Multivalent scaffolds induce galectin-3 aggregation into nanoparticles

Article abstract of DOI:10.3762/bjoc.10.162

Galectin-3 meditates cell surface glycoprotein clustering, cross linking, and lattice formation. In cancer biology, galectin-3 has been reported to play a role in aggregation processes that lead to tumor embolization and survival. Here, we show that lactose-functional-ized dendrimers interact with galectin-3 in a multivalent fashion to form aggregates. The glycodendrimer-galectin aggregates were characterized by dynamic light scattering and fluorescence microscopy methodologies and were found to be discrete particles that increased in size as the dendrimer generation was increased. These results show that nucleated aggregation of galectin-3 can be regulated by the nucleating polymer and provide insights that improve the general understanding of the binding and function of sugar-binding proteins.

Full text of DOI:10.3762/bjoc.10.162

Multivalent scaffolds induce galectin-3 aggregation
into nanoparticles
Candace K. Goodman1, Mark L. Wolfenden1, Pratima Nangia-Makker1,2,
Anna K. Michel1, Avraham Raz1,2 and Mary J. Cloninger*1
Full Research Paper
Open Access
Address:
Department of Chemistry and Biochemistry, Montana State
Beilstein J. Org. Chem. 2014, 10, 1570–1577.
1
doi:10.3762/bjoc.10.162
University, Bozeman, Montana 59717, USA and 2The Departments of
Oncology and Pathology, School of Medicine, Wayne State
Received: 03 March 2014
Accepted: 18 June 2014
Published: 10 July 2014
University, 110 East Warren Avenue, Detroit, Michigan 48201, USA
Email:
Mary J. Cloninger* - mcloninger@chemistry.montana.edu
This article is part of the Thematic Series "Multivalent glycosystems for
nanoscience".
*
Corresponding author
Guest Editor: J.-L. Reymond
Keywords:
dendrimers; galectin-3; glycodendrimers; multivalency; multivalent
glycosylation; protein aggregation
© 2014 Goodman et al; licensee Beilstein-Institut.
License and terms: see end of document.
Abstract
Galectin-3 meditates cell surface glycoprotein clustering, cross linking, and lattice formation. In cancer biology, galectin-3 has been
reported to play a role in aggregation processes that lead to tumor embolization and survival. Here, we show that lactose-functional-
ized dendrimers interact with galectin-3 in a multivalent fashion to form aggregates. The glycodendrimer–galectin aggregates were
characterized by dynamic light scattering and fluorescence microscopy methodologies and were found to be discrete particles that
increased in size as the dendrimer generation was increased. These results show that nucleated aggregation of galectin-3 can be
regulated by the nucleating polymer and provide insights that improve the general understanding of the binding and function of
sugar-binding proteins.
Introduction
The role of multivalency in biology is well established, and made of ~130 amino acids that are arranged in a folded beta-
examples of this phenomenon abound [1]. The ability of multi- sheet structure and have an affinity for β-galactosides [4-7].
valency to enhance weak interactions has been shown in a Galectin-3, one of the most studied members, is commonly up
variety of protein:carbohydrate systems using a wide assort- or down regulated in different cancers and is implicated in
ment of scaffolds and carbohydrates [2]. As research with tumor formation and proliferation, apoptosis, angiogenesis, and
multivalent glycosystems advances, one important target for B cell activation [8-10]. Galectin-3 has been reported to be
potential therapy and understanding is the galectin family of involved in mechanisms that cluster cell surface glycoproteins
proteins [3]. Members of the galectin family share a common [10,11], cross-link receptors [12], and form lattices and larger
conserved sequence carbohydrate recognition domain (CRD) aggregates [13]. Structurally, galectin-3 is composed of one
1570

Beilstein J. Org. Chem. 2014, 10, 1570–1577.
carbohydrate recognition domain and a collagen-like functionalized with a variety of molecules, making these scaf-
N-terminus tail [14].
folds an excellent choice for systematic studies of chemical and
biological phenomena [27,28]. In this investigation, PAMAM
The native quaternary structure of this unique galectin is a dendrimers were functionalized using a methodology similar to
current topic of debate. Brewer et al. found that galectin-3 previous literature [29]. Synthesis of β-lactoside derivative 1
pentamers can be formed at high concentrations of protein [15], was performed as shown in Scheme 1. Lewis acid facilitated
a noncovalent dimer and a monomer form of galectin-3 have glycosylation, which was directed by neighboring group partici-
also been reported [16,17]. Recent anisotropy binding measure- pation of the 2-O-acetyl protecting group, afforded the desired
ments support two types of galectin-3 oligomerization, domin- anomers in good yields. The trichloroacetimidate intermediate
ated by either N- or C-terminal interactions [18], and both N- was formed to enhance coupling.
and C-terminal domains are reported to be required for binding
to targets such as lipopolysaccharide [19,20]. Galectin-3 can
serve as a cellular docking site or a crosslinker for microorgan-
isms binding to pathogens directly [20,21], and galectin-3 can
act as a scaffold for the presentation of ligands such as
lipopolysaccharides into an aggregate that stimulates cellular
responses [19].
Binding affinities have been reported for a series of carbohy-
drate-based ligands to galectin-3, which binds to lactose signifi-
cantly better than to galactose or to N-acetylgalactosamine but
does not bind to mannose [22-24]. Both the glycan ligand and
Scheme 1: Synthesis of isothiocyanato-functionalized lactoside 1.
the topological display on the cell surface are required for high
affinity, selective binding of galectins, as demonstrated in
galectin binding studies with neuroblastoma cells [25].
Syntheses of the carbohydrate-functionalized dendrimers were
performed by addition of compound 1 as shown in Scheme 2.
Here, we demonstrate that glycodendrimers bearing lactose can The functionalized dendrimers were characterized by
be used to form large, discrete aggregates of galectin-3. Since, MALDI–TOF–MS (matrix-assisted laser desorption time of
as noted above, glycan clustering and galectin-3-mediated flight mass spectrometry). The average numbers of sugars that
aggregations have been demonstrated to be important for bio- were incorporated are shown in Scheme 2. The loadings were
logical interactions ranging from bacterial invasion to cancer determined by both the changes in weight average molecular
cellular responses, the development of systems such as glyco- weight (Mw) upon addition of 1 and the changes in Mw upon
dendrimers that can aggregate galectin-3 into nanoparticles in a deacetylation, enabling characterization of the average number
highly controlled fashion is an important area of research. The of sugars per dendrimer [30]. Additional characterization details
study of galectin-3 binding and cluster formation by a series of (including 1H NMR spectra) are provided in Supporting Infor-
glycodendrimers is a central step in the development of a syn- mation File 1.
thetic multivalent antagonist that can intercept and influence
Characterization of dendrimer/galectin-3
galectin-3-mediated cellular processes and may be of clinical
value as a non-cytotoxic drug and/or be developed for cancer aggregates
imaging.
Dynamic light scattering (DLS) was used to characterize the
size and polydispersity of aggregate formation between lactose-
functionalized dendrimers 2, 3, 4, and 5, with galectin-3. Three
concentrations of glycodendrimers (11.5, 3.3 and 0.14 μM)
were added to a constant concentration of galectin-3 (31 μM in
Results and Discussion
Synthesis and characterization of glycoden-
drimers
Poly(amidoamine) (PAMAM) dendrimers are well-defined, PBS) to obtain a ratio of galectin-3 to glycodendrimer of 220:1,
water-soluble, symmetric scaffolds that contain a controlled and 9:1, or 3:1. These ratios were chosen so that results obtained
tuneable number of end groups. The number of end groups is from experiments using a large excess, a significant excess, and
specified by the dendrimer generation and approximately a slight excess of galectin-3 could be compared.
doubles for each subsequent generation [26]. These dendrimers
are commercially available for generations zero, denoted G(0), Regardless of the amount of excess galectin-3 that was used, the
to generation 10, denoted G(10). The amine termini can be size and polydispersity of the aggregates was shown to increase
1571

Beilstein J. Org. Chem. 2014, 10, 1570–1577.
as the nucleating agent for galectin-3 aggregation. For example,
when the concentration of galectin-3 is comparable to the
concentration of glycodendrimer, then galectin-3 is presented
with many different nucleating scaffolds, and smaller particles
are formed. On the other hand, when galectin-3 is present in
large excess, not as many nucleating sites can be incorporated
into each aggregate, which causes the aggregates to be smaller.
A schematic representation of glycodendrimer-mediated
galectin-3 aggregation is shown in Figure 2. This trend was also
observed in other systems. Ottaviani, et al. reported enhanced
aggregation of amyloid peptides at low concentrations of
maltose and maltotriose-functionalized poly(propylene imine)
dendrimers and inhibition at high glycodendrimer concentra-
tions [31]. Previous studies of asialofetuin/galectin-3 aggrega-
tion indicated that the glycoprotein ligand could serve to initiate
aggregation, but carbohydrate binding was not required for all
Scheme 2: Synthesis of carbohydrate-functionalized PAMAM
dendrimers. (Values for m and for x equivalents added are given in
Supporting Information File 1.)
with increasing dendrimer generation (Figure 1 and Table 1). of the galectin-3 lectins that were involved in the interaction.
The largest aggregates were observed for the 9:1 ratio of Some galectin-3/galectin-3 interactions, in addition to carbohy-
galectin-3 to glycodendrimer, and smaller aggregates were drate/galectin-3 interactions, were proposed [18].
formed when a large excess or a very small excess of galectin-3
was used. This trend is logical if the glycodendrimer is serving
Figure 2: Schematic representation of galectin-3/glycodendrimer
aggregates at varying stoichiometries.
A series of control experiments indicate that aggregation is
Figure 1: Effective diameter of galectin-3/glycodendrimer aggregates
induced by the binding of galectin-3 to lactose on the
(
DLS). Final concentration of galectin-3 31 μM; final concentration of
glycodendrimers 0.14, 3.3, 11.5 μM for 220:1, 9:1, 3:1, respectively. 2
blue), 3 (black), 4 (red) and 5 (green). Aggregate size was below the
dendrimers. No observable particles were detected upon addi-
tion of a mannose-functionalized G(4) dendrimer (Table S3 in
Supporting Information File 1). Pre-incubation of galectin-3
(
detection limit for 2 at 0.14 μM.
Table 1: Summary of aggregate characterization.
Compound
No. of particles
Mean diameter (FM, nm)
Avg. effective diameter (DLS, nm)
2
3
4
5
59
240 ± 50
not detectable
560 ± 40
221
137
146
700 ± 290
1070 ± 350
1790 ± 650
1180 ± 80
1620 ± 110
1572

Beilstein J. Org. Chem. 2014, 10, 1570–1577.
with 1 mM lactose solution completely inhibited aggregate for-
mation in the presence of glycodendrimers 3, 4, and 5 (Table S3
in Supporting Information File 1). Titration of a lactose solu-
tion into the solution of preformed aggregates did result in
disassembly, but titration of an equivalent volume of PBS also
resulted in disassembly (Table S3, Supporting Information
File 1). Analysis of concentration effects and kinetics of aggre-
gation and disaggregation are beyond the scope of this report
but are under investigation. Experiments using 5 and truncated
galectin-3, which has only the carbohydrate recognition domain
without the N-terminal domain, did not result in aggregate for-
mation (see Table S3, Supporting Information File 1). No
aggregates were observed for dendrimers 2–5 in solution
without addition of galectin-3, and no aggregates were observed
for galectin-3 when glycodendrimers were absent from the
solution. Taken together, these data support aggregate initiation
as a response to specific carbohydrate binding interactions
between lectin and glycodendrimer. They also reveal the signif-
icance of the N-terminal domain in formation of higher order
aggregates.
The presence of these aggregates was confirmed by epifluores-
cence microscopy using galectin-3 fluorescently labelled with
AlexaFluor 488 (A488gal-3, Figure 3). Following conjugation,
the labelled galectin-3 was dialyzed against PBS to maintain
identical conditions to DLS experiments. Size quantification
using image analysis software (Pixcavator 6.0) and fluorescent
microsphere standards (Dragon Green, Bangs Laboratories,
Inc.) provided similar diameters as those obtained in DLS
experiments (Figure 3, Table 1). The polydispersity calculated
from the micrographs (Figure 4) was higher than that calcu-
lated by DLS, but this is likely due to sampling bias of DLS
measurements as a result of attenuating the incident light
Figure 3: Fluorescence microscopy images of labelled particles.
Microbead standards at similar exposure times are shown in (a)–(d);
(a) 190 nm, inset is a 4× enlargement of selected area, (b) 520 nm,
inset is a 4× enlargement of selected area, (c) 1020 nm and (d) 1900
nm. Aggregates formed after ca. 60 min incubation of Alexa 488
labelled galectin-3 (A488gal-3) with lactose-functionalized dendrimers
are shown in (e)–(h); (e) A488gal-3 and 2; (f) A488gal-3 and 3;
(g) A488gal-3 and 4; (h) A488gal-3 and 5. Exposure times of
lectin–glycodendrimer aggregates are provided in Table S6,
Supporting Information File 1.
(
smaller particles that remain undetected in DLS were observed
in microscopy images). For both methods, the trend of
increasing size and polydispersity with increasing dendrimer
generation was observed.
and protein that are required to form nanoparticle aggregates of
the observed sizes, the aggregates are highly monodisperse.
The aggregate size is remarkably large compared to galectin-3 Although it has been previously determined using turbidity and
and the dendrimers. The largest dendrimer used (generation 6) precipitation assays that carbohydrate-functionalized
has a reported unfunctionalized diameter of 6.8 nm [32,33], and dendrimers induce lectin aggregation, the consistent formation
addition of the sugar adds about 4 nm to the overall diameter of large nanoparticles has to our knowledge not been previ-
according to our DLS results with 5 (Table S3, Supporting ously identified and characterized.
Information File 1). The measured diameter of the CRD domain
from the crystal structure of galectin-3 is roughly 3 nm [34]. The most likely explanation for the formation of large,
The N-terminal domain consists of slightly fewer amino acids monodisperse nanoparticle aggregates from galectin-3/glyco-
but is unstructured. Assuming the unstructured portion dendrimer solutions is as follows. The glycodendrimer serves to
contributes about the same size or slightly more to the diameter nucleate the aggregation process through the specific binding of
of the protein as the CRD, the glycodendrimer complex would lactose into the carbohydrate binding site on galectin-3. Binding
be expected to have a much smaller diameter than the observed of the carbohydrate into the galectin-3 binding site must then be
values. Considering the large number of copies of dendrimer enabling protein–protein interactions. Some of these
1573

Beilstein J. Org. Chem. 2014, 10, 1570–1577.
plex multivalent interactions between galectin-3 and lactose-
functionalized dendrimers is indicated by the observation of
large aggregates in DLS and epifluorescence experiments.
The large and relatively monodisperse nature of the glycoden-
drimer/galectin-3 aggregates that were formed (as determined
by DLS and fluorescence microscopy) was dependent on both
the dendrimer concentration and the generation. Third and
fourth generation glycodendrimers formed smaller, more
monodisperse aggregates, than sixth generation glycoden-
drimers. Aggregates formed at molar ratios of 9:1
galectin:glycodendrimer were largest while 220:1 and 3:1 ratios
produced smaller complexes. The difference in aggregate sizes
may relate to the size and shape complementarity between
dendrimer and lectin or to the interplay of enthalpic and
entropic contributions to aggregate formation as previously
postulated [18,31,36]. Ongoing studies on the aggregate stoi-
chiometry should provide valuable insight on this matter.
Overall, the results presented here indicate that clustering and
aggregation events should be considered in addition to carbohy-
drate binding affinity for galectin-3, and also for other bio-
logical processes that are mediated by multivalent carbohy-
drate–protein interactions.
Experimental
General experimental methods
General reagents were purchased from Acros and Aldrich
Chemical Companies. PAMAM dendrimers were purchased
from Dendritech. Fluorescent microbead standards were
purchased from Bangs Laboratories, Inc.. Dichloromethane was
purified on basic alumina; other solvents were used as received.
Silica gel (32–63 μm “40 micron flash”) for flash column chro-
matography purification was purchased from Scientific Adsor-
bants Incorporated.
Figure 4: Aggregate diameter distribution of fluorescence microscopy
images; 31 μM galectin-3 and 0.14 μM (a) 2, (b) 3, (c) 4 and (d) 5.
5
-Isothiocyanato-3-oxapentyl 2,3,4,6-tetra-O-acetyl-
β-D-galactopyranosyl-[1→4]-2,3,6-tri-O-acetyl-β-D-
glucopyranoside (1)
protein–protein interactions may occur because of intertwining 2,3,4,6-Tetra-O-acetyl-β-D-galactopyranose-[1→4]-1,2,3,6-
of the N-terminal domains that are now in close proximity. tetra-O-acetyl-β-D-glucopyranose (4.4 g, 6.4 mmol) was
However, protein–protein interactions using the carbohydrate dissolved in dry DMF (20 mL). Hydrazine acetate (0.77 g,
recognition domains of galectin-3 after an initial carbohydrate 8.4 mmol) was added and the reaction mixture was heated to
binding event is entirely consistent with a recently proposed 55 °C for 1 h. The mixture was diluted into CH2Cl2 (20 mL)
binding mechanism [18], and is also consistent with proposed and washed with brine (2 × 10 mL) and water (2 × 10 mL),
models for scaffold-mediated nucleation of protein signalling dried over MgSO4, filtered and the solvent was removed in
complexes [35].
vacuo. The residual product was added to a solution of
trichloroacetonitrile (3.34 g, 23.1 mmol) in CH2Cl2 (20 mL).
After cooling the mixture in an ice bath, DBU (60 mg,
Conclusion
In summary, β-lactoside functionalized PAMAM dendrimers 0.32 mmol) was added drop-wise and the mixture was stirred
2
–5 were synthesized and characterized. The presence of com- for 3 h. The reaction mixture was dissolved in CH2Cl2 (30 mL),
1574

Beilstein J. Org. Chem. 2014, 10, 1570–1577.
and the organic layer was washed with brine (2 × 10 mL) and added slowly until the pH was ~7. This neutralized solution was
water (2 × 10 mL), dried over MgSO4, filtered and the solvent placed in a dialysis membrane (MW cutoff 3500 Da) and
was removed in vacuo. The residual product was taken up in dialyzed in 1 L of deionized water for 8 h. The water was
CH2Cl2 (50 mL) with 2-(2-isothiocyantoethoxy)ethanol (0.6 g, changed and let stand for a further 8 h twice more. The
4
mmol) and 4 Å molecular sieves. BF3OEt2 (0.6 g, 4 mmol) remaining liquid in the membrane was frozen and lyophilized to
was added to the mixture over 30 min at 0 °C, and the reaction give a white fluffy solid. This procedure for deacetylation was
mixture was let stir and warmed to room temperature over 2 h. performed in a manner similar to our previously described
The solvent was removed and the residue was taken up in ethyl procedure [29]. Characterization data for dendrimers is
acetate (50 mL), washed with saturated aqueous NaHCO3 solu- provided in Supporting Information File 1.
tion (2 × 20 mL), brine (2 × 20 mL), and water (1 × 20 mL),
dried over MgSO4, filtered and the solvent was removed in NMR spectroscopy
vacuo. The oily residue was purified by silica gel column chro- 1H NMR spectra were recorded on Bruker DPX 300 (300 MHz)
matography with a 60:40 ethyl acetate/hexanes eluent, fol- and Bruker DPX-500 (500 MHz) spectrometers. Chemical
lowed by a 20:1 ethyl acetate/MeOH eluent to yield 2.6 g of shifts are reported in ppm from tetramethylsilane with the
product. 1H NMR (300 MHz, CDCl3) δ 5.33 (d, J = 3.1 Hz, 1H, residual protic solvent resonance as the internal standard (chlo-
H4’), 5.20 (app t, J = 9.3 Hz, 1H, H3), 5.09 (dd, J = 8.1 and roform: δ 7.25 ppm; dimethyl sulfoxide: δ 2.50 ppm). Data are
1
0.1 Hz, 1H, H2’), 4.93 (m, 2H, H2 and H3’), 4.50 (m, 3H, H1, reported as follows: chemical shift, multiplicity (s = singlet, bs
H1’ and H6), 4.06 (m, 4H), 3.89 (m, 6H), 3.78 (m, 3H), 3.63 = broad singlet, d = doublet, t = triplet, q = quartet, p = pentet,
m, 6H), 2.13 (s, 3H), 2.10 (s, 3H), 2.04 (s, 3H), 2.02 (m, 9H), m = multiplet, app = apparent), integration, coupling constants
(
1
.95 (s, 3H). As reported [37].
(in Hz) and assignments. Sample NMR spectra are provided in
Figures S2 through S6, Supporting Information File 1.
General procedure for the synthesis of acetyl-
protected lactose-functionalized dendrimers
MALDI–TOF mass spectrometry
An aqueous solution of amine terminated G(4)-PAMAM MALDI mass spectra were acquired using a Bruker Biflex-III
dendrimer (2.48 g of a 17% w/w solution in water, 421 mg, time-of-flight mass spectrometer. Spectra of all-functionalized
3
1.2 μmol) was lyophilized to leave a foamy residue. 7.02 mL dendrimers were obtained using a trans-3-indolacrylic acid
of DMSO was added to this residue to give a 60 mg/mL solu- matrix with a matrix:analyte ratio of 3000:1 or 1000:1. Bovine
tion of the dendrimer. 0.47 mL of a 300 mM solution of 1 serum albumin (Mw 66,431 g/mol), cytochrome C (Mw 12,361
(
184 mg, 141 μmol) was added to 0.5 mL of the 60 mg/mL g/mol), and trypsinogen (Mw 23,982 g/mol) were used as
G(4) PAMAM dendrimer (30 mg, 2.20 μmol) solution. The external standards. An aliquot corresponding to 12–15 pmol of
mixture was stirred for 8 h at which point a 75 μL aliquot was the analyte was deposited on the laser target. Positive ion mass
collected and lyophilized for MALDI–TOF and NMR analysis. spectra were acquired in linear mode and the ions were gener-
The remainder of the reaction mixture was lyophilized and ated by using a nitrogen laser (337 nm) pulsed at 3 Hz with a
subjected to the deacetylation procedure. This procedure for pulse width of 3 nanoseconds. Ions were accelerated at
carbohydrate functionalization of dendrimers was performed in 19,000–20,000 volts and amplified using a discrete dynode
a manner similar to our previously described procedure [29]. multiplier. Spectra (100 to 200) were summed into a LeCroy
Amounts used in the syntheses of 2–5 are provided in Table S1, LSA1000 high-speed signal digitizer. All data processing was
Supporting Information File 1. Characterization data for ace- performed using Bruker XMass/XTOF V 5.0.2. Molecular mass
tylated precursors of 2–5 are provided in Supporting Informa- data and polydispersities (PDI) of the broad peaks were calcu-
tion File 1.
lated by using the Polymer Module included in the software
package. The peaks were analysed using the continuous mode.
Mw values for 2–5 are provided in Table S2, Supporting Infor-
mation File 1.
General procedure for deacetylation to afford
lactose-functionalized dendrimers
To the lyophilized solid per-O-acetylated dendrimers, 1 mL of
1
:1 water/methanol was added, at which point the dendrimer Dynamic light scattering (DLS)
became a white precipitate. To this mixture was added 0.2 equiv DLS measurements were acquired with the Brookhaven 90Plus
of NaOMe (0.8 M in MeOH) for each peripheral carbohydrate, Particle Size Analyzer equipped with a 15 mW solid state,
and let stir for 3 h. If, at this time, the mixture had not become a 633 nm laser and upgraded APD detector. Scattered light was
clear solution a further 0.2 equiv of NaOMe (0.8 M in MeOH) detected at 90° incidence and optimized to a count rate of
was added and this step was repeated until the mixture became 200–400 kilocounts per second (kcps) through adjustment of a
a clear and colourless solution. Aqueous HCl (0.1 M) was then neutral density filter prior to the sample chamber. The intensity
1575

Beilstein J. Org. Chem. 2014, 10, 1570–1577.
was maximized for samples producing less than 200 kcps. producing galectin-3. This paper is dedicated to the memory of
Temperature control was stabilized at 25 °C, and each sample Chris Wolfenden, who courageously battled cancer.
was scanned for 5 min (3 min for control samples). Autocorre-
lation curves were analyzed via the provided software using the References
method of cumulants (quadratic fit) unless otherwise noted. 1. Mammen, M.; Choi, S.-K.; Whitesides, G. M. Angew. Chem., Int. Ed.
1
998, 37, 2754–2794.
doi:10.1002/(SICI)1521-3773(19981102)37:20<2754::AID-ANIE2754>3
0.CO;2-3
This provided the effective diameter and relative variance
reported below. For dust-free, relatively monodispersed
samples, this analysis provided results similar to NNLS and
CONTIN algorithms. Protocols for preparation of solutions and
a brief discussion of the theory for DLS are provided in
Supporting Information File 1. Representative fitted data is
shown in Supporting Information Fle 1, Figure S7 for 4 and 5.
.
2
.
Choi, S.-K. Synthetic multivalent molecules; Wiley: New York, 2004.
doi:10.1002/0471578908
3. Klyosov, A. A.; Zbigniew, D. P.; Witczak, J.; Platt, D. Galectins; John
Wiley and Sons, 2008. doi:10.1002/9780470378076
4.
Barondes, S. H.; Castronovo, V.; Cooper, D. N. W.; Cummings, R. D.;
Drickamer, K.; Feizi, T.; Gitt, M. A.; Hirabayashi, J.; Hughes, C.;
Kasai, K.; Leffler, H.; Liu, F.-T.; Lotan, R.; Mercurio, A. M.;
Monsigny, M.; Pillai, S.; Poirer, F.; Raz, A.; Rigby, P. W. J.; Rini, J. M.;
Wang, J. L. Cell 1994, 76, 597–598.
Fluorescence microscopy
Fluorescence images were captured on either a Nikon Eclipse
TE2000-U with a 60× oil immersion objective lens (2, 3, 4) or
Olympus BX-61 with a 100× oil immersion objective (3, 4, 5).
Exposure time was optimized for each sample and 20–30
images were taken. These images were combined using Gimp 2
image manipulation software. Fluorescent microsphere stan-
dards (190, 520, 1020, and 1900 nm, Bangs Laboratories, Inc.)
were used to calibrate the measured particle perimeter (pixels)
to particle diameter (nm) for each exposure setting (Supporting
Information File 1, Figure S8, equations in Table S4). Diame-
ters of standards were verified with DLS (Table S5, Supporting
Information File 1). The imaged particle perimeters were deter-
mined through Pixcavator Image Analysis software (Intelligent
Perception). The y-intercept of each calibration curve repre-
sents the lower detection limit for the given exposure time.
Protocols for preparation of samples for fluorescence
microscopy are provided in Supporting Information File 1.
doi:10.1016/0092-8674(94)90498-7
5
.
.
Barondes, S. H.; Cooper, D. N. W.; Gitt, M. A.; Leffler, H. J. Biol. Chem.
1994, 269, 20807–20810.
6
Gray, C. A.; Adelson, D. L.; Bazer, F. W.; Burghardt, R. C.;
Meeusen, E. N. T.; Spencer, T. E. Proc. Natl. Acad. Sci. U. S. A. 2004,
101, 7982–7987. doi:10.1073/pnas.0402669101
7. Kasai, K.; Hirabayashi, J. J. Biochem. 1996, 119, 1–8.
doi:10.1093/oxfordjournals.jbchem.a021192
8
.
.
Ochieng, J.; Furtak, V.; Lukyanov, P. Glycoconjugate J. 2002, 19,
27–535. doi:10.1023/B:GLYC.0000014082.99675.2f
Califice, S.; Castronovo, V.; Van Den Brule, F. Int. J. Oncol. 2004, 25,
83–992.
5
9
9
10.Stillman, B. N.; Hsu, D. K.; Pang, M.; Brewer, C. F.; Johnson, P.;
Liu, F.-T.; Baum, L. G. J. Immunol. 2006, 176, 778–789.
doi:10.4049/jimmunol.176.2.778
1
1.Friedrichs, J.; Manninen, A.; Muller, D. J.; Helenius, J. J. Biol. Chem.
008, 283, 32264–32272. doi:10.1074/jbc.M803634200
2
1
2.Delacour, D.; Greb, C.; Koch, A.; Salomonsson, E.; Leffler, H.;
Le Bivic, A.; Jacob, R. Traffic 2007, 8, 379–388.
doi:10.1111/j.1600-0854.2007.00539.x
13.Nieminen, J.; Kuno, A.; Hirabayashi, J.; Sato, S. J. Biol. Chem. 2007,
Supporting Information
282, 1374–1383. doi:10.1074/jbc.M604506200
1
4.Nangia-Makker, P.; Raz, T.; Tait, L.; Hogan, V.; Fridman, R.; Raz, A.
Cancer Res. 2007, 67, 11760–11768.
Supporting Information File 1
Amounts of reagents used in glycodendrimer syntheses;
characterization data for glycodendrimers; sample
calculations; detailed protocols for galectin-3 isolation and
solution and sample preparations; sample NMR spectra;
characterization data for glycodendrimer aggregates.
doi:10.1158/0008-5472.CAN-07-3233
15.Ahmad, N.; Gabius, H.-J.; André, S.; Kaltner, H.; Sabesan, S.; Roy, R.;
Liu, B.; Macaluso, F.; Brewer, C. F. J. Biol. Chem. 2004, 279,
10841–10847. doi:10.1074/jbc.M312834200
1
1
1
1
6.Ochieng, J.; Platt, D.; Tait, L.; Hogan, V.; Raz, T.; Carmi, P.; Raz, A.
Biochemistry 1993, 32, 4455–4460. doi:10.1021/bi00067a038
7.Kuklinski, S.; Probstmeier, R. J. Neurochem. 1998, 70, 814–823.
doi:10.1046/j.1471-4159.1998.70020814.x
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-10-162-S1.pdf]
8.Lepur, A.; Salomonsson, E.; Nilsson, U. J.; Leffler, H. J. Biol. Chem.
2012, 287, 21751–21756. doi:10.1074/jbc.C112.358002
9.Fermino, M. L.; Polli, C. D.; Toledo, K. A.; Liu, F.-T.; Hsu, D. K.;
Roque-Barreira, M. C.; Pereira-da-Silva, G.; Bernardes, E. S.;
Halbwachs-Mecarelli, L. PLoS One 2011, 6, e26004.
doi:10.1371/journal.pone.0026004
Acknowledgements
This research was supported by NIH GM62444 and NSF
1
214134. We thank Dr. Sandra Halonen and Dr. Trevor
Douglas for making their instruments available for use, Anna
Michel for synthesis work and assistance with Pixcavator, Jenni
Monaco for help preparing the Alexa488 galectin-3 conjugate,
and Julie Sprenger, Julianna Weiel, and Kylene Compton for
20.Quattroni, P.; Li, Y.; Lucchesi, D.; Lucas, S.; Hood, D. W.;
Herrmann, M.; Gabius, H.-J.; Tang, C. M.; Exley, R. M. Cell. Microbiol.
2012, 14, 1657–1675. doi:10.1111/j.1462-5822.2012.01838.x
1576

Beilstein J. Org. Chem. 2014, 10, 1570–1577.
2
1.Kavanaugh, D.; Kane, M.; Joshi, L.; Hickey, R. M.
Appl. Environ. Microbiol. 2013, 79, 3507–3510.
doi:10.1128/AEM.03694-12
License and Terms
2
2.Sörme, P.; Arnoux, P.; Kahl-Knutsson, B.; Leffler, H.; Rini, J. M.;
Nilsson, U. J. J. Am. Chem. Soc. 2005, 127, 1737–1743.
doi:10.1021/ja043475p
This is an Open Access article under the terms of the
Creative Commons Attribution License
(http://creativecommons.org/licenses/by/2.0), which
2
3.Ahmad, N.; Gabius, H.-J.; Sabesan, S.; Oscarson, S.; Brewer, C. F.
Glycobiology 2004, 14, 817–825. doi:10.1093/glycob/cwh095
4.Tellez-Sanz, R.; Garcia-Fuentes, L.; Vargas-Berenguel, A.
Curr. Med. Chem. 2013, 20, 2979–2990.
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
2
doi:10.2174/09298673113209990163
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
2
5.Kopitz, J.; Bergmann, M.; Gabius, H.-J. IUBMB Life 2010, 62, 624–628.
doi:10.1002/iub.358
(http://www.beilstein-journals.org/bjoc)
2
6.Newkome, G. R.; Moorefield, C. N.; Vögtle, F. Dendrimers and
Dendrons: Concepts, Syntheses, Applications; Wiley-VCH: Weinheim,
The definitive version of this article is the electronic one
which can be found at:
2
001. doi:10.1002/3527600612
2
2
2
3
7.Labieniec, M.; Watala, C. Cent. Eur. J. Biol. 2009, 4, 434–451.
doi:10.2478/s11535-009-0056-7
doi:10.3762/bjoc.10.162
8.Pourianazar, N. T.; Mutlu, P.; Gunduz, U. J. Nanopart. Res. 2014, 16,
No. 2342. doi:10.1007/s11051-014-2342-1
9.Wolfenden, M. L.; Cloninger, M. J. Bioconjugate Chem. 2006, 17,
958–966. doi:10.1021/bc060107x
0.Walter, E. D.; Sebby, K. B.; Usselman, R. J.; Singel, D. J.;
Cloninger, M. J. J. Phys. Chem. B 2005, 109, 21532–21538.
doi:10.1021/jp0515683
3
3
3
3
1.Ottaviani, M. F.; Cangiotti, M.; Fiorani, L.; Fattori, A.; Wasiak, T.;
Appelhans, D.; Klajnert, B. Curr. Med. Chem. 2012, 19, 5907–5921.
doi:10.2174/092986712804143259
2.Jackson, C. L.; Chanzy, H. D.; Booy, F. P.; Drake, B. J.; Tomalia, D. A.;
Bauer, B. J.; Amis, E. J. Macromolecules 1998, 31, 6259–6265.
doi:10.1021/ma9806155
3.Tomalia, D. A.; Naylor, A. M.; Goddard, W. A., III.
Angew. Chem., Int. Ed. Engl. 1990, 29, 138–175.
doi:10.1002/anie.199001381
4.Seetharaman, J.; Kanigsberg, A.; Slaaby, R.; Leffler, H.;
Barondes, S. H.; Rini, J. M. J. Biol. Chem. 1998, 273, 13047–13052.
doi:10.1074/jbc.273.21.13047
3
3
3
5.Yang, J.; Hlavacek, W. S. Math. Biosci. 2011, 232, 164–173.
doi:10.1016/j.mbs.2011.06.003
6.Schlick, K. H.; Lange, C. K.; Gillispie, G. D.; Cloninger, M. J.
J. Am. Chem. Soc. 2009, 131, 16608–16609. doi:10.1021/ja904073p
7.Seah, N.; Santacroce, P. V.; Basu, A. Org. Lett. 2009, 11, 559–562.
doi:10.1021/ol802613r
1577