Synthesis and characterization of N-(2-chloro-5-methylthiophenyl)-N′- (3-methylthiophenyl)-N′-[11C]methylguanidine [ 11C]CNS 5161, a candidate PET tracer for functional imaging of NMDA receptors

Article abstract of DOI:10.1002/jlcr.1033

N-methyl-D-aspartate (NMDA) receptors play a key role in excitatory neurotransmission and are linked to a variety of acute and chronic neurodegenerative diseases including epilepsy, schizophrenia, Parkinson's disease and drug abuse. N-(2-chloro-5-methylth

Full text of DOI:10.1002/jlcr.1033

JOURNAL OF LABELLED COMPOUNDS AND RADIOPHARMACEUTICALS
J Label Compd Radiopharm 2006; 49: 163–170.
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jlcr.1033
Research Article
Synthesis and characterization of N-(2-chloro-5-
methylthiophenyl)-N0-(3-methylthiophenyl)-N0-
[11C]methylguanidine [11C]CNS 5161, a candidate PET
tracer for functional imaging of NMDA receptors
Yongjun Zhao, Edward Robins, David Turton, Frank Brady,
Sajinder K. Luthra and Erik Arstad*
Hammersmith Imanet, Cyclotron Building, Du Cane Road, London W12 0NN, UK
Summary
N-methyl-d-aspartate (NMDA) receptors play a key role in excitatory neurotransmis-
sion and are linked to a variety of acute and chronic neurodegenerative diseases
including epilepsy, schizophrenia, Parkinson’s disease and drug abuse. N-(2-chloro-5-
methylthiophenyl)-N0-(3-methylthiophenyl)-N0-methylguanidine (CNS 5161) is a high
affinity ligand (Ki ¼ 1:87 ꢀ 0:25 nM) for the NMDA PCP site, which potentially can
be used for functional imaging of this receptor. Herein we report the synthesis of the
corresponding positron emission tomography (PET) tracer [11C]CNS 5161 by means
of [11C]methylation of the desmethyl guanidine precursor. [11C]CNS 5161 was
synthesized with a decay corrected radiochemical yield of 10% within 45 min after
end of bombardment (EOB). The final product was prepared in a sterile saline
solution suitable for clinical studies with a radiochemical purity of >96% and a
specific activity of 41 GBq/mmol at time of injection. Copyright # 2005 John Wiley
& Sons, Ltd.
Key Words: NMDA; glutamate; CNS 5161; PET; radiotracer
Introduction
NMDA receptors are gated ion-channels that mediate long-term potentiation
by allowing prolonged influx of Ca2+ ions, as well as Na+ and K+ ions, into
the synapse.1,2 They are tightly regulated, with a total of six binding sites for
endogenous ligands controlling the frequency and duration of ion-channel
opening.3 Receptor activation requires depolarization of the neuron on which
they are expressed and simultaneous binding of the co-agonists glutamate and
*Correspondence to: Erik Arstad, Hammersmith Imanet, Cyclotron Building, Du Cane Road, London
W12 0NN, UK. E-mail: erik.arstad@csc.mrc.ac.uk
Copyright # 2005 John Wiley & Sons, Ltd.
Received 28 July 2005
Revised 2 November 2005
Accepted 5 November 2005

164
Y. ZHAO ET AL.
glycine. The exact pharmacological and physiological properties of NMDA
receptors are dictated by their subunit composition. Functional NMDA
receptors are composed of heteromeric combinations of the ubiquitously
expressed NR1 subunit, and the discretely distributed NR2 subunit, which has
four subtypes (NR2A-D). Additional complexity arise from splicing of the
NR1 subunit, which is known to exist in eight different isoforms, and from
co-assembling of one or more NR3 subunits, of which two subtypes have been
described to date (NR3A and NR3B).4 Each subunit expresses distinct binding
sites and different combinations produce receptors with unique permeation,
kinetic and pharmacological properties.
The role of NMDA receptors in pathological process has been extensively
studied and NMDA dysfunction is linked to a variety of acute and chronic
neurodegenerative diseases, including epilepsy, stroke, neuropathic pain,
schizophrenia, Parkinson’s disease, and drug abuse.5–9 In this context, non-
invasive molecular imaging and positron emission tomography (PET) in
particular, represents a unique tool to study NMDA receptors in the living
human brain. Imaging agents targeting the intrachannel PCP site are of
particular interest as they selectively bind to NMDA receptors in the open
state, and their distribution and density should therefore reflect dynamic
changes in the channel itself.10 This can potentially be used to quantify
regional populations of activated receptors in the healthy and diseased brain,
as well as to monitor the function of therapeutic drugs targeting this receptor
system. A number of potential imaging agents for the PCP site have been
synthesized (Figure 1),11 of which the SPECT tracer [123I]CNS 1261 (1) is the
first to show promise in clinical trials.12–14 Candidate PET tracers, including
the related N,N0 diarylguanidine [11C]GMOM (2), have suffered from low
specific binding in addition to the usual pitfalls of rapid metabolism and low
brain uptake.11,15 We believe CNS 5161 (3), a lead therapeutic candidate from
16–18
the N,N0diarylguanidine series,
is more attractive for PET tracer develo-
pment due to its superior affinity for the PCP site (1.87 ꢀ 0.25 nM versus
5.2 ꢀ 0.3 nM for GMOM) and favourable log P of 2.68. Herein, we report
the synthesis and characterization of N-(2-chloro-5-methylthiophenyl)-N0-(3-
methylthiophenyl)-N0[11C]methylguanidine, an 11-carbon labelled derivative
of CNS 5161.
Materials and methods
General
Reagents and solvents were obtained from Aldrich Chemical Co. and Fisher
Scientific and used without further purification unless otherwise noted. NMR
spectroscopy was carried out at the University of London Chemistry
Department on a Varian Gemini-300 (300 MHz) instrument or at Imperial
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 163–170

SYNTHESIS AND CHARACTERIZATION OF [11C]CNS 5161
165
Cl
CH3
N
Cl
CH3
N
H
N
H
N
CH3
N
H
N
I
NH
NH
NH
SCH3
SCH3
SCH3
OCH3
CNS 5161
GMOM
CNS 1261
Ki= 1.87 nM
Ki= 5.2 nM
Ki= 4.65 nM
Figure 1. Structures of selected N,N0-diarylguanidine ligands for the NMDA
PCP site. Data from Dumont et al.18
College, Hammersmith Campus on a JEOL GX500 (500 MHz) instruments
and the chemical shifts are reported in parts per million (ppm) relative to
residual solvent peaks. LRMS, HRMS and MS–MS analyses were performed
by Mass Spectrometry Services at the School of Pharmacy, London. LRMS
experiments were carried out using 50% acetonitrile/0.1% formic acid as
solvent. HRMS and MS–MS experiments were performed using a Thermo-
Quest Navigator mass spectrometer operated under ESI mode using a
Micromass Q-TOF Global Tandem mass spectrometer under ESI mode using
50% acetonitrile/0.1% formic acid as solvent and calibrated using [Glu]-
Fibrinopeptide B as internal standard. Flash column chromatography was
performed on 230 mesh silica gel or an Iso Inc. CombiFlash Companion using
pre-packed 4, 12 and 40 g RediSep silica columns. [11C]CNS 5161 (3) was
purified by reverse-phase HPLC using a Beckman 110B HPLC pump
connected to a SA 6506 UV detector (Severn Analytical) set at 254 nm, a
GM tube and a Servogor 124 twin pen chart recorder (Spectronic). Analytical
HPLC for quality control was carried out on an Agilent 1100 series HPLC
system equipped with a Bioscan GM tube detector.
Synthesis
3-Thiomethylphenylcyanamide (4), 18N-methyl-3-thiomethylphenylcyanamide
(5), 182-chloro-5-thiomethylaniline (6)16 and CNS 5161 (3)16 were synthesized
according to published procedures.
N-(2-chloro-5-methylthiophenyl)-N0-(3-methylthiophenyl)-guanidine HCl (7). To
3-thiomethylphenylcyanamide (4) (0.46 g, 2.81 mmol) was added 2-chloro-5-
thiomethylaniline hydrochloride (6) (0.54 g, 2.56 mmol) and chlorobenzene (2 ml),
and the stirred reaction mixture was heated to 1508C under an inert atmosphere
(N2) for 4 h. The resulting solution was allowed to reach room temperature
and concentrated under reduced pressure. Purification of the residue by means of
flash chromatography (silica, CH2Cl2:MeOH gradient, 0–10%) provided the title
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 163–170

166
Y. ZHAO ET AL.
compound as a white solid (0.72 g, 76%). M.p. 168.5–1708C; 1H NMR (CDCl3) d
(ppm) 2.49 (s, 6H, S–CH3) and 6.8–7.3 (m, 7H, Aryl H).
N-methyl-(2-chloro-5-methylthio)aniline (8). To a solution of (2-chloro-5-
thiomethyl)aniline (0.599 g, 3.75 mmol) in THF (10 ml) cooled in an ice-bath
was slowly added 1 equivalent of n-butyllithium (1 M hexanes, 3.75 mmol)
under stirring. After 15 min methyl iodide (0.532 g, 3.75 mmol) was added
dropwise and the resulting solution was allowed to reach room temperature
over a period of 1 h. Solvents were evaporated and the residue separated
between water (50 ml) and diethyl ether (50 ml). The organic phase was dried
(MgSO4), filtered and concentrated under reduced pressure. The resulting
crude product was purified by means of flash chromatography (silica, 1:1
dichloromethane/hexane) to give the title compound 8 (0.556 g, 79%) as a
colourless oil. 1H NMR (CDCl3) d 2.53 (s, 3H, S-Me), 2.96 (s, 3H, N-Me), 4.40
(br s, 1H, NH), 6.59 (dd, 1H, Aryl H, J ¼ 2:2 and 8.4 Hz), 6.65 (d, 1H, Aryl H,
J ¼ 2:2 Hz), 7.19 (d, 1H, Aryl H, J ¼ 8:4 Hz). The corresponding hydro-
chloride salt was prepared by treating an ice-cold solution of 8 in diethyl ether
(5 ml/mmol 8) with a 2-fold molar excess of HCl in diethyl ether (anhydrous,
2 M). Filtration provided the salt as a bright white solid (0.621 g, 94%).
N-(2-chloro-5-methylthiophenyl)-N0-(3-methylthiophenyl)-N-methylguanidine HCl
(9). To N-methyl-(2-chloro-5-methylthio)aniline (8) hydrochloride (0.100 g,
0.45 mmol) was added 3-methylthiophenylcyanamide (4) (0.074 g, 0.45 mmol),
toluene (0.5 ml) and dichloromethane (0.5 ml) under an inert nitrogen
atmosphere. The reaction mixture was heated to 1208C and concentrated by
allowing the dichloromethane solvent to be removed from the reaction mixture
by distillation. After 12 h the mixture was cooled to room temperature and the
residue purified by column chromatography (silica, chloroform/methanol
gradient, 0–10%). The resulting colourless oil was taken up in a minimum of
chloroform and 9.HCl was precipitated by addition of diethyl ether. The
1
product was isolated by filtration as a white solid (0.054 g, 31%). H NMR
(d6-DMSO) d 2.48 (s, 3H, S-Me), 2.52 (s, 3H, S-Me), 3.38 (s, 3H, N-Me),
7.14–6.97 (br m, 2H, Aryl H), 7.16 (m, 1H, Aryl H), 7.34 (m, 2H, Aryl H), 7.55
(m, 2H, Aryl H).
Cold methylation of N-(2-chloro-5-methylthiophenyl)-N0-(3-methylthiophenyl)-
guanidine (7). To the desmethyl precursor 7 (1.4 mg, 4.14 mmol) added
sodium hydride (4.0 mg, 0.17 mmol) and acetonitrile (0.6 ml). After stirring for
several minutes, methyl iodide (1 ml, 16.06 mmol) was added and the mixture
stirred for a further 10 min. The reaction mixture was diluted with
(NH4)2HPO4 (aq) (10 mM, 0.5 ml) and subjected to semi-preparative HPLC
using a C18 mBondapak column (7.8 ꢁ 300 mm, Waters) with a (NH4)2HPO4
(10 mM)/acetonitrile 50/50 (v/v) solvent system (3 ml/min, l ¼ 254 nm).
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 163–170

SYNTHESIS AND CHARACTERIZATION OF [11C]CNS 5161
167
Samples of compounds 1–3 were concentrated under a flow of nitrogen and
extracted into ethanol for analysis.
Radiolabelling
N-(2-chloro-5-methylthiophenyl)-N0-(3-methylthiophenyl)-N0-[11C]methylguanidine
ð3Þ, [11C]CNS 5161. [11C]Methyl iodide was prepared using the standard ‘wet’
method as described by Turton et al.19 The radio-synthesis was carried out
using a remotely controlled apparatus housed within a lead shielded enclosure.
To the desmethyl precursor 7 (0.5 mg) was added sodium hydride (2 mg) and
acetonitrile (anhydrous, 300 ml) under an inert atmosphere (N2). After stirring
for 1 min, [11C]methyl iodide was distilled into the reaction vial and the
resulting mixture was heated to 708C for 3 min. The crude product was diluted
with water (5 ml) and the resulting suspension was loaded on a reverse phase
HPLC column (4.5 ꢁ 30 mm). The column was washed with water (8 ml)
before the product mixture was eluted onto a semi-preparative C18
mBondapak column (7.8 ꢁ 300 mm, Waters) and purified using a solvent
system of di-ammonium hydrogen orthophosphate (0.05 M)/acetonitrile 4/6
(v/v) (3 ml/min). The fraction containing [11C]CNS 5161 (retention time
ꢂ17 min) was collected into a rotary evaporator flask containing a solution of
ascorbic acid (1 mg) in water (1 ml). The product was evaporated to dryness
under reduced pressure, the residue was dissolved in isotonic saline containing
5% ethanol (10 ml) and the solution sterilized by filtration (25 mm, 0.2 mm
pore Acrodisc filter, Pall corp.) in a class A isolator. Analytical HPLC for
product quality control was performed using a Prodigy C18 analytical column
(4.6 ꢁ 150 mm) from Phenomenex and a solvent system of ammonium
orthophosphate (0.05 M)/acetonitrile 45/55 (v/v) (1 ml/min). The identity
and purity of the product was monitored with UV (l ¼ 257 nm) and a Bioscan
GM tube detector. In this system [11C]CNS 5161 (3) eluted after 4.5 min and
the desmethyl precursor 7 eluted after 6 min.
Results and discussion
CNS 5161 can potentially be labelled with [11C]methyl iodide in three
positions; either of the two S-methyl groups or the N0-methyl guanidine
functionality. We rationalized that labelling of the desmethyl guanidine
functionality provided the simpler route for precursor synthesis and
therefore was preferred. The main concern with this strategy was formation
of the N-[11C]methyl guanidine regioisomer 9, and whether this compound
could be separated from [11C]CNS 5161 (3) in a time-scale suitable for high
activity routine production.
The desmethyl guanidine precursor 7, CNS 5161 (3) and its N-methyl
regioisomer 9 were synthesized using modified literature procedures as
depicted in Scheme 1.
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 163–170

168
Y. ZHAO ET AL.
R'
Cl R'' R'
MeS
NH2
MeS
ii
N
N
N
i
iv
Cl
CN
NH
NR''H2Cl
10
SMe
7: R' = R'' = H
3: R' = CH3, R'' = H
9: R' = H, R'' = CH3
SMe
4: R = H
5: R = CH3
SMe
6: R = H
8: R =CH3
iii
Scheme 1. Synthesis of precursor 7, the N-methylated standard 9 and its N0-
methylated regioisomer 3. Reagents and conditions: (i) BrCN in ether, (ii) NaH
and MeI, (iii) Butyl lithium and MeI and (iv) PhMe, reflux
Commercially available 3-methylthioaniline (10) was reacted with
cyanogen bromide (Caution: highly toxic) as described in the literature to
give the corresponding cyanamide 4.18 For the preparation of CNS 5161
(3), this intermediate was methylated under basic conditions using methyl
iodide. Subsequent condensation of either 4 or 5 with the chloroaniline 616
provided the desmethyl precursor 7 and CNS 5161 (3), respectively.
Initially, we attempted to synthesize the N-methylguanidine regioisomer
9 via the same route, i.e. transformation of the chloroaniline 6 to the
corresponding cyanamide, followed by methylation and subsequent condensa-
tion with the aniline 10. However, the chloroaniline 6 resisted all attempts
to convert it into the corresponding cyanamide and instead the N-
methylguanidine regioisomer 9 was prepared by deprotonation of the
chloroaniline 6 with butyl lithium in THF, subsequent treatment with methyl
iodide to provide the methylated chloroaniline 7, and finally condensation
with the cyanamide 4.
When the desmethyl guanidine precursor 7 was reacted with [11C]methyl
iodide in acetonitrile in the presence of sodium hydride, HPLC analysis
revealed formation of two major radioactive products which co-eluted with
CNS 5161 (3) and the N-methyl regioisomer (9), respectively. When the
reaction was repeated using non-radioactive methyl iodide an additional
product was formed. To verify the nature of the products, samples were
purified by preparative HPLC and subjected to electrospray mass-spectro-
metry (ES-MS) and ES-MS–MS fragmentation analysis. The main product
was found to have an identical mass (M þ Hþ ¼ 352 amu) and fragmentation
pattern to CNS 5161 (3), with C–N0 cleavage giving rise to two characteristic
ions of m=z ¼ 199 and 154 amu (Figure 2). The molecular mass
(M þ Hþ ¼ 352 amu) and fragmentation pattern of the second product were
consistent with the N-methyl isomer 9, with C–N0 cleavage resulting in an ion
of m=z ¼ 231 amu, and C–N cleavage ions of m/z=188 and 165 amu. The
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 163–170

SYNTHESIS AND CHARACTERIZATION OF [11C]CNS 5161
169
Cl
Cl
a
H
H
N
N
S
N
N
S
a
b
NH
NH
S
S
S
CNS 5161 (3)
9
m/z
154
m/z
188 (a)
Cl
S
NH2+
+H2N
C8H12NS
(a)
(a)
C8H11ClNS
Cl
S
H
N
H
C+
NH
C8H8ClN2S1 199
N
S
+C
NH
C8H9N2S1
165 (a)
Cl
S
N
C+
NH
C9H10ClN2S2 213 (b)
Figure 2. Electrospray MS–MS fragmentation pattern of compounds 3 and 9
product unique to the cold reaction was observed to have a molecular mass
(M þ Hþ ¼ 366 amu) consistent with that of a di-methylated species, and we
postulate that this is the N,N0-dimethylguanidine derivative. Overall, the data
confirm that the major product from [11C]methylation of the precursor 7 is
[11C]CNS 5161 (3) and that separation from the N-[11C]methyl regioisomer 9
can be achieved by HPLC.
Encouraged by the results, we developed an automated procedure for high
activity routine production of [11C]CNS 5161, which is given in the
experimental part. [11C]CNS 5161 is now routinely obtained in 10% decay
corrected yield within 45 min after EOB, with a radiochemical purity >96%
and a specific activity of 41 GBq/mmol at time of injection. The biological
evaluation of [11C]CNS 5161 will be published elsewhere.
Conclusions
[11C]CNS 5161 is a candidate PET tracer for functional imaging of the NMDA
receptor. Herein we report that N-[11C]methylation of the N-desmethyl
guanidine precursor is a viable route to [11C]CNS 5161, and describe an
automated synthesis based on this chemistry. The radiosynthesis of [11C]CNS
5161 is highly reproducible and allows for the tracer to be produced in high
specific activity, good radiochemical purity and sufficient quantities for in vivo
evaluations.
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 163–170

170
Y. ZHAO ET AL.
References
1. Dingledine R, Borges K, Bowie D, Traynelis SF. Pharmacol Rev 1999; 51: 7–61.
2. Riedel G, Platt B, Micheau J. Behav Brain Res 2003; 140: 1–47.
3. Siegel GJ, Agranoff BW, Albers RW, Fisher SK, Uhler MD. Basic Neurochem-
istry: Molecular, Cellular and Medical Aspects (6th edn). Lippincott Williams &
Wilkins: Philadelphia, 1999.
4. Kew JNC, Kemp JA. Psychopharmacology 2005; 179: 4–29.
5. Waxman EA, Lynch DR. Neuroscientist 2005; 11: 37–49.
6. Loftis JM, Janowsky A. Pharmacology & Therapeutics 2003; 97: 55–85.
7. Konradi C, Heckers S. Pharmacology & Therapeutics 2003; 97: 153–179.
8. Parson CG, Danysz W, Quack G. Drug News Perspect 1998; 11: 523–567.
9. Heresco-Levy U. Int J Neuropsychopharmacol 2000; 3: 243–258.
10. Bressan RA, Pilowsky LS. Eur J Nucl Med 2000; 27: 1723–1731.
11. Waterhouse RN. Nucl Med Biol 2003; 30: 869–878.
12. Owens J, Tebbutt AA, McGregor AL, Kodama K, Magar SS, Perlman ME,
Robins DJ, Durant GJ, McCulloch J. Nucl Med Biol 2000; 27: 557–564.
13. Bressan RA, Erlandsson K, Mulligan RS, Gunn RN, Cunningham VJ, Owens J,
Ell PJ, Pilowksy LS. Ann N Y Acad Sci 2003; 1003: 364–367.
14. Bressan RA, Erlandsson K, Mulligan RS, Gunn RN, Cunningham VJ, Owens J,
Cullum ID, Ell PJ, Pilowsky LS. Nucl Med Biol 2004; 31: 155–164.
15. Waterhouse RN, Dumont F, Sultana A, Simpson N, Laruelle M. J Label Compd
Radiopharm 2002; 45: 955–964.
16. Hu LY, Guo J, Magar SS, Fischer JB, Burke-Howie KJ, Durant GJ. J Med Chem
1997; 40: 4281–4289.
17. Walters MR, Bradford AP, Fischer J, Lees KR. Br J Clin Pharmacol 2002;
53: 305–311.
18. Dumont F, Sultana A, Waterhouse R. Biorg Med Chem Lett 2002; 12: 1583–1586.
19. Turton DR, Brady F, Pike VW, Selwyn AP, Shea MJ, Wilson RA, Landsheere
CM. Int J Appl Radiat Isot 1984; 35: 337–344.
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 163–170