Gold maftfto-glyconanoparticles: Multivalent systems to block HIV-1 gp120 binding to the lectin DC-SIGN

Article abstract of DOI:10.1002/chem.200900923

The HIV envelope glycoprotein gp120 takes advantage of the highmannose clusters on its surface to target the C-type lectin dendritic cellspecific intracellular adhesion molecule-3-grabbing non-integrin (DCSIGN) on dendritic cells. Mimicking the cluster pr

Full text of DOI:10.1002/chem.200900923

DOI: 10.1002/chem.200900923
Gold Manno-Glyconanoparticles: Multivalent Systems to Block HIV-1 gp120
Binding to the Lectin DC-SIGN+
Olga Martꢀnez-ꢁvila,[a] Karolin Hijazi,[b] Marco Marradi,[a] Caroline Clavel,[a, c]
Colin Campion,[d] Charles Kelly,[b] and Soledad Penadꢂs*[a]
Abstract: The HIV envelope glycopro-
tein gp120 takes advantage of the high-
mannose clusters on its surface to
target the C-type lectin dendritic cell-
specific intracellular adhesion mole-
cule-3-grabbing non-integrin (DC-
SIGN) on dendritic cells. Mimicking
the cluster presentation of oligomanno-
sides on the virus surface is a strategy
for designing carbohydrate-based anti-
viral agents. Bio-inspired by the cluster
presentation of gp120, we have de-
signed and prepared a small library of
multivalent water-soluble gold glycona-
noparticles (manno-GNPs) presenting
truncated (oligo)mannosides of the
gp120. These glyconanoparticles are li-
gands for DC-SIGN, which also inter-
acts in the early steps of infection with
a large number of pathogens through
specific recognition of associated gly-
cans. (Oligo)mannosides endowed with
different spacers ending in thiol groups,
which enable attachment of the glyco-
conjugates to the gold surface, have
been prepared. manno-GNPs with dif-
ferent spacers and variable density of
mannose (oligo)saccharides have been
obtained and characterized. Surface
plasmon resonance (SPR) experiments
with selected manno-GNPs have been
performed to study their inhibition po-
tency towards DC-SIGN binding to
gp120. The tested manno-GNPs com-
pletely inhibit the binding from the
micro- to the nanomolar range, while
the corresponding monovalent manno-
sides require millimolar concentrations.
manno-GNPs containing the disacchar-
ide Mana1-2Mana are the best inhibi-
tors, showing more than 20000-fold in-
creased activity (100% inhibition at
115 nm) compared to the corresponding
monomeric disaccharide (100% inhibi-
tion at 2.2 mm). Furthermore, increas-
ing the density of dimannoside on the
gold platform from 50 to 100% does
not improve the level of inhibition.
Keywords: glycoconjugates · glyco-
nanoparticles · high-mannose oligo-
saccharides · multivalency · surface-
plasmon resonance
high-mannose
undecasaccharide
Man9GlcNAc2 and have tested them as
inhibitors of DC-SIGN binding to
Introduction
carbohydrate-mediated interactions is compensated by clus-
tering of the ligands. Multivalency is essential in carbohy-
drate–protein interactions.[2] Different chemical approaches
have been developed to study carbohydrate interactions, all
Multivalency is ubiquitous in biological interactions, includ-
ing carbohydrate-mediated processes.[1] The low affinity of
[a] Dr. O. Martꢀnez-ꢁvila, Dr. M. Marradi, Dr. C. Clavel,
Prof. Dr. S. Penadꢂs
[d] Dr. C. Campion
Carbohydrate Synthesis Ltd., Abingdon (UK)
[+] DC-SIGN=Dendritic cell-specific intracellular adhesion molecule-3-
Laboratory of GlycoNanotechnology
Biofunctional Nanomaterial Unit
grabbing non-integrin.
CIC biomaGUNE and CIBER-BBN
Parque Tecnolꢃgico, P8 de Miramꢃn 182, 20009 San Sebastiꢄn (Spain)
Fax : (+34)943 00 53 01
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200900923. Synthesis and charac-
terization of neoglycoconjugates GlcC5S and 14–16, linkers 26, 39, and
40, aminoethyl mannosides 27–32, and main intermediates; prepara-
tion and characterization of GlcC5-Au and HO2C-Au GNPs; TEM his-
tograms and size distribution of GNPs 4–7, 8, 9, 9a,b, 10a,b, 11a,b,
E-mail: spenades@cicbiomagune.es
[b] Dr. K. Hijazi, Prof. C. Kelly
Kingꢅs College London, Dental Institute, Oral Immunology
Tower Wing, Guyꢅs Hospital, London, SE1 9RT (UK)
1
12a,b, and 13a, glcC5-Au, and HO2C-Au; H and 13C NMR spectra of
[c] Dr. C. Clavel
neoglycoconjugates 16, 33–38, GlcC5S, and linkers 26, 39, and 40. Sen-
Current address: IBMM, UMR 5247
CNRS-Universitꢂs Montpellier 2 et 1
ENSCM, rue de l’Ecole Normale 8
34296 Montpellier Cedex 5 (France)
sorgrams (Figures S1–S4) of SPR experiments.
9874
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9874 – 9888

FULL PAPER
of which are based on multivalent presentation of carbohy-
drate ligands.[3]
the gold cluster and dissect the structural requirements of
the sugars involved in HIV/DC-SIGN interaction. We pres-
ent the preparation and characterization of a small library
of water-soluble gold glyconanoparticles (manno-GNPs)
functionalized with partial structures of the high-mannose
We have developed new polyvalent systems (glyconano-
particles) consisting of a metallic core, to which self-assem-
bled monolayers of glycoconjugates are covalently
linked.[4–6] Glyconanoparticles (GNPs) are water-soluble bio-
functional gold nanoclusters with a three-dimensional (3D)
polyvalent carbohydrate display and globular shape, chemi-
cally well-defined composition, and an exceptionally small
core size. They can display a large number of carbohydrates
on a reduced surface with a high local concentration of
sugars (100 molecules on a 2 nm core gold) or lower densi-
ties. GNPs are useful tools for investigating carbohydrate-
mediated interactions.[7,8] GNPs functionalized with lactose
neoglycoconjugates have been shown to be efficient anti-ad-
hesion agents, inhibiting the ex vivo metastasis of melanoma
in mice by up to 70%.[8]
The surface unit of the HIV envelope glycoprotein gp120
is heavily glycosylated with N-linked mannose glycans,[9]
which presumably shield the neutralizing epitopes.[10] De-
spite the large variation in the amino acid sequence of
gp120 due to the immune selective pressure, the overall
degree of glycosylation is preserved.[11] The C-type lectin
DC-SIGN (dendritic cell-specific intracellular adhesion mol-
ecule-3 (ICAM-3)-grabbing non-integrin) expressed on the
surface of dermal dendritic cells (DCs) has been implicated
in HIV vaginal transmission.[12–14] DC-SIGN binds specifical-
ly to the oligomannosides on gp120 through protein–carbo-
hydrate interactions in a multivalent and Ca2+-dependent
manner.[15–17] Carbohydrate structures on gp120 are targets
for candidate antiviral agents and vaccines.[18–21] Mimicking
the cluster presentation of the oligomannosides on the virus
surface is a strategy for designing carbohydrate-based antivi-
ral agents. Multivalent systems of mannose oligosaccharides
based on proteins,[22] peptides,[23,24] dendrimers,[25,26] and
other scaffolds have been prepared and tested as inhibitors
of gp120 binding to DC-SIGN or to monoclonal antibody
2G12. GNPs that interfere with gp120 interaction to DC-
SIGN may prevent infection by inhibiting DC-mediated
transmission of HIV to CD4 T cells.
undecasaccharide Man9ACTHNUTRGNE(NUG GlcNAc)2 of gp120 or with a non-
natural heptasaccharide (Figure 1). The (oligo)mannosides
were endowed with different spacers ending in a mercapto
(thiol) group, which enables attachment of the glycoconju-
gates to the gold surface. Manno-GNPs with different
spacers and variable density of mannose (oligo)saccharides
have been prepared and explored (Figure 1). Selected glyco-
nanoparticles bearing the (oligo)mannosides have been
tested as inhibitors of DC-SIGN binding to gp120 by surface
plasmon resonance (SPR). We show that multivalent presen-
tation of a simple Mana1-2Mana disaccharide on the gold
nanoparticle increases the inhibitory activity by more than
four orders of magnitude compared to the monovalent dis-
accharide. Furthermore, increasing the density of dimanno-
side on the gold platform does not improve the inhibitory
potency.
Results and Discussion
Multivalent systems functionalized with oligomannosides
have recently been designed as potential vaccines against
HIV or to provide insights into understanding HIV glycobi-
ology.[19–23,28] Multivalent Man9 clusters[29] and template-as-
sembled oligomannose clusters[23,30] have been prepared and
some of these systems proved more effective in inhibiting
2G12 binding to gp120 than the corresponding subu-
nits.[22,29b,30] It was also demonstrated that glycoclusters con-
taining dimannosides were taken up avidly by dendritic
cells.[24,32]
Preparation of manno-GNPs 1–13 (Figure 1) requires con-
jugation of the (oligo)mannosides to a spacer ending in a
mercapto (thiol) group. The selected (oligo)saccharides and
the corresponding conjugates are displayed in Figure 2.
They are structural motifs of the undecasaccharide Man9-
The glyconanoparticle platform offers advantageous alter-
natives to protein, polymer, or dendrimer scaffolds. Glyco-
nanoparticle technology (glyconanotechnology) allows the
preparation of a great variety of water-soluble glycoclusters
with different ligand densities (high and low loadings) and
variable linkers to modulate rigidity and flexibility and to
confer accessibility to the ligands.[5] The nature (hydrophilic
or hydrophobic), the length, and the flexibility of the spacer
can be selected to control the presentation of the carbohy-
drates on the cluster surface, which influences their accessi-
bility to the ligands and behaviour during molecular recog-
nition events. Furthermore, glyconanotechnology offers the
unique possibility of simultaneous incorporation, in a single
gold cluster, of different ligands in a controlled way.[27]
In this study, GNPs coated with sets of different structural
motifs of the N-linked high-mannose-type glycans of gp120
have been designed to assess the effect of presentation on
ACHTUNGTRNE(NNUG GlcNAc)2, except for the heptasaccharide 32, which results
from adding two mannose residues to the pentasaccharide
31. A major effort has been undertaken to develop efficient
strategies for synthesizing high-mannose oligosaccharidic
structures on gp120.[32] Different protocols have been chosen
for the preparation of the (oligo)mannose conjugates. To
prepare the glycoconjugates, diverse spacers in terms of
their hydrophobic and/or hydrophilic nature have been
used: aliphatic chains (C2 or C5) to impart rigidity to the
GNP, or an amphiphilic mixed aliphatic/polyethylene glycol
linker to impart flexibility and solubility to the nanoparticle.
The spacers have been introduced in the sugar either by
direct glycosylation or by conjugation of alkylamino-func-
tionalized (oligo)saccharides to a linker endowed with
either a carboxylic or an isothiocyanate group. The prepara-
tion of neoglycoconjugates of biologically relevant sugars re-
quires the development of efficient conjugation protocols
Chem. Eur. J. 2009, 15, 9874 – 9888
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9875

S. Penadꢂs et al.
Figure 1. Structure of the high-mannose glycans and schematic representation of gold manno-glyconanoparticles (A, B, and C) and the control nanoparti-
cles (D).
for attaching the carbohydrates to the different linkers. Easy
and standard procedures can be applied for mono- and dis-
accharides, but as the complexity of the sugar increases, the
necessity of setting up a general methodology to obtain vari-
ous neoglycoconjugates with suitable thiol-terminated link-
ers on a laboratory scale becomes a real challenge. Different
conjugation methods have been used to obtain the sugar
conjugates. The monosaccharide conjugates 14–16 (Fig-
ure 2A) were prepared by direct glycosylation of the con-
veniently protected sugar with the appropriate linkers. Be-
cause of the difficulty of applying this strategy to oligosac-
charides, we tried a more versatile approach.
We prepared a series of alkylamino (oligo)mannosides
(18–21 and 27–32) for further coupling with suitably func-
tionalized amphiphilic linkers 26 and 39 (Figure 2B and C).
Peptidic coupling between the amino derivatives and car-
boxylic acid-bearing linker 26 was only partially satisfactory,
due to the formation of mixed disulfides and the require-
ment for an excess of the (oligo)saccharide. Coupling of the
amino-functionalized sugars with isothiocyanate-terminated
linker 39 provided a solution. The latter coupling method al-
lowed great versatility in linker selection as well as the prep-
aration of an entire set of mono-, di-, tri-, tetra-, penta-, and
heptamannoside GNPs (Figure 2C) in high yields.
The resulting glycoconjugates were incorporated at vary-
ing densities onto the surface of the gold nanoclusters.
Three families of manno-GNPs, 1–3, 4–7, and 8–13, were
generated with mannose glycoconjugates obtained by direct
glycosylation (Figure 1A), peptidic coupling (Figure 1B),
and isothiocyanate coupling (Figure 1C), respectively. The
density of (oligo)saccharides was controlled by adjusting the
initial concentration of glycoconjugate with mixed linker 40
(HO2CCH2OEG5C11SH) or glucose conjugates 2,2’-dithio-
bisACTHNUTRGNENG(U ethyl)- or 5,5’-dithio-bisCAHTUNGTRENN(UGN pentyl)-b-d-glucopyranoside
(GlcC2S or GlcC5S) as stealth components. Glucose was
chosen as a biocompatible inert component. Large terminal
ligands, such as oligosaccharides, attached to a short alkane-
thiol may induce disorder in the system.[33] For (oligo)man-
9876
www.chemeurj.org
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9874 – 9888

Gold Manno-Glyconanoparticles
FULL PAPER
Figure 2.
sulfides).
ACHTUNGTRENNUNG(Oligo)mannose neoglycoconjugates and the corresponding linkers (for the sake of clarity, all of the neoglycoconjugates are represented as di-
nose conjugates, longer linkers 26 or 39 were (depending on
the conjugation method) chosen to make the sugars more
accessible to receptors. The aliphatic part of the linker
allows good SAMs packaging and confers rigidity to the
inner organic shell to protect the gold core, while the exter-
nal polyether moiety, due to its flexibility upon solvation in
Chem. Eur. J. 2009, 15, 9874 – 9888
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9877

S. Penadꢂs et al.
water, ensures accessibility to ligands and assists water solu-
bility. Furthermore, SAMs that present poly(ethylene
glycol) units are known to resist the adsorption of pro-
teins.[34,35]
Preparation of neoglycoconjugates by peptidic coupling:
The neoglycoconjugates 22–25 (Figure 2B) were prepared
from the aminopropyl mannosides 18–21 (provided as ace-
tate salts) by peptidic coupling with carboxylated linker 26,
which was in turn synthesized by Jones oxidation[44] of linker
17a. Methanolysis of 26 afforded deprotected linker 40
(Scheme 1A). Conjugation of the unprotected aminopropyl
glycosides 18–21 with the carboxylic linker 26 in DMF using
diisopropylcarbodiimide (DIC) and 1-hydroxybenzotriazole
(HOBt)[45] resulted in a mixture of products (Scheme 1B).
Purification on Sephadex LH-20 was used to remove
excess aminosugars, derivatives of the coupling reagents,
and their sub-products. The integrals of the signal of the thi-
oacetyl group (singlet at around d=2.3 ppm) in the
1H NMR spectra were, in all cases, lower than expected, in-
dicating that the reaction conditions caused partial deprotec-
tion of the thioacetyl group with consequent formation of
mixed disulfide species. Furthermore, a sharp singlet at
around d=4.3 ppm due to the methylene protons a to the
carboxylic acid groups was detected, suggesting the presence
of free carboxylic acid linker derivatives. Attempts to purify
mixtures 22’–25’ by chromatography on silica gel were also
unsuccessful: products 22–25 could not be isolated and a
great loss of material occurred. Each mixture was thus di-
rectly treated with an excess of sodium methoxide to com-
plete the deprotection of the thioacetyl group. After desalt-
ing on a column of Sephadex LH-20, a 2:1 mannose/carbox-
ylic acid ratio for mixtures 22’ and 24’ or a 1:1 ratio for mix-
tures 23’ and 25’ (Scheme 1B) was determined by integra-
tion of the NMR signals of the anomeric protons of the
mannoses and the methylene protons a to the carboxylic
group. These mixtures were used to synthesize hybrid
GNPs. The density of the ligands was conserved in the
GNPs. Because of purification problems, the slow coupling
reaction (24–48 h), and the requirement for an excess of
amino saccharide, we investigated alternative synthetic
methods. A parallel strategy based on the coupling of amino
glycosides with an isothiocyanate-functionalized linker to
form a thiourea linkage was used.
Preparation of neoglycoconjugates by direct glycosylation:
Mannose conjugates 14–16 (Figure 2A), 2,2’-dithio-bis-
A
(GlcC2S),
5,5’-dithio-bis-
ACHTUNGTRENNUNG
jugate LactoEG6C11S were synthesized by direct glycosyla-
tion of the conveniently protected saccharides. The mannose
conjugate 14, functionalized with a two-carbon-atom chain,
was prepared as previously reported with minor modifica-
tions.[36] The mannose glycoconjugate 15[37] was prepared by
a synthetic approach based on Fisher glycosylation.[38] For
the preparation of the neoglycoconjugate 16, the peracety-
lated bromo-mannoside[39] was glycosylated with the spacer
17a[40] in the presence of Hg(CN)2 as a promoter[41] to give
the thioacetyl neoglycoconjugate, which was deacetylated by
methanolysis[42] to yield compound 16 (see the Supporting
Information). The neoglucoconjugate GlcC2S was synthe-
sized as previously reported.[27] Glucose functionalized with
five carbon atoms GlcC5S was efficiently produced by meth-
anolysis of the corresponding thioacetate derivative, which
was in turn prepared as reported in the literature.[43] The
conjugate LactoEG6C11S has been previously synthesized in
our laboratory.[5]
Due to the modest yields obtained with the direct glycosy-
lation method, especially when mixed linkers were used, we
investigated an alternative strategy for the synthesis of gly-
coconjugates 22–25 and 33–38. We chose an approach in
which fully deprotected 1-aminoalkyl glycosides were cou-
pled to the linkers in the final steps. This is a convenient
strategy, particularly when oligosaccharides are involved.
The carboxylic acid- or isothiocyanate-functionalized linkers
26 and 39 were introduced by peptidic coupling or thiourea
linkage formation. In principle, this also allows a greater
versatility in terms of the spacer arms that can be incorpo-
rated.
Scheme 1. Synthesis of A) linker 26 and B) neoglycoconjugates by peptidic coupling. a) CrO3, H2SO4, acetone; b) MeONa, MeOH; c) HOBt, DIC, NEt3,
DMF.
9878
www.chemeurj.org
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9874 – 9888

Gold Manno-Glyconanoparticles
FULL PAPER
Preparation of neoglycoconjugates by formation of a thiour-
ea linkage: The neoglycoconjugates 33–38 (Figure 2C) were
prepared by coupling of the aminoethyl (oligo)mannosides
27–32 with the isothiocyanate linker 39 (Scheme 2). The for-
mation of a thiourea linkage between an amino group and
an isothiocyanate group is a well-established reaction in bio-
conjugation that results in high yields of neoglycoconjugates.
derivatives. Purification on silica gel resulted in low yields of
the disaccharide derivatives. Sephadex LH-20 was thus used
for purification of the thioacetyl-protected neoglycoconju-
gates. The monosaccharide 33, the disaccharide 34, and the
tetrasaccharide 36 glycoconjugates were obtained after
methanolysis in yields of 81, 73, and 81% (referred to two
steps), respectively. Trimannoside 29, pentamannoside 31,
and heptamannoside 32 were
coupled to the isothiocyanate
linker 39 using triethylamine in
water/isopropanol/acetonitrile
(1:1:1) as no reaction occurred
in methanol. Evaporation of
the solvents, trituration of the
crude product with diethyl
ether until complete removal of
the unreacted linker, and de-
salting of the mixture and elimi-
nation of unreacted aminosugar
by chromatography on Sepha-
dex LH-20 furnished the pure
thioacetyl derivatives. Depro-
tection of the thioacetyl group
afforded the corresponding ne-
oglycoconjugates 35, 37, and 38
Scheme 2. Synthesis of A) isothiocyanate linker 39 and B) (oligo)mannose neoglycoconjugates by thiourea for-
mation. a) NaN3, PPh3, BrCCl3, DMF; then CS2, PPh3 (72%, overall yield); b) NEt3, MeOH or c) NEt3,
iPrOH/CH3CN/H2O 1:1:1; d) MeONa, MeOH.
in yields of 72, 73, and 52%
(referred to two steps), respec-
tively. The described reaction
conditions were successfully ap-
The aminoethyl monosaccharide 27 was synthesized in
three steps from the commercial peracetylated mannose:
standard glycosylation with 2-N-Z-ethanolamine in the pres-
ence of BF3·Et2O, deacetylation, followed by hydrogenation
of the benzyloxycarbonyl group[46] (see the Supporting Infor-
mation). The aminoethyl mannosides 28 and 29 were ob-
tained by glycosidation of thiotolyl di- and trimannoside
donors, prepared by Wongꢅs one-pot self-condensation strat-
egy,[32i] with 2-N-Z-ethanolamine as the acceptor. The 1-ami-
noalkyl oligomannosides 30–32 were also synthesized by
Wongꢅs protocol, by glycosylation of thiotolyl di- and tri-
mannoside donors with the ethylamino mannoside building
blocks prepared by a modified Ogawa protocol (see the
Supporting Information).[32a,47]
The isothiocyanate linker 39 was obtained in 72% yield
from the corresponding alcohol 17b[40] by an original and
straightforward conversion of the alcohol group to azide
and subsequent conversion of the azido group into the iso-
thiocyanate functionality by treating the crude intermediate
with carbon disulfide (Scheme 2A). The conversion of the
primary alcohol into the azide group was achieved by a
modification of the described procedure (sodium azide and
triphenylphosphine in CCl4/DMF),[48] replacing tetrachloro-
methane with bromotrichloromethane.[49]
plied to each of the aminoethyl mannosides 27–32 to furnish
products 33–38 in good yields.
Preparation of glyconanoparticles: Three types of manno-
GNPs (1–3, 4–7, and 8–13; Figure 1) were prepared and
characterized by the procedure developed in our laborato-
ry.[5,27] An aqueous solution of tetrachloroauric acid was
added to a methanolic solution of the neoglycoconjugate or
to a mixture of variable proportions of glycoconjugates and
stealth component. The resulting mixture was reduced with
an excess of NaBH4 and the suspension was vigorously
shaken for 2 h at 258C. The supernatant was removed and
the residue was dissolved in milliQ water, purified by dialy-
sis or centrifugal filtration, and characterized by 1H NMR
spectrometry, transmission electron microscopy (TEM), and
ultraviolet spectroscopy (UV) (Figure 3). Glyconanoparti-
cles with 100% density (1–3, 8, and 9) or variable density of
sugars (3a–c, 4–7, 9a,b, 10a,b, 11a,b, 12a,b, and 13a) were
obtained by controlling the ratio of neoglycoconjugate to
stealth component (GlcC2S, GlcC5S, or mixed linker 40) in
the initial aqueous solution. The proportion of the ligands
1
on the gold surface was examined by H NMR before and
after cluster formation (Figure 3C). GNPs functionalized
with 100% glucose (GlcC5-Au), lactose (lactoEG6C11-Au),
or mixed linker (HO2C-Au) (Figure 1D) were also prepared
as control systems.
The resulting GNPs showed an exceptionally small core
(1–2 nm), as demonstrated by TEM analysis. TEM micro-
The isothiocyanate-functionalized linker 39 was coupled
to the monomannoside 27, dimannoside 28, and tetramanno-
side 30 in the presence of triethylamine in methanol
(Scheme 2B) to obtain the neoglycoconjugates as thioacetyl
Chem. Eur. J. 2009, 15, 9874 – 9888
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9879

S. Penadꢂs et al.
Figure 3. Characterization of glyconanoparticles 11a: A) TEM micrograph in H2O; B) size-distribution histogram; C) 1H NMR spectra of mixtures of ne-
oglycoconjugates GlcC5S and tetramannoside 36 before (C-1) and after (C-3) GNP formation, and of the corresponding GNP 11a (C-2); D) UV/Vis
spectrum.
graphs showed uniform dispersion of the GNPs and no ag-
gregation was evident. The UV/Vis spectra were often char-
acterized by a surface plasmon band at around 520 nm,
except in the case of the smallest core-sized GNPs, for
termined by TEM) and elemental analysis, an average mo-
lecular formula was estimated (Table 1). Molecular weights
(MW), calculated according to the literature,[50] were in
agreement with those obtained on the basis of elemental
analyses.
1
which the plasmon was scarcely visible. The H NMR spec-
tra of the GNPs featured broader peaks compared to those
of the corresponding neoglycoconjugates. An example of
GNP characterization is shown in Figure 3.
All of the GNPs were found to be water-soluble (although
solubilization of the thiourea linkage-containing GNPs was
slow), and were stable for months under physiological con-
ditions without flocculation. Based on the gold core size (de-
Manno-glyconanoparticles as inhibitors of DC-SIGN bind-
ing to gp120: Glyconanoparticles (manno-GNPs) with differ-
ent (oligo)mannoside motifs were tested as inhibitors of
gp120 binding to DC-SIGN. (Oligo)mannosides mimicking
carbohydrate structures on gp120 can inhibit DC-SIGN/
gp120 binding by competitive interaction with DC-SIGN.
Table 1. Chemical properties of selected glyconanoparticles.
GNP
Mannoside
Average diameter
of gold
particles [nm]
Average
Average molecular formula
Average
mannoside
copy number
Average
Mw
[a]
number of
gold atoms[a]
4
5
6
7
Mana1-2Man
Mana1-3Man
1.3ꢀ0.6
1.0ꢀ0.4
1.3ꢀ0.5
1.3ꢀ0.5
1.3ꢀ0.4
1.2ꢀ0.5
1.6ꢀ0.4
2.7ꢀ0.5
79
79
79
79
79
79
140
586
(C34H72NO18S)25(C23H45O8S)13Au79
(C34H72NO18S)19(C23H45O8S)19Au79
(C44H83NO23S)25(C23H45O8S)13Au79
(C44H83NO23S)19(C23H45O8S)19Au79
(C34H65N2O15S2)22(C11H21O6S)22Au79
(C34H65N2O15S2)59Au79
25
19
25
19
22
59
–
43398
41376
47451
44190
Mana1-3
N
Mana1-2Mana1-2Man
Mana1-2Man
Mana1-2Man
–
–
9a
9
GlcC5-Au
HO2C-Au
39482[b]
63058[b]
37422[b]
179903
(C11H21O6S)35Au140
(C23H41O8S)135Au586
–
[a] The average number of gold atoms in the cluster, the molecular formulae, and the molecular weights of the nanoparticles were calculated according
to the average gold diameter obtained by TEM.[50] [b] Values confirmed by elemental analysis.
9880
www.chemeurj.org
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9874 – 9888

Gold Manno-Glyconanoparticles
FULL PAPER
Multivalent presentation of (oligo)saccharides on GNPs
may significantly increase their inhibitory potency, thereby
improving their potential as microbicides. SPR-based com-
petitive assays have been used to evaluate selected manno-
GNPs as inhibitors of DC-SIGN binding to gp120. SPR in-
hibition assays were carried out by direct immobilization of
gp120 on the sensor chip surface and binding measurements
of fluid-phase DC-SIGN at a fixed concentration in the
presence of manno-GNPs or free alkyl amino (oligo)manno-
sides at varying stoichiometric ratios.
Free (oligo)mannosides, as well as manno-GNPs, showed
dose-dependent inhibition of DC-SIGN binding to gp120.
The inhibitory activity of free alkyl amino (oligo)manno-
sides was compared to that of methyl a-d-mannopyranoside.
All of the free (oligo)mannosides, except for Mana1-
3Mana1-6Mana and Mana1-4Mana, completely inhibited
the gp120/DC-SIGN interaction at concentrations in the 1–
3 mm range. A higher concentration (25 mm) of a-methyl
mannopyranoside was required for 100% inhibition. On an
equimolar basis (500 mm concentration), disaccharide 18
(90% inhibition) and trisaccharide 20 (85% inhibition)
were the best inhibitors (data not shown). Mana1-2Mana 18
was more effective than Mana1-3Mana 19 (40% inhibition)
or
pentasaccharide
GlcNAcb1-2Mana1-3(GlcNAcb1-
2Mana1-6)Mana(GlcNAc2Man3) (64% inhibition), confirm-
ing a preference of DC-SIGN for a terminal a1!2 link-
age[51] and suggesting that a higher complexity of the oligo-
saccharide does not enhance inhibitory activity. Disacchar-
ide Mana1-2Mana 18 was able to fully inhibit the interac-
tion at a concentration of 2.2 mm. We then asked whether
the potency of the free (oligo)mannosides could be boosted
by presentation in a multivalent display on a gold surface.
GNPs 1, 2, 3, and 3a–c, coated with mannose monosacchar-
ide at variable densities (Figure 1A), inhibited DC-SIGN/
gp120 binding (100%) at 20 mm (data not shown), which is
at least 1000-fold lower than the concentration of methyl a-
d-mannopyranoside required to observe complete inhibi-
tion. Under similar experimental conditions, a glycoden-
drimer with 32 mannose units inhibited 50% of the binding
at sub-millimolar concentrations.[26] Glucose-GNP (GlcC5-
Au) and lactose-GNP (lactoEG6C11-Au) showed no inhibito-
ry activity at concentrations as high as 50 mm, and were used
as negative controls.
Figure 4. Inhibitory activity of Mana1-2Man versus Mana1-2Man-coated
GNP 4. gp120ACTHNUTRGENUG(N CN54) was immobilized by direct amine coupling (~1000
GNPs 4–6, containing Mana1-2Mana, Mana1-3Mana, or
Mana1-2Mana1-2Mana and carboxyl linker 40 at different
densities, fully inhibited binding at concentrations between
0.115 and 4.3 mm on the GNP. In particular, multivalent pre-
sentation of Mana1-2Mana on GNP 4, characterized by 25
units of disaccharide Mana1-2Mana and an average molecu-
lar weight of ~43 kDa, resulted in 20000-fold increased ac-
tivity (100% inhibition at 115 nm) compared to the corre-
sponding monomeric disaccharide 18 (100% inhibition at
2.2 mm) (Figure 4). The level of inhibition (approximately
50%) upon co-injection with 6 nm GNP 4 suggests binding
of more than one molecule of DC-SIGN per particle.
resonance units (RU)). The binding activity of fluid-phase DC-SIGN in
the presence of free Mana1-2Mana (A) was determined and compared
to that of DC-SIGN in the presence of Mana1-2Mana-coated GNP 4
(B). A) Superimposed sensorgrams representing DC-SIGN (100 nm)
binding activity in the absence (dotted curve) and presence (solid curves)
of Mana1-2Mana at (55–2200 mm). B) Dose-dependent gp120/DC-SIGN
binding inhibition by GNP 4. DC-SIGN (50 nm) binding in the presence
of 4 (6–115 nm) compared to that in the presence of GlcC5S-Au at 1 mm
(grey curve) and in the absence of inhibitor (dotted curve). Sensorgrams
shown in this panel were obtained by subtraction of the sensorgram ob-
tained following injection of GNPs alone from that generated by injec-
tion of GNPs in the presence of DC-SIGN. The buffer used was HBS-P
supplemented with 10 mm CaCl2. The flow rate was 20 mLminꢁ1 and the
injection volume was 20 mL. C) Plot of the data from B) showing % in-
hibition at each concentration.
GNP 4 was also more potent than Mana1-2Mana1-
2Mana-GNP 6 and Mana1-3Mana-GNP 5, which displayed
Chem. Eur. J. 2009, 15, 9874 – 9888
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9881

S. Penadꢂs et al.
full inhibitory activity at 1.2 mm and 4.3 mm, respectively. The
presence of an additional mannose unit in position 2 did not
improve the inhibitory activity of the nanoparticle. These
findings support the data of the experiments with free alkyl-
amino (oligo)mannosides, which showed that the complexity
of the mannoside structures does not enhance the inhibitory
activity of the compound and that Mana1-2Mana structures
are more effective inhibitors. Table 2 summarizes the inhibi-
tory activity of selected glyconanoparticles (data analysis of
the best-fit curves and estimated IC50 and IC90 values are
reported in the Supporting Information; Figure S5 and
Table S1).
Table 2. Manno-GNPs as inhibitors of DC-SIGN-ECD binding to immo-
bilized gp120.
GNP Mannoside
Average man-
nose
ACHTNUGETRN[NNUG C100]
[a]/mm
copy number
per
GNP
per manno-
side
4
5
6
Mana1-2Mana
25
19
25
0.115
4.3
1.2
2.9
82
30
Mana1-3Mana
Mana1-2Mana1-
2Mana
9
9a
Mana1-2Mana
Mana1-2Mana
59
22
0.08
0.13
4.7
2.9
[a] The nanoparticle concentration required to inhibit DC-SIGN-ECD
binding to gp120 by 100%. Data shown are based on single measure-
ments.
To confirm that inhibition mediated by GNPs occurs by
direct interaction with DC-SIGN, GNP 4 binding activity to
immobilized DC-SIGN-ECD was demonstrated. Sensor-
grams representing GNP 4 binding at various concentrations
(0.25–2 mm) were fitted to the linked-reaction model with a
KD value of 9.74ꢇ10ꢁ7 m (Figure 5). The surface of GNP 4 is
partially (66%) coated with Mana1-2Mana neoglycoconju-
gate, and the remaining sites are occupied by COOH-termi-
nated linkers. To determine whether the COOH groups con-
tribute to GNP interaction with DC-SIGN, GNP HO2C-Au
containing only COOH linker 40 was tested. No significant
binding activity to DC-SIGN was detected (dotted curve,
Figure 5A). Likewise, HO2C-Au did not exert any inhibitory
effect at concentrations required to observe inhibition by
GNP 4 (data not shown).
The binding affinity of DC-SIGN to a GNP containing
100% Mana1-2Mana-glycoconjugate (GNP 9) was approxi-
mately twofold higher than that observed for GNP 4 (Fig-
ure 5B) with faster on rates and slower kd1, although the
overall level of binding was similar. GNP 9 inhibitory activi-
ty on the gp120/DC-SIGN interaction was also comparable
to that exhibited by GNP 4 (data not shown). This suggests
that a higher density of glycoconjugates does not improve
the efficacy of the GNPs.
Figure 5. Effect of density of Mana1-2Man neoglyconjugates on GNP
binding affinity to DC-SIGN. DC-SIGN-ECD (700 RU) was immobilized
directly on the sensor chip surface. Binding of fluid-phase GNPs at 0.25–
2 mm was then determined. A) Superimposed sensorgrams representing
binding activity of GNP 4 (66% in Mana1-2Man) (solid curves) at vary-
ing concentrations to DC-SIGN. The HO2C-Au GNP at 2 mm (dotted
curve) did not exhibit any binding activity. B) Sensorgrams showing GNP
9 (100% in Mana1-2Man) binding activity to immobilized DC-SIGN. Ki-
netic values estimated by fitting the sensorgrams to the linked reaction
model are indicated. The buffer used was HBS-P supplemented with
10 mm CaCl2. The flow rate was 20 mLminꢁ1 and the injection volume
was 20 mL. Plots showing the deviation of data points from the fit (residu-
als) are shown below each sensorgram.
gp120 in a dose-dependent manner with affinity in the sub-
micromolar range (KD ꢂ10ꢁ8–10ꢁ10 m) (data not shown).
GNP 4 (0.1–1.0 mm) showed the strongest binding affinity,
with a KD value of the order of 10ꢁ11 m (see the Supporting
Information, Figure S4A). SPR analysis suggested that GNP
4 and DC-SIGN do not share binding sites on gp120 (Fi-
gure S4B). The binding activities of dimannose- and glucose-
In experiments in which fluid-phase GNPs alone were in-
jected on immobilized gp120, unexpected behaviour was ob-
served. Most GNPs, including gluco-GNPs and nanoparticle
HO2C-Au, harbouring only the carboxylic linker, bound
9882
www.chemeurj.org
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9874 – 9888

Gold Manno-Glyconanoparticles
FULL PAPER
coated GNPs to gp120 are difficult to interpret. Carbohy-
drate–protein interactions or Ca2+-mediated carbohydrate–
carbohydrate interactions may be the basis of this effect.
Lectin-binding has been described for gp41[52] but not for
gp120. Because gp120 is heavily glycosylated, gp120 interac-
tion with manno-GNPs could be of the carbohydrate–carbo-
hydrate type, as shown for Lewis X-coated GNPs[4,7] and
proposed for the interaction of integrin a5b1 with glyco-
sphingolipids.[52,53] Binding to gp120 by nanoparticles bearing
carboxylic acid groups can also be explained in terms of
electrostatic interactions with positively charged surfaces of
the glycoprotein.[54] Non-specific binding activity by the gold
core can be ruled out as GNPs 1 fully functionalized with
the short linker conjugate ManC2S did not show significant
binding to gp120 (Figure S4A).
Experimental Section
General procedures: All chemicals were purchased as reagent grade
from Sigma–Aldrich, except chloroauric acid (Strem Chemicals), and
were used without further purification. Aminopropyl mannosides 18–21
were provided by Carbohydrate Synthesis Ltd. Reactions were monitored
by thin-layer chromatography (TLC) on silica gel 60 F254 aluminium-
backed sheets (Merck) with visualization under UV (254 nm) and/or by
staining with p-anisaldehyde solution [anisaldehyde (25 mL), H2SO4
(25 mL), EtOH (450 mL), and CH3COOH (1 mL)], 10% H2SO4 solution
in EtOH, ninhydrin solution [ninhydrin (0.25 mL), EtOH (100 mL)] or
phosphomolybdic acid solution [phosphomolybdic acid (13 g), CeACHTUNGTRENNUNG(SO4)2
(10 g), H2SO4 (60 mL), H2O (940 mL)] followed by heating at over
2008C. Size-exclusion column chromatography was performed on Sepha-
dex LH-20 (GE Healthcare). Flash column chromatography (FCC) was
performed on silica gel 60 (0.063–0.200 mm or 0.015–0.040 mm; Merck).
UV/Vis spectra were measured with Perkin–Elmer Lambda 12 or Beck-
man Coulter DU 800 UV/Vis spectrophotometers. Infrared (IR) spectra
were recorded from 4000 to 750 cmꢁ1 with a JASCO FT-IR 410 model
spectrometer; solids were pressed into KBr pellets and oils were subject-
ed to attenuated total reflection (ATR). 1H and 13C NMR spectra were
recorded on Bruker DPX-300 (300 MHz) or Bruker AVANCE
(500 MHz) spectrometers. Chemical shifts (d) are given in ppm relative
to the residual signal of the solvent used. Coupling constants (J) are re-
ported in Hz. Splitting patterns are described by using the following ab-
breviations: br, broad; s, singlet; d, doublet; t, triplet; m, multiplet. Mass
spectra were measured with an Esquire 6000 ESI-Ion Trap spectrometer
from Bruker Daltonics. High-resolution mass spectra (HRMS) were ob-
tained using the MALDI technique with a 4700 Proteomics Analyzer
(Applied Biosystems) operated in MALDI-TOF-TOF configuration.
Samples of the products were dissolved in water, 2,5-dihydroxybenzoic
acid (DHB) was used as a matrix, cesium iodide was added to favour the
ionization process, and polyethylene glycol was used as an internal refer-
ence. Optical rotations were determined with a Perkin–Elmer 341 polar-
imeter. For transmission electron microscopy (TEM) examinations, a
single drop (10 mL) of an aqueous solution (ca. 0.1 mgmLꢁ1 in Milli-Q
water) of the gold glyconanoparticles (GNPs) was placed on a copper
grid coated with a carbon film (Electron Microscopy Sciences). The grid
was left to dry in air for several hours at room temperature. TEM analy-
sis was performed with a Philips CM200 or a Philips JEOL JEM-2100F
microscope, both operating at 200 kV. The average diameters and num-
bers of gold atoms of the GNPs were deduced as described in a previous
study.[50] Laboratory distilled water was further purified using a Milli-Q
reagent grade water system (Millipore).
Because manno-GNPs can be used as potential microbi-
cides that block the binding of HIV and other pathogens to
DC-SIGN-expressing cells and thus prevent infection,[24,55]
we are now investigating their cytotoxicity and their potency
as inhibitors of DC-SIGN-mediated HIV-1 trans-infection in
human T cells.
Conclusions
In summary, we have reported the synthesis and characteri-
zation of a small library of (oligo)mannose glyconanoparti-
cles (manno-GNPs) incorporating diverse neoglycoconju-
gates synthesized either by direct glycosylation (first family:
GNPs 1–3 and 3a–c), by peptide linkage (second family:
GNPs 4–7), or by thiourea linkage (third family: GNPs 8, 9,
9a,b, 10a,b, 11a,b, 12a,b, and 13a). The thiourea strategy
proved to be the most efficient in terms of yields and versa-
tility of the synthetic procedure. The generated GNPs have
been tested by SPR for competitive inhibition of DC-SIGN/
gp120 binding. The results of this study have allowed us to
identify the best inhibitor among these new biocompatible
3D systems and have confirmed a multivalent effect of the
mannosides on the gold surface compared with the corre-
sponding monomer conjugates. The Mana1-2Mana-contain-
ing GNPs were found to be the best inhibitors, with inhibi-
tion values in the nanomolar range. In this study, changing
the density of Mana1-2Mana on GNPs from 50% to 100%
did not improve the inhibitory activity. Values are in agree-
ment with those reported for more complex (oligo)mannose
clusters and dendrimers.[25,26] The gold clusters offer a rigid
scaffold, the size of which can be varied. As has been
shown, the glyconanoparticle technology allows the selec-
tion of length and flexibility of the linkers as well as sugar
density for improved presentation of the (oligo)mannosides.
Furthermore, by means of glyconanotechnology, the GNPs
platform can be simultaneously tailored with different bio-
logically-relevant molecules (such as immunogenic peptides
or other antigens),[27] thereby allowing the preparation of
multifunctional structures as potential carbohydrate-based
systems against HIV.[56]
Recombinant proteins: gp120 from HIV-1 strain CN54 (clade C) was
produced using the baculovirus system and was kindly provided by Ian
Jones (University of Reading, Reading, UK). The full-length extracellu-
lar domain of DC-SIGN (DC-SIGN-ECD) was expressed and purified as
follows: cDNA of iMDDC was used as a template to generate the DNA
fragment encoding the entire extracellular domain of DC-SIGN (residues
70–404; GenBank accession number NP 066978) by PCR with primers
5’-GTCTCGAGATGGAACAATCCAGGCAAGACGCGATCT-3’
(sense) and 5’-TCGGATCCCTACGCAGGAGGGGGGTTTGGGGT-3’
(antisense). The amplified sequence, digested with XhoI and BamHI, was
inserted into pET15b (Novagen, EMD Chemicals, Gibbstown, NJ, USA)
and cloned in E. coli TOP10 (Invitrogen, Paisley, UK). Cloned fragments
were confirmed by DNA sequencing (Advanced Biotechnology Centre,
Imperial College London, London, UK) and compared with GenBank
(accession number NM_021155). For expression, E. coli strain BL21/DE3
(Stratagene, La Jolla, CA, USA) was transformed with recombinant plas-
mid. Protein expression and refolding was performed as described else-
where[57] with minor modifications. Inclusion bodies (from 1 L bacterial
culture) were recovered by centrifugation at 10000 ꢇ g for 20 min at 48C
and solubilized in 8 mL of 100 mm Tris-HCl, pH 8.0, containing 6m urea
(solubilizing buffer) supplemented with 0.01% 2-mercaptoethanol, by
gentle rotation overnight at 48C. The mixture was centrifuged at 20000 ꢇ
g for 30 min at 48C and soluble recombinant protein was isolated by Ni2+
-affinity chromatography. Bound material (recovered by elution with
Chem. Eur. J. 2009, 15, 9874 – 9888
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9883

S. Penadꢂs et al.
200 mm imidazole in solubilizing buffer) was dialyzed against 2 L of
100 mm Tris-HCl, pH 8.0, containing 0.01% 2-mercaptoethanol, 10 mm
CaCl2, and 6m urea, then successively against the same buffer with 4m
urea, 2m urea, and no urea. Final dialysis was against 100 mm Tris-HCl,
pH 8.0, containing 10 mm CaCl2. After dialysis, the insoluble precipitate
was removed by centrifugation at 100000 ꢇ g for 30 min at 48C and re-
folded DC-SIGN contained in the soluble fraction was purified by d-
mannose affinity chromatography as previously described.[58] Fractions
were analyzed by SDS-PAGE and protein concentrations were deter-
mined by densitometric analysis using the GeneSnap software (Syngene,
Cambridge, UK). The identity of the protein was confirmed by LC MS/
MS analysis (MRC Clinical Sciences Centre, Imperial College London,
London, UK).
zation afforded the glycoconjugates 33–38. For the sake of clarity, these
neoglycoconjugates are all named as their disulfides. The thiol/disulfide
ratio is reported in each case on the basis of the integrals of the relevant
1
signals in the corresponding H NMR spectra.
23,23’-Dithiobis[N-(ethyl a-d-mannopyranosyl),N’-(3,6,9,12-tetraoxa-tri-
cosanyl)thiourea] (33): Reaction of mannoside 27 (50.0 mg, 0.224 mmol)
and linker 39 (207.7 mg, 0.448 mmol) afforded neoglycoconjugate 33 in
the form of the disulfide (117.4 mg, 0.182 mmol, 81% over two steps) as
a colourless syrup after purification by passage through Sephadex LH-20
(CH2Cl2/MeOH 4:1). 1H NMR (300 MHz, CD3OD): d=4.78 (s, 1H; 1-
H), 3.90–3.51 (m, 26H), 3.47 (t, J=6.6 Hz, 2H; OCH2CH2CH2), 2.69 (t,
J=6.9 Hz, 2H; CH2SS), 1.72–1.64 (m, 2H), 1.61–1.53 (m, 2H), 1.47–
1.20 ppm (m, 14H); 13C NMR (75 MHz, CD3OD): d=101.8 (d; C-1),
74.8, 72.6, 72.4, 72.1, 71.6, 71.4, 71.2, 70.7, 68.6, 67.4 (t; NHCH2CH2O),
62.9 (t; C-6), 45.3 (brt; CH2NHCS), 39.9 (t; CH2SS), 30.8, 30.7, 30.6,
30.36, 30.27, 29.5, 27.3 ppm; C=S undetected; IR (neat): n˜ =3331 (brs),
2924, 2854, 1647, 1560, 1458, 1348, 1294, 1097 cmꢁ1; HRMS: m/z: calcd
for C28H56N2O10S2Na+ [M+Na]+: 667.3274; found: 667.3273.
Surface plasmon resonance assays
Inhibition studies: gp120 in 10 mm sodium acetate at pH 4.0 was immobi-
lized (approximately 1000 RU, which corresponds to ~1 ng of immobi-
lized ligand on a CM5 sensor chip)[59] on the surface of Flow cell 2 of a
CM5 sensor chip following the standard amine coupling procedure (GE
Healthcare, Uppsala, Sweden). Flow cell 1 (activated and blocked with
ethanolamine) served as a reference (blank) cell. Non-saturation binding
concentrations of DC-SIGN were chosen to increase the sensitivity of the
inhibition assay. Binding of fluid-phase DC-SIGN-ECD at a concentra-
tion of 50 nm was determined in the presence of free alkyl amino (oligo)-
mannosides (0.01–3 mm), methyl a-d-mannopyranoside (0.5–25 mm,
Sigma–Aldrich), or GNPs and compared to the binding activity of DC-
SIGN-ECD alone. GNPs 1, 2, 3, and 3a were tested at concentrations in
the range 0.5–20 mm. For GNPs 5 and 6, concentrations were varied in
the range 0.1–2.0 mm. GNPs 4, 9, and 9a were tested at 6–130 nm.
GlcC5Au and HO2C-Au GNPs were tested at 1 mm and 2 mm, respectively.
For GNPs that showed direct binding activity to gp120, final inhibition
sensorgrams were obtained by subtraction of the sensorgram for injection
of GNPs alone from that generated by injection of GNPs in the presence
of DC-SIGN. Fluid-phase compounds were dissolved in HBS-P buffer
(10 mm HEPES [pH 7.4], 0.15m NaCl, 0.005% v/v surfactant P20; GE
Healthcare) supplemented with 10 mm CaCl2. The flow rate was
20 mLminꢁ1 and the injection volume was 20 mL. After each binding mea-
surement, the surface was regenerated with 10 mm EDTA. Single mea-
surements were carried out for each condition.
23,23’-Dithiobis[N-(ethyl a-d-mannopyranosyl-(1!2)-a-d-mannopyrano-
syl),N’-(3,6,9,12-tetraoxa-tricosanyl)thiourea] (34): Reaction of manno-
side 28 (25.0 mg, 0.065 mmol) and linker 39 (60.1 mg, 0.130 mmol) afford-
ed neoglycoconjugate 34 in the form of the disulfide (38.3 mg,
0.047 mmol, 73% over two steps) as a colourless syrup after purification
by passage through Sephadex LH-20 (MeOH/H2O 9:1). 1H NMR
(500 MHz, CD3OD): d=5.11 (d, J=1.5 Hz, 1H; 1-H), 4.97 (d, J=1.5 Hz,
1H; 1’-H), 3.98–3.96 (m, 1H; 2’-H), 3.90–3.80 (m, 5H), 3.75–3.50 (m,
26H), 3.47 (t, J=6.5 Hz, 2H; OCH2CH2CH2), 2.69 (t, J=7 Hz, 2H;
CH2SS), 1.72–1.63 (m, 2H), 1.61–1.53 (m, 2H), 1.44–1.28 ppm (m, 14H);
13C NMR (125 MHz, D2O): d=104.2 (d; C-1), 100.1 (d; C-1’), 80.5 (d; C-
2), 75.1, 74.8, 72.4, 72.1, 71.9, 71.6, 71.3, 71.1, 70.7, 69.0, 68.8, 67.4, 63.2,
and 63.0 (t, 2C; C-6 and C-6’), 45.3 (brt; CH2NH2), 39.8 (t; CH2SS), 30.7,
30.6, 30.6, 30.3, 30.2, 29.4, 29.4, 27.2 ppm; C=S undetected; IR (neat): n˜ =
3330 (brs), 2923, 2853, 1645, 1556, 1456, 1348, 1297, 1057 cmꢁ1; HRMS:
m/z: calcd for C34H66N2O15S2Na+ [M+Na]+: 829.3802; found: 829.3802.
23,23’-Dithiobis[N-(ethyl a-d-mannopyranosyl-(1!2)-a-d-mannopyrano-
syl-(1!2)-a-d-mannopyranosyl),N’-(3,6,9,12-tetraoxa-tricosanyl)thiourea]
(35): Reaction of the formate salt of mannoside 29 (21.4 mg, 0.036 mmol)
and linker 39 (33.0 mg, 0.071 mmol) afforded neoglycoconjugate 35
(24.4 mg, 0.025 mmol, 72% over two steps) as a white solid after purifica-
tion by passage through Sephadex LH-20 (MeOH/H2O 9:1). 1H NMR
(500 MHz, CD3OD): d=5.28 (d, J=1.5 Hz, 1H; 1-H), 5.10 (d, J=1.5 Hz,
1H; 1’-H), 4.98 (d, J=1.5 Hz, 1H; 1’’-H), 4.05–4.02 (m, 1H; 2-H), 3.99–
3.95 (m, 1H; 2’’-H), 3.89–3.80 (m, 6H), 3.76–3.50 (m, 30H), 3.47 (t, J=
6.5 Hz, 2H; OCH2CH2CH2), 2.69 (t, J=7.0 Hz, 0.3H; CH2SS), 2.49 (t,
J=7.5 Hz, 1.7H; CH2SH), 1.63–1.53 (m, 4H), 1.45–1.35 ppm (m, 14H);
13C NMR (125 MHz, CD3OD): d=104.1 (d; C-1’’), 102.5 (d; C-1), 100.0
(d; C-1’), 80.6 (d; C-2’), 80.2 (d; C-2), 75.02, 75.00, 74.7, 72.43, 72.40, 72.0,
71.92, 71.87, 71.5, 71.3, 71.1, 70.7, 69.2, 69.0, 68.8, 67.4; 63.3, 63.2, and
62.9 (t, 3C; 3ꢇC-6), 45.2 (brt; CH2NH), 35.2 (t; CH2SS), 30.69, 30.63,
30.5, 30.2, 29.4, 27.2, 25.0 ppm (t; CH2SH); C=S undetected; HRMS:
m/z: calcd for C40H76N2O20S2Na+ [M+Na]+: 991.4331; found: 991.4330.
Affinity measurements: DC-SIGN-ECD (500–700 RU) or gp120 (approx-
imately 500 RU) were immobilized in Flow cell 2 on a CM5 sensor chip
as described above. In both cases, Flow cell 1, treated as above, served as
a reference cell. Binding of fluid-phase GNPs was determined over a
range of concentrations (2.3–10.7 mm when determining binding to DC-
SIGN, 0.01–5.0 mm when determining binding to gp120) in HBS-P supple-
mented with 10 mm CaCl2. GlcC5Au and HO2C-Au GNPs were tested at
concentrations as high as 2 mm. The injection volume was 20 mL and the
flow rate was 20 mLminꢁ1. The surface was regenerated with 10 mm
EDTA. Equilibrium dissociation constants (KD) as well as association
(ka) and dissociation constant (kd) rates were calculated using the BIA
evaluation software 4.1 (GE Healthcare). Curves were first fitted to a
single 1:1 binding model and then to more complex binding models, se-
lecting that which gave the best fit as judged by the lowest c2 value and
the best distribution of residuals.
23,23’-Dithiobis[N-(ethyl a-d-mannopyranosyl-(1!2)-a-d-mannopyrano-
syl-(1!2)-a-d-mannopyranosyl-(1!3)-a-d-mannopyranosyl),N’-(3,6,9,12-
tetraoxa-tricosanyl)thiourea] (36): Reaction of mannoside 30 (63.0 mg,
0.0874 mmol) and linker 39 (81.0 mg, 0.175 mmol) afforded neoglycocon-
jugate 36 (80.0 mg, 0.071 mmol, 81% over two steps) as a white solid
after passage through Sephadex LH-20 (MeOH/H2O 9:1). 1H NMR
(500 MHz, D2O): d=5.38 (s, 1H), 5.32 (s, 1H), 5.07 (s, 1H), 4.87 (s, 1H),
4.13–3.59 (m, 44H), 3.51 (brt, 2H; OCH2CH2CH2), 2.73 (t, J=7.1 Hz,
2H; CH2SS), 1.78–1.56 (m, 4H), 1.49–1.28 ppm (m, 14H); 13C NMR
(125 MHz, D2O): d=102.2, 100.62 (overlapped), 99.7, 78.4, 78.1, 73.3,
73.23, 73.17, 72.9, 71.2, 70.3, 70.0, 69.9, 69.7, 69.2, 66.9, 66.8, 65.9, 61.0,
60.9, 60.7, 43.7 (brt; CH2NH), 39.0, 34.1, 29.7, 29.6, 29.5, 29.3, 28.5,
26.1 ppm; C=S undetected; IR (KBr): n˜ =3361 (broad), 2926, 2856, 1646,
General procedure for thiourea coupling
A solution of the respective aminoethyl (oligo)mannoside 27–32 (0.09m,
1 equiv) in methanol (for 27, 28, and 30) or H2O/iPrOH/CH3CN (1:1:1)
(for 29, 31, and 32) was added to a solution of isothiocyanate linker 39
(0.12m, 2 equiv) in methanol or H2O/iPrOH/CH3CN (1:1:1), respectively.
The pH was adjusted to 8–9 with triethylamine and the solution was
stirred for 3–5 h at room temperature. The solvent was then removed
under reduced pressure and the crude residue was triturated with Et2O
to remove the excess linker (except in the case of the neoglycoconjugate
of 27). The resulting thioacetyl derivatives were purified by FCC or on
Sephadex LH-20 (for complete characterization of these intermediates,
see the Supporting Information). The thioacetyl derivatives were treated
with sodium methoxide (1 equiv, 1 n in MeOH). The resulting mixture
was stirred for 2 h at room temperature and then neutralized with 0.1n
HCl. Purification on a column of Sephadex LH-20 followed by lyophili-
1556, 1459, 1352, 1296, 1131, 1058 cmꢁ1
C46H86N2O25S2Na+ [M+Na]+: 1153.4859; found: 1153.4879.
23,23’-Dithiobis[N-(ethyl
(bis(a-d-mannopyranosyl-(1!2)-a-d-manno-
pyranosyl-(1!3,6))-a-d-mannopyranosyl)),N’-(3,6,9,12-tetraoxa-tricosa-
; HRMS: m/z: calcd for
A
ACHTUNGTRENNUNG
9884
www.chemeurj.org
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9874 – 9888

Gold Manno-Glyconanoparticles
FULL PAPER
nyl)thiourea] (37): Reaction of the formate salt of mannoside 31
(18.9 mg, 0.0206 mmol) and linker 39 (19.1 mg, 0.041 mmol) afforded ne-
oglycoconjugate 37 (19.4 mg, 0.015 mmol, 73% over two steps) as a
white solid after purification by passage through Sephadex LH-20
(MeOH/H2O 5:1). 1H NMR (D2O, 500 MHz): d=5.38 (s, 1H), 5.16 (s,
1H), 5.08 (s, 1H), 5.06 (s, 1H), 4.86 (s, 1H), 4.16–3.60 (m, 50H), 3.52
(brt, 2H; OCH2CH2CH2), 2.73 (brt, 0.7H; CH2SS), 2.55 (t, 1.3H, J=
6.9 Hz; CH2SH); 1.80–1.56 (m, 4H), 1.49–1.26 ppm (m, 14H); 13C NMR
(D2O, 125 MHz): d=102.3, 102.3, 100.7, 100.0, 98.0, 78.7, 78.4, 73.2, 72.7,
72.0, 71.1, 70.8, 70.4, 70.3, 69.9, 69.7, 69.2, 69.1, 66.9, 66.7, 65.9, 65.6, 65.3,
61.1, 60.9, 43.7 (brt, 2C; OCH2CH2N), 39.0 (t; CH2SS), 34.0, 29.72, 29.69,
29.5, 29.4, 29.2, 28.4, 26.1, 26.0, 24.3 ppm (t; CH2SH); C=S undetected;
IR (KBr): n˜ =3386 (broad), 2926, 2854, 1642, 1556, 1462, 1366, 1301,
1131, 1057 cmꢁ1; HRMS: m/z: calcd for C52H96N2O30S2Na+ [M+Na]+:
1315.5382; found: 1315.5380.
dark-brown powder. TEM (average diameter and number of gold
atoms): 1.6ꢀ0.5 nm, 140; 1H NMR (500 MHz, D2O): d=4.70 (s, 1H; 1-
H), 4.05–3.00 (m, 6H), 2.11–1.02 ppm (m, 8H); UV/Vis (H2O,
0.1 mgmLꢁ1): l=530 nm (surface plasmon band); elemental analysis
calcd (%) for (C11H22O6S)73Au140 (48 kDa): C 20.01, H 3.36, S 4.86;
found: C 20.16, H 3.22, S 4.70.
ManEG6C11-Au (100%) (3): Reaction of 16 (45.0 mg, 0.071 mmol) with
HAuCl4 (520 mL, 0.025m) and NaBH4 (686 mL, 1 n) gave 3 (1.2 mg) as a
brown powder. TEM (average diameter and number of gold atoms):
1
1.0ꢀ0.9 nm, 79; H NMR (500 MHz, D2O): d=5.60 (s, 1H; 1-H), 4.61 (s,
1H; 2-H), 3.90–3.41 (m, 31H), 2.78–2.69 (m, 2H), 1.69–1.55 (m, 2H),
1.46–1.36 ppm (m, 16H); UV/Vis (H2O, 0.1 mgmLꢁ1): l=525 nm (sur-
face plasmon band).
ManEG6C11-Au-GlcC2 ACHTNUTRGNEUGN(30%) (3a): Reaction of a 3:5 mixture of 16
(30.0 mg, 0.048 mmol) and GlcC2S (26.0 mg, 0.108 mmol) with HAuCl4
(1.3 mL, 0.025m) and NaBH4 (620 mL, 1 n) gave 3a (4.1 mg) as a brown
powder. TEM (average diameter and number of gold atoms): 1.7ꢀ
0.5 nm, 201; 1H NMR (500 MHz, D2O): d=4.98 (brs, 0.3H; 1-H), 4.53
(brm, 1H; 1-H glucose), 4.13–3.25 (brm, 20H), 1.96–1.08 ppm (brm,
5.4H); UV/Vis (H2O, 0.1 mgmLꢁ1): l=525 nm (surface plasmon band).
23,23’-Dithiobis[N-
G
N
pyranosyl-(1!2)-a-d-mannopyranosyl-(1!3,6))-a-d-mannopyrano-
syl)),N’-(3,6,9,12-tetraoxatricosanyl)thiourea] (38): Reaction of the for-
mate salt of mannoside 32 (17.2 mg, 0.0138 mmol) and linker 39 (12.8 mg,
0.0276 mmol) afforded neoglycoconjugate 38 (12.1 mg, 0.0075 mmol,
52% over two steps) as a white solid after purification by passage
through Sephadex LH-20 (MeOH/H2O 5:1). 1H NMR (500 MHz,
CD3OD): d=5.37 (s, 1H), 5.32 (s, 1H), 5.32 (s, 1H), 5.13 (s, 1H), 5.07 (s,
1H), 5.07 (s, 1H), 4.86 (s, 1H), 4.14–3.63 (m, 62H), 3.52 (brt, 2H;
OCH2CH2CH2), 2.73 (brt, 1.1H; CH2SS), 2.56 (t, 0.9H, J=6.9 Hz,
CH2SH), 1.79–1.56 (m, 4H), 1.49–1.25 ppm (m, 14H); 13C NMR (D2O,
125 MHz): d=102.2, 102.2, 100.6, 100.6, 100.6, 100.0, 98.0, 78.5, 78.4, 73.2,
72.7, 71.3, 71.2, 70.3, 69.9, 69.7, 67.0, 66.9, 66.8, 61.0, 43.9 (brt, 2C;
OCH2CH2N), 38.8 (t; CH2SS), 29.8, 29.7, 29.5, 29.4, 29.3, 28.6, 26.1,
24.1 ppm (t; CH2SH); C=S undetected; IR (KBr): n˜ =3395 (broad), 2926,
2854, 1644, 1565, 1463, 1369, 1302, 1131, 1054 cmꢁ1; HRMS: m/z: calcd
for C64H116N2O40S2Na+ [M+Na]+: 1639.6443; found: 1639.6440.
ManEG6C11-Au-GlcC2 ACTHNUTRGNE(NUG 15%) (3b): Reaction of a 15:85 mixture of 16
(19.0 mg, 0.030 mmol) and GlcC2S (41.0 mg, 0.171 mmol) with HAuCl4
(1.46 mL, 0.025m) and NaBH4 (800 mL, 1 n) gave 3b (7.5 mg) as a brown
powder. TEM (average diameter and number of gold atoms): 1.7ꢀ
0.5 nm, 201; 1H NMR (500 MHz, D2O): d=4.98 (brs; 1-H), 4.53 (brm,
1H; 1-H glucose), 4.35–3.12 (brm, 13H), 1.77–1.17 ppm (brm, 2.7H);
UV/Vis (H2O, 0.1 mgmLꢁ1): l=525 nm (surface plasmon band).
ManEG6C11-Au-GlcC2 ACHTNUTRGNEUGN(5%) (3c): Reaction of a 5:95 mixture of 16
(6.0 mg, 0.0095 mmol) and GlcC2S (43.0 mg, 0.179 mmol) with HAuCl4
(1.37 mL, 0.025m) and NaBH4 (980 mL, 1 n) gave 3c (8.2 mg) as a brown
powder. TEM (average diameter and number of gold atoms): 1.4ꢀ
0.4 nm, 116; 1H NMR (500 MHz, D2O): d=5.02 (brs; 1-H), 4.54 (brm,
1H; 1-H glucose), 4.21–3.35 (brm, 10H), 1.62–1.10 ppm (brm, 0.9H);
UV/Vis (H2O, 0.1 mgmLꢁ1): l=525 nm (surface plasmon band).
General procedure for preparation of glyconanoparticles
A 0.012m (3 equiv) methanolic solution of the appropriate disulfide 14–
16 and mixture 22’–25’, or a mixture of disulfide 16 or 33–38 in different
ratios (5:95, 10:90, 15:85, 30:70, or 50:50) with conjugate GlcC2S or
GlcC5S was added to a solution of tetrachloroauric acid (0.025m, 1 equiv)
in water. An aqueous solution of NaBH4 (1m, 22 equiv) was then added
in four portions, with rapid shaking. The black suspension formed was
shaken for an additional 2 h at 258C and then the supernatant was re-
moved and analysed. The residue was dissolved in the minimum volume
of NANOPURE water and purified by dialysis. This solution was loaded
into 5–10 cm segments of SnakeSkin pleated dialysis tubing (Pierce, 3500
MWCO), which were placed in a 3 L beaker of water. The contents of
the beaker were stirred slowly, recharging with fresh distilled water every
3–4 h over the course of 72 h. The solution in the membrane was then
lyophilized to afford the GNP. 1H NMR spectra of the glycoconjugate
mixtures used for the GNP synthesis and of the products recovered from
the supernatant after GNP formation were recorded. The ratio of the li-
gands in the GNPs was confirmed through integration of the signals of
the anomeric protons of the mannoside with respect to those of the
anomeric protons of the glucoside. The particle size distribution of the
gold nanoparticles was evaluated from several TEM micrographs by
means of an automatic image analyser. The average diameter and
number of gold atoms of the GNPs was assigned according to a previous
work.[50] The average molecular formula of the nanoparticles was calcu-
lated on the basis of the average diameter obtained by TEM and con-
firmed by elemental analysis.
Man2-Au-CO2H (66%) (4): Reaction of the disulfide mixture of 22’
(29.2 mg, 0.039 mmol) in a 2:1 mannose/carboxylic acid linker ratio,
HAuCl4 (288 mL, 1 n), and NaBH4 (158 mL, 1 n) gave 4 (7.1 mg) as a
light-brown powder soluble in methanol. TEM (average diameter and
number of gold atoms): 1.3ꢀ0.6 nm, 38; 1H NMR (500 MHz, D2O): d=
5.10 (s, 1H; 1-H), 5.04 (s, 1H; 1’-H), 4.32 (s, 1H; CH2CO2H), 4.09 (s,
3H), 4.01–3.32 (brm, 53H), 2.73 (brm; CH2S), 1.95–1.26 ppm (brm,
29H); UV/Vis (H2O, 0.1 mgmLꢁ1): surface plasmon band not observed.
Man2-Au-CO2H (50%) (5): Reaction of the disulfide mixture of 23’
(17.9 mg, 0.024 mol) in
a 1:1 mannose/carboxylic acid linker ratio,
HAuCl4 (177 mL, 1 n), and NaBH4 (97 mL, 1 n) gave 5 (3.2 mg) as a
brown powder. TEM (average diameter and number of gold atoms):
1
1.0ꢀ0.4 nm, 38; H NMR (500 MHz, D2O): d=5.14 (s, 1H; 1-H), 4.83 (s,
1H; 1’-H), 4.13–4.06 (brm, 4H), 3.99–3.32 (brm, 58H), 2.71 (brm;
CH2S), 1.98–1.19 ppm (brm, 38H); UV/Vis (H2O, 0.1 mgmLꢁ1): surface
plasmon band not observed.
Man3-Au-CO2H (66%) (6): Reaction of the disulfide mixture of 24’
(30.4 mg, 0.036 mmol) in a 2:1 mannose/carboxylic acid linker ratio,
HAuCl4 (262 mL, 1n), and NaBH4 (143 mL, 1 n) gave 6 (3.5 mg) as a
brown powder. TEM (average diameter and number of gold atoms):
1
1.3ꢀ0.5 nm, 79; H NMR (500 MHz, D2O): d=5.29 (s, 1H; 1-H), 5.08 (s,
1H; 1’-H), 5.05 (s, 1H; 1’’-H), 4.31 (s, 1H; CH2CO2H), 4.11 (s, 1H), 4.07
(s, 3H), 3.99–3.32 (brm, 53H), 2.76 (brs; CH2S), 1.96–1.04 ppm (brm,
30H); UV/Vis (H2O, 0.1 mgmLꢁ1): surface plasmon band not observed.
ManC2-Au (100%) (1): Reaction of 14 (45.0 mg, 0.187 mmol) with
HAuCl4 (1.38 mL, 0.025m) and NaBH4 (760 mL, 1 n) gave 1 (5.8 mg) as a
dark-brown powder. TEM (average diameter and number of gold
atoms): 2.0ꢀ0.6 nm, 225; 1H NMR (500 MHz, D2O): d=4.95 (s, 1H; 1-
H), 4.40–3.50 ppm (m, 10H); UV/Vis (H2O, 0.1 mgmLꢁ1): l=520 nm
(surface plasmon band); elemental analysis calcd (%) for
(C8H15O6S)121Au225 (73 kDa): C 15.87, H 2.50, S 5.30; found: C 15.89, H
3.00, S 5.12.
Man3-Au-CO2H (50%) (7): Reaction of the disulfide mixture of 25’
(53.5 mg, 0.077 mmol) in a 1:1 mannose/carboxylic acid linker ratio,
HAuCl4 (563 mL, 1 n), and NaBH4 (309 mL, 1 N) gave 7 (7.1 mg) as a
brown powder. TEM (average diameter and number of gold atoms):
1
1.3ꢀ0.5 nm, 79; H NMR (500 MHz, D2O): d=5.13 (s, 1H; 1-H), 4.91 (s,
1H; 1’-H), 4.83 (s, 1H; 1’’-H), 4.31 (s, 2H; CH2CO2H), 4.18–3.44 (brm,
68H), 2.79 (brs; CH2S), 1.96–1.26 ppm (brm, 38H); UV/Vis (H2O,
0.1 mgmLꢁ1): surface plasmon band not observed.
ManC5-Au (100%) (2): Reaction of 15 (50.0 mg, 0.177 mmol) with
HAuCl4 (1.29 mL, 0.025m) and NaBH4 (686 mL, 1 n) gave 2 (3.6 mg) as a
Chem. Eur. J. 2009, 15, 9874 – 9888
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9885

S. Penadꢂs et al.
Man-Au (100%) (8): Reaction of 33 (78.1 mg, 0.121 mmol) with HAuCl4
(1.61 mL, 0.025m) and NaBH4 (885 mL, 1 n) gave 8 (11.1 mg) as a light-
brown powder. TEM (average diameter and number of gold atoms):
1.3ꢀ0.4 nm, 79; 1H NMR (500 MHz, D2O): d=4.89 (s, 1H; 1-H), 3.98
(brs, 1H; 2-H), 3.94–3.37 (brm, 27H), 2.74 (brm; CH2S), 1.91–0.99 ppm
(brm, 18H); UV/Vis (H2O, 0.3 mgmLꢁ1): surface plasmon band not ob-
served; elemental analysis calcd (%) for (C28H55N2O10S2)40Au79 (41 kDa):
C 32.56, H 5.37, N 2.71, S 6.21; found: C 32.67, H 5.32, N 2.75, S 7.36.
Man4-Au-GlcC5
0.010 mmol) and GlcC5S (25.4 mg, 0.09 mmol) with HAuCl4 (1.344 mL,
0.025m) and NaBH4 (738 mL, 1 n) gave 11b (13.0 mg) as brown
powder. TEM (average diameter and number of gold atoms): 1.4ꢀ
0.7 nm, 116; 1H NMR (500 MHz, D2O): d=5.28 (brs, 1H), 5.23 (brs,
1H), 4.97 (brs, 1H), 4.71 (brs, 1H, partially overlapped by water signal),
4.36 (brm, 9H; 1-H glucose), 4.10–3.13 (brm, 118H), 1.87–1.10 ppm
(brm, 72H); UV/Vis (H2O, 0.1 mgmLꢁ1): l=527 nm (surface plasmon
ACHTUNGTREN(NGNU 10%) (11b): Reaction of a 1:9 mixture of 36 (11.4 mg,
a
band);
elemental
analysis
calcd
(%)
for
Man2-Au (100%) (9): Reaction of 34 (30.0 mg, 0.037 mmol) with HAuCl4
(495 mL, 0.025m) and NaBH4 (271 mL, 1 n) gave 9 (9.0 mg) as a light
brown powder. TEM (average diameter and number of gold atoms):
(C46H85N2O25S2)7(C11H21O6S)59Au116 (47 kDa): C 24.63, H 3.90, N 0.41, S
4.94; found: C 24.40, H 4.36, N 0.77, S 4.53.
1
Man5-Au-GlcC5 ACTHNUGTRNEUNG(50%) (12a): Reaction of a 1:1 mixture of 37 (14.0 mg,
1.2ꢀ0.5 nm, 79; H NMR (500 MHz, D2O): d=5.14 (s, 1H; 1-H), 5.05 (s,
0.011 mmol) and GlcC5S (3.05 mg, 0.011 mmol) with HAuCl4 (288 mL,
0.025m) and NaBH4 (158 mL, 1 n) gave 12a (4.4 mg) as a brown powder.
TEM (average diameter and number of gold atoms): 2.1ꢀ1.5 nm, 309;
1H NMR (500 MHz, D2O): d=5.37 (s, 1H), 5.16 (s, 1H), 5.08 (s, 1H),
5.06 (s, 1H), 4.86 (s, 1H), 4.43 (brm, 1H; 1-H glucose), 4.23–3.18 (brm,
60H), 2.75 (brm; CH2S), 1.89–1.13 ppm (brm, 24H); UV/Vis (H2O,
0.1 mgmLꢁ1): surface plasmon band not observed; elemental analysis
calcd (%) for (C52H95N2O30S2)28(C11H21O6S)28Au309 (105 kDa): C 20.19, H
3.12, N 0.71, S 2.57; found: C 20.12, H 3.53, N 0.89, S 3.03.
1H; 1’-H), 4.09 (m, 1H; 2’-H), 4.05–3.45 (brm, 33H), 2.73 (brm; CH2S),
1.74 (brm, 2H), 1.62 (brm, 2H), 1.55–1.25 ppm (brm, 14H); UV/Vis
(H2O, 0.1 mgmLꢁ1): l=519 nm (surface plasmon band); elemental analy-
sis calcd (%) for (C34H65N2O15S2)59Au79 (63 kDa): C 38.17, H 6.08, N 2.62,
S 5.99; found: C 38.24, H 6.23, N 2.46, S 6.63.
Man2-Au-GlcC5 ACHTUNGTRENNUNG(50%) (9a): Reaction of a 1:1 mixture of 34 (22.9 mg,
0.028 mmol) and GlcC5S (8.0 mg, 0.028 mmol) with HAuCl4 (756 mL,
0.025m) and NaBH4 (415 mL, 1 n) gave 9a (9.4 mg) as a brown powder.
TEM (average diameter and number of gold atoms): 1.3ꢀ0.4 nm, 79;
1H NMR (500 MHz, D2O): d=5.14 (s, 1H; 1-H), 5.05 (s, 1H; 1’-H), 4.96–
4.37 (brm, 1H; 1-H glucose), 4.09 (s, 1H; 2’-H), 4.04–3.21 (brm, 41H),
2.78 (brm; CH2S), 1.89–0.81 ppm (brm, 24H); UV/Vis (H2O,
0.1 mgmLꢁ1): surface plasmon band not observed; elemental analysis
calcd (%) for (C34H65N2O15S2)22(C11H21O6S)22Au79 (40 kDa): C 30.12, H
4.83, N 1.56, S 5.36; found: C 29.93, H 4.88, N 1.80, S 5.93.
Man5-Au-GlcC5 ACTHNUGTRNEUNG(10%) (12b): Reaction of a 1:9 mixture of 37 (5.75 mg,
0.0044 mmol) and GlcC5S (11.3 mg, 0.040 mmol) with HAuCl4 (593 mL,
0.025m) and NaBH4 (326 mL, 1 n) gave 12b (5.3 mg) as a dark-brown
powder. TEM (average diameter and number of gold atoms): 1.8ꢀ
0.3 nm, 201 (double distribution: a small percentage of the GNPs have
an average diameter of 5 nm); 1H NMR (500 MHz, D2O): d=5.37 (s,
1H), 5.17 (s, 1H), 5.08 (s, 1H), 5.06 (s, 1H), 4.85 (s, 1H), 4.42–4.33 (brm,
9H; 1-H glucose), 4.19–3.90 (brm, 124H), 2.12–1.26 ppm (brm, 72H);
UV/Vis (H2O, 0.1 mgmLꢁ1): surface plasmon band not observed; elemen-
tal analysis calcd (%) for (C52H95N2O30S2)5(C11H21O6S)43Au201 (58 kDa): C
15.14, H 2.39, N 0.24, S 2.92; found: C 14.96, H 2.80, N 0.35, S 2.93.
Man2-Au-GlcC5 ACHTUNGTRENNUNG(10%) (9b): Reaction of a 1:9 mixture of 34 (19.0 mg,
0.024 mmol) and GlcC5S (59.7 mg, 0.211 mmol) with HAuCl4 (3.14 mL,
0.025m) and NaBH4 (1.72 mL, 1 n) gave 9b (22.5 mg) as a dark-brown
powder. TEM (average diameter and number of gold atoms): 2.0ꢀ
0.5 nm, 225; 1H NMR (500 MHz, D2O): d=5.14 (brs, 1H; 1-H), 5.04 (s,
1H; 1’-H), 4.47 (d, J=8.0 Hz, 9H; 1-H glucose), 4.09 (brs, 2H; 2’-H),
4.03–3.22 (brm, 105H), 2.79 (t, J=7.5 Hz, 2H; CH2S), 1.88–1.28 ppm
(brm, 72H); UV/Vis (H2O, 0.1 mgmLꢁ1): l=527 nm (surface plasmon
Man7-Au-GlcC5 ACTHNUGTRNEUNG(50%) (13a): Reaction of a 1:1 mixture of 38 (10.49 mg,
0.0065 mmol) and GlcC5S (1.83 mg, 0.0065 mmol) with HAuCl4 (173 mL,
0.025m) and NaBH4 (95 mL, 1 n) gave 13a (6.2 mg) as a brown powder.
TEM (average diameter and number of gold atoms): 1.8ꢀ0.4 nm, 201;
1H NMR (500 MHz, D2O): d=5.37 (s, 1H), 5.32 (brs, 2H), 5.14 (s, 1H),
5.07 (s, 2H) (one anomeric signal overlapped by the solvent signal), 4.21–
3.16 (brm, 72H), 2.77 (brm; CH2S), 1.88–1.13 ppm (brm, 24H); UV/Vis
(H2O, 0.1 mgmLꢁ1): surface plasmon band not observed; elemental anal-
band);
elemental
analysis
calcd
(%)
for
(C34H65N2O15S2)9(C11H21O6S)81Au225 (74 kDa): C 19.33, H 3.10, N 0.34, S
4.27; found: C 19.31, H 3.16, N 0.69, S 4.96.
Man3-Au-GlcC5 ACHTUNGTRENNUNG(50%) (10a): Reaction of a 1:1 mixture of 35 (13.0 mg,
0.013 mmol) and GlcC5S (3.7 mg, 0.013 mmol) with HAuCl4 (358 mL,
0.025m) and NaBH4 (196 mL, 1 n) gave 10a (4.6 mg) as a brown powder.
TEM (average diameter and number of gold atoms): 1.6ꢀ0.4 nm, 140;
1H NMR (500 MHz, D2O): d=5.31 (s, 1H; 1-H), 5.12 (s, 1H; 1’-H), 5.07
(s, 1H; 1’’-H), 4.52–4.21 (brm, 1H; 1-H glucose), 4.13–3.31 (brm, 48H),
2.76 (brm; CH2S), 1.90–1.08 ppm (brm, 24H); UV/Vis (H2O,
0.1 mgmLꢁ1): surface plasmon band not observed; elemental analysis
calcd (%) for (C40H75N2O20S2)62(C11H21O6S)61Au140 (105 kDa): C 36.13, H
5.71, N 1.66, S 5.66; found: C 36.21, H 5.79, N 1.93, S 5.76.
ysis calcd (%) for (C52H95N2O30S2)58(C11H21O6S)58Au201 (150 kDa):
34.91, H 5.31, N 1.09, S 3.73; found: C 34.93, H 5.75, N 1.66, S 3.51.
C
Acknowledgements
The work was supported by the EU (grants EMPRO LSHP-CT2003-
503558 and Glycogold MRTN-CT2004-005645), the Spanish Ministry of
Education and Science (MEC, grant CTQ2008-04638), and the Fondation
Dormeur. O.M.A. thanks the CSIC (I3P program) and MEC for financial
support.
Man3-Au-GlcC5 ACHTUNGTRENNUNG(10%) (10b): Reaction of a 1:9 mixture of 35 (6.8 mg,
0.007 mmol) and GlcC5S (17.8 mg, 0.063 mmol) with HAuCl4 (756 mL,
0.025m) and NaBH4 (513 mL, 1 n) gave 10b (10.0 mg) as a dark-brown
powder. TEM (average diameter and number of gold atoms): 1.8ꢀ
0.4 nm, 201; 1H NMR (500 MHz, D2O, 25 8C): d=5.32 (s, 1H; 1-H), 5.13
(s, 1H; 1’-H), 5.06 (s, 1H; 1’’-H), 4.45 (brm, 9H; 1-H glucose), 4.19–3.23
(brm, 112H), 2.18–1.08 ppm (brm, 72H); UV/Vis (H2O, 0.1 mgmLꢁ1):
surface plasmon band not observed; elemental analysis calcd (%) for
(C40H75N2O20S2)13(C11H21O6S)122Au140 (86 kDa): C 25.85, H 4.12, N 0.42, S
5.49; found: C 25.89, H 4.26, N 0.54, S 5.92.
[1] M. Mammen, S.-K. Choi, G. M. Whitesides, Angew. Chem. 1998,
110, 2908–2953; Angew. Chem. Int. Ed. 1998, 37, 2754–2794.
[2] Y. C. Lee, R. T. Lee, Acc. Chem. Res. 1995, 28, 321–327.
[3] L. L. Kiessling, J. K. Pontrello, M. C. Schuster, in Carbohydrate-
Based Drug Discovery (Ed.: C.-H. Wong), Wiley-VCH, Weinheim,
2003, pp. 575–609.
[4] J. M. de La Fuente, A. G. Barrientos, T. C. Rojas, J. Rojo, J. CaÇada,
A. Fernꢄndez, S. Penadꢂs, Angew. Chem. 2001, 113, 2317–2321;
Angew. Chem. Int. Ed. 2001, 40, 2257–2261.
[5] A. G. Barrientos, J. M. de La Fuente, T. C. Rojas, A. Fernꢄndez, S.
Penadꢂs, Chem. Eur. J. 2003, 9, 1909–1921.
[6] J. M. de La Fuente, S. Penadꢂs, BBA-Gen. Subjects 2006, 1760, 636–
651.
Man4-Au-GlcC5 ACHTUNGTRENNUNG(50%) (11a): Reaction of a 1:1 mixture of 36 (16.9 mg,
0.015 mmol) and GlcC5S (4.23 mg, 0.015 mmol) with HAuCl4 (398 mL,
0.025m) and NaBH4 (218 mL, 1 n) gave 11a (6.0 mg) as a brown powder.
TEM (average diameter and number of gold atoms): 1.9ꢀ0.5 nm, 225;
1H NMR (500 MHz, D2O): d=5.38 (brs, 1H), 5.33 (brs, 1H), 5.08 (s,
1H), 4.86 (s, 1H), 4.47 (d, J=7.5 Hz, 1H; 1-H glucose), 4.24–3.31 (brm,
54H), 1.81–1.13 ppm (brm, 24H); UV/Vis (H2O, 0.1 mgmLꢁ1): l=
520 nm (surface plasmon band); elemental analysis calcd (%) for
(C46H85N2O25S2)56(C11H21O6S)56Au225 (123 kDa): C 31.08, H 4.85, N 1.27, S
4.37; found: C 31.23, H 5.09, N 1.94, S 4.63.
[7] J. M. de La Fuente, S. Penadꢂs, Glycoconjugate J. 2004, 21, 149–163.
9886
www.chemeurj.org
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9874 – 9888

Gold Manno-Glyconanoparticles
FULL PAPER
[8] J. Rojo, V. Dꢀaz, J. M. de La Fuente, I. Segura, A. G. Barrientos,
H. H. Riese, A. Bernad, S. Penadꢂs, ChemBioChem 2004, 5, 291–
297.
H. Song, Y. Wang, N. M. Stamatos, L.-X. Wang, Bioconjugate Chem.
2006, 17, 493–500.
[30] H. Li, L.-X. Wang, Org. Biomol. Chem. 2004, 2, 483–488.
[31] a) L. Bꢂdouet, M.-T. Bousser, N. Frison, C. Boccaccio, J.-P. Abasta-
do, P. Marceau, R. Mayer, M. L. Monsigny, A.-C. Roche, Biosci.
Rep. 2001, 21, 839–855; b) N. Frison, P. Marceau, A.-C. Roche, M.
Monsigny, R. Mayer, Biochem. J. 2002, 368, 111–119.
[9] a) T. Mizuochi, M. W. Spellman, M. Larkin, J. Solomon, L. J. Basa,
T. Feizi, Biochem. J. 1988, 254, 599–603; b) T. Feizi, M. Larkin, Gly-
cobiology 1990, 1, 17–23; c) E. Fenouillet, J. C. Gluckman, I. M.
Jones, Trends Biochem. Sci. 1994, 19, 65–70; d) H. Geyer, C.
Holschbach, G. Hunsmann, J. Schneider, J. Biol. Chem. 1988, 263,
11760–11767; e) J. Lifson, S. Coutre, E. Huang, E. Engleman, J.
Exp. Med. 1986, 164, 2101–2106; f) T. Mizuochi, M. W. Spellman,
M. Larkin, J. Solomon, L. J. Basa, T. Feizi, Biomed. Chromatogr.
1987, 2, 260–270; g) T. Mizuochi, T. J. Matthews, M. Kato, J.
Hamako, K. Titani, J. Solomon, T. Feizi, J. Biol. Chem. 1990, 265,
8519–8524.
[10] a) R. Wyatt, J. Sodroski, Science 1998, 280, 1884–1888; b) R. Wyatt,
P. D. Kwong, E. Desjardins, R. W. Sweet, J. Robinson, W. A. Hen-
drickson, J. G. Sodroski, Nature 1998, 393, 705–711.
[11] M. Zhang, B. Gaschen, W. Blay, B. Foley, N. Haigwood, C. Kuiken,
B. Korber, Glycobiology 2004, 14, 1229–1246.
[32] a) T. Ogawa, K. Kiyoaki, K. Sasajima, M. Matsui, Tetrahedron 1981,
37, 2779–2786; b) T. Ogawa, K. Sasajima, Tetrahedron 1981, 37,
2787–2792; c) T. Ogawa, K. Sasajima, Carbohydr. Res. 1981, 93, 53–
66; d) T. Ogawa, S. Nakabayashi, T. Kitajima, Carbohydr. Res. 1983,
114, 225–236; e) T. Ogawa, T. Nukada, Carbohydr. Res. 1985, 136,
135–152; f) J. R. Merritt, E. Naisang, B. Fraser-Reid, J. Org. Chem.
1994, 59, 4443–4449; g) P. Grice, S. V. Ley, J. Pietruszka, H. Priepke,
Angew. Chem. 1996, 108, 206–208; Angew. Chem. Int. Ed. Engl.
1996, 35, 197–200; h) P. Grice, S. V. Ley, J. Pietruszka, H. Osborn,
H. Priepke, S. Warriner, Chem. Eur. J. 1997, 3, 431–440; i) H.-K.
Lee, C. N. Scanlan, C.-Y. Huang, A. Y. Chang, D. A. Calarese, R. A.
Dwek, P. M. Rudd, D. R. Burton, I. A. Wilson, C.-H. Wong, Angew.
Chem. 2004, 116, 1018–1021; Angew. Chem. Int. Ed. 2004, 43, 1000–
1003.
[12] T. B. H. Geijtenbeek, R. Torensma, S. J. van Vliet, G. C. F. van
Duijnhoven, G. J. Adema, Y. van Kooyk, C. G. Figdor, Cell 2000,
100, 575–584.
[13] S. Pçhlman, F. Baribaud, R. D. Doms, Trends Immunol. 2001, 22,
643–646.
[33] For an exhaustive review about thiolate SAMs on metals, see: J. C.
Love, L. A. Estroff, J. K. Kriebel, R. G. Nuzzo, G. M. Whitesides,
Chem. Rev. 2005, 105, 1103–1169.
[14] Y. van Kooyk, T. B. H. Geijtenbeek, Nat. Rev. Immunol. 2003, 3,
697–709.
[34] M. Mrksich, G. M. Whitesides, Annu. Rev. Biophys. Biomol. Struct.
1996, 25, 55–78.
[15] B. M. Curtis, S. Scharnowske, A. J. Watson, Proc. Natl. Acad. Sci.
USA 1992, 89, 8356–8360.
[35] For information on the importance of PEG chain synthesis, see:
F. A. Loiseau, K. K. ACTHNUTRGNEUNG(Mimi) Hii, A. M. Hill, J. Org. Chem. 2004, 69,
[16] H. Feinberg, D. A. Mitchell, K. Drickamer, W. I. Weis, Science 2001,
294, 2163–2166.
[17] Y. Guo, H. Feinberg, E. Conroy, D. A. Mitchell, R. Alvarez, O.
Blixt, M. E. Taylor, W. I. Weis, K. Drickamer, Nat. Struct. Mol. Biol.
2004, 11, 591–598.
[18] K. D. McReynolds, J. Gervay-Hague, Chem. Rev. 2007, 107, 1533–
1552.
[19] C. N. Scanlan, J. Offer, N. Zitzmann, R. A. Dwek, Nature 2007, 446,
1038–1045.
639–647, and references therein.
[36] D. J. Revell, J. R. Knight, D. J. Blyth, A. H. Haines, D. A. Russell,
Langmuir 1998, 14, 4517–4524.
[37] C.-C. Lin, Y.-C. Yeh, C.-Y. Yang, C.-L. Chen, G.-F. Chen, C.-C.
Chen, Y.-C. Wu, J. Am. Chem. Soc. 2002, 124, 3508–3509.
[38] B. Fraser-Reid, U. E. Udodong, Z. Wu, H. Ottosson, J. R. Merritt,
C. S. Rao, C. Roberts, R. Madsen, Synlett 1992, 927–942.
[39] K. P. R. Kartha, H. J. Jennings, J. Carbohydr. Chem. 1990, 9, 777–
781.
[20] a) J.-E. Hansen, H. Clausen, C. Nielsen, L. S. Teglbjærg, L. L.
Hansen, C. M. Nielsen, E. Dabelsteen, L. Mathiesen, S.-I. Hako-
mori, J. O. Nielsen, J. Virol. 1990, 64, 2833–2840; b) J.-E. Hansen, C.
Nielsen, M. Arendrup, S. Olofsson, L. Mathiesen, J. O. Nielsen, H.
Clausen, J. Virol. 1991, 65, 6461–6467; c) J.-E. S. Hansen, C. M.
Nielsen, C. Nielsen, P. Heegaard, L. R. Mathiesen, J. O. Nielsen,
AIDS 1989, 3, 635–642; d) A. Pashov, M. Perry, M. Dyar, M. Chow,
T. Kieber-Emmons, Curr. Pharm. Des. 2007, 13, 185–201.
[21] D. A. Calarese, H.-K. Lee, C.-Y. Huang, M. D. Best, R. D. Astrono-
mo, R. L. Stanfield, H. Katinger, D. R. Burton, C.-H. Wong, I. A.
Wilson, Proc. Natl. Acad. Sci. USA 2005, 102, 13371–13377.
[22] E. W. Adams, D. M. Ratner, P. H. Seeberger, N. Hacohen, ChemBio-
Chem 2008, 9, 294–303.
[40] C. Pale-Grosdemange, E. S. Simon, K. L. Prime, G. M. Whitesides, J.
Am. Chem. Soc. 1991, 113, 12–20.
[41] A. Garcꢀa-Barrientos, J. J. Garcꢀa-Lopꢂz, J. Isac-Garcꢀa, F. Ortega-
Caballero, C. Uriel, A. Bargas Berenguer, F. Santoyo-Gonzꢄlez,
Synthesis 2001, 1057–1064.
[42] G. Zemplꢂn, Ber. Dtsch. Chem. Ges. 1927, 60, 1555–1564.
[43] T. Buskas, E. Sçderberg, P. Konradsson, B. Fraser-Reid, J. Org.
Chem. 2000, 65, 958–963.
[44] We followed the experimental procedure used for similar substrates
by B. Frisch et al., see: B. Frisch, C. Boeckler, F. Schuber, Bioconju-
gate Chem. 1996, 7, 180–186.
[45] E. Arce, P. M. Nieto, V. Dꢀaz, R. Garcꢀa Castro, A. Bernard, J. Rojo,
Bioconjugate Chem. 2003, 14, 817–823.
[23] J. Wang, H. Li, G. Zou L.-X. Wang, Org. Biomol. Chem. 2007, 5,
1529–1540.
[24] N. Frison, M. E. Taylor, E. Soilleux, M.-T. Bousser, R. Mayer, M.
Monsigny, K. Drickamer, A.-C. Roche, J. Biol. Chem. 2003, 278,
23922–23929.
[46] For other published routes to compound 27, see: a) T. K. Lindhorst,
S. Kçtter, U. Krallmann-Wenzel, S. Ehlers, J. Chem. Soc. Perkin
Trans. 1 2001, 823–831; b) L. E. Strong, L. L. Kiessling, J. Am.
Chem. Soc. 1999, 121, 6193–6196; c) N. Jayaraman, J. F. Stoddart,
Tetrahedron Lett. 1997, 38, 6767–6770.
[25] S.-K. Wang, P.-L. Liang, R. D. Astronomo, T.-L. Hsu, S.-L. Hsieh,
D. R. Burton, C.-H. Wong, Proc. Natl. Acad. Sci. USA 2008, 105,
3690–3695.
[26] G. Tabarani, J. J. Reina, C. Ebel, C. Vives, H. Lortat-Jacob, J. Rojo,
F. Fieschi, FEBS Lett. 2006, 580, 2402–2408.
[47] In the Wong protocol (ref. [32i]), the oligosaccharides were obtained
as aminopentyl (instead of aminoethyl) derivatives.
[48] G. V. S. Reddy, G. V. Rao, R. V. K. Subramanyam, D. S. Iyengar,
Synth. Commun. 2000, 30, 2233–2237.
[49] A. Koziara, J. Chem. Res. 1989, 296–297.
[27] R. Ojeda, J. L. de Paz, A. G. Barrientos, M. Martꢀn-Lomas, S. Pe-
nadꢂs, Carbohydr. Res. 2007, 342, 448–459.
[28] J. G. Joyce, I. J. Krauss, H. C. Song, D. W. Opalka, K. M. Grimm,
D. D. Nahas, M. T. Esser, R. Hrin, M. Feng, V. Y. Dudkin, M. Chas-
tain, J. W. Shiver, S. J. Danishefsky, Proc. Natl. Acad. Sci. USA 2008,
105, 15684–15689.
[50] M. J. Hostetler, J. E. Wingate, C.-J. Zhong, J. E. Harris, R. W.
Vachet, M. R. Clark, J. D. Londono, S. J. Green, J. J. Stokes, G. D.
Wignall, G. L. Glish, M. D. Porter, N. D. Evans, R. W. Murray, Lang-
muir 1998, 14, 17–30.
[51] H. Feinberg, R. Castelli, K. Drickramer, P. H. Seeberger, W. I. Weis,
J. Biol. Chem. 2006, 282, 4202–4209.
[29] a) V. Y. Dudkin, M. Orlova, X. Geng, M. Mandal, W. C. Olson, S. J.
Danishefsky, J. Am. Chem. Soc. 2004, 126, 9560–9562; b) L.-X.
Wang, J. Ni, S. Singh, H. Li, Chem. Biol. 2004, 11, 127–134; c) J. Ni,
[52] A. Alfsen, M. Bomsel, J. Biol. Chem. 2002, 277, 25649–25659.
[53] X. Wang, P. Sun, A. Al-Qamari, T. Tai, I. Kawashima, A. S. Paller, J.
Biol. Chem. 2001, 276, 8436–8444.
Chem. Eur. J. 2009, 15, 9874 – 9888
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9887

S. Penadꢂs et al.
[54] a) M. Moulard, H. Lortat-Jacob, I. Mondor, G. Roca, R. Wyatt, J.
Sodroski, L. Zhao, W. Olson, P. D. Kwong, Q. J. Sattentau, J. Virol.
2000, 74, 1948–1960; b) L. N. Callahan, M. Phelan, M. Mallinson,
M. A. Norcross, J. Virol. 1991, 65, 1543–1550; c) A. R. Neurath, N.
Strick, Y. Y. Li, BMC Infect. Dis. 2002, 2, 27.
[55] A. Cambi, C. G. Figdor, Curr. Opin. Immunol. 2005, 17, 345–351.
[56] For highlights on the design of HIV vaccines, see: a) L.-X. Wang, in
Carbohydrate Drug Design (Eds.: A. A. Klyosov, Z. J. Witczak, D.
Platt), Oxford University Press, Oxford, 2006, pp. 133–160; b) D.
Calarese, C. Escalan, H.-K. Lee, P. Rudd, C.-H. Wong, R. Dwek, D.
Burton, I. A. Wilson, in Carbohydrate Drug Design (Eds.: A. A.
Klyosov, Z. J. Witczak, D. Platt), Oxford University Press, Oxford,
2006, pp. 161–185, and references therein.
[57] D. A. Mitchell, A. J. Fadden, K. Drickamer, J. Biol. Chem. 2001,
276, 28939–28945.
[58] P. Y. Lozach, H. Lortat-Jacob, L. A. de Lacroix, I. Staropoli, S.
Foung, A. Amara, C. Houles, F. Fieschi, O. Schwartz, J. L. Virelizier,
F. Arenzana-Seisdedos, R. Altmeyer, J. Biol. Chem. 2003, 278,
20358–20366.
[59] BIAtechnology Handbook, BIACORE AB, Uppsala, 1998.
Received: April 7, 2009
Published online: August 13, 2009
9888
www.chemeurj.org
ꢆ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9874 – 9888