A new class of bradykinin B1 receptor antagonists with high oral bioavailability and minimal PXR activity

Article abstract of DOI:10.1016/j.bmcl.2007.11.057

The design and synthesis of a novel class of human bradykinin B₁ antagonists featuring difluoroethyl ether and isoxazole carboxamide moieties are disclosed. Compound 7g displayed excellent pharmacokinetic properties, efficient ex vivo receptor occupancy, and low potential for P450 induction via PXR activation.

Full text of DOI:10.1016/j.bmcl.2007.11.057

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 682–687
A new class of bradykinin B1 receptor antagonists with high
oral bioavailability and minimal PXR activity
Dong-Mei Feng,a Robert M. DiPardo,a Jenny M. Wai,a Ronald K. Chang,a
Christina N. Di Marco,a Kathy L. Murphy,b Richard W. Ransom,b Duane R. Reiss,b
Cuyue Tang,c Thomayant Prueksaritanont,c Douglas J. Pettibone,b
Mark G. Bocka and Scott D. Kuduka,*
aDepartment of Medicinal Chemistry, Merck Research Laboratories, West Point, PA 19486, USA
bDepartment of Neuroscience Drug Discovery, Merck Research Laboratories, West Point, PA 19486, USA
cDepartment of Drug Metabolism, Merck Research Laboratories, West Point, PA 19486, USA
Received 25 October 2007; revised 12 November 2007; accepted 15 November 2007
Available online 21 November 2007
Abstract—The design and synthesis of a novel class of human bradykinin B1 antagonists featuring difluoroethyl ether and isoxazole
carboxamide moieties are disclosed. Compound 7g displayed excellent pharmacokinetic properties, efficient ex vivo receptor occu-
pancy, and low potential for P450 induction via PXR activation.
Ó 2008 Published by Elsevier Ltd.
The kinin-related peptides are potent endogenous alge-
sic and pro-inflammatory substances. The kinins medi-
ate a number of acute and chronic inflammatory
responses including vasodilation, edema, cellular infil-
tration, and pain.1–3 These effects are mediated by two
G-protein coupled receptors, B1 and B2.4–8 B2 receptors
are normally present in the periphery in many cell types
involved in pain responses. Bradykinin B1 receptors are
not normally highly expressed, but instead are rapidly
induced in response to inflammatory and painful stim-
uli. The inducible nature of the B1 receptor has led to
the belief that B1 receptors are most relevant during
longer term inflammatory pain conditions, while B2
receptors are important mediators of acute pain and
the early inflammatory response. This hypothesis was
initially supported by studies with peripheral adminis-
tration of peptidyl kinin agonists and antagonists in
non-inflamed, inflamed, and neuropathic pain mod-
els.9–12 Recently, however, studies with B1 knockout
mice have generated a broader view of the potential role
of B1 mechanisms in both non-inflammatory and inflam-
matory pain, acutely and chronically.13,14
We previously reported non-peptide human bradykinin
B1 antagonists15 based on the benzodiazepine,16 dihy-
droquinoxaline,17 diaminopyridine,18 and cyclopropane-
carboxamide scaffolds.19 SAR studies in the latter series
indicated that incorporation of a trifluoroacetamide
group led to analogs that were not substrates for the
xenobiotic efflux pump P-glycoprotein (P-gp), and had
acceptable CNS penetration.20 As a follow up to this
work, we investigated a related series that had excellent
binding affinity, good brain penetration, and reasonable
pharmacokinetic properties as exemplified by com-
pound 1 (Fig. 1).21 However, compound 1 still possessed
certain metabolic liabilities related to the methyl ester
O
CF3
NH
O
NH
F
CO2Me
B
Cl
C
Keywords: Bradykinin; GPCR; Pain; PXR; P-glycoprotein.
*
Cl
Corresponding author. Tel.: +1 215 652 5147; e-mail: scott_d_kuduk
@merck.com
Figure 1. Lead compound 1.
0960-894X/$ - see front matter Ó 2008 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2007.11.057

D.-M. Feng et al. / Bioorg. Med. Chem. Lett. 18 (2008) 682–687
683
moiety, and it was shown to activate the pregnane X
O
S
receptor (PXR).22 Activation of PXR leads to induction
of CYP3A4, a cytochrome P450 enzyme that is respon-
sible for the metabolism of many drugs.23 It is possible,
therefore, that PXR activation could cause potential
drug–drug interactions. In this paper, we describe our
synthetic efforts to modify compound 1 in order to im-
prove its pharmacokinetic profile and eliminate PXR
activity.
O
S
t
-Bu
NH
F
t-Bu
NH
F
X
R1
a, b
X
Cl
R1
R2
Cl
Br
Br
4a-h
3a, b
R2
2a-e
O
R3
c
The compounds described in this study are listed in Ta-
bles 1–4 and the general preparative method used to ac-
cess the compounds is outlined in Schemes 1 and 2.
Commercially available phenols 1a–b were brominated
with NBS followed by alkylation with the appropriate
alkyl triflate or halide to provide 2a–e.
NH
NH
NH2
O
d-f
X
X
R1
R1
Cl
Cl
F
F
The synthesis of compounds 3a and 3b has been previ-
ously described,21 and these were readily converted to
the pinacol boronate derivatives which underwent
smooth Suzuki coupling with 2a–e to produce the pro-
tected biphenyl amines 4a–h (Scheme 2). Deprotection
of the sulfinamide in the standard fashion afforded the
biphenylamines 5a–h. Subsequent EDCI mediated acyl-
ation with 1-(tert-butoxycarbonyl)aminocyclopropane-
carboxylic acid followed by Boc removal and acylation
with the appropriate carboxylic acid afforded target
compounds 6a–f for evaluation.
6a-f; X = CH
7a-i; X = N
8a-b; X = N
R2
R2
5a-h
Scheme 2. Reagents and conditions: (a) pinacoldiboron ester, KOAc,
DMSO, Pd(dppf)C12, 80 °C; (b) Pd(dppf)C12, 80 °C; K2CO3, DMSO;
(c) 4 N HCl dioxane, 0 °C; (d) tert-butoxycarbonyl-aminocyclopro-
pane carboxylic acid, EDCI, HOAt, DMF, TEA; (e) HCl, ethyl
acetate, 0 °C; (f) R3CO2H, EDCI, HOAt, DMF, TEA.
pound 6c as a lead structure for further efforts to modify
the PXR activation potential by replacement of the tri-
fluoroacetamide moiety.
A number of ether replacements were examined for the
metabolically labile C-ring methyl ester as shown in Ta-
ble 1. All compounds displayed nanomolar affinity for
the hBK B1 receptor and had good pharmacokinetics
in the rat. For example, ethyl ethers 6a–b both showed
good affinity, but the addition of the chlorine at the 50
position (R2 = Cl) of the phenyl C-ring had notable
improvement in terms of half-life in rat. The difluoroeth-
yl ether (6c) maintained similar rat pharmacokinetics to
6b, but markedly improved binding potency
(hKi = 0.2 nM) whilst the more hindered isopropyl ether
6d lost about 10-fold in terms of receptor binding. Add-
ing a fluorine to each methyl group (6e) returned signif-
icant binding affinity, but proved to also confer
unacceptable P-gp susceptibility (BA/AB ratio = 5.2).
Interestingly, the trifluoroethyl ether 6f lacked sufficient
permeability (<15) for further evaluation.
After surveying a variety of heterocyclic replacements
(data not shown) for the trifluoroacetamide, a series of
promising isoxazoles was settled on for further examina-
tion. The compounds in Table 2 summarize the results
of this endeavor. First, it should be noted that incorpo-
ration of the B-ring pyridine into compound 6c to pro-
vide 7a was reasonably well tolerated in terms of hBK
B1 binding characteristics, maintained good P-gp and
dog pharmacokinetic properties, but did not reduce
the PXR activation liability for the compound. Replace-
ment of the trifluoroacetamide with a 3-isoxazole car-
boxamide (7b) maintained excellent receptor affinity
and dog pharmacokinetics relative to compound 1, but
more importantly had the desired effect of significantly
reducing the PXR activation to 23% relative to the effect
of rifampicin at 10 lM. Alkyl group incorporation and
modification to the 5-position of the isoxazole (com-
pound 7c–e) eroded receptor binding to a modest extent
and began to confer undesirable susceptibility to P-gp
mediated efflux with larger alkyls (7e).
It has been previously observed that compounds con-
taining a pyridine B-ring had markedly improved phys-
ical properties and enhanced CNS penetration than
compounds containing a phenyl B-ring.21 On this basis
we chose to incorporate the pyridine B-ring into com-
The corresponding 5-isoxazole carboxamide (7f) was
also examined and found to have enhanced binding
affinity (hKi = 0.4 nM) and maintained low PXR activa-
tion potential relative to 1, but was a modest P-gp sub-
strate (BA/AB = 3.9). Incorporation of an electron
donating methoxy group (7g) afforded a desirable P-gp
profile, improved the dog half-life, and exhibited the
lowest PXR activation potential amongst the series
(8% at 10 lM relative to rifampicin). The related ethoxy
isoxazole 7h maintained a similar profile to 7g, while the
3-chloro isoxazole 7i exhibited very good receptor affin-
OR
OH
Cl
Br
Cl
a, b
R2
R2
1a-b
2a-e
Scheme 1. Reagents and condition: (a) NBS, CH3CN, 30 min; (b) i—
NaH, HMPA; ii—ROTf or RI or RBr.

684
D.-M. Feng et al. / Bioorg. Med. Chem. Lett. 18 (2008) 682–687
Table 1. Effects of alkyl ether modifications on BK B1 receptor binding affinities and PK properties
O
CF3
NH
O
NH
F
R1
Cl
R2
R1
R2
Cl
hBK1
0.4
P-gpb
1.8
Papp
18
Rat F%d
38
Rat t1/2
Rat Cl
19
PXR (% act at 10 lM)e
a
c
Compound
1
CO2Me
4.6
77
6a
6b
6c
6d
H
0.9
1.3
0.2
2.3
1.9
1.9
2.1
nd
29
18
26
nd
48
38
nd
nd
1.9
7.5
7.6
nd
16
nd
62
95
nd
O
O
Cl
Cl
Cl
11.5
5.8
nd
F
F
O
F
O
F
6e
6f
Cl
Cl
0.9
1.8
5.2
2.0
28
56
35
5.0
27
1.6
1.1
nd
45
O
F
F
O
6.3
F
a Values represent the numerical average of at least two experiments. Interassay variability was 25% (Ki, nM).
b MDR1 Directional transport ratio (B to A)/(A to B). Values represent the average of three experiments and interassay variability was 20%.
c Passive permeability (10ꢀ6 cm/s).
d F% oral bioavailability, half-life is represented in hours, Cl in mL/min/kg. Sprague–Dawley rats (n = 3). Oral dose = 10 mg/kg, IV dose = 2 mg/kg.
Interanimal variability was less than 20%.
e Relative to rifampicin.
ity with a long dog half-life, but the PXR activation in-
creased considerably.
Moreover, the receptor occupancy of 8b was modestly
improved (Occ90 = 380 nM) over 1 (Occ90 = 450 nM)
and has similar brain to plasma ratio in the monkey.
Derivatives 7c and 7g both had Occ90s commensurate
with the level of binding affinity, exhibit CNS penetra-
tion in monkey, and have low PXR activation. In or-
der to make a distinction among compounds 7c, 7g,
and 8b, the pharmacokinetics for these compounds of
interest were compared across species (Dog and Rhesus
Monkey) and results of this comparison are presented
in Table 4.
To further characterize compounds in Table 2 and re-
lated analogs, ex vivo receptor binding experiments were
carried out. The transgenic rat construct employed in
these studies expresses the hBK B1 receptor broadly
throughout the CNS, including the brain and spinal
cord, under the mouse neuronal specific enolase pro-
moter.24 This rat model has been successfully used to
measure the efficiency with which a compound binds
to the hBK B1 receptor expressed in the CNS. In addi-
tion, studies of pharmacokinetics in rats and brain pen-
etration in African Green Monkey (AGM) were carried
out for selected compounds.
It is clear from the dog and rhesus data shown in Table 4
(as well as the rat data shown in Table 3) that the ether
replacements for the ester led to a dramatic improve-
ment in pharmacokinetic properties across species. The
properties of 5-methyl isoxazole 7c were significantly
better in the dog, but did not show any improvement
over 1 in rhesus. The related 3-methoxy isoxazole 7g
showed superior bioavailability (>90%) and the longest
half-lives in dog (21 h) and rhesus (6.2 h) for any ana-
logs examined in this study. Ethyl ether variant 8b
exhibited low bioavailability in rat (7%) and rhesus
As shown in Table 3, the ethyl ether variant (8a) of
compound 1 showed a modest decrease in B1 receptor
affinity that was reflected in the Occ90 of this analog.
Additionally, 8a has significant PXR activation, and
replacement with the 5-isoxazole carboxamide (8b) dra-
matically reduced this activity showing the trifluoroac-
etamide is a major contributor to the PXR problem.

D.-M. Feng et al. / Bioorg. Med. Chem. Lett. 18 (2008) 682–687
685
Table 2. Effects of isoxazole modifications on BK B1 receptor binding affinity, P-gp transport, Dog pharmacokinetic properties, and PXR activity
O
R3
NH
F
F
O
NH
F
N
O
Cl
Cl
R3
hBK1
1.5
P-gpb
2.2
Papp
Dog t1/2
Dog Cl
2.3
PXR (% act at 10 lM)e
a
c
d
Compound
7a
CF3
23
21
99
O
O
O
O
N
N
N
N
N
7b
7c
7d
7e
7f
1.2
1.4
1.9
2.6
0.4
0.5
1.1
0.2
1.8
1.5
2.0
3.8
3.9
2.0
2.1
1.6
35
18
2.0
4.5
14
23
28
37
58
21
8
N
N
N
O
O
O
O
23
25
18
29
28
25
26
9.4
4.5
6.9
4.0
21
13
5.9
1.2
3.0
2.6
O
O
Cl
7g
7h
7i
13.5
34
37
68
a Values represent the numerical average of at least two experiments. Interassay variability was 25% for the binding assays (Ki, nM).
b MDR1 Directional transport ratio (B to A)/(A to B). Values represent the average of three experiments and interassay variability was 20%.
c Passive permeability (10ꢀ6 cm/s).
d Mongrel dog cassette (n = 2). IV dose = 1 mg/kg. Interanimal variability was less than 20% for all values. Half-life is represented in hours, and Cl in
mL/min/kg.
e Relative to rifampicin.
(3%) with shorter half-lives across species relative to 7g
exemplifying the clear advantage of the difluoroethyl
ether moiety in terms of pharmacokinetic properties.
P-gp profile was not satisfactory with a directional trans-
port ratio of 5.1 due to the presence of the isoxazole in
combination with the methyl ester. It is likely that the ex-
tra heteroatoms present due to the ester of 9 confer this P-
gp susceptibility, while the difluoroethyl ether in 7g bal-
ances out this risk and is devoid of the P-gp liability. Thus,
while the replacement of the trifluoroacetamide with the
isoxazole carboxamide was essential to address the PXR
issue, the ether replacement for the ester was required in
order to make up for the commensurate rise in P-gp
liability
To further establish the reduction in PXR activity was
driven largely by the incorporation of the isoxazole het-
erocycle in lieu of the trifluoroacetamide, the 3-meth-
oxyisoxazole version of compound 1 was prepared in
the form of compound 9. This analog exhibited similar
receptor binding and gratifyingly also showed a marked
decrease in PXR activation relative to 1. However, the

686
D.-M. Feng et al. / Bioorg. Med. Chem. Lett. 18 (2008) 682–687
Table 3. Ex vivo occupancy and CNS penetration in AGM for selected compounds
R3
O
NH
O
NH
F
X
R1
Cl
Cl
Compound R1
R3
X
hBK1
Rat F%b Rat t1/2 Rat Cl Occ90
AGM
PXR
a
(nM)a CNS/Plasmac (% act at 10 lM)d
1
CO2Me
CF3
CF3
CH 0.4
38
8
4.6
5
19
450
1.1
2.2
77
89
8a
N
N
N
N
1.5
0.6
1.4
0.5
11.3
1450
O
N
O
N
O
8b
7c
7g
7
30
44
1.9
3
10
380
940
540
1.4
0.9
0.7
15
28
8
O
O
F
F
N
O
5.1
2.4
F
F
O
O
4.5
a Values represent the numerical average of at least two experiments. Interassay variability was 25% (Ki, nM).
b F% oral bioavailability, half-life is represented in hours, Cl in mL/min/kg. Sprague–Dawley rats (n = 3). Oral dose = 10 mg/kg, IV dose = 2 mg/kg.
Interanimal variability was less than 20%.
c Values are means of two experiments.
d Relative to rifampicin.
Table 4. Dog and Rhesus pharmacokinetics for select compounds
In conclusion, systematic structural modifications to
Compound
Dog PKa
Rhesus PKb
compound 1 led to analogs displaying reduced poten-
tial for P450 induction via activation of the PXR path-
way. Exchange of the methyl ester in compound 1 with
a difluoroethyl ether moiety yielded compounds with
good hBK B1 receptor affinity and marked improve-
ments in pharmacokinetic profiles. Replacement of
the trifluoromethyl functional group in 1 with 3-meth-
oxyisoxazole gave compounds with reduced susceptibil-
ity to P-gp mediated efflux, attenuated PXR activation,
and excellent pharmacokinetic properties. The combi-
nation of all of the above modifications in one hybrid
structure resulted in the discovery of compound 7g
which displayed high hBK B1 receptor binding affinity
and efficiently occupied the hBK B1 receptor in the
CNS of the transgenic mouse. Accordingly, 7g was se-
lected for further evaluation as a potential clinical
candidate.
F (%) t1/2 (h) Cl
F (%) t1/2 (h) Cl
1
19
21
94
55
1.6
9.4
20
20
3.9
1.6
6.2
1.2
11.2
9.1
1.6
15
7c
7g
8b
4.5 25
1.2 93
21
4.5
7.0
3
a Mongrel dogs (n = 2). Oral dose 3 mg/kg, IV dose = 1 mg/kg. Inter-
animal variability was less than 20% for all values and Cl is in mL/min/
kg.
b Rhesus Monkeys (n = 2). Oral dose 3 mg/kg, IV dose = 1 mg/kg.
Interanimal variability was less than 20% for all values and Cl in mL/
min/kg.
N
O
O
O
NH
O
NH
F
CO2Me
Cl
References and notes
1. Campbell, D. J. Clin. Pharmacol. Physiol. 2001, 28, 1060.
2. Erdos, E. G. Cardiovasc. Res. 2002, 54, 485.
3. Marceau, F.; Hess, J. F.; Bachvarov, D. R. Pharmacol.
Rev. 1998, 50, 357.
4. Walker, K.; Perkins, M.; Dray, A. Neurochem. Int. 1995,
26, 1.
5. Ahluwalia, A.; Perretti, M. TIPS 1999, 20, 100.
Compound 9
Cl
hBK1 = 0.4 nM
PXR = 40% @ 10 μM
P-gp = 5.1

D.-M. Feng et al. / Bioorg. Med. Chem. Lett. 18 (2008) 682–687
687
6. Couture, R.; Harrisson, M.; Vianna, R. M.; Cloutier, F.
Eur. J. Pharmacol. 2001, 429, 161.
7. Bock, M. G.; Longmore, J. Curr. Opin. Chem. Biol. 2000,
4, 401.
18. Kuduk, S. D.; Ng, C.; Feng, D. M.; Wai, J.; Chang, R. S.
L.; Harrell, C. M.; Murphy, K. L.; Ransom, R. W.; Reiss,
D.; Prueksaritanont, T.; Tang, C.; Mason, G.; Boyce, S.;
Freidinger, R. M.; Pettibone, D. J.; Bock, M. G. J. Med.
Chem. 2004, 47, 6439.
19. Wood, M. R.; Books, K. M.; Kim, J. J.; Wan, B.-L.;
Murphy, K. L.; Ransom, R. W.; Chang, R. S. L.; Tang,
C.; Prueksaritanont, T.; Detwiler, T. J.; Hettrick, L. A.;
Landis, E. R.; Leonard, Y. M.; Krueger, J. A.; Lewis, S.
D.; Pettibone, D. J.; Freidinger, R. M.; Bock, M. G. J.
Med. Chem. 2006, 48, 1231.
20. Kuduk, S. D.; Di Marco, C. N.; Chang, R. K.; Wood, M.
R.; Schirripa, K. M.; Kim, J. J.; Wai, J. C.; DiPardo, R.
M.; Murphy, K. L.; Ransom, R. W.; Harrell, C. M.; Reiss,
D. R.; Holahan, M. A.; Cook, J.; Hess, J. F.; Sain, N.;
Urban, M. O.; Tang, C.; Prueksaritanont, T.; Pettibone,
D. J.; Bock, M. G. J. Med. Chem. 2007, 50, 272.
21. Kuduk, S. D.; DiPardo, R. M.; Chang, R. K.; Di Marco,
C. N.; Murphy, K. L.; Ransom, R.; Tang, C.; Prueksa-
ritanont, T.; Pettibone, D. J.; Bock, M. G. Bioorg. Med.
Chem. Lett. 2007, 17, 3608.
22. Bertilsson, G.; Heidrich, J.; Svensson, K.; Asman, M.;
Jenderberg, L.; Sydow-Backman, M.; Ohlsson, R.; Post-
lind, H.; Blomquist, P.; Berkenstam, A. Proc. Natl. Acad.
Sci. U.S.A. 1998, 95, 12208.
23. Pichard, L.; Fabre, I.; Fabre, G.; Domergue, J.; Aubert, B.
S.; Mourad, G.; Maruel, P. Drug Metab. Dispos. 1990, 18,
595.
24. Hess, J. F.; Ransom, R. W.; Zeng, Z.; Chang, R. S.;
Hey, P. J.; Warren, L.; Harrell, C. M.; Murphy, K. L.;
Chen, T. B.; Miller, P. J. J. Pharmacol. Exp. Ther.
2004, 310, 488.
8. Perkins, M. N.; Campbell, E.; Dray, A. Pain 1993, 53, 191.
9. Perkins, M. N.; Kelly, D. Neuropharmacol. 1994, 33, 657.
10. Davis, A. J.; Perkins, M. N. Br. J. Pharmacol. 1994, 113, 63.
11. Khasar, S. G.; Miao, F. J.-P.; Levine, J. D. Neurosci. 1995,
3, 685.
12. Mason, G. S.; Cumberbatch, M. J.; Hill, R. G.; Rupniak,
N. M. J. Can. J. Physiol. Pharmacol. 2002, 80, 264.
13. 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.; Calixto, J. B.; Lewin, G. R.;
Bader, M. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 8140.
14. Ferreira, J.; Campos, M. M.; Pesquero, J. B.; Araujo, R.
C.; Bader, M.; Calixto, J. B. Neuropharm. 2001, 41, 1006.
15. For an general review of non-peptide ligands for brady-
kinin receptors, see: Dziadulewicz, E. K. Exp. Opin. Ther.
Patents 2005, 15, 829.
16. Wood, M. R.; Kim, J. J.; Han, W.; Dorsey, B. D.;
Homnick, C. F.; DiPardo, R. M.; Kuduk, S. D.; MacNeil,
T.; Murphy, K. L.; Lis, E. V.; Ransom, R. W.; Stump, G.
L.; Lynch, J. J.; O’Malley, S. S.; Miller, P. J.; Chen, T.-B.;
Harrell, C. M.; Chang, R. S. L.; Sandhu, P.; Ellis, J. D.;
Bondiskey, P. J.; Pettibone, D. J.; Freidinger, R. M.;
Bock, M. G. J. Med. Chem. 2003, 46, 1803.
17. Su, D. S.; Markowitz, K.; DiPardo, R. M.; Murphy, K.
L.; Harrell, C. M.; O’Malley, S. S.; Ransom, R. W.;
Chang, R. S. L.; Ha, S.; Hess, F. J.; Pettibone, D. J.;
Mason, G.; Boyce, S.; Freidinger, R. M.; Bock, M. G. J.
Am. Chem. Soc. 2003, 125, 7516.