New Bifunctional Contrast Agents: Bis-Amide Derivatives of C-Substituted Gd-DTPA

Article abstract of DOI:10.1002/ejic.200300457

In order to expand upon and confirm previous observations on the positive and adverse roles played by the functionalization and substitution of the parent MRI contrast agent Gd-DTPA, two new bi-functionalized contrast agents namely Gd-(S)-EOB-DTPA-BMA and Gd-(S)-C₄Bz-DTPA-BMA have been synthesized, purified and characterized by multinuclear NMR studies. Water 1H nuclear magnetic relaxation dispersion profiles were recorded at several temperatures. τ_R was determined by 2H NMR on the diamagnetic lanthanum complex. τ_M, the water residence time, was obtained by 17O NMR measurements and the stability versus transmetallation by zinc was evaluated by proton relaxometry. The residence time of water molecules in the first coordination sphere of the Gd complexes has confirmed the antagonistic effects of amide functions and C-4 substitution on the reduction of the water residence time, a critical parameter of relaxivity. This C-4 position is thus particularly appropriate for the substitution required by the design of new contrast agents covalently bound to macromolecules or carrying specific vectors.

Full text of DOI:10.1002/ejic.200300457

FULL PAPER
New Bifunctional Contrast Agents: Bis-Amide Derivatives of C-Substituted
Gd-DTPA
Sophie Laurent,[a] Francois Botteman,[a] Luce Vander Elst,[a] and Robert N. Muller*[a]
¸
Keywords: Gadolinium / Zinc / Metallation / Amides
In order to expand upon and confirm previous observations
on the positive and adverse roles played by the functionaliz-
by zinc was evaluated by proton relaxometry. The residence
time of water molecules in the first coordination sphere of
ation and substitution of the parent MRI contrast agent Gd- the Gd complexes has confirmed the antagonistic effects of
DTPA, two new bi-functionalized contrast agents namely
Gd-(S)-EOB-DTPA-BMA and Gd-(S)-C4Bz-DTPA-BMA have
been synthesized, purified, and characterized by multinu-
amide functions and C-4 substitution on the reduction of the
water residence time, a critical parameter of relaxivity. This
C-4 position is thus particularly appropriate for the substitu-
clear NMR studies. Water 1H nuclear magnetic relaxation tion required by the design of new contrast agents covalently
dispersion profiles were recorded at several temperatures. τR
was determined by 2H NMR on the diamagnetic lanthanum
complex. τM, the water residence time, was obtained by 17O
NMR measurements and the stability versus transmetallation
bound to macromolecules or carrying specific vectors.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2004)
Introduction
compounds, both substituted on the ethylenic C-4 carbon.
Gd-(S)-EOB-DTPA-BMA (1) and Gd-(S)-C4Bz-DTPA-
BMA (2) (Figure 1) were synthesized and their gadolinium
complexes were characterized by: i) proton relaxation rate
measurements between 0.5 mT and 7.05 T (NMRD pro-
files), ii) estimation of their rotational correlation times
through the analysis of the deuterium relaxation rates of
their specifically labeled La-analogues, and iii) determi-
nation of the water residence times of their coordinated
water molecules through the analysis of the temperature de-
pendence of the 17O transverse relaxation rates of water.
Their stability versus transmetallation by Zn2ϩ, an import-
ant endogenous ion, was also studied. Finally, all these par-
ameters, compared with those of parent and analogous
compounds, are discussed in the context of the
structureϪrelaxivity relationship.
The efficacy of the gadolinium complexes currently used
as contrast agents in Magnetic Resonance Imaging results
partly from the presence of water molecules in the first co-
ordination sphere of the ion and from the exchange of these
coordinated molecules with the bulk. An exchange rate
which is too slow or too fast limits the efficacy of the con-
trast agent. In a previous report,[1] we have shown that Gd-
DTPA derivatives carrying benzyl groups are characterized
by water exchange rates which differ according to the posi-
tion of the group: the complex C-substituted on an ethy-
lenic carbon i.e. [Gd-(S)-C4Bz-DTPA] has a faster water ex-
change rate and an higher relaxivity, the bis(benzylamide)
(Gd-DTPA-BBzA) has a slightly higher proton relaxivity at
37 °C than the parent compound but the longer residence
time of its coordinated water molecules (τM) clearly limits
the relaxivity at low temperatures and, finally, the bis-ester
derivative (Gd-DTPA-BBzE) behaves like Gd-DTPA. An-
other C-substituted derivative, Gd-(S)-EOB-DTPA, was
also shown to be characterized by an increased water ex-
change rate compared with the parent compound.[2] On the
contrary, amide derivatives of Gd-DTPA or Gd-DOTA are
known to be characterized by slower water exchange rates
than those of their parent structures.[3Ϫ5]
In this work, the influence of the C-substitution on the
behavior of bis-amide derivatives was thus studied on two
[a]
NMR Laboratory, Department of Organic Chemistry,
University of Mons-Hainaut,
7000 Mons, Belgium,
Fax: (internat.) ϩ 32-65-373520
E-mail: robert.muller@umh.ac.be
Figure 1. Chemical structures of Gd-(S)-EOB-DTPA-BMA (1) and
Gd-(S)-C4Bz-DTPA-BMA (2)
Eur. J. Inorg. Chem. 2004, 463Ϫ468
DOI: 10.1002/ejic.200300457
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
463

S. Laurent, F. Botteman, L. Vander Elst, R. N. Muller
FULL PAPER
The longitudinal relaxivity, r1, arises from both short (in-
ner sphere) and long (outer sphere) distance magnetic inter-
actions. The inner sphere contribution can be described by
the SolomonϪBloembergen equations,[6Ϫ7] while the outer
sphere contribution can be accounted for by the Freed
model[8] (see ref.[2] for the description of the models). A
rather large number of structural parameters is included in
those models: q the number of water molecules in the first
coordination sphere, τM the residence time describing the
exchange between coordinated and bulk water, r the dis-
tance between coordinated water protons and the unpaired
electron spin, d the distance of closest approach, D the rela-
tive diffusion constant, τR the rotational correlation time of
the hydrated complex and τs1,2 the longitudinal and trans-
verse relaxation times of the electron. These latter param-
eters are field dependent with τSO being the value of τs1,2 at
zero field and τv the correlation time modulating the inter-
action responsible of the electronic relaxation.
Therefore, the analysis of the proton NMRD curves by a
multi-parametric fitting procedure takes advantage of the
knowledge of some of the parameters.[2,9] In previous
studies,[2,4,9Ϫ12] it has been shown that q, τR, and τM could
be evaluated by specific NMR techniques and reduce the
ambiguity of the analysis of the proton NMRD curves. By
analogy with similar complexes [Gd-(S)-EOB-DTPA (5),
Gd-DTPA-BMA (4) or Gd-DTPA (3)], q was assumed to
be equal to one and d and D310 were set to 0.36 nm and 3.3
ϫ 10Ϫ9 m2·sϪ1 respectively.
Results and Discussion
The bis-amide complexes 1 and 2 (Figure 1) were synthe-
sized by the classical method starting from the reaction of
the pentacarboxylated ligands with acetic anhydride in the
presence of pyridine to produce C-functionalized DTPA
bis-anhydrides. The subsequent amination was carried out
with an excess of methylamine.
NMRD profiles recorded at 310 K show that, in general,
the relaxivities (r1) of all the C-substituted derivatives are
higher than those of Gd-DTPA and Gd-DTPA-BMA (Fig-
ure 2). At low magnetic fields, the relaxivities of the bis-
amide derivatives of the C-substituted complexes are
slightly lower than those of their carboxylate analogues. At
higher fields, however, the relaxivities of the bis-amides 1
and 2 are similar to those of their parent compounds Gd-
(S)-EOB-DTPA (5) or Gd-(S)-C4Bz-DTPA (6) but still
markedly higher than those of the unsubstituted structures
Gd-DTPA (3) or Gd-DTPA-BMA (4). At 0.47 T (20 MHz)
and 310 K, the relaxivity values of Gd-(S)-EOB-DTPA (5),
Gd-(S)-EOB-DTPA-BMA (1), Gd-(S)-C4Bz-DTPA (6),
Gd-(S)-C4Bz-DTPA-BMA (2), Gd-DTPA (3), and Gd-
DTPA-BMA (4) are 5.5, 5.0, 4.8, 4.7, 3.9, and 3.8
s
Ϫ1·mϪ1 respectively.
The determination of τR was performed at five tempera-
tures (from 278 K to 318 K) on the diamagnetic lanthanum
complexes deuterated at the α positions of the carbonyl
groups[13] through the analysis of the deuterium longitudi-
nal rates.[2] At 310 K, the values of τR are fairly close {τR
[La-(S)-[D8]EOB-DTPA-BMA] ϭ 63 Ϯ 7 ps; τR [La-(S)-
[D10]EOB-DTPA] ϭ 66 Ϯ 7 ps; τR [La-(S)-[D8]C4Bz-
DTPA-BMA] ϭ 60 Ϯ 6 ps; τR [La-(S)-[D10]C4Bz-DTPA] ϭ
65 Ϯ 7 ps; τR [La-[D8]DTPA-BMA] ϭ 67 Ϯ 7 ps; and τR
[La-[D10]DTPA] ϭ 60 Ϯ 6 ps}. The activation energies for
the rotations calculated from the evolutions of τR with tem-
perature are comparable for all the complexes (ERϭ 20
Ϫ24 kJ/mol).
Among the two relaxation mechanisms, i.e. the outer
sphere and the inner sphere contributions, only the latter
is influenced by the water residence time and can thus be
quenched at low temperatures while the former always in-
creases when the temperature is lowered. Therefore, as long
as the water residence time is not limiting the propagation
of the relaxation (τM Ͻ T1M where T1M is the relaxation
time of the protons of the coordinated water molecule), the
global relaxivity increases when temperature decreases. If
this condition is not fulfilled, the global relaxivity may re-
ach a plateau or even decrease at low temperatures. The
possible influence of τM on the proton relaxivity was evalu-
ated at 0.47 T over a range of temperatures between 278 K
and 318 K (Figure 3). Like Gd-DTPA itself, its C-func-
tionalized derivatives exhibit a continuous increase in r1
when the temperature is lowered, revealing that τM does
not limit the relaxation rate. On the contrary, the bis-amide
Figure 2. 1H NMRD relaxivity profiles of Gd-(S)-C4Bz-DTPA-
BMA (2) and Gd-(S)-C4Bz-DTPA (6) in water at 310 K ([Gd com-
plex] ϭ 1 m) (top graph) and of Gd-(S)-EOB-DTPA-BMA (1)
and Gd-(S)-EOB-DTPA (5) in water at 310 K ([Gd complex] ϭ
1 m) (bottom graph). The fittings of the NMRD profiles of Gd-
DTPA (3) and Gd-DTPA-BMA (4) are shown for comparison. The
lines correspond to the theoretical fittings of the data points.
464
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 463Ϫ468

New Bifunctional Contrast Agents
FULL PAPER
complexes Gd-DTPA-BMA 4, Gd-(S)-C4Bz-DTPA-BMA Gd-(S)-C4Bz-DTPA-BMA (2) of 472
Ϯ 38 ns and
2, and Gd-(S)-EOB-DTPA-BMA 1 show a clear limitation 429Ϯ23 ns, respectively. These values are between those re-
of their relaxivity by the water exchange below 310 K.
ported for pentacarboxylate complexes {τM310 [Gd-DTPA
(3)] ϭ 143Ϯ25 ns; τM310 [Gd-(S)-EOB-DTPA (5)] ϭ 82 Ϯ
21 ns; τM310 [Gd-(S)-C4Bz-DTPA (6)] ϭ 86 Ϯ 8 ns} and the
bis-amide parent structure (τM310 [Gd-DTPA-BMA (4)] ϭ
967Ϯ36 ns) (Table 1). These results are in good agreement
with the temperature dependence of r1 at 0.47 T and they
confirm that the presence of amide functions on the struc-
ture of the DTPA ligand considerably decreases the ex-
change rate and that a conformational or steric effect, in-
duced by substitution on the C4 of the ligand, counteracts
this unfavorable effect.
The fittings of the proton NMRD profiles were per-
formed with the distance r fixed to the usual value of
0.31 nm for Gd-DTPA (3) and Gd-DTPA-BMA (4). For
the C-functionalized complexes, such a distance results in
τR values significantly larger than those obtained by deu-
terium relaxometry (τ3R10ϭ 112 ps, 91 ps, 109 ps, and 84 ps
for complexes 1, 2, 5, and 6 respectively). Shorter effective
distances (0.291 nm and 0.3 nm) were therefore used to fit
the data (Table 2). For complexes 1 and 5, τR values ob-
tained for the distance r equal to 0.291 nm are in better
agreement with the values obtained by deuterium relaxome-
try, whereas for complexes 2 and 6, both distances
(0.291 nm and 0.3 nm) can account for the experimental re-
sults. The fitted values of τSO and τV (Table 2) are in the
same range for all the complexes and agree with those ob-
tained by the fitting of the 17O data (Table 1). At low mag-
netic fields, the lower relaxivity of the bis-amide C-substi-
tuted complexes 1 and 2, compared with their parent com-
plexes 5 and 6, can be related to their longer τM values
which result in a more marked limitation of the relaxivity
at low rather than at high fields.
Figure 3. Temperature dependence of the proton relaxivities of the
various Gd complexes at 0.47 T.
A quantitative estimation of the residence times of water
molecules in the first coordination spheres of the gadolin-
ium complexes (τM) was obtained from the analysis of the
temperature dependence of the 17O transverse relaxation
rates of solutions containing 13Ϫ50 m of the gadolinium
complexes. Various parameters are involved in the theoreti-
cal fitting of the 17O data: A/h , the hyperfine coupling con-
-
stant between the oxygen nucleus of bound water molecules
and the Gd3ϩ ion; τV, the correlation time modulating the
electronic relaxation of Gd3ϩ; Ev, the activation energy re-
lated to τV; ∆2, the mean-square zero field splitting energy
(∆2 is related to τSO by τSO ϭ [12 ∆2τV]Ϫ1); and ∆H# and
∆S#, the enthalpy and entropy of the water exchange pro-
cess. The theoretical fitting of the transverse relaxation rates
obtained over a temperature range from 280 K to 350 K
gives values of τM310 for Gd-(S)-EOB-DTPA-BMA (1) and
The NMRD profiles of Gd-(S)-EOB-DTPA-BMA (1)
and Gd-(S)-C4Bz-DTPA-BMA (2) at 298 K and 310 K are
shown in Figure 4. The data recorded at 298 K are only
slightly greater than those obtained at 310 K, confirming
that τM is indeed limiting the relaxivity. The parameters ob-
tained from the theoretical fitting are shown in Table 2.
Table 1. Parameters obtained from the fitting of the reduced transverse relaxation rate of 17O
Gd-(S)-EOB-
DTPA-BMA (1)
Gd-(S)-C4Bz-
DTPA-BMA (2)
Gd-
DTPA (3)
Gd-
Gd-(S)-EOB-
DTPA (5)
Gd-(S)-C4
Bz-DTPA (6)
DTPA-BMA (4)
∆2 (1020 sϪ2
τV310 (ps)
)
0.98 Ϯ 0.10
16 Ϯ 2.0
2.99 Ϯ 0.23
0.92 Ϯ 0.03
16 Ϯ 0.4
2.80 Ϯ 0.20
1.08 Ϯ 0.03
23 Ϯ 0.3
3.4 Ϯ 0.1
0.85 Ϯ 0.03
16 Ϯ 0.6
3.16 Ϯ 0.04
1.05 Ϯ 0.12
17 Ϯ 2
4.1 Ϯ 0.5
1.17 Ϯ 0.03
10 Ϯ 0.3
3.19Ϯ 0.06
-
|A/h |
(106 rad sϪ1
)
∆H#
43.9 Ϯ 0.14
17.6 Ϯ 0.23
0.95 Ϯ 1.44
37.1 Ϯ 0.07
Ϫ3.54 Ϯ 0.23
0.90 Ϯ 12.4
51.5 Ϯ 0.3
52.1 Ϯ 0.6
4.5 Ϯ 4.2
48 Ϯ 0.05
24.9 Ϯ 0.2
15.0 Ϯ 10.9
53.5 Ϯ 0.3
63 Ϯ 1.3
7.1 Ϯ 5.9
50.1 Ϯ 0.13
51.8 Ϯ 0.4
8.7 Ϯ 3.9
(kJ/mol)
∆S#
(J/mol K)
EV
(kJ/mol)
τM310 (ns)
472Ϯ 38
54Ϯ27
429 Ϯ 23
58 Ϯ 4
143 Ϯ 25
67 Ϯ 11
967 Ϯ 36
56 Ϯ 4
82 Ϯ 21
58 Ϯ 14
86 Ϯ 8
69 Ϯ 4
310
τ
(ps)[a]
SO
[a]
310
Calculated from the values of ∆2 and τV
.
Eur. J. Inorg. Chem. 2004, 463Ϫ468
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
465

S. Laurent, F. Botteman, L. Vander Elst, R. N. Muller
Table 2. Values of τR, τSO, τV, and r obtained by fitting of the proton NMRD data at 310 K and 298 K
FULL PAPER
T ϭ 310 K
τR (ps)
τM (ns)[a]
τso (ps)
τv (ps)
r (nm)
Gd-(S)-EOB-DTPA-BMA (1)
Gd-(S)-C4Bz-DTPA-BMA (2)
71 (63)[b]
92
472
472
429
429
143
967
82
82
86
86
72
79
76
86
83
74
76
88
77
91
16
21
11
16
23
12
23
28
14
17
0.291
0.3
0.291
0.3
0.31
0.31
0.291
0.3
0.291
0.3
61 (60)[b]
70
Gd-DTPA (3)
Gd-DTPA-BMA (4)
Gd-(S)-EOB-DTPA (5)
57 (60)[b]
64 (67)[b]
71 (66)[b]
86
Gd-(S)-C4Bz-DTPA (6)
61 (65)[b]
72
T ϭ 298 K
τR (ps)
τM (ns)[a]
τso (ps)
τv (ps)
r (nm)
Gd-(S)-EOB-DTPA-BMA (1)
Gd-(S)-C4Bz-DTPA-BMA (2)
84 (86)[b]
97
976
976
798
798
69
78
72
88
17
24
13
15
0.291
0.3
0.291
0.3
73 (83)[b]
82
[a]
[b]
2
τM values were obtained from 17O NMR.
The values in parentheses are those obtained by H NMR spectroscopy on the lantha-
num analogue.
ton paramagnetic relaxation rate of the solution. The time
evolution of the proton relaxation rate was monitored at
0.47 T and 310 K. Previous results showed that, on the one
hand, the C-functionalized carboxylate complexes are kine-
tically and thermodynamically more stable than the parent
compound Gd-DTPA and, on the other hand, that bis-am-
ide derivatives are less stable.[14] For the bis-amide C-substi-
tuted compounds, the C-substitution and the bis-amide
functions have opposite effects. The ‘‘kinetic index’’, which
is the time required to reach 80 percent of the initial R1P
value, and the ‘‘thermodynamic index’’, which is the R1P (t)/
R1P (0) value measured after 3 days, indeed show a faster
and more extensive transmetallation of Gd-(S)-EOB-
Figure 4. 1H NMRD relaxivity profiles of Gd-(S)-C4Bz-DTPA-
BMA (2) (top graph) and Gd-(S)-EOB-DTPA-BMA (1) (bottom
graph) in water at 298 K and 310 K ([Gd complex] ϭ 1 m). The
lines correspond to the theoretical fittings of the data points
Figure 5. Evolution of the R1p (t)/ R1p (t ϭ 0) ratio as a function of
time for Gd-(S)-C4Bz-DTPA (6), Gd-(S)-EOB-DTPA (5), Gd-
DTPA (3), Gd-(S)-C4Bz-DTPA-BMA (2), Gd-(S)-EOB-DTPA-
BMA(1), and Gd-DTPA-BMA (4) (concentrations of Gd com-
plexes and ZnCl2 are 2.5 m in a phosphate buffer of pH ϭ 7, T ϭ
Finally, the transmetallation of the gadolinium complex
by Zn2ϩ has been assessed. In a phosphate buffer (pH ϭ
7), a precipitation of gadolinium ions expelled from the
complex by Zn2ϩ occurs resulting in a decrease of the pro- 310 K, B0 ϭ 0.47 T)
466
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 463Ϫ468

New Bifunctional Contrast Agents
FULL PAPER
Table 3. Values of the kinetic and thermodynamic indexes obtained during the transmetallation process
Kinetic index (min)[a]
Thermodynamic index (%)[b]
Gd-(S)-EOB-DTPA-BMA (1)
Gd-(S)-C4Bz-DTPA-BMA (2)
Gd-DTPA (3)[c]
70
80
260Ϫ280
50Ϫ60
1500
300
13
22
49
9
69
73
[c]
Gd-DTPA-BMA (4)
[c]
Gd-(S)-EOB-DTPA (5)
[c]
Gd-(S)-C4Bz-DTPA (6)
[a]
p
The kinetic index is defined here as the time required to attain 80 % of the initial value of R1 . [b] The thermodynamic index is defined
here as the ratio between R1 at 4320 min and the initial R1 . From ref.[14]
.
p
p [c]
rate of 1 mL/min for 20 min. A fluorescence detector was used to
monitor the elution of the gadolinium complex. Relaxometry: Pro-
ton Nuclear Magnetic Relaxation Dispersion (NMRD) profiles
were recorded with a Field Cycling Relaxometer working between
0.02 and 50 MHz. Additional points at 20 and 300 MHz respec-
tively were obtained on a Bruker Minispec PC-20 thermostatically
controlled by a tetrachloroethylene flow and with a Bruker AMX
DTPA-BMA (1) and Gd-(S)-C4Bz-DTPA-BMA (2) than
that observed for the parent compound Gd-DTPA (3). Only
a slight stabilization was observed with respect to the un-
substituted bis-amide derivative (Figure 5 and Table 3).
1
Conclusion
300 spectrometer. Fitting of the H NMRD curves was performed
with previously described software using minimization routines
(Minuit, CERN Library).[15,16] The deuterium and 17O measure-
Two new bi-functionalized compounds, Gd-(S)-EOB-
DTPA-BMA (1) and Gd-(S)-C4Bz-DTPA-BMA (2), have ments were carried out with 2 mL samples contained in 10-mm
o.d. tubes with a Bruker AMX-300 spectrometer equipped with a
broadband probe. The field homogeneity was optimized on the pro-
ton free induction decay using the decoupling coil. The temperature
was regulated by air or nitrogen flow (Bruker BVT-2000 unit). Deu-
terium longitudinal relaxation rates were measured using an inver-
sion recovery Fourier transform sequence on solutions containing
the labeled diamagnetic lanthanum complex dissolved in deuterium
depleted water (Aldrich, Bornem, Belgium). The experimental data
were fitted with a three-parameter minimization routine. Diamag-
netic transverse relaxation times of 17O water (pH ഠ 6.5) were
been successfully synthesized. The τM values obtained by
17O NMR spectroscopy confirm the lengthening and ad-
verse effect of the amide functions and demonstrate the
beneficial effect of a C-4 substitution on the reduction of
the water residence time. As assessed by the Zn2ϩ transmet-
allation process, the two new bis-amides are less stable than
Gd-DTPA and slightly more stable than Gd-DTPA-BMA
(4). In conclusion, although more difficult to synthesize, C-
4-substituted DTPA complexes are better suited for the co-
valent binding to macromolecules or to specific vectors in measured at natural abundance using
the context of the design of new specific contrast agents for
molecular imaging.
a
CarrϪPurcellϪ
MeiboomϪGill sequence and a two-parameter fit of the data. 17O
transverse relaxation times of water (natural abundance) in solu-
tions containing paramagnetic complexes were calculated directly
from the linewidth. All 17O NMR spectra were proton decoupled.
The theoretical fitting of the data was performed as described pre-
viously.[1,2,9]
Experimental Section
General
Synthesis of Ligands
Chemicals: Gd-DTPA (3), Gd-(S)-EOB-DTPA (5), and the ligand
EOB-DTPA were provided by Schering AG (Berlin, Germany). The
ligand (S)-C4Bz-DTPA and the corresponding gadolinium complex
were obtained as described earlier.[1] The organic reactants and in-
termediates were purchased from Aldrich (Bornem, Belgium).
NMR Spectroscopy: 1H and 13C NMR spectra were recorded with
3,6,9-Tris(carboxymethyl)-4-ethoxybenzyl-1,11-bis(methylamino)-
3,6,9-triazaundeca-1,11-dione [(S)-EOB-DTPA-BMA (1)]: A mix-
ture of (S)-EOB-DTPA (0.4 g), acetic anhydride (0.6 mL), and pyri-
dine (0.5 mL) was heated at 70 °C for 24 h. The suspension was
cooled, a 2  methylamine (methanol solution) was added (3 mL)
and the mixture was stirred for around 24 h. After addition of
a Bruker AMX-300 instrument (Bruker, Karlsruhe, Germany). methanol (50 mL), the mixture was filtered. The filtrate was evapo-
Chemical shifts (in ppm) are referred to the solvent peak. For 13C
rated to dryness under reduced pressure. Water (5 mL) was then
NMR, the methyl signal of tert-butanol was used as reference (δ ϭ added and the pH was adjusted to 2 with 2  HCl. The solution
31.2 ppm). Mass Spectrometry: Mass spectra (LSIMS) were ob-
tained with a VG AUTOSPEC mass spectrometer (VG Analytical,
was passed through a DOWEX AG 50 W-X8 (Hϩ form) cation-
exchange resin which was first eluted with water and the product
Manchester, UK). Samples were dissolved in water and deposited was recovered from the resin by elution with 2  aqueous ammonia.
in a glycerol matrix. Chromatography: High-performance liquid
chromatography (HPLC) was performed with a Waters 600 multi-
solvent delivery system equipped with a Rheodyne injection valve
(20µL loop) and controlled by the Millenium software (Waters,
Milford, USA). A Novapak C18 column (4.56 mm ϫ 150 mm) was
The residue obtained after evaporation of the solution was dis-
solved in acetic acid and precipitated with ethyl acetate. The crude
product was recovered by filtration. Yield: 0.16 g (38 %). 1H
NMR (D2O): δ ϭ 6.9Ϫ6.7 (4 H, AAЈBBЈ, Ph), 3.8 (q, 2 H, CH2),
3.6Ϫ2.8 (m, 19 H, 8 ϫ CH2, CH), 2.5 (s, 3 H, CH3), 2.45 (s, 3 H,
used. Elution was performed with a linear gradient of pure 0.05  CH3), 1.1 (t, 3 H, CH3) ppm. 13C NMR (D2O): δ ϭ 178.1, 171.8,
triethylammonium acetate (pH ϭ 6) to 100 % methanol at a flow 171.2, 170.6, 170.4, 166.1, 114.2, 128.9, 130.4, 65.1, 62.5, 61.1, 58.9,
Eur. J. Inorg. Chem. 2004, 463Ϫ468
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
467

S. Laurent, F. Botteman, L. Vander Elst, R. N. Muller
FULL PAPER
57.3, 57.2, 56.5, 55.1, 53.7, 52.3, 51.8, 26.4, 26.3, 14.3 ppm. LSIMS: and 2.5 m of ZnCl2.[14] The measurements were performed with
[M ϩ 1]ϩ: 554ϩ. HPLC: 13.3 min.
a Bruker Minispec PC 120 (20 MHz) spin analyzer at 310 K.
4-Benzyl-3,6,9-tris(carboxymethyl)-1,11-bis(methylamino)-3,6,9-tri-
azaundeca-1,11-dione [(S)-C4Bz-DTPA-BMA (2)]: This compound
was obtained from (S)-C4Bz-DTPA[1] as described for (S)-EOB-
DTPA-BMA above. Yield: 0.142 g (32 %). 1H NMR (D2O): δ ϭ
7.3Ϫ7.0 (m, 5 H, Ph), 3.45Ϫ2.6 (m, 19 H, 9 ϫ CH2, CH), 2.5 (s, 3
H, CH3), 2.45 (s, 3 H, CH3) ppm. 13C NMR (D2O): δ ϭ 177.9,
171.9, 171.3, 170.6, 170.3, 138.8, 128.6, 128.3, 125.8, 62.4, 61.0,
58.8, 57.2, 57.1, 56.6, 55.2, 53.6, 52.4, 51.7, 26.4, 26.3 ppm. LSIMS:
[M ϩ 1]ϩ: 536ϩ. HPLC: 12.5 min.
Acknowledgments
The authors thank Mrs Patricia de Francisco for her help in pre-
paring the manuscript. This work was supported by the ARC Pro-
gram 95/00Ϫ194 of the French Community of Belgium, and the
Fonds National de la Recherche Scientifique (FNRS) of Belgium.
FB thanks the Fonds pour la Recherche dans l’Industrie et l’Agric-
ulture (FRIA) of Belgium for their financial help.
Complexation: The LaIII or GdIII complexes were prepared by mix-
ing aqueous solutions of equimolar amounts of hydrated LaCl3 or
GdCl3 and the ligand. The pH was adjusted to between 6.5 and 7.
The absence of free La or Gd ions was checked with xylenol orange
indicator. The complexes were passed through a Sep-Pak column
(Waters, Accell Plus QMA Cartridges) and the eluted solutions
were freeze-dried. The purities of the chelates were verified by
HPLC and their masses confirmed by LSIMS.
[1]
´
S. Laurent, L. Vander Elst, S. Houze, N. Guerit, R. N. Muller,
Helv. Chim. Acta 2000, 83, 394Ϫ406.
[2]
[3]
[4]
[5]
L. Vander Elst, F. Maton, S. Laurent, F. Seghi, F. Chapelle, R.
N. Muller, Magn. Reson. Med. 1997, 38, 604Ϫ614.
S. Aime, M. Botta, M. Fasano, S. Paoletti, L. Anelli, F. Uggeri,
M. Virtuani, Inorg. Chem. 1994, 33, 4707Ϫ4711.
E. Toth, D. Pubanz, S. Vauthey, L. Helm, A. E. Merbach,
Chem. Eur. J. 1996, 2, 1607Ϫ1615.
S. Aime, A. Barge, J. I. Bruce, M. Botta, J. A. K. Howard, J.
M. Moloney, D. Parker, A. S. de Sousa, M. Woods, J. Am.
Chem. Soc. 1999, 121, 5762Ϫ5771.
I. Solomon, Phys. Rev. 1955, 99, 559Ϫ565.
N. Bloembergen, J. Chem. Phys. 1957, 27, 572Ϫ573.
J. H. Freed, J. Chem. Phys. 1978, 68, 4034Ϫ4037.
Deuteration: Deuteration of the ligands [(S)-EOB-DTPA-BMA and
(S)-C4Bz-DTPA-BMA] or of the lanthanum complexes at the α
position of the carbonylic groups was performed by the procedure
described by Wheeler and Legg.[13] The ligand or complex (4 mmol)
was dissolved in D2O (40 mL), the pD was adjusted to 10.6 by
addition of K2CO3 and the mixture was heated to reflux for 24 h
with stirring. The pD was then adjusted to 2 (ligand solution) or 6
(lanthanum complex solution) with concentrated hydrochloric acid,
the solution was evaporated to a final volume of 10 mL and the
solid KCl was removed by filtration. Acetone was added to induce
precipitation of the deuterated compounds, which were recovered
by filtration, dissolved in H2O and isolated after lyophilization. The
[6]
[7]
[8]
[9]
´
R. N. Muller, B. Radüchel, S. Laurent, J. Platzek, C. Pierart,
P. Mareski, L. Vander Elst, Eur. J. Inorg. Chem. 1999,
1949Ϫ1955.
[10]
[11]
[12]
´
M. C. Alpoim, A. M. Urbano, C. F. G. C. Geraldes, J. A. Pet-
ers, J. Chem. Soc., Dalton Trans. 1992, 463Ϫ467.
K. Micskei, L. Helm, E. Brücher, A. E. Merbach, Inorg. Chem.
1993, 32, 3844Ϫ3850.
F. Botteman, G. Nicolle, L. Vander Elst, S. Laurent, A. E. Mer-
bach, R. N. Muller, Eur. J. Inorg. Chem. 2002, 2686Ϫ2693.
W. D. Wheeler, J. I. Legg, Inorg. Chem. 1985, 24, 1292Ϫ1297.
S. Laurent, L. Vander Elst, F. Copoix, R. N. Muller, Invest.
Radiol. 2001, 36, 115Ϫ122.
R. N. Muller, D. Declercq, P. Vallet, F. Giberto, B. Daminet,
H. W. Fischer, F. Maton, Y. Van Haverbeke, in Implementation
and operation of an integrated and comprehensive relaxometric
data bank, Proc. ESMRMB, 7th Annual Congress, Strasbourg,
1990, p. 394.
P. Vallet, in Relaxivity of nitroxide stable free radicals. Evalu-
ation by field cycling method and optimisation, PhD Thesis, Uni-
versity of Mons-Hainaut, Belgium, 1992.
1
[13]
[14]
deuteration (Ͼ 95 %) was confirmed by H NMR spectroscopy.
(S)-[D8]EOB-DTPA-BMA: NMR (D2O, pH ϭ 10): δ ϭ 6.9Ϫ6.7 (4
H, AAЈBBЈ, Ph), 3.8 (q, 2 H, CH2), 3.6Ϫ2.8 (m, 11 H, 5 ϫ CH2,
CH), 2.5 (s, 3 H, CH3), 2.45 (s, 3 H, CH3), 1.1 (t, 3 H, CH3) ppm.
[15]
[16]
(S)-[D8]C4Bz-DTPA-BMA: NMR (D2O, pH ϭ 10): δ ϭ 7.4Ϫ7.1
(m, 5 H, Ph), 3.45Ϫ3.0 (m, 5 H, 2 ϫ CH2, CH), 2.9Ϫ2.6 (m, 6 H,
3 ϫ CH2), 2.5 (s, 3 H, CH3), 2.45(s, 3 H, CH3) ppm.
Kinetics of the Transmetallation: The technique is based on the
evolution of the paramagnetic proton longitudinal relaxation rate
Received July 11, 2003
2Ϫ
(R1p) of a phosphate buffer solution (pH, 7, [H2PO4Ϫ] ϩ [HPO4
]
Early View Article
ϩ [PO43Ϫ] ϭ 67 m) containing 2.5 m of gadolinium complex
Published Online December 12, 2003
468
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 463Ϫ468