One-way biohydrogen transfer for oxidation of sec-alcohols

Article abstract of DOI:10.1021/ol800549f

(Chemical Equation Presented) Quasi-irreversible oxidation of sec-alcohols was achieved via biocatalytic hydrogen transfer reactions using alcohol dehydrogenases employing selected ketones as hydrogen acceptors, which can only be reduced but not oxidized. Thus, only 1 equiv of oxidant was required instead of a large excess. For the oxidation of both isomers of methylcarbinols a single nonstereoselective short-chain dehydrogenase/reductase from Sphingobium yanoikuyae was identified and overexpressed in E. coli.

Full text of DOI:10.1021/ol800549f

ORGANIC
LETTERS
2008
Vol. 10, No. 11
2155-2158
One-Way Biohydrogen Transfer for
Oxidation of sec-Alcohols
Iva´n Lavandera,† Alexander Kern,‡ Verena Resch,† Bianca Ferreira-Silva,§
Anton Glieder,‡ Walter M. F. Fabian,§ Stefaan de Wildeman,| and
Wolfgang Kroutil*,†
Research Centre Applied Biocatalysis c/o Department of Chemistry, Organic and
Bioorganic Chemistry, UniVersity of Graz, Heinrichstrasse 28, 8010 Graz, Austria,
Research Centre Applied Biocatalysis c/o Institute for Molecular Biotechnology, Graz
UniVersity of Technology, Petersgasse 14/2, 8010 Graz, Austria, Department of
Chemistry, Organic and Bioorganic Chemistry, UniVersity of Graz, Heinrichstrasse 28,
8010 Graz, Austria, and DSM Pharmaceutical Products, P.O. Box 18, 6160,
MD Geleen, The Netherlands
wolfgang.kroutil@unigraz.at
Received March 10, 2008
ABSTRACT
Quasi-irreversible oxidation of sec-alcohols was achieved via biocatalytic hydrogen transfer reactions using alcohol dehydrogenases employing
selected ketones as hydrogen acceptors, which can only be reduced but not oxidized. Thus, only 1 equiv of oxidant was required instead of
a large excess. For the oxidation of both isomers of methylcarbinols a single nonstereoselective short-chain dehydrogenase/reductase from
Sphingobium yanoikuyae was identified and overexpressed in E. coli.
Oxidation of alcohols to the corresponding carbonyl com-
pounds belongs to the most fundamental processes in organic
chemistry and many recent efforts are devoted to develop
selective, “green”, and catalytic oxidation methods.1 Chemo-
catalytic oxidations of secondary alcohols based on asym-
metric hydrogen-transfer reactions (with Ru, Rh, and Ir)2,3
or Oppenauer oxidations (with Al, Zr, and lanthanides)4
employing acetone as hydrogen acceptor gained considerable
attention. For these reactions, it is still difficult to shift the
equilibrium toward completion and to suppress undesired side
reactions.3c,5 Biocatalytic oxidation methods are less em-
ployed for secondary alcohols6,7 via hydrogen transfer with
a single biocatalyst.8,9 This approach would represent a very
simple, “green”, and elegant method since only a single
enzymesan alcohol dehydrogenase (ADH)sis required to
perform the whole cycle (Scheme 1).
However, as for the chemical variant, the huge excess of
cosubstrate (ketone, aldehyde) required driving the equilib-
rium to the product side represents a serious drawback.
Furthermore, for an efficient oxidation process both enan-
tiomers of sec-alcohols have to be oxidized to access the
† Research Centre Applied Biocatalysis, University of Graz.
‡ Research Centre Applied Biocatalysis, Graz University of Technology.
§ Department of Chemistry, Organic and Bioorganic Chemistry, Uni-
versity of Graz.
(1) (a) For recent references, see: Arends, I. W. C. E. Catalytic oxidations
in Green Chemistry and Catalysis; Sheldon, R. A., Arends, I. W. C. E.,
Hanefeld, U., Eds.; Wiley-VCH: Weinheim, 2007. (b) Lenoir, D. Angew.
Chem., Int. Ed. 2006, 45, 3206. (c) Modern Oxidation Methods; Ba¨ckvall,
J.-E. Ed.; Wiley-VCH: Weinheim, 2004.
| DSM Pharmaceutical Products.
10.1021/ol800549f CCC: $40.75
Published on Web 05/07/2008
2008 American Chemical Society

ketones to access optical pure alcohols.11 As a first approach,
we thought to search for wild-type micro-organisms that
reduce ketones in a nonstereoselective fashion. However, a
possible hit does not mean that we found a micro-organism
containing the desired right enzyme: due to the large number
of alcohol dehydrogenases in organisms it is very likely that
two enzymes with opposite stereopreference mimic a mi-
crobial catalyst with low stereoselectivity. Subsequently, we
hypothesized that an ADH which can reduce ketones with
two bulky substituents has such a large open active site that
ketones with one smaller group will be reduced with low
stereopreference and consequently also the enantioselectivity
for the reverse reactionsthe oxidationswill be low. Testing
our culture collection, we identified Sphingobium yanoikuyae
DSM 690012 as a possible hit. Thus, employing lyophilized
cells of S. yanoikuyae, hexanophenone (n-pentyl phenyl
ketone) was reduced to the corresponding (S)-alcohol
employing either ethanol or 2-propanol as hydrogen donor
with good ee [97 (S) and 89 (S), respectively]. To identify
the involved enzyme, a genomic library of this organism in
E. coli was screened for the oxidation of 2-propanol
employing a fluorescence assay for NAD(P)H formation.13
A short chain dehydrogenase (annotated arbitrary SyADH)
with 262 amino acids and with a preference for NADPH
was identified, cloned, and overexpressed in E. coli. The
overexpressed SyADH showed blue fluorescence, as it was
recently described for an ADH from Vibrio Vulnificus.14 As
a confirmation to have identified the “right” enzyme,
hexanophenone was reduced employing the overexpressed
catalyst with good ee (96%) comparable with the wild-type
micro-organism. Testing less hindered substrates like 2-oc-
tanone 1a (see the Supporting Information), the enzyme
showed good activity and to our delight the desired low
stereoselectivity for the reduction, leading to alcohols in
almost racemic form confirming our hypothesis.
According to these results, this enzyme appeared to be a
perfect candidate to allow complete oxidation of carbinols
with at least one small alkyl-group. Indeed, employing
acetone as hydrogen acceptor in large excess (10 equiv), rac-
2-octanol 1b could be oxidized efficiently with 99% conver-
sion within 24 h. In a following step, alternative hydrogen
acceptors (oxidants) were tested. In addition to acetone,
4-methyl-2-pentanone (methyl isobutyl ketone, MIBK), ac-
etaldehyde, and even pyruvate were accepted, emphasizing
the high flexibility of this enzyme to accept different types
of cosubstrates in the active site. To our delight, the stability
of the catalyst allowed the substrate concentration to be
increased up to 400 g L-1 (Figure 1). Over a broad substrate
concentration range (40 g L-1 to 400 g L-1) the space time
yield did not show significant deviations with acetone;
Scheme 1. Different Modes of Biocatalytic Hydrogen Transfer
corresponding ketone in high yield. Due to the intrinsic
chirality of enzymes and therefore their high enantioselec-
tivity a kinetic resolution will occur in general, delaying or
even disabling the oxidation of the second isomer.10 There-
fore, we faced two challenges: (i) finding a hydrogen transfer
oxidation method favoring product formation and (ii) a single
nonstereospecific enzyme, which oxidizes the substrate and
also recycles the cofactor, thus working in a coupled substrate
approach.
Our first aim was to identify an alcohol dehydrogenase
which can be used for oxidation via hydrogen transfer
showing diminished enantiopreference. This is in sharp
contrast to all other studies performed for the reduction of
(2) Some recent examples: (a) Li, Y.-Y.; Zhang, X.-Q.; Dong, Z.-R.;
Shen, W.-Y.; Chen, G.; Gao, J.-X. Org. Lett. 2006, 8, 5565. (b) Faller,
J. W.; Lavoie, A. R. Org. Lett. 2001, 3, 3703. (c) Persson, B. A.; Larsson,
A. L. E.; Le Ray, M.; Ba¨ckvall, J.-E. J. Am. Chem. Soc. 1999, 121, 1645.
(3) For reviews, see: (a) Wu, X.; Xiao, J. Chem. Commun. 2007, 2449.
(b) Gladiali, S.; Alberico, E. Chem. Soc. ReV. 2006, 35, 226. (c) Noyori,
R.; Ohkuma, T. Angew. Chem., Int. Ed. 2001, 40, 40.
(4) For a review, see: de Graauw, C. F.; Peters, J. A.; van Bekkum, H.
Synthesis 1994, 1007.
(5) Trichloroacetaldehyde has been used as quasi-stoichiometric oxidant
in a zirconium-catalyzed Oppenauer oxidation: Krohn, K.; Knauer, B.;
Ku¨pke, J.; Seebach, D.; Beck, A. K.; Hayakawa, M. Synthesis 1996, 1341.
(6) Some recent reviews: (a) Molinari, F. Curr. Org. Chem. 2006, 10,
1247. (b) Kroutil, W.; Mang, H.; Edegger, K.; Faber, K. AdV. Synth. Catal.
2004, 346, 125. (c) Kroutil, W.; Mang, H.; Edegger, K.; Faber, K. Curr.
Opin. Chem. Biol. 2004, 8, 120. (d) Thomas, S. M.; DiCosimo, R.;
Nagarajan, V Trends Biotechnol. 2002, 20, 238. (e) Nakamura, K.; Matsuda,
T.; Harada, T. Chirality 2002, 14, 703.
(7) For biocatalytic oxidation coupling an ADH with a NAD(P)H
oxidase, see: (a) Riebel, B. R.; Gibbs, P. R.; Wellborn, W. B.; Bommarius,
A. S. AdV. Synth. Catal. 2003, 345, 707. (b) Geueke, B.; Riebel, B.;
Hummel, W. Enzyme Microb. Technol. 2003, 32, 205.
(8) (a) Edegger, K.; Mang, H.; Faber, K.; Gross, J.; Kroutil, W. J. Mol.
Catal. A: Chem. 2006, 251, 66. (b) Edegger, K.; Stampfer, W.; Seisser, B.;
Faber, K.; Mayer, S. F.; Oehrlein, R.; Hafner, A.; Kroutil, W. Eur. J. Org.
Chem. 2006, 1904. (c) Kosjek, B.; Stampfer, W.; van Deursen, R.; Faber,
K.; Kroutil, W. Tetrahedron 2003, 59, 9517. (d) Gorrebeeck, C.; Spanoghe,
M.; Lanens, D.; Lemie`re, G. L.; Dommisse, R. A.; Lepoivre, J. A.;
Alderweireldt, F. C. Recl. TraV. Chim. Pays-Bas 1991, 110, 231. (e) Snijder-
Lambers, A. M.; Vulfson, E. N.; Dodema, H. J. Recl. TraV. Chim. Pays-
Bas 1991, 110, 226. (f) Lemie`re, G. L.; Lepoivre, J. A.; Alderweireldt,
F. C. Tetrahedron Lett. 1985, 26, 4527.
(11) Recent references: (a) Buchholz, S.; Gro¨ger, H. In Biocatalysis in
the Pharmaceutical and Biotechnology Industry; Patel, R. N., Ed.; CRC
Press: Boca Raton, 2007; p 757. (b) de Wildeman, S. M. A.; Sonke, T.;
Schoemaker, H. E.; May, O Acc. Chem. Res. 2007, 40, 1260. (c) Goldberg,
K.; Schroer, K.; Lu¨tz, S.; Liese, A. Appl. Microbiol. Biotechnol. 2007, 76,
237. (d) Goldberg, K.; Schroer, K.; Lu¨tz, S.; Liese, A. Appl. Microbiol.
Biotechnol. 2007, 76, 249.
(9) For employing two enzymes and acetone as hydrogen acceptor see:
Fossati, E.; Polentini, F.; Carrea, G.; Riva, S. Biotechnol. Bioeng. 2006,
93, 1216.
(12) DSM number refers to strains commercially available from DSMZ
(Deutsche Sammlung von Mikroorganismen und Zellkulturen, German
Collection of Microorganisms and Cell Cultures, http://www.dsmz.de/).
(13) Reisinger, C.; van Assema, F.; Schu¨rmann, M.; Hussain, Z.; Remler,
P.; Schwab, H. J. Mol. Catal. B: Enzym. 2006, 39, 149.
(10) Performing the oxidation with two enzymes showing opposite
stereo-preference is of course a possibility to overcome this limitation but
cannot be considered as an elegant method. Furthermore, the corresponding
enzymes with opposite enantio-preference are not always accessible.
(14) Polizzi, K. M.; Moore, D. A.; Bommarius, A. S. Chem. Commun.
2007, 1843.
2156
Org. Lett., Vol. 10, No. 11, 2008

Figure 2. Time course of E. coli/SyADH-catalyzed oxidation of
Figure 1. Dependence of space time yield (STY) on varied substrate
rac-2-octanol 1b (30 g L-1) employing 1.5 equiv of oxidant [acetone
([); MIBK (0); chloroacetone (2)].
concentration (rac-2-octanol, 1b) using as hydrogen acceptor (a)
acetone (15% v v-1, [); (b) MIBK (10% v v-1, 0).
Rhodococcus ruber16 was employed in a kinetic resolution
of rac-1b. Varying the substrate concentration up to 100 g
L-1 just 0.6 equiv of chloroacetone as oxidant was required
to reach full conversion for the prefered (S)-enatiomer (Figure
3) within 24 h, thus leaving enantiopure (R)-1b behind.
additionally, MIBK proved to be rather efficient in a two-
phase system at higher substrate concentration as well.
When investigating the scope of substrates which can be
oxidized by SyADH, we experienced a surprise. Halohydrins
like 1-chloro-2-octanol or 2-chloro-1-phenylethanol were not
oxidized at all. First, we expected that this was a curiosity
of this special enzyme, therefore we tested approximately
60 commercial alcohol dehydrogenases (e.g., ADHs from
Lactobacillus kefir and breVis, ADH-‘A’ from Rhodococcus
ruber, or screening kits from Biocatalytics/Codexis) for the
oxidation of halohydrins with an excess of the corresponding
cofactor. Again for all 60 commercial ADHs no oxidation-
products were detectable via GC, thus obviously chlorohy-
drins cannot be oxidized by ADHs. As a consequence, we
concluded that although R-chloro ketones would be reduced
to 1-chloro-2-alcohols, the latter would not be oxidized
backward anymore. Therefore, if chloroacetone15 is used as
an oxidant instead of acetone the system should become
quasi-irreversible (Scheme 1). Comparing the oxidation of
rac-2-octanol 1b with 1.5 equiv of various oxidants (chlo-
roacetone, MIBK, acetone), it became clear that although
the oxidation with chloroacetone was slower at the beginning
compared to acetone, the equilibrium was shifted toward the
product side employing chloroacetone, while the equilibrium
for acetone was at approximately 50% conversion after 24 h
(Figure 2).
Figure 3. Kinetic resolution of rac-2-octanol 1b at varied substrate
concentrations via quasi-irreversible oxidative hydrogen transfer
employing E. coli/ADH-‘A’ and 0.6 equiv of chloro acetone (t )
24 h).
Consequently, the question was raised as to which other
oxidants might be used and what is the reason for this quasi-
irreversibility. As a first approach, the time course of the
oxidation of various 2-propanol derivatives bearing different
substituents (1-fluoro-2-propanol, 1,3-difluoro-2-propanol,
1,1,1,3,3,3-hexafluoro-2-propanol, 1,3-dichloro-2-propanol,
and 1-bromo-2-propanol) was followed by a UV assay for
the formation of NADPH from NADP+ employing SyADH
and compared with 2-propanol and 1-chloro-2-propanol.17
From all these compounds only 2-propanol led to the
To show the applicability of this methodology, several
racemic aliphatic and aromatic secondary alcohols (2b-7b)
were oxidized to the corresponding ketones at elevated
substrate concentrations (30 g L-1) using just 1.5 equivalents
of chloroacetone (Table 1) as hydrogen acceptor. In case of
more bulky alcohols like 7b lower conversions were obtained
since in this case a kinetic resolution occurs due to the higher
enantioselectivity of the enzyme for such type of substrates.
Performing a preparative oxidation of rac-2-octanol (0.2
g) with just 1.1 equiv of chloroacetone and lyophilized cells
of E. coli/SyADH (0.2 g) at 30 °C, complete conversion
(>99%) was achieved within 24 h.
To show the scope of the biocatalytic one-way concept
with another ADH, the enantioselective ADH-‘A’ from
(15) 100 g of chloroacetone costs 14 euros at Sigma-Aldrich (Austria).
(16) This (S)-selective enzyme has a strong preference for NADH/
NAD+. See: Edegger, K.; Gruber, C. C.; Poessl, T. M.; Wallner, S. R.;
Lavandera, I.; Faber, K.; Niehaus, F.; Eck, J.; Oehrlein, R.; Hafner, A.;
Kroutil, W. Chem. Commun. 2006, 2402.
(17) Pure compounds had to be synthesized, since the commercial
samples had only technical grade.
Org. Lett., Vol. 10, No. 11, 2008
2157

genation of all halo ketones led to products which are about
4-6 kcal mol-1 more favored20 than 2-propanol, which is a
clear explanation for the quasi-irreversibility of the observed
reactions.21
Table 1. Quasi-irreversible Oxidation (t ) 24 h) of
rac-sec-Alcohols (30 g L-1) Catalyzed by Lyophilized E.
coli/SyADH Cells Employing Chloroacetone as Hydrogen
Acceptor (1.5 equiv)
In summary, we have identified an alcohol dehydrogenase
accepting all common oxidants (acetone, acetaldehyde,
MIBK, pyruvate). Due to its low stereoselectitivy for
methylcarbinols, both enantiomers could be oxidized at
reasonable rate. Furthermore, we could show that oxidation
via hydrogen transfer does not necessarily need a huge excess
of a hydrogen acceptor: By employing activated ketones, like
chloroacetone,14 1,1- and 1,3-dichloroacetone, 1,1,1-trichlo-
roacetone, or ethyl acetoacetate, to name just a few pos-
sibilities, only stoichiometric amounts of oxidant were
required. This leads to a dramatically reduced amount of
organic reagents needed. Due to the increasing importance
of biocatalytic transformations in organic synthesis,22 it is
expected that oxidation methods employing enzymes will
gain increasing significance especially, since biocatalysts are
easily accessible, cheap, biodegradable, and show perfect
chemoselectivity.
entry
substrate
convn (%)a
1
2
3
4
5
6
7
rac-1b
rac-2b
rac-3b
rac-4b
rac-5b
rac-6b
rac-7b
>99b
90
>99
95b
97
70b
55b
a Measured by GC. b Using double quantity of catalyst.
formation of NADPH; thus, for all other alcohols no ketone
formation could be detected within the observation time (see
the Supporting Information). It can therefore be assumed,
that all the corresponding ketones represent one-way oxidants
employing alcohol dehydrogenases with NAD(P)+. This was
confirmed experimentally employing SyADH with 1,1- and
1,3-dichloroacetone, 1,1,1-trichloroacetone, and ethyl ac-
etoacetate as stoichiometric oxidants leading to complete
conversion. As a first simplified explanation for the one-
way oxidation, the formation of an additional intramolecular
hydrogen bond18 between the new obtained alcohol moiety
and the electronegative group (F, Cl, Br, CdO19) can be
assumed gaining ∼3-4 kcal mol-1.
Acknowledgment. Financial support by DSM, FFG, and
the Province of Styria is gratefully acknowledged. B.F.-S.
acknowledges the FAPSEP (Fundac¸a˜o de Amparo a Pesquisa
do Estado de Sa˜o Paulo) and the CAPES (Fundac¸a˜o
Coordenac¸a˜o de Aperfeic¸oamento de Pessoal de N´ıvel
Superior).
Supporting Information Available: Gene library con-
struction, cloning and overexpression of SyADH, measure-
ment of kinetic constants, computational details, and analytics
are detailed. This material is available free of charge via the
Internet at http://pubs.acs.org.
Thus, calculating the Gibbs free energies for the hydro-
genation of the ketones 1-fluoro-, 1-chloro-, and 1-bromoac-
etone to the corresponding alcohols and comparing them with
the values for acetone, a clear result was obtained: hydro-
(18) Goldstein, T.; Snow, M. S.; Howard, B. J. J. Mol. Spectrosc. 2006,
236, 1.
OL800549F
(19) In a very recent published patent acetoacetate derivatives were used
as oxidants: Peschko, C.; Stohrer, J. Wacker Chemie AG, Germany, DE
102006009743, A1 20070906, CAN 2007, 147, 321414.
(20) Calculation method used: MP2/cc-pVTZ//MP2/cc-pVDZ.
(21) Eckstein, M. F.; Peters, M.; Lembrecht, J.; Spiess, A. C.; Greiner,
L. AdV. Synth. Catal. 2006, 348, 1591.
(22) (a) Breuer, M.; Ditrich, K.; Habicher, T.; Hauer, B.; Kesseler, M.;
Stu¨rmer, R.; Zelinski, T. Angew. Chem., Int. Ed. 2004, 43, 788. (b)
Schoemaker, H. E.; Mink, D.; Wubbolts, M. G. Science 2003, 299, 1694.
(c) Schmid, A.; Dordick, J. S.; Hauer, B.; Kiener, A.; Wubbolts, M.; Witholt,
B. Nature 2001, 409, 258.
2158
Org. Lett., Vol. 10, No. 11, 2008