Crystallization-induced chiral inversion as the key step for synthesis of (S)-2-acetylthio-3-phenylpropanoic acid from L-phenylalanine

Article abstract of DOI:10.1021/ol0489806

(Chemical Equation Presented) A novel crystallization-induced chiral inversion of (S)-2-bromo-3-phenylpropanoic acid to its (R)-enantiomer with excellent enantiomeric excess (96-99%) is achieved. Optically pure (S)-2-acetylthio-3-phenylpropanoic acid can

Full text of DOI:10.1021/ol0489806

ORGANIC
LETTERS
2004
Vol. 6, No. 19
3233-3235
Crystallization-Induced Chiral Inversion
As the Key Step for Synthesis of
(S)-2-Acetylthio-3-phenylpropanoic Acid
from L-Phenylalanine
,†
Jason G. Chen,† Jingyang Zhu,* Paul M. Skonezny,† Victor Rosso,‡ and
John J. Venit‡
Chemical, Fermentation and Biocatalysis DeVelopment Labs, Bristol-Myers Squibb
Company, P.O. Box 4755, Syracuse, New York 13221, and Process Research and
DeVelopment Department, Bristol-Myers Squibb Company, 1 Squibb DriVe,
New Brunswick, New Jersey 08903
jingyang.zhu@bms.com
Received June 2, 2004
ABSTRACT
A novel crystallization-induced chiral inversion of (S)-2-bromo-3-phenylpropanoic acid to its (R)-enantiomer with excellent enantiomeric excess
(96−99%) is achieved. Optically pure (S)-2-acetylthio-3-phenylpropanoic acid can be prepared in good yield from inexpensive and commercially
available L-phenylalanine via diazotization/bromination, chiral inversion, and thioacetate substitution reactions.
Omapatrilat and Gemopatrilat were developed as vaso-
peptidase inhibitors for treatment of hypertension, congestive
heart failure, and renal disease.1 During the course of
developing these inhibitors, an efficient synthesis of the
common intermediate, (S)-2-acetylthio-3-phenylropanoic acid
(Figure 1, 1), was needed.
An existing synthetic route in the literature for the acid
starts from unnatural D-phenylalanine.2 D-phenylalanine is
first converted to (R)-2-bromo-3-phenylpropanoic acid through
diazotization/bromination. The acid obtained is then subjected
to nucleophilic substitution with potassium thioacetate to
afford 1. Although efficient, this approach is not desirable
due to the high cost of D-phenylalanine.
Figure 1.
† Chemical, Fermentation and Biocatalysis Development Labs.
‡ Process Research and Development Department.
(1) Robl, J. A. U.S. Patent 5,508,272, 1996.
Alternative synthetic routes were investigated in our
(2) Karanewsky, D. S.; Robl, J. A. U.S. Patent 5,552,397, 1996.
laboratory, and we reported one approach to this intermediate
10.1021/ol0489806 CCC: $27.50 © 2004 American Chemical Society
Published on Web 08/19/2004

via enzymatic resolution.3 Here, we report a novel crystal-
lization-induced chiral inversion approach to the acid using
inexpensive L-phenylalanine as starting material.
The desired precursor to 1 is (R)-2-bromo-3-phenyl-
propanoic acid (Scheme 1, 3). Recently, crystallization-
tetraethylammonium bromide (TEAB) as a bromide ion
source.
We then reasoned that under dynamic kinetic resolution
conditions, the undesired (S)-enantiomer could be directly
converted to its (R)-enantiomer. This concept can be
described as a crystallization-induced chiral inversion, and
this would ultimately allow us to start with the inexpensive
L-phenylalanine.
Scheme 1. Crystallization-Induced Dynamic Resolution of 2
It is known that diazotization/bromination of an amino acid
results in retention of configuration. Converting L-phenyla-
lanine via diazotization/bromination5 leads to the (S)-
enantiomer of the precursor (S)-2-bromo-3-phenylpropanoic
acid 4. Chiral inversion of 4 was then investigated using
optimal reaction conditions identified for dynamic resolution
of 2: (R)-bornylamine as a chiral amine, acetonitrile as a
solvent, and TEAB as the bromide ion source. Under these
conditions, 4 was converted to 5, the (R)-bornylamine salt
of 4, with high stereospecificity (>96% ee) and good yield
(78%).
The enrichment of the desired bornylamine salt 5 is likely
due to its lower solubility in acetonitrile. It was found that
the undesired salt, the (R)-bornylamine salt of (R)-2-bromo-
3-phenyl-propanoic acid, was almost three times more
soluble than 5 (0.95 vs 0.35 mg/mL). Since the solubility
for both salts in acetonitrile is low, a relatively long reaction
time (∼48 h) was needed to achieve the desired level of
enantiomeric inversion. The bromide source also appeared
to play an important role in achieving the optimal result.
For example, under similar conditions TEAB gave a higher
yield than TBAB (entries 13 and 2). The differences in
isolation yields may primarily be attributed to the differences
in solubility of the salt in different isolation medium, since
there was no difference in reaction profiles. However, when
no bromide was added, the reaction gave only ∼40% ee
under similar conditions.
(R)-2-bromo-3-phenylpropanoic acid (3) was obtained by
dissolving 5 in water, acidifying with methanesulfonic acid,
and extracting the product with MTBE. The ee of the acid
obtained was usually between 96 and 98%. The (R)-
bornylamine was also recovered by adjusting pH to 10-13
and then extracting the free amine into MTBE. The recovered
bornylamine was ∼97% pure, and the recovery was >92%.
The bromo acid 3 was subsequently converted to 1 by
nucleophilic substitution with KSAc to furnish the desired
common intermediate in ∼87-90% yield. The ee of the acid
obtained was between 92 and 95%. Recrystallization of the
acid in MTBE/heptanes further increased the ee to ∼99%.
induced dynamic resolution (CIDR) via a pair of salt
diastereomers has gained popularity in asymmetric synthesis.4
Preliminary lab studies indicated that some chiral amines
were effective for dynamic kinetic resolution of the racemic
2-bromo-3-phenylpropanoic acid (2), which was available
from our earlier work.3 Among more than 40 chiral amines
screened, (R)-bornylamine was found to be the most effec-
tive.
We further optimized the dynamic kinetic resolution
conditions with various combinations of chiral amines,
solvents, and halide sources. Selected combinations are
compiled in Table 1. Excellent enantiomeric excess (∼96%,
Table 1. Optimization of CIDR Conditions for 26
entry
bromide source7
solvent
yield(%)a
ee (%)8
1
2
3
4
5
6
7
8
9
10
11
12
13
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TMAB
TEAB
THAB
MTOAB
TOAB
TEAB
butyl acetate
acetonitrile
MTBE
ethyl acetate
THF
40.00
45.49
78.93
69.57
nab
44.15
56.19
77.19
77.59
42.81
40.00
64.75
76.0
87.0
93.5
85.1
88.6
na
84.1
78.1
74.3
89.4
89.2
84.5
87.6
96.0
butanol
isobutanol
butyl acetate
butyl acetate
butyl acetate
butyl acetate
butyl acetate
a ceton itr ile
a Yield was calculated on the basis of the chiral amine input and was
not optimized. b No salt formation.
Table 1, entry 13) could be obtained when (R)-bornylamine
was used as the chiral amine, acetonitrile as the solvent, and
(5) Allegrini, P.; Soriato, G. WO Patent 99/42438, 1999.
(6) Typical reaction conditions: (R)-bornylamine (0.95-1 equiv) was
slowly delivered over 24 h to a reaction mixture of 2 (1 equiv) and catalyst
(0.1 equiv) in acetonitrile at 50-60 °C. The reaction mixture was
continuously stirred for an additional 24 h. After cooling to room
temperature, the slurry was filtered, and the solid was washed with
acetonitrile and dried.
(7) Abbreviations for bromide sources: TBAB ) tetrabutylammonium
bromide; TMAB ) tetramethylammonium bromide; TEAB ) tetraethyl-
ammonium bromide; THAB ) tetrahexylammonium bromide; MTOAB )
methyltrioctylammonium bromide; TOAB ) tetraoctylammonium bromide.
(8) Ee was determined by chiral HPLC using a Chiralcel AD column,
250 × 4.6 mm. Mobile phase: 97.9% hexane, 2% absolute ethanol, and
0.1% TFA. Flow rate: 1 mL/min. Detector: UV at 230 nm.
(3) Zhu, J.; You, L.; Zhao, S.; White, B.; Chen, J. G.; Skonezny, P. M.
Tetrahedron Lett. 2002, 43, 7585.
(4) (a) Reider, P. J.; Davis, P.; Hughes, D. L.; Grabowski, E. J. J. Org.
Chem. 1987, 52, 955. (b) Moseley, J. D.; Williams, B. J.; Owen, S. N.;
Verrier, H. M. Tetrahedron: Asymmetry 1996, 7, 3351. (c) Alabaster, R.
J.; Gibson, A. W.; Johnson, S. A.; Edwards, J. S.; Cottrell, I. F. Tetrahedron:
Asymmetry 1997, 8, 447. (d) Maryanoff, C.; Scott, A. L.; Shah, R. D.; Villani
Jr, F. J. Tetrahedron: Asymmetry 1998, 9, 3247. (e) Macdonald, S. J. F.;
Clarke, G. D. E.; Dowle, M. D.;. Harrison, L. A.; Hodgson, S. T.; Inglis,
G. G. A.; Johnson, M. R.; Shah, P.; Upton, R. J.; Walls, S. B. J. Org.
Chem. 1999, 64, 5166. (f) Aelterman, W.; Lang, Y.; Willemsens, B.;
Vervest, I.; Leurs, S.; de Knaep, F. Org. Process. Res. DeV. 2001, 5, 467.
3234
Org. Lett., Vol. 6, No. 19, 2004

Scheme 2. Synthesis of 1 from L-Phenylalanine
The entire scheme, including the recycle of the bornylamine,
Acknowledgment. The authors thank Mr. Jun Qiu for
is depicted in Scheme 2.
preliminary chiral amine screening experiments.
In conclusion, optically pure 3 can be obtained from the
corresponding optically pure 4 via a crystallization-induced
chiral inversion with an appropriate chiral amine, and a route
to 1 utilizing inexpensive L-phenylalanine has been demon-
strated. To the best of our knowledge, this is the first time
that an R-substituted chiral carboxylic acid was synthesized
by inverting its enantiomer through a crystallization-induced
chiral inversion.
Supporting Information Available: Experimental pro-
cedures and characterization data for compounds 1 and 3-5.
This material is available free of charge via the Internet at
http://pubs.acs.org.
OL0489806
Org. Lett., Vol. 6, No. 19, 2004
3235