An efficient approach to the discovery of potent inhibitors against glycosyltransferases

Article abstract of DOI:10.1021/jm100612r

We describe a standardized approach for searching potent and selective inhibitors of glycosyltransferases by high throughput quantitative MALDI-TOFMS-based screening of focused compound libraries constructed by 1,3-dipolar cycloaddition of the desired azidosugar nucleotides with various alkynes. An aminooxy-functionalized reagent with a stable isotope was conjugated with oligosaccharides to afford glycopeptides as acceptor substrates with improved ion sensitivity. Enhanced ionization potency of new substrates allowed for MALDI-TOFMS-based facile and quantitative analysis of enzymatic glycosylation in the presence of glycosyl donor substrates. A non-natural synthetic sugar nucleotide was identified to be the first highly specific inhibitor for rat recombinant α2,3-(N)-sialyltransferase (α2,3ST, IC₅₀ = 8.2 μM), while this compound was proved to become a favorable substrate for rat recombinant α2,6-(N)-sialyltransferase (α2,6ST, K_m = 125 μM). Versatility of this strategy was demonstrated by identification of two selective inhibitors for human recombinant α1,3-fucosyltransferase V (α1,3-FucT, K_i = 293 nM) and α1,6-fucosyltransferase VIII (α1,6-FucT, K_i = 13.8 μM).

Full text of DOI:10.1021/jm100612r

J. Med. Chem. 2010, 53, 5607–5619 5607
DOI: 10.1021/jm100612r
An Efficient Approach to the Discovery of Potent Inhibitors against Glycosyltransferases
Kensaku Hosoguchi,†, Takahiro Maeda,†, Jun-ichi Furukawa,† Yasuro Shinohara,† Hiroshi Hinou,† Mitsuaki Sekiguchi,‡
Hiroko Togame,‡ Hiroshi Takemoto,‡ Hirosato Kondo,§ and Shin-Ichiro Nishimura*,†
†Graduate School of Life Science and Frontier Research Center for Post-Genome Science and Technology, Hokkaido University, N21,
W11, Kita-ku, Sapporo 001-0021, Japan, ‡Shionogi Drug Discovery Innovation Center, Hokkaido University, N21, W11, Kita-ku,
Sapporo 001-0021, Japan, and Shionogi & Co. Ltd., Chuo-ku, Osaka 541-0045, Japan. K.H. and T.M. contributed equally to this work.
§
Received May 19, 2010
We describe a standardized approach for searching potent and selective inhibitors of glycosyltransferases by
high throughput quantitative MALDI-TOFMS-based screening of focused compound libraries constructed
by 1,3-dipolar cycloaddition of the desired azidosugar nucleotides with various alkynes. An aminooxy-
functionalized reagent with a stable isotope was conjugated with oligosaccharides to afford glycopeptides as
acceptor substrates with improved ion sensitivity. Enhanced ionization potency of new substrates allowed for
MALDI-TOFMS-based facile and quantitative analysis of enzymatic glycosylation in the presence of
glycosyl donor substrates. A non-natural synthetic sugar nucleotide was identified to be the first highly
specific inhibitor for rat recombinant R2,3-(N)-sialyltransferase (R2,3ST, IC50 = 8.2 μM), while this com-
pound was proved to become a favorable substrate for rat recombinant R2,6-(N)-sialyltransferase (R2,6ST,
Km = 125 μM). Versatility of this strategy was demonstrated by identification of two selective inhibitors for
human recombinant R1,3-fucosyltransferase V (R1,3-FucT, Ki = 293 nM) and R1,6-fucosyltransferase VIII
(R1,6-FucT, Ki = 13.8 μM).
Introduction
pathways of naturally occurring glycans and share their sub-
strates with both sugar nucleotides and acceptor oligosaccha-
ride structures during biosynthesis and metabolic degradation.
In other words, it is extremely difficult to identify every enzyme
responsible for synthesizing biologically relevant glycoforms
even though a large-scale glycomics would reveal whole
human glycoforms and profiles in their structural alterations
during cellular differentiation, proliferation, carcinogenesis,
malignant alteration, and metastasis.5a-d It is therefore evident
that insight into precise substrate specificity and mechanisms
in catalytic action by major enzymes might become a key to
elicit functional roles of individual enzyme and the hierarchical
relationship of crucial GTs/GHs in the course of biosynthetic
pathway of human normal and disease-relevant glycan struc-
tures.
Specific and strong inhibitors of GTs appear to become
potential tools for the investigation of structure-function re-
lationship of human glycomes and the regulation mechanism of
glycan biosynthesis as well as the discovery research of new class
therapeutic reagents. Actually, extensive efforts have been paid
to the development of such inhibitors for various GTs.6a-j How-
ever, there is no efficient and standardized strategy for searching
promising inhibitors due to the complex and dynamic mechan-
ism in the GT-catalyzed glycosylation reactions that involve
many components such as some sugar nucleotides as donor sub-
strates, multiform acceptor substrates, and a metal ion. Three
dimensional (3D) structural information of enzyme at atomic
resolution obtained by crystallographic or NMR analysis may
be utilized for conclusive identification of the active site amino
acid residues, often noted as conserved key amino acid residues
in the catalytic pocket. Considering the success of some antivirus
therapeutic drugs to inhibit the activity of influenza viral neu-
raminidases,7a,b it is believed that structural biology-based mole-
cular design had often accelerated rational molecular design
Posttranslational modification with glycans is a series of dyna-
mic and complicated biosynthetic processes to control essential
functions of glycoconjugates such as glycoproteins, glycosphin-
golipids, and proteoglycans.1 It is obvious that glycan structures
(glycoforms) of cell surface glycoconjugates have pivotal func-
tions in various cellular recognition systems involving cell
differentiation, development, inflammation, immune response,
tumor metastasis, bacterial/viral infections, and many other
intercellular communication and signal transductions.2a-e
However, glycan expression and its microheterogeneity in com-
mon glycoproteins cannot be predicted because protein glycosy-
lation is not template-driven and is subject to multiple sequential
and competitive enzymatic pathways.3a-c Complex glycoforms
of glycoproteins are synthesized through multistep glycosylation
reactions catalyzed by a variety of glycosyltransferases (GTsa)
sharing several sugar nucleotides in collaboration with many
glycoside hydrolases (GHs) to define further synthetic pathways
of two important subgroups, namely hybrid- and complex-type
glycans.4a,b Therefore, a tremendous number of enzymes
are considered to participate in highly complicated synthetic
*To whom correspondence should be addressed. Phone: þ81-11-706-
9043. Fax: þ81-11-706-9042. E-mail: shin@glyco.sci.hokudai.ac.jp.
aAbbreviation: GTs, glycosyltransferases; GHs, glycoside hydro-
lases; MALDI-TOFMS, matrix-assisted laser desorption/ionization
time-of-flight mass spectrometry; βGal-T1, β1,4-galactosyltransferase;
R2,3ST, R2,3-(N)-sialyltransferase; R2,6ST, R2,6-(N)-sialyltransferase;
R1,3FucT, R1,3-fucosyltransferase; R1,6FucT, R1,6-fucosyltransferase;
UDP-Gal, uridine-50-diphospho-R-D-galactopyranose; CMP-Neu5Ac
(CMP-NANA), cytidine-50-monophospho-5-acetamido-3,5-dideoxy-β-
D-glycero-R-D-galacto-non-2-ulosonic acid; GDP-Fuc, guanosine-50-di-
phospho-β-L-fucopyranose; ATP, adenosine-50-triphosphate; CTP,
cytidine-50-triphosphate; TBTA, tris-(benzyltriazolylmethyl)amine;
SPPS, solid phase peptide synthesis; DEAE, diethylaminoethyl-sephar-
ose; TMS, tetramethylsilane; TSP, trimethylsilylpropionate sodium salt.
r
2010 American Chemical Society
Published on Web 07/16/2010
pubs.acs.org/jmc

5608 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Hosoguchi et al.
of potential compounds leading toward practical therapeutic
reagents. We should consider that general 3D structural infor-
mation is still not enough for understanding dynamic reaction
mechanisms in unique GTs’ catalytic action,8a-h although it is
likelythat3DstructuresofsomeknownGTs3c may be helpful to
design some “static inhibitors” for the target GT. In fact, most
human protein kinases are known to interconvert between at
least two structural conformations, that is, active and inactive,
and the phosphorylation of key amino acid residues of kinases
can shift the balance between these dynamic states.9a,b These
two states are characterized by conformational changes in
flexible activation-loops bearing such as DFG (Asp-Phe-Gly)
motif that borders or blocks the conserved ATP-binding
pockets of active kinase conformations. Thus, a growing num-
ber of kinase inhibitors that bind selectively to inactive con-
formation, namely type II inhibitors, have been developed. The
merit of this type of inhibitors is evident because the inhibitors
that bind selectively to inactive conformation will not compete
with cellular ATP. The rational design of the type II inhibitors
has been therefore a promising strategy toward the practical
kinase inhibitors,10 whereas most of type I inhibitors that
preferentially bind with active conformation must face signifi-
cant competition by cellular ATP as a result of their similar
binding affinity to both conformations.
advent of an efficient and high throughput approach for search-
ing potential inhibitors of human GTs is now strongly required.
In the present communication, we report a standardized proto-
col for the development of potent and selective inhibitors against
GTs by MS-based rapid and quantitative screening of the
focused compound libraries constructed by click chemistry11a-e
between azidosugar nucleotides and various alkynes.
Results and Discussion
A. General Concept. Recently, we have developed mass
spectrometry-based quantitative assay for the characteriza-
tion of human protein kinases and their inhibitors.12 It was
demonstrated that synthetic peptide substrates of clinically
important human kinases, in which C-terminus is conjugated
with tryptophanylarginine (WR) and N-terminus is capped
by acetylation with (CH3CO)2O or (CD3CO)2O, allow for
rapid and highly sensitive MALDI-TOFMS-based kinetic
analyses of recombinant human kinases. It was revealed that
concurrent labeling of designated peptides with stable iso-
tope and ion sensitivity-enhancement reagents greatly facil-
itate multiple-kinase assay and inhibitor profiling using
human lung cancer K562 cells in the presence of known
inhibitors in a quantitative manner.12 The advantage of this
high throughput protocol is evident because doubly probed
peptide sets will accelerate the investigation of the relation-
ship between the expression levels of cellular kinases and
inhibitor sensitivity for any specific kinases without any
radioactive isotope labeling and prerequisite tedious chro-
matographic separation.
Our attention has been directed toward the versatility of this
strategy in a wide range of biological and medical/pharmaceu-
tical sciences. Protein glycosylation by GTs is an alternative
important biosynthetic process for posttranslational modifica-
tions to phosphorylation by protein kinases. Glycosylation
patterns of mammalian cells, namely “glycotypes” affecting
cellular adhesion, seem to be strongly dependent on a variety
of cellular stages because whole cellular glycomes change signi-
ficantly during cell differentiation, proliferation, and malignant
alteration.2,13 Judging from the importance of identifying dis-
ease-related GTs and the specific inhibitors, it is not surprising
that advent of a standard protocol for rapid and real activity-
based characterization of GTs has long been needed. It is clear
that such high throughput screening system should accelerate
discovery research of potent inhibitors against GTs of interest.
We hypothesized that tagging sugars with WR and stable iso-
tope concurrently at suited positions will readily enhance ion-
sensitivity of various glycosyl acceptor substrates14a,b and allow
for rapid and quantitative characterization of GTs reactions by
MALDI-TOFMS-based protocol as outlined in Figure 1A,B.
When a focused compound library made by feasible click reac-
tions11a-e between a designated azidosugar nucleotide and vari-
ous alkynes is available (Figure 1C), high throughput screening
for searching selective inhibitors of GTs can be established by
using an internal standard substrate labeled with stable isotope
moiety (Gal-GlcNAc3-WR-OCD3). For instance, the effect of an
inhibitor on R2,3/2,6-sialyltransferases-catalyzed glycosylation
(sialylation) can be determined quantitatively in the presence of
CMP-Neu5Ac and an acceptor substrate (Gal-GlcNAc3-WR-
OCH3) by comparing signal intensities at m/z 1511.223 and m/z
1514.244 due to the expected products bearing OCH3 or OCD3
group in the absence of inhibitor as indicated in Figure 1B.
B. Focused Compound Library Derived from Azidosugar
Nucleotides. Three azidosugar nucleotides were selected
In our previous study,6i we succeeded in the first rational
molecular design of mechanism-based specific inhibitors for
human β1,4-galactosyltransferase (βGal-T1), a member of the
most important and ubiquitous superfamily of GTs in eukar-
yotes, transfer galactose from uridine-50-diphosphogalactose
(UDP-Gal) to various acceptor substrates bearing terminal
GlcNAc residue(s). Our attention was then focused on the fact
that βGal-T1 undergoes critical conformational changes upon
UDP-Gal binding from an open (inactive) conformation to a
closed (active) conformation conducted by the flexible loop
region near the active site.8c Upon donor substrate binding,
tryptophan 310 (Trp310) involved in the small loop plays an
important role for the conformational changes in the long
flexible loop by moving toward the catalytic pocket to interact
with the complex of donor substrate and Mn2þ ion. This step
seems to be essential for generating new binding pocket desig-
nated for the acceptor substrates. Although it seemed that this
dynamic conformational change at around the catalytic site
makes rational molecular design of a favorable inhibitor for
βGal-T1 difficult, we hypothesized that compounds freezing
molecular motion of this crucial Trp310 residue in the small
loop should interrupt the access of the long flexible loop to the
catalytic pocket. Using a designated suicide substrate to probe
Trp310 residue based on the irreversible reaction mechanism,6i,8c
we revealed that the distance between Trp310 and UDP-Gal
binding pocket of βGal-T1 is estimated to be approximately
˚
15 A in the open conformation. Fortunately, this approach
allowed us to develop the first class specific inhibitor for human
βGal-T1 (Ki = 1.86 μM).6i Recently, Wagner et al. reported
a profound insight into a unique mechanism of a new, base-
modified UDP-Gal derivative that blocks the closure of a
flexible loop in the active site of a human blood group GalT
and acts, toward five different galactosyltransferases, as an
inhibitor, with Ki values in the low micromolar to nanomolar
range.8h It should be, however, emphasized that the above
approach may be available only when the precise 3D structural
formats of the target GTs, in which the flexible loop structures
can be traced, had been already uncovered. At present, it seems
likely that only limited number of human GTs may become
targets of such mechanism-based specific inhibitors.3c Thus,

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5609
Figure 1. Concept of mass spectrometry-based rapid and quantitative assay of glycosyltransferase reactions by means of doubly probed
glycopeptides. (A) General protocol illustrating for direct monitoring of the sialylation of doubly tagged acceptor Galβ1,4(GlcNAcβ1,4)3 by
R2,3/2,6-sialyltransferases in the presence of CMP-Neu5Ac (CMP-NANA). (B) Inhibitory effects of tested compounds determined by spiking
an internal standard substrate tagged with stable isotope (OCD3). The signals at m/z 1511.2 and m/z 1514.2 are due to the molecular weights of
expected products bearing OCH3 or OCD3. (C) High throughput and quantitative screening of large numbers of compounds using a focused
compound library constructed by click reactions between azidosugar nucleotides and commercially available alkynes.
tentatively as key intermediates for the construction of
focused compound library related to CMP-Neu5Ac and
GDP-Fuc analogues (Figure 2) because growing importance
of the functions of various human sialyltransferases and
fucosyltransferases in disease-relevant glycans synthesis
prompted us to challenge the discovery of specific inhibitors
showing the highest level of affinity with these enzymes. Our
attention was then paid to the facts that modifications at C-9
and C-5 positions of CMP-Neu5Ac and C-6 position of
GDP-Fuc should not reduce significantly the original affi-
nity with their binding pockets of GTs for sugar nucleo-
tides.15 It is expected that divergent derivatization at these
carbon atoms of 1-3 with various alkynes might be greatly
beneficial for the discovery research of potential candidates
exhibiting mechanism-based inhibitory effect through the
interaction with unknown key amino acid(s) locating on the
flexible loop near the active site. It is likely that this approach
leads to new competitive inhibitors having additional func-
tion to freeze molecular motion of flexible loop involving
nucleophilic amino acid residue(s) as we have succeeded
rationally in the synthesis of mechanism-based inhibitors
against human recombinant βGal-T1.6i
Schemes 1 and 2 indicate the synthetic route of 1 and 2 from
sialic acid on the basis of combined chemical and enzymatic reac-
tions. 1 was prepared readily from a known 9-azido sialic acid
derivative 416 by treatment with CMP-sialic acid synthetase17 in
the presence of CTP (cytidine-50-triphosphate) in 70% yield. In
case for the synthesis of 2, 5-azidoacetamide sialic acid derivative
818 prepared using 5-azidoacetamido-3,5-dideoxy-D-glycero-β-
D-galacto-non-2-thiophenyl-1-methyl ester (6) was subjected to

5610 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Hosoguchi et al.
Figure 2. Azidosugar nucleotides used in this study and general scheme for the click reaction (A), notably 1,3-dipolar cycloaddition in a mild
aqueous solution, between these azidosugar nucleotides and commercially available alkynes (B) allowed for the construction of focused
compound library.
the enzymatic reaction with CTP under a similar condition with
that of the synthesis of 1 to afford 2 in 50% yield. It is clear that
use of CMP-sialic acid synthetase greatly facilitates synthetic
procedure of various CMP-Neu5Ac analogues and the tolerance
observed in the reactions with 5-azidoacetamide and 9-azido
sialic acid derivatives17 should encourage us to expand this
method to the synthesis of much more complicated sialic acid-
related compounds. 3 was synthesized through a key intermedi-
ate, 6-azide-1,2,3,4-tetra-O-benzoyl-6-deoxy-β-D-galactopyranose
derived from D-galactose as a starting material, according to the
procedure described in our previous report.19 Azidosugar nu-
cleotides were employed for 1,3-dipolar cycloaddition reaction
with commercially available 36 kinds of alkynes containing
typical aromatic, aliphatic, hydrophilic, carboxyl, amino
groups, or steroid scaffolds under a mild aqueous solution
according to the condition reported previously.11 As seen from
the plausible yields of the click reactions determined by ESI-MS
(Supporting Information), it seems that yields in the reaction

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5611
Scheme 1. Synthetic Route of Azidosugar Nucleotide 1 from Sialic Acida
a Reagents and conditions: (a) (i) TFA, MeOH, rt. 12 h, (ii) TsCl, pyridine, rt, 12 h, 56%; (b) NaN3, acetone/H2O, reflux, 10 h, 75%; (c) CMP-sialic
acid synthetase, CTP, pH 9.0, 5 h, 70%.
Scheme 2. Synthetic Route of 2 from Sialic Acida
a Reagents and conditions: (a) (i) Dowex 50W X8 (Hþ), MeOH, rt, 1 d, (ii) Ac2O, pyridine, rt, 3 d, 90%; (b) PhSH, BF3, Et2O, CH2Cl2, rt, 2 d, 89%; (c)
CH3SO3H, MeOH, 60 °C, 2 d, 51%; (d) (i) N3CH2CO2Na, DPPA, Et3N, DMF, rt, 12 h, (ii) NaOMe, MeOH, rt, 1.5 h, 69%; (e) NBS, acetone/H2O, 0 °C
to rt, 1 h, 75%; (f) 1 N NaOH aq, 1 h, qy; (g) (i) CMP-sialic acid synthetase, CTP, pH 9.0, 37 °C, 2 h, (ii) calf intestine alkaline phosphatase (Takara Bio),
pH 9.0, 37 °C, 2 h, 50%.
between 1-3 and alkynes depend significantly on the reactivity
of an azide functional group substituted at C5, C6, or C9
positions of sugar residues, and steric hindrance effect during
the 1,3-dipolar cycloaddition reaction with various alkynes.
Therefore, we tested the effects of click reagents such as CuSO4,
TBTA, sodium ascorbate, and DMSO in order to set an
appropriate criteria in the high throughput screening of the
inhibitory effects by these click products.
aminooxy-WR-OCH3/OCD3, 13 or 14,14a,b because the term-
inal N-acetyllactosamine unit generated by galactosylation with
βGal-T1 should become a highly convenient acceptor substrate
for a lot of major GTs such as R2,3/2,6-sialyltransferases, β1,3-
N-acetylglucosaminyltransferase, and R1,3-fucosyltransferases
known as crucial human glycosyltransferases. In addition, we
employed a commercially available glycoamino acid carrying a
typical biantennary N-glycan core21 as a starting material that
can be converted into 11 and 12, doubly probed functional
glycopeptides, by solid phase glycopeptide synthesis and selec-
tive modification at C-terminal lysine residue with stable isotope
moiety (Supporting Information).
D. MS-Based High Throughput Screening of Glycosyl-
transferases Inhibitors. Focused compound library derived
from 1-3 and A1-A36 were subjected to high throughput
screening of inhibitory effects on sugar transferring activities by
using four glycosyltransferases in the presence of 9-12, desig-
nated acceptor substrates, according to the general protocol
illustrated in Figure 1. Advantage of MS-based protocol used in
this study is that crude click products can be employed directly
for the first screening to survey potent candidates exhibiting
significant inhibition when other components such as alkynes,
CuSO4, tris-(benzyltriazolylmethyl)amine (TBTA), and sodium
ascorbate have no or little influence to this assay system
(Supporting Information).11 As anticipated, 9-12 exhibiting
enhanced ion sensitivity allowed for rapid and large-scale screen-
ing to identify some potential compounds from crude click
products without any purification, notably focused compound
library derived from 36 alkynes and three azidosugar nucleotides
(Figure 2). It was demonstrated that analysis of inhibitory effect
by using large number of compounds can be performed quanti-
tatively by focusing the ion intensities of a pair of expected
glycosylated products at m/z 1511.3 and m/z 1514.3 for R2,3/
R2,6-STs, at m/z 1366.4 and m/z 1369.4 for R1,3-FucT, and at
m/z 2386.9 and m/z 2389.2 for R1,6-FucT (Figure 4).
C. Double Probing of Glycosyl Acceptor Substrates. Carbo-
hydrates generally need to be modified chemically by some
sensitive labeling groups such as photosensitive/fluorescent
groups or radioactive isotopes when they are used as substrates
for monitoring the reactions catalyzed by GTs and GHs. Mass
spectrometry is an attractive analytical method because changes
of the mass (molecular weight) from substrate to product can be
directly detected during enzymatic reactions without special
probing. Because of the poor potentials of the ionization and
ion suppression effect, however, it is usually difficult to detect
ions elaborated from free (unmodified) carbohydrates in the
presence of many other biological substances such as peptides,
nucleotides, lipids, and other metabolites. In the course of our
studies on the comprehensive glycomics based on glycoblotting
technology,13,14b,20a,b it was demonstrated that carbohydrates
can be predominantly detected by tagging the reducing end with
some ion-sensitivity enhancement reagents such as aminooxy-
tryptophanylarginine (ao-WR)14a,b or some similar peptides
having hydrophobic and basic amino acid residues. As illu-
strated in Figure 1, we considered that novel glycosyl acceptors
having such improved ion-sensitivity would greatly facilitate
MS-based protocol for monitoring glycosylation by GTs,
trimming by GHs, and inhibitory effects by large numbers of
synthetic and naturally occurring compounds. Figure 3 shows
the chemical structures of glycosyl acceptor substrates used
in this study. In the present study, 9 and 10 were prepared
from chitotriose as a starting material by conjugating with

5612 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Hosoguchi et al.
Figure 3. Glycosyl acceptor substrate sets designed for the characterization of recombinant rat R2,3-ST/ST3Gal III and R2,6-ST/ST6Gal
I and recombinant human R1,3-FucT/FUT V and R1,6-FucT/FUT VIII. WR represents dipeptide tryptophanyl-arginine moiety as an ion-
sensitivity enhancement reagent.
As summarized in Figure 5, the results of MALDI-TOFMS-
based high throughput screening suggest that 17 compounds
[3 compounds for R2,3-ST (Figure 5A,B), four compounds for
R1,3-FucT (Figure 5E), and 10 compounds for R1,6-FucT
(Figure 5F)] among 108 click entries showed potent inhibitory
effects judged by a tentative criteria (relative inhibition rate
>50%). Most derivatives from 1 did not exhibit significant
inhibition against R2,3/R2,6-STs, and they seemed to become
donor substrates of both enzymes because signals corresponding
to the expected products were detected during the inhibition
assay even in the presence of native CMP-Neu5Ac. In addition,
we could not detect any significant inhibitory effect on recom-
binant R2,6-ST by derivatives from 2, indicating that modifica-
tions both at C-5 and C-9 positions of sialic acid do not influence
dynamic mechanism in the glycosylation catalyzed by this
enzyme. On the contrary, it was also revealed that modification
at C-5 position of CMP-Neu5Ac has strong impact to the
catalytic action by R2,3-ST when bulky steroid analogues such
as A17 and A19 were subjected to the click reaction with 2
(Figure 5B). However, the click product between compound 2
and a similar steroidal alkyne A18 had little inhibitory effect,
indicating that the terminal aromatic moiety of A17 and A19
appears to be crucial in addition to the steroidal skeleton for this
activity rather than R,β-unsaturated ketone structure of A18.
These results suggest that 3D structure of the donor binding site
and dynamic mechanism including function of a flexible loop of
mammalian R2,6-STs might be quite different from those of
R2,3-STs, although both R2,3/R2,6-STs share a natural CMP-
Neu5Ac as a redundant donor substrate. As shown in
Figure 5E, it was revealed that activity of R1,3-FucT is inhibited
significantly by four click products derived from 3 with A6, A17,
A33, or A35 having some bulky alcohols. No steroidal com-
pound showed significant inhibition against R1,6-FucT, while
simple aromatic and aliphatic 10 alkynes, A1, A2, A3, A4, A6,
A8, A9, A21, A22, and A28 provided 3 with potent inhibitory
effects (Figure 5F). Interestingly, any compounds derived from
3 did not become substrate for both R1,3- and R1,6-FucTs in
contrast to the results observed in cases for R2,3-ST/R2,6-ST
and compounds derived from 1 or 2.
To evaluate potent inhibitory effects more precisely, we con-
cluded that c 15 (derived from 2 and A17), 16 (derived from 3
and A17), and 17 (derived from 3 and A2) among 17 hit deriva-
tives should be synthesized by simply scale-up in the click
reactions and employed for further in vitro characterization
(Figure 6). It was demonstrated that 15 is highly promising inhi-
bitor against R2,3-ST (ST3Gal III) (IC50= 8.2 μM) as shown in
Figure 7A. Surprisingly, this inhibitor was proved to be a good
donor substrate for R2,6-ST (ST6Gal I) (Km = 125 μM) from
the LB plot indicated in Figure 7B. These results clearly show
that 15 is the first highly selective inhibitor for mammalian R2,3-
ST. It should also be noted that the binding mode of R2,6-ST
with CMP-Neu5Ac appears to be different from that of R2,3-
ST, in which chemical substitution at C-5 and C-9 positions of
CMP-Neu5Ac did not interfere sugar transfer reaction toward
bulky glycosyl acceptor substrates. This may suggest that major
parts of sialic acid moiety are exposed to outer side from the
CMP-Neu5Ac bound in the catalytic pocket of R2,6-ST,
although cytidine and negative charges locating around the
anomeric carbon seem to be crucial for the interaction with
R2,6-ST.22 In contrast, modification at C-5 position of Neu5Ac

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5613
Figure 4. MALDI-TOFMS-based rapid and highly efficient quantification of inhibitory effects by focused compound library. Ion intensities
of a pair of expected glycosylated products should be determined at m/z 1511.3 and m/z 1514.3 for R2,3/R2,6-STs (A) and (B), at m/z 1366.4 and
m/z 1369.4 for R1,3-FucT (C), and at m/z 2386.9 and m/z 2389.2 for R1,6-FucT (D), respectively.
greatly affected the activity of R2,3-ST, indicating that this class
of enzymes recognize sialic acid residue more precisely than
R2,6-STs through the interaction with C-5 acetamide group as
well as above essential motifs of CMP-Neu5Ac. Considering
that many inhibitors inspired from transition-state analogues
during glycosylation catalyzed by R2,3-/R2,6-STs showed much
higher inhibitory effects on R2,6-STs rather than R2,3-STs,22 we
may conclude that there is no discrepancy between the present
results and these previous pioneering reports. It was also
documented that most bisubstrate analogous inhibitors having
designated pseudo R2,6- or R2,3-glycoside linkage exhibited
potent but a similar level inhibition toward both enzymes.6,22
In addition, it was also reported that CMP shows almost similar
level of inhibitory effects on rat R2,3-ST (Ki = 90 μM)23a,b and
R2,6-ST (65 μM),23b respectively. Therefore, our present data
clearly suggest that modification at C-5 position of Neu5Ac

5614 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Hosoguchi et al.
Figure 5. High throughput screening of inhibition by 108 click trials products from 1-3 and 36 kinds of commercially available alkynes. (A) and (B)
show the effects of compounds derived from 1 and 2 on R2,3-ST, (C) and (D) exhibit the effects of compounds derived from 1 and 2 on R2,6-STs, and
(E) and (F) represent the effects of compounds derived from 3 on R1,3-Fuc and R1,6-Fuc, respectively. In (A-F), the results represented by gray-
colored panel mean the compounds were not consumed as donor substrate, while others (white panel) became substrate during inhibition assay.
without loss of an original amide linkage should allow for ideal
molecular design and further optimization study of highly
selective inhibitors against various R2,3-STs.
Inhibitory effects by 16 and 17 were also estimated as Ki =
293 nM (vs R1,3-FucT) and Ki=13.8 μM (vs R1,6-FucT), res-
pectively (Supporting Information). As anticipated from the
scores in the high throughput screening by means of click deri-
vatives from 3 (Figure 5E,F), both compounds were proved to
be highly potential and selective inhibitors for two important
classes of human FucTs. Difference in the structural features
between 16 and 17 derived from the same azidosugar nucleotide
suggests that the strong inhibition by these compounds against
R1,3-FucT and R1,6-FucT must be conducted in a different
manner, although the formation of triazene ring at C-6 position
of L-fucose residue seems to be required to acquire the minimal
inhibitory activity.6f It is also clear from the results shown in
parts E and F of Figure 5 that all click products derived from 3
were not consumed as donor substrate by both R1,3/R1,6-FucTs.
Given the fact that a similar derivative from 3, 6-deoxy-6-N-
(2-naphthalene-2-yl-acetamide)-β-L-galactopyranos-1-yl-gua-
nosine 50-diphosphate disodium salt,19 becomes an excellent
substrate for R1,3/R1,6-FucTs, difference in the anchor moiety
Figure 6. Selected compounds 15-17 showing the highest inhibi-
between GDP-Fuc and hydrophobic elements, namely amide
tory effects against R2,3-ST, R1,3-FucT, and R1,6-FucT among 108
and triazene ring, markedly influence the dynamic interaction
with these enzymes. Because GDP was proven to inhibit both
human R1,3-FucT (Ki = 16 μM and IC50 = 50 μM)24a,b and
click trials using azidosugar nucleotides 1-3 and 36 kinds of
alkynes. Synthetic procedures and characterizations of these com-
pounds (see, Experimental Section and Supporting Information).

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5615
the amount of enzyme that transfer 1.0 μmol of sialic acid from
CMP-Neu5Ac to N-acetyllactosamine per min at 37 °C, pH 7.4),
R 2,6-(N)-sialyltransferase (R2,6-ST, one unit of this enzyme is
defined as the amount of enzyme that transfer 1.0 μmol of sialic
acid from CMP-Neu5Ac to N-acetyllactosamine per min at 37 °C,
pH 6.0), and recombinant human R1,3-fucosyltransferase V (R1,3-
FucT/FUT V, one unit of this enzyme is defined as the amount of
enzyme that will transfer 1.0 μmol of fucose from GDP-Fuc to
N-acetyllactosamine per min at 37 °C, pH 7.5) were purchased from
Calbiochem (San Diego, CA). Recombinant human R1,6-fucosyl-
transferase VIII (R1,6-FucT/FUT VIII, one unit of this enzyme is
defined as the amount of enzyme that will transfer 1.0 μmol of
fucose from GDP-Fuc to asparagine-linked N-acetyl-β-D-glucosa-
mine of the chitobiose moiety per min at 37 °C, pH 7.5) and re-
combinant human β1,4-galactosyltransferase (β1,4-GalT/βGal-T I,
one unit of this enzyme is defined as the amount of enzyme that will
transfer 1.0 μmol of galactose from UDP-gal to D-glucose per min at
30 °C, pH 8.4 in the presence of R-lactoalbumin) were purchased
from Toyobo Co., Ltd. Recombinant influenza CMP sialic acid
synthetase (CSS, one unit of this enzyme is defined as the amount of
enzyme that transfer 1.0 μmol of CMP from CTP to sialic acid per
min at 37 °C, pH 8.0) was provided from Yamasa Co (Japan).
Recombinant calf intestine alkaline phosphatase (one unit of this
enzyme is defined as the amount of enzyme that decomposes 1.0
μmol of p-nitrophenyl phosphate per min at 37 °C, pH 9.8) was
purchased from Takara Bio.
6-Azide-6-deoxy-β-L-galactopyranos-1-yl-guanosine 50-diphos-
phate disodium salt (3) was synthesized from D-galactose as a start-
ing material according to the procedure reported previously by our
group.19 Acetylene compounds (alkynes represented as A1-A36)
were purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo,
Japan). Solvents and other reagents for chemical synthesis were
purchased from Aldrich Chemical Co. (Milwaukee, WI), Calbio-
chem (San Diago, CA), Seikagaku Co. Ltd. (Tokyo, Japan), Tokyo
Chemical Industry (Tokyo, Japan), and Wako Pure Chemical
Industries Ltd. (Tokyo, Japan) and used without further purifica-
tion. Proton and carbon NMR was recorded with Bruker Lambda
600 MHz (Bruker BioSpin Corp., Germany). Chemical shifts are
given in ppm and referenced to internal TMS (δH 0.00 in CDCl3),
CHCl3 (δH 7.26 in CDCl3), TSP (δH 0.00 in D2O), or CDCl3 (δC
77.00). Assignments in 1H NMR were made by first-order analysis
of the spectra and were verified by H-H COSY and HMQC
experiments. Elemental analyses were performed with MT-6
CHN CORDER (Yanako, Japan). TLC was performed on Merck
precoated plates (20 cm ꢀ 20 cm; layer thickness, 0.25 mm; Silica
Gel 60F254); spots were visualized by spraying a solution of 90:5:5
(v/v/v) MeOH-p-anisaldehyde-concentrated sulfuric acid and heat-
Figure 7. 15 is a heterobifunctional molecule. (A) Relationship bet-
ween concentration of 15 (0.39-200 μM) and relative inhibitory effect to
estimate IC50 against R2,3-ST (0.5 mU) in the presence of CMP-Neu5Ac
(200 μM), and 9 (1 mM). (B) LB plot of the reaction catalyzed by R2,6-
ST (0.5 mU) in the presence of 15 (12.5-200 μM) and 9 (1 mM).
human R1,6-FucT (Ki =36μM)24c in a similar manner, it is clear
that enzyme-selective inhibitors become nice tools for investigat-
ing functions and mechanism of individual glycosyltransferases.
However, it is thought that considerable efforts should be paid
toward practically potent drug candidates through further mo-
lecular design and optimization for controlling properties such as
stability, organ/tissue targeting, and cellular permeability.
Conclusion
An efficient and versatile approach for searching potent
inhibitors against glycosyltransferases was established on the
basis of high throughput mass spectrometry. Double probing
of common glycosyl acceptor substrates with an ion-sensitiv-
ity enhancement reagent and stable isotope tag allowed for
highly sensitive and quantitative screening of large number of
compounds prepared by click reactions.11 Advantage of the
present MS-based protocol is evident because direct monitor-
ing molecular mass of interest would greatly facilitate data
annotation/analysis when the method makes highly sensitive
and quantitative analysis of the target ion possible. Actually,
we have demonstrated that MALDI-TOFMS-based screen-
ing is feasible for the assay using heterogeneous cell lysate in
cases for profiling whole kinases activity of human cancer
K562 cells.12 Therefore, it is ourbelief that the present strategy
would greatly facilitate lead discovery research toward new
generation drug candidates controlling posttranslational
modifications such as protein phosphorylation, glycosylation,
and other important modification processes by means of cell-
based experiments as well as in vitro common assay systems.
1
ing at 180 °C for ca. /2 min, a solution of 95: 5 (v/v) MeOH-
concentrated sulfuric acid and heating at 180 °C for ca. 1/2 min, and
by UV light (256 or 365 nm) when applicable. All new compounds
were examined by HRMS and/or HPLC, and the purity was proven
to be more than 95%. Column chromatography was performed on
Silica Gel 60 (spherical type, particle size 40-50 μm; Wako Pure
Chemical Co. Ltd.), Iatrobeads (6RS-8060, Jatron Laboratories,
Inc.), Sephadex G-15, Sephadex G-10 (GE Healthcare Bio-Sciences
Corp., NJ), or Wako-gel C-18 (Wako Pure Chemical Co. Ltd.) with
the solvent systems specified, and the ratio of solvent systems was
gived in v/v. High performance liquid chromatography (HPLC)
was conducted by HITACHI L-7100 HPLC system equipped
with a Inertsil-ODS column (4.6 mm ꢀ250 mm, GL Sicence Inc.)
and HITACHI L-7420 UV-vis detector.
Samples for MALDI-TOFMS were desalted and concentrated
using 10-μL C18 ZipTips (Millipore) according to the manufac-
turer’s instructions. Typically, samples were dissolved in 1 mL of
90% (v/v) acetonitrile containing trifluoroacetic acid and mixed
with the same volume of a saturated solution of 2,5-dihydroxu-
benzoic acid (DHB) in 33% acetonitrile containing 0.1% trifluoro-
acetic acid. The above mixtures (1 μL) were applied to a polished
stainless steel target MALDI plate and air-dried before analysis in
the mass spectrometer. All measurements were performed using an
Experimental Section
General Methods and Materials. Recombinant rat R2,3-(N)-
sialyltransferase (R2,3-ST, one unit of this enzyme is defined as

5616 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Hosoguchi et al.
Ultraflex TOF/TOF mass spectrometer equipped with a reflector
and controlled by the Flexcontrol 1.2 software package (Bruker
Daltonics GmbsH, Bremen, Germany). Ions generated by a pulsed
UV laser beam (nitrogen laser, λ = 337 nm) were accelerated to a
kinetic energy of 23.5 kV. External calibration of MALDI-mass
spectra was carried out using singly charged monoisotopic peaks of
a mixture of human angiotensin II (m/z 1046.542), bombesin (m/z
1619.823), ACTH (m/z 2465.199), and somatostain 28 (m/z
3147.472). The mixture of these peptides was measured on the
central spot of a 3 ꢀ 3 square with external calibration. To achieve
mass accuracy better than 60 ppm, internal calibration was carried
out by doping the matrix solution with a mixture of the calibration
peptides. Calibration of these mass spectra was performed auto-
matically utilizing a customized macro command of the XMASS
5.1.2 NT software package. The macro command was used for the
calibration of the monoisotopic singly charged peaks of the above-
mentioned peptides.
Synthesis. Cytidine-50-monophospho-5-acetamido-9-azido-3,5,
9-trideoxy-β-D-glycero-r-D-galacto-non-2-ulosonic acid (1). To
the solution of 416 (50 mg, 150 μmol) in buffer (6.0 mL, 100
mM Tris-HCl, pH 8.8, 20 mM MgCl2) was added CTP (3Naþ)
(79 mg, 165 μmol) and CMP-sialic acid synthetase (70 μL of 7
U/100 μL, Campylobacter influenzae, Yamasa Co.) and incubated
at 37 °C for 5 h. During the reaction, 1 N NaOH was added to keep
pH of the reaction mixture. The reaction mixture was filtered
and concentrated, and the residue was purified by reverse phase
column chromatography (WakoGel C-18) and ion exchange
chromatography (Diethylaminoethyl-sepharose (DEAE): 0.05-
0.1 M NH4þHCO3-) to give 1 in 53% yield (52 mg, 80 μmol).
NMR data was identical with that of previous paper.17 1H NMR
(600 MHz, D2O, 27 °C): δ 1.66(dt, 1H, J3ax,4 13.2Hz, J3ax,3eq 13.2
Hz, H-3ax), 2.07 (s, 3H, Ac), 2.50 (dd, 1H, J3eq,4 4.7 Hz, J3ax,3eq
13.2 Hz, H-3 equiv), 3.48 (d, 1H, J7,8 9.6 Hz, H-7), 3.50-3.53 (m,
1H, H-9), 3.64-3.66 (m, 1H, H-90), 4.09 (t, 1H, J5,6 10.4 Hz, J4,5
10.4 Hz, H-5), 4.19-4.23 (m, 2H, H-4, H-8), 4.28 (d, 1H, J5,6 10.5
Hz, H-6), 4.25-4.35 (m, 5H, H-2 of ribose, H-3 of ribose, H-4 of
ribose, H-5 of ribose, H-50 of ribose), 5.99 (d, 1H, J1,2 4.2 Hz, H-1
of ribose), 6.21 (d, 1H, J5,6 7.7 Hz, H-5 of cytosine), 8.07 (d, 1H,
J5,6 7.7 Hz, H-6 of cytosine).
5-Azidoacetamido-3,5-dideoxy-D-glycero-β-D-galacto-non-2-
thiophenyl-1-methyl Ester (6). To a solution of methyl 2-azidoa-
cetate (470 mg, 4.08 mmol) in 95% EtOH (1 mL) was added 1 N
NaOH aq (4.08 mL, 4.08 mmol) and stirred at room temperature
for 3 h. Then the solution was added to 525 (1.59 g, 3.89 mmol) in
95% EtOH (10 mL). After stirring at rt for 7 h, the solution was
concentrated and coevaporated with DMF three times. The residue
was dissolved in DMF (15 mL), and the solution was added to Et3N
(1.14 mL, 8.18 mmol) and diphenylphosphorylazide (880 μL, 4.08
mmol). After stirring at room temperature for 12 h, the reaction
mixture was added to MeOH (58 mL) and NaOMe (28% MeOH
solution, 725 μL, 3.76 mmol). After 1.5 h, the reaction mixture was
neutralized with Dowex 50WX8 (Hþ) and filtered and concentrated
in vacuo. The residue was purified on a column of silica gel (CHCl3/
MeOH = 20/1) to give compound 6 in 69% yield (1.22 g, 2.67
mmol) as a syrup. Rf = 0.6 (CHCl3/MeOH = 3/1). 1H NMR δ
(600 MHz, CD3OD) 2.00 (t, 1H, J3ax,412.0 Hz, H-3ax), 2.70 (dd,
1H, J3ax,3eq 13.6 Hz, J3eq,4 4.68 Hz H-3 equiv), 3.57 (d, 1H, J7,8 9.12
0.44 mmol) at 0 °C. After stirred at 0 °C for 1 h, the reaction
mixture was filtered on a silica gel plaque (CHCl3/MeOH = 3/1)
and concentrated. The residue was purified on a column of silica
gel (CHCl3/MeOH = 10/1 to 5/1) to yield compound 7 (30 mg,
75%). NMR data was identical with published data.18 1H NMR
(600 MHz, CD3OD, 27 °C): δ 1.88 (t, 1H, J3ax,413.2 Hz, J3ax,3eq
13.2 Hz, H-3ax), 2.22 (dd, 1H, J3ax,3eq 12.9 Hz, J3eq,4 4.8 Hz H-3
0
equiv), 3.49 (d, 1H, J7,8 9.0 Hz, H-7), 3.62 (q, 1H, J8,9 5.4 Hz, J9,9
13.0 Hz, H-9), 3.68-3.71 (m, 1H, H-8), 3.78 (s, 3H, Me),
3.79-3.81 (m, 1H, H-90), 3.86-3.97 (m, 3H, H-5, N3-CH2-CO),
4.05-4.10 (m, 2H, H-4, H-6).
5-Azidoacetamido-3,5-dideoxy-D-glycero-r,β-D-galacto-non-
2-ulosonic Acid (8).18 To a solution of 7 (52 mg, 114 μmol) in
H2O (1.0 mL) was added 1 N NaOH aq (125 μL, 125 μmol).
After stirred at rt for 1.5 h, the solution was neutralized with
Dowex 50WX8 (Hþ). After filtration, the solution was concen-
trated in vacuo to yield compound 8 (50 mg, qy). NMR data was
identical with that of previously published.18 1H NMR (600
MHz, D2O, 27 °C): δ 1.91 (t, 1H, J3ax,412.0 Hz, J3ax,3eq 12.0 Hz,
H-3ax), 2.31 (dd, 1H, J3ax,3eq 12.9 Hz, J3eq,4 5.4 Hz H-3 equiv),
0
3.58 (d, 1H, J7,8 9.0 Hz, H-7), 3.67 (q, 1H, J8,9 6.6 Hz, J9,9 12.0
0
Hz, H-9), 3.80-3.82 (m, 1H, H-8), 3.90 (dd, 1H, J8,9 2.4 Hz, J9,9
11.7 Hz, H-90), 4.05 (t, H-5, J5,6 10.2 Hz, J4,5 10.2 Hz), 4.13-4.15
(m, 4H, N3-CH2-CO, H-4, H-6).
Cytidine-50-monophospho-5-azidoacetamide-3,5-dideoxy-β-D-
glycero-r-D-galacto-non-2-ulosonic Acid (2). To the solution of 8
(10 mg, 29 μmol) in buffer (2 mL, 100 mM Tris-HCl, pH 9.0,
20 mM MgCl2) was added CTP (3Naþ) (13.6 mg, 26 μmol) and
CMP-sialic acid synthetase (14.3 μL of 7 U/100 μL) and
incubated at 37 °C for 2 h. The reaction mixture was added
buffer (3 mL, 900 mM Tris-HCl, pH 9.0) and alkaline phospha-
tase (50 μL of 1 U/μL, calf intestine, Takara Bio). After
incubation at 37 °C for 2 h, the reaction mixture was filtered
through a cotton plaque. The solution was applied to a column
of Sephadex G15 and eluted with H2O. Fractions containing the
product were lyophilized. The residue was dissolved in water
and purified by ion exchange chromatography (DEAE:H2O to
0.1 M NH4þHCO3-) to give compound 2 in 51% yield (9.9 mg,
15 μmol). NMR data was in good agreement with that pre-
viously reported.18 1H NMR (600 MHz, D2O, 27 °C): δ 1.72 (dt,
1H, J3ax,4 12.3 Hz, J3ax,3eq 12.3 Hz H-3ax), 2.56 (dd, 1H, J3eq,4
4.2, J3ax,3eq 13.2 Hz, H-3 equiv), 3.50 (d, 1H, J7,8 9.6 Hz, H-7),
0
0
3.68 (q, 1H, J8,9b 6.6, J9,9 12.0 Hz, H-9), 3.94 (d, 1H, J9,9 11.1
Hz, H-90), 4.00 (m, 1H, H-8), 4.08(t, 1H, J5,6 10.8 Hz, J4,5 10.8
Hz, H-5), 4.15-4.21 (m, 3H, H-4, N3-CH2-CO), 4.27-4.33 (m,
4H, H-6, H-4 of ribose, H-5 of ribose, H-50 of ribose), 4.36 (t, 1H,
J1,2 4.2 Hz, J2.3 4.2 Hz H-2 of ribose), 4.40 (t, 1H, J2,3 4.2 Hz, J3,4
4.2 Hz, H-3 of ribose), 6.05 (d, 1H, J1,2 4.8 Hz, H-1 of ribose),
6.19 (d, 1H, J5,6 7.2 Hz, H-5 of cytosine), 8.04 (d, 1H, J5,6 7.2 Hz,
H-6 of cytosine).
Cytidine-50-monophospho-5-(2-(4-((8R,9S,13S,14S,17S)-3,17-
dihydroxy-13-methyl-7,8,9,11,12,13,14,15,16,17-decahydro-6H-
cyclopenta[a]phenanthren-17-yl)-1H-1,2,3- triazol-1-yl)acetamido)-
3,5-dideoxy-β-D-glycero-r-D-galacto-non-2-ulosonic Acid (15). The
solution containing compound 2 (2.0 mM), A17 (2.0 mM), 1 mM
CuSO4, 10 mM sodium ascorbate, 1 mM TBTA, 40% t-BuOH,
10% DMSO, and 20 mM Tris-HCl (pH 7.6) in total volume 4.0
mL was incubated for 6 h at rt. The mixture was concentrated by
speedvac (25 °C, 6 h) and the residue was dissolved in water, and
the solution was filtered by membrane filter. The solution was then
purified with reversed-phase HPLC to give compound 15 (2.69mg,
39%). 1H NMR δ (600 MHz, D2O), 1.06 (s, 3H), 1.27-1.53 (m,
4H), 1.63-1.71 (m, 2H), 1.72 (td, 1H, H-3ax), 1.86 (br, 1H),
1,95-2.00 (br, 2H), 2.16-2.20 (m, 2H), 2.47-2.52 (m, 1H), 2.56
(dd, 1H, H-3 equiv), 2.80 (s, 2H), 3.70 (q, 1H, H-9b), 3.94 (d, 1H,
H-9a), 4.00 (m, 1H, H-8), 4.06 (t, 1H, H-5), 4.21(td, 1H, H-4),
4.29(m, 4H, H-6, H-4 of ribose, H-5a of ribose, H-5b of ribose),
4.34 (t, 1H, H-2 of ribose), 4.39 (br, 1H, H-3 of ribose), 5.38 (s, 2H,
methylene at acetamide), 6.03 (d, 1H, H-1 of ribose), 6.17 (d, 1H,
H-5 of cytosine), 6.65 (s, 1H, aromatic), 6.69 (d, 1H, aromatic),
0
Hz, H-7), 3.68 (q, 1H, J8,9 5.4 Hz, J9,9 11.3 Hz, H-9), 3.77-3.80 (m,
0
0
1H, H-8), 3.83 (dd, 1H, J8,9 2.82 Hz, J9,9 11.3 Hz, H-9 ), 3.93-4.01
(m, 3H, H-5, N3-CH2-CO), 4.17 (td, 1H, J3eq,4 4.62 Hz, J4,5 10.7 Hz,
J
3ax,4 10.8 Hz, H-4), 4.60 (d, 1H, J5,6 10.6 Hz, H-6), 7.33-7.39 (m,
3H, aromatic), 7.59 (d, 2H, aromatic). 13C NMR δ (CD3OD) 42.27
(C-3),53.10-53.14 (methyl ester and methylene of azideacetamide),
54.24 (C-5), 65.21 (C-9), 68.09 (C-4), 70.72 (C-7), 71.46 (C-8), 73.27
(C-6), 91.59 (C-2), 130.02-137.41 (thiophenyl), 170.82 (C-1),
171.43 (C-10). HRMS (FAB) Anal. calcd for C18H25N4O8S
[M þ H]þ 457.1393; found 457.1386.
5-Azidoacetamido-3,5-dideoxy-D-glycero-β-D-galacto-non-2-
ulosonic-1-methyl Ester (7). To a solution of 6 (50 mg, 0.11 mmol)
in acetone (9.1 mL) and water (0.9 mL) was added NBS (78 mg,

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5617
7.16 (d, 1H, aromatic), 8.00 (s, 1H, triazole), 8.02 (s, 1H, H-6 of
cytosine). 13C NMR δ (125 MHz, D2O) 14.44, 23.67, 26.52, 27.57,
29.73, 33.30, 37.49, 39.84, 41.79, 43.55, 47.65, 48.72, 52.93, 63.68,
65.57, 67.42, 69.63, 70.05, 70.35, 72.25, 75.02, 83.23, 83.61, 89.75,
97.31, 125.79, 127.40, 139.49, 142.33, 154.05, 158.53, 166.92,
169.00, 174.94. HRMS (ESI) Anal. calcd for C40H53N7O18P
[M - H]- 950.31847; found 950.31749.
to 10 mM. The standard stock solutions of pure acceptor
substrates 9 and 10 were stored at -20 °C as mother solution,
respectively, HRMS (ESI): 9, calcd for C50H79O25N10 [M þ H]þ
1219.5212, found 1219.5218; 10, calcd for C50H76D3O25N10
[M þ H]þ 1222.5401, found 1222.5406 (see also Supporting In-
formation for the calibration curve to indicate the linearity
between mixed rate and rate of MS intensity).
Compound s 11 and 12 were customized products prepared
by Sigma Genosys (Ishikari, Japan) using common solid phase
peptide synthesis (SPPS) on Fmoc 4-hydroxymethylphenoxya-
cetyl (HMPA)-PEGA resin with Fmoc-Asn(heptasaccharide)-
OH,21 Z-Trp-OH, general Fmoc amino acids, and coupling
reagents (benzotriazol-1-yl-N-tetramethyl-uronium tetrafluor-
oborate (TBTU)/1-hydroxy-1H-benzotriazole (HOBt) or N,N0-
diisopropyl carbodiimide (DIC)/HOBt). In the final step after
releasing the N-glycopeptides from resin by treating with TFA,
normal or deuterated acetyl group was incorporated at ε-amino
group of C-terminus lysine residue by using (CH3CO)2O or
(CD3CO)2O. HRMS (ESI): 11, calcd for C96H143O46N15 [M þ
2H]2þ 1120.9656, found 1120.9633; 12, calcd for C96H140D3O46-
N15 [M þ 2H]2þ 1122.4750, found 1122.4732 (see also Support-
ing Information for demonstrating purity of these compounds
and the calibration curve to indicate the linearity between mixed
rate and rate of MS intensity).
Acceptors 10 and 12 (deuterium-labeled compounds) were
subjected to the preparation of internal standards to be em-
ployed in the following inhibition assays by using corresponding
glycosyltransferases and sugar nucleotides (see Figure 4).
Standard Conditions for Click Reactions Tested in High
Throughput Screening. Condition for 1 and 2: Each click reaction
was carried out in 1: 4: 5 (v/v/v) DMSO-t-BuOH-water (50 μL), in
the presence of alkynes (A1-A36, 2 mM), 1 or 2 (2 mM), CuSO4
(1 mM), tris-(benzyltriazolylmethyl)amine (TBTA, 1 mM),
sodium ascorbate (10 mM), and Tris-HCl (20 mM, pH 7.6).
The mixture was kept at room temperature for 6 h and then
concentrated by speedvac (25 °C, 3 h). To the residue was added
2 mM EDTA (2Naþ) to a total volume 100 μL. The solutions
were stored at -20 °C as stock solutions. The yields of all click
reactions were summarized in Supporting Information.
Condition for 3: 4: 6 (v/v) DMSO-water (25 μL), in the pre-
sence of alkynes (A1-A36, 2 mM), 3 (2 mM), CuSO4 (1 mM),
tris-(benzyltriazolylmethyl)amine (TBTA, 1 mM), and sodium
ascorbate (5 mM). The mixture was kept at room temperature,
and the reactions were monitored by ESI-MS. After 4 h, 25 μL of
water was added to adjust the concentration to 1 mM. The
solutions were stored at -20 °C as stock solutions. The yields of
all click reactions and the effects of free alkynes and reagents on
the glycosylation by R1,3/R1,6-FucTs were summarized in the
Supporting Information.
6-Deoxy-6-(2-(4-((8R,9S,13S,14S,17S)-3,17-dihydroxy-13-
methyl-7,8,9,11,12,13,14,15,16,17-decahydro-6H-cyclopenta-
[a]phenanthren-17-yl)-1H-1,2,3-triazol-1-yl))-β-L-galactopyranos-
1-yl-guanosine 50-Diphosphate Disodium Salt (16). Azidosugar
nucleotide 325b (8.5 mg, 12.6 μmol) was dissolved in 2 mL of
H2O, 1 mL of DMSO, and 1 mL MeOH. The solution was added
alkyne A17 (1.1 equiv, 3.7 mg), CuSO4 (0.5 equiv, 1.0 mg), sodium
ascorbate (5.0 equiv, 12.5 mg), and TBTA (0.16 equiv, 1.1 mg) and
incubated for 1 day at rt. The mixture was subjected to the
purification by reverse phase chromatography (Wako gel C18,
eluted with water) to remove CuSO4 and TBTA, and the fractions
containing product were concentrated. The residue was finally
purified by HPLC equipped Inertsil ODS-3 column (4.6 mm ꢀ
250 mm, GL Science Inc.) and eluted by 3% acetonitrile/water con-
taining 20 mM ammonium bicarbonate. The residue was evapo-
rated to afford white amorphous powdery 16 (5.0 mg, 40%). 1H
NMR δ (D2O) 1.49 (3H, -CH3), 3.68 (2H, H-200, H-300), 3.88 (1H,
H-400), 4.11 (1H, H-500), 4.17 (2H, H-50, H-6R00), 4.31 (1H, H-50),
4.49 (1H, H-40), 4.67-4.76 (4H, H-20, H-30, H-60, H-6β00), 4.95 (1H,
H-100), 5.87 (1H, H-10), 6.47, 6.53, 6.90 (each 1H, aromatic), 8.07
(1H, triazole), 8.25 (1H, H-8). 13C NMR δ (D2O) 65.0 (C-6), 68.8
(C-400), 70.1 (C-40), 71.1 (C-300), 72.1 (C-200), 73.6 (C-60), 73.9
(C-500), 83.7 (C-50), 87.7 (C-10), 98.6 (C-100), 112.6, 115.0, 126.5
(aromatic). HRMS (ESI): calcd for C36H46O17N8P2Na [M - Na]-
947.2359; found 947.2326.
6-Deoxy-6-(1-methyl-4-(4-pentylphenyl)-1H-1,2,3-triazol-1-yl)-
β-L-galactopyranos-1-yl-guanosine 50-Diphosphate Disodium
Salt (17). Azidosugar nucleotide 319 (3.3 mg, 4.9 mmol) was
dissolved in 1:1 (v/v) of THF-water (1 mL). The solution was
added A2 (1.1 equiv, 1.0 mg), CuSO4 (0.4 mg), TBTA (2.5 mg),
and sodium ascorbate (4.9 mg) and incubated at rt for 5 h. The
solution was subjected to the chromatography on Wako-gel C18
and eluted by water. The fractions were concentrated. The residue
was chromatographed on Sephadex G-10 and eluted water. The
fractions were collected and lyophilized to afford white amor-
phous powdery 17 (2.5 mg, 61%). 1H NMR δ (D2O) 0.87 (3H,
-CH3), 1.21-1.31 (4H, CH2 ꢀ 2), 1.49-1.56 (2H, CH2),
2.41-2.55 (2H, CH2), 3.73 (2H, H-200, H-300), 4.03 (2H, H-400,
H-50), 4.19 (1H, H-6R00), 4.27 (1H, H-500), 4.32 (1H, H-6β00), 4.57
(1H, H-40), 4.67 (1H, H-6R0), 4.93 (1H, H-100), 5.75 (1H, H-10),
7.21, 7.71 (each 2H, aromatic), 7.92 (1H, triazole), 8.50 (1H, H-8).
13C NMR δ (D2O) 74.8 (C-100), 89.296 (C-10), 124.81 (triazole).
HRMS (ESI): calcd for C29H39N8Na2O15P2 [M þ H]þ 847.1806;
found 847.1801.
Preparation of Glycosyl Acceptor Substrates. Compound s 9
and 10 were synthesized from chitotriose as a starting material
as follows: Chitotriose (6.2 mg, 0.01 mmol) was dissolved in
1 mL of 100 mM acetate buffer (pH 4.0). The solution was
divided into 10 microfuge tubes by 100 μL. Aminooxy-WR-
OCH3/OCD3(13)/(14) (200 μL of 20 mM) and acetonitrile
(700 μL) were added to each tube. The mixture was reacted at
90 °C without closing the cap to concentrate the solvent. After
confirming that starting material was completely converted into
the target product by monitoring MALDI-TOFMS, the resi-
dues were collected in one microfuge tube by 500 μL of water. To
the solution was added 250 mM Hepes buffer (500 μL, pH 7.2,
containing 20 mM MnCl2), UDP-Gal (2.0 equiv, 12.2 mg), and
Gal-T I (20 mU), and the mixture was incubated at 37 °C. When
the enzymatic galactosylation of the chitotriose derivatives was
completed, the reaction was stopped by heat shock (90 °C,
5 min). The mixture was purified by Sephadex G-10 (eluted
with 0.1% aq AcOH) and concentrated. The syrup was dis-
solved in 1 mL of 0.1% aq AcOH to adjust the concentration
General Procedure for Monitoring Glycosyltransferases Reac-
tions by MALDI-TOFMS. The sample was dropped on the
polished steel sample plate and dried by blower. The laser was
irradiated to the edge of sample crystal because the WR-labeled
compounds were concentrated by edge of crystal. The quantitative
analysis was carried out by means of the ratio of mass intensity due
to H-labeled product and D-labeled internal standard as indicated
in Figure 4. The yield of the glycosylation in the presence of each
click product was calculated from the peak intensities correspond-
ing to product and internal standard as shown in eq 1.
yieldð%Þ ¼ P=In ꢀ 100
ð1Þ
Where P is peak intensity due to the product and In represents the
peak intensity due to the internal standard. The data processing and
calculation of inhibition constant were performed by Microsoft
Excel and Graphpad Prism.
Conditions for High Throughput Inhibition Assay of Click
Products. R2,3/R2,6-STs: To the mixture containing 200 μM
CMP-Neu5Ac, 200 μM click product (A1-A36, calculated from
the concentration of azidosugar nucleotide and alkyne), 1 mM
9, 12 mM sodium cacodylate buffer (pH 7.4), 0.15 mg/mL BSA,

5618 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Hosoguchi et al.
and 0.03% Triton CF-54 was added 0.5 mU R2,3- or R2,6-ST
(total volume was adjusted to 25 μL). The reaction mixture was
incubated at 37 °C for 30 min. After incubation, 100 μL of
CH3CN was added to stop the reaction, and the total volume
was adjusted to 250 μL by addition of 125 μL of water. To the
20 μL of DHB solution (10 mg/mL DHB in 30% CH3CN) was
added 2 μL of each reaction mixture and deuterium labeled
internal standard, and then 0.5 μL of mixed solution was
mounted on MALDI-TOFMS target plate.
R1,3/R1,6-FucTs: To the mixture of GDP-Fuc (200 μM), 9
(1 mM), 30 μM click product (A1-A36, calculated from the
concentration of azidosugar nucleotide and alkyne) in 50 mM
Hepes buffer (20 μL, pH 7.2) was added R1,3Fuc-T (1 mU). The
mixture was incubated at 37 °C for 20 min. In the case of R1,6-
Fuc-T, the reaction was carried out in 50 mM sodium cacodylate
buffer (20 μL, pH 7.5) in the presence of GDP-Fuc (50 μM), 11
(100 μM), click product (A1-A36, 50 μM), and R1,6-Fuc-T
(80 μU). All reactions were stopped by addition of CH3CN and
employed for MALDI-TOFMS assay by the same procedure
described above.
Kinetic Analysis. IC50 of compound 15 againstR2,3-ST: The
inhibitory effect of compound 15 on R2,3-ST was evaluated and
determined as IC50 value by using 11 different concentrations of
15 from 0.39 to 200 μM. The solution containing 200 μM CMP-
Neu5Ac, 1 mM acceptor substrate 9, 40 mM sodium cacody-
late buffer (pH 7.4), 0.5 mg/mL BSA, 0.1% Triton CF-54, and
R2,3-ST (0.5 mU) in a total volume 25 μL was incubated at 25 °C
for 15 min. After incubation, 100 μL of CH3CN was added to
stop the reaction, and 125 μL of water was added to the mixture
to be a total volume 250 μL. To the 20 μL of DHB solution
(10 mg/mL DHB in 30% CH3CN) was added 2 μL of each
reaction mixture and deuterium labeled internal standard,
and then 0.5 μL of mixed solution was mounted on MALDI-
TOFMS target plate.
Supporting Information Available: HPLC profiles of com-
pounds 11 and 12, standard curves for the quantitation of
products, characterization of click reactions, and kinetic analy-
sis data. This material is available free of charge via the Internet
at http://pubs.acs.org.
References
(1) Gabius, H.-J. The Sugar Code-Fundamentals of Glycosciences.
Wiley-Blackwell: Weinheim, 2009.
(2) (a) Zachara, N. E.; Hart, G. W. The emerging significance of O-
GlcNAc in cellular regulation. Chem. Rev. 2002, 102, 431–438. (b)
Ohtsubo, K.; Marth, J. D. Glycosylation in cellular mechanisms of health
and disease. Cell 2006, 126, 855–867. (c) Molinari, M. N-Glycan struc-
ture dictates extension of protein folding or onset of disposal. Nature
Chem. Biol. 2007, 3, 313–320. (d) Arnold, J. N.; Wormald, M. R.; Sim,
R. B.; Rudd, P. M.; Dwek, R. A. The impact of glycosylation on the
biological function and structure of human immunoglobulins. Annu.
Rev. Immunol. 2007, 25, 21–50. (e) Lau, K. S.; Partridge, E. A.;
Grigorian, A.; Silvescu, C. I.; Reinhold, V. N.; Demetriou, M.; Dennis,
J. W. Complex N-glycan number and degree of branching cooperate to
regulate cell proliferation and differentiation. Cell 2007, 129, 123–134.
(3) (a) Taniguchi, N.; Honke, K.; Fukuda, M. Handbook of Glycosyl-
transferase and Related Genes; Springer: Tokyo, 2002. (b) Breton, C.;
Snajdrova, L.; Jeanneau, C.; Koca, J.; Imberty, A. Structures and
mechanisms of glycosyltransferases. Glycobiology 2006, 16, 29R–
37R. (c) Pratt, M. R.; Hang, H. C.; Ten Hagen, K. G.; Rarick, J.;
Gerken, T. A.; Tabak, L. A.; Bertozzi, C. R. Deconvoluting the
functions of polypeptide N-R-acetylgalactosaminyltransferase family
members by glycopeptide substrate profiling. Chem. Biol. 2004, 11,
1009–1016.
(4) (a) Countinho, P. M.; Deleury, E.; Davies, G. J.; Henrissat, B. An
evolving hierarchical family classification for glycosyltransferases.
J. Mol. Biol. 2003, 328, 307–317. (b) Davies, G. J.; Henrissat, B.
Structures and mechanisms of glycosyl hydrolases. Structure 1995, 3,
853–859.
(5) (a) Hollingsworth, M. A.; Swanson, B. J. Mucins in cancer:
protection and control of the cell surface. Nature Rev. Cancer
2004, 4, 45–60. (b) Shriver, Z.; Raguram, S.; Sasisekharan, R. Glyco-
mics: a pathway to a class of new and improved therapeutics. Nature
Rev. Drug Discovery 2004, 3, 863–873. (c) Ludwig, J. A.; Weinstein,
J. N. Biomarkers in cancer staging, prognosis and treatment selection.
Nature Rev. Cancer 2005, 5, 845–856. (d) Dube, D. H.; Bertozzi, C. R.
Glycans in cancer and inflammation;potential for therapeutics and
diagnostics. Nature Rev. Drug Discovery 2005, 4, 477–488.
(6) (a) Schaub, C.; Muller, B.; Schmidt, R. R. New sialyltransferase
inhibitors based on CMP-quinic acid: development of a new
sialyltransferase assay. Glycoconjgate J. 1998, 15, 345–354. (b) Amann,
F.; Schaub, C.; Muller, B.; Schmidt, R. R. New potent sialyltransferase
inhibitors: synthesis of donor and of transition-state analogues of sialyl
donor CMP-Neu5Ac. Chem.;Eur. J. 1998, 4, 1106–1115. (c) Kajihara,
Y.; Kodama, H.; Wakabayashi, T.; Sato, K.-I.; Hashimoto, H. Character-
ization of inhibitory activities and binding mode of synthetic 60-modified
methyl N-acetyl-β-lactosaminide toward rat liver CMP-D-Neu5Ac:
D-galactoside-(2f6)-R-D-sialyltransferase. Carbohydr. Res. 1993, 247,
179–193. (d) Wu, C. Y.; Hus, C. C.; Chen, S. T.; Tsai, Y. C. Soyasaponin I, a
potent and specific sialyltransferase inhibitor. Biochem. Biophys. Res.
Commun. 2001, 284, 466–469. (e) Hinou, H.; Sun, X.-L.; Ito, Y.
Systematic syntheses and inhibitory activities of bisubstrate-type inhibitors
of sialyltransferases. J. Org. Chem. 2003, 68, 5602–5613. (f) Lee, L. V.;
Mitchell, M. L.; Huang, S.-J.; Fokin, V. V.; Sharpless, K. B.; Wong, C.-H. A
potent and highly selective inhibitor of human R1,3-fucosyltransferase via
click chemistry. J. Am. Chem. Soc. 2003, 125, 9588–9589. (g) Izumi, M.;
Wada, K.; Yuasa, H.; Hashimoto, H. Synthesis of bisubstrate and donor
analogues of sialyltransferase and their inhibitory activities. J. Org. Chem.
2005, 70, 8817–8824. (h) Lin, T. W.; Chang, W. W.; Chen, C. C.; Tsai, Y. C.
Stachybotrydial, a potent inhibitor of fucosyltransferase and sialyltransfer-
ase. Biochem. Biophys. Res. Commun. 2005, 331, 953–957. (i) Takaya,
K.; Nagahori, N.; Kurogochi, M.; Furuike, T.; Miura, N.; Monde, K.; Lee,
Y. C.; Nishimura, S.-I. Rational design, synthesis, and characterization of
novel inhibitors for human β1,4-galactosyltransferase. J. Med. Chem.
2005, 341, 6054–6065. (j) Chang, K. H.; Lee, L.; Chen, J.; Li, W. S.
Lithocholic acid analogues, new and potent alpha-2,3-sialyltransferase
inhibitors. Chem. Commun. 2006, 629–631.
K
m of compound 15 to R2,6-ST: The kinetic parameter of Km
value of compound 15 for R2,6-ST was determined at six
different concentrations in a range of 12.5-200 μM. The reac-
tion solution containing compound 7, 1 mM 9, 40 mM sodium
cacodylate buffer (pH 7.4), 0.5 mg/mL BSA, 0.1% Triton CF-
54, and R2,6-ST (0.5 mU) in a total volume 25 μL was incubated
at 37 °C for 20 min. After incubation, 100 μL of CH3CN was
added to stop the reaction, and the mixture was added by 125 μL
of water to be a total volume 250 μL. To the 20 μL of DHB
solution (10 mg/mL DHB in 30% CH3CN) was added 2 μL of
each reaction mixture and deuterium labeled internal standard,
and then 0.5 μL of mixed solution was mounted on MALDI-
TOFMS target plate.
Ki of compound 16 against R1,3-FucT: The assays were
carried out in 50 mM Hepes buffer (20 μL, pH 7.2) in the pre-
sence of GDP-Fuc (50-125 μM), 11 (1 mM), 16 (0.25-2 μM),
and R1,3-FucT (0.1 mU). The mixtures were incubated at 37 °C
for 60 min, and the inhibitory effect by compound 16 was
evaluated by general MALDI-TOFMS protocol described
above (Supporting Information).
Ki of compound 17 against R1,6-FucT: The assays were
carried out in 50 mM sodium cacodylate buffer (20 μL, pH
7.5) in the presence of GDP-Fuc (5-50 μM), 9 (100 μM), 17
(0.5-50 μM), and R1,6-FucT (80 μU). The mixtures were
incubated at 37 °C for 60 min, and the inhibitory effect by
compound 17 was evaluated by general MALDI-TOFMS pro-
tocol described above (Supporting Information).
Acknowledgment. This work was partly supported by a
program grant “The matching program for innovation in
future drug discovery and medical care” from the Ministry of
Education, Culture, Science, and Technology, Japan. We also
thank M. Kikuchi, S. Oka, and T. Hirose at the Center for
Instrumental Analysis, Hokkaido University, for ESI-MS
measurement.
(7) (a) Taylor, G. Sialidases: structures, biological significance and
therapeutic potential. Curr. Opin. Struct. Biol. 1996, 6, 830–837.
(b) Taylor, N. R.; Itzstein, M. Molecular modeling studies on ligand
binding to sialidase from influenza virus and the mechanism of
catalysis. J. Med. Chem. 1994, 37, 616–624.
(8) (a) Chiu, C. P.; Watts, A. G.; Lairson, L. L.; Gilbert, M.; Lim, D.;
Wakarchuk, W. W.; Withers, S. G.; Strynadka, N. C. Structural

Article
analysis of the sialyltransferase CstII from Campylobacter jejuni in
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5619
surfaces using a fucosyltransferase. J. Biol. Chem. 1992, 267,
22356–22361. (b) Washiya, K.; Furuike, T.; Nakajima, F.; Lee, Y. C.;
Nishimura, S.-I. Design of fluorogenic substrates for continuous assay
of sialyltransferase by resonance energy transfer. Anal. Biochem. 2000,
283, 39–48. (c) Sawa, M.; Hsu, T.-L.; Itoh, T.; Sugiyama, M.; Hanson,
S. R.; Vogt, P. K; Wong, C.-H. Glycoproteomic probe for fluorescent
imaging of fucosylated glycans in vivo. Proc. Natl. Acad. Sci. U.S.A.
2006, 103, 12371–12376.
complex with a substrate analog. Nature Struct. Mol. Biol. 2004, 11,
163–170. (b) Fritz, T. A.; Hurley, J. H.; Trinh, L.-B.; Shiloach, J.; Tabak,
L. A. The beginnings of mucin biosynthesis: the crystal structure of UDP-
GalNAc:polypeptide R-N-acetylgalactosaminyltransferase-T1. Proc.
Natl. Acad. Sci. U.S.A. 2004, 101, 15307–15312. (c) Qasba, P. K.;
Ramakrishnan, B.; Boeggeman, E. Substrate-induced conformational
changes in glycosyltransferases. Trends Biochem. Sci. 2005, 30, 53–62.
(d) Fritz, T. A.; Raman, J.; Tabak, T. A. Dynamic association between the
catalytic and lectin domains of human UDP-GalNAc: polypeptide R-N-
acetylgalactosaminyltransferase-2. J. Biol. Chem. 2006, 281, 8613–8619.
(e) Chiu, C. P.; Lairson, L. L.; Gilbert, M.; Wakarchuk, W. W.; Withers,
S. G.; Strynadka, N. C. Structural analysis of the alpha-2,3-sialyltransferase
Cst-I from Campylobacter jejuni in apo and substrate-analogue bound
forms. Biochemistry 2007, 46, 7196–7204. (f) Ni, L.; Chokhawala, H. A.;
Cao, H.; Henning, R.; Ng, L.; Huang, S.; Yu, H.; Chen, X.; Fisher, A. J.
Crystal structures of Pasteurella multocida sialyltransferase complexes
with acceptor and donor analogues reveal substrate binding sites and
catalytic mechanism. Biochemistry 2007, 46, 6288–6298. (g) Rao, F. V.;
Rich, J. R.; Rakic, B.; Buddai, S.; Schwartz, M. F.; Johnson, K.; Bowe, C.;
Wakarchuk, W. W.; DeFrees, S.; Withers, S. G.; Strynadka, N. C. J.
Structural insight into mammalian sialyltransferases. Nature Struct. Mol.
Biol. 2009, 16, 1186–1188. (h) Pesnot, T.; Jørgensen, R.; Palcic, M. M.;
Wagner, G. K. Structure and mechanistic basis for a new mode of
glycosyltransferases inhibition. Nature Chem. Biol. 2010, 6, 321–323.
(9) (a) Schindler, T.; Bornmann, W.; Pellicena, P.; Miller, W. T.;
Clarkson, B.; Kuriyan, J. Structural mechanism for STI-571
inhibition of Abelson tyrosine kinase. Science 2000, 289, 1938–
1942. (b) Liu, Y.; Gray, N. S. Rational design of inhibitors that bind to
inactive kinase conformations. Nature Chem. Biol. 2006, 2, 358–364.
(16) (a) Brossmer, R.; Rose, U.; Kasper, D.; Smith, T. L.; Grasmuk, H.;
Unger, F. M. Enzymic synthesis of 5-acetamido-9-azido-3,5,9-
trideoxy-D-glycero-D-galacto-2-nonulosonic acid, a 9-azido-9-
deoxy derivative of N-acetylneuraminic acid. Biochem. Biophys. Res.
Commun. 1980, 96, 1282–1289. (b) Liu, J. L.-C.; Shen, G.-J.; Ichikawa, Y.;
Rutan, J. F.; Zapata, G.; Willie F. Vann, W. F. V.; Wong, C.-H. Over-
production of CMP-sialic acid synthetase. J. Am. Chem. Soc. 1992, 114,
3901–3910.
(17) Yu, H.; Yu, H.; Karpel, R.; Chen, X. Chemoenzymatic synthesis of
CMP-sialic acid derivatives by one-pot two-enzyme system: com-
parison of substrate flexibility of three microbial CMP-sialic acid
synthetases. Bioorg. Med. Chem. 2004, 12, 6427–6435.
(18) Luchansky, S. J.; Goon, S.; Bertozzi, C. R. Expanding the diversity
of unnatural cell-surface sialic acids. ChemBioChem 2004, 5,
371–374.
(19) Maeda, T.; Nishimura, S.-I. FRET-based direct and continuous
monitoring of human fucosyltransferases activity: efficient synth-
esis of versatile GDP-L-fucose derivatives from abundant D-galac-
tose. Chem.;Eur. J. 2007, 14, 478–487.
(20) (a) Nishimura, S.-I.; Niikura, K.; Kurogochi, M.; Matsushita, T.;
Fumoto, M.; Hinou, H.; Kamitani, R.; Nakagawa, H.; Deguchi,
K.; Miura, N.; Monde, K.; Kondo, H. High-throughput protein
glycomics: combined use of chemoselective glycoblotting and
MALDI-TOF/TOF mass spectrometry. Angew. Chem., Int. Ed.
2004, 44, 91–96. (b) Miura, Y.; Hato, M.; Shinohara, Y.; Kuramoto, H.;
Furukawa, J.; Kurogochi, M.; Shimaoka, H.; Tada, M.; Nakanishi, K.;
Ozaki, M.; Todo, S.; Nishimura, S.-I. BlotGlycoABC, an integrated
glycoblotting technique for rapid and large scale clinical glycomics.
Mol. Cell. Proteomics 2008, 7, 370–377.
€
(10) Klebl, B. M.; Muller, G. Second-generation kinase inhibitors.
Expert Opin. Ther. Targets 2005, 9, 975–993.
(11) (a) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B.
A stepwise Huisgen cycloaddition process: copper(I)-catalyzed
regioselective “ligation” of azides and terminal alkynes. Angew.
Chem., Int. Ed. 2002, 41, 2596–2599. (b) Tornøe, C. W.; Christensen,
C.; Meldal, M. Peptidotriazoles on solid phase: [1,2,3]-triazoles by
regiospecific copper(I)-catalyzed 1,3-dipolar cycloadditions of terminal
alkynes to azides. J. Org. Chem. 2002, 67, 3057–3064. (c) Kolb, H. C.;
Sharpless, K. B. The growing impact of click chemistry on drug
discovery. Drug Discovery Today 2003, 8, 1128–1137. (d) Chan,
T. R.; Hilgraf, R.; Sharpless, K. B.; Fokin, V. V. Polytriazoles as
Copper(I)-Stabilizing Ligands in Catalysis. Org. Lett. 2004, 6, 2853–
2855. (e) Meldal, M.; Tornøe, C. W. Cu-Catalyzed azide-alkyne
cycloaddition. Chem. Rev. 2008, 108, 2952–3015.
(12) Kondo, N.; Nishimura, S.-I. MALDI-TOF Mass spectrometry-
based versatile method for the characterization of protein kinases.
Chem.;Eur. J. 2009, 15, 1413–1421.
(13) Amano, M.; Yamaguchi, M.; Takegawa, Y.; Yamashita, T.;
Terashima, M.; Furukawa, J.-I; Miura, Y.; Shinohara, Y.; Iwasaki,
N.; Minami, A.; Nishimura, S.-I. Threshold in stage-specific
embryonic glycotypes uncovered by a full portrait of dynamic N-
glycan expression during cell differentiation. Mol. Cell. Proteomics
2010, 9, 523–537.
(21) Seko, A.; Koketsu, M.; Nishizono, M.; Enoki, Y.; Ibrahim, H. R.;
Juneja, L. R.; Kim, M.; Yamamoto, T. Occurrence of a sialylgly-
copeptide and free sialylglycans in hen’s egg yolk. Biochim. Bio-
phys. Acta 1997, 1335, 23–32.
€
(22) Jung, K.-H.; Schworer, R.; Schmidt, R. R. Sialyltransferase in-
hibitors. Trends Glycosci. Glycotechnol. 2003, 15, 275–289.
(23) (a) Kleineidam, R. G.; Schmelter, T.; Schwarz, R. T.; Schauer, R.
Studies on the inhibition of sialyl- and galactosyltransferases.
Glycoconjugate J. 1997, 14, 57–66. (b) Whalen, L. J.; McEvoy,
K. A.; Halcomb, R. L. Synthesis and evaluation of phosphoramidate
amino acid-based inhibitors of sialyltransferases. Bioorg. Med. Chem.
Lett. 2003, 13, 301–304.
(24) (a) Prieels, J. P.; Monnom, D.; Dolmans, M. Co-purification of the
Lewis blood group N-acetylglucosaminide R1,4-fucosyltransferase
and N-acetylglucosaminide R1,3-fucosyltransferase from human
milk. J. Biol. Chem. 1981, 256, 10456–10463. (b) Wong, C.-H.;
Dumas, D. P.; Ichikawa, Y.; Koseki, K.; Danishefsky, S. J.; Weston,
B. W.; Lowe, J. B. Specificity, inhibition, and synthetic utility of a
recombinant human a 1,3-fucosyltransferase. J. Am. Chem. Soc. 1992,
114, 7321–7322. (c) Ihara, H.; Ikeda, Y.; Taniguchi, N. Reaction
mechanism and substrate specificity for nucleotide sugar of mammalian
R1,6-fucosyltransferase: a large-scale preparation and characterization
of recombinant human FUT8. Glycobiology 2006, 16, 333–342.
(25) (a) Matsuoka, K.; Oka, H.; Terunuma, D.; Kuzuhara, H. Synthesis
and reactivity of a 5-azido analogue of neuramic acid. Carbohydr.
Lett. 2001, 4, 123–130. (b) Tanaka, H.; Nishiura, Y.; Adachi, M.;
Takahashi, T. Synthetic study of R(2,8) oligosialoside using N-troc
sialyl N-phenyltrifluoroimidate. Heterocycles 2006, 67, 107–112.
(14) (a) Shinohara, Y.; Furukawa, J-i.; Niikura, K.; Miura, N.; Nishi-
mura, S. -I. Direct N-glycan profiling in the presence of tryptic
peptides on MALDI-TOF by controlled ion enhancement and
suppression upon glycan-selective derivatization. Anal. Chem.
2004, 76, 6989–6997. (b) Furukawa, J.; Shinohara, Y.; Kuramoto, H.;
Miura, Y.; Shimaoka, H.; Kurogochi, M.; Nakano, M.; Nishimura, S.-I.
Comprehensive approach to structural and functional glycomics based
on chemoselective glycoblotting and sequential tag conversion. Anal.
Chem. 2008, 80, 1094–1101.
(15) (a) Srivastava, G.; Kaur, K. J.; Hindsgaul, O.; Palcic, M. M.
Enzymatic transfer of a preassembled trisaccharide antigen to cell