A ruthenium(II) complex capable of inducing and stabilizing bcl-2 G-quadruplex formation as a potential cancer inhibitor

Article abstract of DOI:10.1016/j.jinorgbio.2013.12.005

Two ruthenium(II) complexes (Ru-complexes) were synthesized and characterized in this study. The selectivity and ability of the complexes to interact with bcl-2 DNA were investigated here. It turned out that [Ru(ip) ₃](ClO₄)₂·2H₂O (complex 1, ip = 1H-iminazole [4,5-f][1,10] phenanthroline) could induce and stabilize the formations of G-quadruplexes more effectively than [Ru(pip)₃] (ClO₄)₂·2H₂O (complex 2, pip = 2-phenylimidazo-[4,5-f][1,10]phenanthroline) did. Considering the important role of the Ru-complex ligand in inducing and stabilizing the formations of G-quadruplex in our previous studies, we speculate that the overlarge ligand of complex 2 may block its binding affinity for G-quadruplexes. Complex 1 also induced cell apoptosis in in vitro assays. In general, this study provided potentially important information for further development of the Ru-complexes as good inducers and stabilizers of bcl-2 G-quadruplex DNA for cancer treatment.

Full text of DOI:10.1016/j.jinorgbio.2013.12.005

Journal of Inorganic Biochemistry 134 (2014) 1–11
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry
journal homepage: www.elsevier.com/locate/jinorgbio
A ruthenium(II) complex capable of inducing and stabilizing bcl-2
G-quadruplex formation as a potential cancer inhibitor
Jingnan Zhang a,1, Qianqian Yu a,1, Qian Li , Licong Yang , Lanmei Chen , Yanhui Zhou ⁎, Jie Liu ⁎
a
a
a,b
a,
a,
a
Department of Chemistry, Jinan University, Guangzhou 510632, PR China
School of Pharmacy, Guangdong Medical College, Zhanjiang, 524023, PR China
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Two ruthenium(II) complexes (Ru-complexes) were synthesized and characterized in this study. The
selectivity and ability of the complexes to interact with bcl-2 DNA were investigated here. It turned out
Received 23 September 2013
Received in revised form 12 December 2013
Accepted 13 December 2013
Available online 24 December 2013
that [Ru(ip)
and stabilize the formations of G-quadruplexes more effectively than [Ru(pip)
-phenylimidazo-[4,5-f][1,10]phenanthroline) did. Considering the important role of the Ru-complex ligand in in-
3
](ClO
4
)
2
·2H
2
O (complex 1, ip = 1H-iminazole [4,5-f][1,10] phenanthroline) could induce
3
](ClO ·2H O (complex 2, pip =
)
4 2
2
2
ducing and stabilizing the formations of G-quadruplex in our previous studies, we speculate that the overlarge li-
gand of complex 2 may block its binding affinity for G-quadruplexes. Complex 1 also induced cell apoptosis in
in vitro assays. In general, this study provided potentially important information for further development of the
Ru-complexes as good inducers and stabilizers of bcl-2 G-quadruplex DNA for cancer treatment.
Crown Copyright © 2014 Published by Elsevier Inc. All rights reserved.
Keywords:
Ruthenium
Bcl-2
G-quadruplexes
DNA structures
Anticancer
1
. Introduction
G-quadruplexes found throughout the human genome are consid-
interesting to study because of their formations, structures, and stabili-
ties. Many ligands exhibiting structural recognition and high affinity for
G-quadruplex DNA have been characterized for their ability to induce
and stabilize G-quadruplexes. Most G-quadruplex binding ligands re-
ported thus far have planar aromatic molecules, which are capable of in-
ducing interaction with guanine tetrads via π–π interactions. In
addition, they also have side chains directed toward the quadruplex
loops [11–14]. However, the selectivity of these ligands for G-
quadruplex DNA over duplex DNA is mostly poor, usually resulting in
acute toxic and intolerable side effects on normal tissues. Hence, in-
creased selectivity for G-quadruplex and low cytotoxicity are among
the factors to consider in designing drugs targeting the bcl-2 G-
quadruplex for cancer chemotherapy [15–17]. Only a few Ru-
complexes have been currently verified to promote the formation and
stabilization of G-quadruplexes [18]. Shi et al. have reported the
remarkable ability of a novel dinuclear complex to promote antiparallel
G-quadruplex formation [19]. Thomas and co-workers investigated the
binding preferences of a dinuclear ruthenium(II) complex with differ-
ent quadruplex DNA structures. They found that the differences in
quadruplex binding affinity and optical signature are consequences of
the structural features of the quadruplexes [20]. They also reported
that the Ru-complexes displayed sequence selectivity and high-
affinity binding to duplex DNA through groove binding [21]. Our
research group has reported that the Ru-complexes could stabilize G-
rich DNA sequences and induce configurational changes in the DNA,
where the planar size of the ancillary ligands of the Ru-complexes has
an important function in the stabilization and induction of G-
quadruplex structures [22]. In our previous reports, we found that
phenanthroline had a better stabilizing effect on G-quadruplex DNA
ered as potential anticancer targets. Many promoter regions could
form G-quadruplex structures, resulting in gene regulation at the level
of transcription. This suggestion has led to extensive investigations of
the role of G-quadruplex formation in the transcriptional regulation of
oncogenic promoters such as bcl-2 [1], c-myc [2], VEGF [3], HIF-1α [4],
and c-kit [5,6].
The bcl-2 gene product is a mitochondrial membrane protein that
has an essential function in cell survival; it exists in a delicate balance
with other apoptosis-related proteins and functions as an inhibitor of
cell apoptosis [7,8]. With its crucial role in apoptosis regulation, the
bcl-2 gene is already viewed as an important target in anticancer treat-
ment using pro-apoptotic drugs.
The human bcl-2 gene has two promoters, P1 and P2 [9], for control-
ling its transcriptional initiation. The P1 promoter, which is the principal
transcriptional promoter for bcl-2 gene, is a GC-rich region capable of
forming G-quadruplex [10]. The GC-rich region upstream of the P1
promoter has been shown to be critically involved in the regulation of
bcl-2 gene expression, thereby controlling the occurrence of cancer dis-
eases. Numerous studies have focused on understanding quadruplex-
mediated alteration in the molecular recognition of the promoter
regions, thus indicating their influence on disease phenotypes. The in-
teractions between G-quadruplexes and small molecules have become
⁎
Corresponding authors. Tel./fax: +86 20 85220223.
E-mail address: tliuliu@jnu.edu.cn (J. Liu).
These authors contributed equally to this work.
1
0162-0134/$ – see front matter. Crown Copyright © 2014 Published by Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jinorgbio.2013.12.005

2
J. Zhang et al. / Journal of Inorganic Biochemistry 134 (2014) 1–11
compared with bipyridyl as an ancillary ligand of the Ru-complex
23–26]. This finding might suggest that the rigid plane of the ligand
dominates the interaction of the Ru-complex with the G-quadruplex
DNA. Therefore, in the present study, we synthesized and characterized
UV–visible spectra were obtained by using a Shimadzu MPS-2000 spec-
trophotometer. Emission spectra were detected by a Shimadzu RF-5000
luminescence spectrometer (Shimadzu, Japan) and CD (circular
dichroism) spectra were recorded on a Jasco J-810 spectropolarimeter.
FRET (fluorescence resonance energy-transfer) assay was carried out
on Light Cycler 2 real-time PCR (polymerase chain reaction) amplifier
(Roche). Fluorescent quantitation polymerase chain reaction was tested
by Bio-Rad Chrome 4 PCR instrument. Flow cytometry (BD Bioscience,
America) was applied in the determination of cell apoptosis and the
change of mitochondrial membrane potential and pH was measured
by Orion Model 720 A pH.
[
two Ru-complexes: [Ru(ip)
iminazole [4,5-f][1,10] phenanthroline) and [Ru(pip)
complex 2, pip = 2-phenylimidazo-[4,5-f][1,10]phenanthroline),
3
](ClO
4
)
2
·2H
2
O (complex 1, ip = 1H-
3
](ClO ·2H O
4
)
2
2
(
using ip and pip as ancillary ligands, which had larger planarity than
those that were previously used for synthesis. The ability of the two
Ru-complexes to interact with bcl-2 DNA was investigated. In consider-
ation of the abundance of G-quadruplex DNA, including bcl-2 DNA,
telomere DNA, and VEGF, in many kinds of cancer cells, we studied
the selectivity of the two complexes for bcl-2 DNA. We found that com-
plex 1 had a stronger selectivity for bcl-2 DNA and a better ability in sta-
bilizing bcl-2 DNA compared with complex 2. Moreover, complex 1
could induce configurational changes in bcl-2 DNA. The results demon-
strate that the planar size of an ancillary ligand is a key factor in the
interaction of the Ru-complexes with G-quadruplex structures, and an
overlarge ligand of this kind of the Ru-complexes might reduce its bind-
ing affinity for G-quadruplexes.
2.3. Synthesis of ligands and the two complexes
3
RuCl ·nH2O (AR) was obtained from the Kunming Institute
of Precious Metals. 1,10-phenanthroline, 2,2′-bipyridine, form-
aldehyde and benzaldehyde were obtained from Sigma. Synthesis
of 1,10-phenanthroline-5,6-dione, [Ru(ip)
3 4 2 2
](ClO ) ·2H O and
[Ru(pip) ](ClO ·2H O was prepared and characterized according
3
4
)
2
2
to the literature references [30,31].
To further understand the mechanisms associated with the cellular
activities of the Ru-complexes, which had been shown to include
potential antitumor activities [27], aspects relating to cytotoxicity, cellu-
lar localization of the Ru-complex, apoptosis, caspase activity, and
change in mitochondrial membrane potential induced by complex 1 in
HeLa cells were analyzed. The results show that complex 1 displays a
broad-spectrum anti-proliferative activity for various cancer cells, espe-
cially for HeLa cells, and could enter the nucleus effectively. Complex 1
has a function in inducing the depletion of mitochondrial membrane
potential and in the activation of caspases, two events which result in
cell apoptosis. Herein, the synthesis and the G-quadruplex interaction
properties of the two complexes are elucidated.
2.3.1. Synthesis of the ligand “ip”
A mixture of 1, 10-phenanthroline-5,6-dione (0.525 g, 2.5 mmol),
methanal (0.26 ml 36%–38%, 3.5 mmol), ammonium acetate (3.88 g,
50 mmol) and glacial acetic acid (10 mL) was refluxed for 4 h. The
cooled deep red solution was diluted with 40 mL water and neutralized
with ammonium hydroxide to give a yellow precipitate. The precipitate
was collected and purified by column chromatography on silica gel
(60–100 mesh) with ethanol as eluent to give the compound as yellow
+
powder. Yield: 0.378 g, 45%. ESI-MS (CH3OH): m/z = 327 ([M + H] ).
2.3.2. Synthesis of the ligand “pip”
The ligand was prepared by a method similar to that of the ligand ip
2
. Experimental section
by replacing methanal with benzaldehyde (371 mg, 3.5 mmol). Yield:
+
88%. ESI-MS (CH3OH): m/z = 297 ([M + H] ).
2
.1. Reagents and materials
2
.3.3. Synthesis of [Ru(ip)
A mixture of [Ru(bpy)
3
](ClO
Cl
4
)
2
·2H
2
O
DNA oligomers bcl-2 (Pu23): 5′-GGGCGC GGGAGG AATTGG
2
2
]·2H
2
O (0.26 g, 0.5 mmol), ethylene
GCGGG-3′, Mutbcl-2 (MutPu23): 5′-AAACGC AAAAGG AATTAA
ACGGG-3′, F27T: 5′-FAM-CGGGCG CGGGAG GAAGGG GGCGGGAGC-
TAMRA-3′ (Donor fluorophore FAM is 6-carboxy-fluorescein; Acceptor
fluorophore TAMRA is 6-carboxytetramethylrhodamine), VEGF (5′-
GGGCGG GCCGGG GGCGGG-3′), HTG21 (5′-TTAGGG TTAGGG TTAGGG
TTAGGG-3′) and bcl-2 rev DNA were purchased from Shanghai Sangon.
Calf-thymus (CT)-DNA (Sigma; highly polymerized stored at 4 °C; long-
term storage at −20 °C). Concentrations of these oligomers were deter-
mined by measuring the absorbance at 260 nm wavelength after melt-
ing. Single-strand extinction coefficients were calculated from
mononucleotide data using a nearest-neighbor approximation [28,29].
The formations of intramolecular G-quadruplexes were carried out as
follows: the oligonucleotide samples were annealed in different buffers
at 95 °C for 5 min, slowly cooled to room temperature, and then incu-
bated at 4 °C overnight. Buffer A: 10 mM Tris–HCl, pH = 7.4. Buffer B:
glycol (10 mL) and ligand ip (0.168 g, 0.5 mmol) was refluxed for 4 h
under argon. Upon cooling, a red precipitate was obtained by a dropwise
addition of saturated aqueous NaClO solution. The precipitated complex
4
dried under vacuum, and purified by chromatography over alumina (200
meshes), using MeCN-acetonitrile (8:1, v/v) as an eluent, yield: 62%. The
sample shows good solubility in solvents such as MeCN, DMSO and
+
acetone. ESI-MS (CH3OH): m/z = 761 ([M-2ClO
4
-H] ), m/z = 381
2
+ 1
([M-2ClO
(s, 3H, J = 5Hz), 7.97 (d, 6H, J = 5Hz), 7.72 (dd, 6H, J1 = 5Hz,
J2 = 3Hz). Anal. Calc.for C43 12Ru: C, 62.76; H, 4.53; N, 20.43.
Found: C, 62.67; H, 4.45; N, 20.01.
4 3 2
/2] ) H NMR. [(CD ) SO]: ppm δ9.09 (d, 6H, J = 8Hz), 8.60
37
H N
2.3.4. Synthesis of [Ru(pip)
This complex was synthesized in a manner identical to that de-
scribed for [Ru(ip) ](ClO ·2H O, with pip (0.62 g, 2.1 mmol) in
place of ip, yield: 68%. ESI-MS (CH3OH): m/z = 989 ([M-2ClO
3 4 2 2
](ClO ) ·2H O
3
4
)
2
2
+
1
0 mM Tris–HCl, 100 mM KCl, pH = 7.4. Buffer C: 10 mM Tris–HCl,
4
-H] ),
4 3 2
m/z = 495 ([M-2ClO /2] ) H NMR [(CD ) SO]: ppm δ9.01 (d, 6H,
2
+ 1
6
0 mM KCl, pH = 7.4. Other reagents and solvents were purchased
commercial sources unless otherwise specified. Doubly distilled water
was used to prepare the buffer solutions.
J = 8Hz), 8.43 (d, 6H, J = 8Hz), 8.03 (d, 6H,J = 5Hz), 7.76 (dd, 6H,
J1 = 5Hz,J2 = 3Hz), 7.60 (t, 6H, J = 7.5Hz), 7.53 (t, 3H, J = 7.5Hz).
Anal. Calc.for C63
54
H N12Ru: C, 70.05; H, 4.98; N, 15.56. Found: C,
2
.2. Physical measurements
70.15; H, 5.12; N, 15.71.
1H NMR spectra were recorded on a Varian Mercury-plus 300 NMR
2.4. Fluorescent experiments
6
spectrometer with [D ] DMSO as solvent and SiMe4 as an internal
standard at 300 MHz at room temperature. Electrospray ionization
mass spectrometry (ES-MS) was recorded on a LQC system (Finngan
MAT, USA) by using CH3CN as a mobile phase. Microanalysis (C, H,
and N) was carried out with an Elementar Vario EL elemental analyzer.
For the selectivity assay, 4 μM DNA (CT-DNA, G4-TTA, VEGF or bcl-2
DNA) and 2 μM Ru-complex were adequately mixed in buffer B (10 mM
Tris–HCl, 100 mM KCl, pH 7.4) over night in 4 °C. Before the test,
the solution was incubated for about 20 min to achieve initial test

J. Zhang et al. / Journal of Inorganic Biochemistry 134 (2014) 1–11
3
temperature. For the titration test, the emission spectra as well as titra-
tion curves were recorded through a constant concentration of the Ru-
complexes, to which the DNA stock solutions were added step by step
at room temperature. The Ru-complex (3000 μL) in a 1.0 cm path
length quartz cuvette was loaded into the fluorimeter sample
block and the solution was incubated for 5 min before the absorption
spectra were recorded. The titration processes were repeated several
times until no further change was observed in the spectra. The fluores-
cent spectra were recorded in wavelength 500 ~ 800 nm by using a
Shimadzu RF-5000 luminescence spectrometer (Shimadzu, Japan) at
room temperature with excitation wavelength 460 nm. The changes
in the Ru-complex concentration caused by dilution at the end of each
titration were negligible.
PCR detection system in a total reaction volume of 20 μL containing
0.2 μM of labeled oligonucleotide and different concentrations of
the Ru-complexes. Measurements were made in triplicate on a DNA
Opticon Engine (MJ Research) with excitation wavelength at 470 nm
and detection at 530 nm. Fluorescence readings were taken at intervals
of 1 °C over the range 37–99 °C with a constant temperature being
maintained for 30 s prior to each reading to ensure a stable value. For
the competitive FRET assay, the F27T (0.4 μM) was mixed with 1 μM
of the Ru-complexes and 0, 1 and 20 μM of CT-DNA in buffer B and
the measurements were carried out as previously mentioned. Final
analysis of the data was carried out using Origin 7.5 (Origin Lab Corp.).
2.9. PCR stop assay
2
.5. Continuous variation assessment
The oligonucleotide bcl-2 (Pu23): 5′-GGGCGCGGGAGGAATTGGGC
GGG-3′ and the corresponding complementary sequence (bcl-2rev)
were used here. The reactions were performed in 1× PCR buffer,
containing 10 pmol of each oligonucleotide, 0.16 mM dNTP, 2.5 U Taq
polymerase, and various concentrations of the Ru-complexes. Reaction
mixtures were incubated in a thermocycler with the following cycling
conditions: 94 °C for 3 min, followed by 30 cycles of 94 °C for 30 s,
The preparation of two series of the solutions for these experiments:
one with varying concentrations of the Ru-complex diluted in buffer B,
and another one with increasing mole fractions of [Ru-complexes]/
DNA]. The concentration of the Ru–DNA solutions was constant at
0 μM. The emission spectrum was collected from 500 to 750 nm
using a quartz cell with a 1 cm path length at 25 °C. The ΔI values
were calculated by subtracting the fluorescence intensity of the Ru-
complex solution without Pu23 DNA from the fluorescence intensity
of corresponding complex solution with Pu23 DNA at λmax. This value
was plotted versus the complex mole fraction to obtain a Job plot of
ΔI–X. Analysis of the data was carried out using Origin 7.5 (Origin Lab
Corp).
[
1
5
1
8 °C for 30 s, and 72 °C for 30 s. PCR products were then analyzed on
5% native polyacrylamide gels in 1× TBE buffer and silver stained.
2.10. CD study
The CD titration procedure is described as follows: CD titration was
performed at a fixed DNA concentration (2 μM) with various concentra-
tions of the Ru-complexes in different buffers. After each addition of the
Ru-complexes, the reaction was mixed adequately and allowed to equil-
ibrate for at least 5 min prior to a scan. The spectra were accumulated
over the wavelength range of 220 to 330 nm at 25 °C with a bandwidth
of 5 nm by using a Jasco J-810 spectropolarimeter. For each sample, the
spectrum was scanned at least three times. The instrument was flushed
continuously with pure evaporated nitrogen throughout the experi-
ment. The scan for buffer alone was subtracted from the average scan
for each sample.
2
.6. Absorption spectra study
The absorption titrations of the Ru-complexes in buffer B were de-
tected by using a constant Ru-complex concentration (20 μM) to
which increments of the DNA stock solution was added. The volume
of the Ru-complex was 3000 μL. During the titration, aliquots of a
stock solution of Pu23 were also added to each cuvette to eliminate
the absorbance value of DNA itself. The solutions were mixed by repeat-
ed inversion and incubated for 5 min before detection. The titration pro-
cesses were repeated several times until no change was observed in the
spectra indicating an achievement of binding saturation. Emission spec-
tra of the Ru-complexes in the presence of increasing amounts of bcl-2
DNA in buffer B were recorded in wavelength 230 ~ 650 nm by using a
Shimadzu MPS-2000 spectrophotometer at room temperature with ex-
citation wavelength 460 nm. The changes of the Ru-complexes concen-
tration due to dilution at the end of each titration were negligible.
2.11. Cell culture
Cells were cultured in DMEM (Dulbecco's modified Eagle's medi-
um) supplemented with 10% heat inactivated fetal bovine serum,
−
1
−1
1
00 μg mL
penicillin and 100 μg mL
streptomycin. Cells were
maintained at 37 °C in a 5% CO
changed every three days.
2
incubator, and the media were
2
.7. Chromogenic reaction
This assay was for the preparation of the Ru-complex-promoted G-
2.11.1. MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide) assay
quadruplex-hemin DNAzyme. In detail, an equal volume of the Ru-
complex solution (in water) was added to the DNA solutions (20 μM
DNA, 10 mM Tris–HCl, 100 μM EDTA, pH 8.00) allowing the DNA
strands to form the G-quadruplex structure within 40 min. Subsequent-
ly, an equal volume of hemin (dissolved in DMSO) was dissolved in the
above G-quadruplex solutions and kept for 2 h at room temperature to
Cell cytotoxicity was determined by measuring the ability of cells to
transform MTT to a purple formazan dye after incubation with different
Ru-complexes for 48 h. Cells including HeLa, A549, CNE, HepG2 and
MDA-MB-231 were incubated to exponential phase, seeded in 96-well
culture clusters (Costar) at a volume of 100 μL per well and incubated
form the DNAzymes. Finally, 180 μL of 296 μM TMB-1.76 mM H
2
O
2
was
for 24 h at 37 °C in 5% CO
2
. (Cells at exponential growth phase were
4
−1
added as the substrate of 20 μL peroxidatic DNAzyme system. The mix-
ture was kept for 1.5 h at room temperature and photographed with a
digital camera.
pre-diluted to 4 × 10 cells mL ). Then the cells were further incubat-
ed with various concentrations of the Ru-complexes (2.5, 5, 10, 20, and
40 μM) for 48 h. After the treatment, MTT (20 μL, 5 mg mL− ) the
solution was added to each well. After a further period of incubation
1
2
.8. FRET assay
(4 h at 37.8 °C in 5% CO
2
) the suspension was replaced with DMSO
(
150 μL/well) to dissolve the formazan. Finally, the plates were ana-
The fluorescent labeled oligonucleotide F21T was prepared as a
00 μM solution in buffer B and then annealed by heating to 90 °C for
min, followed by cooling to room temperature overnight. Fluores-
lyzed on a microplate reader (SpectroAmaxTM250) at a wavelength of
570 nm (the absorbance of the complexes at this wavelength can be
neglected). All data were from at least three independent experiments
and expressed as mean ± standard deviation.
1
5
cence melting curves were determined with a Bio-Rad iQ5 real-time

4
J. Zhang et al. / Journal of Inorganic Biochemistry 134 (2014) 1–11
2
.11.2. Laser confocal microscopy image analysis
HeLa cells were grown on a laser confocal microscopy 35 mm cul-
concentration by a BCA kit. As for caspase activity assay, the cell lysates
2
were placed in 96-well plates and then the specific caspase substrates
were added. Ac-DEVD-AMC (N-acetyl-Asp-Glu-Val-Asp-7-amino-4-
methylcoumarin) is for caspase-3, Ac-IETD-AMC (Ac-Ile-Glu-Thr-Asp-
AMC) is for caspase-8, and Ac-LEHD-AMC (Ac-Leu-Glu-His-Asp-AMC)
is for caspase-9. The plates were then incubated at 37 °C for 1 h, and cas-
pase activity was determined by fluorescence intensity with the excita-
tion and emission wavelengths set at 380 nm and 440 nm, respectively.
4
2
ture dishes at a density of 1.0 × 10 cells under a 5% CO atmosphere
for 24 h in 37 °C, then were added in the cell layer. Cells were treated
with complex 1 (20 and 6 μM) and incubated for different time intervals
(
3 h, 6 h and 24 h). Before the imaging, the cell nuclei were stained
−
1
with Hoechst 33342 (5 μg mL ) for 20 min. After being washed with
PBS (phosphate buffered saline) twice, the cells were cultured in fresh
medium on a thermo-cell culture FCS2 chamber of Carl Zeiss Cell
Observer (Jena, Germany). Cell images were captured with a monochro-
matic CoolSNAP FX camera (Roper Scientific, USA) and analyzed
by using AxioVision 4.2 software (Carl Zeiss). The Ru-complexes had:
3. Results and discussion
3.1. Fluorescence selectivity of the Ru-complex between quadruplex DNA
4
88 nm excitation, detection at 560–615 nm (green) and 625–754 nm
and non-quadruplex DNA
(
red).
The synthesis routes of the two complexes are detailed in the
Experimental section, and the mass spectra and H NMR spectra data
2
.11.3. Flow cytometric analysis
HeLa cells in exponential growth were incubated without or with
1
shown in Fig. S1 and Fig. S2 distinctly confirm the generation of com-
plexes 1 and 2. The structures of complexes 1 and 2 are presented in
Fig. 1.
To understand the interaction between the Ru-complex and
quadruplex DNA, we firstly investigated the selectivity of the Ru-
complex for quadruplex DNA and non-quadruplex DNA. Four different
DNA sequences including CT-DNA, HTG21, VEGF and bcl-2 DNA
complex 1 at different concentrations (5, 10 and 20 μM) for 24 h, then
harvested by trypsinization, washed twice with ice-cold PBS, resus-
pended in 70% ethanol in PBS, and kept at 4 °C overnight. Before the
analysis, the cells were adjusted to a final density of 1 × 10 cells/mL
in PBS buffer containing RNase A (1 mg mL ) and stained with
6
−
1
−
1
1
0 mg mL of propidium iodide (PI) in darkness for 30 min at 37 °C.
The resulting suspension was then passed through a nylon mesh filter
400 meshes) and analyzed on a Becton Dickinson FACS can. Cell cycle
distribution was calculated using the Modifit-3 program.
(
Pu23) were chosen to react with complexes 1 and 2 to investigate
(
the variations of fluorescence behavior of the two complexes. The selec-
tivities of the two Ru-complexes for different DNA structures were pre-
sented by the fluorescence strength which stood for the binding
affinities via an emission spectroscopy. As shown in Fig. 2, solutions con-
taining different DNAs performed various emission spectra strength. For
complex 1 (Fig. 2A), a slight increase in emission was observed in the
presence of CT-DNA, on the contrary, a great increase in emission was
displayed in the presence of all G-quadruplex DNA sequences, namely,
HTG21, VEGF and bcl-2 DNA. Moreover, the fluorescence intensity of
the group treated with bcl-2 DNA was much stronger than that of the
other two G-quadruplex DNA sequences. As for complex 2 (Fig. 2B),
all groups showed a slight increase in fluorescence regardless of the
highest emission strength of the group dealt with bcl-2 DNA. These re-
sults were attributed to the different affinity abilities of DNA for the Ru-
complexes and indicated that there was a greater overlap between the
aromatic surfaces of the metal complexes and the bases allowing their
interaction when bound to quadruplex as opposed to the duplex DNA
[20,32]. Both complexes 1 and 2 harbored the best selectivity for G-
quadruplex sequences especially for bcl-2 DNA while complex 1 per-
formed better than complex 2. As we know the ligand of complex 2
has a better stretched planarity than that of complex 1, however, the
2
.11.4. Assessment of mitochondrial membrane potential
Detection of mitochondrial depolarization was performed by flow
cytometry using Mitoprobe JC-1 assay kit (Molecular Probes) according
to the manufacturer's instruction. The cells were treated with different
concentrations of complex 1 for 24 h after adherence. After the incuba-
tion, the cells were washed with warm PBS and resuspended in 0.5 mL
6
−1
−1
of PBS (about 1 × 10 cells mL ). 10 μg mL
of JC-1 was added to
the cells for 15 min incubation at 37 °C in the dark before analysis.
The percentage of the green fluorescence from JC-1 monomers was
used to demonstrate the cells that lost mitochondrial membrane poten-
tial. The red and green fluorescence intensities of JC-1 dye were mea-
sured by flow cytometry. A total of 10,000 events were analyzed for
each sample.
2
.11.5. Caspase activity assay
Harvested cell pellets were suspended in cell lysis buffer and incu-
bated on ice for 1 h. After a centrifugation at 10,000 g for 30 min, the
supernatants were collected and immediately measured for protein
3 4 2 2 3 4 2 2
Fig. 1. Structures of the two Ru-complexes: [Ru(ip) ](ClO ) ·2H O (1), [Ru(pip) ](ClO ) ·2H O (2).

J. Zhang et al. / Journal of Inorganic Biochemistry 134 (2014) 1–11
5
A
B
2
C
Ru-complex
CT-DNA
HTG21
VEGF
1
1
1
1
1
2
.8
2
1
1
000
500
000
1500
000
2+CT-DNA
+CT-DNA
+HTG21
+VEGF
+bcl-2
2
+HTG21
2+VEGF
+bcl-2
2.4
2.0
1.6
1
0
0
1
bcl-2
2
.2
.8
.4
5
00
5
00
0
5
0
0.0
00 550 600 650 700
Wavelength/nm
500 550 600 650 700
Wavelength/nm
Complex 1
Complex 2
Fig. 2. Selectivity of the Ru-complex between quadruplex DNA and non-quadruplex DNA. Complexes 1 (A) and 2 (B) were mixed with CT-DNA, HTG21, VEGF and bcl-2 DNA (Pu23) in Tris-
KCl buffer (10 mM Tris–HCl, 100 mM KCl, pH 7.4) respectively. [Ru-complex] = 2 μM, [DNA] = 4 μM. (C) The relative fluorescence strength of complexes 1 and 2 without or with dif-
ferent DNA sequences depicted in a histogram.
observation indicated that the ligand of complex 1 had a better affinity
ability for bcl-2 DNA than that of complex 2. Considering our previous
study that a greater planar area of the ancillary ligand always showed
a higher binding affinity for quadruplex DNA [24], we infer that only a
moderate planarity ligand of the Ru-complex will exhibit the best affin-
ity ability to bcl-2 DNA. Definitely, it needs to be further verified by
more studies.
quadruplex DNA (Pu23). The absorption spectra of these complexes
dealt with DNA were shown in Fig. 4. With an increasing addition of
DNA, both complexes displayed hypochromism around 460 nm and
280 nm, which were conceived as the metal-to-ligand charge transfer
band (MLCT band) and intraligand absorption band (IL band). As
hypochromism is usually observed when a complex binds to DNA
through intercalation because of the strong stacking interaction
between an aromatic chromophore and the DNA base pairs in the inter-
calation mode, so the extent of hypochromism indicates the intercala-
tion binding strength [33,34]. The absorbance spectra were recorded
after each addition. As for complex 1, the values of ΔH (%) were calculat-
ed to be 15.9% (around 460 nm) and 22.3% (around 280 nm) respec-
tively. Likewise, the values of ΔH (%) for complex 2 were calculated
to be 9.1% (around 460 nm) and 29.6% (around 280 nm) respectively.
The MLCT bands and IL band of complexes 1 and 2 showed
hypochromism but no obvious bathochromism. However, another
narrow separated band at about 263 nm was present in complex 1
3
.2. Binding ability of the Ru-complex with quadruplex DNA
In terms of the above results, we chose Pu23 as the object for a fur-
ther understanding of the interaction mechanism. The emission spectra
of complexes 1 (Fig. 3A) and 2 (Fig. 3B) in the presence of increasing
amounts of Pu23 in 10 mM Tris–HCl buffer, 100 mM KCl (pH = 7.4)
were determined. Fluorescence intensities increased gradually with
the incremental concentration of DNA and the extent of fluorescence in-
0
tensity enhancement (I/I ) was 2.37 and 1.57 respectively for com-
(
the second arrow in Fig. 4A). This result indicated probably that the
plexes 1 and 2, which further illustrated that complex 1 showed a
stronger interaction with Pu23. In general, complex 1 shows high selec-
tivity and affinity ability to bcl-2 DNA.
To further validate the meaningful binding stoichiometries of the
Ru-complexes with quadruplex DNA, we analyzed continuous variation
using the luminescence intensities. As shown in Fig. 3C, the X values of
the point of intersection for complexes 1 and 2 with bcl-2 G-quadruplex
were 0.6 and 0.51, respectively. This result indicated that complex 1 or 2
could bind to bcl-2 DNA (Pu23) by a 3:2 or 1:1 binding stoichiometric
ratio.
two complexes could intercalate the G-quadruplex through a π–π stack-
ing interaction. In order to compare the DNA-binding affinities of the
two complexes quantitatively, their intrinsic DNA-binding constants
(K
b
) were calculated by monitoring the changes of the MLCT absorbance
of the two complexes according to the Scatchard equation (Eqs. (1a) to
1c)).
(
Ä
Cb ¼ C nðF−F Þ=ðFmax−F0Þ
ð1aÞ
ð1bÞ
ð1cÞ
t
0
r ¼ C =C
b
DNA
3
.3. Biding absorption spectra studies
r=Cf nK −rK
b
b
Although the fluorescence experiments authenticated the real inter-
action between the Ru-complex and Pu23, it was still necessary to
further investigate the particular mechanism of the interaction.
Absorption spectra measurements were always used to further clar-
ify the nature of the interaction between these Ru-complexes and G-
Here, C
t
is the total compound concentration, F being the observed
being
fluorescence emission intensity at a given DNA concentration, F
the intensity in the absence of DNA, and Fmax is the fluorescence of the
totally bound compound. Binding data were cast into the form of a
0
3
3
2
2
1
1
500
000
500
000
500
000
A
increased by 137%
3000
2500
B
C
increased by 57%
2
00
150
00
1
2
2
1
1
000
500
000
1
5
0
0
5
00
0
5
00
0
5
00 550 600 650 700
Wavelength/nm
500 550 600 650 700
Wavelength/nm
0.0 0.2 0.4 0.6 0.8 1.0
Mole Fraction of Ru-complex
Fig. 3. Emission spectral traces of complexes 1 (A) and 2 (B) in Tris-KCl buffer (10 mM Tris–HCl, 100 mM KCl, pH 7.4) at increasing ratios of [Pu23]/[Ru-complex], [Ru-complex] = 2 μM,
Pu23] = 0–10 μM, EX = 467. Arrows show the emission intensity changes upon increasing DNA concentrations. (C) Job plot using luminescence data for complexes 1 and 2 with bcl-2
[
DNA in 10 mM Tris–HCl buffer, 100 mM KCl (pH = 7.4). Total concentration was maintained at 1 μM, the stoichiometries of complexes 1 and 2 with bcl-2 DNA were about 3:2, 1:1
respectively.

6
J. Zhang et al. / Journal of Inorganic Biochemistry 134 (2014) 1–11
0
.8
B
A
1
0
0
0
0
0
.0
.8
.6
.4
.2
.0
1
2
0.6
0.4
0
0
.2
.0
3
00
400
500
600
300
400
500
600
Wavelength/nm
Wavelength/nm
Fig. 4. Absorption spectra of complexes 1 (A) and 2 (B) ([Ru-complex] = 10 μM) in the presence of Pu23 in Tris-KCl buffer (10 mM Tris–HCl buffer, 100 mM KCl, pH = 7.4,
Pu23] = 0–14 μM). Arrows indicate the changes in absorbance with increasing DNA concentrations. These results are mean values of at least three independent experiments.
[
Scatchard plot of r/C
and C is the free ligand concentration [28]. The intrinsic binding con-
stants K
.7 ± 0.03 × 10 M
constant of complex 1) was larger than Kb2 (binding constant of com-
plex 2) indicating that complex 1 bound to the DNA more tightly than
complex 2 did. All the absorbance spectra were well consistent with
the result of the emission spectroscopic analysis. The higher binding
affinity of complex 1 probably results from the fact that its ancillary li-
gand can intercalate with the base pairs or enter into the grooves within
DNA more easily [21,35,36].
f
versus r, where r is the binding ratio C
b
/[DNA],
3.5. Analysis of bcl-2 G-quadruplex selectivity and binding stability via
FRET assays
f
6
−1
b
of complexes 1 and 2 were 1.05 ± 0.4 × 10 M
and
5
−1
4
(Table 1). The binding constant Kb1 (binding
The thermodynamic selectivity and stabilization activity of complex
1 to G-quadruplex DNA were studied using FRET melting experiment
which has come across as a useful tool to characterize the selectivity
of binding ligand toward quadruplex structure over the duplex [43].
We firstly used the FRET measurements to determine the selectivity of
complex 1 to G-quadruplex DNA and the competitive binding property
of CT-DNA. In a system that contained 0.4 μM F27T oligonucleotide and
1 μM complexes, the melting temperatures increased roughly by
13.1 °C and 6.0 °C respectively (Fig. 6A–C). Interestingly, with an addi-
tion of increasing amounts of CT-DNA, no significant change in the melt-
ing temperatures was observed even at 20 μM CT-DNA. These results
indicated that the Ru-complexes exhibited prompt selectivity to bind
to G-quadruplex structures and not to other duplex DNA.
To further quantify the stabilization of the complexes for bcl-2 DNA,
we conducted another FRET melting assay. Shown in Fig. 6D–E, the
3
.4. Visual detection of G-quadruplex structures induced by
the Ru-complexes
As we have investigated the interaction between the Ru-complexes
and bcl-2 DNA (Pu23), it was necessary to further study the new DNA
structure evoked by the Ru-complexes.
melting temperature (T ) of the G-quadruplex in the absence and
m
Although certain kinds of G-rich sequences have been demonstrated
to form G-quadruplex structures readily at physiological concentrations
presence of increasing amounts of complexes varied obviously. The Tm
of the DNA sequence increased in a dose-dependent manner and
reached a summit value when r was 2. The relevance of ΔTm (the
enhanced melting temperature) vs “r” was depicted in Fig. 6F, obvious-
ly, ΔTm (1) was much higher than ΔTm (2) which was 18.0 °C and 7.9 °C
respectively. All these results showed that complex 1 had a better stabi-
lization for the G-quadruplex, which was consistent with the results of
absorption titration studies showing that complex 1 had the highest
Kb (intrinsic DNA-binding constant) value.
+
+
of Na and K in vitro [37–40], the existence of G-quadruplex struc-
tures in vivo is still controversial [41,42]. The determination of G-
quadruplex structures has great significance for cancer research and
drug development for the lack of direct evidence of this quadruplex
structure in living cells. Then we conducted a facile and visual approach
to detect the existence of G-quadruplexes with the naked eyes.
It is well known that most G-quadruplex DNA can be effectively in-
+
duced by K , and G-quadruplexes have the ability to bind with hemin
to form peroxidase-like DNAzymes which can catalyze H
oxidation of TMB (3,3′,5,5′-tetramethylbenzidine). As shown in Fig. 5,
the bcl-2 DNA-treated mixture produced DNAzymes to catalyze H
2
O
2
-mediated
3.6. PCR stop assay
2 2
O -
In order to further evaluate the efficiency of the complexes stabiliz-
ing G-quadruplex DNA, a PCR stop assay was used to ascertain whether
these complexes were bound to a test oligomer bcl-2 DNA and mutbcl-2
DNA. Fig. 7A illustrated that in the presence of complex 1, the template
sequence could form G-quadruplex structures unable to amplify to the
PCR product. Moreover, the inhibitory effect of complex 1 was enhanced
clearly as the concentration increased from 1.0 to 8.0 μM with even no
PCR product detected at an 8.0 μM concentration. However, complex 2
mediated oxidation of TMB to show an easily identified color change
+
in the presence of K , complex 1 or 2. However, a similar color change
failed to emerge in the CT-DNA mixture in the absence or presence of
complexes 1 and 2. In conclusion, both complexes could induce bcl-2
+
DNA into G-quadruplex structures as well as K . But for complexes in-
cubated with double strand DNA (CT-DNA), the final solution retained
the original color. Obviously, the CT-DNA sequence cannot form the G-
quadruplex structure even in the presence of the Ru-complexes.
Table 1
Absorption spectra and hypochromism of complex 1 and 2.
Complex
λ
max(free)
λ
[nm]
max(bound)
Δλ
[nm]
ΔH
[%]
K
b
[10 M−1
]
6
[nm]
bcl-2
bcl-2+K+ bcl-2+1
CT-DNA+1 bcl-2+2 CT-DNA+2
4
57
275
49
472
84
459
276
250
473
296
2
1
1
1
15.9
22.3
22.6
9.1
1
2
1.05 ± 0.4
Fig. 5. Characterizations of the DNAzyme functions of bcl-2 DNA (Pu23) and CT-DNA in
the presence of K+ (2 mM) and complexes 1 and 2 (500 nm) in the TMB-H
system.
, 794 mM;
2
2 2
O
O
2 2
0.47 ± 0.03
Conditions: TMB, 266 mM in Tris-MES buffer (25 mM MES, pH 5.10); H
DNA, 500 nM; hemin, 500 nM.
2
12
29.6

J. Zhang et al. / Journal of Inorganic Biochemistry 134 (2014) 1–11
7
2
0
B
C
complex+F27T
complex+F27T+2µM CT-DNA
complex+F27T+20µM CT-DNA
1
0
0
.0
.8
.6
1
0
0
0
0
0
.0
.8
.6
.4
.2
.0
A
15
0.4 M F27T
0
+
+
.4 M F27T
1
0
5
0
+
+
1
2
M complex 2
M CT-DNA
1
2
M complex 1
M CT-DNA
0.4
0.2
0.0
and 1 M complex 2
20 M CT-DNA
and 1 M complex 2
and 1 M complex 1
20 M CT-DNA
and 1 M complex 1
+
+
4
0
50 60 70 80 90
Temperature/°C
40
50
60
70
80
Complex 1
Complex 2
Temperature/°C
1
20
15
10
5
D
E
2
1
2
1
0
0
0
0
0
.0
1.0
0.8
0.6
0.4
0.2
0.0
F
.8
.6
.4
.2
.0
F27T
r=0.5
r=1
F27T
r=0.5
r=1.0
r=2.0
r=2
0
4
0
50 60 70 80 90
Temperature/°C
40
50
60
70
80
0.0
0.5
1.0
1.5
2.0
Temperature/°C
[Ru]/[F27T]
Fig. 6. Bcl-2 G-quadruplex selectivity and binding stability of the Ru-complexes. (A–C) Competitive FRET melting curves of F27T (0.4 μM) with 2 μM of the Ru-complexes and 0, 5 and
0 μM of CT-DNA in 10 mM Tris–HCl buffer containing 60 mM KCl. (D–F) The FRET melting curves for experiments carried out with F27T (1 μM in 10 mM Tris–HCl, 60 mM KCl,
5
pH 7.4) and complexes 1 and 2.
hardly inhibited the appearance of the PCR product with a low concen-
tration. These data suggested that complex 1 was an efficient G-
quadruplex binder and inducer even at a low concentration. Besides,
in order to exclude the possibility of enzyme inhibition by the com-
plexes, the parallel experiment was performed using a mutated oligo-
mer (mutbcl-2 DNA, Fig. 7B) instead of bcl-2 DNA in identical
conditions. And the results indicated that these complexes could not
obviously inhibit the Taq polymerase at a comparable concentration.
In general, complex 1 can stabilize G-quadruplex structure more effi-
ciently than complex 2 does.
quadruplexes, it can be used to distinguish parallel and antiparallel G-
quadruplex structures depending on their strand orientation [44–46].
In the absence of salt, the CD spectrum of bcl-2 DNA exhibited a neg-
ative band at 240 nm as well as a major positive band at 260 nm at
room temperature, which probably corresponded to the signal of the
bcl-2 DNA random coil. A minor positive band near 295 nm was also ob-
served. Upon titration with complex 1, dramatic changes in the CD spec-
tra were observed. Bands at 240 and 265 nm gradually disappeared
with the addition of the complex and eventually led to the appearance
of a positive band at 240 nm and a major negative band at 260 nm.
Meanwhile, a new, strong, positive band gradually appeared near
2
85 nm (Fig. 8A). It implied that complex 1 could stabilize the combina-
3
.7. CD spectra analysis
tion of bcl-2 G-quadruplex and tended to form the antiparallel confor-
mation. However, upon addition of complex 2, the bands at 240 and
2
The results indicated that complex 2 could only stabilize the combina-
tion of bcl-2 G-quadruplex to some extent but failed to form the
antiparallel G-quadruplex conformation. In the presence of K , bcl-2
CD spectroscopy is used to investigate the conformational properties
65 nm only had the dose-dependent changes of strength (Fig. 8B).
of the enantiomeric chiral molecules in relation to the G-quadruplex
DNA. Although CD cannot give precise structures of these G-
+
DNA can form the hybrid-type G-quadruplex structure [47]. The CD
spectrum of such preformed structure showed a strong positive band
at 290 nm, 265 nm, and a negative band at 235 nm (Fig. 8C, black
line). Upon titration of complex 1 into this DNA solution, the CD spectra
changed with a weak enhancement of the band at 235 nm and a slight
decrease of the band at 265 nm and at 290 nm (Fig. 8c). By contrast, the
addition of increasing amounts of complex 2 to bcl-2 DNA in 100 mM
KCl buffers resulted in no significant changes in the CD spectra
Concentration (µM)
A
0
1
2
4
6
8
Complex 1
Complex 2
(
Fig. 8D). The results indicated that complex 1 was more efficient in
inducing the formation of G-quadruplexes compared with complex 2.
These data implied that the conformation of the G-quadruplex was
stabilized by K , and neither of the two Ru-complexes could change
the conformation of the G-quadruplex in the presence of high ionic
strength.
In the Tris–HCl buffer solution, the CD spectrum of this structure in
the absence of any complex showed a strong positive band at 261 nm,
and a major negative band at 249 nm. The signals demonstrated that
the conformations of mutbcl-2 were in disorder, not in an orderly
secondary structure. Upon titration of complex 1 or complex 2, the CD
spectra produced scarce changes (Fig. 8E,F). No significant change in
Concentration (µM)
16 32
+
B
0
2
4
8
Complex 1
Complex 2
Fig. 7. Effect of the Ru-complexes on the hybridization of bcl-2 DNA (A) and mutbcl-2 DNA
B) in the PCR stop assay. Complexes 1 and 2 were at 0–15 μM concentrations.
(

8
J. Zhang et al. / Journal of Inorganic Biochemistry 134 (2014) 1–11
6
A
1
B
8
6
4
2
0
2
C
3
2
1
0
1
2
2
1
4
2
0
2
-
-
-
-
-
4
2
20 240 260 280 300 320
Wavelength/nm
2
20 240 260 280 300 320
Wavelength/nm
220 240 260 280 300 320
Wavelength/nm
8
8
E
F
2
D
1
1
1
5
0
5
0
2
4
0
4
8
4
0
-
-
-4
-8
-
5
2
20 240 260 280 300 320 340
Wavelength/nm
240 260 280 300 320 340
Wavelength/nm
240 260 280 300 320 340
Wavelength/nm
Fig. 8. CD spectra of bcl-2 DNA (Pu23) in 10 mM Tris–HCl buffer (pH = 7.4) (A, B) and in 100 mM KCl buffer (pH = 7.4) (C, D) in the presence of increasing amounts of complexes 1 and
. (E, F) CD spectra of mut bcl-2 DNA in 10 mM Tris–HCl buffer (pH = 7.4) in the presence of increasing amounts of complexes 1 and 2. [Ru-complex] = 0–12 μM, [DNA] = 2 μM.
2
the CD spectrum was observed even in elevated concentrations of the
complex. The results illustrated that the two complexes could not in-
duce mutbcl-2 DNA to form a G-quadruplex structure and scarcely
interacted with mutbcl-2 DNA.
3.9. Cellular uptake by laser confocal microscopy
The Ru-complex is an autofluorescent complex that emits green
fluorescence and red fluorescence and allows for the visualization of
its intracellular presence in living cells by laser confocal microscopy
[
48]. To investigate the intracellular accumulation of complex 1, we ex-
3
.8. In vitro cytotoxicity analysis
amined its fluorescence in HeLa cells after incubation for 3, 6 and 24 h.
The intracellular green fluorescence was remarkably high in cells after
3
In vitro assay, we firstly investigated the antitumor property of the
h of treatment, followed by a progressive increase to 6 h. As shown
Ru-complexes using the MTT assay in five types of cancer cells, namely,
human cervical cancer (HeLa), human lung carcinoma (A549), naso-
pharyngeal carcinoma cell (CNE), human hepatocellular liver carcinoma
from the upper panel (Fig. 10A), complex 1 mostly entered the cyto-
plasm within 3 h. However, the Ru-complex concentrated well in the
nucleus within 6 h at 20 μM instead of 5 μM (Fig. 10B). This result sug-
gested that complex 1 might enter the nucleus to interact with DNA
after a 6 h treatment, but the Ru-complex couldn't enter the nucleus
well at a too low concentration. The abilities of the complex to enter
the nuclei might be related to its appropriate hydrophobic ligand and
its affinity for the constituents of the nucleus. By the way, the cells ap-
peared to be dying after 24 h incubation with complex 1 (20 μM),
which indicated that the accumulation of this Ru-complex in the nucle-
us of HeLa cells might lead to the inhibition of DNA transcription and
translation resulting in cell death. Therefore, these results revealed
that the nucleus was one of the cellular targets of the Ru-complex and
complex 1 displayed promising anticancer activities.
(
HepG2), and human breast cancer (MDA-MB-231) cells.
As shown from Table 2, the IC50 values varied obviously between dif-
ferent complexes for a certain cell line. It could be concluded that com-
plex 1 had the most efficient anticancer performance compared with
complex 2 in most human cancer cell lines. Most of the tested cancer
cell lines were susceptible to complex 1 particularly the HeLa cells
which were half inhibited at 11.2 μM treatment concentration. The
morphology changes of HeLa cell induced by complex 1 were also
presented by the light microscope (Fig. 9). With increasing concentra-
tions of complex 1, lots of cells became round, detached cells increased
and adherent cells gradually decreased, then a large number of
suspended cells appeared, even accompanied with cell debris. In partic-
ular, both complexes showed weak cytotoxicity against the normal cells
3
.10. Apoptosis and cell cycle arrest induced by complex 1
(
NIH/3T3) with IC50 values at 183.3 μM and 92.0 μM (not listed in the
Table 2).
In brief, all these results indicated that the complex had relatively
higher selectivity to cancer cells than to normal cell lines and complex
exerted well anti-proliferative activity especially in HeLa cells, which
As we know, the inhibition on cancer cell proliferation by anticancer
drugs could be the results of induction of apoptosis or cell cycle arrest, or
a combination of these two modes. Therefore, we performed PI-flow
cytometric analysis to study the potential mechanisms involved in the
cell death induced by complex 1. The results revealed that exposure of
HeLa cells to different concentrations of complex 1 led to a dose-
dependent increase in sub-G1 cell population which was a symbol of
apoptosis (Fig. 11). However, the proportion of G0/G1 phases increased
in a dose-dependent way, which increased up to 71.15% while that of
the control group was only 54.73%. By the way, no significant changes
in S and G2/M phases were observed in treated cells even in a high treat-
ment concentration. These results indicated that the mode of cell death
induced by complex 1 might be a combination of cell apoptosis and cell
cycle arrest.
1
led this Ru-complex to a good application potential in cancer chemopre-
vention and chemotherapy.
Table 2
IC50 of complexes 1 and 2 in various human cancer cells.
Complex
IC50 (μM)
HeLa
A549
CNE
39.3 ± 0.4
HepG2
MDA-MB-231
1
2
11.2 ± 0.5
22.6 ± 1.0
26.4 ± 1.2
74.6 ± 2.8
11.6 ± 0.3
32.8 ± 0.6
24.4 ± 1.1
14.6 ± 0.7
121.0 ± 4.5

J. Zhang et al. / Journal of Inorganic Biochemistry 134 (2014) 1–11
9
Concentration (µM)
Control
5
10
20
Fig. 9. Effect of the Ru-complexes on cell morphology in HeLa cells. Light microscope examination of HeLa cells was performed after a 48 h treatment with various concentrations of the Ru-
complex. Photographs shown are representative of at least three such fields of view per sample, and three independent samples.
3
.11. Mitochondria dysfunction and caspase-dependent apoptosis induced
induced apoptosis, several caspase inhibitors such as z-VAD-fmk
(general caspase inhibitor), z-IETD-fmk (caspase-8 inhibitor) and z-
LEHD-fmk (caspase-9 inhibitor), were used to examine their effects on
the induced cell apoptosis. As shown from Fig. 12C, the induced cell
apoptosis was remarkably suppressed by pretreatment with z-VAD-
fmk in complex 1-induced cell death, suggesting that complex 1 in-
duced a caspase-dependent apoptosis. Cell apoptosis was also sup-
pressed by caspase-8 inhibitor from 24.43% to 8.5%, which confirmed
the contribution of death receptors to the cell apoptosis. Moreover,
caspase-9 inhibitor significantly decreased apoptotic cell death to 9.5%
indicating the important role of the intrinsic mitochondrial-mediated
apoptotic pathway in complex 1-induced apoptosis. In general, all the
results confirmed that complex 1 might promote apoptosis through
the mitochondria-mediated and caspase-activated pathway in HeLa
cells.
by complex 1
Mitochondria play crucial roles in apoptotic signals originating in
both the extrinsic and intrinsic apoptotic pathways. Mitochondrial
dysfunction is the major key in triggering various apoptotic pathways
associated with the initiation of apoptotic cascades [49,50]. Thus, the
status of mitochondrial membrane potential was investigated in com-
plex 1-treated HeLa cells to further study the mechanism of apoptosis
by JC-1 flow cytometric analysis.
As shown in Fig. 12A, complex 1 induced a dose-dependent increase
in the depletion of mitochondrial membrane potential in HeLa cell lines,
as evidenced by the shift of fluorescence from red to green. The percent-
age of depolarized mitochondria in HeLa cells exposed to 10 and 20 μM
complex increased from 3.4% (control) to 29.1% and 39.4%, respectively.
The induced loss of mitochondrial membrane potential resulted in the
release of mitochondrial content and apoptosis-related factors such as
cytochrome c and other apoptosis-inducing factors, then DNA damage
occurred.
As the loss of mitochondrial membrane potential was associated
with the activation of caspase, we determined the activities of initiator
caspases (caspase-8 and caspase-9) and effector caspase-3. Fig. 12B
showed that complex 1 caused activations of caspase-3, caspase-8,
and caspase-9 in the HeLa cells in a dose-dependent manner, which
suggested that both intrinsic and extrinsic apoptotic pathways were in-
volved in the apoptosis. The activity of caspases-3, 8 and 9 increased
significantly in response to the complex treatments, which indicated
that the role of mitochondria in the induction of cell apoptosis was likely
to be very important. To further evaluate the roles of caspases in the
4. Conclusions
In this study, we designed and synthesized the two Ru-complexes
for consideration as promising candidates for selective targeting of G-
quadruplexes. The results revealed that both the Ru-complexes are
capable of inducing and stabilizing the bcl-2 G-quadruplex. They also
had a very strong binding affinity with significant specificity for
quadruplex over duplex interactions. Complex 1 showed a stronger
ability in stabilizing and higher efficiency in inducing the formation of
G-quadruplexes compared with complex 2, which had an overlarge li-
gand planar conformation. These distinctions between complex 1 and
complex 2 prove that the planar conformation of their cores is crucial
for their binding affinities with G-quadruplex for this kind of complex.
Bright + Green + Hoechst 33342
A
Complex 1
2
0 µM
Control
Bright
3 h
6 h
24 h
Overlay
Red
Hoechst 33342
B
Complex 1
µM,6 h
5
Fig. 10. Cellular localization of the Ru-complex by laser confocal microscope. The upper panel (A) is the real-time imaging of the HeLa cells treated with complex 1 (20 μΜ). Nuclei and
complex 1 were visualized by blue and green fluorescence, respectively; scale bar: 10 μm. The lower panel (B) shows the cellular localization of complex 1 (5 μΜ) within 6 h in HeLa cells.
Nuclei and complex 1 were visualized by blue and red fluorescence, respectively; scale bar: 10 μm.

10
J. Zhang et al. / Journal of Inorganic Biochemistry 134 (2014) 1–11
Concentration (µM)
A
B 80
SubG1
G0/G1
G2/M
S
Control
5
10
20
SubG1:0%
60
SubG1: 5.94%
G0/G1:54.53 %
G2/M: 8.86%
S:36.62 %
SubG1:17.01 %
SubG1:24.43 %
G0/G1:71.15 %
G2/M :4.17 %
S:24.68 %
G0/G1:54.73 %
G2/M:14.99 %
S:30.28 %
G0/G1:64.67 %
G2/M :4.86 %
S:30.48 %
40
2
0
0
0
20
40
60
80
100
120
0
30
60
90
120
0
30
60
90
120
0
40
80
120
160
Control
5
10
20
Concentration (µM)
DNA Content
Fig. 11. Effects of complex 1 on cell apoptosis and cell cycle distribution in HeLa cells. (A) The cells treated with different concentrations of complex 1 for 48 h were collected and stained
with PI after fixation as described in the Experimental section. (B) Cellular DNA histograms were analyzed by the MultiCycle software. Each value represents the mean of three indepen-
dent experiments.
Considering our previous studies that showed that a better stretched li-
gand planarity of the Ru-complexes always facilitated the induction and
stabilization of G-quadruplexes [24], we concluded that an undersized
or oversized planar conformation for a Ru-complex might not benefit
bcl-2G-quadruplex targeting. In effect, a Ru-complex ligand should
have the following characteristics: (i) medium size, (ii) high selectivity
for quadruplex structures, (iii) high specificity characterized by high
binding affinity for G-quadruplexes, (iv) low cytotoxicity, and (v) min-
imal side effects.
biological properties of cells through regulation of gene expression. In
the present study, the biological activities of the Ru-complexes in
carcinoma cells were tested. Interestingly, complex 1 displayed a
broad-spectrum anti-proliferative activity for various cancer cells,
especially for HeLa cells, and could partially enter the nucleus in 3 h to
6 h. Then, by flow cytometry, the induction of apoptotic cell death and
G0/G1 cell cycle arrest was found as probable causes for the inhibition
of HeLa cell growth. Further investigation showed that complex 1 also
induced mitochondria-mediated apoptosis by obviously depleting mi-
tochondrial membrane potential, which was further confirmed by the
results of caspase activation. The detailed mechanism of cell apoptosis
With respect to the high G-quadruplex selectivity of the Ru-
complexes, it was inferred that they might be able to affect the
Concentration (µM)
A
Control
5
10
20
5
.6%
29.1%
3
9.4%
3
.4%
Green Fluorescence
Control
B 2
C
10 µM
50
00
50
00
4
3
2
1
0
20 µM
2
1
1
0
0
0
0
5
0
0
Caspase-3 Caspase-8 Caspase-9
Complex 1 (20 µM)
z-VAD fmk
-
-
+
-
+
+
+
-
+
-
-
+
-
-
-
-
z-IETK fmk
z-LEHD fmk
-
-
-
-
-
-
+
-
-
+
-
-
+
-
-
+
Fig. 12. Complex 1 induces mitochondria-mediated and caspase-dependent apoptosis in HeLa cells. (A) Complex 1 induces depletion of mitochondrial membrane potential. Cells treated
with complex 1 for different concentrations were analyzed by JC-1 flow cytometry. (B) Complex 1 induced caspase activation. Cells were treated with complex for 24 h. Significant
difference between treatment and control groups is shown. (C) Effects of caspase inhibitors (40 μ M) on apoptosis induced by complex 1. Cells were pretreated with z-VAD-fmk (general
caspase inhibitor), z-IETD-fmk (caspase-8 inhibitor), and z-LEHD-fmk (caspase-9 inhibitor) for 2 h followed by co-incubation with complex 1 for 24 h. Apoptotic cells were determined by
flow cytometry.

J. Zhang et al. / Journal of Inorganic Biochemistry 134 (2014) 1–11
11
[
[
5] S. Rankin, A.P. Reszka, J. Huppert, M. Zloh, G.N. Parkinson, A.K. Todd, S. Ladame, S.
Balasubramanian, S. Neidle, J. Am. Chem. Soc. 127 (2005) 10584–10589.
6] H. Fernando, A.P. Reszka, J. Huppert, S. Ladame, S. Rankin, A.R. Venkitaraman, S.
Neidle, S. Balasubramanian, Biochemistry 45 (2006) 7854–7860.
and the relationship between the stabilization of bcl-2 G-quadruplex
and cell apoptosis induced by complex 1 are currently under further in-
vestigation. Based on all our results, we suggest that this Ru-complex
can be a good candidate for selective targeting of bcl-2 G-quadruplex.
This complex must be further evaluated for its efficacy as a chemo-
preventive and chemotherapeutic agent for human cancers.
[
[
[
7] D.T. Chao, S.J. Korsmeyer, Annu. Rev. Immunol. 16 (1998) 395–419.
8] J.M. Adams, S. Cory, Science 281 (1998) 1322–1326.
9] T.S. Dexheimer, D. Sun, L.H. Hurley, J. Am. Chem. Soc. 128 (2006) 5404–5415.
[10] M.I. Onyshchenko, T.I. Gaynutdinov, E.A. Englund, D.H. Appella, R.D. Neumann, I.G.
Panyutin, Nucleic Acids Res. 37 (2009) 7570–7580.
[
[
11] H. Han, L.H. Hurley, Trends Pharmacol. Sci. 21 (2000) 136–141.
12] A.M. Burger, F. Dai, C.M. Schultes, A.P. Reszka, M.J. Moore, J.A. Double, S. Neidle, Can-
cer Res. 65 (2005) 1489–1496.
Abbreviations
[
[
[
13] H. Song, J.T. Kaiser, J.K. Barton, Nat. Chem. 4 (2012) 615–620.
14] J.-T. Wang, Y. Li, J.-H. Tan, L.-N. Ji, Z.-W. Mao, Dalton Trans. 40 (2010) 564–566.
15] L.H. Hurley, Nat. Rev. Cancer 2 (2002) 188–200.
G-quadruplex Guanine quadruplex
bcl-2
Ru-complexes Ruthenium(II) complexes
complex 1 [Ru(ip) ](ClO · 2H O (1)
complex 2 [Ru(pip) ](ClO · 2H O (2)
CT-DNA Calf thymus DNA
b-cell lymphoma-2
[16] S. Neidle, G. Parkinson, Nat. Rev. Drug Discov. 1 (2002) 383–393.
[
17] T.m. Ou, Y.j. Lu, J.h. Tan, Z.s. Huang, K.Y. Wong, L.q. Gu, ChemMedChem 3 (2008)
90–713.
18] J.E. Reed, A.A. Arnal, S. Neidle, R. Vilar, J. Am. Chem. Soc. 128 (2006) 5992–5993.
3
)
4 2
2
6
3
)
4 2
2
[
[19] S. Shi, X. Geng, J. Zhao, T. Yao, C. Wang, D. Yang, L. Zheng, L. Ji, Biochimie 92 (2010)
bpy
pip
2,2′-bipyridine; ip:1H-imidazole[4,5-f][1,10]phenanthroline
2-phenylimidazo-[4,5-f][1,10]phenanthroline
370–377.
[
[
20] T. Wilson, M.P. Williamson, J.A. Thomas, Org. Biomol. Chem. 8 (2010) 2617–2621.
21] A. Ghosh, P. Das, M.R. Gill, P. Kar, M.G. Walker, J.A. Thomas, A. Das, Chem. Eur. J. 17
MLCT band Metal-to-ligand charge transfer band
IL band Intraligand absorption band
(
2011) 2089–2098.
[22] S. Shi, J. Liu, T. Yao, X. Geng, L. Jiang, Q. Yang, L. Cheng, L. Ji, Inorg. Chem. 47 (2008)
910–2912.
[
[
2
TMB
FRET
PCR
CD
3,3′,5,5′-tetramethylbenzidine
Fluorescence Resonance Energy-Transfer
Polymerase chain reaction
23] D. Sun, Y. Liu, D. Liu, R. Zhang, X. Yang, J. Liu, Chem. Eur. J. 18 (2012) 4285–4295.
24] D. Liu, Y. Liu, C. Wang, S. Shi, D. Sun, F. Gao, Q. Zhang, J. Liu, ChemPlusChem 77
(2012) 551–562.
25] Q. Yu, Y. Liu, C. Wang, D. Sun, X. Yang, Y. Liu, J. Liu, PloS One 7 (2012) e50902.
26] Q. Yu, Y. Liu, J. Zhang, F. Yang, D. Sun, D. Liu, Y. Zhou, J. Liu, Metallomics 5 (2013)
[
[
Circular dichroism
MTT
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide
222–231.
[27] A. Levina, A. Mitra, P.A. Lay, Metallomics 1 (2009) 458–470.
[
[
[
28] S. Satyanarayana, J.C. Dabrowiak, J.B. Chaires, Biochemistry 31 (1992) 9319–9324.
29] M. Reichmann, S. Rice, C. Thomas, P. Doty, J. Am. Chem. Soc. 76 (1954) 3047–3053.
30] W. Paw, R. Eisenberg, Inorg. Chem. 36 (1997) 2287–2293.
PBS
Phosphate buffered saline
Ac-DEVD-AMC N-Acetyl-Asp-Glu-Val-Asp-7-amino-4-methylcoumarin
Ac-IETD-AMC N-Acetyl-Ile-Glu-Thr-Asp-7-amino-4-methylcoumarin
Ac-LEHD-AMC N-Acetyl-Leu-Glu-His-Asp-7-amino-4-methylcoumarin
[31] J.-Z. Wu, B.-H. Ye, L. Wang, L.-N. Ji, J.-Y. Zhou, R.-H. Li, Z.-Y. Zhou, J. Chem. Soc., Dalton
Trans. (1997) 1395–1402.
[
32] C. Rajput, R. Rutkaite, L. Swanson, I. Haq, J.A. Thomas, Chem. Eur. J. 12 (2006)
611–4619.
4
[
[
33] J.-G. Liu, Q.-L. Zhang, X.-F. Shi, L.-N. Ji, Inorg. Chem. 40 (2001) 5045–5050.
34] X.W. Liu, J. Li, H. Deng, K.C. Zheng, Z.W. Mao, L.N. Ji, Inorg. Chim. Acta 358 (2005)
Acknowledgments
3311–3319.
[
35] M.R. Gill, J.A. Thomas, Chem. Soc. Rev. 41 (2012) 3179–3192.
This work was supported by the National Natural Science Founda-
tion of China (20871056, 21171070, and 21371075), the Planned Item
of Science and Technology of Guangdong Province (c1011220800060,
c1211220800571), the Natural Science Foundation of Guangdong Prov-
ince and the Fundamental Research Funds for the Central Universities.
[36] S.N. Georgiades, N.H. Abd Karim, K. Suntharalingam, R. Vilar, Angew. Chem. Int. Ed.
9 (2010) 4020–4034.
37] L. Petraccone, J.O. Trent, J.B. Chaires, J. Am. Chem. Soc. 130 (2008) 16530–16532.
[38] H.-Q. Yu, D. Miyoshi, N. Sugimoto, J. Am. Chem. Soc. 128 (2006) 15461–15468.
[39] K.N. Luu, A.T. Phan, V. Kuryavyi, L. Lacroix, D.J. Patel, J. Am. Chem. Soc. 128 (2006)
4
[
9963–9970.
[
[
40] K. Shin-ya, Telomestatin, J. Am. Chem. Soc. 123 (2001) 1262–1263.
41] K. Paeschke, S. Juranek, T. Simonsson, A. Hempel, D. Rhodes, H.J. Lipps, Nat. Struct.
Mol. Biol. 15 (2008) 598–604.
Appendix A. Supplementary data
[
42] K. Paeschke, T. Simonsson, J. Postberg, D. Rhodes, H.J. Lipps, Nat. Struct. Mol. Biol. 12
(
2005) 847–854.
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.jinorgbio.2013.12.005.
[43] D. Sun, R. Zhang, F. Yuan, D. Liu, Y. Zhou, J. Liu, Dalton Trans. 41 (2012) 1734–1741.
[44] Y. Xu, Chem. Soc. Rev. 40 (2011) 2719–2740.
[
45] Y. Xu, T. Ishizuka, K. Kurabayashi, M. Komiyama, Angew. Chem. Int. Ed 48 (2009)
833–7836.
[46] C. Zhao, J. Ren, X. Qu, Langmuir 29 (4) (2013) 1183–1191.
7
References
[
47] S. Burge, G.N. Parkinson, P. Hazel, A.K. Todd, S. Neidle, Nucleic Acids Res. 34 (2006)
[1] J. Dai, T.S. Dexheimer, D. Chen, M. Carver, A. Ambrus, R.A. Jones, D. Yang, J. Am.
5402–5415.
Chem. Soc. 128 (2006) 1096–1098.
2] A.T. Phan, Y.S. Modi, D.J. Patel, J. Am. Chem. Soc. 126 (2004) 8710–8716.
3] D. Sun, K. Guo, J.J. Rusche, L.H. Hurley, Nucleic Acids Res. 33 (2005)
[48] C.A. Puckett, R.J. Ernst, J.K. Barton, Dalton Trans. 39 (2010) 1159–1170.
[49] M. Van Gurp, N. Festjens, G. van Loo, X. Saelens, P. Vandenabeele, Biochem. Biophys.
Res. Commun. 304 (2003) 487–497.
[
[
6
070–6080.
4] R. De Armond, S. Wood, D. Sun, L.H. Hurley, S.W. Ebbinghaus, Biochemistry 44
2005) 16341–16350.
[50] M. Juneja, U. Vanam, S. Paranthaman, A. Bharathan, V.S. Keerthi, J.K. Reena, R.
Rajaram, K.N. Rajasekharan, D. Karunagaran, Eur. J. Med. Chem. 63 (2013) 474–483.
[
(