Efficient chemoenzymatic synthesis of uridine 5′-diphosphate N-acetylglucosamine and uridine 5′-diphosphate N-trifluoacetyl glucosamine with three recombinant enzymes

Article abstract of DOI:10.1080/10826068.2016.1188315

Uridine 5′-diphosphate N-acetylglucosamine (UDP-GlcNAc) is a natural UDP-monosaccharide donor for bacterial glycosyltransferases, while uridine 5′-diphosphate N-trifluoacetyl glucosamine (UDP-GlcNTFA) is its synthetic mimic. The chemoenzymatic synthesis of UDP-GlcNAc and UDP-GlcNTFA was attempted by three recombinant enzymes. Recombinant N-acetylhexosamine 1-kinase was used to produce GlcNAc/GlcNTFA-1-phosphate from GlcNAc/GlcNTFA. N-acetylglucosamine-1-phosphate uridyltransferase from Escherichia coli K12 MG1655 was used to produce UDP-GlcNAc/GlcNTFA from GlcNAc/GlcNTFA-1-phosphate. Inorganic pyrophosphatase from E. coli K12 MG1655 was used to hydrolyze pyrophosphate to accelerate the reaction. The above enzymes were expressed in E. coli BL21 (DE3) and purified, respectively, and finally mixed in one-pot bioreactor. The effects of reaction conditions on the production of UDP-GlcNAc and UDP-GlcNTFA were characterized. To avoid the substrate inhibition effect on the production of UDP-GlcNAc and UDP-GlcNTFA, the reaction was performed with fed batch of substrate. Under the optimized conditions, high production of UDP-GlcNAc (59.51 g/L) and UDP-GlcNTFA (46.54 g/L) were achieved in this three-enzyme one-pot system. The present work is promising to develop an efficient scalable process for the supply of UDP-monosaccharide donors for oligosaccharide synthesis.

Full text of DOI:10.1080/10826068.2016.1188315

Preparative Biochemistry and Biotechnology
ISSN: 1082-6068 (Print) 1532-2297 (Online) Journal homepage: http://www.tandfonline.com/loi/lpbb20
Efficient chemoenzymatic synthesis of uridine 5′-
diphosphate N-acetylglucosamine and uridine
5′-diphosphate N-trifluoacetyl glucosamine with
three recombinant enzymes
Xiaoyan Li, Chen Qi, Peilian Wei, Lei Huang, Jin Cai & Zhinan Xu
To cite this article: Xiaoyan Li, Chen Qi, Peilian Wei, Lei Huang, Jin Cai & Zhinan Xu (2016):
Efficient chemoenzymatic synthesis of uridine 5′-diphosphate N-acetylglucosamine and uridine
5′-diphosphate N-trifluoacetyl glucosamine with three recombinant enzymes, Preparative
Biochemistry and Biotechnology, DOI: 10.1080/10826068.2016.1188315
To link to this article: http://dx.doi.org/10.1080/10826068.2016.1188315
View supplementary material
Accepted author version posted online: 24
May 2016.
Published online: 24 May 2016.
Submit your article to this journal
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=lpbb20
Download by: [University of Nebraska, Lincoln]
Date: 26 May 2016, At: 19:09

Efficient chemoenzymatic synthesis of uridine 5’-diphosphate N-acetylglucosamine
and uridine 5’-diphosphate N-trifluoacetyl glucosamine with three recombinant
enzymes
Xiaoyan Li1, Chen Qi1, Peilian Wei2, Lei Huang1, Jin Cai1, Zhinan Xu1,
1Institute of Bioengineering, College of Chemical and Biological Engineering, Zhejiang
University, Hangzhou, China 2College of Biological and Chemical Engineering, Zhejiang
University of Science & Technology, Hangzhou, China
Corresponding author: Prof. Zhinan Xu Institute of Biological Engineering, College of
Chemical and Biological Engineering, Zhejiang University, Hangzhou 310027, China.
E-mail: lhuangblue@zju.edu.cn
Co-corresponding author: Dr. Huang Lei Institute of Biological Engineering, College of
Chemical and Biological Engineering, Zhejiang University, Hangzhou 310027, China.
E-mail: znxu@zju.edu.cn
Abstract
Uridine 5’-diphosphate N-acetylglucosamine (UDP-GlcNAc) is a natural
UDP-monosaccharide donor for bacterial glycosyltransferases, while Uridine
5’-diphosphate N-trifluoacetyl glucosamine (UDP-GlcNTFA) is its synthetic mimic. The
chemoenzymatic synthesis of UDP-GlcNAc and UDP-GlcNTFA was attempted by three
recombinant enzymes. Recombinant N-acetylhexosamine 1-kinase (NahK) was used to
produce GlcNAc/GlcNTFA-1-phosphate from GlcNAc/GlcNTFA.
N-acetylglucosamine-1-phosphate uridyltransferase (GlmU) from E.coli K12 MG1655
was used to produce UDP-GlcNAc/GlcNTFA from GlcNAc/GlcNTFA-1-phosphate.
Inorganic pyrophosphatase (PPase) from E.coli K12 MG1655 was used to hydrolyze
pyrophosphate to accelerate the reaction. The above enzymes were expressed in E.coli
1

BL21 (DE3) and purified respectively, and finally mixed in one-pot bioreactor. The effects
of reaction conditions on the production of UDP-GlcNAc and UDP-GlcNTFA were
characterized. To avoid the substrate inhibition effect on the production of UDP-GlcNAc
and UDP-GlcNTFA, the reaction was carried out with fed batch of substrate. Under the
optimized conditions, high production of UDP-GlcNAc (59.51 g/L) and UDP-GlcNTFA
(46.54 g/L) were achieved in this three-enzyme one-pot system. The present work is
promising to develop an efficient scalable process for the supply of UDP-monosaccharide
donors for oligosaccharide synthesis.
KEYWORDS: Biosynthesis; N-acetylhexosamine 1-kinase;
N-acetylglucosamine-1-phosphate uridyltransferase; Uridine 5’-diphosphate
N-acetylglucosamine; Uridine 5’-diphosphate N-trifluoacetyl glucosamine; one-pot
system
INTRODUCTION
Heparan sulfate (HS) is a unique class of macromolecules mostly presented on mammalian
cell surface and in the extracellular matrix. It consists of a repeating disaccharide unit
containing glucuronic acid (GlcUA) or iduronic acid (IdoUA) and glucosamine, both
capable of carrying sulfo groups [1]. To investigate the important roles of heparin/HS in
protein-binding interactions, structurally defined heparin/HS oligosaccharides such as
2

antithrombin-binding pentasaccharide has been obtained via chemical total synthesis. But
the synthesis of oligosaccharides larger than octasaccharide is extremely difficult,
especially in case of biological evaluation which required multiple target structures [2]
.
Alternatively, the enzymatic biosynthesis has been applied to obtain heparin and heparin
oligosaccharides, through which the biologically active HS oligosaccharides have been
achieved from bacteria capsular polysaccharides [3–5]. However, the mixture of
polysaccharides with various sizes and sulfation patterns could sometimes interfere the
structure and activity relationship studies.
Recently, the biosynthesis of oligosaccharides with different sulfation patterns and sizes
from a disaccharide building block using glycosyltransferases, HS C5-epimerase, and
sulfotransferases has been reported [6]. This method offers a generic approach to prepare
HS oligosaccharides possessing predictable structures. The critical step in such method is
to utilize bacterial glycosyltransferases and specialized starting molecules (Uridine
5’-diphosphate N-acetylglucosamine (UDP-GlcNAc) and Uridine 5’-diphosphate
N-trifluoacetyl glucosamine (UDP-GlcNTFA)) to build an oligosaccharide backbone from
a disaccharide. Furthermore, 10- and 12-step chemoenzymatic syntheses of two
structurally homogeneous and biologically active ULMW heparins (MW = 1778.5 and
1816.5) based on this method have been reported with 45% and 37% overall yield
respectively [7]. However, the difficulty to obtain these expected starting materials has
3

made the scalable synthesis of these structurally defined HS oligosaccharides very
impractical.
Comparing to the only 15% overall yield of five-step chemical synthesis [8], various
enzymatic syntheses afforded superior capability to efficiently produce UDP-GlcNAc
under certain conditions. For example, a yeast-based method using 5’-UMP and
N-acetylhexosamine (GlcNAc) [9], or enzyme co-immobilized agarose beads [10], or
reconstituted biosynthetic pathway in a cell-free system [11], were applied to synthesis
UDP-GlcNAc. In addition, UDP-GlcNAc has been produced via microbial fermentation of
combined recombinant Escherichia coli and Corynebacterium ammoniagenes [12]
,
Saccharomyces cerevisiae [13], and the probiotic bacterium Lactobacillus casei strain BL23
[14]. However, the production efficiency of these methods is not high enough. Similarly, the
chemical synthetic yield of UDP-GlcNTFAis also very low [15]. The enzymatic synthesis of
UDP-GlcNTFA from GlcNTFA-1-phosphate was catalyzed by
N-acetylglucosamine-1-phosphate uridyltransferase (GlmU), in which
GlcNTFA-1-phosphate was prepared from costly substrate GlcNH2 1-phosphate [6]. To
avoid the limited origin of GlcNH2 1-phosphate, N-acetylhexosamine 1-kinase (EC
2.7.1.162, NahK) cloned from Bifidobacterium longum JCM1217 [16] was applied to
catalyze the direct addition of ATP-derived phosphate to the anomeric position of
GlcNAc/GlcNTFA, forming an important phosphate intermediate. Substrate specificity
study showed that NahK could tolerate diverse modifications at C2 of GlcNAc [17–19]. On
4

the other hand, the N-terminal domain of GlmU (amino acids from M1 to R299) from
Escherichia coli K12 could also catalyze the formation of UDP-GlcNAc from GlcNAc-1-P
and UTP. Subsequent substrate evaluation suggested that only analogs with an amide
linkage at the C2 nitrogen could be recognized by GlmU [20]. Based on these work, an
elaborate route with two-step enzymatic reactions has been designed to accomplish the
production of UDP-GlcNAc using common substrates (GlcNAc, ATP and UTP) [21]. For
enzymatic production processes it is of interest to use high initial substrate concentrations
to obtain higher yields. However, a high concentration of initial substrates (GlcNAc, ATP
and UTP) [22] or the pyrophosphate by-product [23] formed in the enzymatic reactions had a
large negative effect on the enzyme activity and affect the yield of the desired product.
In this work, recombinant NahK GlmU and PPase are used in one-pot to synthesize
UDP–GlcNAc/GlcNTFA (Fig. 1). PPase is added for hydrolyzing the by-product
pyrophosphate (PPi) to drive the reaction forward. The effects of temperature, Mg2+
concentration, pH and substrate concentration on the productivity of
UDP–GlcNAc/GlcNTFA have been investigated respectively. Finally, a fed-batch
approach is used for controlling concentrations of substrates which affect the yield of
UDP–GlcNAc/GlcNTFA.
EXPERIMENTAL
5

Strains, Plasmids, And Materials
Escherichia coli K12 substr. MG1655 (ATCC-700926) was used as the original strain to
amplify PPase and GlmU genes; Live Combined Bifidobacterium, Lactobacillus and
Enterococcus oral powder (commercial brand PeiFeiKang) was purchased from
Sinepharm (Shanghai, China) as the origin to amplify NahK gene. E. coli DH5α and E. coli
BL21 (DE3) were purchased from Invitrogen (Carlsbad, CA, USA). The restriction
enzymes Nco I and Xho I, as well as LA Taq DNA polymerase and T4 DNA ligase, were
purchased from Takara (Dalian, China). GlcNAc was purchased from Aladdin (Shanghai,
China). Ethyl trifluoroacetate (ETFA) and the standard GlcNTFA were purchased from
J&K Scientific Ltd. (Beijing, China).
Plasmids Construction
The target genes of NahK, GlmU and PPase were amplified from corresponding microbial
genomic DNA, respectively. The primers were shown in Table 1. The introduced
restriction sites Nco I and Xho I were underlined. Each target gene was cloned into
pET28a(+) vector via Nco I/ Xho I and transformed into E. coli DH5α competent cells. The
resulting expression plasmids were characterized by restriction mapping and DNA
sequencing (Sangon, Shanghai, China).
Enzyme Expression And Purification
6

Validated expression plasmid was transformed into E. coli BL21 (DE3). A fresh clone of
recombinant E. coli strain was grown in 5 mL LB medium containing 30 mg/L kanamycin
at 37oC and 200 rpm for 12 h, then inoculated into 50 mL fresh LB medium containing 30
mg/L kanamycin and continued culturing at the same conditions. When the OD600 of
culture was reached 1.0, 0.1 mM isopropyl-β-D-thiogalactoside (IPTG) was added to
induce the expression for additional 12 h at 25oC. After harvested by centrifugation at
12,000 × g for 10 min, cells were resuspended in 20 mM sodium phosphate (pH 7.4) and
sonicated on ice. The resulting cell lysate was centrifuged at 12,000 ×g and 4oC for 10 min,
and the supernatant was applied into a nickel affinity chromatography column packed with
10 mL Ni-NTA resin (Qiagen, Hilden, Germany). After washing column with washing
buffer (20 mM sodium phosphate, pH 7.4, 100 mM NaCl, 10 mM imidazole), NahK was
eluted with elution buffer I (20 mM sodium phosphate, pH 7.4, 100 mM NaCl, 500 mM
imidazole), and then desalted by HiTrapTM Desalting HP column with 50 mM Tris-HCl
buffer, pH 7.5. The purified fractions were concentrated with Amicon Ultra and stored in
20% glycerol (50 mM Tris-HCl, pH 7.5) at -20oC. For purification of PPase and GlmU,
cells were resuspended in buffer (20 mM Tris-HCl, pH8.0, 500 mM NaCl, 20 mM
imidazole) and sonicated on ice. Similarly, target enzyme was eluted with elution buffer II
(20 mM Tris-HCl, pH 8.0, 500 mM NaCl, 250 mM imidazole). The fractions containing
PPase were pooled and desalted by HiTrapTM Desalting HP column with 20 mM Tris-HCl
(pH 8.0) and 100 mM KCl, and stored in storage buffer I (20 mM Tris-HCl, pH 8.0, 0.1
mM EDTA, 1 mM DTT, 100 mM KCl, and 50% (v/v) glycerol) at -20oC. The fractions
7

containing GlmU were pooled and desalted by HiTrapTM Desalting HP column with 50
mM Tris-HCl, pH 8.0, and 150 mM NaCl, and stored in storage buffer II (50 mM Tris-HCl,
pH 8.0, 2 mM EDTA, 2 mM DTT, 150 mM NaCl, and 20% (v/v) glycerol) at -20oC.
Quantification Of Purified Protein
The concentration of target enzyme was determined in a 96-well plate using bicinchoninic
acid (BCA) Protein Assay Kit (Sangon, Shanghai, China), with bovine serum albumin as
the standard. The absorbance of each sample was measured at 562 nm by a
spectrophotometer (Varioskan Flash, Thermo Scientific (Waltham, MA, USA)). For PPase,
1 U is the amount of enzyme that will generate 1 μmol of phosphate per minute from PPi
under standard reaction conditions (a 10 min reaction at 25 oC in 100 mM Tris-HCl, pH 7.2,
2 mM MgCl2 and 2 mM PPi in a reaction volume of 0.5 mL). The phosphate concentration
in the mixture was measured as described by Heinonen [24]
.
Chemical Synthesis Of Glcntfa
The chemical synthesis of GlcNTFAis shown as in Fig. 1A. 1.02 g Glucosamine HCl (4.73
mmol) was dissolved in 10 mL MeOH, then 1.00 g Na2CO3 (9.46 mmol) and 1.34 mL
ETFA (9.46 mmol) were sequentially added into the solution. The reaction mixture was
stirred at 45oC overnight and the GlcNTFA was purified by silica gel column
8

chromatography (volume ratio, EtOAc: MeOH=100:1). GlcNTFA samples were analyzed
by HPLC.
Synthesis Of Udp-Glcnac And Udp-Glcntfa
The reaction mixture for synthesis of UDP-GlcNAc/UDP-GlcNTFA contained 1 mM
GlcNAc/GlcNTFA, 1 mM UTP, 1 mM ATP, 10 mM MgCl2, 0.2 mg NahK, 0.2 mg GlmU, 5
U PPase , and 50 mM Tris-HCl (pH 8.0). 1 ml mixture was incubated at 37oC for 30 min,
then boiled for 5 min and centrifuged to remove residues. To optimize the reaction
condition, single factor analysis was executed to investigate the effects of temperature,
Mg2+ concentration, pH on the biosynthesis of UDP-GlcNAc and UDP-GlcNTFA.
Preparative-scale synthesis of UDP-GlcNAc and UDP-GlcNTFA were respectively
performed in 10 mL 200 mM Tris-HCl (pH 8.5) containing different equal dose (20 mM,
30 mM, 40 mM, and 50 mM) of GlcNAc/GlcNTFA, UTP and ATP; 5 mM MgCl2, 2 mg
NahK, 2 mg GlmU and 50 U PPase. The reaction mixture was incubated at 37oC and
monitored by HPLC analysis every two hours.
To avoid possible substrate inhibition effect, a substrate feeding strategy was applied on the
production of UDP-GlcNAc and UDP-GlcNTFA. The reaction was performed in 100 mL
200 mM Tris-HCl solution containing 2 mmol GlcNAc or 2 mmol GlcNTFA, 2 mmol ATP,
9

2 mmol UTP, 5mM MgCl2, 20 mg NahK, 20 mg GlmU, and 500 U PPase, and the pH was
adjusted to 8.5 using 12M HCl. The reaction was monitored by HPLC every half hour.
Another dose of substrates (1 mmol GlcNAc or 1 mmol GlcNTFA, 1 mmol ATP, and 1
mmol UTP) were added when reaction was nearly completed, and the pH was adjusted to
8.5 using 2M NaOH. The reaction volume increased by 2 mL each time as a result of the
addition of substrates and the pH adjustment. The reaction was finally stopped when no
UDP-GlcNAc or UDP-GlcNTFA was produced.
Products Analysis
Samples were detected by the Agilent 1100 Series HPLC equipped with a TC-C18 column
(4.6×250 mm, 5 μm) and a UV detector. GlcNTFA was detected at 217 nm with isocratic
mobile phase (MeOH: H2O=60:40) and 0.5 mL/min flow rate; UDP-GlcNAc was analyzed
at 254 nm using 1 mL/min isocratic flow of 50 mM ammonium formate (pH 4.5)
containing 5% menthol; UDP-GlcNTFA was also analyzed at 254 nm using 1 mL/min
isocratic flow of 50 mM ammonium formate (pH 3.5) containing 5% menthol. The
program was performed at 30 °C for 15 min, and the expected nucleotide sugars UTP and
ADP/ATP were detected at 254 nm. ESI-MS analysis was carried out using a Bruker
LC/MSD Trap mass spectrometer (Bruker, Massachusetts, USA), the collected data was
processed by HP Chemstation. ESI-MS analysis of UDP-GlcNAc was carried out in the
10

positive ion detection mode and ESI-MS analysis of UDP-GlcNTFA was carried out in the
negative ion detection mode.
Separation Of Udp-Glcnac And Udp-Glcntfa Using A Preparative Hplc
The separation was performed by an Agilent 1200 Series HPLC equipped with an Agilent
Eclipse XDB C18 (9.4×250mm, 5 μm) and a UV detector. For UDP-GlcNAc, 2 mL/min
isocratic flow of 20 mM ammonium acetate (pH 4.5) containing 5% menthol was used. For
UDP-GlcNTFA, 2 mL/min isocratic flow of 0.1% formic acid containing 5% menthol was
used. The resultant products were lyophilized and the purity was validated by HPLC
analysis.
RESULTS AND DISCUSSION
Cloning, Expression, And Purification Of Enzymes
In order to reconstruct the enzymatic synthesis pathway of UDP-GlcNAc/GlcNTFA in
vitro, three plasmids pET28a-NahK, pET28a-GlmU and pET28a-PPase were constructed.
The subsequent DNA sequencing has validated the correctness of GlmU and PPase genes.
The nucleotide BLAST result suggested that the NahK amplified from Peifeikang was
identical to NahK from Bifidobacterium longum subsp. JDM301 (GenBank no.
CP002010.1). After enzyme expression and purification, up to 185 mg, 400 mg and 205
11

mg of NahK, GlmU and PPase could be obtained from one liter of E. coli culture,
respectively. The SDS-PAGE analysis showed purified proteins around 41, 50 and 20 kDa
(Fig. 2), which were well matched with the calculated molecular weight of His6-tagged
fusion proteins of NahK, GlmU and PPase (41.1, 50.5 and 19.7 kDa, respectively).
Enzymatic Synthesis Of Udp-Glcnac And Udp-Glcntfa
The synthesis of UDP-GlcNAc and UDP-GlcNTFA mainly involve two steps: the
formation of GlcNAc-1-P or GlcNTFA-1-P catalyzed by NahK, and the transfer of UDP
catalyzed by GlmU. UDP-GlcNAc was detected at the retention time of 3.261 min (Fig.
3A), and subsequent LC-MS analysis gave the signal of UDP-GlcNAc at m/z 607.9 (m/z
calculated for C17H27N3O17P2 (M+H) is 608.1) (Fig. 4A). UDP-GlcNTFA was detected at
the retention time of 4.64 min (Fig. 3B), and the subsequent LC-MS analysis gave the
signal of UDP-GlcNTFA at m/z 660.0 (m/z calculated for C17H24F3N3O17P2 (M-H) is 660.1)
(Fig. 4B).
After the purification using a preparative HPLC, the HPLC analysis of final UDP-GlcNAc
showed 98% purity based on the HPLC peak area; while the HPLC analysis of final
UDP-GlcNTFA showed only 85.34% purity based on the HPLC peak area, with ADP as
the major impurity (data not shown). ADP is not easy to remove efficiently by only using
preparative liquid chromatograph. In previous reports, calf intestinal alkaline phosphatase
12

(CIAP) was used to hydrolyze nucleotides for UDP-GlcNAc production [25] and ATP
regeneration system [26–28] also could be introduced to avoid the accumulation of ADP.
The Effects Of Different Reaction Conditions On The Synthesis Of Udp-Glcnac And
Udp-Glcntfa
The Effect Of Temperature
The temperature not only affects the enzyme activity, but also affects the stability of the
enzyme. The effect of temperature on the synthesis of UDP-GlcNAc/GlcNTFA was shown
in Fig. 5. The fractional conversion increased with the increase of temperature under 37oC
and dramatic decrease was observed when the temperature increased from 40 to 50oC.
Although NahK remains half of its activity during incubation for 30 min at 50°C [16], the
GlmU is extremely unstable [29] so that no UDP-GlcNAc/GlcNTFA is generated at 50oC.
The final determined optimal temperature was 37oC. Overall, the fractional conversion of
UDP-GlcNAc is higher than that of UDP-GlcNTFA at the temperature range of 20-50oC.
The Effect Of Mg2+ Concentration
Both NahK and GlmU required magnesium ions for activity. As shown in Fig. 6, the
optimum Mg2+ concentration for the synthesis of UDP-GlcNAc/GlcNTFA was 5 mM. The
fractional conversion in the presence of 5 mM Mg2+ was about 4~5 times of those in the
13

presence of 0.5 mM Mg2+ and 1mM Mg2+. Increasing the concentration of Mg2+ from 5
mM to 20 mM caused a slight decrease of fractional conversion.
The Effects Of Ph And Buffer System
The effects of pH and buffer system on the synthesis of UDP-GlcNAc/GlcNTFA were
shown in Fig. 7Aand Fig. 7B. Both conversion ratios of UDP-GlcNAc and UDP-GlcNTFA
were highest after 5 min incubation at pH 8.5 in Tris-HCl buffer. The synthesis of
UDP-GlcNAc had a broad pH adaptability while the synthesis of UDP-GlcNTFA had a
narrow one. The conversion ratio dropped quickly with either decrease of the pH to below
7.5 or an increase of the pH to above 9.5. About 50% of the optimal activity was observed
at pH 6.0 in NaH2PO4/Na2HPO4 buffer and at pH 10.5 in Glycine-NaOH buffer for the
synthesis of UDP-GlcNAc. In comparison, 66.3% of the optimal activity was observed at
pH 7.5 in Tris-HCl buffer and 60.4% of the optimal activity was observed at pH 9.0 in
Glycine-NaOH buffer for the synthesis of UDP-GlcNTFA.
The Effect Of Substrate Concentration
The time-course of UDP-GlcNAc (Fig. 8A) production at different initial substrate
concentrations showed that complete conversion of substrate to UDP-GlcNAc took 1h, 3 h,
8h and 18 h, respectively at 20 mM, 30 mM, 40 mM and 50 mM substrate concentration.
The accumulation of UDP-GlcNAc in 0.5 h and 1 h increased with the initial substrate
14

concentrations from 20 mM to 50 mM. At 30 mM substrate concentration, the reaction rate
increased over time, which was also observed at 40 mM and 50 mM substrate
concentration. This results suggested that substrate inhibit the enzyme activity at high
concentration
The time-course of UDP-GlcNTFA (Fig. 8B) production at different initial substrate
concentrations showed that complete conversion of substrate to UDP-GlcNAc took 1 h, 4 h
and 13h, respectively at 20 mM, 30 mM and 40 mM substrate concentration. Only 24.5%
of substrate was converted to UDP-GlcNTFA in 16 h when the initial substrate
concentration was 50 mM. The accumulation of UDP GlcNTFA in 0.5 h and 1 h also
increased with the initial substrate concentrations from 20 mM to 50 mM. The reaction rate
increased over time at 20mM, 30mM and 40mM substrate concentration. All the evidences
suggested substrate inhibition effect.
Scaled-Up Production Of Udp-Glcnac And Udp-Glcntfa With Fed Batch Of Substrate
To remove the substrate inhibition effect on the production of UDP-GlcNAc and
UDP-GlcNTFA , the scaled-up production with fed batch of substrate was executed (Fig.
9A, B). The conversion of initial GlcNAc, ATP and UTP (20 mM) to UDP-GlcNAc took 1
h and the earlier dose of GlcNAc, ATP and UTP was converted to UDP-GlcNAc quickly.
The UDP-GlcNAc production slowed down with Mg2+ consumption after 4.25 h reaction.
15

After 47 mg MgCl2 was added at 5.25 h, the synthesis of UDP-GlcNAc accelerated. The
last dose reaction lasted for approximately 3 h, which might be due to the product
inhibition effect and/or the loss of enzyme activity. Eventually, the final concentration of
produced UDP-GlcNAc reached 98.04 mM (59.51 g/L) after 14.75 h reaction. The
productivity of UDP-GlcNAc is 4.03 g·L-1·h-1 and the conversion ratio was 92.0%.
Similarly, the conversion of initial GlcNTFA, ATP and UTP (20 mM) to UDP-GlcNTFA
also took 1 h and the earlier dose of GlcNTFA, ATP and UTP was converted to
UDP-GlcNTFA quickly in 1 h. After 47 mg MgCl2 was added at 5.0 h, the synthesis of
UDP-GlcNTFAaccelerated. The last dose of GlcNTFA, ATP and UTP took 3.5 h to convert
to UDP-GlcNTFA. The final concentration of produced UDP-GlcNTFA reached 70.41
mM (46.54 g/L) after 14 h reaction. The productivity of UDP-GlcNTFA is 3.32 g·L-1·h-1
and the conversion ratio was 75.5%. The conversion ratio could not reach as high as that of
UDP-GlcNAc because some UDP-GlcNTFA was degraded.
Fed-batch of substrate greatly improves the productivity of UDP-GlcNAc and
UDP-GlcNTFA. The accumulation of UDP-GlcNAc and UDP-GlcNTFA slowed down as
time went by, which might be caused by the product inhibition effect and the loss of
enzyme activity. In the future, immobilization of these enzymes together, such as on the
surface of Ni2+-functionalized beads, can be utilized to further improve the efficiency of
enzyme catalysis and reduce the total cost for large-scale production.
16

CONCLUSIONS
The aim of the present study was to achieve high production of UDP-GlcNAc and
UDP-GlcNTFA. GlcNTFA was chemically synthesized from GlcNAc and ethyl ETFA.
The key enzymes essential for the efficient production of UDP-GlcNAc and
UDP-GlcNTFA were expressed and purified, and finally mixed in one-pot bioreactor. The
effects of reaction conditions on the productivity of UDP-GlcNAc and UDP-GlcNTFA
were characterized. To avoid the substrate inhibition effect on the production of
UDP-GlcNAc and UDP-GlcNTFA, the reaction was carried out with fed batch of substrate.
Under the optimized conditions, high production of UDP-GlcNAc (59.51 g/L) and
UDP-GlcNTFA (46.54 g/L) were achieved in this three-enzyme one-pot system. In this
study, large-scale production of UDP-GlcNAc/GlcNTFA at low cost can be achieved,
which provides the basis for the industrialization of biosynthesis of ultralow molecular
weight heparin.
ACKNOWLEDGEMENTS
This work was financially supported by the National Natural Science Foundation of China
(Grant No. 21276226 and 21306164), and the National High Technology Research and
Development Program of China (2014AA021302, 2012AA021502 and
2012AA022105A).
17

REFERENCES
[1] Peterson, S. P.; Frick, A.; Liu, J. Design of Biologically Active Heparan Sulfate and
Heparin Using an Enzyme-based Approach. Nat. Prod. Rep. 2009, 26(5), 610–627.
[2] Petitou, M.; van Boeckel, C. A. A. A Synthetic Antithrombin III Binding
Pentasaccharide is now a Drug! What Comes next? Angew. Chem. Int. Ed. 2004, 43,
3118–3133.
[3] Chen, J.; Jones, C. L.; Liu, J. Using an Enzymatic Combinatorial Approach to Identify
Anticoagulant Heparan Sulfate Structures. Chem. Biol. 2007, 14(9), 986–993.
[4] Chen, J.; Avci, F. Y.; Munoz, E. M.; McDowell, L. M. et al. Enzymatic Redesigning of
Biologically Active Heparan Sulfate. J. Biol. Chem. 2005, 280, 42817–42825.
[5] Zhang, Z.; McCallum, S. A.; Xie, J., Nieto, L. et al. Solution Structures of
Chemoenzymatically Synthesized Heparin and its Precursors. J. Am. Chem. Soc. 2008,
130, 12998–13007.
[6] Liu, R.; Xu, Y.; Chen, M.; Weiwer, M. et al. Chemoenzymatic Design of Heparan
Sulfate Oligosaccharides. J. Biol. Chem. 2010, 285, 34240-34249.
[7] Xu, Y. M.; Masuko, S.; Takieddin, M.; Xu, H. M. et al. Chemoenzymatic Synthesis of
Homogeneous Ultralow Molecular Weight Heparins. Science. 2011, 334(6055), 498-501.
18

[8] Heidlas, J. E.; Lees, W. J.; Pale, P.; Whitesides, G. M. Gram-scale Synthesis of
Uridine 5'-Diphospho-N-acetylglucosamine: Comparison of Enzymatic and Chemical
Routes. J. Org. Chem. 1992, 57(1), 146-151.
[9] Okuyama, K.; Hamamoto, T.; Ishige, K.; Takenouchi, K.; Noguchi, T. An Efficient
Method for Production of Uridine 5 '-Diphospho-N-acetylglucosamine. Biosci.
Biotechnol. Biochem. 2000, 64(2), 386–392.
[10] Shao, J.; Zhang, J. B.; Nahalka, J.; Wang, P. G. Biocatalytic Synthesis of Uridine 5
'-Diphosphate N-acetylglucosamine by multiple Enzymes Co-immobilized on Agarose
Beads. Chem. Commun. 2002, 21, 2586–2587.
[11] Zhou, J. W.; Huang, L.; Lian, J. Z.; Sheng, J. Y. et al. Reconstruction of the
UDP-N-acetylglucosamine Biosynthetic Pathway in Cell-free System. Biotechnol. Lett.
2010, 32(10), 1481-1486.
[12] Tabata, K.; Koizumi, S.; Endo, T.; Ozaki, A. Production of
UDP-N-acetylglucosamine by Coupling Metabolically Engineered Bacteria. Biotechnol.
Lett. 2000, 22(6), 479–483.
[13] Ying, H. J.; Chen, X. C.; Cao, H. P.; Xiong, J. et al. Enhanced Uridine Diphosphate
N-acetylglucosamine Production using Whole-cell Catalysis. Appl. Microbiol. Biotechnol.
2009, 84(4), 677-683.
[14] Rodriguez-Diaz, J.; Rubio-del-Campo, A.; Yebra, M. J. Metabolic Engineering of
Lactobacillus casei for Production of UDP-N-acetylglucosamine. Biotechnol. Bioeng.
2012, 109(7), 1704-1712.
19

[15] Sala, R. F.; MacKinnon, S. L.; Palcic, M. M.; Tanner, M. E.
UDP-N-trifluoroacetylglucosamine as an Alternative Substrate in
N-acetylglucosaminyltransferase Reactions. Carbohyd. Res. 1998, 306(1-2), 127-136.
[16] Nishimoto, M.; Kitaoka, M. Identification of N-acetylhexosamine 1-Kinase in the
Complete Lacto-N-Biose I/Galacto-N-biose Metabolic Pathway in Bifidobactetium
longum. Appl. Environ. Microbiol. 2007, 73(20), 6444–6449.
[17] Cai, L.; Guan, W. Y.; Kitaoka, M.; Shen, J. et al. A Chemoenzymatic Route to
N-acetylglucosamine-1-phosphate Analogues: Substrate Specificity Investigations of
N-acetylhexosamine 1-Kinase. Chem. Commun. 2009, 20, 2944-2946.
[18] Fang, J. Q.; Guan, W. Y.; Cai, L.; Gu, G. F. et al. Systematic Study on the Broad
Nucleotide Triphosphate Specificity of the Pyrophosphorylase Domain of the
N-acetylglucosamine-1-phosphate Uridyltransferase from Escherichia coli K12. Bioorg.
Med. Chem. Lett. 2009, 19(22), 6429-6432.
[19] Li, Y. H.; Yu, H.; Chen, Y.; Lau, K. et al. Substrate Promiscuity of
N-acetylhexosamine 1-Kinases. Molecules. 2011, 16(8), 6396-6407.
[20] Masuko, S.; Bera, S.; Green, D. E.; Weiwer, M. et al. Chemoenzymatic Synthesis of
Uridine Diphosphate-GlcNAc and Uridine Diphosphate-GalNAc Analogs for the
Preparation of Unnatural Glycosaminoglycans. J. Org. Chem. 2012, 77(3), 1449-1456.
[21] Zhao, G. H.; Guan, W. Y.; Cai, L.; Wang, P. G. Enzymatic Route to
Preparative-Scale Synthesis of UDP-GlcNAc/GalNAc, their analogues and GDP-fucose.
Nat. Protoc. 2010, 5(4), 636-646.
20

[22] Zhou, J. W.; Fan, L. M.; Wei, P. L.; Huang, L. et al. Efficient Production of Uridine
5'-Diphospho-N-acetylglucosamine by the Combination of three Recombinant Enzymes
and Yeast Cells. Prep. Biochem. Biotech. 2010, 40(4), 294-304.
[23] Chen, Y.; Thon, V.; Li, Y. H.; Yu, H. et al. One-pot three-Enzyme Synthesis of
UDP-GlcNAc Derivatives. Chem. Commun. 2011, 47 (38), 10815-10817.
[24] Heinonen, J. K.; Lahti, R. J. A New and Convenient Colorimetric Determination of
Inorganic Orthophosphate and its Application to the Assay of Inorganic Pyrophosphatase.
Anal. Biochem. 1981, 113 (2), 313-317.
[25] Zhai, Y. F.; Liang, M.; Fang, J. Q.; Wang, X. Y. NahK/GlmU Fusion Enzyme:
Characterization and one-step Enzymatic Synthesis of UDP-N-acetylglucosamine.
Biotechnol. Lett. 2012, 34(7), 1321-1326.
[26] Marshall, D. L. ATP Regeneration Using Immobilized Carbamyl Phosphokinase.
Biotechnol. Bioeng. 1973, 15(3), 447-453.
[27] Yan, B. K.; Ding, Q. B.; Ou, L.; Zou, Z. Production of Glucose-6-phosphate by
Glucokinase coupled with an ATP Regeneration System. World. J. Microb. Biot. 2014,
30(3), 1123-1128.
[28] Zhang, J. B.; Wu, B. Y.; Zhang, Y. X.; Kowal, P.; Wang, P. G. Creatine
Phosphate-Creatine Kinase in Enzymatic Synthesis of Glycoconjugates. Org. Lett. 2003,
5(15), 2583-2586.
[29] DeLuca, C.; Lansing, M.; Crescenzi, F.; Martini, I. et al. Overexpression, one-step
Purification and Characterization of UDP-glucose Dehydrogenase and
21

UDP-N-acetylglucosamine Pyrophosphorylase. Bioorgan. Med. Chem. 1996, 4(1),
131-141.
22

Table 1. Oligonucleotides used in this study. The restriction sites Nco I and Xho I are
underlined.
Primer
Sequence（5'to3')
NahK F CGCCCATGGGTATGACTGAAAGCAATGAAGTTTTATTCGG
NahK R CGCCTCGAGCCTGGCAGCCTCCATGATGTCGG
GlmU F CATGCCATGGAAATGTTGAATAATGCTAT
GlmU R CCGCTCGAGCTTTTTCTTTACCGGACG
PPase F GCGCCATGGGAATGAGCTTACTCAACGTCCC
PPase R CGCCTCGAGTTTATTCTTTGCGCGC
23

Figure 1. Chemoenzymic synthesis schedule of UDP-GlcNAc and UDP-GlcNTFA. (A)
Chemical synthesis of GlcNTFA. (B) Enzymatic synthesis of UDP-GlcNAc and
UDP-GlcNTFA. NahK, N-acetylhexosamine 1-kinase cloned from Combined
Bifidobacterium、Lactobacillus and Enterococcus Powder; GlmU,
N-acetylglucosamine-1-phosphate uridylyltransferase cloned from E.coli K12 MG1655;
PPase, E.coli K12 MG1655 inorganic pyrophosphatase.
24

Figure 2. SDS-PAGE analysis of NahK, GlmU and PPase. Lane 1, NahK lysate after
induction; Lane 2, purified NahK; Lane 3, GlmU lysate after induction; Lane 4, purified
GlmU; Lane 5, PPase lysate after induction; Lane 6, purified PPase.
25

Figure 3. HPLC analysis of UDP-GlcNAc and UDP-GlcNTFA from GlcNAc and
UDP-GlcNTFA. (A) Enzymatic synthesis of UDP-GlcNAc from GlcNAc. (B) Enzymatic
synthesis of UDP-GlcNTFA from GlcNTFA.
26

Figure 4. ESI-MS analysis of the product. (A) The spectrum of UDP-GlcNAc (m/z 607.9).
(B) The spectrum of UDP-GlcNTFA (m/z 660.0).
27

Figure 5. Effect of temperature on the synthesis of UDP-GlcNAc/GlcNTFA. The
concentration of ATP, UTP, GlcNAc/GlcNTFAand Mg2+ is 5 mM, the reaction lasts 15 min
in 50 mM Tris-HCl buffer (pH 8.0). All experiments were performed in triplicate.
28

Figure 6. The effect of Mg2+ concentration on the synthesis of UDP-GlcNAc/GlcNTFA.
The concentration of ATP, UTP, GlcNAc/GlcNTFA is 5mM, the reaction lasts 15min at
37oC water bath in 50 mM Tris-HCl buffer (pH8.0). All experiments were performed in
triplicate.
29