The synthesis and structure-activity relationships of 4-aryl-3-aminoquinolin-2-ones: A new class of calcium-dependent, large conductance, potassium (maxi-K) channel openers targeted for post-stroke neuroprotection

Article abstract of DOI:10.1016/S0960-894X(02)00240-8

A series of 4-aryl-3-aminoquinoline-2-one derivatives was synthesized and evaluated as activators of the cloned maxi-K channel mSlo (hSlo) expressed in Xenopus laevis oocytes using electrophysiological methods. A brain penetrable activator of maxi-K channels was identified and shown to be significantly active in the MCAO model of stroke.

Full text of DOI:10.1016/S0960-894X(02)00240-8

Bioorganic & Medicinal Chemistry Letters 12 (2002) 1779–1783
The Synthesis and Structure–Activity Relationships of
4-Aryl-3-aminoquinolin-2-ones: A New Class of
Calcium-Dependent, Large Conductance, Potassium (Maxi-K)
Channel Openers Targeted for Post-Stroke Neuroprotection
Piyasena Hewawasam,a,* Wenhong Fan,a Jay Knipe,c Sandra L. Moon,b
Christopher G. Boissard,b Valentin K. Gribkoffb and John E. Starrett, Jr.a
aDepartment of Chemistry, The Bristol-Myers Squibb Pharmaceutical Research Institute, 5 Research Parkway,
Wallingford, CT 06492, USA
bDepartment of Neuroscience, The Bristol-Myers Squibb Pharmaceutical Research Institute,
5 Research Parkway, Wallingford, CT 06492, USA
cDepartment of Pharmacokinetics and Metabolism, The Bristol-Myers Squibb Pharmaceutical Research Institute,
5 Research Parkway, Wallingford, CT 06492, USA
Received 11 February 2002; accepted 20 March 2002
Abstract—A series of 4-aryl-3-aminoquinoline-2-one derivatives was synthesized and evaluated as activators of the cloned maxi-K
channel mSlo (hSlo) expressed in Xenopus laevis oocytes using electrophysiological methods. A brain penetrable activator of maxi-K
channels was identified and shown to be significantly active in the MCAO model of stroke. # 2002 Elsevier Science Ltd. All rights
reserved.
Stroke is currently recognized as a major cause of adult
disability and death affecting more than 700,000 per
year in the United States alone and over 2 million
annually worldwide.1 Strokes are classified into two
types: ischemic stroke, which results from blockade of
arterial blood flow, and hemorrhagic stroke which is
caused by the rupture (aneurysm) of a blood vessel in
the brain. Acute ischemic stroke represents the most
common form with approximately 80% of all strokes.
Thrombolytic agents have been developed and utilized
to restore blood flow to ischemic neuronal tissue and
tissue plasminogen activator has been proven to be
effective in a limited patient population.2 In the last
decade, numerous neuroprotective mechanisms have
been examined including antagonists of AMPA/
kainate3À5 and N-methyl-d-aspartate (NMDA) excita-
tory amino acid receptors6 and inhibitors of neuronal
adenosine reuptake.7,8
However, to date value of these potential neuroprotec-
tive agents has not been realized due to either a lack of
demonstrated clinical efficacy or because of poor side-
effect profiles.9 Thus, post-stroke neuroprotection
represents one of the most prominent unmet medical
needs in the clinical arena. During ischemic stroke,
neurons at risk are exposed to abnormally high levels of
intracellular calcium (Ca2+), initiating a neurotoxic
cascade.10,11 To protect neurons at risk we have relied
on a strategy designed to reduce abnormally high levels
of Ca2+ entry, by the opening of large-conductance,
calcium-activated potassium (maxi-K) channels and
thereby reducing the excessive release of excitatory amino
acids12 and controlling neuronal hyperexcitability.13,14
Large-conductance, Ca2+-activated potassium (maxi-K
or BK) channels are present in a variety of excitable cell
types including neurons and smooth muscle cells15,16
and play a key role in regulating cell membrane poten-
tial and neuronal excitability.17 As a consequence,
modulators of maxi-K channels have emerged as
potentially useful therapeutic agents for various disease
states associated with both the central nervous system
*Corresponding author. Fax: +1-203-677-7702; e-mail: hewawasp@
bms.com
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00240-8

1780
P. Hewawasam et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1779–1783
and smooth muscle.18,19 Maxi-K channel opening sti-
mulated by Ca2+ in the presence of a pharmacological
opener would act to limit further Ca2+ entry, thus
interrupting the potentially fatal neurotoxic cascade
initiated by abnormally high Ca2+ entry, a circum-
stance encountered during neuronal ischemia.10,11,14,17
Therefore, a maxi-K opener would not be expected to
significantly affect channels in non-ischemic tissue in
which intracellular levels of Ca2+ are low thereby reducing
the potential for undesirable side effects.
As illustrated in Scheme 2, preparation of 4-aryl-3-ami-
noquinolin-2-ones required the 2-amino-benzophenones
(8) as precursors.
The synthesis of 2-aminobenzophenones by a variety
methods has been reviewed.33 The majority of 2-amino-
benzophenones (8) were prepared by the reaction of aryl
esters (7) with ortho-lithiated protected aniline deriva-
tives (6) via the formation of dianion species with tert-
BuLi (2.2 equiv) followed by deprotection of either Boc
(3 N HCl) or COtBu (6 N HCl) groups (Scheme 1). The
pattern of substitution of the 2-aminobenzophenones
prepared by this method was dependent upon the
regiospecificity of the directed metalation.
Electrophysiological and pharmacological properties of
a variety of synthetic and naturally occurring maxi-K
channel openers has recently been reviewed.18À20
Among small molecule maxi-K channel openers, the
benzimidazolone derivatives21À23 NS-004 (1) and NS-
1619 (2) have been studied in some detail, both in vitro
and in vivo.24À27 More recently, a series of 3-arylox-
indole derivatives with maxi-K channel opening activity
has been disclosed.14,28,29 The oxindole derivatives, 3a–b29
and (+)-4 (MaxiPost)14 identified from this series have
been evaluated in a rat model of stroke that involved
permanent occlusion of the middle cerebral artery
(MCAO model).
As shown in Scheme 2, chloroacetylation of 2-amino-
benzophenones (8) with chloroacetyl chloride in the
presence of pyridine gave the corresponding N-(chloro-
acetyl)-2-aminobenzophenone derivatives (9). Upon
heating a solution of 9 in anhydrous pyridine at reflux
for 15–30 min, the initially formed a-pyridinium salt
undergoes cyclodehydration to afford the N-[(4-arylqui-
nolin-2-one)-3-yl]pyridinium chloride (10). Hydrazino-
lysis of 10 with hydrazine hydrate in ethanol at reflux
for 1–2 h provided the desired 4-aryl-3-aminoquinolin-
2-ones 11. Finally, demethylation of the methyl ether
moiety of 11 with BBr3 afforded the desired phenols 12.
In order to mimic the 3-hydroxyl moiety present in 5,
As part of a broad-based effort directed towards the
identification of brain penetrable activators of neuronal
maxi-K channels that would be useful as neuroprotec-
tive agents, we synthesized a series of 4-aryl-3-amino-
quinolin-2-one derivatives. Recently, a series of 4-aryl-
3-hydroxyquinolin-2-ones (5)30 has been disclosed as
maxi-K channel openers, but these openers lack brain
penetrability mainly due to the presence of an ionizable
3-hydroxyl moiety.31 We reasoned that replacement of
the 3-hydroxyl moiety of 5 with non-ionizable amino
groups would enhance the brain penetrability. The
alkylsulfonamido group has been widely used as a
bioisosteric replacement for the hydroxyl group.32
Scheme 2. (a) ClCH2COCl, pyridine, CH2Cl2; 82–91%; (b) pyridine,
.
reflux; 88–98%; (c) NH2NH2 H2O, EtOH, reflux; 76–89%; (d)
R=SO2CF3; (1) (CF3SO2)2O, pyridine, 23 ꢀC; (2) NaOH, THF–
MeOH; 70%; (e) R=SO2Me; (1) MeSO2Cl, pyridine, 50–60 ꢀC; (2)
NaOH, THF–MeOH; 62%; (f) R=Ac; (1) Ac2O, pyridine, 60–70 ꢀC;
(2) NaOH, THF–MeOH; 82%; (g) BBr3, CH2Cl2, 0–23 ꢀC; 90–95%.
Scheme 1. (a) tBuLi (2.2 equiv), THF or ether, À78 to À40 ꢀC; (b) add
7 at À40 ꢀC then warm to 0 ꢀC; 45–82%; (c) 3–6 N HCl, EtOH, reflux;
93–97%.

P. Hewawasam et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1779–1783
1781
the 3-amino group was derivatized as both tri-
fluoromethyl and methyl sulfonamides. Selective mono-
sulfonylation of the amino group of 11 was not possible
even with stoichiometric amounts of the corresponding
sulfonylating agent. Alternatively, persulfonylation of
11 followed by selective deprotection with aqueous
NaOH in THF–MeOH led to the formation of desired
monosulfonylated 3-aminoquinolin-2-one derivatives
13. The acetamide derivative was also prepared by a
similar protocol using excess Ac2O in refluxing pyridine.
Finally, demethylation of the methyl ether moiety of 13
with BBr3 afforded the desired phenols 14. The 3-amino-
quinolin-2-one derivatives prepared by these methods are
compiled in Table 1 along with relevant physicochemical
data.
(except at C-8) markedly enhances channel opening
efficacy. Both 5-CF3 (12e) and 6-CF3 (12d) isomers were
superior to the 7-CF3 (12c) isomer. The 6-chloro ana-
logue, 12f was less efficacious when compared to the 6-
CF3 analogue, 12d. Removal of either the chlorine atom
or both chloro and hydroxyl groups from the 4-aryl
moiety of 12d results in a substantial loss of activity.
Thus, the presence of both a p-chlorophenol element
and an electron-withdrawing group is critical for main-
taining the maxi-K channel opening activity. Derivati-
zation of the 3-amino group of 12d and 12c as the
(trifluoromethyl)sulfonamides, 14a–b resulted in dra-
matic increase in maxi-K channel opening activity.
However, both the methylsulfonamide (14c) and the
acetamide (14d) derivatives were shown to be relatively
less active compared to the parent 3-aminoquinolone,
12d. This result when taken together with the effect of
CF3 moiety indicates an important relationship between
increasing acidity of the 3-amino moiety and enhanced
channel opening activity.
The ability of the target compounds to open maxi-K
channels and increase maxi-K-mediated whole-cell out-
ward K+-currents was assessed by using two-electrode
voltage clamp recording from Xenopus laevis oocytes
expressing cloned mSlo34 (or hSlo35) maxi-K channels,
as described previously.36 All compounds were tested in
at least five different oocytes to evaluate the effect of a
single drug concentration on channel current sensitive
to iberiotoxin (IbTx). The average percentage change in
mSlo (or hSlo) current relative to drug-free control
(100%) was determined for each compound tested. The
results obtained are listed in Table 1 along with data for
NS-004, which allows the efficacy comparison of the 4-
aryl-3-aminoquinolin-2-one derivatives to a prototypical
maxi-K channel opener.
As a prelude to evaluating compounds in animal models
of stroke, the ability of the selected maxi-K openers,
12c, 12d, and 14a, to enter rat brain following intrave-
nous administration was determined. Table 2 shows the
plasma and whole brain concentrations, as well as the
brain/plasma (B/P) ratios of each compound at 15 min
and 2 h after administration of an IV bolus dose (5 mg/
kg). The whole brain concentrations of 12c and 12d
Table 2. Whole brain and plasma concentrations of 12c, 12d, and 14a
after iv administration (5 mg/kg) to male rats
The structure–activity relationships presented in Table 1
provide a rudimentary understanding of the maxi-K
channel opening pharmacophore of 4-aryl-3-aminoqui-
nolin-2-ones. For this preliminary study, the optimal
substitution pattern of the quinoline nucleus was pro-
bed while restricting the 4-aryl moiety to the p-chloro-
phenol, an element present in prototype maxi-K openers
1 and 3–5. Compound 12a, which is unsubstituted on
the quinoline nucleus, was only poorly active. From the
results presented in Table 1, it appears that introduction
of an electron withdrawing substituent such as CF3
Compd
Plasma levela
0.25 h 2 h
Brain levelb
0.25 h 2 h
B/P ratio
0.25 h
2 h
12c
12d
14a
1166Æ145 339Æ14 1467Æ88 1480Æ75
802Æ92 281Æ10 1309Æ52 1493Æ110
1.3
1.6
ND
4.4
5.3
ND
1230Æ125 125Æ10
<50c
<50c
aConcentration as ng/mL; mean of three animalsÆstandard deviation.
bConcentration as ng/g wet weight; mean of three animalsÆstandard
deviation.
cValues below the detection limit of the assay (50 ng/g).
Table 1. Structure and physical properties of 3-amino-4-arylquinolin-2-one derivatives, effect on maxi-K-mediated outward current in Xenopus
laevis oocytes expressing the cloned maxi-K channels mSlo
Compd
R1
R2
R3
R4
R5
R6
R
Mp (ꢀC)a
% Increase in mSlo
current @ 20 mM
11a
11b
12a
12b
12c
12d
12e
12f
12g
14a
14b
14c
14d
1 (NS-004)
H
H
H
CF3
H
H
H
H
H
H
H
H
H
H
CF3
H
H
H
H
H
H
CF3
H
H
H
CF3
H
Cl
CF3
CF3
H
CF3
CF3
H
H
H
H
H
H
CF3
H
H
H
OMe
H
Cl
H
H
H
H
H
H
H
H
H
H
248–250
200–202
265–268
209–212
225–227
233–235
267–268
184–185
244–245
267–270
267–270
263–265
222–224
108Æ3
111Æ2
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
Cl
Cl
Cl
Cl
Cl
Cl
H
Cl
Cl
Cl
Cl
114Æ5
112Æ5
159Æ11
186Æ12
191Æ8
157Æ12
137Æ10
343Æ23
213Æ13
164Æ12
120Æ6
SO2CF3
SO2CF3
SO2Me
COCH3
H
H
H
CF3
H
H
H
H
H
132Æ13b
aAll new compounds exhibited spectroscopic and combustion data in accord with the designated structure.
bReference compound 1 (NS-004) shown to have identical effects on mSlo and hSlo-mediated maxi-K currents.36

1782
P. Hewawasam et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1779–1783
Table 3. Mean whole brain and plasma pharmacokinetic parameters
of 12d after iv administration (4.7 mg/kg) to male rats (N=3 at each
time point)
4-aryl-3-aminoquinolones are brain penetrable maxi-K
channel openers with significant neuroprotective prop-
erties in a rat MCAO model of stroke.
Parameter
Plasma
Brain
B/P ratiob
Cmax (ng/g)
Tmax (h)
AUCa
—
—
2704
7.6
1990Æ324
0.08
19,306
8.1
—
—
7.1
—
References and Notes
T1/2 (h)
1. Williams, G. R.; Jiang, J. G.; Matchar, D. B.; Samsa, G. P.
Stroke 1999, 30, 2523.
2. Fisher, M. J. Thromb. Thrombolysis 1999, 7, 165.
3. Pellegrini-Giampietro, D. E.; Pulsinelli, W. A.; Zukin, R. S.
J. Neurochem. 1994, 62, 1067.
aExpressed as ng-h/mL for plasma and ng-h/g for brain calculated by
the trapezoidal rule over 0–24 h after dosing.
bBased on ratio of AUC values.
4. Wilding, T. J.; Huettner, J. E. Mol. Pharmacol. 1996, 49, 540.
5. Goldberg, M. P.; Strasser, U.; Dugan, L. L. Int. Rev. Neu-
robiol. 1997, 40, 69.
6. Gagliardi, R. J. Arq. Neuropsiquiatr. 2000, 58, 583.
7. Miller, L. P.; Hsu, C. J. Neurotrauma. 1992, 9, 563.
8. Sweeney, M. I. Neurosci. Biobehav. Rev. 1997, 21, 207.
9. De Keyser, J.; Sulter, G.; Luiten, P. G. Trends Neurosci.
1999, 22, 535.
10. Choi, D. W. Trends Neurosci. 1995, 18, 58.
11. Kristian, T.; Siesjo, B. K. Stroke 1998, 29, 705.
12. Choi, D. W. J. Neurobiol. 1992, 23, 1261.
13. Dirnagl, U.; Iadecola, C.; Moskowitz, M. A. Trends Neu-
rosci. 1999, 22, 391.
14. Gribkoff, V. K.; Starrett, J. E., Jr.; Dworetzky, S. I.;
Hewawasam, P.; Boissard, C. G.; Cook, D. A.; Frantz, S. W.;
Heman, K.; Hibbard, J. R.; Huston, K.; Johnson, G.; Krish-
nan, B. S.; Kinney, G. G.; Lombardo, L. A.; Meanwell, N. A.;
Molinoff, P. B.; Myers, R. A.; Moon, S. L.; Ortiz, A.; Pajor,
L.; Pieschl, R. L.; Post-Munson, D. J.; Signor, L. J.; Srinivas,
N.; Taber, M. T.; Thalody, G.; Trojnacki, J. T.; Wiener, H.;
Yeleswaram, K.; Yeola, S. W. Nat. Med. 2001, 7, 471.
15. Latorre, R.; Oberhauser, A.; Labarca, P.; Alvarez, O.
Annu. Rev. Physiol. 1989, 51, 385.
16. Sah, P. Trends Neurosci. 1996, 19, 150.
17. Gribkoff, V. K.; Starrett, J. E.; Dworetzky, S. I. Adv.
Pharmacol. 1997, 37, 319.
18. Starrett, J. E.; Dworetzky, S. I.; Gribkoff, V. K. Curr.
Pharm. Des. 1996, 2, 413.
19. Shieh, C. C.; Coghlan, M.; Sullivan, J. P.; Gopa-
lakrishnan, M. Pharmacol. Rev. 2000, 52, 557.
20. Coghlan, M.; Gopalakrishnan, M.; Carrol, W. A. Annual
Reports in Med. Chem.; Academic: San Diego, 2001; Vol. 36,
p. 11.
21. Olesen, S.-P. Exp. Opin. Invest. Drugs 1994, 3, 1181.
22. Olesen, S.-P.; Munch, E.; Moldt, P.; Drejer, J. Eur. J.
Pharmacol. 1994, 251, 53.
were comparable at both time points and greatly excee-
ded those of 14a. The increase in absolute brain levels
and B/P ratio over 0.25–2 h suggested the accumulation
and possibly prolonged elimination of 12c and 12d from
brain tissue.
In order to fully explore the relationship between brain
and plasma levels of 12d over time, a larger group of
rats was intravenously dosed (iv bolus; 4.7 mg/kg) and
whole brain and plasma concentrations were deter-
mined through 24 h after dosing. The results from this
expanded study (Table 3) verified that 12d rapidly
entered rat brain and attained concentrations, which
exceeded those of plasma over a prolonged interval
(total brain/plasma AUC ratio of 7.1, evaluated over
24 h). Furthermore 12d was eliminated slightly more
slowly from rat brain than from plasma (elimination
half-lives of 8.1 and 7.6 h, respectively).
To determine the ability of 12d to reduce cell loss
resulting from neuronal ischemia, a standard rodent
model of permanent focal ischemia, involving occlusion
of the middle cerebral artery in the spontaneously
hypertensive rat (MCAO model) was employed.37 This
procedure results in a reliably large neocortical infarct
volume that is measured by means of vital dye exclusion
in serial slices through the brain 24 h after middle cere-
bral artery occlusion (MCAO). In this model, effects of
dose response and time course exposure of 12d on neo-
cortical infarct volume were examined. In the dose
response study, highest reduction (9%) in neocortical
infarct volume was observed with 0.001 mg/kg dose
when administered iv 2-h post-MCAO as compared to
vehicle-treated (2% DMSO, 98% propylene glycol)
control. In the time course exposure study with
0.001 mg/kg dose a significant reduction (14%, p <0.05)
in neocortical infarct volume was observed when admi-
nistered as a single IV bolus 30 min after MCAO as
compared to vehicle-treated control.
23. Olesen, S.-P.; Munch, E.; Watjen, F.; Drejer, J. NeuroRe-
port 1994, 5, 1001.
24. McKay, M. C.; Dworetzky, S. I.; Meanwell, N. A.; Ole-
sen, S.-P.; Reinhart, P. H.; Levitan, I. B.; Adelman, J. P.;
Gribkoff, V. K. J. Neurophysiol. 1994, 71, 1873.
25. Xu, X.; Tsai, T. D.; Wang, J.; Lee, E. W.; Lee, K. S. J.
Pharmacol. Exp. Ther. 1994, 271, 362.
26. Edwards, G.; Niederste-Hollenberg, A.; Schneider, J.;
Noack, T.; Weston, A. H. Br. J. Pharmacol. 1994, 113, 1538.
27. Lee, K.; Rowe, I. C.; Ashford, M. L. Eur. J. Pharmacol.
1995, 280, 215.
28. Hewawasam, P.; Meanwell, N. A.; Gribkoff, V. K.;
Dworetzky, S. I.; Boissard, C. G. Bioorg. Med. Chem. Lett.
1997, 7, 1255.
29. Hewawasam, P.; Erway, M.; Moon, S. L.; Knipe, J.;
Weiner, H.; Boissard, C. G.; Post-Munson, D. J.; Gao, Q.;
Huang, S.; Gribkoff, V. K.; Meanwell, N. A. J. Med Chem.
2002, 45, 1487.
In summary, we have identified a novel class of brain
penetrable maxi-K channel openers and demonstrated
that channel opening activity is sensitive to both the
nature and pattern of substitution of both aromatic
elements. The preliminary structure–activity data for
this series indicates the importance of both an electron-
withdrawing substituent on the quinolone nucleus and
the presence of a phenolic hydroxyl for effective expres-
sion of maxi-K channel opening properties. Based on in
vitro and in vivo profiling it has been demonstrated that
30. Sit, S.-Y.; Meanwell, N. A. US Patent 5,892,045 1999.

P. Hewawasam et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1779–1783
1783
31. The pKa value of 3-hydroxyl moiety of 5 is estimated to be in
the range of 8–9. For pKa measurements of 5 and related com-
pounds, please see: Masoud, M. S.; Mohammed, Y. S.; Abdel-
Latif, F. F.; Soliman, E. M. A. Spectrosc. Lett. 1988, 21, 369.
32. Patani, G. A.; LaVoie, E. J. Chem. Rev. 1996, 96, 3147.
33. Walsh, D. A. Synthesis 1980, 677.
35. Dworetzky, S. I.; Trojnacki, J. T.; Gribkoff, V. K. Mol.
Brain Res. 1994, 27, 189.
36. Gribkoff, V. K.; Lum-Ragan, J. T.; Boissard, C. G.; Post-
Munson, D. J.; Meanwell, N. A.; Starrett, J. E.; Kozlowski,
E. S.; Romine, J. L.; Trojnacki, J. T.; McKay, M. C.; Zhong,
J.; Dworetzky, S. I. Mol. Pharmacol. 1996, 50, 206.
37. Tamura, A.; Graham, D. I.; McCulloch, J.; Teasdale,
G. M. J. Cereb. Blood FlowMetab. 1981, 1, 53.
34. Butler, A.; Tsunoda, S.; McCobb, D. P.; Wei, A.; Salkoff,
L. Science 1993, 261, 221.