Inhibition of Mucin-type O-glycosylation through metabolic processing and incorporation of N-thioglycolyl-d-galactosamine peracetate (Ac 5GalNTGc)

Article abstract of DOI:10.1021/ja405189k

Mucin-type O-glycans form one of the most abundant and complex post-translational modifications (PTM) on cell surface proteins that govern adhesion, migration, and trafficking of hematopoietic cells. Development of targeted approaches to probe functions o

Full text of DOI:10.1021/ja405189k

Subscriber access provided by TEXAS STATE UNIV SAN MARCOS
Article
Inhibition of mucin-type O-glycosylation through metabolic processing and
5
incorporation of N-thioglycolyl-D-galactosamine peracetate (AcGalNTGc).
Kavita Agarwal, Rachna Kaul, Monika Garg, Asif Shajahan,
Saroj Kumar Jha, and Srinivasa-Gopalan Sampathkumar
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/ja405189k • Publication Date (Web): 29 Aug 2013
Downloaded from http://pubs.acs.org on August 30, 2013
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.
Journal of the American Chemical Society 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 17
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
Inhibition of mucin-type O-glycosylation through metabolic
processing and incorporation of N-thioglycolyl-D-
galactosamine peracetate (Ac5GalNTGc).
9
Kavita Agarwal, Rachna Kaul, Monika Garg, Asif Shajahan, Saroj Kumar Jha, and Srinivasa-Gopalan
Sampathkumar*
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
Laboratory of Chemical Glycobiology (CGB), National Institute of Immunology (NII), Aruna Asaf Ali Marg, New Delhi
110067, India. E-mail: gopalan@nii.ac.in, Ph: +91-11-26703843.
KEYWORDS: GalNAc analogs, O-glycans, CD43, thiols, and lectins.
ABSTRACT: Mucin-type O-glycans form one of the most abundant and complex post-translational modification (PTM) on cell
surface proteins that govern adhesion, migration, and trafficking of hematopoietic cells. Development of targeted approaches to
probe functions of O-glycans is at an early stage. Amongst several approaches, small molecules with unique chemical functional
groups that could modulate glycan biosynthesis form a critical tool. Herein, we show that metabolism of peracetyl N-acyl-D-
galactosamine derivatives carrying N-thioglycolyl- (Ac5GalNTGc, 1) moiety – but not N-glycolyl- (Ac5GalNGc, 2) and N-acetyl-
(Ac4GalNAc, 3) – through the N-acetyl-D-galactosamine (GalNAc) salvage pathway induced abrogation of MAL-II and PNA
epitopes in Jurkat cells. Mass spectrometry of permethylated O-glycans from Jurkat cells confirmed presence of significant amounts
of elaborated O-glycans (sialyl-T and disialyl-T) which were inhibited upon treatment with 1. O-glycosylation of CD43, a cell sur-
face antigen rich in O-glycans, was drastically reduced by 1 in a thiol-dependent manner. By contrast, only mild effects were ob-
served for CD45 glycoforms. Direct metabolic incorporation of 1 was confirmed by thiol-selective Michael addition reaction of
immunoprecipitated CD43-myc/FLAG. Mechanistically, CD43 glycoforms were unperturbed by peracetylated N-(3-
acetylthiopropanoyl)- (4), N-(4-acetylthiobutanoyl)- (5), and N-methylthioacetyl- (6) galactosamine derivatives, N-thioglycolyl-D-
glucosamine (7, C-4 epimer of 1), and α-O-benzyl 2-acetamido-2-deoxy-D-galactopyranoside (8), confirming the critical require-
ment of both free sulfhydryl and galactosamine moieties for inhibition of mucin-type O-glycans. Similar, yet differential, effects of
1 were observed for CD43 glycoforms in multiple hematopoietic cells. Development of small molecules that could alter glycan
patterns in an antigen-selective and cell-type selective manner might provide avenues for understanding biological functions of
glycans.
INTRODUCTION
Investigation of structure and function of O-linked glycans
present additional challenges compared to N-linked glycans,
due to presence of multiple initiating monosaccharide building
blocks and variety of core structures coupled with lack of con-
sensus sequons, specific biochemical tools, and inhibitors.1
Among various types of glycosylations, mucin-type O-
glycosylation constitutes one of the most abundant on mam-
malian cell surface.2,3 Biosynthesis of mucin-type O-glycans is
initiated by the addition of GalNAc to Ser/Thr by UDP-
Particularly, metabolic glycan engineering (MGE) meth-
odology has been employed successfully for expression of
non-natural carbohydrates on glycans exploiting the permis-
sivity of glycosylation and salvage pathways (e.g. sialylation,
fucosylation, mucin-type O-glycosylation, and β-O-GlcNAc)
using peracetylated monosaccharide precursors.15-18 Peracety-
lated N-azidoacetyl-D-galactosamine (Ac4GalNAz) and N-
glycolyl-D-galactosamine (Ac5GalNGc) are readily metabo-
lized through GalNAc salvage pathway and incorporated in to
mucin-type O-glycans both in vitro and in vivo.19-21 Peracety-
lated non-natural GalNAc analogs (GalNR) are first de-
acetylated intracellularly, converted to UDP-GalNR via
GalNR-1-P, and then transported to ER / Golgi for utilization
by ppGalNAcTs.19
GalNAc:polypeptide
N-acetyl-galactosaminyl-transferases
(ppGalNAcT).4-6 α-O-GalNAc-S/T (Tn-antigen) is elaborated
by addition of Gal, GlcNAc, Fuc, and NeuAc to yield a variety
of complex glycan structures exquisitely orchestrated by a
battery of respective glycosyl transferases. Multiple ap-
proaches could be envisaged for elucidation of glycan depend-
ent functions each with unique merits and limitations: (i) gly-
coprotein specific knock out models lead to loss of both pro-
tein and its glycans;7 (ii) knock-outs of specific glycosyl trans-
ferases results in simultaneous perturbation of multiple protein
substrates;8,9 (iii) efficient O-glycan inhibitors, despite the
drug development potential,10 have been few and far between
with most compounds focused mostly on substrate decoys and
uridine analogs;11-13 and (iv) installation of monosaccharide
building blocks carrying non-natural functional groups ena-
bling bio-orthogonal ligation.14
On one hand, MGE has been exploited for direct incorpo-
ration and imaging of glycans.17 On the other hand, depending
on the specific chemical structure, metabolism of non-natural
analogs is likely to exhibit altered affinities and reaction rates
for glycosylation pathway enzymes with respect to wild-type
substrates. Kinetic studies have shown that UDP-GalNAz is
utilized at one-third the rate as compared to UDP-GalNAc by
ppGalNAcTs.22 In vivo usage of peracetyl 2-acetamido-2,4-
dideoxy-4-fluoro-D-galactosamine (Ac3-4F-GalNAc) is re-
ported to inhibit elaboration of O-glycans via metabolic incor-
poration in P-selectin glycoprotein ligand -1 (PSGL-1/CD162)
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 17
1
2
3
4
5
6
7
8
9
and modulate leukocyte adhesion.23 Other in vitro studies us-
of 1),26 peracetyl benzyl 2-acetamido-2-deoxy-α-D-
galactopyranoside (Ac3-α-BnGalNAc, 8),27 and non-acetylated
α-Ο-BnGalNAc (9) were employed as controls. Metabolic
processing of 1 via GalNAc salvage pathway (Figure 1B) spe-
cifically induced drastic hypoglycosylation of CD43 (leu-
kosialin / sialophorin) in Jurkat (human T-cell leukemia) cells
as a consequence of direct incorporation. We show for the first
time that depending on the nature of the installed chemical
moiety, non-natural GalNAc analogs drastically affect mucin-
type O-glycosylation. These small molecules act as a first step
in manipulating mucin-type O-glycans and may further serve
as valuable probes in probing the relationship between struc-
ture and function of mucins in biological processes.
ing non-natural UDP-GalNAc analogs as donors and recombi-
nant MUC1 protein as an acceptor have shown that relative
incorporation of non-natural epitopes are governed by both
size and nature of functional groups.24,25 Typically, the trade-
off for metabolic processing of non-natural monosaccharide
analogs is an alteration to the full complement of cellular gly-
cosylation patterns.
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
A
Ac5GalNTGc (1)
Ac5GalNGc (2)
Ac4GalNAc (3)
Ac5GalNTPr (4)
Ac5GalNTBut (5)
R = CH2SAc
R = CH2OAc
R = CH3
R = CH2CH2SAc
R = CH2CH2CH2SAc
AcO
AcO
CH2OAc
O
OAc
HN
O
R
Ac4GalNMeTA (6) R = CH2SCH3
RESULTS AND DISCUSSION
CH2OAc
O
HO
HO
AcO
AcO
CH2OAc
O
CH2OH
AcO
AcO
OAc
O
Metabolic effects of non-natural monosaccharide analogs
on glycosylation may be broadly classified in to three catego-
ries, (i) incorporation with minimal effects on overall glycan
structures,15,20,28 (ii) altered glycan structures with no incorpo-
ration,29-32 and (iii) altered glycan structures as a consequence
of metabolic incorporation.23 Due to the specific chemical
properties of the sulfhydryl group, metabolism of 1 was an-
ticipated to impact mucin-type O-glycan biosynthesis in a
unique manner. We chose Jurkat cells as a model, as earlier
studies have shown that 1 is metabolized readily resulting in a
2.5 fold increase in accessible cell surface thiol (CST) levels.26
The oxa-analog 2 and analog 3 (a precursor for ‘natural’ Gal-
NAc) were used as controls for effects due to intracellular
deacetylation and altered metabolic flux in the GalNAc sal-
vage and mucin-type O-glycosylation pathways.
HN
O
AcHN
AcHN
OBn
OBn
α
-O-BnGalNAc (9)
CH2SAc
α
Ac5GlcNTGc (7)
Ac3- -O-BnGalNAc (8)
B
AcO
HO
HO
HO
HO
CH2OAc
O
CH2OH
CH2OH
a
b
O
O
AcO
OAc
OH
HN
O
HN
HN
2-
OPO3
O
O
CH2SAc
CH2SH
GalNTGc
CH2SH
-GalNTGc-1-P
α
1
HO
HO
HO
CH2OH
CH2OH
Ser R = H
c
O
d
O
Thr R = CH3
O
HN
R
HO
HN
HN
O
O
UDP
O
O
CH2SH
CH2SH
Disrupting the sugar coat: Metabolic processing of 1,
but not 2, abrogates MAL-II and PNA epitopes in Jurkat
cells. In order to investigate whether GalNAc analogs affected
overall cell surface glycosylation, we employed lectin stain-
ing.33 Three lectins, namely, Maackia amurensis lectin - II
(MAL-II) (binds to NeuAcα(2→3)Gal), Sambucus nigra ag-
glutinin (SNA) (binds to NeuAcα(2→6)Gal/GalNAc), and
peanut agglutinin (PNA) (binds to Galβ(1→3)GalNAc-S/T,
also known as T-antigen and Thomson-Friedenreich (TF)-
antigen) were used. Jurkat cells were incubated with dimethyl
sulfoxide (DMSO, D) (vehicle), 2, or 1 (100 μM, 48 h),
treated without and with neuraminidase (Neu), stained with
biotinylated-lectins followed by AlexaFluor594-conjugated
streptavidin, and studied by confocal imaging. Remarkably,
binding of MAL-II was abrogated upon incubation with 1, but
not D or 2, indicating thiol-specific effects. Expectedly, treat-
ment with Neu resulted in loss of both MAL-II (Figure 2A)
and SNA binding (SI Figure S1). SNA staining remained unaf-
fected upon incubation with D, 2, or 1 (SI Figure S1). No
binding of peanut agglutinin (PNA) was detected on cells
treated with D, 2, or 1, as PNA epitopes are known to be
masked by terminal sialylation. While cells treated with D or 2
revealed PNA epitopes after Neu treatment, strikingly, cells
treated with 1 showed no PNA binding (Figure 2B). Absence
of PNA binding in cells incubated with 1 even after Neu
treatment, suggested thiol-specific abrogation of T-antigens.
α
α
-UDP-GalNTGc
-O-GalNTGc-S/T
Mucin-type -glycans
O
Figure 1. Metabolic processing through GalNAc salvage path-
way. (A) Structures of peracetylated GalNAc analogs 1 – 8 and 9.
(B) Biosynthetic processing of 1 by (a) intracellular esterases, (b)
GalNAc-1-kinase, (c) UDP-GalNAc pyrophosphorylase, and (d)
polypeptide N-acetylgalactosaminyltransferases (ppGalNAcT)
results in expression of thiol functionality on Tn-antigen, the initi-
ating monosaccharide for biosynthesis of mucin-type O-glycans.
In this context, we wanted to investigate whether non-
natural GalNAc analogs could be utilized as tools to alter the
overall carbohydrate-loading and pattern of mucin-type O-
glycosylation, exploiting MGE. We surmised that depending
on the chemical nature of the non-natural functional group and
possible utilization of UDP-GalNR (R = acyl) as donors by
ppGalNAcTs might result in direct incorporation, which in
turn, might attenuate subsequent elaboration of O-glycans and
neighbour site decoration. Herein, we show thiol-specific ef-
fects on selective inhibition of O-glycosylation both globally
and in selected lymphocyte cell surface antigens induced by
peracetyl N-thioglycolyl-D-galactosamine (Ac5GalNTGc, 1).26
A panel of peracetylated N-acyl-D-galactosamine derivatives
carrying
N-glycolyl-
(Ac5GalNGc,
2),21
N-acetyl-
(Ac4GalNAc, 3), N-(3-acetylthiopropanoyl)- (Ac5GalNTPr, 4),
N-(4-acetylthio-butanoyl)- (Ac5GalNTBut, 5), and N-
methylthioacetyl- (Ac4GalNMeTA, 6) were synthesized (Fig-
ure 1A) (Supporting information (SI)). Additionally, peracetyl
N-thioglycolyl-D-glucosamine (Ac5GlcNTGc, 7 – C-4 epimer
2
ACS Paragon Plus Environment

Page 3 of 17
Journal of the American Chemical Society
1
2
3
and subjected to Neu treatment (Figure 2D). PNA binding was
4
5
6
7
8
9
similar in all cases (100 % for untreated; 88 % and 95 % for
cells treated with D and 3, respectively) except in cells treated
with 1 (16 % of control) which suggested inhibition of T-
antigen structures on cell surface by 1. PNA binding to cells
incubated with no vehicle (U), D, 3, or 1, that were not sub-
jected to Neu treatment showed a similar pattern (SI Figure
S2). We tested the possibility that abrogation of lectin binding
could be a consequence of non-natural disulfides formed due
to expression GalNTGc. Jurkat cells incubated with and with-
out 1 (100 μM, 48 h) followed by treatment with tris-(2-
carboxyethyl)phosphine hydrochloride (TCEP) and maleim-
ide-PEG2-biotin (MB) showed no change in levels of PNA
binding confirming that reduction in PNA binding induced by
1 is indeed a consequence of O-glycan inhibition and ruled out
disulfide mediated changes to PNA binding (SI Figure S2).
Specificity of PNA binding to T-antigen epitopes was con-
firmed by titration with its competitive ligand lactose (SI Fig-
ure S2).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Jurkat cells are generally considered to express low levels
of elaborated O-glycans relative to K562 (human chronic mye-
loid leukemia)36 and HL-60 (human acute myeloid
leukemia).35 This is attributed to (i) a mutation in cosmc (core1
beta3-Gal-T-specific molecular chaperone) – a chaperone es-
sential for proper folding and activity of T-synthase (glycopro-
Figure 2. Thiol-dependent inhibition of expression of MAL-II
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
and PNA epitopes in Jurkat cells induced specifically by 1. Con-
focal imaging of cells incubated with D, 2, or 1 (100 μM, 48 h),
treated without and with Neu, fixed and stained with biotinylated
lectins, (A) MAL-II and (B) PNA followed by AlexaFluor-594-
conjugated streptavidin (red) and DAPI (blue). Images shown are
representative of multiple field views of two replicate samples;
scale bar = 50 μm. (C) MAL-II lectin blots of total lysates of cells
incubated with D, 3, 2, or 1 (100 μM, 60 h) (left panel) and cells
incubated with D or 1 (100 μM, 48 h) followed by treatment with
and without Neu (right panel); β-actin blots are shown as loading
controls. (D) Flow cytometry estimation of PNA binding in cells
incubated with no vehicle (U), D, 3, or 1 (50 μM, 48 h) and sub-
jected to Neu treatment. MFI values are shown on the graph. Error
bars shown are standard deviation of two replicate samples. D,
DMSO; Neu, neuraminidase; U, untreated; MAL-II, Maackia
amurensis lectin – II, PNA, peanut agglutinin, MFI in a.u., mean
fluorescence intensity in arbitrary units.
tein-N-acetylgalactosamine
3-β-galactosyltransferase,
C1GALT)37 and (ii) absence of branched core2 glycans.34 Sur-
prisingly, Jurkat cells (untreated) showed a 13-fold gain in
PNA binding upon Neu treatment (mean fluorescence intensity
(MFI) of 60 arbitrary units (a.u.) without Neu and 776 a.u.
with Neu) (SI Figure S2). Significant PNA binding observed
by both confocal imaging and flow cytometry indicated that
Jurkat cells, nonetheless, do express substantial levels of T-
antigen structures.
In order to test if the thiol-specific effects observed with 1
were caused by cytotoxicity, a viability assay using propidium
iodide (PI) and annexin-V was performed (SI Figure S3). Jur-
kat cells treated with D, 2, and 1, were found to be 90 %, 83
%, and 79 % viable respectively, after incubation for 48 h at
200 μM, indicating that thiol-specific effects are not a result of
compound toxicity. It is noteworthy that abrogation of MAL-II
and PNA epitopes was observed at 100 μM suggesting that 1
inhibits O-glycan elaboration efficiently with minimal cyto-
toxicity.
Lectin blots of total lysates of Jurkat cells incubated with
D, 1, 2, or 3, clearly showed thiol-specific inhibition of MAL-
II epitopes (Figure 2C). MAL-II binding was observed as
prominent bands between 150 – 100 kDa in all cases except in
cells treated with 1. Interestingly the bands at 150-100 kDa
coincide with apparent molecular weights of CD43 and
CD162/PSGL-1, two major cells surface antigens on T-cells
that are rich in O-glycans.34,35 Neu treatment of cells incubated
with D prior to lysis clearly showed loss of MAL-II epitopes;
whereas, MAL-II epitopes were absent in cells treated with 1
even without Neu treatment (Figure 2C). Notably, effects ob-
served in MAL-II blots complement microscopy results and
confirm absence of sialyl-T-antigen structures induced specifi-
cally by 1.
Jurkat cells carry substantial levels of elaborated O-
glycans whose expression is specifically inhibited by 1. For
non-lectin based evidence, we isolated O-glycans from Jurkat
cells and characterized by MALDI-TOF/TOF mass spectrome-
try.38 Mass spectra of permethylated O-glycans from untreated
Jurkat cells clearly showed peaks at m/z 691.29, 895.34 and
1256.46 corresponding, respectively, to sialyl-Tn (disaccha-
ride), sialyl-T (trisaccharide), and disialyl-T (tetrasaccharide)
antigen structures (Figure 3A). Glycan structures were further
confirmed by MS/MS on precursor ion m/z 895.34 which
yielded daughter ions at m/z 398.127 (B-ion, sialyl), 520.226
(Y-ion, T-antigen), and 620.257 (C-ion, sialyl-Gal) daughter
ions (Figure 3B). MS/MS of precursor ion m/z 1256.46
yielded expected daughter ions at m/z 398.15 (B-ion, sialyl),
Corroboration of thiol-specific inhibition of T-antigen was
obtained by flow cytometry estimation of PNA binding to
Jurkat cells incubated with no vehicle (U), D, 1, or 3 for 48 h
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 17
1
2
3
506.21(YY-ion, T-antigen), 620.20 (C-ion, sialyl-Gal), 659.31
thiol-dependent global inhibition of elaboration of O-glycans
4
5
6
7
8
9
(Z-ion, sialyl-Tn), and 881.405 (Y-ion, sialyl-T – linear /
branched) (Figure 3C). It is pertinent to emphasize here that
these structures have earlier been identified and characterized
in CD43 isolated from Jurkat cells through radioactive label-
ing and chromatography.34 Thus, based on MAL-II/PNA bind-
ing and mass spectrometry evidence, it can be concluded that
Jurkat cells do express significant levels of elaborated O-
glycans (notwithstanding cosmc mutation and reduced T-
synthase activity).
induced by 1, we focused on changes to glycoforms of se-
lected cell surface antigens of immunological importance.
CD43 was chosen as a model for mucin-type O-glycoproteins
for the following reasons: (a) CD43 is heavily O-glycosylated
with ~80-90 O-glycans (contributing 50-60% of total molecu-
lar weight) resulting in a bottle-brush like structure for the
extracellular domain,39,40 (b) CD43 is known to be both a
negative regulator and co-receptor for immune activation,7 and
(c) glycoform-specific anti-CD43 antibody reagents were
readily available. Jurkat cells express CD43 abundantly and its
O-glycans have been well characterized.34,41 Three types of
anti-CD43 antibodies, namely, Neu-sensitive (clones CD43-
L60 and CD43-1G10), Neu-resistant (clone CD43-L10), and
glycan-independent (clone CD43-C20, C-terminal) were em-
ployed (Figure 4A).42 Cell surface estimation of CD43 epi-
topes was performed under saturating antibody conditions (SI
Figure S4). Neu-sensitive and Neu-resistant binding of CD43
antibodies were verified by flow cytometry (SI Figure S5).
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
Flow cytometry revealed that Jurkat cells do express abun-
dant levels of CD43-L60 epitopes which was reduced to 26 %
of controls when incubated with 1, but not 2 or vehicle (etha-
nol, E), at a low concentration of 10 μM for 48 h (MFI values
of 237 and 218 a.u., respectively, for E and 2 compared to 61
a.u. for 1) (Figure 4B). A dose dependency study (0 – 50 μM,
48 h) with 1 showed drastic reduction in CD43-1G10 (15 % of
control at 50 μM) and significant reduction in CD43-L10 (50
% of control at 50 μM) levels; in contrast, at 50 μM, both
CD43-1G10 (95 % of control) and CD43-L10 (90 % of con-
trol) levels remained unaffected in cells treated with 2, indicat-
ing thiol-specific interference in cell surface expression of
CD43 glycoforms (Figure 4C).
In order to investigate changes to overall CD43 glyco-
forms western blotting was performed on total lysates from
Jurkat cells incubated with no vehicle (untreated), D, 2, or 1
(100 μM, 48 h) (Figure 4D). Specifically, treatment with 1
induced drastic reduction of CD43-L60 epitopes which ap-
peared as sharp bands at 125 kDa in controls (U, D, and 2).
Similarly, significant reduction of CD43-L10 epitopes (along
with appearance of multiple bands between 125-100 kDa) was
induced specifically by 1, but not U, D, or 2, in agreement
with flow cytometry results (Figure 4C). Strikingly, CD43-
C20 blots revealed appearance of lower molecular weight
bands in the range of 125 – 75 kDa, only in cells treated with
1, but not U, D, or 2. Clearly, treatment with 1 resulted in both
hypo-sialylation (CD43-L60 and CD43-1G10) and hypo-
glycosylation (CD43-L10 and CD43-C20) of CD43 in a thiol-
dependent manner. Further, treatment with analogs 4 and 5 –
carrying acetylthio-moiety with longer chain lengths – did not
have the same effect as 1 on CD43 glycoforms indicating
steric limitations on the thiol-specific effects (Figure 4E). A
qPCR study of Jurkat cells incubated with D, 2 (200 μM), or 1
(10-200 μM) for 48 h revealed that CD43 mRNA levels re-
mained unaffected (SI Figure S6). This confirmed that 1 inter-
fered post-translationally in the glycosylation of CD43 result-
ing in appearance of multiple micro-heterogeneous glycoforms
with lower molecular weights.
Figure 3. MALDI-TOF/TOF spectra revealed presence of elabo-
rated O-glycans in Jurkat cells and their suppression by 1. Mass
spectra of permethylated O-glycans from 35 % acetonitrile in
water fractions isolated from cells treated with (A) no vehicle,
(D) D, and (E) 1 (100 μM, 48 h). MS/MS spectra of selected pre-
cursor ions from untreated cells corresponding to (B) sialyl-T and
(C) disialyl-T structures. Relevant m/z peaks are underlined in
red. Spectra shown are representative of at least two replicate
samples. Monosaccharide symbols shown are as per the Consorti-
um for Functional Glycomics (CFG) guidelines: Gal, yellow cir-
cle; GalNAc, yellow square; NeuAc, pink diamond.
A qualitative analysis of mass spectra revealed that cells
treated with D showed O-glycan peaks similar to untreated
cells (Figure 3D). However, cells incubated with 1 (100 μM,
48 h) showed absence of disialyl-T (m/z 1256. 46) and pres-
ence of very low levels of sialyl-T (m/z 895.34) O-glycans
(Figure 3E), supporting abrogation of MAL-II epitopes. No
evidence for presence of GalNTGc could be obtained in pre-
sumably due to levels of metabolic incorporation below detec-
tion limit.
Ac5GalNTGc (1) drastically alters glycoforms of CD43
in a dose and time dependent manner. Having confirmed
4
ACS Paragon Plus Environment

Page 5 of 17
Journal of the American Chemical Society
1
2
3
natural metabolites with time, and restoration of wild-type
4
5
6
7
8
9
glycosylation patterns. It is pertinent to emphasize here that
effects observed with 1 in Jurkat cells are under competitive
conditions wherein both GALE activity and UDP-GalNAc
levels appear to be normal.43 This is in contrast to earlier stud-
ies with 2 utilizing the mutant CHO-ldlD cell line that does
not biosynthesize UDP-GalNAc endogenously.21
10
11
12
13
14
15
16
17
18
19
20
21
Ac5GalNTGc (1) alters glycoforms selectively in an an-
tigen dependent manner. At the cellular level, apart from the
metabolic processing of 1 and possible utilization of its me-
tabolite UDP-GalNTGc as a donor by ppGalNAcTs, effects of
1 are likely to be governed by the nature of polypeptide accep-
tor substrates as well. This is likely to result in global, yet,
selective effects of metabolic engineering. Thus for example
CD45, another abundant protein on T-cell surface, is known to
carry only 8 – 17 O-glycans as opposed to 80-90 O-glycans on
CD43.41 To study susceptibility of other cluster of differentia-
tion (CD) antigens to MGE, effect of 1 on glycoforms of
CD45 was probed using Neu-sensitive (clone CD45-UCHL-1),
Neu-resistant (clone CD45-4B2), and glycan-independent (C-
terminal, CD45-B8) antibodies. Incubation of Jurkat cells with
1 (10 μM, 48 h) caused significant reduction in CD45-UCHL-
1 epitopes (MFI of 12 and 12.5 a.u for D and 2 respectively
compared to 5.1 a.u. for 1) (Figure 4G) while CD45-4B2 epi-
topes remained unchanged (MFI values of 375, 387, and 376
respectively for D, 2, and 1) (Figure 4H). Western blots using
CD45-B8 revealed no significant change in the bands at 220
kDa upon treatment either with D, 2, or 1 (100 μM, 48 h)
(Figure 4I) in contrast to that observed with CD43-C20 (Fig-
ure 4D). It is quite possible that effects of 1 on CD45 are not
reflected in western blots due to counteracting factors of hypo-
sialylation and hypo-glycosylation. Also, effect of 1 on O-
glycans might be offset by a relatively large number of N-
glycans present on CD45. By contrast, CD43 is known to
carry only a single N-glycan site proximal to plasma mem-
brane. These results illustrate global, yet, selective effects of
metabolic engineering as a function of nature of cellular gly-
coprotein substrates.
Figure 4. Glycoforms of CD43 and CD45 were altered by 1 in a
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
thiol-specific and antigen dependent manner in Jurkat cells. (A)
Schematic representation of a CD43 glycoform showing binding
epitopes of Neu-sensitive (clones 1G10 and L60), Neu-resistant
(clone L10), and glycan-independent (C-terminal, clone C20)
anti-CD43 antibodies. (B) Anti-CD43-L60 binding to cells after
incubation with ethanol (E), 2 or 1 (10 μM, 48 h). (C) Dose de-
pendency of anti-CD43 (1G10 and L10) binding to cells treated
with 1 or 2 (0 – 50 μM, 48 h) as measured by flow cytometry. (D)
Anti-CD43 (L60, L10, and C20) western blots of lysates of cells
incubated with no vehicle (U), D, 2, or 1 (100 μM, 48 h). (E)
Anti-CD43 (L60 and C20) blots of cells incubated with D, 1, 4, or
5 (100 μM, 48 h). (F) Time course of CD43-L60 expression in
lysates of cells treated with 1 (50 μM) for 48 h followed by treat-
ment with D, 1, or 3 (50 μM) at 48 h. Expression of CD45 (G)
UCHL-1 (Neu-sensitive) and (H) 4B2 (Neu-resistant) epitopes in
cells incubated with D, 2, or 1 (10 μM, 48 h). (I) Anti-CD45 (B8,
C-terminal, glycan-independent) western blots of lysates of cells
incubated with D, 2, or 1 (100 μM, 48 h). MFI values in arbitrary
units (a.u.) are shown in parenthesis on histograms. Error bars
shown are standard deviation of two replicate samples. Western
blots shown are representative of at least two replicates with β-
actin shown as loading control. Neu, neuraminidase; U, no anti-
body control; IC, isotype control; MFI, mean fluorescence intensi-
ty.
Ac5GalNTGc (1) causes drastic hypo-glycosylation of
CD43 through metabolic incorporation. Having confirmed
global – yet antigen-selective – thiol-dependent inhibition of
O-glycan elaboration, we investigated whether GalNTGc was
metabolically installed on CD43. A bioinformatics search re-
vealed that CD43 was ideally suited for probing incorporation
of GalNTGc via thiol-selective Michael addition reaction as it
lacks ‘cysteine’ in its polypeptide backbone. Therefore, sus-
ceptibility to thiol-selective ligation reactions could arise only
due to presence of GalNTGc. Indirectly, formation of UDP-
GalNTGc from 1 and its ready utilization by ppGalNAcT was
evident from earlier studies that showed an increase in CST
levels.26 In order to obtain direct evidence, we stably ex-
pressed CD43 carrying myc/FLAG tags at the C-terminus in
Jurkat-E6.1 cells to facilitate immunoprecipitation in a glycan-
independent manner (SI). CD43-L60 and anti-FLAG western
blots confirmed that thiol-specific effects of 1 were consistent
in Jurkat-E6.1-CD43-myc/FLAG cells (Figure 5A) as ob-
served with Jurkat cells (Figure 4D). Anti-FLAG blots showed
characteristic bands at lower molecular weights (125 – 75
kDa) similar to CD43-C20. Cells were incubated with either D
A time course study of CD43-L60 epitopes by western
blotting confirmed differential effects of 1 compared to 2 (SI
Figure S7). To study reversibility of effects of 1 on CD43 gly-
cosylation, Jurkat cells were first incubated with D or 1 (50
μM). After 48 h, cells were treated with D, another dose of 1
(50 μM), or 3 (50 μM). Cells were harvested at 48, 72, and 96
h and probed by western blotting (Figure 4F). Cells treated
with a single dose of 1 exhibited reduction in CD43-L60 levels
up to 72 h but showed significant recovery by 96 h; whereas,
cells treated with two doses of 1 showed sustained reduction in
CD43-L60 levels at 96 h. In the case of cells treated with 1 (at
0 h) followed by 3 (at 48 h) recovery of CD43-L60 levels was
faster at 72 h and showed near-complete recovery by 96 h.
Sustained suppression and recovery of CD43-L60 epitopes –
using both non-natural and natural precursors – illustrate the
facile flexibility of metabolic incorporation, dilution of non-
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 17
1
2
3
4
5
6
7
8
9
(SI Figure S10). Expectedly, Neu treatment increased binding
or 1 (100 μM, 48 h) and CD43-myc/FLAG was immunopre-
cipitated (SI Figure S8). Non-transfected and vector-
transfected Jurkat-E6.1 cells were used as controls. CD43-
myc/FLAG immunoprecipitates on-beads were subjected to
reduction by tris-(2-carboxyethyl)phosphine hydrochloride
(TCEP) followed by thiol-selective Michael addition using
maleimide-PEG2-biotin (MB) and analyzed by western blot-
ting (Figure 5B).44 Strikingly, HRP-avidin blot revealed the
presence of biotin exclusively on CD43-myc/FLAG immuno-
precipitated from cells incubated with 1, but not in controls.
Anti-FLAG blot showed sharp bands at 125 kDa for vehicle-
treated cells while cells treated with 1 showed characteristic
diffuse bands from 125 – 75 kDa (a cross-reactive protein
band at 125 kDa observed in silver stains and anti-FLAG blots
was considered to be non-specific as it eluted from non-
transfected cells as well (SI Figure S8)). Notably, thiol-
selective biotinylation was dependent upon prior reduction by
TCEP indicating that most of the GalNTGc were spontane-
ously oxidized to disulfide forms on cell surface18 and possibly
during sample preparation (SI Figure S9). Thus, thiol-selective
biotinylation of CD43-myc/FLAG confirmed direct metabolic
incorporation of 1 as GalNTGc.
of VVA and HPA in all three cases owing to increased acces-
sibility of Tn structures. Therefore, based on lectin blots of
CD43-myc/FLAG it might be concluded that metabolic incor-
poration of 1 does not seem to affect addition of initiating
GalNAc residues to other S/T sites on CD43, but clearly abro-
gates elaboration by T-synthase and sialyltransferases.
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
Interestingly, ppGalNAcTs possess a ricin-like lectin do-
main that helps in anchoring a pre-existing GalNAc. Relative
distance between active site and lectin domain is thought to
determine the choice of neighbour site for addition of
GalNAc.45 Also, ppGalNAcTs prefer glycopeptides over pep-
tides as substrates and exhibit unique site selectivities.6,46
Speculatively, it is plausible that metabolic incorporation of 1
as GalNTGc on mucin-type O-glycoproteins modulates bind-
ing to lectin domain of ppGalNAcTs. This in turn might affect
subsequent decoration of adjacent sites resulting in reduced
site occupancy reflected in micro-heterogeneous lower mo-
lecular weight glycoforms. It is also known that ppGalNAcTs
exhibit two free Cys side chains that form hydrogen bonds to
UDP-GalNAc.47 The sulfhydryl groups at the donor binding
site could potentially interact with UDP-GalNTGc through
red-ox mechanisms – possibly in an isoform selective manner
– thus modulating kinetics of ppGalNAcT and glycosylation
of their respective polypeptide acceptor substrates.
Both sulfhydryl and galactosamine moieties are critical
for mucin-type O-glycan inhibition. Next, we explored the
mechanistic significance of galactosamine and thiol moieties
for inhibition of O-glycosylation on CD43. Flow cytometry
revealed that Ac4GalNMeTA (6) – wherein sulfur is capped
with a stable methyl group – did not affect CD43-1G10 epi-
topes on Jurkat cells even at 100 μM, while 1 induced a drastic
reduction at 10 μM (Figure 6A). Lack of activity of 6 indi-
cated that intracellular hydrolysis of acetylthio- group to gen-
erate thiol from 1 is critical for inhibition. Similarly,
Ac5GlcNTGc (7) – the C-4 epimer of 1 – did not alter CD43-
1G10 levels even at 100 μM (Figure 6A) suggesting that
GlcNTGc is not readily converted to GalNTGc. Recent studies
have shown that GALE could convert both GalNGc21 and
GalNAz43 to GlcNGc and GlcNAz, respectively. However,
efficiency of the reverse reaction is not known. Absence of
activity with 7 suggests that GALE might not act upon
GlcNTGc.
Figure 5. Abrogation of MAL-II and PNA epitopes on CD43
specifically through metabolic incorporation of 1. (A) Anti-CD43-
L60 and anti-FLAG western blots of lysates of CD43-myc/FLAG
stably transfected Jurkat-E6.1 cells incubated with D, 2, or 1 (100
μM, 48 h). (B) Anti-FLAG and HRP-avidin blots of CD43-
myc/FLAG immunoprecipitates from cells treated with D or 1
(100 μM, 48 h) subjected to TCEP treatment followed by MB. (C)
Lectin blots using MAL-II and PNA of CD43-myc/FLAG
immunoprecipitates from cells incubated with D, 2, or 1 (100 μM,
48 h). For PNA, immunoprecipitates were subjected to Neu treat-
ment prior to electrophoresis. β-actin and anti-FLAG blots are
shown as loading controls. Blots are representative of at least two
replicates. Neu, neuraminidase; N, non-transfected; V, vector
control; TCEP, tris(2-carboxyethyl)phosphine hydrochloride; MB,
maleimide-PEG2-biotin.
Additional proof of inhibition of elaboration of O-glycans
on CD43 induced by 1 was obtained using lectin blots. CD43-
myc/FLAG immunoprecipitates revealed absence of MAL-II
and PNA epitopes on CD43 upon treatment with 1, but not D
or 2 (Figure 5C), consistent with global effects observed by
confocal microscopy, western blots, and flow cytometry (Fig-
ure 2A, 2B, and 2C). Importantly, positive binding of MAL-II
and PNA to CD43 from cells treated with D or 2 is in agree-
ment with earlier work by Fukuda and others wherein CD43
isolated from Jurkat cells were found to carry both linear and
branched sialyl-T and disialyl-T structures.34 Probing for effect
of D, 2, or 1 on Tn-epitopes on CD43-myc/FLAG immuno-
precipitates using Vivia villosa agglutinin (VVA) and Helix
pomatia agglutinin (HPA) revealed no major changes in bind-
ing, except for the lower molecular weight bands induced by 1
A strategy for inhibition of mucin-type O-glycan biosyn-
thesis is through employment of substrate-decoys that compete
with glycoprotein substrates for glycosyl transferases.11,12,48
A
comparative flow cytometry study revealed that unlike 1, per-
acetylated O-Bn-α-GalNAc (8) did not change the levels of
both CD43-1G10 and CD43-L10 epitopes. At 2.0 mM, the
non-acetylated O-Bn-α-GalNAc (9) showed moderate reduc-
tion of CD43-1G10 and no change in CD43-L10 epitopes (SI
Figure S11). This confirmed that 1 does not act by a substrate-
decoy mechanism (Figure 6B). It is noteworthy that substrate-
decoys could affect only downstream of Tn-antigen; whereas,
1 could potentially affect the biosynthesis of Tn-antigen by
modulating ppGalNAcT action.
6
ACS Paragon Plus Environment

Page 7 of 17
Journal of the American Chemical Society
1
2
3
and 50 % of untreated controls, respectively, in Jurkat, THP-1
4
5
6
7
8
9
10
11
12
(human acute monocytic leukemia), K562, and HL60 cells
(Figure 7A). Cells treated with D did not show significant
changes compared to untreated controls. Western blots using
both anti-CD43-L60 and anti-CD43-C20 for vehicle treated
cells showed bands at 125 kDa for Jurkat and K562, while
THP-1 and HL-60 were observed at 130 kDa highlighting
inherent variations among cell lines.49 Treatment with 1 in-
duced abrogation of CD43-L60 in Jurkat and THP-1, mild
reduction in K562, and mild reduction with lower apparent
molecular weights in HL60 cells. Corresponding anti-CD43-
C20 blots showed lower molecular weight bands maximally
for Jurkat, moderately for THP-1 and HL-60, and minimally
for K562 cells induced by 1 (Figure 7B).
Figure 6. Both sulfhydryl and galactosamine moiety are critical
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
for efficient inhibition of O-glycosylation of CD43 in Jurkat cells
by 1. (A) Flow cytometry estimation of anti-CD43-1G10 binding
to cells incubated with 1, 6, or 7 (0 – 100 μM, 48 h). (B) anti-
CD43 (L10 and 1G10) binding to cells incubated with D, 8, or 1
(100 μM, 48 h). (C) Anti-CD43-1G10 binding to cells incubated
with D or 1 (100 μM, 48 h) followed by treatment with TCEP and
MB. Error bars shown are standard deviation of two replicate
samples. D, DMSO; Neu, neuraminidase, TCEP, tris(2-
carboxyethyl)phosphine hydrochloride; MB, maleimide-PEG2-
biotin.
A possibility that lower molecular weight bands observed
in anti-CD43-C20 blots could be a result of ‘hyper-sialylation’
was ruled out as follows: Upon Neu treatment prior to lysis, as
expected, CD43-L60 epitopes were reduced for D and 2, and
abrogated for 1. In CD43-C20 blots, both D and 2 showed
characteristic bands at a higher apparent molecular weight due
to loss of negatively charged sialic acids (SI Figure S12). By
contrast, cells incubated with 1 showed no change upon Neu
treatment confirming severe hypoglycosylation of CD43.
Figure 7. Thiol-dependent effects on CD43 glycoforms induced
by 1 in various human cell lines. (A) Flow cytometry estimation
of anti-CD43-L60 binding in Jurkat, THP-1, K562, and HL-60
cells incubated with no vehicle (U), D, or 1 (100 μM, 48 h). (B)
Anti-CD43 (L60 and C20) western blots of lysates of cells incu-
bated with D or 1 (100 μM, 48 h). Error bars shown are standard
deviation of two replicate samples. Western blots are representa-
tive of at least two replicates with β-actin shown as loading con-
trol. D, DMSO.
In principle, binding of antibodies to cell surface epitopes
could be altered due to oligomerization of glycoproteins medi-
ated by disulfides involving GalNTGc. To test if anti-CD43
binding are affected by MGE, Jurkat cells were incubated with
D or 1 (100 μM, 48 h), followed by reduction of cell surface
disulfides using TCEP, and capping of thiols using MB (Fig-
ure 6C). CD43-1G10 binding remained unaffected even after
reduction and capping confirming that the observed thiol-
specific effects are indeed due to hypoglycosylation of CD43.
Similar results were observed for PNA binding as well (SI
Figure S2).
Differential effects of 1 and a threshold requirement for
modulation of glycoforms among hematopoietic cell lines
could be attributed to inherent load of glycans on CD43. For
instance, mucin-type O-glycans of CD43 isolated from Jurkat
cells carry ~ 83 % of GalNAc and ~ 17 % of elaborated O-
glycans34 which is almost reversed in the case of CD43 from
K562 cells, i.e. ~ 5 % of GalNAc and ~ 95 % of elaborated O-
glycans.36 Moderate effects of 1 on CD43-L60 in K562 com-
pared to acute effects in Jurkat could, at least in part, be ex-
plained by robust T-synthase activity in K562 which is defi-
cient in Jurkat.34 Global effects at a cellular level would also
be governed by differential expression, activity, and prefer-
ence (or the lack thereof) for UDP-GalNTGc by the fourteen
known human isoforms of ppGalNAcTs in various cell lines.
Differential effects of 1 open up an interesting possibility of
exploiting cell-type selective effects to modulate migration,
trafficking, and behavior of leukocytes via modulation of gly-
coforms of cell surface antigens under disease conditions.
CD43 glycoforms are modulated differentially by 1 in
hematopoietic cells. Effects of metabolic processing of non-
natural hexosamine analogs on cell surface glycoforms are
governed by (a) specific chemical structure and reactivity of
the non-natural moiety, (b) affinity of glycosyl transferases to
activated non-natural sugar nucleotide donors, (c) expression
and activity of multiple isoforms of each type of glycosyl
transferases, and (d) nature and properties of acceptor poly-
peptide substrates. Expression of glycosylation machinery
does vary widely among various cell types and also within a
given cell type depending on whether it is activated or not.49,50
Hence, in order to test if effects observed with 1 are general,
we studied a panel of hematopoietic cell lines using changes to
CD43 glycoforms as a read-out sensor. Flow cytometry re-
vealed that treatment with 1 reduced binding of anti-CD43-
L60 in all four cell lines indicating hypo-sialylation of CD43,
albeit to different extents. Upon incubation with 1 (100 μM,
48 h), CD43-L60 levels were reduced to 11 %, 31 %, 42 %
CONCLUSIONS
Our studies clearly show that effects of metabolic process-
ing of non-natural GalNAc analogs in mammalian cells are
governed by the chemical structure of functional groups and
the nature of cellular protein substrates. Expression of MAL-II
and PNA epitopes on Jurkat cells were abrogated in a thiol-
specific manner by 1. Glycoform-specific antibodies revealed
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 17
1
2
3
4
5
6
7
8
9
drastic hypoglycosylation of CD43 induced specifically by 1,
22 °C), followed by washing with PBS (3 × 1 mL). For siali-
but not 2 - 9. Both thiol and galactosamine moieties of 1 were
found to be critical for activity. Metabolic incorporation of 1
was established via thiol-selective biotinylation of CD43-
myc/FLAG. Lectin blots of CD43 immunoprecipitates showed
absence of both MAL-II and PNA epitopes while HPA/VVA
epitopes were unaffected. Glycoforms of CD43 and CD45
were altered differentially by 1 and found to be a function of
polypeptide backbone and glycan load. Differential sensitivity
of hematopoietic cells to 1 could be attributed to inherent dif-
ferences in glycan load and glycosylation machinery. Consid-
ering that cell surface glycans, particularly on CD43, govern
leukocyte trafficking in both immunity and auto-immune dis-
eases,51 it would be interesting to test if effects observed with
1 are translated in vivo. Ability to modulate glycan patterns in
a selective set of immune cells using carbohydrate-based small
molecules may open-up new avenues for understanding bio-
logical functions of glycans in immunity, auto-immune dis-
eases, and cancer.
dase treatment, cells attached to cover slips were incubated
with sialidase (0.005 units per cover slip) from Arthrobacter
ureafaciens in acetate buffer (pH 5.0) for 30 min at 37 °C,
followed by washing with PBS (2 × 1 mL). Attached cells
were blocked with 2.0 % gelatin (Sigma-Aldrich) (30 min, 22
°C), washed with PBS (2 × 1 mL), and incubated with respec-
tive biotinylated lectin (30 min, 22 °C). Lectins were used at a
concentration of 10 μg/mL in PBS for bSNA and bMAL-II;
bPNA was diluted in a buffer containing 10 mM HEPES, pH
7.5, 0.15 M NaCl, 0.1 mM CaCl2 as per suppliers instructions.
Cells were washed with PBS (3 × 1 mL), incubated with Al-
exaFluor-594 streptavidin conjugate (1:1000 dilution of 1.0
mg/mL stock solution) containing DAPI (1.0 μg/mL) (30 min,
22 °C), and washed with PBS (3 × 1 mL). The cover slips
were mounted on microscope slides using ProLong Gold anti-
fade reagent (Life Technologies). Confocal imaging was per-
formed using LSM 510 Meta Laser Scanning Microscope
(Carl Zeiss) on multiple field views and cover slips.
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
EXPERIMENTAL PROCEDURES
Evaluation of PNA binding by flow cytometry: Jurkat
cells were incubated with no vehicle, D, 1, or 3 (50 µM) for 48
h were harvested and washed twice with PBS. Cells (5.0 × 105
cells per sample) were incubated without and with α-(2→3)-
neuraminidase (from Salmonella typhimurium LT2 expressed
in E. coli, 20 units) in 100 μL of PBS at 37 ºC in a water bath.
After 30 min, cells were washed twice with ice-cold lectin
staining buffer (LSB) (10 mM HEPES buffer, containing 150
mM sodium chloride, 0.1 mM calcium chloride, and 0.08 %
sodium azide). Cells were stained using biotin-PNA (5 µg/mL)
in 300 µL LSB for 30 min in ice followed by FITC-avidin
(1:250 dilution) in avidin staining buffer (ASB) (PBS contain-
ing 5 % FBS and 0.1 % sodium azide). After 30 min, cells
were washed in ASB (3 × 0.5 mL), re-suspended in 400 μL of
ASB containing propidium iodide (PI, 2.5 µg / mL) and ana-
lyzed by flow cytometry. Samples were prepared in duplicates
and each sample was counted twice. About 20,000 events
were acquired per sample and PNA binding was analyzed by
gating on PI negative viable cell population.
Chemical synthesis: Peracetylated non-natural GalNAc
analogs 1¸26 2,21, 3, 7,26 and 8,27 were synthesized following
literature. Analogs 4, 5, and 6 were synthesized by analogous
procedures, details of which are available in SI.
Cell culture, antibodies, lectins, enzymes, and reagents.
Standard conditions used for mammalian cell culture are given
in SI. Following anti-human antibodies were employed in this
study: Anti-CD43 (FITC-1G10, FITC-L10, and L60) (BD
Biosciences); goat polyclonal anti-CD43 (C-20) and anti-
CD45 (B-8) (Santa Cruz Biotechnology); anti-CD45 (UCHL-
1), polyclonal anti-FLAG, and anti-β-actin (AC-15) (Sigma-
Aldrich). Purified anti-CD45 (clone 4B2) was a kind gift from
Dr. A. Qadri (NII). Appropriate FITC and HRP-conjugated
secondary and isotype control antibodies were employed.
Normal chicken serum (GIBCO) was a kind gift from Dr. V.
K. Nandicoori (NII). Biotinylated lectins – bSNA, bMAL-II,
bPNA, and bVVA – were obtained from Vector Laboratories,
UK. AlexaFluor488-HPA was obtained from Life Technolo-
gies. Neuraminidases from Vibrio cholerae and Arthrobacter
ureafaciens (Roche), and from Salmonella typhimurium LT2
expressed in E.coli (New England Biolabs (NEB)) were em-
ployed. PNGase-F was purchased from NEB. CD43-
myc/FLAG plasmid was obtained from Origene and trans-
fected using DMRIE-C (Life Technologies). MB was obtained
from ThermoFisher Scientific. Anti-FLAG-M2-agarose beads,
3×FLAG peptide, and gelatin were purchased from Sigma-
Aldrich.
Thiol-selective biotinylation of immunoprecipitated
CD43-myc/FLAG. CD43-myc/FLAG captured on beads were
re-suspended in PBS and treated with TCEP (1.0 mM) for 30
min at 4 °C after which MB was added at a final concentration
of 2.0 mM.44 Beads were rotated at 4 °C for 2 h and washed
five times with lysis buffer. Bound protein was released from
beads either by adding 3×FLAG peptide at a final concentra-
tion of 150 ng/μL and incubating at 4 °C for 30 min on rota-
tion or by boiling in 2× Lämmli buffer for 5 min at 100 °C.
Eluates or supernatants were subjected to 7.5 % SDS-PAGE
followed by silver staining and Western blotting. Blots were
probed with HRP-Avidin (1:40,000 in 3.0 % BSA-PBS, 50
ng/mL final concentration), anti-FLAG (1:5000 in PBS con-
taining 5.0 % NFM, 160 ng/ml final concentration) or anti-
CD43 L60 (1:10,000 in PBS containing 5.0 % NFM, 50 ng/ml
final concentration).
Confocal imaging of lectin stained Jurkat cells. For
washing of suspension-grown Jurkat cells, PBS containing
CaCl2 and MgCl2 was used throughout in order to facilitate
attachment to poly-L-lysine coated cover slips. Jurkat cells (3.0
× 105 cells/mL) in complete medium were incubated with ei-
ther D, 1, or 2 (100 μM). After 48 h, cells were collected,
washed with PBS (2 × 1 mL), re-suspended in PBS (400 μL),
placed on cover slips (pre-coated with poly-L-lysine), and
incubated at 37 °C. After 30 min, cells were gently washed
with PBS (2 × 1 mL), fixed with 2 % formaldehyde (20 min,
ASSOCIATED CONTENT
Supporting Information. Supporting results and discussion,
figures, detailed experimental procedures for synthesis of analogs
8
ACS Paragon Plus Environment

Page 9 of 17
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
(21) Bergfeld, A. K.; Pearce, O. M.; Diaz, S. L.; Lawrence, R.;
Vocadlo, D. J.; Choudhury, B.; Esko, J. D.; Varki, A. J Biol Chem
2012, 287, 28898.
(22) Hang, H. C.; Yu, C.; Pratt, M. R.; Bertozzi, C. R. J Am Chem
Soc 2004, 126, 6.
(23) Marathe, D. D.; Buffone, A., Jr.; Chandrasekaran, E. V.; Xue,
J.; Locke, R. D.; Nasirikenari, M.; Lau, J. T.; Matta, K. L.;
Neelamegham, S. Blood 2010, 115, 1303.
1–8, cell culture, flow cytometry, western blotting,
immunoprecipitation, qPCR, mass spectrometry. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
* E-mail: gopalan@nii.ac.in, Tel: +91-11-26703843.
(24) Pouilly, S.; Bourgeaux, V.; Piller, F.; Piller, V. ACS Chem Biol
2012, 7, 753.
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
(25) Pouilly, S.; Piller, V.; Piller, F. Febs J 2012, 279, 586.
(26) Du, J.; Che, P. L.; Aich, U.; Tan, E.; Kim, H. J.;
Sampathkumar, S. G.; Yarema, K. J. Bioorg Med Chem Lett 2011, 21,
4980.
(27) Paulsen, H.; Volker, P.; Brockhausen, I. Liebigs Annalen der
Chemie 1992, 5, 513.
(28) Keppler, O. T.; Horstkorte, R.; Pawlita, M.; Schmidt, C.; Reut-
ter, W. Glycobiology 2001, 11, 11R.
(29) Rillahan, C. D.; Antonopoulos, A.; Lefort, C. T.; Sonon, R.;
Azadi, P.; Ley, K.; Dell, A.; Haslam, S. M.; Paulson, J. C. Nat Chem
Biol 2012, 8, 661.
(30) Nishimura, S.; Hato, M.; Hyugaji, S.; Feng, F.; Amano, M.
Angew Chem Int Ed Engl 2012, 51, 3386.
(31) Barthel, S. R.; Antonopoulos, A.; Cedeno-Laurent, F.; Schaf-
fer, L.; Hernandez, G.; Patil, S. A.; North, S. J.; Dell, A.; Matta, K. L.;
Neelamegham, S.; Haslam, S. M.; Dimitroff, C. J. J Biol Chem 2011,
286, 21717.
(32) Zandberg, W. F.; Kumarasamy, J.; Pinto, B. M.; Vocadlo, D. J.
J Biol Chem 2012, 287, 40021.
(33) Gupta, G.; Surolia, A.; Sampathkumar, S. G. Omics 2010, 14,
419.
(34) Piller, V.; Piller, F.; Fukuda, M. J Biol Chem 1990, 265, 9264.
(35) Wilkins, P. P.; McEver, R. P.; Cummings, R. D. J Biol Chem
1996, 271, 18732.
(36) Maemura, K.; Fukuda, M. J Biol Chem 1992, 267, 24379.
(37) Ju, T.; Lanneau, G. S.; Gautam, T.; Wang, Y.; Xia, B.; Stowell,
S. R.; Willard, M. T.; Wang, W.; Xia, J. Y.; Zuna, R. E.; Laszik, Z.;
Benbrook, D. M.; Hanigan, M. H.; Cummings, R. D. Cancer Res
2008, 68, 1636.
(38) Sutton-Smith, M.; Dell, A. In Cell Biology; Third Edition ed.;
Celis, S., Ed.; Academic Press: 2006; Vol. 4, p 415.
(39) Fukuda, M.; Tsuboi, S. Biochim Biophys Acta 1999, 1455, 205.
(40) Cyster, J. G.; Shotton, D. M.; Williams, A. F. Embo J 1991,
10, 893.
(41) Clark, M. C.; Baum, L. G. Ann N Y Acad Sci 2012, 1253, 58.
(42) Weber, S.; Ruh, B.; Dippel, E.; Czarnetzki, B. M. Immunology
1994, 82, 638.
(43) Boyce, M.; Carrico, I. S.; Ganguli, A. S.; Yu, S. H.; Hangauer,
M. J.; Hubbard, S. C.; Kohler, J. J.; Bertozzi, C. R. Proc Natl Acad
Sci U S A 2011, 108, 3141.
(44) Sampathkumar, S. G.; Li, A. V.; Yarema, K. J. Nat Protoc
2006, 1, 2377.
(45) Fritz, T. A.; Hurley, J. H.; Trinh, L. B.; Shiloach, J.; Tabak, L.
A. Proc Natl Acad Sci U S A 2004, 101, 15307.
(46) Pratt, M. R.; Hang, H. C.; Ten Hagen, K. G.; Rarick, J.;
Gerken, T. A.; Tabak, L. A.; Bertozzi, C. R. Chem Biol 2004, 11,
1009.
ACKNOWLEDGMENT
Funding support from National Institute of Immunology (NII),
Department of Science and Technology (DST) (SR/S5/OBP-
66B/2008), and Department of Biotechnology (DBT)
(BT/PR11052/BRB/10/631/2008) Government of India is grate-
fully acknowledged. SGS was supported by the Ramalingaswami
Fellowship of DBT. SGS is grateful to Dr. Anne Dell and the
Glyco-TRIC program for training on glycan mass spectrometry.
We thank Dr. Mark Jones (Case Western Reserve University,
Ohio) and Dr. R. P. Roy (NII) for critical reading of the manu-
script. We thank Mr. Sebanta Pokhrel, Mr. Mohd. Aslam, and Mr.
Arnob Nandi for technical help.
REFERENCES
(1) Varki A; Cummings R. D.; Esko J. D., e. a., editors. ; Cold
Spring Harbor (NY): Cold Spring Harbor Laboratory Press; 2009.
Available from: http://www.ncbi.nlm.nih.gov/books/NBK1908/:
2009.
(2) Hang, H. C.; Bertozzi, C. R. Bioorg Med Chem 2005, 13, 5021.
(3) Jentoft, N. Trends Biochem Sci 1990, 15, 291.
(4) Bennett, E. P.; Mandel, U.; Clausen, H.; Gerken, T. A.; Fritz, T.
A.; Tabak, L. A. Glycobiology 2012, 22, 736.
(5) Ju, T.; Otto, V. I.; Cummings, R. D. Angew Chem Int Ed Engl
2011, 50, 1770.
(6) Gill, D. J.; Clausen, H.; Bard, F. Trends Cell Biol 2011, 21, 149.
(7) Manjunath, N.; Correa, M.; Ardman, M.; Ardman, B. Nature
1995, 377, 535.
(8) Ohtsubo, K.; Marth, J. D. Cell 2006, 126, 855.
(9) Schjoldager, K. T.; Vakhrushev, S. Y.; Kong, Y.; Steentoft, C.;
Nudelman, A. S.; Pedersen, N. B.; Wandall, H. H.; Mandel, U.; Ben-
nett, E. P.; Levery, S. B.; Clausen, H. Proc Natl Acad Sci U S A 2012,
109, 9893.
(10) Tarp, M. A.; Clausen, H. Biochim Biophys Acta 2008, 1780,
546.
(11) Kuan, S. F.; Byrd, J. C.; Basbaum, C.; Kim, Y. S. J Biol Chem
1989, 264, 19271.
(12) Sarkar, A. K.; Fritz, T. A.; Taylor, W. H.; Esko, J. D. Proc
Natl Acad Sci U S A 1995, 92, 3323.
(13) Hang, H. C.; Yu, C.; Ten Hagen, K. G.; Tian, E.; Winans, K.
A.; Tabak, L. A.; Bertozzi, C. R. Chem Biol 2004, 11, 337.
(14) Sletten, E. M.; Bertozzi, C. R. Angew Chem Int Ed Engl 2009,
48, 6974.
(15) Kayser, H.; Zeitler, R.; Kannicht, C.; Grunow, D.; Nuck, R.;
Reutter, W. J Biol Chem 1992, 267, 16934.
(16) Campbell, C. T.; Sampathkumar, S. G.; Yarema, K. J. Mol
Biosyst 2007, 3, 187.
(17) Prescher, J. A.; Bertozzi, C. R. Cell 2006, 126, 851.
(18) Sampathkumar, S. G.; Li, A. V.; Jones, M. B.; Sun, Z.;
Yarema, K. J. Nat Chem Biol 2006, 2, 149.
(19) Hang, H. C.; Yu, C.; Kato, D. L.; Bertozzi, C. R. Proc Natl
Acad Sci U S A 2003, 100, 14846.
(47) Tenno, M.; Toba, S.; Kezdy, F. J.; Elhammer, A. P.; Kurosaka,
A. Eur J Biochem 2002, 269, 4308.
(48) Patsos, G.; Robbe-Masselot, C.; Klein, A.; Hebbe-Viton, V.;
Martin, R. S.; Masselot, D.; Graessmann, M.; Paraskeva, C.; Gal-
lagher, T.; Corfield, A. Biochem Soc Trans 2007, 35, 1372.
(49) Carlsson, S. R.; Fukuda, M. J Biol Chem 1986, 261, 12779.
(50) Piller, F.; Piller, V.; Fox, R. I.; Fukuda, M. J Biol Chem 1988,
263, 15146.
(20) Dube, D. H.; Prescher, J. A.; Quang, C. N.; Bertozzi, C. R.
Proc Natl Acad Sci U S A 2006, 103, 4819.
(51) Ford, M. L.; Onami, T. M.; Sperling, A. I.; Ahmed, R.;
Evavold, B. D. J Immunol 2003, 171, 6527.
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 17
1
2
3
4
5
6
7
8
Table of Contents (TOC)
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 17
Journal of the American Chemical Society
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
84x93mm (300 x 300 DPI)
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 17
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
194x226mm (300 x 300 DPI)
ACS Paragon Plus Environment

Page 13 of 17
Journal of the American Chemical Society
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
210x184mm (300 x 300 DPI)
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 17
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
27x8mm (300 x 300 DPI)
ACS Paragon Plus Environment

Page 15 of 17
Journal of the American Chemical Society
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
35x15mm (300 x 300 DPI)
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 17
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
38x17mm (300 x 300 DPI)
ACS Paragon Plus Environment

Page 17 of 17
Journal of the American Chemical Society
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
89x35mm (300 x 300 DPI)
ACS Paragon Plus Environment