Atropisomeric quinazolin-4-one derivatives are potent noncompetitive α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor antagonists

Article abstract of DOI:10.1016/S0960-894X(00)00622-3

Piriqualone (1) was found to be an antagonist of AMPA receptors. Structure-activity optimization was conducted on each of the three rings in 1 to afford a series of potent and selective antagonists. The sterically crowded environment surrounding the N-3 aryl group provided sufficient thermal stability for atropisomers to be isolated. Separation of these atropisomers resulted in the identification of (+)-38 (CP-465,022), a compound that binds to the AMPA receptor with high affinity (IC₅₀ = 36 nM) and displays potent anticonvulsant activity.

Full text of DOI:10.1016/S0960-894X(00)00622-3

Bioorganic & Medicinal Chemistry Letters 11 (2001) 177±181
Atropisomeric Quinazolin-4-one Derivatives are Potent
Noncompetitive ꢀ-Amino-3-hydroxy-5-methyl-4-
isoxazolepropionic Acid (AMPA) Receptor Antagonists
W. M. Welch, F. E. Ewing, J. Huang, F. S. Menniti, M. J. Pagnozzi, K. Kelly,
P. A. Seymour, V. Guanowsky, S. Guhan, M. R. Guinn, D. Critchett, J. Lazzaro,
A. H. Ganong, K. M. DeVries, T. L. Staigers and B. L. Chenard*
Global Research and Development, Groton Laboratories, P®zer Inc., Groton, CT 06340, USA
Received 15 September 2000; accepted 31 October 2000
AbstractÐPiriqualone (1) was found to be an antagonist of AMPA receptors. Structure±activity optimization was conducted on
each of the three rings in 1 to a€ord a series of potent and selective antagonists. The sterically crowded environment surrounding
the N-3 aryl group provided sucient thermal stability for atropisomers to be isolated. Separation of these atropisomers resulted in
the identi®cation of (+)-38 (CP-465,022), a compound that binds to the AMPA receptor with high anity (IC50=36 nM) and
displays potent anticonvulsant activity. # 2001 Elsevier Science Ltd. All rights reserved.
Introduction
Attesting to these concerns, several new compounds
(both competitive and noncompetitive) with improved
solubility have now been reported as the area of AMPA
receptor modulation continues to hold the interest of
the research community.6
The a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic
acid (AMPA) subtype of glutamate receptors is asso-
ciated with fast excitatory synaptic transmission in the
central nervous system. During the course of neuro-
degenerative conditions including ischemic stroke and
epilepsy, hyperactivation of glutamate receptors, and, in
particular, N-methyl-d-aspartate (NMDA) and AMPA
receptors, has been proposed as one of the processes
that contributes to neuron death.1 Consistent with this
hypothesis, antagonists of both receptor subtypes have
been protective in various anticonvulsant models.2
Unfortunately, the prototype quinoxalindiones such as
NBQX3 and YM90K,4 as well as other competitive
AMPA receptor antagonist classes, are generally poorly
soluble and thus dicult to formulate for clinical use.
Allosteric modulators of AMPA receptors based on the
2,3-benzodiazepine nucleus have recently been identi-
®ed. Compounds from this class, such as GYKI 52466
and LY 300164, have also shown bene®t as anti-
convulsants and in stroke models.5 However, poor
solubility and modest potency remain potential issues.
As part of ongoing studies into the pharmacology of
glutamate receptors, we have identi®ed a new antago-
nist template for AMPA receptors, the 3-aryl-2-(pyrid-
2-ylvinyl)-quinazolin-4-ones, exempli®ed by the known
anticonvulsant piriqualone.7 Herein, we describe a
*Corresponding author. Tel.: +1-860-441-4548; fax: +1-860-715-
8234; e-mail: chenardbl@groton.p®zer.com
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00622-3

178
W. M. Welch et al. / Bioorg. Med. Chem. Lett. 11 (2001) 177±181
Scheme 1. Synthesis of quinazolin-4-ones: (a) Ac2O, HOAc, re¯ux; (b) o-toluidine, HOAc, re¯ux; (c) 2-pyridinecarboxaldehyde, ZnCl2, Ac2O,
dioxane, re¯ux.
succinct SAR program that led to the discovery of (S)-
3-(2-chlorophenyl)-2-[2-(6-diethylaminomethyl-pyridin-
2-yl)-vinyl]-6-¯uoro-3H-quinazolin-4-one (CP-465,022),
a potent antagonist that interacts with the receptor
through an allosteric site.8 This compound is highly
selective for the AMPA receptor relative to NMDA or
kainic receptors and displays potent anticonvulsant
activity.
cerebellar granule cells.10 This method successfully
identi®ed the standard competitive and noncompetitive
antagonists, NBQX and GYKI 52466, and was used
exclusively for our primary SAR program. It is impor-
tant to note that NBQX must compete with kainate for
the binding site on the receptor. Thus, potency in this
assay is much weaker than its IC50 for [3H] AMPA
binding displacement might suggest.11 This assay was
adapted to a 96-well format for rapid screening and
piriqualone was identi®ed as a lead. This compound had
an IC50 of 0.50 mM and was found to inhibit AMPA
receptor activity in a manner not competitive with ago-
nist concentration. Subsequent experiments showed that
the GYKI 52466 and piriqualone shared at least partial
overlap at the binding site.12
Chemistry
The synthetic route to piriqualone combines three
structural fragmentsÐanthranilic acid (A ring), ortho
toluidine (B ring), and pyridine-1-carboxaldehyde (C
ring).9 Thus, SAR optimization was eciently con-
ducted through the combination of various commercial
or readily accessible analogues of these simple compo-
nents. In brief, anthranilic acid is converted to the benz-
oxazin-4-one with hot acetic anhydride in acetic acid
(Scheme 1). Subsequent reaction with a suitable aniline
fragment in re¯uxing acetic acid installed the B ring.
Finally, condensation of this resultant 2-methyl-3-aryl-
quinazolin-4-one with pyridine-2-carboxaldehydes pro-
vided the target compounds. The ®nal condensation
reaction was conveniently e€ected with acetic anhydride
and anhydrous zinc chloride in re¯uxing dioxane.
Development of SARs followed from combination of
the three ring fragments noted above. Optimization of
each ring was conducted while generally maintaining the
other two rings constant. In this fashion, a rapid SAR
was established. Limited opportunity for substitution
was found in the A ring (Table 1). Introduction of small
substituents, especially halogen at C-6, gave about a 4-
fold enhancement in activity. Since there was little dif-
ference between the various C-6 halogens, the 6-¯uoro
template was arbitrarily selected to explore structural
modi®cations in the B and C rings.
It soon became apparent that an ortho substituent in the
B ring was essential for good AMPA receptor antago-
nist activity (Table 2). However, 2,6-disubstitution even
with small ¯uorine atoms reduced potency. Substituents
as large as bromine, methyl and tri¯uoromethyl were
Pharmacology and SAR
To capture both competitive and noncompetitive
antagonists in a high throughput screen, a functional
assay of AMPA receptor activity was developed. We
tested compounds for their ability to block kainate-
induced 45Ca+2 in¯ux through AMPA receptors in rat
Table 2. B ring SAR
Table 1. A ring SAR
Entry
X
Y
Z
IC50 (nM)
1
H
H
H
H
H
H
F
F
F
F
CH3
F
Cl
Br
CF3
OCH3
CH3
F
Cl
Br
H
H
H
H
H
H
H
H
H
H
H
CH3
Cl
F
460Æ25
116Æ31
433Æ65
102Æ3
11
12
13
14
15
3
16
17
18
19
20
21
22
Entry
X
IC50 (nM)
1
2
3
4
5
6-H
6-CH3
6-F
6-Cl
7-Cl
460Æ25
1370Æ290
248Æ80
124Æ9
146Æ9
1100Æ600
248Æ80
180Æ30
96Æ8
>3000
6
7
8
9
10
8-Cl
6,8-Cl2
6-Br
8-OCH3
6,7-(OCH3)2
>1000
>1000
65Æ0.8
>1000
>1000
>1000
>300
H
H
H
F
H
CH3
Cl
162Æ34
>1000
>1000
F

W. M. Welch et al. / Bioorg. Med. Chem. Lett. 11 (2001) 177±181
179
well tolerated but did not lead to signi®cant potency
enhancement. However, a methoxy group reduced
activity by 10- to 20-fold.
observed suggesting a minimum 400-fold separation in
potencies at these two glutamate receptors.13 In addition,
no nonglutamate receptor anity was identi®ed at con-
centrations up to 1 mM.14
Up to this point, our SAR development in the A and B
rings showed very limited opportunity to modify the
structure and retain or improve on the activity pro®le.
Equally important, these changes produced compounds
that had relatively poor solubility (observed in handling
but not quantitatively measured). Therefore, we turned
to C ring modi®cations as we sought a means to optimize
SAR and solubility.
In vivo, 38 was very e€ective at blocking seizures in
mice induced by pentalenetetrazol (administered sc) or
AMPA (administered icv, ID50=3.8 and 3.6 mg/kg, sc,
respectively, Table 4).15 In anesthetized rat, 38 reduced
the population spike amplitude in the CA3 region of the
hippocampus (a measure of in vivo synaptic transmis-
sion). At a dose of 2 mg/kg iv, 65.9Æ9.1% blockade was
observed.16
Modi®cation of the pyridyl C ring was carried out on
the template containing the 6-¯uoroquinazolin-4-one A
ring and the ortho chlorophenyl B ring. Though many
changes were tolerated, C-6 proved to be an optimal
substitution site (Table 3). Thus a wide range of sub-
stituents including neutral and H-bond donors and
acceptors generally led to products with sub 100 nM
potency. We were particularly attracted to the amino-
methyl analogue. Though somewhat less potent
(270 nM), it o€ered a handle to enhance solubility.
Atropisomer Characterization
Restricted rotation generally due to the presence of
bulky substituents, especially in ortho substituted
biphenyls and related compounds, can result in stable
isolable chiral structures known as atropisomers.17 Per-
haps the best known atropisomers are the BINAP and
BINOL compounds which have found great utility as
catalysts for asymmetric synthetic processes.18
Mono- and disubstitution of the aminomethyl group
modestly improved potency and yielded several com-
pounds with attractive pro®les. In particular, the di-
ethylaminomethyl analogue 38 (CP-392,110) was
selected for further pro®ling. In patch clamp electro-
physiology experiments, the compound selectively
blocked AMPA receptor mediated inward currents with
an IC50 of 25 nM. No e€ect was observed on NMDA
receptor mediated currents at concentrations up to
1 mM. However, at 10 mM, a 36% current reduction was
Based on previous reports,19 it was highly likely that the
rotational barrier in these compounds would be su-
cient to allow isolation of the distinct atropisomers. In
combination with the quinazolin-4-one C-2 side chain
and the C-4 carbonyl, the ortho B ring SAR strongly
suggested that only one of the potential atropisomers of
our new AMPA receptor antagonists had the appro-
priate topological pro®le to ®t into a putative binding
site. Thus, we investigated whether a single atropisomer
would have sucient thermal stability for isolation. The
atropisomers of 38 were readily separated by chiral
HPLC.20 The individual isomers were further charac-
terized by optical rotation and X-ray crystallography
Table 3. C ring SAR
Entry
R
IC50, nM
17
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
H
CH3
OCH3
CHO
CN
COOCH3
COOH
CH2OH
CH2OAc
96Æ8
36Æ3
31Æ6
97Æ30
244Æ28
400
>1000
58Æ9
Table 4. Pharmacological pro®le of 38 and its atropisomers
45Æ3
38
(+)-38
( )-38
CH2OCH3
CH2OCH2CH3
CH2F
42Æ2
82Æ3
Inhibition of AMPA receptor
mediated 45Ca+2 uptake in rat
cerebellar neurons
74Æ4 nM 38Æ3 nM >300 nM
56Æ13
340Æ70
103Æ18
178Æ31
81Æ11
74Æ4
CH2NH2
CH2NHCH3
CH2N(CH3)2
CH2NHCH2CH3
CH2N(CH2CH3)2
-Pyrrolidine-
-Piperidine-
CH2NCH(CH3)2
CH2N(CH3)CH2CH2N(CH3)2
Blockade of pentalenetetrazol-
induced seizures in mice, sc
ID50 (95% CI)
3.8 mg/kg 1.5 mg/kg >10 mg/kg
(2.1±6.8) (0.31±3.2)
Blockade of ivc AMPA-induced 3.6 mg/kg 3.1 mg/kg Not tested
seizures in mice, sc
ID50 (95% CI)
31Æ4
(3.2±4.1)
(2.1±4.5)
>300
59Æ10
150Æ60
Inverted grid, mice, MED, sc
10.0 mg/kg 5.6 mg/kg Not tested

180
W. M. Welch et al. / Bioorg. Med. Chem. Lett. 11 (2001) 177±181
and had the absolute con®guration illustrated below. To
evaluate thermal stability, solutions of (+)-38 were
maintained at various temperatures and monitored by
HPLC.21 At 37 ꢀC, no equilibration was detected over a
7 day period. At higher temperatures, thermal equili-
bration was observed. For example, at 70 ꢀC, 10%
racemization was noted after 90 h. Importantly, no
equilibration was seen over 1 week in pH 7.4 phosphate
bu€er at 37 ꢀC, suggesting good in vivo stability.
such as (+)-38 will likely play a critical role in de®ning
the role of the AMPA receptor in the treatment of
neurodegenerative and other diseases of the central
nervous system.29
References and Notes
1. (a) Chimirri, A.; Gitto, R.; Zappala, M. Exp. Opin.
Ther. Patents 1999, 9, 557. (b) Lees, G. J. CNS Drugs 1996,
5, 51. (c) Danysz, N.; Parsons, C. G.; Bresink, I.; Quack, G.
Drugs News Perspect. 1995, 8, 261.
2. Lehmann, J. Exp. Opin. Ther. Patents 1997, 7, 1099.
3. Sheardown, M. J.; Nielsen, E. O.; Hansen, A. J.; Jacobsen,
P.; Honore, T. Science 1990, 247, 571.
4. Ohmori, J.; Sakamoto, S.; Kubota, H.; Shimizu-Sasamata,
M.; Okada, M.; Kawasaki, S.; Hidaka, K.; Togami, J.; Fur-
uya, T.; Murase, K. J. Med. Chem. 1994, 37, 467.
Consistent with our previous SAR which taught that a
single ortho substituent was essential for activity, only
one of the atropisomers, (+)-38, blocked AMPA
receptor activity in the primary assay (Table 4). Like-
wise, all the anticonvulsant activity resided in this com-
pound. AMPA receptor antagonists had been reported
to have CNS depressant side e€ects. Thus, we further
evaluated 38 and its atropisomers for inverted grid per-
formance in mice.22 Minimum e€ective doses (MEDs)
for behavioral disruption in these experiments were 2-
to 4-fold higher than the anticonvulsant ID50s. This
separation from side e€ects was consistent with the
previous report.23
5. (a) GYKI 52466: Smith, S.; Durmuller, N.; Meldrum, B. S.
Eur. J. Pharmacol. 1991, 201, 179. (b) Yamaguchi, S.; Don-
evan, S. D.; Rogawski, M. A. Epilepsy Res. 1993, 15, 179. (c)
Donevan, S. D.; Yamaguchi, S.; Rogawski, M. A. J. Pharma-
col. Exp. Ther. 1994, 271, 25. (d) De Sarro, G.; Chimirri, A.;
De Sarro, A.; Gitto, R.; Grasso, S.; Guisti, P.; Chapman, A. G.
Eur. J. Pharmacol. 1995, 294, 411. (e) Smith, S.; Meldrum,
B. S. Stroke 1992, 23, 861. (f) LY 300164: Borowicz, K. K.;
Kleinrok, Z.; Czuczwar, S. J. Naunyn-Schmiedeberg's Arch.
Pharmacol. 2000, 361, 629. (g) Lodge, D.; Bond, A.; O'Neill,
M. J.; Hicks, C. A.; Jones, M. G. Neuropharmacology 1996,
35, 1681.
6. (a) Turski, L.; Huth, A.; Sheardown, M.; McDonald, F.;
Neuhaus, R.; Schneider, H. H.; Dirnagl, U.; Wiegand, F.;
Jacobsen, P.; Ottow, E. Proc. Natl. Acad. Sci. U.S.A. 1998, 95,
10960. (b) Takahashi, M.; Ni, J. W.; Kawasaki-Yatsugi, S.;
Toya, T.; Yatsugi, S.-I.; Shimizu-Sasamata, M.; Koshiya, K.;
Shishikura, J.-I.; Sakamoto, S.; Yamaguchi, T. J. Pharmacol.
Exp. Ther. 1998, 284, 467. (c) Nijholt, I.; Blank, T.; Gra-
felmann, B.; Cepok, S.; Kugler, H.; Spiess, J. Brain Res. 1999,
821, 374. (d) Stutzmann, J.-M.; Bohme, G. A.; Boireau, A.;
Damour, D.; Debono, M. W.; Genevois-Borella, A.; Imper-
ato, A.; Jimonet, P.; Pratt, J.; Randle, J. C. R.; Ribeill, Y.;
Vuilhorgne, M.; Mignani, S. Bioorg. Med. Chem. Lett. 2000,
10, 1133.
Finally, poor solubility has been a recurring issue for
most previously described classes of AMPA receptor
antagonists. It was assumed that the basic C ring side
chain would endow this new series with an improved
solubility pro®le. This has proven to be the case. Pre-
paration of (+)-38 as its highly crystalline mesylate salt
provided a product with excellent aqueous solubility
(180 mg/mL at pH 4.7). Thus formulation, even for iv
administration, should be feasible.
Conclusion
In summary, the quinazolin-4-ones represent a new and
highly selective class of AMPA receptor antagonists.
These quinazolin-4-ones exist as stable isolable atrop-
isomers and all the antagonist activity resides in a single
antipode (S isomer in the case of 38). Within the context
of SAR development, introduction of a basic amine side
chain on the pyridyl C ring produced an optimal com-
bination of potency and aqueous solubility. The end
result of this e€ort was the identi®cation of (+)-38 as
an ideal probe to explore the pharmacology of the
AMPA receptor. Consistent with previous ®ndings,
(+)-38 was a potent anticonvulsant when evaluated
against pentalenetetrazol and AMPA induced seizures
while retaining a modest safety pro®le.
7. Mori, A.; Mukawa, J.; Okusu, H.; Kobayashi, K.; Mizuno,
A. Brain Nerve 1969, 21, 271.
8. (a) Portions of this work have appeared in abstract form:
Welch, W. M.; Huang, J.-H.; Ewing, F. E.; Menniti, F. S.;
Pagnozzi, M. J.; Banker, M. J.; Devries, K. M.; McCarthy, K. E.
Annual Meeting of the Society for Neuroscience, Los Angeles,
CA, 1998; Abstract 44.15. (b) For a related structural series
see: Chenard, B. L.; Menniti, F. S.; Pagnozzi, M. J.; Shenk,
K. D.; Ewing, F. E.; Welch, W. M. Bioorg. Med. Chem. Lett.
2000, 10, 1203.
9. Boltze, K.-H.; Dell, H.-D.; Lehwald, H.; Lorenz, D.;
Ruberg-Schwear, M. Arzneim. Forsch. 1963, 13, 688.
10. Menniti, F. S.; Collins, M.; Ducat, M. F.; Kelly, K. A.;
Pagnozzi, M. J.; Richter, K. E. G. Annual Meeting of the Society
for Neuroscience, Los Angeles, CA, 1998; Abstract 44.12.
11. (a) Sheardown, M. J.; Nielsen, E. é.; Hansen, A. J.;
Jacobsen, P.; Honore, T. Science 1990, 247, 571. (b) Cordi,
A. A.; Desos, P.; Randle, J. C. R.; Lepagnol, J. Bioorg. Med.
Chem. 1995, 3, 129. (c) Ohmori, J.; Shimizu-Sasamata, M.;
Okada, M.; Sakamoto, S. J. Med. Chem. 1996, 39, 3971.
12. Menniti, F. S.; Chenard, B. L.; Collins, M. B.; Ducat,
M. F.; Elliott, M. L.; Ewing, F. E.; Huang, J. I.; Kelly, K. A.;
Lazzaro, J. T.; Pagnozzi, M. J.; Weeks, J. L.; Welch, W. M.;
White, W. F. Mol. Pharmacol. 2000, 58, 1310.
Beyond epilepsy, AMPA receptor antagonists may have
therapeutic bene®t in the treatment of ischemic stroke
(both focal and global),24 CNS trauma,25 pain,26 and
Parkinson's disease27 based on activity in various in
vivo disease models. However, recently, the selectivity
of many of the early prototype AMPA antagonists,
particularly relative to the kainate ionotropic glutamate
receptors, has been called into question.28 Thus, the true
spectrum of activity of a pure AMPA receptor blocker
may be less clear cut than the original in vivo studies
might suggest. Therefore, highly selective compounds

W. M. Welch et al. / Bioorg. Med. Chem. Lett. 11 (2001) 177±181
181
13. Lazzaro, J. T.; Ganong, A. H. Annual Meeting of the Society
for Neuroscience, Los Angeles, CA, 1998; Abstract 44.13.
14. A battery of about 40 transmitter binding sites, ion chan-
nel binding sites, and transmitter reuptake sites were evaluated
by Novascreen, 7170 Standard Drive, Hanover, MD 21076-
1334. No response ꢁ50% was observed at the 1 mM test con-
centration.
15. Seymour, P. A.; Critchett, D.; Guanowsky, V.; Wilkinson,
L. O. Annual Meeting of the Society for Neuroscience, Los
Angeles, CA, 1998; Abstract 44.14.
22. Coughenor, L. L.; McLean, J. R.; Parker, R. B. Pharma-
col. Biochem. Behav. 1977, 6, 351.
23. Ornstein, P. L.; Arnold, M. B.; Augenstein, N. K.; Lodge,
D.; Leander, J. D.; Schoepp, D. D. J. Med. Chem. 1993, 36,
2046.
24. Buchan, A. M.; Lesiuk, H.; Barnes, K. A.; Li, H.; Huang,
Z.-G.; Smith, K. E.; Xue, D. Stroke 1993, 24, I-148.
25. Bernert, H.; Turski, L. Proc. Natl. Acad. Sci. U.S.A. 1996,
93, 5235.
26. Sang, C. N.; Hostetter, M. P.; Gracely, R. H.; Chappell,
A. S.; Schoepp, D. D.; Lee, G.; Witcup, S.; Caruso, R.; Max,
M. B. Anesthesiology 1998, 89, 1060.
16. Side e€ects at higher doses prevented accurate assessment
of an ED50. Manuscript in preparation.
17. March, J. Advanced Organic Chemistry; John Wiley &
Sons: New York, 1992: pp 101±102.
27. Klockgether, T.; Turski, L.; Honore, T.; Zhang, Z.; Gash,
D. M.; Kurlan, R.; Greenamyre, T. Ann. Neurol. 1991, 30,
717.
28. Bleakman, D.; Ogden, A.-M.; Ornstein, P. L.; Hoo, K.
Eur. J. Pharmacol. 1999, 378, 331.
29. Compound (+)-38 at concentration of 0.10 mM com-
pletely blocked AMPA receptor responses in hippocampal
neurons but had no e€ect on kainate receptor responses in
dorsal route ganglia neurons. Menniti, F. S.; Buchan, A.;
Chenard, B.; Critchett, D.; Collins, M. A.; Ganong, A.; Gua-
nowsky, V.; Lazzaro, J.; Lerma, J.; Pagnozzi, M.; Seymour,
P.; Volberg, W.; Welch, W. M. Manuscript in preparation.
18. Noyori, R. In Stereocontrolled Organic Synthesis; Trost,
B. M., Ed.; Blackwell: Oxford, UK, 1994; pp 1±15.
19. (a) Colebrook, L. D.; Giles, H. G. Can. J. Chem. 1975, 53,
3431. (b) Mannschreck, A.; Koller, H.; Stuehler, G.; Davies,
M. A.; Traber, J. Eur. J. Med. Chem.ÐChim. Ther. 1984, 19,
381.
20. HPLC conditions: column: Chiracel OD; mobile phase:
93/7 hexane/ethanol.
21. Newell, L. M.; Sekhar, V. C.; DeVries, K. M.; Staigers, T.
L.; Finneman, J. I. Submitted for publication.