Highly enantioselective reduction of β,β-disubstituted aromatic nitroalkenes catalyzed by Clostridium sporogenes

Article abstract of DOI:10.1021/jo800124v

(Chemical Equation Presented) This is the first report of the use of Clostridium sporogenes extracts for enantioselective reduction of C=C double bonds of β,β-disubstituted (1) and α,β-disubstituted nitroalkenes (3). Crude enzyme preparations reduced aryl derivatives 1a-e and 1h, in 35-86% yield with ≥97% ee. Reduction of (E)- and (Z)-isomers of 1c gave the same enantiomer of 2c (≥99% ee). In contrast, α,β- disubstituted nitroalkene 3a was a poor substrate, yielding (S)-4a in low yield (10-20%), and the ee (30-70% ee) depended on NADH concentration. An efficient synthesis of a library of nitroalkenes 1 is described.

Full text of DOI:10.1021/jo800124v

Jacobsen-type organocatalysts.9 The asymmetric reduction of
CdC double bonds is especially attractive, as up to two
stereogenic centers can be created. Unfortunately, the literature
procedures suffer either from poor environmental acceptability
due to the nature of the catalysts utilized or from the narrow
substrate range that limits the synthesis of desirable products.
As a result, there is considerable interest in developing alterna-
tive enzymatic procedures for the synthesis of optically active
nitro-compounds10 as starting blocks for chiral amines. Alkenes
activated by electron-withdrawing groups, e.g., ketones, alde-
hydes, carboxylic acids, and amides, are reduced stereoselec-
tively by various microorganisms, and the reductions are
catalyzed by flavoenzyme oxidoreductases [EC 1.3.1.x].10 The
family of enoate reductases has received particular attention,
and their potential value has provided the impetus to investigate
several enoate reductases from various aerobic11–13 and anaero-
bic14 microorganisms and plants.15,16
Highly Enantioselective Reduction of
ꢀ,ꢀ-Disubstituted Aromatic Nitroalkenes
Catalyzed by Clostridium sporogenes
Anna Fryszkowska,† Karl Fisher,‡ John M. Gardiner,† and
Gill M. Stephens‡,*
Manchester Interdisciplinary Biocentre, UniVersity of
Manchester, 131 Princess Street,
Manchester M1 7DN, United Kingdom
gill.stephens@manchester.ac.uk
ReceiVed January 25, 2008
Although the reduction of nitroalkenes by various microor-
ganisms has been studied by several groups,16–33 highly enan-
tioselective examples of this reaction are limited to the reduction
of ꢀ,ꢀ-disubstituted nitroalkenes by baker’s yeast17,18 and the
recently reported reductases from Lycopersicon esculentum16
and Saccharomyces carlsbergensis.19 Contrarily, the reduction
of R,ꢀ-disubstituted nitroalkenes has shown only poor enantio-
(9) Martin, N. J. A.; Ozores, L.; List, B. J. Am. Chem. Soc. 2007, 129, 8976–
8977.
(10) Stuermer, R.; Hauer, B.; Hall, M.; Faber, K. Curr. Opin. Chem. Biol.
2007, 11, 203–213.
This is the first report of the use of Clostridium sporogenes
extracts for enantioselective reduction of CdC double bonds
of ꢀ,ꢀ-disubstituted (1) and R,ꢀ-disubstituted nitroalkenes (3).
Crude enzyme preparations reduced aryl derivatives 1a-e and
1h, in 35-86% yield with g97% ee. Reduction of (E)- and
(Z)-isomers of 1c gave the same enantiomer of 2c (g99% ee).
In contrast, R,ꢀ-disubstituted nitroalkene 3a was a poor
substrate, yielding (S)-4a in low yield (10-20%), and the ee
(30-70% ee) depended on NADH concentration. An efficient
synthesis of a library of nitroalkenes 1 is described.
(11) Swiderska, M. A.; Stewart, J. D. J. Mol. Catal. B: Enzym. 2006, 42,
52–54.
(12) Chaparro-Riggers, J. F.; Rogers, T. A.; Vazquez-Figueroa, E.; Polizzi,
K. M.; Bommarius, A. S. AdV. Synth. Catal. 2007, 349, 1521–1531.
29.(13) Mu¨ller, A.; Hauer, B.; Rosche, B. Biotechnol. Bioeng. 2007, 98, 22–
(14) Rohdich, F.; Wiese, A.; Feicht, R.; Simon, H.; Bacher, A J. Biol. Chem.
2001, 276, 5779–5787.
(15) Stueckler, C.; Hall, M.; Ehammer, H.; Pointner, E.; Kroutil, W.;
Macheroux, P.; Faber, K. Org. Lett. 2007, 9, 5409–5411.
(16) Hall, M.; Stueckler, C.; Kroutil, W.; Macheroux, P.; Faber, K. Angew.
Chem., Int. Ed. 2007, 46, 3934–3937.
(17) Ohta, H.; Kobayashi, N.; Ozaki, K. J. Org. Chem. 1989, 54, 1802–
1804.
(18) Ohta, H.; Ozaki, K.; Tsuchihashi, G.-I. Chem. Lett. 1987, 1, 191–192.
(19) Swiderska, M. A.; Stewart, J. D. Org. Lett. 2006, 8, 6131–6133.
(20) Kawai, Y.; Inaba, Y.; Tokitoh, N. Tetrahedron: Asymmetry 2001, 12,
309–318.
Nitroalkanes are important intermediates in organic synthesis,
mainly due to their easy conversion into the corresponding
amines, aldehydes, carboxylic acids, or denitrated compounds.1,2
Chiral nitroalkanes can be obtained by asymmetric conjugate
addition to nitroalkenes3–6 or enantioselective conjugate reduc-
tion of nitroalkenes using chiral transition metal catalysts7,8 or
(21) Kawai, Y.; Inaba, Y.; Hayashi, M.; Tokitoh, N. Tetrahedron Lett. 2001,
42, 3367–3368.
(22) Sakai, K.; Nakazawa, A.; Kondo, K.; Ohta, H. Agric. Biol. Chem. 1985,
49, 2331–2335.
(23) Korbekandi, H.; Mather, P.; Gardiner, J.; Stephens, G. Enzyme Microb.
Technol. 2007, 42, 308–314.
(24) Takeshita, M.; Yoshida, S.; Kohno, Y. Heterocycles 1994, 37, 553–
562.
† Manchester Interdisciplinary Biocentre and the School of Chemistry.
‡ Manchester Interdisciplinary Biocentre and the School of Chemical Engi-
neering and Analytical Science.
(1) Ono, N. The Nitro Group in Organic Synthesis; Wiley-VCH: New York,
2001.
(25) Shaw, P. D. Biochemistry 1967, 6, 2253–2260.
(26) Mori, A.; Ishiyama, I.; Akita, H.; Suzuki, K.; Mitsuoka, T.; Oishi, T.
Chem. Pharm. Bull. 1990, 38, 3449–3451.
(27) Bak, R. R.; McAnda, A. F.; Smallridge, A. J.; Trewhella, M. A. Aust.
J. Chem. 1996, 49, 1257–1260.
(2) Ballini, R.; Petrini, M. Tetrahedron 2004, 60, 1017–1047.
(3) Mampreian, D. M.; Hoveyda, A. H. Org. Lett. 2004, 6, 2829–2832.
(4) Cote, A.; Lindsay, V. N. G.; Charette, A. B. Org. Lett. 2007, 9, 85–87.
(5) Berner, O. M.; Tedeschi, L.; Enders, D. Eur. J. Org. Chem. 2002, 1877–
1894.
(28) McAnda, A. F.; Roberts, K. D.; Smallridge, A. J.; Ten, A.; Trewhella,
M. A. J. Chem. Soc., Perkin Trans. 1 1998, 501, 504.
(29) Meah, Y.; Massey, V. Proc. Natl. Acad. Sci. 2000, 97, 10733–10738.
(30) Kohli, R. M.; Massey, V. J. Biol. Chem. 1998, 273, 32763–32770.
(31) Messiha, H. L.; Munro, A. W.; Bruce, N. C.; Barsukov, I.; Scrutton,
N. S. J. Biol. Chem. 2005, 280, 10695–10709.
(6) Wu, J.; Mampreian, D. M.; Hoveyda, A. H. J. Am. Chem. Soc. 2005,
127, 4584–4585.
(32) Kataoka, M.; Kotaka, A.; Hasegawa, A.; Wada, M.; Yoshizumi, A.;
Nakamori, S.; Shimizu, S. Biosci. Biotechnol. Biochem. 2002, 66, 2651–2657.
(33) Kataoka, M.; Kotaka, A.; Thiwthong, R.; Wada, M.; Nakamori, S.;
Shimizu, S. J. Biotechnol. 2004, 114, 1–9.
(7) Czekelius, C.; Carreira, E. M. Angew. Chem., Int. Ed. 2003, 42, 4793–
4795.
(8) Czekelius, C.; Carreira, E. M. Org. Lett. 2005, 6, 4575–4577.
10.1021/jo800124v CCC: $40.75
Published on Web 05/02/2008
2008 American Chemical Society
J. Org. Chem. 2008, 73, 4295–4298 4295

SCHEME 1. Asymmetric Reduction of ꢀ-Alkyl-ꢀ-aryl- (1)
and r-Alkyl-ꢀ-aryl-nitroalkenes (3) by C. sporogenes
derivatives, with both electron-withdrawing and electron-
donating substituents at the phenyl ring.
Under the reaction conditions employed, the (Z)-isomer was
formed only as a minor product (<5% for R1 ) Ar by NMR
and ∼30% for R1 ) pentyl). However, upon exposure to UV
light, (E)-1c could be transformed to the respective stable (Z)-
isomer, as observed previously.42,43 Racemic nitroalkanes 2 and
4, required as analytical standards, were obtained in a NaBH4-
catalyzed reduction, as described previously.44
Biotransformations. Because substrate delivery and product
isolation from biotransformations in aqueous media can be
troublesome, we used an anaerobic biphasic solvent system
consisting of iso-octane/50 mM phosphate buffer (pH 7.0) at
20:80 v/v ratio. Initially, we employed whole cells of C.
sporogenes DSM 795 with H2 as the electron donor. The
reduction of 1a was enantioselective, yielding (R)-2a in >99%
ee. In contrast, the enantioselectivity for reduction of nitroalkene
3a was poor (e30% ee).
SCHEME 2. Synthesis of (E)- and (Z)-ꢀ,ꢀ-Disubstituted
Nitroalkenes 1a-i
selectivity (e50% ee).20–23 The enoate reductases from anaero-
bic Clostridium species have been extensively investigated by
Simon et al.14,34–36 These enzymes, including the enoate
reductase from C. sporogenes,37–39 proved to be versatile,
stereoselective biocatalysts for the reduction of R,ꢀ-unsaturated
carboxylic acids, ketones, and aldehydes.34 However, these
microorganisms have not been used for bioreduction of ni-
troalkenes, and therefore we wished to explore their potential
for reduction of this class of compounds.
Here, we report the substrate specificity of C. sporogenes
toward aromatic ꢀ-alkyl-ꢀ-aryl- (1) and R-alkyl-ꢀ-aryl-nitroalk-
enes (3), as depicted in Scheme 1. Using crude enzyme prepara-
tions, we were able to obtain stereoselective reduction of
substrates 1a-e in excellent yield.
Synthesis of Substrates. (E)-1-Aryl-1-nitroalkenes of struc-
ture 3 can be synthesized conveniently by base-catalyzed
condensation of nitroalkanes with benzaldehyde under sonication
conditions in very good yields (>80%).40 However, one of the
barriers to the study of the biotransformations of nitroalkenes
is that the standard syntheses of substrates of structure 1 is
relatively complicated or has structural limitations. They can
be prepared from R-methylstyrenes via a two-step method
involving addition of N2O4 and subsequent elimination7 or in a
similar synthetic approach, with a nitration reaction with HNO3
as a first step.9,16,17 The conditions for the latter reaction limit
the synthesis to the derivatives with electron-withdrawing
substituents at the phenyl ring. Therefore, we began by
developing a more convenient, single-step method for the
synthesis of ꢀ,ꢀ-disubstituted nitroalkenes 1 from their respec-
tive R-methylstyrenes (5) via a CAN-catalyzed nitration reac-
tion,41 as shown in Scheme 2. This procedure allowed the
straightforward preparation of a library of nitropropenes (E)-
1a-i in 14-52% yield, giving access to alkyl and aryl
Nitroalkene reductase activity was also retained with similar
enantioselectivity after anaerobic preparation of crude cell-free
extracts. The reaction time was shorter by a factor of 2,
presumably because there was no requirement for the transport
of substrates and products across the cell membrane. Like the
whole cells, the enzyme preparation was sensitive to oxygen.
Nitroalkene reduction could be driven using H2 in the absence
of added cofactors, most probably because an H2-dependent
system for recycling endogenous cofactors was present in the
crude extract. However, the product yield was doubled and
reaction time halved when 20%mol amount of exogenous NADH
was also added, and this procedure was therefore used throughout.
Further optimization of the reaction conditions was attempted
by comparing various other nonpolar solvents (toluene, benzene,
chloroform, dichloromethane, and n-alkanes from pentane to
tetradecane and iso-octane) in 20:80 v/v phase ratio with
phosphate buffer (see Supporting Information, Figure 2). No
apparent loss of enzymatic activity with respect to the reduction
of approximately 1.7 mM 1-nitro-2-phenyl-propene (1a) by 40
mg of crude extract was observed with any of the alkanes tested
(data not shown). With all other solvents, lower yields of product
2a were obtained, probably as a result of enzyme deactivation.
We selected iso-octane as the solvent of choice because it
possesses favorable physical properties, namely, a low boiling
point and good substrate solubility. Increasing the iso-octane
phase ratio to 60% v/v did not result in a decrease in product
yield.
The best yields of (R)-2a were obtained when the reactions
were performed in the pH range 6.5-8.0, and the enantiose-
lectivity was unaffected by pH (see Supporting Information,
Figure 3). In contrast, the enantioselectivity for reduction of
nitroalkene 3a was poor (e30% ee, depending on pH). The
optimum pH for the formation of nitroalkane 4a was signifi-
cantly lower at pH 5.5, but the enantioselectivity for the
reduction of nitroalkene 3a was optimal at pH 6.5-7.0 (30%
ee). Therefore, subsequent reductions were conducted at pH 7.0.
Having optimized the assay conditions at small scale, prepara-
tive-scale reactions with substrates 1a-i and 3a were conducted
anaerobically using C. sporogenes crude extract preparations
(Table 1).
(34) Steinbacher S.; Stumpf, M.; Weinkauf, S.; Rohdich, F.; Bacher, A.;
Simon, H. In FlaVins and FlaVoproteins; Chapman, S., Percham, R., Scrutton,
N. S., Eds.; Rudolf Weber: Berlin, 2002, pp 941-949 and references therein.
(35) Rambeck, B.; Simon, H. Angew. Chem., Int. Ed. 1974, 13, 609.
(36) Simon, H.; Rambeck, B.; Hashimoto, H.; Gu¨nther, H.; Nohynek, G.;
Neumann, H. Angew. Chem., Int. Ed. 1974, 13, 608–609.
(37) Buhler, M.; Giesel, H.; Tischer, W.; Simon, H. FEBS Lett. 1980, 109,
244–246.
(38) Giesel, H.; Machacek, G.; Bayerl, J.; Simon, H. FEBS Lett. 1981, 123,
107–110.
(39) Bader, J.; Simon, H. FEMS Microbiol. Lett. 1983, 20, 171–175.
(40) Osorio-Olivares, M.; Rezende, M. C.; Sepulveda-Boza, S.; Cassels,
B. K.; Fierro, A. Bioorg. Med. Chem. 2004, 12, 4055–4066.
(41) Hwu, R. J.; Chen, K.-L.; Anathan, S. J. Chem. Soc., Chem. Commun.
1994, 1425–1426.
(42) Bluhm, A. L.; Weinstein, J. J. Am. Chem. Soc. 1965, 87, 5511–5512.
(43) Sousa, J. A.; Weinstein, J.; Bluhm, A. L. J. Org. Chem. 1969, 34, 3320–
3323.
(44) Kabalka, G. W.; Varma, R. S. Org. Prep. Proced. Int. 1987, 19, 283–
328.
4296 J. Org. Chem. Vol. 73, No. 11, 2008

TABLE 1. Reduction of 2-Phenyl-1-nitropropenes 1a-h and
1-Phenyl-2-nitropropene 3a by C. sporogenes Crude ExtractI
followed the trend H > F > Cl > Br. p-Methoxy-substituted
nitroalkene 1e was completely converted after 5 days, similar
to the reduction of 1a; however, the isolated yield of the product
was much lower due to formation of byproduct arising from
possible alternative metabolic pathways. These compounds were
tentatively identified as respective propionaldehyde derivatives
(Nef reaction product) by GC-MS but were not fully character-
ized. When the methoxy substituent was in the meta position
(1g), the reaction time increased and the optical purity of the
product 2g decreased (89% ee), while the reduction of the ortho
derivative (1f) led to an even longer reaction time and a total
loss of enantioselectivity, resulting in the formation of racemic
product. Clostridium sporogenes crude extracts also catalyzed
the highly enantioselective reduction of 2-(4′-methyl-thienyl)-
1-nitropropene 1h to its corresponding alkane (S)-2h in excellent
optical purity (98% ee), which shows that the substrate range
also extends to heterocyclic aromatic nitroalkenes. Interestingly,
aliphatic nitroalkene derivatives were very poor substrates, since
the reduction of (E)- and (Z)-2-methyl-1-nitro-octene 1i gave
only trace amounts of product 2i, as monitored by GC-MS.
Furthermore, none of the other aliphatic substrates tested,
namely, 1-nitrocyclohexene and 1-cyclohexyl-2-nitro-propene,
were reduced, as determined by spectrophotometric assays.
The reactions described above were all done using the (E)-
nitroalkenes. The outcome with the corresponding (Z)-isomers
was thus of interest, and we therefore tested the reduction of
(Z)-1c on analytical scale with crude extract, using the same
general conditions. This resulted in the formation of the same
enantiomer as obtained from reduction of (E)-1c, namely, (R)-
2c, with the same high optical purity (g99%), allowing enantio-
convergent reduction. The time required for full conversion of
(Z)-1c was approximately 1.5 times longer than for the (E)-
isomer with crude extract, but preliminary experiments with
partially purifed enzyme showed that the initial rates of reduction
were the same for (Z)-1c and (E)-1c. Similar behavior has been
observed from studies with yeast as a biocatalyst17,20 but is in
striking contrast to the reduction of enoates and enals, where
different enantiomers are obtained using the (E)- and (Z)-isomers
of substrates.10,11,13,15,19 Furthermore, there is still a possibility
that only one isomer is stereoselectively reduced since pure
isomers of 1c equilibrated very slowly to form a mixture of
isomers E/Z: substrate purity of 98% for (E)-1c and 97% for
(Z)-1c, after 5 days in the absence of enzyme under the same
reaction conditions. Thus, the selective reduction of only one
of the isomers would perturb the equilibrium for isomerization,
ensuring a continuing supply of the preferred isomer as a
substrate for the stereoselective reduction. At present, it is not
possible to determine which of these hypotheses explains the
apparent lack of discrimination between the geometrical isomers,
and therefore we are attempting to obtain enzyme of sufficient
purity for crystallographic studies of substrate binding. Never-
theless, the ability to reduce a mixture of geometrical isomers
to the same stereoisomer with no loss of enantioselectivity is
of important practical value because the preparation of a single
isomer is then not required. While this is extremely advanta-
geous, it does mean that it was not possible to control the
absolute configuration of the product (R or S) by the stereo-
chemistry of the alkene.
a By GC-MS and referenced to substrate recovered. b Isolated. c By
HPLC using Chiralcel OD column. d In CHCl3, 30 °C. e Reaction carried
out at analytical scale. f Additional 150 mg of NADH was added after 5
days, and the reaction was continued until it reached full conversion.
g On Chiralcel OJ. h +34.0 (c 0.60, CHCl3) ) +38.0 (c 0.44, MeOH).26
I Conditions: substrate 1 or 3 (150 mg); iso-octane (400 mL); phosphate
buffer pH 7.0 (800 mL, 50 mM); crude extract (2.2 g total protein
equivalent to approximately 3.4 g whole cell dry weight), NADH (150
mg).
The reactions were allowed to proceed until full substrate
conversion was obtained, as determined by GC-MS, or were
stopped after a maximum of 7 days. The optical purity of
products was analyzed by chiral HPLC, and the absolute
configuration was assigned by comparing the optical rotation
sign or HPLC data with literature.3,6,17,26
The bioreduction of aromatic nitroalkenes (E)-1a-e was
highly enantioselective, giving the corresponding propane
derivatives (R)-2 in high optical purity (g97% ee). Product
yields ranged from 35-59%, but the substrates were consumed
with greater efficiency than this. We assume that this was due
to side reactions catalyzed by other enzymes in the cell-free
extract, since the yields increased when the enzyme was partially
purified (results not shown).
In contrast to ꢀ,ꢀ-disubstituted nitroalkenes 1, the reduction
of the R,ꢀ-disubstituted nitroalkene, 1-nitro-2-phenyl-nitropro-
pene 3a, was 5-10 times slower and gave product (S)-4a in
low yield (20%). At preparative scale, the ee was higher (50%
The presence of a substituent in the para position of the
phenyl ring ((E)-1b-e) resulted in an increased reaction time
for total substrate conversion. The efficiency of reduction of
unsubstituted and p-halogen-substituted nitroalkenes 1a-d
J. Org. Chem. Vol. 73, No. 11, 2008 4297

400 mL of anaerobic iso-octane. The vessel was sealed with a suba-
seal stopper and flushed with H2 for 2 min prior to continual stirring
in an anaerobic cabinet. Reactions were monitored daily by
GC-MS until they reached full conversion or were stopped after
a maximum of 7 days. If reactions were observed to slow con-
siderably, e.g., during the reduction of 3a, then an additional 150
mg of NADH was added. Upon termination of the reaction, the
phases were separated and the aqueous layer was extracted with
CHCl3. Combined organic layers were dried (MgSO4) and evapo-
rated to dryness. Column chromatography (hexane f hexane/Et2O,
95:5) gave pure nitroalkane 2 or 4.
(R)-1-Nitro-2-phenyl-propane ((R)-2a). 59% yield as transpar-
ent oil; purified by chromatography (hexane/Et2O, 95:5); 1H NMR
(CDCl3, 400 MHz) δ 1.39 (d, J ) 6.9 Hz, 3H), 3.64 (q, J ) 7.4
Hz, 1H), 4.45-4.59 (m, 2H), 7.23-7.46 (m, 5H); 13C NMR (CDCl3,
100 MHz) δ 18.6, 38.6, 81.9, 126.8, 127.5, 128.9, 140.8; MS (EI)
m/z 165 (1%, M+), 91 (100%); IR (neat) 768, 1382, 1542 cm-1; ee
> 99% (Chiralcel OD, hexane/i-PrOH, 9:1, tR ) 9.6 min., tS )
10.3 min, lit. chromatography data available for (S)-enantiomer8);
[R]D ) +52.0 (c 0.56, CHCl3, 30 °C), lit. +44.3 (c 3.4, CHCl3, 27
°C, 98% ee17) or [R]D ) -48.8 (c 0.97, CHCl3, 20 °C, (S)-
enantiomer, 92% ee3). Anal. Calcd for C9H11NO2: C, 65.44; H,
6.71; N, 8.48. Found: C, 65.48; H, 6.93; N, 8.28. All spectral data
are consistent with the literature values.3,8
General Procedure for Synthesis of ꢀ,ꢀ-Disubstituted Ni-
troalkenes (E)-1a-i in a CAN-Catalyzed Reaction. To a suspen-
sion of R-methylstyrene 5 (1 mmol), NaNO2 (10 mmol), and CAN
(1 mmol) in CHCl3 (10 mL) was added acetic acid (12 mmol)
dropwise. The mixture was sonicated in a sealed flask connected
to a bubbler until the reaction reached completion, as judged by
TLC (5-60 min). CHCl3 (20 mL) was added and solution was
washed with saturated NaHCO3 and water and then dried (MgSO4).
The solvent was evaporated and the product (E)-1 was purified by
flash chromatography (hexane/Et2O, 9:1).
(E)-1-Nitro-2-phenyl-propene (1a). 42% yield as yellow oil;
purified by chromatography (hexane/Et2O, 95:5); 1H NMR (CDCl3,
400 MHz) δ 2.66 (s, 3H), 7.32 (s, 1H), 7.46 (s, 5H); 13C NMR
(CDCl3, 100 MHz) δ 18.6, 126.8, 129.0, 130.3, 136.2, 138.2, 150.0;
MS (EI) m/z 163 (1%, M+), 115 (100%); IR (neat) 1320, 1518
cm-1. Anal. Calcd for C9H9NO2: C, 66.25; H, 5.56; N, 8.58. Found:
C, 66.35; H, 5.64; N, 8.54. All spectral data are consistent with
the literature values.6,8,17
ee) after 5 days than in the small-scale experiments. However,
the conversion was incomplete, so we added additional NADH
(150 mg), and full conversion was obtained after 11 days. This
also improved the optical purity of the product (70% ee),
demonstrating that that enantioselectivity is influenced by the
NADH concentration. Interestingly, during the reduction of 3a,
phenylacetone was also formed in approximately 30% yield.
The difference in enantioselectivity of the bioreduction of
nitroalkenes, high for ꢀ,ꢀ-disubstituted16,17 and low for R,ꢀ-
disubstituted nitroalkenes,20–23 has been observed for many
microorganisms. It has been suggested that the low optical purity
of products possessing substituents at the R-carbon atom (e.g.,
3a) is due to product racemization under the conditions in the
reaction.10,17 However, Kawai et al. showed that the rate of
spontaneous epimerization at the R-carbon of nitroalkanes was
actually very low.20 We found the deuterium exchange of the
R-proton of (()-4a was negligible (<5%) if incubated in D2O/
iso-octane (40:60, v/v) at 30 °C for 7 days. This suggests that
the stereochemical outcome is due to the mechanism of
enzymatic reduction rather than racemization, and further studies
with purified protein are needed to explain this phenomenon.
In summary, the nitroalkene reductase from C. sporogenes
is a versatile biocatalyst for stereoselective reduction of aromatic
and heteroaromatic nitroalkenes, although it does not reduce
alkyl derivatives. Thus, reduction of 2-aryl-1-nitro-propenes
1a-e and 1h was efficient and highly enantioselective, providing
single enantiomers of the respective products (R)-2a-e and (S)-
2h (g97% ee) in 35-86% yield. The absolute configuration of
product was independent of which geometric isomer of substrate
was used. Contrary to the substrates of structure 1, the
bioreduction of 1-nitro-2-phenyl-propene 3a by crude extract
resulted in (S)-4a in low yield, and the optical purity depended
on NADH concentration. The low enantiopurity could not be
explained by racemization.
This is the first report of the use of enzymes from Clostridia
for the highly enantioselective reduction of nitroalkenes with
good yields, thus demonstrating that their substrate range extends
beyond enoates and enals.35,36,38 The mechanism and scope of
the reduction of nitroalkenes and other unsaturated compounds
by C. sporogenes are currently under investigation in our
laboratory.
Acknowledgment. Financial support from the Biotechnology
and Biological Sciences Research Council (grant BBD0028261)
is gratefully acknowledged. We also thank the EPSRC Mass
Spectrometry Service at Swansea for mass spectral analyses.
Experimental Section
General Procedure for Preparative-Scale Bioreduction of
Nitroalkenes 1 or 3. Bioreductions were performed in a 2-L round-
bottom flasks containing 800 mL of anaerobic 50 mM phosphate
buffer pH 7.0 with approximately 2 g (total protein) of crude extract
of C. sporogenes (DSM 795 strain; see Supporting Information),
150 mg of NADH, and 150 mg of substrate 1 or 3a dissolved in
Supporting Information Available: Full characterization
data for the prepared compounds, anaerobic methodology, cell
growth and enzyme preparation. This material is available free
of charge via the Internet at http://pubs.acs.org.
JO800124V
4298 J. Org. Chem. Vol. 73, No. 11, 2008