Topological Defects in Hyperbranched Glycopolymers Enhance Binding to Lectins

Article abstract of DOI:10.1002/chem.201703432

Central scaffold topology and carbohydrate density are important features in determining the binding mechanism and potency of synthetic multivalent of poly- versus monodisperse carbohydrate systems against a model plant toxin (Ricinus communis agglutinin (RCA₁₂₀)). Lower densities of protein receptors favour the use of heterogeneous, polydisperse glycoconjugate presentations, as determined by surface plasmon resonance and dynamic light scattering.

Full text of DOI:10.1002/chem.201703432

A Journal of
Accepted Article
Title: Topological Defects in Hyperbranched Glycopolymers Enhance
Binding to Lectins
Authors: Míriam Salvadó, José Juan Reina, Javier Rojo, Sergio
Castillón, and Omar Boutureira
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201703432
Link to VoR: http://dx.doi.org/10.1002/chem.201703432
Supported by

Chemistry - A European Journal
10.1002/chem.201703432
FULL PAPER
Topological Defects in Hyperbranched Glycopolymers Enhance
Binding to Lectins
[a]
[b]
[b]
[a]
[a]
Míriam Salvadó, José J. Reina, Javier Rojo, Sergio Castillón, and Omar Boutureira*
Abstract: Central scaffold topology and carbohydrate density are
important features in determining both the binding mechanism and
a
b
potency of synthetic multivalent, polydisperse vs. monodisperse
carbohydrate systems to a model plant toxin (RCA120). We found
lower densities of protein receptors favour the use of heterogeneous,
polydisperse glycoconjugate presentations as determined by surface
plasmon resonance (SPR) and dynamic light scattering (DLS).
galectin
mucin
mucin
cross-linking
galectin
aberrant
mucin
STn
Tn
glycosylation
defects
normal cell
cancer cell
Introduction
design &
structure
The complexity of cellular glycocalyx has been shaped over
millions of year of evolution by the action of cell’s
evaluation of a
defect-containing
model
density
glycosylation site
(
glyco)machinery, an intricate pool of carbohydrate-processing
BoltornTM
enzymes that together with a different set of post-translational
modifications define the composition of the cellular membrane,
and its components. Biological systems, and particularly
glycoproteins, are heterogeneous in nature, yet the functions of
redundant glycoforms, which represent structurally similar
glycoproteins that differ in sugar structure, density, and/or
glycosylation site, and decorate the outer surface of cells
remains unknown. Heterogeneous glycoforms typically
participate in a variety of surface recognition events. In one
D
-Gal
c
heterogeneous
(hyperbranched
glycopolymer)
homogeneous
(glycodendrimer)
model lectin
RCA
120
in solution by DLS
on
surface by SPR
example, galectins are able to recruit and cross-link
a
Figure 1. Schematic representation of the binding modes of (a) normal and (b)
cancer cells to galectins due to the reduction/alteration of the surface glycan
shell. (c) Proposed simplified model to evaluate binding mechanism and
potency of both states.
heterogeneous collection of complex glycoconjugates that
[
1]
trigger important signalling events. In another example, the
presence of partially deglycosylated mucin proteins correlates
with disease states via reduction of the high-density shell
surface that reveals truncated, short epitopes (e.g. Tn and STn
antigens). The presence of such composition and spatial
organization defects enables preferential binding modes with
particular lectins and reveals higher-order/supramolecular
glycan level[5] and the central core architecture.[6] Examples by
Cloninger and co-workers highlight the presence of topological
defects that enhance binding towards their targeted receptors.[
In an attempt to mimic Nature and gain insight into the
7]
recognition patterns.[
2
]
Thus, glycoproteins resulting from
heterogeneous
essence
of
multivalent
sugar-protein
aberrant glycosylation patterns and/or with glycosylation defects
render otherwise hidden structures surface exposed and
sufficiently accessible to bind entities such as antibodies, lectins
or pathogens with particularly high avidities (Figure 1a,b).
Chemical glycobiologists have traditionally focused their efforts
on the development of homogeneous multivalent structures to
precisely determine carbohydrate-protein binding events.[ 3 ]
However, recent findings demonstrate that steric clash usually
diminishes potency of homogeneous, yet sterically congested
high-generation glycoconjugates[ 4 ] driving the attention to
interactions we sought to systematically evaluate the binding
mechanism between a series of multivalent homogeneous
(
monodisperse) vs. heterogeneously (polydisperse) presented
glycoligands and a model plant toxin (RCA120). We performed a
thorough analysis via surface plasmon resonance (SPR) and
dynamic light scattering (DLS) of the behaviour of ready
accessible–processable hyperbranched glycopolymers[ as a
suitable family of simplified defect-containing multivalent
glycomaterials with enhanced potency (Figure 1c). These two
techniques are representative of surface vs. solution evaluation
methods, two of the main important sugar-protein binding
events occurring in Nature.
8]
alternative, heterogeneous multivalent structures both at the
[a]
Dr. M. Salvadó, Prof. S. Castillón, Dr. O. Boutureira
Departament de Química Analítica i Química Orgànica
Universitat Rovira i Virgili
C/ Marcel·lí Domingo 1, 43007 Tarragona (Spain)
E-mail: omar.boutureira@urv.cat
Results and Discussion
[b]
Dr. J. J. Reina, Dr. J. Rojo
Glycosystems Laboratory, Instituto de Investigaciones Químicas
A series of multivalent systems were synthesized using two
different polyester polyol central cores (hyperbranched
Boltorn™ H30 and pentaerythritol dendrimer) that allowed for
the presentation of two different D-galactose loadings (16 and
(
4
IIQ), CSIC – Universidad de Sevilla, Av. Américo Vespucio 49,
1092 Sevilla (Spain)
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
32 D-Gal) in a monodisperse and a polydisperse manner
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201703432
FULL PAPER
a
(
Scheme 1). Thus, glycosyl azide
1
was reacted with
HO OH
monodisperse 2, 3 and polydisperse 4, 5 alkyne cores to obtain
M16, M32 and P16 (dispersity (Ð) = 2.19), P32 (Ð = 1.54),
respectively (supporting information, Figures S13 and S14). We
first investigated the aggregation properties and specificity of
RCA120 towards polydisperse and monodisperse ligands in
solution by DLS (Figure 2). Ricinus communis agglutinin
(i)
O
(
ii)
O
HO
N3
OH
1
M16
M32
P16
P32
∼
∼
32
16
32
16
2
3
4
5
(
RCA120) is a heterodimeric D-Gal-specific lectin with four
subunits and a BAAB composition. Whereas in A-chain resides
the cytotoxic activity, the B-chain contains two D-Gal-binding
sites separated by ca. 100 Å and composed by two globular
domains.[ Both lectin and ligands (P16, P32, M16, and M32)
were checked prior to analysis to determine their size and
confirm the absence of aggregates (supporting information,
M16
M
32
P
16
P
32
9]
M16
M32
P16
P32
(94%)
(88%)
(94%)
(89%)
OH
OH
O
OH
HO
O
b
OH OH
OH
OH OH
OH
HO
HO
HO O HO H
HO
O
O
HO
O
OH
OH
O
N
NN
Figure S4). A solution of RCA120 (166 nM, D
EP buffer was treated with increasing ligand concentrations
100–400 nM). At high concentrations, the only aggregates
formed were those with polydisperse ligands, P16 (D = 712 nm)
and P32 (D = 460 nm) while no aggregates were observed for
h
= 6.5 nm) in HBS-
O
HO
O
OH
OH
OH
O
OH
OH
O
O
HO
HO
HO
O
N
N
N
N
OH
O
NNN
O
HO
N
N
HO
O
O
N
N
O
HO
HO
OH
N
N
N
NN
OH
OH
O
N
O
O
OH
N
O
N
N
N
O
O
HO
O
OH O OH
OH
O
O
HO
O
OH
HO OH
O
O
O
OH
O
O
O
N
N
N
OH
NNN
O
O
O
O
O
O
HO
HO
O
O
O
O
O
O
O
N
NN
OH
O
N
N
O
O
O
O
O
(
HO OH
O
O
O
N
O
O
O
O
O
OH
OH
HO
O
O
O
N
N
N
O
O
O
HO
O
O
O
O
O
O
O
O
O
O
OH
O
O
O
OH
OH O OH
O
O
O
O
O
O
O
O
O
O
N
N
O
N
N
N
O
O
O
O
N
HO
O
O
O
O
O
O
O
O
N
HO
O
O
OH
O
N
N
O
O
OH
h
HO OH
N
N
N
O
HO
OH
O
O
O
O
O
O
OH
HO OH
O
O
O
O
O
O
HO
OH
O
O
O
HO
OH
HO
HO
O
O
O
O
O
O
O
O
O
O
O
OH
O
O
N
N
N
N
N
N
O
O
O
O
O
O
N
NN
O
HO
O
O
O
O
O
O
HO
h
O
O
HOHO
OH
O
OH
O
O
O
O
O
O
O
O
O
HO
O
O
O
OH
O
N
N
N
O
HO
O
HO OH
HO
O
N
N
O
O
OH
OH
OH
NN
N
O
N
O
O
HO
HO
O
O
O
O
OH
O
monodisperse ligands, M16 and M32 supporting the fact under
same concentration range and generation, polydisperse ligands
proved more effective to cross-link RCA120 (Figure 2a). For P32,
the largest aggregates were observed at 200 nM (1:1.2
lectin/ligand ratio), confirming that the ligand was acting as the
nucleating agent[ 10 ] while no aggregates were found for
monodisperse ligands presenting a similar carbohydrate loading.
A series of control experiments were also performed to confirm
the binding specificity of RCA120 towards D-Gal residues (Figure
N
N
O
OH
O
HO
OH
O
O
N
NN
N
N
N
HO
O
HO
N
N
HO
OH
N
HO
OH
N
N
N
N
N
N
O
HO
OH
N
N
N
O
O
OH
O
O
O
HO
HO
O
OH
O
OH
OH
O
N
O
OH
HO
HO
OH OH
HO
HO HO
O
OH
HO
HO
OH
O
O
HO
HOHO HO
O
OHHO OH
HO HO
M16
P16
= 7302 Da, M
Ð = 2.19
M
n
= 15963 Da
w
OH OH OH OH
OHOH
OH
HO OH
OH HO
OOHHO OOH OH OH
HO
O
OH
OH
OH
OH OH
OH
O
OH
OH HO
O
O
O
HO
O
OH
OH
HO
HO
OH
HO
HO
O
O
O
N
N
O
O
OH
O
HO
N
HO
O
O
HO
HO OH OH
O
NN
O
NN
O
OH
OH
OH
O
NN
N
N
N
N
N
OH
O
N
N
OH
OH
HO
O
N
N
HO
O
HO
N
N
N
HO
N
N
N
N
O
OH
O
N
N
N
N
N
O
HO OH OH
O
N
OH
HO
O
O
O
HO
NN N
O
O
O
O
O
N
HO
O
O
O
O
O
N
HO
O
OH
OH
O
N
N
O
O
O
N
N
OH
O
N
N
N
O
O
O
O
OH
O
O
O
N
N
O
HO
HO
N
O
O
O
O
O
O
O
HO
OH
OH
HO OH
HO
O
O
O
O
O
O
O
OH
OH
OH
O
O
O
N
O
O
O
N
N
N
O
O
N
NN
HO
HO
O
O
O
N
N
HO
OH
O
O
O
O
O
O
O
O
HO
OH
O
NN
O
O
O
O
O
HO
OH
HO OH
HO OH
O
HO
O
O
O
O
N
O
O
O
O
O
OH
N
O
O
N
N
N
O
O
O
O
N
N
N
O
O
O
O
N
N
O
O
O
N
HO
O
O
O
O
OH
O
O
O
O
O
OOO
OH
OH
N
N
N
N
O
O
O
O
O
HO OHO
HO
N
N
N
O
O
O
O
O
O
O
N
N
O
O
HO OH
O
O
O
O
O
O
O
N
N
N
2b). A solution of P16 with RCA120 in HBS-EP buffer (1:2.4
O
OH
O
O
O
O
O
O
O
O
O
O
O
O
HO
HO
OH
O
O
O
O
OH
O
O
O
HO
O
HO OH
OH
OH
HO OH
N
N
N
O
O
O
O
O
O
O
O
O
N
N
OHO
O
HO
OH
OH
N
N
O
O
O
O
O
O
O
O
O
O
O
O
O
O
OH
O
O
O
O
O
O
HO
OH
N
N
N
O
N
O
O
O
O
NN
HO
HO
OH
N
O
O
O
O
OH
O
O
N
N
O
HO
O
lectin/ligand ratio) was chosen for the study. The solution was
titrated with D-galactose (from 250 μM to 1 mM) and the particle
size was determined. For concentrations of D-galactose up to
O
O
O
O
O
O
O
O
N
O
HO
N
O
O
O
N
HO OH
O
O
O
O
HO O
OH
OH
N
O
O
O
OH
OH
N
O
O
O
O O
O
O
O
O
O
O
N
N
N
O
HO
O
HO
HO
OH
N
N
O
O
O
O
O
O
N
N
N
HO OH
O
O
N
N
N
O
O
HO
O
O
O
O
O
OH
OH
O
O
NN
N
O
O
O
O
O
O
O
OH
HO
OH
O
O
O
N
N
N
N
HO OH
HO
HO
O
O
OO
O
O
N
N
O
HO
O
O
O
O
OH
O
N
OH
O
O
O
OH
O
N
N
O
N
N
O
HO
OHO
O
O
O
N
O
OH
OHO
N
N
OH
HO
OH
N
HO
HO
HO
HO
OH
N
N
NN
N
N
N
N
N
O
N
HO OH
O
N
N
N
N
N
O
HO
NN
N
OHO
N
N
N
N
O
OH
HO
OH
HO
N
N
N
N
N
O
OH
O
OH
O
N
N
O
O
O
N
N
N
N
N
N
OH
HO
O
OH
O
OHHO
HO
OH
O
O
HO
OH
OH
O
HO
HO
O
OH
O
O
OH
O
HO HO
HO
O
OH
O
O
O
OH
OHHO
HO
OH
O
OH
O
O
OH
500 μM, reversion of aggregate formation was observed. While
O
OH
HO
O
HO OH
O
O
OH
HO
HO
HO
HO OH
HO HO
HO
HO OHHO
HO OH
HO
HO
O
OH
HO
OH OH
HO
titration of the same solution with the same amount of D-glucose
as well as the equivalent volume of buffer (HBS-EP) did not
influence aggregate formation, indicating that our ligands were
able to bind selectively to RCA120 due to the presence of non-
reducing D-Gal residues (supporting information, Figure S6).
With these preliminary results in hand, we sought to study how
the monodisperse and polydisperse D-Gal presentation
significantly influences the binding kinetics and mechanism. To
obtain kinetic data of multivalent interactions, SPR direct binding
experiments were performed (Table 1). RCA120 was covalently
attached to a polycarboxyl CM5 sensor chip to generate three
different lectin surfaces at various levels of functionalization
M32
P32
= 4913 Da, M
Ð = 1.54
M
n
= 7561 Da
w
Scheme 1. (a) Preparation of multivalent (glyco)polymers/dendrimers; (i)
CuSO ·5H O, NaAsc, TBTA, 1:1 THF/H O, microwave (µwave) irradiated in a
sealed tube at 60 ºC for 2 h using a CEM-Discover™ single-mode synthesizer
temperature control, fixed hold time off, normal absorption mode, 300 W); (ii)
CuSO ·5H O, NaAsc, 1:1 tBuOH/H O, rt, 72 h for P16 and CuSO ·5H O,
NaAsc, TBTA, 1:1 THF/H O, µwave, 60 ºC, 2 h for P32. (b) Structures of
monodisperse dendrimers (M16, M32) and idealized polydisperse
hyperbranched polymers (P16, P32). NaAsc = sodium ascorbate. TBTA =
tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]amine.
4
2
2
(
4
2
2
4
2
2
(
RCA120-HD, -MD, and -LD where HD, MD, and LD = high,
medium, and low densities, respectively). Binding data were
collected at μL/min flow rate using different analyte
5
concentrations. Prior to analysis, binding tests using RCA120-HD
at different flow rates were performed to discard mass transport
effects that could influence the shape of sensograms.[ 11 ] As
expected, complex binding profiles were reflected in
sensograms with RCA120-MD (chosen for further studies due to
easier surface regeneration) according to the multivalent nature
of the analytes tested. Attempts to fit the results obtained to
conventional binding models failed, yet following the procedure
described by Fernandez-Megia et al.,[ 12 ] a separate kinetic
analysis of the sensograms at early association and late
Figure 2. Particle size measurements by DLS using (a) RCA120/ligands (1:2.4
ratio) in HBS-EP buffer and (b) titrations of RCA120/P16 with D-Gal and D-Glc.
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201703432
FULL PAPER
n
Table 1. Kinetic data of the binding of multivalent M/P-Gal ligands to RCA120
The early association phase (t ≤ 50 s) was fitted to a 1:1
Langmuir binding model and kobs vs. concentration was plotted.
Deviations from the linear behaviour were found for all analytes
4
–1 –1
surfaces with low density (LD) and medium density (MD). kon (x10 M
s ), koff
4
–1
(
x10 s ) and K
D
(nM).
RCA120-LD (1300 RU)[a]
RCA120-MD (6400 RU)[b]
Ligand
k
on
k
off
K
D
k
on1
k
on2
k
off
K
D
P16
M16
P32
M32
1.44
ND
17.9
ND
45
125
ND
10.6
217
1.93
24.9
8.0
6.35
56.2
8.73
18.6
11.7
3.23
2.71
5.93
13.8
ND
32.6
ND
57.5
66.7
ND
9.21
[
a]
Fitted to a 1:1 Langmuir binding model. [b] Separate kinetic analysis of early
association and late dissociation phases. ND=not detected (the analyte does
not remain attached to the chip surface after injection and therefore the
increase in RUs during association is due to a change in the refractive index
caused by the change in solution concentration).
dissociation phases was performed (Figure 3 and supporting
information, Figures S1 and S2). During the dissociation, a two-
phase process was observed. Initially a fast dissociation was
shown followed by an extremely slow dissociation phase
suggesting the presence of both weak and strong binding events.
This fast dissociation is more pronounced in monodisperse than
in polydisperse analytes. More analyte reminds attached to the
surface of the lectin when polydisperse structures are tested. In
both cases, attempts to fit the dissociation phase to conventional
binding models failed. While the mayor degree of heterogeneity
was found at early dissociation times (t = 180–250 s), late
dissociation times (t ≥ 250 s) showed good fitting to a 1:1
Langmuir binding model. For polydisperse analytes, koff values
were found to be lower than for monodisperse. Significantly
different values in koff resulted when comparing multivalent
structures presenting the same number of D-Gal units. M16 and
M32 showed 8.9- and 2.1-fold higher dissociation rates than P16
and P32, respectively. When comparing dispersity, in the case
of polydisperse analytes (P16 and P32) a 1.4-fold slower
dissociation was observed when the number of D-Gal was
increased (Table 1). In contrast, M16 showed a 3.0-fold slower
Figure 3. Representative SPR separate kinetic analysis of the interaction of
P16 and M16 with RCA120-MD. (a, b) Late dissociation phase kinetic analysis:
sensograms and global fitting (red) to pseudo-first-order kinetics. (c, d) Early
association phase kinetic analysis: sensograms and global fitting of the early
association phase (magenta and green) to the integrated rate equation for
pseudo-first-order kinetics. (e, f) Plots of kobs vs. [ligand].
dissociation
than
the
corresponding
3rd
generation
evaluated and two different slopes were observed (Figure 3 and
supporting information, Figures S1 and S2). Low concentrations
present a fast association (largest slope, high order complexes)
while high concentrations present a slow association resulting in
complex kinetic interactions due to the multivalent nature of the
analytes (Table 1). The 2-fold increase of D-Gal residues in
polydisperse analytes (from P16 to P32) resulted in a 5-fold
increase in kon1. This trend was not observed in monodisperse
analytes, when the glycodendrimer surface is coated with 32 D-
monodisperse M32. The same trend was observed for
polydisperse P16 when the early dissociation phase was
evaluated as the percentage of SPR signal decay (t = 180–250
s). For P16 the percentage observed was 1.9- and 1.1-fold lower
than for P32 and M16 (supporting information, Table S1). The
slow dissociation of polydisperse Boltorn™ structures (P16 and
P32) and their low percentage of signal decay from RCA120-MD
surface represented an enhanced binding activity compared to
monodisperse structures (M16 and M32). The significant low
value of off rates could be explained because a nearby analyte
could quickly replace a bound analyte in close proximity
4
–1 –1
Gal units kon1 value for M32 (66.7x10 M s ) decreased in 3.2-
fold compared to M16 (217x104 M–1 s ), which at low
concentrations yielded a kon1 value 20.4-fold higher than P16
–1
(
statistical rebinding), which is favoured when D-Gal is presented
4
–1
4
–1
(
10.6x10 M s–1) and a kon2 (24.9x10 M s–1) 12.9-fold higher
to the receptor in a polydisperse manner due to the higher
accessibility of the sugar epitopes compared to those found in
more crowded, sterically congested monodisperse systems.
Additional stabilization due to the presence of secondary
interactions with unmodified surface groups and/or inner core
may also affect off rates. After an exhaustive evaluation of the
dissociation phase of the sensogram, a kinetic evaluation of the
association was also performed. The linear increase observed
during the association phase, also indicates a strong binding.
at high concentrations. For M32, kon1 and kon2 were only 1.2-fold
higher compared to P32. The shape of sensograms was
indicative of the high increase of binding for monodisperse
compared to polydisperse analytes (at early association a more
pronounced increase is observed in M16 and M32 sensograms
compared to P16 and P32). In hyperbranched polymers (P),
when the number of carbohydrates presented in the surface
increases their effective multivalency is also amplified. This
effect could favor rebinding processes and the fact that one
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201703432
FULL PAPER
analyte could simultaneously bind to two proteins might explain
why the affinity constant increase when the number of
carbohydrates presented is higher. However in the case of
homogeneous glycodendrimers it is known that structures with
higher generations (high valency) not always give better
affinities.[ The drop in affinity has been usually attributed to
steric hindrance between carbohydrate units. In previous studies
the binding constant of D-galactose towards RCA120 has been
might be one of the pivotal reasons for the co-evolution of
glycans to such complex, heterogeneous structures. We
anticipate the data presented herein will be relevant to develop
cheap, ready available multivalent carbohydrate therapeutics
designed and utilized as anti-infective agents against common
human diseases.
13]
determined by isothermal titration calorimetry (ITC) with a K
D
of
Acknowledgements
4
.5x10–4 M.[14] P16, M16, P32, and M32 yielded a K
D
in the nM
range, four orders of magnitude lower than its monomeric
counterpart. These results suggest that our multivalent ligands
are able to simultaneously bind two RCA120 lectins in the MD-
surface.[15] The detailed kinetic evaluation of RCA120-MD gave
important information about the real-time binding of this type of
analytes, showing a heterogeneous binding with distinguishable
weak and strong affinities caused by clustering and rebinding
phenomena. SPR direct binding analysis was also performed in
RCA120-LD for our four analytes, yet conversely to MD-surface
no binding was detected for monodisperse ligands (Table 1).
Exhaustive analysis of the polydisperse analytes was also
performed at different concentrations. The same sensogram
profile was observed as that for RCA120-MD. According to the
dissociation and association phases, the same trend is observed
as in RCA120-MD, a fast dissociation followed by a slow
dissociation and a linear increase in association. However, the
decrease in lectin density showed a less complex binding profile
and the sensograms could be fitted well to a 1:1 Langmuir
binding model. Thus, the RCA120-LD surface showed a single
slope that equals to kon when plotting kobs vs. concentration for
P16 and P32 indicating no binding heterogeneity (supporting
information, Figure S3). The binding efficiency between
generations is maintained but the decrease in lectin density
impedes the analyte to simultaneously bind two RCA120 lectins.
The high affinity binding modes could only be explained by
rebinding and clustering with the lectin secondary binding sites.
Thus, the study of the binding affinities with different lectin
density gave an idea of the importance of sugar presentation vs.
mode of action. Accordingly, this real-time-dependent analysis
suggests that the different binding mechanisms described
depend on the local concentration of analyte and its proximity to
the immobilized lectin and not only to the analyte multivalency
and the lectin surface density.
We thank the European Commission (Marie Curie CIG, O.B.),
MINECO (CTQ2014-58664-R, CTQ2014-52328-P, and RYC-
2015-17705), the European Regional Development Fund,
Generalitat de Catalunya (M.S.), and CSIC (JAEdoc contract,
J.J.R.) for financial support. O.B. is a Ramón y Cajal Fellow.
Keywords: carbohydrates • glycodendrimers • hyperbranched
glycopolymers • molecular recognition • multivalency
[1]
B. Belardi, G. P. O’Donoghue, A. W. Smith, J. T. Groves, C. R. Bertozzi,
J. Am. Chem. Soc. 2012, 134, 9549–9552.
[
2]
3]
K. Godula, C. R. Bertozzi, J. Am. Chem. Soc. 2012, 134, 15732–15742.
a) S. Cecioni, A. Imberty, S. Vidal, Chem. Rev. 2015, 115, 525–561; b)
Y. M. Chabre, R. Roy, Chem. Soc. Rev. 2013, 42, 4657–4708.
a) S. Bhatia, L. Cuellar Camacho, R. Haag, J. Am. Chem. Soc. 2016,
[
[
4]
5]
138, 8654–8666; b) M. Kanai, K. H. Mortell, L. L. Kiessling, J. Am.
Chem. Soc. 1997, 119, 9931–9932; c) P. R. Ashton, E. F. Hounsell, N.
Jayaraman, T. M. Nilsen, N. Spencer, J. F. Stoddart, M. Young, J. Org.
Chem. 1998, 63, 3429–3437; d) D. Pagé, R. Roy, Bioconjugate Chem.
1997, 8, 714–723.
[
a) J. L. Jiménez Blanco, C. Ortiz Mellet, J. M. García Fernández, Chem.
Soc. Rev. 2013, 42, 4518–4531; b) D. Ponader, P. Maffre, J. Aretz, D.
Pussak, N. M. Ninnemann, S. Schmidt, P. H. Seeberger, C.
Rademacher, G. U. Nienhaus, L. Hartmann, J. Am. Chem. Soc. 2014,
136, 2008–2016; c) L. Xue, X. Xiong, K. Chen, Y. Luan, G. Chen, H.
Chen, Polym. Chem. 2016, 7, 4263–271.
[6]
a) Y. Miura, Y. Hoshino, H. Seto, Chem. Rev. 2016, 116, 1673–1692; b)
E. Schaeffer, L. Dehuyser, D. Sigwalt, V. Flacher, S. Bernacchi, O.
Chaloin, J.-S. Remy, C. G. Mueller, R. Baati, A. Wagner, Bioconjugate
Chem. 2013, 24, 1813–1823.
[7]
[8]
[9]
E. K. Woller, E. D. Walter, J. R. Morgan, D. J. Singel, M. J. Cloninger, J.
Am. Chem. Soc. 2003, 125, 8820–8826.
C. C. Lee, J. A. MacKay, J. M. J. Fréchet, F. C. Szoka, Nat. Biotechnol.
2005, 23, 1517–1526.
a) C. Scheibe, S. Wedepohl, S. B. Riese, J. Dernedde, O. Seitz,
ChemBioChem 2013, 14, 236–250; b) V. Wittmann, R. J. Pieters,
Chem. Soc. Rev. 2013, 42, 4492–4503.
Conclusions
[
10] C. K. Goodman, M. L. Wolfenden, P. Nangia-Makker, A. K. Michel, A.
Raz, M. J. Cloninger, Beilstein J. Org. Chem. 2014, 10, 1570–1577.
11] D. G Myszka, Curr. Opin. Biotech. 1997, 8, 50–57.
12] E. M. Munoz, J. Correa, R. Riguera, E. Fernandez-Megia, J. Am. Chem.
Soc. 2013, 16, 5966–5969.
In summary, we found the nature of the core and the display of
ligands enormously affects the binding mechanism and
potency.[ Modifications of the central core of multivalent
systems give relevant binding differences in both DLS and SPR
experiments. Thus, lower densities of receptors (RCA120) as
those found in physiological environments favour the use of
heterogeneous (polydisperse) glycoconjugate presentations –
mimicking inherent defects found in glycoproteins and/or
different glycoforms – that enhance their potency in a match
scenario of interactions suggesting this mutual reinforcement
[
[
4]
[13] F. Lasala, E. Arce, J. R. Otero, J. Rojo, R. Delgado, Antimicrob. Agents
Chemother. 2013, 47, 3970–3972.
[14] T. K. Dam, C. F. Brewer, Chem. Rev. 2002, 102, 387–430.
[15] a) A. Mulder, T. Auletta, A. Sartori, S. Del Ciotto, A. Casnati, R. Ungaro,
J. Huskens, D. N. Reinhoudt, J. Am. Chem. Soc. 2004, 126, 6627–
6636; b) M. Mammen, S.-K. Choi, G. M. Whitesides, Angew. Chem. Int.
Ed. 1998, 37, 2754–2794; Angew. Chem. 1998, 110, 2908–2953.
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201703432
FULL PAPER
FULL PAPER
Core topology and sugar density in
synthetic multivalent, polydisperse vs.
monodisperse carbohydrate structures
affect binding mechanism and potency
to a model plant toxin
M. Salvadó, J. J. Reina, J. Rojo, S.
Castillón, O. Boutureira*
Page No. – Page No.
Topological Defects in
Hyperbranched Glycopolymers
Enhance Binding to Lectins
This article is protected by copyright. All rights reserved.