The Small Molecule 2-Azido-2-deoxy-glucose Is a Metabolic Chemical Reporter of O-GlcNAc Modifications in Mammalian Cells, Revealing an Unexpected Promiscuity of O-GlcNAc Transferase

Article abstract of DOI:10.1021/acschembio.6b00877

Glycans can be directly labeled using unnatural monosaccharide analogs, termed metabolic chemical reporters (MCRs). These compounds enable the secondary visualization and identification of glycoproteins by taking advantage of bioorthogonal reactions. Most widely used MCRs have azides or alkynes at the 2-N-acetyl position but are not selective for one class of glycoprotein over others. To address this limitation, we are exploring additional MCRs that have bioorthogonal functionality at other positions. Here, we report the characterization of 2-azido-2-deoxy-glucose (2AzGlc). We find that 2AzGlc selectively labels intracellular O-GlcNAc modifications, which further supports a somewhat unexpected, structural flexibility in this pathway. In contrast to the endogenous modification N-acetyl-glucosamine (GlcNAc), we find that 2AzGlc is not dynamically removed from protein substrates and that treatment with higher concentrations of per-acetylated 2AzGlc is toxic to cells. Finally, we demonstrate that this toxicity is an inherent property of the small-molecule, as removal of the 6-acetyl-group renders the corresponding reporter nontoxic but still results in protein labeling.

Full text of DOI:10.1021/acschembio.6b00877

Subscriber access provided by Georgetown University | Lauinger and Blommer Libraries
Article
The small molecule 2-Azido-2-deoxy-glucose is a metabolic
chemical reporter of O-GlcNAc modifications in mammalian cells,
revealing an unexpected promiscuity of O-GlcNAc transferase
Balyn W Zaro, Anna R. Batt, Kelly N Chuh, Marisol X Navarro, and Matthew R. Pratt
ACS Chem. Biol., Just Accepted Manuscript • DOI: 10.1021/acschembio.6b00877 • Publication Date (Web): 30 Nov 2016
Downloaded from http://pubs.acs.org on December 4, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Chemical Biology is published by the American Chemical Society. 1155 Sixteenth
Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 11
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
The small molecule 2-Azido-2-deoxy-glucose is a metabolic chemical
reporter of O-GlcNAc modifications in mammalian cells, revealing
an unexpected promiscuity of O-GlcNAc transferase.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Balyn W. Zaro,†,♯Anna R. Batt,†,♯ Kelly N. Chuh,† Marisol X. Navarro,† and Matthew R. Pratt*,†,§
†Department of Chemistry and §Department of Molecular and Computational Biology, University of Southern California,
Los Angeles, CA 90089ꢀ0744
ABSTRACT: Glycans can be directly labeled using unnatural monosaccharide analogs, termed metabolic chemical reporters
(MCRs). These compounds enable the secondary visualization and identification of glycoproteins by taking advantage of
bioorthogonal reactions. Most widelyꢀused MCRs have azides or alkynes at the 2ꢀNꢀacetyl position but are not selective for one
class of glycoprotein over others. To address this limitation, we are exploring additional MCRs that have bioorthogonal functionaliꢀ
ty at other positions. Here, we report the characterization of 2ꢀazidoꢀ2ꢀdeoxyꢀglucose (2AzGlc). We find that 2AzGlc selectively
labels intracellular OꢀGlcNAc modifications, which further supports a somewhat unexpected, structural flexibility in this pathway.
In contrast to the endogenous modification Nꢀacetylꢀglucosamine (GlcNAc), we find that 2AzGlc is not dynamically removed from
protein substrates and that treatment with higher concentrations of perꢀacetylated 2AzGlc is toxic to cells. Finally, we demonstrate
that this toxicity is an inherent property of the smallꢀmolecule, as removal of the 6ꢀacetylꢀgroup renders the corresponding reporter
nonꢀtoxic but still results in protein labeling.
Metabolic chemical reporters (MCRs) of glycans are typically
monosaccharide analogs that contain bioorthogonal functional
groups, such as an alkyne or azide. Once living cells are
and these modifications can be subsequently removed by Oꢀ
GlcNAcase (OGA), rendering OꢀGlcNAcylation dynamic.
14
1ꢀ3
Genetic knockout of either of these enzymes is lethal in
15ꢀ18
treated with one of these analogs, the MCRs can be metaboliꢀ
cally transformed into at least one, if not more, highꢀenergy
sugar donors that can be subsequently utilized by glycosylꢀ
transferases resulting in the incorporation of MCRs into glyꢀ
cans. An ever growing number of bioorthogonal reactions can
then be employed for the selective installation of tags for the
visualization, characterization, and identification of glycoconꢀ
mammals and insects, and the modification plays key roles
19,20
in human disease, including cancer and neurodegeneration.
Here, we demonstrate that 2ꢀazidoꢀ2ꢀdeoxyꢀglucose (2AzGlc)
is also a selective MCR for OꢀGlcNAc modifications. Treatꢀ
ment of cells with the perꢀacetylated reporter Ac42AzGlc (Figꢀ
ure 1A) resulted in the labeling of a variety of proteins by
2AzGlc in mammalian cells but no detectable incorporation of
2AzGlc into cellꢀsurface glycoproteins. Additionally,
bioorthogonal enrichment of proteins that were labeled with
Ac42AzGlc followed by proteomics identified many known
OꢀGlcNAcylated proteins. In contrast to 6AzGlcNAc, 2AzGlc
appears to not be removed from substrate proteins by OGA,
consistent with the enzymatic mechanism that requires anchiꢀ
4
jugates, most notably glycoproteins. In the past five years, we
and others have demonstrated that the majority of glycoprotein
5ꢀ8
MCRs are not specific for one type of glycosylation. For
example, the original MCR for intracellular OꢀGlcNAc modiꢀ
9
fication, Nꢀazidoacetylꢀglucosamine (GlcNAz), labels both Oꢀ
GlcNAcylated proteins and cellꢀsurface glycans (i.e., Nꢀlinked
6,8
21
and mucin Oꢀlinked). Additionally, we have shown that
meric assistance of the Nꢀacetyl group. Additionally, treatꢀ
MCRs that have their bioorthogonal functionality at the Nꢀ
acetyl position, which makes up the vast majority of glycan
reporters, have the potential to be deꢀNꢀacetylated, resulting in
ment of cells with Ac42AzGlc at a high concentration (200
ꢁM) was toxic to cells after overnight exposure. Interestingly,
this toxicity is inherent to the structure of the smallꢀmolecule,
as deprotection of the 6ꢀOꢀacetate to yield Ac32AzGlc reꢀ
moved all cellꢀdeath but retained similar levels of protein laꢀ
beling. Importantly, this observation has been made on other
MCRs by the Yarema lab, which they attribute to changing the
balance between the productive biosynthetic transformation of
reporters to their corresponding donor sugars versus toxic
10
labeling of the protein acetylation pathway. These issues
result in a lack of specificity that can limit the information that
can be gained in glycoprotein visualization experiments or the
number of a specific type of glycoprotein that can be identified
using enrichment and proteomics.
We recently demonstrated that changes to the MCR structure
22
sideꢀreactions.
resulted in the first selective reporter of OꢀGlcNAcylation, 6ꢀ
8
azidoꢀ6ꢀdeoxyꢀNꢀacetylꢀglucosamine
(6AzGlcNAc).
Oꢀ
RESULTS AND DISCUSSION
GlcNAcylation is the addition of the monosaccharide Nꢀ
Given our interesting previous results that structural changes
in the chemical structure of MCRs can change their selectivity,
we set out to explore 2AzGlc, as it lacks the acetamidoꢀ
functionality that is common to the vast majority of monosacꢀ
acetylꢀglucosamine to serine and threonine sideꢀchains of proꢀ
11ꢀ13
teins located in the cytosol, nucleus, and mitochondria. Oꢀ
GlcNAc transferase (OGT) uses the uridine diphosphate sugar
donor UDPꢀGlcNAc to add OꢀGlcNAc to protein substrates,
ACS Paragon Plus Environment

ACS Chemical Biology
charide positions found in glycoproteins and that has been a
Page 2 of 11
1
consistent feature of all previous MCRs. Therefore, we rationꢀ
alized that it may have interesting properties concerning its
metabolism and glycoprotein selectivity. Towards this goal,
we prepared the perꢀOꢀacetylated compound, Ac42AzGlc, in
two steps from glucosamine. These acetates enable diffusion
of the reporter through cellular membranes where they are
removed by lipases and/or hydrolases. NIH3T3 and H1299
cells were then treated with increasing concentrations of
Ac42AzGlc for 6 h, and the corresponding cell lysates were
subjected to the copper(I)ꢀcatalyzed azideꢀalkyne cycloaddiꢀ
tion (CuAAC) with an alkyneꢀrhodamine tag. Inꢀgel fluoresꢀ
cence scanning showed a range of labeled proteins (Figure
1B). In both cell lines, we observed a large increase in labeling
when the concentration of Ac42AzGlc was increased from 150
to 200 ꢁM. Notably, this higher concentration also resulted in
toxicity when the cells were treated for longer periods of time
(i.e., 16 h). This observation is consistent with previous experꢀ
iments by the Bertozzi lab, who commented on the toxicity of
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
23
Ac42AzGlc but did not characterize its labeling. To explore
this toxicity further, the same cell lines were treated with a
range of Ac42AzGlc concentrations for 16 h and their
NAD(P)Hꢀdependent reductase activity was measured using a
commercially available MTS assay (Figure 1C). Interestingly,
an increase in activity was observed for lower concentrations
of Ac42AzGlc, indicating additional metabolic flux into reducꢀ
tive equivalents. However, the cells displayed a dramatic loss
of activity between 150 to 200 ꢁM Ac42AzGlc, consistent with
our morphological observations.
Despite this toxicity, we next set out to characterize the selecꢀ
tivity of Ac42AzGlc as an MCR. To determine if 2AzGlc is
incorporated into cellꢀsurface glycans, H1299 cells were treatꢀ
ed with Ac42AzGlc (200 ꢁM) for 6 h, which is before the onꢀ
set of obvious cellular toxicity. The intact cells were then subꢀ
jected to
a strainꢀpromoted azideꢀalkyne cycloaddition
(SPAAC) with DBCOꢀbiotin, which can only label extracelluꢀ
lar azides, followed by incubation with FITCꢀavidin and analꢀ
ysis by flowꢀcytometry (Figure 2A). As controls, we also simꢀ
ultaneously treated cells with either Ac4GalNAz (200 ꢁM),
8,24
which does label cellꢀsurface glycoproteins,
or
8
Ac36AzGlcNAc (200 ꢁM), which does not. We observed esꢀ
sentially no fluorescence signal from cells treated with either
Ac42AzGlc or Ac36AzGlcNAc, despite strong signal from the
Ac4GalNAzꢀtreated cells, indicating that Ac42AzGlc does not
label cellꢀsurface glycans. We next used Ac42AzGlc in an unꢀ
biased proteomics experiment in NIH3T3 cells, as we have
utilized these cells in the past for the identification of labeled
proteins with other MCRs. Specifically, NIH3T3 cells were
treated in triplicate with Ac42AzGlc or Ac4GlcNAc as a conꢀ
trol (both at 200 ꢁM) for 6 h, again before any signs of toxiciꢀ
ty. The cells were then lysed under denaturing conditions (4%
SDS) and the labeled proteins underwent CuAAC with an
alkyne biotinꢀtag. The labeled proteomes were then reduced
and alkylated with iodoacetamide, followed by incubation
with streptavidinꢀconjugated beads. After extensive washing,
the enriched proteins were subjected to onꢀbead trypsinolysis,
and the resulting peptides were analyzed using LCꢀMS/MS
and identified using Proteome Discoverer and Mascot. Specꢀ
tral counting was used to quantify the proteins enriched in
either sample.
Figure 1. Evaluation of perꢀacetylatedꢀ2ꢀazidoꢀglucose
(Ac42AzGlc) as a metabolic chemical reporter (MCR). (A) Treatꢀ
ment of living cells with MCRs, like Ac42AzGlc, results in modiꢀ
fication of proteins that can then be visualized using bioorthogoꢀ
nal reactions. (B) Ac42AzGlcꢀtreatment results in labeling of a
wide range of proteins. NIH3T3 or H1299 cells were treated with
the indicated concentrations of Ac42AzGlc for 6 h, followed by
CuAAC with alkyneꢀrhodamine and visualization of labeled proꢀ
teins by inꢀgel fluorescence scanning. (C) High concentrations of
Ac42AzGlc are toxic to mammalian cells. NIH3T3 or H1299 cells
were treated with the indicated concentrations of Ac42AzGlc for
16 h and their viability was then measured using an MTS assay.
Quantitated data is the average of three separate biological experꢀ
iments, and error bars represent ±s.e.m. from the mean of biologiꢀ
cal replicates (n = 3).
ACS Paragon Plus Environment

Page 3 of 11
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Figure 2. Ac42AzGlc is an OꢀGlcNAc metabolic chemical reporter. (A) Cell surface glycoproteins are not labeled after treatment with
Ac42AzGlc. H1299 cells were treated with the indicated metabolic chemical reporters (200 ꢁM) for 6 h at which time the cells were colꢀ
lected and subjected to copperꢀfree click chemistry with DBCOꢀbiotin (Click Chemistry Tools). After incubation with FITCꢀavidin, surꢀ
face glycoprotein labeling was analyzed by flow cytometry. (B) Unbiased identification of OꢀGlcNAcylated proteins using 2AzGlc.
NIH3T3 cells were treated in triplicate with Ac42AzGlc (200 ꢁM) or Ac4GlcNAc (200 ꢁM) vehicle for 6 h, followed by CuAAC with
alkyneꢀbiotin, enrichment with streptavidinꢀcoated beads, and onꢀbead trypsinolysis. Proteins that were identified using LCꢀMS/MS are
represented as the total number of positive spectral counts minus the total number of control spectral counts. Representative known Oꢀ
GlcNAcylated proteins are shown. (C) OꢀGlcNAc transferase (OGT) overexpression increases 2AzGlc labeling. NIH3T3 cells that stably
express either HAꢀtagged OGT or an empty plasmid were treated with Ac42AzGlc (200 ꢁM) or DMSO vehicle for 6 h. At this time, the
cell lysates were subjected to CuAAC with alkyneꢀrhodamine and labeled proteins were visualized using inꢀgel fluorescence.
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Proteins were considered to have been labeled by 2AzGlc if
they met certain criteria. First, the proteins must have been
identified by at least 1 peptide in each of the three Ac42AzGlcꢀ
treated samples. Second, the sum of the spectral counts from
the Ac42AzGlcꢀtreated samples must be threeꢀtimes greater
than the sum of the spectral counts for the same protein in the
Ac4GlcNAcꢀtreated sample. Finally, the spectral counts in the
MCRꢀtreated samples must be statistically significantly differꢀ
ent from the control samples (P < 0.05, tꢀtest). With these reꢀ
quirements, we identified 361 proteins (Figure 2B and Table
S1). Importantly, in this list were 265 proteins that had been
previously identified in other proteomics experiments as being
potentially OꢀGlcNAcylated. We then annotated these proteins
based on their subcellular localization, since OꢀGlcNAcylated
proteins will be intracellular while cellꢀsurface glycoproteins
will be extraꢀcellular/lumenal or have a transmembrane doꢀ
main. Using the Uniprot database, we found that Ac42AzGlc
labeled essentially exclusively intracellular or transmembrane
proteins and only resulted in the enrichment of one exclusively
extracellular protein, EGFꢀcontaining fibulinꢀlike extracellular
matrix protein, that could not bear OꢀGlcNAcylation.
plore the generality of 2AzGlc labeling, we next treated a vaꢀ
riety of mammalian cellꢀlines with Ac42AzGlc (200 ꢁM) for 6
h. Inꢀgel fluorescence scanning, after CuAAC with alkyneꢀ
rhodamine, showed strong labeling in all of the cellꢀlines testꢀ
ed (Figure 3), indicating that 2AzGlc is a general MCR for the
visualization and identification of OꢀGlcNAcylated proteins.
We next explored the reversibility of 2AzGlc labeling. As
stated in the introduction, the enzymatic mechanism of OGA
involves anchimeric assistance of the Nꢀacetyl group to generꢀ
ate an oxazoline intermediate. A wealth of carbohydrate chemꢀ
istry has demonstrated that a 2ꢀazido group is incapable of
25
participating in this fashion, raising the likely possibility that
once 2AzGlc is installed by OGT it cannot be removed by
OGA. To directly test this hypothesis, we performed a pulseꢀ
chase experiment. Briefly, HeLa cells were treated with
Ac42AzGlc (200 ꢁM) for 6 h. At this time the media was exꢀ
changed for fresh media containing Ac4GlcNAc (200 ꢁM) and
26
either the OGA inhibitor ThiametꢀG (10 ꢁM) or DMSO vehiꢀ
cle. After an additional 12 h, the corresponding cellꢀlysates
were subjected to CuAAC with alkyneꢀrhodamine and analyꢀ
sis by inꢀgel fluorescence (Figure 4A). In contrast to
6AzGlcNAc labeling that we previously demonstrated is
Next, we used retroviral transformation to generate NIH3T3
cells that stably overexpress an HAꢀtagged OGT or were
transduced with an empty plasmid as a negative control. These
cells were treated with Ac42AzGlc (200 ꢁM) for 6 h, followed
by CuAAC with alkyneꢀrhodamine and analysis using inꢀgel
fluorescence (Figure 2C). Notably, we observed significantly
more labeling in the OGT overexpressing cells, further indiꢀ
cating that 2AzGlc is an MCR for OꢀGlcNAcylation. To exꢀ
8
maintained by ThiametꢀG treatment, we observed no differꢀ
ence in the reduction of 2AzGlc labeling, indicating that it is
not a substrate for OGA. Interestingly, we did, however, obꢀ
serve a relatively rapid loss in 2AzGlc labeling over 12 h, sugꢀ
gesting that the labeled proteins were being turned over by
another mechanism. Proteasomal degradation is one of the
major pathways by which proteins are targeted for turnover.
To determine if 2AzGlcꢀlabled proteins are being destroyed by
ACS Paragon Plus Environment

ACS Chemical Biology
Page 4 of 11
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Ac42AzGlc is a general chemical reporter in a variety of mammalian cell lines. (A) The indicated cell lines were treated
with Ac42AzGlc (200 ꢁM) or DMSO vehicle for 6 h. Total cell lysates were then collected and subjected to CuAAC with alkyneꢀ
rhodamine, and labeled proteins were visualized using inꢀgel fluorescence scanning.
this pathway, we performed another pulseꢀchase experiment.
In this case, the pulse was identical to the one above, but in the
chase, we added either the proteasome inhibitor MGꢀ132 (10
ꢁM) or DMSO control. Inꢀgel fluorescence scanning following
CuAAC with alkyneꢀrhodamine only showed a slight stabiliꢀ
zation of the fluorescent signal (Figure 4B), indicating that the
majority of the 2AzGlcꢀdependent labeling is being removed
by an unknown mechanism.
at all of the concentrations tested (Figure 5C). These results
support an inherent toxicity mechanism that is largely unrelatꢀ
ed to protein labeling by 2AzGlc as an OꢀGlcNAc reporter.
Finally, to explore the potential mechanism underlying the
toxicity of Ac42AzGlcꢀtreatment, we looked to the elegant
experiments performed by the Yarema lab who were exploring
the offꢀtarget effects of monosaccharides modified by shortꢀ
22
chain fatty acids. Although the exact biological mechanism
behind the toxicity of certain acetylated monosaccharides are
not entirely known, the authors demonstrated that the toxicity
of these molecules could be eliminated by revealing the 6ꢀ
hydroxyl functionality, which increases the flux of the carboꢀ
hydrate into cellular biosynthetic pathways. To test whether
this observation could explain the toxicity of Ac42AzGlc, we
synthesized the 6ꢀOH analog (Ac32AzGlc, Figure 5A). First,
we examined whether aqueous conditions would result in miꢀ
gration of the 4ꢀOꢀacetate to the 6ꢀhydroxyl group, as can be
observed under certain reaction conditions. To accomplish
this, we dissolved Ac32AzGlc in a 5:1 mixture of D2O and
deuterated DMSO and incubated the mixture for 16 h at room
temperature. After this length of time the solution was diluted
to 1:1 D2O and deuterated DMSO and analyzed by proton
NMR. Comparison of this spectra to a freshly prepared sample
in the same solvent composition showed essentially no differꢀ
ences (Supplemental Information), demonstrating that any
acetate migration is minimal and below the detection limit of
NMR. Next, treatment with either NIH3T3 or H1299 cells
with Ac32AzGlc resulted in robust labeling of proteins, as
visualized by inꢀgel fluorescence (Figure 5B). However, unꢀ
like Ac42AzGlc, the 6ꢀOH analog showed no cellular toxicity
Figure 4. Characterization of the dynamics of 2AzGlc labeling.
(A) Protein labeling by 2AzGlc is not enzymatically removed by
OꢀGlcNAcase (OGA). HeLa cells were treated with 200 ꢁM
Ac42AzGlc for 6 h. At this time, the media was exchanged for
fresh media containing 200 ꢁM Ac4GlcNAc and either the OGA
inhibitor ThiametꢀG (10 ꢁM) or DMSO. Cells were collected at
the indicated times, subjected to CuAAC with alkyneꢀrhodamine
and analyzed by inꢀgel fluorescence scanning. (B) Labeling by
2AzGlc is not notably stabilized by proteasome inhibition. HeLa
cells were treated with 200 ꢁM Ac42AzGlc for 6 h. At this time,
the media was exchanged for fresh media containing 200 ꢁM
Ac4GlcNAc and either the proteasome inhibitor MG132 (10 ꢁM)
or DMSO. Cells were collected at the indicated times, subjected
to CuAAC with alkyneꢀrhodamine and analyzed by inꢀgel fluoꢀ
rescence scanning.
ACS Paragon Plus Environment

Page 5 of 11
ACS Chemical Biology
not rule out that different proteins are more readily modified
CONCLUSIONS
1
by OGT with a reporter that lacks the 2ꢀacetamido functionaliꢀ
ty versus an MCR like 6AzGlcNAc. Therefore, we suggest
that 6AzGlcNAc be used for studies aimed at the biology of
OꢀGlcNAcylation and that all potentiallyꢀmodified proteins be
confirmed by either antibodies or the chemoenzymatic modifiꢀ
The development of new MCRs for glycans has continued to
grow the chemical toolbox for the investigation of glycobioloꢀ
gy. In particular, these reporters have been key for the visualiꢀ
zation and identification of glycoproteins. We have been foꢀ
cused on understanding the selectivity of different MCRs,
including some that would not be predicted to label proteins
based on the previously established biosynthetic and salvage
pathways. Continuing that theme here, we performed the celꢀ
lular characterization of 2AzGlc as a chemical reporter of glyꢀ
cosylation. Using bioorthogonal chemistry, we show that
2AzGlc labels a range of proteins in total cellꢀlysate (Figure
1B and 3), but does not label cellꢀsurface glycans at a detectaꢀ
ble level (Figure 2A), suggesting that it might be selective for
OꢀGlcNAc modifications. This selectivity was further supꢀ
ported by an unbiased proteomics experiment that used
2AzGlc to enrich a total of 361 proteins, 73.4% of which had
been previously identified in other OꢀGlcNAcylation proteꢀ
omics experiments and only 1 exclusively extracellular or
lumenal glycoprotein (Figure 2C and Table S1). The ability
for 2AzGlc to be transferred by OGT from the corresponding
UDPꢀ2AzGlc sugar donor in cells is somewhat surprising,
given the ternary crystal structures of OGT in complex with its
peptide and donorꢀsugar substrates, as well as the structure
2
3
4
5
6
7
8
9
30
cation strategy developed by the HsiehꢀWilson lab. In sumꢀ
mary, 2AzGlc is the second MCR that has been characterized
to be selective for OꢀGlcNAcylation. It also further highlights
the promiscuity of OGT, which now includes 2AzGlc, as well
5ꢀ9,31
as several Nꢀacetylꢀglucosamine monosaccharides,
and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
could have interesting implications for the modification of
intracellular proteins by natural, non Nꢀacetylꢀglucosamine
monosaccharides that can be biosynthetically transformed to
the corresponding UDPꢀsugar donors.
27
with the corresponding products. These structures show that
the 2ꢀacetamide group of GlcNAc undergoes significant
movement during the enzymatic reaction and that the amide
nitrogen contributes a hydrogenꢀbond to the phosphate of the
UDP leavingꢀgroup. Additionally, the same authors show that
in vitro OGT will accept UDPꢀGalNAc but not UDPꢀGlc or
UDPꢀ2ꢀketoꢀGlc. However, another structural and biochemiꢀ
cal study found that OGT accepts UDPꢀGlcNAcF3 (2ꢀNꢀ
trifluoroacetamide), which they interpreted as evidence that
the electronics of the 2ꢀacetamide are not key for catalysis.
Given these data, the 2ꢀazido group might either be functionalꢀ
ly benign or could participate in weak hydrogen bonds, given
its unique electronics. We also confirmed a statement from
the Bertozzi lab that 2AzGlc is toxic to mammalian cells (Figꢀ
ure 1B), although they did not show any supporting data.
Here, we demonstrate that this toxicity can be eliminated by
deprotection of the 6ꢀposition to reveal the hydroxyl group,
which the Yarema lab has shown increased flux of other monꢀ
osaccharide analogs into the corresponding biosynthetic pathꢀ
ways. Given the mechanism of OGA, we also explored the
reversibility of the 2AzGlc modifications. We used a pulseꢀ
chase experiment to demonstrate that 2AzGlc is not removed
from proteins by OGA (Figure 4A). As stated above, this is
not necessarily surprising, given the requirement for anchimerꢀ
27
28
29
23
21
ic assistance in the enzymatic mechanism. Interestingly, altꢀ
hough loss of 2AzGlc labeling is not OGAꢀdependent, the rate
of loss is similar to other MCRs that are removed by OGA,
Figure 5. The acetylation pattern and not labeling by 2AzGlc is
responsible for cellular toxicity. (A) Proteins can also be labeled
by Ac32AzGlc, which bears a free hydroxyl at the 6ꢀOH position.
(B) Ac32AzGlcꢀtreatment results in labeling of a wide range of
proteins. NIH3T3 or H1299 cells were treated with the indicated
concentrations of Ac32AzGlc for 16 h, followed by CuAAC with
alkyneꢀrhodamine and visualization of labeled proteins by inꢀgel
fluorescence scanning. (C) No concentrations of Ac32AzGlc are
toxic to mammalian cells. NIH3T3 or H1299 cells were treated
with the indicated concentrations of Ac32AzGlc for 16 h and their
viability was then measured using an MTS assay. Quantitated data
is the average of four separate biological experiments, and error
bars represent ±s.e.m. from the mean of biological replicates (n =
4).
8
such as 6AzGlcNAc, suggesting to us that the 2AzGlcꢀlabeled
proteins might be degraded. However, a pulseꢀchase experiꢀ
ment with the proteasomal inhibitor MGꢀ132 only showed a
modest stabilization of the fluorescent signal (Figure 4B),
indicating that another pathway exists to remove these modifiꢀ
cations. Further experiments will be needed to directly test this
hypothesis. Our proteomics data show that the vast majority of
identified proteins have been previously characterized as being
potentially OꢀGlcNAcylated by a variety of chemical and bioꢀ
logical techniques (Table S1 ꢀ proteins highlighted in blue).
This indicates that the specificity for protein substrates beꢀ
tween 2AzGlc and other MCRs is similar. However, we can
ACS Paragon Plus Environment

ACS Chemical Biology
Page 6 of 11
EXPERIMENTAL PROCEDURES
ture 5.18 (d, J = 3.4 Hz, 1H), 4.55 (d, J = 8.0 Hz, 1H), 4.01
1
2
3
4
5
6
7
8
(dd, J = 14.0, 6.0 Hz, 1H), 3.95 (t, J = 9.5 Hz, 1H) 3.86ꢀ3.65
(m, 6H), 3.53ꢀ3.46 (m, 2H), 3.37 (t, J = 9.5 Hz 1H), 3.31ꢀ3.25
General. All reagents used for chemical synthesis were purꢀ
chased from SigmaꢀAldrich, Alfa Aesar or EMD Millipore
unless otherwise specified and used without further purificaꢀ
tion. All anhydrous reactions were performed under argon or
nitrogen atmosphere. Analytical thinꢀlayer chromatography
(TLC) was conducted on EMD Silica Gel 60 F254 plates with
detection by ceric ammonium molybdate (CAM), anisaldeꢀ
hyde or UV. For flash chromatography, 60 Å silica gel (EMD)
13
(m, 1H), 3.20ꢀ3.14 (m, 2H), 0.81 (s, 18H), 0.00 (s, 12H). C
NMR (125 MHz, CDCl3): δ (ppm) 95.89, 91.59, 76.91, 74.47,
72.45, 71.18, 70.08, 66.33, 63.9, 62.99, 25.77, ꢀ5.80. ESIꢀMS
calculated for C12H26N3O5Si [M + H]: 320.1636, found:
319.5972.
1,3,4ꢀTriꢀOꢀAcetylꢀ6ꢀTertꢀButyldimethylsilylꢀ2ꢀAzidoꢀ2ꢀ
1
was utilized. H spectra were obtained at 400, 500, or 600
Deoxyglucose (Ac36TBS2AzGlc, 4): Compound 3 (150 mg,
0.47 mmol) was dissolved in pyridine (2 mL) and cooled to
0°C. Acetic anhydride (200 uL, 2.11 mmol) was added and the
reaction warmed to room temperature over 16 h. The reaction
was diluted in DCM and washed with sodium bicarbonate
(2x), water (2x), and brine (1x). The organic layer was dried
with sodium sulfate, filtered, and concentrated by vacuum
before purification by column chromatography (10% ethyl
acetate:hexanes) to afford the product as a light brown syrup
9
MHz on a Varian spectrometers Mercury 400, VNMRSꢀ500,
or ꢀ600. Chemical shifts are recorded in ppm (δ) relative to
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
solvent. C spectra were obtained at 100, 125 or 150 MHz on
the same instruments.
Cell lines. NIH3T3 cells stably expressing either HAꢀtagged
OꢀGlcNAc Transferase (OGT) or an empty vector were generꢀ
ated using Amphopack293 retroviral packaging cell lines acꢀ
cording to manufacturer procedure (Clontech).
1
(170 mg, 81%). H NMR (400 MHz, CDCl3) of αꢀ, βꢀanomer
Synthesis of chemical reporters. Known compounds Thiaꢀ
metꢀG, Ac4GalNAz, Ac4GlcNAz, Ac42AzGlucose and azidoꢀ
rhodamine were synthesized according to literature proceꢀ
mixture: δ (ppm) 6.28 (d, J = 3.6 Hz, 1H), 5.52 (d, J = 8.6 Hz,
1H), 5.46ꢀ5.40 (m, 1H), 5.14 (dd, J = 10.2, 9.4 Hz, 1H), 5.08ꢀ
5.04 (m, 2H), 3.91ꢀ3.85 (m, 1H), 3.76ꢀ3.70 (m, 1H), 3.69ꢀ
3.58 (m, 6H), 2.16 (d, J = 1.8 Hz, 6H), 2.09 (d, J = 4.7 Hz,
6H), 2.03ꢀ1.99 (m, 6H), 0.86 (s, 18H), 0.04ꢀꢀ0.03, (m, 12H).
8,26,32,33
dures.
Synthesis
of
1,3,4,6-Tetra-O-Acetyl-2-Azido-2-
13
Deoxyglucose (Ac42AzGlc, 2):
C NMR (125 MHz, CDCl3): δ (ppm) 169.7, 169.14, 168.38,
92.23, 89.83, 74.95, 73.01, 72.17, 71.08, 68.01, 62.34, 61.53,
60.19, 25.51, 20.52, ꢀ5.72. ESIꢀMS calculated for
2ꢀAzidoꢀ2ꢀDeoxyglucose (1): Commercially available gluꢀ
cosamine hydrochloride was subjected to azide formation usꢀ
+
34
C18H31N3O8SiNa [M + Na] : 468.1773, found: 468.1766.
ing a diazo transfer reaction according to literature procedure.
1
H NMR (500 MHz, Deuterium Oxide) of βꢀanomer δ 4.71 (d,
1,3,4ꢀTriꢀOꢀAcetylꢀ2ꢀAzidoꢀ2ꢀDeoxyglucose (Ac32AzGlc, 5):
Compound 4 (170 mg, 0.38 mmol) was dissolved in THF (2
mL) and cooled to 0°C. Glacial acetic acid (445 ul) was added
followed by addition of 1.0 M TBAF in THF (7.8 mL) and the
reaction was allowed to warm to room temperature over 16
hours. The solution was diluted with ethyl acetate, washed
with sodium bicarb (2x), water (2x), and brine (1x), and dried
with sodium sulfate. The resulting mixture was filtered, conꢀ
centrated by vacuum, and purified by column chromatography
(40% ethyl acetate:hexanes) to afford product as a yellow syrꢀ
J = 8.1 Hz, 1H), 3.92 – 3.82 (m, 1H), 3.81 – 3.70 (m, 1H),
3.54 – 3.43 (m, 3H), 3.28 (dd, J = 9.6, 8.2 Hz, 1H). The prodꢀ
uct was used in the subsequent reaction without further charꢀ
acterization.
1,3,4,6ꢀTetraꢀOꢀAcetylꢀ2ꢀAzidoꢀ2ꢀDeoxyglucose (Ac42AzGlc,
2): Compound 1 was resuspended in pyridine and acetic anhyꢀ
dride. A catalytic amount of dimethylaminopyridine was addꢀ
ed and reaction stirred at room temperature for 16 h. Upon
completion reaction was concentrated, and the crude was reꢀ
suspended in dichloromethane. The organic layer was washed
with equivolume 1 M HCl (1x), saturated Sodium Bicarbonate
(x2), Water (x2) and brine (1x). The washed organic layer was
then concentrated and subjected to column chromatography to
yield the pure product. NMR characterization of this known
compound was consistent with previous reports. H NMR
(500 MHz, CDCl3) of βꢀanomer δ 5.55 (d, J = 8.5 Hz, 1H),
5.15 – 5.00 (m, 2H), 4.34 – 4.25 (m, 1H), 4.11 – 4.05 (m, 1H),
3.84 – 3.77 (m, 1H), 3.71 – 3.62 (m, 1H), 2.19 (s, 3H), 2.09 (s,
3H), 2.07 (d, 3H), 2.02 (s, 3H).
1
up (107 mg, 85%). H NMR (400 MHz, CDCl3) of αꢀ, βꢀ
anomer mixture: δ (ppm) 6.27 (d, J = 3.7 Hz, 1H), 5.54 (d, J =
8.5 Hz, 1H), 5.32 (dd, J = 10.5, 9.3 Hz, 1H), 4.95ꢀ4.90 (m,
1H), 4.56ꢀ4.47 (m, 2H), 4.21 (dd, J = 12.5, 1.7 Hz, 1H), 4.21
(dd, 12.5, 2.3 Hz, 1H), 3.92ꢀ3.87 (m, 1H), 3.65ꢀ3.48 (m, 6H),
13
23
1
2.20ꢀ2.18 (m, 12H), 2.12 (s, 6H). C NMR (125 MHz,
CDCl3): δ (ppm) 171.28, 170.95, 168.34, 92.35, 89.91, 74.86,
72.73, 72.03, 68.27, 62.16, 59.92, 20.53. ESIꢀMS calculated
+
for C12H18N3O8 [M + H] : 332.1088, found: 332.1085.
Cell culture. NIH3T3 and MEF cells were cultured in high
glucose DMEM media (HyClone, ThermoScientific) enriched
with 10% fetal calf serum (FCS, HyClone, ThermoScientific).
MDAꢀMBꢀ231, MDAꢀMBꢀ468, HeLa and MCFꢀ7 were culꢀ
tured in high glucose DMEM (HyClone, ThermoScientific)
enriched with 10% fetal bovine serum (FBS HyClone, Therꢀ
moScientific). H1299 and HCT15 cells were cultured in RPMI
1640 medium (HyClone, ThermoScientific) enriched with
10% FBS. A549 cells were cultured in Fꢀ12K medium (Hyꢀ
Clone, ThermoScientific) enriched with 10% FBS. MCF10A
and SHꢀSY5Y were cultured in a 1:1 mixture of Fꢀ12 and
DMEM media (HyClone, ThermoScientific) enriched with
10% FBS. All cell lines were maintained in a humidified incuꢀ
bator at 37 ℃ and 5.0% CO2.
Synthesis of 1,3,4-Tri-O-Acetyl-2-Azido-2-Deoxyglucose
(Ac32AzGlc, 5):
6ꢀTertꢀButyldimethylsilylꢀ2ꢀAzidoꢀ2ꢀDeoxyglucose
(6TBS2AzGlc, 3): Compound 1 (668 mg, 3.26 mmol) was
dissolved in dry DMF (8 mL) under nitrogen at room temperaꢀ
ture. Imidazole (667 mg, 9.8 mmol) and tertꢀ
butyldimethylsilyl chloride (613 mg, 4.07 mmol) were added
and the reaction stirred for 16 h. The reaction was diluted in
ethyl acetate and washed with sodium bicarbonate (2x), water
(2x), and brine (1x). The organic layer was dried with sodium
sulfate, filtered, and concentrated by vacuum. The compound
was purified by column chromatography (25% ethyl aceꢀ
tate:hexanes) to afford the product as a light brown syrup (327
1
mg, 31%). H NMR (400 MHz, CDCl3) of αꢀ, βꢀanomer mixꢀ
ACS Paragon Plus Environment

Page 7 of 11
ACS Chemical Biology
Metabolic labeling. To cells at 80ꢀ85% confluency, highꢀ or g at 4 °C). Cells were then resuspended in 200 ꢁL PBS conꢀ
1
2
3
4
5
6
7
8
lowꢀglucose media containing Ac4GlcNAz, Ac4GalNAz,
Ac42AzGlc (1,000 x stock in DMSO), or DMSO vehicle was
added as indicated. For chase experiments, lowꢀglucose media
was replaced with highꢀglucose media supplemented with 200
µM Ac4GlcNAc (Sigma) with or without 20 µM ThiametꢀG.
taining DBCOꢀbiotin (Click Chemistry Tools, 60 ꢁM) for 1 h,
after which time they were washed three times with PBS (5
min, 300 x g at 4 °C) before being resuspended in iceꢀcold
PBS containing fluorescein isothiocynate (FITC) conjugated
avidin (Sigma, 5 ꢁg/mL, 30 min at 4 °C). Cells were then
washed three times in PBS (5 min, 300 x g at 4 °C) before
being resuspended in 400 ꢁL PBS for flowꢀcytometry analysis.
A total of 10,000 cells [dead cells were excluded by treatment
Preparation of NP-40-soluble lysates. The cells were colꢀ
lected by trypsinization and pelleted by centrifugation at for 4
min at 2,000 x g, followed by washing 2x with PBS (1 mL).
Cell pellets were then resuspended in 100 ꢁL of 1% NPꢀ40
lysis buffer [1% NPꢀ40, 150 mM NaCl, 50 mM triethanolaꢀ
mine (TEA) pH 7.4] with Complete, Mini, EDTAꢀfree Proteꢀ
ase Inhibitor Cocktail Tablets (Roche) for 20 min and then
centrifuged for 10 min at 10,000 x g at 4 °C. The supernatant
(soluble cell lysate) was collected and the protein concentraꢀ
tion was determined by BCA assay (Pierce, ThermoScienꢀ
tific).
−1
with propidium iodide (2.5 ꢁg mL in water, 30 min)] were
analyzed on a BD SORP LSRII Flow Cytometer using the 488
nm argon laser.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Biotin Enrichment and On-bead Trypsinolysis. NIH3T3 or
H1299 cellꢀpellets labeled with Ac36AzGlcNAc, Ac32AzGlc
or Ac4GlcNAc for 16 hours were resuspended in 200 ꢁL H2O,
60 ꢁL PMSF in H2O (250 mM), and 500 ꢁL 0.05% SDS buffer
(0.05% SDS, 10 mM TEA pH 7.4, 150 mM NaCl) with Comꢀ
plete Mini protease inhibitor cocktail (Roche Biosciences). To
this was added 8 ꢁL Benzonase (Sigma), and the cells were
incubated on ice for 30 min. Then, 4% SDS buffer (2000 ꢁL)
was added, and the cells were briefly sonicated in a bath soniꢀ
cator followed by centrifugation (20,000 x g for 10 min at 15
°C). Soluble protein concentration was normalized by BCA
Cu(I)-catalyzed [3 + 2] azide-alkyne cycloaddition (Cu-
AAC). Cell lysate (200 ꢁg) was diluted with 1% NPꢀ40 lysis
−1
buffer to obtain a desired concentration of 1 ꢁg ꢁL . Newlyꢀ
made click chemistry cocktail (12 ꢁL) was added to each samꢀ
ple [alkynylꢀrhodamine tag (100 ꢁM, 10 mM stock solution in
DMSO); tris(2ꢀcarboxyethyl)phosphine hydrochloride (TCEP)
(1 mM, 50 mM freshly prepared stock solution in water);
−1
assay (Pierce, ThermoScientific) to 1 mg mL , and 10 mg of
total protein was subjected to the appropriate amount of click
chemistry cocktail containing alkyneꢀPEG3ꢀbiotin (5 mM,
Click Chemistry Tools) for 1 h, after which time 10 volumes
of iceꢀcold MeOH were added. Precipitation proceeded 2
hours at ꢀ20 °C. Precipitated proteins were centrifuged at
5,200 x g for 30 min at 0 °C and washed 3 times with 40 mL
iceꢀcold MeOH, with resuspension of the pellet each time. The
pellet was then airꢀdried for 1 h. To capture the biotinylated
proteins by streptavidin beads, the airꢀdried protein pellet was
resuspended in 2 mL of resuspension buffer (6 M urea, 2 M
thiourea, 10 mM HEPES pH 8.0) by bath sonication. To cap
cysteine residues, 100 ꢁl of freshlyꢀmade TCEP (200 mM
stock solution, Thermo) was then added and the mixture incuꢀ
bated for 30 min, followed by 40 ꢁl of freshly prepared iodoaꢀ
cetamide (1 M stock solution, Sigma) and incubation for a
further 30 min in the dark. Steptavadin beads (250 ꢁL of a
50% slurry per sample, Thermo) were washed 2 X with 1 mL
PBS and 1 X with 1 mL resuspension buffer and resuspended
in resuspension buffer (200 ꢁL). Each sample was combined
with streptavadin beads and incubated on a rotator for 2 h.
tris[(1ꢀbenzylꢀ1ꢀHꢀ1,2,3ꢀtriazolꢀ4ꢀyl)methyl]amine
(TBTA)
(100 ꢁM, 10 mM stock solution in DMSO); CuSO4•5H2O (1
mM, 50 mM freshly prepared stock solution in water)] for a
total reaction volume of 200 ꢁL. The reaction was gently vorꢀ
texed and allowed to sit at room temperature for 1 h. Upon
completion, 1 mL of ice cold methanol was added to the reacꢀ
tion, and it was placed at ꢀ20 °C for 2 h to precipitate proteins.
The reactions were then centrifuged at 10,000 x g for 10 min
at 4 °C. The supernatant was removed, the pellet was allowed
to air dry for 15 min, and then 50 ꢁL 4% SDS buffer (4%
SDS, 150 mM NaCl, 50 mM TEA pH 7.4) was added to each
sample. The mixture was sonicated in a bath sonicator to enꢀ
sure complete dissolution, and 50 ꢁL of 2x SDSꢀfree loading
buffer (20% glycerol, 0.2% bromophenol blue, 1.4% βꢀ
mercaptoethanol, pH 6.8) was then added. The samples were
boiled for 5 min at 97 °C, and 40 ꢁg of protein was then loadꢀ
ed per lane for SDSꢀPAGE separation (Any Kd, Criterion Gel,
BioꢀRad).
In-gel Fluorescence Scanning. Following SDSꢀPAGE sepaꢀ
ration, gels were scanned on a Typhoon 9400 Variable Mode
Imager (GE Healthcare) using a 532 nm for excitation and 30
nm bandpass filter centered at 610 nm for detection.
®
These mixtures were then transferred to Mini BioꢀSpin colꢀ
umns (BioꢀRad) and placed on a vacuum manifold. Captured
proteins were then washed with agitation 5 X with resuspenꢀ
sion buffer (10 mL), 5 X PBS (10 mL), 5x with 1% SDS in
PBS (10 mL), 30x with PBS (1 mL per wash, vacuum applied
between each wash), and 5x 2M urea in PBS (1 mL per wash,
vacuum applied between each wash). Beads were then resusꢀ
pended in 2 M urea in PBS (1 mL), transferred to screwꢀtop
tubes, and pelleted by centrifugation (2000 x g for 2 min). At
this time, 800 ꢁL of the supernatant was removed, leaving a
volume of 200 ꢁL. To this beadꢀmixture was added 2 ꢁL o f
CaCl2 (200 mM stock, 1 mM final concentration) and 2 ꢁL of
4
MTS Assay. NIHꢀ3T3 and H1299 cells (1 x 10 cells) were
plated per well in a 96ꢀwell, white bottom dish 24 hours before
treatment with 0ꢀ200 ꢁM Ac4GlcNAc or Ac32AzGlucose for
16 hours under low glucose media conditions (1 g/L).
®
CellTiter 96 Aqueous NonꢀRadioactive Cell Proliferation
Assay (Promega, Madison, WI) was used according to the
provided protocol. Absorbance at 490 nm was read using a
BioTek Synergy H4 MultiꢀMode Microplate reader.
Flow Cytometry of Cell-Surface Labeling with DBCO-
Biotin. H1299 cells grown in 6ꢀwell plates at 80ꢀ85% confluꢀ
ency were treated with 200 ꢁM Ac4GlcNAc, Ac4GlcNAz,
Ac4GalNAz or Ac32AzGlc in triplicate for 16 hours at which
time media was removed and cells were gently washed with
PBS before being detached from the plate with 1 mM EDTA
in PBS. Cells were collected by centrifugation (5 min, 300 x g
at 4 °C) and were washed three times with PBS (5 min, 300 x
−1
1 mg mL sequence grade trypsin (Promega) and incubated at
37 °C for 18 hours. The resulting mixtures of tryptic peptides
®
and beads were transferred to Mini BioꢀSpin columns (Bioꢀ
Rad) and the eluent was collected by centrifugation (1,000 x g
for 2 min). Any remaining peptides were eluted by addition of
100 ꢁL of 2 M urea in PBS followed by centrifugation as imꢀ
mediately above. The tryptic peptides were then applied to
ACS Paragon Plus Environment

ACS Chemical Biology
Page 8 of 11
(5) Boyce, M., Carrico, I. S., Ganguli, A. S., Yu, S.ꢀH., Hangauer, M.
C18 spin columns (Pierce) according to manufacturer's inꢀ
structions, eluted with 70% acetonitrile in H2O, and concenꢀ
trated to dryness on a speedvac.
J., Hubbard, S. C., Kohler, J. J., and Bertozzi, C. R. (2011) Metabolic
crossꢀtalk allows labeling of Oꢀlinked {beta}ꢀNꢀacetylglucosamineꢀ
modified proteins via the Nꢀacetylgalactosamine salvage pathway.
Proc Natl Acad Sci USA 108, 3141–3146.
(6) Zaro, B. W., Yang, Y.ꢀY., Hang, H. C., and Pratt, M. R. (2011)
Chemical reporters for fluorescent detection and identification of Oꢀ
GlcNAcꢀmodified proteins reveal glycosylation of the ubiquitin ligase
NEDD4ꢀ1. Proc Natl Acad Sci USA 108, 8146–8151.
(7) Bateman, L. A., Zaro, B. W., Chuh, K. N., and Pratt, M. R. (2013)
NꢀPropargyloxycarbamate monosaccharides as metabolic chemical
reporters of carbohydrate salvage pathways and protein glycosylation.
Chem Commun 49, 4328–4330.
(8) Chuh, K. N., Zaro, B. W., Piller, F., Piller, V., and Pratt, M. R.
(2014) Changes in metabolic chemical reporter structure yield a selecꢀ
tive probe of OꢀGlcNAc modification. J Am Chem Soc 136, 12283–
12295.
(9) Vocadlo, D., Hang, H., Kim, E., Hanover, J., and Bertozzi, C.
(2003) A chemical approach for identifying OꢀGlcNAcꢀmodified
proteins in cells. Proc Natl Acad Sci USA 100, 9116–9121.
(10) Zaro, B. W., Chuh, K. N., and Pratt, M. R. (2014) Chemical
Reporter for Visualizing Metabolic CrossꢀTalk between Carbohydrate
Metabolism and Protein Modification. ACS Chem Biol 9, 1991–1996.
(11) Torres, C. R., and Hart, G. W. (1984) Topography and polypepꢀ
tide distribution of terminal Nꢀacetylglucosamine residues on the
surfaces of intact lymphocytes. Evidence for Oꢀlinked GlcNAc. J Biol
Chem 259, 3308–3317.
1
2
3
4
5
6
7
8
LC-MS/MS Proteomic Analysis. Peptides were desalted on a
trap column following separation on a 12cm/75um reversed
phase C18 column (Nikkyo Technos Co., Ltd. Japan). A 3
hour gradient increasing from 10% B to% 45% B in 3 hours
(A: 0.1% Formic Acid, B: Acetonitrile/0.1% Formic Acid)
−1
was delivered at 150 nL min . The liquid chromatography
setup (Dionex, Boston, MA, USA) was connected to an Orꢀ
bitrap XL (Thermo, San Jose, CA, USA) operated in topꢀ5ꢀ
mode. Acquired tandem MS spectra (CID) were extracted
using ProteomeDiscoverer v. 1.3 (Thermo, Bremen, Germany)
and queried against the human Uniprot protein database using
MASCOT 2.3.02 (Matrixscience, London, UK). Peptides fulꢀ
filling a Percolator calculated 1% false discovery rate threshꢀ
old were reported. All LCꢀMS/MS analysis were carried out at
the Proteomics Resource Center at The Rockefeller Universiꢀ
ty, New York, NY, USA. Excel files containing identified
proteins will be made available upon request.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ASSOCIATED CONTENT
Supporting figures and proteomic data tables. This material is
available free of charge via the Internet at http://pubs.acs.org.
(12) Holt, G. D., and Hart, G. W. (1986) The subcellular distribution
of terminal Nꢀacetylglucosamine moieties. Localization of a novel
proteinꢀsaccharide linkage, Oꢀlinked GlcNAc. J Biol Chem 261,
8049–8057.
AUTHOR INFORMATION
Corresponding Author
(13) Zachara, N. E., and Hart, G. W. (2002) The emerging signifiꢀ
cance of OꢀGlcNAc in cellular regulation. Chem Rev 102, 431–438.
(14) Vocadlo, D. J. (2012) OꢀGlcNAc processing enzymes: catalytic
mechanisms, substrate specificity, and enzyme regulation. Curr Opin
Chem Biol 16, 488–497.
(15) Shafi, R., Iyer, S. P., Ellies, L. G., O'Donnell, N., Marek, K. W.,
Chui, D., Hart, G. W., and Marth, J. D. (2000) The OꢀGlcNAc transꢀ
ferase gene resides on the X chromosome and is essential for embryꢀ
onic stem cell viability and mouse ontogeny. Proc Natl Acad Sci USA
97, 5735–5739.
(16) O'Donnell, N., Zachara, N. E., Hart, G. W., and Marth, J. D.
(2004) Ogtꢀdependent Xꢀchromosomeꢀlinked protein glycosylation is
a requisite modification in somatic cell function and embryo viability.
Mol Cell Biol 24, 1680–1690.
(17) Sinclair, D. A. R., Syrzycka, M., Macauley, M. S., Rastgardani,
T., Komljenovic, I., Vocadlo, D. J., Brock, H. W., and Honda, B. M.
(2009) Drosophila OꢀGlcNAc transferase (OGT) is encoded by the
Polycomb group (PcG) gene, super sex combs (sxc). Proc Natl Acad
Sci USA 106, 13427–13432.
(18) Yang, Y. R., Song, M., Lee, H., Jeon, Y., Choi, E.ꢀJ., Jang, H.ꢀJ.,
Moon, H. Y., Byun, H.ꢀY., Kim, E.ꢀK., Kim, D. H., Lee, M. N., Koh,
A., Ghim, J., Choi, J. H., LeeꢀKwon, W., Kim, K. T., Ryu, S. H., and
Suh, P.ꢀG. (2012) OꢀGlcNAcase is essential for embryonic developꢀ
ment and maintenance of genomic stability. Aging Cell 11, 439–448.
(19) Ma, Z., and Vosseller, K. (2013) OꢀGlcNAc in cancer biology.
Amino Acids 45, 719–733.
(20) Yuzwa, S. A., and Vocadlo, D. J. (2014) OꢀGlcNAc and neuroꢀ
degeneration: biochemical mechanisms and potential roles in Alzheiꢀ
mer's disease and beyond. Chem Soc Rev 43, 6839–6858.
(21) Macauley, M. S., Whitworth, G. E., Debowski, A. W., Chin, D.,
and Vocadlo, D. J. (2005) OꢀGlcNAcase uses substrateꢀassisted catalꢀ
ysis: kinetic analysis and development of highly selective mechanismꢀ
inspired inhibitors. J Biol Chem 280, 25313–25322.
(22) Aich, U., Campbell, C. T., Elmouelhi, N., Weier, C. A., Samꢀ
pathkumar, S. G., Choi, S. S., and Yarema, K. J. (2008) Regioisomerꢀ
ic SCFA Attachment to Hexosamines Separates Metabolic Flux from
Cytotoxicity and MUC1 Suppression. ACS Chem Biol 3, 230–240.
(23) Saxon, E., Luchansky, S. J., Hang, H. C., Yu, C., Lee, S. C., and
Bertozzi, C. R. (2002) Investigating cellular metabolism of synthetic
* matthew.pratt@usc.edu
Author Contribution
# These authors contributed equally to this manuscript.
Notes
The authors have no competing financial interests.
ACKNOWLEDGMENT
The authors thank H. Molina (Rockefeller University) for asꢀ
sistance with the proteomic analysis. B.W.Z and K.N.C. were
fellows of the National Science Foundation Graduate Research
Fellowship Program (DGEꢀ0937362). This research was supꢀ
ported by the National Science Foundation (CHEꢀ1506503 to
M.R.P.), Susan G. Komen for the Cure (CCR14299333 to
M.R.P.), and the American Cancer Society (RSGꢀ14ꢀ225ꢀ01ꢀ
CCG to M.R.P.), and in part by the National Cancer Institute
of the US National Institutes of Health (CCSG
P30CA014089). Flow cytometry was performed in the USC
Flow Cytometry Core Facility. LCꢀMS/MS proteomic analysis
was performed at the Rockefeller University Proteomics Reꢀ
source Center.
REFERENCES
(1) Grammel, M., and Hang, H. C. (2013) chemical reporters for bioꢀ
logical discovery. Nat Chem Biol 9, 475–484.
(2) Chuh, K. N., and Pratt, M. R. (2015) Chemical methods for the
proteomeꢀwide identification of posttranslationally modified proteins.
Curr Opin Chem Biol 24, 27–37.
(3) Chuh, K. N., Batt, A. R., and Pratt, M. R. (2016) Chemical Methꢀ
ods for Encoding and Decoding of Posttranslational Modifications.
Cell Chemical Biology 23, 86–107.
(4) Patterson, D. M., Nazarova, L. A., and Prescher, J. A. (2014)
Finding the right (bioorthogonal) chemistry. ACS Chem Biol 9, 592–
605.
ACS Paragon Plus Environment

Page 9 of 11
ACS Chemical Biology
azidosugars with the Staudinger ligation. J Am Chem Soc 124, 14893–
14902.
sugar pyrophosphate participation in catalysis. Nat Chem Biol 8, 969–
974.
1
(24) Hang, H. C., Yu, C., Kato, D. L., and Bertozzi, C. R. (2003) A
metabolic labeling approach toward proteomic analysis of mucinꢀtype
Oꢀlinked glycosylation. Proc Natl Acad Sci USA 100, 14846–14851.
(25) Kerns, R. J., and Wei, P. (2012) Synthetic Methods To Incorpoꢀ
rate αꢀLinked 2ꢀAminoꢀ2ꢀDeoxyꢀDꢀGlucopyranoside and 2ꢀAminoꢀ2ꢀ
DeoxyꢀDꢀGalactopyranoside Residues into Glycoconjugate Strucꢀ
tures, in Glycobiology and Drug Design, pp 235–263. American
Chemical Society, Washington, DC.
(26) Yuzwa, S. A., Macauley, M. S., Heinonen, J. E., Shan, X., Denꢀ
nis, R. J., He, Y., Whitworth, G. E., Stubbs, K. A., McEachern, E. J.,
Davies, G. J., and Vocadlo, D. J. (2008) A potent mechanismꢀinspired
OꢀGlcNAcase inhibitor that blocks phosphorylation of tau in vivo.
Nat Chem Biol 4, 483–490.
(27) Lazarus, M. B., Jiang, J., Gloster, T. M., Zandberg, W. F., Whitꢀ
worth, G. E., Vocadlo, D. J., and Walker, S. (2012) Structural snapꢀ
shots of the reaction coordinate for OꢀGlcNAc transferase. Nat Chem
Biol 8, 966–968.
(28) Schimpl, M., Zheng, X., Borodkin, V. S., Blair, D. E., Ferenꢀ
bach, A. T., ttelkopf, A. W. S. U., Navratilova, I., Aristotelous, T.,
Albarbarawi, O., Robinson, D. A., Macnaughtan, M. A., and van
Aalten, D. M. F. (2012) OꢀGlcNAc transferase invokes nucleotide
(29) Tchertanov, L. (1999) Structural metrics relationships in covaꢀ
lently bonded organic azides. Acta Crystallogr., B 55, 807–809.
(30) Clark, P. M., Dweck, J. F., Mason, D. E., Hart, C. R., Buck, S.
B., Peters, E. C., Agnew, B. J., and HsiehꢀWilson, L. C. (2008) Direct
InꢀGel Fluorescence Detection and Cellular Imaging of OꢀGlcNAcꢀ
Modified Proteins. J Am Chem Soc 130, 11576–11577.
(31) Li, J., Wang, J., Wen, L., Zhu, H., Li, S., Huang, K., Jiang, K.,
Li, X., Ma, C., Qu, J., Parameswaran, A., Song, J., Zhao, W., and
Wang, P. G. (2016) An OGAꢀResistant Probe Allows Specific Visualꢀ
ization and Accurate Identification of OꢀGlcNAcꢀModified Proteins
in Cells. ACS Chem Biol 11, 3002–3006.
(32) Saxon, E., and Bertozzi, C. (2000) Cell surface engineering by a
modified Staudinger reaction. Science 287, 2007–2010.
(33) Charron, G., Zhang, M. M., Yount, J. S., Wilson, J., Raghavan,
A. S., Shamir, E., and Hang, H. C. (2009) Robust Fluorescent Detecꢀ
tion of Protein FattyꢀAcylation with Chemical Reporters. J Am Chem
Soc 131, 4967–4975.
(34) GoddardꢀBorger, E. D., and Stick, R. V. (2007) An Efficient,
Inexpensive, and ShelfꢀStable Diazotransfer Reagent: Imidazoleꢀ1ꢀ
sulfonyl Azide Hydrochloride. Org Lett 9, 3797–3800.
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

ACS Chemical Biology
Page 10 of 11
Authors are required to submit a graphic entry for the Table of Contents (TOC) that, in conjunction with the manuscript title,
should give the reader a representative idea of one of the following: A key structure, reaction, equation, concept, or theorem,
etc., that is discussed in the manuscript. Consult the journal’s Instructions for Authors for TOC graphic specifications.
1
2
3
4
5
Insert Table of Contents artwork here
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10
ACS Paragon Plus Environment

Page 11 of 11
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ToC Figure
82x28mm (300 x 300 DPI)
ACS Paragon Plus Environment