Synthesis and immunochemical studies on a Candida albicans cluster glycoconjugate vaccine

Article abstract of DOI:10.1039/b709912f

The immunoprotective β-mannan of Candida albicans occurs as part of the cell wall phosphomannan N-linked glycoprotein. This macromolecule is composed of an extended α1,6 linked mannopyranan backbone containing α1,2 mannopyranan branches, to which β1,2 man

Full text of DOI:10.1039/b709912f

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Synthesis and immunochemical studies on a Candida albicans cluster
glycoconjugate vaccine
Xiangyang Wu,a Tomasz Lipinski,a Fre´de´ric R. Carrel,b J. James Baileya and David R. Bundlea
Received 29th June 2007, Accepted 3rd September 2007
First published as an Advance Article on the web 17th September 2007
DOI: 10.1039/b709912f
The immunoprotective b-mannan of Candida albicans occurs as part of the cell wall phosphomannan
N-linked glycoprotein. This macromolecule is composed of an extended a1,6 linked mannopyranan
backbone containing a1,2 mannopyranan branches, to which b1,2 mannopyranan epitopes are
attached. The synthesis of b1,2-mannan disaccharides clustered on a glucose core has been achieved as
a way to imitate the multipoint display of b-mannans in the native glycoprotein. The clustered epitopes
were conjugated to tetanus toxoid and bovine serum albumin. Rabbits immunized with tetanus toxoid
cluster glycoconjugate gave good antibody titres for the disaccharide cluster or simple trisaccharide
epitope (coupled to BSA). The anti-sera also showed strong cross-reactivity with a Candida albicans
b-mannan cell wall extract. These immunochemical results are compared with data obtained with
non-cluster disaccharide and trisaccharide glycoconjugate antigens. The same conjugates gave
substantially lower antibody levels when used to immunize mice.
Introduction
Candida albicans, the most common etiologic agent of can-
didiasis, commonly affects immunocompromised patients and
those undergoing long-term antibiotic treatment. The potential
of immunotherapy for this fungal infection is attracting attention
because mortality rates remain relatively high and antifungal
agents with excellent in vitro activity have significant toxicity
issues.1
The major antigenic complex of the Candida albican’s cell
wall is a N-linked glycoprotein. The structure of the cell surface
phosphomannoprotein complex (PMPC) has been proposed by
Suzuki and co-workers.2,3 Differential fractionation of the PMPC
by mild acid hydrolysis of a phosphate bond releases a b1,2-
Fig. 1 Composite structure of the C. albicans phosphomannoprotein
mannan component, an epitope that also occurs linked via a
deduced by Suzuki and co-workers.2,3 2,3 b-Mannan epitopes are linked
glycosidic linkage to the a1,2-mannan chains that branch from
the a1,6-mannan main chain (Fig. 1).
to the a-mannan side chains either via glycosidic bonds or via a
phosphodiester.
Mouse experiments show that a vaccine approach can heighten
host resistance against hematogenously disseminated disease
and against vaginal infection.4–6 Both active immunization and
passive protection by monoclonal antibody to the cell wall b-
mannan afford protection. Short b1,2-mannan homo-oligomers
from disaccharide up to hexasaccharide were shown to inhibit
the binding of two protective, b1,2-mannan specific monoclonal
antibodies reported by the group of Cutler.7 In sharp contrast to
a decades old paradigm first reported by Kabat8,9 we observed
that disaccharide and trisaccharide epitopes were the most potent
inhibitors of these monoclonal antibodies, whereas tetra, penta
and hexsaccharides exhibited rapidly diminishing activities.7 The
paradigm repeatedly observed by others for oligomeric haptens
anticipates a steady increase in inhibitor power with size until an
antigenic determinant is reached that approximates the dimen-
sions and complimentary requirements of an antibody binding
site. In addition to speculation concerning antigen conformation
and the size of the binding site of these monoclonal antibodies,7
we conjectured that a small and readily synthetically accessible
oligosaccharide epitope when incorporated into a glycoconjugate
vaccine might be sufficient to raise antibodies that could confer
protection against Candida albicans infections.10 Subsequent and
yet to be published research (Bundle et al. unpublished data) has
shown that a synthetic conjugate vaccine based on a trisaccharide
epitope conjugated to tetanus toxoid gave high titre anti-sera in
rabbits, but mice responded to the same immunogen with only
modest antibody levels.
aAlberta Ingenuity Centre for Carbohydrate science, Department of Chem-
istry, University of Alberta, Edmonton, AB T6G 2G2, Canada. E-mail:
dave.bundle@ualberta.ca; Tel: +1 780-492-8808
In their approach to generating antigens that simulate multiple
epitope presentation in glycoproteins, work from Danishefsky’s
group has demonstrated that the immune system tends to
recognize clustered motifs of the sialyl T antigen.11 Clustering
bLaboratory of Organic Chemistry, Swiss Federal Institute of Technology
(ETH), Wolfgang-Pauli-Strasse 10, 8093 Zu¨rich, Switzerland. E-mail:
fcarrel@org.chem.ethz.ch
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 3477–3485 | 3477
©

View Article Online
Scheme 1 A disaccharide tether glycoside possessing a terminal thiol and a tetra-epoxypropyl glucopyranoside are the envisaged intermediates required
to generate a clustered antigen.
of oligosaccharide epitopes has been reported for the Vibrio
cholera antigen but without immunological data.12 Clustering of
an oligomannose epitope found on HIV-1 gp120 gave a conjugate
that was reactive with a defined antibody but immunization with
cluster glycoconjugates induced antibodies the majority of which
bound the linker rather than carbohydrate.13 In the search for a
presentation of the C. albicans b-mannan antigenic determinant
that is as immunogenic in mice as it is in rabbits, we have
investigated clustering of oligosaccharide motifs.
Results
We elected to cluster b1,2 linked disaccharide epitopes on a
glucopyranoside scaffold derivatized as a triethyleneglycol glu-
copyranoside (Scheme 1). An azide group would provide a latent
amine for eventual conjugation to protein. Per-allylation of a
tether glucopyranoside followed by epoxidation of the tetra-allyl
ether would allow for efficient coupling with a thiol terminated
mannobiose pentanyl glycoside.
Triethylene glycol glucopyranoside 114 was converted to the
tetra-allyl ether 2 under standard conditions (Scheme 2). Epox-
idation of 2 with meta-chloroperbenzoic acid (m-CPBA) gave the
important tetra-epoxy derivative 3, which was used later to create
the cluster antigen.
The synthesis of disaccharide 10 was accomplished as out-
lined in Scheme 3. The synthesis is based on a reported
procedure10 employing an oxidation–reduction strategy to convert
Scheme 2 Synthesis of epoxide 3.
◦
b-glucopyranoside to b-mannopyranosides. The glycosyl acceptor
415 and glycosyl donor 516 were readily synthesized according
to literature methods. The glycosylation reaction between the
selectively benzylated acceptor 4 and trichloroacetimidate glycosyl
donor 5 was performed by activation with trimethylsilyl trifluo-
romethanesulfonate (TMSOTf) (0.05 equivalents) in CH2Cl2 at
−10 ◦C, to give the required disaccharide 6 in excellent yield.
Deacetylation gave the desired alcohol 7 quantitatively, and
oxidation by DMSO and Ac2O (2 : 1) gave the corresponding
ketone, which was not characterized but immediately reduced with
L-selectride at −78 C in THF to afford the target disaccharide
8 in 80% yield. It should be noted that the pentenyl glycoside
was employed instead of an allyl group because it is stable under
◦
Birch reduction conditions at −78 C. The smooth reduction of
per-benzylated disaccharide 8 and subsequent acetylation to the
peracetate 9 was achieved in 60% yield with sodium in ammonia
without affecting the pentenyl double bond. For the attachment
of disaccharide to the core glucose 3, a terminal thioacetate was
chosen as a versatile functionality from which the cluster could be
readily generated. The protected oligosaccharide 9 was elaborated
3478 | Org. Biomol. Chem., 2007, 5, 3477–3485
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Scheme 3 Synthesis of disaccharide 10.
via photoaddition of thioacetic acid17 to the pentenyl glycoside to
give the key product 10 in 87% yield.
room temperature for 5 h, affording the corresponding half ester
14 in good yield after washing with dichloromethane to remove
excess linker reagent, followed by purification on a reversed-
phase column (Scheme 5). Coupling of half ester 14 to BSA was
performed over 48 h in buffer (pH = 7.5) at ambient temperature.
The BSA conjugate 15 was obtained as a white powder after
dialysis against deionized water followed by lyophilization. In
the same way, half ester 14 was conjugated to tetanus toxoid in
phosphate buffer (pH = 7.2) overnight at ambient temperature.
After dialysis against phosphate buffered saline (PBS) pH = 7.2, a
solution of conjugate 16 was obtained for use as a vaccine. In order
to compare immunological properties of simple and clustered
forms of disaccharide hapten we generated a conjugate 17 with a
lower incorporation of cluster hapten yielding a similar number of
disaccharide haptens to those present in conjugate 19 (Scheme 6).
Lower incorporation was obtained by decreasing the equivalents
of half ester in the reaction mixture. The molar ratio of protein
to activated half ester 14 and observed hapten incorporation are
tabulated (Table 1).
The coupling reaction of thioacetate 10 with epoxide 3 was
smoothly achieved in the presence of potassium carbonate
(K2CO3) in a mixture of methanol and water under argon in
60% yield to provide the mannobiose-cluster 11 (Scheme 4),
which was purified by HPLC and characterized by MALDI-
TOF. Subsequent reduction of the azide group of the cluster
epitope 11 with hydrogen sulfide (H2S) under basic conditions
gave pure amine-cluster 12 after HPLC. The structure of 12
was confirmed by MS spectrometry and 1H NMR spectra which
showed the correct number and identity of anomeric hydrogen
atoms.
Previously, we have demonstrated that coupling of a b1,2-
mannose trisaccharide to bovine serum albumin (BSA) or
tetanus toxoid (TT) could be achieved via a homobifunctional
p-nitrophenyl ester of adipic acid with incorporation rates of ∼20
haptens per mole of BSA and 8–12 moles of trisaccharide per
mole of tetanus toxoid.10,18 Comparable incorporation levels for
conjugation 12 to protein would achieve a 4 fold higher level of
disaccharide hapten substitution. A potentially helpful outcome
that could enhance the antigenicity of the resulting glycoconjugate.
The cluster hapten as its amine derivative 12 was treated with
5 equiv. of the adipate p-nitrophenyl diester 13 in dry DMF at
The degree of incorporation of the cluster hapten on BSA or
tetanus toxoid was established by MALDI-TOF mass spectrom-
etry using sinapinic acid as matrix. Conjugation efficiencies of
between 18 and 23% were achieved. Glycoconjugates of this type
are heterogeneous with respect to the precise number of ligands
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 3477–3485 | 3479
©

View Article Online
Table 1 BSA and tetanus toxoid conjugates
Molar ratio of
protein–half ester
Hapten
incorporated
Incorporation
efficiency (%)
Conjugate
15 (BSA-cluster)
16 (TT-cluster)
17 (TT-cluster)
19 (TT-disac)
21 (TT-trisac)
1 : 24
1 : 20
1 : 10
1 : 20
1 : 20
5.4
4.6
1.8
7
23
23
18
35
40
7.9
Scheme 4 Synthesis of disaccharide cluster 12.
Scheme 6 Structure of disaccharide 18, 19 and trisaccharide conjugates
20, 21.
attached to protein and with regard to the specific lysine residues
that are substituted. The non-integer number n refers to the average
degree of hapten substitution. Unfortunately, the incorporation
rates for cluster hapten 12 were 3–4 fold lower than those we have
observed for the direct coupling of oligosaccharides to proteins
by the same procedure.10 Despite clustering of the epitopes on a
common linker scaffold, the lower degree of conjugation resulted
in effective hapten loading on both BSA and tetanus toxoid that
were similar to those observed for direct coupling of an activated
non-clustered oligosaccharide.
Immunization studies
To place the results of this study in context we draw comparison
with the immune response observed in rabbits in response to
non-clustered disaccharide and trisaccharide epitopes conjugated
to tetanus toxoid conjugate, glycoconjugates 19 and 21. The
data for 19 were collected as part of this study but data for
trisaccharide conjugate 21 are taken from separate studies (Bundle
et al. unpublished data) that involved immunization of 24 rabbits.
Sera from rabbits vaccinated with the two cluster tetanus toxoid
conjugates 16 and 17, and the disaccharide conjugate 19 were
Scheme 5 Synthesis of cluster conjugates 15, 16 and 17.
3480 | Org. Biomol. Chem., 2007, 5, 3477–3485
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
analyzed in ELISA against the cluster BSA conjugate 15, the
disaccharide and trisaccharide BSA conjugates 18 and 20 (Fig. 2).
As well, the sera were titred against the C. albicans native cell wall
antigen. The reactivity pattern for sera obtained by immunization
with either of the cluster glycoconjugates 16 or 17 was similar to
sera from rabbits immunized with disaccharide glycoconjugate
19. All conjugates elicited a strong immune response to the
immunizing antigen. Sera from the group immunized with 19 rec-
ognized the clustered disaccharide antigen, while rabbit antibodies
from the group immunized with either cluster glycoconjugates
16 or 17 also showed strong recognition of the disaccharide
epitope 18. Surprisingly, sera from rabbits immunized with tetanus
toxoid disaccharide conjugate 19 gave titres with the cluster
BSA conjugate 15 that were higher than those obtained with
the homologous disaccharide-BSA antigen 18. We infer that the
clustered form of the disaccharide epitope performs better in
ELISA than disaccharide conjugates due to the higher density
of oligosaccharides on the protein coating the ELISA plate. In
general, both forms of the cluster disaccharide antigen 16 and 17
as well as the disaccharide antigen 19 developed antibodies specific
for the b-mannan disaccharide. For these conjugates, we also
observed that all sera showed less reactivity with the C. albicans
cell wall antigen i.e. about 10% of the activity observed for the
corresponding synthetic BSA glycoconjugates.
In response to the trisaccharide glycoconjugate vaccine 21,
consistently high titre sera were obtained as measured by a solid
phase binding assay employing the trisaccharide conjugated to the
heterologous protein BSA 20. Mean titres against the immunizing
trisaccharide epitope were 1 : 350 000 but 2–3 fold lower against
cluster 15 and non cluster disaccharide epitopes 18 (Fig. 2).
This is consistent with preferred recognition of the immunizing
epitope. The ability of antibodies induced by any of the synthetic
glycoconjugates to recognize and bind to the C. albicans cell wall
b-mannan is of crucial functional significance. To quantitate this,
we employed a cell wall extract from C. albicans to coat ELISA
plates. When this was done, titres against the cell wall antigen were
approximately 10 to 20% of those against the trisaccharide-BSA
conjugate (Fig. 2).
In contrast to antibody responses in rabbits, vaccination of mice
with conjugates 16, 19 and 21 showed significant differences in
immunogenicity of these glycoconjugates (Fig. 3A–C). Glycocon-
jugate 19 carrying disaccharide directly coupled to TT induced the
highest antibody titre and all animals in the group responded to the
vaccination. Lower titres were observed when sera were analysed
against the cluster form of disaccharide 15 and even lower against
the trisaccharide antigen 20 but an almost negligible reaction with
Candida cell wall preparation (Fig. 3A). Conjugate 16 induced
high response to homologus antigen 15 but only in a few animals
in the group. Comparing the response to the two vaccines 16 and
19, the cross-reactivity of sera to 16 with trisaccharide 20 was lower
compared to sera raised to disaccharide conjugate 19 (Fig. 3B).
Surprisingly, glycoconjugate 21 was shown to be less active as an
immunogen, and only 5 mice out of 10 responded to the vaccine.
Two mice showed high titres (Fig. 3C). Only one of these high
titre sera was able to recognize cell wall. No reactivity towards
disaccharide glycoconjugates was observed.
Fig. 2 Serum antibody titres of rabbits immunized with cluster conjugates
16 and 17 with different degrees of substitution (17 n = 1.8, and for
16 n = 4.6) and with disaccharide-tetanus toxoid conjugate 19 and
trisaccharide-tetanus toxoid conjugate 21. Sera were analysed by ELISA
on plates coated with 18 (light grey bar), 15 (black bar), 20 (white bar)
and Candida albicans cell wall extract (dark grey bar). Titres are recorded
as the geometric mean of the antibody levels for groups of 3 rabbits. Error
bars indicate the standard deviation.
Discussion
Extensive studies of the immunochemistry of synthetic com-
pounds related to cancer Tn antigens revealed the existence of
Fig. 3 Graphs show results of ELISA mice sera titration against different antigens. Groups of 10 CD1 mice were vaccinated with disaccharide-TT
conjugate 19 (A), cluster disaccharide-TT conjugate 16 (B) and trisaccharide-TT conjugate 21 (C). For each group of 10 mice, the first 5 mice in each
group were given antigens with alum as adjuvant and the following 5, with Freund’s adjuvant. Sera were collected 10 days after the final immunization
and screened in ELISA against different BSA conjugates 15 (cluster), 18 (disaccharide), 20 (trisaccharide) and Candida albicans cell wall extract as shown
on each panel. For mouse 6 in A, the ELISA titre against the cluster antigen 15 is high and off scale.
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 3477–3485 | 3481
©

View Article Online
different presentation forms of the same sugar epitope. It was
proposed that Tn antigens can be found as clusters of 3–4 conse-
cutive Tn monosaccharides.19 Lewisy antigen can also be displayed
as glycolipid or mucin glycoprotein.20 An antibody able to distin-
guish cluster and single forms of antigens has been described.21
These observations justify the precedence for synthesis of cluster
forms of tumor associated carbohydrate antigens which have
been shown to improve the potency of experimental therapeutic
vaccines.22–25
Based on these reports and our observations we reasoned that
presentation of disaccharide units in a clustered form might re-
semble more adequately the multivalent organization of b-mannan
antigenic determinants in the fungal cell wall mannoprotein. The
Candida albicans cell wall mannoprotein is a complex structure,
composed of two moieties, acid-labile and acid-stable. The exact
organization of the C. albicans cell wall phosphomannoprotein is
not known in fine detail since the structural analysis of Candida
cell walls is inherently difficult and capable of yielding only an
aggregate structure. Nevertheless, a well supported composite
structure has been established (Fig. 1).2,3 This proposes that
multiple b-mannan epitopes are to be found on a single glycan
chain of the Candida phosphomannan. Therefore, synthesis of
cluster glycoconjugates can be a relatively simple way to address
the issue of a defined vaccine that simulates the cell wall antigen.
We reasoned that attachment of b-mannan epitopes to a glucose
core via 9 atom long tethers would fulfill two objectives. First
it would increase the number of b-mannan epitopes displayed
on the conjugate vaccine and second permit their diverse spatial
arrangement. In fact coupling efficiency dropped for the clustered
hapten so that the number of b-mannan units was virtually the
same in both cluster and non-cluster glycoconjugates.
It is not clear why the analyzed sera exhibit approximately 10
fold lower titres with the cell wall. Since the induced antibody
appears to be specific for the b-mannan disaccharide with no
difference if presented as simple disaccharide or cluster antigen,
similar activity would be expected for the cell wall. In general, mice
responded to immunization by producing antibodies that were
more specific for the immunizing antigen and exhibited less cross
reactivity to closely related epitopes than rabbit sera. However,
antibody responses were less consistent among a given group of
animals. It seems unlikely that the rabbit response is directed
in part to the tether, since if this were the case, stronger cross-
reactivity between disaccharide and trisaccharide haptens would
be expected, since all glycoconjugates utilize the same tether.
However, there remains a pressing need to develop a vaccine
construct that is equally immunogenic in mice.
Experimental procedures
General methods
1H NMR spectra were recorded at either 400, 500, or 600 MHz,
and are referenced to the residual protonated solvent peaks; dH
7.24 ppm for solutions in CDCl3, and 0.1% external acetone
(dH 2.225) for solutions in D2O. Mass analysis was performed
by positive-mode electrospray ionization on a hybrid sector-TOF
mass spectrometer and for protein glycoconjugates by MALDI
mass analysis, employing 2,5-dihydroxybenzoic acid (DHB) as
matrix. Analytical thin-layer chromatography (TLC) was per-
formed on silica gel 60-F254 (Merck). TLC detection was achieved
by charring with 5% sulfuric acid in ethanol. All commercial
reagents were used as supplied. Column chromatography used
˚
silica gel (SiliCycle, Quebec City, Quebec, 230–400 mesh, 60 A),
and redistilled solvents. HPLC separations were performed on
a Beckmann C18 semipreparative reversed-phase column with a
combination of methanol and water containing 0.1% HOAC as
eluents. Photoadditions were carried out using a spectroline model
ENF-260 C UV lamp and cylindrical quartz vessels.
(2-[2-(2-Azidoethoxy)ethoxy]ethyl)-2,3,4,6-tetra-O-allyl-b-D-
glucopyranoside (2)
To a stirred solution of compound 1 (674 mg, 2 mmol) in THF
(20 mL) was added allyl bromide (1.69 mL, 20 mmol) and NaH
(560 mg of NaH 60%,14 mmol) by small portions. The reaction
was then refluxed for 90 minutes and subsequently cooled to
room temperature. Water (5 mL) was added dropwise to destroy
excess NaH. The translucent solution was extracted with ethyl
acetate (1 × 100 mL, 3 × 30 mL). The combined organic layers
were dried (MgSO4) and the organic phase was evaporated. The
residue (1.256 g) was chromatographed on silica gel (acetone–
hexane, 1 : 4; Rf = 0.32) to give 2 (648 mg, 65% from 1) as
1
a colorless oil. H-NMR dH (C6D6, 500 MHz): 6.0–5.8 (4H, m,
=
=
OCH2CH CH2), 5.35–5.21 (4H, m, OCH2CH CH2), 5.07–5.03
=
=
(4H, m, OCH2CH CH2), 4.52–4.48 (2H, m, OCH2CH CH2),
3
=
4.36–4.30 (2H, m, OCH2CH CH2), 4.29 (1H, d, J1,2 = 7.7 Hz,
2
=
H1a), 4.23 (1H, m, J = 12.9 Hz, OCH2CH CH2), 4.12 (1H,
2
3
=
dd, J = 12.8 Hz, J = 5.4 Hz, OCH2CH CH2), 3.93–3.87 (3H,
ꢀ
=
m, OCH2CH2O, OCH2CH CH2), 3.63–3.59 (3H, m, H6a, H6a ,
OCH2CH2O), 3.53–3.44 (4H, m, H3a, H4a, OCH2CH2O), 3.43–
3.37 (4H, m, OCH2CH2O, H2a, OCH2CH2O), 3.33–3.27 (3H,
m, OCH2CH2O, H5a), 3.175 (2H, dd, OCH2CH2N3), 2.77 (2H,
dd, OCH2CH2N3); 13C-NMR dC (C6D6, 125 MHz): 136.3, 136.3,
135.9, 135.5, 116.2, 115.8, 115.8, 115.6, 104.1, 84.8, 82.3, 77.9,
75.4, 74.4, 73.7, 73.5, 72.5, 70.9, 70.9, 70.8, 70.2, 69.6, 69.0, 50.7.
Low resolution mass: M + Na+ 520.3. C24H39N3O8Na: requires
520.3.
Conclusions
Clustering of four b-mannan disaccharide epitopes on a glucose
scaffold that is covalently attached to tetanus toxoid yields
glycoconjugates 16 and 17 that are highly immunogenic in rabbits
but significantly less immunogenic in mice. Antibodies raised in
response to these antigens in rabbits bind the C. albicans wall b-
mannan. However, sera from rabbits immunized with the more
easily prepared disaccharide and trisaccharide conjugates 19 and
21 exhibited higher antibody levels against synthetic and native
antigen. As the conjugate of the trisaccharide epitope is simpler to
prepare and yields sera with high titres for the immunizing epitope
and especially the cell wall b-mannan, a glycoconjugate vaccine
based on this trisaccharide glycoconjugate is being developed.
(2-[2-(2-Azidoethoxy)ethoxy]ethyl)-2,3,4,6-tetra-O-(2,3-
epoxypropyl)-b-D-glucopyranoside (3)
To a stirred solution of 2 (208 mg, 0.42 mmol) in CH2Cl2
(2.1 mL) was added, at room temperature, mCPBA (618 mg of
77%, 2.51 mmol, 6 eq) by small portions over 10 minutes. The
3482 | Org. Biomol. Chem., 2007, 5, 3477–3485
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
reaction was stirred at room temperature for 75 minutes during
which time metachlorobenzoic acid precipitated as a white solid.
Dichloromethane (1 mL) was added and the reaction was refluxed
for 105 min. The reaction mixture was diluted with CH2Cl2 (3 mL)
and filtered. The white solid was washed 3 times with CH2Cl2
(3 mL). The combined filtrates were neutralized by extraction with
a saturated solution of NaHCO3 (5 mL). The aqueous phase was
extracted 3 times with CH2Cl2 (10 mL). The combined organic
layers were dried (MgSO4) and then evaporated. The residue was
chromatographed on silica gel (acetone–hexane, 1 : 1.5; Rf =
(758 mg, 0.76 mmol) in dichloromethane (5 mL), methanol (5 mL)
was added sodium methoxide (5 mg), and the solution was
stirred overnight at room temperature. The resulting mixture was
neutralized with IR 120 (H+-form), and concentrated in vacuum.
Column chromatography in n-hexane–ethyl acetate (4 : 1) gave
the disaccharide 7 (722 mg, 100%). [a]D −35.7◦ (c 1.0, CHCl3);
1H-NMR dH (CDCl3, 600 MHz): 7.2–7.42 (30H, m, Ar), 5.92–
=
=
5.85 (1H, m, CH2CH CH2), 5.07–5.11 (2H, m, CH2CH CHa,
=
CH2Ph), 5.0–5.03 (1H, m, CH2CH CHb), 4.90–4.98 (3H, m, 3/2
2
3
CH2Ph), 4.83 (1H, d, J = 11.4 Hz, CH2Ph), 4.73 (1H, d, J =
7.8 Hz, H1b), 4.66 (1H, d, 2J = 12.0 Hz, CH2Ph), 4.49–4.61 (6H,
m, 3 × CH2Ph), 4.41 (1H, s, H1a), 4.28 (1H, d, 3J = 3.0 Hz, H2a),
1
0.28) to give 3 (168 mg, 72%) as a colorless oil. H-NMR dH
3
(CDCl3, 500 MHz): 4.25 (1H, d, J1,2 = 7.7 Hz, H1a), 4.18–
4.04 (1H, m, OCH2C), 4.04–3.92 (3H, m, H6a, H6aꢀ, OCH2C),
3.92–3.80 (1H, m, OCH2C), 3.80–3.60 (13H, m, OCH2CH2N3, 8
H(OCH2CH2O), H5a, 2 × OCH2C), 3.60–3.52 (1H, m, OCH2C),
3.52–3.44 (1H, m, OCH2C), 3.40–3.30 (5H, m, OCH2CH2N3, H3a,
3.96–4.0 (1H, m, OCHH), 3.93 (1H, t, J = 9.6 Hz, H4a), 3.75–
3
3.82 (4H, m, H2b, H6b, H6a, H6ꢀa), 3.68–3.71 (2H, m, H3b, H6ꢀb),
3.49–3.62 (4H, m, H5b, H3a, H4b, OCHH), 3.43 (1H, m, H5a),
2.20 (2H, m, CH2CH CH2), 1.78 (2H, m, OCH2CH2); 13C-NMR
=
H4a, OCH2C), 3.22–3.10 (5H, m, 4 × CH CH2, H2a), 2.82–
dC (125 MHz, CDCl3), 139.1–115.1, 104.1 (1JC-H = 162 Hz, C1b),
100.5 (1JC-H = 156 Hz, C1a), 85.2, 80.3, 76.8, 75.7, 75.4, 75.3, 75.1,
74.8, 74.7, 73.4, 70.0, 69.8, 69.3, 69.2; EMS; M + Na+ 973.44977:
C59H66O11Na+ requires 973.44974.
=
2.74 (4H, m, OCH2C), 2.62–2.54 (4H, m, OCH2C); 13C-NMR dC
(CDCl3, 125 MHz): 103.4, 103.3, 84.9, 84.5, 82.7, 82.5, 78.1, 77.5,
75.0, 71.9, 70.7, 69.8, 68.9, 50.9, 50.5, 44.6, 44.1. Low resolution
mass: M + Na+ 584.2. C24H39N3O12Na: requires 584.2.
Pentenyl 2-O-(3,4,6-tri-O-benzyl-b-D-mannopyranosyl)-3,4,6-
tri-O-benzyl-b-D-mannopyranoside (8)
Pentenyl 2-O-(3,4,6-tri-O-benzyl-2-O-acetyl-b-D-glucopyranosyl)-
3,4,6-tri-O-benzyl-b-D-mannopyranoside (6)
The procedure used was analogous to the preparation of allyl
2-O-(3,4,6-tri-O-benzyl-b-D-mannopyranosyl)-3,4,6-tri-O-benzyl-
b-D-mannopyranoside.10 Disaccharide 7 (630 mg, 0.66 mmol)
was dissolved in freshly distilled dimethyl sulfoxide (10 mL)
and acetic anhydride (5 mL) was added. The resulting solution
was stirred for 18 h at room temperature, and diluted with ethyl
acetate, then washed with water, sodium bicarbonate solution and
a brine solution. Finally, the solution was concentrated at low
pressure to give a yellow syrup. This syrup was dissolved in THF
(10 mL) and then cooled to −78 ◦C under argon. L-selectride
(1 M THF, 2 mL) was added dropwise and the reaction was
stirred for 5 min. The dry ice bath was removed and the reaction
was allowed to warm to room temperature. The reaction mixture
was quenched after 15 min with methanol (2 mL), and diluted
with dichloromethane. Washing with a solution of hydrogen
peroxide (5%) and sodium hydroxide (1 M) followed by sodium
thiosulfate (5%) and sodium chloride solutions gave a clear
colourless organic solution. The resulting solution was dried
over magnesium sulfate and concentrated to a colourless oil.
Column chromatography in n-hexane–ethyl acetate (5 : 2) gave
the disaccharide 8 (504 mg, 80%). [a]D −59.4◦ (c 1.0, CHCl3);
1H-NMR dH (CDCl3, 600 MHz): 7.19–7.42 (30H, m, Ar), 5.92–
The procedure used was analogous to the preparation of allyl
(3,4,6-tri-O-benzyl-2-O-acetyl-b-D-glucopyranosyl)-(1 → 2)-3,4,6-
tri-O-benzyl-b-D-mannopyranoside.10 Glucopyranosyl imidate 5
(650 mg, 1.02 mmol), monosaccharide acceptor 4 (440 mg
˚
0.85 mmol), and activated 4 A molecular sieves (100 mg) were
dried together under vacuum for one hour in a pear-shaped
flask (50 mL). The contents of the flask were then dissolved in
dichloromethane (8 ml). The suspension was stirred for 10 min
at room temperature under argon, and then the temperature was
reduced with a −10 ◦C bath, and trimethylsilyl trifluoromethane-
sulfonate (6 ll) was added dropwise. After 30 min, the reaction
mixture was neutralized with triethylamine and concentrated in
vacuum. Column chromatography in n-hexane–ethyl acetate (4 : 1)
gave the disaccharide 6 (758 mg, 85%). [a]D −27.6◦ (c 1.0, CHCl3);
1H-NMR dH (CDCl3, 600 MHz): 7.2–7.42 (30H, m, Ar), 5.92–
3
=
5.85 (1H, m, CH2CH CH2), 5.17–5.14 (1H, dd, J = 7.8 Hz,
=
9.6 Hz, H2b), 5.7–5.11 (1H, m, CH2CH CHa), 5.02–5.04 (1H, m,
=
CH2CH CHb), 4.76–4.97 (6H, m, H1b, CH2Ph), 4.48–4.6 (7H,
m, CH2Ph), 4.33 (1H, s, H1a), 4.28 (1H, d, 3J = 3.0 Hz, H2a), 3.89–
3.93 (1H, m, OCHa), 3.79 (2H, m, H6a, H6ꢀa), 3.75 (1H, dd, 3J =
8.4 Hz, H3b), 3.69 (1H, m, H6b), 3.58–3.65 (4H, m, H4a, H4b,
H5b, H6ꢀb), 3.44–3.51 (3H, m, H3a, H5a, OCHH), 2.25 (2H, m,
=
=
5.85 (1H, m, CH2CH CH2), 4.99–5.03 (1H, m, CH2CH CHa),
CH2CH CH2), 1.98 (3H, s, Ac), 1.78 (2H, m, OCH2CH2); 13C-
4.93–4.98 (4H, m, H1b, CH2Ph, CH2CH CHb), 4.84–4.90 (2H,
=
=
NMR dC (125 MHz, CDCl3), 138.6–127.5, 114.9, 101.1, 100.7,
83.2, 80.1, 78.2, 75.6, 75.3, 75.1, 74.9, 74.8, 73.6, 73.3, 73.2,
72.3, 70.5, 69.8, 68.8. EMS; M + Na+ 1015.46036: C61H68O12Na+
requires 1015.46030.
m, 2J = 12.0 Hz, CH2Ph), 4.56–4.69 (4H, m, 2 CH2Ph), 4.44–4.50
(5H, m, H2a, 2 × CH2Ph), 4.38 (1H, s, H1a), 4.34 (1H, dd,
3J = 1.2 Hz, 3.0 Hz, H2b), 3.92–3.94 (2H, m, H4b, OCHa),
3.77–3.80 (3H, m, H4a, H6a, H6b), 3.67–3.74 (2H, m, H6ꢀa,
H6ꢀb), 3.56–3.59 (2H, m, H3a, H3b), 3.49–3.52 (1H, m, H5b),
=
3.42–3.47 (2H, m, H5a, OCHb), 2.20 (2H, m, CH2CH CH2),
Pentenyl 2-O-(3,4,6-tri-O-benzyl-b-D-glucopyranosyl)-3,4,6-tri-
O-benzyl-b-D-mannopyranoside (7)
1.78 (2H, m, OCH2CH2); 13C-NMR dC (125 MHz, CDCl3),
138.4–115.1, 101.1 (1JC-H = 162 Hz, C1b), 99.3 (1JC-H = 156 Hz,
C1a), 81.5, 80.4, 75.6, 75.1, 74.4, 74.2, 73.5, 73.3, 70.8, 70.7, 70.1,
69.9, 69.5, 69.2, 67.7; EMS; M + Na+ 973.44991: C59H66O11Na+
requires 973.44974.
The procedure used was analogous to the preparation of al-
lyl (3,4,6-tri-O-benzyl-b-D-glucopyranosyl)-(1 → 2)-3,4,6-tri-O-
benzyl-b-D-mannopyranoside.10 To a solution of disaccharide 6
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 3477–3485 | 3483
©

View Article Online
Pentenyl 2-O-(2,3,4,6-tetra-O-acetyl-b-D-mannopyranosyl)-3,4,6-
tri-O-acetyl-b-D-mannopyranoside (9)
(2-[2-{2-Azidoethoxy}ethoxy]ethyl) 2,3,4,6-tetrakis-(2-hydroxy-
[pentanyl 2-O-(b-D-mannopyranosyl)-b-D-manno-pyranosyl]-3-
thiapropyl)-b-D-glucopyranoside (11)
The perbenzylated disaccharide 8 (80 mg, 0.084 mmol) was
dissolved in THF (2 mL) and t-butanol (2 mL). The solution
was added in one portion to a solution of sodium metal (100 mg)
in ammonia (∼50 mL) and the mixture was stirred with a glass
A stirred solution of 10 (47 mg, 0.06 mmol) and 3 (6.8 mg,
0.012 mmol) in MeOH (2 mL) was flushed with argon for
45 minutes at room temperature. Potassium carbonate (12 mg,
0.087 mmol) was added and the solution stirred for 1 hour at
room temperature. The reaction solution became turbid and five
drops of degassed water was added, and the solution was stirred
at room temperature for 18 h. The reaction mixture was diluted
with MeOH (10 mL) and neutralized with H+-ion exchange resin.
The solution was filtered and the resin was washed with MeOH
(5 mL, 3 times). Solvents were evaporated and the residue was
purified by HPLC on a C18-preparative column under gradient
conditions (a) 5 min → H2O–MeOH (100 : 0), (b) 15 min →
H2O–MeOH (67 : 33), (c) 60 min → H2O–MeOH (0 : 100), (d)
20 min → H2O–MeOH (0 : 100) to give the product 11 (17 mg,
60%) as a white powder. 1H-NMR dH (D2O, 600 MHz): 4.84 (4H, s,
4 × H1b), 4.75 (4H, s, 4 × H1b), 4.54 (1H, m, H1Glc), 4.25 (4H,
m, 4 × H2a), 4.13 (4H, m, 4 × H2b), 3.35–4.05 (77H, m, 4 ×
[H3b, H4b, H5b, H6b-, H6ꢀb, H3a, H4a, H5a, H6a, H6ꢀa, OCH2,
OCH], H6aGlc, H6bGlc, H2Glc, H3Glc, H4Glc, 3 OCH2CH2),
3.24 (1H, m, H5Glc), 2.23–2.8 (16H, m, 4 × CH2SCH2), 1.62–1.68
(16H, m, 4 × OCH2CH2CH2CH2CH2S), 1.45–1.50 (8H, m, 4 ×
OCH2CH2CH2CH2CH2S); MALDI-MS (positive mode, DHB,
H2O): M + K+ 2377.33: C92H167N3O56S4K+ requires 2376.92.
◦
coated stir bar at −78 C. The flask previously containing 8 was
rinsed with THF (2 mL) and t-butanol (2 mL) and this solution
was added to the ammonia solution. After 30 minutes the reaction
was quenched with methanol and the ammonia was allowed to
evaporate at room temperature. The remaining THF was removed
under vacuum and the resulting white solid was taken up in water
(5 mL). The suspension was neutralized with a 5 M acetic acid
solution against pH paper and filtered through a 0.2 lM filter.
The solution was then passed through a C18 Sep-pak cartridge
and eluted with methanol. The crude product was acetylated to
afford compound 9 (35 mg, 60%); [a]D −86.6◦ (c 1.0, CHCl3); 1H-
=
NMR dH (CDCl3, 600 MHz): 5.76–5.84 (1H, m, CH2CH CH2),
5.53 (1H, dd, 3J = 1.2 Hz, 3.6 Hz, H2b), 5.03–5.21 (2H, m,
=
H4a, H4b), 4.95–5.03 (3H, m, H3b, CH2CH CH2), 4.85 (1H,
3
bs, H1b), 4.63 (1H, dd, J = 3.2 Hz, 10.1 Hz, H3a), 4.47 (1H, s,
3
H1a), 4.35 (1H, d, J = 3.0 Hz, H2a), 4.19–4.29 (2H, m, H6a,
H6b), 4.06–4.10 (1H, m, H6ꢀa), 3.98 (1H, m, H6ꢀb), 3.89 (1H,
m, OCHH), 3.56 (1H, m, H5b), 3.51 (1H, m, H5a), 3.41–3.48
=
(1H, m, OCHH), 1.99–2.2 (23H, m, CH2CH CH2, 7 Ac), 1.78
(2H, m, OCH2CH2); 13C-NMR dc (125 MHz, CDCl3), 169.2–
170.8, 137.8, 115.0, 99.8 (1JC–H = 162 Hz, C1b), 97.9 (1JC–H
=
156 Hz, C1a), 72.3, 72.1, 71.9, 71.8, 70.7, 69.4, 68.6, 66.3, 65.1,
62.5, 61.9; EMS; M + Na+ 727.24186: C31H44O18Na+ requires
727.24199.
(2-[2-{2-Aminoethoxy}ethoxy]ethyl) 2,3,4,6-tetrakis-(2-hydroxy-
[pentanyl 2-O-(b-D-mannopyranosyl)-b-D-mannopyranosyl]-3-
thiapropyl)-b-D-glucopyranoside (12)
Compound 11 (29 mg) was dissolved in a mixture solvent of H2O,
pyridine and NEt3 (10 : 1 : 0.3) (10 mL), and H2S was bubbled
through the reaction mixture at room temperature. After 4 h,
TLC indicated that the starting material had reacted. Finally, the
solvents were removed together with excess H2S. The residue was
dissolved in H2O (2 mL), and lyophilized to give the product as
1-Thioacetyl-pentanyl 2-O-(2,3,4,6-tetra-O-acetyl-b-D-
mannopyranosyl)-3,4,6-tri-O-acetyl-b-D-mannopyranoside (10)
Thioacetic acid (18 ll, 0.226 mmol) was added to a solution of
peracetylated disaccharide 9 (53 mg, 0.075 mmol) dissolved in
CH2Cl2 (4 mL) and the reaction mixture was flushed with argon
for 2 min. The reaction was exposed to UV light for 15 min
at which point the reaction was judged to be complete since
NMR indicated all starting material had been consumed. The
mixture diluted in CH2Cl2 (10 mL) was washed with saturated
sodium bicarbonate solution, brine, and the solution was dried
over MgSO4. The solution was evaporated to give the crude
product and this residue was purified by chromatography to afford
1
a white powder. H-NMR dH (D2O, 600 MHz): 4.83 (4H, s, 4 ×
H1b), 4.74 (4H, s, 4 × H1a), 4.55 (1H, m, H1Glc), 4.25 (4H, m, 4 ×
H2a), 4.15 (4H, m, 4 × H2b), 3.24 (1H, m, H2Glc); MALDI-MS
(positive mode, DHB, H2O): M + K+ 2335.64: C92H169NO56S4Na+
requires 2334.93.
9-Aza-10,15-dioxo-15-(4-nitro-phenoxy)-(2-[2-{2-
ethoxy}ethoxy]ethyl) 2,3,4,6-tetrakis-(2-hydroxy-[pentanyl
2-O-(b-D-mannopyranosyl)-b-D-mannopyranosyl]-3-thiapropyl)-
b-D-glucopyranoside (13)
pure 10 (51 mg, 87%). [a]D 84.8◦ (c 1.0, CHCl3); H-NMR dH
1
3
(CDCl3, 600 MHz): 5.54 (1H, d, J = 3.0 Hz, H2b), 5.18–5.21
(2H, m, H4a, H4b), 5.05 (1H, m, H3b), 4.85 (1H, s, H1b), 4.64
To a solution of free amine 12 (3.5 mg, 1.5 lmol) in dry DMF
(1 mL) was added diester 14 (12 mg, 0.03 mmol) under argon
with stirring. After 5 h, TLC indicated almost complete reaction
of the amine 12. The reaction mixture was co-evaporated with
toluene to remove DMF, and the residue was dissolved in CH2Cl2
(5 mL), and washed with H2O (5 mL) containing 1% acetic acid.
The water solution was then applied to a C18-Sep-Pak cartridge.
The cartridge was washed with water and then with methanol
containing 1% acetic acid, to remove hydrophobic compounds
that would be irreversibly absorbed to the silica reverse phase
column. The solution was concentrated at low pressure to afford
3
(1H, dd, J = 10.2 Hz, 3.0 Hz, H3a), 4.47 (1H, s, H1a), 4.35
3
(1H, d, J = 3.0 Hz, H2a), 4.29 (1H, m, H6b), 4.22 (1H, m,
H6a), 4.09 (1H, m, H6ꢀa), 4.01 (1H, m, H6ꢀb), 3.87 (1H, m,
OCHH), 3.61 (1H, m, H5b), 3.50 (1H, m, H5a), 3.42 (1H, m,
OCHH), 2.87 (2H, t, CH2SAc), 2.30 (3H, s, SAc), 1.98–2.2 (21H,
m, 7 × Ac), 1.53–1.70 (4H, m, OCH2CH2CH2CH2CH2SAc), 1.4
(2H, m, OCH2CH2CH2CH2CH2SAc); 13C-NMR dC (125 MHz,
CDCl3), 99.8, 97.9, 72.4, 72.1, 71.9, 70.8, 69.7, 68.6, 66.4, 65.2,
62.6, 62.0; EMS; M + Na+ 803.24013: C33H48O19Na+ requires
803.24027.
3484 | Org. Biomol. Chem., 2007, 5, 3477–3485
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
crude product as a solid. Final purification on a C18 reverse
phase column was accomplished with a water–methanol mixture
containing a 1% acetic acid gradient to yield pure half ester 13
(2.5 mg, 64%). 1H-NMR dH (D2O, 600 MHz): 8.40 (2H, m, C6H4),
7.44 (2H, m, C6H4), 4.83 (4H, s, 4 × H1b), 4.71 (4H, s, 4 × H1a),
4.49 (1H, m, H1Glc), 4.23 (4H, m, 4 × H2a), 4.13 (4H, m, 4 ×
H2b), 3.17 (1H, m, H2Glc); MALDI-MS (positive mode, DHB,
H2O): M + Na+ 2585.8598: C104H180N2O61S4Na+ requires 2585.78.
washing with PBS containing 0.1% Tween (PBST), wells were
filled with 100 lL of serial dilutions of sera (starting from 10−3)
in root of ten order. BSA (0.1%) in PBST was used for dilutions
to prevent non-specific binding. Plates were sealed and incubated
for 2 h at room temperature. After washing with PBST, a reporter
antibody (anti-mouse IgG, HRP conjugate) in 0.1% BSA PBST,
at a dilution of 1/2000 was applied and plates were incubated for
1 h at room temperature. Plates were washed again with PBST
and color developed with a HRP substrate system for 15 min. The
reaction was stopped with 1 M phosphoric acid and absorbance
measured in an ELISA plate reader. Titres were recorded as the
dilution giving an absorbance 0.2 above background at 450 nm.
Cluster glycoconjugates
The general procedure for generating protein–carbohydrate con-
jugates was as follows: BSA (10 mg) was dissolved in phosphate
buffer pH 7.5 (2 mL). Half ester was dissolved in DMF (100 ll)
and this solution was injected slowly into the buffered solution of
protein. The reaction was left for one day at room temperature. The
mixture was then diluted with deionized water and dialysed against
5 changes of deionized water (2 L). Tetanus toxoid conjugates were
dialysed against PBS and stored as a PBS solution, pH = 7.2. The
aqueous solutions of BSA conjugates were lyophilized to white
solids.
References
1 A. Espinel-Ingroff and M. A. Pfaller, in Manual of clinical microbiology,
ed. P. R. Murray, E. J. Baron, M. A. Pfaller, F. C. Tenover and
R. H. Yolken, American Society for Microbiology, Washington, D.C.,
6th edn, 1995.
2 H. Kobayashi, N. Shibata, M. Nakada, S. Chaki, K. Mizugami, Y.
Ohkubo and S. Suzuki, Arch. Biochem. Biophys., 1990, 278, 195–204.
3 N. Shitbata, K. Ikuta, T. Imai, Y. Satoh, S. Richi, A. Suzuki, C. Kojima,
H. Kobayashi, K. Hisamichi and S. Suzuki, J. Biol. Chem., 1995, 270,
1113–1122.
MALDI-MS (positive mode, matrix; sinapinic acid, H2O): BSA-
cluster conjugate 15 (79368), TT-cluster conjugates 16 (165059)
and 17 (157306).
4 Y. Han and J. E. Cutler, Infect. Immun., 1995, 63, 2714–2719.
5 Y. Han, R. P. Morrison and J. E. Cutler, Infect. Immun., 1998, 66,
5771–5776.
6 Y. Han, M. A. Ulrich and J. E. Cutler, J. Infect. Dis., 1999, 179, 1477–
1484.
Rabbit immunization
7 M. Nitz, C. C. Ling, A. Otter, J. E. Cutler and D. R. Bundle, J. Biol.
Chem., 2002, 277, 3440–3446.
Groups of three New Zealand white rabbits (weighing approxi-
mately 3 kg) were immunized with each conjugate absorbed on
alum in PBS as adjuvant. Vaccine formulation in a suspension
of alum (1 mL) containing 300 ug of conjugate was administered
on day 0 by injections of 0.2 mL each to the quadriceps of the
posterior thigh, lumbar muscles (both sides) and 3 subcutaneous
sites. A second set of injections at the same sites were given on day
21. Blood samples were drawn on day 0 (pre-immune) and ten days
after the second injection.
8 E. A. Kabat, Fed. Proc., 1962, 21, 694–701.
9 E. A. Kabat, J. Immunol., 1966, 97, 1–11.
10 X. Wu and D. R. Bundle, J. Org. Chem., 2005, 70, 7381–7388.
11 D. Sames, X. T. Chen and S. J. Danishefsky, Nature, 1997, 389, 587.
12 J. Zhang and P. Kovac, Carbohydr. Res., 1999, 321, 157.
13 J. Ni, H. Song, Y. Wang, N. M. Stamatos and L.-X. Wang, Bioconjugate
Chem., 2006, 17, 493.
14 P. I. Kitov, Y. E. Tsvetkov, L. V. Bakinovskii and N. K. Kochetkov,
Dokl. Chem., 1993, 329, 175–179.
15 (a) M. Nitz, B. W. Purse and D. R. Bundle, Org. Lett., 2000, 2, 2939;
(b) M. Nitz and D. R. Bundle, J. Org. Chem., 2001, 66, 8411.
16 (a) A. B. Charette, N. Turcotte and B. Cote, J. Carbohydr. Chem., 1994,
13, 421; (b) R. R. Schmidt and G. Effenberger, Liebigs Ann. Chem.,
1987, 825.
17 V. Gagnard, A. Leydet, A. More´re, J. Montero, I. Lefe´bvre, G. Gosselin,
C. Pannecouque and E. D. Clercq, Bioorg. Med. Chem., 2004, 12, 1393.
18 X. Wu, C.-C. Ling and D. R. Bundle, Org. Lett., 2004, 6, 4407.
19 H. Nakada, M. Inoue, Y. Numata, N. Tanaka, I. Funakoshi, S. Fukui,
A. Mellors and I. Yamashina, Proc. Natl. Acad. Sci. U. S. A., 1993, 90,
2495–2499.
20 B. W. T. Yin, C. L. Finstad, K. Kitamura, M. J. Frederici, M.
Welshinger, V. Kudryshov, W. J. Hoskins, S. Welt and K. O. Lloyd,
Int. J. Cancer, 1996, 65, 406.
21 S. Zhang, L. A. Walberg, S. Ogata, S. H. Itzkowitz, R. R. Koganty,
M. Reddish, S. S. Gandhi, B. M. Longenecker, K. O. Lloyd and P. O.
Livingston, Cancer Res., 1995, 55, 3364–3368.
22 S. D. Kuduk, J. B. Schwarz, X. T. Chen, P. W. Glunz, D. Sames, G.
Ragupathi, P. O. Livingston and S. J. Danishefsky, J. Am. Chem. Soc.,
1998, 120, 12474.
23 G. Ragupathi, L. Howard, S. Cappello, R. R. Koganty, D. Qiu, B. M.
Longenecker, A. M. Reddish, K. O. Lloyd and P. O. Livingston, Cancer
Immunol. Immunother., 1999, 48, 1–8.
24 P. W. Glunz, S. Hintermann, L. J. Williams, J. B. Schwarz, S. D. Kuduk,
V. Kudryashov, K. O. Lloyd and S. J. Danishefsky, J. Am. Chem. Soc.,
2000, 122, 7273–7279.
Mouse immunization
Three groups of ten CD1 mice were immunized with each
glycoconjugate. Each group of ten mice were divided into two
groups of five mice. One group received vaccine absorbed to
alum and the other group the vaccine emulsified with Freunds
adjuvant. The group immunized with alum deposited vaccine
received a total of 90 lg of cluster glycoconjugate in 300 lL of
vaccine suspension administered intraperitoneally (200 lL) and
subcutaneously (100 lL) on days 0 and 21. Serum was collected
at day 31.
Mice immunized with the vaccine emulsified with Freunds
adjuvant received 90 lg of glycoconjugate in a 1 : 1 mixture of PBS
and adjuvant (200 lL) given at two subcutaneous sites. The first
injection employed a 50 : 50 mixture of complete and incomplete
Freunds adjuvant and the second injection used only incomplete
adjuvant.
ELISA titrations
25 R. Lo-Man, S. Vichier-Guerre, R. Perraut, E. De´riaud, V. Huteau, L.
BenMohamed, O. M. Diop, P. O. Livingston, S. Bay and C. Leclerc,
Cancer Res., 2004, 64, 4987–4994.
Polystyrene 96-well microtitre plates were coated overnight with
BSA glycoconjugates at a concentration of 5 lg mL−1 in PBS. After
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 3477–3485 | 3485
©