New cyclodextrin dimers and trimers capable of forming supramolecular adducts with shape-specific ligands

Article abstract of DOI:10.1039/b812172a

Bridged cyclodextrin dimers and trimers, in which respectively two and three hydrophobic cavities lie in close proximity, display much higher binding affinities and molecular selectivities than do parent cyclodextrins (CDs). By joining βCD units with links inserted at different positions (2-2′, 3-2′, 6-2′ or 6-2′-6″) and interposing spacers of different lengths and shapes, multicavity structures can be synthesized that are precisely tailored to fit specific guest molecules. This enzyme-mimicking strategy can also be used to generate stable supramolecular adducts. A series of CD dimers and trimers was prepared in good yields by carrying out the critical synthetic steps under power ultrasound (US) or microwave (MW) irradiation. Starting from azide and acetylenic CD derivatives, we exploited an efficient MW-promoted Huisgen 1,3-dipolar cycloaddition in the presence of Cu(I) salts. The resulting bridged CD derivatives gave stable adducts with magnetic-resonance-imaging contrast agents (MRI CAs) containing gadolinium(III) chelates. These inclusion complexes were found to be 2 to 3 orders of magnitude more stable than those formed by βCD and to be endowed with high relaxivity values.

Full text of DOI:10.1039/b812172a

View Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
New cyclodextrin dimers and trimers capable of forming supramolecular
adducts with shape-specific ligands
Silvio Aime,a Eliana Gianolio,a Francesca Arena,a Alessandro Barge,b Katia Martina,b George Heropoulosc and
Giancarlo Cravotto*b
Received 16th July 2008, Accepted 17th October 2008
First published as an Advance Article on the web 20th November 2008
DOI: 10.1039/b812172a
Bridged cyclodextrin dimers and trimers, in which respectively two and three hydrophobic cavities lie in
close proximity, display much higher binding affinities and molecular selectivities than do parent
cyclodextrins (CDs). By joining bCD units with links inserted at different positions (2-2¢, 3-2¢, 6-2¢ or
6-2¢-6¢¢) and interposing spacers of different lengths and shapes, multicavity structures can be
synthesized that are precisely tailored to fit specific guest molecules. This enzyme-mimicking strategy
can also be used to generate stable supramolecular adducts. A series of CD dimers and trimers was
prepared in good yields by carrying out the critical synthetic steps under power ultrasound (US) or
microwave (MW) irradiation. Starting from azide and acetylenic CD derivatives, we exploited an
efficient MW-promoted Huisgen 1,3-dipolar cycloaddition in the presence of Cu(I) salts. The resulting
bridged CD derivatives gave stable adducts with magnetic-resonance-imaging contrast agents (MRI
CAs) containing gadolinium(III) chelates. These inclusion complexes were found to be 2 to 3
orders of magnitude more stable than those formed by bCD and to be endowed with high relaxivity
values.
monoacetylenes resulting in the formation of a 1,2,3-triazole
Introduction
bridge.19
Cyclodextrins (CDs) find numerous important applications in
Owing to problems of selectivity, efficiency and purification,
chemical, biological and pharmaceutical technologies.1 Their
monosubstitution and disubstitution of CDs still remain chal-
areas of use include analytical separations,2 food manufacturing,3
lenging targets. We found that non-conventional techniques
enzyme mimics,4 drug carriers,5 and the delivery of medical
(US or MW irradiation) could markedly improve several CD
diagnostic agents.6 They are among the most widely used oligosac-
charide hosts for drug complexation, because CD-encapsulated
drugs usually have a better bioavailability, a longer half-life in the
body, unhindered excretion and no extra toxicity.7
functionalizations, to great advantage in terms of yields and
reaction times.20–22
Mono-O-propargyl-CDs and monoazido-deoxy-CDs proved to
be versatile building blocks for the construction of all types of
homo- and heterodimers, whether joined ‘head-to-head’, ‘head-to-
tail’ or ‘tail-to-tail’. Selective mono- and disubstitutions to provide
azido derivatives could usually be achieved via monotosylation20
as described in our improved protocols that afforded the 3- and
6-monoazido as well as the 6,6¢-diazido derivative.22 The latter is
a suitable building block for preparing CD trimers.
Mono-O-propargylation on the secondary face was not a
straightforward procedure; protective groups had to be inserted at
C-6 and strict operative conditions adhered to in order to minimize
time-consuming chromatographic purifications (Scheme 1). An
alternative strategy for the synthesis of dimers employed a spacer
molecule bearing two azido groups or two acetylenic moieties that
reacted with suitably monosubstituted CDs (Scheme 2).
The key step in the present synthetic protocol is the Huisgen
1,3-dipolar cycloaddition of azides and alkynes, which under
catalysis with copper(I) salts react regioselectively to give 1,4-
disubstituted 1,2,3-triazoles. It is experimentally straightforward
and widely applicable because it shows a high tolerance toward
many functional groups and proceeds to completion in a large
variety of solvents including aqueous t-butanol and DMF. The
catalyst is usually prepared in situ by reduction of Cu(II) salts
(which are less expensive and often purer than Cu(I) salts),
Since the pioneering work by Tabushi8 and Harada,9 much
effort has been devoted to making CD dimers (bis-CDs) with a
variety of functional tethers.10 Although syntheses and properties
of homodimers linked through their primary11 or secondary12 faces
are well documented, few heterodimers and trimers (tris-CDs)13
have been reported to date and not much is known about their
binding properties.
We have previously described improved US- and MW-promoted
protocols for bCD homodimerization through 6-6¢ and 2-2¢
ureido- and thioureido bridges14 or alkenyl bridges built by cross-
metathesis reactions in the presence of 2nd-generation Grubbs
catalyst.15 Very recently we introduced a new, efficient synthetic
protocol16 for preparing homo- and heterodimers of a-, b-
and gCD. This was based on
a
MW-promoted17 Huis-
gen 1,3-dipolar cycloaddition18 of CD monoazides and CD
aDipartimento di Chimica IFM, Universita` di Torino, Via Giuria 7, 10125,
Torino, Italy
bDipartimento di Scienza e Tecnologia del Farmaco, Universita` di Torino,
Via Giuria 9, 10125, Torino, Italy. E-mail: giancarlo.cravotto@unito.it;
Fax: +390116707687; Tel: +390116707684
cInstitute of Organic and Pharmaceutical Chemistry, National Hellenic Res.
Found., Vas. Constantinou Av. 48, Athens, 116 35, Greece
370 | Org. Biomol. Chem., 2009, 7, 370–379
This journal is
The Royal Society of Chemistry 2009
©

View Online
Results and discussion
Our first goal was to prepare a library of CD dimers and trimers
whose CD units would be joined ‘head-to-head’, ‘head-to-tail’ and
‘tail-to-tail’ through a rigid, chemically stable spacer. The second
goal was to investigate their ability to host in their supramolecular
cavities, and thus firmly bind, CAs bearing suitable substituents
(e.g. one or more cyclohexyl groups; see Fig. 1).
The key synthetic step was the Cu(I)-catalyzed Huisgen 1,3-
dipolar cycloaddition by which monoazido and monoacetylenic
CD derivatives could be efficiently linked together through a
1,2,3-triazole moiety. Our speedy MW-promoted protocol was
characterized by short reaction times, good yields and easy work-
up.
We adopted two alternative strategies: the first was to directly
react two CD derivatives bearing an azido and an alkyne group
respectively; the second one used a bifunctional spacer – either
1,3-bis(azidomethyl)benzene or 1,3-bis(propargyloxy)benzene –
to join together two appropriate CD derivatives by generating
two triazole bridges. Scheme 1 resumes the preparation of mono-
propargyl derivatives. 6-O-t-Butyldimethylsilyl-bCD,24 when re-
acted under reflux in anhydrous THF with a quasi-stoichiometric
amount of propargyl bromide (1.1 eq) in the presence of lithium
hydride, yielded the 2-monoalkynyl derivative and traces of the
3-alkylated product. When a moderate excess of bromoalkyne
(3 eq) was used instead, the dialkylated CD was isolated as the
major product.
Scheme 1 Synthesis of monopropargyl CD derivatives. Reagents and
conditions: a) TBDMSCl, imidazole, dry pyridine, stirring, rt., 8 h; b)
LiH, dry THF/DMSO, propargyl bromide, reflux, 4 h.
usually with L-ascorbic acid or sodium L-ascorbate. When this
cycloaddition was carried out under MW irradiation, the reaction
time was dramatically reduced.
The other aim of the present work was to investigate our
newly-synthesized, water-soluble CD multimers as host molecules
for adducts containing Gd(III) chelates. The resulting complexes
should find application as contrast agents (CAs) for MRI diag-
nostic medical investigations. As they have much larger molecular
masses than the CAs themselves, they should be endowed with
markedly higher relaxivities owing to their longer rotational
correlation times and consequently generate better-contrasted
images.
6-Monoazido-bCD and 3-monoazido-a-, b- and gCD (5 and
10–12 respectively) were prepared as recently described by
ourselves20,21 and so was 6A,6D-dideoxy-6A,6D-diazido-bCD (18),
obtained by displacing with sodium azide under MW the primary
(C-6) sulfonic ester group of 6A,D-capped-bCD.
Scheme 2 Synthesis of bis-CDs and tris-CDs via Huisgen cycloaddition. Reagents and conditions: a) CuSO4/ascorbic acid, t-BuOH/H2O (1:1), MW
85 ◦C, 150 Wmax, 40 min; b) AcCl 2% in CH3OH/CH2Cl2 1:2.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 370–379 | 371
©

View Online
efficiently, by chelation with diethylenetriamine-pentaacetic acid
sodium salt. Protecting silyl groups were cleaved off 17 and 19
in a few minutes under acidic conditions (2% acetyl chloride in
methanol). The final products are shown in Scheme 4.
We studied how our CD dimers and trimers interacted with
suitably functionalized CAs for diagnostic Magnetic Resonance
Imaging (MRI),25 viz. the Gd complexes shown in Fig. 1. Ditopic
guest molecules can be expected to bind to bis-CDs more strongly
than monotopic ones. The efficacy of a Gd(III) chelate as CA
for MRI is related to its relaxivity (r1p), which can be much
enhanced by attaching it to a very large molecule such as a bis-
CD.15 To minimize its toxicity, the resulting adduct should be
non-covalent rather than covalent in nature, a condition that can
be satisfied if the Gd(III) chelate contains suitable hydrophobic
groups that will behave as guests in the CD cavities. Relative
binding parameters listed in Table 1 (association constants Ka,
relaxivities Rb of the adducts and number of binding sites n),
were determined by the Proton Relaxivity Enhancement (PRE)
method (for relevant equations see reference 26), which exploits
the increase in relaxation rate of the paramagnetic complex that
is determined by its binding to a macromolecular substrate. The
binding parameters were obtained from two different kinds of
titration: the first, in which a fixed concentration of the Gd
chelate was titrated with variable amounts of CD dimer/trimer
(see example in Fig. 2), allowed an accurate determination of Ka
and Rb; the second, in which a fixed concentration of the CD
dimer/trimer was titrated with the Gd chelate (see example in
Fig. 3), allowed the determination of the number of equivalent,
independent binding sites. In the latter, as the binding affinity
was sufficiently high, the plot of the experimental data showed a
breaking point corresponding to substrate saturation. By fitting
the experimental points with suitable PRE equations, we obtained
the number of binding sites on the CD dimer/trimer. While for
the monotopic Gd-1 complex the number of binding sites was
found to be 1, for all the investigated CDs, for bitopic and tritopic
Gd-2 and Gd-3 complexes the number of binding sites on the
CD dimer/trimer turned out to be fractional (0.5–0.7). Further
support to these findings emerged when the data were displayed
in the form of Scatchard plots (Fig. 4), where the intercept on the
x axis yields the number of binding sites on the substrate.27 To
Fig. 1 Structures of gadolinium complexes used to prepare host–guest
adducts.
Acetylenic and azido derivatives were subjected to cycload-
dition in t-butanol/water (1:1) in the presence of CuSO4 and
L-ascorbic acid under MW (90 ◦C, 150 W).17 The same protocol
was employed to obtain symmetric homodimers (4, 6) using
either of the above-mentioned bifunctional cross-linkers, viz.
1,3-bis(azidomethyl)benzene (AMB) and 1,3-bis(propargyloxy)
benzene (POB). AMB was used in our previous work to prepare
cross-linked CD derivatives.24
When we subjected 5 to cycloaddition with a 2-mono-O-
propargyl-6-TBDMS-CD, we obtained respectively the asym-
metric homodimer 8 in the case of bCD derivative 2, and two
asymmetric heterodimers (7, 9) in the case of a- or gCD derivatives
(1, 3). As we aimed to compare the respective binding properties
of such bis-CDs (joined by unlike faces) with those of a bis-CD
that was joined by like faces, we carried out the cycloaddition
on 3-monoazido-a, -b or -gCD (11) and 2-mono-O-alkylated-
bCD (2).
We then attempted the synthesis of a tris-CD from dipropar-
gylated derivatives (16) and 6-monoazido-bCD (5), but found it
more difficult than those of analogous bis-CDs. The operating
conditions used for the latter tended here to result in lower yields.
Indeed, when carried out in t-butanol/water, the reaction stopped
at the dimer intermediate, which precipitated out of the medium;
in fact, one acetylenic group had quite failed to react. When the
reaction however was repeated in DMF, the final product (17) was
isolated in 68% yield. Because di-O-alkylation generated different
positional isomers, the configuration of our tris-CD is unknown.
On the other hand, when 6A,6D-dideoxy-6A,6D-diazido-bCD (18)
and mono-O-propargyl-6-TBDMS-bCD (2) were reacted in DMF,
the product was identifiable as 6A,6D-bis((4-(6¢-O-hepta-TBDMS-
bCD-2¢-yl)-1H-1,2,3-triazol-1-yl)-6A,6D-dideoxy-bCD (19).
Residual copper was removed from crude reaction mixtures
by filtration through a silica column and crystallisation or, more
Fig. 2 Observed water–proton relaxation rates of 0.53 mM Gd-1 (ꢀ),
0.37 mM Gd-2 (ꢀ) and 0.48 mM Gd-3 (★) solutions as a function of
dimer 4 concentration (20 MHz, 298 K, pH = 7).
372 | Org. Biomol. Chem., 2009, 7, 370–379
This journal is
The Royal Society of Chemistry 2009
©

View Online
Scheme 3 Synthesis of dimers 7–9 and 13–15. Reagents and conditions: a) CuSO4/ascorbic acid, t-BuOH/H2O (1:1), MW 85 ◦C, 150 Wmax, 40 min;
b) AcCl 2% in CH3OH/CH2Cl2 1:2.
formed, in the case of ditopic and tritopic complexes three or
four dimer molecules might be bridged through two or three Gd
complexes (n = 2/3 and n = 3/4 respectively) (Fig. 7).
From inspection of Table 1 we can clearly infer a relationship
between binding parameters and structural features of the inter-
acting compounds. For example, by comparing the interactions
between dimer 4 and Gd-1, Gd-2, and Gd-3 respectively (Fig. 2),
we see that an increase in the number of hydrophobic cyclohexyl
substituents determines an enhancement in the relaxivity of the
adducts; on the other hand the affinity constant increases only
when going from one to two substituents, whereas the addition
of a third one does not influence it further. A closer investigation
of parameters governing the relaxometric behaviour of the three
CD-dimer adducts has been performed by analysing the relative
NMRD (Nuclear Magnetic Resonance Dispersion) profiles.28 This
approach accurately determines the reorientational correlation
time (tR)29 of each adduct, which is strictly related to its molecular
Fig. 3 Observed water-proton relaxation rates of a 0.75 mM dimer 4
solution as a function of Gd-2 concentration ( 20 MHz, 298 K, pH = 7).
dimensions. Fig. 5 shows the NMRD profiles of the adducts of
dimer 4 with Gd-1, Gd-2 and Gd-3. Data were analysed using the
Solomon–Bloembergen–Morgan model,30 assuming (on the basis
explain this difference we hypothesized that, while in the case of
of what is observed with free complexes) one water molecule in the
a monotopic Gd-1 complex a simple 1:1 host–guest adduct was
This journal is The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 370–379 | 373
©

View Online
Scheme 4 Synthesis of trimers 17 and 19. Reagents and conditions: a) CuSO4/ascorbic acid, DMF, MW 85 ◦C, 150 Wmax, 40 min; b) AcCl 2% in
CH3OH/CH2Cl2 1:2.
inner coordination sphere for each Gd(III) complex (q = 1) and
fixing the exchange lifetime (tM) to 200 ns. Both the profile shapes
and tR values determined from the fitting of experimental results
(Table 1) may be explained by the formation of supramolecular
systems of increasing size as we go from Gd-1/dimer 4 to
Gd-2/dimer 4 and Gd-3/dimer 4 inclusion complexes.
For adducts of dimer 4 with Gd-2 and Gd-3, quantitative
analysis of the NMRD profiles was not satisfactory when the
simple inner-outer-sphere model was used; thus we assumed that
the water protons present in the chelate second-coordination
sphere also contributed to observed relaxivities. This contribution
may originate from clusters of water molecules entrapped within
the supramolecular adducts. Such a behaviour is commonly
encountered with adducts formed when a paramagnetic complex
is non-covalently bound at the surface of a protein28 as well as with
the supramolecular “host–guest” adduct formed by poly-bCD and
Gd(III) complexes.31
to be much shorter than tR (Table 1), suggesting that they
modulate the lifetimes of water molecules lying at the surface
of the paramagnetic complex. Moreover, the mere contribution
from the second coordination sphere does not suffice to explain
the high relaxivity values of Gd-2/dimer 4 and Gd-3/dimer 4
adducts compared to the Gd-1/dimer 4 adduct. This considerable
relaxivity enhancement gives strong support to the hypotesis
that ditopic and tritopic complexes may act as bridges linking
two or three dimer molecules, resulting in larger supramolecular
adducts endowed with longer reorientational correlation times and
consequently higher relaxivities (Fig. 7).
Some further considerations may be appropriate regarding
the association constants (Table 1) determined for the in-
teractions of the Gd-2 complex with the other dimers here
reported.
By comparing the binding properties of Gd-2 towards dimers
4 and 6, in which the two CD units are linked by the same
spacer which is however fastened to different positions on the
CD surfaces, we find that the association constant is one order of
magnitude larger when the two CD units are joined by their larger
faces.
From the fitting of experimental data we estimated that
5–6 second-sphere water molecules are present per Gd(III) ion (see
˚
Table 1), at an average distance of 4 A from each paramagnetic
center. This second-sphere contribution was analysed on the basis
of the Solomon–Bloembergen–Morgan model, suitably modified
by introducing a generic correlation time (tss) which dealt with
modulating the dipolar interactions (exchange and/or rotation)
of the second-coordination-sphere water molecules. The tss values
obtained from fitting the experimental NMRD profiles appear
In the case of asymmetric dimers 7, 8 and 9 interacting with
Gd-2, a markedly larger association constant was found when both
CDs were b. The binding affinity determined for this particular
system was one order of magnitude larger than those found for all
other systems.
374 | Org. Biomol. Chem., 2009, 7, 370–379
This journal is
The Royal Society of Chemistry 2009
©

View Online
Table 1 Binding parameters determined by the analysis of NMRD profiles
Gd-1
Gd-2
Gd-3
Dimer 4
Ka = 4.6 ¥ 103 130 M-1
Ka = 3.9 ¥ 104 207 M-1
Rb = 24.3 0.30 mM-1s-1
n = 0.66
Ka = 4.1 ¥ 104 250 M-1
Rb = 12.4 0.20 mM-1s-1
Rb = 27.2 0.25 mM-1s-1
n = 1
n = 0.7
NMRD: tR = 356 12 ps
NMRD: tR = 890 17 ps
NMRD: tR = 930 23 ps
nss = 0
nss = 6
1
nss = 6 0.8
—
tss = 60 ps
tss = 60 ps
Dimer 6
Dimer 7
Dimer 8
Dimer 9
Dimer 14
Trimer 17
Ka = 5.9 ¥ 103 85 M-1
Ka = 5.9 ¥ 103 115 M-1
Rb = 25.8 0.18 mM-1s-1
n = 0.66
—
Rb = 11.5 0.45 mM-1s-1
—
—
—
—
—
—
—
—
—
—
—
—
—
—
n = 1
—
Ka = 1.4 ¥ 104 97 M-1
Rb = 21.7 0.31 mM-1s-1
n = 0.5
—
—
Ka = 1.1 ¥ 104 112 M-1
Ka = 6.1 ¥ 105 223 M-1
Rb = 26.1 0.18 mM-1s-1
n = 0.66
Rb = 10.8 0.2 mM-1s-1
n= 1
Ka = 1.8 ¥ 103 88 M-1
Ka = 3.0 ¥ 103 103 M-1
Rb = 27.2 0.26 mM-1s-1
n = 0.5
Rb = 11.8 0.11 mM-1s-1
n = 1
—
Ka = 1.8 ¥ 104 90 M-1
Rb = 24.3 0.43 mM-1 s-1
n = 0.66
—
—
Ka = 1.4 ¥ 104 95 M-1
Ka = 5.3 ¥ 104 135 M-1
Rb = 27 0.18 mM-1 s-1
n = 0.7
Ka = 6.1 ¥ 103 112 M-1
Rb = 12.8 0.14 mM-1s-1
Rb = 35.6 0.26 mM-1 s-1
n = 1
n = 1
NMRD: tR = 390 88 ps
NMRD: tR = 1 0.23 ns
NMRD: tR = 2 0.34 ns
nss = 0
nss = 5
1
nss = 9 1.5
—
tss = 60 ps
tss = 60 ps
Trimer 19
—
—
—
Ka = 1.9 ¥ 104 113 M-1
Rb = 24 0.9 mM-1s-1
n = 1
—
—
—
Ka= Association constant, Rb= Adduct relaxivity, n = Number of binding sites on the dimer determined by the relaxometric PRE method at 298 K and
neutral pH; tR = Reorientational correlation time; nss = Number of second-sphere water molecules; tss = Second-sphere correlation time.
Moreover, when we compare the binding properties of Gd-2
toward a bis-CD joined by unlike faces (dimer 8) with those shown
toward a bis-CD joined by like faces (dimer 14), we see that the
affinity is one order of magnitude lower in the latter case. These
differences in the affinity constant are certainly due to the different
dimensions and relative orientations of the two CD cages. For this
type of guest molecule, in which the interacting unit is cyclohexane,
the best cavity size is offered by bCD and the best orientation is
presented by the dimer in which the CD units are joined by unlike
faces.
Finally we investigated the binding properties of the three
complexes reported in Fig. 5 with CD trimers 17 and 19. The
strength of interaction seems to remain in the same order of
magnitude as found in the case of dimeric systems, but with Gd-
3/trimer complexes relaxivity values turned out to be even higher.
The NMRD profiles of Gd-1, Gd-2 and Gd-3 adducts with trimer
17 are reported in Fig. 6. Their higher relaxivity values (compared
to those of dimeric adducts) are supported by longer tR values
resulting from the larger dimensions of these supramolecular
systems.
bearing either reactive function. These versatile derivatives were
subjected to a MW-promoted Huisgen cycloaddition that led to
the formation of bis- and tris-CDs. Access to a library of CD
dimers and trimers featuring different linkage positions on CD
surfaces and cavities of different dimensions should enable us
to better tune the interaction between the guest molecule (drug,
diagnostic marker, etc.) and the CD carrier. Our study with medical
diagnostic markers (MRI contrast agents) as guest molecules bore
out this concept. Moreover, the stability constants and relaxivities
found for the adducts of Gd(III) chelates with our CD multimers,
especially the trimers, indicate that these supramolecular systems
are promising candidates for MRI applications.
Experimental
General
Materials and methods. Reactions were monitored by TLC
on Merck 60 F254 (0.25 mm) plates, which were visualized by UV
inspection and/or by heating after a spray with 5% H2SO4 in
ethanol. Merck silica gel was used for column chromatography
(CC). IR spectra were recorded with a Shimadzu FT-IR 8001
spectrophotometer. Unless stated otherwise, NMR spectra were
recorded with a Bruker 300 Av◦ance (300 MHz and 75 MHz for
1H and 13C, respectively) at 25 C; chemical shifts are calibrated
to the residual proton and carbon resonances of the solvent,
CDCl3 (dH = 7.26, dC = 77.0). Chemical shifts (d) are given in
ppm, coupling constants (J) in Hz. MALDI-TOF MS spectra were
Conclusions
A new approach to the preparation of CD dimers and trimers
has been introduced. It employs CDs that are regioselectively
monosubstituted with acetylenic and azido groups, or, alterna-
tively, the same in association with disubstituted aromatic spacers
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 370–379 | 375
©

View Online
Fig. 6 1/T1 NMRD profiles of Gd-1/trimer 17 (ꢀ), Gd-2/trimer 17 (●)
and Gd-3/trimer 17 (ꢀ) adducts, obtained at pH = 7 and 298 K. Solid
curves through the data points were calculated with the parameters
reported in Table 1. The fitting procedure was performed by assuming:
q (number of coordinated water molecules) = 1; r (distance between
the Gd(III) ion and the protons of the coordinated water molecule) =
˚
3.1 A; a (distance between the Gd(III) ion and the outer-sphere water
˚
proton nuclei = 3.8 A; D (solute-solvent diffusion coefficient) = 2.247 ¥
10-5 cm2 s-1; tM (exchange lifetime of the inner-sphere water molecule) was
fixed to the value previously obtained for the free complexes (200 ns).
Fig. 4 Scatchard plots of data obtained from PRE titrations of: A)
0.75 mM dimer 4 solution with Gd-2 and B) 0.375 mM dimer 4 solution
with Gd-1. The graphs clearly point to n = 0.7 and to n = 1 for Gd-2 and
Gd-1 respectively.
Fig. 7 Hypothetical interaction mode in Gd-1/, Gd-2/ and Gd-3/dimer
adducts.
recorded on
a Bruker Reflex III spectrometer, ESI-mass
spectra on a Waters Micromass ZQ equipped with ESI source.
MW-promoted reactions were carried out in a professional oven
from Microsynth–Milestone (Italy). Commercially available
reagents and solvents were used without further purification
unless otherwise noted. Native CDs were kindly provided by
Wacker Chemie. Gd1–3 (Fig. 1) were kindly provided by Bracco
Imaging Spa (Italy).
Synthesis
2I-O-Propargyl-6I–VIIO(t-butyldimethylsilyl)-bCD (2), 61-azido-61-
deoxy-bCD (5) and 3I-deoxy-3I-altro-azido-a, -b and -gCD (10,
11, 12) and biphenyl-4,4¢-disulfonyl-A,D-capped bCD were pre-
pared following published procedures.20,22,32 O-propargyl deriva-
tives of a- and gCD (1, 3) had not so far been described in the
literature.
Fig. 5 1/T1 NMRD profiles of Gd-1/dimer 4 (ꢀ), Gd-2/dimer 4 (ꢀ)
and Gd-3/dimer 4 (★) adducts, as determined at pH = 7 and 298 K. Solid
curves through data points were calculated with parameters reported in
Table 1. The fitting procedure was been performed by assuming: q (number
of coordinated water molecules) = 1; r (distance between the Gd(III) ion
2I-Propargyl-6I,VII-O-TBDMS-aCD
(1). 6I–VII-O-TBDMS-
aCD (3 g, 1.8 mmol) and LiH (43 mg, 5.4 mmol) were dissolved
in anhydrous THF (25 mL). The solution was heated 2 h under
reflux. After it had cooled down to room temperature, 80%
propargyl bromide in toluene (292 mL, 2 mmol) dissolved in
2 mL of THF was added dropwise and the mixture was stirred
overnight at room temperature. The reacted mixture was diluted
with EtOAc, washed with 1M H2SO4 and brine, and finally dried
˚
and the protons of the coordinated water molecule) = 3.1 A; a (distance
between the Gd(III) ion and the outer-sphere water proton nuclei =
-5
3.8 A; D (solute-solvent diffusion coefficient) = 2.247 ¥ 10 cm2 s-1; tM
˚
(exchange lifetime of the inner-sphere water molecule) was fixed to the
value previously obtained for the free complexes (200 ns).
376 | Org. Biomol. Chem., 2009, 7, 370–379
This journal is
The Royal Society of Chemistry 2009
©

View Online
(Na2SO4). The crude residue, purified by CC (CHCl3–acetone)
yielded 1.35 g of derivative 1 (0.8 mmol, yield 44%).
suspended in 5 mL of a t-BuOH/H2O 1:1 mixture. 0.6 mmol of
CuSO4·5H2O and 1.2 mmol of L-ascorbic acid were added. The
mixture was irradiated with MW at constant temperature (90 ◦C,
max power 150 W) and the reaction monitored by TLC until
complete conversion of the starting material was observed. Water
(30 mL) was then added; the precipitate was filtered off and washed
with a cold 40 mM solution of diethylenetriamine-pentaacetic acid
sodium salt (20 mL) to remove copper, and finally with water.
1 is a white powder; Rf = 0.43 (CHCl3/CH3OH 4:1);
n
max(KBr)/cm-1 3420, 2930, 1473, 1361, 1084, 1041 and 835;
dH(300 MHz; CDCl3) 4.5 (br s, 6H, 1-H), 4.04 (br d, J 9.4, 1H,
1¢-H), 3.98 (br d, J 9.4, 1H, 1¢-H), 4.0–3.49 (m, 36H), 2.43 (br. s,
1H, 3¢-H), 1.08 (s, 54H, t-but) and 0.05 (s, 36H, Si-CH3); m/z
(ESI-MS) calcd. for [M + H]+ 1697.4, found 1697.3.
2I-Propargyl-6I,VII-O-TBDMS-cCD
(3). 6I–VII-O-TBDMS-
General procedure for deprotection
gCD (3 g, 1.35 mmol) and LiH (30 mg, 3.75 mmol) were dissolved
in anhydrous THF (25 mL) and the solution was heated 2 h
under reflux. After it had cooled down to room temperature, 80%
propargyl bromide in toluene (221 mL, 1.48 mmol), diluted in
2 mL of THF, was added dropwise and the mixture was stirred
overnight at room temperature. The reacted mixture was diluted
with EtOAc, washed with 1M H2SO4 and brine, and finally dried
(Na2SO4). The crude residue, purified by CC (CHCl3-Acetone)
yielded 1.57 g of derivative 3 (0.702 mmol, yield 52%).
The crude product (1 mmol) was dissolved in CH2Cl2 (20 mL), a
2% solution of AcCl in MeOH (10 mL) was added and the mixture
was stirred overnight at room temperature. Ether (50 mL) was then
added; the precipitate was filtered off, washed with ether (40 mL)
and dried under vacuum.
1,3-Bis((4-(bCD-2¢-yl-methyl)-1H-1,2,3-triazol-1-yl)methyl)-
benzene (4). The reaction was carried out with 2I-O-propargyl-
6I–VII-O-TBDMS-bCD (2) (500 mg, 0.25 mmol) and 1,3-bis-
(azidomethyl) benzene (AMB, 23 mg, 0.12 mmol). The reacted
mixture was diluted with EtOAc, washed with H2O and brine
and finally dried (Na2SO4). The crude product, when purified
by CC (CHCl3/CH3OH = 19:1, 9:1, 4:1), yielded 370 mg of
6I–VIITBDMS b-b dimer (0.09 mmol, yield 75%) as a white powder.
Rf = 0.43 (CHCl3/CH3OH 4:1) and 0.56 (EtOAc/MeOH/H2O =
40:7:5); nmax(KBr)/cm-1 3422, 1473, 1389, 1254, 1086, 1040 and
835; dH(300 MHz; CDCl3) 7.9 (s, 2H, 5-H triazole), 7.4 (m, 1H,
5-H), 7.35–7.3 (m, 3H, 2-H, 4-H, 6-H), 5.53 (s, 4H, -N-CH2-Ph),
4.9 (m, 18H, 1¢-H, overlapped 4 H, triazole-CH2-O), 4.09–3.9
(m, 28H), 3.7–3.4 (m, 56H), 0.88 (s, 126H, t-but), 0.05 (s, 84H,
Si-CH3); dC(75 MHz; CDCl3) 145.1 (C-4 triazole), 135.8 (C-1,3),
130.1 (C-5), 128.8, 128.2, 128.0 (C-2,C-4,C-6), 124.2 (C-5 triazole),
102.7 (C-1¢), 82.1 (C-4¢), 73.8, 73.4, 73.2 (C-2¢,C-3¢,C-5¢), 65.3 (C
triazole -CH2-O), 62.1 (C-6¢), 54.7 (C N-CH2-Ph), 26.4 (C-Me3),
18.7 (C-Me3), -4.0, -4.2 (Si-Me2); m/z (MALDI-TOF MS) calcd.
for [M + Na]+ 4153.053, found 4152.388. 370 mg of 6I–VIITBDMS
b-b dimer (0.09 mmol) were deprotected to obtain 180 mg of b-b
dimer 4 (180 mg, 0.07 mmol, yield 80%).
4 is a white powder. Rf = 0.51 (CH3CN/H2O 2:1); nmax(KBr)/
cm-1 3435, 1490, 1399, 1257, 1090, 1050 and 852; dH(300 MHz;
D2O) 7.9 (s, 2H, 5-H triazole), 7.3 (t, J 7.9, 1H, 5-H), 7.2 (d,
2H, J 7.9, 4,6-H), 7.1 (s, 1H, 2-H), 5.53 (s, 4H, -N-CH2-Ph), 4.9
(m, 14H, 1¢-H), 4,7 (m, 4H, triazole-CH2-O), 3.9–3.7 (m, 56H),
3.6–3.4 (m, 24H), 3.4 (dd, J 9.9 and 3.6, 2H, 2¢-H), 3.2 (t, J 9.9,
2H, 4¢-H); dC(75 MHz; D2O) 144.2 (C-4 triazole); 136.1 (C-1,C-3),
130.4, 128.7, 128.6, 127.9 (C-2,C-4,C-5,C-6), 126.0 (C-5 triazole),
102.2 (C-1¢), 81.8 (C-4¢), 73.8, 73.4, 73.2 (C-2¢,C-3¢,C-5¢), 65.3 (C
triazole-CH2-O), 60.6 (C-6¢), 54.1 (N-CH2-Ph); m/z (MALDI-
TOF MS) calcd. for [M + Na]+ 2555.842, found 2556.074.
3 is a white powder; Rf = 0.25 (CHCl3/CH3OH 4:1);
n
max(KBr)/cm-1 3420, 2930, 1471, 1362, 1086, 1035 and 835;
dH(300 MHz; CDCl3) 5.1–4.9 (m, 8H, 1-H), 4.78 (br d, J 8.1,
2H, 1¢-H), 4.11–3.40 (m, 48H), 2.1 (br. s, 1H, 3¢-H), 0.9 (s, 72H,
t-but) and 0.05 (s, 48H, Si-CH3); m/z (MALDI-TOF MS) calcd.
for [M + Na]+ 2270.120, found 2270.148.
Di-O-propargyl-6I,VII-O-TBDMS-bCD (16). 6I–VII-O-TBDMS-
bCD (3 g, 1.55 mmol) and LiH (37 mg, 4.6 mmol) were dissolved in
anhydrous THF (12 mL). The solution was heated 2 h under reflux.
After it had cooled down to room temperature, 80% propargyl
bromide in toluene (690 mL, 4.65 mmol), diluted in 2 mL of THF,
was added dropwise and the mixture was stirred overnight at room
temperature. The reacted mixture was diluted with EtOAc, washed
with 1M H2SO4 and brine, and finally dried (Na2SO4). The crude
residue, purified by CC (CHCl3-CH3OH 19:1, 9:1, 4:1) yielded
1.24 g of di-O-alkylated-bCD (0.62 mmol, yield 40%).
16 is a white powder; Rf = 0.26 (CHCl3/CH3OH 4:1);
n
max(KBr)/cm-1 3420, 3325, 1473, 1254, 1086, 1040 and 835;
dH(300 MHz; CDCl3) 4.9 (br s, 7H, 1-H), 4.5 (br q, J 16.8, 4H, 1¢-
H), 4.1–3.9 (m, 14H), 3.8–3.5 (m, 28H), 2.4 (br. t, J 2.3, 2H, 3¢-H),
0.88 (s, 63H, t-but) and 0.05 (s, 42H, Si-CH3); m/z (MALDI-TOF
MS) calcd. for [M + Na]+ 2031.996, found 2032.322.
6A,6D-Diazido-6A,6D-dideoxy-bCD (18). The reaction was car-
ried out under magnetic stirring in a professional MW oven,
temperature being monitored with a fibre-optic thermometer.
500 mg (0.35 mmol) of biphenyl-4,4¢-disulfonyl-A,D-capped bCD
and 66 mg (1.05 mmol) of sodium azide were dissolved in 5 ml
of DMF. The mixture was irradiated with MW (120 W) at 85 ◦C
for 40 min. The solvent was then partially evaporated and the
product precipitated with 7 ml of water/acetone 1:10. 403 mg
of pure 6A,6D-diazido-6A,6D-dideoxy-bCD were recovered (yield
96%). Analytical data were in accordance with reported values.33
1,3-Bis((1-(6¢-deoxy-bCD-6¢-yl)-1H-1,2,3-triazol-4-yl)methoxy)-
benzene (6). The reaction was carried out with 61-azido-61-
deoxy-bCD (5) (400 mg, 0.34 mmol) and 1,3-bis(propargyloxy)-
benzene (POB) (23 mg, 0.12 mmol). Acetone (30 ml) was added
to the reacted mixture and the precipitate was filtered off. The
solid was recrystallised from water/acetone 1:2 and 215 mg of
pure b,b dimer 6 were recovered (0.084 mmol, yield 70%).
Analytical data were in accordance with reported values.34
General procedure for Cu-catalyzed Huisgen 1,3-dipolar
cycloadditions
In a 10 mL two-necked round-bottomed flask (equipped with
an optical-fiber thermometer for reactions under MW) 1 mmol
of the alkyl azide and 1 mmol of the acetylenic derivative were
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 370–379 | 377
©

View Online
6
is
a
white powder. Rf
=
0.2 (CH3CN/H2O 2:1);
Water-proton relaxivity measurements
n
max(KBr)/cm-1 3422, 1640, 1389, 1254, 1040 and 835;
Longitudinal water-proton relaxation rates were measured by a
Stelar Spinmaster spectrometer (Mede, PV- Italy) operating at
0.47 T, by means of the standard inversion-recovery technique
(16 experiments, 2 scans). Typical 90◦ pulse width was 7.5 ms and
the reproducibility of the T1 data was 0.5%. Temperature was
controlled ( 0.1 ◦C) with a Stelar VTC-91 air-flow heater equipped
with a copper-constantan thermocouple. The proton 1/T1 NMRD
profiles were measured over a continuum of magnetic field strength
ranging from 0.00024 to 0.47 T (corresponding to 0.01–20 MHz
proton Larmor Frequency) on a Stelar field-cycling relaxometer.
This works under complete computer control with an absolute
uncertainty in 1/T1 of 1%. Data points from 0.47 T (20 MHz)
to 1.7 T (70 MHz) were collected on a Stelar Spinmaster
spectrometer.
dH(300 MHz; D2O) 7.7 (s, 2 H, 5-H triazole), 7.2 (t, J 6.8, 1H,
5-H), 7.1 (s, 1 H, 2-H), 6.8 (d, J 6.8, 1H, 4-H), 6.6 (d, J 6.8, 1H,
6-H), 5.3–4.8 (m, 18H, 1¢-H overlapped 4 H, triazole-CH2-O),
4.0–3.0 (m, 80H), 3.0 (m, 2H), 2.9 (m, 2H); m/z (ESI-MS) calcd.
for [M + Na2]2+ = 1253,4, found 1253.7.
2-(O-Di-(1-(6¢-deoxy-bCD-6¢-yl)-1H-1,2,3-triazol-4-yl)methyl)-
bCD (17). The reaction was carried out in DMF with di-2I-O-
propargyl-6I–VII-O-TBDMS-bCD (2) (414 mg, 0.2 mmol) and 61-
azido-61-deoxy-bCD (5) (525 mg, 0.45 mmol). Acetone (30 ml)
was added to the reacted mixture, the precipitate was filtered
off, washed with a cold 40 mM solution of diethylenetriamine-
pentaacetic acid sodium salt (20 mL) to remove copper, and
finally with water. The 6I–VIITBDMS b-b trimer is a white powder
(618 mg, 0.14 mmol, yield 68%). nmax(KBr)/cm-1 3422, 1474,
1364, 1254, 1082, 1038 and 835; dH(300 MHz; DMSO-d6) 8.0
(s, 2H, 5-H triazole), 5.1–4.6 (m, 25H, 1-H overlapped 4 H,
triazole-CH2-O), 3.8–3,2 (m, overlapped with H2O), 0.88 (s,
63H, t-but), 0.04 (s, 42H, Si-CH3); m/z (MALDI-TOF MS)
calcd. for [M + Na]+ 4350.749, found 4350.952. Starting with
6I–VIITBDMS b-b trimer (618 mg, 0.143 mmol), we obtained trimer
17 (290 mg, 0.082 mmol, yield 58%) as a white powder. Rf = 0.06
(CH3CN/H2O 2:1); nmax(KBr)/cm-1 3422, 1465, 1380, 1270, 1095,
and 850; dH(300 MHz; D2O) 8.0 (s, 1H, 5-H triazole), 5.2–4.9
(m, 25H, 1¢-H overlapped 4 H triazole-CH2-O), 4.1 (br t, J 7.5,
2H), 3.9–3.2 (m, 120H), 3.1 (m, 2H), 2.8 (m, 2H); dC(75 MHz;
D2O) 144.3 (C-4 triazole) 127.3 (C-5 triazole), 102.3–101.8 (C-1¢),
81.7–81.5 (C-4¢), 73.5–73.0 (C-5¢), 72.4–71.8 (C-2¢,C-3¢), 69.8 (C
triazole-CH2-O), 61.3–60.5 (C-6¢); m/z (ESI-MS) m/z (MALDI-
TOF MS) calcd. for [M + Na]+ 3552.144, found 3552.523.
Acknowledgements
This work was supported by the University of Torino and MIUR
(FIRB 2003). It has been carried out under the auspices of the
COST Action D32 (WG 006).
References
1 L. Szente and J. Szejtli, Adv. Drug Deliv. Rev., 1999, 36, 17–28.
2 E. Schneiderman and A. M. Stalcup, J. Chromatogr. B, 2000, 745,
83–102.
3 G. Cravotto, A. Binello, E. Baranelli, P. Carraro and F. Trotta, Curr.
Nutr. Food Sci., 2006, 2, 343–350.
4 R. Breslow, Chem. Record, 2001, 1, 3–11.
5 D. Duchene, D. Wouessidjewe and G. Ponchel, J. Contr. Rel., 1999, 62,
263–268.
6 S. Aime, E. Gianolio, E. Terreno, I. Menegotto, C. Bracco and
G. Cravotto, Magn. Res. Chem., 2003, 7, 800–805; E. Battistini, E.
Gianolio, R. Gref, P. Couvreur, S. Fuzerova, M. Othman, S. Aime, B.
Badet and P. Durand, Chem-Eur. J., 2008, 14, 4551–4561.
7 T. Loftsson, M. E. Brewster and M. Masson, Am. J. Drug Deliv, 2004,
2, 261–275.
8 I. Tabushi, Y. Kuroda and K. Shimokawa, J. Am. Chem. Soc., 1979,
101, 1614–1615.
9 A. Harada, M. Furue and S.-I. Nozakura, Polym. J., 1980, 12, 29–33.
10 Y. Liu, L. Li, H.-L. Zhang and Y. Song, J. Org. Chem., 2003, 68, 527–
536; D.-Q. Yuan, J. Lu, M. Atsumi, J.-M. Yan, M. Kai and K. Fujita,
Org. Biomol. Chem., 2007, 5, 2932–2939; T. Lecourt, J.-M. Mallet and
P. Sinay, Eur. J. Org. Chem., 2003, 4553–4560.
6A,6D-Bis((4-(6¢-bCD-2¢-yl)-1H-1,2,3-triazol-1-yl)-6A,6D-dideoxy-
bCD (18). The reaction was carried out with 6A,6D-dideoxy-
6A,6D-diazido-bCD (12) (80 mg, 0.067 mmol) and 2I-O-propargyl-
6I–VII-O-TBDMS-bCD (2) (264 mg, 0.135 mmol). The solvent was
then partially evaporated and the reacted mixture was diluted
with CHCl3, washed with H2O (3 ¥ 20 ml), dried (Na2SO4)
and evaporated to dryness. The crude product, purified by CC
(CHCl3/CH3OH =, 9:1, 4:1), yielded the desired 6I–VIITBDMS
b-b trimer (252 mg, 0.049 mmol, 73%) as a white powder.
11 H. Yamamura, K. Satake, R. Hirasawa, S. Yamada and M. Kawai,
J. Incl. Phenom., 2007, 57, 261–264; D. T. Pham, P. Clements, C. J.
Easton, J. Papageorgiou, B. L. May and S. F. Lincoln, New J. Chem.,
2008, 32, 712–718.
n
max(KBr)/cm-1 3422, 1475, 1389, 1263, 1090, 1038 and 835;
12 F. Venema, H. F. M. Nelissen, P. Berthault and N. Birlirakis, Chem.
Eur. J., 1998, 4, 2237–2250; S. H. Chiu, D. C. Myles, R. L. Garrell
and J. F. Stoddart, J. Org. Chem., 2000, 65, 2792–2796; Y. Ishimaru, T.
Masuda and T. Iida, Tetrahedron Lett., 1997, 38, 3743–3744.
13 M. Luo, W. Chen, D. Yuan and R. Xie, Synth. Commun., 1998, 28,
3845–3848; J. A. Faiz, N. Spencer and Z. Pikramenou, Org. Biomol.
Chem., 2005, 3, 4239–4245.
14 G. Cravotto, C. Bicchi, S. Tagliapietra, L. Costa, S. Di Carlo and C.
Nervi, Chirality, 2004, 16, 526–533.
15 S. Aime, E. Gianolio, B. Robaldo, A. Barge and G. Cravotto, Org.
Biomol. Chem., 2006, 4, 1124–1130.
16 G. Cravotto, F. Mendicuti, K. Martina, S. Tagliapietra, B. Robaldo,
and A. Barge, Synlett 2008 in press.
17 P. Cintas, K. Martina, B. Robaldo, D. Garella, L. Boffa and G.
Cravotto, Collect. Czech. Chem. Commun., 2007, 72, 1014–1024.
18 R. Huisgen in: 1,3-Dipolar Cycloaddition Chemistry (Ed.: A. Padwa),
Vol. 1, pp. 1. Wiley, New York, 1984.
dH(300 MHz; DMSO-d6) 7.99 (1 H, s, 5-H triazole), 6.0–5.5 (OH,
m); 5.0–4.5 (25 H, m, 1-H overlapped 4 H, triazole-CH2-O),
4.20–3.15 (126 H, m), 0.8 (s, 126 H, t-but), 0.01 (s, 84 H,
Si-CH3); m/z (MALDI-TOF MS) calcd. for [M + Na]+ 5149.4,
found 5149.4. Starting from 6I–VIITBDMS b-b trimer (150 mg,
0.029 mmol), the mixture, stirred overnight at room temperature,
yielded tris-CD 18 (68 mg, 0.02 mmol, yield 68%) as a white
powder. nmax(KBr)/cm-1 3412, 1477, 1390, 1255, 1085, 1038
and 835; dH(300 MHz; D2O) 8.2 (s, 1H, 5-H triazole), 5.3–4.4
(m, 25H, 1-H overlapped 4 H triazole-CH2-O), 4.0–3.5 (126 H,
m), 2.8 (m, 2H); dC(75 MHz; D2O) 143.7, (C-4 triazole), 127.4
(C-5 triazole), 102.2–101.7 (C-1¢), 81.9–81.4 (C-4¢), 73.4 (C-5¢)
72.3–71.8 (C-2¢,C-3¢), 66.2 (C triazole-CH2-O), 60.6–59.3 (C6¢)
m/z (MALDI-TOF MS) calcd. for [M + Na]+ 3552.144 found
3552.417.
19 V. V. Rostovtsev, L. G. Green, V. V. Fokin and K. V. Sharpless, Angew.
Chem. Int. Ed., 2002, 41, 2596–2599.
378 | Org. Biomol. Chem., 2009, 7, 370–379
This journal is
The Royal Society of Chemistry 2009
©

View Online
20 F. Trotta, K. Martina, B. Robaldo, A. Barge and G. Cravotto,
J. Inclusion Phenom, 2007, 57, 3–7.
21 S. Aime, E. Gianolio, A. Barge, S. Tagliapietra and G. Cravotto, J. Inorg.
Biochem., 2006, 100, 931–938.
22 K. Martina, F. Trotta, B. Robaldo, N. Belliardi, L. Jicsinszky and G.
Cravotto, Tetrahedron Lett., 2007, 52, 9185–9189.
23 A. Dietrich, B. Maas, W. Messer, G. Bruche, V. Karl, A. Kauzinger and
A. Mosandl, J. High Res Chromatogr., 1992, 15, 590–593.
24 A. Binello, B. Robaldo, A. Barge, R. Cavalli and G. Cravotto, J. Appl.
Polym. Sci., 2008, 107, 2549–2557.
27 G. Scatchard, Ann. N.Y. Acad. Sci., 1949, 51.
28 L. Banci, I. Bertini, C. Luchinat, Nuclear and Electronic Relax-
ationVCH, Weinheim, 1991, pp 91–122.
29 S. Aime, M. Botta, M. Fasano, E. Terreno, The Chemistry of Contrast
Agents in Medical Magnetic Resonance Imaging, John Wiley & amp;
Sons LTD, (Eds.: A. E. Merbach, E. Toth), 2001, chap. 5, p. 193.
30 N. Bloembergen and L. O. Morgan, J. Chem. Phy.s, 1961, 34, 842–
850.
31 S. Aime, M. Botta, F. Fedeli, E. Gianolio, E. Terreno and P. L. Anelli,
Chem-Eur. J., 2001, 7(24), 5261–5269.
25 S. Aime, M. Botta, M. Fasano, and E. Terreno, in The Chemistry of
Contrast Agents in Medical Magnetic Resonance Imaging (Eds. E. Toth,
A. E. Merbach), Wiley, Chichester, 2001, p. 193.
26 B. G. Jenkins, E. Armstrong and R. B. Lauffer, Magn Reson Med,
1991, 17, 164–178; S. Aime, M. Chiaussa, G. Digilio, E. Gianolio and
E. Terreno, J. Biol. Inorg. Chem., 1999, 4, 766–774.
32 I. Tabushi, K. Yamamura and T. Nabeshima, J. Am. Chem. Soc., 1984,
106, 5267–5270.
33 K. Fujita, H. Yamamura and T. Imoto, J. Org. Chem., 1985, 50, 4393–
4395; E. S. M. Cotner and P. Smith, J. Org. Chem, 1998, 63, 1737–1739.
34 M. Mourer, F. Hapiot, E. Monflier and S. Menuel, Tetrahedron, 2008,
64, 7159–7163.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 370–379 | 379
©