Synthesis of Sialic Acids, Their Derivatives, and Analogs by Using a Whole-Cell Catalyst

Article abstract of DOI:10.1002/chem.201703083

Sialic acids (Sias) are important constituents of cell surface glycans. Ready access to Sias in large quantities would facilitate the development of carbohydrate-based vaccines and small-molecule drugs. We now present a facile method for synthesizing various natural forms and non-natural derivatives or analogs of Sias by using a whole-cell catalyst, which is constructed by adding a plasmid containing necessary enzyme genes into a metabolically engineered strain of Escherichia coli. The flexible substrate tolerance of incorporated enzymes (N-acetylglucosamine 2-epimerase and N-acetylneuraminic acid aldolase) allows the cellular catalyst to convert a wide range of simple and inexpensive sugars into various Sia-related compounds through an easily scalable fermentation process. Further, syntheses using this whole-cell biotransformation in combination with three conventional enzymatic reactions provide a series of complex Sia-containing glycans (sialyloligosaccharides) and their derivatives bearing different substituents. The processes described herein should permit the large-scale and economical production of both Sias and sialyloligosaccharides, and may complement existing chemical and enzymatic strategies.

Full text of DOI:10.1002/chem.201703083

A Journal of
Accepted Article
Title: Synthesis of Sialic Acids, Their Derivatives and Analogues Using
a Whole-Cell Catalyst
Authors: Xun Lv, Hongzhi Cao, Baixue Lin, Wei Wang, Wande Zhang,
Qian Duan, Yong Tao, Xue-Wei Liu, and Xuebing Li
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: Chem. Eur. J. 10.1002/chem.201703083
Link to VoR: http://dx.doi.org/10.1002/chem.201703083
Supported by

10.1002/chem.201703083
Chemistry - A European Journal
FULL PAPER
Synthesis of Sialic Acids, Their Derivatives and Analogues Using
a Whole-Cell Catalyst
Xun Lv,a Hongzhi Cao,b Baixue Lin,c,*Wei Wang,d Wande Zhang,d Qian Duan,d Yong Tao,c Xue-Wei
Liue and Xuebing Lia,f,g,
*
Abstract: Sialic acids (Sias) are important constituents of cell
surface glycans. Ready access to Sias in large quantities would
facilitate the development of carbohydrate-based vaccines and
small-molecule drugs. We now present
a facile method for
synthesizing various natural forms and non-natural derivatives or
analogues of Sias using a whole-cell catalyst, which is constructed
by adding a plasmid containing necessary enzyme genes into a
metabolically engineered strain of Escherichia coli. The flexible
substrate tolerance of incorporated enzymes (N-acetylglucosamine
2-epimerase and N-acetylneuraminic acid aldolase) allows the
cellular catalyst to convert a wide range of simple and inexpensive
sugars into various Sia-related compounds through an easily
scalable fermentation process. Further, syntheses using this whole-
cell biotransformation in combination with three conventional
enzymatic reactions provide a series of complex Sia-containing
glycans (sialyloligosaccharides) and their derivatives bearing
different substituents. The processes described herein should permit
the large-scale and economical production of both Sias and
sialyloligosaccharides, and may complement existing chemical and
enzymatic strategies.
Scheme 1. Synthesis of Neu5Ac using E. coli ΔnanTEK/pNA whole-cell
catalyst. AGE: GlcNAc 2-epimerase (EC 5.1.3.8), gene from Anabaena PCC
7120; ATP: adenosine 5-triphosphate; NanA: Neu5Ac aldolase (EC 4.1.3.3),
gene from E. coli K-12; nanTEK:
a gene cluster encoding NanT (Sia
transporter), NanK (ManNAc kinase), and NanE (ManNAc-6-phosphate
epimerase); pNA: a plasmid carrying NanA and AGE genes.
found at the terminus of cell surface glycans. The structural and
functional diversity of both Sias and Sia-containing glycans
(sialyloligosaccharides) makes these carbohydrates important
synthetic targets for biological research and drug
development.[1,2] Chemical constructions of Sias require complex
protection and coupling strategies for the control of both the
stereochemistry and regiochemistry of bond formation.[36]
Syntheses using enzymes involved in natural Sia metabolism
are straightforward[716] and have been well developed to meet
the needs of mass production, as pioneered by the works of
Marukin Shoyu Co., Ltd. and Kyowa Hakko Bio Co., Ltd.[17,18]
However, the preparation of enzymes and necessary cofactors
requires tedious processes, which can make the enzymatic
approach expensive and limit its application. Another option is
the use of whole-cell catalysis that encapsulates exogenous
Introduction
Sialic acids (Sias) represent
a
class of acidic
monosaccharides with a nine-carbon skeleton that are typically
[a] Dr. X. Lv, Prof. Dr. X. Li
CAS Key Laboratory of Pathogenic Microbiology and Immunology,
Institute of Microbiology, Chinese Academy of Sciences (CAS)
Chaoyang District, Beijing 100101 (China)
E-mail: lixb@im.ac.cn
enzymatic
reactions
into
genetically
engineered
microorganisms.[1923] This method allows the in situ generated
enzymes, cofactors and intermediates to be directly employed
without purification and, therefore, is particularly attractive for
large-scale syntheses.
[b] Dr. H. Cao
National Glycoengineering Research Center, Shandong University
Jinan 250012 (China)
[c] Dr. B. Lin, Dr. Y. Tao
CAS Key Laboratory of Microbial Physiological and Metabolic
Engineering, Institute of Microbiology, Chinese Academy of Sciences
(CAS)
Chaoyang District, Beijing 100101 (China)
E-mail: linbaixue@126.com
We previously developed
a
recombinant bacterium
(Escherichia coli ΔnanTEK/pNA) to convert N-acetylglucosamine
(GlcNAc) into N-acetylneuraminic acid (Neu5Ac), the most
abundant Sia found in nature.[24] The desired synthetic pathway
in this living-cell catalyst (Scheme 1) was constructed by
incorporating a plasmid containing GlcNAc 2-epimerase (AGE)
and Neu5Ac aldolase (NanA) genes into a mutant form of E. coli
K-12 (BW25113), in which the endogenous nanTEK gene
cluster that is associated with Neu5Ac uptake and the competing
N-acetylmannosamine (ManNAc) catabolic pathway were
deleted. Cultivation of the engineered bacteria in the presence of
substrates (GlcNAc and pyruvate) at high concentrations drove
the reversible reactions toward the product, and multigram
quantities of Neu5Ac were obtained in the extracellular fraction
with high efficiency (74.2 g/L, 40% yield from GlcNAc).
[d] W. Wang, W. Zhang, Dr. Q. Duan
School of Materials Science and Engineering, Changchun University of
Science and Technology
Weixing Road, Changchun 130022 (China)
[e] Prof. Dr. X. W. Liu
School of Physical and Mathematical Sciences, Nanyang
Technological University
Singapore 637371 (Singapore)
[f] Prof. Dr. X. Li
Savaid Medical School, University of Chinese Academy of Sciences
Huairou District, Beijing 101408 (China)
[g] Prof. Dr. X. Li
Center for Influenza Research and Early-warning (CASCIRE), Chinese
Academy of Sciences
Chaoyang District, Beijing 100101 (China)
Supporting information for this article is given via a link at the end of
the document.((Please delete this text if not appropriate))
Herein, we report the use of E. coli ΔnanTEK/pNA to
synthesize various natural Sias and non-natural Sia derivatives
This article is protected by copyright. All rights reserved.

10.1002/chem.201703083
Chemistry - A European Journal
FULL PAPER
Figure 1. Structures of sugar substrates (A), epimerization (B) and Sia-related products (C). All of these carbohydrates are the D-enantiomers. Compounds
marked with an asterisk are naturally occurring.
or analogues from simple starting materials (Figure 1). We also
describe an efficient approach for the synthesis of
sialyloligosaccharides using this engineered bacterium in
and the results are summarized in Table 1. In addition to its
natural substrates GlcNAc (1) and ManNAc (10), AGE accepted
a range of GlcNAc derivatives (28) with modifications at
different positions of the sugar backbone, affording their
corresponding epimerization products (1824) in 9%35%
equilibrium yields. No product formation was confirmed for the
AGE-catalysed reactions of all the non-nitrogenous sugars
tested (1116), in agreement with the catalytic mechanism of
AGE reported previously.[28,29] In contrast to AGE, NanA
accepted ManNAc and the non-nitrogenous sugars, but not
GlcNAc and its derivatives as substrates, consistent with
previous studies.[79,15] The aldol addition reactions of 1013 with
pyruvate catalysed by NanA gave isomerically pure Neu5Ac (25),
deaminoneuraminic acid (KDN, 34), and KDN analogues 35 and
36 in high yields, respectively, whereas isomeric mixtures of
products were observed in the reactions with starting aldoses
1416. An inversion of stereoselectivity during the nucleophilic
attack of pyruvate at the anomeric centre of the aldose may
occur in Sia aldolase-catalysed reactions.[8,9] While the isomers
in these mixtures were not identified in the present study, the
above results demonstrate that both AGE and NanA have fairly
flexible substrate tolerances. Additionally, neither AGE nor NanA
accepted N-acetylgalactosamine (GalNAc, 17) as a substrate.
combination with three enzymatic catalysts,
a CMP-Sia
synthetase and two sialyltransferases. The present study is
based on previous findings that many carbohydrate-processing
enzymes, particularly those from microbial sources, have broad
substrate scope and have been extensively used for the
construction of diverse carbohydrate-related molecules.[2527]
Results and Discussion
The substrate specificity of purified AGE and NanA was first
investigated to ensure that the E. coli ΔnanTEK/pNA can convert
structurally diverse aldoses. In nature, AGE catalyses the 2-
epimerization of GlcNAc to form ManNAc using adenosine 5-
triphosphate (ATP) as a cofactor and NanA catalyses the aldol
addition reaction between ManNAc and pyruvate to generate
Neu5Ac. Both reactions are reversible. The characteristics of
substrates acceptable to both AGE[28] and NanA[79,15] have been
well studied previously. Sixteen aldoses were examined here
This article is protected by copyright. All rights reserved.

10.1002/chem.201703083
Chemistry - A European Journal
FULL PAPER
Table 1. Substrate specificity of AGE and NanA.[a]
Table 2. Synthesis of Sias and their derivatives or analogues catalysed by
E. coli ΔnanTEK/pNA.[a]
Product (% yield)[b]
Substrate
Product
Yield (%)[b]
Substrate
AGE
10 (22)
18 (27)
19 (35)
20 (14)
21 (18)
22 (18)
23 (33)
24 (9)
1 (80)

NanA
AGE + NanA
1
2
25
26
40
45
36
32
39
42
55
48
44
39
44
51
36
1

25 (37)
2

26 (36)
3
27
3

27 (22)
4
28
4

28 (18)
5
29
5

29 (25)
6
30
6

30 (42)
7
31
7

31 (40)
8
32
8

32 (13)
9
33
10
11
12
13
14
15
16
17
25 (83)








11
12
13
14
34
34 (92)
35

35 (80)
36

36 (80)
37 + 38

37 + 38 (90)[c]
mixture (63)[c]
mixture (61)[c]

[a] Reactions were typically carried out by cultivating bacteria (OD600 nm = 30)
in the presence of sugar substrate (0.2 mmol) and pyruvate (5 equiv) in
TrisHCl buffer (2 mL). [b] Isolated yields. Products were characterized by
MS and NMR spectroscopy. Characterization data are provided in the SI.



and pyruvate) with the endogenous supply of ATP. A high
concentration of substrates was used to drive the reversible
reactions toward product formation. This engineered bacterium
transformed 19 and 1114 into different natural Sia forms
(2527, 34), their derivatives (2833) and analogues (3538)
(Table 2). In the conversion of GlcNAc (1) and its derivatives
(28), the yields from whole-cell catalysis (32%55%) were
generally higher than those of the corresponding in vitro
enzymatic reactions (13%42%, Table 1, column ‘AGE + NanA’).
This increased catalytic efficiency could primarily be attributed to
the knockout of the nanT gene in the bacteria,[24] which resulted
in the abrogation of uptake of extracellularly transported
products and thus ensured the progress of intracellular reactions
towards the desired direction. The removal of the nanE and
nanK genes in the bacteria could prevent the requisite
epimerization intermediates from entering the bacteria-intrinsic
ManNAc metabolic pathways. This also contributed to
improvement in the efficiency of cell-based conversions of 18.
However, the whole-cell catalysis of non-nitrogenous aldoses
1114 showed much lower yields (36%51%) than the
corresponding enzymatic approaches (80%92%, Table 1,
column ‘NanA’). In addition to the exogenously incorporated Sia-
synthetic pathway, the endogenous metabolic machinery of E.
coli also employs Man (11) or arabinose (Ara, 14) as sources of
carbon and energy.[30,31] It is very likely that their
stereochemically relevant analogues lyxose (Lyx, 12) and 2-
deoxy-glucose (2-DG, 13) were similarly consumed during
[a] Reactions were carried out by incubating the substrate in TrisHCl buffer
(pH 8.0) in the presence of AGE (+ATP), NanA (+pyruvate), or both
enzymes (+ATP and pyruvate). [b] Equilibrium yields were determined by
HPLC. Chromatograms are provided in the Supporting Information (SI). : no
reaction; : not tested. [c] An isomeric product mixture was determined by
MS and NMR spectroscopy after HPLC isolation. Previous reports have
confirmed that the product of aldol addition reaction with Ara (14) is
composed of 3-deoxy-D-manno-2-octulosonic acid (KDO, 37) and its 4-
epimer (38).[8,9]
A one-pot reaction combining the catalysts AGE and NanA
was next examined for its ability to synthesise Sia-related
products from GlcNAc (1) and its derivatives (28). All
compounds 18 could be converted into the expected Neu5Ac
(25) and its derivatives (2632) through this one-pot approach in
13%42% equilibrium yields (Table 1, column ‘AGE + NanA’).
Since NanA alone did not accept 18 as substrates, the one-pot
reaction must proceed via the intermediates 10 and 1824,
which were generated in the AGE-catalysed epimerization
reactions and then coupled in situ to pyruvate in the NanA-
catalysed aldol addition reactions. Taken together with the result
that NanA directly converted non-nitrogenous sugars 1116, this
study of combined AGENanA catalysis provided a basis to
investigate the E. coli ΔnanTEK/pNA for the synthesis of various
Sias and their derivatives or analogues.
The scope of E. coli ΔnanTEK/pNA-catalysed reactions was
explored using the exogenous addition of substrates (sugars
This article is protected by copyright. All rights reserved.

10.1002/chem.201703083
Chemistry - A European Journal
FULL PAPER
normal cell metabolism, leading to decreased efficiencies of cell-
based transformations of these compounds relative to isolated
enzymatic reactions. Nevertheless, the conversion yields of all
sugar substrates tested for the whole-cell catalysis were around
40% (for the space-time yields, see the Table S1). For example,
74.2 g/L of Neu5Ac (25) was afforded from an initial input of
132.7 g/L GlcNAc (1). These values represent the highest level
reported to date for cell-based generation of Neu5Ac.[1923] We
therefore conclude that the E. coli ΔnanTEK/pNA is an efficient
biocatalyst for the synthesis of various Sia-related compounds.
This whole-cell catalysis has particular advantages for
large-scale preparation. The E. coli ΔnanTEK/pNA can be
constructed quickly and economically, as its precursor E. coli K-
12 is a routinely used biological engineering tool with many
genetically modified variants readily available.[32] Further, a large
number of E. coli variants have been used as platforms for
industrial production of diverse molecules and their fermentation
processes have been well developed.[33,34] Most of the whole-cell
reactions in this study were carried out on milligram scales.
However, for the synthesis of several natural Sias, Neu5Ac (25),
N-glycolylneuraminic acid (Neu5Gc, 26), 9-O-acetyl-Neu5Ac
(Neu5,9Ac2, 27), and KDN (34), the reactions were scaled up to
gram levels without apparent decreases in catalytic efficiencies.
We also confirmed that the whole-cell catalyst could be reused
at least 5 times without a significant loss of activity.
The Sia-related products listed in Table 2 may be very
useful standards or probes for relevant biological research and
drug development. Neu5Ac (25) is a common distal end moiety
of mammalian glycoproteins and glycolipids that are involved in
many physiological and pathological processes such as cellular
recognition and tumorigenesis.[1] Neu5Gc (26) is distributed in
non-human mammals but can be metabolically incorporated by
human cells from dietary sources, which results in human
susceptibility to bacterial toxins.[35,36] Neu5,9Ac2 (27) is found in
higher animals and serves as the cell-surface receptor of
influenza C and D viruses.[37,38] KDN (34) occurs in bacteria or
lower vertebrates and is associated with the binding of trout
sperm to egg.[39,40] The eight-carbon Sia analogue KDO (37) is a
component of lipopolysaccharides in Gram-negative bacteria
and is essential to bacterial viability.[41] The non-natural Neu5Ac
derivatives (2833) with substitutions at different positions of the
sugar skeleton can be used as probes in the study of sialidases,
sialyltransferases, and Sia-binding proteins.[25] Complementary
to existing chemical and enzymatic methods,[316] the whole-cell
catalysis illustrated above provides an attractive route to readily
access these valuable compounds.
(4559, Table 3) with a sialyl α2,3- or α2,6-linkage in 66%98%
isolated yield. Among these products, the Neu5Ac-, Neu5Gc-,
Neu5,9Ac2-, and KDN-containing structures (4553 and 59)
represent sequences found in natural glycoconjugates, and the
oligosaccharides derivatives carrying an azide (55) or an alkynyl
group (4750 and 5259) at a specific position on the sugar
carbon backbone can be selectively manipulated downstream in
a chemical synthesis through “click” ligation.[45]
Scheme 2. Synthesis of sialyloligosaccharides using E. coli ΔnanTEK/pNA,
CMP-Sia synthetase (NmCSS), and sialyltransferase (Pd2,6ST or Pm2,3ST)
as catalysts. The cell-catalysed reactions gave Sia-containing mixtures in
TrisHCl buffer after removing the bacteria by centrifugation. Incubation of
these mixtures in the presence of an acceptor, NmCSS, CTP, and Pd2,6ST
(or Pm2,3ST) afforded sialyloligosaccharide products. CMP: cytidine 5'-
monophosphate; CTP: cytidine 5'-triphosphate; PPi: inorganic pyrophosphate.
At present, the large-scale synthesis of such complex
sialyloligosaccharides primarily relies on enzymatic processes,
which typically use ManNAc or its derivatives as starting
materials, and Neu5Ac aldolase, CMP-Sia synthetase, and
sialyltransferase as a catalyst combination.[43,44,46] ManNAc is
commercially available, but its present price is even higher than
that of Neu5Ac (Carbosynth Ltd, Berkshire, UK). This can make
these enzymatic approaches expensive and limit their
application. A GlcNAc 2-epimerase such as AGE could be
included into the catalyst combination to confer the ability to
synthesize sialyloligosaccharides from the inexpensive GlcNAc.
Unfortunately, these enzymes catalyse the reversible conversion
between ManNAc and GlcNAc, with a preference for GlcNAc
formation.[47] The inefficiency of GlcNAc 2-epimerases to
generate ManNAc, as also observed for AGE in this study
(Table 1, column ‘AGE’), precludes their direct application to the
enzymatic synthesis of sialyloligosaccharides. However, our
approach that uses a combined catalysis of whole cell and
conventional enzymes enabled various sialyloligosaccharides to
be prepared in large amounts directly from GlcNAc and the
readily available derivatives of GlcNAc.
The usefulness of E. coli ΔnanTEK/pNA was further
explored
in
this
study
with
the
construction
of
sialyloligosaccharides (Scheme 2). Synthesis was initiated with
the cell-catalysed transformation of a Sia precursor into a Sia,
which was then converted into its nucleotide-activated form
using CMP-Sia synthetase NmCSS,[42] followed by the transfer
of Sia to a glycosyl acceptor using sialyltransferase Pd2,6ST[43]
or Pm2,3ST.[44] The whole-cell and enzymatic reactions were
carried out consecutively in TrisHCl buffer without isolating the
intermediates. Combinations of 8 Sia precursors with 6 glycosyl
acceptors (3944, Figure S1) afforded 15 oligosaccharides
Conclusions
The E. coli ΔnanTEK/pNA represents a new biocatalyst for the
facile synthesis of natural Sias as well as their non-natural
derivatives or analogues. Coupled with a CMP-Sia synthetase
This article is protected by copyright. All rights reserved.

10.1002/chem.201703083
Chemistry - A European Journal
FULL PAPER
and two sialyltransferases, this whole-cell catalyst also allows a
range of sialyloligosaccharides to be prepared efficiently and
economically. The catalytic processes described in this study
are easy to scale up, and the large-scale syntheses of both Sias
and sialyloligosaccharides should benefit their use in biological
and pharmaceutical applications. Given the distinct advantages
of microbial fermentation in the mass production of biomolecules,
engineering of new bacteria-based catalysts that can directly
Table 3. Structure and yield of sialyloligosaccharide products.[a]
Precursor
Product
Yield (%)[b]
NO2
NO2
45
OH
O2C HO
HO
OH
OH
O
O
O
O
1
1
1
1
98
98
95
95
AcHN
HO
O
HO
O
OH
HO
OH
OH
O2C HO
HO
OH
O
synthesize
sialyloligosaccharides
and
even
sialylated
O
O
HO
O
HO
O
AcHN
HO
O
glycoconjugates from simple materials represents a challenge
for the future. Indeed, research efforts along these lines are
ongoing.[48,49]
46
47
OH
HO
OH
OH
HO
O2C HO
O
O
O
O
O
AcHN
HO
HO
HO
OH
NHAc
O2C
HO
HO
Experimental Section
O
O
48
AcHN
HO
HO
O
HO
AcHN
General Information
OH
O2C HO
HO
OH
O
OH
O
O
O
O
O
O
O
1
1
92
95
AcHN
HO
O
All chemical regents and solvents, including compounds 1, 2, 8, and
1017, were obtained from commercial source and used as purchased,
unless otherwise noted. The GlcNAc derivatives 3,[50] 4,[51] 5,[52] 6[53] and
9[54] were synthesized as described previously, and 7 was synthesized as
described in the SI. The glycosyl acceptors (Figure S1), 39,[55] 40,[56]
41,[57] 42,[58] 43,[55] and 44,[59] were synthesized as described previously.
All aqueous solutions were prepared using ultrapure water (≥18 MΩ,
Milli-Q, Millipore). TLC analysis was performed on the Merck silica gel
HO
49
OH
HO
OH
OH
O2C HO
HO
OH
O
O
O
HO
O
HO
AcHN
HO
50
OH
HO
OH
OH
HO
CO2
O
AcHN
HO
NO2
HO
1
2
OH
O
92
89
OH
O
glass plates 60F254
.
Sephadex G-10 used for gel-filtration
HO
O
O
HO
O
HO
chromatography was purchased from GE Healthcare (China). Silica gel
(200–300 mesh) used for column chromatography was obtained from
Qingdao Haiyang Chemical Co., Ltd. 1H NMR (500 MHz) and 13C NMR
(125 MHz) spectra were measured on a Bruker Advance spectrometer
with the signal of deuterated solvents as the internal standard. ESIMS
and HR-ESIMS spectra were recorded on a Bruker Apex IV FTMS or an
Agilent 6520 QTOFMS spectrometer. HPLC was performed on an Agilent
1200 system coupled with a UV detector (detection at 210 nm). Optical
rotations were measured on a Perkin-Elmer 241 polarimeter.
HO
O
51
OH
O2C HO
OH
OH
OH
HO
OH
O
OH
H
O
O
O
O
O
N
HO
HO
HO
NHAc
OH
OH
O
OH
OH
52
53
54
O2C HO
AcO
OH
O
O
O
HO
O
O
3
4
5
6
72
92
66
90
AcHN
HO
O
NHAc
HO
OH
OH
O2C HO
HO
OH
O
O
O
HO
O
HO
O
O
O
O
O
Bacteria and Enzymes
AcHN
F
NHAc
OH
OH
OH
N3
O2C HO
The method used for the construction of E. coli ΔnanTEK/pNA has been
detailed previously[24] and is briefly summarized in the SI. The
procedures used for the preparation of recombinant enzymes (AGE,[24]
OH
O
O
O
HO
O
O
O
O
AcHN
HO
HO
NHAc
HO
OH
55
56
57
58
59
OH
NanA,[24] NmCSS,[42] Pd2,6ST,[43] Pm2,3ST[44]
) have been detailed
F
O2C HO
OH
O
OH
O
previously and are also briefly described in the SI.
O
HO
AcHN
HO
O
NHAc
HO
OH
AGE-Catalysed Epimerization Reaction
OH
O2C HO
F
OH
O
O
O
HO
O
HO
6
7
88
82
AcHN
HO
To a solution of a sugar substrate (18, 1012 or 1417, 0.40.6 mg, 2
μmol), ATP (0.1 mg, 0.2 μmol) and MgCl2 (0.4 mM) in TrisHCl buffer (50
mM, pH 8.0, 100 μL) was added AGE (0.1 U). The mixture was incubated
at 37°C for 6 h. The reaction was then quenched by heating at 100°C for
5 min. After removal of precipitates from the mixture by centrifugation
(6000 rpm, 30 min, 4°C), the supernatant was analysed by HPLC to
determine the equilibrium yield of product (1, 10 and 1824), which
ranged from 9% to 80% as summarised in Table 1. Chromatographic
conditions: Bio-Rad Aminex HPX-87H column (300 × 7.8 mm), aqueous
sulfuric acid solution (6 mM) as eluent, flow rate at 0.3 or 0.35 mL/min,
UV detection at 210 nm.
NHAc
HO
HO
OH
OH
O2C HO
OH
O
OEt
H
O
O
HO
O
O
O
N
HO
HO
NHAc
OH
O
OH
OH
O2C HO
HO
OH
O
O
O
HO
O
O
11
92
O
HO
HO
HO
NHAc
OH
OH
[a] Reactions were carried out generally on a preparative scale using 40-
μmol acceptors and 5-equiv Sia precursors, except for the syntheses of 49
and 50, which were scaled up by 100-fold and gave these compounds in
gram quantities. [b] Isolated yield based on acceptor.
This article is protected by copyright. All rights reserved.

10.1002/chem.201703083
Chemistry - A European Journal
FULL PAPER
NanA-Catalysed Aldol Addition Reaction
reaction was complete. The reaction was then quenched by addition of
an aqueous ethanol solution (95%, 2 mL) and incubation at 4°C for 30
min. The precipitates were removed from the mixture by centrifugation
(6000 rpm, 30 min, 4°C) and the supernatant was passed through a
Sephadex G-10 column eluted with deionised water to yield compounds
4559 (1934 mg, 66%98%), which was characterised by NMR and MS
analyses. For the synthesis of 46 and 47, the reactions were also carried
out on the gram scale. The details of reaction scale and product yield are
provided in Table S2 (in the SI).
A solution of a sugar substrate (18 or 1017, 0.40.6 mg, 2 μmol),
sodium pyruvate (1.1 mg, 10 μmol, 5.0 equiv) and NanA (0.1 U) in
TrisHCl buffer (50 mM, pH 8.0, 100 μL) was incubated overnight at
37°C. The reaction was quenched by adding an aqueous ethanol solution
(95%, 100 μL) followed by incubation at 0°C for 30 min. The precipitates
were then removed from the mixture by centrifugation (6000 rpm, 30 min,
4°C) and the supernatant was analysed by HPLC to determine the
equilibrium yield of product (25, 3436 and the isomeric product
mixtures), which ranged from 61% to 92% as summarised in Table 1.
The isomeric product mixtures were further isolated by HPLC for the
NMR and MS analyses. The HPLC was performed by using the same
conditions as described for the AGE-catalysed epimerization reaction.
Acknowledgements
This work was financially supported by the National Key R & D
Program of China (Grant No. 2017YFD0500200), the National
Natural Science Foundation of China (Grant No. 81661148055
and 31470801), the Open Project Program of the State Key
Laboratory of Microbial Resources, Institute of Microbiology,
CAS (Grant No. SKLMR-20160604), and the Open Project
Program of CAS Key Laboratory of Microbial Physiological and
Metabolic Engineering, Institute of Microbiology, CAS.
One-Pot Reaction Catalysed by the Combination of AGE and NanA
To a solution of a sugar substrate (18, 0.40.6 mg, 2 μmol), ATP (0.1
mg, 0.2 μmol) and MgCl2 (0.4 mM) in TrisHCl buffer (50 mM, pH 8.0,
100 μL) was added AGE (0.1 U). The mixture was incubated at 37°C for
6 h followed by addition of sodium pyruvate (1.1 mg, 10 μmol, 5.0 equiv)
and NanA (0.1 U), and further incubation at 37°C overnight, when TLC
analysis (i-PrOH/H2O/HOAc, 14:6:3, v/v) indicated that reaction was
complete. The reaction was then quenched by addition of an aqueous
ethanol solution (95%, 100 μL) and incubation at 0°C for 30 min. The
precipitates were removed from the mixture by centrifugation (6000 rpm,
30 min, 4°C) and the supernatant was analysed by HPLC to determine
the equilibrium yield of product (2532), which ranged from 13% to 42%
as summarised in Table 1. The HPLC was performed by using the same
conditions as described for the AGE-catalysed epimerization reaction.
Conflict of interest
The authors declare no conflict of interest.
Keywords: biosynthesis • Escherichia coli • N-
acetylglucosamine 2-epimerase • N-acetylneuraminic acid
aldolase • sialic acids
Synthesis of Sia-related Compounds (2538) Catalysed by E. coli
ΔnanTEK/pNA
E. coli ΔnanTEK/pNA was grown in Luria–Bertani medium (pH 7.0)
containing ampicillin (50 μg/mL) with shaking (250 rpm) at 37°C overnight.
The bacteria were then inoculated (1% inoculum) into an auto-induction
ZYM medium[60] followed by incubation on a shaker (250 rpm) at 37oC for
16 h to induce the expression of AGE and NanA. After collection by
centrifugation (8000 rpm, 5 min, 4 °C) and washing with aqueous 0.85%
NaCl solution, the induced bacteria were mixed with a sugar substrate
(19 or 1114), sodium pyruvate, and MgCl2 (10 mM) in TrisHCl buffer
(100 mM, pH 8.0, bacterial concentration: OD600 nm = 30). The reaction
mixture was incubated on a shaker (200 rpm) at 30°C for 16 h. The
bacteria were then removed from the mixture by centrifugation (6000 rpm,
30 min, 4°C) and the supernatant was passed through a Sephadex G-10
column eluted with deionised water to afford compounds 2538, which
was characterised by NMR and MS analyses. The details of reaction
scale and product yield are provided in Table S1 (in the SI).
[1]
[2]
[3]
N. M. Varki, A. Varki. Lab Invest. 2007, 87, 851857.
X. Chen, A. Varki. ACS Chem. Biol. 2010, 19, 163176.
R. Lorpitthaya, S. B. Suryawanshi, S. Wang, K. K. Pasunooti, S. Cai, J.
Ma, X. W. Liu. Angew. Chem. Int. Ed. 2011, 50, 1205412057; Angew.
Chem. 2011, 123, 1226012263.
Z. Hong, L. Liu, C. C. Hsu, C. H. Wong. Angew. Chem. Int. Ed. 2006,
45, 74177421; Angew. Chem. 2006, 118, 75777581.
S. D. Burke, E. A. Voight. Org. Lett. 2001, 3, 237240.
E. A. Voight, C. Rein, S. D. Burke. J. Org. Chem. 2002, 67, 84898499.
M. J. Kim, W. J. Hennen, H. M. Sweers, C. H. Wong. J. Am. Chem.
Soc. 1988, 110, 64816486.
[4]
[5]
[6]
[7]
[8]
[9]
C. H. Lin, T. Sugai, R. L. Halcomb, Y. Ichikawa, C. H. Wong. J. Am.
Chem. Soc. 1992, 114, 1013810145.
C. Augé, B. Bouxom, B. Cavayé, C. Gautheron. Tetrahedron Lett.
1989, 30, 22172220.
[10] S. Hu, J. Chen, Z. Yang, L. Shao, H. Bai, J. Luo, W. Jiang, Y. Yang.
Appl. Microbiol. Biotechnol. 2010, 85, 13831391.
[11] I. Maru, J. Ohnishi, Y. Ohta, Y. Tsukada. Carbohydr. Res. 1998, 306,
575578.
Synthesis of Sialyloligosaccharides (4559) Catalysed by E. coli
ΔnanTEK/pNA, NmCSS and Pd2,6ST (or Pm2,3ST)
[12] O. M. Pearce, A. Varki. Carbohydr. Res. 2010, 345, 12251229
[13] U. Kragl, D. Gygax, O. Ghisalba, C. Wandrey. Angew. Chem. Int. Ed.
1991, 30, 827828; Angew. Chem. 1991, 103, 854855.
[14] J. O. Lee, J. K. Yi, S. G. Lee, S. Takahashi, B. G. Kim. Enzyme Microb.
Technol. 2004, 35, 121125.
A reaction mixture containing the induced E. coli ΔnanTEK/pNA (OD600 nm
= 30), a Sia donor precursors (17 or 11, 3060 mg, 0.2 mmol), sodium
pyruvate (110 mg, 1.0 mmol, 5.0 equiv) and MgCl2 (10 mM) in TrisHCl
buffer (100 mM, pH 8.0, 2 mL) was incubated on a shaker (200 rpm) at
30°C for 16 h. The bacteria were then removed from the mixture by
centrifugation (6000 rpm, 30 min, 4°C) to give a supernatant, to which
were added a glycosyl acceptor (3944, 10–25 mg, 0.04 mmol), CTP (32
mg, 0.06 mmol), NmCSS (0.2 U) and Pd2,6ST (or Pm2,3ST, 0.2 U). The
mixture was further incubated with shaking (140 rpm) at 25°C for 2 h,
when TLC analysis (EtOAc/MeOH/H2O/HOAc, 8:4:2:1, v/v) indicated that
[15]
Q. Chen, L. Han, X. Chen, Y. Cui, J. Feng, Q. Wu, D. Zhu. Enzyme
Microb. Technol. 2016, 92, 99106.
[16] P. Laborda, S. Y. Wang, A. M. Lu, M. He, X. C. Duan, Y. J. Qian, Y. S.
Jung, L. Liu, J. Voglmeir. Adv. Synth. Catal. 2017, DOI:
10.1002/adsc.201700678.
This article is protected by copyright. All rights reserved.

10.1002/chem.201703083
Chemistry - A European Journal
FULL PAPER
[17] Y. Tsukada, Y. Ohta. (Marukin Shoyu Co., Ltd.), EP 0578825 B1, 1997.
[18] S. Koizumi, K. Tabata, T. Endo, A. Ozaki. (Kyowa Hakko Bio Co., Ltd.),
US 2003/0109007 A1, 2003.
[41] F. M. Unger. Adv. Carbohydr. Chem. Biochem. 1981, 38, 323–388.
[42] H. Yu, H. Yu, R. Karpel, X. Chen. Bioorg. Med. Chem. 2004, 12,
64276435.
[19] P. Xu, J. H. Qiu, Y. N. Zhang, J. Chen, P. G. Wang, B. Yan, J. Song, R.
M. Xi, Z. X. Deng, C. Q. Ma. Adv. Synth. Catal. 2007, 349, 16141618.
[20] K. Tabata, S. Koizumi, T. Endo, A. Ozaki. Enzyme Microb. Technol.
2002, 30, 327333.
[21] M. Ishikawa, S. Koizumi. Carbohydr. Res. 2010, 345, 26052609.
[22] F. Tao, Y. Zhang, C. Ma, P. Xu. Sci. Rep. 2011, 1, 142.
[23] Y. C. Lee, H. C. Chien, W. H. Hsu. J. Biotechnol. 2007, 129, 453460
[24] B. X. Lin, Z. J. Zhang, W. F. Liu, Z. Y. Dong, Y. Tao. Appl. Microbiol.
Biotechnol. 2013, 97, 47754784.
[25] C. C. Yu, S. G. Withers. Adv. Synth. Catal. 2015, 357, 16331654.
[26] J. M. Langenhan, B. R. Griffith, J. S. Thorson. J. Nat. Prod. 2005, 68,
16961711.
[27] C. J. Thibodeaux, C. E. Melançon, H. W. Liu. Angew. Chem. Int. Ed.
2008, 47, 98149859; Angew. Chem. 2008, 120, 996010007.
[43] H. Yu, S. Huang, H. Chokhawala, M. Sun, H. Zheng, X. Chen. Angew.
Chem. Int. Ed. 2006, 45, 39383944; Angew. Chem. 2006, 118,
40424048.
[44] H. Yu, H. Chokhawala, R. Karpel, H. Yu, B. Wu, J. Zhang, Y. Zhang, Q.
Jia, X. Chen. J. Am. Chem. Soc. 2005, 127, 1761817619.
[45] V. V. Rostovtsev, L. G. Green, V. V. Fokin, K. B. Sharpless. Angew.
Chem. Int. Ed. 2002, 41, 2596−2599; Angew. Chem. 2002, 114,
27082711.
[46] O. Blixt, J. C. Paulson. Adv. Synth. Catal. 2003, 345, 687−690.
[47] S. Ghosh, S. Roseman. J. Biol. Chem. 1965, 240, 15311536.
[48] T. Antoine, A. Heyraud, C. Bosso, E. Samain. Angew. Chem. Int. Ed.
2005, 44, 13501352; Angew. Chem. 2005, 117, 13741376.
[49] N. Fierfort, E. Samain. J. Biotechnol. 2008, 134, 261265.
[50] G. C. Look, Y. Ichikawa, G. J. Shen, P. W. Cheng, C. H. Wong. J. Org.
Chem. 1993, 58, 4326–4330.
[28]
S. Y. Wang, P. Laborda, A. M. Lu, X. C. Duan, H. Y., Ma, L. Liu, J.
Voglmeir. Catalysts 2016, 6, 212.
[51] S. I. Nishimura, M. Hato, S. Hyugaji, F. Feng, M. Amano. Angew. Chem.
Int. Ed. 2012, 51, 3386–3390; Angew.Chem. 2012, 124, 3442–3446.
[52] K. Ikeda, F. Kimura, K. Sano, Y. Suzuki, K. Achiwa. Carbohydr. Res.
1998, 312, 183–189.
[29] S. Ghosh, S. Roseman. J. Biol. Chem. 1965, 240, 15311536.
[30] A. Markovitz, R. J. Sydiskis, M. M. Lieberman. J. Bacteriol. 1967, 94,
14921496.
[31] D. J. LeBlanc, R. P. Mortlock. J. Bacteriol. 1971, 106, 9096.
[32] J. H. Miller. Trends Biochem. Sci. Libr. Compend. 1993, 18, 193.
[33] S. Y. Lee. Trends Biotechnol. 1996, 14, 98105.
[53] M. L. Shulman, A. Y. Khorlin. Carbohydr. Res. 1973, 27, 141–147.
[54] L. Cai, W. Y. Guan, M. Kitaoka, J. Shen, C. F. Xia, W. L. Chen, P. G.
Wang. Chem. Commun. 2009, 20, 2944–2946.
[34] X. Chen, L. Zhou, K. Tian, A. Kumar, S. Singh, B. A. Prior, Z. Wang.
Biotechnol. Adv. 2013, 31, 12001223.
[55] J. Wei, X. Lv, Y. Lü, G. Yang, L. Fu, L. Yang, J. Wang, J. Gao, S.
Cheng, Q. Duan, C. Jin, X. B. Li, Eur. J. Org. Chem. 2013, 12, 2414–
2419.
[35] E. Byres, A. W. Paton, J. C. Paton, J. C. Löfling, D. F. Smith, M. C.
Wilce, U. M. Talbot, D. C. Chong, H. Yu, S. Huang, X. Chen, N. M.
Varki, A. Varki, J. Rossjohn, T. Beddoe. Nature 2008, 456, 648652.
[36] M. J. Martin, A. Muotri, F. Gage, A. Varki. Nat. Med. 2005, 11, 228232.
[37] G. N. Rogers, G. Herrler, J. C. Paulson, H. D. Klenk. J. Biol. Chem.
1986, 261, 59475951.
[56] B. Thomas, C. Pifferi, G. C. Daskhan, M. Fiore, B N. erthet, O.
Renaudet. Org. Biomol. Chem. 2015, 13, 11529–11538.
[57] K. K. Yeoh, T. D. Butters, B. L. Wilkinson, A. J. Fairbanks. Carbohydr.
Res. 2009, 344, 586–591.
[58] L. Moni, A. Ciogli, I. D'Acquarica, A. Dondoni, F. Gasparrini, A. Marra.
Chem. Eur. J., 2010, 16, 5712–5722.
[38] H. Song, J. Qi, Z. Khedri, S. Diaz, H. Yu, X. Chen, A. Varki, Y. Shi, G. F.
Gao. PLoS Pathog. 2016, 12, e1005411.
[59] R. Daly, G. Vaz, A. M. Davies, M. O. Senge, E. M. Scanlan. Chem. Eur.
J., 2012, 18, 14671–14679.
[39] S. Inoue, K. Kitajima. Glycoconj. J. 2006, 23, 277290
[40] S. Yu, N. Kojima, S. I. Hakomori, S. Kudo, S. Inoue, Y. Inoue. Proc.
Natl Acad. Sci. USA 2002, 99, 28542859.
[60] F. W. Studier. Protein Expres. Purif., 2005, 41, 207234.
This article is protected by copyright. All rights reserved.

10.1002/chem.201703083
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents
FULL PAPER
Xun Lv, Hongzhi Cao, Baixue Lin, Wei
Wang, Wande Zhang, Qian Duan, Yong
Tao, Xue-Wei Liu, Xuebing Li*
Page No. – Page No.
Synthesis of Sialic Acids, Their
Derivatives and Analogues Using a
Whole-Cell Catalyst
Sialic acid-producing cell factory: Engineered Escherichia coli converted simple
and inexpensive sugars into various natural forms and non-natural derivatives or
analogues of sialic acid through an easily scalable fermentation process. Syntheses
using this cellular catalyst in combination with conventional enzymes afforded a
range of complex sialyloligosaccharides without isolation of intermediates.
This article is protected by copyright. All rights reserved.