Full text of DOI:10.1016/j.ica.2017.07.071

Accepted Manuscript
Research paper
Remarkable differences and similarities between the isomeric Mn(II)- cis- and
trans-1,2-diaminocyclohexane-N,N,N′,N′ -tetraacetate complexes
Enikő Molnár, Balázs Váradi, Zoltán Garda, Richárd Botár, Ferenc K. Kálmán,
Éva Tóth, Carlos Platas–Iglesias, Imre Tóth, Ernő Brücher, Gyula Tircsó
PII:
S0020-1693(17)30682-5
http://dx.doi.org/10.1016/j.ica.2017.07.071
ICA 17794
DOI:
Reference:
Inorganica Chimica Acta
To appear in:
Received Date:
Revised Date:
Accepted Date:
30 April 2017
20 July 2017
31 July 2017
Please cite this article as: E. Molnár, B. Váradi, Z. Garda, R. Botár, F.K. Kálmán, E. Tóth, C. Platas–Iglesias, I.
Tóth, E. Brücher, G. Tircsó, Remarkable differences and similarities between the isomeric Mn(II)- cis- and
trans-1,2-diaminocyclohexane-N,N,N′,N′ -tetraacetate complexes, Inorganica Chimica Acta (2017), doi: http://
dx.doi.org/10.1016/j.ica.2017.07.071
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Remarkable differences and similarities between the isomeric Mn(II)- cis-
and trans-1,2-diaminocyclohexane-N,N,N′,N′-tetraacetate complexes
Enikő Molnára, Balázs Váradia, Zoltán Gardaa, Richárd Botára,b, Ferenc K. Kálmána,b,c, Éva
Tóthb, Carlos Platas-Iglesiasd, Imre Tótha, Ernő Brüchera and Gyula Tircsóa*
a
Department of Inorganic and Analytical Chemistry, Faculty of Science and Technology,
University of Debrecen, Debrecen, Egyetem tér 1, H-4010, Hungary
b
Centre de Biophysique Moléculaire, CNRS, Université d’Orléans, rue Charles Sadron,
45071 Orléans, Cedex 2, France
c
Le Studium, Loire Valley Institute for Advanced Studies, 1 Rue Dupanloup, 45000 Orléans,
France
d
Centro de Investigacións Científicas Avanzadas (CICA), Departamento de Química,
Universidade da Coruña, Campus da Zapateira-Rúa da Fraga 10, A Coruña, Spain
This paper is dedicated to Prof. Imre Sóvágó on the occasion of his 70th birthday.
Corresponding author e-mail: gyula.tircso@science.unideb.hu
1

Keywords
Magnetic resonance Imaging (MRI), Contrast Agents (CAs), Manganese(II) complexes,
Stability, Inertness, Solvent exchange, Relaxivity
Abstract
Equilibrium, kinetic (solvent exchange and dissociation of the complex) and relaxometric
studies (1H and 17O NMR) have been performed with the [M(II)(c-cdta)]2− complexes (c-cdta
=
cis-1,2-diaminocyclohexane-N,N,N′,N′-tetraacetic acid, M(II)= Mn(II), Zn(II),
Cu(II),Ca(II), Mg(II)) and the physico-chemical data are compared to the isomeric complexes
with trans-1,2-cdta (t-cdta) with the aim of searching appropriate ligands for Mn(II)
complexation for safe MRI contrast agents. The total basicity (Σ log KiH) of the c-cdta ligand
appears to be very similar to that of the trans- derivative under the conditions applied (I=0.15
o
H
M NaCl and 25 C), but the first two protonation constants notably differ. log K1 is 1.5 log
H
units higher, while the log K2 is 0.8 log units lower than those determined for the trans-
derivative. Similar basicity of the ligands results in similar complex stability (log K[Mn(L)]
values are 14.19(2) and 14.32), whereas the conditional stabilities near to physiological pH
are different (pMn values are 7.82 and 8.68) for the [Mn(c-cdta)]2− and the [Mn(t-cdta)]2−
derivatives, respectively. Dissociation kinetic studies revealed that the [Mn(c-cdta)]2−
298
dissociates 250 times faster than the [Mn(t-cdta)]2− complex. The water exchange rate (kex
)
of [Mn(c-cdta)]2– is ca. 60% higher than that of [Mn(t-cdta)]2–. The differences can likely be
attributed to the different distances between the individual donor atoms, and the arrangement
of the donor atoms around the metal ions in the cis- and trans- isomers. Interestingly, the
relaxivity values of the Mn(II) complexes are very close (r1p = 3.79 mM–1s–1 and 3.62 mM–1s–
1; 20 MHz, 25 °C for the cis- and trans-isomers, respectively). DFT calculations were used to
gain insight into the different properties of the [Mn(c-cdta)]2− and the [Mn(t-cdta)]2−
complexes. The results gained in our studies confirm that the trans-1,2-cyclohexanediamine
“building block” displays better features for further ligand development.
Introduction
The linear and macrocyclic aminopolycarboxylate (APC) complexes are widely used in
medical diagnosis and therapy for the complexation of metal ions applied in the field of
2

nuclear medicine, optical and magnetic resonance imaging (MRI) [1-7]. These different fields
have some special claims regarding the properties of complexes, but the general requirements,
which should be accomplished for all the complexes, are the high stability and kinetic
inertness. In other words, it means that the complexes must not dissociate in the body of
patients during or after the examinations are performed. Considering the very complex nature
of biological fluids which contain a great number of metal ions and complex forming ligands,
this is a very strict requirement. To prevent any interaction between the APC complexes in
use and the endogenous metals or ligands, the structure of APC ligands play very important
role. So the ligands have to be specially designed for the robust complexation of metal ions.
To design a ligand, the chemical properties of the metal ion, its ionic size, electronic structure
and coordination number are considered first to find out how many and which kind of donor
atoms are optimal for the complexation. Some special structural elements may also be
beneficial e.g. the rigidity of the ligand and/or its degree of pre-organization to bind the metal
ion, which can be achieved by the use of suitable functional group(s) and/or by introducing
some cyclic or macrocyclic building blocks.
In MRI several Gd(III) – APC complexes are clinically used as contrast agents (CAs) to
increase the contrast between the healthy and diseased tissues or organs by differently
increasing the relaxation rates of protons (mainly water protons) in the body. The ligands used
for metal ion complexation in CAs are the linear dtpa, the macrocyclic dota and their
derivates [6, 7]. Nowadays there are 6 linear and 3 macrocyclic CAs in clinical use, but in the
last decade some concerns arose with the use of the linear agents. It was recognized that a
new disease, called nephrogenic systemic fibrosis (NSF), was associated with the use of the
Gd3+-based CAs. NSF was developed in patients with severe renal impairment, when the
elimination of the CAs was slow. During the longer residence time of the Gd(III) complex in
the body of patients a small amount of Gd(III) remained deposited, which presumably
triggered the development of NSF [8-10]. More recently, the presence of Gd has been
detected in different organs, including the brain of patients with normal renal function after a
few repeated MR scans using CAs [11-14]. The presence of traces of non-chelated Gd(III) in
the body of patients has raised concerns because of the potential toxicity of the Gd(III) ion.
These problems with the use of the linear Gd(III)-based CAs turned the attention of
researchers to the complexes of Mn(II) as potential contrast agents [15-16]. The relaxation
effects of Mn(II) are known to be similar to those of Gd(III), but on the other hand, Mn(II) is
an essential metal ion and some amount of the intravenously administered Mn(II) can be
3

eliminated relatively rapidly through the hepatobiliary system. Since the coordination number
of Gd(III) is 8-9, while that of Mn(II) is often 6-7, the ligands used in the Gd(III)-based CAs
can not be directly used for the complexation of Mn(II). Some examples are known, when the
Mn(II) complexes with ligands performing well with Gd(III) are stable and inert, but the
absence of inner sphere water molecule(s) results in negligible enhance in the contrast [17].
Another critical point is the lability/inertness of the complexes (the terms lability/inertness in
this manuscript describe dissociation kinetics parameters of the complexes rather than the
solvent exchange kinetics). The majority of the open-chain ligands form very labile
complexes with Mn(II). The most shocking example is that of [Mn(dtpa)]3− which was found
to dissociate within the dead-time of the stopped-flow instrument (ca. 8 msec) in the presence
of Cu(II) as a ligand scavenger [18]. In order to find suitable ligands for Mn(II), a number of
new linear and macrocyclic ligands were synthesized during the last years and the properties
of their complexes were investigated extensively [15-24]. Amongst the hexadentate linear
ligands edta forms the relatively stable [Mn(edta)]2− complex in which a water molecule
occupies the seventh coordination site, but the kinetic inertness of the complex is also low,
because of the flexible structure of the ligand [18]. The edta analogue trans-1,2-cdta
(abbreviated as t-cdta through the text, Figure 1) is more rigid and it forms the more stable
and more inert complex [Mn(t-cdta)]2− which seems to be suitable as a potential CA and
recently was evaluated even in preclinical studies [25-26]. The behaviour of the cdta
derivative ligands are likely similar to the parent ligand, so their Mn(II) complexes are
promising for the preparation of Mn(II)-based MRI CAs [18, 22].
While the complexation properties of the t-cdta ligand have been studied extensively,
there is very few knowledge about the complexes of the other stereoisomer, the cis-1,2-cdta
ligand (abbreviated as c-cdta in the text). The donor atoms of t-cdta form a semi-rigid
preorganized coordination cage, where metal ions of different size may enter and generally
form complexes of high stability [27]. The maximum distance between the two N-donor
atoms in t-cdta is 4 Å, while that in c-cdta is only 3.1 Å [28], so the coordination cage formed
by the donor atoms of the cis- isomer is relatively small, anticipating a low stability of the
complexes with c-cdta and large metal ions. Since the size of the Mn(II) ion is not too large
and there are only very few data in the literature regarding the complexation of c-cdta, we
decided to study its complexes with Mn(II) and with some other endogenous metal ions.
4

HO2C
HO2C
CO2H HO2C
CO2H HO2C
CO2H
CO2H
HO2C
HO2C
CO2H
CO2H
N
N
N
N
N
N
c-cdta
t-cdta
edta
CO2H
CO2H
N
N
N
N
HO
N
O
CO2H
CO2H
4-het-t-cdta
Figure 1. Formulae of the edta, c-cdta, t-cdta, 4-het-t-cdta ligands.
Results and Discussion
The synthesis of the c-cdta ligand (Scheme 1) was accomplished in a two-step reaction
sequence starting from cis-cyclohexane-1,2-diamine, which was alkylated by using ethyl
bromoacetate in the presence of Hünig's base (diisopropylethyl amine, dipea). The
saponification of the tetraethyl ester with NaOH afforded the product which was purified by
recrystallization from acidic (pH=2.0) aqueous solution.
Scheme 1. The synthesis of the c-cdta chelator: i). 4.2 equivalent of ethyl bromoacetate, dipea,
MeCN, 7 h reflux; ii). 6 equivalent of NaOH, EtOH, 24 h reflux; iii). cc HCl till pH = 2.0.
Equilibrium studies
In order to characterize the metal complexes as potential MRI contrast enhancement
agents, one should consider their equilibrium and kinetic behaviour in biofluids, the proton
relaxation effects, together with other parameters which determine the proton relaxation rates
in the presence of these complexes. For the assessment of the equilibrium behaviour we have
5

determined the stability constant of the Mn(II) complex and those of the Mg(II), Ca(II), Zn(II)
and Cu(II) complexes that may compete with the Mn(II) ion for the c-cdta ligand in biofluids.
To obtain information about the kinetic inertness of [Mn(c-cdta)]2− complex, the rates of the
exchange reactions occurring between the complex and the Cu(II) ion have been investigated,
vide infra. These properties are already known for the [Mn(t-cdta)]2− complex, so the
comparison of the behaviour of the two complexes is possible.
The protonation constants of the c-cdta ligand have been determined by pH-
potentiometric titration. The protonation constants (KiH) defined by Equation (1), are
presented in Table 1.
[HiL]
[Hi−1L][H+ ]
KiH
=
where i=1, 2, n
(eqn. 1)
where i = 1, 2, …, 5, [HiL], [Hi-1L] and [H+] are the equilibrium concentrations of the
HiL and Hi-1L ligand species and the H+ ions, respectively. The stability constants of the ML
complexes (KML) are defined by Equation (2):
[ML]
[M][L]
KML
=
(eqn. 2)
In Equation (2) [M], [ML] and [L] are the equilibrium concentrations of the M(II) metal
ion, ML complex and L4- ligand, respectively (the charges of ions will be used only when it is
necessary).
The metal ions and protons compete for the ligand, so at lower pH values protonated
complexes can be formed. The protonation constants of complexes are defined as follows:
[MHiL]
[MHi−1L][H+ ]
(eqn. 3)
KMH L
=
i
where i = 1 or 2, [M(HiL)] and [M(Hi-1L)] are the concentration of the protonated
complexes.
Table 1. Protonation constants of the c-cdta, t-cdta and edta ligands and stability
constants of their Mg(II), Ca(II), Zn(II), Cu(II) and Mn(II) complexes (log K values, 25°C,
0.15 NaCl).
6

c-cdta[a]
c-cdta[b]
10.70
5.21
3.50
2.44
−
t-cdta[c]
t-cdta[b]
edta[c]
9.28
H
H
H
H
H
log K1
log K2
log K3
log K4
log K5
9.54
5.97
11.70
6.12
3.52
2.43
−
11.00(1)
5.20(1)
3.41(1)
2.30(1)
1.41(1)
23.32
6.04
2.72
3.60
1.99
2.52
1.11
1.46
∑ log KiH
21.14
7.61
21.85
8.38
4.44
9.45
4.11
−
23.09
9.14
23.77
11.07
−
log KML
log KMHL
log KML
9.01(2)
4.72(6)
9.65(2)
4.55(5)
14.19(2)
2.85(3)
−
−
3.53
9.53
10.23
3.53
14.32[d]
2.90[d]
1.89[d]
8.68
13.15
−
log KMHL
log KMnL
log KMnHL
log KMnH2L
pMn[e]
2.92
12.46[d]
2.95[d]
17.48
2.8
−
[d]
−
−
−
7.83
15.92
3.23
7.82
−
−
log KML
16.75
2.57
19.37
2.9
17.06(1)
2.76(1)
−
−
log KMHL
log KMH2L
log KML
−
1.50
−
−
−
19.02[f]
19.78[f]
2.91
22.0
3.1
18.3(1)[f]
3.65(3)
−
−
log KMHL
log KMH2L
3.15
−
2.04
1.10
−
−
[a] this work; [b] Ref. [27]; [c] 0.15 M NaCl, 25°C from the diploma work of Veronika Józsa,
2015, University of Debrecen, Debrecen, Hungary; [d] Ref. [18]; [e] calculated by using
cMn=cLig=0.01 mM at pH=7.4 according Tóth et al. [29]; [d] determined by simultaneous
fitting of the UV-vis and pH-potentiometic data.
The protonation constants of c-cdta determined in this work at 25°C in 0.15 M NaCl
solution (Table 1) agree relatively well with those reported in the literature [27] (20°C and 0.1
M KCl), indicating that the interaction of c-cdta with the Na+ and K+ ions is negligible. The t-
H
cdta ligand behaves differently, since its logK1 value in 0.15 M KNO3 is much higher
(11.70) than in 0.15 M NaCl (9.54), because the stability constant of the [Na(t-cdta)]2−
complex is relatively high (log KNaL = 4.4) [27], while the interaction of the ligand with K+ is
very week. The different behaviour of the two ligands indicates that the coordination cage
7

formed by the two iminodiacetate (ida) groups of c-cdta is too small to accommodate the
larger M(I) metal ions, while in the case of t-cdta the coordination cage is larger, where even a
Na+ ion can enter.
The stability constant of the Mg(II)- and [Ca(c-cdta)]2− complexes obtained in this work
are comparable with those published in the literature (Table 1). For the c-cdta complexes of
Mn(II), Zn(II) and Cu(II) there are not any stability constant data in literature. The stability
constants of complexes formed with the c-cdta, t-cdta and edta ligands in 0.15 M NaCl (Table
1) do not differ considerably, because of the formation of the [Na(t-cdta)]3− complex.
Considering the formation of Na+ complexes, it is more correct to compare the log KML values
of c-cdta complexes with those log KML values obtained for the t-cdta complexes at 0.1 M KCl
ionic strength. This comparison shows that the stability constants of the [M(c-cdta)]2−
complexes are lower by 2-3 log units than those of the [M(t-cdta)]2− complexes (Table 1).
The paramagnetic metal ions used as a component of the MRI CAs result in high proton
relaxation rates. Since the free metal ions are toxic, their complexes are used as CAs. The
longitudinal relaxation rate of water protons (1/T1 where T1 is the longitudinal relaxation time)
in the presence of Mn(II) ions is significantly higher than in the presence of [Mn(c-cdta)]2− or
[Mn(t-cdta)]2−. The relaxation effect of paramagnetic ions and complexes are expressed by
their relaxivity values. The relaxivity (r1p, mM-1s-1) is the enhancement of the longitudinal
water proton relaxation rate referred to 1.0 mM concentration of the paramagnetic species.
The relaxivity of the [Mn(H2O)6]2+ ion at 25°C and 20 MHz is 7.94 mM-1s-1, while that of the
[Mn(c-cdta)]2− complex is 3.79 mM-1s-1. This latter r1 value is similar to that of the
[Mn(edta)]2− (3.23 mM-1s-1) and [Mn(t-cdta)]2− (3.62 mM-1s-1) complexes, in which one water
molecule is coordinated in the inner sphere of Mn(II).
The formation of [Mn(c-cdta)]2− complexes could be followed by measuring the water
proton relaxation rates in equilibrium solutions of 1.0 mM Mn(II) and 1.0 mM c-cdta as a
function of pH. At pH<2 only [Mn(H2O)6]2+ is present, resulting in high 1/T1 values (Fig. 2).
With the increase of pH, MnHL and MnL complexes are formed, so that the proton relaxation
rates decrease because the water molecules are displaced from the inner sphere of Mn2+. Once
the formation of [Mn(L)] complex is complete (at about pH>5,) the 1/T1 values are constant
(this constant 1/T1 value is the relaxivity of the MnL complex). In Figure 2 the species
distribution curves (calculated from the equilibrium data for the 1.0 mM Mn(II) - 1.0 mM c-
cdta system) are also presented (together with the 1/T1 values). The data in Figure 2 indicate
8

that the decrease of the proton relaxation rates with the increase of pH occurs parallel with the
formation of the MnHL and MnL complexes. The good agreement in the trend of the data
obtained by pH-potentiometry and relaxometry indicates the reliability of the equilibrium
data.
100
Mn2+
[Mn(L)]
7.5
80
60
40
6.5
5.5
4.5
20
0
[Mn(HL)]
3.5
11.5
1.5
3.5
5.5
7.5
9.5
pH
Figure 2. Species distribution curves and the relaxation rate (1/T1-1/T1w) profile (red points) of
1.021 mM [Mn(c-cdta)]2− as a function of pH (T = 25 oC, I=0.15 M NaCl, 20 MHz).
As it was underlined several times in the literature, reliable stability data for Cu(II) –
APC complexes can be obtained either by competition method with the use of auxiliary ligand
(the equilibrium of the Cu(II) – auxiliary ligand system must be well known), or it can be
accessed by using multiple methods simultaneously (e.g. UV-vis spectrophotometry
applicable also in very acidic conditions and pH-potentiometry) [31]. The absorption spectra
of the [Cu(c-cdta)]2− complex as a function of H+ ion concentration is shown on Figure 3. The
spectrophotometric data were fitted simultaneously with the pH-potentimetric titration data
and the stability constants determined are shown in Table 1. In line with the equilibrium data
obtained for other metal ions studied, the stability constant of the [Cu(c-cdta)]2− complex is
9

lower than the value of the corresponding [Cu(t-cdta)]2− complex determined under identical
conditions (the difference is ca. 0.7 log units).
0.4
0.3
0.2
0.1
0
400 450 500 550 600 650 700 750 800
Wavelenght (nm)
Figure 3. Absorption spectra of the [Cu(c-cdta)]2− system as a function of [H+]. Conditions:
o
[c-cdta4−] = 3.528 mM, [Cu(II)] = 3.410 mM, 25 C, 1.0 M (Na++H+)Cl-, cH+ in the samples
are as follows: 12.1; 18.8; 33.9; 49.0; 75.9; 101.0; 156.8; 205.0; 253.3; 349.8 and 462.4 mM
(from top to bottom).
Dissociation kinetics of [Mn(c-cdta)]2− complex
The fate of the MRI CAs in biological fluids is strongly influenced by their kinetic
inertness, that is an important property in determining the extent of dissociation of the
complex while present in the body. For the Gd(III)-based CAs it was generally assumed that
the deposition of Gd in the body was the result of the dissociation of Gd(III) complexes, that
occurred through transmetallation reactions with endogenous metal ions like Zn(II), Cu(II)
and Ca(II) [33, 34]. To obtain information about the kinetic properties of CAs, the rates of the
metal-exchange reactions are generally studied between the Gd(III) or Mn(II) complex and an
exchanging metal ion, very often Cu(II), because in this case the reaction can easily be
10

followed by spectrophotometry. The kinetic behaviour of [Mn(c-cdta)]2− was studied similarly
by following the exchange reactions depicted in Equation (4):
[Mn(c-cdta)]2− + Cu(II)
[Cu(c-cdta)]2− + Mn(II)
(eqn. 4)
Since the stability constant of [Cu(c-cdta)]2− is much higher than that of [Mn(c-cdta)]2−
(Table 1), reaction (4) takes place to completeness in the presence of Cu(II) excess (10-40
fold of Cu(II) excess was used). Under such conditions the reactions can be treated as pseudo-
first-order processes kinetically. The rate of the exchange reaction (4) is directly proportional
to the concentration of the complex and can be expressed as follows:
d[Mn(L)]t
(eqn. 5)
−
= kobs[Mn(L)]t
dt
In Equation (5) kobs is a pseudo-first-order rate constant, [Mn(L)]t is the total
concentration of complexes containing the [Mn(c-cdta)]2− species. The rates of the exchange
reactions (4) have been studied at four different Cu(II) concentrations in the pH range 4.03 –
5.03. The kobs values obtained are shown in Figure 4.
0.5
10x
20x
30x
40x
0.4
0.3
0.2
0.1
0.0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
cH+ (x105 mol/dm3)
11

Figure 4. The rate constants (kobs) characterizing the exchange reactions of [Mn(c-cdta)]2− (0.4
mM) with Cu(II) measured by stopped-flow method. The Cu(II) concentrations are 4.0 mM
(red), 8.0 mM (green), 12.0 mM (blue) and 16.0 mM (wine-red) (25°C, 0.15 M NaCl).
The kinetic data shown in Figure 4 reveal that the kobs values are directly proportional to
[H+], the exchange reactions showing first-order dependence on the H+ ion concentration. The
dependence of the kobs values on the Cu(II) concentration is surprising, because the increase in
the Cu(II) concentration results in a slight decrease in the reaction rates, so that Cu(II) ions
depress the rate of the exchange reactions. Similar phenomena were observed previously for
the reaction of [Mn(t-cdta)]2− with Cu(II) and for the reactions between [Cd(t-cdta)]2− and
Pb(II) or Cu(II) ions [18, 35]. The effect of the Cu(II) ions on the reaction rates can be
interpreted by the formation of dinuclear [Mn(c-cdta)Cu] complexes, where presumably one
acetate group of [Mn(c-cdta)]2− is coordinated to the Cu(II) ion and so the probability of
protonation decreases (in fact [Mn(c-cdta)Cu] is a “dead-end” complex). Since the reaction
rate does not increase with the increase of the Cu(II) concentration, the exchange reactions do
not take place with the direct attack of the Cu(II) ion. On the basis of the experimental data
we came to the conclusion that the exchange reactions between the [Mn(c-cdta)]2− and Cu(II)
ion occur through the dissociation mechanism. After the rate determining dissociation of the
complex, the released c-cdta ligand rapidly reacts with the Cu(II) ions with the formation of
[Cu(c-cdta)]2−. The dissociation of the [Mn(c-cdta)]2− complex can take place spontaneously
and by the assistance of H+ ions. The spontaneous dissociation is presumably very slow, while
the proton assisted dissociation occurs relatively rapidly at pH values around 4-5. The
protonation of [Mn(c-cdta)]2− presumable occurs at a carboxylate group (log KMHL = 2.85),
but for the dissociation of the complex the proton must be transferred to a N atom, when the
protonated iminodiacetate (ida) group is not coordinated anymore and this intermediate may
dissociate (or the complex may be re-formed by the dissociation of the proton). Considering
the possible reaction pathways, the rate of dissociation of the complex can be expressed by
Equation (6)
[Mn(L)]t
(eqn. 6)
−
= k0[Mn(L)]+k MnHL[Mn(HL)]
dt
12

79 23
where k0 and kMnHL are the rate constants characterizing the dissociation of the MnL and
MnHL complexes, respectively.
Comparing the total concentration of the complex [Mn(c-cdta)]2− ([Mn(L)]t = [Mn(L)] +
[Mn(HL)] + [Mn(L)Cu]), Equation (5) and (6), for the kobs value Equation (7) is obtained:
+
k0 +kMnHL⋅K MnHL⋅[H ]
kobs
=
(eqn. 7)
+
1+ KMnHL [H ]+ KMnLCu [Cu2+ ]
The term KMnHL [H+] in the denominator of Equation (7) is very small in the pH range
4.03 – 5.03, so it can be neglected. For the calculation of the unknown parameters Equation
(7) has been used as follows:
+
k0 +k1⋅[H ]
1+ KMnLCu [Cu2+ ]
(eqn. 8)
kobs
=
where k1 = kMnHL KMnHL
.
By fitting the kobs values presented in Figure 4 to Equation (8) the k0, k1 and KMnLCu
values have been calculated and presented in Table 2. The k0 value is very low, so it can be
neglected.
Table 2. Rate and equilibrium constants characterizing the dissociation rates of the [Mn(c-
cdta)]2−, [Mn(t-cdta)]2− and [Mn(edta)]2− complexes.
t-cdta[a,b]
edta[a,b]
c-cdta
k1 (M-1s-1)
k2 (M-2s-1)
k3 (M-1s-1)
(1.02 0.09)×105
4.0×102
5.2×104
2.3×108
45 8
−
−
−
−
−
KMnLCu
79
13

2.85[c]
0.47
log KH
2.90[c]
12
2.65[c]
0.076
t1/2[d] (h)
[a] Ref [18]; [b] for [Mn(t-cdta)]2- k1 = 3.2×102 M–1s–1 and t1/2 = 15 h while for [Mn(edta)]2- k3
= 3.0×10-1 M–1s–1, k4 = ~4.8×101 M–2s–1 and log KMnHL =3.10 were found in Ref. [36]; [c]
determined by pH-potentiometry and fixed in the data fitting; [d] calculated at pH = 7.4 in the
presence of cCu2+ = 1×10-5 M.
In Table 2, the k1 rate constants and KMnLCu stability constants obtained for the [Mn(c-
cdta)]2− complex are compared with those of the [Mn(t-cdta)]2−. The presented data indicate
that the k1 dissociation rate constant of [Mn(c-cdta)]2− is approximately 250 times higher than
that of [Mn(t-cdta)]2−. This difference is shown in the values of the half-life of dissociation,
that is much longer for complex [Mn(t-cdta)]2−. The results of these studies indicate that the
kinetic inertness of the Mn(II) complex formed with the c-cdta isomer is significantly lower
than that of the t-cdta isomer.
17O and NMRD measurements
The nuclear magnetic relaxation dispersion (NMRD) profile of the [Mn(c-cdta)]2– was
measured at three different temperatures (298, 310 and 323 K) in the frequency range 0.01-80
MHz to gain information on the relaxation properties of the complex (Figure 5). The NMRD
profiles show a single dispersion at 1-10 MHz which is typical of low molecular weight Mn2+
complexes. Their temperature dependence indicates that the relaxivity of the complex is
limited by fast rotation. Several physicochemical parameters affect the relaxivity of a
paramagnetic metal complex, such as the water exchange rate, the electron spin relaxation
parameters and the rotational correlation time. The temperature dependence of the transverse
and longitudinal 17O relaxation rates can provide information on the water exchange rate (kex)
and the rotational correlation time (τR), respectively, while the chemical shift is related to the
number of the water molecules directly coordinated to the paramagnetic metal center (q).
Variable temperature transverse and longitudinal relaxation rates and chemical shifts were
measured for an aqueous solution of [Mn(c-cdta)]2– and for a diamagnetic reference at 9.4 T.
The T1 values showed negligible difference between the Mn(II) complex and the reference,
17
thus were not included in the analysis. The reduced O transverse relaxation rates, 1/T2r, and
chemical shifts, ∆ω , have been fitted according to the Swift-Connick equations, by assuming
r
14

a simple exponential behaviour for the electron spin relaxation. The water exchange rate,
298
kex298, its activation enthalpy, ∆H≠, and 1/T1e were calculated (the activation energy of
electron spin relaxation was fixed to 1 kJ/mol). The hydration number was fixed to q=1 based
-
on analogy to [Mn(t-cdta)]2– and on the relaxivity value measured at 20 MHz (r1p=3.79 mM-1s
1, r
=3.62 [18]) We have attempted to determine the hydration number of the complex
1,[Mn(t-cdta)]2–
by using Gale’s method based on the maximum value of the transverse 17O relaxation rate
however the q obtained (q = 0.50 0.20) seems to be an underestimation since the relaxivity of
the given complex as well as the results of DFT calculations (vide supra) are being consistent
1
with q=1.[37] To characterize the rotational dynamics of the complex, the H NMRD data
were analyzed by the Solomon-Bloembergen-Morgan (SMB) [38] and Freed models [39]
(inner- (IS) and outer-sphere (OS) relaxation mechanisms) and the Swift-Connick [40, 41]
equations related to solvent exchange (see Supporting Information). The parameters for water
exchange were fixed to those determined from the 17O NMR study. The distances between the
metal ion and the inner and outer sphere water protons were also fixed (rMnH = 2.83 Å and
aMnH = 3.6 Å), as well as the diffusion coefficient and its activation energy (DMnH = 26 × 10-10
m2s-1 and EMnH = 18 kJ/mol-1 [29]) and the activation energy of the modulation of the zero-
field-splitting (Ev = 1.0 kJ/mol). The values of the electron relaxation parameters τ 298 and ∆2
v
were found to be 21 2 ps and (1.0 0.1)×1020 s–2, respectively. The best-fit parameters
obtained are given in Table 3 and compared with those of [Mn(t-cdta)]2– and [Mn(4-het-
cdta)]2– complexes [42, 43]. The experimental data and the fitted curves are shown in Figures
5 and 6.
15

17
16.5
16
15.5
15
14.5
14
7.5
2.9
3.1
3.3
3.5
3.7
6
4.5
3
1.5
0
2.9
3.1
3.3
3.5
3.7
1000/T (1/K)
Figure 5. Variable temperature reduced transverse 17O relaxation rates (top) and chemical
shifts (bottom) recorded for [Mn(c-cdta)]2– (B = 9.4 T). The solid lines correspond to the fit of
the data.
17
1
Table 3. Parameters obtained from the analysis of O NMR and H NMRD data for the
[Mn(c-cdta)(H2O)]2– complex.
16

[d]
[Mn(4-het-cdta)]2–
[c]
[Mn(c-cdta)]2–
22.5 0.5
42.0 0.8
+56 4
[Mn(t-cdta)]2–
Parameter
kex298 (x107 s–1)[a]
∆H‡ (kJ mol–1)[a]
∆S‡ (J K-1 mol–1)[a]
AO/ꢀ (106 rad s–1)[a]
14.0
42.5
–
17.6
36.2
–
-40
-42.7 0.9
27.0 0.5
20.4 0.7
74 2
-26.4
0.65
–
1/T1e298 (x107/ s–1)[a]
ErH (kJ mol–1)[b]
τrH298 (ps) [b]
7
26.2
104.9
–
[a] from 17O NMR; [b] from NMRD; [c] Ref. [42]; [d] Ref. [44].
17
As Figure 5 shows, the reduced transverse O relaxation rates undergo a changeover
at ~290 K between the fast and intermediate water exchange regime. Importantly, the water
exchange rate has a significant contribution to the overall correlation time (1/τc = kex + 1/T1e;
kex varying from 10 to 80% in the temperature range 273-338 K), which makes possible to
298
accurately determine kex. The kex value characterizing the water exchange of [Mn(c-cdta)]2–
is ca. 60% higher than that of [Mn(t-cdta)]2–. The high positive activation entropy points to a
dissociative mechanism, as would be expected for a seven-coordinated Mn(II) complex.
On the other hand, kex298 is half of the value determined for [Mn(edta)]2- (47.1×107 s-1)
[45]. This is related to the more rigid structure of the cdta4− ligands which detains the
rearrangement of the coordination environment during the water exchange process. The water
exchange is also several times slower on [Mn(c-cdta)]2– than on the 1,4-do2a or do1a
complexes, having a very different, macrocyclic structure [46]. The rotational correlation time
(τrH298) obtained from the analysis of the relaxivity data is in the usual order of magnitude of
typical low molecular weight complexes.
17

7.5
6
4.5
3
1.5
0
0.01
0.1
1
10
100
proton Larmor frequency (MHz)
Figure 6. 1H NMRD profiles of [Mn(c-cdta)]2– at 25 °C (♦), 37 °C (ꢀ) and 50 °C (ꢀ).
DFT study
Aiming shed light on the reasons behind the different behaviour of the [Mn(c-
cdta)(H2O)]2– and [Mn(t-cdta)(H2O)]2– complexes, we carried out a theoretical DFT study.
These calculations were performed with the explicit inclusion of two second-sphere water
molecules, which were previously found to be necessary for an accurate determination of
Mn-Owater bond distances and 17O hyperfine coupling constants AO/ꢀ [47]. The X-ray structure
of the ammonium salt of [Mn(t-cdta)(H2O)]2– was reported [48]. Our DFT calculations
(M062X/TZVP, see details below) provide bond distances of the metal coordination
environment in excellent agreement with the solid state data (Supporting Information). The
structure obtained using DFT for [Mn(c-cdta)(H2O)]2– is very similar to that of the trans
analogue, with the bond distances in the two complexes differing by > 0.035 Å (Figure 7). An
overlay of the two calculated structures corroborates the very similar coordination
environments of the two structures (Supporting Information). In spite of the minor differences
in the metal coordination environments, DFT calculations indicate that the [Mn(t-
cdta)(H2O)]2– complex is considerably more stable than the cis analogue. The free energy
18

difference between the two complexes favors the trans isomer by 35.4 kJ mol-1, which is
compatible with a higher stability of [Mn(t-cdta)(H2O)]2– by several orders of magnitude. This
is likely the consequence of the more strained geometry of the ligand in [Mn(c-cdta)(H2O)]2–
to accommodate the relatively large Mn2+ ion.
Figure 7. Structures of the [Mn(c-cdta)(H2O)]2–·2H2O (left) and [Mn(t-cdta)(H2O)]2–·2H2O
(right) systems obtained with DFT calculations.
The bond distances involving the metal ion and the coordinated water molecule
(Mn-O5) obtained for the two systems are very similar (2.264 and 2.271 Å for the cis and
trans complexes, respectively). Since the water exchange reaction in these complexes is
expected to follow a dissociative mechanism, one would expect similar activation enthalpies
for the two systems, as the rupture of the Mn-O5 bond to reach the six-coordinated transition
state should require similar energies. Thus, we explored the potential energy surface of the
[Mn(c-cdta)(H2O)]2–·2H2O and [Mn(t-cdta)(H2O)]2–·2H2O systems by increasing the Mn-O5
distances, which eventually led to the six coordinated systems [Mn(c-cdta)]2–·3H2O and
[Mn(t-cdta)]2–·3H2O. The enthalpy difference of the seven- and six-coordinated forms favours
the q = 1 structures by very similar values (24.0 and 23.4 kJ mol-1 for the c-cdta and t-cdta
complexes, respectively), as would be expected considering the similar Mn-O5 bond
distances. However, the corresponding relative free energies differ significantly (15.1 and
16.8 kJ mol-1 for the c-cdta and t-cdta complexes, respectively), and suggest a higher free
energy barrier for the water exchange in [Mn(t-cdta)(H2O)]2–, which agrees with the
experimental evidence. Thus, the faster water exchange rate of [Mn(c-cdta)(H2O)]2– obtained
19

from 17O NMR measurements appears to be related to entropy factors rather than to important
differences in the environments around the coordinated water molecules.
DFT calculations give values of the hyperfine coupling constants AO/ꢀ of -44.9×106
and -44.6×106 rad s-1 for the [Mn(c-cdta)(H2O)]2– and [Mn(t-cdta)(H2O)]2– complexes,
17
respectively. The first value is in excellent agreement with the one obtained from O NMR
measurements in this study (AO/ꢀ = -42.7×106 rad s-1), while the second is somewhat higher
than the experimental value reported for [Mn(t-cdta)]2– using only T2 relaxation data
(-26.4×106 rad s-1) [42]. The close agreement between the experimental and calculated
hyperfine coupling constant of [Mn(c-cdta)(H2O)]2– confirms that the assumption of q = 1 is
correct.
Summary
Our efforts to find suitable ligand for Mn(II) complexation started from equilibrium,
kinetic and relaxometric characterization of Mn(II) complexes formed with commercially
available open-chain ligands a few years ago [18]. The studies performed revealed that the
rigid t-cdta forms relatively inert complex with Mn(II) ion (the half-life of the dissociation is
expected to be 12 hours near to physiological conditions). However, to best of our knowledge
Mn(II) complexes of other rigid edta derivative ligands such as cis-1,2-cyclohexanediamine
or o-phenylenediamine tetraacetates (e.g. c-cdta or phdta) were not studied in detail in this
respect. The combined experimental and DFT study presented in this work show that the
[Mn(c-cdta)(H2O)]2– and [Mn(t-cdta)(H2O)]2– complexes present very similar solution
structures in terms of the geometry around the metal coordination environment. However, the
better pre-organization of the trans derivative for the coordination of relatively large metal
ions such as Mn2+ results in rather different properties.
The thermodynamic, kinetic (solvent and dissociation kinetic data) and relaxation
parameters expressed in terms of water exchanges rates and relaxivity values (at 25 °C) are
collected in the following table. The data collected for these hexadentate chelators indeed
indicate that the Mn(II) complexes of these ligands differ notably. The most pronounced
difference is seen for the dissociation kinetic data as the Mn(II) complex formed by the cis-
derivative ligand shows properties similar to those seen for other labile complexes of linear
edta-type ligands [18] whereas the t-cdta forms relatively inert Mn(II) complex. Moreover,
the stability (especially the conditional stability) behaves alike by making the trans-1,2-
20

cyclohexanediamine unit to be an excellent “building block” when tailoring inert Mn(II)
complexes for safe MRI applications.
Table 4. Comparison of the most important physico-chemical data of the Mn(II) complexes
formed with t-cdta and c-cdta ligands.
[a]
[Mn(c-cdta)]2–
[Mn(t-cdta)]2–
Parameter
log K[Mn(L)]
pMn[b]
14.19
7.82
0.47
22.5
3.79
14.32
8.68
12
t1/2 (h)[c]
kex298 (x107 s–1)
14.0[d]
3.62
298 [e]
r1p
[a] Ref. [18]; [b] pMn values were calculated at pH=7.4 by using 0.01 mM Mn(II) and ligand
concentration as suggested by É. Tóth and co-workers [29]; [c] the half-lives (h) of
dissociation were extrapolated to pH=7.4 by using 0.01 mM Cu(II) ion concentration; [d]
form Ref. [43]; [e] at 25 oC and 20 MHz.
Experimental section
Synthesis
All commercially purchased reagents (Sigma-Aldrich) and solvents (Scharlab) were
used as received and without further purification. NMR spectra were acquired on Bruker
Avance 360 and 400 MHz spectrometers and processed/analyzed using MestreNova 8.1
(Mestrelab Research) or Topspin (Bruker) softwares. The ESI QTOF (Electro Spray
Ionization Quadrupole Time of Flight) spectrometry was performed at the Department of
Applied Chemistry of the University of Debrecen.
a). Synthesis of cis-1,2-CDTA-tetraethyl ester: Ethyl-bromoacetate (6.14 g, 36.8 mmol, 4.2
equiv.) was dissolved in dry acetonitrile and added dropwise to the mixture of 1.00 g (8.76
mmol) cis-1,2-diaminocyclohexane, 5.66 g N,N-diisopropylethylamine (43.8 mmol, 5.0
equiv.) and 1.31 g (8.76 mmol, 1.0 equiv.) of NaI prepared in dry acetonitrile (30 ml). The
reaction mixture was then heated to reflux in N2 atmosphere and kept at this temperature for 7
hours. The precipitate formed in the course of reaction was filtered off from the hot solution
and the filtrate was evaporated under reduced pressure. The resulting oily residue was
dissolved in 40 mL of bi-distilled water and extracted with portions of chloroform (3×40 mL).
The combined organic fraction was dried over anhydrous Na2SO4 and evaporated under
21

reduced pressure. The purification of the title product by flash chromatography on silica gel
by using petrolether and ethyl acetate (increasing the ethyl acetate content of the eluent
gradually form 10:1 to 5:1 by volume) yielded the pure substance (0.82 g which corresponds
to 20 % yield). Rf in 5:1 petrolether : ethyl acetate is 0.38.
1H-NMR: (360.13 MHz, CDCl3): δ = 1.23 (12H, t, J = 7.11 Hz, -O-CH2-CH3); 1,29 (2H, m, -
CH2-); 1.43 (2H, m, -CH2-); 1.62 (2H, m, -CH2-); 1.79 (2H, m, -CH2-); 3.16 (2H, d, J = 6.88
Hz, -CH-); 3.70 and 3.79 (8H, AB doublet, 2JAB = 17.90 Hz, -N-CH2-COO-); 4.11 (8H, q, J =
7.11 Hz, -O-CH2-CH3). 13C-NMR: (90.55 MHz, CDCl3): δ = 14.34 (4C, s, -O-CH2-CH3);
23.44 (2C, s, -CH2-CH2-); 26.88 (2C, s, -CH2-); 53.08 (4C, s, -N-CH2-COO-); 60.27 (4C, s, -
O-CH2-CH3); 60.55 (2C, s, -CH-); 172.58 (-N-CH2-COO-). MS (ESI-Q-TOF): m/z [M+]
calculated for C22H38N2ONa: 481.252; found: 481.251
b). Synthesis of cis-1,2-CDTA: 0.40 g NaOH (10 mmol, 6 equiv.) was dissolved in 1,0 mL
of bi-distilled water and was added portionwise to the solution of 0.75 g (1.64 mmol) cis-1,2-
CDTA-tetraethyl ester dissolved in 15 mL of ethanol. The reaction mixture was heated to 78
◦C and refluxed at this temperature for 24 h. The white precipitate formed during the reaction
(Na4(c-cdta)) in the reaction mixture was filtered off and washed twice with cold ethanol
(2×10 mL), diethyl ether (2×10 mL) and dried to constant weight. The solid obtained was
dissolved in minimal amount of water required to dissolve it (3.0 mL) and pH of the sample
was set to pH=2.02 by addition of concentrated HCl. The white precipitate formed as a result
of acid addition was filtered off and washed with small portions of acidified distilled water
(pH=2.50), cold ethanol (2×5 mL), diethyl ether (2×5 mL) and dried until the weight of the
product become constant. Yield: 0.22 g (39 %).
1H-NMR: (360.13 MHz, D2O): δ = 1,51 (2H, m, -CH2-); 1.74 (2H, m, -CH2-); 1.83 (2H, m, -
2
CH2-); 2.00 (2H, m, -CH2-); 3.51 (2H, m, -CH-); 3.83 and 4.01 (8H, AB doublet, JAB
=
17.39 Hz, -N-CH2-COO-). 13C-NMR: (90.55 MHz, D2O): δ = 22.11 (2C, s, -CH2-CH2-);
23.19 (2C, s, -CH2-); 54.76 (4C, s, -N-CH2-COO-); 61.26 (2C, s, -CH-); 172.81 (-N-CH2-
COO-). MS (ESI-Q-TOF): m/z [M+] calculated for C14H21N2O8NaNa: 391.109; found:
391.109
Equilibrium studies
The chemicals (MCl2 salts) used in the studies were of the highest analytical grade.
The concentration of the stock solutions was determined by complexometric titration using a
22

standardized Na2H2edta solution and appropriate indicators (Patton & Reeder (CaCl2),
Eriochrome Black T (MgCl2 and MnCl2), xylenol orange (ZnCl2), murexid (CuCl2)) [49].
The pH-potentiometric titrations were carried out with a Methrohm 888 Titrando titration
workstation, using a Metrohm 6.0233.100 combined electrode. The titrated solutions (6.00
mL) were thermostated at 25 °C. The samples were stirred and kept under inert gas
atmosphere (N2) to avoid the effect of CO2. The calibration of the electrode was performed by
two point calibration (KH-phthalate (pH = 4.005) and borax (pH = 9.177) buffers). The
calculation of [H+] from the measured pH values was performed with the use of the method
proposed by Irving et al. [50] by titrating a 0.01 M HCl solution (I = 0.15 M NaCl) with a
standardized NaOH solution. The differences between the measured (pHread) and calculated
pH (-log [H+]) values were used to obtain the equilibrium H+ concentrations from the pH-data
obtained in the titrations. The ion product of water was determined from the same experiment
in the pH range 11.4 – 12.0.
The concentration of the c-cdta ligand was determined by pH-potentiometric titration,
comparing the titration curves obtained in the presence and absence of moderate Mn(II)
excess. The protonation constants of c-cdta as well as the stability and protonation constants
of the complexes formed with Mg(II) Ca(II), Mn(II), Cu(II) and Zn(II) were also determined
by pH-potentiometric titration. The metal-to-ligand concentration ratio was 1:1 (the
concentration of the ligand was generally 2.50 − 3.00 mM). The pH-potentiometric titrations
curves were measured in the pH range 1.8 − 11.8 and 44 − 205 mL NaOH–pH data pairs were
collected and fitted.
The stability constant and solution speciation of the [Mn(c-cdta)]2− complex was
1
confirmed using H relaxometry in order to support the stability constant value obtained by
pH-potentiometry (for details about T1 measurements, see the “Relaxivity determination”
section below). 3.00 mL of a 1.02 mM [Mn(c-cdta)]2− sample were titrated with solid NaOH
or gasous HCl (in order to avoid sample dilution) to adjust the pH in the range 1.97 − 7.12,
followed by recording and averaging of 5 − 6 T1 values for each pH data point.
Owing to the high conditional stability of [Cu(c-cdta)]2-, the formation of the complex
was complete (nearly 100%) even at pH = 1.54 (starting point of the pH-potentiometric
titration). For this reason, 12 out-of-cell (batch) samples containing a slight excess of ligand
o
and the Cu(II) ion were prepared (cL = 3.528 mM, cCu2+ = 3.410 mM, 25 C, 0.5 M
(Na++H+)Cl-). The samples, whose acidity was varied in the concentration range of 12.1 − 462
o
mM, were equilibrated for 1 day before recording the absorption spectra at 25 C in Peltier
23

thermostated semi-micro 1 cm Hellma® cells using a Varian CARY 1E, UV-Vis
spectrophotometer. The molar absorptivity of the [Cu(c-cdta)]2-, complex was determined at
21 wavelengths (600 − 800 nm range) by recording the spectra of 1.72×10-3, 2.55×10-3 and
3.48×10-3 M complex solutions, while for the Cu(II) ion, previously published molar
absorptivity values (determined under identical conditions) [51] were used for the data fitting.
The molar absorption coefficients of monoprotonated [CuH(c-cdta)]2- were calculated during
the simultaneous data refinement (UV-vis and pH-potentiometric titration curves). The
protonation (ligand and complexes) and stability constants (complexes) were calculated from
the titration data with the PSEQUAD program [52].
Kinetic studies
o
The dissociation rates of the [Mn(c-cdta)]2− chelate were studied at 25 C (Peltier
thermostated) and 0.15 M NaCl ionic strength by stopped-flow method monitoring the
formation of the Cu(II) complex at 300 nm using an Applied Photophysics DX-17MV
instrument. All dissociation reactions were performed under pseudo-first order conditions
where the exchanging metal ion (Cu(II)) was in 10–40-fold excess relative to the complex
(ccomplex = 2.0×10–4 M, pH range 4.25–5.10). The kinetic studies were carried out in a non-
coordinating buffer to maintain the pH in the samples constant (0.05 M N,-methylpiperazine
H
(nmp) with log K2 = 4.34 under the conditions applied). The pseudo-first-order rate constants
(kobs) were calculated by fitting the absorbance-time data series to equation 9:
−k
t
obs
(eqn. 9)
At = (A0 − Ae )e
+ Ae
where At, A0 and Ae are the absorbance at time t, at the start and at the equilibrium of the
reaction, respectively. The calculations were performed with the computer program
Micromath Scientist, version 2.0 (Salt Lake City, UT, USA) by using a standard least-squares
procedure [53].
1H-, and 17O-NMR relaxometry
Longitudinal (1/T1) and transverse (1/T2) relaxation rates and chemical shifts of an
aqueous solution of the Mn(II) complex (pH=7.4, 10.3 mM) and of a diamagnetic reference
(HClO4 acidified water, pH = 3.3) were measured in the temperature range 273 – 338 K using
a Bruker Avance 400 (9.4 T, 54.2 MHz) spectrometer. The temperature was determined
according to previous calibration by means of ethylene glycol and methanol as standards [54].
24

1/T1 and 1/T2 values were determined by the inversion - recovery and the Carr − Purcell −
Meiboom −Gill spin − echo technique, respectively [55]. The technique of the 17O NMR
measurements has been described previously [56]. To avoid susceptibility corrections of the
chemical shifts, a glass sphere fitted into an 10 mm NMR tube was used to contain the
17
samples. To improve sensitivity, 17O enriched water (10% H2 O, CortecNet) was added to
the solutions to reach around 1% enrichment.
Proton NMRD profiles of the [Mn(c-cdta)]2– complex (1.00 mM, pH = 7.4) were
recorded in aqueous solution on a Stelar SMARTracer Fast Field Cycling relaxometer (0.01 −
10 MHz) and a Bruker WP80 NMR electromagnet adapted to variable field measurements (20
− 80 MHz) and controlled by a SMARTracer PC-NMR console. The temperature was
monitored by a VTC91 temperature control unit and maintained by a gas flow. The
temperature was determined by previous calibration with a Pt resistance temperature probe.
The least-squares fit of the 17O NMR and of the NMRD data was performed using
Visualiseur/Optimiseur [57, 58] running on a MATLAB 8.3.0 (R2014a) platform.
Relaxivity determination
The longitudinal water proton relaxation rate (r1 = 1/T1 - 1/Tw) was measured at 20
MHz with a Bruker Minispec MQ-20 relaxometer (Bruker Biospin, Germany). Samples were
o
thermostated by using a circulating water bath at 25.0 0.2 C. The longitudinal relaxation
times (T1) were measured by using the inversion recovery method (180o − τ − 90o) by
averaging 5 − 6 data points for each concentration point obtained by using 14 different τ
values (τ values ranging between 0 to at least 6 times the expected T1).
DFT calculations
The [Mn(c-cdta)(H2O)]2–·2H2O and [Mn(t-cdta)(H2O)]2–·2H2O systems were fully
optimized by using the hybrid meta-GGA M062X functional and the TZVP basis set [59, 60].
The nature of optimized geometries as true energy minima was confirmed by frequency
analysis. Frequency calculations provided zero-point energies (ZPEs), enthalpies (H) and free
energies (G) at 298.15 K and 1 atm. The relative free energies include non-potential-energy
contributions (zero point energies and thermal terms) obtained through frequency analysis.
Hyperfine coupling constants were calculated on the optimized structures by using the
EPR-III basis set for the ligand atoms and the aug-cc-pVTZ-J basis set for Mn [61, 62].
25

Throughout this study bulk solvent effects were considered by using the integral equation
formalism variant of the polarizable continuum model (IEFPCM) [63].
Abbreviations
MRI
DFT
APC
CA
Magnetic resonance imaging
Density Functional Theory
aminopolycarboxylate
Contrast Agent
dtpa
dota
NSF
edta
Diethylenetriaminepentaacetic acid
1,4,7,10-Tetraazacyclododecane-1,4,7,10-tetraacetic acid
Nephrogenic Systemic Fibrosis
Ethylenediaminetetraacetic acid
trans-1,2-cdta trans-1,2-Diaminocyclohexane-N,N,N′,N′-tetraacetic acid
cis-1,2-cdta
dipea
cis-1,2-Diaminocyclohexane-N,N,N′,N′-tetraacetic acid
N,N-diisopropylethylamine
ida
Iminodiacetate
NMRD
4-het-cdta
Nuclear magnetic relaxation dispersion
4-((1-(2-Hydroxyethyl)-1H[1,2,3]triazol-4-yl)methoxy)methyl-trans-1,2-
diaminocyclohexane-N,N,N’,N’-tetraacetic acid
1,4-do2a
do1a
1,4,7,10-Tetraazacyclododecane-1,4-diacetic acid
1,4,7,10-Tetraazacyclododecane-1-acetic acid
o-Phenylenediamine tetraacetate
phdta
ESI QTOF
nmp
Electro Spray Ionization Quadrupole Time of Flight
N-methylpiperazine
26

ZPEs
Zero-point energies
Appendix A. Supplementary data
Supplementary data associated with this article (1H-, 13C-NMR and MS data of the pure
compounds, comparison of X-ray and DFT structures and the equations related to the solvent
exchange kinetics data refinement) can be found, in the online version, at
http://dx.doi.org/10.1016/xxxxxxxxxxxxxxxxx .
Acknowledgements
This research was funded by the Hungarian Scientific Research Fund (OTKA K-120224
project), Le Studium, Loire Valley Institute for Advanced Studies (Gy.T. and F.K.K.) and the
János Bolyai Research Scholarship (Gy.T. and F.K.K.) of the Hungarian Academy of
Sciences. E. T. acknowledges support of the Ligue contre le Cancer (France). The research
was also supported by the EU and co-financed by the European Regional Development Fund
under the project GINOP-2.3.2-15-2016-00008. C. P.-I. thanks Ministerio de Economía y
Competitivad (CTQ2016-76756-P) for generous financial support.
27

References
[1] Anderson L. J., Welch M. J., Chem. Rev. 1999, 99(9), 2219-34.
[2] Stimmel J.B., Kull F.C. Jr., Nucl. Med. Biol. 1998, 25, 117-125.
[3] Nayak T. K., Brechbiel M. W., Bioconjug. Chem. 2009, 20(5), 825-841.
[4] Parker D., Dickins R. S., Puschmann H., Crossland C. and Howard J. A. K., Chem. Rev.
2002, 102 (6), 1977-2010.
[5] Montgomery C. P., Murray B. S., New E. J., Pal R., Parker D., Acc. Chem. Res., 2009, 42
(7), 925-937.
[6] Caravan P., Ellison J. J., McMurry T. J., Lauffer R. B., Chem. Rev. 1999, 99(9), 2293-
2352.
[7] The Chemistry of Contrast Agents in Medical Magnetic Resonance Imaging, 2nd edition,
Eds. Merbach A. E, Helm L. Tóth É., John Wiley and Sons, London, 2013.
[8] Grobner T., Nephrol. Dial. Transplant. 2006, 21, 1104-1108.
[9] Marckmann P., Skov L., Rossen K., Dupont A., Damholt M. B., Heaf J. G., Thomsen H.
S., J. Am. Soc. Nephrol. 2006, 17, 2359-2362.
[10] Kitajima K., Maeda T., Watanabe S., Ueno Y., Sugimura K., Int. J. Urol. 2012, 19,
806-811.
[11] Kanda T., Fukusato T., Matsuda M., Toyoda K., Oba H., Kotoku J., Haruyama T.,
Kitajima K., Furui S., Radiology 2015, 276, 228-232.
[12] Kanal E., Tweedle, M. F., Radiology 2015, 275, 630-634.
[13] Roberts D. R., Lindhorst S. M., Welsh C. T., Maravilla K. R., Herring M. N., Braun K.
A., Thiers B. H., Davis W. C., Invest Radiol. 2016, 51(5), 280-289.
[14] Birka M., Wentker K. S., Lusmöller E., Arheilger B., Wehe C. A., Sperling M.,
Stadler R., Karst U., Anal. Chem. 2015, 87(6), 3321-8.
[15] Troughton J. S., Greenfield M. T., Greenwood J. M., Dumas S., Wiethoff A. J., Wang
J., Spiller M., McMurry T. J., Caravan P. Inorg. Chem., 2004, 43(20), 6313-23
[16] Wang S., Westmoreland D. T., Inorg. Chem. 2009, 48(2), 719-27.
[17] Drahoš B., Kubíček V., Bonnet C. S., Hermann P., Lukeš I, Tóth É., Dalton Trans.
2011, 40(9), 1945-51.
[18] Kálmán F. K., Tircsó Gy., Inorg. Chem. 2012, 51(19), 10065-17.
[19] Drahos B., Kotek J., Hermann P., Lukes I., Tóth É., Inorg. Chem. 2010, 49, 3224-38.
28

[20] Tei L., Gugliotta G., Fekete M., Kálmán F. K., Botta M., Dalton Trans. 2011, 40(9),
2025-32
[21] Regueiro-Figueroa M., Rolla G. A., Esteban-Gómez D., de Blas A., Rodríguez-Blas
T., Botta M., Platas-Iglesias C., Chem. Eur. J. 2014, 52(20), 17300–17305.
[22] Gale E. M., Atanasova I. P., Blasi F., Ay I., Caravan P., J. Am. Chem. Soc. 2015, 137,
15548-15557.
[23] Forgács A., Pujales-Paradela R., Regueiro-Figueroa M., Valencia L., Esteban-Gómez
D., Botta M., Platas-Iglesias C., Dalton Trans. 2017, 46, 1546-1558.
[24] Phukan B., Patel A. B., Mukherjee C., Dalton Trans. 2015, 44, 12990-12994.
[25] Ussov W.Y., Beljanin M.L. Borodin O.Y., Churin A.A., Bezlepkin A.I., Filimonov
V.D., Medical Visualization (in russian) 2009, 5 121-132.
[26] Ussov W.Y., Bogunetsky A.A., Babokin V.E., Belyanin M.L., Goltsov S.G.,
Filimonov V.D., Tomsk/RU In ECR, 2014 March 6–10, 2014, Vienna, Austria; B-0904.
[27] Critical Stability Constants, Ed. Martell A. E., Smith R. M., Vol. 1-6. Plenum Press,
New York, London, 1974-1989.
[28] Charlier L., Merciny E., Fuger, J., Anal. Chim. Acta, 1985, 178, 299-306.
[29] Drahoš B., Kotek J., Hermann P., Lukeš I., Tóth É., Inorg. Chem. 2010, 49(7), 3224-
3238
[30] Sudmeier J. L., Reilley C. N., Anal. Chem. 1964, 36(9), 1707-712.
[31] Regueiro-Figueroa M., Ruscsák E., Fra L., Tircsó Gy., Tóth I., de Blas A., Rodríguez-
Blas T., Platas-Iglesias C. and Esteban-Gómez D., Eur. J. Inorg. Chem. 2014, 36, 6165-
6173.
[32] Rodríguez-Rodríguez A., Garda Z., Ruscsák E., Esteban-Gómez D., de Blas A.,
Rodríguez-Blas T., Lima L. M. P., Beyler M., Tripier R., Tircsó Gy.. Platas-Iglesias C.,
Dalton Trans., 2015,44, 5017-5031.
[33] Thakral C., Abraham J.L., J. Cutan Pathol. 2009, 36(12),1244-54.
[34] Port M., Idée J.M., Medina C., Robic C., Sabatou M., Corot C., Biometals. 2008,
21(4), 469-90.
[35] Smith G. F., Margerum D. W, Inorg. Chem. 1969, 8(1), 135-138.
[36] Margerum D. W., JPausch. B., Nyssen G. A., Smith G. F., Anal. Chem. 1969, 41(2),
233-238.
[37] Gale E. M., Zhu J., Caravan P. J. Am. Chem. Soc. 2013, 135(49), 18600-18608.
[38] Bloembergen N.; Morgan L. O., J. Chem. Phys. 1961, 34, 842-850.
[39] Freed J. H., J.Chem.Phys. 1978, 68, 4034-4037.
29