Fine-tuning water exchange on GdIII poly(amino carboxylates) by modulation of steric crowding

Article abstract of DOI:10.1039/b506702b

In the objective of optimizing water exchange rate on stable, nine-coordinate, monohydrated Gd^III poly(amino carboxylate) complexes, we have prepared monopropionate derivatives of DOTA^4- (DO3A-Nprop^4-) and DTPA^5- (DTTA-Nprop^5-). A novel ligand, EPTPA-BAA^3-, the bisamylamide derivative of ethylenepropylenetriaminepentaacetate (EPTPA^5-) was also synthesized. A variable temperature ¹⁷O NMR study has been performed on their Gd^III complexes, which, for [Gd(DTTA-Nprop)(H₂O)] ^2- and [Gd(EPTPA-BAA)(H₂O)] has been combined with multiple field EPR and NMRD measurements. The water exchange rates, k _ex²⁹⁸, are 8.0 × 10⁷ s^-1, 6.1 × 10⁷ s^-1 and 5.7 × 10⁷ s ^-1 for [Gd(DTTA-Nprop)(H₂O)]^2-, [Gd(DO3A-Nprop)(H₂O)]^- and [Gd(EPTPA-BAA)(H ₂O)], respectively, all in the narrow optimal range to attain maximum proton relaxivities, provided the other parameters (electronic relaxation and rotation) are also optimized. The substitution of an acetate with a propionate arm in DTPA^5- or DOTA^4- induces increased steric compression around the water binding site and thus leads to an accelerated water exchange on the Gd^III complex. The k_ex values on the propionate complexes are, however, lower than those obtained for [Gd(EPTPA)(H₂O)]^2- and [Gd(TRITA)(H₂O)] ^- which contain one additional CH₂ unit in the amine backbone as compared to the parent [Gd(DTPA)(H₂O)]^2- and [Gd(DOTA)(H₂O)]^-. In addition to their optimal water exchange rate, [Gd(DTTA-Nprop)(H₂O)]^2- has, and [Gd(DO3A-Nprop)(H₂O)]^- is expected to have sufficient thermodynamic stability. These properties together make them prime candidates for the development of high relaxivity, macromolecular MRI contrast agents. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b506702b

View Article Online / Journal Homepage / Table of Contents for this issue
F U L L P A P E R
III
Fine-tuning water exchange on Gd poly(amino carboxylates) by
modulation of steric crowding†
´
Zolt a´ n J a´ szber e´ nyi, Ang e´ lique Sour, Eva T o´ th, Meriem Benmelouka and Andr e´ E. Merbach*
Laboratoire de Chimie Inorganique et Bioinorganique, Ecole Polytechnique F e´ d e´ rale de
Lausanne, ISIC, BCH, CH-1015 Lausanne, Switzerland. E-mail: andre.merbach@epfl.ch;
Fax: 41 21 693 9875; Tel: 41 21 693 9871
Received 12th May 2005, Accepted 22nd June 2005
First published as an Advance Article on the web 12th July 2005
III
In the objective of optimizing water exchange rate on stable, nine-coordinate, monohydrated Gd poly(amino
carboxylate) complexes, we have prepared monopropionate derivatives of DOTA (DO3A-Nprop ) and DTPA
4
−
4−
5−
5
−
3−
(
DTTA-Nprop ). A novel ligand, EPTPA-BAA , the bisamylamide derivative of ethylenepropylenetriamine-
5
−
17
pentaacetate (EPTPA ) was also synthesized. A variable temperature O NMR study has been performed on their
III
2−
Gd complexes, which, for [Gd(DTTA-Nprop)(H
multiple field EPR and NMRD measurements. The water exchange rates, kex , are 8.0 × 10 s , 6.1 × 10 s and
.7 × 10 s for [Gd(DTTA-Nprop)(H O)] , [Gd(DO3A-Nprop)(H O)] and [Gd(EPTPA-BAA)(H O)],
respectively, all in the narrow optimal range to attain maximum proton relaxivities, provided the other parameters
electronic relaxation and rotation) are also optimized. The substitution of an acetate with a propionate arm in
DTPA or DOTA induces increased steric compression around the water binding site and thus leads to an
accelerated water exchange on the Gd complex. The kex values on the propionate complexes are, however, lower
than those obtained for [Gd(EPTPA)(H
the amine backbone as compared to the parent [Gd(DTPA)(H
optimal water exchange rate, [Gd(DTTA-Nprop)(H
2
O)] and [Gd(EPTPA-BAA)(H
2
O)] has been combined with
2
98
7
−1
7
−1
7
−1
2−
−
5
2
2
2
(
5
−
4−
III
2
−
−
2
O)] and [Gd(TRITA)(H
2
O)] which contain one additional CH unit in
2
2
−
−
2
O)] and [Gd(DOTA)(H
O)] has, and [Gd(DO3A-Nprop)(H
2
O)] . In addition to their
2
−
−
2
2
O)] is expected to have
sufficient thermodynamic stability. These properties together make them prime candidates for the development of
high relaxivity, macromolecular MRI contrast agents.
III
Introduction
Gd complexes, like TREN-Me-3,2-HOPO derivatives have
3
shown faster water exchange. However, for bishydrated chelates
a potential drawback is the formation of ternary complexes with
small ligands (carbonate, phosphate, citrate) in biological fluids,
which can erase any relaxivity gain in vivo.
Nine-coordinate, monohydrated Gd poly(amino carboxy-
lates), including all commercial Gd -based MRI contrast
agents, undergo a dissociative, D, or dissociative interchange,
I , water exchange, in contrast to the associative, A, mechanism
on the eight-coordinate aqua ion, [Gd(H
dissociative exchange processes is primarily determined by the
overall charge of the chelate (a more negative charge leads
to faster exchange), and by the steric crowding in the inner
coordination sphere. An increased steric compression around
the inner sphere water molecule will facilitate its leaving which, in
a dissociative process, constitutes the rate determining step. Re-
cently we have shown that increased steric compression around
the water binding site indeed results in a remarkable acceleration
of the water exchange process. Steric compression has been
induced in the acyclic DTPA- or the macrocyclic DOTA-type
complexes by the insertion of an additional CH
the amine backbone of the ligand (EPTPA , TRITA ), or in
the carboxylate arm (DTTA-N prop Scheme 1). While the
elongation of the amine backbone results in almost 2 orders
of magnitude increase in the water exchange rate of the Gd
III
Paramagnetic Gd complexes are extensively applied as contrast
agents in Magnetic Resonance Imaging (MRI). By enhancing
1
the intrinsic contrast, they largely contribute to the excellent
anatomical resolution of the MR images. Currently emerging
applications, such as molecular imaging call for contrast agents
of much higher efficiency than the marketed ones. The rational
design of high efficacy agents has to consider the relation-
ships between structure, dynamics and the relevant parameters
determining relaxation processes. The relaxivity of a contrast
agent is defined as the paramagnetic proton relaxation rate
enhancement of the bulk water protons, referred to 1 mmo-
lar concentration of gadolinium. The Solomon–Bloembergen–
Morgan theory, which relates the observed paramagnetic re-
laxation rate enhancement to microscopic properties, predicts
4
III
III
d
3+ 1
2
O) ] . The rate of
8
−
1
−1
III
proton relaxivities over 100 mM
s
for Gd complexes
provided the three most important influencing factors, rotation,
electron spin relaxation and water exchange are simultaneously
1
optimised. The rotation has to be slowed down which has led
in the past few years to the development of macromolecular
agents. The optimisation of the electron spin relaxation on Gd
2
group either in
III
5−
4−
complexes remains a difficult issue, despite the recent theoretical
ꢀ
5−
5,6
2
advances. The optimal value of the water exchange rate, the
third determining factor for proton relaxivity, is situated in
III
7
−1
a relatively small range (around kex ≈ 5 × 10 s ). In the
−
complex in comparison to the parent [Gd(DOTA)(H
Gd(DTPA)(H
by a propionate on the DTPA ligand gives rise only to a
0-fold increase in kex. The water exchange rates determined
for [Gd(TRITA)(H
higher than the value which would be optimal to attain max-
imum proton relaxivities. Indeed, when [Gd(EPTPA)(H O)]
was covalently linked to different generations of PAMAM
dendrimers, the relaxivity of these macromolecular agents was
2
O)] or
III
currently used Gd -based contrast agents water exchange is
2−
[
2
O)] , the substitution of the central acetate arm
6
−1
1
considerably slower (kex ≈ 1–4 × 10 s ). Some bishydrated
5−
1
−
2−
2
O)] and [Gd(EPTPA)(H O)] are slightly
2
†
Electronic supplementary information (ESI) available: Tables of the
17
variable temperature O NMR, EPR and NMRD data. Figures
of O NMR experimental data and fitted curves for [Gd(DO3A-
Nprop)(H
2
−
17
2
−
17
2
O)] , and O NMR, EPR and NMRD experimental data
O)]. Outline of the Solomon–
and fitted curves for [Gd(EPTPA-BBA)(H
2
III
1
7
Bloembergen–Morgan theory used for the analysis of O NMR, EPR
and NMRD data. See http://dx.doi.org/10.1039/b506702b
partially limited by the too fast water exchange of the Gd
chelate. [Gd(DTTA-N prop)(H O)] was found to ensure a
7
ꢀ
2−
2
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 5
D a l t o n T r a n s . , 2 0 0 5 , 2 7 1 3 – 2 7 1 9
2 7 1 3

View Article Online
The ligand H
3
EPTPA-BAA was obtained by conversion of
14
15
H
5
EPTPA into the bis-anhydride followed by reaction with
two equivalents of amylamine.
The strategy used to synthesize H DO3A-Nprop in the
4
16
literature was to alkylate DO3A-tris(t-butylester) with 3-
bromopropionic acid. Our approach, as for H DTTA-Nprop,
5
involved t-butyl esters. However, purification of the intermediate
and the final product was not obvious. We obtained the
ligand H DO3A-Nprop by repeated additions of t-butyl 3-
4
17
bromopropionate (prepared from 3-bromopropionic acid ) to
commercially available DO3A-tris(t-butylester), and followed by
hydrolysis of the esters.
Equilibrium studies on H
3
EPTPA-BAA
pH-potentiometric titrations have been used to determine the
H
protonation constants of H
eqn. (1)), and the stability constant (KML) of its Gd complex
eqn. (2)).
3
EPTPA-BAA (K , as defined in
i
3+
(
[
i
H L]
H
Ki = [H
i = 0, 1, 2, 3 and 4
ML]
(1)
(2)
+
i−1] [H ]
[
K
ML = [
M] [L]
Scheme 1
3−
The protonation constants of EPTPA-BAA as calculated from
the titration data (81 data points; Fig. 1) are given in Table 1.
H
5−
5−
smaller, and more optimal water exchange rate, however, its
drawback is the reduced thermodynamic (and very likely kinetic)
stability. It was previously reported that the substitution of a ter-
For comparison, logK values of DTPA , DTTA-Nprop and
i
5
−
EPTPA are also presented. The protonation sequence has
1
been established by H NMR titrations for analogous ligands
5
−
5− 18
3− 19,20
5− 21
minal acetate of DTPA by a propionate group (DTTA-Nprop)
results in a smaller decrease of the thermodynamic stability for
like DTPA , DTPA-bis(amides)
or EPTPA , and a
3
−
similar protonation scheme is expected for the EPTPA-BAA .
The first three protonation steps take place predominantly on
the amine nitrogens. The first proton is attached to the central
nitrogen, while the second protonation step occurs on a terminal
amine, accompanied by the partial transfer of the first proton to
the other terminal nitrogen. Then the third proton is probably
shared by the central nitrogen and the central carboxylate, as
8,9
the lanthanide complexes. The monopropionate derivative of
4
−
the macrocyclic DOTA ligand can also be expected to form
lanthanide complexes of thermodynamic and kinetic stability
which is sufficiently high for biomedical applications.
III
Here we report the characterization of the Gd complexes
formed with the propionate derivatives of DTPA (DTTA-
Nprop) and DOTA (DO3A-Nprop) with regard to contrast
agent applications. Additionally, the bisamylamide derivative
5
−
3−
of EPTPA , EPTPA-BAA has been synthesized. The re-
5
−
placement of two carboxylates of EPTPA by amide functions
is expected to result in a ∼10-fold decrease of kex on the
III
Gd complex, as previously proved for amide derivatives of
1
0,11
DTPA- and DOTA-type ligands.
Based on a variable tem-
1
7
1
perature O NMR study, which was combined with H NMRD
2
−
and EPR measurements for [Gd(DTTA-Nprop)(H
2
O)] and
O)] , we have determined the parameters
characterizing water exchange and rotational dynamics on all
2
−
[
Gd(EPTPA-BAA)(H
2
III
3−
three Gd complexes. In the case of the novel EPTPA-BAA ,
we have also measured the protonation constants of the ligand
and the thermodynamic stability constant for the Gd complex.
III
Fig. 1 pH-potentiometric titration curves of the EPTPA-BAA ligand
3+
3+
in the absence (left curve) and presence of Gd (right curve; Gd /ligand
ratio 1 : 1). I = 0.1 M KCl, 25 C.
◦
Results and discussion
H
i
Table 1 Protonation constants (logK ) of various ligands and stability
3+
◦
Experimental
constants (logKML) of their Gd complexes (25 C). The values in
parenthesis correspond to one standard deviation
The synthesis of
5
H DTTA-Nprop, has been previously
13
reported. It was carried out in two steps using acidic alkylating
arms: reaction of diethylenetriamine with 3-chloropropionic
acid first, then with chloroacetate. Each step was followed by
long purification by cation exchange chromatography. Here
we present a new synthetic procedure involving t-butyl esters,
which yields products that are easier to purify. The first crucial
step is the monoalkylation of diethylenetriamine with t-butyl
a
b
c
e
Ligand
DTPA
DTTA-Nprop
EPTPA
EPTPA-BAA
+
L + H
10.41
8.37
4.09
2.51
2.04
22.50
1.80
9.64
8.86
4.52
3.54
2.79
19.74
3.74
10.60
8.92
5.12
2.80
9.08 (0.05)
5.76 (0.07)
3.57 (0.08)
2.22 (0.08)
+
LH + H
+
LH
LH
LH
2
3
4
+ H
+ H
+ H
+
+
3+
d
Gd + L
17.5
13.44(0.03)
+
3
-chloropropionate. By silica gel chromatography, we could
GdL + H
isolate the compound containing one esterified propionate arm.
Further alkylation of the amines with t-butyl bromoacetate,
followed by hydrolysis of the esters gave the ligand H
Nprop.
a
b
c
Ref. 12 I = 0.1 M (CH
3
)
4
NCl. Ref. 8 I = 0.1 M KNO
3
.
Ref. 21 I =
d
e
0
0
.1 M (CH
.1 M KCl.
3
)
4
NNO
3
.
Ref. 6 I = 0.1 M (CH
3
)
4
NCl. This work I =
5
DTTA-
2
7 1 4
D a l t o n T r a n s . , 2 0 0 5 , 2 7 1 3 – 2 7 1 9

View Article Online
5
−
III
in DTPA-bis(amide) derivatives. In comparison to DTPA ,
Gd concentration and to q (the value of the scalar coupling
constant, A/ꢀ, does not change much within the family of simi-
5
−
5−
DTTA-N-prop and EPTPA , all protonation constants of
3
−
H
i
III
17
EPTPA-BAA are significantly lower. This decrease in logK
lar Gd complexes). In the analysis of the O NMR, EPR and
NMRD data, some of the parameters were fixed to common and
brought by the replacement of two carboxylates by amide groups
may result from the formation of hydrogen bonds between amide
hydrogens and amine nitrogens or carboxylate oxygens.
physically meaningful values. For the distances we used rGdO
=
19,20
17
˚
˚
2.5 A (Gd electron spin and O nucleus distance), rGdH = 3.1 A
1
˚
The stability constant of Gd(EPTPA-BAA) was determined
by direct pH-potentiometry (from 50 data points at pH 1.9–
(Gd electron spin and H nucleus distance) and aGdH = 3.5 A
(closest approach of the bulk water protons). The longitudinal
3
+
17
4
.8). The titration curve in the presence of Gd is shown in
O relaxation is related to motions of the Gd-coordinated water
Fig. 1; the stability constant is given in Table 1. The formation
oxygen vector, while the proton relaxation is determined by
motions of the Gd-coordinated water proton vector. For the
ratio of the rotational correlation time of the Gd–Hwater and
Gd–Owater vectors, sRH/sRO , similar values have been found for
+
of protonated [GdH(EPTPA-BAA)(H O)] complex could not
2
be detected. Among the selection of ligands listed in Table 1,
which all similarly contain oxygen and nitrogen donor atoms,
3
−
3+
III
EPTPA-BAA forms the least stable complex with Gd . Due
to the propylene bridge between two amine nitrogens, EPTPA-
type complexes all contain one six-membered chelate ring, less
stable than a five-membered ring. This is the reason why EPTPA-
derivatives form lower stability complexes with metal ions than
various small molecular weight, monohydrated Gd complexes,
both by experimental studies and MD simulations (sRH/sRO
=
25
0.65 ± 0.2). This sRH/sRO ratio, within the given error, is
considered as a common value for the ratio of the two rotational
1
7
correlation times. In the simultaneous analysis of O NMR,
6
2−
the DTPA-analogues. In addition, the replacement of two
EPR and NMRD data for the [Gd(DTTA-Nprop)(H
[Gd(EPTPA-BAA)(H
to 0.65. The quadrupolar coupling constant for the bound water
2
O)] and
carboxylates with amide groups further destabilizes the complex,
as a consequence of the lower negative charge and reduced amine
basicity of amide derivatives.
2
O)] complexes, we fixed the sRH/sRO ratio
2
1/2
oxygen, v(1 + g /3) , was fitted and a value of 10.9 ± 2.5 and
.3 ± 1.6 MHz was found for [Gd(DTTA-Nprop)(H O)] and
[Gd(EPTPA-BAA)(H O)], respectively. The empirical constant
os that characterizes the outer sphere contribution to the
2
−
7
2
1
7
2
O NMR, EPR and NMRD measurements
The water exchange rate was determined for the Gd complexes
1
7
C
O
III
III
chemical shift, was fixed for all three Gd complexes studied
to 0.1. Proton relaxivities only above 5 MHz were included in
the fit, within the validity of the Redfield relaxation theory. The
5
−
of the ligands DTTA-Nprop , DO3A-Nprop and EPTPA-
3
−
17
BAA from a variable temperature O NMR study. Ad-
ditionally, variable temperature EPR spectra were recorded
and proton relaxation rates measured on aqueous solutions
2
limits of the Redfield theory have been previously discussed.
1
7
The experimental O NMR, EPR and NMRD data and the
2
−
2−
of [Gd(EPTPA-BAA)(H
2
O)] and [Gd(DTTA-Nprop)(H
2
O)]
fitted curves for [Gd(DTTA-Nprop)(H O)] are presented in
2
with the objective of assessing all parameters that describe water
exchange, rotation, electronic relaxation and proton relaxivity.
All available experimental data for a given Gd complex, i.e.
Fig. 2; for the two other complexes the figures can be found in the
ESI†. The most relevant parameters obtained in the fit are shown
III
298
in Tables 2 and 3. The diffusion constant, D , and its activation
GdH
the oxygen-17 chemical shifts (Dx
transverse (1/T2r) relaxation rates, the electronic relaxation
rates (1/T2e) and the longitudinal proton relaxivities (r , when
r
), longitudinal (1/T1r) and
17
Table 3 Parameters obtained from the fitting of O NMR data for
l
macrocyclic Gd complexes
III
1
7
measured) were analysed simultaneously. The O NMR data
−
on [Gd(DO3A-Nprop)(H
2
O)] were fitted to the Solomon–
a
b
c
Ligand
DOTA
DO3A-Nprop
TRITA
1
Bloembergen–Morgan equations. For the two other complexes,
where we had variable field EPR and variable temperature
2
98
7
−1
k
ex / 10 s
0.46
54.5
6.1 ± 1.9
39.2 ± 3.4
+35.6 ± 5.7
−3.3 ± 0.5
153 ± 12
22.4 ± 6.1
7.0
270
17.5
−24
−3.8
82
‡
−1
1
7
DH /kJ mol
DS /J mol
NMRD data in addition to the O relaxation rates and chemical
shifts, the electronic relaxation has been described by the Rast–
Borel theory, involving both transient and static zero-field-
‡
−1
−1
K
+65
6
−1
A/ꢀ/10 rad s
−4.0
100
20
298
s
RO /ps
2,22
splitting contributions. Equations used in the fit can be found
in the ESI†.
For all three Gd complexes we assumed one inner sphere
−1
E
R
/kJ mol
21.9
2
98
d
sv /ps
0.65
8.6
III
−1
d
d
E
v
2
/kJ mol
1.0
0.97 ± 0.08
20 −2
D /10
s
—
water molecule (q = 1). This is based (i) on the analogy
III
to previously studied, DTPA- or DOTA-type Gd -chelates
a
c
17
b
Ref. 23 From O NMR, NMRD and EPR data. Present work.
where the ligand possesses eight donor atoms, and (ii) on the
d
Ref. 5. Not obtained. Underlined parameters were fixed in the fit.
1
7
experimental O chemical shifts, which are proportional to the
17
III
Table 2 Parameters obtained from the simultaneous fitting of O NMR, NMRD and EPR data for various acyclic Gd complexes
a
b
ꢀ
c
d
e
d
Ligand
DTPA
DTPA-BMA
DTTA-N prop
DTTA-Nprop
EPTPA
EPTPA–BAA
2
ex
98
7
−1
k
/ 10 s
0.38
52.9
0.043
46.6
+18.9
−3.6
167
21.6
34
3.1
30.8
+2.0
−3.3
121
17.4
18
8.0 ± 1.5
19.5 ± 3.0
−28.2 ± 5.1
−3.4 ± 0.3
107 ± 16
20.0 ± 4.2
7.7 ± 0.5
0.8 ± 0.3
—
33
27.9
+11.0
−3.9
75
17.7
22.4
1
0.76
—
—
—
—
5.7 ± 0.8
28.2 ± 3.7
−2.2 ± 6.0
−3.5 ± 0.8
192 ± 18
19.0 ± 5.9
4.0 ± 0.3
1
‡
−1
DH /kJ mol
DS /J mol
‡
−1
−1
K
+54
6
−1
A/ꢀ/10 rad s
−3.9
115
20
2
98
s
RO /ps
−
1
E
v
R
/kJ mol
/ps
2
98
s
0.10
1.0
−
1
E
v
2
/kJ mol
9
1
20 −2
D /10
s
—
0.38
—
—
—
—
1.2
—
—
g
1.99
0
0.016
0.0280
0.48
1.99 ± 0.01
0.51 ± 0.02
0
1.99 ± 0.02
0.39 ± 0.03
0
10
10
10
−1
a
a
a
a
2
4
6
/10
/10
/10
s
s
s
—
—
—
—
−1
−1
−1
0
0
1
0
2T /10
s
—
0.48 ± 0.01
—
0.38 ± 0.01
a
b
c
17
d
e
17
Ref. 23. Ref. 24 Ref. 6, only from O NMR. Present work. Ref. 6 from O NMR and EPR data. Underlined values were fixed in the fit.
D a l t o n T r a n s . , 2 0 0 5 , 2 7 1 3 – 2 7 1 9
2 7 1 5

View Article Online
6
contribution of 1/T1e to 1/T2m was 6% maximum ). For an
intermediate water exchange rate, however, the contribution of
the electron spin relaxation to the observed transverse relaxation
1
7
rate in O NMR can be important, therefore independent
information from EPR is very helpful in determining exact
water exchange rates. It is also well-known that the traditional
Solomon–Bloembergen–Morgan theory has serious limitations
in describing the magnetic field dependence of the electron
spin relaxation rates. Therefore, it is particularly important that
2
−
in the case of [Gd(DTTA-Nprop)(H
BAA)(H O)] which both present intermediate water exchange
rates, we had independent information on electronic relaxation
2
O)] and [Gd(EPTPA-
2
1
7
from EPR data, and the combined analysis of EPR and
O
NMR (and NMRD) was performed by using an adequate theory
of electronic relaxation, capable of describing field dependences
in a large domain (the EPR data were obtained at B = 0.34 T
(
X-band), 1.25 T (Q-band) and 3.41 T (W-band), whereas the
1
7
O NMR was performed at 9.4 T). On the other hand, the water
−
exchange rate on [Gd(DO3A-Nprop)(H
only from O NMR data. Consequently, the realistic, physically
meaningful error on it is clearly higher; one can estimate it to be
2
O)] was calculated
1
7
∼
± 50%, in contrast to the statistical error of the fit as given in
Table 3.
17
Fig. 2 Temperature dependence of reduced O transverse (᭹), longi-
tudinal (ꢀ) relaxation rates (a), reduced chemical shifts (b) at 9.4 T;
Water exchange
◦
◦
◦
◦
proton relaxivities at 5.1 C (ꢁ), 25.0 C (᭹), 37.1 C (ꢀ) and 49.7 C
Tables 2 and 3 summarize water exchange parameters for a
selection of MRI-related, linear (DTPA-type) and macrocyclic
(DOTA-type) Gd complexes. Despite the structural similarity
(
ꢂ) (c); EPR peak-to-peak line widths (d) and the apparent g-factor (e)
at 9.4 GHz (᭹), 35 GHz (ꢀ) and 94 GHz (only one point: ꢁ-measured,
III
2−
ꢂ-calculated) for [Gd(DTTA-Nprop)(H2O)] . The curves represent the
III
of the ligands, the water exchange rates on their Gd complexes
fit to the experimental data points.
cover three orders of magnitude from the DTPA-bisamide to
−
10
2
5−
energy, EDGdH, were calculated to be (19 ± 2) × 10
s , (20 ± 2) × 10 m s and (23 ± 1) kJ mol , (30 ±
) kJ mol for [Gd(DTTA-Nprop)(H O)] and [Gd(EPTPA-
BAA)(H O)], respectively.
m
EPTPA . The origin of this remarkable variation is the differing
−
1
−10
2
−1
−1
steric compression that the ligand induces around the water
binding site. By varying the distance between the donor atoms
that coordinate to the metal, one can tune the steric crowding in
−
1
2−
3
2
2
27
III
The values of the rotational correlation time obtained from the
the inner coordination sphere. All these nine-coordinate Gd
1
7
−
O longitudinal relaxation rates for [Gd(DO3A-Nprop)(H
2
O)]
complexes undergo a dissociatively activated water exchange, for
which steric crowding is of primary importance. The leaving of
the coordinated water molecule, which is the rate-determining
step, is largely facilitated by an increased steric crowding.
Unfortunately, when one ethylene is replaced by a propy-
lene bridge in the amine skeleton of the ligand, the ther-
modynamic and kinetic stability of the complex is reduced.
1
7
1
and from the O and H longitudinal relaxation rates for
2
−
[
Gd(DTTA-Nprop)(H
2
O)] and [Gd(EPTPA-BAA)(H O)] are
2
in the usual range expected for small molecular weight
2
98
chelates. The s calculated for the bisamylamide [Gd(EPTPA-
RO
BAA)(H
2
O)] complex is only slightly higher than that of
O)] which shows that—as expected—the
[
Gd(DTPA-BMA)(H
2
2
−
five-carbon chain is still too short to promote micellar aggrega-
tion in aqueous solution that would considerably slow down the
rotational motion.
Moreover, the water exchange rate on [Gd(EPTPA)(H
and [Gd(TRITA)(H
2
O)]
−
2
O)] already exceeds the optimal value.
Both problems can be resolved if the steric compression is
introduced by elongation of a carboxylate arm. [Gd(DTTA-
III
It has to be noted that all three Gd complexes studied here
2
−
−
present “intermediate” water exchange rates. This means that
Nprop)(H
2
O)] and [Gd(DO3A-Nprop)(H O)] have interme-
2
the slow exchange regime is not visible in the ln(1/T2r) vs.
diate water exchange rates, which fall exactly in the optimal
range. It is interesting to observe in the DTPA-family that
the position of the propionate arm also has an influence
on the water exchange rate; the complex with the terminal
propionate exchanges its inner sphere water 2.6 times faster
than the analogous chelate with a central propionate. This
structural difference also has a significant consequence on the
1000/T curves, on the other hand, we are far from the “very
fast” exchange, where the electron spin relaxation has negligible
1
7
contribution to the experimentally measured O transverse
III
relaxation rates. For a Gd complex with slow water exchange,
1
7
at low temperatures the reduced transverse O relaxation rates,
/T2r, increase with increasing temperature (slow exchange
1
III
regime). Under such conditions, 1/T2r is determined exclusively
thermodynamic stability of the Gd complex: for the central
by the water exchange rate, and electronic relaxation does not
contribute to the experimental O relaxation rates. Therefore,
propionate the stability constant is three orders of magnitude
1
7
lower than that for the terminal propionate (logKGdL = 16.7 vs.
8,9
even in the lack of EPR data, kex can be determined with high
19.7).
1
7
exactitude. On the other hand, the reduced transverse O relax-
ation rates can also decrease with increasing temperature (fast
exchange regime). Here they are influenced by the relaxation rate
of the coordinated water oxygen, 1/T2m. 1/T2m is determined by
both the rate of water exchange and the rate of electron spin
relaxation. The faster the water exchange, the less influence
the electron spin relaxation has. In the situation of a “very
fast” water exchange, electron spin relaxation has no, or only
a negligible role in determining the experimentally measured
The thermodynamic stability constant has not been reported
−
for [Gd(DO3A-Nprop)(H
2
O)] , however, we can expect that
−
28
it is higher than logKGdL = 19.17 for [Gd(TRITA)(H
2
O)] .
The substitution of an acetate with a propionate arm in
2
−
[Gd(DTPA)(H O)] decreases the stability constant of the
2
complex by 2.7 logK units (Table 1), thus for [Gd(DO3A-
−
Nprop)(H
2
O)] one can expect logKGdL ≈ 22 (logKGdDOTA
=
29
24.7 ). Beside the thermodynamic stability, kinetic inertness is
also very important for biomedical applications of metal com-
1
7
2+
298
ex
2−
O 1/T
2
9
rates, as was typically the case for Eu(H
2
O)
7
(k
=
plexes. No data are available for the linear [Gd(EPTPA)(H
2
O)] .
−1 26
2−
298
8
−1
−
5
.0 × 10 s ) or [Gd(EPTPA)(H
2
O)] (kex = 3.3 × 10 s ;
For [Gd(TRITA)(H
2
O)] , it has been recently proved that
2
7 1 6
D a l t o n T r a n s . , 2 0 0 5 , 2 7 1 3 – 2 7 1 9

View Article Online
the kinetic inertness, characterized by the dissociation half-
using CH
(26%) of compound 1. d
6.5 Hz, 2H), 2.90 (t, J 5.7 Hz, 2H), 2.83 (s, 4H), 2.78 (t, J 5.9 Hz,
2H), 2.52 (t, J 6.5 Hz, 2H), 1.54 (s, 9H); d /ppm (400 MHz,
CDCl ) 172.4, 80.6, 52.6, 49.3, 45.3, 41.9, 36.0, 28.3.
2
Cl
2
/MeOH/NH
3
(50 : 25 : 5) as eluent afforded 0.72 g
III
live of the Gd complex, is three orders of magnitude lower
H
/ppm (400 MHz, CDCl ) 2.94 (t, J
3
−
30
than for [Gd(DOTA)(H
2
O)]
(though still higher than for
2
−
[
Gd(DTPA)(H
2
O)] , the most widely used MRI contrast
C
agent). Similarly to the thermodynamic stability, we can as-
sume a higher kinetic inertness for the propionate complexes
3
Synthesis of compound 2. To a solution of 1 (0.55 g,
.38 mmol) in DMF (12 ml) was added t-butyl bromoacetate
2
−
[
Gd(DTTA-Nprop)(H
2
O)] and more particularly [Gd(DO3A-
2
(
−
2−
Nprop)(H
[
2
O)] as compared to [Gd(EPTPA)(H
2
O)]
or
1.92 g, 9.88 mmol) and K
2
CO (1.38 g, 10.0 mmol) and the
3
−
Gd(TRITA)(H
2
O)] , and this kinetic inertness would certainly
resulting mixture was stirred overnight at room temperature.
After filtration and evaporation of the solvent, the crude product
fulfil the requirements of safe in vivo application.
The water exchange rate on [Gd(EPTPA-BAA)(H O)] is one
2
was extracted with H
phase was dried with Na
Purification through a silica gel column using EtOAc/heptane
2
O (9 ml) and CHCl
3
(16 ml). The organic
2
−
order of magnitude lower than that on [Gd(EPTPA)(H O)] .
2
2
SO and evaporated to dryness.
4
A decrease in kex has been previously observed for all amide
derivatives with respect to the carboxylate parent complexes.
The magnitude of this decrease on replacing one carboxylate
by an amide function was found to be little dependent of the
ligand structure: it was a factor of 3–4 per each carboxylate
replaced in both DTPA- or DOTA-type ligands. Thus the value
(
(
(
1/1) as eluent afforded 1.3 g of compound 2 (yield 80%). m/z
+
MS-ESI) 689 [M + H] ; d
H
/ ppm (400 MHz, CDCl
3
) 3.43
4H, s), 3.32 (2H, s), 3.26 (2H, s), 2.89 (2H, dd, J 7.2 Hz, J
7
7
1
5
.5 Hz), 2.77 (4H, m), 2.70 (4H, s), 2.34 (2H, dd, J 7.2 Hz, J
.5 Hz), 1.43 (36H, s), 1.41 (9H, s);d / ppm (400 MHz, CDCl
72.0, 171.1, 170.9, 170.8, 81.0, 80.8, 80.3, 56.3, 56.0, 53.2, 53.0,
2.6, 52.5, 50.5, 34.8, 28.3, 28.2.
C
3
)
7
−1
of 5.7 × 10 s for [Gd(EPTPA-BAA)(H
3 × 10 s for [Gd(EPTPA)(H O)] follows well this empirical
observation.
2
O)] as compared to
7
−1
2−
3
2
In conclusion, we have demonstrated that by minor, appropri-
ate changes in the ligand structure one can fine-tune the steric
compression around the water binding site in nine-coordinate,
Synthesis of H DTTA-Nprop. A solution of 2 (1.27 g,
5
1.86 mmol) in TFA (20 ml) was stirred at room temper-
ature for 18 h. The light yellow solution was filtered and
evaporated to dryness. The residue was dissolved in water
(50 ml) and evaporated. This operation was repeated twice.
III
monohydrated Gd poly(amino carboxylate) complexes, which
translates to the fine-tuning of the rate of water exchange.
The introduction of one six-membered chelate ring in the
complex always gives rise to an increased steric crowding. The
steric crowding and the consequent acceleration of the water
exchange is more important on the elongation of the amine
Et O (50 ml) was added and then evaporated. This operation
2
was also repeated twice. A beige powder was obtained (1.0 g,
92%). (Found: C, 36.80; H, 4.65; N, 6.78; F, 14.22. Calc.
For C15H25N3O10(CF
N, 7.15; F, 14.55%); m/z (MS-ESI) 408 [M + H] ; d
(400 MHz, D O) 3.12 (2H, t, J 6.9 Hz), 3.35 (2H, t, J 6.1 Hz),
3.40 (2H, t, J 6.6 Hz), 3.67–3.72 (4H, m), 3.479–3.82 (4H, m),
4.27 (2H, s), 4.30 (4H, s). d /ppm (400 MHz, D O) 174.3, 173.9,
169.3, 168.7, 55.5, 54.5, 53.9, 53.5, 52.7, 50.9, 49.7, 49.0, 28.3.
3
COOH)1.5(H
2
O)0.5: C, 36.80; H, 4.72;
5
−
4−
+
backbone (EPTPA , TRITA ) than on the elongation of the
H
/ppm
5
−
4−
carboxylate pending arm (DTTA-Nprop , DO3A-Nprop ).
For the propionate derivative ligands DTTA-Nprop , DO3A-
Nprop we observe a moderate, (optimal) water exchange rate.
This, together with the limited reduction of the stability of
the Gd complexes makes the ligands DTTA-Nprop and
2
5
−
4
−
C
2
III
5−
4
−
DO3A-Nprop prime candidates for the development of high-
relaxivity, macromolecular MRI contrast agents.
3
H EPTPA-BAA
Experimental
The synthetic procedures of compounds H
5
DTTA-Nprop,
H
3
EPTPA-BAA and H DO3A-Nprop are summarized in
4
Schemes 2, 3 and 4. Commercially available chemicals were
reagent grade and were used without further purification. t-Butyl
3
-chloropropionate was synthesized according to a published
3
1
1
4
procedure.
5
H EPTPA was obtained following the literature.
DO3A tris(t-butylester) was purchased from Macrocyclics.
Scheme 3
Synthesis of the ligands
H
5
DTTA-Nprop
Synthesis of compound 4. To a suspension of H
(1.344 g, 3.30 mmol) in pyridine (2 ml) was added acetic
5
EPTPA
3
anhydride in three portions (2.9 ml, 30.88 mmol). The starting
compound dissolved progressively. The solution was heated at
◦
5
0 C for 1 h then stirred at room temperature for 3 h. Et
2
O
(
30 ml) was added, the precipitate was washed with Et O and
2
dried under vacuum to afford a white and yellow crispy solid 4
which was used without further purification.
Synthesis of H EPTPA-BAA. To a solution of EPTPA-
3
bis(anhydride) 4 (1.22 g, 3.30 mmol) in DMF (50 ml) was
added amylamine (0.75 ml, 6.60 mmol) in three portions. The
mixture was stirred for 1 h at 40 C. After cooling, the solvent
◦
Scheme 2
was evaporated. The resulting yellow oil was dissolved in H
2
O
Synthesis of compound 1. t-Butyl 3-chloropropionate (2 g,
2.15 mmol) was added to a refluxing solution of diethy-
lene triamine (3.9 ml, 36.29 mmol) and NaOH (0.486 mg,
(15 ml) and Et O (15 ml). This operation was repeated twice. The
resulting yellow powder was dissolved in hot EtOH (80 ml) and
heptane (5 ml) was added. A white powder precipitated during
2
1
1
2
2.15 mmol) in water (3.6 ml) and reflux was continued for
5 min. After cooling, the resulting yellow solution was extracted
cooling, it was collected by filtration, washed with Et
2
O and
dried in vacuo affording 0.99 g (global yield 55%) of H EPTPA-
3
with CH
2
Cl
2
(10 ml). Purification by silica gel chromatography
BAA as a white powder. (Found: C, 54.57; H, 8.82; N, 12.67; O,
D a l t o n T r a n s . , 2 0 0 5 , 2 7 1 3 – 2 7 1 9
2 7 1 7

View Article Online
2
2.82. Calc. For C25H47N5O8(H
2
O)0.25: C, 54.57; H, 8.70; N,
O) 4.14 (2H, s), 3.90
anion-exchange chromatography column (Bio-Rad AG 1-X4,
converted into HCOO form, 6 ml). The column was washed
−
1
2.73; F, 23.99%); d /ppm (400 MHz, D
H
2
(
2H, s), 3.83 (2H, s), 3.53 (2H, s), 3.51 (2H, s), 3.44–3.35 (6H,
m), 3.28–3.24 (4H, m), 3.18–3.16 (2H, m), 2.24 (m, 2H), 1.56–
.49 (4H, m), 1.32–1.26 (m, 8H), 0.90–0.86 (6H, m). d /ppm
400 MHz, D O) 175.5, 171.5, 170.4, 169.9, 165.2, 58.4, 57.2,
6.7, 55.9, 55.7, 53.6, 53.4, 52.8, 50.8, 39.8, 39.4, 28.5, 28.4,
8.2, 27.9, 26.5, 21.8, 21.7, 19.7, 13.4, 13.3.
with water until the pH of the eluate was neutral and the product
was eluted with a 0.1–0.3 M gradient of HCOOH (total volume
of gradient ≈80 ml). The fractions containing the product
were evaporated and the anion-exchange chromatography was
repeated twice until the product could be isolated. After removal
1
(
5
2
C
2
of the solvents and drying under reduced pressure, H DO3A-
4
Nprop was obtained as a white powder (20 mg, 14%). m/z (MS-
ESI) 419 [M + H] .
+
H
4
DO3A-Nprop
Preparation of the stock solutions
The stock solution of Gd(ClO
ing Gd in a slight excess of HClO
in double distilled water. Its concentration was determined
by complexometric titration with standardized Na EDTA
solution using xylenol orange as indicator (H
edta = ethylene-
diaminetetraacetic acid). The solutions of the GdL complexes
were prepared by mixing equimolar amounts of Gd(ClO and
the ligand. A slight ligand excess (5%) was used and the pH
was adjusted to about 5–5.5 by adding 0.1 M HClO or 0.1 M
4
)
3
was prepared by dissolv-
2
O
3
4
(Merck p.a. 60%)
2
H
2
4
)
4 3
4
NaOH. The absence of free metal was checked in each sample
by the xylenol orange test.
1
7
17
17
In the O NMR samples, O enriched water (Izotec, O:
1
1.4%) was used (final enrichment ∼1–2%) to improve sensitivity
and the pH was checked again. The concentration and pH of
the samples were the following: [Gd(DTTA-Nprop)(H
2
−
2
O)] :
Scheme 4
−
2
−1
17
−3
4
.510 × 10 mol kg , pH = 5.4 ( O NMR), 9.931 × 10 M,
−
2
pH = 5.6 (NMRD), 4.942 × 10 M, pH = 5.3 (EPR);
−
2
−1
Synthesis of compound 5.
added to a stirred suspension of anhydrous MgSO
5.36 mmol) in CH Cl (20 ml) under argon. The resulting
suspension was stirred for 15 min then 3-bromopropionic acid
2.5 g, 16.34 mmol) and t-BuOH (7.48 ml, 81.70 mmol) were
added. The mixture was stirred at room temperature for 18 h.
A saturated aqueous solution of NaHCO (100 ml) was slowly
added and the mixture was stirred until dissolution of MgSO
The product was extracted with CH Cl
(2 × 50 ml). The organic
phase was washed with H O (40 ml), dried with MgSO and
evaporated to dryness. A colorless liquid 5 was obtained (1.2 g,
5%). d /ppm (400 MHz, CDCl ) 3.54 (2H, t, J 13.7 Hz), 2.81
2H, t, J 13.7 Hz), 1.46 (9H, s).
H
2
SO
4
(0.92 ml, 16.34 mmol) was
[
(
Gd(EPTPA-BAA)(H
2
O)]: 3.870 × 10 mol kg , pH = 5.5
1
7
−2
4
(7.87 g,
O NMR), 1.017 × 10 M, pH = 5.6 (NMRD), 4.348 ×
−
−
2
2
−
6
2
2
1
1
0
0
M, pH = 5.5 (EPR); [Gd(DO3A-Nprop)(H
2
O)] : 2.420 ×
−1
17
mol kg , pH = 5.8 ( O NMR).
(
Equilibrium measurements
The protonation constants of EPTPA-BAA and the stability
3
3
−
4
.
3
+
constant with Gd were determined by pH-potentiometric
2
2
◦
titration at 25 C in 0.1 M KCl. A Metrohm Dosimat 665
2
4
automatic burette, a combined glass electrode (C14/02-SC,
reference electrode Ag/AgCl in 3 M KCl, Møeller Scientific
Glass Instruments, Switzerland) and a Metrohm 692 pH/ion-
meter were used for the titrations. The samples (3 ml) were stirred
3
(
H
3
Synthesis of H DO3A-Nprop. To a solution of DO3A tris(t-
butylester) (180 mg, 0.35 mmol) in acetonitrile (3.5 ml) was
4
and N
2
was bubbled through the solutions.
3−
The concentration of EPTPA-BAA (2 mM) was determined
using the titration curves obtained in the presence and absence
added Cs
2
CO
3
(285 mg, 0.87 mmol) and 5 (73.2 mg, 0.35 mmol)
◦
under argon. The resulting suspension was heated at 60 C for
of an excess of CaCl
protons of the ligand dissociate).
2
(cCa/c
L
≈ 40, when all the dissociable
2
days. After evaporation of the solvent, the crude product was
dissolved in water (10 ml) and extracted with CH Cl
(3 × 10 ml).
The organic phase was dried with MgSO and evaporated to
2
2
Protonation constants were determined in 2 mM EPTPA-
4
3−
BAA solutions titrated with standardized KOH solution
dryness. The ES-MS spectra showed the presence of a small pic
of the desired compound. The reaction and the work-up were
repeated twice (with addition of 5). The mass spectra showed
that the pic of the final compound was higher than the starting
material and revealed also the presence of secondary products. A
3+
(
0.05 M). In the EPTPA-BAA-Gd system, the ligand and
+
metal concentrations were both 2 mM. The H concentration
was obtained from the measured pH values using the correction
method proposed by Irving et al. The PSEQUAD program was
32
33
used to calculate the protonation and stability constants.
SiO
2
chromatography eluted with CH
2
2
Cl /MeOH (9/1) allowed
us to obtain 94 mg of the tetra ester in the presence only of the
starting material. This mixture was not purified further and used
for the next step. It was dissolved in trifluoroacetic acid (10 ml)
and stirred at room temperature for 24 h. After evaporation
to dryness the compound was redissolved in water (12 ml) and
1
7
O NMR measurements
1
7
Longitudinal and transverse O relaxation rates and chemical
shifts were measured between 277 and 371 K. The measurements
were performed using a Bruker ARX-400 (9.4 T, 54.2 MHz)
spectrometer. A Bruker VT-1000 temperature control unit
was used to maintain constant temperature, measured by
a substitution technique. The samples were sealed in glass
spheres, adapted to 10-mm NMR tubes, in order to eliminate
susceptibility corrections to the chemical shifts. A HClO
solution (pH = 3.0) was used as external reference. Longitudinal
and transverse relaxation rates, 1/T and 1/T , were obtained
evaporated. This operation was repeated twice. Et O (12 ml) was
2
then added. Evaporation to dryness gave a yellow powder which
was dissolved in a minimum of water and loaded onto a cation-
+
exchange chromatography column (Bio-Rad AG 50W-X4, H
34
form, 6 ml). The column was washed with water until the pH of
the eluate was neutral and the product mixture was eluted in one
4
fraction with NH
3
(0.5 M, 50 ml). This fraction was evaporated
1
2
to dryness, dissolved in water (2 ml) and NaOH 1 M was added
until the pH reached 12. This solution was loaded onto an
by the inversion recovery and the Carr–Purcell–Meiboom–Gill
spin echo technique, respectively.
2
7 1 8
D a l t o n T r a n s . , 2 0 0 5 , 2 7 1 3 – 2 7 1 9

View Article Online
NMRD
6 S. Laus, R. Ruloff, E´ . T o´ th and A. E. Merbach, Chem. Eur. J., 2003,
9
, 3555.
1
The longitudinal H relaxation rates (NMRD profiles) were
obtained at 278, 298, 310, 323 K on a Stelar Spinmaster FFC
fast field cycling NMR relaxometer equipped with a VTC90
´
7
S. Laus, A. Sour, R. Ruloff, E. T o´ th and A. E. Merbach, Chem.
Eur. J., 2005, 11, 3064.
8 D. J. Sawyer and J. E. Powell, Polyhedron, 1989, 8, 1425.
9 V. F. Vasileva, O. Y. Lavrova, N. M. Dyatlova and V. G. Yashunskii,
Zh. Obshch. Khim., 1966, 36, 674.
−
4
temperature control unit (Stelar, Italy) (2 × 10 − 0.47 T,
(
corresponding to a proton Larmor frequency range 0.01–
1
10 G. Gonzalez, D. H. Powell, V. Tissi e` res and A. E. Merbach, J. Phys.
Chem., 1994, 98, 53.
2
0 MHz). At higher fields, the longitudinal H relaxation times
were measured on Bruker Minispecs mq30 (30 MHz), mq40
40 MHz) and mq60 (60 MHz) and on Bruker 50 MHz (1.18 T),
00 MHz (2.35 T) and 200 MHz (4.70 T) cryomagnets connected
´
1
1 E. T o´ th, D. Pubanz, S. Vauthey, L. Helm and A. E. Merbach, Chem.
(
1
Eur. J., 1996, 2, 1607.
12 IUPAC Stability Constants, Academic Software and K. J. Powell,
Release 1.05, Yorks., UK LS21 2PW, 1999.
to a Bruker AC-200 console. In each case, the temperature was
measured by a substitution technique.
1
1
1
1
3 D. J. Sawyer, J. E. Powell and H. R. Burkholder, J. Chromatogr., 1988,
55, 193.
4
4 Y.-M. Wang, C.-H. Lee, G.-C. Liu and R.-S. Sheu, J. Chem. Soc.,
Dalton Trans., 1998, 4113.
EPR
5 C. F. G. C. Geraldes, R. Delgado, A. M. Urbano, J. Costa, F. Jasanada
and F. Nepveu, J. Chem. Soc., Dalton Trans., 1995, (3), 327.
6 D. A. Keire, Y. H. Jang, L. Li, S. Dasgupta, W. A. Goddard III and
J. E. Shively, Inorg. Chem., 2001, 40, 4310.
The EPR spectra were recorded on a Bruker ElexSys spectrom-
eter E500 at X-band (9.4 GHz) (278, 298, 310 and 329 K),
Q-band (34.6 GHz) (277, 299, 312 and 322 K) and a Bruker
ElexSys E680 spectrometer at W-band (94.2 GHz) (286.7 K). A
controlled nitrogen gas flow was used to maintain a constant
temperature, measured by a substitution technique. The peak-
to-peak line width was obtained from the experimental spectra
using the MATLAB program.
17 D. Felder, D. Guillon, R. Levy, A. Mathis, J. F. Nicoud, J. F.
Nierengarten, J. L. Rehspringer and J. Schell, J. Mater. Chem., 2000,
1
0, 887.
1
1
8 P. Letkeman and A. E. Martell, Inorg. Chem., 1979, 18, 1284.
9 H. Imura, G. R. Choppin, W. P. Cacheris, L. A. de Learie, T. J. Dunn
and D. H. White, Inorg. Chim. Acta, 1997, 258, 227.
0 C. F. G. C. Geraldes, A. M. Urbano, M. C. Alpoim, A. D. Sherry,
K.-T. Kuan, R. Rajagopalan, F. Maton and R. N. Muller, Magn.
Reson. Imaging, 1995, 13, 401.
2
Data analysis
1
7
The analysis of O NMR and EPR data was performed either
with a program working on a MATLAB platform version 5.3
or with Scientist for Windows
2
1 Y.-M. Wang, C. H. Lee, G.-C. Liu and R.-S. Sheu, J. Chem. Soc.,
Dalton Trans., 1998, 4113.
22 A. Borel, L. Helm, A. E. Merbach, Molecular Dynamics of Gd(III)
35
ꢀ
R
TM
ꢀR
by Micromath , version
Complexes in Aqueous Solution by HF EPR in Very High Frequency
2
.0. The reported errors correspond to one standard deviation
(
VHF) ESR/EPR Biological Magnetic Resonance, ed. L. J. Berliner,
obtained by the statistical analysis.
Plenum Press, NY, USA, 2004, ch. 6, vol. 22, pp. 207–247.
´
2
2
3 L. Burai, E. T o´ th, H. Bazin, M. Benmelouka, Z. J a´ szber e´ nyi, L. and
Acknowledgements
A. E. Merbach, Dalton Trans., submitted.
4 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.
We thank Dr Robert Ruloff for scientific advice. This research
was financially supported by the Swiss National Science Foun-
dation and the Office for Education and Science (OFES). This
work was carried out in the frame of the EC COST Action D18
and the European-founded EMIL programme (LSHC-2004–
25 F. Yerly, K. I. Hardcastle, L. Helm, S. Aime, M. Botta and A. E.
Merbach, Chem. Eur. J., 2002, 8, 1031.
´
2
2
6 P. Caravan, E. T o´ th, A. Rockenbauer and A. E. Merbach, J. Am.
Chem. Soc., 1999, 121, 10403.
5
03569).
7 J. P. Dubost, J. M. Leger, M. H. Langlois, D. Meyer and M. Schaefer,
C. R. Acad. Sci. Paris, 1991, 312, 349.
2
2
8 E. T. Clarke and A. E. Martell, Inorg. Chim. Acta, 1991, 190, 37.
9 W. P. Cacheris, S. K. Nickle and A. D. Sherry, Inorg. Chem., 1987,
26, 958.
References
´
1
2
E. T o´ th, L. Helm, A. E. Merbach, Relaxivity of Gadolinium(III)
Complexes: Theory and Mechanism in The Chemistry of Contrast
30 E. Balogh, R. Tripier, R. Ruloff and E´ . T o´ th, Dalton Trans., 2005,
´
Agents in Medical Magnetic Resonance Imaging, ed. E. T o´ th, and
1058–1065.
A. E. Merbach, Wiley, Chichester, UK, 2001, pp. 45–120.
31 S. Pavlov, M. Bogavac and V. Arsenijevic, Bull. Soc. Chim. Fr., 1974,
(a) S. Rast, A. Borel, L. Helm, E. Belorizky, P. H. Fries and A. E.
Merbach, J. Am. Chem. Soc., 2001, 123, 2637; (b) S. Rast, P. H. Fries,
E. Belorizky, A. Borel, L. Helm and A. E. Merbach, J. Chem. Phys.,
12, 2985.
32 H. M. Irving, M. G. Miles and L. Pettit, Anal. Chim. Acta, 1967, 28,
475.
2
001, 115, 7554.
33 L. Z e´ k a´ ny, I. Nagyp a´ l, in Computation Methods for Determination
of Formation Constants, ed. D. J. Leggett, Plenum, NY, USA, 1985,
p. 291.
3
4
S. M. Cohen, J. Xu, E. Radkov, K. N. Raymond, M. Botta, A. Barge
and S. Aime, Inorg. Chem., 2000, 39, 5747.
S. Aime, A. Barge, M. Botta, J. A. Howard, R. Kataky, M. P. Lowe,
J. M. Moloney, D. Parker and A. S. de Sousa, Chem. Commun., 1999,
1
34 A. D. Hugi, L. Helm and A. E. Merbach, Helv. Chim. Acta, 1985,
68, 508.
047.
35 F. Yerly, VISUALISEUR 2.2.4. and OPTIMISEUR 2.2.4., 1999,
Institute of Molecular and Biological Chemistry, University of
Lausanne, Switzerland.
´
5
R. Ruloff, E. T o´ th, R. Scopelliti, R. Tripier, H. Handle and A. E.
Merbach, Chem. Commun., 2002, 2630–2631.
D a l t o n T r a n s . , 2 0 0 5 , 2 7 1 3 – 2 7 1 9
2 7 1 9