Exploiting Cofactor Versatility to Convert a FAD-Dependent Baeyer–Villiger Monooxygenase into a Ketoreductase

Article abstract of DOI:10.1002/anie.201907606

Cyclohexanone monooxygenases (CHMOs) show very high catalytic specificity for natural Baeyer–Villiger (BV) reactions and promiscuous reduction reactions have not been reported to date. Wild-type CHMO from Acinetobacter sp. NCIMB 9871 was found to possess an innate, promiscuous ability to reduce an aromatic α-keto ester, but with poor yield and stereoselectivity. Structure-guided, site-directed mutagenesis drastically improved the catalytic carbonyl-reduction activity (yield up to 99 %) and stereoselectivity (ee up to 99 %), thereby converting this CHMO into a ketoreductase, which can reduce a range of differently substituted aromatic α-keto esters. The improved, promiscuous reduction activity of the mutant enzyme in comparison to the wild-type enzyme results from a decrease in the distance between the carbonyl moiety of the substrate and the hydrogen atom on N5 of the reduced flavin adenine dinucleotide (FAD) cofactor, as confirmed using docking and molecular dynamics simulations.

Full text of DOI:10.1002/anie.201907606

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Exploiting Cofactor Versatility to Convert a FAD Dependent
Baeyer−Villiger Monooxygenase into a Ketoreductase
Authors: Jian Xu, Yongzhen Peng, Zhiguo Wang, Yujing Hu, Jiajie Fan,
He Zheng, Xianfu Lin, and Qi Wu
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: Angew. Chem. Int. Ed. 10.1002/anie.201907606
Angew. Chem. 10.1002/ange.201907606
Link to VoR: http://dx.doi.org/10.1002/anie.201907606
http://dx.doi.org/10.1002/ange.201907606

10.1002/anie.201907606
Angewandte Chemie International Edition
COMMUNICATION
Exploiting Cofactor Versatility to Convert a FAD Dependent
Baeyer−Villiger Monooxygenase into a Ketoreductase
Jian Xu+, Yongzhen Peng+, Zhiguo Wang, Yujing Hu, Jiajie Fan, He Zheng, Xianfu Lin, Qi Wu*
Abstract: Cyclohexanone monooxygenases (CHMOs) show very
high catalytic specificity for natural Baeyer−Villiger (BV) reactions,
but promiscuous reduction reactions have not been reported to date.
In order to reach this goal, we focused mechanistically on the flavin
adenine dinucleotide (FAD) as the cofactor of CHMOs, which can
switch between oxidation and reduction states. We achieved such a
novel promiscuous reduction activity of CHMOs by converting
(quinone; Flox), reduced (hydroquinone; Flred) and peroxidized
(peroxyflavin; FlOOH).[11] As outlined in Figure 1a, traditional
oxygenation reactions catalyzed by CHMO begin with the
combination of Flred and O2 to form FlOOH. After one atom of
oxygen is incorporated into ketones, the FlOOH is converted into
FLox by water elimination, and then FLox undergoes reduction by
NADPH leading to the regeneration of Flred. In nature, enoate
reductases from the family of old yellow enzymes (OYEs) are a
class of flavin-containing enzymes that can catalyze asymmetric
reduction of activated C=C bonds by Flred (Figure 1b).[12]
Recently, Hyster and coworkers have discovered phenylacetone
monooxygenase (PAMO) could be converted into a reductase to
catalyze an enantioselective radical dehalogenation reaction.[3a]
Inspired by these advances, we hypothesized that the catalytic
cycle of CHMO could be broken, enabling Flred to be capable of
catalyzing ketone reduction reactions of suitable substrates by
CHMOAcineto into
a
ketoreductase. Rational structure-guided
engineering of CHMOAcineto was implemented in order to drastically
improve the catalytic reduction activity (yield up to 99%) and
stereoselectivity (e.e. up to 99%).
Enzymes are known to be powerful catalysts in organic
synthesis and pharmaceutical chemistry because of their high
efficiency and selectivity.[1] In contrast to traditional
organometallic catalysts, they offer attractive alternatives due to
their low toxicity and mild reaction conditions. However, in most
cases, enzymes catalyze only the respective natural reaction
types. Inducing additional reactivities as “enzymatic promiscuity”
remains a great challenge.[2] Driven by the development of
protein engineering and advances derived from mechanistic
studies, new notable cases of catalytic promiscuity have been
reported recently, leading to higher synthetic utility.[3-6] For
hydride transfer from N5.
(a)
R
N
O
H
N
O
O
NH
O
1a
N
H
O
Flred
OH
O
O
H
P
D
A
N
(R)-2a
O2
example,
nicotinamide-dependent
ketoreductases
were
employed to catalyze asymmetric dehalogenation by irradiating
with light.[5] Cytochrome P450 monooxygenases and
hemoglobins have been engineered into new artificial enzymes
for constructing a series of useful chemical bonds, such as chiral
C-C, C-N, C-Si and C-B bonds.[6] Inspired by these significant
advances, we report herein an unprecedented promiscuous
R
N
R
N
R
N
N
O
O
N
O
O
-H2O
N
O
NH
NH
NH
N
N
H
HO
N
H
O
OH
O
O
O
Flox
FlOOH
O
case of
a Baeyer−Villiger monooxygenase (BVMO) as a
(b)
OYEs
This work
mechanistically novel biocatalyst in enantioselective carbonyl
reduction. To date such reactivity has not been observed in any
other FAD-dependent monooxygenases known in nature.
O
H
N
H
R1 R2
R1 R2
R3 R4
H
R1 R2
O
R1 R2
HN
O
R3 R4
OH
N
R
N
H
As one of the most studied BVMOs,[7] cyclohexanone
monooxygenase (CHMO), which can catalyze in vivo the BV
transformation of cyclohexanone to caprolactone, among many
other reactions, has been successfully engineered to increase
its activity, thermal stability, and enantioselectivity by directed
evolution.[8] However, CHMO is quite specific for its natural
reaction type. Other catalytic activities such as the oxidations of
thioethers or boronic acid were performed in the similar
mechanism with the natural one.[9] In addition, Fraaije and
coworkers converted a BVMO into a NADPH oxidase by protein
engineering.[10] As the active site of CHMO, flavins are extremely
versatile molecules containing multiple states, such as oxidized
Flred
Figure 1. (a) The catalytic cycle of a Baeyer-Villiger reaction and the proposed
mechanism of flavin-dependent carbonyl reduction. (b) Flred-catalyzed
reduction reaction.
Optically active α-hydroxy carbonyl compounds are important
chiral building blocks in pharmaceutical chemistry and chemical
biology.[13] We therefore chose the α-keto ester 1a as an ideal
model substrate for our study. CHMO from Acinetobacter sp.
NCIMB 9871 (CHMOAcineto)
[7, 14] was first tested under anaerobic
conditions with E. coli as host organism. We were delighted to
find that the reduction of 1a appeared to be feasible, although
the observed reaction efficiency was poor (15% yield and 84%
e.e. in favor of (R)-2a, Table 1, entry 1,). Due to the low activity
of CHMOAcineto, we turned our attention to PAMO,[15] and found
the whole-cell system of PAMO displayed a significant increase
in promiscuous activity toward 1a (Table 1, entry 2). However,
the control experiment of E. coli (TOP 10) in the absence of
PAMO also afforded the reduced product 2a with the same
[a]
[+]
J. Xu+, Y. Peng+, Y. Hu, J. Fan, He Zheng, X. Lin & Q. Wu
Department of Chemistry, Zhejiang University
Hangzhou 310027 (China).
E-mail: wuqi1000@163.com
Dr. Z. Wang, Institute of Aging Research, School of Medicine,
Hangzhou Normal University, Hangzhou 311121 (China).
These authors contributed equally to this work.
This article is protected by copyright. All rights reserved.

10.1002/anie.201907606
Angewandte Chemie International Edition
COMMUNICATION
yield and e.e. value as PAMO (Table 1, entry 3). This significant
background reaction prevented us from investigating the
reduction activity of PAMO in the whole-cell system. The
influence of the host cells of CHMOAcineto (BL21-DE3 or BL21-
DE3 with plasmid pET-22b (+)) was also evaluated. Only trace
amounts of reduced product 2a with low stereoselectivity were
observed (Table 1, entry 4-5). Upon isolating, purifying and
testing the enzyme, the e.e. value improved significantly (up to
99%), pointing to the influence of whole-cell system (Table 1,
entry 6). In yet another control experiment with CHMOAcineto
which was devoid of FAD cofactor, no reduced product was
observed.[16] GDH was also tested separately in the model
reaction providing no reduction product.
Table 1: Screening the conditions for the promiscuous reduction of 1a [a]
Entry Condition
Yield (%) e.e. (%)
1
2
3
4
5
6
7
8
CHMOAcineto-WT, BL21-BE3[a]
15
99
99
10
11
6
84
12
12
45
51
99
0
PAMO-WT, Top 10[a]
Top 10[a]
BL21-DE3[a]
BL21-DE3-pET 22b (+)[a]
Purified-CHMOAcineto, GDH[b]
Purified-CHMOAcineto devoid of FAD, GDH[b] [c]
GDH[d]
Figure 3. Assessing the introduction of volume-based rationally designed
mutagenesis at the chosen hot spots. The experiment details were shown in
supporting information.
0
0
0
[a] 1a (0.01 mmol) was dissolved in 20 μL acetonitrile, then added to 2 mL
cell cultures (OD600 ≈ 3.0) under anaerobic condition for 4 h at 20 oC; the
yields and e.e values were determined by chiral HPLC; [b] 0.01 mmol 1a, 20
μL acetonitrile, 1% NADP+, 10 mg glucose, 0.01 μmol CHMO and 1 μmol
GDH were dissolved in 2 mL PBS buffer (50 mM, pH=7.4) under anaerobic
condition for 20 h at 20 oC; [c] treated with urea and KBr, supporting
information; [d] without Purified-CHMO.
Figure 4. CHMOAcineto-catalyzed reduction of α-Keto esters. (a) The progress
curve of reduction catalyzed by WT-CHMO and L143F. (b) TON of WT and
L143F for 1a reduction. (c) The Influence of temperature with L143F. (d)
Reaction yields catalyzed by whole-cell catalyst, cell lysate, or purified enzyme
of L143F. The experiment details were shown in supporting information.
Figure 2. Active site of CHMOAcineto, homology model based on the X-ray
structure of CHMO from Rhodococcus sp. (PDB code: 3UCL[19]
)
series of residues around the binding pocket including L143,
L144, F246, F277, A428, F432, L435 and F505 were chosen as
hotspots for protein engineering to improve the reduction activity
(Figure 2). A focused rationally designed library was constructed
based on a volume-scanning strategy using alanine (A), leucine
(L), and phenylalanine (F). These were regarded as small,
medium and large amino acids, replacing the amino acids
individually at hotspots in the anticipation that this would
optimally modify the space of the substrate-binding pocket. As
With the aim of increasing product formation, we sought to
engineer CHMOAcineto by rational design[17] rather than directed
evolution.[18] If successful, large mutant libraries as in directed
evolution would not have to be screened. As the crystal structure
of CHMOAcineto is not available, we built a homology model
(named as CHMOhomo) using the X-ray structure of CHMO from
Rhodococcus sp. (PDB code: 3UCL),[19] which shows high
sequence similarity around the active sites with CHMOAcineto. A
This article is protected by copyright. All rights reserved.

10.1002/anie.201907606
Angewandte Chemie International Edition
COMMUNICATION
shown in Figure 3, several mutants in the volume-based library
were found to have noticeable effects on the activity and
stereoselectivity. Under the same reaction conditions, L144A,
F246A and F246L displayed slightly increased reduction yields
(25-30%) with high stereoselectivity (80-90% e.e.) compared
with the WT. Remarkably, the stereoselectivity of L277A was
reversed from (R)- into (S)-configurational preference. Among
the hot positions tested, residue L143 was considered as a
crucial spot for improving activity, with the best mutant L143F
exhibiting a notable yield of 62% and an e.e. of 98% (R).
2a, Yield: 95%,
2b, Yield: 99%,
2c, Yield: 61%,
e.e.: 98%
e.e.: 99%
e.e.: 85 %
With the best mutant in hand, the properties of the biocatalytic
system were assessed. The time-reaction curves showed that
the reaction catalyzed by L143F proceeds much faster than WT
(95% vs. 18% conversion in 8 hours, Figure 4a). The catalytic
efficiency of L143F, measured by kinetic parameters (Table S3),
was improved remarkably relative to WT-CHMOAcineto (36.30 s-
2d, Yield: 57%,
2e, Yield: 64%,
2f, Yield: 95%,
e.e.: 93 %
e.e.: 45%
e.e.: 99%
1mM-1 vs 0.30 s-1mM-1). And L143F exhibited
a 20-fold
improvement in TON compared with WT (Figure 4b). We then
investigated the influence of temperature on the formation of
(R)-2a in the whole-cell systems. The results showed that the
structure of CHMOAcineto is unstable with increasing temperature,
the reduction activity diminishing considerably over 30 oC
(Figure 4c). Notably, although the stereoselectivity achieved by
purified CHMOAcineto is excellent, the reaction rates catalyzed by
purified protein or cell lysate proved to be much slower than
those in the whole-cell system (Figure 4d, Figure S3). We
speculate that the cell membrane periplasm may protect enzyme
from possible inactivation by substrate, or keep the enzyme in a
more reducing environment preventing the formation of Fl-OOH
species.[6a,f] Surprisingly, this transformation could proceed both
in aerobic and anaerobic conditions (Table S5). However, the
catalytic activities in aerobic conditions are slightly lower than
anaerobic conditions. Additional experiment (Table S6) ensured
that 1a is hardly to be oxidized by L143F. Thus, the reaction was
considered to be implemented under air. The decrease of
reduction activity may be caused by the formation of FLOOH
and the transformation of uncoupling.
2g, Yield: 51%,
2h, Yield: 35%,
2i, Yield: 77%,
e.e.: 77%
e.e.: 99%
e.e.: 32%
2k, Yield: 76%,
2j, Yield: 94%
2l, Yield: 55%,
e.e.: 98%
e.e.: 94%
e.e.: 99%
2m, Yield: n.d.
[a] Reaction conditions: 1 (0.01 mmol) was dissolved in 20 μL acetonitrile,
then added to 2 mL cell cultures under anaerobic condition for 8-12 h at 20
oC; [b] the ee values of 2 were determined by chiral HPLC.
O
HO
D
(a)
L143F
O
O
GDH, NADP+
Glucose, D2O
O
O
Next, we explored the substrate scope of this unnatural
reduction reaction catalyzed by CHMOAcineto in the whole-cell
system. As illustrated in Table 2, a wide range of differentially
substituted aromatic α-keto esters accommodating electron-rich
and electron-deficient groups were readily converted into the
corresponding alcohols in moderate to good yields and
50% deuterium incorporation
Yield: 38% e.e: 99%
O
HO
D
L144F
O
O
GDH, NADP+
O
O
D1-Glucose, H2O
11% deuterium incorporation
Yield: 18% e.e: 99%
seteroselectivity.
Notably,
substrates
bearing
chlorine
substituents at different positions (ortho-, meta-, para-) of the
phenyl ring were all accepted by L143F and displayed excellent
(R)-selectivity (2b-2d). α-Keto esters characterized by a different
length of the alcohol parts also underwent reduction with high
yield and satisfactory stereoselectivity (2a, 2j, 2k), while the
methyl ester is relatively poor with respect of stereoselectivity
(2i). The reduction of the 1,2-diketone 1l into the α-hydroxy
ketone 2l occurs with moderate yield and high e.e. value (99%).
It is noteworthy that propiophenone (1m) could not be reduced
by CHMOAcineto (2m), probably due to the absence of some
interactions between the β-carbonyl function and enzyme
binding pocket. Large scale reactions of 1a and 1l (100 mg)
were also implemented with good yields (2a: 93%, 2l: 50%) and
(b)
R
N
H
N
R
N
D
N
O
D2O
O
NH
NH
N
H
N
D
O
O
Flred
D-Flred
Figure 5. Kinetic isotope effect studies.
Mechanistic investigations were first performed by isotope
incorporation (Figure 5). When D1-glucose was used to generate
D-Flred, only 18% yield was observed with 99% e.e. value and
11% deuterium incorporation. We considered that the absence
of complete deuterium incorporation was due to the H-D
exchange between N5 of D-Flred and H2O.[20] This was also
confirmed with the transformation in which D2O was used as a
solvent under otherwise standard conditions: 38% yield of 2a
high stereoselectivity (2a: 98% e.e., 2l: 90% e.e.).
Table 2: Substrate scope of the promiscuous reduction reaction[a][b]
This article is protected by copyright. All rights reserved.

10.1002/anie.201907606
Angewandte Chemie International Edition
COMMUNICATION
was obtained with 99% e.e. and 50% deuterium incorporation.
Rate study of deuterium incorporation experiments was
implemented, and the results showed the reducation rate in the
isotope incorporation reaction with D1-glucose proceeds
significantly slower compared with that of D2O (Figure S4). Thus,
the reduction rate with D1-glucose might be much slower than H-
D exchange between the reduced flavin with the solvent (H2O),
leading to less deuterated product (11%). On the other hand, in
the isotope incorporation reaction with D2O, the higher reactive
rate resulted in more deuterated product (50%). This result
supports a mechanism in which Flred is regenerated in situ and
acts as a hydrogen atom donor to enable the reduction of
appropriate carbonyl compounds. Further insight into the
reduction mechanism was obtained by using density functional
theory (DFT) calculations. These results indicate that the
mechanism we propose is also thermodynamically favorable
(∆Go = −13.96 kcal·mol−1, Figure S5 and Table S8).
strategy at hotspots was successfully implemented to improve
this promiscuous function with minimal screening effort. The
origin of enhanced reduction activity of CHMOAcineto mutants was
rationalized by computational simulation. Future research will
focus on extending the application of this new biocatalytic
reaction.
Acknowledgements
The financial support from National Natural Science Foundation
of China (21574113) and Zhejiang Provincial Natural Science
Foundation (LY19B020014) is gratefully acknowledged.
Keywords: Cyclohexanone monooxygenases
•
Catalytic
promiscuity • Protein engineering • Reductase • Volume-based
library
[1]
Reviews of enzyme catalysis in organic and pharmaceutical chemistry:
a) Biocatalysis in the Pharmaceutical and Biotechnology Industries,
(Eds: R. N. Patel), CRC Press, 2006; b) A. Goswami, J. D. Stewart,
Organic Synthesis Using Biocatalysts, Academic Press, 2015; c)
Enzyme Catalysis in Organic Synthesis (Eds: K. Drauz, H. Gröger, O.
May), Wiley-VCH, Weinheim, 2012; d) K. Faber, Biotransformations in
Organic Chemistry: A Textbook, Springer, Stuttgart, 2011; e) A. Liese,
K. Seelbach, C. Wandrey, Industrial Biotransformations, Wiley-VCH,
Weinheim, 2006.
[2]
[3]
a) I. Nobeli, A. D. Favia, J. M. Thornton, Nat. Biotechnol. 2009, 27, 157-
167; b) K. Hult, P. Berglund, Trends Biotechnol. 2007, 25, 231-238: c)
U. T. Bornscheuer, R. J. Kazlauskas, Angew. Chem. Int. Ed. 2004, 43,
6032-6040.
a) B. A. Sandoval, A. J. Meichan, T. K. Hyster, J. Am. Chem. Soc. 2017,
139, 11313-11316; b) K. F. Biegasiewicz, S. J. Cooper, M. A.
Emmanuel, D. C. Miller, T. K. Hyster, Nat. Chem. 2018, 10, 770-775;
c) B. A. Sandoval, S. I. Kurtoic, M. M. Chung, K. F. Biegasiewicz, T. K.
Hyster, Angew. Chem.-Int. Edit. 2019, doi: 10.1002/anie.201902005; d)
J. Vilim, T. Knaus, F. G. Mutti, Angew. Chem.-Int. Edit. 2018, 57,
14240-14244; e) M. Engleder, G. A. Strohmeier, H. Weber, G.
Steinkellner, E. Leitner, M. Müller, D. Mink, M. Schurmann, K. Gruber,
H. Pichler, Angew. Chem.-Int. Edit. 2019, 58, 7480-7484.
[4]
H. Jochens, K. Stiba, C. Savile, R. Fujii, J. G. Yu, T. Gerassenkov, R. J.
Kazlauskas, U. T. Bornscheuer, Angew. Chem.-Int. Edit. 2009, 48,
3532-3535.
Figure 6. MD simulations for gaining insight into the origins of activity
improvement with WT-CHMO (a) and L143F (b).
[5]
[6]
M. A. Emmanuel, N. R. Greenberg, D. G. Oblinsky, T. K. Hyster, Nature
2016, 540, 414-+.
Finally, the docking and MD experiments were implemented to
gain insight into the origins of high activity of L143F in
comparison with WT. As shown in Figure 6, when compared
with WT-CHMO, the nature of the binding pocket is dramatically
altered by mutation L143F, especially the possible T-shaped π-π
interaction between L143F and NADP+.[19] The average distance
between the carbon atom in carbonyl moiety of 1a and the
hydrogen atom on N5 of Flred is clearly shorter in L143F than in
WT (3.8 Å in L143F vs 9.8 Å in WT), according to the analysis of
25 ns trajectory of MD simulation (Figure S6). This is a
requirement for the substrate to undergo hydride transfer from
Flred with formation of the reduction product 2a.
a) S. B. J. Kan, X. Y. Huang, Y. Gumulya, K. Chen, F. H. Arnold,
Nature 2017, 552, 132-+; b) C. K. Prier, R. K. Zhang, A. R. Buller, S.
Brinkmann-Chen, F. H. Arnold, Nat. Chem. 2017, 9, 629-634; c) S. B. J.
Kan, R. D. Lewis, K. Chen, F. H. Arnold, Science 2016, 354, 1048-
1051; d) R. K. Zhang, K. Chen, X. Huang, L. Wohlschlager, H. Renata,
F. H. Arnold, Nature 2019, 565, 67-72; e) I. Cho, Z.-J. Jia, F. H. Arnold,
Science 2019, 364, 575-578; f) A. Tinoco, V. Steck, V. Tyagi, R. Fasan,
J. Am. Chem. Soc. 2017, 139, 5293-5296; g) R. Fasan, Nat. Chem.
2017, 9, 609-611.
[7]
Selected reviews of BVMOs: a) K. Balke, A. Beier, U. T. Bornscheuer,
Biotechnol. Adv., 2018, 36, 247-263; b) J. Dong, E. Fernández-Fueyo,
F. Hollmann, C. E. Paul, M. Pesic, S. Schmidt, Y. Wang, S. Younes, W.
Zhang, Angew. Chem. Int. Ed., 2018, 57, 9238-9261; c) Z. G. Zhang, L.
P. Parra, M. T. Reetz, Chem. Eur. J., 2012, 18, 10160-10172; d) H.
Leisch, K. Morley, P. C. K. Lau, Chem. Rev., 2011, 111, 4165-4222; e)
D. E. Torres Pazmiño, H. M. Dudek, M. W. Fraaije, Curr. Opin. Chem.
Biol., 2010, 14, 138-144.
In summary, we report a mechanistically novel biocatalytic
carbonyl
reduction
catalyzed
by
a
Baeyer−Villiger
monooxygenase (BVMO). It is a new case of a promiscuous
enzyme-catalyzed transformation, enabled by the cofactor FAD
that has not been observed in any other FAD-dependent
monooxygenase known in nature. Whereas WT-CHMOAcineto
displayed only a very limited ability to catalyze the model
reaction 1a→2a, rational design based on a volume-scanning
[8]
Selected studies of protein engineering of BVMOs: a) M. T. Reetz, B.
Brunner, T. Schneider, F. Schulz, C. M. Clouthier, M. M. Kayser,
Angew. Chem.-Int. Edit. 2004, 43, 4075-4078; b) D. J. Opperman, M. T.
Reetz, Chembiochem 2010, 11, 2589-2596; c) S. Schmidt, M. Genz, K.
Balke, U. T. Bornscheuer, J. Biotechnol. 2015, 214, 199-211; d) G. Li,
This article is protected by copyright. All rights reserved.

10.1002/anie.201907606
Angewandte Chemie International Edition
COMMUNICATION
M. J. L. J. Furst, H. R. Mansouri, A. K. Ressmann, A. Ilie, F. Rudroff, M.
D. Mihovilovic, M. W. Fraaije, M. T. Reetz, Org. Biomol. Chem. 2017,
15, 9824-9829; e) A. Stamm, A. Biundo, B. Schmidt, J. Brücher, S.
Lundmark, P. Olsén, L. Fogelström, E. Malmström, U.T. Bornscheuer,
P.O. Syrén, ChemBioChem 2019; DOI: 10.1002/cbic.201900046. f) Y.
Hu, J. Wang, Y. Cen, H. Zheng, M. Huang, X. Lin, Q. Wu, Chem.
Comm. 2019, 55, 2198-2201.
[9]
a) B. P. Branchaud, C. T. Walsh, J. Am. Chem. Soc. 1985, 107, 2153-
2161. b) M. T. Reetz, F. Daligault, B. Brunner, H. Hinrichs, A. Deege,
Angew. Chem.-Int. Edit. 2004, 43, 4078-4081.
[10] P. B. Brondani, H. M. Dudek, C. Martinoli, A. Mattevi, M. W. Fraaije, J.
Am. Chem. Soc. 2014, 136, 16966-16969.
[11] a) S. A. B. Dockrey, A. R. H. Narayan, Tetrahedron 2019, 75, 1115-
1121; b) E. Romero, J. R. G. Castellanos, G. Gadda, M. W. Fraaije, A.
Mattevi, Chem. Rev. 2018, 118, 1742-1769.
[12] a) K. Kitzing, T. B. Fitzpatrick, C. Wilken, J. Sawa, G. P. Bourenkov, P.
Macheroux, T. Clausen, J. Biol. Chem. 2005, 280, 27904-27913; b) H.
S. Toogood, N. S. Scrutton, Curr. Opin.Chem. Biol. 2014, 19, 107-115.
[13] X. Ren, H. Du, J. Am. Chem. Soc. 2016, 138, 810-813.
[14] C. T. Walsh, Y. C. J. Chen, Angew. Chem.-Int. Edit. 1988, 27, 333-343.
[15] M. W. Fraaije, J. Wu, D. Heuts, E. W. van Hellemond, J. H. L. Spelberg,
D. B. Janssen, Appl. Microbiol. Biotechnol. 2005, 66, 393-400.
[16] M. Krzek, H. L. van Beek, H. P. Permentier, R. Bischoff, M. W. Fraaije,
Enzyme Microb. Technol. 2016, 82, 138-143.
[17] a) J. Pleiss in Enzyme Catalysis in Organic Synthesis, 3rd. (Eds.: K.
Drauz, H. Gröger, O. May), Wiley-VCH, Weinheim, 2012, pp. 89-117; b),
T. Ema, Y. Nakano, D. Yoshida, S. Kamata, T. Sakai, Org. Biomol.
Chem. 2012, 10, 6299-6308; c) K. Steiner, H. Schwab, Comput. Struct.
Biotechnol. 2012, 2, e201209010.
[18] Selected reviews of directed evolution: a) N. J. Turner, Nat. Chem. Biol.
2009, 5, 568-574; b) M. T. Reetz, Angew. Chem.-Int. Edit. 2011, 50,
138-174; c) A. Currin, N. Swainston, P. J. Day, D. B. Kell, Chem. Soc.
Rev. 2015, 44, 1172-1239; d) C. A. Denard, H. Ren, H. Zhao, Curr.
Opin. Chem. Biol. 2015, 25, 55-64; e) F. H. Arnold, Angew. Chem. Int.
Edit. 2018, 57, 4143-4148; f) C. Zeymer, D. Hilvert, Annu. Rev.
Biochem. 2018, 87, 131-157; g) U. T. Bornscheuer, B. Hauer, K. E.
Jaeger, U. Schwaneberg, Angew. Chem.-Int. Edit. 2019, 58, 36-40.
[19] B. J. Yachnin, T. Sprules, M. B. McEvoy, P. C. K. Lau, A. M. Berghuis,
J. Am. Chem. Soc. 2012, 134, 7788-7795.
[20] Z. Y. Zhang, L. Chen, L. Liu, X. Y. Su, J. D. Rabinowitz, J. Am. Chem.
Soc. 2017, 139, 14368-14371.
This article is protected by copyright. All rights reserved.

10.1002/anie.201907606
Angewandte Chemie International Edition
COMMUNICATION
COMMUNICATION
Baeyer−Villiger Monooxygenase-
catalyzed carbonyl reduction that
has not been observed in any other
FAD-dependent monooxygenase
known in nature, was demonstrated in
this work. Rational structure-guided
engineering of CHMOAcineto was
implemented in order to drastically
improve the catalytic reduction activity
(yield up to 99%) and stereoselectivity
(e.e. up to 99%).
Jian Xu+, Yongzhen Peng+, Zhiguo
Wang, Yujing Hu, Jiajie Fan, He Zheng,
Xianfu Lin, Qi Wu*
Page No. – Page No.
Exploiting Cofactor Versatility to
Convert an FAD Dependent
Baeyer−Villiger Monooxygenase into
a Ketoreductase
This article is protected by copyright. All rights reserved.