Improved cytotoxicity of pyridyl-substituted thiosemicarbazones against MCF-7 when used as metal ionophores

Article abstract of DOI:10.1007/s10534-015-9905-1

Zinc is the second most abundant transition metal in the human body, between 3 and 10 % of human genes encoding for zinc binding proteins. We have investigated the interplay of reactive oxygen species and zinc homeostasis on the cytotoxicity of the thiose

Full text of DOI:10.1007/s10534-015-9905-1

Biometals (2016) 29:157–170
DOI 10.1007/s10534-015-9905-1
Improved cytotoxicity of pyridyl-substituted
thiosemicarbazones against MCF-7 when used as metal
ionophores
.
Fady N. Akladios Scott D. Andrew
Christopher J. Parkinson
.
Received: 2 December 2015 / Accepted: 10 December 2015 / Published online: 18 December 2015
Ó Springer Science+Business Media New York 2015
Abstract Zinc is the second most abundant transi-
tion metal in the human body, between 3 and 10 % of
human genes encoding for zinc binding proteins. We
have investigated the interplay of reactive oxygen
species and zinc homeostasis on the cytotoxicity of the
thiosemicarbazone chelators against the MCF-7 cell
line. The cytotoxicity of thiosemicarbazone chelators
against MCF-7 can be improved through supplemen-
tation of ionic zinc provided the zinc ion is at a level
exceeding the thiosemicarbazone concentration. Elim-
ination of the entire cell population can be accom-
plished with this regime, unlike the plateau of
reactive oxygen species, suggesting that multiple
mechanisms of cytotoxicity can be associated with
the thiosemicarbazones dependant on the level of
metal ion association.
Keywords Copper Á Zinc Á Cytotoxicity Á Reactive
oxygen species (ROS) Á Thiosemicarbazone
Introduction
cytotoxicity
observed
on
thiosemicarbazone
In our previous study relating to the cytotoxic effect of
pyridylketone thiosemicarbazone chelators on MCF-7
we reported that the cytotoxicity of these species
reached a ceiling beyond which the cells were resistant
to the thiosemicarbazone over an extended concentra-
tion range. Even high concentrations of the thiosemi-
carbazones ([20 lM) failed to result in [90 %
cytotoxicity, whereas the corresponding Cu(II) com-
plexes were significantly more potent with IC90 of the
those complexes around 1 lM (Akladios et al. 2015).
We did not observe a similar enhancement for the iron
complexes of the thiosemicarbazone in that study.
Consequently, we set out to ascertain the effects of
other metals in order to attain meaningful levels of
cytotoxicity towards the MCF-7 cell line.
monotherapy. The cytotoxic effects of copper com-
plexes of the thiosemicarbazone are not enhanced by
zinc supplementation, displacement of copper from
the complex being disfavoured. Treatment of MCF-7
with uncomplexed thiosemicarbazone initiates post
G1 blockade alongside the induction of apoptosis, cell
death being abrogated through subsequent supple-
mentation with zinc ion after drug removal. This
would implicate a metal depletion mechanism in the
cytotoxic effect of the un-coordinated thiosemicar-
bazone. The metal complexes of the species, however,
fail to initiate similar G1 blockade and apparently
exert their cytotoxic effect through generation of
Zinc is the second most abundant transition metal in
the human body. Between 3 and 10 % of human genes
encode for zinc-binding proteins (Andreini et al. 2006;
Anzellotti and Farrell 2008). Consequently, it can be
F. N. Akladios Á S. D. Andrew Á C. J. Parkinson (&)
School of Biomedical Sciences, Charles Sturt University,
Orange, NSW 2800, Australia
e-mail: cparkinson@csu.edu.au
123

158
Biometals (2016) 29:157–170
inferred that zinc is essential for proper structure and
function of a wide range of cellular proteins and that
regulation of cellular zinc levels should play a critical
role in cell survival. Zn2? co-ordination sites in bio-
complexes generally comprise S, N and O components
of amino acids, these substituents also being respon-
sible for stabilisation of those proteins in their correct
folding patterns (Loh 2010). Owing to strong interac-
tion of Zn2? ions with proteins, an increase in the
labile zinc pool can initiate non-specific interactions
resulting in misfolding, oxidative stress and cytotoxic
effects, having significant implications on cancer
development (Butler and Loh 2007). Zinc homeostasis
is tightly controlled through the interaction of the
metal with endogenous thiols, particularly metalloth-
ioneins (MTs), which control zinc trafficking through
cysteine/cystine interconversion (Fig. 1) dependant on
the cellular redox status (Maret and Vallee 1998).
The cellular zinc buffering capacity is, similarly,
dependent on endogenous thiols which are redox
sensitive. Such residues can be modified through
oxidants (which oxidise thiols into disulfides), metal
chelators capable of extruding protein bound zinc
(Fig. 1) or alkylating agents (which alkylate nucle-
ophilic thiol residues preventing the zinc–thiol inter-
action) (Maret 2009). Both zinc complexation and
oxidative pressure result in depletion of the pool of
reducing thiols, suggesting that the interplay of zinc
homeostasis and redox homeostasis would be a fertile
avenue to explore in the development of cytotoxic
agents (Maret 2008).
homeostasis and the intracellular bioavailability of zinc
can also control the p53 folding. Treatment of MCF-7
with the non-thiosemicarbazone based membrane-per-
meable chelator N1,N1,N2,N2-tetrakis(pyridin-2-yl-
methyl)ethane-1,2-diamine (TPEN, compound 2,
Fig. 2) resulted in the extrusion of protein bound zinc
ions resulting in protein misfolding (Verhaegh et al.
´
1998; Meplan et al. 2000). Itwould appear reasonable to
conclude that similar extrusion and misfolding would
proceed with other zinc metalloproteins. In a related
study, it was noted that thiosemicarbazones such as
NSC319726 (Fig. 2) could be used as a delivery vehicle
for zinc ions to assist in a zinc templated refolding of
mutant metalloproteins in TOV112D (Yu et al. 2012).
As follows from the above, the thiosemicarbazones
are implicated in zinc transport. The aim of this study,
therefore, is to evaluate the impact of zinc ions on the
cytotoxicity of thiosemicarbazones against MCF-7
and establish the linkage between zinc supplementa-
tion and redox homeostasis.
Materials and methods
Chemical synthesis
All compounds were synthesised according to stan-
dard protocols reported in the literature (Scheme 1)
(Kalinowski et al. 2007; Richardson et al. 2006;
Bernhardt et al. 2008).
The interplay of redox homeostasis and zinc home-
ostasis in cancer cells has been examined with a focus
on the tumour suppressor p53 protein. The p53 metal-
loprotein is one of the zinc-binding proteins that
contains a single zinc ion near its DNA-binding domain
(DBD) responsible for the co-ordination of p53
tetramer, folding of p53 structure and site specific
DNA binding (Kim et al. 2011). The p53 folding is
dependent on the redox status of cysteine residues in the
DBD which bind the central zinc (Loh 2010). Zinc
Cell culture
Cell culture of MCF-7
Human breast adenocarcinoma cells (MCF-7) were
cultured in RPMI 1640 media supplemented with
10 % fetal calf serum, insulin 1 lg/ml, 1 % strepto-
mycin–penicillin and 1 % glutamine. The culture
flasks were incubated in a humidified atmosphere of
5 % CO2 in air at 37 °C. Cells were harvested upon
reaching 80–90 % confluence using Trypsin–EDTA
for 5 min, diluted with equal volume of fetal calf
serum and then centrifuged at 1500 rpm for 5 min.
Chelation
Oxidation
Reduction
S
SH
S
Zn2+
Cell culture of MRC-5
Zn
S
SH
S
Human lung fibroblast cells (MRC-5) were cultured in
MEM media supplemented with 10 % fetal calf serum
Fig. 1 Perturbation of cellular zinc buffering capacity through
structural modification of zinc-binding thiols (Maret 2009)
123

Biometals (2016) 29:157–170
159
Fig. 2 Structures of a
selection of metal chelators
with reported anticancer
activity (substitution
patterns of 4a–i can be found
in Table 1)
N
S
N
N
S
N
N
N
NH
NH
N
N
N
N
N
N
N
(1) Dp44mT
(2) TPEN
(3) NSC319726
O
N
(CH2)5
R2
N
N
HO
S
O
R3
OH
O
N
NH
O
NH
H2N
N
N
(CH2)5
N
OH
N
HN
HN
N
N
R1
(H2C)5
O
O
(4) a-i
(5)
(6) DFO
R2
N
R2
N
S
N
S
R3
R3
AcO
Cu
NH
N
N
O
H2N NH R2
N
N
N
N
(ii)
(i)
R1
R1
R1
+
S
R3
Scheme 1 Reagents and conditions: (i) Ethanol, acetic acid (cat.), reflux; (ii) DMF, Cu(OAc)2, reflux, 2 h
and 1 % of non-essential amino acids. Cells were
incubated in a humidified atmosphere and harvested
when 90 % confluence was reached.
was removed and replaced with fresh media. After-
wards, 20 lL of MTS reagent was added into each
well of the assay plate containing 100 lL in culture.
The plates were then incubated for 3 h before mea-
suring the absorbance at 492 nm using 96-well plate
reader.
Cell proliferation assay
Thiosemicarbazone chelators (with and without sup-
plementation of varying quantities of copper(II)
acetate or zinc nitrate), Cu(II) complex of (4b) and
Zn complex of (4b) were dissolved in dimethyl
sulfoxide (DMSO) at a stock concentration of
10 mM and were used at the concentrations indicated
by dilution in the RPMI culture media before addition
to MCF-7 or MRC-5 cells.
MRC-5 cells
Approximately 3000 cells were seeded into 96-well
plates and allowed to settle for 48 h before adding
individual compounds or combinations. After 48 h/
37 °C incubation, full confluence in the control wells
was reached and the same assay protocol mentioned in
MCF-7 was followed.
MCF-7 cells
Formation of the zinc complex in solution
Approximately 10,000 cells were seeded into 96-well
plates and allowed to settle for 24 h before adding
individual compounds or combinations with and
without zinc nitrate selected concentrations. After
72 h/37 °C incubation with the compounds, the media
Thiosemicarbazone chelator (4b) and zinc nitrate
heptahydrate were dissolved at stock concentrations
of 10 mM in Tris buffer with 25 % DMSO (pH 7.4).
The stock solutions were then diluted using Tris/
123

160
Biometals (2016) 29:157–170
DMSO buffer to 100 lM of the thiosemicarbazone
and 50–1000 lM of zinc nitrate. The measurements
were the following: thiosemicarbazone (4b) (20 lM),
4b (20 lM) plus zinc nitrate (10 lM), 4b (20 lM)
plus zinc nitrate (20 lM) and 4b (20 lM) plus zinc
nitrate (200 lM). The experiment was carried out
using Nanodrop 2000cÒ spectrophotometer.
Dickinson FACSCalibur). Quantitative analysis of
cell cycle distribution was performed using Flowing
Software 2.5.1.
Determination of reactive oxygen levels resultant
from thiosemicarbazone exposure
Approximately 106 cells were seeded into 10 cm
dishes, and then allowed to settle for 48 h. Afterwards,
the thiosemicarbazone or metal/thiosemicarbazone
combinations were added and left in contact with the
cells for 10, 13 or 22 h. The cells were then harvested
via trypsinisation for 5 min. The cell concentration of
the samples was adjusted to 106 cells per 1 mL of
complete media for FACS analysis. The positive
control used was Menadione (100 lM) which was
incubated with a sample of cells for 1 h. The samples
were then treated with CellROXÒ Green reagent at a
final concentration of 500 nM for 30 min at 37 °C
protected from light. The fluorescence at 488 nm
excitation and corresponding emission was measured
and the data were analysed by FACS (Becton–
Dickinson FACSCalibur) followed by analysis using
Flowing Software 2.5.1.
Spectroscopic determination of Zn2? replacement
from Zn-Ligand chelate
Thiosemicarbazone chelator (4b), copper (II) acetate
monohydrate, zinc nitrate heptahydrate were dis-
solved at stock concentrations of 10 mM in Tris
buffer 10 mM with 25 % DMSO (pH 7.4). The stock
solutions were then diluted using Tris/DMSO buffer to
100 lM of 4b, 100 lM of the Cu (II) acetate and
2000 lM of zinc nitrate. The measurements were the
following: 4b (20 lM), 4b (20 lM) plus Cu (II)
acetate (20 lM) and 4b (20 lM) plus zinc nitrate
(200 lM). The latter mixture of the 4b (20 lM) plus
Zn nitrate (200 lM) was then mixed with Cu (II)
acetate (20 lM) and the absorbance was then mea-
sured every 1.5 min. The experiment was carried out
using Nanodrop 2000cÒ spectrophotometer.
Flow cytometric analysis of chelator treated MCF-
7 cells using propidium iodide (PI) stain
Results and discussion
The influence of combining Zn2?
with thiosemicarbazone metal chelators on MCF-7
cytotoxicity
Approximately 106 cells were seeded into 10 cm
dishes, and then allowed to settle for 48 h. Afterwards,
the drugs/combinations were added and left in contact
with the cells for 24 h followed by preparation of the
samples for FACS analysis. Briefly, the supernatants
were collected into 10 mL tubes, and then centrifuged
separately, whereas the adherent cells were harvested
via trypsinisation for 5 min, collected onto 5 mL of
FBS. The pellets were then resuspended into cold PBS
and combined with the corresponding supernatant
pellets for each drug/combinations and then washed
twice with cold PBS. The cells were then fixed using
1 mL of ice cold 70 % Ethanol for 30 min. The cells
were centrifuged at 1500 rpm for 5 min. After subse-
quent washes with PBS, the cell pellets were then
treated with 50 lL of RNase solution (100 lg/mL) to
ensure that DNA is only stained and incubated for
15 min at 37 °C. The propidium iodide stain (50 lg/
mL) was then added, left in the dark for 30 min and
then the samples were analysed by FACS (Becton–
In an attempt to improve the cytotoxicity of pyridyl
thiosemicarbazones against MCF-7, we studied the
cytotoxicity of the chelator Dp44mT [(see Fig. 2,
structure 1), as a highly potent representative of the
thiosemicarbazone class (Jansson et al. 2010)] sup-
plemented with different metals including Fe2?, Fe3?
,
Cu2? and Zn2?. We rationalized that the thiosemicar-
bazone would be efficient in delivering the metal ion
across the cell membrane. Significantly, the cytotoxic
activity was enhanced in the presence of the redox
active metal Cu2? but not in the presence of iron
(irrespective of oxidation state). Supplementation with
the non redox active Zn2? resulted in enhanced
cytotoxicity, albeit at levels substantially higher than
the copper levels required for similar enhancement
(Fig. 3). Such differences in the cytotoxicity did not
coincide with potency of different preformed
123

Biometals (2016) 29:157–170
161
thiosemicarbazone (Fig. 4b). The optimum zinc ion
concentration for maximum cytotoxicity appeared to
be between 10 and 20 lM. Zinc ion exerted no
cytotoxic effect over the concentration range studied
in the absence of the thiosemicarbazone. Similarly,
zinc ion played no role in the cytotoxic effect of more
classical iron chelators such as deferoxamine (DFO,
Fig. 4c). In a study relating to the renaturation of a
misfolded p53 protein, Yu et al. noted a marginally
enhanced cytotoxicity of a related thiosemicarbazone
(NSC319726, Fig. 2). (Yu et al. 2012) Refolding of a
mutant p53, however, is unlikely to play a role in the
present study where the MCF-7 exhibits wild type p53
status.
As a result of the cytotoxicity enhancement resul-
tant from zinc supplementation, we set out to re-
evaluate the cytotoxicity profile of thiosemicar-
bazones at fixed levels of zinc supplementation. As
expected, improved cytotoxicity was observed for all
thiosemicarbazones subject to the study (For 4b see
Fig. 5a) with full elimination of the cell being
possible—unlike the non zinc supplemented thiosemi-
carbazone (Table 1). As thiosemicarbazone concen-
tration approached the zinc concentration, however, a
lessening in cytotoxic effect was observed for
thiosemicarbazones subject to the study (one excep-
tion). A similar enhancement was observed for the
related chelator, di-2-pyridyl ketone isonicotinoylhy-
drazone, albeit at much higher dosages (Fig. 5b). The
enhancement in cytotoxicity resultant from zinc
supplemementation could be plotted as an ‘‘enhance-
ment factor’’, this being the ratios of the fraction of
cells affected with and without zinc. It can be seen that
Fig. 3 Cytoxicity of a 1 lM solution of the thiosemicarbazone
Dp44mT with metal ion supplementation
complexes of Dp44mT on the leukemia line HL-60
(Bernhardt et al. 2008).
We previously demonstrated that the thiosemicar-
bazone (4b) showed a flattening of the dose response
curve resulting in a fraction affected (Fa) value around
0.5 from a dose of 1 lM through to 25 lM against
MCF-7 cells (Akladios et al. 2015). This plateau in
cytotoxicity was observed with all thiosemicarbazones
in the class with the exception of the relatively inactive
parent (Compound 4a, see Table 1) (Fig. 4a). Since
we had noted that the cytotoxicity of Dp44mT was
dependent on metal ion supplementation and the
enhancement of cytotoxicity was metal ion concen-
tration dependent, we examined the impact of a
concentration gradient of zinc (2.5–20 lM) on the
cytotoxicity of a fixed (1 lM) concentration of the
Table 1 Cytotoxicity data for thiosemicarbazones subject to this study under zinc supplementation conditions
Compound
R1
R2
R3
IC50 SD (lM) (with Zn2? 15 lM)
IC95 (lM) (with Zn2? 15 lM)
4a
4b
4c
4d
4e
4f
2-pyridyl
2-pyridyl
2-pyridyl
2-pyridyl
2-pyridyl
Phenyl
Me
H
H
Not calculated
0.8 0.1
[40
1
Me
Me
H
Et
Not calculated
1.9 0.1
25
3
Ph
H
(4-OMe)Ph
H
2.0 0.13
2.8 0.2
3
Et
H
4
4g
4h
4i
Me
Et
Me
H
1.9 0.2
3
Me
4.3 0.4
6
Me
Ph
H
4.3 0.3
6
The IC50 values of the compounds tested were calculated by GraphPad Prism v5.0 using non-linear regression analysis
123

162
Biometals (2016) 29:157–170
Fig. 4 a Concentration related cytotoxicity profile of thiosemi-
carbazones against MCF-7 used in this study in comparison to
reference materials cisplatin and doxorubicin. b concentration
dependent effect of zinc on cytotoxicity in the presence and
absence of thiosemicarbazone 4b. c Concentration dependent
effect of zinc ions on cytotoxicity of deferoxamine (DFO)
Fig. 5 a Dose related cytotoxicity profile of thiosemicarbazome 4b at varying levels of zinc supplementation. b Dose related
cytotoxicity profile of bis-2-pyridyl ketone isonicotinoylhydrazone at varying levels of zinc supplementation
higher dosage levels of the thiosemicarbazone result in
lesser degrees of enhancement (Fig. 6).
supplementation on the cytotoxicity of copper-
thiosemicarbazone complexes. We rationalized that
the strong association between copper and the
thiosemicarbazone may impact on the ability of the
Our previous study demonstrated the cytotoxicity
difference between the uncomplexed thiosemicar-
bazone and the copper complex of the thiosemicar-
bazone against the MCF-7 cell line (Akladios et al.
2015). Consequently, we examined the effect of zinc
´
complex to transport zinc into the cell (Gaal et al.
2014). On this basis we reasoned that zinc supple-
mentation should play no role in the cytotoxic profile
123

Biometals (2016) 29:157–170
163
Fig. 6 Enhancement of cytotoxicity [EF = (Fa in presence of
zinc)/(Fa in absence of zinc)] of thiosemicarbazones against
MCF-7 in the presence of zinc
Fig. 8 Cytotoxicity of pre-formed zinc complex of 4b (ZnL2)
in the presence and absence of Copper(II)
levels were substantially greater than thiosemicar-
bazone levels (see Fig. 5 above). Addition of cop-
per(II) to the pre-formed zinc complex resulted in
increased cytotoxicity—indicative of the replacement
of zinc with copper. This result suggested that the
preformed zinc complex and the active form of the
thiosemicarbazone under zinc supplementation con-
ditions may not be equivalent (Fig. 8).
Complexation of zinc by the thiosemicarbazone
in solution
Fig. 7 Impact of zinc supplementation on cytotoxicity of
copper complexes of thiosemicarbazones against MCF-7
of such copper complexes. To confirm this, we
examined the dose response curve of the 1:1 copper(II)
complex of the chelator 4b in both the presence and
absence of zinc supplementation. As expected, the
cytotoxic profile of the copper complex was unaltered
on zinc supplementation (Fig. 7). This observation
would appear to support our initial hypothesis that the
copper complex is ineffective as a zinc transport
vehicle.
Due to the differing cytotoxicity profiles of the
preformed zinc complex and the behaviour of the
thiosemicarbazone under zinc supplementation, we
examined the nature of the zinc—thiosemicarbazone
complex in solution. The uncomplexed thiosemicar-
bazone showed an absorption maximum at 329 nm.
Addition of a half equivalent of zinc ion resulted in the
appearance of a new absorption maximum at 405 nm.
No trend towards full complexation of the free
thiosemicarbazone was observed with time—suggest-
ing that the ZnL2 species that was isolated by
Bernhardt et al. (2008) does not form under the mild
and dilute conditions employed in the current study.
Addition of excess zinc results in the same species
with an absorption maximum at 405 nM—suggesting
that the 1:1 complex is the predominant form (Fig. 9).
As expected, addition of copper ions to the matrix
rapidly generated the copper complex of the thiosemi-
carbazone (absorption maximum 422 nm) even in the
presence of excess levels of zinc ion (Fig. 10). This
result is in full concordance with the stability constants
of metal—thiosemicarbazone complexes reported by
Gaal et al. (2014).
Cytotoxicity of the preformed ZnL2 complex
The Zn complex of 4b (ZnL2) was synthesised
according to the method of Bernhardt et al. (2008).
The dose response profile of the preformed complex
was analogous to that observed for the uncoordinated
ligand, reaching a ceiling at around 50 % of cells
killed. It should be noted, however, that solubility
limitations of this complex became apparent at
dosages between 10 and 20 lM. This result was not
surprising as we had observed that the enhancement of
thiosemicarbazone cytoxicity resultant from zinc
supplementation was most significant where zinc
123

164
Biometals (2016) 29:157–170
Fig. 9 The nature of the
zinc-thiosemicarbazone
complex in solution. [Blue
line (no absorption at 405
nm) thiosemicarbazone 4b
(20 lM), Green Line
(median absorption at 405
nm) 4b (20 lM) and Zinc
ions (10 lM), Red
Line (highest absorption at
405 nm) 4b (20 lM) and
Zinc ions (200 lM)]
Fig. 10 a Changes in the UV spectrum of compound 4b
(20 lM) when combined with Cu2? (20 lM) and Zn2?
(200 lM). The blue line (with absorption maximum at
329 nm) represents the ligand, the black line (with absorption
maximum at 405 nm) represents Zn-ligand chelate and the red
line (with absorption maximum at 422 nm) represents Cu-ligand
chelate; b changes in the UV spectrum of Zn-ligand chelate
upon adding Cu2?, the absorbance was measured every 1.5 min
The strong nature of the copper-thiosemicarbazone
complex in solution led us to postulate that the pre-
formed copper complex and additional thiosemicar-
bazone in the presence of zinc may form a synergistic
drug combination with concomitant enhancement of
cytotoxicity and reduction in the levels of the copper
complex required for treatment of the MCF-7 cell.
cytotoxicity and reduction in the levels of the copper
complex required for treatment of the MCF-7 cell.
Copper(II) complexes of thiosemicarbazone chela-
tors generate ROS via redox cycling of Cu2?/1? and
participation in cellular Fenton chemistry, leading to
induction of oxidative stress and cell death (Jansson
et al. 2010). The IC50 of Cu complex of (4b) on MCF-7
was determined to be 0.7 lM. The combination of the
chelator 4b and its Cu complex (0.6 lM) did not show
any improvement in the cytotoxicity irrespective of
the amount of 4b added. The addition of Zn2?
(15 lM) to the combination of both agents, however,
resulted in improvement of cytotoxicity (Fig. 11a).
This effect could be used to lower the concentration of
the copper complex required for complete elimination
of the MCF-7 cells (Fig. 11b). This combination
Combination of chelator and Zn2? with pre-
formed copper complexes
The strong nature of the copper- thiosemicarbazone
complex in solution led us to postulate that the pre-
formed copper complex and additional thiosemicar-
bazone in the presence of zinc may form a synergistic
drug combination with concomitant enhancement of
123

Biometals (2016) 29:157–170
165
Fig. 11 a The impact of
combinations of the
thiosemicarbazone with its
copper complex and zinc on
MCF-7 cytotoxicity.
b Safety of three component
combinations as represented
by lack of cytotoxicity to
MRC-5 fibroblasts
displayed a good safety profile (Fig. 10b) against
MRC-5 fibroblasts, complete elimination of the MCF-
7 being feasible at levels of the combination without
visible damage to the fibroblast.
different Zn2? concentrations (0–30 lM) and the cells
were incubated for further 48 h before submitting the
cells to MTS cell viability assay. Adding Zn2? with
the replacement of media resulted in increased
survival of MCF-7 cells when compared to cells not
subjected to Zn2? supplementation. The increased
survival showed a dose dependence with respect to
Zn2? supplementation levels indicative of a salvage
mechanism. This effect was not as significant at higher
thiosemicarbazone concentrations, presumably due to
the concentration dependence of cellular damage
(Fig. 12a). This result suggests that reversible zinc
extrusion from metalloproteins may be implicated in
cytotoxicity when the thiosemicarbazone is dosed as
the uncoordinated ligand. This observation is in
concordance with an earlier study concluding that
Effect of adding Zn2? after removing the chelator/
Cu complex
We reasoned that one mechanism of cellular damage
resultant from treatment with the thiosemicarbazone
in the absence of metal supplementation may be
intracellular zinc depletion. To evaluate this effect, the
thiosemicarbazone chelator 4b (10 and 15 lM) was
incubated with MCF-7 cells for 24 h; a time at which
cells appeared to be viable. The media was then
removed, replaced with fresh media containing
123

166
Biometals (2016) 29:157–170
Fig. 12 a Effect of zinc
replacement after treatment
of MCF-7 with sub-lethal
levels of thiosemicarbazone
(4b). b Effect of zinc
replacement after a short
term treatment of MCF-7
with the copper complex of
thiosemicarbazone (4b)
p53 folding can be restored in MCF-7 upon treatment
with zinc after intracellular zinc dequestration using
TPEN, the renaturation being attributable to incorpo-
available metals, are implicated in the anticancer
activity of the thiosemicarbazones.
´
ration of zinc in the zinc depleted protein (Meplan
Impact of thiosemicarbazones and metal
complexes on the cell cycle
et al. 2000). The previous study was, however, limited
to p53 rather than overall zinc depletion.
A similar experiment was conducted using the
Cu(II) complex (1:1) of the metal chelator (4b, 1 lM)
which was pre-incubated with MCF-7 cells for 6,12
and 24 h prior to removal and replacement of the
media. A series of zinc concentrations was added to
the fresh media (0–30 lM). Unlike the thiosemicar-
bazone itself, Zn2? had no effect on salvaging cell
viability after treatment with the cytotoxic copper
complex (Fig. 12b), all cells damaged during the
initial treatment proceding to cell death. This would
suggest that reversible metal extrusion plays a limited
role at best in the cytotoxic mechanism of copper
complexes of the thiosemicarbazones, the activity of
which appears rapid and irreversible. It can be
concluded that several cytotoxic mechanisms, depen-
dent on the level of association of the chelator with
Previous studies have shown that metal chelators can
induce cell cycle arrest in cancer cell lines, the iron
selective chelator DFO being reported to initiate G1
arrest (Siriwardana and Seligman 2015). Since com-
pound (4b) demonstrated significant anticancer activ-
ity against MCF-7 and this activity was enhanced in
the presence of Cu2? or Zn2?, we investigated the
effects of the chelator on the cell cycle using flow
cytometry. We reasoned that the apparent lack of a
zinc depletion mechanism in the cytotoxicity resultant
from treatment with the thiosemicarbazone—metal
complex (as indicated from the failure of the zinc
replacement test in Fig. 12 above) may manifest itself
in different staging within the cell cycle.
Untreated MCF7 cells in exponential growth were
characterised by populations of cells in G1 (65 %), S
123

Biometals (2016) 29:157–170
167
Fig. 13 a Population
distribution across phases of
the cell cycle for untreated
MCF-7 cells. b Population
distribution across phases of
the cell cycle for MCF-7
cells treated with
thiosemicarbazone 4b
(10 lM) only. c Population
distribution across phases of
the cell cycle for MCF-7
cells treated with
thiosemicarbazone 4b
(1 lM) plus Zn2? (15 lM).
d Population distribution
across phases of the cell
cycle for MCF-7 cells
treated with
thiosemicarbazone 4b
(1 lM) plus Cu2? (1 lM)
(12 %) and G2/M (23 %) phase of the cell cycle
(Fig. 13). Following treatment of MCF7 cells with
(4b) at a concentration of 10 lM for 24 h, the
proportion of cells in the G1 phase increased to
75 % with a concomitant decrease in the percentage of
cells in the S (9 %) and G2/M (16 %) phases of the cell
cycle. This would appear to be indicative of cell cycle
blockade subsequent to the G1 phase as would be in
accordance with the findings of the DFO study. In
contrast, this G1 block was relieved if the thiosemi-
carbazone was supplemented with Zn2? (15 lM and
Cu2? 1 lM), the proportion of cells in the G1 phase
being identical to the untreated control. The concen-
trations of the agents used in this study were
standardized to a level expected to result in approx-
imately 50 % cell death if the cells were exposed to a
72 h treatment. This data suggests different mecha-
nisms of cytotoxicity for the chelator depending on the
presence and availability of metal ions.
Table 2 Changes in SubG1 peak distribution as result of
chelator/metal treatment
Treatment protocol
Sub G1 (%)
Negative control
2.5
8.4
11
4b-10 lM
4b-1 lM plus Zn21 15 lM
4b-2 lM plus Zn21 15 lM
4b-1 lM plus Cu21 1 lM
25
15.5
incubated with either thiosemicarbazone (4b) alone
or in the presence of additional copper or zinc for 24 h.
Flow cytometry was used to evaluate the cultures for
the presence a sub G1 peak indicative of apoptotic
cells.
In comparison to untreated MCF7, the treated cells
showed a reduction in cell size and side scatter signal
which is characteristic of cells undergoing late apop-
tosis (Darzynkiewicz et al. 2005; Williams 2004).
Untreated MCF7 cells exhibited only a small sub G1
(2.5 %) peak consistent with a low population of
apoptotic cells (Table 2).
Quantification of apoptotic cell death resultant
from thiosemicarbazone treatment
Treatment with the thiosemicarbazone (4b) pro-
duced an around threefold increase in the sub G1
To assess whether G1 blockade induced by the
chelator leads to apoptosis, MCF-7 cells were
123

168
Biometals (2016) 29:157–170
peak (8.4 %) consistent with the induction of
apoptosis. Addition of copper(II) or zinc ions to
the matrix resulted in further increase in apoptosis
as indicated by the increased area of the sub-G1
peak (11 and 15.5 %) respectively. This increase in
sub G1 peak is indicative of a mechanism typical
for DNA damaging agents (King and Radicchi-
Mastroianni 2002; Williams 2004). The magnitude
of apoptotic or sub-G1 peak was dose dependent,
doubling of the thiosemicarbazone concentration
resulting in a 2–3 fold elevation in the sub-G1
population.
Reactive oxygen species resulting
from thiosemicarbazone exposure
In order to verify that cytotoxicity resultant from
thiosemicarbazone treatment was preceded by ele-
vated levels of reactive oxygen species (ROS), we set
out to establish that ROS levels rose early in the
treatment protocol. The levels of ROS detected in both
treated and untreated MCF-7 cells were determined by
flow cytometric detection of CellROXÒ Green dye
incorporation. The CellROXÒ reagent, which is non
fluorescent in its reduced form, exhibits a strong
fluorescent signal on oxidation resultant from contact
with intracellular ROS. Menadione (a known gener-
ator of ROS, 100 lM) was used as a positive control to
verify that ROS could be detected in the MCF-7
system (Fig. 14).
Treatment of MCF-7 cells with the thiosemicar-
bazone (4b, 1 lM) in combination with either Cu2?
(1 lM) or Zn2? (15 lM), was associated with a 4 or 5
fold rise respectively, in the intracellular content of
ROS (Fig. 15a, b). This was in contrast to the levels of
ROS detected in MCF-7 cells treated with the
thiosemicarbazone (4b, 10 lM) alone, or in MCF7
cells supplemented with exogenous Cu2? (1 lM) or
Zn2? (15 lM) in the absence of the thiosemicar-
bazone—all of which showed minimal increase in
intracellular ROS relative to untreated MCF-7 cells.
Forward and side scatter data (Fig. 14) suggested that
MCF-7 cells treated with Zn and Cu complexes also
displayed a larger population of apoptotic cells by
comparison with untreated cells. Similarly, the cyto-
toxic profile of the respective treatments showed the
same trend—suggesting that cytotoxicity proceeds
through an apoptotic mechanism and is subsequent to
Fig. 14 a.1–d.1 Dot plots of forward scatter versus side scatter;
the red population (higher level of forward and side scatter)
represents healthy cells, whereas the blue population (lower left
data population) represents the shrunken cells/debris, a.2–d.2
histograms of untreated and treated MCF-7 cells; region H5
represents the sub G1 population undergoing apoptosis
reactive oxygen generation. It should be noted that the
increase in ROS observed is derived from gated data
which excludes contamination with dead or apoptotic
cells.
123

Biometals (2016) 29:157–170
169
Fig. 15 Comparison of
ROS signals detected by
CellROXÒGreen a 4b-1 lM
plus Cu21 1 lM, b 4b-
1 lM plus Zn21 15 lM,
c Cu2? 1 lM (Green), Zn2?
15 lM (Blue), d 4b-1 lM
Conclusions
the thiosemicarbazones dependant on the level of
metal ion association.
The cytotoxicity of thiosemicarbazone chelators
against MCF-7 can be improved through supplemen-
tation of ionic zinc provided the zinc ion is at a level
exceeding the thiosemicarbazone concentration. Elim-
ination of the entire cell population can be accom-
plished with this regime, unlike the plateau of
Acknowledgments The
authors
acknowledge
the
contribution of Dr Gregg Maynard for assistance in setting up
flow cytometry studies. F Akladios acknowledges the receipt of
an Australian Postgraduate Award (APA). CJP wishes to thank
the CSU Pharmacy Foundation for a grant partially funding this
study. CJP and SDA thank the Kolling Institute (Royal North
Shore Hospital) for the donation and characterization of the
MCF-7 cell line employed in this study.
cytotoxicity
observed
on
thiosemicarbazone
monotherapy. The cytotoxic effects of copper com-
plexes of the thiosemicarbazone are not enhanced by
zinc supplementation, displacement of copper from
the complex being disfavoured. Treatment of MCF-7
with uncomplexed thiosemicarbazone initiates post
G1 blockade alongside the induction of apoptosis, cell
death being abrogated through subsequent supple-
mentation with zinc ion after drug removal. This
would implicate a metal depletion mechanism in the
cytotoxic effect of the un-coordinated thiosemicar-
bazone. The metal complexes of the species, however,
fail to initiate similar G1 blockade and apparently
exert their cytotoxic effect through generation of
reactive oxygen species, suggesting that multiple
mechanisms of cytotoxicity can be associated with
Author contribution CJP and SDA designed the research
programme. FNA conducted all research, analysed data and
assembled the draft manuscript. All authors read and approved
the final content.
References
Akladios FN, Andrew SD, Parkinson CJ (2015) Selective
induction of oxidative stress in cancer cells via synergistic
combinations of agents targeting redox homeostasis.
Bioorgan Med Chem
Andreini C, Banci L, Bertini I, Rosato A (2006) Counting the
zinc-proteins encoded in the human genome. J Proteome
Res 5(1):196–201
123

170
Biometals (2016) 29:157–170
Anzellotti A, Farrell N (2008) Zinc metalloproteins as medicinal
targets. Chem Soc Rev 37(8):1629–1651
Loh SN (2010) The missing zinc: p53 misfolding and cancer.
Metallomics 2(7):442–449
Bernhardt PV, Sharpe PC, Islam M, Lovejoy DB, Kalinowski
DS, Richardson DR (2008) Iron chelators of the
dipyridylketone thiosemicarbazone class: precomplexation
and transmetalation effects on anticancer activity. J Med
Chem 52(2):407–415
Butler JS, Loh SN (2007) Zn2 ? -dependent misfolding of the
p53 DNA binding domain. Biochemistry 46(10):2630–
2639
Maret W (2008) Metallothionein redox biology in the cytopro-
tective and cytotoxic functions of zinc. Exp Gerontol
43(5):363–369. doi:10.1016/j.exger.2007.11.005
Maret W (2009) Molecular aspects of human cellular zinc
homeostasis: redox control of zinc potentials and zinc
signals. Biometals 22(1):149–157
Maret W, Vallee BL (1998) Thiolate ligands in metallothionein
confer redox activity on zinc clusters. Proc Natl Acad Sci
95(7):3478–3482
Darzynkiewicz Z, Huang X, Okafuji M, King MA (2005)
Cytometric methods to detect apoptosis. Cytometry 75:307
´
Meplan C, Richard M-J, Hainaut P (2000) Metalloregulation of
´
´
´
´
}
Gaal A, Orgovan G, Polgari Z, Reti A, Mihucz VG, Bosze S,
Streli C (2014) Complex forming competition and in vitro
toxicity studies on the applicability of di-2-pyridylketone-
4,4,-dimethyl-3-thiosemicarbazone (Dp44mT) as a metal
chelator. J Inorg Biochem 130:52–58
the tumor suppressor protein p53: zinc mediates the
renaturation of p53 after exposure to metal chelators
in vitro and in intact cells. Oncogene 19(46):5227–5236
Richardson DR, Sharpe PC, Lovejoy DB, Senaratne D, Kali-
nowski DS, Islam M, Bernhardt PV (2006) Dipyridyl
thiosemicarbazone chelators with potent and selective
antitumor activity form iron complexes with redox activity.
J Med Chem 49(22):6510–6521
Siriwardana G, Seligman PA (2015) Iron depletion results in Src
kinase inhibition with associated cell cycle arrest in neu-
roblastoma cells. Physiol Rep 3(3):e12341
Verhaegh GW, Parat MO, Richard MJ, Hainaut P (1998)
Modulation of p53 protein conformation and DNA-binding
activity by intracellular chelation of zinc. Mol Carcinog
21(3):205–214
Jansson PJ, Sharpe PC, Bernhardt PV, Richardson DR (2010)
Novel thiosemicarbazones of the ApT and DpT series and
their copper complexes: identification of pronounced redox
activity and characterization of their antitumor activity.
J Med Chem 53(15):5759–5769
Kalinowski DS, Yu Y, Sharpe PC, Islam M, Liao Y-T, Lovejoy
DB, Richardson DR (2007) Design, synthesis, and char-
acterization of novel iron chelators: structure-activity
relationships of the 2-benzoylpyridine thiosemicarbazone
series and their 3-nitrobenzoyl analogues as potent antitu-
mor agents. J Med Chem 50(15):3716–3729
Kim DH, Kundu JK, Surh YJ (2011) Redox modulation of p53:
mechanisms and functional significance. Mol Carcinog
50(4):222–234
King MA, Radicchi-Mastroianni MA (2002) Effects of caspase
inhibition on camptothecin-induced apoptosis of HL-60
cells. Cytometry 49(1):28–35
Williams, O. (2004). Flow cytometry-based methods for apop-
tosis detection in lymphoid cells . In: Apoptosis methods
and protocols. Springer, New York, pp. 31-42
Yu X, Vazquez A, Levine AJ, Carpizo DR (2012) Allele-
specific p53 mutant reactivation. Cancer Cell 21(5):614–
625
123