Screening, Molecular Cloning, and Biochemical Characterization of an Alcohol Dehydrogenase from Pichia pastoris Useful for the Kinetic Resolution of a Racemic β-Hydroxy-β-trifluoromethyl Ketone

Article abstract of DOI:10.1002/cbic.201600101

The stereoselective synthesis of chiral 1,3-diols with the aid of biocatalysts is an attractive tool in organic chemistry. Besides the reduction of diketones, an alternative approach consists of the stereoselective reduction of β-hydroxy ketones (aldols). Thus, we screened for an alcohol dehydrogenase (ADH) that would selectively reduce a β-hydroxy-β-trifluoromethyl ketone. One potential starting material for this process is readily available by aldol addition of acetone to 2,2,2-trifluoroacetophenone. Over 200 strains were screened, and only a few yeast strains showed stereoselective reduction activities. The enzyme responsible for the reduction of the β-hydroxy-β-trifluoromethyl ketone was identified after purification and subsequent MALDI-TOF mass spectrometric analysis. As a result, a new NADP⁺-dependent ADH from Pichia pastoris (PPADH) was identified and confirmed to be capable of stereospecific and diastereoselective reduction of the β-hydroxy-β-trifluoromethyl ketone to its corresponding 1,3-diol. The gene encoding PPADH was cloned and heterologously expressed in Escherichia coli BL21(DE3). To determine the influence of an N- or C-terminal His-tag fusion, three different recombinant plasmids were constructed. Interestingly, the variant with the N-terminal His-tag showed the highest activity; consequently, this variant was purified and characterized. Kinetic parameters and the dependency of activity on pH and temperature were determined. PPADH shows a substrate preference for the reduction of linear and branched aliphatic aldehydes. Surprisingly, the enzyme shows no comparable activity towards ketones other than the β-hydroxy-β-trifluoromethyl ketone.

Full text of DOI:10.1002/cbic.201600101

DOI: 10.1002/cbic.201600101
Full Papers
Screening, Molecular Cloning, and Biochemical
Characterization of an Alcohol Dehydrogenase from Pichia
pastoris Useful for the Kinetic Resolution of a Racemic
b-Hydroxy-b-trifluoromethyl Ketone
[
a, b]
[c]
[b]
[c]
Dalia Bulut,
Nongnaphat Duangdee, Harald Grçger, Albrecht Berkessel, and
[a, b]
Werner Hummel*
The stereoselective synthesis of chiral 1,3-diols with the aid of
biocatalysts is an attractive tool in organic chemistry. Besides
the reduction of diketones, an alternative approach consists of
the stereoselective reduction of b-hydroxy ketones (aldols).
Thus, we screened for an alcohol dehydrogenase (ADH) that
would selectively reduce a b-hydroxy-b-trifluoromethyl ketone.
One potential starting material for this process is readily avail-
able by aldol addition of acetone to 2,2,2-trifluoroacetophe-
none. Over 200 strains were screened, and only a few yeast
strains showed stereoselective reduction activities. The enzyme
responsible for the reduction of the b-hydroxy-b-trifluorometh-
yl ketone was identified after purification and subsequent
MALDI-TOF mass spectrometric analysis. As a result, a new
identified and confirmed to be capable of stereospecific and
diastereoselective reduction of the b-hydroxy-b-trifluoromethyl
ketone to its corresponding 1,3-diol. The gene encoding
PPADH was cloned and heterologously expressed in Escherichia
coli BL21(DE3). To determine the influence of an N- or C-termi-
nal His-tag fusion, three different recombinant plasmids were
constructed. Interestingly, the variant with the N-terminal His-
tag showed the highest activity; consequently, this variant was
purified and characterized. Kinetic parameters and the depend-
ency of activity on pH and temperature were determined.
PPADH shows a substrate preference for the reduction of
linear and branched aliphatic aldehydes. Surprisingly, the
enzyme shows no comparable activity towards ketones other
than the b-hydroxy-b-trifluoromethyl ketone.
+
NADP -dependent ADH from Pichia pastoris (PPADH) was
Introduction
Enzyme-catalyzed reductions of ketones for the production of
chiral alcohols are important tools for the chemical and phar-
maceutical industries. Biocatalytic reductions have developed
and offers many advantages such as the use of water as sol-
[5]
vent for one-pot synthetic approaches.
In particular, the synthesis of 1,3-diols is currently of consid-
erable interest in organic chemistry, because of their use as
building blocks for drugs or biologically active agents contain-
[
1,2]
to become a first-choice approach.
There are many advan-
tages and benefits that biocatalytic asymmetric reductions can
offer, such as highly selective reactions, environmentally
benign processes, and efficient synthesis of complex chiral
[6,7]
ing 1,3-diol subunits.
Recently, it was reported that the
enantioselective organocatalytic aldol addition of acetone to
aldehydes can be combined, in a one-pot procedure, with the
diastereoselective enzymatic reduction of the aldol’s keto func-
tion. As a result, all four stereoisomeric 1,3-diols can be pro-
duced in such a way, with diastereomeric ratios (dr values) typ-
ically >20:1, and virtually perfect enantiomeric purity (ee>
[
3]
products. Enantiopure alcohols are important building blocks
and serve as key intermediates in the synthesis of pharmaceut-
[
4]
[5]
icals, fine chemicals, and natural products. The combination
of biocatalytic and chemocatalytic reactions represents an eco-
nomically and ecologically attractive tool in organic synthesis
[7]
9
8%). The development of a related approach providing
chiral 1,3-diols bearing a trifluoromethyl substituent at the
[8]
[a] D. Bulut, Prof. Dr. W. Hummel
quaternary chiral center is our future aim: the combination of
the organocatalytic aldol process starting with acetone and tri-
fluoroacetophenone with a subsequent biocatalytic reduction
to generate the corresponding chiral 1,3-diols (Scheme 1) in a
stereochemically defined form.
Institute of Molecular Enzyme Technology
Heinrich Heine University of Dꢀsseldorf, Research Centre Jꢀlich
Wilhelm-Johnen-Strasse, 52426 Jꢀlich, (Germany)
[
b] D. Bulut, Prof. Dr. H. Grçger, Prof. Dr. W. Hummel
Faculty of Chemistry, Bielefeld University
Universitꢁtsstrasse 25, 33615 Bielefeld (Germany)
E-mail: werner.hummel@uni-bielefeld.de
To address this aim, we carried out enzyme screening with
the goal of finding an alcohol dehydrogenase (ADH) capable
of reducing b-hydroxy-b-trifluoromethyl ketone 1 (Scheme 1)—
[
c] Dr. N. Duangdee, Prof. Dr. A. Berkessel
Department of Chemistry, Cologne University
Greinstrasse 4, 50939 Kçln (Germany)
referred to from now on as CF -ketone 1—to its corresponding
3
1
,3-diol 2. In this report, we describe the screening, identifica-
Supporting information for this article can be found under http://
dx.doi.org/10.1002/cbic.201600101.
tion, and biochemical characterization of a new ADH from
ChemBioChem 2016, 17, 1 – 11
1
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Scheme 1. ADH-catalyzed synthesis of chiral trifluoromethylated 1,3-diol 2.
The precursor rac-1 can be obtained through the organocatalyzed aldol re-
action between trifluoroacetophenone and acetone.
Pichia pastoris GS115 as a promising enzyme and its applica-
tion for the stereoselective reduction of the sterically demand-
ing racemic trifluoromethylated substrate 1.
Results
Figure 1. A) Purification strategy, and B) SDS-PAGE analysis of PPADH at dif-
ferent stages of purification from P. pastoris cells. Lane M: molecular weight
marker (Prestained Protein Ladder, Thermo Scientific). Lane 1: crude extract
New microbial ADHs were screened for the enantiospecific re-
duction of compound rac-1 for the production of chiral 1,3-
diols (Scheme 1, compounds 1 and 2). A large screening in a
strain collection of over 200 strains of bacteria and yeasts was
performed, resulting in only two strains showing ADH activity
(
10 mg). Lane 2: pool after HIC purification (10 mg). Lane 3: eluate after IEC
purification (10 mg, TCA precipitated). The enriched protein bands that were
cut out are surrounded by boxes and designated as samples 1 and 2.
in the reduction of the CF -ketone. The crude extracts of the
3
two strains Candida boidinii and Candida methanolica tested
positive by gas chromatography for the reduction of the CF3-
ketone 1. It was shown that the enzyme of interest is highly
enantiospecific and reduces only the S enantiomer in the race-
mic mixture of compound rac-1.
These results indicate that the desired ADH seems to be
a rare enzyme with unusual properties. Unfortunately, the ge-
nomes of both strains are unknown, and this makes it difficult
to gain access to a recombinant enzyme. We therefore
screened more strains among the genera Candida and Pichia
and found some more strains, including P. pastoris. Preliminary
tests had already shown a NADPH cofactor preference for re-
purification is shown in Figure 2B. Here, again, only fractions
that tested positive were selected for the subsequent gel filtra-
tion step (data not shown). Although the desired protein could
not be obtained in a homogenous form, we were able to ob-
serve a significant enrichment of two proteins during the pu-
rification procedure. A representative SDS polyacrylamide gel
documenting the purification is shown in Figure 1B.
The two enriched protein bands were cut out of the gel
(Figure 1B, samples 1 and 2), and a tryptic in-gel digestion was
performed. Thanks to the fact that the whole genome se-
quence of P. pastoris GS115 is available in databases, MALDI-
TOF mass spectrometry could be applied. The observed mass
prints of the two protein bands were compared with databas-
es and revealed a glyceraldehyde-3-phosphate dehydrogenase
(score 133, Uniprot ID: C4R0P1) for sample 2, whereas
sample 1 was identified as a putative dihydrokaempferol 4-re-
ductase (score 186, Uniprot ID: C4R4L0). We assumed that the
glyceraldehyde-3-phosphate dehydrogenase is a ubiquitous
and essential enzyme in the glycolytic pathway of glucose and
is therefore widely distributed among eukaryotes and even
prokaryotes, and so we decided to proceed only with the puta-
tive dihydrokaempferol 4-reductase, further designated as
PPADH.
duction of the CF -ketone. In the genome database only the
3
[
9]
[10]
strains P. pastoris GS115 and CBS7435 are sequenced, and
so we chose P. pastoris GS115 for further investigations. We
cultivated this strain, prepared the crude extract by sonication,
and tested it for the reduction of the racemic CF -ketone 1, ob-
3
serving the best levels of conversion out of all tested strains.
Because this strain includes nine annotated ADHs, we decided
to purify the desired CF -ketone-reducing ADH from crude cell
3
extract, by means of several purification steps.
Purification of ADH from P. pastoris GS115
The desired ADH from P. pastoris GS115 was purified from the
crude cell extract by a four-step procedure (Figure 1). The first
purification step consists of an ammonium sulfate precipitation
followed by a hydrophobic interaction chromatography (HIC)
purification. An elution profile of the HIC purification is shown
in Figure 2A.
Recombinant expression of PPADH
To address the question of whether or not the identified
PPADH is the desired dehydrogenase responsible for the enan-
tioselective reduction of compound 1, the gene coding for the
putative dihydrokaempferol 4-reductase (PPADH) was cloned
and expressed in Escherichia coli BL21(DE3), which lacks reduc-
tase activity for substrate rac-1. We used the crude extract of
the heterologously expressed PPADH and observed the desired
Selected fractions were tested for reduction of the CF3-
ketone 1 and analyzed by GC. Fractions that tested positive
were pooled and prepared for the next purification step: ion-
exchange chromatography (IEC). The elution profile for the IEC
reduction of the CF -ketone 1 by GC analysis.
3
&
ChemBioChem 2016, 17, 1 – 11
www.chembiochem.org
2
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Figure 3. A) Influence of expression temperature on soluble expression and
activity of PPADH in E. coli BL21(DE3). Reduction activity was measured with
10 mm substrate rac-1 in Tris·HCl buffer (100 mm, pH 7) and NADPH
(
0.25 mm) at 258C in triplicate; standard deviations are shown in the dia-
Figure 2. Elution profiles of Pichia pastoris wild type crude extract: A) after
ammonium sulfate precipitation during HIC purification, and B) after HIC pu-
rification during IEC purification. UV absorption at 280 nm and the elution
gradient are shown. Gray-shaded boxes show fractions with reduction activi-
gram. B) Influence of an N- or C-terminal affinity tag on the activity of
PPADH (crude extract). Relative activities of PPADH without tag [A], with a
C-terminal His-tag [B], with an N-terminal His-tag [C], or with both C- and N-
terminal His-tags [D] are given; reduction activity was measured with 20 mm
benzaldehyde in Tris·HCl buffer (100 mm, pH 7) and NADPH (0.20 mm) at
2
3
ty for CF -ketone 1.
À1
58C in triplicate; 100% corresponds to an activity of 1.22 Umg in the
crude extract.
In conclusion, we were able to assign the enzyme responsi-
ble for reduction of racemic compound 1 after several purifica-
tion steps and subsequent identification by MALDI-TOF mass
spectrometry.
the N-terminal His-tag variant. Here we can assume that the
fusion of PPADH to the N-terminal His-tag causes a strong im-
provement in activity (ca. 95%) relative to that of the native
enzyme in the crude extracts (Figure 3B).
In order to improve and simplify the expression and purifica-
tion of the ADH, standard cloning techniques were used to
create the recombinant plasmid pET28a-PPADH, corresponding
to a hexahistidine affinity purification motif (His-tag), fused to
the N terminus of the protein. The recombinant plasmid was
expressed in E. coli BL21(DE3) at different temperatures, reveal-
ing best expression and activities at 378C (Figure 3A).
The N-terminal His-tag of the protein enables fast and
simple purification by “immobilized-metal affinity chromatog-
raphy” (IMAC) with Ni-NTA resin material. An overview of the
protein purification is given in Table 1 and in Figure S1 in the
Supporting Information. Our purification procedure achieved
In addition, the influence of N- or C-terminal modification
on the activity of PPADH was investigated (Figure 3B). Our
result revealed that neither an N- nor a C-terminal His-tag has
any negative influence on the activity of PPADH relative to
that of the unmodified wild-type enzyme. In fact, the variant
with an N-terminal His-tag showed much higher activity than
the native enzyme, whereas the variant with the C-terminal
His-tag showed activity similar to that of the wild type. The
variant with both C- and N-terminal His-tags also showed
higher activity than the wild-type enzyme, similar to that of
Table 1. Purification of recombinant PPADH from E. coli BL21(DE3) cells
by IMAC with Ni-NTA.
Purification
step
Total
activity [U]
Specific activity
Purification
factor
Yield
[%]
À1
[a]
[Umg
]
crude extract
purified PPADH
117
78
3.6
37.3
1
10
100
67
[
a] 100% correspond to the total activity [U] of the crude extract.
ChemBioChem 2016, 17, 1 – 11
www.chembiochem.org
3
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
approximately 67% recovery. The resulting specific activity of
to be at pH 5.5 in citrate/phosphate buffer. This broad pH opti-
mum of the reduction reaction is an advantage for the use of
PPADH as a versatile catalyst.
À1
PPADH after the purification was 37.3 Umg , corresponding
to a tenfold improvement in specific activity relative to that of
the crude extract. A single band with an apparent molecular
mass of 41 kDa (Figure S1, lane 2) corresponds to the enzyme,
confirming the calculated molecular mass (41.41 kDa) derived
from the amino acid sequence.
In addition, a study addressing the optimum temperature
was performed. Figure 4B illustrates an almost linear increase
of activity from 20 to 408C with an optimum activity between
40 and 458C. Investigations of the temperature stability of the
purified enzyme were also performed at different temperatures
(4, 20, and 508C). Here we were able to observe good stability
of PPADH at 48C (half activity after 100 days), whereas the
enzyme showed loss of activity at 208C storage after five days
In summary, purification of the recombinant PPADH to ho-
mogeneity (>95%) was achieved through a single purification
step (Figure S1).
(
data not shown). With use of 508C for storage, rapid inactiva-
Dependence of PPADH activity on temperature and pH
tion was found after only 5 min incubation time (data not
shown). In general, long-time storage of PPADH at 48C is rec-
ommended.
To determine the dependence of PPADH activity on pH and
temperature, the reduction of the commercially available ethyl
4-chloroacetoacetate was measured. For the determination of
the pH optimum, four buffers with overlapping pH ranges
were used [citrate/phosphate buffer (pH 4.0 to 6.5), potassium
phosphate buffer (pH 6.0 to 7.5), Tris·HCl buffer (pH 7.0 to 9.0),
and sodium carbonate buffer (pH 9.0 to 10.0)]. As demonstrat-
ed in Figure 4A, the enzyme showed good activity over a wide
pH range (pH 4.0 to 8.0). The optimum activity was determined
Characterization of the kinetic parameters
The kinetic parameters of the purified PPADH (with an N-termi-
nal His-tag) for the substrates benzaldehyde and the racemic
CF -ketone 1, together with the cofactor NADPH, were mea-
3
sured at 258C and pH 7 as described in the Experimental Sec-
tion. For benzaldehyde, the Michaelis–Menten constant (KM)
was determined to be (5.2Æ0.3) mm, and the maximum specif-
À1
ic activity (V ) was (78.8Æ1.6) Umg [corresponds to a k
max
cat
À1
value of (54.4Æ1.1) s ]. For the racemic CF -ketone 1, the Mi-
3
chaelis–Menten constant (K ) was determined to be (2.3Æ
M
0
.1) mm, and the maximum specific activity (V ) was (64.5Æ
max
À1
À1
0
.5) Umg [kcat =(44.5Æ0.4) s ]. Because the K value was de-
M
termined with a racemic mixture and the enzyme reduces only
the S enantiomer (see later results regarding kinetic resolution
of rac-1), the K value would be only half—about 1.15 mm. For
M
the cofactor NADPH, the estimated kinetic constants were:
K =(16.1Æ0.9) mm and V =(84.6Æ1.0) UmgÀ1 [kcat =(58.4Æ
M
max
À1
0
.6) s ].
Substrate specificity of PPADH
The substrate specificity of the enzyme was determined by
using spectrophotometric assays for either the reduction or ox-
idation reaction. The specific activities of the purified enzyme
in the reduction of different linear aliphatic, branched aliphatic,
and aromatic aldehydes and ketones (Table 2) and the oxida-
tion of a variety of alcohols (Table 3) were determined. For the
reduction reactions, activities were compared with those deter-
mined with the standard substrate benzaldehyde (entry 1, spe-
À1
cific activity 38.6 Umg ). In the reduction, the best activity
was obtained with 2-methylvaleraldehyde (entry 6). Other sub-
strates measured are listed in Table 2. In general, linear and
branched aliphatic aldehydes are good substrates for PPADH.
Out of many ketones tested as substrates, compound 1 and
some b-keto esters such as ethyl 4-chloroacetoacetate
(entry 15) and ethyl 2-chloroacetoacetate (entry 14) are re-
duced by the PPADH with high specific activities (38 and
Figure 4. Determination of A) optimum pH, and B) optimum temperature.
The effects of pH and temperature on the enzyme activity of the purified
PPADH in the reduction of ethyl 4-chloroacetoacetate were measured in trip-
licate by monitoring the consumption of NADPH at 340 nm. The standard
deviations of triplicates are shown in each diagram. For the determination
of the optimum pH, four different buffers were used. Symbols: *: citrate/
À1
À1
5.7 Umg , in comparison with 37.5 Umg
for the CF3-
phosphate buffer (pH 4.0 to 6.5, 100 mm),
pH 6.0 to 7.5, 100 mm), ~: Tris·HCl buffer (pH 7.0 to 9.0, 100 mm), *: sodium
carbonate buffer (pH 9.0 to 10.0, 100 mm).
&
: potassium phosphate buffer
ketone 1). With regard to aliphatic ketones, the PPADH showed
almost no activity towards the short-chain compound butan-2-
(
&
ChemBioChem 2016, 17, 1 – 11
www.chembiochem.org
4
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Table 2. Photometrically determined activities for the reduction of vari-
Table 3. Photometrically determined activities for the oxidation of vari-
ous substrates by PPADH. The activities were obtained by use of the stan-
dard spectrophotometric assay; 100% corresponds to 38.6 Umg (puri-
ous substrates by PPADH. The activities were obtained by use of the stan-
dard spectrophotometric assay; 100% corresponds to 6.6 Umg (purified
À1
À1
fied enzyme).
enzyme).
[
b]
[a]
Substrate
Specific activity
Relative activity
[%]
Substrate
Specific activity
Relative
activity [%]
À1
À1
[U mg
]
[Umg
]
1
8
9
1.06
16.0
29.3
1
2
3
4
38.6
100.0
0.2
1
1.94
0.06
2
2
0
1
1.11
2.04
16.7
30.9
0.10
0.3
22
2.53
1.29
1.26
3.62
38.3
19.5
19.1
54.9
[
a]
23
37.5
97.2
2
2
4
5
5
6
68.0
96.1
176.2
248.9
2
2
6
7
6.61
3.79
100.0
57.4
7
8
9
0
28.5
37.5
73.9
97.2
2
2
8
9
2.09
0.57
31.6
8.7
17.8
27.8
1.6
46.1
71.9
4.1
1
3
0
2.16
1.19
32.7
18.0
11
1
1
2
3
0.8
2.1
31
[
a] In addition to the displayed compounds, (R)-hex-5-en-2-ol, 2-phenyl-
propan-1-ol, benzyl alcohol, (S)-2-aminobutan-1-ol, (R)-octan-2-ol, butane-
,2-diol, propane-1,2-diol, 1-octen-3-ol, cyclohexanol, butane-2,3-diol, 1-
4-chlorophenyl)ethanol, (S)-butan-2-ol, pent-1-en-3-ol, cyclohex-2-en-1-ol,
0.5
1.3
1
(
14
5.7
14.7
and (R)-1-phenylethanol were also tested and found to be inactive (activi-
ty <0.5 Umg ).
À1
1
1
1
5
6
7
38.0
0.5
98.3
1.3
À1
one (<0.1 Umg ), but ketones with longer chain lengths (C –
6
C8) are well accepted by the dehydrogenase (entries 11–13). It
seems that the PPADH has special requirements with regard to
substrate acceptance, which need to be further investigated.
For the oxidation, (S)-(+)-hexan-2-ol (entry 26) was used as
3.3
8.5
3
[a] Racemic CF -ketone 1 (20 mm) was used in the assay. [b] In addition to
À1
the displayed compounds, benzylacetone, 2-cyclohexen-1-one, butan-2-
one, fenchone, 3,3,5-trimethylcyclohexanone, propiophenone, (R)-(À)-car-
vone, nonan-2-one, (R)-(+)-pulegone, (1R,4S)-p-menthan-3-one, 4-methyl-
pentan-2-one, 4-methoxy-4-methylpentan-2-one, 4,4-dimethyldihydrofur-
an-2,3-dione, 4-hydroxyhexan-3-one, and acetone were also tested and
a reference substrate (6.61 Umg ). In the oxidation, the best
activity was obtained with (S)-(+)-hexan-2-ol as substrate; ac-
tivities towards other substrates are shown in Table 3. In gener-
al, both primary and secondary alcohols are accepted by the
PPADH. On comparing the reduction and oxidation reactions,
it became obvious that the PPADH favors the reduction reac-
tion.
À1
found to be inactive (activity <0.5 Umg ).
ChemBioChem 2016, 17, 1 – 11
www.chembiochem.org
5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
In gene databases, PPADH is classified as a flavonoid-con-
verting enzyme; more precisely, it is termed a “dihydrokaemp-
ferol reductase”. We thus studied the enzyme-catalyzed con-
version of commercially available dihydrokaempferol (20 mm)
in the presence of NADPH and an NADPH-regenerating system
[
d-glucose and glucose dehydrogenase (GDH)]. Samples taken
after overnight incubation (ca. 20 h) showed no product for-
mation at all by HPLC. Other flavonoids such as naringenin, di-
hydromyricetin, and taxifolin were also tested, but no conver-
sion could be observed, thus suggesting that the proposed
classification for the gene coding for PPADH is not correct.
Enantioselectivity of PPADH
To determine the enantioselectivity of PPADH, some prochiral
ketones were reduced to their corresponding alcohols
(Table 4). It can be seen that all tested substrates (one aromatic
Table 4. Reduction of selected ketones by ADH PPADH. Degrees of con-
version and ee values were determined by chiral GC.
Substrate
Conversion [%]
Selectivity [ee]
Figure 5. Time course of kinetic resolution of racemic trifluoromethylated b-
hydroxy ketone 1 by PPADH [20 mm of racemic substrate 1, 200 mm b-d-glu-
cose, 1 mm NADP , 0.5 U purified PPADH, 5 U GDH, KPi buffer (pH 7, 0.5m),
acetophenone
hexan-2-one
heptan-2-one
93
>99
>99
>99
S, >99%
S, >99%
S, >99%
S, >99%
+
[
a]
258C, 900 rpm]. *: S enantiomer of b-hydroxy ketone 1;
: R enantiomer of
&
[
a]
1; ~: S,S form of diol 2. Concentrations were determined by chiral GC.
octan-2-one
[a] Hydroxy groups of products were acylated to afford trifluoroacetate
derivatives.
and HPLC analysis (Figure S3C). Further investigations of this
KR and its dependence on the reaction time revealed a minimal
decrease in the diastereomeric excess of the (S,S)-1,3-diol
(down to 97% de after 18 h; data not shown).
and three aliphatic ketones) were reduced to their correspond-
ing S-configured alcohols with high enantioselectivity (Table 4),
thus showing that PPADH is an S-specific ADH. We also tested
PPADH for the oxidation of racemic hexan-2-ol, which resulted
in a kinetic resolution in which only one enantiomer was oxi-
dized to the corresponding ketone (Figure S2). The oxidation
of racemic heptan-2-ol and octan-2-ol revealed the same re-
sults as for the kinetic resolution of hexan-2-ol (data not
shown).
Discussion
In this study we report the screening, identification, and char-
acterization of a new NADPH-dependent ADH from P. pastoris
GS115. Over 200 strains from a collection of bacteria and
yeasts were screened for reduction of compound 1 but only
a few Candida and Pichia strains showed enantioselective re-
duction activity towards this substrate. Purification of the de-
sired dehydrogenase from P. pastoris GS115 could be success-
fully performed by means of several purification steps, and the
enzyme could be identified by MALDI-TOF mass spectrometry.
PPADH was (incorrectly) classified as a putative dihydrokaemp-
ferol 4-reductase (UniprotKB). The gene encoding PPADH was
successfully cloned, expressed in E. coli BL21, and purified by
IMAC. We were able to observe that the fusion of an N-termi-
nal His-tag has an influence on the activity of the PPADH in
the crude extracts (Figure 3B). The literature offers some exam-
ples in which enzyme activities are affected by a His-tag for
various reasons: by influencing, for example, the protein solu-
Kinetic resolution of racemic b-hydroxy-b-trifluoromethyl
ketone 1
As we have already mentioned, the newly discovered PPADH
shows a preference towards the S enantiomer for the reduc-
tion of compound rac-1, and so we investigated PPADH for
this asymmetric reduction with simultaneous regeneration of
NADPH by use of glucose and GDH as cofactor regeneration
system (Figure 5). A typical time course of the asymmetric re-
duction of racemic b-hydroxy-b-trifluoromethyl ketone 1 is il-
lustrated in Figure 5; the result is a continuous decrease solely
in the S enantiomer with an accompanying increase in the 1,3-
diol, resulting in nearly complete KR after one hour. Moreover,
the PPADH showed a high stereochemical preference in the
formation of the new second stereogenic center, yielding 1,3-
diol (S,S)-2 with excellent ee and de>99%, together with
[11–16]
bility, protein structure, or substrate-binding affinity.
For
PPADH, further specific studies to elucidate the influence of
the His-tag on the activity will be necessary.
A broad investigation of the substrate specificity, including
aldehydes, ketones, and alcohols, revealed a preference of
PPADH activity for reduction of different aliphatic aldehydes
>
99% ee for the remaining (R)-1, determined at t=2 h by GC
&
ChemBioChem 2016, 17, 1 – 11
www.chembiochem.org
6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
with a medium chain length of C to C (entries 5–8) and to-
both S and R enantiomers of the racemic CF -ketone 1 would
4
8
3
wards the aromatic substrate benzaldehyde (entry 1). A com-
be reduced to afford their corresponding 1,3-diols 2. Therefore,
we applied the newly discovered biocatalyst for the kinetic res-
olution of rac-1 (Figure 5). Our results showed nearly complete
conversion only of the S enantiomer after one hour, with the
newly formed (S,S)-1,3-diol being formed with ee and de
>99% at t=2 h, confirmed by chiral GC and HPLC. Further in-
vestigation of this kinetic resolution and its dependence on
the reaction time revealed a minimal decrease in the diastereo-
meric excess of the (S,S)-1,3-diol (down to 97% after 18 h). We
assume that, after complete consumption of the preferred S
enantiomer of the racemic mixture, the enzyme begins to act
unselectively, with the remaining reduction activity resulting in
minimal acceptance of the R enantiomer, which is reduced to
afford the corresponding (R,S)-diol. From this result and with
regard to the reductions of the different ketones producing
only S-configured alcohols (Table 4) and the enantioselective
oxidation of racemic hexan-2-ol (Figure S2), it can be assumed
that the PPADH is an S-selective ADH.
parison of the unsaturated aldehyde trans-2-hexen-1-al
(entry 8) with hexanal (entry 7) showed a slight preference for
the unsaturated analogue. Also, aromatic aldehydes such as
benzaldehyde (entry 1) and b-keto esters are reduced by
PPADH. The dehydrogenase shows good activity towards ethyl
4-chloroacetoacetate (entry 15), but the activity decreases six-
fold if the location of the chlorine atom is changed from the 4-
to the 2-position (entry 14). In addition, activity drops with in-
creasing size of the alcohol part of the ester (entry 16). Surpris-
ingly, PPADH showed lower activities towards aliphatic ketones
(
entries 11–13) with chain lengths of C –C , relative to reduc-
6 8
tion of rac-1, and almost no activity in the reduction of other
ketones tested. With the substrate hexan-2-one, the PPADH
showed higher activity than with heptan-2-one and octan-2-
one, thus implying that longer chain lengths (>C ) lead to de-
6
creases in activity (entries 11–13). With regard to oxidation re-
actions, PPADH showed a broad substrate acceptance includ-
ing primary and secondary alcohols. Interestingly, introduction
of a double bond into a substrate causes a decrease in the ac-
tivity of PPADH by a factor of two (compare entries 22/23 and
In the literature, only a few examples of kinetic resolutions
by use of dehydrogenases have been described. A glycerol de-
hydrogenase from Gluconobacter oxydans, for example, was
used in the kinetic resolution of glyceraldehyde with 50% con-
26/27). As a consequence, (S)-hexan-2-ol (entry 26) and rac-
[17]
pentan-2-ol (entry 22) as substrates were oxidized twice as rap-
idly as their unsaturated analogues (entries 27 and 23).
version, and an ee >99% was achieved. Furthermore, a one-
pot synthesis of enantiopure 3,4-dihydroisocoumarins through
reductive dynamic kinetic resolution was applied with ADH-A
With regard to the reduction of compound 1, containing
a tertiary alcohol group (Scheme 2), two routes are conceivable
in the presence of an enantioselective ADH. If the ADH or the
enzyme of interest were stereoselective for C-1, a kinetic reso-
lution of the racemate would be possible, because only one of
two enantiomers—in this case the S enantiomer of rac-1—
would be reduced to its corresponding 1,3-diol, resulting in
the enantiomerically pure 1,3-diol 2 and the ketone (R)-1 as
products. Otherwise, if the ADH were nonselective for C-1,
[18]
from Rhodococcus ruber. In addition, whole cells of a geneti-
cally engineered Saccharomyces cerevisiae strain were used as
a biocatalyst for the kinetic resolution of racemic bicyclo[3.3.1]-
[19]
nonane-2,6-dione. Oxidative kinetic resolutions have in most
[
20–22]
cases been applied to racemic alcohols.
Conventionally, ki-
[
23–28]
netic resolutions are performed with hydrolases,
because
of their simple handling without the need for cofactor regener-
ation systems. In addition to these, the described kinetic reso-
lution of the hydroxy ketone rac-1 (Scheme 1) represents
a means of biocatalytic access to enantiomerically pure tertiary
[29]
alcohols through the use of dehydrogenases. Taken togeth-
er, in this study we found and described the new ADH PPADH
from P. pastoris; it can be used for the enantiospecific and dia-
stereoselective reduction of b-hydroxy-b-trifluoromethyl
ketone 1 to its corresponding 1,3-diol 2. Additionally this ADH
is useful for the asymmetric oxidation of racemic secondary
alcohols. We are currently developing a combination of the
organocatalytic aldol reaction with subsequent asymmetric
bioreduction.
Experimental Section
General: If not stated otherwise, all chemicals were purchased
from Sigma–Aldrich or Roth. Centrifugation was carried out with
RC5BPlus (Sorvall, Waltham, USA) and Mikro22 and Rotina 35 R
(Hettich, Tuttlingen, Germany) centrifuges. For GC analysis, a GC-
Scheme 2. Principal stereochemical aspects of the preparative application of
an ADH for the reduction of the racemic CF -ketone 1, demonstrated by two
3
possible routes. If the ADH is selective for C-1, a kinetic resolution is possi-
ble, with a maximum yield of 50% (R)-1’ being achievable. If the ADH is not
selective for C-1, both enantiomers should be reduced to afford the corre-
sponding 1,3-diols.
17A (Shimadzu) was used. Restriction enzymes were purchased
from Thermo Scientific.
Cultivation of microbial strains and isolation of crude extracts:
Strains were obtained from the DSMZ (Deutsche Stammsammlung
von Mikroorganismen und Zellkulturen) and were cultivated under
ChemBioChem 2016, 17, 1 – 11
www.chembiochem.org
7
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
recommended medium and temperature conditions. After cultiva-
tion, cells were harvested by centrifugation at 10000g for 30 min
at 48C. Isolated cells were resuspended in potassium phosphate
buffer (50 mm, pH 7) at a ratio of 1 g (wet weight) of cells per 3 mL
of buffer. Glass beads (ø 0.25–0.50 mm, 1.2 g) were added to a cell
suspension (600 mL). Cells were disrupted (with cooling) at 30 Hz
for 15 min with a mixer mill from Retsch (Haan, Germany). The
crude extract was obtained after centrifugation at 20000g for
Molecular cloning of the alcohol dehydrogenase gene of
PPADH: Genomic DNA isolated from P. pastoris GS115 was used as
template for amplification of the putative dihydrokaempferol 4-re-
ductase with the following primers: 5’-GCG CTT CCA TAT G GTT
TCT AAG G-3’ and 5’-CGG CTC GAG ATT ATT TAT TAG CAC GC-3’
cloned with an N-terminal His-tag (in pET28a, Novagen (Merck))
with recognition sites for NdeI and XhoI. The PCR product was
cloned into pET28a (Novagen) between the NdeI and XhoI restric-
tion sites by standard techniques. To investigate the effects of His-
tags on activity, additional constructs 1) without a His-tag, 2) with
a C-terminal His-tag, and 3) with both N- and C-terminal His-tags
were created by QuikChange PCR.
3
0 min at 48C.
Screening assay for the reduction of the CF -ketone 1: The isolat-
3
ed crude extracts, in combination with an in situ cofactor regenera-
tion system consisting of d-glucose and GDH, were tested for the
Heterologous expression of PPADH: For recombinant protein ex-
pression, E. coli BL21(DE3) cells containing the recombinant plas-
reduction of the CF -ketone. The assay mixture, with a total
3
+
volume of 1 mL, contained crude extract (200 mL), NAD and
NADP (1 mm), potassium phosphate buffer (pH 7.5, 50 mm),
+
mid were cultivated overnight at 378C in lysogeny broth (LB;
À1
1
0 mL) containing kanamycin (50 mgmL ). For expression in coni-
MgCl2 (1 mm), d-glucose (100 mm) and racemic CF -ketone
3
cal flasks, a final cell concentration of 0.05 optical density at
600 nm (OD600) was used to inoculate different amounts of Terrific
Broth (TB) medium containing the corresponding antibiotic for se-
lection. The cultures were cultivated at 378C. At an OD600 of 0.5 to
(
10 mm) as substrate. For cofactor regeneration, GDH from Bacillus
À1
subtilis (ꢀ5 Uml ) was used. The reaction mixture was shaken at
9
00 rpm at 308C overnight (20 h). Reduction of the CF -ketone was
3
examined by GC analysis. For this purpose, samples (100 mL) were
mixed with ethyl acetate (200 mL) for extraction and centrifuged.
0
.7 the production of the recombinant ADH was induced by addi-
tion of isopropyl-b-d-thiogalactopyranoside (IPTG) to a final con-
centration of 0.5 mm. For the determination of the optimal growth
conditions cultures were grown after induction at 20, 24, 30, and
Cultivation of Pichia pastoris GS115, cell disruption, and purifica-
tion of the alcohol dehydrogenase: The strain Pichia pastoris
GS115 was kindly provided by the research group of Prof. Joachim
Ernst (University of Dusseldorf) and was grown in YPD medium at
3
78C and assayed after a period of 20 h.
Purification of recombinant PPADH: The bacterial culture was har-
vested by centrifugation (18000g, 30 min, 48C). A cell suspension
2
58C for three days. After harvesting by centrifugation, the Pichia
cells were resuspended in potassium phosphate buffer (50 mm,
pH 7) at a ratio of 1 g (wet weight) of cells per 3 mL of buffer. The
cell suspension was disrupted with a French Press Cell Disrupter
(25%) in Tris·HCl buffer (pH 7.0, 100 mm) was prepared and dis-
rupted by sonication on ice (2 min, 25% power output, 1 min cool-
ing, three cycles in total). After a centrifugation step (17000g,
(
Glen Mills, Inc., Clifton, NJ, USA) three times at 900 psi with cool-
3
0 min, 48C), the cleared crude extract was used for protein purifi-
ing steps in between. The crude extract was obtained after centri-
fugation (20000g for 45 min at 48C) and used for the ammonium
sulfate precipitation. Firstly, ammonium sulfate was added to the
crude extract to produce a final concentration of 2.47m with
gentle stirring on ice for 1 h and a subsequent centrifugation step
to remove a proportion of the Pichia crude extract proteins. The re-
sulting supernatant was subjected to a second ammonium sulfate
precipitation by addition of ammonium sulfate to a final concen-
tration of 4.25m under the same conditions as described previous-
ly. After centrifugation the supernatant was removed, and the re-
sulting sediment was resuspended in the initial volume with start
buffer [potassium phosphate buffer (100 mm)+ammonium sulfate
cation by immobilized metal affinity chromatography (IMAC, Ni-
NTA, MCLAB, Nimagen, Netherlands). The column was equilibrated
with lysis buffer [NaH PO4 (50 mm), NaCl (300 mm), imidazole
2
(10 mm), pH 8.0]. After loading, unbound proteins were eluted
with wash buffer [NaH PO4 (50 mm), NaCl (300 mm), imidazole
2
(20 mm), pH 8.0]. The PPADH was obtained with an imidazole con-
centration of 250 mm [NaH PO (50 mm), NaCl (300 mm), imidazole
2
4
(250 mm), pH 8.0]. After a desalting step with PD-10 columns (Se-
phadex G 25, GE Healthcare), the purified enzyme was stored in
Tris·HCl buffer (pH 7.0, 100 mm) at 48C prior to use.
Activity assay: A continuous assay with use of UV absorbance at
(
pH 7.2, 1.7m)], and the solution was applied to a Phenyl Sephar-
3
40 nm was employed to monitor the decrease in NADPH concen-
ose 6 Fast Flow column (GE Healthcare), which was equilibrated
with five column volumes (CV) start buffer. The proteins were
eluted with potassium phosphate buffer (50 mm, pH 7) with a gra-
dient of 1.7m to 0m ammonium sulfate concentration. Active frac-
tions were pooled, and the buffer was changed to Tris·HCl buffer
tration during reduction catalyzed by the ADH. Benzaldehyde was
used as standard substrate because it is commercially available.
One unit of activity was defined as the amount of enzyme catalyz-
ing the oxidation of 1 mmol NADPH per minute under standard
conditions (258C, pH 7.0). The reduction assay mixture contained
980 mL substrate solution [substrate (10 mm) in Tris·HCl buffer
(
pH 9, 50 mm) by use of an Ultrafiltration membrane (Mr
0000 Da). After concentration, the protein solution was applied to
1
(pH 7.0, 100 mm), NADPH in distilled water (20 mm, 10 mL), enzyme
a Q Sepharose Fast Flow column (GE Healthcare) that had been
equilibrated with five CV Tris·HCl buffer (50 mm, pH 9). The column
was washed with the same buffer, and the proteins were eluted
with Tris·HCl buffer (50 mm, pH 9) containing NaCl (1m) with use
of a linear gradient from 0m to 1m NaCl. Active fractions were
pooled, equilibrated in potassium phosphate buffer [50 mm+NaCl
solution (10 mL)]. For the determination of the enzyme activity in
the oxidation reaction, the assay mixture contained 980 mL sub-
strate solution [substrate (10 mm) in Tris·HCl buffer (pH 9.0,
+
1
00 mm), NADP in distilled water (20 mm, 10 mL), enzyme solution
(10 mL)]. Reactions were started by addition of the enzyme solution
and measured over 1 min.
(
300 mm), pH 7.2] and applied to a Superdex 75 column (GE
Healthcare), which was equilibrated in potassium phosphate buffer
50 mm+NaCl (300 mm), pH 7.2]. Resulting fractions were tested
Determination of pH and temperature optima: In order to obtain
the pH and temperature optima of the ADH activity, the standard
enzyme assay was performed with ethyl 4-chloroacetoacetate
(10 mm) at different temperatures or pH values instead of benzal-
dehyde. (With benzaldehyde as substrate, the determination of pH
[
for activity (NADPH-dependent reduction of CF -ketone). After
3
each purification step, the protein concentration was determined,
and samples were subjected to SDS-PAGE analysis.
&
ChemBioChem 2016, 17, 1 – 11
www.chembiochem.org
8
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
and temperature optima was not accurate, resulting in high fluctu-
ations in measurements.) For the determination of the pH opti-
mum, four buffers with overlapping pH ranges were applied: cit-
rate/phosphate buffer (pH 4.0 to 6.5, 100 mm), potassium phos-
phate buffer (pH 6.0 to 7.5, 100 mm), Tris·HCl buffer (pH 7.0 to 9.0,
Kinetic resolution of racemic b-hydroxy-b-trifluoromethyl
ketone 1: For the kinetic resolution of rac-1 by PPADH, d-glucose
and GDH from Bacillus subtilis were used for NADPH recycling
under the following conditions: racemic substrate 1 (20 mm), b-d-
+
glucose (200 mm), NADP (1 mm), purified PPADH (0.5 U), GDH
1
00 mm), and sodium carbonate buffer (pH 9.0 to 10.0, 100 mm).
(crude extract, 5 U), KPi-buffer (pH 7, 0.5m) at a final volume of 1–
1
.5 mL, 258C, 900 rpm. samples of the reaction mixture (100 mL)
Determination of kinetic parameters: To determine the kinetic
parameters (KM und Vmax) of the purified PPADH with N-terminal
His-tag, the standard activity assay for reduction was used. All
other parameters were kept constant [258C, Tris·HCl buffer (pH 7.0,
were taken periodically, extracted with ethyl acetate or diethyl
ether (200 mL), and analyzed by chiral GC and HPLC.
Chiral GC analysis: For the enzyme-catalyzed reduction of the dif-
ferent biotransformations, analysis of concentrations and enantio-
meric excess (ee) were determined by gas chromatography with
a GC-17A (Shimadzu) equipped with a b-cyclodextrin column (CP-
Chirasil-Dex CB column, 25 mꢁ0.25 mm, Varian). For analysis of the
1
00 mm)], with only the investigated parameter (cofactor or sub-
strate concentration) being modified. Kinetic parameters were cal-
culated by use of a nonlinear fitting algorithm (Graphpad Prism5
software).
reduction of CF -ketone rac-1, the following method was applied:
3
À1
À1
Protein analysis: Protein determination was performed according
6
3
08C (5 min); 308C min to 1358C; 58C min to 1858C (0 min),
08C min to 1958C (0 min). The corresponding retention times
[30]
À1
to Bradford, with BSA as a standard. SDS-PAGE was carried out
with Bis·Tris gels (4–12%) running in SDS buffer. The PageRuler
Prestained Protein ladder (10–170 kDa, Thermo Scientific) was used
for estimation of molecular weights of proteins. The gel was
stained in SimplyBlue SafeStain (Invitrogen).
were: S enantiomer of CF -ketone 1 11.74 min, R enantiomer of
3
CF -ketone 1 11.94 min, corresponding diol of (S)-CF -ketone
3
3
1
5.90 min. The following GC methods were applied for the differ-
ent biotransformations.
Biotransformations of different ketones and secondary alcohols
coupled with coenzyme regeneration: The enantioselectivity of
PPADH in the reduction of different ketones and the oxidation of
racemic secondary alcohols was examined with use of d-glucose
and GDH from Bacillus subtilis for NADPH recycling (reduction part)
À1
1
2
) Reduction of acetophenone: 1008C (2 min), 58C min to
1
À1
408C (0 min), 208C min to 1958C (1 min). Retention times:
acetophenone 6.65 min, (R)-1-phenylethanol 9.66 min, (S)-1-
phenylethanol 9.94 min.
[31,32]
) Reduction/oxidation of hexan-2-one/hexan-2-ol: 708C (15 min).
Retention times: hexan-2-one 4.58 min, (R)-hexan-2-ol
or NAD(P)H oxidase from Lactobacillus sanfranciscensis (LsNOX)
+
for NADP recycling. LsNOX was applied as a recombinant enzyme
cloned in E. coli. Cloning of the corresponding gene, cultivation of
E. coli, and preparation of the crude extract were carried out as de-
scribed in the literature. The different biotransformations were per-
formed as follows:
1
0.38 min, (S)-hexan-2-ol 10.70.
3) Reduction of heptan-2-one and octan-2-one: 608C (5 min), 58C
À1
min to 1408C (0 min) with following retention times: heptan-
2-one 9.68 min, S enantiomer of heptan-2-yl trifluoroacetate
7
.97 min, R enantiomer of heptan-2-yl trifluoroacetate 8.10 min,
Reduction of acetophenone: Substrate (20 mm), d-glucose
and octan-2-one 12.81 min, S enantiomer of octan-2-yl trifluor-
oacetate 11.18 min, R enantiomer of octan-2-yl trifluoroacetate
+
(150 mm), NADP (1 mm), GDH (crude extract, 3 U), purified PPADH
(0.1 mg) in a final volume of 1 mL potassium phosphate buffer
(0.5m, pH 7.0), incubated at 258C. After 19 h, an aliquot (100 mL) of
11.31 min.
the reaction mixture was extracted with ethyl acetate (200 mL) and
analyzed by chiral GC.
Chiral HPLC analysis: The absolute configuration of (S,S)-2 was
identified by HPLC as previously described, with the following
[33]
Reduction of hexan-2-one, heptan-2-one, and octan-2-one: Sub-
strate solution [900 mL, substrate (10 mm) in Tris·HCl buffer (pH 7.0,
method applied for chiral HPLC: Chiralcel OD-H, n-heptane/isopro-
À1
panol 98:2, 0.5 mLmin , UV 203 nm. Retention times for com-
+
0
.1m), d-glucose (50 mm), NADP (1 mm), GDH (crude extract,
pound rac-1: 23 min (R)-1 and 44 min (S)-1. Retention times for
compounds 2 (1,3-diols): (R,S)-2 31.2 min, (R,R)-2 34.3 min, (S,R)-2
3
U), purified PPADH (0.5 U, related to the reduction activity to-
wards hexan-2-one) at a final volume of 1 mL], incubated at 258C.
After 22 h, 100 mL of each reaction mixture were taken and extract-
ed with ethyl acetate (200 mL) and analyzed by chiral GC. For deter-
mination of the enantioselectivity, the hydroxy groups of the sec-
ondary alcohols of heptan-2-ol and octan-2-ol were acylated to
afford the trifluoroacetate derivatives. For this purpose, the reac-
tion mixture (100 mL) was extracted with chloroform, and the or-
ganic phase was isolated. Acylation was performed by adding tri-
fluoroacetic anhydride (25 mL) for 30 min at 258C with continuous
shaking. Subsequently, potassium phosphate buffer (0.5m, pH 7.0,
8
8.9 min, (S,S)-2 166.6 min (Figure S3).
Acknowledgements
The authors D.B., H.G. and W.H. thank the Federal Ministry of
Education and Research, Germany (Bundesministerium fꢀr Bil-
dung und Forschung, BMBF), for generous support within the
programme “Biotechnologie 2020+: Basistechnologien fꢀr eine
nꢁchste Generation biotechnologischer Verfahren” (Projekt-Fçr-
derkennzeichen: 031A184). Furthermore, the authors would like
to thank Dipl.-Ing. Astrid Wirtz for analysis of flavonoids and
Birgit Henssen for analysis of 1,3-diols by HPLC, Dr. Jens Sproß
for MALDI-TOF mass spectrometry, and Dipl.-Ing. Thorsten Rose-
nbaum for technical assistance.
1
00 mL) was added, and the organic phase was subjected to chiral
GC.
Oxidation of racemic hexan-2-ol: Substrate solution [894 mL, sub-
+
strate (20 mm) in Tris·HCl buffer (pH 7, 0.1m), NADP (1 mm),
NAD(P)H oxidase (LsNOX, crude extract, 5 U), purified PPADH
(
0.5 U, relative to the oxidation activity towards hexan-2-ol) at
a final volume of 1 mL], incubated at 258C. At a time point of 22 h,
00 mL of each reaction mixture were taken, extracted with ethyl
acetate (200 mL), and analyzed by chiral GC.
1
ChemBioChem 2016, 17, 1 – 11
www.chembiochem.org
9
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
[
16] A.-C. Freydank, W. Brandt, B. Drꢄger, Proteins Struct. Funct. Bioinf. 2008,
Keywords: alcohol dehydrogenases · asymmetric reduction
and oxidation · biocatalysis · diols · enantioselectivity · kinetic
resolution
7
2, 173–183.
[17] N. Richter, M. Neumann, A. Liese, R. Wohlgemuth, T. Eggert, W.
Hummel, ChemBioChem 2009, 10, 1888–1896.
[
[
[
[
[
18] J. Mangas-Sꢅnchez, E. Busto, V. Gotor, V. Gotor-Fernꢅndez, Org. Lett.
2
013, 15, 3872–3875.
[1] S. M. A. De Wildeman, T. Sonke, H. E. Schoemaker, O. May, Acc. Chem.
Res. 2007, 40, 1260–1266.
19] M. Carlquist, C.-J. Wallentin, K. Wꢄrnmark, M. F. Gorwa-Grauslund, Tetra-
hedron: Asymmetry 2008, 19, 2293–2295.
20] L. Dai-Huan, W. Dong-Zhi, L. Jin-Ping, J. Pure Appl. Microbiol. 2014, 8,
[2] G. W. Huisman, J. Liang, A. Krebber, Curr. Opin. Chem. Biol. 2010, 14,
122–129.
1183–1189.
[
[
[
[
3] Y. Ni, J.-H. Xu, Biotechnol. Adv. 2012, 30, 1279–1288.
4] R. N. Patel, Coord. Chem. Rev. 2008, 252, 659–701.
5] H. Grçger, W. Hummel, Curr. Opin. Chem. Biol. 2014, 19, 171–179.
6] P. Gupta, N. Mahajan, S. C. Taneja, Catal. Sci. Technol. 2013, 3, 2462–
21] C. R. Silva, J. C. Souza, L. S. Araffljo, E. Kagohara, T. P. Garcia, V. H. Peliz-
zari, L. H. Andrade, J. Mol. Catal. B 2012, 83, 23–28.
22] K. Matsumoto, K. Hashimoto, M. Sakuragi, A. Kusunoki, M. Nogawa, Lett.
Org. Chem. 2011, 8, 536–539.
2480.
[
[
23] C. Sanfilippo, G. Nicolosi, A. Patti, J. Mol. Catal. B 2014, 104, 82–86.
24] Y. Zhang, F. Schaufelberger, M. Sakulsombat, C. Liu, O. Ramstrçm, Tetra-
hedron 2014, 70, 3826–3831.
[7] K. Baer, M. Kraußer, E. Burda, W. Hummel, A. Berkessel, H. Grçger,
Angew. Chem. Int. Ed. 2009, 48, 9355–9358; Angew. Chem. 2009, 113,
9519–9522.
[
[
25] S. Devendran, G. D. Yadav, Chirality 2014, 26, 286–292.
26] A. L. Vlasyuk, V. A. Voblikova, G. D. Gamalevich, E. P. Serebryakov, Russ.
Chem. Bull. 2014, 62, 2041–2045.
[8] N. Duangdee, W. Harnying, G. Rulli, J.-M. Neudçrfl, H. Grçger, A. Berkes-
sel, J. Am. Chem. Soc. 2012, 134, 11196–11205.
[9] K. De Schutter, Y.-C. Lin, P. Tiels, A. Van Hecke, S. Glinka, J. Weber-Leh-
mann, P. Rouzꢂ, Y. van de Peer, N. Callewaert, Nat. Biotechnol. 2009, 27,
[
27] U. T. Bornscheuer, G. W. Huisman, R. J. Kazlauskas, S. Lutz, J. C. Moore, K.
Robins, Nature 2012, 485, 185–194.
561–566.
[
[
[
[
28] M. Yousefi, M. Mohammadi, Z. Habibi, J. Mol. Catal. B 2014, 104, 87–94.
29] M. Mꢃller, ChemBioEng Rev. 2014, 1, 14–26.
[
10] A. Kꢃberl, J. Schneider, G. G. Thallinger, I. Anderl, D. Wibberg, T. Hajek, S.
Jaenicke, K. Brinkrolf, A. Goesmann, R. Szczepanowski, A. Pꢃhler, H.
Schwab, A. Glieder, H. Pichler, J. Biotechnol. 2011, 154, 312–320.
30] T. Bradford, Anal. Biochem. 1976, 72, 248–254.
31] B. R. Riebel, P. R. Gibbs, W. B. Wellborn, A. S. Bommarius, Adv. Synth.
Catal. Adv. Synth. Catal 2002, 344, 1156–1168.
32] B. R. Riebel, P. R. Gibbs, W. B. Wellborn, A. S. Bommarius, Adv. Synth.
Catal. 2003, 345, 707–712.
33] N. Duangdee, Organocatalytic Asymmetric Aldol Reactions with Ketone
Acceptors. PhD thesis, Cologne University, 2013.
[
[
[
[
[
11] C. M. Halliwell, G. Morgan, C. P. Ou, A. E. Cass, Anal. Biochem. 2001, 295,
257–261.
[
12] M. Sabaty, S. Grosse, G. Adryanczyk, S. Boiry, F. Biaso, P. Arnoux, D.
Pignol, BMC Biochem. 2013, 14, 14–28.
13] A. Panek, O. Pietrow, P. Filipkowski, J. Synowiecki, Acta Biochim. Pol.
[
2013, 60, 163–166.
14] K. A. Majorek, M. L. Kuhn, M. Chruszcz, W. F. Anderson, W. Minor, Protein
Sci. 2014, 23, 1359–1368.
15] H. Song, E. J. Park, Y. H. Shin, H. S. Kim, D. H. Na, J. Pharm. Invest. 2012,
Manuscript received: February 19, 2016
Accepted article published: April 28, 2016
Final article published: && &&, 0000
42, 15–19.
&
ChemBioChem 2016, 17, 1 – 11
www.chembiochem.org
10
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

FULL PAPERS
Find the needle in the haystack! A
new alcohol dehydrogenase—PPADH
from P. pastoris—was identified. It is
able to distinguish between the two
enantiomers of the racemic b-hydroxy-
b-trifluoromethyl ketone 1 by efficiently
reducing the S form to the correspond-
ing stereochemically uniform 1,3-diol
D. Bulut, N. Duangdee, H. Grçger,
A. Berkessel, W. Hummel*
&& – &&
Screening, Molecular Cloning, and
Biochemical Characterization of an
Alcohol Dehydrogenase from Pichia
pastoris Useful for the Kinetic
Resolution of a Racemic b-Hydroxy-b-
trifluoromethyl Ketone
(
S,S)-2.
ChemBioChem 2016, 17, 1 – 11
www.chembiochem.org
11
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ