Radiosynthesis of three [11C]ureido-substituted benzenesulfonamides as PET probes for carbonic anhydrase IX in tumors

Article abstract of DOI:10.1016/j.bmcl.2011.09.102

Three ureido-substituted benzenesulfonamides 1a-c have been developed as potent inhibitors for carbonic anhydrase IX, which is overexpressed in hypoxic tumors. In this study, we labeled these unsymmetrical ureas 1a-c using [ ¹¹C]phosgene ([

Full text of DOI:10.1016/j.bmcl.2011.09.102

Bioorganic & Medicinal Chemistry Letters 21 (2011) 7017–7020
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Radiosynthesis of three [11C]ureido-substituted benzenesulfonamides as PET
probes for carbonic anhydrase IX in tumors
Chiharu Asakawa a, Masanao Ogawa a,b, Katsushi Kumata a, Masayuki Fujinaga a, Tomoteru Yamasaki a,
Lin Xie a, Joji Yui a, Kazunori kawamura a, Toshimitsu Fukumura a, Ming-Rong Zhang a,
⇑
a Department of Molecular Probes, Molecular Imaging Center, National Institute of Radiological Sciences, 4-9-1 Anagawa, Inage-ku, Chiba 263-8555, Japan
b SHI Accelerator Service Co., Ltd, 5-9-11 Kitashinagawa, Shinagawa-ku, Tokyo 141-8686, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Three ureido-substituted benzenesulfonamides 1a–c have been developed as potent inhibitors for car-
bonic anhydrase IX, which is overexpressed in hypoxic tumors. In this study, we labeled these unsym-
metrical ureas 1a–c using [11C]phosgene ([11C]COCl2) as a labeling agent with the expectation that
Received 10 August 2011
Revised 21 September 2011
Accepted 27 September 2011
Available online 1 October 2011
[
11C]1a–c could become promising positron tomography probes for imaging carbonic anhydrase IX in
tumors. The strategy for radiosynthesis of [11C]1a–c was to react hydrochloride of anilines 2a–c with
[
11C]COCl2 to give isocyanate [11C]4a–c, followed by a reaction with 4-aminobenzenesulfonamide (3).
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
Carbonic anhydrase IX
[
11C]ureido-substituted
benzenesulfonamide
11C]phosgene
PET
[
Carbonic anhydrase IX (CA IX) is a multidomain protein with
CA IX-specific inhibitors. In particular, 4-{[(3-nitrophenyl)carbam-
oyl]amino}benzenesulfonamide (1a) was shown to significantly in-
hibit the formation of metastases in a model of breast cancer
metastases at a pharmacologic dose of 45 mg/kg, suggesting that
1a is a candidate for the development of novel antimetastatic
drugs. This finding motivated us to label 1a–c (Scheme 1) with pos-
itron-emitting carbon-11 (11C; half life: 20.4 min) as new positron
emission tomography (PET) probes for imaging CA IX in hypoxic
tumors.
the CA subdomain located outside the cell and is one of the most
overexpressed genes in hypoxic conditions.1 CA IX has recently
been shown to be a target for drug development and imaging of
hypoxic tumors, which have high expression of this protein.2–4 It
has been found that CA IX expression was significantly increased
in many types of solid tumors, including glioma/ependymomas,
mesotheliomas, and papillary/follicular carcinomas, carcinomas
of the bladder, uterine cervix, kidneys, esophagus, lungs, head
and neck, breast, brain, and vulva; and squamous/basal cell carci-
nomas.5,6 CA IX has high CO2 hydrase catalytic activity, which
causes CA IX to play a key role in the regulation of pH in tumors.7
Because hypoxic tumors do not respond to the general chemo- and
radiotherapy, inhibitors of CA IX in the hypoxic tumors may be-
come promising anticancer drugs with a novel action mecha-
nism.2–5
Recently, dozens of ureido-substituted benzenesulfonamide
analogs have been developed and show potent inhibition for sev-
eral human CA proteins.8 Of these CA inhibitors, compounds 1a–
c (Scheme 1) had high inhibitory affinity for human CA IX, with
Ki in the range of 0.3–5.4 nM. Moreover, these compounds showed
selectivity for other CA subtypes, such as CA I (Ki: 9–38 nM), CA II
(2.4–1060 nM), CA XII (4.6–50 nM). These data indicate that these
sulfonamides showed more potent inhibition for tumor growth as
PET is a potential and quantitative molecular imaging technique
with high functional sensitivity, which permits repeatable, non-
invasive assessment and understanding of biological and pharma-
cological processes.9 Development of PET probes for imaging
receptors, enzymes, and transporters has enabled PET to quantita-
tively evaluate the expression and pharmacological action of
receptors, etc. in humans, which contributes to disease diagnosis
and provides information about the most effective doses of drug
therapy. We have synthesized several anticancer drugs which tar-
get specific enzymes overexpressed in tumors, such as [11C]gefiti-
nib for epidemic growth factor receptor and tyrosine kinase,10,11
[
11C]sorafenib for vascular endothelial growth factor receptor and
Raf kinase,12 11C]topotecan for topoisomerase I.13 Using these
[
PET probes, we measured in vitro cellular uptake, determined
in vivo pharmacokinetics in animals and performed imaging of tu-
mor-bearing rodents.
In the present study, we synthesized [11C]ureido-substituted
benzenesulfonamides ([11C]1a–c, Scheme 1) with the expectation
⇑
Corresponding author. Tel.: +81 43 382 3708; fax: +81 43 206 3261.
E-mail address: zhang@nirs.go.jp (M.-R. Zhang).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.09.102

7018
C. Asakawa et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7017–7020
R
Ki for CA IX (nM)
SO2NH2
O
C
0.9
5.4
0.3
1a: R = 3-NO2
1b: R = 4-CH3CO
1c: R = 2-CN
N
H
N
H
R
R
SO2NH2
SO2NH2
O
O
11C
11C
+
+
Cl
Cl
N
N
NH2
H2N
H
H
[
[
[
11C]1a: R = 3-NO2
2a-c
3
11C]1b: R = 4-CH3CO
11C]1c: R = 2-CN
Scheme 1. Chemical structure of 1a–c and retrosynthesis of [11C]1a–c.
that these radioligands may become promising PET probes for
imaging CA IX in tumors. Previously, an imaging study of CA IX
with an 177Lu-labeled human monoclonal antibody has been per-
formed on nude mice which bear adenocarcinoma xenografts
expressing rich CA IX.14,15 However, to our knowledge, there was
no report about the development and application of PET probes
for imaging CA IX in hypoxic tumors.
The preparation of [11C]COCl2 for the present radiosynthesis is
routinely performed using
a home-made synthetic unit (>2
times/week).19,20 As shown in Scheme 2, [11C]COCl2 was produced
starting from cyclotron-produced [11C]CO2 via [11C]CH4 and then
[
11C]CCl4 as two intermediates.18,19 After irradiation, [11C]CO2
was recovered from the cyclotron target and concentrated on the
inner space of a stainless steel tube under liquid N2. Release of
Regarding the chemical structures of 1a–c with an unsymmet-
rical urea moiety, we decided to label these compounds using
[
11C]CO2 from the tube and continuous passage of [11C]CO2 with
H2 of 10 mL/min through a heated methanizer at 400 °C gave
[
11C]phosgene ([11C]COCl2) as a labeling agent16,17 (Scheme 1). This
[
11C]CH4.21 A mixture of [11C]CH4 and Cl2 gas was then passed
approach involves convenient and reliable production of
[
through a heated quartz tube at 560 °C with N2 of 50 mL/min to af-
ford [11C]CCl4. [11C]CCl4 was passed through a glass tube22 coated
with oxidizing agents at room temperature to produce [11C]COCl2.
This multi-step process gave [11C]COCl2 at an average of 75% de-
cay-corrected radiochemical yield (n > 50) based on the total
11C]COCl2,18–20 the usefulness of two different amines: 3-nitro
(2a), 4-acetyl (2b) or 2-cyano (2c) anilines and 4-aminobenzene-
sulfonamide (3), and construction of a [11C]urea site, as shown in
the retro-synthetic route.
Prior to radiosynthesis, the non-radioactive benzenesulfona-
mides 1a–c were prepared by heating commercially available iso-
cyanates 4a–c with 3 in anhydrous CH3CN at 50 °C for 2–8 h with
26–50% chemical yields (Scheme 2). Compounds 1a–c were ana-
lyzed by mp, NMR and high resolution MS, which were identical
to those reported previously8 (also see Supplementary data).
[
11C]CO2, which took about 10 min from the end of bombardment.
The convenient and reliable production of [11C]COCl2 paved the
way for radiolabeling with this agent.
Constructing a moiety of unsymmetrical [11C]urea by the reac-
tion of [11C]COCl2 with two different amines has been a challeng-
ing task.23–26 The formation of symmetrical [11C]urea due to the
Chemical Synthesis
SO2NH2
R
R
SO2NH2
O
3
H2N
C
a
N
N
N
C O
H
H
1a: R = 3-NO2
1b: R = 4-CH3CO
1c: R = 2-CN
4a-c
Radiosynthesis
H2, Ni
Cl2
c
I2O5, H2SO4
11CO2
11CCl4
11COCl2
11CH4
b
d
R
R
R
SO2NH2
O
11C
O
11C
e
3
f
+
N 11
C
O
Cl
Cl
.
NH2
N
N
H
HCl
H
11C]4a-c
[
[
[
11C]1a: R = 3-NO2
11C]1b: R = 4-CH3CO
11C]1c: R = 2-CN
.
2a-c HCl
[
Scheme 2. Chemical synthesis and radiosynthesis. Reagents and conditions: (a) CH3CN, 50 °C, 2–8 h, 26–50%; (b) 400 °C, 2 min; (c) 560 °C, 2 min; (d) room temperature,
2 min, 70–80% from [11C]CO2; (e) THF, ꢀ15 °C, 1 min; (f) THF, 60 °C, 3 min, 21–70% (incorporation of total radioactivity in the final reaction mixture).

C. Asakawa et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7017–7020
7019
largely exceeded that of [11C]COCl2, [11C]4a could easily react with
2a to give the symmetrical [11C]urea.
To prevent formation of the symmetrical [11C]urea, we used
hydrochloride of 2a (2aꢁHCl) in place of the free 2a as a starting
material (Scheme 2). Although the nucleophilicity of 2aꢁHCl was
decreased by this treatment, its reactivity remained enough to re-
act with [11C]COCl2 perfectly to give [11C]4a, even at ꢀ15 °C. On the
other hand, due to the decreased reactivity of 2aꢁHCl, [11C]4a did
not further react with excess 2aꢁHCl at ꢀ15 °C. After 1-min trap-
ping of [11C]COCl2, subsequent treatment with 3 gave the desired
[
11C]1a with 70% incorporation yield of the total radioactivity in
the reaction mixture. Moreover, no symmetrical [11C]urea of 2a
or 3 was confirmed in the reaction mixture.
According to the reaction conditions examined here, we carried
out automated synthesis of [11C]1a using a multi-step synthesis
system, which was previously developed in our institute.19,20,27
[
11C]COCl2 was bubbled into
1.02
of 3 (0.35 mg, 2.03
a
solution of 2aꢁHCl (0.18 mg,
mol in 300 mL THF) at ꢀ15 °C for 1 min, followed by reaction
mol in 300 mL THF) at 60 °C for 3 min. After
l
l
the two-step reactions, THF was evaporated by N2 flow. The reac-
tion mixture was diluted and loaded onto a reversed-phase HPLC
system.28 Figure 1A shows a representative HPLC chromatogram
for separation, which shows that [11C]1a was produced as the main
radioactive peak in the reaction mixture. Starting from 18–21 GBq
of [11C]CO2, 0.74–0.88 GBq of [11C]1a was obtained at the end of
synthesis (n = 5). The total synthesis time was averaged as
38 min from the end of bombardment.
The identity of [11C]1a was confirmed by co-injection with non-
radioactive 1a on analytic HPLC.27 In the final product solution, the
radiochemical purity of [11C]1a was higher than 99% (Fig. 1B) and
specific activity was 30 GBq/lmol. No significant UV peaks corre-
sponding to 2a, 3 and other chemical impurities were observed
on the HPLC chart of the finally-formulated product. Moreover,
the radiochemical purity of [11C]1a remained >98% after 90 min at
room temperature, and this product was radiochemically stable
within the period of one PET scan. This analytical result was in com-
pliance with our in-house quality control/assurance specifications.
Table 1 shows the synthetic results of [11C]1a–c. In addition to
Figure 1. Typical chromatograms from the HPLC separation (A) and analysis (B)
used in the radiosynthesis of [11C]1a.
reaction of [11C]COCl2 with two molecules of the same amine is a
main problem, which significantly decreases the labeling efficiency
of unsymmetrical [11C]urea. In the present experiment, after
[
11C]1a, the acetyl analog [11C]1b was also obtained with enough
radioactivity by reacting [11C]COCl2 with hydrochloride of 40-ami-
noacetophenone (2bꢁHCl) in a moderate incorporation yield of
radioactivity (Scheme 2). However, the radioactivity amount of
[
11C]COCl2 gas had been bubbled into THF solution containing a
[
11C]1c, which was obtained by reacting
mixture of 3-nitroaniline (2a; 0.28 mg, 2.03
(0.35 mg, 2.03
lmol) and 3
the cyano analog
[
11C]COCl2 with hydrochloride of 2-aminobenzonitrile (2cꢁHCl),
l
mol) at ꢀ15 °C for 1 min, no radioactive peak cor-
responding to [11C]1a was found in the reaction mixture. Two sym-
metrical [11C]ureas of 2a and 3 were found in the mixture,
respectively. On the other hand, after [11C]COCl2 gas had been
trapped into THF solution of 2a at ꢀ15 °C for 1 min, subsequent
treatment with 3 did not afford [11C]1a. Only the symmetrical
was lower than those of [11C]1a and [11C]1b. By analyzing the reac-
tion mixture, we found that [11C]1c gradually decomposed within
the reaction and purification processes. Although several agents
(ascorbic acid, EtOH, etc.) for preventing radiolysis had been added
to the reaction mixtures, decomposition of [11C]1c over time could
not be stopped.
[
11C]urea of 2a was formed in the reaction mixture. These results
indicate that isocyante [11C]4a (Scheme 2), which was yielded as
an intermediate by reaction of [11C]COCl2 with one molecule of
2a, was difficult to retain in the reaction mixture containing excess
2a within the trapping of [11C]COCl2. Because the mass of 2a
Because it has been reported that 1a had an observatory in vivo
inhibitory effect on breast cancer metastases,8 we selected [11C]1a
for preliminary in vitro evaluation. Cellular uptake of [11C]1a was
measured using HT-29 human colorectal adenocarcinoma cells,15
Table 1
Radiosynthesis results of [11C]1a–c
Radioligand
R
Amount of radioactivitya (MBq)
Radiochemical purity (%)
Specific activityc (GBq/
lmol)
Synthesis time fromd (min)
[
[
[
11C]1a
11C]1b
11C]1c
3-NO2
4-CH3CO
2-CN
810 32
750 81
155 20b
99
97
95
31 10
34 11
38
37
38
2
5
3
39
8
a
b
c
Using 18.5–22.2 GBq of cyclotron-produced [11C]CO2.
The low yield was due to radiolysis of [11C]1c within the reaction and purification processes.
n P 3 at the end of synthesis.
d
n P 3 at the end of bombardment.

7020
C. Asakawa et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7017–7020
which have high CA IX expression and MCF-7 breast tumor cells,15
which have low CA IX expression, respectively. At 15 min and
60 min after incubation with [11C]1a, the uptake level of radioac-
tivity in HT-29 cells was 1.2-fold and 1.6-fold higher than that in
MCF-7 cells, respectively. This result suggests that [11C]1a might
specifically accumulate in CA IX-rich tumor cells, although the dif-
ference in the uptake between HT-29 and MCF-7 cells was not as
high as we anticipated. Because of the short half-life of 11C, we
could not measure the uptake of [11C]1a in both tumor cell lines
until after a longer time. To achieve a greater difference in the cel-
lular uptake between the two cell lines, we will try to label 1a with
longer half-life positron emitters, such as 18F (109 min) and 124I
(4.15 d).
6. Said, H. M.; Supuran, C. T.; Hageman, C.; Staab, A.; Polat, B.; Katzer, A.;
Scozzafava, A.; Anacker, J.; Flentje, M.; Vordemark, D. Curr. Pharm. Des. 2010,
16, 3288.
7. Hilvo, M.; Baranauskiene, L.; Salzano, A. M.; Scaloni, A.; Matulis, D.; Innocenti,
S.; Scozzafava, A.; Monti, S. M.; Di Fiore, A.; De Simone, G.; Lindfors, M.; Janis, J.;
Valjakka, J.; Pastorelova, S.; Pastorek, J.; Kulomaa, M. S.; Nordlund, H. R.;
Supuran, C. T.; Parkkila, S. J. Biol. Chem. 2008, 283, 27799.
8. Pacchiano, F.; Carta, F.; McDonald, P. C.; Lou, Y.; Vullo, D.; Scozzafava, A.;
Dedhar, S.; Supuran, C. T. J. Med. Chem. 2011, 54, 1896.
9. Fowler, J. S.; Volkow, N. D.; Wang, G.-J.; Ding, Y.-S.; Dewey, S. L. J. Nucl. Med.
1999, 40, 1154.
10. Kawamura, K.; Yamasaki, T.; Yui, J.; Hatori, A.; Konno, F.; Kumata, K.; Konno, F.;
Irie, T.; Fukumura, T.; Suzuki, K.; Kanno, I.; Zhang, M.-R. Nucl. Med. Biol. 2009,
36, 239.
11. Zhang, M.-R.; Kumata, K.; Hatori, A.; Takai, N.; Toyohara, J.; Yanamoto, K.;
Yamasaki, T.; Yui, J.; Kawamura, K.; Koike, S.; Ando, K.; Suzuki, K. Mol. Imag.
Biol. 2010, 12, 181.
12. Asagawa, C.; Ogawa, M.; Kumata, K.; Fujinaga, M.; Yamasaki, T.; Kato, K.; Yui, J.;
Kawamura, K.; Hatori, A.; Fukumura, T.; Zhang, M.-R. Bioorg. Med. Chem. Lett.
2011, 21, 2220.
13. Yamasaki, T.; Fujinaga, M.; Kumata, K.; Kawamura, K.; Yui, J.; Hatori, A.; Zhang,
M.-R. Nucl. Med. Biol. 2011, 38, 707.
14. Ahlskog, J. K. J.; Dumelin, C. E.; Trussel, S.; Marlind, J.; Neri, D. Bioorg. Med.
Chem. Lett. 2009, 19, 4851.
15. Ahlskog, J. K. J.; Schliemann, C.; Marlind, J.; Ammar, A.; Pedley, R. B.; Neri, D. Br.
J. Cancer. 2009, 101, 645.
In conclusion, [11C]1a–c were synthesized using [11C]COCl2 as a
labeling agent, via the intermediate preparation of isocyanate
[
11C]4a–c. The labeling route was reliable and reproducible, which
guaranteed enough radioactivity of [11C]1a for in vitro and further
in vivo evaluation. We are preparing animal models bearing sev-
eral tumors and will perform PET imaging with [11C]1a for the
modeled animals. PET imaging with [11C]1a may be a useful tool
to study CA IX in the hypoxic tumors.
16. Miller, P. M.; Long, N. J.; Vilar, R.; Gee, A. D. Angew. Chem., Int. Ed. 2008, 47,
8998.
17. Roeda, D.; Dollé, F. Curr. Top. Med. Chem. 2010, 10, 1680.
18. Nishijima, K.; Kuge, Y.; Seki, K.; Ohkura, K.; Motoki, N.; Nagatsu, K.; Tanaka, A.;
Tsukamoto, E.; Tamaki, N. Nucl. Med. Biol. 2003, 29, 345.
19. Ogawa, M.; Takada, Y.; Suzuki, H.; Nemoto, K.; Fukumura, T. Nucl. Med. Biol.
2010, 37, 73.
20. Takada, Y.; Ogawa, M.; Suzuki, H.; Nemoto, K.; Fukumura, T. Appl. Radiat. Isot.
2010, 68, 1715.
21. Link, J. M.; Krohn, A. J. Labelled Compd. Radiopharm. 1997, 40, 306.
22. This tube, which is produced by Komyo Rikagaku Kogyo (Tokyo, Japan), is
called as a Kitagawa gas detection tube used to measure the CCl4 concentration
in working environment.
Acknowledgments
We are grateful to Mrs. Y. Yoshida and N. Nengaki (SHI Acceler-
ator Service Co., Ltd) for technical support with radiosynthesis. We
also thank the staff of the National Institute of Radiological Sci-
ences for support with the cyclotron operation and radionuclide
production.
23. Conway, T.; Diksic, M. J. Nucl. Med. 1988, 29, 1957.
24. Lidstrom, P.; Benasera, T. A.; Marquez-M, M.; Nilsson, S.; Bergstron, M.;
Langstrom, B. Steroids 1998, 63, 228.
Supplementary data
25. Lemoucheux, L.; Roudlen, J.; Ibezizene, M.; Sobrio, F.; Kasne, M. C. J. Org. Chem.
2003, 68, 7289.
26. Dollé, F.; Martarello, L.; Bramoulle, Y.; Bottlaendler, M.; Gee, A. D. J. Labelled
Compd. Radiopharm. 2005, 48, 501.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.09.102.
27. Fukumura, T.; Suzuki, H.; Mukai, K.; Zhang, M.-R.; Yoshida, Y.; Nemoto, K.;
Suzuki, K. J. Labellled Compd. Radiopharm. 2007, 50, S202.
28. HPLC was performed using a JASCO HPLC system (JASCO, Tokyo). Column: Capcell
Pac C18 (Shiseido, Tokyo), 10 mm ꢂ 250 mm for purification, 10 mm ꢂ 250 mm
for analysis; detector: UV 254 nM.
References and notes
1. Alterio, V.; Hilvo, M.; Di Fiore, A.; Supuran, C. T.; Pan, P.; Pakkila, S.; Scaloni, A.;
Pastorek, J.; Pastorekova, S.; Pedone, C.; Scozzafava, A.; Monti, S. M.; De Simone,
G. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 16233.
[
11C] 1a, purification: CH3CN/H2O = 40/60 (0.1% trifluoroacetic acid (TFA)),
2. Supuran, C. T. Nat. Rev. Drug Disc. 2008, 7, 168–181.
5.0 mL/min, 8.5 min; analysis: CH3CN/H2O = 40/60 (0.1% TFA), 1.0 mL/min,
8.2 min.
3. Ozensoy, O.; De Simone, G.; Supuran, C. T. Curr. Med. Chem. 2010, 17, 1516.
4. De Simone, G.; Supuran, C. T. Biochim. Biophys. Acta. 2010, 1804, 404.
5. Wykoff, C. C.; Beasley, N. J.; Watson, P. H.; Turner, K. J.; Pastorek, J.; Sibtain, A.;
Wilson, G. D.; Turley, H.; Talks, K. L.; Maxwell, P. H.; Pugh, C. W.; Ratcliffe, P. J.;
Harris, A. L. Cancer Res. 2000, 60, 7075.
[
11C] 1b, purification: CH3CN/H2O = 35/65 (0.1% TFA), 4.0 mL/min, 9.7 min;
analysis: CH3CN/H2O = 30/70 (0.1% TFA), 1.5 mL/min, 6.0 min.
11C] 1c, purification: CH3CN/H2O = 45/65 (0.1% TFA), 4.5 mL/min, 12.0 min;
analysis: CH3CN/H2O = 45/65 (0.1% TFA), 1.2 mL/min, 7.3 min.
[