A study on the synthesis and properties of substituted EHBG-Fe(III) complexes as potential MRI contrast agents

Article abstract of DOI:10.1016/j.jorganchem.2014.07.011

Mono- and disubstituted EHBG (HBED) ligands of different kind: forming hydrogen bonds (p-OMe, p-NHAc) and of high lipophilicity (3,4-Me₂, p-Ph) were obtained. The ligands were complexed with Fe(III). Complexes so obtained were studied for their stability in different pH environments; and r₁ relaxivity, which turned out the highest for Fe-EHBG-NHAc, i.e. 0.95 mM^-1 s^-1 in aqueous solution and 1.34 mM^-1 s^-1 for Fe-EHBG-Me₂ in bovine serum. EPR signals were observed at g-factor of 4.27 indicative of Kramer's doublet transitions of a rhombically distorted high-spin paramagnetic center.

Full text of DOI:10.1016/j.jorganchem.2014.07.011

Journal of Organometallic Chemistry 769 (2014) 100e105
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
A study on the synthesis and properties of substituted EHBG-Fe(III)
complexes as potential MRI contrast agents
*
ꢀ
_
Nikodem Kuznik , Grazyna Szafraniec-Gorol, Lidia Oczek, Artur Grucela, Paweł Jewuła,
ꢀ
Anna Kuznik, Pawel Zassowski, Wojciech Domagala
Faculty of Chemistry, Silesian University of Technology, M. Strzody 9, 44-100 Gliwice, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
Mono- and disubstituted EHBG (HBED) ligands of different kind: forming hydrogen bonds (p-OMe, p-
NHAc) and of high lipophilicity (3,4-Me2, p-Ph) were obtained. The ligands were complexed with Fe(III).
Complexes so obtained were studied for their stability in different pH environments; and r1 relaxivity,
which turned out the highest for Fe-EHBG-NHAc, i.e. 0.95 mMꢀ1 sꢀ1 in aqueous solution and
1.34 mMꢀ1 sꢀ1 for Fe-EHBG-Me2 in bovine serum. EPR signals were observed at g-factor of 4.27 indicative
of Kramer's doublet transitions of a rhombically distorted high-spin paramagnetic center.
© 2014 Elsevier B.V. All rights reserved.
Received 28 March 2014
Received in revised form
13 July 2014
Accepted 15 July 2014
Available online 24 July 2014
Keywords:
Fe(III) complexes
EHBG e N,N0-ethylenebis[(2-
hydroxybenzyl)glycine]
HBED e N,N0-di(2-hydroxybenzyl)
ethylenediamine-N,N0-diacetic acid
Contrast agents
Relaxivity
EPR
Introduction
tissue, the investigations on the applications in MRI have not been
pursued further. Recent reports of harmfulness of gadolinium sys-
EHBG - N,N0-ethylenebis[(2-hydroxybenzyl)glycine], known also
as HBED, is a classic hexadentate ligand used particularly with first
row transition metals [1]. As a versatile, yet relatively simple ligand,
it is used as a model for the transferrin binding site [2,3]. EHBG forms
very stable complexes with Fe(III), whose logK reach values of 39
[4,5], and for this reason EHBG has been studied as an iron chelating
agent for iron overload disease (hemochromatosis) [6]. Stability of
its iron complexes and lipophilicity higher than that of EHPG com-
plexes were prerequisites for their potential application as blood
pool MRI contrast agents [7,8]. Moreover, 5-bromo derivative of
HBED was found to bind avidly to human serum albumin (HSA)
[3,4,9]. Despite promising results of relatively high relaxivities of
Fe(III) complexes, high affinity to HSA, high uptake by the tumor
tems, causing nephrogenic systemic fibrosis (NSF) in patients
suffering renal failure [10,11], may be a turning point in the pursuit
for new contrast agents based on endogenous metals. High abun-
dance of iron in nature together its low price, as opposed to gado-
linium being considered a critical metal for global market due to its
limited availability, favor this metal as an ideal alternative. Recog-
nizing the novelty and importance of this research topic, we un-
dertook the study of the properties of selected group of EHBG
derivative ligands with respect to their influence on the longitudinal
relaxation time (T1) and redox properties of their respective Fe(III)
complexes. We designed the ligand to form potential hydrogen
bonds with water in order to enhance their interaction at the second
coordination sphere. On the other hand, the model structure with
phenyl substituents was expected to express slower tumbling rate,
which could increase its relaxative properties. Together with
dimethyl substituted derivative they may express high affinity to
lipophilic targets. The results of these studies, accompanied with
structural discussion of the complexes, are presented in this work.
* Corresponding author. Tel.: þ48 32 2371839; fax: þ48 32 2372094.
E-mail addresses: nikodem.kuznik@polsl.pl, nikodem.kuznik@gmail.com
ꢀ
(N. Kuznik).
http://dx.doi.org/10.1016/j.jorganchem.2014.07.011
0022-328X/© 2014 Elsevier B.V. All rights reserved.

ꢀ
N. Kuznik et al. / Journal of Organometallic Chemistry 769 (2014) 100e105
101
Table 1
Results and discussion
Ligand 1a-1d and complex 2a-2d yields. Purified products.
Synthesis of EHBG derivatives
Ligand
Ligand yield
Complex 2a-2d yield
EHBG-OMe (1a)
EHBG-NHAc (1b)
EHBG-Me2 (1c)
EHBG-Ph (1d)
10
33
19
59
46
22
69
92
Synthesis of EHBG ligands. Ethylenediamine-N,N0-diacetic acid
(EDDA) was synthesized according to Haydock's procedure pre-
sented in Ref. [12]. The product was purified and used for the
synthesis of EHBG ligands according to the reaction sequence
presented in Scheme 1.
Martell described the synthesis of EHBG as a multistep process,
involving reaction of formaldehyde and HCN to N,N0-bis(2-
hydroxybenzyl)ethylamine-1,2-diamine, followed by the hydroly-
sis to diacid [13]. The method was too complicated and not
designed for para substituted derivatives. Later, Wilson presented a
convenient method applying the Mannich condensations [14]. The
latter route was successfully applied by us for the synthesis of EHBG
derivatives, affording the desired products in satisfactory yields,
taking into account the multimolecular one-pot reaction (see
Table 1). The most troublesome step was the separation and puri-
fication of the products. It was necessary to transform them into
zwitterionic form at an appropriate pH value. This afforded them as
semisolid bodies, which were purified in alcohol solutions.
Table 2
Relaxivities of EHBG complexes.
Complex
r1 in H2O
Std. dev. r1
in H2O
r1 in serum
Std. dev. r1
in serum
[mMꢀ1 sꢀ1
]
[mMꢀ1 sꢀ1
]
Fe-EHBG-OMe (2a)
Fe-EHBG-NHAc (2b)
Fe-EHBG-Me2 (2c)
Fe-EHBG-Ph (2d)
Fe-EHPG
0.72
0.95
0.76
0.14
0.81
0.02
0.09
0.04
0.01
0.04
1.31
1.09
1.34
0.81
1.36
0.03
0.03
0.01
0.02
0.05
Measured at 300 MHz, 7.1 T, 22 ꢁC.
so in favour of the carboxylate form, that definitive identification of
the enol one may exceed the sensitivity limits of employed
analytical methods.
Synthesis of EHBG complexes
Properties of EHBG complexes
Iron(III) complexes were synthesized by a modification of an
analogous reaction leading to Bailey's EHPG complexes (Scheme 2)
[15], giving products in varying, but good yields (Table 1). The ob-
tained complexes were purified by means of classic chromatog-
raphy (TLC preparative plates or chromatography columns). Only
one, colored fraction was collected. IR analysis of each complex
revealed a broad band in the range of 3348e3282 cmꢀ1 of weak
intensity. This band was also present for dry, lyophilized complex
samples. This band corresponds to a typical stretching vibration of
eOH group, unexpected in these models. Consequently, we propose
a hypothesis that carboxylate group of the ligand may rearrange
into an appropriate enolate group. This form may be additionally
stabilized by intermolecular hydrogen bonding, as depicted in
Scheme 3. This hypothesis may be verified by X-ray analysis.
Tautomeric enol form of carboxylic ligand was found in the simple
platinum complex [16]. However, to the best of our knowledge, this
was the only report supported by a definite X-ray structure. Un-
fortunately, our attempts to obtain good quality crystals of our
complexes failed. Mass spectra did not clearly reveal the presence
of water adducts. Only in the case of 2b, a peak [M þ K þ H2O]þ of
15% intensity was observed with respect to the dominant [M þ Na]þ
signal. It is also possible that the tautomeric equilibrium is shifted
The behavior of UVeVis spectra versus pH was studied for one
model 2d e Fig. 1.
In the pH range of 1.9e7.4, the absorption maximum at 257 nm
is found the decrease subtly with decreasing pH. This band repre-
sents the phenolate e iron charge transfer transition. Martell
associated it with the non-ionized form of analog EHPG complexes
[17]. Recent studies by Carrano revealed the possibility of septuple
coordination of Fe-EHPG complexes, with a water molecule at the
seventh coordination site [18]. At lower pH, the later suggests the
[Fe-EHPG(H2O)]ꢀ complex form. At very basic conditions (pH > 10)
the 257 nm band shifts batochromically. Martell interpreted this as
deprotonation of phenolic groups, while Carrano puts forward a
deprotonation mechanism leading to monohydroxo species [Fe-
EHPG(HO)]ꢀ. Both hypotheses are in agreement with an anticipated
bathochromic shift for the more negative species. When pH is be-
tween 7.4 and 10.1, a simple equilibrium of the two species is
observed with two isosbestic points at 230 and 275 nm.
The longitudinal relaxivity (r1) of the obtained complexes was
determined at 7.1 T. Aqueous solutions of EHBG complexes (2a-2d)
demonstrated values ranging from 0.14 to 0.95 mMꢀ1 sꢀ1 (see
Table 2). Relaxivity higher than that of Fe-EHPG (measured as a
Scheme 1. Synthetic sequence leading to EHBG ligands.

ꢀ
102
N. Kuznik et al. / Journal of Organometallic Chemistry 769 (2014) 100e105
Scheme 2. Synthesis of EHBG complexes.
reference [8]) was observed for Fe-EHBG-NHAc (2b). In this case,
water molecules are expected to interact with the NHAc substituent
forming a hydrogen bond in the outer coordination sphere. On the
other hand, decreased relaxivity for the complex with phenyl
substituted ligand may come from a minor impact of the tumbling
time in the net relaxivity expression. The phenyl substituent is
expected to increase nucleophilicity of the complex, however. The
obtained values are close to r1 ¼ 1.6 mMꢀ1 sꢀ1 of iron(III) ammo-
nium citrate (also known as FAC e ferric ammonium citrate) used as
an oral contrast agent (Geritol® and Ferriseltz®), however this
result was registered at much lower field of 16 MHz [19]. Trans-
ferrin, an iron protein responsible for iron transport, has an r1 of
2 mMꢀ1 sꢀ1 measured at 20 MHz [20]. Since r1 decreases with
increasing field, it is expected that the relaxivities of Fe-EHBG
complexes will surpass the abovementioned values. Moreover, r1
registered in bovine serum solutions are much higher than those in
aqueous environment. As postulated above, the highest relaxivity
increase is observed for the most lipophilic derivative with phenyl
substituent 2d. This general relation could be associated with lower
mobility of the molecules, hindering their tumbling. Another factor
is the RIME effect e interaction of paramagnetic species with pro-
teins (in serum). The highest relative increase of r1 (almost six fold)
observed for 2d could be a superposition of both effects. On the
other hand, the highly apolar substituents thwart effective inter-
action of water molecules with the complex leading to lowest
relaxivities in both media. Nevertheless, the relaxivity values
comparable to FAC, together with much higher stability constants
[5], and an endogenous character of the metal (eliminating dele-
terious side effects like NSF) make the Fe-EHBG complexes an
interesting complement to the classic gadolinium contrast agents
for MRI.
Scheme 3. Rearrangement of carboxylate group into enolate one, with consequent
EPR measurements in frozen methanol corroborate the dis-
torted rhombohedral geometry of high spin Fe(III) type complex
[21]. A strong, sharp signal at g ¼ 4.27 can be interpreted as middle
and lower Kramer's doublet transitions [22] (Fig. 2), while the
signal at g ¼ 4.21 and a broad band at g z 4.7 is associated with
weak resonance of the ground Kramer's doublets of axial symme-
try. These assignments are fully consistent with Lauffer's structure
of the unsubstituted Fe-EHBG complex shown in Fig. 3 [9]. Octa-
hedral geometry with pseudo C2 symmetry is evident considering
cis coordination of phenolates and that of carboxylate groups in
trans positions. Additionally the different character of phenolate
oxygen atoms and those of carboxylate groups is traced to their
divergent bond lengths (1.887 and 2.053 Å respectively). These
features incur a minor rhombohedral distortion of the geometry
from an ideal octahedron.
hydrogen bonding stabilization.
Redox activity of complexes was assessed using cyclic voltam-
perometry (CV). Phosphate buffered (pH of 7.4) aqueous solutions
of pure complexes were examined, however, no clear redox signals
Fig. 1. UVeVis spectra of complex 2d as a function of pH.

ꢀ
N. Kuznik et al. / Journal of Organometallic Chemistry 769 (2014) 100e105
103
Materials and methods
All liquid chemicals and solvents were dried and distilled prior
to use. EDDA has been synthesized according to Haydock's proce-
dure (13% yield after purification) [12]. NMR spectra were taken on
a Varian Unity Inova 300 MHz and Agilent 400 MHz spectrometers
using solvent signal as an internal reference at room temperature.
Electrospray-ionization mass spectrometry was performed on a
4000 QTrap (Applied Biosystems/MDS Sciex) mass spectrometer.
High resolution mass spectra were registered on a Micromass/
Waters LCT (TOF) spectrometer at the University of Warsaw.
Relaxivity: T1 has been measured on Varian Unity Inova 300 MHz
spectrometer using IR sequence at 22 ꢁC with t1 ¼ 15 s and t2: 0.05,
0.1, 0.5, 0.75, 1, 3 s. Samples of complexes at concentrations of: 0.3,
0.6, 0.9 and 1.2 mmol/dm3 were dissolved in 5% D2O in H2O mixture
and degassed by nitrogen purging. Samples of 2c and 2d were
dissolved in a mixture of 5% D2O and 10% DMSO in deionized water.
The serum solutions were prepared in analogous way, with
deionized water replaced by bovine serum (Sigma, No. B9433).
Based on the slope of 1/T1 vs. molarity plot the relaxivity value was
computed with an appropriate standard error. EPR measurements
were recorded on JEOL JES-FA 200 X-band spectrometer. Samples
dissolved in methanol were measured in a capillary tube to mini-
mize the dielectric losses. Measurements were conducted at 78 K.
Cyclic voltamperometry was performed on CH Instruments 620
Electrochemical Analyzer potentiostat in aqueous phosphate buffer
solutions (0.1 M, pH ¼ 7.4), with 1 mM concentration of each
complex. Ferrocene/ferrocinium redox couple was used a reference
electrode external standard. UVeVis spectra were taken on Hitachi
Y-2910 spectrophotometer at 190e700 nm range in aqueous so-
lutions. IR spectra were registered on Nicolet 6700 ATC-FTIR
spectrophotometer. EHBG ligands were obtained by modification
of Wilson's method [14].
Fig. 2. EPR spectra of frozen 1 mM methanol solution of 2d at 78 K. Instrument pa-
rameters: microwave modulation freq. 100 kHz, power 0.5 mW, modulation width
1 mT, amplitude 100, time const. 0.01 s, sweep time 2 ꢂ 2 min, complex concentration
1 mM.
were observed in the range of ꢀ1÷1.2 V. This observation mitigates
the concerns about undesired intracorporeal (non)catalytic redox
activity of iron centers. Additionally, EHBG was found to display
antioxidant properties, blocking superoxide ion reduction [23].
Both observations cast an optimistic light on potential medical
applications of this class of complexes.
Conclusions
Negatively charged Fe(III) EHBG complexes can be considered
potential MRI contrast agents, complementary to widespread, yet
exogenous, gadolinium compounds. Possible side effects (NSF) of
gadolinium agents, scarce abundance of this metal in nature, as
well as environmental issues, all advocate for development of
complimentary, or even alternative element MRI diagnostic agents.
The EHBG complexes demonstrate comparably good relaxivity
values to FAC e a common ferric oral contrast agent, while their
high stability, lack of redox activity and an endogenous character of
iron are all important merits compared to the classic gadolinium
agents. Further improvement of relaxivity are foreseen by intro-
duction of polar groups capable of forming hydrogen bonds.
Another design pathway holds promise of improvement of lip-
ophilicity, expected from ligands with aryl alkyl derivatives. Bio-
distribution and selectivity studies are the next step to assess the
potential of these complexes in the medical practice field.
EHBG-OMe (1a)
N,N0-Ethylenediaminodiacetic acid (EDDA) 0.98 g (5 mmol) with
water (2 cm3) and methanol (3 cm3) were introduced into a round-
bottomed flask. The mixture was heated for 20 min at 60 ꢁC. Solid
sodium carbonate 0.54 g (5 mmol) was added, followed by water
(3 cm3) and methanol (1.5 cm3). After 30 min of heating, the so-
lution became transparent. Then, paraformaldehyde 0.32
g
(10 mmol) was added and the heating continued for 1 h. A colorless
Fig. 3. X-Ray structure of [FeEHBG]ꢀ complex according to Lauffer's work [9].

ꢀ
N. Kuznik et al. / Journal of Organometallic Chemistry 769 (2014) 100e105
104
solution was cooled and 4-methoxyphenol 1.26 g (10 mmol) in
methanol (5 cm3) was finally added. Reaction mixture was refluxed
for 8 h. After completion, all the volatiles were removed in vacuo.
An orange solid remained that was extracted with acetone
(3 ꢂ 15 cm3) to remove unreacted substrate. Afterward, it was
dissolved in hot ethanol/water (2:1, 15 cm3 total) mixture. The
resulting solution was acidified to pH ¼ 2.5. The product precipi-
tated upon refrigerating, in 24 h. It was filtered, washed with an
ethanol/water (1:1) mixture followed by hot ethanol (10 cm3).
Finally 220 mg of the product was obtained. Yield: 10%. Mp
EHBG-Ph (1d)
N,N0-Ethylenediaminodiacetic acid (EDDA) 1.55 g (8.8 mmol),
methanol (5 cm3), sodium hydroxide (2 cm3, 30% aqueous soln.)
and 6 g (35 mmol) 4-phenylphenol dissolved in 30 cm3 of methanol
were introduced into a round-bottomed flask equipped with
condenser, pH-electrode and a dropping funnel. pH was adjusted to
8.8 with 10% HCl. Following heat-up, 2.6 cm3 (96 mmol) of 37%
formaldehyde in 8 cm3 methanol was added dropwise. During re-
action, pH was maintained at 8.0. The mixture was refluxed for 7 h
followed by alkalization to pH ¼ 10 with 10% sodium hydroxide.
Solvents were evaporated in vacuo, the remaining reddish solid was
washed with acetone (3 ꢂ 25 cm3). The crude, solid sodium salt of
the product was dissolved in water with methanol (1:1) and acid-
ified to pH ¼ 3.0, which resulted in precipitation of the desired
product. After filtration and drying, 2.81 g (5.2 mmol) of the
product was collected. Yield: 59%. 1H NMR (D2O with Na2CO3):
162e164 ꢁC (decomp.). 1H NMR (d6-DMSO):
d 3.21 (s, 4H,
NeCH2eCH2eN), 3.66 (s, 4H, NeCH2eCOOH), 3.66 (s, 6H,
AreOCH3), 4.02 (s, 4H, NeCH2eAr), 6.75e6.89 (m, 6H, Ar); 13C NMR
(d6-DMSO):
d 50.5 (NeCH2eCH2eN), 53.0 (CAreOCH3), 53.5
(NeCH2eAr), 56.0 (CH2eCOOH), 116.0 (CAr), 116.8 (CAr), 117.4 (CAr),
120.6 (CAr), 150.8 (CAreOH), 152.7 (CAreOCH3), 171.1 (COOH).
d
2.86 (s, 4H, eNeCH2eCH2e), 3.35 (s, 4H, eNeCH2eAr), 3.60 (s,
4H, eNeCH2eCOO), 6.91e6.89 (m, 6H, Ar), 7.80e7.64 (m, 10H, Ar).
13C NMR (D2O with Na2CO3):
55.1 (NeCH2eCH2eN), 62.2
(NeCH2eAr), 70.6 (NeCH2eCOOH), 109.2 (CAr), 113.1 (CAr), 118.4
(CAr), 128.7 (CAr), 143.4 (CAr), 145.0 (CAr), 145.7 (CAreOH), 174.4
(COOH).
EHBG-NHAc (1b)
d
N,N0-Ethylenediaminodiacetic acid (EDDA) 1.76 g (10 mmol)
with water (4 cm3) and methanol (6 cm3) were introduced into a
round-bottomed flask. The mixture was heated for 15 min at 60 ꢁC.
Solid sodium carbonate 1.06 g (10 mmol) was added, followed by
paraformaldehyde 0.6 g (20 mmol) and heated for 1 h. The cloudy
solution was cooled and 4-(acetamino)phenol 3.02 g (10 mmol) in
methanol (14 cm3) was finally added. Reaction mixture was
refluxed for 8 h. After completion, all the volatiles were removed in
vacuo. The residue was extracted with acetone (3 ꢂ 20 cm3) to
remove unreacted substrates. Afterward, it was dissolved in hot
ethanol/water (1:1, 15 cm3 total) mixture. The resulting solution
was acidified to pH ¼ 1.5. The product precipitated by reducing the
volume of the solution by half. It was filtered, washed with ethanol/
water (1:1) mixture followed by hot ethanol (10 cm3). Finally, 1.77 g
of the product was obtained. Yield: 33%. 1H NMR (D2O with
General method for preparation of EHBG complexes
0.44 mmol of EHBG ligand in 1.85 cm3 of ethanol was intro-
duced into a round-bottomed flask. The suspension was refluxed
and aqueous solution of FeCl3 (137 mg, 0.84 mmol; 1.85 cm3 H2O)
was added dropwise. A clear red-violet solution was formed and
reflux was continued for another 30 min. On cooling, an aqueous
solution of NaOH (80 mg, 2 mmol; 1.5 cm3 H2O) was added and
the mixture was kept stirred for another 30 min. All volatiles
were removed in a rotary evaporator and the residue was purified
on a chromatography column using methanol:chloroform (2:1)
eluent. Appropriate yields and spectroscopic data are given
below.
Na2CO3): d 2.17 (s, 6H, NHCOCH3), 2.71 (s, 4H, NeCH2eCH2eN), 3.23
(s, 4H, NeCH2eCOOH), 3.63 (s, 4H, NeCH2eAr), 6.74 (d, 2H,
J ¼ 8.6 Hz, Ar), 6.93 (d, 2H, J ¼ 2.5 Hz, Ar), 7.19 (dd, 2H, J ¼ 8.6,
J ¼ 2.5 Hz, Ar); 13C NMR (D2O with Na2CO3):
d 22.5 (NHCOCH3), 48.8
Fe-EHBG-OMe (2a)
(NeCH2eCH2eN), 55.9 (NeCH2eAr), 57.4 (CH2eCOOH), 116.3 (CAr),
122.7 (CAr),123.5 (CAr),124.1 (CAr),128.0 (CAr),154.8 (CAreOH),172.4
(NHCOCH3), 177.8 (COOH).
Complex was purified on preparative TLC silica plate. Yieldꢀ46%.
IR (cmꢀ1): 1256 s (nCeO); 1610vs (neꢀCOO, asym.); 1378 m (neCOO
,
sym); 1225÷1044 sem (nCeN); 2923w (nCeH); 1465 m (dCeH); 1152 s
EHBG-Me2 (1c)
(nCeOeC); 859w, 806vs, 720 s (1,2,4-trisubstituted benzene ring).
HRMS [MꢀNa]ꢀ: for C22H24N2O8Fe calculated: 500.0872, found:
500.0872, EA for C22H24N2O8FeNa calculated C 50.50, H, 4.62 N
5.35, found C 50.55, H, 4.60 N 5.46.
N,N0-Ethylenediaminodiacetic acid (EDDA) 1.5 g (8.5 mmol),
methanol (5 cm3), sodium hydroxide (2 cm3, 30% aqueous soln.)
and 4.2 g (34 mmol) 2,3-dimethylphenol dissolved in 4 cm3
methanol were introduced into a round-bottomed flask equipped
with condenser, pH-electrode and a dropping funnel. pH was
adjusted to 8.2 with 10% HCl. Following heat-up, 2.5 cm3 (93 mmol)
of 37% formaldehyde in 8.2 cm3 methanol was added dropwise.
Then, the mixture was refluxed for 6 h followed by alkalization to
pH ¼ 9.2 with 10% sodium hydroxide. Solvents were evaporated in
vacuo and the remaining reddish solid was washed with acetone
(2 ꢂ 20 cm3). The crude, solid sodium salt of the product was dis-
solved in water and acidified to pH ¼ 3.0, which resulted in pre-
cipitation of the desired product. After filtration and drying, 717 mg
(1.6 mmol) of the product was collected. Yield: 19%. 1H NMR (D2O
Fe-EHBG-NHAc (2b)
Complex was purified on preparative TLC silica plate. Yieldꢀ22%.
IR (cmꢀ1): 1263vs (nCeO); 1613 s (neꢀCOO, asym.); 1371 m (neCOO
,
sym); 3282w (nNeH); 1540 m (dNeH); 1613 s (nC¼O); 1101÷1072 m
(
nCeN); 883 m, 821 s and 725 m (1,2,4-trisubstituted benzene ring).
HRMS [MꢀNa]ꢀ: for C24H26N4O8Fe calculated 554.1100, found
554.1087. EA for C24H26N4O8FeNa calculated C 49.93H 4.54 N 9.70,
found C 50.01H 4.48 N 9.82.
Fe-EHBG-Me2 (2c)
with Na2CO3):
3.17 (s, 4H, NeCH2eAr), 3.50 (s, 4H, eNeCH2eCOO), 6.57 (d, 2H,
Ar), 6.58 (d, 2H, Ar); 13C NMR (d6-DMSO):
12.9 (AreCH3), 20.7
(AreCH3), 49.3 (NeCH2eCH2eN), 53.3 (NeCH2eAr), 54.7
(NeCH2eCOOH), 122.0 (CAr), 124.7 (CAr), 129.5 (CAr), 139.6 (CAr),
154.6 (CAreOH), 172.0 (COOH).
d 2.02 (s, 6H, eCH3), 2.58 (s, 4H, NeCH2eCH2eN),
Complex was purified on silica gel chromatography column with
methanol/chloroform (2:3) eluent and then purified additionally on
preparative TLC silica plate. Yield 69%. IR (cmꢀ1): 1259 m (nCeO);
1604vs (nꢀeCOO, asym.); 1399 s (neꢀCOO, sym); 1229÷1089 sem (nCeN);
2918w (nCeH); 1455 m (dCeH); 799 s (1,2,3,4-tetrasubstituted
d

ꢀ
N. Kuznik et al. / Journal of Organometallic Chemistry 769 (2014) 100e105
105
benzene ring). HRMS [MꢀNa]ꢀ: for C24H28N2O6Fe calculated
References
496.1169, found 496.1285. EA for C24H28N2O6FeNa calculated C
[1] J.A. McCleverty, T.J. Meyer, A.B.P. Lever, Comprehensive Coordination Chem-
istry II: from Biology to Nanotechnology, Elsevier, 2004.
55.51H 5.43 N 5.39, found C 55.54H 5.35 N 5.48.
[2] B. Yang, J. Feng, Y. Li, F. Gao, Y. Zhao, J. Wang, J. Inorg. Biochem. 96 (2003) 416.
[3] A.D. Tinoco, C.D. Incarvito, A.M. Valentine, J. Am. Chem. Soc. 129 (2007) 3444.
[4] B.G. Jenkins, E. Armstrong, R.B. Lauffer, Magn. Reson. Med. 17 (1991) 164.
[5] O.V. Krokhin, O.V. Kuzina, H. Hoshino, O.A. Shpigun, T. Yotsuyanagi,
J. Chromatogr. A 890 (2000) 363.
Fe-EHBG-Ph (2d)
Complex was purified on silica gel chromatography column with
[6] G.M. Brittenham, Alcohol 30 (2003) 151.
methanol/chloroform (1:2) eluent. Yield 92%. IR (cmꢀ1): 1201 m
[7] B.A. Hoener, B.L. Engelstad, E.C. Ramos, H.A. Macapinlac, D.C. Price,
T.R. Johnson, D.L. White, J. Magn. Reson. Imaging 1 (1991) 357.
(
(
nCeO); 1602 s (neCOOe, asym.); 1450 m (neCOOe, sym.); 3268 m
nNeH); 1201÷1038 mew (nCeN); 2924w (nCeH); 894 m, 830 s and
ꢀ
[8] N. Kuznik, P. Jewuła, L. Oczek, S. Kozłowicz, A. Grucela, W. Domagała, Acta
Chim. Slov. 61 (2014) 87.
[9] S.K. Larsen, B.G. Jenkins, N.G. Memon, R.B. Lauffer, Inorg. Chem. 29 (1990)
1147.
[10] P. Yerram, G. Saab, P.R. Karuparthi, M.R. Hayden, R. Khanna, Clin. J. Am. Soc.
Nephrol. 2 (2007) 258.
[11] T. Grobner, Nephrol. Dial. Transplant 21 (2006) 1104.
[12] D.B. Haydock, T.P.C. Mulholland, J. Chem. Soc. C (1971) 2389.
[13] A.E. Martell, R.J. Motekaitis, E.T. Clarke, J. Harrison, Can. J. Chem. 64 (1986)
449.
696 s (1,2,4-trisubstituted benzene ring). HRMS [MꢀNa]ꢀ: for
C
C
32H28N2O6Fe calculated 592.1297, found 592.1290. EA for
32H28N2O6FeNa calculated C 62.45H 4.59 N 4.55, found C 62.56H
4.50 N 4.70.
Acknowledgments
[14] G. Wilson, Aust. J. Chem. 41 (1988) 173.
[15] N. Bailey, D. Cummins, E. McKenzie, J. Worthington, Inorg. Chim. Acta 50
(1981) 111.
[16] M. Galanski, B.K. Keppler, B. Nuber, Angew. Chem. Int. Ed. 34 (1995) 1103.
[17] A.E. Frost, H.H. Freedman, S.J. Westerback, A.E. Martell, J. Am. Chem. Soc. 80
(1958) 530.
[18] M.G. Patch, K.P. Simolo, C.J. Carrano, Inorg. Chem. 22 (1983) 2630.
[19] D.L. Rubin, H.H. Muller, S.W. Young, Magn. Reson. Med. 23 (1992) 154.
[20] S.H. Koenig, C.M. Baglin, R.D. Brown, Magn. Reson. Med. 2 (1985) 283.
[21] R. Viswanathan, M. Palaniandavar, T. Balasubramanian, P.T. Muthiah, J. Chem.
Soc. Dalton Trans. (1996) 2519.
This work has been partially supported by grant no. N N204
030935 of the Polish Ministry of Science and Higher Education. G.
Szafraniec-Gorol acknowledges the DoktoRIS - Scholarship pro-
gram for innovative Silesia, co-financed by the European Union
within the European Social Fund programme.
Appendix A. Supplementary data
[22] M. Scarpellini, A. Casellato, A.J. Bortoluzzi, I. Vencato, A.S. Mangrich, A. Neves,
S.P. Machado, J. Braz. Chem. Soc. 17 (2006) 1617.
[23] A.M. Samuni, M. Afeworki, W. Stein, A.T. Yordanov, W. DeGraff, M.C. Krishna,
J.B. Mitchell, M.W. Brechbiel, Free Radic. Biol. Med. 30 (2001) 170.
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.jorganchem.2014.07.011.