Novel N1-(benzyl)cinnamamidine derived NR2B subtype-selective NMDA receptor antagonists.

Article abstract of DOI:10.1016/S0960-894X(02)01060-0

Novel (E)-N(1)-(benzyl)cinnamamidines were prepared and evaluated as NR2B subtype NMDA receptor ligands. Excellent affinity was achieved by appropriate substitution of either phenyl ring. The 2-methoxybenzyl compound 1h had approximately 1,000-fold lower IC(50) in NR2B than NR2A-containing cells. Replacement of the styryl unit by 2-naphthyl was well tolerated.

Full text of DOI:10.1016/S0960-894X(02)01060-0

Bioorganic & Medicinal Chemistry Letters 13 (2003) 693–696
Novel N1-(Benzyl)cinnamamidine Derived NR2B
Subtype-Selective NMDA Receptor Antagonists
Neil R. Curtis,a,* Helen J. Diggle,a Janusz J. Kulagowski,a Clare London,a
Sarah Grimwood,b Peter H. Hutson,b Fraser Murray,b Pawel Richards,b
Alison Macaulayc and Keith A. Waffordc
aDepartment of Medicinal Chemistry, Merck Sharp & Dohme Research Laboratories, The Neuroscience Research Centre,
Terlings Park, Eastwick Road, Harlow, Essex CM20 2QR, UK
bDepartment of Behavioural Neuroscience, Merck Sharp & Dohme Research Laboratories, The Neuroscience Research Centre,
Terlings Park, Eastwick Road, Harlow, Essex CM20 2QR, UK
cDepartment of Pharmacology, Merck Sharp & Dohme Research Laboratories, The Neuroscience Research Centre,
Terlings Park, Eastwick Road, Harlow, Essex CM20 2QR, UK
Received 2 October 2002; accepted 8 November 2002
Abstract—Novel (E)-N1-(benzyl)cinnamamidines were prepared and evaluated as NR2B subtype NMDA receptor ligands. Excel-
lent affinity was achieved by appropriate substitution of either phenyl ring. The 2-methoxybenzyl compound 1h had ꢀ1000-fold
lower IC50 in NR2B than NR2A-containing cells. Replacement of the styryl unit by 2-naphthyl was well tolerated.
# 2003 Elsevier Science Ltd. All rights reserved.
Glutamate is the major excitatory neurotransmitter
in the mammalian CNS. The N-methyl-d-aspartate
(NMDA) receptor is one of the three families of iono-
tropic glutamate receptor and is a major therapeutic
target for a wide range of clinical indications such as
stroke, epilepsy, neuropathic pain and Parkinson’s dis-
ease.1 The NMDA receptor is a hetero-oligomeric
cation channel complex comprising one or more NR1
subunits, of which there are eight isoforms (NR1a–h),
and at least one of a family of NR2 subunits (NR2A–
D), with different subunit combinations conferring
diverse functional properties.1 NR1 and NR2A are
abundantly expressed throughout rat brain whereas
localisation of NR2B subunits is more restricted, sug-
gesting that agents acting selectively at this subtype may
exhibit a reduced side-effect profile.2,3 NR2B-selective
NMDA antagonists have been developed,4,5 such as the
prototypic compound ifenprodil, Ro25-6981, CP-
101,606 and CI-1041 (Fig. 1), and shown to be effective
in preclinical pain models such as carrageenan-induced
hyperalgesia and mechanical allodynia in nerve ligated
rats.3,6
Screening for novel compounds with significant affinity
for NR2B-containing receptors identified (E)-N1-(ben-
zyl)cinnamamidine (1a) as a promising candidate for
initiation of a medicinal chemistry programme, particu-
larly since it lacked the phenolic group and 4-benzyl-
piperidine moiety common to many NR2B antagonists.
Amidines 1–4 were readily prepared by treatment of
(E)-ethyl cinnamimidate hydrochloride with the appro-
priate amine and crystallizing the product hydrochlo-
ride (Scheme 1) or via Horner–Wadsworth–Emmons
reaction of an amidine phosphonate intermediate
(Scheme 2).
Benzamidines and 2-naphthamidines 6–8 were available
from the corresponding nitrile via the imidate or
directly using either a copper-mediated method7 or a
methylchloroaluminium amide derived from amine
hydrochloride and trimethylaluminium8 (Scheme 3).
The primary assay used to evaluate compounds was
displacement of NMDA receptor-specific [3H]ifenprodil
*Corresponding author. Fax: +44-1279-440390; e-mail: neil_curtis@
merck.com
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(02)01060-0

694
N. R. Curtis et al. / Bioorg. Med. Chem. Lett. 13 (2003) 693–696
Figure 1. Structures of NR2B-selective NMDA receptor antagonists.
Scheme 3. Reagents and conditions: (i) HCl, EtOH; (ii) Ar0CH2NH2,
MeOH; (iii) Ar0CH2NH2, CuCl, EtOH, reflux; (iv) Ar0CH2N-
HAlClCH3, toluene, 80 ꢂC.
Scheme 1. Reagents and conditions: (i) HCl, EtOH, Et2O; (ii)
RCH2NH2, MeOH.
Table 1. [3H]Ifenprodil binding affinities for cinnamamidines 1a–s
Compd
R1
R2
[3H]Ifenprodil Ki (nM)a
1a
1b
1c
1d
1e
1f
1g
1h
1i
1j
1k
1l
1m
1n
1o
1p
1q
1r
1s
H
H
H
H
H
H
H
H
H
H
H
H
H
H
2-Cl
3-Cl
4-Cl
9.0
1.0
0.8
77
1.4
31
1400
0.7
1.9
0.5
0.6
3.6
2.7
1.4
93
2-CF3
3-CF3
4-CF3
2-OCH3
3-OCH3
2-OCH2CH3
2-OCF3
2,3-di-Cl
2,5-di-Cl
3,5-di-Cl
H
Scheme 2. Reagents and conditions: (i) 2-bromomethylnaphthalene,
CHCl3, reflux; (ii) PhCH2NH2, EtOH; (iii) RCHO, K2CO3, THF,
reflux; (iv) RCHO, CH2Cl2, aq NaOH, n-Bu4NBr, rt.
binding to recombinant human NR1a/NR2B receptors
stably expressed in L(tk-) cells.9 Thus, in the [3H]ifen-
prodil binding assay amidine 1a had Ki 9 nM, which
was comparable to that determined in-house9 for (ꢁ)-
Ro25-6981 (Ki 10 nM) and (ꢁ)-CP-101,606 (Ki 11 nM).
In patch-clamp electrophysiology studies 1a demon-
strated selectivity for inhibition of the response to glu-
tamate/glycine in NR1a/NR2B expressing cells (IC50
70ꢁ6 nM, n=4) over that observed in NR1a/NR2A
cells (no effect up to 3 mM, n=4).10 In the discussion,
NR1a/NR2B has been abbreviated to NR2B for the
sake of brevity.
H
2-Cl
3-Cl
4-Cl
4-F
3,4-di-F
H
H
H
H
12
52
5.7
1.7
aDisplacement of [3H]ifenprodil binding to recombinant human
NR1a/NR2B receptors stably expressed in L(tk-) cells (ref 9). Values
represent the geometric mean of 3 to 5 determinations.
suggesting that a hydrophobic interaction was respon-
sible for the improved affinity over 1a rather than elec-
tronic factors. Dichloro substitution (1l–n) offered
no particular advantage over the 2- or 3- monochloro
analogues.
Initial investigation of the lead amidine 1a focussed on
substitution of either phenyl ring (Table 1). Substitution
of the benzyl moiety (R2 substituent) showed that 4-
substitution was poorly tolerated; for example 4-chloro
(1d) gave an order of magnitude reduction in NR2B
affinity and 4-trifluoromethyl (1g) a 150-fold reduction.
A 3-chloro group (1c) gave a 10-fold improvement, yet
the larger 3-trifluoromethyl (1f) substituent was less well
tolerated. A variety of 2-substituents showed improved
affinity over the parent to provide subnanomolar com-
pounds, for example, 2-chloro (1b) or 2-alkoxy (1h, 1j),
Preliminary investigation of substitution on the styryl
phenyl ring of 1a (R1 substituent) suggested less scope
for improving NR2B affinity than with changes made to
the benzyl moiety (Table 1). 3-Chloro was tolerated
(1p), but 4- and 2-chloro had 5- and 10-fold lower affinity
than 1a, respectively. 4-Fluoro (1r) was incorporated to

N. R. Curtis et al. / Bioorg. Med. Chem. Lett. 13 (2003) 693–696
Table 2. Variation of chain length and replacement of the styryl
group
695
Compd
R1
n
R2
[3H]Ifenprodil
Ki (nM)a
1a
2
1
2
3
1
1
1
2
1
1
1
1
H
H
9.0
97
Figure 2. Representative substituted N1-(benzyl)cinnamamidine and
N1-(benzyl)naphthamidine compounds displaying low nanomolar
NR2B receptor affinity.
amidine functional group by amine, amide12 or imidazole
significantly attenuated activity (data not shown).
Extension of the benzyl moiety of 1a by one or two
carbon atoms to give the corresponding phenethyl and
phenylpropyl compounds, 2 and 3, respectively, reduced
NR2B affinity 10-fold (Table 2). Replacement of the
phenyl ring of the styryl unit by cyclohexyl (4) retained
equivalent activity.
3
H
120
19
4
H
5
H
530
10,000
55
In order to obviate the potential Michael-acceptor
reactivity of 1a, the double bond was hydrogenated to
afford 5. Unfortunately, this was accompanied by a 50-
fold drop in binding affinity. Instead, the styryl group
was truncated to phenyl by deleting the double bond.
However, N-(benzyl)benzamidine (6) was essentially
devoid of NR2B affinity. Interestingly, significant activ-
ity could be restored by homologation of the benzyl
group to give phenethyl compound 7, suggesting some
tolerance for the position of the amidine functionality
along the chain linking the two aromatic groups.
6
H
7
H
8a
8b
8c
8d
H
32
2-OCH3
2-Cl
3-Cl
1.3
Replacement of the styryl group of 1a with 2-naphthyl
(8a) gave a modest reduction in affinity. However, the
activity could be improved by benzyl substitution found
to be beneficial in the cinnamamidine series. Hence,
2-methoxy and 3-chlorobenzyl naphthamidines, 8b and
8d respectively, had low nanomolar activity without the
potential Michael-acceptor liability.
7.3
1.6
aSee corresponding footnote in Table 1.
In conclusion, investigation of the lead N1-(benzyl)cinna-
mamidine 1a (Ki 9 nM) demonstrated that NR2B affinity
could be improved by aromatic substitution of either
phenyl ring, as exemplified by the 2-methoxybenzyl com-
pound 1h (Ki 0.7 nM) and 3,4-difluorocinnamamidine 1s
(Ki 1.7 nM; Fig. 2). The styryl group could be replaced
by 2-naphthyl as in 8b (Ki 1.3 nM) and 8d (Ki 1.6 nM).
Selectivity over NR2A was good, where determined; three
orders of magnitude in the case of 1h. Tritiation of 1h has
provided an NR2B-selective radioligand and further ela-
boration of the compounds identified in this study has led
to the discovery of orally efficacious NR2B-selective
NMDA receptor antagonists.11
block a potential site of metabolism, and was found to
be well tolerated. Indeed, 3,4-difluoro subsitution (1s)
gave 5-fold higher affinity than the unsubstituted parent
structure 1a.
NMDA subtype selectivity for this series of cinnam-
amidines was further illustrated by patch-clamp elec-
trophysiology. Thus, 1h had IC50 28.6ꢁ2.6 nM (n=5)
for inhibition of the response to glutamate/glycine in
NR1a/NR2B expressing cells, compared with approxi-
mately 26 mM (n=6) in NR1a/NR2A cells.10
Having identified several high affinity substituted N1-
(benzyl)cinnamamidine NR2B antagonists of potential
use as biological tools, notably 1h,11 we turned our
attention to variation of the amidine group, the overall
chain length and the styryl unit. Attempts to replace the
Acknowledgements
The authors are indebted to Alan Watt and Steve
Thomas for analytical support.

696
N. R. Curtis et al. / Bioorg. Med. Chem. Lett. 13 (2003) 693–696
8. Garigipati, R. S. Tetrahedron Lett. 1990, 31, 1969.
References and Notes
9. Grimwood, S.; Richards, P.; Murray, F.; Harrison, N.;
Wingrove, P. B.; Hutson, P. H. J. Neurochem. 2000, 75, 2455.
10. Priestley, T.; Laughton, P.; Myers, J.; Le Bourdelles, B.;
Kerby, J.; Whiting, P. J. Mol. Pharmacol. 1995, 48, 841.
11. Claiborne, C. F.; McCauley, J. A.; Libby, B. E.; Curtis, N.
R.; Diggle, H. J.; Kulagowski, J. J; Michelson, S. R.; Ander-
son, K. D.; Claremon, D. A.; Freidinger, R. M.; Bednar, R.
A.; Mosser, S. D.; Gaul, S. L.; Connolly, T. M.; Condra, C.
L.; Bednar, B.; Stump, G. L.; Lynch, J. J.; Macaulay, A.;
Wafford, K. A.; Koblan, K. S.; Liverton, N. J. Bioorg. Med.
Chem. Lett. see following communication. doi:10.1016/S0960-
894X(02)01061-2.
1. Dingledine, R.; Borges, K.; Bowie, D.; Traynelis, S. F.
Pharmacol. Rev. 1999, 51, 7.
2. Monyer, H.; Burnashev, N.; Laurie, D. J.; Sakmann, B.;
Seeburg, P. H. Neuron 1994, 12, 529.
3. Boyce, S.; Wyatt, A.; Webb, J. K.; O’Donnell, R.; Mason,
G.; Rigby, M.; Sirinathsinghji, D.; Hill, R. G.; Rupniak,
N. M. J. Neuropharmacology 1999, 38, 611.
4. Chenard, B. L.; Menniti, F. S. Curr. Pharm. Des. 1999, 5,
381.
5. Nikam, S. S.; Meltzer, L. T. Curr. Pharm. Des. 2002, 8,
845.
12. The amide analogue of 1a (C¼NH replaced by C¼O) had
NR2B Ki 3.2 mM. N-(Phenylalkyl)cinnamide NR2B subtype-
selective antagonists have been reported; J. Med. Chem. 1999,
42, 3412.
6. Chizh, B. A.; Headley, P. M.; Tzschentke, T. M. Trends
Pharmacol. Sci. 2001, 22, 636 and references cited therein.
7. Rousselet, G.; Capdavielle, P.; Maumy, M. Tetrahedron
Lett. 1993, 34, 6395.