Convergent in situ Generation of Both Transketolase Substrates via Transaminase and Aldolase Reactions for Sequential One-Pot, Three-Step Cascade Synthesis of Ketoses

Article abstract of DOI:10.1002/cctc.201901756

We describe an efficient three-enzyme, sequential one-pot cascade reaction where both transketolase substrates are generated in situ in a convergent fashion. The nucleophilic donor substrate hydroxypyruvate was obtained from l-serine and pyruvate by a transaminase-catalyzed reaction. In parallel, three different (2S)-α-hydroxylated aldehydes, l-glyceraldehyde, d-threose, and l-erythrose, were generated as electrophilic acceptors from simple achiral compounds glycolaldehyde and formaldehyde by d-fructose-6-phosphate aldolase catalysis. The compatibility of the three enzymes was studied in terms of temperature, enzyme ratio and substrate concentration. The efficiency of the process relied on the irreversibility of the transketolase reaction, driving a shift of the reversible transamination reaction and securing the complete conversion of all substrates. Three valuable (3S,4S)-ketoses, l-ribulose, d-tagatose, and l-psicose were obtained in good yields with high diastereoselectivity.

Full text of DOI:10.1002/cctc.201901756

Accepted Article
Title: Convergent in situ Generation of Both Transketolase Substrates
via Transaminase and Aldolase Reactions for Sequential One-
Pot, Three-Step Cascade Synthesis of Ketoses
Authors: Laurence Hecquet, Franck Charmantray, Pere Clapés, Wolf-
Dieter Fessner, Thierry Gefflaut, Christine Guérard Hélaine,
and Marion Lorillière
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.201901756
Link to VoR: http://dx.doi.org/10.1002/cctc.201901756
A Journal of
www.chemcatchem.org

10.1002/cctc.201901756
ChemCatChem
FULL PAPER
Convergent in situ Generation of Both Transketolase Substrates
via Transaminase and Aldolase Reactions for Sequential One-Pot,
Three-Step Cascade Synthesis of Ketoses
Marion Lorillière,[a] Christine Guérard-Hélaine,[a] Thierry Gefflaut,[a] Wolf-Dieter Fessner,[c] Pere
Clapés,[b] Franck Charmantray,*[a] Laurence Hecquet*[a]
Abstract: We describe an efficient three-enzyme, sequential one-
pot cascade reaction where both transketolase substrates are
generated in situ in a convergent fashion. The nucleophilic donor
substrate hydroxypyruvate was obtained from L-serine and pyruvate
by a transaminase-catalyzed reaction. In parallel, three different
(2S)-α-hydroxylated aldehydes, L-glyceraldehyde, D-threose, and L-
erythrose, were generated as electrophilic acceptors from simple
achiral compounds glycolaldehyde and formaldehyde by D-fructose-
6-phosphate aldolase catalysis. The compatibility of the three
enzymes was studied in terms of temperature, enzyme ratio and
substrate concentration. The efficiency of the process relied on the
irreversibility of the transketolase reaction, driving a shift of the
reversible transamination reaction and securing the complete
conversion of all substrates. Three valuable (3S,4S)-ketoses,
L-ribulose, D-tagatose, and L-psicose were obtained in good yields
with high diastereoselectivities.
acceptor, to produce a corresponding Cn+2 ketose (Scheme 1).
In the pentose phosphate metabolism, TK catalyzes a ketol
transfer reaction for the reversible equilibration of ketoses and
aldoses carrying a terminal phosphate ester group. Previous in
vitro studies showed that non-phosphorylated aldose
compounds can also be used as TK substrates, yielding the
equivalent free ketoses in one step. In particular, irreversible
release of carbon dioxide from hydroxypyruvate (HPA) as donor
kinetically drives the conversion, rendering TK a powerful tool for
the asymmetric synthesis of ketoses and related acyloin
compounds.7,8
Introduction
Scheme 1. Irreversible reaction catalyzed by TK using hydroxypyruvate (HPA)
as donor substrate with (2R)-hydroxyaldehydes as acceptor substrates.
The need for greener and more sustainable chemical
production1 has seen enzymatic strategies to emerge as a
powerful approach to the eco-friendly, highly selective synthesis
of valuable chiral compounds. To compete with the productivity
of traditional methods, the use of two or even more enzymes in
cascade can considerably improve the efficiency of a multistage
synthesis by obviating the isolation of intermediates, thus saving
time, resources, reagents and energy, while reducing waste.2−4
Cascade reactions can be performed along a simultaneous one-
pot strategy when all the enzyme requirements are compatible.
To meet limitations, such as substrate/product/reagent inhibition
or incompatibility of reaction conditions (pH, temperature), a
telescoped, sequential one-pot procedure can be used.2
Chiral polyols, regarded as highly valuable compounds in
various fields, can be obtained by stereoselective carbon-carbon
bond formation.5,6 Especially, transketolase (TK, EC 2.2.1.1) is a
powerful thiamine diphosphate-dependent biocatalyst that allows
a two-carbon chain elongation by transferring an α-hydroxy
carbonyl (ketol) group from a ketone donor to an aldehyde
The studies on TK acceptor substrate specificity revealed that
non-phosphorylated α-hydroxylated aldehydes with short carbon
chains (C2-C3) gave the highest TK activities. The new
asymmetric center at C3 of the Cn+2 ketose product is formed
with high stereoselectivity for an (S) configuration. In addition,
the enzyme is highly enantioselective with chiral α-hydroxy
aldehydes, and converts the (2R)-epimers with high preference,
thus leading to (3S, 4R)-configured ketoses. For biocatalytic
applications, mesophilic TKs from Saccharomyces cerevisiae
and from Escherichia coli have been largely used and optimized
by mutagenesis.9,10 We have identified the first thermostable TK
from Geobacillus stearothermophilus (TKgst
significantly improved stability at high temperature,11 more
robustness towards non-usual reaction conditions12 and
broadened substrate spectrum obtained by in vitro evolution
)
that offers
a
towards
(2S)-hydroxylated,13 aliphatic14
and
aromatic
aldehydes15 and also towards HPA analogs as new donor
substrates.16
The purpose of this study was to optimize the synthesis of
highly
valuable
natural
(3S,4S)-ketoses
from
(2S)-
[a]
Prof. Dr. L. Hecquet,* Dr. F. Charmantray,* Dr, M. Lorillière, Dr, C.
Guérard-Hélaine, Prof. Dr. T. Gefflaut
Université Clermont Auvergne, CNRS, SIGMA Clermont,
Institut de Chimie de Clermont-Ferrand (ICCF)
F63000 Clermont-Ferrand, France
hydroxyaldehydes. While the synthesis of such compounds is
inaccessible using mesophilic TKs, we could recently show that
TKgst catalysis enabled best yields upon reaction at 60°C in a
reasonable time frame (8-24 h).11,13 To make this TKgst
-
catalyzed process greener and more sustainable, a major
challenge was to generate the donor HPA as well as the
acceptor substrates in situ in a one-pot procedure, not only to
avoid the costly purchase or chemical synthesis of these
compounds, but also to counteract their limited stability at 60°C
(Scheme 2). HPA is particularly unstable at high temperature
but can be produced in situ from L-serine (L-Ser) by
transamination to a ketoacid as amino group acceptor.17 For this
purpose, we recently introduced the thermostable L-α-
transaminase from Thermosinus carboxydivorans (TAtca), which
was identified, characterized and coupled with thermostable
E-mails: Laurence.hecquet@uca.fr, Franck.Charmantray@uca.fr
Prof. Dr. P. Clapés
Biotransformation and Bioactive Molecules Group, Instituto de
Química Avanzada de Cataluꢀa, IQAC−CSIC Jordi Girona 18-26,
08034 Barcelona, Spain
[b]
[c]
Prof. Dr. W.-D. Fessner
Institut für Organische Chemie und Biochemie, Technische
Universität Darmstadt, Alarich-Weiss-Strasse 4,
64287 Darmstadt, Germany
Supporting information for this article is given via a link at the end of the
document.(Please delete this text if not appropriate)
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901756
ChemCatChem
FULL PAPER
18
TKgst
.
The targeted (2S) hydroxyaldehydes – L-glyceraldehyde
which resulted in a 60% conversion of GA after 24 h at 25 °C.
Because the wild-type FSAeco is a thermostable enzyme (half-life
8 days, 30 h and 16 h, at 55 °C, 65 °C and 75 °C, respectively),
we tested if reactions could be promoted at higher temperatures.
Indeed, reactions carried out at 50 °C led to a complete
conversion of GA within 19 h.
1, D-threose 2, and L-erythrose 3 – can be synthesized from
cheap non-chiral starting materials such as glycolaldehyde (GA)
and formaldehyde (FA) by enzymatic aldol reactions catalyzed
by D-fructose-6-phosphate aldolase from Escherichia coli
(FSAeco; wild-type or variants).19 The compatibility of FSA and
TK-catalyzed reactions has been previously demonstrated for
the synthesis of phosphorylated sugars.20 Here, we have
investigated the best conditions (i.e., temperature, enzyme
ratios, substrate concentrations) to combine the three individual
reactions catalyzed by TA, FSA and TK into a convergent one-
pot procedure. This procedure was then applied to the synthesis
of three rare (3S, 4S) ketoses, namely L-ribulose 4,21 D-tagatose
5,22 and L-psicose 6,23 which are highly valuable compounds in
the pharmaceutical and food sectors.
D-Threose (2S, 3R) 2 or L-erythrose (2S, 3S) 3 were both
obtained by the condensation of two GA molecules with
catalysis by two different FSAeco variants, A129G or
A165G/S166P, respectively, which control an opposite
stereoselectivity at C3.19 The synthesis of 2 was fast, but
compromised by (i) the retroaldolization of 2 releasing GA, and
(ii) the condensation of 2 to GA yielding D-idose. To avoid these
undesired complications, use of GA at 200 mM with 4 mg of
A129G at 25 °C led to clean formation of D-threose 2 with 98%
conversion of GA in 24 h, without formation of D-idose as shown
by NMR analysis of the reaction mixture. To reduce the reaction
time, we also tested elevated temperatures, but this caused the
denaturation of the FSAeco variant. Aldehyde 3 was obtained
from the self-addition of GA catalyzed by FSAeco variant
A165G/S166P allowing the formation of the (2S, 3S)
configuration, which is inaccessible with WT FSAeco. The affinity
of this variant for GA being very low, about six times more
enzyme than in the previous reaction (25 mg) was required to
convert 96% of GA (200 mM) in 10 days at 25 °C.19 Again, a
higher temperature was tried to shorten reaction time but also
caused enzyme precipitation.
Results and Discussion
The convergent reaction strategy illustrated in Scheme 2 entails
the practical selectivity problems that the substrates FA and/or
GA involved in the FSA reaction for the generation of L-
glyceraldehyde 1, D-threose 2 and L-erythrose 3 are also good
substrates of TKgst, and could also be consumed by undesired
transamination from L-Ser. Therefore, the aldolase catalyzed
synthesis of TKgst acceptors 1–3 in the first stage had to be
uncoupled from the remaining steps. Then, in a telescoped
second stage, all other enzymes and substrates could be added
simultaneously for in situ generation of HPA donor and for the
final carboligation leading to L-ribulose 4, D-tagatose 5 and L-
psicose 6.
Table 1. Conditions and results of FSAeco catalyzed reactions for the synthesis
of aldehydes 1–3.
GA
mM
FA
mM
FSAeco
mg
T °C
Reaction
time
Product
GA
conversion
[d]
%
50
100
4[a]
50°C
9h
24h
100
98
200
200
-
-
4[b]
25°C
25°C
Scheme 2. One-pot cascade reaction for the synthesis of ketoses 4–6 by TKgst
with convergent in situ generation of HPA using TAtca and aldehydes 1–3
25[c]
10 days
96
using FSAeco
.
In situ generation of aldehydes 1–3 by FSAeco
[a] wild-type FSAeco
. [b] variant FSAeco A129G. [c] variant FSAeco
A165G/S166P. [d] Determined by HPLC and quantitative 1H NMR relative to
3-trimethylsilyl-2,2,3,3-tetradeuteropropionate (TSP-d4) as internal standard
24
FSAeco belongs to the class I aldolases and catalyzes in vivo
the reversible stereoselective addition of dihydroxyacetone to D-
glyceraldehyde-3-phosphate to form D-fructose-6-phosphate.
The aldolase showed a remarkable tolerance to a large panel of
nucleophilic substrates, such as non-phosphorylated ketones,
and aldehydes such as FA, acetaldehyde and GA.25 In addition,
efficient FSAeco variants could be obtained by enzyme
engineering to broaden the nucleophile scope.26 The enzymatic
syntheses of 1, 2 and 3 have been recently achieved with wild-
type and variants of FSAeco using cheap, achiral aldehydes FA
and GA.19 The original procedures had to be slightly modified for
the purpose of this work. The three FSA-catalyzed reactions
Finally, HPLC and NMR analysis validated that the three
aldehydes 1, 2 and 3 were obtained with excellent yield and with
high diastereoselectivity (>95%). Subsequently, the reaction
mixtures obtained were subjected directly without purification to
the subsequent synthesis of the corresponding ketoses 3, 4, and
5.
Synthesis of ketoses 4–6 from aldehydes 1–3 and in situ
generated HPA by TAtca coupled withTKgst
1
were monitored by HPLC and by quantitative H NMR analysis
to observe the conversion of aldehydes and the simultaneous
formation of the ketose products (ESI).
Aldehydes 1–3 present a (2S)-configuration, which is opposite to
the strongly preferred natural (2R)-substrates of WT TKgst
.
Previously, we showed that increased conversion of non-natural
(2S)-substrates could be promoted by high temperature (60 °C),
The synthesis of L-glyceraldehyde 1 from FA and GA was
19
catalyzed by wild-type-FSAeco
.
To prevent the self-
condensation of GA leading to the production of D-threose 2, the
strategy required an excess of FA (100 mM) over GA (50 mM),
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901756
ChemCatChem
FULL PAPER
Table 1. Conditions and results of one-pot cascade reactions for ketoses 4–6 synthesis from aldehydes 1–3 by coupling TAtca and TKgst
Ser
mM
Pyr
mM
Aldehyde
(mM)[a]
TAtca/TKgs
(mg/mg)
Ketose
product
Reaction
time (h)
Aldehyde
Conversion
rate %[a]
>95
Isolated
yield (%)
de
(%)[b]
150
150
150
50
1-(50)
2- (50)
3- (50)
4.8 / 6
4.8 / 10
4.8 /10
24
96
96
53
55
49
>95
>95
>95
100
100
>95
>95
[a] Reaction mixtures obtained previously with FSAeco, aldehyde concentrations were quantified by HPLC and quantitative 1H NMR. [b] Determined using
quantitative 1H NMR relative to 3-trimethylsilyl-2,2,3,3-tetradeuteropropionate (TSPd4) as internal standard [b] Diastereoisomeric excess (de) determined by 1H
and 13C NMR 13b
however, which was also resulting in HPA degradation over
time.11,13a To overcome this problem, we found an efficient
alternative to using synthetic HPA, consisting of the generation
of HPA in situ by using a thermostable transaminase from
Thermosinus carboxydivorans (TAtca). The thermostability of
both TAtca and TKgst at 60 °C and the irreversibility of the TK
reaction, allowing an equilibrium shift of the TA reaction towards
HPA, are of particular interest. Thus, TAtca with L-Ser and
pyruvate (Pyr) as substrates, pyridoxal phosphate (PLP) as
cofactor, and TKgst, with both its cofactors thiamine diphosphate
(ThDP) and MgCl2, were simultaneously added to the aldol
reaction mixture containing aldehydes 1–3, obtained with FSAeco
in the first stage. We observed that all substrates and reagents
assayed at 60 °C were found to be stable except Pyr, which
slowly decomposed with a half-life of 4 days.18
In order to study the influence of different enzyme quantities and
substrate concentrations, small scale reaction mixtures were
monitored in situ by 1H NMR analysis against 3-trimethylsilyl-
2,2,3,3-tetradeuteropropionate (TSP-d4) as an internal standard.
Thereby, we could quantify the individual conversion rates of
aldehydes and Pyr along with the formation of ketose products.
The reaction mixtures containing aldehydes 1, 2 and 3 obtained
by FSAeco catalysis were diluted to a final concentration of 50
mM. The ratio TAtca/TKgst was adjusted to avoid an accumulation
of HPA while allowing its gradual conversion into ketoses 4–6.
The best conversion rates of aldehydes were reached by using
1.5 equivalents of L-Ser in combination with 1 equivalent of Pyr.
In the case of D-threose 2 and L-erythrose 3, for which TKgst
shows lower activities, a corresponding larger quantity of TKgst
(4.8 mg TAtca/10 mg TKgst) had to be used. To avoid the known
inhibition of TAtca by Pyr at concentrations above 50 mM,18 this
substrate was added in two portions, the second after complete
disappearance of the first fraction.
(de > 95%). The absolute configurations of 4–6 were further
confirmed by comparing their optical rotations with those
reported in the literature. These characterization data confirmed
that the three TKgst-catalyzed reactions each led to a single
ketose with the expected L-erythro (3S, 4S) configuration,
consistent with the stereoselectivity observed with other
aldehyde acceptors by TKgst catalysis at elevated
temperatures.10,12,13
Conclusions
The synthesis of three (3S, 4S)-ketoses, L-ribulose 4, D-tagatose
5 and L-psicose 6, which are highly valuable compounds, was
performed from in situ generated (2S)-α-hydroxy aldehydes 1–3
and HPA using an efficient sequential one-pot, three-step
cascade sequence catalyzed by FSAeco and thermostable TAtca
and TKgst. This convergent procedure allowed circumventing the
instability of HPA and the cost of expensive aldehyde
precursors. The efficiency of this one-pot, three-step cascade
process relied on the demonstrated compatibility of the three
enzyme types. It allowed for an in situ generation of aldehydes
catalyzed by FSAeco variants from achiral, cheap aldehydes GA
and FA. Moreover, also the sensitive HPA could be generated
by TAtca from L-Ser coupled with its irreversible consumption in a
TKgst reaction, thus avoiding its accumulation and
decomposition. In addition, the thermostability of TAtca and TKgst
enabled the process to be performed at 60 °C, resulting in
increased activities towards the (2S)-configured α-
,
hydroxyaldehydes, which are poor TKgst substrates at 25 °C.
Consequently, excellent conversion rates in a reasonable time
(24–96 h) were attained. This environmentally friendly,
straightforward approach offers a useful alternative to the
conventional chemical synthesis of these compounds, and
seems applicable to a wider scope of related products, using
other wild-type or variant carboligation catalysts.
Under such adjusted conditions, we observed the total
conversion of aldehydes 1–3 and simultaneously the quantitative
formation of the corresponding ketose products. Finally,
L-ribulose 4, D-tagatose 5 and L-psicose 6 were isolated in good
yields of 53%, 55% and 49%, respectively. It is noteworthy that
the isolated yields of compounds 4–6 obtained by following the
novel cascade process were 4–5 times higher than those
obtained with synthetic Li-HPA and commercially available
substrates 1–3, which were reported as 11%,13a 13%,13b and
10%,13b respectively. The three L-erythro (3S, 4S) ketoses were
obtained in optically pure form with high diastereoselectivities
Experimental Section
General experimental. Thermosinus carboxydivorans DSM 14886 was
purchased from DSMZ, E. coli strains BL21(DE3)pLysE strain from
Invitrogen. Reagents for molecular biology were obtained from Life
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901756
ChemCatChem
FULL PAPER
Technologies. Sodium chloride was from Roth. L-Ser, Pyr, imidazole,
ammonium sulphate and potassium chloride were purchased from Alfa
Aesar. Pyridoxal 5-phosphate monohydrate (PLP) was from Acros
Organics. L-Glyceraldehyde was from Molekula.. Ni-NTA resin was from
Qiagen. The 96-well microplates were shaken and incubated in a
Titramax 1000 incubator (Heidolph). Lyophilization was carried out with a
Triad Labconco dryer. UV-visible absorbances were measured using a
Safire II-Basic plate reader from Tecan and a Perkin Elmer Lambda 25
(d, J = 1.2 Hz, 1H, H-1), 4.23 (m, 1H, H-3), 4.21 (m, 1H, H-4’), 4.03 (m,
1H, H-2), 3.94 (dd, J = 9.2, 2.7 Hz, 1H, H-4). 13C NMR (101 MHz, D2O): δ
(ppm) (α-D-threo-1,4-furanose): 103.4 (C-1), 81;9 (C-2), 76.4 (C-3), 74.3
(C-4); (β-D-threo-1,4-furanose): 97.9 (C-1), 77.4 (C-2), 76.1 (C-3), 71.8
(C-4); (hydrate linear form): 91.1 (C-1), 74.5 (C-2), 72.1 (C-3), 64.2 (C-4).
Synthesis of L-erythrose 3. GA (200 mM) was dissolved in H2O (5 mL)
and the pH was adjusted to 8 with 0.1 M NaOH. FSAeco A165G/S166P
mutant (5 mg) was added to initiate the reaction, and the mixture was
stirred at 25 C, 900 rpm. After 2, 4, 6 and 8 days, FSAeco mutant
A165G/S166P (5 mg) was added. The reaction was monitored by in situ
1H NMR and HPLC. After quasi-complete disappearance of GA (96%, 11
days), the reaction was stopped and the purity of L-erythrose 3 was
analysed by 1H NMR and HPLC.
(2S, 3S)-2,3,4-trihydroxy-butyraldehyde (L-erythrose) 3. The title
compound was obtained in solution in unbuffered water, Rf 0.32
(CH2Cl2:CH3OH, 90:10 v:v). HPLC: tr = 13,74 min. NMR data were
identical to those previously reported27,28 (ratio α-L-erythro-1,4-
furanose/β-L-erythro-1,4-furanose/ hydrate linear form = 62/27/11); lit.
ratio α-L-erythro-1,4-furanose/β-L-erythro-1,4-furanose/linear hydrate
form = 63/20/10).9 13C NMR (101 MHz, D2O) δ (ppm) (β-L-erythro-1,4-
furanose): 102.4 (C-1), 77.6 (C-2), 72.5 (C-4), 71.7 (C-3); δ (α-L-erythro-
1,4-furanose): 96.8 (C-1), 72,9 (C-4), 72,4 (C-2), 70.6 (C-3); δ (ppm)
(hydrate linear form): 90.7 (C-1), 74.8 (C-2), 73.0 (C-3), 63.8 (C-4).
spectrophotometer
enabling
Peltier-effect
temperature
control.
Macherey-Nagel GmbH & Co KG 60 F254 silica gel TLC plates and
Macherey-Nagel GmbH & Co KG 60/40-63 mesh silica gel for liquid flash
chromatography were used. NMR spectra were recorded in D2O and
CD3OD on a Bruker Avance 400 spectrometer. Optical rotation was
determined with a P-2000 JASCO PTC-262 polarimeter at the given
temperature and wavelength (Na-D-line: λD = 589 nm) in a cell 10 cm
long. Optical rotations ([α]D) values are given in dm-1.g-1.cm3. HPLC
analyses were performed on an AcclaimTM 120 C18 5 µm 120 Å (4.6 ×
250 nm) column from VWR (USA).
In situ 1H NMR measurements. Reactions were monitored using
quantitative in situ 1H NMR relative to 3-trimethylsilyl-2,2,3,3-
tetradeuteropropionate (TSP-d4) as internal standard. Aliquots of reaction
mixtures (450 µL) were mixed with 50 µL of TSP-d4 (50 mM).
HPLC measurements. HPLC analyses and derivatisation were
performed according to literature.1 Samples (30 µL) were injected and
eluted in the following conditions: solvent system (A) 0.1% (v/v) aqueous
trifluoroacetic acid (TFA) and (B) 0.1% (v/v) TFA in CH3CN/H2O (4:1),
gradient elution 10–95% B in 30 minutes, flow rate 1 mL min-1, detection
at 215 nm, column temperature 30 C. Reaction samples (10 µL) were
mixed with 50 µL of a solution of O-benzylhydroxylamine hydrochloride
(21.1 mg mL–1; 0.14 mmol mL–1) in pyridine:methanol:water (33:15:2).
Samples were then diluted in methanol (500 µL), centrifuged (5 min.,
14,500 rpm, 25 °C) and the supernatant directly analysed by HPLC.
General procedure for ketose 4–6 synthesis. ThDP (0.1 mM), MgCl2·6
H2O (1 mM), PLP (0.2 mM) and Pyr (50 mM) were dissolved in H2O and
the pH was adjusted to 7 with 0.1 M NaOH. To this stirred solution was
added TKgst (6 mg in the case of L-glyceraldehyde; 10 mg in the case of
D-threose and L-erythrose) and TAtca (4.8 mg), and the mixture was
stirred for 20 minutes at 60°C. In another flask, the reaction mixture
containing (2S) α-hydroxy aldehydes 1, 2 or 3 obtained previously with
FSAeco was diluted to obtain a concentration of 50 mM (except for L-
glyceraldehyde) and L-Ser (150 mM) were mixed and the pH adjusted to
7 with 0.1 M NaOH. After pre-incubation of enzymes, cofactors and Pyr,
L-Ser and (2S) α-hydroxy aldehydes (50–100 mM) in the reaction mixture
were added, and the mixture was stirred at 60 °C. The final volume was
20 mL. The pH was maintained at 7 by adding 0.1 M HCl using a pH stat
(Radiometer Analytical). The reaction was monitored by measuring Pyr
and (2S) α-hydroxy aldehydes consumption by in situ 1H NMR. In the
case of D-threose and L-erythrose, the complete disappearance of Pyr
was observed after 48 h and a second portion of Pyr (50 mM) was
added. After total conversion of Pyr and (2S)-α-hydroxy aldehydes (24–
96 h), silica was added to the solution, and the suspension was
concentrated to dryness under reduced pressure and loaded onto a flash
silica column. After silica gel chromatography using CH2Cl2:CH3OH v:v
90:10 as eluent (for L-ribulose 4) and EtOAc:CH3OH v:v 100:0–90:10 as
eluent (for D-tagatose 5 and L-psicose 6), compounds 4, 5 and 6 were
isolated.
(3S,4S)-1,3,4,5-tetrahydroxypentan-2-one (L-ribulose) 4. The title
compound was isolated as a yellow oil; yield: 84 mg, 56 %. Rf 0.43
(DCM/CH3OH, 8:2). [α]D20= +15.8 (c 0.1, H2O) lit. [α]D20 = +16.5 ± 1.5 (c
0.1 H2O).38 NMR data were identical to those previously described30 (ratio
α-anomer/β-anomer/linear form = 63/23/14; lit. ratio α-anomer/β-
anomer/linear form = 58/24/18).30 13C NMR (101 MHz, D2O): δ (ppm)
(linear form) 213.9 (C-2), 76.8 (C-3), 74.0 (C-4), 67.9 (C-1), 62.7 (C-5); δ
(ppm) (α-anomer) 104.0 (C-2), 72.9 (C-5), 71.9 (C-3), 64.2 (C-4), 71.6
(C-1); δ (ppm) (β-anomer) 107.1 (C-2), 77.2 (C-3), 72.2 (C-5), 72.0 (C-
4), 63.9 (C-1). m/z HRMS found [M + Cl]- 185.0212, C5H10O5Cl requires
185.0217.
Preparative-scale enzymatic cascade synthesis
Wild-type and variants of L-fructose-6-phosphate aldolase from E. coli
(FSAeco
)
19 wild-type L-α-transaminase from Thermosinus carboxydivorans
and wild-type TK from Geobacillus stearothermophilus (TKgst)
18
11
(TAtca
)
were produced from recombinant cells and purified as previously
described (SI).
Synthesis of L-glyceraldehyde 1. GA (50 mM) and FA (100 mM) were
dissolved in H2O (20 mL), and the pH was adjusted to 8 with 0.1 M
NaOH. Wild-type FSAeco (8 mg) was added to initiate the reaction, and
the mixture was stirred at 50 C, 100 rpm. The reaction was monitored by
in situ 1H NMR and HPLC. After complete disappearance of GA (100%
conversion, 19 h), the reaction was stopped and the purity of L-
glyceraldehyde 1 was analysed by 1H NMR and HPLC.
(2S)-2,3-dihydroxy-propionaldehyde (L-glyceraldehyde) 1. The title
compound was obtained in solution in unbuffered water Rf 0.82
(CH2Cl2:CH3OH, 80:20 v:v). HPLC: tr = 15.78 min. NMR data were
identical to those previously reported.27 1H NMR (400 MHz, D2O): δ
(ppm) (hydrate form): 4.92 (d, J = 5.2 Hz, 1H, C-1), 3.73 (dd, J = 11.1,
2.7 Hz, 1H, H-3), 3.61 (m, 1H, H-3’), 3.56 (m, 1H, H-2). 13C NMR (101
MHz, D2O): δ (ppm) (hydrate form): 98.7 (C-1), 74.0 (C-2), 61,6 (C-3).
Synthesis of D-threose 2. GA (200 mM) was dissolved in H2O (5 mL)
and the pH was adjusted to 8 with 0.1 M NaOH. FSAeco A129G mutant (2
mg) was added to initiate the reaction, and the mixture was stirred at
25 C, 900 rpm. The reaction was monitored by in situ 1H NMR and
HPLC. After quasi-complete disappearance of GA (98% conversion, 24
h), the reaction was stopped and the purity of D-threose 3 was analysed
by 1H NMR and HPLC.
(2S,3R)-2,3,4-trihydroxy-butyraldehyde (D-threose) 2. The title
compound was obtained in solution in unbuffered water,. Rf 0.36
(EtOAc:CH3OH, 95:5 v:v). HPLC: tr = 13.96 min. NMR data were identical
to those previously reported28 (ratio α-furanose/β-furanose/hydrate linear
(3S,4S,5R)-1,3,4,5,6-pentahydroxyhexan-2-one (D-tagatose) 5. The
title compound was isolated as a white powder; yield: 93 mg, 52%. Rf
0.31 (DCM/CH3OH, 8:2). [α]D25 = −2.9 (c 2, H2O) lit. [α]D20 = −2.5.31 NMR
data were identical to those previously described32 (ratio α-D-tagato-2.6-
pyranose/β-D-tagato-2.6-pyranose/α-D-tagato-2.5-furanose/β-D-tagato-
2.6-furanose = 72/22/2/4.; lit. ratio α-D-tagato-2.6-pyranose/β-D-tagato-
2.6-pyranose/α-D-tagato-2.5-furanose/β-D-tagato-2.6-furanose
=
79/14/2/5).33 13C NMR (101 MHz, D2O)
δ
(ppm) (α-D-tagato-2.6-
pyranose): 99.0 (C-2), 71.8 (C-4), 70.6 (C-3), 67.2 (C-5), 64.7 (C-1), 63.1
(C-6); δ (ppm) (β-D-tagato-2.6-pyranose): 99.1 (C-2), 70.6 (C-4), 70.1 (C-
5), 64.5 (C-3), 64.3 (C-1); 60.9 (C-6); δ (ppm) (α-D-tagato-2.5-furanose):
105.8 (C-2), 80.0 (C-5), 77.5 (C-3), 71.9 (C 4), 63.2 (C-1), 62.6 (C-6); δ
(ppm) (β-D-tagato-2.6-furanose): 103.4 (C-2), 80.9 (C-5), 71.8 (C-4), 71.5
(C3), 63.4 (C-1), 61.8 (C-6). m/z HRMS found [M + Cl]- 215.0319,
C6H12O6Cl requires 215.0322.
form
= 52/38/10;lit. ratio α-anomer/β-anomer/linear hydrate form =
52/38/10).29 1H NMR (400 MHz, D2O): δ (ppm) (hydrate linear form): 5.02
(d, J = 6.3 Hz, 1H, H-1), 3.88 (ddd, J = 7.5, 5.0, 2.7 Hz, 1H, H-3), 3.65 (m,
1H, H-4), 3.63 ppm (m, 1H, H-4’), 3.45 ppm (dd, J = 6.2 , 2.7 Hz, 1H, H-
2); (α-D-threo-1,4-furanose): 5.40 (d, J = 4.2 Hz, 1H, H-1), 4.30 (m, 1H,
H-3), 4.17 (m, 1H, H-4), 4.05 (m, 1H, H-2); (β-D-threo-1,4-furanose): 5.24
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901756
ChemCatChem
FULL PAPER
[11] J. Abdoul Zabar, I. Sorel, V. Hélaine, F. Charmantray, T. Devamani, D.
Yi, V. de Berardinis, D. Louis, P. Marlière, W.-D. Fessner, L. Hecquet,
Adv. Synth. Catal. 2013, 355, 116-128.
[12] G. Ali, T. Moreau, C. Forano, C. Mousty, V. Prevot, F. Charmantray, L.
Hecquet, ChemCatChem. 2015, 7, 3163–3170.
(3S,4S,5S)-1,3,4,5,6-pentahydroxyhexan-2-one (L-psicose) 6. The title
compound was isolated as a white powder; yield: 92 mg, 51%. Rf 0.31
(DCM/CH3OH, 8:2). [α]D20 = −2.3 (c 0.1 H2O) lit. [α]D20 = −2.4.34 Data for
D-psicose: +3.02 (c = 1.16 in H2O) 44, +3.1 (c = 1.62 in H2O).35 NMR data
were identical to those previously described for the D enantiomer:32,36
[13] a) J. Abdoul Zabar, M. Lorillière, D.Yi, S. Thangavelu, T. Devamini, L.
(ratio
α-D-psico-2.6-pyranose/β-D-psico-2.6-pyranose/α-D-psico-2.5-
Nauton, F. Charmantray, V. Hélaine, W.-D. Fessner, L. Hecquet, Adv.
furanose/β-D-psico-2.6-furanose = 25/24/37/13; lit. ratio α-D-psico-2.6-
pyranose/β-D-psico-2.6-pyranose/α-D-psico-2.5-furanose/β-D-psico-2.6-
furanose = 22/24/39/15).42 1H NMR (400 MHz, D2O) δ (ppm) 3.43 (d, J =
11.7 Hz), 3.87–3.51 (m), 4.11–3.91 (m), 4.09 (s), 4.33 (dd, J = 7.6, 4.7
Hz). 13C NMR (101 MHz, D2O) δ (ppm) (α-D-psico-2,6-pyranose) 98.6 (C-
2), 72.6 (C-4), 66.8 (C-5), 66.4 (C-3), 64.0 (C-1), 58.9 (C-6); δ (ppm) (β-
D-psico-2,6-pyranose) 99.3 (C-2), 71.12 (C-3), 69.9 (C-5), 66.0 (C-4),
65.1 (C-6), 64.9 (C-1); δ (ppm) (α-D-psico-2,5-furanose) 104.2 (C-2), 83.6
(C-5), 71.2 (C-4), 71.2 (C-3), 64.2 (C-1), 62.2 (C-6); δ (ppm) (β-D-psico-
2,6-furanose) 106.5 (C-2), 83.6 (C-5), 75.5 (C-3), 71.9 (C-4), 63.7 (C-6),
63.3 (C-1). m/z HRMS found [M + Cl]- 215.0318, C6H12O6Cl requires
215.0322.
Synth. Catal. 2015, 357, 1715-1720; b)
Catalysis. 2019, 9, 4754-4763.
[14] T. Saravanan, T. Devamani, F. Charmantray, L. Hecquet , W.-D.
Fessner, Chem. Commun. 2015, 51, 480-483.
,
,
,
,
ACS
[15] T. Saravanan, M. L. Reif, D. Yi, M. Lorillière, F. Charmantray, L.
Hecquet, W.-D. Fessner, Green Chem. 2017, 19, 481-489.
[16] T. Saravanan, S. Junker, M. Kickstein, J. Hegen, Hein, M. K. Link, S.
Witt, M. Lorillière, L. Hecquet, W.-D. Fessner, Angew. Chem. Int. Ed.
2017, 56, 5358-5362.
[17] L. Hecquet, J. Bolte and C. Demuynck. Tetrahedron. 1996, 52, 8223-
8232; b) M. Bawn, F. Subrizi, G. J. Lye, T. D. Sheppard, H. C. Hailes,, J.
M. Ward.. Enzyme Microb. Technol. 2018, 116, 16-22.
[18] M. Lorillière, M. De Sousa, F. Bruna, E. Heuson, T. Gefflaut, V. de
Berardinis, T. Saravanan, D. Yi, W.-D. Fessner, F. Charmantray, L.
Hecquet, Green Chem. 2017, 19, 425-435.
[19] A. Szekrenyi, X. Garrabou, T. Parella, J. Joglar, J.;Bujons, P. Clapés,
Nat.Chem. 2015, 7 (9), 724–729.
[20] C. Guérard-Hélaine, M. De Sousa Lopes Moreira, N. Touisni, L .
Hecquet, M. Lemaire and V. Hélaine. Adv. Synth. Catal. 2017, 359,
2061-2065.
[21] a) A. Kylmä, T. Granström, M. Leisola , Appl. Microbiol. Biotechnol.
2004, 63, 584–591; b) K. Vanhessche, E. Van der Eyckenand, M.
Vandewalle, Tetrahedron Lett. 990, 31, 2337-2340; c) G. Gumina, G.Y.
Song, C.K. Chu, FEMS Microbiol. Lett. 2001, 202, 9–15.
[22] D.-K. Oh, Appl. Microbiol. Biotechnol. 2007, 76, 1–8.
[23] a) Y. Terami, K. Uechi, S. Nomura, N. Okamoto, K. Morimoto, G.
Takata., Biosci. Biotechnol. Biochem. 2015, 79, 1725-9; b) D. Rao, P.
Gullapalli, A. Yoshihara , S.F. Jenkinson, K. Morimoto, G. Takata, K.
Akimitsu , S. Tajima, G.W. Fleet, K. Izumori, J. Biosci. Bioeng. 2008,
106, 473-80.
Acknowledgements
The authors thank the French National Research Agency (grant ANR-13-
IS07-0003-01 to L.H), the Ministerio de Economía y Competitividad
(MINECO), the Fondo Europeo de Desarrollo Regional (FEDER) (grant
RTI2018-094637-B-I00), COST Action CM1303 “Systems Biocatalysis”
(financial support of a short-term scientific mission in Barcelona to M.L)
and ERACoBioTech (European Union’s Horizon 2020 research and
innovation programme under grant agreement No [722361] to L.H., P.C
and W.D.F).
Keywords: biocatalysis, transketolase, transaminase, fructose-6-
phosphate aldolase, hydroxypyruvate, ketoses, stereoselectivity.
[24] M. Schurmann, G. A. Sprenger, J. Biol. Chem. 2001, 276, 11055–
11061.
[25] R. Roldán, I. Sanchez-Moreno, T. Scheidt, V. Hélaine, M. Lemaire, T.
Parella, P. Clapés, W.-D. Fessner, C. Guérard-Hélaine, Chem. Eur. J.
2017, 23, 5005–5009.
[1]
[2]
R. A.; Sheldon, J. M. Woodley Chem. Rev. 2018, 118, 801–838.
J. H. SchrittwieSer, S. Velikogne, M. Hall W. Kroutil, Chem. Rev. 2018,
118, 270–348.
[26] a) D. Güclü, A. Szekrenyi, X. Garrabou, M. Kickstein, S. Junker, P.
Clapés, W.-D. Fessner, ACS Catal. 2016, 6, 1848–1852. b) R. Roldán,
K. Hernandez, J. Joglar, J. Bujons, T. Parella, I. Sánchez-Moreno, V.
Hélaine, M. Lemaire, C. Guérard-Hélaine, W.-D. Fessner, ACS Catal.
2018, 8, 8804–8809.
[27] X. Garrabou, J. A. Castillo, C. Guérard-Hélaine, T. Parella, J. Joglar, M.
Lemaire, P. Clapés, Angew. Chem. Int. Ed. Engl. 2009, 48, 5521–
5525.
[3]
[4]
J. M. Sperl, V. Sieber, ACS Catal. 2018, 8, 2385–2396.
S. Wu, Z. Li, ChemCatChem. 2017, 10, 2164–2178.
[5]
P. Clapes, W.-D. Fessner, G. Sprenger, A. K. Samland, Curr. Opin.
Chem. Biol. 2010, 14, 154-167.
[6]
[7]
M. Müller , D. Gocke , M. Pohl , FEBS J. 2009, 27, 2894-2904.
a) Y. Kobori, D. C. Myles, G. M. Whitesides, J. Org. Chem. 1992,
57,5899-5907; b) S. R. Marsden, L. Gjonaj, S. J. Eustace, U.
Hanefeld, ChemCatChem.2017, 9, 1808 – 1814.
[28] A. S. Serianni, E. L. Clark, R. C. Barker, Carbohydr. Res. 1979, 72, 79–
[8]
[9]
a) T. Ziegler, A. Straub, F. Effenberger, Angew. Chem. Int. Ed.
Engl.1988, 100, 737–738; b) L. Hecquet, J. Bolte, C. Demuynck,
Tetrahedron. 1996, 52, 8223–8232; c) F. T. Zimmermann, A.
Schneider, U. Schörken, G. A. Sprenger, W.-D. Fessner,
Tetrahedron: Asymmetry. 1999, 10, 1643-1646; d) N. J. Turner, Curr.
Opin. Biotechnol. 2000, 11, 527–531; e) F. Charmantray, V. Helaine,
B.; Legeret, L. Hecquet, J. Mol. Catal. B: Enzym. 2009, 57, 6-9.
a) J. L. Galman, D. Steadmann, S. Bacon, P. Morris, M. E. B. Smith J.;
Ward, P. A. Dalby, H. C. Hailes, Chem. Commun. 2010, 46, 7608-
7610; b) A. Cázares, J. L. Galman, L. G. Crago M. E. Smith J.
Strafford, L. Ríos-Solís, G. L. Lye, P. A. Dalby, H. C. Hailes, Org.
Biomol. Chem. 2010, 8, 1301-1130; c) F.; Subrizi, M. Cárdenas-
Fernández, G. J. Lye, J. M. Ward, P. A. Dalby, T. D. Sheppard, H. C.
Hailes, Green Chem. 2016, 18, 3158-3165.
91.
[29] C. André, J. Bolte, C. Demuynck, Tetrahedron: Asymmetry. 1998, 9,
1359–1367.
[30] T. Vuorinen and A. S. Serianni, Carbohydr. Res. 1991, 209, 13–31.
[31] K. Yoshihara, Y. Shinohara, T. Hirotsu and K. Izumori, J. Biosci.
Bioeng. 2006, 101, 219–222.
[32] J.-P. Sauvage, S. Chapelle and J.-F. Verchère, Carbohydr. Res. 1992,
237, 23 –32.
[33] S. Köpper and S. Freimund, Helv. Chim. Acta. 2003, 86, 827–843.
[34] K. Takeshita, T. Shimonishi and K. Izumori, J. Biosci. Bioeng. 1996, 81,
212–215.
[35] M. Steiger and T. Reichstein, Helv. Chim. Acta. 1936, 19, 184–189.
[36] D. Enders and C. Grondal, Angew. Chem. Int. Ed. 2005, 8, 1210–1212.
[10] A. Ranoux, S. K. Karmee, J. Jin, A; Bhaduri, A. Caiazzo, I.W. Arends,
and U. Hanefeld, ChemBioChem. 2012, 13, 1921-1931.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901756
ChemCatChem
FULL PAPER
FULL PAPER
Marion
Lorillière,[a],
Christine
Efficient sequential one-pot, three-
step cascade reactions: in situ
generation of both transketolase
Guérard-Hélaine, Thierry Gefflaut,
Wolf-Dieter Fessner, [c] Pere Clapés,
[b]
Franck Charmantray,*[a] Laurence
FSA
substrates
catalyzed
by
a
Hecquet*[a]
Pyruvate
L-alanine
transaminase (TA) and D-fructose-6-
phosphate aldolase (FSA) from L-
Serine and achiral compounds,
TA
TK
Page No. – Page No.
CO2
glycolaldehyde
(GA)
and
formaldehyde (FA).
Convergent in situ Generation of
Both Transketolase Substrates via
Transaminase and Aldolase
Reactions for Sequential One-Pot,
Three-Step Cascade Synthesis of
Ketoses
This article is protected by copyright. All rights reserved.