Parallel synthesis of a series of subtype-selective NMDA receptor antagonists

Article abstract of DOI:10.1016/S0960-894X(00)00035-4

A series of 1-(heteroarylthioalkyl)-4-benzylpiperidines was rapidly synthesized through the use of parallel synthesis to investigate the binding affinity for the NR1A/2B receptor subtype. (C) 2000 Published by Elsevier Science Ltd. All rights reserved.

Full text of DOI:10.1016/S0960-894X(00)00035-4

Bioorganic & Medicinal Chemistry Letters 10 (2000) 527±529
Parallel Synthesis of a Series of Subtype-Selective NMDA
Receptor Antagonists
Tracy F. Gregory, a,* Jon L. Wright, a Lawrence D. Wise, a Leonard T. Meltzer, b
Kevin A. Serpa, b Christopher S. Konkoy, c Edward R. Whittemore c
and Richard M. Woodward c
aDepartment of Chemistry, Parke-Davis Pharmaceutical Research, Division of Warner-Lambert Company, Ann Arbor, MI 48105, USA
bDepartment ofTherapeutics, Parke-DavisPharmaceuticalResearch, Division ofWarner-Lambert Company, AnnArbor, MI48105, USA
cCoCensys, Inc., 213 Technology Drive, Irvine, CA 92618, USA
Received 13 October 1999; accepted 11 January 2000
AbstractÐA series of 1-(heteroarylthioalkyl)-4-benzylpiperidines was rapidly synthesized through the use of parallel synthesis to
investigate the binding anity for the NR1a/2B receptor subtype. # 2000 Published by Elsevier Science Ltd. All rights reserved.
Parkinson's disease a€ects 1% of the population over 50
years of age.1 It is a progressive degenerative central
nervous system disorder, which has many debilitating
e€ects such as muscle rigidity, resting tremors and
slowness or poverty of movement. Parkinson's disease
results from degeneration of dopaminergic neurons that
lie within the substantia nigra. The treatment of choice
has been dopamine replacement therapy, using l-di-
hydroxyphenylalanine (l-DOPA). However, long term
treatment often results in adverse side e€ects such as
dyskinesias. It has been shown that N-methyl-d-aspar-
tate (NMDA) antagonists can potentiate the e€ects of
l-DOPA in animal models of Parkinson's disease. The
use of non-selective NMDA antagonists also results in
several side e€ects, the most common being ataxia,
sedation and cognitive impairments. The recent dis-
covery of multiple subtypes of NMDA receptors, which
are di€erentially expressed throughout the brain, may
allow separation of therapeutic activity from adverse
side e€ects.
and NR1a/2C IC50s >100 mM). Compound 1 was also
shown to potentiate the e€ects of l-DOPA when
administered interperitoneally but not orally in the
6-hydroxydopamine-lesioned (6-OHDA) rat, a model of
Parkinson's disease.3 A direct analogue (2) was pre-
pared and it maintained good binding activity and
selectivity (NR1a/2A IC50=35.0 mM, NR1a/2B
IC50=0.17 mM, NR1a/2C IC50 >1 0 0 mM). Due to the
known high metabolism of the phenol moiety,4 our
objective was to identify phenol replacements that might
show oral activity in vivo. The hydrogen bonding char-
acter of the phenol moiety was needed to maintain good
binding activity: therefore, we focused our synthesis on
heterocyclic replacements which contain similar H-
bonding character.
Utilizing parallel synthesis, we were able to rapidly
synthesize analogues of 2 to explore the binding anity
for the NMDA NR1a/2B receptor subtype. The general
synthesis is shown in Scheme 1.5
We recently reported a novel series of acetylene-linked
NMDA NR1a/2B subtype-selective antagonists,2 in
which compound 1 showed sub-micromolar potency at
the NR1a/2B receptor subtype (IC50=0.1 mM) expres-
sed in Xenopus oocytes and good selectivity (NR1a/2A
A standard SN2 reaction between the appropriate bro-
mide and the 4-benzyl piperidine resulted in the alcohol.
This alcohol was reacted with thionyl chloride to give
the chlorides in 60ꢀ80% yield. Standard solutions (0.5 M)
of the chlorides in acetonitrile were prepared and sub-
sequently transferred (4 mL) to ten 28Â95 mm vials.
Solutions (0.5 M) of each thiol were prepared and 4 mL
transferred to the appropriate vial. Potassium carbonate
(303 mg) was added separately to each reaction vial.
*Corresponding author. Tel.: +1-734-622-4696; fax: +1-734-622-
5165; e-mail: tracy.gregory@wl.com
0960-894X/00/$ - see front matter # 2000 Published by Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00035-4

528
T. F. Gregory et al. / Bioorg. Med. Chem. Lett. 10 (2000) 527±529
The vials were ¯ushed with nitrogen, sealed and warmed
with mixing to 100 ꢁC on a J-KemTM heater/shaker.
After 18 h at 100 ꢁC the reactions were cooled, ®ltered
and concentrated. The crude mixtures were puri®ed in
parallel by MPLC on silica gel (10 g) resulting in the
desired products in 50ꢀ95% yield.
Table 1. NR1a/2B receptor potency for analogues of 2
Compound
HetAr
NR1a/2A2
IC50 (mM)
NR1a/2B2
IC50 (mM)
NR1a/2C2
IC50 (mM)
To initiate our SAR study, we explored the e€ects on
NR1a/2B antagonist potency by replacing the phenol
of 2 with heteroaryl groups. We identi®ed several
compounds that were potent at NR1a/2B receptors
(Table 1). The best potency correlated with the presence
of a H-bond donor, as in compound 4. Lack of a
hydrogen bond donor, as in compound 9, resulted in a
decrease in the binding potency. Larger heteroaryl
groups that contain a H-bond donor (such as in 11) also
resulted in a reduction of receptor binding.
2
35.0 (1)
0.10‹0.02 (4) >100 (1)
3
4
>100 (1)
0.93‹0.3 (3)
>100 (1)
>100 (1)
>100 (1) 0.035‹0.01
5
6
7
>100 (1)
>100 (1)
70.0 (1)
0.07‹0.03 (3) >100 (1)
0.13‹0.01 (3) >100 (1)
0.15‹0.02 (3) >100 (1)
In Table 2 the NR1a/2B activity was examined for the
higher chain homologues. This change was not well tol-
erated at NR1A/2B receptors, resulting in a reduction
of potency by a factor of 4- to 8-fold (e.g. compound 4
versus 14). However, exceptions in the reductions were
seen, such as compound 20, where an increase in
binding potency was observed. This binding increase
may be the result of additional rotational freedom in the
molecule.
8
>100 (1)
>100 (1)
>100 (1)
0.21‹0.02 (3) >100 (1)
9
4.8‹0.7 (3)
7.4‹1.0 (3)
>100 (1)
>100 (1)
Compound 4 is a potent NR1a/2B antagonist and was
subjected to further testing in the 6-hydroxydopamine-
lesioned rat. Like many other compounds in this series,
4 showed no activity when administered at 10 mg/kg PO
in the 6-OHDA rat model. The lack of oral activity may
be due to ®rst pass metabolism of the sulfur linkage or
the inability of the compounds to cross the blood brain
barrier.
10
11
12
>100 (1)
75 (1)
0.24‹0.03 (3) >100 (1)
0.38‹0.03 (3) >100 (1)
In conclusion, through the use of parallel synthesis we
rapidly investigated the thio ether linked heteraryl
Scheme 1.

T. F. Gregory et al. / Bioorg. Med. Chem. Lett. 10 (2000) 527±529
Table 2. NR1a/2B activity for analogues of 2
529
groups and identi®ed several potent and selective
NMDA NR1a/2B receptor antagonists. As with the
previous acetylene-linked series, a hydrogen bond donor
was required for potent NR1a/2B activity.2 An ethylene
link (n=2) between the piperidine and the thio aryl
system was preferred. While we found replacements for
the phenol moiety of 2 that maintained in vitro activity,
oral activity in vivo was not achieved.
Compound
n
HetAr
NR1a/2A2
IC50 (mM)
NR1a/2B2
IC50 (mM)
NR1a/2C2
IC50 (mM)
3
2
3
3
>100 (1) 0.93‹0.3 (3)
>100 (1)
>100 (1)
>100 (1)
13
14
85.0 (1)
4.3‹1.0 (3)
References and Notes
77.0 (1) 0.63‹0.1 (3)
1. Cooper, A. J.; Carroll, C. B.; Mitchell, I. J. CNS Drugs
1998, 9, 421.
2. All compounds were tested for inhibitory activity at NR1a/
2A, NR1a/2B and NR1a/2C receptors in frog oocytes as pre-
viously described. Values are mean SEM taken from 1 to 4
experiments, as noted in parentheses: Wright, J. L.; Gregory,
T. F.; Bigge, C. F.; Boxer, P. A.; Serpa, K.; Meltzer, L. T.;
Wise, L. D.; Cai, S. X.; Hawkinson, J.; Whittemore, E.;
Woodward, R.; Zhou, Z. J. Med. Chem. 1999, 42, 2469.
3. 6-OHDA rat model: rats were lesioned by injection of 6-
hydroxydopamine into the right medial forebrain bundle
rostral to the substantia nigra. After lesioning, rats are condi-
tioned with l-DOPA methyl ester (10 mg/kg SC, mixed with
s-carbidopa 1 mg/kg to prevent peripheral decarboxylation) to
establish a stable response. Vehicle or the test compound was
administered orally at the same time as l-DOPA in a rando-
mized, crossover manner, with 1 week between treatments.
The total number of contraversive rotations observed follow-
ing dosing with the test compound+l-DOPA was compared
with the response in the same rats to vehicle+l-DOPA, using
a paired t-test.
15
16
3
3
77.0 (1) 0.61‹0.04 (3) >100 (1)
51.0 (1) 0.63‹0.04 (3) >100 (1)
17
18
3
3
>100 (1)
1.0‹0.2 (4)
>100 (1)
73.0 (1) 0.56‹0.03 (3) >100 (1)
19
20
3
3
>100 (1)
6.4‹1.0 (3)
>100 (1)
>100 (1)
61.0 (1) 0.91‹0.2 (4)
21
22
3
3
>100 (1)
64.0 (1)
1.6‹0.2 (3)
1.0‹0.2 (4)
>100 (1)
51.0 (1)
4. Mistry, M.; Houston, J. B. Drug Metab. Dispos. 1985, 13,
740.
5. All new compounds had satisfactory 1H NMR, IR, MS and
microanalysis.