A Supramolecular polymer as a self-assembling polyvalent scaffold

Full text of DOI:10.1002/anie.200900143

Angewandte
Chemie
DOI: 10.1002/anie.200900143
Supramolecular Polyvalency
A Supramolecular Polymer as a Self-Assembling Polyvalent Scaffold**
Marion K. Mꢀller and Luc Brunsveld*
Adequate recognition and strong affinities in biological
systems are mainly the result of polyvalent interactions.[1]
The polyvalent recognition elements are typically supra-
molecular assemblies in a self-reorganizing biological envi-
ronment, such as the cell membrane. Induction or inhibition
of these types of interactions provides entry into targeting
diseases and infections related to them.[1] Small-molecule
inhibitors usually do not feature the affinity required for
effective competition. Therefore, synthetic polyvalent inhib-
itors have been generated, with a large molecular diversity in
their scaffolds.[2] Nevertheless, the question of how to design
and synthesize a macromolecule that optimally mimics and
matches the arrangement of its multiple targets[3] in the
membrane and adapts itself to the dynamics of these targets,
is essentially still unsolved. Ligand placing, polymer folding,
and active adjustment of ligand positioning are topics that
need to be addressed for the generation of polyvalent systems
that respond to the dynamic rearrangement of the biological
interaction partners. Synthetic supramolecular systems are
self-assembling and dynamic scaffolds that might feature
these properties. Vesicular architectures[4] and pseudopoly-
rotaxanes[5] have been shown to be promising supramolecular
architectures in this respect, with applications in, for example,
biomedical engineering[4] and bacterial detection.[5,6] There is,
however, a need for new self-assembling synthetic systems,
with diverse topology, composition, and assembly dynamics.
Herein we show for the first time that synthetic supramolec-
ular polymers[7] are ideal polyvalent scaffolds to target
polyvalent biological systems. The design, synthesis, and
biological evaluation of a biocompatible, auto-fluorescent,
polyvalent columnar supramolecular polymer are presented.
The discotic monomers reversibly assemble into a columnar
polymer at low concentrations in water and can be decorated
with ligands to target carbohydrate–lectin polyvalent inter-
actions.[8] The ligand density in the polymer can be controlled
by reversible exchange of monomers. This supramolecular
polymer features strong binding to bacteria, is easily detected
by fluorescence microscopy, and the control over its ligand
density allows easy optimization of bacterial clustering
(Figure 1).
Figure 1. Schematic representation of the formation of polyvalent
columnar supramolecular polymers. A) ratio of monomers controls the
composition of the supramolecular polymer, B) binding of bacteria,
which results in clustering and detection of the bacteria.
Disc-shaped molecule 1 (Scheme 1) assembles into col-
umnar supramolecular polymers,[9] at low concentrations in
water[10] and other polar media.[11] The decoration of the
supramolecular polymer with a shell of solubilizing ethylene
glycol chains ensures water-solubility and prevents unspecific
interactions with biological matter. Upon polymerization,
these molecules become highly fluorescent. Based on 1,
compound 2 was designed featuring three selectively intro-
duced azide functionalities at the periphery of the molecule,
envisaged to provide a flexible platform for modifications
with biological ligands (Scheme 1).[12] Compound 2 was
synthesized in a multistep approach in a convergent fashion
(see Supporting Information for synthesis). Propargyl-a-d-
mannopyranoside was selected as the ligand for the inter-
action with bacterial lectins and prepared in three steps using
the tin tetrachloride catalyzed glycosidation procedure.[13] For
the threefold modification of 2 with this mannose derivative,
standard conditions of 2 mol% CuSO4, 5 mol% of sodium
ascorbate, and 15 equivalents of 1-propargyl-mannose per
azide residue in a water/tBuOH mixture were applied.[14] In
contrast to many previous reports on the copper-catalyzed
azide–alkyne cycloaddition for sugars,[15] the modification of 2
to 3 by this reaction was relatively slow. Approximately 30%
of the azide functionalities reacted rapidly within the first
hours, but the complete functionalization of all groups
required longer times (up to four weeks). Apparently the
polymerization of 2/3 under the reaction conditions in water
[*] Prof. Dr. L. Brunsveld
Laboratory of Chemical Biology,
Department of Biomedical Engineering, TU Eindhoven
Eindhoven (The Netherlands)
Fax: (+31)40-247-8367
E-mail: l.brunsveld@tue.nl
Homepage: http://www.bmt.tue.nl/cb
M. K. Mꢀller, Prof. Dr. L. Brunsveld
Department of Chemical Biology, Max Planck Institute for Molecular
Physiology and Chemical Genomics Centre
Otto-Hahn-Strasse 15, 44227 Dortmund (Germany)
[**] L.B. was supported via a Sofja Kovalevskaja Award of the Alexander
von Humboldt Foundation and the BMBF. We kindly thank Prof.
Paul Orndorff (North Carolina State University) for providing
bacterial strains ORN178 and ORN208.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200900143.
Angew. Chem. Int. Ed. 2009, 48, 2921 –2924
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2921

Communications
Figure 2. Microscopy pictures in a) brightfield and b) fluorescence
(lex =360 nm, lem =490 nm) mode on E. coli incubated with A) water,
B) inert discotic 1, C) mannose discotic 3.
demonstrate the absence of unspecific binding of supra-
molecular polymers of 1.
The specificity of the interaction of 3 with the mannose-
binding FimH receptors on the bacterial surface was eval-
uated through binding studies on two E. coli strains, ORN178
and ORN208.[17] These strains differ in their mannose-bind-
ing-properties owing to the over-expression or suppression of
the FimH receptor, respectively. Mixtures of 3 and ORN178
featured a strong fluorescence response where the bacteria
were located, whereas no binding of 3 to ORN208 could be
detected (Figure 2 in the Supporting Information). The
supramolecular polymers thus bind to the bacteria through
the mannose–FimH receptor interaction selectively and do
not induce unspecific binding.
The polyvalent effect of the supramolecular polymers was
evaluated on dilute bacteria samples, with the bacteria
present in non-aggregated state. When these samples were
incubated with 3 the bacteria clustered and a colocalized
fluorescence of the bacteria could be observed. Control
experiments with either 1 or water showed dispersed bacteria
only (Figure 3). Apparently the supramolecular polymers are
Scheme 1. Discotic compounds: 1 with inert glycol side chains, 2 with
azide functionalities for attachment of ligands, 3 mannose functional-
ized.
results in molecular crowding (see below) of the reactive
functional groups.[16] The dynamic rearrangement of the
discotic monomers nevertheless allows full conversion of
the free azide groups over time, probably at the chain-ends of
the supramolecular polymer. The pure product 3 could be
isolated after size-exclusion chromatography, separating the
excess of mannose, in 80% yield.
All discotic compounds dissolve readily in water and form
supramolecular polymers (Kass 1 = 108 mÀ1).[10] The binding of
polyvalent supramolecular polymers built up out of 1 or 3 and
mixtures thereof to bacteria was investigated by microscopy
studies with the E. coli strain BL 21a. The bacteria were
cultivated in LB (lysogeny broth) media, washed, resus-
pended and incubated with either nonfunctionalized discotic
1, mannosylated discotic 3, or with water, for 1 h at room
temperature.[6] The total concentration of discotic monomers
1 and 3 was kept dilute at 10À7 m. After washing and mounting
on glass slides, binding to the bacteria was evaluated with a
fluorescent microscope (Figure 2), taking advantage of the
strong auto-fluorescence of the discotic monomers when
present as supramolecular polymer.[10,11] Only when the
bacteria were incubated with mannose-modified discotic 3
could a strong fluorescence of the bacterial aggregates be
observed, colocalizing with the brightfield image. Both the
control experiments with nonfunctionalized scaffold 1 and
with water did not show fluorescence. These results show that
the supramolecular polymers of 3 bind to the bacteria and
Figure 3. Microscopy pictures in brightfield mode on dilute bacterial
samples incubated with a) 100% 1, b) 99/1 1/3, and c) 100% 3.
long enough to function as polyvalent cross-linking scaffolds,
similar to conventional polymers.[6] The supramolecular
polymers are furthermore most probably stabilized by the
polyvalent interaction with the bacteria, which increases the
degree of polymerization.
To evaluate the influence of ligand density, supramolec-
ular polymers consisting of mixtures of 1 and 3 were
generated and similarly evaluated at dilute bacteria concen-
trations. These supramolecular copolymers induced bacterial
aggregation at all mixture compositions evaluated (Figure 3
and Supporting Information Figure 3). A functional-mono-
2922 www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 2921 –2924

Angewandte
Chemie
mer percentage as low as 1% of 3 still resulted in the
formation of bacterial aggregates. Evidently, only a limited
number of mannose functionalities at the periphery of the
supramolecular polymer are required to induce clustering of
the bacteria around the columns. Furthermore, the supra-
molecular polymers are sufficiently long to bridge the
distances between the few functional monomers 3 to induce
bacterial clustering (Figure 1). Interestingly, the aggregates
observed for the supramolecular polymers containing pre-
dominantly nonfunctionalized discotic 1 were typically larger
than those induced by polymers consisting of 3 only. This
result indicates that optimal binding to bacteria occurs when
only a limited number of mannose-functionalized monomers
is present, surrounded by nonfunctionalized monomers.
“Overcrowding” of the supramolecular polymer with ligands,
thus results in a sub-optimal binding. The supramolecular
nature of the polymer allows for easy adjustments and
optimization of this monomer composition.
stronger than the reference compound (24.9 per molecule 3),
showing the polyvalent inhibition of the supramolecular
polymers. Typically, small trivalent ligands do not show
significant valency-corrected enhancement effects to higher
valency in specific lectin–ligand assays, such as this ELLA
assay, as they cannot span the distances between the binding
sites on the ConA.[20] For these small scaffolds, higher
valencies improve the total potency, but do not improve
valency-corrected potency, which results merely from the
“cluster glycoside effect”.[18,21] The supramolecular polymers
formed by 3 allow effective polyvalent binding up to high
valencies and show high potency per ligand. Apparently,
because of the polymeric nature of 3, the compounds are
capable of spanning the distance between the mannose
binding sites on ConA.
With the mannose–lectin interaction as an example, we
have shown that columnar supramolecular polymers are
effective polyvalent scaffolds for binding and detecting
bacteria. The simple generation of polymers with different
monomer compositions showed that decreasing the amount
of mannose component actually enhanced bacterial aggrega-
tion, by reducing steric crowding. The self-assembly into
polymers enhances the potency of the monomers significantly
more than is to be expected on the basis of the cluster
glycoside effect alone. Supramolecular polymers are ideal
systems to generate polyvalent architectures for binding and
modulation of biological interactions. The reversible self-
assembly of monomers into polymers provides control over
ligand density, polymeric architecture, and environmental
response to the biological interaction partner, not accessible
with covalent polymeric systems.
A competition experiment between 3 and mannose on
binding to bacteria was performed to examine the strength of
the polyvalent binding of the supramolecular polymer. To
allow the most effective competition by the mannose, the
bacteria were first incubated with different concentrations of
mannose for ten minutes, after which discotic 3 was added
Figure 4. Microscopy pictures in fluorescence mode (lex =360 nm,
lem =490 nm) of bacteria incubated with 3 (10À7 m) and mannose as a
competitor. a) pure 3 (10À7 m); b) 3 (10À7 m) and mannose
(3ꢁ10À4 m); c) 3 (10À7 m) and mannose (3ꢁ10À1 m).
Received: January 9, 2009
Published online: March 17, 2009
Keywords: carbohydrates · fluorescence · polyvalency ·
.
self-assembly · supramolecular polymers
(Figure 4 and Supporting Information Figure 4). The bacteria
remained fluorescent over the whole mannose concentration
regime studied (up to a 106 fold excess of mannose).
Apparently the supramolecular polymers feature highly
efficient polyvalent binding to the bacteria in this assay.
To quantify the polyvalent binding of the supramolecular
polymers of 3, an enzyme linked lectin assay (ELLA) was
performed.[18] Mannose-coated polyvalent structures are
known to competitively inhibit Concanavalin A (ConA)
binding to the yeast cell surface receptor mannan. ConA is
a tetramer at neutral pH value, containing four spatially well
separated binding sites (6.5 nm) for oligosaccharides.[19]
Experiments using horseradish peroxidase labeled ConA
(HRP-Con A) as the lectin and yeast mannan as the surface-
fixed-ligand were carried out in 96-well plates. After pre-
incubation with different concentrations of 3, or methyl a-d-
mannopyranoside as the reference compound, binding of
HRP-ConA to mannan was measured photospectrometri-
cally (see Supporting Information). The IC50 value for the
methyl glycoside was around 3000 mm in our assay and the
IC50 value for 3 around 120 mm (360 mm, valency corrected).
The relative valency-corrected binding of 3 was 8.3 times
[1] M. Mammen, S.-K. Choi, G. M. Whitesides, Angew. Chem. 1998,
110, 2908 – 2953; Angew. Chem. Int. Ed. 1998, 37, 2754 – 2794.
[2] S.-K. Choi, Synthetic Multivalent Molecules, Wiley-VCH, New
York, 2004.
[3] P. I. Kitov, J. M. Sadowska, G. Mulvey, G. D. Arnstrong, H. Ling,
N. S. Pannu, R. J. Read, D. R. Bundle, Nature 2000, 403, 669 –
672.
[4] a) G. C. L. Wong, J. X. Tang, A. Lin, Y. Li, P. A. Janmey, C. R.
Safinya, Science 2000, 288, 2035 – 2039; b) G. Thoma, A. G.
Katopodis, N. Voelcker, R. O. Duthaler, M. B. Streiff, Angew.
Chem. 2002, 114, 3327 – 3330; Angew. Chem. Int. Ed. 2002, 41,
3195 – 3198; c) G. A. Silva, C. Czeisler, K. L. Niece, E. Beniash,
D. A. Harrington, J. A. Kessler, S. I. Stupp, Science 2004, 303,
1352 – 1355; d) B.-S. Kim, D.-J. Hong, J. Bae, M. Lee, J. Am.
Chem. Soc. 2005, 127, 16333 – 16337; e) J.-H. Ryu, E. Lee, Y.
Lim, M. Lee, J. Am. Chem. Soc. 2007, 129, 4808 – 4814; f) K.
Rajangam, M. S. Arnold, M. A. Rocco, S. I. Stupp, Biomaterials
2008, 29, 3298 – 3305.
[5] a) A. Nelson, J. M. Belitsky, S. Vidal, C. S. Joiner, L. G. Baum,
J. F. Stoddart, J. Am. Chem. Soc. 2004, 126, 11914 – 11922;
b) J. M. Belitsky, A. Nelson, J. D. Hernandez, L. G. Baum, J. F.
Stoddart, Chem. Biol. 2007, 14, 1140 – 1151.
Angew. Chem. Int. Ed. 2009, 48, 2921 –2924
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2923

Communications
[6] M. D. Disney, J. Zheng, T. M. Swager, P. H. Seeberger, J. Am.
Chem. Int. Ed. 2002, 41, 2596 – 2599; c) J. M. Baskina, C. R.
Bertozzi, QSAR Comb. Sci. 2007, 26, 1211 – 1219.
Chem. Soc. 2004, 126, 13343 – 13346.
[7] a) L. Brunsveld, B. J. B. Folmer, E. W. Meijer, R. P. Sijbesma,
Chem. Rev. 2001, 101, 4071 – 4097; b) P. Y. W. Dankers, E. W.
Meijer, Bull. Chem. Soc. Jpn. 2007, 80, 2047 – 2073.
[8] a) F. P. Schwarz, K. D. Puri, R. G. Bhat, A. J. Surolia, J. Biol.
Chem. 1993, 268, 7668 – 7677; b) D. K. Mandal, N. Kishore, C. F.
Brewer, Biochemistry 1994, 33, 1149 – 1156.
[9] L. Brunsveld, E. W. Meijer, A. E. Rowan, R. J. M. Nolte, Top.
Stereochem. 2003, 24, 373 – 423.
[10] L. Brunsveld, B. G. G. Lohmeijer, J. A. J. M. Vekemans, E. W.
Meijer, Chem. Commun. 2000, 2305 – 2306.
[11] a) L. Brunsveld, H. Zhang, M. Glasbeek, J. A. J. M. Vekemans,
E. W. Meijer, J. Am. Chem. Soc. 2000, 122, 6175 – 6182; b) L.
Brunsveld, B. G. G. Lohmeijer, J. A. J. M. Vekemans, E. W.
Meijer, J. Inclusion Phenom. Macrocyclic Chem. 2001, 41, 61 –
64.
[12] a) C. W. Tornøe, C. Christensen, M. Meldal, J. Org. Chem. 2002,
67, 3057 – 3064; b) V. V. Rostovtsev, L. G. Green, V. V. Fokin,
K. B. Sharpless, Angew. Chem. 2002, 114, 2708 – 2711; Angew.
[13] R. J. Kaufman, R. S. Sidhu, J. Org. Chem. 1982, 47, 4941 – 4947.
[14] W. D. Sharpless, P. Wu, T. V. Hansen, J. G. Lindberg, J. Chem.
Educ. 2005, 82, 1833 – 1836.
[15] R. J. Pieters, D. T. S. Rijkers, R. M. J. Liskamp, QSAR Comb.
Sci. 2007, 26, 1181 – 1190.
[16] Modification reactions of discotic 1 analogues in the molecular
dissolved state in dichloromethane proceed rapidly.
[17] S. L. Harris, P. A. Spears, E. A. Havell, T. S. Hamrick, J. R.
Horton, P. E. Orndorff, J. Bacteriol. 2001, 183, 4099 – 4102.
[18] J. J. Lundquist, E. J. Toone, Chem. Rev. 2002, 102, 555 – 578.
[19] J. H. Naismith, C. Emmerich, J. Habash, S. J. Harrop, J. R.
Helliwell, W. N. Hunter, J. Raferty, A. J. Kalb, J. Yariv, Acta
Crystallogr. Sect. D 1994, 50, 847 – 858.
[20] a) J. B. Corbell, J. J. Lundquist, E. J. Toone, Tetrahedron: Asym-
metry 2000, 11, 95 – 111; b) S. G. Gouin, E. Vanquelef, J. M.
Garcia Fernandez, C. O. Mellet, F.-Y. Dupradeau, J. Kovensky, J.
Org. Chem. 2007, 72, 9032 – 9045.
[21] Y. C. Lee, R. T. Lee, Acc. Chem. Res. 1995, 28, 321 – 327.
2924 www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 2921 –2924