Synthesis of an acromelic acid a analog-based 11C-labeled PET tracer for exploration of the site of action of acromelic acid a in allodynia induction

Article abstract of DOI:10.1016/j.bmcl.2011.02.018

A novel ¹¹C-labeled PET (positron emission tomography) tracer, which was designed based on the (phenylthio)pyrrolidine derivative that can competitively block the acromelic acid A-induced allodynia, was synthesized. A protocol in which methylat

Full text of DOI:10.1016/j.bmcl.2011.02.018

Bioorganic & Medicinal Chemistry Letters 21 (2011) 2017–2020
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis of an acromelic acid A analog-based 11C-labeled PET tracer
for exploration of the site of action of acromelic acid A in allodynia induction
Masakatsu Kanazawa a, Kyoji Furuta a, , Hisashi Doi b, Tomoko Mori b, Toshiaki Minami c,
⇑
Seiji Ito d, Masaaki Suzuki b,
⇑
a Field of Biological Molecular Sciences, United Graduate School of Drug Discovery and Medical Information Sciences, Gifu University, 1-1 Yanagido, Gifu 501-1193, Japan
b RIKEN Center for Molecular Imaging Science (CMIS), 6-7-3 Minatojima-minamimachi, Kobe 650-0047, Japan
c Department of Anesthesiology, Osaka Medical College, 2-7 Daigakumachi, Takatsuki 569-8686, Japan
d Department of Medical Chemistry, Kansai Medical University, 10-15 Fumizono, Moriguchi 570-8506, Japan
a r t i c l e i n f o
a b s t r a c t
A novel 11C-labeled PET (positron emission tomography) tracer, which was designed based on the (phen-
ylthio)pyrrolidine derivative that can competitively block the acromelic acid A-induced allodynia, was
synthesized. A protocol in which methylation by palladium-mediated coupling of the boronate derivative
with [11C]CH3I and deprotection of the protected amino acid moiety are successively performed in one-
pot within 5 min was established for the synthesis of the tracer. The tracer is potentially useful as a tool to
investigate the mode of action of acromelic acid A in the induction of allodynia.
Article history:
Received 5 January 2011
Revised 2 February 2011
Accepted 4 February 2011
Available online 25 February 2011
Keywords:
Acromelic acid A
Allodynia
Ó 2011 Elsevier Ltd. All rights reserved.
PET tracer
Food poisoning due to a toadstool, Clitocybe acromelalga, causes
severe allodynia accompanied by erythema and burning pain at
one’s hands and feet.1 Allodynia refers to painful paresthesia in re-
sponse to tactile stimuli, which are normally perceived as innocu-
ous. The responsible toxin for the poisoning has not been identified
for a long time. However, Minami and Ito recently reported that
acromelic acid A2,3 (Fig. 1), a minor constituent isolated from C.
acromelalga, induces allodynia in mice by intrathecal (i.t.) adminis-
tration.4 This finding raises the possibility that acromelic acid A is
at least one of the key substances for the C. acromelalga poisoning
in humans.5
HOOC
NH
R
S
COOH
COOH
O
COOH
COOH
N
H
N
H
2: R = H
3a: R = 11CH3
3b: R = CH3
1, Acromelic acid A
Figure 1. Chemical structures of acromelic acid A, its analogs and a PET tracer.
Allodynia, together with spontaneous pain and hyperalgesia, is
the cardinal symptom of intractable neuropathic pain, such as pos-
therpetic neuralgia and complex regional pain syndrome. Although
the pathogenic mechanism of allodynia is not fully understood, the
involvement of prostaglandin, glutamate and ATP receptors in the
induction and maintenance of allodynia is reported.6–8 Thus, it has
tamic acid, and is known to cause the strong depolarization of neu-
rons.9,10 Accumulated studies have shown that kainoids bind to
and activate kainate and AMPA receptors, but not the NMDA sub-
type.11,12 Furthermore, acromelic acid A has a higher affinity for
AMPA receptor than kainate receptor,11 and its in vivo behavioral
and pathological effects are different from those of kainic acid.13,14
Based on these findings, acromelic acid A is considered to evoke
allodynia through AMPA receptor. However, extensive evidence
from pharmacological studies including the glutamate receptor
antagonism suggests the existence of an acromelic acid A-specific
receptor distinct from the already known ionotropic glutamate
receptors including AMPA receptor.4,15,16 If we can identify such
a receptor involved in the induction of allodynia, it may provide
a novel target to develop promising analgesic drugs for use in
the treatment of neuropathic pain. Thus, our objective is to prepare
been shown that i.t. administration of AMPA (
a-amino-3-hydroxy-
5-methylisoxazole-4-propionic acid) and NMDA (N-methyl-
D
-
aspartate), each being a specific agonist for the AMPA and NMDA
receptor subtypes among the ionotropic glutamate receptors,
respectively, cause allodynia in mice.7 Acromelic acid A belongs
to the class of kainoids, which include a structure like that of glu-
⇑
Corresponding authors. Tel.: +81 58 230 6503 (K.F.).
E-mail addresses: furuta@gifu-u.ac.jp (K. Furuta), suzuki.masaaki@riken.jp (M.
Suzuki).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.02.018

2018
M. Kanazawa et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2017–2020
a molecular probe for the identification and functional analysis of
the putative receptor for acromelic acid A.
PET tracers. Thus, approximately 5–6 min can be allocated to the
sequence of methylation–deprotection in the present case when
securing time for the post-reaction operations. This means that
the deprotection must be accomplished in one or fewer minutes
because the rapid methylation reportedly proceeds in about
5 min.24 In this context, we have devised a route sequentially con-
ducting the methylation and deprotection in one-pot in which the
organoboron compound 4 is utilized as the precursor of the PET
probe 3a (Scheme 1). The amino and two carboxy groups of 4 are
protected with trifluoroacetyl and methyl (as esters) groups,
respectively. Trifluoroacetamide and methyl ester groups have a
similar reactivity toward hydroxide,26 and therefore, both protect-
ing groups can easily be removed by treating with an aqueous alka-
line solution in a single operation. Furthermore, since the standard
protocol of the rapid methylation uses N,N-dimethylformamide
(DMF) as the solvent, we consider that the deprotection can be
achieved in one-pot by direct addition of an alkaline solution to
the rapid methylation reaction mixture without workup.
The allodynia-inducing effect of acromelic acid A can be evalu-
ated only by behavioral pharmacological studies in animals. Partic-
ularly, it should be emphasized that i.t. injected acromelic acid A
exerts its effect at extremely low doses (1 ag to 10 fg/mouse),4 sug-
gesting a lower level of expression of the receptor responsible for
allodynia induction. There is currently no information on what
kind of cells express the receptor that is activated by acromelic
acid A to induce allodynia. Thus, determining the target tissues
and types of cells in which acromelic acid A exerts its effect
in vivo is the first step to identify the receptor. Based on these cir-
cumstances, we intended to address this issue by using positron
emission tomography (PET), which enables the in vivo kinetic anal-
ysis of biologically active substances with an ultra-high sensitivity.
In this paper, we describe the design and synthesis of a PET probe
targeting the site of action of acromelic acid A.
The labeling of acromelic acid A itself by a positron-emitting
radionuclide is not an easy task because of its definitive molecular
structure. However, we previously reported that the simplified
acromelic acid A analog 2,17,18 with a phenylthio group instead of
pyridone ring at C-4 of the pyrrolidine skeleton (Fig. 1), potently
suppresses the allodynia induced by acromelic acid A.19 Thus the
acromelic acid A-induced allodynia was blocked in a dose-depen-
dent manner with an ID50 value of 2.19 fg/mouse, when 2 was i.t.
injected simultaneously with acromelic acid A (1 fg/mouse).19
The effect of 2 is specific to allodynia, not affecting thermal or
mechanical nociception or inflammatory pain. These facts strongly
suggest that both 2 and acromelic acid A possibly share the target
receptor with almost equal affinity. Additionally, the simplified
structure of 2 makes the structural modification as biochemical
probes easy. Consequently, we considered 2 to be an appropriate
compound for the basis of a PET probe design.
Before synthesizing the PET tracer 3a, we decided to prepare
and evaluate the biological activity of the nonradioactive molecule
3b. Thus, the intermediary compound 6 was synthesized from
trans-4-hydroxy-L-proline according to the reported proce-
dure.18,27,28 The reaction of 6 with di-p-tolyl disulfide in the pres-
ence of PBu3 afforded the coupling product 7 in 97% yield.29,30
Subsequent hydrolysis of 7 was achieved by heating in hydrochlo-
ric acid to give the desired compound 3b as a white solid in 95%
yield after ion-exchange chromatography and lyophylization
(Scheme 2).
We next assessed the effect of 3b on the acromelic acid A-in-
duced allodynia in mice to determine whether the designed mole-
cule is eligible for the PET tracer targeting the receptor in question.
First, we confirmed that 3b alone, like 2, did not induce allodynia
by the i.t. administration at 1–100 fg/mouse (data not shown).
The allodynic response of mice to tactile stimuli was then assessed
when 3b was i.t. injected simultaneously with acromelic acid A
(1 fg/mouse).4,19 As expected, 3b blocked the acromelic acid A-in-
duced allodynia in a dose-dependent manner with a potency
equivalent as 2 (Fig. 2),19 providing the validity of the design of
3a as the PET tracer.
Following this promising result, we next intended to establish
the one-pot, two-step procedure for preparing the PET tracer 3a.
To start with, the precursor organoboron derivative 4 was prepared
according to the pathway illustrated in Scheme 3. The phosphine-
mediated coupling of the preceding intermediary compound 6
with bis(4-bromophenyl) disulfide gave the (4-bromophenyl)thio
derivative 8 in 96% yield.29,30 Acidic hydrolysis of the amide and es-
ter moieties of 8 followed by re-esterification of the resulting acids
with methanol and re-amidation with trifluoroacetic anhydride re-
sulted in the formation of the protected (4-bromophenyl)thio
derivative 11 with the overall yield of 75% from 8 in three steps.
Finally, the palladium-catalyzed coupling of 8 with bis(pinacola-
to)diboron afforded the desired precursor 4 in 70% yield.31
By using the precursor 4, we examined the one-pot methyla-
tion-deprotection protocol for preparing the PET tracer under cold
conditions (Scheme 4). The rapid methylation of 4 was conducted
In general, 11C and 18F are utilized as positron-emitting radio-
nuclides for the synthesis of PET tracers, and a variety of methods
introducing such nuclides into chemical agents have been estab-
lished.20–23 Judging from the structure of 2 and applicability of
such labeling reactions, we designed 3a as a candidate for a PET
probe, in which the structural change from 2 is slight (only substi-
tution by a methyl) assuming not affecting the biological activity.
For the introduction of [11C]CH3, we planned to adopt our rapid
methylation of organoboron compounds with [11C]CH3I under pal-
ladium catalysis.24 The corresponding boron precursor, however, is
favorable to be protected at the amino acid moiety in its structure
because the free secondary amino group and carboxylate are
nucleophilic and may affect the efficiency of the methylation at
the boron moiety with [11C]CH3I. Based on this circumstance, an
additional deprotection process after rapid methylation is required
to obtain the desired PET tracer.
For preparation of a PET tracer with high radioactivity, it is gen-
erally assumed that the total synthesis time including labeling
reactions, HPLC purification and radiopharmaceutical formulation,
should be within about two times that of the half-life of a positron-
emitting nuclide.25 Since the half-life of 11C is 20.4 min, the total
synthesis time of 40 min meets the criteria for the 11C-labeled
O
11CH3
methylation
S
COOCH3
B
S
COOCH3
COOCH3
11CH3
S
COOH
COOH
deprotection
O
COOCH3
CF3
N
N
N
H
O
O
CF3
4
5
3a
one-pot
Scheme 1. Paradigm for one-pot, two-step synthesis of the acromelic acid A analog-based PET tracer.

M. Kanazawa et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2017–2020
2019
HO
COOt-Bu
COOCH3
HO
O
O
B
S
COOCH3
COOCH3
COOH
N
N
H
COPh
N
trans-4-hydroxy-L-proline
6
O
CF3
4
H3C
a
S
COOt-Bu
COOCH3
H3C
S
COOH
COOH
i, ii (one-pot)
N
COPh
N
H
7
3b
H3C
b
S
COOH
COOH
Scheme 4. One-pot synthesis of 3b by palladium-catalyzed coupling of 4 with MeI
followed by alkaline hydrolysis. Reagents and conditions: (i) MeI, Pd2(dba)3, P(o-
tolyl)3, K2CO3, DMF, 100 °C, 5 min; (ii) 2 M NaOH, 100 °C, 1 min, then HCOOH, 0 °C;
HPLC analysis, 93%.
N
H
3b
Scheme 2. Synthesis of 3b. Reagents and conditions: (a) di-p-tolyl disulfide, PBu3,
toluene, 90 °C, 97%; (b) 6 M HCl, reflux, 95%.
have succeeded in establishing the rapid one-pot, two-step proce-
dure applicable to the synthesis of the PET tracer 3a.
Based on the above-mentioned protocol, we performed the syn-
thesis of the PET tracer 3a. The precursor 4, Pd2(dba)3, P(o-tolyl)3
and K2CO3 were dissolved in anhydrous DMF in a vial under an
Ar atmosphere. [11C]CH3I, prepared as previously described,23
was transported into the vial by a stream of helium over a period
of 2 min at room temperature. After the solution was heated at
100 °C for 4 min, an aqueous NaOH solution was introduced into
the vial and heating was continued for another 1 min. Then, the
mixture was poured into a separate vial containing a solution of
formic acid in acetonitrile. The reaction vial was washed with
DMF/water and the washing was added to the acetonitrile solution.
The resulting mixture was subjected to reverse-phase HPLC purifi-
cation. The desired fractions were collected in a flask containing
25% ascorbic acid, and the organic solvent was removed under re-
duced pressure. The radiotracer was dissolved in saline. The total
synthesis time including HPLC purification and radiopharmaceuti-
cal formulation was 33 min. The radioactivity of 3a after formula-
tion of an injection solution for a PET study was up to 5.25 GBq. The
decay-corrected radiochemical yield based on the radioactivity of
trapped [11C]CH3I was 34–43%. The specific radioactivity of 3a
was 82 GBq/mmol. The chemical identity of 3a was confirmed by
co-injection with the authentic sample of 3b by analytical HPLC.
The chemical purity analyzed at 254 nm and the radiochemical
purity of 3a were both greater than 99%. A summary of the one-
pot synthesis of 3a is illustrated in Scheme 5.
Figure 2. Inhibitory effect of 3b on the acromelic acid A-induced allodynia. Each
dose of 3b was intrathecally injected with acromelic acid A (1 fg/mouse) into the
subarachnoid space of conscious mice. The effect of 3b on the allodynia was
evaluated 15 min after injection. Data are expressed as percent of the control (open
bar) in which the score for allodynia induced by acromelic acid A alone was taken as
100%. Values are the means of six mice.
following the reported standard procedure: CH3I dissolved in DMF
was added to a solution of 4, Pd2(dba)3, P(o-tolyl)3 and K2CO3 in
DMF, and the resulting mixture was heated at 100 °C for 5 min.24
To this was added an aqueous NaOH solution at the same temper-
ature, then heating was continued for a further 1 min. The reaction
vial was then quickly immersed in an ice-bath, and formic acid was
injected to quench the reaction. The resulting mixture was directly
subjected to HPLC analysis, confirming the formation of 3b in 93%
yield.32 Accordingly, it was proved that the deprotection step suf-
ficiently proceeds within 1 min, as expected. Taken together, we
In conclusion, we successfully synthesized an acromelic acid A
analog-based PET tracer by developing an efficient protocol in
which methylation with [11C]CH3I and deprotection are succes-
sively performed in one-pot within 5 min. The PET tracer is usable
HO
COOt-Bu
COOCH3
Br
S
COOt-Bu
COOCH3
Br
S
COOH
COOH
b
c
a
N
N
N
H
.
COPh
HCl
COPh
6
8
9
O
Br
S
COOCH3
COOCH3
B
S
COOCH3
COOCH3
Br
S
COOCH3
COOCH3
d
e
O
N
N
N
H
.
HCl
COCF3
COCF3
10
11
4
Scheme 3. Synthesis of the boron derivative 4. Reagents and conditions: (a) bis(4-bromophenyl) disulfide, PBu3, toluene, 90 °C, 96%; (b) 6 M HCl, reflux; (c) CH3OH, SOCl2,
reflux; (d) (CF3CO)2O, Et3N, DMAP, CH2Cl2, 75% (three-steps from 8); (e) bis(pinacolato)diboron, Pd(dba)2, PCy3, AcOK, dioxane, 100 °C, 70%.

2020
M. Kanazawa et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2017–2020
O
O
B
S
COOCH3
COOCH3
CF3
11CH3
S
COOH
COOH
i: methylation ii: hydrolysis iii: separation
4 min 1 min
N
N
H
O
one-pot, 5 min
within 40 min
4
3a
specific radioactivity: 82 GBq/µmol
Scheme 5. One-pot synthesis of 3a. Reagents and conditions: (i) Pd2(dba)3, P(o-tolyl)3, K2CO3, DMF, 11CH3I, 100 °C, 4 min; (ii) 2 M NaOH, 100 °C, 1 min, then HCOOH/CH3CN;
(iii) HPLC, reverse-phase, 0.5% HCOOH in H2O/CH3CN (gradient elution from 85/15 to 80/20), then concentration and radiopharmaceutical formulation.
6. Ito, S.; Okuda-Ashitaka, E.; Minami, T. Neurosci. Res. 2001, 41, 299.
7. Minami, T.; Matsumura, S.; Okuda-Ashitaka, E.; Shimamoto, K.; Sakimura, K.;
as a probe to explore the site of action of acromelic acid A in vivo,
and eventually to identify the target receptor involved in the
Mishina, M.; Mori, H.; Ito, S. Brain Res. 2001, 895, 178.
induction of allodynia. Imaging studies using the PET tracer in ro-
dents are now in progress. Neuropathic pain with allodynia as a
prominent symptom is an intractable disease to which anti-inflam-
matory agents and opioids are not effective. Ongoing PET studies
using our tracer will hopefully provide important clues to the elu-
cidation of the molecular mechanism of allodynia.
8. Tsuda, M.; Shigemoto-Mogami, Y.; Koizumi, S.; Mizokoshi, A.; Kohsaka, S.;
Salter, M. W.; Inoue, K. Nature 2003, 424, 778.
9. Ishida, M.; Shinozaki, H. Brain Res. 1988, 474, 386.
10. Ishida, M.; Shinozaki, H. Br. J. Pharmacol. 1991, 104, 873.
11. Kwak, S.; Aizawa, H.; Ishida, M.; Shinozaki, H. Neurosci. Lett. 1992, 139, 114.
12. Bunch, L.; Krogsgaard-Larsen, P. Med. Res. Rev. 2009, 29, 3.
13. Shinozaki, H.; Ishida, M.; Gotoh, Y.; Kwak, S. Brain Res. 1989, 503, 330.
14. Kwak, S.; Aizawa, H.; Ishida, M.; Shinozaki, H. Life Sci. 1991, 49. PL91.
15. Tsuji, K.; Nakamura, Y.; Ogata, T.; Mitani, A.; Kataoka, K.; Shibata, T.; Ishida, M.;
Shinozaki, H. Neuroscience 1995, 68, 585.
Acknowledgments
16. Kwak, S.; Nakamura, R. Brain Res. 1995, 702, 61.
17. Baldwin, J. E.; Pritchard, G. J.; Williamson, D. S. Bioorg. Med. Chem. Lett. 2000, 10,
1927.
18. Baldwin, J. E.; Pritchard, G. J.; Williamson, D. S. Tetrahedron 2001, 57, 7991.
19. Soen, M.; Minami, T.; Tatsumi, S.; Mabuchi, T.; Furuta, K.; Maeda, M.; Suzuki,
M.; Ito, S. Eur. J. Pharmacol. 2007, 575, 75.
This work was supported in part by a Grant-in-Aid for Scientific
Research from Japan Society for the Promotion of Science. The
authors thank Dr. Yasuyoshi Watanabe, Director of RIKEN Center
for Molecular Imaging Science (CMIS), for his strong support and
valuable advice.
20. Miller, P. W.; Long, N. J.; Vilar, R.; Gee, A. D. Angew. Chem., Int. Ed. 2008, 47,
8998.
21. Ametamey, S. M.; Honer, M.; Schubiger, P. A. Chem. Rev. 2008, 108, 1501.
22. Cai, L.; Lu, S.; Pike, V. W. Eur. J. Org. Chem. 2008, 2853.
23. Suzuki, M.; Sumi, K.; Koyama, H.; Siqin; Hosoya, T.; Takashima-Hirano, M.; Doi,
H. Chem. Eur. J. 2009, 15, 12489.
Supplementary data
24. Doi, H.; Ban, I.; Nonoyama, A.; Sumi, K.; Kuang, C.; Hosoya, T.; Tsukada, H.;
Suzuki, M. Chem. Eur. J. 2009, 15, 4165.
25. Fowler, J. S.; Wolf, A. P. Acc. Chem. Res. 1997, 30, 181.
26. Wuts, P. G. M.; Greene, T. W. Greene’s Protective Groups in Organic Synthesis, 4th
ed.; Wiley, 2007. Chapter 7.
27. Baldwin, J. E.; Bamford, S. J.; Fryer, A. M.; Rudolph, M. P. W.; Wood, M. E.
Tetrahedron 1997, 53, 5233.
28. Furuta, K.; Wang, G. X.; Minami, T.; Nishizawa, M.; Ito, S.; Suzuki, M.
Tetrahedron Lett. 2004, 45, 3933.
29. Nakagawa, I.; Hata, T. Tetrahedron Lett. 1975, 16, 1409.
30. Cleary, D. G. Synth. Commun. 1989, 19, 737.
31. Ishiyama, T.; Miyaura, N. J. Organomet. Chem. 2000, 611, 392.
32. The yield of 3b was determined based on HPLC using 5-methylresorcinol as an
internal standard.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.02.018.
References and notes
1. Saviuc, P. F.; Danel, V. C.; Moreau, P. A.; Guez, D. R.; Claustre, A. M.; Carpentier,
P. H.; Mallaret, M. P.; Ducluzeau, R. J. Toxicol. Clin. Toxicol. 2001, 39, 403.
2. Konno, K.; Shirahama, H.; Matsumoto, T. Tetrahedron Lett. 1983, 24, 939.
3. Konnno, K.; Hashimoto, K.; Ohfune, Y.; Shirahama, H.; Matsumoto, T. J. Am.
Chem. Soc. 1988, 110, 4807.
4. Minami, T.; Matsumura, S.; Nishizawa, M.; Sasaguri, Y.; Hamanaka, N.; Ito, S. Br.
J. Pharmacol. 2004, 142, 679.
5. Saviuc, P. F.; Danel, V. C. Toxicol. Rev. 2006, 25, 199.