Monovalent and bivalent N-fucosyl amides as high affinity ligands for Pseudomonas aeruginosa PA-IIL lectin

Article abstract of DOI:10.1016/j.carres.2010.03.012

The adhesion of bacteria to human glycoconjugates can be inhibited by soluble glycomimetics that compete with the natural target. Four monovalent and one divalent α-fucosyl amides have been tested for their affinity for a fucose-binding lectin from Pseudomonas aeruginosa. Isothermal calorimetric titrations demonstrated that they bind to the lectin in the micromolar range, with highest affinity for the divalent ligand. Molecular modelling established that, compared to J -fucoside compounds, the glycomimetic amide group resulted in the loss of water-bridged hydrogen bonds that could be partially compensated by additional contact of the aglycone with the protein surface.

Full text of DOI:10.1016/j.carres.2010.03.012

Carbohydrate Research 345 (2010) 1400–1407
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Monovalent and bivalent N-fucosyl amides as high affinity ligands
for Pseudomonas aeruginosa PA-IIL lectin
b,
Manuel Andreini a, Marko Anderluh a, Aymeric Audfray b, Anna Bernardi a, , Anne Imberty
*
*
a Universita’ degli Studi di Milano, Dipartimento di Chimica Organica e Industriale, via Venezian 21, I-20133 Milano, Italy
b CERMAV, UPR5301 CNRS (Affiliated with Université de Grenoble and belonging to ICMG), 601 rue de la chimie, BP 53, 38041 Grenoble, France
a r t i c l e i n f o
a b s t r a c t
Article history:
The adhesion of bacteria to human glycoconjugates can be inhibited by soluble glycomimetics that com-
pete with the natural target. Four monovalent and one divalent -fucosyl amides have been tested for
Received 29 January 2010
Received in revised form 11 March 2010
Accepted 14 March 2010
a
their affinity for a fucose-binding lectin from Pseudomonas aeruginosa. Isothermal calorimetric titrations
demonstrated that they bind to the lectin in the micromolar range, with highest affinity for the divalent
ligand. Molecular modelling established that, compared to J -fucoside compounds, the glycomimetic
amide group resulted in the loss of water-bridged hydrogen bonds that could be partially compensated
by additional contact of the aglycone with the protein surface.
Available online 17 March 2010
Keywords:
Glycosyl amides
Fucose
Ó 2010 Elsevier Ltd. All rights reserved.
Neo-glycoconjugates
Lectin
Pseudomonas aeruginosa
1. Introduction
in adhesion to host glycan.12 Preliminary clinical assays performed
on human and mice with P. aeruginosa pneumonia demonstrated
N-Glycosyl amides are currently under intense scrutiny as po-
that administration of fucose or fucoside reduces bacterial load
in the lungs.13,14 The lectin therefore appears as an attractive target
for therapeutic application and several groups have developed high
affinity glycomimetics and glycodendrimers that can compete with
host glycoconjugates and inhibit bacterial adhesion.15–20
tential effectors of carbohydrate-binding proteins.1 Our group has
reported on the synthesis of a-N-glycosyl amides as mimics of a-
glycosides, potentially endowed with metabolic stability.2–6 NMR
analysis4 revealed that these compounds appear to maintain the
normal pyranose conformation of the monosaccharide involved
and thus can be regarded as bona fide structural mimetics. Further-
A small group of
a-fucosyl amides have been tested for their
affinity for PA-IIL using isothermal calorimetric titrations and were
found to bind in the micromolar range. These results constitute
more, we also showed7 that an
a-fucosyl amide anchor could be
used to create a Lewis-x mimic capable of engaging the carbohy-
drate recognition domain (CRD) of DC-SIGN, a dendritic cell recep-
tor with mannose and fucose specificities, which has been
implicated in the onset of HIV infection.8 This work established
proof of principle that
a-glycosyl amides can perform as effective
mimics of -fucosides also in high affinity, tight binding proteins.
a
2. Results and discussion
A group of 5 -N-fucosyl amides 1–5 were considered in the
a-fucosyl amides as effective functional mimics of
a-fucosides.
The CRD of DC-SIGN, however, has a typical large, shallow and
a
solvent-exposed lectin binding site, which could be rather permis-
sive in terms of structural requirements for bound ligands. Pseudo-
monas aeruginosa PA-IIL lectin, on the contrary, presents a much
more demanding case. This soluble bacterial lectin binds with an
present work (Table 1). The first three compounds, 1–3, were ob-
tained via modifications of the Lewis-x mimic previously described
by our group.7 Compound 4 represents a simple linear amide gen-
erated with b-alanine and compound 5, a dimeric form of the lat-
ter. All ligands were synthesized using our modification7 of
DeShong’s procedure (Scheme 1)21 starting from the known
unusually strong micromolar affinity to L-fucose in a tight binding
site which requires two Ca2+ ions.9–11 P. aeruginosa is the causative
agent of lung infections that are the major cause of mortality for
cystic fibrosis patients and PA-IIL has been proposed to be involved
2,3,4-tri-O-acetyl-fucosyl azide, 7 (9:1
a
/b mixture).7 Thus, treat-
ment of 7 with Ph3P in nitroethane at reflux yielded the oxazoline
8, which was acylated in situ with the pyridyl thioester of (1S,6R)-
6-(benzyloxycarbonylamino) cyclohex-3-enecarboxylic acid7 or
with the pyridyl thioester of benzyloxycarbonyl-b-alanine,22 to af-
ford 9 and 10, in 64% and 52% yield, respectively.
* Corresponding author. Tel.: +33 4 76 03 7636; fax: +33 4 7654 7203.
E-mail addresses: anna.bernardi@unimi.it (A. Bernardi), imberty@cermav.cnrs.fr
(A. Imberty).
0008-6215/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2010.03.012

M. Andreini et al. / Carbohydrate Research 345 (2010) 1400–1407
1401
OAc
OAc
OAc
OAc
O
OAc
OAc
O
Me
Me
e
c
HN
HN
H2N
O
O
OAc
OAc
O
9
11
OAc
OAc
BzCHN
b
OAc
H
O
Me
O
Me
X
N
8
OAc
6 X=OAc
OAc
a
7
X=N3
OAc
O
d
O
Me
NHR
10 R=Cbz
HN
e
12
R=H
Scheme 1. Synthesis of
a-N-fucosyl amides. Reagents and conditions: (a) TMSOTf 0.4 mol equiv, TMSN3, DCM, rt, 94%; (b) Ph3P, EtNO2, reflux, 12 h; (c) pyridyl thioester of
(1S,6R)-6-(benzyloxycarbonylamino) cyclohex-3-enecarboxylic acid, CuCl2, 20 h, 40 °C; (d) pyridyl thioester of Cbz-b-alanine, CuCl2, 20 h, 40 °C; (e) H2, Pd–C.
The nitrogen-protecting group was removed to provide amines
11 and 12 which were functionalized with standard procedures, as
shown in Scheme 2, to afford the final compounds 1–5.
Compounds 1 to 5 were tested for their interaction with PA-IIL
by titration microcalorimetry (Table 1). In all ITC binding experi-
ments, the first injections resulted in large exothermic peaks, indi-
cating an enthalpy-driven interaction (Fig. 1). After titration, no
residual heat of binding was observed.
The stoichiometry of the interaction between PA-IIL and com-
pounds 1 to 3 is close to one, in agreement with the crystal struc-
ture. Compound 4 displays a higher stoichiometry, close to 1.4,
which could be indicative of some nonspecific binding. Interest-
ingly, compound 5, which is a dimeric form of 4, has a stoichiom-
etry lower than 1 and almost half of that of compound 4,
indicating that this compound is able to bind two PA-IIL binding
sites.
As opposed to a a-fucosyl amides have been
-C-fucosides,25,26
reported to adopt the native 1C4(L) chair conformation.4,7 The diag-
nostic large coupling constant value (J2-3 = 10.6 Hz) between fucose
H-2 and H-3 protons in 1–4 allows to confirm this feature for the
compounds examined. The conformation of the cis-b-aminoacid
fragment in 1 and 2 is also of concern. For these compounds, AM-
*
BER calculations predicted a single chair conformation for the
amino acid fragment, featuring the carboxy group in the equatorial
position and the amino group in the axial position (Fig. 2b and c).
Coupling constant analysis allowed for confirmation of the model-
ling results.
It was immediately clear that the water molecule that is con-
served in all structures of PA-IL complexes cannot be maintained
in the presence of N-fucosyl amides. As confirmed by molecular
dynamics study,27 this water molecule receives hydrogen bond
from two amide nitrogens from the protein main chain and gives
two hydrogen bonds to O-6 and O-1 of the fucose residue
(Fig. 2a). The NH group at anomeric position perturbs this network
and creates steric conflicts. The absence of this tightly bound water
For the compounds containing one N-fucosyl amide group, the
dissociation constants vary from 1.2 to 2.1
ligands being compound 1. Such values are between those mea-
sured for -methyl fucoside (KD 0.38 M) or previously obtained
for fucose11 (KD 2.9
M). Compound 5 that contains two N-fucosyl
amide groups is the highest affinity ligand with a dissociation con-
stant of 0.62 M. Analysis of the thermodynamic contribution indi-
lM, the higher affinity
a
l
molecules rationalizes the differences observed between a-methyl
l
fucoside and 4, which is the simplest N-fucosyl amide in this study.
The higher number of stabilizing contacts (hydrogen bonds) when
the water is present explains the stronger enthalpy term (16 kJ/
l
cates that compounds 1 to 4 have a strong enthalpy of binding but
also present favourable entropy contribution. Protein–carbohy-
drate interactions generally present unfavourable entropy contri-
bution,23 but the two-calcium lectins of PA-IIL family often
display a favourable entropy contribution, which has been attrib-
uted to the release of the calcium-coordinating water molecules
upon glycan binding.24 The dimeric compound 5 presents very
strong binding enthalpy that corresponds to twice that of com-
pound 4 (Fig. 1 and Table 1). This term is partly counterbalanced
by an unfavourable entropy term. Comparison of thermodynamic
contribution of 4 and 5 clearly confirms that 5 is able to bind to
two PA-IIL binding sites. The modest gain in affinity indicates that
this compound does not bind to two sites of the same PA-IIL tetra-
mer but more likely bridges two different lectins.
mol in favour of a-methyl fucoside), which is only partly compen-
sated by the freezing of this water that is entropically unfavourable
(12 kJ/mol in favour of N-fucosyl amine). The presence of the water
molecule when binding O-fucose derivatives appears to be of
strong influence on the affinity.
Except for this bridging water molecule, the N-fucosyl amides
establish the same hydrogen bond network between fucose and
PA-IIL binding site as observed in other fucosides (Fig. 2). The
CO group at the glycosidic linkage establishes an additional
hydrogen bond with Ser23. For compounds 1–3, the CO of the
other amide linkage is also predicted to establish hydrogen bond
to this serine. Additional hydrophobic interactions can be estab-
lished between the aromatic ring of these three compounds and
the protein surface. The aglycon portions of compounds 4 and 5
are more flexible and it is not possible to predict the stabilizing
interactions that could occur in the vicinity of the sugar binding
site.
The data show that, despite its tight constraints, the PA-IIL
binding site can accommodate an
form a complex with low micromolar affinity. To rationalize the
decrease in affinity observed going from -methyl fucoside to
a-fucosyl amide moiety and
a
a
-Fucosides are well known for their chemical instability and
mimics of -fucosides have been actively explored. Recently,
both C-fucosides and simple -fucoside ligands bearing various
heterocycles in the aglycon portion were used to synthesize high
affinity multivalent clusters bearing multiple copies of the -fu-
cose epitope.28 The results we report here show that
-N-fucosyl
the amide in our series, a molecular docking study was performed
for 1 and 2 in the binding site of PA-IIL. These two compounds
were built in their low energy conformation, taking into account
the information from 600 MHz spectra NMR for selecting the cor-
rect ring shape of the cyclohexane moiety.7
a-L
a-L
L
a

1402
M. Andreini et al. / Carbohydrate Research 345 (2010) 1400–1407
Table 1
Isothermal calorimetric titration of PA-IIL with 1–5
Compound
n
KA (104 Mꢀ1
)
KD
(lM)
D
G (kJ/mol)
D
H (kJ/mol)
TDS (kJ/mol)
O
O
HO
HO
NH
HN
1.02 ( 0.02)
84 ( 2)
1.2
ꢀ33.8
ꢀ27.1 ( 0.1)
6.7
O
HO
OH
1
OH
O
O
NH
HN
1.11 ( 0.01)
48 ( 3)
2.1
ꢀ32.4
ꢀ22.8 ( 0.1)
9.6
O
OH
HO
2
N
O
O
HO
NH
HN
O
1.13a
47.1
2.1
ꢀ32.3
ꢀ32.8
ꢀ22.2
ꢀ19.3
10.1
HO
OH
3
N
H
HO
HO
O
O
O
1.39a
56.2
1.8
13.5
N
N
H
H
4
HO
O
O
NH HN
O
O
5
0.80 ( 0.01)
161 ( 3)
0.62
0.38
ꢀ35.4
ꢀ36.6
ꢀ38 ( 1)
ꢀ2.6
NH
O
NH
O
HO
OH
HO
OH
OH
OH
α−Me-fucoside
a
-Me-fucoside
1.05 ( 0.01
261 ( 4)
ꢀ35.5 ( 0.6)
1.1
a
Only one experiment performed.
amides can be used to the same effect. Although compared to
O-fucosides, N-fucosyl amides appear to suffer from the presence
of the NH group at the anomeric position, their activity can be
rescued by additional interactions created by the amide side
chain in the vicinity of the fucose binding site. In the future, it
would be of interest to quantify this interaction by direct com-
parison with the corresponding glycoside or anomeric ester. At
3. Experimental
3.1. General
Solvents were dried by standard procedures: CH2Cl2, MeOH and
Et3N were dried over calcium hydride; pyridine was dried over
activated molecular sieves. Reactions requiring anhydrous condi-
tions were performed under nitrogen. 1H and 13C and spectra were
recorded at 400 MHz on a Bruker AVANCE-400 instrument. Chem-
ical shifts (d) for 1H and 13C spectra are expressed in parts per
million (ppm) relative to internal Me4Si as a standard. Signals were
abbreviated as s, singlet; br s, broad singlet; d, doublet; t, triplet; q,
quartet; m, multiplet. Mass spectra were obtained with a Bruker
ion-trap Esquire 3000 apparatus (ESI ionization) or with a VG-Ana-
lytical Autospec Q mass spectrometer or with a FT-ICR Mass Spec-
trometer APEX II & Xmass software. Thin layer chromatography
the same time a-N-fucosyl amides do not correspond to a known
biological chemotype, and therefore, like C-glycosides, they
ought to be stable to the action of hydrolytic enzymes. Informa-
tion on how to design better inhibitors can now be gathered
from the reported docking protocols, which can be used to prior-
itize in silico a series of possible side chains. The simple synthe-
sis and easy functionalization of these compounds together with
their chemical stability make them very interesting candidates
for antagonism or modulation of fucose-binding lectins.

M. Andreini et al. / Carbohydrate Research 345 (2010) 1400–1407
1403
O
O
HO
HO
NH
a,b
HN
O
OH
O
O
1
OH
AcO
AcO
NH
H2N
O
OAc
O
11
HO
NH
c,b
d,b
HN
O
O
HO
HO
OH
2
N
O
O
HO
NH
HN
OH
3
N
H
O
O
O
O
e,b
HO
NH
NHAc
AcO
AcO
NH
NH2
HO
OH
OAc
4
12
HO
HO
O
O
O
N
H
N
H
HO
HO
f,b
HN
O
5
O
HO
N
H
O
HO
Scheme 2. Functionalization of a-N-fucosyl amides. Reagents: (a) 2-acetoxy-benzoylchloride; (b) cat. MeONa, MeOH; (c) HBTU, nicotinic acid; (d) HBTU, indolacetic acid; (e)
Ac2O, pyridine; (f) succinic anhydride, THF, then PyBOP, Et3N, CH2Cl2.
(TLC) was carried out with pre-coated Merck F254 silica gel plates.
Flash chromatography (FC) was carried out with Macherey–Nagel
Silica Gel 60 (230–400 mesh). Unless otherwise indicated, auto-
mated chromatography was carried out on silica gel with a Biotage
System SP1 (KP-Sil cartridge). Compounds 9 and 11 were previ-
ously described.7
carboxyl)-2,3,4-tri-O-acetyl-a-L
-fucopyranosylamine. 1H NMR
(CDCl3): d (ppm) = 0.68 (d, 3H, J5-6 = 6.3 Hz, H6), 1.32–2.05 (m,
8H, CH2cycl), 1.89, 1.94, 2.09, 2.25 (4 ꢁ s, 12H, 4ꢁ–C(O)CH3), 2.77
(m, 1H, H9), 3.60 (m, 1H, H5), 4.26 (m, 1H, H10), 4.99 (s, 1H, H4),
5.24–5.32 (m, 2H, H2, H3), 5.82 (dd, 1H, J1-2 = 5.2 Hz, J1-NH = 8.7 Hz,
H ), 7.04, 7.23, 7.27, 7.80 (4 ꢁ m, 5H, HAr); 13C NMR (CDCl3): d
1
(ppm) = 16.1 (C6), 20.9, 21.5 (4ꢁ–C(O)CH3), 22.0, 24.7, 27.8, 29.2
(4ꢁCH2cycl), 45.0 (C9), 48.8 (C10), 65.8 (C5), 66.6 (C2), 68.3 (C3),
71.0 (C4), 74.5 (C1), 123.4, 126.7 (2 ꢁ CAr), 128.3 (Cipso), 130.1,
132.3 (2 ꢁ CAr), 148.2 (Cipso), 165.7 (C12), 169.7, 169.9, 170.6,
170.97 (4 ꢁ –C(O)CH3), 174.8 (C8); ESI-MS: m/z = 599 [M+Na]+,
100%; Rf = 0.30 (hexane–EtOAc 1:1).
3.2. N-((1S,2R)-2-(2-Hydroxybenzamido)cyclohexanecarboxyl)-
a-
L-fucopyranosyl amine 1
2-Acetoxybenzoic acid (14 mg, 0.08 mmol) was refluxed in 1 mL
of toluene in the presence of oxalyl chloride (3 mmol) during 3 h.
The solution was then evaporated and added to a solution of 11
(30 mg, 0.072 mmol) and Et3N (2 equiv) in 1 mL of THF. The solu-
tion was stirred at room temperature for 18 h, the solvent was
evaporated and the crude dissolved with CH2Cl2. The solution
was washed with a satd aq NaHCO3, 10% citric acid and an aqueous
saturated NaCl solution. The solvent was evaporated under
vacuum and the residue was purified by automated chromatogra-
Deprotection under Zemplen’s conditions afforded 1 in quantita-
tive yield. 1H NMR (D2O): d (ppm) = 0.78 (d, 3H, J5-6 = 6.5 Hz, H6),
1.48–2.05 (m, 8H, CH2cycl), 3.00 (m, 1H, H9), 3.61 (dq, 1H,
J5-6 = 6.5 Hz, J4-5 <1 Hz, H5), 3.67 (dd, 1H, J3-4 = 3.4 Hz, H4), 3.83 (dd,
1H, J2-3 = 10.6 Hz, H3), 4.00 (dd, 1H, J1-2 = 5.6 Hz, H2), 4.56 (m, 1H,
H
10), 5.48 (d, 1H, H1), 7.08, 7.49, 7.84 (3 ꢁ m, 4H, HAr); 13C NMR
(D2O): d (ppm) = 16.1 (C6), 21.6, 23.9, 24.2, 30.4 (4 ꢁ CH2cyl), 45.6
(C9), 49.6 (C10), 66.8 (C2), 67.9 (C5), 70.4 (C3), 72.3 (C4), 77.7 (C1),
118.0 (CAr), 118.2 (Cipso), 121.3, 130.4, 135.0 (3 ꢁ CAr), 157.3 (Cipso),
phy using
a hexane–EtOAc gradient to yield 50% (21 mg,
0.038 mmol) of N-((1S,2R)-2-(2-acetoxybenzamido)cyclohexane-

1404
M. Andreini et al. / Carbohydrate Research 345 (2010) 1400–1407
Figure 1. Titration microcalorimetry of PA-IIL by 4 (a) and 5 (b) with titration curves (top) and integration of heat released with the solid line representing the best least-
squares fit (bottom). Titration performed at 25° by 30 automatic injections of 10 L compound (0.5 mM) added every 300 s to PA-IIL containing cell (0.05 mM).
l
Figure 2. Complex of PA-IIL with fucose (a), 1 (b) and 2 (c). Complex with fucose and hydrogen bond network (yellow dotted lines) of crystalline water (green dotted lines) is
adapted from.10 Complexes with 1 and 2 result from the docking study. For (b) and (c) hydrogen atoms are not represented for clarity. (For interpretation of the references to
colour in this figure legend, the reader is referred to the web version of this article.)
169.1, 178.6 (2 ꢁ C@O); ESI-MS: m/z = 431.5 [M+Na]+, 30%; FT-ICR
to the reaction mixture and the organic phase was washed with
0.5 M NaOH (10 mL), 1 M KHSO4 (10 mL), water (10 mL) and brine
(10 mL). The organic phase was dried over Na2SO4, filtered and the
solvent evaporated under vacuum. The residue was purified via
automated chromatography (EtOAc) to yield 77% (22.8 mg,
0.035 mmol) of N-((1S,2R)-2-(3-pyridinecarboxamido)cyclohexan-
(ESI) calcd. for
C
20H28N2O7Na [M+Na]+: 431.17942, found:
431.17927; Rf = 0.75 (CHCl3–MeOH 4:1).
3.3. N-((1S,2R)-2-(3-Pyridinecarboxamido)cyclohexanecarboxyl)-
a-
L-fucopyranosyl amine 2
ecarboxyl)-2,3,4-tri-O-acetyl-
a-L-fucopyranosylamine.
To a solution of 11 (30 mg, 0.072 mmol) in 0.7 mL of CH2Cl2,
1H NMR (CDCl3): d (ppm) = 0.89 (d, 3H, J5-6 = 6.3 Hz, H6), 1.36–
0
0
0
0
0
0
Et3N (0.26 mmol, 3 equiv), nicotinic acid (13.2 mg, 0.08 mmol,
1.5 equiv) in 0.7 mL of CH2Cl2 were added. Subsequently, HBTU
(0.08 mmol, 1.5 equiv) was added and the reaction mixture was
stirred at room temperature. After 18 h, 10 mL of CH2Cl2 was added
1.48 (m, 3H, H4 , H5 ), 1.61–1.72 (m, 3H, H3 ax, H4 , H5 , H6 ax), 1.92
0
0
(m, 1H, H6 eq), 2.06 (m, 1H, H3 eq), 1.93, 1.96, 2.08 (2 ꢁ s, 3H,
0
0
2 ꢁ –(O)CCH3), 2.79 (m, 1H, H1 ), 3.87 (m, 1H, H5 ), 4.30 (m, 1H,
0
H2 ), 5.13 (m, 1H, H4), 5.24 (dd, 1H, J2-3 = 11.0 Hz, J3-4 = 3.1 Hz,

M. Andreini et al. / Carbohydrate Research 345 (2010) 1400–1407
1405
H3), 5.31 (dd, 1H, J1-2 = 5.2 Hz, J2-3 = 11.0 Hz, H2), 5.85 (dd, 1H, J1-2
5.2 Hz, J1-7 = 8.1 Hz, H1), 7.18 (d, 1H, J1-7 = 8.1 Hz, H7), 7.35 (d, 1H,
=
disappearance of the starting material and appearance of the
oxazoline 8. The reaction mixture was used directly in the next
step without isolation. In a separate vessel, the pyridyl thioester
of b-alanine22 (0.147 g, 0.465 mmol; 1.3 equiv) and CuCl2 H2O
(0.079 g, 0.465 mmol; 1.3 equiv) were dissolved in 1 mL of EtNO2
and added to the solution of 8. The reaction mixture was heated
to 40 °C and monitored by TLC (CHCl3/EtOAc, 1/1). After 20 h, the
mixture was filtered through a Celite pad, and Celite was washed
abundantly with EtOAc. The filtrate was washed with an aqueous
solution of NH3–NH4Cl (pH 9), then with water to neutral pH.
The organic phase was dried over Na2SO4, and the solvent was
evaporated under vacuum. The residue was purified by automated
chromatography (hexane–EtOAc gradient 7:3?1:1) to obtain
0.092 g (0.186 mmol) of N-(N0-benzyloxycarbonyl-b-alanyl)-2,3,4-
0
J2 -9 = 9.2 Hz, H9), 7.38 (dd, 1H, J14-15 = 4.9 Hz, J15-16 = 7.9 Hz, H15),
8.0 (d, 1H, J15-16 = 7.9 Hz, H16), 8.72 (d, 1H, J14-15 = 4.9 Hz, H14),
9.10 (pseudo-s, 1H,
H
12); 13C NMR (100 MHz, CDCl3):
d
0
(ppm) = 16.0 (C6), 20.5, 20.6, 20.7 (3 ꢁ –(O)CCH3), 22.3, 23.3 (C4 ,
0
0
0
0
0
C5 ), 27.4 (C6 ), 29.2 (C3 ), 44.58 (C1 ), 48.62 (C2 ), 65.6 (C5), 66.2
(C2), 68.1 (C3), 70.5 (C4), 74.2 (C1), 123.4 (C15), 134.8 (C16),
148.3 (C12), 152.1 (C14), 165.1 (C10), 169.3, 170.4, 170.6 (3 ꢁ –
(O)CCH3), 174.8 (C8); ESI-MS: m/z = 542 [M+Na]+, 83%; Rf = 0.30
(EtOAc).
Deprotection under Zemplen0s conditions afforded 2 in 95%
yield. 1H NMR (D2O): d (ppm) = 0.70 (3H, d, J5-6 = 6.5 Hz, H6),
1.35–1.91 (m, 8H, CH2CYCL), 2.91 (m, 1H, H9), 3.55 (m, 1H, H5),
3.36 (pseudo-d,1H, J4-5 = 3.4 Hz, H4), 3.77 (dd, 1H, J2-3 = 10.6 Hz,
H3), 3.92 (dd, 1H, J1-2 = 5.7 Hz, H2), 4.43 (m, 1H, H10), 5.43 (d, 1H,
H1), 7.52 (dd, 1H, J = 5.0 Hz, J = 7.7 Hz, HAr), 8.05 (d, 1H, J = 8.0 Hz,
tri-O-acetyl-a-L
-fucopyranosyl amine 10 (52%). 1H NMR
(400 MHz, CDCl3): d (ppm) = 1.13 (d, 3H, J6-5 = 6.4 Hz, H6), 2.01–
2.19 (3 s, 9H, 3 ꢁ –C(O)CH3), 2.55 (br s, 2H, C(O)CH2CH2NHAc),
3.53 (br dd, 2H, J1 = 6.4 Hz, J2 = 6.8 Hz C(O)CH2CH2NHAc), 3.95
(dq, 1H, J5-6 = 6.4 Hz, J5-4 <1 Hz, H5), 5.08–5.17 (m, 2H, H3, H4),
H
Ar), 8.69 (m, 2H, HAr); 13C NMR (D2O): d (ppm) = 16.2 (C6), 20.6,
23.6, 30.3 (4 ꢁ CH2CYCL), 45.6 (C9), 49.7 (C10), 66.7 (C2), 67.9 (C5),
70.3 (C3), 72.1 (C4), 77.5 (C1), 137.1, 148.2, 152.4 (4 ꢁ CAr), 169.1
(C12), 178.6 (C8); ESI-MS: m/z = 394.4 [M+H]+, 43%; FT-ICR (ESI)
5.25 (br s, 2H, C(O)OCH2Ph), 5.39 (dd + br s, 2H, J2-3 = 11.0 Hz, J2-1
=
5.6 Hz, H2, C(O)CH2CH2NHAc), 5.89 (dd, 1H, J1-2 = 5.6 Hz, J1-
NH = 7.6 Hz, H1), 6.85 (d, 1H, J1-NH = 7.6 Hz, a-NHC(O)); Rf = 0.35
calcd for
C
19H28N3O6 [M+H]+: 394.19781, found: 394.19801;
Rf = 0.33 (CHCl3–MeOH 4:1).
(CHCl3–MeOH 50:1).
To a solution of 10 (0.1 g, 0.2 mmol) in MeOH (3 mL), 10% Pd–C
(0.015 g) was added and the suspension was stirred for 18 h under
a H2 atmosphere. The catalyst was filtered through a Celite pad and
the solvent evaporated. The crude N-(b-alanyl)-2,3,4-tri-O-acetyl-
3.4. N-((1S,2R)-2-(3-Indolacetamido)cyclohexanecarboxyl)-a-L-
fucopyranosyl amine 3
Using the same procedure described for 2, compound 11
(30 mg, 0.072 mmol) and 3-indolacetic acid (13.9 mg, 0.08 mmol)
were condensed using HBTU (1.5 equiv) to yield 68% (28 mg,
0.049 mmol) of N-((1S,2R)-2-(3-indolacetamido)cyclohexanecarb-
a-L-fucopyranosyl amine 12 was used without further purification
for the synthesis of 4 and 5. An analytical sample was purified by
automated chromatography (KP-NH cartridge) and characterized.
1H NMR (400 MHz, CDCl3): d (ppm) = 1.09 (d, 3H, J6-5 = 6.4 Hz,
H6), 1.93–2.11 (3 s, 9H, 3 ꢁ –C(O)CH3), 2.33 (t, 2H, J = 4.4 Hz,
C(O)CH2CH2NHAc), 2.93–3.05 (m, 2H, C(O)CH2CH2NHAc), 3.93
(dq, 1H, J5-6 = 6.4 Hz, J5-4 <1 Hz, H5), 5.13 (dd, 1H, J3-2 = 11.2 Hz,
J3-4 = 3.2 Hz, H3), 5.20 (dd, 1H, J4-3 = 3.2 Hz, J4-5 <1 Hz, H4), 5.31
(dd, 1H, J2-3 = 11.2 Hz, J2-1 = 5.2 Hz, H2), 5.83 (dd, 1H, J1-2 = 5.2 Hz,
oxyl)-2,3,4-tri-O-acetyl-
d (ppm) = 0.86 (d, J5-6 = 6.4 Hz, H6), 1.31–2.10 (m, 8H, H3 , H4 , H5 ,
a-L
-fucopyranosylamine. 1H NMR (CDCl3):
0
0
0
0
0
H6 ), 1.99, 2.00, 2.14 (3 ꢁ s, 3H, 3 ꢁ –C(O)CH3), 2.84 (m, 1H, H1 ),
0
3.68 (m, 2H, H11), 3.83 (m, 1H, H5), 4.22 (m, 1H, H2 ), 5.14 (m,
1H, H4), 5.34 (dd, 1H, J1-2 = 5.2 Hz, J2-3 = 11.2 Hz, H2), 5.43 (dd,
1H, J2-3 = 11.2 Hz, J3-4 = 3.2 Hz, H3), 5.90 (dd, 1H, J1-2 = 5.2 Hz, J1-7
8.4 Hz, H1), 6.59 (d, 1H, H9), 7.07–7.51 (m, 5H, HAr), 7.89 (d, 1H,
J1-7 = 8.4 Hz, H7), 8.75 (s, 1H, H14); 13C NMR (CDCl3):
=
J1-NH = 8.2 Hz, H1), 9.40 (d, 1H, J1-NH = 8.2 Hz, a
-NHC(O)); 13C
NMR (100 MHz, CDCl3, DEPT): d (ppm) = 16.2 (C6), 20.72, 20.73,
20.75 (3 ꢁ –(O)CCH3), 37.5 (CH2), 38.0 (CH2), 65.7 (CH), 66.3
(CH), 68.3 (CH), 70.7 (CH), 74.2 (CH), 170.8, 172.1; ESI-MS: m/
z = 361.0 [M+H]+, 100%.
d
(ppm) = 16.10 (C6), 20.7, 20.8, 20.9 (3 ꢁ –(O)CCH3), 22.4, 22.7
0
0
0
0
0
0
(C4 , C5 ), 26.9 (C6 ), 29.2 (C3 ), 33.5 (C11), 44.9 (C1 ), 47.5 (C2 ), 65.6
(C5), 66.2 (C2), 68.2 (C3), 70.7 (C4), 74.3 (C1), 108.6 (C12), 111.5,
118.4, 122.5, 123.5 (C16, C17, C18, C19), 126.9 (C20), 136.2 (C15),
169.5, 170.5, 170.7 (3 ꢁ –(O)CCH3), 171.7 (C10), 174.4 (C8); ESI-
MS: m/z = 594 [M+Na]+, 100%; Rf = 0.68 (EtOAc).
3.6. N-(N0-Benzyloxycarbonyl-b-alanyl)-
a-L-fucopyranosyl
amine 4
Deprotection under Zemplen0s conditions afforded 3 in quanti-
tative yield. 1H NMR (CD3OD): d (ppm) = 1.14 (d, 3H, J5-6 = 6.4 Hz,
H6), 1.30–1.84 (m, 8H, CH2cycl), 2.84 (m, 1H, H9), 3.74–3.91 (m,
5H, CH2INDOL, H3, H4, H5), 4.07 (dd, 1H, J1-2 = 5.7 Hz, J2-3 = 10.6 Hz,
H2), 4.39 (m, 1H, H10), 5.56 (d, 1H, H1), 7.24–7.39 (m, 3H, HAr),
7.61 (d, 1H, J = 8.2 Hz, HAr), 7.70 (d, 1H, J = 7.9 Hz, HAr); 13C NMR
(CD3OD): d (ppm) = 16.5 (C6), 21.0, 23.6, 23.8, 30.5 (4 ꢁ CH2cyl),
33.2 (CH2INDOL), 45.7 (C9), 48.8 (C10), 66.8 (C2), 68.0 (C5), 70.3
(C3), 72.2 (C4), 77.5 (C1), 108.4 (Cipso-14), 112.7, 119.2, 120.3,
122.9, 125.6 (5 ꢁ CAr), 128.6, 137.1 (2 ꢁ Cipso), 174.8 (C12), 178.4
(C8); ESI-MS: m/z = 446.3 [M+H]+, 100%; FT-ICR (ESI) calcd for
The crude amine was dissolved in CH2Cl2 (1 mL), pyridine
(0.02 mL, 1.2 equiv) and Ac2O (0.025 mL, 1.2 equiv) were added
and the solution was stirred overnight. The mixture was diluted
with CH2Cl2, extracted with water, diluted HCl, water. The organic
phase was evaporated and the crude purified by flash chromatog-
raphy (CH2Cl2–MeOH 20:1) to yield 73 mg (90% over the two steps)
of acetamide as a white foam. 1H NMR (400 MHz, CDCl3): d
(ppm) = 1.18 (d, 3H, J6-5 = 5.6 Hz, H6), 2.01–2.19 (4 s, 12H, 3 ꢁ –
C(O)CH3 + NHC(O)CH3), 2.57 (br t, 2H, C(O)CH2CH2NHAc), 3.50–
3.62 (m, 2H, C(O)CH2CH2NHAc), 4.10 (br q, 1H, J5-6 = 5.6 Hz, H5),
5.28 (br s, 2H, H4, H3), 5.37–5.42 (m, 1H, H2), 5.94 (dd, 1H, J1-2
5.6 Hz, J1-NH = 7.6 Hz, H1), 6.24 (br s, 1H, C(O)CH2CH2NHAc), 7.07
(d, 1H, J1-NH = 7.6 Hz, -NHC(O)).
=
C
23H32N3O6 [M+H]+: 446.22911, found: 446.22895; Rf = 0.11
(CHCl3–MeOH 4:1).
a
Deprotection under Zemplen’s conditions afforded 4 in 75% yield.
1H NMR (CD3OD): d (ppm) = 1.90 (d, 3H, J6-5 = 5.6 Hz, H6), 1.93 (s, 3H,
NHC(O)CH3), 2.50 (t, 2H, J = 6 Hz, C(O)CH2CH2NHAc), 3.43 (t, 2H,
J = 6 Hz, C(O)CH2CH2NHAc), 3.65 (br s, 1H, H4), 3.73–3.80 (br s, 2H,
H5, H3), 3.93–4.10 (m, 1H, H2), 5.55 (dd, 1H, J1-2 = 5.6 Hz,
J1-NH = 7.6 Hz, H1), 8.05 (br s, 1H, C(O)CH2CH2NHAc), 8.35 (d, 1H,
3.5. N-(b-Alanyl)-2,3,4-tri-O-acetyl-a-L-fucopyranosyl amine 12
To a solution of fucosyl azide 7 (0.113 g, 0.358 mmol; 1 equiv)
in dry EtNO2 (5 mL), grounded activated molecular sieves (4 Å)
were added. PPh3 (0.103 g, 0.394 mmol; 1.1 equiv) dissolved in
EtNO2 (5 mL) was added and the mixture was refluxed for 18 h.
The reaction was monitored by TLC (CHCl3/EtOAc = 1/1) to observe
J1-NH = 7.6 Hz,
21.15 (NHC(O)CH3), 35.1 (C(O)CH2CH2NHAc), 35.4 (C(O)CH2CH2N-
a
-NHC(O)); 13C NMR (CD3OD): d (ppm) = 15.5 (C6),

1406
M. Andreini et al. / Carbohydrate Research 345 (2010) 1400–1407
HAc), 66.5 (C2), 77.2, 70.0, 72.3 (C4), 77.1 (C1), 172.0, 173.6; ESI-MS:
MAV), and information from high resolution NMR. Partial charges
were derived by MNDO calculations. High resolution crystal struc-
ture of PA-IIL/fucose complex (pdb code 1GZT) was used as the
template for docking study with inclusion of all hydrogen atoms
and proper partial charges and parameters for the calcium ions
as previously described.29 Fucose moiety of compounds 1 and 2
were superimposed on the position of fucose in the crystalline
complex. Conformational analysis of rotatable bond of the aglycon
was performed with CO–NH linkage maintained in a trans confor-
mation. Conformations with non-steric conflicts were energy min-
imized using the Tripos force-field30 with PIM energy parameters31
for the fucose moiety, and geometry optimisation of the ligand and
side chains in the binding site. The docking approach was per-
formed in the presence or absence of the crystalline water mole-
cule present in the binding site.
m/z = 277 [M+H]+, 100%.
3.7. Dimer 5
To a solution of 10 (0.07 g, 0.14 mmol) in MeOH (2 mL), 10% Pd–C
(0.01 g) was added and the suspension was stirred for 18 h under a
H2 atmosphere. The catalyst was filtered through a Celite pad and
the solvent evaporated. The crude amine was dissolved in 2 mL of
dry THF and succinic anhydride (0.007 g, 0.07 mmol, 0.5 equiv)
was added to the ice-cooled solution. After 30 min, a solution of Py-
BOP (0.054 g, 0.1 mmol, 0.75 equiv) in CH2Cl2 (1 mL) was added
dropwise and the mixture was stirred at room temperature over-
night. The solvent was evaporated; the residue dissolved in CH2Cl2
and was washed with a satd aq NaHCO3 and brine. The organic phase
was dried with Na2SO4 and the solvent evaporated. The crude was
purified by flash chromatography (CH2Cl2/MeOH gradient, from
20:1?9:1) to yield 35 mg (63%) of protected dimer. 1H NMR
(400 MHz, CD3OD): d (ppm) = 1.12 (d, 6H, J6-5 = 6.4 Hz, 2 ꢁ H6),
2.01–2.19 (3 s, 18H, 6 ꢁ –C(O)CH3), 2.45 (s, 4H, C(O)CH2CH2C(O)),
2.52 (t, 4H, J = 6.4 Hz, 2 ꢁ C(O)CH2CH2NHAc), 3.46 (td, 4H,
J1 = 6.4 Hz, J2 = 2.8 Hz, 2 ꢁ C(O)CH2CH2NHAc), 4.07 (dq, 2H,
J5-6 = 6.4 Hz, J5-4 <1 Hz, 2 ꢁ H5), 5.23 (dd, 2H, J2-1 = 5.6 Hz,
J2-3 = 11.2 Hz, 2 ꢁ H2), 5.30 (br d, J4-3 = 3.4 Hz, 2 ꢁ H4), 5.49 (dd,
2H, J3-4 = 3.4 Hz, J3-2 = 11.2 Hz, 2 ꢁ H3), 5.85 (d, 2H, J1-2 = 5.6 Hz,
2 ꢁ H1).
Acknowledgements
This paper was supported by the FIRB program CHEM-PROFAR-
MANET (RBPR05NWWC) by the ANR (PCV08_322689 GLYCOAST-
RISK), by Vaincre la Mucoviscidose and by GDR Pseudomonas.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.carres.2010.03.012.
Deprotection under Zemplen’s conditions afforded the dimer 5
in 78% yield. 1H NMR (400 MHz, CD3OD): d (ppm) = 1.19 (d, 6H,
J6-5 = 6.8 Hz, 2 ꢁ H6), 2.42–2.56 (m, 8H, 2 ꢁ C(O)CH2CH2NHAc,
C(O)CH2CH2C(O)), 3.39–3.52 (m, 4H, 2 ꢁ C(O)CH2CH2NHAc), 3.67
(br d, 2H, J4-3 = 2.8 Hz, 2 ꢁ H4), 3.79–3.82 (m, 4H, 2 ꢁ H3 + 2 ꢁ H5),
References
1. Norris, P. Curr. Topics Med. Chem. 2008, 8, 101–113.
2. Bianchi, A.; Bernardi, A. Tetrahedron Lett. 2004, 45, 2231–2234.
3. Bianchi, A.; Russo, A.; Bernardi, A. Tetrahedron: Asymmetry 2005, 16, 381–386.
4. Bianchi, A.; Bernardi, A. J. Org. Chem. 2006, 71, 4565–4577.
3.97 (dd, 2H, J2-3 = 10.0 Hz, J2-1 = 5.4 Hz, 2 ꢁ H2), 5.58 (d, 2H, J1-2
=
5.4 Hz, 2 ꢁ H1); 13C NMR (100 MHz, CD3OD): d (ppm) = 15.5 (C6),
30.7, 35.3, 35.5, 66.6 (C2), 67.3, 70.0, 71.9 (C4), 77.0 (C1), 173.5,
173.8; TOF MS ES+: m/z = 551.3 [M+H]+, 573.2 [M+Na]+, 100%;
HRMS (TOF ES+): calculated for C22H39N4O12: 551.2564, found
551.2572.
5. Nisic, F.; Bernardi, A. Carbohydr. Res. 2008, 343, 1636–1643.
6. Nisic, F.; Andreini, M.; Bernardi, A. Eur. J. Org. Chem. 2009, 5744–5751.
7. Timpano, G.; Tabarani, G.; Anderluh, Ml.; Invernizzi, D.; Vasile, F.; Potenza, D.;
Nieto, P. M.; Rojo, J.; Fieschi, F.; Bernardi, A. ChemBioChem 2008, 9, 1921–1930.
8. Geijtenbeek, T. B.; Kwon, D. S.; Torensma, R.; Vliet, S. J.; van Duijnhoven, G. C.;
Middel, J.; Cornelissen, I. L.; Nottet, H. S.; KewalRamani, V. N.; Littman, D. R.;
Figdor, C. G.; van Kooyk, Y. Cell 2000, 100, 587–597.
9. Mitchell, E.; Houles, C.; Sudakevitz, D.; Wimmerova, M.; Gautier, C.; Perez, S.;
Wu, A. M.; Gilboa-Garber, N.; Imberty, A. Nat. Struct. Biol. 2002, 9, 918–921.
10. Mitchell, E. P.; Sabin, C.; Šnajdrová, L.; Pokorná, M.; Perret, S.; Gautier, C.; Hofr,
3.8. ITC (isothermal titration microcalorimetry) analysis
ˇ
C.; Gilboa-Garber, N.; Koca, J.; Wimmerová, M.; Imberty, A. Proteins: Struct.
ITC experiments were performed with a VP-ITC isothermal
titration calorimeter (Microcal). The experiments were carried
out at 25 °C. Ligands and PA-IIL were dissolved in the same buffer
of 0.1 M Tris with 0.03 mm CaCl2 at pH 7.5. The protein concentra-
Funct. Bioinform. 2005, 58, 735–748.
11. Sabin, C.; Mitchell, E. P.; Pokorná, M.; Gautier, C.; Utille, J.-P.; Wimmerová, M.;
Imberty, A. FEBS Lett. 2006, 580, 982–987.
12. Imberty, A.; Wimmerova, M.; Mitchell, E. P.; Gilboa-Garber, N. Microb. Infect.
2004, 6, 222–229.
13. Hauber, H. P.; Schulz, M.; Pforte, A.; Mack, D.; Zabel, P.; Schumacher, U. Int. J.
Med. Sci. 2008, 5, 371–376.
tion in the microcalorimeter cell (1.4 mL) was set to 50
lM. A total
of 30 injections of 10 L of sugar solution at concentrations of
l
14. Chemani, C.; Imberty, A.; de Bentzman, S.; Pierre, P.; Wimmerová, M.; Guery, B.
P.; Faure, K. Infect. Immun. 2009, 77, 2065–2075.
0.5 mm were added at intervals of 5 min whilst stirring at
310 rpm. Control experiments performed by injection of buffer into
the protein solution yielded insignificant heats of dilution. The
experimental data were fitted to a theoretical titration curve using
15. Marotte, K.; Sabin, C.; Préville, C.; Moumé-Pymbock, M.; Wimmerova, M.;
Mitchell, E. P.; Imberty, A.; Roy, R. ChemMedChem 2007, 2, 1328–1338.
16. Morvan, F.; Meyer, A.; Jochum, A.; Sabin, C.; Chevolot, Y.; Imberty, A.; Praly, J.
P.; Vasseur, J. J.; Souteyrand, E.; Vidal, S. Bioconjugate Chem. 2007, 18, 1637–
1643.
17. Deguise, I.; Lagnoux, D.; Roy, R. New J. Chem. 2007, 31, 1291–1299.
18. Marotte, K.; Préville, C.; Sabin, C.; Moumé-Pymbock, M.; Imberty, A.; Roy, R.
Org. Biomol. Chem. 2007, 5, 2953–2961.
19. Johansson, E. M.; Crusz, S. A.; Kolomiets, E.; Buts, L.; Kadam, R. U.; Cacciarini,
M.; Bartels, K. M.; Diggle, S. P.; Camara, M.; Williams, P.; Loris, R.; Nativi, C.;
Rosenau, F.; Jaeger, K. E.; Darbre, T.; Reymond, J. L. Chem. Biol. 2008, 15, 1249–
1257.
software supplied by Microcal, with
(association constant) and n (number of binding sites per mono-
mer) as adjustable parameters. Free energy change ( G) and entro-
S) were determined from the standard
DH (enthalpy change), Ka
D
py contributions (T
D
equation:
D
G ¼
D
H ꢀ T
DS
20. Kolomiets, E.; Swiderska, M. A.; Kadam, R. U.; Johansson, E. M.; Jaeger, K. E.;
Darbre, T.; Reymond, J. L. ChemMedChem 2009, 4, 562–569.
21. Damkaci, F.; DeShong, P. J. Am. Chem. Soc. 2003, 125, 4408–4409.
22. Adamczyk, M.; Fishpaugh, J. R. Tetrahedron Lett. 1996, 37, 4305–4308.
23. Dam, T. K.; Brewer, C. F. Chem. Rev. 2002, 102, 387–429.
24. Imberty, A.; Mitchell, E. P.; Wimmerová, M. Curr. Opin. Struct. Biol. 2005, 15,
525–534.
where T is the absolute temperature. All experiments were per-
formed with c values 100 < c < 200 (c = KaM, where M is the initial
concentration of the macromolecule).
3.9. Docking calculations
25. Carpintero, M.; Bastida, A.; Garcia-Junceda, E.; Jiménez-Barbero, J.; Fernandez-
Mayoralas, A. Eur. J. Org. Chem. 2001, 4127–4135.
Three-dimensional structure of 1 and 2 were built using the
Sybyl software editor (Tripos inc) starting from the fucose struc-
ture available from the monosaccharide database (Glyco3D, CER-
26. Dondoni, A.; Catozzi, N.; Marra, A. J. Org. Chem. 2004, 69, 5023–5036.
ˇ
ˇ
27. Mishra, N. K.; Kulhánek, P.; Šnajdrová, L.; Petrek, M.; Imberty, A.; Koca, J.
Proteins 2008, 72, 382–392.

M. Andreini et al. / Carbohydrate Research 345 (2010) 1400–1407
1407
28. Imberty, A.; Chabre, Y. M.; Roy, R. Chem. Eur. J. 2008, 14, 7490–7499.
29. Nurisso, A.; Kozmon, S.; Imberty, A. Mol. Simul. 2008, 34, 469–479.
30. Clark, M.; Cramer, R. D. I.; van den Opdenbosch, N. J. Comput. Chem. 1989, 10,
982–1012.
31. Imberty, A.; Bettler, E.; Karababa, M.; Mazeau, K.; Petrova, P.; Pérez, S. In
Perspectives in Structural Biology; Vijayan, M., Yathindra, N., Kolaskar, A. S., Eds.;
Indian Academy of Sciences and Universities Press: Hyderabad, 1999; pp 392–
409.