Conformational constraints: Nature does it best with sialyl Lewis x

Article abstract of DOI:10.1002/ejoc.201200744

Selectins play a key role in the inflammatory cascade. The interaction with their physiological ligands containing the tetrasaccharide sialyl Lewis ^x (sLe^x) leads to the recruitment of leukocytes from the vascular system to the site of injury. To facilitate the interaction under the shear stress conditions present in the blood vessel, the conformation of sLe^x is stabilized via lipophilic inter-residual contacts. sLe ^x and two analogs were synthesized and evaluated for selectin binding, average conformation, and conformational dynamics. We could show that the methyl group in L-fucose is optimally suited to stabilize the sLe ^x core through an interaction with the β-face of D-galactose and thus enables binding to the selectins under shear stress conditions.

Full text of DOI:10.1002/ejoc.201200744

Job/Unit: O20744
/KAP1
Date: 20-08-12 11:25:50
Pages: 7
FULL PAPER
DOI: 10.1002/ejoc.201200744
Conformational Constraints: Nature Does It Best with Sialyl Lewisx
[b]
[a]
[a]
Alexander Titz,[a][‡] Alberto Marra, Brian Cutting, Martin Smiesko,
ˇ
George Papandreou,[a][‡‡] Alessandro Dondoni,[b] and Beat Ernst*[a]
Keywords: Glycoproteins / Conformational analysis / Carbohydrates / Stacking interactions
Selectins play a key role in the inflammatory cascade. The
interaction with their physiological ligands containing the
tetrasaccharide sialyl Lewisx (sLex) leads to the recruitment
of leukocytes from the vascular system to the site of injury.
To facilitate the interaction under the shear stress conditions
present in the blood vessel, the conformation of sLex is stabi-
lized via lipophilic inter-residual contacts. sLex and two ana-
logs were synthesized and evaluated for selectin binding,
average conformation, and conformational dynamics. We
could show that the methyl group in
suited to stabilize the sLex core through an interaction with
the β-face of -galactose and thus enables binding to the se-
L-fucose is optimally
D
lectins under shear stress conditions.
tant role. Under the influence of shear stress present in
blood vessels, preferably the fraction of the oligosaccharide
ligand that is preorganized in the bioactive conformation
binds to the receptor. Because information regarding the
bound conformation of sLex is available from trNOE NMR
experiments[7] and X-ray crystallography,[8] a possible strat-
egy to high-affinity mimetics is their preorganization in the
bioactive conformation. In an initial attempt, the GlcNAc
moiety was replaced with cyclic 1,2-trans-diols.[9,10] A sig-
nificant reduction in the affinity associated with the replace-
ment of the GlcNAc moiety by conformationally flexible
diols impressively demonstrated the effect of preorganiza-
tion. However, an improved affinity was achieved when the
Lewisx (Lex) core was stabilized by steric constraints, that
is, with an equatorial methyl substituent adjacent to the
linking position of the l-fucose residue.[10] Because this ad-
ditional methyl substituent is not in direct contact with the
protein surface (deduced from the sLex/E-selectin X-ray
structure[8] and STD-NMR investigations[10b]), the in-
creased affinity was assigned to a higher degree of preor-
ganization of the core in its bioactive conformation.
Introduction
The recruitment of leukocytes to inflamed tissue medi-
ated by the cell adhesion molecules E-, P-, and L-selectin is
essential for immune defense. The interaction of selectins
leads to the characteristic tethering and rolling of leuko-
cytes on the vascular endothelium, followed by firm ad-
hesion and migration to the site of injury.[1] However, ex-
cessive extravasation of leukocytes can cause acute or
chronic reactions, as observed in reperfusion injury, stroke,
or rheumatoid arthritis.[2] Therefore, the antagonism of the
selectins is generally considered to be a potential thera-
peutic approach for the treatment of inflammatory dis-
eases.[3]
The known physiological ligands of E-, P-, and L-selectin
contain the carbohydrate epitope sialyl Lewisx [sLex (1a),
Figure 1].[4] Although the affinity of sLex to E-selectin is in
the micro- to millimolar range – Kd values between 107 and
1800 μm are reported[5] – the sLex motif serves as a lead
structure in the search for high-affinity selectin antago-
nists.[6]
For dynamic non-equilibrium processes, such as E-selec-
tin-mediated rolling of leukocytes on activated endothelial
cells, conformational entropy is expected to play an impor-
Results and Discussion
In this communication, we investigated a second con-
straint contributing to the stabilization of the Lex core. Nu-
merous reports reviewed the replacement of the l-Fuc moi-
ety in sLex antagonists.[6,11–13] Thus, when in glycopeptide
2 l-fucose (rIC50 = 1) was replaced by d-arabinose (2b,
rIC50 = 2.8), a substantial loss of affinity resulted, whereas
with l-galactose (2c, rIC50 = 1.25) affinity remained nearly
unchanged (Figure 1).[13b]
[a] Institute of Molecular Pharmacy, University of Basel
Klingelbergstrasse 50-70, 4056 Basel, Switzerland
Fax: +41-61-267 1552
E-mail: beat.ernst@unibas.ch
Homepage: www.pharma.unibas.ch
[b] Laboratorio di Chimica Organica, Dipartimento di Chimica
Università di Ferrara
Via L. Borsari 46, 44100 Ferrara, Italy
[‡] Current address: Zukunftskolleg and Department of Chemistry,
University of Konstanz,
Universitätsstrasse 10, 78457 Konstanz, Germany
[‡‡]Current address: 10 Maurice Ct., Kendall Park, NJ 08824, USA
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201200744.
According to Lemieux, monosaccharides can establish –
besides polar interactions – lipophilic contacts via their ring
methine protons. Some monosaccharides even display ex-
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1

Job/Unit: O20744
/KAP1
Date: 20-08-12 11:25:50
Pages: 7
B. Ernst et al.
FULL PAPER
Figure 1. The carbohydrate epitope recognized by the selectin family. In sLex (1a) and sLex derivatives 1b and 1c (containing a modified
fucose residue) the influence of an intramolecular lipophilic stabilization was studied. Glycopeptide derivatives 2a–c with modified fucose
were reported by Kunz et al. to improve their biological half-life.[13b] Relative IC50 values (rIC50) were calculated by dividing the IC50 of
the substance of interest by the IC50 of the reference compound, here 2a.
tended lipophilic patches, also called their lipophilic face.[14] matic groups in glycoconjugates was reported by Terraneo
Because of the unusually high conformational stability of et al.[23]
the Lex core,[15] we hypothesized that in sLex (1a) the methyl
For our study on the stabilization of the core of sLex
group of the l-fucose moiety establishes a lipophilic contact by an intramolecular hydrophobic contact, we synthesized
with the β-face of d-galactose and thereby contributes to derivatives 1a–c (Figure 1). When the l-fucose moiety in 1a
the stability of the core. Because the 5-methyl group of l- is replaced by d-arabinose (Ǟ1b), a loss of intramolecular
fucose is not in direct contact with the protein,[8,16] no con- stabilization should result. In contrast, a phenyl group
tribution to the enthalpy is expected. In contrast, in arabin- (Ǟ1c) might restore this effect. For the synthesis of 1a and
ose derivative 1b where this stabilizing effect is no longer 1b, trisaccharide 18[24] was glycosylated with fucosyl-
present, affinity should be markedly reduced.
transferase FucT-III by using the glycosyl donors GDP-Fuc
Hydrophobic contacts of carbohydrates with their bind- (Ǟ1a) or GDP-Ara (Ǟ1b), respectively (Scheme 1, for ex-
ing partners have been broadly investigated.[17] Thus, in ga- perimental details see Supporting Information).[25]
lectins[18] or carbohydrate-specific antibodies[19] aromatic
For the synthesis of derivative 1c, phenyllithium was
side chains of amino acids mediate hydrophobic contacts added diastereoselectively to d-mannofuranose diacetonide
with a carbohydrate ligand. In another example, the align- 3, following strategy reported by Mekki et al.
a
ment of the β-face of galactosides with toluene was re- (Scheme 2).[26] The free hydroxy groups in 4 were trans-
ported,[20] which occurs in a manner similar to that of the formed into allyl ethers (i.e., 5). Cleavage of the acetonides
hydroxyphenyl side chain of tyrosine in an Artocarpus hir- under acidic conditions (Ǟ6) followed by iodate-mediated
sute lectin.[21] Furthermore, by NMR and molecular model- oxidative cleavage of the terminal diol yielded bis-allylated
ing, Jiménez-Barbero and Bartik could show the modes of 5-C-phenylarabinose derivative 7 in high yield. After cleav-
stacking interaction of a variety of monosaccharides with age of the allyl ethers, 8 was acetylated (Ǟ9) and treated
either phenol or aromatic amino acid side chains.[22] Finally, with ethyl mercaptan under Lewis acid catalysis. 5-C-Phen-
intramolecular conformational stabilization as a result of ylarabinose donor 11 was finally obtained by replacing the
such stacking interactions between carbohydrates and aro- acetate protection in 10 by benzyl ethers.
Scheme 1. Reagents and conditions: (a) FucT-III, GDP-Fuc (89%), (b) FucT-III, GDP-Ara (81%).
2
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O20744
/KAP1
Date: 20-08-12 11:25:50
Pages: 7
Conformational Constraints: Nature Does It Best with Sialyl Lewisx
Scheme 2. Reagents and conditions: (a) PhLi, Et2O, –78 to –20 °C, 20 h, 76%, 4:1dr; (b) AllBr, NaH, DMF, 0 °C to r.t., 1 h, 93%;
(c) HOAc/H2O (4:1), r.t., 10 h, quant.; (d) NaIO4, CH2Cl2, aq. NaHCO3, 0 °C to r.t., 1 d, 77%; (e) Pd/C, CSA, dioxane/H2O, 95 °C, 2 d;
(f) Ac2O, DMAP, pyridine, r.t., 45 min, 69% (over 2 steps, α/β = 1:2); (g) EtSH, CH2Cl2, TMSOTf, 0 °C to r.t., 3 h, 65% (α/β = 1:2);
(h) 1. NaOMe, MeOH, r.t., 3 h; 2. NaH, BnBr, DMF, 0 °C to r.t., 1 h, 88% (over 2 steps).
Scheme 3. Reagents and conditions: (a) BH3·NMe3, AlCl3, H2O, THF, r.t., 2 h, 92%; (b) DMTST, 4 Å MS, CH2Cl2, 0 °C, 2.5 d, 68%;
(c) H2NC(S)NH2, 2,6-di-tert-butylpyridine, DMF, 70 °C, 22 h, 63%; (d) 1. 11, Br2, CH2Cl2, 0 °C, 30 min; 2. 16, Et4NBr, 4 Å MS, DMF/
CH2Cl2, r.t., 3 d, 37%; (e) H2 (1 atm), Pd/C, dioxane/H2O, r.t., 14 h; (f) NaOMe, MeOH, r.t., 12 h, 59%.
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3

Job/Unit: O20744
/KAP1
Date: 20-08-12 11:25:50
Pages: 7
B. Ernst et al.
FULL PAPER
For the assembly of tetrasaccharide 1c (Scheme 3), the ronment clearly demonstrate the flexibility of the arabinose
benzylidene in the chloroacetylated GlcNAc derivative residue in 1b vs. the conformational stability in derivatives
12[24] was selectively reduced[27] to yield acceptor 13, which 1a and 1c.
was glycosylated with donor 14[28] to give trisaccharide 15.
The chloroacetyl group was then selectively removed by
using thiourea, and resulting acceptor 16 was glycosylated
with 5-C-phenylarabinose donor 11. Tetrasaccharide 17 was
then deprotected by treatment with base and careful hydro-
genolysis to selectively yield 1c without affecting the steri-
cally inaccessible benzylic position of the 5-C-phenylarabi-
nose.
In a competitive binding assay,[29] the affinity of sLex (1a)
and its derivatives 1b and 1c for E-selectin was determined.
sLex (1a) showed an affinity of 1 mm (= rIC50 of 1).[30] Ara-
binoside 1b was inactive up to 10 mm (rIC50 Ͼ 10), whereas
5-C-phenylarabinoside 1c could partially restore binding to
E-selectin (rIC50 = 2.8).
To rationalize the biological affinities, the preferred con-
formation of sLex and its two analogs in aqueous solution
was studied by jump-symmetrized ROESY[31] NMR experi-
ments (Figure 2). The average distances between the axial
H5 of l-fucose and its analogs (Ǟ1a–c) and the axial H2
Figure 3. Affinities of selectin ligands 1a–c determined in a compet-
of the d-galactose moiety were determined by analysis of
itive binding assay[29] Whereas 1a and 1c show comparable affin-
the build-up curves of the rotating frame nuclear Over-
ities, arabino derivative 1b did not show binding to E-selectin. The
average inter-residue distances were measured by ROESY NMR
hauser effect. In sLex (1a) and arabinoside 1b, the average
experiments between the axial H2G and the axial H5 of l-Fuc, d-
distance between those two nuclei is nearly identical (1a:
Ara, or 5-C-phenyl-d-Ara. The dynamic range of distances between
2.54 Å, 1b: 2.49 Å), whereas in 5-C-phenylarabinoside 1c
these two nuclei in 1a, 1b, and 1c was assessed in a 12-ns molecular
(2.75 Å) it is increased by 10%.
dynamics simulation and are plotted against time.
It should be noted that the population of 1b at in-
ternuclear distances of 5–7 Å results from torsion of the
glycosidic linkage of the arabinose moiety and not from a
ring flip as a consequence of a reduced number of equato-
rial substituents compared to 1a and 1c (for the conforma-
tional stability of fucose and its derivatives, see the Sup-
porting Information).
Conclusions
Sialyl Lewisx (1a) and analogs 1b and 1c were synthe-
Figure 2. Selective ROESY experiments measuring the transfer of
sized and analyzed for E-selectin binding. Although the
magnetization from the axial H5 of l-fucose or its analogs in selec-
average conformation of the core in sLex (1a) and arabino-
side 1b as determined by NMR spectroscopy are similar,
their bioactivities differ dramatically. Because the methyl
group of l-fucose does not participate in protein binding,
the loss of activity of 1b is related to an increased flexibility
of the core, as observed from molecular dynamics simula-
tions. When the methyl group of l-fucose was replaced by
the lipophilic phenyl group (1a Ǟ 1c), biological activity
was regained as a result of the increased conformational
stability originating from the inter-residual contact between
l-fucose and 5-C-phenyl-d-Ara with the galactose moiety.
Nature has chosen sLex (1a) as a binding epitope for the
tin antagonists 1a, 1b, and 1c to H2G at equal mixing times. The
distance measurement was calibrated on the ROE intensities of the
axial H5 – axial H3 ROEs in the 1C4 chair conformation of the
fucose/fucose derivatives (equal intensities are indicated by boxes
of equal height on H3F, H3A, and H3P for 1a, 1b, and 1c, respec-
tively). Inter-residue distances between the axial H5 of l-Fuc, d-
Ara, or 5-C-phenyl-d-Ara and axial H2G can be calculated from
the analysis of the build-up curves of the ROE transfer. The dis-
tance differences among the selectin ligands are illustrated by the
height of the boxes fixed at the value for 1a. Superscript A is used
for arabinose, F for fucose, G for galactose, and P for 5-C-phenyl-
arabinose.
To gain deeper insight into the dynamic range of the dis-
tances between the fucose-derived residues and the d-galac- selectins for a number of reasons. First, specificity is gained
tose moiety, molecular dynamics (MD) simulations were through the large number of hydrogen bonds involved in
performed. MD simulations (Figure 3) in an aqueous envi- the interaction.[8,32] In blood vessels, leukocytes bind to the
4
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O20744
/KAP1
Date: 20-08-12 11:25:50
Pages: 7
Conformational Constraints: Nature Does It Best with Sialyl Lewisx
cross- and auto-relaxation rates.[37] The extent to which accurate
cross-relaxation rates can be determined by this manner depends
upon how well the auto-relaxation rate can be defined. Alterna-
tively, it is possible to remove the dependence on the auto-relax-
ation by dividing the target-peak by the source-peak for each value
of the mixing time.[38] The resulting function is a hyperbolic tan-
gent, the argument of which is the product of the cross-relaxation
rate and the mixing time. For the longest mixing times performed
and highest rate of cross-relaxation expected for the compounds
studied herein, the hyperbolic tangent function is indistinguishable
from a linear function, hence offering the potential to apply linear
regression to extract the cross-relaxation rate. The above procedure
resulted in values that were well described by linear functions. Re-
moval of autorelaxation through the conversion of biexponential
into hyperbolic tangent functions has as well been recently applied
to determine accurate relaxation rates in cross-correlation measure-
ments.[39]
selectins under shear stress, conditions where preorganiza-
tion of the pharmacophoric groups of sLex (1a) in their
bioactive conformation[7] is essential for success. Our data
are a further example of lipophilic saccharide interac-
tions[14] and document for the first time a lipophilic inter-
residual stabilization of an oligosaccharide structure.
Experimental Section
Conformational Analysis by NMR: The samples for the ROESY
analysis consisted of ca. 5 mg of either 1a, 1b, or 1c, solvated in
99.8% D2O (Armar Chemicals) at pH ≈ 7.0 (uncorrected for D2O)
and were measured without the addition of a buffer. Shigemi NMR
tubes were used to reduce the sample volume (200 μL) needed for
measurement. Measurements were performed at 25 °C using a
Bruker DMX 500 NMR spectrometer. Chemical shifts were refer-
enced with respect to earlier work,[10a] which assigned a chemical
shift of 4.60 ppm to the H5F resonance of 1a.
The doubly-selective homonuclear Hartmann–Hahn scheme[33] was
used to selectively transfer magnetization from H6F to H5F. This
scheme allowed a highly selective transfer of magnetization from
H6F to H5F through their scalar coupling. The selective excitation
of H5F allowed an accurate quantification of this resonance by
avoiding the excitation of residual H2O, which has a similar chemi-
cal shift. To remove any remaining magnetization from H6F, a se-
lective gradient echo at the frequency of H5F was applied. A 200 ms
REBURP[34] 180° refocusing pulse was applied to the H5F reso-
nance. The REBURP pulse was sandwiched by a pair of Gaussian
shaped gradients of 1 ms each and an amplitude of 20 G/cm. This
additional spectral filter ensured that the observed ROESY[35]
peaks were due to magnetization that originated from the H5F reso-
nance.[36]
Molecular Modeling: The 3D structures of all compounds were gen-
erated using MacroModel.[40] Next, a conformational search was
performed to identify the global minimum conformation by sam-
pling a total of 10 000 structures (MacroModel, mixed torsional/
low-mode sampling method, extended torsional sampling, OPLS
2005 force-field,[41] implicit water solvent model). A periodic
boundary system was created by placing the global minimum in a
box of preorganized TIP3P water molecules. The system charge
was neutralized and sodium and chloride ions were added to reach
a physiological electrolyte concentration of 0.15 m. Special atten-
tion was paid to building a proper solvation shell around the solute:
First, the solvent environment was minimized by using a gradient
criterion of 0.1 kcal/mol followed by a 24 ps molecular dynamics
(MD) simulation, so that the water molecules could reorganize
around the solute (the geometry of the solute was kept fixed). The
whole system was then completely minimized by using a gradient
criterion of 0.05 kcal/mol. A 12-ns MD simulation was performed
by using NPT ensemble and standard conditions (T = 300 K, p =
101.325 kPa) with frames sampled every 1.2 ps. All MD simula-
tions were done using Desmond.[42] For the statistical analysis,
structural data (dihedral angle and interatomic distance values)
were determined from the 9Ј950 frames collected during the MD
run (first 50 frames were skipped due to equilibration of the sys-
tem).
The jump-symmetrized CW-ROESY variation of the ROESY se-
quence was used in all experiments to minimize TOCSY arti-
facts.[31] This sub-element of the pulse sequence was inserted fol-
lowing the selective gradient echo. During the ROESY period, the
transmitter frequency was shifted up or downfield during the first
or second half of the mixing-time, respectively. The highfield spin
lock was applied at 4.9 ppm and the lowfield at 0.9 ppm. The spin
lock was a rectangular pulse of 2 kHz amplitude. For each com-
pound, 10 experiments were run to record a build-up curve of the
ROE transfer. The 10 experiments were sampled with increasing
durations of the spin lock, beginning after 50 ms, and repeated af-
ter each 50 ms increment, resulting in a 500 ms spin lock duration
for the final experiment.
Supporting Information (see footnote on the first page of this arti-
cle): Synthetic and analytical details of the described structures.
Traces of the molecular dynamics simulation of ring dihedral angles
of the fucose residue in 1a, the arabinose residue in 1b, and phenyl-
arabinose residue in 1c. Copies of the 1H NMR and 13C NMR
spectra.
Following the application of the spin lock, the transmitter was re-
turned to the center of the spectrum, at 2.9 ppm, and the FID
measured using 4096 complex points to sample a bandwidth of
7 ppm. To achieve a high signal-to-noise ratio, 1024 scans were
measured for each mixing time. Using a prescan delay of 3 s, on
average the experiments lasted approximately 1.2 h each. The NMR
spectroscopic data were analyzed using XWINNMR version 3.0
operating on a Silicon Graphics O2. The spectra were apodized
with an exponential decay function with 2 Hz line broadening. An
additional advantage of the selective experiments was the lack of
signal overlap, which allowed to integrate the signals without inter-
ference from other resonances.
Acknowledgments
The authors gratefully acknowledge financial support of the Swiss
National Science Foundation (grant no. 200020-103875/1) and Gly-
coMimetics, Inc., Gaithersburg, MD, USA.
[1] a) G. S. Kansas, Blood 1996, 88, 3259–3287; b) A. Varki, Proc.
Natl. Acad. Sci. USA 1994, 91, 7390–7397.
[2] a) S. A. Mousa, Drugs Future 1996, 21, 283–289; b) S. A.
Mousa, D. A. Cheresh, Drug Discovery Today 1997, 2, 187–
199.
To determine the internuclear distances, the rotating-frame cross-
relaxation rates were calculated from the build up curves. Tradi-
tionally the cross-relaxation rate is determined from fitting the
spectra to a biexponential function that depends upon both the
[3] B. Ernst, J. L. Magnani, Nat. Rev. Drug. Discov. 2009, 8, 661–
677.
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5

Job/Unit: O20744
/KAP1
Date: 20-08-12 11:25:50
Pages: 7
B. Ernst et al.
FULL PAPER
[4] a) G. Walz, A. Aruffo, W. Kolanus, M. Bevilacqua, B. Seed,
Science 1990, 250, 1132–1135; b) M. L. Phillips, E. Nudelman,
F. C. A. Gaeta, M. Perez, A. K. Singhal, S. I. Hakomori, J. C.
Paulson, Science 1990, 250, 1130–1132; c) M. J. Polley, M. L.
Phillips, E. Wayner, E. Nudelman, A. K. Singhal, S. Hakomori,
J. C. Paulson, Proc. Natl. Acad. Sci. USA 1991, 88, 6224–6228.
2007, 9, 4444–4451; b) K. N. Rao, C. G. Suresh, U. V. Katre,
S. M. Gaikwad, M. I. Khan, Acta Crystallogr., Sect. D: Biol.
Crystallogr. 2004, 60, 1404–1412.
[22] S. Vandenbussche, D. Díaz, M. C. Fernández-Alonso, W. Pan,
S. P. Vincent, G. Cuevas, F. J. Cañada, J. Jiménez-Barbero, K.
Bartik, Chem. Eur. J. 2008, 14, 7570–7578.
[5] M. K. Wild, M.-C. Huang, U. Schulze-Horsel, P. A. [23] G. Terraneo, D. Potenza, A. Canales, J. Jiménez-Barbero, K. K.
van der Merwe, D. Vestweber, J. Biol. Chem. 2001, 276, 31602–
31612 and references cited therein.
Baldridge, A. Bernardi, J. Am. Chem. Soc. 2007, 129, 2890–
2900.
[6] a) E. E. Simanek, G. J. McGarvey, J. A. Jablonowski, C.-H. [24] M. Kiso, S. Ishida, (Otsuka Pharmaceutical Co., Ltd., Japan),
Wong, Chem. Rev. 1998, 98, 833–862; b) J. H. Musser, N. Rao,
M. Nashed, F. Dasgupta, S. Abbas, A. Nematalla, V. Date, C.
Foxall, D. Asa, P. James, D. Tyrrell, B. K. Brandley, Trends
Recept. Res. 1993, 33–40; c) B. Ernst, H. C. Kolb, O. Schwardt
in The Organic Chemistry of Sugars (Eds.: D. E. Levy, P.
Fügedi), CRC Press/Taylor & Francis, Boca Raton, 2006, pp.
828–862.
JP2000086687, 2000.
[25] a) G. Baisch, R. Öhrlein, Bioorg. Med. Chem. 1997, 5, 383–
391; b) G. Baisch, R. Öhrlein, A. Katopodis, Bioorg. Med.
Chem. Lett. 1997, 7, 2431–2434.
[26] B. Mekki, G. Singh, R. H. Wightman, Tetrahedron Lett. 1991,
32, 5143–5146.
[27] A. A. Sherman, Y. V. Mironov, O. N. Yudina, N. E. Nifantiev,
Carbohydr. Res. 2003, 338, 697–703.
[7] a) M. Rinnbauer, B. Ernst, B. Wagner, J. Magnani, A. J. Benie,
T. Peters, Glycobiology 2003, 13, 435–443; b) K. Scheffler, B. [28] Compound 14 was prepared in analogy to A. Kameyama, H.
Ernst, A. Katopodis, J. L. Magnani, W. T. Wang, R. Weisem-
ann, T. Peters, Angew. Chem. 1995, 107, 2034; Angew. Chem.
Int. Ed. Engl. 1995, 34, 1841–1844.
Ishida, M. Kiso, A. Hasegawa, J. Carbohydr. Chem. 1989, 8,
799–804; see also A. Bhunia, O. Schwardt, H. Gäthje, G.-P.
Gao, S. Kelm, A. J. Benie, M. Hricovini, T. Peters, B. Ernst,
ChemBioChem 2008, 9, 2941–2945.
[8] W. S. Somers, J. Tang, G. D. Shaw, R. T. Camphausen, Cell
2000, 103, 467–479.
[29] a) G. Thoma, J. L. Magnani, R. Oehrlein, B. Ernst, F. Schwa-
rzenbach, R. O. Duthaler, J. Am. Chem. Soc. 1997, 119, 7414–
7415; b) G. Weitz-Schmidt, D. Stokmaier, G. Scheel, N. E. Ni-
fantЈev, A. B. Tuzikov, N. V. Bovin, Anal. Biochem. 1996, 238,
184–190.
[30] The relative IC50 value (rIC50) was calculated by dividing the
IC50 of the substance of interest by the IC50 of the reference
compound 1a.
[9] a) M. J. Bamford, M. Bird, P. M. Gore, D. S. Holmes, R. Priest,
J. C. Prodger, V. Saez, Bioorg. Med. Chem. Lett. 1996, 6, 239–
244; b) J. C. Prodger, M. J. Bamford, M. I. Bird, P. M. Gore,
D. S. Holmes, R. Priest, V. Saez, Bioorg. Med. Chem. 1996,
4, 793–801; c) J. C. Prodger, M. J. Bamford, P. M. Gore, D. S.
Holmes, V. Saez, P. Ward, Tetrahedron Lett. 1995, 36, 2339–
2342.
[10] a) G. Thoma, J. L. Magnani, J. T. Patton, B. Ernst, W. Jahnke, [31] J. Schleucher, J. Quant, S. J. Glaser, C. Griesinger, J. Magn.
Angew. Chem. 2001, 113, 1995; Angew. Chem. Int. Ed. 2001,
40, 1941–1945; b) D. Schwizer, J. Patton, B. Cutting, M. Smie-
sko, B. Wagner, A. Kato, C. Weckerle, F. P. C. Binder, S. Rab-
bani, O. Schwardt, J. L. Magnani, B. Ernst, Chem. Eur. J. 2012,
18, 1342–1351.
Reson., Ser. A 1995, 112, 144–151.
[32] R. U. Lemieux, Chem. Soc. Rev. 1989, 18, 347–374.
[33] B. Boulat, R. Konrat, I. Burghardt, G. Bodenhausen, J. Am.
Chem. Soc. 1992, 114, 5412–5414.
[34] H. Geen, R. Freeman, J. Magn. Reson. 1991, 93, 93–141.
[35] A. A. Bothner-By, R. L. Stephens, J. Lee, C. D. Warren, R. W.
Jeanloz, J. Am. Chem. Soc. 1984, 106, 811–813.
[36] B. Cutting, Chimia 2006, 60, 28–32.
[37] a) J. Cavanagh, W. Fairbrother, A. G. Palmer III, N. Skelton
(Eds.), Protein NMR Spectroscopy: Principles and Practice, Ac-
ademic Press, 1995; b) A. Kumar, G. Wagner, R. R. Ernst, K.
Wuethrich, J. Am. Chem. Soc. 1981, 103, 3654–3658; c) E. Chi-
arparin, P. Pelupessy, B. Cutting, T. R. Eykyn, G. Bodenhau-
sen, J. Biomol. NMR 1999, 13, 61–65.
[38] a) H. Desvaux, N. Birlirakis, C. Wary, P. Berthault, Mol. Phys.
1995, 86, 1059–1073; b) H. Desvaux, C. Wary, N. Birlirakis, P.
Berthault, Mol. Phys. 1995, 86, 1048–1058; c) S. Macura, R. R.
Ernst, Mol. Phys. 1980, 41, 95–117.
[39] a) E. Chiarparin, P. Pelupessy, R. Ghose, G. Bodenhausen, J.
Am. Chem. Soc. 1999, 121, 6876–6883; b) P. Lundstroem,
F. A. A. Mulder, M. Akke, Proc. Natl. Acad. Sci. USA 2005,
102, 16984–16989; c) A. Majumdar, R. Ghose, J. Biomol. NMR
2004, 28, 213–227.
[40] MacroModel, Version 9.6, Schrödinger LLC, New York, NY,
2008.
[41] G. A. Kaminski, R. A. Friesner, J. Tirado-Rives, W. L. Jorgen-
sen, J. Phys. Chem. B 2001, 105, 6474–6487.
[42] a) Maestro-Desmond Interoperability Tools, Version 2.0, D. E.
Shaw Research, New York, NY, 2008; b) Desmond Molecular
Dynamics System, Version 2.0, D. E. Shaw Research, New
York, NY, 2008.
[11] N. Kaila, B. E. Thomas, Med. Res. Rev. 2002, 22, 566–601.
[12] a) G. Baisch, R. Ohrlein, A. Katopodis, M. Streiff, F. Kolb-
inger, Bioorg. Med. Chem. Lett. 1997, 7, 2447–2450; b) M.
Rosch, H. Herzner, W. Dippold, M. Wild, D. Vestweber, H.
Kunz, Angew. Chem. 2001, 113, 3954; Angew. Chem. Int. Ed.
2001, 40, 3836–3839.
[13] a) J. Y. Ramphal, Z.-L. Zheng, C. Perez, L. E. Walker, S. A.
DeFrees, F. C. A. Gaeta, J. Med. Chem. 1994, 37, 3459–3463;
b) C. Filser, D. Kowalczyk, C. Jones, M. K. Wild, U. Ipe, D.
Vestweber, H. Kunz, Angew. Chem. 2007, 119, 2155; Angew.
Chem. Int. Ed. 2007, 46, 2108–2111.
[14] R. U. Lemieux, Acc. Chem. Res. 1996, 29, 373–380.
[15] H. C. Kolb, B. Ernst, Chem. Eur. J. 1997, 3, 1571–1578.
[16] B. Cutting, unpublished STD NMR spectroscopic data.
[17] a) N. K. Vyas, Curr. Opin. Struct. Biol. 1991, 1, 732–740; b) M.
Muraki, Protein Pept. Lett. 2002, 9, 195–209; c) Y. Ferrand,
M. P. Crump, A. P. Davis, Science 2007, 318, 619–622; d) L.
Bautista-Ibanez, K. Ramirez-Gualito, B. Quiroz-Garcia, A.
Rojas-Aguilar, G. Cuevas, J. Org. Chem. 2008, 73, 849–857.
[18] J. V. Pratap, A. Arockia Jeyaprakash, P. Geetha Rani, K. Sekar,
A. Surolia, M. Vijayan, J. Mol. Biol. 2002, 317, 237–247.
[19] M. Cygler, D. R. Rose, D. R. Bundle, Science 1991, 253, 442–
445.
[20] J. Screen, E. C. Stanca-Kaposta, D. P. Gamblin, B. Liu, N. A.
Macleod, L. C. Snoek, B. G. Davis, J. P. Simons, Angew. Chem.
2007, 119, 3718; Angew. Chem. Int. Ed. 2007, 46, 3644–3648.
[21] a) E. C. Stanca-Kaposta, D. P. Gamblin, J. Screen, B. Liu, L. C.
Snoek, B. G. Davis, J. P. Simons, Phys. Chem. Chem. Phys.
Received: June 2, 2012
Published Online: ᭿
6
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O20744
/KAP1
Date: 20-08-12 11:25:50
Pages: 7
Conformational Constraints: Nature Does It Best with Sialyl Lewisx
Conformation Analysis
To facilitate the interaction under the shear
stress conditions in the blood vessel, the
conformation of sLex is stabilized via
lipophilic inter-residual contacts. The
methyl group in fucose is optimally suited
to stabilize the sLex core through an inter-
action with the galactose moiety.
A. Titz, A. Marra, B. Cutting,
M. Smiesˇko, G. Papandreou, A. Dondoni,
B. Ernst* ........................................... 1–7
Conformational Constraints: Nature Does
It Best with Sialyl Lewisx
Keywords: Glycoproteins / Conformational
analysis / Carbohydrates / Stacking interac-
tions
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
7