Practical synthesis of a potent bradykinin B1 antagonist via enantioselective hydrogenation of a pyridyl N-acyl enamide

Article abstract of DOI:10.1021/jo802772d

(Chemical Equation Presented) A practical and efficient synthesis of bradykinin B₁ antagonist 1 is described. A convergent strategy was utilized which involved synthesis of three fragments: 3, 6, and 7. Cross coupling of fragments 6 and 7 followed by amidation with 3 enabled efficient synthesis of 1 in 19 steps total, a 35% overall yield from commercially available pyridine 10. The key to the success of the synthesis was the development of a fluorodenitration step to install the fluorine in pyridine 7 and a catalytic enantioselective hydrogenation of N-acyl enamide 9 to set the stereochemistry.

Full text of DOI:10.1021/jo802772d

Practical Synthesis of a Potent Bradykinin B1 Antagonist via
Enantioselective Hydrogenation of a Pyridyl N-Acyl Enamide
Paul D. O’Shea,*,† Danny Gauvreau,† Francis Gosselin,† Greg Hughes,† Christian Nadeau,†
Ame´lie Roy,† and C. Scott Shultz‡
Department of Process Research, Merck Frosst Centre for Therapeutic Research, P.O. Box 1005,
Pointe-Claire-DorVal, Que´bec H9R 4P8, Canada, and Merck Research Laboratories, P.O. Box 2000,
Rahway, New Jersey 07065
paul_oshea@merck.com
ReceiVed December 19, 2008
A practical and efficient synthesis of bradykinin B1 antagonist 1 is described. A convergent strategy was
utilized which involved synthesis of three fragments: 3, 6, and 7. Cross coupling of fragments 6 and 7
followed by amidation with 3 enabled efficient synthesis of 1 in 19 steps total, a 35% overall yield from
commercially available pyridine 10. The key to the success of the synthesis was the development of a
fluorodenitration step to install the fluorine in pyridine 7 and a catalytic enantioselective hydrogenation
of N-acyl enamide 9 to set the stereochemistry.
Introduction
its pharmacological properties. Herein, we report our efforts to
develop a practical, chromatography-free, enantioselective
synthesis suitable for the preparation of 1 on multikilogram
scale.
Bradykinin is a kinin that mediates the physiological processes
accompanying acute and chronic pain and inflammation.1 Two
classes of bradykinin receptors, B1 and B2, are known to be
activated upon kinin release.2 The B2 receptor is present in many
cell types and tissues under normal physiological conditions and
is believed to be responsible for the immediate acute pain
response following tissue injury.3 In contrast, the B1 receptor
is not normally expressed in most tissues but is induced in
response to inflammation, tissue damage, or bacterial infection.4
This makes the B1 receptor an attractive drug target that could
be potentially useful in the management of pain and inflam-
mation. Compound 1 has been identified as a potent and
selective antagonist of bradykinin B1.5 We were interested in
preparing kilogram quantities of 1 in an effort to further explore
Compound 1 contains a diverse range of functionality, and
we were particularly interested in devising a convergent
synthesis based on assembling fragments of similar complexity.
Our retrosynthetic analysis is shown in Figure 1. We envisioned
that 1 could be assembled in a convergent manner via cross
coupling of aryl boronic acid 6 and bromopyridine 7 followed
by acylation with isoxazole acid 3. The key chiral center could
be installed via asymmetric hydrogenation of pyridine enamide
9, which in turn could be prepared from commercially available
trisubstituted pyridine 10.
Results and Discussion
† Merck Frosst Centre for Therapeutic Research.
‡ Merck Research Laboratories.
Synthesis of Aryl Boronic Acid 6. Our synthesis began with
bromination of commercially available 2,4-dichlorophenol. A
clean reaction was obtained using NBS in various solvents;
(1) Bock, M. G.; Longmore, J. Curr. Opin. Chem. Biol. 2000, 4, 401.
(2) Regoli, D.; Barabe´, J. Pharmacol. ReV. 1980, 32, 1.
(3) Bock, M. G.; Hess, J. F.; Pettibone, D. J. In Annual Reports in Medicinal
Chemistry; Doherty, A. M., Ed.; Elsevier: New York, 2003; Vol. 38, p 111.
(4) (a) Calixto, J. B.; Cabrini, D. A.; Ferreira, J.; Campos, M. M. Pain
2000, 87, 1. (b) Calixto, J. B.; Cabrini, D. A.; Ferreira, J.; Campos, M. M.
Curr. Opin. Anaesthesiol. 2001, 14, 519. (c) Marceau, F. Immunopharma-
cology 1995, 30, 1.
(5) Feng, D.-M.; DiPardo, R. M.; Wai, J. M.; Chang, R. K.; Di Marco, C. N.;
Murphy, K. L.; Ransom, R. W.; Reiss, D. R.; Tang, C.; Prueksaritanont, T.;
Pettibone, D. J.; Bock, M. G.; Kuduk, S. D. Bioorg. Med. Chem. Lett. 2008, 18,
682.
10.1021/jo802772d CCC: $40.75
Published on Web 05/20/2009
2009 American Chemical Society
J. Org. Chem. 2009, 74, 4547–4553 4547

O’Shea et al.
SCHEME 1. Synthesis of Aryl Boronic Acid 6
A number of 2-substituted pyridines were prepared from 10
to evaluate the reactivity of various groups toward cyanation.
The corresponding 2-chloride and 2-triflate were prepared but
were found to be unreactive in the subsequent cyanation step
under a variety of reaction conditions. Initial experiments with
the 2-bromo pyridine 14 indicated that selective cyanation was
possible at the 2-position (over the 5-position), and efforts were
focused on the development of an efficient procedure for its
preparation. Treatment of 10 with NBS/Ph3P8 or P2O5/Bu4NBr9
following reported procedures did not lead to useful yields of
14 (Table 1, entries 1, 2). 2-Bromo pyridines have also been
prepared by treatment of pyridones with POBr3 in DMF.10
However, treatment of 10 using these conditions led to low
yields of the desired bromide. Switching to toluene as the solvent
at 110 °C led to 14 in 80% yield. Interestingly, the addition of
10 mol % of DMF to toluene increased the yield of the desired
product to 92% and enabled a complete reaction at 90 °C within
16 h. It is important to add a solution of POBr3 in toluene slowly
to the pyridine at 90 °C to avoid a significant exotherm that
was observed at 80 °C when all the reagents were mixed at
room temperature and then heated to the reaction temperature.
FIGURE 1. Retrosynthetic analysis.
however, separation of the product from succinimide was
problematic, and 11 could not be isolated pure. To solve the
isolation problems, the bromination was investigated using Br2
in presence of a base. Thus, addition of t-BuNH26 to a degassed
solution of bromine and phenol 8 in toluene at -50 °C in the
absence of light gave a clean reaction to 11. The reaction was
quenched with NaHSO3, and following aqueous workup,
bromide 11 was obtained in 97% yield (Scheme 1).
TABLE 1. Bromination of Pyridine 10
An investigation of possible electrophiles for the alkylation
of phenol 11 indicated that mesylate 12 gave excellent results.
The mesylate was prepared in a straightforward manner in 95%
yield from commercially available 2,2-difluoroethanol (Scheme
1). Phenol 11 was then alkylated in DMF using K2CO3 at 90-95
°C for 16 h to yield difluoroethylether 13 in 93% yield.
Treatment of bromide 13 with i-PrMgCl7 at 0 °C cleanly
generated the corresponding aryl Grignard reagent. Addition of
triisopropyl borate followed by acidic workup gave the desired
boronic acid 6 in an excellent 90% isolated yield (Scheme 1).
Synthesis of Chiral Pyridine 7. The preparation of trisub-
stituted pyridine 7 poses an interesting synthetic challenge, and
a number of potential routes were investigated. The route chosen
is outlined in Scheme 2, starting from commercially available
5-bromo-3-nitropyridin-2-ol 10.
entry
reagent
solvent
T
time yield
1
2
3
4
NBS/Ph3P
dioxane
100 °C 15 h 15%
P2O5/Bu4NBr toluene
100 °C
110 °C
3 h 68%
4 h 80%
POBr3
POBr3
toluene
toluene/DMF (10 mol %)
90 °C 16 h 92%
Transition-metal mediated displacement of aryl halides by a
cyanide ion is a convenient method for the preparation of aryl
nitriles.11 We investigated a number of reaction conditions and
(8) (a) Sugimoto, O.; Mori, M.; Tanji, K.-I. Tetrahedron Lett. 1999, 40, 7477.
(b) Sugimoto, O.; Mori, M.; Tanji, K.-I. HelV. Chim. Acta 2001, 84, 1112.
(9) Kato, Y.; Okada, S.; Tomimoto, K.; Mase, T. Tetrahedron Lett. 2001,
42, 4849.
(10) Quallich, G. J.; Fox, D. E.; Friedmann, R. C.; Murtiashaw, C. W. J.
Org. Chem. 1992, 57, 761.
(11) Ellis, G. T.; Romney-Alexander, T. M. Chem. ReV. 1987, 87, 779.
(6) Pearson, D. E.; Wysong, R. D.; Breder, C. V. J. Org. Chem. 1967, 32,
2358.
(7) Knochel, P.; Dohle, W.; Kneisel, F. F.; Kopp, K.; Korn, T.; Sapountzis,
I.; Vu, V. A. Angew. Chem., Int. Ed. 2003, 42, 4302.
4548 J. Org. Chem. Vol. 74, No. 12, 2009

Synthesis of a Potent Bradykinin B1 Antagonist
SCHEME 2. Synthesis of Pyridine 7
found that treatment of dibromopyridine 14 with CuCN in
propionitrile at 90 °C gave excellent results. Thus, a solution
of 14 in propionitrile was treated with CuCN and heated to 90
°C. The reaction was complete after aging for 15 h at 90 °C,
and following aqueous workup, pyridine nitrile 15 was obtained
in 94% yield. It was deemed essential to remove residual copper
to avoid detrimental effects on the downstream chemistry, and
this was easily achieved by washing the toluene solution of 15
with saturated aqueous NaCl.
A key step in our strategy involved installation of the fluorine
at the 3-position of the pyridine ring. There are a number of
available methodologies such as diazotization of amines in the
presence of HF12 or thermal decomposition of diazonium salts13
to prepare fluoropyridines. However, these methodologies
require specialized equipment and pose significant safety
concerns, making them unattractive for large scale experiments.
the water content reduced the level of the hydroxy pyridine 17
(Table 2, entry 4) leading to a higher yield. It was also found
that the addition of H2SO4 (2 mol %) significantly reduced the
low-level impurities and improved the product color. Thus, the
optimal conditions were found to be the addition of a DMF
solution of 15 to a predried (molecular sieves) solution of TBAF
in DMF containing H2SO4 (2 mol %) at -35 °C (Table 2, entry
5). The reaction was complete within 30 min at -30 °C, and
fluoropyridine 16 was isolated in 78% yield following precipita-
tion by addition of aqueous 1 N HCl to the reaction mixture
and filtration. The material was recrystallized from 2-propanol
in 92% yield to remove residual quaternary ammonium salts
which interfered with the performance of the subsequent
enamide formation and hydrogenation steps (Scheme 2).
The synthesis of optically active R-methylpyridyl amines has
received much attention in the literature,16 and a number of
approaches including enzymatic resolution17 and diastereose-
lective reduction of chiral imines18 have been reported. An
attractive methodology for the preparation of chiral amines is
via enantioselective hydrogenation of N-acyl enamides. Rhodium-
catalyzed enantioselective hydrogenation of aryl N-acyl enam-
ides has been achieved with high levels of enantioselectivity
using a number of chiral phosphine ligands providing access to
a variety of chiral benzyl amines.19 A survey of the literature
revealed scant examples of the synthesis and subsequent
enantioselective hydrogenation of the corresponding pyridyl
N-acyl enamides.20 Despite their broad utility in asymmetric
hydrogenation, general methodologies for the synthesis of N-acyl
enamides still remain a significant challenge. Methodologies
involving organometallic addition to nitriles,21 a Heck reaction
of aryl triflates with N-vinyl acetamide,22 and Pd-catalyzed
coupling of enol triflates or tosylates with amides and carbam-
ates23 have been reported. We were interested in examining
organometallic methyl addition to nitrile 16 followed by in situ
acylation of the intermediate metallo-imine 18 as a straightfor-
ward approach to access N-acyl enamide 9 (Table 3). An initial
TABLE 2. Fluorodenitration of Pyridine 15
entry solvent TBAF T (°C)
time
conversion 15 17 yielda
1
2
3
4
5
THF
CH3CN 1.5 equiv
DMF 1.5 equiv -15
3.0 equiv -10 120 min
95%
99%
95%
88%
90%
15% 50%
11% 78%
10% 75%
2% 85%
1% 89%
20
30 min
60 min
15 min
45 min
DMFb 1.5 equiv -15
DMFb 3.0 equiv -30
a Assay yield measured by HPLC. b Predried DMF solution of TBAF.
Therefore, we were interested in evaluating fluorodenitration
as a potentially milder and scaleable method to prepare 16.
Fluorodenitrations have been reported for electron-poor aryl
substrates14 using a number of fluoride sources and have been
recently extended to pyridines using TBAF as the fluoride
source.15 Pyridine 15 underwent fluorodenitration in a number
of solvents using TBAF to give the desired fluoride 16. The
main byproduct observed was the corresponding 3-hydroxy
pyridine 17 (Table 2, entries 1-3). It was found that treatment
of a DMF solution of TBAF with molecular sieves to reduce
(16) Chelucci, G.; Baldino, S.; Chessa, S. Tetrahedron 2006, 62, 619.
(17) Skupinska, K. A.; McEachern, E. J.; Baird, I. R.; Skerlj, R. T.; Bridger,
G. J. J. Org. Chem. 2003, 68, 3546.
(18) Chelucci, G. Tetrahedron: Asymmetry 2005, 16, 2353.
(19) (a) Tang, W.; Zhang, X. Chem. ReV. 2003, 103, 3029. (b) Zeng, Q. H.;
Hu, X. P.; Duan, Z. C.; Liang, X. M.; Zheng, Z. J. Org. Chem. 2006, 71, 393.
(c) Enthaler, S.; Hagemann, B.; Junge, K.; Erre, G.; Beller, M. Eur. J. Org.
Chem. 2006, 2912. (d) Stemmler, R. T.; Bolm, C. Tetrahedron Lett. 2007, 48,
6189.
(20) (a) Cataviela, C.; Mayoral, J. A.; Mele´ndez, E.; Oro, L. A.; Pinillos,
M. T.; Uson, R. J. Org. Chem. 1984, 49, 2503. (b) Bozell, J. J.; Vogt, C. E.;
Gozum, J. J. Org. Chem. 1991, 56, 2584. (c) Do¨bler, C.; Kreuzfeld, H. J.;
Michalik, M.; Krause, H. W. Tetrahedron: Asymmetry 1996, 7, 117. (d) Wang,
B.; Wang, T. Patent WO2006/123113, 2006.
(21) Savarin, C. G.; Boice, G. N.; Murry, J. A.; Corley, E.; DiMichele, L.;
Hughes, D. Org. Lett. 2006, 18, 3903.
(12) Fukuhara, T.; Yoneda, N.; Suzuki, A. J. Fluorine Chem. 1987, 38, 435.
(13) (a) Roe, A.; Hawkins, G. F. J. Am. Chem. Soc. 1947, 69, 2443. (b)
Hawkins, G. F.; Roe, A. J. Org. Chem. 1949, 14, 328. (c) Link, W. J.; Borne,
R. F.; Sethiff, F. L. J. Heterocycl. Chem. 1967, 4, 641. (d) Milner, D. J. Synth.
Commun. 1992, 22, 73.
(14) (a) Finger, G. C.; Kruse, C. W. J. Am. Chem. Soc. 1956, 78, 6034. (b)
Adams, D. J.; Clark, J. H.; Nightingale, D. J. Tetrahedron 1999, 55, 7725.
(15) Kuduk, S. D.; DiPurdo, R. M.; Bock, M. G. Org. Lett. 2005, 7, 577.
(22) Harrison, P.; Meek, G. Tetrahedron Lett. 2004, 45, 9277.
(23) Brace, G. N.; Holmes, I. P. Synthesis 2005, 3229.
J. Org. Chem. Vol. 74, No. 12, 2009 4549

O’Shea et al.
TABLE 4. Enantioselective Hydrogenation of Enamide 9a
TABLE 3. Methyl Addition to Pyridine 16
entry
Me- (1.5 equiv)
Ac2O (equiv)
18a
9b
19c
1
2
3
4
5
6
7
8
MeMgBr (3 M THF)
MeMgI (3 M THF)
MeLi (1.6 M Et2O)
MeLi-LiBr (1.5 M Et2O)
MeMgCl (3 M THF)
MeMgCl (3 M THF)
MeMgCl (3 M THF)
MeMgCl (3 M THF)
2
2
ND
ND
ND
ND
97
29 ND
16 ND
2
3
ND
2
<2 ND
2
55 2.8
2.5
5
97
69
2
98
79 0.6
10
>99 97 0.2
a % conversion measured as methyl ketone by HPLC. b Assay yield
by HPLC. c % measured by HPLC.
investigation of solvents indicated that toluene was superior for
this reaction relative to methyl-t-butyl ether (MTBE), THF, 1,2-
dimethoxyethane, and 1,2-dichloroethane. The source of methyl
anion proved crucial with methylmagnesium bromide or iodide,
methyllithium, or methyllithium-lithium bromide complex,
affording poor yields of desired product (Table 3, entries 1-4).
Methylmagnesium chloride gave high conversion to the desired
metallo-imine (measured as the corresponding methyl ketone
by HPLC); however, the yield of the corresponding enamide
was only modest following quenching with 2 equiv of Ac2O
(Table 3, entry 5). Increasing the amount of Ac2O to 10 equiv
gave improved results, with reactions reaching 97% HPLC assay
yield of enamide 9 (Table 3, entries 6-8). Interestingly, using
this large excess of Ac2O led to lower formation of bis-
acetamide impurity 19, typically observed at <0.5 A% by HPLC
analysis. Thus, methylmagnesium chloride (3 M THF) was
added to a solution of 16 in toluene at -10 °C. After ∼30 min
complete conversion to 18 was observed by HPLC. Ac2O was
added at -10 °C and the mixture aged for 18 h to obtain 9 in
97% yield following workup. We observed variable performance
of enamide in the enantioselective hydrogenation step depending
on the method used to work up the reaction mixture. A
systematic analysis of workup procedures was performed to
correlate the effects of acetate, chloride, and magnesium ions
with respect to hydrogenation efficiency. We found that samples
with high acetate levels (∼1.5 wt %) performed poorly in the
hydrogenation step. Thus, working up the reaction by quenching
with 0.5 M aqueous NaHCO3 followed by washing with water,
10% aq Na2SO4, and water gave material of a consistent quality
which underwent efficient asymmetric hydrogenation.
a All reactions were run either using preformed catalyst or prepared
from 10% COD2RhBF4 and 10.5% ligand. The hydrogenations were
carried out at 90 psi in MeOH (25 mL/g of 1) at 25 °C and were
complete in 18 h.
FIGURE 2. Impurities from Suzuki coupling.
99% were observed with the BisP28 24 and Tangphos29 25
ligands (Table 4, entries 5, 6). On the basis of several factors
including cost, availability, and reaction performance we chose
to develop the Tangphos/(COD)2RhBF4 catalyst system for this
transformation.
Using (S,S,R,R)-Tangphos with Rh(COD)2[BF4], full conver-
sion was achieved at catalyst loadings as low as 0.1 mol %
while maintaining excellent enantioselectivity (>99%) using a
sample of 9 purified by column chromatography. However, the
reaction rate and catalyst performance were dependent upon the
quality of enamide 9, and upon scale up (and without chroma-
tography) an increased catalyst charge (ca. 0.3 mol %) was
We began our investigation of the enantioselective hydroge-
nation of enamide 9 by completing a comprehensive screen
using a variety of chiral phosphines and rhodium sources at 10
mol % and loading in methanol at 90 psi hydrogen and 25 °C.
Clean hydrogenation with high enantiomeric excess was ob-
served for a variety of Rh-phosphine catalysts, and a selection
of the best results are summarized in Table 4. Good ee (84%)
was observed using DiPAMP24 ligand 20 (Table 4, entry 1).
High ee’s of 92%, 94%, and 96% were obtained using MeBPE25
21, Catasium MN26 derivative 22, and EtDuphos27 23, respec-
tively, as ligands (Table 4, entries 2, 3, 4). However, ee’s of
(24) (a) Knowles, W. S. Acc. Chem. Res. 1983, 16, 106. (b) Knowles, W. S.;
Sabacky, M. J.; Vineyard, B. D.; Weinkauff, D. J. J. Am. Chem. Soc. 1975, 97,
2567.
(25) Burk, M. J.; Fraster, J. E.; Harlow, R. L. Organometallics 1990, 9, 2653.
(26) Holz, J.; Zayas, O.; Jiao, H.; Baumann, W.; Spannenberg, A.; Monsees,
A.; Riermeier, T. H.; Almena, J.; Kadyrov, R.; Bo¨rner, A. Chem.sEur. J. 2006,
12, 5001.
(27) (a) Burk, M. J. J. Am. Chem. Soc. 1991, 113, 8518. (b) Burk, M. J.;
Feaster, J. E.; Nugent, W. A.; Harlow, R. L. J. Am. Chem. Soc. 1993, 115, 10125.
(28) Miyazaki, T.; Sugaware, M.; Danjo, H.; Imamoto, T. Tetrahedron Lett.
2004, 45, 9341.
(29) Tang, W.; Zhang, X. Angew. Chem., Int. Ed. 2002, 41, 1612.
4550 J. Org. Chem. Vol. 74, No. 12, 2009

Synthesis of a Potent Bradykinin B1 Antagonist
SCHEME 3. Synthesis of Biaryl 27
SCHEME 4. Synthesis of Isoxazole 3
required to achieve full conversion. It is not atypical for the
performance of low catalyst loading reactions to be impacted
by the quality of the materials used. However, it is expected
that following further optimization of the workup conditions
for enamide 9 and hydrogenation reaction parameters the catalyst
loading would be reduced. Thus, treatment of 9 in MeOH with
(S,S,R,R)-TangphosRh(COD)[BF4] and 6.9 bar of H2 at 25 °C
for 4 h gave the desired amide 7 in 94% yield and 99.5% ee.
Synthesis of Biaryl Amine 2. A study of the Suzuki coupling
between aryl boronic acid 6 and bromopyridine amide 7
identified the optimal reaction conditions as using Pd(OAc)2
and PPh3 in a 1:2 ratio with 2 M K3PO4 in refluxing THF, using
chromatographed 7 (Scheme 3). However, the use of crude 7
(directly from the hydrogenation reaction without workup) gave
poor results in the Suzuki coupling requiring >6 mol % of Pd
to obtain >95% conversion. A correlation was observed between
the level of residual Rh in 7 and the performance in the cross
coupling. This necessitated the development of a workup to
reduce Rh levels. Attempts to purify 7 and reduce Rh levels by
extractive workup or crystallization were unsuccessful. However,
Rh levels were significantly reduced by treatment with silica
gel and activated carbon. Thus, 7 was purified by filtration of
an isopropyl acetate (IPAc) solution through a pad of silica gel
followed by treatment of the filtrate with activated carbon (Darco
KB-B). Material obtained from this protocol contained <20 ppm
Rh and performed reproducibly in the Suzuki coupling, requiring
∼3 mol % Pd to obtain complete conversion. While this loading
may appear high, further optimization of the reaction conditions
would be expected to lower the required catalyst levels
significantly.
soon as it reaches >99% conversion, and this was achieved by
cooling to <45 °C, where formation of the 28 and 29 did not
occur. Thus, heating a mixture of 6 and 7 with Pd(OAc)2 (3
mol %) and PPh3 (6 mol %) in THF with aqueous K2PO3 for
1 h afforded complete conversion to biaryl 26. Following
extractive workup, 26 was obtained in 97% HPLC assay yield
as a THF solution which was used directly in the amide
hydrolysis step.
Hydrolysis of the acetamide 26 to the primary amine 2 was
completed in a straightforward manner by addition of 5 equiv
of aqueous 6 N HCl to the THF solution of 26. Heating at 65
°C for 15 h gave the desired amine in 95% assay yield by HPLC.
Salt formation of the amine 2 was used as a key purification
step to remove triaryl impurities as well as residual palladium.
Moreover, this step was required to afford a stable form of the
amine. Amine 2 as a free base was unstable at 20 °C and
degraded at a rate of >1 A% loss per day by HPLC analysis. A
screen of acids indicated that the (S)-mandelate salt 27 was
stable and crystalline. Isolation of 27 following precipitation
gave high purity salt and reduced residual Pd levels to ∼100
ppm. Thus, a water/2-propanol solution of (S)-mandelic acid
was added to a 2-propanol solution of 2. The mixture was heated
to 75 °C and cooled over 16 h to 20-25 °C, and the precipitate
was isolated by filtration to give 27 in 82% yield.
Synthesis of Isoxazole 3. Alkylation of isoxazole 30 with
MeI30 suffered from poor regioselectivity (∼1.5:1) leading to a
low yield of the desired O-methyl isomer 31 (Scheme 4).
Methylation of isoxazole 30 with 1.1 equiv of dimethyl sulfate
in the presence of solid K2CO3 in DMF at -5 °C gave much
improved selectivity (85:15) and afforded the desired O-methyl
isoxazole 31 in 84% HPLC assay yield. Addition of aqueous 2
N NaOH directly to the alkylation reaction mixture in DMF
led to hydrolysis of the methyl ester in ∼30-60 min at rt, and
following aqueous workup the desired acid 32 was obtained as
a toluene solution in 82% yield (by HPLC) over two steps. The
An investigation of the Pd loading and reaction times
indicated that longer reaction times (>2 h) significantly increased
the formation of triaryl impurities 28 and 29 (Figure 2). The
formation of the triaryls is more significant after the bromo cross
coupling is complete, so it is important to stop the reaction as
J. Org. Chem. Vol. 74, No. 12, 2009 4551

O’Shea et al.
temperature rose to 6 °C. After aging 17 h, the mixture was warmed
to 18 °C, and 1 N HCl (52.5 L) was added over 50 min. After
aging 16 h, the mixture was cooled to 2.5 °C and filtered, and the
cake was washed with DMF/water (10% (v/v), 2 × 6 L) to afford,
after drying in a vacuum oven at 35 °C for 40 h, 2.26 kg of 16
(91.6 wt %, 78% yield), which was obtained as a brown solid. A
2.06 kg portion of crude 16 (1.88 kg at 91.6 wt %, 9.46 mol) was
suspended in 2-propanol (9.4 L) in a visually clean and dry 50 L
RBF. With mechanical stirring, the mixture was warmed to 67 °C
over 20 min to give a dark brown solution. The mixture was cooled
to 37 °C over 1 h at which point solids began to precipitate. Water
(200 mL) was added, and the mixture was aged for 5 min before
adding additional water (18.6 L) over 75 min. The mixture was
cooled to 18 °C and filtered, and the cake was washed with H2O/
2-propanol (2:1 (v/v), 2 × 5 L) and water (2 × 5 L). The solids
were dried at 35 °C under vacuum for 20 h to afford 16 at 1.73 kg
and 92% yield. An analytically pure sample was obtained by
sublimation and was characterized as follows. Mp 103-105 °C;
1H NMR (400 MHz, acetone-d6) δ 8.74 (s, 1 H), 8.36 (dd, 1 H, J
) 1.8, 8.4 Hz); 13C NMR (125 MHz, acetone-d6) δ 161.4 (d, J )
274 Hz), 149.5 (d, J ) 4 Hz), 192.2 (d, J ) 21 Hz), 126.2 (d, J )
4 Hz), 121.7 (d, J ) 16 Hz), 113.6 (d, J ) 5 Hz); 19F NMR (377
MHz, acetone-d6) δ -120.5 (d, J ) 9 Hz). Anal. Calcd for
C6H2BrFN2: C, 35.85; H, 1.00; N, 13.94. Found: C, 35.53; H, 0.83;
N, 13.93. HRMS (ESI) calcd for C6H2BrFN2 [M - H] 198.9312,
found 198.9337; IR (film, cm-1) 3030, 1572, 1435, 1407, 1283,
1155, 1084, 908. HPLC, Zorbax Rx-C8 4.6 mm × 25 cm column;
eluents, A, 0.1% aqueous H3PO4; B, acetonitrile; 2 mL/min;
gradient, A/B 70:30 to 5:905 over 20 min; λ ) 220 nm; temperature,
35 °C; tR, 16 ) 6.2 min.
N-methyl isomer was highly water-soluble and readily separated
in the aqueous layer during workup.
Peptide coupling was performed under Schotten-Bauman
conditions with the first activation of the isoxazole acid 32 using
oxalyl chloride in MTBE, followed by addition over 1 h to a
solution of cyclopropane amino ester 33 in aqueous K2CO3 at
0-5 °C. Following removal of the volatiles the dipeptide ester
34 was isolated as a solid in 88% yield. Hydrolysis of the ethyl
ester was performed with aqueous NaOH in MeOH at rt and
gave the desired dipeptide acid 3 in 84% yield and high purity
(99.7 wt % by HPLC).
SCHEME 5. End Game
Synthesis End Game. Mandelate salt 27 was treated with
K2CO3 to give free amine 2 in quantitative yield. Coupling with
acid 3 was effected using 1-ethyl-3-(3-dimethylaminopropyl)-
carbodiimide hydrochloride (EDC·HCl) and 10 mol % 1-hy-
droxybenzotriazole (HOBT) with N-methyl morpholine (NMM)
as base (Scheme 5). Thus, to a DMF solution of 2 and 3 were
added HOBT and NMM, followed by portion-wise addition of
EDC·HCl. After 1 h at 30 °C the coupling reaction was
complete. The product was precipitated directly from the
reaction mixture by addition of water to give 1 as a white solid
in 96% yield and >99% purity by HPLC.
In summary, a practical and efficient chromatography-free
synthesis of bradykinin B1 inhibitor 1 was developed. A
convergent strategy was utilized involving synthesis of three
fragments 3, 6, and 7. Cross coupling of 6 and 7 followed by
amidation with 3 enabled efficient assembly of 1. The key to
the success of the synthesis was the development of a fluo-
rodenitration step to install the fluorine in pyridine 7 and a
catalytic enantioselective hydrogenation of N-acyl enamide 9
to set the stereochemistry. The synthesis was completed in 19
steps total and 35% overall yield (in 9 steps) from commercially
available pyridine 10.
N-[1-(5-Bromo-3-fluoropyridin-2-yl)vinyl]acetamide (9). Au-
thor: A 50 L RBF was charged with nitrile 16 (1.30 kg, 1.35 kg @
96.4 wt %, 6.48 mol) and toluene (13 L). The batch was cooled to
-10 °C, and MeMgCl (3.24 L, 9.72 mol, 3 M in THF) was added
over 35 min during which time the temperature rose to 0 °C. The
batch was allowed to stir at -10 °C for 1 h. HPLC showed >99.8%
conversion to the magnesium amide adduct as ascertained by
hydrolysis of an aliquot to the corresponding ketone. Acetic
anhydride (6.13 L, 64.8 mol) was added over 30 min via an addition
funnel. An exotherm was observed during addition of the first
portion of Ac2O (-10 to 0 °C). The exotherm subsided rapidly
afterward, and the internal temperature returned to -10 °C. The
batch was stirred at -10 °C for 18 h. HPLC analysis showed >99%
conversion to enamide 9. The reaction was quenched with 0.5 M
NaHCO3 (6.5 L), the batch was stirred at rt for 30 min, and the
layers were cut. The upper organic layer was washed with water
(6.5 L), 10% aqueous Na2SO4 (2 × 6.5 L), and water (2 × 6.5 L).
The batch was concentrated, flushed with toluene (6.5 L), and
flushed with MeOH (6.5 L). HPLC assay indicated 1.64 kg of 9,
with 97% yield as an oil. A pure sample (solid) was obtained
Experimental Section
1
following chromatography. Mp 46-47 °C; H NMR (400 MHz,
5-Bromo-3-fluoropyridine-2-carbonitrile (16). A 20 L round-
bottomed flask (RBF) was charged with TBAF (10.3 kg, 39.5 mol)
and DMF (9 L) under an inert atmosphere. The mixture was stirred
for 6 h to dissolve the TBAF. KF analysis showed 8.2% water.
Molecular sieves (4 Å, 8-12 mesh, 3.44 kg) were added, and the
mixture was stirred for 9 h. KF analysis showed 6.5% water. The
mixture was decanted off of the sieves by vacuum transfer into a
100 L reaction flask equipped with a mechanical stirrer, a
thermocouple, and an addition funnel. The sieves were washed with
DMF (2 × 4 L). Sulfuric acid (0.014 L, 0.263 mol) was added,
and the mixture was cooled to -38 °C. A solution of nitropyridine
15 in DMF/IPAc (45.4 wt %, 93/7 DMF/IPAc, 6.61 kg, 13.2 mol)
was added via addition funnel over 20 min maintaining the internal
temperature below -33 °C. HPLC analysis of a sample after 10
min showed an 85:8 ratio of product to starting material. After aging
20 min, 2 N HCl (10.8 L) was added over 30 min. The internal
CDCl3) δ 8.84 (bs, 1H), 8.44 (d, 1H, J ) 0.8 Hz), 7.68 (dd, 1H, J
) 1.8, 11.0 Hz), 6.77 (d, 1H, J ) 4.8 Hz), 5.88 (s, 1H), 2.20 (s,
3H); 13C NMR (125 MHz, CDCl3) δ 169.0, 157.0 (J ) 269 Hz),
144.5 (J ) 5 Hz), 138.8 (J ) 7.9 Hz), 133.1 (J ) 7.7 Hz), 127.6
(J ) 23.8 Hz), 118.9 (J ) 3.2 Hz), 105.9 (J ) 18.0 Hz), 24.9; 19
F
NMR (377 MHz, CDCl3) δ -113.4; HRMS (ES) calcd for
C9H8BrFN2O [M + H]+ 258.9882, found 258.9880; IR (thin film,
cm-1) 3351, 3057, 1686, 1506, 897. HPLC, Zorbax Rx-C8 4.6 mm
× 25.0 cm column; eluents, A, 0.1% aqueous H3PO4; B, acetonitrile;
2 mL/min; gradient, A/B 70:30 to 5:95 over 25 min; λ ) 220 nm;
temperature, 35 °C; tR, enamide 9 ) 5.22 min, ketone ) 5.92 min,
bis-acetamide 19 ) 7.37 min.
N-[(1R)-1-(5-Bromo-3-fluoropyridin-2-yl)ethyl]acetamide (7).
In a nitrogen filled glovebox (<10 ppm O2), (S,S,R,R)-Tangphos
(12.89 g, 45.0 mmol) was combined with (COD)2RhBF4 (17.40 g,
42.8 mmol). Methanol (400 mL) was added to dissolve, and the
solution was aged for 1 h. A portion of the catalyst solution (177.4
g, 21.7 mmol ((S,S,R,R)-Tangphos)Rh(COD)[BF4]) was transferred
(30) Schlewer, G.; Krogsgaard-Larsen, P. Acta Chem. Scand., Ser. B 1984,
B38, 815.
4552 J. Org. Chem. Vol. 74, No. 12, 2009

Synthesis of a Potent Bradykinin B1 Antagonist
to a 500 mL stainless steel vessel. The catalyst solution was further
diluted with methanol (100 mL). To a separate 150 mL stainless
steel vessel was added an additional charge of methanol (100 mL).
These two vessels were connected with a ball valve separating the
two vessels. The enamide 9 (3.66 kg, 54 wt % in MeOH, 1.97
assay kg, 7.6 mol) was drawn into a 40 L stirred autoclave via
vacuum followed by a methanol (15 L) rinse. The solution was
then degassed with nitrogen (3×). The stainless steel vessels
containing the catalyst solution were connected to the autoclave
via flexible tubing. The autoclave was placed under partial vacuum,
and the catalyst solution was drawn into the autoclave followed by
the MeOH rinse (100 mL). The solution was degassed with H2
(6.9 bar, 3×) and the final pressure adjusted to 1.4 bar. The reaction
temperature was set to 25 °C and agitation initiated. The reaction
pressure was increased to 6.9 bar after 20 min and aged for 4 h.
The reaction was sampled and conversion determined to be >99.5%
with 99.5% ee by HPLC. The HPLC assay yield of 7 was 1.87 kg,
93.9%. A second reaction was performed which yielded 1.1 kg,
94%, 99.5% ee and combined with the first MeOH solution for
workup. The MeOH was removed in vacuo, and the residue was
flushed with IPAc (12 L). Additional IPAc (3 L) was added to
obtain a 2 mL/g solution. The dark solution was filtered through a
pad of silica gel (8 kg) and the pad rinsed with IPAc (80 L).
Activated carbon Darco KB-B (1.5 kg) was added, and the mixture
was stirred under N2 at rt for 16 h. The slurry was filtered on a pad
of filter aid Solka floc, and the cake was washed with IPAc. The
solvent was removed in vacuo to afford 7 as a yellow solid (2.9
kg, 97% yield by HPLC assay). Mp 94-95 °C; 1H NMR (500 MHz,
CDCl3) δ 8.43 (s, 1H), 7.59 (dd, 1H, J ) 8.6, 1.8 Hz), 6.75 (br s,
1H), 5.44 (quint, 1H, J ) 6.7 Hz), 2.04 (s, 3H), 1.42 (d, 3H, J )
6.7 Hz); 13C NMR (125 MHz, CDCl3) δ 169.2, 155.3 (d, J ) 264
Hz), 148.1 (d, J ) 15 Hz), 145.7, 126.4 (d, J ) 21 Hz), 118.6,
44.0, 23.3, 21.3; 19F NMR (470 MHz, CDCl3) δ -123.0; HRMS
7
calcd for C9H10BrFN2O [M + Li] 267.0121, found 267.0124; IR
(KBr, cm-1) 3280, 1635, 1540, 1455, 1400, 1372; [R]2D0 - 103.66
(MeOH, c 12.2). HPLC, Inertsil ODS-3 4.6 mm × 25.0 cm column;
eluents, A, 0.1% aqueous H3PO4; B, acetonitrile; 1.5 mL/min;
gradient, A/B 20:80 to 10:90 over 20 min; λ ) 210 nm; temperature,
25 °C; tR, acetamide 7 ) 7.5 min, enamide 9 ) 9.8 min. Chiral
HPLC, Chiralpak AD-H 4.6 mm × 25.0 cm column; eluents, A,
n-heptane; B, EtOH; 2 mL/min; isocratic, A/B 90:10; λ ) 224 nm;
temperature, 20 °C; tR, (R)-7 ) 9.5 min, enamide 9 ) 14.0 min,
(S)-7 ) 23.0 min.
Supporting Information Available: Experimental proce-
dures for compounds 1, 2, 6, 11, 12, 13, 14, 15, 26, 27, 31, 32,
and 34. Copies of 1H, 13C, and 19F NMR spectra. This material
is available free of charge via the Internet at http://pubs.acs.org.
JO802772D
J. Org. Chem. Vol. 74, No. 12, 2009 4553