One-Pot Cascade Synthesis of (3S)-Hydroxyketones Catalyzed by Transketolase via Hydroxypyruvate Generated in Situ from d-Serine by d-Amino Acid Oxidase

Article abstract of DOI:10.1002/adsc.201900109

We described an efficient in situ generation of hydroxypyruvate from d-serine catalyzed by a d-amino acid oxidase from Rhodotorula gracilis. This strategy revealed an interesting alternative to the conventional chemical synthesis of hydroxypyruvate starting from toxic bromopyruvate or to the enzymatic transamination from l-serine requiring an additional substrate as amino acceptor. Hydroxypyruvate thus produced was used as donor substrate of transketolases from Escherichia coli or from Geobacillus stearothermophilus catalyzing the stereoselective formation of a carbon?carbon bond. The enzymatic cascade reaction was performed in one-pot in the presence of d-serine and appropriate aldehydes for the synthesis of valuable (3S)-hydroxyketones, which were obtained with high enantio- and diastereoselectivity and in good yield. The efficiency of the process was based on the irreversibility of both reactions allowing complete conversion of d-serine and aldehydes. (Figure presented.).

Full text of DOI:10.1002/adsc.201900109

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Title: One-Pot Cascade Synthesis of (3S)-Hydroxyketones Catalyzed
by Transketolase via Hydroxypyruvate Generated in situ from D-
Serine by D-Amino Acid Oxidase
Authors: Laurence Hecquet, Franck Charmantray, Loredano
Pollegioni, Wolf-Dieter Fessner, Nazim OCAL, Marion
Lorilliere, Felipe Bruna, and Mélanie L'enfant
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: Adv. Synth. Catal. 10.1002/adsc.201900109
Link to VoR: http://dx.doi.org/10.1002/adsc.201900109

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
10.1002/adsc.201900109
Advanced Synthesis & Catalysis
FULL PAPER
DOI: 10.1002/adsc.201((will be filled in by the editorial staff))
One-Pot Cascade Synthesis of (3S)-Hydroxyketones Catalyzed
by Transketolase via Hydroxypyruvate Generated in situ from
D-Serine by D-Amino Acid Oxidase
Mélanie L’enfant,a Felipe Bruna,a Marion Lorillière,a Nazim Ocal,a Wolf-Dieter
Fessner,b Loredano Pollegioni,c Franck. Charmantray,a* Laurence Hecqueta*
a
Université Clermont Auvergne, CNRS, SIGMA Clermont, Institut de Chimie de Clermont-Ferrand (ICCF) F-63000
Clermont-Ferrand (France) e-mail: laurence.hecquet@uca.fr, franck.charmantray@uca.fr
Institut für Organische Chemie und Biochemie, Technische Universität Darmstadt, 64287 Darmstadt (Germany)
Department of Biotechnology and Life Sciences, Università degli Studi dell’Insubria, Varese (Italy)
b
c
Received: ((will be filled in by the editorial staff))
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201######.((Please
delete if not appropriate))
Abstract. We described an efficient in situ generation of of a carbon-carbon bond. The enzymatic cascade reaction
hydroxypyruvate from D-serine catalyzed by a D-amino acid was performed in one-pot in the presence of D-serine and
oxidase from Rhodotorula gracilis. This strategy revealed an appropriate aldehydes for the synthesis of valuable (3S)-
interesting alternative to the conventional chemical synthesis hydroxyketones, which were obtained with high enantio-
of hydroxypyruvate starting from toxic bromopyruvate or to and diastereoselectivity and in good yield. The efficiency of
the enzymatic transamination from L-serine requiring an the process was based on the irreversibility of both
additional substrate as amino acceptor. Hydroxypyruvate reactions allowing complete conversion of D-serine and
thus produced was used as donor substrate of transketolases aldehydes.
from
Escherichia
coli
or
from
Geobacillus
stearothermophilus catalyzing the stereoselective formation
Keywords: biocatalysis; transketolase; D-amino acid
oxidase; hydroxypyruvate; ketoses; stereoselectivity
catalyzed decarboxylation of HPA is a powerful tool
for such an approach. Indeed, TK, a thiamine
dependent enzyme, catalyzes the subsequent transfer
of the ketol group of HPA to an aldehyde acceptor
leading to a (3S)-ketose by the stereoselective
formation of the C2-C3 bond.[1] The coupling of the
TA and the TK reactions in a one pot has been
developed producing the expected ketose products.[4-
Introduction
Hydroxypyruvate (HPA) is a key substrate for several
HPA-dependent enzymes including aldolases[1] and
transketolases (TK)[2] used in biocatalysis for the
synthesis of valuable monosaccharides and analogues
by stereoselective carbon-carbon bond formation. For
synthetic purposes conducted at gram-scale, the main
problem is the cost of the commercial HPA. This
compound can also be synthetized by hydrolysis of
toxic bromopyruvic acid in the presence lithium
hydroxide.[3] Enzymatic in situ generation of HPA
has been previously described, in particular from L-
serine using transaminases (TA) (Scheme 1A).[4-6]
The reversibility of the TA-reaction requires the
simultaneous coupling to an irreversible enzyme-
catalyzed reaction to shift the overall equilibrium
toward HPA formation and to avoid its accumulation
and degradation, particularly at high temperature,
and/or in buffered solution.[5] The irreversible TK-
6]
The major drawback of this approach is the
requirement of an-keto acid as an amino group
acceptor, which is a second substrate generating the
corresponding amino acid as an additional by-product
if not recycled. In addition, the equilibrium shift
toward the formation of the product needs an excess
of one of the two TA substrates, i.e., D-serine or -
keto acid, and in the latter case inhibition of TA was
observed.[5,6]
To circumvent all these limitations, another
strategy consists in using an aminoacid oxidase
(AAO) to catalyze the oxidative deamination of
serine into HPA (Scheme 1B). Substrate specificity
1
This article is protected by copyright. All rights reserved.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
10.1002/adsc.201900109
Advanced Synthesis & Catalysis
studies of AAOs have shown that the best substrates
for LAAOs are hydrophobic aminoacids[7] while
a wide range of aldehydes leading to the synthesis of
various (3S)-hydroxyketones. In this study, we have
selected the mesophilic TK from Escherichia coli
(TKeco)[18d-h] and the thermostable TK from
Geobacillus stearothermophilus (TKgst)[19a,20] to be
coupled with DAAORg in a one-pot two step
procedure, starting only from the two substrates D-
serine and an appropriated aldehyde acceptor for
obtaining valuable (3S)-hydroxyketones.
DAAOs display
a
broad aminoacid spectrum
including D-serine. Hence, we have selected the latter
enzyme family for the approach developed in this
study.
Results and Discussion
Properties of DAAO from R. gracilis
Scheme 1: In situ generation of HPA from TA or
DAAO-catalyzed reaction coupled with TK-catalyzed
reaction.
DAAORg was overexpressed in E. coli and then
extracted and purified by Ni2+ chelating affinity
column chromatography as reported earlier.[10] In
order to use DAAORg under the best conditions for
the synthesis of HPA in situ from D-serine, we first
determined at analytical scale the stability of the
recombinant flavoenzyme over time at different
temperatures and its kinetic parameters using D-serine
as substrate in the presence of oxygen. During
turnover the tightly bound FAD cofactor is reduced
by the D-amino acid to FADH2 and then reoxidized
by O2 generating hydrogen peroxide. Since the
accumulation of hydrogen peroxide can lead to an
enzyme inhibition, addition of catalase is required for
the dismutation of hydrogen peroxide to water and
oxygen (Scheme 2).
DAAOs (EC 1.4.3.3) are flavin adenine
dinucleotide (FAD)-containing flavoenzymes that
catalyze the deamination of D-amino acids to their
imino acid counterparts with concomitant reduction
of FAD which is subsequently re-oxidized by
molecular oxygen with generation of hydrogen
peroxide. The released imino acid spontaneously
hydrolyses to the corresponding -keto acid and
ammonia (Scheme 2). DAAO activity has been
identified in many organisms, ranging from fungi to
humans.[8] The well-known pig kidney DAAO has
been first reported for use in biocatalytic
applications,[9-11] but the enzyme from the yeast
Rhodotorula gracilis (DAAORg) showed much higher
turnover number, better stability under a wide range
of reaction conditions, and a larger active site to
accommodate various substrates.[12,13] Due to these
properties, the yeast enzyme has been used in a
variety of biotechnological processes at industrial
scale such as the production of glutaryl-7-
aminocephalosporanic acid from cephalosporin C,[14]
the resolution of racemic mixtures of amino acids,[15]
and the production of α-keto acids from natural and
synthetic D-amino acids.[16] DAAORg accepts the polar
amino acid D-serine with higher specific activity
compared to DAAO from pig kidney.[9] However, this
enzyme has never been investigated to produce the α-
keto acid HPA from D-serine, especially at
preparative scale.
The aim of this work is to study the reaction
catalyzed by DAAORg in the presence of D-serine as
the substrate to evaluate the in situ synthesis of HPA
at preparative scale. It should be noted that D-serine is
commercially available at approximatively the same
price as L-serine and can be produced at industrial
scale by a microbial process.[17] To validate the
strategy, the DAAORg-catalyzed reaction will be first
studied independently and then coupled with TK,
requiring and consuming HPA as donor substrate.
The irreversibility of TK-catalyzed reaction and the
stereoselectivity of this enzyme are major advantages.
On this side, wild-type or evolved TKs from
mesophilic[18] or thermophilic[5,6b,19] sources have been
reported as efficient biocatalysts for the conversion of
Scheme 2. Reaction of DAAORg coupled with L-
lactate dehydrogenase (LDH) in the presence of
NADH.
DAAORg activity was determined by a coupled
reaction with L-lactate-dehydrogenase (L-LDH)
allowing the reduction of HPA in the presence of
NADH (Scheme 2). The disappearance of NADH
was followed over time by spectrophotometry at 340
nm (LDH assay).
An optimum DAAORg activity was previously
described in a 37-45°C temperature range and a 8.0-
8.5 pH range.[10] Under those conditions, we observed
that DAAORg fully maintained its activity during 25
hours at 25°C and at pH 7.5 while at higher
temperature the enzymatic activity decreases
drastically (Figure 1).
2
This article is protected by copyright. All rights reserved.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
10.1002/adsc.201900109
Advanced Synthesis & Catalysis
serine completely into HPA and making this strategy
an efficient synthetic way for obtaining HPA at
preparative scale.
120
100
80
60
40
20
0
Conversion of D-serine into HPA catalyzed by
DAAORg
For HPA synthesis from D-serine at preparative scale,
catalase was added dropwise to the reaction mixture,
because addition of a specific quantity of catalase as a
single dose at the beginning of the reaction was found
to be insufficient to ensure the total dismutation of
hydrogen peroxide during the entire reaction period,
thus causing a decrease of DAAO activity. Oxygen
was bubbled into the reactor at a rate of 10 mL . min-
0
5
10
Time (h)
15
20
25
Figure 1: DAAORg thermostability. DAAORg activity
was determined by the LDH assay in the presence of
D-serine (50 mM) and O2 at ♦ 25°C, ▲ 30°C, ● 40°C,
■ 60°C. 100 % corresponds to initial activity (37.5 U
. mg-1 at 25°C and pH 7.5).
1
at atmospheric pressure. The reaction was followed
by quantification of HPA and D-serine over time by
spectrophotometry in aliquots from the reaction
mixture. After the completion of the reaction, HPA
and D-serine quantification was performed by in situ
1H NMR analysis of the reaction mixture.
The DAAORG kinetic parameters were
determined using the LDH assay described above.
The kinetic parameters toward D-serine were
approximatively the same than those reported in the
literature using a polarographic assay method.[11] This
latter assay has been used for the determination of
kinetic parameters of other amino acids (Table 1).
D-Serine at 50 mM led to almost complete
conversion (90%) in 24 hours while HPA
quantification corresponded to 78% of initial D-serine
concentration (Figure 2). This result may be
explained by a slight HPA degradation over time as
previously observed.[19] In addition, we noted that
DAAORg activity decreased during the reaction (SI)
due to an inhibition by HPA already described with
2-oxo-4-methylthiobutyric acid, the keto acid
produced from D-methionine[21a] and other -keto
acids analogs of amino acids.[21b] Indeed, conversion
of D-serine into HPA was 66 and 90% at 250 mM
and 50 mM, respectively, clearly showing an
inhibition of DAAORg by the ketoacid. To avoid HPA
accumulation and to limit DAAORg inhibition, the
DAAORg-catalyzed reaction was therefore coupled
with an irreversible reaction that consumes HPA as
substrate such as a TK-catalyzed reaction.
Table 1. Apparent kinetic parameters of DAAORg
toward different D-amino acids (at air saturation).
D-amino acids
KM
Vmax
U.mg-1
115
Vmax/KM
U.mg1.mM-1
639
kcat/KM
mM-1.s-1
416
mM
D-methionine[10]
D-phenylalanine[10] 0.29
D-leucine[10]
D-alanine[10]
D-serine
0.18
144
497
323
0.50
120
240
156
0.83
140
169
110
13.7[10]
14.0±1 a)
18.9
61[10]
48±7 a)
195
4.4[10]
3.5±0.8 a)
10.3
3.0
2.3±0.5
6.7
D-valine[10]
D-cysteine[10]
21.4
130
6.1
4.0
Hydroxyketone synthesis from D-serine and
various aldehydes by DAAORG /TKgst coupling
a) Measurements with D-serine were determined using
LDH assay by varying the substrate concentration at
25°C, pH 7.5.
The one-pot, two-step cascade reaction catalyzed by
DAAORg and TK was performed using two substrates:
D-serine and an appropriated aldehyde as TK acceptor
substrate. To validate the strategy, various aldehydes
were chosen, such as non-hydroxylated (1, 2) and
hydroxylated (3-7) aldehydes (Table 2). For better
efficiency, TKs from different sources were used
depending on the aldehyde. Previous substrate
specificity studies showed that for wild-type TKs best
DAAORg offers more substrate flexibility
compared to other DAAOs.[8,11] Indeed, based on the
KM values, the best affinities were apparent for D-
methionine, D-phenylalanine, D-leucine, and D-
alanine. Further amino acids are also accepted with
lower affinities such as D-serine, D-valine, and D-
cysteine. The Vmax observed for D-serine is
approximately half of the value obtained with the
reference substrate D-alanine. Even if D-serine is not
the best substrate for DAAORg, the KM and Vmax
values are more suitable for biocatalytic applications
compared to those obtained toward L-serine
converted into HPA by TA-catalyzed reaction (KM
substrates
are
(2R)-configured
hydroxylated
aldehydes with hydroxylated short carbon chain
22.4 mM; Vmax 3.3 U mg-1; Vmax/KM 0.14 U mg-1 .
mM).5 In addition, the DAAORg catalyzed reaction is
irreversible offering the possibility to convert D-
.
.
3
This article is protected by copyright. All rights reserved.