini, R. Gijsbers, S. Celen,
Z. Debyser, L. Mortelmans, A. M. Verbruggen, G. M. Bormans,
J. Med. Chem. 2007, 50, 6627–6637.
[20] S. K. Chitneni, C. M. Deroose, H. Fonge, R. Gijsbers, N. Dyubankova,
J. Balzarini, Z. Debyser, L. Mortelmans, A. M. Verbruggen, G. M.
Bormans, Bioorg. Med. Chem. Lett. 2007, 17, 3458–3462.
[21] S. K. Chitneni, T. De Ruymaeker, J. Balzarini, A. M. Verbruggen, G.
M. Bormans, J. Labelled Compd. Radiopharm. 2007, 50,
649–655.
[22] Z. Janeba, J. Balzarini, G. Andrei, R. Snoeck, E. De Clereq,
M. J. Robins, J. Med. Chem. 2005, 48, 4690–4696.
[23] M. Ubasawa, H. Takashima, K. Sekiya, Chem. Pharm. Bull. 1995, 43,
142–143.
The technical assistance of Mrs Lizette van Berckelaer is
gratefully acknowledged. We thank Peter Vermaelen, Humphrey
Fonge and Li Xin Jin for their help during biodistribution studies.
This work was supported by the SBO grant (IWT-30 238) of the
Flemish Institute supporting Scientific-Technological Research in
industry (IWT), the IDO grant (IDO/02/012) of the Katholieke
Universiteit Leuven, in part by the European Commission (EC) –
FP6-project DiMI, LSHB-CT-2005-512146, the R. Descartes-Prize-
2001 of the EC (HPAW-2002-90001) and by the BioMacs of the
Katholieke Universiteit Leuven.
[24] D. D. Dishino, M. J. Welch, M. R. Kilbourn, M. E. Raichle, J. Nucl. Med.
1983, 24, 1030–1038.
[25] R. C. Young, R. C. Mitchell, T. H. Brown, C. R. Ganellin, R. Griffiths,
M. Jones, K. K. Rana, D. Saunders, I. R. Smith, N. E. Sore, T. J. Wilks,
J. Med. Chem. 1988, 31, 656–671.
REFERENCES
[1] I. Penuelas, J. F. Boan, J. M. Marti-Climent, B. Sangro, G. Mazzolini,
J. Prieto, J. A. Richter, Mol. Imaging Biol. 2004, 6, 225–238.
[2] S. S. Gambhir, Nat. Rev. Cancer 2002, 2, 683–693.
[26] A. R. Fritzberg, W. P. Whitney, C. C. Kuni, W. Klingensmith, Int.
J. Nucl. Med. Biol. 1982, 9, 79–82.
[27] P. H. Elsinga, N. H. Hendrikse, J. Bart, W. Vaalburg, A. van Waarde,
Curr. Pharm. Design 2004, 10, 1493–1503.
[28] G. Luoni, C. McGuigan, G. Andrei, R. Snoeck, E. De Clercq,
J. Balzarini, Bioorg. Med. Chem. Lett. 2005, 15, 3791–3796.
[3] I. Serganova, R. Blasberg, Nucl. Med. Biol. 2005, 32, 763–780.
[4] H. R. Herschman, Crit. Rev. Oncol. Hematol. 2004, 51, 191–204.
[5] M. M. Alauddin, P. S. Conti, S. M. Mazza, F. M. Hamzeh, J. R. Lever,
Nucl. Med. Biol. 1996, 23, 787–792.
www.jlcr.org
Copyright r 2008 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2008, 51 159–166