Design of glycosyltransferase inhibitors targeting human O-GlcNAc transferase (OGT)

Article abstract of DOI:10.1039/c4md00063c

Inhibition of glycosyltransferases requires the design of neutral inhibitors to allow cell permeation in contrast to their natural dianionic substrates. O-GlcNAc transferase (OGT) is a key enzyme involved in dynamic glycosylation of cytosolic and nuclear proteins in competition with phosphorylation. Designing OGT inhibitors is of prime interest for the better understanding of its biological implications. Introduction of a pyridine moiety as a pyrophosphate surrogate was evaluated, which provided moderate in vitro inhibition of OGT. Docking studies highlighted some key features for the binding of the designed inhibitors to the catalytic site of OGT where the carbohydrate moiety did not occupy its natural position but rather turned away and pointed to the solvent outside the catalytic pocket. Further investigation with cellular assays did not provide inhibition of OGT. This lack of OGT inhibition was rationalized with a permeation assay which revealed the sequestration of the inhibitors at the membrane. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4md00063c

MedChemComm
View Article Online
View Journal | View Issue
CONCISE ARTICLE
Design of glycosyltransferase inhibitors targeting
human O-GlcNAc transferase (OGT)†
Cite this: Med. Chem. Commun., 2014,
5, 1172
Shuai Wang,a David L. Shen,bc Dominique Lafont,a Anne-Sophie Vercoutter-Edouart,d
Marlene Mortuaire,d Yun Shi,c Ofelia Maniti,a Agnes Girard-Egrot,a Tony Lefebvre,d
`
`
a
B. Mario Pinto,c David Vocadlobc and Sebastien Vidal
*
´
Inhibition of glycosyltransferases requires the design of neutral inhibitors to allow cell permeation in
contrast to their natural dianionic substrates. O-GlcNAc transferase (OGT) is a key enzyme involved in
dynamic glycosylation of cytosolic and nuclear proteins in competition with phosphorylation. Designing
OGT inhibitors is of prime interest for the better understanding of its biological implications. Introduction
of a pyridine moiety as a pyrophosphate surrogate was evaluated, which provided moderate in vitro
inhibition of OGT. Docking studies highlighted some key features for the binding of the designed
inhibitors to the catalytic site of OGT where the carbohydrate moiety did not occupy its natural position
but rather turned away and pointed to the solvent outside the catalytic pocket. Further investigation with
cellular assays did not provide inhibition of OGT. This lack of OGT inhibition was rationalized with a
permeation assay which revealed the sequestration of the inhibitors at the membrane.
Received 17th February 2014
Accepted 20th March 2014
DOI: 10.1039/c4md00063c
www.rsc.org/medchemcomm
b-N-acetylglucosaminyltransferase
or
O-GlcNAc
trans-
Introduction
ferase).15–18 The glycoside hydrolase O-GlcNAcase (OGA, exo-b-
N-acetylglucosaminidase) catalyzes removal of the GlcNAc
moiety from proteins.19 OGT (EC. 2.4.1.255) is assigned to
the GT41 family of the glycosyltransferase superfamily within
the CAZY (Carbohydrate-Active enZYme) classication
system.20 GT41 family members adopt a GTB fold, which
comprises two Rossman-like domains separated by a cle.
These GT41 enzymes are inverting glycosyltransferases,
meaning that the overall glycosyl transfer reaction proceeds
with inversion of stereochemistry at the anomeric center. OGT
possesses a series of tetratricopeptide repeats (TPRs), which
are generally involved in protein–protein interactions.21 In
mammals, three OGT isoforms have been described and they
differ primarily in the number of their TPRs.18 Nuclear and
cytoplasmic OGT (ncOGT), the best-characterized and the
longest isoform, contains 13.5 TPRs, sOGT only contains 2.5
TPRs and mOGT, which is generated by alternative splicing,
contains 8.5 TPRs and a MTS (Mitochondrial Targeting
Sequence).
O-GlcNAcylation (O-linked b-N-acetylglucosaminylation) is an
abundant post-translational modication found predominantly
on cytosolic and nuclear proteins.1 O-GlcNAcylation is
conserved among all metazoans studied to date,2–8 and it has
been proposed to regulate fundamental cellular processes
including cell signaling,9 cell cycle,10,11 and circadian clock
function.12
In contrast to the more structurally complex glycosylation
occurring within the secretory pathway, O-GlcNAcylation is
dynamic in that it can be installed and removed from proteins
several times during the lifetime of the polypeptide back-
bone.13,14 Transfer of the N-acetylglucosamine (GlcNAc) residue
from the nucleotide–sugar UDP-GlcNAc onto target proteins is
catalyzed by the glycosyltransferase known as O-GlcNAc trans-
ferase (OGT, uridine diphospho-N-acetylglucosamine:polypeptide
a
´
´
Institut de Chimie et Biochimie Moleculaires et Supramoleculaires, CO2-Glyco and
´
´
GEMBAS, UMR 5246, CNRS, Universite Claude Bernard Lyon 1, Universite de Lyon,
43 Boulevard du 11 Novembre 1918, F-6922 Villeurbanne, France. E-mail: sebastien.
vidal@univ-lyon1.fr; Fax: +33 472 448 109; Tel: +33 472 448 349
The regulation of OGT and the mechanism by which it
recognizes modication sites remains an open question,
though the recently solved structure of a large fragment of OGT
is greatly facilitating progress in this area.22 The emerging roles
of OGT in fundamental cellular processes as well as diseases
including for example cancer and neurodegeneration make
tools that can perturb the OGT function of considerable
interest. The generation of potent and selective inhibitors of
OGT is consequently a topic of current interest. Relatively few
compounds, however, have been identied that can inhibit this
bDepartment of Chemistry, Simon Fraser University, Burnaby, British Columbia,
Canada V5A 1S6
cDepartment of Molecular Biology and Biochemistry, Simon Fraser University,
Burnaby, British Columbia, Canada V5A 1S6
d
´
Unite de Glycobiologie Structurale et Fonctionnelle, UMR CNRS/USTL no 8576, IFR
´
¨
147, Universite des Sciences et Technologies de Lille, Avenue Mendeleıev, 59655
Villeneuve d'Ascq, France
† Electronic supplementary information (ESI) available: Experimental procedures
for the synthesis of compounds 2–14 and NMR spectra, cellular OGT inhibition
assays, and permeation assays. See DOI: 10.1039/c4md00063c
1172 | Med. Chem. Commun., 2014, 5, 1172–1178
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
enzyme23,24 making the design of OGT inhibitors a continuing
challenge.
The rational design of GT inhibitors generally involves the
synthesis of acceptor, bi-substrate, and donor analogues. Donor
analogues of UDP-GlcNAc can bind tightly, but the dianionic
pyrophosphate bridge limits the permeability of such
compounds for use in cells. Several donor analogues have been
assessed as inhibitors including UDP-C-GlcNAc, UDP-S-GlcNAc,
and UDP-5S-GlcNAc (Fig. 1). These compounds, however,
cannot permeate into cells making the design of donor
analogues, in which the pyrophosphate bridge is replaced, of
considerable interest. Alloxan is a cell permeable inhibitor of
OGT but suffers from promiscuous inhibition of multiple
targets, which leads to cellular toxicity. A benzoxazolinone
derivative was identied using high-throughput screening24 and
was later shown to be an irreversible OGT inhibitor.25
The present work focuses on the design of OGT inhibitors
following a glycomimetic approach26–28 as analogues of the
natural UDP-GlcNAc substrate. The syntheses will use common
azido-functionalized precursors and conjugations will be ach-
ieved through either azide–alkyne cycloaddition or Staudinger–
Vilarrasa amide bond formation (Fig. 2). The amide or triazole
functionalities were chosen as potential hydrogen bond donors
or acceptors in view of additional and benecial interactions
with the enzyme's active site. The resulting N-glycosidic link-
ages were also selected due to their analogy with natural N-
linked oligosaccharides and also better stability in vivo as
recently reported for N-glycosyl triazoles.29,30
Fig. 2 Structures of three UDP-GlcNAc analogues and an overview of
the retrosynthetic strategies to access these compounds.
Results and discussion
Synthesis of UDP-GlcNAc analogues
Starting from 6-bromopicolinic acid 1, two pathways using the
same synthesis procedures afforded the UDP-GlcNAc
analogues 8 and 11 with variations at the positions of the
Scheme 1 Synthesis of two triazole/amide containing UDP-GlcNAc
analogues 8 and 11. Reagents and conditions: (a) PMe3, DIC, HOBt, and
Fig.
1 Structures of UDP-GlcNAc and (left) known non-neutral THF. (b) Me3SiC^CH, Pd(PPh3)4, CuI, iPr2NH, and PhMe. (c) nBu4NF
inhibitors of OGT based on UDP-GlcNAc analogues and (right) neutral and MeOH. (d) CuSO4, sodium ascorbate, and tBuOH–H2O and (e)
OGT inhibitors identified to date.
NaOMe and MeOH.
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun., 2014, 5, 1172–1178 | 1173

View Article Online
MedChemComm
Concise Article
amide and triazole conjugations (Scheme 1). The Staudinger–
Vilarrasa conjugation31–33 of 6-bromo-picolinic acid 1 with
glycosyl azide 2 afforded the desired N-acylated glycopyranosyl
amine 3 in 95% yield. Sonogashira coupling of this aryl
bromide with trimethylsilylacetylene afforded intermediate 4.
Purication of the silyl-protected intermediate 4 was trouble-
some and this held for the other Sonogashira reactions per-
formed in this series. As a result, cleavage of the silyl groups
was carried out directly to provide alkyne derivative 5. This
two-step process was used in all cases, providing the desired
alkynes in good yields. CuAAC conjugation with 20,30-di-O-
acetyl-50-azido-50-deoxy-uridine 634 afforded the acetylated
derivative 7, which was converted to the UDP-GlcNAc analogue
´
8 under Zemplen conditions. Similarly, 6-bromopicolinic acid
1 was converted to alkyne 9,35 which was used in the CuAAC
coupling with glycosyl azide 2 to provide the acetylated
precursor 10. Final deacetylation gave the second UDP-GlcNAc
analogue 11.
Fig. 3 Structures of the six analogues (8, 11, 14, 15, 16 and 17) and two
UDP analogues (18 and 19) evaluated as OGT inhibitors and of a UDP-
Gal analogue (20) used as a positive control in permeation assays.
The UDP-GlcNAc analogues incorporating two triazole
moieties were prepared from alkyne 12 (Scheme 2).35 CuAAC
conjugation36 with glycosyl azide 2 led to acetylated precursor
Enzymatic studies (IC50 and Ki measurements) with human
OGT in vitro
´
13, which was converted under Zemplen conditions to the third
UDP-GlcNAc analogue 14.
These eight synthesized NDP-sugar analogues were evaluated as
inhibitors of human OGT in vitro (Table 1). These compounds
were rst assayed at 50 mM concentration and the inhibition
was measured at this concentration. This 50 mM concentration
was selected as a rst pass since it facilitates the detection of
genuine inhibitors rather than compounds having indirect
effects on the OGT activity. This initial assay revealed that most
compounds showed little inhibition at 50 mM. Nevertheless,
three analogues (15, 17 and 19) showed greater than 10%
inhibition of hOGT suggesting that these could be legitimate
OGT inhibitors and therefore these compounds were selected
for further study. Surprisingly, the most promising compounds
are the UDP-Glc analogues 15 and 17 while the UDP-GlcNAc
analogues 8, 11 and 14 were quite poor inhibitors. The third
compound of interest is the UDP analogue 19. Measurement of
the Ki value for the best inhibitor (15) by Dixon analysis was
carried out (see ESI†), providing a value of 422 mM, which
though moderate in potency represents a reasonable level of
inhibition for such a derivative. The Dixon analysis performed,
however, does not allow us to distinguish whether the inhibitor
is competitive or not.
Using these approaches, three UDP-GlcNAc analogues 8,
11, and 14 were synthesized using a minimum number of
synthesis steps in good overall yields (Fig. 3). A series of
UDP-Glc analogues 15–17 were also prepared35 using the same
synthesis procedures (Fig. 3). These compounds incorporated
the same amide and triazole moieties. The rationale for
preparing these was to use them as negative controls in the
inhibition assays. Two UDP analogues 18 and 19 were also
synthesized and selected as potential OGT inhibitors since
UDP is a known inhibitor of glycosyltransferases including
OGT. Finally, the UDP-Gal analogue 20 (Fig. 3) was synthe-
sized in a parallel study on galactosyltransferase inhibition.35
This analogue was used as a positive control in the perme-
ation assays in comparison to the inhibitors tested in the
present study.
Table 1 Enzymatic inhibition of hOGT
% Inhibition
Analogue
Inhibitor
at 50 mM
IC50 (mM)
Ki (mM)
UDP-GlcNAc
8
3.0 ꢀ 0.2
9.7 ꢀ 0.7
6.7 ꢀ 0.4
18 ꢀ 1
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
11
14
15
16
17
18
19
UDP-Glc
UDP
340 ꢀ 20
420 ꢀ 10
4.2 ꢀ 0.2
13.3 ꢀ 0.9
7.6 ꢀ 0.1
15.5 ꢀ 0.9
n.d.
n.d.
1600 ꢀ 300
n.d.
710 ꢀ 20
n.d.
n.d.
n.d.
Scheme
2 Synthesis of a bis-triazole containing UDP-GlcNAc
analogue 14. Reagents and conditions: (a) CuI, iPr2NEt, and DMF and
(b) NaOMe and MeOH.
1174 | Med. Chem. Commun., 2014, 5, 1172–1178
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
Fig. 4 Overview of the best docking poses in comparison with the
natural substrate UDP-GlcNAc (grey: OGT, grey sticks: UDP-GlcNAc;
cyan sticks: 8; yellow sticks: 15; red sticks: 17; blue sticks: 19). Different
regions of UDP-GlcNAc are labeled in black.
Fig. 6 Inhibitor binding within the Glc/GlcNAc portion (grey sticks:
UDP-GlcNAc; cyan sticks: 8; yellow sticks: 15; red sticks: 17; blue
sticks: 19; all other lines represent OGT residues). Dashed lines indicate
possible hydrogen bonds from different inhibitors. The acetamido
group of UDP-GlcNAc and 8 is also labeled in grey and cyan,
respectively.
Modeling inhibitors within the active site of OGT
To gain insight into the binding of these compounds to OGT,
docking of several of these inhibitors including 8, 15, 17, and 19
was carried out using the known structure of OGT in complex
with UDP-GlcNAc (PDB 4GZ5) as the template. The best ranking
pose was selected from the docking of each OGT inhibitor
(Fig. 4). The uridine portion of all four inhibitors showed near
perfect overlap, with almost identical binding modes to that of
UDP-GlcNAc. This also agrees with X-ray structures showing
relatively static binding modes of the uridine portion in
different ligands.37
The triazole linkers in 15, 17, and 19 share the same orien-
tation and location, with N2 accepting one hydrogen bond from
the hydroxyl group of Thr922 and N3 accepting another from
the amide backbone of Thr921 (Fig. 5). These polar interactions
from the triazole resemble those from the b-phosphate of UDP-
GlcNAc, and the absence of such hydrogen bonds in 8 could
contribute to its weak inhibitory activity.
However, the binding mode of the natural substrate was not
mimicked with respect to the GlcNAc residue (Fig. 6). The bro-
mopyridine moiety in 19 partially overlapped with the GlcNAc
ring of UDP-GlcNAc, with its ring nitrogen accepting a hydrogen
bond from the Thr921 hydroxyl group. Yet, an orientation
perpendicular to that of GlcNAc was adopted by the pyridine ring
of 15 and 17; 8 did not contact this space at all. In the glucose–
GlcNAc end of the molecules, both 15 and 17 could form
hydrogen bonds with His498, His499, and Lys634, whereas 8
could be involved in a hydrogen bonding interaction with His499
(Fig. 6). Of note, the acetamido group of 8 showed no direct
interaction with the enzyme, and its presence may also have
forced this molecule to adopt a different and less energetically
favorable conformation compared to its glucose counterpart 15.
In summary, compound 19 exhibited the best substrate
mimicry in terms of binding geometry, while the other three
inhibitors extended too far past the sugar-binding pocket and
Fig. 5 Detailed comparison of substrate and inhibitor binding (grey placed the pyranose moiety in a completely different region of
sticks: UDP-GlcNAc; cyan sticks: 8; yellow sticks: 15; red sticks: 17;
blue sticks: 19; all other lines represent OGT residues). (a) Hydrogen
bonds (blue dashed lines) were formed with N2 and N3 of the triazole;
the polar interaction between the b-phosphate of UDP-GlcNAc and
the enzyme active site. These modeling studies therefore reveal
why the change of the sugar residue from that of the natural
substrate GlcNAc to Glc does not have a signicant effect on
binding even though OGT has an established preference for
GlcNAc over Glc.22
Thr921 is also shown (grey dashed line) for comparison. (b) The pyri-
dine nitrogen of 19 accepted a hydrogen bond.
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun., 2014, 5, 1172–1178 | 1175

View Article Online
MedChemComm
Concise Article
evaluate the potential applications of such inhibitors but no
activity could be observed on different cell lines even at high
concentrations (up to 500 mM). A permeation assay designed
using liposomes showed that these inhibitors do not permeate
through the liposome membrane, allowing us to conclude that
the lack of inhibition observed in cellular assays was probably
due to the poor permeability of the compounds through the cell
membrane. The level of moderate inhibition observed, coupled
with the docking studies, suggests that the linkers identied in
this study serve as a replacement for the pyrophosphate unit but
that they fail to position the sugar residue in the appropriate
position. Furthermore, these compounds lack the needed
permeability, suggesting that further changes are needed to
improve both the orientation of the sugar mimicking element
as well as to enhance the permeability. Further renement in
the rationale design of glycosyltransferase inhibitors is still
necessary to obtain cell permeable inhibitors with potential
applications in cell assays or even in vivo.
Cellular OGT inhibition and permeation assays
Given that derivative 15 shows some inhibition of OGT, the next
logical step was to evaluate these derivatives under cell culture
conditions to see whether they could block the OGT action in a
biological setting. Compounds 8, 11, and 14–18 were also
selected since the permeability of these compounds into cells
and their cellular efficacy are difficult to estimate solely on the
basis of in vitro inhibition data. The inhibition of O-GlcNAcy-
lation in a cell based assay was not observed for any of the
designed inhibitors (see ESI†). This lack of inhibition at the cell
level could be the result of either (1) poor permeation of the
inhibitors through the cell membrane or (2) decomposition or
elimination of the inhibitors by the cell machinery. To clarify
the question of permeation of these inhibitors, additional
studies were performed. We have designed an assay in which
the cell membrane is mimicked by a lipid bilayer of liposomes.
The penetration of organic molecules into the liposomes was
detected through their intrinsic UV properties. Two candidates
were tested here: UDP-Gal analogue 20, synthesized in a parallel
study on galactosyltransferases35 and used in this work as a
model compound for permeation assays, and UDP-Glc analogue
15 designed herein (see ESI†). The permeation of the molecules
could not be detected and the conclusion of this evaluation is
that these compounds are either adsorbed at the surface of the
liposome or could not cross the lipidic bilayer.
Experimental section
General methods
All reagents were obtained from commercial sources and used
without further purication. Dichloromethane was distilled
over CaH2. Methanol was distilled over Mg/I2. All reactions were
performed under an argon atmosphere unless otherwise stated.
Thin-layer chromatography (TLC) was carried out on aluminum
sheets coated with silica gel 60 F254 (Merck). TLC plates were
inspected using UV light (l ¼ 254 nm) and developed by treat-
ment with a mixture of 10% H2SO4 in EtOH–H2O (1 : 1 v/v)
followed by heating. Silica gel column chromatography was
performed with silica gel Si 60 (40–63 mm). NMR spectra were
recorded at 293 K, unless otherwise stated, using Bruker 300,
400 or 500 MHz spectrometers. Chemical shis are referenced
relative to deuterated solvent residual peaks. The following
abbreviations are used to explain the observed multiplicities: s,
singlet; d, doublet; t, triplet; q, quadruplet; m, multiplet and bs,
broad singlet. Complete signal assignments were based on 1D
and 2D NMR (COSY, HSQC and HMBC correlations). High
resolution (HR-ESI-QToF) mass spectra were recorded using a
Bruker MicroToF-Q II XL spectrometer. Optical rotations were
measured using a Perkin Elmer polarimeter and values are
Conclusions
The synthesis of NDP-sugar analogues incorporating a pyridine
moiety as a replacement for the pyrophosphate moiety was
achieved through a combination of Staudinger–Vilarrasa and
azide–alkyne cycloaddition. The compounds obtained were
then evaluated as substrate-based inhibitors of OGT mimicking
UDP-GlcNAc. In vitro assays on human OGT provided only poor
to moderate inhibition with the best compound displaying an
inhibition constant (Ki) value of 420 mM. Docking studies were
then performed to better rationalize the inhibition pattern
observed for these inhibitors. The inhibition studies do not
enable the inhibitors to be assigned as competitive, however,
the modelling studies support binding at the active site.
Specically, the uracil moiety of these molecules was always
found to bind in an almost identical manner as seen for the
natural UDP-GlcNAc substrate and is likely responsible for most
of the binding to the enzyme. The pyrophosphate group of the
natural substrate forms H-bonds with the enzyme's active site
given in 10ꢁ1 deg cm2 gꢁ1
.
General protocol for Staudinger–Vilarrasa conjugation
and the triazole moieties of these analogues could reproduce The acid (0.466 mmol, 1 eq.) and HOBt (0.839 mmol, 1.8 eq.)
such interactions. Nevertheless, a major change was observed were co-evaporated with toluene (3 ꢂ 5 mL) and THF (3 ꢂ 5
for the position of the carbohydrate moiety (either GlcNAc or mL). The mixture was dried under vacuum for 1 h. The mixture
Glc) which pointed out into the solvent and therefore was was dissolved in dry THF (4 mL) under argon and cooled to 0 ꢃC.
positioned outside the binding pocket. This observation DIC (0.839 mmol, 1.8 eq.) was added dropwise at 0 ꢃC. Aer
explains why the glucose-based analogue was also an inhibitor addition, the ice-bath was removed and the reaction was stirred
like the expected GlcNAc analogue. This observation was very at r.t. for 30 min. Meanwhile, the azide (0.699 mmol, 1.5 eq.)
similar to our previous report of galactosyltransferases in which was dissolved in dry THF (4 mL) under argon and cooled to 0 ꢃC.
the carbohydrate moiety was also located outside of the enzy- PMe3 (0.932 mmol, 2 eq.) was added and the reaction was stir-
me's catalytic site based on crystallographic studies of enzyme– red at 0 ꢃC. Aer 30 min, the solution was transferred into a
inhibitor complexes.35 Cellular assays were then performed to ask containing the acid–HOBt solution at 0 C. The ask was
ꢃ
1176 | Med. Chem. Commun., 2014, 5, 1172–1178
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
washed with THF (4 mL) and the solution was transferred. The mM was used). The reaction was initiated by the addition of the
resulting reaction mixture was stirred at 0 ꢃC for 1 h, then enzyme with a pipette and incubated at 37 ꢃC for 1 h (a time for
allowed to reach r.t. and stirred for an additional 16 h. The which linear rates are obtained under these assay conditions).
reaction mixture was diluted with water (60 mL) and extracted The reactions were placed on ice aer 1 h of incubation and
with EtOAc (4 ꢂ 60 mL). The combined organic layers were then immediately applied to a 1.5 by 3 cm piece of nitrocellulose
washed with sat. Na2CO3 (60 mL), H2O (60 mL) and brine (60 membrane (BioRad) and allowed to air dry. The quantity of
mL), dried (Na2SO4) and concentrated. The residue was puried protein loaded onto each piece of nitrocellulose was at least ten
by silica gel column chromatography (PE to EtOAc) to afford the times less than the binding capacity of the membrane (as
desired amide.
detailed in the manufacturer's protocol). The membranes were
washed with four consecutive large volumes (100 mL) of PBS
and air dried. The pieces of membrane were loaded into scin-
tillation vials, 4 mL of scintillation uid (Amersham) was
added, and the levels of tritium were quantied using a liquid
scintillation counter (Beckman LS6000). All assays were done in
triplicate (2 minutes for sample counting). The Ki value is
calculated from an equation generated based on the establish-
ment of the Dixon plot of each inhibitor and where it is inter-
General protocol for Sharpless CuAAC conjugation
To a solution of the alkyne (0.19 mmol, 1 eq.) and the azide (0.19
mmol, 1 eq.) in tBuOH–H2O (1/1, 5.6 mL/280 mL) were added
CuSO4 (0. 11 mmol, 0.6 eq.) and sodium ascorbate (0.228 mmol,
1.2 eq.). The reaction was stirred at 35 ꢃC for 24 h then diluted
with water (20 mL) and extracted with CH2Cl2 (3 ꢂ 30 mL). The
combined organic layers were dried (Na2SO4) and concentrated.
The residue was puried by silica gel column chromatography
to afford the desired triazole.
sected with the 1/Vmax
.
Docking studies
The O-GlcNAc transferase (OGT) crystal structure (PDB: 4GZ5) was
protonated by Molprobity,39 and chain B was extracted aer
removal of water, ions, and the substrate to serve as the model
macromolecule in the docking study. All four OGT inhibitors
were built with the ribose and GlcNAc (Glc) ring conformations
identical to those of the natural substrate in the crystal struc-
ture.37 The macromolecule and four inhibitors were then pro-
cessed by AutoDockTools to remove non-polar hydrogens.40 A grid
box centered on the position of the natural substrate was con-
General protocol for Meldal CuAAC conjugation
iPr2NEt (0.036 mmol, 0.25 eq.) was added into a ask containing
the azide (0.14 mmol, 1 eq.), the alkyne (0.14 mmol, 1 eq.) and
CuI (0.01 mmol, 0.1 eq.) in DMF (2 mL). The reaction was stirred
at r.t. overnight. Aer 24 h, the solution was diluted with EtOAc
(50 mL), washed with sat. Na2CO3 (2 ꢂ 25 mL) and H2O (30 mL).
The combined aqueous layers were extracted with EtOAc (3 ꢂ 30
mL). The combined organic layers were dried (Na2SO4) and
concentrated. The residue was puried by silica gel column
chromatography to afford the desired triazole.
˚
structed using 24 ꢂ 24 ꢂ 24 points and a spacing of 1 A. Subse-
quently, each inhibitor was docked into OGT by AutoDock Vina,41
which automatically calculates the grid maps and clusters the
results, with the value of the exhaustiveness parameter being 32.
General protocol for Sonogashira reaction
Bromo-arene (1.0 mmol, 1 eq.), Pd(PPh3)4 (0.1 mmol, 0.1 eq.)
and CuI (0.1 mmol, 0.1 eq.) were dissolved in toluene (25 mL)
and the solution was degassed with argon. Then, trimethylsi-
lylacetylene (3.0 mmol, 3 eq.) and diisopropylamine (2.2 mmol,
2.2 eq.) were added. The reaction was stirred for 48 h at r.t.
protected from light then poured into sat. NH4Cl (100 mL). The
aqueous layer was extracted with CH2Cl2 (3 ꢂ 100 mL). The
combined organic layers were washed with H2O (100 mL) and
brine (100 mL), dried (Na2SO4) and concentrated. The residue
was then puried by silica gel column chromatography to afford
the desired product.
Acknowledgements
´
The authors thank the Universite Claude Bernard Lyon 1 and
the CNRS for nancial support. S.V. is grateful for nancial
support from the COST Action CM-1102 MultiGlycoNano. S.W.
`
´
thanks the Ministere de l'Enseignement Superieur et de la
Recherche for a PhD stipend under the international program.
Dr F. Albrieux, C. Duchamp and N. Henriques are gratefully
acknowledged for mass spectrometry analyses. A.S.V.E. and T.L.
´
thank the CNRS and the Comite du Nord de la Ligue Contre le
Cancer for nancial supports. D.J.V. and B.M.P. thank the
Natural Sciences and Engineering Council of Canada (NSERC)
for their support. D.J.V. thanks the Canada Research Chairs
(CRC) program for support as the CRC in Chemical Glycobiol-
ogy and NSERC for the award of an EWR Steacie Memorial
Fellowship. Wesley Zandberg and Lehua Deng are thanked for
preparing the UDP-5S-GlcNAc.
Radioactive OGT inhibition assays
The ability of small molecules to inhibit the OGT activity was
assessed using radiolabelled [3H]-UDP-GlcNAc (American
Radiolabel) as the donor and recombinant nup62 as the
acceptor.38 Assays contained 10 mM nup62 and 1 mM UDP-
GlcNAc for the assay (constant specic activity of 0.8 Ci mmolꢁ1
of [3H]-UDP-GlcNAc) or 50 mM UDP-GlcNAc (constant specic
activity of 0.35 Ci mmolꢁ1 of [3H]-UDP-GlcNAc), 100 nM hOGT,
and various concentrations of the inhibitor (for Ki values a
range of concentrations including 1000, 500, 250, 125, 62.5 and
31.25 mM were used and for IC50 values a range from 3000 to 100
Notes and references
1 G. W. Hart, M. P. Housley and C. Slawson, Nature, 2007, 446,
1017–1022.
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun., 2014, 5, 1172–1178 | 1177

View Article Online
MedChemComm
Concise Article
¨
2 D. Nolte and U. Muller, Mamm. Genome, 2002, 13, 62–64.
3 G. D. Holt, C. M. Snow, A. Senior, R. S. Haltiwanger, L. Gerace
and G. W. Hart, J. Cell Biol., 1987, 104, 1157–1164.
22 M. B. Lazarus, Y. Nam, J. Jiang, P. Sliz and S. Walker, Nature,
2011, 469, 564–567.
23 T. M. Gloster, W. F. Zandberg, J. E. Heinonen, D. L. Shen,
L. Deng and D. J. Vocadlo, Nat. Chem. Biol., 2011, 7, 174–181.
24 B. J. Gross, B. C. Kraybill and S. Walker, J. Am. Chem. Soc.,
2005, 127, 14588–14589.
4 N. E. Olszewski, C. M. West, S. O. Sassi and L. M. Hartweck,
Biochim. Biophys. Acta, 2010, 1800, 49–56.
5 D. A. R. Sinclair, M. Syrzycka, M. S. Macauley, T. Rastgardani,
I. Komljenovic, D. J. Vocadlo, H. W. Brock and B. M. Honda, 25 J. Jiang, M. B. Lazarus, L. Pasquina, P. Sliz and S. Walker,
Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 13427–13432. Nat. Chem. Biol., 2012, 8, 72–77.
6 Y. Perez-Cervera, G. Harichaux, J. Schmidt, F. Debierre- 26 P. Sears and C.-H. Wong, Angew. Chem., Int. Ed., 1999, 38,
Grockiego, V. Dehennaut, U. Bieker, E. Meurice, 2300–2324.
T. Lefebvre and R. T. Schwarz, Amino Acids, 2011, 40, 847– 27 D. C. Koester, A. Holkenbrink and D. B. Werz, Synthesis,
856. 2010, 2010, 3217–3242.
7 A. Shen, H. D. Kamp, A. Grundling and D. E. Higgins, Genes 28 B. Ernst and J. L. Magnani, Nat. Rev. Drug Discovery, 2009, 8,
Dev., 2006, 20, 3283–3295. 661–677.
8 J. d. J. Perez, N. D. Udeshi, J. Shabanowitz, S. Ciordia, 29 C. J. Carroux, J. Moeker, J. Motte, M. Lopez, L. F. Bornaghi,
¨
´
S. Juarez, C. L. Scott, N. E. Olszewski, D. F. Hunt and
J. A. Garcia, Virology, 2013, 442, 122–131.
9 Q. Zeidan and G. W. Hart, J. Cell Sci., 2010, 123, 13–22.
K. Katneni, E. Ryan, J. Morizzi, D. M. Shackleford,
S. A. Charman and S.-A. Poulsen, Bioorg. Med. Chem. Lett.,
2013, 23, 455–459.
10 S. Olivier-Van Stichelen, L. Drougat, V. Dehennaut, I. El 30 C. J. Carroux, G. M. Rankin, J. Moeker, L. F. Bornaghi,
Yazidi-Belkoura, C. Guinez, A. M. Mir, J. C. Michalski,
A. S. Vercoutter-Edouart and T. Lefebvre, Oncogenesis,
2012, 1, e36.
K. Katneni, J. Morizzi, S. A. Charman, D. Vullo,
C. T. Supuran and S.-A. Poulsen, J. Med. Chem., 2013, 56,
9623–9634.
11 L. Drougat, S. Olivier-Van Stichelen, M. Mortuaire, 31 H. Chapuis and P. Strazewski, Tetrahedron, 2006, 62, 12108–
F. Foulquier, A.-S. Lacoste, J.-C. Michalski, T. Lefebvre and 12115.
A.-S. Vercoutter-Edouart, Biochim. Biophys. Acta, 2012, 1820, 32 J. Bures, M. Martin, F. Urpi and J. Vilarrasa, J. Org. Chem.,
1839–1848. 2009, 74, 2203–2206.
12 M.-D. Li, H.-B. Ruan, M. Hughes, J.-S. Lee, J. Singh, S. Jones, 33 J. Garcia, F. Urpi and J. Vilarrasa, Tetrahedron Lett., 1984, 25,
M. Nitabach and X. Yang, Cell Metab., 2013, 17, 303–310. 4841–4844.
13 E. P. Roquemore, M. R. Chevrier, R. J. Cotter and G. W. Hart, 34 H. Chen, J. Zhao, Y. Li, F. Shen, X. Li, Q. Yin, Z. Qin, X. Yan,
´
Biochemistry, 1996, 35, 3578–3586.
14 R. S. Haltiwanger and G. A. Philipsberg, J. Biol. Chem., 1997,
272, 8752–8758.
15 R. S. Haltiwanger, G. D. Holt and G. W. Hart, J. Biol. Chem.,
1990, 265, 2563–2568.
16 B. D. Lazarus, D. C. Love and J. A. Hanover, Glycobiology,
2006, 16, 415–421.
17 R. Hurtado-Guerrero, H. C. Dorfmueller and D. M. F. van
Aalten, Curr. Opin. Struct. Biol., 2008, 18, 551–557.
18 D. J. Vocadlo, Curr. Opin. Chem. Biol., 2012, 16, 488–497.
19 L. Wells, Y. Gao, J. A. Mahoney, K. Vosseller, C. Chen,
Y. Wang, P. Zhang and J. Zhang, Bioorg. Med. Chem. Lett.,
2011, 21, 574–576.
35 S. Wang, J. A. Cuesta-Seijo, D. Lafont, M. M. Palcic and
S. Vidal, Chem.–Eur. J., 2013, 19, 15346–15357.
36 M. Meldal and C. W. Tornoe, Chem. Rev., 2008, 108, 2952–
3015.
37 M. B. Lazarus, J. Jiang, T. M. Gloster, W. F. Zandberg,
G. E. Whitworth, D. J. Vocadlo and S. Walker, Nat. Chem.
Biol., 2012, 8, 966–968.
38 D. L. Shen, T. M. Gloster, S. A. Yuzwa and D. J. Vocadlo, J.
Biol. Chem., 2012, 287, 15395–15408.
A. Rosen and G. W. Hart, J. Biol. Chem., 2002, 277, 1755– 39 V. B. Chen, W. B. Arendall III, J. J. Headd, D. A. Keedy,
1761.
R. M. Immormino, G. J. Kapral, L. W. Murray,
J. S. Richardson and D. C. Richardson, Acta Crystallogr.,
Sect. D: Biol. Crystallogr., 2010, 66, 12–21.
20 B. L. Cantarel, P. M. Coutinho, C. Rancurel, T. Bernard,
V. Lombard and B. Henrissat, Nucleic Acids Res., 2009, 37,
D233–D238.
21 M. Jinek, J. Rehwinkel, B. D. Lazarus, E. Izaurralde,
J. A. Hanover and E. Conti, Nat. Struct. Mol. Biol., 2004, 11,
1001–1007.
40 G. M. Morris, R. Huey, W. Lindstrom, M. F. Sanner,
R. K. Belew, D. S. Goodsell and A. J. Olson, J. Comput.
Chem., 2009, 30, 2785–2791.
41 O. Trott and A. J. Olson, J. Comput. Chem., 2010, 31, 455–461.
1178 | Med. Chem. Commun., 2014, 5, 1172–1178
This journal is © The Royal Society of Chemistry 2014