Synthesis and characterization of a paramagnetic sialic acid conjugate as probe for magnetic resonance applications

Article abstract of DOI:10.1016/j.carres.2012.03.002

Magnetic Resonance Imaging (MRI) using paramagnetic systems as contrast agents is receiving increased attention as diagnostic tool in the clinic. At the same time, NMR of paramagnetic systems can also be applied in biochemical fields; for example, the use of Paramagnetic Relaxation Enhancement (PRE) allows structure refinement and the analysis of transient dynamic processes involved in macromolecular complex formation. Herein we report the synthesis and computational characterization of a new DOTA-like sialic acid conjugate, which can be used both in MRI and PRE applications when coordinated to a suitable paramagnetic metal.

Full text of DOI:10.1016/j.carres.2012.03.002

Carbohydrate Research 354 (2012) 21–31
Contents lists available at SciVerse ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Synthesis and characterization of a paramagnetic sialic acid conjugate
as probe for magnetic resonance applications
Davide Bini a, Maria Gregori b, , Ugo Cosentino c, Giorgio Moro a, Angeles Canales e, Alice Capitoli a,
Jesús Jiménez-Barbero a,d, Laura Cipolla a,
⇑
a Department of Biotechnology and Biosciences, University of Milano-Bicocca, Piazza della Scienza 2, 20126 Milano, Italy
b Department of Experimental Medicine, University of Milano-Bicocca, via Cadore 48, 20052 Monza, Italy
c Department of Environmental Sciences, University of Milano-Bicocca, Piazza della Scienza 1, 20126 Milano, Italy
d Centro de Investigaciones Biológicas, CSIC, Ramiro de Maeztu 9, 28040 Madrid, Spain
e Dep. Química Orgánica I, Facultad Ciencias Químicas, Universidad Complutense de Madrid, Avenida Complutense s/n, 28040 Madrid, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Magnetic Resonance Imaging (MRI) using paramagnetic systems as contrast agents is receiving increased
attention as diagnostic tool in the clinic. At the same time, NMR of paramagnetic systems can also be
applied in biochemical fields; for example, the use of Paramagnetic Relaxation Enhancement (PRE) allows
structure refinement and the analysis of transient dynamic processes involved in macromolecular com-
plex formation. Herein we report the synthesis and computational characterization of a new DOTA-like
sialic acid conjugate, which can be used both in MRI and PRE applications when coordinated to a suitable
paramagnetic metal.
Received 13 January 2012
Received in revised form 29 February 2012
Accepted 3 March 2012
Available online 17 March 2012
Keywords:
Sialic acid
Molecular modelling
Magnetic Resonance Imaging
Paramagnetic Relaxation Enhancement
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
cancer, therefore suggesting its involvement in the malignant pro-
cess. These results indicate that high expression of sialidase in can-
N-Acetylneuraminic acid (Fig. 1, NeuAc, 1) represents the most
ubiquitous member of the sialic acid family of derivatives found on
cell surface glycoproteins and glycolipids.1 It plays important roles
in many physiological and pathological processes2 such as regula-
tion of biofluid and mucin viscosity,3 and disaggregation of cells by
repulsive effects.4 Moreover, as terminal substituents, sialic acids
are ideally suited to participate in protein–carbohydrate interac-
tions that mediate cell surface recognition phenomena. Indeed, cell
surface sialosides are known to serve as ligands for microbial tox-
ins,5,6 bacterial and viral adhesins,7 and for mammalian lectins
responsible for cell–cell adhesion.8 In addition, aberrant glycosyla-
tion is known to be a common feature of cancer cells9 and, in par-
ticular, alteration of sialic acids is thought to be a characteristic
associated with malignant properties10 including invasiveness
and metastatic potential. Sialidases, which catalyses the removal
of sialic acid residues from glycoproteins and glycolipids, have also
been suggested to play important roles in many biological pro-
cesses through regulation of sialic acid contents in glycoconju-
gates.11 Altered expression of sialidase has been observed in
cer cells leads to protection against programmed cell death,
probably by modulation of gangliosides.12 It has been shown that
levels of mRNA of human plasma membrane-associated sialidase
(Neu3) are increased in human colon cancer by 3- to 100-fold com-
pared to adjacent non-tumour mucosa.13 These findings highlight
Neu3 as a potential target for diagnosis and therapy of certain type
of cancers.
Nuclear magnetic resonance (NMR) techniques involving para-
magnetic systems are today valuable tools that can find applica-
tions in relevant fields, such as Magnetic Resonance Imaging
(MRI), and in biochemical issues14 with the use of Paramagnetic
Relaxation Enhancement (PRE). Among different diagnostic tech-
niques, Magnetic Resonance Imaging (MRI)15 using paramagnetic
systems as contrast agents is receiving increased attention because
it is non-invasive and provides information about biological struc-
ture and function of whole organisms over time.16,17
Due to the electronic properties of the gadolinium ion, Gd(III)
complexes are currently employed in clinical practice as contrast
agents in MRI techniques.14,15 Contrast agents improve the con-
trast in images by enhancing the nuclear magnetic relaxation rates
of the water protons in the tissues where they are distributed. The
mechanism of relaxation enhancement, that is, relaxivity, involves
dipolar interactions between the magnetic moment of the metal
⇑
Corresponding author. Tel.: +39 02 64482152; fax: +39 02 64483565.
E-mail address: laura.cipolla@unimib.it (L. Cipolla).
Present address.
0008-6215/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.carres.2012.03.002

22
D. Bini et al. / Carbohydrate Research 354 (2012) 21–31
H3C
HN
O
H3C
HN
O
OH
OH
O
O
HO
HO
H
5
H
3
HO
O-
HO
O
OH
OH
H3C
HN
O
4
O
O
O
OH
O
O-
N
O
O
2
OH
OH
O
O
HO
(1) NeuAc
H
N
1
HO
OH
HO
Gd3+
OH
HO
O
OH
N
N
N
N
N
N
N
O
O-
O-
N
N
N
[Gd(Sial)]- (4)
O
OH HO
O
O
OH HO
(3) HE-DO3A
O
(2) Sial
Sialidase
Figure 1. Molecular structures of neuraminic acid (NeuAc, 1), the sialic acid
conjugate Sial (2) and of HE-DO3A (3).
O-
O
OH
OH2
N
N
ion and the nuclear spin of the water protons. Octadentate polya-
mino carboxylate ligands (PAC) such as DOTA (DOTA = 1,4,7,10-
tetrakis(carboxymethyl)-1,4,7,10-tetraaza-cyclododecane), form
kinetically and thermodynamically stable complexes with gadolin-
ium: the ninth coordination site of Gd allows rapidly exchanging
water molecules to transmit the paramagnetic relaxation effect
to the bulk solvent. Today, the search for new Gd-based contrast
agents is oriented toward developing compounds that possess
binding ability toward specific biological targets.15 By conjugating
a suitable targeting vector to the metal chelate it may be possible
to target tissue type, cell type, individual proteins and even specific
nucleic acid sequences. Among the different targeting strategies,
there is an approach proposed for constructing ‘conditional’ MRI
contrast agents, that is, contrast agents that switch states between
low-relaxivity (inactive) and high-relaxivity (active) by the action
of a specific enzyme.18
Gd3+
N
N
O
O-
O-
O
[Gd(HE-DO3A)] (5)
Figure 2. Structure of [Gd(Sial)]ꢁ (4) and proposed activation mechanism (includ-
ing atomic numbering defining the torsion).
s
The conjugate 2, when complexed to a suitable paramagnetic
metal such as gadolinium ([Gd(Sial)]ꢁ, 4, Fig. 2) can be used as a
conditional MRI contrast agent, relying on the hypothesis that
selective activation of the contrast agent in the tumour sites over-
expressing Neu3 will occur. The approach capitalizes on the mod-
ulation relaxation rate that can be achieved by altering the
hydration number (q) of the contrast agent, as reported in the lit-
erature on analogous examples.18,25 Before the cleavage, the sugar
moiety should prevent the contrast agent activity, hindering coor-
dination of a water molecule to the ninth coordination site of the
gadolinium ion (q = 0). In the presence of sialidases, the neuraminic
acid should be cleaved from the chelate allowing coordination of a
water molecule (q = 1) and restoring the contrast agent activity
(Fig. 2). Hence, exploiting differences in sialidase expression and
activity in cancer areas, complex 4 would be cleaved by endoge-
nous overexpressed sialidases. This hypothesis is supported by
the fact that several glycosidases have been reported to cleave
other DOTA-monosaccharide conjugates.25,26
The same sialic acid conjugate 2 can find applications also in
PRE techniques when complexed to different paramagnetic metals.
NMR spectra may usually be obtained from complexes of paramag-
netic lanthanide ions with organic ligands.27 Such spectra
commonly show paramagnetic shifts, mainly of dipolar (pseudo-
contact) origin, of between 0 and 20 ppm, which are approximately
proportional to 1/r3 (r = Ln–H distance). Large low-field shifts can
be produced in the receptor spectrum, and hence yield structural
information in addition to simplifying the spectrum itself. More-
over, NMR of lanthanide derivatives helps in capturing long time
scale motions. As highlighted previously, sialic acids are typically
found at the non-reducing termini of many glycan chains where
it is responsible for carbohydrate mediated intercellular and intra-
cellular events. A suitable paramagnetic neuraminic acid complex
might help in deciphering relevant ligand/receptor interactions
occurring in recognition phenomena involving sialic acids.28
On the other hand, relevant advances have been achieved in the
use of PRE in structure refinement and in the analysis of transient
dynamic processes involved in macromolecular complex forma-
tion.19 The PRE technique is based on magnetic dipolar interactions
between a nucleus (e.g., 1H) and the unpaired electrons of a para-
magnetic centre, and results in an increase in the relaxation rate of
the nuclear magnetization.20 For a given electron–nucleus distance
r, the magnitude of the PRE is proportional to r-6, a relationship
analogous to that between the magnitude of the NOE and interpro-
ton distance. However, because the magnetic moment of the un-
paired electron is large, the PRE effects are large and can provide
long-range distance information, extending in the case of Mn2+
,
for example, up to ꢀ35 Å. Recent applications include for example
rapid elucidation of protein–protein,21 and protein–oligosaccha-
ride interactions and mapping of binding sites.22 Independently
from the magnetic resonance technique of choice, the system
necessitates the attachment of an extrinsic paramagnetic group
to the (macro)molecule of interest through appropriate chemical
modification.
To date only a few examples have been focused on paramag-
netic sialic acid conjugates, and on detection of sialic acids,23 while
several examples of targeted MRI contrast agents toward other gly-
can structures have been reported.24 Herein we report the synthe-
sis of a new DOTA-like sialic acid conjugate, named Sial (2, Fig. 1),
which can be used both in MRI and PRE applications, when coordi-
nated to a suitable paramagnetic metal. The sialic acid conjugate 2
consists of the HE-DO3A ligand (HE-DO3A = 10-(2-hydroxyethyl)-
1,4,7,10-tetraazacyclododecane 1,4,7-tetraacetate, 3, Fig. 1) cova-
lently a-linked to a sialic acid moiety via the hydroxyethyl moiety.

D. Bini et al. / Carbohydrate Research 354 (2012) 21–31
23
tion is stably occupied by one water molecule, even if one water
exchange event occurred during the simulation period (20 ns). Fig-
ure 3 shows the radial distribution function (rdf) of the water oxy-
gen atoms around the ion in the two conformations and Table 1
reports the distance of the first maximum in the rdf. In C1, the max-
imum of the first peak appears at 3.93 Å, thus confirming that no
water molecules coordinate the ion; in C2, the first peak presents
a maximum at 2.55 Å (peak integration provides a coordination
number of 0.83), thus highlighting the quite stable coordination
of one water molecule to the gadolinium ion.
Trajectory analysis (Table 1) shows that the torsions in the five-
member rings involving gadolinium and acetate arms show only
small fluctuations around the average values in the two conforma-
tions. On the other hand, the alcoholic arm adopts the same orien-
tation of the acetate arms in C1, while in C2 it stably adopts an
opposite orientation, although interconversions are here observed
during the simulation (Table 1 and Fig. 4).
Average gadolinium–ligand bond distances (Table 1) compare
well with experimental values in the analogue [Gd-HPDO3A] com-
pound32 (HP-DO3A = 10-(2-hydroxypropyl)-1,4,7,10-tetraazacy-
clododecane 1,4,7-tetraacetate): as already observed,31 the Gd–O
average MD bond distance values are well in line with experimen-
tal ones, while the Gd–N bond distances are slightly greater than
the experimental values.
2. Results and discussion
To evaluate if the designed sialic acid complex 4 could be a
promising candidate for MRI applications, we investigated the pos-
sible conformational equilibria by computational methods.
2.1. Computational studies on [Gd(Sial)]ꢁ
Conformational equilibria in aqueous solution of [Gd(Sial)]ꢁ
have been investigated by ab initio calculations and classical MD
simulations. Solvent effects have been included in the ab initio cal-
culations by means of a continuum approach,29 the polarizable
continuum model (PCM),30 and in MD simulations by explicit
water molecules. MD simulations of the [Gd(Sial)]ꢁ system were
performed describing interactions of the gadolinium ion with the
eight donor atoms of the PAC ligand by a valence force field, while
interactions at the ninth-coordination site of the gadolinium ion
that involve water molecules, or other donor atoms of the sialic
residue, are modelled by non bonded (electrostatic and van der
Waals) potential. Force field parameters for gadolinium–ligand
interactions were those previously developed.31
Preliminary Molecular Mechanics calculations (see Computa-
tional Methods) showed that the (A+) and (Aꢁ) diastereoisomeric
conformations present similar energies, lower than those of B(+)
and (Bꢁ) (see Computational Methods for IUPAC nomenclature of
the four conformations). The (A+) conformation was slightly more
stable and thus was selected for successive investigation. The most
MD simulations show C1 to be the most stable conformation, C2
being 8.5 kcal molꢁ1 above C1
(
D
Etot in Table 2). Analysis of the
energetic contributions shows that although the internal energy
term ( ESia) favours C2 over C1, solvent interactions ( ESia-wat) pre-
vail in determining the overall C1 stability.
relevant conformational degree of freedom in [Gd(Sial)]ꢁ is the
torsion (Gd1–O2–C3–C4 in Fig. 2), defining the relative orientation
of the sialic acid moiety with respect to the coordination cage.
Three starting conformations, differing in the orientation of the sia-
lic acid, were used as starting points for simulations, with
of 0°, 90°, and 180°.
Each conformation was submitted to MD simulations in aque-
ous solution. In the first and third runs the torsion presents on
average values close to the starting ones, that is, ꢁ27° and 172°,
respectively (Table 4); in the second run rapidly (within 40 ps)
increases to a mean value close to that of the third run. Thus MD
simulation indicates the presence of two stable conformations,
here labelled as C1 and C2, in which the different orientations of
the sialic moiety cause different coordination situations around
the ion. In C1, the ninth coordination position is stably occupied
by one carboxylic oxygen of the sialic moiety, with an average
Gd–O5 distance of 2.26 Å, while in C2, the ninth coordination posi-
s
D
D
To check the reliability of MD results, two representative snap-
shots from the C1 and C2 trajectories have been submitted to C-
PCM ab initio geometry optimization. In both calculations, the
presence of one water molecule is considered in the molecular sys-
tem: in C1 the water molecule is not coordinated to the ion and in-
volved in hydrogen bond with the ligand, while it is coordinated to
the ion in C2.
The geometries of the C1 and C2 conformations of
[Gd(Sial)(H2O)]ꢁ calculated at the C-PCM HF/3-21G level are re-
ported in Figure 5. C-PCM ab initio calculations and MD results
provide essentially the same geometrical parameters (Table 1).
Also the relative energies, calculated at the C-PCM PBE1PBE/6-
311G⁄⁄ level, agree with MD results (Table 2): the C1 conformation
is the most stable one and the solvent interactions are decisive in
determining the overall C1 stability. In fact, analysis of the C-PCM
s value
s
s
Table 1
Values of the main geometrical parameters (see Fig. 2 for atom numbering) of the investigated conformations of [Gd(Sial)]ꢁ from MD simulations and ab initio calculations
[Gd(Sial)]ꢁ
MD simulations
[Gd(Sial)(H2O)]ꢁ
C-PCM HF/3-21G
[Gd(HP-DO3A)]
X-ray
C1
C2
C1
C2
Bond distances (Å)
Gd–N
Gd–O
Gd–O2
Gd–O5
2.732 (.176)
2.341 (.059)
2.509 (.058)
2.264 (.066)
3.93a
2.743 (.191)
2.327 (.073)
2.491 (.058)
—
2.751 (.070)
2.346 (.020)
2.516
2.741 (.016)
2.337 (.011)
2.566
—
2.460
2.648 (0.007)
2.348 (0.035)
2.397
—
2.507
2.378
Gd–Owater
2.55a
Torsions (°)
sb
ꢁ27.1 (7.1)
172.3 (7.4)
ꢁ6.9
163.5
—
Acetate arm
N–C–C–O
Gd–N–C–C
ꢁ31.0 (8.3)
ꢁ30.0 (7.9)
ꢁ29.4 (9.6)
ꢁ29.8 (11.5)
ꢁ23.5 (12.2)
41.1 (6.3)
40.6 (6.6)
37.5 (5.4)
37.6 (4.5)
34.1 (4.7)
Alcoholic arm
N–C–C–O2
Gd–N–C–C
ꢁ53.3 (6.5)
42.6 (22.1)
ꢁ9.3 (16.8)
ꢁ59.0
61.3
ꢁ21.2
ꢁ40.4
35.2 (9.2)
45.6
49.2
For comparison, values from the X-ray structure of [Gd(HP-DO3A)] are reported.32 Standard deviations in parenthesis.
a
Gd–Owater distance from the first maximum in the rdf of the water oxygen atoms around the ion.
b
The
s torsion is defined by atoms: Gd1–O2–C3–C4 (see Fig. 2 for atomic numbering).

24
D. Bini et al. / Carbohydrate Research 354 (2012) 21–31
1.5
1.5
1
(a)
(b)
1
0.5
0
0.5
0
2
4
6
8
10
12
14
16
2
4
6
8
10
12
14
16
Figure 3. Radial distribution function of water oxygen atoms around the gadolinium ion in the (a) C1 and (b) C2 [Gd(Sial)]ꢁ conformations.
90
60
30
0
Table 2
τ(°)
Relative energies (D
E, kcal molꢁ1) of C2 conformation
(a)
with respect to C1, calculated from MD simulations and
ab initio calculationsa
D
E (C2–C1)
MD simulationsb
D
D
D
ESia
ESia-wat
ETot
ꢁ5.1 (10.0)
13.5 (22.2)
8.5 (24.3)
Ab initio calculationsc
E°
DDGsol
Gtot
D
ꢁ0.8
6.4
5.6
D
Standard deviations from MD in parentheses.
-30
-60
-90
a
Absolute energy values are reported in Table 1S of
the Supplementary data.
b
D
ESia, relative internal energy; DESia-wat, relative
interaction energy with solvent water molecules;
Etot ESia ESia-wat, relative total energy.
Calculations at the C-PCM PBE1PBE/6-311G⁄⁄ level
on C-PCM HF/3-21G optimized geometries of
[Gd(Sial)(H2O)]ꢁ.
E°, in vacuo relative total energies;
DDGsol Gtot
relative solvation free energies;
E° + DD
sol, relative total free energies in solution.
D
=
D
+ D
c
t(ps)
D
,
D
=
D
G
0
5000
10000
15000
20000
90
τ(°)
energetic contributions shows that in vacuo (DE° values) C2 is
(b)
more stable than C1, while solvent interactions (DDGsol values)
favour C1.
60
30
The C1 conformation, both in ab initio calculations and MD sim-
ulations, does not present a coordinated water molecule, that is, it
is a ‘closed’ form corresponding to a low-relaxivity (MRI ‘inactive’)
state. The C2 conformation, significantly less stable than C1, pre-
sents one coordinated water molecule and should represents an
‘open’ form corresponding to a high-relaxivity (MRI ‘active’) state.
Thus, these results suggest that in the most stable conformation of
[Gd(Sial)]ꢁ, the gadolinium ion is not coordinated by a water mol-
ecule, and thus [Gd(Sial)]ꢁ can exist in an inactive form, able to
switch to the active form by enzymatic cleavage.
0
-30
-60
-90
2.2. Chemical synthesis
The retrosynthetic scheme for the synthesis of complex 4 is out-
lined in Scheme 1. In general, it should be noted that sialic acid
manipulation is particularly challenging due to its complex chem-
ical structure in comparison with other monosaccharides. Key
steps are the glycosylation reaction, which requires stereochemical
t(ps)
0
5000
10000
15000
20000
control toward the formation of the
occurring in sialosides, and the DOTA monoalkylation reaction
with 20-bromoethyl
-sialoside.
a-glycosidic linkage, naturally
Figure 4. Alcoholic arm: trend of the N–C–C–O2 (black) and Gd–N–C–C (grey)
torsions along the trajectory in conformations (a) C1, and (b) C2.
a

D. Bini et al. / Carbohydrate Research 354 (2012) 21–31
25
Figure 5. Molecular geometries of the (a) C1 and (b) C2 conformations of [Gd(Sial)(H2O)]ꢁ calculated at the C-PCM HF/3-21G level.
First steps of the synthesis involve functional group protection
according to literature procedure33 (see Supplementary data for
details).
from a neighbouring C-3 substituent often leads to low stereose-
lectivity and the formation of side-products through a competing
elimination reaction.34 In this context, we concentrated on the
choice of the donor, the promoter and the solvent. The first
glycosylation procedure that has been tried was a Koenigs–Knorr
reaction35 via the peracetylated b-sialyl chloride methyl ester
(see Supplementary data for details).36 However, sialyl chloride
did not give the desired glycosylation product. Thus phenyl and to-
lyl thiosialosides were also tested as donors,37 without any encour-
aging result. As a last attempt, peracetylated derivative 6, was used
as glycosyl donor, as already reported,34b in the presence of differ-
ent Lewis acids as promoters (see Table 5S). The best conditions
To have access to the
able sialyl donor with 2-bromoethanol in the presence of a pro-
moter should be optimized toward -stereoselection. The
a-sialyl conjugate, glycosylation of a suit-
a
complicated molecular architecture of sialic acids imparts a sub-
stantial degree of difficulty in the glycosylation reaction. Even after
appropriate protection, the presence of the anomeric carboxylic
acid destabilizes the formation of the oxocarbenium ion intermedi-
ate through an inductive electron withdrawing effect. The steric
constraints at the C-2 reactive site and the lack of an assistance
OH
OH
O
OH
O
O
O
AcHN
AcHN
HO
O
AcHN
HO
H
H
HO
H
O-
-O
OH
O
O
N
O
O
OH
HN
O
HO
O
O
O
N
HO
HO
HO
OH
OH
OH
N
N
N
Gd3+
N
N
NH
HN
N
N
O-
-O
O
O
OH
HO
OAc
O
OAc
AcHN
AcO
O
OAc
H
AcHN
AcO
O
O
AcHN
AcO
OCH3
HN
H
O
N
AcO
H
OCH3
O
OAc
O
OCH3
OAc
AcO
O
AcO
OAc
OAc
Br
NH
HN
OH
O
AcHN
HO
H
COOH
OH
HO
OH
Scheme 1. Proposed retrosynthesis for [Gd(Sial)]ꢁ.

26
D. Bini et al. / Carbohydrate Research 354 (2012) 21–31
use donor 6 and BF3ꢂOEt2 as the promoter and acetonitrile as the
In a last attempt, we decided to try the conjugation between
sialoside 7 and the unprotected cyclene moiety, by simply stirring
compound 7 with cyclen overnight at rt.25 Surprisingly, no poly-
alkylated by-products were observed (Scheme 3). The best solvent
for the alkylation reaction was CHCl3 both in terms of yield and
reaction rate, among the several tried (dichloromethane, DMF,
CCl4, acetone and DMSO). Polyalkylation products were observed
only when this reaction was performed at reflux temperature.
The only by-product was the elimination product 2S (see
Scheme 1S).
solvent (Table 5S and entry 13), as it is well known that it can help
in affording a-anomeric glycosides through a kinetically controlled
b-nitrilium ion intermediate.38
However, the low stereoselectivity (
a
/b ꢀ 1:1) obtained in the
previous conditions, prompted us to vary other parameters such
as the promoter/acceptor molar ratio and the reaction time. First,
an increase of the promoter/acceptor ratio from 1 to 5 caused a de-
crease in the
a/b ratio. On the other hand, reduction of reaction
time from 3 days to 3 h resulted in an increase in the
a/to b ratio
(Table 3).
Compound 10 was deacetylated under Zémplen conditions (Na,
MeOH)41 and the methyl ester hydrolysed with NaOH 1 M in
H2O.42 Complete alkylation of 12 to yield 2 was achieved in 24 h
by the addition of 3 equiv of bromoacetic acid in H2O with K2CO3
as the base (Scheme 3). This reaction led also to the formation of
trace amounts of incomplete alkylation of the cyclene moiety
and of the esterification on the sialic acid’s carboxylic group. Final-
ly the paramagnetic metal was coordinated to the Sial (2) using
GdCl3 in H2O.
An additional parameter was finally considered, after a recent
publication which reports that a highly -stereoselective sialyla-
tion (27:1 /b) can be achieved by adding a non reactive compound,
a
a
N-methyldiacetamide (DAMA) capable of modifying the supramo-
lecular aggregation state of the glycosyl donor, thus influencing
anomeric stereoselectivity.39 Hence, the glycosylation reaction
was also tried in the presence of DAMA, using different molar con-
centrations of the donor, as summarized in Table 4. However, we
did not observe a significant increase in the stereoselection but
unexpectedly observed a strong reduction in the formation of the
competitive elimination product could be achieved. In summary,
2.3. NMR studies
best conditions found to obtain the
in Table 4 (entry 2), which give 2:1
With the
-20-bromoethylsialoside in hand we proceeded with
the introduction of the HE-DO3A moiety, by alkylation of the suit-
ably protected derivative 8 (Scheme 2). The reaction did not afford
any product. Alternatively, different selective protection of three
out of four nitrogen atoms of the cyclen was also attempted
(BCl3, B(NMe2)3, POCl3),40 but none of them afforded the desired
product.
a
a
-sialoside 7 are those reported
/b ratio and 47% yields.
1D proton NMR spectra were collected for the sialic acid conju-
gate Sial (2) in the presence of increasing amounts of Dy3+ (298 K,
600 MHz with cryoprobe). Addition of the paramagnetic ion gave
rise to huge line broadening of all the proton signals of the mole-
cule, even with a low amount of Dy3+, at a molar ratio 2.5:1, sialic
conjugate:metal (Fig. 6).
a
1H–13C HSQC spectra of 2 acquired at a molar ratio 2.5:1, sialic
conjugate:Dy3+ (Fig. 7) displayed fewer signals than in the absence
of the metal, because only 9-CH2 and the acetamido CH3 group
Table 3
Conditions for glycosylation reactions
Entry
Reagents
Ratio
Concn (M)
Time
T
Results
1
2
3
4
5
6
BF3OEt2–BrEtOH
BF3OEt2–BrEtOH
BF3OEt2–BrEtOH
BF3OEt2–BrEtOH
BF3OEt2–BrEtOH
BF3OEt2–BrEtOH
0.05
0.05
0.5
5
5
1
0.12
0.12
0.12
0.12
0.12
0.12
3 days
3 days
3 days
3 days
5 h
rt
Degradation
Degradation
Degradation
Y = 33% (only b)
0 °C
0 °C
rt
rt
rt
Y = 31% (a/b 1:1)
/b 2:1)
3 h
Y = 30% (
a
Table 4
Conditions for glycosylation reactions using DAMA
Entry
Reagents
Eq
Concn (M)
Time
T
Yield (
a/b)
1
2
BF3OEt2–BrEtOH + DAMA
BF3OEt2–BrEtOH + DAMA
10/10/1
10/10/1
0.016
0.12
1 h + 3 h
1 h + 3 h
rt
rt
Y = 38% (1.5:1)
Y = 47% (2:1)
OAc
O
AcHN
AcO
H
CO2Me
AcO
OAc
AcHN
OAc
O
O
O
Ot-Bu
Ot-Bu
AcHN
OAc
Ot-Bu
a
AcO
b
X
O
AcO
H
CO2Me
H
CO2Me
O
AcO
AcO
N
N
N
N
O
OAc
OAc
OAc
6
7
NH
N
N
N
Br
t-BuO
Ot-Bu
O
O
9
t-BuO
O
O
8
Scheme 2. Reagents and conditions: (a) BF3OEt2, BrEtOH, DAMA, CH3CN; (b) 8, CHCl3, rt.

D. Bini et al. / Carbohydrate Research 354 (2012) 21–31
27
OAc
OH
O
AcHN
AcHN
AcO
HO
H
CO2Me
H
CO2Me
AcO
CO2Me
OAc
O
HO
O
O
AcHN
O
OAc
AcO
OH
H
a
b
AcO
N
HN
N
HN
O
OAc
NH HN
NH HN
Br
7
c
10
11
OH
OH
OH
AcHN
AcHN
AcHN
HO
O
HO
H
CO2H
HO
H
CO2H
HO
H
HO
HO
O-
O
O
O
N
N
O
O
N
N
O
OH
O
O
O-
N
O
O
OH
OH
OH
d
e
N
HN
N
N
Gd3+
NH HN
N
HO
OH
O
O-
O-
O
Sial (2)
12
[Gd(Sial)] (4)
Scheme 3. Reagents and conditions: (a) cyclen, CHCl3; (b) Na met., MeOH; (c) NaOH 1 M, H2O; (d) BrCH2COOH, K2CO3, H2O; (e) GdCl3ꢂ6H2O, H2O.
could be detected. The loss of most of the signals could be due to
the presence of complex mixtures of conformational isomers, as
described by Grzesiek and co-workers43 in other DOTA-like sys-
tems. In order to improve the quality, a new spectrum was re-
corded at a lower temperature (280 K) and at a higher sialic
conjugate:metal ratio.
detection was performed with a solution containing concentrated
H2SO4–EtOH–H2O in a ratio of 5:45:45. Flash column chromatogra-
phy was performed on silica gel 230–400 mesh (Merck). NMR
spectra were recorded at 400 MHz (1H) and at 100.57 MHz (13C)
on a Varian Mercury instrument or at 600 MHz (1H) and at
150.91 MHz (13C) on a Bruker spectrometer equipped with a cryo-
probe. Spectral widths of 15,000 Hz in F1 and 5000 Hz in F2 were
employed, with 32 scans for each of the 256 t1 increments and a
relaxation delay of 1.5 s. Chemical shifts are reported in parts per
million downfield from TMS as an internal standard; J values are
given in Hz. Mass spectra were recorded on a System Applied Bio-
systems MDS SCIEX instrument (Q TRAP, LC/MS/MS, turbon ion
spray) or on a System Applied Biosystem MDS SCIEX instrument
(Q STAR elite nanospray). ESI full MS were recorded with a Thermo
LTQ instrument by direct inlet; relative percentages are shown in
brackets. Elemental analyses (C, H, N) were performed with a Per-
kin–Elmer series II 2400 analyzer or with a Perkin–Elmer CHNS/O
2400 analyse, and all synthesized compounds showed a purity of
more than 95%.
Acquisition of 1H–13C HSQC spectra of 2 at a 3.5:1 molar ratio
and 280 K (600 MHz cryoprobe) allowed the detection of most of
the sialic conjugate signals (Fig. 8). In these conditions, only CH-
6 signal basically disappeared, while CH-7 significantly decreased
its intensity. This is also in agreement with the presence of flexibil-
ity under these experimental conditions.
As the ionic radius of Dy3+ is somewhat smaller than that of
Gd3+
, calculations have been also performed on the [Dy(S-
ial)(H2O)]ꢁ system. The same C1 and C2 conformations have been
found also in this case. Moreover, the relative total free energies
in solution
on C-PCM HF/3-21G optimized geometries, show that C1 is the
most stable conformation (
tot = 5.9 kcal molꢁ1), as found for
the Gd3+ system ( tot = 5.6 kcal molꢁ1).
DG
tot, calculated at the C-PCM PBE1PBE/6-311G⁄⁄ level
D
G
D
G
Comparison of the H/C–metal ion distances in the C1 and C2
conformations found in the computational analysis (Fig. 5) with
the observed experimental NMR data did not allow a clear-cut
non-ambiguous discrimination between the two conformations.
Nevertheless in conformation C2, the metal is rather close to the
C3 protons and a decrease in the intensity of these hydrogens
should be observed. Because no major changes in the intensity of
these signals are observed in the corresponding HSQC, a low partic-
ipation of this conformer should be expected.
3.2. Methyl (2R,4S,5R,6R)-4-(acetyloxy)-2-(2-bromoethoxy)-5-
acetamido-6-[(1R,2S)-1,2,3-tris(acetyloxy)propyl]oxane-2-
carboxylate (7)
To a solution of compound 6 (33 mg, 0.0626 mmol) in dry MeCN
(500
0.0626 mmol) were added. The mixture was left under stirring
L, 0.626 mmol) was added. After
lL), BrEtOH dry (44 lL, 0.626 mmol) and DAMA (7 lL,
for 1 h and then BF3ꢂOEt2 (79
l
2 h satd aq NaHCO3 was added to neutrality and the mixture ex-
tracted with EtOAc. The usual work up followed by flash chroma-
tography (CHCl3–EtOH 98.5:1.5) afforded compound
7 as a
3. Materials and methods
3.1. General
yellowish oil (17 mg, 47%, 33% de). 1H NMR (CDCl3): d = 5.37 (m,
1H, 8-H), 5.31 (d, J = 8.7 Hz, 1H, NHAc), 5.12 (d, J = 9.4 Hz, 1H, 7-
H), 4.87 (m, 1H, 4-H), 4.27 (dd, J = 2.6, 12.4 Hz, 1H, 9a-H), 4.08
(m, 4H, 5-H, 6-H, 9b-H, OCH2a), 3.81 (s, 3H, OCH3), 3.59 (ddd,
J = 5.5, 7.4, 11.2 Hz, 1H, OCH2 b), 3.44 (m, 2H, CH2Br), 2.63 (dd,
J = 4.7, 12.9 Hz, 1H, 3a-H). 13C NMR (CDCl3): d = 171.2, 171.1,
171.0, 170.4 (s, CH3COO), 168.2 (s, CH3CON), 98.9 (s, 2-C), 72.1,
All solvents were dried over molecular sieves, for at least 24 h
prior to use. When dry conditions were required, the reaction
was performed under Ar atmosphere. Thin-layer chromatography
(TLC) was performed on Silica Gel 60 F254 plates (Merck). The

28
D. Bini et al. / Carbohydrate Research 354 (2012) 21–31
Figure 6. 1H NMR spectra of 2 acquired at different molar ratios with Dy3+
.
with Et2O. The residue obtained after water evaporation was redis-
solved in CHCl3 to achieve the precipitation of the unreacted cy-
clen, that is filtered on gooch. The desired product 10 (49 mg)
was obtained as yellow oil (86%). 1H NMR (CDCl3): d = 5.33 (m,
3H, 8-H, NH, 7-H), 4.86 (m, 1H, 4-H), 4.28 (dd, J = 8.4, 10.6 Hz,
1H, 9a-H), 4.05 (m, 3H, 5-H, 6-H, 9b-H), 3.81 (m, 5H, OCH3,
OCH2), 2.86 (m, 18H, NCH2), 2.61 (br s, 1H, 3a-H), 2.14 (s, 6H, 2
OAc), 2.02 (m, 7H, 2 OAc, 3b-H), 1.87 (s, 3H, NHAc). 13C NMR
(CDCl3): d = 171.0, 170.8, 170.3, 170.2, 170.1 (5s, COO), 168.1 (s,
CON), 98.7 (s, 2-C), 72.2, (d, 6-C), 68.9, 67.9, 67.3 (3d, 4-C, 7-C, 8-
C), 62.5 (t, 9-C), 54.2 (t, 20-C), 53.0 (d, 5-C), 51.2, 47.8, 46.4, 45.5,
45.2, 38.3, 38.0 (9t, NCH2 cyclen, 200-C), 49.3 (q, COOCH3), 25.0 (t,
3-C), 23.2, (q, CH3CO-N), 21.2, 21.2, 20.9, 20.8 (4q, CH3COO). Calcd
for C30H51N5O13 (689.76): C, 52.24; H, 7.45; N, 10.15. Found: C,
52.22; H, 7.44; N, 10.16.
3.4. Methyl (2R,4S,5R,6R)-5-acetamido-4-hydroxy-2-[2-
(1,4,7,10-tetraazacyclododecan-1-yl)ethoxy]-6-[(1S,2S)-1,2,3-
trihydroxypropyl]oxane-2-carboxylate (11)
To a solution of 10 (30 mg, 0.043 mmol) in dry MeOH (1.7 mL)
catalytic Na was added. After 3 days the solvent was evaporated.
Compound 11 is a yellowish oil. 13C NMR (CDCl3): d = 184.0 (s,
COOMe), 177.6 (s, CH3CON), 103.3 (s, 2-C), 75.4, 74.6, 71.2, 71.1
(4d, 4-C, 6-C, 7-C, 8-C), 65.3, 59.9 (2t, 9-C, 20-C), 55.2 (d, 5-C),
47.6, 47.2, 46.0, 45.9, 45.9, 44.1 (9 t, NCH2 cyclen, 200-C), 26.0, (q,
CH3CON). Calcd for C22H43N5O9 (521.61): C, 50.66; H, 8.31; N,
13.43. Found: C, 50.65; H, 8.32; N, 13.44.
Figure 7. 1H–13
C HSQC spectra of 2 acquired at a molar ratio 2.5:1, sialic
conjugate:Dy3+
.
(d, 6-C), 69.2, 68.7, 67.4 (d, 4-C, 7-C, 8-C), 65.2, 62.6, (t, 9-C, 20-C),
53.1 (q, COOCH3), 49.4 (d, 5-C), 38.0, (t, 200-C), 30.6, (t, 3-C), 21.4,
(q, CH3CON), 21.0, (q, CH3COO). Calcd for C22H32BrNO13 (598.40):
C, 44.16; H, 5.39; N, 2.34. Found: C, 44.18; H, 5.38; N, 2.35.
3.5. (2R,4S,5R,6R)-5-Acetamido-4-hydroxy-2-[2-(1,4,7,10-
tetraazacyclododecan-1-yl)ethoxy]-6-[(1S,2S)-1,2,3-
trihydroxypropyl]oxane-2-carboxylic acid (12)
3.3. Methyl (2R,4S,5R,6R)-4-(acetyloxy)-5-acetamido-2-[2-
(1,4,7,10-tetraazacyclododecan-1-yl)ethoxy]-6-[(1R,2S)-1,2,3-
tris(acetyloxy)propyl]oxane-2-carboxylate (10)
To a solution of crude 11 in H2O (1.45 mL) 116 lL of NaOH 1 M
were added. After 3 days the solvent was evaporated. Compound
12 is a yellowish oil. 13C NMR (CDCl3): d = 184.2 (s, COOH), 171.1
(s, CH3CON), 103.2 (s, 2-C), 77.4, 75.0, 73.2, 71.0 (4d, 4-C, 6-C, 7-
C, 8-C), 65.3, 64.7 (2t, 9-C, 20-C), 55.5 (d, 5-C), 53.6, 47.6, 47.3,
46.0, 45.4 (10t, NCH2 cyclen, 200-C, 3-C), 25.997 (q, CH3CON). MS
(TOF): m/z = 508.3 [M+H]+. Calcd for C21H41N5O9 (507.59): C,
49.69; H, 8.14; N, 13.80. Found: C, 49.67; H, 8.15; N, 13.81.
To a solution of compound 7 (50 mg, 0.083 mmol) in dry CHCl3
(1 mL) cyclen (29 mg, 0.167 mmol) was added and the reaction
was left stirring overnight at rt. The solvent was then evaporated
under reduced pressure. The crude residue was dissolved in a small
amount of water and the unreacted starting material was extracted

D. Bini et al. / Carbohydrate Research 354 (2012) 21–31
29
Figure 8. 1H–13C HSQC spectra of 2 and 2 complexed to Dy3+ at a molar ratio 3.5:1, sialic conjugate–Dy3+
.
3.6. (2R,4S,5R,6R)-5-Acetamido-4-hydroxy-6-[(1S,2S)-1,2,3-
trihydroxypropyl]-2-{2-[4,7,10-tris(carboxymethyl)-1,4,7,10-
tetraazacyclododecan-1-yl]ethoxy}oxane-2-carboxylic acid (2)
3.7. [Gd{(2R,4S,5R,6R)-5-Acetamido-4-hydroxy-6-[(1S,2S)-1,2,3-
trihydroxypropyl]-2-{2-[4,7,10-tris(carboxymethyl)-1,4,7,10-
tetraazacyclododecan-1-yl]ethoxy}oxane-2-carboxylic acid}]
(4)
Crude 12 was dissolved in H2O (230 lL), adjusting pH to 8 with
HCl 6 N and K2CO3, and bromoacetic acid (5 mg, 0.035 mmol) was
added. The reaction was left stirring at room temperature for
2 days, filtered on gooch and the solvent was then evaporated un-
der reduced pressure giving the desired crude product 2. MS (TOF):
m/z = 680.3 [MꢁH]ꢁ. Calcd for C27H47N5O15 (681.70): C, 47.57; H,
6.95; N, 10.27. Found: C, 47.59; H, 6.96; N, 10.28.
To a solution of crude 2 in H2O (600
l
L) GdCl3ꢂ6H2O (4 mg,
0.0096 mmol) was added and pH adjusted to 5.5. The reaction
was left stirring at room temperature for 3 days monitoring by
MS. To precipitate the metal in excess pH was brought up to 9 with
NaOH 0.1 M and the mixture was then centrifuged. After adjusting
pH to 7 the solution was freeze-dried giving crude 4 as a white pow-

30
D. Bini et al. / Carbohydrate Research 354 (2012) 21–31
der. MS (TOF): m/z = 835.2 [M]ꢁ. Calcd for C27H43GdN5O15 (834.92):
C, 38.84; H, 5.19; N, 8.39. Found: C, 38.82; H, 5.18; N, 8.38.
calculations were performed with the PBE1PBE functional,56 using
the 6-311G⁄⁄ basis sets for the ligand.57 Solvent effects were evalu-
ated by the C-PCM variant58 of GAUSSIAN 03.59 In line with the united
atom topological model (UATM),60 the solute cavity is built as an
envelope of spheres centred on atoms or atomic groups with appro-
priate radii. For the gadolinium(III) ion, the previously parameter-
ized radius61 was used, neglecting the gadolinium dispersion and
repulsion parameters due to their negligible influence in aqueous
solution. All the UATM radii were scaled by a factor of 1.2 in the cal-
culation of electrostatic contributions, while unscaled values were
used to calculate other contributions. To avoid convergence prob-
lems during geometry optimization, the non-electrostatic contribu-
tions to the energy and energy gradient, viz., cavitation, dispersion,
and repulsion contributions, were omitted. Due to the slow conver-
gence the optimization was, in some cases, stopped when the con-
vergence parameters were less than twice the default values. For
this reason frequency analysis was not performed to characterize
the stationary points; thus, the final geometries correspond to sta-
ble conformations for the chosen minimization algorithm, rather
than true minima. As PCM results have free energies status, total
free energy in solution is labelled as Gtot, in vacuo total energy as
3.8. Computational methods
[Gd(Sial)]ꢁ geometry was built from the crystallographic struc-
ture of [Gd(HP-DO3A)].32 In the crystallographic cell of [Gd(HP-
DO3A)], two diastereoisomeric conformations, namely
K(dddd)
and (kkkk), are present and differ only in the configuration of
K
the cyclen: these two conformations are labelled as (A+) and
(Aꢁ), respectively. Their geometries were modified in order to ob-
tain the two other possible diastereoisomeric conformations,
namely the
D(dddd) and D(kkkk) ones, presenting opposite orienta-
tions of the arms, labelled as (B+) and (Bꢁ). Then, the sialic moiety
was connected obtaining the four [Gd(Sial)]ꢁ diastereoisomeric
conformations: these were optimized in vacuo at the MM level
by the appropriate force fields44 in the Sybyl software package.45
On the basis of their relative energies, the (A+) conformation was
selected for further analysis.
3.9. Molecular dynamics simulations
E°, and solvation free energy as
DG
sol. Total free energies in solution
Three starting conformations, differing in the orientation of the
sialic moiety, were considered for simulations. Each complex was
hydrated in a periodic cubic box, approximately 42 Å long per edge,
containing about 2500 Extended Simple Point Charge (SPC/E) water
molecules.46 No counter-ion was added to the [Gd(Sial)]ꢁ system.
The simulations were performed by the GROMACS (version 3.3)
package,47 using the previously developed force fields31 (Table 2S).
Non-bonding interactions were evaluated using a twin range
cut-off. The pair list within the shorter-range cut-off (9.0 Å) was
evaluated at each step, whereas the pair list within the longer
cut-off (14.0 Å) was updated every five steps. To correct for the ne-
glect of electrostatic interactions beyond the 14.0 Å cut-off, a reac-
obtained from single point calculations at optimized geometries in
solution include both the electrostatic and non-electrostatic contri-
butions of all atoms, the exception being the dispersion and repul-
sion contributions of gadolinium, which are omitted. All the
calculations were done using the GAUSSIAN 03 package.59
4. Conclusions
We proposed here the synthesis of a DOTA-like sialic acid con-
jugate that, upon complexation with suitable paramagnetic metals,
can be used both for MRI and PRE applications. Both theoretical
and experimental characterization of this system suggest that the
sialic moiety can prevent the coordination of the water molecule
to the paramagnetic ion; as a result, the complex should exist in
a low-relaxivity (MRI ‘inactive’) state prone to be switched to a
MRI ‘active’ state upon cleavage of the sugar moiety by the siali-
dase. The conjugate, however, exists in solution in different confor-
mations (as deduced form line broadening in the 1H NMR
spectrum), thus limiting the efficiency of the system.
tion field (RF) correction48 with
eRF = 78.0 was used. A preliminary
steepest descendent optimization of the solvated complexes was
done. Each system was then submitted to a MD run. To maintain
constant temperature and pressure, the Berendsen algorithm49
was applied. The complex and the solvent were independently
coupled to a temperature bath (298 K) with a coupling time of
0.1 ps. The pressure was held at 1 bar, with a coupling time of
0.5 ps. The isothermal compressibility was 4.5 ꢃ 10ꢁ5 barꢁ1. Bond
lengths within the complexes involving hydrogen atoms were con-
strained using the LINCS algorithm;50 the bond lengths and angle
in the water molecules were constrained using the SETTLE algo-
rithm.51 Time step was 2.0 fs; simulation time was 20.0 ns for each
run; equilibration time was 0.1 ns. Table 3S shows the relevant
simulation parameters. Coordinates were stored every 0.5 ps.
Atomic charges used in the MD simulations were calculated by fit-
ting the in vacuo HF/6-311G⁄⁄ molecular electrostatic potential
(MEP) by means of the Merz–Kollman method.52 The charges of
atoms with the same MM atom-type and similar chemical environ-
ment were then averaged using RESP software.53 The list of the
atomic charges is reported in Table 4S of the Supplementary data.
Acknowledgements
We gratefully acknowledge financial supports from the
University of Milano-Bicocca (FAR2008) and computational
facilities from CILEA (Consorzio Interuniversitario Lomabardo per
l’Elaborazione Automatica), Programma Operativo Regione
Lombardia Ob. 3 Fondo Sociale Europeo 2000–2006, Sovvenzione
Globale Progetto Ingenio A0001127/2007 and Consorzio
Interuniversitario Nazionale ‘Metodologie e Processi Innovativi di
Sintesi’ (C.I.N.M.P.I.S.).
Supplementary data
3.10. Ab initio calculations
Supplementary data (computational details: atomic charges,
parameter set, simulation parameters) associated with this article
can be found, in the online version, at http://dx.doi.org/10.1016/
j.carres.2012.03.002.
Ab initio calculations on [Gd(Sial)(H2O)]ꢁ use the quasi-relativ-
istic effective core potential (ECP) of Dolg et al.,54 and the related
[5s4p3d]-GTO valence basis sets for the gadolinium ion. This ECP
includes 46+4fn electrons in the core, leaving the outermost 11 elec-
trons to be treated explicitly. Following the previously adopted
computational procedure,44,55 geometry optimizations were per-
formed at the restricted HF level using the 3-21G basis set for the
ligand atoms. On the optimized structures, DFT single point energy
References
1. (a) Schauer, R. Zoology 2004, 107, 49–64; (b) Corfield, A. P.; Schauer, R. In
Schauer, R., Ed.; Sialic Acids, Chemistry, Metabolism and Function in Cell
Biology Monograph; Springer: New York, 1982; Vol. 2, p 5.

D. Bini et al. / Carbohydrate Research 354 (2012) 21–31
31
2. Varki, A.; Cummings, R.; Esko, J.; Freeze, H.; Stanley, P.; Bertozzi, C.; Hart, G.;
Etzler, M. In Essentials of Glycobiology, 2nd ed.; Cold Spring Harbor Laboratory
Press: New York, 2008.
29. Tomasi, J.; Persico, M. Chem. Rev. 1994, 94, 2027–2094.
30. Amovilli, C.; Barone, V.; Cammi, R.; Cances, E.; Cossi, M.; Mennucci, B.; Pomelli,
C. S.; Tomasi, J. Adv. Quantum Chem. 1998, 32, 227–261.
3. Ahmad, F.; McPhile, P. Int. J. Biochem. 1980, 11, 91–96.
4. Kemp, R. B. J. Cell Sci. 1970, 6, 751–766.
31. Cosentino, U.; Pitea, D.; Moro, G.; Saracino, G. A. A.; Villa, A. PCCP 2009, 11,
3943–3950.
5. Severi, E.; Hood, D. W.; Thomas, G. H. Microbiology 2007, 153, 2817–2822.
6. Karlsson, K.-A. Curr. Opin. Struct. Biol. 1995, 5, 622–635.
32. Kumar, K.; Chang, C. A.; Francesconi, L. C.; Dischino, D. D.; Malley, M. F.;
Gougoutas, J. Z.; Tweedle, M. F. Inorg. Chem. 1994, 33, 3567–3575.
33. Kononov, L. O.; Magnusson, G. Acta Chem. Scand. 1998, 52, 141–144.
34. (a) Ress, D. K.; Linhardt, R. J. Curr. Org. Synth. 2004, 1, 31–46; (b) Boons, G.-J.;
Demchenko, A. V. Chem. Rev. 2000, 100, 4539–4565.
7. (a) Ito, Y.; Gaudino, J. J.; Paulson, J. C. Pure Appl. Chem. 1993, 65, 753–762; (b)
Paulson, J. C. In The Receptors; Conn, M., Ed.; Academic Press: New York, 1985;
Vol. 2, p 131; (c) Olofsson, S.; Bergstrom, T. Ann. Med. 2005, 37, 154–172.
8. (a) Springer, T. A.; Larsky, L. A. Nature 1991, 349, 196–197; (b) Paulson, J. C. In
Adhesion: Its Role in Inflammatory Diseases; Harlan, J. M., Liu, D. Y., Eds.; WH
Freeman: New York, 1992; p 19; (c) Chabre, Y. M.; Roy, R. In The Sugar Code.
Fundamentals of Glycosciences; Gabius, H.-J., Ed.; Wiley-VCH: Weinheim, 2009.
9. (a) Yogeeswaran, G.; Salk, P. Science 1981, 212, 1514–1516; (b) Fogel, M.;
Altevogt, P.; Schirrmacher, V. J. Exp. Med. 1983, 157, 371–376; (c) Collard, J. G.;
Schijven, J. F.; Bikker, A.; La Riviere, G.; Bolscher, J. G. M.; Roos, E. Cancer Res.
1986, 46, 3521–3527; (d) Dennis, J. W.; Lafarté, S.; Waghorne, C.; Breitman, M.
L.; Kerbel, R. S. Science 1987, 236, 582–585; (e) Passaniti, A.; Hart, G. W. J. Biol.
Chem. 1988, 263, 7591–7603.
35. Koenigs, W.; Knoor, E. Chem. Ber. 1901, 34, 957–981.
36. De Smet, L. C. P. M.; Pukin, A. V.; Stork, G. A.; De Vos, C. H. R.; Visser, G. M.;
Zuilhof, H.; Sudhölte, E. J. R. Carbohydr. Res. 2004, 339, 2599–2605.
37. Lu, K. C.; Tseng, S. Y.; Lin, C. C. Carbohydr. Res. 2002, 337, 755–760.
38. (a) Pougny, J.; Sinaÿ, P. Tetrahedron Lett. 1976, 45, 4073–4076; (b) Braccini, I.;
Derouet, C.; Esnault, J.; Herve du Penhoat, C.; Mallet, J.-M.; Michon, V.; Sinaÿ, P.
Carbohydr. Res. 1993, 246, 23–41; (c) Schmidt, R. R.; Castro-Palomino, J. C.; Retz,
O. Pure Appl. Chem. 1999, 71, 729–744.
39. Kononov, L. O.; Malysheva, N. N.; Orlova, A. V. Eur. J. Org. Chem. 2009, 611–616.
40. (a) Guillaume, D.; Marshall, G. R. Synth. Commun. 1998, 28, 2903–2906; (b)
Maillet, M.; Kwok, C. S.; Noujaim, A. A.; Snieckus, V. Tetrahedron Lett. 1998, 39,
2659–2662; (c) Bender, J. A.; Meanwell, N. A.; Wang, T. Tetrahedron 2002, 58,
3111–3128.
10. (a) Kim, Y. J.; Varki, A. Glycoconjugate J. 1997, 14, 569–576; (b) Hakomori, S.
Cancer Res. 1996, 56, 5309–5318; (c) Hakomori, S.; Yamamura, S.; Handa, A. K.
Ann. N.Y. Acad. Sci. 1998, 845, 1–10.
11. Miyagi, T.; Wada, T.; Yamaguchi, K.; Hata, K. Glycoconjugate J. 2003, 20, 189–
198.
12. Hakomori, S. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 10231–10233.
13. Kakugawa, Y.; Wada, T.; Yamaguchi, K.; Yamanami, H.; Ouchi, K.; Sato, I.;
Miyagi, T. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 10718–10723.
41. Zèmplen, G. Chem. Ber. 1926, 59, 1254–1266.
42. Rencurosi, A.; Poletti, L.; Guerrini, M.; Russo, G.; Lay, L. Carbohydr. Res. 2002,
337, 473–483.
43. Häussinger, D.; Huang, J.-R.; Grzesiek, S. J. Am. Chem. Soc. 2009, 131, 14761–
14767.
14. (a) Cheng, H.; Markley, J. L. Ann. Rev. Biophys. Biomol. Struct. 1995, 24, 209–237;
(b) Bertini, I.; Luchinat, C.; Licciolini, M. Methods Enzymol. 2001, 339, 314–340;
(c) Ubbink, M.; Worral, J. A. R.; Canters, G. W.; Groenen, E. J. J.; Huber, M. Ann.
Rev. Biophys. Biomol. Struct. 2002, 31, 393–422; (d) Sharp, R. R. Nucl. Magn.
Reson. 2005, 34, 553–596.
15. See for example: (a) Meade, T. J. Acc. Chem. Res. 2009, 42, 821; (b) Yoo, B.; Pagel,
M. D. Front. Biosci. 2008, 13, 1733–1752.
16. (a) Caravan, P.; Ellison, J. J.; McMurry, T. J.; Lauffer, R. B. Chem. Rev. 1999, 99,
2293–2352; (b) Merbach, A. E.; Toth, E. The Chemistry of Contrast Agents in
Medical Magnetic Resonance Imaging; John Wiley and Sons Ltd: West Sussex,
2001.
44. (a) Cosentino, U.; Moro, G.; Pitea, D.; Villa, A.; Fantucci, P. C.; Maiocchi, A.;
Uggeri, F. J. Phys. Chem. A 1998, 102, 4606–4614; (b) Villa, A.; Cosentino, U.;
Pitea, D.; Moro, G.; Maiocchi, A. J. Phys. Chem. A 2000, 104, 3421–3429.
45. Tripos Inc.: 1699 S. Hanley Road, St. Louis, MO, 63144, SYBYL 7.2, 2006.
46. Berendsen, H. J. C.; Grigera, J. R.; Straatsma, T. P. J. Phys. Chem. 1987, 91, 6269–
6271.
47. (a) Berendsen, H. J. C.; van der Spoel, D.; van Drunen, R. Comput. Phys. Commun.
1995, 91, 43–56; (b) Lindahl, E.; Hess, B.; van der Spoel, D. J. Mol. Model. 2001, 7,
306–317.
48. Tironi, I. G.; Sperb, R.; Smith, P. E.; van Gunsteren, W. F. J. Chem. Phys. 1995, 102,
5451–5459.
17. (a) Woods, M.; Kovacs, Z.; Sherry, A. D. J. Supramol. Chem. 2002, 2, 1–15; (b)
Aime, S.; Geninatti Crich, S.; Gianolio, E.; Giovenzana, G. B.; Tei, L.; Terreno, E.
Coord. Chem. Rev. 2006, 250, 1562–1579.
18. Louie, A. Y.; Huber, M. M.; Ahrens, E. T.; Rothbacher, U.; Moats, R.; Jacobs, R. E.;
Fraser, S. E.; Meade, T. J. Nat. Biotechnol. 2000, 18, 321–325.
49. Berendsen, H. J. C.; Postma, J. P. M.; van Gunsteren, W. F.; Di Nola, A.; Haak, J. R.
J. Chem. Phys. 1984, 81. 3684–360.
50. Hess, B.; Bekker, H.; Berendsen, H. J. C.; Fraaije, J. G. E. M. J. Comput. Chem. 1997,
18, 1463–1472.
51. Miyamoto, S.; Kollman, P. A. J. Comput. Chem. 1992, 13, 952–962.
52. Besler, B. H.; Merz, K. M.; Kollman, P. A. J. Comput. Chem. 1990, 11, 431–439.
53. Bayly, C. I.; Cieplak, P.; Cornell, W. D.; Kollman, P. A. J. Phys. Chem. 1993, 97,
10269–10280.
54. Dolg, M.; Stoll, H.; Savin, A.; Preuss, H. Theor. Chim. Acta 1989, 75, 173–194.
55. Cosentino, U.; Villa, A.; Pitea, D.; Moro, G.; Barone, V.; Maiocchi, A. J. Am. Chem.
Soc. 2002, 124, 4901–4909.
56. Adamo, C.; Barone, V. J. Chem. Phys. 2002, 116, 5933–5940.
57. Description of the basis sets and explanation of standard levels of theory can be
found in: Foresman, J. B.; Frisch, A. E. Exploring Chemistry with Electronic
Structure Methods, 2nd ed.; Gaussian Inc.: Pittsburgh, PA, 1996.
58. (a) Barone, V.; Cossi, M. J. Phys. Chem. A 1998, 102, 1995–2001; (b) Cossi, M.;
Rega, N.; Scalmani, G.; Barone, V. J. Comput. Chem. 2003, 24, 669–681.
59. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.;
Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.;
Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.;
Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken,
V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A.
J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G.
A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.;
Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe,
M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A.
GAUSSIAN 03, Revision C.02; Gaussian, Inc.: Wallingford CT, 2004.
60. Barone, V.; Cossi, M.; Tomasi, J. J. Chem. Phys. 1997, 107, 3210–3221.
61. Cosentino, U.; Villa, A.; Pitea, D.; Moro, G.; Barone, V. J. Phys. Chem. B 2000, 104,
8001–8007.
19. Clore, G. M.; Tang, C.; Iwahara, J. Curr. Opin. Struct. Biol. 2007, 17, 603–616. and
references cited therein.
20. Bloembergen, N.; Morgan, L. O. J. Chem. Phys. 1961, 34, 842–850.
21. (a) Mahoney, N. M.; Rastogi, V. K.; Cahill, S. M.; Girvin, M. E.; Almo, S. C. J. Am.
Chem. Soc. 2000, 122, 7851–7852; (b) Mal, T. K.; Ikura, M.; Kay, L. E. J. Am. Chem.
Soc. 2002, 124, 14002–14003; (c) Gross, J. D.; Moerke, N. J.; Van deer Haar, R.;
Lugovsky, A. A.; Sachs, A. B.; McCarthy, J. E.; Wagner, G. Cell 2003, 115, 739–
750; (d) Card, P. B.; Erbel, P. J. A.; Gardner, K. H. J. Mol. Biol. 2005, 353, 664–677.
22. (a) Johnson, P. E.; Brun, E.; MacKenzie, L. F.; Withers, S. G.; McIntosh, L. P. J. Mol.
Biol. 1999, 287, 609–625; (b) Jain, N. U.; Venot, K.; Umemoto, H.; Leffler, H.;
Prestegard, J. H. Protein Sci. 2001, 10, 2393–2400; (c) Macnaughtan, M. A.;
Kamar, M.; Alvarez-Manilla, G.; Venot, A.; Glushka, J.; Pierce, J. M.; Prestegard,
J. H. J. Mol. Biol. 2007, 366, 1266–1281.
23. (a) Lemieux, G. A.; Yarema, K. J.; Jacobs, C. L.; Bertozzi, C. R. J. Am. Chem. Soc.
1999, 121, 4278–4279; (b) Frullano, L.; Rohovec, J.; Aime, S.; Maschmeyer, T.;
Prata, M. I.; Pedroso de Lima, J. J.; Geraldes, C. F. G. C.; Peters, J. A. Chem. Eur. J.
2004, 10, 5205–5217; (c) Shelke, S. V.; Cutting, B.; Jiang, X.; Koliwer-Brandl, H.;
Strasser, D. S.; Schwardt, O.; Kelm, S.; Ernst, B. Angew. Chem., Int. Ed. 2010, 49,
5721–5725.
24. Cipolla, L.; Gregori, M.; So, P.-W. Curr. Med. Chem. 2011, 18, 1002–1018.
25. Duimstra, J. A.; Femia, F. J.; Meade, T. J. J. Am. Chem. Soc. 2005, 127, 12847–12855.
26. (a) Urbanczyk-Pearson, L. M.; Femia, F. J.; Smith, J.; Parigi, G.; Duimstra, J. A.;
Eckermann, A. L.; Luchinat, C.; Meade, T. J. Inorg. Chem. 2008, 47, 56–68; (b)
Urbanczyk-Pearson, L. M.; Meade, T. J. Nat. Protocols 2008, 3, 341–350.
27. Briggsg, J.; Frosti, H.; Hart, A.; Moss, G. P.; Stanifor, M. L. Chem. Commun. 1970,
749–750.
28. (a) Bernardi, A.; Arosio, D.; Potenza, D.; Sánchez-Medina, I.; Mari, S.; Cañada, F.
J.; Jiménez-Barbero, J. Chem. Eur. J. 2004, 10, 4395–4406; (b) Politi, M.; Alvaro-
Blanco, J.; Groves, P.; Prieto, A.; Leal, J. A.; Cañada, F. J.; Jiménez-Barbero, J. Eur.
J. Org. Chem. 2006, 2067–2073; (c) Ernst, B.; Magnani, J. L. Nat. Rev. Drug Disc.
2009, 8, 661–677.