Comprehensive evaluation of the physicochemical properties of Ln III complexes of aminoethyl-DO3A as pH-responsive T1-MRI contrast agents

Article abstract of DOI:10.1002/chem.201304063

N-Substituted aminoethyl groups were attached to 1,4,7,10- tetraazacyclododecane-1,4,7-triacetic acid (DO3A) with the aim to design pH-responsive Ln^III complexes based on the pH-dependent on/off ligation of the amine nitrogen to the metal ion. The following ligands were synthesized: AE-DO3A (aminoethyl-DO3A), MAE-DO3A (N-methylaminoethyl-DO3A), DMAE-DO3A (N,N-dimethylaminoethyl-DO3A) and MEM-AE-DO3A (N-methoxyethyl-N- methylaminoethyl-DO3A). The physicochemical properties of the Ln^III complexes were investigated for the evaluation of their potential applicability as magnetic resonance imaging (MRI) contrast agents. In particular, a ¹H and ¹⁷ONMR relaxometric study was carried out for these Gd^III complexes at two different pH values: at basic pH (pendant amino group coordinated to the metal centre) and at acidic pH (protonated amine, not interacting with the metal ion). Eu^III complexes allow one to estimate the number of inner-sphere water molecules through luminescence lifetime measurements and obtain some structural information through variable-temperature (VT) high-resolution ¹HNMR studies. Equilibria between differently hydrated species were found for most of the complexes at both acidic and basic pH. The thermodynamic stability of Ca^II, Zn^II, Cu^II and Ln^III complexes and kinetics of formation and dissociation reactions of Ln^III complexes of AE-DO3A and DMAE-DO3A were investigated showing stabilities comparable to currently approved Gd^III-based CAs. In detail, higher total basicity (Σlog K_i^H) and higher stability constants of Ln ^III complexes were found for AE-DO3A with respect to DMAE-DO3A (i.e., log K_Gd-AE-DO3A=22.40 and log K_Gd-DMAE-DO3A=20.56). The transmetallation reactions of Gd^III complexes are very slow (Gd-AE-DO3A: t_1/2=2.7×10⁴h; Gd-DMAE-DO3A: 1.1×10⁵h at pH7.4 and 298K) and occur through proton-assisted dissociation. pH Switches: N-Substituted aminoethyl-1,4,7,10- tetraazacyclododecane-1,4,7-triacetic acid (DO3A)-based complexes were prepared to obtain pH-sensitive T₁-MRI contrast agents. A detailed evaluation of the physicochemical properties of Ln^III complexes was undertaken by different techniques, highlighting the structural variations triggered by the protonation of the pendant amine and the effect of the substituents on the corresponding nitrogen atom (see figure).

Full text of DOI:10.1002/chem.201304063

DOI: 10.1002/chem.201304063
Full Paper
&
Magnetic Resonance Imaging
Comprehensive Evaluation of the Physicochemical Properties of
LnIII Complexes of Aminoethyl-DO3A as pH-Responsive T1-MRI
Contrast Agents**
Zsolt Baranyai,[a] Gabriele A. Rolla,[b] Roberto Negri,[c] Attila Forgꢀcs,[a]
Giovanni B. Giovenzana,*[c] and Lorenzo Tei*[b]
Abstract: N-Substituted aminoethyl groups were attached
to 1,4,7,10-tetraazacyclododecane-1,4,7-triacetic acid (DO3A)
with the aim to design pH-responsive LnIII complexes based
on the pH-dependent on/off ligation of the amine nitrogen
to the metal ion. The following ligands were synthesized:
AE-DO3A (aminoethyl-DO3A), MAE-DO3A (N-methylami-
cules through luminescence lifetime measurements and
obtain some structural information through variable-temper-
ature (VT) high-resolution 1H NMR studies. Equilibria be-
tween differently hydrated species were found for most of
the complexes at both acidic and basic pH. The thermody-
namic stability of CaII, ZnII, CuII and LnIII complexes and kinet-
ics of formation and dissociation reactions of LnIII complexes
of AE-DO3A and DMAE-DO3A were investigated showing
stabilities comparable to currently approved GdIII-based CAs.
In detail, higher total basicity (SlogKiH) and higher stability
constants of LnIII complexes were found for AE-DO3A with
respect to DMAE-DO3A (i.e., logKGd-AE-DO3A =22.40 and
logKGd-DMAE-DO3A =20.56). The transmetallation reactions of
GdIII complexes are very slow (Gd-AE-DO3A: t1/2 =2.7ꢁ104 h;
Gd-DMAE-DO3A: 1.1ꢁ105 h at pH 7.4 and 298 K) and occur
through proton-assisted dissociation.
noethyl-DO3A),
DMAE-DO3A
(N,N-dimethylaminoethyl-
DO3A) and MEM-AE-DO3A (N-methoxyethyl-N-methylami-
noethyl-DO3A). The physicochemical properties of the LnIII
complexes were investigated for the evaluation of their po-
tential applicability as magnetic resonance imaging (MRI)
1
contrast agents. In particular, a H and 17O NMR relaxometric
study was carried out for these GdIII complexes at two differ-
ent pH values: at basic pH (pendant amino group coordinat-
ed to the metal centre) and at acidic pH (protonated amine,
not interacting with the metal ion). EuIII complexes allow
one to estimate the number of inner-sphere water mole-
Introduction
CAs operate by reducing the longitudinal relaxation time (T1)
of water protons, leading to a local brightening of the image;
the ability to reduce T1 is measured by a parameter called re-
laxivity (r1p). The relaxivity of a paramagnetic metal complex is
dependent on several parameters, and can be improved by:
1) slowing the rotational motion of the paramagnetic centre
encasing the latter in a large molecule (typically by covalent-
or non-covalent binding to physiological or synthetic biocom-
patible macromolecules);[2] 2) increasing the number (q) of
water molecules directly coordinated to the metal ion and
their rate of exchange with bulk water;[3] 3) improving the
local concentration of paramagnetic metal centres, linking the
single chelates in multimeric systems[4] or collecting them in
well-defined nanoparticles.[5]
MRI diagnostics takes advantage of specific contrast agents
(CAs) to improve the quality of the registered image.[1] Two
families of contrast agents, namely paramagnetic chelates and
superparamagnetic nanoparticles have been used in the last
three decades as T1 and T2 CAs, respectively, and a detailed
knowledge of their potentiality has been gained. Paramagnetic
[a] Dr. Z. Baranyai, A. Forgꢀcs
Department of Inorganic and Analytical Chemistry
University of Debrecen, H-4010 Debrecen Egyetem tꢁr 1 (Hungary)
[b] Dr. G. A. Rolla, Dr. L. Tei
Dipartimento di Scienze e Innovazione Tecnologica
Universitꢂ del Piemonte Orientale “Amedeo Avogadro”
Viale T. Michel 11, 15121 Alessandria (Italy)
E-mail: lorenzo.tei@mfn.unipmn.it
The chemistry of paramagnetic complexes is well summar-
ized in dedicated reviews[6] and books,[7] and the results of the
activity in this field is witnessed by hundreds of derivatives, de-
vised to optimize the chemical and magnetic properties.[8] The
observation of toxic effects related to long term dissociation[9]
of the metal ion from the corresponding complex prompted
research efforts aimed at the enhancement of both the ther-
modynamic and kinetic stability of the chelates, generally tack-
led by structural reinforcement of the ligand structures.[10] The
[c] Dr. R. Negri, Prof. G. B. Giovenzana
Dipartimento di Scienze del Farmaco
Universitꢂ del Piemonte Orientale “Amedeo Avogadro”
Largo Donegani 2/3, 28100 Novara (Italy)
E-mail: giovenzana@pharm.unipmn.it
[**] DO3A=1,4,7,10-tetraazacyclododecane-1,4,7-triacetic acid
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201304063.
Chem. Eur. J. 2014, 20, 2933 – 2944
2933
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
recent development of molecular imaging applications,[11] in
which imaging probes are required to report specific changes
in physical or (bio)chemical parameters led to an extensive re-
search for “responsive” MRI probes. Recent reviews summarize
the wide array of responsive paramagnetic chelates, designed
to monitor different parameters such as pH,[12] pO2,[13] the con-
centration of glucose,[14] metal ions[15] and enzymes.[16] pH rep-
resents one of the first and most important targets: significant
variations from physiological values may be associated with
a number of pathologies, cancer being the most representa-
tive.[17] Several different pH-sensitive CAs were reported in the
scientific literature. Gd-based CAs usually rely on the variation
of number of inner-sphere water molecules (q)[18] or reorienta-
tional correlation time (tR)[19] to obtain a pH-dependent behav-
iour. PARACEST agents[20] show a large pH sensitivity elicited by
a mechanism involving chemical exchange of amide protons
of the paramagnetic complex.[21] Among the Gd-based CAs,
a macrocyclic DO3A-like system embodying a b-sulfonami-
doethyl group (Gd-DO3A-SA)[18] represents a prototypical ex-
ample of the strategy involving the modulation of the coordi-
nation to affect the relaxivity of the complex. The deprotona-
tion of the sulfonamide in the side arm triggers the coordina-
tion of the metal by this donor group, replacing coordinated
water molecules. We have recently reported a versatile exten-
sion of this interesting reversible mechanism, by using a ligand
bearing a side arm containing a simple amino group, whose
protonation/deprotonation leads to a modification of the cor-
responding complex structure and dynamics. The simple
amino group allows us to finely modulate the pK of this
change by N-substitution with suitable alkyl groups.[22] Recent-
ly, the pH-dependent PARACEST effect of the EuIII complex with
amino- and dimethylaminoethyl-DO3As has been reported
showing an additional interesting application of this class of li-
gands.[23]
series. Substitution with simple methyl groups leads to MAE-
DO3A (N-methylaminoethyl-DO3A) and DMAE-DO3A (N,N-di-
methylaminoethyl-DO3A), designed to explore the effect of
the sequential substitution on the exocyclic nitrogen atom.
The fourth ligand (MEM-AE-DO3A, N-methoxyethyl-N-methyl-
aminoethyl-DO3A) has a tertiary amino group in which one of
the N-substituents carries an additional weak donor atom
(ether oxygen), devised to gain a supplementary influence on
the metal coordination sphere, with the side arm acting as po-
tential bidentate, although maintaining the overall basicity
profile of the ligand family.
Ligands AE-DO3A[24] and DMAE-DO3A[22] were synthesized
according to literature procedures. The preparation of ligands
MAE-DO3A and MEM-AE-DO3A starts from N-methylethanol-
amine (Scheme 2). The latter is N-alkylated with benzyl bro-
Herein we report a comprehensive evaluation of physico-
chemical properties of selected LnIII complexes of aminoethyl-
DO3As, relevant to their potential application as MRI contrast
agents, and highlighting the key role of N-alkyl substituents on
the solution structure and dynamics of the corresponding
complexes.
Results and Discussion
Synthesis
The aminoethyl-functionalized DO3A-like ligands employed for
the formation of paramagnetic complexes of this work are de-
picted in Scheme 1. The ligand AE-DO3A (aminoethyl-DO3A)
represents the parent (N-unsubstituted) compound of this
Scheme 2. Preparation of MAE-DO3A and MEM-AE-DO3A.
mide to give the tertiary ethanolamine 1, converted into the
mesylate 2 by treatment with methanesulfonyl chloride and
triethylamine. Compound 2 was directly used for the alkylation
of the secondary amine group of DO3A tri-tert-butyl ester[25] (3,
DO3A(tBu)3), operating in acetonitrile in the presence of finely
powdered potassium carbonate and leading to the intermedi-
ate 4.
Removal of the benzyl group on the distal nitrogen atom
was accomplished through catalytic transfer hydrogenation
Scheme 1. Aminoethyl-functionalized DO3A complexes.
Chem. Eur. J. 2014, 20, 2933 – 2944
www.chemeurj.org
2934
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
(HCOONH4, Pd/C, MeOH heated at reflux) and the product was
treated with trifluoroacetic acid (TFA) directly after the isolation
from the hydrogenation step, giving the desired ligand MAE-
DO3A.
In parallel, N-methylethanolamine was N-alkylated with tosy-
late 5, which was easily obtained by treatment of 2-methoxy-
ethanol with p-toluenesulfonyl chloride in the presence of trie-
thylamine. Attempted mesylation of the alcoholic group of 6
with MsCl/Et3N resulted in complete conversion to the
b-chloroamine 7, likely by displacement of the first formed me-
sylate by the chloride anion with the anchimeric assistance of
the oxygen atom of the remote methoxy group, suitably
placed to form a six-membered oxonium intermediate. Never-
theless, the alkylating nature of 7 was exploited to introduce
the desired structural residue on the macrocyclic ring of DO3A
by treatment with DO3A(tBu)3 (3), leading to the protected
ligand 8. Final removal of the tert-butyl groups from the esters
yielded the ligand MEM-AE-DO3A.
Figure 1. Relaxivity (r1p) variation as a function of pH for the Gd-aminoethyl-
pH dependence
~
&
^
DO3A complexes ( =Gd-AE-DO3A, =Gd-MEM-AE-DO3A, =Gd-MAE-
*
DO3A, =Gd-DMAE-DO3A).
The pH dependence of the relaxivity (r1p) of GdIII complexes of
this class of DO3A-like ligands was measured to assess the via-
bility of these complexes as responsive pH agents. Since an im-
portant issue to this end is the extent of the variation of r1p
(Dr1p) in a clinically relevant pH range (6–8), the possibility to
modulate the basicity and the steric hindrance of the pendant
amino groups in this series of GdIII complexes is an important
feature that can give rise to interesting results. In fact, the
presence of the amine leads to a protonation equilibrium that
can influence the coordination cage of the GdIII complex and
thus its hydration state. As shown previously for Gd-AE-DO3A
and Gd-DMAE-DO3A, the GdIII complexes reported herein ex-
hibit pH-dependent relaxivity due to a change in the hydration
state associated with the reversible on/off ligation of the
amine nitrogen donor. The pH range and the degree of varia-
tion of r1p vary markedly in the series of complexes passing
from Gd-AE-DO3A, which bears a pendant arm containing
a primary amine, to Gd-DMAE-DO3A and Gd-MEM-AE-DO3A,
in which tertiary amino groups are present. The pKaH of the
amino functionalities can be estimated by fitting the relative
sigmoidal curves, characteristic of a typical two-state equilibri-
um process. The relaxivity variation as a function of pH is re-
ported in Figure 1. The extimated pKaH range from 5.95 for Gd-
AE-DO3A to 7.82 for Gd-DMAE-DO3A, reflecting the different
basicity of the substituted amino groups. Since for MRI pH-re-
sponsive applications the variation of r1p has to occur around
the physiological pH (between 6 and 7.4), the pKaH of 6.96 for
Gd-MEM-AE-DO3A is the most appropriate.
basic pH and the Dr1 is reduced (Figure S1 in the Supporting
Information).
Concerning the relaxivity values, for all the complexes r1p at
acidic pH is roughly similar, ranging from 5.7 to 6.2 mmꢀ1 sꢀ1
(20 MHz and 298 K). By increasing the pH, the relaxivity steadi-
ly decreases proportionally to the decrease in the hydration
state q reaching a limiting value of 4.1 mmꢀ1 sꢀ1 for Gd-AE-
DO3A and about 3 mmꢀ1 sꢀ1 at pH 9 for the other GdIII com-
plexes. The Dr1p observed with these GdIII complexes ranges
from 1.6 mmꢀ1 sꢀ1 in case of Gd-AE-DO3A to 3.4 mmꢀ1 sꢀ1 for
Gd-MEM-AE-DO3A. These r1p variations are lower than those
observed for Gd-DO3A-sulfonamidoethyl systems (Dr1p
ꢁ6 mmꢀ1
s
ꢀ1), in which there is a more evident switch from
q=2 to q=0 hydration states of the GdIII complex as a conse-
quence of the on/off ligation switch. In fact, for low-molecular-
weight monohydrated GdIII-chelates (q=1), as the clinically
used MRI contrast agents, the relaxivity assumes a value of
about 4–5 mmꢀ1 sꢀ1 measured at 20 MHz and 298 K. Among
our complexes, the r1p value of 4.1 mmꢀ1 sꢀ1 for Gd-AE-DO3A
at pH>7.5 approximately indicates a q=1 GdIII complex,
whereas for the other complexes the relaxivity-limiting values
at acid and basic pH are not easily attributable to GdIII com-
plexes with a distinct hydration state. For this reason, the hy-
dration number of the complexes at both acidic and basic pH
was independently determined by luminescence lifetime meas-
urements on analogue EuIII complexes following a well-estab-
lished method.[27] The emission lifetimes of the Eu(5D0) excited
levels were measured in D2O and H2O solutions of the com-
plexes, and were used to calculate the number of coordinated
water molecules q (Table 1). The luminescence lifetimes deter-
mined for all four Eu complexes at acidic pH with the proton-
ated pendant amino groups lead to a fractional value of hydra-
tion number of about 1.3, revealing the possible presence of
an equilibrium in solution involving a nine-coordinated species
A preliminary analysis of the activity of these probes in Sero-
norm was performed to obtain an indication of the applicabili-
ty of this class of GdIII complexes as pH-sensitive probes in
vivo, in which it is recognized that a number of endogenous
anions can displace coordinated water molecules of Gd-DO3A-
like complexes masking the effect that is active in H2O.[26] Thus,
the measure of the pH dependence of r1p for Gd-MEM-AE-
DO3A in Seronorm showed that pKaH =7.7 is slightly shifted to
Chem. Eur. J. 2014, 20, 2933 – 2944
www.chemeurj.org
2935
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ordination isomers characterized with the same conformation
of the macrocyclic ring but with different orientation of the
side arms (i.e., capped square-antiprismatic geometry, SAP, and
capped twisted square antiprismatic geometry, TSAP).[28,31] In
solution, the conformation isomers of Ln-DOTA complexes can
give rise to exchange processes through macrocyclic ring in-
version ((llll)Ð(dddd)) and acetate arms rotation (LÐD) in
the temperature range of 0–1008C.[28,31] Detailed NMR structur-
al studies on Ln-DOTA-like complexes indicated that the rela-
tive population of the SAP and TSAP isomers is affected by the
size of the LnIII ions.[28,31]
Table 1. Lifetimes of the Eu(5D0) excited states in Eu-AE-DO3A, Eu-MAE-
DO3A, Eu-DMAE-DO3A, Eu-MEM-AE-DO3A complexes and hydration
number (q).
Acidic pH[a]
tD O
Basic pH[a]
tD O
Complex
tH O
q[b]
tH O
q[b]
2
2
2
2
[ms]
[ms]
[ms]
[ms]
Eu-AE-DO3A
0.48
0.48
0.50
0.49
1.47
1.37
1.54
1.35
1.4
1.3
1.3
1.3
0.57
0.65
0.80
0.71
1.41
1.03
1.61
1.30
1.0
0.4
0.4
0.4
Eu-MAE-DO3A
Eu-DMEA-DO3A
Eu-MEM-AE-DO3A
[a] The pH/pD values used for Eu-AE-DO3A are 4.5 and 7.5; for all the
other complexes pH/pD=5 and
9
were used; [b] qEu =1.2(1/tH O
ꢀ
The substitution of one acetic arm with aminoethyl moieties
reduces the symmetry removing the proton equivalence and
likely decreases the rigidity of the coordination cage due to
2
1/tD Oꢀ0.25) as reported in ref. [27].
2
1
the higher flexibility of the ethylene group. The H NMR spec-
with q=2 and an eight-coordinated species with q=1. This
may be due to the formation of hydrogen bonds between the
protonated amine and the inner-sphere water molecules that
hinders the coordination of a second water molecule to the
metal centre. The results obtained at basic pH when the free
amino group can bind the lanthanide ion are different for Eu-
AE-DO3A with respect to the other EuIII complexes. In fact,
whereas for Eu-AE-DO3A the data are consistent with the pres-
ence of one inner-sphere water molecule, for the other EuIII
complexes the hydration number results again fractional (q=
0.4), which might be attributed to an equilibrium between two
differently hydrated species (q=0 and q=1) with eight- and
nine-coordination geometries.
tra of Eu-AE-DO3A and Eu-MAE-DO3A at 270, 280 and 290 K
(Figures S2 and S3, the Supporting Information) show one set
of signals that can be interpreted by the predominance of one
isomer in solution at pH 10. In fact, the chemical shifts of the
four axial ring protons (Ha: d=24.7, 13.3, 13.3, and 12.1 ppm
for Eu-AE-DO3A and Ha: d=33.9, 32.8, 31.0, and 23.1 ppm for
Eu-MAE-DO3A), are respectively similar to those of the TSAP
and SAP isomers of Eu-DOTA,[28] indicating the prevalence of
different isomers for the two complexes. On the other hand,
the 1H NMR spectra of Eu-DMAE-DO3A (Figure S4, the Sup-
porting Information) contain two sets of well-separated signals
indicating the presence of both SAP and TSAP isomers. The
chemical shifts of the axial ring protons of the major Ha: d=
38.5, 36.4, 31.7, and 28.0 ppm and minor H’a: d=19.9, 19.1,
18.1, and 16.8 ppm isomers of Eu-DMAE-DO3A are similar to
those of the SAP and TSAP isomers of Eu-DOTA, respectively.
By comparing the integral of the two set of signals of axial
ring-protons, the ratio between the SAP and TSAP isomers for
Eu-DMAE-DO3A under these conditions can be estimated as
about 68:32. In conclusion, the fractional hydration number
can be explained by a mixture of q=0 and q=1 isomers with
a different coordination number. However, an alternative inter-
pretation could be given by the presence of q=1 complexes
with a longer Ln–water bond as reported earlier.[32]
1H NMR spectroscopic study on EuIII complexes
To get more insight, the structural and dynamic properties of
Eu-AE-DO3A, Eu-MAE-DO3A and Eu-DMAE-DO3A were inves-
tigated by using 1H NMR spectroscopy at pH 10 and at low
temperatures (270, 280 and 290 K). Basic pH was chosen as the
coordination of the amino group in the EuIII complexes would
result in a more rigid structure with C1 symmetry. On the other
hand, the solution structure of the protonated Eu-HAE-DO3A,
Eu-HMAE-DO3A and Eu-HDMAE-DO3A complexes likely re-
sembles that of the corresponding Ln-DO3A complexes charac-
terized by a fast fluxional motion even at low temperature.
The solution structure of the LnIII complexes with this class
of DO3A-like ligands is expected to be similar to that of the
corresponding Ln-DOTA (DOTA=1,4,7,10-tetraazacyclodode-
cane-1,4,7,10-tetraacetic acid) complexes, which was analysed
in the solid state by using X-ray diffraction and in solution by
1H- and 13C NMR spectroscopy.[28–31] The detailed structural
characterization of Ln-DOTA complexes indicates that, because
of their rigid structure, the helicity of the four identically ori-
ented ethylene groups of the macrocycle results in two confor-
mation geometries (llll or dddd) of the Ln-DOTA complexes.
Similarly, the acetate groups can be oriented with two different
helicities labelled by L and D. Considering all possible con-
formers, two enantiomer pairs (L(llll) and D(dddd); D(llll)
Relaxometric properties
Since the four GdIII complexes discussed in this work have re-
laxivities that depend on pH with significant variation of r1p
from basic to acidic pH, the relaxometric properties were inde-
pendently studied in detail in the acidic range (pH 4.5–5) and
in the basic range (pH 8–9.5). The value of r1p of a GdIII com-
plex depends on the magnetic field strength, temperature, and
on several important structural and dynamic molecular param-
eters of the metal complex that describe the magnetic interac-
tion between the solvent nuclei and GdIII.[6] The measurement
of the longitudinal relaxation rates of the water proton over
a wide range of magnetic fields is the best way to obtain the
set of parameters governing the relaxivity. Thus, r1p as a func-
tion of proton Larmor frequency (nuclear magnetic resonance
dispersion, NMRD) was measured for all GdIII complexes at
both acidic and basic pH at 298 and 310 K over the range
1
and L(dddd)) may be formed in solution. Therefore, H- and
13C NMR studies on several Ln-DOTA complexes revealed the
presence of two sets of signals associated to two different co-
Chem. Eur. J. 2014, 20, 2933 – 2944
www.chemeurj.org
2936
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 4. Top : 1/T1 1H NMRD relaxation data for Gd-DMAE-DO3A at pH 5
Figure 2. Top : 1/T1 1H NMRD relaxation data for Gd-AE-DO3A at pH 4.5 (left)
and 7.5 (right), 298 K ( ) and 310 K ( ). Bottom: variable-temperature 17O re-
!
&
!
(left) and 9.5 (right), 298 K ( ) and 310 K ( ). Bottom: variable-temperature
17O reduced transverse relaxation rates at 11.7 T and at pH 5 (left) and 9.5
(right) for a 17.1 mm solution of Gd-DMAE-DO3A. The solid lines represent
the best results of the fitting to the experimental points (Table 2).
&
duced transverse relaxation rates at 11.7 T and at pH 4.5 (left) and 7.5 (right)
for a 13.3 mm solution of Gd-AE-DO3A. The solid lines represent the best re-
sults of the fitting to the experimental points (Table 2).
Figure 3. Top : 1/T1 1H NMRD relaxation data for Gd-MAE-DO3A at pH 5 (left)
Figure 5. Top : 1/T1 1H NMRD relaxation data for Gd-MEM-AE-DO3A at pH 4.5
and 9.5 (right), 298 K ( ) and 310 K ( ). Bottom: variable-temperature 17O re-
duced transverse relaxation rates at 11.7 T and at pH 5 (left) and 9.5 (right)
for a 12.8 mm solution of Gd-MAE-DO3A. The solid lines represent the best
results of the fitting to the experimental points (Table 2).
!
&
!
&
(left) and 9 (right), 298 K ( ) and 310 K ( ). Bottom: variable-temperature
17O reduced transverse relaxation rates at 11.7 T and at pH 4.5 (left) and 9
(right) for a 25.2 mm solution of Gd-MEM-AE-DO3A. The solid lines repre-
sent the best results of the fitting to the experimental points (Table 2).
0.01–70 MHz corresponding to magnetic field strengths vary-
ing between 2.34ꢁ10ꢀ4 and 1.64 T (Figure 2, Figure 3, Figure 4
and Figure 5).
and Freed’s[34] equations for the inner- and outer-sphere
proton relaxation mechanisms, respectively, and with the
Swift–Connick theory for 17O relaxation.[35] Some parameters
were fixed to reasonable values, which are well-accepted as
constant for this class of complexes: 1) the distance of closest
approach of the outer-sphere water protons (a=4.0 ꢃ); 2) the
relative diffusion coefficient of solute and solvent (298D=2.24ꢁ
To obtain more accurate information on the kinetics of the
water exchange, the temperature dependence of the 17O NMR
transverse relaxation rate, R2 was also measured for all Gd com-
plexes at 11.7 T on 13–25 mm solutions of the complexes at
pH 4.5–5 and pH 8–9.5.
10ꢀ5 cm2 sꢀ1); 3) the metal-bound water proton distance (rGdH
=
NMRD and 17O NMR spectroscopic data were analysed ac-
cording to the established theory of paramagnetic relaxation
expressed with the Solomon–Bloembergen–Morgan (SBM)[33]
3.1 ꢃ); 4) the Gd-17O scalar coupling constant (A/ꢀh=
ꢀ3.5 radsꢀ1). The number of coordinated water molecules q
was also set to the value determined by luminescence meas-
Chem. Eur. J. 2014, 20, 2933 – 2944
www.chemeurj.org
2937
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1
perature indicates higher tm values that can limit H
inner-sphere 1/T1 as exemplified by Equation (1):[7]
Table 2. Selected parameters obtained from the analysis of the 1/T1 NMRD profiles
(298 K) and 17O NMR (11.7 T) data for Gd-AE-DO3A, Gd-MAE-DO3A, Gd-DMAE-DO3A
and Gd-MEM-AE-DO3A and related complexes (Gd-DO3A[36] and Gd-DOTAMA-En[37]).[a]
1
1
D2
[1019 sꢀ2
tV
/
ð1Þ
pH
q
tR
kex
[ꢁ106 sꢀ1
DHm
[kJmolꢀ1
T1IS tM þ T1M
[ps]
]
]
]
[ps]
7.5
4.5 1.4
9.5 0.4
1
64ꢂ2
1.8ꢂ0.3
8.3ꢂ0.2
4.9ꢂ0.1
27.4ꢂ0.2 3.1ꢂ0.3 14ꢂ1
52.0ꢂ0.6 6.2ꢂ0.5 20ꢂ2
45.8ꢂ0.3 1.6ꢂ0.4 34ꢂ2
Gd-AE-DO3A
A more accurate estimate of this effect was as-
sessed by 17O NMR experimental data, especially at
basic pH for the mono-aqua Gd-AE-DO3A and for
the other Gd complexes present at about 40% in the
q=1, nonacoordinate form (Figures 2–5). The trend
along the series highlights that the water exchange
rate is slower (1.8ꢁ106 sꢀ1) for Gd-AE-DO3A, when
the primary amine coordinates the metal centre and
it increases moving to the coordination of a secon-
dary amine with Gd-MAE-DO3A (4.9ꢁ106 sꢀ1) and of
a tertiary amine with Gd-DMAE-DO3A (8.1ꢁ106 sꢀ1).
84ꢂ2
66ꢂ2
Gd-MAE-DO3A
Gd-DMAE-DO3A
Gd-MEM-AE-DO3A
5.0 1.3 115ꢂ2
0.65ꢂ0.05 22.3ꢂ0.6 3.4ꢂ0.2 28ꢂ1
9.5 0.4
69ꢂ4
8.1ꢂ0.1
29.6ꢂ0.3 3.5ꢂ0.5 19ꢂ2
5.0 1.3 107ꢂ2
0.72ꢂ0.05 24.0ꢂ0.7 5.9ꢂ0.4 22ꢂ1
72ꢂ4 13.4ꢂ0.5 46.0ꢂ0.7 8.2ꢂ0.5 13ꢂ1
9.5 0.4
5.0 1.3 120ꢂ2
0.68ꢂ0.01 20.1ꢂ0.5 3.7ꢂ0.2 28ꢂ1
Gd-DOTAMA-En
Gd-DO3A
7.4
1
79
66
1.1
6.2
34
44
3.8
4.6
11
14
7.4 1.8
[a] For the parameters rGd-H, a, 298D and A/ꢀh the values of 3.1, 4.0 ꢃ, 2.25ꢁ10ꢀ5 cm2 sꢀ1
ꢀ3.5 radsꢀ1 were used, respectively. DOTAMA-En=DOTA monoamide-ethylene-
,
diamine
urements on the corresponding EuIII complex. The best-fit pa-
rameters are listed in Table 2 and compared with those of Gd-
DO3A[36] for data obtained at acidic pH and of a representative
neutral Gd-DOTA-monoamide complex,[37] which resembles
more closely the coordination cage of our GdIII complexes at
basic pH when the amines are bound to the metal centre. The
NMRD profiles of all GdIII complexes at both acidic and basic
pH are characterized by a simple shape featuring a plateau at
low fields, a single dispersion centred around 4–8 MHz and an-
other plateau in the high-field region (>20 MHz), quite typical
1
of rapidly tumbling complexes. Fitting the H NMRD data al-
lowed the confirmation of hydration and assessment of rota-
tional motion of the GdIII complexes in solution. Rotational cor-
relation times (tR) of the GdIII complexes at basic pH were well
in agreement with those of other low molecular weight GdIII
chelates increasing with increasing the molecular weight
within the series, from Gd-AE-DO3A, which contains a primary
Figure 6. Temperature dependence of r1p for Gd-AE-DO3A, Gd-MAE-DO3A,
Gd-DMAE-DO3A, and Gd-MEM-AE-DO3A (pH 7.0, 20 MHz).
amine, to Gd-MAE-DO3A, which bears a secondary dialkyla-
mine, to Gd-DMAE-DO3A and Gd-MEM-AE-DO3A, carrying ter-
tiary amines. On the contrary, slightly higher tR values were ob-
tained for the protonated species at acidic pH. This could be
due either to a larger solvation sphere around the protonated
amine, which causes an increase of the hydrodynamic radius
of the species in solution, or to the moving away from the GdIII
centre of the aminoethyl substituent upon its protonation,
which results in a less compact molecule with higher molecu-
A further increase was observed for Gd-MEM-AE-DO3A (13.4ꢁ
106 sꢀ1). This behaviour can be attributed both to an increase
in basicity of the amine, as already shown by the differences in
pKaH reported previously, and to the steric hindrance of the
amino group. In fact, the coordination between the amine and
the GdIII ion causes a destabilisation of the GdIIIꢀO(water) bond
as expected on the basis of the stereoelectronic effect of the
alkyl groups on the pendant amine. A similar trend was report-
ed in the case of the GdIII complexes with DO3AM (1,4,7,10-tet-
raazacyclododecan-1,4,7-triacetamide) and Me-DO3AM (10-
1
lar volume. The analysis of both H NMRD and 17O NMR data
yielded electronic spin parameters (D2, tV) in good agreement
with literature values. The most interesting finding is related to
a significant variation in water exchange regime observed
within the series and upon pH modulation. A qualitative indi-
cation can be also drawn from the temperature dependence
of the relaxivity at both acidic and basic pH that was measured
for all complexes to further confirm the presence of fast or
slow water exchange regimes (Figure 6). In fact, a steep expo-
nential decay with temperature is indication of fast exchange
methyl-1,4,7,10-tetraazacyclododecan-1,4,7-triacetamide),
in
which a secondary and a tertiary amine, respectively, coordi-
nate the metal centre and induce a 2.5-fold increase in the
water-exchange rate.[38] In the case of Gd-MEM-AE-DO3A, the
bulkier methoxyethyl group may be responsible for a further
increase of kex. It should be noted that only the q=1 fraction
of the GdIII complexes at basic pH was considered in the fitting
of the 17O NMR data, whereas at acidic pH the presence of
1
regime, in which a low value of tm (kex =1/tm) does not limit H
inner-sphere relaxivity, whereas a shallower variation with tem-
Chem. Eur. J. 2014, 20, 2933 – 2944
www.chemeurj.org
2938
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
a mixture of q=1 and q=2 allowed us to obtain average
values of parameters of the two states. In general, the analysis
of the 17O data of the GdIII complexes at acidic pH allows us to
emphasize a particularly low water-exchange rate for Gd-MAE-
DO3A, Gd-DMAE-DO3A, and Gd-MEM-AE-DO3A confirmed
also by the shape of the temperature-dependence curve
(Figure 6), which resembles closely that of the triply charged
Gd-DOTAM (1,4,7,10-tetraazacyclododecan-1,4,7,10-tetraaceta-
mide) complexes.[39] This result suggests a possible hydrogen-
bonding interaction between the coordinated water molecule
and the protonated amine, either directly or involving an inter-
mediate water molecule. A similar behaviour was observed for
ternary complexes between Gd-DO3AM and fluoride or phos-
phate anions.[12] On the other hand, a possible explanation for
the fast water-exchange rate determined at acidic pH for Gd-
AE-DO3A is that the protonated complex is present in a coordi-
nation isomer with a fast kex influencing the average tm value
dramatically. In fact, it is well-known that the two TSAP and
SAP isomers of Gd-DOTA-like derivatives are associated with
markedly different inner-sphere water-exchange dynamics.[40]
½HiLꢃ
KiH
¼
ð2Þ
½Hiꢀ1Lꢃ½Hþꢃ
in which i=1, 2…6. The protonation scheme of DOTA-like li-
gands was well characterized by both spectroscopic and po-
tentiometric methods.[41–43] Thus, the first and second protona-
tion processes occur at two diagonal ring nitrogen atoms,
whereas the third and fourth protonations take place at the
carboxylate groups attached to the non-protonated ring nitro-
gen atoms due to larger charge separation, H-bond network
and electrostatic repulsion between the protonated donor
atoms.[41,43]
A comparison of protonation constants of AE-DO3A and
DMAE-DO3A with those of DOTA and DO3A, indicates that
logK1 and logK2 of AE-DO3A are the highest (Table 3),
whereas logK4 and logK5 are comparable with those of
DOTA and DO3A. The higher first and second protonation con-
stants of AE-DO3A might be explained by the formation of
a H-bond between the protonated ring nitrogen and the pri-
mary amino group stronger than that formed between the
protonated ring nitrogen and acetate groups in DOTA.[41] On
H
H
H
H
H
H
the other hand, in case of DMAE-DO3A, logK1 and logK2 are
lower than those of AE-DO3A, possibly due to a weaker H-
bond between the protonated ring nitrogen atoms and the di-
methylamino group of the side chain for its steric hindrance.
Although the pKa of simple ethyl- and dimethylethylamines are
10.7 and 10.4, respectively,[44] the third protonation of AE-
DO3A and DMAE-DO3A should take place at the nitrogen
atom of the pendant amino group. These values lower than
the free isolated amines can be explained by the electrostatic
repulsion between the first, second and the entering third
proton, and by the presence of a H-bond between the proton-
ated ring-nitrogen atoms and the amino group of the pendant
Solution equilibrium studies
The assessment of the thermodynamic stability of LnIII com-
plexes with this class of aminoethyl-DO3A ligands is of particu-
lar interest because they behave as heptadentate ligands in
acidic conditions and octadentate at higher pH values when
the pendant amine coordinates the metal ion. These two coor-
dination environments are characteristic of DO3A- and DOTA-
like ligands, respectively. In this regard, the stability constants
(logKLnL) of Ln-DOTA-like complexes are 3–4 orders of magni-
tude higher than those of the corresponding Ln-DO3A-like
complexes. The solution thermodynamic and kinetic study was
carried out only on AE-DO3A and DMAE-DO3A ligands be-
cause they represent the two extremes of the series of this
class of DO3A-like ligands having primary to tertiary amines on
the pendant arms.
H
H
arm. Finally, logK4 and logK5 correspond to the protonation
of two carboxylate groups. The SlogKiH values (Table 3) indi-
cate that the total basicity of AE-DO3A and DMAE-DO3A is
higher than that of DOTA and DO3A because of the relatively
high protonation constant of the pendant amine nitrogen
atom. Since the protonation of the different donor atoms does
not take place independently, the total basicity of ligands indi-
rectly affects the stability of its metal complexes. Therefore,
the thermodynamic stability of AE-DO3A and DMAE-DO3A
complexes, in the case of the coordination of the pendant
amino group, should be comparable or even higher than
those of DOTA and DO3A complexes. The coordination of this
amine to the metal ion may play an important role in the com-
plexation, since it increases the denticity of the ligand from
heptadentate to octadentate. On the other hand, it is well-
known that non-ionic donor atoms are more weakly bound to
LnIII ions than ionic donor atoms, so the equilibrium study of
the metal complexes of AE-DO3A and DMAE-DO3A would
give new insights into the effects on the complex stability of
the interaction between the non-protonated amino group and
the metal ions.
Protonation equilibria: The protonation constants of H3AE-
DO3A and H3DMAE-DO3A, as defined by Equation (2), were
determined by pH potentiometry. The logKiH values obtained
are listed and compared with those of H4DOTA and H3DO3A in
Table 3 (standard deviations are shown in parentheses).
Table 3. Protonation constants of H3AE-DO3A, H3DMAE-DO3A, H3DO3A,
and H4DOTA (0.1m KCl, 298 K).
H3AE-DO3A
H3DMAE-DO3A
H3DO3A[a]
H4DOTA[a]
H
H
H
H
H
H
logK1
logK2
logK3
logK4
logK5
logK6
12.49 (2)
10.38 (2)
8.80 (2)
4.03 (2)
1.70 (3)
–
11.92 (1)
9.84 (2)
8.72 (2)
4.25 (2)
1.93 (2)
–
11.99
9.51
4.30
3.63
1.84
–
11.41
9.83
4.38
4.63
1.92
1.58
33.75
SlogKiH
37.40
36.83
31.26
Complexation properties: The stability and protonation con-
stants of AE-DO3A and DMAE-DO3A complexes with several
[a] Ref. [42].
Chem. Eur. J. 2014, 20, 2933 – 2944
www.chemeurj.org
2939
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 4. Stability (logKML) and protonation (logKMHiL) constants of the CaII, ZnII, CuII, and LnIII complexes of the AE-DO3A, DMAE-DO3A, DO3A and DOTA
ligands (0.1m KCl, 298 K).
H3AE-DO3A
H3DMAE-DO3A
H3DO3A[a]
H4DOTA[a]
H3AE-DO3A
H3DMAE-DO3A
H3DO3A[a]
H4DOTA[a]
CaL
14.00 (5)
7.44 (5)
–
12.68 (1)
8.72 (1)
5.20 (3)
20.09 (5)
8.75 (3)
3.92 (2)
2.63 (2)
24.80 (7)
8.69 (3)
3.79 (2)
1.47 (2)
18.50 (9)
12.57
4.60
–
21.57
3.47
2.07
–
25.89
3.65
1.81
–
16.11
3.67
–
LaHL
LaH2L
LaH3L
GdL
GdHL
GdH2L
GdH3L
LuL
6.64 (3)
8.21 (3)
–
–
CaHL
CaH2L
ZnL
*6.27
*7.61 (2)
*4.91 (1)
22.40 (6)
5.91 (7)
*5.58 (3)
20.56 (1)
7.98 (4)
21.45 (6)
7.48 (3)
4.10 (1)
2.14 (5)
25.36 (4)
7.35 (3)
3.67 (2)
1.41 (8)
20.23 (3)
20.21
4.12
3.49
2.56
24.83
4.12
3.57
0.87
21.7[b]
21.56
–
*5.93
24.7[b]
–
ZnHL
ZnH2L
ZnH3L
CuL[c]
CuHL
CuH2L
CuH3L
LaL
*7.25 (2)
*4.89 (5)
22.09 (4)
6.53 (7)
*6.01 (4)
20.63 (8)
7.73 (3)
21.44
–
*5.69
25.4[b]
–
LuHL
LuH2L
LuH3L
*7.46 (2)
*4.44 (4)
*5.71 (3)
18.63
[a] Ref. [42]; [b] ref. [48]; [c] measured by spectrophotometry (298 K). [*]=Stability constants of the protonated *Ln(HiL) intermediate *KLn(HiL)
[Ln(HiL)]/[Ln3+][HiL], where i=3 for AE-DO3A and DMAE-DO3A; i=2 for DO3A and DOTA.
=
metal ions were investigated by pH-potentiometric, UV/Vis
It must be highlighted that the protonation and/or stability
constants of GdIII-, CeIII- and CuII complexes with AE-DO3A and
DMAE-DO3A were independently determined by using other
experimental methods and very similar logK values were ob-
tained, thus confirming the accuracy of the measured data
(see the Supporting Information for details). In particular,
logKGdL =22.36(8) and logKGdHL =5.95(2) (for L=AE-DO3A) and
logKGdL =20.72(7) and logKGdHL =7.82(3) (for L=DMAE-DO3A)
were determined by measuring the relaxivities of equilibrium
solutions of GdIII/AE-DO3A and GdIII/DMAE-DO3A systems in
separate samples (out-of-cell method) in the pH range 2.8–4.0
(see the Supporting Information, Figure S7, 20 MHz, 0.1m KCl,
298 K). Moreover, protonation constants of Ce-AE-DO3A
(logKCeHL =6.33(3)) and Ce-DMAE-DO3A (logKCeHL =8.20(8)), Cu-
1
spectrophotometric and H NMR relaxometric methods (298 K,
0.1m KCl; Table 4). The formation of the protonated *Ln(HiL) in-
termediates were also considered in our equilibrium studies as
already reported for Ln-DOTA complexes.[45–47] The experimen-
tal details and the equations used for the evaluation of the
equilibrium data are summarized in the Supporting Informa-
tion.
The stability constants of CaII- and LnIII complexes with AE-
DO3A are generally about 1–2 orders of magnitude higher
than those of the corresponding DMAE-DO3A and DO3A com-
plexes. The logKML of ZnII- and CuII complexes of AE-DO3A are
similar to those observed for DO3A and higher than those
with DMAE-DO3A and DOTA probably due to its higher first
and second protonation constants. The similar stability of CaII-
and LnIII complexes with DMAE-DO3A and DO3A indicates
that the pendant dimethylamino group does not affect signifi-
cantly the logKML values likely due to its steric hindrance in the
coordination of the metal ions.
AE-DO3A (logKCuHL =7.64(5)) and Cu-DMAE-DO3A (logKCuHL
=
8.74(6)) were determined by measuring the UV/Vis absorption
spectra at different pH values (see the Supporting Information,
Figures S5, S6, S9 and S10). Additionally, the pendant amino ni-
trogen atom of CaII-, ZnII- and CuII complexes with AE-DO3A
and DMAE-DO3A can be protonated leading to logKMHL values
The stability constants of Ln-AE-DO3A and Ln-DMAE-DO3A
complexes increase from LaIII to GdIII, then they remain practi-
cally constant for the lanthanides with lower size (Table 4) as
already found for Ln-DO3A and Ln-DOTA complexes.[7] This
result clearly indicates that the size match between the lantha-
nide ions and the coordination cage determined by the ring-
nitrogen atoms, the acetate oxygen atoms, and the amino N-
atom is optimal for GdIII. The protonation constants of the
amino group of Ln-AE-DO3A and Ln-DMAE-DO3A complexes
depends on the nature of the LnIII ion (Table 4) as also ob-
served for Ln-DO3A-SA complexes.[42] In particular, logKLnHL for
Ln-AE-DO3A are in the range 5.9–6.7, 2–3 orders of magnitude
lower than the correspondent protonation constant of the free
H
(Table 4) similar to logK3 of the free ligands. This finding indi-
cates that in these complexes the pendant amines are either
uncoordinated or are only weakly coordinated giving rise to
a structure very similar to analogous DO3A complexes. More
details on the structural behaviour of Cu-AE-DO3A and Cu-
DMAE-DO3A are given in the Supporting Information in Figur-
es S10–S12. Finally, these CaII-, ZnII- and CuII complexes, similar-
ly to those of the DOTA and DO3A, form di- and triprotonated
species at lower pH values, in which one and two free carbox-
ylate groups are protonated, respectively.
Kinetic studies
H
ligand (logK3 ). These findings indicate that the interaction be-
tween the LnIII ions and the free amine is strong and can be re-
sponsible for the higher logKLnL of Ln-AE-DO3A with respect
to Ln-DO3A. On the other hand, logKLnHL of Ln-DMAE-DO3A
are in the range of 7.7–8.2, decreasing from LaIII to LuIII, and
Complex formation kinetics: The complex formation reactions
between AE-DO3A and DMAE-DO3A and LnIII ions were found
to be slow in the pH range 4–6 in which the protonated Ln-
HAE-DO3A and Ln-HDMAE-DO3A species are present. It is
well-established that the formation of the LnIII complexes with
macrocyclic ligands is generally slow;[45–47,49–53] for example, the
introduction of LnIII ions into the rigid and preformed coordi-
H
closely similar to logK3 of the free ligand. Thus, it can be as-
sumed that in this case the interaction between dimethylami-
no group and LnIII ions is relatively weak.
Chem. Eur. J. 2014, 20, 2933 – 2944
www.chemeurj.org
2940
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
nation cage of DOTA-like ligands is slow because of the forma-
tion of the unusually stable diprotonated intermediates, *Ln-
H2DOTA, which could be detected by spectrophotometry,[45,49]
1H NMR[50] and luminescence spectroscopy.[46,47] The lumines-
cence lifetime study of the intermediate *Eu-H2DOTA revealed
that the number of H2O molecules directly coordinated to the
EuIII was between four and five, suggesting the out-of-cage co-
ordination by four carboxylates with two diagonal nitrogen
atoms of the ring being protonated.[46,47,49,50] The stability con-
stants of the intermediates were also calculated from the data
obtained by kinetic,[50] direct pH-potentiometric,[45] UV/Vis[45]
and luminescence spectroscopy measurements.[46,47] The first
step of the complexation was shown to be the rapid formation
of the diprotonated *Ln-H2DOTA intermediate in equilibrium
with a monoprotonated *Ln-HDOTA intermediate. The depro-
tonation of the latter, that is an OHꢀ-assisted reaction is the
rate-determining step of Ln-DOTA formation.[45–47,49,51–53]
The rates of the formation reactions of CeIII- and EuIII com-
plexes with AE-DO3A and DMAE-DO3A were studied by spec-
trophotometry in the presence of 5–40 fold excess of LnIII ions
in the pH range 4.5–6.0. The plots of kobs values versus [LnIII]
gave saturation curves, which clearly indicates the formation of
reaction intermediates (see the Supporting Information, Figur-
es S11 and S12). This phenomenon can be interpreted by the
rapid formation of a triprotonated intermediate that can be
deprotonated and rearranged to the final product (Ln-HAE-
DO3A and Ln-HDMAE-DO3A) in a slow, rate-determining step.
As already shown in the thermodynamic study section, the for-
mation of the Ln-AE-DO3A and Ln-DMAE-DO3A complexes
occurs through the formation of a triprotonated *Ln-H3L out-
of-cage complex (intermediate) in which three carboxylate
groups are coordinated to the LnIII ion and two diagonal ring
and the amino nitrogen atoms are protonated. The rate-deter-
mining step of the formation of the Ln-HAE-DO3A and Ln-
HDMAE-DO3A complexes is then the deprotonation of a ring
nitrogen of the *Ln-H2AE-DO3A and *Ln-H2DMAE-DO3A inter-
mediates and the rearrangement to the final product. The pro-
posed mechanism for the formation of Ln-HAE-DO3A and Ln-
HDMAE-DO3A complexes is shown in Scheme 3. The defini-
tions and equations used for the evaluation of the kinetic data
are reported in the Supporting Information.
from the kinetic data are 2.2(1) and 1.8(1) for AE-DO3A and
2.2(1) and 2.5(1) for DMAE-DO3A, respectively. These values
are higher than the conditional stability constants of the dipro-
tonated *Ln-H2-1,7-DO2A (1,7-DO2A=1,4,7,10-tetraazacyclodo-
decane-1,7-diacetic acid) intermediates (*Ce-H2-1,7-DO2A: 1.98,
*Yb-H2-1,7-DO2A: 1.60)[52] and lower than those reported for
the diprotonated *Ln-H2DO3A and *Ln-H2DOTA intermediates
(*Gd-H2DO3A: 3.48, *Ce-H2DOTA: 4.4, *Eu-H2DOTA: 4.3).[45,51]
The electrostatic repulsion between the LnIII ion and the pro-
tonated pendant amino group may be responsible for the re-
*
duction of KCLnðH LÞ values for *Ln-H3AE-DO3A and *Ln-
3
H3DMAE-DO3A. However, it can be assumed that the structure
of *Ln-H3AE-DO3A and *Ln-H3DMAE-DO3A, that is, the
number of the acetate arms coordinated to the LnIII ion, is simi-
lar to those of *Ln-H2DO3A intermediates.
The rate constants kf as a function of [OHꢀ] obtained for the
formation of the CeIII- and EuIII complexes of AE-DO3A and
DMAE-DO3A are shown in Figure 7. These data indicate that kf
displays a second-order dependence on the OHꢀ concentra-
tion, expressed by Equation (3):
2
kf ¼ kOH½OHꢀꢃ þ kOH2 ½OHꢀꢃ
ð3Þ
This kind of rate law is quite unusual as, typically, kf for Ln-
DOTA-like complexes shows a first-order dependence on
[OHꢀ].[45–47,49–51,53] For these systems kOH and kOH2 are the rate
constants characterizing the OHꢀ-assisted deprotonation and
rearrangement of the out-of-cage intermediates to the final
complexes. Similar second-order dependence of kf on [OHꢀ]
was reported for the formation of LnIII complexes with DOTA-
phosphonate derivatives.[54] The rate constants kOH and kOH2
were calculated by fitting kf rate constants (Figure 7) to Equa-
tion (3) and are reported in Table 5 and Table 6 compared to
analogous data of Ln-DOTA and Ln-DOTA-like complex forma-
tion.
The first-order dependence of kf on [OHꢀ], expressed by the
kOH rate constants, can be interpreted by the rapid deprotona-
Scheme 3. The suggested pathway for the formation of Ln-HAE-DO3A and
Ln-HDMAE-DO3A complexes.
*
*
In Scheme 3, KLCnðH LÞ, KLHnðH LÞ and kf are the conditional sta-
3
2
bility constants of *Ln-HAE-DO3A and *Ln-HAE-DO3A, the pro-
tonation constants of *Ln-H2AE-DO3A and *Ln-H2DMAE-DO3A
and the rate constants characterising the deprotonation and
rearrangement of the intermediates to the Ln-HAE-DO3A and
Ln-HDMAE-DO3A complexes. The conditional stability con-
Figure 7. The kf rate constants for the formation reactions of the a) Ln-HAE-
III
DO3A, and b) Ln-HDMAE-DO3A complexes of CeIII
tion of [OHꢀ] (0.1m KCl, 298 K).
(
) and Eu ( ) as a func-
&
^
stants (*KLCnðH LÞ) of *Ce-H3L and *Eu-H3L intermediates obtained
3
Chem. Eur. J. 2014, 20, 2933 – 2944
www.chemeurj.org
2941
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Dissociation kinetics: The kinetic inertness of lanthanide com-
plexes is an important issue for their applicability in medical di-
agnosis since their products of dissociation, both free LnIII ions
and ligands, are toxic.[7] In vitro studies of the rate of dissocia-
tion reactions of GdIII complexes may provide important infor-
mation concerning their kinetic behaviour under physiological
conditions. The dissociation of Ln-DOTA and Ln-DOTA-like
complexes is very slow and generally occurs through a proton-
assisted pathway without the involvement of endogenous
metal ions like ZnII and CuII.[45,55,56] The kinetic stabilities of GdIII
complexes are characterized either by the rates of their dissoci-
ation measured in 0.1m HCl or by the rates of the transmetal-
lation reaction, occurring in solutions with ZnII and CuII or EuIII
ions.[42,45,55–59] To obtain information about the kinetic inertness
of GdIII complexes formed with this class of DO3A-like ligands
bearing aminoethyl pendant arms, the metal-exchange reac-
tions of Gd-AE-DO3A and Gd-DMAE-DO3A complexes with
CuII were investigated at high CuII concentrations (10–20-fold
excess) to guarantee the pseudo-first-order conditions. These
reactions [Eq. (4)] were followed by spectrophotometry on the
absorption band of the Cu-AE-DO3A and Cu-AE-DO3A at
300 nm in the pH range 3.2–5.2 (see the Supporting Informa-
tion, Figures S13 and S14).
Table 5. The kOH [mꢀ1 sꢀ1
DO2A and DOA3P complexes with CeIII and EuIII (0.1m KCl, 298 K).
] of AE-DO3A, DMAE-DO3A, DOTA, DO3A, 1,7-
CeIII
EuIII
AE-DO3A3ꢀ
DMAE-DO3A3 ꢀ
DOTA4ꢀ [a]
(1.4ꢂ0.1)ꢁ105
(1.7ꢂ0.2)ꢁ106
(1.8ꢂ0.1)ꢁ106
1.1ꢁ107
2.1ꢁ107 (GdIII)
2.5ꢁ105 (YbIII)
2.2ꢁ104 (GdIII)
(6.5ꢂ0.1)ꢁ105
3.5ꢁ106
DO3A3ꢀ [b]
–
1,7-DO2A2ꢀ [c]
DOA3P7ꢀ [d]
2.8ꢁ105
–
[a] Ref. [45]; [b] ref. [51]; [c] ref. [53]; [d] ref. [54]. DO3AP=1,4,7,10-tetra-
azacyclododecane-4,7,10-tris(carboxymethyl)-1-methylphosphonate
Table 6. The kOH2 [mꢀ2 ꢀ1] rate constants characterizing the formation of
s
AE-DO3A, DMAE-DO3A and DOA3P complexes formed with CeIII and EuIII
(0.1m KCl, 298 K).
CeIII
EuIII
AE-DO3A3ꢀ
DMAE-DO3A3ꢀ
DOA3P7ꢀ [a]
(1.1ꢂ0.1)ꢁ1014
(4.2ꢂ0.1)ꢁ1014
–
(9.7ꢂ0.5)ꢁ1014
(4.8ꢂ0.2)ꢁ1014
3.8ꢁ1011
[a] Ref. [54].
GdL þ Cu2þ Ð CuL þ Gd3þ
ð4Þ
The definitions and equations used for the evaluation of the
kinetic data are reported in the Supporting Information. The
rate constants and the protonation constants characterizing
the transmetallation reactions of Gd-AE-DO3A and Gd-DMAE-
DO3A with CuII are listed in Table 7 and are compared with the
tion of the triprotonated to the diprotonated *Ln-H2AE-DO3A
and *Ln-H2DMAE-DO3A intermediates in the equilibrium reac-
tion, followed by the release of the second proton in the OHꢀ-
assisted rate-determining step and a final structural rearrange-
ment of the complex. The kOH values of our systems
(Table 5) are lower than those of Ln-DO3A complexes
probably due to the presence of the positively
Table 7. Rate and equilibrium constants and half-lives (t1/2 =ln2/kd) of dissociation at
pH 7.4 for the dissociation reactions of Gd-AE-DO3A, Gd-HDMAE-DO3A, Gd-DO3Aꢀ
and Gd-DOTAꢀ complexes (298 K).
charged side chain leading to lower stabilities of the
formed triprotonated *Ln(H3L) intermediates. The
second-order dependence of kf rate constants on
[OHꢀ] can be explained by assuming the OHꢀ-assist-
ed deprotonation of the aminoethyl pendant arm,
which can then act as a general base for the proton
transfer and rearrangement of the out-of-cage inter-
mediates to the final Ln-HAE-DO3A and Ln-HAE-
DO3A complexes. The validity of general base cataly-
sis was proved for Ln-DOTA and some Ln-DO3A-like
complexes,[51,53] for which the rate-controlling step is
the deprotonation of a *LnꢀH2L intermediate fol-
lowed by a structural rearrangement with the LnIII ion
settling into the “coordination cage” formed by the
ring nitrogen atoms and the carboxylate oxygen
Gd-AE-DO3A
Gd-HDMAE-DO3A
Gd-DO3A[a]
Gd-DOTA[b]
ionic strength
k0 [sꢀ1
0.1m KCl
0.1m KCl
(2ꢂ8)ꢁ10ꢀ8
(4.3ꢂ0.3)ꢁ10ꢀ2
–
9.5ꢁ107
(pH-pot)
1.7ꢁ10ꢀ9
1.1ꢁ105
0.1m KCl
0.15m NaCl
6.74ꢁ10ꢀ11
1.83ꢁ10ꢀ6
–
]
(ꢀ1ꢂ5)ꢁ10ꢀ8
0.18ꢂ0.03
(8.0ꢂ0.8)ꢁ102
8.1ꢁ105
–
k1 [mꢀ1 sꢀ1
k2 [mꢀ1 sꢀ1
KHGdL
]
]
2.3ꢁ10ꢀ2
–
115
14[c]
(pH-pot)
kd [sꢀ1] pH 7.4
t1/2 [h] pH 7.4
6.9ꢁ10ꢀ9
9.2ꢁ10ꢀ10
2.1ꢁ105
7.3ꢁ10ꢀ14
2.64ꢁ109
2.7ꢁ104
[a] Ref. [6]; [b] ref. [55]; [c] ref. [45].
atoms of the ligand.[45–47,49,51–53] Compared with the DOTA-
phosphonate system, the kOH2 for Ln-HAE-DO3A and Ln-
HDMAE-DO3A are three orders of magnitude higher than for
Gd-DOA3P probably due to the different basicity of the pend-
ant amino- (AE-DO3A: logK3H =8.8, DMAE-DO3A: logK3H =8.7,
Table 3) and phosphonate groups (DOA3P: logK3H =7.7,
logK4H =6.33, logK5H =5.13).[54]
corresponding values for Gd-DOTA and Gd-DO3A. The half-
lives (t1/2) of dissociation of GdIII complexes calculated at pH 7.4
are included in Table 7.
Considering the pH range investigated (3.2–5.2) and the dif-
ferent speciation of the two GdL complexes, the transmetalla-
tion reactions occur between CuII ions and the monoprotonat-
ed Gd-HDMAE-DO3A, Gd-HAE-DO3A and the deprotonated
Gd-AE-DO3A complexes. For Gd-AE-DO3A the reaction takes
place through spontaneous dissociation of the deprotonated
Chem. Eur. J. 2014, 20, 2933 – 2944
www.chemeurj.org
2942
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
(k0) and monoprotonated (k1) complexes, whereas the monop-
rotonated complex can also dissociate through the proton-as-
sisted (k2) pathway. On the other hand, for Gd-DMAE-DO3A, k0
and k1 characterize the spontaneous and proton-assisted disso-
ciation of the monoprotonated Gd-HDMAE-DO3A complex
that predominates in this pH range. This behaviour is similar to
Gd-DO3A and Gd-DOTA, in which k0 and k1 characterize the
spontaneous and proton-assisted dissociation of GdL. For the
direct comparison of the kinetic inertness, the dissociation
half-lives (t1/2, hours) of GdL were calculated for pH 7.4 by
using the rate and protonation constants reported in Table 7.
The data indicate that t1/2 for Gd-AE-DO3A is 4-, 8-, and 105
times shorter than for Gd-HDMAE-DO3A, Gd-DO3A and Gd-
DOTA complexes, respectively, whereas the dissociation rate
for Gd-HDMAE-DO3A is only two times faster than that of Gd-
DO3A. The interpretation of dissociation kinetic data is some-
how similar to that reported for Gd-DOTA and Gd-
DO3A.[42,45,55–57] The complex dissociation mechanism typically
involves proton transfer from a protonated species to a ring ni-
trogen, causing the release of GdIII from the macrocyclic cavity.
The proton transfer occurs more likely in a less rigid complex
(i.e., Ln-DO3A), which can dissociate much faster than Ln-
DOTA.[42,45,55–57] We have already assumed that the protonation
of Gd-AE-DO3A and Gd-DMAE-DO3A takes place at the amino
group of the pendant arm, so the resulting monoprotonated
complexes highly resemble Gd-DO3A. Moreover, the faster dis-
sociation of Gd-HAE-DO3A and Gd-HDMAE-DO3A complexes
with respect to Gd-DO3A can be explained by the easier
proton transfer mediated by the protonated pendant amino
group. The lower basicity of the aminoethyl pendant arm than
the dimethylaminoethyl is responsible for the lower kinetic in-
ertness of Gd-HAE-DO3A since the conjugated acid transfers
more efficiently the proton to the ring nitrogen. Finally, it is
noteworthy that the t1/2 values of Gd-AE-DO3A and Gd-
HDMAE-DO3A are much higher than the half-lives of dissocia-
tion of the open-chain complexes, Gd-DTPA-BMA or Gd-DTPA
5) weaker coordination of the amino group to the metal ion;
6) higher kinetic inertness.
The behaviour of the protonated complexes was also inves-
tigated in detail, highlighting properties different from those
of the simple correspondent DO3A complexes. Therefore, it
was shown that the protonated amine on the pendant arm in-
fluences both the relaxometric and kinetic properties probably
through the formation of a network of hydrogen bonds be-
tween the ammonium ion and the coordinated water mole-
cule. This was mirrored by experimental evidence of both
water-exchange rate reduction and faster dissociation of the
GdIII complex. An unusual mechanism for complex kinetic for-
mation was proposed, which involves a second-order depend-
ence of the rate constant on [OHꢀ] explained by the action of
the amine as general base in the deprotonation and rearrange-
ment steps.
Overall, this class of ligands constitutes a promising platform
for developing new pH-sensitive probes. A better understand-
ing of the effect of amine N-substitution was gained. Moreover,
the possibility to insert different alkyl groups on the pendant
amine can be exploited to prepare a library of compounds
useful: 1) to increase the relaxivity variation in response to pH
through amplification procedures such as anchoring or encap-
sulation into nanoparticles or protein interaction; 2) eventually
to link the relaxivity change to the concentration of the com-
plex through the preparation of dual imaging reporters.
Acknowledgements
This research was supported by the European Union and the
State of Hungary, co-financed by the European Social Fund in
the framework of TꢄMOP 4.2.4. A/2-11-1-2012-0001 “National
Excellence Program” (grant no.: A2-MZPD-12-0038; Z.B.); the
Italy-Hungary Intergovernmental S&T Cooperation Program
(HU11MO2-TET_10-1-2011-0202); the “Compagnia di San
Paolo” (CSP-2012 NANOPROGLY Project). The research was also
supported by the EU and co-financed by the European Social
Fund under the project ENVIKUT (grant no.:TꢄMOP-4.2.2.A-11/
1/KONV-2012-0043) (A.F.). This study was performed under the
auspices of EU-COST Action TD1004. We thank Fabio Carniato
for his contribute in luminescence lifetime measurements and
Nikolett Gajdics for her help in the equilibrium and kinetic ex-
periments.
(DTPA=diethylenetriamine-N,N,N’,N’’,N’’-pentaacetic
acid;
DTPA-BMA= diethylenetriamine-N,N,N’,N’’,N’’-pentaacetic acid
bis-methylamide) (t1/2 =50 h and 305 h, respectively).[55]
Conclusion
The physicochemical properties of a series of aminoethyl mac-
rocyclic complexes were comprehensively investigated with
a multi-technique approach. The focus of this work was to ex-
plore solution behaviour and the potential of such complexes
as pH-responsive MRI T1 contrast agents. In particular, the in-
creased substitution on the pendant amino group, passing
from a primary amine to a tertiary one, was found to affect
several characteristics of the complexes due to increased steric
hindrance and to higher basicity of the metal-bound N-donor
atom. The following changes were identified: 1) a shift to
higher pH values of the pKaH and larger Dr1p from basic to acid
pH; 2) structural modification of the complexes and presence
of equilibria between differently hydrated species; 3) faster
water-exchange rates when the amino group is coordinated to
the metal centre; 4) lower total basicity of the ligands;
Keywords: contrast agents · lanthanides · NMR spectroscopy ·
macrocyclic ligands · thermodynamics
[1] C. F. G. C. Geraldes, S. Laurent, Contrast Media Mol. Imaging 2009, 4, 1–
23.
[2] a) S. Aime, M. Botta, M. Fasano, S. G. Crich, E. Terreno, J. Biol. Inorg.
Chem. 1996, 1, 312–319; b) R. B. Lauffer, D. J. Parmelee, S. U. Dunham,
H. S. Ouellet, R. P. Dolan, S. Witte, T. J. McMurry, R. C. Walovitch, Radiolo-
gy 1998, 207, 529–538.
[3] A. Datta, K. N. Raymond, Acc. Chem. Res. 2009, 42, 938; b) S. Aime, L.
Calabi, C. Cavallotti, E. Gianolio, G. B. Giovenzana, P. Losi, A. Maiocchi, G.
Palmisano, M. Sisti, Inorg. Chem. 2004, 43, 7588.
[4] H. Kobayashi, M. W. Brechbiel, Curr. Pharm. Biotechnol. 2004, 5, 539–
549.
Chem. Eur. J. 2014, 20, 2933 – 2944
www.chemeurj.org
2943
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[5] W. J. M. Mulder, G. J. Strijkers, G. A. F. Van Tilborg, D. P. Cormode, Z. A.
Fayad, K. Nicolay, Acc. Chem. Res. 2009, 42, 904–914.
[32] S. Aime, M. Botta, D. Parker, J. A. G. Williams, J. Chem. Soc. Dalton Trans.
1996, 17–23.
[6] a) R. B. Lauffer, Chem. Rev. 1987, 87, 901–927; b) P. Caravan, J. J. Ellison,
T. J. McMurry, R. B. Lauffer, Chem. Rev. 1999, 99, 2293–2352; c) P. Her-
[33] L. Banci, I. Bertini, C. Luchinat, Nuclear and Electron Relaxation, The Mag-
netic Nucleus-Unpaired Electron Coupling in Solution, VCH, Weinheim,
1991; N. Bloembergen, L. O. Morgan, J. Chem. Phys. 1961, 34, 842.
[34] L. P. Hwang, J. H. Freed, J. Chem. Phys. 1975, 63, 4017; J. H. Freed, J.
Chem. Phys. 1978, 68, 4034.
[35] T. J. Swift, R. E. J. Connick, J. Chem. Phys. 1962, 37, 307–312; D. H.
Powell, O. M. Ni Dhubhghaill, D. Pubanz, L. Helm, Y. S. Lebedev, W.
Schlaepfer, A. E. Merbach, J. Am. Chem. Soc. 1996, 118, 9333.
[36] S. Aime, M. Botta, S. G. Crich, G. B. Giovenzana, R. Pagliarin, M. Sisti, E.
Terreno, Magn. Reson. Chem. 1998, 36, S200.
[37] L. Tei, G. Gugliotta, Z. Baranyai, M. Botta, Dalton Trans. 2009, 9712–
9714.
[38] M. Botta, S. Aime, A. Barge, G. Bobba, R. S. Dickins, D. Parker, E. Terreno,
Chem. Eur. J. 2003, 9, 2102–2109.
[39] A. L. Thompson, D. Parker, D. A. Fulton, J. A. K. Howard, S. U. Pandya, H.
Puschmann, K. Senanayake, P. A. Stenson, A. Badari, M. Botta, S. Aveda-
no, S. Aime, Dalton Trans. 2006, 5605–5616.
[40] M. Woods, M. Botta, S. Avedano, J. Wang, A. D. Sherry, Dalton Trans.
2005, 3829–3837.
ˇ
ˇ
mann, J. Kotek, V. Kubꢅcek, I. Lukes, Dalton Trans. 2008, 3027–3047.
[7] The Chemistry of Contrast Agents in Medical Magnetic Resonance Imag-
ing, 2nd ed. (Eds.: A. Merbach, L. Helm, E. Toth), Wiley, New York, 2013.
[8] a) P. Caravan, Chem. Soc. Rev. 2006, 35, 512–523; b) S. Aime, D. D. Cas-
telli, S. G. Crich, E. Gianolio, E. Terreno, Acc. Chem. Res. 2009, 42, 822–
831.
[9] a) H. S. Thomsen, S. K. Morcos, P. Dawson, Clin. Radiol. 2006, 61, 905;
b) S. Aime, P. Caravan, J. Magn. Res. Imaging 2009, 30, 1259–1267.
[10] a) H.-S. Chong, K. Garmestani, L. H. Bryant Jr., M. W. Brechbiel, J. Org.
Chem. 2001, 66, 7745–7750; b) Z. Baranyai, F. Uggeri, G. B. Giovenzana,
A. Bꢆnyei, E. Brꢇcher, S. Aime, Chem. Eur. J. 2009, 15, 1696–1705.
[11] S. Achilefu, Chem. Rev. 2010, 110, 2575–2578, and whole thematic
issue.
[12] a) R. Hovland, C. Gløgard, A. J. Aasen, J. Klaveness, J. Chem. Soc. Perkin
Trans. 2 2001, 929–933; b) S. Aime, D. D. Castelli, E. Terreno, Angew.
Chem. 2002, 114, 4510–4512; Angew. Chem. Int. Ed. 2002, 41, 4334–
4336.
[13] S. Aime, M. Botta, E. Gianolio, E. Terreno, Angew. Chem. 2000, 112, 763–
766; Angew. Chem. Int. Ed. 2000, 39, 747–750.
[14] S. Zhang, R. Trokowski, A. D. Sherry, J. Am. Chem. Soc. 2003, 125,
15288–15289.
[41] J. F. Desreux, E. Merciny, M. F. Lincoln, Inorg. Chem. 1981, 20, 987.
[42] A. Takꢀcs, R. Napolitano, M. Purgel, A. Cs. Bꢆnyei, L. Zꢆkꢀny, E. Brꢇcher, I.
Tꢉth, Zs. Baranyai, S. Aime, Inorg. Chem. 2014, submitted.
[43] A. Bianchi, L. Calabi, C. Giorgi, P. Losi, P. Mariani, P. Paoli, P. Rossi, B. Val-
tancoli, M. Virtuani, J. Chem. Soc. Dalton Trans. 2000, 697.
[44] A. E. Martell, R. M. Smith, Critical Stability Constants, Vol. 4, Plenum, New
York, 1974.
[45] ꢊ. Tꢉth, E. Brꢇcher, I. Lꢀzꢀr, I. Tꢉth, Inorg. Chem. 1994, 33, 4070.
[46] S. L. Wu, W. D. Horrocks, Inorg. Chem. 1995, 34, 3724.
[47] J. Moreau, E. Guillon, J. C. Pierrard, J. Rimbault, M. Port, M. Aplincourt,
Chem. Eur. J. 2004, 10, 5218.
[48] W. P. Cacheris, S. K. Niekle, A. D. Sherry, Inorg. Chem. 1987, 26, 958.
[49] E. Brꢇcher, G. Laurenczy, Z. S. Makra, Inorg. Chim. Acta 1987, 139, 141.
[50] X. Wang, T. Jin, V. Comblin, A. Lopez-Mut, E. Merciny, J. F. Desreux, Inorg.
Chem. 1992, 31, 1095.
[51] K. Kumar, M. F. Tweedle, Inorg. Chem. 1993, 32, 4193.
[52] L. Burai, I. Fabian, R. Kiraly, E. Szilagyi, E. Brꢇcher, J. Chem. Soc. Dalton
Trans. 1998, 243.
[15] E. L. Que, C. J. Chang, Chem. Soc. Rev. 2010, 39, 51–60.
[16] a) A. Louie, H. M. Huber, E. T. Ahrens, U. Rothbacher, R. Moats, R. E.
Jacobs, S. E. Fraser, T. J. Meade, Nat. Biotechnol. 2000, 18, 321–325; b) B.
Yoo, M. D. Pagel, J. Am. Chem. Soc. 2006, 128, 14302–14303.
[17] R. J. Gillies, N. Raghunand, M. L. Garcia-Martin, R. A. Gatenby, IEEE Eng.
Med. Biol. Mag. 2004, 23, 57–64.
[18] M. P. Lowe, D. Parker, O. Reany, S. Aime, M. Botta, G. Castellano, E. Gia-
nolio, R. Pagliarin, J. Am. Chem. Soc. 2001, 123, 7601–7609.
[19] S. Aime, M. Botta, S. G. Crich, G. Giovenzana, G. Palmisano, M. Sisti,
Chem. Commun. 1999, 1577–1578.
[20] a) S. Viswanathan, Z. Kovacs, K. N. Green, S. J. Ratnakar, A. D. Sherry,
Chem. Rev. 2010, 110, 2960–3018; b) S. R. Zhang, M. Merritt, D. E.
Woessner, R. E. Lenkiski, A. D. Sherry, Acc. Chem. Res. 2003, 36, 783–790.
[21] a) S. Zhang, K. Wu, A. D. Sherry, Angew. Chem. 1999, 111, 3382–3384;
Angew. Chem. Int. Ed. 1999, 38, 3192–3194; b) D. L. Longo, W. Dastrꢈ,
G. Digilio, J. Kneupp, S. Langereis, S. Lanzardo, S. Prestigio, O. Stein-
bach, E. Terreno, F. Uggeri, S. Aime, Magn. Reson. Chem. 2011, 65, 202–
211.
[53] E. Szilꢀgyi, ꢊ. Tꢉth, Z. Kovꢀcs, J. Platzek, B. Radꢇchel, E. Brꢇcher, Inorg.
Chim. Acta 2000, 298, 226.
[54] F. K. Kꢀlmꢀn, PhD Dissertation, University of Debrecen, 2007.
[55] Z. Baranyai, Z. Palinkas, F. Uggeri, A. Maiocchi, S. Aime, E. Brucher,
Chem. Eur. J. 2012, 18, 16426.
[22] G. B. Giovenzana, R. Negri, G. A. Rolla, L. Tei, Eur. J. Inorg. Chem. 2012,
2035–2039.
[56] ꢊ. Tꢉth, R. Kiraly, J. Platzek, B. Raduchel, E. Brꢇcher, Inorg. Chim. Acta
1996, 249, 191.
ˇ
ˇ
[23] T. Krchovꢀ, J. Kotek, D. Jirꢀk, J. Havlꢅckovꢀ, I. Cꢅsarovꢀ, P. Hermann,
Dalton Trans. 2013, 42, 15735–15747.
[24] A. K. Mishra, J.-F. Chatal, New J. Chem. 2001, 25, 336–339.
[25] D. A. Moore, Org. Synth. 2008, 85, 10–14.
[26] a) J. I. Bruce, R. S. Dickins, L. J. Govenlock, T. Gunnlangsson, S. Lopinski,
M. P. Lowe, D. Parker, J. J. B. Perry, S. Aime, M. Botta, J. Am. Chem. Soc.
2000, 122, 9674; b) L. Burai, R. Kiraly, E. Toth, E. Brucher, Magn. Reson.
Imaging 1997, 38, 146.
[57] P. Wedeking, K. Kumar, M. F. Tweedle, Magn. Reson. Imaging 1992, 10,
641.
[58] L. Sarka, L. Burai, E. Brꢇcher, Chem. Eur. J. 2000, 6, 719.
[59] L. Sarka, L. Burai, R. Kiraly, L. Zekany, E. Brꢇcher, J. Inorg. Biochem. 2002,
91, 320.
[60] K. Kumar, T. Jin, X. Wang, J. F. Desreux, M. F. Tweedle, Inorg. Chem. 1994,
33, 3823.
[27] A. Beeby, I. M. Clarkson, R. S. Dickins, S. Faulkner, D. Parker, L. Royle,
A. S. de Sousa, J. A. G. Williams, M. Woods, J. Chem. Soc. Perkin Trans. 2
1999, 493–503.
[61] H. M. Irving, M. G. Miles, L. Pettit, Anal. Chim. Acta 1967, 38, 475–488.
[62] L. Zꢆkꢀny, I. Nagypꢀl, in Computational Method for Determination of For-
mation Constants (Ed.: D. J. Legett), Plenum, New York, 1985, p.291.
[28] S. Aime, M. Botta, G. Ermondi, Inorg. Chem. 1992, 31, 4291–4299.
[29] S. Hoeft, K. Roth, Chem. Ber. 1993, 126, 869–873.
[30] V. Jacques, F. Desreux, Inorg. Chem. 1994, 33, 4048.
[31] S. Aime, M. Botta, M. Fasano, M. P. M. Marques, C. F. G. C. Geraldes, D.
Pubanz, A. E. Merbach, Inorg. Chem. 1997, 36, 2059–2068.
Received: October 16, 2013
Published online on February 7, 2014
Chem. Eur. J. 2014, 20, 2933 – 2944
www.chemeurj.org
2944
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim