10910 J. Am. Chem. Soc., Vol. 123, No. 44, 2001
Izumi et al.
Table 1. One-Step Purification of R-2,3-Sialyltransferase from 1 L
chemical sialylation reaction often results in low yields and low
stereoselectivity, and thus remains one of the most difficult
glycosylation reactions.6-12 Enzymatic sialylation using sialyl-
transferases therefore has been considered to be an attractive
alternative.1 The narrow substrate specificity of mammalian
sialyltransferases, however, limits the usefulness of these en-
zymes in synthesis. The commercially available mammalian
R-2,3-sialytransferase from rat liver, for example, only accepts
N-acetyllactosamine as substrate; monosaccharides, sulfate-
containing oligosaccharides, and glycopeptides are not accept-
able.1e Bacterial sialyltransferases appear to have a broader
substrate specificity. The R-2,6-sialyltransferase from Photo-
bacterium damse, for example, has been shown to catalyze the
transfer of sialic acid to a terminal galactose which is fucosylated
or sialylated at the 2 or 3 position, respectively.13 The R-2,3
sialytransferase from Neisseria meningitidis has been shown to
sialylate the terminal galactose of N-acetyllactosamine, lactose,
and R- or â-galactosides.14a,b A viral R-2,3-sialyltransferase was
shown to sialylate fucosylated acceptors such as Lewis a and
Lewis x.14c A bacterial R-2,8/2,9-sialyltransferase was shown
to accept various oligosialic acids with alternate R-2,8 and R-2,9
linkages.14d Other microbial transferases have also been
reported.14e However, the donor and acceptor specificities of
all these microbial sialyltransferases, except the one from N.
meningitidis, have not been well studied and their synthetic
usefulness has not been well demonstrated. In this work, we
exploit the substrate specificity of this microbial sialyltransferase
using synthetic donor and acceptor analogues and derivatives,
with particular emphasis on the development of enzymatic
sialylation methods for the synthesis of important sialyl
conjugates which are difficult or impractical to synthesize by
other means.
of Culture by Affinity Column Chromatography
total
protein
(mg)
total
specific
activity
(U/mg)
activity
yield
(%)
step
(µmol/ h)a
cell free ext.
pellet
after affinity column
3600
12100
60.5
1587
921
181
0.44
0.08
3.00
100
11.4
(from supernatant)
a The activity was measured based on the acceptor lactose (1 mM
CMP-sialic acid, 1 mM acceptor, 10 mM MgCl2, 50 mM Tris buffer,
pH 7.5). One unit (U) of the enzyme activity is the formation of 1
µmol of product per hour.
The expressed protein was active, but was not able to bind to
the Ni-affinity colunm. Using pET vector we are able to express
the R-2,3-sialyltransferase with the histidine tag at the C-
terminus. The expressed protein was more active and able to
bind to the Ni-colunm. To purify the enzyme, the cells were
passed through a french press followed by centrifugation. The
supernatant was found to contain about 60% of the enzyme
activity and the remaining activity was found in the cell pellet.
The supernatant was passed through the Ni-colunm and eluted
with 100 mM imidazole. The enzyme isolated through this
simple procedure was shown to be quite homogeneous on SDS-
page electrophoresis. It was thus used directly for determination
of its specific activity and substrate specificity (Table 1).
Enzymatic Analysis. In general, the assay for acceptor
specificity was carried out with CMP-[14C]Neu5Ac as donor
substrate. Sialylated products were separated from CMP-[14C]-
Neu5Ac by passing through Dowex 1×8 resin (phosphate form)
and the isotope content was counted by scintillation counter.
This method is, however, not effective for the analysis of
acceptors containing negative charges or hydrophobic groups
due to the problem of separation. Alternative methods were used
for qualitative analysis to find synthetically acceptable substrates.
Thus, for evaluation of acceptors carrying a hydrophobic
aglycon, cold CMP-Neu5Ac was used and the product was
isolated with a SepPak 18 solid-phase extraction cartridge and
analyzed by electrospray mass spectrometry (ESI-MS). For
evaluation of sulfate-containing oligosaccharides as acceptors,
the reaction was analyzed by TLC with use of resorcinol to
visualize the sialylated product. For compounds carrying a long
aliphatic chain, the assay was carried out in the presence of
0.02% Triton X-100. To make sure that these qualitative
methods are reliable, the sialylated product was further char-
acterized by ESI-MS to ensure that sialylation does occur. The
relative intensity of the sialylated product was then used to
estimate the enzyme activity. Alternatively, the relative rates
can be determined by using a pH indicator (e.g. phenol red, ∆ꢀ
) 5.6 × 104 M-1 cm-1) in a minimal concentration of buffer
(∼2 mM). Several saccharides and derivatives were quickly
surveyed by using one of these methods to determine their
relative acceptor specificity. All numbers are based on lactose
acceptor, the specific activity of which was determined to be 3
U/mg (1 U ) 1 µmol product formed per hour).
Results and Discussion
Preparation of r-2,3-Sialyltransferase. The gene coding for
the enzyme from Neisseria gonorrheae was initially cloned into
pRSET vector (Invitrogen Co., Carlsbad, CA) for expression
of the protein containing a hexahistidine tag at the N-terminus.
(6) Boons, G.-J.; Demchenko, A.-V. Chem. ReV. 2000, 100, 4539-4565.
(7) Okamoto, K.; Goto, T. Tetrahedron 1990, 46, 5835-5857.
(8) DeNinno, M. P. Synthesis 1991, 583-593.
(9) Kanie, O.; Hindsgaul, O. Curr. Opin. Struct. Biol. 1992, 2, 674-
681.
(10) (a) Erce´govic, T.; Magnusson, G. J. Org. Chem. 1995, 60, 3378-
3384. (b) Erce´govic, T.; Magnusson, G. J. Org. Chem. 1996, 61, 179-
184. (c) Hossain, N.; Magnusson, G. Tetrahedron Lett. 1999, 40, 2217-
2220. (d) Ito, Y.; Nunomura, S.; Shibayama, S.; Ogawa, T. J. Org. Chem.
1992, 57, 1821-1831.
(11) Castro-Palomino, J. C.; Tsvetkov, Y. E.; Schneider, R.; Schmidt,
R. R. Tetrahedron Lett. 1997, 38, 6837-6840.
(12) Wang, Z.; Zhang, X.-F.; Ito, Y.; Nakahara, Y.; Ogawa, T. Bioorg.
Med. Chem. 1996, 4, 1901-1908.
(13) Kajihara, Y.; Yamamoto, T.; Nagae, H.; Nakashizuka, M.; Sakak-
ibara, T.; Terada, I. J. Org. Chem. 1996, 61, 8632-8635.
(14) (a) Gilbert, M.; Cunningham, A.-M.; Watson, D. C.; Martin, A.;
Richards, J. C.; Wakarchuk, W. W. Eur. J. Biochem. 1997, 249, 187-194.
(b) Wakarchuk, W. W.; Martin, A.; Jennings, M. P.; Moxon, E. R.; Richards.
J. C. J. Biol. Chem. 1996, 271, 19166-19173. (c) Sujino, K.; Jackson, R.
J.; Chan, N. W. C.; Tsuji, S.; Palcic, M. Glycobiology 2000, 10, 313-320.
(d) Shen, G.-J.; Datta, A. K.; Izumi, M.; Koeller, K. M.; Wong, C.-H. J.
Biol. Chem. 1999, 274, 35139-35146. (e) Bozue, J. A.; Tullius, M. V.;
Wang, J.; Gibson, B. W.; Munson, R. S., Jr. J. Biol. Chem. 1999, 274,
4106-4114. Gilbert, M.; Watson, D. C.; Cunningham, A.-M.; Jennings,
M. P.; Young, N. M.; Wakarchuk, W. W. J. Biol. Chem. 1996, 271, 28271-
28276. Gilbert, M.; Brisson, J. R.; Karwaski, M. F.; Michniewicz, J.;
Cunningham, A.-M.; Wu, Y.; Young, N. M.; Wakarchuk, W. W. J. Biol.
Chem. 2000, 275, 3896-3909. Gilbert, M.; Bayer, R.; Cunningham, A.-
M.; DeFees, S.; Gao, Y.; Watson, D.; Young, N. M.; Wakarchuk, W. W.
Nat. Biotechnol. 1998, 16, 769-772. Wakarchuk, W. W.; Watson, D.; St.
Michael, F.; Li, J.; Wu, Y.; Brisson, J. R.; Young, N. M.; Gilbert, M. J.
Biol. Chem. 2001, 276, 12785-12790.
Synthesis of Substrates. We are particularly interested in
developing enzymatic methods for the sialylation of sulfate-
containing oligosaccharides, glycopeptides, and glycolipids, as
these sialosides exhibit important functions1e,5 but are difficult
to synthesize with traditional glycosylation methods. The
phenolic sulfate group is too acid sensitive to be compatible
with other protecting groups manipulations. The preparations
of potential substrates shown in Table 2 were performed as
follows. Allyl 6-O-sulfo-N-acetyllactosaminide (7) was synthe-
sized as shown in Scheme 1. Glycosylation of allyl 6-O-