A new bis(3-hydroxy-4-pyridinone)-IDA derivative as a potential therapeutic chelating agent. Synthesis, metal-complexation and biological assays

Article abstract of DOI:10.1039/b409357g

A new bis(3-hydroxy-4-pyridinone) derivative of iminodiacetic acid, imino-bis(acetyl(1-(3′-aminopropyl)-3-hydroxy2-methyl-4-pyridinone)), IDAPr(3,4-HP)₂, has been prepared and studied in its interaction with a set of hard metal ions. This tetradentate ligand presents a much higher chelating efficiency for trivalent hard metal ions (Fe, Ga, Al) than the monodentate derivative Deferriprone, namely at the diluted conditions prevailing in physiological conditions and at low clinical doses. A similar behaviour was also observed for the complexation with Zn(II) but at a significantly lower extent. This compound presents a moderate hydrophilic character at physiological pH (log D = -1.72). In vivo assays showed much more rapid clearance of ⁶⁷Ga from most tissues of metal-loaded mice than the drug Deferriprone and the radioactivity excretion occurs mostly through the kidneys. Therefore, results from in vitro and in vivo studies indicated good perspectives for this compound to be a potential decorporating agent for hard metal ions in overload situations without depletion of essential metal ions such as zinc.

Full text of DOI:10.1039/b409357g

View Article Online / Journal Homepage / Table of Contents for this issue
F U L L P A P E R
A new bis(3-hydroxy-4-pyridinone)-IDA derivative as a potential
therapeutic chelating agent. Synthesis, metal-complexation and
biological assays
M. Amélia Santos,*a Sofia Gama,a Lurdes Gano,b Guilhermina Cantinhoc and Etelka Farkasd
a
Centro de Química Estrutural, Instituto Superior Técnico, Complexo I, 1049-001 Lisboa,
Portugal. E-mail: masantos@ist.utl.pt; Fax: +351 21 846 44 55; Tel: +351 218419273
Instituto Tecnológico e Nuclear, Estrada Nacional N° 10, 2686-953 Sacavém, Portugal
Instituto de Medicina Nuclear, Faculdade de Medicina de Lisboa, R. Egas Moniz,
b
c
1600 Lisboa, Portugal
Department of Inorganic and Analytical Chemistry, University of Debrecen,
d
H-4010 Debrecen, Hungary
Received 21st June 2004, Accepted 24th September 2004
First published as an Advance Article on the web 12th October 2004
A new bis(3-hydroxy-4-pyridinone) derivative of iminodiacetic acid, imino-bis(acetyl(1-(3-aminopropyl)-3-hydroxy-
2-methyl-4-pyridinone)), IDAPr(3,4-HP)2, has been prepared and studied in its interaction with a set of hard metal
ions. This tetradentate ligand presents a much higher chelating efficiency for trivalent hard metal ions (Fe, Ga,
Al) than the monodentate derivative Deferriprone, namely at the diluted conditions prevailing in physiological
conditions and at low clinical doses. A similar behaviour was also observed for the complexation with Zn(II) but
at a significantly lower extent. This compound presents a moderate hydrophilic character at physiological pH
(logD = −1.72). In vivo assays showed much more rapid clearance of 67Ga from most tissues of metal-loaded mice
than the drug Deferriprone and the radioactivity excretion occurs mostly through the kidneys. Therefore, results
from in vitro and in vivo studies indicated good perspectives for this compound to be a potential decorporating agent
for hard metal ions in overload situations without depletion of essential metal ions such as zinc.
chelating therapy of hard metal ions, namely because they can
also efficiently bind to other essential metal ions, notably zinc
and calcium, causing their depletion in vivo.
Introduction
The relevance of specific metal chelators in medicine has
increased in recent years, namely compounds with high affinity
As part of an ongoing project aimed at exploring the poly-
towards “hard” M3+ metal ions such as iron and the group 13
denticity of 3,4-HP derivatives, in addition to the development
metals (aluminium, gallium and indium), due to their potential
of tris-hydroxypyridinonates (hexadendate ligands),9 we sought
application, either for controlling situations of metal toxicity in
to explore tetradentate bis-hydroxypyridinone derivatives,
the body (e.g. Fe, Al) or for clinical diagnosis and chemotherapy
having two 3,4-HP binding groups appended to various mole-
cular scaffolds with the potential existence of extra-functional
groups for molecular interaction with biological sites. Following
(e.g. 67,68Ga, 111In). The 3-hydroxy-4-pyridinones (3,4-HP),
having a hydroxy group ortho to a ketone functionality, are a
class of hydroxypyridinones with high ability to complex metal
this molecular-designing option, firstly we used a cyclohexane
cations with large charge/ionic radius ratios and they have been
backbone with two 3,4-HP units and one carboxylate group as
strongly suggested as good candidates for medicinal chemistry.1–3
pendant arms [KEMPPr(3,4-HP)2, see Chart 1]. Although this
In particular, 1,2-dimethyl-3-hydroxy-4-pyridinone (3,4-DMHP,
commercially available as Deferriprone) has already been
clinically used as an orally active iron-chelating drug in thalas-
semic patients, in substitution of Desferrioxamine (DFO), the
well established tris-hydroxamate derivative.4 However, one
important limitation to the clinical application of bidentate
metal decorporating agents such as Deferriprone emerges from
the great ligand concentration dependence of their chelating
efficacy as compared to those of multidentate ligands, which
can be used at much lower clinical doses. Therefore, in order
to minimize drug-induced toxicity, considerable efforts have
been made on the design of new ligands with high denticity
and presumably high ability for formation of stable 1:1 (M:L)
complexes with that type of metal ions.1,3,5–8 However, most of
the tris-hydroxypyridinones studied so far are of 3-hydroxy-
2-pyridinone (3,2-HP) type, which are known to present less
affinity for the hard metal ions at the physiological pH and to
involve more laborious methods of synthesis than the corres-
ponding 3,4-HP derivatives. On the other hand, polyamino-
carboxylic compounds, such as ethylenediaminetetracetic acid
(EDTA) and diethylenetriaminepentaacetic acid (DPTA), have
good ability to form 1:1 complex with a large variety of metal
ions and they are zwitterionic species with very good solubility
properties. However, the lack of specificity of these compounds
for metal-complexation limits their clinical applications for
Chart 1
3 7 7 2
D a l t o n T r a n s . , 2 0 0 4 , 3 7 7 2 – 3 7 8 1
T h i s j o u r n a l i s
©
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4

View Article Online
186–189 °C. 1H NMR (CDCl3) d (ppm): 8.14 (1H, d, 6-HPy), 7.42
(5H, s, Ph), 7.11 (1H, d, 5-HPy), 5.13 (2H, s, CH2Ph), 4.31 (2H,
t, CH2NPy), 3.03 (2H, t, CH2NH2), 2.36 (2H, t, CH2CH2CH2),
2.12 (3H, s, CH3). m/z (FAB-MS) 273 (M + 1).
tetradentate ligand showed reasonably good ability to form 1:1
complexes with the set of M2+ and M3+ metal ions,10,11 there was
some apparent strain associated to the metal wrapping by the
hydroxypyridone moieties and also water-solubility limitations.
Therefore, we have decided to change the ligand topology by
choosing another type of molecular scaffold, namely linear
polyaminocarboxylic derivatives.
N-Benzylimino-bis(acetyl(1-(3-aminopropyl)-3-benzyloxy-2-
methyl-4-pyridinone)). Free 1-(3-aminopropyl)-3-benzyloxy-
2-methyl-4-pyridinone was obtained by adding a solution of
KOH (525 mg, 9.36 mmol) in dry methanol (5 mL) to a solu-
tion of 1-(3-aminopropyl)-3-benzyloxy-2-methyl-4-pyridinone
hydrochloride (1.45 g, 4.68 mmol) in dry methanol (20 mL),
under nitrogen, in a water-ice bath. The mixture was allowed
to stir for 15 min and then KCl was filtered out. In a two-neck
round flask, N-benzyliminodiacetic acid (494 mg, 2.21 mmol)
was dissolved in dry THF (50 mL) and the solution was kept
under nitrogen at ca. 0 °C. Then, N-methylmorpholine (495 mg,
4.89 mmol) and ethylchloroformate (579 mg, 5.34 mmol)
were added, the mixture was left under stirring for 2 h and
the N-methylmorpholine hydrochloride salt was filtered out.
To this solution, the free 1-(3-aminopropyl)-3-benzyloxy-
2-methyl-4-pyridinone solution was dropwise added under
stirring and the solution was allowed to stir for 3.5 h keeping
the temperature at ca. 0 °C. The solution was filtered and the
solvent was removed in vacuum. The solid residue was purified
by flash chromatography on a silica-gel column. The isolated
product was recrystallized from dry methanol–ether (g = 45%).
1H NMR (CDCl3) d (ppm): 8.70 (2H, d, 6-HPy), 7.3 (15H, s,
Ph), 6.29 (2H, d, 5-HPy), 5.09 (4H, s, OCH2Ph), 3.80 (4H, t,
NCH2CH2CH2NHCO), 3.67(2H, s, NCH2Ph), 3.20(6H, s, CH3),
2.12–2.02 (4H + 4H, m, CH2NHCO + COCH2NCH2CO), 1.79
(4H, t, CH2CH2CH2). m/z (FAB-MS): 732 (M + 1).
Herein we report the preparation of a new tetradentate
derivative, IDAPr(3,4-HP)2 (see Chart 1) which has the iminodi-
acetic acid (IDA) as molecular scaffold and two 3,4-HP binding
groups attached to its carboxylic groups, thus leaving the amino
group free for further potential interaction with biological
sites or for appending other receptor probes. We examine its
physico-chemical properties in aqueous solution, alone and in
the presence of a set of metal ions, and also results of in vivo
assays. In particular, equilibrium studies in aqueous solution,
mostly based on potentiometric and spectroscopic (UV-
1
Vis, H NMR, 27Al NMR) measurements, are conducted to
investigate the binding affinities towards the proton and a set
of trivalent (Fe, Ga, Al) metal ions, and also Zn(II) due to its
availability in biological systems. Due to the great importance
of the lipo/hydrophilic character of a drug for the assessment
of its bioavailability, the distribution coefficient of IDAPr(3,4-
HP)2 between 1-octanol and a Tris buffered (pH 7.4) aqueous
solution is also calculated. The ability of the new chelating agent
to mobilize 67Ga from radiotracer overloaded mice and the bio-
distribution of the 67Ga-IDAPr(3,4-HP)2 complex is further
assayed to evaluate potential pharmaceutical applications of
this new bis(3-hydroxy-4-pyridinone) compound, in comparison
with Deferriprone.
Experimental
Materials
Imino-bis(acetyl(1-(3-aminopropyl)-3-hydroxy-2-methyl-4-
pyridinone)). To a solution of N-benzylimino-bis(acetyl(1-(3-
aminopropyl)-3-benzyloxi-2-methyl-4-pyridinone)) (837 mg,
1.14 mmol) in dry methanol (10 mL) was added 10% Pd/C
(282 mg) and the mixture was stirred under H2 (1.5 atm) for 4 h,
at room temperature. After filtration, the solvent was evaporated
under reduced pressure and the product (489 mg) was obtained
as a white powder, which was recrystallised from methanol–ether
(g = 93%), mp 95–97 °C. 1H NMR (CDCl3) d (ppm): 7.67 (2H, d,
6-HPy), 6.53(2H, d, 5-HPy), 4.14(4H, t, NCH2CH2CH2NHCO),
3.32 (4H + 4H, m, CH2NHCO + COCH2NHCH2CO), 2.43
(6H, s, CH3), 2.03 (4H, t, CH2CH2CH2). m/z (FAB-MS): 462
(M + 1). Calc. for C22N5H31O6·3H2O·0.8 MeOH: C, 52.38; H,
7.78; N, 12.94. Found: C, 52.60; H, 7.31; N, 12.68%.
Analytical grade reagents were used as supplied. Whenever
necessary, solvents were dried according to standard methods.12
All the chemical reactions were TCL controlled.
Synthesis of the ligand
3-Benzyloxy-2-methyl-4-pyrone. To a solution of 3-hydroxy-
2-methyl-4-pyrone (10.00 g; 79 mmol) containing the equivalent
amount of a 7 M NaOH aqueous solution (12 mL) in methanol
(100 mL), benzyl chloride (10.6 mL; 92 mmol) was dropwise
added and the mixture was left to reflux for 6 h. After cooling,
the reaction mixture was filtered and the filtrate was evaporated
under vacuum. The residual oil was taken in dichloromethane
(50 mL) and washed with 5% NaOH aqueous solution
(5 × 20 mL) and finally with water. The organic solution was
dried with anhydrous sodium sulfate and the solvent was
evaporated to dryness to obtain the pure product as a pale oil
(g = 85%). 1H NMR (CDCl3) d (ppm): 7.57 (1H, d, 6-HPy); 7.32
(5H, s, Ph); 6.33 (1H, d, 5-HPy); 5.14 (2H, s, CH2Ph); 2.06 (s, 3H,
CH3). m/z (FAB-MS): 217 (M + 1).
Potentiometric measurements
The potentiometric measurements were performed with a
Metrohm 6.0234.100 combined electrode and a Dosimat
765 burette controlled by computer, at T = 25 ± 0.1 °C and
I = 0.1 M (KNO3), using a CRISON emf-Meter. Measuring the
emf in the cell, [H+] of the solutions was calculated taking into
account that E = E° + Qlog[H+] + Ej. E° and Q were obtained
daily by calibration of the electrode applying Gran’s method13
to a strong acid/strong base (HNO3/KOH 0.1 M) titration in
the acid range at the same ionic strange, and Ej has a negligible
value at the used experimental conditions. The Kw value used
in the calculations was 10−13.78 and it was also determined from
a strong acid/strong base titration using the alkaline range
data. Atmospheric CO2 was excluded from the system with a
purging stream of N2. The ligand (H2L) concentration in the
sample was 1.75 × 10−3 M. The exact concentration of the
ligand stock solution and the protonation constants was deter-
mined from the fitting analysis of the potentiometric data with
the HYPERQUAD 2000 computer program.14 Complexation
studies were performed for each metal ion by potentiometric
titrations with the same computer program, although in some
cases combined with spectrophotometric measurements (see
below). To study the formation of the species with 1:1 and the
(3-Aminopropyl)-3-benzyloxy-2-methyl-4-pyridinone.
3-Benzyloxy-2-methyl-4-pyrone (9.37 g, 43.59 mmol) and
1,3-diaminopropane (3.8 mL, 45.6 mmol) were left refluxing
(T = 70–75 °C), in a EtOH–water mixture (20/15) mL with
2 M NaOH (4 mL) for 24 h. After cooling, 2 M HCl was
added until ca. pH = 1 and the ethanol was evaporated. To the
remaining residue, water was added (50 mL) and this solution
was extracted with ether (2 × 50 mL). The aqueous phase was
basified with 10 M NaOH until pH  12 and then it extracted
with dichloromethane (5 × 50 mL). The organic solution was
dried over anhydrous sodium sulfate and the solvent was
evaporated to dryness. That residue was taken into methanol
(2 mL). Acidification of that solution until ca. pH 1 with
HCl-saturated methanol gave a white precipitate which was
recrystallized from dry methanol–acetonitrile to give the pure
product as the corresponding hydrochloride salt (g = 40%), mp
D a l t o n T r a n s . , 2 0 0 4 , 3 7 7 2 – 3 7 8 1
3 7 7 3

View Article Online
of injected dose per total organ (% ID/organ). For blood, bone
and muscle, total activity was calculated assuming, as previ-
ously reported,17 that these organs constitute 7, 10 and 40% of
the total weight, respectively. However, the accuracy of these
values is not very important because they are used to calculate
the organ uptakes in all the experimental results which are then
taken for comparison purposes.
2:3 stoichiometries, 1:1 and 1:2 metal ion-to-ligand ratios were
used, respectively. Since the 2:3 species is supposed to be formed
at a higher pH than the 1:1 species, in that case, a ligand excess
was used to suppress hypothetical precipitation due to hydrolytic
species. The metal ions were provided from stock solutions of
the corresponding nitrate salts.
The selection of the equilibrium models was based on the
critical analysis of the weighted residuals, the statistical para-
meters (v2, r)15 and graphical comparisons between the experi-
mental and simulated potentiometric curves.
Other measurements
1H NMR spectra were recorded using a Varian Unity
300 spectrometer at 25 °C. The chemical shifts are reported in
ppm (d) from sodium 3-(trimethylsilyl)-[2,2,3,3-2D4]propionate
(DSS) internal reference in D2O solutions and tetramethylsilane
(TMS) in organic solvents. Some abbreviations are used to
characterize the peaks: d, doublet; s, singlet; t, triplet. Solution
magnetic susceptibilities were measured in D2O solution, as a
function of pD,19 by the Evans method20 (vM, calculated per
molar concentration of iron) and the corresponding magnetic
moments were calculated (lM = 2.84(vMT )1/2,21 for both the
Fe:L stoichiometries, under the same experimental conditions
used for potentiometry, at T = 298 K. Melting points were
measured using a Leica Galen III hot stage apparatus and they
are uncorrected values. Elemental analysis was performed on
a Fisons EA1108 CHNF/O instrument and the mass spectra
results were recorded on a VG TRIO-2000 GC/MS instrument.
Spectrophotometric measurements
Spectral determinations were made with a Perkin-Elmer
Lambda 9 spectrophotometer using 1.0 cm matched quartz cells
at T = 25.0 °C and constant ionic strength conditions. Solutions
of the metal complexes were generated in situ by addition of
a standard metal ion solution to the ligand solution. The pH
values were measured using a Crison emf-Meter equipped with
a Metrohm 6.0234.100 combined electrode. For very acidic
conditions (pH < 2), the stability constants of the correspond-
ing iron(III) and gallium(III) complexes were calculated from the
spectrophotometric data, using the PSEQUAD program,16 and
the fitting parameter accepted was below 2 × 10−2.
Distribution coefficients
The distribution coefficient of the ligand in octanol/water (Tris
buffered solution, pH = 7.4) at ca. 25 °C was determined by
the “shakeflask” method, as previously described17 and based
on established procedures.18 The ligand concentrations were
evaluated by UV spectrophotometry, measuring the absorbance
of its benzenoid bands (p–p*) at k = 280 nm. The distribution
coefficients, D (or logD), were obtained as the ratio of the
ligand concentrations in the organic phase to that in the aqueous
phase.
Results and discussion
Synthesis
This ligand was synthesized through a straightforward method
outlined in Scheme 1. Two equivalents of 1-(3-aminopropyl)-
3-benzyloxy-2-methyl-4-pyridinone were condensed with
N-benzyliminodiacetic acid, upon its previous activation with
ethylchloroformate in THF. Purification by flash silica-gel
chromatography column afforded the bis-hydroxypyridinone
N,O-protected intermediate, IDABzPr(3,4-HPBz)2, in 45%
yield as a beige solid. Removal of the benzyl protecting groups
was performed by standard hydrogenolysis (1.5 atm H2 over
10% Pd/C) to give the final product, IDAPr(3,4-HP)2, as a white
powder (g = 93%).
Biodistribution studies
Animal experiments were carried out with groups of
3–5 female mice CD-1 (randomly bred Charles River, from
CRIFFA, Spain) weighing approximately 25 g. Animals were
intravenously injected with 100 lL (5–10 MBq) of 67Ga-citrate
via the tail vein (Group 1). A separated group of animals was
simultaneously intraperitoneally injected with 0.5 lmol of the
ligand in 100 lL saline solution (Group 2). A third group was
intraperitoneally injected with 0.5 lmol of the ligand in 100 lL
saline solution, 30 min after the administration of 67Ga-citrate
(Group 3). Finally, another group was intravenously injected
with 100 lL (5–10 MBq) of the 67Ga-complex previously
prepared (Group 4). Mice were maintained on normal diet
ad libitum. At 5 min, 15 min, 30 min, 1 h, 24 h and 48 h post-
administration, animals were killed by cervical dislocation.
The radioactive dosage administered and the radioactivity in
the sacrificed animal were determined by counting in a dose
calibrator (Aloka, Curiemeter IGC-3, Aloka, Tokyo, Japan).
The difference between the radioactivity in the injected and the
sacrificed animal was assumed to be due to excretion. Tissue
samples of main organs were then removed for counting in a
gamma counter (Berthold LB2111, Berthold Technologies,
Germany). Biodistribution results were expressed as percent
Protonation
In its neutral form, the ligand [IDAPr(3,4-HP)2, hereafter also
designed as H2L] has two dissociable protons, corresponding
to the pyridinone hydroxy groups, but the fully protonated
form of the ligand has three extra N–H protons associated
to two pyridinium and one ammonium groups. The fitting
analysis of the corresponding potentiometric titration curve
(Fig. 1) by the HYPERQUAD 2000 program14 gave the global
protonation constants which allowed the calculation of the
stepwise protonation constants (logKi) presented in Table 1. For
comparison purposes, this table also includes the protonation
constants reported in the literature for the bidentate 1,2-
dimethyl-3-hydroxy-4-pyridinone (3,4-DMHP).22 Analysis of
Table 1 shows that the calculated logKi for IDAPr(3,4-HP)2
are according to chemical evidences. In fact, the first two
protonation constants (9.94, 9.44) are easily attributed to the
hydroxypyridinone hydroxy groups and the third one (5.42) is
attributed to the amine group, as suggested by its similarity with
Scheme 1 Synthesis procedure for the ligand.
3 7 7 4
D a l t o n T r a n s . , 2 0 0 4 , 3 7 7 2 – 3 7 8 1

View Article Online
Table 1 Stepwise protonation constants for IDAPr(3,4-HP)2 and DMHP; global formation constants (logb) for their metal complexes (M = Fe, Ga,
Al, Zn) at I = 0.1 M KNO3 and 25.0 °C; pM values for these compounds and some relevant synthetic and biological ligands
logb
Compound
logKi
(p,q,r)
FepHqLr
GapHqLr
AlpHqLr
ZnpHqLr
(2,−2,2)
(1,−1,1)
(2,3, 3)
(2,2,3)
(2,1,3)
(2,0,3)
(1,3,1)
(1,2,1)
(1,1,1)
(1,0,1)
pM
44.00(9)
—
—
32.05(6)
—
18.23(9)
84.97(1)
80.16(4)
74.64(4)
68.44(4)
33.21(9)
—
28.83(7)
24.30(8)
22.9
89.53(2)
85.54(3)
79.90(9)
7426(6)
33.59(2)
—
31.16(3)
26.16(4)
25.8
76.46(2)
71.72(3)
66.21(3)
60.18(3)
30.44(6)
27.71(5)
25.37(2)
20.35(3)
18.8
9.94(1)
9.44(1)
5.42(2)
3.54(2)
3.11(3)
27.28(1)
IDAPr(3,4-HP)2
18.92(6)
13.30(3)
9.7
7.19
13.53
—
(1,0,1)
(1,0,2)
(1,0,3)
15.10
26.61
35.88
13.17
25.43
35.76
12.20
23.25
32.62
9.77
3.68
3,4-DMHPa
KEMPPr(3,4-HP)2b
DTPAc
pM
pM
pM
pM
pM
pM
19.3
20.5
24.6
24.3
26.5
20.3e
19.4
18.8
20.9
18.8
22.4
20.3f
16.1
18.1
15.2
13.2
19.3
14.5g
6.2
9.5
14.8
17.9
6.6
—
—
—
—
—
—
DOTAc
DBFd
Transferrin
pM = −log[M] with CL/CM = 10 and CM = 10−6 M. a Ref. 21. b Ref. 11. c Ref. 31. d Refs. 32 and 33. e Ref. 34. f Ref. 35. g Ref. 36.
Fig. 1 Potentiometric titration curves of IDAPr(3,4-HP)2, alone
(CL = 5.998 × 10−4 M) and in presence of the three-charged metal
ions (M3+), at conditions of: ca. 1:2 metal–ligand system molar
ratio and CFe = 2.844 × 10−4 M (1), CGa = 2.982 × 10−4 M (2), CAl=
2.972 × 10−4 M (3); ca. 1:1 metal–ligand ratio and CFe = 5.918 × 10−4
M
(4), CGa = 5.540 × 10−4 M (5), CAl = 5.631 × 10−4 M (6). T = 25.0 °C,
I = 0.1 M, KNO3).
the corresponding value for iminodihydroxamic acid (IDHA,
logK1 = 5.66).23 The fairly acidic character of this protonated
amine group should be due to some electron-withdrawing effect
of the amide groups in its vicinity as well as to some stabilization
of the conjugated base by hydrogen bonding (amide –NH to
N–amine) through the formation of stable five-membered rings.
The two remaining protonation constants (logKi = 3.54, 3.11)
are undoubtedly attributed to pyridinium protons. Therefore,
based on the calculated protonation constants, the neutral form
of the ligand (H2L) should be the major species at the physi-
ological pH (Fig. 2(a)).
Fig. 2 Concentration distribution curves of the free ligand IDAPr(3,4-
HP)2 (CL = 5.998 × 10−4 M) (a) and of the complexes formed in
the system Zn(II)–IDAPr(3,4-HP)2 at (1:1) metal to ligand ration,
CM = 1.8 × 10−3 M (b).
deprotonation could not be completely finished because precipi-
tation occurred (pH  7), probably due to formation of neutral
species. The deprotonation of the ammonium group is not
greatly affected by the presence of the metal ion, thus suggesting
that the amine group is not involved in the coordination to
the metal ion. The first species to be formed (ZnH3L) has
the metal ion coordinated by one hydroxypyridinone moiety
(HP), thus keeping one amine and two HP protons (hydroxy
and pyridinium). Therefore, when ZnH3L releases both these
HP protons (in a cooperative way) ZnHL is formed (two HP
moieties are coordinated to the metal ion and the amine still
remains protonated). A brief analysis of the global formation
Metal complexation
Although the main interest of the present metal-complexation
studies is centred on a set of three-charged hard metal ions
with biological interest (Fe, Al, Ga), the fact that zinc is an
essential element present in biological systems led us to make
a preliminary evaluation of the interaction of this ligand with
Zn(II) by potentiometry. Analysis of the potentiometric titration
curve for the 1:1 metal-to-ligand ratio conditions showed
that the complexation started only at pH  3.5 and the last
D a l t o n T r a n s . , 2 0 0 4 , 3 7 7 2 – 3 7 8 1
3 7 7 5

View Article Online
constants (logbML) calculated for the 1:1 (M:L) complexes
with IDAPr(3,4-HP)2 shows they are very close to the values
1
obtained for the corresponding 1:2 complexes, logbML , with
2
the mono-hydroxypyridinone derivative (Deferriprone; Table 1),
thus indicating identical hydroxypyridonate coordination mode.
However, due to differences on the denticity and the acidity of
these ligands, accurate comparison of their metal-complexation
effectiveness is better based on the corresponding pM values
(pM = −log[M2+] for pH = 7.4 at micromolar concentration
of the metal ion and 10-fold ligand excess), assuming that no
precipitation takes place. Therefore, IDAPr(3,4-HP)2 presents
higher affinity for zinc than the corresponding mono-derivatives
but it is still quite low (pZn = 9.7) and, at the physiological pH,
the major complex species is in the neutral form (ZnL, Fig. 2).
To study the complexation behaviour of IDAPr(3,4-HP)2
towards the M3+ hard metal ions (M = Fe, Al, Ga) in aqueous
solutions, potentiometric and spectrophotometric techniques
were used at various metal-to-ligand molar ratios. Representative
potentiometric titration curves are shown in Fig. 1 for the ligand
alone and in the presence of each of those M3+ metal ions, at
the 1:2 and 1:1 metal-to-ligand molar ratios. For each of these
binary systems there is a considerable drop of the pH at the
beginning of the titration, thus indicating high affinity of the
ligand for these metal ions. A more detailed analysis of the (1:1)
titration curves shows the existence of breaks at a = 4 (a = mol
of base per mol of the ligand), indicating that at ca. pH = 3.5
four protons have been released (presumably from two hydroxy
and two pyridinium groups). Accordingly, there was formation
of the MHL species, having the metal ion coordinated to both
hydroxypyridinone moieties but the amine group being still
protonated. However, analysis of the (1:1) titration curves
showed that one extra deprotonation process is occurring (ca.
one equivalent of extra base consumption), which should be due
to some hydrolytic process. As expected, this process was not
observed in the presence of ligand excess.
Fitting analysis of the potentiometric and spectrophoto-
metric titration data, for each ligand–metal system at both the
metal-to-ligand molar ratios, leads to a speciation model of the
complex species whose global stability constants are listed in
Table 1.
Spectrophotometric titrations of the IDAPr(3,4-HP)2–Fe
system with base were also performed, at both the stoichio-
metric conditions used for potentiometric measurements, to aid
the understanding of the complexation model. Analysis the pH
dependency of UV-Vis electronic spectra (Fig. 3) suggests the
existence of different equilibrium models for each stoichiometric
condition.
Spectrophotometric titrations further indicate that the
complexation process starts at ca. pH 0.4. Therefore, at the
beginning of the potentiometric titrations (ca. pH = 2),
the complex formation was already complete. Therefore to
study the stability constants of the corresponding species it
was required to go to a lower pH and this study could not be
performed by potentiometry because we could not measure the
pH value. However, below pH 2, the pH could be accurately
calculated based on the analytical concentration of the proton
and the equilibrium process could be followed and studied by
spectrophotometry. In fact, analysis of the spectrophotometric
spectra (Fig. 3) shows that from the beginning of the titration
(pH = 0.4), the formation of the monochelated and other
species could be easily followed by the changes of the wavelength
and absorbance values which are in agreement with the species
distribution diagram. A detailed analysis of Fig. 3 indicates that
the spectral parameters for the species with maximum concen-
tration at ca. pH 1 (FeH3L; kmax = 562 nm, e = 2079 M−1 cm−1)
and 1.8 (FeHL; kmax = 510 nm, e = 3593 M−1 cm−1) correspond
to the CT absorption bands of mono- and bis-chelated species,
respectively, in agreement with literature reports for Fe(III)–(3-
hydroxy-4-pyridinone) complexes.22 Above this pH value, both
the stoichiometric systems presented blue-shifts of the absorp-
tion bands, thus indicating further increase of the degree of
Fig. 3 Absorption spectra registered at various pH values for the
Fe(III)–IDAPr(3,4-HP)2 system at CM = 1.99 × 10−4 M and two different
metal-to-ligand molar ratios: 1:2 (a) and 1:1 (b).
coordination around each metal ion, despite eventual differ-
ences in the metal environment. For higher pH conditions,
potentiometric measurements could be used to calculate the
stability constants of other species, but holding constant the
values previously determined by spectrophotometry.
Regarding the 1:2 stoichiometric conditions (Fig. 4(a)), above
ca. pH 3.4 and up to ca. pH 9.1, another species starts being
formed with kmax gradually shifting to 460 nm (e = 5460 M−1
cm−1), a spectral characteristic of tris-chelated species, thus
indicating the formation of 2:3 (M:L) dimeric species (Fe2HxL3;
x = 0–3). In this case, it can be assumed that one of the three
ligand molecules could bridge two bis-chelated iron centers to
complete their coordination sphere (Scheme 2(a)).
For the 1:1 stoichiometric conditions, analysis of the pH
dependency of the absorption spectra (Fig. 4(b)) shows the
existenceof abis-chelatedspecies(kmax = 510nm)withmaximum
formation at ca. pH = 2.06 and it should correspond to the
hydroxypyridinonate-to-iron CT band of the FeHL species.
Above pH ca. 2.5 and up to pH 5.5 the amino group is expected
to be deprotonated with formation of FeL. Comparison of its
spectral parameters (kmax = 475 nm, e = 3734 M−1 cm−1) with
those of the previous species (FeHL) suggests the existence of
some strengthening of the metal interaction; probably there is
also formation of soluble mixed hydroxo species, FeHL(OH),
instead of the assumed neutral FeL species, but they cannot be
differentiated by potentiometry. Above this pH (pH 5.5–7.2)
there is a further hypsochromic shift and absorbance increase
of that band, which parallels the above referred one equivalent
extra consumption of base in the potentiometric titration.
This result suggests the formation of mixed hydroxo–ligand
complexes which, according to the calculated spectral para-
meters (kmax = 465 nm, e = 7120 M−1 cm−1, per complex), should
be attributed to di-iron complex [Fe2(OH)2L2].24 Under this
assumption, each iron centre would be coordinated by two
hydroxypyridinone moieties of each ligand and one hydroxy
3 7 7 6
D a l t o n T r a n s . , 2 0 0 4 , 3 7 7 2 – 3 7 8 1

View Article Online
pH and conditions of the complexation studies. At T = 298 K,
the pD dependence of the magnetic moment (lM) is illustrated
in Fig. 4. For the 1:1 stoichiometry (Fig. 4(b)), at ca. pH 6 there
is a large drop of lM (from ca. 5.0 to ca. 3.1 lB) which should
result from a strong anti-ferromagnetic interaction between two
Fe(III) sites, thus giving support to the assumed formation of
a binuclear species with two bridging hydroxo groups. In fact,
for ca. pH < 6, lM values are close to the magnetic moments
reported for non-interacting mono-ferric complexes in the
solid state (5.9 lB),26 but for ca. pH  6 the value calculated for
the magnetic moment (ca. 3.1 lB) is close to that found upon
the formation of l2-oxo–diiron(III) complexes (lM = 3.4 lB, at
300 K).25 Concerning the 1:2 stoichiometric conditions, a steady
decreasing of lM with the pH (for pH = 4, lM  4.5 lB) seems
to parallel the formation of the dimeric species (Fe2HiL3). Such
behaviour can be attributed to a particular structure adopted
by that complex, in which the metal centres are close enough
to have some anti-ferromagnetic interaction and/or also the
presence of diamagnetic impurities such as minor l2-hydroxo–
diiron(III) species.
Since EPR spectra from the available apparatus require much
higher concentrations and much lower temperatures than those
used for potentiometric or magnetic moment measurements,
with concomitant precipitation and changes on the speciation
model, they could not be used to support the dimeric species.
Concerning the IDAPr(3,4-HP)2–Al(III) and IDAPr(3,4-
HP)2–Ga(III) systems, analysis of their equilibrium models
(Table 1 and Fig. 5) shows that, in agreement with the profiles
of the potentiometric titration curves for the 1:1 stoichiometric
conditions (Fig. 1), the hydrolytic processes for the aluminium
system seem to parallel those for the ferric system, albeit, for
the gallium system, they are present at a much lower extent.
Since none of these Ga or Al complexes present d–d transi-
tions or CT bands (and the ligand bands do not change too
much upon complexation), the spectrophotometric techniques
while used for the Ga complexation (low pH conditions) could
not give any valuable help to support the models. Fortunately
27Al NMR spectroscopy is usually able to discriminate between
aluminium-coordinating modes, namely between octahedral
(d = −10–40 ppm) and tetrahedral (d = 60–80 ppm).27 Therefore
27Al NMR spectra were obtained at the same concentration
range used in potentiometry for the 1:1 (M:L) molar ratio,
aimed at getting some clue to the geometry of the complexes.
Two representative spectra recorded at two pD values and
T = 25 °C are depicted in Fig. 6.
Fig. 4 Concentration distribution curves of the Fe-containing
species for the system [Fe(III)–IDAPr(3,4-HP)2] at 1:2 (M:L),
CM = 2.844 × 10−4 M) (a); and 1:1 (M:L), CM = 5.918 × 10−4 M (b);
at 25.0 °C, pH dependence of: kmax of UV-Vis spectra of [Fe(III)–
IDAPr(3,4-HP)2] complexes, I = 0.1 M HNO3 (); pD dependence of
the molar magnetic moments ().
These spectra present a considerable amount of noise because
27Al NMR spectrometry requires much higher concentrations
than those used in potentiometry and increasing analytical
concentration would affect the concentration distribution (the
hydrolytic reactions could be reduced). Analysis of the spectra
depicted in Fig. 6 shows, at very acidic conditions, the existence
of a very sharp peak (d = 0 ppm), undoubtedly ascribed to the
very symmetrical hexa-aqua octahedral complex [Al(H2O)6]3+.
All the remaining peaks are more or less broad and they should
correspond to Al–L complex species. A further analysis of these
spectra suggests that at ca. pH 2.5 the complexation has already
started, with the formation of a complex species (d = 66.7 ppm),
presumably with unsymmetrical tetrahedral geometry, and
eventually slow exchange processes. According to the species
distribution, they should be mainly ascribed to bis-chelated
hydroxypyridinonate species.
Scheme 2 Structural diagrams for the complexes: Fe2H3L3 (a),
Fe2(OH)2L2 (b).
The peak appearing at 39.2 ppm (ca. pH 5.57) should be
attributed to a complex with octahedral geometry around
the aluminium. Although this chemical shift is close to that
reported for a tris-hydroxypyridonate coordination (d = 37 ppm,
W1/2 = 920 Hz),28 such a complete hydroxypyridinonate coordi-
nation cannot be considered for an Al-bis-hydroxypyridinone
system at 1:1 metal-to-ligand molar ratio. However, these
spectral data can be ascribed to an octahedral di-bridged
di-(hydroxo-ligand–Al) [Al2(OH)2L2] complex, according to the
proposed speciation model. Based on known chemical shifts
group (due to the hydrolysis of one water-coordinated mole-
cule), but each one bridging the two metallic centers, accord-
ing to Scheme 2(b). However, the presence of such l2-hydroxo
bridged bonds in the core of dinuclear species is supposed to
induce some distortion from the octahedral symmetry because
the Fe–Ohydroxo distances are expected to be shorter than those
of Fe–OHP.
Aiming at gain some light on the assumed equilibrium
models the magnetic susceptibility was measured and the
magnetic moment was calculated in about the same range of
D a l t o n T r a n s . , 2 0 0 4 , 3 7 7 2 – 3 7 8 1
3 7 7 7

View Article Online
Fig. 5 Concentration distribution curves of the complexes formed for the systems Ga(III)–IDAPr(3,4-HP)2 (a) and Al(III)–IDAPr(3,4-HP)2 (b), at
two different metal-to-ligand molar ratios (M:L): 1:2, CM = 2.844 × 10−4 M (1) and 1:1, CM = 5.918 × 10−4 M (2).
presents a much higher metal chelating efficacy than the
monohydroxypyridinone derivative, 3,4-DMHP (Deferriprone)
(DpGa = 3.5). Thus, it can be predicted as a better scavenger
for this type of hard metal ions then that drug. It also presents
a higher chelating ability than the dihydroxypyridinonate
analogue, KEMPPr(3,4-HP)2,11 which may due to the higher
flexibility of the present ligand, the main motivation its design.
Furthermore, IDAPr(3,4-HP)2 shows a much higher chelating
efficacy and specificity for these M3+ metal ions than some of the
polyaminocarboxylate ligands with clinical applications (DPTA
and DOTA).31 The present ligand shows a chelating strength
similar to that of DFB.32,33 for this set of metal ions, albeit higher
Fig. 6 27Al NMR spectra of [Al(III)–IDAPr(3,4-HP)2] complexes at
for millimolar concentrations.
Although from complexation data with transferrin (Table 1)
it seems likely that IDAPr(3,4-HP)2 and DFB can compete
1:1 metal-to-ligand molar ratio, CAl = 6.0 × 10−4 M and indicated pD
values.
with apotransferrin for this set of metal ions,34–36 the actual
for Al-coordination with three hydroxypyridinones (37 ppm)
ability and mechanism of DFB to remove iron from transferrin
and four hydroxy groups (80.5 ppm), the contribution of two
remains unclear. However, since the monohydroxypyridinone
hydroxypyridinonates and one hydroxo group per aluminium
Deferriprone proved to be able to mobilize iron from several
centre could account for the observed chemical shift, thus
iron-containing proteins, such as ferritin and transferrin,37
supporting the proposed speciation model. Above ca. pH 6 there
an identical role, but with eventually higher efficacy, can be
was precipitation under our experimental conditions and so no
postulated for the present dihydroxypyridinone.
further spectra are obtained.
Regarding the Ga(III) complexation, the best fitting of the
equilibrium data were obtained with the model suggested by
Molecular modelling
UV-spectrophotometry titration at low pH and potentiometry
for pH > 2, although for the Al(III) complexation a good model
was obtained only based on the potentiometric data.
In order to get a further insight on the dimeric ferric complex
structures of the present tetradentate ligand and better
understanding of the magnetic properties presented by ferric
solutions, brief molecular dynamics/mechanics calculations
were performed to determine the most probable structure and
the Fe–Fe distance for the Fe2H−2L2 complex. We have used
the CERIUS2 program38 with the default force-fields39 and a
previously reported approach,40 based on the X-ray structure
data for ferric complexes with the monohydroxypyridone 3,4-
DMHP.41 This modelling study (Fig. 7) helps in the elucidation
of the structure, but it presents a shorter distance between the
ferric centers (dFe–Fe = 2.75 Å) than that found in the literature
for a series of l2-hydroxo–diiron(III) complexes (3.0–3.2 Å),42
which could be mostly attributed to the force fields used in this
brief simulation.
The rationalisation of those results is in good agreement with
the speciation model listed in Table 1 and with the fact that the
deprotonation of non-coordinated amino groups and coordi-
nated water molecules can overlap. Furthermore, the studies on
the complexation with these M3+ metal ions in aqueous solution,
revealed a trend in the equilibrium constants, logbML, for Al <
Ga < Fe, which correlates well with the logKMOH values reported
for the same metal ions (11.27, 11.0, 8.47, respectively),29 thus
reflecting the rather ionic nature of this type of complexes, as
previously reported for other hydroxypyridinone derivatives.30
Comparative analysis of complexation ability
Further conclusions about the metal-affinity of this chelator
at physiological pH can be made based on the analysis of the
calculated pM3+ values (which are directly related with the
chelating strength) and comparison with the corresponding
values for a set of usual synthetic and biological ligands
presented in Table 1. It is clearly shown that, for the micromolar
conditions that prevail in biological systems, IDAPr(3,4-HP)2
Distribution coefficient
The lipo/hydrophilic character of IDAPr(3,4-HP)2 was
evaluated based on its distribution coefficient (logD) between
1-octanol and aqueous phase (Tris buffered at pH = 7.4).
Although, at pH = 7.4, this compound is mainly in the
neutral form, due to the low pKa of the amino group, it is
3 7 7 8
D a l t o n T r a n s . , 2 0 0 4 , 3 7 7 2 – 3 7 8 1

View Article Online
Table 2 Biodistribution data in percent of injected dose per total organ (% I.D. ± SD) of 67Ga–citrate (Group 1), 67Ga–citrate with simultaneous
intraperitoneal injection of IDAPr(3,4-HP)2 (Group 2), 67Ga–citrate with intraperitoneal injection of IDAPr(3,4-HP)2 30 min later (Group 3) and
67Ga–IDAPr(3,4-HP)2 (Group 4), 1 and 24 h after intravenous administration in female mice (n = 3–5)
% I.D. ± SD
Group 1
1 h
Group 2
1 h
Group 3
1 h
Group 4
1 h
Organ
24 h
4.0 ± 2.0
24 h
0.4 ± 0.1
24 h
0.6 ± 0.2
24 h
0.1 ± 0.01
0.2 ± 0.1
0.2 ± 0.1
0.01 ± 0.0
<0.01
0.01 ± 0.0
0.3 ± 0.0
0.2 ± 0.1
0.6 ± 0.3
0.04 ± 0.0
97.4 ± 0.6
Blood
Liver
Intestine
Spleen
Heart
3.2 ± 1.9
1.8 ± 0.2
9.8 ± 1.6
0.1 ± 0.0
0.1 ± 0.0
0.3 ± 0.2
0.9 ± 0.1
11.3 ± 0.7
17.1 ± 7.5
0.8 ± 0.3
17.2 ± 3.6
3.1 ± 1.3
1.8 ± 0.4
1.6 ± 0.2
0.2 ± 0.04
0.2 ± 0.02
0.5 ± 0.1
3.9 ± 1.0
5.2 ± 0.05
2.3 ± 0.2
0.4 ± 0.0
78.4 ± 7.7
3.0 ± 0.2
1.2 ± 0.2
2.0 ± 0.4
0.1 ± 0.02
0.1 ± 0.0
0.3 ± 0.1
1.0 ± 0.2
4.0 ± 0.4
2.5 ± 0.4
0.5 ± 0.2
71.3 ± 6.8
1.3 ± 0.4
0.5 ± 0.0
0.4 ± 0.0
0.02 ± 0.0
0.03 ± 0.0
0.1 ± 0.0
0.6 ± 0.1
2.5 ± 0.8
1.2 ± 0.7
0.1 ± 0.0
76.0 ± 5.8
5.1 ± 1.9
4.3 ± 0.7
0.2 ± 0.1
0.1 ± 0.0
0.4 ± 0.1
1.3 ± 0.3
7.4 ± 0.8
21.8 ± 3.2
0.5 ± 0.2
35.0 ± 7.2
0.7 ± 0.1
1.0 ± 0.3
0.08 ± 0.0
0.04 ± 0.0
0.1 ± 0.0
1.0 ± 0.0
0.8 ± 0.2
1.5 ± 0.2
0.2 ± 0.0
96.2 ± 0.4
1.3 ± 0.3
1.0 ± 0.2
0.1 ± 0.0
0.04 ± 0.0
0.08 ± 0.0
1.0 ± 0.2
1.4 ± 0.2
4.4 ± 1.0
0.2 ± 0.1
92.4 ± 1.0
Lung
Kidney
Muscle
Bone
Stomach
Excretion
Fig. 8 Biodistribution data in most relevant organs, expressed as %
I.D./total organ for 67Ga–IDAPr(3,4-HP)2 at 5 min, 15 min, 30 min,
1 h, 24 h and 48 h, after intravenous administration in female mice
(n = 3–5).
Fig. 7 Low-energy conformation simulated for the dimeric complex
Fe2H−2L2. For simplification of the picture, all the hydrogen atoms are
removed.
more hydrophilic (logD = −1.7 ± 0.02) than the corresponding
monomeric species (logD = −1.03, for 3,4-DMHP),43 although
slightly less then KEMPPr(3,4-HP)2, which is negatively charged
(logD = −1.96 ± 0.02).10 The high hydrophilic character of
IDAPr(3,4-HP)2 may be also due to existence of some hydrogen
bonding between the amine group and the water molecules.
Therefore, differences in target biological sites and clearance
pathways are expected for these compounds.
biodistribution profile, radioactivity is mainly excreted from the
whole animal body by the kidneys at a high rate (more than 75%
in the first hour). ITLC analyses of mice urine samples, collected
at sacrifice time, indicated that the radioactivity is mostly
excreted as the 67Ga complex. This observation demonstrates
the high in vivo stability of this complex.
Therefore, this favourable biodistribution pattern (with a high
rate of excretion and no significant uptake in any particular
organ) as well as the high affinity of the ligand to coordinate
these hard trivalent metal ions, suggested a good capability of
the ligand to be potentially used to complex and remove in vivo
overdoses of unwanted metal ions. To evaluate this ability two
different experiments were carried out with the 67Ga radiotracer.
In a first set of animals, 67Ga–citrate was intravenously adminis-
tered immediately followed by intraperitoneal administration
of 0.5 lmol of ligand in saline solution, while another set of
animals were administered with the same solution of ligand
only 30 min after the 67Ga–citrate injection. Therefore biological
distributions of the radiotracer in the presence or the absence of
the ligand, at 1 h and 24 h, were compared (Table 4).
These biodistribution studies clearly evidence that the
administration of the ligand highly interferes in the normal
tissue distribution profile of the tracer leading to a rapid
clearance from most tissues and a highly increased excretion
rate of radioactivity from the animal body, as compared with
the excretion of 67Ga in the citrate form. Such enhancement is
probably due to the higher metal-binding affinity of the ligand
as compared to that of citrate (pGa = 18.8).44 Comparative
analysis of the biodistribution profiles of the radiotracer
with administration of ligand solution, either simultaneously
or 30 min later, vs. biodistribution of 67Ga–IDAPr(3,4-HP)2
Biological assays
In vivo assays were carried out in female mice to investigate
both the biodistribution of the 67Ga–IDAPr(3,4-HP)2 complex
and the efficacy of the chelating agent in the mobilisation of
gallium in 67Ga–citrate loaded mice, as models for other hard
metal ions. The synthesis of the radioactive 67Ga-complex with
IDAPr(3,4-HP)2 was accomplished with labelling efficiencies
higher than 98%, as indicated by ITLC analysis. High radio-
chemical purity was obtained immediately after addition of the
metal to the ligand solution at room temperature. Thus, 100 lL
of a 67Ga–IDA(Pr-3,4-HP)2 solution, as equivalent to 0.5 lmol
of ligand, were administered to each animal for biodistribution
studies. Both the chelator and the 67Ga-complex were well
tolerated by animals at the administered doses with no obvious
side effects. Tissue distribution data of radionuclide complex up
to 48 h after administration, expressed as percentage of injected
dose per total organ, can be overviewed in the histogram
depicted in Fig. 8.
Biodistribution results show that the complex is quite
efficiently cleared from the blood via both the hepatobiliar and
renal pathways. Clearance from most organs including soft
tissues like muscle, and bone is also quite rapid. Based on this
D a l t o n T r a n s . , 2 0 0 4 , 3 7 7 2 – 3 7 8 1
3 7 7 9

View Article Online
demonstrates that there are no significant differences in the
excretion rate. Clearance from most organs is always highly
enhanced. However, at 1 h after administration of the ligand
solution, there is a higher accumulation of activity in main
organs such as blood and excretory organs when compared
to that of the 67Ga complex administration. These slight
differences are probably due to some delay in the mobilisation
of the metal.
The evidence that the administration of ligand solution
induces a fastest elimination rate of 67Ga from most tissues and
whole animal body suggested its potential interest to be used in
vivo to complex with trivalent metals that may be overdosed in
some clinical situations. Comparison of our results with those
found in the literature45 for the 67Ga complex with Deferriprone
supports this hypothesis since a more rapid clearance and
urinary excretion of the complex relative to the radiotracer
were also found. To evaluate the in vivo behaviour and the
ability of Deferriprone to coordinate in vivo with gallium in
our animal model, a set of animal experiments were carried out
by administration of 67Ga citrate followed by intraperitoneal
injection of 0.5 lmol of Deferriprone in saline solution and
another group of animals were administered with the same
amount of solution 30 min after administration of the radio-
tracer as we previously did with our ligand. Biodistribution was
evaluated at 1, 24 and 48 h.
to a iminodiacetic acid skeleton, has been prepared, studied in
its complexation behaviour towards a set of hard metal ions and
bio-assayed in mice. Equilibrium solution results demonstrate
that, for a wide range of pH and the diluted conditions prevail-
ing in vivo, this new compound possesses a higher efficacy for
chelating hard metal ions (e.g. Fe3+, Ga3+, Al3+) than the mono-
hydroxypyridinone Deferriprone (a drug clinically used) and
also some bioligands (e.g. citric acid and transferrin). Results
of biological studies in mice indicated that the 67Ga-complex is
very stable in vivo without apparent toxicity and it presents a
much more rapid clearance from most tissues than the citrate
complex. As a gallium decorporating agent in mice it has also
been proved to present a higher efficacy than Deferriprone.
Therefore, independently of the mechanism involved in the
in vivo gallium transfer and its dependence on other features
than the favourable lipo/hydrophilic balance and the chelating
strength, this set of results indicates there are good perspectives
for this new compound to be a potential decorporating agent
for this type of hard metal ions in overload situations without
depletion of essential metal ions such as zinc.
Acknowledgements
The authors thank to the Portuguese Fundação para a Ciência
e Tecnologia (FCT) (project POCTI/35344/00), the Hungarian
Scientific Research Fund (OTKA) (OTKA T034674) and COST
D21/001 program for financial support.
This preliminary study also indicated an increase in the
excretion of the radiotracer when Deferriprone was adminis-
tered (about 60.3 and 87.4% at 1 h and 48 h after administration,
respectively). A comparative analysis of these results with data
from our ligand is shown in the histogram of Fig. 9.
References
1 Z. D. Liu and R. C. Hider, Med. Res. Rev., 2002, 22, 26.
2 R. A. Yokel, Coord. Chem. Rev., 2002, 228, 97.
3 M. A. Santos, Coord. Chem. Rev., 2002, 228, 187.
4 G. J. Kontoghiorghes, Lancet, 1985, 817.
5 M. Meyer, J. R. Telford, S. M. Cohen, D. J. White, J. Xu and K. N.
Raymond, J. Am. Chem. Soc., 1997, 119, 10093.
6 B. L. Rai, H. Khodr and R. C. Hider, Tetrahedron, 1999, 55,
1129.
7 Y. Sun, M. J. Motekaitis and A. E. Martell, Inorg. Chim. Acta, 1998,
281, 60.
8 A. J. Johnson, B. O’Sullivan and K. N. Raymond, Inorg. Chem.,
2000, 39, 2652.
9 M. A. Santos, R. Grazina, L. Gano, S. Gama, Port. Pat., PT102660,
2003.
10 M. A. Santos, R. Grazina, P. Buglyó, S. Gama and E. Farkas,
Polyhedron, 2002, 21, 2609.
11 M. A. Santos, R. Grazina, L. Gano and E. Farkas, unpublished
work.
12 Purification of Laboratory Chemicals, ed. W. L. F. Armarego and
D. D. Perrin, Butterworth-Heinemann Press, Oxford, 4th edn.,
1999.
13 F. J. C. Rossotti and H. Rossotti, J. Chem. Educ., 1965, 42, 375.
14 P. Gans, A. Sabatini and A. Vacca, Talanta, 1996, 43, 1739.
15 A. Sabatini, A. Vacca and P. Gans, Talanta, 1974, 21, 53.
16 L. Zékány and I. Nagypál, in Computational Methods for the
Determination of Stability Constants, ed. D. Leggett, Plenum Press,
New York, 1985.
Fig. 9 Biodistribution data in percent of injected dose per total organ
(% I.D. ± SD) of 67Ga–citrate (A), 67Ga–citrate with simultaneous
intraperitoneal injection of IDAPr(3,4-HP)2 (B) or Deferriprone (C),
67Ga–citrate with intraperitoneal injection of IDAPr(3,4-HP)2 (D) or
Deferriprone (E) 30 min later; 67Ga–IDAPr(3,4-HP)2 (F), 1 h after
intravenous administration in female mice (n = 3–5).
17 M. A. Santos, R. Grazina, A. Q. Neto, G. Cantinho, L. Gano and
L. Patrício, J. Inorg. Biochem., 2000, 78, 303.
18 A. Leo, C. Hansch and D. Elkins, Chem. Rev., 1971, 71, 526.
19 A. K. Covington, M. Paabo, R. A. Robinson and R. G. Bates, Anal.
Chem., 1968, 40, 700.
20 D. F. Evans, J. Chem. Soc., 1959, 2003.
In this study it was found when IDA(Pr-3,4-HP)2 was
administered a more rapid clearance from organs and excretion
of radioactivity was observed than in case of Deferriprone.
Although the mechanism involved in the biodistribution in
mice is not clear, namely the actual ability to mobilize gallium
from transferrin, the present results support our expectations in
terms of its high chelating capacity and allow us to suggest the
potential usefulness of our compound for decorporating metals
associated to pathologic situations.
21 E. N. Sloth and C. S. Garner, J. Chem. Phys., 1954, 22, 2064.
22 E. T. Clarke and A. E. Martell, Inorg. Chim. Acta, 1992, 191, 57.
23 R. Karlicek and J. Majer, Collect. Czech. Chem. Commun., 1972, 37,
151.
24 T. Justel, M. Muller, T. Weyhermuller, C. Kressel, E. Bill,
P. Hildebrandt, M. Lengen, M. Grodzicki, A. X. Trautwein,
B. Nuber and K. Wieghardt, Chem. Eur. J., 1999, 5, 793.
25 M. Gerloch, J. Lewis, F. E. Mabbs and A. Richards, J. Chem. Soc. A,
1968, 112.
26 W. H. Armstrong, M. E. Roth and S. J. Lippard, J. Am. Chem. Soc.,
1987, 109, 6318.
27 F. R. Venema, J. A. Peters and H. van Bekkum, J. Chem. Soc.,
Dalton Trans., 1990, 2137.
28 L. Simpson, S. J. Rettig, J. Trotter and C. Orvig, Can. J. Chem., 1991,
69, 893.
Conclusions
A new dihydroxypyridinone derivative, IDAPr(3,4-HP)2, having
two 3-hydroxy-4-pyridinone (3,4-HP) chelating units attached
3 7 8 0
D a l t o n T r a n s . , 2 0 0 4 , 3 7 7 2 – 3 7 8 1

View Article Online
29 A. E. Martell and R. M. Smith, Critical Stability Constants, Plenum
Press, New York, 1989, First Supplement, pp. 428–429.
30 M. A. Santos, M. Gil, S. Marques, L. Gano, G. Cantinho and
S. Chaves, J. Inorg. Biochem., 2002, 92, 43.
31 Critically Selected Stability Constants of Metal Complexes Database,
ed. A. E. Martell, R. M. Smith and R. J. Motekaitis, College Station
TX, Version 4.0, 1997.
32 B. J. Hernlem, L. M. Vane and G. D. Sayles, Inorg. Chim, Acta, 1996,
244, 179.
33 E. A. Enyedy, I. Pocsi and E. Farkas, unpublished results.
34 R. J. Motekaitis and A. E. Martell, Inorg. Chim. Acta, 1991, 183,
71.
38 CERIUS2 Program, Version 3.0, 1998, Molecular Simulation Inc.,
Cambridge.
39 A. K. Rappe, C. J. Casewit, K. S. Colwell, W. A. Goddard and
W. M. Skiff, J. Am. Chem. Soc., 1992, 114, 10024.
40 M. A. Esteves, M. C. T. Vaz, M. L. S. Gonçalves, E. Farkas and
M. A. Santos, J. Chem. Soc., Dalton Trans., 1995, 2565.
41 E. T. Clarke, A. E. Martell and J. Reibenspies, Inorg. Chim. Acta,
1992, 196, 177.
42 S. M. Gorun and S. J. Lippard, Inorg. Chem., 1991, 30, 1625.
43 R. A. Yokel, A. K. Datta and E. G. Jackson, J. Pharmacol. Exp.
Ther., 1991, 257, 100.
44 A. E. Martell and R. M. Smith, Critical Stability Constants, Plenum
Press, New York, 1977, vol. 3, p. 161.
45 D. A. Clevette, D. M. Lyster, W. O. Nelson, T. Rihela, G. A. Webb
and C. Orvig, Inorg. Chem., 1990, 29, 667.
35 H. R. Harris and V. L. Pecoraro, Biochemistry, 1983, 22, 292.
36 W. R. Harris and J. Sheldon, Inorg. Chem., 1990, 29, 119.
37 G. Faa and G. Crisponi, Coord. Chem. Rev., 1999, 184, 291.
D a l t o n T r a n s . , 2 0 0 4 , 3 7 7 2 – 3 7 8 1
3 7 8 1