Asymmetric bioreduction of C=C bonds using enoate reductases OPR1, OPR3 and YqjM: Enzyme-based stereocontrol

Article abstract of DOI:10.1002/adsc.200700458

Three cloned enoate reductases from the "old yellow enzyme" family of flavoproteins were investigated in the asymmetric bioreduction of activated alkenes. 12-Oxophytodienoate reductase isoenzymes OPR1 and OPR3 from Lycopersicon esculentum (tomato), and YqjM from Bacillus subtilis displayed a remarkably broad substrate spectrum by reducing α,β-unsaturated aldehydes, ketones, maleimides and nitroalkenes. The reaction proceeded with absolute chemoselectivity-only the conjugated C=C bond was reduced, while isolated olefins and carbonyl groups remained intact-with excellent stereoselectivities (ees up to >99%). Upon reduction of a nitroalkene, the stereochemical outcome could be determined via choice of the appropriate enzyme (OPR1 versus OPR3 or YqjM), which furnished the corresponding enantiomeric nitroalkanes in excellent ee. Molecular modelling suggests that this "enzyme-based stereocontrol" is caused by subtle differences within the active site geometries.

Full text of DOI:10.1002/adsc.200700458

FULL PAPERS
DOI: 10.1002/adsc.200700458
Asymmetric Bioreduction of C=C Bonds using Enoate Reduc-
tases OPR1, OPR3 and YqjM: Enzyme-Based Stereocontrol
MØlanie Hall,a Clemens Stueckler,a Heidemarie Ehammer,b Eva Pointner,b
Gustav Oberdorfer,c Karl Gruber,c Bernard Hauer,d Rainer Stuermer,d
Wolfgang Kroutil,a Peter Macheroux,b and Kurt Fabera,*
a
Department of Chemistry, Organic and Bioorganic Chemistry, University of Graz, Heinrichstrasse 28, 8010 Graz, Austria
Fax : (+43)-316-380-9840; e-mail: kurt.faber@uni-graz.at
Institute of Biochemistry, Graz University of Technology, Petersgasse 12, 8010 Graz, Austria
Department of Chemistry, Structural Biology, University of Graz, Humboldtstrasse 50, 8010 Graz, Austria
BASF AG, GVF/E-B9, 67056 Ludwigshafen, Germany
b
c
d
Received: September 19, 2007; Published online: February 6, 2008
Abstract: Three cloned enoate reductases from the of a nitroalkene, the stereochemical outcome could
“old yellow enzyme” family of flavoproteins were in- be determined via choice of the appropriate enzyme
vestigated in the asymmetric bioreduction of activat- (OPR1 versus OPR3 or YqjM), which furnished the
ed alkenes. 12-Oxophytodienoate reductase isoen- corresponding enantiomeric nitroalkanes in excellent
zymes OPR1 and OPR3 from Lycopersicon esculen- ee. Molecular modelling suggests that this “enzyme-
tum (tomato), and YqjM from Bacillus subtilis dis- based stereocontrol” is caused by subtle differences
played a remarkably broad substrate spectrum by re- within the active site geometries.
ducing
a,b-unsaturated
aldehydes,
ketones,
maleimides and nitroalkenes. The reaction proceed-
ed with absolute chemoselectivity – only the conju- Keywords: asymmetric bioreduction; enoate reduc-
gated C=C bond was reduced, while isolated olefins tase; nitroalkenes; old yellow enzyme; stereocomple-
and carbonyl groups remained intact – with excellent mentary process; a,b-unsaturated carbonyl com-
stereoselectivities (ees up to >99%). Upon reduction pounds
Introduction
action”. In the subsequent “oxidative half reaction”
the hydride is transferred onto Cb of the substrate,
The asymmetric reduction of C=C bonds goes hand in while an essential Tyr residue (conserved along the
hand with the creation of (up to) two chiral carbon OYE family) adds a proton onto Ca from the oppo-
centres and is thus one of the most widely employed site side. As a consequence of this mechanism, the ad-
strategies for the production of chiral materials.
Whereas cis-hydrogenation using transition metal-
based homogeneous catalysts has been developed to
an impressive standard,[1] stereocomplementary asym-
metric trans-hydrogenation, which resembles the con-
jugate (Michael-type) addition of a hydride onto a
C=C bond bearing an electron-withdrawing substitu-
ent is still at the stage of development.[2]
The biocatalytic counterpart for the stereoselective
reduction of activated alkenes is catalysed by enoate
reductases [EC 1.3.1.X],[3–5] members of the “old
yellow enzyme” family.[6] These enzymes are ubiqui-
tous in nature and their catalytic mechanism has been
investigated in great detail (Scheme 1).[3,7] The reac-
tion proceeds via a ping-pong bi-bi mechanism. First,
the flavin cofactor is reduced at the expense of a nico-
Scheme 1. Asymmetric bioreduction of activated alkenes
bearing an electron-withdrawing group (EWG) by enoate
tinamide cofactor NAD(P)H in the “reductive half re- reductases.
Adv. Synth. Catal. 2008, 350, 411 – 418
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
411

MØlanie Hall et al.
FULL PAPERS
dition of [H2] proceeds in a trans-fashion with abso- whole-cell bioreduction.[10,18] However, when NADH
lute stereospecificity.[8] Overall, the reaction resem- was recycled using the FDH/formate system, the alde-
bles an asymmetric conjugate (Michael-type) addition hyde moiety was rapidly reduced to furnish (S)-citro-
of a chiral hydride onto an enone and thus, non-acti- nellol (3,7-dimethyloct-6-ene-1-ol, ee >95%) and
vated C=C bonds are completely unreactive.[9,10,18] In nerol (3,7-dimethylocta-2,6-dien-1-ol) as the major
order to avoid the requirement for external products (entry 2). This drawback – presumably
NAD(P)H recycling, the vast majority of asymmetric caused by prim-ADH activities present in commercial
C=C-bioreductions were performed using whole fer- FDH preparations [crude NAD+-specific FDH prepa-
menting cells, usually bakerꢁs yeast.[11] Although the rations from Jülich Fine Chemicals (No. 09.11 and
stereoselectivities thus obtained were often excellent, 24.11)] – was successfully circumvented by switching
the chemoselectivities with respect to C=C versus to the GDH/glucose system (entry 3). Alternatively,
C=O reduction were severely impeded by the action NADPH could be recycled via G6PDH/glucose 6-
of competing carbonyl reductases, which not only de- phosphate (entry 4). The short-chain (E)-2-methypen-
pleted the substrate (by forming allylic alcohols as tenal (2a) was selected as an a-substituted a,b-unsatu-
dead-end products) but also the product (furnishing rated aldehyde analogue; this substrate was reduced
saturated alcohols due to “over-reduction”). Since with poor stereoselectivity by the three enzymes, al-
enoate reductases and alcohol dehydrogenases though with opposite stereoselectivities (entry 5):
depend on the same cofactor, redox decoupling of whereas OPR1 furnished (R)-2-methylpentanal (2b)
both enzyme activities is hardly possible. Just recently, in moderate ee (47%), OPR3 produced the opposite
this major drawback was circumvented by the use of enantiomer with poor stereoselectivity (19% ee).
cloned enoate reductases from various yeasts.[12,13]
With YqjM, the saturated aldehyde (R)-2b was ob-
The electron-withdrawing substituent attached to tained almost racemic (10% ee). The apparent large
the C=C bond (EWG, Scheme 1) usually consists of a selectivity difference between the b- (1a) and a-sub-
carbonyl group, such as an aldehyde, ketone, carbox- stituted enals (2a) might be due to racemisation of
ylic acid, ester, anhydride, lactone or imide.[5] Due to the latter catalysed by enoate reductases.[19] A related
the electronic similarity of the nitro and the carboxy switch of stereopreference for citral (1a) versus 2-
groups, it is not surprising that also nitroolefins are methylpentenal (2a) was recently observed with sev-
sufficiently activated. The remarkable catalytic pro- eral OYEs from yeast.[13] Next, a,b-unsaturated cyclic
miscuity[14] of enoate reductases was recently demon- ketones 3a–6a were investigated with respect to the
strated by the trans-bioreduction of ynones.[15] In influence of the ring size and the position of the sub-
search for (oxygen-)stable enoate reductases possess- stituent on activity and selectivity. 2-Methylcyclopen-
ing a broad substrate tolerance, we recently investi- tenone (3a) was reduced to (S)-3b in reasonable to
gated 12-oxophytodienoate reductase isoenzymes excellent stereoselectivity by all enzymes in up to
OPR1 and OPR3 from Lycopersicon esculentum 94% ee (entries 6–9). At present, the remarkable var-
(tomato).[16] Herein we report on the substrate toler- iation of stereoselectivities depending on the type of
ance of OPR1, OPR3 and the “old yellow enzyme” cofactor (or recycling system) used cannot be ex-
homologue YqjM from Bacillus subtilis, which is in- plained on a molecular basis, but similar effects have
volved in the oxidative stress response.[17] In order to been observed with related enzymes and are presuma-
obtain reasonably broad information on the substrate bly due to “allosteric effects”.[12,20] Much to our sur-
selectivity pattern, substrates were selected for the (i) prise, the ring size of the substrate had a tremendous
nature of the activating group, (ii) chain-length or effect on the stereochemical outcome of the bioreduc-
ring size and (iii) relative position of substituents at tion. In contrast to 3a/(S)-3b, a switch of stereoprefer-
the C=C bond, and (iv) the influence of the cofactor ence was observed when reducing 2-methylcyclohexe-
type (NADH vs. NADPH) on activity or selectivity.
none (4a), where all enzymes furnished (R)-4b with
up to 93% ee (entries 10–13). This effect was most
pronounced with YqjM using the NADP+/G6PDH re-
cycling system, which furnished (S)-3b and (R)-4b in
94% and 93% ee, respectively. In order to verify this
Results and Discussion
Due to the presence of multiple C=C and C=O unexpected steric effect of an additional CH2 moiety
bonds, the bioreduction of citral (1a) in a chemo-, on the stereopreference, the absolute configuration of
regio-, and stereoselective fashion represents a chal- 3b and 4b was double-checked using various methods
lenging task. OPR1, OPR3 and YqjM reduced the (see Experimental Section) and verified by molecular
a,b-unsaturated C=C-moiety in a highly selective modelling (data from docking studies not shown).
fashion at the expense of NADH or NADPH to yield
The position of the methyl substituent at the C=C
(S)-citronellal (1b) in >95% ee. (Table 1, entry 1). bond turned out to be crucial: Whereas both a-substi-
Both the non-activated C=C bond and the aldehyde tuted cycloalkenones (3a, 4a) were quickly converted
moiety remained untouched, which is in contrast to with good stereoselectivities (entries 6–13), the corre-
412
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 411 – 418

Asymmetric Bioreduction of C=C Bonds using Enoate Reductases
FULL PAPERS
Adv. Synth. Catal. 2008, 350, 411 – 418
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
413

MØlanie Hall et al.
FULL PAPERS
Scheme 2. Enzyme-based stereocontrol.
sponding b-analogues (5a, 6a) turned out to be “diffi- 10b with excellent stereoselectivity (ee up to 98%),
cult” substrates (entries 14–21). They were not (or OPR3 and YqjM delivered the opposite enantiomer
only marginally) transformed by OPR1 or YqjM, also in up to 93% ee. This reversal of stereoselectivity is
OPR3 showed low activities, albeit with excellent ste- remarkable bearing in mind that the active site archi-
reoselectivity (ee >99%).
tectures of isozymes OPR1 and OPR3 are highly con-
The reduction of ketoisophorone (7a) to levodione served (55% sequence identity, 70% homology). Al-
(7b) was of special interest, since the latter is an im- though similar “enzyme-based stereocontrol” has
portant building block for the synthesis of zeaxanthin been reported before,[25] this complete reversal of ste-
on an industrial scale (entries 22–25).[21] With all three reoselectivity is remarkable in its magnitude. The dif-
enzymes, 7a was chemoselectively reduced to (R)-7b ference in the DDG# values[26] accounting for the for-
in up to >99% ee, again, the type of cofactor or recy- mation of opposite enantiomers in 99 and 93% ee, re-
cling system seems to play an important role on the spectively, amounts to 5.1 kcalMÀ1, which is derived
stereoselectivity.
from the “destruction” of the stereoselectivity for
a-Methylmaleimides 8a and 9a turned out to be ex- (R)-10b (99% ee, 3.14 kcalMÀ1) and the “generation”
cellent substrates, which were readily reduced to the of stereoselectivity in the opposite sense to (S)-10b
saturated counterparts (8b, 9b) by all three enzymes (97% ee, 1.96 kcalMÀ1). Furthermore, YqjM surpris-
(entries 26–33). However, our expectation that the ingly displayed an enhanced stereoselectivity (along
stereoselectivity could be altered (or possibly even with higher conversion) using NADH as cofactor (ee
switched) via substrate engineering[22] through varia- 92%, entry 34), while only 70% ee were reached with
À
tion of the size of the N substituent, i.e., N H versus NADPH (entry 35).
À
N Ph, failed: both products showed the same (R)-
configuration and were formed in 92 to >99% ee.
In order to rationalise this switch of stereoprefer-
ence in the reduction of 1-nitro-2-phenylpropene
Nitroalkenes are readily reduced by enoate reduc- (10a) by isoenzymes OPR1 and OPR3, modelling
tases forming chiral nitroalkanes, which can be easily studies were pereformed based on the crystal struc-
converted to useful intermediates.[23] Due to the acidi- tures of both proteins.[27,29] Despite the 55% sequence
ty of the Ca-proton, a-substituted nitroalkanes race- identity and 70% similarity, they show some subtle,
mise spontaneously,[24] thus 1-nitro-2-phenylpropene but crucial differences in active site architecture. In
(10a) was chosen as model substrate (entries 34–37). both cases, docking calculations with Autodock 3.0[28]
Much to our surprise, a striking switch of stereopre- yielded clear binding modes for 10a which differ
ference was observed (Scheme 2): while OPR1 re- mainly in the position and orientation of the phenyl
duced the substrate to the corresponding (R)-product group (Figure 1). Whereas the catalytically active resi-
Figure 1. Docking of 1-nitro-2-phenylpropene (10a) within the active site of OPR1 (left) and OPR3 (right). The sphere rep-
resents the hydride being added from the flavin onto Cb of the substrate.
414
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 411 – 418

Asymmetric Bioreduction of C=C Bonds using Enoate Reductases
FULL PAPERS
clohexenone (4a) and 2-methylcyclohexanone (4b) were
provided by BASF (Ludwigshafen), (R)- and (S)-citronellal
(1b), (R)-3-methylcyclopentanone (5b), N-phenyl-2-methyl-
maleimide (9a), NAD+ and ammonium formate were from
Aldrich. 4-Ketoisophorone (7a) was purchased from ABCR,
2-methyl-2-cyclopentenone (3a) was from Acros, citraconic
anhydride was purchased from Alfa Aesar, 3-methyl-2-cy-
clopentenone (5a), ammonium acetate and glucose were
from Fluka, NADH, NADPH and NADP+ were purchased
from Biocatalytics, glucose 6-phosphate and glucose 6-phos-
phate dehydrogenase were obtained from Biochemica, for-
mate dehydrogenase and glucose dehydrogenase were from
Jülich Chiral Solutions.
The open reading frame of Lycopersicon esculentum
OPR1 was cloned into pET-21a and expressed as a C-termi-
nal hexahistidine tagged protein in E. coli BL21 cells. The
expressed recombinant protein was purified on an Ni-NTA
affinity column (Invitrogen) according to the manufacturerꢁs
protocol. Lycopersicon esculentum OPR3 and YqjM from
Bacillus subtilis were expressed and purified as recently re-
ported.[17,29]
GC-MS analyses were performed on an HP 6890 Series
GC system equipped with a 5973 mass selective detector
and a 7683 Series injector using a (5% phenyl)-methylpoly-
siloxane capillary column (HP-5Msi, 30 m, 0.25 mm ID,
0.25 mm film). GC-FID analyses were carried out on a
Varian 3800 using H2 as carrier gas (14.5 psi). HPLC analy-
ses were performed using a Shimadzu system equipped with
a Chiralcel OD-H column (25 cm, 0.46 cm). Circular dichro-
ism spectra were measured on a JASCO spectropolarimeter
J-715. NMR spectra were measured on a Bruker AMX spec-
trometer at 360 MHz.
dues, such as the flavin and the O binding site of the
substrate are highly conserved throughout the OYE
family, loop regions, especially b-strands b3 and b6,[27]
which create the specificity site for substrate recogni-
tion, show almost no conservation. In OPR1
(Figure 1, left), Tyr-78, Lys-79 and Tyr-358 build up
the entrance of the active site cavity forming a deep
pocket on the bottom that accommodates the phenyl
ring of 10a. In OPR3, however, Lys-79 is altered to
Pro-75, which results in a conformational change of
this loop and forces Phe-74 (equivalent to Tyr-78 in
OPR1) to adopt a different orientation. This confor-
mational change reduces the distance between the
two residues (Phe-74 and Tyr-370) flanking the sub-
strateꢁs b-substituent, leading to a pocket with a flat
bottom, thus forcing the substrate to turn around and
to bind in the opposite orientation: the smaller
pocket on the bottom now accommodates the methyl
group with the phenyl moiety pointing upwards, so
that the hydride attacks the Cb atom of 10a from the
opposite side of the C=C-bond to furnish the (S)-en-
antiomer (Figure 1, right).
Conclusions
Isoenzymes OPR1 and OPR3 of 12-oxophytodienoate
reductase from Lycopersicon esculentum (tomato),
and the “old yellow enzyme” homologue YqjM from
Bacillus subtilis displayed a remarkably broad sub-
strate spectrum in the asymmetric bioreduction of
olefins bearing a carbonyl or nitro group as activating
substituent. Thus, enals, enones, maleimides and nitro-
olefins were reduced with excellent stereoselectivities
to the corresponding saturated products at the ex-
pense of NADH or NADPH, which were equally well
Synthesis of Substrates and Reference Material
(6R)-Levodione (7b): Ketoisophorone was reduced to levo-
dione using bakerꢁs yeast yielding the (6R)-enantiomer[21] in
62% ee.
2-Methylmaleimide (8a): Ammonium acetate (2 g,
accepted. Cofactor recycling using NADH/glucose/ 26.0 mmol) and citraconic anhydride (0.9 mL, 10 mmol)
were added to acetic acid (5 mL) and heated under reflux
for 2 h. The solution was cooled down to room temperature
and evaporated. After the addition of ice/water (20 mL) to
the dark syrupy residue, the aqueous phase was extracted
with EtOAc (8”5 mL) and CH2Cl2 (2”5 mL). The com-
bined organic phases were evaporated to give a yellow solid
that was purified by silica flash chromatography (EtOAc/
light petroleum, 1:1) to afford 8a as white crystals; yield:
300 mg (27%);[30] mp 1058C; 1H NMR (DMSO): d=1.93–
1.94 (3H, d, J=1.8 Hz), 6.48–6.50 (1H, m), 10.74 (br. s,
GDH or NADPH/glucose 6-phosphate/G6PDH was
successful, but the NADH/formate/FDH system
turned out to be plagued by competing carbonyl re-
duction presumably caused by alcohol dehydrogenas-
es in commercial FDH preparations. In contrast to
whole-cell transformations, the reaction showed high
chemoselectivity for the conjugated C=C bond, while
isolated olefins and carbonyl groups remained un-
touched. With 1-nitro-2-phenylpropene, isoenzymes
OPR1 and OPR3 behaved in a sharp stereocomple- NH).
mentary fashion by producing opposite enantiomeric
rac-2-Methylsuccinimide (8b): 2-Methylmaleimide (8a,
107 mg, 0.96 mmol) was dissolved in THF (10 mL) and was
hydrogenated at atmospheric pressure at room temperature
in the presence of 10% Pd/C (5 mg) as catalyst. After 20 h,
the mixture was filtered through Celite and evaporated to
afford of rac-8b; yield: 97 mg (0.86 mmol, 89%); mp 658C;
1H NMR (360 MHz, DMSO): d=1.35–1.37 (3H, d, J=
7.2 Hz); 2.35–2.42 (1H, m); 2.89–3.00 (2H, m); 8.80 (br. s,
NH).
products.
Experimental Section
General Remarks
Citral (1a), 2-methylpentenal (2a), 2-methylpentanal (2b), 2-
methylpentanol, 2-methylcyclopentanone (3b), 2-methylcy-
rac-N-Phenyl-2-methylsuccinimide (9b):
methylmaleimide (9a, 50 mg, 0.27 mmol) was dissolved in in
N-Phenyl-2-
Adv. Synth. Catal. 2008, 350, 411 – 418
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
415

MØlanie Hall et al.
FULL PAPERS
ethyl acetate (5 mL) and was hydrogenated at atmospheric
pressure at room temperature using 10% Pd/C (2.8 mg) as
catalyst. After 24 h, the mixture was filtered through Celite
and evaporated to afford rac-9b; yield: 48 mg (0.25 mmol,
94%); 1H NMR (CDCl3): d=1.47 (d, 3H, J=7 Hz), 2.52
(dd, 1H, J=17.4 Hz, J=4 Hz), 3.01–3.10 (m, 1H), 3.11 (dd,
1H, J=17.3 Hz, J=9.2 Hz), 7.29–7.51 (m, 5H).
genase or glucose 6-phosphate dehydrogenase, 10 U). The
mixture was shaken at 308C and 140 rpm for 24 h and
worked up as described above.
Analytical Procedures
Determination of conversion: Conversions for citronellal
(1b) and 2-methylcyclohexanone (4b) were analysed by GC-
FID using a PEG-phase capillary column (Varian CP-Wax
52 CB, 30 m, 0.25 mm, 0.25 mm), detector temperature
2508C, split ratio 20:1. Programme: 1008C, hold for 2 min,
158CminÀ1 to 2108C, hold for 2.5 min. Retention times were
as follows: citronellal (1b) 5.21 min, neral (Z-1a) 7.10 min,
geranial (E-1a) 7.53 min, 4b 5.21 min and 4a 7.10 min. Con-
versions of 2-methylpentenal (2a), 2-methylcyclopentenone
(3a), 3-methylcyclopentenone (5a), ketoisophorone (7a), N-
phenyl-2-methylmaleimide (9a) and 1-nitro-2-phenylpropene
(10a) were determined using a 6% cyanopropyl-phenyl
phase capillary column (Varian CP-1301, 30 m, 0.25 mm,
0.25 mm), detector temperature 2508C, split ratio 30:1. Tem-
perature programme for 2a: 408C hold 2 min, 208CminÀ1 to
1808C, hold 1.5 min. Retention times: 2b 4.47 min and 2a
5.44 min. Temperature programme for 3a and 5a: 808C hold
10 min, 308CminÀ1 to 2008C, hold 2 min. Retention times:
3b 4.25 min, 3a 5.82 min, 5b 4.44 min, 5a 8.77 min. Tempera-
ture programme for ketoisophorone 7a: 1108C hold 5 min,
308CminÀ1 to 2008C, hold 2 min. Retention times: 7a
6.78 min and 7b 7.28 min. Temperature programme for N-
l-Nitro-2-phenylpropene (10a): To a stirred mixture of
acetic anhydride (40 mL) and 65% nitric acid (5.28 g) was
added 2-phenylpropene (3.2 mL, 12.2 mmol) at 08C. After
20 min, the solution was poured into water (180 mL) and
stirred for additional 30 min. The organic layer was washed
with saturated aqueous NaHCO3, water and then dried
(Na2SO4). Removal of the solvent under reduced pressure
gave an oily residue of crude 2-acetoxy-l-nitro-2-phenylpro-
pane, which was used without purification. A solution of the
nitroacetate in triethylamine (15 mL) and chloroform
(30 mL) was stirred for 3 h at room temperature. After addi-
tion of 2 N HCl (30 mL), the mixture was extracted with di-
chloromethane and dried (Na2SO4). Evaporation of the sol-
vent followed by silica gel chromatography (hexane/ethyl
1
acetate, 20:1) afforded 10a; yield: 25%; H NMR (CDCl3):
d=2.66 (d, 3H, J=1.3 Hz), 7.32 (d, 1H, J=1.3 Hz), 7.46 (s,
3
3
5H). J-type HMBC: JH-1,C-3 =6.0Æ0.3 Hz, JH-1,C-Ar ꢀ5.2 Hz.
1-Nitro-2-phenylpropane rac-(10b): trans-b-Nitrostyrene
(0.45 g, 3 mmol) in 20 mL dry ether was added to methyl-
magnesium iodide (5 mL of a 3M solution, 15 mmol) in
40 mL of ether at À208C. Within 10 min, the solution was
added to ice cold 5% aqueous HCl solution and stirred for
30 min. The solution was extracted with CH2Cl2, dried over
MgSO4, filtered and the solvent was evaporated to give rac-
10b; yield: 22%; 1H NMR (CDCl3): d= 1.4 (d, 3H, J=
6.9 Hz), 3.62–3.72 (m, 1H), 4.51–4.60 (m, 2H), 7.24–7.38 (m,
5H).
phenyl-2-methylmaleimide
9a:
1108C
hold
2 min,
308CminÀ1 to 2108C, hold 6 min. Retention times: 9a
8.78 min and 9b 9.90 min. Temperature programme for 1-
nitro-2-phenylpropene (10a): 1208C hold 3 min, 108CminÀ1
to 1808C, 208CminÀ1 to 2208C, hold 2 min. Retention times:
10b 8.88 min and 10a 9.56 and 10.27 min (E/Z-isomers).
Conversion of 2-methylmaleimide (8a) was determined
using a modified b-cyclodextrin capillary column (Chiraldex
B-TA, 40 m, 0.25 mm), detector temperature 2008C, split
ratio 10:1. Programme: 1308C hold 2 min, 108CminÀ1 to
1608C, hold 6 min. Retention times: 8a 6.15 min, (S)-8b and
(R)-8b 8.74 and 8.89 min, respectively.
General Procedure for the Enzymatic Bioreduction
of Substrates 1a–10a
An aliquot of OPR1, OPR3 or YqjM (protein purity >90%,
protein content 75–275 mg/mL) was added to a Tris-HCl
buffer solution (0.8 mL, 50 mM, pH 7.5) containing the sub-
strate (5 mM) and the cofactor NADH or NADPH
(15 mM). N-Phenyl-2-methylmaleimide (8a) was added as a
0.5M DMF solution (1% final DMF concentration) to over-
come its poor solubility in water. The mixture was shaken at
308C and 140 rpm. After 48 h, products were extracted with
EtOAc (2”0.5 mL) containing 0.05% (v/v) of 1-octanol (for
1a/1b) or (R)-limonene (for 3a/3b-10a/10b) as internal GC
standard. The combined organic phases were dried
(Na2SO4) and the resulting samples were analysed on achiral
GC. Products were identified by comparison with authentic
reference materials (which were either commercially avail-
able or were independently synthesised) via co-injection on
GC-MS and achiral GC.
Determination of enantiomeric excess and absolute con-
figuration: Citronellal (1b): The enantiomeric excess was de-
termined using a modified b-cyclodextrin capillary column
(Hydrodex-b-TBDAc, 25 m, 0.25 mm). Detector tempera-
ture 2008C, injector temperature 1808C, split ratio 20:1.
Temperature programme for 1b: 408C hold 2 min, 48CminÀ1
to 1208C, hold 1 min, 208CminÀ1 to 1808C, hold 3 min. Re-
tention times: (S)-1b and (R)-1b 19.84 and 19.97 min, re-
spectively.
2-Methylpentanal (2b) was analysed as the corresponding
prim-alcohol (2-methylpentanol) after reduction with
NaBH4 in MeOH. The enantiomeric excess was determined
using a modified b-cyclodextrin column (Hydrodex-b-6-
TBDM, 25 m, 0.25 mm), split ratio 100:1. Temperature pro-
gramme for 2-methylpentanol: 708C hold 9 min, 108CminÀ1
to 1608C, hold 10 min. Retention times: (R)- and (S)-2-
methylpentanol 10.42 and 11.05 min, respectively.
General Procedure for Cofactor Recycling
OPR1, OPR3 or YqjM was added to a Tris-HCl buffer solu-
tion (0.8 mL, 50 mM, pH 7.5) containing the substrate
(5 mM), the oxidised form of the cofactor (NAD+ or
NADP+, 100 mM), the cosubstrate (ammonium formate, glu-
cose or glucose 6-phosphate, 20 mM) and the corresponding
recycling enzyme (formate dehydrogenase, glucose dehydro-
Enantiomeric excesses of 3b, 4b, 5b and 8b were deter-
mined using a modified b-cyclodextrin capillary column
(Chiraldex B-TA, 40 m, 0.25 mm). Detector temperature
2008C, injector temperature 1808C, split ratio 25:1. Temper-
ature programme for 3b: 708C hold 8 min, 108CminÀ1 to
416
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 411 – 418

Asymmetric Bioreduction of C=C Bonds using Enoate Reductases
FULL PAPERS
808C, hold 2 min, 308CminÀ1 to 1808C, hold 2 min. Reten-
tion times: (R)-3b and (S)-3b 10.35 and 10.62 min, respec-
tively. Temperature programme for 4b: 808C hold 2 min,
58CminÀ1 to 1058C, 108CminÀ1, hold 4 min. Retention
times: (R)-4b and (S)-4b 6.34 and 6.47 min, respectively.
Temperature programme for 5b: 708C hold 10 min,
308CminÀ1 to 1808C, hold 4 min. Retention times: (S)-5b
and (R)-5b 11.19 and 11.27 min, respectively. Temperature
programme for 8b: 1308C hold 2 min, 108CminÀ1 to 1608C
hold 6 min. Retention times: (S)-8b and (R)-8b 8.74 and
8.89 min, respectively.
The enantiomeric excess of 6b and 10b was determined
using a b-cyclodextrin capillary column (CP-Chirasil-DEX
CB, 25 m, 0.32 mm, 0.25 mm film). Temperature programme
for 6b: 908C hold 2 min, 48CminÀ1 to 1158C, 208CminÀ1 to
1808C, hold 2 min. Retention times: (R)-6b and (S)-6b 6.42
and 6.74 min, respectively. Temperature programme for 10b:
1058C hold 5 min, 18CminÀ1 to 1158C, hold 1 min,
208CminÀ1 to 1808C, hold 2 min. Retention times: (S)-10b
and (R)-10b 7.90 and 8.08 min, respectively.
Knowles, Angew. Chem. 2002, 114, 2096–2107; Angew.
Chem. Int. Ed. 2002, 41, 1998–2007.
[2] a) J. W. Yang, M. T. Hechavarria Fonseca, N. Vignola,
B. List, Angew. Chem. 2004, 117, 110–112; Angew.
Chem. Int. Ed. 2004, 44, 108–110; b) J. B. Tuttle, S. G.
Ouellet, D. W. C. MacMillan, J. Am. Chem. Soc. 2006,
128, 12662–12663; c) N. J. A. Martin, L. Ozores, B.
List, J. Am. Chem. Soc. 2007, 129, 8976–8977.
[3] R. E. Williams, N. C. Bruce, Microbiology 2002, 148,
1607–1614.
[4] S. Steinbacher, M. Stumpf, S. Weinkauf, F. Rohdich, A.
Bacher, H. Simon, in: Flavins and flavoproteins, (Eds.:
S. K. Chapman, R. N. Perham, N. S. Scrutton), Weber,
Berlin, 2002, pp 941–949.
[5] R. Stuermer, B. Hauer, M. Hall, K. Faber, Curr. Opin.
Chem. Biol. 2007, 11, 203–213.
[6] O. Warburg, W. Christian, Biochem. Z. 1933, 266, 377–
411.
[7] P. A. Karplus, K. M. Fox, V. Massey, FASEB J. 1995, 9,
1518–1526.
The absolute configurations of 1b–7b were determined by
co-injection with reference materials of known absolute con-
figuration.[31–33] The absolute configurations of 3b and 4b
were additionally confirmed via comparison of the CD spec-
tra using independently synthesised reference material.
Enantioenriched 3b or 4b obtained by reduction of 3a or 4a
(25 mg), respectively, using YqjM was dissolved in dioxane
(1 mL) and the solution was analysed on a spectropolarime-
ter in a 1 mm plexiglas cuvette. The scan was performed be-
tween 330 and 250 nm. 3b showed CD [q]296 =+3.2 and 4b
displayed CD [q]291 =À39.3, which are in agreement with lit-
erature data: (S)-3b [q]298 =+4143;[34] (R)-4b [q]286 =+
24.7;[35] the corresponding racemates gave a flat baseline.
The enantiomeric excess of 9b was determined on HPLC
using n-heptane/EtOH, 95:5 (isocratic) at 188C. Retention
times: (R)-9b and (S)-9b 27.15 min and 29.10 min, respec-
tively. The absolute configurations of 8b and 9b were deter-
mined via comparison of the CD spectra using independent-
ly synthesised reference material. Enantioenriched 8b
(13 mg) or 9b (16 mg) obtained by reduction of 8a or 9a, re-
spectively, using OPR1 was dissolved in cyclohexane or, re-
spectively, CHCl3 (3 mL) and the solution was analysed on a
spectropolarimeter in a 5 mm plexiglas cuvette. The scan
was performed between 300 and 230 nm. 8b showed CD
[q]251 =À5.3 and 9b showed CD [q]272 =À14, which are in
agreement with literature data: (R)-8b CD [q]253 =À840,
(R)-9b [q]272 =À76.9;[36,37] the corresponding racemates gave
a flat baseline.
[8] K. Shimoda, D. I. Ito, S. Izumi, T. Hirata, J. Chem.
Soc., Perkin Trans. 1 1996, 355–358.
[9] C. Fuganti, P. Grasselli, J. Chem. Soc., Chem.
Commun. 1979, 995–997.
[10] A. Müller, B. Hauer, B. Rosche, J. Mol. Catal. B:
Enzym. 2006, 38, 126–130.
[11] a) I. L. Gatfield, Adv. Biochem. Eng. Biotechnol. 1997,
55, 221–238; b) S. Servi, Synthesis 1990, 1, 1–25; c) C.
Fuganti, Pure Appl. Chem. 1990, 62,1449–1452.
[12] a) J. F. Chaparro-Riggers, T. A. Rogers, E. Vazquez-Fig-
ueroa, K. M. Polizzi, A. S. Bommarius, Adv. Synth.
Catal. 2007, 349, 1521–1531; b) M. A. Swiderska, J. D.
Stewart, J. Mol. Catal. B: Enzym. 2006, 42, 52–54;
M. A. Swiderska, J. D. Stewart, Org. Lett. 2006, 8,
6131–6133; c) M. Wada, A. Yoshizumi, Y. Noda, M.
Kataoka, S. Shimizu, H. Takagi, S. Nakamori, Appl.
Environ. Microbiol. 2003, 69, 933–937.
[13] A. Müller, B. Hauer, B. Rosche, Biotechnol. Bioeng.
2007, 98, 22–29.
[14] K. Hult, P. Berglund, Trends Biotechnol. 2007, 25, 231–
236.
[15] A. Müller, R. Stürmer, B. Hauer, B. Rosche, Angew.
Chem. 2007, 119, 3380–3382; Angew. Chem. Int. Ed.
2007, 46, 3316–3318.
[16] M. Hall, C. Stueckler, W. Kroutil, P. Macheroux, K.
Faber, Angew. Chem. 2007, 119, 4008–4011; Angew.
Chem. Int. Ed. 2007, 46, 3934–3937.
[17] K. Kitzing, T. B. Fitzpatrick, C. Wilken, J. Sawa, G. P.
Bourenkov, P. Macheroux, T. Clausen, J. Biol. Chem.
2005, 280, 27904–27913.
[18] M. Hall, B. Hauer, R. Stuermer, W. Kroutil, K. Faber,
Tetrahedron: Asymmetry 2006, 17, 3058–3062.
[19] I. Thanos, A. Deffner, H. Simon, Biol. Chem. Hoppe-
Seyler 1988, 369, 451–460.
Acknowledgements
We would like to thank Nicole Müller, Tea Pavkov and Klaus
Zangger for their excellent technical assistance.
[20] B. J. Brown, Z. Deng, A. Karplus, V. Massey, J. Biol.
Chem. 1998, 273, 32753–32762.
[21] a) H. G. W. Leuenberger, W. Boguth, E. Widmer, R.
Zell, Helv. Chim. Acta 1976, 59, 1832–1849; M. Kata-
References
oka, A. Kotaka, A. Hasegawa, M. Wada, A. Yoshizumi,
S. Nakamori, S. Shimizu, Biosci. Biotechnol. Biochem.
2002, 66, 2651–2657.
[1] a) R. Noyori, Angew. Chem. 2002, 114, 2108–2123;
Angew. Chem. Int. Ed. 2002, 41, 2008–2022; b) W. S.
Adv. Synth. Catal. 2008, 350, 411 – 418
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
417

MØlanie Hall et al.
FULL PAPERS
[22] a) A. de Raadt, H. Griengl, Curr. Opin. Biotechnol.
2002, 13, 537–542; b) B. V. Nguyen, O. Nordin, C.
Vçrde, E. Hedenstrçm, H. E. Hçgberg, Tetrahedron:
Asymmetry 1997, 8, 983–986; c) K. Adachi, S. Kobaya-
shi, M. Ohno, Chimia 1986, 40, 311–314.
[27] C. Breithaupt, J. Strassner, U. Breitinger, R. Huber, P.
Macheroux, A. Schaller, T. Clausen, Structure 2001, 9,
419–429.
[28] G. M. Morris, D. S. Goodsell, R. S. Halliday, R. Huey,
W. E. Hart, R. K. Belew, A. J. Olson, J. Comput. Chem.
1998, 19, 1639–1662.
[29] C. Breithaupt, R. Kurzbauer, H. Lilie, A. Schaller, J.
Strassner, R. Huber, P. Macheroux, T. Clausen, Proc.
Natl. Acad. Sci. USA 2006, 103, 14337–14342.
[30] P. Kuehne, M. Hesse, Tetrahedron 1993, 49, 4575–4580.
[31] V. Bertolasi, O. Bortolini, M. Fogagnolo, G. Fantin, P.
Pedrini, Tetrahedron: Asymmetry 2001, 12, 1479–1483.
[32] M. A. Swiderska, J. D. Stewart, J. Mol. Catal. B:
Enzym. 2006, 42, 52–54.
[23] N. Ono, in: The Nitro Group in Organic Synthesis, (Ed:
H. Feuer), Wiley-VCH, New York, 2001.
[24] K. Sakai, A. Nakazawa, K. Kondo, H. Ohta, Agric.
Biol. Chem. 1985, 49, 2331–2335.
[25] a) Y. Ijima, K. Matoishi, Y. Terao, N. Doi, H. Yanaga-
wa, H. Ohta, Chem. Commun. 2005, 877–879; b) A. O.
Magnusson, M. Takwa, A. Hamberg, K. Hult, Angew.
Chem. 2005, 117, 4658–4661; Angew. Chem. Int. Ed.
2005, 44, 4582–4585; c) M. T. Reetz, B. Brunner, T.
Schneider, F. Schulz, C. M. Clouthier, M. M. Kayser,
Angew. Chem. 2004, 116, 4167–4170; Angew. Chem.
Int. Ed. 2004, 43, 4075–4078; d) P. F. Mugford, S. M.
Lait, B. A. Keay, R. J. Kazlauskas, ChemBioChem
2004, 5, 980–987; e) M. A. Phillips, M. R. Wildung,
D. C. Williams, D. C. Hyatt, R. Croteau, Arch. Bio-
chem. Biophys. 2003, 411, 267–276; f) S. Wang, M. M.
Kayser, H. Iwaki, P. C. K. Lau, J. Mol. Catal. B:
Enzym. 2003, 22, 211–218; g) E. Ferre, M. H.
N’Guyen, J. Le Petit, G. Gil, Biocatal. Biotransform.
2002, 20, 311–317; h) T. Schubert, W. Hummel, M. R.
Kula, M. Müller, Eur. J. Org. Chem. 2001, 4181–4187.
[26] M. Norin, K. Hult, A. Mattson, T. Norin, Biocatalysis
1992, 7, 131–147.
[33] H. Ohta, N. Kobayashi, K. Ozaki, J. Org. Chem. 1989,
54, 1802–1804.
[34] J. J. Partridge, N. K. Chadha, M. R. Uskokovic, J. Am.
Chem. Soc. 1973, 95, 532–540.
[35] a) A. I. Meyers, D. R. Williams, G. W. Erickson, S.
White, M. Druelinger, J. Am. Chem. Soc. 1981, 103,
3081–3087; b) M. Duraisamy, H. M. Walborsky, J. Am.
Chem. Soc. 1983, 105, 3252–3264.
[36] T. Polonski, J. Chem. Soc., Perkin Trans. 1 1988, 629–
637.
[37] T. Hirata, A. Takarada, M. E. F. Hegazy, Y. Sato, A.
Matsushima, Y. Kondo, A. Matsuki, H. Hamada, J.
Mol. Catal. B: Enzym. 2005, 32, 131–134.
418
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 411 – 418