Biocompatible copper(I) catalysts for in vivo imaging of glycans

Article abstract of DOI:10.1021/ja106553e

The Cu(I)-catalyzed azide-alkyne cycloaddition (CuAAC) is the standard method for bioorthogonal conjugation. However, current Cu(I) catalyst formulations are toxic, hindering their use in living systems. Here we report that BTTES, a tris(triazolylmethyl)amine-based ligand for Cu(I), promotes the cycloaddition reaction rapidly in living systems without apparent toxicity. This catalyst allows, for the first time, noninvasive imaging of fucosylated glycans during zebrafish early embryogenesis. We microinjected embryos with alkyne-bearing GDP-fucose at the one-cell stage and detected the metabolically incorporated unnatural sugars using the biocompatible click chemistry. Labeled glycans could be imaged in the enveloping layer of zebrafish embryos between blastula and early larval stages. This new method paves the way for rapid, noninvasive imaging of biomolecules in living organisms.

Full text of DOI:10.1021/ja106553e

Published on Web 11/09/2010
Biocompatible Copper(I) Catalysts for in Vivo Imaging of
Glycans
David Soriano del Amo,† Wei Wang,† Hao Jiang,† Christen Besanceney,†
Amy C. Yan,† Matthew Levy,† Yi Liu,‡ Florence L. Marlow,# and Peng Wu*,†
Department of Biochemistry and DeVelopmental and Molecular Biology, Albert Einstein College
of Medicine, YeshiVa UniVersity, 1300 Morris Park AVe, Bronx, New York 10461, United States,
and The Molecular Foundry, Lawrence Berkeley National Laboratory, One Cyclotron Road,
MS 67R6110, Berkeley, California 94720, United States
Received July 23, 2010; E-mail: peng.wu@einstein.yu.edu
Abstract: The Cu(I)-catalyzed azide-alkyne cycloaddition (CuAAC) is the standard method for bioorthogonal
conjugation. However, current Cu(I) catalyst formulations are toxic, hindering their use in living systems.
Here we report that BTTES, a tris(triazolylmethyl)amine-based ligand for Cu(I), promotes the cycloaddition
reaction rapidly in living systems without apparent toxicity. This catalyst allows, for the first time, noninvasive
imaging of fucosylated glycans during zebrafish early embryogenesis. We microinjected embryos with alkyne-
bearing GDP-fucose at the one-cell stage and detected the metabolically incorporated unnatural sugars
using the biocompatible click chemistry. Labeled glycans could be imaged in the enveloping layer of zebrafish
embryos between blastula and early larval stages. This new method paves the way for rapid, noninvasive
imaging of biomolecules in living organisms.
Introduction
cycloaddition (CuAAC) is the quintessential bioorthogonal click
reaction for chemical biologists (Figure 1a).8,9 This transforma-
The recent development of bioorthogonal click chemistry has
led to an explosion of interest in the selective covalent labeling
of biomolecules in cells and living organisms.1,2 In these labeling
reactions, one of the two bioorthogonal functional groups is first
incorporated into target biomolecules via genetic3 or metabolic
approaches.4,5 A biophysical probe, functionalized in a comple-
mentary fashion, is introduced in the second step, allowing
detection or isolation of the target of interest. To minimize
perturbations to the physiological state of the cells or organisms
probed, an ideal ligation reaction must proceed in water at
neutral pH and at temperatures between 25 and 37 °C without
any cytotoxic effects. Further, the reactive partners participating
in this transformation must be inert to the native functional
groups present in the biological system.6,7
tion is accelerated by approximately 7 orders of magnitude
compared to the uncatalyzed version.10,11 As a ligand-assisted
process, the reaction is further accelerated by Cu(I)-stabilizing
ligands, i.e., tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]amine
(TBTA, 1, Figure 1b,c) and tris(3-hydroxypropyltriazolylm-
ethyl)amine (THPTA, 3).12,13 The fast reaction kinetics and
exquisite selectivity of CuAAC have gained it widespread
utilization in chemical biology and materials science.14,15
To date, however, the use of CuAAC in living systems has
been hampered by the toxicity associated with the catalyst
formulation (CuSO4 or CuBr + sodium ascorbate + TBTA).7
TBTA, the ligand utilized in the optimized CuAAC conditions
to stabilize the Cu(I) oxidation state, has very poor water
solubility, which mandates the use of high Cu loading (0.2-1
mM) to achieve reasonable reaction rates. Free Cu(I) ions that
escape from the coordination sphere of TBTA promote the
generation of reactive oxygen and nitrogen species and induce
detrimental consequences to cellular metabolism.16 For example,
Escherichia coli cells that incorporated azidohomoalanine into
Few chemical reactions satisfy both the bioorthogonal and
click requirements. Discovered by the Sharpless-Fokin and
Meldal groups in 2002, the Cu(I)-catalyzed azide-alkyne
† Department of Biochemistry, Albert Einstein College of Medicine,
Yeshiva University.
# Developmental and Molecular Biology, Albert Einstein College of
Medicine, Yeshiva University.
‡ The Molecular Foundry, Lawrence Berkeley National Laboratory.
(1) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem., Int. Ed.
2001, 40, 2004–2021.
(8) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. Angew.
Chem., Int. Ed. 2002, 41, 2596–2599.
(9) Tornoe, C. W.; Christensen, C.; Meldal, M. J. Org. Chem. 2002, 67,
3057–3064.
(2) Baskin, J. M.; Bertozzi, C. R. QSAR Comb. Sci. 2007, 26, 1211–
1219.
(10) Wu, P.; Fokin, V. V. Aldrichim. Acta 2007, 40, 7–17.
(11) Meldal, M.; Tornoe, C. W. Chem. ReV. 2008, 108, 2952–3015.
(12) Chan, T. R.; Hilgraf, R.; Sharpless, K. B.; Fokin, V. V. Org. Lett.
2004, 6, 2853–2855.
(3) Wang, L.; Schultz, P. G. Angew. Chem., Int. Ed. 2004, 44, 34–66.
(4) Laughlin, S. T.; Agard, N. J.; Baskin, J. M.; Carrico, I. S.; Chang,
P. V.; Ganguli, A. S.; Hangauer, M. J.; Lo, A.; Prescher, J. A.;
Bertozzi, C. R. Methods Enzymol. 2006, 415, 230–250.
(5) Dieterich, D. C.; Lee, J. J.; Link, A. J.; Graumann, J.; Tirrell, D. A.;
Schuman, E. M. Nat. Protoc. 2007, 2, 532–540.
(13) Hong, V.; Presolski, S. I.; Ma, C.; Finn, M. G. Angew. Chem., Int.
Ed. 2009, 48, 9879–9883.
(14) Kolb, H. C.; Sharpless, K. B. Drug DiscoV. Today 2003, 8, 1128–
1137.
(6) Prescher, J. A.; Bertozzi, C. R. Nat. Chem. Biol. 2005, 1, 13–21.
(7) Sletten, E. M.; Bertozzi, C. R. Angew. Chem., Int. Ed. 2009, 48, 6974–
6998.
(15) Lutz, J. F. Angew. Chem., Int. Ed. 2007, 46, 1018–1025.
(16) Gaetke, L. M.; Chow, C. K. Toxicology 2003, 189, 147–163.
9
10.1021/ja106553e  2010 American Chemical Society
J. AM. CHEM. SOC. 2010, 132, 16893–16899 16893

A R T I C L E S
Soriano del Amo et al.
Figure 1. Copper(I)-catalyzed azide-alkyne cycloaddtion (CuAAC) is accelerated by Cu(I)-stabilizing ligands. (a) CuAAC of azides and terminal alkynes
to form 1,4-disubstituted 1,2,3-triazoles. (b) Structures of CuAAC-accelerating ligands. (c) Conversion-time profiles of CuAAC in the presence/absence of
accelerating ligands. Reaction conditions: propargyl alcohol (50 µM), 3-azido-7-hydroxy-coumarin (100 µM), CuSO4 (75 µM), 0.1 M potassium phosphate
buffer (pH 7.0)/DMSO 95:5, sodium ascorbate (2.5 µM), room temperature. Error bars represent the standard deviation of three replicate experiments.
their outer membrane protein OmpC survived the initial
treatment with 100 µM CuBr for 16 h but were no longer able
to divide.17 Similarly, greater than 90% of mammalian cells
underwent apoptosis and cell lysis within 20 min when treated
with 1 mM Cu(I) under optimized CuAAC conditions.7 Ze-
brafish embryos exhibited a similar sensitivity to Cu(I). When
embryos were treated with 1 mM CuSO4, 1.5 mM sodium
ascorbate, and 0.1 mM TBTA ligand, all the embryos were dead
within 15 min.7 As presently formulated, labeling of biomol-
ecules via CuAAC is not feasible in living systems.
To improve upon the biocompatibility of the azide-alkyne
cycloaddition, Bertozzi and co-workers developed a copper-
free [3+2] cycloaddition by employing ring strains as an
alternative means for alkyne activation.18,19 Among the cy-
cloalkynes examined, a difluorinated cyclooctyne, DIFO,20 and
a biarylazacyclooctynone, BARAC,21 showed rapid kinetics in
biomolecular labeling experiments. DIFO-fluorophore conju-
gates are particularly sensitive for imaging azide-tagged glycans
within complex biological systems, including live cells,20 C.
elegans,22 and zebrafish embryos,23,24 with very low background
fluorescence. However, recent in vivo studies revealed that
DIFO-based probes bind to mouse serum albumin nonspecifi-
cally, presumably via covalent-bond formation between the
cyclooctyne and cysteine residues.25 In addition, the construction
of these cyclooctyne-based probes usually involves multistep
linear syntheses, which can be a challenge.20,26 A major goal
in this field is to identify a new copper catalyst formulation
that can promote rapid azide-alkyne cycloaddition in living
systems without cytotoxicity.
Results and Discussion
In nature, copper is a bioessential element and the second
most abundant transition metal in the human organism.27 With
Cu(II)/Cu(I) redox potential between 0.0 and 0.8 V, copper-
containing enzymes are prevalent, participating particularly in
reactions involving dioxygen transport and utilization,27-30 as
well as in the degradation of unwanted side products of O2
metabolism such as O2 radicals.31 The activities of these
•-
enzymes are elegantly orchestrated by the ligands surrounding
the copper ions in the active sites. Applying lessons from nature,
we sought to design a new ligand for Cu(I) that could extend
the utilization of CuAAC to living systems. When coordinating
with Cu(I), the ligand would engage in forming an active copper
catalyst to promote the azide-alkyne cycloaddition at micro-
molar Cu(I) concentrations, while sequestering the copper-
associated cytotoxicity.
To develop a nontoxic Cu(I) catalyst that is suitable for
applications in living systems, we screened a library of 14 TBTA
analogues (Figure S1, Supporting Information), most of which
showed improved water solubility, except 4, 8, 11, and 15. From
all monomeric TBTA analogues that are water-soluble, we
discovered that the higher number of bulky tert-butyl groups a
ligand bears, the faster the corresponding cycloaddition reaction
becomes (Figure S2d, Supporting Information). This screen led
to the discovery of a bis(tert-butyltriazoly) ligand, BTTES (2,
Scheme 1), which contains the ideal balance between reactivity
and solubility. This ligand was the most effective in promoting
the CuAAC among all water-soluble ligands screened (Figure
S2d) and dramatically accelerated the rate of the azide-alkyne
cycloaddition by coordination with the in situ generated Cu(I)
(Figure 1c). It also bears a sulfate functionality designed to
(17) Link, A. J.; Vink, M. K.; Tirrell, D. A. J. Am. Chem. Soc. 2004, 126,
10598–10602.
(18) Agard, N. J.; Prescher, J. A.; Bertozzi, C. R. J. Am. Chem. Soc. 2004,
126, 15046–15047.
(19) Jewett, J. C.; Bertozzi, C. R. Chem. Soc. ReV. 2010, 39, 1272–1279.
(20) Baskin, J. M.; Prescher, J. A.; Laughlin, S. T.; Agard, N. J.; Chang,
P. V.; Miller, I. A.; Lo, A.; Codelli, J. A.; Bertozzi, C. R. Proc. Natl.
Acad. Sci. U.S.A. 2007, 104, 16793–16797.
(21) Jewett, J. C.; Sletten, E. M.; Bertozzi, C. R. J. Am. Chem. Soc. 2010,
132, 3688–3690.
(22) Laughlin, S. T.; Bertozzi, C. R. ACS Chem. Biol. 2009, 4, 1068–1072.
(23) Laughlin, S. T.; Baskin, J. M.; Amacher, S. L.; Bertozzi, C. R. Science
2008, 320, 664–667.
(27) Kaim, W.; Rall, J. Angew. Chem., Int. Ed. 1996, 35, 43–60.
(28) Solomon, E. I.; Tuczek, F.; Root, D. E.; Brown, C. A. Chem. ReV.
1994, 94, 827–856.
(24) Baskin, J. M.; Dehnert, K. W.; Laughlin, S. T.; Amacher, S. L.;
Bertozzi, C. R. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 10360–10365.
(25) Chang, P. V.; Prescher, J. A.; Sletten, E. M.; Baskin, J. M.; Miller,
I. A.; Agard, N. J.; Lo, A.; Bertozzi, C. R. Proc. Natl. Acad. Sci.
U.S.A. 2010, 107, 1821–1826.
(29) Magnus, K. A.; Tonthat, H.; Carpenter, J. E. Chem. ReV. 1994, 94,
727–735.
(30) Solomon, E. I.; Chen, P.; Metz, M.; Lee, S. K.; Palmer, A. E. Angew.
Chem., Int. Ed. 2001, 40, 4570–4590.
(31) Pierre, J.-L.; Chautemps, P.; Refaif, S.; Beguin, C.; El Marzouki, A.;
Serratrice, G.; Saint-Aman, E.; Rey, P. J. Am. Chem. Soc. 1995, 117,
1965–1973.
(26) Poloukhtine, A. A.; Mbua, N. E.; Wolfert, M. A.; Boons, G. J.; Popik,
V. V. J. Am. Chem. Soc. 2009, 131, 15769–15776.
9
16894 J. AM. CHEM. SOC. VOL. 132, NO. 47, 2010

Biocompatible Cu(I) Catalysts for Imaging Glycans in Vivo
A R T I C L E S
Scheme 1. Synthesis of BTTES
minimize the cell membrane permeability of the coordinated
copper. BTTES was synthesized via CuAAC by reacting
2-azidoethanol with N,N-bis(triazolmethyl)propargylamine capped
with bulky tert-butyl groups (Scheme 1). In a fluorogenic assay
(Figure S2a), 75 µM CuSO4 premixed with BTTES at various
ratios (between 2:1 and 6:1, [ligand]:[copper]) yielded >50%
cycloaddition product within 20 min (Figure S2b). By contrast,
the TBTA-Cu(I)-catalyzed reaction was significantly slower,
leveling off with <20% yield, and the reaction rate decreased
gradually when more than 1 equiv of the ligand was used (Figure
S2c).
LNCaP cells, a human prostate adenocarcinoma cell line, bearing
alkynyl (SiaNAl) or azido sialic acid (SiaNAz) residues within
their cell-surface glycans were reacted with biotin-azide or
biotin-alkyne (100 µM), catalyzed by various concentrations
of the copper catalyst for 1-2.5 min at room temperature and
then stained with Alexa Fluor 488-streptavidin and analyzed
by flow cytometry. As shown in Figure 2c,d, cells displaying
alkyne or azide showed a dose-dependent increase in fluores-
cence upon treatment with the catalyst. Cells lacking alkynyl
or azido residues showed only background labeling (Figure 2b;
Figure S5, Supporting Information). Similar results were
observed for HEK 293T, a human embryonic kidney cell line,
Jurkat, and Pro-5 CHO cells (Figures S6-S8, Supporting
Information). In all cell lines profiled, significantly higher
fluorescence was detected in cells displaying alkyne on the cell
surface than the corresponding azide-displaying counterparts,
with robust labeling achieved within 1 min. This observation
is consistent with our previous discovery that Ac4ManNAl
metabolism is more efficient than that of Ac4ManNAz in these
mammalian cell lines.25
Biocompatibility Evaluation of the BTTES-Cu(I) Complex. As
the first step toward extending the use of CuAAC to living
systems, we sought to evaluate the cytotoxicity of the new
ligand-copper complex by 7-aminoactinomycin D (7-AAD)
staining and flow cytometry analysis. Jurkat cells, a human T
lymphocyte cell line, were incubated with 50 µM peracetylated
N-(4-pentynoyl)mannosamine (Ac4ManNAl) or peracetylated
N-azidoacetylmannosamine (Ac4ManNAz) to introduce the
corresponding sialic acid (SiaNAl or SiaNAz) into their cell-
surface glycoconjugates.32 Three days later, the cells were
reacted with biotin-azide (or biotin-alkyne, 100 µM), sodium
ascorbate (2.5 mM), and CuSO4 (25-75 µM) premixed with
BTTES or in the absence of BTTES. The reactions were
quenched after 5 min with bathocuproine sulfonate (BCS), a
biocompatible copper chelator.33 Significant cell death was
observed when the cells were treated with the in situ generated
Cu(I) alone, and cell death increased along with increasing
copper concentrations (Figure S3, Supporting Information).
Interestingly, we observed that cells bearing alkynes on their
surface underwent apoptosis much faster than their azide-bearing
counterparts, presumably due to the formation of the reactive
Cu(I)-acetylide on their surface. In stark contrast, when cells
were treated with the BTTES-Cu(I) complexes, cell death
decreased dramatically. In the presence of BTTES-Cu(I) 6:1
complex, cell death was completely suppressed to the same level
as for untreated cells. Similar results were obtained using Pro-5
Chinese hamster ovary (CHO) cells (Figure S4, Supporting
Information).
To measure the kinetics of the BTTES-promoted CuAAC on
the cell surface, we reacted SiaNAz-expressing Pro-5 CHO cells
with biotin-alkyne and the BTTES-Cu(I) catalyst ([Cu] ) 75
µM) for 2-17 min, followed by staining with Alexa Fluor
488-streptavidin. As quantified by flow cytometry analysis, the
cell-associated Alexa Fluor 488 fluorescence reached a plateau
around 14 min, indicating the completion of the ligation reaction
between biotin-alkyne and the cell surface reagent-accessible
SiaNAz (Figure 3). Importantly, during the 17-min course of
the reaction, no cell apoptosis was observed (Figure S9,
Supporting Information).
To evaluate if BTTES-Cu(I)-catalyzed click chemistry
causes any long-term perturbation to the treated cells, we labeled
Jurkat and HEK 293T cells bearing SiaNAz or SiaNAl residues
with biotin-alkyne or biotin-azide in the presence of the
catalyst ([Cu] ) 75 µM) for 3 min. The reactions were then
quenched with BCS. The catalyst-treated cells were cultured
for 3-4 days. In control experiments, we cultured untreated
cells and cells treated with in situ generated Cu(I) in the absence
of ligand. Viable cells, based on Trypan Blue assay, were
counted each day. Cells treated with the BTTES-Cu(I) catalyst
proliferated at similar rates as untreated cells (Figure 4a). By
contrast, all cells labeled in the absence of BTTES underwent
lysis within 24 h of the labeling experiments. Similar results
were obtained using an alamarBlue cell viability assay (Figure
4b). Taken together, the discovery of BTTES creates a nontoxic
and highly efficient Cu(I) catalyst for azide-alkyne cycload-
BTTES-Cu(I)-Catalyzed Azide-Alkyne Cycloaddition Al-
lows Rapid Detection of Sialylated Glycans on Live Cell Surface.
Next, we investigated the utility of the BTTES-Cu(I) catalyst
for live cell labeling (the catalyst formulation, BTTES-CuSO4
([ligand]:[Cu] ) 6:1) complex, and 2.5 mM sodium ascorbate).
(32) Chang, P. V.; Chen, X.; Smyrniotis, C.; Xenakis, A.; Hu, T.; Bertozzi,
C. R.; Wu, P. Angew. Chem., Int. Ed. 2009, 48, 4030–4033.
(33) Mohindru, A.; Fisher, J. M.; Rabinovitz, M. Nature 1983, 303, 64–
65.
9
J. AM. CHEM. SOC. VOL. 132, NO. 47, 2010 16895

A R T I C L E S
Soriano del Amo et al.
Figure 2. Detection of glycoconjugates on the surface of live cells via biocompatible CuAAC. (a) Schematic representation of metabolic labeling and
detection of cell-surface sialic acids using Ac4ManNAl and Ac4ManNAz and BTTES-Cu(I)-catalyzed click chemistry. (b-d) Flow cytometry data of cell
surface labeling experiments described in (a) using LNCaP cells. (b) Cells were treated with biotin-azide or biotin-alkyne (100 µM) in the presence of the
BTTES-Cu(I) catalyst ([Cu] ) 75 µM) for 1 or 2.5 min, respectively, before probing with Alexa Fluor 488-streptavidin conjugates. In all cases, cells
cultured in the absence of sugar displayed mean fluorescence intensity (MFI, arbitrary units) values <15. (c) Cells were labeled with biotin-azide (100 µM)
for 1 min in the presence of 25-75 µM Cu(I). (d) Cells were labeled with biotin-alkyne (100 µM) for 2.5 min in the presence of 25-75 µM Cu(I). Error
bars represent the standard deviation of three replicate experiments. Solid line, + Ac4ManNAl (c) or Ac4ManNAz (d); dashed line, no sugar.
The first class of glycans we chose to explore in developing
zebrafish was fucosylated glycans. Specific terminal glycan
fucosylation confers unique properties to cell surface glyco-
conjugates and is often regulated in cellular differentiation and
embryogenesis.34 There is evidence that Lewis X (Lex), an R1,3
fucosylated trisaccharide, serves as a biomarker for murine
pluripotent stem cells, in which it plays an important role in
adhesion, guiding the migration of cells in the pre-implantation
embryo.34,35 Studies have also shown that during development
of the zebrafish nervous system, neuroepithelial cells require
fucosylated glycans to guide the migration of vagus motor
neuron progenitors in the developing hindbrain.36 Despite the
physiological significance of these fucosylated glycans, there
is currently no method for imaging them in live cells or
organisms.37,38
In nature, fucosylated glycans are synthesized by fucosyl-
tranferases, enzymes that transfer the activated fucose from
GDP-fucose (GDP-Fuc) to the acceptor substrates. GDP-Fuc
is formed through either a de novo biosynthetic pathway or a
salvage pathway.39 In the de novo pathway, GDP-mannose is
converted into GDP-Fuc via GDP-mannose 4,6-dehydratase
(GMD) and GDP-keto-6-deoxymannose 3,5-epimerase/4-reduc-
Figure 3. Analysis of CuAAC kinetics on the surface of CHO cells by
flow cytometry. CHO cells were incubated for 3 days in untreated medium
or medium supplemented with 50 µM Ac4ManNAz, followed by labeling
with 100 µM biotin-alkyne in the presence of BTTES-Cu(I) catalyst ([Cu]
) 75 µM). The reaction was quenched at various time points with BCS
and probed with Alexa Fluor 488-streptavidin conjugates. Error bars
represent the standard deviation of three replicate experiments.
dition, setting the stage to test its use for in vivo imaging of
azide- and alkyne-tagged biomolecules.
Microinjection of GDP-FucAl Allows Imaging of Fucosylated
Glycans during the Blastula and Tissue Segmentation Periods of
Zebrafish Embryogenesis. The development of a multicellular
organism as it grows from a single zygote to a complex system
of tissues is accompanied by complex changes of glycosylation
on the cell surface. To test if the biocompatible copper catalyst
could be extended to image the dynamic changes of glycans in
vivo, we took advantage of the rapid embryonic development
and optical clarity of the zebrafish embryo as a vertebrate model
system.
(34) Becker, D. J.; Lowe, J. B. Glycobiology 2003, 13, 41R–53R.
(35) Saito, S.; Liu, B.; Yokoyama, K. Hum. Cell 2004, 17, 107–115.
(36) Ohata, S.; Kinoshita, S.; Aoki, R.; Tanaka, H.; Wada, H.; Tsuruoka-
Kinoshita, S.; Tsuboi, T.; Watabe, S.; Okamoto, H. DeVelopment 2009,
136, 1653–1663.
(37) Hsu, T. L.; Hanson, S. R.; Kishikawa, K.; Wang, S. K.; Sawa, M.;
Wong, C. H. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 2614–2619.
(38) Sawa, M.; Hsu, T. L.; Itoh, T.; Sugiyama, M.; Hanson, S. R.; Vogt,
P. K.; Wong, C. H. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 12371–
12376.
(39) Ma, B.; Simala-Grant, J. L.; Taylor, D. E. Glycobiology 2006, 16,
158R–184R.
9
16896 J. AM. CHEM. SOC. VOL. 132, NO. 47, 2010

Biocompatible Cu(I) Catalysts for Imaging Glycans in Vivo
A R T I C L E S
Figure 4. BTTES-Cu(I)-catalyzed click chemistry has no long-term perturbation to Jurkat and HEK cells. (a) Cell growth curve after click chemistry
treatment. (b) Cell viability measured using an alamarBlue assay (Invitrogen). Error bars represent the standard derivation for three replicates.
tase (also named FX) (Figure 5a). By contrast, the salvage
pathway begins with the uptake of free fucose, which is
converted into fucose-1-phosphate by fucokinase, and thence
to GDP-fucose by GDP-fucose pyrophosphorylase (Figure 5b).
In vertebrates, the de novo biosynthetic pathway usually
accounts for 90% of the total GDP-Fuc production, whereas
the remaining 10% of GDP-Fuc is contributed by the salvage
pathway. Discovered by Wong and Bertozzi, the salvage
pathway of cultured mammalian cells can be hijacked by
unnatural fucose analogues. Their studies demonstrated that
6-azidofucose is cytotoxic, preventing its use in live cell imaging
via the cyclooctyne-based copper-free click chemistry.40 How-
ever, FucAl, a 5-alkyne-modified fucose analogue, is well
tolerated by several mammalian cell lines.37,38
To image fucosylated glycans in zebrafish embryos, we
conceived a strategy of bypassing the salvage pathway and using
the downstream GDP-FucAl directly as a metabolic substrate.
We synthesized GDP-FucAl chemoenzymatically41 and micro-
injected one-cell stage embryos with 20 pmol of GDP-FucAl
(Figure 5c). We dechorionated the embryos and reacted them
at 10 h post fertilization (hpf) with Alexa Fluor 488-azide (100
µM) in the presence of the BTTES-Cu(I) catalyst ([Cu] ) 50
µM). Immediately following a 3-min reaction, we were able to
observe robust labeling of the treated embryos, and only
background fluorescence was detected for control embryos
microinjected with GDP-Fuc (Figure 5d; Figure S10, Supporting
Information). We also reacted embryos at the blastula period
(2.5 hpf) and achieved strong labeling of fucosides after a 3-min
click reaction; this labeling remained visible at 24 hpf (Figure
5e).
To examine the developmental period during which a single
dose of GDP-FucAl could yield a detectable alkyne-dependent
fluorescent signal, we extended our labeling time course to 80
hpf, at which point the zebrafish has completed most of its
morphogenesis and reached the early larval stage. Specifically,
we examined 80 hpf larvae, which are distinguishable by
progressive development of the pectoral fins, the median fin
fold, and the pigmentation pattern produced by the melanophores
(Figure S11, Supporting Information). After labeling with Alexa
Fluor 488-azide in the presence of the BTTES-Cu(I) catalyst,
we could observe robust labeling for the zebrafish larva at 80
hpf (Figure 5f), indicating that the exogenously supplied GDP-
FucAl persisted and remained accessible for the labeling
reaction. As revealed by examining individual z-planes obtained
using confocal fluorescence microscopy, only the enveloping
layer of the blastodisc was labeled by the Alexa fluorophore
due to the low tissue penetration depth of the click chemistry
reagents (Movie S1, Supporting Information).
Recently, Finn and co-workers reported that THPTA (3), a
water-soluble member of the tris(triazolylmethyl)amine family,
is a superior ligand to TBTA for cowpea mosaic virus
bioconjugation.13 To evaluate if this ligand could be used for
in vivo labeling and to compare its labeling efficiency to that
of BTTES, we conducted labeling experiments using zebrafish
embryos metabolically labeled with GDP-FucAl (n ) 24). At
regular intervals starting from the tissue segmentation period,
we reacted the alkyne-bearing embryos with Alexa Fluor
488-azide in the presence of the THPTA-Cu(I) catalyst under
the same conditions as specified for the BTTES-Cu(I) catalyst.
In contrast to the BTTES-Cu(I)-mediated reaction, we could
detect only very weak fluorescence signal above the background
with the THPTA-Cu(I) catalyst (Figure S13, Supporting
Information). The exact mechanism of the different catalytic
activities between these structurally similar ligands is a complex
(40) Rabuka, D.; Hubbard, S. C.; Laughlin, S. T.; Argade, S. P.; Bertozzi,
C. R. J. Am. Chem. Soc. 2006, 128, 12078–12079.
(41) Wang, W.; Hu, T.; Frantom, P. A.; Zheng, T.; Gerwe, B.; Del Amo,
D. S.; Garret, S.; Seidel, R. D., III; Wu, P. Proc. Natl. Acad. Sci.
U. S. A. 2009, 106, 16096–16101.
9
J. AM. CHEM. SOC. VOL. 132, NO. 47, 2010 16897

A R T I C L E S
Soriano del Amo et al.
Figure 5. In vivo imaging of fucosylated glycans during early zebrafish embryogenesis via BTTES-Cu(I)-catalyzed click chemistry. (a) GDP-Fuc de novo
biosynthetic pathway. (b) GDP-Fuc salvage pathway. (c) Microinjection combined with the BTTES-Cu(I)-catalyzed click chemistry enables the labeling of
fucosylated glycans in zebrafish embryos. Zebrafish embryos were microinjected with a single dose of GDP-FucAl and allowed to develop to 10 hpf (d), 2.5
hpf (e), and 80 hpf (f). The embryos were then reacted with Alexa Fluor 488-azide catalyzed by BTTES-Cu(I) and imaged using confocal microscopy. The
maximum intensity z-projection fluorescence image for (e) was acquired at 24 hpf. Scale bar: 100 µm.
and unresolved issue42 that is currently under thorough inves-
tigation in our laboratories.
suggesting that the defects may be due to microinjection rather
than the click reaction (Figure S12, Supporting Information).
The BTTES-Cu(I) catalyst, combined with yolk-cell micro-
injection, allows for the first time rapid imaging of fucosylated
glycans in the enveloping layer of zebrafish embryos as early
as 2.5 h post fertilization. Since transplantation can be easily
performed from labeled donor embryos to unlabeled hosts
beginning with blastula stages (4 hpf) through the onset of
gastrulation (5.3 hpf), exploiting this manipulation may allow
us to follow the allocation of these glycans in specific lineages
of the embryo during development.
Importantly, this work adds the alkynyl functionality to the
bioorthogonal reagent repertoire of in vivo imaging. As the
synthesis of BTTES can be easily scaled up to multigrams and
many azide- and alkyne-functionalized reagents are commercially
available, only genetic or metabolic manipulations are required to
generate the tagged biomolecules in vivo for this new procedure
of click modification. Thus, the chemical tools reported here can
be directly applied for studying other sectors of the glycome and
be generalized for dynamic in vivo imaging or profiling of other
biomolecules, i.e., proteins, lipids, and cofactors, many of which
are already targets of the canonical click chemistry in vitro.14
Conclusion
Imaging the glycome within a cellular environment is now
possible using new tools from the emerging field of bioorthogo-
nal click chemistry.43 With the discovery of the tris(triazolyl-
methyl)amine-based ligand BTTES, we have identified a
practical and biocompatible click reaction that allows the
imaging of glycan biosynthesis during zebrafish early develop-
ment. Robust labeling, which requires 1 h of reaction time when
using the cyclooctyne-based copper-free click chemistry, was
achieved within minutes.24 We observed little toxicity or
developmental abnormalities resulting from treatments with the
Cu(I) catalyst through four days post click reaction. Among the
120 embryos microinjected with the nucleotide sugar analogues
and treated with the click reagents, less than 12 embryos
exhibited minor developmental defects, i.e., impaired posterior
body development characterized by a shorter anterior-posterior
axis (Figure S11, Supporting Information). Comparable phe-
notypes were observed in 10% of embryos that were injected
but were not treated with the click reagents (n ∼ 1000),
Experimental Section
(42) For mechanistic insights regarding CuAAC, see: Presolski, S. I.; Hong,
Labeling of Alkynyl or Azido Sialic Acids on Mammalian Cell
Surface with Biotinylated Probes via CuAAC and Analyzing by
Flow Cytometry. Jurkat, CHO, HEK, and LNCaP cells were
incubated for 3 days in untreated medium or medium containing
V.; Cho, S.-H.; Finn, M. G. J. Am. Chem. Soc. 2010, 132, 14570–
14576.
(43) Laughlin, S. T.; Bertozzi, C. R. Proc. Natl. Acad. Sci. U.S.A. 2009,
106, 12–17.
9
16898 J. AM. CHEM. SOC. VOL. 132, NO. 47, 2010

Biocompatible Cu(I) Catalysts for Imaging Glycans in Vivo
A R T I C L E S
50 µM Ac4ManNAz or Ac4ManNAl. The cells then were distributed
into a 96-well round-bottom tissue culture plate (0.4-0.5 million
cells/well), pelleted (300g, 3 min), and washed two times with 200
µL of labeling buffer (PBS, pH 7.4, containing 1% FCS). Cells
were then resuspended in 92 µL of labeling buffer (stored at 4 °C),
followed by addition of 100 µM biotin-alkyne (for azide-bearing
cells) or biotin-azide (for alkyne-bearing cells), 2.5 mM sodium
ascorbate, and BTTES-CuSO4 complex (CuSO4 concentrations
from 20 to 75 µM, BTTES-CuSO4 ratio from 5:1 to 7:1) to the
buffer at room temperature. The reactions were quenched by adding
2 µL of copper chelator BCS (50 mM) at various time points. The
cells were then pelleted, washed three times with labeling buffer
(4 °C), and resuspended in the same buffer containing 1 µg/mL
Alexa Fluor 488-streptavidin (Invitrogen). After a 30-min incubation
on ice (in the dark), the cells were washed three times with 200
µL of cold labeling buffer and then resuspended in 400 µL of FACS
buffer (Hank’s balanced salt solution, pH ) 7.4, 1% BCS, 2 µg/
mL 7-AAD, 0.2% NaN3) for flow cytometry analysis.
AlamarBlue Cell Viability and Cell Growth Assay. Jurkat and
HEK 293T cells were cultured in medium or medium containing
50 µM Ac4ManNAz or Ac4ManNAl. After 72 h, all cells were
harvested, with HEK cells harvested by trypsinization (trypsin
EDTA, 1×). Cells were washed twice with 10 mL of PBS (1%
FBS) and resuspended in the same buffer to a final concentration
of 1 M viable cells per 100 µL. Cells cultured with Ac4ManNAz
were reacted with or without biotin-alkyne, and cells cultured with
Ac4ManNAl were reacted with or without biotin-azide in a 100
µL reaction containing the BTTES-CuSO4 complex (75 µM
CuSO4, 450 µM BTTES), and 2.5 mM freshly prepared sodium
ascorbate at room temperature. As a positive apoptosis control to
determine the copper toxicity in the absence of the ligand, cells
(cultured in the presence of the unnatural sugars) were reacted with
75 µM CuSO4 and 2.5 mM sodium ascorbate. After 3 min, all
reactions were quenched with BCS (1 mM). Untreated cells cultured
in medium or medium containing Ac4ManNAz or Ac4ManNAl were
included as negative controls. Following the reactions cells were
washed twice with 0.5 mL of PBS (1% FBS) and once with 0.5
mL of complete medium. Cells were resuspended in 1 mL of
complete medium, and total cell numbers were determined. In most
cases, ∼75% cells were recovered. All cells were then normalized
to the lowest concentration. Cell viability was determined by
alamarBlue colorimetric assay (Invitrogen). Briefly, treated or
untreated Jurkat were seeded at 80 000 cells/100 µL complete
medium, and HEK 293T cells were seeded at 10 000 cells/100 µL
of complete medium. AlamarBlue stock solution was diluted to
10% and incubated with the Jurkat cells and HEK 293T cells for
6 and 20 h, respectively, at which time the percentage of the reduced
form of resorufin (alamarBlue) reached ∼50% (the percentage of
the reduced form of resorufin is proportional to cell viability). The
absorbance of resorufin was measured at 540 (Red) and 600 (Ox)
nm. To determine cell growth following the reaction, Jurkat and
HEK 293T cell were seeded at 60 000 and 125 000 cells/mL,
respectively. The viable cells were counted every 24 h for 3-4
days using the Trypan blue dye exclusion method.
Metabolic Labeling of Zebrafish Embryos by Microinjection
with GDP-FucAl and Detection by the BTTES-Cu(I)-Catalyzed
Click Chemistry. Zebrafish embryos at the one-cell or two-cell stage
were microinjected in the yolk with 1 nL of a 20 mM solution of
GDP-FucAl (or GDP-fucose) and either rhodamine-dextran (5%
w/v) or phenol red (0.1% w/v) in 0.2 M KCl as a tracer. The
embryos were allowed to develop to 5 hpf, at which point they
were manually dechorionated. The dechorionated embryos were
cultured in E3 embryo medium at 28 °C for an additional 3 h before
transferring to 1% agarose-coated 96-well plates containing 92 µL
of embryo medium using a fire-polished wide-bore Pasteur pipet.
Alexa Fluor 488-azide (100 µM from a 1.75 mM stock in H2O)
was added to each well, followed by BTTES-CuSO4 6:1 complex
([Cu] ) 50 µM from 2.5 mM stock). The solutions were gently
shaken, and freshly prepared sodium ascorbate (2.5 mM from 100
mM stock in embryo medium) was added to initiate the click
reactions. After 3 min, the reaction was quenched with BCS (1
mM from 50 mM stock in water) and diluted immediately with
100 µL of embryo medium. The treated embryos were washed two
times with 15 mL of embryo medium and were anesthetized with
0.2% (w/v) tricaine in embryo medium. After mounting on a slide
or a 50 mm glass bottom dish (Matek) in 1.2% ultralow melt
agarose in E3 medium, the embryos were ready for imaging by
confocal microscopy.
Acknowledgment. This work was partially supported by the
National Institutes of Health (GM080585, P.W.), the Mizutani
Foundation for Glycoscience (P.W.), and startup funds from Albert
Einstein College of Medicine. C.B. is supported by the NIH training
grant T32 GM007491. Part of the ligands synthesis was performed
as a User Project at the Molecular Foundry, Lawrence Berkeley
National Laboratory, which was supported by the Office of Science,
Office of Basic Energy Sciences, U.S. Department of Energy, under
contract DE-AC02-05 CH11231. We thank Prof. Valery Fokin for
providing six azide and alkyne precursors for initial ligand
screening, Prof. Pamela Stanley for providing Pro-5 CHO cells,
Spartak Kalinin for zebrafish care, and Liana M. Klivansky for help
on chemical synthesis. P.W. thanks Prof. Carolyn Bertozzi for
insightful discussion of copper-free click chemistry, Prof. Valery
Fokin for discussion of the mechanism of CuAAC, Prof. Erik Snapp
for assistance in fluorescence image acquiring and processing, and
Prof. Carolyn Bertozzi, Prof. Vern Schramm, and Dr. Penny Drake
for critical comments on the manuscript.
Supporting Information Available: Synthetic procedures,
spectral data for BTTES, protocol for kinetic measurement,
experimental procedures for flow cytometry, fluorescent mi-
croscopy, and zebrafish husbandry and strains. This material is
available free of charge via the Internet at http://pubs.acs.org.
JA106553E
9
J. AM. CHEM. SOC. VOL. 132, NO. 47, 2010 16899