Design and SAR of selective T-type calcium channel antagonists containing a biaryl sulfonamide core

Article abstract of DOI:10.1016/j.bmcl.2007.11.103

T-type calcium channel antagonists were designed using a protocol involving the program SPROUT and constrained by a ComFA-based pharmacophore model. Scaffolds generated by SPROUT were evaluated based on their ability to be translated into structures that were synthetically tractable. From this exercise, a novel series of potent and selective T-type channel antagonists containing a biaryl sulfonamide core were discovered.

Full text of DOI:10.1016/j.bmcl.2007.11.103

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 474–478
Design and SAR of selective T-type calcium channel
antagonists containing a biaryl sulfonamide core
Jon J. Hangeland,* Daniel L. Cheney, Todd J. Friends, Stephen Swartz, Paul C. Levesque,
Adam J. Rich, Lucy Sun, Terry R. Bridal, Leonard P. Adam, Diane E. Normandin,
Natesan Murugesan and William R. Ewing
Bristol-Myers Squibb R& D, Princeton, NJ 08543-5400, USA
Received 17 October 2007; accepted 27 November 2007
Available online 3 December 2007
Abstract—T-type calcium channel antagonists were designed using a protocol involving the program SPROUT and constrained by a
ComFA-based pharmacophore model. Scaffolds generated by SPROUT were evaluated based on their ability to be translated into
structures that were synthetically tractable. From this exercise, a novel series of potent and selective T-type channel antagonists con-
taining a biaryl sulfonamide core were discovered.
Ó 2007 Elsevier Ltd. All rights reserved.
Calcium channels open in response to membrane depo-
larization causing an increase in intracellular calcium,
thereby activating many crucial physiological processes
such as muscle contraction, hormone secretion, neuro-
transmission, synaptic plasticity, regulation of
enzymatic activities and gene expression.1 Calcium
channels are widely distributed. For example, T-type
channels can be found in neurons, heart, kidney, smooth
and skeletal muscle, sperm, and endocrine tissues such
as the adrenal and pituitary glands and the pancreas.
T-type calcium channels are thought to be involved in
autonomic nervous functions, and in regulation of
cardiovascular activities such as heart rate, arterial and
venous smooth muscle innervation and tone.2
O
CH3
O
O
N
N
F
CH3
HN
1
O
N+
O–
O
H3C
H3C
O
P
O
N
O
Drugs such as Mibefradil3 and Efonidipine4 (1 and 2,
respectively, in Fig. 1), which selectively block T-type
over L-type calcium channels, have been shown to be
useful in treating a variety of disease states. For exam-
ple, Mibefradil has been shown to be useful for the treat-
ment of hypertension and angina, without showing the
side-effects associated with predominant L-type channel
blockers such as negative inotropy, reflex tachychardia,
vasoconstrictive hormone release or peripheral edema.5
Furthermore, multiple animal studies with Mibefradil
O
H3C
N
CH3
H
2
Cl
A
O
O
O
4 N
N
C
B
CH3
H3C
N
O
O
H
Keywords: T-type calcium channel antagonists; Molecular modelling;
Sulfonamide; COMFA; SPROUT.
3
Figure 1. Structures of Mibefradil3 (1), efonidipine4 (2), and com-
pound 3.10
*
Corresponding author. Tel.: +1 609 818 4958; fax: +1 609 818
3560; e-mail: jon.hangeland@bms.com
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.11.103

J. J. Hangeland et al. / Bioorg. Med. Chem. Lett. 18 (2008) 474–478
475
have demonstrated its potential to be cardioprotective.
Cl
For example, Mibefradil was shown to improve left
ventricular remodeling and reduce capillary density in
cardiomyopathic hamsters (CMH),6a to reduce myocar-
dial fibrosis in aldosterone treated rats,6b and to im-
prove myocardial function in dogs following coronary
perfusion pressure induced ischemia.6c Other studies
have pointed to its potential to be renal protective.
For example, treatment with Mibefradil caused a greater
increase in renal microvascular efferent blood flow ver-
sus afferent blood flow in dogs,7a reduced glomular dam-
age and protenuria in deoxycorticosterone acetate
(DOCA) treated hypertensive rats,7b,c and prevented
L-NAME-exacerbated nephrosclerosis in spontaneously
hypertensive rats.7d In addition, unlike L-type calcium
blockers, Mibefradil has been shown to be potentially
useful in the treatment of heart failure.8a For example,
Mibefradil improved cardiac function in rat model of
chronic heart failure8b and helped to normalize Ca2+ uti-
lization and improve inotropic effects of b-adrenergic
stimulation in hypertrophied rat myocardium.8c More
recently, both antagonists9a,b and dual action agonist/
antagonists9c of the T-type calcium channel have been
shown to have potent anti-proliferative activity against
a variety of cancer cell lines. In light of the potential in-
crease in benefit shown by mixed T-type and L-Type cal-
cium channel blockers over predominantly L-type
calcium channel blockers in the treatment of hyperten-
sion and other disorders, a program to discover proprie-
tary and selective T-type calcium channel blockers was
initiated.
X
R
N
N
CH3
H3C
O
O
4: R = H, X = CO (5%)
5: R = H, X = SO2 (30%)
6: R = CH2CO2CH(CH3)2
X = SO2 (17%)
Figure 2. Structures of compounds 4–6. The values in parentheses
represent percent inhibition at 1 lM inhibitor concentration in the T-
EP assay.16
model.12 The output contained >1000 hypothetical scaf-
folds that were categorized into families according to
their structural similarity. The largest families were eval-
uated for their ability to be translated from the hypo-
thetical scaffold provided by SPROUT into real,
synthetically accessible chemical structures suitable for
SAR evaluation. Two families satisfied this criterion;
one was translated into a biphenyl core (Fig. 3B) and
the other into a biaryl sulfonamide core (Fig. 3C).
Importantly, both provided an attachment point (de-
noted X in Fig. 3B and C) for the ester side chain C con-
tained in compound 3. Thus, these two cores provided
the ability to recapitulate and explore each of the key
binding elements, A–C, defined by the CoMFA model.12
To corroborate the choice of these two cores, an addi-
tional modeling exercise was carried out. First, small
molecule X-ray crystal structures from the public
domain containing either the biphenyl or biaryl sulfon-
amide core were surveyed. The most frequently observed
conformation for each was used to validate models of
the biphenyl series (e.g., structure 7, Fig. 4A) and of
the biaryl sulfonamide series (e.g., structure 8, Fig. 4B)
having key torsion angles in their lowest energy confor-
mation. Each validated structure was overlaid onto
compound 3, which was constrained according to the
CoMFA model used to construct the SPROUT query.
This study showed that both cores adopted low energy
conformations that displayed the key binding elements
A–C similarly to compound 3. However, the atom-for-
atom correspondence between the biphenyl core and
compound 3 was less than the correspondence for the
biaryl sulfonamide core and compound 3 (Fig. 4A and
B). In order to compare the two cores experimentally,
synthesis of examples of each was initiated.
The design of biaryl sulfonamide calcium channel antag-
onists grew out of an effort to simplify the core structure
of T-type calcium channel antagonists like compound 3
(Fig. 1; T-EP IC50 = 0.25 lM, RA IC50 = 2.1 lM),16,17
which contain a chiral dihydropyrimidinone (DHP).10
Comparative molecular field analysis (CoMFA)11 of
representative examples from the DHP series, including
compound 3, and related series suggested that the DHP
core functioned primarily to display peripheral binding
elements in the proper orientation, specifically rings A
and B and, to a lesser extent, the ester side chain C.12
The first attempt to simplify the core and to eliminate
the chiral center was carried out by straightforward
deletion of portions of the DHP core and the ester side
chain. This approach produced compounds 4 and 5
(Fig. 2, synthesis not shown) that were inactive at a con-
centration of 1.0 lM. Re-installation of the ester side
chain, and hence the chiral center, gave compound 6
(Fig. 2, synthesis not shown), which did not show im-
proved potency.
The first set of compounds evaluated for antagonist
activity against the T-type calcium channel were
designed to extend incrementally a phenyl ring away
from the core by varying the length of the linker to a
position that would be likely to overlap with ring B in
compound 3 (see Fig. 3A–C). In addition, a few exam-
ples in each series were designed to potentially extend
beyond ring B in compound 3. The selection of the
X-group was based on the preference for an isopropyl
ester in the DHP series as in compound 3.10 Thus, X
in the biphenyl series was chosen to be the carboxy
In an attempt to identify a new, more rigid core, a mod-
eling exercise was carried out using the program
SPROUT.13,14 The input was constructed from com-
pound 3, where the DHP core, the ester side chain,
and the first two methylene units of the side chain con-
˚
taining the biaryl ether were deleted, leaving a 5 A gap
to be filled in by SPROUT (Fig. 3A). The retained por-
tions of the structure, which contained binding elements
A and B, were constrained according to the CoMFA

476
J. J. Hangeland et al. / Bioorg. Med. Chem. Lett. 18 (2008) 474–478
A
B
A
X
B
LINKER
5
C
A
N
X
S
LINKER
B
O
O
Figure 3. (A) SPROUT input showing the portion of compound 3 retained where the atoms connected by SPROUT are depicted as purple spheres.
(B) Biphenyl series. (C) Biaryl sulfonamide series.
Cl
A
B
Cl
O
O
O
O
O
O
S
O
N
N
H
O
N
H
O
O
7
8
Figure 4. (A) Overlay of compound 3 (magenta) with structure 7 (salmon). (B) Overlay of compound 3 (magenta) with structure 8 (salmon).
Table 1. T-type patch clamp and rat aorta strip data
isopropyl group, whereas it was chosen to be the isopro-
pyl acetate group in the biaryl sulfonamide series (Table
1).
a
Compound
R
IC50 (lM)
T-EP16 RA17
(6)
RA/
T-EP
12a
12b
16a
16b
16c
16d
16e
CH2CH2Ph
Synthesis of the biphenyl series (Scheme 1) began with
conversion of 2-amino-4-nitrobenzoic acid (9) to 2-
iodo-4-nitrobenzoic acid.15 The resulting iodobenzoic
acid was converted to the isopropyl ester and the nitro
group was reduced to provide isopropyl 4-amino-2-
iodobenzoate (10). Coupling of 10 with 3-chlorophenyl
boronic acid provided the biphenyl aniline 11, which
was condensed with a diverse set of carboxylic acids,
CH2N(CH3)CO2CH2Ph (24)
CH3
(0)
CH2Ph
CH2CH2Ph
(40)
0.54
>30
15
>55
62
CH2N(CH3)CO2CH2Ph 0.24
CH2CH2NHCO2CH2Ph 0.42
17
41
a Values contained in parentheses are percent inhibition at 1 lM
inhibitor concentration.

J. J. Hangeland et al. / Bioorg. Med. Chem. Lett. 18 (2008) 474–478
477
Cl
Cl
NO2
O
I
O
O
a-c
d
e
HO
O
O
O
O
O
NH2
NH2
NH2
N
H
R
9
10
11
12a-b
Scheme 1. Synthesis of the biphenyl series. Reagents and conditions: (a)15 NaNO2, 9 N H2SO4, NaI, urea (50%); (b) isopropanol, cat. H2SO4, reflux
(66%); (c) Fe°, HOAc, H2O, rt (80%); (d) 3-chlorophenyl boronic acid, PdCl2(dppf), K3PO4, dioxane, 100 °C (67%); (e) diverse acids, EDCI, HOAt,
N-Me-morpholine, CH2Cl2/DMF 5:2, rt (50–90%).
Cl
Cl
Cl
d
O
S
a-c
O
O
O
Cl
O
+
N+
O–
N
N
O
S
N
H
R
O
S
NH2
O
NH2
O
O
O
O
13
14
15
16a-e
Scheme 2. Synthesis of the biaryl sulfonamide series. Reagents and conditions: (a) TEA, CH2Cl2, rt (85%); (b) isopropyl bromoacetate, NaH, DMF,
rt (50%); (c) Fe°, HOAc, H2O, rt (90%); (d) diverse acids, EDCI, HOAt, 4-Me-morpholine, CH2Cl2, rt, 12 h (40–70%).
herein exemplified by compounds 12a and 12b. Synthesis
of the biaryl sulfonamide series (Scheme 2) was initiated
by coupling 3-chloroaniline (13) with 3-nitrophenyl-sul-
fonyl chloride (14) followed by sequential alkylation of
the sulfonamide nitrogen with isopropyl bromoacetate
and reduction of the nitro group to give aniline 15,
which was coupled with a diverse set of acids to provide
compounds 16a–e. Synthesis of compounds 17a–g (see
Table 2), where the isopropyl acetate group is replaced
by other side chains, was accomplished by straightfor-
ward extension of the chemistry used to synthesize com-
pounds 16a–e.
73% inhibition, respectively, at 1 lM concentration,
and had IC50 values equal to 0.24 and 0.42 lM, respec-
tively. Importantly, both were selective for the T-type
channel (62- and 42-fold, respectively). For comparison,
Mibefradil inhibits the T-type and L-type channels with
IC50 values of 0.13 and 7 lM, respectively (data for each
obtained by patch clamp technique),3 whereas the T-EP
and RA values for the individual enantiomers of com-
pound 3 were 0.18 and 2.1 lM and 0.65 and 3.8 lM,
respectively.18 Importantly, this series provided a clear
SAR, which was absent from the biphenyl series. For
example, compounds 16a–16c, which contain aliphatic
side chains of increasing length, showed incremental
improvement in antagonist potency from inactive (16a)
to having an IC50 = 0.54 lM (16c). While the SAR is
limited to relatively few examples, there was also an
indication that there exists an optimal R-group length,
which is suggested by comparing compounds 16c–e
(T-EP IC50 = 0.54, 0.24, and 0.42 lM, respectively).
Compounds were first evaluated for their ability to inhi-
bit the T-type calcium channel at a concentration of
1.0 lM using a patch clamp assay (T-EP).16 T-EP IC50
values were determined for compounds that gave per-
cent inhibition P50% at this concentration. Compounds
with IC50 values <1 lM were further evaluated in a rat
aorta strip assay (RA),17 which served to measure func-
tional L-type calcium channel antagonist activity. Selec-
tivity of the compounds for the T-type calcium channel
was estimated by calculating the ratio (RA IC50)/(T-EP
IC50).
Additional SAR studies (Table 2) with the biaryl sulfon-
amide series focused on replacing the acetic ester side
chain (X in Fig. 3C). Replacing the isopropyl ester of
16d with a carboxylic acid (17a) or ethyl ester (17b)
resulted in compounds with significantly lower potency.
Isosteric replacement of the isopropyl acetate of 16d
with a purely aliphatic side chain (17c) gave a compound
equal in potency (IC50 = 0.18 lM, 67-fold T-type selec-
tive), whereas replacement of the isopropyl ester with
N-isopropyl amide (17d) reduced the potency by almost
10-fold. Addition of a basic side chain (17f) resulted in a
substantial loss in potency. The most potent and selec-
tive compound discovered in this series (17e;
IC50 = 0.11 lM, 109-fold T-type selective) contained a
benzyl group in place of the isopropyl acetic acid side
chain.
The first set of compounds from the biphenyl series
showed poor inhibition. Two examples, compounds
12a and 12b, are representative of the entire set, and
showed only 6% and 24% inhibition at 1 lM concentra-
tion, respectively. The percent inhibition of all other
compounds submitted from the biphenyl series fell with-
in this range, and no clear SAR was apparent. This ser-
ies was discontinued as a result. In contrast, the first set
of compounds submitted from the biaryl sulfonamide
series contained potent T-type channel antagonists.
For example, compounds 16d and 16e gave 78% and

478
J. J. Hangeland et al. / Bioorg. Med. Chem. Lett. 18 (2008) 474–478
Table 2. T-type patch clamp and rat aorta strip data
Saruta, T. J. Hypertens. 2001, 19, 2031; (b) Baylis, C.; Qiu,
C.; Engels, K. Am. J. Kidney Dis. 2001, 38, 1292; (c)
Karam, H.; Clozel, J.-P.; Bruneval, P.; Gonzalez, M.-F.;
Cl
´
Menard, J. Hypertension 1999, 34, 673; (d) Qiu, C.;
Bruneval, P.; Roeckel, A.; Heudes, D.; Van Huyen, J.-P.
D.; Roux, S. J. Hypertens. 1999, 17, 1489.
8. For a review see: (a) Clozel, J.-P.; Ertel, E. A.; Ertel, S. I.
Proc. Assoc. Am. Phys. 1999, 111, 429; (b) Mulder, P.;
Vincent, R.; Compagnon, P.; Henry, J.-P.; Lallemand, F.;
O
CH3
N
N
O
X
S
N
H
O
O
O
´
Clozel, J.-P.; Koen, R.; Mace, B.; Thuillez, C. J. Am. Coll.
Cardiol. 1997, 29, 416; (c) Min, J.-Y.; Meissner, A.; Wang,
J.; Morgan, J. P. Exp. Biol. Med. 2002, 227, 336.
9. (a) McCalmont, W. F.; Heady, T. N.; Patterson, J. R.;
Lindenmuth, M. A.; Haverstick, D. M.; Gray, L. S.;
Macdonald, T. L. Bioorg. Med. Chem. Lett. 2004, 14,
3691; (b) Lee, J. Y.; Seong, J. P.; Sung, J. P.; Lee, M. J.;
Rhim, H.; Seo, S. H.; Kim, K.-S. Bioorg. Med. Chem.
Lett. 2006, 16, 5014; (c) McCalmont, W. F.; Rapperson, J.
R.; Lindenmuth, M. A.; Heady, T. N.; Haverstick, D. M.;
Gray, L. S.; Macdonald, T. L. Bioorg. Med. Chem. 2005,
13, 3821.
a
Compound
X
IC50 (lM) RA/
T-EP
T-EP16 RA17
16d
17a
17b
17c
17d
17e
17f
CH2CO2i-Pr
CH2CO2H
CH2CO2Et
0.24
(11)
(35)
0.18
1.7
15
62
CH2CH2CH2CH(CH3)2
CH2COHNi-Pr
CH2Ph
12
67
>30 >17
12 109
0.11
CH2CON(H)CH2CH2N(CH3)2 (22)
10. Murugesan, N.; Gu, Z.; Fadnis, L.; Swartz, S.; Levesque,
P.; Bridal, T.; Sun, L.; Rich, A.; Ewing, W.R. unpublished
results.
a Values contained in parentheses are percent inhibition at 1 lM
inhibitor concentration.
11. For a review see: Kaminski, J. J. Adv. Drug Delivery Rev.
1994, 14, 331.
Application of the design paradigm described in this
paper resulted in the discovery of novel, potent T-type
calcium channel antagonists, the biaryl sulfonamides,
which are selective for the T-type calcium channel.19
The successful translation from design to identification
of potent antagonists may be attributed to combining
the CoMFA constrained SPROUT design work with
validation of the low energy conformations through
comparison with X-ray crystal structures of small mole-
cules containing the biaryl sulfonamide core. Thus, the
biaryl sulfonamides disclosed herein provide an addi-
tional chemotype that may be used to discover T-type
selective antagonists for the treatment of cardiovascular
and other disorders.
12. Cheney, D. L.; Hangeland, J. J.; Friends, T. J.; Swartz, S.;
Levesque, P. C.; Rich, A. J.; Sun, L.; Bridal, T.R.; Adam,
L.P.; Normandin, D. E.; Murugesan, N.; Ewing, W. R. J.
Comp. Aid. Mol. Des. submitted for publication.
13. NOTE: The stereochemistry of compound 3, and its
analogs, used in this exercise was arbitrarily chosen to be
the S-configuration at C-4 of the DHP ring.
14. Gillet, V. J.; Newell, W.; Mata, P.; Myatt, G.; Sike, S.;
Zsoldos, Z.; Johnson, A. P. J. Chem. Inf. Comput. Sci.
1994, 34, 207.
15. Katritzky, Alan R.; Savage, Paul G.; Gallos, John K.;
Durst, Dupont H. Org. Prep. Proced. Int. 1989, 21,
157.
16. T-type calcium channel assay conditions (T-EP): Whole cell
currents were recorded from a HEK-293 stable cell line
transfected with the human a1H clone using standard whole
cell patch clamp techniques. Cells were held at À80 mV and
stepped to À30 mV at 0.1 Hz frequency to activate inward
Ca2+ current using 5 mM Ca2+ as the charge carrier. All
records were obtained at room temperature (21 °C). Data
were digitized at 5 kHz and sampled at 10 kHz. Peak inward
currents were measured before and after addition of
compound to the bathing solution.
Acknowledgment
The authors would like to thank Dr. Guixue Yu for his
helpful suggestions and encouragement throughout the
course of this work.
References and notes
17. Rat aorta strip assay conditions (RA): Aortas from rats
were dissected, cut into rings, and the endothelium
denuded by rubbing. The rings were then attached to
force transducers in a tissue bath containing a physiolog-
ical saline solution. The rings were incubated with KCl to
achieve a near maximal contraction, and test compounds
were added at increasingly higher doses to determine the
IC50 for inhibition of force.
18. NOTE: The absolute configurations for the individual
enantiomers of compound 3 were not determined.
19. Additional selective T-type calcium channel antagonists
have been reported since the completion of this work: (a)
Ku, I. W.; Cho, W.; Doddareddy, M. R.; Jang, M. S.;
Keum, G.; Less, J.-H.; Chung, B. Y.; Kim, Y.; Rhim, H.;
Kang, S. B. Bioorg. Med. Chem. Lett. 2006, 16, 5244; (b)
Kim, H. S.; Kim, Y.; Doddareddy, M. R.; Seo, S. H.;
Rhim, H.; Tae, J.; Pae, A. N.; Choo, H.; Cho, Y. S.
Bioorg. Med. Chem. Lett. 2007, 17, 476.
1. Catterall, W. A. Annu. Rev. Cell Dev. Biol. 2000, 16, 521.
2. Perez-Reyes, E. Physiol. Rev. 2003, 83, 117.
3. Clozel, J.-P.; Ertel, E. A.; Ertel, S. I. J. Hypertens. 1997,
15(Suppl. 5), S17.
4. (a) Masuda, Y.; Tanaka, S. Cardiovasc. Drug Rev. 1994,
12, 123; (b) Tanaka, H.; Shigenobu, K. Cardiovasc. Drug
Rev. 2002, 20, 81.
5. (a) Sandmann, S.; Unger, T. J. Clin. Basic Cardiol. 1999,
2, 187; (b) Glasser, S. P. J. Clin. Pharmacol. 1998, 38, 659.
6. (a) Villame, J.; Massicotte, J.; Jasmin, G.; Dumont, L.
Cardiovasc. Drugs Ther. 2001, 15, 41; (b) Ramires, F. J.
A.; Sun, Y.; Weber, K. T. J. Mol. Cell. Cardiol. 1998, 30,
475; (c) Clozel, J.-P.; Danken, L.; Oserrieder, W.
J. Cariovasc. Pharm. 1989, 14, 713.
7. (a) Honda, M.; Hayashi, K.; Matsuda, H.; Kubota, E.;
Tokuyama, H.; Okubo, K.; Takamatsu, I.; Ozawa, Y.;