J Biol Inorg Chem
out to examine the nature of that coordinating group. The
1,3-diformylphenyl bis-thiosemicarbazones with a central
phenolic system (Bis 2 a–c, Fig. 2) demonstrated cytotox-
icity profiles (Fig. 3f) analogous to the pyridyl substituted
bis-thiosemicarbazones (Bis 1a–c, Fig. 2) and were, once
again, superior to the TSC lacking the central coordination
site.
We then evaluated the effect of replacing the sulphur
atom of the bis-thiosemicarbazone component with oxy-
gen as illustrated in Bis (3a). The already weak activity of
Bis (1a) was diminished further, rendering the compound
totally inactive.
n-butyl as in Bis (1e), the cytotoxic activity was largely lost.
The cytotoxicity of some of the bis-thiosemicarbazones in
this study was observed to drop at concentrations >1 µM
(when exceeding the 1:1 ratio of ligand to copper, Fig. 3g).
We rationalised that this decline in cytotoxicity could be
attributed to formation of higher order Cu(II) complexes
(not the 1:1 complex) with either poor cellular penetration
profiles or with lower cytotoxic activity. Upon supplement-
ing the non pyridyl chelators Bis(1k–m) with Cu2+ ions,
we did not observe any improvement relative to the unsup-
plemented ligands, indicating that the presence of central
metal binding motif is a key feature for the cytotoxicity of
bis-thiosemicarbazone-Cu(II) complexes.
We previously demonstrated that the cytotoxicity of
Dp44mT to MCF-7 cells was significantly improved when
supplemented with Cu2+ ions (1 µM), and we attributed
this to the efficiency of the ligand in delivering Cu2+ ions
across cell membranes and subsequent participation in
intracellular Fenton chemistry [16]. Treatment of MCF-7
with Cu2+ in the absence of ligand at levels up to 10 µM
resulted in no effect on cell viability over the time course
of this assay. In an attempt to improve the cytotoxicity of
the bis-thiosemicarbazone chelators subject to the current
work, the same approach of Cu2+ ion supplementation
was adopted and the cytotoxicity to MCF-7 cells was re-
evaluated over a ligand concentration range between 0.01
and 10 µM. The formation of the metal complex could be
confirmed through the shift in the UV absorption maxi-
mum to longer wavelength. HPLC–MS analysis of the
solution adduct resultant from the combination of Bis (1d)
with cupric ion was indicative of a species of the form
Cu2L [calculated for Cu2(C15H21N7S2), [Cu2L+] requires
488.98864, 490.98676, 492.98491, found 488.98807,
490.98641, 492.98491 (Variance 1.1 ppm)]. Generally, sup-
plementation with Cu2+ (1 µM) enhanced the cytotoxicity
of Bis (1a–1j) with full elimination of cancer cells being
possible at low chelator concentrations (Fig. 3g). The high-
est enhancement observed was with the chelator Bis (1g)
supplemented with Cu2+ (>150 fold improvement of IC50,
IC50 of the Cu(II) complex 0.27 µM). We observed, how-
ever, that improvement was limited by the type and size of
substitution on the N4-termini of pyridyl bis-thiosemicar-
bazones. In the presence of exogenous copper (analogous
to formation of an extracellular copper complex) the pyri-
dyl bis-thiosemicarbazones could be considered to be a
class of equal or better potency against MCF-7 when com-
pared to the Dp44mT reference.
Unlike bis-thiosemicarbazones with the pyridyl core, the
cytotoxicity of bis-thiosemicarbazones with a phenolic core
(Bis 2a–c) did not exhibit useful cytotoxic profiles—even
after Cu supplementation. Only the bis-thiosemicarbazone
lacking substitution on the N4-terminus (Bis 2a, Fig. 3h)
showed a significant cytotoxicity enhancement on Cu sup-
plementation (IC50 0.2 µM, Table 1), suggesting that size
or lipophilicity of the N-terminus was a critical driver in
the cytotoxicity of this series. The copper supplemented
Bis (2a) showed potency superior to Dp44mT under similar
conditions.
We subsequently evaluated the cytotoxicity of the bis-
thiosemicarbazones synthesised against the MelRm mela-
noma cell line. Unlike MCF-7 cells, MelRm cells were sen-
sitive to the reference iron chelator DFO (IC50 6 µM). It was
not, therefore, surprising to see Dp44mT showing potent
anticancer activity against MelRm (IC50 0.3 µM) with full
killing achieved around 5 µM (Fig. 4a). The N4-unsub-
stituted pyridyl bis-thiosemicarbazone (Bis 1a) showed
weak cytotoxic profile similar to that demonstrated against
MCF-7 (Fig. 4b). Monsubstitution generally improved the
cytotoxicity against MelRm relative to the unsubstituted
derivative Bis (1a). Monosubstituted analogues Bis (1b, 1f
and 1g) demonstrated moderate cytotoxicity improvement
(IC50 0.8–5.0 µM) (Fig. 4b, c) whereas the phenyl substi-
tuted derivative Bis (1c) showed only limited cytotoxicity
(IC50 >15 µM). We noted also that Bis (1f), with the poten-
tially ligating pyridyl motif on the N4 termini, showed non
classical plateauing of the dose response curve (Fig. 4c)
analogous to that observed for Dp44mT against MCF-7
(Fig. 4a). Disubstitution of the N4-termini of pyridyl bis-
thiosemicarbazones resulted in compounds with moder-
ate potency against MelRm and lower IC50 values were
obtained using compounds bearing cyclic amines with a
polar terminus (morpholine and N-methyl piperazine in Bis
1i and Bis 1j, respectively). Replacing the central pyridyl
core in Bis (1a–d) with the phenyl core in Bis (1k–m) or
with phenolic core in Bis (2a–c) resulted in inactive com-
pounds. In general, however, the cytotoxic activity of the
Bis- series was inferior to the Dp44mT reference.
Unsubstituted Bis(1a) showed slight improvement upon
Cu supplementation compared to the unsupplemented
chelator. Disubstitution with bulky and hydrophobic sub-
stituents was shown to affect the improvement level as
exemplified by Bis (1h) where the cytotoxicity levelled off
at around 50 % of cells killed from 1 µM drug concentra-
tion. Upon increasing the size of the dialkyl substituents to
1 3