Efficient solid-phase synthesis of trifunctional probes and their application to the detection of carbohydrate-binding proteins

Article abstract of DOI:10.1021/ol0523188

(Chemical Equation Presented) An efficient solid-phase synthesis of trifunctional probes containing a photoreactive group, a reporter tag, and a carbohydrate ligand was developed. Labeling studies with these probes demonstrate that specific lectins can be

Full text of DOI:10.1021/ol0523188

ORGANIC
LETTERS
2005
Vol. 7, No. 24
5477-5480
Efficient Solid-Phase Synthesis of
Trifunctional Probes and Their
Application to the Detection of
Carbohydrate-Binding Proteins
Myung-ryul Lee, Da-Woon Jung, Darren Williams, and Injae Shin*
Department of Chemistry, Yonsei UniVersity, Seoul 120-749, Korea
injae@yonsei.ac.kr
Received September 23, 2005
ABSTRACT
An efficient solid-phase synthesis of trifunctional probes containing a photoreactive group, a reporter tag, and a carbohydrate ligand was
developed. Labeling studies with these probes demonstrate that specific lectins can be labeled with high sensitivity and selectivity. This
technique serves as a powerful tool for the rapid detection and profiling of lectins.
In the post-genomic era, numerous studies of the cellular
functions of proteins have been performed in order to
elucidate the nature of biological processes and to discover
novel therapeutic agents. Among the myriad of proteins
found in nature, those that bind carbohydrates (lectins) are
extremely important since their specific interactions with
glycans play key roles in a wide variety of physiological
and pathogenic processes.1 Analysis of the human genomic
sequence suggests that about 100 gene products are lectins.2
However, the binding properties of only about half of these
proteins have been relatively well characterized; the functions
of the remaining proteins in this family have not yet been
evaluated. To fully understand the biological implications
of carbohydrate-protein interactions in living organisms and
to help develop novel carbohydrate-based drugs, it is
imperative to identify and characterize these unknown lectins
and determine their binding specificities. Moreover, a
quantitative assessment of the variations in lectins expressed
in different cell states and cell types is essential for
elucidating their biological roles and identifying disease-
related markers.
A new carbohydrate microarray-based technology has
emerged recently for the rapid analysis of carbohydrate-
protein recognition events.3,4 This technology has the po-
tential for broad research fields, such as functional glycomics,
diagnosis of diseases, and drug discovery. On the other hand,
several chemical strategies have been exploited to detect
carbohydrate-related proteins in cells, including carbohydrate-
(3) (a) Lee, M.-r.; Shin, I. Org. Lett. 2005, 7, 4269. (b) Lee, M.-r.; Shin,
I. Angew. Chem., Int. Ed. 2005, 44, 2881. (c) Park, S.; Lee, M.-r.; Pyo,
S.-J.; Shin, I. J. Am. Chem. Soc. 2004, 126, 4812. (d) Nimrichter, L.; Gargir,
A.; Gortler, M.; Altstock, R. T.; Shtevi, A.; Weisshaus, O.; Fire, E.; Dotan,
N.; Schnaar, R. L. Glycobiology 2004, 14, 197. (e) Disney, M. D.; Seeberger,
P. H. Chem. Biol. 2004, 11, 1701. (f) Blixt, O.; Head, S.; Mondala, T.;
Scanlan, C.; Huflejt, M. E.; Alvarez, R.; Bryan, M. C.; Fazio, F.; Calarese,
D.; Stevens, J.; Razi, N.; Stevens, D. J.; Skehel, J. J.; van Die, I.; Burton,
D. R.; Wilson, I. A.; Cummings, R.; Bovin, N.; Wong, C.-H.; Paulson, J.
C. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 17033. (g) Park, S.; Shin, I.
Angew. Chem., Int. Ed. 2002, 41, 3180. (h) Houseman, B. T.; Mrksich, M.
Chem. Biol. 2002, 9, 443. (i) Fukui, S.; Feizi, T.; Galustian, C.; Lawson,
A. M.; Chai, W. Nat. Biotechnol. 2002, 20, 1011. (j) Wang, D.; Liu, S.;
Trummer, B. J.; Deng, C.; Wang, A. Nat. Biotechnol. 2002, 20, 275.
(1) (a) Roth, J. Chem. ReV. 2002, 102, 285. (b) Bertozzi, C. R.; Kiessling,
L. L. Science 2001, 291, 2357.
(2) Drickamer, K.; Taylor, M. E. Genome Biol. 2002, 3, 1.
10.1021/ol0523188 CCC: $30.25
© 2005 American Chemical Society
Published on Web 11/01/2005

binding proteins, carbohydrate transporters and carbohydrate-
processing enzymes. For example, biotinylated photoaffinity
labels have been developed to detect lectins and a fructose
or glucose transporter.5,6 In addition, biotinylated mechanism-
based inhibitors have also been designed and utilized to
identify glycosidases.7,8 However, difficulties are still en-
countered when attempts are made to apply this methodology
to the identification of unknown cellular lectins. Thus,
although these chemical approaches have been used suc-
cessfully to identify carbohydrate-related proteins, more
efficient methods that are suitable for a systematic study of
carbohydrate-binding proteins in cells are required. Toward
this end, we have designed an approach for readily detecting
and profiling cellular lectins, which relies on trifunctional
probes that contain a photoreactive group, a reporter tag and
a carbohydrate ligand. These probes, prepared by a highly
efficient solid-phase methodology, play a key role in a new
technique to detect lectins in cells.
The initial step in the general route used to synthesize
monovalent carbohydrate probes involves attachment of the
safety-catch linker-coupled acid 1 to a PS-PEG amine resin
to produce 2 (Scheme 1). Since coupling of succinic acid or
Scheme 1. Reaction Sequence for Preparation of Monovalent
Carbohydrate Probesa
The general structure and strategy for preparation of
trifunctional probes developed in this effort are shown in
Figure 1. In the procedure, a photoreactive group is
Figure 1. Strategy for preparation of trifunctional probes.
incorporated into the probe for irreversible, covalent labeling
of weakly bound lectins, in particular, cell-surface lectins
and a reporter (biotin or fluorophore) is inserted to detect
and quantify the labeled protein. The general synthetic
strategy involves an initial coupling step to covalently link
the carbohydrate ligand and photoreactive group to a solid
support, and a final step in which the desired probe is
released from the resin by reaction with an amine-linked
reporter tag after activation of a safety-catch linker.9
a Abbreviations: Boc ) tert-butyloxycarbonyl; BOP ) 1-benzo-
triazolyloxytris(dimethylamino)phosphonium hexafluorophosphate;
Fmoc ) 9-fluorenylmethoxycarbonyl; HOBt ) 1-hydroxybenzo-
triazole; NEM ) N-ethylmorpholine; NMP ) N-methylpyrrolidi-
none; PS-PEG ) polystyrene-polyethyleneglycol; TFA ) trifluoro-
acetic acid.
(4) For reviews, see: (a) Shin, I.; Park, S.; Lee, M.-r. Chem. Eur. J.
2005, 11, 2894. (b) Shin, I.; Cho, J. W.; Boo, D. W. Comb. Chem. High
Throughput Screening 2004, 7, 565. (c) Feizi, T.; Chai, W. Nat. ReV. Mol.
Cell. Biol. 2004, 5, 582. (d) Feizi, T.; Fazio, F.; Chai, W.; Wong, C.-H.
Curr. Opin. Struct. Biol. 2003, 13, 637. (e) Ortiz Mellet, C.; Garcia
Ferna´ndez, J. M. ChemBioChem 2002, 3, 819. (f) Love, K. R.; Seeberger,
P. H. Angew. Chem., Int. Ed. 2002, 41, 3583.
(5) (a) Hatanaka, Y.; Kempin, U.; Park, J.-P. J. Org. Chem. 2000, 65,
5639. (b) Ilver, D.; Arnqvist, A. O¨ gren, J.; Frick, I.-M.; Kersulyte, D.;
Incecik, E. T.; Berg, D. E.; Covacci, A.; Engstrand, L.; Bore´n, T. Science
1998, 279, 373. (c) Lee, R. T.; Lee, Y. C. Biochemistry 1986, 25, 6835.
(6) (a) Romaniouk, A. V.; Silva, A.; Feng, J.; Vijay, I. K. Glycobiology
2004, 14, 301. (b) Tsai, C.-S.; Li, Y.-K.; Lo, L.-C. Org. Lett. 2002, 4, 3607.
(c) Kipp, H.; Kinne, R. H.; Lin, J.-T. Anal. Biochem. 1997, 245, 61. (d)
Shailubhai, K.; Illeperuma, C.; Tayal, M.; Vijay, I. K. J. Biol. Chem. 1990,
265, 14105.
succinic anhydride to the safety-catch linker-connected resin
was inefficient, a preformed handle strategy was employed.10
After removal of the t-Bu group in 2, coupling to 4,7,10-
trioxa-1,13-tridecanediamine (3) was performed to generate
the amine terminated resin 4. Reaction of 4 with the
benzophenone (BP)-conjugated serine derivative 5 gave 6.
Sequential Boc deprotection, reaction with succinic anhy-
(7) (a) Yang, J.; Dowden, J.; Tatiboue¨t, A.; Hatanaka, Y.; Holman, G.
D. Biochem. J. 2002, 367, 533. (b) Hashimoto, M.; Yang, J.; Holman, G.
D. ChemBioChem 2001, 2, 52.
(8) Activities of carbohydrate-processing enzymes were also analyzed
by MS technology using biotinylated carbohydrate probes. Gerber, S. A.;
Scott. C. R.; Turecek, F.; Gelb, M. H. Anal. Chem. 2001, 73, 1651.
(9) (a) Backer, B. J.; Ellman, J. A. J. Am. Chem. Soc. 1994, 116, 11171.
(b) Backer, B. J.; Virgilio, A. A.; Ellman, J. A. J. Am. Chem. Soc. 1996,
118, 3055.
(10) Wang, G. T.; Li, S.; Wideburg, N.; Krafft, G. A.; Kempf, D. J. J.
Med. Chem. 1995, 38, 2995.
5478
Org. Lett., Vol. 7, No. 24, 2005

dride, and coupling with 2-aminoethyl R-L-fucopyranoside
or 2-aminoethyl R-D-mannopyranoside yielded 7. Finally,
after activation of the safety-catch linker in 7 with iodo-
acetonitrile, reaction with Cy3-tether-NH2 afforded the
monovalent carbohydrate probes 8 and 9 (Figure 2).
Scheme 2. Reaction Sequence for Preparation of Trivalent
Carbohydrate Probes
Figure 2. Structures of synthesized trifunctional probes.
In general, interactions between monovalent carbohydrates
and proteins are weak. Consequently, multivalent carbohy-
drate probes are required in order to facilitate labeling of
the lectins in cells.11 In the sequence we have developed for
preparation of trivalent carbohydrate probes, the tether-
conjugated amine resin 4 was reacted with FmocLys(Boc)OH
to give 10 (Scheme 2). Fmoc deprotection in 10 and
subsequent coupling with FmocLys(Fmoc)OH produced
the selectively protected trisamine resin 11. The amine
formed by removal of the Boc group in 11 was then coupled
with 5 to produce 12. Subjection of 12 to the chemical
processes used to transform 6 to monovalent carbohydrate
probes resulted in formation of trivalent probes 14-18
(Figure 2).
HPLC analyses of the monovalent and trivalent carbohy-
drate probes showed that the solid-phase synthetic routes
were highly efficient. It is worth mentioning that in contrast
to monovalent probes, which were easily separated from
excess Tag-tether-NH2 by HPLC, most of the trivalent probes
were difficult to separate from excess Tag-tether-NH2. In
these cases, the probes were purified after they were first
reacted with Boc2O (Figure 3).
The efficiency of protein labeling by the monovalent and
trivalent carbohydrate probes were evaluated. The fucose (8,
15) and mannose (9, 17) probes containing Cy3 were
preincubated with A. aurantia (AA) and ConA, respectively,
for 20 min, and the resulting mixtures were irradiated for
5-30 min at 4 °C. Analysis of fluorescence intensities of
the proteins after gel separation showed that the amounts of
protein labeling by the trivalent probes 15 and 17 were higher
than those induced by the monovalent counterparts 8 and
9.12 These observations demonstrate that the efficiencies for
protein labeling by the trivalent probes are greater than those
by the monovalent probes.
(11) Mannen, M.; Choi, S.-K.; Whitesides, G. M. Angew. Chem., Int.
Ed. 1998, 37, 2754.
(12) See the Supporting Information.
Org. Lett., Vol. 7, No. 24, 2005
5479

Figure 3. HPLC chromatograms (a) after release of 15 by Cy3-
tether-NH2 and (b) after reaction of the released mixture with
Boc2O.
Figure 4. (a) Quantitation of AA labeling under two different
conditions (lane 1, Coomassie blue staining; lane 2, Cy3 fluores-
cence detection; lane 3, Cy5 fluorescence detection): BSA, bovine
serum albumin; OVA, ovalbumin; PT, P. tetrangonolobus; TV, T.
Vulgaris. (b) Selective labeling of AA by 15 in cell lysates (left,
fluorescence image; right, silver staining).
Next, the detection sensitivity of the fluorescence assay
for protein labeling was compared with that obtained by using
the Western blotting, silver, and Coomassie blue staining
techniques. Comparison of the results of labeling experi-
ments, performed by using AA with Biotin-tri-Fuc (14) and
Cy3-tri-Fuc (15), showed that the detection limit (0.4 ng of
AA) for the fluorescence assay was superior to that for the
Western blotting (3.7 ng of AA), silver (11 ng of AA), and
Coomassie blue staining (100 ng of AA) techniques.12
Investigations have been carried out to test application of
the trifunctional probes for rapid determination of the amount
of specific lectins present in two different cell states. To
evaluate quantitative aspects of the technique, the mixture
of five proteins (BSA, OVA, AA, PT, and TV) containing
800 ng of AA or 200 ng of AA was separately incubated
with Cy3-tri-Fuc (15) or Cy5-tri-Fuc (16), respectively, and
then irradiated. The mixtures were combined and then
subjected to protein gel separation. The fucose probes
selectively labeled AA, and the fluorescence intensities of
Cy3 and Cy5 were found to have a ratio of 4:1 (Figure 4a).13
This result shows that the trifunctional probes can be used
to accurately and rapidly quantify lectins. In addition, this
methodology was employed in the detection of an exogenous
lectin in E. coli lysates to demonstrate its potential applica-
tions in large-scale proteomic studies. Accordingly, cell
lysates (0, 0.25, and 0.5 µg of proteins) containing AA (60
ng) were incubated with 15 and then irradiated. As shown
in Figure 4b, AA was selectively labeled by the probe.
In conclusion, an efficient method for synthesizing doubly
labeled carbohydrate probes to detect cellular lectins has been
developed. This synthetic pathway is applicable to the
preparation of various chemical probes, including peptides
and small molecules.14 The results show that probes prepared
using this methodology label proteins in a highly sensitive
and selective manner. Therefore, they may be useful in the
study of poorly characterized lectins predicted from genomic
studies. In addition, the probes can be also used for profiling
carbohydrate-binding proteins in different cell types or cell
states.
Acknowledgment. This work was supported by grants
from the NRL program and the SRC program of MOST/
KOSEF (R11-2000-070-04003-0).
Supporting Information Available: Synthesis of car-
bohydrate probes and labeling experiments. This material is
available free of charge via the Internet at http://pubs.acs.org.
OL0523188
(13) A 1:1 mixture of Cy3 and Cy5 exhibits a fluorescence intensity
ratio of 1:1. For additional quantitative protein labeling experiments, see
the Supporting Information.
(14) (a) Videlock, E. J.; Chung, V. K.; Mohan, M. A.; Strok, T. M.;
Austin, D. J. J. Am. Chem. Soc. 2004, 126, 3730. (b) Kumar, V.; Aldrich,
J. V. Org. Lett. 2003, 5, 613.
5480
Org. Lett., Vol. 7, No. 24, 2005