Expanding the reaction space of aldolases using hydroxypyruvate as a nucleophilic substrate

Article abstract of DOI:10.1039/c6gc02652d

Aldolases are key biocatalysts for stereoselective C-C bond formation allowing access to polyoxygenated chiral units through direct, efficient, and sustainable synthetic processes. The aldol reaction involving unprotected hydroxypyruvate and an aldehyde offers access to valuable polyhydroxy-α-keto acids. However, this undescribed aldolisation is highly challenging, especially regarding stereoselectivity. This reaction was explored using, as biocatalysts, a collection of aldolases selected from biodiversity. Several enzymes that belong to the same pyruvate aldolase Pfam family (PF03328) were found to produce the desired hexulosonic acids from hydroxypyruvate and d-glyceraldehyde with complementary stereoselectivities. One of them was selected for the proof of concept as a biocatalytic tool to prepare five (3S,4S) aldol adducts through an eco-friendly process.

Full text of DOI:10.1039/c6gc02652d

View Article Online
View Journal
Green
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: V. de Berardinis,
C. Helaine, E. Darii, K. Bastard, V. Hélaine, A. Mariage, J. Petit, N. Poupard, I. Sanchez-Moreno, M. Stam, T.
Gefflaut, M. Salanoubat and M. Lemaire, Green Chem., 2016, DOI: 10.1039/C6GC02652D.
This is an Accepted Manuscript, which has been through the
Volume 18 Number
7 7 April 2016 Pages 1821–2242
Royal Society of Chemistry peer review process and has been
accepted for publication.
Green
Chemistry
Accepted Manuscripts are published online shortly after
Cutting-edge research for a greener sustainable future
www.rsc.org/greenchem
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1463-9262
CRITICAL REVIEW
G. Chatel et al.
Heterogeneous catalytic oxidation for lignin valorization into valuable
chemicals: what results? What limitations? What trends?
rsc.li/green-chem

Please do not adjust margins
Page 1 of 10
Green Chemistry
View Article Online
DOI: 10.1039/C6GC02652D
Journal Name
ARTICLE
Expanding the reaction space of aldolases using hydroxypyruvate
as nucleophile substrate
Véronique de Berardinis,a† Christine Guérard-Hélaine,b† Ekaterina Darii,a Karine Bastard,a Virgil
Hélaine,b Aline Mariage,a Jean-Louis Petit,a Nicolas Poupard,b Israel Sanchez-Moreno,b Mark
Stam,a Thierry Gefflaut,b Marcel Salanoubat*,a† and Marielle Lemaire*,b†
Received 00th January 20xx,
Accepted 00th January 20xx
DOI: 10.1039/x0xx00000x
www.rsc.org/
Aldolases are key biocatalysts for stereoselective C-C bond formation allowing access to polyoxygenated chiral units
through direct, efficient, and sustainable synthetic processes. Aldol reaction involving unprotected hydroxypyruvate and
an aldehyde offers access to valuable polyhydroxy-α-keto acids. However, this undescribed aldolisation is highly
challenging, especially regarding stereoselectivity. This reaction was explored using, as biocatalysts, a collection of
aldolases selected from biodiversity. Several enzymes that belong to the same pyruvate aldolase Pfam family (PF03328),
were found to produce the desired hexulosonic acids from hydroxypyruvate and D-glyceraldehyde with complementary
stereoselectivities. One of them was selected for proof of concept as biocatalytic tool to prepare five (3S,4S) aldol adducts
through an eco-friendly process.
complete atom economy to the 3,4-dihydroxy-2-oxo-acid
moiety along with the control of two stereogenic centres. To
our knowledge, despite the tremendous developments in the
Introduction
Asymmetric aldol reactions play a pivotal role in the synthesis
of natural compounds and drug precursors.1 For many years,
aldolases catalysing stereoselective C-C bond formation have
been considered essential for synthetic applications.2 Indeed,
due to enzyme selectivity, no protection-deprotection steps
are needed and an optimal atom economy is reached.
Moreover, biocatalysed aldolisation reactions are performed
under mild conditions, generally without need of any co-
solvent and are therefore highly valuable for the development
of green synthetic processes. Although these biocatalysts are
known to accept a broad range of aldehydes as electrophiles,
and therefore have been extensively used, most of them are
strictly dependent on a sole nucleophile substrate. A notable
exception is fructose-6-phosphate aldolase (FSA) which is able
to catalyse aldolisations in a high stereoselective manner with
several nucleophiles.3 As already pointed out by Turner et al. in
their report on biocatalytic retrosynthesis, there is room for
new C-C bond forming enzymes to construct more complex
molecules as this category of biocatalysts is underused.4 In this
context, enzymatic aldol reactions involving hydroxypyruvate
(HPA) as a nucleophile could give access in one step and with
field of aldol reactions, no equivalent of such a reaction was
reported in the literature.5 Moreover, as highlighted by
Mlynarski et al,6 even aldol reactions with ketoacids such as
pyruvate and its ester derivatives remain little described. To
date, synthetic access to polyhydroxy-α-keto acids was mainly
restricted to 6-carbon chain compounds prepared by oxidation
of natural sugars7 and to multistep protection-deprotection
routes from sugar pool.8
Following an aldolisation strategy involving HPA, a variety of
ulosonic acids and analogues of biological interest9,10 could be
prepared in a sustainable manner (Scheme 1). For instance,
glycero-
-talo-2-octulosonic acid (KO),8,10 analogue of the well-
known 3-deoxy- -manno-2-octulosonic acid (KDO), is an acid
resistant component of Gram-negative bacterial
D-
D
D
lipopolysaccharides (LPS) involved in immune response and
pathogen recognition. Moreover, a variety of derivatives can
be readily obtained from the α-ketoacid moiety in one or few
steps (Scheme 1).1a,11 This includes hydroxy- aldehydes,
ketones, acids, esters or tetronic acids analogues of vitamin C
as well as amino acids such as polyoxamic acids found in
peptidyl nucleoside antibiotics. Thus, HPA-aldolases would
complement the already well known biocatalysts such as
threonine aldolases (leading to
-aminoacids) or DHAP
a.CEA, DRF, IG, Genoscope, UMR 8030, 91057 Evry, France; CNRS, UMR8030, 91057
Evry, France; Université d’Evry Val d’Essonne, UMR 8030, 91057 Evry, France.
b.Clermont Université, Université Blaise Pascal, ICCF BP 10448, F-63000 Clermont-
Ferrand, France; CNRS, UMR 6296, BP 80026, F-63177 Aubière, France.
† These authors contributed equally to this work.
Electronic Supplementary Information (ESI) available: [details of any
supplementary information available should be included here]. See
DOI: 10.1039/x0xx00000x
aldolases (leading to hydroxyketones) for which the main
drawback is a lack of stereoselectivity, or DHA aldolases
(leading to hydroxy-ketones or aldehydes) offering a limited
access to the D
-threo configuration.12
We describe herein several natural aldolases selected from
genomic databases, able to efficiently catalyse the aldolisation
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 2 of 10
ARTICLE
Journal Name
reaction between HPA and aldehydes with complementary 2-keto-3-deoxy-L-rhamnonate aldolase RhmA (P76469)).16,17
View Article Online
DOI: 10.1039/C6GC02652D
stereoselectivities. One of them, providing the hardly accessed The resulting pool of 22 enzymes were His-tagged and purified
(3S,4S) configuration, was chosen for further applications in (Method S5) for further studies. Interestingly, the three E. coli
synthesis.
enzymes were shown to promote the formation of a
hexulosonic acid detected by LC/MS in levels close to that
measured with the 19 new aldolases identified by screening.
Moreover, in accordance with their annotation, all of the 22
enzymes proved also active with PA as a nucleophile as shown
by LC/MS detection of the 3-deoxyulosonic acid formed in the
presence of PA and GA (Table 1).
Scheme 1 Retrosynthetic approach to access polyfunctionalised structures from
the ketoacid moiety.
NMR characterisation of aldolisation products
The next steps were to confirm that the enzymes were
catalysing the expected reaction through the characterisation
of aldol adducts and to have a first estimation of yields and
stereoselectivities, in order to select the most promising
enzymes. Therefore, small scale reactions with the 22
aldolases used directly as cell lysates were performed with 0.5
mmol. of each HPA and D-GA. The reactions were run at pH 8
in Gly-Gly buffer for 24 h. Concomitantly, a series of reactions
with PA as the nucleophile substrate and blank reactions
without enzyme were carried out in the same conditions.
Titration of HPA using a standard spectrophotometric assay
(monitoring of NADH absorbance in presence of Lactic
Dehydrogenase (LDH)) indicated less than 10% degradation of
HPA without aldolase over 24 h. Considering that 2-
Results and discussion
Screening of aldolase collection
Among established strategies to provide enzymes possessing
new or improved properties, the exploration of the protein
natural diversity is one of the most attractive.13,14 A general
enzymatic screening was performed using HPA as the
nucleophile in the aldol reaction to discover HPA aldolases
from a collection of selected natural enzymes.15 A sequence-
driven approach was used to build a library representative of
ketogluconic acid (1) was the only commercially available
hexulosonic acid standard, 13C NMR analysis of the crude
reaction mixtures was preferred over HPLC analyses to
estimate yields and selectivities (Method S6). Configurations at
C3 and C4 positions of hexulosonic acids were assigned by
comparison with 13C NMR spectra of known compounds. (3S,
the functional diversity including Class
I
(enamine
intermediate) and II (enolate intermediate) aldolase families
(Table S1, Methods S1 and S2). Only threonine aldolases were
excluded, as they do not catalyse addition of a keto-
nucleophile.1,12 The process led to a selection of 571 proteins
covering most of the selected family’s diversity to be screened
(Figure S1).
4R) configuration was determined by comparison with
1
whereas the other configurations were assigned after chemical
oxidative decarboxylation giving known aldonic acids (Scheme
2).18 When PA was used as nucleophile, oxidative
decarboxylation to 3-deoxyaldonic acids18b was also chosen to
assign the 4R or 4S configuration of the known aldol
5 .
and 6 19
The biocatalytic potential of each candidate protein was
assessed in cell free extract after addition of HPA and D,L-
Table 2 summarises the results obtained with the 22 enzymes
using HPA or PA as nucleophile.
glyceraldehyde (GA) (Method S3). GA was chosen since it is a
common electrophile substrate for most aldolases. The aldol
formation (hexulosonic acid upon reaction with HPA) was
monitored through LC/MS analysis by detection of the desired
product through Multiple Reaction Monitoring (MRM): m/z
Only one aldolase (Q2K7V6, Table 2, entry 17) was not
confirmed to react with HPA. This correlates with the LC/MS
analysis (Table 1), showing that this enzyme is one of the less
active towards HPA. Among the 22 enzymes, 10 allowed the
complete conversion of HPA and thus constitute efficient
biocatalysts. Regarding stereoselectivity, hexulosonic acids
were always obtained as a mixture of two or three stereomers
193103.9 and m/z 19358.9 characteristic transitions.
Commercially available 2-ketogluconic acid was used as a
standard (Method S4). Noteworthy this MS-based analysis
allows the detection of all possible stereoisomers.
This screening revealed 19 enzymes able to catalyse the
formation of a hexulosonic acid (Table 1). They are annotated
mainly as putative 4-hydroxy-2-oxo-heptane-1,7-dioate
except in one case (B2T1L6, entry 11) where
1 was the sole
detected product. Other enzymes as, for example, B4XH86
(entry 4) or A5VAX1 (entry 8) showed the same selectivity
towards the (3S,4R) isomer
isomer. Conversely, A5VH82 (entry 5) and B2TGJ0 (entry 3)
gave mainly the (3S,4S) isomer . The (3R,4S) isomer was
formed with A0A081HJP9 (entry 1) and Q9RYM1 (entry 18).
However, and were also detected as minor products in
both cases. The last (3R,4R) configuration corresponding to
1 which was obtained as the major
aldolase (HpcH/HpaI), and putative 5-keto-4-deoxy-D-glucarate
3
2
aldolases (GarL/DDGA) falling into the class II family of pyruvic
acid (PA) aldolase.16 For comparison three well-known and
biochemically characterized class II PA aldolases from E. coli
were included in the study: GarL (P23522), HpcH (B1IS70) and
1
3
D-
ribonate was never evidenced in our experiments. However,
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 10
Green Chemistry
Please do not adjust margins
Journal Name
ARTICLE
considering the low sensitivity of 13C NMR analysis, we cannot
exclude that this isomer could have been formed in low
amounts.
View Article Online
DOI: 10.1039/C6GC02652D
Scheme
2
Configuration assignments of the stereoisomers obtained by
biocatalysed aldolisation of
D
-glyceraldehyde and hydroxypyruvate
Table 1 LC/MS analysis of HPA aldolase catalysed reactions . Values correspond to ratios of signals in sample with over expressed aldolases versus E. coli cell free extract as
reference
Entry
UniprotKB id
Organism
HPA + GA
PA + GA
annotation
uncharact. proteina
Hpch aldolase
Hpch aldolase
Hpch aldolase
Hpch aldolase
Garl aldolase
aldolase
Hpch aldolase
Hpch aldolase
Hpch aldolase
Garl aldolase
HOA
Garl aldolase
Hpch aldolase
Garl aldolase
Hpch aldolase
Hpch aldolase
Hpch aldolase
Hpch aldolase
RhmA
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
A0A081HJP9
Q0K203
B2TGJ0
B4XH86
A5VH82
B7NJZ1
A9CJL2
A5VAX1
A9E0U5
A3SLS0
B2T1L6
Q16DB5
Q1QUJ4
A7HU24
A9BQY3
A3K1H1
Q2K7V6
Q9RYM1
I7DKY0
Pseudomonas aeruginosa
Ralstonia eutropha
Burkholderia phytofirmans
Actinobacillus pleuropneumoniae
Sphingomonas wittichii
Escherichia coli IAI39
Agrobacterium tumefaciens
Sphingomonas wittichii
Oceanibulbus indolifex
Roseovarius nubinhibens
Burkholderia phytofirmans
Roseobacter denitrificans
Chromohalobacter salexigens
Parvibaculum lavamentivorans
Delftia acidovorans
35
33
45
22
48
21
6
34
30
11
8
8
9
5
8
5
6
113
119
121
20
24
114
32
88
125
135
16
49
137
22
99
99
49
111
131
72
75
126
Sagittula stellata
Rhizobium etli
Deinococcus radiodurans
Phaeobacter gallaeciensis
Escherichia coli K12
16
27
36
26
25
P76469
P23522
B1IS70
Escherichia coli K12
Escherichia coli C
GarL aldolase
Hpch aldolase
(HPA): hydroxy pyruvic acid; (PA): pyruvic acid; (GA): glyceraldehyde; RhmA: rhamnonate aldolase; Hpch aldolase: 2,4-dihydroxyhept-2-ene-1,7-dioic acid aldolase or
hpch/hpai aldolase family protein; a: putative uncharacterized protein; Garl aldolase : 2-dehydro-3-deoxyglucarate aldolase ; HOA: 4-hydroxy-2-oxovalerate aldolase
(wrong annotation in UniprotKB).
When using PA as nucleophile, all tested enzymes gave the 4R hydroxyl group present in HPA seems to have an influence on
isomer
A3K1H1, and Q2K7V6 (entries 3, 5, 14, 16 and 17)
only detected product whereas B4XH86 and A5VAX1 (entries 4
and 8) predominantly gave the 4S isomer . A comparison of
5
. Notably, in the case of B2TGJ0, A5VH82, A7HU24, the binding mode of
D-GA within the active site.
5
was the
4
the results obtained using HPA or PA reveals that several
enzymes (B7NJZ1, A3SLS0, B2T1L6 and P23522, entries 6, 10,
11 and 21) gave a different relative configuration at C4
depending on the nucleophile substrate. As D-GA was used in
both cases as electrophile, this result indicates that the extra
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 4 of 10
View Article Online
DOI: 10.1039/C6GC02652D
Journal Name
ARTICLE
Table 2 Hexulosonic acids§ isolated from
D
-GA and HPA (
1-
3) or PA (
4
,5
) after aldolisation catalysed by 22 selected enzymes aldolases
From HPA From PA
Entry
UniprotKB id
1
2
3
4
5
A0A081HJP9*
Q0K203*
B2TGJ0*
B4XH86*
A5VH82*
B7NJZ1*
A9CJL2
6
7
8
9
A5VAX1*
A9E0U5
A3SLS0
10
11
12
13
14
15
16
17
18
19
20
21
22
B2T1L6*
Q16DB5
Q1QUJ4
A7HU24
A9BQY3
A3K1H1
Q2K7V6
Q9RYM1*
I7DKY0
P76469
P23522
B1IS70*
* Complete conversion of HPA as detected by NMR. White: no detection;
<10;
10<>30;
30<>50;
50<>75;
>75, ratio estimated by 13C NMR on crude
sample; in bold: known E.coli aldolases included as reference proteins.
Preliminary characterization of selected aldolases
respectively. In order to compare the relative aldolase activity
of each enzyme with HPA and PA, glycolaldehyde was chosen
Three aldolases (A5VH82 from Sphingomonas wittichii, B2T1L6
from Burkholderia phytofirmans and A0A081HJP9 from
as the electrophile substrate to avoid
a
possible
enantiopreference that could be observed towards a chiral
aldehyde. The nucleophile disappearance was monitored using
LDH and NADH. Interestingly, close activities towards both
nucleophiles were found for A5VH82 whereas A0A081HJP9
and B2T1L6 gave 3.3 and 2.5 fold faster reactions with PA
respectively (Table S2, Method S8).
Pseudomonas
aeruginosa)
with
complementary
stereoselectivities were selected and purified by IMAC
(Method S6) for further characterisation. The three aldolases
were shown to efficiently catalyse the decarboxylation of
oxalacetic acid which is a general property of known PA-
aldolases. For convenience, this reaction was used to measure
the activity of the three enzyme, by using a standard direct
assay using LDH and NADH (Method S7).20 Various divalent
cations (Mg2+, Co2+ and Mn2+) efficiently restored the activity
of the purified enzymes, all of them showing a slight
preference for Mn2+. Furthermore, inhibition by EDTA
confirmed that the three enzymes belong to the class-II
aldolase family. Decarboxylation activities of 12, 20 and 101
U/mg were measured for B2T1L6, A5VH82, and A0A081H5P9
Synthetic applications with selected aldolases
Small preparative scale reactions were then performed with
HPA and
separated by ion exchange chromatography. Using B2T1L6 as
biocatalyst, the (3S,4R) isomer was isolated as the unique
product with 55% yield. Compound (3R,4S) was obtained as
D-GA and the resulting aldols were purified and
1
2
the major product with 30% yield when using A0A081HJP9.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 4
Please do not adjust margins

Page 5 of 10
Green Chemistry
Please do not adjust margins
Journal Name
Finally, (3S,4S) isomer
ARTICLE
3
was the main compound produced that HPA is widely used as nucleophile in this famiVlyiew(sAerteiclFeiOgnulirnee
DOI: 10.1039/C6GC02652D
with A5VH82 (68% isolated yield). This latter configuration was S2, Tables S4 and S5) in spite of their probable diverse
previously difficult to obtain with known aldolases. Indeed, physiological function. Structural analysis of PF03328 family
among the four stereo-complementary DHAP aldolases, was then conducted and 1481 tridimentional homology
tagatose 1,6-bisphosphate aldolase is not highly selective for models were built successfully (Method S9). The obtained
the (3S,4S)-configuration.1,21 Aldolase A5VH82 thus appears as structures were aligned and the residue conservation was
an alternative catalyst to access this configuration more calculated showing that the active site is very well conserved
efficiently. Therefore, in order to investigate the electrophile among the family (Figure S3), again corroborating a wide use
substrate scope of A5VH82, other reactions were carried out of HPA. To explore the ability of aldolases belonging to other
using four additional electrophile substrates:
lactaldehyde and -erythrose. These aldehydes were extended collection were experimentally re-investigated using
strategically chosen to enable the characterization of the both PA and HPA as nucleophiles. While 18 additional HPA
L-GA,
D- and
L-
pyruvate-aldolase families to use HPA, enzymes from our
D
corresponding aldol adducts 6-9 by comparison with known aldolases were revealed for PF03328 family (Table S6), 60
compounds (Scheme 3).22 Each aldehyde (1 mmol) was reacted selected representatives of the others aldolase families
over 20 h with HPA (0.75 mmol) in the presence of A5VH82 (PF00701 (NanA/DapA), PF01081 (KDPG) and PF00682
(2,5 mg) following a standard protocol and products were (MhpE/HOA)) were shown to be unable to use HPA to produce
purified by ion exchange chromatography (see experimental). hexulosonic acids (Table S7). However, the use of HPA as a
Compound
13C NMR analysis, this product exists in solution as a mixture of since their active site is conserved with PF03328 members23
four cyclic forms (
and β anomeric forms of furanose and and moreover, E. coli MhpE aldolase23b was described to be
pyranose cycles).18a The (3S,4S) configuration and de > 95% inactive with GA as electrophile. The discovery of other HPA
were readily assessed since -ribonate was the only product aldolases in PF00682 might then be possible with other
detected by NMR after decarboxylation.22 Starting from
6 was isolated from L-GA in 61% yield. As shown by nucleophile for the class II MhpE aldolases cannot be excluded

L
D-
aldehyde electrophiles.
lactaldehyde, a mixture of compounds 7a and 7b was isolated
in 51% yield. Three isomers, not separated by ion exchange
chromatography, were evidenced by 1H NMR analysis. A
NOESY experiment showed that the two major ones were the
Experimental
1- Reactions using the 22 selected aldolases with HPA or PA and D-
GA.

and β furanic forms of the (3S,4S) aldol 7a whereas the
minor one was observed under one furanic form of the (3S,4R)
aldol 7b. In the case of -lactaldehyde, compound was
isolated in 47% yield. H NMR analysis showed an equimolar
mixture of and β furanic anomers of . Configuration
assignment was done after decarboxylation and lactonisation.
Decarboxylation afforded single compound whose
lactonisation gave the known 5-deoxy- -ribono-1,4-lactone
confirming the (3S,4S) configuration.22b Finally, compound
was isolated in 57% yield when -erythrose was used as
electrophile. Only one pyranic form was observed by NMR.
L
8
General reaction protocol
1
To a solution of D-GA (45 mg, 0.5 mmol,) in a 10 mL final
volume of 50 mM Gly-Gly buffer, pH 8, were added HPA or PA
(0.5 mmol) and 0.5 mL of each aldolase lysate prepared as
described in Method S2. After stirring for 24 h, the reaction
mixture was poured onto a Dowex 1X8 column (200-400 mesh,
AcO- form, 4 mL). The column was washed with H2O (10 mL)
before elution with 1 M HCO2H. Fractions containing the
products (as checked by TLC using 6/5: NH4OH/EtOH as eluent)
were combined and concentrated under reduced pressure.
The residue was dissolved in H2O (5 mL) and concentrated
again. This latter operation was repeated to remove HCO2H.
The resulting products mixture was then analysed by NMR.
Alternatively, products 1, 2 and 3 were separated using the
same protocol and a 0.1 – 1 M HCO2H gradient for column
elution.
2-ketogluconic acid 1: NMR spectra were identical to those
obtained from an authentic commercially available sample
(from Sigma-Aldrich).
Abbreviations p, p, f; f, ax and eq are used respectively
for pyranose, pyranose, furanose, furanose, axial

8
a
L
9
D
Upon decarboxylation, D-mannonate was obtained confirming
again the (3S,4S) configuration.22a Importantly, these
unprecedented aldolisation reactions involving HPA as the
nucleophile were conducted with respect to green chemistry
principles. Indeed, all these compounds 1-9 were prepared at
RT, in water as the unique solvent and from biobased
materials. Their purification required only reusable ion
exchange resins and diluted aqueous acid or base solutions as
eluents.
Protein sequence analysis
The 22 protein sequences were compared pairwise revealing
their sequence diversity with identities ranging from 30 to 72%
(Table S3). Regarding the functional gene annotation, all of
them were part of the same Pfam protein family (PF03328
named HpcH/HpaI aldolase/citrate lyase family). Noteworthy,
the gene cluster organizations of these HPA/PA aldolases are
diverse and not similar to HpcH, GarL or RhmA E. coli gene
neighborhoods. This diversity of genomic contexts suggests
and equatorial.
2-ketogalactonic acid
1
2
: H NMR (400 MHz, D2O) δ 3.77 (d, J =
5.3 Hz, 1H, H6eq), 3.74 – 3.55 (m, H3, H4, H5, H6ax, 4H).
13C NMR (100 MHz, D2O):

171.84 (C1); 96.19 (C2); 72.36 (C3
C4); 72.40 (C4); 69.16 (C5); 62.46 (C6). HRMS ESI-: m/z calcd.
for [C6H9O7] = 193.0354; found 193.0334.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 6 of 10
View Article Online
DOI: 10.1039/C6GC02652D
Journal Name
ARTICLE
Table 3 A5VH82 aldolase from Sphingomonas wittichii catalysed reactions using HPA as nucleophile with various electrophiles
Electrophile
D-GA
Product§
Yield* (%)
68
Configurations
de (%)†
(3S,4S,5R)
> 95
3
6
L-GA
61
51
(3S,4S,5S)
> 95
55
(3S,4S,5R)
+
(3S,4R,5R)
7a
7b
8
D-lactaldehyde
L-lactaldehyde
D-erythrose
47
57
(3S,4S,5S)
> 95
> 95
(3S,4S,5R,6R)
9
*: After purification by ion exchange resin; †: determined by NMR (> 95 : no other isomer detected; 55 : ratio of the methyl intensity signals)
Scheme 3 Chemical routes for configuration assignments of 6-9.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 6
Please do not adjust margins

Page 7 of 10
Green Chemistry
Please do not adjust margins
Journal Name
ARTICLE
H4
: 1H NMR (400 MHz, D2O) δ 4.04 (d, J = 2.8 1H, H6ax
Hz, 1H, H3), 3.83 – 3.75 (m, 3H, H4, H5, H6ax), 3.60 – 3.55 (m, H6eq p), 3.87-3.70 (m, 6H, 2H6
(m, 1H, H5
p).13C NMR (100 MHz, D2O): δ 176.21 (C1
171.93 (C1); 96.54 (C2); 70.89 (C3 172.56 (C1 p), 172.32 (C1 f or f), 171.86 (C1 f or f), 103.75
f), 100.32 (C2 f), 96.75 (C2 p), 95.92 (C2 p), 84.30
f), 83.55 (C5 f), 76.03 (C3 f), 72.97 (C3 f), 71.16 (C3
f or C4 f, or C3 p, or C4 p, or C5 p), 70.07 (C4
f or C4 f, or C3 p, or C4 p, or C5
f), 67.88 (C3p or C4p or C5 p), 65.53 (C3p or
NMR spectra were C4p or C5 p), 64.59 (C5 p), 61.87 (C6 f or C6 f), 61.15
p), 59.20 (C6 f or C6 f), 58.51 (C6 p). HRMS ESI-: m/z

f), 4.10 (d, J = 4.7 Hz, 1H, H3

f), 4.02 (dd, J =V1ie2w.A7r,tic1le.4OnHlinze,
2-ketogulonic acid
3

p), 3.96 (d, J = 2.3 Hz, 1H,DOHI4: 10.1039/C6GC02652D

p), 3.90 (m, 1H,
f, H5 f, H5 f) 3.64
p),


f, 2H6



1H, H6eq)


13C NMR (100 MHz, D2O):






); 70.42 (C4); 65.68 (C5); 62.49 (C6). HRMS ESI-: m/z calcd. for (C2
[C6H9O7] = 193.0354; found 193.0331. (C5
2-keto-3-deoxy-D-gluconic acid
to those described in literature.19 (see SI for 13C NMR 69.97 (C4













p),
p),
4
: NMR spectra were identical 70.94 (C4






p), 68.43 (C4f
description).
or C4


2-keto-3-deoxy-D-galactonic acid
5
:





identical to those described in literature.19 (see SI for 13C NMR (C6



D
description)
calcd. for [C6H9O6] = 193.0348, found: 193.0336; [
(c 2.5, H2O).
] 25 = + 14.9
Oxidative decarboxylation of 1-5
5-deoxy-
(20 mg, 0.1 mmol) in 1H NMR (400 MHz, D2O) δ 4.44 (qd, J = 6.4, 2.7 Hz, H5’), 4.40
H2O (3 mL), pH=7.5, were added 10 M H2O2 (40 μL, 0.4 mmol). (d, J = 4.8 Hz, H3), 4.37 (d, J = 4.6 Hz, H3’), 4.24 (qd, J = 6.5, 3.4
The reaction mixture was stirred for 12 at room Hz, H5), 4.22 (d, J = 8.3 Hz, H3’’), 4.13 (dd, J = 4.8, 3.4 Hz, H4),
D-lyxo-2-hexulosonic acids 7a and b: 68 mg.
To a solution of α-keto acid 1, 2, 3, 4 or 5
h
temperature. Dimethylsulfide (60 μL, 0.8 mmol) was then 3.99 (dd, J = 4.6, 2.7 Hz, H4’), 3.93 (qd, J = 8.3, 6,2 Hz, H5’’),
added and the reaction mixture was stirred for 2 h before 3.78 (t, J = 8.3 Hz, H4’’), 1.32 (d, J = 6.2 Hz, H6’’), 1.28 (d, J = 6.5
concentration under reduced pressure to give aldonic acids18 Hz, H6), 1.25 (d, J = 6.4 Hz, H6’).
in quantitative yield.
13C NMR (101 MHz, D2O) δ 176.31 (C1), 175.61 (C1‘), 174.84
(C1‘‘), 103.91 (C2‘), 99.98 (C2), 99.12 (C2‘‘), 79.28, 78.43,
77.74, 77.19, 76.61, 74.33, 73.91, 72.05 (C3, C3‘, C3‘‘, C4, C4‘,
C4‘‘, C5, C5‘, C5‘‘), 18,82 (C6‘‘), 14.38 (C6), 13.74 (C6‘). HRMS
NMR spectra were identical to those described in literature.18
(see SI for 13C NMR description)
2- Reactions using purified B2T1L6, A0A081HJP9 and A5VH82 ESI-: m/z calcd. for [C6H9O6] = 177.0399, found: 177.0397.
with HPA and various aldehydes
5-deoxy-L-ribo-2-hexulosonic acid 8: 84 mg.
1H NMR (400 MHz, D2O) δ 4.27 (d, J = 5.8 Hz, H3), 4.14 – 4.06
(m, H5, H5’), 4.07 (d, J = 4.5 Hz, H3’), 4.02 (dd, J = 8.1, 4.5 Hz,
H4’), 3.86 (t, J = 5.8 Hz, H4), 1.29 (d, J = 6.1 Hz, H6’), 1.21 (d, J =
6.4 Hz, H6).
General reaction protocol
To a 0.5 M aqueous solution of HPA (1.5 mL, 0.75 mmol) were
successively added 300 µL of 100 mM MnCl2 (0.03 mmol), a
145 mM aqueous solution of aldehyde (l-glyceraldehyde, d- or
l-lactaldehyde or d-erythrose, 7 mL, 1 mmol), 50 mM GlyGly
buffer, pH 8 (2.7 mL, 0.135 mmol) and an aqueous solution of
purified aldolase (B2T1L6, A0A081HJP9 or A5VH82, 4 mL, 50 U
corresponding to oxalacetate decarboxylation activity). The
reaction mixture was stirred over 20 h until complete
disappearance of HPA as evidenced by spectrophotometric
LDH/NADH assay. After centrifugation, the products were
isolated from the supernatant as described above by ion
exchange chromatography, using a 0.1-1 M HCO2H gradient.
Yields are given in the following table
13C NMR (100 MHz, D2O) δ 172.41 (C1‘), 171.83 (C1), 103.54
(C2), 99.96 (C2‘), 79.74 (C5), 79.45 (C5‘), 76.38 (C3‘), 75.16
(C4‘), 74.74 (C4), 72.61 (C3), 18.75 (C6‘), 17.66 (C6). HRMS ESI-:
m/z calcd. for [C6H9O6] = 177.0399, found: 177.0411; [
9.1 (c 2.8, H2O).
D
]
25 = +
D
-manno-2-heptulosonic acid 9: 128 mg.
1H NMR (400 MHz, D2O) δ 4.08 (d, J = 3.4 Hz, H3), 3.89 (dd, J =
9.7, 3.4 Hz, H4), 3.86 (dd, J = 9.2, 4.6 Hz, H7A), 3.82 – 3.73 (m,
H6, H7B), 3.64 (t, J = 9.7, H5).
13C NMR (100 MHz, D2O) δ 172.40 (C1), 96.39 (C2), 73.75 (C6),
71.00 (C3), 70.43 (C4), 66.14 (C5), 60.70 (C7). HRMS ESI-: m/z
calcd. for [C7H11O8] = 223.0459, found: 223.0472; [
13.2 (c 0.8, H2O).
D
]
= +
25
Aldehyde
Aldolase
(Uniprot KB id)
A0A081HJP9
B2T1L6
Products (Yield, %)
Determination of configurations of 6-9
D-GA
D-GA
1 (15), 2 (30), 3 (12)
1 (55)
1 (traces), 3 (68)
6 (61)
7a + 7b (51)
8 (47)
9 (57)
Compounds
6
and
9
were decarboxylated following the
to give known
-ribonate,18a
protocol already described for
1
-
5
L
and D
-mannonate22a respectively. NMR spectra were identical
D-GA
A5VH82
to those described in literature. (see SI for 13C NMR
description)
L-GA
A5VH82
D-Lactaldehyde
L-Lactaldehyde
D-erythrose
A5VH82
A5VH82
A5VH82
Compound
8 was also decarboxylated and submitted to a
subsequent lactonisation step: the crude residue obtained
after H2O2 oxidation was dissolved in H2O (5 mL) and Dowex
50WX-8 resin (100-200 mesh, H+ form, 0.6 mL) was added to
the solution. After gentle stirring for 1 h, the resin was filtered
L-ribo-2-hexulosonic acid 6 : 89 mg.
1H NMR (400 MHz, D2O) δ 4.30 (dd, J = 8.0, 4.7 Hz, 1H, H4
4.25 (d, J = 5.4 Hz, 1H, H3 f), 4.15 (s, 1H, H3 p), 4.14 (m, 1H,

f),


This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 8 of 10
ARTICLE
Journal Name
Sprenger, A.K. Samland, W.-D. Fessner Chem. EurV.ieJw. A2r0tic1le1O, n1lin7e
,
out and the solution concentrated under reduced pressure to
give the known 5-deoxy-
-ribono-1,4-lactone.22b
2623-2632.
N. J. Turner, E. O'Reilly Nature Chem. DBOioI:l.120.0101339,/C6GC02652D
9, 285–288.
L
4
5
NMR spectra were identical to those described in literature.
V. Bisai, A. Bisai, V. K. Singh, Tetrahedron 2012, 68, 4541-
(see SI for 13C NMR description)
4580.
6
(a) M. A. Molenda, S. Bas, O. El-Sepelgy, M. Stefaniak, J.
Mylnarski Adv. Synth. Catal. 2015, 357, 2098-2104; (b) O. El-
Sepelgy, D. Schwarzer, P. Oskwarek, J. Mlynarski, Eur. J. Org.
Chem. 2012, 2012, 2724-2727.
Conclusions
7
(a) C. E. Chan-Thaw, L. E. Chinchilla,; S. Campisi, G. A. Botton,
, 4189–
This sequence-driven approach for genome mining has given L. Prati, N. Dimitratos, A. Villa, ChemSusChem 2015,
8
4194. (b) J. C. González, M. F. Mangiameli, A. C. Asis, S. Bellú, L.
F. Sala, Polyhedron 2013, 49, 84–92. (c) S. N. Sidorov, I. V.
Volkov, V. A. Davankov, M. P. Tsyurupa, P. M. Valetsky, L. M.
Bronstein, R. Karlinsey, J. W. Zwanziger, V. G. Matveeva, E. M.
Sulman, N. V. Lakina, E. A. Wilder, R. J. Spontak J. Am. Chem. Soc.
access to several aldolases from the same pyruvate-aldolase
family (PF03328) active with HPA as nucleophile, including
GarL, HpcH, and RhmA from E. coli. One S,S stereoselective
“HPA aldolase” was used in organic synthesis, thus
demonstrating its utility as complementary tool to the existing 2001, 123, 10502–10510.
8
(a) B. Pokorny, S. Muller-Loennies, P. Kosma, Carbohydr. Res.
biocatalysts. To our knowledge these newly discovered
aldolases are the first biocatalysts described so far which are
able to add HPA on a variety of aldehydes thus offering a
stereoselective access to 3,4-dihydroxy-2-keto acids.
2014, 391, 66-81. (b) M. Reiner and R. R. Schmidt Tetrahedron:
Asymmetry 2000, 11, 319–335.
9
D. Gillingham, P. Stallforth, A. Adibekian, P. Seeberger, D.
Hilvert, Nat. Chem. 2010, , 102-105.
2
Remarkably, this aldol reaction has no equivalent in classical 10 (a) H. S. Chung, C. R. H. Raetz, Proc. Natl. Acad. Sci. U S A
2011, 108 510-515. (b) M. Hashimoto, M. Ozono, M.
,
organic synthesis. Using one of these new biocatalysts, seven
new compounds sharing the β-hydroxy- -ketoacid moiety
were prepared in one step with good yields and
stereoselectivity, and respecting the green chemistry
Furuyashiki, R. Baba, S. Hashiguchi, Y. Suda, K. Fukase, Y.
Fujimoto J. Biol. Chem. 2016, doi: 10.1074/jbc.M116.751271.
11 (a) Y. Lee, Y. Park, M. Kim, S. Jew, H. Park, J. Org. Chem.
2011, 76, 740-743; (b) M. Lingamurthy, A. Rajender, B. V. Rao,

principles. This study has efficiently extended the chemist’s Tetrahedron Lett. 2013, 54, 6189-6191.
12 P. Clapés in Science of Synthesis, Biocatalysis in Organic
Synthesis vol 2 (Eds K. Faber, W.-D. Fessner, N. J. Turner) Georg
Thiem Verlag, 2015, pp 31-82.
toolbox with versatile enzymes for use in sustainable organic
processes.
13 (a) G. Yang, Y. Ding Bioorg. Med. Chem. 2014, 22, 5604-5612.
(b) T. Davids, M. Schmidt, D. Bottcher, U. Bornscheuer, Curr.
Opin. Chem. Biol. 2013, 17, 215-220.
Acknowledgements
14 (a) M. Höhne, S. Schätzle, H. Jochens, K. Robins, U. T
This work was supported by the Agence Nationale de la
Recherche (ANR), Genozyme program (ANR-08-GENM-026).
We also wish to thank M. Besnard Gonnet, A. Debard and V.
Pellouin for excellent technical assistance and Prof. A.
Zaparucha for fruitful discussions. We are deeply indebted to
A. Tolonen for improving the manuscript.
Bornscheuer Nature Chem. Biol. 2010, 6, 807–813. (b) S. Zhao, R.
Kumar, A. Sakai, M. W. Vetting, B. M. Wood, S. Brown, J. B.
Bonanno, B. S. Hillrich, R. D. Seidel, P. C. Babbitt, S. C. Almo, J. V.
Sweedler, J. A. Gerlt, J. E. Cronan, M. P. Jacobson Nature, 2013,
502, 698-702. (c) K. Bastard, A. A. T. Smith, C. Vergne-Vaxelaire,
A. Perret, A. Zaparucha, R. De Melo-Minardi, A. Mariage, M.
Boutard, A. Debard, C. Lechaplais, C. Pelle, V. Pellouin, N.
Perchat, J.-L. Petit, A. Kreimeyer, C. Medigue, J. Weissenbach, F.
Artiguenave, V. De Berardinis, D. Vallenet, M. Salanoubat Nature
Chem. Biol., 2013, 10, 42-49.
Notes and references
15 C. Guérard-Hélaine, V. de Berardinis, M. Besnard-Gonnet, E.
Darii, M. Debacker, A. Debard, C. Fernandes, V. Hélaine, A.
Mariage, V. Pellouin, A. Perret, J.-L. Petit, M. Sancelme, M.
Lemaire, M. Salanoubat ChemCatChem, 2015, 7, 1871-1879.
16 (a) W. Wang, S. Seah, Biochemistry 2005, 44, 9447-9455; (b)
T. Izard, N. Blackwell, EMBO J. 2000, 19, 3849-3856.
§
The ulosonic acids are presented following Fischer and Cram
conventions, however several hemiketal cyclic forms were
detected on NMR spectra.
1
(a) D. Enders, A. A. Narine, J. Org. Chem. 2008, 73, 7858-
7870; (b) M. Markert, R. Mahrwald, Chem. Eur. J. 2008, 14, 40-
48; (c) J. Mlynarski, B. Gut, Chem. Soc. Rev. 2012, 41, 587-596.
17 (a) D. Rea, V. Fulop, T. Bugg, D. Roper, J. Mol. Biol. 2007, 373,
866-876; (b) W.-J. Wang, S. Y. K. Seah, FEBS Lett. 2008, 582,
3385-3388; (c) W.-J. Wang, P. J. Baker, S. Y. K. Seah,
Biochemistry 2010, 49, 3774-3782; (d) B. K. Hubbard, M. Koch,
2
(a) P. Clapés, X. Garrabou, Adv. Synth. Catal. 2011, 353,
2263-2283; (b) P. Clapés, W. D. Fessner, G. A. Sprenger, A. K.
Samland, Curr. Opin. Chem. Biol. 2010, 14, 154-167; (c) M.
Müller, Adv. Synth. Catal. 2012, 354, 3161-3174; (d) M. Brovetto,
D. R. J. Palmer, P. C. Babbitt, J. A. Gerlt, Biochemistry 1998, 37
,
D. Gamenara, P. Mendez, G. Seoane, Chem. Rev. 2011, 111
,
14369-14375; (e) D. Rea, R. Hovington, J. Rakus, J. Gerlt, V.
Fulop, T. Bugg, D. Roper, Biochemistry 2008, 47, 9955-9965; (f) J.
F. Rakus, A. A. Fedorov, E. V. Fedorov, M. E. Glasner, B. K.
Hubbard, J. D. Delli, P. C. Babbitt, S. C. Almo, J. A. Gerlt,
Biochemistry 2008, 47, 9944-9954; (g) S. V. Smirnov, N. N.
Samsonova, A. E. Novikova, N. G. Matrosov, N. Y. Rushkevich, T.
Kodera, J. Ogawa, H. Yamanaka, S. Shimizu, FEMS Microbiol.
Lett. 2007, 273, 70-77.
4346-4403; (e) A. Bolt, A. Berry, A. Nelson, Arch. Biochem.
Biophys. 2008, 474, 318-330; (f) A. K. Samland, G. A. Sprenger,
Appl. Microbiol. Biotechnol. 2006, 71, 253-264.
3
(a) M. Schürmann, G. A. Sprenger, J. Biol. Chem. 2001, 276,
11055-11061; (b) X. Garrabou, J. A. Castillo, C. Guérard-Hélaine,
T. Parella, J. Joglar, M. Lemaire, P. Clapés, Angew. Chem. Int. Ed.
2009, 48, 5521-5525; (c) M. Sugiyama, Z. Hong, P.-H. Liang, S. M.
Dean, L. J. Whalen, W. A. Greenberg, C.-H. Wong, J. Am. Chem.
Soc. 2007, 129, 14811-14817. (d) D. Güclü, A. Szekrenyi, X.
Garrabou, M. Kickstein, S. Junker, P. Clapés, W.-D. Fessner, ACS
18 (a) A. S. Serianni, H. A. Nunez, R. Barker, J. Org. Chem. 1980,
45, 3329-3341; (b) S.-Y. Han, M. M. Joullie, V. V. Fokin, N. A.
Petasis, Tetrahedron: Asymmetry 1994, 5, 2535-2562.
Catal. 2016,
6, 1848-1852. (e) M. Rale, S. Schneider, G.A.
8 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 9 of 10
Green Chemistry
Please do not adjust margins
Journal Name
ARTICLE
19 (a) H. J. Lamble, N. I. Heyer, S. D. Bull, D. W. Hough, M. J.
Danson J. Biol. Chem. 2003, 278, 34066–34072. (b) H. J. Lamble,
M. J. Danson, D. W. Hough, S. D. Bull, Chem. Commun. 2005,
View Article Online
DOI: 10.1039/C6GC02652D
124–126. (c) C. Augé, V. Delest Tetrahedron: Asymmetry 1993, 4,
1165-1168. (d) R. Plantier-Royon, F. Cardona, D. Anker, G.
Condemine, W. Nasser, J. Robert-Baudouy J. Carbohydr. Chem.
1991, 10, 787-811.
20 H. Hara, E. Masai, K. Miyauchi, Y. Katayama, M. Fukuda, J.
Bacteriol. 2003, 185, 41-50.
21 D. R. Hall, C. S. Bond, G. A. Leonard, C. I. Watt, A. Berry, W.
N. Hunter J. Biol. Chem. 2002, 277, 22018–22024.
22 (a) F. Pezzotti, M. Therisod, Carb. Res. 2006, 341, 2290-2292.
(b) J. Refaat, M.N. Samy, S.Y. Desoukey, M.A. Ramadan, S.
Sugimoto, K. Matsunami, M.S. Kamel, Med. Chem. Res. 2015, 24
2939-2949.
,
23 (a) P. J. Baker, S. Y. K. Seah, J. Am. Chem. Soc. 2012, 134, 507-
513; (b) W.-J. Wang, S. Mazurkewich, M. S. Kimber, S. Y. K. Seah,
J. Biol. Chem. 2010, 285, 36608-36615.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 9
Please do not adjust margins

Green Chemistry
Page 10 of 10
View Article Online
DOI: 10.1039/C6GC02652D
Expanding the reaction space of aldolases using hydroxypyruvate as nucleophile
substrate
V. de Berardinis, C. Guérard-Hélaine, E. Darii, K. Bastard, V. Hélaine, A. Mariage, J.-L.
Petit, N. Poupard, I. Sanchez-Moreno, M. Stam, T. Gefflaut, M. Salanoubat,* M. Lemaire*
An unprecedented biocatalytic activity consisting of converting hydroxypyruvate as
nucleophile substrate in an aldol reaction was revealed for the first time. To extend the
chemist’s toolbox with versatile enzymes, a sequence-driven approach to find new enzymes
from biodiversity was applied. Hydroxypyruvate was chosen in order to bring out efficient
aldolases that could be considered as complementary tools to the known dihydroxyacetone,
dihydroxyacetone phosphate and pyruvate-dependent aldolases. Several hits were found in
one Pfam family, one of them was chosen to demonstrate its potentiality as an efficient
stereoselective biocatalyst.