A new class of bradykinin 1 receptor antagonists containing the piperidine acetic acid tetralin core

Article abstract of DOI:10.1016/j.bmcl.2006.01.069

The bradykinin 1 (B1) receptor is upregulated during times of inflammation and is important for maintaining inflamed and chronic pain states. Blocking this receptor has been shown to reverse and/or ameliorate pain and inflammation in animal models. In this report, we describe a new class of B1 receptor antagonists that contain the piperidine acetic acid tetralin core. A structure-activity relationship for these analogs is described in this paper. The most potent compounds from this class have IC₅₀s < 20 nM in a B1 receptor functional assay. One of these compounds, 13g, shows modest oral bioavailability in rats.

Full text of DOI:10.1016/j.bmcl.2006.01.069

Bioorganic & Medicinal Chemistry Letters 16 (2006) 2071–2075
A new class of bradykinin 1 receptor antagonists containing
the piperidine acetic acid tetralin core
Christopher Fotsch,a,* Gloria Biddlecome,b Kaustav Biswas,a Jian Jeff Chen,a
Derin C. D’Amico,a Robert D. Groneberg,d Nianhe Bruce Han,a Feng-Yin Hsieh,c
Augustus Kamassah,b Gondi Kumar,c Dianna Lester-Zeiner,b Qingyian Liu,a
David A. Mareska,d Babak Bobby Riahi,a Yueh-Ju Judy Wang,b Kevin Yang,a
James Zhan,d Joe Zhu,a Eileen Johnson,b Gordon Ngb and Benny C. Askewa
aAmgen Inc., Department of Chemistry Research and Development, One Amgen Center Drive, Thousand Oaks, CA 91360, USA
bAmgen Inc., Department of Neurosciences, One Amgen Center Drive, Thousand Oaks, CA 91360, USA
cAmgen Inc., Department of Pharmacokinetics and Drug Metabolism, One Amgen Center Drive, Thousand Oaks, CA 91360, USA
dArray BioPharma, 3200 Walnut Street, Boulder, CO 80301, USA
Received 21 December 2005; revised 16 January 2006; accepted 17 January 2006
Available online 7 February2006
Abstract—The bradykinin 1 (B1) receptor is upregulated during times of inflammation and is important for maintaining inflamed
and chronic pain states. Blocking this receptor has been shown to reverse and/or ameliorate pain and inflammation in animal
models. In this report, we describe a new class of B1 receptor antagonists that contain the piperidine acetic acid tetralin core. A
structure–activity relationship for these analogs is described in this paper. The most potent compounds from this class have
IC50s < 20 nM in a B1 receptor functional assay. One of these compounds, 13g, shows modest oral bioavailability in rats.
Ó 2006 Elsevier Ltd. All rights reserved.
The kinins and their cognate receptors play a prominent
role in pain and inflammation. The bradykinin subtype
2 (B2) receptor is constitutively expressed in the central
and peripheral nervous system, vascular and tissue
endothelium, and on a number of inflammatory cells.1,2
In contrast, the bradykinin subtype 1 (B1) receptor is
only expressed at very low levels or not at all under basal
conditions, but it is rapidly upregulated following tissue
injury and during inflammation.3–5 Although the B1 and
B2 receptors have similar signaling cascades, they have
very different patterns of regulation. When the B2
receptor is activated by its ligands, bradykinin (BK) or
kallidin, it is rapidly desensitized and internalized.6
The B1 receptor, on the other hand, is thought to
maintain chronic inflammation and pain, since the
receptor is not desensitized and internalized when
activated by its ligands, des-Arg(9)-BK or des-Arg(10)-
kallidin.1 The B1 receptor, therefore, is an attractive
drug target since it is important for maintaining
inflammation and chronic pain states.
Several groups have demonstrated that peptide antago-
nists to the B1 receptor inhibit inflammation and pain in
animal models. In rats for example, B1 receptor peptide
antagonists reversed or prevented the development of
hyperalgesia induced by Freund’s adjuvant, UV light,7
and zymosan.8 In the mouse streptozotocin-induced
diabetic model, a B1 receptor peptide antagonist attenu-
ated thermal hyperalgesia.9 A systemically administered
B1 receptor peptide antagonist also inhibited nocicep-
10
tive spinal reflex in hyperalgesic rabbits induced with
Freund’s adjuvant. More recently, in a carrageenan-
induced hyperalgesia model, a subcutaneous injection
of a B1 peptide antagonist (des-Arg10-[Leu9]-kallidin)
reversed mechanical hyperalgesia.11
In the past few years, non-peptide, small molecule
antagonists to the B1 receptor have appeared in the lit-
erature. Compounds from these reports inhibited noci-
ceptive spinal reflex in rabbits,12 hyperalgesia in mice,13
and prevented neuropathic pain in rats.14 This last study
used a compound named SSR-240612 (Sanofi), which is
Keywords: Bradykinin 1; Antagonists; Tetralin.
*
Corresponding author. Tel.: +1 8054477746; fax: +1 8054801337;
e-mail: cfotsch@amgen.com
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.01.069

2072
C. Fotsch et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2071–2075
reported to be in Phase I clinical trials for the treatment
of chronic pain.
structure hoping to bias a bioactive conformation. To
that end, we replaced the b-phenylalanine with a piper-
idine-2-acetic acid group and the benzyl amine group
with a tetralin ring system, which resulted in a com-
pound with two less rotatable bonds. Besides adding
conformational constraints, we removed two of the
phenyl rings and alkylated the terminal amine. The re-
sult was compound 2, which was threefold more potent
(IC50 = 1.5 lM) than compound 1. Encouraged by these
results, we prepared additional analogs in this series,
which we describe in this publication.16
Given the evidence supporting the involvement of B1 in
inflammation and pain, we launched an effort to identify
orally available non-peptide B1 antagonists. In this re-
port, we describe the discovery and SAR of a new class
of non-peptide antagonists exemplified by compound 2.
A pharmacophore found on many non-peptide B1
receptor antagonists contains an aryl sulfonamide, a
linker group, and a basic group.15 A compound identi-
fied early in our research program has a similar motif
(see Fig. 1). Compound 1 contains a naphthyl sulfon-
amide attached to a b-phenylalanine (linker group) that
is appended to a benzyl amine (basic group). This com-
pound had an IC50 = 4.8 lM in our human B1 function-
al assay (calcium influx). To improve the potency of this
compound, we added conformational constraints to the
A synthetic approach that provided analogs of 2 is out-
lined in Scheme 1. The tetralin portion was constructed
from commercially available 6-methoxy tetralone (3).
Demethylation of 3 using HBr provided phenol 4, which
was converted to the corresponding triflate 5. Carbonyl-
ation of 5 with carbon monoxide in the presence of meth-
anol gave methyl ester 6. CBS reduction17 of 6 gave the
S-alcohol 7 in >99% ee. Treating alcohol 7 with diphe-
nylphosphoryl azide and diazobicycloundecene gave
the R-azide 8 with inversion.18 Exhaustive reduction of
8 with LiAlH4 gave the amino alcohol 9. This product
was coupled with the commercially available Boc-pro-
tected R-piperidine acetic acid (10) to give 11. Benzylic
oxidation of 11 with manganese dioxide followed by
reductive amination gave compound 12. Deprotection
of 12 gave an intermediate that was sulfonylated in the
last step to give an arylsulfonamide. Compounds 13a–q
all have the tert-butyl amine on the tetralin, but the aryl-
sulfonamide group is varied. Compounds 14a–j hold
O
O
S
N
H
O
S
N
H
O
HN
N
O
O
N
H
NH2
2
1
IC50 = 1500 nM
IC50 = 4800 nM
Figure 1. Design of a conformationally constrained analog.
O
O
HO
O
O
O
O
a
c
b
d
OCH3
OH
OTf
OMe
OMe
3
4
5
6
7
e
H2N
N3
O
O
f
N
OH
O
OH
O
OMe
10
9
8
g
Ar
O
S
O
O
H
N
O
O
N
O
N
H
H
N
N
N
O
j, k
h, i
O
O
R2
N
R2
N
R1
OH
R1
11
12
13, R1 = tBu, R2 = H, Ar = varied
14, R1 and R2 varied, Ar = 3CF3Ph
Scheme 1. Reagents and conditions: (a) 48% HBr (88%); (b) Tf2O, Et3N, CH2Cl2 (quant.); (c) 2 mol% Pd(OAc)2, 2 mol% dppp, CO, DMF, MeOH,
65 °C (65%); (d) 5 mol% (R)-methyl CBS reagent, 1.2 equiv BH3ÆDMS, toluene, À20 °C (81%); (e) 1.2 equiv. DPPA, DBU, toluene 0 °C (92%); (f)
2.1 equiv LiAlH4, THF (73%); (g) HOBT, EDC, DMF; (h) MnO2, THF; (i) NaBH(OAc)3, amine, DCE; (j) HCl, EtOAc; (k) ArSO2Cl, Et3N,
CH2Cl2.

C. Fotsch et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2071–2075
2073
constant the meta-trifluoromethylphenylsulfonamide
while the amine groups are varied.
We also replaced the aryl ring with a set of heteroaryl
compounds (see Table 2). The benzothiophene analog,
13m, was the most potent compound in this set with an
IC50 = 66 nM, while the benzthiadiazole analog, 13n,
was threefold less potent than 13m. The potency of the
quinoline and isoquinoline analogs, 13o and 13p, was
affected by the position of nitrogen. The quinoline analog
hadan IC50 = 440 nM. However, the isoquinoline analog,
13p, which differs from 13o only by a small change in the
position of the ring nitrogen, was 10-fold less potent than
13o. The N-methylimidazole analog, 13q, had significant-
ly reduced activity (IC50 > 40 lM).
Our first step was to explore the structure–activity rela-
tionship (SAR) of the aryl group on the sulfonamide
(see Table 1).19 Analogs were prepared that contained
substituents at the 4-position of the aryl ring (13a–f).
The most potent compound in this set contained the
4-methyl group, compound 13a (IC50 = 130 nM). Ana-
logs with larger groups at the 4-position were tolerated,
but none were as potent as 13a (see analogs 13b–d).
Compounds with smaller groups at the 4-position had
even higher IC50s. The 4-chloro analog, 13e, was sixfold
less potent than 13a, which suggests that potency was
more affected by electronics than by size since the methyl
group and chloro group have similar van der Waals
radii. Even the analog with the 4-methoxy group, 13f,
was 10-fold less potent than 13a, which was surprising
since the 4-butoxy analog, 13d, was only fivefold less
potent than 13a.
We then explored alternatives to the NH-tert-butyl
group on 13g by preparing a group of analogs with var-
ious other amines on the tetralin ring (see Table 3). Ana-
logs containing branched alkyl groups were 3- to 18-fold
less potent than 13g (see compounds 14a–c). Com-
pounds that either had an amine with an electron-with-
drawing or an electron-donating group were also
prepared. The trifluoroethyl analog, 14d, was 54-fold
less potent than 13g, but the methoxyethyl analog,
14e, was only fourfold less potent than 13g. Compounds
containing amines with small cycloalkyl groups
(14f–14h) all had similar potencies (IC50 = 30–50 nM).
Exploring substitutions on the 3-position we found that
the 3-chloro analog, 13h, had an IC50 = 73 nM, and the
3-trifluoromethyl analog, 13g, had an IC50 = 13 nM,
which was more than 180-fold more potent than com-
pound 2. To see if potency could be improved by adding
the 4-methyl group to these two analogs, we prepared
compounds 13i and 13k. No improvement in potency
was seen for the 3-trifluoromethyl-4-methyl analog,
compound 13k (IC50 = 16 nM vs 13 nM for 13a). How-
ever, the 3-chloro-4-methyl analog, 13i, was threefold
more potent (IC50 = 22 nM) than the 3-chloro analog
13h.
Table 2.
Ar
O
O
S
N
H
N
O
Table 1.
N
H
X
O
S
Compound
Ar
hB1 binding
a
IC50 (nM)
hB1 functional
a
IC50 (nM)
H
O
N
N
O
13m
14
1
66
8
S
N
S
N
N
13n
13o
13p
13q
150 20
220 50
610 130
>100,000
180 70b
440 250c
4300 110
>40,000
H
Compound
X
hB1 binding
a
IC50 (nM)
hB1 functional
a
IC50 (nM)
N
2
H
4-Me
1700 220
93 27
2400 1200
130 50
210 30
540 140c
590 70
810 150
1400 300b
13 8b
13a
13b
13c
13d
13e
13f
13g
13h
13i
13k
4-CF2CF3
4-Ph
360 30
690 80b
1200 260b
560 70
780 170b
23 8b
4-O(CH2)3CH3
4-Cl
4-OMe
N
3-CF3
3-Cl
N
N
120 30
15 5b
73 10
22 9b
3-Cl, 4-Me
3-CF3, 4-Me
11
3
16
7
a Values reported represent an average of two determinations for each
compound unless otherwise noted.
b n = 4.
c n = 3.
a Values reported represent an average of two determinations for each
compound unless otherwise noted.
b n = 4.
c n = 3.

2074
C. Fotsch et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2071–2075
Table 5. In vitro potency and rat pharmacokinetics of compounds
Table 3.
13g21
hB1
binding
IC50 (nM) IC50 (nM)
hB1
t1/2 Cla
Vssa
(mL kgÀ1
Oral
%Fb
a
O
O
functional (h) (mL hÀ1 kgÀ1
)
)
F3C
S
N
H
N
23 13
a Dosed iv (1 mg kgÀ1) in DMSO.
b Dosed po (10 mg kgÀ1) in 2% HPMC/1% Tween 80 in water.
8
8
2
2200
5400
19%
O
R
pound 15a. This single modification from a methylene to
an oxygen in the right-hand ring system resulted in a
compound that was ca. sevenfold less potent than 13g.
Contracting the ring size from a piperidine to pyrroli-
dine ring resulted in compound 15b, which was 10-fold
less potent than 15a.
Compound R
hB1 binding hB1 functional
a
IC50 (nM) IC50 (nM)
a
13g
14a
14b
14c
14d
14e
14f
14g
14h
14i
NH-tert-butyl
NH-iso-propyl
23 8b
72
13 8b
110 30
230 70
42 18b
1
NHCH2-tert-Butyl
NH-iso-butyl
NH–CH2CF3
59 12
13 2b
In summary, we have prepared several compounds with
low-nanomolar IC50s in the human B1 receptor func-
tional assay. In addition, all compounds were selective
for the B1 receptor since the human B2 receptor IC50s
were all >20,000 nM. In this series of compounds, a
bulky electron-withdrawing group on the 3-position on
the arylsulfonamide is preferred. Small alkyl and halo
groups are tolerated at the 4-position, but larger groups
are less potent. On the amine portion of this series, the
groups preferred were secondary amines with branched
alkyl groups. Less preferred were those amines with
three carbons or fewer. Changes to the tetralin also
resulted in less potent compounds.
1900 1500c 700 480d
NH–CH2CH2OCH3
NH–CH2-cyclo-propyl 16
21 5b
46 12
31
47 18
54
160 43
1
2
5
NH-cyclo-propyl
NH-cyclo-butyl
N(CH3)2
48
58 14
62 13
4
14j
Azetidine
1100 230 4300 340
a Values reported represent an average of two determinations for each
compound unless otherwise noted.
b n = 4.
c n = 32.
d n = 188.
Finally, we examined the pharmacokinetics of 13g in
rats (see Table 5) to determine how this compound
might perform in vivo. This compound had a half-life
of 2 h and oral bioavailability of 19%. The large volume
of distribution (>5 L kgÀ1) suggests that the compound
may reach the peripheral tissue where the B1 receptor
would be expressed during times of inflammation. Our
efforts to modify this compound to improve the potency
and pharmacokinetic profile will be reported in future
publications.
Table 4.
O
F3C
S
N
O
H
N
O
X
(
)
n
N
H
References and notes
Compound
n
X
hB1 binding
a
IC50 (nM)
hB1 functional
a
IC50 (nM)
1. Marceau, F.; Regoli, D. Nat. Rev. Drug Discov. 2004, 3,
845.
2. Howl, J.; Payne, S. J. Expert Opin. Ther. Targets 2003, 7,
277.
3. Calixto, J. B.; Medeiros, R.; Fernandes, E. S.; Ferreira, J.;
Cabrini, D. A.; Campos, M. M. Br. J. Pharmacol. 2004,
143, 803.
4. Dray, A. Can. J. Physiol. Pharmacol. 1997, 75, 704.
5. Pesquero, J. B.; Araujo, R. C.; Heppenstall, P. A.; Stucky,
C. L.; Silva, J. A., Jr.; Walther, T.; Oliveira, S. M.;
Pesquero, J. L.; Paiva, A. C. M.; Calixto, J. B.; Lewin, G.
R.; Bader, M. Proc. Natl. Acad. Sci. U.S.A. 2000, 97,
8140.
6. Marceau, F.; Bachvarov, D. R. Clin. Rev. Allergy Immun.
1998, 16, 385.
7. Perkins, M. N.; Campbell, E.; Dray, A. Pain 1993, 53, 191.
8. Belichard, P.; Landry, M.; Faye, P.; Bachvarov, D. R.;
Bouthillier, J.; Pruneau, D.; Marceau, F. Immunopharma-
cology 2000, 46, 139.
13g
15a
15b
2
2
1
CH2
O
23
8
13
95 40
8
160 30c
O
1500 10b
960 90
a Values reported represent an average of four determinations for each
compound unless otherwise noted.
b n = 2.
c n = 6.
Finally, two analogs containing smaller tertiary amines
were tested. Dimethylamine derivative, 14i, was 12-fold
less potent than 13g, and the analog with the azetidine
group, 14j, was 330-fold less potent than 13g.
Derivatives containing alternatives to the piperidine and
tetralin were also investigated (see Table 4). Replacing
the tetralin ring on 13g with a chroman ring20 gave com-
9. Gabra, B. H.; Sirois, P. Neuropeptides 2003, 37, 36.

C. Fotsch et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2071–2075
2075
10. Mason, G. S.; Cumberbatch, M. J.; Hill, R. G.; Rupniak,
N. M. J. Can. J. Physiol. Pharmacol. 2002, 80, 264.
11. Manning, B. H.; Dong, H.; Magal, E.; Johnson, E. J.
Inflammatory pain in the conscious rabbit: a new, efficient
method for measuring mechanical hyperalgesia in the hind
paw. Abstract of papers, 11th World Congress on Pain,
Seattle: IASP Press, 2005.;
12. Kuduk, S. D.; Ng, C.; Feng, D.-M.; Wai, J. M. C.; Chang,
R. S. L.; Harrell, C. M.; Murphy, K. L.; Ransom, R. W.;
Reiss, D.; Ivarsson, M.; Mason, G.; Boyce, S.; Tang, C.;
Prueksaritanont, T.; Freidinger, R. M.; Pettibone, D. J.;
Bock, M. G. J. Med. Chem. 2004, 47, 6439.
13. Fox, A.; Kaur, S.; Li, B.; Panesar, M.; Saha, U.; Davis, C.;
Dragoni, I.; Colley, S.; Ritchie, T.; Bevan, S.; Burgess, G.;
McIntyre, P. Br. J. Pharmacol. 2005, 144, 889.
14. Gougat, J.; Ferrari, B.; Sarran, L.; Planchenault, C.;
Poncelet, M.; Maruani, J.; Alonso, R.; Cudennec, A.;
Croci, T.; Guagnini, F.; Urban-Szabo, K.; Martinolle, J.-
P.; Soubrie, P.; Finance, O.; Le Fur, G. J. Pharmacol. Exp.
Ther. 2004, 309, 661.
17. Corey, E. J.; Bakshi, R. K.; Shibata, S.; Chen, C. P.;
Singh, V. K. J. Am. Chem. Soc. 1987, 109, 7925.
18. Thompson, A. S.; Humphrey, G. R.; DeMarco, A. M.;
Mathre, D. J.; Grabowski, E. J. J. J. Org. Chem. 1993, 58,
5886.
19. When describing the SAR, we focus our discussion on the
data obtained from the functional assay. The binding data
are also included for reference. In general, the potency
trends were similar between the two assays. See the
following reference for a description of the assays:
Groneberg, R. D.; Zhan, J.; Askew, B.; D’Amico, D.;
Han, N.; Fotsch, C. H.; Liu, Q.; Riahi, B.; Zhu, J.; Yang,
K.; Chen, J. J.; Nomak, R. PCT Int. Appl. WO
2004092164, Chem. Abstr. 2004, 141, 379814.
20. Compounds containing a chroman ring system were
prepared by procedures that are analogous to those found
in Scheme 1.
21. Analytical data for compound 13g: 1H NMR (CD3OD)
d = 8.15 (m, 2H), 7.96 (d, 1H, J = 7.6), 7.81 (t, 1H,
J = 9), 7.1–7.2 (m, 3H), 4.99 (t, 1H, J = 4), 4.66 (q, 1H,
J = 5), 3.80 (dd, 1H, J = 4, 14), 3.64 (s, 2H), 3.12 (dt,
1H, J = 2, 14), 2.78 (m, 2H), 2.51 (d, 2H, J = 8), 1.2–1.9
(m, 10 H), 1.19 (s, 9 H); LC–MS 100% pure at 220 and
254 nm; m/z (+ion) = 566.2 (MH+); m/z (Àion) = 610.2
(M+HCO2À).
15. Marceau, F. Trends Pharmacol. Sci. 2005, 26, 116.
16. We initially tried to replace the sulfonamide on 2 with a
carboxamide. The Ki of this compound was greater than
100 lM, so we focused only on the sulfonamides in this
report.