Highly efficient chemoenzymatic synthesis of β1-3-linked galactosides

Article abstract of DOI:10.1039/c0cc02850a

A novel d-galactosyl-β1-3-N-acetyl-d-hexosamine phosphorylase cloned from Bifidobacterium infantis (BiGalHexNAcP) was used with a recombinant E. coli K-12 galactokinase (GalK) for efficient one-pot two-enzyme synthesis of T-antigens, galacto-N-biose (Galβ1-3GalNAc), lacto-N-biose (Galβ1-3GlcNAc), and their derivatives.

Full text of DOI:10.1039/c0cc02850a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Highly efficient chemoenzymatic synthesis of b1–3-linked galactosidesw
Hai Yu,a Vireak Thon,a Kam Lau,a Li Cai,b Yi Chen,a Shengmao Mu,a Yanhong Li,a
Peng George Wangb and Xi Chen*a
Received 28th July 2010, Accepted 16th August 2010
DOI: 10.1039/c0cc02850a
A novel D-galactosyl-b1–3-N-acetyl-D-hexosamine phosphorylase
cloned from Bifidobacterium infantis (BiGalHexNAcP) was
used with a recombinant E. coli K-12 galactokinase (GalK) for
efficient one-pot two-enzyme synthesis of T-antigens, galacto-N-
biose (Galb1–3GalNAc), lacto-N-biose (Galb1–3GlcNAc), and
their derivatives.
Carbohydrate phosphorylases catalyze the reversible
formation of monosaccharide-1-phosphate from an oligo-
saccharide or a polysaccharide and an inorganic phosphate.
Their ability in catalyzing the reverse reaction has been used
for the synthesis of oligosaccharides. Based on their protein
sequence similarity, carbohydrate phosphorylases have been
grouped into several families of glycoside hydrolases (GH13,
GH65, GH94, GH112) and glycosyltransferases (GT3, GT4,
GT20, GT35) in the Carbohydrate Active Enzymes (CAZy)
database (http://www.cazy.org).6 Among them, D-galactosyl-
b1–3-N-acetyl-D-hexosamine phosphorylase (GalHexNAcP,
EC 2.4.1.211), a member of CAZy glycoside hydrolase
GH112 family, was initially reported in Bifidobacterium
bifidum DSM 20082 in 1999.7 It reversibly phosphorolyzes
GNB and LNB to produce a-D-galactose-1-phosphate
(Gal-1-P) and the corresponding N-acetyl-D-hexosamine. The
partially purified enzyme (MW: 140 kDa)7 was used in the
synthesis of Galb1–3GlcNAcaOR containing different
aglycones with low yields.8 The GalHexNAcP gene was later
found in Clostridium perfringens ATCC13124, Vibrio vulnificus
CMCP6, and several strains of bifidobacteria and was believed
to contribute to the intestinal colonization of the bacteria.9
The enzyme has been cloned and used with a sucrose
phosphorylase, a UDP-glucose-hexose-1-phosphate uridylyl-
transferase, and a UDP-glucose 4-epimerase for one-pot
four-enzyme high-yield large-scale production of LNB or
GNB from sucrose and GlcNAc or GalNAc in the presence
of UDP-glucose and phosphate.10
b1–3-Linked galactosides such as galacto-N-biose (GNB,
Galb1–3GalNAc) and lacto-N-biose (LNB, Galb1–3GlcNAc)
are key carbohydrate structures in nature. They are themselves,
or are part of, important carbohydrate epitopes involved in cell
adhesion, signalling, fertilization, differentiation, development,
and cancer metastasis. For example, GNB with an a-configuration
at the anomeric carbon of GalNAc (Galb1–3GalNAca) is
named Thomsen–Friedenreich antigen (TF or T-antigen,
Galb1–3GalNAcaSer/Thr) disaccharide. It is the glycan of
the Core 1 mucin-type O-GalNAc glycopeptides/glycoproteins
and can be branched/extended to form Core 2 as well as
extended Core 1 and Core 2 glycans.1 On the other hand,
GNB with a b-configuration at the anomeric carbon of
GalNAc (Galb1–3GalNAcb) is an essential part of the
carbohydrate moieties of complex glycosphingolipids such as
GA1, GM1, GD1, GT1, GQ1, and GP1, etc. In comparison,
LNB is a common structure presented broadly in human milk
oligosaccharides (e.g. lacto-N-tetraose and its sialylated/
fucosylated derivatives) and in the carbohydrate moieties
(e.g. type I glycans, Lewis a, sialyl Lewis a, and Lewis b
antigens) of glycoproteins and glycolipids.1,2
Crystal structure of a recombinant GalHexNAcP from
Bifidobacterium longum JCM1217 reveals a partially broken
TIM barrel fold resembling a thermophilic b-galactosidase.11
Nevertheless, the acceptor substrate specificity of GalHexNAcP
has not been studied in detail and its application in efficient
synthesis of GNB and LNB derivatives has not been fully
To use these b1–3-linked galactosides as probes or glycan
building blocks for biological studies, GNB and LNB have been
synthesized by chemical3 and enzymatic approaches using
b-glycosidases4 or b1–3-galactosyltransferases.5 Nevertheless,
these methods have limitations. Chemical synthesis requires
multiple tedious protection–deprotection and purification
procedures. Glycosidase-catalyzed reactions usually result in
low yields and poor regioselectivity. The formation of
b1–3-linked galactosidic bonds using b1–3-galactosyltransferases
has been proven efficient, but its application is hampered by
limited access to a sufficient amount of highly efficient glycosyl-
transferases and the required sugar nucleotide, uridine
50-diphosphate galactose (UDP-Gal).
Scheme 1 Synthetic scheme for one-pot two-enzyme synthesis of
b1–3-linked galactosides. GalK, E. coli K-12 galactokinase;
BiGalHexNAcP, Bifidobacterium infantis D-galactosyl-b1–3-N-acetyl-
D-hexosamine phosphorylase.
a Department of Chemistry, University of California,
One Shields Avenue, Davis, California 95616, USA.
E-mail: chen@chem.ucdavis.edu; Fax: +1 530-752-8995;
Tel: +1 530-754-6037
b Departments of Biochemistry and Chemistry, Ohio State University,
Columbus, OH 43210, USA
w Electronic supplementary information (ESI) available: Experimental
details for cloning and characterization of BiGalHexNAcP, chemical
and enzymatic synthesis, NMR and HRMS data. See DOI: 10.1039/
c0cc02850a
Scheme
2
One-pot two-enzyme synthesis of T-antigen Galb1–
3GalNAca1-O-Ser/Thr.
c
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 7507–7509 7507

View Article Online
Table 1 One-pot two-enzyme synthesis of b1–3-linked galactosides from Gal, ATP, and N-acetyl-hexosamine (HexNAc) derivatives using GalK
and BiGalHexNAcP
Acceptors
Products
Yields (%)
Acceptors
Products
Yields (%)
95
93
96
94
95
92
93
91
94
92
86
78
69
NR
NR
NR
NR
84
NR
87
explored. Here, we report a highly efficient one-pot two-enzyme
approach for the synthesis of diverse b1–3-linked galactosides,
including T-antigens, GNB, LNB, and their derivatives using a
novel D-galactosyl-b1–3-N-acetyl-D-hexosamine phosphorylase
cloned from Bifidobacterium infantis (BiGalHexNAcP)
and a recombinant galactokinase (GalK) cloned from E. coli
K-12.12
Bifidobacterium longum subsp. infantis ATCC 1569713 by
polymerase chain reaction and cloned into pET15b vector.
Optimal expression of the recombinant N-His6-tagged protein
(MW: 86.5 kDa) was achieved by incubating E. coli Origamit
B(DE3) cells containing desired plasmids at 25 1C for
18–20 h with vigorous shaking (250 rpm) after the addition
of isopropyl-1-thio-b-D-galactopyranoside (IPTG) (0.1 mM).
Ni2+–NTA column purification using an AKTA FPLC
system yielded 55 mg pure protein per litre of cell culture.
To obtain BiGalHexNAcP, Blon_2174 gene (GenBank
NC_011593) was amplified from the genomic DNA of
c
7508 Chem. Commun., 2010, 46, 7507–7509
This journal is The Royal Society of Chemistry 2010

View Article Online
As shown in Scheme 1, the synthesis of GNB, LNB, and
their derivatives was carried out using a one-pot two-enzyme
system containing BiGalHexNAcP and a recombinant E. coli
GalK. GalK catalyzed the formation of a-D-Gal-1-P from
inexpensive D-Gal in the presence of ATP. The in situ generated
a-D-Gal-1-P was used as the donor substrate by the
phosphorylase to form diverse b1–3-linked galactosides.
Excess amounts of D-Gal and ATP (1.2 or 1.5 equiv.) were
used to drive the reaction towards disaccharide formation.
HPLC-based pH profile study of BiGalHexNAcP revealed
that its catalytic activity was optimum in a relatively narrow
pH range of 5.0 to 6.5. Extremely low activity was observed
when the pH of the reaction reached 7.0 or higher (see ESIw).
For GalK, it prefered a pH higher than 7.0 for optimum
activity, showed medium activity at pH 6.5 and low activity at
pH lower than 6.5. Therefore, pH 6.5 was chosen for the
one-pot two-enzyme reactions. The reactions were carried out at
37 1C for 48 h. Products were purified using the combination of
size exclusion chromatography and silica gel chromatography.
As shown in Table 1, BiGalHexNAcP exhibits promiscuous
acceptor substrate specificity and has comparable levels of
activity toward GlcNAc and GalNAc-based structures. For
example, free GlcNAc (1) and GalNAc (11), their b-glycosides
(2 and 12, respectively) and a-glycosides (3 and 13, respectively)
are superb acceptors for BiGalHexNAcP to produce LNB
(21–23) and GNB (31–33) disaccharides in excellent 92–96%
yields. Apparently, the configurations of the C-4 hydroxyl
group and the anomeric center of the N-acetyl hexosamine do
not affect the activity of BiGalHexNAcP.
and Galb1–3GalNAca1-O-Thr. As shown in Scheme 2,
incubating GalNAca1-O-Ser 41 or GalNAca1-O-Thr 42 with
Gal and ATP in the presence of GalK and BiGalHexNAcP
successfully produced the desired disaccharide products 43
and 44 in excellent 92% and 91% yields, respectively.
In summary, taking advantage of the acceptor substrate
promiscuity of BiGalHexNAcP, we have developed a highly
efficient one-pot two-enzyme approach for the synthesis of
diverse b1–3-linked galactosides. Compared to galactosyl-
transferase-catalyzed approaches, the BiGalHexNAcP-catalyzed
reactions do not require the use, in situ generation, or regeneration
of expensive sugar nucleotides, and thus are more efficient and
simplified systems for producing galactosides. We believe that
this synthetic route will contribute greatly to obtaining and
elucidating the important roles of b1–3-galactosides as well as
b1–3-galactoside-containing glycans and glycoconjugates.
This work was supported by NIH grants R01GM076360,
U01CA128442 (to X. Chen), and R01HD061935 (to P. G.
Wang). X. Chen is an Alfred P. Sloan Research Fellow, a
Camille Dreyfus Teacher-Scholar, and a UC-Davis Chancellor’s
Fellow. We thank Professor David Mills at the University
of California-Davis for providing us the genomic DNA of
Bifidobacterium longum subsp. infantis ATCC 15697.
Notes and references
1 Essentials of Glycobiology, ed. A. Varki, R. D. Cummings,
J. D. Esko, H. H. Freeze, P. Stanley, C. R. Bertozzi, G. W. Hart
and M. E. Etzler, Cold Spring Harbor Laboratory Press,
New York, 2nd edn, 2008.
Similarly, N-acetyl group derivatization with small acyl
groups or derivatives does not affect the BiGalHexNAcP
activity significantly. For example, GlcNTFA 4 and GalNTFA
14 with an N-trifluoroacetyl group or GlcNAcN3 5 and
GalNAcN3 15 with an N-azidoacetyl group in the hexosamines
are excellent acceptors to produce disaccharides 24–25 and
34–35 in 91–94% yields. However, larger N-acyl groups
differentiated glucosamine and galactosamine-based acceptors.
For example, GlcNPr 6 and GlcNBu 7 with an N-propyl and
an N-butyl group, respectively, at glucosamine, are good
acceptors for BiGalHexNAcP to synthesize disaccharides 26
and 27 in 86% and 78% yields, respectively. The synthetic
yield decreases moderately as the size of the N-acyl group on
the glucosamine increases. In contrast, the corresponding
galactosamine derivative GalNPr 16 leads to disaccharide 36
in a moderate 69% yield while GalNBu 17 is not a suitable
acceptor for BiGalHexNAcP. A bulky N-benzoyl group prevents
both glucosamine and galactosamine derivatives (8 and 18) from
being suitable BiGalHexNAcP acceptors. Quite interestingly,
GlcN3 9 and GalN3 19, the C2 derivatives of GlcNAc and
GalNAc with the N-acetyl group being replaced by a relatively
small azido group (–N3), are not tolerable acceptors.
2 (a) C. Kunz, S. Rudloff, W. Baier, N. Klein and S. Strobel, Annu.
Rev. Nutr., 2000, 20, 699–722; (b) J. Holgersson and J. Lofling,
Glycobiology, 2006, 16, 584–593; (c) T. Feizi, Immunol. Rev., 2000,
173, 79–88; (d) C. R. Bertozzi and L. L. Kiessling, Science, 2001,
291, 2357–2364.
3 M. Wilstermann and G. Magnusson, Carbohydr. Res., 1995, 272, 1–7.
4 (a) L. Hedbys, E. Johansson, K. Mosbach and P. O. Larsson,
Carbohydr. Res., 1989, 186, 217–223; (b) A. Vetere, M. Miletich,
M. Bosco and S. Paoletti, Eur. J. Biochem., 2000, 267, 942–949;
(c) A. D’Almeida, M. Ionata, V. Tran, C. Tellier, M. Dion and
C. Rabiller, Tetrahedron: Asymmetry, 2009, 20, 1243–1246.
5 (a) G. Baisch, R. Ohrlein, M. Streiff and F. Kolbinger, Bioorg.
Med. Chem. Lett., 1998, 8, 751–754; (b) D. M. Su, H. Eguchi,
W. Yi, L. Li, P. G. Wang and C. Xia, Org. Lett., 2008, 10,
1009–1012; (c) Z. Wang, M. Gilbert, H. Eguchi, H. Yu,
J. Cheng, S. Muthana, L. Zhou, P. G. Wang, X. Chen and
X. Huang, Adv. Synth. Catal., 2008, 350, 1717–1728.
6 B. L. Cantarel, P. M. Coutinho, C. Rancurel, T. Bernard, V. Lombard
and B. Henrissat, Nucleic Acids Res., 2009, 37, D233–D238.
7 D. Derensy-Dron, F. Krzewinski, C. Brassart and S. Bouquelet,
Biotechnol. Appl. Biochem., 1999, 29(Pt 1), 3–10.
8 E. Farkas, J. Thiem, F. Krzewinski and S. Bouquelet, Synlett,
2000, 728–730.
9 (a) M. Kitaoka, J. Tian and M. Nishimoto, Appl. Environ.
Microbiol., 2005, 71, 3158–3162; (b) J. Xiao, S. Takahashi,
M. Nishimoto, T. Odamaki, T. Yaeshima, K. Iwatsuki and
M. Kitaoka, Appl. Environ. Microbiol., 2010, 76, 54–59.
10 (a) M. Nishimoto and M. Kitaoka, Biosci., Biotechnol., Biochem.,
2007, 71, 2101–2104; (b) M. Nishimoto and M. Kitaoka,
Carbohydr. Res., 2009, 344, 2573–2576.
BiGalHexNAcP also shows good tolerance towards
C6-modification. Both 6-deoxy-GlcNAc (GlcNAc6Deoxy) 10
and 6-azido-6-deoxy-GlcNAc (GlcNAc6N3) 20 are very good
acceptors for the enzyme to synthesize disaccharides 30 and 40
in high yields (84 and 87%, respectively).
11 M. Hidaka, M. Nishimoto, M. Kitaoka, T. Wakagi, H. Shoun and
S. Fushinobu, J. Biol. Chem., 2009, 284, 7273–7283.
12 X. Chen, J. Fang, J. Zhang, Z. Liu, J. Shao, P. Kowal, P. Andreana
and P. G. Wang, J. Am. Chem. Soc., 2001, 123, 2081–2082.
13 D. A. Sela, J. Chapman, A. Adeuya, J. H. Kim, F. Chen,
T. R. Whitehead, A. Lapidus, D. S. Rokhsar, C. B. Lebrilla,
J. B. German, N. P. Price, P. M. Richardson and D. A. Mills,
Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 18964–18969.
The BiGalHexNAcP and the one-pot two-enzyme system
have also been successfully applied in the efficient synthesis of
biologically important T-antigens Galb1–3GalNAca1-O-Ser
c
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 7507–7509 7509