Stereodivergent Synthesis of Carveol and Dihydrocarveol through Ketoreductases/Ene-Reductases Catalyzed Asymmetric Reduction

Article abstract of DOI:10.1002/cctc.201801391

Chiral carveol and dihydrocarveol are important additives in the flavor industry and building blocks in the synthesis of natural products. Despite the remarkable progress in asymmetric catalysis, convenient access to all possible stereoisomers of carveol and dihydrocarveol remains a challenge. Here, we present the stereodivergent synthesis of carveol and dihydrocarveol through ketoreductases/ene-reductases catalyzed asymmetric reduction. By directly asymmetric reduction of (R)- and (S)-carvone using ketoreductases, which have Prelog or anti-Prelog stereopreference, all four possible stereoisomers of carveol with medium to high diastereomeric excesses (up to >99 %) were first observed. Then four stereoisomers of dihydrocarvone were prepared through ene-reductases catalyzed diastereoselective synthesis. Asymmetric reduction of obtained dihydrocarvone isomers by ketoreductases further provide access to all eight stereoisomeric dihydrocarveol with up to 95 % de values. In addition, the absolute configurations of dihydrocarveol stereoisomers were determined by using modified Mosher's method.

Full text of DOI:10.1002/cctc.201801391

Accepted Article
Title: Stereodivergent Synthesis of Carveol and Dihydrocarveol
through Ketoreductases/Ene-Reductases Catalysed Asymmetric
Reduction
Authors: Jiyang Guo, Rui Zhang, Jingping Ouyang, Feiting Zhang,
Fengyu Qin, Guigao Liu, Wenhe Zhang, Hengyu Li, Xiaohong
Ji, Xian Jia, Bin Qin, and Song You
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201801391
Link to VoR: http://dx.doi.org/10.1002/cctc.201801391
A Journal of
www.chemcatchem.org

ChemCatChem
10.1002/cctc.201801391
FULL PAPER
Stereodivergent Synthesis of Carveol and Dihydrocarveol
through Ketoreductases/Ene-Reductases Catalysed Asymmetric
Reduction
Jiyang Guo,†,[a] Rui Zhang,†,[b] Jingping Ouyang, Feiting Zhang, Fengyu Qin, Guigao Liu, Wenhe
[c]
[a]
[a]
[b]
[a]
[a]
[b]
[c]
[b]
[a]
Zhang, Hengyu Li, Xiaohong Ji, Xian Jia, Bin Qin,* and Song You*
Dedication ((optional))
Abstract: Chiral carveol and dihydrocarveol are important additives
in the flavor industry and building blocks in the synthesis of natural
products. Despite the remarkable progress in asymmetric catalysis,
convenient access to all possible stereoisomers of carveol and
complex terpene natural products[2] and the preparation of chiral
ligands.[3] In addition, these compounds, such as carveol, could
be derived into permeation enhancers for transdermal drug
delivery.[4]
dihydrocarveol remains
a
challenge. Here, we present the
Owing to the existence of multiple stereogenic centers in their
structures, these chiral natural products have various
stereoisomers (Figure 1). Despite optically pure (4R)-carvone
((4R)-1) and (4S)-carvone ((4S)-1) could be isolated from
corresponding essential oil of spearmint (Mentha spicata) and
caraway (Carum carvi),[5] carveol (2), dihydrocarvone (3), and
stereodivergent synthesis of carveol and dihydrocarveol through
ketoreductases/ene-reductases catalysed asymmetric reduction. By
directly asymmetric reduction of (R)- and (S)-carvone using
ketoreductases, which have Prelog or anti-Prelog stereopreference,
all four possible stereoisomers of carveol with medium to high
diastereomeric excesses (up to >99%) were first observed. Then four
stereoisomers of dihydrocarvone were prepared through ene-
reductases catalysed diastereoselective synthesis. Asymmetric
reduction of obtained dihydrocarvone isomers by ketoreductases
further provide access to all eight stereoisomeric dihydrocarveol with
up to 95% de values. In addition, the absolute configurations of
dihydrocarveol stereoisomers were determined by using modified
Mosher’s method.
dihydrocarveol (4) are usually isolated as
a mixture of
stereoisomers.[6] Meanwhile, most stereoisomers of carveol,
dihydrocarvone, and dihydrocarveol are not commercial available
now. Since different stereoisomers of these compounds have
distinct odor quality/intensity and biological effects,[7] and also can
be used as starting materials for synthesis of different bioactive
natural products,[2a] generation every possible stereoisomer of
these compounds through stereodivergent synthesis[8] is
therefore highly desired.
Introduction
Carvone (1, Figure 1), carveol (2), dihydrocarvone (3), and
dihydrocarveol (4) are naturally occurring terpenoids that have
monocyclic skeletons and could be found in the essential oils of
Mentha spicate, Carum carvi and so on. These odorous
compounds, especially carvone and carveol, are useful and
valuable additives in the flavor and fragrance industry.[1]
Moreover, optically pure carvone and dihydrocarvone could be
used as versatile chiral building blocks in the total synthesis of
†
[
a]
b]
J. Guo, F. Zhang, F. Qin, W. Zhang, H. Li, Prof. S. You
School of Life Sciences and Biopharmaceutical Sciences
Shenyang Pharmaceutical University
1
03 Wenhua Road, Shenhe District, Shenyang 110016 (P.R. China)
E-mail: yousong207@163.com
R. Zhang, G. Liu, X. Ji, Dr. B. Qin
†
Figure 1. Asymmetric reduction of carvone (1) by ketoreductases and ene-
reductases. In the whole-cell biotransformation reaction, the presence of
ketoreductases and ene-reductases at the same time, and the un-acceptance
of carveol (2) by ene-reductases lead to the yielding of mixtures. On the other
hand, stereodivergent synthesis of carveol (2) could be achieved by
complementary ketoreductases catalysed reaction, while cascade reduction by
complementary ene-reductases and ketoreductases might enable the access to
stereodivergent synthesis of dihydrocarveol (4).
[
Wuya College of Innovation
Shenyang Pharmaceutical University
103 Wenhua Road, Shenhe District, Shenyang 110016 (P.R. China)
E-mail: to-qinbin@163.com, binqin@syphu.edu.cn
J. Ouyang, Prof. X. Jia
School of Pharmaceutical Engineering
Shenyang Pharmaceutical University
[
b]
†]
103 Wenhua Road, Shenhe District, Shenyang 110016 (P.R. China)
[
These authors contributed equally to this work
Supporting information for this article is given via a link at the end of
the document.((Please delete this text if not appropriate))
To prepare optically pure carveol, dihydrocarvone, and
dihydrocarveol, reduction of (4R)-carvone and (4S)-carvone has
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801391
FULL PAPER
previously been carried out using various whole-cell
biocatalysts,[9] such as bacteria,[9a] yeast,[9b,9c] fungi,[9d,9e] and plant
cells.[9f] However, the microbiological biotransformation of
carvone usually gives a mixture of C=O bond and activated C=C
bond reduced products, while the obtained carveol,
dihydrocarvone, and dihydrocarveol are also mixtures of
stereoisomers.[9] These un-desired results are particularly due to
stereocomplementary
reduction
of
halogen-substituted
acetophenones.[14] Based on this study, here we present the
stereodivergent synthesis of carveol (2) and dihydrocarveol (4).
Through directed reduction of (4R/S)-carvone (1) by
ketoreductases and sequential reduction (4R/S)-carvone (1) by
ene-reductases/ketoreductases, all four possible stereoisomers
of carveol (2) with medium to high diastereomeric excesses (41%
to >99% de) were observed while each of the eight stereoisomeric
dihydrocarveol (4) were also obtained with up to 95% total de
values.
(i) the presence of various competing ene-reductases (EREDs,
for activated C=C bond reduction) and ketoreductases (KREDs,
for C=O bond reduction) in cells, (ii) the different stereoselectivity
for each of these reductases, and (iii) the acceptance of
dihydrocarvone (3) by ketoreductases while the un-acceptance of
carveol (2) by ene-reductases (Figure 1). Although some works Results and Discussion
showed that recombinant enzymes could improve the selectivity
for (cascade) reduction of carvone (1),[10] for example, Zhu group
In the ketoreductase-substrate complex model, the attack of a
presented the synthesis of optically active (1R,2S,4R)-
dihydrocarveol.[10] However, none of these studies reported the
production of all possible stereoisomers, especially stereoisomers
of carveol (2) and dihydrocarveol (4). In addition, wild-type ene-
reductases and ketoreductases usually display specific
stereopreference, thus engineered enzymes with switched
stereopreference are needed.[11] Therefore, it’s still a challenge to
perform the synthesis complementary stereoisomers of carveol
and dihydrocarveol by using a biocatalytic route.
Promoted by the remarkable advance in bioinformation-based
enzyme mining and enzyme engineering,[12] several
stereocomplementary enzymes toward desired substrates have
been reported, such as lipase,[13a] ketoreductases,[13b] ene-
reductases,[13c] transaminases,[13d,13e] monooxygenases,[13f] and
so on. For example, Bornscheuer group identified various
hydride from coenzyme (NADPH/NADH) to the carbonyl of an
unsymmetrical ketone substrate would lead to the asymmetric
reduction of this substrate and the production of corresponding
optically active alcohol.[15] Usually, the smaller substituent of this
ketone substrate would be located in the smaller binding pocket
of ketoreductase, while the larger substituent is placed in the
larger pocket. For reduction the C=O bond of carvone (1), the
fragment of carvone that contains methyl group is smaller than
the fragment that contains isopropenyl group (Figure 2 and Figure
S1 in Supporting information). Although the fragments of the two
sides of the carbonyl are not significantly different, reported
stereoselective
reduction
of
almost-symmetrical
3-
oxacyclopentanone/3-thiacyclopentanone[16] and 4-alkylidene
cyclohexanone[17] indicated that ketoreductases could recognize
the slight difference between smaller and larger substituents of an
unsymmetrical substrate. On the other hand, the Prelog rule,
which works fine for linear ketones, also works in some cyclic
ketones.[18] Therefore, asymmetric reduction of unsymmetrical
carvone (1) by ketoreductases with Prelog and anti-Prelog
preference might produce carveol diastereoisomers with opposite
hydroxy groups at C-2 position. (Figure 2)
(
S)-selective and (R)-selective transaminases through in silico
approach and protein engineering,[13d,13e] Arnold group developed
diverse variants of heme proteins for stereodivergent
cyclopropanation of unactivated alkenes.[13f] In our recent work,
several
stereopreference were also observed by genome mining and
protein engineering, which could be used for
ketoreductases
with
Prelog
or
anti-Prelog
Figure 2. Stereodivergent synthesis of carveol (2) through asymmetric reduction of unsymmetrical carvone (1) by Prelog and anti-Prelog ketoreductases. In the
(4R)-carvone-ketoreductase model, the hydride from the coenzyme (NADPH) attack the carbonyl of (4R)-1 from the face that directly below the page. The different
(flipped) orientations of carvone in the binding pockets of Prelog and anti-Prelog ketoreductases would yield carveol with opposite configurations at C-2 position
(Figure S1 in Supporting information).
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801391
FULL PAPER
diastereoisomers, achiral GC was thus used to determine the
conversions and diastereomeric excess (de) values. Fortunately,
from the GC chromatogram, the Prelog enzymes displayed
opposite diastereoselectivity in comparison with anti-Prelog
enzymes at C-2 position, indicating the Prelog rule worked fine in
these cyclic ketones. As predicted by following the Prelog rule,
the Prelog reduction of (4R)-1 would yield (2R,4R)-2 as the
product, while anti-Prelog reduction of (4R)-1 could produce
Table 1. Asymmetric Reduction of (4R)-1 and (4S)-1 by Ketoreductases.
Conv
%][a]
By-P.
[%][b]
de
[%]
Substrate
Enzyme
Product
[c]
[
LfSDR1
V186A/G92F/E141L
2
4
5
6
4
95
OH
(2S,4R)-2 (Figure 2 and Figure S1). GC analysis also showed that
LfSDR1
V186A/G92E/E141L
7
52
the configuration of carveol observed by Prelog enzymes was the
4
same as the carveol produced by NaBH reduction (Figure S5),
LfSDR1 M50[d]
80
56
9
3
90
(
2R,4R)-2
which was previously reported as (2R,4R)-carveol [19], indicating
O
LfSDR1 M50[e]
the Prelog enzymes also gave predicted (2R,4R)-carveol as the
0
90
product.
Moreover,
the
Prelog
enzyme
LfSDR1
BmSDR11
BmSDR11 S95A
LfSDR1
trace
trace
2
>99
>99
>99
>99
>99
V186A/G92F/E141L variant displayed high diastereoselectivity
(95% de) for producing the first carveol stereoisomer (2R,4R)-2
(
4R)-1
OH
89
95
75
74
(Table 1). On the other hand, the second carveol stereoisomer
(2S,4R)-2 with high de values were easily observed through
reduction of (4R)-1 by several anti-Prelog ketoreductases. For
example, both V186W variant of LfSDR1 and S95A variant of
BmSDR11 displayed the ability to yield (2S,4R)-2 with >99% de
values (Table 1).
LfSDR1 V186F
LfSDR1 V186W
trace
trace
(
2S,4R)-2
LfSDR1
V186A/G92F/E141L
NA[f]
trace
NA
OH
2R,4S)-2
OH
With the stereocomplementary reduction of (4R)-1 in hand, we
next investigated the asymmetric reduction of its enantiomer (4S)-
carvone ((4S)-1) using same Prelog and anti-Prelog
ketoreductases.[14a] The absolute stereochemistry of observed
carveol diastereoisomers (2R,4S)-2 and (2S,4S)-2 could be easily
proposed, since the configurations of their “mirror-image”
stereoisomers (2S,4R)-2 and (2R,4R)-2 have been determined.
As show in Table 1, asymmetric reduction of (4S)-1 by Prelog
LfSDR1 V186A/G92E/E141L variant gave (2R,4S)-2 as the
product with a very low de value (5%). To improve the selectivity,
two polar residues D146 and K206 in the binding pocket of
LfSDR1 were mutated to non-polar V and L to facilitate the
accommodation of substrate (Figure S2). Fortunately, the LfSDR1
V186A/G92F/E141L/D146V/K206L variant (LfSDR1 M50)
displayed enhanced 41% diastereoselectivity toward (4S)-1,
giving (2R,4S)-2 as the product by following the Prelog rule. The
conversion in the reduction of (4R)-1 by this variant was also
improved. Meanwhile, anti-Prelog enzyme LfSDR1 V186W
variant exhibited high diastereoselectivity for reduction of (4S)-1,
affording (2S,4S)-2 with 92% de value. Therefore, all the four
possible stereoisomers of carveol (2) were observed through
asymmetric reduction of (4R)-1 and (4S)-1 by Prelog and anti-
Prelog ketoreductases (Table 1). Notably, three of four carveol
stereoisomers with high de values (>90%) could be prepared
when the best ketoreductase for each substrate was used. In
addition, some by-products were found in the reduction of
carvone. Since crude cell lysates of ketoreductases were used,
we proposed that the host’s endogenous ene-reductases in E. coli
cells (NemA homologue) might catalyse the activated C=C bond
reduction of carvone, yielding dihydrocarvone that could be
accepted as substrate by ketoreductases to give dihydrocarveol
as by-products. Therefore, when purified LfSDR1 M50 and
V186W variants were applied to catalyse the reduction of
corresponding (4R)-1 and (4S)-1, no by-products were detected
while the stereoselectivity was maintained at high (Table 1).
LfSDR1
V186A/G92E/E141L
72
4
5
5
(
O
LfSDR1 M50[d]
96
41
BmSDR11
BmSDR11 S95A
LfSDR1
NA
NA
NA
75
trace
trace
trace
3
NA
NA
NA
93
(
4S)-1
LfSDR1 V186W
LfSDR1 V186W[e]
(2S,4S)-2
79
0
93
[
a] Conversions (in percent) were determined by achiral GC analysis. [b] By-
products dihydrocarveol were determined by achiral GC analysis. [c]
Diastereomeric excess (de) values (in percent) were determined by achiral
GC analysis. The absolute configuration of (2R,4R)-2 was identified by
),[19]
comparing the retention time of GC with authentic sample (by NaBH
4
inspiring the absolute configurations of its three diastereoisomers. The de
values were calculated as de=[(Major-Minor)/(Major+Minor)]×100. [d] The
sequence ofLfSDR1 M50 variant was V186A/G92F/E141L/D146V/K206L. [e]
Reaction was performed using purified enzyme (See Supporting information
for details). [f] NA, No measurable activity. Reaction condition: Experiments
were performed at analytical scale using cell lysate of E. coli expressing
SDRs (0.2 g wet cells) or purified SDRs (10 μM) with GDH (0.1 g wet cells or
μM), 5 mM carvone, 44 mM glucose, 0.4 mM NADP , 10 vol % EtOH, and
PBS buffer (100 mM, pH 6.0) at 30 °C for 12 hours.
+
5
Inspired by the variety of ketoreductases, we first set out on the
stereocomplementary reduction of (4R)-carvone ((4R)-1). Several
previously obtained short-chain reductases (SDRs) with Prelog or
anti-Prelog preference (Figure S1) toward halogen-substituted
acetophenones in our recent study[14a] were initially investigated
at analytical scale (5 mM substrate). Since the starting (4R)-
carvone ((4R)-1) is optically pure (>99% ee), which means the
resultant
carveol
(2R,4R)-2
and
(2S,4R)-2
were
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801391
FULL PAPER
Figure 3. Stereodivergent synthesis of dihydrocarveol (4) through asymmetric reduction of carvone (1) by ene-reductases and ketoreductases.
Table 2. Asymmetric Reduction of (4R)-1 and (4S)-1 by Ene-Reductases.
Co-
Solv
Conv
[%][a]
de
[%]
Substrate
Enzyme
Product
[b]
PETNR
PETNR
NamA
YqjM
EtOH
PE[c]
PE
>99
>99
>99
98
70
93
94
94
O
O
(1R,4R)-3
PE
OYE1 W116A
PE
PE
PE
PE
PE
>99
>99
>99
>99
>99
51
66
59
70
96
O
OYE1 W116A/T37C
OYE1 W116A/T37S
OYE1 W116A/T37A
OYE1 W116A/T37A[d]
(4R)-1
(
1S,4R)-3
Figure 4. Structure of OYE1 W116A variant (PDB ID: 4K7V[21]). (a) (4R)-
carvone ((4R)-1), shown as yellow sticks) in the active site of OYE1 W116A
variant. (b) Active site surface of the OYE1 W116A with bound (4R)-1 (shown
as yellow sticks). Residue T37 (shown as blue spheres) was directly below the
isopropenyl group of (4R)-1. The cofactor FMN is shown as white sticks.
Substrate binding residues H191 and N194 are shown as green sticks. Catalytic
residue Y196 is shown as cyan stick while previously engineered A116 residue
is shown as magenta sticks.
PETNR
PETNR
NamA
YqjM
EtOH
PE
>99
>99
>99
8
86
93
94
91
O
O
PE
(
1R,4S)-3
PE
TsER C25G/I67T
O
PE
>99
78
(4S)-1
Next, we turned to create all eight possible stereoisomers of
dihydrocarveol (4). To achieve this, four optically pure
OYE1 W116A
PE
>99
93
stereoisomers
of
dihydrocarvone
(3)
are
needed.
(1S,4S)-3
Stereocomplementary reduction of these dihydrocarvone
stereoisomers by previous ketoreductases might provide access
to all eight possible stereoisomers of dihydrocarveol (Figure 3). In
recent years, the number of studies combining different
biocatalysts or combining chemo- and biocatalysis,[20] especially
ketoreductases with other enzymes[21] has substantially
increased. Thus, combining ene-reductases with ketoreductases
was performed to create dihydrocarveol stereoisomers. The
biocatalytic reduction of activated C=C bond of carvone (1) to give
dihydrocarvone (3) has been well studied by several ene-
[
a] Conversions and [b] Diastereomeric excess (de) values were determined
by achiral GC analysis. The absolute configurations of obtained
dihydrocarvone were identified by comparing with literature data. [c] PE,
petroleum ether. [d] Reaction was performed using purified enzyme.
Reaction condition: Experiments were performed at analytical scale using cell
lysate of E. coli expressing EREDs (0.2 g wet cells) or purified EREDs (10
μM) with GDH (0.1 g wet cells or 5 μM), 5 mM carvone, 44 mM glucose, 0.4
+
mM NADP , 10 vol % EtOH or PE, and PBS buffer (100 mM, pH 6.0) at 25
°
C for 12 hours.
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801391
FULL PAPER
reductases.[10,13c,22,23] For example, (1R,4R)-3 and (1R,4S)-3
could be obtained through asymmetric reduction of corresponding
To prepare optical pure dihydrocarvone (3), we cloned or
synthesized the genes for some ene-reductases, and then
expressed them in E. coli. Biocatalytic reduction of (4R)-carvone
and (4S)-carvone was first performed using cell lysates of these
ene-reductases. As shown in Table 2, the de values of obtained
dihydrocarvone stereoisomers (1R,4R)-3 and (1R,4S)-3 by
PETNR were corresponding 69% and 86% when ethanol was
used as co-solvent. Luckily, the diastereoselectivity was improved
to about 93% when petroleum ether was used as co-solvent for
(
4R)-1 and (4S)-1 by various ene-reductases, such as
OYE1,[22,23a] PETNR[23b], and so on.[23d] However, only a few ene-
reductase variants, such as OYE W116A/V[22] and TsER
C25G/I67T,[23b] displayed switched stereoselectivity to yield
dihydrocarvone with S configuration at C-1 position. In addition,
the observed (1S,4R)-3 by OYE W116A/V variants had medium
de values (about 50%),[22] which was not sufficient for the next
reduction by ketoreductases.
PETNR (Table
2
and Figure S6). Similarly, the third
dihydrocarvone stereoisomer (1S,4S)-3 with 93% de value was
also observed when OYE1 W116A variant (Table 2) was used for
reduction of (4S)-1 and we expected the de values of three
observed dihydrocarvone stereoisomers to be sufficient for further
reduction by ketoreductases.
Table 3. Asymmetric reduction of (1R,4R)-3 and (1S,4R)-3 by ketoreductases.
Conv
[
de
de
Substrate
Enzyme
Product
%][a]
[%]
96
98
[b]
[%]
92
91
[c]
Protein engineering was then carried out to improve the
diastereoselectivity of OYE1 W116A variant for reduction of (4R)-
BmSDR5
OH
96
1, which gave the last dihydrocarvone stereoisomer (1S,4R)-3
LfSDR1 M50
>99
>99
with insufficient 51% de value. From the crystal structure of OYE1
W116A variant in complex with (4R)-1 (PDB ID: 4K7V),[22] we
noticed that the residue T37 in the hydrophobic binding pocket
created by W116A was directly below the isopropenyl group of
O
(1R,2R,4
R)-4
BmSDR5 A187F/F188I
>99
9
BmSDR11
BmSDR11 S95A
LfSDR1
>99
>99
>99
>99
88
>99
97
93
89
OH
(4R)-1 (Figure 4). We hypothesized that substituting of T37 by
(1R,4R)-3
similar C, smaller S and A residues could reshape or further
enlarge the binding pocket, which might improve the
diastereoselectivity toward (4R)-1. As shown in Table 2, the
diastereoselectivity for reduction of (R)-1 by W116A/T37C,
W116A/T37S, and W116A/T37A variants were improved from
>99
>99
97
93
(
1R,2S,4
R)-4
LfSDR1 V186W
93
BmSDR5
35
OH
51% to 66%, 59%, and 70%, respectively. When purified OYE1
BmSDR5[d]
W116A/T37A variant was used, (1S,4R)-3 with 95% de value was
obtained (Table 2 and Figure S6), which was sufficient for next
reduction by ketoreductases. Thus, asymmetric reduction of (R)-
>99
89
97
93
LfSDR1 M50
BmSDR5 A187F/F188I
BmSDR11
81
19
(
1S,2R,4
R)-4
O
99
30
NA[e]
31
1 and (S)-1 by ene-reductases could give all four possible
stereoisomers of dihydrocarvone (3) with >90% de values.
Motivated by these results, we next performed biocatalytic
reduction of observed dihydrocarvone stereoisomers (3) using
ketoreductases with Prelog or anti-Prelog stereopreference,
which previously enabled the access to stereodivergent synthesis
of carveol (2). Since these ketoreductases could accept carvone
as substrates, the next reduction was performed in the sequential
mode. (1R,4R)-3 with 93% de value from PETNR catalysed
reaction was first extracted and used as the substrate for
ketoreductases. Similar as reduction of (4R)-1, Prelog enzymes
>99
82
79
OH
(
1S,4R)-3
LfSDR1
66
21
LfSDR1 V186W
98
94
46
(
1S,2S,4
R)-4
LfSDR1 V186W[d]
>99
98
95
[
a] Conversions and [b] diastereomeric excess (de) values were determined
by achiral GC analysis. The absolute configurations were proposed and then
determined by using modified Mosher’s method. The de values for the major
of two resultant dihydrocarveol stereoisomers (from major dihydrocarvone)
were calculated, for example, the de value of (1R,2R,4R)-4 for BmSDR5 was
calculated as {[(1R,2R,4R)-4 - (1R,2S,4R)-4]/[(1R,2R,4R)-4 + (1R,2S,4R)-
(
(
BmSDR5, LfSDR1 M50 variant) and anti-Prelog enzymes
BmSDR5 A187F/F188I variant, BmSDR11, LfSDR1, and so on)
4
]}×100. [c] Since the starting dihydrocarvone was not optically pure, the total
also displayed opposite diastereoselectivity for reduction the C=O
bond of (1R,4R)-3. As shown in Table 3, although the total de
values were limited by the starting (1R,4R)-3 (93% de), some
ketoreductases still displayed excellent selectivity. For example,
both Prelog BmSDR5 and anti-Prelog LfSDR1 V186W variant had
de values for the major of four resultant dihydrocarveol stereoisomers were
calculated, for example, the total de of (1R,2R,4R)-4 for BmSDR5 was
calculated as {[(1R,2R,4R)-4 - (1R,2S,4R)-4 - (1S,2R,4R)-4 - (1S,2S,4R)-
4]/[(1R,2R,4R)-4 + (1R,2S,4R)-4) + (1S,2R,4R)-4 + (1S,2S,4R)-4]}×100. [d]
Reactions were performed using purified ene-reductases and keto-
reductases. [e] NA, not available, the major (1R,2S,4R)-4 was less than the
sum of its three diastereoisomers. Reaction condition: Experiments were
performed at analytical scale using cell lysate of E. coli expressing SDRs (0.2
g wet cells) or purified SDRs (10 μM) with GDH (0.1 g wet cells or 5 μM), 5
>95% diastereoselectivity toward (1R,4R)-3, yielding (1R,2R,4R)-
4
and (1R,2S,4R)-4, respectively. Notably, same SDRs display
different selectivity toward carvone and dihydrocarvone. This was
reasonable since the methyl groups of dihydrocarvone isomers
might have steric interactions with the active pockets of SDRs.
+
dihydrocarvone (from EREDs reaction), 44 mM glucose, 0.4 mM NADP , 10
vol % EtOH, and PBS buffer (100 mM, pH 6.0) at 30 °C for 12 hours.
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801391
FULL PAPER
been obtained through next reduction by ene-reductases and
ketoreductases.
Table 4. Asymmetric reduction of (1R,4S)-3 and (1S,4S)-3 by ketoreductases.
Conv
%][a]
de
[%]
de
[%]
With four stereoisomers of dihydrocarveol in hand, we next
performed asymmetric reduction of (1R,4S)-3 and (1S,4S)-3 to
prepare the other four stereoisomers. (1R,4S)-3 with 93% de
value derived from (4S)-1 by PETNR was first used as substrate
for ketoreductases. As show in Table 4, some anti-Prelog
ketoreductases, such as BmSDR5 A187F/F188I variant,
BmSDR11 S95A variant, and LfSDR1 V186W variant displayed
high diastereoselectivity toward (1R,4S)-3. For example, >99%
diastereoselectivity was found in LfSDR1 V186W variant
catalysed reaction, which gave the product (1R,2S,4S)-4 with 92%
total de value. On the other hand, the tested Prelog
ketoreductases exhibited medium diastereoselectivity toward
Substrate
Enzyme
Product
OH
[b]
[c]
[
LfSDR1
V186A/G92F/E141L
>99
52
46
LfSDR1
V186A/G92E/E141L
>99
78
70
O
LfSDR1 M50
(1R,2R,4
S)-4
>99
>99
>99
9
56
83
92
87
92
>99
60
79
75
91
47
58
44
96
98
56
74
86
80
85
92
54
71
73
90
37
43
41
88
90
LfSDR1 M51d
BmSDR5 A187F/F188I
BmSDR11
(1R,4S)-3
OH
(1R,4S)-3. The best enzyme LfSDR1 M51 variant displayed 83%
BmSDR11 S95A
LfSDR1 V186W
BsSDR8
>99
>99
73
(1R,2S,4
S)-4
diastereoselectivity and gave the product (1R,2R,4S)-4 with 74%
total de value. Then the last dihydrocarvone (1S,4S)-3 with 93%
de value observed from carvone (4S)-1 by OYE1 W116A variant
was tested using ketoreductases. Surprisingly, all the previously
used Prelog ketoreductases showed low activity and
diastereoselectivity toward (1S,4S)-3. Luckily, F146W variant of
BsSDR8, a SDR enzyme had different structural skeleton with
LfSDR1, BmSDR5, and BmSDR11 (Figure S2),[ exhibited 91%
diastereoselectivity toward (1S,4S)-3 and gave the product
OH
BsSDR8 S184Y
BsSDR8 F146Y
BsSDR8 F146W
BmSDR5 A187F/F188I
BmSDR11
87
71
(1S,2R,4
S)-4
O
95
14a]
73
OH
(1S,2R,4S)-4 with 89% total de value (Table 4). In contrast, the
31
(1S,4S)-3
last dihydrocarveol stereoisomer (1S,2S,4S)-4 with high de value
was easily observed by some anti-Prelog ketoreductases. For
BmSDR11 S95A
LfSDR1
97
(1S,2S,4
S)-4
87
example,
diastereoselectivity toward (1S,4S)-3 and gave the desired
1S,2S,4S)-4 with 90% total de value. Therefore, all eight
LfSDR1
V186W
variant
displayed
98%
LfSDR1 V186W
>99
(
stereoisomers of dihydrocarveol were obtained by ene-reductase-
and ketoreductase-based reduction of (4R)-1 and (4S)-1.
[
a] Conversions (in percent). [b] Diastereomeric excess (de) values (in
percent). The de values for the major of two resultant dihydrocarveol
stereoisomers (from major dihydrocarvone) were calculated. [c] The total de
values for the major of four resultant dihydrocarveol stereoisomers were
calculated. [d] The sequence of LfSDR1 M51 variant was
V186A/G92E/E141L/D146V/K206L.
(
1S,4R)-3 with 95% de value observed from reduction of (4R)-
by OYE1 W116A/T37A variant was then used for next reduction
by previous ketoreductases. As shown in Table 3, some Prelog
and anti-Prelog ketoreductases displayed high
1
diastereoselectivity toward (1S,4R)-3. Surprisingly, the total de
values of observed (1S,2R,4R)-4 and (1S,2S,4R)-4 were far less
than the 95% de value of their substrate (1S,4R)-3. For example,
Prelog BmSDR5 WT and anti-Prelog LfSDR1 V186W variant
gave (1S,2R,4R)-4 with 35% de value and (1S,2S,4R)-4 with 46%
de value, respectively. Control-experiments using E. coli cell
lysate revealed that the host’s endogenous enzymes could
catalyse the isomerization of (1S,4R)-3 into (1R,4R)-3, leading to
the decrease of (1S,4R)-3 and low de values of reduced products
by ketoreductases. Therefore, purified BmSDR5 WT and LfSDR1
V186W variant, both displayed >93% diastereoselectivity toward
Figure 5. The absolute configurations of the C-2 hydroxyl groups in (1R,2S,4S)-
4 and (1S,2S,4S)-4 were determined through the application of modified
Mosher’s method,[25] by comparing the differences in the 1H NMR chemical
shifts of the (R)- and (S)-MPA esters (5R/S, 6R/S).
To test the potential in practical utility of this process, we next
performed preparative-scale synthesis of dihydrocarveol
stereoisomers (1R,2S,4R)-4, (1R,2S,4S)-4, and (1S,2S,4S)-4.
Sequential reduction of (4R)-1 using PENTR and LfSDR1 V186W
was conducted on a 25-mg scale, affording the corresponding
(
(
1S,4R)-3, were used to prevent the isomerization. As expected,
1S,2R,4R)-4 and (1S,2S,4R)-4 with respective 93 % and 95%
total de values were observed. Taken together, now
dihydrocarveol stereoisomers (1R,2R,4R)-4, (1R,2S,4R)-4,
(
4
1R,2S,4S)-4 in 48% isolated yield. Similarly, (1R,2S,4S)-4 with
0% isolated yield and (1S,2S,4S)-4 with 52% isolated yield were
observed when corresponding PETNR and LfSDR1 V186W,
(1S,2R,4R)-4, and (1S,2S,4R)-4 derived from (4R)-carvone have
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801391
FULL PAPER
OYE1 W116A and LfSDR1 V186W were used for sequential
reduction of (4S)-1 on 25-mg scale. The carveol stereoisomer
For example, only two of eight dihydrocarveol stereoisomers were
obtained in a previous study.[10] Through ketoreductase-based
direct reduction and ene-reductase/ketoreductase-based
sequential reduction of (R)-carvone and (S)-carvone,
complementary stereoisomers of carveol and dihydrocarveol
were observed in this study. Furthermore, as show in Figure 6,
most of these stereoisomers could be accessed with >90% total
de values when the best enzyme was used for each substrate.
We also noticed that the anti-Prelog enzyme LfSDR1 V186W
variant displayed high stereoselectivity toward all the six-tested
ketone substrates. However, carveol stereoisomer (2R,4S)-2 and
dihydrocarveol stereoisomer (1R,2R,4S)-4 were observed with
medium 41% de value and 74% de value, respectively (Figure 6).
Interestingly, their substrates were corresponding (4S)-1
and(1R,4S)-3, both of which have isopropenyl groups with S
configurations and should be accepted by Prelog enzymes to
afford alcohols. The moderate stereoselectivity is possibly due to
the unfavorable steric interactions between methyl (or
isopropenyl) group and the attacking hydride from coenzyme
(Figure 7a and 7b, Figure S3 and S4), or the insufficient
stereoselectivity of tested ketoreductases with Prelog preference.
Since the used ketoreductases in this study were not specifically
designed and engineered for carvone and dihydrocarvone, while
only four SDR enzymes (LfSDR1, BmSDR5, BmSDR11, and
BsSDR8) and their variants were used, we proposed further
(2S,4R)-2 was also obtained with 73% isolated yield when
LfSDR1 V186W variant was used for direct reduction of (4R)-1. In
addition, the de values of obtained dihydrocarveol stereoisomers
(1R,2S,4R)-4, (1R,2S,4S)-4, (1S,2S,4S)-4, and carveol
stereoisomer (2S,4R)-2 were 93%, 92%, 91%, and >99%,
respectively, which was maintained at high as analytical-scale
reaction.
Since the configurations of C-2 hydroxyl groups in the observed
dihydrocarveol stereoisomers were proposed by following the
Prelog rule,[24] we then determined their absolute configurations.
As show in Figure 5, absolute stereochemistry at C-2 position of
(
1R,2S,4S)-4 and (1S,2S,4S)-4 derived from (4S)-carvone were
established by the application of modified Mosher’s method.[25]
R)- and (S)-MPA esters at C-2 alcohol of (1R,2S,4S)-4 (5R and
(
5
S) were first synthesized and isolated by following standard
1
procedures. From the H NMR spectra of 5R and 5S (Figure S11-
S16), the ΔδRS (δR - δ ) values were calculated, allowing the
assignment of the absolute configuration of C-2 as S (Figure 5),
which was same as the predicted configuration. Similarly, the
absolute stereochemistry at C-2 of (1S,2S,4S)-4 was also
assigned as the predicted S. Because (1R,2S,4S)-4 and
S
(
(
1R,2R,4S)-4 were derived from same substrate (1R,4S)-3, while
1S,2S,4S)-4 and (1S,2R,4S)-4 were derived from same
substrate (1S,4S)-3, the determined configurations of (1R,2S,4S)- screening, rational design and protein engineering of
4
and (1S,2S,4S)-4 could inspire the stereochemistry of their
ketoreductases might overcome this puzzle. On the other hand,
the ene-reductases enzymes with about 93% diastereoselectivity
in cell lysate also limited the total de values of observed
diastereoisomers. Meanwhile, the absolute stereochemistry the
other four stereoisomers of dihydrocarveol could be further
identified since they were “mirror-image” derived for (1R,4R)-3
and (1S,4R)-3.
Chiral carveol (2) and dihydrocarveol (4) have potential
application in the flavor industry and could be used as building
blocks for synthesis of bioactive natural products. However,
access to all four possible stereoisomers of carveol and all eight
possible stereoisomers of dihydrocarveol are not available now.
dihydrocarveol
stereoisomers.
Reaction
optimization,
engineering of ene-reductases, and removal of host’s
endogenous enzymes might provide access to the optically pure
dihydrocarvone isomers and further achieve the preparation of
dihydrocarveol complementary stereoisomers with high de
values.
Figure 6. All possible stereoisomers of carveol (2) and dihydrocarveol (4) observed from stereodivergent synthesis. (a) Four possible stereoisomers of carveol (2)
with 41% to >99% de values were observed from direct reduction of carvone by ketoreductases with Prelog or anti-Prelog stereopreference. The racemic carveol
(
2) was commercial authentic standard. (b-c) Eight possible stereoisomers of dihydrocarveol (4) with 74% to 95% de values were obtained from reduction of carvone
by ene-reductases and ketoreductases. The mixture of dihydrocarveol stereoisomers was prepared through the reduction of commercial dihydrocarvone (mixture
of stereoisomers) by NaBH
4
.
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801391
FULL PAPER
EREDs (0.2 g wet cells) or purified EREDs (10 μM) and GDH (0.1 g wet
cells or 5 μM), glucose (8 mg, 44 mM), and NADP+ (0.3 mg, 0.4 mM) in
900 µL PBS buffer (100 mM, pH 6.0). The mixture was incubated at 25 °C
and 200 rpm for 12 h, and then extracted twice with ethyl acetate. The
combined organic layers were concentrated in vacuo. The de values of the
products were determined by achiral GC analysis. For sequential reduction
by SDRs, the extract of dihydrocarvone by EREDs were evaporated and
dissolved in ethanol. Without purification, the obtained dihydrocarvone
was used as substrate for SDRs, following same procedure as direct-
reduction of carvone.
Acknowledgements
Figure 7. Docking of (4S)-1 in the active site of LfSDR1 M50 variant model and
This work was financially supported by the Young Elite Scientists
Sponsorship Program by CAST (2016QNRC001), the National
Natural Science Foundation of China (81602993), Science and
Technology Research Projects from the Ministry of Education of
the People’s Republic of China (213007A), and Project from the
Department of Education of Liaoning Province.
(1R,4S)-3 in the active site of LfSDR1 M51 variant model. The substrates are
shown as yellow sticks and spheres. The catalytic residues (Ser and Tyr) and
cofactor NADP+ are shown as white sticks. The key residues as green sticks.
See Figure S3 and S4 for more details.
Conclusions
Stereodivergent asymmetric synthesis is a valuable approach that Conflicts of interest
can allow efficient access to all stereoisomers of a desired
product. In this research, we developed a stereodivergent
strategy for the synthesis of all possible stereoisomers of carveol
and dihydrocarveol by using ketoreductases and ene-reductases.
The asymmetric reduction of (R)-carvone and (S)-carvone
catalysed by Prelog and anti-Prelog stereopreferred
ketoreductases first gave all four possible stereoisomers of
carveol with medium to high diastereomeric excesses (up to
The authors declare no conflict of interest.
Keywords: stereodivergent synthesis • carveol • dihydrocarveol
•
ene-reductases • ketoreductases • asymmetric reduction
[1]
a) E. Brenna, C. Fuganti, F. G. Gatti, S. Serra, Chem. Rev. 2011, 111,
4036-4072; b) J. L. Bicas, A. P. Dionísio, G. M. Pastore, Chem. Rev.
>99% de). Then previously reported and newly engineered ene-
2009, 109, 4518-4531.
reductases with R or S selectivity were applied to synthesize four
stereoisomers of dihydrocarvone. The observed dihydrocarvone
were further accessed to all eight stereoisomeric dihydrocarveol
with 74% to 95% total de values through asymmetric reduction by
Prelog and anti-Prelog ketoreductases. In addition, the
preparative-scale synthesis of dihydrocarveol and carveol
stereoisomers indicates the potential in practical utility of this
process. This stereodivergent approach with further engineered
enzymes could also serve a start for controlled, green, and
efficient synthesis of other valuable chiral alcohols.
[2]
a) Z. G. Brill, M. L. Condakes, C. P. Ting, T. J. Maimone, Chem. Rev.
2017, 117, 11753-11795; b) P. Finkbeiner, K. Murai, M. Röpke, R.
Sarpong, J. Am. Chem. Soc. 2017, 139, 11349-11352; c) L. J. Nannini,
S. J. Nemat, E. M. Carreira, Angew. Chem., Int. Ed. 2018, 57, 823-826;
d) H. Chu, G. Dünstl, J. Felding, P. S. Baran, Bioorg. Med. Chem. Lett.
2018, 28, 2705-2707;
[
3]
4]
A. J. Smaligo, S. Vardhineedi, O. Kwon, ACS Catal. 2018, 8, 5188-5192.
a) B. Sapra, S. Jain, A. K. Tiwary, AAPS J. 2008, 10, 120-132; b) A.
Herman, A. P. Herman, J. Pharm. Pharmacol. 2015, 67, 473-485.
J. M. Finefield, D. H. Sherman, M. Kreitman, R. M. Williams, Angew.
Chem., Int. Ed. 2012, 51, 4802-4836.
[
[5]
[
6]
7]
H. J. Bouwmeester, J. Gershenzon, M. C. J. M. Konings, R. Croteau,
Plant Physiol. 1998, 117, 901-912.
[
a) T. J. Leitereg, D. G. Guadagni, J. Harris, T. R. Mon, R. Teranishi,
Nature 1971, 230, 455-456; b) R. Bentley, Chem. Rev. 2006, 106, 4099-
Experimental Section
4112.
Asymmetric reduction of carvone. The enzymatic reduction of carvone
[
8]
9]
S. Krautwald, E. M. Carreira, J. Am. Chem. Soc. 2017, 139, 5627-5639.
a) M. R. Cramarossa, A. Nadini, M. Bondi, P. Messi, U. M. Pagnoni, L.
Forti, C. R. Chim. 2005, 8, 849-852; b) M. Goretti, C. Ponzoni, E. Caselli,
E. Marchigiani, M. R. Cramarossa, B. Turchetti, P. Buzzini, L. Forti,
Enzyme Microb. Technol. 2009, 45, 463-468; c) M. Goretti, B. Turchetti,
M. Cramarossa, L. Forti, P. Buzzini, Molecules 2013, 18, 5736-5748; d)
R. A. M. d. Santos, F. d. O. Souza, E. J. Pilau, C. Porto, J. E. Gonçalves,
A. J. B. d. Oliveira, R. A. C. Gonçalves, Biocatal. Biotransform. 2017, 1-
(
1) by SDRs was carried out by using a NADPH recycle system as
[
described previously.[14] Carvone (1) (5 mM) in 100 µL ethanol were added
to a mixture of cell lysate of E. coli Rosetta2 (DE3) harbouring SDRs (0.2
g wet cells) or purified SDRs (10 μM) and GDH (0.1 g wet cells or 5 μM),
+
glucose (8 mg, 44 mM), and NADP (0.3 mg, 0.4 mM) in 900 µL PBS buffer
(100 mM, pH 6.0). The mixture was incubated at 30 °C and 200 rpm for 12
h, and then extracted twice with ethyl acetate. The combined organic
layers were concentrated in vacuo. Diastereomeric excess (de) values of
the products were determined by achiral GC analysis. The enzymatic
reduction of carvone (1) by EREDs were also carried out by using a
NADPH recycle. Carvone (1) (5 mM) in 100 µL ethanol or petroleum ether
was added to a mixture of cell lysate of E. coli Rosetta2 (DE3) harbouring
5; e) M. S. van Dyk, E. van Rensburg, I. P. B. Rensburg, N. Moleleki, J.
Mol. Catal. B: Enzym. 1998, 5, 149-154; f) A. Speicher, H. Roeser and
R. Heisel, J. Mol. Catal. B: Enzym. 2003, 22, 71-77; g) I. L. Hook, S.
Ryan, H. Sheridan, Phytochemistry 2003, 63, 31-36.
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801391
FULL PAPER
[
[
[
10] X. Chen, X. Gao, Q. Wu, D. Zhu, Tetrahedron: Asymmetry 2012, 23, 734-
19, 1954-1958; d) J. Ou-yang, W. Zhang, F. Qin, W. Zuo, S. Xu, Y. Wang,
B. Qin, S. You, X. Jia, Org. Biomol. Chem., 2017, 15, 7374-7379.
[19] C. Gally, B. M. Nestl, B. Hauer, Angew. Chem., Int. Ed. 2015, 54, 12952-
12956.
738.
11] P. F. Mugford, U. G. Wagner, Y. Jiang, K. Faber, R. J. Kazlauskas,
Angew. Chem., Int. Ed. 2008, 47, 8782-8793.
12] a) K. L. Bren, J. Am. Chem. Soc. 2017, 139, 14331-14334; b) U.
Bornscheuer, G. Huisman, R. Kazlauskas, S. Lutz, J. Moore, K. Robins,
Nature 2012, 485, 185-194.
[20] a) J. Muschiol, C. Peters, N. Oberleitner, M. D. Mihovilovic, U. T.
Bornscheuer, F. Rudroff, Chem. Commun., 2015, 51, 5798-5811; b) S.
P. France, L. J. Hepworth, N. J. Turner, S. L. Flitsch, ACS Catal., 2017,
7, 710-724; c) J. H. Schrittwieser, S. Velikogne, M. Hall, W. Kroutil,
Chem. Rev., 2018, 118, 270-348; d) H. Gröger, W. Hummel, Curr. Opin.
Chem. Biol., 2014, 19, 171-179; e) F. Rudroff, M. D. Mihovilovic, H.
Gröger, R. Snajdrova, H. Iding, U. T. Bornscheuer, Nat. Catal., 2018, 1,
12-22; f) S. Gandomkar, A. Żądło-Dobrowolska, W. Kroutil,
ChemCatChem, 2018, DOI: 10.1002/cctc.201801063.
[13] a) Q. Wu, P. Soni, M. T. Reetz, J. Am. Chem. Soc. 2013, 135, 1872-
1
881; b) M. Hall, A. S. Bommarius, Chem. Rev. 2011, 111, 4088-4110;
c) S. K. Padhi, D. J. Bougioukou, J. D. Stewart, J. Am. Chem. Soc. 2009,
31, 3271-3280; d) I. V. Pavlidis, M. S. Weiß, M. Genz, P. Spurr, S. P.
Hanlon, B. Wirz, H. Iding, U. T. Bornscheuer, Nat. Chem. 2016, 8, 1076-
082; e) M. Höhne, S. Schätzle, H. Jochens, K. Robins, U. T.
1
1
Bornscheuer, Nat. Chem. Biol. 2010, 6, 807-813; f) A. M. Knight, S. B. J.
Kan, R. D. Lewis, O. F. Brandenberg, K. Chen, F. H. Arnold, ACS Cent.
Sci. 2018, 4, 372-377.
[21] a) K. Tauber, M. Fuchs, J. H. Sattler, J. Pitzer, D. Pressnitz, D.
Koszelewski, K. Faber, J. Pfeffer, T. Haas, W. Kroutil, Chem. – Eur. J.,
2013, 19, 4030-4035; b) F. G. Mutti, T. Knaus, N. S. Scrutton, M. Breuer,
N. J. Turner, Science, 2015, 349, 1525-1529; c) E. Liardo, N. Ríos-
Lombardía, F. Morís, J. González-Sabín, F. Rebolledo, Org. Lett., 2016,
18, 3366-3369.; d) M. J. Rodríguez-Álvarez, N. Ríos-Lombardía, S.
Schumacher, D. Pérez-Iglesias, F. Morís, V. Cadierno, J. García-
Álvarez, J. González-Sabín, ACS Catal., 2017, 7, 7753-7759; e) E.
Brenna, M. Crotti, M. DeꢀPieri, F. G. Gatti, G. Manenti, D. Monti, Adv.
Synth. Catal., 2018, DOI: 10.1002/adsc.201800299.
[
14] a) F. Qin, B. Qin, W. Zhang, Y. Liu, X. Su, T. Zhu, J. Ouyang, J. Guo, Y.
Li, F. Zhang, J. Tang, X. Jia, S. You, ACS Catal. 2018, 8, 6012-6020; b)
F. Qin, B. Qin, T. Mori, Y. Wang, L. Meng, X. Zhang, X. Jia, I. Abe, S.
You, ACS Catal. 2016, 6, 6135-6140.
[
15] C. M. Nealon, M. M. Musa, J. M. Patel, R. S. Phillips, ACS Catal. 2015,
5, 2100-2114.
[
16] a) E. L. Noey, N. Tibrewal, G. Jiménez-Osés, S. Osuna, J. Park, C. M.
Bond, D. Cascio, J. Liang, X. Zhang, G. W. Huisman, Y. Tang, K. N.
Houk, Proc. Natl. Acad. Sci. U. S. A. 2015, 112, E7065-E7072; b) Z. Sun,
R. Lonsdale, A. Ilie, G. Li, J. Zhou, M. T. Reetz, ACS Catal. 2016, 6,
[22] Y. A. Pompeu, B. Sullivan, J. D. Stewart, ACS Catal. 2013, 3, 2376-2390.
[23] a) A. Fryszkowska, H. Toogood, M. Sakuma, J. M. Gardiner, G. M.
Stephens, N. S. Scrutton, Adv. Synth. Catal. 2009, 351, 2976-2990; b)
N. Nett, S. Duewel, A. A. Richter, S. Hoebenreich, ChemBioChem 2017,
18, 685-691; c) Y. Fu, K. Castiglione, D. Weuster-Botz, Biotechnol.
Bioeng. 2013, 110, 1293-1301; d) H. S. Toogood, N. S. Scrutton, ACS
Catal. 2018, 8, 3532-3549.
1
598-1605.
17] R. Agudo, G.-D. Roiban, M. T. Reetz, J. Am. Chem. Soc. 2013, 135,
665-1668.
18] a) R. Agudo, G.-D. Roiban, M. T. Reetz, J. Am. Chem. Soc. 2013, 135,
665-1668; b) G. de Gonzalo, I. Lavandera, K. Faber, W. Kroutil, Org.
[
[
1
1
[24] R. S. Cahn, C. Ingold, V. Prelog, Angew. Chem., Int. Ed. 1966, 5, 385-
415.
Lett., 2007, 9, 2163-2166; c) I. Lavandera, B. Höller, A. Kern, U. Ellmer,
A. Glieder, S. de Wildeman, W. Kroutil, Tetrahedron: Asymmetry, 2008,
[25] T. R. Hoye, C. S. Jeffrey, F. Shao, Nat. Protoc. 2007, 2, 2451.
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201801391
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
FULL PAPER
Jiyang Guo, Rui Zhang, Jingping
Ouyang, Feiting Zhang, Dr. Fengyu Qin,
Guigao Liu, Wenhe Zhang, Hengyu Li,
Xiaohong Ji, Prof. Xian Jia, Dr. Bin Qin,*
Prof. Song You*
Page No. – Page No.
By directly asymmetric reduction of (R)- and (S)-carvone using ketoreductases, all
four possible stereoisomers of carveol with medium to high diastereomeric
excesses (up to >99%) were first observed. Then previously studied and newly
engineered ene-reductases were applied to synthesis four stereoisomers of
dihydrocarvone. Asymmetric reduction of obtained dihydrocarvone isomers by
ketoreductases further provide access to all eight stereoisomeric dihydrocarveol.
Stereodivergent Synthesis of Carveol
and Dihydrocarveol through
Ketoreductases/Ene-Reductases
Catalysed Asymmetric Reduction
This article is protected by copyright. All rights reserved.