Emulation of racemase activity by employing a pair of stereocomplementary biocatalysts

Article abstract of DOI:10.1002/chem.200700528

Racemization is the key step to turn a kinetic resolution process into dynamic resolution. A general strategy for racemization under mild reaction conditions by employing stereoselective biocatalysts is presented, in which racemization is achieved by employing a pair of stereocomplementary biocatalysts that reversibly interconvert an sp³ to a sp² center. The formal interconversion of the enantiomers proceeds via a prochiral sp ² intermediate the formation of which is catalyzed either by two stereocomplementary enzymes or by a single enzyme with low stereoselectivity. By choosing appropriate reaction conditions, the amount of the prochiral intermediate is kept to a minimum. This general strategy, which is applicable to redox enzymes (e.g., by acting on R2CHOH and R₂CHNHR groups) and lyase-catalyzed addition-elimination reactions, was proven for the racemization of secondary alcohols by employing alcohol dehydrogenases. Thus, enantiopure chiral alcohols were used as model substrates and were racemized either with highly stereoselective biocatalysts or by using (rarely found) non-selective enzymes.

Full text of DOI:10.1002/chem.200700528

FULL PAPER
DOI: 10.1002/chem.200700528
Emulation of Racemase Activity by Employing a Pair of
Stereocomplementary Biocatalysts
[
a]
[b]
[b]
[c]
Christian C. Gruber, Bettina M. Nestl, Johannes Gross, Petra Hildebrandt,
[
c]
[a]
[a]
Uwe T. Bornscheuer, Kurt Faber, and Wolfgang Kroutil*
Abstract: Racemization is the key step
to turn a kinetic resolution process into
dynamic resolution. A general strategy
for racemization under mild reaction
conditions by employing stereoselec-
tive biocatalysts is presented, in which
racemization is achieved by employing
a pair of stereocomplementary biocata-
tion of which is catalyzed either by two
stereocomplementary enzymes or by a
single enzyme with low stereoselectiv-
ity. By choosing appropriate reaction
conditions, the amount of the prochiral
intermediate is kept to a minimum.
This general strategy, which is applica-
ble to redox enzymes (e.g., by acting
on R CHOH and R CHNHR groups)
2
2
and lyase-catalyzed addition–elimina-
tion reactions, was proven for the race-
mization of secondary alcohols by em-
ploying alcohol dehydrogenases. Thus,
enantiopure chiral alcohols were used
as model substrates and were race-
mized either with highly stereoselective
biocatalysts or by using (rarely found)
non-selective enzymes.
Keywords: alcohol dehydrogenase ·
biotransformations · dynamic reso-
3
lysts that reversibly interconvert an sp
2
to a sp center. The formal interconver-
lution
·
racemization
·
redox
sion of the enantiomers proceeds via a
chemistry · secondary alcohols
2
prochiral sp intermediate the forma-
Introduction
stable compounds, such as secondary alcohols and chiral
amines, are more difficult to racemize. Previously, this has
Recently, considerable efforts have been made to improve
been accomplished by redox processes that are mediated by
[1]
[3a–d,4]
and develop novel methods for racemization. Racemiza-
tion is needed in asymmetric synthesis for the recycling of
undesired enantiomers that are obtained from kinetic reso-
transition-metal complexes.
Due to the high specificity
of biosynthetic pathways, nature has, in general, little need
for racemization and thus the number of “true” racemases is
[2]
[3e]
lution, as well as for the preparation of chiral alcohol and
very limited. Thus only a few specialized enzymes—race-
[
3,4]
amine functionalities by dynamic resolution.
zation of stereolabile compounds, such as cyanohydrins,
hemi(thio)acetals, a-substituted carbonyl compounds or a-
aryl-substituted hydantoins, can easily be achieved by mild
The racemi-
mases—are known to catalyze the racemization of a-hydrox-
ycarboxyllic acids (such as mandelate and derivatives), (N-
[1,3e]
A
C
H
T
R
E
U
N
G
acyl) a-amino acids, and hydantoins.
However, no de-
fined enzymes have been described for the racemization of
[1,4c,5]
[1,6]
acid or base catalysis.
In contrast, stereochemically
secondary alcohols or chiral primary amines.
[
a] C. C. Gruber, Prof. K. Faber, Prof. W. Kroutil
Department of Chemistry, Organic and Bioorganic Chemistry
University of Graz, Heinrichstrasse 28, 8010 Graz (Austria)
Fax : (+43)316-380-9840
Results and Discussion
Inspired by the racemization mechanism of chiral primary
amines and secondary alcohols by transition-metal cataly-
sis via a prochiral intermediate, as well as a consequence
of our own research on biocatalytic racemization, we de-
veloped a racemization strategy based on stereoselective
bio)catalysts. We envisaged applying biocatalysts that inter-
convert a chiral center to a prochiral intermediate in analo-
gy to transition-metal catalyzed racemization (Scheme 1).
As indicated in Scheme 1a, the chiral catalyst has a cer-
tain stereopreference to transform the prochiral compound
E-mail: Wolfgang.Kroutil@uni-graz.at
[7]
[
b] Dr. B. M. Nestl, Ing. J. Gross
[
6]
Research Centre Applied Biocatalysis Petersgasse 14
8
010 Graz (Austria)
[
c] P. Hildebrandt, U. T. Bornscheuer
(
Department of Biotechnology & Enzyme Catalysis
Institute of Biochemistry, Greifswald University
Felix-Hausdorffstr. 4, 17487 Greifswald (Germany)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Chem. Eur. J. 2007, 13, 8271 – 8276
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8271

epimerization of diastereomers represent an extension of
the concept.
Overall, the following prerequisites for biocatalytic race-
mization have to be fulfilled:
2
Scheme 1. P=compound with a prochiral center, sp hybrid; S, R=enan-
tiomers, sp hybrid. a) Desymmetrization of a prochiral compound P by a
3
1) One must have a matching pair of (highly selective) bio-
catalysts that possess opposite stereopreferences for the
reversible transformation of both enantiomers via the
corresponding prochiral intermediate.
stereoselective catalyst furnishes a single product enantiomer S; b) rever-
sible transformation of enantiomer S to the prochiral intermediate P by a
stereoselective catalyst “destroying” the chiral center; c) racemization of
R or S either by employing two (bio)catalysts with opposite stereoprefer-
ence or a single (bio)catalyst with little (or no) stereopreference.
2
) Alternatively, a single biocatalyst with incomplete stereo-
selectivity will equally be effective.
3
) The equilibrium between substrate enantiomers and pro-
chiral intermediates can be controlled by adjustment of
the reaction conditions (e.g., the redox balance, etc.).
P into enantiomer S in a desymmetrization reaction. If the
process is reversible, one can start from a single enantiomer
4) In case a cofactor is required and two enzymes are em-
ployed, free exchange of the (identical) cofactor between
the biocatalysts has to be ensured.
(
e.g., S), of which the chiral centre becomes destroyed
during the transformation to prochiral P (Scheme 1b). To
achieve racemization, two reversible stereocomplementary
transformations (PQS, and PQR) have to take place
5) For redox-mediated racemization with cofactors, external
cofactor recycling should not be required, because the
[10]
(
Scheme 1c). The equilibrium between the substrate enan-
net redox balance of this process is zero.
tiomers, S and R, and the prochiral intermediate P can be
kept on the substrate side by choosing the appropriate reac-
tion conditions (e.g., redox-balance, etc.).
To prove this general strategy, we envisaged employing al-
cohol dehydrogenases (ADHs) for the racemization of sec-
[11]
On first glimpse, a single chiral catalyst that possesses per-
fect stereoselectivity is not suitable. Fortunately, no catalyst
is absolutely stereoselective and during the interconversion
of S to P, a small amount of R will always be formed due to
ondary alcohols via the prochiral ketone intermediate. As
ADHs are highly enantioselective in general, only enantio-
mer S is oxidized to ketone P (Scheme 1b). After the ther-
+
modynamic ketone–alcohol–NAD(P)H–NAD(P) equilibri-
[8]
“selectivity mistakes” that finally lead to racemization.
um is reached, the velocity of oxidation and reduction are
identical and (in the case that the ADH shows perfect ste-
reoselectivity) the ketone is reduced back to the same sub-
strate enantiomer, S. Consequently, by employing a second
ADH with the opposite stereopreference, ketone P can be
reduced to enantiomer R. If both processes are in equilibri-
The rate of racemization is a function of the stereoselectiv-
ity of the chiral catalyst; in other words, the number of “se-
lectivity mistakes” that occur. However, most biocatalysts
show such high stereoselectivity that racemization which is
based on a single catalyst would take an inordinately long
reaction time; this would be insufficient to allow dynamic
[12]
um, overall racemization is achieved (Scheme 2).
The
[9]
kinetic resolution, which requires fast racemization rates.
ketone-to-alcohol ratio depends on the amount and ratio of
the oxidized and reduced form of the cofactor added, that
This bottleneck can be overcome by employing two differ-
ent biocatalysts that possess opposite stereopreferences in
order to provide reaction rates of a similar magnitude for
both enantiomers (Scheme 1c); this ensures fast racemiza-
tion. For evident reasons, the fastest rates are achieved by
employing either a pair of stereocomplementary biocatalysts
with equal activity or a single nonstereospecific catalyst.
Due to the intrinsic excellent regio- and chemoselectivity of
the enzymes, one can expect that such racemization process-
es will not affect other functional groups within the mole-
cule, and are therefore “clean”; this avoids the formation of
undesired side-products.
+
is, NAD(P) versus NAD(P)H. Hence, the ketone-to-alco-
hol ratio can be controlled by the amount of cofactors ap-
plied to the system. By keeping the concentration of
+
NAD(P) to a minimum, the ratio of ketone to alcohol is
This general concept is applicable for all substrates of cen-
3
tral chirality, in which an sp center is enzymatically trans-
2
formed to an sp center in a reversible fashion. Examples
for such enzymes are redox enzymes that act on R CHOH
2
and R CHNHR groups (such as alcohol dehydrogenases
2
and a-amino acid dehydrogenases) and C=C bonds (enoate
reductases). Alternatively, lyase-catalyzed addition–elimina-
tion reactions that involve C=C bonds (for instance hydra-
tases, ammonia lyases) can be employed. Applications for
Scheme 2. Biocatalytic racemization of secondary alcohols 1a–1h and
acyloins 1i and 1j via the ketone intermediate by employing alcohol de-
hydrogenase(s).
8272
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 8271 – 8276

Racemization by Biocatalysts
FULL PAPER
kept on the alcohol side in the ketone–alcohol–NAD(P)H–
NAD(P) equilibrium.
Table 1. Biocatalytic racemization of secondary alcohols and acyloins by
employing pairs of stereocomplementary alcohol dehydrogenases.
[
a]
+
[
b]
[c]
Enzymes
ADH-“A” and LK-ADH (S)-1a
R)-1a
(R)-1b
Substrate ee [%]
(Di)Ketone [%]
For the racemization experiments, two commercially
[13]
available ADHs, ADH-“A” from Rhodococcus ruber and
ADH from Lactobacillus kefir (LK-ADH) were em-
rac
rac
rac
rac
rac
rac
rac
rac
rac
2.3
1.3
<0.1
8
0.9
3.4
2.4
5.1
2.6
0.3
(
[
[
d,f]
d,f]
[14]
[15]
ployed.
LK-ADH is known
to possess opposite (anti-
(
(
R)-1c
S)-1d
Prelog) stereospecificity to ADH-“A”. The racemization of
enantiopure (R)- and (S)-2-octanol (1a) was investigated by
monitoring the decline of enantiomeric excess over time
[f]
[
f]
(R)-1d
(S)-1e
(
(
(
(
(
(
(
R)-1e
S)-1 f
S)-1g
R)-1g
S)-1h
R)-1i
R)-1j
(
Figure 1).
[
f]
[
d]
19.2 (S)
25.0 (R)
>99.5 (S)
70.3( R)
>99.5 (R)
[
d]
1.3
[
d]
13.9
n.d.
[
d]
d]
[e]
[
8.6
3.4
2.2
<0.1
2.6
3.4
[
e]
RE-ADH and LK-ADH
(S)-1a
rac
rac
[
e]
(
(
(
(
(
(
(
(
R)-1a
R)-1b
R)-1c
S)-1e
R)-1e
S)-1h
R)-1i
R)-1j
[
[
d]
d]
40.2 (R)
32.2 (R)
35.4 (S)
23.9 (R)
>99.5 (S)
91.4 (R)
30.1 (R)
À1
[
d]
[
d]
5.7
23.1
n.d.
36.2
[
e]
[
d]
[
[
a] Reaction conditions: Substrate 2.5 gL , 24 h, 3 08 C, buffer pH 7.5.
b] The ee is based on GC (1a–1h) and HPLC (1i–1j); the ee of the
starting material was >99.5%. rac means ee <5%. [c] Determined by
GC/HPLC. [d] 1/5th of the enzyme was applied. [e] The diketone was not
detectable by HPLC. [f] 72 h.
Figure 1. Racemization of (R) and (S)-2-octanol (1a) by employing a
combination of two ADHs that possess opposite stereopreference
[4c,16]
(
Prelog-ADH-“A” from Rhodococcus ruber and anti-Prelog-LK-ADH
ed enone.
Nevertheless, 1i,1j were successfully race-
from Lactobacillus kefir). Experiments with single ADHs served as con-
trols.
mized by using pairs of isolated ADHs (Table 1). A clean
racemization procedure for the cyclic acyloin 1j is of special
interest, because the elimination reaction for this substrate
is promoted even by exposure to neutral silica gel during
column chromatography.
After 28 h, an enantiomeric excess (ee) of <15% for both
(
S)- and (R)-1a was achieved, while the amount of 2-octa-
none (2a) was kept below 5% by choosing an appropriate
amount and ratio of NADH/NAD ; this proves that the
For certain applications, the use of two enzymes might be
impossible, most notably if enzymes of opposite stereopre-
ference are not available. To overcome this limitation, we
envisaged the use of a single enzyme for racemization. How-
ever, as shown in Figure 1 and Table 2 (Entries 1 and 2), no
significant racemization was observed when a highly selec-
tive single ADH was employed. Only when enantiopure
(R)-2-octanol (1a) was incubated with the stereoselective
ADH-“A” for an extended period of 14 days, racemization
+
strategy is successful. As expected, no significant racemiza-
tion was observed by using only a single stereoselective
ADH.
Besides 1a, a range of aliphatic and aromatic secondary
alcohols 1b–1g were successfully racemized by employing
this two-enzyme system (Table 1). It is worth noting that
substrates that containing carbon–carbon multiple bonds
such as sulcatol (1e), 1-octene-3-ol (1 f), and 1-octyn-3-ol
(
1g), which would be susceptible to reduction by using tran-
Table 2. Racemization of (R)-1a (ee>99.5) by employing single alcohol
dehydrogenases.
sition-metal-catalyzed racemization, were interconverted in
a clean fashion without side-reactions.
[a]
Entry
Enzyme
t [h]
ee [%]
99.2 (R)
Ketone [%]
Additionally, racemization was proven for a combination
of another commercially available ADH, that is, Prelog-
ADH from Rhodococcus erythropolis (RE-ADH), and anti-
Prelog-selective LK-ADH. Only substrate 1h could not be
racemized, either with the combination of ADH-“A”/LK-
ADH or RE-ADH/LK-ADH.
Acyloins represent a group of substrates that are rather
difficult to racemize, because their chemical racemization is
often plagued by elimination as a major side-reaction, due
to the ease of formation of a resonance-stabilized conjugat-
1
2
3
4
5
6
7
8
ADH-“A”
LK-ADH
E. coli extract
ADH-“A”
Control
KRED-118
KRED-119
PF-ADH
24
24
24
336
336
72
<0.1
1.1
0.3
2.1
<0.1
0.7
98.0 (R)
>99.5 (R)
82.0 (R)
99.8 (R)
65.5 (R)
44.3( R)
rac
[
c]
[d]
[
b]
[b]
72
24
1.2
5.3
[
a] Determined by GC. [b] NADPH was used instead of NADH. [c] Cell-
free extract of E. coli. [d] No racemization took place with the (S)-enan-
tiomer.
Chem. Eur. J. 2007, 13, 8271 – 8276
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8273

W. Kroutil et al.
was observed (Table 2, Entry 4, ee 82%), while the control
experiment in the absence of enzyme showed no significant
decrease of optical purity (Table 2, Entry 5). This demon-
strates that even a single highly selective catalyst is effective,
albeit at a slow rate.
R CHOH or R CHNHR groups, or in general for substrates
2
2
3
in which an sp center can be enzymatically transformed to
2
an sp center in a reversible fashion. Examples include
redox enzymes (alcohol dehydrogenases, a-amino acid dehy-
drogenases, enoate reductases, etc.) and lyases (hydratases,
ammonia lyases, etc.). The feasibility of the concept was suc-
cessfully proven for the racemization of secondary alcohols
by employing alcohol dehydrogenases.
For fast racemization, a catalyst with low stereoselectivity
is required. However, because the primary goal during the
last decades was to identify highly selective ADHs for the
reduction of ketones, little is published about nonselective
enzymes. After testing approximately 60 ADHs, we identi-
fied three candidates that reduce 2-octanone (2a) with
modest or no stereoselectivity. The enzymes of interest were
This approach to biocatalytic racemization is widely appli-
cable and allows the interconversion of enantiomers under
mild (physiological) conditions and thus ensures “clean” and
[20]
efficient isomerization processes.
Due to the recent re-
[17]
[18]
KRED-118, KRED-119
and PF-ADH.
Due to their
ports of applications of alcohol dehydrogenases in organic
[21]
low stereoselectivity in the reduction mode, they were
solvents, it can be expected that this racemization proto-
col can be combined with a lipase to perform a dynamic ki-
netic resolution. Alternatively, the following two processes
are feasible: 1) racemization and resolution can be per-
formed simultaneously in separated vessels that are connect-
ed by a membrane or related techniques or 2) a stepwise
process, in which the racemization and the resolution step
chosen for racemization experiments: (R)-1a was successful-
+
ly racemized in the presence of limiting amounts of NAD /
NADH (Table 2, entries 6–8). These data show that racemi-
zation is also feasible by employing a single “non-racemase”
enzyme that shows imperfect stereoselectivity. A detailed in-
vestigation of the racemization of both enantiomers of 2-oc-
tanol by employing a single ADH (PF-ADH) was carried
out by monitoring the decline of ee over a period of 48 h;
this showed that (R)-1a and (S)-1a were successfully race-
mized (Figure 2). It is worth noting that both graphs are
symmetrical, thus no preference for one enantiomer was ob-
served.
[2]
are performed sequentially in cycles.
Experimental Section
Substrates and reference compounds: Enantiopure compounds 1a, 1d–
1
g were obtained from Alfa Aesar (Lancaster, Frankfurt/Main Germa-
ny), enantiopure 1b, 1c, 1h and ketones 2a–2 f, 2h were purchased from
Sigma–Aldrich. Compounds 1i–1j and 2i–2j were synthesized as previ-
[
6a]
ously described. Reference compound 2g was synthesized by enzymat-
TM
ic oxidation of 1g by employing whole cells of E. coli Tuner (DE3)/
pET22b+-ADH-“A” with hydrogen transfer by using acetone as a hy-
[
22]
drogen acceptor as previously described.
Enzymes: RE-ADH (#68482, alcohol dehydrogenase from Rhodococcus
erythropolis) and LK-ADH (#05643, alcohol dehydrogenase from Lacto-
bacillus kefir) were obtained from Fluka. KRED ketoreductases/alcohol
[17]
dehydrogenases were obtained from BioCatalytics Inc.
Lyophilized
ADH from Pseudomonas fluorescens (PF-ADH) was prepared from the
cell-free extract of E. coli that contained the overexpressed enzyme as
[
18]
previously reported. ADH-“A” was commercially available from Bio-
Catalytics Inc. For the experiments described, the following cell-free
TM
preparation was employed: E. coli Tuner (DE3)/pET22b+-ADH-“A”
was grown as previously described in LB-amp medium (250 mL) with ad-
2
+
À1 [13a]
ditional Zn
(
(100 mgL ).
The medium (ODꢀ3) was centrifuged
Jouan KE22i, AK-500.11, 20 min, 8000 rpm, 48C) and the cell debris was
resuspended in buffer (50 mL, 50 mm Tris–HCl pH 7.5). The cells were
disrupted by ultrasonication (Branson, S250D CE, 200W, 5 mm spike,
Figure 2. Racemization of (R)- and (S)-2-octanol (1a) by employing a
single ADH from Pseudomonas fluorescens (PF-ADH) possesing low ste-
reoselectivity.
5
0 mL tubes, 1 s impulse, 2 s pause, amplitude 50%, 16 min, 48C) and
centrifuged (Jouan KE22i, AK-100.21, 20 min, 13000 rpm, 4 8C). The su-
pernatant was transferred to an Erlenmeyer flask (250 mL) and kept at
6
1
58C for 25 min. After centrifugation (Jouan KE22i, AK-100.21, 20 min,
3000 rpm, 4 8C) the supernatant was used for the experiments.
Conclusion
Representative procedures for racemization employing two enzymes
ADH-’A’ and LK-ADH: ADH-“A”, isolated from Rhodococcus ruber
+
A racemization strategy is presented that is based on the re-
versible interconversion of substrate enantiomers via the
corresponding prochiral intermediate by using either a pair
of stereocomplementary biocatalysts, or a single enzyme
that shows low stereoselectivity. Due to the increasing im-
portance of biocatalytic transformations in asymmetric or-
[23]
[22]
(
0.15 U, 5 mg), LK-ADH (0.2 U, 0.5 mg), and a mixture of NAD
+
À1
À1
and NADH (final concentration 0.7 mmolmL NAD and 1.2 mmolmL
NADH) were dissolved in phosphate buffer (50 mm, pH 7.5, total volume
1
mL). The reaction was started by addition of enantiopure (R)-1a
À1
(
2.5 mL, 26 mmolmL , ee >99.9%). After 24 h of shaking (130 rpm) at
3
08C the mixture was extracted with EtOAc (500 mL) and centrifuged to
achieve phase separation.
[19]
ganic synthesis, novel methods are needed. This general
concept is applicable for substrates that bear, for instance,
RE-ADH and LK-ADH: RE-ADH (0.05 U, 2 mg), LK-ADH (0.04 U,
+
0.1 mg), and a mixture of NAD and NADH (final concentration
8274
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 8271 – 8276

Racemization by Biocatalysts
FULL PAPER
À1
+
À1
0
.7 mmolmL NAD and 1.2 mmolmL NADH) were dissolved in phos-
7, 4523–4526; e) J. Paetzold, J.-E. Bäckvall, J. Am. Chem. Soc. 2005,
127, 17620–17621; f) A. Parvulescu, D. De Vos, P. Jacobs, Chem.
Commun. 2005, 5307–5309; g) A.-B. Fransson, L. Boren, O. Pamies,
J.-E. Bäckvall, J. Org. Chem. 2005, 70, 2582–2587; h) L. Veum, L. T.
Kanerva, P. J. Halling, T. Maschmeyer, U. Hanefeld, Adv. Synth.
Catal. 2005, 347, 1015–1021; i) J. Limanto, A. Shafiee, P. N. Devine,
V. Upadhyay, R. A. Desmond, B. R. Foster, D. R. Gauthier, Jr.,
R. A. Reamer, R. P. Volante, J. Org. Chem. 2005, 70, 2372–2375;
j) S. Akai, K. Tanimoto, Y. Kita, Angew. Chem. 2004, 116, 1407–
phate buffer (50 mm, pH 7.5, total volume 1 mL). The reaction was start-
ed by the addition of enantiopure (R)-1a (2.5 mL, 26 mmolmL , ee
99.9%). After 24 h of shaking (130 rpm) at 308C the mixture was ex-
tracted with EtOAc (500 mL) and centrifuged to achieve phase separa-
À1
>
tion.
Representative procedures for racemization employing a single enzyme
+
KRED-118: KRED-118 (0.1 mg) and a mixture of NAD and NADH
final concentration 0.7 mmolmL NADP and 1.2 mmolmL NADPH)
were dissolved in phosphate buffer (50 mm, pH 7.5, total volume 1 mL).
À1
+
À1
(
1
410; Angew. Chem. Int. Ed. 2004, 43, 1407–1410; k) L. Veum, U.
Hanefeld, Tetrahedron: Asymmetry 2004, 15, 3707–3709; l) C. Paicz,
P. Tähtinen, M. Tosa, C. Majdik, F.-D. Irimie, L.-T. Kanerva, Tetra-
hedron 2004, 60, 10533–10540; m) J. A. Chaplin, M. D. Levin, B.
Morgan, N. Farid, J. Li, Z. Zhu, J. McQuaid, L. W. Nicholson, C. A.
Rand, M. J. Burk, Tetrahedron: Asymmetry 2004, 15, 2793–2796;
n) B. Martin-Matute, M. Edin, K. Bogar, J.-E. Bäckvall, Angew.
Chem. 2004, 116, 6697–6701; Angew. Chem. Int. Ed. 2004, 43, 6535–
The reaction was started by the addition of enantiopure (R)-1a (2.5 mL,
À1
2
6 mmolmL , ee >99.9%). After 24 h of shaking (130 rpm) at 308C the
mixture was extracted with EtOAc (500 mL) and centrifuged to achieve
phase separation.
+
PF-ADH: Crude PF-ADH preparation (4 mg) and a mixture of NAD
and NADH (final concentration 0.7 mmolmL NAD and 1.2 mmolmL
À1
+
À1
NADH) were dissolved in phosphate buffer (50 mm, pH 7.5, 1 mL). The
reaction was started by the addition of either enantiopure (S)-1a (2.5 mL,
6
539; o) A. Dijksman, J. M. Elzinga, Y.-X. Li, I. W. C. E. Arends,
R. A. Sheldon, Tetrahedron: Asymmetry 2002, 13, 879–884.
5] a) M.-C. Gutierrez, R. Furstoss, V. Alphand, Adv. Synth. Catal.
À
À1
2
6 mmolmL 1, ee>99.9%) or enantiopure (R)-1a (2.5 mL, 26 mmolmL ,
[
[
ee>99.9%). After incubation (130 rpm) at 308C, the reaction mixture
was extracted with EtOAc (500 mL).
2
005, 347, 1051–1059; b) M. Inagaki, J. Hiratake, T. Nishioka, J.
Oda, J. Am. Chem. Soc. 1991, 113, 9360–9361.
6] a) B. M. Nestl, W. Kroutil, K. Faber, Adv. Synth. Catal. 2006, 348,
2 4
General workup pr ocedure : The organic phase was dried (Na SO ) and
filtered and the alcohol was acetylated by allowing it to react with acetic
anhydride (100 mL) and DMAP (0.5 mg) for 2 h. After aqueous workup,
the organic phase was analyzed by GC on a chiral stationary phase.
8
73–876; b) B. M. Nestl, S. M. Glueck, M. Hall, W. Kroutil, R.
Stuermer, B. Hauer, K. Faber, Eur. J. Org. Chem. 2006, 4573–4577.
7] J. S. M. Samec, J.-E. Bäckvall, P. G. Andersson, P. Brandt, Chem.
Soc. Rev. 2006, 35, 237–248.
[
[
8] According to Eliel et al., racemization is an entropy-driven thermo-
À1
À1
dynamically downhill process: DS=Rln2=1.38 calmol
K ; see:
E. Eliel, S. H. Wilen, M. P. Doyle, in Basic Organic Stereochemistry,
Wiley, New York, 2001, pp. 102–105.
9] M. Kitamura, M. Tokunaga, R. Noyori, J. Am. Chem. Soc. 1993,
Acknowledgements
[
1
15, 144–152.
This study was performed in cooperation between projects financed by
the Austrian Science Fund (FWF) P18537-B03 (W.K.) and K. Faber/
BASF AG (Ludwigshafen). In addition financial support by the FWF, the
FFG, the City of Graz and the Province of Styria is gratefully acknowl-
edged. BioCatalytics Inc. (http://www.biocatalytics.com) kindly provided
recombinant alcohol dehydrogenases (KRED kit).
[
10] This strategy is also applicable with the same prerequisites for chiral
metal/chemocatalysts; however, for these methods achiral catalysts
are easily accessible, whereas no achiral peptide is available from
nature.
[11] For reviews see: a) K. Nakamura, T. Matsuda, in Enzyme Catalysis
in Organic Synthesis, Vol. 3 (Eds.: K. Drauz, H. Waldmann) Wiley-
VCH, Weinheim, 2002, pp. 991–1047; b) K. Nakamura, T. Matsuda,
T. Harada, Chirality 2002, 14, 703–708; c) R. N. Patel, Curr. Opin.
Biotechnol. 2001, 12, 587–604; d) W. Kroutil, H. Mang, K. Edegger,
K. Faber, Curr. Opin. Chem. Biol. 2004, 8, 120–126; e) M.-R. Kula,
U. Kragl, in Stereoselective Biocatalysis (Ed.: R. N. Patel), Marcel
Dekker, New York, 2000, pp. 839–866.
12] Dehydrogenases were suspected to be involved in the racemization
of alcohols in nature, but this was never proven: E. Adams, Adv.
Enzymol. Relat. Areas Mol. Biol. 1976, 44, 69–138.
13] a) K. Edegger, C. C. Gruber, T. M. Poessl, S. R. Wallner, I. Lavan-
dera, K. Faber, F. Niehaus, J. Eck, R. Oehrlein, A. Hafner, W. Krou-
til, Chem. Commun. 2006, 2402–2404; b) W. Stampfer, B. Kosjek, C.
Moitzi, W. Kroutil, K. Faber, Angew. Chem. 2002, 114, 1056–1059;
Angew. Chem. Int. Ed. 2002, 41, 1014–1017.
[
1] a) A. N. Parvulescu, P. A. Jacobs, D. E. Vos, Chem. Eur. J. 2007, 13,
034–2043; b) B. Martin-Matute, M. Edin; K. Bogar, F. B. Kaynak,
2
J.-E. Bäckvall, J. Am. Chem. Soc. 2005, 127, 8817–8825; c) A. Ber-
kessel, M. L. Sebastian-Ibarz, T. N. Müller, Angew. Chem. 2006, 118,
6
717–6720; Angew. Chem. Int. Ed. 2006, 45, 6567–6570; d) S.
Wuyts, K. De Temmerman, D. E. De Vos, P. A. Jacobs, Chem. Eur. J.
005, 11, 386–397; e) P. Hoyos, M. Fernµndez, J. V. Sinisterra, A. R.
[
2
Alcµntara, J. Org. Chem. 2006, 71, 7632–7637; f) E. J. Ebbers,
G. J. A. Ariaans, J. P. M. Houbiers, A. Bruggink, B. Zwanenburg,
Tetrahedron 1997, 53, 9417–9476.
[
[
2] a) O.-J. Park, S.-H. Lee, T.-Y. Park, W.-G. Chung, S.-W. Lee, Org.
Process Res. Dev. 2006, 10, 588–591; b) S. Wuyts, K. De Temmer-
man, D. E. De Vos, P. A. Jacobs, Chem. Commun. 2003, 1928–1929;
c) B. Larissegger-Schnell, S. M. Glueck, W. Kroutil, K. Faber, Tetra-
hedron 2006, 62, 2912–2916; d) B. Larissegger-Schnell, W. Kroutil,
K. Faber, Synlett 2005, 12, 1936–1938; e) P. Gadler, S. M. Glueck,
W. Kroutil, B. M. Nestl, B. Larissegger-Schnell, B. T. Ueberbacher,
S. R. Wallner, K. Faber, Biochem. Soc. Trans. 2006, 34, 296–300.
3] For reviews see: a) O. Pamies, J.-E. Bäckvall, Trends Biotechnol.
[14] Although the ADH from Lactobacillus kefir (LK-ADH) is usually
used with NADPH (see W. Hummel, Appl. Microbiol. Biotechnol.
1990, 34, 15–19), we noticed in a separate experiment that the
enzyme preparation from Fluka also accepts NADH. Therefore,
LK-ADH was compatible with ADH-“A” and RE-ADH. For mu-
tants that accept NADH see: W. Hummel, B. Riebel (Forschungs-
zentrum Jülich) WO 9947684, 1999 [Chem. Abstr. 1999, 131,
239737].
[
2
1
004, 22, 130–135; b) O. Pamies, J.-E. Bäckvall, Chem. Rev. 2003,
03, 3247–3261; c) H. Pellissier, Tetrahedron 2003, 59, 8291–8327;
d) M. J. Kim, Y. Ahn, J. Park, Curr. Opin. Biotechnol. 2002, 13, 578–
[15] a) C. W. Bradshaw, W. Hummel, C.-H. Wong, J. Org. Chem. 1992,
57, 1532–1536; b) W. Hummel, B. Riebel, Ann. N. Y. Acad. Sci.
1996, 799, 713–716.
5
87; e) B. Schnell, K. Faber, W. Kroutil, Adv. Synth. Catal. 2003,
3
45, 653–666.
[
4] For recent examples see: a) V. Zimmermann, M. Beller, U. Kragl,
Org. Process Res. Dev. 2006, 10, 622–627; b) Y. Asano, S. Yamagu-
chi, J. Am. Chem. Soc. 2005, 127, 7696–7697; c) P. Ödman, L. A.
Wessjohann, U. T. Bornscheuer, J. Org. Chem. 2005, 70, 9551–9555;
d) N. Kim, S.-B. Ko, M. S. Kwon, M.-J. Kim, J. Park, Org. Lett. 2005,
[16] a) T. Taniguchi, K. Ogasawara, Chem. Commun. 1997, 1399–1400;
b) S. Easwar, S. B. Desai, N. P. Argade, K. N. Ganesh, Tetrahedron:
Asymmetry 2002, 13, 1367–1371; c) A. S. Demir, H. Hamamci, P.
Ayhan, A. N. Duygu, A. C. Igdir, D. Capanoglu, Tetrahedron: Asym-
metry 2004, 15, 2579–2582.
Chem. Eur. J. 2007, 13, 8271 – 8276
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8275

W. Kroutil et al.
[
[
17] KRED-118 and KRED-119 were obtained from BioCatalytics Inc.
http://www.biocatalytics.com/.
18] a) P. Hildebrandt, A. Musidlowska, U. T. Bornscheuer, J. Altenbuch-
ner, Appl. Microbiol. Biotechnol. 2002, 59, 483–487; b) P. Hilde-
brandt, T. Riermeier, J. Altenbuchner, U. T. Bornscheuer, Tetrahe-
dron: Asymmetry 2001, 12, 1207–1210.
racemase see: c) G. J. A. Conceicao, P. J. S. Moran, J. A. R. Rodri-
gues, Tetrahedron: Asymmetry 2003, 14, 2327–2330. For lactate race-
mase see: d) J. S. Pepple, D. Dennis, Biochim. Biophys. Acta 1976,
429, 1036–1040.
[21] a) M. M. Musa, K. I. Ziegelmann-Fjeld, C. Vieille, J. G. Zeikus, R. S.
Phillips, Angew. Chem. 2007, 119, 3151–3154; Angew. Chem. Int.
Ed. 2007, 46, 3091–3094; b) G. de Gonzalo, I. Lavandrea, K. Faber,
W. Kroutil, Org. Lett. 2007, 9,2163–2166; c) D. M.-R. De Temino,
W. Hartmeier, M. B. Ansorge-Schumacher, Enzyme Microb. Tech-
nol. 2005, 36, 3 –9.
[
19] a) H. E. Schoemaker, D. Mink, M. G. Wubbolts, Science 2003, 299,
1
694–1697; b) M. Breuer, K. Ditrich, T. Habicher, B. Hauer, M.
Kesseler, R. Stürmer, T. Zelinski, Angew. Chem. 2004, 116, 806–
43; Angew. Chem. Int. Ed. 2004, 43, 788–824.
8
[
20] The enzymatic racemization of hydroxy compounds (such as acetoin,
lactate, and benzoin), which was presumed to proceed through the
catalysis of racemases (whose existence remains still unclear) could
be explained by a redox process that involve dehydrogenases. For
acetoin racemase see: a) M. Voloch, M. R. Ladisch, V. W. Rodwell,
G. T. Tsao, Biotechnol. Bioeng. 1983, 25, 173–183; b) M. B. Taylor,
E. Juni, Biochim. Biophys. Acta 1960, 39, 448–457. For benzoin
[22] K. Edegger, H. Mang, K. Faber, J. Gross, W. Kroutil, J. Mol. Catal.
A 2006, 251, 66–70.
[23] 1 U corresponds to the amount of enzyme that reduces 1 mmol ace-
tophenone per minute at pH 7.0 and 258C to phenylethanol.
Received: April 4, 2007
Published online: July 17, 2007
8276
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 8271 – 8276