Semisynthetic lectin-4-dimethylaminopyridine conjugates for labeling and profiling glycoproteins on live cell surfaces

Article abstract of DOI:10.1021/ja4043214

Glycoproteins on cell surfaces play important roles in biological processes, including cell-cell interaction/signaling, immune response, and cell differentiation. Given the diversity of the structure of glycans, labeling and imaging of selected glycoproteins are challenging, although several promising strategies have been developed recently. Here, we design and construct semisynthetic reactive lectins (sugar-binding proteins) that are able to selectively label glycoproteins. Congerin II, an animal galectin, and wheat germ agglutinin are conjugated with 4-dimethylaminopyridine (DMAP), a well-known acyl transfer catalyst by our affinity-guided DMAP method and Cu(I)-assisted click chemistry. Selective labeling of glycoproteins is facilitated by the DMAP-tethered lectin catalysts both in vitro and on living cells. Two-dimensional polyacrylamide gel electrophoresis (2D-PAGE) analysis enabled us to isolate labeled glycoproteins that are uniquely exposed on distinct cell lines. Furthermore, the combination of immunoprecipitation with mass spectrometry (MS)-fingerprinting techniques allowed us to characterize 48 glycoproteins endogenously expressed on HeLa cells, and some low-abundant glycoproteins, such as epidermal growth factor receptor (EGFR) and neuropilin-1, were successfully identified. Our results demonstrate that semisynthetic DMAP-tethered lectins provide a new tool for labeling and profiling glycoproteins on living cells.

Full text of DOI:10.1021/ja4043214

Article
pubs.acs.org/JACS
Semisynthetic Lectin−4-Dimethylaminopyridine Conjugates for
Labeling and Profiling Glycoproteins on Live Cell Surfaces
Takahiro Hayashi,† Yedi Sun,† Tomonori Tamura,† Keiko Kuwata,‡ Zhining Song,† Yousuke Takaoka,†
and Itaru Hamachi*,†
†Department of Synthetic Chemistry and Biological Chemistry, Graduate School of Engineering, Kyoto University, Katsura, Kyoto
615-8510, Japan
‡Institute of Transformative Bio-Molecules (WPI-ITbM), Nagoya University, Chikusa, Nagoya 464-8601, Japan
S
* Supporting Information
ABSTRACT: Glycoproteins on cell surfaces play important roles in
biological processes, including cell−cell interaction/signaling, immune
response, and cell differentiation. Given the diversity of the structure
of glycans, labeling and imaging of selected glycoproteins are
challenging, although several promising strategies have been developed
recently. Here, we design and construct semisynthetic reactive lectins
(sugar-binding proteins) that are able to selectively label glycoproteins.
Congerin II, an animal galectin, and wheat germ agglutinin are
conjugated with 4-dimethylaminopyridine (DMAP), a well-known acyl
transfer catalyst by our affinity-guided DMAP method and Cu(I)-assisted click chemistry. Selective labeling of glycoproteins is
facilitated by the DMAP-tethered lectin catalysts both in vitro and on living cells. Two-dimensional polyacrylamide gel
electrophoresis (2D-PAGE) analysis enabled us to isolate labeled glycoproteins that are uniquely exposed on distinct cell lines.
Furthermore, the combination of immunoprecipitation with mass spectrometry (MS)-fingerprinting techniques allowed us to
characterize 48 glycoproteins endogenously expressed on HeLa cells, and some low-abundant glycoproteins, such as epidermal
growth factor receptor (EGFR) and neuropilin-1, were successfully identified. Our results demonstrate that semisynthetic
DMAP-tethered lectins provide a new tool for labeling and profiling glycoproteins on living cells.
containing reactive handles (e.g., azide, alkyne, or ketone/
aldehyde groups), followed by bioorthogonal probe ligation,
which enabled selective labeling of glycans on living cells and
even in whole organisms.7 As an alternative approach, Hsieh-
Wilson et al. developed a chemoenzymatic method to probe
glycans in living systems. They used bacterial glycosyltrans-
ferases to tether bioorthogonal reactive handles on glycans,
which can then be detected by subsequent bioorthogonal
ligation.8 Although these techniques are undoubtedly powerful,
they require metabolic engineering that might perturb the
natural cellular conditions and can detect only a handful of
mono- or disaccharides. Thus, it is highly desirable to develop
new methods to selectively label endogenous glycoproteins
with a variety of glycan structures.
INTRODUCTION
■
Glycoproteins play important roles in numerous biological
processes, such as cell−cell interactions/signaling, immune
response, and cell differentiation. It is known that unusual
expression levels of glycoproteins are strongly correlated with
progression of diseases, including cancers and autoimmune
diseases.1,2 Therefore, a number of cell surface glycoproteins
are therapeutic targets or important biomarkers.3 Despite the
pivotal roles of cell surface glycoproteins, chemical tools for
their precise and comprehensive analysis are still limited. This is
largely due to the structural heterogeneity and dynamic change
of the glycans within glycoproteins and lack of conventional
genetic and biochemical methods for precise control of
glycans.4
We previously reported that 4-dimethylaminopyridine
(DMAP) tethered to a small molecule ligand can be a powerful
tool for selective protein labeling even under aqueous
physiological conditions because of a proximity effect guided
by ligand−protein interactions, which is termed affinity-guided
DMAP (AGD) chemistry.9 Using a series of ligand-tethered
DMAP catalysts and thiophenylester acyl donors, we selectively
labeled several sugar-binding proteins (lectins), such as
congerin II (congII), and other soluble proteins in test tubes,
Detection and analysis of cell surface glycoproteins are
conventionally achieved by tagging strategies (e.g., biotinylation
and hydrazide chemistry) coupled with advanced mass
spectrometric teqniques.5,6 Unfortunately, these methods lack
selectivity between different types of glycans and are often
unable to obtain crucial information about the dynamic change
and spatial distribution of glycoproteins. Recently, two
promising methods have been proposed that overcome these
drawbacks. The first and most extensively used method is based
on a combination of metabolic labeling and bioorthogonal
ligation.7a Pioneering works by Bertozzi and co-workers
exploited the metabolic uptake of unnatural monosaccharides
Received: May 1, 2013
Published: July 26, 2013
© 2013 American Chemical Society
12252
dx.doi.org/10.1021/ja4043214 | J. Am. Chem. Soc. 2013, 135, 12252−12258

Journal of the American Chemical Society
Article
as well as a G-protein-coupled receptor on the surface of living
cells.9 In comparison to other affinity-labeling methods,10 this
catalyst-based approach is unique in the way that probe-
appended labeling reagent itself is less reactive but activated in
the presence of ligand-tethered catalyst and transferred to
nucleophilic amino acid residues near the ligand binding site of
target proteins. Thus, AGD chemistry provides a way to
selectively tag a protein of interest with high efficiency and site
specificity.
Here, we describe a new approach to label glycoproteins in
test tubes and on living cells using DMAP-tethered reactive
lectins. The design principle of the reactive lectin is based on
rational coupling of the ability of lectins to recognize sugars
with the catalytic activity of DMAP, as illustrated in Figure 1.
glycoproteins with complex glycan structures in their native
environment.
RESULTS AND DISCUSSION
■
Design, Preparation, and Characterization of DMAP-
Tethered Lectins. To prove the principle of the reactive
lectin−DMAP conjugate, we initially used congII,11 a member
of the galectin family, as a sugar recognition module. In the
present semisynthetic lectin, the well-established acyl transfer
catalyst DMAP was bioconjugated to congII and lectin−glycan
interactions facilitate selective acylation of target glycoproteins
(Figure 1). CongII−DMAP conjugates were prepared in two
steps: introduction of an azide group, a bioorthogonal reactive
handle, by AGD chemistry, followed by Cu(I)-catalyzed click
chemistry to attach DMAP moieties (Figure 2A). We designed
two types of DMAP tethers bearing one or four DMAP
group(s) jointed via a rigid hexaproline linker (hereafter called
1DMAP and 4DMAP, respectively, 6a and 6b in Figure 2C).
The linker was expected to keep the DMAP moieties away
from congII and efficiently reach the target protein.12
In the first step, congII was treated with azide-appended
thiophenyl ester 5 in the presence of lactose-tethered DMAP
1a (Figure 2B), according to our protocol reported previously.9
Progress of the reaction was monitored by matrix-assisted laser
desorption−ionization time-of-flight mass spectrometry
(MALDI−TOF MS), and after 5 h of incubation at room
temperature, we confirmed that new peaks appeared at m/z 15
500 (+257) and 15 754 (+511), which corresponded to the
calculated mass of congII modified with one (calculated m/z
+256) and two (calculated m/z +512) azide handle(s),
respectively (80% yield according to the MALDI−TOF MS;
Figure 3A). Azide-modified congII was subsequently reacted
with alkyne-appended DMAP (6a and 6b in Figure 2C) using
Cu(I)-assisted click chemistry to yield DMAP-tethered congII
(Figure 2A). Upon incubation at room temperature for 2 h,
Figure 1. Schematic representation of glycoprotein labeling with a
DMAP-tethered lectin.
Two reactive lectins with selectivity for specific sugars were
constructed using our AGD strategy. Using the DMAP-tethered
lectins and probe-bearing acyl donors, we selectively labeled
fetuin and asialofetuin (ASF) in test tubes through lectin−
glycan interactions. Because of its efficiency and biocompati-
bility, this new approach allowed for labeling and profiling of
endogenous glycoproteins on living animal cells, demonstrating
the potential of reactive lectins to identify selected
Figure 2. (A) Two-step conjugation of DMAP catalysts to lectins and chemical structures of (B) lactose-tethered and [N-acetylglucosamine
(GlcNAc)]3-tethered DMAP catalysts 1a and 1b, respectively, and azide-type acyl donor 5 and (C) alkyne-appended DMAP catalysts with one and
four DMAP group(s) (6a and 6b, respectively).
12253
dx.doi.org/10.1021/ja4043214 | J. Am. Chem. Soc. 2013, 135, 12252−12258

Journal of the American Chemical Society
Article
Figure 3. (A) MALDI−TOF MS spectra of native congII (top), congII−5 (middle), and congII−5+6b (bottom) and (B) crystal structure of
lactose-bound congII. The red and blue residues are Tyr51 and Lys78 labeled by AGD chemistry, and the green molecule is lactose (PDB ID 1IS4).
Figure 4. (A) Chemical structure of HC-type acyl donor 8. (B) Fluorescence and silver-stained gel images of ASF labeled by congII−4DMAP. (C)
pH dependence of the labeling reaction with an incubation time of 5 h. (D) Selective labeling of ASF by congII−4DMAP in the presence of ASF,
HRP, and CAI.
quantitative conversion of azide-modified congII to congII−
DMAP was demonstrated by the appearance of new peaks at
m/z 17 239 (+1974) and 19 223 (+3946), along with the
disappearance of the peaks from the azide-modified congII
species (Figure 3A and see Figure S2 of the Supporting
Information). The modification sites of congII were determined
by conventional peptide mapping, revealing that mainly lysine
78, which is located at the entrance of the sugar-binding crevice,
was modified (63%), along with tyrosine 51 (23%), which is
positioned close to the sugar-binding pocket (Figure 3B and see
Figures S3 and S4 of the Supporting Information).9a Such
distribution of the labeling sites may be reasonably ascribed to
the flexible linker of DMAP catalyst 1a. The total isolation
yields of these two reactions were typically 50%, and the purity
of DMAP-tethered congII was 81% (native, 19%; 4DMAP×1,
54%; and 4DMAP×2, 27%).
Following the same scheme, we also modified wheat germ
agglutinin (WGA), another lectin showing affinity for GlcNAc
and sialic acid, with azide using GlcNAc3-tethered DMAP 1b.13
This conjugate was then converted to DMAP-tethered WGA by
the click reaction with alkyne-appended DMAP 6b (see Figure
S5A of the Supporting Information). The labeling sites were
identified as lysine 33 (50%) and lysine 149 (27%) (see Figures
S5B and S6 of the Supporting Information). Similar to the case
of congII, both sites are proximal to the sugar-binding pocket of
WGA (see Figure S5B of the Supporting Information).13 The
total yields were approximately 60%, and the purity of WGA−
4DMAP was 71% (native, 29%; 4DMAP×1, 70%; and
4DMAP×2, 1%) (see Figure S5A of the Supporting
Information).
Glycoprotein Labeling by DMAP-Tethered Lectin in
Test Tubes. With the DMAP−lectin conjugates in hand, we
subsequently examined their ability to label glycoproteins, such
as ASF, a glycoprotein displaying terminal galactose moieties as
major glycoforms (see Figure S7A of the Supporting
Information). In the presence of congII−4DMAP and
hydroxycoumarin (HC)-appended acyl donor 8 (Figure 4A),
a new fluorescent band corresponding to HC-modified ASF
appeared in the sodium dodecyl sulfate−polyacrylamide gel
electrophoresis (SDS−PAGE) fluorescence image of the
reaction mixture, and the intensity of this band increased for
8 h (Figure 4B and see Figure S7B of the Supporting
Information). We also noticed that additional fluorescent bands
from congII−4DMAP resulted from self-labeling during this
reaction (see Figure S7C of the Supporting Information). It
should be noted that a fluorescent band from HC-modified
ASF was not observed for the reaction in the absence of
DMAP-tethered congII or in the presence of excess lactose, a
competitive ligand for congII with Ka of 2.3 × 104 M−1 (lanes 6
and 7 in Figure 4B),11c demonstrating that the labeling reaction
was effectively driven by the sugar−lectin interaction.14 The
congII−1DMAP catalyst did not readily label ASF (see Figure
S7D of the Supporting Information). Therefore, the multivalent
DMAP catalyst is required for efficient labeling, which is in
good agreement with our previous report.9c The labeling
reaction proceeded around physiological pH (7−8), with an
increased rate at higher pH, suggesting that deprotonation of
certain nucleophilic amino acid side chains accelerated the acyl
transfer reaction (Figure 4C and see Figure S8 of the
Supporting Information). Furthermore, we were able to
12254
dx.doi.org/10.1021/ja4043214 | J. Am. Chem. Soc. 2013, 135, 12252−12258

Journal of the American Chemical Society
Article
Figure 5. (A) Chemical structure of FL-type acyl donor 9, (B) CLSM images, and (C) western blots of Hela cells labeled by congII−4DMAP and
acyl donor 9. Red arrowheads indicate glycoproteins labeled by congII−4DMAP. N and D denote treatment of cells with neuraminidase and a
deglycosylation mixture of PNGase, O-glycosidase, neuraminidase, β1−4 galactosidase, and β-N-acetylglucosaminidase, respectively, prior to the
labeling reaction. (D) 2D-PAGE western blotting analysis of HeLa (top panel) and COS7 (bottom panel) cells treated with congII−4DMAP and
acyl donor 9.
determine the labeled amino acid residues of ASF by in-gel
digestion and deglycosylation followed by liquid chromatog-
raphy−tandem mass spectrometry (LC−MS/MS) analysis.
HC-labeling occurred at four Lys residues (Lys57, Lys67,
Lys232, and Lys238) and one Ser residue (Ser65) of ASF (see
Figure S9 and S10 of the Supporting Information). These
results demonstrated that DMAP-tethered congII allows for
covalent labeling of glycoproteins without any structural
information (i.e., three-dimentional structure).
The glycan selectivity of the labeling reaction catalyzed by
congII−4DMAP was subsequently evaluated using a custom-
made protein mixture containing ASF, another glycoprotein-
bearing mannose-terminated sugar group, horseradish perox-
idase (HRP), and non-glycoprotein carbonic anhydrase I
(CAI). Incubation of the protein mixture with congII−
4DMAP and acyl donor 8 resulted in a fluorescent band
corresponding to ASF to appear, revealing the glycan-based
selectivity of the congII−4DMAP catalyst (Figure 4D).15 In
contrast, substantial labeling of ASF did not occur in the
absence of congII−4DMAP (Figure 4D).
Using WGA−4DMAP, we confirmed that the fluorescent
labeling of fetuin, a glycoprotein with sialic acid moieties at its
terminals, occurred in the presence of acyl donor 8. A new
fluorescent band corresponding to HC-modified fetuin
appeared in the SDS−PAGE fluorescence image of the reaction
mixture, and the intensity of this band increased for 8 h (see
panels A and B of Figure S11 of the Supporting Information).
This band was not observed in the absence of DMAP-tethered
WGA or in the presence of excess GlcNAc3, a competitive
ligand for WGA with Ka of 1.1 × 104 M−1 (see lanes 6 and 7 in
Figure S11A of the Supporting Information).13c These
behaviors are quite similar to those observed for congII−
4DMAP. We then compared the glycan selectivity of the two
DMAP-tethered lectins. A mixture of fetuin and ASF was
incubated in the presence of each lectin−DMAP (i.e., congII−
4DMAP or WGA−4DMAP) and acyl donor 8. Distinct band
patterns were obtained with congII−4DMAP and WGA−
4DMAP; that is, both fetuin and ASF were labeled by WGA−
4DMAP, whereas only ASF was labeled by congII−4DMAP.
This result is consistent with the fact that WGA recognizes
both GlcNAc and sialic acids, whereas congII only binds to
galactose/lactose moieties. Taken together, these observations
demonstrate that the design of our DMAP-tethered lectins is
valid and the selectivity of glycoprotein labeling can be
controlled by the lectin scaffold in the DMAP-tethered lectin.
On-Cell Labeling of Glycoproteins by Lectin−DMAP.
Exploiting the reaction driven by sugar recognition, we finally
used the lectin−DMAP conjugates to selectively label
glycoproteins endogenously expressed on living cells. Cultured
HeLa and COS7 cells were treated with fluorescein (FL)-
appended acyl donor 9 (Figure 5A) in the presence of congII−
4DMAP for 90 min, during which time the cell morphology did
not changed. As shown in Figure 5B and Figure S12A of the
Supporting Information, strong fluorescence was predom-
inantly observed from the cell membrane of both HeLa and
COS7 and this fluorescence remained even after extensive
washing with buffer solution containing lactose. In contrast, no
or minimal fluorescence was detected in the absence of congII−
4DMAP or the presence of excess lactose (Figure 5B). These
results indicate that covalent labeling with 9 was greatly
facilitated by sugar−lectin interactions even on the live cell
surface, similar to the test tube experiments.
To further analyze these FL-labeled proteins, we performed
western blotting analysis of the cells after treatment with
congII−4DMAP and acyl donor 9. Several bands were clearly
stained by an anti-FL antibody at 65, 75, 100, and 130−250
kDa for HeLa (red arrowheads next to lane 1 in Figure 5C) and
at 130 kDa for COS7 (see Figure S12B of the Supporting
Information). Consistent with the confocal laser scanning
microscope (CLSM) images, no appreciable bands were
observed in the absence of congII−4DMAP or presence of
excess lactose (lanes 4 and 5 in Figure 5C and see lanes 4 and 5
in Figure S12B of the Supporting Information). Interestingly,
we found that pretreatment of these cells with neuraminidase
(sialidase) or glycosidase cocktail resulted in the loss or
downshift of most of these bands, strongly suggesting that these
FL-labeled bands originate from glycosylated proteins (lanes 2
and 3 in Figure 5C and see lanes 2 and 3 in Figure S12B of the
Supporting Information).
12255
dx.doi.org/10.1021/ja4043214 | J. Am. Chem. Soc. 2013, 135, 12252−12258

Journal of the American Chemical Society
Article
Table 1. Cell Surface Glycoproteins on HeLa Cells Labeled by CongII−4DMAP and WGA−4DMAP
a
protein
MW (kDa)
glycosylation
membrane localization
protein function
EGFR
integrin α-5
NP-1
134
116
103
85
+
+
+
+
+
+
+
single-pass type I
single-pass type I
single-pass type I
single-pass type II
single-pass type I
single-pass type I
single-pass type II
proliferation
cell adhesion
coreceptor
TfR
iron uptake
CD44
82
cell adhesion
cell adhesion
dephosphorylation
intercellular adhesion molecule 1 (ICAM1)
PLAP
59
58
a
Only proteins that are confirmed by immunoblots are listed.
We next labeled the surface of HeLa cells with WGA−
in normal cell proliferation as well as tumor growth and
progression, and its glycosylation is regarded as important for
its function (i.e., dimerization).17 Neuropilin-1 (NP-1), a
membrane-bound coreceptor for receptor tyrosin kinases, was
recently shown to activate vascular endothelial growth factor
receptors (VEGFRs) via interaction with galectin-1 to regulate
the neuron regeneration, angiogenesis, and tumor metastasis.18
Transferrin receptor (TfR) was recently reported to mediate
infection of arenaviruses, causative agents of hemorrhagic
fever.19 Placental-type alkaline phosphatase (PLAP) has long
been known as a tumor marker for seminoma and ovarian
cancer.20 Notably, some low-abundant glycorpoteins, such as
EGFR and NP-1, were labeled and characterized by our
methods,21 and most of these proteins (EGFR, TfR, PLAP, and
integrin α-5) were also identified on prostate cancer cells using
metabolic labeling.22 These results strongly validate the
potential utility of our lectin−DMAP for glycoproteomic
analysis of live cells.
4DMAP and acyl donor 9 and compared the data from western
blotting analysis to that obtained using congII−4DMAP.
Similar to the case of congII−4DMAP, labeled protein bands
were detected at molecular weights ranging from 65 to 250 kDa
and no significant labeling occurred in the absence of WGA−
4DMAP or presence of excess inhibitor GlcNAc3 (see Figure
S13 of the Supporting Information). It should be noted that the
overall intensity of the bands from the reaction with WGA−
4DMAP were higher and the number of clear bands was greater
compared to those from the congII−4DMAP experiment. This
suggests that the catalytic activity of WGA−4DMAP is higher
than that of congII−4DMAP and/or that sialylated glyco-
proteins are expressed on tumor cell surfaces more than
galactosylated glycoproteins.
To further analyze the FL-labeled glycoproteins, two-
dimensional polyacrylamide gel electrophoresis (2D-PAGE)
combined with western blotting was subsequently carried out
using anti-FL antibody. Figure 5D shows representative
blotting images for the whole cell lysates of HeLa and COS7
cells treated with congII−4DMAP and acyl donor 9. Well-
resolved blotting patterns were observed on the 2D-PAGE
blots. To our surprise, the band pattern was explicitly different
for HeLa and COS7 cells. That is, H1−H7 were selectively
labeled in HeLa cells, whereas C1−C3 were exclusively labeled
in COS7 cells (Figure 5D), indicating the presence of different
glycoproteins and/or glycoforms on these two cell lines. This is
presumably due to the distinct glycosylation states or different
expression levels of corresponding glycoproteins in these two
types of cells. The blotting image of HeLa cells labeled by
WGA−4DMAP showed a slightly different pattern from that
using congII−4DMAP; that is, several new bands at ∼65, 75,
110, and 130 kDa were observed (see Figure S14 of the
Supporting Information). This suggests that the two lectin−
DMAP species have distinct catalytic activity and/or that
sialylated glycoproteins are more abundant than galactosylated
glycoproteins on HeLa cells.
We then sought to identify the proteins labeled by the
present method. The labeled proteins were isolated by either
immunoprecipitation followed by SDS−PAGE or 2D-PAGE
experiments (as mentioned above). The isolated protein bands
were then subjected to in-gel tryptic digestion followed by LC−
MS/MS analysis. We found that 26 glycoproteins were labeled
by both lectin−DMAP catalysts, and 8 and 14 glycoproteins
were uniquely labeled by congII−4DMAP and WGA−4DMAP,
respectively (see Figure S15 of the Supporting Information).16
We confirmed some of the labeled proteins by a reciprocal
immunoblot using an antibody for each protein (see Figure S16
of the Supporting Information), which is summarized in Table
1. Among the identified proteins, epidermal growth factor
receptor (EGFR), a receptor tyrosin kinase, plays crucial roles
CONCLUSION
■
We have developed semisynthetic lectin− conjugates as a new
chemical tool for efficient labeling and analysis of glycoproteins.
The rational coupling of the glycan recognition ability of lectins
with the reactivity of organocatalyst DMAP accelerated the acyl
transfer reaction driven by sugar−lectin recognition, which
allowed for not only covalent labeling of glycoproteins in test
tubes but also glycoprotein profiling of live mammalian cells.
The present example establishes a powerful design strategy for
unique semisynthetic enzymes that is applicable under live cell
conditions.23 Given that we transferred the weak glycan binding
ability of lectins to stable covalent labeling, this method may
also be used to identify lectins with unknown partners
(glycoproteins). Future work includes construction of a
DMAP−lectin library with varied glycan selectivity to facilitate
detailed analysis of dynamic changes in glycoprotein expression
during cell differentiation and/or tumor progression. We
envision that such effort will contribute to discovery of new
glycoproteins as potential biomarkers/therapeutic targets and
provide new functional insight into biological processes.
EXPERIMENTAL SECTION
■
Materials and Methods. All chemical and biochemical reagents
were purchased from commercial sources (Wako Pure Chemical, TCI
Chemical, Sigma-Aldrich, and Watanbe Chemical) and were used
without further purification. Thin-layer chromatography (TLC) was
performed on silica gel 60 F254 precoated aluminum sheets (Merck)
and visualized by fluorescence quenching or ninhydrin staining.
Chromatographic purification was performed using flash column
chromatography on silica gel 60 N (neutral, 40−50 μm, Kanto
Chemical).
Preparation of DMAP-Tethered CongII and WGA. The
expression and purification of recombinant congII was carried out as
12256
dx.doi.org/10.1021/ja4043214 | J. Am. Chem. Soc. 2013, 135, 12252−12258

Journal of the American Chemical Society
Article
described previously.11b The congII concentration was calculated on
the basis of the molar extinction coefficient at 280 nm of 11 500 M−1
cm−1. Purified protein solution (50 μM) was incubated with DMAP
catalyst 1a (100 μM) and acyl donor 5 (250 μM) in 50 mM N-2-
hydroxyethylpiperazine-N′-2-ethanesulfonic acid (HEPES) (pH 8.0)
at room temperature for 5 h. The conjugation reaction was monitored
by MALDI−TOF MS to confirm the conversion to azide-labeled
congII. The protein solution was then subjected to gel filtration on
performed with a CLSM (FV1000 IX81, Olympus) equipped with a
60×, numerical aperture (NA) = 1.40 oil objective. Fluorescence
images were acquired using the 488 nm excitation derived from an Ar
laser. To minimize internalization of membrane proteins, samples were
maintained at 10 °C with Peltier-type stage cooling unit (Japanhight-
ech, Fukuoka, Japan) during image acquisition.
For western blotting analysis, cells were washed twice with ice-cold
PBS, lysed with radio immunoprecipitation assay (RIPA) buffer, and
mixed with an equal volume of 2× SDS−PAGE loading buffer
containing 100 mM DTT. The samples were applied to SDS−PAGE
and electrotransferred onto immune-blot polyvinylidene fluoride
(PVDF) membranes (Biorad, Hercules, CA), followed by blocking
with 5% nonfat dry milk in Tris-buffered saline (TBS) containing
0.05% Tween (Sigma-Aldrich, St. Louis, MO). The membranes were
stained with rabbit anti-fluorescein antibody (Abcam, ab19491),
followed by a HRP-conjugated goat anti-rabbit immunoglobulin G
(IgG) (Santa Cruz, SC-2004). The membranes were developed with
Chemi-Lumi One (Nacalai Tesque, Kyoto, Japan).
2D-PAGE Analysis of HeLa and COS7 Cells. Protein extracts of
HeLa and COS7 in sample buffer containing 40 mM Tris base, 7 M
urea, 2 M thiourea, 1% (w/v) ABS-14, 0.001% BPB, 2 mM
tributylphoshine, and 0.2% (w/v) Bio-Lyte 3/10 ampholyte were
applied to a 7 cm immobilized pH gradient (IPG) strip at pH 3−10 or
5−8. Rehydration was carried out for 12 h at 50 V using a PEOTEAN
i12 IEF system, followed by isoelectric focusing (IEF). All IEF
equipment and reagents were purchased from Biorad. Immediately
after IEF, IPG strips were equilibrated with buffer solutions I and II
containing 375 mM Tris-HCl (pH 8.8), 6 M urea, 20% glycerol, 2%
SDS, and 130 mM DTT for buffer solution I or 137 mM
iodoacetamide for buffer solution II. The equilibrated IPG strips
were placed on the top of acrylamide SDS−PAGE minigels, sealed
with low-melt agarose. After second-dimension separation, proteins
were detected by silver staining or western blotting, as described
above.
̈
Toyopearl HW-40 (Toso, Tokyo, Japan) using an AKTA purifier
system (GE Healthcare) equilibrated with 50 mM HEPES at pH 7.2 to
remove the remaining labeling reagents (see Figure S1 of the
Supporting Information). The protein fractions were monitored at 280
nm and bicinchoninic acid (BCA) assay (Pierce, Rockford, IL),
collected using Frac-920 (GE Healthcare) (see Figure S1 of the
Supporting Information), and if necessary, concentrated by Amicon
Ultra-4 centrifugal filter device (Millipore Corporation, Billerica, MA).
The purified azide-labeled congII was then incubated with alkyne-
appended DMAP 6a and 6b in the presence of 0.5 mM Cu(II)-[tris-
(ethylestertriazolylmethyl)amine] (TEtOTA)24 and 1 mM ascorbate
for 2 h at room temperature. The conversion to DMAP-tethered
congII was monitored by MALDI−TOF MS. The protein solution was
then subjected to gel filtration, and the protein fractions were collected
as described above (see Figure S1B of the Supporting Information).
The protein concentration and modification yield were determined by
BCA assay and ultraviolet−visible (UV−vis) absorption with a molar
extinction coefficient at 280 nm of 18 000 M−1 cm−1 per DMAP
group.
WGA, a mixture of WGA1 and WGA3 isomers, was purchased from
Funakoshi and used without further purification. The WGA
concentration was calculated on the basis of the molar extinction
coefficient at 280 nm of 25 500 M−1 cm−1. The azide-labeling,
following Click reaction, and gel filtration purification were performed
by the same procedure as congII (see Figure S1 of the Supporting
Information).
Glycoprotein Labeling by Lectin−DMAP in Test Tubes. ASF
from fetal calf serum (Sigma-Aldrich, St. Louis, MO) (1 μM) was
incubated with congII−DMAPs (6 μM) and HC-type acyl donor 8 (8
μM) with or without 15 mM lactose inhibitor in 50 mM HEPES (pH
8.0) at 17 °C. For mixed protein experiments, ASF, HRP, and CAI
from human erythrocytes (Sigma-Aldrich, St. Louis, MO) at final
concentrations of 100 μg/mL for ASF and 200 μg/mL for HRP and
CAI were incubated with congII−4DMAP (6 μM) and acyl donor 8 (8
μM) under the same conditions as above. At indicated time points,
each sample was mixed with an equal volume of 2× SDS−PAGE
loading buffer [125 mM Tris-HCl, 4% SDS, 20% glycerol, and 0.01%
bromophenol blue (BPB) at pH 6.8] with or without 100 mM
dithiothreitol (DTT) and boiled in a heatblock for 3 min. The samples
were subjected to SDS−PAGE, and HC-labeled proteins were
detected by an in-gel fluorescence image. The fluorescence band
intensity was quantified using HC-modified bovine serum albumin
(BSA) as a standard. After fluorescence imaging, the gels were stained
by silver stain (Wako, Osaka, Japan).
Glycoprotein Labeling by CongII−4DMAP on Live HeLa and
COS7 Cells. HeLa and COS7 cells were cultured in Dulbecco’s
modified Eagle’s medium (DMEM) (Sigma-Aldrich, St. Louis, MO)
supplemented with 10% fetal bovine serum and antibiotic−antimycotic
(Gibco, Carlsbad, CA) and incubated in a 5% CO2 humidified
chamber at 37 °C. Prior to labeling experiments, HeLa and COS7 cells
(3 × 104 cells) were incubated on 35 mm dishes treated with
polystyrene (BD Falcon, Franklin Lakes, NJ) at 37 °C. After 24 h, the
media was removed, and the cells were washed twice with phosphate-
buffered saline (PBS). To the cells were added labeling mixtures
containing congII−4DMAP and acyl donor 9 in HBS (20 mM
HEPES, 107 mM NaCl, 6 mM KCl, 2 mM CaCl2, and 1.2 mM MgSO4
at pH 8.0) and incubated at 4 °C for 90 min. For glycosidase-treated
samples, the cells were washed once with DMEM and incubated with
100 units/mL neuraminidase or deglycosylation mix (New England
Biolabs) in DMEM at 37 °C for 2 h prior to the labeling reactions.
For imaging experiments, the cells were washed further with ice-
cold HBS buffer containing 15 mM lactose 3 times. Cell imaging was
ASSOCIATED CONTENT
* Supporting Information
Experimental procedures, including chemical synthesis of
DMAP catalysts and acyl donors, modification site determi-
nations, and protein identifications. This material is available
free of charge via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
E-mail: ihamachi@sbchem.kyoto-u.ac.jp.
Notes
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Dr. Shigeki Kiyonaka at Kyoto University for
experimental support and helpful discussion. Takahiro Hayashi,
Yedi Sun, and Tomonori Tamura acknowledge the Japan
Society for the Promotion of Science (JSPS) Fellowships for
Young Scientists.
REFERENCES
■
(1) Dube, D. H.; Bertozzi, C. R. Nat. Rev. Drug Discovery 2005, 4,
477−488.
(2) Rudd, P. M.; Elliott, T.; Cresswell, P.; Wilson, I. A.; Dwek, R. A.
Science 2001, 291, 2370−2376.
(3) Fuster, M. M.; Esko, J. D. Nat. Rev. Cancer 2005, 5, 526−542.
(4) Lowe, J. B.; Marth, J. D. Annu. Rev. Biochem. 2003, 72, 643−691.
(5) Shin, B. K. J. Biol. Chem. 2002, 278, 7607−7616.
(6) Zhang, H.; Li, X.-J.; Martin, D. B.; Aebersold, R. Nat. Biotechnol.
2003, 21, 660−666.
(7) (a) Prescher, J. A.; Dube, D. H.; Bertozzi, C. R. Nature 2004, 430,
873−877. (b) Mahal, L. K.; Yarema, K. J.; Bertozzi, C. R. Science 1997,
12257
dx.doi.org/10.1021/ja4043214 | J. Am. Chem. Soc. 2013, 135, 12252−12258

Journal of the American Chemical Society
Article
276, 1125−1128. (c) Laughlin, S. T.; Baskin, J. M.; Amacher, S. L.;
Bertozzi, C. R. Science 2008, 320, 664−667. (d) Laughlin, S. T.;
Bertozzi, C. R. ACS Chem. Biol. 2009, 4, 1068−1072. (e) Dehnert, K.
W.; Beahm, B. J.; Huynh, T. T.; Baskin, J. M.; Laughlin, S. T.; Wang,
W.; Wu, P.; Amacher, S. L.; Bertozzi, C. R. ACS Chem. Biol. 2011, 6,
547−552.
(20) Lange, P. H.; Millan, J. L.; Stigbrand, T.; Vessella, R. L.;
Ruoslahti, E.; Fishman, W. H. Cancer Res. 1982, 42, 3244−3247.
(21) (a) Eiblmaier, M.; Meyer, L. A.; Watson, M. A.; Fracasso, P. M.;
Pike, L. J.; Anderson, C. J. J. Nucl. Med. 2008, 49, 1472−1479. (b) Jin,
Q.; Alkhatib, B.; Cornetta, K.; Alkhatib, G. Virology 2010, 396, 203−
212.
(22) (a) Hanson, S. R.; Hsu, T.-L.; Weerapana, E.; Kishikawa, K.;
Simon, G. M.; Cravatt, B. F.; Wong, C.-H. J. Am. Chem. Soc. 2007, 129,
7266−7267. (b) Hubbard, S. C.; Boyce, M.; McVaugh, C. T.; Peehl, D.
M.; Bertozzi, C. R. Bioorg. Med. Chem. Lett. 2011, 21, 4945−4950.
(23) (a) Ballell, L.; Alink, K. J.; Slijper, M.; Versluis, C.; Liskamp, R.
M. J.; Pieters, R. J. ChemBioChem 2004, 6, 291−295. (b) Chen, Z.;
Vohidov, F.; Coughlin, J. M.; Stagg, L. J.; Arold, S. T.; Ladbury, J. E.;
Ball, Z. T. J. Am. Chem. Soc. 2012, 134, 10138−10145.
(8) (a) Khidekel, N.; Arndt, S.; Lamarre-Vincent, N.; Lippert, A.;
Poulin-Kerstien, K. G.; Ramakrishnan, B.; Qasba, P. K.; Hsieh-Wilson,
L. C. J. Am. Chem. Soc. 2003, 125, 16162−16163. (b) Clark, P. M.;
Dweck, J. F.; Mason, D. E.; Hart, C. R.; Buck, S. B.; Peters, E. C.;
Agnew, B. J.; Hsieh-Wilson, L. C. J. Am. Chem. Soc. 2008, 130, 11576−
11577. (c) Chaubard, J.-L.; Krishnamurthy, C.; Yi, W.; Smith, D. F.;
Hsieh-Wilson, L. C. J. Am. Chem. Soc. 2012, 134, 4489−4492.
(9) (a) Koshi, Y.; Nakata, E.; Miyagawa, M.; Tsukiji, S.; Ogawa, T.;
Hamachi, I. J. Am. Chem. Soc. 2008, 130, 245−251. (b) Sun, Y.;
Takaoka, Y.; Tsukiji, S.; Narazaki, M.; Matsuda, T.; Hamachi, I. Bioorg.
Med. Chem. Lett. 2011, 21, 4393−4396. (c) Wang, H.; Koshi, Y.;
Minato, D.; Nonaka, H.; Kiyonaka, S.; Mori, Y.; Tsukiji, S.; Hamachi, I.
J. Am. Chem. Soc. 2011, 133, 12220−12228.
(24) Zhou, Z.; Fahrni, C. J. J. Am. Chem. Soc. 2004, 126, 8862−8863.
(10) (a) Hayashi, T.; Hamachi, I. Acc. Chem. Res. 2012, 45, 1460−
1469. (b) Takaoka, Y.; Ojida, A.; Hamachi, I. Angew. Chem., Int. Ed.
2013, 52, 4088−4106.
(11) (a) Shirai, T.; Matsui, Y.; Shionyu-Mitsuyama, C.; Yamane, T.;
Kamiya, H.; Ishii, C.; Ogawa, T.; Muramoto, K. J. Mol. Biol. 2002, 321,
879−889. (b) Ogawa, T.; Ishii, C.; Suda, Y.; Kamiya, H.; Muramoto,
K. Biosci., Biotechnol., Biochem. 2002, 66, 476−480. (c) Mamoto, K.;
Kagawa, D.; Sato, T.; Ogawa, T.; Nishida, Y.; Kamiya, H. Comp.
Biochem. Physiol., Part B: Biochem. Mol. Biol. 1999, 123, 33−45.
(12) Sato, S.-I.; Kwon, Y.; Kamisuki, S.; Srivastava, N.; Mao, Q.;
Kawazoe, Y.; Uesugi, M. J. Am. Chem. Soc. 2007, 129, 873−880.
(13) (a) Muraki, M.; Ishimura, M.; Harata, K. Biochim. Biophys. Acta
2002, 1569, 10−20. (b) Schwefel, D.; Maierhofer, C.; Beck, J. G.;
Seeberger, S.; Diederichs, K.; Moller, H. M.; Welte, W.; Wittmann, V.
̈
J. Am. Chem. Soc. 2010, 132, 8704−8719. (c) Bains, G.; Lee, R. T.;
Lee, Y. C.; Freire, E. Biochemistry 1992, 32, 12624−12628.
(14) We examined the individual roles of congII−4DMAP×1 and
congII−4DMAP×2 in the ASF labeling using the congII−4DMAP×1
sample containing a little congII−4DMAP×2 (native, 43%;
4DMAP×1, 51%; and 4DMAP×2, 6%), which showed that the
contribution from congII−4DMAP×2 was minimal in both labeling
yield and kinetics. On the other hand, the attempt to assign the role of
two 4DMAP×1 species [i.e., 4DMAP×1(Lys78) and 4DMAP×1(-
Tyr51)] in acyl transfer chemistry was unsuccessful, because we were
unable to separate the two 4DMAP×1 species. However, the facts that
the lectin−glycoprotein interaction is dynamic and that the linker of
the DMAP catalyst possesses some structural flexibility may allow for
compensation of such differences in the labeling site.
(15) A weak fluorescence band corresponding to HC-labeled CAI
was observed even in the absence of congII−4DMAP, suggesting some
non-catalytic reaction of CAI with the acyl donor. Such a reaction is
unavoidable in the test tube experiments, where the concentrations of
proteins are relatively high (i.e., a few micromolars).
(16) It should be noted that the current approach is not designed to
detect all glycoproteins, because it requires nucleophilic amino acid
residues, such as Lys, Tyr, and Ser, adjacent to glycans recognized by
the lectin−DMAPs.
(17) (a) Hanash, S. M.; Pitteri, S. J.; Faca, V. M. Nature 2008, 452,
571−579. (b) Takahashi, M.; Yokoe, S.; Asahi, M.; Lee, S. H.; Li, W.;
Osumi, D.; Miyoshi, E.; Taniguchi, N. Biochim. Biophys. Acta 2008,
1780, 520−524. (c) Liu, Y.-C.; Yen, H.-Y.; Chen, C.-Y.; Chen, C.-H.;
Cheng, P.-F.; Juan, Y.-H.; Chen, C.-H.; Khoo, K.-H.; Yu, C.-J.; Yang,
P.-C.; Hsu, T.-L.; Wong, C.-H. Proc. Natl. Acad. Sci. U.S.A. 2011, 108,
11332−11337.
(18) Hsieh, S. H.; Ying, N. W.; Wu, M. H.; Chiang, W. F.; Hsu, C. L.;
Wong, T. Y.; Jin, Y. T.; Hong, T. M.; Chen, Y. L. Oncogene 2008, 27,
3746−3753.
(19) Radoshitzky, S. R.; Abraham, J.; Spiropoulou, C. F.; Kuhn, J. H.;
Nguyen, D.; Li, W.; Nagel, J.; Schmidt, P. J.; Nunberg, J. H.; Andrews,
N. C.; Farzan, M.; Choe, H. Nature 2007, 446, 92−96.
12258
dx.doi.org/10.1021/ja4043214 | J. Am. Chem. Soc. 2013, 135, 12252−12258