Synthesis and bladder smooth muscle relaxing properties of substituted 3-amino-4-aryl-(and aralkyl-)cyclobut-3-ene-1,2-diones

Article abstract of DOI:10.1016/j.bmcl.2005.03.073

We have reported on the design, synthesis, and biological characterization of (R)-4-[3,4-dioxo-2-(1,2,2-trimethyl-propylamino)-cyclobut-1-enylamino]-3- ethyl-benzonitrile (1),¹ a novel, potent, and selective adenosine 5′-triphosphate-sensitive

Full text of DOI:10.1016/j.bmcl.2005.03.073

Bioorganic & Medicinal Chemistry Letters 15 (2005) 2495–2501
Synthesis and bladder smooth muscle relaxing properties of
substituted 3-amino-4-aryl-(and aralkyl-)cyclobut-3-ene-1,2-diones
John A. Butera,a,* Douglas J. Jenkins,a Joseph R. Lennox,a Jeffrey H. Sheldon,b
N. Wesley Norton,b Dawn Wargab and Thomas M. Argentierib
aChemical and Screening Sciences, Wyeth Research, CN 8000, Princeton, NJ 08543, USA
bUrologic Diseases, Womenꢀs Health Research Institute, Wyeth Research, 500 Arcola Road, Collegeville, PA 19426, USA
Received 10 January 2005; revised 17 March 2005; accepted 17 March 2005
Available online 12 April 2005
Abstract—We have reported on the design, synthesis, and biological characterization of (R)-4-[3,4-dioxo-2-(1,2,2-trimethyl-propyl-
amino)-cyclobut-1-enylamino]-3-ethyl-benzonitrile (1),1 a novel, potent, and selective adenosine 50-triphosphate-sensitive potassium
(KATP) channel opener with potential utility for the treatment of urge urinary incontinence (UUI). Excising the aniline-derived
nitrogen atom of 1 or replacing it with an aralkyl group, led to bladder smooth muscle relaxant chemotypes 3 and 4, respectively.
Prototype compounds in these series were found to produce significant increases in an iberiotoxin (IbTx)-sensitive hyperpolarizing
current, thus suggesting that these relatively modest structural modifications resulted in a switch in the mechanism of action of these
smooth muscle relaxants from KATP channel openers to activators of the large-conductance Ca2+-activated potassium channel
(BKCa). We report herein the syntheses and biological evaluation of a series of substituted 3-amino-4-aryl-(and aralkyl-)cyclo-
but-3-ene-1,2-diones.
Ó 2005 Elsevier Ltd. All rights reserved.
Transmembrane potassium channels, a diverse family of
pore-forming proteins that allow for selective perme-
ation of potassium ions across the cell membrane, con-
tinue to provide useful molecular targets for
controlling the physiological function of excitable cells.
Recent reviews have provided useful perspectives of
potassium channel sub-type, structure, function and re-
lated medicinal chemistry, and SAR.2–5 It has been
shown that bladder smooth muscle may contain several
sub-type families of potassium channels and that modu-
lation of these channels with specific ligands produces
significant detrusor muscle relaxation in vitro and in
vivo.6 Potassium channel openers (KCOs) with specifi-
city towards bladder tissue may show promise as poten-
tial drugs to treat urge urinary incontinence (UUI), a
condition characterized by spontaneously hyperactive
bladder smooth muscle.
tent and bladder-selective KATP-channel openers pos-
sessing striking oral efficacy in a rat hypertrophied
bladder model of UUI,1,7 metabolic stability studies
(in vitro and in vivo) on 1a and 1b suggested a rapid
turnover to the primary metabolite, a 4-amino-benzonit-
rile derivative. Concern over potential long-term expo-
sure to these compounds prompted us to focus our
synthetic efforts on structural modifications of the aryl
vinylogous amide bond. Numerous approaches were
implemented and several have been reported: (i) homolo-
gation of the anilino-cyclobutenedione linker with con-
comitant re-optimization of SAR to afford development
track compound 2 (KCO-616),8,9 (ii) acylation of the
anilino-nitrogen to afford acyl analogs 5,10 which slows
but does not prevent formation of the aniline metabo-
lite, and (iii) incorporation of the aryl amine and alkyl
amine moieties onto numerous heterocyclic scaffolds
(6) to circumvent the metabolic cleavage.11 While ap-
proach (i) was successful and resulted in the identifica-
tion, characterization, and development of a series of
nonhydrolyzing benzylamino cyclobutenediones exem-
plified by KCO-616, concomitant strategies involved
the excision of the labile vinylogous amide bond or its
replacement with a carbon-based variant such as an ole-
fin, methylene, or substituted methylene unit. Interest-
ingly, preliminary cell-electrophysiological data suggest
Subsequent to the discovery and characterization of
diaminocyclobutenediones 1a and 1b (Fig. 1), two po-
Keywords: Bladder instability; Urge urinary incontinence; Potassium
channel opener; BKCa channel opener; Bladder relaxant.
*
Corresponding author. Tel.: +1 732 274 4289; fax: +1 732 274
4129; e-mail: buteraj@wyeth.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.03.073

2496
J. A. Butera et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2495–2501
Heterocyclic
R1
R2
Scaffold
N
H
N
H
6
O
O
O
O
O
O
NC
Cl
O
R1
R2
N
N
N
Acyl
5
N
N
H
N
H
H
H
H
R
Cl
CH3
2
1a R = H
1b R = Et
O
O
Z
O
R1
N
H
R1
R2
N
R2
H
Z=CH2, CH=CH, C(C=O)R6
3
4
Figure 1. Attempts to modulate the metabolic fate of cyclobutenedione leads (1a and 1b) via structural modification of the aryl vinylogous amide
bond led to several novel chemotypes possessing in vitro and in vivo bladder smooth muscle relaxant properties.
that these seemingly modest structural modifications re-
sulted in a switch in the mechanism of action of these
smooth muscle relaxants from KATP channel openers
to activators of the large-conductance Ca2+-activated
potassium channel (BKCa). This report will detail our ef-
forts to synthesize a series of substituted 3-amino-4-aryl,
3-amino-4-aralkyl, and 3-amino-4-vinyl cyclobut-3-ene-
1,2-diones, represented by templates 3 and 4, and their
biological evaluation as bladder smooth muscle
relaxants.12
Synthesis of benzyl analogs (9) was accomplished simi-
larly by treating stannane 7 with an appropriate benzyl
bromide in the presence of benzyl(chloro)bis-(triphenyl-
phosphine)palladium(II), cuprous iodide, under trans-
halogenation conditions using sodium iodide to afford
the benzyl-substituted cyclobutenedione, which was
converted to final product 9 by treatment with a primary
or secondary amine in ethanol.
Compounds possessing an alkanoyl group appended
from the benzylic carbon of chemotype 9 were prepared
(Scheme 2) by inverse dropwise addition of the pre-
formed enolate of the appropriate ketone 10 (potassium
bis(trimethylsilyl)amide at ꢀ78 °C) into a cold solution
of diethoxysquaric acid. Adduct 11 was then treated
with one equivalent of amine in EtOH at 0 °C to afford
compound 12.
Cross-coupling reaction conditions (Scheme 1) using 3-
isopropoxy-4-tri-n-butyl-stannyl)-3-cyclobutene-1,2-dione
(7)13 and an appropriate aryl halide (X = CH) or 3-pyri-
dyl halide (X = N) in the presence of benzyl(chloro)bis-
(triphenylphosphine)palladium(II) and cuprous iodide
afforded the Stille product,14 which was subsequently
treated with a primary or secondary amine in ethanol
to give the 3-amino-4-phenyl-(or pyridyl)-cyclobut-3-
ene-1,2-dione (8).
A third linker was explored where the aniline nitrogen of
1 was replaced with a C@C double bond. Thus, iodo-
I
O
O
R1
(Ph3P)2Pd(CH2Ph)Cl
CuI DMF / 25ºC
O
O
1)
2)
X
R1
R2
R2
R3
R3
N
Bu3Sn
O
R4
EtOH / 70ºC
X
HN
R4
7
8
R1
1)
2)
Br
R2
O
O
R1
R2
(Ph3P)2Pd(CH2Ph)Cl
R3
N
CuI NaI DMF
R3
R4
EtOH / 70ºC
9
HN
R4
Scheme 1. Synthesis of 3-amino-4-phenyl- and 3-amino-4-pyridyl-cyclobut-3-ene-1,2-diones 8 and benzyl analogs 9.

J. A. Butera et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2495–2501
2497
R3
R4
KN(TMS)2
R1
R2
O
O
O
O
HN
O
THF / Et2O -78 ºC
RT
R1
R2
R1
R2
R3
R6
OEt
N
O
O
EtOH / 0ºC RT
R4
10
O
R6
11
O
R6
12
EtO
OEt
Scheme 2. Synthesis of acyl benzyl analogs 12.
O
O
(Ph3P)2Pd(CH2Ph)Cl
CuI DMF / 25ºC
R1
R2
CrCl2 CHI3 THF
RT, Dark
R1
R2
O
R1
R2
H
I
7
O
13
14
15
O
O
R3
R4
HN
R3
N
R1
R2
R4
EtOH / ∆
16
Scheme 3. Synthesis of styrenylcyclobutenedione analogs 16.
methylation of aldehyde 13 (Scheme 3) in the presence
of iodoform and chromium(II) chloride at rt (in the
absence of light)15 gave the corresponding vinyl iodide
14, which was reacted with stannane 7 in the presence
of benzyl(chloro)bis-(triphenylphosphine)-palladium(II)
and cuprous iodide in DMF to afford styrenylcyclo-
butenedione derivative 15. Displacement of the isoprop-
oxy group with an amine in EtOH at reflux gave
vinylogous amide 16.
was still ꢁ25-fold less potent than 1a. The in vitro relax-
ant effects of both 8.1 and 8.2 were found to be reversed
by iberiotoxin, a selective blocker of the large conduc-
tance, Ca2+-activated potassium channel (BKCa). An-
other distinguishing feature of compound 8.2 was its
ꢁ2.6-fold selectivity for bladder tissue versus aortic tis-
sue in contrast to the diaminocyclobutenediones, which
trended toward aortic selectivity in vitro.
Our initial survey of linker replacements for the anilino
NH of 1 was widened to examine CH2 (9.1 and 9.2),
CH(COCH3) (12.1 and 12.2), and a trans-olefin (16.1
and 16.2). These changes resulted in loss of activity
when compared to 1a except in the case of 16.1
(IC50 = 2.94 lM). Due to their interesting smooth mus-
cle relaxant properties, their unique structure, and their
amenability to synthesis, we chose to focus our efforts
on generating a series of substituted phenyl cyclobutene-
diones and evaluating them as bladder smooth muscle
relaxants. The data are shown in Table 2.
Our initial biological evaluation utilized a functional as-
say, which measured the compoundꢀs ability to relax
bladder smooth muscle. In vitro bladder smooth muscle
relaxant properties were determined for all test com-
pounds in tissue baths utilizing KCl pre-contracted rat
detrusor muscle strips.17 The data for prototype com-
pounds examining linker variants (Z = direct bond,
CH2, CHCOCH3, and CH@CH) are shown in Table 1.
We began our assessment of potential linker replace-
ments for the anilino NH of squarate leads 1 by holding
the alkyl amino group constant as the 1,2,2-trimethyl-
propyl moiety, the optimal substituent identified from
our previously reported SAR optimization effort in this
series.1 Direct excision of the anilino NH in 1a, resulted
in a drastic reduction (ꢁ100-fold) in smooth muscle
relaxant properties as seen when comparing 1a with
8.1. Interestingly, in contrast with diaminocyclobutene-
diones 1a, 1b, and 2, the relaxation resulting from treat-
ment with phenyl cyclobutenedione 8.1 was not reversed
by the sulfonylurea glyburide, a specific blocker of the
KATP channel. This suggested that the smooth muscle
relaxation was probably due to an alternative mecha-
nism of action. Replacement of the 4-cyano moiety of
8.1 with a 4-methoxy group (8.2) reinstated smooth
muscle relaxant properties although the compound
Studying the effects of a-gem-disubstitution on the
alkylamino side, the 1,2,2-trimethylpropyl-amino group
in 8.2 was replaced with a 1,1-dimethyl-2-phenyl ethyl
moiety to afford 8.3, a compound with 8-fold improve-
ment in bladder relaxant properties, which still main-
tains its ꢁ2.6-fold selectivity toward bladder tissue
versus aortic rings. Compound 8.10, another gem-disub-
stituted analog, was found to exhibit the most selectivity
for bladder tissue with an IC50 ratio of ꢁ15. Again prob-
ing the effects of exchanging the 4-methoxy group, the 4-
CF3 and 4-Br analogs (8.4 and 8.5, respectively) were
found to be much less potent than 8.2, 8.3, and 8.10.
Moving the methoxy substituent from the 4- to the 3-po-
sition (8.9 and 8.11) afforded molecules devoid of blad-
der relaxant properties. In fact, compound 8.9 further

2498
J. A. Butera et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2495–2501
Table 1. In vitro effects of linker (Z) variants of KATP channel openers 1 and 2 on pre-contracted rat bladder detrusor muscle strips and aortic rings
(IC50, lM)
O
O
R1
R2
R3
Z
N
H
Compound
Number
Z
R1
R2
R3
IC50 bladder nb % Reversed by IC50 aorta
nb
(lM)a
glyburidec
(lM)d
1a
1b
NH
NH
4-CN
4-CN
2,4-Di-Cl 6-Me 1,1-Dimethyl-1-propyl
H
(R) 1,2,2-Trimethylpropyl
(R) 1,2,2-Trimethylpropyl
0.37 0.07
0.09 0.02
0.1 0.03
4
5
4
4
4
3
2
2
3
5
3
110
77
100
0
0.064 0.015
0.017 0.004
0.042 0.014
NTe
3
2-Et
2
2
CH2–NH
f
—
—
3
8.1
8.2
9.1
9.2
12.1
12.2
16.1
16.2
4-CN
H
H
H
H
H
H
H
H
(R,S) 1,2,2-Trimethylpropyl >30
(R,S) 1,2,2-Trimethylpropyl 9.5 1.8
(R,S) 1,2,2-Trimethylpropyl >30
(R,S) 1,2,2-Trimethylpropyl >30
(R,S) 1,2,2-Trimethylpropyl >30
(R,S) 1,2,2-Trimethylpropyl 16.2 6.82
(R,S) 1,2,2-Trimethylpropyl 2.94 0.62
(R,S) 1,2,2-Trimethylpropyl >30
—
5
4-OCH3
4-CN
4-OCH3
0
24.3 6.3
NT
NT
CH2
CH2
0
—
—
—
—
—
—
0
CH(COCH3) 4-OCH3
CH(COCH3) 4-CF3
0
NT
NT
0
CH@CH
CH@CH
4-CN
4-Br
0
25
NT
NT
a Drug concentration SE (lM) that relaxed KCl-induced contractions in rat detrusor strips by 50%.
b Number of experiments.
c Percent of the observed relaxation that was reversed by the addition of the glyburide, a selective KATP channel blocker.
d Drug concentrations SE (lM) that relaxed KCl-induced contractions in rat aortic rings by 50%.
e Not tested.
f Direct bond.
Table 2. In vitro effects of phenyl cyclobutenedione derivatives 8 on pre-contracted rat bladder detrusor muscle strips and aortic rings (IC50, lM)
O
O
R3
N
R1
R4
X
Compound
number
X
R1
R3
R4
IC50 bladder nb % Reversal by IC50 aorta
nb
(lM)a
glyburidec
(lM)d
8.1
8.2
CH 4-CN
CH 4-OCH3
CH 4-OCH3
CH 4-CF3
CH 4-Br
(R,S) 1,2,2-Trimethylpropyl
(R,S) 1,2,2-Trimethylpropyl
1,1-Dimethyl-2-phenyl-ethyl
(R,S) 1,2,2-Trimethylpropyl
(R,S) 1,2,2-Trimethylpropyl
1,1-Dimethyl-1-propyl
H
H
H
H
H
H
H
H
H
H
H
H
H
>30
4
4
2
7
3
2
2
4
2
4
2
2
2
4
0
NTe
—
5
9.5 1.8
1.2 0.03
>30
0
24.3 6.3
3.25 0.35
NT
8.3
8.4
0
2
0
—
—
—
—
—
—
2
8.5
8.6
25.7 1.1
>30
0
NT
NT
N
H
0
8.7
8.8
CH 4-OCH3
CH 4-OCH3
CH 3-OCH3
CH 4-OCH3
CH 3-OCH3
CH 4-OCH3
(R)-1-Phenyl-ethyl
27 0.3
>30
Contractedf
2.5 0.49
>30
0
NT
NT
2-(3-Trifluoromethyl-phenyl)-ethyl
1,1-Dimethyl-1-propyl
1,1-Dimethyl-1-propyl
1,1-Dimethyl-1-propyl
2-Phenyl-1-propyl
0
8.9
—
0
NT
8.10
8.11
8.12
8.13
8.14
37.65 6.05
NT
NT
0
—
—
—
3
>30
Contractedf
0
CH 4-(3-Phenylpropoxy) 1,1-Dimethyl-1-propyl
CH 4-OCH3 2-Propyl
—
0
NT
46.2 7.8
CH3 11.99 2.4
a Drug concentration SE (lM) that relaxed KCl-induced contractions in rat detrusor strips by 50%.
b Number of experiments.
c Percent of the observed relaxation that was reversed by the addition of the glyburide, a selective KATP channel blocker.
d Drug concentrations SE (lM) that relaxed KCl-induced contractions in rat aortic rings by 50%.
e Not tested.
f Drug caused an additional contraction of rat detrusor strips beyond that seen after treatment with KCl.
contracted the bladder strips. The underlying mecha-
nism for this observation is not well understood; how-
ever the data could suggest that 8.9 is a potassium
channel blocker. Replacement of the substituted phenyl
group with a 3-pyridyl moiety (8.6) also resulted in loss
of in vitro bladder relaxant activity.
Compounds bearing a phenethyl group, which lack the
a-gem disubstitution (i.e., 8.8 and 8.12) and those bear-
ing an a-methyl benzyl group (8.7) demonstrated a loss
of activity. Also, as suggested by 8.14, methylation of
the vinylogous amide N is not tolerated. Finally, homolo-
gation of the 4-methoxy group of 8.10 to a larger

J. A. Butera et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2495–2501
2499
substituent such as a 3-phenyl-propoxy moiety (8.13)
also converts the compound into a smooth muscle
contractor.
larizing current consistent with activation of the large
conductance Ca2+-activated potassium channel (BKCa).
Additional studies to better understand the intrinsic
bladder selectivity of these derivatives and to more fully
evaluate their potential for use in the treatment of urge
urinary incontinence are underway.
The compound with the highest measured intrinsic selec-
tivity for bladder tissue, 3-(1,1-dimethyl-propylamino)-
4-(4-methoxy-phenyl)-cyclobut-3-ene-1,2-dione (8.10),
was selected for further evaluation in electrophysiologi-
cal studies to confirm the underlying mechanism of ac-
tion of its observed smooth muscle relaxing properties.
Voltage-clamp studies (Fig. 2) on isolated human bron-
chial smooth muscle cells18 showed that compound 8.10
at a concentration of 1 lM increased outward current
above +40 mV. This increase in outward current was re-
versed back to control levels by addition of the selective
BKCa channel blocker, iberiotoxin.
References and notes
1. Butera, J. A.; Antane, M. M.; Antane, S. A.; Argentieri, T.
M.; Freeden, C.; Graceffa, R. F.; Hirth, B. H.; Jenkins, D.;
Lennox, J. R.; Matelan, E.; Norton, N. W.; Quagliato, D.;
Sheldon, J. H.; Spinelli, W.; Warga, D.; Wojdan, A.;
Woods, M. J. Med. Chem. 2000, 43(6), 1187.
2. Coghlan, W. J.; Caroll, W. A.; Gopalakrishnan, M. J.
Med. Chem. 2001, 44(11), 1627.
3. Gribkoff, V. K.; Starrett, J. E. In Ann. Rep. Med. Chem.;
Doherty, A. M., Ed.; 2002; Vol. 37, pp 237–246.
4. Aguilar-Bryan, L.; Clement, J. P.; Gonzalez, G.; Kunjil-
war, K.; Babenko, A.; Bryan, J. Physiol. Rev. 1998, 78,
227.
The intrinsic in vitro selectivity for bladder tissue ob-
served with this novel class of KCOs provided a compel-
ling reason to expand the SAR. A more focused
electrophysiological evaluation of these compounds on
various isolated smooth muscle cell types may be re-
quired to better understand this measured selectivity.
5. Bryan, J.; Aguilar-Bryan, L. Curr. Opin. Cell. Biol. 1997,
9, 553.
6. (a) Zografos, P.; Li, J. H.; Kau, S. T. Pharmacology
(Basel) 1992, 45, 216; (b) Bonev, A. D.; Nelson, M. T.
Am. J. Physiol. 1993, 264, C1190; (c) Malmgren, A.;
Andersson, K. E.; Sjogren, C.; Andersson, P. O. J. Urol.
1989, 142, 1134; (d) Nurse, D. E.; Restorick, J. M.;
Mundy, A. R. Br. J. Urol. 1991, 68, 27; (e) Howe, B. B.;
Halterman, T. J.; Yochim, C. L.; Do, M. L.; Pettinger, S.
J.; Stow, R. B.; Ohnmacht, C. J.; Russell, K.; Empfield, J.
R.; Trainor, D. A.; Brown, F. J.; Kau, S. T. J. Pharmacol.
Exp. Ther. 1995, 274, 884.
7. Wojdan, A.; Freeden, C.; Woods, M.; Oshiro, G.; Spinelli,
W.; Colatsky, T. J.; Sheldon, J.; Norton, N. W.; Warga,
D.; Antane, M.; Antane, A.; Butera, J. A.; Argentieri, T.
M. J. Pharmacol. Exp. Ther. 1999, 289, 1410.
8. Gilbert, A. M.; Antane, M. M.; Argentieri, T. M.; Butera,
J. A.; Francisco, G. D.; Freeden, C.; Gundersen, E. G.;
Graceffa, R. F.; Herbst, D.; Hirth, B. H.; Lennox, J. R.;
McFarlane, G.; Norton, N. W.; Quagliato, D.; Sheldon, J.
H.; Warga, D.; Wojdan, A.; Woods, M. J. Med. Chem.
2000, 43, 1203.
In conclusion, numerous strategies were implemented
with the goal of attenuating the metabolic instability
of the vinylogous amide bond in the potent diaminocy-
clobutenedione KATP channel openers 1a and 1b.
Replacement of the labile NH linker with carbon-based
linkers (i.e., methylene, substituted methylene, olefin)
was not tolerated. Excision of the NH linker to afford
a directly bound aryl-cyclobutenedione generated a no-
vel scaffold (8), which possessed smooth muscle relaxing
properties that were not glyburide sensitive. Optimiza-
tion of the alkylamine substituent led to compound
8.10. Voltage-clamp studies on isolated human bron-
chial smooth muscle cells suggested that compound
8.10 caused an iberiotoxin-sensitive increase in hyperpo-
2.00
9. Butera, J. A.; Argentieri, T. M. Drugs Fut. 2000, 25,
239.
Control, n=3
10. Antane, M. M.; Antane, S. A.; Argentieri, T. A.; Butera, J.
A.; Francisco, G. P.; Freeden, C.; Gilbert, A. M.;
Graceffa, R. F.; Gundersen, E. G.; Herbst, D. R.; Hirth,
B. H.; Ho, D. M.; Jenkins, D.; Lennox, J. R.; Matelan, E.;
McFarlane, G. R.; Norton, N. W.; Oshiro, G.; Primeau, J.
L.; Quagliato, D. A.; Sheldon, J. H.; Spinelli, W.; Warga,
D.; Wojdan, A.; Woods, M. Abstract of Papers, 218th
National Meeting of the American Chemical Society, New
Orleans, LA; American Chemical Society: Washington
DC, 1999; Abstract MEDI 17.
Compound 8.10 1 µM, n=3
1.50
Compound 8.10 + IbTX 100 nM, n=3
*
1.00
*
*
0.50
11. Data not shown in this manuscript.
12. All compounds gave satisfactory spectral data. The
preparations of compounds 8.3, 9.2, 12.2, and 16.2 will
serve as sample experimentals for the syntheses of directly
bound phenyl squarates (8), benzyl squarates (9), alka-
noyl-substituted benzyl squarates (12), and styrenyl
squarates (16), respectively. Compound 8.3: To a cooled
(0 °C) solution of 3-isopropoxy-4-(tri-n-butylstannyl)-3-
cyclobutene-1,2-dione13 (8.00 g, 18.65 mmol) in DMF
(40 mL) was added 4-iodoanisole (4.85 g, 20.72 mmol).
Benzylchlorobis-(triphenylphosphine)palladium(II) (0.942
g, 1.24 mmol) and cuprous iodide (0.355 g, 1.87 mmol)
0.00
-20
0
20
40
60
80
100
V Test (mV)
Figure 2. Current–voltage relationship for outward current before
(control, s, n = 3) and after compound 8.10 (1 lM, d, n = 3). This
increase was reversed by the BKCa channel blocker iberiotoxin
(100 nM, m, n = 3). Compound 8.10 was associated with a significant
increase in outward current above +40 mV (*p 6 0.05). Data were
normalized to peak control current.

2500
J. A. Butera et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2495–2501
were added, and the mixture was stirred at rt overnight.
trituration with petroleum ether afforded 1.18 g (39%) of
3-[2-oxo-1-(4-trifluoromethyl-phenyl)-propyl]-4-ethoxy-cy-
clobut-3-ene-1,2-dione as a light yellow solid: 1H NMR
(DMSO-d6): d 11.51 (s, 1H), 7.60 (ABq, 2H), 7.24 (ABq,
2H), 4.59 (q, 2H), 1.89 (s, 3H), 1.19 (t, 3H). This
intermediate (0.350 g, 1.07 mmol) and 2-amino-3,3-di-
methylbutane (0.13 mL, 0.998 mmol) were stirred together
in ethanol (6 mL) at rt overnight. Diethyl ether (25 mL)
was added and the precipitated product was collected by
filtration. It was stirred in diethyl ether/petroleum ether
overnight, filtered, and dried in vacuo to afford 0.15 g
(37%) of 3-[2-oxo-1-(4-trifluoromethyl-phenyl)-propyl]-
4-(1,1,1-trimethyl-propylamino)-cyclobut-3-ene-1,2-dione
(12.2) as an off-white solid (1H NMR in CDCl3 suggested
the presence of both the keto and enol forms in about an
Dilution with diethyl ether (200 mL) followed by aqueous
work up with NH4Cl, 10% KF, and brine afforded crude
product, which was dissolved in hot ethyl acetate, decol-
orized (charcoal), and filtered. The filtrate was treated with
hexane and allowed to cool. 3-Isopropoxy-4-(4-methoxy-
phenyl)-cyclobut-3-ene-1,2-dione precipitated as a light
1
yellow solid (2.46 g, 54%): H NMR (DMSO-d6): d 7.89
(d, 2H), 7.15 (d, 2H), 5.45 (hept, 1H), 3.82 (s, 3H), 1.48 (d,
6H). To
a solution of the intermediate (0.150 g,
0.609 mmol) in ethanol (3 mL) was added 1,1-dimethyl-
2-phenyl-ethylamine (0.182 g, 1.22 mmol). The mixture
was stirred at 70 °C overnight, then filtered hot through a
pad of silica gel. The filtrate was concentrated and the
resulting residue was recrystallized from ethyl acetate/
hexanes to give 0.170 g (83%) of 3-(1,1-dimethyl-2-phenyl-
ethylamino)-4-(4-methoxy-phenyl)-cyclobut-3-ene-1,2-dione
(8.3) as a tan solid: mp 150–151 °C; 1H NMR (DMSO-d6):
d 8.28 (s, 1H), 8.00 (m, 2H), 7.28 (m, 2H), 7.26 (m, 1H),
7.13 (m, 2H), 7.09 (m, 2H), 3.82 (s, 3H), 3.10 (s, 2H), 1.42
(s, 6H); MS (m/z) 335 [M+]. Anal. Calcd for C21H21NO3:
C, 75.20; H, 6.31; N, 4.18. Found: C, 74.35; H, 6.41; N,
3.90. Compound 9.2: To a heterogeneous mixture of CuI
(178 mg, 0.932 mmol) and NaI (2.09 g, 14.0 mmol) in
anhydrous DMF (5.0 mL) was added 4-methoxybenzyl
chloride. The reaction mixture, was stirred for 30 min at
rt, whereupon a solution of 3-isopropoxy-4-(tri-n-butyl-
stannyl)-3-cyclobut-3-ene-1,2-dione (4.00 g, 9.32 mmol) in
DMF (5.0 mL) was added, followed by addition of the
benzylchlorobis-(triphenylphosphine)palladium(II) cata-
lyst (530 mg, 0.699 mmol). Upon stirring at rt for 2 h,
the mixture was diluted with ethyl acetate (250 mL), and
subjected to aqueous work up with saturated NH4Cl, 10%
KF and brine (100 mL). The organic phase was dried over
Na2SO4/charcoal and submitted to flash chromatography
(40% ether–petroleum ether) affording 3-(4-methoxy-benz-
1
8:1 ratio): mp 178.2–179.8 °C; H NMR (CDCl3): d 11.74
(s, 1H), 7.73 (ABq, 2H), 7.39 (ABq, 2H), 3.92 (m, 1H),
1.86 (s, 3H), 0.93 (d, 3H), 0.69 (s, 9H); MS (m/z) 381 [M+].
Anal. Calcd for C20H22NO3F3: C, 62.98; H, 5.81; N, 3.67.
Found: C, 62.67; H, 5.72; N, 3.56. Compound 16.2: To a
heterogeneous mixture of CrCl2 (3.98 g, 32.4 mmol) in
THF (40 mL) at 0 °C was added via syringe pump over 1 h
(in the total absence of light) a solution of 4-bromobenz-
aldehyde (1.00 g, 5.40 mmol) and CHI3 (4.25 g,
10.8 mmol) in THF (20 mL). The cold bath was removed,
and the mixture stirred at rt for 1 h. The reaction mixture
was diluted with 1:1 ether–hexanes (200 mL), then filtered
through a short pad of SiO2, and concentrated to give
crude (E)-1-iodo-2-(4-bromophenyl)ethylene. The com-
pound (5.4 mmol) was used as is and treated with CuI
(0.103 g, 0.54 mmol), benzylchlorobis-(triphenylphos-
phine)palladium(II) (0.307 g, 0.42 mmol), and 3-isoprop-
oxy-4-(tri-n-butylstannyl)-3-cyclobutene-1,2-dione (3.11
mL, 8.27 mmol) in DMF (18 mL) at rt (in the dark). An
aqueous workup and filtration through a plug of silica gel
afforded 0.298 g (17%) of 4-isopropoxy-3-[(E)-2-(4-bromo-
yl)-4-isopropoxy-cyclobut-3-ene-1,2-dione
as
golden
phenyl)-vinyl]-cyclobut-3-ene-1,2-dione as a dark oil,
which was used directly in the next step. To this
intermediate (0.288 g, 0.898 mmol) in ethanol (3.5 mL)
brown oil (1.04 g, 43%): 1H NMR (CDCl3): d 7.20
(ABq, 2H), 6.85 (ABq, 2H), 5.38 (hept, 1H), 3.83 (s,
2H), 3.79 (s, 3H), 1.43 (d, 6H). In a manner similar to
compound 8.3, the intermediate (350 mg, 1.34 mmol) was
dissolved in anhydrous isopropyl alcohol (7.0 mL) and
was treated with 2-amino-3,3-dimethylbutane (272 mg,
2.69 mmol) at rt, affording 0.124 g (31%) of 3-(4-methoxy-
benzyl)-4-(1,2,2-trimethyl-propylamino)cyclobut-3-ene-1,2-
dione (9.2) one-quarter hydrate (1H NMR in DMSO-d6
suggested the presence of amide rotamers in a ratio of
approximately 3:1): mp 137.4–138.1 °C; 1H NMR
(DMSO-d6): d 8.63 (d, 1H), 7.16 (ABq, 2H), 6.87 (ABq,
2H), 3.90 (m, 1H), 3.70 (s, 2H), 1.15 (d, 3H), 0.84 (s, 9H);
MS (m/z) 301 [M+]. Anal. Calcd for C18H23NO3Æ0.25
H2O: C, 70.68; H, 7.74; N, 4.58. Found: C, 70.43; H, 7.77;
N, 4.55. Compound 12.2: A solution of 1-[(4-trifluoro-
methyl)phenyl]-2-propanone16 (1.86 g, 9.208 mmol) in
THF (10 mL) was added dropwise (under N2) to a cooled
(ꢀ78 °C) solution of potassium bis(trimethylsilyl)amide
(19.3 mL; 0.5 M in toluene, 9.67 mmol) in THF/diethyl
ether (1:1 ratio, 80 mL). The mixture stirred at ꢀ78 °C for
15 min and was then stirred at rt for 2.5 h. The enolate
solution was cooled to ꢀ78 °C and added by cannula to a
cooled (ꢀ78 °C) flask containing diethyl squarate
(1.50 mL, 10.13 mmol) in THF/diethyl ether (1:1 ratio,
20 mL). The reaction was stirred for 15 min at ꢀ78 °C and
was then allowed to warm to rt over a 1 h. The reaction
was concentrated to give a residue, which was partitioned
between 0.1 N HCl and ethyl acetate. The organic phase
was washed with brine, dried (MgSO4), and concentrated
to give crude product. Purification by flash column
chromatography (2:1 hexanes/ethyl acetate) followed by
was
added
2-amino-3,3-dimethylbutane
(241 lL,
1.80 mmol). The mixture was stirred at rt overnight,
diluted with methanol and filtered. The solid was washed
with cold methanol, and dried in vacuo at 60 °C to afford
3-[(E)-2-(4-bromo-phenyl)-vinyl]-4-(1,2,2-trimethyl-propyl-
amino)-cyclobut-3-ene-1,2-dione (127 mg, 39%) (16.2) as
a hygroscopic yellow solid (1H NMR in DMSO-d6
suggested the presence of amide rotamers in a ratio of
approximately 4:1): mp 192.3–192.9 °C; 1H NMR
(DMSO-d6): d 8.87 (d, 1H), 7.71 (m, 5H), 7.46 (d, 1H),
4.02 (dq, 1H), 1.19 (d, 3H), 0.90 (s, 9H); IR (KBr) 3310,
2970, 1760, 1705, 1575, 1480, 1430, 1300, 1220, 1160, 1070,
1005, 975, 875, 820, 800 cmꢀ1; MS (m/z) 361/363 [M+].
Anal. Calcd for C18H20BrNO2: C, 59.68; H, 5.56; N, 3.87.
Found: C, 59.55; H, 5.38; N, 3.72.
13. Liebeskind, L. S.; Fengl, R. W. J. Org. Chem. 1990, 55,
5359.
14. Goure, W. F.; Wright, M. E.; Davis, P. D.; Labadie, S. S.;
Stille, J. K. J. Am. Chem. Soc. 1984, 106, 6417.
15. Takai, K.; Nitta, K.; Utimoto, K. J. Am. Chem. Soc. 1986,
108, 7408.
16. Saari, W. S.; Williams, J.; Britcher, S. F.; Wolf, D. E.;
Kuehl, F. A. J. Med. Chem. 1967, 10, 1008.
17. Protocol for determination of in vitro relaxant properties
in bladder and aortic tissue reported in: Butera, J. A.;
Antane, S. A.; Hirth, B. H.; Lennox, J. R.; Sheldon, J. H.;
Norton, N. W.; Warga, D.; Argentieri, T. M. Bioorg. Med.
Chem. Lett. 2001, 11, 2093.
18. Brief cell EP protocol: Human bronchial smooth muscle
cells were acquired from Clonetics (Walkersville, MD) and

J. A. Butera et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2495–2501
2501
cultured in house. Cells were removed from the culture
dishes by brief exposure to trypsin and then placed directly
into the tissue bath for study. Cells were studied using
standard voltage clamp technique19 in 32 °C physiological
salt solution (PSS) that contained (mM): NaCl (118.4),
KCl (5.0), CaCl2 (2.5), MgSO4 (1.2), KH2PO4 (1.2),
NaHCO3 (24.9), and D-glucose (11.1) gassed with O2–
CO2, 95/5 to achieve a pH of 7.4. The pipette solution
contained (mM): KCl (126.0), MgCl2Æ6H2O (4.5), ATP
Mg salt (4.0), GTP tris salt (0.3), creatine PO4 (14.0), D-
glucose (9.0), EGTA (9.0), HEPES (9.0). The pH was
adjusted to 7.4 with KOH. Electrodes had tip resistances
of 2–4 MX. Whole cell recordings were made using an
Axon Instruments 200A patch clamp amplifier (3 kHz
high frequency cut-off) and pClamp software (Axon
Instruments, Foster City, CA). Currents were evoked
from a holding potential of ꢀ50 mV and then ramping the
membrane voltage from ꢀ100 to +80 mV at a rate of
3.4 mV/s. Stable baseline recordings were made for a
minimum of 5 min after which time drug was added to the
superperfusate. Currents were again given a minimum of
5 min to stabilize before measurements were made. After
drug administration, iberiotoxin was also added to the
superperfusate to reverse any effects on BKCa current.20
Data were analyzed using a 586-based personal computer
and pClamp (Axon Instruments, Foster City, CA) soft-
ware. Measured currents were normalized to peak control
current and comparisons were made between control and
treatment for each cell. Data are represented as mean
SEM for each treatment group and analyzed for statistical
significance utilizing the paired Studentꢀs t-test at the
p 6 0.05 level of significance.
19. Hamill, O. P.; Marty, A.; Neher, E.; Sakmann, B.;
Sigworth, F. J. Pflug. Arch.—Eur. J. Physiol. 1981, 391,
85.
20. Galves, A.; Gimenez-Gallego, G.; Reuben, J. P.; Roy-
Contancin, L.; Feigenbaum, P.; Kaczorowski, G. J.;
Garcia, M. L. J. Biol. Chem. 1990, 265, 11083.