Two subtle amino acid changes in a transaminase substantially enhance or invert enantiopreference in cascade syntheses

Article abstract of DOI:10.1002/cbic.201500074

Amine transaminases (ATAs) are powerful enzymes for the stereospecific production of chiral amines. However, the synthesis of amines incorporating more than one stereocenter is still a challenge. We developed a cascade synthesis to access optically active

Full text of DOI:10.1002/cbic.201500074

DOI: 10.1002/cbic.201500074
Communications
Two Subtle Amino Acid Changes in a Transaminase
Substantially Enhance or Invert Enantiopreference in
Cascade Syntheses
Lilly Skalden,[a] Christin Peters,[a] Jonathan Dickerhoff,[b] Alberto Nobili,[a] Henk-Jan Joosten,[c]
Klaus Weisz,[b] Matthias Hçhne,[d] and Uwe T. Bornscheuer*[a]
Amine transaminases (ATAs) are powerful enzymes for the ste-
reospecific production of chiral amines. However, the synthesis
of amines incorporating more than one stereocenter is still a
challenge. We developed a cascade synthesis to access optical-
ly active 3-alkyl-substituted chiral amines by combining two
asymmetric synthesis steps catalyzed by an enoate reductase
and ATAs. The ATA wild type from Vibrio fluvialis showed only
modest enantioselectivity (14% de) in the amination of (S)-3-
methylcyclohexanone, the product of the enoate-reductase-
catalyzed reaction step. However, by protein engineering we
created two variants with substantially improved diastereo-
selectivities: variant Leu56Val exhibited a higher R selectivity
(66% de) whereas the Leu56Ile substitution caused a switch in
enantiopreference to furnish the S-configured diastereomer
(70% de). Addition of 30% DMSO further improved the selec-
tivity and facilitated the synthesis of (1R,3S)-1-amino-3-methyl-
cyclohexane with 89% de at 87% conversion.
cade reaction. This biomimetic approach represents an exciting
recent development in “white biotechnology”.[3] In a cascade,
intermediate downstream and purification steps are unneces-
sary, and so cascades also represent a very promising approach
from a green chemistry point of view. For example, Sehl et al.
used the combination of a thiamine diphosphate (ThDP)-de-
pendent acetohydroxy acid synthase and amine transaminases
to produce the 1,2-amino alcohols pseudo- and norpseudo-
ephedrine.[4] Several other cascades with up to 13 different en-
zymes have been reported recently, thus showing the general
applicability of this strategy.[5] A great advantage is the modu-
larity of this approach, because different combinations of en-
zymes with distinct enantiopreferences result in a spectrum of
products with different diastereomeric configurations.
We were interested in the synthesis of ring-substituted exo-
cyclic amines because they are employed as building blocks
for the production of pharmaceuticals and fine chemicals.[6]
Typically, in reactions employing ATAs, acyclic ketones contain-
ing one large and one small substituent apiece have been
used as substrates,[7] and the molecular mechanisms governing
substrate recognition and enantioselectivity are fairly well un-
derstood: the active sites of an R- and an S-selective ATA differ
in the localization of the small and the large binding pocket,
thus forcing the ketone into one specific binding mode, lead-
ing to the formation of a distinct amine enantiomer.
The enantioselective synthesis of chiral primary amines has ad-
vanced remarkably during the last decade,[1] and some enzy-
matic processes can now outperform well-established chemical
processes. One benchmark example is the manufacture of Sita-
gliptin, the active pharmaceutical ingredient (API) of Januvia,
which demonstrates the potential of amine transaminase tech-
nology.[2]
In contrast, only a few reports on the amination of (substi-
tuted) cyclic ketones have been published.[6b,8] Structurally,
cyclic ketones are challenging with respect to activity and ste-
reoselectivity[9] because they do not possess the well-defined
large and small substituents that are usually important for ach-
ieving highly selective transamination. A few cyclic substrates
with large substituents (e.g., 2-(1H-benzoimidazol-2-yl)-1-meth-
ylpiperidin-4-one,[6b] 2-[2-(3,4-dimethoxyphenyl)ethoxy]cyclo-
hexanone,[6a] or methyl (3-oxocyclohexyl)acetate[8]) were inves-
tigated recently, but cyclic compounds only bearing single
small alkyl substituents have not been explored. Therefore, we
envisioned the synthesis of 1-amino-3-methylcyclohexanes (3)
as model compounds by combining two asymmetric synthesis
steps based on 1) an enoate reductase (ERED), and 2) an amine
transaminase and starting from 3-methylcyclohex-2-enone (1,
Scheme 1).
Many APIs contain multiple stereocenters, and their prepara-
tion is therefore challenging. An elegant solution with regard
to atom economy is the combination of different (bio)catalytic
asymmetric synthesis reactions facilitating the stepwise crea-
tion of several stereogenic centers. If reaction conditions are
compatible, the synthesis can be carried out as a one-pot cas-
[a] L. Skalden, C. Peters, A. Nobili, Prof. Dr. U. T. Bornscheuer
Biotechnology and Enzyme Catalysis, Institute of Biochemistry
Felix-Hausdorff-Strasse 4, 17487 Greifswald (Germany)
E-mail: uwe.bornscheuer@uni-greifswald.de
[b] J. Dickerhoff, Prof. Dr. K. Weisz
Analytical Biochemistry, Institute of Biochemistry
Felix-Hausdorff-Strasse 4, 17487 Greifswald (Germany)
[c] Dr. H.-J. Joosten
Bio-Prodict
Nieuwe Marktstraat 54E, 6511 AA Nijmegen (The Netherlands)
An efficient cascade requires: 1) a highly enantioselective
ERED, 2) that the precursor 1 should not be an ATA substrate
(or otherwise, 1 must be fully consumed in the first step), to
avoid the formation of byproducts, and 3) that the ATA should
introduce the amino group with the desired enantiopreference
[d] Prof. Dr. M. Hçhne
Protein Biochemistry, Institute of Biochemistry
Felix-Hausdorff-Strasse 4, 17487 Greifswald (Germany)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cbic.201500074.
ChemBioChem 0000, 00, 0 – 0
1
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Communications
after 18 h. The observed compo-
sition of 30:26:40:4 for the four
amine diastereomers 3a–3d
clearly shows a low enantioselec-
tivity in the amination reaction.
Remarkably, the ratios of the
(1R)- to the (1S)-amines of 30:26
and 40:4 indicate that ATA-Vfl in-
troduces the amino group with
R enantiopreference (see below;
Figure S4A and B in the Sup-
porting Information). This is in
contrast with the well-docu-
Scheme 1. The combination of an ERED with an ATA in a cascade reaction enables access to optically pure
1-amino-3-methylcyclohexanes. Glucose dehydrogenase (GDH) can be used for cofactor recycling in the ERED-cat-
alyzed reaction, whereas lactate dehydrogenase (LDH) and GDH allow a shift of the equilibrium in the ATA-cata-
lyzed reaction. Note that different combinations of EREDs and ATAs with complementary enantiopreferences are
required to obtain all four diastereomers. The boxed diastereomers were obtained in this study.
mented[16]
S enantiopreference
of ATA-Vfl and the previously re-
ported high S selectivity in the
conversion of the related six-
membered cyclic ketone methyl
and with high enantioselectivity, preferentially regardless of
the configuration of the 3-methyl group. Enoate reductases,
which are flavin-mononucleotide-containing (FMN-containing)
NAD(P)H-dependent oxidoreductases, are well known to con-
vert a/b-unsaturated ketones or aldehydes into the corre-
sponding enantiopure compounds.[10] ATAs introduce amine
groups, often with excellent enantioselectivities,[7b,11] into a
wide range of ketones, but enantioselectivity drops drastically
if the size difference between the two substituents on either
side of the carbonyl group is too small or if the ketones are
small cyclic molecules such as our target ketone 2. If (dia)ste-
reoselectivity is not sufficient for a ketone bearing a chiral
center in the a- or b-position, this can in principle be improved
by protein engineering. One example was shown by Limanto
et al., who used ATA to convert 2-[2-(3,4-dimethoxyphenyl)-
ethoxy]cyclohexanone;[6a] after three rounds of evolution selec-
tive enzymes for both chiral centers were found.
(3-oxocyclohexyl)acetate.[8] Obviously, the type and size of sub-
stituents at the b-position in a cyclic ketone such as 2 strongly
influence enantioselectivity and -preference in the amination
reaction.
To identify amino acid positions governing enantioselectivity,
we modeled the internal aldimine intermediate of 2 into the
structure of ATA-Vfl by using YASARA[17] (Figure 1A). From
visual inspection, we chose Leu56 as target for mutagenesis
because its side chain would be in contact with the methyl
substituent in the PLP intermediates of 3b and 3c, so muta-
tions of Leu56 might induce a change in selectivity. Analysis of
the amino acid distribution at position 56 by use of the 3DM
database[14] identified six amino acids (Ile, Val, Ala, Trp, Tyr, Phe)
besides leucine as most frequently occurring in PLP fold type I
proteins (Figure 1B). In a subset containing only subfamilies
harboring S-selective ATA sequences, leucine is found in 90%
of the sequences. These six mutants were generated by Mega-
Whop PCR, overexpressed in E. coli Bl21 (DE3), and purified by
His-tag affinity chromatography (Figures S1–S3). Only the
Leu56Ile, Leu56Val, and Leu56Ala mutants showed sufficient
activity. Because of its lower activity, in combination with enan-
tioselectivities that were similar to those of the Leu56Val
mutant, the Leu56Ala mutant was also excluded.
The aim of this work was to demonstrate the applicability of
the envisioned cascade and, in this context, to identify enan-
tiocomplementary ATAs, preferably with high stereoselectivity
towards 2. Identification of suitable ATAs can be achieved by
methods of protein discovery[7b,12] and protein engineer-
ing,[2a,13] which can be guided by bioinformatics tools. One of
these tools is the 3DM software package, which connects
a high-quality, structure-guided sequence alignment of a pro-
tein superfamily[14] with bioinformatics analysis tools. 3DM inte-
grates data from various sources, including sequences, struc-
tural information, protein ligand contacts, and mutational data
from the literature, using a unified amino acid residue number-
ing system for all proteins. We have previously used this plat-
form to guide the degree of randomization for the creation of
“small, but smart” libraries in multiple-site saturation mutagen-
esis experiments.[15] In this work, we have used information
from 3DM to guide the identification of variants of the ATA
from Vibrio fluvialis (ATA-Vfl) with suitable stereoselectivity and
enantiopreference for the synthesis of different diastereomers
of 3, because an initial screening had identified this ATA as
having sufficient activity. Unfortunately, ATA-Vfl exhibited only
moderate selectivity, with rac-2 having been fully converted
GC analysis revealed that variant Leu56Ile showed an im-
proved 1R enantioselectivity in the amination of the R enantio-
mer of rac-2 (Table 1, entry 2). Interestingly, the enzyme ami-
nated (S)-2 with inverted enantiopreference. The Leu56Val mu-
tation improved the R enantioselectivity in the amination of
(S)-2, but had no effect on the R enantiomer of 2 (Table 1,
entry 3, Figure S4B). In the case of 2, the carbonyl group is
part of a flexible aliphatic ring: 2 is an almost symmetric com-
pound because only a small methyl group is present in the b-
position. Thus, the typical steric constraints originating from
the interaction of a substrate bearing well-defined small and
large substituents with the enzyme’s active site[7a,16b] do not
apply for 2. This difference could lead to the switch of the
commonly observed stereoselectivity of the ATA from V. fluvia-
lis resulting in the R configuration here. We were thus able to
show that subtle changes in the active site geometry produced
&
ChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
2
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communications
Figure 1. A) View of the active site of ATA-Vfl together with the modeled external aldimine of 2. The PLP intermediate and Leu56 are shown as magenta and
green sticks, respectively. The cyan spheres indicate which hydrogen atom is replaced by a methyl group if one of the amines 3a–3d is bound to PLP. Note
that the side-chain Arg415 can also adopt a conformation flipping outside the active site.[20] B) Amino acid distribution of PLP fold type I proteins at position
56 of ATA-Vfl. In the 3DM database this position corresponds to number 46.
periments. In accordance with
Table 1. Diastereomeric purities of the products of the transamination of substrate 2 and of the simultaneous
cascade reactions starting from 1 with employment of OYE as the ERED and ATA-Vfl or its mutants in the ab-
sence or presence of the organic solvents DMF or DMSO.
the results described above, ap-
plication of the Leu56Val variant
increased the diastereomeric
purity of (1R,3S)-3a from 14% de
(WT) to 65% de (mutant). In con-
trast, the mutant Leu56Ile gave
the (1S,3S)-amine 3b with 70%
de (Table 1, entries 8 and 9, Fig-
ures S5–S7).
Diastereomeric composition
Enzyme
Substrate DMSO DMF
t
3a
3b
3c
3d
de
Conversion
[%]
[%]
[%]
[h] (1R,3S) (1S,3S) (1R,3R) (1S,3R) [%][a]
1
2
3
4
5
6
7
8
9
ATA-Vfl
ATA Leu56Ile
ATA Leu56Val
OYE+ATA-Vfl
OYE+ATA Leu56Ile
OYE+ATA Leu56Val
OYE+ATA-Vfl
OYE+ATA Leu56Ile
OYE+ATA Leu56Val
rac-2
rac-2
rac-2
0
0
0
1
1
1
0
0
0
1
1
1
0
0
0
1
1
1
0
0
0
18 30
18
26
47
16
44
86
17
43
85
18
12
41
6
40
46
39
4
n.d.
5
n.a.
n.a.
n.a.
12
71
66
14
70
65
76
99
99
99
99
97
99
97
95
98
88
81
87
7
18 40
48 56
48 14
48 83
48 57
48 15
48 82
72 88
72 59
72 94
1
1
1
1
1
1
1
1
1
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
Because the ERED reaction ex-
clusively produced the S enan-
tiomer of 2, it was possible to
quantify the selectivity of the
amine transaminase by deter-
mining the ratio of cis and trans
10 OYE+ATA-Vfl
11 OYE+ATA Leu56Ile
12 OYE+ATA Leu56Val
30
30
30
18
89
1
products 3a and 3b. All H NMR
n.d.: not detected. [a] The enantioselectivity of the ERED was >99% ee for the S enantiomer.
resonances of both diastereo-
mers in the mixture were as-
signed through DQF-COSY ex-
by small side chain variations at position 56 (located at the
boundary of the two binding pockets) significantly affected
diastereoselectivity.
periments. The downfield-shifted H1 protons of the cis and the
trans diastereomer significantly differ in their chemical shifts,
by nearly 0.5 ppm, and were subsequently used for a stereo-
chemical assignment with a one-dimensional NOESY version
employing selective refocusing (Figure 2). Thus, on excitation
of the H1 resonance at 2.62 ppm in the selective 1D NOESY ex-
periment, NOEs develop for H3 and H5a as well as for the two
directly neighboring H2a and H6a protons (Figure 2B). This
places H1 and H3 on the same side of the cyclohexane ring
system in close proximity, thus confirming the cis form with
both substituents being in an equatorial position. On the other
hand, upon excitation of the other H1 resonance at 3.08 ppm
a noticeable NOE develops for the methyl group at position 7
in addition to the two H2a and H6a neighboring protons (Fig-
ure 2C). These NOEs, indicative of short interproton distances,
reveal a trans configuration with the methyl group oriented on
the same side as H1. Based on the unambiguous assignment
To provide access to highly optically pure diastereomers, we
then investigated the one-pot reaction starting with the reduc-
tion of the a/b-unsaturated ketone 1 by using the enoate
reductases from Saccharomyces carlsbergensis (OYE) or from
Pseudomonas putida (XenB). This afforded optically pure (S)-2
with >99% ee[18] (Figure S8). Because OYE showed quantitative
conversion, this enzyme was used for further investigations. To
shift the equilibrium of the amination reaction, the well-estab-
lished lactate dehydrogenase/glucose dehydrogenase (LDH/
GDH) system was used;[16a,19] it also served to recycle the
NADPH required for the ERED. Interestingly, 1 was not convert-
ed by ATAs, so the process could be performed as a cascade
reaction in a simultaneous mode (both enzymes being present
from the beginning). This setup was used in all subsequent ex-
ChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
3
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Communications
of the two stereogenic centers to yield particular diastereo-
mers of the cyclic, alkyl-substituted amine 3. Thus, we consider
this cascade to represent a valuable synthetic strategy for the
production of compounds of this class. Mutations leading to
very small variations in the active site can strongly influence
the diastereoselectivity of the amine transaminase from V. flu-
vialis, and this is further improved by use of co-solvents.
Acknowledgements
We thank the European Union (KBBE-2011-5, grant no. 289350)
and the DFG (grant no. Bo1862/6-1) for financial support. We are
grateful to Prof. Jon D. Stewart (University of Florida, USA) for
the gene encoding OYE1 and Prof. Byung-Gee Kim (Seoul Nation-
al University, Korea) for the gene encoding the ATA from V. fluvia-
lis.
Keywords: amine transaminases
enantiopreference · enzyme catalysis · protein engineering
·
cascade synthesis
·
Figure 2. NMR spectra of the commercial 3-methylcyclohexylamine standard.
A) Normal 1D spectrum. B), C) 1D NOESY spectra of B) the cis, and C) the
trans diastereomer with selective excitation of the corresponding H1 protons
at 2.62 and 3.08 ppm, respectively. The two antiphase multiplets at high
field in (B) are artifacts arising from polarization transfer between H1 and
the two scalar coupled H2b and H6b protons.
[1] a) M. Hçhne, U. T. Bornscheuer, ChemCatChem 2009, 1, 42–51; b) H.
Kohls, F. Steffen-Munsberg, M. Hçhne, Curr. Opin. Chem. Biol. 2014, 19,
180–192; c) J. Rudat, B. Brucher, C. Syldatk, AMB Express 2012, 2, 11;
d) M. Malik, E.-S. Park, J.-S. Shin, Appl. Microbiol. Biotechnol. 2012, 94,
1163–1171; e) P. Tufvesson, J. Lima-Ramos, J. S. Jensen, N. Al-Haque, W.
Neto, J. M. Woodley, Biotechnol. Bioeng. 2011, 108, 1479–1493.
[2] a) 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, Science 2010, 329, 305–309; b) A. A. Desai,
Angew. Chem. Int. Ed. 2011, 50, 1974–1976; Angew. Chem. 2011, 123,
2018–2020.
of the H1 protons of the two diastereomers, a preparative-
scale cascade reaction (77% isolated yield) with the ERED and
the ATA-Leu56Ile was found to result in a preference for the
trans product 3b over the cis product 3a (Figure S9).
The dependence of enantioselectivity on the type of solvent
is well known in traditional chemical synthesis,[21] but has also
been described for enzyme reactions, including for ATAs.[22]
Therefore we sought to identify suitable co-solvents that
might further improve the enantioselectivity of the ATA reac-
tion. Indeed, addition of DMSO or DMF during the transamina-
tion of rac-2 (with ATA-Vfl and its mutants) results in increased
formation of the cis products (Table 1, Figure S10).
[3] a) J. Muschiol, C. Peters, N. Oberleitner, M. Mihovilovic, U. T. Bornscheu-
er, F. Rudroff, Chem. Commun. 2014, DOI: 10.1039/C4CC08752F; b) E.
Ricca, B. Brucher, J. H. Schrittwieser, Adv. Synth. Catal. 2011, 353, 2239–
2262; c) R. C. Simon, N. Richter, E. Busto, W. Kroutil, ACS Catal. 2014, 4,
129–143; d) I. Oroz-Guinea, E. Garcꢁa-Junceda, Curr. Opin. Chem. Biol.
2013, 17, 236–249; e) V. Kçhler, N. J. Turner, Chem. Commun. 2015, 51,
450–464.
[4] T. Sehl, H. C. Hailes, J. M. Ward, R. Wardenga, E. von Lieres, H. Offer-
mann, R. Westphal, M. Pohl, D. Rother, Angew. Chem. Int. Ed. 2013, 52,
6772–6775; Angew. Chem. 2013, 125, 6904–6908.
The cascade reactions (Scheme 1) were then also investigat-
ed with the co-solvents DMSO and DMF. Fortunately, all of the
enzymes (ATA, OYE, LDH, and GDH) are active in the presence
of DMSO. Although OYE tolerated 30% DMSO, as reported
earlier,[23] the enzyme was inactive in 30% DMF. High levels of
DMSO increased the preference for the cis product 3a in all
three cases. We were pleased to find that, by adding 30%
DMSO, the diastereomeric excess of the cis product (1R,3S)-3a
could be substantially improved to 76 and 89% de in the pres-
ence of ATA-Vfl or the mutant Leu56Val, respectively. Because
mutant Leu56Ile forms the trans product 3b, DMSO addition
had no useful effect for preparative purposes, but we observed
a switch in diastereoselectivity towards a slight preference for
the cis product 3a (Table 1, entry 11). As described in other
studies, such as the hydrolysis of a,a-disubstituted malonate
esters,[24] the mechanisms of these solvent effects are currently
not understood.
[5] I. Ardao, A.-P. Zeng, Chem. Eng. Sci. 2013, 87, 183–193.
[6] a) J. Limanto, E. R. Ashley, J. Yin, G. L. Beutner, B. T. Grau, A. M. Kassim,
M. M. Kim, A. Klapars, Z. Liu, H. R. Strotman, M. D. Truppo, Org. Lett.
2014, 16, 2716–2719; b) Z. Peng, J. W. Wong, E. C. Hansen, A. L. A.
Puchlopek-Dermenci, H. J. Clarke, Org. Lett. 2014, 16, 860–863.
[7] a) J.-S. Shin, B.-G. Kim, J. Org. Chem. 2002, 67, 2848–2853; b) F. Steffen-
Munsberg, C. Vickers, A. Thontowi, S. Schꢂtzle, T. Tumlirsch, M. Sveden-
dahl Humble, H. Land, P. Berglund, U. T. Bornscheuer, M. Hçhne, Chem-
CatChem 2013, 5, 150–153.
[8] E. Siirola, F. G. Mutti, B. Grischek, S. F. Hoefler, W. M. F. Fabian, G. Grogan,
W. Kroutil, Adv. Synth. Catal. 2013, 355, 1703–1708.
[9] M. Hçhne, K. Robins, U. T. Bornscheuer, Adv. Synth. Catal. 2008, 350,
807–812.
[10] H. S. Toogood, J. M. Gardiner, N. S. Scrutton, ChemCatChem 2010, 2,
892–914.
[11] D. Koszelewski, I. Lavandera, D. Clay, G. M. Guebitz, D. Rozzell, W. Kroutil,
Angew. Chem. Int. Ed. 2008, 47, 9337–9340; Angew. Chem. 2008, 120,
9477–9480.
[12] a) M. Hçhne, S. Schꢂtzle, H. Jochens, K. Robins, U. T. Bornscheuer, Nat.
Chem. Biol. 2010, 6, 807–813; b) J.-S. Shin, B.-G. Kim, Biotechnol. Bioeng.
1997, 55, 348–358.
Our results demonstrate the advantage of a cascade based
on the use of EREDs and ATAs for the successive introduction
[13] A. Nobili, F. Steffen-Munsberg, H. Kohls, I. Trentin, C. Schulzke, M.
Hçhne, U. T. Bornscheuer, ChemCatChem 2015, 7, 757–760.
&
ChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
4
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communications
[14] R. K. Kuipers, H.-J. Joosten, W. J. H. van Berkel, N. G. H. Leferink, E. Rooij-
en, E. Ittmann, F. van Zimmeren, H. Jochens, U. Bornscheuer, G. Vriend,
V. A. P. Martins dos Santos, P. J. Schaap, Proteins Struct. Funct. Bioinf.
2010, 78, 2101–2113.
Bornscheuer, Biotechnol. Adv. 2015; DOI: 10.1016/j.biotechadv.
2014.12.012; b) F. Steffen-Munsberg, C. Vickers, A. Thontowi, S. Schꢂtzle,
T. Meinhardt, M. Svedendahl Humble, H. Land, P. Berglund, U. T. Born-
scheuer, M. Hçhne, ChemCatChem 2013, 5, 154–157.
[15] R. Kourist, H. Jochens, S. Bartsch, R. Kuipers, S. K. Padhi, M. Gall, D.
Bçttcher, H.-J. Joosten, U. T. Bornscheuer, ChemBioChem 2010, 11,
1635–1643.
[16] a) J.-S. Shin, B.-G. Kim, Biotechnol. Bioeng. 1999, 65, 206–211; b) K. S.
Midelfort, R. Kumar, S. Han, M. J. Karmilowicz, K. McConnell, D. K. Gehl-
haar, A. Mistry, J. S. Chang, M. Anderson, A. Villalobos, J. Minshull, S. Go-
vindarajan, J. W. Wong, Protein Eng. Des. Sel. 2013, 26, 25–33; c) B.-K.
Cho, H.-Y. Park, J.-H. Seo, J. Kim, T.-J. Kang, B.-S. Lee, B.-G. Kim, Biotech-
nol. Bioeng. 2008, 99, 275–284.
[21] a) S. Arseniyadis, A. Valleix, A. Wagner, C. Mioskowski, Angew. Chem. Int.
Ed. 2004, 43, 3314–3317; Angew. Chem. 2004, 116, 3376–3379; b) M.
Node, D. Hashimoto, T. Katoh, S. Ochi, M. Ozeki, T. Watanabe, T. Kaji-
moto, Org. Lett. 2008, 10, 2653–2656.
[22] C. S. Fuchs, M. Hollauf, M. Meissner, R. C. Simon, T. Besset, J. N. H. Reek,
W. Riethorst, F. Zepeck, W. Kroutil, Adv. Synth. Catal. 2014, 356, 2257–
2265.
[23] D. Clay, C. Winkler, G. Tasnꢃdi, K. Faber, Biotechnol. Lett. 2014, 36, 1329–
1333.
[17] E. Krieger, G. Koraimann, G. Vriend, Proteins Struct. Funct. Bioinf. 2002,
47, 393–402.
[24] M. E. Smith, S. Banerjee, Y. Shi, M. Schmidt, U. T. Bornscheuer, D. S. Mas-
terson, ChemCatChem 2012, 4, 472–475.
[18] C. Peters, R. Kçlzsch, M. Kadow, L. Skalden, F. Rudroff, M. D. Mihovilovic,
U. T. Bornscheuer, ChemCatChem 2014, 6, 1021–1027.
[19] N. Nakajima, K. Tanizawa, H. Tanaka, K. Soda, J. Biotechnol. 1988, 8,
243–248.
Received: February 10, 2015
[20] a) F. Steffen-Munsberg, C. Vickers, H. Kohls, H. Land, H. Mallin, A. Nobili,
L. Skalden, T. van den Bergh, H.-J. Joosten, P. Berglund, M. Hçhne, U. T.
Published online on && &&, 0000
ChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
5
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

COMMUNICATIONS
L. Skalden, C. Peters, J. Dickerhoff,
A. Nobili, H.-J. Joosten, K. Weisz,
M. Hçhne, U. T. Bornscheuer*
&& – &&
Two Subtle Amino Acid Changes in
a Transaminase Substantially Enhance
or Invert Enantiopreference in
Cascade Syntheses
A single mutation, a big difference: An
amine transaminase (ATA) and an en-
oate reductase (ERED) combined in a
cascade reaction allowed the produc-
tion of amines containing two chiral
centers. Single-point mutations of the
ATA substantially improved (L56V) or
even inverted (L56I) its selectivity and
gave products with high diastereomeric
excess.
&
ChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
6
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!