Evaluation of 2-methyl-3-hydroxy-4-pyridinecarboxylic acid as a possible chelating agent for iron and aluminium

Article abstract of DOI:10.1039/b717269a

In view of a possible application to Fe and Al chelation therapy, 2-methyl-3-hydroxy-4-pyridinecarboxylic acid (DT2) was synthesised, and its complex formation, electrochemical and cytotoxic properties were studied. The complexing properties of DT2 towards Fe(iii) and Al(iii) were investigated in aqueous 0.6 m (Na)Cl at 25 °C by means of potentiometric titrations, UV-vis spectrophotometry, and ¹H NMR spectroscopy. DT2 is a triprotic acid (H₃L⁺) having pK_a1 = 0.47, pK_a2 = 5.64 and pK_a3 = 11.18. The metal-ligand complexes observed in solution and their corresponding stability constants (logβ values) are the following: FeLH (19.38), FeL (16.01), FeLH_-1 (12.28), FeL ₂H₂ (37.29), FeL₃H₃ (53.41), FeL₃H₂ (47.99), FeL₃H (41.21) and FeL ₃ (34.1); AlLH (17.43), AlL₂H₂ (33.74), AlL₂H (27.6), AlL₃H₃ (48.72), AlL ₃H₂ (42.67), AlL₃H (35.8) and AlL₃ (27.92). The complex formation between DT2 and Fe(ii) was studied by UV-vis: the weak complex FeLH (logβ = 15.8) was detected. DT2 shows a lower complexation efficiency with Fe(iii) and Al(iii) than that of other available chelators, but higher than that of its non-methylated analogue 3-hydroxy-4-pyridinecarboxylic acid (DT0). The electrochemical behaviour of DT2 was investigated by means of cyclic voltammetry, indicating that the oxidation of the ligand proceeds through a two electron process with a CECE mechanism. Voltammetric curves suggest that the oxidation or the reduction of DT2 in vivo is unlikely. According to the thermodynamic data, also the Fe(iii)-DT2 complexes do not undergo redox cycling at physiological pH. Amperometric titrations of solutions containing Fe(iii) and DT2 at pH = 5 indicated the same Fe(iii): ligand stoichiometric ratio as calculated from potentiometric data. The toxicity of DT2 and of other simple hydroxypyridinecarboxylic acids was investigated in vitro and no cytotoxic activity was observed (IC₅₀ > 0.1 mM) on cancer cell lines and also on primary human cells, following a three day exposure. The Royal Society of Chemistry.

Full text of DOI:10.1039/b717269a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Evaluation of 2-methyl-3-hydroxy-4-pyridinecarboxylic acid as a possible
chelating agent for iron and aluminium
a
b
c
c
a
a
Annalisa Dean, Maria Grazia Ferlin, Paola Brun, Ignazio Castagliuolo, Denis Badocco, Paolo Pastore,
d
a
a
Alfonso Venzo, G. Giorgio Bombi and Valerio B. Di Marco*
Received 7th November 2007, Accepted 4th January 2008
First published as an Advance Article on the web 14th February 2008
DOI: 10.1039/b717269a
In view of a possible application to Fe and Al chelation therapy, 2-methyl-3-hydroxy-4-
pyridinecarboxylic acid (DT2) was synthesised, and its complex formation, electrochemical and
cytotoxic properties were studied. The complexing properties of DT2 towards Fe(III) and Al(III) were
◦
investigated in aqueous 0.6 m (Na)Cl at 25 C by means of potentiometric titrations, UV-vis
1
+
spectrophotometry, and H NMR spectroscopy. DT2 is a triprotic acid (H
3
L ) having pKa1 = 0.47,
pKa2 = 5.64 and pKa3 = 11.18. The metal–ligand complexes observed in solution and their
corresponding stability constants (log b values) are the following: FeLH (19.38), FeL (16.01), FeLH−1
(
12.28), FeL
AlL (33.74), AlL
complex formation between DT2 and Fe(II) was studied by UV-vis: the weak complex FeLH (log b =
5.8) was detected. DT2 shows a lower complexation efficiency with Fe(III) and Al(III) than that of
2
H
2
(37.29), FeL
3
H
3
(53.41), FeL
3
H
2
(47.99), FeL
3
H (41.21) and FeL
3
(34.1); AlLH (17.43),
2
H
2
2
H (27.6), AlL
3
H
3
(48.72), AlL
3
H
2
(42.67), AlL
3
H (35.8) and AlL
3
(27.92). The
1
other available chelators, but higher than that of its non-methylated analogue 3-hydroxy-4-
pyridinecarboxylic acid (DT0). The electrochemical behaviour of DT2 was investigated by means of
cyclic voltammetry, indicating that the oxidation of the ligand proceeds through a two electron process
with a CECE mechanism. Voltammetric curves suggest that the oxidation or the reduction of DT2 in
vivo is unlikely. According to the thermodynamic data, also the Fe(III)–DT2 complexes do not undergo
redox cycling at physiological pH. Amperometric titrations of solutions containing Fe(III) and DT2 at
pH = 5 indicated the same Fe(III) : ligand stoichiometric ratio as calculated from potentiometric data.
The toxicity of DT2 and of other simple hydroxypyridinecarboxylic acids was investigated in vitro and
no cytotoxic activity was observed (IC50 > 0.1 mM) on cancer cell lines and also on primary human
cells, following a three day exposure.
Introduction
far for the replacement of DFO and L1 is ICL 670A (deferasirox),
which is commercially available in the European Union since 2006
under the trade name “Exjade” (Novartis).
The overload of metal ions may produce several patholo-
gies because of toxic degenerative processes in the involved
5
An ideal chelating agent should display negligible or low
1,2
parenchyma. Fe overload is a common adverse consequence of
the chronic transfusional therapies for thalassemic patients, and Al
overload may be observed in uremic patients. Fe and Al overload in
central nervous system is involved in neurodegenerative disorders
such as Alzheimer’s and Parkinson’s diseases.
toxicity, high stability of the Fe(III) and Al(III) complexes at
physiological conditions, low affinity towards essential metal
ions to reduce toxic side effects, low molecular mass (less than
4
00 Da) to allow oral administration, a low affinity towards
Fe(II) and no redox property to avoid redox activity in vivo. We
Current treatment options for metal ions overload are designed
to remove tissue deposits which cause the toxic effects. At the
present time, chelation therapy is the most efficient therapeutic
7,8
recently proposed hydroxypyridinecarboxylic acids as potential
chelating agents for Fe and Al, as they have several of the above
described requirements for an ideal chelator. They have a low
molecular weight, and they have a very low affinity towards
3
approach. The chelators presently used for Fe and Al overload
therapies, DFO (desferal) and L1 (deferiprone), have several
drawbacks. Multidisciplinary research for alternative molecules is
9
Zn(II). The affinity of 3-hydroxy-4-pyridinecarboxylic acid (DT0,
7
previously indicated as 3H4P, Fig. 1) towards Fe(III) and Al(III)
3–6
actively pursued. The most promising compound proposed so
was much lower than that of L1. This affinity was significantly
increased by the 1-methyl substitution in the pyridinic ring of
DT0, i.e. for 1-methyl-3-hydroxy-4-pyridinecarboxylic acid (DT1,
a
Dipartimento di Scienze Chimiche, Universit a` di Padova, via Marzolo 1,
10
10,11
3
5131, Padova, Italy. E-mail: valerio.dimarco@unipd.it
previously indicated as 1M3H4P, Fig. 1).
To our knowledge,
b
Dipartimento di Scienze Farmaceutiche, Universit a` of Padova, via Marzolo
the electrochemical properties of hydroxypyridinecarboxylic acids
and of their metal–ligand complexes have not been investigated.
As regards toxicity, data are available for only a limited number of
hydroxypyridinecarboxylic acids: DT0 displays negligible toxicity
5
, 35131, Padova, Italy
c
Dipartimento di Istologia, Microbiologia e Biotecnologie Mediche, via
Gabelli 63, 35121, Padova, Italy
d
CNR, Istituto di Scienze e Tecnologie Molecolari, via Marzolo 1, 35131,
12,13
Padova, Italy
towards animals, and it was proposed as an aspirin-like drug.
In
This journal is © The Royal Society of Chemistry 2008
Dalton Trans., 2008, 1689–1697 | 1689

View Article Online
Fig. 1 3-Hydroxy-4-pyridinecarboxylic acid (DT0), 1-methyl-3-hydroxy-
4
-pyridinecarboxylic acid (DT1), 2-methyl-3-hydroxy-4-pyridinecarboxylic
+
acid (DT2), shown in the completely protonated forms (H
3
L for DT2).
addition, 4-pyridoxic acid (2-methyl-3-hydroxy-5-hydroxymethyl-
4
-pyridinecarboxylic acid), the main metabolite of vitamin B6, is
14
non-toxic.
In this paper we evaluated 2-methyl-3-hydroxy-4-
pyridinecarboxylic acid (DT2, Fig. 1) as a new possible
chelating agent for Fe and Al. DT2 is not commercially available
and it was synthesised. Its complexing properties towards Fe(III)
and Al(III) were studied by means of potentiometric, UV-vis
Scheme 1 Synthetic route to DT2 (4).
1
and H NMR measurements. The Fe(II)–DT2 complexation was
4
1
.60 (q, 1H, J = 7.2 Hz, CHCH
3
), 7.20 (br s, 1H, NH), 8.18 (s,
studied by UV-vis. Cyclic voltammetry (CV) experiments were
performed in order to evaluate the electrochemical behaviour of
DT2, and to obtain independent Fe(III)–DT2 complexing data.
The in vitro cytotoxicity of DT0, DT1 and DT2 was assessed on
human cancer cell lines and normal human fibroblasts.
H, CHO).
Synthesis of 4-methyl-5-ethoxyoxazole (3). A suspension of
P O (20 g), Celite (6 g) and MgO (6 g) was vigorously stirred,
2
5
until it became homogeneous, in a 500 mL round bottom flask with
50 mL CHCl at room temperature under N . A solution of ethyl
N-formyl-2-aminopropanoate (4 g, 27.19 mmol) in 50 mL CHCl
1
3
2
3
Experimental
was added drop wise and the stirring was continued for 30 min.
The mixture was heated to reflux for 20 h. A saturated solution
Synthesis of 2-methyl-3-hydroxy-4-pyridinecarboxylic acid (DT2)
(
300 mL) of NaHCO was slowly added and the mixture was
3
Melting points were determined on a Gallenkamp MFB 595
cooled and filtered. The filtrate was extracted with chloroform, the
organic phase was washed with water, and dried with anhydrous
1
0
10M/B capillary melting point apparatus. H NMR spectra were
recorded on a Bruker (300 MHz) spectrometer, using the sol-
vents indicated. Chemical shifts are reported in (ppm) downfield
from tetramethylsilane as internal reference. Coupling constants
are given in Hertz. In the case of multiplets, chemical shifts
were measured starting from the approximate center. Elemental
analyses were performed on a 240B Perkin-Elmer elemental
analyser; results agreed with the calculated values within ± 0.4%.
Mass spectra were obtained on a Applied Biosystems Mariner
Na
Yield 50%; R
d 1.32 (t, 3H, J = 7.2 Hz, CH
H, J = 7.2 Hz, CH CH ), 7.25 (s, 1H, 2-H).
2
SO . On removing the solvent, a dark brown oil was obtained.
4
1
f
0.8 (chloroform–methanol, 9 : 1); H NMR (CDCl
3
)
2
CH ), 2.00 (s, 3H, CH ), 4.05 (q,
3
3
2
2
3
Synthesis of 2-methyl-3-hydroxy-4-pyridinecarboxylic acid (4).
In a100mL flask, 3g(23.6mmol) of oxazole 3 were reacted with 3 g
2.86 mL, 41.23 mmol) of acrylic acid at room temperature. After
(
System 5220 LC/Ms (nozzle potential 250.00). CHCl was made
3
1
5 min the formation of a precipitate was observed which resulted
anhydrous by molecular sieves 4A, 8 to 12 mesh. Starting materials
and reagents were purchased from Aldrich Chimica, and solvents
from Carlo Erba, Fluka and Lab-Scan.
◦
complete after some hours at 4 C. The thin white crystalline
product was collected, re-crystallized from water and dried under
vacuum at 40 C. Yield 60%; mp 320–325 C (decomp.) (lit. 329–
◦
◦
15
◦
1
3
32 C); H NMR (D
2
O, NaOH) d 2.18 (s, 3H, CH ), 6.81 (d, 1H
3
Synthesis of ethyl 2-aminopropanoate hydrochloride (1) (see
also Scheme 1). A solution of commercial DL-alanine (3 g,
2
2
13
J
6,5 = 5.72 Hz, 6-H), 7.30 (d, 1H, J5,6 = 5.72 Hz, 5-H); C NMR
O, NaOH) d 20.12 (CH ), 121.14 (5-C), 131.57 (4-C), 135.95
6-C), 152.01 (2-C), 157.60 (3-C), 178.76 (CO); HRMS [MH ] m/z
NO 153.1213); elemental analysis: calc.
: C, 54.90; H, 4.61; N, 9.15; found: C, 54.91; H, 4.54;
(
(
D
2
3
3
3.67 mmol) in 150 mL absolute ethanol and 3 mL HCl 37%
+
was refluxed for 5 h. The solvent was evaporated and the oily
1
154.0448 (calc. for C
for C NO
N, 9.06%.
7
H
7
3
residue was dried under vacuum. Yield 95%; H NMR (D
2
O) d
7
H
7
3
1
4
.41 (t, 3H, CH
.47 (q, 1H, CH).
2
CH
3
), 1.70 (d, 3H, CHCH
3
), 4.28 (q, 2H, CH ),
2
Synthesis of ethyl N-formyl-2-aminopropanoate (2). 4 g (26.04
mmol) of ethyl 2-aminopropanoate hydrochloride and 13 mL
78.12 mmol, 11.58 g) of triethyl orthoformate were refluxed
Thermodynamic study
(
All potentiometric measurements were performed using a Ra-
diometer ABU93 Triburette apparatus. UV-vis spectra were
recorded using a Perkin-Elmer Lambda 20 spectrophotometer,
for 1.5 h. The produced ethanol and excess orthoformate were
removed by atmospheric distillation to give a light orange oil. Yield
1
1
8
(
5%; H NMR (CDCl
3
) d 1.28 (t, 3H, J = 7.2 Hz, CH
2
CH
3
), 1.43
CH ),
H NMR spectra with a Bruker DRX-400 spectrometer operating
d, 3H, J = 7.2 Hz, CHCH
3
), 4.21 (q, 2H, J = 7.2 Hz, CH
2
3
at 400.13 MHz. All analyte concentrations were expressed in the
1
690 | Dalton Trans., 2008, 1689–1697
This journal is © The Royal Society of Chemistry 2008

View Article Online
Table 1 UV-vis measurements
3
3
−3
−1
−1
Solution
DT2
10 C
M
/m
10 CDT2/m
pH
Path cell/cm
1
k/nm
10 e/mol kg cm
—
0.0976
0.24–1.58
300, 312, 315, 330
7.0 ± 0.2 (H
at 315 nm)
1.65 ± 0.04 (FeLH)
3
L), 6.9 ± 0.1 (H
2
L)
(
3+
Fe + DT2
0.481
0.178
0.455
0.148
0.25–1.82
2.56–3.56
0.5
1
420
300, 315, 330
3+
Al + DT2 (I series)
8.38 ± 0.04 (AlLH), 15.2 ± 0.3
2 2
(
AlL
H
) (at 330 nm)
3+
Al + DT2 (II series)
0.253
203
0.564
0.986
3.48–5.38
2.25–3.69
0.1
1
2+
II
Fe + DT2
407
1.26 ± 0.07 (Fe LH)
−
1
molality scale (mol kg of water, m). For potentiometric and UV-
vis measurements, working solutions of HCl (0.2 m), NaOH (0.1
the low complex formation rate was attributed to the inertness of
3
+
Al(H
2
O)
6
and to the intramolecularly bonded phenolic hydrogen
and 0.2 m), AlCl
m, containing HCl 0.3 m), were prepared and standardised as
described previously.
a 0.006 m working solution (containing NaOH 0.01 m to ensure
3
(0.04 m, containing HCl 0.2 m) and FeCl (0.05
3
of the ligand. Experimental details regarding the treatment of the
slow kinetics during the titrations are reported.
23
16,17
DT2 was used as synthesized to prepare
UV-vis measurements are summarised in Table 1. When the
pH of the solution was below 2, its value was computed from
3
+
+
water solubility). The Mohr’s salt (Carlo Erba, 99% min., Fe
.01%) was used for the UV-vis scans of Fe(II)–DT2 solutions.
The ionic strength of all solutions was adjusted to 0.6 m (0.594 M)
≤
the stoichiometric concentration of HCl, because the [H ] mod-
0
ifications produced by the other species were negligible under
these conditions. In other cases the pH was measured with the
same electrodes and procedures as for potentiometric titrations.
Solutions for the Fe(II)–DT2 complexation study were degassed
16
1
(
Na)Cl. Solutions for H NMR measurements were prepared by
dissolving weighed amounts of DT2, AlCl (Carlo Erba, 98% min.)
and NaCl in D O (Aldrich, 99.9% atom D). Internal reference was
Me SiCD CD COONa (Aldrich).
3
2
by N prior iron addition. More UV-vis spectra were collected
2
3
2
2
sequentially for the same sample to ensure that no Fe(II) oxidation
occurred during the wavelength scan. In the investigated acidic
pH range, solutions were stable for at least 15 min (longer time
stability was not investigated). The pKa1 for the ligand (Table 1),
the log b values for some metal–ligand complexes (Table 2) and
the values of e (molal absorbtivity coefficient; selected values are
given in Table 1) were computed by the program PITMAP at the
given wavelengths.
◦
Potentiometric measurements were carried out at 25.0 ± 0.1 C;
duplicate potentiometric measurements were performed using
two different glass electrodes (Radiometer pHG201 and VWR
18,19
662–1792) and an Ag/AgCl/0.6 m NaCl reference electrode
with a J-shaped junction. Besides glass electrode calibration,
base standardisation, and ligand standardisation experiments,
titrations of metal–ligand mixtures were performed: metal ion to
ligand ratios were 1 : 1, 1 : 2, 1 : 3, 1 : 4, 1 : 5, 1 : 6, 1 : 7 and 1 : 8,
16
1
◦
H NMR spectra were obtained at 25 C unless differently
specified. Chemical shift values are given in d units with reference
to internal Me SiCD CD COOD. Suitable integral values for the
−
4
−4
Fe(III) concentrations ranged from 3.1 × 10 to 9.1 × 10 m,
and Al(III) concentrations ranged from 3.2 × 10 to 9.8 × 10
m. All stability constants were calculated using the computer pro-
−
4
−4
3
2
2
proton spectra were obtained by a pre-scan delay of 10 s. The
assignment of the proton resonances was performed by standard
chemical shift correlations and NOESY measurements when
necessary. Spectra were collected for DT2 at various pD values
20
gram PITMAP. The stability constants for metal/hydroxo com-
plexes have been taken from the literature: log bFeOH = −2.87,
17
log bFe(OH) = −6.16, log bFe(OH) = −12.16, log bFe(OH) = −22.16,
2
3
4
21
log bFe (OH) = −2.9, log bFe (OH) = −6.3, log bFe (OH) = −48.9,
in D O solutions containing NaCl 0.6 m and the free ligand,
2
2
2
3
4
12
34
log bAlOH = −5.52, log bAl(OH) = −11.3, log bAl(OH) = −17.3,
or the ligand + Al(III). The pH was measured with a Crison
5014 combined glass electrode previously calibrated in buffered
solutions at pH = 4 and 7. In order to correct for isotopic
and solvent effects, 0.41 pH units were added to the pH meter
2
3
log bAl(OH) = −23.46, log bAl (OH) = −13.57, log bAl (OH)
=
4
2
3
4
13
32
2
21
−
109.2. Some of the above constants (given by Flynn ) refer
to ionic strengths which differ significantly from (Na)Cl 0.6 m.
Nevertheless, in the data elaboration of our metal–ligand titrations
we found that metal-hydroxo species have always a negligible
concentration, apart at pH values very close to the start of
hydroxide precipitation (points discarded in the calculation).
In the case of Al(III)–DT2 mixtures, after the addition of
the metal ion and of the ligand to an acidic solution, the
measured e.m.f. drifted and reached a constant value only after
ca. 1 h, suggesting a slow complex formation. The kinetics of
the Al–ligand reaction remained quite slow (equilibration within
ca. 15 min after each NaOH addition) at all examined pH
values. A similar behaviour was observed also for the Al(III)
25
readings, i.e. pD rather than pH values were measured.
a
Table 2 Acidic properties of DT2
Species
pK
a
n
+
H
H
HL
3
2
L
L
0.47 ± 0.02
5.640 ± 0.001
11.178 ± 0.005
3
9
9
UV-vis
Potentiometry
Potentiometry
−
a
For potentiometric data, n is the number of experiments from which
the data were obtained. Values are shown as the mean and standard
deviation of the mean. The latter is merely a measure of the repeatability
of our results, not of their accuracy. For UV-vis data, n is the number of
wavelengths from which the data were obtained; a weighted mean and the
corresponding weighted standard deviation were calculated from the data.
7,10,23
complexes of other hydroxypyridinecarboxylic acids.
finding is in agreement with the results obtained by Baiocchi
and Mentasti in a study of the kinetics of the reaction between
This
24
Al(III) and 3-hydroxy-2-pyridinecarboxylic acid; in that paper,
This journal is © The Royal Society of Chemistry 2008
Dalton Trans., 2008, 1689–1697 | 1691

View Article Online
Electrochemistry
at room temperature originating a not isolable bicyclic Diels–
Alder adduct which promptly changed into the DT2 (4) which
precipitated (40%).
Cyclic voltammetry (CV) experiments were carried out with a
home-made three-electrode cell having a 3-mm glassy carbon (GC)
working electrode, a SCE reference and a Pt counter electrode. The
potential waveform applied to the working electrode was generated
by an AUTOLAB PGSTAT100 (Quebec, Canada) potentiostat.
A Crison GLP 21 pH-meter and an Ingold combined glass
electrode were used for pH measurements. The GC electrode was
mechanically polished with SiC 4000 grits, rinsed with water, and
conditioned at −1.0 V for 10–20 s. Solutions were degassed for
Acidity constants of DT2
Potentiometric titration of the ligands alone allowed the compu-
tation of the pK
a
value for the HL and H L form (for simplicity,
2
charges will be generally omitted from the formulae, except in
the Tables), corresponding to the deprotonation of the NH and
OH groups, respectively. Values are reported in Table 2. pKa2 and
pKa3 values for DT2 can be compared with the ones obtained
5
min prior each experiment at the chosen pH.
7
for DT0, 4.71 and 10.29, respectively. The significant (almost 10-
Cytotoxicity tests
fold) decrease of the acidity in DT2 with respect to DT0 can be
explained by the inductive effect of the electron-donating methyl
group. Very similar effects of the methyl substitution are reported
in the literature. For example, 3-hydroxypyridin-4(1H)-one and 3-
Human cancer cell lines (OVCAR, OE33, A549, HeLa, obtained
from American Type Culture Collection, ATCC) and normal
human fibroblasts were cultured in DMEM medium (Gibco). In
all cases the growth medium was supplemented with penicillin
hydroxy-2-methylpyridin-4(1H)-one OH groups have pK
a
= 9.01
−
1
−1
(
100 units mL ), streptomycin (10 lg mL ), and 10% heat
◦
32
◦
33
(
25 C, KNO
3
0.1 M) and 9.80 (25 C, NaCl 0.15 M), and
inactivated foetal calf serum (Euroclone). All the cell cultures were
pyridine and 2-methylpyridine NH groups have pK
a
= 5.33 and
◦
let grown in 25 mL Falcon bottles at 37 C under continuous flow
◦
34,35
6
.06 (25 C, NaClO
4
0.1 M),
respectively.
of a 5% carbon dioxide and moisture enriched atmosphere.
The cytotoxic activity of DT0, DT1 and DT2, was
determined using a standard 3-[4,5-dimethylthiazol-2-yl]-2,5-
diphenyltetrazodium bromide (MTT)-based colorimetric assay
Accurate acidity constants at pH values lower than ca. 1.5–2
cannot be obtained from potentiometric measurements, because
in these conditions the pH modification due to the acid–base
equilibria is negligible. Therefore, the pK
a
value for the H L
3
26
(
Sigma), according to the method described by Alley et al.
form, corresponding to the deprotonation of the COOH group,
was determined by means of UV-vis measurements (Table 2). Its
high acidity is in agreement with values previously observed for
Quadruplicate cultures were employed for each treatment. The
5
cells (5 × 10 /well) were seeded in 96-well microplates in the ap-
◦
propriate growth medium (100 lL) and then incubated at 37 C in a
7,10
other hydroxypyridinecarboxylic acids, and it can be explained
5
% carbon dioxide controlled atmosphere. After 24 h the medium
by both the inductive effect exerted by the positively charged
nitrogen, and by the intramolecular bond which is formed when
was removed and replaced with a fresh medium containing the
compounds to be evaluated, at concentrations ranging between
7
the carboxylate deprotonates.
0.01 and 100 lM (all the tested compounds were initially dissolved
in DMSO to obtain a 1 M stock solution and then further diluted
in Tris HCl, pH = 8). After 72 h, each well was supplemented with
Metal–ligand complexes: potentiometric results
−
1
1
0 lL of a 5 mg mL MTT solution (3-(4,5-dimethylthiazole-2-
Potentiometric data were elaborated in a sequential manner, as
yl)-2,5-diphenyltetrazolium bromide, Sigma Chemical Co.). After
further 5 h, 100 lL of a sodium dodecyl sulfate (SDS) in HCl
7
described previously, to avoid the difficulties produced by the
strong correlation between the parameters to be optimised in
the presence of a large number of metal–ligand species. The
stoichiometry and the values of the stability constants of the
metal–ligand complexes identified in solution are reported in
Table 3. The distribution diagrams for Fe(III)–DT2 and Al(III)–
0
.1 M solution were added. Finally, after an additional 18 h, the
optical absorbance was measured at 570 nm using a LX300 Epson
Diagnostic microplate reader. The cytotoxic effect of each tested
compound was evaluated by the ratio between the number of living
cells present in the sample and in a blank treated with the solvent
only.
−
4
DT2, as computed for solutions containing 5 × 10 m metal
−
3
ion and 2 × 10 m ligand, are reported in Fig. 2(a) and (b),
respectively.
Results and discussion
In the case of Fe(III), the free metal ion predominates only below
pH = 0.5, and upon increasing the pH the sequential formation of
Synthesis of 2-methyl-3-hydroxy-4-pyridinecarboxylic acid (DT2)
FeLH, FeL
latter (FeL
minor species except at 1 : 1 ligand-to-metal ratio. The pK
of FeL (5.42, 6.78 and 7.1) can be easily calculated from data
in Table 3 (e.g., pKa(FeL3
2
H
2
, FeL
3
H
3
, and of the deprotonation products of the
DT2 was synthesized following a described synthetic route
to 3-hydroxypyridines via
3
H
2
, FeL
3
H and FeL
3
) is observed. FeL and FeLH−1 are
values
a
5-ethoxyoxazole intermediate
a
27–29
(
Scheme 1).
4-Methyl-5-ethoxyoxazole (3) was easily obtained
3
H
3
from DL-alanine which was first esterified with absolute ethanol
and hydrochloric acid yielding the ester hydrochloride 1 (95%)
and then N-formylated with triethyl orthoformate to compound
2
H
)
= log b(FeL
H
)
− log b(FeL
H )
), and they
H occurs at the pyridinic
3
3
3
3 2
confirm that the deprotonation of FeL
3
3
30
nitrogen. On the contrary, the low pK
3.73) suggest that FeL and FeLH−1 arise from the deprotonation
of a Fe(III)-coordinated water molecule. Precipitation of Fe(OH)
a
values of FeLH (3.37,
31
(85%). The cyclization reaction of 2 was carried out using P
2
O
5
in dry CHCl and in the presence of MgO (50%). In the absence
3
3
of MgO the reaction did not run. Compound 3 and acrylic acid
were mixed directly without solvent and the mixture was stirred
begins at pH between 5 and 7, depending on the stoichiometric
concentrations in the starting solution.
1
692 | Dalton Trans., 2008, 1689–1697
This journal is © The Royal Society of Chemistry 2008

View Article Online
Table 3 Stability constants for Fe(III)–DT2 and Al(III)–DT2 complexes,
not experimentally observed in the presence of a three-fold or
higher excess ligand.
◦
3+
2−
+
at 25 C in aqueous (Na)Cl 0.6 m (reactions: m M + l L + h H
ꢀ
3m−2l+h
2−
M
m
L
l
H
h
). L is the ligand in the completely deprotonated form
(
Fig. 1)
Metal–ligand complexes: UV-vis results
Fe(III)–DT2
Al(III)–DT2
Selected UV-vis spectra for Fe(III)–DT2 and Al(III)–DT2 solutions
at several pH values are reported in Fig. 3(a) and (b), respectively.
All spectra show the typical absorption of substituted pyridinic
rings. The increase of pH causes a bathochromic shift of the
main peak at 315 nm, and a general increase of the absorption
coefficients. These effects are in agreement with the gradual
formation of less positively charged, thus electron-richer, species
at higher pH values. In the presence of Fe(III), two weak charge-
transfer peaks appear between 400 and 500 nm.
Species
log b
n
log b
n
2+
a
MLH
19.38 ± 0.01
1
17.43 ± 0.03
14
3
a
1
—
7.57 ± 0.01
+
ML
16.01 ± 0.04
12.28 ± 0.09
37.29 ± 0.04
8
6
17
MLH−1
—
+
ML
2
H
2
33.74 ± 0.04
14
3
6
8
6
6
6
a
3
3.71 ± 0.08
ML
ML
ML
ML
ML
2
3
3
3
3
H
H
H
H
—
27.6 ± 0.1
3
53.41 ± 0.04
47.99 ± 0.06
41.21 ± 0.09
34.1 ± 0.1
9
9
9
8
48.72 ± 0.04
42.67 ± 0.04
35.8 ± 0.1
−
−
2
2
3
−
27.92 ± 0.06
a
Obtained by UV-vis (see text).
Fig. 3 UV-vis spectra for Fe(III)–DT2 (a) (pH = 0.25, 0.39, 0.60, 0.79,
0
3
.98, 1.18, 1.36, 1.59 and 1.82) and for Al(III)–DT2 solutions (b) (pH =
.48, 4.01, 4.23, 4.33, 4.42, 4.72, 4.82 and 5.38). Other details are reported
in Table 1 and in the Experimental section.
Fig. 2 Distribution diagrams of the most important Fe(III) (a) and Al(III)
◦
(
C
b) species in the presence of DT2 in aqueous (Na)Cl 0.6 m, T = 25 C;
−
4
−3
UV-vis data for Fe(III)–DT2 allowed the determination of the
stability constant of FeLH (Table 3), which cannot be obtained
from potentiometric titrations because this complex is already
formed at very acidic pH values (Fig. 2(a)). In the case of Al(III)–
DT2, UV-vis data were obtained to confirm the potentiometric
M
= 5 × 10 m, CDT2 = 2 × 10 m. Dashed lines show the theoretical
pH for hydroxide precipitation.
Al(III)–DT2 solutions contain the same main species observed
for Fe(III), with a similar pH-dependent distribution. The pK
values of AlL (6.05, 6.90 and 7.84) are similar to those of
FeL as in both cases the deprotonation occurs at the pyridinic
nitrogen. The acidity of AlL (pK
a
3
H
3
results. The stability constants determined for AlLH and AlL
2
H
2
3
H
3
(Table 3) are in reasonable agreement with the corresponding
potentiometric ones, thus suggesting the absence of systematic
errors in the results.
UV-vis spectra of Fe(II)–DT2 solutions (not reported), similarly
to those of Fe(III)–DT2, show a weak complex absorption centered
at 407 nm and an even weaker absorption at 480 nm. By increasing
the pH, the intensity of these peaks increases, and their shape
2
H
2
a
= 6.10) does not indicate
whether the deprotonation occurs at the pyridinic nitrogen or at
the Al(III)-coordinated water. At basic pH values all the complexes
are destabilised owing to the formation of Al(OH)
investigated concentration range, the precipitation of Al hydroxide
predicted on the basis of the solubility product of Al(OH) ) was
4
. In the
(
3
This journal is © The Royal Society of Chemistry 2008
Dalton Trans., 2008, 1689–1697 | 1693

View Article Online
remains unchanged. Data elaboration allowed to detect one
ferrous complex, FeLH (hereafter called Fe LH, to distinguish
a number of diastereoisomers in solution (AlL
eight diastereoisomers). By increasing the temperature, the broad
2
H may form up to
2
II
it from the corresponding ferric FeLH complex), with a stability
constant log b = 15.8 ± 0.2.
signal at d 2.4 becomes narrower (inset of Fig. 4(b)), indicating that
◦
at 25 C AlL
2
2
H diastereoisomers interconvert with a slow kinetics
with respect to the NMR time scale. As the peaks of the free ligand
are narrow at all temperatures, the AlL isomers interconversion
1
2
H
2
Al(III)–ligand complexes: H NMR results
should proceed through an intramolecular mechanism.
At pD 5.9 the signals of AlLH almost disappear (as clearly
1
In the H NMR spectra of DT2 in D
2
O (pD = 3.1), the singlet
of the CH
3
protons is observed at d 2.68, and the signals of H(6)
visible in the aliphatic region), those of AlL
2
2
H are still observable,
and H(5) are at d 8.08 (d) and 8.05 (d), respectively. All signals
move upfield by increasing the pD value: for example, at pD = 9.0
and other broad multiplets appear. The most significant peaks
can be seen in the aliphatic region at d 2.36, 2.34, 2.25 and 2.23.
CH
3
, H(6) and H(5) resonate at d 2.46, 7.56 and 7.96, respectively.
These signals are attributed to AlL
are not distinguishable by NMR, and are collectively referred
as “AlL ”). These four signals are independent from each
3
H
3
, AlL
3
H
2
and AlL
3
H (which
The chemical shift change is higher for H(6) which is closest to the
deprotonating pyridinic nitrogen.
3
H
x
In the presence of Al(III), the resonances of the free ligands
are observed together with new peaks due to the complexes at all
the investigated pD values (Fig. 4). At pD = 3.1 the new signals
are observed at d 8.03 (partially overlapped with the peaks of the
free ligand), 7.81 and 2.62, having the same multiplicity and very
similar coupling constants as the signals of the free ligand. This
set of signals belongs clearly to a single species, which according
to the potentiometric data is AlLH.
other, i.e. they are not two doublets, as demonstrated by spectra
obtained at different operating frequency (inset of Fig. 4(c)). The
presence of four methyl signals for one species is due to the
presence of diastereoisomers for AlL
it may chelate the metal, forming AlL
configurations. If the three ligands of AlL
ion in the same configurations, a symmetrical diastereoisomer
(AlL x-sym) forms, and the three CH groups are chemically
equivalent. If one ligand chelates differently from the other two, an
unsymmetrical diastereoisomer (AlL
CH groups give rise to different signals in the NMR spectrum.
3
H
x
. DT2 is asymmetric, thus
, in two different spatial
chelate the metal
3
H
x
3
H
x
3
H
3
Peaks of AlLH are still observed at pD 4.6, where a new set
of broad aromatic signals appears. In the aliphatic proton signal
region a very broad signal appears at ca. d 2.4, too. These signals
3
Hx-unsym) forms, and the three
3
are attributed to AlL
2
H
2
. Their broadness is due to the presence of
The four signals show the same intensity: this can be explained
1
−4
−4
Fig. 4 H NMR spectra of D
“
2
O solutions containing Al(III) (3.00 × 10 m) and DT2 (9.50 × 10 m), pD = 3.1 (a), 4.6 (b), 5.9 (c), 9.0 (d). “L”, “*”,
2 2 3 x
+ and “#”, denote the signals of the free ligand and of three different complexes (AlLH, AlL H and AlL H ), respectively. “i” are impurities of DT2.
The sensitivity of spectra in the aliphatic zone was enhanced by a factor of 5. The inset to (b) shows the spectrum of Al(III)–DT2 at pD 4.6 recorded at
◦
5
0 C. The inset to (c) shows the spectrum of the same solution at pD 6.2 with a 300 MHz instrument.
1
694 | Dalton Trans., 2008, 1689–1697
This journal is © The Royal Society of Chemistry 2008

View Article Online
by supposing that the formation microconstant of AlL
AlL x-unsym are very similar, so that their ratio in solution is 1 : 3.
An identical finding (four signals of AlL with the same integration
ratios) was observed for Al(III)–DT1 solutions at similar pD
3
H
x-sym and
acids are weaker Fe(III) and Al(III) chelators than L1 at all pH
values. It is also important to note that DT2 forms significantly
more stable complexes with Fe(III) than DT0 does, especially at
slightly acidic and neutral pH values, the pFe difference being more
than one log unit. On the other hand, the same methyl substitution
did not produce any increase in the Al(III) chelation strength, as
pAl values of DT2 are practically identical to those of DT0. We
do not yet have an explanation for the different effect of the 2-
methyl substitution for the two metal ions. The study of other
alkylated 3-hydroxy-4-pyridinecarboxylic acids, now in progress,
should better clarify this open question.
3
H
3
1
0
values.
At pD = 9.0 the NMR analysis demonstrates that a significant
fraction of the ligand is still bound to the metal ion, as expected
on the basis of potentiometric results. The symmetrical and
3
x
unsymmetrical diastereoisomers of AlL H are probably still
present, but aliphatic signals collapse into a more intense signal
at d 1.93 and a smaller peak at d 2.24. Also the aromatic region
shows a more intense multiplet centered at around d 7.7 and a
less intense one at d 7.90. According to potentiometric results,
At the same time, the increase of metal–ligand complex stability
due to methyl substitution depends on its ring position, being more
significant if the methyl is bound at position 1 (DT1) rather than at
position 2 (DT2). This difference can be explained by comparing
3
x
at pD 9.0 practically all AlL H is deprotonated (Fig. 2(b)): this
observation can explain the significant high-field shift of AlL
signals at pD 9.0 with respect to pD 5.9.
3
H
x
the pK value of the OH group for the two ligands: for DT1 this
a
10
value (6.63) is 4.5 log units lower than for DT2, thus indicating
that in DT1 the proton competes less efficiently with the metal
ion than it does in DT2. Anyway, the study of other 3-hydroxy-
4-pyridinecarboxylic acids should be performed to allow a better
understanding of this result.
Metal–ligand complexes: discussion
As previously mentioned, the methylation of the pyridinic ring at
the nitrogen increased the affinity of the ligand towards Fe(III) and
10,11
II
II
Al(III).
Here, the effect of the 2-methyl substitution has to be
From the stability constant of the Fe LH complex, the pFe
evaluated. To this aim, pM vs. pH (pM = −log[M]) was computed
for DT2, DT1 and DT0 (Fig. 5): higher pM values indicate
stronger metal–ligand complexes. Computation is performed at
value in the presence of DT2 is computed to be slightly larger than
6 at all pH values, i.e. DT2 is expected to bind only negligible
amounts of Fe(II) in vivo.
−
6
metal and ligand stoichiometric concentrations (10 m metal ion
−
5
36
and 5 × 10 m ligand) which are representative of physiologic
Electrochemistry of DT2
conditions. The pFe and pAl plots for L1 at the same conditions
speciation data from Clarke and Martell ) are reported for com-
37
The electrochemical behaviour of DT2 was studied by CV at
different pH values (3, 5, 7 and 12) and scan rates in the range 50–
(
parison. Fig. 5 demonstrates that all hydroxypyridinecarboxylic
−
1
8
00 mV s . CV measurements demonstrated diffusive irreversible
1
/2
waves. From the slope of a I vs. v plot the value of the
p
experimental current function W can be evaluated:
1
/2
W = I
p
/Am C
where I
p
is the peak current, v is the scan rate, A is the electrode
area and C is the concentration of DT2. The dependence of W
on pH is shown in Fig. 6. The number of exchanged electrons
in the ligand oxidation can be determined by comparing the W
values of the ligand with those of ferrocyanide (W = 0.71 A
−
1/2
1/2
−2
−1
V
s
cm M ) and 1,4-hydroquinone (e.g. at pH = 3,
−1/2 1/2 −2 −1
W = 1.59 A V
s
cm M ), namely of a purely one- and
Fig. 6 Experimental current function W vs. pH for DT2. E
p
values (mV)
Fig. 5 pFe (a) and pAl (b) plots for several ligands (see text); C
M
= 10−
6
refer to v = 200 mV s . Experimental conditions: Britton–Robinson
−1
m, Cligand = 5 × 10− m.
5
36
buffer; CDT2 = 3 × 10 M.
−3
This journal is © The Royal Society of Chemistry 2008
Dalton Trans., 2008, 1689–1697 | 1695

View Article Online
38
two-electron process, under the same experimental condition.
DT2 shows an intermediate behaviour between the one- and two-
electron systems, indicating a two-electron process conditioned
by acid–base equilibria and coupled chemical reactions. The W
variation with C (not reported here) confirms the presence of
chemical equilibria in solution. Similar findings were observed by
38
Terriero et al. in an electrochemical investigation of the DT2-like
salicylic acid.
On the basis of the above results a CECE mechanism for
the oxidation of DT2 may be suggested. The electrochemical
irreversibility of this reaction does not allow to obtain a value
for its standard reduction potentials. However, the high values of
E
p
of the free ligand at physiological pH (Fig. 6: E
p
= 1.02 V vs.
−
1
SCE at pH 7 and 200 mV s ) suggest that DT2 cannot undergo
redox reactions at in vivo conditions.
Electrochemistry of Fe(III) complexes with DT2
The complex formation with Fe(III) was studied with the CV
technique at pH = 2 and 5. The Britton–Robinson buffer was not
used to avoid significant Fe(III) complexation with phosphate ion.
At pH = 2 the buffer capacity of water solutions is high enough
to maintain a constant pH value during the CV measurements. At
pH = 5 acetic buffer was used.
−3
Fig. 8 CV at rotating disk electrode of Fe(III) (CFe(III) = 3 × 10 M) in the
presence of different amounts of DT2 at pH 2; rotation speed = 500 rpm.
Other experimental conditions as in Fig. 6. Inset: titration curves obtained
from the steady state waves of Fig. 8. Reading potential: 0.32 V (᭹); 1.38 V
(᭺).
The amperometric titration of DT2 with Fe(III) performed at
DT2 stoichiometry is 3 : 2.5. According to potentiometric data,
the stoichiometric ratio at these conditions should be 3 : 2.9.
The disagreement is probably due to the different ionic strength
employed to obtain the two sets of data and to a different pH scale
definition.
pH = 5 by reading I
p
at E = E
p
is reported in Fig. 7 together with
the corresponding CV waves. The complex exhibited a 1.2 : 3 Fe :
DT2 stoichiometry. This ratio agrees very well with the description
obtained from potentiometric data, according to which the mean
stoichiometric ratio at pH = 5 is 1.19 : 3.
The reduction of the Fe(III)–DT2 complexes leads to the
II
formation of a mixture of free ligand and of Fe LH. At pH = 2 and
at the chosen Fe(III) and DT2 concentrations, thermodynamic data
allow to compute a theoretical potential shift of −104 mV (with
3
+
2+
respect to the potential of free Fe /Fe couple). The experimental
shift evidenced in Fig. 8, −280 mV, is larger because of the presence
of kinetic effects, i.e. because of the irreversibility of the reduction
wave.
39
◦
According to Merkofer et al., metal complexes with E values
larger than 0.39 V or smaller than 0.28 V cannot redox cycle in vivo.
Our thermodynamic data allow to calculate the standard reduction
potentials of the Fe(III)–DT2 complexes, see Table 4. Values cannot
be considered exact, because they were obtained summing the
◦
3+
2+
E of free Fe /Fe which refers to zero ionic strength and the
contribution given by our thermodynamic constants which refer
◦
to NaCl 0.6 m. Nevertheless, E values in Table 4 indicate that
redox cycling is unlikely for Fe(III)–DT2 solutions at physiological
pH, where FeL
Fig. 2(a)).
3
3
H and its deprotonation products predominate
Fig. 7 CV scans relative to additions of Fe(III) to DT2 (CDT2 = 3 ×
(
−
3
−1
1
0
M). Experimental conditions: v = 200 mV s ; acetic buffer at pH 5.
Inset: titration curve in terms of I of DT2 vs. Fe(III) concentration at
p
pH 5. Reading at E = E
p
.
In vitro cytotoxic activity
The amperometric titration of Fe(III) with DT2 performed at
pH = 2 is reported in Fig. 8. At this pH value the solutions
containing Fe(III) and DT2 may be studied with a rotating disk
electrode which gives two advantages: (i) elimination of reaction
products; (ii) possibility of monitoring the complex stoichiometry
from both the decrease of the Fe(III) wave at E = 0.32 V and
the increase of the DT2 wave, after the end-point, at E = 1.38 V
Analyzing the percentage of living cells with respect to cells ex-
posed to DT0, DT1, or DT2, it was apparent that each compound,
when used at concentrations up to 100 lM, showed no cytotoxic
activity on the investigated cancer cell lines (OVCAR, OE33, A549,
HeLa) and also on primary human fibroblasts, following a three-
day exposure. Concentrations higher than 100 lM could not be
investigated because of the toxic effects due to the pH variation in
the culture medium (produced by the solvent additions). Therefore
(
see Fig. 8). The inset of Fig. 8 indicates that at pH 2 the Fe :
1
696 | Dalton Trans., 2008, 1689–1697
This journal is © The Royal Society of Chemistry 2008

View Article Online
◦
Table 4 Standard reduction potentials (E ) of Fe(III)–DT2 complexes.
3 C. Hershko (Guest Editor), Sem. Hematol. 2005, 42, Issue 2, Supple-
◦
For the E computation, pK
a
values of DT2 were taken from Table 2,
log b values of Fe(III)–DT2 complexes were taken from Table 3, log b of
ment 1.
4 M. J. Cunningham and D. G. Nathan, Curr. Opin. Hematol., 2005, 12,
II
◦
Fe LH = 15.8, E Fe3+ /Fe2+ = 0.771 V
129.
5
R. C. Hider and T. Zhou, Ann. N. Y. Acad. Sci., 2005, 1054, 141.
◦
reaction
E /V (vs. NHE)
6 R. A. Yokel, Coord. Chem. Rev., 2002, 228, 97.
7
V. B. Di Marco, R. A. Yokel, M. G. Ferlin, A. Tapparo and G. G.
III
−
II
Bombi, Eur. J. Inorg. Chem., 2002, 2648.
8 V. B. Di Marco, R. A. Yokel, H. Li, A. Tapparo and G. G. Bombi,
Inorg. Chim. Acta, 2004, 357, 3753.
9 V. B. Di Marco, A. Tapparo, A. Dolmella and G. G. Bombi, Inorg.
Chim. Acta, 2004, 357, 135.
10 V. B. Di Marco, A. Dean, M. G. Ferlin, R. A. Yokel, H. Li, A. Venzo
and G. G. Bombi, Eur. J. Inorg. Chem., 2006, 1284.
11 V. B. Di Marco, A. Dean, R. A. Yokel, H. Li and G. G. Bombi,
Polyhedron, 2006, 26, 3227.
2 H. Tandeloo, F. Broekman, V. Stroer and J. Siemelink, Neth. Pat. Appl.,
70 11,749, 1972.
Fe LH + e → Fe LH
+0.56
−0.05
−0.64
+0.16
−0.13
−0.62
−1.03
−1.42
III
−
II
Fe L + e + H
2
O → Fe LH + OH
III
−
II
Fe LH−1 + e + 2 H
2
O → Fe LH +2 OH
II
III
−
Fe L
2
3
3
3
3
H
H
H
2
3
2
+ e → Fe LH + HL
− II
III
Fe L
+ e → Fe LH + 2 HL
III
−
II
Fe L
+ e + H
2
O → Fe LH + 2 HL + OH
III
−
II
Fe L
H + e + 2 H
2
O → Fe LH + 2 HL + 2 OH
III
−
II
Fe L
+ e + 3 H
2
O → Fe LH + 2 HL + 3 OH
1
we could not calculate an IC50 value (concentration of chemical
resulting in 50% inhibition of cell growth).
1
1
3 W. O. Foye and J. M. Kauffman, Chim. Ther., 1967, 2, 462.
4 H. G. Kraft, L. Fiebig and R. Hotovy, Arzneimittelforschung (Drug
Res.), 1961, 11, 922.
Conclusions
15 T. Matsuo and T. Miki, Chem. Farm. Bull., 1972, 20, 806.
1
1
1
6 V. B. Di Marco, A. Tapparo and G. G. Bombi, Ann. Chim. (Rome),
999, 89, 397.
7 V. B. Di Marco, A. Tapparo and G. G. Bombi, Ann. Chim. (Rome),
001, 91, 595.
1
DT2 was synthesised, and it was evaluated as new potential
chelating agent for Fe(III) and Al(III). DT2 forms strong complexes
in aqueous solution with both metal ions, and it binds Fe(II)
very weakly. The CV measurements and the thermodynamic data
suggest that DT2 and its Fe(III) complexes cannot undergo redox
reactions at in vivo conditions. DT0, DT1 and DT2 show no
significant cytotoxic activity on cancer cell lines and also on
primary human cells (IC50 > 0.1 mM), following a three day
exposure.
Despite these several positive properties of DT2, it must be
noticed that its complexes with Fe(III) and Al(III) are weaker than
those formed by L1 at all pH values. Therefore, DT2 is probably
not as effective metal ion chelator in vivo as DFO and L1. The goal
for future work is to prepare hydroxypyridinecarboxylic acids with
increased Fe(III) and Al(III) affinity.
The complexation strength of DT2 towards Fe(III) is signifi-
cantly higher than that of unsubstituted DT0, and slightly lower
than that of DT1. This suggests that the 1,2 bis-alkylation of
the pyridinic ring should lead towards a stronger Fe(III) chelator.
As regards Al(III), only the 1-methyl substitution increased
significantly the affinity of Al(III) complexes, therefore it is less
clear whether further alkylation of pyridinic ring will result in a
significant increase of the complexation strength. The synthesis
of the 1,2-dimethyl derivative of DT0, and the thermodynamic,
electrochemical and cytotoxicity studies on this compound, are in
progress.
2
8 G. J. Janz, Silver–Silver Halide Electrodes, in: Reference Electrodes:
Theory and Practice, ed. D. J. G. Ives and G. J. Janz, Academic Press,
New York–London, 1961.
9 G. Biedermann and L. G. Sill e´ n, Ark. Kemi, 1953, 5, 425.
0 V. B. Di Marco, PhD Thesis, University of Padova, 1998.
1
2
21 C. M Flynn, Jr., Chem. Rev., 1984, 84, 31.
¨
2
2
2 L. O. Ohman, Inorg. Chem., 1988, 27, 2565.
3 V. B. Di Marco, A. Tapparo and G. G. Bombi, Ann. Chim. (Rome),
1
999, 89, 535.
2
4 C. Baiocchi and E. Mentasti, Ann. Chim. (Rome), 1981, 71, 631.
25 A. K. Covington, M. Paabo, R. A. Robinson and R. G. Bates, Anal.
Chem., 1968, 40, 700.
2
6 M. C. Alley, D. A. Scudiero, A. Monks, M. L. Hursey, M. J. Czerwinski,
D. L. Fine, B. J. Abbott, J. G. Mayo, R. H. Shoemaker and M. R. Boyd,
Cancer Res., 1988, 48, 589.
27 G. Y. Kondrat’eva and C. H. Huang, Dokl. Akad. Nauk SSSR, 1965,
64, 939.
8 B. A. Johnsen and K. Undheim, Acta Chem. Scand., Ser. B, 1983, 37,
27.
29 M. Y. Karpeiskii and V. L. Florent’ev, Russ. Chem. Rev., 1969, 38, 540.
1
2
1
3
3
3
0 T. Chancellor and C. Morton, Synthesis, 1994, 10, 1023.
1 R. G. F. Giles and C. P. Taylor, Arkivoc, 2002, 149–165.
2 R. C. Scarrow, P. E. Riley, K. Abu-Dari, D. L. White and K. N.
Raymond, Inorg. Chem., 1985, 24, 954.
33 D. J. Clevette, D. M. Lyster, W. O. Nelson, T. Rihela, G. A. Webb and
C. Orvig, Inorg. Chem., 1990, 29, 667.
3
4 Y. Faev, V. de Sharnin and V. Shormanov, Zh. Neorg. Khim., 1997, 42,
1
224.
35 K. Kahmann, H. Sigel and H. Erlenmeyer, Helv. Chim. Acta, 1964, 47,
754.
1
3
3
3
6 T. Kiss and E. Farkas, Perspec. Bioinorg. Chem., 1996, 3, 199.
7 E. T. Clarke and A. E. Martell, Inorg. Chim. Acta, 1992, 191, 57.
8 A. A. J. Terriero, J. M. Luco, L. Sereno and J. Raba, Talanta, 2004, 62,
References
2
47.
1
2
N. C. Andrews, N. Engl. J. Med., 1999, 341, 1986.
P. O. Ganrot, Environ. Health Perspect., 1986, 65, 363.
39 M. Merkofer, R. Kissner, R. C. Hider, U. T. Brunk and W. H. Koppenol,
Chem. Res. Toxicol., 2006, 19, 1263.
This journal is © The Royal Society of Chemistry 2008
Dalton Trans., 2008, 1689–1697 | 1697