Probing carbohydrate-lectin recognition in heterogeneous environments with monodisperse cyclodextrin-based glycoclusters

Article abstract of DOI:10.1021/jo201797b

A series of β-cyclodextrin (βCD)-scaffolded glycoclusters exposing heterogeneous yet perfectly controlled displays of α-mannosyl (α-Man) and β-lactosyl (β-Lact) antennas were synthesized to probe the mutual influence of varying densities of the saccharide

Full text of DOI:10.1021/jo201797b

Article
pubs.acs.org/joc
Probing Carbohydrate-Lectin Recognition in Heterogeneous
Environments with Monodisperse Cyclodextrin-Based Glycoclusters
Marta Gomez-García,† Juan M. Benito,*,‡ Anna P. Butera,† Carmen Ortiz Mellet,†
́
Jose M. García Fernandez,‡ and Jose L. Jimenez Blanco*,†
́
́
́
́
†Departamento de Química Organ
Spain
‡Instituto de Investigaciones Químicas, CSIC - Universidad de Sevilla, Amer
́
ica, Facultad de Química, Universidad de Sevilla, Profesor García Gonzal
́
ez 1, E-41012 Sevilla,
́
ico Vespucio 49, Isla de la Cartuja, E-41092 Sevilla, Spain
S
* Supporting Information
ABSTRACT: A series of β-cyclodextrin (βCD)-scaffolded glycoclusters
exposing heterogeneous yet perfectly controlled displays of α-mannosyl
(α-Man) and β-lactosyl (β-Lact) antennas were synthesized to probe the
mutual influence of varying densities of the saccharide motifs in the binding
properties toward different plant lectins. Enzyme-linked lectin assay
(ELLA) data indicated that the presence of β-Lact residues reinforced
binding of α-Man to the mannose-specific lectin concanavalin A (Con A)
even though homogeneous β-Lact clusters are not recognized at all by this
lectin, supporting the existence of synergic recognition mechanisms
(heterocluster effect). Conversely, the presence of α-Man motifs in the
heteroglycoclusters also resulted in a binding-enhancing effect of β-Lact
toward peanut agglutinin (PNA), a lectin strongly binding multivalent
lactosides but having no detectable affinity for α-mannopyranosides, for
certain architectural arrangements. Two-site, sandwich-type ELLA data
corroborated the higher lectin clustering efficiency of heterogeneous glycoclusters compared with homogeneous displays of the
putative sugar ligand with identical valency. A turbidity assay was also consistent with the previous observations. Most revealingly,
the lectin cross-linking ability of heterogeneous glycoclusters was sensitive to the presence of high concentrations of the non-
ligand sugar, strongly suggesting that “mismatching” saccharide motifs may modulate carbohydrate-lectin specific recognition in a
lectin-dependent manner when present in highly dense displays together with the “matching” ligand, a situation frequently
encountered in biological systems.
Multivalent glyco-constructs taking heterogeneity into
consideration are rather scarce. Statistical grafting of different
saccharidic wedges onto polymeric,8 dendritic,9 or self-
assembled scaffolds10 affords polydisperse materials in which
the relative position between the sugar components is not
defined. On the other hand, orthogonal functionalization of
dendritic cores is concomitantly associated, in most cases, to
low-throughput synthetic strategies,11 whereas combinatorial
approaches12 are so far limited to low valency derivatives. Only
a few reports have investigated the role that “mismatching”
carbohydrates might exert during the recognition of “matching”
glycoligands by a specific lectin.8c,9b,10a,11d,12a In a previous
communication we have reported a synthetic methodology for
the preparation of β-cyclodextrin (βCD)-centered heteroglyco-
clusters13 that allows sampling monodisperse conjugates with
perfectly defined densities and spatial orientation of different
saccharidic antennas, e.g., α-D-mannopyranoside (α-Man) and
β-D-glucopyranoside (β-Glc) residues.14 Notably, the affinity of
these clusters toward the α-Man-specific lectin concanavalin
INTRODUCTION
■
Specific interactions between carbohydrate ligands and
carbohydrate-binding proteins (lectins) are known to play
key roles in a plethora of fundamental processes in cell daily
life.1 Despite being characteristically weak, Nature has managed
to efficiently exploit these interactions through presentation of
both the binding motifs and the recognition sites in multiple
copies. Affinity enhancements exceeding those expected from
the simple addition of individual interactions can be achieved in
this way, a phenomenon known as the cluster or multivalent
glycoside effect.2 First noted by Lee and co-workers,3 the cluster
effect has been extensively investigated. Synthetic polyconju-
gates with well-defined structures have contributed to unravel
the mechanisms at work,4 leading eventually to useful tools for
biotechnological5 or therapeutic purposes.6 The great majority
of these artificial systems incorporate a single sugar motif,
which strongly contrasts with the heterogeneity generally
encountered in biological systems. Actually, differences in the
surface density of a particular sugar ligand would be expected to
be affected by the relative proportion and location of the
neighbor saccharides, which might affect the affinity and selectivity
toward a complementary receptor.7
Received: August 31, 2011
Published: December 20, 2011
© 2011 American Chemical Society
1273
dx.doi.org/10.1021/jo201797b | J. Org. Chem. 2012, 77, 1273−1288

The Journal of Organic Chemistry
Article
A (Con A) was found to be amplified by the presence of the
β-Glc motifs, which are supposed to be irrelevant for this
particular lectin, in highly dense glycocluster presentations.
This unprecedented binding-enhancing role of “mistmatching”
carbohydrates, termed the heterocluster effect,15 was further
confirmed by several monitoring techniques, including com-
petitive enzyme-linked lectin assay (ELLA) isothermal titration
calorimetry (ITC) and surface plasmon resonance (SPR).16
Thermodynamic analysis indicated that the heterocluster
effect relies on a favorable entropic contribution to the free
energy of binding. It was speculated that the “mismatching”
sugar residue could facilitate sliding of the “matching” epitopes
over the lectin binding site by transient exchange.14,16 Whether
or not this phenomenon is a peculiarity of α-Man−Con A
binding in the presence of β-Glc or might this synergistic
binding mechanism be also operating in other carbohydrate-
lectin pairs remains intriguing. In order to address this question,
a series of monodisperse homo- and heteromultivalent βCD-
centered glycoclusters featuring α-Man and β-lactoside (β-Lact)
units have been now synthesized. Their binding affinities
toward two different plant lectins, namely, the mannose-specific
lectin Con A and the lactose-binding lectin peanut agglutinin
(PNA), have been evaluated by ELLA, two-site ELLA, and
turbidimetry. Both lectins have been extensively used for the
purification and characterization of glycoconjugates in a variety
of research areas, including the separation and structural
analysis of oligosaccharides or glycopetides cleaved from
glycoproteins, the pattern analysis for tissue comparison, or
the detection of different glycoforms in glycoproteins.17 Results
are discussed as a function of the total carbohydrate density, the
relative proportion of the α-Man and β-Lact units, and the
nature of the lectin.
RESULTS AND DISCUSSION
■
General Synthetic Strategy and Building Blocks. The
synthesis of the new series of α-Man and β-Lact homo- and
heteroglycoclusters has been carried out by implementing a
modular convergent strategy that takes advantage of (i) the
stepwise radical addition of thiols to double bonds (ene-thiol
addition)18 for the construction of glycodendrons and (ii) the
amine-isothiocyanate coupling reaction (thiourea-forming
reaction) to generate the hyperbranched structure.19 The
ene-thiol addition proceeds with anti-Markovnikok regio-
selectivity and permits the sequential incorporation of
saccharidic wedges onto a polyene branching element that
can be further armed with an isothiocyanate group (Figure 1A),
while the thiourea-forming reaction is very well suited for
multiconjugation purposes (Figure 1B).20 Triallylated penta-
erythritol 121 and per-(C-6)-cysteaminyl-βCD 222 (Figure 1C)
were chosen as the central building blocks. In this manner,
a series of C7-symmetric homo- and heterofunctional glycoclusters
with a total (α-Man + β-Lact) 21-valency became accessible.
The azo-bis(isobutironitrile) (AIBN)-initiated radical addi-
tion of either the β-Lact (3)23 or α-Man (4)24 thiosugars to
1 led to the homotrivalent dendrons 5 and 616 in 71% and 83%
yield, respectively (Scheme 1). Triflyl activation of the pri-
mary hydroxyl group in 5 and 6 followed by azide anion
displacement (→ 7 and 8)16 and isothiocyanation using the
triphenylphosphine−carbon disulfide (TPP−CS2) system25 led
to the corresponding isothiocyanate-armed dendrons 9 and
10.16 The fully unprotected trivalent compounds triLact-OH
and triMan-OH16 were also prepared as control compounds
(Scheme 1).
Figure 1. Schematic representation of the modular convergent strategy
implemented to build monodisperse CD-scaffolded hetero-
glycoclusters highlighting the key steps (A, ene-thiol addition; B,
amine-isothiocyanate addition) and the corresponding building
blocks (C).
Radical addition of thiosugars 3 and 4 to 1 can be
experimentally controlled to preferentially obtain partially
glycosylated products by simply adjusting the amount of
reactive thiol. Thus, by reacting 0.45 equiv of thiol 3 with the
triallylated scaffold 1 a binary mixture of the easily separable
mono- (11, 71%) and dilactosylated (12, 25%) derivatives
could be obtained. Further AIBN-mediated reaction of these
conjugates with the mannosyl thiol 4 furnished the
heterotrivalent adducts 13 and 14, respectively, which were
transformed into the isothiocyanate-armed derivatives 17 and
18 via the corresponding azide intermediates 15 and 16
following a reaction sequence parallel to that above commented
for the homogeneous counterparts 9 and 10. Conventional
deacetylation of 13 and 14 was also effected to obtain the
corresponding heterotrivalent models MandiLact-OH and
LactdiMan-OH (Scheme 2).
In order to compare the multivalent effect in high-density
versus low-density conjugates, monovalent building blocks were
required. For that purpose, the strategy depicted in Scheme 3
was implemented. Starting from monoallylated propyleneglycol
19,26 the lactosylated and mannosylated derivatives 20 and 21
were obtained. Conventional deacetylation furnished Lact-OH
1274
dx.doi.org/10.1021/jo201797b | J. Org. Chem. 2012, 77, 1273−1288

The Journal of Organic Chemistry
Article
Scheme 1. Synthesis of Trivalent Homoglycodendrons and Building Blocks
Scheme 2. Synthesis of Trivalent Heteroglycodendrons and Building Blocks
Scheme 3. Synthesis of Monovalent Ligands and Building
Blocks
Synthesis of βCD-Scaffolded Homo- and Heterogly-
coconjugates. The thiourea-forming reaction was purposely
chosen for the key multiconjugation step because it is totally
chemoselective in the presence of free hydroxyl groups, is high
yielding, and has no solvent restrictions or byproduct
formation, and furthermore the thiourea linkages are bio-
compatible and physiologically stable. The series of isothio-
cyanate-armed derivatives 9, 10, 17, and 18 feature all possible
combinations of α-mannosyl and β-lactosyl antennas on a
trivalent dendron. Amplification of these patterns by coupling
with the heptacysteaminyl βCD derivative 2 furnished a diverse
set of displays of per-O-acetylated α-Man and β-Lact epitopes
(28−31) with a global 21-valency (Scheme 4). Similarly, the
monovalent building blocks 23 and 27 afforded the
corresponding homogeneous heptavalent glycoclusters 32 and
33. TLC and NMR monitoring of the reaction mixtures
evidenced outstanding chemical yields in all cases. Nevertheless,
chromatographic purification of the high molecular weight
hemiacetates handicapped the final isolation yields in certain
cases (see Experimental Section). Final acetyl cleavage by
and Man-OH,16 respectively, while the sequence hydroxyl
triflation → azide displacement (to give 22 and 26)16 followed
by reaction with the CS2-TPP system furnished the corre-
sponding isothiocyanate-armed building blocks 23 and 27.16
mixed Zemplen transesterification-saponification reaction
́
quantitatively furnished the target fully unprotected multivalent
glycoconjugates triLact-CD triMan-CD, MandiLact-CD,
LactdiMan-CD, Lact-CD, and Man-CD.
1275
dx.doi.org/10.1021/jo201797b | J. Org. Chem. 2012, 77, 1273−1288

The Journal of Organic Chemistry
Article
Scheme 4. Synthesis of CD-Scaffolded Homo- and Heteroglycoclusters
The homogeneity and purity of all structures were confirmed
by mass spectrometry, NMR spectroscopy, and combustion
1
analysis. H and 13C NMR spectra of the final conjugates in
D2O showed the typical line broadening associated with
restricted rotation at the pseudoamide NH−C(S) bonds,27
which remained evident also at elevated temperatures (333−
353 K). The drastic decrease of motion at the central core
region of the macromolecular conjugates provokes an increase
in the relaxation time for the corresponding carbon atoms that
translates into much lower intensities in the 13C NMR spectra
in comparison with the carbons of the peripheral glycoligands.
Nevertheless, both the 13C and 1H NMR spectra were
consistent with the expected C7 symmetry for homogeneously
substituted βCD-centered clusters. In the case of the
homoglycoconjugates triMan-CD and triLact-CD, proton
spectra showed two and three spin systems, respectively,
corresponding to the cyclodextrin core and the mannosyl (for
triMan-CD) or glucosyl and galactosyl units (for triLact-CD)
in 1:3 ratio. For heteroglycoclusters LactdiMan-CD and
MandiLact-CD, four different spin systems were distinguished,
their relative intensities perfectly matching the expected
mannose/lactose/CD relative proportions (Figure 2).
Evaluation of Lectin Binding Affinity by ELLA. First, the
Con A binding avidity of the α-Man/β-Lact glycoclusters was
evaluated by enzyme-linked lectin assay (ELLA).28 This test
measures the ability of a soluble saccharide to inhibit the
association between a labeled lectin (here Con A or PNA lectin
labeled with horseradish peroxidase, Con A-HRP or PNA-
HRP) and a ligand immobilized on the microtiter well (here
yeast mannan or a lactose-functionalized polyacrylamide
polymer). The location of the glycodendrons at the primary
face of the CD platform provides a well-defined topology in
which the saccharide epitopes share the same space region,
featuring a highly dense glycosylated surface (Figure 3, top).
Considering that the HRP label is a rather big protein,
(40 kDa) this architecture likely prevents two lectin moieties
from approaching, as previously demonstrated in analogous
systems,14 resulting in 1:1 binding stoichiometries. On the
other hand, molecular simulations14,29 indicated that the largest
Figure 2. Stacked 1H (below) and 1D TOCSY NMR (500 MHz, D2O,
333 K) spectra of compound MandiLact-CD.
1276
dx.doi.org/10.1021/jo201797b | J. Org. Chem. 2012, 77, 1273−1288

The Journal of Organic Chemistry
Article
interaction of the multivalent conjugate with a single
carbohydrate binding site in the lectin.
In addition to the compounds depicted in Scheme 4, the
dimannosylated conjugate diMan-OH, still bearing an allyl
group instead of the lactosyl unit, and the corresponding CD-
scaffolded 14-valent homologue diMan-CD16 were included in
the binding assay against Con A-HRP as additional controls
(Chart 1). The concentration of glycoclusters required to
Chart 1. Structures of the Homogeneous Divalent and
Tetradecavalent Glycoclusters diMan-OH and diMan-CD
Used As Control Compounds in ELLA Experiments
achieve 50% inhibition, the IC50 values (Table 1) are assumed
to be inversely proportional to the lectin-saccharide free energy
Table 1. Inhibition of Yeast Mannan−Con A Binding by
Homo- and Heteromultivalent Glycoclusters Determined by
ELLA
Lact
units
Man
rel
Man molar rel
j
a
compound
units IC50 (μM) potency
potency
Me-α-Man
Man-OH
0
0
2
1
1
1
865 30b
800 35b
370 25c
1
1
1.1
2.3
1.1
2.3
MandiLact-
OH
diMan-OH
0
1
2
2
319 25d
340 30c
2.5
2.5
1.3
1.3
LactdiMan-
OH
triMan-OH
lactose
0
1
3
0
0
7
7
46 5e
nij
18.8
6.3
triLact-OH
Man-CD
1
nij
0
67 5f
50 5g
12.9
17.3
1.8
2.5
MandiLact-
CD
14
Figure 3. Schematic representation of the MandiLact-CD heteroge-
neous glycocluster (top) and 3D representation obtained by molecular
modeling (bottom; an axial view from the primary face of the βCD
core and a lateral view are shown at the right and left corners,
respectively). The CD platform is depicted as a space-filled CPK
model, whereas sticks have been used for the rest of the molecule.
β-Lactosyl and α-mannosyl residues are colored blue and green,
respectively.
diMan-CD
0
7
14
14
76 8f
14 2h
11.4
61
0.8
4.4
LactdiMan-
CD
triMan-CD
Lact-CD
0
7
21
0
5.5 0.5i
157
7.5
nik
nik
triLact-CD
21
0
a
Relative values are compared to methyl-α-D-mannopyranoside
distance between the external sugar units (ca. 6 nm) is shorter
than the distance between two adjacent recognition sites in
Con A30 or PNA31 lectins (Figure 3, bottom). This situation
should prevent contributions to the binding energy associated
to the chelate effect (e.g., simultaneous interaction of
glycotopes in the same glycocluster molecule with more than
one lectin binding site).9 In such circumstances, differences in
binding affinities can be ascribed, essentially, to the effect of
differences in the relative densities of the sugar motifs in the
(Me-α-Man, IC50 865 30 μM).32 The IC50 values are expressed as
mean values SD obtained from at least five independent determi-
b−i
nations.
Differences between data with different superscript letters
j
are statistically significant (P < 0.001). No inhibition at ca. 4 mM.
k
No inhibition at ca. 1 mM.
of binding. For comparative purposes, they were normalized to
the affinity shown by methyl α-D-mannopyranoside in a parallel
experiment (IC50 865 30 μM).32 In agreement with previous
1277
dx.doi.org/10.1021/jo201797b | J. Org. Chem. 2012, 77, 1273−1288

The Journal of Organic Chemistry
Article
results,12b,33 a substantial amplification of Con A-HRP avidity
was observed for conjugates where the mannosyl units were
presented in triads, and little, if any, differences were noticed
among mono- or dimannosylated dendrons, regardless of the
presence or absence of other saccharidic or nonsaccharidic
elements. The mannose-devoid ligands Lact-OH or triLact-
OH did not bind at all Con A-HRP, in agreement with the
known lectin specificity. Lactose, therefore, has no apparent
influence in the recognition of mannose by Con A when the
total sugar density is low.
The scenario was substantially different in the hyperbranched
CD-scaffolded glycocluster series. Despite lactose not being
itself recognized by Con A (no binding observed for the 7- or
21-valent homogeneous conjugates Lact-CD or triLact-CD),
the presence of lactosyl residues significantly enhanced the
relative mannose binding potency (i.e., from 1.8 to 2.5 in
mannose molar basis when going from Man-CD to MandiLact-
CD; from 0.8 to 4.4 from diMan-CD to LactdiMan-CD).
Actually, the IC50 of LactdiMan-CD (14 μM) was 5.4-fold lower
than that of diMan-CD, both bearing 14 α-D-mannopyranoside
units, and was fairly close to the value measured for triMan-CD
(5.5 μM), with 21 α-D-mannopyranoside residues. The lactose-
induced affinity increases are slightly less pronounced than those
previously measured for β-D-glucopyranose,16 indicating that the
heterocluster effect is dependent on the nature of the secondary
“mismatching” sugar.
surrounded by mannose as in LactdiMan-CD (3.3-fold relative
to lactose) than in the homogeneous 21-valent conjugate triLact-
CD (only 2.6-fold). This result is most striking considering that
1:2 lactose/mannose relative proportion was the least efficient in
the trivalent series. The heptalactoside LactdiMan-CD is actually
as potent as the tetradeca lactoside MandiLact-CD as a PNA
ligand, meaning that the individual lactose motifs are recognized
twice more efficiently in the first case. The particular structural
features of PNA, with binding sites that are more exposed to the
bulk solvent and significantly smaller than those of Con A,35
surely determine their different behavior.
Evaluation of Lectin Clustering Capability by Two-
Site ELLA. To evaluate the lectin clustering abilities of the new
glycoclusters, a two-site “sandwich” ELLA experiment was
performed.28b Unlabeled (therefore cross-linkable) lectin (Con
A or PNA) was first laid down onto a microtiter well. Preformed
complexes of the glycoclusters at different concentrations with the
corresponding HRP-labeled lectin (Con A-HRP or PNA-HRP)
were then added. The relative amounts of bound HRP-labeled
lectin obtained are represented in Figures 4 and 5, with the
The results for the competitive binding inhibition test using
PNA-HRP are summarized in Table 2. In agreement with the
Table 2. Inhibition of Polylactoside−PNA Binding by
Homo- and Heteroglycoclusters Determined by ELLA
Lact
Man
rel
Lact molar
b−g
IC50
1150 65b
a
compound
lactose
units units
(μM) potency
rel potency
1
1
0
2
1
1
Figure 4. Relative cross-linking efficiencies of CD-glycoclusters against
Con A at different concentrations.
LactdiMan-
OH
24% inhibition
at 1.7 mM
500 35c
MandiLact-
OH
2
1
2.3
3.6
1.2
1.2
triLact-OH
Man-OH
3
0
0
7
7
0
1
320 25d
nih
nih
77 8e
50 6f
triMan-OH
Lact-CD
3
0
14.9
23
2.1
3.3
LactDiMan-
CD
14
MandiLact-
CD
14
7
50 7f
23
1.6
2.6
triLact-CD
Man-CD
21
0
0
7
21 2g
nii
nii
54.8
triMan-CD
0
21
a
Relative values are compared to lactose (IC50 1150 65 μM). The
IC50 values are expressed as mean values SD obtained from at least
b−g
Differences between data with
different superscript letters are statistically significant (P < 0.001). No
inhibition at ca. 2 mM. No inhibition at ca. 1 mM.
five independent determinations.
h
Figure 5. Relative cross-linking efficiencies of CD-glycoclusters against
PNA at different concentrations.
i
expected lectin specificity,34 mannosylated conjugates were not
recognized at all by PNA. The presence of mannose residues in
combination with lactose in the trivalent derivatives was found
to be irrelevant. In the CD-scaffolded series, participation of
mannosyl residues in lectin recognition became evident. Thus,
the IC50 value decreases from 77 to 50 μM when going from
Lact-CD to LactdiMan-CD. In a lactose molar basis, the
recognition of the lactose binding motif is more efficient when
maximum for the corresponding homogeneous 21-valent
glycocluster triMan-CD (for Con A, Figure 4) or triLact-CD
(for PNA, Figure 5) at the highest concentration (250 μM) set at
100%. The residual value (11−15%) observed without previous
addition of ligands can be explained by HRP-protein exchanges
with unlabeled lectin at the surface of the well. Relative cross-
linking values in the range 10−18%, which were reached in the
experiments using methyl α-D-mannopyranoside, lactose, or the
1278
dx.doi.org/10.1021/jo201797b | J. Org. Chem. 2012, 77, 1273−1288

The Journal of Organic Chemistry
Article
trivalent glycodendrons (data not shown), must be considered
within the background.
As expected, homogeneous CD-centered glycoclusters
mismatching the known lectin selectivity, that is, triMan-CD
and Man-CD when assayed against PNA and triLact-CD and
Lact-CD when assayed against Con A, showed no clustering
capabilities. For matching pairs, the linking potential of the 21-
valent cluster was much higher as compared with the 7-valent
counterparts (1.5- to 2-fold at 250 μM). The clustering abilities
of the Man/Lact-CD heteroglycoclusters were, in all cases,
between these limits. Most importantly, the pairs MandiLact-
CD/Con A and LactdiMan-CD/PNA consistently exhibited a
higher tendency to cross-link as compared with the correspond-
ing Man-CD/Con A and Lact-CD/PNA pairs, even though
they have identical valency in a mannose molar basis, in agree-
ment with the enhanced lectin-binding capabilities in hetero-
geneous environments observed by classical ELLA.
Evaluation of Lectin Aggregation Capability by
Turbidity Assay. To test whether the differences in binding
simultaneously two lectin molecules by the glycoclusters
correlated with their relative capacity to promote the formation
of three-dimensional aggregates, a kinetic turbididy assay was
carried out. Turbidity measurements can be used to monitor
the formation of cross-linked complexes in real time.36 For that
purpose, the ligands were added to a solution of Con A or PNA
in PBS (pH 7.3), and the turbidity of the mixture was screened.
The initial rate of precipitation (Vi) was determined by linear
fits of the initial portion of the data (Figures 6 and 7).
Figure 7. Absorption changes of PNA in PBS (pH 7.3, 1 mg/mL,
50 μL) at 490 nm upon addition of lactose-containing CD-glycoclusters
(250 μM, 50 μL) as a function of time. D-Mannose (100 mM, 50 μL)
and lactose (100 mM, 2 × 50 μL) were sequentially added to the
mixtures after 12, 18, and 24 min. Addition of lactose (100 mM, 100 μL)
after 12 min led to basal absorption values in all cases. Values are
expressed as mean values SD obtained from at least three independent
determinations.
Table 3. Initial Aggregation Rates (Vi) of Con A and PNA
Lectins in PBS (pH 7.3, 1 mg/mL, 50 μL) by CD-Centered
a
Homo- and Heteroglycoclusters (250 μM, 50 μL) at 25 °C
Lact
units
Man
units
Vi, Con A (AU
Vi, PNA (AU
−1
compound
min−1)
min )
triMan-CD
triLact-CD
0
21
14
21
0
0.437
0
0
0.340
0.151
MandiLact-
CD
7
0.249
LactdiMan-
CD
7
14
0.369
0.111
Man-CD
Lact-CD
0
7
7
0
0.168
0
0
0.080
a
No aggregation was observed for Con A or PNA in combination with
any of the trivalent homo and heteroglycocendrons depicted in
Schemes 1 and 2, respectively.
cross-linking the corresponding lectin than the heptavalent
homogeneous derivatives Man-CD and Lact-CD, even though
the accessibility of the recognition motifs must be higher in the
later. Altogether, those data strongly support an active contribution
of the mismatching sugar on the binding of highly dense
heteroglycoclusters to both Con A and PNA.
Figure 6. Absorption changes of Con A (1 mg/mL, 50 μL) at 490 nm
upon addition of mannose-containing CD-glycoclusters (250 μM,
50 μL) as a function of time. Lactose (100 mM, 50 μL) and
D-mannose (100 mM, 2 × 50 μL) were sequentially added to the
mixtures after 12, 18, and 24 min. Addition of D-mannose (100 mM,
100 μL) after 12 min led to basal absorption values in all cases. Values
In order to assess whether the observed enhanced capacity of
heterogeneous mannose/lactose CD-glycoclusters to aggregate
the mannose specific lectin Con A was lactose-specific, 50 μL of
a 100 mM solution of lactose in PBS was added to the well
containing the suspensions of MandiLact-CD/Con A and
LactdiMan-CD/Con A after 12 min. In both cases a significant
decrease of the optical density, corresponding to a decrease in
turbidity of about 20%, was observed. In contrast, addition of
lactose had no effect on the aggregates of Con A with the
homogeneous mannosyl clusters triMan-CD and Man-CD.
Further addition of 50 μL of a 100 mM solution of mannose in
PBS (18 min) fully reverted aggregation in the case of the
heptamannosyl conjugates MandiLact-CD and Man-CD and
strongly decreased turbidity for the 14- and 21-valent (in
mannose molar basis) clusters LactdiMan-CD and triMan-CD;
a second addition of the mannose solution (24 min) was
necessary to completely disrupt the aggregates in these cases
(Figure 6). Parallel experiments in which the addition of 100 μL
of 100 mM solution of mannose in PBS to the suspensions was
are expressed as mean values
SD obtained from at least three
independent determinations.
The corresponding data (Table 3) were consistent with the
previous observations by two-site ELLA. Thus, no precipitation
occurred for trivalent dendrons or for mismatched CD-centered
glycocluster-lectin pairs. The 21-valent homogeneous glyco-
clusters were very efficient at promoting fast aggregation of
the matching lectin. For heterogeneous glycoclusters, the aggre-
gation capacity increased with the putative ligand valency
(α-D-mannopyranosyl for Con A and β-lactosyl for PNA). The
turbidimetry data also indicated that the heteroglycoclusters
with seven copies of the matching ligand, MandiLact-CD for
Con A and LactdiMan-CD for PNA, were more efficient at
1279
dx.doi.org/10.1021/jo201797b | J. Org. Chem. 2012, 77, 1273−1288

The Journal of Organic Chemistry
Article
effected directly after 12 min resulted in instantaneous rever-
sion for all homogeneous and heterogeneous glycoclusters
(data not shown).
function of the relative proportion of the matching-mismatch-
ing elements. Previous thermodynamic data pointed to an
entropic origin for the heterocluster effect, probably through
facilitating sliding of the matching sugar over the binding site by
transiently binding to the lectin.14 The fact that this process is
someway disrupted in the presence of the mismatching sugar in
the solution fits with this hypothesis. It should be emphasized,
however, that the mismatching sugar was not recognized by the
lectin either in multivalent presentation or at the high
concentration used in the competitive turbidimetric experi-
ment. This might indicate that transient binding occurs at the
vicinity of the primary binding site and is only relevant when
the matching epitope is already bound.
The turbidimetry data and initial aggregation rates in the case
of PNA (Figure 7 and Table 3) closely reproduced the relative
affinities obtained by ELLA (Table 2), with the heterogeneous
glycoclusters having similar capacity to bind to the lectin despite
MandiLact-CD having double lactose-base valency compared
with LactdiMan-CD. A similar experiment was carried out to
assess whether the mismatching α-D-mannopyranosyl units were
contributing to PNA aggregation in a mannose-specific manner.
Notably, addition of 50 μL of a 100 mM solution of mannose in
PBS to the suspensions (12 min) had a much more pronounced
effect in the LactdiMan-CD/PNA than in the MandiLact-CD/
PNA aggregates. This result is consistent with the more pro-
nounced mannose-induced PNA binding efficiency previously
determined by ELLA (relative affinities in lactose molar basis
3.3 vs 1.6 for LactdiMan-CDas compared to MandiLact-CD,
Table 2). Addition of excess lactose to the precipitates resulted
in clear solutions in all cases, demonstrating the lactose-specific
character of the aggregation.
Investigation of multivalent recognition is a complicated task
because it usually reflects an intricate overlapping of a number
of microscopic effects, the individual contributions to the
overall process being difficult to dissect.37 In our case, the
molecular design and experimental setup was conceived to
exclude contributions from simultaneous binding to several
lectin binding sites (chelate effect). The β-cyclodextrin scaffold
allows building highly dense carbohydrate surfaces of nano-
metric dimensions where the relative proportions of the two
glycotope constituents, D-mannose and lactose, can be strictly
controlled while keeping constant overall topologies. Never-
theless, binding efficiency might be influenced by a number of
other effects, such as subsite binding or steric congestion
between glycoligands.36,38 Actually, the ensemble of data clearly
indicates that the binding affinity of each of this sugars by a
specific lectin (Con A or PNA) is not independent of the
presence of the second sugar when both are closely packed at
high valency presentation. The effect is observable when
considering the interaction of the multivalent display with a
single lectin binding site (ELLA), the capacity to clusterize the
lectin (two-site ELLA), and the ability to promote three-
dimensional aggregates (turbidity assay).
Most relevant is that in the case of pairs of compounds
having identical valency of the matching ligand, e.g., the
heptamannosylated conjugates MandiLact-CD and Man-CD
when assayed against Con A, the derivative bearing in addition
the mismatching epitope showed enhanced lectin binding
affinity, even though steric considerations would be unfavor-
able. This strongly suggests an active participation of the
mismatching sugar in the recognition of the glycocluster. The
fact that the aggregation potential is decreased in the presence
of high concentrations of this mismatching sugar in the solution
further supports that this effect is specific in nature.
The current body of evidence does not allow unequivocal
deciphering of the mechanism at work for the heterocluster
effect. In the case of the α-D-mannopyranose−Con A associa-
tion, quantitatively significant differences have been encoun-
tered dependent on the nature of the mismatching sugar
(β-D-glucopyranose or lactose). For identical binary propor-
tions of carbohydrates, the effect is also lectin-dependent, being
more pronounced in the case of Con A as compared with PNA.
In each case, the optimal increase on binding efficiency is a
CONCLUSION
■
In this study, we implemented a synthetic methodology to
elaborate a library of molecularly well-defined multivalent
glycoclusters with a precise display of α-Man and β-Lact
antennas by using β-cyclodextrin as the central platform. The
resulting monodisperse architectures are very compact, with all
dendron branches oriented toward the same face of the βCD
cone and the sugar residues sharing an area of ca. 6 nm
diameter. These highly dense sugar patches have proven very
useful to assess the influence that different distributions of
“matching” and “mismatching” sugar ligands play in the specific
recognition by the α-D-mannopyranose- and lactose-binding
lectins Con A and PNA, respectively. The ensemble of results
supports that the vision that gives all the credit to the main
actors (the receptor and its putative binding epitope) is
probably too simplistic. Saccharidic ligands that are not
specifically recognized by a certain lectin might influence the
recognition process of the “matching” sugar motif, by
mechanisms that cannot be explained only in terms of steric
effects, when forced to share the same space region. Altogether,
these results support that the proposed heterocluster effect is not
just a curiosity restricted to Con A but can also influence the
binding mechanisms of other lectins. The differences observed
in the present study surely arise from structural differences in
the binding sites of the selected lectins. Thus, Con A is known
to possess an extended binding site, and although it cannot
accommodate galactose in the primary site, this monosacchar-
ide residue can bind to a secondary site when attached to
mannose.39 This might also happen in highly dense conju-
gates and, simultaneously, facilitate sliding of the mannosyl
ligands.36,40 Actually, combining primary and secondary
carbohydrate and noncarbohydrate ligands in multivalent
displays has already proven to be a very powerful strategy to
increase affinity toward biologically relevant receptors.41 Given
the fact that heterogeneity is probably the main feature of the
cellular milieu, it is very likely that the heterocluster effect is
also operative in biological systems. Further research in this
direction is currently sought in our laboratories.
EXPERIMENTAL SECTION
■
General Methods. Native and horseradish peroxidase-labeled
concanavalin A (Con A and HRP-Con A) and peanut agglutinin from
Arachis hypogaea (PNA and HRP-PNA), mannan from Saccharomyces
cerevisiae, and all other common reagents and materials were
purchased from commercial sources. Optical rotations were measured
at room temperature in 1-cm or 1-dm tubes. Infrared (IR) spectra
were recorded on a FTIR spectrophotometer. H (and 13C NMR)
1
spectra were recorded at 300 (75.5 for 13C) and 500 (125.7 for 13C)
1
1
MHz instruments. 1D H TOCSY, 2D COSY, and H−13C HMQC
1280
dx.doi.org/10.1021/jo201797b | J. Org. Chem. 2012, 77, 1273−1288

The Journal of Organic Chemistry
Article
experiments were used to assist NMR assignments. Thin-layer
chromatography (TLC) was carried out on aluminum sheets coated
with Kieselgel 60 F254, with visualization by UV light and by charring
with 10% H2SO4 or 0.2% ninhydrin. Column chromatography was
carried out on silica gel 60 (230−400 mesh). Gel permeation
chromatography (GPC) of the fully unprotected CD adducts was
carried out on a Sephadex G-25 (eluent H2O) column attached to a
fraction collector system using a UV detector set at 248 nm. The
operating conditions of FAB mass spectra were the following: the
primary beam consisted of Xe atoms with a maximum energy of 8 keV;
the samples were dissolved in thioglycerol, and the positive ions were
separated and accelerated over a potential of 7 keV; NaI was added as
cationising agent. MALDI-TOF mass spectra were acquired on a
spectrometer operating in the positive-ion mode with an accelerating
voltage of 28 keV. Samples were dissolved in H2O at mM
concentration and mixed with a standard solution of 2,5-dihydroxy-
benzoic acid (DHB; 10 mg/mL in 10% aq EtOH, 2 mL) in 1:1 v/v
relative proportions; 1 μL of the mixture was loaded onto the target
plate and then allowed to air-dry at room temperature. Elemental
was added. The mixture was stirred at room temperature for 3 h and
then concentrated. The resulting residue was dissolved in DCM,
washed with water, dried (MgSO4), concentrated, and purified by
column chromatography (1:2 → 1:3 petroleum ether/EtOAc). Yield:
0.20 g (71%); Rf = 0.54 (1:3 petroleum ether/EtOAc); [α]D = −9.1 (c
1.0 in DCM); IR (KBr) νmax 2105 cm−1; 1H NMR (500 MHz, CDCl3)
δ 5.30 (d, 3 H, J3′,4′ = 3.5 Hz, H-4′Lact), 5.05 (dd, 3 H, J2′,3′ = 10.5 Hz,
J1′,2′ = 7.9 Hz, H-2′Lact), 4.91 (dd, 3 H, H-3′Lact), 4.87 (t, 3 H, J1,2 = J2,3
9.9 Hz, H-2Lact), 5.16 (t, 3 H, J3,4 = 9.6 Hz, H-3Lact), 4.45 (d, 3 H,
H-1′Lact), 4.43 (d, 3 H, H-1Lact), 4.42 (dd, 3 H, J6a,6b = 12.0 Hz, J5,6a
=
=
1.0 Hz, H-6aLact), 4.09 (dd, 3 H, J6a′,6b′ = 11.8 Hz, J5′,6a′ = 7.2 Hz, H-
6a′Lact), 4.04 (dd, 3 H, J5,6b = 5.3 Hz, H-6bLact), 4.03 (dd, 3 H, J5′,6b′ = 7.2
Hz, H-6b′Lact), 3.84 (t, 3 H, H-5′Lact), 3.74 (t, 3 H, J4,5 = 9.9 Hz, H-4Lact),
3.57 (ddd, 3 H, H-5Lact), 3.39 (t, 6 H, 3JH,H = 6.0 Hz, H-3Pent), 3.25 (s,
3
2 H, CH2N3), 3.24 (s, 6 H, H-1Pent), 2.67, 2.64 (2 dt, 6 H, JH,H
=
7.0 Hz, 2JH,H = 14.0 Hz, H-5Pent), 1.92−2.12 (7 s, 63 H, MeCO), 1.80
(m, 6 H, H-4Pent); 13C NMR (125.7 MHz, CDCl3) δ 170.3−169.1
(CO), 101.1 (C-1′Lact), 83.8 (C-1Lact), 76.7 (C-5Lact), 76.2 (C-4Lact),
73.8 (C-3Lact), 71.0 (C-3′Lact), 70.7 (C-5′Lact), 70.4 (C-2Lact), 69.6 (C-
1Pent), 69.5 (C-3Pent), 69.1 (C-2′Lact), 66.6 (C-4′Lact), 62.2 (C-6Lact),
60.8 (C-6′Lact), 52.0 (CH2N3), 45.4 (Cq), 30.0 (C-4Pent), 27.4 (C-
5Pent), 21.0−20.5 (MeCO); FABMS m/z 2262 [M + Na]+. Anal. Calcd
for C92H131N3O54S3: C 49.35, H 5.90, N 1.88. Found: C 49.29, H 5.84,
N 1.76.
́
analyses were performed at the Instituto de Investigaciones Quimicas
(Sevilla, Spain).
Azo-bis(isobutyronitrile), dichloromethane, trifluoromethanesul-
fonic anhydride, triphenylphosphine, p-toluensulfonic chloride, and
N,N-dimethylformamide are indicated by the acronyms AIBN, DCM,
Tf2O, TPP, TsCl, and DMF, respectively. 2,3,6,2′,3′,4′,6′-Hepta-O-
acetyl-1-thio-β-lactose (3) and 2,3,4,6-tetra-O-acetyl-1-thio-α-D-
mannopyranose (4) were prepared from the corresponding per-O-acetates
in three steps by transformation into the corresponding glycosyl
halides, treatment with thiourea, and subsequent hydrolysis of the
resulting isothiouronium salt with potassium metabisulfite
(K2S2O5).23,24 Heptaamine 222 and allylated derivatives 121 and 1926
were prepared following the reported procedures. Mannosylated
building dendrons 6, 8, 10, 24−27, Man-OH, and triMan-OH and
CD conjugates 29, 33, Man-CD, and triMan-CD were obtained as
previously described.16 The synthesis of the lactosylated acrylamide
polymer used as reference in ELLA experiments with PNA-HRP is
depicted in Scheme S1 in the Supporting Information.
2,2,2-Tris[5-(2,3,6,2′,3′,4′,6′-hepta-O-acetyl-β-lactosylthio)-
2-oxapentyl]ethanol (5). A solution of 1 (124 mg, 0.48 mmol),
3 (1.9 g, 2.9 mmol, 2 equiv) and AIBN (48 mg, 0.3 mmol) in dry
dioxane (20 mL) was stirred under Ar at 75 °C for 3 h. Cyclohexene
(1.0 mL) was then added, the solvents were removed under reduced
pressure, and the residue was purified by column chromatography
using 1:2 → 1:3 petroleum ether/EtOAc and then EtOAc as eluent.
Yield: 0.76 g (71%); Rf1= 0.20 (1:3 petroleum ether/EtOAc); [α]D =
−6.4 (c 1.0 in DCM); H NMR (500 MHz, CDCl3) δ 5.30 (d, 3 H,
J3′,4′ = 3.4 Hz, H-4′Lact), 5.16 (t, 3 H, J2,3 = J3,4 = 9.7 Hz, H-3Lact), 5.05
(dd, 3 H, J2′,3′ = 10.4 Hz, J1′,2′ = 8.0 Hz, H-2′Lact), 4.91 (dd, 3 H, H-
3′Lact), 4.87 (t, 3 H, J1,2 = 9.7 Hz, H-2Lact), 4.46 (d, 3 H, H-1′Lact), 4.44
(d, 3 H, H-1Lact), 4.43 (dd, 3 H, J6a,6b = 10.0 Hz, J5,6a = 2.3 Hz, H-
6aLact), 4.09 (dd, 3 H, J6a′,6b′ = 12.5 Hz, J5′,6a′ = 7.0 Hz, H-6a′Lact), 4.05
(dd, 3 H, J5,6b = 5.2 Hz, H-6bLact), 4.04 (dd, 3 H, J5′,6b′ = 7.0 Hz,
2,2,2-Tris[5-(2,3,6,2′,3′,4′,6′-hepta-O-acetyl-β-lactosylthio)-
2-oxapentyl]ethyl Isothiocyanate (9). To a solution of azide 7
(0.44 g, 0.2 mmol) in dry dioxane (10 mL) were added TPP (57 mg,
0.22 mmol) and CS2 (0.12 mL, 1.98 mmol) under Ar. The solution
was stirred at room temperature for 24 h. Then the solvents were
evaporated, and the residue was purified by column chromatography
using 1:2 petroleum ether/EtOAc as eluent. Yield: 0.44 g (95%); Rf =
0.38 (1:3 petroleum ether/EtOAc); [α]D = −10.5 (c 1.0 in DCM); IR
(KBr) νmax 2191, 2108 cm−1; 1H NMR (500 MHz, CDCl3) δ 5.32 (d,
3 H, J3′,4′ = 3.0 Hz, H-4′Lact), 5.20 (t, 3 H, J2,3 = J3,4 = 9.9 Hz, H-3Lact),
5.09 (dd, 3 H, J2′,3′ = 10.5 Hz, J1′,2′ = 7.7 Hz, H-2′Lact), 4.95 (dd, 3 H, H-
3′Lact), 4.92 (t, 3 H, J1,2 = 9.9 Hz, H-2Lact), 4.58 (3 H, dd, J6a,6b = 12.0
Hz, J5,6a = 1.5 Hz, H-6aLact), 4.48 (d, 3 H, H-1′Lact), 4.47 (d, 3 H, H-
1Lact), 4.11 (3 H, dd, J6′a,6′b = 11.5 Hz, J5′,6′a = 7.5 Hz, H-6′aLact), 4.06
(3 H, dd, J5,6b = 5.0 Hz, H-6bLact), 4.05 (3 H, dd, J5′,6′b = 8.5 Hz, H-
6′bLact), 3.87 (dd, 3 H, H-5′Lact), 3.79 (t, 3 H, J4,5 = 9.5 Hz, H-4Lact),
3.61 (ddd, 3 H, H-5Lact), 3.56 (s, 2 H, CH2NCS), 3.44 (t, 6 H, 3JH,H
5.7 Hz, H-3Pent), 3.32 (s, 6 H, H-1Pent), 2.71, 2.68 (2 dt, 6 H, JH,H
=
=
2
13.1 Hz, 3JH,H = 7.3 Hz, H-5Pent), 2.14−1.95 (6 s, 63 H, MeCO), 1.83
(m, 6 H, H-4Pent); 13C NMR (125.7 MHz, CDCl3) δ 170.1−168.9
(CO), 130.3 (NCS), 101.0 (C-1′Lact), 83.6 (C-1Lact), 76.7 (C-5Lact),
76.2 (C-4Lact), 73.8 (C-3Lact), 71.0 (C-3′Lact), 70.7 (C-5′Lact), 70.4 (C-
2Lact), 69.6 (C-1Pent), 69.5 (C-3Pent), 69.1 (C-2′Lact), 66.6 (C-4′Lact),
62.2 (C-6Lact), 60.8 (C-6′Lact), 45.8 (CH2NCS), 45.7 (Cq), 29.8
(C-4Pent), 27.2 (C-5Pent), 20.8−20.3 (MeCO); FABMS m/z 2278
[M + Na]+. Anal. Calcd for C93H131NO54S4: C 49.53, H 5.85, N 0.62.
Found: C 49.40, H 5.66, N 0.60.
2,2,2-Tris(5-β-lactosylthio-2-oxapentyl)ethanol (triLact-OH).
H-6b′Lact), 3.84 (t, 3 H, H-5′Lact), 3.74 (t, 3 H, J4,5 = 9.7 Hz, H-4Lact),
3
́
Conventional Zemplen deacetylation of 5 (0.20 g, 90 μmol) gave
3.58 (ddd, 3 H, H-5Lact), 3.56 (s, 2 H, CH2OH), 3.40 (t, 6 H, JH,H
=
triLact-OH. Yield: 107 mg (90%); Rf = 0.13 (6:3:1 MeCN/H2O/
5.8 Hz, H-3Pent), 3.34 (s, 6 H, H-1Pent), 2.70 (bs, 1 H, OH), 2.67, 2.64
1
3
2
NH4OH); [α]D = −11.0 (c 1.0 in H2O); H NMR (500 MHz, D2O)
(2 dt, 6 H, JH,H = 6.0 Hz, JH,H = 13.0 Hz, H-5Pent), 2.11−1.92 (7 s,
63 H, MeCO), 1.79 (m, 6 H, H-4Pent); 13C NMR (125.7 MHz, CDCl3)
δ 170.3−169.1 (CO), 101.1 (C-1′Lact), 83.7 (C-1Lact), 76.7 (C-5Lact),
76.2 (C-4Lact), 73.8 (C-3Lact), 71.2 (C-1Pent), 71.0 (C-3′Lact), 70.7 (C-
5′Lact), 70.3 (C-2Lact), 69.7 (C-3Pent), 69.1 (C-2′Lact), 66.6 (C-4′Lact), 65.4
(C-1Pent), 62.2 (C-6Lact), 60.8 (C-6′Lact), 45.0 (Cq), 29.9 (C-4Pent), 27.3
(C-5Pent), 21.0−20.5 (MeCO); FABMS m/z 2237 [M + Na]+. Anal.
Calcd for C92H132O55S3: C 49.90, H 6.01. Found: C 49.92, H 6.04.
2,2,2-Tris[5-(2,3,6,2′,3′,4′,6′-hepta-O-acetyl-β-lactosylthio)-
2-oxapentyl]ethyl Azide (7). To a solution of alcohol 5 (0.28 g,
0.13 mmol) in dry DCM (1.5 mL) were added pyridine (60 μL) and
Tf2O (26 μL, 0.16 mmol) under Ar at −25 °C. The solution was
stirred for 40 min at −25 °C, diluted with DCM (1.0 mL), washed with
cold saturated aqueous NaHCO3, dried (MgSO4), and concentrated.
The residue was dissolved in DMF (3 mL), and NaN3 (25 mg, 0.38 mmol)
δ 4.48 (d, 3 H, J1,2 = 10.0 Hz, H-1Lact), 4.38 (d, 3 H, J1′,2′ = 8.0 Hz,
H-1′Lact), 3.89 (dd, 3 H, J6a,6b = 12.05 Hz, H-6aLact), 3.85 (d, 3 H,
H-4′Lact), 3.73 (dd, 3 H, H-6bLact), 3.71 (dd, 3 H, J6′a,6′b = 11.5 Hz, J5′,6′a
3.0 Hz, H-6′aLact), 3.67 (dd, 3 H, J5′,6ab′ = 3.5 Hz, H-6′bLact), 3.64 (m,
3 H, H-5′Lact), 3.60 (dd, 3 H, J3′,4′ = 3.5 Hz, H-3′Lact), 3.59 (t, 3 H, J4,5
=
=
9.0 Hz, H-4Lact), 3.58 (t, 3 H, J3,4 = 9.0 Hz, H-3Lact), 3.53 (m, 6 H,
H-3Pent), 3.52 (ddd, 3 H, J5,6a = 2.0 Hz, J5,6b = 4.5 Hz, H-5Lact), 3.50
(s, 2 H, CH2OH), 3.47 (dd, 3 H, J2′,3′ = 10.0 Hz, H-2′Lact), 3.38 (m, 6
H, H-1Pent), 3.30 (dd, 3 H, J2,3 = 9.0 Hz, H-2Lact), 2.77, 2.73 (2 dt, 6 H,
J4′,5′ = 7.0 Hz, J5a′,5b′ = 14.0 Hz, H-5Pent), 1.86 (m, 6 H, H-4Pent);
13C NMR (75.5 MHz, D2O) δ 102.8 (C-1′Lact), 85.3 (C-1Lact), 78.6
(C-5Lact), 78.0 (C-4Lact), 75.7 (C-3Lact), 75.3 (C-5′Lact), 72.4 (C-3′Lact),
71.9 (C-2Lact), 70.8 (C-2′Lact), 69.9 (C-3Pent), 69.2 (C-1Pent), 68.5 (C-
4′Lact), 61.3 (CH2OH), 60.9 (C-6′Lac), 60.1 (C-6Lac), 44.9 (Cq), 29.1
1281
dx.doi.org/10.1021/jo201797b | J. Org. Chem. 2012, 77, 1273−1288

The Journal of Organic Chemistry
Article
(C-4Pent), 26.7 (C-5Pent); FABMS m/z 1353 [M + Na]+. Anal. Calcd
for C50H90O34S3: C, 45.10, H, 6.81. Found: C, 44.83, H, 6.55.
2-[5-(2,3,6,2′,3′,4′,6′-Hepta-O-acetyl-β-lactosylthio)-2-oxa-
pentyl]-2,2-bis(2-oxapent-4-enyl)ethanol (11) and 2,2-Bis[5-
(2,3,6,2′,3′,4′,6′-hepta-O-acetyl-β-lactosylthio)-2-oxapentyl]-2-
(2-oxapent-4-enyl)ethanol (12). A solution of 3 (1.78 g, 2.73 mmol),
1 (0.5 g, 1.95 mmol) and AIBN (128 mg, 0.78 mmol) in dry dioxane
(25 mL) was stirred under Ar at 75 °C for 1 h. Cyclohexene (2.6 mL)
was then added, the solvents were removed under reduced pressure,
and the products were separated by column chromatography using
1:2 → 1:3 petroleum ether/EtOAc and then EtOAc as eluent.
Data for 11. Yield: 0.88 g (71%); Rf = 0.49 (1:2 petroleum ether/
H-2Man), 5.26 (t, 2 H, J3,4 = J4,5 = 9.5 Hz, H-4Man), 5.20 (d, 2 H, H-
1Man), 5.19 (dd, 2 H, H-3Man), 5.16 (t, 1 H, J2,3 = J3,4 = 9.5 Hz, H-3Lact),
5.05 (dd, 1 H, J2′,3′ = 9.5 Hz, J1′,2′ = 7.5 Hz, H-2′Lact), 4.90 (dd, 1 H, H-
3′Lact), 4.87 (t, 1 H, J1,2 = H-2Lact), 4.45 (m, 2 H, H-1Lact, H-6aLact), 4.44
(d, 1 H, H-1′Lact), 4.33 (ddd, 2 H, J5,6a = 5.0 Hz, J5,6b = 2.0 Hz, H-5Man),
4.26 (dd, 2 H, J6a,6b = 12.0 Hz, H-6aMan), 4.08 (dd, 1 H, J6a′,6b′ = 11.0 Hz,
J5′,6a′ = 7.0 Hz, H-6a′Lact), 4.05 (dd, 2 H, H-6bMan), 4.04 (dd, 1 H,
J6a,6b = 12.0 Hz, H-6bLact), 4.03 (dd, 1 H, J5′,6b′ = 7.0 Hz, H-6b′Lact), 3.83
(t, 1 H, H-5′Lact), 3.75 (t, 1 H, J3,4 = 9.5 Hz, H-4Lact), 3.74 (t, 1 H, J4,5
9.5 Hz, H-4Lact), 3.59 (m, 2 H, CH2OH), 3.58 (ddd, 1 H, J5,6b
=
=
5.0 Hz, J5,6a = 2.5 Hz, H-5Lact), 3.42 (2 t, 6 H, 3JH,H = 6.5 Hz, H-3Pent),
3.35 (s, 6 H, H-1Pent), 2.66, 2.62 (2 dt, 6 H, 2JH,H = 12.5 Hz, 3JH,H = 6.5 Hz,
H-5Pent), 1.91−2.12 (8s, 45 H, MeCO), 1.80 (m, 6 H, H-4Pent); 13C
NMR (125.7 MHz, CDCl3) δ 170.0−169.1 (CO), 101.1 (C-1′Lact),
83.7 (C-1Lact), 82.7 (C-1Man), 76.7 (C-5Lact), 76.2 (C-4Lact), 73.8 (C-3Lact),
71.4, 71.3 (C-1Pent), 71.1 (C-2Man), 71.0 (C-3′Lact), 70.7 (C-5′Lact), 70.3
(C-2Lact), 69.7, 69.6 (C-3Pent), 69.4 (C-3Man), 69.1 (C-2′Lact), 69.0 (C-5Man),
66.6 (C-4′Lact), 66.3 (C-4Man), 65.4 (CH2OH), 62.4 (C-6Man), 62.2
(C-6Lact), 60.8 (C-6′Lact), 45.1 (Cq), 29.9, 29.7 (C-4Pent), 28.2, 27.3
(C-5Pent), 20.9−20.8 (MeCO); FABMS m/z 1661 [M + Na]+. Anal.
Calcd for C68H100O39S3: C 49.87, H 6.15. Found: C 49.84, H 5.94.
2,2-Bis[5-(2,3,6,2′,3′,4′,6′-hepta-O-acetyl-β-lactosylthio)-2-
oxapentyl]-2-[5-(2,3,4,6-tetra-O-acetyl-α-D-mannopyranosylth-
io)-2-oxapentyl]ethanol (14). A solution of 12 (0.60 g, 0.39 mmol),
4 (0.28 g, 0.78 mmol) and AIBN (12 mg, 77 μmol) in dry dioxane
(8.4 mL) was stirred under Ar at 75 °C for 30 min. Cyclohexene (0.26 mL)
was added, the solvents were removed under reduced pressure, and
the residue was purified by column chromatography using 1:2 →
1:3 petroleum ether/EtOAc as eluent. Yield: 0.31 g (42%); Rf = 0.19
(1:3 petroleum ether/EtOAc); [α]D = +9.7 (c 1.2 in DCM); 1H NMR
(500 MHz, CDCl3) δ 5.30 (d, 2 H, J3′,4′ = 3.5 Hz, H-4′Lact), 5.29 (dd, 1
H, J2,3 = 3.5 Hz, J1,2 = 1.5 Hz, H-2Man), 5.27 (t, 1 H, J3,4 = J4,5 = 9.5 Hz,
H-4Man), 5.20 (bs, 1 H, H-1Man), 5.20 (dd, 1 H, H-3Man), 5.17 (t, 2 H,
J2,3 = J3,4 = 9.5 Hz, H-3Lact), 5.06 (dd, 2 H, J2′,3′ = 10.5 Hz, J1′,2′ = 8.0 Hz,
H-2′Lact), 4.91 (dd, 2 H, H-3′Lact), 4.88 (t, 2 H, J1,2 = 9.5 Hz, H-2Lact),
4.45 (d, 2 H, H-1′Lact), 4.43 (m, 4 H, H-1Lact, H-6aLact), 4.33 (ddd, 1 H,
J5,6a = 5.5 Hz, J5,6b = 2.0 Hz, H-5Man), 4.28 (dd, 1 H, J6a,6b = 12.5 Hz, H-
6aMan), 4.09 (dd, 2 H, J6a′,6b′ = 11.5 Hz, J5′,6a′ = 7.0 Hz, H-6a′Lact), 4.06
(dd, 1 H, J5,6b = 5.0 Hz, H-6bMan), 4.05 (dd, 2 H, J6a,6b = 12.0 Hz, H-
6bLact), 4.04 (dd, 2 H, J5′,6b′ = 7.0 Hz, H-6b′Lact), 3.84 (t, 2 H, H-5′Lact),
3.75 (t, 2 H, J4,5 = 9.5 Hz, H-4Lact), 3.59 (m, 2 H, CH2OH), 3.58 (ddd,
1
EtOAc); [α]D = −9.4 (c 1.0 in DCM); H NMR (500 MHz, CDCl3)
δ 5.81 (ddt, 2 H, OCH2CH), 5.30 (d, 1 H, J3′,4′ = 3.5 Hz, H-4′Lact),
5.24 (dq, 2 H, 3JH,H = 17.5 Hz, CHa), 5.22 (dq, 2 H, 3JH,H = 10.5 Hz,
2JH,H = 1.5 Hz, CHb), 5.15 (t, 1 H, J2,3 = J3,4 = 9.0 Hz, H-3Lact), 5.04
(dd, 1 H, J1′,2′ = 8.0 Hz, H-2′Lact), 4.90 (dd, 1 H, J2′,3′ = 10.5 Hz,
H-3′Lact), 4.87 (t, 1 H, J1,2 = 9.5 Hz, H-2Lact), 4.43 (d, 1 H, H-1′Lact),
4.41 (d, 1 H, H-1Lact), 4.40 (dd, 1 H, J6a,6b = 11.5 Hz, J5,6a = 1.9 Hz,
H-6aLact), 4.07 (dd, 1 H, J6a′,6b′ = 11.5 Hz, J5′,6a′ = 7.5 Hz, H-6a′Lact), 4.04
(dd, 1 H, J5,6b = 5.0 Hz, H-6bLact), 4.03 (dd, 1 H, J5′,6b′ = 7.5 Hz,
3
4
H-6b′Lact), 3.89 (dt, 4 H, JH,H = 5.5 Hz, JH,H = 1.2 Hz, CH2CH),
3.82 (t, 1 H, H-5′Lact), 3.73 (t, 1 H, J4,5 = 9.5 Hz, H-4Lact), 3.64 (m,
3
2 H, CH2OH), 3.57 (ddd, 1 H, H-5Lact), 3.42 (t, 2 H, JH,H = 6.0 Hz,
H-3Pent), 3.41 (m, 6 H, H-1Pent, CH2OAll), 2.82 (bt, 1 H, OH), 2.68,
3
2
2.62 (2 dt, 2 H, JH,H = 6.5 Hz, JH,H = 13.0 Hz, H-5Pent), 1.91−2.10
(7 s, 21 H, MeCO), 1.79 (m, 2 H, H-4Pent); 13C NMR (125.7 MHz,
CDCl3) δ 170.8−168.6 (CO), 134.7 (CH), 116.9 (CH2), 101.2
(C-1′Lact), 83.8 (C-1Lact), 76.6 (C-5Lact), 76.3 (C-4Lact), 73.7 (C-3Lact),
72.4 (OCH2CH), 71.3 (CH2OAll, C-1Pent), 70.9 (C-3′Lact), 70.6
(C-5′Lact), 70.2 (C-2Lact), 69.6 (C-3Pent), 69.0 (C-2′Lact), 66.5 (C-4′Lact),
65.8 (CH2OH), 62.2 (C-6Lact), 60.8 (C-6′Lact), 44.8 (Cq), 29.8
(C-4Pent), 27.4 (C-5Pent), 21.0−20.6 (MeCO); FABMS m/z 931
[M + Na]+. Anal. Calcd for C40H60O21S: C 52.85, H 6.65. Found:
C 52.73, H 6.69.
Data for 12: Yield: 0.76 g (25%); Rf = 0.25 (1:2 petroleum ether/
EtOAc); [α]D = −26.2 (c 1.0 in DCM); 1H NMR (500 MHz, CDCl3)
δ 5.82 (ddt, 2 H, OCH2CH), 5.31 (d, 2 H, J3′,4′ = 3.5 Hz, H-4′Lact),
3
2
5.22 (dq, 1 H, JH,H = 10.4 Hz, JH,H = 1.5 Hz, CHb), 5.18 (t, 2 H,
3
J2,3 = J3,4 = 9.5 Hz, H-3Lact), 5.14 (dq, 1 H, JH,H = 17.2 Hz, CHa),
5.07 (dd, 2 H, J1′,2′ = 8.0 Hz, H-2′Lact), 4.91 (dd, 2 H, J2′,3′ = 10.5 Hz,
H-3′Lact), 4.90 (t, 2 H, J1,2 = 9.5 Hz, H-2Lact), 4.45 (dd, 2 H, J6a,6b = 12.0 Hz,
J5,6a = 1.5 Hz, H-6aLact), 4.44 (d, 2 H, H-1′Lact), 4.43 (d, 2 H,
H-1Lact), 4.10 (dd, 2 H, J6a′,6b′ = 12.0 Hz, J5′,6a′ = 7.5 Hz, H-6a′Lact), 4.05
(dd, 2 H, J5,6b = 5.0 Hz, H-6bLact), 4.04 (dd, 2 H, J5′,6b′ = 7.5 Hz,
3
2 H, J5,6a = 2.0 Hz, H-5Lact), 3.41 (2 t, 6 H, JH,H = 6.0 Hz, H-3Pent),
3.35 (s, 6 H, H-1Pent), 2.69, 2.63 (2 dt, 6 H, 2JH,H = 13.0 Hz, 3JH,H = 6.5
Hz, H-5Pent), 1.92−2.13 (10 s, 54 H, MeCO), 1.81 (m, 6 H, H-4Pent);
13C NMR (125.7 MHz, CDCl3) δ 170.0−169.1 (CO), 101.1 (C-1′Lact),
83.7 (C-1Lact), 82.7 (C-1Man), 76.7 (C-5Lact), 76.2 (C-4Lact), 73.8 (C-3Lact),
3
4
H-6b′Lact), 3.92 (dt, 2 H, JH,H = 5.5 Hz, JH,H = 1.3 Hz, CH2CH),
3.84 (t, 2 H, H-5′Lact), 3.75 (t, 2 H, J4,5 = 9.5 Hz, H-4La3ct), 3.63 (s, 2 H,
CH2OH), 3.59 (ddd, 2 H, H-5Lact), 3.43 (t, 4 H, JH,H = 6.0 Hz,
H-3Pent), 3.40 (m, 6 H, H-1Pent, CH2OAll), 2.88 (bs, 1 H, OH), 2.69,
71.3, 71.1 (C-1Pent), 71.2 (C-2Man), 71.0 (C-3′Lact), 70.7 (C-5′Lact
)
,
70.3 (C-2Lact), 69.7, 69.6 (C-3Pent), 69.5 (C-3Man), 69.1 (C-2′Lact),
69.0 (C-5Man), 66.6 (C-4′Lact), 66.3 (C-4Man), 65.4 (CH2OH), 62.4
(C-6Man), 62.2 (C-6Lact), 60.8 (C-6′Lact), 45.0 (Cq), 29.9, 29.5
(C-4Pent), 28.2, 27.3 (C-5Pent), 21.0−20.8 (MeCO); FABMS m/z
1947 [M + Na]+. Anal. Calcd for C80H116O47S3: C 49.89, H 6.07.
Found: C 49.86, H 5.96.
3
2
2.66 (2 dt, 4 H, JH,H = 7.0 Hz, JH,H = 13.5 Hz, H-5Pent), 2.13−1.94
(7 s, 42 H, MeCO), 1.81 (m, 4 H, H-4Pent); 13C NMR (125.7 MHz,
CDCl3) δ 171.3−170.9 (CO), 134.7 (CH), 116.6 (CH2), 101.1
(C-1′Lact), 83.8 (C-1Lact), 76.7 (C-5Lact), 76.2 (C-4Lact), 73.8 (C-3Lact),
72.4 (OCH2CH), 71.2, 71.0 (CH2OAll, C-1Pent), 70.8 (C-3′Lact),
70.7 (C-5′Lact), 70.3 (C-2Lact), 69.7 (C-3Pent), 69.1 (C-2′Lact), 66.6
(C-4′Lact), 65.8 (CH2OH), 62.2 (C-6Lact), 60.8 (C-6′Lact), 44.9 (Cq),
29.9 (C-4Pent), 27.3 (C-5Pent), 20.8−20.5 (MeCO); FABMS m/z 1583
[M + Na]+. Anal. Calcd for C66H96O38S2: C 50.76, H 6.20. Found:
C 50.66, H 6.01.
2-[5-(2,3,6,2′,3′,4′,6′-Hepta-O-acetyl-β-lactosylthio)-2-oxa-
pentyl]-2,2-bis[5-(2,3,4,6-tetra-O-acetyl-α-D-mannopyrano-
sylthio)-2-oxapentyl]ethyl Azide (15). Compound 15 was pre-
pared from 13 (0.34 g, 0.21 mmol) by triflation with Tf2O (43 μL,
0.26 mmol) in dry DCM (2.5 mL) in the presence of pyridine (100 μL)
and displacement of the resulting triflate with NaN3 (42 mg, 0.64 mmol)
in DMF (4.0 mL), following the procedure above-described for the
preparation of 7. Yield: 0.28 g (71%); IR (KBr) νmax 2103 cm−1;
Rf = 0.46 (1:3 petroleum ether/EtOAc); [α]D = +37.0 (c 1.0 in
2-[5-(2,3,6,2′,3′,4′,6′-Hepta-O-acetyl-β-lactosylthio)-2-oxa-
pentyl]-2,2-bis[5-(2,3,4,6-tetra-O-acetyl-α-D-mannopyrano-
sylthio)-2-oxapentyl]ethanol (13). A solution of 11 (1.25 g, 1.38
mmol), 4 (2.0 g, 5.53 mmol) and AIBN (90 mg, 0.55 mmol) in dry
dioxane (30 mL) was stirred under Ar at 75 °C for 30 min.
Cyclohexene (1.8 mL) was then added, the solvents were removed
under reduced pressure, and the residue was purified by column
chromatography using 1:2 → 1:3 petroleum ether/EtOAc as eluent.
Yield: 0.77 g (35%); Rf = 0.19 (1:3 petroleum ether/EtOAc); [α]D =
1
DCM); H NMR (500 MHz, CDCl3) δ 5.29 (d, 1 H, J3′,4′ = 3.5 Hz,
H-4′Lact), 5.28 (d, 2 H, J2,3 = 4.0 Hz, H-2Man), 5.25 (t, 2 H, J3,4 = J4,5
9.5 Hz, H-4Man), 5.20 (s, 2 H, H-1Man), 5.19 (dd, 2 H, H-3Man), 5.15
(t, 1 H, J2,3 = J3,4 = 9.5 Hz, H-3Lact), 5.03 (dd, 1 H, J2′,3′ = 10.5 Hz, J1′,2′
=
=
7.5 Hz, H-2′Lact), 4.90 (dd, 1 H, H-3′Lact), 4.86 (t, 1 H, J1,2 = 9.5 Hz,
H-2Lact), 4.43 (d, 1 H, H-1′Lact), 4.42 (d, 1 H, H-1Lact), 4.41 (dd, 1 H,
J6a,6b = 12.5 Hz, H-6aLact), 4.32 (ddd, 2 H, J4,5 = 9.5 Hz, J5,6a = 5.0 Hz,
J5,6b = 2.0 Hz, H-5Man), 4.26 (dd, 2 H, J6a,6b = 12.5 Hz, H-6aMan), 4.07
1
+20.9 (c 1.0 in DCM); H NMR (500 MHz, CDCl3) δ 5.29 (d, 1 H,
J3′,4′ = 3.0 Hz, H-4′Lact), 5.28 (dd, 2 H, J2,3 = 3.5 Hz, J1,2 = 1.5 Hz,
1282
dx.doi.org/10.1021/jo201797b | J. Org. Chem. 2012, 77, 1273−1288

The Journal of Organic Chemistry
Article
(1 H, dd, J6a′,6b′ = 11.5 Hz, J5′,6a′ = 7.0 Hz, 6a′Lact), 4.02 (1 H, dd, J5′,6b′
=
(CO), 128.5 (NCS), 101.1 (C-1′Lact), 83.7 (C-1Lact), 82.7 (C-1Man),
76.8 (C-5Lact), 76.2 (C-4Lact), 73.8 (C-3Lact), 71.2 (C-2Man), 71.0 (C-
3′Lact), 70.7 (C-5′Lact), 70.3 (C-2Lact), 69.7 (C-1Pent), 69.6 (C-3Pent), 69.5
(C-3Man), 69.1 (C-2′Lact), 69.0 (C-5Man), 66.6 (C-4′Lact), 66.3 (C-4Man),
62.4 (C-6Man), 62.2 (C-6Lact), 60.8 (C-6′Lact), 45.9 (CH2NCS), 45.8
(Cq), 29.9, 29.5 (C-4Pent), 28.3, 27.3 (C-5Pent), 20.9−20.8 (MeCO);
FABMS m/z 1702 [M + Na]+. Anal. Calcd for C69H99NO38S4: C
49.37, H 5.94, N 0.83. Found: C 49.30, H 5.72, N 0.80.
7.0 Hz, H-6b′Lact), 4.03 (2 H, dd, H-6bMan), 4.02 (1 H, dd, H-6bLact),
3.82 (t, 1 H, H-5′Lact), 3.72 (t, 1 H, J4,5 = 9.5 Hz, H-4Lact), 3.56 (ddd,
3
1 H, J5,6b = 5.5 Hz, J5,6a = 2.0 Hz, H-5Lact), 3.40 (2 t, 6 H, JH,H = 6.0
Hz, H-3Pent), 3.26 (s, 6 H, H-1Pent), 3.25 (m, 2 H, CH2N3), 2.65, 2.61
(2 dt, 6 H, 2JH,H = 12.5 Hz, 3JH,H = 5.5 Hz, H-5Pent), 1.93−2.11 (8 s, 45
H, MeCO), 1.80 (m, 6 H, H-4Pent); 13C NMR (125.7 MHz, CDCl3)
δ 170.3−169.5 (CO), 101.1 (C-1′Lact), 83.7 (C-1Lact), 82.7 (C-1Man),
76.6 (C-5Lact), 76.2 (C-4Lact), 73.8 (C-3Lact), 71.1 (C-2Man), 70.9 (C-3′Lact),
70.7 (C-5′Lact), 70.3 (C-2Lact), 69.6 (C-1Pent), 69.3 (C-3Pent, C-3Man), 69.1
(C-2′Lact), 69.0 (C-5Man), 66.6 (C-4′Lact), 66.3 (C-4Man), 62.4 (C-6Man),
62.2 (C-6Lact), 60.8 (C-6′Lact), 51.9 (CH2N3), 45.4 (Cq), 29.9, 29.5
(C-4Pent), 28.3, 27.4 (C-5Pent), 21.0−20.6 (MeCO); FABMS m/z 1684
[M + Na]+. Anal. Calcd for C68H99N3O38S3: C 49.12, H 6.00; N 2.53.
Found: C 48.84, H 5.62, N 2.47.
2,2-Bis[5-(2,3,6,2′,3′,4′,6′-hepta-O-acetyl-β-lactosylthio)-2-
oxapentyl]-2-[5-(2,3,4,6-tetra-O-acetyl-α-D-mannopyranosylth-
io)-2-oxapentyl]ethyl Azide (16). Compound 16 was prepared
from 14 (0.26 g, 0.13 mmol) by triflation with Tf2O (26 μL, 0.16 mmol)
in dry DCM (3.0 mL) in the presence of pyridine (64 μL) and
displacement of the resulting triflate with NaN3 (26 mg, 0.41 mmol) in
DMF (2.5 mL), following the procedure above-described for the
preparation of 7. Yield: 194 mg (75%); IR (KBr) νmax 2103 cm−1; Rf =
0.40 (1:3 petroleum ether/EtOAc); [α]D = +10.0 (c 1.0 in DCM); 1H
NMR (500 MHz, CDCl3) δ 5.32 (d, 2 H, J3′,4′ = 3.5 Hz, H-4′Lact), 5.31
2,2-Bis[5-(2,3,6,2′,3′,4′,6′-hepta-O-acetyl-β-lactosylthio)-2-
oxapentyl]-2-[5-(2,3,4,6-tetra-O-acetyl-α-D-mannopyranosylth-
io)-2-oxapentyl]ethyl Isothiocyanate (18). Compound 18 was
obtained by isothiocyanation of azide 16 (194 mg, 99 μmol) with TPP
(29 mg, 0.11 mmol) and CS2 (55 μL, 0.91 mmol) following the
procedure above-described for the preparation of 9. Yield: 174 mg
(89%); Rf = 0.53 (1:3 petroleum ether/EtOAc); [α]D = +8.7 (c 1.0 in
DCM); IR (KBr) νmax 2191, 2108 cm−1; 1H NMR (500 MHz, CDCl3)
δ 5.31 (d, 2 H, J3′,4′ = 3.5 Hz, H-4′Lact), 5.30 (dd, 1 H, J2,3 = 3.5 Hz,
J1,2 = 1.5 Hz, H-2Man), 5.28 (t, 1 H, J3,4 = J4,5 = 10.0 Hz, H-4Man), 5.23
(d, 1 H, H-1Man), 5.21 (dd, 1 H, H-3Man), 5.17 (t, 2 H, J2,3 = J3,4 = 10.0 Hz,
H-3Lact), 5.07 (dd, 2 H, J2′,3′ = 10.5 Hz, J1′,2′ = 8.0 Hz, H-2′Lact), 4.92
(dd, 2 H, H-3′Lact), 4.89 (t, 2 H, J1,2 = 10.0 Hz, H-2Lact), 4.46 (d, 2 H,
H-1′Lact), 4.45 (dd, 2 H, J6a,6b = 12.0 Hz, J5,6a = 2.0 Hz, H-6aLact), 4.44
(d, 2 H, H-1Lact), 4.34 (ddd, 1 H, J5,6a = 5.0 Hz, J5,6b = 2.0 Hz, H-5Man),
4.28 (dd, 1 H, J6a,6b = 12.0 Hz, H-6aMan), 4.10 (2 H, dd, J6′a,6′b = 11.0 Hz,
J5′,6′a = 7.0 Hz, H-6′aLact), 4.07 (dd, 1 H, H-6bMan), 4.06 (dd, 2 H, H-
6bLact), 4.05 (dd, 2 H, J5′,6′b = 7.0 Hz, H-6′bLact), 3.85 (t, 2 H, H-5′Lact),
3.76 (t, 2 H, J4,5 = 10.0 Hz, H-4Lact), 3.59 (ddd, 2 H, H-5Lact), 3.52 (s, 2 H,
(dd, 1 H, J2,3 = 3.4 Hz, J1,2 = 1.7 Hz, H-2Man), 5.29 (t, 1 H, J3,4 = J4,5
9.9 Hz, H-4Man), 5.23 (d, 1 H, H-1Man), 5.22 (dd, 1 H, H-3Man), 5.18 (t,
2 H, J2,3 = J3,4 = 9.4 Hz, H-3Lact), 5.07 (dd, 2 H, J2′,3′ = 10.4 Hz, J1′,2′
7.9 Hz, H-2′Lact), 4.93 (dd, 2 H, H-3′Lact), 4.89 (t, 2 H, J1,2 = 9.4 Hz,
H-2Lact), 4.46 (d, 2 H, H-1′Lact), 4.45 (d, 2 H, H-1Lact), 4.35(ddd, 1 H,
J5,6a = 5.2 Hz, J5,6b = 1.9 Hz, H-5Man), 4.45 (dd, 1 H, J6a,6b = 12.0 Hz,
J5,6a = 1.9 Hz, H-6aLact), 4.29 (dd, 1 H, J6a,6b = 12.2 Hz, H-6aMan), 4.10
(dd, 2 H, J6a′,6b′ = 11.1 Hz, J5′,6a′ = 7.3 Hz, H-6a′Lact), 4.07 (dd, 1 H,
=
3
=
CH2NCS), 3.43 (2 t, 6 H, JH,H = 6.0 Hz, H-3Pent), 3.31 (s, 7 H,
2
H-1PentMan), 3.30 (s, 4 H, H-1PentLact), 2.69, 2.65 (2 dt, 6 H, JH,H
=
14.0 Hz, 3JH,H = 7.0 Hz, H-5Pent), 2.14−1.93 (6 s, 54 H, MeCO), 1.83
(m, 6 H, H-4Pent); 13C NMR (125.7 MHz, CDCl3) δ 170.6−169.1
(CO), 130.3 (NCS), 101.1(C-1′Lact), 83.7 (C-1Lact), 82.7 (C-1Man),
76.7 (C-5Lact), 76.2 (C-4Lact), 73.8 (C-3Lact), 71.2 (C-2Man), 71.0 (C-3′Lact),
70.7 (C-5′Lact), 70.3 (C-2Lact), 69.7 (C-1Pent), 69.5 (C-3Pent), 69.4 (C-3Man),
69.1 (C-2′Lact), 69.0 (C-5Man), 66.6 (C-4′Lact), 66.3 (C-4Man), 62.4
(C-6Man), 62.2 (C-6Lact), 60.8 (C-6′Lact), 45.9 (CH2NCS), 45.7 (Cq),
29.8, 29.5 (C-4Pent), 28.3, 27.3 (C-5Pent), 20.9−20.7 (MeCO); FABMS
m/z 1990 [M + Na]+. Anal. Calcd for C81H115NO46S4: C 49.46, H
5.89, N 0.71. Found: C 49.23, H 5.75, N 0.70.
H-6bMan), 4.06 (dd, 2 H, J5,6b = 5.3 Hz, H-6bLact), 4.05 (dd, 2 H, J5′,6b′
=
7.3 Hz, H-6b′Lact), 3.85 (t, 2 H, H-5′Lact), 3.76 (t, 1 H, J4,5 = 9.4 Hz,
3
H-4Lact), 3.59 (ddd, 2 H, H-5Lact), 3.42 (2 t, 6 H, JH,H = 5.8 Hz, H-
3Pent), 3.28 (s, 2 H, CH2N3), 3.27, 3.26 (2 s, 6 H, H-1Pent), 2.69, 2.66
2
3
(2 dt, 6 H, JH,H = 13.7 Hz, JH,H = 7.5 Hz, H-5Pent), 1.93−2.14 (6 s,
54 H, MeCO), 1.85 (m, 6 H, H-4Pent); 13C NMR (125.7 MHz, CDCl3)
δ 170.3−169.1 (CO), 101.1 (C-1′Lact), 83.8 (C-1Lact), 82.7 (C-1Man), 76.7
(C-5Lact), 76.2 (C-4Lact), 73.8 (C-3Lact), 71.2 (C-2Man), 71.0 (C-3′Lact), 70.7
(C-5′Lact), 70.4 (C-2Lact), 69.6 (C-1Pent), 69.5 (C-3Pent, C-3Man), 69.1 (C-
2′Lact), 69.0 (C-5Man), 66.6 (C-4′Lact), 66.3 (C-4Man), 62.4 (C-6Man), 62.2
(C-6Lact), 60.8 (C-6′Lact), 52.0 (CH2N3), 45.4 (Cq), 29.9, 29.6 (C-4Pent),
28.3, 27.4 (C-5Pent), 20.9−20.6 (MeCO); FABMS m/z 1973 [M + Na]+.
Anal. Calcd for C80H115N3O46S3: C, 49.25, H, 5.94; N, 2.15. Found: C,
49.09, H, 6.05; N, 2.17.
2,2-(5-β-Lactosylthio-2-oxapentyl)-2-bis(5-α-D-mannopyra-
nosylthio-2-oxapentyl)ethanol (LactdiMan-OH). Conventional
Zemplen deacetylation of 13 (118 mg, 72 μmol) gave LactdiMan-
́
OH. Yield: 62 mg (95%); Rf = 0.17 (6:3:1 MeCN/H2O/NH4OH);
[α]D = +32.0 (c 1.0 in H2O); 1H NMR (500 MHz, D2O) δ 5.23 (bs, 2
H, H-1Man), 4.47 (d, 1 H, J1,2 = 9.9 Hz, H-1Lact), 4.37 (d, 1 H, J1,2 = 7.9
Hz, H-1′Lact), 3.98 (dd, 2 H, J2,3 = 3.1 Hz, J1,2 = 1.3 Hz, H-2Man), 3.93
(dd, 2 H, J3,4 = 9.7 Hz, H-3Man), 3.89 (dd, 1 H, J6a,6b = 12.0 Hz, J5,6a
=
1.9 Hz, H-6aLact), 3.86 (d, 1 H, J3,4 = 4.4 Hz, H-4′Lact), 3.80 (dd, 2 H,
J6a,6b = 12.3 Hz, J5,6a = 2.1 Hz, H-6aMan), 3.73 (dd, 1 H, J5,6b = 2.3 Hz,
H-6bLact), 3.70 (M, 6 H, H-5Man, H-6′aLact, H-6′bLact, H-6bMan), 3.63
(M, 5 H, H-5′Lact, H-4Man, CH2OH), 3.61, 3.59 (2 t, 2 H, J2,3 = 9.8 Hz,
J3,4 = 9.8 Hz, J4,5 = 9.7 Hz, H-3Lact, H-4Lact), 3.58 (dd, 1 H, J2,3 = 10.5
Hz, H-3′Lact), 3.54 (t, 6 H, 3JH,H = 5.7 Hz, H-3Pent), 3.52 (ddd, 1 H, H-
5Lact), 3.47 (t, 1 H, H-2′Lact), 3.37 (M, 6 H, H-1Pent), 3.31 (t, 1 H, H-
2Lact), 2.70, 2.68 (2 dt, 6 H, JH,H = 13.3 Hz, JH,H = 6.4 Hz, H-5Pent),
1.85 (M, 6 H, H-4Pent); 13C NMR (125.7 MHz, D2O) δ 103.0 (C-
1′Lact), 85.5 (C-1Lact), 85.0 (C-1Man), 78.7 (C-4Lact), 78.3 (C-5Lact), 75.9
(C-3Lact), 75.4 (C-5′Lact), 73.3 (C-3Man), 72.6 (C-3′Lact), 72.1 (C-2Lact),
71.9 (C-2Man), 71.2 (C-5Man), 71.0 (C-2′Lact), 70.0 (C-3Pent), 69.6, 69.4
(C-1Pent), 68.6 (C-4′Lact), 67.1 (C-4Man), 61.6 (CH2OH), 61.1 (C-
6′Lact), 60.9 (C-6Man), 60.3 (C-6Lact), 45.1 (Cq), 29.3 (C-4PentLact), 28.6
(C-4PentMan), 27.7 (C-5PentMan), 26.8 (C-5PentLact); FABMS m/z 1191
[M + Na]+. Anal. Calcd for C44H80O29S3: C, 45.20, H, 6.90. Found: C,
45.21, H, 6.81.
2-[5-(2,3,6,2′,3′,4′,6′-Hepta-O-acetyl-β-lactosylthio)-2-oxa-
pentyl]-2,2-bis[5-(2,3,4,6-tetra-O-acetyl-α-D-mannopyrano-
sylthio)-2-oxapentyl]ethyl Isothiocyanate (17). Compound 17
was obtained by isothiocyanation of azide 15 (0.61 g, 0.37 mmol) with
TPP (106 mg, 0.40 mmol) and CS2 (0.22 mL, 3.7 mmol) following
the procedure above-described for the preparation of 9. Yield: 0.54 g
(87%); Rf = 0.61 (1:3 petroleum ether/EtOAc); [α]D = +33.4 (c 1.0 in
DCM); IR (KBr) νmax 2191, 2106 cm−1; 1H NMR (500 MHz, CDCl3)
δ 5.32 (d, 2 H, J3′,4′ = 3.5 Hz, H-4′Lact), 5.31 (bd, 2 H, J2,3 = 3.5 Hz, H-
2Man), 5.29 (t, 2 H, J3,4 = J4,5 = 9.9 Hz, H-4Man), 5.23 (bs, 2 H, H-1Man),
5.22 (dd, 2 H, H-3Man), 5.18 (t, 1 H, J2,3 = J3,4 = 9.5 Hz, H-3Lact), 5.07
(dd, 1 H, J2′,3′ = 10.4 Hz, J1′,2′ = 8.2 Hz, H-2′Lact), 4.93 (dd, 1 H, H-
3′Lact), 4.90 (t, 1 H, J1,2 = 9.5 Hz, H-2Lact), 4.46 (d, 1 H, H-1′Lact), 4.45
(d, 1 H, H-1Lact), 4.44 (m, 1 H, H-6aLact), 4.35 (ddd, 2 H, J5,6a = 5.0
Hz, J5,6b = 2.0 Hz, H-5Man), 4.29 (dd, 2 H, J6a,6b = 12.1 Hz, H-6aMan),
4.10 (dd, 1 H, J6′a,6′b = 11.2 Hz, J5′,6′a = 6.8 Hz, H-6′aLact), 4.07 (dd, 2 H,
H-6bMan), 4.06 (dd, 1 H, J6a,6b = 12.0 Hz, J5,6b = 5.6 Hz, H-6bLact), 4.05
(dd, 1 H, J5′,6′b = 6.8 Hz, H-6′bLact), 3.85 (t, 1 H, H-5′Lact), 3.76 (t, 1 H,
J4,5 = 9.5 Hz, H-4Lact), 3.60 (ddd, 1 H, J5,6a = 1.9 Hz, H-5Lact), 3.54
2
3
2,2-Bis(5-β-lactosylthio-2-oxapentyl)-2-(5-α-D-mannopyra-
nosylthio-2-oxapentyl)ethanol (MandiLact-OH). Conventional
(m, 2 H, CH2NCS), 3.45, 3.44 (2 t, 6 H, 3JH,H = 5.9 Hz, H-3Pent), 3.32
́
Zemplen deacetylation of 14 (107 mg, 56 μmol) gave MandiLact-
2
(s, 4 H, H-1Pent), 3.31 (s, 2 H, H-1Pent), 2.69, 2.65 (2 dt, 6 H, JH,H
=
OH. Yield: 62 mg (95%); Rf = 0.17 (6:3:1 MeCN/H2O/NH4OH);
1
14.0 Hz, 3JH,H = 7.0 Hz, H-5Pent), 1.93−2.14 (6 s, 45 H, MeCO), 1.85
(m, 6 H, H-4Pent); 13C NMR (125.7 MHz, CDCl3) δ 170.3−169.1
[α]D = +32.0 (c 1.0 in H2O); H NMR (300 MHz, D2O) δ 5.25
(s, 1 H, H-1Man), 4.50 (d, 2 H, J1,2 = 9.9 Hz, H-1Lact), 4.40 (d, 2 H,
1283
dx.doi.org/10.1021/jo201797b | J. Org. Chem. 2012, 77, 1273−1288

The Journal of Organic Chemistry
Article
J1′,2′ = 7.57 Hz, H-1′Lact), 4.01(d, 1 H, J2,3 = 2.3 Hz, H-2Man), 3.93 (dd,
20.8, 20.7, 20.6, 20.5 (MeCO); FABMS m/z 945 [M + Na]+. Anal.
Calcd for C39H54O21S2: C 50.75, H 5.90, S, 6.95. Found: C 50.81,
H 5.84, S, 6.68.
1 H, J3,4 = 9.5 Hz, H-3Man), 3.91 (bdd, 2 H, J6a,6b = 12.6 Hz, J5,6a
=
1.9 Hz, H-6aLact), 3.87 (d, 2 H, J3′,4′ = 3.3 Hz, H-4′Lact), 3.83 (dd, 1 H,
J6a,6b = 12.3 Hz, J5,6a = 2.3 Hz, H-6aMan), 3.76 (bd, 2 H, H-6bLact), 3.75
(m, 1 H, H-5Man), 3.74 (d, 1 H, H-6bMan), 3.73 (d, 2 H, H-6′aLact), 3.70
(d, 2 H, H-6′bLact), 3.67 (dd, 2 H, H-5′Lact), 3.62 (t, 1 H, J4,5 = 9.5 Hz,
7-(2,3,6,2′,3′,4′,6′-Hepta-O-acetyl-β-lactosylthio)-4-oxahep-
tyl Azide (22). A mixture of 21 (0.61 g, 0.66 mmol) and NaN3
(130 mg, 2.0 mmol) in dry DMF (8 mL) was vigorously stirred at
80 °C for 2 h. The reaction mixture was concentrated, and the resulting
residue was dissolved in DCM (15 mL), washed with H2O (15 mL),
dried (MgSO4) and purified by column chromatography (3:2 petroleum
ether/EtOAc). Yield: 0.51 g (96%); Rf = 0.24 (1:1 petroleum ether/
EtOAc); [α]D = −10.2 (c 1.0 in DCM); IR (NaCl) νmax 2099 cm−1; 1H
NMR (500 MHz, CDCl3) δ 5.33 (bd, 1 H, J3′,4′ = 3.5 Hz, H-4′), 5.19
(t, 1 H, J2,3 = J3,4 = 9.5 Hz, H-3), 5.09 (dd, 1 H, J1′,2′ = 8.0 Hz, J2′,3′ = 10.
Five Hz, H-2′), 4.94 (dd, 1 H, J2′,3′ = 10.5 Hz, J3′,4′ = 3.5 Hz, H-3′), 4.92
(dd, 1 H, J1,2 = J2,3 = 9.5 Hz, H-2), 4.46 (m, 3 H, H-1, H-1′, H-6a),
4.09 (m, 3 H, H-6b, H-6′a, H-6′b), 3.86 (bt, 1 H, H-5′), 3.77 (t, 1 H,
J3,4 = J4,5 = 9.5 Hz, H-4), 3.60 (m, 1 H, H-5), 3.47 (t, 4 H, 3JH,H = 6.0
Hz, H-3Hept, H-5Hept), 3.37 (m, 2 H, CH2N3), 2.71 (m, 2 H, H-7Hept),
2.14, 2.10, 2.05, 2.04, 2.03, 1.95 (6s, 21 H, MeCO), 1.83 (m, 4 H, H-
2Hept, H-6Hept); 13C NMR (125.7 MHz, CDCl3) δ 170.3, 170.2, 170.1,
170.0, 169.7, 169.6, 169.0 (CO), 101.0 (C-1′), 83.7 (C-1), 76.7 (C-5),
76.2 (C-4), 73.8 (C-3), 71.0 (C-3′), 70.7 (C-5′), 70.3 (C-2), 69.0 (C-
2′, C-5Hept), 67.4 (C-3Hept), 66.6 (C-4′), 62.2 (C-6), 60.8 (C-6′), 48.44
(C-1Hept), 29.9 (C-6Hept), 29.1 (C-2Hept), 27.2 (C-7Hept), 20.8, 20.7,
20.6, 20.5 (MeCO); FABMS m/z 816 [M + Na]+. Anal. Calcd for
C32H47N3O18S: C 48.42, H 5.97, N 5.29, S, 4.04. Found: C 48.19,
H 5.68, N 5.02, S, 3.76.
H-4Man), 3.61 (dd, 2 H, J3′,4′ = 3.3 Hz, H-3′Lact), 3.60 (t, 2 H, J4,5
=
9.6 Hz, H-4Lact), 3.59 (t, 2 H, J2,3 = 9.6 Hz, H-3Lact), 3.56 (s, 2 H, H-
3PentMan), 3.55 (m, 2 H, H-5Lact), 3.55 (m, 4 H, H-3PentLact), 3.52 (s,
2 H, CH2OH), 3.49 (dd, 2 H, J2′,3′ = 10.0 Hz, H-2′Lac), 3.41 (s, 2 H, H-
1PentMan), 3.40 (m, 4 H, H-1PentLac), 3.33 (t, 2 H, J2,3 = 9.6 Hz, H-2Lac),
2.77 (m, 4 H, H-5PentLact), 2.70 (m, 2 H, H-5PentMan), 1.89 (s, 2 H,
H-4PentMan), 1.88 (m, 4 H, H-4PentLac); 13C NMR (75.5 MHz, D2O)
δ 102.9 (C-1′Lac), 85.4 (C-1Lac), 85.0 (C-1Man), 78.7 (C-5Lact), 78.2
(C-4Lact), 75.8 (C-3Lact), 75.4 (C-5′Lact), 73.2 (C-3Man), 72.6 (C-3′Lact),
72.1 (C-2Lact), 71.9 (C-2Man), 71.2 (C-5Man), 71.0 (C-2′Lact), 70.0 (C-
3Pent), 69.4 (C-1Pent), 68.6 (C-4′Lact), 67.0 (C-4Man), 61.5 (CH2OH),
61.0 (C-6′Lact), 60.9 (C-6Man), 60.3 (C-6Lact), 45.1 (Cq), 29.2 (C-4PentLact),
28.5 (C-4PentMan), 26.8 (C-5PentLact), 27.7 (C-5PentMan); FABMS m/z 1191
[M + Na]+. Anal. Calcd for C44H80O29S3: C, 45.20, H, 6.90. Found: C,
45.21, H, 6.81.
7-(2,3,6,2′,3′,4′,6′-Hepta-O-acetyl-β-lactosylthio)-4-oxahep-
tanol (20). A solution of 19 (0.18 g, 1.53 mmol), 3 (1.2 g, 1.83 mmol),
and AIBN (69 mg, 0.42 mmol) in dry dioxane (6 mL), under Ar, was
stirred at 75 °C for 3 h. Cyclohexene (1.4 mL) was then added, the
solvents were removed under reduced pressure and the residue was
purified by column chromatography using 100:1 → 40:1 DCM/
MeOH as eluent. Yield: 0.79 g (67%); Rf = 0.15 (1:2 petroleum ether/
EtOAc); [α]D = −123.8 (c 1.0 in DCM); 1H NMR (500 MHz,
7-(2,3,6,2′,3′,4′,6′-Hepta-O-acetyl-β-lactosylthio)-4-oxahep-
tyl Isothiocyanate (23). To a solution of azide 22 (0.46 g,
0.58 mmol) in dioxane (14 mL) were added TPP (166 mg, 0.63 mmol)
and CS2 (0.35 mL, 5.76 mmol). The reaction mixture was stirred at
room temperature under Ar for 22 h and then concentrated, and the
residue was purified by column chromatography using 1:1 petroleum
ether/EtOAc as eluent. Yield: 0.34 mg (72%); Rf = 0.20 (1:1 EtOAc/
petroleum ether); [α]D = −12.6 (c 1.0 in DCM); IR (NaCl) νmax 2119
CDCl3) δ 5.33 (d, 1 H, J3′,4′ = 3.5 Hz, H-4′), 5.19 (t, 1 H, J2,3 = J3,4
9.7 Hz, H-3), 5.09 (dd, 1 H, J1′,2′ = 8.0 Hz, J2′,3′ = 10. Five Hz, H-2′),
4.94 (dd, 1 H, J2′,3′ = 10.5 Hz, J3′,4′ = 3.5 Hz, H-3′), 4.92 (dd, 1 H, J1,2
=
=
J2,3 = 9.7 Hz, H-2), 4.46 (m, 3 H, H-1, H-1′, H-6a), 4.09 (m, 3 H, H-
6b, H-6′a, H-6′b), 3.86 (bt, 1 H, H-5′), 3.77 (t, 1 H, J3,4 = J4,5 = 9.7 Hz,
H-4), 3.73 (t, 2 H, 3JH,H = 5.9 Hz, CH2OH), 3.61 (m, 1 H, H-5), 3.58
3
1
cm−1; H NMR (500 MHz, CDCl3) δ 5.33 (dd, 1 H, J3′,4′ = 3.5 Hz,
(t, 2 H, JH,2H = 5.9 Hz, H-3Hept), 3.50 (m, 2 H, H-5Hept), 2.75, 2.69
3
(2 dt, 2 H, JH,H = 13.0 Hz, JH,H = 7.2 Hz, H-7Hept), 2.14, 2.10, 2.05,
H-4′), 5.20 (t, 1 H, J2,3 = J3,4 = 9.5 Hz, H-3), 5.10 (dd, 1 H, J1′,2′ = 8.0 Hz,
J2′,3′ = 10.5 Hz, H-2′), 4.95 (dd, 1 H, J2′,3′ = 10.5 Hz, J3′,4′ = 3.5 Hz, H-3′),
4.93 (dd, 1 H, J1,2 = J2,3 = 9.5 Hz, H-2), 4.47 (m, 3 H, H-1, H-1′, H-6a),
4.10 (m, 3 H, H-6b, H-6′a, H-6′b), 3.86 (bt, 1 H, H-5′), 3.78 (t, 1 H,
2.04, 2.03, 1.95 (6 s, 21 H, MeCO), 1.85 (m, 2 H, H-6Hept), 1.81 (m,
2 H, H-2Hept); 13C NMR (125.7 MHz, CDCl3) δ 170.3, 170.1, 170.0,
169.7, 169.1 (CO), 101.0 (C-1′), 83.6 (C-1), 76.7 (C-5), 76.2 (C-4),
73.7 (C-3), 70.9 (C-3′), 70.7 (C-5′), 70.2 (C-2), 69.8 (C-3Hept), 69.2
(C-5Hept), 69.1 (C-2′), 66.6 (C-4′), 62.2 (C-6), 61.6 (C-1Hept), 60.8
(C-6′), 32.0 (C-2Hept), 29.8 (C-6Hept), 27.1 (C-7Hept), 20.8, 20.7, 20.6,
20.5 (MeCO); FABMS m/z 791 [M + Na]+. Anal. Calcd for C32H48O19S:
C 49.99, H 6.29, S, 4.17. Found: C 49.83, H 6.18, S, 3.89.
3
J3,4 = J4,5 = 9.5 Hz, H-4), 3.63 (t, 2 H, JH,H = 6.0 Hz, CH2NCS),
3.60 (m, 1 H, H-5), 3.51 (t, 2 H, 3JH,H = 6.0 Hz, H-3Hept), 3.49 (t, 2 H,
3JH,H = 6.0 Hz, H-5Hept), 2.72 (m, 2 H, H-7Hept), 2.14, 2.13, 2.09, 2.08,
2.06, 2.05, 2.04, 2.03, 2.02 1.95 (7 s, 21 H, MeCO), 1.93 (m, 2 H, H-
2Hept), 1.85 (m, 2 H, H-6Hept); 13C NMR (125.7 MHz, CDCl3) δ 170.3,
170.2, 170.1, 170.0, 169.7, 169.6, 169.0 (CO), 101.1 (C-1′), 83.7 (C-1),
76.7 (C-5), 76.2 (C-4), 73.8 (C-3), 71.0 (C-3′), 70.7 (C-5′), 70.3 (C-2),
69.2, 69.1 (C-2′, C-5Hept), 66.8 (C-3Hept), 66.6 (C-4′), 62.2 (C-6), 60.8
(C-6′), 42.14 (C-1Hept), 30.1 (C-2Hept), 29.8 (C-6Hept), 27.2 (C-7Hept),
20.8, 20.7, 20.6, 20.5 (MeCO); FABMS m/z 832 [M + Na]+. Anal.
Calcd for C33H47NO18S2: C 48.94, H 5.85, N 1.73, S 7.92. Found: C
48.63, H 5.65, N 1.45, S, 7.61.
7-(2,3,6,2′,3′,4′,6′-Hepta-O-acetyl-β-lactosylthio)-4-oxahep-
tyl p-Toluenesulfonate (21). To a solution of 20 (0.69 g, 0.9 mmol)
in dry DCM (14 mL), tosyl chloride (0.26 g, 1.35 mmol) and DMAP
(165 mg, 1.35 mmol) were added. The reaction mixture was stirred at
room temperature for 24 h, diluted with DCM (35 mL) and washed
with H2O (50 mL). The organic layer was separated, dried (MgSO4)
and evaporated, and the residue was purified by column chromatog-
raphy (1:1 petroleum ether/EtOAc). Yield: 0.65 g (78%); Rf = 0.52
General Procedure for the Preparation of Lactose-Contain-
ing CD-Scaffolded Glycoclusters. A solution of 2 (20 mg, 11 μmol)
in H2O (1 mL) was adjusted to pH 8−9 with solid NaHCO3 and
stirred for 16 h at room temperature. A solution of the corresponding
isothiocyanate (115 μmol, 1.5 equiv) in acetone (1 mL) was then
added, and the reaction mixture was stirred at room temperature for
24−48 h. Acetone was then evaporated under reduced pressure, the
remaining aqueous suspension was freeze-dried, and the solid residue
was purified by column chromatography, using MeCN → 10:1 MeCN/
H2O as eluent, to give the corresponding hemiacetylated C7-symmetric
adducts. For the notation of atoms in NMR assignments, see Figure S1 in
Supporting Information.
Heptakis{6-deoxy-6-{2-{3-{2,2,2-tris[5-(2,3,6,2′,3′,4′,6′-
hepta-O-acetyl-β-lactosylthio)-2-oxapentyl]ethyl}thioureido}-
ethyl}thio}cyclomaltoheptaose (28). Reaction time: 48 h. Yield:
86 mg (45%), Rf = 0.53 (10:1:1 MeCN/H2O/NH4OH); [α]D = +0.1
(c 1.0 in MeOH); 1H NMR (500 MHz, MeOD, 323 K) δ 5.35 (d, 21
H, J3′,4′ = 3.0 Hz, H-4′Lact), 5.20 (t, 21 H, J2,3 = J3,4 = 8.5 Hz, H-3Lact),
1
(2:1 petroleum ether/EtOAc); [α]D = −121.4 (c 1.0 in DCM); H
NMR (500 MHz, CDCl3) δ 7.78, 7.34 (2d, 4 H, Ph), 5.33 (dd, 1 H,
J3′,4′ = 3.5 Hz, J4′,5′ = 1.0 Hz, H-4′), 5.19 (t, 1 H, J2,3 = J3,4 = 9.5 Hz, H-
3), 5.09 (dd, 1 H, J1′,2′ = 8.0 Hz, J2′,3′ = 10. Five Hz, H-2′), 4.94 (dd,
1 H, J2′,3′ = 10.5 Hz, J3′,4′ = 3.5 Hz, H-3′), 4.92 (dd, 1 H, J1,2 = J2,3
9.5 Hz, H-2), 4.46 (m, 3 H, H-1, H-1′, H-6a), 4.09 (m, 5 H, H-6b,
H-6′a, H-6′b, CH2OTs), 3.86 (bt, 1 H, H-5′), 3.78 (t, 1 H, J3,4 = J4,5
=
=
3
9.5 Hz, H-4), 3.60 (m, 1 H, H-5), 3.42 (t, 2 H, JH,H = 6.0 Hz,
H-3Hept), 3.38 (m, 2 H, H-5Hept), 2.70, 2.63 (2 dt, 2 H, 2JH,H = 13.0 Hz,
3JH,H = 7.0 Hz, H-7Hept), 2.44 (s, 3 H, MePh), 2.14, 2.13, 2.09, 2.08,
2.06, 2.05, 2.04, 2.03, 2.02 1.95 (7 s, 21 H, MeCO), 1.88 (m, 2 H,
H-6Hept), 1.78 (m, 2 H, H-2Hept); 13C NMR (125.7 MHz, CDCl3)
δ 170.3, 170.2, 170.1, 170.0, 169.7, 169.6, 169.0 (CO), 144.7, 133.2,
129.84, 127,9 (C−Ar), 101.1 (C-1′), 83.7 (C-1), 76.7 (C-5), 76.2
(C-4), 73.8 (C-3), 71.0 (C-3′), 70.7 (C-5′), 70.4 (C-2), 69.1 (C-2′,
C-5Hept), 67.6 (C-1Hept), 66.6 (C-4′), 66.1 (C-3Hept), 62.2 (C-6), 60.8
(C-6′), 29.9 (C-6Hept), 29.3 (C-2Hept), 27.2 (C-7Hept), 21.6 (MePh),
1284
dx.doi.org/10.1021/jo201797b | J. Org. Chem. 2012, 77, 1273−1288

The Journal of Organic Chemistry
Article
5.10 (bd, 21 H, J2′,3′ = 10.0 Hz, H-3′Lact), 5.00 (m, 21 H, H-2′Lact), 4.99
(m, 7 H, d, H-1), 4.87 (t, 21 H, J1,2 = J2,3 = 9.5 Hz, H-2Lact), 4.73 (m,
(m, 7 H, H-6a), 3.01 (m, 7 H, H-6b), 2.91 (m, 14 H, CH2SCyst), 2.79,
2
3
2.76 (2 dt, 42 H, JH,H = 13.5 Hz, JH,H = 7.0 Hz, H-5Pent), 2.15−2.03
(s, 315 H, MeCO), 1.93 (m, 42 H, m, H-4Pent); 13C NMR (125.7 MHz,
MeOD, 325 K) δ 182.2 (CS), 172.3−171.2 (CO), 103.7 (C-1), 102.0
(C-1′Lact), 85.8(C-4), 84.8 (C-1Lact), 84.0 (C-1Man), 78.1 (C-5Lact), 77.6
(C-4Lact), 75.6 (C-3Lact), 74.3 (C-3), 73.7 (C-2, C-5), 72.8 (C-3′Lact),
72.6 (C-2Man), 72.1 (C-2Lact, C-1Pent), 71.9 (C-5′Lact), 71.2 (C-3Man),
71.0 (C-3Pent), 70.9 (C-2′Lact), 70.6 (C-5Man), 68.8 (C-4′Lact), 67.8 (C-4Man),
63.9 (C-6Lact), 63.8 (C-6Man), 62.4 (C-6′Lact), 47.1 (Cq, CH2NCyst), 41.8
(CH2NCyst), 34.7 (C-6), 33.9 (CH2SCyst), 31.4 (C-4PentLact), 30.8
(C-4PentMan), 29.6 (C-5PentMan), 28.2 (C-5PentLact), 21.3−20.4 (MeCO).
Anal. Calcd for C539H798N14O294S35: C, 48.67, H, 6.05, N, 1.47. Found:
C, 48.40, H, 5.86, N, 1.31.
Heptakis{6-deoxy-6-{2-{N′-[7-(2,3,6,2′,3′,4′,6′-hepta-O-ace-
tyl-β-lactosylthio)-4-oxaheptyl]thioureido}ethylthio}}-
cyclomaltoheptaose (32). Reaction time: 48 h. Yield: 26 mg (37%);
Rf = 0.45 (10:1:1 MeCN/H2O/NH4OH). The compound was purified
by column chromatography using 10:1 MeCN/H2O as eluent. Com-
pound 32 was deprotected without any further characterization.
General Procedure for the Deprotection of Hemiacetylated
CD-Scaffolded Glycoclusters. Deacetylation was effected by
treatment with 1 N NaOMe in MeOH (0.1 equiv per mol of acetates)
at room temperature. After 5 min a white precipitate appeared, which was
redissolved by addition of H2O. The solution was stirred for 15 min,
neutralized using Amberlite IR-120 (H+) ion-exchange resin, demineral-
ized with Duolite MB-6113 (H+, OH−) ion-exchange resin and freeze-
dried to give the fully unprotected conjugates. Analytically pure samples
for lectin-binding studies were obtained by gel permeation chromatog-
raphy (Sephadex G-25, H2O).
21 H, H-1′Lact), 4.72 (bs, 21 H, H-1Lact), 4.50 (bd, 21 H, dd, J6a,6b
=
12.0 Hz, H-6aLact), 4.16 (m, 21 H, H-6bLact), 4.13 (m, 21 H, H-6′aLact),
4.11 (m, 21 H, dd, H-6′bLact), 3.97 (m, 7 H, H-5), 3.89 (m, 21 H, H-
4Lact), 3.84 (m, 21 H, H-5′Lact), 3.82 (m, 7 H, H-3), 3.79 (m, 21 H, H-
5Lact), 3.71 (m, 14 H, CH2NCyst), 3.52 (m, 42 H, H-3Pent), 3.47 (m,
14 H, H-2, H-4), 3.40 (m, 44 H, H-1Pent, CH2NHCyst), 3.22 (m, 7 H,
H-6a), 2.99 (m, 7 H, H-6b), 2.90 (m, 42 H, CH2SCyst), 2.77 (m, 42 H,
H-5Pent), 2.11−1.94 (m, 441 H, MeCO), 1.90 (m, 42 H, m, H-4Pent);
1D TOCSY (500 MHz, MeOD, 323 K, irradiation at H-2′Lact) δ 5.35
(d, 21 H, J3′,4′ = 3.0 Hz, H-4′Lact), 5.10 (bd, 21 H, J2′,3′ = 10.0 Hz,
H-3′Lact), 5.00 (m, 21 H, H-2′Lact), 4.73 (m, 21 H, H-1′Lact), 4.13 (m,
21 H, H-6′aLact), 4.11 (m, 21 H, H-6′bLact), 3.84 (m, 21 H, H-5′Lact);
1D-TOCSY (500 MHz, MeOD, 323 K, irradiation at H-3Lact) δ = 4.87
(t, 21 H, J1,2 = J2,3 = 9.5 Hz, H-2Lact), 4.72 (bs, 21 H, H-1Lact), 4.50 (bd,
21 H, J6a,6b = 12.0 Hz, H-6aLact), 4.16 (m, 21 H, H-6bLact), 3.89 (m, 21
H, H-4Lact), 3.79 (m, 21 H, H-5Lact); 13C NMR (125.7 MHz, MeOD,
323 K) δ 184.0 (CS), 172.0−171.0 (CO), 103.6 (C-1), 101.8
(C-1′Lact), 86.0 (C-4), 84.7 (C-1Lact), 77.9 (C-5Lact), 77.4 (C-4Lact), 75.4
(C-3Lact), 74.4 (C-3), 74.2 (C-2, C-5), 72.4 (C-3′Lact), 72.0 (C-2Lact), 71.7
(C-5′Lact), 71.0 (C-1Pent), 70.8 (C-2′Lact, C-3Pent), 68.5 (C-4′Lact), 63.8
(C-6Lact), 62.2 (C-6′Lact), 45.7 (Cq, CH2NCyst), 40.6 (CH2NCyst), 34.7
(C-6), 33.9 (CH2SCyst), 31.2 (C-4Pent), 28.5 (C-5Pent), 21.2−19.6
(MeCO). Anal. Calcd for C708H1025N14O406S35: C, 49.01, H, 5.95, N,
1.13. Found: C, 48.70, H, 5.56, N, 1.00.
Heptakis{6-deoxy-6-{2-{3-{2,2-bis{[5-(2,3,6,2′,3′,4′,6′-hepta-
O-acetyl-β-lactosylthio)-2-oxapentyl]-2-[5-(2,3,4,6-tetra-O-ace-
tyl-α-D-mannopyranosylthio)-2-oxapentyl]}ethyl}thioureido}-
ethylthio}}cyclomaltoheptaose (30). Reaction time: 16 h. Yield:
Heptakis{6-deoxy-6-{2-{N′-{2,2,2-tris[5-(β-lactosylthio)-2-
oxapentyl]ethyl}thioureido}ethylthio}}cyclomaltoheptaose
119 mg (70%), Rf = 0.57 (10:1:1 MeCN/H2O/NH4OH); [α]D
+22.5 (c 1.3 in MeOH); H NMR (500 MHz, MeOD, 313 K) δ 5.45
=
1
(triLact-CD). Yield: 56 mg (99%); [α]D = −1.0 (c 1.0 in H2O); H
1
NMR (500 MHz, D2O, 343 K) δ 5.47 (bs, 7 H, H-1), 4.89 (d, 21 H,
J1,2 = 9.0 Hz, H-1Lact), 4.83 (d, 21 H, J1′,2′ = 8.5 Hz, H-1′Lact), 4.32 (bd,
21 H, J6a,6b = 11.5 Hz, H-6aLact), 4.30 (d, 21 H, J3′,4′ = 4.5 Hz, H-4′Lact),
(m, 7 H, H-1Man), 5.34 (m, 21 H, H-4′Lact, H-2Man), 5.27 (t, 7 H, J3,4
=
J4,5 = 10.0 Hz, H-4Man), 5.10 (m, 21 H, H-3Man, H-3Lact), 5.09 (bd,
14 H, J2,3 = 9.7 Hz, H-3′Lact), 4.99 (t, 14 H, J1,2 = 8.1 Hz, H-2′Lact), 4.85
(t, 14 H, J1,2 = 9.8 Hz, H-2Lact), 4.82 (m, 7 H, H-1), 4.72 (m, 28 H,
H-1′Lact, H-1Lact), 4.50 (bd, 14 H, dd, J6a,6b = 10.6 Hz, H-6aLact), 4.31
(m, 14 H, H-5Man, H-6aMan), 4.13 (m, 63 H, H-6bMan, H-6′aLact, H-
6′bLact, H-6bLact, H-5′Lact), 3.98 (m, 7 H, H-5), 3.95 (m, 14 H, H-4Lact),
3.80 (m, 21 H, H-3, H-5Lact), 3.65 (m, 14 H, CH2NCyst), 3.54 (m, 56 H,
H-2, H-4, H-3Pent), 3.40 (m, 56 H, H-1Pent, CH2NHCyst), 3.22 (m, 7 H,
H-6a), 3.01 (m, 7 H, H-6b), 2.89 (m, 14 H, CH2SCyst), 2.78 (m, 42 H,
H-5Pent), 2.15−2.05 (m, 378 H, MeCO), 1.95 (m, 42 H, m, H-4Pent);
13C NMR (125.7 MHz, MeOD) δ 183.5 (CS), 172.3−171.2 (CO),
103.8 (C-1), 102.0 (C-1′Lact), 84.8 (C-4), 84.2 (C-1Lact), 83.6 (C-1Man),
77.9 (C-5Lact), 77.7 (C-4Lact), 75.5 (C-3Lact), 74.6 (C-3), 74.5 (C-2,
4.29 (m, 7 H, H-5), 4.18 (bd, 21 H, H-6bLact), 4.15 (m, 42 H, H-6′aLact
,
H-6′bLact), 4.12 (m, 21 H, H-5′Lact), 4.09 (bs, 14 H, CH2NCyst), 4.02
(dd, 21 H, J2′,3′ = 8.5 Hz, H-3′Lact), 4.02 (m, 42 H, H-3Lact, H-4Lact),
4.05 (m, 7 H, H-3), 3.97 (m, 42 H, H-3Pent), 3.95 (m, 21 H, H-5Lact),
3.97 (m, 7 H, H-2), 3.95 (m, 56 H, H-1Pent, CH2NCyst), 3.94 (t, 21 H,
H-2′Lact), 3.76 (t, 21 H, J2,3 = 9.0 Hz, H-2Lact), 3.54 (t, 7 H, J3,4 = J4,5
=
9.0 Hz, H-4), 3.51 (m, 7 H, H-6a), 3.32 (m, 7 H, H-6b), 3.25 (m, 14
H, CH2SCyst), 3.19 (m, 42 H, H-5Pent), 2.29 (m, 42 H, H-4Pent); 13C
NMR (125.7 MHz, D2O, 343 K) δ 181.8 (CS), 103.5 (C-1′Lact), 102.6
(C-1), 85.8 (C-1Lact), 84.8 (C-4), 79.2 (C-4Lact), 76.4 (C-5Lact), 75.8
(C-5′Lact, C-3Lact), 73.2 (C-3′Lact), 72.9 (C-3), 72.7 (C-2Lact), 71.5 (C-
2′Lact, C-3Pent), 70.6 (C-1Pent, C-2, C-5), 69.1 (C-4′Lact), 61.4 (C-6′Lact),
61.1 (C-6Lact), 44.9 (Cq, CH2NCyst, CH2NCyst), 32.5 (C-6), 31.2
(CH2SCyst), 29.9 (C-4Pent), 27.4 (C-5Pent); MALDI-TOFMS m/z
11197 [M + K]+. Anal. Calcd for C419H731N14O259S35: C, 44.51, H,
6.60, N, 1.76. Found: C, 44.63, H, 6.35, N, 1.53.
C-5), 72.8 (C-3′Lact), 72.5 (C-2Man), 71.8 (C-2Lact, C-1Pent, C-5′Lact
)
,
71.2 (C-3Man, C-3Pent), 70.8 (C-2′Lact), 70.1 (C-5Man), 68.7 (C-4′Lact),
67.7 (C-4Man), 63.8 (C-6Lact, C-6Man), 62.3 (C-6′Lact), 45.7 (Cq,
CH2NCyst), 40.8 (CH2NCyst), 35.0 (C-6), 34.0 (CH2SCyst), 31.2 (C-
4PentLact), 30.8 (C-4PentMan), 29.5 (C-5PentMan), 28.7 (C-5PentLact), 21.3−
20.4 (MeCO). Anal. Calcd for C623H910N14O350S35: C, 48.85, H, 5.99,
N, 1.28. Found: C, 48.81, H, 5.83, N, 1.11.
Heptakis{6-deoxy-6-{2-{3-{2-{[5-(2,3,6,2′,3′,4′,6′-hepta-O-
acetyl-β-lactosylthio)-2-oxapentyl]-2,2-bis[5-(2,3,4,6-tetra-O-
acetyl-α-D-mannopyranosylthio)-2-oxapentyl]}ethyl}-
thioureido}ethylthio}}cyclomaltoheptaose (31). Reaction time:
16 h. Yield: 32 mg (22%); Rf = 0.55 (10:1:1 MeCN/H2O/NH4OH);
[α]D = +39.9 (c 1.0 in MeOH); 1H NMR (500 MHz, MeOD, 323 K)
δ 5.37 (m, 14 H, H-1Man), 5.34 (d, 7 H, J3′,4′ = 3.5 Hz, J4′,5′ = 0.0 Hz, H-
Heptakis{6-deoxy-6-{2-N′-{2,2-bis[5-(β-lactosylthio)-2-oxa-
pentyl]-2-[5-(α-D-mannopyranosylthio)-2-oxapentyl]ethyl}-
thioureido}ethylthio}}cyclomaltoheptaose (MandiLact-CD).
1
Yield: 77 mg (99%); [α]D = +35.0 (c 0.7 in H2O); H NMR (500
MHz, D2O, 333 K) δ 5.60 (s, 7 H, H-1Man), 5.40 (bs, 7 H, H-1), 4.82
(d, 14 H, J1,2 = 8.0 Hz, H-1Lact), 4.75 (d, 14 H, J1′,2′ = 7.5 Hz, H-1′Lact),
4.36 (m, 7 H, H-2Man), 4.25 (bd, 14 H, J6a,6b = 12.0 Hz, H-6aLact), 4.24
(m, 7 H, H-5Man), 4.23 (d, 14 H, J3′,4′ = 5.0 Hz, H-4′Lact), 4.19 (m,
14 H, H-5, H-3), 4.13 (m, 21 H, H-3Man. H-6aMan, H-6bMan), 4.10 (bd,
14 H, H-6bLact), 4.06 (m, 35 H, H-6′aLact, H-6′bLact, H-4Man), 4.01 (m,
4′Lact), 5.32 (bd, 14 H, J2,3 = 3.5 Hz, H-2Man), 5.25 (t, 14 H, J3,4 = J4,5
10.5 Hz, H-4Man), 5.20 (dd, 14 H, H-3Man), 5.19 (t, 7 H, J2,3 = J3,4
9.0 Hz, H-3Lact), 5.09 (dd, 7 H, J2′,3′ = 10.0 Hz, H-3′Lact), 4.99 (dd, 7 H,
J1′,2′ = 7.5 Hz, H-2′Lact), 4.87 (t, 7 H, J1,2 = 9.0 Hz, H-2Lact), 4.69 (d, 7
H, H-1′Lact), 4.68 (d, 7 H, H-1Lact), 4.48 (bd, 7 H, dd, J6a,6b = 10.9 Hz,
=
=
28 H, H-5′Lact, CH2NCyst), 3.97 (m, 7 H, H-2), 3.95 (m, 28 H, H-3Lact
,
H-4Lact), 3.94 (dd, 7 H, J2′,3′ = 10.5 Hz, H-3′Lact), 3.89 (m, 42 H,
H-3Pent), 3.88 (m, 63 H, H-1Pent, H-4, CH2NCyst), 3.87 (m, 14 H, H-5Lact),
3.85 (t, 7 H, H-2′Lact), 3.68 (t, 14 H, J2,3 = 9.5 Hz, H-2Lact), 3.53 (m,
7 H, H-6a), 3.31 (m, 7 H, H-6b), 3.24 (m, 42 H, CH2SCyst), 3.12 (m,
42 H, H-5Pent), 2.24 (m, 14 H, H-4Pent); 13C NMR (125.7 MHz, D2O,
333 K) δ 180.1 (CS), 103.5 (C-1′Lact), 102.6 (C-1), 85.9 (C-1Lact),
H-6aLact), 4.39 (bdd, 14 H, J5,6a = 5.0 Hz, H-5Man), 4.28 (bd, 14 H, J6a,6b
12.0 Hz, H-6aMan), 4.14 (bd, 7 H, H-6bMan), 4.13 (m, 21 H, H-6′aLact
=
,
H-6′bLact, H-6bLact), 3.95 (m, 7 H, H-5′Lact), 3.85 (m, 7 H, H-4Lact), 3.81
(m, 7 H, t, H-3), 3.75 (m, 7 H, H-5Lact), 3.72 (m, 14 H, CH2NCyst), 3.54
(m, 56 H, H-2, H-4, H-3Pent), 3.41 (m, 56 H, H-1Pent, CH2NHCyst), 3.21
85.6 (C-1Man), 85.0 (C-4), 79.2 (C-4Lact), 76.4 (C-5Lact), 75.8 (C-3Lact
C-5′Lact), 73.8 (C-5Man), 73.2 (C-3′Lact), 72.8 (C-3, C-2Lact), 72.5
(C-2Man), 71.9 (C-3Man), 71.5 (C-2′Lact, C-3Pent), 70.7 (C-1Pent, C-2, C-5),
,
1285
dx.doi.org/10.1021/jo201797b | J. Org. Chem. 2012, 77, 1273−1288

The Journal of Organic Chemistry
Article
69.1 (C-4′Lact), 67.6 (C-4Man), 61.5 (C-6′Lact, C-6Man), 61.1 (C-6Lact), 45.0
(Cq, CH2NCyst, CH2NCyst), 34.0 (C-6), 32.8 (CH2SCyst), 30.1 (C-4PentLact),
29.6 (C-4PentMan), 28.4 (C-5PentMan), 27.5 (C-5PentLact); MALDI-TOFMS
m/z 10114 [M + H]+. Anal. Calcd for C371H660N14O226S33: C 44.60,
H 6.66, N 1.96. Found: C, 44.11, H, 6.29, N, 1.71.
Heptakis{6-deoxy-6-{2-{N′-{2-[5-(β-lactosylthio)-2-oxapen-
tyl]-2,2-bis[5-(α-D-mannopyranosylthio)-2-oxapentyl]ethyl}-
thioureido}ethylthio}}cyclomaltoheptaose (LactdiMan-CD).
Yield: 22 mg (99%); [α]D = +48.3 (c 0.6 in H2O); 1H NMR
(500 MHz, D2O, 343 K) δ 5.66 (s, 14 H, H-1Man), 5.46 (bs, 7 H, H-1),
4.88 (d, 7 H, J1,2 = 10.0 Hz, H-1Lact), 4.82 (d, 7 H, J1′,2′ = 8.0 Hz, H-
1′Lact), 4.41 (s, 14 H, H-2Man), 4.30 (m, 14 H, H-5Man), 4.29 (bd, 7 H,
J6a,6b = 11.0 Hz, H-6aLact), 4.28 (d, 7 H, J3′,4′ = 4.5 Hz, H-4′Lact), 4.27
(m, 7 H, H-5), 4.19 (m, 14 H, H-3Man), 4.18 (bd, 7 H, H-6bLact), 4.13
(m, 14 H, H-6′aLact, H-6′bLact), 4.12 (s, 28 H, H-6aMan, H-6bMan), 4.10
(m, 21 H, H-5′Lact, CH2NCyst), 4.05 (m, 7 H, H-3), 4.03 (m, 14 H,
were evaporated and the resulting residue was purified by column
chromatography using 9:1 DCM/MeOH. Yield: 0.49 g (50%); Rf =
1
0.55 (9:1 DCM/MeOH); [α]D = +12.0 (c 1.0 in DCM); H NMR
(500 MHz, CDCl3) δ 5.33 (d, 1 H, J3′,4′ = 3.3 Hz, H-4′Lact), 5.19 (t,
1 H, J2,3 = J3,4 = 9.2 Hz, H-3Lact), 5.09 (dd, 1 H, J2′,3′ = 10.4 Hz, J1′,2′
=
7.9 Hz, H-2′Lact), 4.95 (dd, 1 H, H-3′Lact), 4.92 (t, 1 H, J1,2 = 9.2 Hz,
H-2Lact), 4.57 (dd, 1 H, J6a,6b = 11.0 Hz, J5,6a = 1.8 Hz, H-6aLact), 4.51
(d, 1 H, H-1′Lact), 4.49 (d, 1 H, H-1Lact), 4.12 (dd, 1 H, J6′a,6′b = 11.1 Hz,
J5′,6′a = 6.1 Hz, H-6′aLact), 4.08 (dd, 1 H, J5,6b = 4.0 Hz, H-6bLact), 4.07
(dd, 1 H, J5′,6′b = 7.4 Hz, H-6′bLact), 3.88 (dd, 1 H, H-5′Lact), 3.79 (t, 1 H,
J4,5 = 9.2 Hz, H-4Lact), 3.66 (ddd, 1 H, H-5Lact), 2.98 (m, 1 H, NCHa),
2.91 (m, 2 H, NCHb, SCHa), 2.76 (m, 1 H, SCHb), 2.14−1.95 (7 s,
21 H, MeCO); 13C NMR (125.7 MHz, CDCl3) δ 172.0−169.0 (CO
ester), 101.2 (C-1′Lact), 84.0 (C-1Lact), 76.9 (C-5Lact), 76.1 (C-4Lact),
73.8 (C-3Lact),71.0 (C-3′Lact), 70.9 (C-5′Lact), 70.3 (C-2Lact), 69.3 (C-2′Lact
)
,
66.68 (C-4′Lact), 62.1 (C-6 Lact), 61.0 (C-6′Lact), 41.6 (NCH2), 33.9
(SCH2), 21.1−20.7 (MeCO); FABMS m/z 718 (100%, [M + Na]+).
Anal. Calcd for C28H41NO17S: C 48.34, H 5.94, N 2.01. Found: C 49.19,
H 5.81, N 1.95.
H-4Man), 4.02 (dd, 7 H, J2′,3′ = 10.5 Hz, H-3′Lact), 4.01 (m, 14 H, H-3Lact
H-4Lact), 3.94 (m, 42 H, H-3Pent), 3.93 (m, 7 H, H-5Lact), 3.95 (m, 63 H,
H-1Pent, H-2, CH2NCyst), 3.92 (t, 7 H, H-2′Lact), 3.75 (t, 7 H, J3,4 = J4,5
,
=
9.5 Hz, H-4), 3.73 (t, 7 H, J2,3 = 9.5 Hz, H-2Lact), 3.66 (m, 7 H, H-6a),
3.21, 3.18 (2 dt, 42 H, 2JH,H = 13.0 Hz, 3JH,H = 6.0 Hz, H-5Pent), 3.38 (m,
7 H, H-6b), 3.32 (m, 42 H, CH2SCyst), 2.30 (m, 42 H, H-4Pent); 13C NMR
(125.7 MHz, D2O, 343 K) δ 182.1 (CS), 103.6 (C-1′Lact), 102.7 (C-1),
86.0 (C-1Lact), 85.7 (C-1Man), 85.2 (C-4), 79.4 (C-4Lact), 76.5 (C-5Lact),
75.9 (C-5′Lact, C-3Lact), 73.8 (C-5Man), 73.4 (C-3′Lact), 72.6 (C-3,
C-2Lact), 72.5 (C-2Man), 71.9 (C-3Man), 71.6 (C-2′Lact, C-3Pent), 70.8
(C-1Pent, C-2, C-5), 69.1 (C-4′Lact), 67.6 (C-4Man), 61.5 (C-6′Lact, C-6Man),
61.1 (C-6Lact), 44.8 (Cq, CH2NCyst), 40.5 (CH2NCyst), 34.0 (C-6), 33.1
(CH2SCyst), 30.2 (C-4PentLact), 29.8 (C-4PentMan), 28.5 (C-5PentMan),
27.6 (C-5PentLact); MALDI-TOFMS m/z 8902 [M + Na]+. Anal.
Calcd for C329H588N14O189S35: C 44.47, H 6.67, N 2.20. Found: C
44.55, H 6.43, N 2.02.
Heptakis{6-deoxy-6-{2-{N′-[7-(β-lactosylthio)-4-oxaheptyl]-
thioureido}ethylthio}}cyclomaltoheptaose (Lact-CD). Reaction
time: 48 h. Deacetylation was achieved following the general pro-
cedure above-described. Yield: 21 mg (99%); [α]D = −51.0 (c 0.5 in
H2O); 1H NMR (500 MHz, D2O, 343 K) δ 5.52 (bs, 7 H, H-1), 4.96
(d, 7 H, J1,2 = 10.0 Hz, H-1Lact), 4.88 (d, 7 H, J1′,2′ = 8.0 Hz, H-1′Lact),
4.37 (bd, 7 H, J6a,6b = 12 Hz, H-6aLact), 4.36 (bs, 7 H, H-4′Lact), 4.36−
3.90 (m, 2H, H-3, H-4); 4.32 (bs, 7 H, H-5), 4.23 (bd, 7 H, H-6bLact),
4.20 (m, 14 H, H-6′aLact, H-6′bLact), 4.14 (m, 7 H, H-5′Lact), 4.12 (bs,
14 H, CH2NCyst), 4.08 (dd, 7 H, J2′,3′ = 8.5 Hz, J3′,4′ = 4.5 Hz, H-3′Lact),
4.08 (m, 14 H, H-3Lact, H-4Lact), 4.10 (m, 7 H, H-2), 4.01 (m, 7 H, H-
5Lact), 3.96 (m, 7 H, H-2′Lact), 3.81 (t, 7 H, J2,3 = 9.0 Hz, H-2Lact),
3.60−3.30 (m, 14 H, H-6a, H-6b), 3.37 (bs, 14 H, CH2SCyst), 3.92 (bs,
14 H, H-1Hept), 2.30 (m, 14 H, H-2Hept), 4.01 (m, 14 H, H-3Hept), 4.02
(m, 14 H, H-5Hept), 2.35 (m, 14 H, H-6Hept), 3.23 (m, 14 H, H-7Hept);
13C NMR (125.7 MHz, D2O, 343 K) δ 181,0 (CS), 103.5 (C-1′Lact),
102.6 (C-1), 85.9 (C-1Lact), 85.0 (C-4), 79.2 (C-4Lact), 79.2 (C-3Lact),
76.4 (C-5Lact), 75.8 (C-5′Lact), 73.2 (C-3′Lact), 72.8 (C-2Lact), 72.7
(C-3), 71.5 (C-2′Lact), 69.8 (C-5Hept), 69.6 (C-2, C-5), 69.2 (C-4′Lact),
68.7 (C-3Hept), 61.5 (C-6′Lact), 61.1 (C-6Lact), 44.3 (CH2NCyst), 41.8
(C-1Hept), 33.9 (C-6), 32.9 (CH2SCyst), 30.1 (C-6Hept), 29.0 (C-2Hept),
27.3 (C-7Hept); MALDI-TOFMS m/z 5219 [M - 2H + 3Na]+. Anal.
Calcd for C189H336N14O105S21: C 44.01, H 6.57, N 3.80, S, 13.05.
Found: C 43.68, H 6.20, N 3.46, S, 12.70.
Synthesis of the Reference Lactosylated Polyacrylamide
Polymer for ELLA against PNA. The lactosylated polyacrylamide
polymer (S4) used as reference ligand in the competitive ELLA experi-
ments using PNA was prepared from 2,3,6,2′,3′,4′,6′-hepta-O-acetyl-1-
(2-aminooethyl)thio-β-lactose (S1) by formation of the corresponding
acrylamide derivative (S2), de-O-acetylation (→ S3), and final co-
polymerization with acrylamide. The corresponding synthetic scheme
is depicted in the Supporting Information (Scheme S1).
2,3,6,2′,3′,4′,6′-Hepta-O-acetyl-1-(2-acrylamidoethyl)thio-β-
lactose (S2). A solution of compound S1 (0.38 g, 0.55 mmol) and
triethylmine (0.32 mL, 3.2 mmol, 6 equiv) in DCM (35 mL) was
cooled to 0 °C. Acryloyl chloride (57 μL, 1.25 equiv) in DCM (15 mL)
was added dropwise while the temperature was allowed to reach
room temperature. The reaction mixture was stirred for 30 min.
Methanol (2 mL) was then added, and the reaction mixture was stirred
at room temperature for a further 1 h. Water (25 mL) was then added,
and the organic phase was then washed successively with 0.5 M HCl
(2 × 25 mL), saturated aqueous NaHCO3 and water. The dried
(MgSO4) organic phase was filtered and evaporated to dryness. The
residue was purified by column chromatography using 1:1 EtOAc/
DCM. Yield: 0.24 g (60%); Rf = 0.26 (1:1 EtOAc/DCM); [α]D
=
1
+16.0 (c 1.0 in DCM); H NMR (500 MHz, CDCl3) δ 6.29 (t, 1 H,
3JH,H = 5.1 Hz, NH), 6.27 (dd, 1 H, 3JH,H = 10.5 Hz, 2JH,H = 1.5 Hz, 
CHa), 6.12 (dd, 1 H, 3JH,H = 17.0 Hz, CH), 5.62 (dd, 1 H, CHb),
5.32 (d, 1 H, J3′,4′ = 3.3 Hz, H-4′Lact), 5.18 (t, 1 H, J2,3 = J3,4 = 9.2 Hz,
H-3Lact), 5.08 (dd, 1 H, J2′,3′ = 10.4 Hz, J1′,2′ = 5.7 Hz, H-2′Lact), 4.94
(dd, 1 H, H-3′Lact), 4.88 (t, 1 H, J1,2 = 9.8 Hz, H-2Lact), 4.57 (dd, 1 H,
J6a,6b = 12.1 Hz, J5,6a = 1.9 Hz, H-6aLact), 4.48 (d, 1 H, H-1′Lact), 4.45 (d,
1 H, H-1Lact), 4.11 (dd, 1 H, J6′a,6′b = 11.1 Hz, J5′,6′a = 6.3 Hz, H-6′aLact),
4.06 (dd, 1 H, J5′,6′b = 7.7 Hz, H-6′bLact), 4.04 (dd, 2 H, J5,6b = 5.1 Hz,
H-6bLact), 3.84 (dd, 1 H, H-5′Lact), 3.74 (t, 1 H, J3,4 = J4,5 = 9.5 Hz,
H-4Lact), 3.60 (m, 2 H, H-5Lact, NCHa), 3.43 (m, 1 H, NCHb), 2.89
(m, 1 H, SCHa), 2.77 (m, 1 H, SCHb), 2.13−1.94 (7 s, 21 H,
MeCO); 13C NMR (125.7 MHz, CDCl3) δ 170.5−169.0 (CO ester),
165.6 (CO amide), 130.6 (=CH), 126.7 (=CH2), 101.1 (C-1′Lact), 83.8
(C-1Lact), 76.8 (C-5Lact), 75.9 (C-4Lact), 73.6 (C-3Lact), 71.0 (C-3′Lact),
70.9 (C-5′Lact), 70.0 (C-2Lact), 69.1 (C-2′Lact), 66.6 (C-4′Lact), 61.7 (C-6 Lact),
60.7 (C-6′Lact), 39.1 (NCH2), 30.8 (SCH2), 20.8−20.5 (MeCO); FABMS
m/z 772 [M + Na]+. Anal. Calcd for C31H43NO18S: C, 49.66, H, 5.78;
N 1.87. Found: C, 49.50, H, 5.60, N, 1.76.
1-(2-Acrylamidoethyl)thio-β-lactose (S3). Conventional Zem-
plen
́
deacetylation of S2 (122 mg, 0.16 mmol) gave S3 (84 mg, 99%).
Rf = 0.49 (6:3:1 MeCN/H2O/NH4OH); [α]D3= −3.0 (c 1.0 in H2O);
1H NMR (500 MHz, D2O) δ 6.15 (dd, 1 H, JH,H = 10.5 Hz, JH,H
=
3
2
3
17.1 Hz, CH), 6.06 (dd, 1 H, JH,H = 0.9 Hz, JH,H = 17.1 Hz, 
CHa), 5.64 (dd, 1 H, 2JH,H = 0.9 Hz, 3JH,H = 10.1 Hz, CHb), 4.45 (d,
1 H, J1′,2′ = 9.9 Hz, H-1′Lact), 4.32 (d, 1 H, J1,2 = 7.8 Hz, H-1Lact), 3.83
(dd, 1 H, J6a,6b = 12.3 Hz, J5,6a = 1.5 Hz, H-6aLact), 3.79 (d, 1 H, J3′,4′
=
3.1 Hz, H-4′Lact), 3.66 (dd, 1 H, J5,6b = 5.4 Hz, H-6bLact), 3.63 (m, 2 H,
H-6′Lact), 3.59 (m, 1 H, H-5′Lact) 3.54 (m, 3 H, H-3Lact, H-3′Lact, H-
4Lact), 3.48 (m, 1 H, H-5Lact), 3.41 (m, 3 H, H-2Lact, NCH2), 3.23 (t,
1 H, J2,3 = J4,5 = 9.9 Hz, H-2′Lact), 2.85 (dt, 1 H, 2JH,H = 13.3 Hz, 3JH,H
=
7.0 Hz, SCHa), 2.75 (dt, 1 H, SCHb); 13C NMR (125.7 MHz, D2O)
δ 165.6 (CO), 129.9 (=CH), 127.5 (=CH2), 102.9 (C-1′Lact), 85.3
2,3,6,2′,3′,4′,6′-Hepta-O-acetyl-1-(2-aminooethyl)thio-β-lac-
tose (S1). A suspension of cysteamine hydrochloride (0.20 g, 1.8 mmol)
in dry toluene (3 mL) was sonicated until complete solution under
Ar. Lactose octaacetate (0.5 g, 0.74 mmol) and BF3·Et2O (0.46 mL,
4.39 mmol) were then added and the reaction mixture was vigorously
stirred at 50 °C for 16 h. Et3N (1.5 mL) was then added, the solvents
(C-1Lact), 78.7 (C-5Lact), 78.1 (C-4Lact), 75.7 (C-3Lact), 75.4 (C-5′Lact
)
,
72.5 (C-3′Lact), 72.0 (C-2′Lact), 71.0 (C-2Lact), 68.6 (C-4′Lact), 61.0 (C-
6′Lact), 60.2 (C-6Lact), 39.6 (NCH2), 29.2 (SCH2); FABMS m/z 478
[M + Na]+. Anal. Calcd for C17H29NO11S: C, 44.83, H 6.42, N 3.08,
S 7.04 . Found: C 44.82, H 6.19, N 2.97, S 6.98.
1286
dx.doi.org/10.1021/jo201797b | J. Org. Chem. 2012, 77, 1273−1288

The Journal of Organic Chemistry
Article
Poly[acrylamide-co-[N-[2-(β-lactosylthio)ethyl]acrylamide]
(S4). Compound S3 (15 mg, 33 μmol) and acrylamide (22 mg,
0.31 mmol) were dissolved in deoxygenated H2O (0.7 mL), and to the
solution were added ammonium persulfate (12 μL, 50 mg/mL
solution in H2O) and N,N,N′N′-tetramethylethylenediamine
(TMEDA, 4 μL, 0.027 mmol, 0.08 equiv). The reaction mixture was
stirred at 90 °C for 12 min under nitrogen. The procedure was
repeated until complete disappearance of the starting materials (TLC).
Then, the reaction mixture was cooled to room temperature, diluted
with H2O (10 mL), and lyophilized. The residue was dissolved in H2O
(20 mL), extensively dialyzed (2 kD cutoff) against pure H2O (5 ×
2 L), and lyophilized to furnish lactose polymer S4 (19 mg, 52%). 1H
NMR (300 MHz, D2O) δ 4.45 (bs, 1 H, J1,2 = 7.8 Hz, H-1′Lact), 4.30
(bd, 1 H, J1,2 = 7.8 Hz, H-1Lact), 3.84−3.75 (m, 2 H, H-6aLact, H-4′Lact),
0.25 mM of PBS. The ligands were added in serial 2-fold dilutions
(50 μL/well) in PBS and incubated at 37 °C. After 1 h, horseradish
peroxidase-labeled Con A or PNA lectin (Con A-HRP or PNA-HRP,
50 μL/well of 200-fold dilution of a 1 mg/mL stock solution in PBS)
was added to the microtiter plates, which were incubated for an
additional 1 h at 37 °C. The plates were washed with PBS, and 50 μL/well
of ABTS (0.25 mg/mL) in citrate-phosphate buffer (0.2 M, pH 4.0 with
0.015% H2O2) was added. The reactions were stopped after 30 min by
adding 50 μL/well of 1 M H2SO4, and the optical density was measured at
405 nm relative to 570 nm.
Turbidity Assay. To 50 μL of a solution of Con A or PNA lectin
in PBS (pH 7.3, containing 0.1 mM Ca2+ and 0.1 mM Mn2+) was
added the ligand of interest (50 μL of a 250 μM solution in PBS). The
time-dependent turbidity kinetics were recorded by measuring the
absorption coefficient at 490 nm at intervals of 1 min for 30 min. After
12 min, addition of mannose in the experiments run with Con A or
lactose in experiments run with PNA (100 mM, 100 μL) restored the
clear solution state. In parallel experiments, lactose (100 mM, 50 μL)
and mannose (100 mM, 2 × 50 μL), for experiments run with Con A,
or mannose (100 mM, 50 μL) and lactose (100 mM, 2 × 50 μL), for
experiments run with PNA, were sequentially added to the mixtures
after 12, 18, and 24 min. In all cases the last addition fully restored the
clear solution state.
3.70−3.20 (m, 1 H, H-2Lact, H-3Lact, H-4Lact, H-5Lact, H6bLact, H-2′Lact
,
H-3′Lact, H-5′Lact, H-6′Lact, NCH2), 2.75 (m, 2 H, SCH2), 2.20−2.00 (m,
∼10 H, CHCO), 1.65−1.35 (m, ∼20 H, CH2CHCO); 13C NMR
(75.5 MHz, D2O) δ 179.0 (CO), 103.0 (C-1′Lact), 85.5 (C-1Lact), 78.9
(C-5Lact), 78.3 (C-4Lact), 76.0 (C-3Lact), 75.6 (C-5′Lact), 72.7 (C-3′Lact),
72.1 (C-2′Lact), 71.2 (C-2Lact), 68.8 (C-4′Lact), 61.3 (C-6′Lact), 60.4
(C-6Lact), 42.3−41.8 (CHCO, NCH2), 36.1−34.2 (CH2CHCO), 29.5
(SCH2).
Enzyme-Linked Lectin Assays (ELLA). Nunc-Inmuno plates
(MaxiSorp) were coated overnight with yeast mannan or poly-
[acrylamide-co-2-(acrylamido)ethylthio-β-lactoside)] (S4) at 100 μL/
well diluted from a stock solution of 10 μg/mL in 0.01 M phosphate
buffer saline (PBS, pH 7.3 containing 0.1 mM Ca2+ and 0.1 mM Mn2+)
at room temperature. The wells were then washed three times with
300 μL of washing buffer (containing 0.05% (v/v) Tween 20)
(PBST). The washing procedure was repeated after each of the
incubations throughout the assay. The wells were then blocked with
150 μL/well of 1% BSA/PBS for 1 h at 37 °C. After washing, the wells
were filled with 100 μL of serial dilutions of peroxidase labeled lectins
(Con A and Peanut agglutinin) from 10−1 to 10−5 mg/mL in PBS and
incubated at 37 °C for 1 h. The plates were washed and 50 μL/well of
2,2′-azinobis-(3-ethylbenzothiazoline-6-sulfonic acid) diammonium
salt (ABTS) (0.25 mg/mL) in citrate buffer (0.2 M, pH 4.0 with
0.015% H2O2) was added. The reaction was stopped after 20 min by
adding 50 μL/well of 1 M H2SO4 and the absorbances were measured
at 415 nm. Blank wells contained citrate-phosphate buffer. The
concentration of lectin displaying absorbances between 0.8 and 1.0 was
used for inhibition experiments. In order to carry out the inhibition
experiments, each inhibitor was added in a serial of 2-fold dilutions
(60 μL/well) in PBS with 60 μL of the desired lectin-peroxidase
conjugate concentration on Nunclon (Delta) microtiter plates and
incubated for 1 h at 37 °C. The above solutions (100 μL) were then
transferred to the mannan-coated microplates, which were incubated
for 1 h at 37 °C. The plates were washed, and the ABTS substrate was
added (50 μL/well). Color development was stopped after 20 min,
and the absorbances were measured. The percentage of inhibition was
calculated as follows:
Data Analysis. All experiments were done in quintuplicate (ELLA)
or triplicate (two-site ELLA and turbidity assay). Significance testing
using Student’s t test and regression analysis was done using Anova
software. Significance was defined with P < 0.001.
ASSOCIATED CONTENT
* Supporting Information
■
S
Scheme S1 depicting the reaction sequence for the synthesis of
the lactosylated acrylamide polymer used in ELLA, Figure S1
with the notation used for NMR in the case of compound 32,
and Figures S2−S30 reproducing the 1H and 13C NMR spectra
of newly synthesized compounds are provided. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Email: jljb@us.es; juanmab@iiq.csic.es.
■
ACKNOWLEDGMENTS
This work was supported by the Spanish Ministerio de Innovacion
y Ciencia (MICINN, contract numbers SAF2010-15670, and
CTQ2010-15848), the Junta de Andalucia, and the European
■
́
́
Union (FEDER and FSE). We also thank the CITIUS for
technical support.
REFERENCES
■
(1) (a) Gabius, H. Biochim. Biophys. Acta 2002, 1572, 163−164.
(b) Bertozzi, C. R.; Kiessling, L. L. Science 2001, 291, 2357−2364.
(c) Murrey, H. E.; Hsieh-Wilson, L. C. Chem. Rev. 2008, 108, 1708−
1731. (d) Collins, B. E.; Paulson, J. C. Curr. Opin. Chem. Biol. 2004, 8,
617−625.
% Inhibition = (A
/A
− A )
(with inhibitor)
(no inhibitor)
× 100.
(no inhibitor)
Results in triplicate were used for plotting the inhibition curves for
each individual ELLA experiment. Typically, the IC50 values
(concentration required for 50% inhibition of the lectin coating
polysaccharide association) obtained from several independently
performed tests were in the range of 15%. Nevertheless, the relative
inhibition values calculated from independent series of data were
highly reproducible.
Two-Site ELLA (Sandwich Assay). For Con A, Nunc-Inmuno
plates (MaxiSorp) microtitration plates were coated with yeast mannan
and blocked with BSA as above-described. Con A lectin was then added at
100 μL/well from a stock solution of 5 μg/mL in 0.01 M phosphate
buffer (PBS, pH 7.3) for 2 h at 37 °C. In the case of PNA, the plates were
coated directed with the lectin and then blocked with BSA. The
synthesized multivalent glycoclusters were used as stock solutions of
(2) Lundquist, J. J.; Toone, E. J. Chem. Rev. 2002, 102, 555−578.
(3) Lee, Y. C.; Lee, R. T. Acc. Chem. Res. 1995, 28, 321−327.
(4) For recent comprehensive reviews of several aspects of the
glycoside cluster effect, see: (a) Chabre, Y. M.; Roy, R. Adv. Carbohydr.
Chem. Biochem. 2010, 63, 165−393. (b) Lahmann, M. Top. Curr.
Chem. 2009, 288, 17−65. (c) Jayaraman, N. Chem. Soc. Rev. 2009, 38,
3463−3483. (d) Deniaud, D.; Julienne, K.; Gouin, S. G. Org. Biomol.
Chem. 2011, 9, 966−979.
(5) (a) Gupta, G.; Surolia, A.; Sampathkumar, S. Omics 2010, 14,
419−436. (b) Gemeiner, P.; Mislovicova, D.; Tkac, J.; Svitel, J.;
Paetoprsty, V.; Hrabarova, E.; Kogan, G.; Kozar, T. Biotechnol. Adv.
2009, 27, 1−15. (c) McNicholas, S.; Rencurosi, A.; Lay, L.; Mazzaglia, A.;
Sturiale, L.; Perez, M.; Darcy, R. Biomacromolecules 2007, 8, 1851−1857.
1287
dx.doi.org/10.1021/jo201797b | J. Org. Chem. 2012, 77, 1273−1288

The Journal of Organic Chemistry
Article
(d) Kizhakkedathu, J. N.; Creagh, A. L.; Shenoi, R. A.; Rossi, N. A. A.;
Brooks, D. E.; Chan, T.; Lam, J.; Dandepally, S. R.; Haynes, C. A.
Biomacromolecules 2010, 11, 2567−2575. (e) Hoai, N.; Sasaki, A.;
Sasaki, M.; Kaga, H.; Kakuchi, T.; Satoh, T. Biomacromolecules 2011,
12, 1891−1899.
(19) Jimenez Blanco, J. L.; Ortega-Caballero, F.; Ortiz Mellet, C.;
Garcia Fernandez, J. M. Beilstein J. Org. Chem. 2010, 6, 20.
(20) For recent examples on the use of the thiourea-forming reaction
for multiconjugation of cyclodextrin platforms to different functional
elements, see : (a) Men
Vierling, P.; Santoyo-Gonzal
J. M. J. Org. Chem. 2011, 76, 5882−5894. (b) Ortiz Mellet, C.; Benito,
J. M.; García Fernandez, J. M. Chem. Soc. Rev. 2011, 40, 1586−1680.
(c) Ortiz Mellet, C.; Benito, J. M.; García Fernan
́
dez-Ardoy, A.; Guilloteau, N.; Di Giorgio, C.;
́
ez, F.; Ortiz Mellet, C.; García Fernandez,
́
(6) Kitov, P. I.; Sadowska, J. M.; Mulvey, G.; Armstrong, G. D.; Ling,
H.; Pannu, N. S.; Read, R. J.; Bundle, D. R. Nature 2000, 403, 669−
672.
́
́
dez, J. M. Chem.
(7) Horan, N.; Yan, L.; Isobe, H.; Whitesides, G. M.; Kahne, D. Proc.
Natl. Acad. Sci. U.S.A. 1995, 96, 11782−11786.
Eur. J. 2010, 16, 6728−6742. (d) Díaz-Moscoso, A.; Vercauteren, D.;
Rejman, J.; Benito, J. M.; Ortiz Mellet, C.; De Smedt, S. C.; García
(8) (a) Thoma, G.; Patton, J. T.; Magnani, J. L.; Ernst, B.; Oehrlein,
R.; Duthaler, R. O. J. Am. Chem. Soc. 1999, 121, 5919−5929.
(b) Ladmiral, V.; Mantovani, G.; Clarkson, G. J.; Cauet, S.; Irwin, J. L.;
Haddleton, D. M. J. Am. Chem. Soc. 2006, 128, 4823−4830.
(c) Miyachi, A.; Dohi, H.; Neri, P.; Mori, H.; Uzawa, H.; Seto, Y.;
Nishida, Y. Biomacromolecules 2009, 10, 1846−1853.
́
Fernandez, J. M. J. Controlled Release 2010, 143, 318−325.
(21) Lubineau, A.; Malleron, A.; Le Narvor, C. Tetrahedron Lett.
2000, 41, 8887−8891.
́
(22) Defaye, J.; Ortiz Mellet, C.; Garcia Fernandez, J. M.; Gomez-
García, M.; Chmurski, K.; Yu, J. X. WO2004/087768, 2004.
(23) Fulton, D. A.; Stoddart, J. F. J. Org. Chem. 2001, 66, 8309−8319.
(24) Matta, K. L.; Girotra, R. N.; Barlow, J. J. Carbohydr. Res. 1975,
43, 101−109.
(25) Garcia-Moreno, M. I.; Diaz-Perez, P.; Benito, J. M.; Ortiz
Mellet, C.; Defaye, J.; Garcia Fernandez, J. M. Carbohydr. Res. 2002,
337, 2329−2334.
(26) Dufour, M.; Gramain, J. C.; Husson, H. P.; Sinibaldi, M. E.;
Troin, Y. Synth. Commun. 1992, 22, 189−200.
(9) (a) Wolfenden, M. L.; Cloninger, M. J. J. Am. Chem. Soc. 2005,
127, 12168−12169. (b) Wolfenden, M. L.; Cloninger, M. J.
Bioconjugate Chem. 2006, 17, 958−966.
(10) (a) Vico, R. V.; Voskuhl, J.; Ravoo, B. J. Langmuir 2011, 27,
1391−1397. (b) Voskuhl, J.; Stuart, M. C. A.; Ravoo, B. J. Chem.Eur.
J. 2010, 16, 2790−2796.
(11) (a) Pourceau, G.; Meyer, A.; Vasseur, J.; Morvan, F. J. Org.
Chem. 2009, 74, 1218−1222. (b) Deguise, I.; Lagnoux, D.; Roy, R.
New J. Chem. 2007, 31, 1321−1331. (c) Katajisto, J.; Karskela, T.;
Heinonen, P.; Loennberg, H. J. Org. Chem. 2002, 67, 7995−8001.
(27) Ortiz-Mellet, C.; Benito, J. M.; Garcia Fernandez, J. M.; Law, H.;
Chmurski, K.; Defaye, J.; O’Sullivan, M. L.; Caro, H. N. Chem.Eur. J.
1998, 4, 2523−2531.
(d) Ortega-Munoz, M.; Perez-Balderas, F.; Morales-Sanfrutos, J.;
̃
(28) (a) Corbell, J. B.; Lundquist, J. J.; Toone, E. J. Tetrahedron:
Asymmetry 2000, 11, 95−111. (b) Gouin, S. G.; Vanquelef, E.; Garcia
Fernandez, J. M.; Ortiz Mellet, C.; Dupradeau, F.; Kovensky, J. J. Org.
Chem. 2007, 72, 9032−9045.
Hernandez-Mateo, F.; Isac-Garcia, J.; Santoyo-Gonzalez, F. Eur. J. Org.
Chem. 2009, 2454−2473.
(12) (a) Dulery, V.; Renaudet, O.; Wilczewski, M.; Van der Heyden,
A.; Labbe, P.; Dumy, P. J. Comb. Chem. 2008, 10, 368−371.
(29) Gouin, S. G.; Garcia Fernandez, J. M.; Vanquelef, E.;
(b) Ramstrom, O.; Lohmann, S.; Bunyapaiboonsri, T.; Lehn, J.-M.
̈
Dupradeau, F.; Salomonsson, E.; Leffler, H.; Ortega-Munoz, M.;
̃
Chem.Eur. J. 2004, 10, 1711−1715.
Nilsson, U. J.; Kovensky, J. ChemBioChem 2010, 11, 1430−1442.
(30) Naismith, J. H.; Emmerich, C.; Habash, J.; Harrop, S. J.;
Helliwell, J. R.; Hunter, W. N.; Raftery, J.; Kalb, A. J.; Yariv, J. Acta
Crystallogr., Sect. D 1994, 50, 847−858.
(13) Cyclodextrins have been widely used for the construction of
homogeneous glycoclusters. For reviews and selected recent examples,
see: (a) Ortiz Mellet, C.; Defaye, J.; Garcia Fernandez, J. M. Chem.
Eur. J. 2002, 8, 1982−1990. (b) Vargas-Berenguel, A.; Ortega-
Caballero, F.; Casas-Solvas, J. M. Mini-Rev. Org. Chem. 2007, 4, 1−14.
(c) Bernardes, G. J. L.; Kikkeri, R.; Maglinao, M.; Laurino, P.; Collot,
M.; Hong, S. Y.; Lepenies, B.; Seeberger, P. H. Org. Biomol. Chem.
2010, 8, 4987−4996. (d) Díaz-Moscoso, A.; Guilloteau, N.; Bienvenu,
(31) (a) Banerjee, R.; Das, K.; Raviashankar, R.; Suguna, K.; Surolia,
A.; Vijayan, M. J. Mol. Biol. 1996, 259, 281−296. (b) Ravishankar, R.;
Suguna, K.; Surolia, A.; Vijayan, M. Acta Crystallogr., Sect. D 1999, 55,
1375−1382.
(32) Baussanne, I.; Benito, J. M.; Ortiz Mellet, C.; Garcia Fernandez,
J. M.; Defaye, J. ChemBioChem 2001, 2, 777−783.
(33) (a) Benito, J. M.; Gomez-Garcia, M.; Ortiz Mellet, C.;
Baussanne, I.; Defaye, J.; Garcia Fernandez, J. M. J. Am. Chem. Soc.
2004, 126, 10355−10363. (b) Ashton, P. R.; Hounsell, E. F.;
Jayaraman, N.; Nilsen, T. M.; Spencer, N.; Stoddart, J. F.; Young, M.
J. Org. Chem. 1998, 63, 3429−3437.
C.; Men
Gourrierec, L.; Di Giorgio, C.; Vierling, P.; Defaye, J.; Ortiz Mellet, C.;
García Fernan
́ ́
dez-Ardoy, A.; Jimenez Blanco, J. L.; Benito, J. M.; Le
́
́
dez, J. M. Biomaterials 2011, 32, 7263−7273.
(e) Grunstein, D.; Maglinao, M.; Kikkeri, R.; Collot, M.; Barylyuk,
̈
K.; Lepenies, B.; Kamena, F.; Zenobi, R.; Seeberger, P. H. J. Am. Chem.
Soc. 2011, 133, 13957−13966.
(34) Lotan, R.; Skutelsky, E.; Danon, D.; Sharon, N. J. Biol. Chem.
1975, 250, 8518−8523.
(35) Ravishankar, R.; Thomas, C. J.; Suguna, K.; Surolia, A.; Vijayan,
M. Proteins 2001, 43, 260−270.
(14) Gomez-Garcia, M.; Benito, J. M.; Rodriguez-Lucena, D.; Yu,
J.-X.; Chmurski, K.; Ortiz Mellet, C.; Gutierrez Gallego, R.; Maestre,
A.; Defaye, J.; Garcia Fernandez, J. M. J. Am. Chem. Soc. 2005, 127,
7970−7971.
(15) The term heterocluster effect refers to the increase in the intrinsic
binding affinity of a carbohydrate ligand towards a lectin in the
presence of a second sugar that itself is not a ligand for this lectin
(intrinsic affinity means in a per-ligand basis, keeping the same overall
topology and preventing contributions to binding from aggregation
phenomena).
(36) Gestwicki, J. E.; Cairo, C. W.; Strong, L. E.; Oetjen, K. A.;
Kiessling, L. L. J. Am. Chem. Soc. 2002, 124, 14922−14933.
(37) Mammen, M.; Choi, S.-K.; Whitesides, G. M. Angew. Chem., Int.
Ed. 1998, 37, 2754−2794.
(38) Lindhorst, T. K.; Bruegge, K.; Fuchs, A.; Sperling, O. Beilstein J.
Org. Chem. 2010, 6, 801−809.
(39) Williams, T. J.; Homer, L. D.; Shafer, J. A.; Goldstein, I. J.;
Garegg, P. J.; Hultberg, H.; Iversen, T.; Johansson, R. Archiv. Biochem.
Biophys. 1981, 209, 555−564.
(16) Gomez-Garcia, M.; Benito, J. M.; Gutierrez-Gallego, R.;
Maestre, A.; Ortiz Mellet, C.; Garcia Fernandez, J. M.; Jimenez
Blanco, J. L. Org. Biomol. Chem. 2010, 8, 1849−1860.
(17) (a) Glycosciences−Status and Perspectives; Gabius, H.-J., Gabius,
S., Eds.; Chapman & Hall: Weinheim, 1997. (b) Goldstein, I. J.;
Poretz, R. D. In The Lectins−Properties, Functions and Applications in
Biology and Medicine; Liener, I. E., Sharon, N., Goldstein, I. J., Eds.;
Academic Press: Orlando, 1986; pp 33−243.
(40) Cairo, C. W.; Gestwicki, J. E.; Kanai, M.; Kiessling, L. L. J. Am.
Chem. Soc. 2002, 124, 1615−1619.
(41) Guiard, J.; Fiege, B.; Kitov, P. I.; Peters, T.; Bundle, D. R.
Chem.Eur. J. 2011, 17, 7438−7441.
(18) (a) Dondoni, A. Angew. Chem., Int. Ed. 2008, 47, 8995−8997.
(b) Hoyle, C. E.; Bowman, C. N. Angew. Chem., Int. Ed. 2010, 49,
1540−1573.
1288
dx.doi.org/10.1021/jo201797b | J. Org. Chem. 2012, 77, 1273−1288