Self-activating nuclease and anticancer activities of copper(ii) complexes with aryl-modified 2,6-di(thiazol-2-yl)pyridine

Article abstract of DOI:10.1039/c3dt50395j

Three mononuclear copper complexes [Cu(PDTP)Cl₂] (PDTP = 4-phenyl-2,6-di(thiazole-2-yl)pyridine, CuPDTP), [Cu(ADTP)Cl₂] (ADTP = 4-(anthracen-9-yl)-2,6-di(thiazole-2-yl)pyridine, CuADTP) and [Cu(BFDTP)Cl₂] (BFDTP = 4-(benzofuran-2-yl)-2,6-di(thiazole-2-yl) pyridine, CuBFDTP) were synthesized and characterized. The X-ray single crystallography results indicated that the Cu(ii) ions showed slightly distorted square pyramid coordination environments, and the ligands deviated from ideal planarity in all three compounds. Based on the DNA binding studies, it was demonstrated that these three complexes exhibited weak DNA binding strengths, which were most likely groove binding modes. CuPDTP, CuADTP and CuBFDTP induced efficient DNA cleavage in the dark without the addition of external catalysts (oxidant or reductant). In contrast, in the presence of reducing or oxidizing agents, the nuclease activities increased more than 10-fold. Mechanistic investigations revealed the participation of reactive oxygen species, which can be trapped by ROS radical scavengers and ROS sensors. In the same experimental conditions, the free ligands and CuCl₂ did not display any DNA cleaving activity. This result indicates that the complexes, rather than their components, play a significant role in the nuclease reaction process and that DNA cleavage may be initiated in an oxidative pattern. The proposed mechanism was attributed to the in situ activation of molecular oxygen by the oxidation of the copper complexes. In the MTT cytotoxicity studies, the three Cu(ii) complexes exhibited an antitumor activity against the HeLa, BEL-7402 and HepG2 tumor cell lines. The HeLa cells treated with Cu(ii) complexes demonstrated marked changes in their nuclear morphology, which were detected by Hoechst 33258 nuclear staining and acridine orange/ethidium bromide (AO/EB) staining assays. Nuclear chromatin cleavage also was observed from alkaline single-cell gel electrophoresis (comet assay).

Full text of DOI:10.1039/c3dt50395j

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Self-activating nuclease and anticancer activities
of copper(II) complexes with aryl-modified
2,6-di(thiazol-2-yl)pyridine†
Cite this: Dalton Trans., 2013, 42, 11576
Lüying Li,a Kejie Du,a Yi Wang,a Haina Jia,a Xiaojuan Hou,a,b Hui Chao*a and
Liangnian Ji*a
Three mononuclear copper complexes [Cu(PDTP)Cl2] (PDTP = 4-phenyl-2,6-di(thiazole-2-yl)pyridine,
CuPDTP), [Cu(ADTP)Cl2] (ADTP = 4-(anthracen-9-yl)-2,6-di(thiazole-2-yl)pyridine, CuADTP) and [Cu-
(BFDTP)Cl2] (BFDTP = 4-(benzofuran-2-yl)-2,6-di(thiazole-2-yl)pyridine, CuBFDTP) were synthesized and
characterized. The X-ray single crystallography results indicated that the Cu(II) ions showed slightly dis-
torted square pyramid coordination environments, and the ligands deviated from ideal planarity in all
three compounds. Based on the DNA binding studies, it was demonstrated that these three complexes
exhibited weak DNA binding strengths, which were most likely groove binding modes. CuPDTP, CuADTP
and CuBFDTP induced efficient DNA cleavage in the dark without the addition of external catalysts
(oxidant or reductant). In contrast, in the presence of reducing or oxidizing agents, the nuclease activities
increased more than 10-fold. Mechanistic investigations revealed the participation of reactive oxygen
species, which can be trapped by ROS radical scavengers and ROS sensors. In the same experimental con-
ditions, the free ligands and CuCl2 did not display any DNA cleaving activity. This result indicates that the
complexes, rather than their components, play a significant role in the nuclease reaction process and that
DNA cleavage may be initiated in an oxidative pattern. The proposed mechanism was attributed to the
in situ activation of molecular oxygen by the oxidation of the copper complexes. In the MTT cytotoxicity
studies, the three Cu(II) complexes exhibited an antitumor activity against the HeLa, BEL-7402 and HepG2
tumor cell lines. The HeLa cells treated with Cu(II) complexes demonstrated marked changes in their
nuclear morphology, which were detected by Hoechst 33258 nuclear staining and acridine orange/ethi-
dium bromide (AO/EB) staining assays. Nuclear chromatin cleavage also was observed from alkaline
single-cell gel electrophoresis (comet assay).
Received 9th February 2013,
Accepted 31st May 2013
DOI: 10.1039/c3dt50395j
www.rsc.org/dalton
them potential value in the treatment of cancer and for appli-
cations in biotechnology. Generally, metal-mediated DNA clea-
Introduction
2+
Since the bis-phenanthroline copper complex Cu(phen)2 was vage occurs through one of three different pathways, including
first reported as an effective DNA cleaving agent,1 copper- photocleavage,4,5 DNA hydrolysis6 and oxidative cleavage.7
based chemical nucleases have attracted significant attention. Photoactivated DNA cleavage is limited by the effectiveness of
A number of artificial nucleases containing redox-active Cu(II) the light penetration and the dark toxicity in photodynamic
as their active sites have been synthesized.2,3 Copper(II) com- therapy, and metal hydrolases tend to show low cleaving
plexes possess diverse structures and have a reactivity that efficiency.8,9 Thus, synthetic chemical nucleases exhibiting oxi-
enables them to effectively cleave DNA and RNA, which gives dative activity are the main focus of the majority of studies on
artificial DNAses.10 The 1,10-phenanthroline copper complex
[Cu(phen)2]2+ was reported to oxidatively cleave DNA in the
presence of H2O2 by forming “oxo”-Cu(III) active species.11 Guo
aMOE Key Laboratory of Bioinorganic and Synthetic Chemistry, State Key Laboratory
et al. reported a very active oxidative system of multinuclear
copper complexes with the ability to cleave DNA using
reductants.12–14 However, as pharmaceutical agents, these
efficient oxidative cleavers require the addition of external
reagents (e.g., ascorbic acid or hydrogen peroxide) to initiate
the desired reactions, which limits their applications
of Optoelectronic Materials and Technologies, School of Chemistry and Chemical
Engineering, Sun Yat-Sen University, Guangzhou 510275, P. R. China.
E-mail: ceschh@mail.sysu.edu.cn, cesjln@mail.sysu.edu.cn
bHuaihua Medical College, Huaihua 418000, P. R. China
†Electronic supplementary information (ESI) available: Fig. S1–S10. CCDC
numbers 869871, 869872 and 869873. For ESI and crystallographic data in CIF
or other electronic format see DOI: 10.1039/c3dt50395j
11576 | Dalton Trans., 2013, 42, 11576–11588
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
in vivo.9,15 As a result, the development of a self-activating aldehydes and 2-acetylthiazole was stirred for several hours at
system requiring no further activation to induce the double- room temperature. This one-pot method is facile and readily
strand breakage of DNA is desired.
applied to the synthesis of pyridine derivatives containing
Recently, copper complexes possessing inherently diverse sterically hindered aryl substitutions. The Cu(II) complexes
structural and redox properties have become attractive agents were prepared in high yields by mixing stoichiometric
for DNA cleavage due to their potential self-activating nuclease amounts of CuCl2·2H2O and the tridentate ligands PDTP,
activity.16–24 Some copper complexes containing hydroxyl-rich ADTP or BFDTP in a methanol solution. The elemental ana-
ligands or phthalate ligands have the ability to spontaneously lyses of all the complexes are in agreement with the corres-
generate active oxygen by reactions involving concomitant ponding molecular formula. The compositions of the
copper and ligand redox cycles, which initiate DNA clea- complexes have been further confirmed by their ES-MS
vage.16,17 Reedijk et al. found that a copper–Hpyramol complex spectra. In the mass spectra, the molecular ion [M − Cl]+ was
cleaved DNA in the absence of reductants by attacking mul- observed, and the measured molecular weights were consistent
tiple nucleotide sites.18,19 In this case, it was suggested that with the expected values.
DNA cleavage was initiated by the dehydrogenation of the
CuPDTP, CuADTP and CuBFDTP were characterized by
ligand upon metal center coordination with the participation X-ray single crystallography. The Cu(II) center was found to be
of non-diffusible radicals. Although the detailed cleavage coordinated by one N atom (N2) from pyridine and two
mechanism remains unclear, all of the reactions appear to be N atoms (N1, N3) from the dithiazole of the ligands. An
activated by ligand-based metal ions. In the field of clinical ORTEP perspective view is shown in Fig. 1. The coordination
anti-tumor treatment, bleomycin (BLM) was observed to geometry of Cu in the complex is described as a distorted
trigger both single and double-strand DNA cleavage.25 In the square pyramid according to the Addison–Reedijk geometric
presence of Fe(II) and O2, BLM induces DNA damage without
any external reductant by the formation of Fe–OOH, which is
an activated intermediate.26 In the structure, bisthiazole,
which is a unit of BLM, is a DNA binding domain that signifi-
cantly enhances the DNA binding affinity and efficiently pro-
motes the cleavage of two strands of DNA.27,28 The title Cu(II)
complexes, [Cu(PDTP)Cl2] (PDTP = 4-phenyl-2,6-di(thiazole-2-
yl)pyridine, CuPDTP), [Cu(ADTP)Cl2] (ADTP = 4-(anthracen-9-
yl)-2,6-di(thiazole-2-yl)pyridine, CuADTP) and [Cu(BFDTP)Cl2]
(BFDTP
=
4-(benzofuran-2-yl)-2,6-di(thiazole-2-yl)pyridine,
CuBFDTP), were inspired by BLM and are similar to other thia-
zole-containing synthetic DNA nucleases.29,30 Coordinated to
the copper, the thiazole ligands were designed and synthesized
to be applicable in the study of self-activating nuclease activity.
Considering that the majority of the current copper complexes
must be initiated by an excess of exogenous agents to cleave
DNA, the present work is the first example of thiazole ligand-
containing copper complexes that exhibit self-activating oxi-
dative cleavage.
Results and discussion
Synthesis and characterization
The methods used to prepare the ligands and complexes are
shown in Scheme 1. The ligands were synthesized according to
the literature31 with small modifications. The solution of aryl
Fig. 1 ORTEP representations of the complex cations, CuPDTP (top), CuADTP
(middle) and CuBFDTP (bottom) showing 30% probability ellipsoid plots.
Scheme 1 The synthesis of the ligands and the Cu(II) complexes.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 11576–11588 | 11577

View Article Online
Paper
Dalton Transactions
criterion32 in which the parameter τ, calculated using the
observed L–M–L′ basal angles, is an index of the degree of tri-
gonality of the structure. Within the structural continuum
between the trigonal bipyramidal and square pyramidal limit-
ing geometries, τ is equal to zero for a perfect square pyrami-
dal or tetragonal geometry; τ becomes unity for a perfect
trigonal bipyramidal arrangement. Based on this calculation,
τ = 0.015, 0.13 and 0.094 for CuPDTP, CuADTP and CuBFDTP,
respectively. CuPDTP crystallizes in the monoclinic space
group (P21/n). Each asymmetric unit contains two Cu(II) ions,
two PDTP ligands and four chloride ions. CuADTP and
CuBFDTP crystallize in the orthorhombic and triclinic space
groups, respectively. The coordination geometries of CuADTP
and CuBFDTP are similar to that of CuPDTP. The equatorial
positions of the distorted square pyramids are occupied by one
chloride atom and a ADTP/BFDTP ligand, while the axial posi-
tion is occupied by another chloride atom. The Cu–N and Cu–
Cl bond distances are in the normal range for this coordi-
nation geometry. Due to the constrained bite of the tridentate
Fig. 2 The emission profiles of 100 μM CT-DNA-bound EB in a 10 mM phos-
■
phate buffer (pH 7.2) at room temperature upon the addition of CuPDTP ( ),
▲
▼
CuADTP ( ), CuBFDTP ( ). [EB] = 10 μM, [Cu] = 0–24 μM.
in the DNA melting temperature (Tm). The melting curves of
CT-DNA in the absence and presence of CuPDTP, CuADTP and
CuBFDTP are presented in Fig. S1 (ESI†). The thermal dena-
turation profiles of native DNA in the presence of the Cu(II)
complexes, obtained by plotting the UV absorbance at 258 nm
as a function of temperature, show that the DNA melting temp-
erature decreased by 1.1, 2.3, and 0.8 °C, respectively. This
indicates that CuPDTP, CuADTP and CuBFDTP may bind to
the DNA in groove sites and destabilize the DNA helix. This is
observation is different from that seen when DNA-intercalative
metal complexes bind to DNA, as DNA-intercalative metal com-
plexes stabilize the double helix.36,40,41 The small decreases in
the DNA denaturation temperature (ΔTm) in the presence of
CuPDTP, CuADTP and CuBFDTP further verify the conclusion
obtained in the EB displacement assay.
Circular dichroic spectral studies provide information
regarding the DNA conformational changes as well as the
interaction strength between the metal complexes and DNA.
The CD spectrum of CT-DNA exhibits a positive band at 275 nm
due to base stacking and a negative band at 245 nm due to the
right-handed helicity of the B-DNA form, which is quite sensi-
tive to the mode of DNA interaction with small molecules. The
simple groove binding and electrostatic interaction of the com-
plexes with DNA show less or no perturbation of the base stack-
ing and helicity bands, while the intercalator enhances the
intensities of both bands.42–44 From the CD spectra (Fig. S2,
ESI†), it was observed that the DNA CD bands were weakly modi-
fied by increasing amounts of the Cu(II) complexes, confirming
that the double helical structure is only slightly perturbed by
the interaction with the Cu(II) complexes. This observation was
consistent with the above spectroscopic studies. Given that the
nuclease activity was not affected by high salt concentrations
(Fig. S3, ESI†), we speculate that the Cu(II) complexes do not
bind the DNA through electrostatic interactions. Similar to the
oxidative cleaving agent [Cu(phen)2]2+,45 the Cu(II) complexes
likely interact with the DNA through groove binding.
ligands, the Cu–N bond length to the central ring [Cu–N2
=
1.977(3)–1.995(4) Å] is shorter than those to the terminal rings
[2.035(3)–2.068(4) Å], which is typical for the coordination of
conjugated terimine systems.33–35
The coordination angles reflect the strong distortion caused
by the bulk hindrance of the ligand. The dihedral angles of
dithiazole and pyridine are 6° and 1°, indicating that the DTP
ligand, which contains two thiazoles and a pyridine, shows
good planarity. Thus, to discuss the planarity of the ligand, the
focus should be directed to their different substitutional
groups. In these three compounds, the dihedral angles bet-
ween pyridine and phenyl, anthracen-9-yl or benzofuran-2-yl
are 10°, 78° and 5°, respectively. These results suggest that
the overall planarity of the ligands decreases as follows:
CuBFDTP > CuPDTP ≫ CuADTP.
DNA binding
The three Cu(II) complexes were non-emissive in aqueous solu-
tion and in the presence of calf thymus DNA. Hence, the
binding of the Cu(II) complexes with DNA cannot be directly pre-
sented in the emission spectra. It has been previously reported
that the enhanced fluorescence can be quenched, at least par-
tially, by the addition of a second molecule.36–38 Ethidium
bromide (EB) emits intense fluorescence in the presence of
DNA, and the extent of the fluorescence quenching of EB bound
to DNA is used to determine the extent of the binding between
the second molecule and the DNA.37 The fluorescence quench-
ing of EB bound to DNA by the Cu(II) complexes is shown in
Fig. 2. The ability of the present complexes to quench the EB
emission decreased in the order of CuBFDTP (37.1%) > CuPDTP
(29.8%) > CuADTP (10.8%). Compared with DNA intercalative
Cu(II) complexes,38 these slight changes in fluorescence indi-
cated that the Cu(II) complexes cannot displace the EB molecules
from the DNA base pairs, suggesting the primarily electrostatic
or groove DNA binding39 nature of the Cu(II) complexes.
Chemical nuclease activity
According to previous reports,40,41 the intercalation of the The cleavage of plasmid pBR322 DNA induced by CuPDTP,
metallointercalators generally results in prominent increases CuADTP and CuBFDTP was performed in the absence of
11578 | Dalton Trans., 2013, 42, 11576–11588
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
distamycin (100 μM), while the methyl green (100 μM) addition
has no apparent effect on the DNA cleavage of these com-
plexes. The results indicate the minor groove binding propen-
sity of CuPDTP, CuADTP and CuBFDTP (Fig. S5, ESI†). The
DNA cleavage experiments were performed at different pH
values (pH = 5.5–9.5) and it was found that the pH of the solu-
tion exhibited little effect on the DNA cleavage activity of the
Cu(II) complexes (Fig. S6, ESI†). It was also noticed that in the
presence of the thiazole ligands, CuPDTP, CuADTP and
CuBFDTP can exhibit self-activating oxidative cleavage;
however, when replacing the thiazole with pyridine, such a be-
havior was not observed in the Cu(II) complexes (Fig. S7, ESI†).
Similar to DNA enzymes,46 DNA cleavage induced by
CuPDTP, CuADTP and CuBFDTP is dependent on temperature
and time. At 37 °C, all of the Cu(II) complexes demonstrated
significantly higher nuclease activities than at lower tempera-
tures (Fig. S8, ESI†). This result follows the trend that cleavage
reactions are more active at higher temperatures. The cleavage
efficiency was also enhanced with time (Fig. 5); the amounts of
plasmid DNA gradually decreased due to the degradation of
the SC DNA (form I) to the NC form (form II) and L DNA
(form III). The double-strand DNA cleavage was completed
within 2 h and the plasmid DNA ultimately disappeared. The
kinetic characterization of the DNA cleavage by CuPDTP,
CuADTP and CuBFDTP showed that the time-dependent disap-
pearance of form I DNA and the appearance of form II and
form III DNA followed pseudo-first-order kinetic profiles
(Fig. 5b).47 However, these profiles cannot be fitted with a
single exponential function (Fig. S9, ESI†). This result indicates
that the cleavage did not occur through a simple hydrolytic
pathway. Furthermore, the DNA cleavage products failed to be
religated by the enzyme T4 DNA ligase, giving additional evi-
dence of nonhydrolysis. Additionally, all of the experiments
were performed in dark, so the action of photo-induced clea-
vage can be excluded. All of the observations support the
Fig. 3 The gel electrophoresis of pBR322 DNA cleavage catalyzed by CuPDTP,
CuADTP and CuBFDTP in phosphate buffer at 37 °C for 1.5 h. (a) Lane 1: DNA
control (40 μM), lanes 2–4: DNA + CuPDTP (25, 50, 100 μM), lanes 5–7: DNA +
CuADTP (25, 50, 100 μM), lanes 8–10: DNA + CuBFDTP (25, 50, 100 μM).
(b) Lane 1: DNA control (40 μM); lane 2: DNA + H2O2 (100 μM); lanes 3 and 4:
DNA + CuPDTP (5, 10 μM) + H2O2 (100 μM); lanes 5 and 6: DNA + CuADTP (5,
10 μM) + H2O2 (100 μM); lanes 7 and 8: DNA + CuBFDTP (5, 10 μM) + H2O2
(100 μM).
external cofactor agents (either oxidant or reductant) in the
dark. The Cu(II) complexes were found to promote the cleavage
of plasmid pBR322 DNA from the supercoiled form (I) to the
nicked form (II) and linear form (III) (Fig. 3a). No DNA clea-
vage was observed for the control in which the metal complex
was absent (lane 1). When the concentrations of the three
Cu(II) complexes were increased (lanes 2–10), the amount of
form I of pBR322 DNA diminished gradually, whereas the
amounts of form II and III increased. The Cu(II) complexes are
able to induce significant cleavage of the plasmid DNA at the
concentration of 25 μM. At the concentration of 100 μM,
CuPDTP, CuADTP and CuBFDTP almost promote the complete
conversion of DNA from form I to forms II and III. Similar
cases were also observed in the light or in an argon atmos-
phere (Fig. S4, ESI†). In the presence of an oxidant (100 μM
H2O2) under aerobic conditions (Fig. 3b), only 5 μM of
CuPDTP, CuADTP and CuBFDTP can degrade 60–80% of DNA.
In the presence of a reducing agent (250 μM ascorbic acid),
10 μM of the Cu(II) complexes are sufficient to induce 60–70%
of the plasmid DNA cleavage (Fig. 4). This observation indi-
cates that the chemical nuclease activities of the Cu(II) com-
plexes are significantly promoted with the addition of either
oxidants or reductants.
The DNA groove binding preference of the complexes was
studied using the DNA major groove binder methyl green and
the DNA minor groove binder distamycin. A significant inhi-
bition in the chemical nuclease activity of complexes CuPDTP,
CuADTP and CuBFDTP was observed in the presence of
Fig. 4 The agarose gel electrophoresis of the Cu(II) complexes showing the oxi-
dative cleavage of DNA in the presence of ascorbic acid (HA, 250 μM). Lane 1:
DNA control (40 μM); lane 2: DNA + HA; lane 3: DNA + CuCl2 (500 μM) + HA;
lanes 4–7: DNA + CuPDTP (5, 10, 15, 20 μM) + HA; lanes 8–11: DNA + CuADTP
(5, 10, 15, 20 μM) + HA; lanes 12–15: DNA + CuBFDTP (5, 10, 15, 20 μM) + HA.
Fig. 5 (a) Time-dependent experiments of pBR322 DNA (40 μM bp) cleavage
catalyzed by CuBFDTP (20 μM) over a period of 120 min at 37 °C in phosphate
buffer (pH 7.0). Lane 1: DNA control (40 μM), lanes 2–8: DNA + 20 μM
CuBFDTP (incubating time 5, 20, 40, 60, 80, 100, 120 min). (b) A plot of the
changes of form I, form II and form III with time.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 11576–11588 | 11579

View Article Online
Paper
Dalton Transactions
hypothesis that CuPDTP, CuADTP and CuBFDTP most likely
cleaved the plasmid DNA by a self-activating oxidative mechan-
ism, which is similar to the manner in which bleomycin
cooperates with a redox metal (FeII) to trigger DNA cleavage
without exogenous agents.
Fig. 7 The agarose gel electrophoresis of CuBFDTP (10 μM) cleaving DNA
(40 μM) in the presence of various radical scavengers under redox conditions for
1.5 h at 37 °C. Lane 1: DNA control, lanes 2 and 11: no inhibitor ([CuBFDTP] =
10 μM); lanes 3 and 12: CuBFDTP + NaN3 (100 mM); lanes 4 and 13: CuBFDTP
+ histidine (100 mM); lanes 5 and 14: CuBFDTP + DMSO (100 mM); lanes 6 and
A mechanistic investigation of the DNA cleavage reaction
To explore the cleavage mechanism, the copper-mediated DNA
cleavage was investigated with several different potential
radical scavengers to identify the intermediate reactive oxygen
species (ROS) in the absence of external cofactors (Fig. 6 and
Fig. S10, ESI†). DMSO and mannitol were used as hydroxyl
radical (HO˙) scavengers; sodium azide and L-histidine were
used as singlet oxygen (1O2) scavengers; and superoxide dismu-
tase (SOD) was used as an O2˙− radical scavenger. The DNA
breaks mediated by CuPDTP, CuADTP and CuBFDTP were not
affected by the presence of the hydroxyl radical inhibitors
DMSO and mannitol, suggesting that diffusible hydroxyl rad-
icals were not involved in the DNA damage mechanism. The
single oxygen scavengers NaN3 and histidine partially
decreased the cleavage activity, each to a different extent.
Sodium azide was significantly more effective than histidine,
which is likely due to the NaN3 chelation of the metal ion,
which indicates that singlet oxygen may participate in the
nuclease reaction. However, in the dark, it was observed that
the singlet oxygen scavengers NaN3 and histidine had no effect
on the cleavage activity (Fig. S11, ESI†). The metal chelator
EDTA also completely inhibited cleavage activity, indicating
that this DNA cleavage mechanism is a metal-assisted reaction.
DNA cleaving activity is completely repressed by catalase, an
enzyme that disproportionates H2O2 to yield H2O + O2,48
implying that H2O2 as a reactive oxygen species is responsible
for DNA cleavage. In contrast, SOD, an enzyme capable of con-
verting the superoxide radical to H2O2, promotes DNA
damage. This is additional proof that H2O2 participates in the
cleavage. The cleavage is not dependent on the carbon scaven-
ger 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO), ruling out
the existence of a carbon radical in the reaction. The ROS
inhibitor studies have indicated that DNA cleavage can be
15: CuBFDTP
+ mannitol (50 mM); lanes 7 and 16: CuBFDTP + SOD
(500 U mL−1); lanes 8 and 17: CuBFDTP + catalase (1000 U mL−1); lanes 9 and
18: CuBFDTP + EDTA (10 mM); lanes 10 and 19: CuBFDTP + TEMPO (20 mM).
most likely that CuPDTP, CuADTP and CuBFDTP cleaved the
DNA through a self-activating oxidative pathway.
To further test the involvement of H2O2 in the cleavage reac-
tion, parallel cleavage experiments were performed. Under
redox reaction conditions, CuPDTP, CuADTP and CuBFDTP
were found to form the same ROS species involved in the reac-
tion, and H2O2 was the crucial active intermediate involved in
the DNA cleavage (Fig. 7 and Fig. S12, ESI†). The resulting
nuclease activity and mechanism under redox conditions
strongly indicate a “self-activating” mechanism, which is most
likely initiated by the copper(II) complex through an oxidative
pathway.49–51 However, under the present conditions, because
the high valent copper species were unstable, evidence for
CuBFDTP+ was difficult to observe. Further studies are cur-
rently underway to clarify the cleavage mechanism.
The detection of reactive oxygen species
The ROS radical also was detected by various radical indi-
cators. The possibility of hydroxyl radical formation in the
presence of the copper complexes was evaluated using rhoda-
mine B dye. This reaction was performed in a 10 mM phos-
phate buffer at pH 7.2 under aerobic conditions. Based on our
observation, there was no obvious decrease in the absorbance
of the dye at 552 nm in the presence of the copper complexes
(Fig. S13a, ESI†). The electronic spectra of rhodamine B was
slightly increased after incubation with CuADTP due to the
assembly of a copper complex in the rhodamine B solution.
Under the same conditions, a parallel experiment, using FeCl2
and H2O2 (the Fenton conditions),16 displayed a gradual decrease
with the reaction time, verifying that the hydroxyl radical for-
mation resulted in dye degradation. The diffusing hydroxyl rad-
icals were not trapped during the detection of rhodamine B
degradation, suggesting that the OH˙ radical was not included
in the reaction. Unlike the H2O2 and O2 activating copper nucle-
ase, which produces OH˙ by an internal electron transfer to
initiate DNA cleavage,52 CuPDTP, CuADTP and CuBFDTP cleaved
1
initiated by reactive oxygen species O2 (only in the light) and
H2O2 but not diffusing OH˙ radicals. Considering that these
types of active species mostly occur in oxidative cleavage, it is
Fig. 6 The agarose gel electrophoresis of CuBFDTP (100 μM) cleaving DNA
(40 μM) in the presence of various radical scavengers under “self-activating” DNA without the participation of OH˙ radicals.
When using 1,3-diphenylisobenzofuran (DPBF),53 the emis-
conditions for 1.5 h at 37 °C. (a) Lane 1: DNA control; lane 2: no inhibitor
([CuBFDTP] = 100 μM); lane 3: CuBFDTP + NaN3 (100 mM); lane 4: CuBFDTP +
histidine (100 mM); lane 5: CuBFDTP + DMSO (100 mM); lane 6: CuBFDTP +
mannitol (50 mM); lane 7: CuBFDTP + SOD (500 U mL−1); lane 8: CuBFDTP +
catalase (1000 U mL−1); lane 9: CuBFDTP + EDTA (10 mM); lane 10: CuBFDTP +
TEMPO (20 mM).
1
sion is quenched in the presence of O2. The changes in the
1
dye emission can be used to evaluate the formation of O2. In
a typical experiment, DPBF was incubated with the Cu(II) com-
plexes while monitoring the fluorescence at 479 nm. The DPBF
11580 | Dalton Trans., 2013, 42, 11576–11588
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
fluorescence consumption vs. time profile is shown in activity of the Cu(II) complex may be attributed to its stronger
Fig. S13b (ESI†). As a positive control, the emission of DPBF ability to cleave DNA. The findings of the in vitro cytotoxic
with [Ru(bpy)3]2+ under aerobic conditions resulted in a activities further confirm that the DNA cleavage induced by
decrease in the fluorescence intensity of DPBF. This indicates Cu(II) complexes leads to cell death.
that the dye was consumed in the presence of the singlet
oxygen radicals under aerobic conditions. Similar to [Ru-
Nuclear morphology
(bpy)3]2+, CuPDTP, CuADTP and CuBFDTP resulted in
Cell death can be divided into two types, necrosis (accidental
decreased fluorescence of DPBF with time, indicating the for-
cell death) and apoptosis (programmed cell death).56 Necrotic
cells undergo cell lysis and lose their membrane integrity,
inducing severe inflammation.57 Apoptotic cells, however, are
transformed into small membrane-bound vesicles (apoptotic
bodies), which are engulfed in vivo by macrophages, and no
inflammatory response is found.58 The harmless removal of
cells (e.g., cancer cells) is one consideration in chemother-
apy.59 Therefore, the induction of apoptosis is one of the con-
siderations in the development of anticancer drugs. To
identify the possible involvement of apoptosis, HeLa cells
treated with the Cu(II) complexes were stained with Hoechst
33258.60 As shown in Fig. 8 for the HeLa cells treated with
CuPDTP, CuADTP and CuBFDTP, a significant increase in the
apoptotic nuclear morphology, such as extensive chromatin
aggregation or nuclear condensation, was observed in the
treated cells; in contrast, the nuclear contours of normal HeLa
cells stained with Hoechst 33258 stained evenly. The number
of abnormal cells generated at 24 h varied according to
CuADTP > CuBFDTP > CuPDTP. The higher apoptosis-indu-
cing activity of CuADTP is consistent with its stronger double-
strand DNA cleavage ability.
mation of 1O2 in the reaction system.
To test whether the H2O2 intermediate was formed by the
copper complexes in aqueous solution, 2′,7′-dichlorofluores-
cein (DCF) was examined for the formation of H2O2.54 The
nonfluorescent DCFH can be rapidly oxidized to emit the
strongly fluorescent DCF in the presence of H2O2. In our experi-
ment, when incubated with CuPDTP, CuADTP and CuBFDTP,
DCFH consistently generated significant fluorescence
(Fig. S13c, ESI†). CuADTP showed the highest H2O2 quantum
yield and exhibiting the strongest amount of emission. In the
negative contrasting experiment, the DCFH dye was pre-incu-
bated with catalase (1000 U mL−1) to remove H2O2. To our sur-
prise, the fluorescence intensity did not change significantly.
The observations indicated that the oxidation of DCFH was
directly inhibited by the addition of catalase. This result pro-
vided additional proof that H2O2 is an active intermediate that
is generated in the reaction system and that it is an active
species with respect to catalyzed DNA cleavage.
Cytotoxicity
The cytotoxicities of CuPDTP, CuADTP and CuBFDTP were
measured using a standard MTT assay55 against three human-
derived tumor cell lines with the clinical anti-tumor agent cis-
platin as a control. Under similar conditions, the target com-
plexes were tested on HeLa human cervix carcinoma cell lines,
BEL-7402 and the HepG2 hepatoma carcinoma cell line. The
IC50 values for the cell lines incubated with the complexes are
summarized in Table 1. CuPDTP, CuADTP and CuBFDTP dis-
played medium toxicities against all three tumor lines, and in
all cases, the activities of the complexes were higher than
those of the corresponding free ligands. The cytotoxic activity
studies in vitro indicate that the Cu(II) complexes have excellent
activities on the BEL-7402 and HepG2 cells but showed a sig-
It is well-established that acridine orange (AO) is able to
penetrate the membranes of viable cells, while EB is excluded.
Therefore, the emission of green fluorescence from AO is a
convenient marker for viable cells, and the emission of red
fluorescence from EB indicates a loss of cell viability.61 Micro-
photographs of the AO/EB-stained HeLa cells, which were pre-
treated for 24 h with the three Cu(II) complexes, showed that
CuPDTP, CuADTP and CuBFDTP induced the apoptosis of the
HeLa cells (Fig. 9). Some of the apoptotic cells were also
detected in secondary necrosis. As shown, the three Cu(II)
nificantly lower activity than cisplatin for the HeLa cells (IC50
=
0.014 mM, which was calculated by the same experimental
method). Hence, the above results indicate that the Cu(II) com-
plexes are active on specific cancer cells. The better cytotoxic
Table 1 The IC50 values (mM) for the Cu(II) complexes and cisplatin against cer-
vical cancer cells (HeLa), hepatoma carcinoma cells (BEL-7402) and hepatocellu-
lar carcinoma cells (HepG2) over a period of 24 h
IC50 (mM)
Cisplatin
CuPDTP
CuADTP
CuBFDTP
Fig. 8 HeLa cells were stained with Hoechst 33258 and observed under a flu-
orescence microscope: (a) untreated cells, (b) treated with the Cu(II) complexes
(25 μM), (c) treated with the Cu(II) complexes (50 μM).
HeLa
BEL-7402
HepG2
0.014
0.020
0.026
0.059
0.028
0.036
0.036
0.021
0.027
0.043
0.026
0.033
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 11576–11588 | 11581

View Article Online
Paper
Dalton Transactions
Fig. 10 Drug-induced double-strand DNA breaks in HeLa cells. Cells were
untreated (a) or treated with (b) CuADTP (15 μM), (c) CuADTP (30 μM),
(d) CuADTP (60 μM) for 2 h. The detection of the DNA breaks was done using
alkaline single-cell gel electrophoresis.
Fig. 9 HeLa cells were stained by AO/EB and observed under a fluorescence
microscope: (a) untreated cells, (b) treated with the Cu(II) complexes (25 μM),
(c) treated with the Cu(II) complexes (50 μM); incubated at 37 °C and 5% CO2
for 24 h. The arrows point to the cells representing certain cell viability status: L
points to the live cells; A points to the apoptotic cells; and N points to the necro-
tic cells.
appearance of “broom-like” tails, indicating the presence of
severe DNA damage (Fig. 10d). The outcome of the comet
assay showed that the DNA of a single cell underwent degra-
dation as a consequence of direct DNA damage or rapid apop-
tosis. The high level of DNA damage induced by the Cu(II)
complexes reinforces the above results obtained using the
MTT assay and the nuclear staining assay.
complexes (15 μM) can induce apoptosis, which is observed as
the condensation of the HeLa cell nuclei after 24 h. Upon the
treatment of the cells with significant amounts of the Cu(II)
complexes (60 μM), the cells exhibited condensed orange
nuclei (Fig. 9), which is a hallmark of late apoptotic cells;
however, the necrotic cells displayed structurally intact nuclei
with even orange staining. The orange nucleus was due to the
loss of membrane integrity in the late apoptotic and necrotic
cells, allowing EB to stain the nucleus. The results further
support the idea that the CuPDTP, CuADTP and CuBFDTP
complexes caused significant HeLa cell apoptosis. The mor-
phological changes observed with CuPDTP, CuADTP and
CuBFDTP suggest that the cells are committed to death in
such a way that both apoptotic and necrotic cells increase in
number in a concentration-dependent manner. The ability of
the Cu(II) complexes to induce apoptotic cell death is consist-
ent with the results from the Hoechst 33258 staining.
The intracellular reactive oxygen species study
To elucidate the relationship between cytotoxicity and the
generation of reactive oxygen species (ROS), CuPDTP, CuADTP
and CuBFDTP were exposed to HeLa human cervix carcinoma
cancer cells, which had been pre-treated with the intracellular
ROS indicator 2′,7′-dichlorofluorescein diacetate (DCFH-DA).
In the presence of endogenously generated ROS, DCFH-DA is
oxidized to release the fluorophore 2′,7′-dichlorofluorescein
(DCF).54 The results show an increase in the cellular ROS
levels after treatment for 12 h with the complexes and are
expressed in comparison to the ROS levels of the unexposed
controls (Fig. 11). The amount of the intracellular ROS species
Comet assay
Our present experiment has clearly shown that these Cu(II)
complexes induced DNA strand cleavage by generating ROS
such as 1O2 and H2O2. Additionally, this mechanism is
believed to be the molecular basis of their cytotoxic activity.
Alkaline single-cell gel electrophoresis (comet assay) can give a
perspective of the changes that occur in the chromatin organ-
ization within a single cell, which is considered a more accu-
rate way of detecting early nuclear changes in
a cell
population.62 For example, as shown in Fig. 10, no broom-like
tails were observed in the untreated cells (Fig. 10a). Treatment
with 15 μM Cu(II) complex led to the appearance of obscure
Fig. 11 The DCFH-DA detection of the ROS in HeLa cells treated with the Cu(II)
“halos” around the nuclei of the HeLa cells (Fig. 10b), illustrat- complexes CuPDTP (1), CuADTP (2) and CuBFDTP (3). The HeLa cells were
incubated with the Cu(II) complexes for 12 h, which was followed by incubation
ing that DNA damage occurred in these cells; additionally,
when 30 μM Cu(II) complex was introduced, the nuclei began
to shrink (Fig. 10c). At Cu(II) complex concentrations of up to
with 10 μM DCFH-DA for 30 min. (a) The level of intracellular ROS was detected
by a DCHF-DA assay using flow cytometric analysis. The percentage of ROS-over-
expressing cells under different treatments was detected. (b) HeLa cells showing
60 μM, a sudden change could be observed: the nuclei almost
an increase in the ROS level when treated with the Cu(II) complexes viewed
disappeared, and this was further accompanied by the using a fluorescence microscope.
11582 | Dalton Trans., 2013, 42, 11576–11588
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
generated by the Cu(II) complexes follows the order of CuADTP system (Finnigan MAT, USA), and the quoted m/z values given
> CuBFDTP > CuPDTP, which is the same as that observed for in this work are for the major peaks in the isotope distri-
the IC50 values. This reveals that CuADTP, which exhibits the bution. UV-visible (UV-vis) spectra were recorded using a
strongest double-strand DNA cleavage ability, activates the Perkin-Elmer Lambda 850 spectrophotometer, and the fluo-
ROS-generating machinery and generates the highest amount rescence spectra were recorded using a Perkin-Elmer LS55
of ROS when treated with cancer cells, leading to apoptosis. Fluorescence spectrometer. CD spectra were measured using a
Thus, it is clear that the Cu(II) complexes act as efficient anti- JASCO-J810 spectrometer equipped with a Peltier temperature-
cancer agents by an oxidative stress mechanism.
controlling programmer ( 0.1 °C).
The synthesis of 4-phenyl-2,6-di(thiazole-2-yl)pyridine (PDTP)
2-Acetylthiazole (1.28 g, 10 mmol) was added into a solution of
benzaldehyde (0.53 g, 5 mmol) in EtOH (10 mL). KOH pellets
(0.77 g, 85%, 10 mmol) and NH3 (aq., 1.49 mL, 25%, 20 mmol)
were then added to the mixture. After stirring at room temp-
erature for 4 h, a silver-white solid was collected by filtration
and washed with EtOH (3 × 10 mL). The yield was 81%. The
ligand was used in the next step to obtain the copper complex
without further purification. Anal. calcd for C17H11N3S2 (%): C,
63.53; H, 3.45; N, 13.07; found (%): C, 63.51; H, 3.46; N, 13.02.
ES-MS (CH3OH) m/z: 322 [M + H]+. 1H NMR (500 MHz,
d6-DMSO): δ 8.42 (s, 1H); 8.08 (d, 2H, J = 6 Hz); 7.96 (d, 2H, J =
6 Hz); 7.94 (d, 2H, J = 7 Hz); 7.58 (dd, 1H, J = 8 Hz); 7.56 (dd,
2H, J = 7 Hz).
Conclusions
In conclusion, the Cu(II) complexes CuPDTP, CuADTP and
CuBFDTP, possessing an extensive heterogeneous ring,
present an impressive plasmid DNA cleaving ability, which
triggers double-strand DNA breaks in the absence of any
exogenous oxidants or reductants. After groove binding with
DNA, the complexes cleaved the DNA strands by a self-activat-
ing oxidative mechanism involving the generation of ROS. It is
proposed that molecular oxygen is reduced to hydrogen per-
oxide in situ in a mechanism that is likely initiated by the
redox of Cu(II) during self-activating oxidative cleavage. The
three Cu(II) complexes show high antitumor activities to HeLa,
BEL-7402 and HepG2 tumor cells. Nuclear chromatin cleavage
has been observed from Hoechst 33258 and AO/EB staining
assays. Alkaline single-cell gel electrophoresis (comet assay)
provides evidence that the Cu(II) complexes indeed induce
DNA fragmentation, which is further evidence of apoptosis.
We hope the results will be of value in further understanding
the DNA cleavage and damage induced by Cu(II) complexes, as
well as laying the foundations for the discovery of new antitu-
mor drugs.
The synthesis of 4-(anthracen-9-yl)-2,6-di(thiazole-2-yl)pyridine
(ADTP)
Anthracen-9-yl-carboxaldehyde (1.03 g, 5 mmol) was dissolved
in an EtOH (20 mL) solution for 1 h, and then 2-acetylthiazole
(1.28 g, 10 mmol), KOH (0.77 g, 85%, 10 mmol) and NH3 (aq.,
1.49 mL, 25%, 20 mmol) were added to the system. After stir-
ring at room temperature overnight, a brown powder was col-
lected by filtration and washed with excess ethanol (3 ×
10 mL). The yield was 65%. Anal. calcd for C25H15N3S2 (%): C,
71.23; H, 3.59; N, 9.97; found (%): C, 71.18; H, 3.62; N, 9.96.
ES-MS (CH3OH) m/z: 422 [M + H]+. 1H NMR (500 MHz, d6-
DMSO): δ 8.32 (dd, 2H, J = 8 Hz); 8.22 (m, 2H); 8.18 (s, 2H);
8.01 (dd, 2H, J = 7.5 Hz); 8.69 (d, 1H, J = 8.8 Hz); 7.61 (m, 4H);
7.50 (m, 2H).
Experimental section
Materials
2-Acetylthiazole, benzaldehyde, anthracen-9-yl-carboxaldehyde,
and CuCl2·2H2O were purchased from Alfa Aesar. Calf thymus
DNA and supercoiled plasmid pBR322 DNA were purchased
from Sigma and Fermentas, respectively. Ultrapure water was
used for all experiments. 2′,7′-Dichlorofluorescein (DCFH) and
1,3-diphenylisobenzofuran (DPBF) were purchased from
Sigma. The agarose gels and ethidium bromide (EB) were pur-
chased from Sangon. T4 DNA ligase, agarose, a gel DNA Extrac-
tion Kit and EcoRI endonuclease were purchased from Takara
Co., Ltd. All commercial solvents were used without further
purification.
The synthesis of 4-(benzofuran-2-yl)-2,6-di(thiazole-2-yl)-
pyridine (BFDTP)
This compound (pale yellow) was synthesized in an identical
manner to that described for PDTP using benzo[b]furan-2-car-
boxaldehyde (0.73 g, 5 mmol) in place of benzaldehyde. The
yield was 55%. Anal. calcd for C19H11N3OS2 (%): C, 63.14; H,
3.07; N, 11.63; found (%): C, 63.11; H, 3.12; N, 11.60. ES-MS
(CH3OH) m/z: 362 [M + H]+. 1H NMR (500 MHz, d6-DMSO):
δ 8.58 (s, 2H); 8.09 (d, 2H, J = 6 Hz); 8.03 (s, 1H); 7.97 (d, 2H,
J = 6 Hz); 7.76 (d, 2H, J = 8 Hz); 7.46 (d, 1H, J = 7 Hz); 7.35
(t, 1H, J = 5 Hz).
Physical measurements
Elemental analyses (C, H and N) were carried out using a Vario
EL elemental analyzer. H NMR spectra were recorded using a
1
The synthesis of Cu(PDTP)Cl2 (CuPDTP)
Varian Mercury-Plus 300 NMR spectrometer or a Varian The ligand PDTP (0.322 g, 1 mmol) was dissolved in 20 mL
INOVA500NB 500 NMR spectrometer with (CD3)2SO as the methanol solution, and CuCl2·2H2O (0.17 g, 1 mmol) in
solvent at room temperature and TMS as the internal standard. methanol was added dropwise to the solution. After stirring
Electrospray mass spectra (ES-MS) were recorded using an LCQ the mixture for 4 h at 50 °C, the solution turned clear green.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 11576–11588 | 11583

View Article Online
Paper
Dalton Transactions
The resulting solution was filtered while it was hot, and the fil- with the full-matrix least-squares technique using the SHELXS
trate was left unperturbed in air to allow slow evaporation. 9763 and SHELXL 9763 program packages. Anisotropic thermal
Dark-green crystals were obtained after one week. The yield parameters were applied to all non-hydrogen atoms except the
was 62%. Anal. calcd for C17H11Cl2N3S2Cu (%): C, 44.79; H, guest molecules. The organic hydrogen atoms were generated
2.43; N, 9.22; found (%): C, 44.75; H, 2.46; N, 9.19. ES-MS geometrically. The crystal data and refinement parameters for
(CH3OH) m/z: 421 [M − Cl]+. UV-vis spectra (DMSO) [λmax, nm the complexes are summarized in Table 2. Selected bond
(ε, M−1 cm−1)]: 292 (34 900), 334 (21 500).
lengths and angles are given in Table 3.
Detailed crystallographic data for the crystal structural ana-
lysis have been collected in the Cambridge Crystallographic
Data Centre, given by the CCDC reference numbers 869871,
869872 and 869873, which contain the supplementary crystal-
lographic data for the present paper.
The synthesis of Cu(ADTP)Cl2 (CuADTP)
The ADTP ligand is poorly soluble in methanol; thus, the use
of a prolonged stirring time was required to dissolve the
ligand. ADTP (0.422 g, 1 mmol) was dissolved in 20 mL metha-
nol. After the ADTP was dissolved completely, CuCl2·2H2O
(0.17 g, 1 mmol) in methanol was added dropwise to the solu-
tion. Stirring was continued for 5 h, and the solution was fil-
tered while it was hot. The filtrate was kept for developing
crystals. The yield was 53%. Anal. calcd for C25H15Cl2N3S2Cu
(%): C, 54.01; H, 2.72; N, 7.56; found (%): C, 53.97; H, 2.75; N,
7.54. ES-MS (CH3OH) m/z: 521 [M − Cl]+. UV-vis spectra
(DMSO) [λmax, nm (ε, M−1 cm−1)]: 301 (27 900), 331 (20 700),
369 (12 700), 389 (11 700).
DNA binding experiments
The DNA binding experiments were performed at room temp-
erature. All spectroscopic titration measurements were carried
out in a 10 mM phosphate buffer at pH 7.2.
The fluorescence quenching experiments were conducted
by adding small aliquots of a 100 μM Cu(II) complex solution
to samples containing 4 μM ethidium bromide (EB) and 80 μM
DNA in buffer. The samples were excited at 340 nm, and the
emission was observed between 500 and 700 nm.
The synthesis of Cu(BFDTP)Cl2 (CuBFDTP)
The Tm was determined by monitoring the UV absorbance
of DNA at 260 nm in the absence or presence of the Cu(II) com-
plexes using a Perkin Elmer Lambda-850 UV-vis spectrometer
equipped with an auto-temperature controller system. The Tm
was obtained by fitting (A − A0)/(Af − A0) versus temperature,
where Af, A0 and A are the final, initial and observed absor-
bances at 260 nm, respectively.
Circular dichroism spectroscopic studies were performed
using a JASCO J-810 automatic recording spectropolarimeter. All
determinations were performed at room temperature with 1 cm
This compound was synthesized in a manner identical to that
described for CuPDTP using BFDTP (0.362 g, 1 mmol) in place
of PDTP. Green crystals were collected after 1 week. The yield
was 40%. Anal. calcd for C19H11Cl2N3OS2Cu (%): C, 46.02; H,
2.24; N, 8.47; found (%): C, 45.99; H, 2.28; N, 8.46. ES-MS
(CH3OH) m/z: 461 [M − Cl]+. UV-vis (DMSO) [λmax, nm (ε, M−1
cm−1)]: 314 (25 400), 331 (22 100).
X-ray crystallography
The intensity data were collected using a Bruker Apex CCD pathway cells. The CD spectra were run from 400–220 nm at a
area-detector diffractometer (Mo Kα). Absorption corrections speed of 100 nm min−1, and the data were recorded at an interval
were applied by using a multi-scan program SADABS. The of 0.1 nm. The buffer background was automatically subtracted,
structures were solved with direct methods and were refined and the data were accumulated 3 times and then averaged.
Table 2 Crystallographic data for CuPDTP, CuADTP and CuBFDTP
CuPDTP
CuADTP
CuBFDTP
Empirical formula
Crystal system
Space group
a/Å
b/Å
c/Å
α/°
β/°
C17H11Cl2CuN3S2
Monoclinic
P21/n
18.5932(18)
8.4641(8)
23.526(2)
90
102.327(2)
90
3617.1(6)
8
C25H15Cl2CuN3S2
Orthorhombic
P212121
10.241(3)
10.290(3)
23.602(6)
90
C19H11Cl2CuN3OS2
Triclinic
P1
ˉ
7.9078(11)
11.1470(16)
11.5729(16)
111.368(2)
92.003(2)
90.218(2)
949.3(2)
2
90
90
γ/°
V/Å3
2487.3(11)
4
1.285
Z
μ/mm−1
1.739
1.674
1.668
1.735
Dc/g cm−3
1.527
θ range (°)
1.27–26.00
7026/0.0338
0.0447
0.1094
0.0686
1.73–25.99
4883/0.0372
0.0496
0.1383
0.0609
1.89–25.99
3688/0.0213
0.0362
0.0895
0.0453
Unique reflns/Rint
Final R [I > 2σ(I)] R1
Final R [I > 2σ(I)] wR2
R indices (all data) R1
R indices (all data) wR2
0.1318
0.1477
0.0972
11584 | Dalton Trans., 2013, 42, 11576–11588
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Table 3 Selected bond lengths (Å) and angles (°) for CuPDTP, CuADTP and CuBFDTP
CuPDTP
CuADTP
CuBFDTP
Cu(1)–N(2)
Cu(1)–N(3)
Cu(1)–N(1)
Cu(1)–Cl(2)
Cu(1)–Cl(1)
1.977(3)
2.035(3)
2.056(3)
2.2520(11)
2.4665(11)
Cu(1)–N(2)
Cu(1)–N(3)
Cu(1)–N(1)
Cu(1)–Cl(1)
Cu(1)–Cl(2)
1.995(4)
2.052(4)
2.068(4)
2.2203(15)
2.4988(14)
Cu(1)–N(2)
Cu(1)–N(1)
Cu(1)–N(3)
Cu(1)–Cl(2)
Cu(1)–Cl(1)
1.987(2)
2.054(2)
2.053(2)
2.2307(9)
2.4556(9)
N(2)–Cu(1)–N(3)
N(2)–Cu(1)–N(1)
N(3)–Cu(1)–N(1)
N(2)–Cu(1)–Cl(2)
N(3)–Cu(1)–Cl(2)
N(1)–Cu(1)–Cl(2)
N(2)–Cu(1)–Cl(1)
N(3)–Cu(1)–Cl(1)
N(1)–Cu(1)–Cl(1)
Cl(2)–Cu(1)–Cl(1)
78.26(12)
78.05(12)
154.30(12)
155.17(9)
99.16(9)
98.15(9)
100.50(9)
98.14(9)
N(2)–Cu(1)–N(3)
N(2)–Cu(1)–N(1)
N(3)–Cu(1)–N(1)
N(2)–Cu(1)–Cl(1)
N(3)–Cu(1)–Cl(1)
N(1)–Cu(1)–Cl(1)
N(2)–Cu(1)–Cl(2)
N(3)–Cu(1)–Cl(2)
N(1)–Cu(1)–Cl(2)
Cl(1)–Cu(1)–Cl(2)
78.14(15)
78.71(16)
155.36(17)
163.18(12)
100.53(12)
99.17(13)
92.11(12)
96.38(12)
92.78(13)
104.69(6)
N(2)–Cu(1)–N(1)
N(2)–Cu(1)–N(3)
N(1)–Cu(1)–N(3)
N(2)–Cu(1)–Cl(2)
N(1)–Cu(1)–Cl(2)
N(3)–Cu(1)–Cl(2)
N(2)–Cu(1)–Cl(1)
N(1)–Cu(1)–Cl(1)
N(3)–Cu(1)–Cl(1)
Cl(2)–Cu(1)–Cl(1)
78.01(9)
78.18(9)
152.75(9)
158.36(8)
97.76(7)
99.48(7)
98.57(7)
100.39(7)
96.09(7)
103.06(3)
95.81(9)
104.30(4)
Nuclease activities
methanol solution were used to monitor the changes in the
luminescence as a function of time. ex = 405 nm, em = 479 nm.
As a control experiment, [Ru(bpy)3]2+ was reported to generate
1O2 as a result of DPBF consumption, which showed a signifi-
cant decrease in fluorescence. In our experiment, the fluo-
rescence intensity of all the copper complexes was observed to
decrease in the presence of DPBF, suggesting the probable for-
mation of 1O2 during the nuclease reaction.
The reactions were carried out with 40 μM pBR322 DNA and
Cu(II) complexes in a 10 mM phosphate buffer at pH 7.2. The
copper complexes were initially prepared in DMF, then diluted
in phosphate buffer. A fixed amount of DNA was mixed with
the copper complex solution and diluted by phosphate buffer
to a total volume of 10 μL. In the nuclease activity mechanism
studies, different ROS radical scavengers were added to the
DNA solution before the complexes. The following procedure
is similar to the typical nuclease activity study. The samples
were incubated for 1.5 h at 37 °C in the dark, followed by
quenching with 2 μL of loading buffer (0.25% bromophenol
blue, 1.86% EDTA and 50% glycerol). Electrophoresis was com-
pleted at 80 V loading on agarose gel (0.8%) for 1.5 h in TBE
buffer (89 mM Tris, 89 mM H3BO3, 2 mM EDTA). After staining
with an EB solution (1 μg mL−1), the bands were visualized
under UV light and photographed with an Alpha Innotech
IS-5500 Chemiimager.
Similar to the mechanism of DPBF when used as a 1O2
sensor, DCFH is oxidized by H2O2 and generates dehydro-form
2′,7′-dichlorofluorescein (DCF), emitting significant green fluo-
rescence. The H2O2 radical detection was performed by adding
the copper complexes to the DCFH methanol solution and
recording the signal changes in fluorescence (ex = 470 nm,
em = 525 nm). To further verify the involvement of H2O2 in the
cleaving process, the Cu(II) complexes were added to a catalase
pre-incubated DCFH solution, and the emission levels were
observed. As a negative control system, the catalase–DCFH
mixture exhibited insufficient changes in the fluorescence
upon the addition of the copper complexes due to the
reduction of H2O2 radicals.
ROS radical detection experiments
The generation of ROS in the cleavage reaction was detected by
dye degradation or by using sensitive radical sensors. Hydroxyl
radical formation was quantified by rhodamine B (10 µM)
degraded in the presence of Cu(II) complexes (100 µM) in phos-
phate buffer under aerobic conditions. The degradation of the
dye was monitored by the UV absorbance of rhodamine B at
552 nm. A parallel experiment was carried out with a mixture
of FeCl2 (100 µM) and H2O2 (10 µM) (the Fenton conditions) as
the control. The control was used to verify that the dye
degrades in the presence of hydroxyl radicals.
Cell lines and cell culture
Human cancer cell lines, including cervical carcinoma HeLa
and hepatocellular carcinoma HepG2, were purchased from
American Type Culture Collection (ATCC, Manassas, VA). The
human hepatocellular carcinoma BEL-7402 cell line was
obtained from the Cell Bank (Cell Institute, Sinica Academica
Shanghai, Shanghai, China). All cell lines were maintained in
either RPMI-1640 or DMEM medium supplemented with fetal
bovine serum (10%), penicillin (100 units per mL) and strepto-
mycin (50 units per mL) at 37 °C in a CO2 incubator (95% rela-
tive humidity, 5% CO2).
The fluorescent probes DPBF53 and DCFH54 have been
reported to detect 1O2 and H2O2, respectively. These two
probes were used to evaluate the ROS species in our
Cytotoxicity
experiment.
1
In the presence of O2, DPBF is activated, forming its non- In vitro cytotoxicity tests were carried out using a standard
emissive derivative. From the fluorescence measurements, the MTT
(3-(4,5-dimethylthiazole-2-yl)-2,5-diphenyltetrazolium
level of fluorescence consumption can be identified. Mixtures bromide) colorimetric assay.55 The cells were plated in 96-well
of [Ru(bpy)3]2+ or the Cu(II) complexes with DPBF in a microassay culture plates (1 × 104 cells per well) and grown
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 11576–11588 | 11585

View Article Online
Paper
Dalton Transactions
overnight at 37 °C in a 5% CO2 incubator. The test compounds Intracellular reactive oxygen species (ROS) measurement
were then added to the wells to achieve final concentrations
The formation of intracellular ROS in the cells was measured
according to the manufacturer’s protocol for the Reactive
Oxygen Species Assay Kit (Beyotime Institute of Biotechnology,
China). The cultured HeLa cells were treated with an IC50 con-
ranging from 10−6 to 10−4 M. The control wells were prepared
by the addition of culture medium (100 μL). The wells contain-
ing culture medium without cells were used as blanks. The
plates were incubated at 37 °C in a 5% CO2 incubator for 48 h.
centration of the Cu(II) complexes and the untreated cells were
Upon completion of the incubation step, the stock MTT dye
maintained as the control. After incubation for 12 h, the cells
solution (20 μL, 5 mg mL−1) was added to each well. After 4 h
were harvested and washed twice, then resuspended in 10 μM
2,7-dichlorodihydrofluorescein diacetate (DCFH-DA) and incu-
bated at 37 °C for 30 min. The levels of intracellular ROS were
of incubation, buffer (100 μL) containing N,N-dimethylform-
amide (50%) and sodium dodecyl sulfate (20%) was added to
solubilize the MTT formazan. The optical density of each well
examined with flow cytometry (FACSCanto II, BD Biosciences,
was then measured on a microplate spectrophotometer at a
USA) and an inverted fluorescence microscope (Zeiss Axio
wavelength of 490 nm. The IC50 value was determined from
Observer D1). The wavelength of excitation was 485 nm, and
plots of % viability against the dose of compound added.
the fluorescence was measured at 530 nm. The level of intra-
cellular ROS was expressed as the mean fluorescence intensity.
Data acquisition and analysis were performed using BD FACS-
Diva software v6.0.
Hoechst 33258 nuclear staining60
The control and the HeLa cells were treated with the Cu(II)
complexes at the noted concentrations for 24 h in the dark. At
the end of the incubation, the cells attached to the glass plates
were washed twice with ice-cold PBS and then were fixed with Acknowledgements
4% paraformaldehyde in PBS for 10 min. A solution of
The authors thank the National Science of Foundation of
Hoechst 33258 (20 μg mL−1) was added to stain the cells for
China (no. 21071155, 21172273, 21171177), the National
10 min in the dark. After washing twice with PBS, the cells
Science
of
Foundation
of
Guangdong
Province
were examined for condensed nuclei under an inverted fluo-
rescence microscope (Zeiss Axio Observer D1).
(9351027501000003), the Research Fund for the Doctoral
Program of Higher Education (20110171110013), the State Key
Laboratory of Optoelectronic Materials and Technologies
(2010-ZY-4-5) and Sun Yat-Sen University for financial
support. X. J. H. thanks the support of the Science and Tech-
nology Department of Hunan Province (2010FJ4090).
Ethidium bromide/acridine orange (AO/EB) staining61
A monolayer of HeLa cells was incubated in the absence or
presence of the Cu(II) complexes at a concentration of 50 μM at
37 °C and 5% CO2 for 24 h. After 24 h, the cells were stained
with an AO/EB solution (100 μg mL−1 AO, 100 μg mL−1 EB).
The samples were observed under an inverted fluorescence
microscope (Zeiss Axio Observer D1).
Notes and references
1 D. S. Sigman, D. R. Graham, V. D. Aurora and A. M. Stern,
J. Biol. Chem., 1979, 254, 12269–12272.
2 C. Marzano, M. Pellei, F. Tisato and C. Santini, Anti-Cancer
Agents Med. Chem., 2009, 9, 185–211.
Alkaline single-cell gel electrophoresis (comet assay)62
An alkaline single-cell gel electrophoresis (comet assay) was
performed to measure the extent of DNA fragmentation
induced by the Cu(II) complexes in the HeLa cells after 24 h.
Briefly, 1 × 104 cells per mL of treated and untreated HeLa
cells were mixed with 0.8% low-melting-point agarose at a
ratio of 1 : 10 (v/v) and spread on slides precoated with 1%
normal agarose. The embedded cells were lysed in a precooled
lytic solution (2.5 M NaCl, 100 mM EDTA, 10 mM Tris base
and 1% Triton X-100, 10% DMSO; the last two compounds
were added fresh, pH 10) at 4 °C for 120 min, rinsed, and equi-
librated in an alkaline electrophoresis buffer (0.3 M NaOH and
1 mM EDTA, pH 13). The electrophoresis was performed in an
ice bath at 25 V and 300 mA for 20 min, and the slides were
neutralized with 0.4 M Tris-HCl (pH 7.5), stained with PI solu-
tion (2 μg mL−1) for 5 min in the dark and analyzed using an
inverted fluorescence microscope (Zeiss Axio Observer D1).
3 J. A. Cowan, Curr. Opin. Chem. Biol., 2001, 5, 634–642.
4 T. Bortolotto, P. P. Silva, A. Neves, E. C. Pereira-Maia and
H. Terenzi, Inorg. Chem., 2011, 50, 10519–10521.
5 T. K. Goswami, B. V. S. K. Chakravarthi, M. Roy,
A. A. Karande and A. R. Chakravarty, Inorg. Chem., 2011, 50,
8452–8464.
6 N. Mitic, S. J. Smith, A. Neves, L. W. Guddat, L. R. Gahan
and G. Schenk, Chem. Rev., 2006, 106, 3338–3363.
7 Q. Jiang, N. Xiao, P. F. Shi, Y. G. Zhu and Z. J. Guo, Coord.
Chem. Rev., 2007, 251, 1951.
8 L. Tjioe, A. Meininger, T. Joshi, L. Spiccia and B. Graham,
Inorg. Chem., 2011, 50, 4327–1972.
9 D. D. Li, F. P. Huang, G. J. Chen, C. Y. Gao, J. L. Tian,
W. Gu, X. Liu and S. P. Yan, J. Inorg. Biochem., 2010, 104,
431–441.
The DNA contents in the head and tail were quantified using 10 S. M. Hadi, M. F. Ullah, U. Shamim, S. H. Bhatt and
the “comet score” from Autocomet.
A. S. Azmi, Chemotherapy, 2010, 56, 280–284.
11586 | Dalton Trans., 2013, 42, 11576–11588
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
11 W. K. Pogozelski and T. D. Tullius, Chem. Rev., 1998, 98,
1089–1108.
and P. Saarenketo, J. Chem. Soc., Dalton Trans., 2000, 1447–
1462.
12 Y. M. Zhao, J. H. Zhu, W. J. He, Z. Yang, Y. G. Zhu, Y. Z. Li, 35 H. Chao, G. Yang, G. Q. Xue, H. Li, H. Zhang,
J. F. Zhang and Z. J. Guo, Chem.–Eur. J., 2006, 12, 6621–
6629.
I. D. Williams, L. N. Ji, X. M. Chen and X. Y. Li, J. Chem.
Soc., Dalton Trans., 2001, 1326–1331.
13 J. W. Chen, X. Y. Wang, Y. Shao, J. H. Zhu, Y. G. Zhu, 36 M. J. Waring, J. Mol. Biol., 1965, 13, 269–282.
Y. Z. Li, Q. Xu and Z. J. Guo, Inorg. Chem., 2007, 46, 3306– 37 C. L. Liu, J. Y. Zhou and H. B. Xu, J. Inorg. Biochem., 1998,
3312.
14 C. Tu, Y. Shao, N. Gan, D. Xu and Z. J. Guo, Inorg. Chem., 38 X. L. Wang, H. Chao, H. Li, X. L. Hong, L. N. Ji and X. Y. Li,
2004, 43, 4761–4766. J. Inorg. Biochem., 2004, 98, 423–429.
15 B. K. Santra, P. A. N. Reddy, G. Neelakanta, S. Mahadevan, 39 G. Yang, J. Z. Wu, L. Wang, L. N. Ji and X. Tian, J. Inorg.
71, 1–6.
M. Nethaji and A. R. Chakravarty, J. Inorg. Biochem., 2002,
89, 191–196.
16 S. S. Tonde, A. S. Kumbhar, S. B. Padhye and R. J. Butcher,
J. Inorg. Biochem., 2006, 100, 51–60.
Biochem., 1997, 66, 141–144.
40 G. A. Neyhart, N. Grover, S. R. Smith, W. A. Kalsbeck,
T. A. Fairly, M. Cory and H. H. Thorp, J. Am. Chem. Soc.,
1993, 115, 4423–4428.
17 A. Kellett, M. O’Connor, M. McCann, M. McNamara, 41 F. Gao, H. Chao, F. Zhou, X. Chen, Y. F. Wei and L. N. Ji,
P. Lynch, G. Rosair, V. McKee, B. Creaven, M. Walsh, J. Inorg. Biochem., 2008, 102, 1050–1059.
S. McClean, A. Foltyn, D. O’Shea, O. Howe and 42 S. Mahadevan and M. Palaniandavar, Inorg. Chem., 1998,
M. Devereux, Dalton Trans., 2011, 40, 1024–1027. 37, 3927–3934.
18 P. U. Maheswari, S. Roy, H. den Dulk, S. Barends, G. P. van 43 B. Saha, M. M. Islam, S. Paul, S. Samanta, S. Ray,
Wezel, B. Kozlevcár, P. Gamez and J. Reedijk, J. Am. Chem.
Soc., 2006, 128, 710–711.
19 P. U. Maheswari, M. van der Ster, S. Smulders, S. Barends,
C. R. Santra, S. R. Choudhury, B. Dey, A. Das, S. Ghosh,
S. Mukhopadhyay, G. S. Kumar and P. Karmakar, J. Phys.
Chem. B, 2010, 114, 5851–5861.
G. P. van Wezel, C. Massera, S. Roy, H. den Dulk, P. Gamez 44 L. Y. Li, H. N. Jia, H. J. Yu, K. J. Du, Q. T. Lin,
and J. Reedijk, Inorg. Chem., 2008, 47, 3719–3727.
20 K. Li, L. H. Zhou, J. Zhang, S. Y. Chen, Z. W. Zhang,
K. Q. Qiu, H. Chao and L. N. Ji, J. Inorg. Biochem., 2012,
113, 31–39.
J. J. Zhang, H. H. Lin and X. Q. Yu, Eur. J. Med. Chem., 45 T. B. Thederahn, M. D. Kuwabara, T. A. Larsen and
2009, 44, 1768–1772. D. S. Sigman, J. Am. Chem. Soc., 1989, 111, 4941–4946.
21 E. Lamour, S. Routier, J. L. Bernier, J. P. Catteau, 46 A. Sreedhara, J. D. Freed and J. A. Cowan, J. Am. Chem. Soc.,
C. Bailly and H. Vezin, J. Am. Chem. Soc., 1999, 121,
1862–1869.
22 K. Ghosh, P. Kumar, N. Tyagi, U. P. Singh, V. Aggarwal and
M. C. Baratto, Eur. J. Med. Chem., 2010, 45, 3770–3779.
2000, 122, 8814–8824.
47 J. T. Wang, Q. Xia, X. H. Zheng, H. Y. Chen, H. Chao,
Z. W. Mao and L. N. Ji, Dalton Trans., 2010, 39, 2128–
2136.
23 K. Ghosh, P. Kumar, N. Tyagi, U. P. Singh and N. Goel, 48 S. Borah, M. S. Melvin, N. Lindquist and R. A. Manderville,
Inorg. Chem. Commun., 2011, 14, 489–492. J. Am. Chem. Soc., 1998, 120, 4557–4562.
24 K. Ghosh, P. Kumar, V. Mohan, U. P. Singh, S. Kasiri and 49 J. A. Cowan, Curr. Opin. Chem. Biol., 2001, 5, 634–642.
S. S. Mandal, Inorg. Chem., 2012, 51, 3343–3345.
25 R. M. Burger, Chem. Rev., 1998, 98, 1153–1170.
50 Y. Jin and J. A. Cowan, J. Am. Chem. Soc., 2005, 127, 8408–
8415.
26 L. Worth, B. L. Frank, D. F. Christner, M. J. Absalon, 51 S. Mahapatra, J. A. Halfen, E. C. Wilkinson, L. Que Jr. and
J. Stubbe and J. W. Kozarich, Biochemistry, 1993, 32, 2601–
2609.
27 S. J. Sucheck, J. F. Ellena and S. M. Hecht, J. Am. Chem.
Soc., 1998, 120, 7450–7460.
W. B. Tolman, J. Am. Chem. Soc., 1994, 116, 9785–9786.
52 Q. Zhu, Y. Lian, S. Thyagarajan, S. E. Rokita,
K. D. Karlin and N. V. Blough, J. Am. Chem. Soc., 2008,
130, 6304–6305.
28 A. T. Abraham, X. Zhou and S. M. Hecht, J. Am. Chem. Soc., 53 H. J. Yu, S. M. Huang, L. Y. Li, H. N. Jia, H. Chao,
2001, 123, 5167–5175.
29 H. Sasaki, Chem. Pharm. Bull., 2007, 55, 1762–1767.
Z. W. Mao, J. Z. Liu and L. N. Ji, J. Inorg. Biochem., 2009,
103, 881–890.
30 M. Gras, B. Therrien, G. Suss-Fink, A. Casini, F. Edafe and 54 L. R. Silveira, L. Pereira-Da-Silva, C. Juel and Y. Hellsten,
P. J. Dyson, J. Organomet. Chem., 2010, 695, 1119–1125.
31 J. H. Wang and G. S. Hanan, Synlett, 2005, 1251–1254.
Free Radical Biol. Med., 2003, 35, 455–464.
55 T. Mosmann, J. Immunol. Methods, 1983, 65, 55–63.
32 A. W. Addison, T. N. Rao, J. Reedijk, J. van Rijn and 56 I. Vermes and C. Haanen, Adv. Clin. Chem., 1984, 31, 177–
C. C. Verschoor, J. Chem. Soc., Dalton Trans., 1984, 1349–
1356.
33 G. D. Storrier, S. B. Colbran and D. C. Craig, J. Chem. Soc.,
Dalton Trans., 1998, 1351–1364.
34 N. W. Alock, P. R. Barker, J. M. Haider, M. J. Hannon,
C. L. Painting, Z. Pikramenou, E. A. Plummer, K. Rissanen
246.
57 R. van Furth and T. L. Van Zwet, J. Immunol. Methods, 1988,
108, 45–51.
58 J. S. Savill, P. M. Henson, J. E. Henson, C. Haslett,
M. J. Walport and A. H. Wyllie, J. Clin. Invest., 1989, 83,
865–875.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 11576–11588 | 11587

View Article Online
Paper
Dalton Transactions
59 N. A. Thornberry and Y. Lazebnik, Science, 1998, 281, 1312–1316.
60 S. Chen, J. Lin, S. Liu, Y. Liang and S. Lin-Shiau, Toxicol.
Sci., 2008, 102, 138–149.
L. A. Leinwand, Cold Spring Harbor Laboratory Press,
New York, ch. 15, vol. 1, 1998.
62 A. Stang and I. Witte, Mutat. Res., Genet. Toxicol. Environ.
Mutagen., 2009, 675, 5–10.
61 G. P. Amarante-Mendes, E. Bossy-Wetzel, T. Brunner,
D. Finucane, D. R. Green and S. Kasibhatla, in Cells: A lab- 63 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystal-
oratory manual, ed. D. L. Spector, R. D. Goldman and
logr., 2008, 64, 112–122.
11588 | Dalton Trans., 2013, 42, 11576–11588
This journal is © The Royal Society of Chemistry 2013