Altering the substrate specificity of reductase CgKR1 from Candida glabrata by protein engineering for bioreduction of aromatic α-keto esters

Article abstract of DOI:10.1002/adsc.201300775

A versatile keto ester reductase CgKR1, exhibiting a broad substrate spectrum, was obtained from Candida glabrata by genome data mining. It showed the highest activity toward an aliphatic β-keto ester, ethyl 4-chloro-3-oxobutanoate (COBE), but much lower activity toward bulkier α-keto esters with an aromatic group, such as methyl ortho- chlorobenzoylformate (CBFM) and ethyl 2-oxo-4-phenylbutyrate (OPBE). By rational design of the active pocket, the substrate specificity of the reductase was significantly altered and this tailor-made reductase showed a much higher activity toward aromatic α-keto esters (～7-fold increase in k _cat/K_m toward CBFM) and lower activity toward aliphatic keto esters (～12-fold decrease in k_cat/K_m toward COBE). Meanwhile, the thermostability of the reductase was enhanced by a consensus approach. Such improvements may yield practical catalysts for the asymmetric bioreduction of these aromatic α-keto esters

Full text of DOI:10.1002/adsc.201300775

COMMUNICATIONS
DOI: 10.1002/adsc.201300775
Altering the Substrate Specificity of Reductase CgKR1 from
Candida glabrata by Protein Engineering for Bioreduction of
Aromatic a-Keto Esters
Lei Huang,+a Hong-Min Ma,+a Hui-Lei Yu,a,* and Jian-He Xua,*
a
Laboratory of Biocatalysis and Synthetic Biotechnology, State Key Laboratory of Bioreactor Engineering, East China
University of Science and Technology, 130 Meilong Road, Shanghai 200237, Peopleꢀs Republic of China
Fax : (+86)-21-6425 0840; e-mail: huileiyu@ecust.edu.cn or jianhexu@ecust.edu.cn
+
These authors contributed equally to this work.
Received: September 13, 2013; Revised: February 13, 2014; Published online: May 13, 2014
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201300775.
Abstract: A versatile keto ester reductase CgKR1,
exhibiting a broad substrate spectrum, was obtained
from Candida glabrata by genome data mining. It
showed the highest activity toward an aliphatic b-
keto ester, ethyl 4-chloro-3-oxobutanoate (COBE),
but much lower activity toward bulkier a-keto
esters with an aromatic group, such as methyl
ortho-chlorobenzoylformate (CBFM) and ethyl 2-
oxo-4-phenylbutyrate (OPBE). By rational design
of the active pocket, the substrate specificity of the
reductase was significantly altered and this tailor-
made reductase showed a much higher activity
toward aromatic a-keto esters (~7-fold increase in
kcat/Km toward CBFM) and lower activity toward
aliphatic keto esters (~12-fold decrease in kcat/Km
toward COBE). Meanwhile, the thermostability of
the reductase was enhanced by a consensus ap-
proach. Such improvements may yield practical cat-
alysts for the asymmetric bioreduction of these aro-
matic a-keto esters
gineering methods have been applied to generate bio-
catalysts with higher activity, improved thermostabili-
ty and/or better selectivity to strengthen the advan-
tages of biocatalysts and to extend their application in
the chemical and pharmaceutical industries.[3]
Recently, we discovered a versatile keto ester re-
ductase from Candida glabrata (CgKR1), which ex-
hibited a broad substrate spectrum.[4] It showed the
highest activity (114 U/mg protein) toward ethyl 4-
chloro-3-oxobutanoate (COBE, 10, as shown in
Figure 2). In contrast, when the substrates were aro-
matic a-keto esters with a bulky phenyl group, such
as methyl ortho-chlorobenzoylformate (CBFM, 1)
(Scheme 1) and ethyl 2-oxo-4-phenylbutyrate (OPBE,
4), the activity of CgKR1 decreased obviously by
nearly one order of magnitude. In order to enhance
the activity of CgKR1 toward the synthesis of these
useful chiral alcohols, a rational engineering strategy
was adopted to construct NDT libraries[5] of amino
acid residues located at the substrate binding site.
Given the absence of crystallographic data, three-
dimensional models of CgKR1 and Gre2p[6] were pre-
dicted by homology modeling using the crystal struc-
tures of a carbonyl reductase (SsCR) from Sporobolo-
Keywords: aromatic a-keto esters; protein engi-
neering; reductases; substrate specificity
Chiral alcohols are frequently required as important
intermediates for the introduction of chiral centers
into the pharmaceuticals, flavors, aroma and agricul-
tural chemicals, and specialty materials.[1] Enantiose-
lective ketone reduction is a reliable, scalable and
straightforward route to optically active alcohols. Bio-
catalysts are becoming preferred for ketone reduction
and the application of reductases in the commercial
synthesis of chiral alcohols has undergone a revolution
Scheme 1. Asymmetric reduction of methyl ortho-chloroben-
zoylformate (CBFM) with recombinant cells of E. coli/
over the past several years.[2] Furthermore, protein en- pCgKR1 and BmGDH.
Adv. Synth. Catal. 2014, 356, 1943 – 1948
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1943

COMMUNICATIONS
Lei Huang et al.
Table 1. Kinetic parameters of wild-type CgKR1 and variants with substrates CBFM and COBE.
Enzyme
CBFM
kcat [sÀ1]
COBE
Km [mM]
kcat/Km [sÀ1 mMÀ1]
Km [mM]
kcat [sÀ1]
kcat/Km [sÀ1 mMÀ1]
WT
2.25Æ0.31
5.84Æ1.05
1.53Æ0.33
1.97Æ0.28
74.0Æ2.0
918Æ72
56.0Æ4.0
482Æ30
32.9
157
36.6
245
1.00Æ0.24
12.3Æ0.8
0.60Æ0.08
10.7Æ1.6
310Æ26
442Æ12
162Æ6.0
288Æ32
310
36
270
27
M1[a]
M2[b]
M3[c]
[a]
The mutations are F92L/F94V.
The mutations are I99Y/G174A.
[b]
[c]
The mutations are F92L/F94V/I99Y/G174A.
myces salmonicolor.[7] In our previous work,[4] homol- mutated variant (CgKR1M1) exhibited higher activity
ogy modeling and docking analyses were performed than either of the two single-mutated variants, with-
to gain insights into the high selectivity of the out any substrate inhibition. To establish whether the
enzyme. The substrate molecule of methyl ortho- increased activity of the mutant M1 (F92L/F94V) was
chlorobenzoylformate (CBFM) was docked into the caused by the enhanced binding affinity and/or the
pocket of the homology modeled structure. According higher turnover rate of substrate, kinetic parameters
to the proposed catalytic mechanism of the short- for the enzyme-catalyzed reduction of CBFM were
chain alcohol dehydrogenase,[7] the carbonyl oxygen determined (Table 1). The results showed that the ap-
atom of the ketone substrate forms hydrogen bonds proximately 5-fold increase in activity (kcat/Km) of
with both Tyr and Ser residues and it is protonated CgKR1M1 resulted from the 12-fold increase in kcat
from the Tyr residue, followed by the attack of a hy- and the 2-fold increase in Km. This accounts for the
drogen atom from the C-4 atom of NADPH toward observed increase in specific activity of the mutant
the carbonyl carbon atom of the substrate. In the (109 U/mg protein) as compared to that of CgKR1-
structure of CgKR1/CBFM complex (Supporting In- WT (16 U/mg protein).
formation, Figure S1A), Ser134 and Tyr175 stabilize
Although the activity of the double-mutated
the substrate with hydrogen bonds. The aryl ring is mutant M1 toward CBFM and OPBE had increased,
embedded in a hydrophobic cavity next to the catalyt- its thermostability became worse. Importantly, en-
ic center, while the a-ester group is located in a small hancement of the enzymeꢀs thermostability results in
cavity mainly composed of F92, F94, N224 as well as increased catalytic lifetime or total turnover number
NADPH. It is obvious that the substrate binding (TTN).[9] Thus, it is necessary to improve the thermo-
pocket of CgKR1 is bigger than that of Gre2p, which stability of this mutated reductase. A number of pro-
might explain the higher activity of CgKR1 toward tein-engineering strategies has been employed in an
CBFM. It implies that the binding pocket could be attempt to improve protein stability.[10] Taking advant-
modified to change the substrate specificity, perhaps age of the large number of available protein sequen-
by rational design. To investigate the effect of differ- ces, the semi-rational “consensus approach” is a well-
ent amino acid side chains at the three selected posi- established strategy to improve protein thermostabili-
tions (F92, F94, N224) on CgKR1 activity, each of the ty.[11] The sequences that we picked range in amino
selected residues was replaced by 12 other residues acid identity from 31% to 62% with respect to the
(Phe, Leu, Ile, Val, Cys, Arg, Ser, Gly, Tyr, His, Asn, wild-type, which represent a significantly wide scope
and Asp). Three single-site libraries were generated (Supporting Information, Table S1).
with the help of NDT degenerate primers, comprising
From the output of the consensus analysis, six
12 codon variants, with a balanced mixture of aliphat- (S37T, I99Y, G174A, T153V, Y215F and H249F) of
ic and aromatic, polar and non-polar, positively and the 352 positions were distinct under a 50% consensus
negatively charged residues.[8]
cut-off. The single mutants and some combined mu-
Among the three libraries, two variants, F92L and tants were analyzed by a thermostability and activity
F94V, were identified, displaying higher activities assay (Supporting Information, Table S2).
toward CBFM, while there were no significant find-
Variant M2 (I99Y/G174A) was selected for further
ings in the N224 library compared to the wild-type. research considering both the thermostability and ac-
Although the leucine substitution engendered the de- tivity. Thus, M1 and M2 were combined for a quadru-
sired activity, the variant F92L was inhibited to some ple-mutated mutant M3 (F92L/F94V/I99Y/G174A).
extent by the high concentration of substrate CBFM Just as expected, the thermostability of M3 was en-
15
50
in further research (Km =0.17 mM, Kis =0.24 mM). hanced as shown by the increase in T (defined as
Nevertheless, we tried to combine the two mutations the temperature at which heat treatment for 15 min
of F92L and F94V. To our surprise, this new double- reduced the initial activity by 50%) by 2.38C as com-
1944
asc.wiley-vch.de
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2014, 356, 1943 – 1948

COMMUNICATIONS
Altering the Substrate Specificity of Reductase CgKR1
keto esters tested were 3~5 times higher (Figure 2).
Furthermore, substrates 1 and 10 were selected as the
representatives of aromatic keto esters and aliphatic
keto esters for the kinetic constants measurement.
The enhanced activity of CgKR1M3 for CBFM was
mainly due to its higher kcat value (7 times higher
than that of the wild-type, Table 1). Meanwhile, the
reduced activity of CgKR1M3 for COBE was mainly
due to its lower affinity (11-fold higher Km than that
of the wild-type, Table 1). To determine the stereose-
lectivity of the bioreduction by CgKR1M3, 0.5 mL re-
actions were performed using lyophilized CgKR1M3
cells in the presence of BmGDH and glucose for the
regeneration of NADPH.[12] Analysis of the stereose-
lectivity by HPLC or GC (Supporting Information,
Table S4) indicated that CgKR1M3 had almost the
same stereoselectivity toward the majority of tested
substrates except one aliphatic keto ester (substrate
Figure 1. Thermostability of the purified WT and variants
M1, M2 and M3 of CgKR1 as displayed by the residual ac-
tivity curves.
15
pared to the wide-type (T : 41.88C) and 3.98C to M1 6) (Figure 2).
50
15
50
(TT : 40.28C) (Figure 1). The kinetic study showed
To demonstrate the synthetic potential of variant
that mutant M3 exhibited a slightly lower Km value CgKR1M3, the bioreduction of CBFM by either
for CBFM and ~8-fold increase in kcat, when com- CgKR1 wild-type or variant M3 was performed on
pared to those of the wide-type.
a 100-mL scale for the preparation of methyl (R)-
Using the purified enzyme, the substrate profile of ortho-chloromandelate [(R)-CMM], a key chiral inter-
M3 was explored in comparison with the wild-type mediate for the synthesis of (S)-clopidogrel. The same
enzyme. The activities of M3 for aliphatic keto esters amount of lyophilized cells of E. coli/pCgKR1 WT
were drastically lower than those of the wild-type and M3 were utilized to transform CBFM at 100 gLÀ1
enzyme while the activities toward all the aromatic a- in 100 mL phosphate buffer with 1.5 equivalents of
Figure 2. Specific activity (in unit of U/mg protein, upper) and stereoselectivity (% ee, down) of CgKR1-WT and variant
CgKR1M3 toward aromatic and aliphatic keto esters.
Adv. Synth. Catal. 2014, 356, 1943 – 1948
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1945

COMMUNICATIONS
Lei Huang et al.
In order to unravel the differences in substrate spe-
cificity between CgKR1M3 and CgKR1-WT, we mod-
eled the CBFM into the active sites of CgKR1 WT
and M3. Analysis of the most favorable docking poses
of CgKR1 revealed striking differences in the sub-
strate binding pocket. The binding pocket of M3 var-
iant (Figure 4A) is obviously larger than that of wild-
type (Figure 4B). The substitution of the large amino
acid F92/F94 with smaller leucine and valine appa-
rently expands the cavity packaging the ester group,
and seems to allow the substrates to adopt the catalyt-
ic conformation more easily and thus contributes to
the enhanced kcat observed experimentally as kcat is
positively related to the rate of formation of covalent
intermediates,[13] which is a determinant for the in-
creased activity.[14] Meanwhile, the significant increase
in the Km for ethyl 4-chloro-3-oxobutanoate (COBE)
can contribute to the decreased activity toward ali-
phatic keto esters. The expanded cavity enabling the
accommodation of the phenyl group may be compara-
tively large for the smaller aliphatic group resulting in
its lower affinity (Figure 4C, D).[15] Given that both
substrates 1 and 10 are converted to the correspond-
Figure 3. Progress curves of asymmetric reduction of CBFM
&
with lyophilized cells of E. coli/pCgKR1-WT ( ) and E.
À1
~
coli/pCgKR1M3 ( ). Reaction conditions: CBFM 100 gL ,
d-glucose 1.5 equiv., lyophilized cells of E. coli/pCgKR1
1.5 g, lyophilized BmGDH powders 1.5 g, KPB 100 mL
(pH 6.0, 100 mM), 258C. pH was kept at 6.0 with 1M
Na2CO3.
glucose, and lyophilized powders of BmGDH (cell- ing R-enantiomers, the binding modes for 1 and 10
free extract) was used to complete the requisite re- are supposed to be that the aromatic substituent of
generation of cofactor by catalyzing the oxidation of 1 and the ester moiety of substrate 10 occupy the
glucose. The initial rates were relatively slow because same location. It suggests the different location of the
of the impermeability of the lyophilized cells. The ester groups of these two substrates, which gives an
speeds rose after ~0.5 h along with the increased per- explanation to the much more favorable binding of
meability. When the biocatalyst loading was 15 gLÀ1, the enzyme with the aromatic substrates after opening
CgKR1-WT and CgKR1M3 resulted in >99% conver- up the small binding pocket. The mutant site Ile99 is
sion within 4 h and 6 h, respectively (Figure 3). The at the surface of CgKR1 whereas Gly174 is in the
lyophilized cells were then lowered to 7.5 gLÀ1, in inside of CgKR1 near the surface (Figure 5). This
spite of the slightly longer time, CgKR1M3 could finding underscores the important role that the pro-
transform all the substrate into optically pure (R)- tein surface plays on stability as found in many other
CMM after 4.5 h while CgKR1-WT only resulted in enzymes with improved thermostability.[16] I99Y may
89% conversion. When the enzyme loading was fur- increase the hydrogen bonds with water, or increase
ther decreased to 5 gLÀ1, CgKR1-WT and CgKR1M3 the hydrophilicity of the enzyme.[17] Some degree of
gave 57% and 81% conversions, respectively the stabilization may due to the packing effect or the
(Table 2).
increased hydrophobic interaction as reported for
Table 2. Asymmetric reduction of CBFM with lyophilized cells of E. coli/pCgKR1 WT and M3.[a]
Entry
1
Enzyme
Substrate [gLÀ1]
Cell [gLÀ1][b]
Time [h]
Conversion [%][c]
ee [%)][d]
WT
M3
WT
M3
WT
M3
100
100
100
100
100
100
15
15
7.5
7.5
5.0
5.0
6
4
5
4.5
10
10
>99
>99
89
>99
57
98.7 (R)
98.7 (R)
98.7 (R)
98.7 (R)
98.7 (R)
98.7 (R)
2
3
81
[a]
Reaction conditions: CBFM (100 gLÀ1), d-glucose (1.5 equiv.), lyophilized cells of E. coli/pCgKR1, lyophilized BmGDH
powders (the same amount as lyophilized cells of E. coli/pCgKR1, in excess), 100 mL KPB (pH 6.0, 100 mM), 258C, pH
was kept at 6.0 with 1M Na2CO3.
The quantity of lyophilized cells of E. coli/pCgKR1.
Determined by GC analysis.
[b]
[c]
[d]
Determined by HPLC analysis.
1946
asc.wiley-vch.de
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2014, 356, 1943 – 1948

COMMUNICATIONS
Altering the Substrate Specificity of Reductase CgKR1
Figure 4. Molecular docking of substrates CBFM and COBE into the active site of CgKR1 WT (A, C) and mutant CgKR1
F92L/F94V (B, D). The cavity and crevice of CgKR1 are represented by a grey surface.
binding pocket, which presumably facilitates the bio-
reduction of aromatic a-keto esters. Further improve-
ment of the enzyme thermostability via protein engi-
neering is currently underway.
Experimental Section
Generation of Mutagenesis
Mutations were introduced by PCR into the pET28-CgKR1
template DNA[4] using the QuickChange (Stratagene) ac-
cording to the manufacturerꢀs instructions. The primers used
in this study are listed in the Supporting Information, Table
S3. Additional combination mutations were introduced in
subsequent rounds of PCR. Mutagenesis was confirmed by
DNA sequencing (Shanghai Sunny Biotechnology Co. Ltd,
China). The mutant proteins were expressed and purified
similarly as the wild-type.
15
50
T
Determinations
Figure 5. The model structure of CgKR1. Catalytic triad:
Ser/Tyr/Lys.
15
50
T
data were obtained by filling a row in a 96-well plate
with 100 mL per well of enzyme at 1 mgmLÀ1. The PCR
plate was sealed with microseal, and a thermocycler was
used to apply a temperature gradient for 15 min. The PCR
plate was then immediately cooled on ice. Then, the residual
G174A.[18] In this case the activity and thermostability
of CgKR1 were enhanced by these mutations.
In summary, the activity and thermostability of
CgKR1 toward aromatic a-keto esters were improved
by a structure-based rational design approach and
consensus approach. Here we have confirmed that
Phe92 and Phe94 are important in substrate recogni-
tion of CgKR1 while Ile99 and Gly174 are influential
residues for CgKR1 thermostability. CgKR1M3 dis-
plays higher activity and thermostability simultane-
ously toward the examined aromatic a-keto esters. Ki-
netic analysis and substrate docking could partially
provide insights into the mechanism of the mutation
effects: the mutations at positions 92 and 94 cause sig-
activity of the enzymes was measured with the standard pro-
15
tocol. The T was estimated as the temperature at which
50
heat treatment for 15 min reduced the initial activity by
50%.[19]
Consensus Approach
The sequence alignment was submitted to the online Com-
parative Sequence Analysis (http://coot.embl.de/Alignment/
consensus.html), and the threshold was set at 50% (Support-
ing Information, Figure S2). Wild-type residues were mutat-
ed to the consensus amino acid if they also fit a number of
criteria, such as not destroying salt bridges or helix, located
nificant changes in the geometry of the substrate- in more than 6 ꢂ from the cofactor binding site.[11a] The con-
Adv. Synth. Catal. 2014, 356, 1943 – 1948
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1947

COMMUNICATIONS
Lei Huang et al.
firmed and putative or probable amino acid sequences were
obtained by BLASTP programs on NCBI using amino acid
sequence of CgKR1 as the query and aligned with CLUS-
TALW software.
[9] a) T. A. Rogers, A. S. Bommarius, Chem. Eng. Sci.
2010, 65, 2118–2124; b) I. Wu, F. H. Arnold, Biotechnol.
Bioeng. 2013, 110, 1874–1883.
[10] a) D. Timo, S. Marlen, B. Dominique, U. T. Bornsche-
uer, Curr. Opin. Chem. Biol. 2013, 17, 215–220;
b) M. T. Reetz, M. Bocola, J. D. Carballeira, D. X. Zha,
A. Vogel, Angew. Chem. 2005, 117, 4264–4268; Angew.
Chem. Int. Ed. 2005, 44, 4192–4196; c) N. Declerck, M.
Machius, G. Wiegand, R. Huber, C. Caillardin, J. Mol.
Biol. 2000, 301, 1041–1057; d) G. P. Zhu, C. Xu, M. K.
Teng, L. M. Tao, X. Y. Zhu, C. J. Wu, J. Hang, L. W.
Niu, Y. Z. Wang, Protein Eng. 1999, 12, 635–638;
e) A. M. Rojkova, A. G. Galkin, L. B. Kulakova, A. E.
Serov, P. A. Savitsky, V. V. Fedorchuk, V. I. Tishkov,
FEBS Lett. 1999, 445, 183–188; f) A. V. Gribenko,
M. M. Patel, J. J. Liu, S. A. McCallum, C. Y. Wang, G. I.
Makhatadze, Proc. Natl. Acad. Sci. USA 2009, 106,
2601–2606.
Acknowledgements
This work was financially supported by the National Natural
Science Foundation of China (No. 21276082), Ministry of
Science and Technology, P. R. China (Nos. 2011CB710800 &
2012AA022201), and Shanghai Commission of Science and
Technology (No. 11431921600).
References
[1] Y. Ni, J. H. Xu, Biotechnol. Adv. 2012, 30, 1279–1288.
[2] a) G. W. Huisman, J. Liang, A. Krebber, Curr. Opin.
Chem. Biol. 2010, 14, 122–129; b) K. Ishihara, H. Ya-
mamoto, Jpn. Kokai Tokkyo Koho JP2005218348, 2005;
c) L. J. Giver, L. M. Newman, E. Mundorff, G. W. Huis-
man, S. J. Jenne, J. Zhu, B. Behrouzian, J. H. Grate, J.
Lalonde, PCT Int. Appl. WO2008042876A2, 2008.
[3] a) E. T. Johnson, S. Ryu, H. Yi, B. Shin, H. Cheong, G.
Choi, Plant J. 2001, 25, 325–333; b) E. Campbell, S.
Chuang, S. Banta, Protein. Eng. Des. Sel. 2012, 26, 181–
186.
[11] a) E. V. Figueroa, J. C. Riggers, A. S. Bommarius,
ChemBioChem 2007, 8, 2295–2301; b) M. Anbar, O.
Gul, R. Lamed, U. O. Sezerman, E. A. Bayer, Appl.
Environ. Microbiol. 2012, 78, 3458–3464; c) D. L.
Zhang, F. Y. Zhu, W. C. Fan, R. S. Tao, H. Yu, Y. L.
Yang, W. H. Jiang, S. Yang, Appl. Microbiol. Biotech-
nol. 2011, 90, 1361–1371.
[12] N. D. Shen, Y. Ni, H. M. Ma, L. J. Wang, C. X. Li,
G. W. Zheng, J. Zhang, J. H. Xu, Org. Lett. 2012, 14,
1982–1985.
[13] A. Westerbeek, W. Szymanski, H. J. Wijma, S. J. Mar-
rink, B. L. Feringa, D. B. Janssena, Adv. Synth. Catal.
2011, 353, 931–944.
[4] H. M. Ma, L. L. Yang, Y. Ni, J. Zhang, C. X. Li, G. W.
Zheng, H. Y. Yang, J. H. Xu, Adv. Synth. Catal. 2012,
354, 1765–1772.
[14] L. F. Godinho, C. R. Reis, R. van Merkerk, G. J. Poelar-
ends, W. J. Quax, Adv. Synth. Catal. 2012, 354, 3009–
3015.
[5] a) M. T. Reetz, L. W. Wang, M. Bocola, Angew. Chem.
2006, 118, 1258–1263; Angew. Chem. Int. Ed. 2006, 45,
1236–1241; b) M. T. Reetz, M. Bocola, L. W. Wang, J.
Sanchis, A. Cronin, M. Arand, J. Y. Zou, A. Archelas,
A. L. Bottalla, A. Naworyta, S. L. Mowbray, J. Am.
Chem. Soc. 2009, 131, 7334–7343; c) A. G. Sandstrom,
Y. Wikmark, K. Engstrom, J. Nyhlen, J. E. Backvall,
Proc. Natl. Acad. Sci. USA 2012, 109, 78–83; d) C. K.
Savile, J. M. Janey, E. C. Mundorff, J. C. Moore, S. Tam,
W. R. Jarvis, J. C. Colbeck, A. Krebber, F. J. Fleitz, J.
Brands, P. N. Devine, G. W. Huisman, G. J. Hughes, Sci-
ence 2010, 329, 305–309; e) M. T. Reetz, S. Wu, Chem.
Commun. 2008, 43, 5499–5501.
[15] P. T. Xie, T. D. Hurley, Protein Sci. 1999, 8, 2639–2644.
[16] a) A. Jakoblinnert, A. V. D. Wittenboer, A. V. Shi-
vange, M. Bocola, L. Heffele, M. A. Schumacher, U.
Schwaneberg, J. Biotechnol. 2013, 165, 52–62; b) R.
Martinez, F. Jakob, R. Tu, P. Siegert, K. H. Maurer, U.
Schwaneberg, Biotechnol. Bioeng. 2013, 110, 711–720;
c) A. Bosshart, S. Panke, M. Bechtold, Angew. Chem.
Int. Ed. 2013, 52, 9673–9676; d) C. Vieille, G. J. Zeikus,
Microbiol. Mol. Biol. Rev. 2001, 65, 1–43.
[17] M. J. McLachlan, T. W. Johannes, H. M. Zhao, Biotech-
[6] a) T. Ema, N. Okita, S. Ide, T. Sakai, Org. Biomol.
Chem. 2007, 5, 1175–1176; b) T. Ema, S. Ide, N. Okita,
T. Sakai, Adv. Synth. Catal. 2008, 350, 2039–2044.
[7] S. Kamitori, A. Iguchi, A. Ohtaki, M. Yamada, K. Kita,
J. Mol. Biol. 2005, 352, 551–558.
[8] M. T. Reetz, D. Kahakeaw, R. Lohmer, ChemBioChem
2008, 9, 1797–1804.
nol. Bioeng. 2008, 99, 268–274.
[18] T. W. Johannes, R. D. Woodyer, H. M. Zhao, Appl. En-
viron. Microbiol. 2005, 71, 5728–5734.
[19] M. T. Reetz, J. D. Carballeira, A. Vogel, Angew. Chem.
2006, 118, 7909–7915; Angew. Chem. Int. Ed. 2006, 45,
7745–7751.
1948
asc.wiley-vch.de
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2014, 356, 1943 – 1948