Synthesis of two 3,5-disubstituted sulfonamide catechol ligands and evaluation of their iron(III) complexes for use as MRI contrast agents

Article abstract of DOI:10.1021/jm0501984

Two 3,5-disubstituted sulfonamide catechol ligands were synthesized. Tris(ligand) iron(III) complexes were prepared and investigated as MRI contrast agents. Longitudinal relaxivity (r1) values were determined for the complexes. The r1 values in water were

Full text of DOI:10.1021/jm0501984

7482
J. Med. Chem. 2005, 48, 7482-7485
Synthesis of Two 3,5-Disubstituted Sulfonamide Catechol Ligands and
Evaluation of Their Iron(III) Complexes for Use as MRI Contrast Agents
Daniel D. Schwert,*,† Nicholas Richardson,‡ Gongjun Ji,† Bernd Radu¨chel,§ Wolfgang Ebert,§ Peter E. Heffner,†
Rick Keck, and Julian A. Davies†
Department of Chemistry, University of Toledo, 2801 W. Bancroft Street, Toledo, Ohio 43606, Department of Chemistry and
Physics, Wagner College, Staten Island, New York 10301, Research Laboratories of Schering AG, 13342 Berlin, Germany, and
Department of Urology, Medical College of Ohio, Toledo, Ohio 43614
Received March 3, 2005
Two 3,5-disubstituted sulfonamide catechol ligands were synthesized. Tris(ligand) iron(III)
complexes were prepared and investigated as MRI contrast agents. Longitudinal relaxivity
(r1) values were determined for the complexes. The r1 values in water were substantially higher
than those of typical six-coordinate iron(III) complexes. The r1 values in plasma under the
same conditions increased. The iron(III) complexes were administered to rats, and the kidney
and liver signal intensities were measured by T1-weighted MR imaging experiments.
Introduction
(negative contrast). Currently one manganese(II)- and
four gadolinium(III)-based T1 (positive) contrast agents
are available in the United States for clinical imaging.
No iron(III)-based positive (but Feridex I.V. as a T2
agent) contrast agents are available.5
In part the lack of iron(III)-based contrast agents can
be attributed to the accessibility of the iron(II)/iron(III)
redox couple to physiological systems, leading to the
need for coordinatively saturated complexes. This elimi-
nates the possibility of inner-sphere coordination of
water molecules, potentially limiting the effectiveness
of the contrast agents.
Iron(III)-tris(tironate) was investigated as a contrast
agent for MRI.6 This compound was intended to maxi-
mize second-sphere interactions with water molecules
through hydrogen bonding to the oxygen atoms of the
Fe-O-R linkages while maintaining the coordinative
saturation of the iron center. The longitudinal (T1)
relaxivity (r1) value of the iron(III)-tris(tironate) far
exceeded the values reported for typical six-coordinate,
coordinatively saturated iron(III) complexes, suggesting
that second-sphere coordination of water may indeed
be important with this contrast agent. However, the
complex was found to be highly toxic, possibly due to
the inherent incompatibility of the nine-minus charge
carried by the contrast agent with biological systems.
This toxicity was not believed to be due to the release
of iron(III) from the complex as the intact metal-ligand
complex was detected in the urine of the animals
injected. In addition, this charge necessitates nine
counterions per contrast agent anion, increasing the
osmolality of the injection solution.
In an attempt to remedy this situation, two 3,5-
disubstituted catecholate derivatives have been pre-
pared. The iron(III)-tris(catecholate) complexes of these
ligands each carry a three-minus charge, reducing the
osmotic stress caused by the injection solution. The r1
values of these complexes have been determined in
water and plasma at 0.47 T (20 MHz) and 37 °C (Bruker
Minispec pc 20), and the kidney and liver SI enhance-
ments obtained by T1-weighted MR imaging experi-
ments (GE Signa 1.5 T) on rats injected with the agents
Magnetic resonance imaging (MRI) has received much
attention over the past two decades as a technique for
the diagnosis of disease states and to test organ function
in the body.1 The signal intensity (SI) from different
tissues varies with the density of protons in the tissues
and both the longitudinal (T1) and transverse (T2)
relaxation times of those protons. While the inherent
image contrast of different body tissues is sufficient in
some cases, it may be necessary to use a contrast agent
to improve the quality of the image and aid in diagnosis.
Contrast agents act by changing the relaxation times
and/or the proton density within a tissue so that the
objective tissue may be differentiated from adjacent
tissues.
Paramagnetic metal ions have been investigated as
MRI contrast agents because they change the T1 and
T2 relaxation times of water protons.2 Due to concerns
about toxicity of metal aquo ions, chelating ligands are
often employed in contrast agent development. In ad-
dition to decreasing the toxicity of the metal ions,
ligands may be functionalized in order to provide tissue
specificity. Unfortunately, as the ligands occupy coor-
dination sites on the metal ion the number of water
molecules directly bound to the metal center decreases
and therefore the overall effectiveness of the contrast
agent decreases.
Paramagnetic gadolinium(III),3 manganese(II),4 and
iron(III) have received the most attention in MRI
contrast agent research. When used in T1-weighted
MRI, these agents act by increasing the SI of the target
organs and tissues (positive contrast).2 Other agents,
such as superparamagnetic iron oxides (SPIO) and
perfluorocarbon solutions, act by decreasing the SI
* To whom correspondence should be addressed. Currently at
Division of Chemistry, Indiana University Southeast, 4201 Grant Line
Rd., New Albany, IN 47150-6405. Phone: 812-941-2462. E-mail:
dschwert@ius.edu.
† University of Toledo.
‡ Wagner College.
§ Schering AG.
Medical College of Ohio.
10.1021/jm0501984 CCC: $30.25 © 2005 American Chemical Society
Published on Web 10/20/2005

Brief Articles
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23 7483
Figure 1. Structure of Gd-EOB-DTPA.
Scheme 1. Synthesis of Disubstituted Sulfonamide
Catechols
Table 1. Relaxivity Data for Iron(III)-Based Contrast Agent
Complexes and Gd-EOB-DTPA
r1 (H2O)
r1 (plasma)
compound
(mmol-1 s-1
)
(mmol-1 s-1
)
Fe(2)3
1.9
2.0
5.3
4.0
3.0
8.7
Fe(3)3
Gd-EOB-DTPAa
a From ref 7.
were investigated. These results were compared with
those obtained in MR imaging experiments with gado-
linium-(4S)-4-(4-ethoxybenzyl)-3,6,9-tris(carboxylato-
methyl)-3,6,9-triazaundecanoic acid, disodium salt (Gd-
EOB-DTPA, Primovist, Figure 1), a liver-specific con-
trast agent in the late phase of clinical development.7,8
Figure 2. MRI scans of rats pre and postinjection of Fe(2)3.
(a) Pre- (left) and 4 min postinjection (right) images showing
pronounced enhancement (168%) of the kidney (arrow). (b) Pre-
(left) and 4 min postinjection (right) images showing pro-
nounced (53.0%) enhancement of the liver (arrow). Images are
adjacent 3 mm slices separated by a 1.5 mm intersection gap.
Results and Discussion
molecule bound directly to the metal center. In plasma,
Gd-EOB-DTPA also had a higher r1 value than the
iron(III) complexes but the percentage increases in r1
values was greater for one of the iron(III)-based contrast
agents. The r1 increase for Gd-EOB-DTPA was 64%
while the relaxivities of Fe(2)3 and Fe(3)3 increased by
110% and 50%, respectively. Part of the increase in r1
is due to the higher viscosity of plasma. The rest of the
relaxivity increase is likely due to some binding of the
contrast agents to plasma proteins.
Animal Experiments. Female Fischer 344 rats were
injected intravenously with solutions of contrast agent
and the liver and kidney SI postinjection were measured
and compared with the preinjection values. An example
of a typical animal scan for the iron(III) complexes is
shown in Figure 2. A similar scan for an animal injected
with Gd-EOB-DTPA has been previously published.10
The injected solutions had a concentration of 45.0 mM
and the applied dosage was 0.1 mmol Fe per kg body
weight (bw). Slight variations in the weights of the rats
resulted in marginally different final injection concen-
trations.
Synthesis. The synthesis of compound 1 was adapted
from a previous published procedure.9 Two different
disubstituted sulfonamide catechol ligands (H22 and
H23, Scheme 1) were synthesized for use as chelating
agents for iron(III) as potential MRI contrast agents.
Alkyl chain length was varied to determine the effect
of such variation on the in vitro relaxivity values and
on the in vivo SI enhancements of the iron(III) com-
plexes (Na3Fe(2)3 and Na3Fe(3)3) of these ligands. Two
synthetic routes were used to generate the metal
complexes. For relaxivity experiments, the trisodium
salts of the tris(ligand) complexes of iron(III) were
synthesized using Fe(acac)3 as the source of iron(III),
and the complexes were isolated. For the animal experi-
ments, the metal complexes were generated in situ by
the addition of 1 mol equiv of FeCl3 to 3 mol equiv of
ligand in aqueous solution, and the pH of the solution
was adjusted to 7.2 with sodium hydroxide.
Relaxivity Experiments. The results of the relax-
ivity measurements are shown in Table 1. The mea-
sured values of r1 for the iron(III) complexes are higher
than those reported for typical six-coordinate, coordi-
natively saturated iron(III) complexes.5 As expected, the
iron(III) catecholate complexes have lower r1 values
than Gd-EOB-DTPA due to the higher number of
unpaired electrons on gadolinium(III) and the one water
The results of the experiments with 0.1 mmol/kg bw
contrast agent solution are shown in Figure 3 and
Figure 4. Note that the error bars represent the
standard deviation values of the voxel intensities and
are not a representation of experimental error but
rather a measure of the interindividual differences

7484 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23
Brief Articles
enhancements by Fe(2)3 (P > 0.7). This result is impres-
sive considering the large differences in in vitro r1
values between Fe(2)3 and Gd-EOB-DTPA. The other
iron(III)-based contrast agent, Fe(3)3, provided signifi-
cantly (P < 0.001) lower liver SI enhancement than Gd-
EOB-DTPA.
Conclusions
Two new catechol-based ligands were synthesized,
complexed with iron(III), and investigated as MRI
contrast agents. The r1 values of the complexes in water
were found to be greater than those of typical six-
coordinate, coordinatively saturated iron(III) complexes.
The iron(III) complexes showed increases in r1 value of
at least 50% when measured in plasma. Investigations
are currently underway to determine the protein bind-
ing of the complexes in order to determine the cause of
the increase in r1 values. The iron(III) complexes
injected into animals at 0.1 mmol/kg bw showed greater
kidney SI enhancements than Gd-EOB-DTPA, a
promising liver-specific contrast agent. One of the iron-
(III) complexes, Fe(2)3, provided liver SI enhancements
comparable to Gd-EOB-DTPA when injected at 0.1
mmol/kg bw.
Figure 3. SI enhancements of the kidneys after injection of
Fe(2)3 (circle, eight rats, 0.101 ( 0.006 mmol/kg), Fe(3)3
(square, four rats, 0.100 ( 0.008 mmol/kg), and Gd-EOB-
DTPA (triangle, five rats, 0.10 ( 0.01 mmol/kg). Error bars
represent the standard deviation of the data.
Experimental Section
All reagents (each above 97% purity except for dimethyl-
amine [40% in water]) were purchased from Aldrich Chemical
Co. (Milwaukee, WI) and were used without further purifica-
tion. All 1H and 13C{1H} NMR data were recorded on a Varian
VXR-400 spectrometer. Tetramethylsilane and 3-(trimethyl-
silyl)propionic-2,2,3,3-d4 acid were used as internal standards
in organic and aqueous samples, respectively. Elemental
analysis was performed by the Instrumentation Center at the
University of Toledo.
3,5-Bis(dichlorosulfonyl)catechol (1). The synthesis of
compound 1 was adapted from a previously published proce-
dure. A 1000-mL, three-necked, round-bottomed flask was
equipped with a nitrogen inlet, a thermometer, and a magnetic
stir bar. Chlorosulfonic acid (100 mL, 1.50 mol) was added to
the flask and heated to 110 °C under nitrogen. Ten grams of
catechol (0.0908 mol) was slowly added to the flask. The
resulting solution was heated for 90 min and then cooled to
room temperature. The nitrogen inlet was removed, and a
water condenser was attached. The flask was placed in a dry
ice/acetone/water bath and cooled to below -10 °C. Concen-
trated hydrochloric acid (100 mL) was slowly added, keeping
the temperature below 0 °C at all times. After all of the acid
was added, 300 mL of diethyl ether was added. The organic
layer was collected, and the aqueous layer was washed with
diethyl ether (10 × 100 mL). The ether solution was dried over
sodium sulfate and filtered, and then the solvent was removed
by rotary evaporation. An orange/brown oil was recovered
(10.4, 0.0339 mol, 37% yield). 1H NMR in acetone-d6 (ppm):
7.8 (d, 1H, ArH), 8.0 (d, 1H, ArH). 13C{1H} NMR in acetone-
d6 (ppm): 118.4, 120.1, 130.5, 134.6, 148.7, 152.7 (C6H2).
3,5-Bis(diethylsulfonamide)catechol (H22). A 1000-mL
round-bottomed flask was equipped with a Claisen adapter,
reflux condenser, and an addition funnel. Diethylamine (42.4
mL, 0.410 mol) was dissolved in 100 mL of diethyl ether and
cooled to 0 °C in an ice bath. Compound 1 (10.0 g, 0.0326 mol),
dissolved in 100 mL of diethyl ether, was added slowly. The
reaction was allowed to come to room temperature and stirred
overnight. The ether was removed by rotary evaporation.
Concentrated HCl (80 mL) was added to the resulting solid.
The aqueous layer was washed with CH2Cl2 (6 × 150 mL).
The CH2Cl2 solution was dried over sodium sulfate and
filtered, and the solvent was removed by rotary evaporation
leaving a black solid. The black solid was extracted with hot
Figure 4. SI enhancements of the liver after injection of Fe-
(2)3 (circle, eight rats, 0.101 ( 0.006 mmol/kg), Fe(3)3 (square,
four rats, 0.100 ( 0.008 mmol/kg), and Gd-EOB-DTPA
(triangle, five rats, 0.10 ( 0.01 mmol/kg). Error bars represent
the standard deviation of the data.
between animals. Of the two iron(III) complexes, it was
found that Fe(2)3 provided significantly (P < 0.05)
higher liver SI enhancement and greater kidney SI
enhancement over the course of the experiments. The
levels of SI enhancements for both the liver and the
kidney after injection of Fe(2)3 remained relatively
constant over the time course of the experiments.
Removal of one methylene group on the alkyl side chains
(resulting in Fe(3)3) resulted in lower kidney and liver
image SI enhancements and a faster decrease in the
levels of SI. This is possibly due to the rapid clearance
of the contrast agent from the bloodstream thus an early
peak in SI enhancement may be missed.
In both cases, the iron(III)-based contrast agents
provided evidence of greater kidney SI enhancements
as compared to Gd-EOB-DTPA at 0.1 mmol/kg bw
over the course of the MRI scans (for Fe(2)3 and Gd-
EOB-DTPA, P < 0.001; for Fe(3)3 and Gd-EOB-
DTPA, P < 0.10). The liver SI enhancements with Gd-
EOB-DTPA were found to be comparable to the

Brief Articles
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23 7485
hexanes (10 × 1000 mL). The combined hexanes extracts were
cooled in the freezer. The product precipitated as white crystals
(3.3 g, 0.0087 mol, 26% yield). 1H NMR in CDCl3 (ppm): 1.10
(t, 6H, CH3), 1.13 (t, 6H, CH3), 3.21 (q, 4H, CH2), 3.38 (q, 4H,
CH2), 7.4 (d, 1H, ArH), 7.6 (d, 1H, ArH). 13C{1H} NMR in
CDCl3 (ppm): 14.0 (CH3), 14.3 (CH3), 42.3 (CH2), 42.4 (CH2),
116.9, 118.1, 123.7, 133.0, 145.7, 146.5 (C6H2). Anal. (C14H24-
N2O6S2) C, H, N.
3,5-Bis(dimethylsulfonamide)catechol (H23). A 1000-
mL round-bottomed flask was equipped with a Claisen adapter,
reflux condenser, and an addition funnel. Dimethylamine (52
mL, 40% in water, 0.410 mol) was dissolved in 100 mL of
diethyl ether and cooled to 0 °C in an ice bath. Compound 1
(10.0 g, 0.0326 mol) dissolved in 100 mL of diethyl ether was
added slowly. The reaction was allowed to come to room
temperature and stirred overnight. The ether was removed
by rotary evaporation. Concentrated HCl (80 mL) was added
to the resulting solid. The aqueous layer was washed with CH2-
Cl2 (6 × 150 mL). The CH2Cl2 solution was dried over sodium
sulfate and filtered, and the solvent was removed by rotary
evaporation. The product was recrystallized using CH2Cl2/
hexanes (1.1 g, 0.0034 mol, 10% yield). 1H NMR in CDCl3
(ppm): 2.7 (s, 6H, CH3), 2.8 (s, 6H, CH3), 7.4 (d, 1H, ArH), 7.5
(d, 1H, ArH). 13C{1H} NMR in CDCl3 (ppm): 37.7 (CH3), 37.9
(CH3), 118.0, 119.1, 119.6, 127.7, 146.6, 146.7 (C6H2). Anal.
(C10H16N2O6S2) C, H, N.
Preparation of Iron(III) Complexes for Relaxivity
Experiments. For relaxivity experiments, Fe(2)3 and Fe(3)2
were isolated as trisodium salts. The appropriate ligand (3
mmol) was dissolved in water. Sodium hydroxide (0.0800 g, 2
mmol) and Fe(acac)3 (0.3532 g, 1 mmol) were added, and the
solution was refluxed for 24 h. The solution was allowed to
cool and the pH adjusted with NaOH to pH 7 and extracted
with diethyl ether (6 × 150 mL). The solution was filtered,
and the solvent was removed by lyophilization. Anal. (C52H80N6-
FeNa3O25S6 for 2, C40H66N6FeNa3O25S6 for 3) C, H, N.
Preparation of MRI Injection Solutions. For a 45.0-mM
injection solution, compound H22 (0.257 g, 0.675 mol) or
compound H23 (0.219 g, 0.675 mmol) and NaOH (0.0540 g, 1.35
mmol) were dissolved in 3 mL of water. Ferric chloride
hexahydrate (0.0608 g, 0.225 mol) was added with stirring.
The pH was adjusted to 7.2 using NaOH, and the volume was
adjusted to 5 mL. Injection solutions of Gd-EOB-DTPA at a
concentration of 45.0 mM were made by the dissolution of the
disodium salt of Gd-EOB-DTPA (Schering, AG, Berlin,
Germany) in water.
Five postinjection scans were then acquired using the same
scan parameters as the pre-injection scans. The postinjection
scans were separated by an approximately five-second delay
between scans.
SI measurements were made in operator-defined regions of
interested (ROI) on the pre- and postcontrast images. For each
animal imaged, three or four liver image slices were analyzed
and one image slice was analyzed for each kidney. The ROI
data were then converted to percent enhancement values (eq
1). The percent enhancement values for the liver and kidney
image slices were then averaged and standard deviation values
were determined. The significance of the differences in the data
was determined using Student’s t test (95% confidence inter-
val). The data were plotted against time, with the completion
of the injection being time ) 0 min. The time of completion of
the scans (approximately 4, 8, 12, 16, and 20 min) was used
as the time of the image, even though the image is actually
an average of the data acquired over the time of the scan.
ROIpostinj - ROIpreinj
percent enhancement )
× 100% (1)
ROIpreinj
Acknowledgment. We thank Carol Okenka, Tina
Kinkaid, and Kathy Sbrocchi for their assistance in the
animal experiments. Financial support from Schering
AG is acknowledged. We thank Amy Eshleman for
assistance with the statistical analyses.
Supporting Information Available: A table listing the
elemental analysis for the target compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1) Rinck, P. A. Magnetic Resonance in Medicine; 3rd ed.; Blackwell
Scientific: Oxford, 1993; pp 155-174.
(2) Lauffer, R. B. Paramagnetic metal complexes as water proton
relaxation agents for NMR imaging: theory and design. Chem.
Rev. 1987, 87, 901-927.
(3) Caravan, P.; Ellison, J. J.; McMurry, T. J.; Lauffer, R. B.
Gadolinium(III) chelates as MRI contrast agents: structure,
dynamics, and applications. Chem. Rev. 1999, 99, 2293-2352.
(4) Schwert, D. D.; Davies, J. A.; Richardson, N. Non-gadolinium-
based MRI contrast agents. In Topics in Current Chemistry;
Krause, W., Ed.; Springer: New York, 2001; Vol. 221, pp 165-
200.
(5) Richardson, N.; Davies, J. A.; Radu¨chel, B. Iron(III)-based
contrast agents for magnetic resonance imaging. Polyhedron
1999, 18, 2457-2482.
Relaxivity Experiments. The prepared lyophilized samples,
Fe(2)3 and Fe(3)3 were dissolved in either water or plasma to
give samples at three concentrations (0.25, 0.5, and 1.0 mmol
Fe/L). The T1 relaxation rate of each sample was measured
by use of a Minispec P20 (Bruker, Rheinstetten, Germany) at
20 MHz (0.47 T) and 37 °C. Linear regression analysis of
relaxation rates R1 ) 1/T1 versus concentration gave the
relaxivity values (r1) of each complex in water and plasma.
Animal Experiments. Animal experiments were per-
formed on a Signa 1.5-T superconducting MR scanner (GE
Medical Systems, Milwaukee, WI). Two Fisher 344 rats were
anesthetized with 65 mg/kg bodyweight of sodium pentobar-
bital and placed in a send-and-receive head coil. A localizer
image was obtained followed by a sagittal preinjection T1-
weighted spin-echo (SE) pulse sequence with TR ) 300 ms
and TE ) 20 ms. The matrix was 256 × 256 over a 20-cm field
of view. Three signals were acquired of 3-mm slices with a
1.5-mm intersection gap. The total scan time was 3 min and
56 s. The contrast agent (0.5 mL per animal) was injected
through a 27-gauge butterfly needle in the tail vein over 2 min
and 15 s followed by a 0.75-1.00 mL saline flush over 45 s.
(6) Davies, J. A.; Dutremez, S. G.; Hockensmith, C. M.; Keck, R.;
Richardson, N.; Selman, S.; Smith, D. A.; Ulmer, C. W., II.;
Wheatley, L. S.; Zeiss, J. Iron-based second-sphere contrast
agents for magnetic resonance imaging: development of a model
system and evaluation of iron(III) tris(tironate) complex in rats.
Acad. Radiol. 1996, 3, 936-945.
(7) Weinmann, H.-J.; Schuhmann-Giampieri, G.; Schmitt-Willich,
H.; Vogler, H.; Frenzel, T.; Gries, H. A new lipophilic gadolinium
chelate as a tissue-specific contrast medium for MRI. Magn.
Reson. Med. 1991, 22, 233-237.
(8) Schmitt-Willich, H.; Brehm, M.; Ewers, C. L. J.; Michl, G.;
Mu¨ller-Fahrnow, A.; Petrov, O.; Platzek, J.; Radu¨chel, B.; Su¨lzle,
D. Synthesis and physicochemical characterization of a new
gadolinium chelate: The liver-specific magnetic resonance imag-
ing contrast agent Gd-EOB-DTPA. Inorg. Chem. 1999, 38, 1134-
1144.
(9) Pollak, J.; Gebauer-Fulnegg, E. U¨ ber die einwirkung von clor-
sulfonsa¨re auf phenole. Monatsch. Chem. 1926, 47, 109-118.
(10) Schuhmann-Giampieri, G.; Schmitt-Willich, H.; Press: W.-R.;
Negishi, C.; Weinmann, H.-J.; Speck, U. Preclinical evaluation
of Gd-EOB-DTPA as a contrast agent in MR imaging of the
hepatobiliary system. Radiology 1992, 183, (1), 59-64.
JM0501984