Microarrays of synthetic heparin oligosaccharides

Article abstract of DOI:10.1021/ja057584v

We present the first preparation of microarrays containing synthetic heparin oligosaccharides in order to elucidate the heparin-protein interactions involved in a variety of biological processes. For this purpose, we have developed a novel linker strategy

Full text of DOI:10.1021/ja057584v

Published on Web 02/08/2006
Microarrays of Synthetic Heparin Oligosaccharides
Jose L. de Paz, Christian Noti, and Peter H. Seeberger*
Laboratory for Organic Chemistry, Swiss Federal Institute of Technology (ETH) Zurich,
Wolfgang-Pauli-Str. 10, HCI F315, 8093 Zurich, Switzerland
Received November 7, 2005; E-mail: seeberger@org.chem.ethz.ch
Carbohydrate microarrays1 are a powerful platform to screen
interactions involving these molecules in a high-throughput manner.2
The chip-based format offers important advantages over classical
methods, such as the ability to screen several thousand binding
events on a single glass slide and the miniscule amounts of both
analyte and ligand required for one experiment.
One of the most complex classes of carbohydrates, heparin-like
glycosaminoglycans (HLGAGs), plays a key role in regulating
biological processes, including growth factor interactions, virus
entry, and angiogenesis by binding to a host of proteins.3 However,
the structure-function relationships of HLGAGs are very poorly
understood due to the chemical complexity and heterogeneity of
this type of biopolymer. The use of microarrays of synthetic heparin
oligosaccharides can substantially improve the understanding of
heparin-protein interactions, opening an opportunity for the
discovery of novel therapeutic interventions for a variety of disease
states.
Figure 1. General strategy for the preparation of microarrays containing
synthetic heparin oligosaccharides.
Scheme 1. Preparation of Sulfated Monosaccharides 4 and 9a
Herein, we report the creation and use of microarrays containing
synthetic heparin oligosaccharides. A novel linker chemistry is
compatible with the protecting-group manipulations required for
the synthesis of HLGAGs following solution phase or solid phase
assembly. The covalently attached, defined oligosaccharides can
be used for the rapid analysis of HLGAG-protein interactions as
exemplified by incubation with acidic and basic fibroblast growth
factors (FGF-1 and FGF-2), two important heparin-binding pro-
teins.4
Initially, we designed a suitable linker strategy to attach synthetic
heparin molecules to a chip surface. Synthetic oligosaccharides
obtained by automated solid phase synthesis are released as terminal
pentenyl glycosides5 (Figure 1) via olefin cross-metathesis. We
elected an amine-terminated thiol to elongate the n-pentenyl
glycosides before sulfated oligosaccharides are generated and
printed onto glass slides.
a Reagents and conditions: (a) HS(CH2)2NHZ, AIBN, THF, 75 °C, 82%;
(b) LiOH, H2O2, then KOH, 88%; (c) SO3‚Py, Py, 85%; H2, Pd/C, 95%;
(d) HS(CH2)2NHZ, AIBN, THF, 75 °C, 95%; (e) LiOH, H2O2; KOH;
KHSO5, MeOH, 88%; (f) PMe3, THF, NaOH, 96%; (g) SO3‚Py, Py, 84%;
H2, Pd/C, 99%.
Two monosaccharides 1 and 5 (Scheme 1) served as models to
develop and optimize the linker chemistry. First, radical extension
of the pentenyl moiety using 2-(benzyloxycarbonylamino)-1-
ethanethiol was worked out. Thermal activation6 at 75 °C in the
presence of a catalytic amount of AIBN gave the best yields among
a host of reaction conditions that were tested. Treatment of amine-
terminated 2 and 6 with lithium hydroperoxide and then KOH
hydrolyzed the acyl and methoxycarbonyl groups. Simultaneous
oxidation of the sulfide afforded 3 and 77 in good yield. Introduction
of the sulfate groups that are frequent in heparin oligosaccharides
and are placed following the assembly process was achieved by
treatment with the SO3‚Py complex. Global deprotection was
executed by hydrogenolysis to give the sulfated monosaccharide
4. To demonstrate that the reduction of azide protecting groups
and simultaneous O- and N-sulfation followed by hydrogenolysis
can be carried out, glucosamine 7 was transformed via Staudinger
reduction to afford 9 in high yield. Thereby, a method to equip
synthetic heparin oligosaccharides for covalent immobilization onto
chip surfaces was established.
A series of fully protected di-, tetra-, and hexasaccharides
containing the GlcNSO3(6-OSO3)-IdoA(2-OSO3) repeating unit of
the major sequence of heparin was derived as part of our ongoing
program concerned with the automated solid phase synthesis of
heparin oligosaccharides.8 Application of the new linker strategy
furnished the sulfated oligosaccharides 10, 11, and 12 (Figure 2).
With a straightforward route to procure amine-functionalized
sugars at hand, the challenge to covalently attach these molecules
on a suitable glass surface was addressed. Amine-reactive CodeLink
slides that are coated with a hydrophilic polymer containing
N-hydroxysuccinimide esters performed best due to the three-
dimensional nature of the polymer. The immobilization chemistry
was initially tested using manual spotting of 0.5 µL of heparin
oligosaccharides 10, 11, and 12 at concentrations ranging from 5
9
2766
J. AM. CHEM. SOC. 2006, 128, 2766-2767
10.1021/ja057584v CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
Following incubation with FGF, primary and secondary antibod-
ies were utilized to detect any bound protein on the slides using a
standard fluorescent slide scanner. FGF-2 bound best to 11 and 12
based on the highest fluorescence intensities, while monosaccharides
were not bound (see Supporting Information). Similar results were
obtained for FGF-1, but interestingly, monosaccharide 4 exhibited
a spot intensity comparable to that of longer oligosaccharides
(Figure 3). The presence of a 2,4-O-sulfation pattern, not found in
nature, may be responsible for this result. Also, this result illustrates
the possibility to employ the heparin microarrays to discover
inhibitors for heparin-protein interactions.
Figure 2. Heparin oligosaccharides 10, 11, and 12 ready for immobilization
on a chip surface.
In conclusion, we have developed a new method for the
preparation of microarrays displaying synthetic heparin oligosac-
charides derived by solution and solid phase assembly methods.
Strategic placement of an orthogonally protected amine linker was
key to the success of the array construction. The potential of the
new methodology was demonstrated by probing the carbohydrate
affinity of two heparin-binding proteins, FGF-1 and FGF-2, that
are implicated in the development and differentiation of several
tumors. On the basis of the method disclosed here, the construction
of diverse heparin microarrays that can be used to rapidly screen
heparin-protein interactions has become possible. Heparin arrays
are expected to fundamentally impact the establishment of structure-
activity relationships for heparin sequences.
Acknowledgment. This research was supported by ETH Zurich
and the European Commission (Marie Curie Fellowship for J.L.P.).
We thank Dr. Jens Sobek (Functional Genomics Center, Zurich)
for assistance in the construction of microarrays.
Supporting Information Available: Experimental procedures for
the production of the heparin microarrays, synthesis of compounds 4
and 9, and the complete ref 2d. This material is available free of charge
via the Internet at http://pubs.acs.org.
Figure 3. Top: Microarray after incubation with FGF-1. Bottom: Fluo-
rescence signal observed for each arrayed carbohydrate binding to FGF-1
at 500 µM. Sodium phosphate buffer served as a negative control.
References
(1) For reviews, see: (a) Love, K. R.; Seeberger, P. H. Angew. Chem., Int.
Ed. 2002, 41, 3583-3586. (b) Feizi, T.; Fazio, F.; Chai, W.; Wong, C.-
H. Curr. Opin. Struct. Biol. 2003, 13, 637-645. (c) Shin, I.; Park, S.;
Lee, M. Chem.sEur. J. 2005, 11, 2894-2901.
mM to 50 µM. After incubation and quenching of the remaining
activated esters, the slides were ready for screening experiments
using pure growth factor FGF-1 or FGF-2. These important
members of a family of heparin-binding proteins are involved in
developmental and physiological processes, including cell prolifera-
tion, differentiation, morphogenesis, and angiogenesis.9 After
removing unbound protein by washing, slides were subsequently
incubated with anti-human FGF polyclonal antibodies. Finally, the
human FGF captured by the heparin microarray was visualized by
using a secondary antibody labeled with an Alexa 546 dye. Strongly
fluorescent signals at tetra- and hexasaccharide positions were in
agreement with the minimal structural requirements to bind both
FGF-1 and FGF-2.10
Miniaturization of the spotting process using an arraying robot
to construct heparin microarrays for high-throughput experiments
was now simple. Oligosaccharides 4, 9, 10, 11, and 12 along with
2′-aminoethyl-2-acetamido-R-D-glucopyranoside (GlcNAc) and 2′-
aminoethyl-â-D-galactopyranoside (Gal)11 were used in concentra-
tions ranging from 2 mM to 250 µM (Figure 3). The robot delivered
1 nL of carbohydrate-containing solutions to create spots with a
diameter of ∼200 µm. All samples were printed in 15 replicates to
generate an array of 480 spots (Figure 3).
(2) For examples, see: (a) Wang, D.; Liu, S.; Trummer, B. J.; Deng, C.;
Wang, A. Nat. Biotechnol. 2002, 20, 275-281. (b) Fukui, S.; Feizi, T.;
Galustian, C.; Lawson, A. M.; Chai, W. Nat. Biotechnol. 2002, 20, 1011-
1017. (c) Disney, M. D.; Seeberger, P. H. Chem. Biol. 2004, 11, 1701-
1707. (d) Blixt, O.; et al. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 17033-
17038. (e) Ratner, D. M.; Adams, E. W.; Su, J.; O’Keefe, B. R.; Mrksich,
M.; Seeberger, P. H. ChemBioChem 2004, 5, 379-383. (f) Disney, M.
D.; Seeberger, P. H. Chem.sEur. J. 2004, 10, 3308-3314. (g) Lee, M.;
Shin, I. Org. Lett. 2005, 7, 4269-4272.
(3) (a) Capila, I.; Linhardt, R. J. Angew. Chem., Int. Ed. 2002, 41, 390-412.
(b) Noti, C.; Seeberger, P. H. Chem. Biol. 2005, 12, 731-756.
(4) Conrad, H. E. Heparin-Binding Proteins; Academic Press: San Diego,
CA, 1998.
(5) Plante, O. J.; Palmaci, E. R.; Seeberger, P. H. Science 2001, 291, 1523-
1527.
(6) Buskas, T.; Soderberg, E.; Konradsson, P.; Fraser-Reid, B. J. Org. Chem.
2000, 65, 958-963.
(7) In the case of 6, additional treatment with oxone was necessary for the
complete conversion of sulfide into sulfone 7.
(8) (a) Orgueira, H. A.; Bartolozzi, A.; Schell, P.; Litjens, R. E. J. N.; Palmacci,
E. R.; Seeberger, P. H. Chem.sEur. J. 2003, 9, 140-169. (b) Noti, C.;
De Paz, J. L.; Seeberger, P. H. Unpublished results.
(9) Pellegrini, L. Curr. Opin. Struct. Biol. 2001, 11, 629-634.
(10) Raman, R.; Sasisekharan, V.; Sasisekharan, R. Chem. Biol. 2005, 12, 267-
277.
(11) Monosaccharide 4 without sulfate groups was used as an additional
negative control (see Supporting Information).
JA057584V
9
J. AM. CHEM. SOC. VOL. 128, NO. 9, 2006 2767