4-Alkylidenyl glutamic acids, potent and selective GluR5 agonists

Article abstract of DOI:10.1016/S0960-894X(00)00346-2

Twenty-four 4-alkylidene glutamic acids were synthesised and tested as potential subtype selective GluR5 and 6 ligands. It was found that a critical size of alkylidene group gave potent and selective GluR5 receptor agonists. LY339624 had K(i)s of 0.0326 and > 100 μM on GluR5 and 6 receptors, respectively. (C) 2000 Published by Elsevier Science Ltd. All rights reserved.

Full text of DOI:10.1016/S0960-894X(00)00346-2

Bioorganic & Medicinal Chemistry Letters 10 (2000) 1807±1810
4-Alkylidenyl Glutamic Acids, Potent and Selective GluR5 Agonists
S. Richard Baker,a,* David Bleakman,b Jesus Ezquerra,b Barbara A. Ballyk,c
Michele Deverill,c Ken Ho,c Rajender K. Kamboj,c Ivan Collado,d
Carmen Domõnguez,d Ana Escribano,d Ana I. Mateo,d
Concepcion Pedregald and Almudena Rubiod
aEli Lilly and Company Ltd, Lilly Research Centre, Erl Wood Manor, Windlesham, Surrey GU20 6PH, UK
bLilly Research Laboratories, Indianapolis, IN 46285, USA
cAllelix Biopharmaceuticals Inc., 6850 Goreway Drive, Mississauga, ON Canada L4V 1V7
dLilly, S.A. Avda. de la Industri, 30, 28108 Alcobendas, Spain
Received 27 March 2000; revised 13 June 2000; accepted 14 June 2000
AbstractÐTwenty-four 4-alkylidene glutamic acids were synthesised and tested as potential subtype selective GluR5 and 6 ligands. It
was found that a critical size of alkylidene group gave potent and selective GluR5 receptor agonists. LY339624 had Kis of 0.0326 and
>100 mM on GluR5 and 6 receptors, respectively. # 2000 Published by Elsevier Science Ltd. All rights reserved.
l-Glutamate (Glu) is the major excitatory neuro-
transmitter in the mammalian central nervous system
(CNS). Glutamate receptors are subdivided into iono-
tropic (GluRs)1,2 and metabotropic receptors (mGluRs).3,4
The ionotropic receptors mediate fast synaptic trans-
mission through ligand-gated ion channels. These iono-
tropic receptors are subdivided into three classes on the
basis of their sensitivity to NMDA (N-methyl-d-aspar-
tate), KA (kainate) and AMPA (a-amino-3-hydroxy-5-
methyl-4-isoxazolepropionate), from which they take
their names.5 Studies have shown there are ®ve distinct
kainate receptor subunits: GluR5, GluR6, GluR7, KA1,
and KA2 (for a review see ref 6). GluR5, 6 and 7, when
expressed homomerically, form functional ion channels
activated by kainate, domoate or glutamate.7,8 Since
these individual kainate binding proteins have di€erent
distributions within the CNS,9,10 subtype selective kai-
nate ligands may be expected to show distinct pharma-
cology and may thus have therapeutic potential.
Unfortunately, there are very few kainate subtype
selective ligands, hence the therapeutic potential of such
agents has not been realised. In an earlier paper, we
reported that the 4-substituted glutamate analogue
LY339434 is a potent and selective GluR5 receptor
agonist (Ki 14.8 nM).11 This paper extends the struc-
ture±activity relationship (SAR) to a number of alkyli-
dene glutamate analogues (Fig. 1), some of which are
potent and selective GluR5 receptor agonists. This
study also gives insight into the di€erences in the recep-
tor binding sites of GluR5 and 6.
Chemistry
The compounds were prepared as shown in Scheme 1 by
.
BF3 OEt2 mediated aldol condensation of S-ethyl pyro-
glutamate 1 with aldehydes and ketones.12 The resultant
diastereomeric mixtures of the alcohols 2 were dehy-
drated13 to give predominately the E isomers of the
4-alkylidenepyroglutamates 3 which were puri®ed by
¯ash column chromatography. Hydrolysis of 3 was
accomplished with aqueous LiOH in THF to produce
the N-BOC-protected diacid followed by removal of the
BOC group with a saturated solution of HCl in ethyl
acetate at room temperature to give the ®nal products 4.
Biological Results
The compounds were tested in triplicate in a radioligand
binding assay, using cell membranes prepared from
HEK293 cells expressing the respective human receptor
subtypes. Results are expressed as Ki values in mM. In
Table 1, we report the e€ects of linear and branched alkyl
groups. The compounds with the smallest groups, e.g.,
hydrogen (5) and methyl (6) are equiactive on GluR5
*Corresponding author. Fax: +44-1276-853525; e-mail: baker_s_richard
@lilly.com
0960-894X/00/$ - see front matter # 2000 Published by Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00346-2

1808
S. R. Baker et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1807±1810
chains show similar potency to their straight-chain ana-
logues at GluR5, but the activity at GluR6 is further
reduced, making these even more selective, with anity
ratios of greater than 1000. These encouraging results
with larger alkyl groups led us to investigate the e€ect of
having a cycloalkyl or aryl group either directly bonded
to the ole®n or attached via a spacer group (Table 2).
Figure 1.
Further increasing the size of the substituents on the
ole®n causes the GluR5 activity to be reduced with only
the methyl cyclopentyl compound (15) showing good
potency. Finally, we investigated the e€ects of having two
substituents on the double bond, as shown in Table 3.
and 6 receptors. Further increasing the size of the alkyl
group greatly reduces the activity at GluR6 and in con-
trast the anity at GluR5 increases slightly. The net
e€ect is that compounds with the larger alkyl chains,
e.g., n-propyl (8) and n-butyl (9) show greater than 700-
fold selectivity for GluR5 over GluR6. The introduction
of an additional double bond, as in compound 10, has
little e€ect on activity. Compounds with branched alkyl
The SAR trend was similar to that seen with the mono-
substituted compounds. There was an initial decrease in
activity with the methyl compound (22) compared with
the unsubstituted analogue (5) but, as the size of the
alkyl group increased, the potency on GluR5 improved,
Scheme 1.
Table 1.
Table 2.
Compound
no.
Structure
Ki (mM) Ki (mM)
Ratio
GluR6/GluR5
Compound
no.
Structure
Ki (mM) Ki (mM) Ratio
GluR5 GluR6 GluR6/GluR5
GluR5
GluR6
5
6
0.27
2.2
0.45
2.5
1.67
1.14
15
16
0.152
87.6
576
6.06 >100.01
>16.5
7
LY310683
0.0605
18.0
298
8
0.0804
0.150
57.2
125.0
16.3
711
833
319
17
18
41.2
193.0
205.0
4.7
9
17.1
29.1
12.0
10
0.0511
19
20
>100.0
>100.0
>3.4
100
11
LY339637
0.0544
0.0936
82.5
1517
12
LY339675
>500.0
>5342
>1.0
>1.0
13
14
0.224
634.0
2870
21
>10.0
10
0.0857
>100.0
>1167

S. R. Baker et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1807±1810
1809
Table 3.
GluR5 ligand was the Z isopropyl compound LY339675
(12) which had a Ki value greater than 10 mM across the
kainate and AMPA receptors tested. Disappointingly,
the most potent compound LY339624 (25) had sig-
ni®cant anity for GluR4 receptor subtype, with a Ki
value of 0.533 mM, while the E isopropyl compound
LY339637 (11) had signi®cant anity at GluR1 and
GluR7 receptors. Although none of these compounds was
as potent as LY339434, they did show di€erent anity
pro®les and, in some cases, better subtype selectivity.
Compound
no.
Structure
Ki (mM) Ki (mM)
GluR5
Ratio
GluR6 GluR6/GluR5
22
13.9
>400.0
207.0
>28.8
859
23
LY339687
0.241
0.470
Finally we investigated whether the compounds were
agonists or antagonists using patch clamp electro-
physiology on the transfected HEK 293 cells, as shown
in Table 5.
24
>100.0
>100.0
>213
25
LY339624
0.0326
>3060
All the compounds tested were found to be agonists on
GluR5 receptors with ecacy similar to or slightly less
than kainic acid. LY339624 was also tested on rat dorsal
root ganglion (DRG) neurons (thought to express
GluR5 receptors) (EC50 1.05 mM) to con®rm that the
compound was an agonist on a native preparation as
well as on the HEK 293 GluR5 construct.
26
27
28
0.236
3.62
5.13
>100.0
46.3
>424
12.8
In summary, the GluR5 receptor binding of glutamate
is enhanced by the introduction of a lipophilic 4-alkyli-
dene group of a critical size. In contrast, the potency of
glutamic acid analogues at GluR6 receptors is reduced
by the introduction of a 4-alkylidene group. This
important di€erence could be useful in the design of
subtype selective GluR5 receptor ligands.
255.0
49.7
reaching a maximum with the cyclopentyl compound
(25). The more bulky substituted cyclopentyl com-
pounds 26 and 27, and the cyclohexyl compound (28),
were signi®cantly less potent, again demonstrating that
there is a clear limit to the size of alkyl group the GluR5
receptor will accept. The most potent and selective com-
pounds were then tested further on the other kainate and
AMPA receptors, as shown in Table 4. The most selective
Experimental
General procedure: synthesis of 4-alkylidene
pyroglutamates
The 4-alkylidenepyroglutamates 3 were prepared from
protected pyroglutamate as reported earlier12 and then
Table 4.
Compound
GluR1
Ki (mM)
GluR2
Ki (mM)
GluR4
Ki (mM)
GluR5
Ki (mM)
GluR6
Ki (mM)
KA2
Ki (mM)
GluR7
Ki (mM)
Kainate
7.5Æ2.0 (8)
1.36Æ0.26 (3)
>10.0
12.2Æ2.7 (4)
0.940Æ0.09 (3)
>10.0
1.71Æ0.17 (3)
0.177Æ0.022 (3)
0.701Æ0.046 (3)
0.0148
0.0324Æ0.0013 (4)
1.106Æ0.02 (3)
15.41
0.08Æ0.002 (5)
0.010Æ0.002 (3)
Glutamate
LY339434
LY339637
LY339675
LY339687
LY339624
0.868Æ0.0219 (3)
0.750Æ0.077 (3)
0.789Æ0.083 (3)
>10.0
5.6
10.3
4.71
0.533
>10.0
7.50
82.2
21.4
10.0
0.617
0.518
10.5
12.4
5.73
0.70
3.70
44.9
280.9
0.0544
0.0936
0.241
0.0326
82.5
500.0
207.4
>100.0
113.4
>100.0
1.10
3.68
Table 5.
Compound
GluR5
GluR6
DRG cells
EC50 (mM)
Rmax (%)
EC50 (mM)
Rmax (%)
EC50 (mM)
Rmax (%)
Kainate
16.2Æ1.0 (4)
75.0Æ7.5 (15)
2.5Æ0.86 (3)
14.2Æ1.64 (3)
6.2Æ3.8 (4)
100 (4)
0.7Æ0.1 (6)
25Æ2 (8)
>100 (5)
>300 (5)
>100
100
n.d.
n.d.
n.d.
n.e.
12.0Æ0.1 (4)
35.2Æ0.2 (5)
0.8Æ0.2 (6)
n.d.
100
51Æ10 (5)
57Æ5 (6)
n.d.
Glutamate
LY339434
LY310683
LY339624
79Æ2 (15)
80Æ0.08 (3)
94Æ4.0 (4)
79Æ12.9 (4)
1.05
n.d.

1810
S. R. Baker et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1807±1810
deprotected as follows: To a solution of the pyrogluta-
mate 3 (2 mmol) in THF (15 mL) was added a 2.5 N aqu-
eous solution of LiOH (14.4 mL, 36mmol). The mixture
was stirred at room temperature for 4 h and then acidi®ed
to pH 2 with 1 N HCl solution and extracted with diethyl
ether (3Â20mL). The combined organic layers were dried
over Na2SO4 and concentrated in vacuo to give an oily
residue, which was fully deprotected by treatment with a
saturated HCl solution in ethyl acetate for 1 h at room
temperature. After evaporation to dryness, the resulting
white solid was triturated with diethyl ether. The amino
acids 4 were isolated either as hydrochlorides or as zwit-
terions by treatment of a methanolic solution of the
hydrochloride with propylene oxide.
4. Nakanishi, S.; Masu, M. Annu. Rev. Biophys. Biomol.
Struct. 1994, 23, 319.
5. Sommer, B.; Seeburg, P. H. Trends Pharmacol. Sci. 1992,
13, 291.
6. Bleakman, D.; Lodge, D. Neuropharmocology 1998, 37,
1187.
7. Korczak, B.; Nutt, S. L.; Fletcher, E. J.; Hoo, K. H.;
Elliott, C. E.; Rampersand, V.; McWhinnie, E. A.; Kamboj,
R. K. Receptors Channels 1995, 3, 41.
8. Hoo, K. H.; Nutt, S.; Fletcher, E. J.; Elliot, C.; Korczak,
B.; Deverill, M.; Rampersad, Y.; Fantaske, R. P.; Kamboj, R.
K. Receptors Channels 1994, 2, 327.
9. Hollmann, M.; Heinemann, S. Annu. Rev. Neurosci. 1994,
17, 31.
10. Tolle, T. R.; Berthele, A.; Zieglgansberger, W.; Seeber, P.
H.; Wisden, W. J. Neurosci. 1993, 13, 5009.
11. Small, B.; Thomas, J.; Kemp, M.; Hoo, K.; Ballyk, B.;
Deverill, M.; Ogden, A. M.; Rubio, A.; Pedregal, C.; Bleak-
man, D. Neuropharmocology 1998, 37, 1261.
12. Ezquerra, J.; Pedregal, C.; Rubio, A.; Yruretagoyena, B.;
Carreno, M. C.; Escribano, A.; Ruano, J. L. G. J. Org. Chem.
1995, 60, 2925.
Binding studies and electrophysiology
The binding and electrophysiological studies were per-
formed as described by Clarke et al.14
13. The dehydration was carried out with mesyl chloride and
triethylamine except where R=tBu when it was necessary to
use Burgess reagent: Burgess, E. M.; Penton, H. R.; Taylor, E.
A. J. Org. Chem. 1973, 38, 26.
14. Clarke, V. R. J.; Ballyk, B. A.; Hoo, K. H.; Mandelzys,
A.; Pellizzari, A.; Bath, C. P.; Thomas, J.; Sharpe, E. F.;
Davies, C. H.; Ornstein, P. L.; Schoepp, D. D.; Kamboj, R.
K.; Collingridge, G. L.; Lodge, D.; Bleakman, D. Nature
1997, 389, 599.
References and Notes
1. Collingridge, G. L.; Lester, R. A. Pharmacol. Rev. 1989, 40,
143.
2. Cunningham, M. D.; Ferkany, J. W.; Enna, S. J. Life Sci.
1994, 54, 135.
3. Hollmann, M.; Heinemann, S. Annu. Rev. Neurosci. 1994,
17, 31.