Synthesis and in vitro evaluation of 18F-labelled S-fluoroalkyl diarylguanidines: Novel high-affinity NMDA receptor antagonists for imaging with PET

Article abstract of DOI:10.1016/j.bmcl.2010.01.052

Two S-[¹⁸F]fluoroalkylated diarylguanidines were synthesized and evaluated in vitro as potential tracers for imaging of N-methyl-d-aspartate receptors (NMDARs) with positron emission tomography (PET). [¹⁸F]1 and [¹⁸F]10 we

Full text of DOI:10.1016/j.bmcl.2010.01.052

Bioorganic & Medicinal Chemistry Letters 20 (2010) 1749–1751
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and in vitro evaluation of 18F-labelled S-fluoroalkyl
diarylguanidines: Novel high-affinity NMDA receptor antagonists
for imaging with PET
Edward G. Robins a, Yongjun Zhao a, Imtiaz Khan b, Anthony Wilson b, Sajinder K. Luthra a, Erik Årstad a,
*
a GE Healthcare MDx Research, MRC Cyclotron Building, Hammersmith Hospital, London W12 0NN, United Kingdom
b GE Healthcare MDx Research, The Grove Centre, Amersham HP7 9LL, United Kingdom
a r t i c l e i n f o
a b s t r a c t
Two S-[18F]fluoroalkylated diarylguanidines were synthesized and evaluated in vitro as potential tracers
for imaging of N-methyl- -aspartate receptors (NMDARs) with positron emission tomography (PET).
Article history:
Received 17 November 2009
Revised 5 January 2010
Accepted 6 January 2010
Available online 20 January 2010
D
[
18F]1 and [18F]10 were synthesized by [18F]fluoroethylation and [18F]fluoromethylation of the thiol pre-
cursor 6, respectively. [18F]1 is a promising candidate NMDAR PET tracer, with low nanomolar affinity for
the NMDA PCP-site, high selectivity and moderate lipophilicity.
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Glutamate
NMDA
PET
Fluorine-18
Imaging
N-Methyl-
D
-aspartate receptors (NMDARs) are ligand-gated
tightly controlled and highly complex function of NMDARs repre-
sents a formidable challenge for drug development, which is
further compounded by the lack of validated tracers for quantifica-
tion of NMDAR function in vivo. Over the last two decades, there
has been a concerted effort to develop tracers for imaging of
NMDARs with positron emission tomography (PET) and single pho-
ton emission computed tomography (SPECT).7 A number of PCP
site ligands have been labelled with either 11C, 18F or 123I and eval-
uated for imaging of NMDAR function, including derivatives of (+)-
MK801,8–11 memantine,12,13 ketamine14 and phencyclidine15,16
heteromeric ion channels that regulate fast neurotransmission by
allowing prolonged influx of Ca2+, Na+ and K+ ions, into the syn-
apse. NMDARs are widely distributed in the mammalian brain
and play a key role in neurodevelopment, neuroplasticity and neu-
roprotection.1,2 NMDARs have been extensively studied for their
role in pathological conditions, and NMDAR dysfunction has been
implicated in a wide range of neurological and psychiatric diseases,
most notably ischaemic stroke, epilepsy, neuropathic pain, Alzhei-
mer’s disease, drug addiction, and schizophrenia.1–3,7 Despite long-
term efforts to develop NMDAR antagonists for therapeutic inter-
vention, clinical trials have so far been unsuccessful.3,4
NMDARs are regulated by six binding sites for endogenous li-
gands, which control the frequency and duration of ion channel
opening. NMDAR activation requires simultaneous binding of glu-
tamate and the co-agonist glycine, as well as depolarisation of the
post-synaptic terminal where they are expressed. This leads to a
conformation change and release of an intra-channel bound Mg2+
cation, which open the channel for ion influx. The exact combina-
tion of subunits and splice variants defines the pharmacological
and physiological properties of the individual NMDAR subtypes
(NR2A–D and NR3A–B) as described in recent reviews.2,3,5,6 The
18
NH
2
F
11
NH CH
NH
18
N
S
18
3
F
O
CH
3
F
18
18
11
18
[
F]F-MEM
IC = 6000 nM
50
[
F]FE-TCP
= 38 nM
[
C]Ketamine
= 1200 nM
[
F]MK 801
= 79 nM
K
K
K
d
d
d
11
Cl
H
CH
N
Cl
CH
3
H
3
11
N
O
CH
N
N
SCH
3
CH
N
3
H
3
123
N
I
NH
NH
NH
SCH
3
SCH
3
123
11
[
11
[
I]CNS 1261
C]GMOM
[
C]CNS 5161
K
= 4.2 nM
K
d
= 5.2 nM
K
= 1.9 nM
*
Corresponding author at present address: Institute of Nuclear Medicine,
d
d
University College London, 235 Euston Road, London NW1 2BU, United Kingdom.
Tel.: +44 (0) 2076792344.
Chart 1. Examples of previously reported NMDA PCP tracers for imaging with PET
E-mail address: e.arstad@ucl.ac.uk (E. Årstad).
and SPECT.
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.01.052

1750
E. G. Robins et al. / Bioorg. Med. Chem. Lett. 20 (2010) 1749–1751
(Chart 1). Imaging studies with these tracers have produced incon-
clusive results, which have been attributed to high non-specific
binding, poor metabolic stability or insufficient affinity.11,17,18
N-(1-Naphthyl)-N0-(3-[123I]-iodophenyl)-N0-methylguanidine
([123I]CNS 1261) is arguably the most promising NMDAR tracer re-
ported to date.19,20 It has high-affinity (Kd 4.2 0.4 nM) and selec-
tivity for the NMDA PCP site and moderate lipophilicity (Log D
2.19). In clinical studies, the brain regional values of volume of dis-
tribution (Vt) of [123I]CNS 1261 correlate with NMDAR density, and
there is evidence of at least partial displacement of the tracer in
healthy subjects following administration of (S)-ketamine, a drug
known to compete for the NMDA PCP site.21–23 To enable high res-
olution quantitative imaging of NMDARs with PET, we and others
have developed 11C-labelled diarylguanidines structurally related
to CNS 1261, including [11C]CNS 5161 (Ki = 1.9 0.3 nM)24,25 and
The affinity of the S-fluoroalkyl guanidines 1 and 10 for the
NMDA PCP site was measured by inhibition of [3H]N-[1-(2-thie-
nyl)cyclohexyl]-3,4-piperidine ([3H]TCP) binding in rat cerebral
cortex homogenates (MDS Pharma Services, Taiwan).32 Non-
radioactive CNS 5161 was included as a positive control. As shown
in Table 1, the S-fluoroalkyl substituents are well tolerated for this
compound class and 1 and 10 effectively retain the affinity of the
parent compound CNS 5161 (Kd 1.78 0.1 nM). Our results suggest
that the 2-fluoroethyl substituent in 1 causes a modest drop in
affinity (Kd 2.35 0.2 nM), whereas the fluoromethyl group in 10
increases the affinity slightly (Kd 1.38 0.2 nM). The measured
affinity of CNS 5161 in this study is in good agreement with previ-
ously published data (Kd 1.87 0.25 nM),27 and for all compounds
Hill coefficients close to one were obtained suggesting binding to a
single site on the NMDA receptor.
[
11C]GMOM (Ki = 5.2 0.3 nM).26
The guanidines 1 and 10 were radiolabelled by alkylation of the
thiol precursor 6 with [18F]fluoroethyl tosylate and [18F]fluorom-
ethyl tosylate, respectively (Scheme 2). 2-[18F]Fluoroethyl tosylate
(74–126 MBq) was prepared in 10–17% radiochemical yield (non-
decay-corrected, 90 min from end of bombardment) by a modifica-
Herein, we report the synthesis and in vitro evaluation of S-fluo-
roalkyl diarylguanidines labelled with 18F (half-life 110 min).27 As
the half-life of 11C (20.4 min) restricts applications to centres with
cyclotron facilities, development of 18F-labelled tracers is required
to enable widespread PET studies and routine clinical applications.
18F also offers the potential for longer data acquisition and the use
of bolus/infusion paradigms, as have been developed for [123I]CNS
1261.28
tion of the procedure reported by Block et al.33 18F]Fluoromethyl
[
tosylate (148–192 MBq) was obtained in 20–26% radiochemical
yield (non-decay-corrected, 90 min from end of bombardment)
using the method reported by Neal et al.34 In both cases, [18F]fluor-
oalkylation of 6 was carried out in acetonitrile in the presence of
Cs2CO3 at 110 °C for 15 min. Following purification by reversed
phase semi-preparative HPLC, the guanidines [18F]1 and [18F]10
were concentrated on a C18 Sep-Pak cartridge, eluted with ethanol
We initially attempted to prepare S-fluoroethyl diarylguanidine
1 by condensation of the S-fluoroethyl thioether 2 with cyanamide
3 (Scheme 1). 4-Chloro-3-amino-thiophenol 429 was obtained in
good yield by tin(II) chloride mediated reduction of 4-chloro-3-ni-
tro-benzenesulfonyl chloride (5), and cyanamide 3 was obtain in
two steps from 3-(thiomethyl)aniline as previously described.30
Alkylation of 4 with 2-fluoroethyl tosylate afforded the S-fluoro-
ethyl thioether 2 in good yield. In our hands, reaction of the cyan-
amide 3 with the hydrochloride salt of 2 led to decomposition of
the starting materials and failed to yield the desired guanidine 1.
Attempted reaction of the aniline 4 with 3 to provide the guanidine
6 was also unsuccessful, possibly due to a competing reaction of
the thiol group with the cyanamide functionality, or alternatively,
decomposition of the aniline 4 under the relatively harsh condi-
tions employed. In the light of the encountered difficulties, the
thiol group was protected by alkylation of 4 with benzyl bromide
to yield the aniline 7. Under optimised conditions, condensation
of 7 with cyanamide 3 provided the corresponding guanidine 9
in 74% yield. Using a modification of the procedure reported by
Simão et al.,31 deprotection of the benzyl thioether 9 with alumin-
ium(III) chloride afforded the thiol 6 in 64% yield. Finally, alkyl-
ation of 6 with 2-fluoroethyl tosylate and fluorochloromethane
proceeded in high yield to provide the target compounds 1 and
10, respectively.
(200 lL) and formulated for injection in either 0.9% w/w isotonic
saline or phosphate buffered saline. On standing at room tempera-
ture, [18F]10 was observed by analytical radio-HPLC to slowly
decompose to give a polar radioactive product consistent with
[
[
[
18F]fluoride. After standing for 2 h the radiochemical purity of
18F]10 was reduced from >95% to ꢀ85%. Due to its low stability
18F]10 was excluded from further characterisation. No decomposi-
tion was observed for [18F]1. The target [18F]fluoroalkylated guani-
dine
radiochemical yields from
rected), with radiochemical purity >95% and specific activity in
the range of 1.13–2.50 GBq/ mol. The total synthesis time was
[
18F]1 (30–67 MBq) was obtained in 4–9% overall
18F]fluoride (n = 6, non-decay-cor-
[
l
3–4 h from end of bombardment. The modest specific activity
achieved reflects the use of low levels of radioactivity in these
experiments.
Lipophilicity is a good indicator of the ability of tracers to pas-
sively diffuse across the blood–brain barrier, and is also known to
Table 1
Binding affinity to the NMDA PCP site
a
Ligand
Kd (nM)
gH
Cl
Cl
Cl
Cl
Me
N
H
NO2
NH2
NH2
N
SMe
iv
1
10
CNS 5161
2.35 0.2
1.38 0.2
1.78 0.1
1.01 0.02
0.92 0.14
0.93 0.003
iii
i
Me
N
NH
MeS
SO2Cl
SH
SBn
CN SBn
9
a
4
5
7
Data are means SEM. Each experiment was repeated in duplicate.
3
v
ii
Cl
Me
N
Cl
Me
N
Cl
H
H
Cl
Me
N
N
SMe
N
SMe
vi
NH2
H
N
SMe
i
3
18F OTs
n
ii
NH
TsO OTs
NH
F
n
NH
SH
6
F
S
S
n
F
S
n
6
2
1 n = 2
10 n = 1
[
[
18F]1 n = 2
18F]10 n = 1
Scheme 1. Synthesis of S-fluoroalkyl guanidines. Reagents and conditions: (i) SnCl2,
concd HCl, reflux 1 h; (ii) 2-fluoroethyl tosylate, NaOMe, MeOH, 60 °C, 12 h; (iii) n-
BuLi, BnBr, THF, 0 °C to rt, 1 h; (iv) 160 °C, N2, 3 h; (v) AlCl3, toluene, 2 h; (vi) NEt3,
CH2ClF(g), CH2Cl2, rt, or K2CO3, acetone, 2-fluoroethyl tosylate, reflux, 72 h.
Scheme 2. Radiosynthesis of 18F-labelled diarylguanidines. Reagents and condi-
tions: (i) K18F/Kryptofix 2.2.2, MeCN, 90 °C, 15 min; (ii) thiol precursor 6, Cs2CO3,
MeCN, 110 °C, 15 min.

E. G. Robins et al. / Bioorg. Med. Chem. Lett. 20 (2010) 1749–1751
7. Waterhouse, R. N. Nucl. Med. Biol. 2003, 30, 869. and references therein.
1751
affect non-specific binding and metabolic stability. The distribu-
tion coefficient (log D7.4) of 1 was found to be 2.49 0.1,35 which
is well within the range considered ideal for brain tracers, and as
expected, similar to the log D value reported for CNS 5161 (log D7.5
2.68).26
The 2-fluoroethyl substituted guanidine 1 was subjected to a
broad receptor binding assay (MDS Pharma Services, Taiwan) to
determine its selectivity for the NMDA PCP site. At a concentration
8. Kiesewetter, D. O.; Finn, R. D.; Rice, K. C.; Monn, J. A. Int. J. Rad. Appl. Instrum. [A]
1990, 41, 139.
9. Andersson, Y.; Tyrefors, N.; Sihver, S.; Onoe, H.; Watanabe, Y.; Tsukada, H.;
Långström, B. J. Label. Compd. Radiopharm. 1998, 41, 567.
10. Brady, F.; Luthra, S. K.; Pike, V. W. Appl. Radiat. Isot. 1989, 40, 325.
11. (a) Brown, D. R. P.; Wyper, D. J.; Owens, J.; Patterson, J.; Kelly, R. C.; Hunter, R.;
McCulloch, J. J. Psychiatr. Res. 1997, 31, 605; (b) Owens, J.; Wyper, D. J.;
Patterson, J.; Brown, D. R. P.; Elliott, A. T.; Teasdale, G. M.; McCulloch, J. Nucl.
Med. Commun. 1997, 18, 149.
12. Samnick, S.; Ametamey, S.; Leenders, K. L.; Vontobel, P.; Quack, G.; Parsons, C.
G.; Neu, H.; Schubiger, P. A. Nucl. Med. Biol. 1998, 25, 323.
of 1 lM, significant binding was observed to 10 out of 60 receptors
and ion channels investigated, most notably the cardiac potassium
channel (hERG, 56% inhibition) and voltage gated sodium channel
(site 2, 47% inhibition). Moderate to low binding was observed to
Sigma 1 and 2 receptors, the norepinephrine transporter, GABAA,
5-HT5A and opiate receptors. However, at 10 nM concentration
no significant binding was observed, confirming that the radioli-
gand [18F]1 has high selectivity for the NMDA PCP binding site.
In summary, introduction of S-fluoroalkyl substituents on the
N-aryl group of CNS 5161 is well tolerated and retain the affinity
and selectivity of the parent compound for the NMDA PCP binding
site. Labelling of the thiol precursor 6 with 2-[18F]fluoroethyl and
13. Ametamey, S. M.; Bruehlmeier, M.; Kneifel, S.; Kokic, M.; Honer, M.; Arigoni,
M.; Buck, A.; Burger, C.; Samnick, S.; Quack, G.; Schubiger, P. A. Nucl. Med. Biol.
2002, 29, 227.
14. Shiue, C.-Y.; Vallabhahosula, S.; Wolf, A. P.; Dewey, S. L.; Fowler, J. S.; Schlyer,
D. J.; Arnett, C. D.; Zhou, Y.-G. Nucl. Med. Biol. 1997, 24, 145.
15. Haradahira, T.; Sasaki, S.; Maeda, M.; Kobayashi, K.; Inoue, O.; Tomita, U.;
Nishikawa, T.; Suzuki, K. J. Label. Compd. Radiopharm. 1998, 41, 843.
16. Orita, K.; Sasaki, S.; Maeda, M.; Hashimot, A.; Nishikawa, T.; Yugami, T.; Umezu,
K. Nucl. Med. Biol. 1993, 20, 865.
17. Sihver, S.; Sihver, W.; Andersson, Y.; Murata, T.; Bergström, M.; Onoe, H.;
Matsumura, K.; Tsukada, H.; Oreland, L.; Långström, B.; Watanabe, Y. J. Neural
Transm. 1998, 105, 117.
18. Blin, J.; Denis, A.; Yamaguchi1, T.; Crouzel1, C.; MacKenzie, E. T.; Baron, J. C.
Neurosci. Lett. 1991, 121, 183.
19. Owens, J.; Tebbutt, A. A.; McGregor, A. L.; Kodama, K.; Magar, S. S.; Perlman, M.
E.; Robins, D. J.; Durant, G. J.; McCulloch, J. Nucl. Med. Biol. 2000, 27, 557.
20. Knol, R. J. J.; De Bruin, K.; Van Eck-Smit, B. L. F.; Pimlott, S.; Wyper, D. J.; Booij, J.
Synapse 2009, 63, 557.
21. Erlandsson, K.; Bressan, R. A.; Mulligan, R. S.; Gunn, R. N.; Cunningham, V. J.;
Owens, J.; Wyper, D.; Ell, P. J.; Pilowsky, L. S. Nucl. Med. Biol. 2003, 30, 441.
22. Stone, J. M.; Erlandsson, K.; Årstad, E.; Bressan, R. A.; Squassante, L.; Teneggi, V.;
Ell, P. J.; Pilowsky, L. S. Nucl. Med. Biol. 2006, 33, 239.
[
[
18F]fluoromethyl tosylate provided the radioligands [18F]1 and
18F]10, respectively, which to the best of our knowledge are the
first low nanomolar 18F-labelled NMDA PCP site antagonists re-
ported to date. In particular, the 2-fluoroethyl substituted guani-
dine [18F]1 warrants further evaluation as a potential PET tracer
for imaging of NMDAR activity in vivo.
23. Stone, J. M.; Erlandsson, K.; Årstad, E.; Squassante, L.; Teneggi, V.; Bressan, R. A.;
Krystal, J. H.; Ell, P. J.; Pilowsky, L. S. Psychopharmacology 2008, 197, 401.
24. Zhao, Y.; Robins, E. G.; Turton, D.; Brady, F.; Luthra, S. K.; Årstad, E. J. Label.
Compd. Radiopharm. 2006, 49, 163.
Acknowledgements
25. Biegon, A.; Gibbs, A.; Alvarado, M.; Ono, M.; Taylor, S. Synapse 2007, 61, 577.
26. Waterhouse, R. N.; Dumont, F.; Sultana, A.; Simpson, N.; Laruelle, M. J. Label.
Compd. Radiopharm. 2002, 45, 955.
We thank cyclotron operators at Hammersmith Imanet Ltd for
providing 18F.
27. Robins, E.G.; Arstad, E. Patent WO2006136846A1, 2006.
28. Bressan, R. A.; Erlandsson, K.; Mulligan, R. S.; Gunn, R. N.; Cunningham, V. J.;
Owens, J.; Cullum, I. D.; Ell, P. J.; Pilowsky, L. S. Nucl. Med. Biol. 2004, 31, 155.
29. Allert, R. Chem. Ber. 1881, 14, 1434.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2010.01.052.
30. Hu, L. Y.; Guo, J.; Magar, S. S.; Fischer, J. B.; Burke-Howie, K. J.; Durant, G. J. J.
Med. Chem. 1997, 40, 4281.
31. Simão, D.; Alves, H.; Belo, D.; Rabaça, S.; Lopes, E. B.; Santos, I. C.; Gama, V.;
Duarte, M. T.; Henriques, R. T.; Novais, H.; Almeida, M. Eur. J. Inorg. Chem. 2001,
3119.
32. In vitro ligand binding assays were performed by MDS Pharma Services Ltd,
Taiwan. http://www.mdsps.com/.
References and notes
1. Dingledine, R.; Borges, K.; Bowie, D.; Traynelis, S. F. Pharmacol. Rev. 1999, 51, 7.
2. Lau, C. G.; Zukin, R. S. Nat. Rev. Neurosci. 2007, 8, 413.
33. Block, D.; Coenen, H. H.; Stöcklin, G. J. Label. Compd. Radiopharm. 1988, 25, 201.
34. Neal, T. R.; Apana, S.; Berridge, M. S. J. Label. Compd. Radiopharm. 2005, 48, 557.
3. Kalia, L. V.; Kalia, S. K.; Salter, M. W. Lancet Neurol. 2008, 7, 742.
4. (a) Walters, M. R.; Bradford, A. P.; Fischer, J.; Lees, K. R. Br. J. Clin. Pharmacol.
2002, 53, 305; (b) Gamzu, E. Eur. Neuropsychopharm. 1995, 5, 171; (c) Muir, K.
W.; Grosset, D. G.; Gamzu, E.; Lees, K. R. Br. J. Clin. Pharmacol. 1994, 38, 33.
5. Paoletti, P.; Neyton, J. Curr. Opin. Pharmacol. 2007, 7, 39.
35. The n-octanol:10 mM sodium phosphate buffer partition (Log D7.4
)
determination for compound 1 was performed by Syngene International Ltd
following the method described by Mondal, S. K.; Mondal, N. B.; Banerjee, S.;
Mazumder, U. K. Indian J. Pharmacol., 2009, 41, 176.
6. Cavara, N. A.; Hollmann, M. Mol. Neurobiol. 2008, 38, 16.