Synthesis and potassium channel opening activity of substituted 10H-benzo[4,5]furo[3,2-b]indole- and 5,10-dihydro-indeno[1,2-b]indole-1-carboxylic acids

Article abstract of DOI:10.1016/S0960-894X(01)00385-7

Compounds in a structurally novel series of substituted 10H-benzo[4,5]furo[3,2-b]indole-1-carboxylic acids and related 5,10-dihydro-indeno[1,2-b]indole-1-carboxylic acids were prepared and shown to possess potent, bladder-selective smooth muscle relaxant

Full text of DOI:10.1016/S0960-894X(01)00385-7

Bioorganic & Medicinal Chemistry Letters 11 (2001) 2093–2097
Synthesis and Potassium Channel Opening Activity of
Substituted 10H-Benzo[4,5]furo[3,2-b]indole- and
5,10-Dihydro-indeno[1,2-b]indole-1-carboxylic Acids
John A. Butera,a,* Schuyler A. Antane,a Bradford Hirth,a Joseph R. Lennox,a
Jeffrey H. Sheldon,b N. Wesley Norton,b Dawn Wargab and Thomas M. Argentierib
aMedicinal Chemistry, Chemical Sciences, Wyeth-Ayerst Research, CN 8000, Princeton, NJ 08543, USA
bUrologic Diseases, Women’s Health Discovery Research, Wyeth-Ayerst Research, 145 King of Prussia Road,
Radnor, PA 19087, USA
Received 21 March 2001; accepted 15 May 2001
Abstract—Compounds in a structurally novel series of substituted 10H-benzo[4,5]furo[3,2-b]indole-1-carboxylic acids and related
5,10-dihydro-indeno[1,2-b]indole-1-carboxylic acids were prepared and shown to possess potent, bladder-selective smooth muscle
relaxant properties and thus are potentially useful for the treatment of urge urinary incontinence. Electrophysiological studies using
rat detrusor myocytes have demonstrated that prototype compound 7 produces a significant increase in hyperpolarizing current,
which is iberiotoxin (IbTx)-reversed, thus consistent with activation of the large-conductance Ca2+-activated potassium channel
(BKCa). # 2001 Elsevier Science Ltd. All rights reserved.
Modulation of potassium channels continues to provide
potentially useful approaches to treat pathological con-
ditions involving cell hyperactivity.1 By virtue of their
ability to hyperpolarize cells and bring the membrane
potential further from the threshold for activation of
voltage-gated Ca2+ channels, potassium channel open-
ers (KCOs) reduce excitability and relax smooth muscle
cells. The structure–function relationship of physiologi-
cally relevant potassium channel subtypes has been
reviewed.2 It has been shown that bladder smooth mus-
cle contains several types of potassium channels and
that their activation by a variety of KCOs leads to
detrusor muscle relaxation in vitro and in vivo.3 Tissue-
specific agents that selectively activate these bladder
channels may show promise as effective therapeutics for
the treatment of urge urinary incontinence. We have
recently reported on a novel class of 1,2-diaminocyclo-
butene-3,4-diones as bladder selective KATP channel
openers demonstrating potent in vivo bladder relaxant
properties at doses which caused minimal hemodynamic
effects.4 As part of our continuing efforts focused on
structural modifications of prototypical KATP agents,
celikalim (1), a benzopyran-based antihypertensive
KATP channel opener was chosen as a starting point for
new synthesis (Fig. 1). A series of 4-indazolinonyl deri-
vatives (2) was prepared and selected compounds were
shown to possess potent bladder relaxant properties
that were mediated via activation of the KATP channel.5
Ring-contracted 2,3-dihydro-benzofuran analogues 3
were subsequently targeted for synthesis.
The initial attempt to reductively cyclize hydrazone 5
(Scheme 1) in the presence of NaCNBH3 and acetic acid
afforded only trace amounts of desired dihydro-benzo-
furan target 6, but afforded instead the Fischer-indole
product 10H-benzo[4,5]furo[3,2-b]indole (7).
Figure 1. Structural modification of prototype KATP channel opener
celikalim (1) afforded indazole leads (2). Ring contracted analogues (3)
were subsequently targeted.
*Corresponding author. Fax: +1-732-274-4129; e-mail: buteraj@war.
wyeth.com
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00385-7

2094
J. A. Butera et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2093–2097
The compound was evaluated for its relaxant properties
in both rat detrusor tissue and thoracic aortic rings and
was found to be a highly selective bladder relaxant
(IC50=15 mM) in vitro (IC50 ratio aorta/bladder=8).
While relaxation of bladder tissue associated with our
earlier KATP channel activator leads was readily
reversed by addition of glyburide (a selective KATP
channel blocker), relaxation induced by 7 was only
reversed by addition of iberiotoxin, a selective blocker
of the large conductance, Ca2+-activated potassium
channel (BKCa). Due to its novel structure and unique
in vitro profile, a series of benzofuro-indoles and
indeno-indoles related to 7 was generated and the com-
pounds were evaluated for their smooth muscle relaxant
properties.
Scheme 1. An attempt to reductively cyclize hydrazone 5 to produce
2-benzofuranyl-indazolone 6 gave instead Fischer-indole product 7,
which was evaluated as a bladder relaxant.
Scheme 2. Synthesis of 10H-benzo[4,5]-furo[3,2-b]indoles (template A) and 5,10-dihydro-indeno[1,2-b]indoles (templates B and C): (a) H2O, ethanol,
20 ꢀC; (b) HCO2H, microwave irradiation; (c) DMF, NaH, MeI, À5 to 20 ꢀC; (d) THF, H2O, LiOH, 20 ꢀC; (e) Et2O, PCl5, 20 ꢀC; (f) Et2O, NH3; (g)
H2SO4, CH3OH, 100 ꢀC; (h) THF, LiAlH4, 20 ꢀC; (i) CH3CN, Dess–Martin periodinane, 20 ꢀC.
Table 1. In vitro effects of tetracycles A, B, and C on precontracted rat bladder smooth muscle strips and rat aortic rings (IC50, mM)
Compd
Template
X
Y
R1
R2
IC50 bladdera
nb
IC50 aortac
nb
7
A
A
A
A
A
A
A
A
A
B
O
O
O
O
O
O
O
O
O
NH
NH
NH
NH
Br
I
Cl
NO2
H
Br
Br
Br
Br
Br
NO2
Br
I
Cl
OH
OH
OH
OH
OH
OH
NH2
OCH3
H
OH
OH
OH
OH
OH
OH
OH
15.1Æ4.7
6.1Æ3.0
5.8Æ1.5
6.3Æ2.6
30Æ6.5
>30
>30
>30
>30
8.6Æ3.0
5.2Æ1.8
4.4Æ0.95
10.1Æ3.2
22.6Æ1.8
>30
8
8
6
8
4
2
2
4
2
7
4
8
5
5
2
4
118Æ 21.8
128Æ16.9
268Æ82.3
125Æ13.8
NTd
9
3
4
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
3
NH
—
—
—
—
—
4
4
6
—
—
—
—
NCH3
NH
NH
NH
NH
NT
NT
NT
NT
CH2
C(CH3)2
NH
NH
NH
NH
NH
210Æ45.0
17.2Æ2.3
109Æ12.4
NT
B
NH
C
C
C
C
C
CH2
CH2
CH2
CH2
CH2
NT
NT
NT
SO2NH2
OCF3
15.4Æ4.8
aDrug concentrationÆSE (mM) that relaxed KCl-induced contractions in rat detrusor strips by 50%.
bNumber of experiments.
cDrug concentrationsÆSE (mM) that relaxed KCl-induced contractions in rat aortic rings by 50%.
dNot tested.

J. A. Butera et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2093–2097
2095
Figure 2. Current voltage relationship calculated from measured cur-
rent from rat bladder smooth muscle cells. Compound 7 (10 mM, n=3)
was associated with an increase in outward current above +30 mV. A
3.5Â increase in outward current was observed at +80 mV. These
increases were reversed by iberiotoxin (100 nM).
Treatment (Scheme 2) of an appropriately substituted
benzofuranone (8, X=O) or indanone (8, X=C) with 2-
hydrazinobenzoic acid in aqueous media afforded
phenyl hydrazones 10, which were subjected to micro-
wave-facilitated6,7 Fischer-indole cyclization in formic
acid to afford a series of 10H-benzo[4,5]-furo[3,2-
b]indoles (template A) or 5,10-dihydro-indeno[1,2-
b]indoles (template B). Regioisomeric indeno-indoles
(template C) were generated by condensing substituted
phenylhydrazines (11) with indanone carboxylic acids
(12) followed by microwave-facilitated Fischer-indole
cyclization of 13. The N-methylindole analogue 18 and
carboxylic acid analogues 19–21 were prepared from 7
using standard transformations described in Scheme 2.
All target compounds were evaluated in vitro for their
ability to relax KCl precontracted rat detrusor muscle
strips.8 To ascertain intrinsic tissue selectivity, selected
compounds were screened in similar tissue baths using
rat aortic rings. The data are shown in Table 1.
The effect of the 8-bromo substituent was studied first.
While removal of the bromine atom (17) resulted in 2-
fold loss of in vitro activity, replacement with I, Cl, or
NO2 (14, 15, and 16, respectively) resulted in over 2-fold
increase in bladder relaxant potency. Of these five ana-
logues, 8-Cl derivative 15 was the most bladder selective
smooth muscle relaxant (ratio=46). The free indole N–
H and carboxylic acid proton are critical for in vitro
bladder relaxing properties as demonstrated by the loss
of activity associated with N-methylation (18) and con-
version of the acid to a primary amide (19), ester (20),
or aldehyde (21). The oxygen atom of the benzofuran
portion of the tetracycle could be replaced by carbon
(22 and 23) with retention of both bladder relaxant
potency and selectivity over aortic tissue. Regioisomeric
5,10-dihydro-indeno[1,2-b]indoles 24–26 and 28
retained in vitro bladder relaxant activity. In general,
the same types of substituents are tolerated in place of
bromine. A direct comparison of regioisomers 22 and 24
reveals about a 2-fold preference for the 8-bromo-5,10-
dihydro-indeno[1,2-b]indole-1-carboxylic acid versus the
3-bromo-5,10-dihydro-indeno[1,2-b]indole-6-carboxylic
acid with surprisingly similar tissue selectivities for both
compounds.
Figure 3. Representative tracings recorded from rat bladder smooth
muscle cells. Cells were held at À10 mV and stepped to +80 mV in 10
mV increments for 1000 ms. The top panel shows control currents.
Compound 7 (10 mM, n=3) increased outward currents as illustrated
in the center panel. After exposure to iberiotoxin (100 nM), recorded
currents returned to control levels (lower panel). Steps at 0, 20, 40, 60,
and 80 mV are displayed.
To investigate the underlying mechanism of action of
this novel series of selective bladder smooth muscle
relaxants, several cell electrophysiological studies were
undertaken with prototype compound 7. Voltage clamp
studies9 on isolated rat detrusor myocytes (control rest-
ing membrane potential=À45Æ5 mV) showed that
exposure to 7 (10 mM) increased outward current above
+30 mV test potentials (Fig. 2). A 3.5-fold increase in
outward current was observed at +80 mV. Concurrent
exposure with iberiotoxin, a selective BKCa channel
blocker, reversed these increases back to control level
currents. Representative whole cell voltage clamp tra-
cings are shown in Figure 3.

2096
J. A. Butera et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2093–2097
The electrophysiological data indicate that furano-
indole 7 can activate a hyperpolarizing current, con-
sistent with the large-conductance calcium-dependent
potassium channel, in isolated rat bladder cells. The
resulting transmembrane hyperpolarization would be
expected to underlie, in part, the bladder smooth muscle
relaxing properties of this compound. Of major interest
is the apparent selectivity for the bladder versus aortic
smooth muscle. Other investigators have demonstrated
that vascular smooth muscle cells contain BKCa chan-
nels; therefore the bladder selectivity cannot simply be
attributed to the lack of BKCa channels in the aortic
tissues. Evaluation of the electrophysiological effects of
these compounds on isolated vascular smooth muscle
cells will be required to understand the difference in
potency between bladder and aortic smooth muscle.
4. (a) 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, 1187. (b) 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, 1202. (c) Butera, J. A.;
Argentieri, T. M. Drugs Future 2000, 25, 239. (d) Wojdan, A.;
Freeden, C.; Woods, M.; Oshiro, G.; Spinelli, W.; Colatsky,
T. J.; Sheldon, J. H.; Norton, N. W.; Warga, D.; Antane,
M. M.; Antane, S. A.; Butera, J. A.; Argentieri, T. M. J.
Pharmacol. Exp. Ther. 1999, 289, 1410.
5. Antane, S. A.; Butera, J. A.; Argentieri, T. M.; Norton, N.
W.; Zebick, D. M. Abstracts of Papers, 207th National Meet-
ing of the American Chemical Society, 1994; American Che-
mical Society: Washington, DC, 1994; MEDI 227.
In conclusion, an initial attempt to prepare a structu-
rally modified analogue of benzopyranyl-based KATP
channel opener 2 afforded the unexpected Fischer-
indole product 7. Assessment of its biological properties
revealed a unique bladder selective in vitro smooth
muscle relaxing profile consistent with activation of the
BKCa channel. A series of related 10H-benzo[4,5]-
6. Abramovitch, R. A.; Bulman, A. Synlett 1992, 10, 795.
7. All compounds gave satisfactory spectral data. The pre-
parations of compounds 7, 22, and 24 serve as sample experi-
mentals for the syntheses of templates A, B, and C,
respectively. To a solution of 5-bromo-3(2H)-benzofuranone10
(3.10 g, 14.6 mmol) in ethanol (100 mL) was added a solution
of 2-hydrazinobenzoic acid hydrochloride (5.49 g, 29.1 mmol)
in deionized water (200 mL). The mixture was stirred for 1 h at
20 ꢀC and then cooled to 0 ꢀC. Vacuum filtration and drying
in vacuo afforded 3.65 g (72%) of o-[(2,3-dihydro-5-bromo-
furo[3,2-b]indoles
and
5,10-dihydro-indeno[1,2-
b]indoles was generated utilizing a microwave-induced
Fischer-indole cyclization as the key step. Compounds 7
and 14–28 were evaluated for their ability to relax KCl
pre-contracted rat detrusor strips and thoracic aortic
rings. Test compounds were found to be highly bladder
selective with aorta/bladder IC50 ratios ranging from 8
to 46-fold. Voltage clamp studies on isolated rat bladder
myocytes suggest that prototype furano-indole 7 causes
an iberiotoxin-sensitive increase in hyperpolarizing current
consistent with activation of the BKCa channel. Studies are
in progress to further evaluate this series for potential use
in the treatment of urge urinary incontinence.
benzofuran-3-ylidene)hydrazino]-benzoic acid as
a brown
solid: mp 195 ꢀC (dec) which was used without further pur-
ification. The hydrazone (0.500 g, 1.51 mmol) was suspended
in formic acid (2 mL, 96%) and was irradiated for 2 min in a
closed cap Teflon vessel in a microwave oven (700 W). The
mixture was vacuum filtered hot and the solid was dried in
vacuo to yield 0.271 g (57%) of 7 as a yellow solid: mp 312–
313 ꢀC; H NMR (DMSO-d6) d 13.36 (s, 1H), 11.64 (s, 1H),
1
8.27 (d, J=2.1 Hz, 1H), 8.07 (d, J=7.9 Hz, 1H), 7.92 (dd,
J1=7.5 Hz, J2=1.0 Hz, 1H), 7.70 (d, J=8.7 Hz, 1H), 7.50 (dd,
J1=8.7 Hz, J2=2.1 Hz, 1H), 7.27 (dd, J1=7.9 Hz, J2=7.5 Hz,
1H); IR (KBr): 3420, 1685 cmÀ1; MS (m/z) 329 (M+). Anal.
calcd for C15H8BrNO3: C, 54.57; H, 2.44; N, 4.24; found: C,
54.22; H, 2.32; N, 4.30. To a solution of 6-bromoindanone11
(0.447 g, 2.12 mmol) in ethanol (100 mL) was added a solution
of 2-hydrazinobenzoic acid hydrochloride (0.800 g, 4.24
mmol) in deionized water (50 mL). The mixture was stirred for
1 h then cooled to 0 ꢀC. The precipitated o-[(2,3-dihydro-6-
bromoinden-3-ylidene)hydrazino]-benzoic acid was vacuum
filtered and dried in vacuo, yield: 0.628 g (86%) of a yellow
solid: mp 186 ꢀC (dec) which was used as is. The hydrazone
(0.620 g, 1.80 mmol) in formic acid (2 mL, 96%) was irra-
diated for 2 min in a closed cap Teflon vessel of a microwave
oven (700 W). The mixture was vacuum filtered hot and the
solid was dried in vacuo to yield 0.360 g (61%) of 22 as a yel-
References and Notes
1. (a) Lawson, K. Kidney International 2000, 57, 838. (b)
Atwal, K. Med. Res. Rev. 1992, 12, 569. (c) Primeau, J.;
Butera, J. Curr. Pharm. Des. 1995, 1, 391. (d) Quast, U. Trends
Pharmacol. Sci. 1993, 14, 332. (e) Longman, S. D.; Hamilton,
T. C. Med. Res. Rev. 1992, 12, 73. (f) Lawson, K. Pharmacol.
Ther. 1996, 70, 39.
2. (a) Aguilar-Bryan, L.; Clement, J. P.; Gonzalez, G.; Kun-
jilwar, K.; Babenko, A.; Bryan, J. Physiol. Rev. 1998, 78, 227.
(b) Bryan, J.; Aguilar-Bryan, L. Curr. Opin. Cell. Biol. 1997, 9,
553. (c) Garcia, M. L.; Hanner, M.; Knaus, H.-G.; Koch, R.;
Schmalhofer, W.; Slaughter, R. S.; Kaczorowski, G. J. Adv.
Pharmacol. 1997, 39, 425. (d) McDonough, S.; Lester, H. A.
DrugDev. Res. 1994, 33, 190. (e) Wible, B. A.; Brown, A. M.
DrugDev. Res. 1994, 33, 225. (f) Edwards, G.; Weston, A. H.
Expert Opin. Invest. Drugs 1996, 5, 1453.
low solid: mp 245–247 ꢀC (dec); H NMR (DMSO-d6) d 13.43
1
(s, 1H), 11.73 (s, 1H), 8.32 (d, J=2.0 Hz, 1H), 7.86 (d, J=7.9
Hz, 1H), 7.78 (dd, J1=7.6 Hz, J2=0.8 Hz, 1H), 7.49 (d, J=8
Hz, 1H), 7.36 (dd, J1=8.0 Hz, J2=2.0 Hz, 1H), 7.16 (dd,
J1=7.9 Hz, J2=7.6 Hz, 1H), 3.71 (s, 2H); IR (KBr): 3460,
1650 cmÀ1
;
MS (m/z) 327 (M+). Anal. calcd for
3. (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.
C16H10BrNO2: C, 58.56; H, 3.07; N, 4.27; found: C, 58.62; H,
2.83; N, 4.22. 1-Oxo-4-indancarboxylic acid12 (0.528 g, 2.98
mmol) and 4-bromophenylhydrazine (0.566 g, 3.00 mmol) in
formic acid (2 mL, 96%) were irradiated for 2 min in a closed cap
Teflon vessel of a microwave oven (700 W). The mixture was
vacuum filtered. The crude product was dissolved in acetone/
diethyl ether (1:1) and treated with decolorizing carbon, filtered,
concentrated and dried in vacuo to yield 0.530 g (54%) of 24 as a
white solid: mp 328–330 ꢀC (dec); 1H NMR (DMSO-d6) d 13.43

J. A. Butera et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2093–2097
2097
(s, 1H), 11.85 (s, 1H), 7.78–7.83 (m, 3H), 7.50 (dd, J1=7.89 Hz,
J2=7.47 Hz, 1H), 7.43 (d, J=8.3 Hz, 1H), 7.16 (dd, J1=8.5 Hz,
J2=2.1 Hz, 1H), 4.00 (s, 2H); IR (KBr): 3440, 1690 cmÀ1; MS (m/
z) 327 (M+). Anal. calcd for C16H10BrNO2: C, 58.56; H, 3.07; N,
4.27; found: C, 58.57; H, 2.88; N, 4.30.
8. In vitro studies protocol as described in ref 4a. Briefly:
female Sprague–Dawley rats (Charles River, Wilmington,
MA, USA; 250–350 g) were rendered unconscious via inhala-
tion of CO2 and exsanguinated. The entire bladder was
removed and placed into room temperature PSS of the fol-
lowing composition (mM): NaCl (118.4), KCl (4.7), CaCl2
(2.5), MgSO4 (1.2), KH2PO4 (1.2), NaHCO3 (24.9) and d-glu-
cose (11.1) gassed with O2–CO2, 95%/5% to achieve a pH of
7.4. The dome of the bladder was isolated from the trigone
region and the mucosa was removed. The detrusor was then
cut into strips 4–5 mm wide by 10 mm long. One end was
secured to the bottom of a 10 mL tissue bath and the other to
contracted with an additional 15 mM KCl and again allowed
to equilibrate for approximately 90 min. Compounds were
administered directly into the tissue baths as cumulative con-
centrations and responses allowed to reach steady state. Sig-
nals were digitized (486 based personal computer, 12-bit
resolution, 1 s sampling interval, custom software) for on-line
analysis. Since isolated bladder strips contract with irregular
frequency and amplitude, a 5-min area-under-the-contraction
curve was used to assess contractility after achieving steady
state for each concentration. Aortic ring studies were per-
formed in a similar matter.
9. Cell electrophysiological studies protocol as described in
ref 4d.
10. Ellingboe, J. W.; Alessi, T. R.; Dolak, T. M.; Nguyen,
T. T.; Tomer, J. D.; Guzzo, F.; Bagli, J. F.; McCaleb, M. L. J.
Med. Chem. 1992, 35, 1176.
11. Adamczyk, M.; Watt, D. S.; Netzel, D. A. J. Org. Chem.
1984, 49, 4226.
a
Grass isometric force transducer (Grass Instruments,
Quincy, MA, USA). Tissues were pretensioned (0.25–0.5 g)
and allowed to equilibrate for 30 min. After which, strips were
12. Aono, T.; Araki, Y.; Imanishi, M.; Noguchi, S. Chem.
Pharm. Bull. 1978, 26, 1153.