Synthesis of 2-[(4-[18F]Fluorobenzoyloxy)methyl]-1,4- naphthalenedione from 2-hydroxymethyl 1,4-naphthoquinone and 4-[ 18F]fluorobenzoic acid using dicyclohexyl carbodiimide

Article abstract of DOI:10.1002/jlcr.1932

2-[(4-[¹⁸F]Fluorobenzoyloxy)methyl]-1,4-naphthalenedione ([ ¹⁸F]1) was synthesised as a putative hypoxia imaging agent from 2-hydroxymethyl 1,4-naphthoquinone (7) and 4-[¹⁸F]fluorobenzoic acid ([¹⁸F]8) using dicyclohexyl carbodiimide (DCC) to activate [ ¹⁸F]8. This coupling reaction was fast and gave quantitative yields. Further investigations are warranted on the use of DCC as a coupling agent in Positron Emission Tomography. The synthesis including HPLC purification and reformulation has been fully automated on a modified FDG synthesiser with two reactor vials. [¹⁸F]1 was produced in a radiochemical yield of 27 ± 5%, with a radiochemical purity of 97.5% and a specific activity of 78.4-134.5 GBq/μmol at the end of synthesis (n = 23). The total synthesis time including reformulation was 65 min. [¹⁸F]1 was found to be stable in plasma and saline, but underwent rapid metabolism in a phase 1 metabolite assay using rat S9 liver fractions. An in vivo evaluation of [ ¹⁸F]1 in transplanted, hypoxic SK-RC-52 tumour-bearing BALB/c nude mice revealed the tumour-to-muscle ratio to be 2.4 ± 0.1 at 2 h post-injection.

Full text of DOI:10.1002/jlcr.1932

Research Article
Received 8 June 2011,
Revised 10 August 2011,
Accepted 17 August 2011
Published online 23 September 2011 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/jlcr.1932
Synthesis of 2-[(4-[18F]Fluorobenzoyloxy)
methyl]-1,4-naphthalenedione from
2-hydroxymethyl 1,4-naphthoquinone
and 4-[18F]fluorobenzoic acid using
dicyclohexyl carbodiimide
a,d
Uwe Ackermann,a,b Duanne Sigmund, Shinn Dee Yeoh,a
*
Angela Rigopoulos,c Graeme O’Keefe,a,b Glenn Cartwright,c
Jonathan White,d Andrew M. Scott,a,b,c and Henri J. Tochon-Danguya,b
2-[(4-[18F]Fluorobenzoyloxy)methyl]-1,4-naphthalenedione ([18F]1) was synthesised as a putative hypoxia imaging agent
from 2-hydroxymethyl 1,4-naphthoquinone (7) and 4-[18F]fluorobenzoic acid ([18F]8) using dicyclohexyl carbodiimide
(DCC) to activate [18F]8. This coupling reaction was fast and gave quantitative yields. Further investigations are warranted
on the use of DCC as a coupling agent in Positron Emission Tomography. The synthesis including HPLC purification and
reformulation has been fully automated on a modified FDG synthesiser with two reactor vials. [18F]1 was produced in a
radiochemical yield of 27 ꢀ 5%, with a radiochemical purity of 97.5% and a specific activity of 78.4–134.5 GBq/mmol at the
end of synthesis (n = 23). The total synthesis time including reformulation was 65 min. [18F]1 was found to be stable in
plasma and saline, but underwent rapid metabolism in a phase 1 metabolite assay using rat S9 liver fractions. An in vivo
evaluation of [18F]1 in transplanted, hypoxic SK-RC-52 tumour-bearing BALB/c nude mice revealed the tumour-to-muscle
ratio to be 2.4 ꢀ 0.1 at 2 h post-injection.
Keywords: fluorine-18; PET chemistry; 4-fluorobenzoate; tumor hypoxia; DCC; SK-RC-52 tumors
as the biologically relevant moiety responsible for the trapping in
Introduction
the hypoxic cell.7–9 However, in our laboratory we have synthe-
Imaging of hypoxic tissue is of great significance in oncology
because hypoxic tumours are more resistant to radiotherapy
and chemotherapy than normoxic tumours.1 For the planning
of cancer therapy, the measurement of the hypoxic fraction of
tumours is therefore of critical importance.2,3
The gold standard for the measurement of oxygen partial
pressure (pO2) is by using a polarographic oxygen electrode.4
Unfortunately, this is an invasive technique and therefore is not
suitable for routine clinical application. Positron Emission Tomo-
graphy (PET) is noninvasive and offers the potential to measure
physiological processes in vivo. To date, the most commonly
used radiotracers for PET imaging of tumour hypoxia are the
nitroimidazole-based compounds [18F]FMISO and [18F]FAZA.5,6
sised two fluorine-18-labelled haloethyl sulfoxides, which we
have investigated in vivo and in vitro as a new class of hypoxia
imaging agents.10
Although these compounds showed great promise in the
imaging of hypoxic tissue in a middle cerebral artery occlusion
stroke model in rats, the clinical use of these compounds was
deemed unsuitable because their structure is closely related
to that of highly toxic nitrogen mustards.
In our search for potential novel hypoxia imaging agents, we
have found in the literature that derivatives of 2-hydroxymethyl
aCentre for PET, Austin Health, Melbourne, Australia
However, slow accumulation in hypoxic tissue and slow clearance bThe University of Melbourne, School of Medicine, Dentistry and Health Sciences,
from normoxic tissue result in a low target-to-background ratio
Melbourne, Australia
and a 2 h delay between tracer administration and actual scan-
ning of the patient. Also, [18F]FMISO only shows uptake in
tumours with a pO2 value below 10 mm Hg, and a differential
assessment of tumour hypoxia is therefore not possible. These
shortcomings of the existing tracers have sparked the develop-
ment of new tracers for hypoxia imaging. Most novel tracers
reported in the literature have retained the 2-nitroimidazole core
cLudwig Institute for Cancer Research, Melbourne-Austin Branch, Melbourne,
Australia
dThe University of Melbourne, Bio21 Institute, Melbourne, Australia
*Correspondence to: Uwe Ackermann, Centre for PET, Austin Health, Melbourne,
Australia.
E-mail: ackerman@petnm.unimelb.edu.au
J. Label Compd. Radiopharm 2011, 54 788–794
Copyright © 2011 John Wiley & Sons, Ltd.

U. Ackermann et al.
naphthoquinones have the ability to either release cytotoxic
The cold standard for tracer identification and measurement
compounds under hypoxic conditions or become cytotoxic of specific activity was synthesised from 4-[19F]fluorobenzoic
themselves.11,12 Figure 1 shows the mechanism by which these acid (8) and 2-hydroxymethyl 1,4-naphthoquinone (7) using
quinones can release xenobiotic compounds.
dicyclohexyl carbodiimide (DCC) for the esterification step
The aim of this project was to couple 2-hydroxymethyl 1, (Figure 3).
4-naphthoquinone (7) to 4-[18F]fluorobenzoic acid ([18F]8), and
The cold standard was characterised by NMR and high-resolution
to investigate the in vitro and in vivo properties of this putative MS. The elution profile of [19F]1 was determined using an LCMS
tracer using hypoxic SK-RC-52 tumours.
system for future independent quality control of the radiotracer.
The UV active peak at 18.3 min showed a mass spectrum consistent
with the structure of 1. Characteristic ions were [C7H5FO2 +
NaCH3CN(H2O)1]+ (m/z= 222), [M-C7H3FO2 +H(H2O)5]+ (m/z= 263),
and [M + H]+ (m/z= 311).
Results and discussion
Chemistry
Radiolabelling
Precursor and F-19 standard synthesis
4-[18F]fluorobenzoic acid was produced from (4-ethoxycarbonylphenyl)
trimethylammonium triflate (9) via an aromatic nucleophilic
substitution mechanism using a dried [18F]KF/Kryptofix com-
plex.16 After hydrolysis of the intermediate ethyl 4-[18F]
fluorobenzoate ([18F]10) with tetramethylammonium hydroxide
followed by addition of acetic acid, 4-[18F]fluorobenzoic acid
([18F]8) was obtained in a radiochemical yield of 34 ꢀ 5%. The
compound was trapped on a C-18 SepPak and eluted with
acetonitrile into a second reactor vial. [18F]1 was obtained by
reacting quinone 7 with 4-[18F]fluorobenzoic acid ([18F]8) in
the presence of DCC (Figure 4).
Semi-preparative HPLC separation was carried out using an
isocratic 60:40 acetonitrile/0.1 M ammonium formate mobile
phase on an Alltech Apollo™ C-18 column at a flow rate of
4 mL/min. The retention time of the tracer was 16 min. After HPLC
separation, the radiotracer was reformulated in 10 mL of 10%
ethanol/saline.
Figure 2 shows the synthesis of 2-hydroxymethyl 1,4-naphthoquinone
(7), which was achieved by following the procedures reported in the
literature.
Commercially available 1,4-naphthoquinone (2) was reacted
with SnCl2 and MeOH/HCl to form 1,4-dimethoxynaphthalene
(3).13 Bromination of 1,4-dimethoxynaphthalene (3) gave 2-bromo
1,4-dimethoxynaphthalene (4), which was subsequently converted
to 1,4 dimethoxynaphthalene-2-carbaldehyde (5) via a Grignard
reaction.14 Reduction of this compound to 2-hydroxymethyl
1,4-dimethoxynaphthalene (6) was achieved with sodium borohy-
dride as reducing agent.15 The desired 2-hydroxymethyl 1,4-
naphthoquinone (7) was then synthesised from 2-hydroxymethyl
1,4-dimethoxynaphthalene (6) by cerium ammonium nitrate
oxidation.15
O
OR
O
Independent quality control was performed using a radio-HPLC
system. The radioactive and UV active peaks observed were at
identical retention times as the cold standard (18.3 min).
Radiochemical yield was 27 ꢀ 5% (2.16 ꢀ 0.4 GBq from 8.1 GBq
of [18F]fluoride), radiochemical purity was 97.5%, and specific
activity was 78.4–134.5 GBq/mmol at the end of synthesis (n = 23).
Two-electron bioreduction
+ OR-
O
OH
Figure 1. Release of cytotoxic compounds under bioreductive conditions.
OCH3
OCH3 O
OCH3
O
Br
i
ii
iii
H
OCH3
OCH3
OCH3
O
5
4
2
3
iv
OCH3 OH
O
OH
v
OCH3
O
7
6
Reagents and conditions: i: HCl, MeOH, SnCl2; ii: Br2, Fe, CHCl3; iii: Mg, DMF, THF; iv: NaBH4; v: CAN, CH3CN
Figure 2. Synthesis of 2-hydroxymethyl 1,4-naphthoquinone precursor.
J. Label Compd. Radiopharm 2011, 54 788–794
Copyright © 2011 John Wiley & Sons, Ltd.
www.jlcr.org

U. Ackermann et al.
O
O
OH
O
OH
O
O
O
DCC/RT
+
F
F
O
8
7
[
19F]1
Figure 3. Synthesis of cold standard [19F]1.
O
O
OEt
O
OEt
O
OH
O
O
O
18F
i
ii
iii
+ NMe3
18F
18F]10
18F
[
18F]8
9
[
[
18F]1
Reagents and conditions: i: [18F]KF/kryptofix, 110 C; ii: NMe4OH, 90 C; iii: 2-hydroxymethylnaphthoquinone, DCC, 40 C
°
°
°
Figure 4. Synthesis of [18F]1.
The radiosynthesis including reformulation was fully automated controlled radical reaction, this catalytic system has the ability to
using a modified FDG synthesiser with two reactor vials.17 The total hydroxylate xenobiotic compounds. When conducted in vitro,
synthesis time including reformulation was 65 min.
It is worth noting that the addition of acetic acid after base generating system.22 In this study, we have decided to use an
hydrolysis is crucial to the success of the synthesis. Attempts to NADPH-generating system, which consisted of NADP, MgCl2
this assay requires the addition of either NADPH or an NADPH-
ꢂ
produce [18F]1 by omitting this step were unsuccessful.
6H2O, glucose 6-phosphate, and glucose 6-phosphate dehydro-
To our knowledge, this is also the first time that a DCC- genase.23 Since a maximum incubation time of 2 h is generally
mediated coupling was used for the synthesis of a PET radiotracer. considered sufficient for fluorine-18-labelled tracers, we used
The quantitative yield and short reaction time may make this an four time points (30, 60, 90, and 120 min) to measure tracer
interesting method for other applications.
stability.24 For each time point, 300 mL of a solution of S9 liver
fractions, the NADPH-generating system, and the reformulated
radiotracer were incubated at 37 ꢃC in an Eppendorf tube. After
the required incubation time, the reaction was quenched by
the addition of methanol, the sample centrifuged and analysed
by metabolite HPLC. Experiments were done in triplicates for sta-
tistical reliability.
Our assay revealed that the compound underwent rapid
metabolism, with only 14.4% of the tracer intact after 60 min
and 3.4% of the tracer still intact at the end of the observation
period of 120 min. Figure 5 shows the average per cent of intact
tracer over time with error bars. The percentage of unchanged 1
was fitted using Microsoft Excel software. The following single
exponential equation:
In vitro studies
Lipophilicity
The octanol/water coefficient was measured and the logP was
calculated to be 2.7, thus indicating that [18F]1 has a lipophilicity
suitable for crossing cell membranes. In the literature, the logP
values for FAZA and FMISO were calculated using HPLC retention
times and were found to be 1.1 and 2.6, respectively.18 However,
when measured using octanol/water distribution, the logP for
FMISO was found to be ꢁ0.33.19 In our laboratory, with the
octanol/water distribution method, the logP for FMISO was 0.05.
Stability and metabolite studies
y ¼ 100 ꢂ eꢁ0:0304t
The compound was found to be stable in human plasma, saline,
and potassium phosphate buffer (pH 7.4) for a period of more where y is the percentage of intact tracer was found to describe
than 2 h.
the radiotracer kinetics with R2 = 0.9756. From this formula, a bio-
As a model for radiotracer metabolism in vivo, we used a logical half-life of 22.8 min could be calculated.
phase 1 metabolism in vitro assay using rat S9 liver fractions.20
At least six polar radioactive metabolites could be detected
Phase 1 metabolism is predominantly catalysed by the cytochrome over the observed period of the assay. Figure 6 shows a repre-
P450 (CyP450) family of enzymes, which is found on the endo- sentative chromatogram of the S9 liver fraction assay at 60 min
plasmic reticulum, with the active site facing the cytosol.21 All post-incubation.
members of this family have a haeme group, which enables them
The retention time of [18F]1 under the conditions of the HPLC
to bind oxygen and receive electrons from NADPH. Through a conditions was 17.6 min. 4-[18F]Fluorobenzoic acid (8), which we
www.jlcr.org
Copyright © 2011 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2011, 54 788–794

U. Ackermann et al.
100
90
80
70
60
50
40
30
20
10
0
mice investigated was found to be 2.4 ꢀ 0.1. Figure 8 shows a
representative image of the uptake of [18F]1 in a BALB/c nude
mouse bearing a 495 mm3 SK-RC-52 tumour.
Average
Expon. (Average)
After imaging, the oxygen partial pressure in the centre and
periphery of all tumours was measured using an oxygen elec-
trode.25,26 All tumours were found to have pO2 levels of below
10 mm Hg in the central region, thus making them a good model
for tumour hypoxia. Owing to the heterogeneity of tumours, the
pO2 value measured at the periphery was found to be greater
than 10 mm Hg. The small-animal PET images also revealed that
the majority of the radioactivity was found in the bowel, and we
believe that clearance of [18F]1 occurs mainly through this organ.
Although tumour hypoxia imaging with [18F]1 seems feasible,
we believe that modifications to the chemical structure of this
lead compound are necessary to improve the in vivo and
in vitro properties. Depending on the substituents attached to
the 1,4- naphthoquinone core, reduction potentials between
ꢁ0.16 V and ꢁ0.42 V have been reported in the literature.27 The
reported value for [18F]FMISO (ꢁ0.389 V) falls well within this
range, and 1,4-naphthoquinone-based imaging agents therefore
offer the exciting prospect of fine-tuning hypoxia selectivity.6
The observed tumour-to-muscle ratio of [18F]1 (2.4 ꢀ 0.1) is lower
than the literature reported value for [18F]FMISO uptake (4.4 ꢀ 1)
in mice bearing a C3H mammary carcinoma.28 Potentially, this
may be improved if irreversible trapping of the imaging agent
in hypoxic tissue can be achieved.
0
20
40
60
80
100
120
Time (min)
Figure 5. Metabolism of [18F]1 over time with S9 liver fractions.
expected to be a major metabolite, was observed at 11.9 min. As
shown in Figure 7, this metabolite can either be formed by
simple hydrolysis of the starting material [18F]1 or by oxidation
of the benzylic function followed by elimination of [18F]8.
It is therefore surprising that the radioactive peak correspond-
ing to [18F]8 represents only 12.5% of the total radioactivity of all
metabolites at 60 min post-incubation. The appearance of a
polar metabolite at 11.4 min may point to hydroxylation of 4
[18F]fluorobenzoic acid ([18F]8), which could occur either before
or after cleavage from the 2-[(4-[18F]fluorobenzoyloxy)methyl]-
1,4-naphthalenedione ([18F]1) parent molecule.
A control incubation with the NADPH-generating system only
showed that [18F]1 was stable towards reduction by NADPH. We
have also found that [18F]FMISO does not undergo phase 1
metabolism with our S9 liver fraction assay over a period of 2 h
and is also stable towards NADPH reduction.
Experimental
General
No-carrier-added [18F]fluoride was produced by the 18O(p,n)18F
nuclear reaction with a 10 MeV proton beam generated by the
IBA Cyclone 10/5 cyclotron in a titanium target using [18O]H2O
at Austin Health, Centre for PET. Typical irradiation parameters were
20 mA for 30min, which resulted in 5.4–8.1 GBq (146–219 mCi) of
[18F]fluoride being transferred into the synthesis module. These
In vivo experiments
In vivo testing of the compound was carried out using trans- yields represent around 25% of the theoretical yield and are
planted SK-RC-52 tumours, which were grown in the shoulders explained by the use of recycled [18O]H2O of unknown isotopic
of BALB/c nude mice.25 A static, 10 min frame 2 h after tracer enrichment. Isolation of the [18F]fluoride ion from [18O]H2O was
administration was acquired for four mice to measure the achieved by trapping on a QMA ion exchange column. Elution of
tumour-to-muscle ratio. The tumour-to-muscle ratio for the four the column with
a
solution containing anhydrous K2CO3
mV(x10)
RAD
1.2
1.1
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
min
0.0
2.5
5.0
7.5
10.0
12.5
15.0
17.5
Figure 6. Representative HPLC chromatogram of [18F]1 metabolite assay after 60 min.
J. Label Compd. Radiopharm 2011, 54 788–794
Copyright © 2011 John Wiley & Sons, Ltd.
www.jlcr.org

U. Ackermann et al.
O
O
OH
O
O
O
OH
18F
18F
50
4
P
y
C
H
O
P
D
A
18F]8
N
/
[
2
O
O
O
O
18F
O
OH
h
y
d
r
o
l
a
s
e
[
18F]1
18F
18F]8
[
Figure 7. Potential metabolic pathways for the formation of [18F]8.
tumor
tumor
Figure 8. CT (left) and PET (centre) co-registered (right) sagittal view of [18F]1 uptake in an SK-RC-52 tumour-bearing mouse.
(0.025 mmol) and 20 mg of Kryptofix 2.2.2 (0.053 mmol) in 0.4 mL of 150 ꢂ 4.6 mm). Acetonitrile (A) and water (B) with 0.1% formic
acetonitrile plus 0.2 mL of water followed by repeated (3 times, acid were used as the mobile phase, and a gradient elution
1.5 mL each) azeotropic evaporation with acetonitrile to dryness technique was used for the analysis: 0–18 min: 45–90% A,
gave the anhydrous [18F]fluoride complex used in the labelling 18–30 min: isocratic 90% A at a flow rate of 0.5 mL/min. The
experiments.
same system was used for independent quality control. How-
Solvents were purchased from Merck and used as received. ever, the MS was replaced with an in-house built flow-through
Reagents were purchased from Sigma-Aldrich and used without detector with a Geiger-Müller tube for the detection of radio-
further purification. Flash chromatography was performed on a active compounds.
Grace Davison Reveleris™ Flash Chromatography System using
For radiotracer stability studies, a Shimadzu HPLC system
equipped with a 20 mL injection loop, a SPD-20A UV-Vis detector,
Reveleris™ Flash Cartridges.
Semi-preparative HPLC was performed using a Shimadzu LC- and two LC-20AD solvent pumps for high-pressure mixing of mobile
10AS isocratic pump equipped with a 5 mL injection loop and a phases were used. The stationary phase was a Grace Davison Vision
reversed phase column (Alltech Apollo™ C-18, 5 mm, reversed phase column (5 mm, 150 ꢂ 4.6 mm). Acetonitrile (A) with
10 ꢂ 250 mm). An isocratic mobile phase of 60:40 acetonitrile/ 0.1% formic acid and water (B) with 0.1% formic acid were used as
0.1 M ammonium formate was used as eluent at a flow rate of the mobile phase, and a gradient elution technique was used for
4 mL/min. Detection of chemical compounds was achieved with the analysis: 0–18 min: 5–90% A, 18–30 min: isocratic 90% A at a flow
a Shimadzu SPD-6AV UV detector (254 nm) and a Geiger-Müller rate of 0.5 mL/min. For the detection of radioactive compounds, a
tube as radiodetector.
Bioscan dual BGO coincidence detector was used. A CAT Thermo
For LCMS measurements, a Shimadzu 2010 LCMS system Shaker SH26.3 was used to incubate the radiotracer with rat S9 liver
equipped with a 5 mL injection loop, an SPD-20A UV-Vis detector, fractions and the NADPH-generating system. Statistical calculations
and two LC-20AD solvent pumps for high-pressure mixing of were performed using Microsoft Excel 2003.
mobile phases were used. The stationary phase was a
Phenomenex Gemini C-18 reversed phase column (10 mm, NMR spectrometer operating at 500 and 125 MHz, respectively.
1H and 13C NMR spectra were recorded on a Varian Unity 500
www.jlcr.org
Copyright © 2011 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2011, 54 788–794

U. Ackermann et al.
Chemistry
and the supernatant analysed by metabolite HPLC. No radioac-
tive metabolites could be detected.
Cold chemistry
Metabolite studies using rat S9 liver fractions
2-[(4-Fluorobenzoyloxy)methyl]-1,4-naphthalenedione. 4-Fluorobenzoic
acid (8) (140 mg, 1 mmol) and DCC (247 mg, 1.2 mmol) were
dissolved in 5 mL of dry DMSO, and the solution was stirred at
room temperature for 5 min. 2-Hydroxy 1,4-naphthoquinone
(7) (188 mg, 1 mmol) in 3 mL of dry acetonitrile was added,
and the reaction mixture was stirred for 24 h. Then, 10 mL of
water was added, and the mixture was purified using the Reve-
leris™ system with a 12 g Reveleris reversed phase C-18 column
as stationary phase and an acetonitrile (A)/water (B) gradient
system as mobile phase. The gradient system used was
0–3 min: 45% A, 3–19.8 min: 74% A, 19.8–22.5 min: 100% A,
and 22.5–30 min: 100% A. The peak at 14.8 min was collected,
and the solvent evaporated to give 264 mg (85%) of pure 1.
1H NMR (CDCl3) d: 5.41 (s, 2H), 6.99 (s, 1H), 7.158–7.236 (m, 4H),
7.78–7.89 (m, 2H), 8.103–8.121 (m, 2H).
Preparation of the NADPH-generating system. In 4 mL of 100 mM
potassium phosphate (pH 7.4) were dissolved 37.2 mg
(0.132 mmol) of glucose 6-phosphate, 39.8 mg (0.052 mmol) of
NADP, and 11.5 U of glucose 6-phosphate dehydrogenase. A
solution of 26.8 mg (0.132 mmol) of MgCl2 ꢂ 6H2O in 1 mL of
water was added to the mixture, and the mixture was stored
on ice.
S9 liver fraction assay. For each of the four time points (30, 60, 90,
and 120 min), 15 mL of rat S9 liver fractions (20 mg/mL), 40 mL of
the NADPH-generating system, 15 mL of the reformulated radio-
tracer, and 230 mL of 100 mM potassium phosphate buffer
(pH 7.4) (total volume 300 mL) were combined in an Eppendorf
tube and incubated at 37 ꢃC using a thermo shaker. After the
required incubation time, the reaction was quenched by adding
150 mL of methanol. The sample was then centrifuged for 10 min
at 11 000 rcf, and 20 mL of the supernatant was injected into the
metabolite HPLC.
13C NMR (CDCl3) d: 60.50, 115.75, 115.92, 125.51, 126.41,
126.52, 131.93, 132.38, 132.46, 133.66, 133.99, 134.19, 145.21,
164.72, 165.12, 167.15, 184.04, 184.48.
HRMS (C18H11FO4) (M + Na) measured (m/z): 333.05319, calcd.:
333.05391.
LogP measurement. LogP was measured by mixing 8 MBq (typi-
cally around 40 mL of the final formulation) of the tracer with
1 g each of 1-octanol and phosphate buffer (0.1 M, pH 7.4) in a
test tube. The test tube was vortexed for 3 min at room tempera-
ture, followed by centrifugation for 5 min at 9300 rcf. Two
weighed samples (0.5 g each) from the 1-octanol and buffer
layers were then measured for radioactivity using a Wallac
Wizard well counter. The partition coefficient was calculated
from the ratio of cpm/g of 1-octanol to that of buffer.
Radiochemistry
4-[18F]Fluorobenzoic acid ([18F]8). (4-Ethoxycarbonylphenyl)
trimethylammonium triflate (9) (5 mg, 20 mmol) in 1 mL of DMSO
was added to the dried [18F]KF/Kryptofix complex, and the
mixture was heated to 110 ꢃC for 15 min. Twenty microlitres of
an aqueous 1 M solution of tetramethylammonium hydroxide
(20 mmol) in 1 mL of acetonitrile was then added. The mixture
was heated to 110 ꢃC for 15 min in an open vial to evaporate
excess water through azeotropic distillation with acetonitrile
and hydrolyse the intermediate ethyl 4-[18F]fluoro benzoate
([18F]10). Then, 3 mL of 6% acetic acid and 6 mL of water were
added, and 4-[18F]fluorobenzoic acid ([18F]8) was loaded on a
C18 SepPak cartridge. After washing with 5 mL of water and
drying of the C-18 SepPak cartridge, 4-[18F]fluorobenzoic acid
([18F]8) was eluted with 1 mL of acetonitrile into a second reactor
vial.
In vivo animal experiments
All animal experiments were approved by the Austin Health
animal ethics committee.
Generation of tumour transplants
SK-RC-52 tumours were first grown by subcutaneously injecting
suspensions of 6 ꢂ 106 SK-RC-52 cells into the flank of BALB/c
nude mice. Tumours were allowed to grow to a size of
300 mm3, transplanted into the shoulder of BALB/c nude mice,
and allowed to grow to a size of ~500 mm3 for imaging studies.
2-[(4-[18F]Fluorobenzoyloxy)methyl]-1,4-naphthalenedione ([18F]1).
To the second reactor vial was then added a mixture 2 mg of 2
hydroxymethyl 1,4-naphthoquinone (7) (0.01 mmol) and 20 mg
of DCC (0.097 mmol) in 1 mL of acetonitrile. After heating to
40 ꢃC for 10 min, 3 mL of aqueous 0.1 M ammonium formate
was added, and the reaction mixture was purified using the
semi-preparative HPLC system as described above. The retention
time of [18F]1 was 16 min. Reformulation in 10% ethanol gave
2-(1,4-naphthoquinonyl)methyl 4-[18F]fluorobenzoate ([18F]1) in
27 ꢀ 5% radiochemical yield with a specific radioactivity of
78.4–134.5 GBq/mmol and a radiochemical purity of 97.5%.
Independent quality control was carried out using a radio-HPLC
system.
Small animal imaging
Imaging studies were performed using a Philips Mosaic small-
animal PET scanner. Animals were injected with 12.95 MBq of
radiotracer in 100 mL of the final formulation and anaesthetized
using isoflurane delivered by the Minerva Biovet animal imaging
system before scanning. The Minerva Biovet system also pro-
vides a temperature-stabilised environment for the animals
during the induction and imaging stages. Two hours after injec-
tion, a 10-min static frame was acquired and images were recon-
structed using the RAMLA3D algorithm.29 Following PET image
acquisition, the animals were relocated to a Philips Gemini PET/
CT and CT scanned at 90 kVp with a 150-mA tube current at
In vitro studies
Stability in saline and plasma
The in vitro stability of [18F]1 was assessed by incubating 1 mCi of 0.5 s per rotation. The acquired CT had a pixel size of 0.098 mm
radioactivity at 37 ꢃC in 2 mL of saline or plasma over a period of and a slice thickness of 0.6 mm. The resultant PET and CT images
2 h. Two millilitres of acetonitrile was added to the plasma were then imported into the PMOD analysis system for spatial
sample to precipitate the proteins. The sample was spun down, alignment and subsequent volume-of-interest analysis.
J. Label Compd. Radiopharm 2011, 54 788–794
Copyright © 2011 John Wiley & Sons, Ltd.
www.jlcr.org

U. Ackermann et al.
[6] M. Piert, H.-J. Machulla, M. Picchio, G. Reischl, S. Ziegler, P. Kumar,
H.-J. Wester, R. Beck, A. J. B. McEwan, L. I. Wiebe, M. Schwaiger, J.
Nucl. Med. 2005, 46, 106–113.
[7] L. Dubois, W. Landuyt, L. Cloetens, A. Bol, G. Bormans, K. Haustermans,
D. Labar, J. Nuyts, V. Grégoire, L. Mortelmans, Eur. J. Nucl. Med. Mol.
Imaging 2009, 36, 209–218.
[8] J. van Loon, M. H. M. Janssen, M. Öllers, H. J. W. L. Aerts, L. Dubois,
M. Hochstenbag, A.-M. C. Dingemans, R. Lalisang, B. Brans, B. Windhorst,
G. A. van Dongen, H. Kolb, J. Zhang, D. De Ruysscher, P. Lambin, Eur. J.
Nucl. Med. Mol. Imaging 2010, 37, 1663–1668.
[9] C. J. Koch, J. S. Scheuermann, C. Divgi, K. D. Judy, A. V. Kachur,
R. Freifelder, J. S. Reddin, J. Karp, J. B. Stubbs, S. M. Hahn, J. Driesbaugh,
D. Smith, S. Prendergast, S. M. Evans, Eur. J. Nucl. Med. Mol. Imaging
2010, 37, 2048–2059.
Polarographic pO2 measurements
After imaging, animals were humanely euthanized, and the
oxygen partial pressure in the tumour was measured using a
polarographic oxygen electrode.26 Tumour pO2 was measured
with a two-channel time-resolved luminescence-based optical
oxygen-sensing probe (Oxylite 2000, Oxford Optronix, Oxford, UK)
or Oxylite probe. The probes (230 mm o.d.) were precalibrated by
the manufacturer (ꢀ0.7 mmHg or <ꢀ10% of actual pO2, whichever
was greater). To further ensure correct pO2 readings in the
experiments, the probe was checked in normal saline and again
in animals just killed to ensure a 0 mm Hg recording.
[10] C. L. Falzon, U. Ackermann, N. Spratt, H. J. Tochon-Danguy, J. White,
D. Howells, A. M. Scott, J Labelled Cpd Radiopharm 2006, 49,
1089–1103.
Conclusion
2-(1,4-Naphthoquinonyl)methyl 4-[18F]fluorobenzoate ([18F]1) has
been synthesised as a putative hypoxia tracer from 2-hydroxymethyl
1,4-naphthoquinone (7) and 4-[18F]fluorobenzoic acid ([18F]8)
[11] I. Antonini, T.-S. Lin, L. A. Cosby, Y.-R. Dai, A. C. Sartorelli, J. Med.
Chem. 1982, 25, 730–735.
[12] C. Flader, J. Liu, R. F. Borch, J. Med. Chem. 2000, 43, 3157–3167.
[13] K. Karichiappan, D. Wege, Aust. J. Chem. 2000, 53, 743–747.
using DCC to activate [18F]8. The synthesis has been automated [14] H. Uno, J. Org. Chem. 1986, 51, 350–358.
using a modified FDG synthesiser. [18F]1 has been produced in a
radiochemical yield of 27 ꢀ 5%, with a radiochemical purity of
97.5% and a specific activity of 78.4–134.5GBq/mmol at the end
of synthesis.
[15] K. Kobayashi, M. Ushida, T. Uneda, K. Yoneda, M. Tanmatsu,
O. Morikawa, H. Konishi, J. Chem. Soc. Perkin. Trans. 1. 2001,
2977–2982.
[16] G. Tang, W. Zeng, M. Yu, G. Kabalka, J. Labelled Compd. Radiopharm.
2008, 51, 68–71.
[17] S. D. Yeoh, U. Ackermann, H. J. Tochon-Danguy, J. Sachinidis,
R. Mulligan, S. Poniger, J. Label. Compd Radiopharm. 2009, 52, Sup-
plement 1 S289
[18] P. Kumar, D. Stypinski, H. Xia, A. J. B. McEwan, H.-J. Machulla,
L. I. Wiebe, J. Labelled. Compds. Radiopharm. 1999, 42, 3–16.
The results of our in vivo evaluation indicate that imaging of
tumour hypoxia with radiolabelled quinones is feasible. How-
ever, improvements to the overall structure of [18F]1 are required
to make this molecule more stable towards liver metabolism and
achieve irreversible trapping in hypoxic cells. The formation of [19] R. Bejot, V. Kersemans, C. Kelly, L. Carroll, R. C. King, V. Gouverneur,
[18F]1 from the intermediate acid [18F]8 using DCC was fast
Nucl. Med. Biol. 2010, 37, 565–575.
[20] S. Yoshihara, M. Makishima, N. Suzuki, S. Ohta, Toxicol. Sci., 2001, 62,
and quantitative. We therefore believe that the use of DCC as a
coupling reagent in PET deserves further investigation.
221–227.
[21] C. J. Patten, Drug Discov. Today: Technol. 2006, 73–78.
[22] N. Plant, Drug Discov. Today. 2004, 9, 328–336.
[23] D. C. Ackley, K. T. Rockich, T. R. Baker, Methods in Pharmacology and
Acknowledgements
Toxicology: Optimization in Drug Discovery: In Vitro Methods, 1st
Edition 2004, 151–162, ISBN: 978-1-61737-499-9, Springer, Humana
This work was supported by a grant from the National Health
Press.
and Medical Research Council project (grant no. 469002) and a
[24] T. L. Ross, M. Honer, P. Y. H. Lam, T. L. Mindt, V. Groehn, R. Schibli,
grant from the Perpetual Trustees H&L Hecht Trust.
P. A. Schubiger, S. M. Ametamey, Bioconjugate Chem. 2008, 19,
2462–2470.
[25] N. Lawrentschuk, A. M. Poon, S. S. Foo, L. G. Putra, C. Murone,
I. D. Davis, D. M. Bolton, A. M. Scott, BJU Int. 2005, 96, 540–6.
[26] N. Lawrentschuk, F. T. Lee, G. Jones, A. Rigopoulos, A. Mountain,
G. O’Keefe, A. T. Papenfuss, D. M. Bolton, I. D. Davis, A. M. Scott,
Urol. Oncol. 2009, Jun 12. doi:10.1016/j.urolonc.2009.03.028
[Epub ahead of print]
[27] J. Koyama, I. Morita, N. Kobayashi, T. Osakai, H. Hotta, J. Takayasu,
H. Nishino, H. Tokuda, Cancer Lett. 2003, 201, 25–30.
[28] T. Grönroos, L. Bentzen, P. Marjamäki, R. Murata, M. R. Horsman,
S. Keiding, O. Eskola, M. Haaparanta, H. Minn, O. Solin, Eur. J.
Nucl. Med. Mol. Imaging. 2004, 31, 513–520.
References
[1] J. Overgaard, J. R. Horsmann, Sem Radiat Oncol. 1969, 6, 10–21.
[2] J. G. Rajendran, K. R. G. Hendrickson, A. M. Spence, M. Muzi, K. A. Krohn,
D. A. Mankoff, Eur. J. Nucl. Med. Mol. Imaging. 2006, 33, S44-S53.
[3] D. M. Brizel, R. K. Dodge, R. W. Clough, M. W. Dewhirst, Radiother.
Oncol. 1999, 53, 113–117
[4] B. Gagel, P. Reinartz, E. DiMartino, M. Zimny, M. Pinkawa, P. Maneschi,
S. Stanzel, K. Hamacher, H. H. Coenen, M. Westhofen, U. Buell, M. J.
Eble, Q. J. Nucl. Med. 2001, 45, 138–188.
[5] H. J. Tochon-Danguy, J. I. Sachinidis, F. Chan, J.G. Chan, C. Hall,
L. Cher, S. Stylli, J. Hill, A. Kaye, A. M. Scott, Nucl. Med. Biol. 2002,
29, 191–197.
[29] J. Browne, A. B. de Pierro, IEEE Trans. Med. Imaging, 1996, 15,
687–699.
www.jlcr.org
Copyright © 2011 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2011, 54 788–794