4-Aryl-3-(mercapto)quinolin-2-ones: Novel maxi-K channel opening relaxants of corporal smooth muscle

Article abstract of DOI:10.1016/j.bmcl.2004.06.051

A novel series of 4-aryl-3-(mercapto)quinolin-2-one derivatives was prepared and evaluated as openers of the cloned maxi-K channel hSlo expressed in Xenopus laevis oocytes by utilizing electrophysiological methods. The effect of these maxi-K openers on co

Full text of DOI:10.1016/j.bmcl.2004.06.051

Bioorganic & Medicinal Chemistry Letters 14 (2004) 4479–4482
4-Aryl-3-(mercapto)quinolin-2-ones: novel maxi-K channel
opening relaxants of corporal smooth muscle
a
b
b
Piyasena Hewawasam,a, Wenhong Fan, Deborah A. Cook, Kimberly S. Newberry,
*
Christopher G. Boissard,b Valentin K. Gribkoff,b John Starretta and Nicholas J. Lodgeb
aDepartment of Chemistry, The Bristol-Myers Squibb Pharmaceutical Research Institute, 5 Research Parkway,
Wallingford, CT 06492, USA
bNeuroscience/Genitourinary Drug Discovery, The Bristol-Myers Squibb Pharmaceutical Research Institute,
5 Research Parkway, Wallingford, CT 06492, USA
Received 6 May 2004; accepted 16 June 2004
Available online 17 July 2004
Abstract—A novel series of 4-aryl-3-(mercapto)quinolin-2-one derivatives was prepared and evaluated as openers of the cloned
maxi-K channel hSlo expressed in Xenopus laevis oocytes by utilizing electrophysiological methods. The effect of these maxi-K open-
ers on corporal smooth muscle was studied in vitro using isolated rabbit corpus cavernosum. In vivo efficacy has been demonstrated
with a selective maxi-K opening relaxant in a rat model of erectile function.
Ó 2004 Elsevier Ltd. All rights reserved.
Erectile dysfunction (ED) has been recognized as a com-
mon condition that afflict over 100 million men world-
wide.1 Clinically, ED is defined as the consistent
inability to achieve and maintain an erection adequate
for satisfactory sexual activity.1 Severity of dysfunction
can vary from moderate to complete ED depending on
the age group and other health conditions such as diabe-
tes, hypertension, and heart disease.2 A variety of topi-
cal and oral drug therapies have been developed to
treat ED during the past two decades.3,4 Sildenafil (Via-
gra) and other phosphodiesterase type 5 (PDE5) inhibi-
tors are effective oral agents for the treatment of male
ED. However, there continues to be a need for new
agents that have a reduced likelihood of cardiovascular
or other side effects,5,6 and that may treat patients
refractory to PDE5 inhibitors.
whereby lowering of cytosolic calcium (Ca2+) leads to
the relaxation of corporal smooth muscle.10 However,
in general the majority of these pathways depend upon
the accumulation of cyclic nucleotides cAMP and
cGMP or hyperpolarization of corporal smooth
muscle cells via activation of potassium channels.10
Maxi-K channels are the most prominent subtype
among the several different K+-channels that are present
in corporal smooth muscle cells.11,12 Therefore, activa-
tion of maxi-K channels represents an important and
attractive mechanism for controlling corporal smooth
muscle function.11–13 Opening of maxi-K+ channels
would result in membrane hyperpolarization, thereby
closing the voltage-gated Ca2+-channels with consequent
lowering in cytosolic Ca2+ and smooth muscle relaxa-
tion.
It is generally believed that impaired relaxation of the
corpus cavernosum smooth muscle is one of the major
causes of penile erectile dysfunction in a great majority
of impotent men.7–9 Thus, precise modulation of corpo-
ral smooth muscle function plays a important role in
developing effective and improved treatments for erectile
dysfunction. There are several mechanistic pathways
In pursuit of identifying orally bioavailable and water-
soluble activators of maxi-K channels that would be
useful in smooth muscle relaxation, we synthesized a ser-
ies of 4-aryl-3-(mercapto)quinolin-2-one derivatives as
shown in Scheme 1.
Prior to this work a series of 4-aryl-3-hydroxyquino-
lin-2-ones (1) has been disclosed as maxi-K channel
openers with antibacterial activity.14 A related series
of 4-aryl-3-aminoquinolin-2-ones (2) has been identi-
fied as brain penetrable maxi-K channel openers with
*
Corresponding author. Tel.: +1-203-677-7815; fax: +1-203-677-
7702; e-mail: hewawasp@bms.com
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.06.051

4480
P. Hewawasam et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4479–4482
Table 1. Water solubility of compounds 6, 7, 8a–i, effect on maxi-K
H
N
H
N
O
mediated outward current in Xenopus laevis oocytes expressing the
cloned maxi-K channels hSlo and relaxation of isolated rabbit corpus
cavernosum pre-contracted with phenylephrine
O
R1
XH
OR
F3C
(CH2)nOH
OR
Compd
Solubilitya
mg/mL
% hSlo current
@ 20lM
% Inhibition of
force @ 10lMb
R2
Cl
3, R = H, CH3
8a
6
<0.001
<0.001
231 23
Insol @ 20lM
73 7 @ 5lM
22 (n=2)
NT
1, X = O; R = H, CH3
2, X = NH; R = H, CH3
7
8b
0.003
0.004
0.003
0.194
0.092
0.097
>1.9
197
5
NT
43 10 (n=5)
23 5 (n=3)
Chart 1.
348 29
242 16
246 13
223 21
256 17
192 11
neuroprotective properties.15 More recently, a series of
4-aryl-3-(hydroxyalkyl)quinolin-2-ones (3) has been dis-
covered as novel maxi-K channel opening relaxants of
corporal smooth muscle16 (Chart 1).
8c
( )-8d
(S)-8e
(R)-8f
( )-8g
31 10 (n=6)
45 10 (n=6)
41 7 (n=5)
55 6 (n=12)
34 3 (n=8) @ 1lM
59 7 (n=5)
Acylation of the 2-aminobenzophenone derivative 416
with bromoacetyl bromide gave the corresponding
amide 5a. Demethylation of methyl ether moiety of 5a
with BBr3 provided the corresponding phenol 5b. Upon
reaction of 5b with tert-butyl mercaptan in the presence
of NaH (2.2equiv) in THF, initially formed tert-butyl-
thioacetamide intermediate undergoes cyclo-dehydra-
tion to afford the quinolinone 6. Treatment of 6 with
BBr3 in DCM afforded the desired 3-(mercapto)quinol-
inone 7. Selective S-alkylation of the 3-mercapto moiety
with respective alkylating agents17 in the presence of
NaH in THF gave the quinolinone derivatives 8a–i
(Scheme 1). The 4-aryl-3-(mercapto)quinolin-2-one
derivatives18 prepared by the methods illustrated in
Scheme 1 are listed in Table 1 along with solubility data.
(S)-8h
(R)-8i
>1.6
>1.6
173 9
33 5 (n=6) @ 1lM
45 7 (n=4)
162 11
132 13
30 6 (n=6) @ 1lM
31 3 (n=5)
NS-004
a Solubility in HPLC-grade water as measured by UV absorptivity
method.
b Percentage inhibition of isometric force in response to a test com-
pound in isolated rabbit corpus cavernosum pre-contracted with
phenylephrine (3lM). Vehicle produced approximately 13% relaxa-
tion.
The effect of the target compounds on whole-cell out-
ward K+-currents was determined by using two-elec-
trode voltage clamp recordings from Xenopus laevis
oocytes expressing cloned hSlo19 maxi-K channels, as
described previously.20 All compounds were tested in a
minimum of five different oocytes to evaluate the effect
of a single drug concentration (20lM) on outward
K+-current sensitive to iberiotoxin (IbTx). The average
percentage change in hSlo current relative to drug-free
control (100%) was determined for each compound
tested. The results obtained are presented in Table 1.
H
N
O
NH2
O
F3C
Br
F3C
F3C
F3C
O
OR1
OMe
a
Cl
Cl
5a; R1 = Me
4
b
c
5b; R1 = H
In this preliminary study a variation of 3-mercapto sub-
stitution was examined while restricting the quinoline
nucleus to 4-(5-chloro-2-hydroxyphenyl)-6-(trifluoro-
methyl)quinolin-2-one, an element identified from pro-
totype maxi-K openers 1–3. In order to identify more
soluble maxi-K openers polar groups such as alcohols
and aminoalcohols were incorporated into the 3-merca-
pto moiety. As can be seen from the data presented in
Table 1, all compounds including the parent 3-mercapto
derivative (7) were shown to be effective openers of
maxi-K channels. However, parent compound 7 and
both tert-butyl and methyl thioethers of 7 showed extre-
mely poor aqueous solubility. Hydroxyethyl derivative
8b was the most effective maxi-K opener identified from
this study. However, both hydroxyethyl and hydroxy-
propyl derivatives (8b,c) lacked adequate aqueous solu-
bility. Incorporation of additional hydroxyl groups (8d–
f) into the side chain somewhat improved the aqueous
solubility (<100lg/mL). Although replacement of termi-
nal hydroxyl moiety with a dialkylamino group slightly
diminished the maxi-K opening activity a significant
enhancement in aqueous solubility (>1mg/mL) was
H
N
H
N
O
O
SH
OH
F3C
StBu
OH
d
Cl
e
Cl
7
6
8a; R = CH3
8b; R = (CH2)2OH
8c; R = (CH2)3OH
H
N
O
8d; R = CH2(CHOH)CH2OH
8e; R = CH2(C*HOH)CH2OH
8f; R = CH2(C*HOH)CH2OH
8g; R = CH2(CHOH)NEt2
8h; R = CH2(C*HOH)NEt2
8i; R = CH2(C*HOH)NEt2
SR
OH
Cl
8a-i
Scheme 1. Reagents and conditions: (a) ClC(O)CH2Br, pyridine,
DCM, 0–23°C; 96%; (b) BBr3, DCM, À78–23°C; 91%; (c) tBuSH,
NaH, THF, 23–67°C; 87%; (d) BBr3, DCM, À78–0°C; 100%; (e) RX
(X=Br, Cl), NaH, THF, 23°C, 16h.

P. Hewawasam et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4479–4482
4481
observed. The racemic amino alcohol (8g) and individ-
ual enantiomers (8h,i) were equally effective openers
with comparable aqueous solubility as their free base
form. The different salt forms of these amino alcohols
exhibited 40–100 times higher aqueous solubility.
for the analysis of data with p<0.05 considered signifi-
cant.
Amino alcohol 8g (1mg/kg iv) produced a significant in-
crease in electrically-stimulated ICP (Fig. 1). The vehicle
PEG400 had no effect on ICP (n=6; not shown). MABP
increased following injection of 8g (Fig. 1). However, a
similar increase was observed in a separate group of ani-
mals given only vehicle (Fig. 1). The a1/a2 adrenergic re-
ceptor antagonist phentolamine was evaluated as a
comparitor; Vasomax (phentolamine mesylate) has been
shown to be effective in the treatment of erectile dys-
function.24 Phentolamine (0.03mg/kg iv) did not affect
electrically-stimulated ICP but produced a 13% fall in
MABP (n=6). A higher dose of phentolamine (0.3mg/
kg iv) produced a significant increase in ICP, but only
at the 65min post-dose time-point (Fig. 2). MABP in
these animals decreased by 32% and then recovered
slowly (Fig. 2). Saline did not affect blood pressure (sep-
arate group of animals; Fig. 2) or electrically-evoked
ICP (n=6; not shown). Thus, relative to phentolamine,
8g produced an increase in ICP with a less effect on
blood pressure.
The functional effects of maxi-K openers on smooth
muscle relaxation were evaluated using a rabbit corpus
cavernosum tissue strip assay.21 Corpus cavernosum
strips were isolated from the rabbit penis and suspended
in tissue baths containing warm (37°C) physiological
salt solution. Isometric force was measured using stand-
ard methods. Tissue strips were stimulated with the
a-agonist, phenylephrine (3lM) and allowed to reach
a steady level of force prior to the addition of test com-
pounds. The results are expressed as the percentage inhi-
bition of phenylephrine-induced force as compared to
vehicle control (Table 1).
As shown in Table 1, a majority of maxi-K openers
produced greater than a 30% reduction of contractile
force at concentration of 10lM. In direct comparison,
a prototype maxi-K opener, NS-004 showed 31% reduc-
tion at 10lM. In particular, racemic amino alcohol 8f
inhibited the contractile response by over 50% at
10lM. The (S)-enantiomer 8h was slightly more effec-
tive than (R)-enantiomer 8i in rabbit corporal tissue
assay. However, despite their robust maxi-K channel
opening activity some of the compounds displayed poor
inhibition of the contractile response. For an example,
maxi-K openers (8a,c) were shown to be less effective
as relaxants. Thus, a weak correlation between maxi-K
channel opening activity and the ability to relax phenyl-
ephrine-induced contractions in rabbit corporal tissue
strip assay is observed. It is, therefore, plausible that
other mechanisms contribute to the in vitro relaxant re-
sponses.
200
180
160
140
120
100
80
80
Treatment
70
60
50
40
30
20
10
0
*
*
*
*
*
60
20
35
50
65
Mean 5
Control
Stim
Time (minutes)
Figure 1. Effect of 8g (1mg/kg, iv) on intracavernous pressure (ICP)
response elicited by electrical stimulation of the cavernous nerve at
various time points following compound administration. ICP is
indicated by the bars (n=6). Ã indicates p<0.05. Blood pressure in
The in vivo effects of racemic amino alcohol 8g were
evaluated using a rat model of erectile function as previ-
ously described in the literature.22 Male Fischer 344 rats
(280–350g) were anesthetized with sodium pentobar-
bital. One carotid artery was cannulated to allow the
measurement of mean arterial blood pressure (MABP).
Both jugular veins were also cannulated for the admin-
istration of test compound as well as a constant infusion
of sodium pentobarbital to maintain the plane of anes-
thesia. The trachea was cannulated to allow for artificial
ventilation. Body temperature was maintained at
37 0.5°C using a heating blanket connected to a rectal
probe thermistor. The right corpora cavernosa was
exposed and cannulated for the measurement of intra-
cavernous pressure (ICP). The cavernous nerve was iso-
lated and a stainless steel bipolar electrode was placed
around the nerve to allow for electrical stimulation.
MABP and ICP were continuously monitored. The cav-
ernous nerve was electrically stimulated at a frequency
of 20Hz (typically 0.3mA pulses of 0.22ms duration)
for a period for 30s. A second control stimulation was
given 15min after the first and then either vehicle
(PEG 400)23 or compound was administered iv.
Repeated stimulations were performed at 5, 20, 35, 50,
and 65min post treatment. StudentÕs t-test was used
response to 8g (-d-) (n=7) and PEG 400 vehicle (- -) (n=6).
ꢀ
70
200
60
180
Treatment
*
50
40
160
140
120
100
80
30
20
10
0
0
1
2
4
60
Mean
5
20
35
50
65
Control
Time (minutes)
Stim
Figure 2. Effect of phentolamine on intracavernous pressure (ICP)
response elicited by electrical stimulation of the cavernous nerve at
various time points following compound administration. ICP is
indicated by the bars (n=6). Ã indicates p<0.05. Blood pressure in
response to phentolamine (-d-) (n=6) and saline vehicle (- -) (n=6).
ꢀ

4482
P. Hewawasam et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4479–4482
13. Christ, G. J.; Brink, P. R.; Melman, A.; Spray, D. C. Int.
J. Impot. Res. 1993, 5, 77.
14. Sit, S.-Y.; Meanwell, N. A. U.S. Patent 5,892,045, 1999.
15. Hewawasam, P.; Fan, W.; Knipe, J.; Moon, S. L.;
Boissard, C. G.; Gribkoff, V. K.; Starrett, J. E., Jr.
Bioorg. Med. Chem. Lett. 2002, 12, 1779.
16. Hewawasam, P.; Fan, W.; Ding, M.; Flint, K.; Cook, D.;
Goggins, G. D.; Myers, R. A.; Gribkoff, V. K.; Boissard,
C. G.; Dworetzky, S. I.; Starrett, J. E., Jr.; Lodge, N. J.
J. Med. Chem. 2003, 46, 2819.
In summary, we have identified a novel series of effective
maxi-K channel openers and demonstrated that channel
opening activity can be retained or enhanced by deriva-
tizing the 3-mercapto moiety. A great majority of these
maxi-K openers were shown to be effective relaxants of
pre-contracted rabbit corpus cavernosal strips in vitro.
Introduction of amino alcohol side chain imparts excel-
lent water solubility to these maxi-K opening relaxants.
Furthermore, we have demonstrated the in vivo efficacy
of these maxi-K openers in a rat model of erectile func-
tion. In conclusion, we have identified a series of water-
soluble maxi-K channel opening relaxants of corporal
smooth muscle that offer potential as agents for the
treatment of male ED.
17. Alkylating agents used for the S-alkylation of 3-mercapto
group are as follows: (i) CH3Br; (ii) Br(CH2)nOH (n=2
and 3); (iii) ( )-3-bromo-1,2-propanediol; (iv) (S)-(+)-3-
chloro-1,2-propanediol and (R)-(À)-3-chloro-1,2-propane-
diol; reaction of both racemic and enantiomeric epi-
chlorohydrin with diethylamine provided the requisite
alkylating agents, 1-chloro-3-(diethylamino)-2-propanol
derivatives.
Acknowledgements
18. All new compounds were characterized by 1H NMR,
LRMS, and elemental analysis.
19. Dworetzky, S. I.; Trojnacki, J. T.; Gribkoff, V. K. Mol.
Brain Res. 1994, 27, 189.
We thank Gregory D. Goggins for technical support.
20. Gribkoff, V. K.; Lum-Ragan, J. T.; Boissard, C. G.; Post-
Munson, D. J.; Meanwell, N. A.; Starrett, J. E.; Kozlow-
ski, E. S.; Romine, J. L.; Trojnacki, J. T.; McKay, M. C.;
Zhong, J.; Dworetzky, S. I. Mol. Pharmacol. 1996, 50, 206.
21. New Zealand white rabbits (3.4–3.7kg) were euthanized
by sodium pentobarbital (100mg/kg) overdose. The penis
was collected from each animal, the outer tunica was
removed and the corpus cavernosum smooth muscle cut
into strips 3–5mm in length. The strips were suspended in
tissue baths containing a modified Krebs Henseleit bicar-
bonate buffer (37°C; pH7.4) gassed with 95%O2/5%CO2.
The strips were allowed to equilibrate for at least one hour
at a resting tension of 1.0g. Tension was measured using a
Grass FT03 force transducer coupled to a BIOPAC Data
Acquisition system and AcqKnowledge analysis software.
22. Rehman, J.; Chenven, E.; Brink, P.; Peterson, B.; Walcott,
B.; Wen, Y. P.; Melman, A.; Christ, G. Am. J. Physiol.
1997, 272, H1960–H1971.
References and notes
1. NIH Consensus Conference. Impotence. NIH Consensus
Development Panel on Impotence. J. Am. Med. Assoc.
1993, 270, 83.
2. Feldman, H. A.; Goldstein, I.; Hatzichristou, D. G.;
Krane, R. J.; McKinlay, J. B. J. Urol. 1994, 151, 54.
3. Garcia-Reboll, L.; Mulhall, J. P.; Goldstein, I. Drugs
Aging 1997, 11, 140.
4. Lue, T. F. J. Formos. Med. Assoc. 1999, 98, 233.
5. Kloner, R. A. Am. J. Cardiol. 2000, 86, 57.
6. Vobig, M. A.; Klotz, T.; Staak, M.; Bartz-Schmidt, K. U.;
Engelmann, U.; Walter, P. Lancet 1999, 353, 375.
7. Carrier, S.; Brock, G.; Kour, N. W.; Lue, T. F. Urology
1993, 42, 468.
8. Andersson, K. E.; Wagner, G. Physiol. Rev. 1995, 75,
191.
23. Despite the aqueous solubility of some of the compounds
in this study, the aqueous solubility of 8g was insufficient
to allow use of a water vehicle in the in vivo studies. PEG
400 was found to be the best vehicle for in vivo animal
studies.
9. Christ, G. J. Urol. Clin. North Am. 1995, 22, 727.
10. Saenz de Tejada, I. Int. J. Impot. Res. 2000, 12(Suppl 4),
34.
11. Christ, G. J.; Spray, D. C.; Brink, P. R. J. Androl. 1993,
14, 319.
24. Becker, A. J.; Stief, C. G.; Machtens, S.; Schultheiss, D.;
Hartmann, U.; Truss, M. J. Urol. 1998, 159, 1214.
12. Fan, S. F.; Brink, P. R.; Melman, A.; Christ, G. J. J. Urol.
1995, 153, 818.