Halogenated 2′-benzoylpyridine thiosemicarbazone (XBpT) chelators with potent and selective anti-neoplastic activity: Relationship to intracellular redox activity

Article abstract of DOI:10.1021/jm200924c

Iron chelators of the 2′-benzoylpyridine thiosemicarbazone (BpT) class show substantial potential as anticancer agents. To explore structure-activity relationships, new BpT analogues were designed that incorporated halogen substituents on the noncoordinat

Full text of DOI:10.1021/jm200924c

ARTICLE
pubs.acs.org/jmc
Halogenated 20-Benzoylpyridine Thiosemicarbazone (XBpT) Chelators
with Potent and Selective Anti-Neoplastic Activity: Relationship to
Intracellular Redox Activity
Christian Stefani,† Gaya Punnia-Moorthy,† David B. Lovejoy,† Patric J. Jansson,† Danuta S. Kalinowski,†
Philip C. Sharpe,‡ Paul V. Bernhardt,‡ and Des R. Richardson*,†
†Iron Metabolism and Chelation Program, Department of Pathology, University of Sydney, Sydney, NSW, 2006, Australia
‡Centre for Metals in Biology, School of Chemistry and Molecular Biosciences, University of Queensland, Brisbane, Qld 4072 Australia
S Supporting Information
b
ABSTRACT: Iron chelators of the 20-benzoylpyridine thiosemicarbazone (BpT)
class show substantial potential as anticancer agents. To explore structureꢀ
activity relationships, new BpT analogues were designed that incorporated
halogen substituents on the noncoordinating phenyl group (XBpTs). These
XBpT ligands exhibited potent antiproliferative activity with some analogues
exceeding that of the parent BpT compound. Importantly, there was an
appreciable therapeutic index in vitro, as mortal cells were significantly less
affected by these chelators relative to neoplastic cells. The addition of a halogen
led to a halogen-specific increase in the redox potential of XBpTꢀFe complexes.
Probing for chelator-induced intracellular reactive oxygen species (ROS) with
the fluorescent probe, 20,70-dichlorofluorescein, revealed a 1.5ꢀ4.7-fold increase
in fluorescence upon incorporation of Cl, Br, or I to the parent analogues.
Furthermore, an important structureꢀactivity relationship was deduced where the addition of halogens led to a positive correlation
between intracellular ROS generation and antiproliferative activity in the more hydrophilic BpT parent compounds.
’ INTRODUCTION
However, these findings have led to research examining novel Fe
chelators as cancer chemotherapeutics.25,26 More hydrophobic
Fe chelators, such as 3-aminopyridine-2-carboxaldehyde thiose-
micarbazone (3-AP; Figure 1A), have since been investigated as
potential anticancer therapeutics. In fact, 3-AP has been inves-
tigated in several phase I and phase II clinical trials.27ꢀ29
The traditional use of iron (Fe) chelators is for the treatment
of Fe overload diseases.1ꢀ4 However, their use in cancer therapy
has emerged as a potentially fruitful area of investigation.5ꢀ7 Iron
is essential for proliferation, and the increased requirement for Fe
in rapidly dividing cells leads to an enhanced sensitivity of tumor
cells to Fe-deprivation as compared to normal cells. This
increased requirement for Fe is reflected in cancer cells relative
to their normal counterparts by the increased expression of the
transferrin receptor 1, which is involved in Fe uptake from
transferrin (Tf).8
Iron chelators can sequester Fe from biological systems and,
hence, are able to inhibit the activity of Fe-requiring proteins,
including ribonucleotide reductase, a key enzyme that is involved
in the rate-limiting step of DNA synthesis.9ꢀ11 However, the
potent antiproliferative activity of some Fe chelators is likely
mediated by their effects on multiple molecular targets.12ꢀ15
Chelators with potent antiproliferative activity, such as the
thiosemicarbazones, also demonstrate the ability to form Fe
complexes capable of redox cycling.16ꢀ20 This leads to the
generation of reactive oxygen species (ROS) such as hydroxyl
radicals, which induce cellular injury including DNA damage.16ꢀ20
Previous studies using the Fe chelator, desferrioxamine (DFO;
Figure 1A), showed that it induced modest antitumor efficacy
in vitro and in vivo.5,15,16 The poor membrane permeability of
DFO limits its activity and application in cancer therapy.21ꢀ24
Chelators previously developed within our laboratory,
namely, the di-20-pyridylketone thiosemicarbazone (DpT) series,
exhibited potent and selective antiproliferative activity.18 In
particular, di-2-pyridylketone 4,4-dimethyl-3-thiosemicarbazone
(Dp44mT; Figure 1A) demonstrated the highest activity of
all ligands within this series.18 Dp44mT showed marked and
selective activity in vivo, reducing the growth of a murine M109
lung cancer by approximately 50% after 5 days of treatment,
while having little effect on hematological indices.18 In addition,
initial studies indicated that the Fe complex was redox-active
within cells.18 More recently, Dp44mT has shown marked and
selective activity in vivo against a panel of human tumor
xenografts in nude mice.30 However, at high nonoptimal doses,
this agent leads to cardiac fibrosis and weight loss.30 The
electrochemistry of the Fe complexes of the DpT analogues
showed facile interconversion between the FeII and the FeIII
states within the range accessible to cellular oxidants and
Received: July 13, 2011
Published: August 16, 2011
r
2011 American Chemical Society
6936
dx.doi.org/10.1021/jm200924c J. Med. Chem. 2011, 54, 6936–6948
|

Journal of Medicinal Chemistry
ARTICLE
Electrochemistry of the Fe complexes of the BpT analogues
revealed lower FeIII/II redox potentials (+99 to +180 mV vs
NHE) than their DpT homologues.33 The lower electron-with-
drawing property of the phenyl moiety in the BpT series relative
to the pyridyl ring in the DpT series probably contributes toward
the lowered FeIII/II redox potentials of the BpTꢀFe complexes.33
This factor leads to their higher redox activity, as shown by
ascorbate oxidation and benzoate hydroxylation studies, which
demonstrated that the FeIIIꢀBpT complexes were able to redox
cycle and stimulate redox reactions to a greater degree than the
corresponding FeIIIꢀDpT complexes.33
Considering the marked activity of the BpT analogues33 and
to gain further insight into chelator structureꢀactivity relation-
ships, we investigated the impact of incorporating halogens on
the phenyl ring to generate 2-(40-halobenzoyl)pyridine thiose-
micarbazone (XBpT; Figure 1B) analogues. In particular, we
have chosen four BpT analogues of increasing lipophilicity,
namely: BpT, 20-benzoylpyridine 4-methyl-3-thiosemicarbazone
(Bp4mT), Bp4eT, and Bp44mT, each of which were haloge-
nated on the noncoordinating phenyl ring with either F, Cl, Br,
or I (Figure 1B). Halogenation is expected to both enhance the
lipophilicity of the molecule and also to subtly alter the electro-
chemistry of the chelatorꢀiron complex. The major findings
from these studies are the link between antiproliferative
efficacy and redox activity and the structureꢀactivity relation-
ships that pertain to this, namely: (i) the correlation between
lipophilicity and intracellular redox stress and (ii) establishment
of the concept of a threshold for redox activity and its relationship
to antiproliferative efficacy. In addition, the potencies of the
new halogenated compounds are in general superior to the
parent analogues with an appreciable therapeutic index being
observed.
Figure 1. (A) Chemical structures of the iron chelators: 3-AP,
DFO, 2-hydroxy-1-naphthaldehyde isonicotinoyl hydrazone (NIH),
and di-2-pyridylketone 4,4-dimethyl-3-thiosemicarbazone (Dp44mT).
(B) Chemical structures of members of the 20-benzoylpyridine thiose-
micarbazone (BpT) series of ligands and their halogenated counterparts.
The current studies investigated the impact of incorporating halogen
atoms (F, Cl, Br, or I) on the phenyl ring of the BpT series ligands to
generate XBpT analogues, 20-benzoylpyridine 4-methyl-3-thiosemicar-
bazone (Bp4mT) analogues, 20-benzoylpyridine 4-ethyl-3-thiosemicar-
bazone (Bp4eT) analogues, and 20-benzoylpyridine-4,4-dimethyl-3-
thiosemicarbazone (Bp44mT) analogues.
’ RESULTS AND DISCUSSION
Synthesis and Spectroscopy. The XBpT analogues were all
prepared by high-yielding Schiff base condensation reactions
following previous protocols for related thiosemicarbazones
reported from our laboratory.31,33 The compounds were only
sparingly soluble in water. However, they exhibited greater
solubility in polar aprotic solvents such as DMF, MeCN, and
DMSO. The 1H NMR spectral properties of the XBpT ligands
are very similar to those of the BpT analogues.33
Structural Characterization. As representative members of
the XBpT series, the crystal structures of FBp44mT, ClBp44mT,
BrBp44mT, and IBp44mT have been determined (Figure 1S
AꢀD in the Supporting Information). Crystal data are in Table
1S (Supporting Information), and selected bond lengths and
angles are included in Table 2S (Supporting Information). The
most obvious variations are to the CꢀX (X = F, Cl, Br, or I) bond
length due to the increasing covalent radius of the halogen. An
unexpected, but significant, variation across the series is found in
the N3ꢀC13 bond, which increases (from the F through to the I
analogue; Table 2S in the Supporting Information). Interest-
ingly, this was not correlated with any other variations in bond
lengths. The C13ꢀS1 bonds are consistent with double bonds,31
while the N3ꢀC13 bond is typical of a thioamide (intermediate
bond order). In each structure, intramolecular H-bonding
reductants (∼+100 to +250 mV vs NHE),31 making them ideal
catalysts of Fenton chemistry. This suggests that the antitumor
activity of these chelators is due, at least in part, to both their
ability to bind intracellular Fe and form redox-active complexes,
inducing oxidative damage.31,32
Substitution of the noncoordinating 2-pyridyl moiety on the
DpT series with a phenyl ring generated the 20-benzoylpyridine
thiosemicarbazone (BpT; Figure 1B) series. The BpT series
exhibited enhanced antiproliferative efficacy relative to their DpT
homologues.33 In particular, 20-benzoylpyridine 4-ethyl-3-thio-
semicarbazone (Bp4eT; Figure 1B) demonstrated the highest
activity of all ligands within this series in vitro (IC50, 0.002
μM).33 Recent studies in vivo using nude mice bearing a human
DMS-53 lung cancer xenograft showed that the BpT series
analogue, 20-benzoylpyridine-4,4-dimethyl-3-thiosemicarbazone
(Bp44mT), can be administered intravenously or orally with
potent antitumor efficacy.34 Importantly, in contrast to Dp44mT,
Bp44mT has the marked advantage of not causing cardiac toxicity
or weight loss, even when administered at much higher doses.34
(N3ꢀH N1) is a feature that fixes the conformation of the
3 3 3
ligand in a way that is unsuitable for coordination through the
N1, N2, and S1 set of donors. This is discussed below when
6937
dx.doi.org/10.1021/jm200924c |J. Med. Chem. 2011, 54, 6936–6948

Journal of Medicinal Chemistry
ARTICLE
Table 1. Partition Coefficients (Log Pcalcd) and FeIII/II Redox
Potentials (MeCN:H2O 7:3) of the XBpT Series Chelators
and Their Ferrous and Ferric Complexesa
partition coefficient
FeII/III redox potential
(mV vs NHE)
ligand
NIH
(log Pcalcd)
2.63
2.19
2.25
2.41
2.81
3.08
3.6
Dp44mT
BpT
113
104
131
137
142
138
79
FBpT
ClBpT
BrBpT
IBpT
Bp4mT
FBp4mT
ClBp4mT
BrBp4mT
IBp4mT
Bp4eT
2.77
2.93
3.33
3.6
93
103
101
104
66
4.13
3.11
3.26
3.66
3.93
4.46
3.14
3.3
Figure 2. Cyclic voltammograms of 1 mM solutions of [Fe(Bp44mT)2]+,
[Fe(FBp44mT)2]+, [Fe(BrBp44mT)2]+, and [Fe(IBp44mT)2]+ (from
top to bottom) showing the impact of incorporation of a halogen atom
on the FeIII/II redox potential. Sweep rate, 100 mV sꢀ1; solvent MeCN:
H2O (70:30) with 0.1 M Et4NClO4.
FBp4eT
ClBp4eT
BrBp4eT
IBp4eT
91
95
94
102
63
Bp44mT
FBp44mT
ClBp44mT
BrBp44mT
IBp44mT
Electrochemistry. The electrochemical properties of the Fe
complexes of the XBpT analogues constituted a crucial part of
this study due to the relevance of this property to their biological
activity.33 Indeed, the antiproliferative activity of related ligands
reported by us previously is likely linked with their capability to
undergo Fenton chemistry upon complexation with intracellular
Fe.20,31,33,35,36 All compounds exhibited totally reversible one
electron FeIII/II couples in MeCN:H2O (7:3) at sweep rates
between 50 and 500 mV sꢀ1. This solvent combination was
chosen to ensure that (i) sufficient solubility, as these com-
pounds are largely insoluble in water, and (ii) to enable compar-
isons with similar compounds (i.e., BpT analogues) reported in
our previous studies where the same solvent was used.33 Other
solvent combinations such as DMF:H2O (4:1) also gave similar
electrochemistry. The redox potentials of the Fe complexes are
collected in Table 1 and may be compared with those of the
[Fe(BpT)2]+ complexes.33
82
3.7
99
3.97
107
4.5
98
a Comparable data for the parent BpT compounds, Dp44mT and NIH,
are also shown. Log Pcalcd values were calculated using the program
ChemBioDraw Ultra v 11.0.1 using Crippen's fragmentation procedure.42
considering the structures of the Fe complexes and their biolo-
gical activity as antitumor agents.
Fe Coordination Chemistry. The FeIII complexes of each
series were isolated and characterized. Their electronic spectral
properties mirror those of the respective [FeIII(BpT)2]+
analogues.33 As relative examples, the crystal structures of
[FeIII(ClBp44mT)2]ClO4 and of [FeIII(FBp44mT)2](ClO4)
3
MeCN were determined, and the complex cations are shown in
Figure 1SE,F (Supporting Information). The coordination
spheres of the Fe centers are very similar, and the bond lengths
are consistent with low spin FeIII complexes.35 There are major
differences in the bond lengths within the N3ꢀC13ꢀS1 group
brought about by deprotonation of N3 and coordination of S1 in
both complexes (Table 2S in the Supporting Information). It is
apparent that C13ꢀS1 takes on single bond character, while
C13ꢀN3 shortens by comparison with the structures of the
neutral free ligands (Table 2S in the Supporting Information).
The anionic ligands are coordinated in a tridentate manner
dominantly in their imine-thiolate resonance form. The dimethy-
lamino groups of the ligands are coplanar with the ꢀC(S)NN
group in these structures and in those of the free ligands. Note
that the conformations of the coordinated ligands (with the three
donor atoms in a syn disposition) are distinctly different to those
of the free ligands. These data demonstrating Fe complexation
are significant, as the antitumor efficacy of this general class of
ligands is due to two major properties, namely, their cellular
Fe-binding efficacy and ability to form cytotoxic, redox-active Fe
complexes.33
The anodic shift in the FeII/III redox potentials of the XBpT
complexes in comparison to their corresponding BpTꢀFe com-
plexes is apparent from Figure 2 and Table 1. As an example, the
cyclic voltammograms of four FeIII complexes, namely,
[Fe(FBp44mT)2]+, [Fe(ClBp44mT)2]+, [Fe(BrBp44mT)2]+,
and [Fe(IBp44mT)2]+ bearing the same terminal substituent
(ꢀNMe2) are shown in Figure 2. Generally, the redox potentials
of the halo-substituted complexes were anodically shifted by
14ꢀ44 mV relative to their BpT counterparts, as would be
expected from the electron-withdrawing, inductive effects of
these substituents. However, smaller than expected shifts of
7ꢀ16 mV were observed within the four halo-substituted com-
plexes of each set (e.g., [Fe(XBpT)2]+/0 X = F, Cl, Br, or I). The
only exception to this latter effect was [Fe(BrBp44mT)2]+/0
,
with an anodic shift of 25 mV relative to its F- analogue.
The effect of the identity of the N4 substituent on the FeIII/II
redox potential was similar within the four halo-substituted com-
plexes of each set (e.g., [Fe(XBpT)2]+/0 X = F, Cl, Br, or I). That
is, methylation, ethylation, or dimethylation atN4 led to a cathodic
shift of 34ꢀ49 mV relative to the [Fe(XBpT)]+ parent complex.
6938
dx.doi.org/10.1021/jm200924c |J. Med. Chem. 2011, 54, 6936–6948

Journal of Medicinal Chemistry
ARTICLE
Table 2. IC50 (μM) Values of XBpT Series Chelators at
Inhibiting the Growth of SK-N-MC Neuroepithelioma Cells
and Mortal MRC-5 Lung Fibroblasts as Determined by the
MTT Assaya
and moderate antiproliferative activity;37 and Dp44mT, a che-
lator with high antiproliferative activity extensively studied for
cancer therapy16,18 (Figure 1A). These results are presented
in Table 2. This study identified all of the N4 substituted
XBpT analogues [i.e., 20-(400-halobenzoyl)pyridine 4-methyl-3-
thiosemicarbazone (XBp4mT), 20-(400-halobenzoyl)pyridine
4-ethyl-3-thiosemicarbazone (XBp4eT), and 20-(400-haloben-
zoyl)pyridine 4,4-dimethyl-3-thiosemicarbazone (XBp44mT)]
as having high antiproliferative activity (IC50, 0.007ꢀ0.024
μM), which was similar to that observed for Dp44mT (IC50,
0.004 μM), but significantly (p < 0.001) more effective than
either DFO or NIH. All of the XBpT ligands examined were
significantly (p < 0.01) more active than DFO (IC50, 22.7 μM),
and excluding FBpT, these halogenated ligands were significantly
(p < 0.01) more active than NIH (IC50, 2.53 μM). Our previous
studies examining the DpT ligands against a wide range of
neoplastic cell types in vitro indicated that these agents show
broad and marked antiproliferative activity against 28 different
neoplastic cell types.30 Additionally, our DpT and BpT chelators
demonstrate antitumor activity against a range of human tumor
xenografts in vivo in nude mice.18,30,34 Hence, we would expect
that the very closely related XBpT series to show similar broad
anticancer activity.
The addition of halogens of increasing atomic weight to the
parent BpT ligand led to the most marked and significant (p <
0.001) increase in antiproliferative activity against SK-N-MC
neuroepithelioma cells, where the IC50 decreased 32-fold from
5.59 (BpT) to 0.17 μM (IBpT; Table 2). This marked enhance-
ment in efficacy is believed to be due to the effect of the halogens
on the overall lipophilicity of the ligand (Table 1). The addition
of Cl, Br, or I to the Bp4mT ligand led to a more modest but
significant (p < 0.01) increase in antiproliferative activity, where
the IC50 decreased 2.4-fold from 0.024 (Bp4mT) to 0.010 μM
(IBp4mT; Table 2). In contrast, halogenation of the more
lipophilic Bp4eT (IC50, 0.012 μM) and Bp44mT (IC50, 0.009
μM) produced ligands of comparable antiproliferative efficacy.
These observations suggested an important structureꢀactivity
relationship. That is, for the BpT and Bp4mT series ligands,
addition of halogens to the noncoordinating phenyl group
generated analogues with lipophilicities optimal for membrane
permeability and antiproliferative activity. However, for the
Bp4eT and Bp44mT series of chelators, an optimal lipophilicity
appeared to have already been achieved. Hence, the addition of a
halogen had little effect on antiproliferative efficacy, which was
already marked for the parent ligands (i.e., Bp4eT and Bp44mT).
Plotting the octanol:water partition coefficients (log Pcalcd) for
the chelators against their antiproliferative activity (IC50) in
SK-N-MC cells demonstrated that an optimal lipophilic range
existed, with the ligands possessing the highest antiproliferative
activity having log Pcalcd values between 3.1 and 4.5, with optimal
efficacy being observed at a log Pcalcd of ∼4 (Figure 3A). It should
be noted that for this analysis the XBpT analogues and the
respective parent compound (BpT) were excluded due to their
low activity. Considering this, it is likely that compounds of this
class with log Pcalcd values outside 3.1ꢀ4.5 will either have
difficulty traversing the cell membrane or become trapped within
membranes due to their high affinity for the lipid environment.
Previous studies have also indicated the importance of the
relationship between optimal lipophilicity and antiproliferative
activity.16,37 In fact, a similar parabolic relationship between the
lipophilicity and the efficacy of iron chelation has been observed
examining a series of tridentate aroylhydrazone ligands.38
IC50 (μM)
ligand
DFO
SK-N-MC
MRC-5
p value therapeutic index
22.7 ( 1.55
2.53 ( 0.18
NIH
Dp44mT
BpT
0.004 ( 0.001
5.589 ( 0.299
2.534 ( 0.169
>25
>25
p < 0.001
p < 0.001
FBpT
ClBpT
BrBpT
IBpT
0.356 ( 0.026 16.66 ( 0.407 p < 0.001
0.321 ( 0.072 8.198 ( 1.498 p < 0.005
0.172 ( 0.046 6.980 ( 1.662 p < 0.005
0.024 ( 0.000 2.321 ( 0.208 p < 0.001
0.022 ( 0.001 2.468 ( 0.008 p < 0.001
47
26
41
Bp4mT
FBp4mT
96
112
148
93
ClBp4mT 0.013 ( 0.002 1.967 ( 0.155 p < 0.001
BrBp4mT 0.011 ( 0.002 1.011 ( 0.167 p < 0.001
IBp4mT
Bp4eT
0.010 ( 0.001 0.959 ( 0.084 p < 0.001
0.012 ( 0.001 0.738 ( 0.100 p < 0.001
0.010 ( 0.001 0.732 ( 0.108 p < 0.001
0.009 ( 0.001 0.586 ( 0.062 p < 0.001
0.009 ( 0.002 0.498 ( 0.038 p < 0.001
0.010 ( 0.001 0.461 ( 0.023 p < 0.001
0.009 ( 0.001 1.012 ( 0.106 p < 0.001
98
64
FBp4eT
ClBp4eT
BrBp4eT
IBp4eT
Bp44mT
71
65
53
46
116
102
96
FBp44mT 0.008 ( 0.001 0.794 ( 0.171 p < 0.001
ClBp44mT 0.007 ( 0.001 0.706 ( 0.008 p < 0.001
BrBp44mT 0.008 ( 0.001 0.571 ( 0.093 p < 0.001
72
IBp44mT
0.009 ( 0.002 0.633 ( 0.004 p < 0.001
68
a Cells were seeded and allowed to attach to wells for 24 h/37 °C and
then incubated for 72 h/37 °C with control medium or the chelators.
Results are means ( SDs (three experiments). The p values were
determined using Student's t test and compared the activity of the ligand
using normal and neoplastic cells. Therapeutic index values represent
the ratio of MRC-5 to SK-N-MC IC50 values; a higher therapeutic index
value indicates greater selectivity for neoplastic cells.
Overall, the anodic shift in potential resulting from incorpora-
tion of a halogen to the noncoordinating phenyl group is offset by
the cathodic shift resulting from alkylation at N4. In relation to
biological activity, all measured FeII/III redox potentials lie within
the range accessible to both cellular oxidants and reductants, and
both the ferric and the ferrous forms are chemically stable.
Hence, like their BpT counterparts,33 these halogen-substituted
compounds could redox cycle to generate intracellular ROS
(examined below).
’ BIOLOGICAL STUDIES
Antiproliferative Activity against Tumor Cells. The ability
of the XBpT series to inhibit cellular proliferation was assessed
using SK-N-MC neuroepithelioma cells, as the effect of Fe
chelators on their growth has been well characterized.31,33,37
These novel ligands were compared to their parent compounds,
the BpT series,33 and a number of relevant positive control
chelators. These included DFO, which is used for the treatment
of Fe overload; 2-hydroxy-1-naphthylaldehyde isonicotinoyl
hydrazone (NIH), a ligand with high iron mobilizing activity
6939
dx.doi.org/10.1021/jm200924c |J. Med. Chem. 2011, 54, 6936–6948

Journal of Medicinal Chemistry
ARTICLE
activity having log Pcalcd values between 3.1 and 4.5 (Figure 3B).
Again, the XBpT analogues and the BpT parent ligand were
excluded to enable a comparison to neoplastic cells (Figure 3A).
Chelator-Mediated 59Fe Efflux from Cells. To determine the
potential role of Fe chelation efficacy in the antiproliferative
activity of the XBpT analogues, the ability of these chelators to
mobilize intracellular 59Fe from prelabeled SK-N-MC neuroe-
pithelioma cells was examined (Figure 4A). The release of 59Fe
mediated by these ligands was compared to that of their parent
BpT compounds, as well as a number of positive controls,
including the chelators DFO, NIH, and Dp44mT (Figure 1A),
which have been characterized in detail in this cell type.9,18,37
Both of the positive controls, NIH and Dp44mT, showed high
59Fe mobilization activity, releasing 55 ( 1 and 53 ( 1% of
intracellular 59Fe, respectively (Figure 4A). In comparison, the
control that was culture media alone mediated 10 ( 1% of
intracellular 59Fe release. As shown previously,18,33 the clinically
used chelator, DFO, demonstrated limited 59Fe mobilization
efficacy, resulting in the release of only 23 ( 1% of intracellular
59Fe (Figure 4A). All 16 members of the XBpT analogues
demonstrated significant 59Fe mobilization activity, resulting in
the release of 34ꢀ56% of cellular 59Fe. The XBp4mT series of
ligands, FBp4mT, ClBp4mT, and BrBp4mT, demonstrated
the greatest 59Fe chelation efficacy (51ꢀ56% of cellular 59Fe
released; Figure 4A), having comparable 59Fe mobilization
activity to NIH and Dp44mT. The XBp4eT and XBp44mT
ligands showed very similar activity that was slightly less than
their homologous Bp4mT counterparts, leading to the release of
34ꢀ48% of cellular 59Fe.
Figure 3. Relationship between the antiproliferative activity and lipo-
philicity of the chelators using: (A) SK-N-MC neuroepithelioma cells or
(B) MRC-5 lung fibroblasts. The IC50 and log Pcalcd values for the
XBp4mT, XBp44mT, and XBp4eT analogues and their parent com-
pounds (Bp4mT, Bp44mT, and Bp4eT) were plotted. However, the
XBpT analogues and the respective parent compound (BpT) were
excluded due to their high IC50 values.
Within each series, incorporation of a halogen moiety led to a
molecular weight-dependent decline in 59Fe mobilization activ-
ity, the starkest of which was observed for the XBp4eT and
XBp44mT series. Within all four groups of BpT analogues, the
addition of I to the noncoordinating phenyl group of the parent
compound resulted in a significant (p < 0.001) decrease of Fe
chelation efficacy (Figure 4A). In fact, the iodinated chelator
represents the most lipophilic halogenated ligand in each group
(Table 1), and as such, may become entrapped within mem-
branes through their high affinity for the lipid environment,
reducing 59Fe efflux. It should be noted that the chelator-
mediated increase in cellular 59Fe mobilization was not mediated
by their cytotoxic effects, as the cells remained >99% viable
(as judged by Trypan blue staining) within the very short 3 h
incubation used in these assays.
Antiproliferative Activity against Normal Cells. For a
compound to be considered as an antitumor agent, it must
exhibit potent antiproliferative activity against neoplastic cells,
while leaving normal cells unaffected. To determine whether the
compounds exhibited selectivity toward neoplastic cells, the
effect on the proliferation of mortal MRC-5 fibroblasts was
assessed (Table 2). Comparison of the antiproliferative activity
of the chelators between SK-N-MC neuroepithelioma cells and
MRC-5 fibroblasts showed that the neoplastic cells were sig-
nificantly (p < 0.005ꢀ0.001) more sensitive than the normal
cells, suggesting an appreciable therapeutic index. In contrast to
the results found for the SK-N-MC tumor cells, where only
halogenation of the BpT and Bp4mT analogues led to increased
antiproliferative activity, it was of interest that using mortal
MRC-5 cells, halogenation increased activity in all series. How-
ever, it was notable that as found for SK-N-MC cells, the increase
in antiproliferative efficacy upon halogenation was most pro-
nounced for the BpT series (Table 2).
The in vitro therapeutic index was also calculated. This
parameter represents the ratio of the IC50 values of MRC-5 to
SK-N-MC cells (i.e., IC50 MRC-5/IC50 SK-N-MC; Table 2),
with higher values representing greater selectivity against tumor
cells. The XBp4mT series showed the greatest selectivity, with
ClBp4mT being the most selective ligand examined. Plotting the
octanol:water log Pcalcd for the chelators against IC50 found in
mortal MRC-5 cells revealed that as found for neoplastic SK-N-
MC cells (Figure 3A), a very similar optimal lipophilic range
exists, with all ligands possessing the highest antiproliferative
Chelator-Mediated Inhibition of Cellular 59Fe Uptake
from 59Fe-Tf. The ability of the XBpT series chelators to inhibit
59Fe uptake from the Fe-binding protein, Tf, by SK-N-MC
neuroepithelioma cells was also assessed (Figure 4B). This is
important, as inducing Fe deprivation and thus antiproliferative
activity involves both increasing Fe mobilization and preventing
Fe uptake from Tf.37 As demonstrated in our previous work,9,18
the positive controls, NIH and Dp44mT, were found to effec-
tively reduce 59Fe uptake to 6 ( 1 and 4 ( 1% of the control,
respectively (Figure 4B). In contrast, the hydrophilic chelator,
DFO, exhibited far less activity, as shown previously,37ꢀ39
reducing 59Fe uptake to only 74 ( 1% of the control (Figure 4B).
All members of the XBpT series markedly reduced 59Fe uptake
to between 4 and 32% of the control (Figure 4B), being
significantly (p < 0.01) more effective than DFO. All of the
ligands examined, except BpT, FBpT, and ClBpT, were signifi-
cantly (p < 0.05) less effective than both Dp44mT and NIH at
inhibiting 59Fe uptake from 59Fe-Tf. The XBpT series of ligands
6940
dx.doi.org/10.1021/jm200924c |J. Med. Chem. 2011, 54, 6936–6948

Journal of Medicinal Chemistry
ARTICLE
Figure 4. Effect of the XBpT series chelators on (A) 59Fe mobilization from prelabeled SK-N-MC neuroepithelioma cells and (B) 59Fe uptake from
59Fe-Tf by SK-N-MC neuroepithelioma cells. Results are means ( SDs of three experiments with three determinations in each experiment. The
relationship between (C) 59Fe uptake and lipophilicity (log Pcalcd) and (D) 59Fe mobilization and lipophilicity (log Pcalcd) for each examined ligand.
were the most effective at reducing 59Fe uptake from 59Fe-Tf,
decreasing it to 4ꢀ12% of the control (Figure 4B). As observed
in the 59Fe efflux experiments (Figure 4A), incorporation of a
halogen again led to a molecular weight-dependent reduction in
activity, with the I-containing analogue having the least activity
at preventing 59Fe uptake in each case. This relationship can
be seen most clearly when 59Fe efflux and uptake results are
plotted against the lipophilicity (log Pcalcd) of the chelators
(Figure 4C,D). Within each series, a linear relationship was
observed between 59Fe efflux or 59Fe uptake relative to lipophi-
licity, demonstrating the important role of optimum lipophilicity
in their ability to bind cellular iron pools that are crucial for
proliferation.5,16,26
Ascorbate Oxidation Studies. The electrochemical studies
reported in Table 1 demonstrated the facile interconversion
between the FeIII and FeII states at potentials accessible to
biological oxidants and reductants. To determine if redox activity
played a role in the compounds' antiproliferative activity, their
ability to catalyze the oxidation of a physiologically relevant
substrate, ascorbate, was evaluated. The in vitro oxidation of
ascorbate by Fe complexes of the XBpT series was examined, in
both comparison to their parent BpT analogues and the positive
control, EDTA (Figure 5). EDTA was included as its activity has
been well characterized in this assay.33,35,40
In all experiments, results were expressed as iron-binding
equivalents (IBEs). This was due to the different coordination
modes of the ligands to Fe, that is, EDTA is hexadentate and
6941
dx.doi.org/10.1021/jm200924c |J. Med. Chem. 2011, 54, 6936–6948

Journal of Medicinal Chemistry
ARTICLE
analogues to catalyze FeIII-mediated ascorbate oxidation at IBE
ratios of 0.1, 1, or 3, with all compounds being similar to EDTA in
each case (Figure 5AꢀC). The ascorbate oxidation activity of the
XBp44mT series could not be measured with certainty, as under
the reaction conditions used, the compounds precipitated from
solution. Hence, these results are not shown.
Collectively, these ascorbate oxidation results indicate that the
incorporation of a halogen to the noncoordinating phenyl group
of the BpT series does not appreciably affect their ability to
mediate ascorbate oxidation; hence, their redox activity remained
marked. These results are also in agreement with the electro-
chemical data presented, which demonstrated that the Fe com-
plexes of the XBpT series exhibited comparable redox potentials
to their parent BpT counterparts (Table 1), enabling ascorbate
oxidation.
Intracellular ROS Assay. To confirm the redox activity of the
XBpT analogues as suggested by the electrochemical and ascor-
bate oxidation studies, further studies assessed their potential to
generate ROS intracellularly. The degree of cellular ROS was
examined by probing for oxidation of the nonfluorescent mole-
cule, 20,70-dichlorodihydrofluorescein (H2DCF), which becomes
fluorescent when oxidized to 20,70-dichlorofluorescein (DCF).
The use of this probe represents an established protocol for
assessing intracellular redox stress.18,41 Hydrogen peroxide
(H2O2) was included as a positive control.18
Incubation of cells with all analogues studied (except BpT) led
to a shift of the DCF fluorescence peak to the right relative to
control cells. Indeed, as appropriate examples in Figure 6A, we
present flow cytometry profiles of the parent compound of each
series (BpT, Bp4mT, Bp4eT, and Bp44mT) and their respective
I-substituted analogues. This shift is marked for the halogenated
analogues (shaded in gray) relative to the parent compounds and
reflects an increase in the intracellular oxidation of H2DCF to
fluorescent DCF. The DCF fluorescence peaks for each com-
pound evaluated were quantified and are shown in Figure 6B.
These data reveal a significant (p < 0.01) increase in DCF
fluorescence for the parent compounds Bp4mT, Bp4eT, and
Bp44mT of 114 ( 6, 138 ( 17, and 156 ( 28% relative to the
control following a 3 h incubation at 25 μM. This observation
probably reflects their increasing lipophilicity and access to
intracellular compartments. The positive control, hydrogen
peroxide (50 μM), produced a significant (p < 0.001) increase
in DCF oxidation to 289 ( 4% of control following a 30 min/
37 °C incubation.
Figure 5. Effect of the FeIII complexes of the (A) XBpT, (B) XBp4mT,
and (C) XBp4eT series and their parent compounds on ascorbate
oxidation. Chelators at iron-binding equivalent (IBE) ratios of 0.1, 1, and
3 were incubated in the presence of FeIII (10 μM) and ascorbate
(100 μM). The UVꢀvis absorbance at 265 nm was recorded after 10 and
40 min, and the difference between the time points was calculated.
Comparison of these chelators is made to the positive control, EDTA.
The XBp44mT series of compounds could not be measured due to
precipitation of the ligand using these assay conditions. Results are
means ( SDs (three experiments).
The incorporation of halogens to the phenyl group of the
parent ligands (except FBpT and FBp44mT) resulted in a
significant (p < 0.01) increase in DCF fluorescence over both
the control and the parent ligand (Figure 6B). In all cases, the
iodinated analogues led to the greatest increase in H2DCF
oxidation in each series, with an increase of 220 ( 11, 293 (
13, 370 ( 21, and 469 ( 39% relative to control medium being
observed for IBpT, IBp4mT, IBp4eT, and IBp44mT, respec-
tively. The most active chelators at mediating H2DCF oxidation
were the ꢀBr and ꢀI analogues of Bp4eT and Bp44mT, which
increased oxidation to 326ꢀ469% of the control and were
significantly (p < 0.01) more effective than H2O2.
The influence of halogenation on the oxidation of H2DCF by
the BpT analogues can be divided into two main effects: (i)
incorporation of F (except FBpT), which led to a slight increase
in DCF fluorescence relative to the parent analogue, and (ii)
incorporation of Cl, Br, or particularly I, which led to a significant
(p < 0.01) enhancement in DCF fluorescence versus the parent
forms 1:1 ligand:Fe complexes, while the BpT and XBpT
analogues are tridentate (Figure 1S in the Supporting
Information),33 resulting in 2:1 ligand:Fe complexes. In the
present study, a range of ligand:Fe IBE ratios were used, namely,
0.1, 1, or 3. An IBE of 0.1 represents an excess of Fe, where one
hexadentate or two tridentate ligands are in the presence of 10 Fe
atoms. An IBE of 1 results in complete filling of the coordination
sphere, while an IBE of 3 represents an excess of the ligand, where
three hexadentate or six tridentate ligands are in the presence of
one Fe atom.31
In good agreement with our previous studies, EDTA markedly
increased ascorbate oxidation at an IBE of 1 and 3, respectively,
while showing little activity at an IBE of 0.1 (Figure 5AꢀC). This
is in agreement with previous studies demonstrating the ability of
the EDTAꢀFe complex to undergo facile redox cycling and
catalyze ascorbate oxidation.33,35,40 Generally, halogenation did
not markedly affect the ability of the BpT, Bp4mT, or Bp4eT
6942
dx.doi.org/10.1021/jm200924c |J. Med. Chem. 2011, 54, 6936–6948

Journal of Medicinal Chemistry
ARTICLE
Figure 6. Comparison of the intracellular production of ROS by the parent and XBpT chelators (25 μM) as demonstrated by the oxidation of
nonfluorescent H2DCF to fluorescent DCF following a 3 h incubation at 37 °C. The DCF probe fluorescence was quantified by flow cytometry in SK-N-
MC cells. The positive control, H2O2 (50 μM), was incubated with cells for 30 min/37 °C. (A) As relative examples of the flow cytometry results,
presented are the following: control cells (solid line) or cells incubated with the parent BpT ligand (dashed line) or iodinated BpT ligand (shaded area).
(B) Quantification of the flow cytometric results for all examined ligands showing the percentage increase in DCF fluorescence over that observed for
control cells. Results are means ( SDs (n = 6). (C) The relationship between DCF fluorescence and lipophilicity (log Pcalcd).
compound. Plotting these values against the calculated log Pcalcd
of the ligand, we observed a positive linear relationship between
DCF fluorescence and lipophilicity for all halogenated analogues
examined (Figure 6C). Hence, the enhanced permeability of the
chelator due to halogen addition probably plays an important
role in the increased intracellular redox activity observed. This
factor appears crucial in terms of interpreting these data, as there
was little difference in the ability of these analogues to oxidize
ascorbate (Figure 5).
Assessment of H2DCF oxidation as a function of antiproli-
ferative activity (IC50) against SK-N-MC cells revealed a positive
correlation between these parameters for the XBpT (r2 = 0.68)
and XBp4mT (r2 = 0.67) series of halogenated ligands (data not
shown). Hence, it can be hypothesized that increasing the
lipophilicity of these ligands by halogenation led to ligands
entering the optimum lipophilic range (log Pcalcd 3.1ꢀ4.5;
Table 1 and Figure 3), which resulted in enhanced permeability,
intracellular redox activity (Figure 6), and antiproliferative
efficacy (Table 2). However, no such correlation between
H2DCF oxidation as a function of antiproliferative activity was
observed for the more lipophilic XBp4eT and XBp44mT series
of ligands (r2 = 0.20 and 0.05, respectively). This lack of
relationship for the latter ligands suggests that the already
optimal lipophilicity of the parent compounds (i.e., Bp4eT and
Bp44mT; Table 1 and Figure 3) meant that while halogenation
results in increased intracellular redox activity, this did not
potentiate antitumor activity. This may indicate that other factors
in addition to a certain threshold of redox activity are essential
for the antiproliferative efficacy of the XBp4eT and XBp44mT
series. Indeed, it is known that chelators act by multiple
6943
dx.doi.org/10.1021/jm200924c |J. Med. Chem. 2011, 54, 6936–6948

Journal of Medicinal Chemistry
ARTICLE
mechanisms to inhibit tumor cell growth, including their effects
on molecules involved in DNA synthesis (i.e., ribonucleotide
reductase),9ꢀ11 cell cycle progression (i.e., cyclin D1 and
p21),12,14,15 and growth and metastasis suppression (i.e., p53
and NDRG1).13,15,16
range (i.e., log Pcalcd 3.1ꢀ4.5; Table 1 and Figure 3A,B), leading
to increased intracellular redox activity and antiproliferative
efficacy. In contrast, the addition of halogens to the more
lipophilic XBp4eT and XBp44mT analogues resulted in no
correlation between intracellular redox activity and antiprolifera-
tive efficacy, despite the increase in H2DCF oxidation. This was
probably because the lipophilicity of the parent compound was
already within the optimal range (Table 1 and Figure 3A,B), with
a further increase due to halogenation not resulting in a
substantial benefit to antiproliferative activity.
’ CONCLUSIONS
The incorporation of a halogen into the phenyl group of the
BpT series of ligands generated chelators with potent antiproli-
ferative activity against neoplastic cells. Importantly, an appreci-
able in vitro therapeutic index was observed as tumor cells were
more sensitive to the effects of the chelators than mortal cells.
The increased antineoplastic activity of some of the XBpT series
can, in part, be attributed to their enhanced lipophilicity due to
the presence of a halogen group. This effect was most clearly
evident with the more hydrophilic chelators (i.e., XBpT and
XBp4mT series), as demonstrated by the fact that the antipro-
liferative activity of the parent BpT ligand against SK-N-MC cells
increased 32-fold by the addition of I (Table 2). The addition of
halogens to the more lipophilic parent ligands (i.e., Bp4eT and
Bp44mT) led to no marked improvement of antiproliferative
activity against tumor cells. Potentially, this latter effect could be
because the parent compounds were already sufficiently lipophi-
lic to gain intracellular access to crucial Fe pools and that a further
increase in lipophilicity did not enhance this. By examining the
library of synthesized compounds, we established that the optimal
lipophilic (log Pcalcd) range was between 3.1 and 4.5, with the
greatest antiproliferative efficacy being evident at log Pcalcd ∼ 4
(Figure 3). Hence, these latter data support the reasoning above,
namely, that halogenation of the BpT and Bp4mT parent ligands
led to analogues that approached or were within this optimal
lipophilic range, while the lipophilicity of the Bp4eT and Bp44mT
parents were already within the optimal range (Table 1).
Generally, the XBpT analogues exhibited a significant ability
to mobilize intracellular 59Fe from prelabeled cells and to inhibit
59Fe uptake from the serum Fe-binding protein Tf, being far
more effective than the clinically used chelator, DFO. However,
in comparison to their respective parent ligands, halogenation led
to reduced 59Fe mobilization activity and a decreased ability to
inhibit 59Fe uptake from Tf. Hence, increased lipophilicity
(log P) within each series led to decreased Fe chelation efficacy
(Figure 4). This could possibly reflect accumulation of the ligand
or its Fe complex in subcellular organelles or entrapment in cell
membranes due to their high affinity for the lipid environment.
The incorporation of a halogen led to a halogen-specific
increase in the redox potential of XBpTꢀFe complexes. Their
higher redox potentials did not appreciably affect their ability to
mediate ascorbate oxidation. However, incorporation of Cl, Br,
or I significantly enhanced H2DCF oxidation between 1.5- and
4.7-fold relative to all of the parent analogues. In addition, a
positive correlation (r2 = 0.835) was observed between DCF
fluorescence and lipophilicity (log Pcalcd; Figure 6C). Consider-
ing these latter results, the permeability of the chelator appears to
be a crucial factor for the higher intracellular redox activity
observed using the DCF assay. This is supported by the fact
that halogenation of the parent chelators did not significantly
affect ascorbate oxidation, where no permeability barrier was
involved (i.e., the cell membrane).
’ EXPERIMENTAL PROCEDURES
All commercial reagents were used without further purification. DFO
was obtained from Novartis (Basel, Switzerland). The 2-(4-haloben-
zoyl)pyridines were obtained from Rieke Metals Inc. (Lincoln, United
States). The chelators, Dp44mT and NIH, were prepared and char-
acterized according to previously described methods.31,35 Combustion
analysis (C, H, N) of the ligands and complexes was performed to
confirm g95% purity.
Chelators of the XBpT Series: General Synthesis. The
ligands were synthesized by the following general procedure: 2-(4-
Halobenzoyl)pyridine (0.01 mol) was dissolved in 15 mL of ethanol/
methanol, and subsequently, 1.00 g (0.01 mol) of the appropriate
thiosemicarbazide was dissolved in warm water (15 mL), and the two
solutions were then mixed. Glacial acetic acid (5ꢀ6 drops) was added,
and the mixture was gently refluxed for 2ꢀ5 h. The mixture was cooled
to room temperature and allowed to stand in a refrigerator overnight to
ensure complete precipitation. The product was filtered off and washed
with ethanol (10 mL) followed by diethyl ether (10 mL).
FBpT. Yield, 77%. Anal. calcd for C13H11N4SF: C, 56.92%; H, 4.04%;
N, 20.42%. Found: C, 56.31%; H, 4.09%; N, 19.77%. 1H NMR (DMSO-
d6): 8.67 (dq, 1H), 8.15 (dq, 1H), 8.39 (s, 1H), 8.05 (dd, 1H),
7.97ꢀ7.54 (m, 3H), 7.45ꢀ7.33 (dt, 3H). IR (cmꢀ1): 3391w, 3290m,
3151w, 3057w, 1640w, 1587s, 1509s, 1478s, 1421m, 1362w, 1329m,
1255m, 1223m, 1159m, 1108s, 1090m, 1057w, 1047w, 999w, 942w,
851s, 835s, 799m, 740m, 659m, 623m, 604m, 586m, 563m. MS (EI)
m/z 274 (M+).
ClBpT.CH3OH. Yield, 82%. Anal. calcd for C14H15N4OSCl: C,
52.09%; H, 4.68%; N, 17.36%. Found: C, 52.41%; H, 4.67%; N,
1
17.60%. H NMR (DMSO-d6): 8.53 (dq, 1H), 8.22 (dq, 1H), 8.39
(s, 1H), 8.05 (dd, 1H), 7.87ꢀ7.44 (m, 3H), 7.40ꢀ7.23 (dt, 3H). IR
(cmꢀ1): 3393w, 3276m, 3130w, 3096w, 1620w, 1590s, 1551w, 1473s,
1421m, 1397w, 1326m, 1256m, 1158w, 1108s, 1089s, 1065w, 1050w,
1014m, 949w, 854s, 827s, 796m, 779w, 734w, 678w, 657w, 633w, 620w,
598m, 523s. MS (EI) m/z 290 (M+).
BrBpT. Yield, 85%. Anal. calcd for C13H11N4SBr: C, 46.58%; H,
1
3.31%; N, 16.71%. Found: C, 46.1%; H, 3.16%; N, 16.36%. H NMR
(DMSO-d6): 8.63 (dq, 1H), 8.22 (dq, 1H), 8.43 (s, 1H), 8.23 (dd, 1H),
7.90ꢀ7.54 (m, 3H), 7.43ꢀ7.29 (dt, 3H). IR (cmꢀ1): 3405w, 3277m,
3155w, 3096w, 1620w, 1590s, 1539w, 1473s, 1421s, 1391w, 1327m,
1257w, 1178w, 1158w, 1108s, 1089s, 1070w, 1050w, 1009m, 948w,
854s, 825s, 796m, 750w, 729w, 720w, 690w, 656w, 620w, 598m, 523m.
MS (EI) m/z 335 (M+).
IBpT 1/2H2O. Yield, 92%. Anal. calcd for C13H13N4OSI: C, 39.91%;
3
H, 3.09%; N, 14.30%. Found: C, 39.83%; H, 2.89%; N, 14.00%. IR
(cmꢀ1): 3416w, 3295m, 3156w, 1620w, 1588s, 1535w, 1472s, 1421m,
1388w, 1365w, 1326m, 1256w, 1170w, 1107m, 1088m, 1005m, 962w,
948w, 853s, 823m, 794m, 748w, 713w, 682w, 655w, 632w, 620w, 597m,
520m, 492m. 1H NMR (DMSO-d6): 8.53 (dq, 1H), 8.22 (dq, 1H), 8.33
(s, 1H), 8.51 (dd, 1H), 7.80ꢀ7.54 (m, 3H), 7.33ꢀ7.29 (dt, 3H). MS
(EI) m/z 382 (M+).
Collectively, the current results demonstrate that the addition
of halogens to the more hydrophilic XBpT analogues (i.e., XBpT
and XBp4mT) increased their lipophilicity into the optimal
FBp4mT 1/2H2O. Yield, 79%. Anal. calcd for C14H14N4O0.5SF: C,
3
56.55%; H, 4.75%; N, 18.84%. Found: C, 55.96%; H, 4.52%; N, 18.60%.
6944
dx.doi.org/10.1021/jm200924c |J. Med. Chem. 2011, 54, 6936–6948

Journal of Medicinal Chemistry
ARTICLE
1H NMR (DMSO-d6): 8.75 (d, 1H, pyr), 8.61 (d, 1H, pyr), 8.21 (t, 1H,
pyr), 7.53 (m, 3H, pyr), 7.15 (m, 3H, pyr), 7.21 (m, 1H, pyr), 3.04 (d,
3H,-CH3). IR (cmꢀ1): 3307 m, 3146w, 2940w, 1601m, 1541s, 1509m,
1482s, 1452m, 1407m, 1360w, 1314m, 1226s, 1159m, 1113m, 1095m,
1047m, 996w, 942w, 845m, 814m, 797m, 735m, 663m, 648m, 620w,
592m, 511m. MS (EI) m/z 288 (M+).
(DMSO-d6): 8.70 (dd, 1H), 8.45 (dd, 1H), 8.12ꢀ7.84 (dd, 2H),
7.66ꢀ7.44 (dd, 4H), 7.33 (dt, 1H), 3.35 (s, 3H), 3.25 (s, 3H). IR
(cmꢀ1): 3059m, 2925w, 1587m, 1541s, 1506s, 1461s, 1424m, 1371m,
1314s, 1234s, 1154s, 1118s, 1065m, 1005m, 966m, 908s, 835s, 794m,
723s, 699m, 658w, 612w, 581m, 494m. MS (EI) m/z 302 (M+).
ClBp44mT 1/2H2O. Yield, 90%. Anal. calcd for C15H16N4O0.5SCl: C,
3
ClBp4mT H2O. Yield, 81%. Anal. calcd for C14H15N4OSCl: C,
54.96%; H, 4.92%; N, 17.09%; S, 9.78%. Found: C, 55.22%; H, 4.65%; N,
17.49%; S, 9.77%. 1H NMR (DMSO-d6): 8.70 (dd, 1H), 8.46 (dd, 1H),
8.12ꢀ7.84 (dd, 2H), 7.66ꢀ7.44 (dd, 4H), 7.33 (dt, 1H), 3.35 (s, 3H),
3.25 (s, 3H). IR (cmꢀ1): 3058m, 2925w, 1585m, 1541s, 1506s, 1460m,
1424w, 1371m, 1314s, 1234s, 1154s, 1118s, 1065m, 1005m, 966m, 909s,
835s, 795m, 723s, 699m, 658w, 611w, 581w. MS (EI) m/z 318 (M+).
3
52.09%; H, 4.68%; N, 17.36%. Found: C, 52.71%; H, 4.11%; N,
1
17.57%. H NMR (DMSO-d6): 8.63 (d, 1H, pyr), 8.52 (d, 1H, pyr),
8.16 (t, 1H, pyr), 7.33 (m, 3H, pyr), 7.09 (m, 3H, pyr), 7.41 (m, 1H,
pyr), 3.14 (d, 3H,-CH3). IR (cmꢀ1): 3357m, 3124w, 3054w, 1580m,
1548s, 1480s, 1427m, 1397m, 1369w, 1312m, 1269w, 1235s, 1161w,
1115m, 1105m, 1089m, 1039m, 1011w, 950w, 835s, 820m, 800w, 775w,
751w, 721w, 655w, 639m, 621m, 576m. MS (EI) m/z 304 (M+).
BrBp44mT 1/2H2O. Yield, 95%. Anal. calcd for C15H16N4O0.5SBr: C,
3
48.39%; H, 4.33%; N, 15.05%; S, 8.61%. Found: C, 48.31%; H, 3.96%; N,
15.27%; S, 8.53%. 1H NMR (DMSO-d6): 8.69 (dd, 1H), 8.50 (dd,1H),
8.27ꢀ7.81 (dd, 2H), 7.63ꢀ7.58 (dd, 4H), 7.44 (dt, 1H), 3.25 (s, 3H),
3.13 (s, 3H). IR (cmꢀ1): 3080w, 2940w, 1586m, 1537s, 1487m, 1465w,
1445w, 1374s, 1333s, 1310s, 1229s, 1176s, 1122s, 1068m, 1009m, 962m,
907m, 834s, 797s, 747m, 690w, 650m, 606w, 446w. MS (EI) m/z 363
(M+).
BrBp4mT 1/2H2O. Yield, 83%. Anal. calcd for C14H14N4O0.5SBr: C,
3
46.94%; H, 3.94%; N, 15.64%. Found: C, 46.5%; H, 3.50%, N, 15.50%.
1H NMR (DMSO-d6): 8.43 (d, 1H, pyr), 8.22 (d, 1H, pyr), 8.16 (t, 1H,
pyr), 7.33 (m, 3H, pyr), 7.11 (m, 3H, pyr), 7.56 (m, 1H, pyr), 3.08 (d,
3H,-CH3). IR (cmꢀ1): 3357m, 3124w, 3054w, 1590m, 1548s, 1480s,
1427m, 1397m, 1369w, 1312m, 1269w, 1235s, 1161w, 1114m, 1105m,
1089m, 1039m, 1010w, 950w, 835s, 820m, 801w, 775w, 751w, 721w,
655w, 639m, 621m, 576m. MS (EI) m/z 348 (M+).
IBp44mT. Yield, 96%. Anal. calcd. for C15H15N4SI: C, 43.91%; H,
3.68%; N, 13.66%; S, 7.81%. Found: C, 43.53%; H, 3.56%; N, 13.39%; S,
7.88%. 1H NMR (DMSO-d6): 8.66 (dd, 1H), 8.32 (dd,1H), 8.22ꢀ7.89
(dd, 2H), 7.33ꢀ7.55 (dd, 4H), 7.44 (dt, 1H), 3.25 (s, 3H), 3.13 (s, 3H).
IR (cmꢀ1): 3060w, 2931w, 1584m, 1533s, 1506m, 1446w, 1375s, 1330s,
1313s, 1228s, 1169m, 1117s, 1060m, 1008m, 958m, 908m, 839s, 803m,
778m, 733m, 655m, 611w, 492w. MS (EI) m/z 410 (M+).
IBp4mT 1/2H2O. Yield, 88%. Anal. calcd for C14H14N4O0.5SI: C,
3
41.49%; H, 3.48%; N, 13.83%. Found: C, 41.34%; H, 3.16% N, 13.69%.
1H NMR (DMSO-d6): 8.54 (d, 1H, pyr), 8.13 (d, 1H, pyr), 8.19 (t, 1H,
pyr), 7.33 (m, 3H, pyr), 7.23 (m, 3H, pyr), 7.44 (m, 1H, pyr), 3.10 (d,
3H,-CH3). IR (cmꢀ1): 3357m, 3124w, 3054w, 1595m, 1548s, 1480s,
1427m, 1397m, 1364w, 1312m, 1269w, 1235s, 1161w, 1112w, 1100w,
1089m, 1039m, 1011w, 950w, 836m, 820m, 800w, 775w, 750w, 721w,
639w, 621w, 576m, 498m. MS (EI) m/z 396 (M+).
General Synthesis of [FeIII(XBpT)]+ Complexes. The iron
complexes were prepared by the following general method. The appropriate
thiosemicarbazone (3.5 mmol) was dissolved in 15 mL of ethanol. Minimal
amounts of MeCN were added dropwise to aid dissolution. Once dissolved,
0.36 g of Et3N was added to the solution. Then, 0.81 g (1.7 mmol) of
FBp4eT. Yield, 94%. Anal. calcd for C15H15N4SF: C, 59.58%; H,
5.00%; N, 18.53%. Found: C, 59.63%; H, 4.99%; N, 18.40%. 1H NMR
(DMSO-d6): 8.55 (t, 1H), 8.05 (td, 1H), 7.73ꢀ7.66 (td, 1H),
7.47ꢀ7.42 (m, 3H), 7.34 (dt, 1H), 3.65 (m, 2H), 1.06 (t, 3H). IR
(cmꢀ1): 3282m, 3090w, 2977w, 2937w, 1599m, 1541s, 1475s, 1433m,
1420m, 1371w, 1310s, 1218s, 1155m, 1113m, 1094w, 1052m, 1001w,
931m, 848s, 813m, 794m, 738s, 682m, 658m, 621w, 452m. MS (EI) m/z
302 (M+).
Fe(ClO4)3 6H2O was added, and the mixture was gently refluxed for 60
3
min. After it was cooled, the dark brown powder was filtered off and washed
with ethanol (10 mL) followed by diethyl ether (10 mL).
[Fe(FBpT)2]ClO4 3H2O. Yield, 33%. Anal. calcd for C26H26ClFe-
3
N8O7S2F2: C, 41.31%; H, 3.47%; N, 14.82%. Found: C, 41.10%; H,
3.71%; N, 14.50%. IR (cmꢀ1): 3308w, 1602s, 1506s, 1482s, 1433s,
1324s, 1228s, 1208s, 1146s, 1092s, 956w, 851m, 784s, 755m, 720m,
650m, 627s.
ClBp4eT 1/2H2O. Yield, 93%. Anal. calcd for C15H16N4O0.5SCl: C,
3
54.96%; H, 4.92%; N, 17.09%; S, 9.78%. Found: C, 55.38%; H, 4.78%; N,
17.53; S, 9.47%. IR (cmꢀ1): 3303m, 3100w, 2965w, 2935w, 1585m,
1536s, 1479s, 1432m, 1310s, 1251w, 1217s, 1154w, 1114m, 1091m,
1055m, 1013w, 930w, 846m, 811m, 794m, 745m, 676m, 658w, 621w,
501s. 1H NMR (DMSO-d6): 8.55 (t, 1H), 8.05 (td, 1H), 7.73ꢀ7.66 (td,
1H), 7.47ꢀ7.42 (m, 3H), 7.34 (dt, 1H), 3.65 (m, 2H), 1.06 (t, 3H). MS
(EI) m/z 318 (M+).
[Fe(ClBpT)2]ClO4 CH3CH2OH. Yield, 37%. Anal. calcd for
3
C28H25Cl3FeN8O5S2: C, 43.07%; H, 3.36%; N, 14.35%; S, 8.21%.
Found: C, 42.7%; H, 4.05%; N, 14.34%; S, 7.89%. IR (cmꢀ1): 3274w,
1591s, 1493s, 1471s, 1443s, 1422s, 1326s, 1227m, 1147s, 1107s, 1089s,
1014s, 854s, 828s, 797s, 705w, 651w, 624s.
[Fe(BrBpT)2]ClO4 H2O. Yield, 38%. Anal. calcd for C26H22ClFe-
3
N8O5S2Br2: C, 37.10%; H, 2.63%; N, 13.31%; S, 7.62%. Found: C,
36.90%; H, 2.55%; N, 13.19%; S, 7.59%. IR (cmꢀ1): 3279w, 1610s,
1590s, 1493s, 1424vs, 1322s, 1207s, 1147s, 1104s, 1011s, 942m, 853m,
840m, 830m, 797s, 747s, 728w, 655s, 624s.
BrBp4eT 1/2H2O. Yield, 91%. Anal. calcd for C15H16N4O0.5SBr: C,
3
48.39%; H, 4.33%; N, 15.05%; S, 8.61%. Found: C, 47.93%; H, 4.05%; N,
14.87%; S, 8.35%. 1H NMR (DMSO-d6): 8.51 (t, 1H), 8.33 (td, 1H),
7.80ꢀ7.56 (td, 1H), 7.47ꢀ7.33 (m, 3H), 7.24 (dt, 1H), 3.55 (m, 2H),
1.08 (t, 3H). IR (cmꢀ1): 3277mw, 3114w, 2970w, 2929w, 1591m,
1532s, 1471s, 1420m, 1392w, 1310s, 1252w, 1216s, 1151w, 1115m,
1053m, 1011w, 928w, 835m, 811w, 796w, 777w, 737m, 654w, 619w,
498w. MS (EI) m/z 364 (M+).
[Fe(IBpT)2]ClO4 2CH3CN. Yield, 39%. Anal. calcd for C32H29ClFe-
3
N11O4S2I2: C, 36.04%; H, 2.62%; N, 14.01%. Found: C, 36.18%; H,
2.72%; N, 13.62%. IR (cmꢀ1): 3354w, 1608s, 1592s, 1492s, 1428vs,
1320s, 1207s, 1144s, 1092vs, 1007s, 943w, 852s, 797s, 748w, 726w, 624s.
[Fe(FBp4mT)2]ClO4. Yield, 37%. Anal. calcd for C28H24ClFe-
N8O4S2F2: C, 46.07%; H, 3.31%; N, 15.35%; S, 8.78%. Found: C,
46.06%; H, 3.13%; N, 15.59%; S, 8.93%. IR (cmꢀ1): 3383w, 1602w,
1548s, 1519s, 1495s, 1449s, 1400s, 1338s, 1297s, 1269s, 1234s, 1204s,
1161s, 1104vs, 970w, 850m, 834m, 785s, 756w, 623s.
IBp4eT 1/2H2O. Yield, 94%. Anal. calcd for C15H16N4O0.5SI: C,
3
42.97%; H, 3.85%; N, 13.36%. Found: C, 43.05%; H, 3.57%; N,
13.33%. 1H NMR (DMSO-d6): 8.44 (t, 1H), 8.22 (td, 1H),
7.76ꢀ7.56 (td, 1H), 7.41ꢀ7.32 (m, 3H), 7.28 (dt, 1H), 3.34 (m,
2H), 1.08 (t, 3H). IR (cmꢀ1): 3284m, 3168w, 3078w, 2972w, 2929w,
1587m, 1533s, 1469s, 1421m, 1368m, 1310s, 1251w, 1216s, 1194s,
1150w, 1114m, 1098m, 1058m, 1007m, 955w, 928m, 830m, 809m,
798w, 777w, 736m, 676w, 617w, 494m. MS (EI) m/z 410 (M+).
FBp44mT. Yield, 96%. Anal. calcd for C15H15N4SF: C, 59.58%; H,
5.00%; N, 18.53%. Found: C, 59.93%; H, 4.99%; N, 18.30%. 1H NMR
[Fe(ClBp4mT)2]ClO4. Yield, 41%. Anal. calcd for C28H24Cl3Fe-
N8O4S2: C, 44.08%; H, 3.17%; N, 14.69%; S, 8.41%. Found: C,
44.11%; H, 3.17%; N, 14.85%; S, 8.48%. IR (cmꢀ1): 3392w, 1594w,
1548s, 1509s, 1448s, 1403s, 1335s, 1297w, 1267s, 1204s, 1164s, 1091vs,
1015w, 969w, 841m, 809w, 785m, 756w, 723w, 623s.
6945
dx.doi.org/10.1021/jm200924c |J. Med. Chem. 2011, 54, 6936–6948

Journal of Medicinal Chemistry
ARTICLE
[Fe(BrBp4mT)2]ClO4. Yield, 42%. Anal. calcd for C28H24ClFe-
N8O4S2Br2: C, 39.48%; H, 2.84%; N, 13.16%; S, 7.53%. Found: C,
39.60%; H, 2.91%; N, 13.40%; S, 7.38%. IR (cmꢀ1): 3396w, 1588w,
1545s, 1508s, 1447s, 1402s, 1333s, 1297w, 1267s, 1204s, 1164s, 1092vs,
1012m, 968m, 839m, 807w, 785m, 756w, 718w, 622s.
Partition coefficients of the free ligands were determined by ChemBioDraw
v.11.0.1. (CambridgeSoft) using Crippen's fragmentation procedure.42
Crystallography. Intensity data were acquired on an Oxford
Diffraction Gemini Ultra S CCD diffractometer. Data reduction was
performed with the CrysAlisPro program (version 171.33.42). The
structures were solved by direct methods with SHELX-86 and refined by
full matrix least-squares with SHELXL.43 Thermal ellipsoid diagrams
were generated with ORTEP3.44 All calculations were carried out within
the WinGX package.45
Biological Studies. Cell Culture. Chelators were dissolved in
DMSO as 10 mM stock solutions and diluted in medium containing
10% fetal calf serum (Commonwealth Serum Laboratories, Melbourne,
Australia) so that the final [DMSO] < 0.5% (v/v). At this final
concentration, DMSO had no effect on proliferation, 59Fe uptake, or
59Fe mobilization from cells, as shown previously.37 The human SK-N-
MC neuroepithelioma cell line and mortal human MRC5 fibroblasts
(American Type Culture Collection, Manassas, VA) were grown by
standard procedures37 at 37 °C in a humidified atmosphere of 5% CO2/
95% air in an incubator (Forma Scientific, Marietta, OH).
Effect of the Chelators on Cellular Proliferation. The effect of the
chelators and complexes on cellular proliferation was determined by the
[1-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium] (MTT) assay
using standard techniques.37,46 The SK-N-MC cell line was seeded in 96-
well microtiter plates at 1.5 ꢁ 104 cells/well in medium containing
human 56Fe2-Tf (1.25 μM) and chelators at a range of concentrations
(0ꢀ25 μM). Control samples contained medium with Fe2-Tf (1.25
μM) without the ligands. The cells were incubated at 37 °C in a
humidified atmosphere containing 5% CO2 and 95% air for 72 h. After
this incubation, 10 μL of MTT (5 mg/mL) was added to each well and
further incubated at 37 °C/2 h. After solubilization of the cells with 100
μL of 10% SDS-50% isobutanol in 10 mM HCl, the plates were read at
570 nm using a scanning multiwell spectrophotometer. The inhibitory
concentration (IC50) was defined as the chelator concentration neces-
sary to reduce the absorbance to 50% of the untreated control. Using this
method, absorbance was shown to be directly proportional to cell
counts, as shown previously.37
[Fe(IBp4mT)2]ClO4 CH3CN. Yield, 44%. Anal. calcd for C30H27ClFe-
3
N9O4S2I2: C, 36.51%; H, 2.76%; N, 12.77%; S, 6.50%. Found: C,
36.69%; H, 2.47%; N, 12.28%; S, 6.72%. IR (cmꢀ1): 3394w, 1587w,
1544s, 1505s, 1447s, 1400s, 1332s, 1297w, 1266s, 1204s, 1163m,
1088vs, 1007s, 967m, 837w, 806w, 784m, 755w, 717w, 622s.
[Fe(FBp4eT)2]ClO4. Yield, 31%. Anal. calcd for C30H28ClFe-
N8O4S2F2: C, 47.54%; H, 3.72%; N, 14.78%; S, 8.46%. Found: C,
47.63%; H, 3.71%; N, 14.50%; S, 8.14%. IR (cmꢀ1): 3383w, 1602w,
1543s, 1517s, 1493s, 1427s, 1334s, 1299s, 1266s, 1238s, 1198m, 1160s,
1094vs, 960w, 850m, 834m, 783s, 751w, 625s.
[Fe(ClBp4eT)2]ClO4 CH3CN. Yield, 36%. Anal. calcd for C30H28Cl3-
3
FeN8O4S2: C, 46.20%; H, 3.76%; N, 15.15%. Found: C, 46.16%; H,
3.73%; N, 14.90%. IR (cmꢀ1): 3401w, 1593w, 1540s, 1506s, 1493s,
1422s, 1331s, 1295s, 1261s, 1197m, 1153s, 1091vs, 1014m, 968w, 840w,
802w, 752w, 623s.
[Fe(BrBp4eT)2]ClO4 CH3CN. Yield, 40%. Anal. calcd for C32H31ClFe-
3
N9O4S2Br2: C, 41.74%; H, 3.39%; N, 13.69%; S, 6.96%. Found: C,
41.80%; H, 3.77%; N, 13.97%; S, 7.17%. 3398w, 1586w, 1535s, 1504s,
1473s, 1423s, 1332s, 1308s, 1216m, 1150s, 1115vs, 1012s, 968w, 836m,
797w, 737w, 621s.
[Fe(IBp4eT)2]ClO4. Yield, 42%. Anal. calcd for C32H29ClFeN11O4-
S2I2: C, 37.00%; H, 3.77%; N, 11.51%; S, 6.58%. Found: C, 37.16%; H,
3.72%; N, 11.63%; S, 6.49%. IR (cmꢀ1): 3357w, 1595w, 1541s, 1502s,
1475m, 1440s, 1331s, 1294s, 1262s, 1196s, 1149s, 1100vs, 1008s, 968m,
835w, 802w, 780w, 754w, 716w, 622s.
[Fe(FBp44mT)2]ClO4 CH3CN. Yield, 28%. Anal. calcd for C32H31-
3
ClFeN9O4S2F2: C, 48.10%; H, 3.91%; N, 15.78%; S, 8.02%. Found: C,
47.87%; H, 3.63%; N, 15.40%; S, 8.09%. IR (cmꢀ1): 2938w, 1602w,
1543s, 1510s, 1460s, 1437s, 1394s, 1345s, 1302s, 1263, 1205s, 1154w,
1094vs, 977w, 920s, 783m, 748w, 723m, 692s, 663w, 633w, 620s.
[Fe(ClBp44mT)2]ClO4. Yield, 38%. Anal. calcd for C30H28Cl3Fe-
N8O4S2: C, 45.56%; H, 3.57%; N, 14.17%; S, 8.11%. Found: C,
45.53%; H, 3.49%; N, 14.21%; S, 8.44%. IR (cmꢀ1): 2936w, 1599w,
1542s, 1508s, 1461s, 1437s, 1395s, 1345s, 1302s, 1263, 1207s, 1154w,
1091vs, 978w, 921s, 781m, 748m, 725w, 792s, 663w, 632w, 622s.
[Fe(BrBp44mT)2]ClO4. Yield, 43%. Anal. calcd for C30H28ClFe-
N8O4S2Br2: C, 40.95%; H, 3.21%; N, 12.74%; S, 7.29%. Found: C,
41.22%; H, 3.38%; N, 12.33%; S, 7.24%. IR (cmꢀ1): 2937w, 1600w,
1541s, 1510s, 1463s, 1436s, 1393s, 1345s, 1302s, 1260, 1210s, 1151w,
1092vs, 978w, 922s, 781m, 748w, 724m, 690s, 663w, 632w, 622s.
[Fe(IBp44mT)2]ClO4. Yield, 42%. Anal. calcd for C30H28ClFeN8O4S2I2:
C, 37.00%; H, 2.90%; N, 11.51%; S, 6.58%. Found: C, 37.44%; H, 2.99%;
N, 11.83%; S, 6.89%. IR (cmꢀ1): 2912w, 1589w, 1542s, 1505s, 1473s,
1436s, 1383s, 1346s, 1302s, 1262, 1210s, 1149m, 1099vs, 974w, 922s,
780w, 752w, 720m, 681s, 662w, 632w, 622s.
Preparation of 56Fe- and 59Fe-Tf. Human Tf (Sigma) was labeled
with 56Fe or 59Fe (Dupont NEN, MA) to produce 59Fe2-Tf and 56Fe2-
Tf, respectively, as previously described.39,46 Unbound 59Fe was re-
moved by passage through a Sephadex G25 column and exhaustive
vacuum dialysis against a large excess of 0.15 M NaCl buffered to pH 7.4
with 1.4% NaHCO3 by standard methods.46
Effect of Chelators on 59Fe Efflux from Cells. Iron efflux experiments
examining the ability of various chelators to mobilize 59Fe from SK-N-
MC cells were performed using established techniques.37,47 Briefly,
following prelabeling of cells with 59Fe2-Tf (0.75 μM) for 3 h/37 °C, the
cell cultures were washed on ice four times with ice-cold PBS and then
subsequently incubated with each chelator (25 μM) for 3 h/37 °C. The
overlying media containing released 59Fe was then separated from the
cells using a Pasteur pipet. The radioactivity was measured in both the
cell pellet and the supernatant using a γ-scintillation counter (Wallac
Wizard 3, Turku, Finland). In these studies, the novel ligands were
compared to the previously characterized chelators, DFO, NIH, and
Dp44mT.
1
Physical Methods. H NMR (400 MHz) spectra were acquired
using a Bruker Avance 400 NMR spectrometer with DMSO-d6 as the
solvent and internal reference (Me2SO: 1H NMR δ 2.49 ppm and 13
C
NMR δ 39.5 ppm vs TMS). Infrared spectra were measured with a
Varian Scimitar 800 FT-IR spectrophotometer with compounds being
dispersed as KBr discs. Cyclic voltammetry was performed using a
BAS100B/W potentiostat. A glassy carbon working electrode, an
aqueous Ag/AgCl reference, and a Pt wire auxiliary electrode were used.
All complexes were at ca. 1 mM concentration in MeCN:H2O 70:30 v/v.
This solvent combination was used to ensure solubility of all com-
pounds. The supporting electrolyte was Et4NClO4 (0.1 M), and the
solutions were purged with nitrogen prior to measurement. All poten-
tials are cited versus the normal hydrogen electrode (NHE) by addition
of 196 mV to the potentials measured relative to the Ag/AgCl reference.
Effect of Chelators at Preventing 59Fe Uptake from 59Fe2-Tf by Cells.
The ability of the chelator to prevent cellular 59Fe uptake from the serum
Fe transport protein, 59Fe2-Tf, was examined using established
techniques.18,48 Briefly, cells were incubated with 59Fe2-Tf (0.75 μM)
for 3 h/37 °C in the presence of each of the chelators (25 μM). The cells
were then washed four times with ice-cold PBS, and internalized 59Fe
was determined by standard techniques by incubating the cell monolayer
for 30 min/4 °C with the general protease, Pronase (1 mg/mL;
Sigma).37,49,50 The cells were removed from the monolayer using a
plastic spatula and centrifuged at 14000 rpm/1 min. The supernatant
6946
dx.doi.org/10.1021/jm200924c |J. Med. Chem. 2011, 54, 6936–6948

Journal of Medicinal Chemistry
ARTICLE
represents membrane-bound, Pronase-sensitive 59Fe that was released
by the protease, while the Pronase-insensitive fraction represents inter-
nalized 59Fe.49,50 The novel ligands were compared to the previously
characterized chelators, DFO, NIH, and Dp44mT.
Ascorbate Oxidation Assay. An established protocol was used to
measure ascorbate oxidation.2,9,51 Briefly, ascorbic acid (100 μM) was
prepared immediately prior to an experiment and incubated in the
presence of FeIII (10 μM; added as FeCl3), a 50-fold molar excess of
citrate (500 μM), and the chelator (1ꢀ60 μM). The absorbance at
265 nm was measured after 10 and 40 min at room temperature, and the
decrease of intensity between these time points was calculated.40 The
results of these experiments were expressed in terms of IBEs due to the
varying denticity of the chelators examined.
Intracellular ROS Measurements. Intracellular ROS generation was
measured using H2DCF-DA.18,41 H2DCF-DA is hydrolyzed by intra-
cellular esterases to H2DCF, which leads to it becoming trapped within
the cytosol. Cellular oxidants localized to the cytosol oxidize nonfluor-
escent H2DCF to the fluorescent product, DCF.18,41 SK-N-MC cells
were incubated with 30 nM H2DCF-DA for 30 min/37 °C and then
washed twice with ice-cold PBS. The cells were then treated with either
the positive control, H2O2 (50 μM) for 30 min/37 °C, or a chelator (25
μM) for 3 h/37 °C. Cells were collected for flow cytometric assessment,
and intracellular ROS was detected as an increase in green cytosolic DCF
fluorescence using a FACS Canto flow cytometer (Becton Dickinson,
Lincoln Park, NJ). In these studies, 10000 events were acquired for every
sample. Data analysis was performed using FlowJo software v7.5.5 (Tree
Star Inc., Ashland, OR).
H2DCF, 20,70-dichlorodihydrofluorescein; IBE, iron-binding
equivalent; XBpT, 20-(400-halobenzoyl)pyridine thiosemicar-
bazone; XBp4mT, 20-(400-halobenzoyl)pyridine 4-methyl-
3-thiosemicarbazone; XBp4eT, 20-(400-halobenzoyl)pyridine
4-ethyl-3-thiosemicarbazone; XBp44mT, 20-(400-halobenzoyl)-
pyridine 4,4-dimethyl-3-thiosemicarbazone; NIH, 2-hydroxy-
1-naphthaldehyde isonicotinoyl hydrazone; ROS, reactive oxy-
gen species; Tf, transferrin
’ REFERENCES
(1) Bergeron, R. J.; Huang, G. F.; Weimar, W. R.; Smith, R. E.;
Wiegand, J.; McManis, J. S. Desferrithiocin analogue based hexacoordi-
nate iron(III) chelators. J. Med. Chem. 2003, 46, 16–24.
(2) Bergeron, R. J.; Wiegand, J.; McManis, J. S.; Bussenius, J.; Smith,
R. E.; Weimar, W. R. Methoxylation of desazadesferrithiocin analogues:
Enhanced iron clearing efficiency. J. Med. Chem. 2003, 46, 1470–1477.
(3) Chua, A. C. G.; Ingram, H. A.; Raymond, K. N.; Baker, E.
Multidentate pyridinones inhibit the metabolism of nontransferrin-
bound iron by hepatocytes and hepatoma cells. Eur. J. Biochem. 2003,
270, 1689–1698.
(4) Liu, J.; Obando, D.; Schipanski, L. G.; Groebler, L. K.; Witting,
P. K.; Kalinowski, D. S.; Richardson, D. R.; Codd, R. Conjugates of
desferrioxamine B (DFOB) with derivatives of adamantane or with
orally available chelators as potential agents for treating iron overload.
J. Med. Chem. 2010, 53, 1370–1382.
(5) Buss, J. L.; Torti, F. M.; Torti, S. V. The role of iron chelation in
cancer therapy. Curr. Med. Chem. 2003, 10, 1021–1034.
(6) Torti, S. V.; Torti, F. M.; Whitman, S. P.; Brechbiel, M. W.; Park,
G.; Planalp, R. P. Tumor cell cytotoxicity of a novel metal chelator. Blood
1998, 92, 1384–1389.
Statistical Analysis. Experimental data were compared using Stu-
dent's t test. Results were expressed as means ( SDs (number of
experiments) and considered to be statistically significant when p < 0.05.
(7) Turner, J.; Koumenis, C.; Kute, T. E.; Planalp, R. P.; Brechbiel,
M. W.; Beardsley, D.; Cody, B.; Brown, K. D.; Torti, F. M.; Torti, S. V.
Tachpyridine, a metal chelator, induces G(2) cell-cycle arrest, activates
checkpoint kinases, and sensitizes cells to ionizing radiation. Blood 2005,
106, 3191–3199.
(8) Larrick, J. W.; Cresswell, P. Modulation of cell-surface iron
transferrin receptors by cellular density and state of activation.
J. Supramol. Struct. 1979, 11, 579–586.
’ ASSOCIATED CONTENT
S
Supporting Information. Crystal data, selected bond
b
lengths, and crystal structures of XBp44mT ligands and their
Fe complexes. This material is available free of charge via the
Internet at http://pubs.acs.org.
(9) Chaston, T. B.; Lovejoy, D. B.; Watts, R. N.; Richardson, D. R.
Examination of the antiproliferative activity of iron chelators: Multiple
cellular targets and the different mechanism of action of triapine
compared with desferrioxamine and the potent pyridoxal isonicotinoyl
hydrazone analogue 311. Clin. Cancer Res. 2003, 9, 402–414.
(10) Cooper, C. E.; Lynagh, G. R.; Hoyes, K. P.; Hider, R. C.;
Cammack, R.; Porter, J. B. The relationship of intracellular iron chelation
to the inhibition and regeneration of human ribonucleotide reductase.
J. Biol. Chem. 1996, 271, 20291–20299.
(11) Nyholm, S.; Mann, G. J.; Johansson, A. G.; Bergeron, R. J.;
Graslund, A.; Thelander, L. Role of ribonucleotide reductase in inhibi-
tion of mammalian-cell growth by potent iron chelators. J. Biol. Chem.
1993, 268, 26200–26205.
(12) Fu, D.; Richardson, D. R. Iron chelation and regulation of the
cell cycle: 2 mechanisms of posttranscriptional regulation of the
universal cyclin-dependent kinase inhibitor p21(CIP1/WAF1) by iron
depletion. Blood 2007, 110, 752–761.
(13) Le, N. T. V.; Richardson, D. R. Iron chelators with high
antiproliferative activity up-regulate the expression of a growth inhibi-
tory and metastasis suppressor gene: a link between iron metabolism and
proliferation. Blood 2004, 104, 2967–2975.
’ AUTHOR INFORMATION
Corresponding Author
*Tel: +61-2-9036-6548. Fax: +61-2-9036-6549. E-mail:
d.richardson@med.usyd.edu.au.
’ ACKNOWLEDGMENT
Financial support from the National Health and Medical
Research Council (Senior Principal Research Fellowship and
Project Grants to D.R.R.) and Australian Research Council
(DP0773027 to P.V.B. and D.R.R. and DP1096029 to P.V.B.)
is gratefully acknowledged. P.J.J. (10/CDF/2-15) and D.S.K.
(08/ECF/1-30) thank the Cancer Institute NSW for Fellowship
support. C.S. is the grateful recipient of an Australian Postgrad-
uate Award from the University of Sydney.
’ ABBREVIATIONS USED
3-AP, 3-aminopyridine-2-carboxaldehyde thiosemicarbazone;
DCF, 20,70-dichlorofluorescein; DFO, desferrioxamine; BpT, 20-
benzoylpyridine thiosemicarbazone; Bp4mT, 20-benzoylpyridine
4-methyl-3-thiosemicarbazone; Bp4eT, 20-benzoylpyridine 4-ethyl-3-
thiosemicarbazone; Bp44mT, 20-benzoylpyridine 4,4-dimethyl-
3-thiosemicarbazone; DpT, di-20-pyridylketone thiosemicarbazone;
Dp44mT, di-20-pyridylketone 4,4-dimethyl-3-thiosemicarbazone;
(14) Nurtjahja-Tjendraputra, E.; Fu, D.; Phang, J. M.; Richardson,
D. R. Iron chelation regulates cyclin D1 expression via the proteasome:
A link to iron deficiency-mediated growth suppression. Blood 2007,
109, 4045–4054.
(15) Yu, Y.; Kalinowski, D. S.; Kovacevic, Z.; Siafakas, A. R.; Jansson,
P. J.; Stefani, C.; Lovejoy, D. B.; Sharpe, P. C.; Bernhardt, P. V.;
Richardson, D. R. Thiosemicarbazones from the old to new: Iron
6947
dx.doi.org/10.1021/jm200924c |J. Med. Chem. 2011, 54, 6936–6948

Journal of Medicinal Chemistry
ARTICLE
chelators that are more than just ribonucleotide reductase inhibitors.
J. Med. Chem. 2009, 52, 5271–5294.
(16) Kalinowski, D. S.; Richardson, D. R. The evolution of iron
chelators for the treatment of iron overload disease and cancer.
Pharmacol. Rev. 2005, 57, 547–583.
(17) Barnham, K. J.; Masters, C. L.; Bush, A. I. Neurodegenerative
diseases and oxidative stress. Nature Rev. Drug Discovery 2004,
3, 205–214.
(18) Yuan, J.; Lovejoy, D. B.; Richardson, D. R. Novel di-2-pyridyl-
derived iron chelators with marked and selective antitumor activity: In
vitro and in vivo assessment. Blood 2004, 104, 1450–1458.
(19) Halliwell, B.; Gutteridge, J. M. C. Oxygen free-radicals and iron
in relation to biology and medicine—Some problems and concepts.
Arch. Biochem. Biophys. 1986, 246, 501–514.
synthesis, and characterization of novel iron chelators: Structure-activity
relationships of the 2-benzoylpyridine thiosemicarbazone series and
their 3-nitrobenzoyl analogues as potent antitumor agents. J. Med. Chem.
2007, 50, 3716–3729.
(34) Yu, Y.; Suryo Rahmanto, Y.; Richardson, D. R. Bp44mT: An
orally-active iron chelator of the thiosemicarbazone class with potent
anti-tumor efficacy. Br. J. Pharmacol. 2011, June 9. DOI: 10.1111/j.1476-
5381.2011.01526.x. [Epub ahead of print] PMID:21658021.
(35) Bernhardt, P. V.; Caldwell, L. M.; Chaston, T. B.; Chin, P.;
Richardson, D. R. Cytotoxic iron chelators: Characterization of the
structure, solution chemistry and redox activity of ligands and iron
complexes of the di-2-pyridyl ketone isonicotinoyl hydrazone (HPKIH)
analogues. J. Biol. Inorg. Chem. 2003, 8, 866–880.
(36) Richardson, D. R.; Kalinowski, D. S.; Richardson, V.; Sharpe,
P. C.; Lovejoy, D. B.; Islam, M.; Bernhardt, P. V. 2-Acetylpyridine
thiosemicarbazones are potent iron chelators and antiproliferative
agents: Redox activity, iron complexation and characterization of their
antitumor activity. J. Med. Chem. 2009, 52, 1459–1470.
(20) Chaston, T. B.; Watts, R. N.; Yuan, J.; Richardson, D. R. Potent
antitumor activity of novel iron chelators derived from di-2-pyridylk-
etone isonicotinoyl hydrazone involves fenton-derived free radical
generation. Clin. Cancer Res. 2004, 10, 7365–7374.
(21) Blatt, J.; Stitely, S. Antineuroblastoma activity of desferoxamine
in human cell-lines. Cancer Res. 1987, 47, 1749–1750.
(37) Richardson, D. R.; Tran, E. H.; Ponka, P. The potential of iron
chelators of the pyridoxal isonicotinoyl hydrazone class as effective
antiproliferative agents. Blood 1995, 86, 4295–4306.
(22) Donfrancesco, A.; Deb, G.; Dominici, C.; Angioni, A.; Caniglia,
M.; Desio, L.; Fidani, P.; Amici, A.; Helson, L. Deferoxamine, cyclopho-
sphamide, etoposide, carboplatin, and thiotepa (D-CECAT)—A new
cytoreductive chelation-chemotherapy regimen in patients with
advanced neuroblastoma. Am. J. Clin. Oncol.-Cancer Clin. Trials 1992,
15, 319–322.
(23) Donfrancesco, A.; Deb, G.; Dominici, C.; Pileggi, D.; Castello,
M. A.; Helson, L. Effects of a single course of deferoxamine in
neuroblastoma patients. Cancer Res. 1990, 50, 4929–4930.
(24) Estrov, Z.; Tawa, A.; Wang, X. H.; Dube, I. D.; Sulh, H.; Cohen,
A.; Gelfand, E. W.; Freedman, M. H. In vitro and in vivo effects of
deferoxamine in neonatal acute-leukemia. Blood 1987, 69, 757–761.
(25) Aouad, F.; Florence, A.; Zhang, Y.; Collins, F.; Henry, C.; Ward,
R. J.; Crichton, R. R. Evaluation of new iron chelators and their
therapeutic potential. Inorg. Chim. Acta 2002, 339, 470–480.
(26) Richardson, D. R. Molecular mechanisms of iron uptake by cells
and the use of iron chelators for the treatment of cancer. Curr. Med.
Chem. 2005, 12, 2711–2729.
(27) Karp, J. E.; Giles, F. J.; Gojo, L.; Morris, L.; Greer, J.; Johnson,
B.; Thein, M.; Sznol, M.; Low, J. A phase I study of the novel
ribonucleotide reductase inhibitor 3-aminopyridine-2-carboxaldehyde
thiosemicarbazone (3-AP, Triapine(R)) in combination with the
nucleoside analog fludarabine for patients with refractory acute leukemias
and aggressive myeloproliferative disorders. Leuk. Res. 2008, 32, 71–77.
(28) Ma, B.; Goh, B. C.; Tan, E. H.; Lam, K. C.; Soo, R.; Leong, S. S.;
Wang, L. Z.; Mo, F.; Chan, A. T. C.; Zee, B.; Mok, T. A multicenter phase
II trial of 3-aminopyridine-2-carboxaldehyde thiosemicarbazone (3-AP,
Triapine (R)) and gemcitabine in advanced non-small-cell lung cancer
with pharmacokinetic evaluation using peripheral blood mononuclear
cells. Invest. New Drugs 2008, 26, 169–173.
(29) Mackenzie, M. J.; Saltman, D.; Hirte, H.; Low, J.; Johnson, C.;
Pond, G.; Moore, M. J. A Phase II study of 3-aminopyridine-2-
carboxaldehyde thiosemicarbazone (3-AP) and gemcitabine in advanced
pancreatic carcinoma. A trial of the Princess Margaret Hospital Phase II
consortium. Invest. New Drugs 2007, 25, 553–558.
(38) Ponka, P.; Richardson, D. R.; Edward, J. T.; Chubb, F. L. Iron
chelators of the pyridoxal isonicotinoyl hydrazone class. Relationship of
the lipophilicity of the apochelator to its ability to mobilize iron from
reticulocytes in vitro. Can. J. Physiol. Pharmacol. 1994, 72, 659–666.
(39) Richardson, D. R.; Ponka, P. The iron-metabolism of the
human neuroblastoma cell - Lack of relationship between the efficacy
of iron chelation and the inhibition of DNA-synthesis. J. Lab. Clin. Med.
1994, 124, 660–671.
(40) Chaston, T. B.; Richardson, D. R. Interactions of the pyridine-
2-carboxaldehyde isonicotinoyl hydrazone class of chelators with iron
and DNA: Implications for toxicity in the treatment of iron overload
disease. J. Biol. Inorg. Chem. 2003, 8, 427–438.
(41) Myhre, O.; Andersen, J. M.; Aarnes, H.; Fonnum, F. Evaluation
of the probes 20,70-dichlorofluorescin diacetate, luminol, and lucigenin as
indicators of reactive species formation. Biochem. Pharmacol. 2003,
65, 1575–1582.
(42) Ghose, A. K.; Crippen, G. M. Atomic physicochemical param-
eters for 3-dimensional-structure-directed quantitative structure-activ-
ity-relationships. 2. Modeling dispersive and hydrophobic interactions. J.
Chem. Inf. Comput. Sci. 1987, 27, 21–35.
(43) Sheldrick, G. M. A short history of SHELX. Acta Crystallogr.,
Sect. A 2008, 64, 112–122.
(44) Farrugia, L. J. ORTEP-3 for windows—A version of ORTEP-III
with a graphical user interface (GUI). J. Appl. Crystallogr. 1997, 30, 565.
(45) Farrugia, L. J. WinGX suite for small-molecule single-crystal
crystallography. J. Appl. Crystallogr. 1999, 32, 837–838.
(46) Richardson, D. R.; Milnes, K. The potential of iron chelators of
the pyridoxal isonicotinoyl hydrazone class as effective antiproliferative
agents 0.2. The mechanism of action of ligands derived from salicylal-
dehyde benzoyl hydrazone and 2-hydroxy-1-naphthylaldehyde benzoyl
hydrazone. Blood 1997, 89, 3025–3038.
(47) Baker, E.; Richardson, D.; Gross, S.; Ponka, P. Evaluation of the
iron chelation potential of hydrazones of pyridoxal, salicylaldehyde and
2-hydroxy-1-naphthylaldehyde using the hepatocyte in culture. Hepatol-
ogy 1992, 15, 492–501.
(48) Becker, E.; Richardson, D. R. Development of novel aroylhy-
drazone ligands for iron chelation therapy: 2-Pyridylcarboxaldehyde
isonicotinoyl hydrazone analogs. J. Lab. Clin. Med. 1999, 134, 510–521.
(49) Richardson, D. R.; Baker, E. The uptake of iron and transferrin
by the human-malignant melanoma cell. Biochim. Biophys. Acta 1990,
1053, 1–12.
(30) Whitnall, M.; Howard, J.; Ponka, P.; Richardson, D. R. A class of
iron chelators with a wide spectrum of potent antitumor activity that
overcomes resistance to chemotherapeutics. Proc. Natl. Acad. Sci. U.S.A.
2006, 103, 14901–14906.
(31) Richardson, D. R.; Sharpe, P. C.; Lovejoy, D. B.; Senaratne, D.;
Kalinowski, D. S.; Islam, M.; Bernhardt, P. V. Dipyridyl thiosemicarba-
zone chelators with potent and selective antitumor activity form iron
complexes with redox activity. J. Med. Chem. 2006, 49, 6510–6521.
(32) Kalinowski, D. S.; Richardson, D. R. Future of toxicology-iron
chelators and differing modes of action and toxicity: The changing face
of iron chelation therapy. Chem. Res. Toxicol. 2007, 20, 715–720.
(33) Kalinowski, D. S.; Yu, Y.; Sharpe, P. C.; Islam, M.; Liao, Y. T.;
Lovejoy, D. B.; Kumar, N.; Bernhardt, P. V.; Richardson, D. R. Design,
(50) Richardson, D. R.; Baker, E. Intermediate steps in cellular iron
uptake from transferrin - Detection of a cytoplasmic pool of iron, free of
transferrin. J. Biol. Chem. 1992, 267, 21384–21389.
(51) Dean, R. T.; Nicholson, P. The action of 9 chelators on iron-
dependent radical damage. Free Radical Res. 1994, 20, 83–101.
6948
dx.doi.org/10.1021/jm200924c |J. Med. Chem. 2011, 54, 6936–6948