← Back
Design, synthesis and in vitro bioactivity of mixed ligand Ru(II) complexes bearing the fluoroquinolone antibacterial agents
Transition Metal Chemistry
https://doi.org/10.1007/s11243-019-00341-3
Design, synthesis and in vitro bioactivity of mixed ligand Ru(II)
complexes bearing the fluoroquinolone antibacterial agents
Ramadevi Pulipaka1 · Soumya R. Dash1 · Priyanka Khanvilkar1 · Sarmita S. Jana2 · Ranjitsinh V. Devkar2 ·
Debjani Chakraborty1
Received: 7 April 2019 / Accepted: 29 June 2019
© Springer Nature Switzerland AG 2019
Abstract
Mixed ligand Ru(II) phenanthroline complexes of the type [Ru(1,10-phen) Flq]ClO (RPFlq-1-3) and “piano-stool”-type
2 4
Ru(II) arene complexes [Ru(η6-p-cymene)Cl(Flq)] (RAFlq-1-3), where Flq = fluoroquinolone, have been synthesized, char-
acterized and studied for their anticancer potential. DFT calculations were in line with the proposed structures, wherein
the fluoroquinolones are coordinated to the metal through the ring carbonyl and one of the carboxylic oxygen atoms in a
bidentate fashion. Binding efficacies of the synthesized complexes with bovine serum albumin (BSA) and CT-DNA were
studied spectroscopically, and it has been established that the arene complexes, though have moderate binding propensities
to CT-DNA (K = 0.8–1.7 × 103 M−1), have 1 02–103-fold better binding efficacies toward BSA (K = 3.2 × 105–2.1 × 106 M−1)
b a
due to the presence of the hydrophobic arene moiety. These results further prompted a study in their in vitro cytotoxicity
assay on A-549 non-small cell lung cancer and MCF7 breast cancer cell lines. Furthermore, gene expression studies on BAX
and BCL-2 genes and FACS analysis confirmed apoptosis as the mode of cell death.
Introduction 1. Ligand exchange kinetics Ruthenium(II) and
ruthenium(III) complexes have ligand exchange kinet-
The chemotherapeutic success of platinum-based antican- ics similar to those of platinum(II) complexes making
cer drug cisplatin is limited due to its inefficiency against them a possible alternative that display similar biologi-
many common types of cancer. The underlying problems cal effects to platinum(II) drugs. Ligand exchange is an
are poor solubility, drug resistance of cancerous cells, and important determinant of biological activity, and very
toxic side effects like nausea, neurotoxicity and kidney dam- few metal drugs reach the biological target without
age. Ruthenium complexes have been found to be promising being modified. Most metallodrugs undergo interactions
alternatives in overcoming cisplatin resistance with a low with macromolecules such as DNA, proteins or even
general toxicity and increased selectivity. The three major with water [1, 2].
properties of ruthenium that are well suited for pharmaco- 2. Physiologically accessible oxidation states Ruthenium
logical applications are: can possess variable oxidation states (II, III and IV)
under physiologically relevant conditions. These oxi-
dation states are easily interconvertible inside the cell,
due to its low energy barrier for the same. “Activation
by reduction” is a theory based on the fact that, due
Electronic supplementary material The online version of this to higher effective nuclear charge, Ru(III) complexes
article (https ://doi.org/10.1007/s1124 3-019-00341 -3) contains are more inert than Ru(II) complexes. Cancerous cells,
supplementary material, which is available to authorized users.
owing to their higher metabolic rate and remoteness
* from the blood supply, provide a chemically reducing
Debjani Chakraborty
debchak23@gmail.com environment than the healthy cells. Under biological cir-
cumstances of low oxygen concentration, acidic pH and
1 Department of Chemistry, The Maharaja Sayajirao high levels of glutathione, the Ru(II/III) redox poten-
University of Baroda, Vadodara, India
tial can be altered, and thus, Ru(III) complexes can be
2 Department of Zoology, The Maharaja Sayajirao University readily reduced to Ru(II) complexes. Thus, an admin-
of Baroda, Vadodara, India
Vol.:(0112 33456789)
Transition Metal Chemistry
istration of ruthenium in relatively inert III oxidation are the cause of their extensive use in clinical research [11].
state, caused minimal damage to the healthy cells, but is Fluoroquinolones, in addition to their antibacterial activ-
reduced to the active II oxidation state in vivo in cancer ity, were also shown to exhibit tumor-inhibiting properties
cells leading to target-based cell death [3]. and are suitable as ligands, featuring an O,O-chelate motif
3. Iron mimic Ability of ruthenium to mimic iron in bind- [12–15].
ing to many biological molecules is believed to be con- Since the approach to attach a bioactive ligand to a Ru(II)
tributing factor toward its low toxicity and high target center has been previously successfully used [16, 17] and
selectivity. Ruthenium complexes are capable of bind- keeping in mind the various biological properties of fluo-
ing to transferrin as effectively as iron can. Tumor cells roquinolones, we have prepared ruthenium complexes with
usually over-express transferrin receptors compared to lemofloxacin (Flq-1), levofloxacin (Flq-2) and ciprofloxacin
normal tissue because of their higher iron requirement (Flq-3). Herein, we describe an extended study comprising
which provides tumor cells to be selectively treated the synthesis and characterization of [Ru(phen) (Flq1-3)]
2
with the ruthenium-based drug. A transferrin-mediated ClO (RPFlq-1-3) and [Ru(p-cym)(Flq1-3)Cl] complexes
4
uptake by cancer cells has been observed for ruthenium (RAFlq-1-3) of the fluoroquinolones and their reactivity
complexes [4]. toward the DNA and the serum transport protein bovine
serum albumin (BSA) as well as their anticancer activity on
Apart from the pharmacological properties, the radio- human cancer cell lines.
physical properties of 97Ru can be applied to radiodiagnostic
imaging [5, 6]. With the development of new technology,
Experimental
such as photodynamic therapy (PDT), Ru(II) complexes can
be photophysical and bioactive leading to the design of light-
Reagents and materials
activated biocatalysts, improving the efficacy and selectivity
of Ru(II) complexes as anticancer drugs, as well as allow-
ing for the elucidation of their mechanism of action. The Analytical grade chemicals and solvents were used for the
Ru(II)–polypyridyl compound, TLD-1433, recently entered synthesis and characterization of the complexes. Fluoro-
phase IB clinical trials as a PDT agent in patients with blad- quinolone ligands Flq-1-3 were procured as gift samples
der cancer in 2015 [7]. from Alembic pharmaceuticals (Vadodara, Gujarat, India)
Organometallic ruthenium complexes bearing a π-bonded with 99% HPLC purity. The precursors [Ru(phen) Cl ]
2 2
arene ligand and other N,N, N,O and O,O chelating ligands and [Ru(η6-p-cymene)Cl ] were prepared according to
2 2
have attracted attention as promising anticancer agents the procedure cited in the literature [18–20]. α-terpinene,
[8, 9]. Arenes are known to stabilize 2+ oxidation state 1,10-phenathroline and sodium perchlorate were purchased
of ruthenium; hence, research has been done to study the from Sigma-Aldrich, Qualigens and Acros Organics (USA),
potential of Ru(II) arene complexes as anticancer agents. respectively. R uCl .3H O and BSA (bovine serum albumin)
3 2
Also, good aqueous solubility and the inertness of the arene were purchased from Hi-media. CT-DNA, trisodium citrate
ligand toward displacement under physiological conditions and EB (ethidium bromide) were purchased from SRL (Sisco
provide an advantage for medicinal use of “half-sandwich” Research Laboratory, Mumbai, India). A549 and MCF7 cell
Ru(II) mono-arene complexes [10]. The stability of these lines were procured from NCCS, Pune, Maharashtra, India.
complexes is due to the electron donation of arene to the
Instrumentation
empty ruthenium 4d orbitals and back-donation of filled 4d
electron into the vacant arene orbital. This factor is influ-
enced by the presence of a strong donor or acceptor ligand. 1H NMR spectra were recorded on Bruker 400 MHz NMR
Fluoroquinolones are a family of synthetic antibiotics Spectrophotometer. ESI mass spectra of the complexes were
with potent bactericidal activities and broad spectrum activ- recorded on Applied Biosystem API 2000 Mass spectrom-
ity against many clinically important pathogens which are eter. C, H and N elements of the complexes were estimated
responsible for a variety of infections. The mechanism of using Thermo Scientific Flash 2000 elemental analyzer.
action of quinolones is through the inhibition of bacterial Infrared spectra (400–4000 cm−1) were recorded on Perki-
gyrase, an enzyme involved in DNA replication, recombi- nElmer R0X-1 FTIR with samples prepared as KBr pellets.
nation and repair, thus stopping cell growth. The affinity of The molar conductance of the complexes was measured in
fluoroquinolones to the metal ions seems to be crucial to DMSO at 1 0−3 M concentration using Toshniwal conductiv-
their contribution to the clinical field. Though certain disad- ity bridge-type CLOI/O1A with a dip-type conductivity cell.
vantages and limitations are there, properties such as excel- UV–visible spectra for DNA binding studies were
lent oral absorption, prolonged half-lives, efficacy, excellent recorded in DMSO solutions at concentrations around
tissue penetration and significant entry into phagocytic cells 1 0−3 M on PerkinElmer Lambda-35 dual-beam UV–Vis
1 3
Transition Metal Chemistry
spectrophotometer. Fluorescence spectra for competitive formula C H ClFN O Ru; anal. found: C, 54.05; H, 3.59;
42 36 7 8
DNA binding and BSA binding studies were recorded in N, 10.18. Calc.: C, 54.69; H, 3.93; N, 10.63. ESI–MS m/z:
solution on JASCO FP-6300 fluorescence spectrophotometer. 822.3 ( M+-ClO −), 98.9 (ClO −); FTIR (KBr, ν/cm−1):
4 4
OriginPro 8 software was used to analyze the data ν 1724, ν 1623, ν 1337, Δν
(pyridone)C=O COOassym COOsym COO
obtained from DNA/BSA complexes interaction (titration) 286, ν 650.
Cl–O
experiments and for plotting. ORCA program package (ver-
sion 4.0.1.2) was used for geometry optimization. [Ru(phen) 2 (Flq‑3)]ClO 4 (RPFlq‑3) [Ru(phen) 2 Cl 2 ] 2 H 2 O
(0.0528 mmol, 30.0 mg) and Flq-3 (0.0528 mmol, 17.0 mg)
General synthetic procedure of [Ru(phen) (Flq1‑3)]
2 were used to synthesize RPFlq-3. Solubility: DMSO, DMF.
ClO complexes—(RPFlq‑1‑3)
4 Yield 62.14%; molecular weight 892.27 g/mol; molecular
formula C H ClFN O Ru; anal. found: C, 54.57; H, 3.65;
41 34 7 7
[Ru(phen) Cl ] 2 H O and the fluoroquinolone ligands Flq- N, 10.58. Calc.: C, 55.19; H, 3.84; N, 10.99. ESI–MS m/z:
2 2 2
1-3 in 1:1 mol ratio were mixed and refluxed in 3 ml of 792.3 ( M+-ClO −), 98.9 (ClO −); FTIR (KBr, ν/cm−1):
4 4
ethanol/water (2:1) mixture for 7 h to yield a clear red solu- ν 1717, ν 1627, ν 1340, Δν
(pyridone)C=O COOassym COOsym COO
tion. After cooling, a saturated aqueous solution of NaClO 287, ν 668.
4 Cl-O
was added drop-wise and stirred for 2 h at r.t. The reaction
General synthetic procedure of [Ru(p‑cym)(Flq1‑3)
mixture was sealed under nitrogen and cooled at 0 °C for
Cl] complexes—(RAFlq‑1‑3)
overnight. On addition of water, immediate reddish brown
precipitates were obtained which were filtered, washed with
water and diethyl ether and dried. Figure 1 shows the general [Ru(p-cym)(Flq1-3)Cl] (RAFlq-1-3) were prepared by a
synthetic route of complexes RPFlq-1-3. typical µ-chlorido-bridge splitting reaction of [Ru(η6-p-
cymene)Cl ] . To a solution of [Ru(η6-p-cymene)Cl ] (in
2 2 2 2
[Ru(phen) 2 (Flq‑1)]ClO 4 (RPFlq‑1) [Ru(phen) 2 Cl 2 ] 2 H 2 O 2.5 ml C H 2 Cl 2 ), a solution of the ligand Flq-1-3 in 2.5 ml
(0.0528 mmol, 30.0 mg) and Flq-1 (0.0528 mmol, 18.5 mg) methanol was added with stirring in 1:2 ratio, respectively.
were used to synthesize RPFlq-1. Solubility: DMSO, DMF. The reaction mixture was left on stirring overnight (20–24 h)
Yield 54.29%; molecular weight 914.30 g/mol; molecular at room temperature and then left for slow evaporation at
formula C H ClF N O Ru; anal. found: C, 52.01; H, 3.93; r.t. The reddish brown crystalline precipitate was filtered,
41 37 2 7 7
N, 10.23. Calc.: C, 52.86; H, 4.08; N, 10.42. ESI–MS m/z: washed with pet ether and C H Cl and dried in air. The com-
2 2
814.3 ( M+-ClO −), 98.9 (ClO −); FTIR (KBr, ν/cm−1): plexes so obtained were recrystallized from dichloromethane
4 4
ν 1721, ν 1617, ν 1329, Δν and ether which resulted in reddish brown crystals but not
(pyridone)C=O COOassym COOsym COO
288, ν 651. of the single-crystal quality. Figure 2 shows the general syn-
Cl–O
thetic route for the preparation of RAFlq-1-3 complexes.
[Ru(phen) 2 (Flq‑2)]ClO 4 (RPFlq‑2) [Ru(phen) 2 Cl 2 ] 2 H 2 O
(0.0528 mmol, 30.0 mg) and Flq-2 (0.0528 mmol, 19.0 mg) [Ru(η6‑p‑cym)(Flq‑1)Cl] (RAFlq‑1) RAFlq-1 was synthe-
were used to synthesize RPFlq-2. Solubility: DMSO, DMF. sized by reaction of [Ru(η6-p-cymene)Cl ] (0.049 mmol,
2 2
Yield 66.78%; molecular weight 922.30 g/mol; molecular 30.0 mg) and ligand Flq-1 (0.098 mmol, 34.6 mg). Soluble
Fig. 1 General synthetic route to complexes RPFlq-1-3
1 3
Transition Metal Chemistry
Fig. 2 General synthetic route to complexes RAFlq-1-3
in almost all organic solvents like DMSO, MeOH and Geometry optimization
C H Cl . Yield 56.1%; molecular weight 624.10 g/mole;
2 2
molecular formula C H ClF N O Ru; anal. found: C, All the density functional theory (DFT) calculations were
27 35 2 3 3
51.25; H, 5.32; N, 6.68. Calc.: C, 51.96; H, 5.65; N, 6.73. carried out in ORCA program package (version 4.0.1.2)
ESI–MS m/z: 588.2 (M+-Cl); δ (400 MHz, DMSO-d ) [21]. All the geometries were optimized using BP86 func-
H 6
5.82–5.77 (dd, 4H, p-cym Ar–H), 2.85–2.79 (m, 1H, p-cym- tional [22–24] along with Ahlrichs’ split-valence double-ξ
iso-prop-CH), 2.08 (s, 3H, p-cym Ar-CH ), 1.22 (d, 6H, basis set def2-SV(P) [25] and def2-J [26] auxiliary basis set,
3
p-cym-iso-prop-(CH ) ); FTIR (KBr/cm−1): ν 2960, except for Ru atom for which def2-ECP [27] was employed.
3 2 (Ar)C-H
ν 1721, ν 1623, ν 1391, Δν The resolution of identity (RI) approximation [28] was used
(pyridone)C=O COOassym COOsym COO
232. to reduce the computational time. To consider the solvent
effect of methanol, a conductor-like polarizable continuum
[Ru(η6‑p‑cym)(Flq‑2)Cl] (RAFlq‑2) Reaction of [Ru(η6- model (CPCM) [29] was also employed. All the models were
p-cymene)Cl ] (0.049 mmol, 30.0 mg) and ligand Flq- visualized using Avogadro software [30].
2 2
2 (0.098 mmol, 35.4 mg) yielded RAFlq-2. Soluble in
organic solvents like DMSO, MeOH and CH Cl . Yield DNA binding experiments
2 2
63.8%; molecular weight 632.11 g/mole; molecular for-
mula C H ClFN O Ru; anal. found: C, 52.74; H, 5.37; UV absorption studies
28 34 3 4
N, 6.59. Calc.: C, 53.20; H, 5.42; N, 6.65. ESI–MS m/z:
596.2 ( M+-Cl); δ (400 MHz, DMSO-d ) 5.82–5.76 (dd, UV–Vis spectroscopy was used to investigate the nature of
H 6
4H, p-cym Ar–H), 2.82–2.77 (m, 1H, p-cym-iso-prop-CH), interaction of the synthesized complexes with CT-DNA and
2.08 (s, 3H, p-cym Ar-CH ), 1.19 (d, 6H, p-cym-iso-prop- to calculate the binding constant (K ). Experiments were
3 b
(CH ) ); FTIR (KBr/cm−1): ν 2959, ν 1713, performed with constant complex concentration and varying
3 2 (Ar)C-H (pyridone)C=O
ν 1623, ν 1337, Δν 286. DNA concentration within. Stock solutions of the complexes
COOassym COOsym COO
were diluted with tris buffer to get the final concentration
[Ru(η6‑p‑cym)(Flq‑3)Cl] (RAFlq‑3) RAFlq-3 was synthe- (30 μM). While measuring the absorption, equal increments
sized by reaction of [Ru(η6-p-cymene)Cl ] (0.049 mmol, of CT-DNA were added at different ratios to both the com-
2 2
30.0 mg) and ligand Flq-3 (0.098 mmol, 32.4 mg). Soluble in pound solution and the reference solution to eliminate the
DMSO, MeOH and CH Cl . Yield 59.4%; molecular weight absorbance of CT-DNA itself.
2 2
602.08 g/mole; molecular formula C H ClFN O Ru; anal.
27 32 3 3
found: C, 52.99; H, 5.21; N, 6.73. Calc.: C, 53.86; H, 5.36; Competitive binding studies with EB using fluorescence
N, 6.98. ESI–MS m/z: 567.2 ( M+-Cl); δ (400 MHz, DMSO- spectroscopy
H
d ) 5.82–5.76 (dd, 4H, p-cym Ar–H), 2.83–2.77 (m, 1H,
6
p-cym-iso-prop-CH), 2.08 (s, 3H, p-cym Ar-CH ), 1.19 (d, Fluorescence spectroscopy has been employed to study the
3
6H, p-cym-iso-prop-(CH ) ); FTIR (KBr/cm−1): ν competitive binding efficacy of the complexes with CT-
3 2 (Ar)C-H
2959, ν 1724, ν 1629, ν 1337, DNA, i.e., to examine whether the complexes can displace
(pyridone)C=O COOassym COOsym
Δν 292. EB from DNA–EB complex. The DNA–EB complex was
COO
1 3
Transition Metal Chemistry
prepared by adding known concentration of EB (66.6 μM) incubated at 37 °C in a 5% CO incubator for 48 h. Upon
2
and DNA (169.6 μM) in tris buffer. The solution of each completion of the incubation period, MTT dye solution
complex was added step by step into the solution of the (prepared using serum-free culture medium) was added
DNA–EB complex. The effect of addition of the complex to each well to a final concentration of 0.5 mg/ml. After
was obtained by recording the variation in the fluorescence 4 h of incubation with MTT, the culture media were dis-
emission spectra of the DNA–EB complex measured at carded and the wells were washed with phosphate buffer
609 nm (λ = 524 nm). saline (Hi-Media, India Pvt., Ltd.), followed by addition
ex
of DMSO to dissolve the formazan crystals so formed and
Viscosity measurements subsequent incubation for 30 min. The optical density of
each well was measured spectrophotometrically at 563 nm
Cannon–Ubbelohde viscometer was used to measure the using Biotek-ELX800MS universal ELISA reader (BioTek
relative viscosity of DNA. The viscosity measurements instruments, Inc., Winooski, VT). The IC values were
50
of DNA (200 µM) solutions were carried out at a constant determined by plotting the percentage viability versus
temperature of 32.0 ± 0.1 °C in the presence of complexes concentration on a logarithmic graph and reading off the
RAFlq-1 and RPFlq-1 (representative from each series) concentration at which 50% of cells remained viable rela-
at [complex]/[DNA] ratio of 0, 0.04, 0.08, 0.12, 0.16 and tive to the control. Each experiment was repeated at least
0.20 in Tris–HCl buffer (pH 7.2). Digital stopwatch with three times to obtain mean values.
least count of 0.01 s. was used for flow time measurement
with accuracy of ± 0.1 s. The flow time of each sample was
Gene expression studies
measured three times, and an average flow time was cal-
culated. Data are presented as (η/η )1/3 versus [complex]/
0
[DNA], where η is the viscosity of DNA in the presence of Real-time PCR (RT-PCR) with GAPDH as a control was
complex and η is the viscosity of DNA alone. Viscosity used to study the expression of apoptosis-related genes,
0
values were calculated from the observed flow time of DNA- Bax and Bcl-2. The studies were carried out on A549 and
containing solutions (t) corrected for that of the buffer alone MCF7 cell lines where the IC values of the complexes
50
(t ), η = (t − t )/t [31]. from MTT assay were taken as dosage for the treatment.
0 0 0
Total RNA was isolated using TRIzol reagent (Invitrogen,
BSA binding experiments
California, USA). cDNA was synthesized by reverse tran-
scription of 1 μg of total RNA using iScript cDNA Syn-
Steady-state fluorescence spectroscopy was used to study thesis Kit (BIORAD, California, USA). PCR was carried
the protein-binding capability of the complexes under study. out according to manufacturer’s instructions using SYBR
Tryptophan fluorescence quenching experiments were car- Green Master Mix kit (Invitrogen, California, USA).
ried out using bovine serum albumin (BSA, 16.6 μM) in Cycler conditions were as follows: initial denaturation at
buffer (containing 15 mM trisodium citrate and 150 mM 95 °C for 3 min followed by 35 reaction cycles (30 s at
NaCl at pH 7.0). The influence of addition of increasing 94 °C, 30 s at 55 °C and 30 s at 72 °C) and final cycle at
concentrations of the complexes was monitored by recording 72 °C for 10 min. Statistical analysis was performed by
the reduction in the emission intensity (quenching) of the using one-way ANOVA. Primers used for this study are
tryptophan residue of BSA at 343 nm [32]. Fluorescence listed in ESM 1.
spectra were recorded from 300 to 500 nm at an excitation
wavelength of 296 nm.
FACS analysis for detection of apoptosis
Cytotoxicity
Annexin V FITC/PI dual staining assay was used, to quantify
Standard 3-(4,5-dimethylthiazole)-2,5-diphenyltetraazo- apoptosis, according to the manufacturer’s protocol (Invitro-
lium bromide (MTT) assay was used [33]. A549 (to be gen, California, USA). A549 and MCF7 cells were treated
treated with RPFlq-1-3) and MCF7 cells (to be treated with the complexes at their IC values for 48 h. The cells
50
with RAFlq-1-3) (5.0 × 103 cells/well) were placed in from each well were then centrifuged, washed with PBS
two separate 96-well culture plates (Tarsons India Pvt., and suspended in 100 µl buffer. Five microliters of annexin
Ltd.) and grown overnight at 37 °C in a 5% CO incuba- V FITC conjugate and 10 µl of propidium iodide solution
2
tor. Compounds to be tested were then added to the wells were added to each cell suspension and incubated for 10 min
to achieve final concentrations ranging from 10 to 500 μg/ at room temperature in dark. The samples were analyzed
ml. Control wells were prepared by the addition of cul- on flow cytometer (MoFlo™ Cytomation, Modular Flow
ture medium without the compounds. The plates were then Cytometer) using CellQuest software.
1 3
Transition Metal Chemistry
Results and discussion 270–290 nm due to the intraligand n → π* transition which
have blueshifted on complexation with the metal ion as com-
Characterization pared to the free ligand. The third broadband observed within
325–335 nm, also due to an intraligand n → π* transition,
The stretching bands at 1680–1700 cm−1 and ν O–H remains fairly unchanged as seen in the free ligand. The λ max
at 3420–3462 cm−1 attributable to the free –COOH values of all the transitions taking place in the complexes are
ν group in the fluoroquinolone ligands were not tabulated in Table 1.
C=O carboxylate
observed in the spectra of complexes. ν asymmetric All the three complexes RPFlq-1-3 have molar
O–C–O
and symmetric stretching vibrations were obtained as two conductances (1 × 10−3 M in DMSO) in the range of
strong characteristic bands in the range 1617–1630 and 25–28 Ω−1 cm2 mol−1 (Table 1), suggesting 1:1 electrolytic
1330–1400 cm−1, respectively. The separation frequency behavior.
Δν = ν – ν values were found to be in the range The ESI–MS spectra of RPFlq-1-3 complexes show
COO asym COOsym
232–292 cm−1 [34, 35], suggesting a monodentate coordina- peaks at m/z values equivalent to (M+-ClO 4 −) in the pos-
tion mode of the carboxylato group of the fluoroquinolone itive-ion spectra, while m/z peaks corresponding to the
ligand. The band at ~ 1685 cm−1 of free C = O stretching perchlorate anion (ClO 4 −) in the negative-ion spectra. The
of the pyridone ring in the fluoroquinolones has shifted to spectra of RAFlq-1-3 show m/z peaks corresponding to
1700–1725 cm−1 in the complexes, suggesting binding of ( M+-Cl) values. The m/z values of all the complexes are in
the ligand to the metal center through the pyridone carbonyl well agreement with the proposed compositions. The m/z
oxygen atom. values along with their fragments are provided in Table 2,
The UV–Vis spectra of the complexes RPFlq-1-3
(Fig. 3a) show bands in the region 250–300 nm correspond-
Table 1 UV–Vis peak assignments and conductance measurements
ing to the intraligand π → π* transitions, whereas bands
of complexes
obtained as low-intensity shoulders in the region 360–380 nm
are due to the ligand-centered n → π* transitions. In addition, Com- Intraligand MLCT d–d tran- Λ m
pound transitions (nm) dπ–π* sitions (Ω−1 cm2 mol−1)
metal-centered dπ → π* MLCT and d–d transitions are also
transitions (nm)
π–π* n–π*
observed in the region 480–500 nm and 700–705 nm, respec- (nm)
tively. In the case of RAFlq-1-3 complexes, three major
RPFlq-1 282 360 491 704 25.6
bands in the wavelength range 200–400 nm are obtained
RPFlq-2 279 374 485 704 28.1
(Fig. 3b). The first band within 200–250 nm corresponds to
RPFlq-3 275 359 493 700 26.4
an intraligand π → π* transition owing to the aromatic rings
RAFlq-1 224 281, 328 – – –
present in the arene ligand (p-cymene) as well as the fluo-
RAFlq-2 225 288, 334 – – –
roquinolone ligand. A medium-intensity band is observed at
RAFlq-3 224 271, 329 – – –
Fig. 3 UV–Vis spectra of complexes a RPFlq-1-3 and b RAFlq-1-3
1 3
Transition Metal Chemistry
Table 2 m/z values of complexes showing fragmentation Any interaction between the compounds and DNA may per-
turb the ligand-centered transitions of the compounds giv-
Compound Calculated Positive-ion peaks Negative-
mass (g/mol) ion peaks ing rise to spectral changes that may give evidence of the
[M+–ClO] [M+–Cl] ClO− existing interaction mode [38]. The UV spectra of all the
4 4
complexes ( 10−6 M) were recorded in the absence and pres-
RPFlq-1 914.30 814.3 – 98.9
ence of varying CT-DNA concentration (at a ratio of 0–6.0)
RPFlq-2 922.30 822.3 – 98.9
within. The spectra of the complexes in the presence of CT-
RPFlq-3 892.27 792.3 – 98.9
DNA showed significant hypochromism (Fig. 5), speculative
RAFlq-1 624.10 – 588.2 –
of intercalative mode of binding of the compounds to DNA.
RAFlq-2 632.11 – 596.2 –
The magnitude of binding was determined in terms
RAFlq-3 602.08 – 567.2 – of binding constant (K ) obtained from the Mehan’s
b
equation (Fig. 5, inset) [39]. The K values (Table 4)
b
for the complexes RPFlq-1-3 are in the range of
and the mass spectra of all the complexes are provided in 6.1 × 103–1.0 × 104 M−1, and those of RAFlq-1-3 are in the
supplementary material (ESM 2). Furthermore, the com- range of 0.8 × 103–1.7 × 103 M−1. It is apparent from the
position and purity of the complexes have been confirmed K values that the planar phenanthroline rings attached to
b
by their C, H, N elemental analysis. the metal center in RPFlq-1-3 complexes make them bet-
The 1H NMR spectra of RAFlq-1-3 (provided as sup- ter DNA binders (intercalators) compared to their arene
plementary material ESM 3) showed 6-proton doublet at counterparts (RAFlq-1-3 complexes). The titration plots of
δ = 1.19 ppm due to the two methyl branches of isopropyl remaining complexes are provided in supplementary mate-
group [CH(CH ) ], 3-proton singlet at δ = 2.08 ppm owing rial ESM 4.
3 2
to the methyl group para to the isopropyl group, 1-proton
multiplet at δ = 2.85 ppm attributed to –CH of the isopro- Competitive binding studies with ethidium bromide using
pyl group and 4-proton doublet of doublet at δ = 5.8 ppm fluorescence spectroscopy
assigned to the 4 Ar protons of p-cymene. These signals are
consistent with those obtained for unbound p-cymene, but The intercalative mode of binding of the complexes to the
with small shifts. The peak arising due to carboxylic O–H CT-DNA has been further confirmed using competitive bind-
proton in the free Flq ligands disappeared in the spectra of ing studies with ethidium bromide (EB). Ethidium bromide
the complexes, suggesting a coordination of the carboxy- is a typical indicator of intercalation, by emitting intense
late oxygen to the ruthenium center. The remaining peaks fluorescence when bound to CT-DNA [40, 41]. An inclu-
can be ascribed to the coordinated Flq ligand. The peak at sion of a second molecule with competitive binding ability
δ = 2.503 ppm, observed in all the three spectra, is the sol- to DNA may lead to the displacement of EB from DNA–EB
vent peak owing to DMSO-d used as solvent. The 1H NMR complex resulting in a decrease in the DNA-induced EB
6
results confirm the presence of p-cymene and Flq coordi- emission [42]. The emission spectra of DNA–EB complex
nated to the ruthenium metal center, and hence the sug- (λ = 546 nm, λ = 609 nm) in the absence and presence of
ex em
gested structure. Due to the low solubilities of the RPFlq-1- increasing amounts of complexes (at a ratio of 0–2.0) have
3 complexes, their NMR spectra could not be recorded with been recorded. A decrease in the intensity of the emission
good resolution and hence are not discussed here. band at 609 nm was observed, indicating the competition
In the absence of X-ray crystallography data, the DFT of the compounds with EB in binding to DNA (Fig. 6). The
calculations for geometry optimization provide a great observed quenching of DNA–EB fluorescence suggests that
insight into the molecular structure. Optimized structures they displace EB from the DNA–EB complex and interact
of the complexes are shown in Fig. 4. The metal–ligand with DNA by intercalation.
bond lengths, bond angles and the single-point energies The relative binding of complexes to CT-DNA was deter-
are tabulated in Table 3. The metal–ligand bond lengths mined by calculating the quenching constant (K ) from the
SV
and bond angles are in well agreement with the values in slopes of straight lines obtained from the Stern–Volmer
the literature [36, 37]. equation [43]. The Stern–Volmer quenching plots (Fig. 6,
inset) illustrate that the quenching of EB bound to DNA
DNA binding studies by complexes is in good agreement (R = 0.99) with the lin-
ear Stern–Volmer equation, and the Stern–Volmer quench-
Electronic absorption titration ing constant (K ) values are given in Table 4. Complexes
SV
RPFlq-1-3 show quenching constant values in the range
Investigation of a possible interaction of a drug with the of 3.9 × 103–2.6 × 104 M−1, whereas for RAFlq-1-3 it is
DNA can be done with the help of electronic spectroscopy. 4.1 × 103–4.3 × 103 M−1. This observation is in agreement
1 3
Transition Metal Chemistry
Fig. 4 Optimized structures of
the complexes under study
with the K b values, indicating better binding of RPFlq-1-3 complexes. The viscosity of DNA is sensitive to length
complexes compared to RAFlq-1-3 complexes. The titration changes and is regarded as the least ambiguous and the
plots of remaining complexes are provided in supplementary most critical clues of a DNA binding mode in solution [44].
material ESM 5. In general, intercalating agents are expected to elongate
the double helix to accommodate the ligands in between
Viscosity measurements the base pairs, leading to an increase in the viscosity of
DNA. In contrast, a complex that binds exclusively in the
In order to further confirm the modes of binding of com- DNA grooves typically causes less pronounced (positive or
plexes to CT-DNA, viscosity measurements of DNA solu- negative) or no changes in DNA solution viscosity [45, 46].
tions were carried out in the presence and absence of these The effects of complexes RAFlq-1, RPFlq-1 and classical
1 3
Transition Metal Chemistry
Table 3 Metal ligand bond
Bond lengths (in Å) RAFlq-1 RAFlq-2 RAFlq-3
lengths and bond angles of
complexes under study obtained Part A: Metal ligand bond lengths of complexes RAFlq-1-3 and RPFlq-1-3
from geometry optimization
Ru–O1 (COO) 2.10 2.08 2.07
Ru–O2 (C=O) 2.12 2.08 2.08
Ru–Cl 2.45 2.44 2.44
Ru-p-cym (π ring) 1.65 1.65 1.65
RPFlq-1 RPFlq-2 RPFlq-3
Ru–O1 (COO) 2.10 2.07 2.07
Ru–O2 (C=O) 2.09 2.06 2.06
Ru–N1 2.04 2.03 2.03
Ru–N2 2.06 2.06 2.06
Ru–N3 2.03 2.03 2.03
Ru–N4 2.06 2.05 2.05
Bond angles (in degrees) RAFlq-1 RAFlq-2 RAFlq-3
Part B: Metal ligand bond angles of complexes RAFlq-1-3 and RPFlq-1-3
Cl–Ru–O1 89.0 84.7 84.8
Cl–Ru–O2 89.4 84.5 84.9
O1–Ru–O2 87.7 89.8 89.8
RPFlq-1 RPFlq-2 RPFlq-3
N1–Ru–N2 80.6 80.6 80.6
N1–Ru–N3 94.9 94.8 94.8
N1–Ru–N4 95.7 97.9 97.9
N2–Ru–N3 96.6 97.7 97.7
N2–Ru–N4 175.2 177.7 177.8
N3–Ru–N4 80.6 80.7 80.7
N1–Ru–O1 169.9 173.4 173.3
N1–Ru–O2 89.8 88.2 88.1
N2–Ru–O1 89.4 93.0 92.9
N2–Ru–O2 85.4 88.1 88.0
N3–Ru–O1 87.8 87.8 87.8
N3–Ru–O2 175.1 173.8 173.9
N4–Ru–O1 94.4 88.5 88.6
N4–Ru–O2 97.7 93.6 93.7
O1–Ru–O2 87.7 89.8 89.8
The numbering of atoms in the first compound of each series is shown in Fig. 4 as a reference for viewing
the bond lengths and bond angles followed in the table
intercalator EB on the viscosities of CT-DNA solution are association or denaturation, leading to changes in the tryp-
shown in Fig. 7. With increasing [complex]/[DNA] ratios, tophan environment of BSA. By the addition of increasing
the relative viscosities of CT-DNA increased gradually, complex concentrations to a solution of the protein (at a ratio
suggesting a characteristic intercalative mode of binding of 0–2.0), a significant decrease in the fluorescence intensity
which is in accordance with the previous findings. of BSA (Fig. 8a, b) was observed as a result of binding inter-
action of the complexes under study to BSA. The new peak
BSA binding studies
observed in Fig. 8a is owing to the intrinsic fluorescence
of RAFlq-1-3 complexes that exhibits hyperchromism with
BSA solutions, owing to their tryptophan residues, exhibit a increased concentration of the compounds, and is not an
strong fluorescence emission peak at 343 nm on excitation at attribute of the binding interactions with BSA.
296 nm [32, 47]. A binding interaction of a drug molecule The Stern–Volmer quenching constant (K ) values
SV
with BSA may change the protein conformation, subunit obtained from the Stern–Volmer plot (Fig. 8c, d) for the
1 3
Transition Metal Chemistry
Fig. 5 UV absorption spectra of a RAFlq-1 and (b) RPFlq-1 at of [DNA]/(ε A − ε f ) versus [DNA] for a RAFlq-1 and b RPFlq-1. The
increasing concentrations of CT-DNA, the arrow shows decrease in slope-to-intercept ratio gave K b values
intensity upon increasing concentration of the CT-DNA; Inset: plot
Table 4 K b and K SV values of complexes complexes interacting with BSA (Table 5) suggest good
binding propensity of the complexes with the serum pro-
Compound K b (M−1) K SV (M−1) tein. The association binding constant (K ) and the number
a
RPFlq-1 6.1 × 103 2.6 × 104 of binding sites per albumin (n) can be calculated using
RPFlq-2 1.0 × 104 3.9 × 103 double-logarithmic equation [48]. The double-logarithmic
RPFlq-3 7.5 × 103 2.0 × 104 plot for all the complexes was found to be linear (Fig. 8e,
RAFlq-1 0.8 × 103 4.2 × 103 f), and the K and n have been obtained from the intercept
a
RAFlq-2 1.8 × 103 4.3 × 103 and slope, respectively, which are tabulated in Table 5. The
RAFlq-3 1.7 × 103 4.1 × 103 association binding constant so calculated indicates that the
Fig. 6 Plot of fluorescence emission intensity I versus wavelength λ on increasing concentration of the complex. Inset: Stern–Volmer
for CT-DNA–EB complex at different concentrations of a RAFlq-1 quenching plot of DNA–EB emission in the presence of a RAFlq-1
and b RPFlq-1, the arrow shows decrease in fluorescence intensity and b RPFlq-1. The slope gave K SV values
1 3
Transition Metal Chemistry
values of the complexes were consistent with their IC val-
50
ues, which is indicative of the fact that proteins may be the
cellular targets.
Effect of the Ru(II) complexes on gene expression
MCF7 and A549 cancer cells were exposed to the complexes
RAFlq-2 and RPFlq-2, respectively, for 48 h to assess the
expression levels of the pro-apoptotic (Bax) and anti-apop-
totic (Bcl-2) genes in the presence of the two complexes. It
was observed (Fig. 10) that expression levels of Bax signifi-
cantly increased in both the cancer cell lines after treatment
with the respective complex, suggesting that they possibly
trigger apoptosis of the cancer cells. Alternatively, expres-
sion levels of Bcl-2 in both the cancer cell lines treated with
the complexes were low, further indicating higher vulner-
Fig. 7 Effect of increasing amounts of the complexes RAFlq-1,
RPFlq-1and ethidium bromide (EB) on the relative viscosity of ability for trigger of apoptosis.
CT-DNA (200 µM) in Tris–HCl buffer at 32 (± 0.1) °C. [Complex]/
[DNA] = 0, 0.04, 0.08, 0.12, 0.16 and 0.20
Effect of the Ru(II) complexes on apoptosis (FACS
analysis)
arenes RAFlq-1-3 interact with the protein tenfold better
than the phenanthroline complexes RPFlq-1-3. In general,
the K and K values suggest good binding propensities of Flow cytometry was used to carry out apoptosis study to
a SV
the complexes to BSA. Moreover, the linearity of the dou- differentiate the cell types like viable cells, pro-apoptotic,
ble-logarithmic plots reveals that only one of the tryptophan early apoptotic cells and late apoptotic cells in the presence
residues on BSA protein is interacting with the compounds of the complexes. The measurement of annexin V–propid-
[49]. Also, the availability of only one binding site on the ium iodide dual stain as an indicator for apoptosis has been
protein is indicated by the n values of the complexes which taken along with a dye exclusion test to establish the integ-
average out to be 1. The titration plots of remaining com- rity of the cell membrane. The A549 cells were incubated
plexes are provided in supplementary material ESM 6. with RPFlq-2 and MCF7 cells with RAFlq-2 complexes
(showing the lowest IC values in their respective series)
50
Cytotoxicity
for 48 h at their IC concentrations followed by staining
50
with annexin V and PI. The results are shown in Fig. 11. It
In vitro cytotoxicity of the complexes RPFlq-1-3 and was observed that the viable cells (lower left quadrant) did
RAFlq-1-3 was evaluated on A-549 human lung cancer and not bind to either annexin V or PI, the early apoptotic cells
MCF7 human breast cancer cell line, respectively, by expos- (lower right quadrant) bound to annexin V but did not bind
ing the cells to the said complexes for 48 h at different con- PI, and the late apoptotic cells (upper right quadrant) were
centrations (dosages). The percentage of live cells was found positive for both annexin V and PI. Dead cells were observed
to decrease with increasing concentration of the complexes in the upper left quadrant. 3.3% apoptotic A549 cells and
(Fig. 9). The IC values given in Table 6 are much less than 2.7% apoptotic MCF7 cells were found within the control
50
that of NAMI-A, a Ru(II) complex under phase II clinical cells, whereas the cells treated with RPFlq-2 and RAFlq-
trial having IC in the range of 550–750 µM for various 2 contained 19.3% and 12.8% apoptotic cells, respectively.
50
cancer cell lines on treatment for 48 h [50, 51] and RAPTA, a These results reveal that the ruthenium complexes investi-
well-studied ruthenium–arene complex with IC50 > 1600 µM gated herein can induce apoptosis of the cancer cells.
for MCF7 cell line on treatment for 72 h [52]. However, they
were found to be much less active than cisplatin.
Conclusion
The in vitro anticancer activity of the compounds was
found to be inconsistent with their DNA binding abilities.
The different order of DNA binding affinity and cytotoxicity Ru(II) complexes of 1,10-phenanthroline and p-cymene con-
means various other targets and mechanisms are involved. taining the fluoroquinolone antibacterial agents, lemofloxacin,
DNA binding and cleavage need not be the only pathway for levofloxacin and ciprofloxacin, were synthesized and charac-
their anticancer activity. However, the BSA binding constant terized using various spectral techniques. Spectral and DFT
1 3
Transition Metal Chemistry
1 3
Transition Metal Chemistry
◂Fig. 8 Plot of fluorescence emission intensity versus wavelength for Table 6 IC
50
values of complexes obtained from MTT assay
BSA at increasing concentrations of a RAFlq-1, b RPFlq-1, the
arrow shows decrease in the fluorescence intensity with increasing Compound Cell line IC 50 (µM)
concentration of the complex; Stern–Volmer plot for the quenching of
RPFlq-1 A549 394
BSA fluorescence by c RAFlq-1-3, d RPFlq-1-3.; double-logarith-
mic plot for the quenching of BSA fluorescence by e RAFlq-1-3, f RPFlq-2 A549 71.5
RPFlq-1-3 to determine the binding constant K a and number of bind- RPFlq-3 A549 308
ing sites n
RAFlq-1 MCF7 209
RAFlq-2 MCF7 185
studies reveal coordination of the fluoroquinolones to the RAFlq-3 MCF7 222
metal center via the carboxylate and carbonyl oxygen. The RAPTA MCF7 > 1600
complexes have good binding propensities toward the DNA NAMI MCF7 750
and BSA as revealed by the binding studies. The RPFlq-1-3
complexes were found to be better DNA binders owing to
the planar phenanthroline rings attached to the metal center
capable of intercalating within the DNA base pairs, whereas
the RAFlq-1-3 complexes were found to have better binding
affinities toward BSA due to their hydrophobic nature owing
to the arene moiety (p-cymene). The complexes were found
to be more cytotoxic toward A549 and MCF7 cancer cells as
compared to NAMI-A and RAPTA. The gene expression stud-
ies and FACS analysis provided valuable evidence on the role
of the complexes in triggering apoptosis in cancer cells and
revealed the merits of the Ru(II) metal complexes containing
fluoroquinolones as potent anticancer agents.
Table 5 K a, K SV and n values for complexes
Compound K a (M−1) K SV (M−1) n
RPFlq-1 9.1 × 104 7.4 × 104 1.2
RPFlq-2 6.8 × 105 9.0 × 104 1.2
RPFlq-3 4.2 × 104 6.1 × 104 1.2 Fig. 10 Expression levels of pro-apoptotic gene (Bax) and anti-
RAFlq-1 3.2 × 105 4.9 × 104 1.2 apoptotic gene (Bcl-2) were studied using quantitative real-time
RAFlq-2 2.1 × 106 9.6 × 104 1.3 PCR. The CT values were determined and transformed into fold
change in expression. Statistical analysis was performed using one-
RAFlq-3 3.5 × 105 3.2 × 104 1.0 way ANOVA, for n = 3, p < 0.005**
Fig. 9 Percentage cell viability of a RAFlq-1-3 on MCF7 human breast cancer cell line, b RPFlq-1-3 on A549 human lung cancer cell line at
48-h incubation. Each point is the mean ± standard error obtained from two independent experiments
1 3
Transition Metal Chemistry
Fig. 11 Annexin V staining shows induction of apoptosis of A549 early apoptotic cells bound to Annexin V but excluded PI (lower right
and MCF7cells treated with RPFlq-2 and RAFlq-2 complexes, quadrant), and late apoptotic cells were both Annexin V- and PI-pos-
respectively. The percent of apoptotic cells was detected by analyz- itive (upper right quadrant); the upper left quadrant contains the dead
ing Annexin V and PI binding with the help of flow cytometry. cells
Viable cells did not bind to Annexin V or PI (lower left quadrant),
Acknowledgements The authors gratefully acknowledge the Uni- References
versity Grants Commission RFSMS-BSR fellowship [UGC No.
F.4-1/2006 (BSR)/5-68/2007 (BSR)], New Delhi, for financial assis-
1. Allardyce CS, Dyson PJ (2001) Platinum Metals Rev. 45:62
tance. The authors are thankful to the Head, Department of Chem-
2. Lentzen O, Moucheron C, Kirsch-De Mesmaeker A (2005) Metal-
istry and Department of Zoology, Faculty of Science, The Maharaja
lotherapeutic drugs & metal-based diagnostic agents. Wiley, West
Sayajirao University of Baroda, for providing us with the necessary
Sussex, pp 359–378
laboratory facilities and required instrumentation facilities to carry
3. Schluga P (2006) Dalton Trans 14:1796
out the research work. The authors thank DBT-MSUB-ILSPARE
4. Pongratz M, Schluga P, Jakupec M, Arion VB, Hartinger C, All-
project, Dr. Vikram Sarabhai Science Block, M. S. University of
maier G, Keppler BK (2004) J Anal At Spectrom 19:46
Baroda, for providing the ESI MS analysis of the complexes and
5. Zanzi I, Srivastava SC, Meinken GE, Robeson W, Mausner LF,
FACS analysis.
Fairchild RG, Margouleff D (1989) Nucl Med Biol 16:397
6. Srivastava SC (1996) Semin Nucl Med 26:119
Compliance with ethical standards
7. Zeng L, Gupta P, Chen Y, Wang E, Ji L, Chao H, Zhe-Sheng C
(2017) Chem Soc Rev 46:5771
Conflict of interest The authors declare that they have no conflict of 8. Barcelo M, Garcia A, Terroon A, Molins E, Prieto MJ, Moreno
interest. V, Martinez J, Llado V, Lopez I, Gutierrez A, Escriba PV (2007)
J Inorg Biochem 101:649
1 3
Transition Metal Chemistry
9. Hartinger CG, Dyson PJ (2009) Chem Soc Rev 38:391 35. Turel I (2002) Coord Chem Rev 232(1–2):27
1 0. Y.Yaw Kai, Chem. Comm. 2005, 38, 4764 36. Ya-Wen T, Yun-Fan C, Yong-Jie L, Kuan-Hung C, Lin C-H, Jui-
11. Soni K (2012) Indo Glob J Pharm Sci 2:43 Hsien H (2018) Molecules 23:159
12. Ronald AR, Low DE (2003) Fluoroquinolone antibiotics. Birkhäu- 37. Adebayo AA, Ajibade PA (2016) J Chem 2016:15, Article ID
ser, Basel, p 250 3672062
13. Kwok Y, Sun D, Clement JJ, Hurley LH (1999) Anti-Cancer Drug 38. Son G, Yeo J, Kim M, Kim S, Holmen A, Akerman B, Norden B
Des 14:443 (1998) J Am Chem Soc 120:6451
14. Yu H, Kwok Y, Hurley LH, Kerwin SM (2000) Biochemistry 39. Pyle AM, Rehmann JP, Meshoyrer R, Kumar CV, Turro NJ, Bar-
39:10236 ton JK (1989) J Am Chem Soc 111:3051
15. Rinky S, Ravirajsinh NJ, Menaka CT, Ranjitsinh VD, Debjani C 40. Zhao G, Lin H, Zhu S, Sun H, Chen Y (1998) J Inorg Biochem
(2012) Transit Met Chem 37:541 70:219
16. Ramadevi P, Rinky S, Sarmita SJ, Ranjitsinh VD, Debjani C 41. Dhar S, Nethaji M, Chakravarty AR (2005) J Inorg Biochem
(2017) J Organomet Chem 833:80 99:805
17. Ramadevi P, Rinky S, Sarmita SJ, Ranjitsinh VD, Debjani C 42. Pasternack RF, Cacca M, Keogh B, Stephenson TA, Williams AP,
(2015) J Photochem Photobio AChem 305:1 Gibbs FJ (1991) J Am Chem Soc 113:6835
18. Sullivan BP, Salmon DJ, Meyer TJ (1978) Inorg Chem 17:3334 43. Lakowicz JR, Weber G (1973) Biochemistry 12:4161
19. Bennett MA, Smith AK (1974) J Chem Soc Dalton Trans (2):233 44. Satyanarayana S, Dabrowiak JC, Chaires JB (1993) Biochemistry
20. Bennett MA, Huang TN, Matheson TW, Smith AK (1983) Inorg 32:2573
Synth 21:74 45. Kelly JM, Tossi AB, McConnell DJ (1985) Nucleic Acids Res
21. Neese F (2012) Wiley Interdiscip Rev Comput Mol Sci 2:73 13:6017
22. Perdew JP (1986) Phys Rev B 33:8822 46. Suh D, Chaires JB (1995) Bioorg Med Chem 3:723
23. Becke AD (1988) Phys Rev A 38:3098 47. Wang Y, Zhang H, Zhang G, Tao W, Tang S (2007) J Lumines-
24. Becke AD (1988) Phys Rev A 38:3098 cence 126:211
25. Weigend F, Ahlrichs R (2005) Phys Chem Chem Phys 7:3297 48. Ahmad B, Parveen S, Khan RH (2006) Biomacromolecules
26. Weigend F (2006) Phys Chem Chem Phys 8:1057 7:1350
27. Andrae D, Haeussermann U, Dolg M, Stoll H, Preuss H (1990) 49. Mishra B, Barik A, Priyadarsini KI, Mohan H (2005) J Chem Sci
Theor Chim Acta 77:123 117:641
28. Vahtras O, Almlöf J, Feyereisen MW (1993) Chem Phys Lett 50. Petra H, Bock K, Atil B, Hoda MA, Korner W, Bartel C, Jun-
213:514 gwirth U, Keppler BK, Michael M, Berger W, Gunda K (2010) J
29. Cossi M, Rega N, Scalmani G, Barone V (2003) J Comput Chem Biol Inorg Chem 15:737
24:669 51. Tan C, Hu S, Liu J, Liangnian J (2011) Eur J Med Chem 46:1555
30. Hanwell MD, Curtis DE, Lonie DC, Vandermeersch T, Zurek E, 52. Wee HA, Elisa D, Claudine S, Rosario S, Lucienne J, Dyson PJ
Hutchison GR (2012) J Cheminform 4:17 (2006) Inorg Chem 45:9006
31. Patel MN, Patel CR, Joshi HN, Thakor KP (2014) Spectrochim
Acta, Part A 127:261 Publisher’s Note Springer Nature remains neutral with regard to
32. Lakowicz JR (1999) Principles of fluorescence spectroscopy, 2nd jurisdictional claims in published maps and institutional affiliations.
edn. Plenum Press, New York
33. Mosmann T (1983) J Immunol Methods 65:55
34. Deacon GB, Phillips R (1980) J Coord Chem Rev 33:227
1 3