Identification of a nonpeptidic and conformationally restricted bradykinin B1 receptor antagonist with anti-inflammatory activity

Article abstract of DOI:10.1021/jm061224g

We report the discovery of chroman 28, a potent and selective antagonist of human, nonhuman primate, rat, and rabbit bradykinin B1 receptors (0.4-17 nM). At 90 mg/kg s.c., 28 decreased plasma extravasation in two rodent models of inflammation. A novel met

Full text of DOI:10.1021/jm061224g

J. Med. Chem. 2007, 50, 607-610
607
A major drawback for the preclinical assessment of human
B1 (hB1) antagonists is the cross-species divergence in phar-
macology and, in particular, between human and rodent.
Transgenic rats8 expressing the human receptor have been used
to address interspecies differences in B1 pharmacology. While
useful, in vivo results derived from humanized rat studies are
potentially complicated by the physiologically regulated rodent
B1 (rB1) receptor expressed in the background. A preferred
approach to drug development is to conduct target validation
studies with an agent that is active across species. Indeed, there
are reports of B1 antagonists with comparable potencies against
rB1 and hB1 receptors. For example, researchers at Sanofi
published a series of dipeptoid sulfonamides such as 19 (Figure
1, hB1 Ki ) 0.48 nM; rat ileum pA2 ) 9.4). Additionally,
scientists at Merck disclosed two series10,11 that are essentially
equipotent at both hB1 and rB1 receptors. Ultimately, the need
for potency at nonprimate receptors is required to establish
preclinical in vivo efficacy. Accordingly, our goal was to
identify novel, selective, and potent hB1 antagonists with
sufficient rB1 potency and pharmacokinetic properties to
demonstrate anti-inflammatory efficacy in a rodent model of
inflammation.
Identification of a Nonpeptidic and
Conformationally Restricted Bradykinin B1
Receptor Antagonist with Anti-Inflammatory
Activity
Derin C. D’Amico,*,† Toshi Aya,† Jason Human,†
Christopher Fotsch,† Jian Jeffrey Chen,† Kaustav Biswas,†
Bobby Riahi,† Mark H. Norman,†
Christopher A. Willoughby,†,‡ Randall Hungate,†
Paul J. Reider,† Gloria Biddlecome,§ Dianna Lester-Zeiner,§
Carlo Van Staden,§ Eileen Johnson,| Augustus Kamassah,|
Leyla Arik,| Judy Wang,| Vellarkad N. Viswanadhan,
Robert D. Groneberg,# James Zhan,# Hideo Suzuki,#
Andras Toro,# David A. Mareska,# David E. Clarke,#
Darren M. Harvey,# Laurence E. Burgess,# Ellen R. Laird,#
Benny Askew,X and Gordon NgO
Chemistry Research and DeVelopment, Neuroscience, HTS/
Molecular Pharmacology, Molecular Structure and Design, and
Inflammation, Amgen Inc., One Amgen Center DriVe, Thousand
Oaks, California 91320, and Array BioPharma, 3200 Walnut Street,
Boulder, Colorado 80301
The majority of small molecule B1 antagonists have a basic
amine and a lipophilic sulfonamide as exemplified by R1 and
R2 (1, Figure 1), an observation also noted by Marceau.12 The
requirement for a basic residue probably stems from a favorable
interaction with either of the two acidic residues within
extracellular domain 4 (EC4), namely, Glu273 or Asp291. In
addition, EC4 has been shown to be important for species-
specific pharmacology.11 Initial lead identification efforts sought
to optimize the spatial placement between a basic amine (R3)
and the 2-naphthylsulfonamide (R4) of â-phenylalanine. From
these investigations, compound 2 was identified as a moderately
potent antagonist of the hB1 receptor (IC50 ) 359 nM) as
determined by a B1-mediated calcium mobilization assay.
Furthermore, compound 2 had excellent selectivity over the B2
subtype (hB1 Ki ) 382 nM vs hB2 Ki ) 56 000 nM).
Subsequent studies showed that 4-(R) contained the activity
observed with 2 (hB1 Ki ) 132 nM (SEM ) 16 nM, n ) 4);
hB1 IC50 ) 447 nM); 3-(S) was devoid of functional potency
up to 4000 nM. A second lead, compound 6 (IC50 ) 520 nM),
was obtained by extending the distance from the sulfonamide
to the basic amine by one methylene unit. With these leads in
hand, efforts to improve potency were initiated.
ReceiVed October 19, 2006
Abstract: We report the discovery of chroman 28, a potent and
selective antagonist of human, nonhuman primate, rat, and rabbit
bradykinin B1 receptors (0.4-17 nM). At 90 mg/kg s.c., 28 decreased
plasma extravasation in two rodent models of inflammation. A novel
method to calculate entropy is introduced and ascribed ∼30% of the
gained affinity between “flexible” 4 (Ki ) 132 nM) and “rigid” 28 (Ki
) 0.77 nM) to decreased conformational entropy.
Kinins (bradykinin-related peptides) are potent vasoactive
molecules implicated in inflammation and pain that mediate their
acute and chronic actions via bradykinin receptors, which are
either constitutively expressed (B2) or inducible (B1).1,2 The
B2-antagonist icatibant (HOE-140) is being developed as an
anti-inflammatory agent for osteoarthritis.3 While rodent hy-
pertension following systemic HOE-140 administration has been
observed,4 it is unclear whether this effect would limit human
use. A phase II clinical trial looking at the effects of HOE-140
following intra-articular injection (into the knee) was initiated.
The route of administration may have been chosen, among other
factors, to mitigate potential B2-related hypertensive effects.
Reports of B1 antagonists in the clinic, however, are not
available. The preclinical role of B1 signaling in inflammation,5,6
pain,6 edema, and loss of function7 has been described. Of
relevance to our in vivo models (vide infra), B1-knockout mice
lacked the B1-agonist-induced plasma extravasation (edema)
response observed in their wild-type litter mates. Based on these
results, a B1-selective antagonist would be useful as an anti-
inflammatory agent.
One strategy that has been successfully used by medicinal
chemists to increase potency for a given receptor involves
decreasing the flexibility of a lead compound.13 In doing so,
the entropic penalty associated with adopting a preferred binding
conformation is minimized. Accordingly, the benzylic position
of 4 was “fixed” to the arene ring via a six-membered
carbocycle. The resulting tetralin 5 was ∼75-fold more potent
(hB1 IC50 ) 6 nM) than the acyclic 4. Interestingly, when a
similar cyclization was applied to phenethylamine 6 (to afford
bicycle 7), a modest increase in potency was observed (IC50:
570 nM f 380 nM). This communication reports the synthesis
and biological evaluation of a series of constrained analogs such
as tetralin 5.
* To whom correspondence should be addressed. Amgen, Inc., One
Amgen Center drive, MS-B29-4-2-D, Thousand Oaks, California 91320.
Tel.: 805-447-7354. Fax: 805-480-3016. E-mail: ddamico@amgen.com.
† Chemistry Research and Development, Amgen, Inc.
‡ Deceased (12/31/2002).
§ HTS/Molecular Pharmacology, Amgen, Inc.
| Neuroscience, Amgen, Inc.
Molecular Structure and Design, Amgen, Inc.
Compounds 2-4 were prepared as shown in Scheme 1.
Monoprotected diamine 8 was alkylated with 1,5-dibromo-
pentane to give piperidine 9. The Boc protecting group was
removed and the resulting amine, 10, was coupled with (()-11
# Array BioPharma.
X Current address: Serono Research Institute, Rockland, MA.
O Inflammation, Amgen, Inc.
10.1021/jm061224g CCC: $37.00 © 2007 American Chemical Society
Published on Web 01/23/2007

608 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4
Letters
Table 1
binding Ki, mean (SEM)
hB1 hB2
hB1 IC50, nM
mean (SEM)
cmpd
X-Y
C2′ C3
24 -CH2-
R
S
R
S
R
S
R
S
R 1.6 (0.7)a
R 241, 327
49 200 (600)b 5.6 (0.7)c
13 300 (1000)b 183 (2)a
25 -CH2-
26 -CH2CH2-
27 -CH2CH2-
28 -OCH2-
29 -OCH2-
30 -OCH2-
31 -OCH2-
32 -N(CH3)SO2-
33 -N(CH3)SO2-
34 -H, H-
R 0.24 (0.03)c 57 000 (1600)h 5.5 (0.7)g
R 31, 39
15 000 (1300)c 86 (6)a
R 0.77 (0.06)f >50 000j
3.4 (0.1)i
418 (89)e
519 (85)c
1700 (300)a
R 17 (4)b
>20 000g
ND
S 144, 143
S 1900, 2900 ND
S
R
R
R 15.80, 15.87 16 000 (1000)b 14 (0.4)a
R 1.3, 1.5
35 000 (3500)b 10.7 (0.2)a
ND
302 (45)d
Figure 1. hB1 antagonists and their associated potencies (IC50 ( SEM).
R 17 (4)a
Scheme 1a,b
a Number of measurements (NM): 4. b NM: 7. c NM: 12. d NM: 16.
e NM: 24. f NM: 31. g NM: 48. h NM: 248. i NM: 1118. j NM: 9188.
-CH2CH2-) were obtained from THF-MTBE, which afforded
high quality diffraction data. Using the method of Flack,16 it
was possible to the assign the (R)-configuration to the single
stereocenter. This conclusion verified that the (R)-CBS reduction
produced the expected (S)-tetralin alcohol and that a single
inversion was realized in the 18 f 19 sequence. Amines 20
were coupled with either acid 11 or 12 using silica-supported
carbodiimide (Si-DCC) and N-hydroxysuccinimide (NHS) to
afford amides 21. Manganese dioxide oxidation of benzylic
alcohols 21 afforded aldehydes 22 in good yield, which when
treated with piperidine/NaB(OAc)3H/HOAc, afforded amines
23. Diastereomeric mixtures of 23 were separated at this point
using preparative c-SFC. Without exception, the tested com-
pounds shown in Table 1 were obtained with >99% de, ee,
and purity.
a Reagents and conditions: (a) Br(CH2)5Br, K2CO3, CH3CN; (b) HCl,
MeOH; (c) 11, HOBt, EDC, DMF; (d) 12, HOBt, EDC, DMF; (e) Chiralcel
b
OJH, ee > 99.5%. R3 and R4 are defined in Figure 1.
Scheme 2a,b
The binding affinities for the hB1 and hB2 receptor subtypes
were next determined for the new analogs (Table 1). Tetralin
26 and chroman 28 showed the highest affinity for hB1 (Ki )
0.24 and 0.77 nM, respectively) and selectively bound this
subtype with >105 orders of magnitude over the hB2 receptor.
These cyclic analogs had affinities that were significantly (p <
0.01) lower than the acyclic 4 (Ki ) 132 nM), as determined
by Student’s t-test at the 95% confidence level. Subsequent
studies attribute this 170- to 550-fold affinity gain primarily to
enthalpic factors and secondarily to decreased conformational
entropy (vide infra). Because a high degree of selectivity over
hB2 was uniformly achieved, subsequent biological discussions
will focus on hB1 functional potency using the calcium flux
assay.
In this SAR study, the structural variables examined were
the polarity and conformational flexibility of the analogs and
the configuration present at C2′ and C3. The most potent
compounds were tetralin 26, chroman 28, indane 24, and sultams
32 and 33 (hB1 IC50 ) 3 - 14 nM). The polarity within these
five analogs varied between sultam 32 and tetralin 26 (∆cLogP
) 1.4), which when plotted against their associated potencies,
hinted at a correlation between lipophilicity and potency.
Development of this SAR was ended, because the potencies of
sultam 32 and tetralin 26 were statistically equivalent (Student’s
t-test at 95% confidence). A significant difference (p < 0.0001),
however, was observed between the acyclic (2, 4, and 34) and
cyclic compounds (24, 26, 28, 32, and 33); Chroman 28 was
125-fold more potent than the initial lead, compound 4. The
strategy to increase activity relative to compound 4, therefore,
a Reagents and conditions: (a) (R)-CBS, toluene, -10 °C; (b) NaBH4,
MeOH; (c) (S)-CBS, toluene, -10 °C; (d) DPPA, toluene, 0 °C f room
temperature; (e) LiAlH4, THF, -15 °C; (f) 11 or 12, Si-DCC, NHS, THF-
CH2Cl2; (g) MnO2, CH2Cl2; (h) piperidine, NaB(OAc)3H, HOAc, DCE; (i)
b
Chiralcel OJH, de and ee > 99.5%. R3 and R4 are defined in Figure 1.
to afford amide 2 as a racemic mixture. This mixture was
separated using chiral supercritical fluid chromatography (c-
SFC) to afford analogs 3 and 4 with >99% ee. Comparison of
the retention times (from c-SFC) of the separated enantiomers
with that of 4 (from chiral acid 12) proved that the R-form
contained the hB1 potency in racemic amide 2.
The synthesis of the constrained analogs, such as 5, required
the preparation of a series of novel bicyclic amines (Scheme
2). Known ketones 13-17 were reduced either with NaBH4 to
afford racemic alcohols 18 or with Corey-Bakshi-Shibata14
(CBS) conditions to give the corresponding chiral alcohols 18.
The (S)-alcohols were synthesized using (D)-leucine as the chiral
auxiliary in the CBS reduction ((R)-CBS), whereas the (R)-
alcohols were synthesized with (L)-leucine as the chiral control-
ler ((S)-CBS). The enantiomeric purities were 96-98%, as
determined by c-SFC analysis. Conversion of the hydroxyl
groups into azides 19 with stereochemical inversion was
accomplished using diphenylphosphoryl azide (DPPA).15 No
detectable loss of enantiomeric purity (via c-SFC) was observed
in the conversion of 18 f 19. Both the azido- and ester- groups
were reduced with LiAlH4 to afford the corresponding amino
alcohols 20. To confirm the stereochemical outcome of the
sequence, an X-ray study was conducted on a representative
example. Specifically, crystals of aminotetralinol 20 (X-Y )
was successful as five potent analogs were identified (IC50
14 nM).
<
One consequence of the conformational-restriction strategy
was the generation of an additional stereocenter at C2′ (as in
23, Scheme 2), which when coupled to the center at C3 afforded

Letters
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4 609
four possible diastereomers. The C2′-(R) and C3-(R) isomers
proved to be the optimal configuration for potency in each of
the ring systems (i.e., 24, 26, 28, and 33). However, there was
no preference for C2′-(R) in the sultam series; the C2′-(S)
diastereomer, 32, was equipotent to 33. This finding contrasts
the trend observed in the indane, tetralin, and chroman analogs;
activity decreased by 16- to 100-fold when C2′ was changed
from (R) to (S). The statistically significant rank order of potency
as a function of both C2′ and C3 was RR > SR > RS > SS (i.e.,
28 > 29 > 30 > 31). The (S)-configuration, whether at C2′ or
C3, resulted in decreased potency.
Studies in B1-transfected CHO cells showed that chroman
28 had comparable potency (IC50 ( SEM (n)) at the human
(3.4 ( 0.1 nM (1118)), rabbit (17 ( 4 nM (10)) rat- (4 ( 1 nM
(8)), and African green monkey B1 receptors (0.40 ( 0.02 nM
(17)). Additionally, the compound prevented a B1 agonist-
induced (Lys-desArg9-BK) contraction of isolated human
umbilical vein with a pA2 ) 7.4 (40 nM; Supporting Informa-
tion, section 3e). Having obtained rB1 potency, the rodent
pharmacokinetic properties of chroman 28 were measured
(Supporting Information, section 4a). Compound 28 had an
apparent clearance (Cl/F) of 5 L when dosed intravenously (i.v.)
or subcutaneously (s.c.), a plasma fraction unbound (fub) of 0.032
and a s.c. bioavailability of 100%. Taken together, plasma levels
in excess of the presumed pharmacologically relevant drug
concentration (IC50/fub ) 125 nM) could be obtained for 5-6 h
following a 90 mg/kg s.c. dose.
Figure 2. Energy profiles and Newman projection (I) about torsion
“a” for 35-39.
ligand-receptor binding event. Herein, we describe a method
that uses Shannon entropy,17 an equation that has been used to
characterize protein18 and polymer conformations.19 The affini-
ties for two acyclic analogs, 4 and 34, along with those of the
cyclic analogs 24, 26, and 28 formed the basis set for an ab
initio study. To simplify the calculations, the following structural
approximations were used: 4 ≈ N-acetylbenzylamine (35), 34
≈ N-acetyl-R-methylbenzylamine (36), 24 ≈ N-acetyl-2-amino-
indane (37), 26 ≈ N-acetyl-2-aminotetralin (38), and 28 ≈
N-acetyl-2-aminochroman (39). The calculations showed the
rotational freedom about bond “a” (Figure 2) was restricted in
the bicyclic ring systems 37-39 and, to a lesser extent, in the
methyl-substituted benzylamine 36. Specifically, cyclic analogs
37-39 had two gas-phase minima separated by 3.5 kcal mol-1
energy located at -105° (Figure 2, I) and -60°. The acyclic
36 likewise contained two minima, but with a diminished ∆E
(1.5 kcal mol-1). The two minima observed in 36-39 became
degenerate in compound 35 (minima at -150° and 30°) due to
symmetry (σ).
Chroman 28 was studied in the rat pleurisy and the reverse
passive Arthus (RPA) models of inflammation. Both models
afford the same endpoint, namely, plasma extravasation (edema)
within the pleural cavity, but differ in the way the inflammation
is initiated. Carrageenan was used in the pleurisy model and
rabbit IgG complexation to goat anti-rabbit IgG was used in
the RPA model. The latter model is thought to mimic an
immune-based inflammatory condition but without engaging the
host’s immune system. Chroman 28 (90 mg/kg, s.c.) reduced
the collected exudate by 22% relative to control (Dunnett’s
method, p ) 0.0003, Supporting Information, section 4b, vii-
viii). The unbound-drug concentration for the 90 mg/kg cohort
was 169 ( 15 nM (SEM, n ) 8, Supporting Information, section
4b, ix). As a control, the Cox-1/2 inhibitor indomethacin (Indo)
was administered at the highest dose not associated with gastric
bleeding (3 mg/kg per oral, p.o.). As observed for chroman
28, Indo reduced edema formation by 28% (p < 0.0001). In
the RPA model, reduction in exudate volume was observed at
30 (14%, p ) 0.0002), 60 (10%, p ) 0.005), and 90 mg/kg
(26%, p < 0.0001) relative to saline treated animals (Supporting
Information, section 4c, vi-vii). The responses observed at the
30 and 60 mg/kg doses were not statistically different from each
other. Indo at 3 mg/kg p.o., however, was without effect in the
RPA model. Apparently, the two models afford distinct inflam-
matory states, one that is Indo-sensitive (pleurisy) and one that
is Indo-insensitive (RPA). Unlike Indo, chroman 28 was
efficacious in both models.
The conformational entropy for 35-39 was next calculated
by converting their respective energy profiles (Figure 2) into
probability functions (eq 1). The entropy term, T∆S, was
obtained by applying the Shannon relationship (eq 2) to the
derived probability functions.
pæ ) e(-Eæ /kT)/∑i e(-Eæ /kT)
(1)
(2)
(3)
i
i
i
T∆S ) RT(-∑i pæ ln[pæ ])
i
i
BFE ) RT(Ln(hB1Ki))
where pæ is the probability that a molecule lies with a dihedral
i
angle æ, Eæ is the energy for that state, R is the universal gas
i
constant (1.987 cal mol-1 K-1), T is temperature in Kelvin, Ki
is the affinity in moles L-1, and BFE is the binding free energy.
This method calculates the total number of conformations a
molecule can exist in by integrating the probability that a
molecule will exist in a conformation with angle æ, within the
limits æ ) 0 to 360°. Using this approach on otherwise
equivalent compounds, it can be shown that a molecule with a
single, narrow energy well would have a lower calculated T∆S
than one with several accessible minima or one that has a single,
wide energy well.
To better understand the reasons for the 550-fold increase in
Ki (4 f 26), and to aid the development of a pharmacophore
model, a computational study was performed. Entropy is often
used to rationalize the increased affinity associated with the
receptor binding of more- versus less-rigid ligands; flexible
molecules can exist in more distinct conformations and,
therefore, the probability of existing in any one (i.e., the
receptor-preferred conformer) is lower than that of a rigid
molecule. Surprisingly, the literature provided no examples that
have quantitatiVely measured the entropic component in a
The calculated T∆S (kcal mol-1) for 35-39 as a function of
the “a” bond were (in decreasing order): T∆S35F(a) ) 1.82;
T∆S36F(a) ) 1.52; T∆S37F(a) ) 1.38; T∆S39F(a) ) 1.29; and
T∆S38F(a) ) 1.22. These data correlated well with the increased
affinity observed for their corresponding full structures (4 >
34 > 24 > 28 > 26, r2 ) 0.97, for plots see Supporting
Information 2c). The exceptional linearity is hard to ignore
because it is statistically significant (p ) 0.0025). However,

610 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4
Letters
(4) Carini, F.; Guelfi, M.; Lecci, A.; Tramontana, M.; Meini, S.; Giuliani,
S.; Montserrat, X.; Pascual, J.; Fabbri, G.; Ricci, R.; Quartara, L.;
Maggi, C. A. Cardiovascular effects of peptide kinin B2 receptor antag-
onists in rats. Can. J. Physiol. Pharmacol. 2002, 80 (4), 310-322.
(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. Hypoalgesia and
altered inflammatory responses in mice lacking kinin B1 receptors.
Proc. Natl. Acad. Sci. U.S.A. 2000, 97 (14), 8140-8145.
(6) Ferreira, J.; Beirith, A.; Mori, M. A. S.; Araujo, R. C.; Bader, M.;
Pesquero, J. B.; Calixto, J. B. Reduced nerve injury-induced
neuropathic pain in kinin B1 receptor knock-out mice. J. Neurosci.
2005, 25 (9), 2405-2412.
because the hB1 binding assay did not differentiate between
analogs 24, 26, and 28, the danger of overinterpretation exists.
There was a significant difference between the cyclic and the
acylic analogs (i.e., 26 and 4, vide supra), and therefore, the
change in BFE for tetralin 26 relative to 4 is significant
(∆BFE26F(4) ) BFE26 - BFE4 ) 3.7 kcal mol-1). The entropy
term of ∆BFE26F(4) was -0.6 kcal mol-1 (T∆S38F(a) - T∆S35F(a)
)
or about 20% of the 550-fold gain in affinity. While relevant
to SAR pertaining to bond “a”, the value for T∆S35F(a)
underestimates the total entropic effect because it does not
include the contributions due to bond “b” (Figure 2). Taking
(7) Sharma, J. N.; Buchanan, W. W. Pathogenic responses of bradykinin
system in chronic inflammatory rheumatoid disease. Exp. Toxicol.
Pathol. 1994, 46 (6), 421-33.
into account both bonds afforded T∆S35F(a,b) ) 2.3 kcal mol-1
,
which decreased the change in entropy for 35 f 38 to -1.1
kcal mol-1 (∆(T∆S38F(a,b)) ) T∆S38F(a) - T∆S35F(a,b)). To put
the numbers into perspective, a 1.1 kcal mol-1 decrease in
entropy corresponds to an ∼7-fold increase in affinity; it is not
the major factor behind the 550-fold gain. Enthalpy (∆∆H26F(4)),
on the other hand, accounted for 2.6 kcal mol-1 of the 3.7 kcal
(8) Hess, J. F.; Ransom, R. W.; Zeng, Z.; Chang, R. S. L.; Hey, P. J.;
Warren, L.; Harrell, C. M.; Murphy, K. L.; Chen, T.-B.; Miller, P.
J.; Lis, E.; Reiss, D.; Gibson, R. E.; Markowitz, M. K.; DiPardo, R.
M.; Su, D.-S.; Bock, M. G.; Gould, R. J.; Pettibone, D. J. Generation
and characterization of a human bradykinin receptor B1 transgenic
rat as a pharmacodynamic model J. Pharmacol. Exp. Ther. 2004,
310 (2), 488-497.
(9) 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. SSR240612 [(2R)-2-[((3R)-3-(1,3-benzodioxol-5-yl)-3-{[(6-meth-
oxy-2-naphthyl)sulfonyl]amino}propanoyl)amino]-3-(4-{[2R,6S)-2,6-
dimethylpiperidinyl]methyl}phenyl)-N-isopropyl-N-methylpropan-
amide hydrochloride], a new nonpeptide antagonist of the bradykinin
B1 receptor: Biochemical and pharmacological characterization J.
Pharmacol. Exp. Ther. 2004, 309 (2), 661-669.
(10) 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. Benzodiazepines as potent and selective bradykinin B1
antagonists. J. Med. Chem. 2003, 46 (10), 1803-1806.
(11) Su, D.-S.; Markowitz, M. 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. S.; Boyce, S.; Freidinger, R.
M.; Bock, M. G. Discovery of a potent, nonpeptide bradykinin B1
receptor antagonist. J. Am. Chem. Soc. 2003, 125 (25), 7516-7517.
(12) Marceau, F. A possible common pharmacophore in the nonpeptide
antagonists of the bradykinin B1 receptor. Trends Pharmacol. Sci.
2005, 26 (3), 116-118.
mol-1 increase in ∆BFE26F(4) (∆∆H26F(4) ) ∆BFE26F(4)
+
∆(T∆S38F(a,b))). The presence of favorable enthalpic interactions
(e.g., Van der Waals) between the atoms required to make the
cycle (X-Y, Figure 2) and the receptor are the primary factor
underlying the affinity gain between acyclic 4 (Ki ) 132 nM)
and cyclic 26 (Ki ) 0.24 nM). Decreased conformational
flexibility, while significant, contributed ∼30% of the BFE
difference between the two molecules.
In summary, chroman 28 was identified as a potent B1-
selective antagonist with in vitro activity across multiple species
and anti-inflammatory activity in vivo. A computational study
suggested that while decreased conformational entropy cor-
related with increased affinity, favorable enthalpic interactions
between the atoms required to make the molecule more rigid
and the receptor accounted for the 550-fold affinity enhancement
(4 f 26). Finally, it should be noted that potent hB1 antagonists
occupied conformation I (Figure 2, æ ) -105 ( 25°) >90%
of the time. It seems likely that this orientation lies close to the
receptor-preferred binding conformation.
(13) Cannon, J. G. Analog design. In Burger’s Medicinal Chemistry and
Drug DiscoVery, 5th ed.; Wolff, M. E., Ed.; John Wiley and Sons,
Inc.: New York, 1995; Vol. 1: Principles and Practice, pp 788-
791.
(14) Corey, E. J.; Bakshi, R. K.; Shibata, S. Highly enantioselective borane
reduction of ketones catalyzed by chiral oxazaborolidines. Mechanism
and synthetic implications. J. Am. Chem. Soc. 1987, 109 (18), 5551-
5553.
Acknowledgment. The authors express our gratitude to M.
P. Seed, S. K. McMaster, and D. A. Willoughby for conducting
the in vivo assays. Additionally, the reviewer’s comments were
helpful, thought-provoking, and very much appreciated.
Supporting Information Available: Detailed biological meth-
ods and results, synthetic experimentals, and computational data
and methodologies (61 pages). This material is available free of
charge via the Internet at http://pubs.acs.org.
(15) Lautens, M.; Rovis, T. Selective functionalization of 1,2-dihydro-
naphthalenols leads to a concise, stereoselective synthesis of sertra-
line. Tetrahedron 1999, 55 (29), 8967-8976.
(16) Flack, H. D. On enantiomorph-polarity estimation. Acta Crystallogr.,
Sect. A 1983, A39 (6), 876-81.
(17) Shannon, C. E. Prediction and entropy. Bell Syst. Tech. J. 1951, 50-
64.
References
(1) Dray, A.; Perkins, M. Bradykinin and inflammatory pain. Trends
Neurosci. 1993, 16 (3), 99-104.
(18) Sundaram, K.; Viswanadhan, V. N.; MacElroy, R. D. Information
contained in protein shapes. Int. J. Pept. Protein Res. 1983, 21, 107-
117.
(19) Viswanadhan, V. N.; Mattice, W. L. Assessment of bond rotation
interdependence in polymer chains: An information theory approach.
Macromolecules 1987, 20, 685-688.
(2) Proud, D.; Kaplan, A. P. Kinin formation: Mechanisms and role in
inflammatory disorders. Annu. ReV. Immunol. 1988, 6 (1), 49-83.
(3) Carini, F.; Guelfi, M.; Lecci, A.; Tramontana, M.; Meini, S.; Giuliani,
S.; Montserrat, X.; Pascual, J.; Fabbri, G.; Ricci, R.; Quartara, L.;
Maggi, C. A. Cardiovascular effects of peptide kinin B2 receptor antag-
onists in rats. Can. J. Physiol. Pharmacol. 2002, 80 (4), 310-322.
JM061224G