Uncommon overoxidative catalytic activity in a new halo-tolerant alcohol dehydrogenase

Article abstract of DOI:10.1002/cctc.202001112

Alcohol dehydrogenases (ADH) are versatile and useful enzymes employed as biocatalysts, especially for the selective oxidation of primary and secondary alcohols, and for the reduction of carbonyl moieties. A new alcohol dehydrogenase (HeADH-II) has been identified from the genome of the halo-adapted bacterium Halomonas elongata, which proved stable in the presence of polar organic solvents and salt exposure. Unusual for this class of enzymes, HeADH-II lacks enantiopreference and is capable of oxidizing both alcohols and aldehydes, enabling a direct overoxidation of primary alcohols to carboxylic acids. HeADH-II was coupled with a NADH-oxidase from Lactobacillus pentosus (LpNOX) to increase the process yields and allowing recycling of the cofactor. The enzymatic oxidation of primary alcohols was also paired with in situ condensation of the intermediate aldehydes with hydroxylamine to prepare the corresponding aldoximes, with particular attention to perillartine (a powerful sweetener), whose enzymatic synthesis starting from natural sources, leads to an equally natural product.

Full text of DOI:10.1002/cctc.202001112

The European Society Journal for Catalysis
Supported by
Accepted Article
Title: Uncommon overoxidative catalytic activity in a new halo-tolerant
alcohol dehydrogenase
Authors: Martina Contente, Noemi Fiore, Pietro Cannazza, David
Roura Padrosa, Francesco Molinari, Louise Gourlay, and
Francesca Paradisi
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.202001112
Link to VoR: https://doi.org/10.1002/cctc.202001112
01/2020

10.1002/cctc.202001112
ChemCatChem
FULL PAPER
Uncommon overoxidative catalytic activity in a new halo-tolerant
alcohol dehydrogenase
Martina L. Contente,[a] Noemi Fiore, [b] Pietro Cannazza,[b] David Roura Padrosa,[c] Francesco
Molinari,[b]* Louise Gourlay, [d] Francesca Paradisi [a,c]
*
[a]
[b]
Dr. M. L. Contente, Prof. F. Paradisi
School of Chemistry
University of Nottingham
University Park, Nottingham, NG7 2RD, UK
N. Fiore, P. Cannazza, Prof. F. Molinari
Department of Food, Environmental and Nutritional Sciences (DeFENS)
University of Milan
via Mangiagalli 25, 20133 Milan, Italy
E-mail: francesco.molinari@unimi.it
[c]
[d]
Dr. D. Roura Padrosa, Prof. F. Paradisi
Department of Chemistry and Biochemistry
University of Bern
Freiestrasse 3, 3012 Bern, Switzerland
E-mail: francesca.paradisi@dcb.unibe.ch
Dr. Louise Gourlay
Department of Biosciences (DBS),
University of Milan
via via Celoria 26, 20133 Milan, Italy
Supporting information for this article is given via a link at the end of the document.
Abstract: Alcohol dehydrogenases (ADH) are versatile and useful
enzymes employed as biocatalysts, especially for the selective
oxidation of primary and secondary alcohols, and for the reduction of
carbonyl moieties. A new alcohol dehydrogenase (HeADH-II) has
been identified from the genome of the halo-adapted bacterium
Halomonas elongata, which proved stable in the presence of polar
organic solvents and salt exposure. Unusual for this class of enzymes,
HeADH-II lacks enantiopreference and is capable of oxidizing both
alcohols and aldehydes, enabling a direct overoxidation of primary
alcohols to carboxylic acids. HeADH-II was coupled with a NADH-
oxidase from Lactobacillus pentosus (LpNOX) to increase the process
yields and allowing recycling of the cofactor. The enzymatic oxidation
of primary alcohols was also paired with in situ condensation of the
intermediate aldehydes with hydroxylamine to prepare the
corresponding aldoximes, with particular attention to perillartine (a
powerful sweetener), whose enzymatic synthesis starting from natural
sources, leads to an equally natural product.
biocatalysts capable of overoxidative activity towards primary
alcohols under mild conditions is essential for green and
sustainable chemistry.[8] Classically, the oxidation of primary
alcohols is chemoselective and leads to the exclusive formation
of aldehydes. Further oxidation is generally obtained by coupling
ADHs with an aldehyde dehydrogenase (AldDH) or by using
whole cell systems, such as acetic acid bacteria.[10,11] Alcohol
oxidases represent another way to access aldehydes and
ketones starting from primary and secondary alcohols,
respectively. In these reaction, molecular oxygen is reduced to
hydrogen peroxide. In order to avoid enzyme deactivation, a
second enzyme (e.g. catalase) is usually added, particularly on
preparative scale.[12,13] However, oxidation of aldehydes via ADH
has been observed with horse liver dehydrogenase (HLADH),
which catalyzes the dismutation of short-chain aliphatic
aldehydes, in the presence of an excess of NAD+.[6] The complex
HLADH-NAD+ firstly performs the oxidation of the aldehyde
generating the carboxylic acid, while the newly formed complex
(HLADH-NADH) binds another aldehyde molecule, catalyzing its
reduction to alcohol. In this equilibrium, the hydration of the
aldehyde plays an important role since the free form is the
substrate for reduction, whereas the hydrate gem-diol is the actual
substrate for oxidation. With simple aliphatic aldehydes, which
freely interconvert between the two forms (free and gem-diol one),
an equivalent amount of alcohols and carboxylic acids is formed
at the end of the process with HLADH, thus promoting an
enzymatic Cannizzaro reaction.[7] Könst et al. screened 70
different ADHs and only 9 showed significant accumulation of acid,
demonstrating that this activity is rather an exception.[9] In
Introduction
Alcohol dehydrogenases (ADHs) have acquired increasing
interest as versatile chemo- and enantioselective biocatalysts.[1-3]
While classic ADH-mediated oxidation of alcohols to the
corresponding carbonyl compounds are well known,[4-6] just a few
examples of complete (primary) alcohol oxidation to carboxylic
acids have been reported to date, often involving complex
cascade enzymatic reactions.[7-9] The identification of novel
1
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001112
ChemCatChem
FULL PAPER
particular, a mutant ADH (ADH-9V1), with lower binding affinity
towards NADH than the wild-type enzyme, was able to better
catalyze aldehyde oxidation. The conversion was further
increased by recycling the cofactor with an H2O-forming
NAD(P)H-oxidase (NOX) from Lactobacillus sanfranciscensis and
performing the reaction at pH 9.0 where a higher amount of gem-
diol is present. Under optimized conditions (21 h), ADH-9V1
catalyzed the oxidation of primary alcohols into carboxylic acids
with conversions ranging from 9 to 37% for aromatic substrates,
and 70-73% for simple aliphatic alcohols.[9]
from Lactobacillus pentosus (LpNOX) (10 U/mL) and FAD (0.1
mM) (Scheme 1).
Scheme 1. Oxidation of cinnamyl alcohol and cinnamaldehyde with cofactor
recycling
To combine the two different enzymes in an one-pot reaction,
New enzymes with improved solvent tolerance may be found
in extremophiles.[14] ADHs with greater stability towards polar
organic solvents have been expressed from halophilic
microorganisms[15] and from amplified genomes from unculturable
species occurring in brine pool.[16,17] In this work, an ADH gene
(HeADH-II) identified from the genome of Halomonas elongata
was codon optimized and expressed in E. coli; Halomonas
elongata is a halo-adapted bacterium which has been previously
used as source of enzymes with remarkable solvent-tolerance
and stability under operating conditions.[18-21] HeADH-II was
characterized and used as biocatalyst, proving its ability to oxidize
both primary alcohols and aldehydes, showing enhanced stability
towards water-miscible co-solvents (e.g., DMSO).
Size exclusion chromatography indicated that the enzyme
assembles as a tetramer, and in-silico modelling (together with a
preliminary low-resolution crystal structure), suggests that
HeADH-II exhibits the canonical monomeric fold of the zinc-
dependent ADHs (See Supporting Information). Comparison of
the overall fold revealed similarities with ADH enzymes of
mammalian origin that also catalyze alcohol overoxidations.
the reaction conditions were set so that also LpNOX (employed to
re-oxidize the cofactor) could optimally perform; LpNOX is active
between pH 5.0-9.0 and 25-50 °C.[23] Therefore, preparative
oxidation of both cinnamyl alcohol and cinnamaldehyde were
firstly carried out at different pHs (5.0-8.0) and 30 °C. The reaction
leads to the accumulation of aldehyde, as also seen in the
experiments without cofactor regeneration with a maximum
conversion around 93-95%, while cinnamic acid was produced
only in small amounts. The biotransformation did not proceed
further after the apparent equilibrium was reached within one hour,
regardless of the medium pH. This behavior may be ascribed to
the low stability of LpNOX, with subsequent interruption of
cofactor recycling. To check this hypothesis, fresh LpNOX (10 U)
was added after each hour, showing that oxidation proceeded to
carboxylic acid with conversions up to 99% (See Supporting
Information Figure S3), again independently of the pH.
The oxidative reaction with fed-batch addition of LpNOX was
then applied to different aromatic alcohols and aldehydes (Table
1).
Results and Discussion
Table 1. Oxidation of different primary alcohols using HeADH-II.
The activity and stability of HeADH-II were firstly evaluated
using the reduction of cinnamaldehyde to cinnamyl alcohol as
standard reaction (See Supporting Information). The highest
activity of HeADH-II was found at pH 5.0 in a temperature range
of 45-65 °C. The presence of salts has been well tolerated; the
highest activity was surprisingly recorded in the presence of 3 M
NaCl; remarkably, the addition of DMSO had minor effect on the
enzyme performance since only 8% of the original activity was
lost in the presence of 10% (v/v) DMSO, and just 18% with 20%
(v/v) of the same co-solvent. The highest stability was found at
25 °C and pH 8.0; high stability was also observed in the presence
of NaCl and KCl and DMSO (10-20%), confirming the possibility
of using HeADH-II with water-miscible co-solvents or relatively
high salt concentrations.
Reduction of cinnamaldehyde (20 mM) into cinnamyl alcohol,
with a stoichiometric amount of NADH, was carried out at pH 8.0,
30 °C (best compromise between activity and stability) in the
presence of HeADH-II (1 U/mL), showing 65% molar conversion.
Relatively high concentrations of DMSO (15%) were used to
better solubilize cinnamaldehyde, which is only slightly soluble in
water (1.05 g/L, 8 mM at 20 °C).[22]
Reaction conditions: a) 20 mM substrate in DMSO (15% v/v), 0.1 mM NAD+,
0.1 mM FAD, HeADH-II (1 U/mL) in Tris-HCl buffer 100 mM, pH 8.0, fed-batch
addition of LpNOX (10 U) every hour; b) 20 mM substrate in DMSO (15% v/v),
0.1 mM NAD+, 0.1 mM FAD, HeADH-II (1 U/mL), LpNOX (10 U/mL), in Tris-HCl
buffer 100 mM, pH 8.0 at 30 °C.
Entry
Substrate
Aldehyde
Acid
(%)
Time
(h)
4
(%)
1
2
3
4
5
6
7
Cinnamyl alcohola
-
10
5
-
> 98
90
Benzyl alcohola
16
16
8
Piperonyl alcohola
95
4-Methoxybenzyl alcohola
2-Phenyl-1-ethanolb
3-Phenyl-1-propanolb
(R,S)-2-Phenyl-1-propanolb
> 98
> 98
> 98
> 98
-
2
-
2
-
3
The reverse reaction (oxidation of cinnamyl alcohol) was
performed in the presence of stoichiometric amounts of NAD+ at
different pHs. Oxidation generally occurred with formation of cin-
namaldehyde and, unexpectedly, cinnamic acid, albeit with low
conversion of the substrate (Scheme 1). To increase the yields of
the oxidations, bioconversions were carried out with HeADH-II (1
U/mL) in the presence of NAD+ (0.1-10 mM), a NADH-oxidase
The bioconversion of benzyl alcohol derivatives (Entries 2-4,
Table 1) occurred in a manner similar to what was observed with
cinnamyl alcohol, with an initial accumulation of aldehyde and
complete oxidation into carboxylic acids with fed-batch addition of
LpNOX. For fast-reacting substrates (Entries 5-7, Table 1), acid
2
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001112
ChemCatChem
FULL PAPER
formation was complete in short timescales without further
addition of LpNOX. Surprisingly, despite the short reaction time,
no trace of enantioselectivity was observed with racemic 2-
phenyl-1-propanol (Entry 7), which was completely oxidized in 3
hours obtaining the corresponding acid as racemate. With respect
to the previous reported methodology, involving a mutated ADH
with lower affinity for NADH,[9] the new strategy allows a 4-fold
higher substrate loading (20 mM) while the aromatic acids were
obtained with greater yields and in shorter reaction times.
Reaction conditions: 20 mM substrate in DMSO (15% v/v), 0.1 mM NAD+, 0.1
mM FAD, HeADH-II (1 U/mL), LpNOX (10 U/mL), in Tris-HCl buffer 100 mM, pH
8.0 at 30 °C. Reaction volume 7-10 mL.
HeADH-II was subsequently employed for the direct oxidation
of aldehydes into the corresponding carboxylic acids with high
conversion (Table 2).
Aldehyde oxidation was generally rapid, and no further
To further investigate the substrate scope of this new enzyme, addition of LpNOX was necessary to obtain high yields in short
secondary alcohols were also tested under the above described
reaction conditions (Figure 1); no fed-batch addition of LpNox was
reaction times. Transient formation of aldehydes in the two-step
enzymatic oxidation of primary alcohols at pH 8.0 led us to explore
the possibility of condensing the intermediate aldehydes with
hydroxylamine, thus furnishing the corresponding aldoximes.[25]
One-pot preparation of aldoximes from primary alcohols is an
interesting methodology to trap aldehydes which often present a
poor stability limiting their production in high yields.[26,27] In this
instance, three different substrates were selected (Table 3).
necessary. Virtually all reported ADHs display
a strong
enantiopreference towards either R or S secondary alcohols,[24]
but while HeADH-II showed activity towards (R,S)-1-phenyl
ethanol (1a, Figure 1, 65% m.c., 2 h), a standard substrate for this
enzymatic class, no enantiopreference was observed. No
conversion was obtained with the bulkier (R,S)-1-phenyl-2-
propanol (1b, Figure 1).
Table3. Oxidation of primary alcohols using HeADH-II in the presence of
NH2OH.
Entry
Substrate
Aldoxime (%)
Time (h)
1
2
3
Cinnamyl alcohol
Benzyl alcohol
Perillyl alcohol
86
86
82
3
6
2
Figure 1. Oxidation of secondary alcohols using HeADH-II. Reaction
conditions: 20 mM substrate in DMSO (15% v/v), 0.1 mM NAD+, 0.1 FAD,
HeADH-II (1 U/mL), LpNOX (10 U/mL), in Tris-HCl buffer 100 mM, pH 8.0 at
30 °C.
Reaction conditions: 20 mM substrate in DMSO (15% v/v), NH2OH.HCl (24 mM),
0.1 mM NAD+, 0.1 mM FAD, HeADH-II (1 U/mL), LpNOX (10 U/mL), in Tris-HCl
buffer 100 mM, pH 8.0 at 30 °C. Reaction volume 7-10mL.
HeADH-II was assayed also in the reductive direction with
acetophenone (20 mM) in the presence of a stoichiometric
amount of NADH. Again, the enzyme showed no
enantioselectivity and the corresponding 1-phenyl ethanol was
obtained as a completely racemic product (70% m.c., 6 h) (See
Supporting information).
The biotransformations performed in the presence of NH2OH
allowed the recovery of the aldoximes with conversions >80%. It
should be noted that aldoximes have a wide range of applications
as intermediates in organic synthesis. They can be reduced to
amines,[28] oxidized to nitrile oxides,[29] and undergo a Beckman
rearrangement leading to the corresponding amides. Finally,
aldoximes can be converted to nitriles through a dehydration
reaction.[30] Synthesis of perillartine from perillyl alcohol (Entry 3,
Table 3, Scheme 2) was investigated as a molecule of interest in
the food industry. It is typically employed as non-protein
sweetener since it is 2000-fold sweeter than sucrose. Perillartine
is usually chemically prepared, starting from perillaldehyde
extracted from plants (e.g., Perilla frutescens).[31] Particular
attention to the synthesis of this compound was promoted by the
need to develop natural processes for the preparation of food
additives: the obtainment of perillartine through the enzymatic
Table2. Oxidation of different aldehydes using HeADH-II.
Entry
Substrate
Acid
(%)
Time (h)
conversion of
a
natural substrate allows to label and
1
2
3
4
5
6
7
Cinnamaldehyde
94
6
8
8
8
2
2
2
commercialize the corresponding product as natural too, thus
increasing its market value.
Benzaldehyde
> 98
> 98
> 98
> 98
> 98
>99
Piperonal
4-Methoxybenzaldehyde
Phenylacetaldehyde
3-Phenylpropionaldehyde
(R,S)-2-Phenylpropanal
3
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001112
ChemCatChem
FULL PAPER
Scheme 2. Preparation of perillartine from perillyl alcohol
shown that the proton relay is mediated by residues T48 (T44 in
HeADH-II), the 2'-hydroxyl of the nicotinamide ribose in NAD(H),
and the imidazole of His47, with the latter acting as the catalytic
base in the reaction. Given the sequence and structural
conservation of analogous residues in HeADH-II, a similar
mechanism remains to be elucidated.
The one-pot enzymatic synthesis of perillartine was carried
out also on a larger 50 mL scale, maintaining a 20 mM substrate
concentration. After 2 h biotransformation the final product was
recovered with 78% of isolated yield.
Concerning horse liver ADH (HLADH), the zinc coordinating
cysteines and histidine residues were matched (local sequence
The 3D structure of HeADH-II: in silico modelling
ID 35.7% over 41 equivalenced residues);
a coordinating
In order to understand the structural arrangement of HeADH-
II that may explain its peculiar overoxidative activity, a model was
created with MODELLER,[32] using as query the holo-form of the
bacterial ADH from Brucella melitensis (BmADH, PDB entry
3MEQ) which has over 70% sequence identity and is also a
tetramer, as observed for HeADH-II in solution by size exclusion
chromatography studies (See Supporting Information Figure S6).
In addition, a preliminary crystal structure of the apo-HeADH-II
was achieved at low resolution (See Supporting Information
Figure S8 and S9).
glutamate was not observed in the structure of HLADH, although
the authors predicted its role computationally.[34] The authors also
predicted the role of S48 (T44 in HeADH-II) in proton shuttling.
As mentioned, ADHs normally exhibit very high
enantioselectivities, closely related to the shape and size of the
catalytic cleft which tends to display clearly differentiated small
and large pockets. Such architecture favors efficient binding of
just one stereoisomer (in the oxidative direction), and promotes
the hydride transfer only to one face of the carbonyl molecule (in
the reductive direction).[35] In contrast, HeADH-II appears to lack
any enantiopreference with the tested substrates in either
catalytic direction. Only a few wild-type ADHs have been found to
have reduced enantiopreference[36,37] while protein engineering
can lead to a less selective variants.[38-40] The catalytic cleft of
HeADH-II appears to be symmetric, with undistinguishable
pockets, unlike the majority of ADH structures reported (including
that of BmADH).[37] To illustrate the differences in the catalytic cleft,
RCR ((R)-selective carboxyl reductase from Candida
parapsilosis) and TeSADH (non-enantioselective secondary
alcohol dehydrogenase from Thermoanaerobacter ethanolicum)
were compared to HeADH-II.[41,42] Despite a less than 30% of
sequence homology (See Supporting Information, Figure S10A),
the three enzymes exhibit a similar folding and a very similar
architecture of the active site (See Supporting Information, Figure
S10B), mainly differing in the size of their pockets (Figure 2A). In
RCR, the two pockets are clearly distinguishable with a small
pocket limited by F285 and W286. Upon mutation to alanine of
either of those (F285A or W285A), the additional space causes
the loss of enantioselectivity for (R,S)-2-phenyl-1-propanal.[41] The
HeADH-II quasi symmetric catalytic site displays a I290 and a
V291 which leave enough space in the upper pocket for bulky
substituent (Figure 2B). In addition, T44 and specially the
orientation of W91, extend the size of the other pocket, which
mimics RCR and TeSADH. Both positions have been previously
probed to investigate the stereoselectivity of HLADH.[39]
Specifically, the presence of a phenylalanine in the corresponding
W91 position, oriented towards the catalytic site, reduces the size
of the pocket (See Supporting information, Figure S11). Notably,
W91 in HeADH-II appeared to be quite static during the short MD
simulation, staying parallel to the opening of the catalytic cleft, as
in TeSADH.
The He-ADH-II active site analysis: overoxidation ability and
enantioselectivity
The HeADH-II chain contains two zinc ions bound at each
subdomain (See Supporting Information Figure S8 and Figure S9).
The catalytic zinc cation, involved in substrate binding, is bound
at the nucleotide binding subdomain surrounded by C42, C152,
T44 and H67. Another residue in close vicinity to the zinc binding
site is E66. Interestingly, the involvement of a glutamate in zinc
coordination was observed as an atypical feature in human liver
χχ ADH (PDB entry 1TEH). In agreement with the overoxidative
activities of HeADH-II, the human enzyme catalyzes substrate
oxidation of long-chain alcohols, and also S-hydroxymethyl-
glutathione.[33]
HeADH-II and human liver χχ ADH share a low sequence
identity of 28.9% (for 304/380 fitted residues), and a structural
identity of 99.7% (See Supporting Information Figure S8),
although the local sequence identity (41 equivalenced residues)
surrounding the three matched active site residues, increases to
48.8%, reflecting increased conservation in this region.
Direct coordination of E68 (human χχ ADH) has been
proposed to represent a reaction intermediate that assists in
product release and the water exchange step during catalysis.
Accordingly, this glutamate is replaced by a coordinating water
molecule in chain A of the human χχ ADH. The resolution of our
data, however, was not sufficient to visualize water molecules.
The oxidation of long-chain alcohols, accommodated by a
larger than usual active site in human liver χχ ADH, is suggested
to occur via a catalytic mechanism similar to that of class I ADH
that requires a significant decrease in the pKa of the alcohol
substrate due to direct coordination with the zinc ion. As observed
upon superposition of the two enzymes, HeADH-II also has a
comparably large active site (See Supporting Information Figure
S8). The catalytic mechanism of human liver χχ ADH is proposed
to employ H47 (H67 in HeADH-II) as the catalytic base, with a
double role also in NAD(H) binding. During the mechanism, a
proton-relay pathway is suggested, involving transfer of the
substrate’s hydroxyl proton, with passage from inside the active
site to the solvent. In class I ADHs, mutagenesis studies have
4
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001112
ChemCatChem
FULL PAPER
Docking studies of HeADH-II with both (R)- and (S)-1-
phenylethanol were performed; both enantiomers could be fitted
in the active site with identical energies, very similar
conformations, and catalytically plausible orientations (Figure 2B).
could offer major advantages in hydrogen borrowing cascades for
the synthesis of chiral amines from the racemic alcohols.[43]
To further optimize the process, in situ recycling of the
cofactor was performed by adding a second enzyme (LpNOX)
that allowed the generation of a self-sustaining enzymatic system
with the use of catalytic amounts of NAD+. Unfortunately, the
NADH-oxidase from Lactobacillus pentosus (LpNOX) was found
to be unstable at prolonged reaction times, so fed-batch
experiments were carried out to obtain a two-step transformation
with almost complete conversion.
Further applications
particularly focused on the enzymatic preparation of perillartine
(deriving from the condensation of aldehydes with hydroxylamine)
were performed. This reaction proceeds at room temperature and
under the atmospheric pressures. Using an NAD+ regenerating
system, catalytic amount of the cofactor could be employed. As
the only by-product is water, this system can be considered
sustainable and applicable for the preparation of natural food
ingredients.
Experimental Section
Figure 2. A) Comparison of the catalytic pockets of HeADH-II, TeSADH and
RCR. The labelled residues which are defining each of the pockets are coloured
in light cyan or yellow and their surface is shown. B) Detail of the docking of
(R)- and (S)- 1-phenylethanol. The S enantiomer is shown in blue and the R in
green. The docking energies obtained were -5.4 kcal/mol and -5.6 kcal/mol,
respectively.
General
NMR spectra were recorded on a Varian Gemini 300 MHz
spectrometer using the residual signal of the deuter-ated solvent
as internal standard. 1H chemical shifts (δ) are expressed in ppm,
and coupling constants (J) in hertz (Hz). Merck Silica gel 60 F254
plates were used for analytical TLC; flash column
chromatography was performed on Merck Silica gel (200–400
mesh).
Conclusion
We successfully characterized and applied a promiscuous
halophilic ADH (HeADH-II) that can oxidize both alcohols and
aldehydes to the corresponding carboxylic acid. Based on
observations made between the crystal structure of HeADH-II, it
shares structural characteristics with mammalian ADHs that also
carry out oxidative reactions. Such structural similarities and
common functional activities may imply a shared catalytic
mechanism, although further studies are required. This new
enzyme demonstrated great activity and stability at high salt and
polar solvent concentrations, usually employed for solubilizing
organic compounds in water media. Enzymes derived from
extremophiles, such as the halo-adapted bacterium Halomonas
elongata, are particularly suitable for industrial processes thanks
to their adaptation to harsher reaction environments. HeADH-II
showed broad substrate specificity: different aromatic primary
alcohols and aldehydes were fully converted to the final carboxylic
acid. Moreover, the oxidation of racemic 1-phenylethanol to the
corresponding ketone was obtained with good conversion, thus
demonstrating enzyme versatility. Interestingly, with this substrate,
no enantiopreference was observed; the lack of enantioselectivity
was confirmed also when the enzyme was used for the reduction
of acetophenone. This uncommon characteristic could be
explained by the symmetry observed in the catalytic pockets of
HeADH-II which allows both (R)- and (S)-1-phenyl ethanol to fit
as a mirror images in a catalytically plausible conformation. While
the substrates tested in this work are limited, considering that no
conversion was obtained with (R,S)-1-phenyl-2-propanol,
HeADH-II could be included with only a few other examples in the
list of non-stereoselective ADHs for the tested substrates which
Chemicals
All reagents and solvents were obtained from commercial
suppliers and were used without further purification.
Cloning, expression and purification
The synthetic gene encoding for HeADH-II from Halomonas
elongata, directly cloned into vector pET-28b(+) was purchased
from BaseClear. The gene, optimized for the expression in E. coli,
presented an N-terminal his-tag. The resulting pET28b-HeADH-II
was transformed into chemically competent E. coli BL21 (DE3)
pLYsS for heterologous expression. Plates of E. coli BL21 (DE3)
pLYsS pET28b-HeADH-II were grown overnight at 37 °C in LB-
agar medium supplemented with 25 μg/mL kanamycin and 35
μg/mL chloramphenicol. 200 mL of autoinduction medium: 10 g/L
N-Z amine, 5 g/L yeast extract, 50 mL KH2PO4 1 M, 50 mL
K2HPO4 1 M, 25 mL (NH4)2SO4 1 M, 2 mL MgSO4 1 M, 2 mL trace
element solution (FeSO4 50 mM, CaCl2 20 mM, MnCl2 10 mM,
ZnSO4 10 Mm, CoCl2 2 mM, CuCl2 2 mM, NiCl2 2 mM, HCl 60 mM,
NaMoO4 2 mM, H3BO3 2 mM), 20 mL 5052 solution (250 g/L
glycerol, 25 g/L glucose, 100 g/L α-lactose), 25 μg/mL kanamycin,
35 μg/mL chloramphenicol) were inoculated with a single colony.
The cultures were further incubated for 16 h at 37 °C, 120 rpm.
Cells were then harvested by centrifugation (15 min, 5000 rpm,
4 °C), washed once with deionized water and stored at -20 °C; 10
g of pellet were resuspended in 50 mL loading buffer (Tris-HCl
5
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001112
ChemCatChem
FULL PAPER
100 mM pH 8.0, imidazole 20 mM, NaCl 100 mM) and sonicated
(6 cycles of 2 min each, in ice, with 1 min interval). Cell debris
were harvested by centrifugation (45 min, 15000 rpm, 4 °C).
Chromatography was performed using ÄKTA Purifier (GE
Healthcare) with a 1 mL column HisTrapTM FF (GE Healthcare)
pre-loaded with NiSO4 (100 mM). Briefly, the column was
equilibrated with loading buffer and the filtered crude extract
loaded; column was then washed with loading buffer; finally, the
adsorbed enzyme was eluted with elution buffer (Tris-HCl 100 mM
pH 8.0, imidazole 250 mM, NaCl 100 mM).
8.0. To determining conversion of the enzymatic reactions 100 µL
aliquots were quenched with HCl 0.1 M at different reaction times
and extracted with 100 µL of AcOEt for TLC (n-Hexane/AcOEt
9:1). Samples, after evaporation, were resuspended in the mobile
phase for HPLC analysis (See Supporting Information).
One-pot reaction for the preparation of aldoximes
One-pot biotransformations were carried out at 20-mg scale using
20 mM alcohol as substrate in DMSO (15% v/v), NH2OH.HCl (24
mM), 0.1 mM NAD+, 0.1 mM FAD, HeADH-II (1 U/mL), LpNOX
(10 U/mL), in Tris-HCl buffer 100 mM, pH 8.0 at 30 °C. 150 µL
aliquots were extracted with AcOEt at different reaction times and
analyzed by gas-chromatography (See Supporting Information).
Crude extract, pellet and pure protein were analyzed by SDS-
PAGE (See Supporting information, Figure S4). The fractions
showing the presence of a band of the expected size (38 kDa)
were pooled, dialyzed against Tris-HCl buffer 100 mM, pH 8.0 and
stored at 4 °C. Typically, starting from 10 g of wet cell paste, it
was possible to obtain 5 mg of pure protein.
Acknowledgements
The plasmid pET28a(+) containing the gene encoding for LpNOX
was kindly provided by Prof. Sieber (University of München). The
enzyme was expressed and purified as previously reported.[44]
This project was supported by the European Union’s Horizon
2020 research and innovation programme under the Marie
Skłodowska-Curie grant agreement no. 792804 AROMAs-FLOW
(M. L. C.). L. J. G. was supported by Linea 2 Università degli Studi
di Milano funding.
Crude extract, pellet and pure protein were analyzed by SDS-
PAGE (See Supporting information, Figure S5). The fractions
showing the presence of a band of the expected size (52 kDa)
were pooled, dialyzed against Tris-HCl buffer 100 mM, pH 8.0 and
stored at 4 °C. Typically, starting from 2 g of wet cell paste, it was
possible to obtain 10 mg of pure protein.
Keywords: Alcohol dehydrogenase • Halomonas elongata •
Overoxidation • Biocatalysis • Aldoxymes synthesis
HeADH-II activity assays
Bibliography
Activity measurements were performed spectrophotometrically at
340 nm by determining the formation or depletion of NADH at
25 °C in a half-microcuvette (total volume 1 mL) for 2 min. One
unit (U) of activity is defined as the amount of enzyme which
catalyzes the consumption of 1 μmol of NAD+/NADH per minute
under reference conditions, namely 1 mM NAD+/NADH, 4 mM
cinnamyl alcohol/cinnamaldehyde, correct amount of HeADH-II in
Tris-HCl 100 mM, pH 8.0. Specific activity was 3 U/mg and 2.8
U/mg, respectively.
[1] D. Monti, G. Ottolina, G. Carrea, S. Riva, Chem. Rev. 2011,
111, 4111–4140.
[2] Y. G. Zheng, H. H. Yin, D. F. Yu, X. Chen, X. L. Tang, X. J.
Zhang, Y. P. Xue, Y. J. Wang, Z. Q. Liu, Appl. Microbiol.
Biotechnol. 2017, 101, 987–1001.
[3] M. L. Contente, F. Paradisi, Nat. Catal. 2018, 1, 452-459.
[4] J. An, Y. Nie, Y. Xu, Crit. Rev. Biotechnol. 2019, 366–379.
[5] J. Liu, S. Wu, Z. Li, Curr. Opin. Chem. Biol. 2018, 43, 77–
86.
LpNOX activity assay
Activity measurements were performed spectrophotometrically at
340 nm by determining the depletion of NADH at 25 °C in a half-
microcuvette (total volume 1 mL) for 2 min. One unit (U) of activity
is defined as the amount of enzyme which catalyzes the
consumption of 1 μmol of NADH per minute under reference
conditions, namely 1 mM NADH, correct amount of LpNOX in
Tris-HCl, 100 mM, pH 8.0. Specific activity was 50 U/mg.
[6] L. P. Olson, J. Luo, O. Almarsson, T. C. Bruice,
Biochemistry 1996, 35, 9782–9791.
[7] a) C. Wuensch, H. Lechner, S. M. Glueck, K. Zangger, M.
Hall, K. Faber, ChemCatChem 2013, 5, 1744–1748. b) E.
Tassano, K. Faber, M. Hall, Adv. Synth. Catal. 2018, 360, 2742 –
2751.
[8] J. Dong, E. Fernandez-Fueyo, F. Hollmann, C. E. Paul, M.
Pesic, S. Schmidt, Y. Wang, S. Younes, W. Zhang., Angew.
Chem. Int. Ed. 2018, 57, 9238–9261.
Biotransformations
Molar conversion and enantioselectivity towards different
aromatic alcohols and aldehydes were determined by performing
biotransformations at 20-mg scale, using an enzyme-coupled
system (LpNOX) for cofactor recycling. Oxidations were carried
out in 10-mL screw-capped test tubes with HeADH-II (1 U/mL),
LpNOX (10 U/mL), NAD+ (0.1 mM), FAD (0.1 mM), substrate (20
mM), DMSO (20% v/v) suspended in Tris-HCl buffer 100 mM pH
[9] P. Könst, H. Merkens, S. Kara, S. Kochius, A. Vogel, R.
Zuhse, D. Holtmann, I. W. C. E. Arends, Hollmann F., Angew.
Chem. Int. Ed. 2012, 51, 9914–9917.
6
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001112
ChemCatChem
FULL PAPER
[10] D. Romano, R. Villa, F. Molinari, ChemCatChem 2012, 4,
739–749.
[32] B. Webb B, A. Sali, Curr. Protoc. Bioinformatics 2016, 54,
5.6.1-5.6.37.
[11] D. Romano, M. Contente, T. Granato, W. Remelli, P.
Zambelli, F. Molinari, Monatsh. Chem. 2013, 14, 735–737.
[12] F. Molinari, Curr. Org. Chem. 2006, 10, 1247–1263.
[13] M. Pickl, M. Fuchs, S. M. Glueck, K. Faber, Appl Microbiol
Biotechnol 2015, 99, 6617–6642.
[33] Z. N. Yang, W. F. Bosron, T. D. Hurley, J. Mol. Biol. 1997,
265, 330–343.
[34] L. P. Olson, J. Luo, Ö. Almarsson, T. C. Bruice,
Biochemistry 1996, 35, 9782–9791.
[35] F. M. Dickinson, K. Dalziel, Nature 1967, 214, 31–33.
[36] I. Lavandera, A. Kern, V. Resch, B. Ferreira-Silva, A.
Glieder, W. M. F. Fabian, S. de Wildeman, W. Kroutil, Org. Lett.
2008, 10, 2155–2158.
[14] A. Trincone, Mar. Drugs 2011, 9, 478–499.
[15] D. Alsafadi, S. Alsalman, F. Paradisi, Org. Biomol. Chem.
2017, 15, 9169–9175.
[16] S. W. Grötzinger, R. Karan, E. Strillinger, S. Bader, A.
Frank, I. S. Al Rowaihi, A. Akal, W. Wackerow, J. A. Archer, M.
Rueping, D. Weuster-Botz, M. Groll, J. Eppinger, S. T. Arold, ACS
Chem. Biol. 2018, 13, 161−170.
[37] F. Chen, Y. Liu, G. Zheng, J. Xu, ChemCatChem 2015, 7,
3838-3841.
[38] M. A. Maria-Solano, A. Romero-Rivera, S. Osuna, Org.
Biomol. Chem. 2017, 15, 4122-4129.
[17] A. L. Akal, R. Karan, A. Hohl, I. Alam, M. Vogler, S. W.
Grötzinger, J. Eppinger, M. Rueping, FEBS Open Bio 2019, 9,
194–205.
[39] K. Kim, B. V. Plapp, Chem-Biol Interact. 2017, 276, 77–87.
[40] J. M. Patel, M. M. Musa, L. Rodriguez, D. A. Sutton, V. V.
Popik, R. S. Phillips, Org. Biomol. Chem. 2014, 12, 5905–5910.
[41] S. Wang, Y. Nie, Y. Xu, R. Zhang, T. P. Ko, C. H. Huang, H.
C. Chan, R. T. Guo, R. Xiao, Chem. Commun. 2014, 50, 7770–
7772.
[18] L. Cerioli, M. Planchestainer, J. Cassidy, D. Tessaro, F.
Paradisi, J. Mol. Catal. B: Enzym. 2015, 120, 141–150.
[19] M. Planchestainer, M. L. Contente, J. Cassidy, F. Molinari,
L. Tamborini, F. Paradisi, Green Chem. 2017, 9, 372–375.
[20] D. Roura-Padrosa, V. De Vitis, M. L. Contente, F. Molinari,
F. Paradisi, Catalysts 2019, 9, 232-242.
[42] Y. Korkhin, A. J. Kalb, M. Peretz, O. Bogin, Y. Burstein, F.
Frolow J. Mol. Biol., 1998, 278, 967–981
[43] F. G. Mutti, T. Knaus, N. S. Scrutton, M. Breuer, N. J.
Turner, Science 2015, 349, 1525–1529.
[21] M. L. Contente, F. Dall’Oglio, L. Tamborini, F. Molinari, F.
Paradisi, ChemCatChem 2017, 9, 3843-3848.
[44] C. Nowak, B. Beer, A. Pick, T. Roth, P. Lommes, V. Sieber,
Front. Microbiol. 2015, 6, 957-965.
[22] S. N. Ashakirin, M. Tripathy, U. K. Patil, A. B. A. Majeed, Int
J. Pharm. Sci. 2017, 8, 2333-2340.
[23] A. A. Koesoema, D. M. Standley, T. Senda, T. Matsuda,
Appl. Microbiol. Biotechnol. 2020, 104, 2897–2909
[24] J. D. Zhang, Z. M. Cui, X. J. Fan, H. L. Wu, H. H. Chang,
Bioprocess Biosyst. Eng. 2016, 39, 603-611.
[25] P. Zambelli, A. Pinto, D. Romano, E. Crotti, P. Conti, L.
Tamborini, R. Villa, F. Molinari, Green Chem. 2012, 14, 2158-
2161.
[26] H. Kanno, R. J. K. Taylor, Synlett 2002, 1287-1290.
[27] A. R. Klasat, F. Kazemi, K. Nourbakhsh Phosphorus, Sulfur,
Silicon Relat. Elem., 2004, 179, 1809-1812.
[28] P. Bandgar, S. M. Nikat, P. P. Wadgaonkar Synth Commun.
1995, 25, 863-869.
[29] B. A. Mendelsohn, S. Lee, S. Kim, F. Teysseier, V. S.
Aulakh, M. A. Ciufolini, Org. Lett. 2009,11, 1539-1542.
[30] Y. T. Li, B. S. Liao, H. P. Chen, S. T. Liu, Synthesis 2011,
2639-2643.
[31] R. K. Saini, S. Sravan Kumar, S. Priyanka, K. Kamireddy,
P. Giridhar in Biotechnological Production of Natural Ingredients
for Food Industry, 1st ed, Bentham Science Publisher Sharjah,
2016, pp 21-59.
7
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001112
ChemCatChem
FULL PAPER
Entry for the Table of Contents
Insert graphic for Table of Contents here.
HeADH-II
An unusual alcohol dehydrogenase has been identified from the halo-adapted bacterium H. elongata. HeADH-II shows a great stability
to polar solvents and high salt concentration as well as uncommon characteristics such as the lack of enantiopreference and the
capability of overoxidizing alcohols to carboxylic acids. To increase the process yields and allowing cofactor recycling, HeADH-II was
coupled with a NADH-oxidase from L. pentosus (LpNOX).
researcher Twitter usernames: @ParadisiResLab
8
This article is protected by copyright. All rights reserved.