Asymmetric bioreductions of β-nitro acrylates as a route to chiral β2-amino acids

Article abstract of DOI:10.1021/ol062612f

(Chemical Equation Presented) Reductions of β-nitroacrylates by Saccharomyces carlsbergensis old yellow enzyme is the key step in a concise route to optically active β²-amino acids. The enzymatic reductions occur with 87-96% ee, with larger sub

Full text of DOI:10.1021/ol062612f

ORGANIC
LETTERS
2
006
Vol. 8, No. 26
131-6133
Asymmetric Bioreductions of
Acrylates as a Route to Chiral
Acids
â
-Nitro
2
6
â
-Amino
Magdalena A. Swiderska and Jon D. Stewart*
Department of Chemistry, UniVersity of Florida, GainesVille, Florida 32611
jds2@chem.ufl.edu
Received October 23, 2006
ABSTRACT
Reductions of
â
-nitroacrylates by Saccharomyces carlsbergensis old yellow enzyme is the key step in a concise route to optically active
2
â
-amino acids. The enzymatic reductions occur with 87−96% ee, with larger substrates providing greater stereoselectivities. This work extends
enantioselective enzymatic alkene reductions to include acyclic systems with weakly coordinating substituents.
2
2
-Substituted derivatives of 3-aminopropionic acid (â -amino
potential of enzyme-mediated alkene reductions to supply
chiral intermediates. Here, we describe how the best-studied
enzyme in this classsSaccharomyces carlsbergensis old
yellow enzymescan be used in stereoselective â -amino acid
synthesis.
acids) have become important synthetic targets since peptides
containing these residues can assume different structures than
those of R-amino acids and their peptide bonds resist protease
2
1
degradation. In contrast to R-amino acids, for which a
variety of reliable, high-yielding methods have been devised,
â -amino acids remain challenging synthetic targets. Recent
The required â-nitroacrylates 3a-g were prepared by a
two-step sequence (Scheme 1). Henry reactions were carried
out using a solid-phase base along with the appropriate
R-keto ester (either commercially available or prepared by
2
2
attention has focused on catalytic strategies involving
3
4
conjugate additions to amino- and nitroacrylates. Asym-
metric hydrogenation of aminoacrylates has also been
reported,5 and this can provide very high enantioselectivities
12
13a
the method of Macritchie et al. ) and nitroalkane. Spectral
,6
data for nitroaldol adducts 2a-e,g matched those reported
7
8
9-11
13,14
in favorable cases. We and others
have investigated the
previously.
These were converted to the corresponding
*
(
Phone: 352-846-0743. Fax: 352-846-2095.
1) Lee, H.-S.; Park, J.-S.; Kim, B. M.; Gellman, S. H. J. Org. Chem.
003, 68, 1575-1578.
(7) Huang, H.; Liu, X.; Deng, J.; Qiu, M.; Zheng, Z. Org. Lett. 2006, 8,
3359-3362.
2
(8) Swiderska, M. A.; Stewart, J. D. J. Mol. Catal. B: Enzym. 2006, 42,
52-54.
(
2) (a) Cardillo, G.; Tomasini, C. Chem. Soc. ReV. 1996, 117-128. (b)
Sewald, N. Angew. Chem., Int. Ed. 2003, 42, 5794-5795. (c) Lelais, G.;
Seebach, D. Biopolymers 2004, 76, 206-243.
(9) Kataoka, M.; Kotaka, A.; Hasegawa, A.; Wada, M.; Yoshizumi, a.;
Nakamori, S.; Shimizu, S. Biosci. Biotechnol. Biochem. 2002, 66, 2651-
2657.
(10) Wada, M.; Yoshizumi, A.; Noda, Y.; Kataoka, M.; Shimizu, S.;
Takagi, H.; Nakamori, S. Appl. EnViron. Microbiol. 2003, 69, 933-937.
(11) M u¨ ller, A.; Hauer, B.; Rosche, B. J. Mol. Catal. B: Enzym. 2006,
38, 126-130.
(12) Macritchie, J. A.; Silcock, A.; Willis, C. L. Tetrahedron: Asymmetry
1997, 8, 3895-3902.
(13) (a) Ballini, R.; Fiorini, D.; Palmieri, A. Tetrahedron Lett. 2004,
7027-7029. (b) Palacios, F.; de los Santos, J.; Aparacio, D. ARKIVOC
2005, 405-414. (c) Du, D.-M.; Lu, S.-F.; Fang, T.; Xu, J. J. Org. Chem.
2005, 70, 3712-3715. (d) Li, H.; Wang, B.; Deng, L. J. Am. Chem. Soc.
2006, 128, 732-733.
(
3) (a) Sibi, M. P.; Patil, K. Angew. Chem., Int. Ed. 2004, 43, 1235-
1
2
5
238. (b) Sibi, M. P.; Tatamidani, H.; Patil, K. Org. Lett. 2005, 7, 2571-
573. (c) Sibi, M. P.; Patil, K. Tetrahedron: Asymmetry 2006, 17, 516-
19.
(4) (a) Eilitz, U.; Lessmann, F.; Seidelmann, O.; Wendisch, V. Tetra-
hedron: Asymmetry 2003, 14, 189-191. (b) Eilitz, U.; Lessmann, F.;
Seidelmann, O.; Wendisch, V. Tetrahedron: Asymmetry 2003, 14, 3095-
3
097. (c) Rimkus, A.; Sewald, N. Org. Lett. 2003, 5, 79-80.
(
5) Saylik, D.; Campi, E. M.; Donohue, A. C.; Jackson, W. R.; Robinson,
A. J. Tetrahedron: Asymmetry 2001, 12, 657-667.
6) Elaridi, J.; Thaqi, A.; Prosser, A.; Jackson, W. R.; Robinson, A. J.
Tetrahedron: Asymmetry 2005, 16, 1309-1319.
(
1
0.1021/ol062612f CCC: $33.50
© 2006 American Chemical Society
Published on Web 12/01/2006

reduction products by chiral-phase chromatography, the crude
materials were hydrogenated in the presence of Raney-Ni
to the corresponding amines (75-85% yield). Enantiomer
separations were then possible by chiral-phase GC following
derivatization with trifluoroacetic anhydride. Good optical
purities were obtained from 2-alkyl-substituted nitroalkenes
Scheme 1
1
9
3c-f; by contrast, 3-alkyl-substituted products were obtained
in essentially racemic form (Table 1). The absolute configu-
Table 1. â-Nitroacrylate Reductions by S. carlbergensis Old
Yellow Enzyme
starting
compd
conversion (%)
after 8 h
% ee
(absolute config)
a
[R]Db
(
(
(
E)-3a
E)-3b
Z)-3c
>98
50
>98
>98
>98
>98
8 (R)
8 (R)
91 (R)
94 (R)
96 (R)
-13.0 (c 0.94)
-2.7 (c 1.0)
+1.2 (c 1.0)
-1.6 (c 1.0)
nitroacrylates via mesylate derivatives.15 Both 2-alkyl-
(Z)-3d
(
(
Z)-3e
Z)-3f
substituted nitroacrylates 3a and b were obtained predomi-
nantly in the (E)-form, whereas the (Z)-isomers predominated
for 3-substituted alkenes 3c-g.16 Because olefin geometry
directly impacted the stereoselectivity of enzymatic reduc-
tions, the major alkene isomers were chromatographically
enriched (>95% geometric purity).
a
Determined by chiral-phase GC following nitro group reduction and
derivatization with trifluoroacetic anhydride. b Measured in aqueous solution
at the indicated concentrations from hydrochloride salts at room tempera-
ture.enzyme under the same conditions after 48 h.
8
20a
Biocatalytic reductions utilized S. carlsbergensis old
yellow enzyme that had been purified by affinity chroma-
rations were assigned by the direction of optical rotations
of the free â -amino acids as their hydrochloride salts
17
2
tography to avoid possible complications from Escherichia
coli alkene reductases. NADPH was supplied by a cofactor
regeneration system (glucose-6-phosphate/baker’s yeast glu-
cose-6-phosphate dehydrogenase).
(obtained by acid hydrolysis in 88-95% yields). Overall
2
yields of â -amino acids from â-nitroalkenes ranged from
57 to 73%.
Preliminary studies had revealed that olefin isomerization
was more rapid under alkaline conditions, and pH 6.93 was
selected to minimize this side-reaction while maintaining
acceptable enzyme efficiency. A 2-fold molar excess of
â-cyclodextrin (relative to the nitroacrylate) was also in-
cluded to enhance substrate solubility under aqueous condi-
tions. Unfortunately, the solubility of the â-cyclodextrin/3g
complex was still too low for efficient reduction, and no
reaction was observed in this case. Both the substrates (3a-f
and glucose-6-phosphate) and the two enzymes were added
portionwise to enhance the longevity of the processes, which
were carried out at room temperature.18 Reactions were
monitored by GC/MS, and complete substrate consumption
was observed after ca. 8 h in all cases except for (E)-3b.
NMR and GC analysis of the crude reaction products
verified that only the olefins had been reduced and the nitro
groups remained intact. No significant levels of side products
were observed, and yields ranged from 74 to 98%. Because
it was not possible to determine the optical purities of the
The results of our previous studies of alkyl-substituted
-cyclohexenone reductions by old yellow enzyme were
8
2
2
consistent with the net trans-addition of H mechanism
20
elucidated by Massey and Karplus (Scheme 2). Hydride
â-addition (from reduced FMN) occurs from the bottom face
18) Reaction mixtures contained final concentrations of NADP+ (20
(
µmoles, 15 mg), glucose-6-phosphate (1.27 mmoles, 429 mg), glucose-6-
phosphate dehydrogenase (500 µg), nitroacrylate (25mM), and purified OYE
(
20-40 µg) in 100 mM KPi, pH 6.95 in total volumes of 50 mL.
Conversions were carried out at room temperature. Reaction components
except for KPi buffer) were added in 10 equal portions every 45 minutes
(
and the mixtures were sampled for GC analysis periodically. After nearly
all of the substrate had been consumed, the reaction mixture was extracted
with Et2O (3 × (5 × reaction volume)). The combined organic extracts
were washed with brine (1 volume) and water (1 volume), dried with
MgSO4, and concentrated in vacuo.
(19) Crude reduction products (ca. 1.5 mmol) were hydrogenated at 500
psi in the presence of Raney nickel (200 mg) in EtOH (50 mL) at room
temperature. After 16 h, the resulting solution was filtered through Celite
and the solvent was evaporated. A portion of the residue (50 mg) was
dissolved in 6 M HCl and the solution was held at reflux overnight. The
solution was concentrated under reduced pressure to afford a yellow oil,
which was washed with EtOAc to remove any nonpolar impurities. Water
2
was removed by rotary evaporator to yield the â -amino acids as
(
14) Spectral data for 2f: 1H NMR (CDCl3) δ 4.85 (d, J ) 13.5 Hz,
hydrochloride salts. Spectral data matched those reported previously:
Nejman, M.; Sliwinska, A.; Zwierzak, A. Tetrahedron 2005, 61, 8536-
8541. Gangadhar, N.; Huber, V. J.; Lum, C.; Goodman, B. A. Org. Lett.
2000, 2, 3527-3529. Sammis, G. M.; Jacobsen, E. N. J. Am. Chem. Soc.
2003, 125, 4442-4443. Lee, H. S.; Park, J. D.; Kim, D. H. Bull. Korean
Chem. Soc. 2003, 24, 467-472.
1
H), 4.70 (d, J ) 13.3 Hz, 1H), 4.39 (m, 2H), 3.61 (s, 1H), 2.0 (m, 1H),
.36 (t, J ) 7.1 Hz, 3H), 1.0 (d, J ) 6.9 Hz, 3H), 0.92 (d, J ) 6.9 Hz, 3H);
C NMR (CDCl3) δ 173.1, 80.3, 77.7, 63.1, 34.3, 17.0, 16.4, 14.2.
1
13
(
(
15) Melton, J.; McMurry, J. E. J. Org. Chem. 1975, 40, 2138-2139.
16) Olefin configurations for 3c-g were assigned based on the chemical
shift values of the allylic protons, e.g., 2.11 ppm for (Z)-3c versus 2.60
ppm for the methyl ester of (E)-3c (ref 15a), as described by Denmark and
Marcin (J. Org. Chem. 1993, 58, 3850-3856.).
(20) (a) Kohli, R. M.; Massey, V. J. Biol. Chem. 1998, 273, 32763-
32770. (b) Xu, D.; Kohli, R. M.; Massey, V. Proc. Natl. Acad. Sci. U.S.A.
1999, 96, 3556-3561. (c) Brown, B. J.; Hyun, J. w.; Duvvuri, S.; Karplus,
P. A.; Massey, V. J. Biol. Chem. 2002, 277, 2138-2145. (d) Fox, K. M.;
Karplus, P. A. Structure 1994, 2, 1089-1105.
(17) Abramovitz, A. S.; Massey, V. J. Biol. Chem. 1976, 251, 5321-
5
326.
6132
Org. Lett., Vol. 8, No. 26, 2006

the reaction milieu. Conversions were not optimized during
these preliminary investigations, and it is likely that adjusting
the catalyst level and/or reaction pH might further improve
the biotransformations.
Scheme 2
Old yellow enzyme-mediated reductions of 3-alkyl-
substituted â-nitroacrylates 3a,b yielded essentially racemic
products. Given the highly stereoselective R-protonation
observed in 2-cyclohexenone reductions, this result was
surprising. Post-reduction racemization is unlikely since
Kawai has shown that similar R-alkyl-substituted nitro
compounds are configurationally stable at neutral pH23 and
our own studies in D
2
O buffer further support this notion.
Instead, we suspect that R-protonation likely occurs after
2
4
a
product release from the active site. The pK values of the
nitronate intermediates and the phenol of Tyr 196 make
proton transfer thermodynamically unfavorable. Indeed,
Massey has shown that mutation of Tyr 196 to Phe rendered
old yellow enzyme incapable of enone reductions while
having minimal impact on nitroalkene reductions.2 Stereo-
selective protonation of the nitronate intermediates will
0a
while R-protonation likely involves the phenol side-chain
of Tyr 196. Carbonyl activation is achieved via hydrogen
bonding by the side chains of His 191 and Asn 194 (not
shown). For the acyclic â-nitroacrylates investigated here,
analogous binding could occur in which one nitro oxygen
occupies the same location as the carbonyl oxygen and the
alkene is positioned similarly (Scheme 2). This substrate
binding orientation was verified by carrying out the enzy-
require an active site general acid with a better-matched pK
a
value near 7.0.
In conclusion, our results have uncovered a new applica-
tion for S. carlsbergensis old yellow enzyme in synthesizing
2
optically active â -amino acids. The synthetic route is concise
and utilizes inexpensive starting materials. The major dif-
ficulties lie in suppressing alkene isomerization prior to
reduction and ensuring active site protonation of the nitronate
intermediate. Efforts to accomplish both of these goals are
underway.
2
matic reductions of (E)-3b and (E)-3c in D O. All MS and
NMR data were consistent with deuterium incorporation only
on the nitro-bearing carbon in both cases. Isotope wash-in
following reduction was ruled out by observing that the
alkene reduction product of (E)-3a failed to incorporate
detectable levels of deuterium under the same conditions after
Acknowledgment. This work was supported financially
by the National Science Foundation (CHE-0615776) and by
a Ruegamer Foundation Fellowship to M.A.S. The assistance
of Dr. Bette Jo Brown of Professor Vincent Massey’s
laboratory (University of Michigan) in supplying the old
yellow enzyme overexpression plasmid and Mr. Daniel Serra
48 h.
Net trans-addition of H
2
to 3c-f would lead to the
observed (R)-products, and this model provides a useful
method for predicting the outcomes of additional nitroalkene
reductions by old yellow enzyme.21 High stereoselectivity
requires alkene configurational stability and this could be
the main reason that â-nitroacrylates with smaller alkyl
substituents afforded slightly lower enantiopurities. Within
(University of Florida) for his helpful discussions are also
gratefully acknowledged.
Supporting Information Available: Experimental pro-
7
h, initially pure (Z)-3d incubated in buffer alone afforded
cedures, spectral and chromatographic data used to determine
2
2
three byproducts in an aggregate yield of ca. 50%. These
observations underscore the need to reduce â-nitroacrylates
rapidly and minimize their exposure to the aqueous reaction
conditions. Fortunately, the reduction products are stable to
2
chemical and optical purities of the â -amino acid final
products, and spectral data for new compounds 3c-g. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(21) In a preliminary experiment, E-3d was reduced by old yellow
OL062612F
enzyme with largely (S)-stereoselectivity, as would be expected from the
model in Scheme 2. This suggests that it may be possible to prepare the
enantiomeric series of â-amino acids by using the alternate olefin isomers.
(23) Kawai, Y.; Inaba, Y.; Tokitoh, N. Tetrahedron: Asymmetry 2001,
(
22) The three byproducts were identified by GC/MS as (E)-3d, the
12, 309-318.
alternate acrylate alkene regioisomer of 3d, and the water addition product
from 3d (in racemic form).
(24) Meah, Y.; Massey, V. Proc. Natl. Acad. Sci. U.S.A. 2000, 97,
10733-10738.
Org. Lett., Vol. 8, No. 26, 2006
6133