NMR relaxometric study of new Gd(III) macrocyclic complexes and their interaction with human serum albumin.

Article abstract of DOI:10.1039/b313677a

Five novel Gd(iii) complexes based on the structure of the heptadentate macrocyclic 1,4,7,10-tetraazacyclododecane-1,4,7-triacetic acid (DO3A) ligand have been synthesized and their (1)H and (17)O NMR relaxometric properties investigated in detail. The co

Full text of DOI:10.1039/b313677a

View Article Online / Journal Homepage / Table of Contents for this issue
NMR relaxometric study of new GdIII macrocyclic complexes and
their interaction with human serum albumin
Mauro Botta,*a Silvio Quici,*b Gianluca Pozzi,b Giovanni Marzanni,c Roberto Pagliarin,c
Serena Barrad and Simonetta Geninatti Crichd
a Dipartimento di Scienze e Tecnologie Avanzate, Corso Borsalino 54, 15100 Alessandria, Italy.
E-mail: mauro.botta@unito.it; Fax: (+39) 0131-287416; Tel: (+39) 0131-287431
b CNR-Istituto di Scienze e Tecnologie Molecolari (ISTM), Via C. Golgi 19, I – 20133 Milano,
Italy
c Dipartimento di Chimica Organica e Industriale dell’Università, Via C. Golgi 19,
I – 20133 Milano, Italy
d Dipartimento di Chimica IFM, Via Pietro Giuria 7, 10125 Torino, Italy
Received 29th October 2003, Accepted 15th December 2003
First published as an Advance Article on the web 21st January 2004
Five novel Gd() complexes based on the structure of the heptadentate macrocyclic 1,4,7,10-tetraazacyclododecane-
1,4,7-triacetic acid (DO3A) ligand have been synthesized and their 1H and 17O NMR relaxometric properties
investigated in detail. The complexes have been functionalised on the secondary nitrogen atom of the macrocyclic
ring with different pendant groups for promoting their ability to interact non-covalently with human serum albumin
(HSA). The analysis of the proton relaxivity, measured as a function of pH and magnetic field strength, have revealed
that the three complexes bearing a poly(ethylene glycol) (PEG) chain possess a single coordinated water molecule,
whereas the complexes functionalised with 1-[3-(2-hydroxyphenyl)]-propyl and 1-[3-(2-carboxyphenyloxy)]-propyl
pendant groups have two inner sphere water molecules. The water exchange rates, measured by variable temperature
17O NMR, cover a broad range of values (from 18 to 770 ns) as a function of their charge, the chemical nature of the
substituent and its ability to organize a second sphere of hydration near the water(s) binding site. All the complexes
have shown some degree of interaction with HSA, with a stronger binding affinity measured for those bearing an
aromatic moiety on the pendant group. However, upon binding the expected relaxation enhancement has not been
observed and this has been explained with the displacement of the coordinated water molecules by the protein and
formation of ternary adducts.
cyclic ligand is based on the 1,4,7,10-tetraazacyclododecane-
1,4,7-triacetic acid (DO3A) scaffold and it therefore contains
Introduction
The polyaminopolycarboxylate paramagnetic complexes have
been intensively investigated as possible contrast agents (CA)
for magnetic resonance imaging (MRI).1–3 These substances are
able to markedly increase the water proton relaxation rate in
the tissues where they distribute thus enhancing the contrast
between healthy and pathological districts. The GdIII chelates
are particularly suitable for this purpose because of the high
magnetic moment and long electronic relaxation times of the
metal ion, which results in a strong dipolar interaction with the
water proton nuclei. The efficiency of this interaction depends
on several structural and dynamic parameters of the metal
complex, including the number q of inner sphere water mole-
cules, their rate of exchange (k = 1/τM), their distance r from the
metal centre, the rotational mobility of the whole complex
described in terms of the correlation time τR, the electronic
relaxation times of the paramagnetic metal ion (T1,2e), the
relative diffusion coefficients of solute and solvent (D), and the
distance of minimum approach between the water molecules
in the outer coordination sphere of the complex and the
paramagnetic centre (a).4 The relaxation efficacy of a given
paramagnetic complex is described by its relaxivity, the
increment of the water protons’ relaxation rate induced by one
millimolar concentration of the metal chelate. A deep under-
standing of the relationship between the solution structure of
the complexes and their relaxation parameters is crucial in
order to enhance the contrast efficiency of these agents.5 We
have now synthesized and characterized, by relaxometric
techniques, a series of GdIII complexes with new tetraaza
macrocyclic heptadentate ligands functionalised with different
pendant groups at the secondary nitrogen atom. The macro-
seven potential donor atoms: the four nitrogen atoms of the
macrocycle and three oxygen atoms of the carboxylate groups.6
Since gadolinium ions are commonly nine coordinate, two
water molecules are expected to complete the inner co-
ordination sphere of the paramagnetic metal cation, as found
for the corresponding lanthanide complexes of DO3A.7,8 At
the imaging fields (20–60 MHz proton Larmor frequency) the
relaxivity of the low molecular weight GdIII chelates are largely
determined by the term qτR/r6 and by τM, which has an optimal
value of about 30 ns at 298 K. In the case of GdDO3A a τM of
about 90 ns was calculated by 17O NMR data, short enough so
as not to limit the observed relaxivity.8 The present work is
aimed at studying the effects of the introduction of pendant
groups potentially able to bind human serum albumin (HSA)
through non-covalent interactions on the relaxometric proper-
ties of the complexes. Pendant groups consisting of poly-
ethyleneglycol (PEG) chains were chosen for the following
reasons: i) to increase the solubility of the complexes in water;
ii) to exploit the increase of the molecular weight for enhancing
the inner sphere relaxivity, which strongly depends on the value
of τR; iii) to check the possibility that the PEG chain favours the
organization of a well defined second sphere of hydration, thus
contributing to further increasing the relaxivity; iv) to assess, in
the case the previous hypothesis is verified, if a modulation
of the value of τM occurs; v) to investigate the extent of the
binding with HSA and the concomitant gain in relaxivity.
For a better evaluation of the role of the PEG pendant
groups, relaxometric properties of similar complexes char-
acterized by the presence of propylene chains terminated with
aromatic substituents have been also studied.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 7 0 – 5 7 7
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
570

View Article Online
Scheme 1
products. Decomplexation of the sodium complex of 8 was
Results and discussion
Synthesis of the ligands
achieved by continuous extraction of the complex, after ClϪ for
Ϫ
ClO4 counterion exchange in H2O/CH3OH = 9 : 1 (v/v) with
N-alkylation of 1,4,7,10-tetraazacyclododecane-1,4,7-triacetic
acid tris-(1,1-dimethylethyl) ester 69 with suitable reagents,
followed by deprotection of the tert-butyl ester functions,
allows a convenient access to the new ligands 1–5 (Scheme 1).
However, alkylation conditions and purification of the inter-
mediate triester had to be optimized for any single ligand. Thus,
ligands 1 and 3 were prepared by alkylation of 6 with meth-
anesulfonates 7 and 11, respectively, carried out in CH3CN and
solid Na2CO3 as a base. Subsequent treatment of the crude
triesters 8 and 12, dissolved in CH2Cl2, with 30% aqueous
NaClO4 afforded the corresponding sodium complexes which
were isolated by column chromatography. The preparation of
the complexes is necessary in order to simplify the chromato-
graphic purification and allow higher yields of isolated
n-pentane10 to give the free ligand 8 in 58% yield. This last
compound was hydrolyzed with CF3COOH at room temper-
ature to afford 1 quantitatively as a white solid. Compound 12
was analogously isolated and characterized as NaClO4 complex
in 85% yield after purification by column chromatography. This
product was first decomplexed and then hydrolyzed with
CF3COOH, as described in the case of 8, to afford ligand 3 in
quantitative yield. On the other hand, alkylation of 6 with
PEG-350 monomethylether methanesulfonate 9 was carried out
in DMF at 110 ЊC for 24 h to give triester 10 in 70% yield as an
oil. Hydrolysis of 10 by CF3COOH in CH2Cl2 at room tem-
perature and crystallization of the residue with THF/n-heptane
yielded 2 as a light white solid. Alkylation of 6 with 2-(3-
bromopropyl)phenyl benzoate 13 was carried out in CH3CN at
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 7 0 – 5 7 7
571

View Article Online
reflux and solid Na2CO3 as a base to give triester 14 in 58%
yield, whereas triester 16, obtained by alkylation of 6 with
methyl 2-(3-bromopropoxy)-benzoate 15, was conveniently
isolated and characterized after treatement with NaClO4.
Hydrolysis of 14 and 16 with NaOH in 96% EtOH at reflux
directly afforded ligands 4 and 5 in quantitative yields. It is
also worth noting that the alkylating agents employed here
were prepared in high yields from commercially available or
easily obtained precursors, which is particularly helpful in
view of possible large-scale preparation of the new ligands.
Indeed, methanesulfonates 7, 9 and 11 were prepared from
triethyleneglycol monobutylether, PEG-350 monomethylether
(commercially available reagents) and triethyleneglycol mono-
2-methoxyphenyl ether 17,11 respectively, following standard
procedures as described in the Experimental section.
2-(3-Bromopropyl)phenyl benzoate 13 was prepared from
3-(2-hydroxyphenyl)-1-propanol 1912 through benzoylation
of the phenolic hydroxyl group followed by bromination of
the protected phenol 20 with N-bromosuccinimide (NBS) and
triphenylphosphine (Scheme 2). Methyl 2-(3-bromopropoxy)-
benzoate 15 was prepared by alkylation of commercially
available methyl salicylate with an excess of 1,3-dibromo-
propane in refluxing acetonitrile and solid Na2CO3 as a base.
Fig. 1 Plot of 1H relaxivity (r1p; mMϪ1 sϪ1) at 20 MHz and 25 ЊC versus
molecular weight for several Gd() complexes with q = 2.14 The linear
relationship represents an indication that the compounds possess the
same number of inner sphere water molecules.
explained by the fact that the substituent is involved in the
coordination rather than by the predominance in solution
of octacoordinated monoaquo species. The pH dependence of
the relaxivity, measured at 25 ЊC and 20 MHz, confirms the
different behaviour of the complexes Gdؒ1–3 and Gdؒ4–5. In
the former, r1p is nearly constant in the range 4–12, whereas for
the latter a significant decrease is observed from 8–10. In q = 1
complexes the coordinated water molecule cannot be displaced
by the carbonate anions originating from dissolved CO2 in the
aerated solutions and the relaxivity does not change with
pH.7,15 In q = 2 complexes the chelating carbonate anions can
remove the two inner sphere water molecules and the relaxivity
decreases to a value close to that typical of pure outer sphere
complexes (ca. 3.0 mMϪ1 Ϫ1) (Fig. 2).7,15 The small inflection
s
point at pH close to 9 observed in the pH profile in the case of
the complexes Gdؒ1–3 could result either from the presence in
solution of a low concentration of the q = 2 species or from
competition of carbonate with PEG for the coordination site(s)
on the metal ion.
Scheme 2
1H NMR relaxometric studies
The parent complex GdDO3A has been shown to be present in
aqueous solution as a mixture of two coordination isomers
differing in the hydration number, i.e. q = 2 (predominant
species) and q = 1 (largely minor species).8 The relaxivity at
25 ЊC and 20 MHz is 6.0 mMϪ1 sϪ1, about 28% higher than for
the monoaquo GdDOTA complex (4.7 mMϪ1 Ϫ1).1–3 Under
s
identical experimental conditions the relaxivities of the GdIII
complexes with 1–3 are 3.9, 4.6 and 4.1 mMϪ1 sϪ1, respectively.
This suggests a lower hydration for Gdؒ1–3 complexes (q < 2)
that may arise from partial displacement of the coordinated
water molecules by oxygen atom(s) of the PEG chain. A similar
effect has recently been observed in the case of PEG-functional-
ised complexes based on the hydroxypyridonate (HOPO)
binding unit.13 Although we cannot exclude the occurrence
of an effective fractional q value arising from a mixture of
coordination isomers with q = 0 (low concentration) and q = 1
(predominant), we assume for simplicity that the complexes are
monoaquo species. Support for this hypothesis has been gained
from the pH-dependent behaviour of the relaxivity (vide infra).
On the other hand, the r1p values of Gdؒ4–5 (25 ЊC, 20 MHz)
are 7.4 and 7.5 mMϪ1 sϪ1, respectively and thus fully consistent
with the relaxivity expected for diaquo complexes. This is
clearly demonstrated by the linear plot of the relaxivity versus
the molecular weight for a series of GdIII complexes with q = 2
(Fig. 1).14
Fig. 2 Plot of 1H relaxivity (r1p; mMϪ1 sϪ1) at 20 MHz and 25 ЊC versus
pH for Gdؒ1 (filled circles) and Gdؒ4(open circles).
17O NMR relaxometric studies
The analysis of the temperature dependence of the transverse
relaxation rate of the 17O water nuclei is a well established
procedure for obtaining an accurate evaluation of the inner
sphere water molecules’ exchange dynamics.16 Moreover, in
favorable cases (intermediate or fast exchange conditions) the
electronic relaxation parameters can also be derived with good
accuracy. We have measured the 17O variable-temperature (VT)
profiles at 2.1 (or 9.4) T and pH = 7 for the complexes Gdؒ1–5
and analyzed the data in terms of the established theory (Swift–
Connick equations). The derived best fit parameters are listed
in Table 1 and representative profiles are shown in Fig. 3. The
mean residence lifetimes show a marked dependence on the
nature of the pendant group: their values (at 298 K) are within
the wide interval 18–770 ns. The parent GdDO3A complex
presents a τM value of ca. 90 ns, shorter by a factor of three with
respect to GdDOTA and GdDTPA and the exchange mech-
The linear dependence of r1p from the molecular weight
implies that all the complexes present the same hydration
number and that their relaxivity is predominantly determined
by the value of τR, which is strongly correlated with the molec-
ular dimension of the complexes. Thus, relatively low values
of r1p are only observed for those complexes featuring oxygen
atoms associated with the PEG chain and this is better
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 7 0 – 5 7 7
572

View Article Online
Table 1 Parameters obtained from least-squares fits of 17O NMR Data at 2.1 T and pH = 7
τM/ns
∆HM/kJ molϪ1
τV/ns
∆HV/kJ molϪ1
∆2/1019 sϪ2
Gdؒ1
Gdؒ2
Gdؒ3
Gdؒ4
Gdؒ5
180
216
500
770
18
44.0
49.0
50.1
56.0
40.8
19
20
18
10
12
8.5
5.0
8.5
8.0
7.0
13
6
9
12
13
Table 2 Best fit relaxation parameters determined by analysis of
NMRD profiles at 25 ЊC and pH = 7a
Gdؒ1
Gdؒ2
Gdؒ3
Gdؒ4
Gdؒ5
∆2/1019 sϪ2
τV/ps
τR/ps
12
18
90
3.10
6.0
18
86
9.5
18
85
13.0
13
120
3.00
12
11
116
r/Å
3.10
3.16
3.00
a The parameters a and D have been fixed in the fitting procedure to the
standard values of 3.8 Å and 2.24 × 10Ϫ5 cm2 sϪ1, respectively.
Fig. 3 Temperature dependence of the paramagnetic contribution
(R2p) to the transverse 17O water relaxation rate of Gdؒ3 (33 mM; open
circles) and Gdؒ5 (10 mM; filled circles) at 2.1 T and pH = 7.
anism has been found to be of a dissociative nature.7 The
exchange lifetimes for Gdؒ1–5 cover a broad range of values
which are a function of the N-substituent. In related macro-
cyclic octadentate complexes, the τM value has been found to be
modulated by the relative proportion of the isomeric species
present in their aqueous solutions,17 the overall electric charge,18
the nature of the counterion19 and the accessibility of the sol-
vent in the proximity of the gadolinium cation.20 In the present
case we have to consider: i) the partial water displacement by
the PEG group. The q = 1 complexes often show longer τM
values; ii) the ability of the PEG chain to promote the form-
ation of a second hydration shell next to the metal center.13 This
network of H-bonded solvent molecules may involve the bound
water molecule(s) and slow down either the departure of the
leaving and/or the approach of the incoming water molecules;
iii) the different basicity of the nitrogen atom of the macrocycle
bearing the pendant group. This effect has been recently shown
to modulate the water exchange rate in related GdDO3A deriv-
atives.21 Particularly striking is the difference in τM between
Gdؒ4 and Gdؒ5, which is difficult to rationalize based on the
available data. We may only make the hypothesis of a possible
involvement of the hydroxy group of the phenolic moiety in an
H-bonding interaction with the coordinated water molecules
that increases their exchange lifetimes by a factor of about 3–4.
On the other hand, the much shorter τM of Gdؒ5 could result
from the predominance in solution of a conformational isomer
endowed with fast water exchange rate and from the presence,
around neutral pH, of an overall negative charge arising from
the deprotonation of the carboxylic group of the aromatic
moiety.
Fig. 4 Nuclear magnetic relaxation dispersion (NMRD) profiles for
Gdؒ3 (open circles) and Gdؒ4 (filled circles) at 25 ЊC and pH = 7. The
dotted line represents the calculated NMRD profile of Gdؒ4 with a τM
value lowered by one order of magnitude.
of this molecular weight and the parameters describing the
electronic relaxation are quite similar to those evaluated from
17O VT experiments. On the other hand, the experimental
profiles of Gdؒ4–5 were nicely fitted with similar parameters,
but by assuming a hydration number q = 2. A limiting effect of
the exchange lifetime on the relaxivity is observed only in the
case of Gdؒ4 where τM is of the order of µs (Fig. 4).
Interaction with HSA
The non covalent binding of Gd() complexes with HSA can
be conveniently studied by relaxometric methods on a low
resolution NMR spectrometer operating at a fixed frequency.23
Titration of a dilute aqueous solution of the complex with the
protein causes an increase of the relaxation rate when a binding
interaction occurs, as a result of the increase of the reorien-
tational correlation time of the macromolecular adduct. An
analysis of the data according to the proton relaxation
enhancement (PRE) method provides good estimates of the
association constant, KA, and of the relaxivity of the bound
complex, r1bp.23 All the complexes investigated show some degree
of interaction with HSA, as shown in Fig. 5. The hydrophilic
complexes Gdؒ1–2 present only a weak interaction with the
protein and this might be related to the lack of a specific, strong
binding site (estimated KA of ca. 1500 MϪ1 by assuming the
presence of a single class of equivalent binding sites). More
likely, the binding for these complexes simply involves charged
domains on the surface of the protein. On the other hand,
Gdؒ3–5 present aromatic groups able to interact more strongly
with the well defined hydrophobic sites (I and II) of HSA. In
fact, in this latter case, the association constants are of the same
NMRD
The analysis of the magnetic field dependence of the water
proton longitudinal relaxation times represents a powerful tool
for investigating the relaxometric properties of paramagnetic
compounds.22 The NMRD profiles of Gdؒ1–3 are consistent
with the results of the previous experiments as they reproduce
quite closely the behaviour of complexes analogous with q = 1.2
The analysis of the data was performed in terms of the
Solomon–Bloembergen–Morgan (SBM) theory of para-
magnetic relaxation, by using for τM the values previously
obtained (see Table 1), and yielded the parameters listed in
Table 2. Representative profiles for Gdؒ3 and Gdؒ4 are shown in
Fig. 4. The rotational correlation times are typical of complexes
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 7 0 – 5 7 7
573

View Article Online
of organic compounds were recorded using Bruker AC 300 or
Bruker AC 200 spectrometers. Melting points (uncorrected)
were determined with a Büchi SMP-20 capillary melting point
apparatus.
Note: Perchlorate salts are potentially dangerous substances.
Although we never experienced any problems during manipul-
ation of the perchlorate complexes of the ligands described
here, they should be handled with care.
Triethyleneglycol-monobutylether methanesulfonate 7
A solution of methanesulfonyl chloride (6.3 g, 55 mmol) in
pyridine (10 cm3) was added dropwise to a stirred solution of
triethyleneglycol-monobutyl ether (10.3 g, 55 mmol) in pyridine
(60 cm3), the temperature being kept between 0 and 5 ЊC. The
reaction mixture was stirred for 4 h at 0 ЊC. After addition of ice
(150 g) and acidification with 37% aqueous HCl, the mixture
was extracted with CH2Cl2 (3 × 50 cm3). The organic phase was
washed with water, brine and then dried over Na2SO4. Evapor-
ation of the solvent afforded 7 (13.9 g, 96%) as a yellowish thick
oil. δH (300 MHz, CDCl3, Me4Si) 0.89 (3 H, t, J 7.3), 1.35 (2 H,
m), 1.54 (2 H, m), 3.05 (3 H, s), 3.43 (2 H, m), 3.50–3.80 (10 H,
m), 4.36 (2 H, m).
Fig. 5 Longitudinal 1H water proton relaxation rates as a function of
[HSA] for Gdؒ1 (0.12 mM, filled circles), Gdؒ3 (0.18 mM, open circles)
and Gdؒ5 (0.12 mM, squares) at 20 MHz and 25 ЊC.
order of magnitude of those previously found for GdDOTA
derivatives bearing benzyloxypropionic groups (KA ≈ 6750,
5870 and 9290 MϪ1 for Gdؒ3–5, respectively).24 It must be noted
that the relaxivity values of the adducts are rather similar and
much lower than those expected for macromolecular para-
magnetic complexes (r1bp = 6.6, 6.8, 6.6, 10.6 and 12.5 mMϪ1 sϪ1
for Gdؒ1–5, respectively). This suggests that the binding might
involve water displacement by carboxylic groups of the HSA
with formation of ternary complexes lacking any coordinated
water molecules. This has been previously observed in related
complexes with q > 1 and represents a strong limitation for the
attainment of high relaxivity through the lengthening of τR.25–27
10-(3,6,9-Trioxatridecyl)-1,4,7,10-tetraazacyclododecane-1,4,7-
triacetic acid tris(1,1-dimethylethyl) ester 8
A suspension of 1,4,7,10-tetraazacyclododecane-1,4,7-triacetic
acid tris(1,1-dimethylethyl) ester 6 (0.51 g, 1.0 mmol), methane-
sulfonate 7 (0.35 g, 1.2 mmol) and, Na2CO3 (0.26 g, 2.5 mmol)
in anhydrous CH3CN (20 cm3) was refluxed for 72 h under
magnetic stirring. The mixture was cooled to room temper-
ature, filtered and the filtrate was evaporated. The residue was
dissolved in CH2Cl2 (50 cm3), washed with water (2 × 30 cm3)
and then with 30% aqueous NaClO4 (3 × 30 cm3). The solvent
was removed under reduced pressure and the product purified
by column chromatography (silica gel, CH2Cl2/MeOH 98 : 2) to
afford the sodium perchlorate complex of 8 as yellowish thick
oil. This product was dissolved in methanol (30 cm3) and stirred
overnight at room temperature in the presence of solid KCl
(0.23 g, 3.0 mmol). The suspension was filtered through Celite®
and the solvent was removed under reduced pressure. The
residue was dissolved in methanol (3 cm3) and transferred in
a liquid–liquid extractor. Water (30 cm3) was added and the
mixture was continuously extracted overnight with n-pentane
(200 cm3). The solvent was removed under reduced pressure to
afford 8 (0.41 g, 58%) as a yellowish oil. δH (300 MHz, CDCl3,
Me4Si) 0.9 (3 H, t, J 8.7), 1.34 (2 H, m), 1.44 (27 H, s), 1.60 (2 H,
m), 2.66 (4 H, m), 2.79 (12 H, br s), 3.27 (m, 4H), 3.42–3.48 (m,
4H), 3.54–3.63 (12 H, m); δC (75 MHz, CDCl3, Me4Si) 14.2,
19.6, 28.6, 32.1, 52.2, 52.3, 52.6, 53.2, 55.1, 55.2, 56.6, 57.1,
62.1, 69.4, 69.6, 70.4, 70.8, 70.9, 71.0, 71.5, 72.9, 81.1, 171.4;
m/z (FABϩ) 703 (M ϩ Hϩ. C36H70N4O9 requires 702).
Conclusions
The relaxometric properties of GdDO3A derivatives bearing an
appendage group on the secondary nitrogen atom of the
macrocycle show a marked dependence on the nature of the
substituent. Lower r1p values are found for the complexes
functionalised with a PEG chain and this is the result of a
partial water displacement by an oxygen atom of the PEG that
lowers the hydration number of the metal ion. As a con-
sequence, the q = 1 and q = 2 complexes exhibit different
pH-dependent relaxivity behaviour. In the diaquo complexes
the water molecules are displaced by the chelating carbonate
anions present in the aerated solutions at basic pH and r1p
significantly decreases. This effect is substantially reduced in the
case of the monoaquo complexes.
The mean residence lifetime of the bound water molecule(s)
is modulated as a function of the nature of the pendant sub-
stituent. This, in turn, reflects the differences in the structure
and dynamics of the H-bonded array of water molecules in the
outer hydration sphere of the complexes.
All the complexes show some degree of binding affinity to
HSA, with large differences in the KA values. Clearly, the design
of functionalised DO3A derivatives has to be optimised in view
of the attainment of high relaxivity for the corresponding
Gd-complexes. The introduction of a PEG chain in the substi-
tuent has to be carefully designed so as to avoid coordination to
the metal centre and water displacement. A possibility is the
introduction of a rigid spacer. The nature of the pendant
aromatic group also needs to be optimised in order to enhance
the targeting ability towards HSA. Finally, it is plausible that
the presence of a negatively charged group on the substituent
may help to decrease the ease of formation of ternary
complexes with bioactive anions in physiological conditions.
10-(3,6,9-Trioxatridecyl)-1,4,7,10-tetraazacyclododecane-1,4,7-
triacetic acid 1
A solution of triester 8 (154 mg, 0.22 mmol) in trifluoroacetic
acid (5 cm3) was stirred for 4 h at room temperature. The
solvent was removed under reduced pressure and the residue
was triturated with diethyl ether (50 cm3) to yield 1 (117 mg,
97%) as a white solid. δH (300 MHz, CD3OD, Me4Si) 0.93 (3H,
t, J 7), 1.39 (2 H, m), 1.55 (2 H, m), 3.06 (8 H, m), 3.38–3.58
(24 H, m), 3.85 (2 H, t, J 4.2), 3.98 (2 H, s); m/z (FABϩ) 535
(M ϩ Hϩ. C24H46N4O9 requires 534).
Experimental
Polyethyleneglycol (PEG) 350-monomethylether methane-
sulfonate 9
Solvents were purified by using standard methods and dried if
necessary. All commercially available reagents were used as
received. TLC was carried out on silica gel Si 60-F254. Column
chromatography was carried out on silica gel Si 60, mesh size
0.040–0.063 mm (Merck, Darmstadt, Germany). NMR spectra
A solution of anhydrous PEG 350-monomethylether (5.0 g,
14.3 mmol) and triethylamine (1.82 g, 18 mmol) in dry CH2Cl2
(30 cm3) was cooled to 0 ЊC. A solution of methanesulfonyl
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 7 0 – 5 7 7
574

View Article Online
chloride (2.31 g, 17.7 mmol) in CH2Cl2 (20 cm3) was slowly
added under stirring and the reaction was then maintained at
4 ЊC for 24 h. The solvent was evaporated and the crude
material was purified by column chromatography (neutral
aluminium oxide, CH3CN) to give 9 (6.15 g, 97%) as a viscous
oil. δH (300 MHz, CDCl3, Me4Si) 3.06 (3 H, s), 3.35 (3 H, s),
3.50–3.76 (26 H, m), 4.33–4.38 (2 H, m).
mg, 85%) as a white solid. δH (300 MHz, CDCl3, Me4Si) 1.42
(9 H, s). 1.44 (18 H, s), 2.10–2.80 (18 H, m), 2.98 (4 H, s),
3.20–3.30 (2 H, m), 3.57 (2 H, m), 3.67 (4 H, m), 3.82 (5 H, m),
4.13 (2 H, t, J 4), 6.87–6.90 (4 H, m); δC (75 MHz, CDCl3,
Me4Si) 28.3, 28.4, 49.8, 52.5, 55.9, 56.2, 56.7, 67.1, 68.5, 70.0,
70.1, 82.5, 112.4, 114.0, 121.2, 121.9, 172.3, 172.8; m/z (FABϩ)
776 ([M – ClO4]ϩ. C39H68N4O10NaClO4 requires 875).
10-[PEG 350 monomethyl ether]-1,4,7,10-tetrazacyclododecane-
1,4,7-triacetic acid tri tert-butyl ester 10
10-{1-[8-(2-Methoxyphenoxy)-3,6-dioxa]-octyl}-1,4,7,10-tetra-
azacyclododecane-1,4,7-triacetic acid 3
A solution of 1,4,7,10-tetraazacyclododecane-1,4,7-triacetic
acid tris(1,1-dimethylethyl) ester 6 (0.8 g, 1.55 mmol) and PEG
350 monomethyl ether methanesulfonate 9 (0.69 g, 1.55 mmol)
in dry DMF (20 cm3) was heated for 24 h under nitrogen at
110 ЊC. The solvent was removed under reduced pressure and
the residue was purified by flash chromatography on silica gel
(gradient elution of concentrated NH3 and CH2Cl2–MeOH
9 : 1) to give 10 (0.91 g, 70%) as a pale yellow oil. δH (200 MHz,
CDCl3, Me4Si) 1.40–1.48 (27 H, m), 2.26–2.90 (18 H, m), 3.02
(4 H, s), 3.12 (2 H, s), 3.37 (3 H,s), 3.48–3.73 (26 H, m); δC (50.3
MHz, CDCl3, Me4Si) 28.1, 49.6, 50.5, 52.8, 55.6, 56.3, 58.9,
67.2, 69.7, 70.4, 71.8, 82.1, 172.5, 172.8; m/z (EI) 836 (Mϩ.
C41H80N4O13 requires 836), 792, 748, 704, 660, 616, 514.
A solution of the perchlorate complex 12 (0.31 g, 0.35 mmol) in
methanol (30 cm3) was stirred overnight at room temperature
in the presence of solid KCl (0.38 g, 5 mmol). The suspension
was filtered through Celite® and the solvent was removed under
reduced pressure. The residue was dissolved in methanol (3 cm3)
and transferred in a liquid–liquid extractor. Water (30 cm3) was
added and the mixture was continuously extracted overnight
with n-pentane (200 cm3). The solvent was removed under
reduced pressure affording a residue (0.25 g) which was
dissolved in trifluoroacetic acid (5 cm3) and stirred at room
temperature for 4 hours. The solvent was removed under
reduced pressure and the residue was triturated with diethyl
ether (50 cm3) to yield 3 (0.19 g, 93%) as a white solid. δH (300
MHz, CD3OD, Me4Si) 2.90–3.00 (8 H, m), 3.40–3.49 (8 H, m),
3.60–3.74 (10 H, m), 3.83 (3 H, m), 3.86 (2 H, m), 3.92 (2 H, m),
4.07 (2 H, m), 4.14 (2 H, m), 6.90 (4 H, m); δH (75 MHz,
CD3OD, Me4Si) 52.2, 52.6, 53.7, 55.0, 55.9, 56.5, 67.0, 69.9,
70.8, 71.2, 71.4, 113.8, 115.4, 122.3, 122.8, 149.7, 150.9, 169.2,
174.2;
10-[PEG 350 monomethylether]-1,4,7,10-tetrazacyclododecane-
1,4,7-triacetic acid 2
A solution of 10 (0.9 g, 1.07 mmol) in trifluoroacetic acid
(2 cm3) and CH2Cl2 (8 cm3) was stirred for 12 h at room tem-
perature. Solvents were removed under reduced pressure
and the residue was taken up in THF (10 cm3). Addition of
n-heptane to this solution allowed the formation of a light
precipitate which was collected by centrifugation to afford
2 (0.6 g, 90%) as a white solid. δH (200 MHz, D2O, Me4Si)
2.62–3.00 (16 H, m), 3.15 (3 H, s), 3.20–3.73 (34 H, m); δC (50.3
MHz, D2O, Me4Si) 50.1, 50.7, 53.2, 54.1, 55.1, 56.1, 56.9, 60.4,
67.0, 71. 9, 72.3, 73.3, 170.8, 176.3; m/z (FABϩ) 669 (M ϩ Hϩ.
C29H56N4O13 requires 668), 691 (M ϩ Naϩ).
m/z (FABϩ) 585 (M ϩ Hϩ. C27H44N4O10 requires 584), 607
(M ϩ Naϩ).
3-(2-Hydroxyphenyl)-1-propanol 19
A solution of commercially available 3,4-dihydrocoumarin 18
(7.41 g, 50 mmol) in dry Et2O (40 cm3) was added dropwise at
room temperature to a stirred suspension of LiAlH4 (2.28 g,
60 mmol) in dry Et2O (40 cm3). The reaction mixture was
refluxed for 4 h and then stirring was continued at room
temperature overnight. The excess of LiAlH4 was destroyed by
careful addition of THF/H2O (1 : 1 v/v, 5 cm3); the mixture was
stirred for 30 min, then 5% aqueous H2SO4 (30 cm3) was added
dropwise until the gas evolution ceased. The organic layer was
separated and the aqueous phase was extracted with Et2O
(2 × 30 cm3). The combined organic layers were washed with
H2O (30 cm3), 5% aqueous NaHCO3 (30 cm3) and brine (30
cm3). The organic phase was dried over Na2SO4 and the solvent
evaporated in vacuo to afford a yellow residue which was
purified by column chromatography (silica-gel, CH2Cl2/AcOEt
7 : 1) to give 19 (6.62 g, 87%) as a pale yellow oil. This
compound had been previously synthesized following a differ-
ent procedure.12
Triethyleneglycol-mono-2-methoxyphenylether methane-
sulfonate 11
A solution of methanesulfonyl chloride (1.34 g, 11.7 mmol) in
anhydrous CH2Cl2 (10 cm3) was added dropwise to a stirred
solution of triethyleneglycol-mono-2-methoxyphenylether 1711
(3 g, 11.7 mmol) and [2,2,2]-diazabicyclooctane (DABCO) (4 g,
35 mmol) in anhydrous CH2Cl2 (160 cm3), the temperature
being kept between 0 and 5 ЊC. The reaction mixture was stirred
for 4 h at 0 ЊC. After addition of ice (100 g) and acidification
with 37% aqueous HCl, the mixture was extracted with CH2Cl2
(3 × 50 cm3), and the organic phase was washed with water,
brine and then dried over Na2SO4. Evaporation of the solvent
afforded 11 (3.6 g, 92%) as a colourless viscous oil. δH (300
MHz, CDCl3, Me4Si) 3.05 (3 H, s), 3.59–4.21 (12 H, m), 3.85
(3 H, s), 6.90 (4 H, m).
δH (300 MHz, CDCl3, Me4Si) 1.83–1.91 (2 H, m), 2.45 (1 H,
s), 2.77 (2 H, t, J 6.6), 3.63 (2 H, t, J 5.9 Hz), 6.82–6.88(2 H, m),
7.00 (1 H, s), 7.08–7.15 (2 H, m); δC (75 MHz 300 MHz, CDCl3,
Me4Si) 25.8, 32.8, 61.4, 116.3, 121.2, 127.9, 128.0, 131.0, 154.8.
10-{1-[8-(2-Methoxyphenoxy)-3,6-dioxa]-octyl}-1,4,7,10-tetra-
azacyclododecane-1,4,7-triacetic acid tris(1,1-dimethylethyl)
ester sodium perchlorate complex 12
2-(3-Hydroxypropyl)phenyl benzoate 20
A suspension of 1,4,7,10-tetraazacyclododecane-1,4,7-triacetic
acid tris(1,1-dimethylethyl) ester 6 (257 mg, 0,50 mmol), meth-
anesulfonate 11 (167 mg, 0,55 mmol) and Na2CO3 (212 mg,
2.0 mmol) in anhydrous CH3CN (20 cm3) was refluxed 72 h
under magnetic stirring. The mixture was cooled to room
temperature, filtered and the filtrate was evaporated. The
residue was dissolved in CH2Cl2 (50 cm3), washed with water
(2 × 30 cm3) and then with 30% aqueous NaClO4 (3 × 30 cm3).
The solvent was removed under reduced pressure and the
product purified by column chromatography (silica gel, CH2Cl2/
MeOH 95/5) to afford the sodium perchlorate complex 12 (371
Neat benzoyl chloride (0.58 cm3, 5 mmol) was added via syringe
to a stirred solution of 19 (1.52 g, 10 mmol) and triethylamine
(2 cm3) in anhydrous CH2Cl2 (20 cm3). The solution was stirred
at room temperature for 15 min then a second portion of
benzoyl chloride (0.58 cm3, 5 mmol) was added and stirring was
continued overnight. The reaction mixture was transferred into
a separatory funnel, diluted with CH2Cl2 (30 cm3) and washed
with H2O (2 × 30 cm3) and brine (30 cm3). The organic phase
was dried over Na2SO4 and the solvent evaporated to afford a
yellow residue which was purified by column chromatography
(silica-gel, CH2Cl2/AcOEt 9 : 1) to give 20 (1.83 g, 71.4%) as a
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 7 0 – 5 7 7
575

View Article Online
colourless thick oil. δH (300 MHz, CDCl3, Me4Si) 1.83–1.92
(2 H, m), 2.69 (2 H, t, J 6.95), 3.62 (2 H, t, J 5.20), 7.13–7.33
(4 H, m), 7.50–7.68 (3 H, m), 8.20–8.22 (2 H, m); δC (75 MHz
300 MHz, CDCl3, Me4Si) 26.2, 32.7, 61.9, 122.4, 126.1, 127.1,
128.6, 129.3, 130.0, 130.3, 133.6, 133.7, 149.1, 165.2.
residue was purified by column chromatography (silica gel,
CH2Cl2) to afford 15 (1,72 g, 63%) as a colourless oil. δH (300
MHz, CDCl3, Me4Si) 2.35 (2 H, m), 3.70 (2 H, t, J 6.2), 3.88 (3
H, s), 4.18 (2 H, t, J 5.6), 6.97(2 H, m), 7.46 (1 H, m), 7.81 (1 H,
dd, J 8.0, 2.0).
2-(3-Bromopropyl)phenyl benzoate 13
10-{1-[3-(2-Carbomethoxyphenyl)]-propyl}-1,4,7,10-tetraaza-
cyclododecane-1,4,7-triacetic acid tris(1,1-dimethylethyl) ester
sodium perchlorate complex 16
Solid N-bromosuccinimide (NBS) (0.21 g, 1.2 mmol) was added
portionwise to a stirred solution of 2-(3-hydroxypropyl)phenyl
benzoate 20 (0.25 g, 1 mmol) and triphenylphosphine (0.31 g,
1.2 mmol) in dry CH2Cl2 (15 cm3). During the addition the
temperature was kept at 0 ЊC and then the reaction mixture was
maintained at this temperature and under stirring for 1 hour.
The solvent was removed under reduced pressure and the
brown waxy solid residue was taken up in diethyl ether (50 cm3).
The red brown solid thus formed was collected by filtration, the
filtrate was concentrated and purified by column chromato-
graphy (silica gel, light petroleum/diethyl ether = 9 : 1) to afford
13 (0.22 g, 68%) as a colourless oil. δH (300 MHz, CDCl3,
Me4Si) 2.15 (2 H, m). 3.36 (2 H, t, J 5.1), 3.76 (2 H, t, J 8.6),
7.15–7.30 (4 H, m), 7.45–7.60 (3 H, m), 8.20–8.22 (2 H, m),
A suspension of 1,4,7,10-tetraazacyclododecane-1,4,7-triacetic
acid tris(1,1-dimethylethyl) ester 6 (0.257 g, 0,5 mmol), alkyl
bromide 15 (0.138 g, 0.5 mmol) and Na2CO3 (0.212 g, 2 mmol)
in anhydrous CH3CN (20 cm3) was refluxed for 40 h under
magnetic stirring. The mixture was cooled to room temper-
ature, filtered and the filtrate was evaporated. The residue was
dissolved in CH2Cl2 (50 cm3), washed with water (2 × 30 cm3)
and then with 30% aqueous NaClO4 (3 × 30 cm3). The solvent
was removed under reduced pressure and the residue purified
by column chromatography (silica gel, CH2Cl2/MeOH 95 : 5) to
afford the complex 16 (0.31 g, 76%) as a white solid. δH (300
MHz, CDCl3, Me4Si) 1.39 (9 H, s), 1.40 (18 H, s), 1.95 (2 H, m),
2.10–3.50 (26 H, m), 3.85 (3 H, s), 4.03 (2 H, t, J 9), 6.92 (2 H,
m), 7.41 (1 H, m), 7.78 (1 H, dd, J 8.0, 1.8); δC (75 MHz, CDCl3,
Me4Si) 28.2, 28.3, 28.5, 50.5, 51.5, 52.2, 56.2, 57.0, 67.7, 82.7,
83.1, 94.2, 94.4, 113.7, 120.6, 120.8, 132.0, 133.8, 158.4, 173.0,
174.0; m/z (FABϩ) 729 ([M Ϫ ClO4]ϩ. C37H62N4O9NaClO4
requires 828).
10-{1-[3-(2-Carboxyphenyl)]-propyl}-1,4,7,10-tetraazacyclo-
dodecane-1,4,7-triacetic acid tris(1,1-dimethylethyl) ester 14
A suspension of 1,4,7,10-tetraazacyclododecane-1,4,7-triacetic
acid tris(1,1-dimethylethyl) ester 6 (0.32 g, 0,63 mmol), alkyl
bromide 13 (0.20 g, 0,63 mmol) and Na2CO3 (0.21 g, 2 mmol) in
anhydrous CH3CN (20 cm3) was refluxed for 24 h under
magnetic stirring. The mixture was cooled to room temper-
ature, filtered and the filtrate was evaporated. The residue was
dissolved in CH2Cl2 (50 cm3) and washed with water (2 × 30
cm3). The solvent was removed under reduced pressure and the
residue purified by column chromatography (silica gel, CH2Cl2/
MeOH 95 : 5) to afford compound 14 (0.276 g, 58%) as a white
solid. δH (300 MHz, CDCl3, Me4Si) 1.37 (9 H, s), 1.41 (18 H, s),
1.60–2.38 (6 H, m), 2.40–2.80 (8 H, m), 2.90–3.20 (14 H, m),
7.10–7.20 (4 H, m), 7.53 (2 H, m), 7.67 (1 H, m), 8.20 (2 H, m);
δC (75 MHz 300 MHz, CDCl3, Me4Si), 28.1, 28.4, 28.7, 50.0,
51.3, 53.2, 53.6, 54.3, 56.1, 56.7, 82.7, 83.0, 122.8, 126.6, 127.6,
129.1, 130.0, 130.3, 133.8, 134.2, 149.3, 165.3, 172.9, 173.8;
m/z (FABϩ) 754 (M ϩ Hϩ. C42H64N4O8 requires 753), 776
(M ϩ Naϩ).
10-{1-[3-(2-Carbomethoxyphenyl)]-propyl}-1,4,7,10-tetraaza-
cyclododecane-1,4,7-triacetic acid 5
A solution of 16 (0.29 g, 0.35 mmol) in methanol (30 cm3) was
stirred overnight at room temperature in the presence of solid
KCl (0.38 g, 5 mmol). The suspension was filtered through
Celite® and the solvent was removed under reduced pressure.
The residue was dissolved in methanol (3 cm3) and transferred
intoaliquid–liquidextractor.Afteradditionofwater(30cm3)the
mixture was continuously extracted overnight with n-pentane
(200 cm3). The solvent was removed under reduced pressure to
afford a residue, which was stirred at reflux for 4 h with NaOH
(0.086 g, 2.3 mmol) in 96% EtOH (20 cm3). Evaporation of the
solvent afforded a residue which was dissolved in water (5 cm3),
acidified to pH = 1 with 10% aqueous HCl and evaporated in
vacuo. The residue was taken up with absolute ethanol (10 cm3),
the solid was removed by filtration and the filtrate was evapor-
ated under reduced pressure to yield 5 (0.17 g, 89%) as a white
solid. δH (300 MHz, CD3OD, Me4Si) 2.27 (2 H, m), 3.13 (4 H,
m), 3.50 (18 H, m), 4.18 (2 H, t, J 5), 4.25 (2 H, s), 7.05 (2 H, m),
7.50 (1 H, m), 7.82 (1 H, dd, J 8, 2); 13C NMR (CD3OD, 75
MHz) 175.1, 170.1, 136.1, 134.3, 122.8, 115.4, 67.0, 56.2, 54.3,
53.9, 51.6, 51.1, 50.8, 50.0, 25.1; m/z (FABϩ) 525 (M ϩ Hϩ.
C24H36N4O9 requires 524).
10-{1-[3-(2-Hydroxyphenyl)]-propyl}-1,4,7,10-tetraazacyclo-
dodecane-1,4,7-triacetic acid 4
A solution of triester 14 (0.25 g, 0.33 mmol) and NaOH (0.16 g,
4 mmol) in 96% EtOH (20 cm3) was stirred at reflux for 4 h.
Evaporation of the solvent afforded a residue which was dissol-
ved in water (5 cm3) and acidified to pH = 1 with 10% aqueous
HCl. The precipitated benzoic acid was washed out with CHCl3
(2 × 30 cm3) and the aqueous phase was evaporated in vacuo.
The residue was taken up with absolute ethanol (10 cm3), the
undissolved solid was removed by filtration and the solvent was
evaporated under reduced pressure to yield 4 (0.12 g, 76%) as a
white solid. δH (300 MHz, CD3OD, Me4Si) 2.12 (2 H, m), 2.70
(2 H, t, J 6), 2.88 (2 H, m), 3.10 (8 H, m), 3.40 (8 H, m), 3.63
(4 H, m), 4.23 (2 H, s), 6.75 (2 H, m), 7.06 (2 H, m); m/z (FABϩ)
481 (M ϩ Hϩ. C28H36N4O7 requires 480), 503 (M ϩ Naϩ).
Synthesis of the Gd(III) complexes
The complexes were prepared by mixing stoichiometric
amounts of the ligand and of gadolinium chloride and by
adjusting the pH to 7 with NaOH. The solutions were kept at
room temperature under vigorous stirring for about one hour
until the pH stabilized. The compounds were then purified by
precipitation by addition of acetone. The absence of free metal
ion was assessed by the constant value of the water proton
relaxation rate, measured at 25 ЊC and 20 MHz, after addition
of a small excess of the free ligand.
2-(3-Bromopropoxy)benzoic acid methyl ester 15
A suspension of methyl salicylate (1.52 g, 10 mmol), 1,3
dibromopropane (10 g, 50 mmol) and K2CO3 (3.12 g, 20 mmol)
in dimethylformamide (30 cm3) was stirred at 70 ЊC for 16 h.
The reaction mixture, diluted with water (100 cm3), was trans-
ferred into a separatory funnel and extracted with diethyl ether
(3 × 50 cm3). The combined organic layers were washed with
10% aqueous NaOH (40 cm3) and water (2 × 50 cm3), dried
with Na2SO4 and the solvent was evaporated in vacuo. The
Relaxometric measurements
Water proton relaxivity measurements. The water proton 1/T 1
longitudinal relaxation rates (20 MHz, 25 ЊC) were measured
with a Stelar Spinmaster Spectrometer (Mede, Pv, Italy) on 0.1–
1
1.5 mM aqueous solutions of the complexes. H spin-lattice
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 7 0 – 5 7 7
576

View Article Online
3 P. Caravan, J. J. Ellison, T. J. McMurry and R. B. Lauffer, Chem.
Rev., 1999, 99, 2293–2352.
relaxation times T 1 were acquired by the standard inversion-
recovery method with typical 90Њ pulse width of 3.5 µs, 16
experiments of 4 scans. The reproducibility of the T 1 data was
1%. The temperature was controlled with a Stelar VTC-91 air-
flow heater equipped with a copper–constantan thermocouple
(uncertainty of 0.1 ЊC). Titration experiments with HSA
(Sigma, St. Louis, Mo, USA): different amounts of the protein
were added to a dilute (0.1–0.2 mM) aqueous solution of the
complexes at pH = 7.2 and the water proton relaxation rate was
measured after each addition at 25 ЊC. The starting pH was
adjusted by either HCl or KOH. Moreover, the pH of the
solutions was controlled before and after the measurement. The
1/T 1 nuclear magnetic relaxation dispersion profiles of water
protons were measured over a continuum of magnetic field
strength from 0.00024 to 0.28 T (corresponding to 0.01–12
MHz proton Larmor frequency) on the fast field-cycling Stelar
Spinmaster FFC relaxometer installed at the “Laboratorio
Integrato di Metodologie Avanzate”, Bioindustry Park del
Canavese (Colleretto Giacosa, To, Italy). The relaxometer
operates under complete computer control with an absolute
uncertainty in the 1/T 1 values of 1%. Additional data points
at 20 and 90 MHz were recorded on the Stelar Spinmaster and
on a JEOL EX-90 spectrometer, respectively. The concentration
of the aqueous solutions of the complexes utilized for the
measurements was in the range 0.5–1.0 mM.
4 É. Tóth, L. Helm and A. E. Merbach, in The Chemistry of Contrast
Agents in Medical Magnetic Resonance Imaging, eds. A. E. Merbach
and É. Tóth, John Wiley & Sons, Chichester (UK), 2001, Chapter 2.
5 J. A. Peters, J. Huskens and D. J. Raber, Prog. Nucl. Magn. Reson.
Spectrosc., 1996, 28, 283–350.
6 C. A. Chang, L. C. Francesconi, M. F. Malley, K. Kumar,
J. Z. Gougoutas, M. F. Tweedle, D. W. Lee and L. J. Wilson, Inorg.
Chem., 1993, 32, 3501–3508.
7 S. Aime, M. Botta, S. Geninatti Crich, G. Giovenzana, R. Pagliarin,
M. Sisti and E. Terreno, Magn. Reson. Chem., 1998, 36, S200–S208.
8 É. Tóth, O. M. Ni Dhubhghaill, G. Besson, L. Helm and
A. E. Merbach, Magn. Reson. Chem., 1999, 37, 701–708.
9 R. S. Ranganathan, E. R. Marinelli, R. Pillai and M. F. Tweedle,
WO 95/27705, Chem. Abstr., 1995, 124, 146216.
10 S. Quici, G. Marzanni, M. Cavazzini, P. L. Anelli, M. Botta,
E. Gianolio, G. Accorsi, N. Armaroli and F. Barigelletti, Inorg.
Chem., 2002, 41, 2777–2784.
11 C. Almansa, A. Moyano and F. Serratosa, Tetrahedron, 1991, 47,
5867–5876.
12 J. O. Knipe, P. J. Vasques and J. K. Coward, J. Am. Chem. Soc., 1982,
104, 3202–3209.
13 D. M. J. Doble, M. Botta, J. Wang, S. Aime, A. Barge and
K. N. Raymond, J. Am. Chem. Soc., 2001, 123, 10758–10759.
14 S. M. Cohen, J. Xu, E. Radkov, K. N. Raymond, M. Botta, A. Barge
and S. Aime, Inorg. Chem., 2000, 39, 5747–5756.
15 M. Botta, S. Aime, A. Barge, G. Bobba, R. S. Dickins, D. Parker and
E. Terreno, Chem. Eur. J., 2003, 9, 2102–2109.
16 U. Frey, A. E. Merbach, D. H. Powell, in Dynamics of Solutions and
Fluid Mixtures by NMR, ed. J.-J. Delpuech, John Wiley & Sons,
Chichester (UK), 1995, pp. 263–307.
17 S. Aime, A. Barge, M. Botta, A. S. De Sousa and D. Parker, Angew.
Chem., Int. Ed., 1998, 37, 2673–2675.
18 S. Aime, M. Botta, M. Fasano and E. Terreno, Acc. Chem. Res.,
1999, 32, 941–949.
19 A. Barge, M. Botta, D. Parker and H. Puschmann, Chem. Commun.,
2003, 1386–1387.
VT 17O relaxation measurements. Variable-temperature 17O
NMR measurements were recorded on a JEOL EX-90 (2.1 T)
spectrometer, equipped with a 5 mm probe, by using a D2O
external lock. Experimental settings were: spectral width 10000
Hz, pulse width 7 µs (90Њ), acquisition time 10 ms, 1000 scans
and no sample spinning. Aqueous solutions of the para-
magnetic complexes (pH = 6.3) containing 2.6% of 17O isotope
(Yeda, Israel) were used. The observed transverse relaxation
rates (R2) were calculated from the signal width at half height.
20 S. Aime, A. Barge, A. S. Batsanov, M. Botta, D. Delli Castelli,
F. Fedeli, A. Mortillaro, D. Parker and H. Puschmann, Chem.
Commun., 2002, 1120–1121.
21 E. Terreno, P. Boniforte, M. Botta, F. Fedeli, L. Milone,
A. Mortillaro and S. Aime, Eur. J. Inorg. Chem., 2003, 3530–3533.
22 S. H. Koenig and R. D. Brown III, Prog. Nucl. Magn. Reson.
Spectrosc., 1990, 22, 487–567.
23 S. Aime, M. Botta, M. Fasano and E. Terreno, in The Chemistry of
Contrast Agents in Medical Magnetic Resonance Imaging, eds.
A. E. Merbach and É. Tóth, John Wiley & Sons, Chichester (UK),
2001, Chapter 5.
1
Theory. The theory behind the analysis of the water H and
17O relaxation properties can be found in several recent
publications.1,18,28
Acknowledgements
24 S. Aime, M. Botta, M. Fasano, S. Geninatti Crich and E. Terreno,
J. Biol. Inorg. Chem., 1996, 1, 312–319.
25 S. Aime, A. Barge, M. Botta, J. A. K. Howard, R. Kataky,
M. P. Lowe, J. M. Moloney, D. Parker and A. S. De Sousa, Chem.
Commun., 1999, 1047–1048.
26 S. Aime, E. Gianolio, E. Terreno, G. B. Giovenzana, R. Pagliarin,
M. Sisti, G. Palmisano, M. Botta, M. P. Lowe and D. Parker, J. Biol.
Inorg. Chem., 2000, 5, 488–497.
We thank MIUR (programme “Chemistry: Contrast and Shift
Agents and Luminescent Probes”: Legge 95/95), the EC/ESF
COST Action D-18 for support and the Bioindustry Park
of Canavese (Ivrea, Italy) for access to the field-cycling
relaxometer.
27 S. Aime, M. Botta, J. I. Bruce, V. Maniero, D. Parker and E. Terreno,
Chem. Commun., 2001, 115–116.
28 D. H. Powell, O. M. Ni Dhubhghaill, D. Pubanz, L. Helm,
Y. S. Lebedev, W. Schlaepfer and A. E. Merbach, J. Am. Chem. Soc.,
1996, 118, 9333–9346.
References
1 R. B. Lauffer, Chem. Rev., 1987, 87, 901–927.
2 S. Aime, M. Botta, M. Fasano and E. Terreno, Chem. Soc. Rev.,
1998, 27, 19–29.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 7 0 – 5 7 7
577