C-disaccharides as probes for carbohydrate recognition - Investigation of the conformational requirements for binding of disaccharide mimetics of sialyl Lewis X

Article abstract of DOI:10.1002/ejoc.200600825

A set of C-disaccharide analogs was designed to probe the recognition of a known O-disaccharide mimetic of sialyl Lewis X, to P-selectin. The synthesis of the C-glycosides centered on the de novo construction of the galactose residue via an oxocarbenium i

Full text of DOI:10.1002/ejoc.200600825

FULL PAPER
DOI: 10.1002/ejoc.200600825
C-Disaccharides as Probes for Carbohydrate Recognition – Investigation of the
Conformational Requirements for Binding of Disaccharide Mimetics of Sialyl
Lewis X
Richard W. Denton,[a,b] Xuhong Cheng,[a,b] Kurissery A. Tony,[a,b] Anna Dilhas,[a,b]
Juan José Hernández,[c] Angeles Canales,[c] Jesús Jiménez-Barbero,[c] and
David R. Mootoo*[a,b]
Keywords: Carbohydrates / Glycomimetics / Selectin / C-Glycoside / Oxocarbenium ion / Conformation
A set of C-disaccharide analogs was designed to probe the
recognition of a known O-disaccharide mimetic of sialyl
Lewis X, to P-selectin. The synthesis of the C-glycosides cen-
tered on the de novo construction of the galactose residue via
an oxocarbenium ion/enol ether cyclization. Conformational
analysis was performed by a combination of NMR spec-
troscopy and molecular mechanics (MM) and molecular dy-
namics (MD) calculations. The inhibition of P-selectin bind-
ing was evaluated in a P-selectin Biacore assay. At 12 mM,
the O-glycoside showed 48% inhibition of binding, while the
C-glycoside analogs exhibited between 25–31% inhibition.
This data is discussed within the context of the active confor-
mation of sLex and the conformational behavior of these li-
gands.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2007)
ported to be five and forty times more active than sLex
against E- and P-selectin, respectively, in a cell-based as-
say.[15,16] The design of 2 was based on a sLex-selectin bind-
ing model, and it was suggested that the galactose and man-
nose segments mimic the galactose and fucose residues in
sLex, respectively (Figure 2).[17–21] We were interested in the
C-glycoside analogs of 2 for two reasons. First, C-glycosides
are more hydrolytically stable than O-glycosides and could
be more practical for drug development.[22–27] Second, the
C-glycoside framework allows for design of analogs with
different conformational properties with respect to the in-
tersaccharide linkages, thereby leading to ligands with dif-
ferent spatial orientation between receptor contacts in the
galactose and mannose segments.[28] Evaluation of the se-
lectin affinity of such structures could provide insight on
the optimal conformational requirements for binding of this
disaccharide framework.[29,30] Against this backdrop we ini-
tiated an investigation on the synthesis, conformational
analysis and P-selectin binding of the O- and C-disaccha-
rides 2–5 (Figure 1). P-selectin has attracted special atten-
tion as a therapeutic target because it is expressed on the
endothelium within minutes after stimulation and therefore
inhibition of sLex-P-selectin binding could allow for early
disruption of the inflammation cascade.[14,31,32] We have
previously described the synthesis of the C-glycoside 3 and
precursors to the conformationally restrained analogs 4 and
5, and the conformational behavior of 2 and 3.[33–36] Herein,
we report the synthesis and conformational analysis of 4
and 5, and the P-selectin binding of the four disaccharide
probes 2–5.
Introduction
The interaction of the sialyl Lewis X (sLex) tetrasaccha-
ride 1 on the surface of leukocytes and E- and P-selectin
on the vascular endothelieum occurs at the onset of the
recruitment of leukocytes during the inflammation response
(Figure 1).[1,2] This low affinity interaction slows the leuko-
cytes to a roll, after which higher affinity integrin-based
binding ensues leading to infiltration of the leukocytes into
the underlying tissue. The interaction of sLex and the selec-
tins has attracted considerable interest as a therapeutic tar-
get in view of the increase in selectin expression during in-
flammation, the association of unregulated inflammation
with several pathological states (such as stroke, reperfusion
injury and cardiac and allergic diseases), and the fact that
sLex-selectin binding is low affinity relative to other cellular
adhesion processes.[3,4]
One approach in the regulation of sLex-selectin binding
is the development of small molecule mimetics of sLex
which can act as selectin antagonists.[5–14] An attractive lead
compound is the 1,1-linked disaccharide 2, which is re-
[a] Department of Chemistry, Hunter College,
695 Park Avenue New York, NY 10021, USA
Fax: +1-212-772-5332
E-mail: dmootoo@hunter.cuny.edu
[b] Department of Chemistry, The Graduate Center, CUNY, 365,
5th Avenue, New York, NY 10016, USA
[c] Centro de Investigaciones Biológicas, CSIC,
Ramiro de Maeztu 9, 28040 Madrid, Spain
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Eur. J. Org. Chem. 2007, 645–654
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
645

D. R. Mootoo et al.
FULL PAPER
Figure 2. Model for sLex-P- Selectin binding.[21]
tected hydroxy acid precursor 13 (instead of the MOM de-
rivative 8) in the C-glycoside synthesis, thereby leading to
the PMB C-disaccharide 14. Conversion of 14 to the diol 12
was best accomplished by first converting 14 to the acetate
derivative, followed by DDQ-mediated removal of the PMB
ether, and deacetylation of the product. The analogous syn-
thesis of the diol 10 from the corresponding PMB-protected
hydroxy acid precursor also proceeded smoothly.
Figure 1. Disaccharide mimetics of sLex.
Results
Synthesis
The synthesis of 3–5 entailed the de novo construction
of the galactose residue and centered on a novel oxocarb-
enium ion/enol ether cyclization approach to C-glycosides.
In this vein, we have previously reported the synthesis of
3, and 9 and 11, potential precursors to conformationally
restrained analogs 4 and 5 (Scheme 1).[33,34] The initial plan
was to convert the hydroxy-MOM acetals 9 and 11 in a
single step to their methylene acetal derivatives.[37] However,
the acidic conditions required for this transformation led
to competing cleavage of the isopropylidene and silyl ether
Scheme 1. Reagents and conditions: (a) ref.[34] and supporting in-
protecting groups. Therefore, in an alternative stepwise formation; (b) Me2BBr, DTBMP, CH2Cl2, –78 to 0 °C, 72%; (c)
6, DCC, DMAP, PhH, 87%; (d) Tebbe reagent, 79%; (e) MeOTf,
strategy, selective removal of the MOM acetals in 9 and 11
DTBMP, CH2Cl2, 87%; (f) BH3·DMS, then Na2O2, 75%; (g) (i)
to afford the diols 10 and 12, was next attempted. Dimethyl-
Ac2O, DMAP, EtOAc; (ii) DDQ, CH2Cl2/H2O; (iii) NaOMe,
boron bromide was found to be the optimal promoter for
MeOH, 74%.
this transformation, but this procedure was only useful for
the conversion of 9 to 10. Application of these conditions
The methylene acetal derivatives 15 and 16 were obtained
to 11 led to a mixture of several products, with a low yield by individual treatment of 10 and 12 with a mixture of di-
of 12. The complications with the MOM protecting group bromomethane and aqueous sodium hydroxide under
were avoided by using the p-methoxybenzyl (PMB) pro- phase-transfer conditions (Scheme 2).[38] These reactions
646
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 645–654

C-Disaccharides as Probes for Carbohydrate Recognition
FULL PAPER
were sometimes accompanied by small amounts of desi-
lylated products, which could be easily reprotected. It was
also found that small amounts of methyl ether side products
were obtained, and these were suppressed by performing
the reaction in the presence of 2,3-dimethylbutene. In prep-
aration for the final alcohol protecting group modifications,
the isopropylidene residues in 15 and 16 were selectively
removed to give the respective diols 17 and 18. For charac-
terization purposes 17 and 18 were transformed via
straightforward alcohol protecting group changes, to their
peracetates 19 and 20. The stereochemistry of the aglycon
segment and the configuration at the intersaccharide car-
bon were assigned on the basis of vicinal J values. Thus,
J1,2 = 10.0, J2,3 = 10.0, J3,4 = 3.0, J4,5 = 0 Hz for 19, and
J1,2 = 9.8, J2,3 = 9.8, J3,4 = 3.0, J4,5 = 0 Hz for 20 are consis-
Scheme 3. Reagents and conditions: (a) Bu2SnO, PhCH3,
BrCH2CO2Me, nBu4NI, 21 (98%), 22 (77%); (b) aq. NaOH,
EtOH, (c) H2, Pd/C, HCOOH, MeOH, then aq. NaOH.
Conformational Analysis
tent with the 3,4-O-isopropylidene-β-C-galacto motif.[39]
A
J1,1Ј value of 9.8 Hz and a NOE between H2 of the galac-
tose residue and the intersaccharide proton for 20 pointed
strongly to an equatorial-like attachment of the mannose
residue onto a chair-like dioxane ring. The corresponding J
value for 19 (6.5 Hz) is somewhat larger than expected for
equatorial–axial arrangement of vicinal protons on a chair-
like dioxane. It appears that the bulky pseudo-axial substit-
uent results in a distorted, half-chair-like geometry, leading
to the unexpectedly large J value. These stereochemical con-
clusions were subsequently corroborated by NMR analysis
of 25 and 26, the deacetylated derivatives of 19 and 20,
respectively (vide infra).
Using a combination of NMR spectroscopy and molecu-
lar mechanics (MM) and molecular dynamics (MD) calcu-
lations, we had previously investigated the conformational
properties of 2 and 3, and their respective analogs without
the glycolate residue at position 3 of the galactose segment,
23 and 24 (Figure 3).[36] These studies indicated that the
presence or absence of glycolate substituent did not have a
significant effect on the conformational behavior with re-
spect to the intersaccharide torsions. Therefore, in the pres-
1
ent study, for simplicity in H NMR signal resolution, we
examined 25 and 26 and assumed that the conformational
properties of 23–26 would mimic those of 2–5, respectively.
Scheme 2. Reagents and conditions: (a) nBu4NBr, CH2Br2, 50%
aq. NaOH, 15 (86%), 16 (83%); (b) MeOH, HCl, 17 (68%), 18
(64%); (c) nBu4NF, THF, (d) H2, Pd/C, HCOOH, MeOH; (e)
Ac2O, EtOAc, DMAP.
Figure 3. Analogs for conformational analysis.
To summarize our earlier result, the C-glycoside 24 in
water populates five different conformational families A–E,
Finally, the diols 17 and 18 were transformed to 4 and 5 defined by the glyconic torsions (Φgal, Φman).[40] [Figure 4].
by a reaction sequence that was similar to that used in our In comparison, the O-glycoside 23 exists primarily in con-
earlier synthesis of 3 (Scheme 3)[33] Thus, selective dibu- formational family A with very minor populations of con-
tyltin oxide mediated alkylation of 17 and 18 with methyl formational families B, D and E.
bromoacetate led to selective 3-O-alkylation followed by in
As was the case for 23 and 24, molecular mechanics and
situ lactonization to give 21 and 22, respectively. Exposure dynamics calculations on the conformationally restrained
of these products to aqueous sodium hydroxide led to con- analogs 25 and 26 were performed using the MM3* force
comitant saponification and desilylation to give the corre- field (MACROMODEL 7.1).[41,42] However, unlike the case
sponding dihydroxy acids, which were subjected to hydro- of the more flexible C-glycoside 24, for which a time-
genolysis. The target compounds 4 and 5 were obtained af- averaged restrained MD protocol was implemented, no re-
ter purification using both reverse and normal phase straints were used for modeling the conformational behav-
chromatography and lyophilization from aqueous solutions. ior of 25 and 26. Obviously, the presence of the cyclic acetal
Eur. J. Org. Chem. 2007, 645–654
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
647

D. R. Mootoo et al.
FULL PAPER
the experimental J values as before indicated that 26 exists
as a 20:80 ratio of A/B. Thus, in the case of 26, the presence
of conformation A is overestimated by the MM3* calcula-
tions. Observed strong key NOE contacts (H-intersac/H1-
man and H1-gal/H2-man) are also in agreement with a
major presence of conformation B.
Figure 4. Conformations for 23 and 24. Structures and exact dihe-
drals are shown for C-glycoside 24. The % populations of the con-
formational families for 23 and 24 are given in parentheses.
Figure 5. Conformational analysis for 25.
in these latter structures severely restricts the conforma-
tional mobility around the Gal pseudoglycosidic linkage
and only the torsional degree of freedom around the Man
linkage remains. Thus, the problem is highly simplified, and
the conformers around Φman were generated and optimized
with MM3* with Φgal left free during the minimization pro-
cess. The GB/SA solvation model for water was used.[43]
The probability distribution was calculated from the energy
values according to a Boltzmann function at 300 K.
Three low energy minima were obtained for 25 (Figure 5).
The lowest and one of the high-energy conformations
(Φgal,Φman: 175, 51 and 151, –59, respectively), corre-
sponded to conformational families D and E that were pre-
viously observed for 23 and 24. The highest energy confor-
mation, F (Φgal, Φman: 168, 140) was not observed for 23
and 24. Intersaccharide coupling constants for these three
conformations were obtained from the Karplus–Altona re-
lationship[44] and compared with the values measured from
1
the H NMR (Figure 5). Accordingly, 25 exists predomi-
Figure 6. Conformational analysis for 26.
nantly in conformation D (i.e. ca. 85%), with minor popula-
tions of E and F (i.e. 10 and 5%, respectively). Thus the
relative population distribution was in qualitative agree-
ment with the result predicted by MM3* calculations. The
observation of strong key NOE contacts (H-intersac/H3-
P-Selectin Binding
The competitive binding of the O-disaccharide 2 and the
man, H-intersac/H5-man, OCH-axO- H1-man and H2-gal/ C-disaccharides 3–5 to a soluble truncated form of human
H1-man) are also in agreement with a high population of P-selectin was next evaluated in a Biacore assay with an
conformation D. The MM3* calculations for 26 indicated immobilized monomeric truncated form of human PSGL-1
low and high-energy minima corresponding to conforma- as the reference ligand.[45] At 12 m, 2, 3, 4 and 5 showed
tional families A and B, respectively (Figure 6). Analysis of 48, 26, 25 and 31% inhibition, respectively.[46] The IC50 of
648
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 645–654

C-Disaccharides as Probes for Carbohydrate Recognition
FULL PAPER
sLex under these conditions was previously found to be
15 m.[14] It should be noted that our result for the O-disac-
charide 2 is in disagreement with an earlier study in which
the binding of 2 to P-selectin was found to be 40 times
greater than sLex.[16] This inconsistency could be due to the
fact that the latter investigation used a dynamic cell-based
assay,[47] but this explanation is not completely satisfying
because the IC50 values for sLex in both the Biacore and
cell-based measurement are similar (ca. 8 vs. 15 m, respec-
tively).
Discussion
It was anticipated that the different relative spatial posi-
tioning of the two sugar residues in 2–5 would lead to sig-
nificantly different activities. In so far as % inhibition could
be used as a measure of binding, this information could
help define an optimal conformation for binding of the di-
saccharide framework. However, the similarity of activity
Figure 7. Comparison of binding of sLex, and conformations B and
D.
of 2–5 makes any such conclusions somewhat conjectural. unrestrained Φman torsion. Thus, for 4, in which Φgal is re-
Nevertheless, analysis of possible bound conformations in strained in approximately 170°, a Φman of 70° gives a D-like
terms of the sLex-P-selectin recognition model may be in- conformation (Φgal, Φman; 170°, 70°), which positions the
sightful with respect to the design of more active P-selectin carboxylate oxygen in 4 close to the Tyr48 binding site for
ligands (Figure 2). In this regard, the more conformation- Neu-COO– in sLex. The respective COO–/Tyr48–O dis-
ally rigid analogs 4 and 5, which have more limited modes tances for this conformation and that of bound sLex are 4.8
of binding, are an appropriate starting point. Accordingly, and 3.6 Å. In this D-like conformation Gal-6-O in 4, is far-
the galactose and mannose residues of 4 and 5 can individu- ther away from the carboxylate oxygen of Glu94, compared
ally, but not simultaneously mimic the interaction of the with Gal-6-O in sLex (3.5 vs. 2.5 Å), while the Gal-4-O/
galactose and fucose residues in sLex, respectively (Fig- Tyr92-O distance is closer for 4 compared with sLex (2.5 vs.
ure 7). The selection of which of the two subunits should 3.5 Å). For 5, in which Φgal is restrained to 60°, a Φman of
be docked into the appropriate carbohydrate domain of P- 60° leads to conformation B, which places the carboxylate
selectin was guided by the crystal structure of the sLex-P- of the glycolate in very similar proximity to Tyr48-O com-
selectin complex and existing structure activity data, which pared with the carboxylate in sLex (3.4 vs. 3.6 Å, respec-
suggests that the fucose residue accounts for the major part tively). However, in conformation B the oxygen atoms at
of the overall binding energy of sLex.[6,21,48,49] Therefore, we position 4 and 6 of the galactose segment is at a much fur-
speculate that the binding of the mannose segment of these ther distance from Tyr92-O and Glu94-COO–, respectively,
disaccharide frameworks is conserved, and closely mimics compared with the corresponding galactose positions in
that of the fucose in sLex (i.e. the 2-, 3- and 4-OH of the sLex, and apparently out of binding range (i.e. approxi-
mannose residue map to the 4-, 3- and 2-OH of fucose) mately 7 Å for gal-4-O/Tyr92-O and 9 Å for gal-6-O/Glu94-
[Figure 7]. In the case of the conformationally restrained COO–). Instead, Gal-6-O in conformation B is relatively
analogs 4 and 5, this means that individual galactose sub- close to Tyr92-O (5.2 Å), and appears more likely to inter-
stituents must occupy different spatial positions relative to act with this residue. Therefore, to summarize, the salient
the mannose segment, and would therefore interact dif- points on the possible binding of B and D type frameworks
ferently with the receptor sites in the galactose-binding do- are an essentially identical mode of binding of their man-
main of sLex. Inspection of the sLex-P selectin complex sug- nose segments, but differences in the receptor interactions
gests these to be Tyr48, Tyr92 and Glu94, which interact in their galactose segments, i.e. 4 appears capable of a rela-
with Neu-COO-, Gal-4-OH and Gal-6-OH of sLex.[21]
tively strong Gal-4-OH/Tyr92 and weak COO–/Tyr48 and
The assumption that the mannose mimics the fucose of Gal-6-OH/Glu94 interactions, whereas for 5 a strong
sLex, the conformational restraints on 4 and 5, and the no- COO–/Tyr48, and weak Gal-6-OH/Tyr92 appear to be the
tion that the relative position of the carboxylate of neur- case. Because unbound 4 and 5 favor conformations like D
aminic acid in sLex and the fucose is critical for binding, and B, respectively, and therefore expected to incur rela-
point to conformations like D (Φgal, Φman; 180°, 60°) and tively small reorganizational energetic costs on binding,
B (Φgal, Φman; 60°, 60°) as possible bound orientations for their similar binding might be an indication that these ga-
4 and 5, respectively (Figure 7). Considering the P-selectin lactose interactions are relatively weak compared to the
sites in the vicinity of the galactose-binding domain of sLex, binding of the mannose segment. As stated earlier, this situ-
conformations like B and D appear to give the best fit to ation is analogous to the apparent dominant binding of the
the receptor, relative to their rotamers with respect to the fucose residue in sLex. However, an alternative explanation
Eur. J. Org. Chem. 2007, 645–654
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
649

D. R. Mootoo et al.
FULL PAPER
that was correlated with TLC mobility. Chromatograms were ob-
served under UV (short and long wavelength) light, and/or were
visualized by heating plates that were dipped in a solution of am-
monium(VI) molybdate tetrahydrate (12.5 g) and cerium(IV) sul-
fate tetrahydrate (5.0 g) in 10% aqueous sulfuric acid (500 mL).
Optical rotations were recorded at 25 °C at 589 nm (sodium D-
line). 1H NMR spectra were recorded on 300, 400 and 500 MHz
instruments. Spectra were obtained for CDCl3, C6D6 and D2O
solutions with CHCl3, C6H6 and HOD, respectively, as internal
standards.
is that the combined set of carboxylate and galactose inter-
actions in D and is similar in magnitude to that for B. Ana-
logs of 4 and 5 with appropriate deletions of alcohol groups
may be useful in probing this issue.
It is appropriate to consider the binding of conforma-
tionally unrestrained analogs 2 and 3 in terms of conforma-
tions B and D. Because the O-glycoside 2 is preorganized
in conformation A (Φgal, Φman; ca. 60°, –60°), which is
closer in enthalpy to conformation B than D, a B-type
bound conformation for 2 might be more favorable than a
D-type conformation. Interestingly, minor variation
in the dihedral angles for the idealized B conformation
(Φgal, Φman; 60°, 60°), which would be energetically less de-
manding for 2 (compared with the more rigid framework
5), allows for B-like conformations that have closer COO–/
2,6:8,12-Dianhydro-1,3,4,5-tetra-O-benzyl-13-O-tert-butyldiphenyl-
silyl-10,11-O-isopropylidene-D-lyxo-D-galacto-D-manno-tridecitol
(10): Dimethylboron bromide (0.18 mL, 1.85 mmol) was added at
–78 °C, to a mixture of MOM-protected C-glycoside 9[50] (320 mg,
0.31 mmol), 2,6-di-tert-butyl-4-methylpyridine (252 mg,
1.23 mmol), freshly activated, powdered 4-Å molecular sieves
Tyr48-O and Gal-6-O/Tyr92-O contacts. Thus the larger en- (624 mg) and anhydrous CH2Cl2 (20 mL). The reaction mixture
was stirred at this temperature for 30 min, at room temp. for an
additional 30 min, quenched by addition of a 1:1 mixture of THF
and saturated aqueous NaHCO3 (10 mL), and extracted with di-
ethyl ether. After washing of the organic layer with 10% aqueous
NaHSO4 and brine, the organic layer was dried (Na2SO4), filtered
and concentrated under reduced pressure. FCC of the crude prod-
uct afforded 10 (221 mg, 72% yield based on 17) as a colorless
thalpy cost (compared to 5), required for the O-glycoside 2
to adopt a B-like conformation may be offset by a more
intimate fit of 2 to the receptor. The larger entropy cost
(relative to 5) that the flexible C-glycoside 3 would incur on
binding in a B-like conformation, may be similarly compen-
sated. Thus, a B-like active conformation for 2 and 3 may
also account for the similarity of their binding compared
with 5 (and 4).
1
oil; Rf = 0.17 (30% EtOAc/petroleum ether). H NMR (500 MHz,
C6D6): δ = 0.95 (s, 9 H), 1.25 (s, 3 H), 1.40 (s, 3 H), 3.13 (br. s, 1
H, D2O ex), 3.25 (dd, J = 5.5, 9.0 Hz, 1 H), 3.45 (dd, J = 3.0,
10.0 Hz, 1 H), 3.49 (t, J = 5.0 Hz, 1 H), 3.56 (d, J = 6.0 Hz, 1 H),
3.75–3.93 (m, 5 H), 4.00 (m, 2 H), 4.11 (m, 1 H), 4.27 (dd, J = 2.0,
4.8 Hz, 1 H), 4.30 (dd, J = 4.37 Hz, 1 H), 4.45–4.63 (m, 9 H), 7.25–
7.42 (m, 26 H), 7.70 (m, 4 H) ppm. 13C NMR (75 MHz, CDCl3):
δ = 26.6, 27.1, 28.6, 63.1, 69.4, 71.0, 71.2, 71.5, 72.4, 72.8, 73.1,
73.3, 73.8, 74.2, 75.0, 75.8, 76.3, 77.9, 80.3, 109.4, 127.7–129.8 (sev-
eral lines), 133.7, 135.7, 137.9, 138.2, 138.5, 138.6 ppm. FAB-
Summary
C-disaccharides with different conformational properties
about the intersaccharide linker were synthesized and evalu-
ated together with their parent O-disaccharide for binding
to P-selectin. These analogs were found to have comparable
activity to sLex. However, the small differences in their ac- HRMS calcd. for C60H70O11NaSi [M + Na] 1017.4585, found
1017.4557.
tivity did not permit any clear conclusions on a favored
bound conformation of the disaccharide framework. Pos-
sible bound conformations were considered based on the
torsional constraints of conformationally restrained ana-
logs and the X-ray structure of the sLex-P-selectin complex.
2,6:8,12-Dianhydro-1,3,4,5-tetra-O-benzyl-13-O-tert-butyldiphenyl-
silyl-10,11-O-isopropylidene-7-O-p-methoxybenzyl-D-lyxo-D-gulo-D-
manno-tridecitol (14): DCC (1.39 g, 6.74 mmol) was added at 0 °C
to a solution of thioacetal alcohol 6[33] (1.26 g, 2.48 mol), acid 13
That such conformationally different analogs show similar (2.10 g, 2.92 mmol, see supporting information), and DMAP
activity as sLex, raises questions about the extent to which
(59 mg, 0.49 mmol) in anhydrous benzene (30 mL). The reaction
the substituents on the galactose residue of sLex contribute
was warmed to room temp. and stirred for 3 h. The mixture was
diluted with diethyl ether and filtered. The filtrate was washed with
0.1  HCl and brine, dried (Na2SO4), filtered, and evaporated un-
der reduced pressure. The residue was purified by FCC to give the
to binding. Investigations aimed at a clearer understanding
of the selectin binding of this disaccharide framework,
using more finely tuned analogs of 4 and 5 are underway
and will be reported in due course. On a general note the
C-glycosides used in this study represent conformational
mimetics of O-glycosides that could find wider use as re-
cognition probes.
derived ester (2.61 g, 87 %) as a colorless oil; Rf = 0.55 (15 %
1
EtOAc/petroleum ether). H NMR (500 MHz, C6D6): δ = 1.05 (s,
9 H), 1.41 (s, 3 H), 1.47, (s, 3 H), 3.42 (dd, J = 3.0, 10.0 Hz, 1 H),
3.51 (dd, J = 3.0, 7.7 Hz, 1 H), 3.61 (dd, J = 4.7, 10.0 Hz, 1 H),
3.74 (m, 4 H), 3.87 (m, 3 H), 3.99 (t, J = 3.4 Hz, 1 H), 4.14–4.45
(m, 10 H), 4.57 (m, 3 H), 5.34 (m, 2 H), 6.75 (d, J = 8.6 Hz, 2 H),
7.12–7.55 (m, 28 H), 7.65 (m, 4 H) ppm. 13C NMR (75 MHz,
CDCl3): δ = 26.7, 27.0, 27.5, 55.3, 62.2, 69.3, 71.8, 72.0, 72.2, 72.9,
73.4, 74.3, 74.7, 75.7, 78.0, 79.1, 84.7, 111.6, 114.0, 127.3–129.8
(several lines), 132.7, 132.8, 133.0, 133.4, 135.5, 135.6, 138.3, 138.5,
138.7, 159.5, 169.6 ppm. FAB-HRMS calcd. for C73H80O12SiSNa
[M + Na] 1231.5038, found 1231.5037.
Experimental Section
General: Solvents were purified by standard procedures or used
from commercial sources as appropriate. Petroleum ether refers to
the fraction of petroleum ether boiling between 40 and 60 °C. Ether
refers to diethyl ether. Unless otherwise stated thin layer
chromatography (TLC) was done on 0.25-mm thick precoated sil-
ica gel 60 (HF-254) aluminum sheets and flash column chromatog-
raphy (FCC) was performed using Silica Gel 60 (32–63 mesh). Elu-
tion for FCC usually employed a stepwise solvent polarity gradient
Tebbe reagent in THF (12.4 mL, 0.5 , 6.2 mmol) was added drop-
wise at –78 °C under argon, to a solution of ester from the previous
step (1.61 g, 1.33 mmol) and pyridine (0.10 mL) in anhydrous 3:1
toluene:THF (27.0 mL). The reaction mixture was warmed to room
650
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 645–654

C-Disaccharides as Probes for Carbohydrate Recognition
FULL PAPER
temp., stirred at this temperature for 2 h, then slowly poured into
aqueous 1  NaOH at 0 °C. The resulting suspension was extracted
with diethyl ether, and the organic phase was washed with brine,
dried (Na2SO4), filtered, and concentrated under reduced pressure.
The residue was purified by FCC on basic alumina (Brockmann I,
150 mesh) to give the enol ether derivative (1.11 g, 79%) as a color-
less oil; Rf = 0.60 (10 % EtOAc/petroleum ether). 1H NMR
(500 MHz, C6D6): δ = 1.17 (s, 9 H), 1.56 (s, 3 H), 1.60 (s, 3 H),
3.25 (s, 3 H), 3.57 (dd, J = 5.0, 11.0 Hz, 1 H), 3.68 (d, J = 11.0 Hz,
1 H), 3.75 (dd, J = 5.0, 9.5 Hz, 1 H), 3.80 (m, 1 H), 3.96 (d, J =
9.5 Hz, 1 H), 4.03 (m, 2 H), 4.19–4.29 (m, 5 H), 4.36–4.52 (m, 6
H), 4.56 (A, of ABq, J = 11.5 Hz, ∆δ = 0.16 ppm, 1 H), 4.61 (A
of ABq, J = 12.5 Hz, ∆δ = 0.41 ppm, 1 H), 4.78, (B of ABq, J =
11.5 Hz, ∆δ = 0.16 ppm, 1 H), 4.81, (d, J = 9.5 Hz, 1 H), 4.85 (d,
J = 6.0 Hz, 1 H), 4.97 (B of ABq, J = 11.0 Hz, ∆δ = 0.41 ppm, 1
H), 6.27 (d, J = 6.0 Hz, 1 H), 6.71 (d, J = 8.0 Hz, 2 H), 6.95–7.28
(m, 29 H), 7.32 (s, 1 H), 7.39 (t, J = 7.0 Hz, 2 H), 7.46 (d, J =
7.0 Hz, 1 H), 7.76 (m, 4 H) ppm. 13C NMR (75 MHz, C6D6): δ =
27.6, 28.1, 55.3, 62.4, 70.6, 71.3, 71.9, 74.0, 75.0, 75.4, 75.8, 76.1,
76.4, 78.4, 81.0, 85.0, 88.2, 112.8, 114.5, 126.8–129.5 (several lines),
130.5, 132.2, 133.3, 133.8, 134.0, 135.4, 136.4, 139.7, 140.0, 157.8,
160.2 ppm. ESIHRMS calcd. for C72H82O13SiSNa [M + Na]
1229.5414, found 1229.5422.
11 Hz, 1 H), 6.77 (d, J = 8.5 Hz, 2 H), 7.11 (d, J = 8.5 Hz, 2 H),
7.17–7.44 (m, 26 H), 7.75 (m, J = 13.5 Hz, 4 H) ppm. 13C NMR
(75 MHz, CDCl3): δ = 26.7, 27.1, 28.6, 55.3, 63.1, 70.1, 70.5, 71.8,
72.1, 72.5, 73.5, 74.6, 74.8, 75.2, 79.6, 80.5, 82.2, 109.3, 114.0,
127.7–129.8 (several lines), 130.1, 133.6, 133.8, 135.6, 135.7, 138.0,
138.4, 138.6, 159.4 ppm. ESIHRMS calcd. for C68H78O12SiNa [M
+ Na] 1137.5161, found 1137.5160.
1,3,4,5-Tetra-O-benzyl-13-O-tert-butyldiphenylsilyl-2,6:8,12-dianhy-
dro-10,11-O-isopropylidene-D-lyxo-D-gulo-D-manno-tridecitol (12):
The PMB-protected C-glycoside 14 (155 mg, 0.13 mmol) was dis-
solved in ethyl acetate (10.0 mL) and treated with acetic anhydride
(0.12 mL, 1.32 mmol) and DMAP (19.0 mg, mmol) for 30 min.
CH3OH (0.5 mL) was then added to the reaction mixture, and the
solvent evaporated in vacuo. FCC of the residue afforded the ace-
tate derivative (156 mg, 98%) as a colorless oil; Rf = 0.34 (20%
EtOAc/petroleum ether). 1H NMR (500 MHz, CDCl3): δ = 1.10 (s,
9 H), 1.36 (s, 3 H), 1.63 (s, 3 H), 2.06 (s, 3 H), 3.38 (dd, J = 3.0,
9.0 Hz, 1 H), 3.60 (t, J = 8.8 Hz, 1 H), 3.68 (dd, J = 6.8, 11.0 Hz,
1 H), 3.75 (dt, J = 2.5, 7.5 Hz, 1 H), 3.83–4.00 (m, 7 H), 4.02–4.09
(m, 4 H), 4.18 (dd, J = 2.0, 3.3 Hz, 1 H), 4.32–4.85 (m, 8 H), 5.34
(dd, J = 7.5, 9.7 Hz, 1 H), 6.75 (d, J = 8.6 Hz, 2 H), 7.11 (d, J =
8.6 Hz, 2 H), 7.21–7.43 (m, 26 H), 7.75 (m, 4 H) ppm. 13C NMR
(75 MHz, CDCl3): δ = 21.5, 26.6, 27.1, 27.9, 55.2, 70.3, 70.4, 71.8,
72.8, 73.6, 73.8, 74.7, 75.0, 75.2, 75.4, 75.9, 76.5, 77.4, 78.3, 80.1,
109.9, 113.9, 127.5–130.5 (several lines), 133.7, 133.8, 135.7, 138.5,
138.6, 138.7, 159.4, 169.2 ppm.
A mixture of enol ether from the previous step (1.10 g, 0.91 mmol),
2,6-di-tert-butyl-4-methylpyridine (2.24 g, 10.9 mmol), and freshly
activated, powdered 4-Å molecular sieves (2.0 g) in anhydrous
CH2Cl2 (30 mL) was stirred for 15 min at room temp. under argon
and then cooled to 0 °C. Methyl triflate (1.02 mL, 9.12 mmol) was
introduced, and the mixture was warmed to room temp. and stirred
for an additional 16 h, at which time Et3N (1.30 mL) was added.
The mixture was diluted with diethyl ether, washed with saturated
aqueous NaHCO3 and brine, dried (Na2SO4), filtered and evapo-
rated under reduced pressure. The residue was purified by FCC on
basic alumina (Brockmann I, 150 mesh) to give the glycal product
(0.92 g, 87% based on recovered starting material) as a clear oil,
Rf = 0.27 (10 % EtOAc/petroleum ether). 1H NMR (500 MHz,
C6D6): δ = 1.30 (s, 9 H), 1.50 (s, 3 H), 1.69 (s, 3 H), 3.38 (s, 3 H),
3.91 (dd, J = 2.3, 13.0 Hz, 1 H), 3.98 (dd, J = 4.7, 13.0 Hz, 1 H),
4.08 (dd, J = 3.0, 9.0 Hz, 1 H), 4.19 (m, 3 H), 4.32 (m, 3 H), 4.40
DDQ (62 mg, 0.27 mmol) was added to a solution of the product
from the previous step (155 mg, 0.13 mmol) in CH2Cl2 (10 mL).
The reaction mixture was stirred at room temp. for 2 h, then diluted
with saturated aqueous NaHCO3, extracted with CH2Cl2, dried
(Na2SO4), and evaporated under reduced pressure. FCC of the resi-
due afforded the derived alcohol (107 mg, 77%) as a colorless oil;
Rf = 0.33 (20 % EtOAc/petroleum ether). 1H NMR (500 MHz,
CDCl3): δ = 1.05 (s, 9 H), 1.34 (s, 3 H), 1.46 (s, 3 H), 1.99 (s, 3 H),
3.06 (d, J = 3.5 Hz, D2O exchange, 1 H) 3.66–3.89 (m, 9 H), 3.94
(dd, J = 3.5, 9.0 Hz, 1 H), 4.00 (m, 1 H), 4.24 (br. s, 1 H), 4.30 (d,
J = 6.0 Hz, 1 H), 4.48–4.63 (m, 7 H), 4.77 (d, J = 11.5 Hz, 1 H),
5.30 (t, J = 5.5 Hz, 1 H), 7.21–7.41 (m, 26 H), 7.68 (m, 4 H) ppm.
13C NMR (75 MHz, CDCl3): δ = 21.3, 26.1, 27.1, 62.8, 68.1, 69.8,
(t, J = 5.3 Hz, 1 H), 4.46 (d, J = 9.0 Hz, 1 H), 4.50 (d, J = 7.2 Hz, 71.2, 71.9, 72.1, 73.3, 73.5, 74.1, 74.2, 75.1, 75.3, 76.2, 78.1, 78.3,
1 H), 4.58–4.66 (m, 5 H), 4.69–4.77 (m, 5 H), 5.06 (d, J = 11.4 Hz, 81.4, 82.2, 110.1, 127.6–129.8 (several lines), 133.6, 133.7, 135.7,
1 H), 5.18 (d, J = 2.0 Hz, 1 H), 6.86 (d, J = 8.6 Hz, 2 H), 7.11–
138.5, 138.7, 170.0 ppm. ESIHRMS calcd. for C62H72O12SiNa [M
7.50 (m, 28 H), 7.90 (m, 4 H) ppm. 13C NMR (125 MHz, C6D6): + Na] 1059.4695, found 1059.4691.
δ = .26.8, 26.9, 28.6, 54.5, 63.4, 69.5, 70.3, 71.0, 71.8, 71.9, 72.2,
73.5, 73.8, 74.5, 75.5, 76.1, 76.5, 77.4, 80.2, 101.2, 110.2, 114.0,
127.2–129.9 (several lines), 135.8 (two signals), 139.3, 153.2, 159.7
ppm. ESIHRMS calcd. for C68H76O11SiNa [M + Na] 1119.5050,
found 1119.5047.
NaOMe (50 mg, 1.14 mmol) was added to a solution of the product
from the previous step (106 mg, 0.11 mmol) in anhydrous CH3OH
(5 mL). The reaction was stirred at room temp. for 2 h and the pH
was then adjusted to 7 by addition of 2  methanolic HCl. The
mixture was concentrated under reduced pressure and the residue
purified by FCC to give 12 (100 mg, 98%) as a colorless oil; Rf =
BH3·Me2S (1.05 mL, 10.9 mmol) was added at 0 °C to a solution
of the glycal from the previous step (0.80 g, 0.73 mmol) in anhy-
drous THF (20 mL) under an atmosphere of argon. The mixture
was warmed to room temp. and stirred for an additional 3 h at this
temperature. At that time the solution was cooled to 0 °C and
treated with a mixture of 3  NaOH (4.4 mL) and 30% aqueous
H2O2 (0.88 mL) for 0.5 h. The mixture was diluted with diethyl
ether, washed with saturated aqueous NaHCO3 and brine, dried
(Na2SO4), filtered and evaporated under reduced pressure. The resi-
due was purified by FCC to give 14 (0.52 g, 75% based on reco-
vered starting material) as a clear oil; Rf = 0.40 (30% EtOAc/petro-
1
0.51 (25% EtOAc/petroleum ether). H NMR (500 MHz, CDCl3):
δ = 1.07 (s, 9 H), 1.34 (s, 3 H), 1.51 (s, 3 H), 2.42 (d, J = 9.5 Hz,
D2O exchange, 1 H), 3.34 (dd, J = 2.0, 9.0 Hz, 1 H), 3.44 (t, J =
9.0 Hz, 1 H), 3.63 (m, 2 H), 3.77 (dd, J = 2.5, 8.5 Hz, 1 H), 3.78–
4.00 (m, 6 H), 4.09 (m, 1 H), 4.15 (t, J = 2.5 Hz, 1 H), 4.21 (dd, J
= 2.5, 9.5 Hz, 1 H), 4.26 (d, J = 3.0 Hz, 1 H), 4.48–4.71 (m, 8 H),
7.13–7.45 (m, 26 H), 7.71 (m, 4 H) ppm. 13C NMR (125 MHz,
CDCl3): δ = 26.6, 27.0, 28.5, 63.1, 69.2, 69.9, 72.2, 72.6, 73.6, 73.7,
74.0, 74.1, 75.1, 80.0 (two signals), 109.5, 127.8–129.9 (several
lines), 133.5, 133.7, 135.7, 135.9, 137.3, 138.2, 138.4 ppm. FAB-
HRMS calcd. for C60H70O11SiNa [M + Na] 1017.4588, found
1017.4585.
1
leum ether). H NMR (500 MHz, C6D6): δ = 1.08 (s, 9 H), 1.40 (s,
3 H), 1.55 (s, 3 H), 3.56 (m, 3 H), 3.71–3.83 (m, 9 H), 3.90–4.00
(m, 4 H), 4.07 (apparent t, J = 5.8 Hz, 1 H), 4.22–4.36 (m, 4 H), 2,6:8,12-Dianhydro-1,3,4,5-tetra-O-benzyl-13-O-tert-butyldiphenyl-
4.40 (dd, J = 2.5, 8.0 Hz, 1 H), 4.47–4.68 (m, 7 H), 4.81 (d, J = silyl-10,11-O-isopropylidene-7,9-O-methylene- -lyxo- -galacto-
D
D
D-
Eur. J. Org. Chem. 2007, 645–654
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
651

D. R. Mootoo et al.
FULL PAPER
manno-tridecitol (15): A mixture of diol 10 (260 mg, 0.26 mmol),
nBu4Br (42 mg, 0.13 mmol), CH2Br2 (0.91 mL, 13.3 mmol) and
2,3-dimethylbutene (0.70 mL) was stirred at 65 °C for 15 min. 20%
aqueous NaOH (8 mL) was then added dropwise to the reaction
mixture and stirring continued at 65 °C for 3 h. The mixture was
then cooled to room temp. and extracted with diethyl ether. The
0.17 mmol) provided 18 (92 mg, 64% based on recovered 16) as a
colorless oil; Rf = 0.44 (40% EtOAc/petroleum ether). 1H NMR
(500 MHz, CDCl3): δ = 0.95 (s, 9 H), 2.44 (d, J = 6.0 Hz, 1 H,
D2O ex), 2.83 (br. s 1 H, D2O ex), 2.94 (t, J = 9.0 Hz, 1 H), 3.25
(t, J = 5.0 Hz, 1 H), 3.50 (m, 2 H), 3.59 (m, 2 H), 3.73 (m, 4 H),
3.86 (m, 2 H), 4.07 (m, 1 H), 4.11 (d, J = 3.0 Hz, 1 H), 4.35–4.61
organic layer was washed with water, dried (Na2SO4) and concen- (m, 10 H), 4.97 (d, J = 6.0 Hz, 1 H), 7.08–7.35 (m, 26 H), 7.56 (m,
trated under reduced pressure. FCC afforded 15 (204 mg, 86 %
4 H) ppm. 13C NMR (75 MHz, CDCl3): δ = 27.2, 63.6, 69.8, 70.0,
based on recovered 10) as a colorless oil; Rf = 0.44 (15% EtOAc/
petroleum ether). H NMR (500 MHz, CDCl3): δ = 1.05 (s, 9 H), 78.8, 80.4, 93.8, 127.5–130.1 (several lines), 132.9, 133.1, 135.7,
72.0, 72.2, 72.3, 72.6, 72.9, 73.6, 75.5, 76.1, 76.6, 77.4, 77.5, 78.0,
1
1.39 (s, 3 H), 1.51 (s, 3 H), 3.49 (dd, J = 6.9, 10.6 Hz, 1 H), 3.63
(dd, J = 5.7, 10.1 Hz, 1 H), 3.74 (dd, J = 5.7, 10.1 Hz, 1 H), 3.81
(m, 3 H), 3.90 (m, 2 H), 3.96 (m, 2 H), 4.06 (dd, J = 5.4, 7.6 Hz,
1 H), 4.24 (dd, J = 7.8, 10.5 Hz, 1 H), 4.30–4.58 (m, 11 H), 4.82
(A of ABq, J = 5.9 Hz, ∆δ = 0.59 ppm, 1 H), 5.42 (B of ABq, J =
5.9 Hz, ∆δ = 0.59 ppm, 1 H), 7.15–7.40 (m, 26 H), 7.68 (m, 4 H)
ppm. 13C NMR (125 MHz, CDCl3): δ = 26.7, 27.0, 28.6, 62.0, 62.8,
68.9, 70.5, 71.5, 72.1, 72.6, 72.7, 73.5, 73.8, 73.9, 74.7, 74.8, 75.2,
76.0, 76.8, 77.6, 91.2, 109.9, 127.6–130.0 (several lines), 133.6,
133.7, 135.7, 135.8 (two signals), 138.5, 138.6, 138.7, 138.9 ppm.
ESIHRMS calcd. for C61H71O11Si [M + H] 1007.4766, found
1007.4810.
135.8, 138.6, 138.8 ppm. FAB-HRMS calcd. for C58H66O11NaSi
[M + Na] 989.4272, found 989.4254.
1,3,4,5,10,11,13-Hepta-O-acetyl-2,6:8,12-dianhydro-7,9-O-methyl-
ene-D-lyxo-D-galacto-D-manno-tridecitol (19): Colorless oil; Rf =
0.30 (40% EtOAc/petroleum ether); (500 MHz, C6D6): δ = 1.57,
1.65, 1.67,1.68, 1.73, 1.96 (all s, 21 H), 3.33 (t, J = 6.0 Hz, 1 H),
3.47 (dd, J = 6.5, 10.0 Hz, 1 H), 4.00 (dd, J = 6.0, 11.6 Hz, 1 H),
4.07 (q, J = 5.7 Hz, 1 H), 4.19 (m, 2 H), 4.51 (t, J = 10.0 Hz, 1 H),
4.56 (m, 3 H), 4.71 (d, J = 6.0 Hz, 1 H), 5.06 (d, J = 6.0 Hz, 1 H),
5.19 (dd, J = 3.0, 10.0 Hz, 1 H), 5.47 (t, J = 6.0 Hz, 1 H), 5.55 (d,
J = 3.0 Hz, 1 H), 5.70 (dd, J = 3.5, 6.0 Hz, 1 H), 5.78 (dd, J = 3.0,
6.0 Hz, 1 H) ppm. 13C NMR (75 MHz, CDCl3): δ = 20.7, 20.9,
21.0 (two signals), 21.1 (two signals), 61.8, 62.3, 67.4, 68.0, 68.1,
68.2, 68.7, 70.7, 71.1, 71.9, 72.6, 73.1, 75.5, 90.5, 169.3, 169.7, 170.0
(two signals), 170.2, 170.5 (two signals) ppm. FAB-HRMS calcd.
for C28H38O18Na [M + Na] 685.1956, found 685.1952.
2,6:8,12-Dianhydro-1,3,4,5-tetra-O-benzyl-13-O-tert-butyldiphenyl-
silyl-10,11-O-isopropylidene-7,9-O-methylene-D-lyxo-D-gulo-D-
manno-tridecitol (16): Application of the procedure that was used
for 10 to diol 12 (260 mg, 0.26 mmol) provided 16 (175 mg, 83%
based on recovered 12); colorless oil; Rf = 0.50 (20% EtOAc/petro-
1,3,4,5,10,11,13-Hepta-O-acetyl-2,6:8,12-dianhydro-7,9-O-methyl-
ene-D-lyxo-D-gulo-D-manno-tridecitol (20): Colorless oil; Rf = 0.29
1
leum ether). H NMR (500 MHz, CDCl3): δ = 1.05 (s, 9 H), 1.34
(s, 3 H), 1.55 (s, 3 H), 3.03 (t, J = 9.0 Hz, 1 H), 3.36 (t, J = 9.0 Hz,
1 H), 3.66 (m, 2 H), 3.74 (dd, J = 2.0, 9.5 Hz, 1 H), 3.80 (m, 3 H),
3.86–4.00 (m, 4 H), 4.04 (q, J = 5.0 Hz, 1 H), 4.18 (d, J = 3.5 Hz,
1 H), 4.35 (d, J = 4.0 Hz, 1 H), 4.44–4.58 (m, 7 H), 4.63 (d, J =
10.5 Hz, 1 H), 4.65 (A of ABq, J = 6.5 Hz, ∆δ = 0.39 ppm, 1 H),
5.04 (B of ABq, J = 6.5 Hz, ∆δ = 0.39 ppm, 1 H), 7.17–7.43 (m,
26 H), 7.69 (m, 4 H). 13C NMR (75 MHz, CDCl3): δ = 26.6, 27.2,
28.8, 62.7, 69.7, 69.3, 70.6, 71.9, 72.2, 72.8, 73.4, 73.8, 75.4, 76.0,
77.0, 80.1, 80.4, 93.7, 109.8, 127.5–129.9 (several lines), 133.7,
133.8, 135.7, 135.8, 138.7, 138.8 ppm. FAB-HRMS calcd. for
C61H70O11SiNa [M + Na] 1029.4584, found 1029.4585.
(50% EtOAc/petroleum ether). 1H NMR (500 MHz, C6D6): δ =
1.63, 1.68, 1.73,1.74, 1.75, 1.78 (all s, 21 H), 3.23 (t, J = 9.8 Hz, 1
H), 3.30 (t, J = 6.5 Hz, 1 H), 3.69 (t, J = 9.8 Hz, 1 H), 3.77 (dd, J
= 2.2, 9.8 Hz, 1 H), 3.99 (dd, J = 6.5, 10.8 Hz, 1 H), 4.10 (m, 2 H),
4.28 (m, 2 H), 4.46 (s, 1 H), 4.56 (dd, J = 5.0, 11.0 Hz, 1 H), 4.62
(d, J = 6.5 Hz, 1 H), 5.15 (dd, J = 3.0, 9.8 Hz, 1 H), 5.47 (d, J =
3.0 Hz, 1 H), 5.72 (t, J = 9.0 Hz, 1 H), 5.84 (br. s, 1 H), 5.96 (dd,
J = 4.0, 9.0 Hz, 1 H) ppm. 13C NMR (75 MHz, CDCl3): δ = 20.8
(two signals), 20.9 (two signals), 21.0, (two signals), 21.2, 61.7, 62.9,
67.3, 68.4 (two signals), 69.9, 71.1, 72.8, 73.7, 74.6, 75.5, 75.7, 80.4,
93.5, 169.7, 168.8 (two signals), 169.9, 170.0, 170.3, 170.7 ppm.
2,6:8,12-Dianhydro-1,3,4,5-tetra-O-benzyl-13-O-tert-butyldiphenyl-
silyl-7,9-O-methylene-D-lyxo-D-galacto-D-manno-tridecitol (17): A
Lactone 21: A mixture of diol 17 (54 mg, 0.06 mmol), Bu2SnO
(27 mg, 0.11 mmol), and anhydrous toluene (5 mL) was heated at
reflux using a Dean–Stark set-up for 1 h. The solution was then
evaporated in vacuo and the residue was dissolved in dry toluene
(3 mL). nBu4NI (22 mg, 0.06 mmol) and methyl 2-bromoacetate
(0.10 mL, 1.08 mmol) were added and the solution heated at reflux
for 1 h, at which time the volatiles were removed under reduced
pressure. FCC of the residue afforded lactone 21 (55 mg, 98%) as
saturated solution of HCl in diethyl ether (0.1 mL) was added to a
solution of 15 (200 mg, 1.29 mmol) in dry CH3OH (30 mL). The
reaction mixture was stirred at room temp. for 2.5 h then neutral-
ized with a solution of NaOMe in methanol. Removal of the vola-
tiles under reduced pressure and FCC of the residue provided 17
(130 mg, 68%) as a colorless oil; Rf = 0.53 (40% EtOAc/petroleum
1
1
ether). H NMR (500 MHz, CDCl3): δ = 0.95 (s, 9 H), 2.49 (d, J
a colorless oil; Rf = 0.35 (20% EtOAc/petroleum ether). H NMR
= 7.5 Hz, 1 H, D2O ex), 2.92 (d, J = 11 Hz, 1 H), 3.02 (d, J =
10.0 Hz, 1 H, D2O ex), 3.21 (s, 1 H), 3.52 (m, 2 H), 3.59 (m, 1 H),
3.78 (m, 3 H), 3.83 (t, J = 10.0 Hz, 1 H), 3.88–4.00 (m, 3 H), 4.05
(dd, J = 3.0, 9.8 Hz, 1 H), 4.20–4.52 (m, 10 H), 4.83 (A of ABq, J
= 5.5 Hz, ∆δ = 0.45 ppm, 1 H), 5.38 (B of ABq, J = 5.5 Hz, ∆δ =
0.45 ppm, 1 H), 7.00–7.38 (m, 26 H), 7.58 (m, 4 H) ppm. 13C NMR
(75 MHz, CDCl3): δ = 27.1, 63.0, 66.4, 69.4, 69.7, 70.4, 71.9, 72.4,
72.8, 73.2, 73.4, 73.5, 74.7, 74.8, 75.0, 75.4, 80.0, 92.1, 127.6–129.9
(several lines), 133.4, 133.5, 135.7, 137.9, 138.1, 138.3, 138.7 ppm.
FAB-HRMS calcd. for C58H66O11NaSi [M + Na] 989.4272, found
989.4251.
(500 MHz, CDCl3): δ = 1.41 (s, 9 H), 3.42 (t, J = 7.0 Hz, 1 H),
3.38 (dd, J = 6.5, 9.0 Hz, 1 H), 3.58 (dd, J = 3.5, 10.0 Hz, 1 H),
3.84–4.48 (m, 16 H), 4.51 (d, J = 12.0 Hz, 1 H), 4.56 (d, J = 3.0 Hz,
1 H), 4.61 (dd, J = 1.5, 6.5 Hz, 1 H), 4.66 (dd, J = 1.5, 9.5 Hz, 1
H), 4.76 (d, J = 12 Hz, 1 H), 4.93 (d, J = 6.0 Hz, 1 H), 4.98 (t, J
= 10.0 Hz, 1 H), 5.74 (d, J = 6.0 Hz, 1 H), 7.04–7.36 (m, 26 H),
7.82 (m, 4 H) ppm. 13C NMR (75 MHz, CDCl3): δ = 27.1, 60.9,
61.2, 68.4, 69.0, 70.5, 71.9, 72.1, 72.7, 73.2, 73.6, 74.2, 74.3, 74.7,
78.2, 91.7, 127.2–130.0 (several lines), 133.0, 133.2, 135.6, 138.3,
138.4, 138.5, 138.7, 166.5 ppm. FAB-HRMS calcd. for
C60H66O12NaSi [M + Na] 1029.4221, found 1029.4197.
2,6:8,12-Dianhydro-1,3,4,5-tetra-O-benzyl-13-O-tert-butyldiphenyl-
Lactone 22: Application of the procedure that was used for 21 to
diol 18 (90 mg, 0.09 mmol) provided 22 (75 mg, 93% based on re-
covered 18) as a colorless oil; Rf = 0.33 (20% EtOAc/petroleum
silyl-7,9-O-methylene-D-lyxo-D-gulo-D-manno-tridecitol (18): Appli-
cation of the procedure that was used for 17 to 16 (175 mg,
652
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 645–654

C-Disaccharides as Probes for Carbohydrate Recognition
FULL PAPER
ether), 1H NMR (500 MHz, CDCl3): δ = 1.06 (s, 9 H), 3.55 (t, J = 139.8, 178.4 ppm. FAB-HRMS calcd. for C44H51O13 [M + H]
9.5 Hz, 1 H), 3.61–3.82 (m, 6 H), 3.90–3.97 (m, 4 H), 4.14 (d, J = 787.3330, found 787.3327.
7.0 Hz, 1 H), 4.22 (s, 1 H), 4.35–4.51 (m, 11 H), 4.68 (d, J = 6.5 Hz,
The product from the previous step (37 mg, 0.05 mmol) was sub-
jected to the hydrogenation procedure that was described for 4.
1 H), 4.88 (d, J = 3.0 Hz, 1 H), 5.10 (d, J = 6.5 Hz, 1 H), 7.15–
7.44 (m, 26 H), 7.67 (m, 4 H) ppm. 13C NMR (75 MHz, CDCl3):
Compound 5 (17 mg, 84%) was obtained as an amorphous solid.
δ = 27.2, 60.6, 61.1, 68.4, 69.5, 71.0, 71.9, 72.3, 72.8, 73.5, 73.6,
1
[α]D = +41.6 (c = 0.30, H2O). H NMR (500 MHz, D2O): δ = 3.32
74.1, 74.2, 75.2, 75.8, 76.1, 77.5, 79.4, 93.8, 127.6–130.1 (several
(t, J = 9.4 Hz, 1 H), 3.60 (t, J = 8.6 Hz, 1 H), 3.65 (dd, J = 2.5,
lines), 132.9, 133.2, 135.6, 135.7, 138.3, 138.5, 138.6, 138.7, 166.8
12.4 Hz, 1 H), 3.68–3.71 (m, 8 H), 3.97 (dd, J = 3.6, 8.6 Hz, 1 H),
4.02 (dd, J = 2.7, 9.6 Hz, 1 H), 4.10–4.15 (m, 5 H), 4.17 (m, 3 H),
ppm. C60H66O12SiNa [M + Na] 1029.4222, found 1029.4221.
4.79 (A of ABq, J = 6.5 Hz, ∆δ = 0.29 ppm, 1 H), 5.08 (B of ABq,
J = 6.5 Hz, ∆δ = 0.29 ppm, 1 H) ppm. 13C NMR (125 MHz, D2O):
δ = 61.2, 61.5, 66.4, 67.5 (two signals), 68.6, 71.3, 72.7, 76.1, 77.1,
77.4, 79.2, 79.3, 79.7, 93.2, 178.3 ppm. FAB-HRMS calcd. for
C16H26O13Na 449.1286, found 449.1295.
Sodium (2,6:8,12-Dianhydro-7,9-O-methylene-D-lyxo-D-galacto-D-
manno-tridecit-10-yloxy)ethanoate (4): Lactone 21 (99 mg,
0.10 mmol) was dissolved in ethanol (5 mL) and treated with 3 
NaOH (2 mL). After 2 h the solvent was removed under reduced
pressure, and the residue purified by FCC to give the dihydroxy
sodium salt resulting from saponification of the lactone and cleav-
age of the silyl ether (56 mg, 73%), as a colorless oil; Rf = 0.46
(30% CH3OH/acetone). 1H NMR (500 MHz, CD3OD): δ = 3.54
(dd, J = 3.0, 9.5 Hz, 1 H), 3.59 (t, J = 5.3 Hz, 1 H), 3.66 (dd, J =
7.0, 9.5 Hz, 1 H), 3.73 (m, 3 H), 3.80 (m, 2 H), 3.86 (m, 1 H), 3.91
(q, J = 5.3 Hz, 1 H), 4.11 (m, 2 H), 4.27 (A of ABq, ∆δ = 0.11 ppm,
J = 16.5 Hz, 1 H), 4.36 (apparent t, J = 5.5 Hz, 1 H), 4.38 (B of
Supporting Information (see also the footnote on the first page of
this article): Experimental procedures and physical data for 7, 8, 9,
I
11, 13 and their synthetic precursors. H and 13C NMR charts for
selected new compounds, procedures and data for conformational
analysis on 25 and 26. P-selectin binding results for 2–5.
ABq, ∆δ = 0.11 ppm, J = 16.5 Hz, 1 H), 4.51–4.60 (m, 13 H), 4.77 Acknowledgments
(d, J = 6.0 Hz, 1 H), 5.29 (d, J = 6.0 Hz, 1 H), 7.20–7.37 (m, 20
H) ppm. 13C NMR (100 MHz, CD3OD): δ = 61.7, 68.1, 68.4, 68.7,
This investigation was supported by National Institutes of Health
70.1, 71.7, 72.7, 72.9, 73.1, 73.6, 74.0, 74.2, 74.5, 76.4, 79.3, 80.3,
90.4, 127.4, 128.1, 138.6, 174.1 ppm. ESIHRMS calcd. for
C44H51O13 787.3330, found 787.3358.
(NIH) (grant R01-GM57865). Support from the National Center
for Research Resources of the NIH (“Research Centers in Minority
Institutions” award RR-03037), which supports the infrastructure
and instrumentation of the Chemistry Department at Hunter Col-
lege, and from Minority Biomedical Research Support, Research
Initiative for Scientific Enhancement (MBRS-RISE) (award
GM60665), are also acknowledged.
A mixture of the product from the previous step (52 mg,
0.06 mmol), 10% Pd on carbon (100 mg), formic acid (0.1 mL) and
CH3OH (3.0 mL) was stirred under hydrogen (balloon), for 12 h.
The reaction mixture was purged with argon, filtered through a
bed of celite and the filtrate concentrated under reduced pressure.
The residue was dissolved in ethanol (5 mL) and treated with 3 
NaOH (1 mL). After stirring at room temp. for 2 h the mixture was
evaporated in vacuo and residue purified by sequential FCC on
C18 silica gel (CH3OH/H2O) and Sephadex LH-20 (H2O). Lyophi-
lization of the eluate provided 4 as an amorphous solid (27 mg,
[1] J. C. Paulson, in: Adhesion: Its role in inflammatory diseases
(Eds.: J. M. Harlan, D. Y. Liu, W. H. Freeman), New York,
1992, p. 19–42 and ref. cited.
[2] T. A. Springer, Cell 1994, 76, 301–314.
[3] S. A. Mousa, Drugs Future 1996, 21, 283–289.
[4] S. R. Chhabra, A. S. A. Rahim, B. Kellam, Mini-Rev. Med.
Chem. 2003, 3, 679–687.
1
90%). [α]D = +13.5 (c = 0.40, H2O). H NMR (500 MHz, D2O): δ
= 3.61 (dt, J = 2.6, 7.4 Hz, 1 H), 3.65–3.74 (m, 5 H), 3.75 (d, J =
2.6 Hz, 1 H), 3.81 (m, 2 H), 3.87 (dd, J = 3.5, 8.3 Hz, 1 H), 4.13
(ABq, J = 16.3 Hz, ∆δ = 0.09 ppm, 2 H), 4.16 (d, J = 3.1 Hz, 1
H), 4.20 (t, J = 3.1 Hz, 1 H). 4.26 (t, J = 9.9 Hz, 1 H), 4.31 (dd, J
= 3.4, 9.3 Hz, 1 H), 4.50 (dd, J = 5.9, 9.3 Hz, 1 H), 4.91 (A of
ABq, J = 6.8 Hz, ∆δ = 0.18 ppm, 1 H), 5.09 (B of ABq, J = 6.8 Hz,
∆δ = 0.18 ppm, 1 H) ppm. 13C NMR (125 MHz, D2O): δ = 60.7,
61.7, 66.5, 67.6, 68.0, 68.5, 69.0, 70.6, 72.3, 72.4, 72.5, 76.4, 79.5,
80.4, 89.6, 178.3 ppm. FAB-HRMS calcd. for C16H26O13Na
449.1288, found 449.1271.
[5] Reviews: R. Roy, in: Carbohydrates in Drug Design (Eds.: Z. J.
Witczak, K. A. Nieforth), Marcel Dekker, New York, 1997, p.
83–135.
[6] E. E. Simanek, G. J. McGarvey, J. A. Jablonowski, C.-H.
Wong, Chem. Rev. 1998, 98, 833–862.
[7] N. Kaila, B. E. Thomas IV, Med. Res. Rev. 2002, 22, 566–601.
[8] Selected recent studies: H. C. Kolb, B. Ernst, Chem. Eur. J.
1997, 3, 1571–1578.
[9] G. Magnani, J. L. Thoma, J. T. Patton, B. Ernst, W. Jahnke,
Angew. Chem. Int. Ed. 2001, 40, 1941–1945.
[10] G. Thoma, R. Banteli, W. Jahnke, J. L. Magnani, J. T. Patton,
Angew. Chem. Int. Ed. 2001, 40, 3644–3647.
Sodium (2,6:8,12-Dianhydro-7,9-O-methylene-D-lyxo-D-gulo-D-
[11] S. Hanessian, V. Mascitti, O. Rogel, J. Org. Chem. 2002, 67,
manno-tridecit-10-yloxy)ethanoate (5): Lactone 22 (75 mg,
0.07 mmol) was subjected to the similar saponification procedure
that was described for lactone 21. The corresponding dihydroxy
sodium salt (38 mg, 67%) was obtained as a colorless oil; Rf = 0.28
(30% CH3OH/acetone). 1H NMR (500 MHz, CD3OD): δ = 3.16
(t, J = 9.0 Hz, 1 H), 3.45 (dd, J = 3.0, 9.5 Hz, 1 H), 3.53 (t, J =
6.3 Hz, 1 H), 3.66 (d, J = 5.0 Hz, 1 H), 3.71 (dd, J = 5.0, 11.5 Hz,
1 H), 3.77 (m, 2 H), 3.91 (d, J = 9.5 Hz, 1 H), 3.96 (q, J = 5.0 Hz,
1 H), 4.00 (m, 1 H), 4.06 (m, 3 H), 4.28 (m, 2 H), 4.48–4.69 (m, 10
H), 4.73 (d, J = 6.5 Hz, 1 H), 5.04 (d, J = 6.5 Hz, 1 H), 7.25–7.43
(m, 20 H) ppm. 13C NMR (75 MHz, CD3OD): δ = 63.0, 68.7, 70.8,
71.0, 73.1, 73.6, 73.8, 74.0, 74.5, 74.9, 76.6, 77.2, 78.7, 79.1, 80.7,
81.2, 81.9, 94.5, 128.5, 128.7, 128.8, 129.0, 129.1, 129.4, 139.7,
3346–3354.
[12] S. M. Chervin, J. B. Lowe, M. Koreeda, J. Org. Chem. 2002,
67, 5654–5662.
[13] G. Thoma, F. Schwarzenbach, Helv. Chim. Acta 2003, 86, 855–
864.
[14] N. Kaila, W. S. Somers, B. E. Thomas, P. Thakker, K. Janz, S.
DeBernardo, S. Tam, W. J. Moore, R. Yang, W. Wrona, P. W.
Bedard, D. Keith Jr., J. C. Crommie, D. H. H. Tsao, J. C. Alva-
rez, H. Ni, E. Marchese, J. T. Patton, J. L. Magnani, R. T.
Camphausen, J. Med. Chem. 2005, 48, 4346–4357.
[15] K. Hiruma, T. Kajimoto, G. Weitz-Schmidt, I. Ollmann, C.-H.
Wong, J. Am. Chem. Soc. 1996, 118, 9265–9270.
[16] K. Shibata, K. Hiruma, O. Kanie, C.-H. Wong, J. Org. Chem.
2000, 65, 2393–2398.
Eur. J. Org. Chem. 2007, 645–654
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
653

D. R. Mootoo et al.
FULL PAPER
[17] For early models for sLex-selectin recognition: R. M. Cooke,
R. S. Hale, S. G. Lister, G. Shah, M. P. Weir, Biochemistry 1994,
33, 10591–10596.
[18] P. Hensley, P. J. McDevitt, I. Brooks, J. J. Trill, J. A. Field, D. E.
McNulty, J. R. Connor, D. E. Griswold, N. V. Kumar, K. D.
Kopple, S. A. Carr, B. J. Dalton, K. Johanson, J. Biol. Chem.
1994, 269, 23949–23958.
[19] K. Scheffler, B. Ernst, A. Katopodis, J. L. Magnani, W. T.
Wang, R. Weisemann, T. Peters, Angew. Chem. Int. Ed. Engl.
1995, 34, 1841–1844.
[20] L. Poppe, G. S. Brown, J. S. Philo, P. V. Nikrad, B. H. Shah, J.
Am. Chem. Soc. 1997, 119, 1727–1736.
[21] For recent X-ray structures which are in general agreement
with ref.[17–19]: W. S. Somers, J. Tang, G. D. Shaw, T. Cam-
phausen, Cell 2000, 103, 467–479.
[22] For reviews on synthesis and biological applications of C-gly-
cosides: M. H. D. Postema, C-Glycoside Synthesis, CRC Press,
Boca Raton, Fl, 1995.
[23] D. Levy, C. Tang, The Synthesis of C-Glycosides, Pergamon,
Oxford, 1995.
[24] J.-M. Beau, T. Gallagher, Top. Curr. Chem. 1997, 187, 1–54.
[25] H. Togo, W. He, Y. Waki, M. Yokoyama, Synlett 1998, 700–
717.
[33]
[34]
[35]
[36]
X. Cheng, N. Khan, D. R. Mootoo, J. Org. Chem. 2000, 65,
2544–2547.
R. W. Denton, D. R. Mootoo, ACS Symp. Ser. 2005, 896, 121–
135.
J. L. Asensio, F. J. Cañada, X. Cheng, N. Khan, D. R. Mootoo,
J. Jiménez-Barbero, Chem. Eur. J. 2000, 6, 1035–1041.
V. G a r c ía-Aparicio, M. Fernández-Alonso, J. Angulo, J. L.
Asensio, F. J. Cañada, J. Jiménez-Barbero, D. R. Mootoo, X.
Cheng, Tetrahedron Asymmetrey 2005, 16, 519–527.
N. A. Powell, W. R. Roush, Org. Lett. 2001, 3, 453–456.
K. S. Kim, W. A. Szarek, Synthesis 1978, 48–50.
P. L. Barili, G. Catelani, F. D’Andrea, E. Mastrorilli, J. Car-
bohydr. Chem. 1997, 16, 1001–1010.
[37]
[38]
[39]
[40]
[41]
The torsion angles Φgal and Φman are defined as H1Gal–C1Gal
–
X–C1Man and H1Man–C1Man–X–C1Gal, where X represents the
intersaccharide “O” or “C”.
F. Mohamadi, N. G. J. Richards, W. C. Guida, R. Liskamp, M.
Lipton, C. Caufield, G. Chang, T. Hendrickson, W. C. Still, J.
Comput. Chem. 1990, 11, 440–467.
[42] N. L. Allinger, Y. H. Yuh, J. H. Lii, J. Am. Chem. Soc. 1989,
111, 8551–8559.
[26] R. J. Linhardt, X. Yuan, Curr. Top. Med. Chem. 2005, 5, 1393–
[43] W. C. Still, A. Tempzyk, R. Hawley, T. Hendrickson, J. Am.
Chem. Soc. 1990, 112, 6127–6129.
1430.
[27] M. H. D. Postema, J. L. Piper, R. L. Betts, Synlett 2005, 1345–
1358.
[44] C. A. G. Haasnoot, F. A. A. M. de Leeuw, C. Altona, Tetrahe-
dron 1980, 36, 2783–2794.
[28] J. Jimenez-Barbero, J. F. Espinosa, J. L. Asensio, F. J. Canada,
A. Poveda, Adv. Carbohydr. Chem. Biochem. 2001, 56, 235–284.
[29] For examples of conformationally restrained di- and tri-saccha-
rides as recognition probes: D. R. Bundle, R. Alibés, S. Nilar,
A. Otter, M. Warwas, P. Zhang, J. Am. Chem. Soc. 1998, 120,
5317–5318.
[30] N. Navarre, N. Amiot, A. Van Oijen, A. Imberty, A. Poveda, J.
Jiménez-Barbero, A. Cooper, M. A. Nutley, G.-J. Boons, Chem.
Eur. J. 1999, 5, 2281–2294.
[45] See ref.[14] and supporting information for a detailed protocol
of the binding experiments.
[46] At 6 mM 2, 3, 4 and 5 exhibited 33, 12, 10 and 13% inhibition,
respectively.
[47] For a discussion on the pitfalls in the evaluation of selectin
binding, see ref.[14] and G. Kretzschmar, A. Toepfer, C. Huls,
M. Krause, Tetrahedron 1997, 53, 2485–2494.
[48] A. G. Toepfer, G. Kretzschmar, E. Bartnik, Tetrahedron Lett.
1995, 36, 9161–9164.
[31] J. G. Geng, M. P. Bevilacqua, K. L. Moore, T. M. McIntyre,
S. M. Prescott, J. M. Kim, G. A. Bliss, G. A. Zimmerman, R. P.
McEver, Nature 1990, 343, 757–760.
[32] P. Mehta, R. D. Cummings, R. P. McEver, J. Biol. Chem. 1998,
273, 32506–32513.
[49] A. G. Toepfer, G. Kretzschmar, Bioorg. Med. Chem. Lett. 1997,
7, 1311–1316.
[50] See ref.[34] and supporting information.
Received: September 20, 2006
Published Online: November 23, 2006
654
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 645–654