Fe(III) salEen derived Schiff base complexes as potential contrast agents

Article abstract of DOI:10.1016/j.ica.2015.04.026

Three iron(III) complexes with ligands derived from N-ethyl-N-(2-aminoethyl)salicylaldiminate (H, 1; 5-Br, 2; 3-OMe, 3 substituents at the phenyl group) were prepared and the X-ray crystal structures of 1 and 2 are reported. NMR studies of solutions of th

Full text of DOI:10.1016/j.ica.2015.04.026

Accepted Manuscript
Fe(III) SalEen derived Schiff base complexes as potential contrast agents
Bernardo de P. Cardoso, Ana I. Vicente, Joseph B.J. Ward, Pedro J. Sebastião,
Fabián Vaca Chávez, Sónia Barroso, Alexandra Carvalho, Stephen J. Keely,
Paulo N. Martinho, Maria José Calhorda
PII:
S0020-1693(15)00214-5
http://dx.doi.org/10.1016/j.ica.2015.04.026
ICA 16519
DOI:
Reference:
Inorganica Chimica Acta
To appear in:
Received Date:
Revised Date:
Accepted Date:
13 February 2015
13 April 2015
20 April 2015
Please cite this article as: B. de P. Cardoso, A.I. Vicente, J.B.J. Ward, P.J. Sebastião, F.V. Chávez, S. Barroso, A.
Carvalho, S.J. Keely, P.N. Martinho, M.J. Calhorda, Fe(III) SalEen derived Schiff base complexes as potential
contrast agents, Inorganica Chimica Acta (2015), doi: http://dx.doi.org/10.1016/j.ica.2015.04.026
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.

Fe(III) SalEen derived Schiff base complexes
as potential contrast agents
Bernardo de P. Cardoso a, Ana I. Vicente a, Joseph B. J. Ward b,*, Pedro J. Sebastião
c,d,*, Fabián Vaca Chávez c, Sónia Barroso e, Alexandra Carvalho f, Stephen J. Keely b,
Paulo N. Martinho a,*, Maria José Calhorda a
aCentro de Química e Bioquímica, DQB, Faculdade de Ciências, Universidade de
Lisboa, Campo Grande, 1749-016 Lisboa, Portugal
Homepage: http://intheochem.fc.ul.pt/index.html
bMolecular Medicine Laboratories, Royal College of Surgeons in Ireland Education and
Research Centre, Beaumont Hospital, Beaumont, Dublin 9, Ireland
cCondensed Matter Physics Centre, Universidade de Lisboa, Lisbon, Portugal
dDepartamento de Física, Instituto Superior Técnico, Universidade de Lisboa, Lisbon,
Portugal
eCentro de Química Estrutural, Instituto Superior Técnico, Universidade de Lisboa, Av.
Rovisco Pais, 1049-001 Lisboa, Portugal
fCENIMAT-I3N/DCM/FCT/UNL, Campus da Caparica, Caparica, Portugal
*
Corresponding authors
Email addresses: seosamh.ward@gmail.com (J. B. J. Ward),
pedro.jose.sebastiao@tecnico.ulisboa.pt (P. J. Sebastião), pnmartinho@fc.ul.pt (P. N.
Martinho)
Keywords: Fe(III) complexes / Evans’ method / Contrast agent / NMRD / cell toxicity
Abstract Three iron(III) complexes with ligands derived from N-ethyl-N-(2-
aminoethyl)salicylaldiminate (H, 1; 5-Br, 2; 3-OMe, 3 substituents at the phenyl group)
were prepared and the X-ray crystal structures of 1 and 2 are reported. NMR studies of
solutions of these complexes in DMSO allowed for investigation of their magnetic
behaviour and paramagnetic relaxation contribution. The relaxivities measured ranged
from 0.35 to 0.80 mM-1 s-1 for proton Larmor frequencies from 0.01 to 300 MHz, in
agreement with those known for other iron(III) based contrast agents. Biological studies
on colonic epithelial T84 cell monolayers showed that the compounds exert toxic effects
only at concentrations higher than 100 µM while coincidently reducing colonic
epithelial secretory function. These two features make these complexes good candidates
for further development in order to be used as MRI contrast agents.
1. Introduction
Magnetic Resonance Imaging (MRI) is a widely used technique for visualising
anatomical soft tissue, using non-ionizing radiation and displaying excellent spatial
resolution [1]. However, it has a limited sensitivity when compared with other imaging
techniques (e.g. optical or nuclear) [2,3], due to the abundance of the water signal
monitored [4]. Thus, efforts to improve this shortcoming focus mainly on chemical
substances which enhance MRI contrast [5].
The field of MRI contrast agents has been dominated by trivalent gadolinium (Gd3+)
complexes [5–8] as they are paramagnetic with seven unpaired electrons [5]. However,
1

concerns regarding Gd3+ toxicity [9–11] are directing research towards the use of other
metals as paramagnetic contrast agents. This resulted in an expansion of non-lanthanide
based contrast agents (e. g.: Mn(II) [12–16], Fe(II)/Fe(III) [17–24], Co(II)/Co(III)
[25,26] and Ni(II) [27]). Among these, iron is a biocompatible metal and
paramagnetism is observed for high-spin (HS) Fe(II) and both high-spin and low-spin
(LS) Fe(III) octahedral species [28,29]. Recently, the approaches to iron based contrast
agents focused on either design of molecular compounds [19,23] or development of
nanoparticles [14,30,31].
Schiff bases were shown to be a versatile class of ligands for coordination chemistry
[32]. These ligands can be tailored to contain groups bearing labile protons (e. g. OH or
NH) thus increasing the solvent NMR relaxivity (r) produced by the contrast agent. This
enhancement is magnetic field dependent and the parameters that describe the
paramagnetic relaxation can be obtained, for instance, through a magnetic field
dependence of the spin-lattice relaxivity (r1) study usually called Nuclear Magnetic
Relaxation Dispersion (NMRD) [16]. This effect is observed by measuring solvent spin-
lattice relaxation rates over a broad range of magnetic fields corresponding to proton
Larmor frequencies (νL,H) from 10 kHz to hundreds of MHz, in general.
We have been investigating this class of ligands aiming at finding new examples of
paramagnetic systems using Fe(III) [33]. Given that these complexes contain iron and
have protons that can easily exchange with the solvent we decided to investigate their
potential as contrast agents.
Administration of contrast agents is commonly done either orally or intravenously.
Ingested contrast agents will, inevitably, come into contact with the epithelial lining of
the gastrointestinal (GI) tract. The GI epithelial barrier is a monolayer of cells that
serves several key functions, such as the transporting of fluid and electrolytes to and
from the gut lumen. The absorption and secretion of fluids across the epithelial barrier is
driven by osmotic gradients, which in turn are generated by the active transport of ions.
Cl- secretion, the primary driving force for fluid secretion, is promoted by hormones and
neuroimmune agonists, which bind to cell surface receptors and increase cytosolic
second messengers, most notably cyclic AMP (cAMP) and Ca2+. These second
messengers then interact with the transport proteins of the Cl- secretory pathway to
stimulate fluid secretion into the gut lumen [34]. Employing the newly synthesised
potential contrast agents, we show their effects on colonic epithelial cell survival and
function.
Here we report the synthesis of compounds of the general formula [Fe(R-salEen)2]Cl
(salEen = N-ethyl-N-(2-aminoethyl)salicylaldiminate; R = H, 1; 5-Br, 2; 3-OMe, 3) and
present the X-ray crystal structures for both [Fe(H-salEen)2]Cl⋅0.5H2O (1) and [Fe(Br-
salEen)2]Cl⋅0.5H2O (2). The magnetic behaviour in solution assessed by the Evans’
method [35], NMRD profiles in DMSO solutions, cell toxicities and their effects on Cl-
secretion.
2. Experimental
2.1. Syntheses
N-ethylethylenediamine, salicylaldehyde, 5-bromosalicylaldehyde, o-vanillin,
iron(III) chloride hexahydrate, iron(II) chloride and all the solvents were purchased and
used without further purification. IR transmittance spectra were obtained on a Nicolet
Nexus 6700 FTIR spectrophotometer in the 400-4000 cm-1 range with 4 cm-1 resolution
using KBr pellets.
2

2.1.1. [Fe(salEen)2]Cl·xH2O (1·xH2O): Salicylaldehyde (214 µL, 2 mmol) was
added to a solution of N-ethylethylenediamine (210 µL, 2 mmol) in methanol (30 mL)
and left stirring for 15 minutes. Iron(III) chloride hexahydrate (270 mg, 1 mmol) was
added to the previous mixture and left stirring for 1 h. The mixture was filtered and
needle shaped dark brown crystals of 1·0.5H2O were obtained by slow evaporation in
methanol (81 mg, 14 %). IR (KBr): νmax/cm-1 3095 (νNH, m), 2924-2853 (νCH, m), 1629
(νC=N, s), 1598 (δC=C, m), 1296 (νC-N, s). Anal. found (calcd) for C22H34ClFeN4O4·2H2O:
C 51.99 (51.83); H 6.32 (6.72); N 10.95 (10.99).
2.1.2. [Fe(5-Br-salEen)2]Cl·xH2O (2·xH2O): The same procedure as for the synthesis
of compound 1 was used, starting from N-ethylethylenediamine (210 µL, 2 mmol) in
methanol (30 mL), 5-bromosalicylaldehyde (402 mg, 2 mmol) and iron(II) chloride
(127 mg, 1 mmol). The mixture was filtered and needle shaped dark brown crystals of
2·0.5H2O were obtained by slow evaporation in methanol (85 mg, 13 %). IR (KBr):
νmax/cm-1 3078 (νNH, m), 2970-2858 (νCH, m), 1633 (νC=N, s), 1590 (δC=C, m), 1295 (νC-
N, s). Anal. found (calcd) for C24H32Br2ClFeN4O4·2H2O: C 39.09 (39.52); H 4.33
(4.57); N 8.22 (8.38).
2.1.3. [Fe(3-OMe-salEen)2]Cl (3): The same procedure used to synthesise compound
1 was used for compound 3 synthesis, starting from N-ethylethylenediamine (210 µL, 2
mmol) in methanol (30 mL), o-vanillin (304 mg, 2 mmol) and iron(II) chloride (127 mg,
1 mmol). A dark green microcrystalline solid was filtered, washed with methanol and
dried under vacuum (116 mg, 20 %). IR (KBr): νmax/cm-1 3063 (νNH, m), 2994-2831
(νCH, w), 1633 (νC=N, s), 1598 (δC=C, m), 1308 (νC-N, s). Anal. found (calcd) for
C24H34ClFeN4O4: C 53.48 (53.89); H 6.50 (6.60); N 10.39 (10.48).
2.2. General procedures for X-ray crystallography
Crystals suitable for single-crystal X-ray analysis were obtained for 1·0.5H2O and
2·0.5H2O, as described in the synthetic procedures. The data were collected using
graphite monochromated Mo-Kα radiation (λ = 0.71073 Å) on a Bruker AXS-KAPPA
APEX II diffractometer equipped with an Oxford Cryosystem open-flow nitrogen
cryostat. Cell parameters were retrieved using Bruker SMART software and refined
using Bruker SAINT on all observed reflections. Absorption corrections were applied
using SADABS [36]. The structures were solved and refined using direct methods with
program SIR2004 [37] using WINGX-Version 1.80.01[38] SHELXL [39,40] system of
programs. All non-hydrogen atoms were refined anisotropically and the hydrogen atoms
were inserted in idealised positions and allowed to refine riding on the parent carbon
atom. The molecular diagrams were drawn with ORTEP-3 for Windows [41] included
in the software package.
Both structures refined to a good convergence, even though crystals of 2·0.5H2O
were of poor quality presenting low ratio of observed/unique reflections. SIMU and
ISOR restraints were applied in order to prevent some atoms from turning non-positive-
definite.
It was not possible to locate the hydrogen atoms of the H2O molecules in both
structures of 1·0.5H2O and 2·0.5H2O.
Flack parameter x obtained for 2·0.5H2O is meaningless (0.44(2)). The crystal is
most likely a 50/50 inversion twin. No absolute configuration determination is possible.
For crystallographic experimental data and structure refinement parameters see Table 1.
Data for structures 1 and 2 were deposited in CCDC under the deposit numbers CCDC
3

1018927-1018928 and can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.uk/data_request/cif.
Table 1
Selected crystallographic experimental data and structure refinement parameters for
1·0.5H2O and 2·0.5H2O.
1·0.5H2O
2·0.5H2O
Empirical
formula
C44H60Cl2Fe2N8O5 C44H56Br4Cl2Fe2N8O5
Formula weight
Temperature
(K)
963.60
150(2)
1279.21
150(2)
Crystal system
Orthorhombic
Fdd2
Monoclinic
P21
Space group
a(Å)
18.230(4)
53.905(11)
9.8130(18)
90
10.282(2)
25.841(6)
10.606(2)
90
b(Å)
c(Å)
α(°)
90
116.136(10)
90
β(°)
90
γ(°)
V(Å3)
9643(3)
8, 1.327
0.763
2529.8(9)
2, 1.679
3.889
Z, ρcalc (g cm-3)
µ(mm-1)
Crystal size
Crystal colour
Crystal shape
0.12×0.12×0.04
brown
0.20×0.04×0.04
brown
plate
prism
Refl. collected
Unique refl.
[R(int)]
26883
21449
4187 [0.0984]
7157 [0.0989]
0.0591
0.1207
1.046
0.0845
0.2011
1.043
R1 [I>2σ(I)]
wR2 [I>2σ(I)]
GooF
2.3. NMR Studies
2.3.1. Evans’ method: Magnetic measurements of complexes 1 and 2 in solution
were performed at room temperature (rt) by 1H NMR using the Evans’ method [35] on a
Bruker Avance 400 spectrometer operating at 400.14 MHz at a constant temperature of
298.15 K. The measurements for each compound at different concentrations ([c] = 100,
50, 10 and 5 mM) were performed in standard 5 mm NMR tubes containing the
paramagnetic samples dissolved in DMSO-d6 with an inert reference of 0.03 % TMS,
against a reference insert tube filled with the same solvent (0.03 % TMS in DMSO-d6)
and their shift measured in Hz.
2.3.2. NMRD measurements: The proton spin−lattice relaxation time, T1, was
4

measured over νL,H, range from 10 kHz to 300 MHz. In the range 10 kHz − 8.9 MHz,
the data were obtained with a home-developed fast field-cycling (FFC) relaxometer
operating with a polarization and detection fields of 0.215 T and a switching time less
than 3 ms [42]. For frequencies between 10 and 91 MHz, a variable-field iron-core
magnet and a 300 MHz Bruker Avance II spectrometer were used and T1 was measured
applying the inversion recovery sequence.
2.3.3. Self-diffusion measurements: The diffusion coefficient, D, expressed in m2 s-1,
was measured using the stimulated spin-echo sequence [43] on a Bruker Avance II 300
MHz spectrometer equipped with a Diff 30 gradient unit. D was obtained from the echo
decay using the stimulated echo sequence according to
ꢈ
ꢈ ꢈ
ꢆ
ꢌ
ꢍꢎ
ꢀ = ꢀꢁꢂꢃꢄꢅ −ꢇ ꢉ ꢊ ꢋ Δ − ꢊ/3
Eq. (1)
where γ is the proton’s gyromagnetic ratio, g is the magnetic field gradient strength, δ =
1 ms is the gradient pulse length, ∆ = 20 ms is the delay between the gradients pulses.
2.4. Biological essays
2.4.1. Sample preparation: Stock solutions (1 M) of compounds 1-3 were prepared
by dissolving these in non-anhydrous DMSO. Solutions of different concentrations (1
nM, 100 nM, 10 µM, 100 µM and 500 µM) were further prepared by dissolving aliquots
of the stock solution in Ringer’s solution containing (in mM): 40 Na+, 5.2 K+, 1.2 Ca2+,
−
−
1.2 Mg2+, 119.8 Cl−, 25 HCO3 , 0.4 H2PO4 , 2.4 HPO42−, and 10 glucose.
2.4.2. Electrophysiological Measurements: T84 cell monolayers, cultured as
previously described [44], were mounted in Ussing chambers (aperture = 0.6 cm2),
voltage-clamped to zero potential difference and bathed in Ringer’s solution, and
monitored for changes in short circuit current (∆Isc) using a VCC MC8 Voltage Clamp
(Physiological Instruments, San Diego, CA). Under such conditions secretagogue-
induced ∆Isc reflect changes in electrogenic Cl- secretion [45]. Results were normalised
and expressed as ∆Isc (µA cm-2).
2.4.3. Transepithelial electrical resistance (Rte) measurements: Rte, Ω cm2 used as a
measure of intestinal epithelial barrier integrity was measured directly using an
epithelial volt meter (EVOM2).
2.4.4. Lactate dehydrogenase (LDH) Assay: Lactate dehydrogenase (LDH) release,
used as a measure of toxicity, from T84 intestinal epithelial cells, was measured using a
commercially available kit (Sigma-Aldrich, Gillingham, UK). T84 cell monolayers were
cultured on permeable supports for 7 – 14 days until the Rte had stabilised. Cells were
then treated bilaterally with compounds 1-3 for 24 hours. Following this, equal aliquots
of apical and basolateral culture media were analysed for the presence of LDH by
spectrophotometry.
2.4.5. Statistical Analysis: All biological data are expressed as mean standard error
of the mean (sem) for a series of n experiments. Student’s t-test was used to compare
paired data. One-way analysis of variance with the Student–Newman–Keuls (SNK)
multiple comparisons post-test was used when three or more groups of data were
compared. p values ≤ 0.05 were considered to be statistically significant.
5

2.5. Stability tests in aqueous solutions
The stability of aqueous solutions of complexes 1-3 was followed by UV-vis and NMR.
UV-vis spectra of a 1×10-4 mol L-1 aqueous solution of [Fe(H2O)6]3+, the three
compounds in DMSO (1×10-4 mol L-1) and the three complexes in H2O (obtained by
dilution of a 1×10-3 mol L-1 stock DMSO solution with H2O to 1×10-4 mol L-1) were
recorded on a Shimadzu 50/60 Hz spectrometer. The aqueous solutions of the three
complexes were recorded at t = 0 h, t = 24 h, t = 48 h and t = 196 h. NMR spectra of the
three complexes were also recorded in DMSO-d6 and in a mixture of DMSO-d6/D2O
(70/30). NMR studies were carried out for all three compounds at times t = 0 h, t = 24 h
and t = 196 h. Stability studies show that the most stable compound to hydrolysis is by
far complex 2 and the least stable complex 1. These studies can be found in the
supplementary information.
3. Results and Discussion
3.1. Syntheses
The synthesis of complexes 1-3 was made as described in Scheme 1. Complex 1 was
first reported by Clément et al. [46] and 1·xH2O is now reported. Complex 2 was
synthesised with crystallisation water molecules (2·xH2O) and complex 3 was obtained
in its unsolvated form. The complexes were characterised by elemental analysis, IR and
complexes 1·0.5H2O and 2·0.5H2O were also structurally characterised by single
crystal X-ray diffraction. The IR spectra of all three complexes presented the
characteristic C=N stretching band at 1633 cm-1, which confirms the successful
formation of the imine. Furthermore, the IR spectra of both 1·xH2O and 2·xH2O
indicate the presence of water molecules (Fig. S2 and Fig. S3)
Scheme 1. General method used in the synthesis of complexes 1-3.
3.2. X-ray studies
Complex 1 crystallises in the orthorhombic system, space group Fdd2, with one
[Fe(salEen)2]+ cation, one Cl− anion and half a water molecule in the asymmetric unit.
6

An ORTEP view of 1·0.5H2O is depicted in Fig. 1 and selected bond distances and
angles (data obtained at 150 K) are listed in Table 2. The iron(III) is octahedrally
coordinated by two oxygen, two nitrogen(amine) and two nitrogen(imine) atoms
belonging to two salEen ligands bound in the meridional coordination mode. The Fe-O
bond lengths (1.885(4) and 1874(4) Å) are shorter than the Fe-N(imine) (1.917(5) and
1.927(5) Å) distances, which are shorter than the Fe-N(amine) ones (2.036(5) and
2.069(5) Å). Compared to the literature, these values suggest a LS Fe(III) centre [33,47–
50].
The crystal structure of 1·0.5H2O comprises alternating layers of [Fe(salEen)2]+
cations and Cl− anions parallel to the ac plane with H2O molecules located in the void
spaces (Fig. 2). The layers are interconnected through short intermolecular contacts
between neighbouring cations and through hydrogen bonds between the amine groups
of the salEen cations and the Cl− anions (N1-H1⋅⋅⋅Cl1 and N3-H3⋅⋅⋅Cl1, d(D⋅⋅⋅A) =
3.269(6) and 3.192(6) Å, respectively).
Fig. 1. ORTEP-3 diagram of 1·0.5H2O with co-crystallised H2O, using 30 %
probability level ellipsoids. Hydrogen atoms are omitted for clarity.
7

Fig. 2. Mercury packing diagram of 1·0.5H2O viewed along the c axis showing the
layers of [Fe(salEen)2]+ cations alternating with Cl− anions and H2O molecules.
Hydrogen bonds between the amine groups of the salEen cations and the Cl− anions are
represented as light blue dashed lines.
Complex 2 crystallises in the monoclinic system, space group P21, with two
[Fe(salEen)2]+ cation, two Cl− anions, and one water molecule in the asymmetric unit.
An ORTEP view of 2·0.5H2O (only one of the molecules, molecule 1, and the co-
crystallised water molecule are shown) is depicted in Fig. 3 and selected bond distances
and angles for both molecules 1 and 2 (data obtained at 150 K) are listed in Table 2.
Similarly to 1·0.5H2O, the coordination around each iron(III) centre is distorted
octahedral with two salEen ligands bound in the meridional coordination mode. In both
molecules 1 and 2, the Fe-O bond lengths are shorter than the Fe-N(imine) distances,
which are shorter than the Fe-N(amine) as observed for complex 1. Although molecules
1 and 2 are very similar, they are crystallographically different: the Fe-O and Fe-N bond
distances are longer in molecule 2 than in molecule 1. The values observed for these
molecules suggest a LS Fe(III) centre in molecule 1 and a HS Fe(III) centre in molecule
2 [33,47–50].
The crystal packing of 2·0.5H2O depicted in Fig. 4 shows alternating layers of
[Fe(salEen)2]+ cations and Cl− anions parallel to the ac plane. Co-crystallised H2O
molecules are located between the Cl− anions. As observed in the structure of 1·0.5H2O
the layers are interconnected through short intermolecular contacts between
neighbouring cations and through hydrogen bonds between the amine groups of the
salEen cations and the Cl− anions (N1-H1⋅⋅⋅Cl1 and N3-H3⋅⋅⋅Cl1). The H2O molecule is
hydrogen bonded to both anions of molecules 1 and 2 (Table 3).
8

Fig. 3. ORTEP-3 diagram of 2·0.5H2O (molecule 1) with co-crystallised H2O, using 30
% probability level ellipsoids. Hydrogen atoms are omitted for clarity.
Fig. 4. Mercury packing diagram of 2·0.5H2O viewed along the c axis, showing the
layers of [Fe(salEen)2]+ cations alternating with Cl− anions and H2O molecules.
Hydrogen bonds between the amine groups of the salEen cations and the Cl− anions are
represented as light blue dashed lines.
Table 2
Selected bond lengths (Å) and angles (°) for 1·0.5H2O and 2·0.5H2O.
2·0.5H2O
1·0.5H2O
molecule 1 molecule 2
Fe(1)-O(1)
1.885(4)
1.866(11)
1.955(12)
9

Fe(1)-O(2)
1.874(4)
2.069(5)
1.927(5)
2.036(5)
1.917(5)
93.1(2)
93.4(2)
89.1(2)
86.7(2)
93.4(2)
177.6(2)
177.4(2)
86.7(2)
84.1(2)
95.9(2)
89.1(2)
175.3(2)
95.0(2)
82.6(2)
91.3(2)
1.870(12)
2.048(13)
1.939(13)
2.052(15)
1.903(15)
93.2(5)
1.938(12)
2.192(14)
2.077(17)
2.223(14)
2.097(15)
94.9(5)
Fe(1)-N(1)
Fe(1)-N(2)
Fe(1)-N(3)
Fe(1)-N(4)
O(2)-Fe(1)-O(1)
O(2)-Fe(1)-N(4)
O(1)-Fe(1)-N(4)
O(2)-Fe(1)-N(2)
O(1)-Fe(1)-N(2)
N(4)-Fe(1)-N(2)
O(2)-Fe(1)-N(3)
O(1)-Fe(1)-N(3)
N(4)-Fe(1)-N(3)
N(2)-Fe(1)-N(3)
O(2)-Fe(1)-N(1)
O(1)-Fe(1)-N(1)
N(4)-Fe(1)-N(1)
N(2)-Fe(1)-N(1)
N(3)-Fe(1)-N(1)
95.0(6)
86.4(5)
85.5(5)
93.7(5)
86.6(5)
96.0(6)
93.7(5)
87.9(6)
178.2(6)
177.6(5)
88.4(6)
177.0(6)
163.8(5)
94.2(6)
83.3(6)
79.7(6)
95.1(6)
86.9(5)
97.7(6)
88.6(5)
177.0(5)
97.5(5)
165.5(5)
100.6(5)
77.7(6)
83.3(5)
91.6(6)
86.0(6)
Table 3
Hydrogen bonds between the cationic and anionic layers in 1·0.5H2O and 2·0.5H2O.
angle (°)
distance (Å)
D⋅⋅⋅H H⋅⋅⋅A
1·0.5H2O
D−H⋅⋅⋅A
D−H⋅⋅⋅A
D⋅⋅⋅A
177
177
0.91
0.91
2.36
2.28
3.269(6)
3.192(6)
N(1)-H(1)⋅⋅⋅Cl(1)
N(3)-H(3)⋅⋅⋅Cl(1)
2·0.5H2O
molecule 1
171
163
0.91
0.91
2.32
2.38
3.227(16)
3.264(18)
3.251
N(1)-H(1)⋅⋅⋅Cl(1)
N(3)-H(3)⋅⋅⋅Cl(1)
O(3)⋅⋅⋅Cl(1)
molecule 2
162
159
0.91
0.91
2.34
2.42
3.220(16)
3.286(19)
3.141
N(1)-H(1)⋅⋅⋅Cl(1)
N(3)-H(3)⋅⋅⋅Cl(1)
O(3)⋅⋅⋅Cl(1)
DFT calculations (Gaussian09 [51], PBE0 [52,53], 6-31G** [54–58] and
sdd+polarization for Fe) were performed in order to estimate the energy difference
between the HS and the LS forms of complexes 1 and 2, using three models in the
geometry optimization: the cation, the cation with two N-H•••Cl hydrogen bonds, and
10

this salt with one water molecule hydrogen bonded to the chloride ion (O-H•••Cl). The
introduction of the chloride modifies the bond lengths by ~0.01 to 0.07 Å, while the
effect of the water molecule is even smaller. The second model (cation and chloride)
gives the best agreement with the experimental distances (details, Table S1 and Fig. S5
in supplementary material). For this model of complex 2, the Fe-O distances are 1.945
(HS) and 1.887 Å (LS), the Fe-N(imine) distances 2.123 (HS) and 1.931 Å (LS), and
the Fe-N(amine) distances 2.207 (HS) and 2.041 Å (LS), in good agreement with the
distances observed in molecule 1 (LS) and molecule 2 (HS) and their assignment. The
HS-LS energy difference is 0.7 kcal mol-1 lower for complex 2. It is thus more likely to
find the coexistence of the two structures in complex 2. On the other hand, the model is
very simple and the value is indicative.
3.3. NMR studies
The magnetic behaviour in solution at rt of all three complexes was assessed by
NMR spectroscopy (magnetic molar susceptibility-χm and spin lattice relaxation time-
T1). Unfortunately the solubility of compound 3 did not allow a correct determination of
both χmT and T1 values.
χmT values determined by the Evans’ method (Table 4, Fig. S6) show that at
different concentrations their values remain practically constant, with χmT changing
from an average value of 4 cm3 mol-1 K for compound 1 to 3 cm3 mol-1 K for compound
2. Taking into account the spin only values for an octahedral Fe(III) complex for both
HS and LS configurations we can conclude that while compound 1 is essentially HS (≈
90 %), compound 2 shows a mixture of spin states (HS ≈ 64 %). A similar ring
substituent effect has been reported by Tweedle and Wilson [59] and was also recently
observed by our group [33].
Table 4
χmT values obtained from 1H NMR results in DMSO-d6 for 1 and 2 at rt.
χmT (cm3 mol-1 K)
[c]
(mM)
5
1
(R = H)
3.7
2
(R = 5-Br)
3.0
4.1
3.1
10
4.1
2.8
2.9
50
100
--*
* Several attempts gave irreproducible results.
To assess the potential of the complexes as MRI contrast agents, we used NMRD to
determine their proton spin-lattice relaxivity enhancement effect (r1) and to gain insight
into the relaxation mechanisms of the solvent molecules. The source of relaxation
increase is the modulation by dynamics of the intermolecular dipole-dipole interaction
between the nuclear spin moment of the solvent molecules and the electron magnetic
moment of the magnetic species [16].
In general, the experimental measured T1 can be written as
ꢏ
ꢏ
ꢆ ꢎ
+ ꢓ ꢔ
=
Eq. (2)
ꢏ
ꢐ
ꢑ
ꢐ
ꢑꢒ
11

where T1d is the observed relaxation time of the solvent (DMSO) in the absence of
paramagnetic molecules (s), [c] is the concentration of the ionic species (mM). r1
measures the efficiency of the agent referred to 1 mM concentration and it is expected
to be concentration independent (mM-1 s-1) [60].
1H T1 can be affected by the presence of magnetic ions in two ways: (i) the so-called
inner-sphere (is) relaxation, associated to the protons bound temporarily to ions or ion
complexes, and (ii) the outer-sphere (os) relaxation, which applies to protons that move
or diffuse close to magnetic ions or particles [61][16]. Then, the relaxivity defined in
Eq. (2) can be written as the sum of the two contributions
ꢓ = ꢓꢕꢖ + ꢓꢁꢖ
Eq. (3)
ꢏ
ꢏ
ꢏ
Fig. 5. r1 dependence with νL,H at concentrations 1, 10 and 100 mM for compound 2.
The inset shows the relaxation rate for the same samples.
Fig. 5 shows r1 at three different concentrations (1, 10 and 100 mM) for complex 2,
using Eq. (2) with T1d = 0.52 s-1 for a reference sample of DMSO. For 100 mM the best
match with the other two r1 data sets was obtained using a different concentration, [c’] =
35 mM.
The inset in Fig. 5 shows the relaxation rate (1/T1) for the three different
concentrations studied. A clear effect on total measured relaxation rate is observed with
increasing concentration.
Fig. 6 shows the results of 1/T1 vs. [c] for compound 2, [2], in the range 0-10 mM at
three different νL,H (100 kHz, 41 MHz and 300 MHz). A clear linear dependence is
observed, thus r1 can be obtained using Eq. (2). The inset plot also shows the values of
1/T1 at 100 mM where a clear deviation from the linear dependency is observed. This
deviation justifies the use of [c’] to obtain r1 dispersion in Fig. 5. These r1 values are on
the same order of magnitude respect to other Fe(III) complexes reported by Richardson
et al. [23].
12

Fig. 6. 1/T1 vs [2] at three different νL,H (100 kHz, 41 MHz and 300 MHz) and the
corresponding r1 values. The inset plot also shows the values of 1/T1 for 100 mM where
a clear deviation from the linear behaviour is observed. The dotted lines are a guide for
the eyes.
The fact that r1 dispersions for 1, 10 and 100 mM in Fig. 5 show some differences
suggests that deviations from Eq. (2) might be expected.
Fig. 7. Fit of the r1 dispersion for 1 mM of compound 2 using Eqs. (2) and (3). The
contributions of the is (dashed) and os (dash dot) to the total r1 are also shown.
Fig. 7 shows the fit of the 1 mM solution for 2 using Eq. (2) with both models for is
and os for paramagnetic complexes [16]. Details on the relaxation models and fitting
parameters are presented in the supplementary material [62]. From the is model the
value obtained for the distance between a Fe3+ ion and a communicating DMSO
molecule was rFeH = 4.70 0.07 Å. From the contribution of the os model, the minimum
distance for the interaction between an Fe3+ ion and an interacting DMSO molecule was
found to be RFeH = 22 1 Å. This distance was obtained from the measured diffusion
constant of the DMSO molecules D = 7.1 × 10-10 m2 s-1, which is slightly smaller than D
13

= 7.4 × 10-10 m2 s-1 of the pure solvent. Additionally the correlation time of the
electronic zero-field-splitting (ZFS) relaxation τv = 1.2 0.2 ps, and the mean squared
fluctuation of the ZFS ∆2 = (4.8 0.2) × 1020 s-2 were estimated from the fitting
parameters.
The r1 dispersions for the two remaining concentrations (10 and 100 mM) were also
fitted using the same model obtaining different values for τv and for the correlation time
associated with the inner sphere relaxation, τmH (see supplementary material). As a
consequence, the is and os paramagnetic relaxation contributions are affected, pointing
to a possible state of aggregation of individual molecules.
3.4. Biological tests
According to the NMR studies these compounds show potential as MRI contrast
agents and it is important to estimate their toxicity and effect on cell regulation.
Treatment of T84 cells with the compounds 1-3 for 24 hours resulted in varying
degrees of LDH release. Of the three compounds studied, compound 3 was the least
toxic and significantly increased LDH release from T84 monolayers only at
concentrations of 1 mM (see supplementary material). Compound 1 was the second
most toxic, causing a small, but significant, increase in LDH release at 500 µM (see
supplementary material). Compound 2 was the most toxic causing a high degree of cell
lysis at 500 µM (Fig. 8). Overall, these complexes exert toxic effects only at high
concentrations.
14

***
***
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
a)
***
Control 1nM 100nM
1mM Lysate
10µM 100µM 500µM
[2] 24 hours
1600
1400
1200
1000
800
600
400
200
0
b)
Control
1 nM
100 nM
10 µM
100 µM
500 µM
1mM
*
***
**
**
**
**
R
***
***
25
0
5
10
15
20
30
Time (hours)
Fig. 8. High concentrations of compound 2 exert toxic effects on T84 monolayers. a)
T84 monolayers were treated bilaterally with compound 2 for 24 hours. An equal aliquot
of apical and basolateral medium was used to measure LDH release from the cells (n =
4 – 8). b) T84 monolayers were treated with compound 2 at various concentrations (1
nM to 1 mM) over a 24 hr period during with Rte was measured (n = 4 – 8).
Having ascertained that each of the three compounds caused LDH release when used
at high concentrations, we next wished to determine whether the compounds caused a
concomitant reduction in barrier function, measured as Rte. Compound 3 (500 µM, 24
hrs) reduced Rte to 430.7 Ω cm2 compared to controls (812.5 Ω cm2) (n = 4 – 8, p ≤
0.01). Similarly to 3, compound 1 (500 µM, 24 hrs) reduced Rte to 599.1 58.6 Ω cm2
compared to controls (1155.3 42.2 Ω cm2; n = 5 – 6, p ≤ 0.001), whereas compound 2
(500 µM, 24 hrs) abolished Rte, while compound 2 (100 µM, 24 hrs) reduced Rte to
371.7 55.6 µ cm2 compared to control (889.2 59.5 Ω cm2; n = 4 – 8, p ≤ 0.01). Thus
the compounds analysed reduce Rte across T84 colonic epithelial monolayers at
concentrations greater than 10 µM.
15

CCh (100 µM)
FSK (10 µM)
a)
Control
2 (100 µM, 24 hrs)
100
50
0
I
0
5
10 15 20 25 30 35 40
Time (mins)
140
120
100
80
b)
60
*
40
I
CCh (100 µM)
FSK (10 µM)
20
0
Control 1 nM 100 nM
10 µM 100 µM
[2]
Fig. 9. Compound 2 attenuates Cl- secretion across T84 cells. a) T84 cell monolayers
were treated bilaterally with compound 2 (100 µM) in serum- free medium for 24 hrs.
Cells were then washed and mounted in Ussing chambers, and Isc responses to CCh
(100 µM) and FSK (10 µM) were measured (n = 6 – 8). b) T84 monolayers were treated
with compound 2 at various concentrations (1 nM to 100 µM) for 24 hrs before Isc
responses to CCh and FSK were measured (n = 6 – 8).
Following barrier function measurements we went on to determine whether the
compounds altered transepithelial Cl- secretion across T84 monolayers, a primary
function of colonic epithelial cells. Pre-treatment of T84 cell monolayers with the
compounds 1-3 for 24 hrs significantly attenuated subsequent Cl- secretory responses to
the Ca2+- and cAMP-dependent secretagogues, carbachol (CCh) and forskolin (FSK),
Fig. 9. In 3 (500 µM)-treated monolayers, peak responses to CCh and FSK were 46.4
2.6 µA cm-2 (n = 4 – 8; ns) and 45.8 6.7 µA cm-2 (n = 4 – 8; p ≤ 0.01) compared to
those in control cells of 76.2 12.7 µA cm-2 and 100.5 13.0 µA cm-2, respectively. In
1 (500 µM)-treated monolayers, peak responses to CCh and FSK were 45.3 8.8 µA
cm-2 (n = 6; ns) and 55.6 5.6 µA cm-2 (n = 6; p ≤ 0.001) compared to those in control
cells of 74.2 11.0 µA cm-2 and 135.8 9.5 µA cm-2 respectively. Finally, in 2 (100
µM)-treated monolayers, peak responses to CCh and FSK were 32.5 6.3 µA cm-2 (n =
6 – 8; ns) and 47.0 4.6 µA cm-2 (n = 6 – 8; p ≤ 0.05) compared to those in control cells
of 86.2
18.8 µA cm-2 and 107.1
15.3 µA cm-2 respectively. The effects of
16

compounds 1-3 were concentration-dependent with anti-secretory effects occurring at
concentrations which concomitantly reduced barrier function and increased LDH
release. These results suggest that the reduction in agonist-stimulated Cl- secretion from
T84 cells elicited by compounds 1–3 may be due to their toxic effects. While the
reductions in Cl- secretion appear to coincide with increased toxicity, further
investigations of this effect would elucidate what components of the Cl- secretory
pathway are being affected by the compounds.
4. Conclusions
In this work we reported the synthesis and characterization of three iron(III)
complexes with ligands derived from N-ethyl-N-(2-aminoethyl)salicylaldiminate (H, 1;
5-Br, 2; 3-OMe, 3 substituents at the phenyl group).
Investigation of the magnetic behaviour of compounds 1 and 2 in DMSO solutions
showed that while compound 1 is predominantly in the HS state, compound 2 shows a
mixture of spin states. NMRD measurements showed that scarcely explored Fe(III)
complexes reduce the T1 as commonly observed for other contrast agents. The relaxivity
behaviour can be explained by a sum of inner and outer sphere spin-lattice relaxation
contributions. Biological assessment of cell toxicity and regulation showed that all three
compounds have a toxic effect only at high concentrations (> 100 µM), and the effects
reflect a reduction in colonic epithelial secretory function.
The hypothesis of association between individual molecules without spin coupling,
suggested by the analysis of the spin-lattice relaxation data, will be explored as a future
development of the present work.
Finally, we can conclude that, despite the low toxicity, further studies are needed in
order to evaluate the potential of these complexes as MRI contrast agents. It would be
interesting to translate the NMR studies into a biological medium to assess their MRI
potential.
Acknowledgements
We thank Fundação para Ciência
e
Tecnologia for financial support
(UID/MULTI/00612/2013, PEst-OE/QUI/UI0536/2011, PEst-OE/FIS/UI0261/2011,
PEst-OE/FIS/UI0261/2014) and fellowships to PNM and S. Barroso
(SFRH/BPD/73345/2010 and SFRH/BPD/7394/2010). S. Barroso is grateful to Centro
de Química Estrutural for the access to crystallography facilities.
Supplementary material
Data for structures 1⋅0.5H2O and 2⋅0.5H2O were deposited in CCDC under the
deposit numbers CCDC 1018927-1018928 and can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.uk/data_request/cif.
Supplementary data associated to this article can be found in the online version at …
Bibliography
[1] W. Semmler, M. Schwaiger, eds., Molecular Imaging I, Springer, 2008.
[2] T.F. Massoud, S.S. Gambhir, Genes Dev. 17 (2003) 545.
17

[3] M. Rudin, R. Weissleder, Nat. Rev. Drug Discov. 2 (2003) 123.
[4] D.W. McRobbie, E.A. Moore, M.J. Graves, M.R. Prince, MRI: From Picture to
Proton, 2nd ed., Cambridge University Press, New York, 2006.
[5] C.J. Jones, J.R. Thornback, Medicinal Applications of Coordination Chemistry,
Royal Society of Chemistry, Cambridge, 2007.
[6] H.S. Thomsen, S.K. Morcos, T. Almén, M.-F. Bellin, M. Bertolotto, G. Bongartz,
O. Clement, P. Leander, G. Heinz-Peer, P. Reimer, F. Stacul, A. van der Molen,
J.A. Webb, Eur. Radiol. 23 (2013) 307.
[7] Y. Xiao, Y.-J. Wu, W.-J. Zhang, X.-J. Li, F.-K. Pei, Chinese J. Anal. Chem. 39
(2011) 757.
[8] S.-P. Lin, J.J. Brown, J. Magn. Reson. Imaging 25 (2007) 884.
[9] T. Grobner, Nephrol. Dial. Transplant 21 (2006) 1104.
[10] T. Grobner, Nephrol. Dial. Transplant. 21 (2006) 1745.
[11] R. Yacoub, K. Kandukurti, M. Panesar, ISRN Transplant. 2013 (2013) 1.
[12] M.F. Wendland, NMR Biomed. 17 (2004) 581.
[13] A.C. Silva, N.A. Bock, Schizophr. Bull. 34 (2008) 595.
[14] A.G. Roca, R. Costo, A.F. Rebolledo, S. Veintemillas-Verdaguer, P. Tartaj, T.
González-Carreño, M.P. Morales, C.J. Serna, J. Phys. D. Appl. Phys. 42 (2009)
224002.
[15] A.C. Silva, J.H. Lee, I. Aoki, A.P. Koretsky, NMR Biomed. 17 (2004) 532.
[16] A. Merbach, L. Helm, É. Tóth, The Chemistry of Contrast Agents in The
Chemistry of Contrast Agents in Magnetic Resonance Imaging, 2nd ed., John
Wiley & Sons, Ltd., 2013.
[17] S.J. Dorazio, J.R. Morrow, Eur. J. Inorg. Chem. 2012 (2012) 2006.
[18] S.J. Dorazio, P.B. Tsitovich, K.E. Siters, J.A. Spernyak, J.R. Morrow, J. Am.
Chem. Soc. 133 (2011) 14154.
[19] S.J. Dorazio, J.R. Morrow, Inorg. Chem. 51 (2012) 7448.
[20] P.B. Tsitovich, J.R. Morrow, Inorganica Chim. Acta 393 (2012) 3.
[21] S.J. Dorazio, P.B. Tsitovich, S. a Gardina, J.R. Morrow, J. Inorg. Biochem. 117
(2012) 212.
[22] D.L. White, Magn. Reson. Med. 312 (1991) 309.
18

[23] N. Richardson, J.A. Davies, B. Radüchel, Polyhedron 18 (1999) 2457.
[24] D.D. Schwert, N. Richardson, G. Ji, B. Radüchel, W. Ebert, P.E. Heffner, R.
Keck, J.A. Davies, J. Med. Chem. 48 (2005) 7482.
[25] P.B. Tsitovich, J.A. Spernyak, J.R. Morrow, Angew. Chem. Int. Ed. Engl. 52
(2013) 13997.
[26] S.J. Dorazio, A.O. Olatunde, J.A. Spernyak, J.R. Morrow, Chem. Commun.
(Camb). 49 (2013) 10025.
[27] A.O. Olatunde, S.J. Dorazio, J.A. Spernyak, J.R. Morrow, J. Am. Chem. Soc.
134 (2012) 18503.
[28] C.E. Housecroft, A.G. Sharpe, Inorganic Chemistry, 2nd ed., Pearson Education
Limited, 2005.
[29] V. Stavila, M. Allali, L. Canaple, Y. Stortz, C. Franc, P. Maurin, O. Beuf, O.
Dufay, J. Samarut, M. Janier, J. Hasserodt, New J. Chem. 32 (2008) 428.
[30] X. Zhao, H. Zhao, Z. Chen, M. Lan, J. Nanosci. Nanotechnol. 14 (2014) 210.
[31] Y. Jun, J.-H. Lee, J. Cheon, Angew. Chem. Int. Ed. Engl. 47 (2008) 5122.
[32] S. Kumar, D.N. Dhar, P.N. Saxena, J. Sci. 68 (2009) 181.
[33] P.N. Martinho, A.I. Vicente, S. Realista, M.S. Saraiva, A.I. Melato, P. Brandão,
L.P. Ferreira, M. de D. Carvalho, J. Organomet. Chem. (2014).
[34] K.E. Barrett, S.J. Keely, Annu. Rev. Physiol. 62 (2000) 535.
[35] D.F. Evans, J. Chem. Soc. (1959) 2003.
[36] G.M. Sheldrick, SADABS,Empiracal Absorpt. Correct. Program, Univ.
Göttingen (1995) based on the methods of Blessing.
[37] M.C. Burla, R. Caliandro, M. Camalli, B. Carrozzini, G.L. Cascarano, L. De
Caro, C. Giacovazzo, G. Polidori, R. Spagna, J. Appl. Crystallogr. 38 (2005) 381.
[38] L.J. Farrugia, J. Appl. Crystallogr. 32 (1999) 837.
[39] G. Sheldrick, Göttingen, Ger. (1997).
[40] G.M. Sheldrick, Acta Crystallogr. A. 64 (2008) 112.
[41] L.J. Farrugia, J. Appl. Crystallogr. 30 (1997) 565.
[42] D.M. Sousa, G.D. Marques, J.M. Cascais, P.J. Sebastião, Solid State Nucl. Magn.
Reson. 38 (2010) 36.
19

[43] W.S. Price, Concepts Magn. Reson. 9 (1997) 299.
[44] J.B.J. Ward, K. Lawler, S. Amu, C.T. Taylor, P.G. Fallon, S.J. Keely, FASEB J.
25 (2011) 535.
[45] C.A. Cartwright, J.A. McRoberts, K.G. Mandel, K. Dharmsathaphorn, J. Clin.
Invest. 76 (1985) 1837.
[46] C.N. Field, M. Boillot, R. Clément, J. Mater. Chem. 8 (1998) 283.
[47] C. Faulmann, K. Jacob, S. Dorbes, S. Lampert, I. Malfant, M.L. Doublet, L.
Valade, J.A. Real, Inorg. Chem. 46 (2007) 8548.
[48] A. Tissot, R. Bertoni, E. Collet, L. Toupet, M.-L. Boillot, J. Mater. Chem. 21
(2011) 18347.
[49] C. Faulmann, J. Chahine, L. Valade, G. Chastanet, J.-F. Létard, D. de Caro, Eur.
J. Inorg. Chem. 2013 (2013) 1058.
[50] C.-F. Sheu, S.-M. Chen, G.-H. Lee, Y.-H. Liu, Y.-S. Wen, J.-J. Lee, Y.-C.
Chuang, Y. Wang, Eur. J. Inorg. Chem. 2013 (2013) 894.
[51] Gaussian 09, Revision A.02, M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E.
Scuseria, M.A. Robb, J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci,
G.A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov,
J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven,
J. Montgomery, J. A., J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E.
Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K.
Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N.
Rega, M.J. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J.
Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C.
Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A.
Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, Ö. Farkas, J.B.
Foresman, J. V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian, Inc., Wallingford CT,
2009.
[52] J. Perdew, Phys. Rev. B 33 (1986) 8822.
[53] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 78 (1997) 1396.
[54] R. Ditchfield, J. Chem. Phys. 54 (1971) 724.
[55] W.J. Hehre, J. Chem. Phys. 56 (1972) 2257.
[56] P.C. Hariharan, J.A. Pople, Theor. Chim. Acta 28 (1973) 213.
[57] P.C. Hariharan, J.A. Pople, Mol. Phys. 27 (1974) 209.
[58] M.S. Gordon, Chem. Phys. Lett. 76 (1980) 163.
20

[59] M.F. Tweedle, L.J. Wilson, J. Am. Chem. Soc. 98 (1976) 4824.
[60] E.J. Werner, A. Datta, C.J. Jocher, K.N. Raymond, Angew. Chem. Int. Ed. Engl.
47 (2008) 8568.
[61] P. Gillis, A. Roch, R.A. Brooks, J. Magn. Reson. 137 (1999) 402.
[62] P.J. Sebastião, Eur. J. Phys. 35 (2014) 015017.
21

Synopsis for the Graphical Abstract
Three new Fe(III) compounds of general formula [Fe(R-salEen)2]Cl (R = H, 1; 5-Br, 2; 3-
OMe, 3) have been synthesised. The relaxivities for 2 are in agreement with those known for
other iron(III) contrast agents and showed low toxic effects, thus making these good
candidates for MRI contrast agents.

Highlights
•
•
•
Three new Fe(III) complexes of salen derivatives were synthesised.
X-ray crystal structures of two complexes reported.
The energy difference between spin states was calculated as ~7 kcal.mol-1
(DFT).
•
•
The Fe(III) complexes reduce T1 as commonly observed for other contrast
agents.
The compounds have a toxic effect only at high concentrations.
22