Artificial multi-enzyme networks for the asymmetric amination of sec-alcohols

Article abstract of DOI:10.1002/chem.201202666

Various artificial network designs that involve biocatalysts were tested for the asymmetric amination of sec-alcohols to the corresponding α-chiral primary amines. The artificial systems tested involved three to five redox enzymes and were exemplary of a range of different sec-alcohol substrates. Alcohols were oxidised to the corresponding ketone by an alcohol dehydrogenase. The ketones were subsequently aminated by employing a ω-transaminase. Of special interest were redox-neutral designs in which the hydride abstracted in the oxidation step was reused in the amination step of the cascade. Under optimised conditions up to 91 % conversion of an alcohol to the amine was achieved. Trickle-down effect: The asymmetric amination of sec-alcohols to the corresponding α-chiral primary amines was performed with a biocatalytic cascade whereby the various steps were interconnected through the cofactors/cosubstrates. In a redox-neutral cascade and under optimised conditions, up to 91 % conversion of an alcohol to the amine was achieved. Copyright

Full text of DOI:10.1002/chem.201202666

DOI: 10.1002/chem.201202666
Artificial Multi-Enzyme Networks for the Asymmetric Amination of sec-
Alcohols
Katharina Tauber,[a] Michael Fuchs,[a] Johann H. Sattler,[a] Julia Pitzer,[a]
Desiree Pressnitz,[a, b] Dominik Koszelewski,[a, b] Kurt Faber,[a] Jan Pfeffer,[c]
Thomas Haas,[c] and Wolfgang Kroutil*[a]
Abstract: Various artificial network de-
signs that involve biocatalysts were
tested for the asymmetric amination of
sec-alcohols to the corresponding a-
chiral primary amines. The artificial
systems tested involved three to five
redox enzymes and were exemplary of
a range of different sec-alcohol sub-
strates. Alcohols were oxidised to the
corresponding ketone by an alcohol de-
hydrogenase. The ketones were subse-
quently aminated by employing a w-
transaminase. Of special interest were
redox-neutral designs in which the hy-
dride abstracted in the oxidation step
was reused in the amination step of the
cascade. Under optimised conditions
up to 91% conversion of an alcohol to
the amine was achieved.
Keywords: alcohols · amines · dom-
ino reactions · enzymes · oxidation
Introduction
quential steps was chosen as the base for a cascade: 1) in
the first step the alcohol was oxidised by an alcohol dehy-
drogenase[6] (ADH) to the corresponding ketone, which was
Metabolic pathways in microorganisms have successfully
been engineered to enhance the production of natural target
metabolites[1] or to produce novel metabolites by introduc-
ing foreign enzymes into organisms.[2] a-Chiral amines rep-
resent an important functional group in the pharmaceutical
and agrochemical industry.[3] Unfortunately, microorganisms
are not known to transform commonly available sec-alcohol
moieties to the corresponding desired primary amines, as re-
cently described for metal catalysts.[4,5] Consequently, a
novel artificial reaction pathway had to be designed from
scratch and various network possibilities were investigated
in vitro.
then 2) aminated by
a
w-transaminase[7] (w-TA;
Scheme 1).[8,9]
Results and Discussion
To achieve the formal substitution of an alcohol by an
amine moiety a linear pathway that encompassed two se-
Scheme 1. Basic reaction sequence for the amination of sec-alcohols.
[a] K. Tauber, M. Fuchs, J. H. Sattler, J. Pitzer, D. Pressnitz,
Dr. D. Koszelewski, Prof. K. Faber, Prof. W. Kroutil
Department of Chemistry
For both steps additional cofactors/cosubstrates were re-
quired, whereby for the oxidation step redox equivalents
had to be supplied in the form of nicotinamide adenine di-
nucleotide phosphate (NAD(P)+), which had to be recycled;
for the transamination step, l-alanine was chosen as the
amine donor, whereby the formed coproduct pyruvate had
to be removed to shift the equilibrium to the product side.[7c]
Furthermore, the chiral centre of the sec-alcohol had to be
considered, thus the stereopreference and selectivity of the
ADH employed had to match the absolute configuration of
the substrate: for instance, in the case of racemic alcohols,
both enantiomers should be oxidised, which might require
two enantio-complementary ADHs. However, the stereo-
Organic and Bioorganic Chemistry, University of Graz
Heinrichstrasse 28, 8010 Graz (Austria)
Fax : (+43)316-380-9840
E-mail: Wolfgang.Kroutil@uni-graz.at
[b] D. Pressnitz, Dr. D. Koszelewski
ACIB GmbH, c/o Heinrichstrasse 28
8010 Graz (Austria)
[c] J. Pfeffer, T. Haas
Evonik Industries AG
CREAVIS Technologies & Innovation
Paul-Baumann-Strasse 1, 45772 Marl (Germany)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201202666.
4030
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 4030 – 4035

FULL PAPER
chemistry of the product amine was controlled exclusively
by w-transaminase.
As a first approach, an artificial catalyst cascade was de-
signed, whereby the oxidation and the amination steps were
independent with respect to the recycling of the cofactor
and cosubstrate (Scheme 2).
Encouraged by these results, a redox-neutral catalyst net-
work[15–17] that consisted of only three enzymes (instead of
five) and avoided external redox reagents, thereby reducing
the complexity of the system, was envisioned (Scheme 3).
Scheme 3. Redox-neutral catalyst network for amination of sec-alcohols
consuming ammonium as sole reagent.
Scheme 2. Amination of sec-alcohols employing redox-independent oxi-
dation and amination steps.
We call the system a ꢁnetworkꢂ because the individual reac-
tion steps of the cascade are interlinked not only by means
of the transformed substrate, but also through the cofactors
and cosubstrates. The idea was to reuse the hydride abstract-
ed in the oxidation step for the formal reductive amination
step. This was achieved by employing NADH-dependent
ADHs in the oxidation step; the NADH formed during oxi-
dation could serve as reducing agent for the recycling of the
amine-donor alanine by reductive amination of pyruvate,
thereby again providing NAD+ for the subsequent oxidation
cycle. In contrast to the first approach, the oxidation is di-
rectly linked to the reductive amination, thereby giving a
redox-neutral process at the expense of ammonia as the
only reagent (Scheme 2).
A first test that employed crude enzyme preparations led
to promising amine formation (20% of (S)-1a was convert-
ed). During optimisation studies, 5 equiv of l-Ala proved to
be sufficient, as no increase of amine formation was found
at higher concentrations. Additional increases in ammonium
concentration should show a similar beneficial effect on the
equilibrium. Indeed, higher conversions to the amine were
observed by applying 4 equiv of ammonium chloride or am-
monium formate; higher concentrations, however, led to
lower conversions. Very low concentrations of pyridoxal-5’-
phosphate (PLP, 0.3 mm) led to good conversions, whereas
higher amounts (3 mm) lowered the conversion.[18] The same
effect was observed for the cofactor NADH. Whereas low
amounts of NADH (0.1 mm) were beneficial, higher
amounts (2.5 mm) retarded the system remarkably. The pH
optimum was determined to be between 7 and 7.5, which is
the usual optimum of the ADHs employed. To increase sub-
strate solubility, different organic solvents (e.g., 1,2-dime-
thoxyethane, or DMSO) were tested, but no major positive
effect was observed, whereas better mixing (450 rpm instead
of 120 rpm, 1.5 mm radius) improved the results.
Pyruvate was recycled to alanine by an NADH-dependent
alanine dehydrogenase from Bacillus subtilis[10] (AlaDH);
consequently, the ADH-catalysed oxidation step had to be
NADP+-dependent to avoid short-circuits between the two
cofactor recycling systems that run simultaneously beside
each other (Scheme 2). For the recycling of NADP+, an
NADPH oxidase from Bacillus subtilis[11] (YcnD) was em-
ployed. For the recycling of NADH, a commercial formate
dehydrogenase (FDH) was used. Testing this cascade con-
cept with various substrates led indeed to amine formation
by employing the crude enzyme preparations (Table 1).
Since racemic alcohols were employed in combination with
one enantioselective ADH, the amine formation was limited
theoretically by 50%, since a perfect enantioselective ADH
could oxidise only 50% of the substrate.
Table 1. Amination of sec-alcohols by means of the biocatalyst network
of Scheme 2.[a]
Substrate
ADH
wTA[b]
Ketone b[c] [%]
(S)-Amine c[d] [%]
rac-1a[d]
rac-2a
rac-3a
rac-4a
LB-ADH
ADH-005
LB-ADH
LB-ADH
BM
BM
ArS
BM
5
5
1
2
32
23[e]
10[f]
24
[a] Reaction conditions: phosphate buffer (50 mm, pH 7.5), NADH
(0.5 mm), NADPH (0.5 mm), l-alanine (5 equiv), ammonium formate
(4 equiv), pyridoxal-5’-phosphate (PLP, 1 mm), transaminase (in lyophi-
lised E. coli cells, 20 mg), AlaDH (crude, 0.63 U, 50 mL), ADH [ADH-
005: 77.5 U, 50 mL, Codexis; (R)-selective LB-ADH from Lactobacillus
brevis, 0.13 U, 50 mL, Codexis], NADPH-oxidase YcnD (1.67 U, 10 mL),
FDH (Codexis, 10 U), substrate (50 mm); 1 mL total volume, 308C and
450 rpm for 24 h using an Eppendorf Thermomixer Comfort. [b] BM=
transaminase from Bacillus megaterium,[12,13] ArS=Arthrobacter sp.[12,14]
[c] Composition measured by GC-FID analysis, peak areas. [d] NADH
(1 mm), NADPH (1 mm), purified AlaDH (0.16 U). [e] 94% ee.
[f]>99% ee.
Under optimised conditions, amine formation was further
increased to 64 and 55% for the nonchiral substrates 8a and
9a (Table 2).
Chem. Eur. J. 2013, 19, 4030 – 4035
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4031

W. Kroutil et al.
Table 2. Redox-neutral amination of sec-alcohols employing the concept of Sche-
BM-wTA. In this setup, 43% (S)-amine was ob-
tained (19% ketone) as determined by GC. After
extraction and crystallisation as the corresponding
HCl salt, 35% (S)-2c was obtained (30 mg,
94% ee).
me 3.[a]
Substrate
ADH[b]
wTA[c]
Ketone b[d] [%]
Amine c[d] [%]
(S)-1a
rac-1a[e,l]
(S)-2a
rac-3a
rac-4a
(S)-5a
rac-6a
rac-7a
8a[g,h]
ADH-A
Vf
25
35
32
5
10
47
13
<1
–
47[b]
50
ADH-A[f] and ADH-007
ADH-A[f]
BM
CV
ArS
BM
Vf
BM
ArS
BM
BM
54[j]
9[k]
31
To gain a better understanding of the network
system of Scheme 3, the experiment was repeated
for substrate (S)-1a by employing purified enzymes.
In this case, only 11% amine formation (Table 3,
entry 1) was found, whereas the experiment with
crude enzyme preparations led to 47% amine for-
mation by using comparable enzyme activities
(Table 2). To exclude the possibility that the type of
buffer had any negative influence,[22] the experiment
was repeated in piperazine-N,N’-bis(2-ethanesulfon-
ic acid) (PIPES) buffer, thus leading to comparable
results. In the previous experiments that employed
crude enzyme preparations, a significant amount of
ketone was formed. However, according to the
amount of NAD+ added, a maximum of 2–4% of
ketone could be formed. Therefore the additional
ADH-A
ADH-A
ADH-A[f]
25[i]
28
10
55
64
ADH-A[f]
ADH-A
ADH-A
ADH-A and ADH-007
9a[g,h]
14
[a] Reaction conditions: phosphate buffer (50 mm, pH 7.5), NADH (1 mm), l-alanine
(5 equiv), ammonium chloride (4 equiv), PLP (1 mm), transaminase (lyophilised cells
containing overexpressed enzyme, 20 mg), AlaDH (0.62 U, crude, 50 mL), ADH
(ADH-A thermally treated, 2.11 U, 50 mL; ADH-007, 1.2 U, 10 mg, Codexis), sub-
strate (50 mm); 1 mL total volume, 308C and 450 rpm for 24 h using an Eppendorf
Thermomixer Comfort. [b] ADH-A=(S)-selective NADH-ADH from Rhodococcus
ruber DSM 44541;[19] ADH 007=(R)-selective NADH-ADH, Codexis. [c] BM=trans-
aminase from Bacillus megaterium, ArS=Arthrobacter sp., Vf=Vibrio fluvialis,[12,20]
CV=Chromobacterium violaceum.[12,21] [d] Composition measured using GC-FID,
peak areas. [e] 1 mm NAD+. [f] ADH-A thermally treated (4.23 U, 100 mL). [g] puri-
fied AlaDH (0.16 U, 10 mL). [h] l-Alanine (10 equiv), ammonium chloride (2 equiv).
[i] 98% ee (S). [j] 78% ee (S). [k]>99% ee (S). [l] Reaction performed at 208C.
Table 3. Variation of the catalyst network in Scheme 4 for amination of
(S)-1a by employing purified enzymes (ADH-A, AlaDH, wTA-Vf).[a]
Starting from optically pure (S)-1a with ADH-A gave
47% of amine 1c. A similar conversion (50%) was obtained
from rac-1a by using a combination of (R)- and (S)-selective
ADHs. In general, it was observed that amine formation
went hand-in-hand with high ketone formation, thus indicat-
ing that unwanted side reactions took place (not depicted in
Scheme 2; see below). Interestingly, a decrease in ee of the
amine over time was also observed, which could most likely
be attributed to constantly running back-and-forth reactions
as in an equilibrating system.[15e]
Following the reaction over time for (S)-2a revealed that
after six hours the reaction slowed down, whereby 10% of
the alcohol still remained, which did not further decrease
(Figure 1). Nevertheless, the ketone was still slowly trans-
formed to the amine. Amine conversion stopped or reached
equilibrium at 60% amine formation.
Entry
Additional enzyme
Ketone 1b [%][b]
Amine 1c [%][b]
1
2
3
4
5
–
5
12
48
10
14
11
34
45
58
71
NOX-2[c]
NOX-2 5ꢃ[c]
LDH[d]
LDH 2ꢃ[d]
[a] General reaction conditions: wTA from Vibrio fluvialis (purified,
lyophilised powder, 20 mg), AlaDH from Bacillus subtilis (purified,
0.16 U, 10 mL), ADH-A from Rhodococcus ruber (purified, 0.72 U,
50 mL), 1a (50 mm); phosphate buffer (50 mm, pH 7.5), NADH (0.5 mm),
NAD+ (0.5 mm), l-alanine (5 equiv), ammonium chloride (4 equiv), PLP
(1 mm), 1 mL total volume, 308C and 450 rpm for 24 h using an Eppen-
dorf Thermomixer Comfort. [b] Determined by GC-FID, peak areas.
[c] NADH oxidase from Streptococcus mutans, 0.48 and 2.42 U, respec-
tively (10 mL, 50 mL). [d] LDH from rabbit muscle, 1655 and 3310 U (5
and 10 mg, respectively, Jꢄlich). In the case of LDH, no ammonium
chloride or AlaDH were added.
The scalability of the reaction was shown for (S)-2a
(65 mg) at 10 mL total volume by employing ADH-A and
redox equivalents were suspected to be provided by
NADH-oxidases in the crude enzyme preparations derived
from the E. coli host cells, which oxidised NADH to NAD+
before its hydride was consumed in the reductive amination
(Scheme 4a).
To verify this hypothesis, the NADH-oxidase NOX-2
from Streptococcus mutans[23] was tested in the reaction cas-
cade with purified enzymes, which led indeed to improved
amine formation of up to 45% but also, as expected, to in-
creased ketone formation (12–48%, Table 3, entries 2 and
3).
In another attempt, the network was redesigned in such a
way that pyruvate was not recycled to alanine but reduced
to lactate by a lactate dehydrogenase (LDH) at the expense
of the hydride (NADH) gained in the oxidation step (Sche-
me 4b). This network was still redox-neutral; however, the
Figure 1. Time course of the redox-neutral catalyst network employing
(S)-2a as substrate.
4032
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 4030 – 4035

Artificial Multi-Enzyme Networks
FULL PAPER
Conclusion
A redox-neutral catalyst network for the transformation of
sec-alcohols to the corresponding primary amines was suc-
cessfully accomplished. In an ideal system, any NADH-oxi-
dase activity has to be minimised, which makes the applica-
tion of E. coli as host difficult due to the NADH oxidases
present.[24] The best results were obtained using a lactate de-
hydrogenase to remove pyruvate, since the highest amounts
of amine were observed with this system. Employing an
amino acid dehydrogenase to recycle pyruvate back to ala-
nine showed certain advantages, since it requires only am-
monia and water is the sole byproduct. This process repre-
sents a cost-efficient and redox-neutral, benign route to
highly valuable a-chiral amines. Additionally, the transfor-
mation of an alcohol to an amine with inversion of configu-
ration by the appropriate choice of stereo-complementary
enzymes can be seen as a reagent-saving alternative to Mit-
sunobu protocols, thus avoiding triphenylphosphane and the
associated purification issues.[25]
Scheme 4. Variations of the catalyst network for amination of sec-alco-
hols: a) NADH-oxidase (NOX-2) leads to internal recycling of NAD+;
b) lactate dehydrogenase (LDH) for removal of pyruvate.
Experimental Section
overall reaction now resembled the transformation of alco-
hol plus alanine to amine plus lactate. By contrast, in the
original version (Scheme 3), alcohol plus ammonium were
transformed to amine plus water.
Materials and methods: Substrates and reference materials were pur-
chased from commercial sources: 1a, (S)-1a, 1b, 1c, 2a, 2b, 2c, (R)-2c,
3b, 4b, 5b, 5c, (S)-5c, 6b, 6c, 9a, 9b and 9c from Sigma Aldrich; (S)-2c,
(S)-5a, 6a, 8a, 8b and 7c from Fluka; (S)-1c, (R)-1c, 3a, 7b and (S)-2a
from Lancaster; and 3c from BASF. Compounds 4a and 7a were syn-
thesised starting from the ketone.[26] All other chemicals were purchased
from Sigma Aldrich or Fluka and used as received. Solvents were ob-
tained from Roth.
This variation of the redox-neutral catalyst network that
involved LDH led to high amine formation (up to 71%)
with purified enzymes (Table 3, entries 4 and 5). Conse-
quently, the network that involved LDH was tested for the
crude enzyme preparations for selected substrates (Table 4).
All substrates tested were transformed with significant im-
proved conversions and reached up to 91% conversion for
substrate 8a (Table 4, entry 5). Ketone formation was ele-
vated again due to the NADH oxidases present in E. coli
host cells.
Enzymes were purchased from Codexis (ADH 005, ADH 007, FDH) and
previously Jꢄlich Fine Chemicals (LDH, l-lactate dehydrogenase from
rabbit muscle, now Codexis, #20.10, 331 UmgÀ1) or overexpressed in E.
coli BL 21 (DE3) cells and used as lyophilised cells (wTA-BM, wTA-Vf,
wTA-AT-S, wTA-CV),[12] crude extract (AlaDH) or purified as described
(ADH-A,[27] YcnD[11]) or by using His-tag (NOX, AlaDH). Alcohol de-
hydrogenase 007: 0.12 UmgÀ1, #42.10. In the case of ADH-LB,[28] a crude
extract of the wild-type enzyme from Lactobacillus brevis was employed.
l-Lactate dehydrogenase from rabbit muscle was commercially available
from Jꢄlich Fine Chemicals, now Codexis, #20.10, 331 UmgÀ1
.
General procedure for the cascade with independent cofactor
regeneration: First a stock solution that contained phosphate
buffer (50 mm, pH 7.5), NADH (0.5 mm), NADPH (0.5 mm), l-
alanine (5 equiv), ammonium formate (4 equiv), PLP (1 mm),
alanine dehydrogenase (cell-free crude extract, 50 mL per
sample, 0.63 U), FDH (10 UmLÀ1), YcnD (purified 10 mL per
sample, 1.67 U) and transaminase (20 mg cells per sample) was
prepared. Stock solutions (950 mL) were distributed onto Ep-
pendorf vials (2 mL) and ADH (ADH-005, 77.5 U, 50 mL,
Jꢄlich Fine Chemicals; or LB-ADH from Lactobacillus brevis,
crude, 0.13 U, 50 mL) and the substrate (50 mm) were added.
All samples were incubated for 24 h at 308C and 450 rpm
using an Eppendorf Thermomixer Comfort instrument.
Table 4. Redox-neutral amination of sec-alcohols employing lactate dehydrogenase,
crude ADH-A and crude w-TAs (Scheme 4b).[a]
Entry Substrate wTA Ketone b [%][b] Amine c [%][b] ee of (S)-amine c [%]
1
2
3
4
5
6
7
8
(S)-1a
(S)-1a
(S)-2a
(S)-2a
8a
BM
Vf
BM
Vf
BM
Vf
Vf
26
27
27
31
<1
2
63
64
48
64
91
54
30
85
20
88
96
96
n.a.[c]
n.a.[c]
n.a.[c]
n.a.[c]
8a
9a
9a
44
10
BM
[a] General reaction conditions: wTA (lyophilised whole cells, 20 mg), ADH-A from
Rhodococcus ruber (ADH-A thermally treated, 4.22 U, 100 mL); LDH from rabbit
muscle (3310 U, 10 mg), substrate (50 mm); phosphate buffer (50 mm, pH 7.5), NADH
(0.5 mm), NAD+ (0.5 mm), l-alanine (5 equiv), PLP (1 mm), 1 mL total volume, 308C
and 450 rpm for 24 h using an Eppendorf Thermomixer Comfort in 1.5 mL reaction
tubes in the horizontal position. [b] Determined by GC-FID analysis, peak areas.
[c] Not applicable.
Workup: The reaction was stopped by the addition of NaOH
(100 mL, 10m) and ethyl acetate (0.9 mL). The solution was
mixed thoroughly and centrifuged for 5 min at 13000 rpm.
Then the organic layer was separated and dried over sodium
sulfate, which was then removed by centrifugation (3 min,
13000 rpm; Heraeus Biofuge pico). The supernatant was sub-
jected to gas chromatography/flame ionization detector (GC-
Chem. Eur. J. 2013, 19, 4030 – 4035
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4033

W. Kroutil et al.
FID) analysis and the products were identified by comparison to authen-
tic reference material.
search (BMBF: project BioxAmine). J.H.S. received funding from the
European Union Seventh Framework Programme FP7/2007-2013 under
grant agreement number 245144 (AmBioCas). D.P. acknowledges the
Federal Ministry of Economy, Family and Youth (BMWFJ), the Federal
Ministry of Traffic, Innovation and Technology (bmvit), the Styrian Busi-
ness Promotion Agency (SFG), the Standortagentur Tirol and ZIT, the
Technology Agency of the City of Vienna, through the COMET-Funding
Program managed by the Austrian Research Promotion Agency FFG.
Derivatisation for determination of enantiomeric excess: A solution of
DMAP (5–10 mg) in acetic anhydride (100 mL) was added to the samples
in AcOEt. Then they were shaken at 218C and 150 rpm for 3 h. Water
was added (500 mL, to destroy excess amounts of acetic anhydride) and
the samples were shaken for 3 h at 218C. For workup the samples were
centrifuged (3 min, 13000 rpm; Heraeus Biofuge pico) and the organic
phase was dried over sodium sulfate.
General procedure for cascade with internal cofactor regeneration: A
stock solution that contained phosphate buffer (50 mm, pH 7.5), NADH
(1 mm), l-alanine (5 or 10 equiv), ammonium chloride (4 equiv), pyridox-
al phosphate (1 mm), alanine dehydrogenase (0.62 U, crude cell-free ex-
tract, 50 mL per sample) and transaminase (lyophilised cells 20 mg per
sample) was prepared. Stock solutions (950–900 mL) were distributed
onto Eppendorf vials (2 mL) and ADH (ADH-A thermally treated,
2.11 U, 50 mL; ADH-007, 1.2 U, 10 mg, Codexis) and substrate (50 mm)
were added. All samples were incubated for 24 h at 308C (for samples
that contained wTA-CV: 208C) and 450 rpm using an Eppendorf Ther-
momixer Comfort. Workup was carried out as described above.
[1] a) J. W. Lee, D. Na, J. M. Park, J. Lee, S. Choi, S. Y. Lee, Nat. Chem.
Biol. 2012, 8, 536–546; b) J. D. Keasling, Science 2010, 330, 1355–
1358; c) G. Stephanopoulos, Science 2007, 315, 801–804; d) V. F.
Wendisch, M. Bott, B. J. Eikmanns, Curr. Opin. Microbiol. 2006, 9,
268–274.
[2] a) W. S. Glenn, E. Nims, S. E. OꢂConnor, J. Am. Chem. Soc. 2011,
133, 19346–19349; b) J. W. Lee, H. U. Kim, S. Choi, J. Yi, Curr.
Opin. Biotechnol. 2011, 22, 758–767; c) Sang Yup Lee K. L. J. Prath-
er, C. H. Martin, Curr. Opin. Biotechnol. 2008, 19, 468–474.
[3] a) T. B. Nguyen, H. Bousserouel, Q. Wang, F. Guꢅritte, Adv. Synth.
Catal. 2011, 353, 257–262; b) A. Henseler, M. Kato, K. Mori, T.
Akiyama, Angew. Chem. 2011, 123, 8330–8333; Angew. Chem. Int.
Ed. 2011, 50, 8180–8183; c) V. N. Wakchaure, J. Zhou, S. Hoffmann,
B. List, Angew. Chem. 2010, 122, 4716–4718; Angew. Chem. Int. Ed.
2010, 49, 4612–4614; d) R. I. Storer, D. E. Carrera, Y. Ni, D. W. C.
MacMillan, J. Am. Chem. Soc. 2006, 128, 84–86; e) T. C. Nugent, M.
El-Shazly, Adv. Synth. Catal. 2010, 352, 753–819; f) T. C. Nugent in
Chiral Amine Synthesis: Methods, Developments and Applications
(Ed.: T. C. Nugent), Wiley-VCH, Weinheim, 2010, pp. 225–245;
g) T. Sun, G. Hou, M. Ma, X. Zhang, Adv. Synth. Catal. 2011, 353,
253–256; h) X.-Y. Zhou, M. Bao, Y.-G. Zhou, Adv. Synth. Catal.
2011, 353, 84–88; i) H. U. Blaser, F. Spindler, M. Studer, Appl.
Catal. A 2001, 221, 119–143.
[4] Selected recent examples: a) M. Lafrance, M. Roggen, E. M. Car-
reira, Angew. Chem. 2012, 124, 3527–3530; Angew. Chem. Int. Ed.
2012, 51, 3470–3473; b) S. Imm, S. Bꢆhn, M. Zhang, L. Neubert, H.
Neumann, F. Klasovsky, J. Pfeffer, T. Haas, M. Beller, Angew.
Chem. 2011, 123, 7741–7745; Angew. Chem. Int. Ed. 2011, 50, 7599–
7603; c) D. Pingen, C. Mꢄller, D. Vogt, Angew. Chem. 2010, 122,
8307–8310; Angew. Chem. Int. Ed. 2010, 49, 8130–8133; d) R. Ka-
wahara, K.-i. Fujita, R. Yamaguchi, J. Am. Chem. Soc. 2010, 132,
15108–15111; e) O. Saidi, A. J. Blacker, M. M. Farah, S. P. Marsden,
J. M. J. Williams, Angew. Chem. 2009, 121, 7511–7514; Angew.
Chem. Int. Ed. 2009, 48, 7375–7378.
General procedure for the cascade that employed LDH recycling: A
stock solution that contained phosphate buffer (50 mm, pH 7.5), NADH
(0.5 mm), NAD+ (0.5 mm), l-alanine (5 equiv), LDH (5–10 mg), transa-
minase (20 mg purified lyophilised enzyme) and ADH-A (100 mL) was
charged with the substrate (50 mm). All samples were incubated for 24 h
at 308C and 450 rpm using an Eppendorf Thermomixer Comfort.
Workup was carried out as described above.
Determination of conversion: The determination of the conversion was
carried out by GC (Varian 3800 using hydrogen as carrier gas) analysis
on an achiral stationary phase using a 14% cyanopropylphenyl-phase ca-
pillary column (J&W Scientific DB-1701; 30 mꢃ0.25 mm, 0.25 mm film),
with an injection and detection temperature of 220 and 2508C, respec-
tively, and a 20:1 split ratio.
Temperature program for 1, 2, 5 and 6: 708C, hold 1 min, 108CminÀ1 to
908C, hold 2 min and 208CminÀ1 to 1608C, hold 4 min. Retention times
[min]: 1a 10.1, 1b 9.8, 1c 9.2; 2a 6.0, 2b 5.8, 2c 4.8; 5a 7.7, 5b 7.5, 5c
6.8; 6a 8.7, 6b 8.6, 6c 7.9.
Temperature program for 3 and 4: 908C, hold for 2 min, 108CminÀ1, to
1808C, hold for 2 min. Retention times [min]: 3a 11.2, 3b 10.7, 3c 10.2;
4a 11.0, 4b 11.2, 4c 10.4.
Temperature program for 7: 1108C, hold 0 min, 108CminÀ1 to 2008C,
hold 8 min. Retention times [min]: 7a 11.1, 7b 10.9, 7c 10.1.
Temperature program for 8 and 9: 508C, hold 2 min, 108CminÀ1 to 908C,
hold 2 min and 208CminÀ1 to 1608C, hold 0 min. Retention times [min]:
8a 6.3, 8b 6.6, 8c 5.0; 9a 4.6, 9b 4.7, 9c 3.4. Amines and ketones were
identified by co-injection with commercially available reference material.
[5] First reports on metal-catalysed amination of alcohols: a) R. Grigg,
T. R. B. Mitchell, S. Sutthivaiyakit, N. Tongpenyai, J. Chem. Soc.
Chem. Commun. 1981, 611–612; b) Y. Watanabe, Y. Tsuji, Y.
Ohsugi, Tetrahedron Lett. 1981, 22, 2667–2670.
Determination of enantiomeric excess: Values of ee were measured by
GC analysis on a chiral stationary phase using a b-cyclodextrin capillary
column (CP-Chirasil-DEX CB, 25 mꢃ0.32 mm, 0.25 mm film), split ratio
10:1. The temperature of the injector and detector for chiral measure-
ments were 180 and 2508C, respectively.
Temperature program for 1, 2 and 5: 708C, hold 2 min, 158CminÀ1 to
1308C, hold 2 min and 108Cmin to 1608C, hold 5 min. Retention times
[min]: (S)-1c 21.3, (R)-1c 21.5; (S)-2c 14.0, (R)-2c 14.5; (S)-1c 21.3, (R)-
1c 21.5; (S)-5c 17.3, (R)-5c 17.5.
Temperature program for 3: 1008C, hold 2 min, 58CminÀ1 to 1308C, hold
2 min and 108CminÀ1 to 1808C, hold 10 min. Retention times [min]: (S)-
3c 17.1, (R)-3c 17.3. Amines were identified by co-injection with deriva-
tised commercially available reference material.
[6] a) F. Hollmann, I. W. C. E. Arends, K. Buehler, A. Schallmey, B.
Bꢄhler, Green Chem. 2011, 13, 226–265; b) T. Matsuda, R. Yamana-
ka, K. Nakamura, Tetrahedron: Asymmetry 2009, 20, 513–557; c) K.
Goldberg, K. Schroer, S. Lꢄtz, A. Liese, Appl. Microbiol. Biotech-
nol. 2007, 76, 237–248; d) K. Edegger, H. Mang, K. Faber, J. Gross,
W. Kroutil, J. Mol. Catal. A 2006, 251, 66–70; e) W. Kroutil, H.
Mang, K. Edegger, K. Faber, Adv. Synth. Catal. 2004, 346, 125–142.
[7] For recent reviews on w-transaminases, see: a) M. S. Malik, E.-S.
Park, J.-S. Shin, Appl. Microbiol. Biotechnol. 2012, 94, 1163–1171;
b) S. Mathew, H. Yun, ACS Catal. 2012, 2, 993–1001; c) P. Tufves-
son, J. Lima-Ramos, J. S. Jensen, N. Al-Haque, W. Neto, J. M. Wood-
ley, Biotechnol. Bioeng. 2011, 108, 1479–1493; d) D. Koszelewski, K.
Tauber, K. Faber, W. Kroutil, Trends Biotechnol. 2010, 28, 324–332;
e) N. J. Turner, M. Truppo in Chiral Amine Synthesis: Methods, De-
velopments and Applications (Ed.: T. C. Nugent), Wiley-VCH,
Weinheim, 2010, pp. 431–459; f) H. C. Hailes, P. A. Dalby, G. J. Lye,
F. Baganz, M. Micheletti, N. Szita, J. M. Ward, Curr. Org. Chem.
2010, 14, 1883–1893; g) J. Ward, R. Wohlgemuth, Curr. Org. Chem.
2010, 14, 1914–1927; h) M. Hçhne, U. T. Bornscheuer, ChemCat
Chem 2009, 1, 42–51.
Acknowledgements
The research that led to these results received funding from Evonik In-
dustries AG and the German Federal Ministry of Education and Re-
4034
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 4030 – 4035

Artificial Multi-Enzyme Networks
FULL PAPER
[8] Redox-neutral biocatalytic transformations of alcohols to amines
have only been reported for a-hydroxy acids to the corresponding
a-amino acids that involve amino acid dehydrogenases and alcohol
dehydrogenases: a) transformation of mandelic acid to l-phenylgly-
cine: V. Resch, W. M. F. Fabian, W. Kroutil, Adv. Synth. Catal. 2010,
352, 993–997; b) transformation of lactate to alanine: C. Wandrey,
E. Fiolitakis, U. Wichmann, Ann. N. Y. Acad. Sci. 1984, 434, 091–
094; as well as for prim-alcohols: c) J. H. Sattler, M. Fuchs, K.
Tauber, F. G. Mutti, K. Faber, J. Pfeffer, T. Haas, W. Kroutil, Angew.
Chem. Int. Ed. 2012, 51, 9156–9159.
[9] prim-Alcohols to amines were transformed in a cascade that consist-
ed of an alcohol oxidase and a w-transaminase: M. Fuchs, K.
Tauber, J. Sattler, H. Lechner, J. Pfeffer, W. Kroutil, K. Faber, RSC
Adv. 2012, 2, 6262–6265.
[10] F. G. Mutti, C. S. Fuchs, D. Pressnitz, N. G. Turrini, J. H. Sattler, A.
Lerchner, A. Skerra, W. Kroutil, Eur. J. Org. Chem. 2012, 1003–
1007.
[11] a) C. V. Voss, C. C. Gruber, K. Faber, T. Knaus, P. Macheroux, W.
Kroutil, J. Am. Chem. Soc. 2008, 130, 13969–13972; b) A. Morokut-
ti, A. Lyskowski, S. Sollner, E. Pointner, T. B. Fitzpatrick, C. Kratky,
K. Gruber, P. Macheroux, Biochemistry 2005, 44, 13724–13733.
[12] D. Koszelewski, M. Gçritzer, D. Clay, B. Seisser, W. Kroutil, Chem-
CatChem 2009, 2, 73–77.
[13] R. L. Hanson, B. L. Davis, Y. Chen, S. L. Goldberg, W. L. Parker,
T. P. Tully, M. A. Montana, R. N. Patel, Adv. Synth. Catal. 2008, 350,
1367–1375.
[16] For a recent review on cascades with redox steps, see: J. H. Schritt-
wieser, J. Sattler, V. Resch, F. G. Mutti, W. Kroutil, Curr. Opin.
Chem. Biol. 2011, 15, 249–256.
[17] Selected recent reviews on biocatalytic cascades: a) E. Ricca, B.
Brucher, J. H. Schrittwieser, Adv. Synth. Catal. 2011, 353, 2239–
2262; b) N. J. Turner, Curr. Opin. Chem. Biol. 2010, 14, 115–121;
c) Multi-Step Enzyme Catalysis (Ed.: E. Garcia-Junceda), 1st ed.,
Wiley-VCH, Weinheim, 2008; d) F. Hollmann, K. Hofstetter, A.
Schmid, Trends Biotechnol. 2006, 24, 163–171.
[18] K. E. Cassimjee, C. Branneby, V. Abedi, A. Wells, P. Berglund,
Chem. Commun. 2010, 46, 5569–5571.
[19] a) K. Edegger, C. C. Gruber, T. M. Poessl, S. R. Wallner, I. Lavan-
dera, K. Faber, F. Niehaus, J. Eck, R. Oehrlein, A. Hafner, W. Krou-
til, Chem. Commun. 2006, 2402–2404; b) B. Kosjek, W. Stampfer,
M. Pogorevc, W. Goessler, K. Faber, W. Kroutil, Biotechnol. Bioeng.
2004, 86, 55–62.
[20] J.-H. Seo, D. Kyung, K. Joo, J. Lee, B.-G. Kim, Biotechnol. Bioeng.
2011, 108, 253–263.
[21] U. Kaulmanna, K. Smithies, M. E. B. Smith, H. C. Hailes, J. M.
Ward, Enzyme Microb. Technol. 2007, 41, 628–637.
[22] a) S. Schꢆtzle, F. Steffen-Munsberg, A. Thontowi, M. Hçhne, K.
Robins, U. T. Bornscheuer, Adv.Synth. Catal. 2011, 353, 2439–2445;
b) K. E. Cassimjee, M. S. Humble, V. Miceli, C. G. Colomina, P. Ber-
glund, ACS Catal. 2011, 1, 1051–1055.
[23] M. Higuchi, M. Shimada, J. Matsumoto, Y. Yamamoto, A. Rhaman,
Y. Kamio, Biosci. Biotechnol. Biochem. 1994, 58, 1603–1607.
[24] a) S. Kerscher, S. Drçse, V. Zickermann, U. Brandt, Results Probl.
Cell Differ. 2008, 45, 185–222; b) G. Unden, J. Bongaerts, Biochim.
Biophys. Acta Bioenerg. 1997, 1320, 217–234.
[14] S. Pannuri, S. V. Kamat, A. R. M. Garcia, (Cambrex North Bruns-
wick Inc.) WO 2006/063336 A2, 2006.
[15] Redox-neutral sequential cascades that exclusively involve enzymes
have been reported in Ref. [8] and for the combination of an alcohol
dehydrogenase and an enoate reductase: a) S. Gargiulo, D. J. Opper-
man, U. Hanefeld, I. W. C. E. Arends, F. Hollmann, Chem.
Commun. 2012, 48, 6630–6632; b) B. Boonstra, D. A. Rathbone,
C. E. French, E. H. Walker, N. C. Bruce, Appl. Environ. Microbiol.
2000, 66, 5161–5166; the combination of an alcohol dehydrogenase
and a Baeyer–Villiger mono-oxygenase: c) R. Gagnon, G. Grogan,
S. M. Roberts, R. Villa, A. J. Willetts, J. Chem. Soc. Perkin Trans. 1
1995, 1505–1511; d) A. J. Willetts, C. J. Knowles, M. S. Levitt, S. M.
Roberts, H. Sandey, N. F. Shipston, J. Chem. Soc. Perkin Trans. 1
1991, 1608–1610; the combination of two alcohol dehydrogenases to
achieve racemisation: e) C. C. Gruber, B. M. Nestl, J. Gross, P. Hil-
debrandt, U. T. Bornscheuer, K. Faber, W. Kroutil, Chem. Eur. J.
2007, 13, 8271–8276.
[25] a) K. C. K. Swamy, N. N. B. Kumar, E. Balaraman, K. V. P. P. Kumar,
Chem. Rev. 2009, 109, 2551–2651; b) D. L. Hughes, R. A. Reamer,
J. J. Bergan, E. J. J. Grabowski, J. Am. Chem. Soc. 1988, 110, 6487–
´
6491; c) O. Mitsunobu, Synthesis 1981, 1, 1–28; d) E. Slusarska, A.
Zwierzak, Liebigs Ann. Chem. 1986, 402–405; e) O. Mitsunobu, M.
Yamada, Bull. Chem. Soc. Jpn. 1967, 40, 2380–2382.
[26] N. Becker, E. M. Carreira, Org. Lett. 2007, 9, 3857–3858.
[27] K. Tauber, M. Hall, W. Kroutil, W. M. F. Fabian, K. Faber, S. M.
Glueck, Biotechnol. Bioeng. 2011, 108, 1462–1467.
[28] W. Hummel, Adv. Biochem. Eng./Biotechnol. 1997, 58, 145–184.
Received: July 26, 2012
Revised: December 19, 2012
Published online: January 22, 2013
Chem. Eur. J. 2013, 19, 4030 – 4035
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4035