The design, synthesis and evaluation of high affinity macrocyclic carbohydrate inhibitors

Article abstract of DOI:10.1039/b416105j

Carbohydrate-protein interactions have been investigated for a model system of a monoclonal antibody, SYA/J6, which binds a trisaccharide epitope of the O-polysaccharide of the Shigella flexneri variant Y lipopolysaccharide. The thermodynamics of binding

Full text of DOI:10.1039/b416105j

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
The design, synthesis and evaluation of high affinity macrocyclic
carbohydrate inhibitors†
Robert S. McGavin,a Rod A. Gagne,b Mary C. Chervenakc and David R. Bundle*a
a Department of Chemistry, University of Alberta, Edmonton, Alberta, Canada T6G 2G2.
E-mail: dave.bundle@ualberta.ca
b Lonza Biologics, 101 International Drive, Portsmouth, NH 03801.
E-mail: Rodney.Gagne@lonza.com
c The Dow Chemical Company, UCAR Emulsion Systems, 3524 Yadkinville Rd. #317,
Winston-Salem, NC 27106. E-mail: CherveMC@dow.com
Received 18th October 2004, Accepted 9th May 2005
First published as an Advance Article on the web 10th June 2005
Carbohydrate–protein interactions have been investigated for a model system of a monoclonal antibody, SYA/J6,
which binds a trisaccharide epitope of the O-polysaccharide of the Shigella flexneri variant Y lipopolysaccharide. The
thermodynamics of binding for the methyl glycoside of the native trisaccharide epitope, Rha-Rha-GlcNAc (1) to
SYA/J6 over a range of temperatures exhibits strong, linear enthalpy–entropy compensation and a negative heat
capacity change (DCp = −152 cal mol−1 degree−1). At 293 K the free energy of association is the sum of favourable
enthalpy and entropy contributions (DH = −3.9 kcal mol−1 and −TDS = −2.9 kcal mol−1). Crystal structures for
SYA/J6 Fab detailed the position of the native trisaccharide epitope, Rha-Rha-GlcNAc (1), and facilitated a strategy
to design a tighter binding, low molecular weight ligand. This involved pre-organization of the native trisaccharide 1
in its bound conformation by addition of intramolecular constraints (a b-alanyl or glycinyl tether). ELISA
measurements indicated that the glycinyl tethered trisaccharide 2 was not an optimal candidate for further analysis,
while microcalorimetry provided data showing that the b-alanyl tethered trisaccharide 3 displayed a 15-fold increase
in affinity for SYA/J6. Tethering of 3 resulted in a favourable entropic contribution to binding, relative to the native
trisaccharide 1 (−TDDS = −1.2 kcal mol−1). Potential energy and dynamics calculations using the AMBER Plus
force fields indicated that trisaccharide 3 adopted a rigid conformation similar to that of the bound conformation of
the native trisaccharide epitope. While this strategy resulted in modest free energy gains by minimizing losses due to
conformational entropy, thermodynamic data are consistent with significant contributions from solvent
reorganization.
bination of structural and thermodynamic data for a range
of monodeoxy ligands provides a clear appreciation of those
residues that generate crucial interactions. Our initial attempts
Introduction
Carbohydrate–protein interactions are found to have dissocia-
tion constants (KD) in the milli- to micromolar range1 with only a
utilized a trisaccharide–antibody Fab complex, wherein the
few exceptions that typically employ multivalent presentations.2
branching 3,6-dideoxy hexose residue of the trisaccharide was
While several physiologically significant interactions of this type
buried and the hydroxymethyl groups of the other two hexose
could benefit from intervention by high affinity compounds, the
residues did not contact protein.6a,b A second antibody Fab–
KD of the relevant carbohydrate–protein interaction must be
trisaccharide complex has been mapped in similar detail and,
interestingly, introduction of either monodeoxy or chloro mon-
significantly improved if design of small ligands as potential
drug candidates is to be realized.3
odeoxy functions created ligands that exhibited 10–25-fold gains
The low intrinsic affinity of these interactions is general and
in affinity.9 It was of interest to investigate tethering of such
its origins have been the topic of much speculation.1,4 Several ex-
compounds and to attempt to create high avidity ligands.
planations have been advanced to account for such weak binding
The monoclonal antibody SYA/J6 was produced following
and include loss of conformational entropy when flexible glyco-
hyper-immunization with a killed, whole cell vaccine.10 The
sidic linkages are constrained in the bound state.1 Several groups,
including ours, have attempted to investigate this phenomenon
antibody recognizes the O-polysaccharide of the cell wall
lipopolysaccharide of Shigella flexneri variant Y.11 The O-
and probe the validity of the latter explanation by pre-ordering
polysaccharide is composed of linear tetrasaccharide repeat-
ligands in their bound conformation.5–8 Pre-ordering attempts
ing units (where the letters ABCD represent the positions of
have generally involved introducing conformational bias that
the monosaccharide residues within the repeat):12
favour the bound conformation5 or intramolecular tethering
between functional groups of the ligand that do not contact
[→ 2)-a-L-Rhap(1 → 2)-a-L-Rhap(1 → 3)-a-L-
A
B
protein.6,7 To date, increases in binding affinity reported as free
energy gains have been limited to 0.5–1.0 kcal mol−1. Some
results failed to demonstrate any gains from entropy effects.
We have favoured investigation of ligand–protein complexes
for which crystal structures are available and for which detailed
titration microcalorimetry data has been collected. The com-
Rhap(1 → 3)-b-D-GlcNAcp(1 →]n
C
D
Two crystal structures of SYA/J6 complexed with an ABCDAꢀ
pentasaccharide and with a 2ꢀ-deoxy trisaccharide derivative
(BCꢀD) have been described in a report that details the molec-
ular interactions of the system.11 Of the five residues of the
pentasaccharide, it is chiefly the BCD residues that make the
dominant interactions with the protein surface. This was also
noticed during the refinement of the crystal structure, where
a less optimised solution only provided electron densities for
† Electronic supplementary information (ESI) available: determination
of DCp for SYA/J6 bound to native trisaccharide and entropy–enthalpy
compensation: −DH vs. −TDS for trisaccharide 1 and SYA/J6. See
http://www.rsc.org/suppdata/ob/b4/b416105j/
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 7 2 3 – 2 7 3 2
2 7 2 3
©

View Article Online
the BCD residues.11 It was postulated that the flanking A and Aꢀ
residues were more mobile and thus did not participate in the key
interactions between the SYA/J6 Fab and the pentasaccharide.9c
One proposal to explain the low intrinsic affinity of
carbohydrate–protein interactions invokes the loss of rotational
freedom about the glycosidic bond when an oligosaccharide
ligand is bound by protein. The magnitude of this penalty
has been estimated at about 0.6 kcal mol−1 per torsion angle,
although work by Whitesides and co-workers suggests this
number may be smaller.13 Observation of the pentasaccharide in
the binding groove of the SYA/J6 Fab indicates that the methyl
groups belonging to the 2-acetamido moiety of residue D and
the C-6 methyl group of residue B extend away from the protein
surface into bulk solvent. We identified these sites as an optimal
position at which to introduce intramolecular tethering. In order
to achieve pre-organization in the bound conformation the size
of the tether would have to be modified.
Scheme 1 Reagents and conditions: (a) EtSH, BF3–O(Et)2, CH2Cl2,
77%; (b) NaOMe, MeOH, 100%; (c) TBDPS-Cl, imidazole, DMF, 81%;
(d) BnBr, NaH, THF, 86%.
In this study, the native trisaccharide 1 was tethered by
connecting residue B to residue D via glycinyl or b-alanyl tethers
(2 and 3, respectively, Fig. 1). A linkage methodology was used
that employed L-mannose instead of L-rhamnose as residue
B to facilitate oxidation and amide bond formation to ligate
the tether. Acyclic derivatives (4 and 5) are based on the b-
alanyl-tethered 3, but with “cut” tethers. These compounds were
prepared to make available analogues with atom arrangements
similar to their cyclic counterpart, especially the amide at the 6-
position of residue B. The activity of these ligands were initially
screened using a competitive solid phase assay, followed by more
detailed analysis of the binding parameters using isothermal
titration microcalorimetry (ITC) to determine the free energy,
enthalpy and entropy of binding and also the heat capacity
for the interaction of the native ligand with the antibody.14
Molecular dynamics calculations and quantitative NOE
measurements were used to evaluate the solution structure of
the tethered trisaccharide 3.
by NIS–AgOTf, gave trisaccharide 15 (78% yield). The anomeric
configuration of both newly synthesized glycosidic linkages
1
were assigned as a based on JCH coupling constants (167.3
and 171.1 Hz).18 Partial deprotection to afford intermediate
core trisaccharide 17 was achieved by treatment of 15 with
ethylene diamine and n-BuOH at reflux to afford the free amine
16. Finally, tetrabutylammonium fluoride (TBAF)19 mediated
removal of silyl ether 16 gave the amino alcohol trisaccharide
17.
Parallel synthesis of trisaccharides 18 and 19 employed
standard peptide coupling conditions (TBTU, HOBt, N-
ethylmorpholine)20 and N-Fmoc protected glycine or N-Fmoc
protected b-alanine. Oxidation of the resultant primary alcohol
of residue B used a TEMPO based system employing phase
transfer conditions21 to give the carboxylic acids 20 and 21 in
moderate yield (Scheme 3). The Fmoc group was cleaved by
20% piperidine in DMF, followed by treatment with TBTU,
HOBt and N-ethylmorpholine to effect macrolactamization in
good yield (65% for 22 and 60% for 23). Global deprotection via
hydrogenolysis with Pd(OH2) in wet methanol afforded the de-
protected cyclic trisaccharides 2 and 3. Both were characterized
by NMR and mass spectroscopy.
The synthesis of the acyclic variants of 3 originated from
amine 16 (Scheme 4). Acylation of 16 with either acetic or
propionic anhydride, followed by TBAF mediated desilylation
afforded primary alcohols 24 and 25. Oxidation was again
accomplished with a TEMPO–NaOCl system and the crude
uronic acids were each treated with carbonyldiimidazole (CDI)
without further purification. The reactions were quenched with
ammonia or methylamine to produce the protected bis-amides
26 and 27 respectively. Hydrogenolysis using Pd(OH)2 produced
the deprotected isomeric trisaccharides 4 and 5 (78% and 81%).
The inhibitory power of trisaccharides 2 to 5 were initially
screened using a competitive ELISA.22 Biotin labelled O-
polysaccharide competed with increasing concentrations of
inhibitors for binding to antibody immobilized on microtitre
plates. Under the conditions of the assay the concentration
of inhibitor required to reach 50% inhibition (IC50) provides
Results
The L-mannosyl thioglycoside donor 10 was prepared in five
steps from L-mannose (Scheme 1). Peracetylated L-mannose
(6) was reacted with ethanethiol and boron trifluoride di-
ethyl etherate in CH2Cl2.15 Acetate protecting groups were re-
moved under Zemplen conditions (NaOMe–MeOH) to produce
thioglycoside 8, followed by regioselective protection with t-
butylchlorodiphenylsilane and imidazole in DMF. Exhaustive
benzylation by treatment of triol 9 with benzyl bromide and
sodium hydride in THF afforded the fully blocked donor 10.
Glycosylation of the known acceptor 1116 with the previously
reported rhamnosyl thioglycoside donor 1217 initiated the linear
synthesis of the core trisaccharide 17 (Scheme 2). This reaction
proceeded smoothly and rapidly under promotion with N-
iodosuccinimide (NIS) and silver triflate (AgOTf) to afford dis-
accharide 13 in 88% yield. Transesterification with a methanolic
solution of sodium methoxide afforded disaccharide acceptor
14. Glycosylation by the L-mannose donor 10, again mediated
Fig. 1 Native trisaccharide (1), cyclic trisaccharides containing glycinyl- (2, n = 1) and b-alanyl- (3, n = 2) tethers. Acyclic derivatives 4 and 5, with
their “cut” tethers in isomeric positions, are also displayed. The rings are labelled BCD from the non-reducing end to the reducing end.
2 7 2 4
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 7 2 3 – 2 7 3 2

View Article Online
Scheme 2 Reagents and conditions: (a) NIS, AgOTf, CH2Cl2, 88%; (b) NaOMe, MeOH, 92%; (c) NIS, AgOTf, CH2Cl2, 78%; (d) NH2CH2CH2NH2,
t-BuOH, 115 ◦C, 90%; (e) TBAF, THF, 60 ◦C, 86%.
Scheme 3 Reagents and conditions: (a) TBTU, HOBt, NEM, DMF and Fmoc-Gly-OH or Fmoc-b-Ala-OH, 86% for 18, 91% for 19; (b) TEMPO,
NaOCl, n-Bu4NCl, NaHCO3, KBr, NaCl, 43% for 20, 63% for 21; (c) (i) 20% piperidine in DMF; (ii) TBTU, HOBt, NEM, DMF, 65% for 22, 60%
for 23; (d) H2, Pd(OH)2, MeOH, 74% for 2, 73% for 3.
an estimate of the disassociation constant. The results are
listed in the second column of Table 1. The b-alanyl-tethered
trisaccharide 3 showed a marked increase in activity relative to 1,
consistent with the envisioned pre-organization of the ligand in
the bound state conformation. Conversely, the glycinyl-tethered
2 lost approximately an order of magnitude of binding affinity,
suggesting that the shorter tether induced a conformation that
diverged substantially from that required in the bound state.
Both acyclic variants 4 and 5 exhibited an almost equal 5-
fold decrease in affinity relative to that of native trisaccharide
1. The most likely explanations were either that a chemical
incompatibility of the polar uronic amide installed on the L-
mannosyl residue with the protein surface or a direct steric clash
between the unligated forms of the tether (Table 1).
Microcalorimetry was employed to establish the thermody-
namics of the ligand–antibody interactions. The binding of
native ligand 1 was investigated over a range of temperatures.
Constant pressure heat capacities, DCp, were determined by
Table 1 Competitive ELISA and ITC data for the native trisaccharide (1) and synthesized trisaccharide derivatives (2 through 5)
Compound
IC50/lM
KA/mol−1
DG/kcal mol−1
DH/kcal mol−1
−TDS/kcal mol−1
1
2
3
4
5
17
137
5
94
92
1.1 × 105
—
−6.8 0.2
—
−3.9 0.1
—
−2.9 0.1
—
1.5 0.05 × 106
2.0 × 104
2.9 × 104
−8.3 0.1
−4.2 0.05
−4.1 0.05
−2.3
−5.8
−3.5
−6.0
−3.1
−2.9
Standard deviations are calculated from duplicate measurements. Errors in the determination of DH are 2.5%.
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 7 2 3 – 2 7 3 2
2 7 2 5

View Article Online
Scheme 4 Synthesis of acyclic trisaccharides 4 and 5. Reagents and conditions: (a) acetic anhydride, MeOH; (b) propionic anhydride, MeOH; (c)
TBAF, THF, 88% for 25, 89% for 26 (two steps); (d) TEMPO, KBr, NaHCO3, NaOCl, MeCN (pH 10.5 with NaOH); (e) CDI, CH2Cl2, then NH3 or
MeNH2, 68% for 27, 67% for 28 (two steps); (f) H2, Pd(OH)2, MeOH, H2O, 81% for 4, 78% for 5.
evaluating DH between 285 to 309 K and were found to be
negative (DCp = −152 cal mol−1 degree−1).† The enthalpy and
entropy exhibit strong linear compensation so that the entropic
contribution to the binding interaction approaches zero as the
temperature increases, and the enthalpy of binding equals the
free energy of binding at a temperature of 318 K.†
Isothermal titration calorimetry was performed on the
tightest binding trisaccharides (1, 3 through5), since ELISA data
provides only relative or approximate association constants.13
The KA of the tethered trisaccharide 3 determined by calorimetry
was even larger than that predicted by the ELISA data and the
free energy gain relative to 1 (DDG) was approximately −1.5 kcal
mol−1. In contrast to previously published data11 from our group,
the current measurements were recorded at 293 K (Table 1).
To explain the binding data, and to confirm that cyclic 3 was
able to adopt a bioactive conformation, molecular dynamics
calculations were performed using Insight II with the AMBER
Plus force field.23 A dielectric constant of 80 was used to
simulate the presence of water. Fig. 2 illustrates the ␾ vs. w
plots for both glycosidic bonds in the native trisaccharide 1,
tethered trisaccharide 3 and acyclic amide 5. Compound 4 was
omitted from the calculation due to its extensive similarity with
5 (Fig. 2). Tethered trisaccharide 3 shows remarkable similarity
to the conformation of 1 and also indicates a narrower range
of conformational freedom, despite a slight deviation on the
w2 axis away from the conformation of 1. While maintaining a
similar conformation to that of 1, acyclic 5 (and by implication 4)
populated a broader range of conformational space. The shorter
glycinyl tether resulted in a conformation for trisaccharide 2 that
was distinctly different than that of native 1.
Estimates of the solution conformation of unbound lig-
ands were inferred from NOE data obtained from T-ROESY
experiments. These data were used to calculate inter-proton
distances. The values correlated well with distances found in the
bound conformation of the BCD trisaccharide with SYA/J6 and
conformations predicted by dynamics calculations. To illustrate
the BC and CD residues, respectively.
this, the global minimum energy structure of 3 was superimposed
Fig. 2 ␾ vs. w plots for (from top) 1, 3 and 5 as measured from MD
simulation. Subscripts BC and CD refer to the glycosidic bond between
on residue C of the BCD portion of the trisaccharide element
of the ABCDAꢀ pentasaccharide–SYA/J6 complex taken from
reference 11 (Fig. 3). This clearly shows that the conformation
of 3 (blue) correlates closely with that of the bound BCD
trisaccharide (red).
Discussion
We have previously shown that the native trisaccharide BCD
essentially fills the binding site of the monoclonal antibody
SYA/J6.9c,11 Consistent with this interpretation, when the
2 7 2 6
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 7 2 3 – 2 7 3 2

View Article Online
how water plays a crucial role in mediating protein–carbohydrate
hydrogen bonds. The different roles of water can profoundly
influence the thermodynamic parameters.26–29 Antibody–protein
complexes have been studied where water is either eliminated
from the binding site or conserved to mediate antigen–antibody
hydrogen bonds.28,29 In addition, it seems probable that the tether
itself may induce solvent re-organization in the bound state,
since the tether will certainly displace water at the interface of
the ligand and the protein complex24 and possibly also make new
van der Waals contacts with the protein. While examination of
the crystal structure of bound oligosaccharide complexes with
Fab offered no suggestions that the amide used for ligation
of the tether to residue B of the trisaccharide 3 would make
any significant contacts with the protein surface, there is some
evidence that the tether in the vicinity of the amide group of
residue D could make contacts with the protein and generate
either favourable or unfavourable interactions with the binding.
Both types of interactions could result in favourable enthalpic
effects.4c,24,25
Fig. 3 Super-imposition of the conformation of 1, taken from the
SJA/J6 co-complex, and the global energy minimum of 3 as calculated
by MD simulations.
epitope is extended by a single residue at either end to give
first the tetrasaccharide (ABCD) corresponding to the biological
repeating unit of the polysaccharide, the free energy of binding
increased by −0.4 kcal mol−1 and by a further −0.2 kcal mol−1 as
an additional rhamnose residue extends the oligosaccharide to
give pentasaccharide ABCDAꢀ. Interestingly, for each additional
hexose residue the enthalpic change becomes increasingly detri-
mental and entropy changes are more favourable by ∼1.5 kcal
mol−1.11 Significantly, the addition of two hexose residues yields
no more than −0.6 kcal mol−1 of additional free energy for
association. Of most interest to the following discussion are
the increases in favourable entropy as the size of the epitope
increased, with this phenomenon being ascribed to displacement
of water molecules at the periphery of the binding pocket. This
underscores the importance of solvent reorganization as a major
factor in the association process. Similar conclusions have been
reached for other carbohydrate–protein systems by Lemieux
et al.4,24 and from direct estimates of solvent effects by Chervenak
and Toone.25
Conclusions
We have demonstrated that a pre-organized trisaccharide in-
hibitor with an optimized tether length is able to mimic the
conformation of the bound, native acyclic trisaccharide epitope.
A relative free energy change (DDG = −1.5 0.3 kcal mol−1)
determined by ITC measurements arose from a favourable
entropy effect (−TDDS = −1.2 0.15 kcal mol−1). These data
suggest that the inherent flexibility of saccharides is not the pri-
mary cause of the low intrinsic affinity of carbohydrate–protein
interactions. The successful design of the tethered form of this
carbohydrate epitope provides us with an opportunity to probe
potential strategies to create higher affinity carbohydrate ligands
by combining tethering with functional group modifications.
These findings are reported in our following paper.
Experimental
The free energy gain for the tethered trisaccharide 3 relative to
1 (DDG) was approximately −1.5 kcal mol−1. At the temperature
of the experiment, 293 K, this gain was of predominantly
entropic origin (Table 1). Ignoring other competing effects, this
would suggest that the maximum entropy gain to be achieved by
pre-organizing a ligand would fall in the range 1–1.5 kcal mol−1.
However, the competing effects discussed below suggest this
number should be regarded as an upper limit. The magnitude
of these entropy gains for organizing four single bonds over
two glycosidic linkages would tend to agree with the estimates
suggested by Mammen et al.13
Molecular dynamics calculations confirmed that 3 was con-
strained in a conformation that is virtually super-imposable on
the bound conformation of native ligand 1 (Fig. 2). Trisaccharide
2 is inactive since its tether is too short and it cannot adopt
the bound conformation. Low activities of acyclic 5 (and by
implication 4) could be explained by the potential for significant
steric interaction between the methyl groups of the NHAc and
the N-methyl amide of 5.
The role of water and the influence of the tether complicate
the attempt to place an estimate on the entropy gains associated
with successfully pre-ordering the oligosaccharide. It is generally
impossible to de-convolute the competing enthalpy and entropy
contributions and assign their origin to precise structural
features of the carbohydrate–protein complex. To consider only
a few of the likely processes, it could be anticipated that the
tether might affect solvation of the complex, while the ligand
also displaces water from the binding site. The negative heat
capacity that characterizes binding of 1 is often correlated
with hydrophobic effects and there is clear evidence that the
favourable entropy of binding for trisaccharide 1 and larger
structures is associated with displacement of water.11 In this
and other carbohydrate–antibody complexes6a,9c we have studied
General methods
Analytical thin layer chromatography (tlc) was performed on
silica gel 60-F254 (Merck). Detection of compounds on TLC
plates was achieved by charring with 5% sulfuric acid in ethanol.
The optical rotations were measured with a Perkin-Elmer 241
polarimeter for samples in a 10 cm cell at 22 2 ◦C. [a]D values
are given in units of 10−1 deg cm2 g−1. All commercial reagents
were used as supplied. Column chromatography used silica gel
(SiliCycle) and solvents that were distilled. High performance
liquid chromatography (HPLC) was performed using a Waters
HPLC system that consisted of a Waters 600S controller, 626
pump and 486 tuneable absorbance detector. HPLC separations
were performed on a Beckmann C18 semi-preparative reversed-
phase column with acetonitrile and water as eluents. 1H NMR
spectra were recorded at either 300, 360, 500 or 600 MHz, and
are referenced to internal standards of the residual protonated
solvent peaks; dH 7.24 ppm for solutions in CDCl3 or to 0.1%
external acetone (dH 2.225 ppm) for solutions in D2O. 13C NMR
spectra (HMQC) were recorded at 125 MHz and are referenced
to internal CDCl3 (dC 77.0 ppm) or to external acetone (dC
31.07 ppm). Electro-spray mass spectrometry and microanalyses
were performed by the analytical services of this department.
Ethyl 2,3,4,6-tetra-O-acetyl-1-thio-a–L-mannopyranoside (7).
Anhydrous CH2Cl2 (100 mL) was added to L-mannose pentaac-
etate (6) (2.1 g, 5.38 mmol) along with ethanethiol (1.20 mL,
16.2 mmol, 0.839 g mL−1). Boron trifluoride diethyl etherate
(2.00 mL, 15.8 mmol, 1.12 g mL−1) was added and stirred under
an argon atmosphere for 2 h at 0 ◦C, followed by 16 h at rt.
Saturated NaHCO3 (aq.) was added and the solution was stirred
for 2 h. The organic layer was dried over Na2SO4, concentrated
and the product was purified by silica gel chromatography using
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 7 2 3 – 2 7 3 2
2 7 2 7

View Article Online
˚
toluene–ethyl acetate, 3 : 1, as the eluent to afford as a white solid
(1.6 g, 77%): [a]2D2 −110.6 (c 0.9, CH2Cl2). 1H NMR (300 MHz,
CDCl3) d: 5.34–5.25 (m, 4H, H-1, H-2, H-3, H-4), 4.38 (ddd,
1H, J4,5 = 9.9 Hz, J5,6a = 5.3 Hz, J5,6b = 2.3 Hz, H-5), 4.29 (dd,
1.27 mmol) and 4 A molecular sieves (1.5 g) and stirred under
an argon atmosphere for 14 h at rt. The solution was cooled to
−78 ◦C, silver trifluoromethanesulfonate (85 mg, 0.33 mmol)
and N-iodosuccinimide (742 mg, 3.30 mmol) were added and
the solution was allowed to reach rt. The resulting dark purple
solution was then filtered through celite and the resulting
solution was washed with saturated Na2S2O3 (100 mL) and
dried with Na2SO4. The solvent was evaporated and the product
was purified using silica gel chromatography (hexane–ethyl
acetate, 2 : 1). The protected disaccharide 13 was isolated as
1H, J5,6a = 5.3 Hz, J6a,6b = 12.1 Hz, H-6a), 4.08 (dd, 1H, J5,6b
=
2.3 Hz, J6a,6b = 12.1 Hz, H-6b), 2.60–2.66 (m, 2H, SCH2CH3),
2.16 (s, 3H, OAc), 2.03 (s, 3H, OAc), 2.02 (s, 3H, OAc), 1.97 (s,
3H, OAc), 1.28 (t, 3H, SCH2CH3). Anal. calcd for C16H24O9S:
C, 48.97; H, 6.16; S, 8.17%. Found: C, 48.86; H, 6.17; S, 8.24%.
ES HRMS: (M + Na): 415.1039, found: 415.1033.
a white solid (872 mg, 88%): [a]2D2 −24.0 (c 0.9, CH2Cl2). H
1
Ethyl 1-thio-a-L-mannopyranoside (8). Distilled methanol
(100 mL) was added to the tetraacetate 7 (1.5 g, 3.82 mmol)
along with sodium (105 mg, 4.56 mmol) and stirred for 18 h
at rt. Strongly-acidic cationic exchange resin was added and
stirred until the solution became neutral. The solution was
filtered through celite and concentrated. Alcohol 8 was obtained
NMR (300 MHz, CDCl3) d: 7.85–7.68 (m, 4H, Phth), 7.32–7.14
(m, 13H, aromatic), 6.90–6.95 (m, 2H, aromatic), 5.55 (s, 1H,
ꢀ
ꢀ
CH–Ph), 5.19 (d, 1H, J1,2 = 8.6 Hz, H-1), 5.08 (dd, 1H, J2 ,3
=
3.5 Hz, J3 ,4 = 9.5 Hz, H-3ꢀ), 4.66 (dd, 1H, J2,3 = 9.7 Hz, J3,4
=
ꢀ
ꢀ
10.3 Hz, H-3), 4.59 (d, 1H, J1 ,2 = 1.8 Hz, H-1ꢀ), 4.46–4.50 (m,
2H, CH2–Ph), 4.41 (dd, 1H, J5,6a = 4.0 Hz, J6a,6b = 12.5 Hz,
H-6a), 4.30 (dd, 1H, J1,2 = 8.5 Hz, J2,3 = 10.2 Hz, H-2), 3.96 (dq,
ꢀ
ꢀ
1
without purification as a white solid (0.86 g, 100%): H NMR
(600 MHz, D2O) d: 5.33 (d, 1H, J1,2 = 1.6 Hz, H-1), 4.05 (dd, 1H,
J1,2 = 1.6 Hz, J2,3 = 3.3 Hz, H-2), 4.00 (ddd, 1H, J4,5 = 9.7 Hz,
J5,6a = 2.3 Hz, J5,6b = 6.2 Hz, H-5), 3.89 (dd, 1H, J5,6a = 2.3 Hz,
1H, J4 ,5 = 9.7 Hz, J5 ,6 = 6.2 Hz, H-5ꢀ), 3.94 (d, 1H, CH2–Ph),
3.83–3.86 (m, 1H, H-6b), 3.76 (d, 1H, CH2–Ph), 3.68–3.73 (m,
ꢀ
ꢀ
ꢀ ꢀ
ꢀ
ꢀ
ꢀ ꢀ
2H, H-4, H-5), 3.51 (dd, 1H, J1 ,2 = 2.0 Hz, J2 ,3 = 3.5 Hz,
J6a,6b = 12.4 Hz, H-6a), 3.79 (dd, 1H, J5,6b = 6.1 Hz, J6a,6b
=
H-2ꢀ), 3.42 (s, 3H, OMe), 3.35 (t, 1H, J3 ,4 = J4 ,5 = 9.5 Hz,
ꢀ
ꢀ
ꢀ ꢀ
12.4 Hz, H-6b), 3.78 (dd, 1H, J2,3 = 3.3 Hz, J3,4 = 9.7 Hz, H-
3), 3.68 (dd, 1H, J3,4 = J4,5 = 9.7 Hz, H-4), 2.66–2.71 (m, 2H,
SCH2CH3), 1.28 (t, 3H, SCH2CH3). ES HRMS: (M + Na):
247.0616, found: 247.0619.
H-4 ), 1.90 (s, 3H, OAc), 0.79 (d, 3H, J5 ,6 = 6.2 Hz, H-6ꢀ). 13
C
ꢀ
ꢀ ꢀ
NMR (125 MHz, CDCl3) d: 99.5 (JC-1,H-1 = 164.6 Hz, C-1),
ꢀ
ꢀ
ꢀ
97.7 (JC-1 ,H-1 = 167.3 Hz, C-1 ). Anal. calcd for C44H45NO12: C,
67.77; H, 5.82; N, 1.80%. Found: C, 67.40; H, 5.83; N, 1.74%.
ES HRMS: (M + Na): 802.2839, found: 802.2832.
Ethyl 6-O-tert-butyl-diphenylsilyl-1-thio-a-L-mannopyrano-
side (9). Dry DMF (30 mL) was added to alcohol 8 (700 mg,
3.12 mmol) along with t-butylchlorodiphenylsilane (1.60 mL,
6.15 mmol) and imidazole (425 mg, 6.24 mmol) and stirred for
20 h at rt. The volatiles were then removed and CH2Cl2 (100 mL)
was added to the residue. The solution was then washed with
5% HCl (aq.) (50 mL), dried over Na2SO4 and filtered. The
solvent was evaporated and the product was purified by silica
gel chromatography (toluene–acetone–methanol, 7.5 : 1 : 0.5).
Silyl ether 9 was isolated as a colourless oil (1.2 g, 81%): [a]D22
−81.7 (c 1.0, MeOH). 1H NMR (300 MHz, CDCl3) d: 7.69–7.34
(m, 10H, aromatic), 5.28 (d, 1H, J1,2 = 2.8 Hz, H-1), 4.00–4.06
(m, 2H, H-2, H-5), 3.94–3.78 (m, 4H, H-3, H-4, H-6a, H-6b),
2.50–2.58 (m, 2H, SCH2CH3), 1.21 (t, 3H, SCH2CH3), 1.04 (s,
9H, t-Bu). Anal. calcd for C24H34O5SSi: C, 62.30; H, 7.41%.
Found: C, 61.83; H, 7.49%. ES HRMS: (M + Na): 485.1794,
found: 485.1790.
Methyl 3-O-(2,4-di-O-benzyl-a-L-rhamnopyranosyl)-4,6-O-
benzylidene-2-deoxy-2-phthalimido-b-D-glucopyranoside
(14).
Distilled methanol (100 mL) was added to the acetate 13
(535 mg, 0.686 mmol), sodium (50 mg, 2.17 mmol) was added
and the reaction was allowed to stir for 18 h at rt. Strongly-
acidic cationic exchange resin was then added to neutralize
the solution that was subsequently filtered through celite. The
solvent was evaporated and the product was purified by silica
gel chromatography (toluene–ethyl acetate, 5 : 1). Alcohol 14
was isolated as a white solid (466 mg, 92%): [a]2D2 −37.2 (c
1
1.0, CH2Cl2). H NMR (600 MHz, CDCl3) d: 7.88–7.68 (m,
4H, Phth), 7.32–7.14 (m, 13H, aromatic), 6.86–6.91 (m, 2H,
aromatic), 5.56 (s, 1H, CH–Ph), 5.19 (d, 1H, J1,2 = 8.5 Hz,
H-1), 4.70 (d, 1H, CH2–Ph), 4.64 (d, 1H, J1 ,2 = 1.3 Hz, H-1ꢀ),
4.60 (dd, 1H, J2,3 = 9.6 Hz, J3,4 = 10.4 Hz, H-3), 4.46 (d, 1H,
CH2–Ph), 4.41 (dd, 1H, J5,6a = 4.3 Hz, J6a,6b = 10.3 Hz, H-6a),
4.26 (dd, 1H, J1,2 = 8.5 Hz, J2,3 = 10.2 Hz, H-2), 3.78–3.82
(m, 4H, H-6b, H-3ꢀ, H-5ꢀ, CH2–Ph), 3.63–3.67 (m, 3H, H-4,
ꢀ
ꢀ
Ethyl 2,3,4-tri-O-benzyl-6-O-tert-butyl-diphenylsilyl-1-thio-a-
L-mannopyranoside (10). Anhydrous THF (100 mL) was added
to alcohol 9 (950 mg, 2.05 mmol) along with sodium hydride
(250 mg, 10.4 mmol) and stirred under an argon atmosphere for
15 min. Benzyl bromide (1.20 mL, 10.1 mmol) was added and
the solution was stirred at 60 ◦C under an argon atmosphere for
24 h. The resulting yellow, cloudy solution was then cooled,
methanol (6 mL) was carefully added and the solution was
stirred for 15 min. The solvent was evaporated and the residue
was taken up in CH2Cl2, washed with water and dried with
Na2SO4. After concentration the crude product was purified
by silica gel chromatography (hexane–ethyl acetate, 15 : 1).
Compound 10 was isolated as a yellow oil (1.3 g, 86%): [a]D22
ꢀ
ꢀ
H-5, CH2–Ph), 3.42 (s, 3H, OMe), 3.35 (dd, 1H, J1 ,2 = 1.6 Hz,
J2 ,3 = 3.8 Hz, H-2ꢀ), 3.05 (t, 1H, J3 ,4 = J4 ,5 = 9.5 Hz, H-4ꢀ),
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ ꢀ
2.01 (d, 1H, OH), 0.72 (d, 3H, J5 ,6 = 6.2 Hz, H-6ꢀ). Anal. calcd
for C42H43NO11: C, 68.37; H, 5.87; N, 1.90%. Found: C, 68.14;
H, 6.02; N, 1.88%. ES HRMS: (M + Na): 760.2734, found:
760.2748.
ꢀ
ꢀ
Methyl 3-O-(2,4-di-O-benzyl-3-O-[2,3,4-tri-O-benzyl-6-O-
tert-butyl-diphenylsilyl-a-L-manno-pyranosyl]-a-L-rhamnopyra-
nosyl)-4,6-O-benzylidene-2-deoxy-2-phthalimido-b-D-glucopyra-
noside (15). Anhydrous CH2Cl2 (70 mL) was added to
mannosyl donor 10 (425 mg, 0.593 mmol) and acceptor
14 (336 mg, 0.455 mmol) and stirred for 10 h under argon
1
−52.6 (c 1.2, CH2Cl2). H NMR (300 MHz, CDCl3) d: 7.75–
7.15 (m, 25H, aromatic), 5.37 (d, 1H, J1,2 = 1.3 Hz, H-1), 4.92–
4.56 (m, 6H, CH2–Ph), 4.00–4.07 (m, 3H, H-4, H-5, H-6a), 3.89
(dd, 1H, J5,6b = 1.2 Hz, J6a,6b = 10.4 Hz, H-6b), 3.88 (dd, 1H,
J2,3 = 3.2 Hz, J3,4 = 9.7 Hz, H-3), 3.68 (dd, 1H, J1,2 = 1.6 Hz,
J2,3 = 3.1 Hz, H-2), 2.56–2.60 (m, 2H, SCH2CH3), 1.20 (t, 3H,
SCH2CH3), 1.04 (s, 9H, t-Bu). Anal. calcd for C45H52O4SSi: C,
73.73; H, 7.15%. Found: C, 73.85; H, 7.20%. ES HRMS: (M +
Na): 755.3202, found: 755.3202.
˚
with 4 A molecular sieves (1 g). The solution was cooled to
−78 ◦C, silver trifluoromethanesulfonate (30 mg, 0.12 mmol)
and N-iodosuccinimide (267 mg, 1.19 mmol) were added
and the solution was allowed to reach rt. The resulting dark
purple solution was then filtered through celite and the resultant
solution was washed with saturated Na2S2O3, dried with Na2SO4
and filtered. The solvent was evaporated and the product was
purified by silica gel chromatography (hexane–ethyl acetate, 2
: 1). Trisaccharide 15 was isolated as a white solid (500 mg,
78%): [a]2D2 −112.3 (c 1.1, CH2Cl2). 1H NMR (600 MHz, CDCl3)
d: 7.82–6.78 (m, 44H, aromatic), 5.54 (s, 1H, CH–Ph), 5.19
(d, 1H, J1,2 = 8.5 Hz, H-1), 5.12 (bs, 1H, H-1ꢀꢀ), 4.92 (d, 1H,
Methyl 3-O-(3-O-acetyl-2,4-di-O-benzyl-a-L-rhamnopyrano-
syl)-4,6-O-benzylidene-2-deoxy-2-phthalimido-b-D-glucopyranoside
(13). Anhydrous CH2Cl2 (125 mL) was added to glycosyl
donor 1217 (710 mg, 1.65 mmol), acceptor 1116 (523 mg,
2 7 2 8
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 7 2 3 – 2 7 3 2

View Article Online
CH2–Ph), 4.65 (d, 1H, CH2–Ph), 4.56 (dd, 1H, J2,3 = 9.6 Hz,
0.13 mmol) and N-ethylmorpholine (34 lL, 0.26 mmol, 0.905 g
mL−1) and stirred for 18 h at rt. The solvents were evaporated
and the product was purified by silica gel chromatography
(toluene–ethyl acetate–methanol, 7.5 : 2 : 0.5). Amide 18 was
isolated as a white solid (76 mg, 86%): [a]2D2 −13.2 (c 0.7,
J3,4 = 10.4 Hz, H-3), 4.53 (d, 1H, J1 ,2 = 1.3 Hz, H-1ꢀ), 4.51
(dd, 1H, J5,6a = 4.3 Hz, J6a,6b = 10.3 Hz, H-6a), 4.36–4.44 (m,
ꢀ
ꢀ
6H, CH2–Ph), 4.36 (t, 1H, J3
= J4
= 9.6 Hz, H-4ꢀꢀ), 4.18
ꢀꢀ ꢀꢀ
ꢀꢀ ꢀꢀ
,5
,4
(dd, 1H, J1,2 = 8.5 Hz, J2,3 = 10.1 Hz, H-2), 3.97 (dd, 1H,
J5
= 2.5 Hz, J6a
= 11.5 Hz, H-6aꢀꢀ), 3.76–3.86 (m, 7H,
CH2Cl2). H NMR (300 MHz, CDCl3) d: 7.69–7.08 (m, 38H,
1
ꢀꢀ ꢀꢀ
ꢀꢀ ꢀꢀ
,6b
,6a
H-6b, H-3ꢀ, H-5ꢀ, H-3ꢀꢀ, H-5ꢀꢀ, H-6bꢀꢀ, CH2–Ph), 3.60–3.70 (m,
aromatic), 6.43–6.45 (m, 1H, NH-Glc-NH2), 5.68 (bt, 1H,
NH-Fmoc), 5.48 (s, 1H, CH-Ph), 5.18 (bs, 1H, H-1ꢀꢀ), 4.96
(bs, 1H, H-1ꢀ), 4.80 (d, 1H, CH2–Ph), 4.60–4.24 (m, 10H, H-1,
CH2–Ph), 4.28 (dd, 1H, J5,6a = 5.7 Hz, J6a,6b = 10.5 Hz, H-6a),
4H, H-4, H-5, H-2ꢀꢀ, CH2–Ph), 3.42 (s, 3H, OMe), 3.26 (m, 2H,
H-2ꢀ, H-4ꢀ), 1.04 (s, 9H, t-Bu), 0.74 (d, 3H, J5 ,6 = 6.2 Hz, H-6ꢀ).
13C NMR (125 MHz, CDCl3) d: 99.5 (JC-1,H-1 = 165.0 Hz, C-1),
ꢀ
ꢀ
99.5 (JC-1
= 171.1 Hz, C-1ꢀꢀ), 97.6 (JC-1 ,H-1 = 168.0 Hz,
4.12 (bt, O–CH2–CH[Fmoc]), 4.01 (dd, 1H, J2 ,3 = 2.8 Hz,
ꢀꢀ
ꢀꢀ
ꢀ
ꢀ
ꢀ ꢀ
,H-1
C-1ꢀ). Anal. calcd for C85H89NO16: C, 72.47; H, 6.37; N, 0.99%.
Found: C, 72.24; H, 6.24; N, 1.01%. ES HRMS: (M + Na):
1430.5838, found: 1430.5848.
J3 ,4 = 9.7 Hz, H-3ꢀ), 4.00 (t, 1H, J3 = J4 = 9.3 Hz, H-4ꢀꢀ),
ꢀ
ꢀ
ꢀꢀ ꢀꢀ
,4
ꢀꢀ ꢀꢀ
,5
3.92 (dd, 1H, J2 ,3 = 2.9 Hz, J3 ,4 = 9.0 Hz, H-3ꢀꢀ), 3.80–3.88
ꢀ
ꢀ
ꢀ ꢀ
(m, 5H, H-2ꢀ, H-5ꢀ, H-6aꢀꢀ, linker), 3.70 (t, 1H, J1
= J2 ,3
=
ꢀꢀ ꢀꢀ
ꢀ
ꢀ
,2
1.9 Hz, H-2ꢀꢀ), 3.61–3.69 (m, 4H, H-3, H-4, H-6b, H-6bꢀꢀ), 3.52
Methyl 2-amino-3-O-(2,4-di-O-benzyl-3-O-[2,3,4-tri-O-ben-
zyl-6-O-tert-butyl-diphenylsilyl-a-L-mannopyranosyl]-a-L-rham-
nopyranosyl)-4,6-O-benzylidene-2-deoxy-b-D-glucopyranoside
(16). Ethylenediamine (3 mL) and t-butyl alcohol (30 mL)
were added to 15 (450 mg, 0.319 mmol) and stirred under an
argon atmosphere for 20 h at 115 ◦C. The resulting pale yellow
solution was then cooled to rt. The solvent was evaporated
and the product was purified by silica gel chromatography
(toluene–ethyl acetate, 5 : 1). Amine 16 was isolated as a white
solid (368 mg, 90%): [a]2D2 −75.2 (c 0.9, CH2Cl2). 1H NMR
(600 MHz, CDCl3) d: 7.82–7.14 (m, 40H, aromatic), 5.50 (s,
(t, 1H, J3 ,4 = J4 ,5 = 9.3 Hz, H-4ꢀ), 3.40–3.45 (m, 6H, H-2, H-5,
ꢀ
ꢀ
ꢀ ꢀ
O–CH2–CH[Fmoc], H-5ꢀꢀ, H-6bꢀꢀ), 3.38 (s, 3H, OMe), 0.73 (d,
3H, J5 ,6 = 6.0 Hz, H-6ꢀ). ES HRMS: (M + Na): 1341.5511,
ꢀ
ꢀ
found: 1341.5512.
Methyl 3-O-(2,4-di-O-benzyl-3-O-[2,3,4-tri-O-benzyl-a-L-
mannopyranosyl]-a-L-rhamnopyranosyl)-4,6-O-benzylidene-2-
deoxy-2-(N-b-fluorenylmethoxycarbonyl-b-alanyl)-amido-b-D-
glucopyranoside (19)
Amine 17 was prepared under similar conditions as for the syn-
thesis of 18, employing Fmoc-protected b-alanine as the amino
acid. The product was purified by silica gel chromatography
(toluene–ethyl acetate–methanol, 7.5 : 2 : 0.5) to give the title
compound as a white solid (105 mg, 91%): [a]2D2 −21.0 (c 0.7,
1H, CH–Ph), 5.22 (bs, 1H, H-1ꢀꢀ), 5.08 (d, 1H, J1 ,2 = 1.7 Hz,
H-1ꢀ), 4.97 (d, 1H, CH2–Ph), 4.40–4.71 (m, 9H, CH2–Ph), 4.34
(dd, 1H, J5,6a = 4.5 Hz, J6a,6b = 10.6 Hz, H-6a), 4.20 (t, 1H,
ꢀ
ꢀ
J3 = J4 = 9.4 Hz, H-4ꢀꢀ), 4.08 (d, 1H, J1,2 = 8.0 Hz, H-1),
ꢀꢀ ꢀꢀ
,4
ꢀꢀ ꢀꢀ
,5
1
3.95–4.07 (m, 3H, H-3ꢀ, H-5ꢀ, H-3ꢀꢀ), 3.90 (dd, 1H, J5
=
CH2Cl2). H NMR (300 MHz, CDCl3) d: 7.69–7.08 (m, 38H,
ꢀꢀ ꢀꢀ
,6a
= 11.4 Hz, H-6aꢀꢀ), 3.73–3.77 (m, 3H, H-6b, H-2ꢀꢀ,
aromatic), 6.00 (bt, 1H, NH–Fmoc), 5.52 (s, 1H, CH–Ph), 5.11
ꢀꢀ ꢀꢀ
4.2 Hz, J6a
,6b
(d, 1H, J1
= 2.2 Hz, H-1ꢀꢀ), 4.86 (bs, 1H, H-1ꢀꢀ), 4.80 (d,
H-5ꢀꢀ), 3.61–3.69 (m, 2H, H-2ꢀ, H-6bꢀꢀ), 3.56 (t, 1H, J2,3 = J3,4
=
ꢀꢀ ꢀꢀ
,2
1H, CH2–Ph), 4.60–4.24 (m, 10H, H-1, CH2–Ph), 4.27 (dd,
1H, J5,6a = 5.5 Hz, J6a,6b = 10.5 Hz, H-6a), 4.08 (bt, O–CH2–
9.0 Hz, H-3), 3.50 (s, 3H, OMe), 3.45–3.51 (m, 3H, H-4, H-5,
H-4ꢀ), 2.62 (t, 1H, J1,2 = J2,3 = 8.4 Hz, H-2), 1.02 (s, 9H, t-Bu),
CH[Fmoc]), 4.01 (dd, 1H, J2 ,3 = 2.8 Hz, J3 ,4 = 9.7 Hz, H-3ꢀꢀ),
0.82 (d, 3H, J5 ,6 = 6.1 Hz, H-6ꢀ). Anal. calcd for C77H87NO14Si:
C, 72.33; H, 6.86; N, 1.10%. Found: C, 71.80; H, 6.77; N, 1.09%.
ES HRMS: (M + Na): 1278.5974, found: 1278.5968.
ꢀ
ꢀ ꢀ
4.01 (t, 1H, J3
= J4
=ꢀ 9.3 Hz, H-4ꢀꢀ), 3.80–3.90 (m, 6H,
ꢀ
ꢀ
ꢀꢀ ꢀꢀ
ꢀꢀ ꢀꢀ
,5
,4
H-2ꢀ, H-5ꢀ, H-3ꢀꢀ, H-6aꢀꢀ, linker), 3.73–3.77 (m, 3H, H-3, H-6b,
H-6bꢀꢀ), 3.66–3.69 (m, 2H, H-4, linker), 3.67 (t, 1H, J1
=
ꢀꢀ ꢀꢀ
,2
J2 ,3 = 2.6 Hz, H-2ꢀꢀ), 3.56–3.58 (m, 1H, linker), 3.52 (t, 1H,
Methyl 2-amino-3-O-(2,4-di-O-benzyl-3-O-[2,3,4-tri-O-ben-
zyl-a-L-mannopyranosyl]-a-L-rhamnopyranosyl)-4,6-O-benzyli-
dene-2-deoxy-b-D-glucopyranoside (17). Anhydrous THF
(50 mL) was added to silyl ether 16 (310 mg, 0.242 mmol)
along with tetrabutylammonium fluoride (480 lL, 0.480 mmol,
1.0 mol L−1 in◦THF) and stirred under an argon atmosphere
for 15 h at 60 C. The resulting pale yellow solution was then
cooled to rt and evaporated. The solution was then dissolved
in CH2Cl2 and washed with water. The organic phase was dried
with Na2SO4 and filtered. The solvent was evaporated and
the product was purified by silica gel chromatography using
(toluene–ethyl acetate–methanol, 7.5 : 2 : 0.5). Alcohol 17 was
isolated as a white solid (217 mg, 86%): [a]2D2 −59.5 (c 0.8,
ꢀ
ꢀꢀ
J3 ,4 = J4 ,5 = 9.3 Hz, H-4ꢀ), 3.37–3.42 (m, 5H, H-5, O–CH2–
ꢀ
ꢀ
ꢀ ꢀ
CH[Fmoc], H-5ꢀꢀ, H-6bꢀꢀ), 3.30 (s, 3H, OMe), 3.24–3.27 (m, 1H,
ꢀ
ꢀ
H-2), 2.16–2.19 (m, 2H, linker), 0.73 (d, 3H, J5 ,6 = 6.0 Hz, H-
6ꢀ). Anal. calcd for C79H84N2O17: C, 71.15; H, 6.35; N, 2.10%.
Found: C, 71.27; H, 6.24; N, 2.20%. ES HRMS: (M + Na):
1355.3667, found: 1355.3675
Methyl 3-O-(2,4-di-O-benzyl-3-O-[2,3,4-tri-O-benzyl-a-L-
mannopyranosyluronic acid]-a-L-rhamnopyranosyl)-4,6-O-ben-
zylidene-2-deoxy-2-(N-a-fluorenyl-methoxycarbonyl-glycinyl)-
amido-b-D-glucopyranoside (20). Distilled CH2Cl2 (4 mL)
and TEMPO (4 mg, 0.02 mmol) were added to alcohol 18
(60 mg, 0.045 mmol) and a solution of potassium bromide
(8 mg, 0.07 mmol) and tetrabutylammonium chloride (10 mg,
0.036 mmol) in saturated NaHCO3 (aq.) (1.5 mL) was added.
The biphasic solution was cooled to 0 ◦C and a solution of
sodium hypochlorite (1.5 mL, 5% solution in water), saturated
NaHCO3 (aq.) (0.6 mL) and saturated NaCl (aq.) (1.3 mL)
was added dropwise over 45 min. The CH2Cl2 solution was
acidified with 5% HCl (aq.) (5mL), washed with water and
dried with Na2SO4. The solvents were removed and the product
was purified by silica gel chromatography (toluene–ethyl
acetate–methanol, 4 : 4 : 1). Carboxylic acid 20 was isolated
1
CH2Cl2). H NMR (300 MHz, CDCl3) d: 7.45–7.16 (m, 30H,
ꢀꢀ ꢀꢀ
aromatic), 5.52 (s, 1H, CH–Ph), 5.30 (d, 1H, J1
= 1.6 Hz,
,2
H-1ꢀꢀ), 5.14 (d, 1H, J1 ,2 = 1.6 Hz, H-1ꢀ), 4.88 (d, 1H, CH2–Ph),
ꢀ
ꢀ
4.60 (m, 9H, CH2–Ph), 4.31 (dd, 1H, J5,6a = 4.9 Hz, J6a,6b
=
10.5 Hz, H-6a), 4.19 (d, 1H, J1,2 = 7.8 Hz, H-1), 3.96–4.02 (m,
2H, H-2ꢀꢀ, H-3ꢀꢀ), 3.85–3.89 (m, 2H, H-5ꢀ, H-6aꢀꢀ), 3.61–3.69 (m,
5H, H-4, H-6b, H-2ꢀ, H-5ꢀꢀ, H-6bꢀꢀ), 3.53–3.58 (m, 3H, H-3, H-4ꢀ,
H-4ꢀꢀ), 3.51 (s, 3H, OMe), 3.40 (ddd, 1H, J4,5 = 9.7 Hz, J5,6a
=
4.9 Hz, J5,6b = 1.2 Hz, H-5), 2.79 (t, 1H, J1,2 = J2,3 = 8.7 Hz, H-2),
1.10 (d, 3H, J5 ,6 = 6.1 Hz, H-6ꢀ). Anal. calcd for C61H69NO14:
C, 70.43; H, 6.69; N, 1.35%. Found: C, 70.06; H, 6.68; N, 1.31%.
ES HRMS: (M + Na): 1062.4616, found: 1062.4628.
ꢀ
ꢀ
1
as a white solid (26 mg, 43%): [a]2D2 +0.4 (c 0.6, MeOH). H
NMR (300 MHz, CDCl3) d: 7.75–7.00 (m, 38H, aromatic),
5.42 (bs, 1H, CH–Ph), 5.18 (bs, 1H, H-1ꢀꢀ), 5.04 (bs, 1H, H-1ꢀ),
4.83 (bd, 1H, CH2–Ph), 4.63–4.34 (m, 10H, H-1, CH2–Ph),
4.30–3.20 (m, 19H, H-2, H-3, H-4, H-5, H-6a, H-6b, H-2ꢀ, H-3ꢀ,
H-4ꢀ, H-5ꢀ, H-2ꢀꢀ, H-3ꢀꢀ, H-4ꢀꢀ, H-5ꢀꢀ, linker, O–CH2–CH[Fmoc],
Methyl 3-O-(2,4-di-O-benzyl-3-O-[2,3,4-tri-O-benzyl-a-L-
mannopyranosyl]-a-L-rhamnopyranosyl)-4,6-O-benzylidene-2-
deoxy-2-(N-a-fluorenylmethoxycarbonyl-glycinyl)-amido-b-D-
glucopyranoside (18). Dry DMF (20 mL) was added to amine
17 (70 mg, 0.067 mmol) along with N-a-Fmoc-L-glycine (40 mg,
0.14 mmol), TBTU (43 mg, 0.134 mmol), HOBt (18 mg,
ꢀ
ꢀ
O–CH2–CH[Fmoc]), 3.32 (s, 3H, OMe), 0.73 (d, 3H, J5 ,6
=
5.7 Hz, H-6ꢀ). Anal. calcd for C78H80N2O18: C, 70.26; H, 6.05; N,
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 7 2 3 – 2 7 3 2
2 7 2 9

View Article Online
2.10%. Found: C, 69.90; H, 6.44; N, 2.01%. ES HRMS: (M +
Found: H, 6.48; N, 2.62%. ES HRMS: (M + Na): 1129.4674,
Na): 1355.5304, found: 1355.5314.
found: 1129.4684.
Methyl 3-O-(2,4-di-O-benzyl-3-O-[2,3,4-tri-O-benzyl-a-L-
mannurono]-a-L-rhamnopyranosyl)-4,6-O-benzylidene-2-deoxy-
2-(2-N-fluorenylmethoxycarbonyl-b-alanyl)-amido-b-D-glucopy-
ranoside (21). Alcohol 19 (85 mg, 0.064 mmol) was treated
in an identical fashion to 18, to afford the title compound
that was purified by silica gel chromatography (toluene–ethyl
acetate–methanol, 4 : 4 : 1) to give carboxylic acid 21 as a
white solid (54 mg, 63%): [a]2D2 +13.6 (c 1.0, MeOH). 1H NMR
(300 MHz, CDCl3) d: 7.69–7.06 (m, 38H, aromatic), 5.42 (bs,
1H, CH-Ph), 5.18 (bs, 1H, H-1ꢀꢀ), 4.85 (bs, 1H, H-1ꢀ), 4.80 (d,
1H, CH2–Ph), 4.60–4.24 (m, 10H, H-1, CH2–Ph), 4.30–3.20
(m, 19H, H-2, H-3, H-4, H-5, H-6a, H-6b, H-2ꢀ, H-3ꢀ, H-4ꢀ,
H-5ꢀ, H-2ꢀꢀ, H-3ꢀꢀ, H-4ꢀꢀ, H-5ꢀꢀ, linker, O–CH2–CH[Fmoc],
O–CH2–CH[Fmoc]), 3.20 (s, 3H, OMe), 2.16–2.21 (m, 2H,
Methyl 2-(2-aminoacetamido)-3-O-(3-O-[a-L-mannopyranosyl-
uronic acid]-a-L-rhamnopyranosyl)-2-deoxy-b-D-glucopyranoside
lactam (2). Distilled methanol (15 mL) and distilled water
(1 mL) were added to the protected cyclic trisaccharide 22
(16 mg, 0.014 mmol) along with palladium(II) hydroxide (23 mg,
20% on carbon) and the suspension was stirred under a hydrogen
atmosphere for 24 h at rt. The resulting solution was then filtered
through celite and the solvent was evaporated. Distilled water
(5 mL) was added to the residue and the solution was passed
through an RP-18 Sep-Pak cartridge. The product was purified
by reversed-phase HPLC using water–acetonitrile, (0–10%). The
deprotected cyclic trisaccharide 2 was lyophilized to a white solid
ꢀꢀ ꢀꢀ
(6 mg, 74%): 1H NMR (600 MHz, D2O) d: 5.22 (d, 1H, J1
=
,2
1.9 Hz, H-1ꢀꢀ), 4.99 (d, 1H, J1 .2 = 1.5 Hz, H-1ꢀ), 4.76 (d, 1H,
ꢀ
ꢀ
linker), 0.73 (d, 3H, J5 ,6 = 5.7 Hz, H-6ꢀ). Anal. calcd for
C79H82N2O18: H, 6.13; N, 2.08%. Found: H, 5.59; N, 2.11%. ES
HRMS: (M + Na): 1369.5460, found: 1369.5464.
J1,2 = 8.4 Hz, H-1), 4.14 (d, 1H, J4 = 9.5 Hz, H-5ꢀꢀ), 3.95–4.00
ꢀ
ꢀ
ꢀꢀ ꢀꢀ
,5
(m, 2H, H-5ꢀ, H-6a), 3.92–3.94 (m, 3H, H-2, H-3ꢀꢀ, H-4ꢀꢀ), 3.84
(bt, 1H, J1
= J2
= 3.2 Hz, H-2ꢀꢀ), 3.83 (dd, 1H, J2 .3
=
=
ꢀꢀ ꢀꢀ
ꢀꢀ ꢀꢀ
ꢀ
ꢀ
,2
,3
3.3 Hz, J3 ,4 = 9.9 Hz, H-3ꢀ), 3.79 (dd, 1H, J1 ,2 = 2.0 Hz, J2 ,3
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ ꢀ
Methyl 2-(2-aminoacetamido)-3-O-(2,4-di-O-benzyl-3-O-[2,3,4-
tri-O-benzyl-a-L-mannopyranosyl-uronic acid]-a-L-rhamnopyra-
nosyl)-4,6-O-benzylidene-2-deoxy-b-D-glucopyranoside lactam
(22). Dry DMF (12 mL) and piperidine (2 mL) were added to
carboxylic acid 20 (36 mg, 0.027 mmol) and stirred for 1 h at
rt. The solvents were evaporated and dry DMF (10 mL) was
added to the residue along with TBTU (17 mg, 0.054 mmol),
HOBt (7 mg, 0.054 mmol) and N-ethylmorpholine (14 lL,
0.11 mmol, 0.905 g mL−1) and stirred for 24 h at rt. The solvents
were evaporated and the product was purified by silica gel
chromatography (toluene–ethyl acetate–methanol, 7.5 : 2 :
0.5). Lactam 22 was isolated as a white solid (19 mg, 65%):
3.2 Hz, H-2ꢀ), 3.76 (dd, 1H, J5,6b = 5.8 Hz, J6a,6b = 12.2 Hz,
H-6b), 3.63–3.67 (m, 1H, linker), 3.50–3.58 (m, 4H, H-4ꢀ, H-3,
H-4, linker), 3.50 (s, 3H, OMe), 3.46 (ddd, 1H, J5,6a = 2.3 Hz,
ꢀ
ꢀ
J5,6b = 5.2 Hz, J4,5 = 10.1 Hz, H-5), 1.29 (d, 3H, J5 ,6 = 6.4 Hz,
H-6ꢀ). ES HRMS: (M + Na): 577.1857, found: 577.1863.
Methyl 2-(2-aminopropionamido)-3-O-(3-O-[a-L-mannopyra-
nosyluronic acid]-a-L-rhamnopyranosyl)-2-deoxy-b-D-glucopy-
ranoside lactam (3). The protected cyclic trisaccharide 23
(32 mg, 0.029 mmol) was deprotected and purified in an
analogous fashion as 2 to give the title compound as a
lyophilized white solid (12 mg, 73%): [a]2D2 −36.3 (c 0.5, H2O).
1
[a]2D2 −27.1 (c 0.4, CH2Cl2). H NMR (300 MHz, CDCl3) d:
1H NMR (600 MHz, D2O) d: 5.11 (d, 1H, J1 = 1.5 Hz, H-1ꢀꢀ),
ꢀꢀ ꢀꢀ
,2
4.77 (d, 1H, J1 .2 = 2.2 Hz, H-1ꢀ), 4.41 (d, 1H, J1,2 = 8.6 Hz,
ꢀ
ꢀ
7.50–7.13 (m, 30H, aromatic), 7.00 (bt, 1H, NH–Fmoc), 6.20
(d, 1H, NH–Glc–NH2), 5.51 (s, 1H, CH–Ph), 5.24 (d, 1H,
H-1), 4.09 (dd, 1H, J1 = 1.5 Hz, J2 = 3.2 Hz, H-2ꢀꢀ), 4.04
ꢀꢀ ꢀꢀ
,2
ꢀꢀ ꢀꢀ
,3
J1
= 2.7 Hz, H-1ꢀꢀ), 5.03 (d, 1H, J1 .2 = 1.6 Hz, H-1ꢀ), 4.98
= 9.5 Hz, H-5ꢀꢀ), 3.95–3.99 (m, 2H, H-5ꢀ, H-6a),
ꢀꢀ ꢀꢀ
ꢀ
ꢀ
ꢀꢀ ꢀꢀ
(d, 1H, J4
,5
,2
(d, 1H, J1,2 = 8.3Hz, H-1), 4.71–4.38 (m, 10H, CH2Ph), 4.30
3.92 (dd, 1H, J1,2 = 8.5 Hz, J2,3 = 10.3 Hz, H-2), 3.89 (t, 1H,
= 9.5 Hz, H-4ꢀꢀ), 3.85 (dd, 1H, J3
= 9.3 Hz,
ꢀꢀ ꢀꢀ
ꢀꢀ ꢀꢀ
ꢀꢀ ꢀꢀ
,4
(dd, 1H, J5,6a = 5.0 Hz, J6a,6b = 10.7Hz, H-6a), 4.23–4.26 (m,
J3
J2
= J4
,4
,5
1H, linker), 4.14 (d, 1H, J4
= 9.3 Hz, H-5ꢀꢀ), 4.08 (dd, 1H,
= 3.3 Hz, H-3ꢀꢀ), 3.83 (dd, 1H, J2 .3 = 3.3 Hz, J3 ,4
=
ꢀꢀ ꢀꢀ
ꢀꢀ ꢀꢀ
ꢀ
ꢀ
ꢀ ꢀ
,5
,3
J2 .3 = 3.0 Hz, J3 ,4 = 9.7 Hz, H-3ꢀ), 3.90–4.00 (m, 4H, H-6b,
9.9 Hz, H-3ꢀ), 3.79 (dd, 1H, J1 ,2 = 2.2 Hz, J2 ,3 = 3.2 Hz, H-2ꢀ),
3.76 (dd, 1H, J5,6b = 5.8 Hz, J6a,6b = 12.2 Hz, H-6b), 3.65–3.70
(m, 1H, NH–CH2–CH2–NH), 3.52–3.57 (m, 4H, H-4ꢀ, H-3,
H-4, NH-CH2–CH2–NH), 3.50 (s, 3H, OMe), 3.46 (ddd, 1H,
J5,6a = 2.2 Hz, J5,6b = 6.1 Hz, J4,5 = 8.1 Hz, H-5), 2.58–2.61 (m,
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ ꢀ
H-5ꢀ, H-3ꢀꢀ, H-4ꢀꢀ), 3.59 (m, 2H, H-3, H-2ꢀꢀ), 3.86 (t, 1H, J3,4
=
J4,5 = 9.1 Hz, H-4), 3.79 (bt, 1H, J1 ,2 = J2 ,3 = 2.3 Hz, H-2ꢀ),
3.54–3.57 (m, 2H, H-5, H-4ꢀ), 3.45 (s, 3H, OMe), 3.02–3.07 (m,
1H, linker), 2.92 (bq, 1H, J1,2 = J2,3 = J2,NH 8.6 Hz, H-2), 0.83
ꢀ
ꢀ
ꢀ ꢀ
(d, 3H, J5 ,6 = 6.2 Hz, H-6ꢀ). Anal. calcd for C63H68N2O15: C,
69.22; H, 6.27; N, 2.56%. Found: C, 69.35; H, 6.02; N, 2.64%.
ES HRMS: (M + Na): 1115.4517, found: 1115.4527.
2H, NH–CH2–CH2-NH), 1.27 (d, 3H, J5 ,6 = 6.4 Hz, H-6ꢀ). ES
HRMS: (M + Na): 591.2013, found: 591.2006.
ꢀ
ꢀ
ꢀ ꢀ
Methyl 4,6-O-benzylidene-2-propionamido-3-O-(2,4-di-O-
benzyl-3-O-(2,3,4-tri-O-benzyl-a-L-mannopyranosyl)-a-L-rhamno-
Methyl 2-(3-aminopropionamido)-3-O-(2,4-di-O-benzyl-3-O-
[2,3,4-tri-O-benzyl-a-L-mannopyranosyluronic acid]-a-L-rham-
nopyranosyl)-4,6-O-benzylidene-2-deoxy-b-D-glucopyranoside
lactam (23). Carboxylic acid 21 (40 mg, 0.030 mmol) was
treated in an identical fashion that produced 22 to afford the
title compound that was purified by silica gel chromatography
(toluene–ethyl acetate–methanol, 7.5 : 2 : 0.5). Lactam 23
was isolated as a white solid (20 mg, 60%): [a]2D2 +12.1 (c 0.5,
pyranosyl)-b-D-glucopyranoside (24). To
a suspension of
amine 16 (0.457 g, 0.39 mmol) in MeOH (15 mL) was added
propionic anhydride (56 lL, 0.44 mmol). The reaction was
stirred for 2 h until clear, and K2CO3 (100 mg) was added.
The solvent was removed under a vacuum and the residue
was dissolved in CH2Cl2 treated with excess TBAF in THF
(5 mL). After 6 h the residue was evaporated and washed with
water. The organic layer was dried with Na2SO4, filtered and
evaporated to dryness. The residue was chromatographed (6 :
4 : 1 cyclohexane–ethyl acetate–acetone) to afford a white
foam (0.383 g, 88%): [a]2D2 −38.68 (c = 0.99, CHCl3). 1H NMR
(600 MHz, CDCl3) d 7.49–7.15 (m, 30H, Ar–H), 6.14 (bd, 1H,
J = 7.4 Hz, EtC(O)NH–), 5.49 (s, 1H, PhCHO2–), 5.21 (s,
1H, H-1ꢀꢀ), 4.92 (d, 1H, Jgem = 14.4 Hz, PhCH2O–), 4.91 (s,
1
CH2Cl2). H NMR (600 MHz, CDCl3) d: 7.50–7.13 (m, 30H,
aromatic), 6.89 (bs, 1H, NH–Fmoc), 5.48 (s, 1H, CH–Ph), 5.42
(d, 1H, NH-Glc-NH2), 5.14 (d, 1H, J1 = 2.3 Hz, H-1ꢀꢀ), 4.91
ꢀꢀ ꢀꢀ
,2
(d, 1H, J1 .2 = 1.3 Hz, H-1ꢀ), 4.71–4.38 (m, 10H, CH2Ph), 4.30
ꢀ
ꢀ
(dd, 1H, J5,6a = 5.0 Hz, J6a,6b = 10.7Hz, H-6a), 4.22 (d, 1H,
J1,2 = 8.6 Hz, H-1), 4.04 (d, 1H, J4 = 9.3 Hz, H-5ꢀꢀ), 3.92 (bt,
ꢀꢀ ꢀꢀ
,5
ꢀ
ꢀ
1H, J1,2 = J2,3 = 8.6 Hz, H-2), 3.96 (dd, 1H, J2 .3 = 3.0 Hz,
J3 ,4 = 9.7 Hz, H-3ꢀ), 3.89 (t, 1H, J3 = J4 = 9.3 Hz, H-4ꢀꢀ),
1H, H-1ꢀ), 4.61–4.51 (m, 7H, PhCH2O–), 4.39 (d, 1H, Jgem
=
ꢀ
ꢀ
ꢀꢀ ꢀꢀ
,4
ꢀꢀ ꢀꢀ
,5
3.70–3.80 (m, 4H, H-6b, H-5ꢀ, H-3ꢀꢀ, NH–CH2–CH2–NH),
12.2 Hz, PhCH2O–), 4.36–4.27 (m, 3H, PhCH2O-, H-6a, H-3),
3.59–3.69 (m, 2H, H-3, H-2ꢀꢀ), 3.76 (t, 1H, J3,4 = J4,5 = 9.1 Hz,
4.14 (dd, 1H, J = 2.8, 9.8 Hz, H-3ꢀꢀ), 3.99 (m, 1H, H-3ꢀꢀ),
ꢀꢀ
H-4), 3.79 (bt, 1H, J1 ,2 = J2 ,3 = 2.3 Hz, H-2ꢀ), 3.52–3.58
(m, 4H, H-4ꢀ, H-5, NH–CH2–CH2–NH), 3.45 (s, 3H, OMe),
3.88–3.81 (m, 2H, H-5ꢀ, H-6a ), 3.78 (dd, 1H, J1,2 ≈ J2,3
=
ꢀ
ꢀ
ꢀ ꢀ
2.3 Hz, H-2ꢀ), 3.77–3.71 (m, 4H, H-6b, H-2ꢀꢀ, H-4ꢀꢀ, H-5ꢀꢀ),
ꢀꢀ
2.37–2.39 (m, 2H, NH–CH2–CH–NH), 0.81 (d, 3H, J5 ,6
=
3.68–3.62 (m, 1H, H-6b ), 3.53–3.47 (m, 3H, H-4, H-5, H-4ꢀ),
ꢀ
ꢀ
6.2 Hz, H-6ꢀ). Anal. calcd for C64H70N2O15: H, 6.37; N, 2.53%.
3.43 (s, 3H, OCH3), 3.16 (ddd ≈ q, J = 9.3 Hz, H-2), 2.04–1.91
2 7 3 0
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 7 2 3 – 2 7 3 2

View Article Online
(0.136 g, 67%) as a white foam. [a]2D2 = −32.99 (c = 0.97, CHCl3).
1H NMR (600 MHz, CDCl3) d 7.48–7.09 (m, 30 H, Ar–H), 6.49
(bd, 1H, J = 7.1 Hz, –NHAc), 6.44 (bd, 1H, NHCH3), 5.52 (s,
1H, PhCHO2–), 5.24 (d, 1H, J = 1.4 Hz, H-1ꢀꢀ), 5.04 (d, 1H, J =
1.50 Hz, H-1ꢀ), 4.81 (d, 1H, Jgem = 10.9 Hz, PhCH2O–), 4.70 (d,
1H, J = 8.3 Hz, H-1), 4.64–4.49 (m, 7H, PhCH2O–), 4.37–4.30
(m, 3H, PhCH2O-, H-6a), 4.13 (dd, 1H, J = 2.7, 9.7 Hz, H-3ꢀ),
(m, 2H, –CH2CH3), 1.01 (t, 3H, 7.6 Hz, –CH2CH3), 0.84 (d,
3H, J = 7.62 Hz, H-6ꢀ). Anal. calcd for C64H73NO15: C, 68.68;
H, 6.57; N, 1.25%. Found C, 68.53; H, 6.64; N, 1.30%. ES MS:
(M + Na): 1118.5, found 1118.5.
Methyl 4,6-O-benzylidene-2-acetamido-3-O-(2,4-di-O-benzyl-
3-O-(2,3,4-tri-O-benzyl-a-L-mannopyranosyl)-a-L-rhamnopyra-
nosyl)-b-D-glucopyranoside (25). Amine 16 (0.496 g,
0.429 mmol) was treated analogously for the production
of 22 using acetic anhydride (42 lL, 0.451 mmol) as the
acylation agent. Similar work-up and chromatography gave the
target compound (0.413 g, 89%) as white foam: [a]2D2 = −36.79
(c = 1.08, CHCl3). 1H NMR (600 MHz, CDCl3) d 7.52–7.14 (m,
30H, Ar–H), 6.39 (bd, 1H, J = 7.3 Hz), 5.51 (s, 1H, PhCHO2–),
5.22 (d, 1H, J = 1.0 Hz, H-1ꢀꢀ), 4.94 (d, 1H, J = 1.7 Hz, H-1ꢀꢀ),
4.92 (d, 1H, Jgem = 11.2 Hz, PhCH2O–), 4.78 (d, 1H, J =
8.33 Hz, H-1), 4.66–4.52 (m, 7H, PhCH2O–), 4.43–4.32 (m, 3H,
H2O-, H-6a), 4.21 (dd, 1H, J2,3 ≈ J3,4 = 9.2 Hz, H-3), 4.14 (dd,
1H, J = 2.8, 9.7 Hz, H-3ꢀ), 3.91 (dd, 1H, J = 3.0, 8.6 Hz, H-3ꢀꢀ),
4.08 (bdd, 1H, J2,3 ≈ J3,4 = 9.2 Hz, H-3), 4.03 (dd, 1H, J3,4
≈
J4,5 = 9.0 Hz, H-4), 3.90–3.85 (m, 2H, H-5ꢀ, H-5ꢀꢀ), 3.83 (dd,
1H, J = 2.9, 9.7 Hz, H-3ꢀꢀ), 3.76 (dd, 1H, Jvic ≈ Jgem = 10.36 Hz,
H-6b), 3.72 (dd, 1H, J1 ,2 ≈ J2 ,3 = 2.2 Hz, H-2ꢀ), 3.68 (dd, 1H,
ꢀ
ꢀ
ꢀ ꢀ
J1 ≈ J2 = 2.6 Hz, H-2ꢀꢀ), 3.56 (dd, 1H, J3,4 ≈ J4,5 = 9.2 Hz,
ꢀꢀ ꢀꢀ
,2
ꢀꢀ ꢀꢀ
,3
H-4), 3.53–3.41 (m, 6H, OCH3, H-2, H-5, H-4ꢀ), 2.68 (d, 3H,
J = 4.9 Hz, –NHCH3), 1.84 (s, 3H, –NHCOCH3), 0.91 (d, 3H,
J = 6.2 Hz, H-6ꢀ). Anal. calcd for C64H72N2O15: C, 69.30; H,
6.54; N, 2.53%. Found: C, 69.27; H, 6.63; N, 2.71%. ES HRMS:
(M + Na): 1131.4830, found: 1131.4837.
Methyl 2-propionamido-3-O-(3-O-(6-amino-a-L-mannopyra-
nuronyl)-a-L-rhamnopyranosyl)-b-D-glucopyranoside (4). The
hydrogenolysis of amide 26 (56.1 mg, 50.6 lmol) was performed
as described for that of protected lactam 22, using identical
workup and purification to afford a clear glass was lyophilized
to white powder (23.4 mg, 81%): [a]2D2 = −63.54 (c = 0.54, H2O).
1H NMR (600 MHz, D2O) d 8.22 (bd, 1H, –NHCOEt), 5.02 (s,
1H, H-1ꢀꢀ), 4.81 (s, 1H, H-1ꢀ), 4.12 (d, 1H, J = 8.7 Hz, H-1), 4.12
ꢀ
ꢀꢀ
ꢀ
ꢀ
ꢀ ꢀ
3.88–3.84 (m, 2H, H-5 , H-6a ), 3.81 (dd, 1H, J1 ,2 ≈ J2 ,3
=
2.5 Hz, H-2ꢀ), 3.78–3.72 (m, 4H, H-6b, H-2ꢀꢀ, H-4ꢀꢀ, H-5ꢀꢀ), 3.65
ꢀꢀ
(dd, 1H, Jgem = 11.5 Hz, Jvic = 7.3 Hz, H-6b ), 3.55–3.47 (m,
3H, H-4, H-5, H-4ꢀꢀ), 3.42 (s, 3H, OCH3), 3.27 (bddd, 1H, J1,2
≈
J2,3 ≈ J2,N–H = 9.0 Hz, H-2), 2.74 (bs, 1H, 6ꢀꢀ-OH), 1.76 (s,
3H, NHCOCH3), 0.86 (d, 3H, J = 6.2 Hz, H-6ꢀ). Anal. calcd
for C63H71NO15: C, 69.92; H, 6.61; N, 1.29%. Found C, 69.80;
H, 6.38; N, 1.35%. ES HRMS: (M + Na): 1104.4721, found
1104.4719.
(d, 1H, J = 9.8 Hz, H-5ꢀꢀ), 4.08 (dd, 1H, J1 ≈ J2 = 1.7 Hz,
ꢀꢀ ꢀꢀ
,2
ꢀꢀ ꢀꢀ
,3
H-2ꢀꢀ), 4.03 (dq, 1H, J = 9.8, 6.2 Hz, H-5ꢀ), 3.93 (d, 1H, Jgem
=
12.3 Hz, H-6a), 3.91 (d, 1H, J = 3.3, 9.7 Hz, H-3ꢀꢀ), 3.87 (dd,
1H, J1 ,2 ≈ J2 ,3 = 1.4 Hz, H-2ꢀ), 3.84–3.77 (m, 3H, H-2, H-3ꢀ,
H-4ꢀꢀ), 3.74 (dd, 1H, Jgem = 12.3 Hz, Jvic = 6.0 Hz, H-6b), 3.59
(dd, 1H, J3,4 ≈ J4,5 = 8.5 Hz, H-3), 3.55–3.43 (m, 6H, OCH3,
H-4, H-5, H-4ꢀ), 2.27 (q, 2H, J = 7.7 Hz, COCH2CH3), 1.24 (d,
3H, J = 6.2 Hz, H-6ꢀ), 1.11 (t, 2H, J = 7.7 Hz, COCH2CH3).
ES HRMS: (M + Na): 593.2164, found: 593.2159.
Methyl 4,6-O-benzylidene-2-propionamido-3-O-(2,4-di-O-
benzyl-3-O-(2,3,4-tri-O-benzyl-6-amino-a-L-mannopyranuronyl)-
a-L-rhamnopyranosyl)-b-D-glucopyranoside (26). To a stirred
solution of alcohol 24 (0.217 g, 0.198 mmol) in acetonitrile
ꢀ
ꢀ
ꢀ ꢀ
◦
(5 mL) at 0 C was added TEMPO (0.031 g, 0.198 mmol) and
an aqueous solution (1.30 mL, pH 10.5 with NaOH) containing
KBr (0.198 mmol) and NaHCO3 (0.040 mmol). NaOCl (5–6%,
1.50 mL) was added dropwise and the reaction stirred for
1 h before the addition of HCl (1 M, 15 mL) and CH2Cl2
(15 mL). The organic layer was removed, washed with water,
dried with Na2SO4 and evaporated to dryness. The residue was
dissolved in dry CH2Cl2 (5 mL) under Ar, carbonyldiimidazole
(0.064 g, 0.396 mmol) was added and the mixture was stirred for
45 min before the introduction of ammonia gas (XS). After an
additional hour, the ammonia was allowed to evaporate and the
solvent was removed under a vacuum. Column chromatography
(2 : 3 : 1 cyclohexane–ethyl acetate–acetone) afforded the target
compound as a clear glass (0.149 g, 68%): [a]2D2 = −22.72 (c =
Methyl 2-acetamido-3-O-(3-O-(6-N-methylamino-a-L-mann-
opyranuronyl)-L-(rhamnopyranosyl)-b-D-glucopyranoside
(5).
The hydrogenolysis of amide 27 (92.4 mg, 83.3 lmol) was
performed as described for that of protected lactam 22, using
identical workup and purification to afford a clear glass that
was lyophilized to white powder (37.0 mg, 78%): [a]2D2 = −50.9
1
(c = 0.64, H2O). H NMR (600 MHz, D2O) 5.21 (d, 1H, J =
1.7 Hz, H-1ꢀꢀ), 4.85 (d, 1H, J = 1.7 Hz, H-1ꢀ), 4.52 (d, 1H, J =
8.6 Hz, H-1), 4.14 (d, 1H, J = 9.2 Hz, H-5ꢀꢀ), 4.12 (dd, 1H, J =
1.7, 3.3 Hz, H-2ꢀꢀ), 4.06 (dq, 1H, J = 6.22, 9.81 Hz, H-5ꢀ), 3.98
(dd, 1H, Jgem = 10.2 Hz, Jvic = 2.2 Hz, H-6a), 3.95 (dd, 1H,
J = 3.3, 9.7 Hz, H-3ꢀꢀ), 3.88 (dd, 1H, J = 2.0, 2.30 Hz, H-2ꢀ),
3.87–3.77 (m, 4H, H-2, H-6b, H-3ꢀ, H-4ꢀꢀ), 3.62 (dd, 1H, J = 8.5,
10.0 Hz, H-3), 3.59–3.48 (m, 6H, OCH3, H-4, H-5, H-4ꢀ), 2.86
(s, 3H, –NHCH3), 1.24 (d, 3H, J = 6.2 Hz, H-6ꢀ). ES HRMS:
(M + Na): 593.2170, found 593.2171.
1
1.08, CHCl3). H NMR (600 MHz, CDCl3) d 7.48 7.11 (m,
30H, Ar–H), 6.58 (bs, 1H, –CONH2a), 6.26 (bd, 1H, J =
7.6 Hz, –NHCOEt), 5.51 (s, 1H, PHCHO2–), 5.46 (bs, 1H,
–CONH2b), 5.39 (d, 1H, J = 1.8 Hz, H-1ꢀꢀ), 4.97 (d, 1H, J =
1.7 Hz, H-1ꢀ), 4.82 (d, 1H, Jgem = 10.7 Hz, PhCH2O–), 4.69 (d,
1H, J = 8.3 Hz, H-1), 4.59–4.49 (m, 6H, PhCH2O–), 4.39–4.32
(m, 3H, PhCH2O-, H-6a), 4.18–4.13 (m, 2H, H-3, H-3ꢀ), 4.09
Purification of monoclonal antibody SYA/J6. Antibody was
purified from ascites fluid by centrifugation (30 min, 64 000 g)
to pellet cells and fatty tissue. After filtration first through a
Millex AP 20 prefilter (Millipore) and then through a Millex-GV
0.22 lm low-binding sterilization filter (Millipore), the filtrate
was loaded onto a Sepharose Protein-A (Pharmacia Biotech)
column equilibrated with running buffer (50 mM Tris, 150 mM
NaCl, 0.02% NaN3, adjusted to pH 8.0). The column was
washed with running buffer until serum proteins were eluted
(absorbance at 280 nm below 0.1). Antibody was then eluted
with citrate buffer (100 mM citric acid, 150 mM NaCl, 0.02%
NaN3, adjusted to pH 3.0). Fractions with an absorbance greater
than 0.1 were collected, pooled and dialyzed (24 h) against
the initial Tris buffer for calorimetry measurements, or against
phosphate buffer saline (PBS) for ELISA measurements.
(dd, 1H, J3 ,4 ≈ J4 ,5 = 9.0 Hz, H-4ꢀ), 4.02 (d, 1H, J = 9.2 Hz,
H-5ꢀꢀ), 3.89 (dq, 1H, J = 6.1, 9.5 Hz, H-5ꢀ), 3.85 (dd, 1H, J =
2.9, 8.5 Hz, H-3ꢀꢀ), 3.77–3.72 (m, 2H, H-6b, H-2ꢀ), 3.69 (dd, 1H,
ꢀ
ꢀ
ꢀ ꢀ
ꢀꢀ ꢀꢀ
ꢀꢀ ꢀꢀ
ꢀꢀ ꢀꢀ
ꢀꢀ ꢀꢀ
J1
≈ J2
= 2.6 Hz, H-2ꢀꢀ), 3.54 (dd, 1H, J3
≈ J4
=
,2
,3
,4
,5
9.3 Hz, H-4ꢀꢀ), 3.51–3.44 (m, 3H, H-4, H-5, H-4ꢀ), 3.42 (s, 3H,
OCH3), 2.10, 2.01 (m, 2H, –NHCOCH2CH3), 1.01 (t, 3H,
–NHCOCH2CH3), 0.84 (d, 3H, J = 6.1 Hz, H-6). Anal. calcd
for C64H72N2O15: C, 69.30; H, 6.54; N, 2.53%. Found: C, 69.15;
H, 6.48; N, 2.82%. ES MS: (M + Na): 1131.5, found: 1131.5.
Methyl 4,6-O-benzylidene-2-acetamido-3-O-(2,4-di-O-benzyl-
3-O-(2,3,4-tri-O-benzyl-6-N-methylamino-a-L-mannopyranuronyl)-
a-L-rhamnopyranosyl)-b-D-glucopyranoside (27). Using the
same protocol for the production of 26, alcohol 25 (0.179 g,
0.182 mmol) was oxidized and converted to its corresponding
amide using CDI (0.061 g, 0.377 mmol) and methylamine
(0.10 mL, 2.0 M solution in THF) to afford the title compound
Enzyme-linked immunosorbent assays. Direct ELISA was
carried out with a PBS solution of protein A purified
SYA/J6 antibody coated on microtitre plates (1 lg mL−1) and
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 7 2 3 – 2 7 3 2
2 7 3 1

View Article Online
O-polysaccharide coupled to biotin.21 Streptavidin–HRP conju-
gate (Sigma Chemical Co., St. Louis, Mo.) served as the disclos-
ing reagent. Inhibition experiments with increasing concentra-
tion of synthetic inhibitors were done in triplicate according to
a previously reported protocol.22,30
3 (a) K.-A. Karlsson, Trends Pharmacol. Sci., 1991, 12, 1; (b) M. S.
Mulligan, J. C. Paulsen, S. De Frees, Z.-L. Zheng, J. B. Lowe and
P. A. Ward, Nature, 1993, 364, 149; (c) C. A. A. van Boeckel and
M. Petitou, Angew. Chem., Int. Ed. Engl., 1993, 32, 1671; (d) J. C.
McAuliffe and O. Hindsgaul, Chem. Ind., 1997, 170.
4 (a) R. U. Lemieux, Chem. Soc. Rev., 1989, 18, 347; (b) R. U. Lemieux,
L. T. J. Delbaere, H. Beierbeck and U. Spohr, Host–Guest Molecular
Interactions: From Chemistry to Biology, Wiley, London, UK, 1991,
p. 231; (c) R. U. Lemieux, Acc. Chem. Res., 1996, 29, 373.
5 (a) H. C. Kolb, Bioorg. Med. Chem. Lett., 1997, 7, 2629; (b)
W. Jahnke, H. C. Kolb, M. J. J. Blommers, J. L. Magnani and B.
Ernst, Angew Chem., Int. Ed. Engl., 1997, 36, 2603.
6 (a) D. R. Bundle, R. Alibes, S. Nilar, A. Otter, M. Warwas and
P. Zhang, J. Am. Chem. Soc., 1998, 120, 5317; (b) R. Alibes, D. R.
Bundle, J. Org. Chem.1998, 63, 6288; (c) S. A. Wacowich-Sgarbi and
D. R. Bundle, J. Org. Chem., 1999, 64, 9080; (d) P. Zhang and D. R.
Bundle, Isr. J. Chem., 2000, 40, 189.
7 (a) N. Navarre, A. H. van Oijen and G. J. Boons, Tetrahedron Lett.,
1997, 38, 2023; (b) N. Navarre, N. Amiot, A. van Oijen, A. Imberty,
A. Poveda, J. Jimenez-Barbero, A. Cooper, M. A. Nutley and G. J.
Boons, Chem. Eur. J., 1999, 5, 2281; (c) M. C. Galan, A. P. Venot,
J. Glushka, A. Imberty and G. J. Boons, J. Am. Chem. Soc., 2002,
124, 5964; (d) M. C. Galan, A. P. Venot, J. Glushka, A. Imberty and
G. J. Boons, Org. Biomol. Chem., 2003, 1, 3891.
Isothermal titration microcalorimetry. Dialyzed antibody
was concentrated and equilibrated against the buffer used for
calorimetry in CentriPrep units (Amicon). Antibody concentra-
tion was determined spectrophotometrically, using a calculated
extinction coefficient of 1.53 mg mL−1.22 Dry saccharide samples
were prepared as a 0.50 mM stock solution and assayed for car-
bohydrate concentration as previously described.25a Isothermal
titration measurements were made using the Microcal VP-ITC
titration microcalorimeter.6a,25a Purified antibody (IgG, 30 lM)
was placed in the cell and titrated with ligand. In all cases,
the value C, defined as the product of the binding constant K
and the concentration of binding sites, was in the range of 1–
500 and ligand concentrations were such that the final ligand
concentration was at least 10KD. Titrations of the antibody and
data processing were carried out as previously described.6a,25a
The measurements were evaluated by the ORIGIN software
package (MicroCal Inc., Northampton, MA) and data were
processed using a single binding site model that assumes no
cooperativity between sites.
Heat capacity measurements were made using the Microcal
MCS titration microcalorimeter. The calorimeter cell held a
volume of 1.3215 mL. For all titrations, solubilized antibody
was placed in the cell and titrated with ligand. Antibody
concentrations ranged from 20 to 100 lM. Titrations of the
antibody were carried out at pH 8.0 in 50 mM Tris buffer
augmented with 150 mM NaCl and 0.02% NaN3. All ligands
were dissolved in an identical buffer. Typically, for antibody–
carbohydrate concentrations, 30 8 lL injections of ligand, 20 s
in duration, were made, with 4 min intervals between injections.
The data was processed as previously described using the
ORIGIN software package.6a Non-negligible heats of dilution
were subtracted prior to data processing. Constant pressure
heat capacities, DCp, were determined by evaluating DH as
a function of T. Heat capacity is rigorously defined as the
partial derivative of DH with respect to temperature at constant
pressure, i.e. DCp = (dH/dT)P. Over short temperature ranges
DCp is predicted to be linear, and can be approximated as
D(DH)/DT, or (DH1 − DH2)/(T1 − T2).
8 M. Wilstermann, J. Balogh and G. Magnusson, J. Org. Chem., 1997,
62, 3659.
9 (a) H. R. Hanna and D. R. Bundle, Can. J. Chem., 1993, 71,
125; (b) F.-I. Auzanneau and D. R. Bundle, Carbohydr. Res., 1993,
247, 195; (c) D. R. Bundle, in Carbohydrates, ed. S. Hecht, Oxford
University Press Inc., Oxford, 1998, p. 370.
10 D. R. Bundle, Pure Appl. Chem., 1989, 61, 1171.
11 N. K. Vyas, M. N. Vyas, M. C. Chervenak, M. A. Johnson, B. M.
Pinto, D. R. Bundle and F. A. Quiocho, Biochemistry, 2003, 41,
13575.
12 N. I. A. Carlin, A. A. Lindberg, K. Bock and D. R. Bundle,
Eur. J. Biochem., 1984, 139, 189.
13 M. Mammen, E. I. Shakhnovich, J. M. Deutch and G. Whitesides,
J. Org. Chem., 1998, 63, 3168.
14 (a) D. R. Bundle and B. W. Sigurskjold, Methods Enzymol., 1994,
247, 288; (b) B. W. Sigurskjold, E. Altman and D. R. Bundle,
Eur. J. Biochem., 1991, 197, 239.
15 V. Pozsgay and H. J. Jennings, J. Org. Chem., 1988, 53, 4042.
16 N. K. Kochetkov, N. E. Byramova, Yu. E. Tsvetkov and L. V.
Backinowsky, Tetrahedron, 1985, 41, 3363.
17 G. H. Veeneman, S. H. van Leeuwen, H. Zuurmand and J. H. van
Boom, J. Carbohydr. Chem., 1998, 9, 783.
18 K. Bock, I. Lundt and C. Pedersen, Tetrahedron Lett., 1973, 14, 1037.
19 E. J. Corey and A. Venkateswarlu, J. Am. Chem. Soc., 1972, 94, 6190.
20 R. Knorr, Tetrahedron Lett., 1989, 30, 1927.
21 N. I. Davis and S. L. Flitsch, Tetrahedron Lett., 1993, 34, 1181.
22 P. J. Meikle, N. M. Young and D. R. Bundle, J. Immunol. Methods,
1990, 132, 255.
Acknowledgements
23 S. W. Homans, Biochemistry, 1990, 29, 9110.
24 R. U. Lemieux, M.-H. Du and U. Spohr, J. Am. Chem. Soc., 1994,
116, 9803.
25 (a) M. Chervenak and E. Toone, J. Am. Chem. Soc., 1994, 116, 10533;
(b) M. C. Chervenak and E. J. Toone, Biochemistry, 1995, 34, 5685.
26 S. D. Sharrow, N. V. Novotny and M. J. Stone, Biochemistry, 2003,
42, 6302.
27 S. D. Sharrow, K. A. Diamonds, M. A. Goodman, M. V. Novotny
and M. J. Stone, Protein Sci., 2005, 14, 249.
28 X. Ysern, B. A. Fields, T. N. Bhat, F. A. Goldbaum, W. Dall’Acqua,
F. P. Schwarz, P. J. Poljak and R. A. Mariuzza, J. Mol. Biol., 1994,
238, 496.
Financial support was provided by research grants to D. R. Bun-
dle from the Natural Science and Engineering Research Council
(NSERC) and the University of Alberta. R. S. McGavin received
NSERC, Alberta Heritage Foundation for Medical Research
and Alberta Research Council postgraduate scholarships. The
authors thank Mrs Joanna Sadowska for the preparation and
purification of monoclonal antibody.
References
1 J. P. Carver, Pure Appl. Chem., 1993, 65, 763.
2 P. I. Kitov, J. M. Sadowska, G. Mulvey, G. D. Armstrong, H. Ling,
N. S. Pannu, R. J. Read and D. R. Bundle, Nature, 2000, 403, 669.
29 F. A. Goldbaum, F. P. Schwartz, E. Eisenstein, A. Cauerhff, R. A.
Mariuzza and R. J. Poljak, J. Mol. Recognit., 1996, 9, 6.
30 E. Vorberg and D. R. Bundle, J. Immunol. Methods, 1990, 132, 81.
2 7 3 2
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 7 2 3 – 2 7 3 2