Electroenzymatic asymmetric reduction of rac-3-methylcyclo-hexanone to (1S,3S)-3-methyleyclohexanol in organic/aqueous media catalyzed by a thermophilic alcohol dehydrogenase

Article abstract of DOI:10.1002/adsc.200700059

Electrochemical regeneration of nicotinamide cofactors has been discussed as a promising, clean, and sustainable technology since the 1980s. However, most concepts for the coupling of this technology to enzymes suffer from low productivities, insufficient

Full text of DOI:10.1002/adsc.200700059

COMMUNICATIONS
DOI: 10.1002/adsc.200700059
Electroenzymatic Asymmetric Reduction of rac-3-Methylcyclo-
hexanone to (1S,3S)-3-Methylcyclohexanol in Organic/Aqueous
Media Catalyzed by a Thermophilic Alcohol Dehydrogenase
a
a
a,b,
Volker Hçllrigl, Katja Otto, and Andreas Schmid *
a
Chair of Chemical Biotechnology, Department of Biochemical and Chemical Engineering, University of Dortmund,
Emil-Figge-Strasse 66, 44221 Dortmund, Germany
Phone: (+49)-231-755-7380; fax: (+49)-231-755-7382; e-mail: andreas.schmid@bci.uni-dortmund.de
Institute for Analytical Sciences (ISAS), Bunsen-Kirchhoff-Strasse 11, 44139 Dortmund, Germany
b
Received: January 24, 2007
Supporting information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
À1 À1 [2]
Abstract: Electrochemical regeneration of nicotin-
amide cofactors has been discussed as a promising,
clean, and sustainable technology since the 1980s.
However, most concepts for the coupling of this
technology to enzymes suffer from low productivi-
ties, insufficient stabilities, or are difficult to scale
up. We have developed an electrochemical cell for
the efficient regeneration of NAD(P)H, which can
be coupled to a reduction reaction catalyzed by the
thermophilic alcohol dehydrogenase from Thermus
sp. Octane, as second organic phase avoided prod-
uct inhibition and allowed for the production of
(41.6 gL h ). However, the production of chiral
compounds is fairly difficult if only electrochemical
means are employed as no intrinsic stereoselective se-
[
3]
lection pressure is present. Due to their high stereo-
selectivity, broad versatility, and high specificity, oxi-
doreductases, like dehydrogenases have a high poten-
tial for efficient, asymmetric synthesis reactions. The
main drawback for industrial application of these en-
zymes is their dependence on reduction equivalents
like NAD(P)H in stoichiometric amounts, which
makes the introduction of regeneration systems neces-
sary. Thus, a combination of electrochemical cofac-
tor regeneration coupled to cofactor-dependent enzy-
matic catalysis should reveal strong synergistic effects.
However, many problems arise at the interface be-
tween the biological and the electrochemical system.
Enzymes often get inactivated at the electrode sur-
face, the rate of the cofactor regeneration is too low,
compared to the reaction rate of the biocatalyst, or
the coupling efficiency is insufficient. This became
evident in a recent survey by Ruinatscha et al., in
which electroenzymatic processes were evaluated
with regard to their application potential. Out of
thirteen reviewed reactions, only three showed suffi-
cient process performance to be of industrial rele-
vance, while the others displayed insufficient produc-
tivity or process stability.
[4]
[
5]
(
1S,3S)-3-methylcyclohexanol at a diastereomeric
excess of 96% from the corresponding racemic
À1 À1
ketone with a productivity of 0.13 gL h and a
current efficiency of 85%. After 10 h, the experi-
ment was actively terminated and the final product
À1
concentration reached was 1.32 gL . In our opin-
ion this concept defines a new state of the art in
electroenzymology and provides a strong basis for
applications in organic synthesis.
[
6]
Keywords: alcohol dehydrogenase; electrochemis-
try; enantioselectivity; enzyme catalysis; in situ
product extraction; reactor design
Here, we describe the electroenzymatic production
of enantiomerically-pure (1S,3S)-3-methylcyclohexa-
Electroenzymology, the combination of electrochem- nol [(1S,3S)-2] from rac-3-methylcyclohexanone (rac-
istry and biocatalysis, is a highly interesting alterna- 1) as outlined in Figure 1, which fulfills the minimal
tive for the production of chiral compounds and has requirements for a preparative process, as defined by
been discussed as a promising and sustainable tech- Ruinatscha et al. The thermophilic NAD-dependent
[
1]
nology since the 1980s. Electrochemistry provides a alcohol dehydrogenase from Thermus sp. (TADH)
cheapand clean source of reduction equivalents over
a very broad range of reaction conditions and has al- was regenerated via [Cp*Rh
served as a biocatalyst. The reduced cofactor NADH
2+
A
H
R
U
G
A
H
R
U
ready been used for the production of enzymatically lective electrochemical mediator for the regeneration
[
2,7,8]
active
NAD(P)H
at
high
productivities of nicotinamide cofactors.
This prevented unspe-
Adv. Synth. Catal. 2007, 349, 1337 – 1340
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1337

Volker Hçllrigl et al.
COMMUNICATIONS
Figure 1. Reaction scheme of the electroenzymatic reduction reaction.
+
cific reduction of NAD at the electrode, which dic compartment filled with 100 mM Bis-Tris buffer,
would have lead to enzymatically inactive cofactor pH 6 at 608C, which provides optimal reaction condi-
[
9]
[11]
species. For the reaction system, an electrochemical tions for TADH-catalyzed reductions. In addition,
cell was designed as shown in Figure 2. The cathodic the decomposition rate of NADH in the presence of
2+
and anodic compartments were separated by a Nafion [Cp*Rh
110 anion exchanger membrane. Both electrodes buffer (0.0137 mMh ) than in other buffer systems
were made of reticulated vitreous carbon, a glassy for example, sodium phosphate buffer (0.024 mMh ).
carbon-like material, which provides a very large sur- Furthermore, by using Bis-Tris buffer interactions be-
face to volume ratio of 66 cm cm , resulting in an tween the enzyme and [Cp*Rh
active surface area of 3300 cm available for the elec- prevented so that no loss of enzyme activity was de-
trochemical reaction. The remaining void volume is tected over a period of 7 h (data not shown).
A
H
R
U
G
A
H
R
U
À1
1
À1
2
À3
2+
A
H
R
U
G
A
H
R
U
G
2
almost 97%, which makes this material very suitable
For system investigations, initially, 40 mM rac-1
for scale-up, as no significant backpressure is generat- were applied as substrate (Table 1). The maximal pro-
[
10]
ed. The biotransformation took place in the catho-
Table 1. Summary of electroenzymatic conversions.
[
a]
rac-1 Product
A
H
R
U
(1S,3S)- Time de
Productivity Current
À1 À1
[mM] extraction 2 [mM] [h]
[%] [gL h ]
yield
%]
[
4
1
1
0
-
-
3.7
3.2
11.6
6.3
5.85 96
10.4 96
96
0.07
0.06
0.13
74
67
85
0 [
b]
[c]
[b]
[b]
50
octane
[
a]
Determined via GC. For further details see Supporting In-
formation.
Based on organic phase.
[
[
b]
c]
Ratio octane:aqueous phase of 1:1.5 was applied.
ductivity for the formation of (1S,3S)-2 was achieved
À1 À1
within the first hours reaching 0.27 gL h , while at
a product concentration of 3.7 mM the rate decreased
and after 6.3 h no further product was formed and
À1 À1
thus an overall productivity of 0.07 gL h
was
reached. The product was almost entirely the desired
1S,3S)-2 at a diastereomeric excess (de) of 96% (see
Supporting Information) with a current efficiency of
(
Figure 2. Set-upof the electroenzymatic reactor. CE: counter
electrode, WE: working electrode, RE: reference electrode. 74%. With 10 mM rac-1 as substrate under otherwise
1338
asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 1337 – 1340

Electroenzymatic Asymmetric Reduction of rac-3-Methylcyclohexanone
COMMUNICATIONS
similar conditions, almost the same final product con- its adsorption to the Nafion membrane. This was al-
centration of 3.2 mM was reached (Table 1 and Sup- ready noticed in control experiments without enzyme,
porting Information). From these findings it can be where only cofactor regeneration was investigated.
concluded that TADH exhibits product inhibition Here, the initial high NADH production rate of
À1
with respect to 2.
13 mMh could not be maintained throughout the
In order to prevent product inhibition and to reach entire electrolysis, as there was a loss of regeneration
higher final product concentrations, octane was used power. This was also indicated by a drop in catalytic
as second phase for in situ product extraction. This current, which could be partly restored by adding rho-
concept has already been applied in electrochemistry dium mediator to the reaction solution. Currently in-
[
12]
to prevent the formation of by-products. The parti- vestigations are on the way to elucidate this loss of re-
tioning coefficients Kp [defined as (octane)/c- generation power. A solution could be the application
(aqueous)] for substrate and product were determined of a different membrane material for the separation
c
ACHTREUNG
ACHTREUNG
[
11]
previously (K =13 for rac-1 and 10 for rac-2).
of the anodic and cathodic compartments to avoid ad-
p
Except for the addition of octane as organic phase to sorption of the mediator.
the cathode side (1:1.5 octane:aqueous) the two-
In summary, a concept for a scalable reactor to
liquid phase reactions were performed as described couple electrochemical cofactor regeneration effi-
above. Under these conditions, the overall productivi- ciently to the enzymatic production of chiral alcohols
À1 À1
ty was 0.13 gL h (based on organic phase) reach- with in situ product extraction was developed. The
ing a final product concentration of 11.6 mM in the process performance of this system is already in an in-
octane phase (de=96%) with a current efficiency of teresting range even though the potential is not yet
85% (Figure 3 and Table 1). After 4.5 h the product fully exploited. Overall, the key advantage of this
formation rate as well as the regeneration rate of the hybrid approach is its flexibility towards reaction pa-
NADH declined. After addition of the rhodium medi- rameters, organic and other solvents, and its scalabili-
ator, the rates recovered, although the initial high ty. This characterizes it as a suitable platform technol-
values were not reached again. An explanation for ogy for the application of a wide range of oxidoreduc-
2+
this apparent deactivation of [Cp*Rh
A
H
R
U
G
A
H
R
U
G
Experimental Section
Unless otherwise stated, all chemicals were purchased from
Sigma Aldrich (Buchs, Switzerland) at the highest grade
available and were used without further purification.
2+
[
Cp*Rh
A
T
E
N
A
H
R
U
G
[
7,13]
ously.
Biocatalyst Preparation
TADH was recombinantly produced in E. coli BL21 (DE3)
[11]
pLysS pASZ2 in a KLF2000 bioreactor (Bioengineering,
[
14]
Wald, Switzerland) using 1 L Terrific Broth
as medium
À1
supplemented with ampicillin (100 mgmL ) and chloram-
phenicol (20 mgmL ). The cells were grown to a cell dry
weight of 0.5 gL and induced with 0.5 mM isopropyl b-d-
À1
À1
1
-thiogalactopyranoside. After 6 h cells were harvested by
centrifugation at 8200”g, 48C, and 20 min in a Sorvall RC-
B centrifuge (Thermo Electron Corporation, Langensel-
5
bold, Germany). The cell pellet was resuspended in 20 mM
Tris pH 7.5 and directly heat treated at 758C for 20 min. In-
soluble fraction and precipitated proteins were separated by
ultracentrifugation in a Sorvall Discovery 90SE centrifuge
(Kendro Laboratory Products GmbH, Langenselbold, Ger-
many) at 91500”g, 48C, 1 h and the supernatant was applied
to a Fractogel EMD TMAE (Merck KGaA, Darmstadt,
Germany) anion exchanger column equilibrated with
20 mM Tris, pH 7.5. TADH was eluted with a linear gradient
of 20 mM Tris, pH 7.5 and 2M NaCl. Fractions containing
TADH were pooled, sterile filtered and stored at À808C
until use.
Figure 3.
A
C
H
T
R
E
U
N
G
(1S,3S)-2 formation during electroenzymatic reduc-
tion of rac-1. General conditions: 100 mM Bis-Tris, pH608C
6
.0, V =150 mL, V
=100 mL, T=608C, flow rate=
octane
2+ +
aq
À1
6.75 mLmin , c
A
C
H
T
R
E
U
N
G
([Cp*Rh
A
H
R
N
(bpy)
A
H
R
U
G
ACHTREUNG
2
=
0.5 mM, c
A
C
H
T
R
E
U
N
G
(rac-1)
=150 mM. (*) c
ACHTREUNG
octane
(
start of the electrolysis by application of À800 mV vs. Ag/
AgCl. Dotted arrow indicates the addition of 50 mM
2+
[Cp*Rh
A
C
H
T
R
E
U
N
G
(bpy)(H O)] .
A
C
H
T
R
E
U
N
G
2
Adv. Synth. Catal. 2007, 349, 1337 – 1340
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1339

Volker Hçllrigl et al.
COMMUNICATIONS
Electrochemical Cell
References
The electrochemical cell was made entirely of glass. The
anodic and cathodic compartments (50 mL each) were sepa-
rated by means of a Nafion 1110 membrane (DuPont de
Nemours, Bad Homburg, Germany). Both electrodes consist
of reticulated vitreous carbon (ERG Materials and Aero-
space Corporation, Oakland, USA) at a porosity of 100 ppi.
[
1] A. Schmid, F. Hollmann, B. Bühler, in: Enzyme Cataly-
sis in Organic Synthesis 2nd edn., Vol. III, (Eds.: K.
Drauz, H. Waldmann), Wiley-VCH, Weinheim, Germa-
ny, 2002, pp. 1108–1170.
[
[
2] K. Vuorilehto, S. Lütz, C. Wandrey, Bioelectrochemistry
2
004, 65, 1–7.
3] T. Nonaka, T. Fuchigami, in: Organic Electrochemistry,
th edn., (Eds.: H. Lund, O. Hammerich), Marcel
100 mM Bis-Tris, pH608C 6.0 and 200 mM Na-phosphate,
pH608C 7.0 were used for the cathodic and anodic compart-
ments, respectively. The cell was constructed as a three-elec-
trode set-upwith an RE-5B Ag/AgCl reference electrode
4
Dekker, New York, 2001, pp. 1051–1102.
[
4] Z. Li, J. B. van Beilen, W. A. Duetz, A. Schmid, A. de
(
Bioanalytical Systems Inc., W. Lafayette, USA). The reac-
tion was controlled by an Autolab PGSTAT302 potentiostat
Deutsche Metrohm GmbH & Co. KG, Filderstadt, Germa-
Raadt, H. Griengl, B. Witholt, Curr. Opin. Chem. Biol.
2
002, 6, 136–144; A. J. Straathof, S. Panke, A. Schmid,
(
Curr. Opin. Biotechnol. 2002, 13, 548–556.
5] W. Hummel, Trends. Biotechnol. 1999, 17, 487–492.
6] R. Ruinatscha, V. Hçllrigl, K. Otto, A. Schmid, Adv.
Synth. Catal. 2006, 348, 2015–2026.
ny) to maintain the cathodic potential at À800 mV vs. Ag/
AgCl. Both cathodic and anodic storage vessels were de-
gassed with nitrogen and heated to 608C. For in situ product
extraction octane was used as second organic phase at a
ratio of 1:1.5 with respect to the aqueous phase.
[
[
[
[
[
7] F. Hollmann, B. Witholt, A. Schmid, J. Mol. Catal. B:
Enzym. 2002, 19, 167–176.
8] E. Steckhan, S. Herrmann, R. Ruppert, E. Dietz, M.
Frede, E. Spika, Organometallics 1991, 10, 1568–1577.
9] H. Jaegfeldt, Bioelectrochem. Bioenerg. 1981, 8, 355–
370.
Analysis
Protein concentration was determined using the Bradford
[15]
assay
(Biorad) with a standard curve made from BSA.
[
10] J. M. Friedrich, C. Ponce-De-Leon, G. W. Reade, F. C.
Walsh, J. Electroanal. Chem. 2004, 561, 203–217; J.
Wang, Electrochimica Acta 1981, 26, 1721–1726.
The course of electroenzymatic reductions was followed by
analyzing samples via GC on a chiral Rt-bDex-sm (Restek)
column.
[
[
11] F. Hollmann, A. Kleeb, K. Otto, A. Schmid, Tetrahe-
dron: Asymmetry 2005, 16, 3512–3519.
12] a) P. Krishnan, V. G. Gurjar, J. Appl. Electrochem.
1
995, 25, 792–796; b) T. Raju, K. Kulangiappar, M. A.
Acknowledgements
Kulandainathan, G. K. Shankar, A. Muthukumaran,
Electrochim. Acta 2005, 51, 356–360; c) L. D. Pachꢁn,
C. J. Elsevier, G. Rothenberg, Adv. Synth. Catal. 2006,
348, 1705–1710; d) T. Raju, K. Kulangiappar, M. A.
Kulandainathan, U. Uma, R. Malini, A. Muthukumar-
an, Tetrahedron Lett. 2006, 47, 4581–4584;
The authors would like to acknowledge Dr. G. Weber (Insi-
tute of Analytical Sciences, ISAS, Dortmund), Dr. K.-M.
Mangold (Karl-Winnacker-Institut der DECHEMA e.V.,
Frankfurt), and Prof. J. Jçrissen (Department of Biochemical
and Chemical Engineering, University of Dortmund, Dort-
mund) for valuable discussions and their support in designing
the electrochemical cell. We would like to thank T. Olma for
supplying the Nafion membrane and T. Schmitges for practi-
cal assistance. This project was financially supported by the
Deutsche Bundesstiftung Umwelt (ICBio Az: 13123–32).
[13] U. Kçlle, B. S. Kang, P. Infelta, P. Comte, M. Grꢂtzel,
Chem. Ber. 1989, 122, 1869–1880.
[14] J. Sambrook, D. W. Russell, Molecular Cloning: A Lab-
oratory Manual, 3rd edn., Cold Spring Laboratory
Press, New York, 2001.
[15] M. M. Bradford, Anal. Biochem., 1976, 72, 248–254.
1340
asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 1337 – 1340