Two enantiocomplementary ephedrine dehydrogenases from arthrobacter sp. TS-15 with broad substrate specificity

Article abstract of DOI:10.1021/acscatal.9b00621

The recently identified pseudoephedrine and ephedrine dehydrogenases (PseDH and EDH, respectively) from Arthrobacter sp. TS-15 are NADH-dependent members of the oxidoreductase superfamily of short-chain dehydrogenases/reductases (SDRs). They are specific for the enantioselective oxidation of (+)-(S) N-(pseudo)ephedrine and (-)-(R) N-(pseudo)ephedrine, respectively. Anti-Prelog stereospecific PseDH and Prelog-specific EDH catalyze the regio- A nd enantiospecific reduction of 1-phenyl-1,2-propanedione to (S)-phenylacetylcarbinol and (R)-phenylacetylcarbinol with full conversion and enantiomeric excess of >99%. Moreover, they perform the reduction of a wide range of aryl-aliphatic carbonyl compounds, including ketoamines, ketoesters, and haloketones, to the corresponding enantiopure alcohols. The highest stability of PseDH and EDH was determined to be at a pH range of 6.0-8.0 and 7.5-8.5, respectively. PseDH was more stable than EDH at 25 °C with half-lives of 279 and 38 h, respectively. However, EDH is more stable at 40 °C with a 2-fold greater half-life than at 25 °C. The crystal structure of the PseDH-NAD⁺ complex, refined to a resolution of 1.83 ?, revealed a tetrameric structure, which was confirmed by solution studies. A model of the active site in complex with NAD⁺ and 1-phenyl-1,2-propanedione suggested key roles for S143 and W152 in recognition of the substrate and positioning for the reduction reaction. The wide substrate spectrum of these dehydrogenases, combined with their regio- A nd enantioselectivity, suggests a high potential for the industrial production of valuable chiral compounds.

Full text of DOI:10.1021/acscatal.9b00621

Subscriber access provided by BUFFALO STATE
Article
Two enantiocomplementary Ephedrine Dehydrogenases
from Arthrobacter sp. TS-15 with broad substrate specificity
Tarek Shanati, Cameron Lockie, Lilian Beloti, Gideon Grogan, and Marion B. Ansorge-Schumacher
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.9b00621 • Publication Date (Web): 29 May 2019
Downloaded from http://pubs.acs.org on May 30, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 13
ACS Catalysis
1
2
3
4
5
6
7
8
9
Two Enantiocomplementary Ephedrine Dehydrogenases from
Arthrobacter sp. TS-15 with Broad Substrate Specificity
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Tarek Shanati,a Cameron Lockie,b Lilian Beloti,b Gideon Grogan,b* Marion Ansorge-Schumachera*
aDepartment of Molecular Biotechnology, Technische Universität Dresden, 01062 Dresden.
bDepartment of Chemistry, University of York, Heslington York YO10 5DD UK.
ABSTRACT: The recently identified pseudoephedrine and ephedrine dehydrogenases from Arthrobacter sp. TS-15, PseDH and
EDH, are NADH-dependent members of the oxidoreductase superfamily of short-chain dehydrogenases/reductases (SDRs). They are
specific for the enantioselective oxidation of (+)-(S) N-(pseudo)ephedrine and (-)-(R) N-(pseudo)ephedrine, respectively. Anti-Prelog
stereospecific PseDH and Prelog-specific EDH catalyse the regio- and enantiospecific reduction of 1-phenyl-1,2-propanedione to (S)-
phenylacetylcarbinol and (R)-phenylacetylcarbinol with full conversion and enantiomeric excess of >99%. Moreover, they perform
the reduction of a wide range of aryl aliphatic carbonyl compounds, including keto amines, ketoesters and haloketones, to the
corresponding enantiopure alcohols. The highest stability of PseDH and EDH was determined to be at a pH range of 6.0-8.0 and 7.5-
8.5, respectively. PseDH was more stable than EDH at 25 °C, with half-lives of 279 h and 38 h respectively. However, EDH is more
stable at 40 °C with a two-fold greater half-life than at 25 °C. The crystal structure of the PseDH-NAD+ complex, refined to a
resolution of 1.83 Å, revealed a tetrameric structure, which was confirmed by solution studies. A model of the active site in complex
with NAD+ and 1-phenyl-1,2-propanedione suggested key roles for S143 and W152 in recognition of the substrate and positioning
for the reduction reaction. The wide substrate spectrum of these dehydrogenases, combined with their regio- and enantioselectivity,
suggests high potential for the industrial production of valuable chiral compounds.
KEYWORDS: alcohol dehydrogenase, asymmetric hydrogen transfer, chiral carbonyl reduction, enantioselective alcohol
synthesis, Phenylacetylcarbinol.
Introduction
TS-15 (DSM 32400).20 The enzymes, Pseudoephedrine
dehydrogenase (PseDH) and Ephedrine dehydrogenase (EDH)
catalyse asymmetric hydrogen transfer to 1-phenyl-1,2-
propanedione (Scheme 1), producing (S)-PAC and (R)-PAC
Chiral secondary alcohols are major synthons in the production
of active pharmaceutical intermediates.1 A commonly practiced
biotransformation for their synthesis under mild reaction
conditions is the asymmetric transfer hydrogenation of
respectively.
O
prochiral
ketones
using
NAD(P)H-dependent
dehydrogenases,2-4 which exhibit high optical selectivity.5,6
Dehydrogenases are divided into three structural superfamilies:
Short- (SDR), medium- (MDR) and long-chain (LDR)
dehydrogenases/reductases.7 According to genomic analyses,
about 25% of all dehydrogenases belong to the superfamily of
SDRs.8 Their structural versatility offers a wide functional
spectrum,9,10 which is exploited in several industrial
applications. The NAD(P)H-dependent SDRs from
Lactobacillus brevis (LbADH), Lactobacillus kefir DSM 20587
(LkADH), Aromatoleum aromaticum EbN1 (Phenylethanol
dehydrogenase [PED]) and Ralstonia sp. DSM 6428
(RasADH), are well-known examples of biocatalysts used in
the production of non-racemic aromatic alcohols such as (R)-
and (S)-phenylethanol.11-14 The crystal structure of these four
dehydrogenases have been determined.15-18 Although
bioreductions of 1,2-diaryl-1,2-diketones have been reported
O
NADH
NADH
NAD+
OH
PseDH
EDH
NAD+
OH
O
O
S
S
R
R
( )-phenyacetylcarbinol ( -PAC)
(
)-phenyacetylcarbinol ( -PAC)
ee
> 99% yield,
>99%
ee
> 99% yield,
>99%
Scheme 1. Catalytic reduction of 1-phenyl-1,2-
propanedione 1a.
The identification of such dehydrogenases is advantageous, as
(S)-PAC can serve as a precursor in the synthesis of many
pharmaceuticals, such as (+)-(S,S)-pseudoephedrine.
Here, we provide a detailed insight into both the functional and
structural characteristics of PseDH and EDH, including
definition of their wide and synthetically useful substrate
spectra and stereoselectivity, studies on kinetics and stability,
and an X-ray crystallographic structure of PseDH in complex
with NAD+, refined to a resolution of 1.83Å.
using wild-type strains of yeasts such as Pichia glucozyma,19
,
recombinantly expressed, isolated enzymes for the regio- and
enantioselective reduction of aryl aliphatic 1,2‐diketones such
as
1-phenyl-1,2-propanedione,
to
make
(S)-
phenylacetylcarbinol [(S)-PAC], are currently not available.
We have recently described the isolation of two
enantiocomplementary dehydrogenases from Arthrobacter sp.
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 13
Results and Discussion
PseDH
EDH
1
2
3
4
5
6
7
8
120
100
80
60
40
20
0
120
100
80
60
40
20
0
Molecular mass. The calculated molecular masses based on the
amino acid sequences of the native enzymes of a single subunit
were 28.17 kDa and 25.93 kDa for PseDH and EDH,
respectively (Figure 1). The relative molecular mass of the
recombinant enzyme of a single subunit was estimated by SDS-
PAGE to be about 33 kDa and 27 kDa for PseDH and EDH,
respectively. According to the results of the Blue native PAGE,
the smallest oligomeric state seemed to be a dimer and a
tetramer form for PseDH and EDH, respectively. The
predominant oligomeric states were determined to be the
tetramer and octamer for PseDH and a tetramer in the case of
EDH. In agreement with the results of the In-gel activity assay,
the EDH tended to form higher oligomeric states, up to a
hexadecamer. The largest band observed with PseDH had the
size of a dodecamer, whereas the smallest active oligomer was
a dimer.
0
20
40
60
0
20
40
60
80
100
temperature [°C]
temperature [°C]
Figure 2: Temperature optima of PseDH and EDH measured
spectrophotometrically at 340 nm by the reduction of 1-phenyl-1,2-
propanedione, using NADH in potassium phosphate buffer at pH
7.5 (100 mM).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In order to investigate the thermal stability, the half-lives (t1/2)
of the enzymes were determined in the temperature range
between 5 and 60 °C (Figure 3). PseDH exhibited a reasonable
stability in the range of 15 to 25 °C (t1/2 of 379.4 and 279.7 h,
respectively). Rapid enzyme deactivation was observed at 40
°C (t1/2 of 0.78 h).
EDH
PseDH
100
80
60
40
20
0
800
600
400
200
0
0
20
40
60
80
0
10
20
30
40
50
temperature [°C]
temperature [°C]
Figure 3: Storage stability of PseDH and EDH in the temperature
range of 5 to 60 °C using 100 mM potassium phosphate buffer pH
7.5. The enzymatic activity was determined spectrophotometrically
at 340 nm following the consumption of NADH and after the
addition of the substrate 1-Phenyl-1,2-propanedione. All stability
curves are presented in Supporting Information Figure S1.
Figure 1: SDS-PAGE and blue native PAGE of purified PseDH
(1) and EDH (2) (both recombinant from E. coli) with the molecular
marker (M). A) SDS-PAGE analysis of PseDH and EDH with a
molecular mass of 33 kDa and 27 kDa, respectively. B) In-gel
activity assay based on NBT activity staining before the addition of
Coomassie blue. Both enzymes show different active high
oligomeric states. C) Blue native PAGE of the enzymes after
Coomassie blue staining to highlight all protein bands including the
marker.
The other dehydrogenase EDH demonstrated an unusual
behaviour concerning its thermal stability with a short t1/2 of 2.6
h at 5 °C and a higher stability at 40 °C (t1/2 of 82.8 h). In order
to verify these results, the experiments were carried out twice,
using enzyme from different cultivations, in triplicate, and each
gave the same results. The cold denatured protein of EDH
generated insoluble aggregates, which might be explained by its
tendency to form higher oligomeric states as described in the
previous paragraph. The t1/2 of EDH at 25 °C was lower than of
PseDH (half-life time 38.3 h).
Effect of temperature on the stability and activity. The
influence of the temperature on both dehydrogenases was
studied spectrophotometrically at 340 nm based on the
consumption of NADH in the reduction of 1-phenyl-1,2-
propanedione. The increase of the temperature between 4 and
25 °C showed 4-fold higher initial activity of PseDH, whereas
EDH demonstrated twice the activity at 25 °C comparing to the
activity at 4 °C (Figure 2). PseDH and EDH reached their
maximal initial activity at 35 and 70 °C, respectively.
Interestingly, PseDH exhibited 1.5-fold higher activity in the
range of 25 to 35 °C, whereas EDH demonstrated 14-fold
higher initial activity with an increase of the temperature from
25 to 70 °C. These results suggest higher thermal stability for
EDH in comparison to PseDH.
Effect of pH on the stability and activity. Reductions with
PseDH and EDH displayed pH-optima of 6.5 and 6.0,
respectively. However, EDH exhibited a higher tolerance than
PseDH at acidic pH values in the reduction reaction. In contrast,
PseDH demonstrated a broader range of pH optimum than EDH
in the oxidation direction. (Figure 4). The stability of PseDH
and EDH regarding the pH of the buffer was studied in a range
between pH 5.0 and 9.5 (Figure 5). Depending on the pH of the
storage buffer PseDH was stable in a pH-range between 6.0 and
8.0 (t1/2113-193 h). Outside this pH-range the stability of PseDH
dropped rapidly, to under 10 h. Compared to PseDH, the
stability of EDH was shifted toward basic pH-values between
7.5 and 8.5 (t1/234-21 h). Outside this range the stability of EDH
dropped quickly.
ACS Paragon Plus Environment

Page 3 of 13
ACS Catalysis
EDH (reduction)
PseDH (reduction)
PseDH (oxidation)
EDH (oxidation)
1
2
3
4
5
6
7
8
9
15
12
9
18
15
12
9
6
5
4
3
2
1
0
6
5
4
3
2
1
0
6
6
3
3
0
0
4
5
6
7
8
9
4
5
6
7
8
9
10 11 12
6
7
8
9
10
11
12
5
6
7
8
9
10 11 12
pH
pH
pH
pH
Figure 4: PseDH and EDH activity (SA) as a function of pH in reduction and oxidation reactions using the following buffer solutions at a
concentration of 100 mM. Symbols: (●) citrate-phosphate buffer, (■): potassium phosphate buffer and (▲): glycine-NaOH buffer. The
activities were followed using 1-phenyl-1,2-propanedione for the reduction assay and using (+)-(S,S)-Pseudoephedrine and (-)-(R,S)-
Ephedrine as substrates for PseDH and EDH, respectively.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Based on the results illustrated in Figure 5. the optimal storage
pH was 7.5, at which both dehydrogenases had the highest
stability. Regarding the activity and the stability of both
dehydrogenases, within a reasonable pH range they are useful
for sequential redox-reaction cascades, which require different
pH values appropriate to those of PseDH and EDH.
of benzaldehyde and pyruvic acid using variants of pyruvate
decarboxylase from Acetobacter pasteurianus.22,23
Hence, the results in the presented study indicated that both
enzymes performed the reduction of aromatic ketones with a
strict regioselectivity for the first carbonyl function on the α-
position close to the aromatic ring. After full conversion,
PseDH reduced 1a to (S)-phenylacetylcarbinol [(S)-PAC] with
excellent enantioselectivity (ee >99%), whereas (R)-
phenylacetylcarbinol [(R)-PAC] resulted with high
enantioselectivity (ee >99%) when EDH was employed
(Scheme 1).
PseDH
EDH
250
200
150
100
50
40
30
20
10
0
The reduction with EDH followed the Prelog rule, whereas
PseDH exhibited a strict anti-Prelog enantioselectivity. The
catalysed reduction of diketones using both enzymes was
irreversible. Moreover, there was no further reduction to the
diols. Both enzymes are active on broad range of different aryl-
alkyl ketones e.g. haloketones, ketoamines, diketones and
ketoesters. Furthermore, they accepted various several types of
aryl group including phenyl-, pyridyl-, thienyl-, and furyl-rings.
The calculated activity of the catalysed reduction of ketones 1a-
26a was determined by monitoring NADH depletion and is
presented in Table 1. The resulting products were identified
chromatographically via HPLC and GC (See Figure S3 in
Supporting Information). The explored ketones are
representative of symmetric and asymmetric aryl-aryl
diketones, asymmetric aryl-aryl ketones, and aryl-alkyl
ketones. The results revealed that the presence of an aromatic
ring is essential for the activity of both dehydrogenases.
However, the smallest aryl-alkyl ketone acetophenone 26a was
not accepted as substrate with either enzymes. However, the
presence of halogens in the halogenated acetophenones 15a and
16a resulted in activity with EDH. The longer propyl chain of
11a resulted in a lower activity with EDH compared to the
model substrate. Further substrates, such as aryl β-diketone 12a
and aryl ketonitrile 22a, were reduced only by EDH.
0
5
6
7
8
9
10
5
6
7
8
9
10
pH
pH
Figure 5: The stability of the PseDH and EDH based on different
pH values between 5.0-9.5 at 25 °C employing 100 mM of
following buffer systems. Symbols: (●): citrate-phosphate buffer,
(■): potassium phosphate buffer and (▲): glycine-NaOH buffer.
The enzymatic activity was determined spectrophotometrically
using the reductive assay after adding the substrate 1-phenyl-1,2-
propanedione. Deactivation curves are given in Supporting
Information Figure S2.
Substrate specificity and stereoselectivity of PseDH and
EDH. The cofactor preference for both PseDH and EDH was
examined using the asymmetric reduction of aryl-aliphatic
ketones and by the oxidation of isomers of ephedrine as model
reactions both dehydrogenases displayed no apparent activity
with NADP/(H) and full catalytic activity with NAD/(H). The
preference for NADH makes the utilization of both
dehydrogenases PseDH and EDH more attractive for industrial
applications, since NADH is more stable and less expensive
than NADPH under operational reaction conditions.21
The biotransformation of various aryl-aliphatic ketones was
explored separately with both recombinant dehydrogenases in
100 mM phosphate buffer at pH 7.5 and 25 °C. As a model
substrate we chose the aromatic α-diketone 1-phenyl-1,2-
propanedione 1a, due to its high reactivity and appropriate
solubility in water. Furthermore, its stereoselectively reduced
products are important building blocks in industrial
applications. For example, (S)-PAC was obtained with up to ee
97% and 78% yield purity and with predominant impurities of
benzaldehyde and the regioisomer of PAC, 2-
hydroxypropiophenone (HPP), via an enzymatic C–C coupling
In the case of PseDH there was no activity with substrates 14a,
15a, 11a, 12a or 22a, suggesting that the presence of a
functional group on the alkyl chain, such as an amine, certain
halogen, or ketone, is crucial for the activity of this enzyme.
Thus, PseDH could reduce various halogenated substrates, 3a,
4a, 5a and 16a. In the case of double halogenated substrate 5a
PseDH reduced it with a 0.2-fold lower specific activity than for
1a. The ketoamine 10a was reduced to the corresponding (S)-
and (R)-α-alcoholamines with 0.17-fold and 0.40-fold lower
specific activity than the model substrate, using PseDH and
EDH respectively.
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 13
Table 1: Substrate spectrum of PseDH and EDH for the reduction of several aromatic ketones with the determined specific
activity (SA) and enantiomeric excess.
1
2
3
Substrate
PseDH
EDH
4
5
6
7
8
9
SA (Unit mg-1)
55.47
ee %
SA (Unit mg-1)
64.02
ee %
O
O
O
O
1a
2a
3a
4a
1-Phenyl-1,2-propanedione
Methyl benzoylformate
99 (S)
99 (R)
O
O
Cl
160.12
16.13
5.41
99 (S)
n.d.
46.56
35.21
8.00
98 (R)
n.d.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
O
Ethyl 2-chlorobenzoylacetate
2-Chloro-1-phenyl-1-propanone
O
n.d.
n.d.
Cl
O
5a
6a
7a
8a
9a
10a
2-Bromo-1-(3-chlorophenyl)-1-
propanone
12.58
9.52
n.d.
5.10
n.d.
Cl
Br
O
1,2-Indandione
1,2-Naphthoquinone
Isatin
n.d.
50.05
13.89
7.84
n.d.
O
O
21.96
66.99
9.49
n.d.
n.d.
O
O
89 (S)
99 (S)
n.d. (S)
52 (R)
99 (R)
95 (R)
O
N
O
2-Aminoacetophenone
4-Chlorobenzil
25.79
6.50
N
Cl
5.26
O
O
O
11a
12a
13a
14a
15a
16a
17a
18a
1-Phenyl-1-propanone
1-Phenyl-1,3-butanedione
Phenylglyoxal
n.a.
n.a.
1.55
13.64
12.10
4.45
8.71
9.44
2.81
1.74
99 (R)
98 (R)
99 (R)
99 (R)
98 (R)
85 (R)
24 (R)
70 (S)
O
O
n.a.
n.a.
O
O
O
5.51
n.a.
n.d.
O
3-Chloropropiophenone
2-Chloroacetophenone
2-Bromoacetophenone
Phenyl-2-pyridinylmethanone
n.a.
Cl
n.a.
n.a.
Cl
Br
O
3.75
0.04
0.20
37 (S)
5 (S)
60 (R)
O
N
O
(4-Chlorophenyl)-2-
pyridinylmethanone
N
Cl
ACS Paragon Plus Environment

Page 5 of 13
ACS Catalysis
19a
Benzil
O
2.40
99 (S)
47.06
99 (R)
1
2
O
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
O
20a
21a
22a
23a
2,2'-Furil
0.66
0.58
n.a.
99 (S)
99 (S)
n.a.
19.95
46.77
6.87
99 (R)
99 (R)
99 (R)
99 (R)
O
O
S
O
O
2,2'-Thenil
S
S
O
O
2-Thenoylacetonitrile
4,4′-Difluorobenzil
N
F
1.50
99 (S)
48.20
O
O
F
24a
25a
26a
2,2′-Dichlorobenzil
4,4′-Dimethylbenzil
Acetophenone
10.08
97 (S)
99 (S)
n.a.
27.91
3.12
n.a.
94 (R)
99 (R)
n.a.
Cl
O
O
O
Cl
2.93
n.a.
O
O
Both PseDH and EDH reduced the aryl-alkyl α-ketoester 2a
with a higher rate than 1a. Asymmetric aryl-aryl ketones 17a
and 18a, and symmetric bulky-bulky α-diketones 19a, 20a,
21a, 23a, 24a and 25a could be also reduced by both
dehydrogenases. Compared to other dehydrogenases, such as
the NADPH-dependent RADH from Ralstonia sp.,24 PseDH
is strictly regiospecific for the α-diketone and therefore there
is no need to use the alpha-hydroxy ketone as a starting
material to obtain the desired (S)-alcohol as the product. There
is, to our knowledge, no other reported ADH with the activity
described here, which transforms 1-phenyl-1,2-propanedione
to (S)-PAC. While there is a considerable number of alcohol
dehydrogenases for the reduction of prochiral ketones to
secondary alcohols,25-28 they do not usually exhibit such
diverse substrate spectra as those observed for PseDH and
EDH. Those two novel NADH dependent carbonyl reductases
with broad substrate specificity suggest that they have
excellent potential for the production of key materials in the
pharmaceutical industry, including (+)-Pseudoephedrine,
Ezetimibe, Doluxitine and Merabegron.
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Although oxidation of the hydroxyl group using either PseDH
or EDH was not be detected for α-hydroxyketone [(R)-PAC],
both enzymes could stereoselectively oxidise α-
alcoholamines such as the isomers of ephedrine.
Moreover, the novelty of those two dehydrogenases is
demonstrated through their strict regio- and enantioselectivity
in the reduction of α-diketones such as 1a to (S)-PAC, (ee
>99%) and (R)-PAC (ee >99%) using PseDH and EDH,
respectively. No isomers, such as 2-hydroxypropiophenone,
or any diols were observed after full conversion. Aromatic α-
ketoamines could also be reduced to the corresponding (S)-
and (R)-alcoholamines with enantioselectivity of 99% using
PseDH and EDH, respectively (Table 1). In the case of α-
haloketones, the enantioselectivity was more dependent on the
nature of the halogen atom. Whereas the enantiomeric excess
of 1-phenyl-2-chloroethanol was ee >99% for both enzymes,
1-phenyl-2-bromoethanol was produced with
a lower
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 13
enantioselectivity ee 37% and ee 85% with PseDH and EDH,
1
2
3
4
5
6
respectively. According to this result there is presumably a
steric or electronic hindrance with both enzymes using the
larger halogen atom. A ketoreductase from Bacillus sp.
ECU0013 also exhibited contrary behaviour toward
chlorinated and brominated aryl ketones, resulting in higher
enantiomeric excess with brominated ketones.29 The bulky-
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
bulky symmetric α-diketones and α-ketoesters were reduced
enantioselectively with ee>99% with both enzymes. The
absence of a functional group on the alpha position to the
carbonyl group did not affect the enantioselectivity with EDH,
which was still high, with ees of 99%, 98%, and 99%
observed for (S)-1-phenyl-1-propanol, (S)-4-hydroxy-4-
phenyl-2-butanone,
and
(S)-3-(2-thienyl)-3-
hydroxypropanenitrile, respectively. By comparison, the
ketoreductase KRED1-Pglu from Pichia glucozyma displayed
lower selectivity than EDH for the reduction of β-ketonitrile
22a.30 The reduction of the asymmetric bulky-bulky ketone
10a revealed more details about the affinity toward
halogenated aromatic rings. Based on the reference materials,
EDH had a significant affinity for non-halogenated aromatic
rings, whereas PseDH accepted both rings without significant
preference. In the case of non-symmetrical aryl-aryl ketone
17a both PseDH and EDH displayed higher affinity towards
the benzene ring than the pyridine ring. The substitution of the
benzene ring with a chloride atom inverted the substrate-
protein interaction by both dehydrogenases resulting in a
higher preference toward the pyridine ring. Thus, (R)-alcohol
resulted from using PseDH (ee 60%) for the reduction of 18a.
In contrast, EDH reduced the same substrate 18a to the (S)-
alcohol with ee 70%, which was higher than that reported for
a
commercial ketoreductase from Merck Research
Laboratories (ee 60%).31
The functional versatility and strict enantioselectivity of
Figure 6. Structure of PseDH determined to a resolution of 1.83 Å. A: Tetrameric structure of PseDH, with monomers shown in ribbon
PseDH and EDH are noteworthy, since different sterically
format in gold, blue, coral and green; B: Structure of monomer of PseDH in complex with NAD+ in cylinder format with carbon atoms in
demanding ketones are reduced to the corresponding
grey; C: Structure of PseDH-NAD+ complex modelled with 1a (carbon atoms in coral) in the active site. The -carbonyl group is shown
secondary alcohols with high regio- and enantiomeric
making interactions with the side chains of S141, S143 and Y155 at distances of 3.0, 3.0 and 3.6 Å respectively. The -carbonyl group is
excesses.
shown interacting with the side chains of S143 at a distance of 3.3 Å. The side chain of S143 is in two conformations.
code 3AWD;33; 34% sequence ID; rmsd 1.6 Å over 257 C-α
atoms) and the human retinal dehydrogenase (PDB code
1YDE;34; 30% sequence ID rmsd 1.3 Å over 254 C-α atoms).
Superimposition of the PseDH monomer with 3AWD revealed
few significant differences in secondary or tertiary structure,
save for an extended loop in the region A197 to S207. As
expected, PseDH contained the typical catalytic tetrad of the
superfamily SDR consisting of N113, S141, Y155 and K159
residues.35-37 K159 interacts with the nicotinamide ribose of the
cofactor and decreases the pKa of the -OH group of Y155
initiating the hydrogen transfer. Y155 acts as catalytic base and
Crystal structure of PseDH. PseDH was crystallised in the
presence of both NAD+ and substrate 1a, and the structure was
solved in complex with the cofactor using a monomer of a
gluconate 5-dehydrogenase from Thermotoga maritima (PDB
code 1VL8, 37% sequence ID) as the model. The crystals were
in space group C2221 and there were six monomers,
representing one-and-a-half tetramers, in the asymmetric unit.
The tetramer is shown in Figure 6A and the monomer in Figure
6B. Analysis of the monomer structure using the DALI server32
suggested that the most similar structures were a short-chain
dehydrogenase gox2181 from Gluconobacter oxydans (PDB
ACS Paragon Plus Environment

Page 7 of 13
ACS Catalysis
S141 stabilizes the substrate. N113 interacts with K159 via a may be expected, and also S143 as well as W152 are substituted
1
2
3
4
5
6
7
8
water molecule to create a network of hydrogen bonds between
cofactor, catalytic side-chains and water molecules.38 Although
peaks of electron density were observed in the difference map
adjacent to the nicotinamide ring of NAD+, these were not
sufficient for modelling the substrate 1a. 1a was therefore
docked into the active centre of PseDH using Autodock Vina.
The model shows the expected interactions between the α-
carbonyl group to be reduced and the side chains of Y155,
which is thought to be the proton donor to the nascent alkoxide
in the reduction reactions, and to S141 which appear to anchor
the carbonyl group in the correct orientation for reduction.35-38
The β-carbonyl group also makes an interaction with the side
chain of S143 (Figure 6C).
in the (R)-selective EDH by an alanine and a tyrosine residue
respectively (Figure S5, S6 in Supporting Information).
A comparison with 30 structures of SDRs from bacterial species
in the PDB suggests that S143 is most commonly a small
hydrophobic residue, such as G, A or V, whereas W152 is most
commonly an H, M or L, although in SDRs from Burkholderia
vietnamiensis (5IF3) and Mycobacterium smegmatis (3PK0) the
residue is W.39 However, there are no reports of chiral ketone
reductions by either of these enzymes that might confirm a W
in this position as essential for (S)-PAC stereopreference.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Further comparison of both dehydrogenases with other well-
studied anti-Prelog specific ADHs, for exampleLbADH and
LkADH, as well as the Prelog specific ADHs, PED and
RasARH, is provided in Supporting Information (Table S1,
S2). Among these ADHs, the amino acid sequence of PseDH
demonstrates the highest relevance to EDH and RasADH (Id
32% and 31.5%), respectively. EDH shares higher identity with
PED (47%). Interestingly, S143 in PseDH is replaced with polar
residue E in LbADH and LkADH. The other prelog specific
ADHs share a hydrophobic residue A analogous to this position.
Moreover, D149 is conserved among the anti-Prelog specific
ADHs, whereas this position is filled with hydrophobic
residues, such as I and L in the Prelog specific ADHs (Figure
S5).
The poor activity of PseDH with non-functionalised
acetophenone derivatives such as 1-phenyl-1-propanone 11a
could be a consequence of the inhibited interaction between the
β-substituent and the side chain of S143. 1a is positioned within
the active site such that delivery of the pro-(S) hydride from
NADH will be to the (re)-face, resulting in the (S)-enantiomer
of Phenylacetylcarbinol being formed as the product, as
determined experimentally.
The side chain of W152 forms a stacking interaction with the
aromatic ring of 1a. Although the active site residues that
determine enantioselectivity in PseDH and in EDH have not yet
been determined, the amino acid sequence alignment of PseDH
and in EDH shows that, while Y155 and S141 are conserved, as
Table 2: Steady-state kinetic parameters of different substrates for PseDH and EDH. Non-linear regressions of the kinetic data
are given in Supporting Information Figure S4.
PseDH
EDH
Substrate
Km [mM]
Ki [mM]
kcat [s-1]
Km [mM]
Ki [mM]
kcat [s-1]
1a
2a
3.12±0.23
7.01±0.23
34.24±1.02
109.18±2.34
0.06±0.01
0.19±0.02
30.43±1.18
40.05±3.05
1.46±0.22
3a
3.94±0.37
8.74±0.66
12.71±1.25
0.93±0.15
0.05±0.01
0.81±0.07
0.14±0.01
0.05±0.00
9.89±0.39
3.8±0.21
0.06±0.01
0.02±0.00
0.00±0.00
2.29±0.05
0.1±0.02
0.51±0.07
0.6±0.2
0.16±0.04
0.1±0.02
37.38±6.73
7.31±1.02
2.54±0.05
31.11±2.66
7.94±0.45
4.36±0.22
27.13±8.13
3.48±0.11
0.77±0.01
15.37±11.74
9.49±0.83
15.74±0.67
4a
5a
9.52±0.53
6.55±0.43
12.76±0.64
43.95±1.43
5.44±0.22
3.75±0.22
6a
7a
8a
9a
23.05±3.49
1.53±0.8
10a
11a
12a
13a
NADH
0.17±0.01
0.01±0.00
0.2±0.17
5.67±1.08
0.07±0.01
0.26±0.27
EDH, respectively, indicating weaker affinity of PseDH for
this substrate. Furthermore, EDH exhibited substrate
inhibition with some substrates, including 2a, 3a, 4a and 9a.
Kinetic parameters of PseDH and EDH. As a basis for
future reaction design, kinetic parameters for the target
substrates 1a-13a were determined (Table 2). The results of
the kinetic studies demonstrate a dependence of Michaelis-
Menten constants (Vmax and Km) on the steric and electronic
properties of the tested substrates.
These substrate inhibitions were also observed during the
oxidation
of
(-)-(R,S)-ephedrine
and
(-)-(R,R)-
pseudoephedrine with EDH, and may have a significant
physiological role in the feedback regulation by the redox-
reactions in the metabolism of ephedrine (data not published
yet). However, both enzymes exhibited high substrate affinity
for the tested ketones, since the Km values were between 0.01
According to these data EDH mostly exhibited lower Km and
Vmax values compared to PseDH. The Km values of the model
substrate 1a were 3.12 mM and 0.06 mM for PseDH and
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 13
mM and 8.74 mM. The maximal reaction velocity was (600 µM). Kinetic parameters (Km and Vmax) were calculated
1
2
3
4
5
6
7
8
observed in reductions of the aryl α-ketoester 2a with both
enzymes.
by fitting to experimental data using nonlinear regression in
GraphPad Prism 8 (ε340 0.00622 L μmol−1 cm−1). All
substrates were diluted in an appropriate amount of DMSO.
The concentration of the purified enzymes was determined at
280 nm using a NanoDrop® ND-1000 (ThermoFisher
Scientific, Bremen, Germany).
Conclusions
The novel dehydrogenases PseDH and EDH are promising
tools for green chemistry, as they permit for the first time the
oxidation of the isomers of ephedrine and the regio- and
enantioselective reduction of sterically demanding substrate
1a to give the desired (S)-PAC and (R)-PAC. Both PseDH and
EDH belong to the family of SDRs and use NADH as
cofactor. Based on the original substrate ephedrine the
substrate spectrum of these enzymes was extended to
arylaliphatic haloketones, diketones, ketoesters, aryl-aryl
ketones and bulky-bulky α-diketones. (Scheme 2). Their
excellent stereoselectivity, coupled with their broad substrate
specificity, makes them especially attractive biocatalysts for
industrial applications.
Enzyme preparation. The genes coding for PseDH and EDH
were overexpressed in E. coli T7 Shuffle (DE3) after insertion
into plasmid pET19b. Resulting enzymes were equipped with
an N-terminal 10X histidine –tag.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Gene overexpression and protein purification. The main
cultures of 500 mL TB medium containing 100 µg mL-1
ampicillin were inoculated with an overnight starter culture
and incubated at 30 °C with shaking at 200 rpm, until an
optical density of 0.8 (measured at 600 nm) was reached.
Subsequently, the cultures were induced for gene
overexpression with 100 µM and 30 µM Isopropyl-β-D-
thiogalactoside (IPTG) for PseDH and EDH, respectively.
After overnight overexpression at 20 °C with shaking at 180
rpm, the cells were harvested by centrifugation (4 °C, 8000 x
g, 10 min) and stored at -20 °C. The cell pellets were
resuspended with 20 mL of 25 mM potassium phosphate
buffer pH 7.5 containing 500 mM NaCl and 50 mM
imidazole.
O
OH
PseDH
R3
R1
R3
R1
R2
NAD+
NADH
R2
The disruption of the cells was carried out on ice by
ultrasonication for 3 x 1 min with 1 min breaks (Sartorius
Labsonic M, Settings:Cycle 0.6, Amplitude 100%). After
O
OH
R3
EDH
NADH
R1
R3
R1
NAD+
addition of
1
µg mL-1 DNAseI (AppliChem
R2
R2
GmbH,Darmstadt, Germany) the crude extracts were
centrifuged twice at 20,000 x g for 30 min at 4 °C.
R1 = Ar
R2 = G, O, N, X
R3 = H, Ar, Alk
The clear lysate was loaded onto a 5 mL Histrap FF column
charged with nickel to purify the enzymes. The column was
washed with 120 mM imidazole for 5 column volumes, after
which the enzymes were eluted with a linear gradient up to
500 mM imidazole.
Scheme 2: Asymmetric hydrogen transfer for a wide range
of substrates.
Due to the observed inhibition by some substrates, a suitable
reaction design should be carried out to achieve the full
potential of the enzymes. Moreover, the determined enzymes’
stabilities are a good starting point for further stabilisation
studies, crystal structure of PseDH also provides a valuable
platform on which to base engineering studies directed
towards the further broadening of substrate specificity and
improving the stability of this enzyme.
The elution buffer was then exchanged with storage buffer of
100 mM potassium phosphate buffer pH 7.5 using a desalting
column (PD-10, Sephadex G-25 M, GE Healthcare).
SDS- and blue Native-PAGE. SDS-PAGE was performed
by using the procedure of Laemmli40 with a 12% separation
gel. The SDS-PAGE electrophoresis was carried out for 1 h at
25 °C and 180 V. Blue Native-PAGE was conducted
according to a modified protocol of Schägger41 with a gradient
gel (4-16%). The electrophoresis was performed for 4 h at 15
°C and 120 V. The loaded protein amounts on all gels were
ca. 13 µg. The molecular masses of subunits and oligomers
were determined by the Rf values driven from the marker
proteins. The calibration curves were obtained using
molecular mass marker ColorPlus™ Prestained Protein
Ladder (New England Biolabs) and Serva native marker
(SERVA Electrophoresis) for SDS-PAGE and Blue Native-
PAGE, respectively.
Material and Methods
Chemicals, media and bacterial strains. Unless otherwise
stated, all chemicals and oligonucleotides were purchased
from Sigma-Aldrich Merck (St Louis, MO, USA). Molecular
biotechnological tools and E. coli strains were purchased from
New England Biolabs GmbH (Frankfurt am Main, Germany).
Cofactors and media components were purchased from Carl
Roth GmbH & Co. (Karlsruhe,Germany).
Steady-state kinetic assays. Initial activities were
determined photometrically at 340 nm for the reduction of
ketones using a Cary 60 UV-Vis spectrophotometer (Agilent,
Waldbronn, Germany) following the absorbance decrease of
NADH. Kinetic parameters were measured at 25 °C in 1 mL
of 100 mM potassium phosphate buffer pH 7.5. The reduction
assay mixture contained 250 µM NADH, variable
concentrations of candidate ketones and an appropriate
concentration of the enzymes. Initial rate data for cofactor
NADH could be determined only for EDH by fixing the
concentration of the diketone 1-phenyl-1,2-propanedione
In-gel activity assay. A modified protocol of Wittig42 was
performed to show the active migrated bands of Native-
PAGE. The gel was washed twice for 30 min at 25 °C with 50
mL of 100 mM phosphate buffer pH 7.5. For staining, the gel
was incubated overnight at 25 °C in 50 mL 100 mM phosphate
buffer pH 7.5 containing 2.5 mM NAD+, 0.5 mM (-)-
ephedrine, 0.5 mM (+)-pseudoephedrine and 1.5 mM
nitrotetrazolium blue chloride (NTB).
Product analytics. The extraction of phenylethanolamine
was performed under basic conditions and with equal volumes
ACS Paragon Plus Environment

Page 9 of 13
ACS Catalysis
of ethyl acetate. The resulting stereoisomers of yield a clear lysate. The lysate was loaded onto a 5 mL
1
2
3
4
5
6
7
8
phenylethanolamine were separated using a chiral column
CycloSil-B 30 m length, ID 0.25 mm, film thickness 0.25 µm
(Agilent Inc. USA) on a gas chromatography system
(Shimadzu GC2010) with a flame ionisation detector and
nitrogen as carrier gas, applying a temperature gradient for 15
min from 130-190 °C with a ramp of 4 °C min-1. Retention
times for 2-aminoacetophenone, (S)-(+)- phenylethanolamine
and (R)-(-)- phenylethanolamine were 11.5 min, 13.1 min, and
13.3 min, respectively.
HisTrap™ FF column charged with nickel. PseDH was eluted
using a gradient of 50 – 500 mM imidazole over 10 column
volumes. Peak fractions containing PseDH, as determined by
SDS-PAGE were pooled and concentrated to a volume of 2
mL. The concentrated protein was loaded onto a HiLoad
16/60 Superdex™ 75 Prep Grade gel filtration column and
eluted using the buffer. Fractions containing PseDH were
pooled, concentrated to 20 mg mL-1 and stored at 4 °C for
crystallisation trials.
9
All other products were extracted with n-hexane and analysed
using HPLC (Knauer GmbH, Berlin, Germany) with a
normal-phase CHIRALPAK IB-column 4.6×250 mm, particle
size: 5 µm (Daicel Chemical Industries, Ltd). The mobile
phase n-hexane:isopropanol was applied with appropriate
ratios and a flow rate of 1 mL min-1. The absorbance of the
products was detected at wavelengths between 220-270 nm.
For further details about HPLC methods see Table S3 in the
Supporting Information. Chiral separations are demonstrated
in Figure S3A-AZ. The absolute configurations of the
products were determined via polarimeter model 343
(PerkinElmer, Inc., USA) and by analogy to other authentic
standards of S- and R-Benzoin and -Furoin. In order to
determine specific rotation values of the reduced products (4-
Pure PseDH, at a concentration of 2 mg mL-1, was subjected
to crystallisation trials using a range of commercially
available screens in 96-well plates using 300 nL drops with
150 nL protein solution and 150 nL of precipitant solution.
The most promising hits were obtained in conditions
containing 0.1 M Bis-tris buffer pH 5.5 and 25% (w/v) PEG
3350 and 200 mM magnesium chloride. Larger crystals were
grown using the hanging-drop vapour diffusion method in 24-
well plate Linbro dishes and using crystallisation drops of 1
µL enzyme solution (at a concentration of 20 mg mL-1) plus 1
µL mother liquor. Superior crystals were obtained in drops
containing 0.1 M Bis-tris buffer pH 5.5 and 25% (w/v) PEG
3350 and 200 mM magnesium chloride with 5 mM NADH
and in the presence of 5 mM 1a. Crystals were flash-cooled in
liquid nitrogen without the addition of a cryoprotectant, and
tested for diffraction using a Rigaku Micromax-007HF fitted
with Osmic multilayer optics and a MARRESEARCH
MAR345 imaging plate detector. The crystals had the space
group C2221. Crystals that displayed diffraction of better than
3 Å resolution were retained for data collection at the
synchrotron.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
chlorophenyl)-2-pyridinylmethanone
and
phenyl-2-
pyridinylmethanone at 586 nm and 25 °C in dichloromethane
at concentration of 10 mg mL-1. The specific values are
presented in Table S4 (Supporting Information).
The identification of the reduced products (S)-/(R)-
phenylacetylcarbinol was performed using GC-MS after 40
mL reaction assay (Agilent 6890N GC / HP 5973N MSD
equipped with an Optima 35 MS column). A constant
temperature program at 130 °C for 30 min was applied.
substrate and product peaks were detected at 9.89 min and
11.08 min, respectively. (Figure S7, S8 Supporting
Information).
1H-NMR spectroscopy (400 MHz, Bruker DRX-500) was
used to analyse the reaction products (S)-phenylacetylcarbinol
and (R)-phenylacetylcarbinol using chloroform-d (1H: δ7.26).
Resulting spectra are provided in the Supporting Information
(Figure S9, S10).
Data collection and processing, structure solution and
refinement for PseDH. A dataset for a crystal of PseDH in
complex with NADH was collected on beamline I04 at the
Diamond Light Source Synchrotron in Oxford, U.K. Data,
which were collected to 1.83 Å, were processed and integrated
using XDS43 and scaled using SCALA44 as part of the Xia2
processing system.45 Data collection statistics are given in
Table 3. The structure was solved with the program
MOLREP,46 using
a monomer of the gluconate 5-
dehydrogenase from Thermotoga maritima (PDB code 1VL8)
as a model. The solution contained six molecules in the
asymmetric unit, consisting of one-and-a-half tetramers. The
structure was built and refined using iterative cycles within
the programs COOT47 and REFMAC48 respectively. After
building the protein and water molecules, clear density was
observed for the cofactor NADH in five out of the six
monomers. The structure was refined to Rcryst and Rfree
values of 16.1 and 18.1% respectively, and was validated
upon deposition within the Protein DataBank (PDB).
Refinement statistics are presented in Table 1. The
Ramachandran plot for the structure showed 96.6% of
residues to be situated in preferred regions, 3.0% in allowed
regions and 0.4 % outliers. Coordinates for the PseDH NADH
complex structure have been deposited in the PDB with the
accession code 6QHE.
Crystallisation of Pseudoephedrine dehydrogenase. For
the purposes of crystallisation, pure PseDH was generated in
the following way: E. coli BL21 (DE3) cells were first
transformed with the plasmid pET19b containing the PseDH
gene. The cultures were spread-plated onto LB agar
containing 30 mg mL-1 ampicillin and grown overnight at 37
°C. Single colonies were used to inoculate 10 mL starter
cultures of LB medium containing 30 mg mL-1 ampicillin.
These starter cultures were incubated at 37 °C overnight with
shaking and each was then used to inoculate 500 mL cultures
of LB medium containing 30 mg mL-1ampicillin, which were
then incubated at 37 °C with shaking at 180 rpm until the
optical density A600 reached 0.5. At this point gene
expression was induced by the addition of 1 mM IPTG, after
which the cultures were incubated at 16 °C overnight. Cells
from 2 L combined culture were harvested by centrifugation
at 4225 x g for 20 min using a Sorvall RC5B centrifuge and
resuspended in 50 mL of 50 mM Tris buffer (pH 7.0),
containing 100 mM NaCl, 20% (v/v) glycerol and 0.1% (v/v)
Tween 20 with 20 mM imidazole (‘buffer’). The cells were
disrupted using ultrasonication at an amplitude of 14,000
microns for 3 x 1 min with 1 min breaks at 4°C, after which
the suspension was centrifuged at 26,892 x g for 30 min to
Table 3. Data Collection and Refinement Statistics for
PseDH. Numbers in brackets refer to data for highest
resolution shells.
PseDH
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 13
Beamline
Diamond I04
NAD+ B (Å2)
29
1
2
3
4
5
Docking
Wavelength (Å)
Resolution (Å)
0.97950
Automated docking was carried out using AUTODOCK
VINA 1.1.2.49 The structure of PseDH was prepared using the
AUTODOCK utility script. The coordinates for 1a were
prepared using ACEDRG within ccp4.50 A monomer model
of PseDH was used, with the appropriate pdbqt file prepared
in AUTODOCK Tools. The active site of PseDH was
contained in a grid of 24 x 24 x 24 with 0.375Å spacing,
centred around the NAD+ nicotinamide ring and was
generated using AutoGrid in the AUTODOCK Tools
interface. The number of runs for the genetic algorithm was
set to 10 and other docking parameters were set to default
values. Docking was performed by VINA, and so the posed
dockings were below 2Å rmsd. The results were visualised in
AUTODOCK Tools 1.5.6, where ligand conformations were
ranked based on VINA energy, but also criteria established by
the known mechanism of short-chain ADHs7 and also the
experimentally-determined enantioselectivity of PseDH for
substrate 1a. These dictated that the only poses considered
were those in which the carbonyl of 1a interacted with the
phenolic hydroxyl of the catalytic Y155 and the conserved
active site serine S141, and in which the (re)-face of the
carbonyl of 1a was presented to the C4 atom of the
nicotinamide ring of NAD+, and thus resulting in the
experimentally observed (S)-PAC product.
52.92-1.83 (1.86-1.83)
6
7
8
9
Space Group
Unit cell (Å)
C2221
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a = 80.73; b = 119.04; c =
348.58
 =  =  = 90.00°
No. of molecules in the
asymmetric unit
6
Unique reflections
Completeness (%)
Rmerge (%)
147773 (7240)
100.0 (100.0)
0.06 (0.75)
ASSOCIATED CONTENT
Rp.i.m.
0.04 (0.42)
Supporting Information. HPLC product detection list. HPLC,
GC, NMR and GC-MS chromatograms, raw spectrophotometric
data of stability, and kinetic parameters. Polarimetric data. Chart
of ion effect. This material is available free of charge via the
Internet at http://pubs.acs.org.
Multiplicity
8.1 (8.2)
16.8 (2.6)
<I/(I)>
AUTHOR INFORMATION
Corresponding Author
*E-mail: marion.ansorge@tu-dresden.de,
gideon.grogan@york.ac.uk
Overall B factor from Wilson 25
plot (Å2)
CC1/2
1.00 (0.84)
Notes
The authors declare no competing financial interest.
Rcryst/ Rfree (%)
16.2/18.3
International patent application WO 2019/002459 A1
(published on 3rd January 2019) includes protection of
enzymes EDH and PseDH for commercial exploitation.
r.m.s.d 1-2 bonds (Å)
r.m.s.d 1-3 angles (o)
0.013
1.60
Acknowledgements
We thank the Friedrich-Ebert-Stiftung e.V., Bonn, Germany
for generous personal support of T.S. We thank the Brazilian
Government for a fellowship to L.B. under the Coordination
for the Improvement of Higher Education Personnel (CAPES)
scheme.
Avge main chain B (Å2)
Avge side chain B (Å2)
29
34
References
(1) Sheldon, R. A.; Woodley, J. M. Role of biocatalysis in
sustainable chemistry. Chem. Rev. 2018, 118, 801–838.
(2) Moore, J. C.; Pollard, D. J.; Kosjek, B.; Devine, P. N. Advances
in the enzymatic reduction of ketones. Acc. Chem. Res. 2007, 40,
1412–1419.
(3) Wildeman, S. M. D.; Sonke, T.; Schoemaker, H. E.; May, O.
Biocatalytic reductions: from lab curiosity to “first choice. Acc.
Chem. Res. 2007, 40, 1260–1266.
Avge water B (Å2)
37
ACS Paragon Plus Environment

Page 11 of 13
ACS Catalysis
(4) Simon, R. C.; Mutti, F. G.; Kroutil, W. Biocatalytic synthesis
of enantiopure building blocks for pharmaceuticals. Drug. Discov.
Today. Technol. 2013, 10, 37–44.
(5) Goldberg, K.; Schroer, K.; Lütz, S.; Liese, A. Biocatalytic
ketone reduction-a powerful tool for the production of chiral
alcohols-part I: processes with isolated enzymes. Appl. Microbiol.
Biotechnol. 2007, 76, 237–248.
emulsion system with Candida utilis cells. Appl. Microbiol.
Biotechnol. 2006, 70, 170–175.
1
2
3
4
(23) Sehl, T.; Bock, S.; Marx, L.; Maugeri, Z.; Walter, L.;
Westphal, R.; Vogel, C.; Menyes, U.; Erhardt, M.; Müller, M.; Pohl,
M.; Rother, D. Asymmetric synthesis of (S)-phenylacetylcarbinol –
closing a gap in C–C bond formation. Green. Chem. 2017, 19, 380–
384.
5
6
7
8
(6) Huisman, G. W.; Liang, J.; Krebber, A. Practical chiral alcohol
manufacture using ketoreductases. Curr. Opin. Chem. Biol. 2010, 14,
122–129.
(7) Kavanagh, K. L.; Jörnvall, H.; Persson, B.; Oppermann, U.
Medium- and short-chain dehydrogenase/reductase gene and protein
families: the SDR superfamily: functional and structural diversity
within a family of metabolic and regulatory enzymes. Cell. Mol. Life.
Sci. 2008, 65, 3895–3906.
(8) Kallberg, Y.; Persson, B. Prediction of coenzyme specificity in
dehydrogenases/reductases: a hidden Markov model-based method
and its application on complete genomes. FEBS. J. 2007, 273, 1177–
1184.
(9) Oppermann, U.; Filling, C.; Hult, M.; Shafqat, N.; Wu, X.;
Lindh, M.; Shafqat, J.; Nordling, E.; Kallberg, Y.; Persson, B.;
Jörnvall, H. Short-chain dehydrogenases/reductases (SDR): the 2002
update. Chem. Biol. Interact. 2003, 143–144, 247–253.
(10) Kallberg, Y.; Oppermann, U.; Jörnvall, H.; Persson B. Short-
chain dehydrogenases/reductases (SDRs). Eur. J. Biochem. 2002,
269, 4409–4417.
(11) Hummel, W. Large-scale applications of NAD(P)-dependent
oxidoreductases: recent developments. Trends. Biotechnol. 1999, 17,
487–492.
(12) Weckbecker, A.; Hummel, W. Cloning, expression, and
characterization of an (R)-specific alcohol dehydrogenase from
Lactobacillus kefir. Biocatal. Biotransformation. 2006, 24, 380389.
(13) Kniemeyer, O.; Heider, J. (S)-1-phenylethanol dehydrogenase
of Azoarcus sp. strain EbN1, an enzyme of anaerobic ethylbenzene
catabolism. Arch. Microbiol. 2001, 176, 129–135.
(14) Lavandera, I.; Kern, A.; Ferreira-Silva, B.; Glieder, A.; de
Wildeman, S.; Kroutil, W. Stereoselective bioreduction of bulky-
bulky ketones by a novel ADH from Ralstonia sp. J. Org. Chem.
2008, 73, 6003–5.
(15) Schlieben, N. H.; Niefind, K.; Muller, J.; Riebel, B.; Hummel,
W.; Schomburg. D. Atomic resolution structures of R-specific
alcohol dehydrogenase from Lactobacillus brevis provide the
structural bases of its substrate and cosubstrate specificity. J. Mol.
Biol. 2005, 349, 801–813.
(16) Noey, E. L.; Tibrewal, N.; Jiménez-Osés, G.; Osuna, S.; Park,
J.; Bond, C. M.; Cascio, D.; Liang, J.; Zhang, X.; Huisman, G. W.;
Tang, Y.; Houk, K. N. Origins of stereoselectivity in evolved
ketoreductases. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, E7065–72.
(17) Man, H; Kędziora, K.; Kulig, J.; Frank, A.; Lavandera, I.;
Gotor-Fernández, V.; Rother, D.; Hart, S.; Turkenburg, J. P.; Grogan,
G. Structures of alcohol dehydrogenases from Ralstonia and
Sphingobium spp. reveal the molecular basis for their recognition of
‘bulky–bulky’ ketones. Top. Catal. 2014, 57, 356–365.
(18) Höffken, H. W.; Duong, M.; Friedrich, T.; Breuer, M.; Hauer,
B.; Reinhardt, R.; Rabus, R.; Heider J. Crystal structure and enzyme
kinetics of the (S)-specific 1-phenylethanol dehydrogenase of the
denitrifying bacterium strain EbN1. Biochemistry. 2006, 45, 82–93.
(19) Hoyos, P.; Sansottera, G.; Fernández, M.; Molinari, F.;
Sinisterra, J. V.; Alcántara, A. R. Enantioselective monoreduction of
different 1,2-diaryl-1,2-diketones catalysed by lyophilised whole
cells from Pichia glucozyma. Tetrahedron. 2008, 64, 7929–7936.
(20) Shanati, T.; Ansorge-Schumacher, M. Enzymes and methods
for the stereoselective reduction of carbonyl compounds, oxidation
and stereoselective reductive amination – for the enantioselective
preparation of alcohol amine compounds. WO 2019; 002459 A1.
PCT/EP2018;067407, 2019. p27.
(24) Kulig, J.; Frese, A.; Kroutil, W.; Pohl, M.; Rother, D.
Biochemical characterization of an alcohol dehydrogenase from
Ralstonia sp. Biotechnol. Bioeng. 2013, 110, 1838–1848.
(25) He, S.; Wang Z.; Zou Y.; Chen, S.; Xu, X. Purification and
9
characterization of
oxidoreduction of
Process. Biochem. 2014, 49, 1107–1112.
a
novel carbonyl reductase involved in
aromatic β-amino ketones/alcohols.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(26) Peng, G. J.; Kuan, Y.C.; Chou, H. Y.; Fu, T. K.; Lin, J. S.;
Hsu, W. H.; Yang, M. T. Stereoselective synthesis of (R)-
phenylephrine using recombinant Escherichia coli cells expressing a
novel short-chain dehydrogenase/reductase gene from Serratia
marcescens BCRC 10948. J. Biotechnol. 2014, 170, 6–9.
(27) Peng, G. J.; Cho, Y. C.; Fu, T. K.; Yang, M. T.; Hsu, W. H.
Enantioselective synthesis of (S)-phenylephrine by recombinant
Escherichia
coli
cells
expressing
the
short-chain
dehydrogenase/reductase gene from Serratia quinivorans BCRC
14811. Process. Biochem. 2013. 48, 1509–1515.
(28) Chen, X. H.; Ping, W.; Wang, X. T.; Zong, M. H.; Lou, W.
Y. A novel carbonyl reductase with anti-Prelog stereospecificity from
Acetobacter sp. CCTCC M209061: Purification and characterization.
PLOS. 2014, 9, e94543.
(29) Ni, Y.; Li, C. X.; Zhang, J.; Shen, N. D.; Bornscheuer U. T.;
Xu J. H. Efficient reduction of Ethyl 2‐Oxo‐4‐phenylbutyrate at 620
g⋅L−1 by a bacterial reductase with broad substrate spectrum. Adv.
Synth. Catal. 2011, 353, 1213–1217.
(30) Contente, M. L.; Serra, I.; Molinari, F.; Gandolfi, R.; Pinto,
A.; Romano, D. Preparation of enantiomerically enriched aromatic β-
hydroxynitriles and halohydrins by ketone reduction with
recombinant ketoreductase KRED1-Pglu, Tetrahedron. 2016, 72,
3974–3979.
(31) Truppo, M. D.; Pollard, D.; Devine, P. Enzyme-catalyzed
enantioselective diaryl ketone reductions, Org. Lett. 2007, 9, 335–
338.
(32) Holm L.; Laakso, L. M. Dali server update. Nucleic. Acids.
research. 2016, 44, W351–W355.
(33) Liu, X.; Yuan, Z.; Adam Yuan, Y.; Lin, J.; Wei, D.
Biochemical and structural analysis of Gox2181, a new member of
the SDR superfamily from Gluconobacter oxydans. Biochem.
Biophys. Res. Commun. 2011, 415, 410–415.
(34) Lukacik, P.; Keller, B.; Bunkoczi, G.; Kavanagh, K. L.;
Kavanagh, K.; Lee, W. H.; Hwa Lee, W.; Adamski, J.; Oppermann,
U. Structural and biochemical characterization of human orphan
DHRS10 reveals
a
novel cytosolic enzyme with steroid
dehydrogenase activity. Biochem. J. 2007, 402, 419–427.
(35) Jörnvall, H.; Persson, B.; Krook, M.; Atrian, S.; Gonzàlez-
Duarte,
R.;
Jeffery,
J.;
Ghosh,
D.
Short-chain
dehydrogenases/reductases (SDR). Biochemistry. 1995, 34, 6003–
6013.
(36) Tanabe, T.; Tanaka, N.; Uchikawa, K.; Kabashima, T.; Ito,
K.; Nonaka, T.; Mitsui, Y.; Tsuru, M.; Yoshimoto, T. Roles of the
Ser146, Tyr159, and Lys163 residues in the catalytic action of
7alpha-hydroxysteroid dehydrogenase from Escherichia coli. J.
Biochem. 1998, 124, 634–641.
(37) Price, A. C.; Zhang, Y. M.; Rock, C. O.; White, S. W.
Cofactor-induced conformational rearrangements establish
a
catalytically competent active site and a proton relay conduit in
FabG. Structure. 2004, 12, 417–428.
(38) Filling, C.; Berndt, K. D.; Benach, J.; Knapp, S.; Prozorovski,
T.; Nordling, E.; Ladenstein, R.; Jörnvall, H.; Oppermann, U. Critical
(21) Li, A.; Ye, L.; Wu, H.; Yang, X.; Yu, H. Characterization of
an excellent anti-Prelog short-chain dehydrogenase/reductase
EbSDR8 from Empedobacter brevis ZJUY-1401. J. Mol. Catal. B.
Enzym. 2015, 122, 179–187.
(22) Satianegara, G.; Breuer, M.; Hauer, B.; Rogers. P. L.; Rosche,
B. Enzymatic (R)-phenylacetylcarbinol production in a benzaldehyde
residues
for
structure
and
catalysis
in
short-chain
dehydrogenases/reductases. J. Biol. Chem. 2002, 277, 25677–25684.
(39) Baugh, L.; Phan, I.; Begley, D. W.; Clifton, M. C.; Armour,
B.; Dranow, D. M.; Taylor, B. M.; Muruthi, M. M.; Abendroth, J.;
Fairman, J. W.; Fox, D.; Dieterich, S. H.; Staker, B. L.; Gardberg, A.
S.; Choi, R.; Hewitt, S. N.; Napuli, A. J.; Myers, J.; Barrett, L. K.;
Zhang, Y.; Ferrell, M.; Mundt, E.; Thompkins, K.; Tran, N.; Lyons-
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 13
Abbott, S.; Abramov, A.; Sekar, A.; Serbzhinskiy, D.; Lorimer, D.;
Buchko, G. W.; Stacy, R.; Stewart, L. J.; Edwards, T. E.; Van
Voorhis, W. C.; Myler, P. J. Increasing the structural coverage of
tuberculosis drug targets. Tuberculosis. (Edinb). 2015, 95, 142–148.
(40) Lämmli, U.K. Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature. 1970, 227, 680–
685.
(41) Schägger, H.; von Jagow, G. Blue native electrophoresis for
isolation of membrane protein complexes in enzymatically active
form. Anal. Biochem. 1991, 199, 223–231.
(42) Wittig, I.; Karas, M.; Schägger, H. High resolution clear
native electrophoresis for In-gel functional assays and fluorescence
studies of membrane protein complexes. Mol. Cell. Proteomics.
2007, 6, 1215–1225.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(43) Kabsch, W. XDS. Acta. Crystallogr. Sect. D. Biol.
Crystallogr. 2010, 66, 125–132.
(44) Evans, P. R. An introduction to data reduction: Space-group
determination, scaling and intensity statistics. Acta. Crystallogr.
Sect. D. Biol. Crystallogr. 2011, 67, 282–92.
(45) Winter, G. An expert system for macromolecular
crystallography data reduction. J. Appl. Crystallogr. 2010, 43, 186–
190.
(46) Vagin, A.; Teplyakov, A. MOLREP: An automated program
for molecular replacement J. Appl. Cryst. 1997, 30, 1022–1025.
(47) Emsley, P.; Cowtan, K. Model-building tools for molecular
graphics. Acta. Crystallogr. Sect. D. Biol. Crystallogr. 2004, 60,
2126–2132.
(48) Murshudov, G. N.; Vagin, A. A.; Dodson, E. J. Refinement
of macromolecular structures by the maximum likelihood
method. Acta. Crystallogr. Sect. D. Biol. Crystallogr. 1997, 53, 240–
255.
(49) Trott, O.; Olson, A. J. AutoDock Vina: Improving the speed
and accuracy of docking with a new scoring function, efficient
optimization, and multithreading. J. Comput. Chem. 2010, 31, 455–
461.
(50) Winn, M. D.; Ballard, C. C.; Cowtan, K. D.; Dodson, E. J.;
Emsley, P.; Evans, P. R.; Keegan, R. M.; Krissinel, E. B.; Leslie, A.
G.; McCoy, A.; McNicholas, S. J.; Murshudov, G. N.; Pannu, N. S.;
Potterton, E. A.; Powell, H. R.; Read, R. J.; Vagin, A.; Wilson K.S.
Overview of the CCP4 suite and current developments. Acta.
Crystallogr. D. Biol. Crystallogr. 2011, 67, 235–242.
ACS Paragon Plus Environment

Page 13 of 13
ACS Catalysis
1
2
3
4
5
6
Graphical abstract:
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment