Self-assembled carbohydrate-based micelles for lectin targeting

Article abstract of DOI:10.1039/c0sm01411g

Biocompatible low-polydispersity micelles designed for lectin targeting have been prepared by spontaneous self-assembly in water of macromolecular glycosylated amphiphiles. Propargyl-blactoside and N-acetyl-b-D-glucosaminide were conjugated by copper-cata

Full text of DOI:10.1039/c0sm01411g

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Soft Matter
Cite this: Soft Matter, 2011, 7, 3453
www.rsc.org/softmatter
PAPER
Self-assembled carbohydrate-based micelles for lectin targeting†
ꢀ ab
Alexandre G. Dal Bo, Valdir Soldi,b Fernando C. Giacomelli,c Bruno Jean,a Isabelle Pignot-Paintrand,a
a
Redouane Borsali and Sebastien Fort
a
*
ꢀ
Received 3rd December 2010, Accepted 21st January 2011
DOI: 10.1039/c0sm01411g
Biocompatible low-polydispersity micelles designed for lectin targeting have been prepared by
spontaneous self-assembly in water of macromolecular glycosylated amphiphiles. Propargyl-b-
lactoside and N-acetyl-b-D-glucosaminide were conjugated by copper-catalyzed Huisgen cycloaddition
to azide-terminated PEG 900 stearate. Upon dissolution in water, the resulting amphiphiles
immediately self-assemble into highly regular micelles having a mean diameter of 10 nm. Dynamic
Light Scattering (DLS), Transmission Electron Microscopy (TEM) and Small-Angle X-ray Scattering
(SAXS) were used to investigate the structure of the self-assembled saccharidic amphiphiles micelles.
The presence of the carbohydrate epitopes on the surface of the micelles and their bioavailability for
lectin targeting were also demonstrated by light scattering measurements. Specific interaction of the
GlcNac and Lac residues with Wheat Germ Agglutinin (WGA) and Peanut Agglutinin (PNA)
respectively, unveils potential applications of such carbohydrate-derived surfactants as simple and
site-specific vectorization systems for drug delivery.
interaction with specific cell receptors for a site-directed delivery
Introduction
of the drugs.
Hierarchical self-assembly of synthetic molecules is a subject of
great current interest and a challenging topic of interdisciplinary
research in chemistry, biology, and materials science. Amphi-
philic molecules, namely those that carry both hydrophilic and
hydrophobic parts within one structure self-assemble into
a variety of supramolecular objects including micellar structures,
ellipsoids, disks, cylinders, vesicles, and lamellae.1–6 This
assembly usually depends strongly on the molecular architecture,
concentration, solvent environment and is driven by non-cova-
lent interactions such as hydrogen bonding, p–p stacking or
electrostatic interactions.7 In particular, spherical micelles have
attracted much attention for their use as drug carrier.8 Encap-
sulation of hydrophobic drugs into self-assembled particles has
been shown to enhance their water-solubility thus reducing their
toxicity and to contribute to more sustained release profiles.9–11
The latest developments in this field focus on the introduction of
functional groups at the surface of the drug carrier such as
peptides,12 antibodies13 or carbohydrates14 aiming at the
Over the last decade, multivalent display of carbohydrates at
the cell surface has been demonstrated to play a critical role in
many biological events. Carbohydrate–protein interactions are
involved in standard as well as pathological processes including
fertilization, viral and bacterial infection, inflammatory
response, cell adhesion, and metastatic spreading.15 Carbohy-
drate-based nanoparticles have thus recently received much
attention as site-directed drug carriers. The preparation of
particles from amphiphilic graft- or block-copolymers contain-
ing a polysaccharidic scaffold including dextran, chitosan, hya-
luronic acid or water-soluble cellulose derivatives is abundantly
documented.16,17 Although biomass-derived polysaccharides
provide a valuable scaffold for the elaboration of nanoparticles,
they barely interact with specific cell-surface carbohydrate
receptors. The binding of lectins typically involves only the two
or three terminal sugar residues of the mammalian glycans
including galactose, mannose, N-acetyl-neuraminic acid, fucose
or N-acetyl-glucosamine.18 Recently, Kim et al. reported a series
of carbohydrate amphiphiles prepared around PEG scaffolds
and polyaromatic hydrophobic blocks.19 Self-assembled nano-
objects of different sizes and morphologies were obtained in
water depending on the molecular architecture as well as the
volume fraction of each constituent, and the bioavailability of
the sugar units for Escherichia coli receptors was confirmed.
Although the biocompatibility and the immunogenicity of such
drug carrier need to be investigated further into details, the
perspectives offered by these glycosylated macromolecular
amphiphiles are extremely broad.
aCentre de Recherches sur les Macromol
eꢀcules Veꢀgeꢀtales
ꢀ
(CERMAV-CNRS) (affiliated with Universite Joseph Fourier Grenoble
1 and member of the Institut de Chimie Moleꢀculaire de Grenoble), BP
53, F-38041 Grenoble Cedex 9, France. E-mail: sebastien.fort@cermav.
cnrs.fr; Fax: +33 (0)4 76 54 72 03; Tel: +33 (0)4 76 03 76 64
bDepartamento de Quꢀımica, Universidade Federal de Santa Catarina,
88040-900 Florianoꢀpolis, SC, Brazil
cCentro de Cie^ncias Naturais e Humanas, Universidade Federal do ABC,
09210-170 Santo Andreꢀ, SP, Brazil
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c0sm01411g
This journal is ª The Royal Society of Chemistry 2011
Soft Matter, 2011, 7, 3453–3461 | 3453

View Article Online
Aiming at the development of new self-assembled nano-
particles whose surface is decorated by carbohydrates with
potential applications for the site-directed vectorization of active
molecules, we have designed a series of carbohydrate-containing
amphiphilic macromolecules from a synthetic PEG ester.
Poly(ethylene glycol) and its derivatives have several physico-
chemical properties that make them especially important in
various biological, chemical and pharmaceutical settings where
one can cite the wide range of solubility in both organic and
aqueous media, the lack of toxicity and immunogenicity, non-
biodegradability and ease of excretion from living organisms.20
PEG is often used as a covalent modifier of a variety of substrates
producing conjugates that combine some of the properties of
both the starting substrate and the polymer. Blood lifetime of
PEGylated liposomes, nanoparticles, and proteins can be
significantly extended and their uptake by reticuloendothelial
system (RES) organs, liver and spleen diminished.21–23 Fatty acid
esters of PEGs, commonly known as macrogolglycerides are
produced industrially and used in a variety of lipid-based
formulations in cosmetic and pharmaceutical applications.
Currently, these lipidic excipients are used as bioavailability
enhancers in, for example, proxyphylline release from matrix
hard gelatin capsules,24 oral formulation of nicotine,25 and
dermal application of cyclosporin A.26
experiments. High-resolution mass spectrometry measurements
were performed on a Micromass ZABSpec-Tof spectrometer at the
Centre Regional de Mesures Physiques de l’Ouest CRMPO
(Rennes, France). The infrared (IR) spectra were recorded using
a Perkin-Elmer Spectrum RXI FTIR spectrometer.
Synthesis
Propargyl 2-acetamido-2-deoxy-b-D-glucopyranoside29 and
propargyl b-lactoside30 were synthesized as previously reported
in the literature.
Mono-tosyl-PEG 900 (1). To a chilled (0 ꢀC) solution of PEG
900 (8.0 g, 8.9 mmol) in anhydrous methylene chloride (250 mL)
was added silver oxide (3.1 g; 13.4 mmol) and potassium iodide
(0.6 g; 3.6 mmol). Recrystallized tosyl chloride TsCl (1.78 g;
9.4 mmol) was then added in one portion. The reaction was
monitored by TLC and after completion (ꢁ2 hours), the mixture
was filtered over pad of Celiteꢀ and evaporated under reduced
pressure to give a colorless oily product. Pure 1 (5.16 g; 55%) was
isolated after purification by silica gel flash chromatography
using methylene chloride : methanol (9 : 1 v/v). 1H NMR:
(300 MHz, CDCl3, ppm): d ¼ 7.78 (d, 2H, J ¼ 8.5 Hz, arom.),
7.31 (d, 2H, J ¼ 8.5 Hz, arom.), 4.12 (t, 2H, J ¼ 4.4 Hz,
CH2OTs), 3.71–3.55 (m, ꢁ78H, CH2O), 2.42(s, 3H, CH3).13C
NMR: (75 MHz, CDCl3, ppm): d ¼ 144.8, 133.0, 129.8, 128.0,
72.5, 70.7, 70.5, 70.3, 69.2, 68.6, 61.7, 21.6. MALDI-TOF MS
calcd for C47H88O23S: m/z 1052; found: m/z 1075 [M + Na]+.
The present work reports the synthesis, self-assembly, and
interaction study of new amphiphiles targeting lectins prepared
by click-chemistry grafting of propargylated carbohydrates onto
azido-PEG900-stearate. Stearic acid, one of the most common
fatty acids, is widely used as a lubricant, in soaps, cosmetics, and
food packaging. It is a component of Solutol HS1527 and Gelucir
50/1328 which has been proved to be an effective stabilizer for the
preparation of lipid particles. On the other hand, PEG having an
average molecular weight of 900 was selected to act as a flexible
hydrophilic spacer of appropriate size to provide amphiphiles
with a hydrophobic volume fraction close to 20% that is required
for the formation of spherical micelles.10
Mono-azide-PEG 900 (2). A solution containing 1 (2.5 g,
2.4 mmol) and NaN3 (0.618, 9.5 mmol) dissolved in anhydrous
DMF (25 mL) was heated at 60 ꢀC in an oil bath for 18 hours and
then concentrated. The residue was purified by silica gel flash
column chromatography using methylene chloride : methanol
1
(9 : 1 v/v) as eluent to yield pure azide PEG 2 (1.86 g, 85%). H
NMR: (300 MHz, CDCl3, ppm): d ¼ 3.71–3.58 (m, ꢁ78H,
CH2O), 3.37 (m, 2H, CH2N3). 13C NMR: (75 MHz, CDCl3,
ppm): d ¼ 72.5, 70.6, 70.30, 70.0, 61.7, 50.7. FT-IR spectra
(g, cmꢂ1): azide absorbance peak 2100 cmꢂ1. MALDI-TOF MS
calcd for C40H81N3O20: m/z 923.61, found: m/z 946.56 [M + Na]+.
Experimental section
Materials
All reagents were of commercial grade and they were used as
received unless otherwise noticed. Triticum vulgaris (wheat germ)
lectin (WGA) and Arachis hypogaea (peanut) lectin (PNA) were
obtained from EY Laboratories, Inc. The solvents were dried
and distilled according to literature procedures before use. The
reactions were monitored by TLC using Silica Gel 60 F254
precoated plates (E.Merck, Darmstadt) and the detection was
achieved by exposure to iodine vapors or by charring with
sulfuric acid solution 3 : 45 : 45 H2SO4 : MeOH : H2O. For flash
chromatography, E. Merck Silica Gel 60 was used.
Azide-PEG 900-stearate (3). N,N’-Dicyclohexylcarbodiimide
(0.20 g,
1 mmol) and 4-dimethylaminopyridine (DMAP)
(0.024 g, 0.2 mmol) were added into a stirred solution of stearic
acid (0.85 g, 3.0 mmol) and mono-azide-PEG 900 (2) (0.92 g,
1 mmol) in anhydrous methylene chloride (30 mL). After
48 hours at room temperature under N2 atmosphere, the mixture
was filtered over pad of Celiteꢀ and evaporated under reduced
pressure to give a colorless oily product. Pure stearate 3 (0.976,
82%) was isolated after purification by silica gel flash column
chromatography using methylene chloride : methanol (9 : 1 v/v)
1
as eluent. H NMR: (400 MHz, MeOD, ppm): d ¼ 4.21 (t, 2H,
Characterization of the amphiphiles
J ¼ 4.8 Hz), 3.71–3.64 (m, ꢁ82H, CH2O), 3.37 (t, 2H, J ¼ 4.9
Hz), 2.36–2.26 (m, 2H), 1.62–1.60 (m, 2H), 1.29 (m, 28H), 0.90 (t,
3H, J ¼ 6.6 Hz, CH3). 13C NMR: (100MHz, DMSO-d6, ppm):
d ¼ 172.9, 69.9, 69.3, 68.4, 63.1, 50.1, 33.5, 31.4, 30.7, 29.1, 28.9,
28.8, 28.5, 24.5, 22.2, 14.0. FT-IR spectra: azide absorbance peak
2100 cmꢂ1. MALDI-TOF MS calcd for C58H115N3O21: m/z 1189,
found: m/z 1212 [M + Na]+.
The NMR spectra were recorded on a Bruker AC 300 or Bruker
Avance 400 spectrometer at 25 ꢀC and they were calibrated against
theresidualsignalofthesolvent.Protonchemicalshiftsarereported
in ppm relative to external SiMe4 (0 ppm). Low-resolution mass
spectrawererecordedonaBrukerDaltonicsAutoflexapparatusfor
MALDI and on a Waters Micromass ZQ spectrometer for ESI
3454 | Soft Matter, 2011, 7, 3453–3461
This journal is ª The Royal Society of Chemistry 2011

View Article Online
N-Acetyl-b-D-glucosaminyl-PEG 900-stearate conjugate (4).
Copper sulfate (53.4 mg, 0.34 mmol) and sodium ascorbate
(73.0 mg, 0.37 mmol) were added into a stirred solution of
propargyl-2-acetamido-2-deoxy-b-D-glucopyranoside (0.113 g,
0.44 mmol) and azide-PEG 3 (0.4 g, 0.34 mmol) in water/THF
(1 : 1 v/v). The mixture was heated at 40 ꢀC and the reaction was
monitored by TLC. After 24 hours, the solution was concen-
trated under vacuum. The residue was taken up in methylene
chloride, filtered and concentrated. Pure C18PEG900GlcNAc 4
(0.35 g, 72%) was isolated after silica gel flash column chroma-
tography using methylene chloride : methanol (8 : 2 v/v) as
spectroscopy using pyrene (Py) as a probe.31 Measurements of
steady-state fluorescence of pyrene were performed in Milli-Q
water at 25.0 ꢀC. Stock solution of pyrene (1 mM) was prepared
in dry ethanol. An aqueous solution of pyrene (10ꢂ6 M) was then
prepared by adding 0.1 mL of the pyrene stock solution into
100 mL of Milli-Q water. The solutions were then prepared from
the stock solution containing pyrene (10ꢂ6 M) and they were
allowed to equilibrate for at least 12 h prior to the acquisition of
emission spectra. This procedure minimizes errors in the emis-
sion intensity occurring in alternative methodologies.
The steady-state fluorescence spectra of pyrene were recorded
on a Perkin-Elmer LS50B Spectrofluorimeter equipped with
a thermostatted cell holder set at 25.0 ꢀC. Both slits of excitation
and emission monochromators were adjusted for 2.5 nm. The
samples were excited at 335 nm and the emission intensity spectra
were recorded at 372.8 nm (I1) and 384.0 nm (I3). Typically, the
fluorescence spectrum was recorded after the addition of each
microlitre of amphiphilic solution. The relative intensities were
corrected for each sample dilution prepared from the pyrene
stock solution. The I1/I3 ratio was estimated by taking into
account the ratio of the maximum peak intensity.
1
eluent. H-NMR (400 MHz, DMSO-d6, ppm): d ¼ 7.93 (s, 1H,
triazolyl), 7.49 (d, 1H, J ¼ 8.6 Hz, NH), 4.80–4.40 (m, 8H), 4.13
(t, 2H, J ¼ 4.2 Hz), 3.82 (t, 2H, J ¼ 4.8 Hz) 3.61–3.50 (m, 82H,
CH2O), 3.15 (2H, J ¼ 4.8 Hz), 2.28 (t, 2H, J ¼ 4 Hz), 2.08
(s, 11H), 1.78 (s, 3H), 1.5 (m, 2H), 1.24 (m, 18H), 0.86 (t, 3H,
J ¼ 6.6Hz, CH3). 13C NMR: (75 MHz, DMSO-d6, ppm): d ¼
172.8, 168.9, 143.5, 124.2, 100.2, 77.0, 74.2, 71.6, 70.7, 69.7, 69.6,
69.5, 68.7, 68.3, 67.9, 62.9, 61.2, 61.1, 55.3, 33.4, 31.2, 29.0, 28.8,
28.6, 28.4, 24.4, 23.0, 22.0, 13.9. ESI-HRMS calcd for
C69H132N4O27: m/z 1471.89767 [M + Na]+; found: 1471.8958.
b-Lactosyl-PEG 900-stearate conjugate (5). Copper sulfate
(61 mg, 0.39 mmol) and sodium ascorbate (79.2 mg, 0.40 mmol)
were added to a stirred solution of propargyl-b-lactoside
(0.190 g, 0.5 mmol) and azide-PEG900 3 (0.456 g, 0.39 mmol) in
Dynamic light scattering (DLS). The size and polydispersity of
the nanoparticles were accessed using DLS measurements. The
experiments were performed using an ALV Laser Goniometer,
which consists of a cylindrical 22 mW HeNe linear polarized
laser with l ¼ 632.8 nm and an ALV-5000/ALV Multiple Tau
Digital Correlator with a 125 ns initial sampling time.32 The
samples were kept at 25 ꢀC. The accessible scattering angle of the
equipment ranges from 20ꢀ to 150ꢀ. All samples were systemat-
ically studied at 90ꢀ and some of them were studied at different
scattering angles varying from 40ꢀ to 140ꢀ. The solutions were
put in ordinary 10 mm in diameter glass cells. The minimum
sample volume required for an experiment was 1 mL. The data
were acquired with the ALV-Correlator Control Software and
the counting time for each sample was in average 300 s. The
distributions of relaxation times—A(t)—were obtained by using
the CONTIN analysis of the auto-correlation function C(q,t).33
The relaxation frequency, G ¼ 1/s is a function of the scattering
angle.34 The apparent diffusion coefficient (Dapp) of the nano-
particles at a given copolymer concentration (Cp) is calculated
from eqn (1).
ꢀ
water/THF (1 : 1 v/v). The mixture was heated at 40 C and the
reaction was monitored by TLC. After 24 hours, the solution
was concentrated under vacuum. The residue was taken up in
methylene chloride, filtered and concentrated. Pure
C18PEG900Lac 5 (0.403 g, 67%) was isolated after silica gel flash
column chromatography using methylene chloride : methanol
1
(7 : 3 v/v) as eluent. H NMR: (400 MHz, DMSO-d6, ppm):
d ¼ 8.03 (1H, triazolyl), 4.91–4.32 (m, 12H), 4.12 (t, 2H, J ¼ 4.9
Hz), 3.82 (2H, J ¼ 4.8 Hz, CH2O) 3.58–3.50 (m, 78H, CH2O),
3.33 (2H, J ¼ 4.8 Hz), 2.29–2.27 (m, 2H), 1.5 (m, 2H), 1.24 (m,
28H), 0.86 (t, 3H, J ¼ 6.6Hz, CH3). 13C NMR: (75 MHz,
DMSO-d6, ppm): d ¼ 172.68, 143.39,124.35, 103.72, 101.71,
80.62, 77.06, 75.41, 74.85, 73.18, 73.03, 71.53, 70.48, 69.68,
69.54, 69.46, 68.61, 68.23, 68.03 67.81, 62.89, 61.53, 60.51,
60.29, 49.25, 33.34, 31.18, 28.90, 28.75, 28.56, 28.29, 24.36,
21.96, 13.83. ESI-HRMS calcd for C75H143N3O33: m/z
1636.95016 [M + Na]+; found: 1636.9507.
ꢀ
where q is the wave vector defined as
G
q2
ꢀ
q/0 ¼ Dapp
(1)
Nanoparticles preparation
ꢁ ꢂ
The nanoparticles suspensions (aqueous micellar solutions—
Cp ¼ 0.1–1.5 mg mLꢂ1) were prepared by direct dissolution of
the amphiphiles 3, 4 and 5 in Milli-Q water or in phosphate
buffered saline solution (PBS, 10 mM, pH 7ꢀ.2, 1 mM CaCl2,
1 mM MnCl2) and stirred for 24 hours at 25 C. The solutions
were then filtered using 0.45 mm pore size nylon membrane
filters in order to remove dust and large non-micellar aggre-
gates.
4pn
q
q ¼
sin
(2)
l
2
l being the wavelength of the incident laser beam (632.8 nm),
n the refractive index of the sample, and q the scattering angle.
The hydrodynamic radius (RH) (or diameter, 2RH) was calcu-
lated from the Stokes–Einstein relation given in eqn (3).
KBT
KBT
RH
¼
q2 ¼
(3)
6phG
6phDapp
Characterization of the nanoparticles
where kB is the Boltzmann constant, T is the temperature of
the sample and h is the viscosity of the solvent (water in this
case).35
Fluorescence spectroscopy. The critical micelle concentration
(CMC) of 3,
4 and 5 was determined by fluorescence
This journal is ª The Royal Society of Chemistry 2011
Soft Matter, 2011, 7, 3453–3461 | 3455

View Article Online
Transmission electron microscopy (TEM). To prepare the TEM
samples, 4 mL of an aqueous micellar solution was dropped onto
a glow-discharged carbon-coated copper grid. Then 4 mL of 2%
(w/v) uranyl acetate negative stain was added prior to complete
drying. After few minutes, the liquid in excess was blotted with
a filter paper and the grid was allowed to dry. The specimens
were observed using a Philips CM200 microscope operated at
80 kV. Images were recorded on Kodak SO163 films, and the
negatives were digitized off-line using a Kodak Mega plus CCD
camera.
Results and discussion
Synthesis of the amphiphiles
The synthesis, as depicted in Scheme 1, is initiated with selective
monotosylation of PEG 900 in the presence of silver oxide
(Ag2O) and a catalytic amount of potassium iodide (KI) as
reported by Bouzide et al.38 This method, which proved to be an
excellent protocol to mediate monotosylation of symmetrical
diols and oligo(ethylene glycol)s (OEGs), provided the
sulfonated PEG 1 in fairly good yield (55%). Substitution of the
tosyl group in the presence of sodium azide provided compound
2 in 85% yield. The absorption peak of the N3 group at 2100 cmꢂ1
observed by infrared spectroscopy and the 13C signal of the
methylene carbon adjacent to the azide group at 50.7 ppm seen
by NMR confirmed the efficiency of the transformation. The
stearic group was further introduced by an esterification reaction
promoted by dicyclohexylcarbodiimide leading to 82% yield of
the ‘‘ready-to-click’’ C18PEG900N3 3. Propargyl b-glycosides of
N-acetyl-glucosamine and lactose were introduced at the polar
head of the surfactant by copper-catalyzed ꢀHuisgen cycloaddi-
tion.39,40 The reaction was carried out at 40 C with the copper/
ascorbate catalytic system in a water/THF mixture to ensure
a good solubility of the amphiphile. Glycosylated conjugates
C18PEG900GlcNAc 4 and C18PEG900Lac 5 were respectively
isolated in 72 and 67% yields after purification by silica gel
chromatography.
Small-angle X-ray scattering (SAXS). The SAXS measure-
ments were performed at the French CRG Beamline D2AM-
BM02 of the European Synchrotron Radiation Facility (ESRF).
The wavelength (l) of the incoming beam and the sample-to-
detector distance were chosen in such way that the q-range,
q being equal to (4p/l)sin(q/2) (q is the scattering angle), could be
covered from 0.15 to 3.0 nmꢂ1. The samples were loaded into
sealed borosilicate capillaries (ꢁ2 mm diameter). The collimated
beam crossed the samples and was scattered to an indirect illu-
mination CCD detector (Princeton Instruments). In all cases the
2D-images were found to be isotropic and they were corrected by
taking into account the detector dark noise and normalized by
the sample transmission. The I(q) vs. q curves resulting from the
360ꢀ azimuthal integration of the 2D-patterns were further cor-
rected by the subtraction of the scattering of the pure solvent
(water). The I(q) vs. q scattering profile of the nanoparticles could
be fitted by using the spherical copolymer micelle model devel-
oped by Pedersen and Gerstenberg.36 The fitting procedures and
other analyses were performed using the SASfit software, which
makes use of the least-square fitting approach for minimizing the
squared chi (c2) parameter. The SASfit software package was
developed by J. Kohlbrecher and it is available online.37
Although the amphiphilic behavior of the molecules compli-
cates their characterization, the structure of 4 and 5 could be
unambiguously confirmed by mass spectrometry and NMR
spectrometry in DMSO, a good solvent for each part of the
molecule. Both 1H and 13C NMR spectra displayed characteristic
signals of the aliphatic chains, the ethylenoxide parts and the
carbohydrate units as well as the signals of the triazolyl ring.
Scheme 1 Synthetic strategy used in the preparation of C18PEG900GlcNAc 4 and C18PEG900Lac 5: (a) TsCl, Ag2O, KI, CH2Cl2, 55%; (b) NaN3, DMF,
85%; (c) DCC, DMAP, stearic acid, CH2Cl2, 82%; (d) CuSO4, sodium ascorbate, propargyl-2-N-acetamido-2-deoxy-b-D-glucopyranoside, H2O/THF,
72%; (e) CuSO4, sodium ascorbate, propargyl b-lactoside, H2O/THF, 67%.
3456 | Soft Matter, 2011, 7, 3453–3461
This journal is ª The Royal Society of Chemistry 2011

View Article Online
Fig. 1 depicts as representative example the 13C NMR spectra of
C18PEG900Lac 5 in (DMSO-d6).
Self-assembly of the amphiphiles
Fluorescence spectroscopy. The CMC is the critical micelle
concentration above which the amphiphile form micelles. Fluo-
rescent probe techniques such as those using pyrene (Py) have
been widely employed for monitoring the formation of amphi-
philic aggregates and to determine the CMC of the surfactants.
The I1/I3 ratio from Py fluorescence decreases for different
systems as the total amphiphilic concentration increases,
reflecting the incorporation of Py in hydrophobic domains due to
aggregate formation. Fig. 2 shows the I1/I3 intensity ratios
derived from Py emission spectra recorded for solutions of
C18PEG900N3 3 (filled circles) and C18PEG900GlcNAc 4 (open
circles). The CMC determined from the interception of the
horizontal and the steep parts of the curves were 0.033 mg mLꢂ1
for 3 and 0.055 mg mLꢂ1 for C18PEG900GlcNAc 4. The same
procedure was used to determine the CMC of the C18PEG900Lac
5 (0.060 mg mLꢂ1—data not shown). As expected, no significant
changes were observed between the unmodified and the glyco-
sylated surfactants since the hydrophilic character of the
different molecules is dictated by the PEG part.
Fig. 2 Pyrene fluorescence ratio I1/I3 as a function of the concentration
measured in water at 25 ꢀC for C18PEG900N3
3 (C) and
C
18PEG900GlcNAc 4 (B).
depicts the distribution of RH by taking into account the
contribution of the volume of particles instead of the light
scattering contribution where it can be easily noticed that the
presence of large aggregates can be neglected as clearly evidenced
by the TEM images (hereafter). The same rationalization is also
valid for the other protein-free explored systems.
Fig. 4 shows typical autocorrelation functions C(q,t) measured
at different scattering angles and the respective distributions of
the relaxation times A(t) at 90ꢀ as revealed by CONTIN analysis
for 0.5mg mLꢂ1 C18PEG900GlcNAc 4 and 0.5 mg mLꢂ1
C18PEG900Lac 5 dissolved in water at 25 ꢀC. The insets depict the
typical angular variation of the frequency G ¼ 1/s measured as
a function of q2 indicating a diffusive behavior of the scattering
particles.
Dynamic light scattering (DLS). Fig. 3 shows the autocorre-
lation function C(q,t) measured at 90ꢀ and the respective distri-
bution of the relaxation times A(t) as revealed by CONTIN
analysis for 0.5 mg mLꢂ1 solution of C18PEG900N3 3 in water.
The non-glycosylated intermediate 3 self-assembles in aqueous
solution into low-polydispersity nanoparticles (RH ¼ 10.4 nm).
Besides the intensity contribution related to the presence of well-
defined micelles, it can be noticed the presence of a light scat-
tering contribution attributed to the formation of large aggre-
gates with loose structure. However, the light scattering intensity
is heavily weighted by the particle mass and size implying that
DLS reports an intensity-average distribution. The inset in Fig. 3
The hydrodynamic radius of 4 and 5 was calculated by using
the Stokes–Einstein relation and one can notice that the values
are practically the same as compared to the value obtained before
the click chemistry reaction (amphiphile 3). The hydrodynamic
Fig. 1 13C NMR spectra of C18PEG900Lac 5 in (DMSO-d6).
This journal is ª The Royal Society of Chemistry 2011
Soft Matter, 2011, 7, 3453–3461 | 3457

View Article Online
demonstrating that the size of the micelles is concentration
independent in both cases (for C > CMC).
Transmission electron microscopy study. The TEM images
(Fig. 6) revealed for all the amphiphiles the formation of nano-
sized spherical structures in the range of 10 nm. Only a very small
number of large aggregates can be observed in the micrographs
suggesting that even though the dynamic light scattering tech-
nique is very sensitive to the presence of large aggregates, their
real number in the current systems is negligible and as mentioned
by several authors, the morphology, size and structure of
nanoparticles are mainly dependent on the balance of forces of
attraction and repulsion between the different blocks of the
amphiphiles. The spherical micelles found here are compatible
with the volumetric fraction of the hydrophobic region
(4) < 1/3.4,10,41 Indeed, the calculated hydrophobic volumetric
fraction of 3, 4 and 5 based on the density of each constituent was
equal to 0.28, 0.24 and 0.23 respectively, thus spherical micelles
are expected.4,10,35
Fig. 3 Typical autocorrelation function C(q,t) measured at q ¼ 90ꢀ and
the respective distribution of relaxation times A(t) as revealed by CON-
TIN analysis for an aqueous 0.5 mg mLꢂ1 solution of azido terminated
C
18PEG900N3 3 in water at 25 ꢀC. In the inset it is given the respective
distribution of RH by considering the contribution of the particles related
to their total volume.
Small-angle X-ray scattering (SAXS). In addition to TEM
images, SAXS measurements were performed in order to probe
the size, shape and internal structure of the scattering particles.
The scattering intensity I(q) of an isotropic solution of highly
regular particles embedded in a matrix with a constant electron
density, after normalization with the background scattering of
the solvent, is given by:
sizes of the particles are equal to 2RH ¼ 10.5 nm and 2RH ¼ 10.2 nm
for C18PEG900GlcNAc 4 and C18PEG900Lac 5 respectively.
Fig. 5 summarizes the values obtained for 2RH as a function of
the concentration of C18PEG900GLcNAc 4 and C18PEG900Lac 5
I(q) ¼ NP(q)S(q)
(4)
Wherein N is the number of scattering particles per unit volume,
P(q) is the form factor of an individual particle, and S(q) is the
structure factor arising from long-range correlations between
scattering centers. For widely separated (dilute) systems, S(q) ꢁ1,
and I(q) is consequently proportional to the form factor P(q) that
reflects the size and shape of the scattering objects.
Fig. 7 shows the SAXS patterns for 40 mg mLꢂ1 solution of
C18PEG900GlcNAc 4 (top) and C18PEG900Lac 5 (bottom) in
water at 25 ꢀC. Such copolymer concentration gives a reasonable
Fig. 4 Autocorrelation functionsC(q,t) measuredat scatteringangles 50ꢀ
(B), 90ꢀ (C), and 130ꢀ (,), and distributions of the relaxation times A(t)
at 90ꢀ as revealed by CONTIN analysis for (a) 0.5 mg mLꢂ1 C18PEG900Lac
5 and (b) 0.5 mg Lꢂ1 C18PEG900GLcNAc 4 in water at 25 ꢀC.
Fig.
18PEG900GLcNAc 4 (B) and C18PEG900Lac 5 (C) concentration
measured in water at 25 ꢀC.
5
Hydrodynamic diameter (2RH) as
a
function of
C
3458 | Soft Matter, 2011, 7, 3453–3461
This journal is ª The Royal Society of Chemistry 2011

View Article Online
flexible PEG chains. SAXS spectra in Fig. 7 clearly exhibit high
amplitude form-factor oscillations that can only be observed in
the case of well-defined particles with a narrow size distribution.
The SAXS profiles could satisfactorily be fitted only by using the
so-called spherical copolymer micelle model formerly developed
by Pedersen and Gerstenberg.36 Such a model describes the
scattering of micelles consisting of a homogeneous spherical core
with a radius (Rc) surrounded by corona chains that follow
Gaussian statistics with dimension (RG). The model has a large
number of fitting parameters, namely: RG, Rc, Nagg (aggregation
number of the micelles), d (explained below) and the excess
scattering length density of the core and corona forming blocks
(bcore and bchain). Therefore, it is usually not possible to get
a single set of fitting parameters if some of the parameters are not
preset. Hence, during the fitting procedures the parameters bchain
(7.30 ꢃ 10ꢂ12 cm) and d were held fixed. The parameter d was
preset at d ¼ 1, which mimics the nonpenetration of the corona
chains into the core. Therefore, the starting point of the Gaussian
chains is displaced to a value R0 ꢁ Rc + RG away from the center
of the particle.
Fig. 6 TEM observation of the self-assemblies: low-polydispersity
spherical nanoparticles formed in water. Structures were visualized after
negative staining (a) C18PEG900GlcNAc 4 and (b) C18PEG900Lac 5. Scale
bar corresponds to 50 nm.
signal-to-noise statistic and the presence of association colloids is
suggested by the pronounced X-ray scattering intensity in the
low-q region (q / 0). The high-q range (q > 1.5 nmꢂ1) profile for
the self-assembled nanoparticles is dominated by the so-called
‘‘blob’’ scattering. This is due to the scattering contribution
coming from the PEG chains dissolved in the micellar corona
that dominates the X-ray scattering profile in the high-q region,
which is ideally dictated by the Debye function and therefore by
a nearly qꢂ2 dependence depending on the conformation of the
The fitting parameters were therefore, Nagg, Rc, RG and bcore
.
In Fig. 7, the solid red lines correspond to the best fits achieved
by using the spherical copolymer micelle model. The high quality
of the fits can be straightforwardly noticed and is further attested
by c2 values that remained below 2.5, thus corroborating that
such a model can properly reproduce the whole experimental
curves. Besides, we have observed that in some cases the very
low-q region was not fitted to the same degree of accuracy, which
is due to the presence of a small number of very large aggregates
with loose structure as already evidenced by the DLS experi-
ments.
Since Rc and Nagg were determined, the core surface area
available per PEG chain in the micellar corona (Ac) can be
quantitatively determined as:
4pR2c
Ac ¼
(5)
Nagg
All the parameters extracted from the SAXS curve fittings are
given in Table 1.
The conformation of the PEG chains should itself depend on
such physical chemical variables. One can notice that a reduction
in the micellar core radius is balanced by the increase in RG. A
larger RG was determined for C18PEG900Lac 5 whose micelles
have a smaller Rc and Ac and thus forces the corona PEG chains
to be slightly more stretched. The whole micellar dimension
(Rmic) calculated as Rc + 2RG was found to be equal to 5.9 nm
(C18PEG900GLcNAc 4) and 6.3 nm (C18PEG900Lac 5) which
reasonably matches with the RH values extracted from DLS.
Specific interactions of the nanoparticles with WGA and PNA
lectins
Specific interactions with lectins were further evidenced by DLS
confirming the bioavailability of the sugar residues on the surface
of the micellar nanoparticles. Fig. 8 shows the autocorrelation
functions C(q,t) and the distributions of the relaxation times A(t)
at a scattering angle of 90ꢀ as revealed by CONTIN analysis of
0.5 mg mLꢂ1 solution of C18PEG900Lac 5 before and after
addition of non-binding or binding proteins.
Fig. 7 Small-angle X-ray scattering profiles for 40 mg mLꢂ1 solutions of
C
18PEG900GlcNAc 4 (top) and C18PEG900Lac 5 (bottom) in water at
25 ꢀC.
This journal is ª The Royal Society of Chemistry 2011
Soft Matter, 2011, 7, 3453–3461 | 3459

View Article Online
Table 1 Parameters extracted from the fit of the SAXS curves in Fig. 7 using the spherical copolymer micelle model. Ac was calculated from eqn (5)
Amphiphile
Rc/nm
RG/nm
La/nm
Rmicb/nm
Nagg
bcore/10ꢂ12cm
Ac/nm2
C18PEG900GlcNAc
C18PEG900Lac
2.7
2.3
1.6
2.0
3.2
4.0
5.9
6.3
41
34
ꢂ1.8
ꢂ1.1
2.2
1.9
a
b
L ¼ 2RG. Rmic ¼ Rc ¼ L.
The addition of WGA, which has no affinity for lactose, does
not perturb the system at all. The autocorrelation curve as well as
the distribution of relaxation times remained as before the
addition of WGA. However, as shown in Fig. 8b, the interaction
of lactosyl-covered spherical micelles with PNA lectin signifi-
cantly increases the micellar hydrodynamic radius RH from 10.8
nm to 14.8 nm. It clearly evidences the specific interaction of the
protein with lactose units that are on the surface of the nano-
particles.
Additionally, in Fig. 9 are shown the autocorrelation functions
and the distributions of relaxation times for 0.5 mg mLꢂ1
C18PEG900GlcNAc 4 at 10 mM phosphate buffer saline (pH 7.2)
and 150 mM NaCl, 0.1 mM CaCl2 and 0.1 mM MnCl2 at 37 ꢀC
Fig. 9 Typical autocorrelation functions C(q,t) and distributions of the
relaxation times A(t) at scattering angle of 90ꢀ as revealed by CONTIN
analysis functions of solutions of C18PEG900GlcNAc 4 (0.5 mg mLꢂ1) in
10 mM phosphate buffer (pH 7.2) and 150 mM NaCl, 0.1 mM CaCl2 and
0.1mM MnCl2 at 37 ꢀC, in absence (B) and in the presence (C) of WGA
(2 mL, 1.0 mg Lꢂ1).
in the absence (open circles) and in the presence of WGA lectin (2
mL, 1.0 g.Lꢂ1) (filled circles).
The DLS measurements confirmed a pronounced change in
the micellar hydrodynamic radius and a significant increase in
the size of the micellar aggregates for 4 and 5. These experiments
clearly demonstrate that PNA and WGA lectins specifically
interact with N-acetyl-glucosamine and lactose particles,
respectively, thus supporting the assumption that the carbohy-
drates are exposed at the surface of the micelles.
Conclusions
The synthesis of carbohydrate-functionalized PEG900–stearate
amphiphiles and their self-assembly in water are reported for the
first time. N-Acetyl-glycosamine and lactose have been efficiently
introduced by azide–alkyne click chemistry at the polar extremity
of PEG–stearate. Both glycoconjugates spontaneously self-
assemble in water into spherical micelles having a mean diameter
of 10 nm and a narrow polydispersity as evidenced by DLS,
SAXS and TEM experiments. The ability of the particles to
interact with specific carbohydrate binding proteins has been
further probed by dynamic light scattering thus confirming the
possible application of such glycoconjugates as site-directed drug
delivery systems. The negligible influence of the carbohydrate
groups on the size and morphology of the micelles suggests that
the present system could be easily applied to other saccharidic
ligands targeting different receptors. Drug encapsulation and
release are now being considered to definitely validate the
potential of these nanoparticles as possible vectorization system.
Fig. 8 Typical autocorrelation functions C(q,t) and distributions of the
relaxation times A(t) at scattering angle of 90ꢀ as revealed by CONTIN
analysis functions of solutions of C18PEG900Lac 5 (0.5 mg mLꢂ1) in 10
mM PBS buffer, pH 7.2 containing 150 mM NaCl, 0.1 mM CaCl2 and 0.1
mM MnCl2 at 37 ꢀC, in the absence (B) and in the presence of protein
(C), (a) WGA (6 mL, 1.0 mg mLꢂ1), (b) PNA (4 mL, 1.0 mg mLꢂ1).
3460 | Soft Matter, 2011, 7, 3453–3461
This journal is ª The Royal Society of Chemistry 2011

View Article Online
19 B. S. Kim, D. J. Hong, J. Bae and M. Lee, J. Am. Chem. Soc., 2005,
127, 16333–16337.
Acknowledgements
20 P. Bailon and C.-Y. Won, Expert Opin. Drug Delivery, 2009, 6, 1–16.
21 G. S. Kwon and K. Kataoka, Adv. Drug Delivery Rev., 1995, 16,
295–309.
22 K. Kataoka, A. Harada and Y. Nagasaki, Adv. Drug Delivery Rev.,
2001, 47, 113–131.
23 A. Rosler, G. W. M. Vandermeulen and H. A. Klok, Adv. Drug
Delivery Rev., 2001, 53, 95–108.
24 V. Ratsimbazafy, E. Bourret, R. Duclos and C. Brossard, Eur.
J. Pharm. Biopharm., 1999, 48, 247–252.
We acknowledge the financial support from the CNRS and
CAPES-COFECUB (project 620/08). The ESRF is acknowl-
edged for supply of beam time (Proposal 02 01 784). The tech-
nical assistance of C. Rochas during the experiments at ESRF is
greatly acknowledged. F.C.G. acknowledges FAPESP (grant
2010/06348-0). Stephanie Boullanger is acknowledged for the
mass spectrometry analyses.
25 J. T. Green, B. K. Evans, J. Rhodes, G. A. O. Thomas, C. Ranshaw,
C. Feyerabend and M. A. H. Russell, Br. J. Clin. Pharmacol., 1999,
48, 485–493.
References
26 H. S. Tran, D. Malli, F. A. Chrzanowski, P. M. Puc, M. S. Matthews
and C. W. Hewitt, J. Surg. Res., 1999, 83, 136–140.
27 K. Buszello, S. Harnisch, R. H. Muller and B. W. Muller, Eur.
J. Pharm. Biopharm., 2000, 49, 143–149.
28 A. Fini, J. R. Moyano, J. M. Gines, J. I. Perez-Martinez and
A. M. Rabasco, Eur. J. Pharm. Biopharm., 2005, 60, 99–111.
29 D. L. Ma, T. Y. T. Shum, F. Y. Zhang, C. M. Che and M. S. Yang,
Chem. Commun., 2005, 4675–4677.
30 H. B. Mereyala and S. R. Gurrala, Carbohydr. Res., 1998, 307, 351–
354.
31 K. Kalyanasundaram and J. K. Thomas, J. Am. Chem. Soc., 1977, 99,
2039–2044.
32 B. Berne and R. Pecora, Dynamic Light Scattering: With Applications
to Chemistry, Biology, and Physics, Dover Publications, Mineola
N.Y., 2000.
33 S. W. Provencher, Makromol. Chem., 1979, 180, 201–209.
34 M. Tammer, L. Horsburgh, A. P. Monkman, W. Brown and
H. D. Burrows, Adv. Funct. Mater., 2002, 12, 447–454.
35 C. Giacomelli, V. Schmidt and R. Borsali, Macromolecules, 2007, 40,
2148–2157.
36 J. S. Pedersen and M. C. Gerstenberg, Macromolecules, 1996, 29,
1363–1365.
37 Espace.
38 A. Bouzide, N. LeBerre and G. Sauve, Tetrahedron Lett., 2001, 42,
8781–8783.
1 Y.-b. Lim, K.-S. Moon and M. Lee, Chem. Soc. Rev., 2009, 38, 925–
934.
2 G. Riess, Prog. Polym. Sci., 2003, 28, 1107–1170.
3 S. Forster and T. Plantenberg, Angew. Chem., Int. Ed., 2002, 41, 689–
714.
4 D. E. Discher and A. Eisenberg, Science, 2002, 297, 967–973.
5 F. S. Bates, Science, 1991, 251, 898–905.
6 C. Giacomelli, V. Schmidt, K. Aissou and R. Borsali, Langmuir, 2010,
26, 15734–15744.
7 L. Lindoy, Self-Assembly in Supramolecular Systems, Royal Society
of Chemistry, Cambridge, 2000.
8 T. M. Allen, Science, 2004, 303, 1818–1822.
9 A. J. Almeida and E. Souto, Adv. Drug Delivery Rev., 2007, 59, 478.
10 Y. B. Lim, K. S. Moon and M. Lee, J. Mater. Chem., 2008, 18, 2909–
2918.
11 W. J. Li and F. C. Szoka, Pharm. Res., 2007, 24, 438–449.
12 X. L. Wu, J. H. Kim, H. Koo, S. M. Bae, H. Shin, M. S. Kim,
B.-H. Lee, R.-W. Park, I.-S. Kim, K. Choi, I. C. Kwon, K. Kim
and D. S. Lee, Bioconjugate Chem., 2010, 21, 208–213.
13 Z. Pang, W. Lu, H. Gao, K. Hu, J. Chen, C. Zhang, X. Gao, X. Jiang
and C. Zhu, J. Controlled Release, 2008, 128, 120–127.
14 A. Suo, J. Qian, Y. Yao and W. Zhang, Int. J. Nanomed., 2010, 5,
1029–1038.
15 A. Varki, Glycobiology, 1993, 3, 97–130.
16 C. Lemarchand, R. Gref, S. Lesieur, H. Hommel, B. Vacher,
A. Besheer, K. Maeder and P. Couvreur, J. Controlled Release,
2005, 108, 97–111.
17 H. Yamamoto, Y. Kuno, S. Sugimoto, H. Takeuchi and
Y. Kawashima, J. Controlled Release, 2005, 102, 373–381.
18 N. Sharon, Glycobiology, 2004, 14, 53R–62R.
39 H. C. Kolb, M. G. Finn and K. B. Sharpless, Angew. Chem., Int. Ed.,
2001, 40, 2004–2021.
40 C. W. Tornøe, C. Christensen and M. Meldal, J. Org. Chem., 2002,
67, 3057–3064.
41 L. F. Zhang and A. Eisenberg, Macromolecules, 1996, 29, 8805–8815.
This journal is ª The Royal Society of Chemistry 2011
Soft Matter, 2011, 7, 3453–3461 | 3461