Discovery of 4'-[(imidazol-1-yl)methyl]biphenyl-2-sulfonamides as dual endothelin/angiotensin II receptor antagonists.

Article abstract of DOI:10.1016/S0960-894X(03)00018-0

A series of 4'-[(imidazol-1-yl)methyl]biphenylsulfonamides has potent antagonist activity against both angiotensin II AT(1) and endothelin ET(A) receptors. Such dual-acting antagonists could have utility in the treatment of hypertension, heart failure, an

Full text of DOI:10.1016/S0960-894X(03)00018-0

Bioorganic & Medicinal Chemistry Letters 13 (2003) 1093–1096
Discovery of 40-[(Imidazol-1-yl)methyl]biphenyl-2-sulfonamides as
Dual Endothelin/Angiotensin II Receptor Antagonists
John E. Tellew,a,* Rose Ann F. Baska,b Sophie M. Beyer,b Kenneth E. Carlson,b
Lyndon A. Cornelius,a Leena Fadnis,a Zhengxiang Gu,a Bridgette L. Kunst,a,y
Mark C. Kowala,b Hossain Monshizadegan,b Natesan Murugesan,a Carol S. Ryan,b
Maria T. Valentine,b Yifan Yangb and John E. Macora
aDepartment of Discovery Chemistry, Bristol-Myers Squibb Pharmaceutical Research Institute, Princeton, NJ 08543-5400, USA
bDepartment of Metabolic and Cardiovascular Drug Discovery, Bristol-Myers Squibb Pharmaceutical Research Institute,
Princeton, NJ 08543-5400, USA
Received 16 September 2002; revised 31 December 2002; accepted 7 January 2003
Abstract—A series of 40-[(imidazol-1-yl)methyl]biphenylsulfonamides has potent antagonist activity against both angiotensin II
AT1 and endothelin ETA receptors. Such dual-acting antagonists could have utility in the treatment of hypertension, heart failure,
and other cardiovascular diseases in a broad patient population. Certain compounds in the present series are orally active in a rat
model of angiotensin II-mediated hypertension.
# 2003 Elsevier Science Ltd. All rights reserved.
Angiotensin II subtype 1 (AT1) receptor antagonists are
clinically useful for the management of hypertension
and heart failure,1 while endothelin subtype A (ETA)
receptor antagonists show promise for treatment of
similar indications.2 We have been engaged in a pro-
gram to develop single agents with the ability to antag-
onize both AT1 and ETA receptors,3 and our efforts
resulted in the discovery of BMS-248360, a potent
(human ETA Ki=2 nM; rat AT1 Ki=6 nM) dual-action
receptor antagonist.4 BMS-248360 bears structural fea-
tures of both irbesartan, a marketed AT1 receptor
antagonist, and BMS-193884, an ETA receptor antago-
nist discovered at Bristol-Myers Squibb (BMS).5 We
sought to expand on this work by merging features of
various other reported AT1 receptor antagonists with
those of BMS-193884. In this communication, we
report that potent dual antagonist activity has been
found in an imidazole series related to the AT1 recep-
tor antagonist losartan. Furthermore, certain com-
pounds in this new class possess promising oral efficacy
in a rat model of angiotensin II-mediated hypertension.
With few exceptions, angiotensin II receptor antagonists
are structurally similar to Dupont’s biphenyl-2-tetrazole
series, exemplified by EXP-3174.6 These compounds are
generally selective for the AT1 receptor subtype over the
AT2 subtype. Many biphenyl 40 substituents are known,
and certain structural features are required for good
AT1 receptor binding affinity.7 Notable is the require-
ment for a hydrogen-bond acceptor in the space occu-
pied by the imidazole C-5 substituent in EXP-3174. In
contrast, endothelin receptor antagonism has been
found in a wide range of chemotypes,8 including a series
of biphenyl-2-sulfonamides discovered at BMS.5,9 These
compounds, exemplified by BMS-193884, are ETA
receptor subtype-selective and superficially can be seen to
resemble AT1 receptor antagonists. An acidic heteroaryl-
substituted sulfonamide is required for ETA receptor
*Corresponding author at current address: Department of Medicinal
Chemistry, Neurocrine Biosciences, Inc., 10555 Science Center Drive,
San Diego, CA 92121, USA. Tel.: +1-858-320-7805; fax: +1-858-658-
7619; e-mail: jtellew@neurocrine.com
yMaiden name: Bridgette L. Kane.
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00018-0

1094
J. E. Tellew et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1093–1096
antagonism, and particularly good activity is found with
a 3,4-dimethyl-5-isoxazolyl pharmacophore.
and N-3 regioisomeric alkylation products, which were
separated by silica gel chromatography. Structural
assignments for these products were determined by H
NMR and were based on literature precedents.10b,c
1
Combining the AT1 and ETA receptor antagonist
chemotypes requires four significant structural com-
promises. First, an ‘AII group’ at the 40 position must
be tolerated by the ETA receptor. Second, the AT1
receptor must tolerate the specific heteroaryl sulfona-
mide required for ETA antagonism. Third, we have
found that ETA antagonism is enhanced by certain
biphenyl 20 substituents, yet these substituents may
interfere with AT1 antagonism. Finally, the resultant
dual antagonist must possess oral bioavailability, which
is difficult to achieve in large molecules (i.e., MW >500).
Since the biphenyl scaffold, the heteroaryl sulfonamide,
and a 40 ‘AII group’ were all required elements, we sought
to find the smallest effective 40 and 20 substituents. The
(imidazol-1-yl)methyl 40 substituent present in losartan
and its active metabolite EXP-3174 is one of the smallest
high-potency ‘AII groups’ reported to date. Although the
anionic carboxylate found in EXP-3174 would not be
conducive to ETA receptor antagonism5 or to achieve-
ment of pharmacokinetic goals, replacement of this group
with neutral hydrogen bond acceptors was precedented.10
Our strategy for the optimization of binding affinity13 was
as follows: First, optimize for AT1 receptor antagonism
through choice of imidazole substituents with a 20-unsub-
stituted biphenyl. Second, improve ETA receptor antagon-
ism by introducing a 20 substituent, which we hoped could
be done without compromising AT1 receptor affinity.
Since various researchers have successfully substituted
amides10 for the carboxylate of EXP-3174, the 5-car-
boxamides became our starting point for imidazole
substituent SAR. The primary carboxamide analogue 1
(Table 1), which possesses the identical 2-butyl and
4-chloro substituents found in losartan, showed moder-
ate activity at the AT1 receptor, but its ETA receptor
binding affinity was poor. Adjusting the imidazole C-2
substituent to propyl and the C-4 substituent to ethyl
resulted in compound 3, which possessed AT1 receptor
binding affinity within the range we sought, along with
moderate ETA receptor binding affinity.
Target compounds were prepared by variations of the
synthetic route described for compound 17 (Scheme 1).
At a minimum, compounds were characterized by H
NMR and LC/MS analysis. The preparation of boronic
acid intermediate A has been described.11 All imidazole
precursors were prepared according to literature pro-
cedures.6,12 The alkylation of imidazoles with inter-
mediate B or analogues generally gave a mixture of N-1
An N,N-dimethylamide (4) showed diminished AT1
receptor binding, but the N-methylamide (5) regained
much of the AT1 receptor affinity found in 3, while also
achieving the best ETA receptor binding of any com-
pound in the 20-hydrogen series. Replacement of the
5-carboxamide group by a methyl ester (6 vs 2) or an
acetyl group (7 vs 2) did not improve activity. Since our
aim was to keep molecular weight (volume) to a mini-
mum, we concluded that the imidazole substitution
patterns in 3 and 5 were optimal or close to optimal for
monocyclic imidazoles.
1
We also investigated fusion of the imidazole C-4 and
C-5 substituents using the (oxo)cycloheptimidazole
group reported by workers at Kotobuki Seiyaku.12d The
resultant compound (8, Table 1) had an AT1 receptor
binding affinity of 8 nM, approximately as good as the
best monocyclic compounds shown in Table 1. How-
ever, ETA receptor binding was somewhat poorer.
Table 1. Variation of imidazole substituentsa
Compd
R2
R4
R5
AT1
Ki (nM)
ETA
Ki (nM)
Scheme 1. Reagents and conditions: (a) pyridine, DMAP, 55 ꢀC, 16 h,
96%; (b) MEMCl, K2CO3, DMF, 0 ꢀC to rt, 14 h, 92%; (c) (i) n-BuLi,
THF, ꢁ90 ꢀC, 15 min, (ii) B(OMe)3, ꢁ90 ꢀC to 0 ꢀC, 1 h, (iii) HCl (aq),
rt, 1 h, 85%; (d) NBS, CCl4, (BzO)2, reflux, 8 h, 58%; (e) NaOMe,
DMF, rt, 3 h, 82%; (f) DIBAL-H, THF, 0 ꢀC, 2 h, 88%; (g) toluene/
EtOH (2:1), Na2CO3 (aq), compound A, Pd(Ph3P)4, 85 ꢀC, 2 h, 83%;
(h) NaBH4, MeOH, 0 ꢀC, 1 h, 95%; (i) CBr4, Ph3P, DMF, 0 ꢀC, 4 h,
72%; (j) K2CO3, DMF, compound B, rt, 24 h, 39%; (k) (i) HCl,
dioxane/EtOH (2:1), 60 ꢀC, 8 h, (ii) KOH, H2O/MeOH (1:2), 65 ꢀC, 30
h, 95% crude (l) CDI, THF, rt, 2 h, then NH3 (aq), rt, 10 min, 38%.
1
2
3
4
5
6
7
8
BuCl
Pr
Pr
Pr
Pr
Pr
Pr
Pr
CONH
24 (4)
21 (1)
7 (2)
51 (10)
13 (4)
23 (10)
40 (1)
8 (4)
415 (170)
40 (30)
55 (12)
55 (25)
16 (6)
460 (130)
40 (20)
105 (15)
2
Cl
Et
Et
Et
Cl
Cl
CONH2
CONH2
CONMe2
CONHMe
CO2Me
COMe
CH2CH2CH2CH2C(=O)
aStandard deviations are given in parentheses.

J. E. Tellew et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1093–1096
1095
SAR at the biphenyl 20 position was investigated initi-
Table 3. Variation of biphenyl 20 substituents in monocyclic seriesa
ally in the fused imidazole series (Table 2). In the work
that led to discovery of BMS-248360, a pyrrolidine-
based 20 substituent appeared to be optimal, providing
potent AT1 and ETA receptor affinity. The substituent
had little effect on AT1 receptor affinity, but it enhanced
ETA receptor affinity in BMS-248360 by approximately
40-fold.4 In the present fused imidazole series, addition
of this substituent (9, Table 2) did indeed dramatically
improve ETA receptor binding over the 20 unsubstituted
analogue 8. However, it also severely diminished AT1
receptor binding.
Compd
R5
R
AT1
Ki (nM)
ETA
Ki (nM)
16
17
18
19
20
21
5
22
23
24
CONH2
CONH2
CONH2
CONH2
CONH2
CONH2
CONHMe
CONHMe
CONHMe
CONHMe
Methoxymethyl
Methyl
Ethoxymethyl
Chloro
Ethyl
0.6 (0.2)
2 (1)
3 (1)
35 (2)
1.3 (0.4)
2 (0.6)
13 (3)
8 (3)
4 (2)
5 (2)
20 (6)
700 (300)
7 (4)
2-Fluoroethoxymethyl
H
8 (3)
This difference in behavior from BMS-248360 could be
related to the difference in the placement of the hydrogen-
bond acceptor groups on the two 40 heterocycles. In the
imidazolinone BMS-248360, the carbonyl oxygen is
directly attached to the 5-membered ring, whereas in the
imidazole series (both monocyclic and fused) there is an
intervening carbon atom. This difference probably
translates into a subtle difference in the optimal bi-
phenyl torsion angle for AT1 receptor-binding. A large
20 substituent might interact with the AT1 receptor in
such a way as to force the torsion angle out of the
optimal range for the imidazole but not for the imida-
zolinone. Alternately, a large 20 substituent might have a
direct steric interaction with the 40 substituent, probably
via the 40 heterocycle’s C-2 alkyl group. If so, then this
interaction is tolerated with the imidazolinone but not
with the imidazole.
16 (6)
7 (3)
5 (2)
Methoxymethyl
Methyl
Ethoxymethyl
4 (1)
14 (4)
22 (2)
aStandard deviations are given in parentheses.
binding, in addition to the expected improvement in
ETA receptor binding. For three of the best 20 sub-
stituents, the corresponding N-methylcarboxamide
compounds were also prepared (Table 3). These com-
pounds (22–24) were indistinguishable from the corres-
ponding primary carboxamides in ETA receptor
binding. However, AT1 receptor affinity diminished
when the 20 substituent was larger than methyl.
To screen for in vivo oral efficacy, we studied selected
compounds in a rat angiotensin II oral pressor assay
using conscious Sprague–Dawley rats.14 In this assay,
10 was only poorly efficacious. In contrast, 17 and 23
both had potency (area-over-the-curve (AOC) value and
maximum percent decrease of pressor response, Table 4)
similar to that of BMS-248360, or approximately half
that of irbesartan. Caco-2 cell permeability rates for
both 17 and 23 were relatively low (Table 4), suggesting
that poor absorption may be limiting their in vivo effi-
cacy. Nonetheless, these results suggest that further
optimization of this series could yield highly-potent,
orally active dual receptor antagonists.
Fortunately, smaller 20 substituents were better toler-
ated by the AT1 receptor (Table 2). The best substituent
in the fused imidazole series appeared to be methyl (10),
though other small lipophilic groups such as ethoxy-
methyl (14) and methoxymethyl (15) were similarly
effective. In the monocyclic imidazole carboxamide ser-
ies (Table 3), the same small 20 substituents were even
more effective, providing single digit nanomolar binding
at both receptors (17 vs 10; 16 vs 15). This series
demonstrated that a properly-chosen biphenyl 20 sub-
stituent can have a subtle positive effect on AT1 receptor
Table 2. Variation of biphenyl 20 substituents in fused imidazole
seriesa
In conclusion, we have combined the features of the
AT1 receptor antagonist losartan with those of
BMS-193884, an ETA-selective endothelin receptor
antagonist. The resultant 40-[(imidazol-1-yl)methyl]-
biphenylsulfonamides have potent in vitro activity.
Inclusion of certain 20 substituents on the biphenyl core
Table 4. In vivo evaluation of selected compoundsa
Compd
R
H
AT1
Ki (nM)
ETA
Ki (nM)
Compd
Rat AII pressor
effect po,
30 mmol/kg (AOC units)b
Max BP % Caco-2 Pc,
decrease
nm/s
pH 5.5/7.4
8
9
8 (4)
105 (15)
1 (0.6)
360 (140)
Irbesartan
BMS-248360
10
17
23
11,900
5800
1200
5300
6700
84 (4)
45 (14)
23 (13)
47 (17)
64 (6)
10
11
12
13
14
15
Methyl
Fluoro
Cyanomethyl
Hydroxymethyl
Ethoxymethyl
Methoxymethyl
2 (1)
16 (2)
16 (5)
30 (10)
5 (1)
13 (4)
20 (6)
15 (7)
14 (3)
20 (7)
17 (6)
71/40
59/37
aValues are means of four experiments, standard deviation is given in
parentheses.
bDefined as the integral from 0 to 240 min on the graph of % inhibi-
tion of pressor response versus time.
4 (1)
aStandard deviations are given in parentheses.

1096
J. E. Tellew et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1093–1096
of these compounds enhances both AT1 and ETA
receptor binding affinity. Compounds 17 and 23 are
orally active in a rat model of angiotensin-mediated
hypertension.
7. Wexler, R. R.; Greenlee, W. J.; Irvin, J. D.; Goldberg,
M. R.; Prendergast, K.; Smith, R. D.; Timmermans,
P. B. M. W.M. J. Med. Chem. 1996, 39, 625.
8. Webb, M. L.; Meek, T. D. Med. Res. Rev. 1997, 17, 17.
9. (a) Stein, P. D.; Hunt, J. T.; Floyd, D. M.; Moreland, S.;
Dickenson, K. E. J.; Mitchell, C.; Liu, E. C.-K.; Webb, M. L.;
Murugesan, N.; Dickey, J.; McMullen, D.; Zhang, R.; Lee,
V. G.; Serafino, R.; Delaney, C.; Schaeffer, T. R.; Kozlowski,
M. J. Med. Chem. 1994, 37, 329. (b) Stein, P. D.; Floyd, D. M.;
Bisaha, S.; Dickey, J.; Girotra, R.; Gougoutas, J. Z.;
Kozlowski, M.; Lee, V. G.; Liu, E. C.-K.; Malley, M. F.;
McMullen, D.; Mitchell, C.; Moreland, S.; Murugesan, N.;
Serafino, R.; Webb, M. L.; Zhang, R.; Hunt, J. T. J. Med.
Chem. 1995, 38, 1344.
Acknowledgements
We thank Anthony M. Marino of the Department of
Metabolism and Pharmacokinetics, Bristol-Myers
Squibb Pharmaceutical Research Institute, for per-
forming Caco-2 cell permeability assays.
10. (a) Santella, J. B., III; Duncia, J. V.; Ensinger, C. L.;
VanAtten, M. K.; Carini, D. J.; Wexler, R. R.; Chiu, A. T.;
Wong, P. C.; Timmermans, P. B. M. W. M. Bioorg. Med.
Chem. Lett. 1994, 4, 2235. (b) Yanagisawa, H.; Amemiya, Y.;
Kanazaki, T.; Shimoji, Y.; Fujimoto, K.; Kitahara, Y.; Sada,
T.; Mizuno, M.; Ikeda, M.; Miyamoto, S.; Furukawa, Y.;
Koike, H. J. Med. Chem. 1996, 39, 323. (c) Deprez, P.; Guil-
laume, J.; Becker, R.; Corbier, A.; Didierlaurent, S.; Fortin,
M.; Frechet, D.; Hamon, G.; Heckmann, B.; Heitsch, H.;
Kleeman, H.-W.; Vevert, J.-P.; Vincent, J.-C.; Wagner, A.;
Zhang, J. J. Med. Chem. 1995, 38, 2357.
References and Notes
1. (a) Burnier, M. Circulation 2001, 103, 904. (b) Ashton,
W. T. Exp. Opin. Invest. Drugs 1994, 3, 1105.
2. (a) Elliott, J. D.; Ohlstein, E. H.; Peishoff, C. E.; Ellens,
H. M.; Lago, M. A. Pharm. Biotechnol. 1998, 11, 113. (b)
Spieker, L. E.; Noll, G.; Ruschitzka, F. T.; Luscher, T. F.
J. Am. Coll. Cardiol. 2001, 37, 1493. (c) Dao, H. H.; Moreau,
P. Expert Opin. Invest. Drugs 2001, 10, 1937.
3. Workers at Merck have reported on a series of phenylacetic
acid derivatives with dual endothelin/angiotensin II receptor
antagonist activity: Walsh, T. F.; Fitch, K. J.; Williams, D. L.,
Jr.; Murphy, K. L.; Nolan, N. A.; Pettibone, D. J.; Chang,
R. S. L.; O’Malley, S. S.; Clineschmidt, B. V.; Veber, D. F.;
Greenlee, W. J. Bioorg. Med. Chem. Lett. 1995, 5, 1155.
4. Murugesan, N.; Tellew, J. E.; Gu, Z.; Kunst, B. L.; Fadnis,
L.; Cornelius, L. A.; Baska, R. A. F.; Beyer, S. M.; Mon-
shizadegan, H.; Dickinson, K.; Panchal, B.; Valentine, M. T.;
Chong, S.; Morrison, R. A.; Carlson, K. E.; Powell, J. R.;
Moreland, S.; Barrish, J. C.; Kowala, M. C.; Macor, J. E.
J. Med. Chem. 2002, 45, 3829.
5. Murugesan, N.; Gu, Z.; Stein, P. D.; Bisaha, S.; Spergel, S.;
Girotra, R.; Lee, V. G.; Lloyd, J.; Misra, R. N.; Schmidt, J.;
Mathur, A.; Stratton, L.; Kelly, Y. F.; Bird, E.; Waldron, T.;
Liu, E. C.-K.; Zhang, R.; Lee, H.; Serafino, R.; Abboa-Offei,
B.; Mathers, P.; Giancarli, M.; Seymour, A. A.; Webb, M. L.;
Moreland, S.; Barrish, J. C.; Hunt, J. T. J. Med. Chem. 1998,
41, 5198.
11. Murugesan, N.; Barrish, J. C.; Gu, Z.; Morrison, R. A.
PCT International Application WO 98/33780, 1998.
12. (a) Watson, S. P. Synth. Commun. 1992, 22, 2971. (b)
Carini, D. J. PCT International Application WO 92/00977,
1992. (c) Paul, R.; Brockman, J. A.; Hallett, W. A.; Hanifin,
J. W.; Tarrant, M. E.; Torley, L. W.; Callahan, F. M.; Fabio,
P. F.; Johnson, B. D.; Lenhard, R. H.; Schaub, R. E.; Wissner,
A. J. Med. Chem. 1985, 28, 1704. (d) Yanagisawa, T.;
Ueyama, N.; Kawai, T.; Sonegawa, M.; Baba, H.; Mochizuki,
S.; Kozakai, K.; Tomiyama, T. Bioorg. Med. Chem. Lett.
1993, 3, 1559.
13. Binding to the human ETA receptor was evaluated by
incubating test compounds with CHO-K1 cells expressing the
human ETA receptor in the presence of 0.05 nM 125I-labelled
endothelin 1 according to ref 4. The Kd for endothelin 1 was
0.05 nM in this assay. Compounds were evaluated for AT1
binding by incubation with rat aortic smooth muscle cells and
0.2 nM 125I-labeled Sar-Ile-angiotensin II according to ref 4.
The Kd for angiotensin II was 0.094 nM in this assay. Each
reported Ki value is the mean of at least three individual
experiments.
6. Carini, D. J.; Duncia, J. V.; Aldrich, P. E.; Chiu, A. T.;
Johnson, A. L.; Pierce, M. E.; Price, W. A.; Santella, J. B.,
III; Wells, G. J.; Wexler, R. R.; Wong, P. C.; Yoo, S.-E.;
Timmermans, P. B. M. W.M. J. Med. Chem. 1991, 34,
2525.
14. Artursson, P.; Borchardt, R. Pharm. Res. 1997, 14, 1655
Pure ETA receptor antagonists such as BMS-193884 are in-
active in this assay.