Two prodrugs of potent and selective GluR5 kainate receptor antagonists actives in three animal models of pain

Article abstract of DOI:10.1021/jm0491952

Amino acids 5 and 7, two potent and selective competitive GluR5 KA receptor antagonists, exhibited high GluR5 receptor affinity over other glutamate receptors. Their ester prodrugs 6 and 8 were orally active in three models of pain: reversal of formalin-i

Full text of DOI:10.1021/jm0491952

4200
J. Med. Chem. 2005, 48, 4200-4203
Letters
Chart 1
Two Prodrugs of Potent and Selective
GluR5 Kainate Receptor Antagonists
Actives in Three Animal Models of Pain
Esteban Dominguez,*,§ Smriti Iyengar,†
Harlan E. Shannon,† David Bleakman,† Andrew Alt,†
Brian M. Arnold,† Michael G. Bell,†
Thomas J. Bleisch,† Jennifer L. Buckmaster,†
Ana M. Castano,§ Miriam Del Prado,§
Ana Escribano,§ Sandra A. Filla,† Ken H. Ho,‡,#
Kevin J. Hudziak,† Carrie K. Jones,†
Jose A. Martinez-Perez,§ Ana Mateo,§
Brian M. Mathes,† Edward L. Mattiuz,†
Ann Marie L. Ogden,† Rosa Maria A. Simmons,†
Douglas R. Stack,† Robert E. Stratford,†
Mark A.Winter,† Zhipei Wu,† and Paul L. Ornstein†
Centro de Investigacio´n Lilly,
S. A. Avda. De la Industria 30, 28108-Alcobendas,
Madrid, Spain, Lilly Research Laboratories, Eli Lilly and
Company, Indianapolis, Indiana 46285, and Allelix
Biopharmaceuticals, Mississauga, Ontario L4V1P1, Canada
(DRG) cell bodies are associated with primary afferent
neurons. It was the early work of Agrawal and Evans
that identified that C-fiber afferents possess KA recep-
tors.4 The biophysical and pharmacological profile of KA
receptors in DRG neurons suggests that they likely
comprise GluR5 homomers.5 Recent studies in spinal
cord slices have implicated KA receptors, specifically the
GluR5 receptor subtype, in pain transmission.6 Sang et
al. showed that the decahydroisoquinoline 1 (LY293558),7
a competitive AMPA/KA antagonist, significantly re-
duced pain intensity and unpleasantness in an experi-
mentally induced human pain study and reduced clini-
cal pain in a study of evoked pain.8 Simmons et al.
showed that 2 (LY382884),9 a selective GluR5 KA
receptor antagonist, was efficacious in the formalin
test10 (an animal model of persistent pain).11 More
recently, Filla et al. showed that 3, a highly selective
competitive GluR5 KA receptor antagonist, exhibited
high GluR5 receptor affinity and selectivity over other
glutamate receptors. Its diethyl ester prodrug 4 was
orally active in two animal models of migraine pain
(Chart 1).12
Received October 6, 2004
Abstract: Amino acids 5 and 7, two potent and selective
competitive GluR5 KA receptor antagonists, exhibited high
GluR5 receptor affinity over other glutamate receptors. Their
ester prodrugs 6 and 8 were orally active in three models of
pain: reversal of formalin-induced paw licking, carrageenan-
induced thermal hyperalgesia, and capsaicin-induced mechan-
ical hyperalgesia.
Glutamate is the primary neurotransmitter in the
mammalian central nervous system. Knowledge of the
existence of subtypes of glutamate receptors was ad-
vanced by the observations that excitatory amino acids
(EAA) such as N-methyl-D-aspartate (NMDA), quis-
qualate, R-amino-3-hydroxy-5-methyl-4-isoxazole pro-
pionic acid (AMPA), and kainic acid (KA) showed
different potencies on subsets of neurons. This informa-
tion, coupled with recent molecular biology studies, led
to the current classification of glutamate receptors into
metabotropic glutamate receptors1 (G-protein-coupled)
and ionotropic glutamate receptors2 (ligand-gated ion
channels). The latter receptors are defined, on the basis
of subtype selective agonists, as NMDA, AMPA, and KA.
Of the AMPA and KA ionotropic glutamate receptors
cloned, subunits GluR1-4 are AMPA-sensitive, while
GluR5-7 and KA1-2 are AMPA-insensitive and KA-
preferring.3
We describe here the in vitro activity of 5 and 7 as
potent and selective GluR5 KA receptor antagonists,
and the oral activity of their ester prodrugs 6 and 8, in
three animal models of pain.
The preparation of 5 (LY458545), 6 (LY467711), 7
(LY457691), and 8 (LY525327) is outlined in Scheme 1.
All compounds were derived from the ketone intermedi-
ate 9, the synthesis of which has previously been
described.13 Treatment of 9 with TMSI selectively
removed the methyl carbamate protecting group, and
the resulting amine was protected again as a Boc
derivative. Then, successful reduction of the ketone was
achieved with sodium borohydride in the presence of
cerium chloride, which provided the hydroxy ester 10
with the desired stereochemistry at C-6, as shown in
Scheme 1.7b The resulting alcohol was converted to 11
under typical Mitsunobu reaction conditions14 (PPh3,
DEAD). The benzonitrile 11 was converted to the
tetrazole by treatment with neat tri-n-butyltin azide at
Transfer of nociceptive information from the periph-
ery to the spinal cord occurs via C-fiber primary
afferents. Within dorsal roots, dorsal root ganglion
* To whom correspondence should be addressed. Phone: +34 91
6633406. Fax: +34 91 6633411. E-mail: dominguez_esteban@lilly.com.
§ Centro de Investigacio´n Lilly.
† Lilly Research Laboratories.
‡ Allelix Biopharmaceuticals.
# Present address: Lilly Research Laboratories, Eli Lilly and
Company, Indianapolis, IN, 46285.
10.1021/jm0491952 CCC: $30.25 © 2005 American Chemical Society
Published on Web 05/21/2005

Letters
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13 4201
Scheme 1a
Figure 1. Inhibition of glutamate (100 µM)-invoked calcium
influx by 5 in the human GluR5(Q), GluR5/6, and GluR5/KA2
receptors stably transfected in HEK293 cells, with mean value
( SEM for n ) 3.
a Conditions: (a) TMSI, CH2Cl2, room temp; (b) Boc2O, Et3N,
CH2Cl2, room temp (92% yield two steps); (c) CeCl3, EtOH, NaBH4,
from -78 °C to room temp. (60% yield); (d) 2-cyanophenol, PPh3,
DEAD, Py, room temp. (51% yield); (e) (i) n-Bu3SnN3, 80 °C; (ii)
2.5M LiOH, THF, 50 °C; (iii) HCl (g)/EtOAc room temp. (50% yield,
three steps); (f) i-BuOH/ HCl(g), (92% yield), (g) 2-aminobenzoni-
trile, NaBH(OAc)3, CH2Cl2, room temp. (32% yield); (h) (i)
n-Bu3SnN3, 80 °C; (ii) 6N HCl reflux (56% yield, two steps); (i)
2-ethylbutyl alcohol/ HCl(g), (87% yield).
80 °C for 3 days followed by ester hydrolysis (2.5 M
LiOH), protecting group removal (HCl(g), EtOAc), and
purification by ion exchange chromatography to yield
the amino acid 5 as a free base (97% purity). Addition-
ally, acid 5 was further esterified to provide ester 6 as
the hydrochloride salt (95% purity) under standard
conditions with isobutyl alcohol.15
Figure 2. Inhibition of glutamate (100 µM)-invoked calcium
influx by 7 in the human GluR5(Q), GluR5/6, and GluR5/KA2
receptors stably transfected in HEK293 cells, with mean value
( SEM for n ) 3.
Reductive amination of 9 with 2-aminobenzonitrile in
the presence of NaBH(OAc)3 provided exclusively 12 in
moderate yield. Tetrazole formation and final treatment
with 6N HCl and purification by ion exchange chroma-
tography provided the desired amino acid 7 as a free
base (98% purity). The amino acid 7 was esterified with
2-ethylbutyl alcohol to yield 8 as a hydrochloride salt
(97% purity).
Amino acids 5 and 7 were evaluated in ligand binding
studies for their effectiveness in displacing binding of
[3H]AMPA to recombinant human AMPA receptors and
[3H]KA to recombinant human KA receptors expressed
in HEK 293 cell membranes.16 Table 1 shows the
affinity of 5 and 7 for the various cloned human AMPA
and KA receptors compared with the competitive AMPA/
KA antagonist 1, the selective GluR5 KA receptor
antagonist 2, and the quinoxalidinedione AMPA/KA
antagonist 13 (NBQX), a standard tool used in the study
of AMPA receptors.17
Compounds 5 and 7 behaved as antagonists, inhibit-
ing glutamate-evoked calcium influx in the human
GluR5(Q), GluR5/6, and GluR5/KA2 receptor stably
transfected in HEK293 cell membranes, with IC50
values of 0.65 ( 0.10 and 0.44 ( 0.08, 0.38 ( 0.07 and
0.24 ( 0.04, and 0.78 ( 0.01 and 0.65 ( 0.04 µM,
respectively (Figures 1 and 2).16,18
We evaluated the oral efficacy of these GluR5 KA
receptor antagonists in three animal models of pain.
Parent amino acids 5 and 7 themselves exhibited poor
oral bioavailability in the rat; consequently, we evalu-
ated esters 6 and 8 as potential prodrugs of 5 and 7,
respectively.
Oral dosing of 6 and 8 showed rapid absortion, and
their oral bioavailabilities were determined to be 24%
and 41%, respectively. Plasma concentrations of 5 and
7 1 h after dosing following 10 mg/kg oral dose of 6 and
8 were 840 and 938 ng/mL, respectively. On the basis
Table 1. In Vitro Binding Affinities of AMPA/KA Antagonists for Recombinant Human AMPA and KA Receptorsa
compd
GluR2b
GluR5(Q)c
GluR5/6c
GluR5/KA2c
GluR6c
GluR6/KA2c
1
2
2.21 ( 0.48
>100
4.16 ( 1.39
17.13 ( 3.60
1.69 ( 0.42
7.37 ( 1.50
1.55 ( 0.59
31.75 ( 3.49
11.73 ( 1.74
2.10 ( 0.76
10.17 ( 6.12
1.00 ( 0.60
5.98 ( 1.14
1.37 ( 0.87
13.71 ( 3.63
6.12 ( 2.29
16.2 ( 4.0
>100
>100
>100
>100
>100
>100
>100
>100
0.6 ( 0.1
>100
5
8.26 ( 2.03
>100
7.89 ( 2.16
60.9 ( 22.4
9.86 ( 1.93
>100
>100
6
>100
7
5.51 ( 1.76
>100
0.21 ( 0.03
>100
8
>100
13
26.6 ( 7.0
13.2 ( 1.9
a Affinities for receptors (Ki, µM) were determined in vitro by radioligand binding assays using HEK 293 cell membranes expressing
the appropriate human AMPA or KA receptor. Each value is the mean ( SEM of at least three determinations. b [3H]AMPA was used as
the high-affinity radioligand. c [3H]KA was used as the high-affinity radioligand.

4202 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13
Letters
Figure 5. Dose-related reversal of capsaicin-induced mechan-
ical hyperalgesia by 6 and 8 in comparison with morphine in
rats (n ) 6 per group).
Figure 3. Effects of 6 and 8 on formalin-induced late-phase
paw-licking pain behavior 1 h after oral administration in
fasted male Sprague Dawley rats. Late phase behavior is
defined as 15-40 min after intraplantar injection of 50 µL of
5% formalin in the hind paw. Data are expressed as percent
of control (mean ( SEM). Morphine (oral) is plotted as a
comparator: (/) p < 0.05 significantly different from vehicle.
Figure 6. Effects of 6 and 8 on performance on the rotorod 1
and 2 h after oral administration in fasted male Sprague
Dawley rats. Baseline performance was 40 s. Data are
expressed as time on rotorod in seconds (mean ( SEM): (/) p
< 0.05 significantly different from no drug.
pain models occurred at several doses that did not show
significant performance deficits in the rotorod test.
In summary, 6 and 8, the ester prodrugs of amino
acids 5 and 7, demonstrated oral efficacy in three well-
established animal models of pain.
Figure 4. Dose-related reversal of carrageenan-induced
thermal hyperalgesia by 6 and 8 in comparison with ibuprofen
in rats (n ) 6 per group).
Supporting Information Available: Experimental pro-
cedures, including analytical and spectral data, for the prepa-
ration of 5-8 and 10-12 and experimental details for the
functional studies in recombinant human GluR5 KA receptors
and for the three animal models of pain. This material is
available free of charge via the Internet at http://pubs.acs.org.
of these results, we concluded that esters 6 and 8 would
act as prodrugs of amino acids 5 and 7.
In the formalin model, 6 and 8 produced dose-
dependent reversal of formalin-induced late-phase paw-
licking pain behavior and were highly efficacious after
oral administration (Figure 3). Statistically significant
effects were evident beginning at 3 mg/kg po with 6 and
5 mg/kg po with 8. In comparison, morphine was much
less potent and efficacious, with a minimal effective dose
of 40 mg/kg po (Figure 3).
Both 6 and 8 produced dose-related reversals of
carrageenan-induced thermal hyperalgesia with mini-
mal effective doses (MEDs) of 1.0 and 3.0 mg/kg po,
respectively (Figure 4). In comparison, the nonsteroidal
antiinflammatory drug ibuprofen also reversed carra-
geenan-induced hyperalgesia with an MED of 1000 mg/
kg po.
In addition, 6 and 8 also produced dose-related
reversals of capsaicin-induced mechanical hyperalgesia
with MEDs of 1.0 and 10 mg/kg po, respectively (Figure
5). Morphine also produced a dose-related reversal of
capsaicin-induced hyperalgesia with an MED of 3.0 mg/
kg sc.
Compound 6 caused deficits in the rotorod test of
neurological function at 30 mg/kg po, whereas 8 showed
performance deficits beginning at 20 mg/kg po (Figure
6). Importantly, the effects of 6 and 8 in the persistent
References
(1) Schoepp, D. D.; et al. Trends Pharmacol. Sci. 1993, 14, 13.
(2) Collingridge, G. L.; Lester, R. A. J. Excitatory amino acid
receptors in the vertebrate central nervous system. Pharmacol.
Rev. 1989, 40, 143-210.
(3) (a) Bleakman, D.; Lodge, D. Neuropharmacology of AMPA and
KA receptors. Neuropharmacology 1998, 37, 1187-1204. (b)
Hollman, M.; Heinemann, S. Cloned glutamate receptors. Annu.
Rev. Neurosci. 1994, 17, 31-108. (c) Seeburg, P. H. The molec-
ular biology of mammalian glutamate receptor channels. Trends
Neurosci. 1993, 16, 359-365.
(4) Agrawal, S. G.; Evans, R. H. The primary afferent depolarizing
action of KA in the rat. Br. J. Pharmacol. 1986, 87, 345-355.
(5) (a) Huettner, J. E. Glutamate receptor channels in DRG
neurons: activation by KA and quisqualate and blockade of
desensitization by concanavalin. Neuron 1990, 5, 255-266. (b)
Partin, K. M.; Patneau, D. K.; Winters, C. A.; Mayer, M. L.;
Buonanno, A. Selective modulation of desensitization at AMPA
versus KA receptors by cyclothiazide and concanavalin A.
Neuron 1993, 11, 1069-1082.
(6) (a) Li, P.; Wilding, T. J.; Kim, S. J.; Calejesan, A. A.; Huettner,
J. E.; Zhuo, M. Kainate-receptor-mediated sensory synaptic
transmission in mammaliam spinal cord. Nature 1999, 397,
161-164. (b) Procter, M. J.; Houghton, A. K.; Faber, E. S. L.;
Chizh, B. A.; Ornstein, P. L.; Lodge, D.; Headley, P. M. Actions
of KA and AMPA selective glutamate receptor ligands on
nociceptive processing in the spinal cord. Neuropharmacology
1998, 37, 1287-1297. (c) Evans, R. H.; Long, S. K. Primary
afferent depolarization in the rat spinal cord is mediated by
pathways utilizing NMDA and non-NMDA receptors. Neurosci.
Lett. 1989, 100, 231-236. (d) Evans, R. H.; Evans, S. J.; Pook,

Letters
P. C.; Sunter, D. C. A comparison of excitatory amino acid
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13 4203
(10) Simmons, R. M. A.; Li, D. L.; Hoo, K. H.; Deverill, M.; Ornstein,
P. L.; Iyengar, S. Kainate GluR5 receptor subtype mediates the
nociceptive reponse to formalin in the rat. Neuropharmacology
1998, 37, 25-36.
(11) Heapy, C. G.; Jamieson, A.; Russell, N. J. W. Afferent C-fiber
and A-delta activity in models of inflammation. Br. J. Pharma-
col. 1987, 90, 164P.
antagonists acting at primary afferent C fibers and motoneurons
of the isolated spinal cord of the rat. Br. J. Pharmacol. 1987,
91, 531-537.
(7) (a) Ornstein, P. L.; Arnold, M. B.; Augenstein, N. K.; Lodge, D.;
Leander, J. D.; Schoepp, D. D. (3SR,4aRS,6RS,8aRS)-6-[2(1H-
Tetrazol-5-yl)ethyl]decahydroisoquinoline-3-carboxylic acid:
A
(12) Filla, S. A.; Winter, M. A.; Johnson, K. W.; Bleakman, D.; Bell,
M. G.; Bleisch, T. J.; Castan˜o, A. M.; Clemens-Smith, A.; Del
Prado, M.; Dieckman, D. K.; Dominguez, E.; Escribano, A.; Hoo,
K. H.; Hudziak, K. J.; Katofiasc, M. A.; Martinez-Perez, J. A.;
Mateo, A.; Mathes, B. M.; Mattiuz, E. L.; Ogden, A. M. L.;
Phebus, L. A.; Stack, D. R.; Stratford, R. E.; Ornstein, P. L. Ethyl
(3S,4aR,6S8aR)-6-(4-Ethoxycarbonylimidazol-1-ylmethyl)decahy-
droisoquinoline-3-carboxylic ester: A prodrug of a GluR5 KA
receptor antagonist active in two animal models of acute
migraine. J. Med. Chem. 2002, 45, 4383-4383.
(13) Ornstein, P. L.; Arnold, M. B.; Augenstein, N. K.; Paschal, J.
W. Syntheses of 6-oxodecahydroquinoline-3-carboxylates. Useful
intermediates for the preparation of conformationally defined
excitatory amino acid antagonists. J. Org. Chem. 1991, 56,
4388-4392
(14) Mitsunobu, O.; Yamada, M. Bull. Chem. Soc. Jpn. 1967, 40,
2380.
structurally novel, systemically active, competitive AMPA recep-
tor antagonist. J. Med. Chem. 1993, 36, 2046-2048. (b) Ornstein,
P. L.; Arnold, M. B.; Allen, N. K. Bleisch, T.; Borromeo, P. S.;
Lugar, C. W.; Leander, J. D.; Lodge, D.; Schoepp, D. D.
Structure-activity studies of 6-(tetrazolylalkyl)-substituted
decahydroisoquinoline-3-carboxylic acid AMPA receptor antago-
nists. 1. Effects of stereochemistry, chain length, and chain
substitution. J. Med. Chem. 1996, 39, 2219-2231. (c) Ornstein,
P. L.; Arnold, M. B.; Allen, N. K.; Bleisch, T.; Borromeo, P. S.;
Lugar, C. W.; Leander, J. D.; Lodge, D.; Schoepp, D. D.
Structure-activity studies of 6-(tetrazolylalkyl)-substituted
decahydroisoquinoline-3-carboxylic acid AMPA receptor antago-
nists. 2. Effects of distal acid bioisosteric substitution, absolute
stereochemical preferences, and in vivo activity. J. Med. Chem.
1996, 39, 2232-2244.
(8) (a) Sang, C. N.; Hostetter, M. P.; Gracely, R. H.; Chappell, A.
S.; Schoepp, D. D.; Lee, G.; Whitcup, S.; Caruso, R.; Max, M. B.
AMPA/kainate antagonist LY293558 reduces capsaicin-evoked
hyperalgesia but not pain in normal skin in humans. Anesthe-
siology 1998, 89, 1060-1067. (b) Gilron, I.; Max, M. B.; Lee, G.
Booher, S. L.; Sang, C. N.; Chappell, A. S.; Dionne, R. A. Effects
of the 2-amino-3-hydroxy-5-methyl-isoxazole-propionic acid/KA
antagonist LY293558 on spontaneous and evoked postoperative
pain. Clin. Pharmacol. Ther. 2000, 68, 320-327.
(9) Bleisch, T.; Ornstein, P. L.; Allen, N. K.; Wright, R. A.; Lodge,
D.; Schoepp, D. D. Structure-activity studies of aryl-spaced
decahydroisoquinoline-3-carboxylic acid AMPA receptor antago-
nists. Bioorg. Med. Chem. Lett. 1997, 7, 1161-1166.
(15) Patent WO 01/02367 A2, 2001 (on prodrugs).
(16) Small, B.; Thomas, J.; Kemp, M.; Hoo, K.; Ballyk, B.; Deverill,
M.; Ogden, A. M.; Rubio, A.; Pedregal, C.; Bleakman, D.
LY339434, a GluR5 KA receptor agonist. Neuropharmacology
1998, 37, 1261-1267.
(17) Sheardown, M. J.; Nielsen, E. Ø.; Hansen, A. J.; Jacobsen, P.;
Honore´, T. 2,3-Dihydroxy-6-nitro-7-sulfamoyl-benzo(F)quinoxa-
line: a neuroprotectant for cerebral ischemia. Science 1990, 247,
571-574.
(18) For experimental methods, see Supporting Information.
JM0491952