One-pot Synthesis of 4-Aminocyclohexanol Isomers by Combining a Keto Reductase and an Amine Transaminase

Article abstract of DOI:10.1002/cctc.201900733

The efficient multifunctionalization by one-pot or cascade catalytic systems has developed as an important research field, but is often challenging due to incompatibilities or cross-reactivities of the catalysts leading to side product formation. Herein we report the stereoselective preparation of cis- and trans-4-aminocyclohexanol from the potentially bio-based precursor 1,4-cyclohexanedione. We identified regio- and stereoselective enzymes catalyzing reduction and transamination of the diketone, which can be performed in a one-pot sequential or cascade mode. For this, we identified regioselective keto reductases for the selective mono reduction of the diketone to give 4-hydroxycyclohexanone. The system is modular and by choosing stereocomplementary amine transaminases, both cis- and trans-4-aminocyclohexanol were synthesized with good to excellent diastereomeric ratios. Furthermore, we identified an amine transaminase that produces cis-1,4-cyclohexanediamine with diastereomeric ratios >98 : 2. These examples highlight that the high selectivity of enzymes enable short and stereoselective cascade multifunctionalizations to generate high-value building blocks from renewable starting materials. Introduction.

Full text of DOI:10.1002/cctc.201900733

Accepted Article
Title: One-pot Synthesis of 4-Aminocyclohexanol Isomers by
Combining a Keto Reductase and an Amine Transaminase
Authors: Olha Sviatenko, Nicolás Ríos-Lombardía, Francisco Morís,
Javier González-Sabín, Kollipara Venkata Manideep, Simon
Merdivan, Sebastian Günther, Philipp Süss, and Matthias
Höhne
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201900733
Link to VoR: http://dx.doi.org/10.1002/cctc.201900733
A Journal of
www.chemcatchem.org

10.1002/cctc.201900733
ChemCatChem
FULL PAPER
One-pot Synthesis of 4-Aminocyclohexanol Isomers by
Combining a Keto Reductase and an Amine Transaminase
Olha Sviatenko,[a] Nicolás Ríos-Lombardía,[b] Francisco Morís,[b] Javier González-Sabín,*[b] Kollipara
Venkata Manideep,[a] Simon Merdivan,[c] Sebastian Günther,[c] Philipp Süss,[d] and Matthias Höhne*[a]
Abstract: The efficient multifunctionalization by one-pot or cascade
catalytic systems has developed as an important research field, but
is often challenging due to incompatibilities or cross-reactivities of
the catalysts leading to side product formation. Herein we report the
stereoselective preparation of cis- and trans-4-aminocyclohexanol
from the potentially bio-based precursor 1,4-cyclohexanodione . We
identified regio- and stereoselective enzymes catalyzing reduction
and transamination of the diketone, which can be performed in a
one-pot sequential or cascade mode. For this, we identified
regioselective keto reductases for the selective mono reduction of
the diketone to give 4-hydroxycyclohexanone. The system is
field.[5]
In this study we focus on the synthesis of a 1,4-amino
alcohol, namely 4-aminocyclohexanol 1. A modular route giving
access to both stereoisomers is desirable: trans-1 is a valuable
precursor of drugs such as ambroxol (a secretolytic agent) and
lomibuvir (an HCV protease inhibitor),[6,7] and also of novel
bioactive molecules with a potential clinical interest.[8] Cis-1 is
required as the building block for substituted isoquinolone
derivatives[9] and certain spirocyclic compounds.[10] Although 1 is
a
rather simple molecule, no efficient (and highly)
stereoselective synthesis of cis- or trans-1 has been reported:
The hydrogenation of paracetamol by applying mono- and
bimetallic catalysts yielded 1 with cis:trans ratios of (35-51):(49-
65)% of 4-aminocyclohexanol.[13] Additionally, the ambroxol
isomers containing a cis- or trans-1 moiety were prepared
starting from cyclohexa-1,3-diene via palladium-catalyzed 1,4-
diacetoxylation and allylic substitution with the subsequent
hydrogenation of the 4-aminocylohex-2-enol moieties.[14]
modular
and
by
choosing
stereocomplementary
amine
transaminases, both cis- and trans-4-aminocyclohexanol were
synthesized with good to excellent diastereomeric ratios.
Furthermore, we identified an amine transaminase that produces cis-
1,4-cyclohexanediamine with diastereomeric ratios >98:2. These
examples highlight that the high selectivity of enzymes enable short
and stereoselective cascade multifunctionalizations to generate
high-value building blocks from renewable starting materials.
Introduction
Due to their excellent regio- and stereoselectivity, enzymes
as catalysts often shorten synthetic routes, as demonstrated in
the synthesis of norephedrine, where a two-step enzymatic
approach was developed as an alternative to the seven-step
classical synthesis.[15]
Chiral amino alcohols are widely used as building blocks for
many pharmaceuticals such as adrenergic agents[1] or drugs
against cardiovascular, renal, and inflammatory diseases.[2]
Additionally, they have also found extensive application as
ligands and chiral auxiliaries in asymmetric organic synthesis.[3]
They are exemplary for many pharmaceuticals: Their syntheses
are waste-intensive processes due to the complexity of the
target molecules and the necessity to install multiple functional
groups and two or more stereocenters in one building block[4].
Therefore, the efficient multifunctionalization of simple starting
materials by a cascade of regio- and stereoselective reactions is
very attractive, and the development of chemoenzymatic one-
pot and tandem reactions has grown as an important research
A) Bio-based route to cyclohexane-1,4-dione 2
O
O
O
O
Δ
1. NaOEt, EtOH
2. H2SO4, H2O
O
O
O
O
H2O
O
O
O
O
2
B) Envisioned cascades towards 4-aminocyclohexanol 1
OH
possible
side product
OH
KRED
OH
route A
5
ATA
KRED
O
3
O
NH2
[a]
O. Sviatenko, K. V. Manideep, Dr. M. Höhne
Institut of Biochemistry
University of Greifswald
ATA
KRED
ATA
NH2
Felix-Hausdorff-Str. 4, 17489 Greifswald, Germany
E-mail: matthias.hoehne@uni-greifswald.de
www.protein-biochemistry.de
Dr. N. Ríos-Lombardía, Dr. F. Morís, Dr. J. González-Sabín
EntreChem SL
O
2
OH
cis- or trans-1
route B
NH2
[b]
O
4
possible
side product
Vivero Ciencias de la Salud, Santo Domingo de Guzmán, s/n 33011
Oviedo (Asturias), España
NH2
E-mail: jgsabin@entrechem.com
S. Merdivan, S. Günther
Institute of Pharmacy,
University of Greifswald
Friedrich-Ludwig-Jahn-Str. 17, 17489 Greifswald, Germany
Dr. Philipp Süss
6
[c]
[d]
Scheme 1. A) Cyclohexane-1,4-dione 2 can be produced from bio-based
succinic acid esterified with bioethanol. B) Proposed cascades towards cis-
and trans-4-aminocyclohexanol 1. The employed enzymes keto reductase
(KRED) and amine transaminase (ATA) must be selective to avoid formation
of by-products (diol 5 and diamine 6).
Enzymicals AG, Walther-Rathenau-Str. 49a, 17489 Greifswald,
Germany
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900733
ChemCatChem
FULL PAPER
With these precedents in mind, we envisaged two one-pot
enzymatic procedures for the stereoselective synthesis of cis-
and trans-4-aminocyclohexanol (Scheme 1) starting from 1,4-
cyclohexanedione 2. This diketone is a very simple, symmetric
undesired diol 5 (Table 1, entries 1, 2 and 4). Particularly, KRED
from Lactobacillus kefir (LK-KRED) and KRED-P2-C11
(Codexis) emerged as the best candidates leading to the highest
conversions (>93%) and the best ratio between the target 3 and
by-product 5 (entries 1 and 4).
starting material that can be obtained from succinic acid,[16]
a
renewable feedstock produced by engineered E. coli cells from
biomass.[17] The first route (route A) would involve the formation
of 4-hydroxycyclohexanone 3 from diketone 2 via ketoreductase-
catalyzed (KRED) monoreduction coupled with a subsequent
transamination by an amine transaminase (ATA). The second
option (route B) is to switch the order of steps of the previous
sequential approach: a selective monoamination of 2 yields 4-
aminocyclohexanone 4, followed by reduction of the remaining
carbonyl group of the intermediate to give 1. Ideally, these
sequences can be run as one-pot or cascade reaction using
stereoselective enzymes to obtain 1 in 100% theoretical yield
and without isolating any intermediate.
Table 1. Reduction of 1,4-cyclohexanedione 2 employing KREDs.
Enz. conc.
(mg/mL)
Entry
Enzyme
[3] (mM)[d]
[5] (mM)[d]
1[a]
2[b]
3[c]
4[c]
5[c]
LK-KRED
RR-KRED
P1-A04
0.1
45.1
44.6
37.7
47.2
36.5
1.2
0.8
5.3
0.5
1.4
Although this looks straightforward, several issues have to
be addressed: The challenge of route A is to find a selective
KRED that ideally stops after one reduction step to avoid
formation of the unwanted diol side-product 5. This requires a
selective catalyst efficiently discriminating between the very
similar di- and hydroxyketones 2 and 3. To the best of our
knowledge, no efficient chemical method for the direct selective
monoreduction of 2 is available. For accessing cis-1 or trans-1
product in the second step, stereoselective ATAs with
complementary stereopreferences have to be identified. Ideally,
the ATA would act only on the hydroxyketone 3, but not on the
diketone substrate to avoid the formation of 4 and possibly
diamine 6 as impurities. In the alternative route B analogous
constraints on substrate- and stereoselectivities of the used
catalysts apply: ATA must convert 2, but not the amino ketone 4,
to prevent formation of by-product 6; accessing cis-1 and trans-1
in the next step requires stereoselective KREDs. Additionaly,
KREDs should not convert 2 to avoid hydroxyketone or diol side
products. Although water is the natural enzyme environment, a
number of challenges must be addressed to implement such
enzymatic cascades depicted in Scheme 1. For reasons of
compatibility, parameters including pH, temperature, substrate
concentration, and co-solvents must be optimized, but also
specific enzyme activities and stabilities should be balanced and
inhibition avoided.
0.27
0.55
0.04
0.55
P2-C11
P2-D11
[a] Reaction conditions: 2 (50 mM), 1.0 mM NAD(P)+, propan-2-ol (100 mM,
0.76% v/v), 50 mM sodium phosphate buffer, pH 7.0, 30°C, 700 rpm, 12 h.
Lactobacillus kefir (LK)-KRED was produced in E. coli and applied as crude
cell lysate.
[b] Reaction conditions: the same as in [a], except pH 8.0. Rhodococcus ruber
(RR)-KRED was produced in E. coli and applied as crude cell lysate.
[c]
Reaction conditions: the same as in [a], except 125 mM potassium
phosphate buffer, 1.25 mM magnesium sulfate. Enzymes were used as
supplied in the Codexis® KRED screening kit.
[d] 3 (product) and 5 (by-product) concentrations were determined by GC.
To optimize the reduction towards high conversions, we
varied the percentage of propan-2-ol ranging from 0.76% to 40%
(v/v). Interestingly, the best performances remained those
achieved with the lowest percentage of propan-2-ol as reductant.
Thus, although the ratio 3:5 increased slightly for LK- KRED and
remarkably for KRED-P2-C11 to a higher percentage of the
monoalcohol, the conversion gradually declined with higher
percentages of propan-2-ol (Table S7, Supporting Information).
It is worth highlighting not only the selectivity, but also the high
degree of conversion that is reached in the reduction of 2 with
the low propan-2-ol content of 2 equivalents, contrasting with
previously reported enzymatic keto reductions of other
substrates.[18] This makes this reaction more attractive from an
economic point of view. Compared to previous attempts, this is
the first efficient enzymatic access to produce hydroxyketone
Results and Discussion
For the first step of route A (Scheme 1), we screened 24
KREDs from the Codexis® KRED Screening Kit and four
recombinantly expressed KREDs available in our laboratory
collection with
a
photometric NAD(P)H-assay employing
3,[19] which itself represents
a valuable building block for
compounds 2-4 as substrates. Half of the enzymes converted 2
and 3 with no or only a slight preference of the diketone 2.
However, there are also notably exceptions with significant
syntheses of compounds against immune disorders,[20] viral[21]
and neurological[22] diseases. Compared to our successful
monoreduction of 2 (route A), no KRED was identified having
activity towards 4 which prompted us to discard route B
(Scheme 1). Alternatively, we were also able to synthesize 3
selectivity and high activity towards
2 (See Supporting
Information). Next, suitable hits with a preference for 2 were
employed in biocatalytic reductions of 50 mM of 2 using 2
equivalents of propan-2-ol for cofactor recycling. We used this
low excess of the electron-donating propan-2-ol to prevent
‘over’reduction to the diol, but still to enable full reduction of one
keto-moiety to investigate the regio selectivity of the enzymes.
Three KREDs showed excellent activity and selectivity during
reduction of 2, and we observed only small amounts of 1-4% of
from the diol
5 by monooxidation employing laccases or
following a recently reported methodology for the oxidation of
sterically hindered alcohols[23] with NaOCl and AZADO, but
substrate conversions were below 50% (see the SI for full
details).
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900733
ChemCatChem
FULL PAPER
After optimizing the first step of the cascade using the
KREDs, we screened our collection of 82 recombinant ATAs
(wild type ATAs and variants) towards amination of 3 using
isopropylamine or alanine as amino donors. For the pre-
screening we produced ATAs in microtiter plates and analyzed
reactions by TLC with NBD-Cl derivatization to increase
sensitivity.[24] Among wild-type enzymes, only the (S)-selective
ATAs from Ruegeria pomeroyi (ATA-3HMU), Ruegeria sp. (ATA-
3FCR), and Chromobacterium violaceum (ATA-Cvi) were able to
aminate 3 (Table S1, SI). From our variant collection, 9 out of 28
variants of ATA-3FCR and ATA-3HMU converted 3 (Table S2,
SI) and several variants from the ATA Vibrio fluvialis (ATA-Vfl),
despite the inactivity of the wild type (Table S1, SI).
selective ATA are not known, what prevented the construction of
a model for comparison.
After the initial ATA-screening results, we aimed to increase
trans-selectivity by optimizing reaction conditions (pH,
temperature, DMSO concentration) for the two best ATA
candidates 113 and 234. Although the selectivity towards trans-1
formation increased significantly for ATA 113 at higher pH
values (Table S10, entries 1 and 3, SI), activities diminished
markedly. On the contrary, ATA-234 was able to keep high
conversions to the product at the same level as during the
screening phase, and an improved selectivity for trans-1
(cis/trans 20:80). We decided to select ATA-234 for further
preparative scale experiments, as it also showed higher stability
and activity under the investigated conditions.
Next, the most promising ATAs were purified (to reduce the
possibility of any unspecific reactions of cell lysate enzymes with
3) and employed in biocatalytic reactions to analyze conversion
and stereoselectivity. Most of these ATAs afforded 1 with
conversions higher than 60% using isopropylamine (Table S4,
SI). Interestingly, all of them exhibited marked cis-selectivity. In
particular, the variant ATA-3HMU W63Y led to 92% of the cis-
isomer of 1 (Table S4, entry 12, SI), and a variant of ATA-3FCR
designed for accepting bulky substrates (ATA-3FCR-4M)[25]
containing four amino acid substitutions afforded 1 with high
(>99%) cis-selectivity. Regarding reactions with D,L-alanine as
amino donor, four of the previous enzymes were active but led
to poor levels of 1 despite significant consumption of the
substrate (Table S5, SI). This substrate depletion also occurred
in the absence of an ATA, and the control reactions revealed
that the glucose dehydrogenase (used for cofactor recycling)
partially catalyzes reduction of 3 to the diol 5.
Water molecules in the active
site entrance
F92
F91
4-hydroxyl group of 3
Y63
L62
Y156
Cα
4-H atom of 3
Alternatively, we assayed the panel of engineered ATAs
from Codexis® with the aim of finding a trans-selective enzyme.
Most of the ATAs turned out to be very active and catalyzed the
amination of 3 with isopropylamine with high conversion, again
predominantly forming the cis-isomer of 1 (Table S6, SI). For
example, the ATAs 200, 237, 251 and 254 led to cis-1 with
perfect stereoselectivity and conversions of >95% (Table S6,
entries 1, 2, 4, 5, SI). On the other hand, only three enzymes,
namely ATAs 113, 234 and 415, exhibited slight trans-selectivity
(Table S6, entries 10, 17, 20, SI), yielding 1 with cis:trans ratios
of up to 3:7.
PLP
Figure 1. Modelled quinoid intermediate in ATA-3HMU W63Y formed after the
condensation of 3 (magenta sticks) and PMP (black sticks). Subsequent
catalytic steps will lead to cis-1. Note that for obtaining the trans-isomer, 3
must bind in an orientation where its 4-hydroxyl group and 4-H atom swap
their place. This would hinder solvation of the 4-hydroxyl group. The surface
shows the active site entrance tunnel, red spheres are oxygen atoms of water
molecules (added by the YASARA program prior to energy minimization). For
a discussion of the other conformations please see the Supporting Information
and Figure S8.
Intrigued by our observation that most transaminases
showed cis-selectivity, we conducted molecular modeling to
understand the molecular reasons for this trend. During catalysis,
3 can be bound to pyridoxamine-5’-phosphate in two different
orientations that will lead to the cis- or trans-configuration upon
transamination. We modeled the quinoid intermediate leading to
cis-1 (Fig. 1) in the active site of the ATA-3HMU W63Y variant.
The 4-hydroxyl group of the bound 3 (modeled in both chair
conformations) can be easily solvated because it faces the
solvent molecules that fill the active site entrance tunnel. For
trans-selective transamination, 3 must be bound in a way
orienting its 4-hydroxyl group towards the wall of the active site
tunnel (it would take the place of the hydrogen atom shown in
Fig. 1). Then it would face the hydrophobic amino acids L62 and
F91, what impedes its solvation and thus is energetically less
favorable. Because these two residues are highly conserved (as
revealed from a multiple sequence alignment containing >1000
proteins with sequence identities of 50-99 % with ATA-3HMU),
this explains the general cis-selectivity of wild-type PLP-fold type
I ATA. Unfortunately, the amino acid sequences of the trans-
Once both catalytic steps were reliably established, we
explored their combination in a one-pot sequential approach.
After the reduction of 80 mg 2 by LK-KRED was completed after
24-48 h (see SI for full details), we added the transamination
reaction components containing ATA and co-substrates and
adjusted the buffer to the previously identified pH optimum.
Pleasantly, the impact of the KRED and NADPH on the
bioamination step was negligible and the target amino alcohols
cis- or trans-1 were obtained with ≈90% conversion (Table 2,
entries 1-3). The reaction mixtures contained small amounts
(<9%) of the diol by-product 5 and the ATAs maintained good to
excellent stereoselectivities under the one-pot conditions,
although ATA-3FCR-4M performed slightly better in terms of
selectivity when used in an isolated reaction. Since
aminoalcohol 1 is a very hydrophilic compound, several work-up
strategies were tested for its isolation from the aqueous medium.
Commonly, derivatization with (Boc)2O helps to increase
hydrophobicity but due to high isopropylamine concentrations
would require using a large excess of the reagent. Solid-phase
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900733
ChemCatChem
FULL PAPER
extraction with Oasis WCX (weak cation exchanger) or Sep-
Pak® tC18 Silica columns was not working because 1 started to
elute from the columns already during the washing step. The
precipitation of
1
with organic acids, such as 3,3-
diphenylpropionic acid was also possible,[26] but required more
optimization steps due to high solubility of the salt. Thus, the
best isolation approach was an optimized extraction protocol
and crystallization as hydrochloride (see SI), facilitating ≈75 %
isolated yield in most cases.[27] Product purification by column
chromatography worked best after Boc derivatization (see
Supporting Information).
Table 2. Synthesis of cis- and trans-1 using one-pot sequential and cascade
Figure 2. Cascade synthesis of cis-4-aminocyclohexanol employing LK-KRED
reactions.
and ATA-3FCR-4M.
Entry Reaction
mode
C[b]
[%]
1
5
[%]
6
[%]
Scale
[mg 2]
Notably, the highest stereoisomeric ratio of cis-1 of 99:1 was
obtained with ATA-200 (entry 6). Here, we did not observe
transaminase inhibition by the diketone 2, but the ATA also
formed 1,4-cyclohexanediamine 6 as by-product, as identified by
NMR. Therefore, ATA-200 gives superior results in the
sequential mode. Reaction and protein engineering might help to
increase regioselectivity for this enzyme. In summary, our
approach is the first preparation of 1,4-amino alcohols by a two-
enzyme one-pot reaction and therefore extends the literature-
known synthesis methods for obtaining 1,2-[11] and 1,3-amino
alcohols.[12]
ATA
cis:trans[a]
1
sequential
3FCR-4M[d] 91
9
0
80
80:20
99:1
2
3
sequential
sequential
ATA-200[f] 85
ATA-234[f] 88
9
4
6
0
80
80
20:80
4
cascade
cascade
3FCR-4M[e] 98
2
3
0
0
160
160
80:20
5
ATA-234[f] 89
ATA-200[f] 44
25:75
99:1
Due to its broader substrate specifitiy, ATA-200 can be
applied for the synthesis of the diamine 6: Starting from 80 mg of
2, transamination furnished the diamine with nearly quantitative
6
7
cascade
ATA only
0
0
48
99
160
80
ATA-200[f]
0
98:2[c]
1
conversion (entry 7). Remarkably, H-NMR revealed that the cis
[a] Substrates and products, cis:trans ratios were determined with HPLC and
GC.
isomer was formed with excellent cis:trans ratio of 98:2 (Table 2,
Entry 7, see Figure S23 for NMR spectra). Cis-6 is used as
building block for tyrosine kinase inhibitor synthesis[28] and as
ligand (1,4-dach) in the platinum complex kiteplatin [Pt(1,4-
dach)Cl2], which recently has been shown to be active against
colon cancer cells lines that are resistant to cisplatin and
oxaliplatin.[29] Furthermore, cis- and trans-6 are used to modify
the properties of polymers such as polyamide[30]. To the best of
our knowledge, the preparation of cis-6 by transamination
represents its first highly stereoselective synthesis. The potential
of ATA200 for synthesizing chiral diamines is remarkable and
will be explored in the future, as only few ATAs are known that
can form this class of compounds. One example is a variant of
the Paracoccus denitrificans transaminase that converts the
cyclic isosorbide diketone to give the diamine with low yields.[31]
[b]
Percentage of total product (cis/trans)-1 formed from 3 as detected by
HPLC/GC.
[c] Cis:trans ratio was evaluated by 1H-NMR (see SI)
[d] ATA-3FCR-4M was produced in E. coli and applied as purified enzyme.
[e] ATA-3FCR-4M was produced in E. coli and applied as cell lysate.
[f] Commercial ATAs were applied as lyophilized enzymes.
Importantly, the one-pot reactions were also successful if
carried out as cascade reaction. By balancing the used enzyme
amounts, diol formation could be further suppressed to 0-3%
(entries 4-6). As the tandem reactions required using one
common pH for both steps, stereoselectivites decreased for TA-
3FCR-4M and ATA-234, but not when ATA-200 was used. A
time-course analysis (Figure 2) of the reactions revealed the
following challenges of the cascade mode: As expected, the
activity of the LK-KRED for the reduction decreased after the
first 20 h of the reaction, when 90% of the diketone 2 was
converted. At the same time, the amine transaminase activity
Conclusions
In conclusion, we developed a totally enzymatic one-pot
synthesis of cis-1 from
a potentially renewable bio-based
was very low: in the beginning, the hydroxy ketone
3
substrate. The molecular architecture of the active site of PLP
fold type I ATA causes their general cis-selectivity. We identified
only three trans-selective ATA variants, which also facilitated the
preparation of trans-1, although with lower stereo selectivity. Our
approach is an attractive alternative to the established 3-5-step
preparation of rac-1, starting from phenol or nitrobenzene (See
Scheme S2 in the Supporting Information). The approach can
be operated in sequential or in tandem mode and represents an
efficient functionalization of a simple starting material by a
sequence of precisely arranged reactions. It is a nice example
how cross-reactivity – a common challenge in cascades – can
accumulated and its conversion was less than 20% when 90%
of the diketone was consumed. However, the transamination
rate started to increase when the concentration of the diketone 2
was below 10 mM, indicating that 2 acted as an inhibitor of the
transaminase.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900733
ChemCatChem
FULL PAPER
[7]
[8]
[9]
I. Ando, N. Ogura, Y. Toyonaga, K. Hirahara, T. Shibata, T. Noguchi,
Intervirology 2013, 56, 302–309.
be avoided by identifying two regioselective enzymes that
discriminate between two very similar substrates such as the
employed diketone and hydroxyketone 2 and 3. The developed
synthesis avoids intermediate product isolation, metal catalysis,
minimizes by-product formation and affords high reaction yields.
S. J. Mishra, S. Ghosh, A. R. Stothert, C. A. Dickey, B. S. J. Blagg,
ACS Chem Biol. 2017, 12, 244-253.
O. Plettenburg, A. Hofmeister, D. Kadereit, J. Brendel, M. Loehn, J.-M.
Altenburger, US8796458, 2014.
[10] W. Yao, J. Zhuo, C. Zhang, US9371323, 2016.
[11] L. Ríos-Solís, P. Morris, C. Grant, A. O. O. Odeleye, H. C. Hailes, J. M.
Ward, P. A. Dalby, F. Baganz, G. J. Lye, Chemical Engineering
Science 2015, 122, 360–372.
Experimental Section
[12] H. Kohls, M. Anderson, J. Dickerhoff, K. Weisz, A. Córdova, P.
Berglund, H. Brundiek, U. T. Bornscheuer, M. Höhne, ChemCatChem
2015, 357, 1808-1814.
All experimental details on analytics, enzyme production and screening,
and all biocatalysis reactions are given in the Supporting Information.
Below follows a representative procedure for the concurrent cascade
synthesis of cis-4-aminocyclohexanol 1: The reaction was performed at a
1.5 mmol scale (concentration of 1,4-cyclohexanedione was 50 mM) in
an opened one-necked flask equipped with a magnetic stirring bar
(stirring at 250 rpm) and a thermometer (at 30°C). Freshly isolated cell
lysate of LK-KRED (0.2 mg/mL) and ATA-200 (2 mg/mL) were added to
30 mL of 100 mM potassium phosphate buffer, pH 7.5, containing 1 mM
NADP+, 0.76% (v/v) isopropanol (100 mM), 1 mM MgCl2, 500 mM
isopropylamine, 1 mM PLP, and 2% v/v DMSO. After 48 hours, the
reaction was stopped after 44% conversion (Table 2) and the product
was isolated as described in the Supporting Information and afforded
light yellow crystals of cis-4-aminocyclohexanol hydrochloride (99:1%
cis:trans ratio).
[13] M. Cerro-Alcarón, A. Guerrero-Ruίz, I. Rodrίguez-Ramos, Catalysis
Today 2004, 93-95, 395-403.
[14] A. L. E. Larsson, R. G. P. Gatti, J.-E. Bäckvall, J. Chem. Soc., Perkin
Trans. 1997, 1, 2873-2877.
[15] T. Sehl, H. C. Hailes, J. M. Ward, R. Wardenga, E. von Lieres, H.
Offermann, R. Westphal, M. Pohl, D. Rother, Angew. Chem. Int. Ed.
2013, 52, 6772–6775; Angew. Chem. 2013, 125, 6904–6908.
[16] A. T. Nielsen, W. R. Carpenter, Org. Synth. 1965, 45, 25.
[17] C. Thakker, I. Martinez, K.-Y. San, G. N. Bennet, Biotechnol. J. 2012, 7,
213-224.
[18] a) S. García-Cerrada, L. Redondo-Gallego, F. J. Martínez-Olid, J. A.
Rincón, P. García-Losada, Org. Process Res. Dev., 2017, 21 (5), 779–
784; b) X. Ju, Y. Tang, X. Liang, M. Hou, Z. Wan, J. Tao, Org. Process
Res. Dev., 2014, 18, 827−830.
[19]
Only one approach was recently reported that facilitates a selective
reduction, but needs a large amount of whole cells (10 g wet cells for
40 mg of substrate). S. Krishnan, S. Narayan, A. Chadha, AMB
Express 2016, 6, 1-15.
Acknowledgements
The work is done within BIOCASCADES project. The
BIOCASCADES project is supported by the European
Commission under the Horizon 2020 program through the Marie
Skłodowska-Curie actions: ITN-EID under the Environment
Cluster. Grant Agreement #634200. We thank to Prof. U. T.
Bornscheuer for his continuous support and fruitful discussions.
[20] B. A. Sherer, N. Brugger, WO2017106607, 2017.
[21] S.-Y. Sit et al, WO2015157483, 2015.
[22]
E. Busto, L. Martínez-Montero, V. Gotor, V. Gotor-Fernández, Eur. J.
Org. Chem. 2013, 4057–4064.
[23] E. Liardo, N. Ríos-Lombardía, F. Morís, F. Rebolledo, J. González-
Sabín, ACS Catal. 2017, 7, 4768-4774.
[24] a) G. Bahrami, B. Mohammadia, J. Chromatogr. B 2007, 850, 400–404;
b) S. T. Ulu, Spectrochim. Acta, Part A 2009, 72, 138–143.
Keywords: amino alcohols • asymmetric catalysis • enzyme
[25] I. V. Pavlidis, M. S. Weiss, M. Genz, P. Spurr, S. P. Hanlon, B. Wirz, H.
Iding, U. T. Bornscheuer, Nat. Chem. 2016, 8, 1076–1082.
catalysis • amine transaminase • ketoreductase
[26] D. Hülsewede, M. Tänzler, P. Süss, A. Mildner, U. Menyes, J. von
Langermann, Eur. J. Org. Chem. 2018, 2130–2133.
[1]
[2]
G. Gübitz, B. Pierer, W. Wendelin, Chirality 1992, 4, 333-337.
R. J. Albers, L. Ayala, S. S. Clareen, M. Mercedes Delgado Mederos, R.
Hilgraf, S. G. Hegde, K. Hughes, A. Kois, V. Plantevin-Krenitsky, M.
McCarrick, L. Nadolny, M. S. S. Palanki, K. Sahasrabudhe, J. Sapienza,
C. Vista, Y. Satoh, M. K. Sloss, E. Sudbeck, J. Wright, US20090275564,
2009.
[27] Additionally,
a method for upgrading stereoisomeric excess of 4-
aminocyclohexanol by crystallization has been reported: B. Lehmann,
EP0909753, 2004.
[28] V. Simov, S. V. Deshmukhc, C. J. Dinsmoreb, F. Elwooda, R. B.
Fernandeza, Y. Garciaa, C. Gibeaua, H. Gunaydina, J. Jungd, J. D.
Katza, B. Kraybille, B. Lapointea, S. B. Patela, T. Siua, H. Sua, J. R.
Young, Bioorg. Med. Chem. Lett. 2016, 26, 1803–1808.
[29] a) J. D. Hoeschele, H. D. H. Showalter, A. J. Kraker, W.L. Elliott, B. J.
Roberts, J. W. Kampf, J. Med. Chem. 1994, 37, 2630–2636, b) N.
Margiotta, C. Marzano, V. Gandin, D. Osella, M. Ravera, E. Gabano, J.
A. Platts, E. Petruzzella, J. D. Hoeschele, G. Natile, J. Med. Chem.
2012, 55, 7182–7192; c) B. J. Pages, J. Sakoff, J. Gilbert, Y. Zhang, S.
M. Kelly, J. D. Hoeschelee, J. R. Aldrich-Wright, Dalton Trans. 2018, 47,
2156-2163.
[3]
[4]
a) D. J. Ager, I. Prakash, D. R. Schaad, Chem. Rev. 1996, 96, 835–
875; b) R. Noyori, M. Kitamura, Angew. Chem. Int. Ed. 1991, 30, 49-69;
Angew. Chem. 1991, 103, 34–55.
Approx. 2/3 of the 2576 chiral drugs listed in a collection of all FDA-
approved drugs contain more than one stereocenter (see
http://drugcentral.org and O. Ursu, J. Holmes, J. Knockel, C. G. Bologa,
J. J. Yang, S. L. Mathias, S. J. Nelson, T. I. Oprea,, Nucleic Acids Res.
2017, 45: D932–D939.)
[5]
For chemo-enzymatic cascades, see: a) H. Gröger, W. Hummel, Curr.
Opin. Chem. Biol. 2014, 19, 171-179; b) F. Rudroff, M. D. Mihovilovic,
H. Gröger, R. Snajdrova, H. Iding, U. T. Bornscheuer, Nat. Catal. 2018,
1, 12-22.
[30] a) G. Zhang, G.-M. Yan, H.-H. Ren, Y. Li, X.-J. Wang, J. Yang, Polym.
Chem. 2016, 7, 44-53; b) Y. Inoue, H. Nishino, Y. Watanabe,
JP2008298722, 2008.
For multi-enzymatic cascades, see: c) E. Ricca, B. Brucher, J. H.
Schrittwieser, Adv. Synth. Catal. 2011, 353, 2239-2262; d) R. C. Simon,
N. Richter, E. Busto, W. Kroutil, ACS Catal. 2014, 4, 129-143; e) J.
Muschiol, C. Peters, N. Oberleitner, M. D. Mihovilovic, U. T.
Bornscheuer, F. Rudroff, Chem. Commun. 2015, 51, 5798-5811; f) S. P.
France, L. J. Hepworth, N. J. Tuner, S. L. Flitsch, ACS Catal. 2017, 7,
710-724; g) J. H. Schrittwieser, S. Velikogne, M. Hall, W. Kroutil, Chem.
Rev. 2018, 118, 270-348.
[31] A. Lerchner, S. Achatz, C. Rausch, T. Haas, A. Skerra, ChemCatChem
2013, 5, 3374–3383.
Supporting information is available at
[6]
M. Malerba, B. Ragnoli, Expert Opin. Drug Metab. Toxicol. 2008, 4,
119-129.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900733
ChemCatChem
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
Avoiding cross-reactivity: Identifying
selective keto reductases and amine
transaminases facilitated the synthe-
sis of 4-aminocyclohexanol isomers.
The starting diketone can be prepared
from renewable resources, and its
enzymatic functionalization is possible
in a one-pot or concurrent cascade
approach. The 4-aminocyclohexanol
isomers are valuable motifs for the
synthesis of pharmaceuticals for the
treatment of cardiovascular, respira-
tory, viral diseases, diabetes and
obesity.
Olha Sviatenko, Nicolás Ríos-
Lombardía, Francisco Morís, Javier
González-Sabín,* Simon Merdivan,
Sebastian Günther, Philipp Süss, and
Matthias Höhne*
Page No. – Page No.
Selective synthesis of 1,4-amino
alcohols by an enzymatic approach
This article is protected by copyright. All rights reserved.