Loop-Grafted Old Yellow Enzymes in the Bienzymatic Cascade Reduction of Allylic Alcohols

Article abstract of DOI:10.1002/cbic.201500604

The enzymatic reduction of C=C bonds in allylic alcohols with Old Yellow Enzymes represents a challenging task, due to insufficient activation through the hydroxy group. In our work, we coupled an alcohol dehydrogenase with three wild-type ene reductases - namely nicotinamide-dependent cyclohex-2-en-1-one reductase (NCR) from Zymomonas mobilis, OYE1 from Saccharomyces pastorianus and morphinone reductase (MR) from Pseudomonas putida M10 - and four rationally designed β/α loop variants of NCR in the bienzymatic cascade hydrogenation of allylic alcohols. Remarkably, the wild type of NCR was not able to catalyse the cascade reaction whereas MR and OYE1 demonstrated high to excellent activities. Through the rational loop grafting of two intrinsic β/α surface loop regions near the entrance of the active site of NCR with the corresponding loops from OYE1 or MR we successfully transferred the cascade reduction activity from one family member to another. Further we observed that loop grafting revealed certain influences on the interaction with the nicotinamide cofactor.

Full text of DOI:10.1002/cbic.201500604

DOI: 10.1002/cbic.201500604
Communications
Loop-Grafted Old Yellow Enzymes in the Bienzymatic
Cascade Reduction of Allylic Alcohols
Sabrina Reich, Bettina M. Nestl, and Bernhard Hauer*[a]
The enzymatic reduction of C=C bonds in allylic alcohols with
Old Yellow Enzymes represents a challenging task, due to in-
sufficient activation through the hydroxy group. In our work,
we coupled an alcohol dehydrogenase with three wild-type
ene reductases—namely nicotinamide-dependent cyclohex-2-
en-1-one reductase (NCR) from Zymomonas mobilis, OYE1 from
Saccharomyces pastorianus and morphinone reductase (MR)
from Pseudomonas putida M10—and four rationally designed
b/a loop variants of NCR in the bienzymatic cascade hydroge-
nation of allylic alcohols. Remarkably, the wild type of NCR was
not able to catalyse the cascade reaction whereas MR and
OYE1 demonstrated high to excellent activities. Through the
rational loop grafting of two intrinsic b/a surface loop regions
near the entrance of the active site of NCR with the corre-
sponding loops from OYE1 or MR we successfully transferred
the cascade reduction activity from one family member to an-
other. Further we observed that loop grafting revealed certain
influences on the interaction with the nicotinamide cofactor.
toductus SA-01.[6] Here, the alcohol dehydrogenase (ADH) con-
verts the allylic alcohol into the corresponding aldehyde, with
the aid of NAD+ as cofactor, and this is followed by the reduc-
tion of the formed enal intermediate with the NADH-depen-
dent ene reductase and the further ADH-catalysed reduction
of the formed aldehyde to generate the corresponding alcohol
Scheme 1. Bienzymatic three-step cascade reaction for the reduction of cin-
namyl alcohol through the coupling of an alcohol dehydrogenase (ADH)
with ene reductase wild-type enzymes and loop-grafted variants.
Asymmetric hydrogenation is an important tool in synthetic
organic chemistry because it can be performed with a large va-
riety of substrates, as well as with experimental simplicity and
high levels of enantiocontrol. Asymmetric hydrogenation has
been applied to Rh- and Ru-catalysed enantioselective reduc-
tions of ketones/aldehydes, imines, activated olefins, aromatic
heterocycles and allylic alcohols.[1] Allylic alcohols are abundant
in natural sources, including essential oils, and are widely used
as starting materials and/or major components in the food, fra-
grance and pharmaceutical industries. Recently, several multi-
enzymatic cascade reactions using ene reductases, alcohol de-
hydrogenases and Baeyer–Villiger monooxygenases, employed
either as cell-free or whole-cell catalytic systems, have been
described.[2–4] Further, biocatalytic approaches for the asym-
metric reduction of allylic alcohols have been reported. Bakers’
yeast has been used for the selective reduction of cinnamyl al-
cohols and derivatives in a whole-cell system.[5]
product (Scheme 1).
Old Yellow Enzymes are a large and widespread enzyme
family known to catalyse the reduction of a wide range of sub-
strates with use of a nicotinamide cofactor.[8,9] The natural
asymmetric bioreduction proceeds by a ping-pong bi–bi mech-
anism, divided into reductive and oxidative half reactions (Fig-
ure S1 in the Supporting Information).[10] The substrate scope
of ene reductases, however, is limited to compounds contain-
ing an electron-withdrawing group that activates the C=C
bond, thus allowing the trans-specific enzymatic reduction.
Substrates with non-activating functional groups, such as allylic
alcohols, cannot directly be reduced by these enzyme cata-
lysts.
Ene reductase family members are based on a (b/a) triose-
phosphate isomerase (TIM) barrel structure. The active site
containing the prosthetic flavin mononucleotide (FMN) is locat-
ed at the carboxy-terminal end of the barrel and surrounded
by diverse and variable b/a surface loop regions.[11] Recently,
we succeeded in demonstrating that loop-engineered variants
of the nicotinamide-dependent cyclohex-2-en-1-one reductase
(NCR) from Zymomonas mobilis[11,12] with different loop lengths
(and thus amino acid compositions) alter the activity and the
overall stability of this enzyme.[13,14] Given the broad substrate
range of natural ene reductases, we set out to explore whether
this diversity could be leveraged by rational loop engineering
of NCR for the reduction of challenging allylic alcohols. There-
fore, two b/a surface loop regions near the entrance of the
Furthermore, the groups of Sacchetti and Hollmann present-
ed a three-step in vitro enzymatic redox isomerisation cascade
reaction based on alcohol dehydrogenase and ene reductase
enzyme catalysts.[6,7] The ene reductases applied in these two
studies include the two bakers’ yeast isoenzyme reductases
OYE2 and OYE3[7] and the thermophilic TsER from Thermus sco-
[a] Dr. S. Reich, Dr. B. M. Nestl, Dr. B. Hauer
Institute of Technical Biochemistry, Universität Stuttgart
Allmandring 31, 70569 Stuttgart (Germany)
E-mail: bernhard.hauer@itb.uni-stuttgart.de
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cbic.201500604.
ChemBioChem 2016, 17, 561 – 565
561
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
active site were selected for loop grafting experiments on the
basis of structure and sequence alignments of the three well-
known and described ene reductase family members: NCR,
OYE1 from Saccharomyces pastorianus[11,15] and morphinone
reductase (MR) from Pseudomonas putida M10[16,17] (Figures 1
and S2). The three selected ene reductases possess overall
sequence identities between 29.4% (NCR/OYE1) and 42.4%
(NCR/MR). The monomeric reductase NCR served as enzyme
scaffold for the generation of a total of four single loop-grafted
variants, named loop A_OYE1, loop A_MR, loop B_OYE1 and
loop B_MR.
In order to facilitate the asymmetric reduction of allylic alco-
hols in a bienzymatic cascade, we tested several commercially
available alcohol dehydrogenases for their ability to catalyse
the oxidation and reduction of a small set of substrates by
using NADH/NAD+ (data not shown). The commercially avail-
able ADH equine from Sigma–Aldrich turned out to be the
most active alcohol dehydrogenase (Table S1). Wild-type en-
zymes and the generated loop variants were purified and ap-
plied in in vitro cascade reactions with the ADH equine togeth-
er with cinnamyl alcohol, geraniol or (À)-perillyl alcohol as sub-
strates.
Of the three wild-type enzymes expressed, purified and
tested, only OYE1 and MR showed good to excellent activities
towards the reduction of the three allylic substrates. Surpris-
ingly, the activity of NCR wild type was fairly low (Table 1).
When the bienzymatic one-pot reduction reactions of the
three allylic alcohol substrates were performed with the gener-
ated NCR loop variants, however, good to excellent product
formation could be observed.
The grafting of the corresponding loop A and loop B regions
from OYE1 and MR, respectively, into the NCR scaffold had
therefore resulted in NCR variants with new activities. It is also
interesting to note that the reduction of (S)-(À)-perillyl alcohol
resulted in the formation of two different diastereomers of the
resulting shisool alcohol (Figure S3). The two wild-type en-
zymes OYE1 and MR demonstrated opposite stereochemical
behaviour in the reduction of (S)-(À)-perillyl alcohol. From liter-
ature data on the reduction of perillyl aldehyde and car-
Figure 1. Sequence and structure alignment of selected ene reductases NCR,
OYE1 and MR. Top: Excerpt of a multiple sequence alignment. Two loops, A
and B (highlighted in bold), near the active site were selected for loop graft-
ing. Bottom: Structural superposition of loop A (light blue) and loop B (dark
blue) of the three ene reductases NCR (PDB ID: 4A3U, cyan), MR (PDB ID:
1GWJ, orange) and OYE1 (PDB ID: 1OYA, purple). Catalytically active tyrosine
and prosthetic FMN are shown as black sticks. Pictures are generated with
PyMol.
Table 1. Product formation in in vitro one-pot bienzymatic reductions of cinnamyl alcohol, geraniol and (S)-(À)-perillyl alcohol with the selected ADH
equine and seven purified (purity >95%) ene reductase wild-type enzymes and variants.
Product formation [%]
Substrates
cinnamyl alcohol
geraniol
(S)-(À)-perillyl alcohol[a]
Products
wild type
3-phenylpropan-1-ol
citronellol
trans-shisool
cis-shisool
NCR
OYE1
MR
4.3Æ0.4
3.3Æ0.6
1.3Æ0.1
95.4Æ0.1
12.0Æ5.3
0.1Æ0.1
0.5Æ0.1
71.4Æ2.4
97.5Æ0.6
85.9Æ0.4
99.1Æ0.1
25.9Æ1.0
A_OYE1
A_MR
B_OYE1
B_MR
57.6Æ8.5
97.6Æ0.7
94.4Æ1.1
30.7Æ1.3
10.3Æ1.0
20.9Æ4.1
42.0Æ4.8
45.4Æ1.2
94.2Æ0.2
93.4Æ0.4
82.9Æ1.4
6.2Æ1.0
3.4Æ0.1
3.4Æ0.7
3.8Æ0.7
84.2Æ0.9
loop variants of NCR
Reactions were performed in triplicate in 24 h reaction time. [a] The reduction of (S)-(À)-perillyl alcohol resulted in the formation of two possible different
diastereomers (Figure S3).
ChemBioChem 2016, 17, 561 – 565
www.chembiochem.org
562
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
vone,[18,19] we assume that OYE1 produced nearly exclusively
trans-shisool (95.4%) and MR cis-shisool (71.4%, see Table 1).
Additional experiments should clarify these observations in
detail. By looking at the stereoselectivity of the generated
loop-grafted variants, we also observed that the ability to pro-
duce cis-shisool was transferred from MR to NCR during the
loop grafting of the loop B region (84.2% cis-shisool was
formed with loop B_MR). In contrast, the grafting of the loop A
region had no impact on the stereoselective output of the
enzyme. These results go in hand with results related to the in-
fluence of b/a surface loop regions on enzyme selectivity (un-
published data) previously obtained in our lab.
the reduced NADH cofactor was used to enable complete con-
version of the substrate. As model substrate cinnamaldehyde
was employed. This substrate demonstrated the highest con-
version rates in the reduction reaction with all enzymes tested
(Figure 2).
As expected, the highest level of product formation was ob-
tained with the NCR wild type in a reaction setup without any
oxidised cofactor present at the beginning of the biotransfor-
mation. Interestingly, NCR was found to be quite sensitive to-
wards the presence of oxidised NAD+ in the reaction mixture.
An increase in NAD+ led to a considerably decreased activity
These results served as the basis for examinations of NCR,
OYE1 and MR wild-type enzymes as well as the four loop-graft-
ed variants in the reduction of the formed activated aldehyde
intermediates. We thus tested the wild-type and loop-grafted
variants in the reduction of the two substrates cinnamalde-
hyde and geranial, corresponding intermediates in the cascade
reaction. Interestingly, NCR wild type clearly demonstrated the
highest product formation with both intermediate aldehydes
tested. Decreased reduction rates were obtained for OYE1 and
MR wild types and the generated NCR loop variants (Figure 2).
Therefore, we conclude that the grafting of b/a loop regions
from the less active biocatalysts OYE1 and MR into the NCR
Table 2. Product formation in the conversion of cinnamaldehyde with
NADH present in the reaction mixture, together with the percentage
alterations in product yields with different amounts of NAD+ present at
the beginning of the reaction (NADH/NAD+ ratio 4:1 and 1:1).
Cinnamaldehyde
reduction
reaction
Cofactor ratio NADH/NAD+
4:1 1:1
Yield of reduced Alteration in yield of reduced
products products relative to 1:0 [%]
1:0
wild type
NCR
OYE1
MR
98.4Æ1.6
À61.5
À80.7
À63.0
À47.3
À55.2
À56.4
À42.9
À67.0
59.4Æ0.8
34.9Æ0.7
92.6Æ4.6
89.0Æ8.8
56.6Æ0.2
39.1Æ2.1
À36.2
À38.1
À35.4
+0.2
loop variants A_OYE1
A_MR
B_OYE1
B_MR
À22.8
À2.0
For all performed reactions NADH was used in excess amount in order to
achieve complete conversion of the substrate. However, further reduction
of hydrocinnamaldehyde to the corresponding alcohol 3-phenylpropan-1-
ol has been observed with excess amounts of NADH. For detailed infor-
mation and deviations see Scheme S1 and Table S3.
of NCR (À80.7% with a ratio of 1:1, Table 2). MR and OYE1
wild-type enzymes and loop-grafted variants loop A_OYE1,
loop A_MR, loop B_OYE1 and loop B_MR demonstrated more
flexible behaviour than NCR towards the presence of oxidised
NAD+ (minor alterations in yield of reduced products, Table 2).
Because NADH represents an opportunistic substrate in the
ping-pong bi–bi reaction mechanism, we assume that the NCR
wild type should be inhibited by the oxidised NAD+. The
higher flexibility of OYE1 and MR towards the oxidised cofactor
could be transferred to the NCR backbone through the graft-
ing of the two b/a surface loops. These results indicate that
the loops of the ene reductases are possibly involved in the
interaction both with the reduced and with the oxidised nico-
tinamide cofactor. The exact type of interaction between the
b/a loop regions and the nicotinamide cofactor is currently
under investigation in our laboratory.
Figure 2. Initial activities for the reduction of the formed intermediates cin-
namaldehyde (black) and geranial (grey) with wild-type enzymes NCR, OYE1
and MR (solid) and the four designed loop-grafted variants loop A_OYE1,
loop A_MR, loop B_OYE1 and loop B_MR (dashed lines). Reactions were per-
formed in triplicate in 2.5 h reaction time (Table S2).
scaffold resulted in reduced activities for these substrates.
Despite the fact that NCR wild type was able to catalyse the
reduction of the aldehyde compounds, a reduced activity cou-
pled to ADH was observed. Thus, we assume that the nicotina-
mide cofactor might directly influence the activity of ene re-
ductases.[20] During the natural bioreduction, as well as in the
bienzymatic cascade setup, the required reduced nicotinamide
cofactor (NADH) is consumed both by the ene reductase and
by the alcohol dehydrogenase. In parallel, the oxidised form
(NAD+) accumulates in the reaction broth and might modulate
the redox balance of this cascade reaction. In order to eluci-
date the impact of the oxidised cofactor NAD+ on the reduc-
tion activity of ene reductase enzymes, biotransformation reac-
tions with different reduced/oxidised nicotinamide cofactor
ratios were performed. In all biotransformations an excess of
In conclusion, the grafting of b/a surface loop regions be-
tween different ene reductases enables the alteration and the
transfer of new reaction activities within these family members.
Furthermore, differences in the interaction with the nicotina-
mide cofactor were noted. We expect that the proposed con-
cept of loop grafting of flexible surface loop regions will stimu-
ChemBioChem 2016, 17, 561 – 565
www.chembiochem.org
563
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
OYE1 proteins was observed at a concentration of 850 mm ammo-
nium sulfate. The purification was followed by a filtration and
desalting step with Vivaspin ultrafiltration spin columns (Vivaspin
10 kDa, Sartorius, Gçttingen, Germany) and concentrated in
Tris·HCl reaction buffer (50 mm, pH 7.5). The purified enzyme was
stored at À208C until further use.
late the engineering of enzymes with modified and improved
enzyme properties.
Experimental Section
General remarks: Restriction endonucleases, Pfu DNA polymerase
and T4 DNA ligase were purchased from Fermentas (Thermo Scien-
tific). Primers were obtained from Metabion (Matrinsried, Germany).
Unless otherwise specified, chemicals were obtained from Fluka,
Sigma–Aldrich, Carl Roth, SAFC (Hamburg, Germany) or BASF SE.
GC-FID analyses were carried out with a Shimadzu GC-2010 instru-
ment equipped with an AOC-20i auto injector and a HP-5 capillary
column (Agilent technologies, 30 m”0.25 mm”0.25 mm) with H2
as carrier gas (1.38 mLminÀ1). Standard molecular biology tech-
niques such as overlapping extension PCR, restriction digestion, li-
gation, heat shock transformation as well as protein concentration
determination were used as previously described.[11,12]
Biotransformation reactions: All biotransformation reactions were
carried out with purified enzyme (protein purity >95%). Biotrans-
formations were performed with NADH (5 mm), purified enzyme
(100 mgmLÀ1), ADH equine (0.2 U, Sigma–Aldrich) and Tris·HCl
(1 mL, 50 mm, pH 7.5). The biotransformation was started by
adding substrate [2 mm (cinnamyl alcohol, geraniol, (S)-(À)-perillyl
alcohol]. After incubation (24 h, 308C, 180 rpm), the reaction was
stopped by extraction with methyl-tert-butylether (MTBE; 2”
500 mL). Activity tests towards cinnamaldehyde were performed
with cinnamaldehyde (2 mm), NADH (2.5 mm) and purified protein
(50 mgmLÀ1) in Tris·HCl (1 mL, 50 mm, pH 7.5) for 2.5 h at 308C and
180 rpm, followed by extraction with MTBE (2”500 mL). Assays
with varying cofactor concentrations contained cinnamaldehyde
(2 mm), NADH/NAD+ cofactor (1:0, 4:1, 1:1, 5 mm) and purified
enzyme (100 mgmLÀ1) in Tris·HCl (1 mL, 50 mm, pH 7.5) for 24 h at
308C and 180 rpm, and were stopped by extraction with MTBE (2”
500 mL).
Protein expression and purification: The two wild-type enzymes
NCR and MR, as well as the four loop-grafted variants, were cloned
in pET-28a(++) and expressed in TB medium with kanamycin
(50 mgmLÀ1) in Escherichia coli BL21(DE3). Enzyme expression was
induced at an OD600 of 0.5–0.6 with IPTG (0.2 mm) at 308C for 20 h.
After harvesting by centrifugation (9000g, 30 min, 48C), the cells
were disrupted with a French press (EmulsiFlex-C5, Avestin, Mann-
heim, Germany) and purified by immobilised metal affinity chroma-
tography with an ¾KTA system. OYE1 wild-type enzyme was
cloned into a pDHE vector and also expressed in E. coli BL21(DE3)
under the following conditions: 378C, 180 rpm in TB medium
(400 mL) containing ampicillin (100 mg). When the cultures
reached an OD600 of 1.3–1.6, they were supplemented with l-
rhamnnose (0.2%) for protein expression. After 14 h of incubation
at 308C (160 rpm), cells were harvested by centrifugation (8000g,
15 min, 48C) and resuspended in potassium phosphate buffer
(50 mm, pH 7.4) containing phenylmethanesulfonyl fluoride (PMSF,
0.1 mm). Cell pellets were disrupted in two or three cycles with
a French press (EmulsiFlex-C5, Avestin, Mannheim, Germany) at
48C. The resulting crude extracts were centrifuged (37000g,
30 min, 48C), and the supernatants with the soluble proteins were
recovered. Protein purification was performed in three steps:
1) ammonium sulfate precipitation, 2) fast protein liquid chroma-
tography (FPLC), and 3) hydrophobic interaction chromatography
(HIC). Protein precipitation of the lysate was performed with am-
monium sulfate (24% final concentration). Subsequently, the lysate
was centrifuged (7000g) for 15 min at 48C, and the protein pellet
was discarded. The ammonium sulfate concentration was then in-
creased to an end concentration of 45%, followed by centrifuga-
tion (7000g, 15 min, 48C), and the supernatant was discarded. Af-
terwards the proteins were resuspended in potassium phosphate
buffer (50 mm, pH 7.8). FPLC was carried with Q-Sepharose FF col-
umns (GE Healthcare) packed to a volume of 275 mL and a maxi-
mum flow of 20 mLminÀ1. The column was washed (10 mLminÀ1
working flow) by a step gradient protocol with potassium phos-
phate buffer (50 mm, pH 7.8) containing ammonium sulfate (1.4m).
The elution of the OYE1 proteins was observed at a concentration
of 400 mm ammonium sulfate. In addition to the characteristic
total protein detection at 280 nm, OYE1 was identified by its ab-
sorbance at 455 nm. HIC was performed with a phenyl Sepharose
HP column (GE Healthcare, Freiburg, Germany). The column was
packed with a maximum flow of 18 mLminÀ1 to a volume of
240 mL. The elution buffer contained potassium phosphate
(50 mm, pH 7.2) and ammonium sulfate (1.4m). The elution of the
GC-FID analytics: Levels of conversion were determined from the
percent area of the product. Samples were directly analysed from
the MTBE phase without derivatisation by using a HP-5 capillary
column (30 m”0.25 mm”0.25 mm, Agilent) and one of two differ-
ent temperature programs. The program for cyclic substrates (cin-
namyl alcohol, (S)-(À)-perillyl alcohol and cinnamaldehyde) was
708C, 158CminÀ1 to 2008C, 308CminÀ1 to 3208C, hold 1 min. The
program for the aliphatic substrate (geraniol) was 3 min at 608C,
108CminÀ1 to 1508C, 508CminÀ1 to 3008C, hold 1 min. The injector
temperature was 2508C. Compounds were detected by flame ioni-
sation detection(FID) at 3258C and identified by coelution with
standards.
Acknowledgements
We gratefully acknowledge financial support from the Landes-
graduiertenfçrderung (LSFG) Baden–Württemberg and the Euro-
pean Community’s Seven Framework Program (FP7/2007–2013)
for the Innovative Medicine Initiative under Grant Agreement
no. 115360 (Chemical manufacturing methods for the 21st centu-
ry pharmaceutical industries, Chem21).
Keywords: allylic alcohols · biocatalysis · cascade reactions ·
loop grafting · oxidoreductases · reduction
[1] R. Noyori, Angew. Chem. Int. Ed. 2002, 41, 2008–2022; Angew. Chem.
2002, 114, 2108–2123.
[2] N. Oberleitner, C. Peters, F. Rudroff, U. T. Bornscheuer, M. D. Mihovilovic,
J. Biotechnol. 2014, 192, 393–399.
[3] N. Oberleitner, C. Peters, J. Muschiol, M. Kadow, S. Saß, T. Bayer, P.
Schaaf, N. Iqbal, F. Rudroff, M. D. Mihovilovic, U. T. Bornscheuer, Chem-
CatChem 2013, 5, 3524–3528.
[4] C. Peters, R. Koelzsch, M. Kadow, L. Skalden, F. Rudroff, M. D. Mihovilov-
ic, U. T. Bornscheuer, ChemCatChem 2014, 6, 1021–1027.
[5] F. Luo, P. Wang, Y. Gong, Tetrahedron Lett. 2010, 51, 1693–1695.
[6] S. Gargiulo, D. J. Opperman, U. Hanefeld, I. W. C. E. Arends, F. Hollmann,
Chem. Commun. 2012, 48, 6630–6632.
ChemBioChem 2016, 17, 561 – 565
www.chembiochem.org
564
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
[7] E. Brenna, F. G. Gatti, D. Monti, F. Parmeggiani, A. Sacchetti, ChemCat-
Chem 2012, 4, 653–659.
[16] K. Saito, D. J. Thiele, M. Davio, O. Lockridge, V. Massey, J. Biol. Chem.
1991, 266, 20720–20724.
[17] P. C. E. Barna, T. Messiha, H. L. Petosa, C. Bruce, N. C. Scrutton, N. S.
Moody, J. Biol. Chem. 2002, 277, 30976–30983.
[8] K. Stuermer, R. Hauer, B. Hall, M. Faber, Curr. Opin. Chem. Biol. 2007, 11,
203–213.
[18] C. E. French, N. C. Bruce, Biochem. J. 1995, 312, 671–678.
[19] S. K. Padhi, D. J. Bougioukou, J. D. Stewart, J. Am. Chem. Soc. 2009, 131,
3271–3280.
[20] N. J. Mueller, C. Stueckler, B. Hauer, N. Baudendistel, H. Housden, N. C.
Bruce, K. Faber, Adv. Synth. Catal. 2010, 352, 387–394.
[21] M. Hall, C. Stueckler, W. Kroutil, P. Macheroux, K. Faber, Angew. Chem.
Int. Ed. 2007, 46, 3934–3937; Angew. Chem. 2007, 119, 4008–4011.
[9] N. S. Toogood, H. S. Gardiner, J. M. Scrutton, ChemCatChem 2010, 2,
892–914.
[10] R. M. Kohli, V. Massey, J. Biol. Chem. 1998, 273, 32763–32770.
[11] K. M. Fox, P. A. Karplus, Structure 1994, 2, 1089–1105.
[12] A. Müller, B. Hauer, B. Rosche, Biotechnology 2007, 98, 22–29.
[13] M. Hall, C. Stueckler, B. Hauer, R. Stuermer, T. Friedrich, M. Breuer, W.
Kroutil, K. Faber, Eur. J. Org. Chem. 2008, 1511–1516.
[14] S. Reich, H. W. Hoeffken, B. Rosche, B. M. Nestl, B. Hauer, ChemBioChem
2012, 13, 2400–2407.
Manuscript received: November 12, 2015
Accepted article published: January 14, 2016
Final article published: February 23, 2016
[15] S. Reich, N. Kress, B. M. Nestl, B. Hauer, J. Struct. Biol. 2014, 185, 228–
233.
ChemBioChem 2016, 17, 561 – 565
www.chembiochem.org
565
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim