Role of Knoevenagel condensate pyrazolone derivative Schiff base ligated transition metal complexes in biological assay and cytotoxic efficacy

Article abstract of DOI:10.1002/aoc.3792

A new bioessential Knoevenagel condensate Schiff base ligand (L) was synthesized by the reaction of 3-(4-hydroxy-3-methoxybenzyl)pentane-2,4-dione and 4-aminoantipyrine. The ligand forms monomeric divalent transition metal complexes (1–4) which were chara

Full text of DOI:10.1002/aoc.3792

Received: 2 January 2017
DOI: 10.1002/aoc.3792
Revised: 5 February 2017
Accepted: 12 February 2017
F U L L PA P E R
Role of Knoevenagel condensate pyrazolone derivative Schiff base
ligated transition metal complexes in biological assay and
cytotoxic efficacy
Rajakkani Paulpandiyan | Alagarraj Arunadevi | Natarajan Raman
Research Department of Chemistry, VHNSN
College, Virudhunagar ‐ 626 001 Tamil
Nadu, India
A new bioessential Knoevenagel condensate Schiff base ligand (L) was synthe-
sized by the reaction of 3‐(4‐hydroxy‐3‐methoxybenzyl)pentane‐2,4‐dione and
4‐aminoantipyrine. The ligand forms monomeric divalent transition metal com-
plexes (1–4) which were characterized using spectral and analytical data. All
these complexes have the general formula [ML]Cl2, where M = Co(II), Ni(II),
Cu(II) and Zn(II). They are electrolytic in nature and adopt square planar geom-
etry. The binding propensity of these complexes with calf thymus DNA was
investigated using absorption spectrophotometric titration, cyclic voltammetry
and viscosity measurements. The binding constant values imply that the com-
plexes bind with DNA via intercalation mode. The in vitro antibacterial and anti-
fungal activities reveal that the complexes have good antimicrobial efficacy
against a set of pathogens. The nucleolytic cleavage activity of these complexes
on pUC18 DNA was investigated using agarose gel electrophoresis. Also, the
in vitro cytotoxicity of the synthesized complexes against a panel of human
tumour cell lines (MCF‐7 and HeLa) and normal cell lines (NHDF and HEK)
was assayed using the MTT method. Interestingly, complex 1 exhibits more
potent anticancer activity than cisplatin and other complexes.
Correspondence
Natarajan Raman, Research Department of
Chemistry, VHNSN College, Virudhunagar
– 626 001, Tamil Nadu, India.
Email: ramchem1964@gmail.com
KEYWORDS
human tumour cell lines, intercalation mode, Knoevenagel condensate Schiff base, MTT assay, nucleolytic
cleavage
1 | INTRODUCTION
with metal complexes by covalent as well as non‐covalent
interactions. A nitrogen base of DNA like guanine N7 can
Cancer is reported as a malignant disease causing swift and
uncontrolled proliferation of cells. As such, defective cellular
apoptosis is essentially involved in the development of
cancer.[1] It is major challenge to treat cancer using an
appropriate strategy while inducing apoptosis. Anticancer
chemotherapy primarily targets damage to DNA and triggers
processes such as cell death and prevention of cellular repro-
duction.[2] Moreover, cell death may be triggered due to
apoptosis, since the cell cycle is being constantly prevented.
DNA is identified as a potential biomolecular target dur-
ing anticancer drug treatment. It is commonly found to bind
replace the labile covalent binding sites of a ligand employed
in a metal complex. Platinum‐based anticancer drugs capable
of covalently binding to DNA are considered to function effi-
ciently for the extensive treatment of various tumours includ-
ing testicular, colorectal and lung cancers.[3–5] In particular,
the DNA‐targeting anticancer chemotherapy drug cisplatin
is specifically recommended for the treatment of testicular
and ovarian cancers. The results of clinical trials of cisplatin
reveal that it is hampered due to the substantial side effects
that arise from its binding mode with DNA via an intrastrand
crosslink formed between neighbouring guanine residues to a
Appl Organometal Chem. 2017;e3792.
wileyonlinelibrary.com/journal/aoc
Copyright © 2017 John Wiley & Sons, Ltd.
1 of 11
https://doi.org/10.1002/aoc.3792

2 of 11
PAULPANDIYAN ET AL.
pair of soft purine nitrogen atoms by covalent binding.[6] In
the past two decades, non‐platinum metal‐based anticancer
complexes including those of nickel(II),[7] copper(II),[8]
zinc(II)[9] and manganese(II)[10] with potent anticancer activ-
ity have been explored in detail. In this respect, much effort
has been directed towards the design and synthesis of new
metal‐based drugs which possess less toxic, more
efficacious, target‐specific and non‐covalent DNA
binding.[11]
Moreover, the design of new metal‐based drugs must be
concerned with enhanced selectivity and distinct modes of
DNA interaction like non‐covalent interaction which mimic
the mode of interaction of proteins with DNA.[12] DNA makes
an essential contribution to life processes, since it carries her-
itage information and induces the biological synthesis of pro-
teins and enzymes by the replication and transcription of
genetic information in living cells. DNA is certainly found
to be a target substance while using metal complexes, since
it provides a variety of options for potential metal binding
sites.[13,14] These binding sites employed for the direct cova-
lent coordination to a metal centre are typically electron‐rich
DNA bases or phosphate groups.
Based on the above considerations, we have synthesized a
Schiff base ligand (L) derived from the condensation reaction
of 3‐(4‐hydroxy‐3‐methoxybenzyl)pentane‐2,4‐dione with
4‐aminoantipyrine as well as its respective Co(II), Ni(II),
Cu(II) and Zn(II) complexes of the structure type [ML]Cl2
(where L is a tetradentate Knoevenagel condensate ligand).
Their scope and merits were studied on the basis of character-
izations like UV absorption titration, viscosity measurements,
cyclic voltammetry and electrophoresis technique to probe
the DNA interaction of these metal complexes. Further, all
the synthesized compounds were screened for their in vitro
antimicrobial activity against various bacterial and fungal
strains. The cytotoxicity of the synthesized compounds was
investigated against human cancer cell lines such as breast
adenocarcinoma (MCF‐7) and cervical carcinoma (HeLa)
cell lines and non‐cancerous human normal cell lines such
as dermal fibroblast (NHDF) and embryonic kidney (HEK)
cell lines using MTT assay.
2 | EXPERIMENTAL
In terms of non‐covalent interactions, DNA tends to inter-
act with transition metal complexes by intercalation and
groove binding or electrostatic surface binding. In particular,
metal complexes containing planar aromatic ligands bind to
DNA by intercalation, which is of interest to researchers.[15]
Such types of metallointercalators act as strongly mutagenic
and exhibit potential chemotherapeutic activity, which is in
accord with their DNA binding affinity.[16] In fact any metal
complex binding to DNA via an intercalative mode has prom-
ising applications. Incidentally, most studies are associated
with complexes of ruthenium and selectively with other metal
complexes. In order to develop an original metal complex with
new structure–pharmacological activity relationship, it is very
critical to understand its DNA binding properties and its pro-
spective interaction leading to cytotoxicity in tumour cells.
Currently, chemotherapeutic Knoevenagel condensate
Schiff bases are the focus of biochemists. Condensation of the
active methylene group of a β‐diketone with an aldehyde sub-
strate affords a non‐enolizable Knoevenagel condensate which
can effectively react with amines to form quantitativelya Schiff
base ligand which possesses favourable coordination sites for
the formation of stable transition metal complexes.[17]
Pyrazol‐3‐one can generate a set of promising compounds with
extensive pharmacological properties including analgesic,
antipyretic and antirheumatic activities.[18] Indeed, a Schiff
base ligand of pyrazol‐3‐one is liable to exhibit enhanced
biological activities as a requisite as a pharmacophore in vari-
ous pharmacologically active agents.[19] In addition, the
Knoevenagel condensate Schiff base metal complexes devel-
opedinrecenttimeshavebeenexploredwidelyinvariousappli-
cations like agriculture, industry, pharmaceutical, etc.[20,21]
The materials and methods involved in the DNA binding
interaction and DNA cleavage efficacy investigations, antimi-
crobial screening and cell proliferation assay are given in the
supporting information (S1).
2.1 | Synthesis of Knoevenagel condensate
β‐Diketone ligand
The
non‐enolizable
β‐diketone
3‐(4‐hydroxy‐3‐
methoxybenzyl)pentane‐2,4‐dione was prepared using the
following procedure. Acetylacetone (1.0 g, 10 mmol) and
4‐hydroxy‐3‐methoxybenzaldehyde (1.52 g, 10 mmol) were
dissolved in ethanol (40 ml) in the presence of piperidine
(0.05 ml). The reaction mixture was stirred homogeneously
for ca 5 h with occasional cooling and incubated in a refrig-
erator for two days. The yellow‐coloured crystalline solid
obtained from the cold reaction mixture was filtered and
washed with ethanol followed by an excess of petroleum
ether to eliminate the unreacted precursors. Washing was
repeated three times and the compound was recrystallized
from ethanol. This compound was used as one component
for the synthesis of the new Schiff base (L).
Yield 78%; yellow colour. Anal. Calcd for C13H16O4
(%): C, 66.1; H, 6.8. Found (%): C, 65.6; H, 6.1. FT‐IR (KBr,
cm−1): 1710 (C═O), 3420 (─OH). 1H NMR (DMSO‐d6,
δ, ppm): 6.8–7.2 (m, Ar‐H, 3H), 2.2 (s, C─CH3, 6H), 2.6
(s, ─CH2, 2H), 3.2 (s, ─CH, 1H), 3.8 (s, OCH3, 3H), 9.6 (s,
OH, 1H). UV–visible (DMSO, cm−1 (transition)): 35 122
(π–π*).

PAULPANDIYAN ET AL.
3 of 11
cm−1): 1575 (─C═N), 1624 (─C═O), 3420 (─OH), 505
(M─O), 445 (M─N). Λm (Ω−1 mol−1 cm2): 148. μeff
(BM): 1.82. UV–visible (DMSO, cm−1 (transition)): 16
181 (d–d).
[CoL]Cl2 (2). Yield 60%; pale green colour. Anal. Calcd
for CoC35H36N6O4Cl2 (%): C, 57.2; H, 4.9; N, 11.4; Co, 8.0.
Found (%): C, 56.8; H, 4.8; N, 11.0; Co, 7.9. FT‐IR (KBr,
cm−1): 1580 (─C═N), 1630 (─C═O), 3410 (─OH), 515
(M─O), 460 (M─N). Λm (Ω−1 mol−1 cm2): 125. μeff
(BM): 3.24. UV–visible (DMSO, cm−1 (transition)): 16
313 (d–d).
2.2 | Synthesis of Schiff Base (L)
The dissolution of 3‐(4‐hydroxy‐3‐methoxybenzyl)pentane‐
2,4‐dione (1.1 g, 5 mmol), prepared as described above, with
4‐aminoantipyrine (2.0 g, 10 mmol) was carried out in etha-
nol (40 ml) and refluxed for ca 4 h with vigorous stirring and
allowed to attain room temperature. Then the reaction mix-
ture was poured into crushed ice to afford orange crystals,
which were filtered and recrystallized from ethanol and dried
under vacuum. The route for the synthesis and the proposed
molecular structure of L with respect to spectral data are
shown in Scheme 1.
[NiL]Cl2 (3). Yield 68%; brownish yellow colour. Anal.
Calcd for NiC35H36N6O4Cl2 (%): C, 57.3; H, 4.9; N, 11.4;
Ni, 8.0. Found (%): C, 57.0; H, 4.8; N, 11.1; Ni, 7.7. FT‐IR
(KBr, cm−1): 1572 (─C═N), 1636 (─C═O), 3400 (─OH),
520 (M─O), 452 (M─N). Λm (Ω−1 mol−1 cm2): 139. UV–
visible (DMSO, cm−1 (transition)): 16 393 (d–d).
[ZnL]Cl2 (4). Yield 66%; pale yellow colour. Anal.
Calcd for ZnC35H36N6O4Cl2 (%): C, 56.7; H, 4.9; N, 11.3;
Zn, 8.8. Found (%): C, 56.2; H, 4.7; N, 11.2; Zn, 8.7. FT‐
IR (KBr, cm−1): 1570 (─C═N), 1635 (─C═O), 3422
Yield 72%; orange colour. Anal. Calcd for C35H36N6O4
(%): C, 69.5; H, 6.0; N, 13.9. Found (%): C, 68.3; H, 5.8;
N, 13.6. FT‐IR (KBr, cm−1): 1654 (C═O), 1590 (C═N),
1
3425 (─OH). H NMR (DMSO‐d6, δ, ppm): 7.1–7.5 (m,
Ar‐H, 13H), 3.4 (s, C═C─CH3, 6H), 2.4 (s, N═C─CH3,
6H), 3.1 (s, N─CH3, 6H), 6.8 (s, HC═C, 1H), 3.8 (s,
OCH3, 3H), 9.5 (s, OH, 1H). 13C NMR (DMSO‐d6, δ,
ppm): 118.9 (C1 and C4), 146.9 (C2), 148.1 (C3), 108.5
(C5), 114.4 (C6), 55.9 (C7), 123.2 (C8), 124.3 (C9), 157.4
(C10), 18.01 (C11), 144.1 (C12), 151.3 (C13), 163.0 (C14),
10.18 (C15), 35.96 (C16), 134.8 (C17), 126.8 (C18 and
C22), 130.6 (C19 and C21), 129.1 (C20). UV–visible
(DMSO, cm−1 (transition)): 35 842 (π–π*), 29 412 (n–π*).
1
(─OH), 512 (M─O), 465 (M─N). H NMR (DMSO‐d6, δ,
ppm): 7.1–7.5 (m, Ar‐H, 13H), 3.4 (s, C═C─CH3, 6H),
2.4 (s, N═C─CH3, 6H), 3.1 (s, N─CH3, 6H), 6.8 (s, HC═C,
1H), 3.8 (s, OCH3, 3H), 9.5 (s, OH, 1H). 13C NMR (DMSO‐
d6, δ, ppm): 117.4 (C1 and C4), 146.4 (C2), 148.6 (C3),
108.0 (C5), 114.9 (C6), 56.9 (C7), 122.5 (C8), 124.7 (C9),
155.4 (C10), 18.32 (C11), 144.1 (C12), 151.1 (C13), 163.4
(C14), 10.33 (C15), 36.12 (C16), 135.2 (C17), 127.1
(C18 and C22), 130.7 (C19 and C21), 129.5 (C20). Λm
(Ω−1 mol−1 cm2): 128. μeff (BM): diamagnetic. UV–visible
(DMSO, cm−1 (transition)): 32 256 (ligand–metal charge
transfer).
2.3 | Synthesis of metal complexes (1–4)
A solution of M(II) chloride (M = Cu, Co, Ni and Zn) in eth-
anol (1 mmol) was mixed with an ethanolic solution of Schiff
base L (1 mmol) and the resultant mixture was refluxed for ca
3 h. The progress of the reaction was continuously monitored
with the aid of TLC. The solid complex precipitated from the
solution was filtered off and washed thoroughly with ethanol
followed by petroleum ether and dried over anhydrous CaCl2
under vacuum.
3 | RESULTS AND DISCUSSION
[CuL]Cl2 (1). Yield 62%; brown colour. Anal. Calcd for
CuC35H36N6O4Cl2 (%): C, 56.9; H, 4.9; N, 11.4; Cu, 8.6.
Found (%): C, 56.2; H, 4.8; N, 11.1; Cu, 8.5. FT‐IR (KBr,
All the metal(II) complexes (1–4) were obtained upon reac-
tion between metal ion and ligand at a metal‐to‐ligand molar
SCHEME 1 Route for the synthesis of
Knoevenagel condensate Schiff base (L)

4 of 11
PAULPANDIYAN ET AL.
ratio of 1:1. These synthesized complexes exist as stable
compounds at room temperature which are soluble in DMF
and DMSO, while insoluble in water and many common
organic solvents. Analytical data of these complexes support
the proposed general formula [ML]Cl2. The experimental ele-
mental analyses data of the compounds are found to be in
good agreement with the calculated data. The magnetic sus-
ceptibility data imply the monomeric nature of these com-
plexes. In addition, the high conductance values of these
metal complexes indicate their electrolytic properties due to
the existence of counter ions in the proposed structure of
complexes 1–4. The proposed molecular structure of the
Knoevenagel condensate Schiff base complexes is depicted
in Figure 1.
3.2 | Electronic absorption spectra and
magnetic moments
The absorption spectra and magnetic data can afford ready
and reliable information about the ligand arrangement around
the metal centre in transition metal complexes. This is
employed as an essential tool to distinguish between the
geometries of complexes. Electronic spectra of L and its com-
plexes were recorded in the range 200–1100 nm at room tem-
perature in DMSO medium. The absorption spectra of L and
its metal complexes are shown in Figure S2. The absorption
spectrum of L exhibits two characteristic bands at 35 842
and 29 412 cm−1 which correspond to the intraligand charge
transfer transitions of π → π* and n → π*, respectively. These
two transitions are shifted due to complex formation, either a
bathochromic or hypsochromic shift.
In the case of complex 1, a strong absorption d–d band
2
2
appears at 16 181 cm−1, assigned to B1g → A1g transition,
which is characteristic of square planar geometry.[25] This is
further supported by the magnetic susceptibility value (1.82
BM). Moreover, a d–d band at 16 393 cm−1 for complex 2
3.1 | Fourier transform infrared (FT‐IR)
spectra
FT‐IR spectra of ligand L and its respective metal complexes
1–4, recorded in the region 400–4000 cm−1, are shown in
Figure S1. The spectra of the complexes are compared with
that of the free ligand in order to identify the coordination
sites which contribute to chelation. The azomethine nitrogen
(─C═N) stretching frequency appears at 1590 cm−1 for the
free ligand whereas for the metal complexes the signal shifts
towards lower wavenumber around 1562–1580 cm−1. The
shift in frequency occurs due to the coordination of imine
groups towards the metal centre.[22] Similarly the band corre-
sponding to the carbonyl oxygen (─C═O) stretching fre-
quency at 1654 cm−1 (L) is shifted to 1624–1638 cm−1 due
to the coordination of carbonyl oxygen atom of the ligand
to the metal centre. A strong broad band is noticed in the
range 3400–3425 cm−1 for both the free ligand and its com-
plexes, which is due to the phenolic ─OH group of vanillin
moiety that implies the unbounded nature in coordination.[23]
The spectrum of each metal complex exhibits slightly less
intense bands in the regions 445–460 and 505–520 cm−1
which indicate the formation of (M─N) and (M─O) bonds,
respectively.[24]
1
1
represents the A1g → B1g transition, which also implies
square planar geometry. In fact the magnetic moment value
(3.24 BM) of this d7 Co(II) ion complex supports the square
planar stereochemistry around the metal centre.[26] Similarly,
the d–d band at 16 393 cm−1 for complex 3 can also be
assigned to 1A1g → 1B1g transition and square planar geome-
try is identified from the observed diamagnetic value of this
complex. Obviously, complex 4 does not show any band in
the visible region due to the diamagnetic behaviour of the
Zn(II) ion in the complex.
3.3 | 1H NMR and 13C NMR spectra
1
In the case of H NMR spectra, the peak assignment for the
proton resonances was based on their peak multiplicity,
intensity pattern and correlation of the integration values of
the protons with the predicted pattern. The H NMR spectra
of L and its diamagnetic Zn(II) complex 4 (Figures S3 and
S4) were recorded in DMSO‐d6 using tetramethylsilane as
1
1
the internal standard. The H NMR spectrum of L shows
peaks at 7.1–7.5 ppm (m, 13H) and 9.5 ppm (s, 1H) attrib-
uted to the phenyl multiplet and the phenolic ─OH group
of the 4‐hydroxy‐3‐methoxybenzaldehyde moiety of L. It
also shows the following signals: C═C─CH3 at 3.4 ppm
(s, 6H), N═C─CH3 at 2.4 ppm (s, 6H), N─CH3 at 3.1 ppm
(s, 6H), HC═C at 6.8 ppm (s, 1H) and OCH3 at 3.8 ppm
(s, 3H). The signal for the phenolic ─OH proton is observed
in the spectrum of Zn(II) complex 4 which reveals that ─OH
proton exists free from sequestration. Likewise there is no
significant alteration perceived with the remaining signals
of this spectrum.
FIGURE 1 Proposed structure of the Knoevenagel condensate Schiff
base complexes

PAULPANDIYAN ET AL.
5 of 11
The 13C NMR spectra of L and its complex 4 are depicted
in Figures S5 and S6. The signals of aromatic carbons of L
are identified in the range 119–148 ppm. Moreover, the imine
group (C═N) and carbonyl (C═O) carbons are observed at
157 and 163 ppm. In the spectrum of complex 4, it is
observed that these peaks are shifted to downfield region,
specifying the coordination of (C═N) and (C═O) groups
towards the metal centre. There is no such marked difference
noted with other signals in this spectrum.
plane σ‐bonding) and β2 (covalent in‐plane π‐bonding) were
calculated from the following equations:
À
Á
A∥
0:036
3
7
α2 ¼
þ g∥−2:0027 þ ðg⊥−2:0027Þ þ 0:04 (2)
ꢀ
ꢁ
À
Á
E
β2 ¼ g∥−2:0027
(3)
−8λα2
where λ = −828 cm−1 for the free copper ion and E is the
electronic transition energy. In the case of complexes,
α2 = 0.5 specifies absolute covalent bonding, whereas
α2 = 1.0 implies pure ionic bonding. In the present case,
the calculated values of α2 (0.74) and β2 (0.96) indicate
that the in‐plane σ‐bonding and in‐plane π‐bonding are
appreciably covalent and are consistent with strong in‐plane
π‐bonding in this complex. Based on the above parameters,
complex 1 is presumed to exist in square planar geometry.
Thus, the EPR data for complex 1 provide evidence to cor-
roborate the conclusion drawn on the basis of electronic spec-
trum and magnetic moment value as well.
3.4 | Electron paramagnetic resonance (EPR)
spectra
The X band EPR spectrum of complex 1 was recorded in
DMSO medium at room temperature and liquid nitrogen
temperature at a frequency of 9.1 GHz under a magnetic
field strength of 3000 G (Figure S7). The spin Hamiltonian
parameters of complex 1 were calculated and are summa-
rized in Table 1. Based on the spectral data, it is noted that
A|| (131) > A⊥ (26); g|| (2.18) > g⊥ (2.04) > 2.0027, which
reveal that the unpaired electron is most likely to localize
in dx2−y2 orbital of Cu(II) ion, scaffolding the square planar
geometry.[27] In accordance with g factors, the geometric
parameter G is determined, indicating a measure of
exchange interaction between Cu(II) centres in polycrystal-
line compound:
3.5 | Mass spectra
Electron spray ionization mass data of the free ligand and its
complexes were obtained to determine their molecular weight
and study their fragmentation patterns. The mass spectra of L
and its complex 1 are displayed in Figure S8. The mass spec-
trum of L shows the molecular ion peak at m/z 604 with
respect to [C35H36N6O2]+ ion. Also the spectrum exhibits
peaks for fragments at m/z 481, 228, 188, 123 and 107, indi-
cating the possible fragmented species [C28H29ON6O2]+,
[C13H14N3O]+, [C11H12N2O]+, [C7H7O2]+ and [C7H7O]+.
The mass spectrum of [CuL]+ complex shows molecular
ion peak at m/z 669 [M + 1] which is equivalent to its molec-
ular weight. Complex 1 exhibits additional fragment ion
peaks at m/z 604, 481, 228, 188, 123 and 107 corresponding
to [C35H36N6O2]+, [C28H29ON6O2]+, [C13H14N3O]+,
[C11H12N2O]+, [C7H7O2]+ and [C7H7O]+. Based on the
observed m/z values or fragmentation patterns of L and its
complex, the stoichiometry of the complex is [ML]Cl2.
g∥−2:0027
g⊥−2:0027
G ¼
(1)
This measures the exchange interaction between the copper
centres in polycrystalline solid. The resultant value
(G = 4.5) for the exchange interaction parameter for complex
1 verifies a negligible exchange interaction among Cu–Cu in
complex 1 according to Hathaway and Tomlinson.[28]
The degree of geometrical distortion (g||/A||) is observed
as an index of deviation from idealized geometry; a value
below 120 cm−1 is typically known for planar complexes,
whereas the range 120 to 180 cm−1 is characteristic of minor
to moderate distortion and 180–250 cm−1 indicates a major
distortion towards tetrahedron.[29] The g||/A|| value obtained
(166 cm−1) for the synthesized complex 1 is in good agree-
ment with appreciable deviation from planarity which is fur-
ther supported by the bonding parameter (α2) value which is
less than unity. The covalency parameters α2 (covalent in‐
3.6 | DNA binding studies
3.6.1 | Absorption spectral titrations
UV–visible absorption spectroscopy is commonly used to
analyse the binding mode of metal complexes with
DNA.[30] DNA can bind to metal complexes via covalent
and/or non‐covalent interactions. In the case of covalent
binding, a labile ligand of a complex can be substituted by
nitrogen base of DNA such as guanine (N7), whereas the
non‐covalent interactions include intercalative, electrostatic
and groove (surface) binding of metal complexes outside of
TABLE 1 Spin Hamiltonian parameters of complex 1 in DMSO
solution at liquid nitrogen temperature
g‐tensor
A (× 10−4 cm−1
)
Complex 1
g||
g⊥
giso
A||
A⊥
Aiso g||/A||
G
[CuL]Cl2
2.18 2.04 2.09 131
26
61 166 4.5

6 of 11
PAULPANDIYAN ET AL.
the DNA helix, along major/minor groove.[31] The interaction
between a complex and DNA is assumed to perturb the
ligand‐centred transitions of the complex. In this context,
absorption spectra of the complexes were obtained in the
presence or absence of calf thymus DNA (CT DNA) at vari-
ous concentrations. Hyperchromism and hypochromism are
two different spectral features of DNA concomitant to its
double helix structure, wherein hyperchromism arises from
the alternation of the secondary structure of DNA and
hypochromism arises from the contraction of DNA in its
helix axis as well as from the conformational variation of
DNA.[32] During intercalation, π* orbital of the intercalated
ligand can couple with the π orbital of DNA base pairs which
leads to bathochromism due to the reduction of π → π* tran-
sition energy. On the other hand, the coupling of π orbital
favours partial electron filling, thus inhibiting the transition
probabilities and concomitantly showing hypochromism.[33]
The absorption spectra of complex 1 in the absence and
presence of CT DNA in the UV region are shown in
Figure 2. The absorption parameters of complexes 1–4 bound
to CT DNA are listed in Table 2. A marginal hypochromic
shift in the intraligand transition is identified with all four
complexes upon addition of CT DNA in increasing quantity
with a red shift (2–4 nm). This implies that these complexes
interact with CT DNA through the intercalative mode with a
strong interaction between the complexes and the base pairs
of DNA.[34] The planarity and extended aromaticity of the
Knoevenagel condensate Schiff base ligand unit promote the
stacking of the molecule among the DNA bases. This study
shows the trend of hypochromism among the synthesized
complexes follows the order 1 > 4 > 3 > 2, with the binding
constant (Kb) varying between 2.6 × 105 and 4.5 × 105 M−1
(Table 2). Binding constant values of these complexes indi-
cate that complex 1 has the maximum binding affinity since
TABLE 2 Electronic absorption parameters for interaction of CT
DNA with complexes
λmax (nm)
Δλ
(nm)
H
Kb
Complex
Free
Bound
(%)a
(× 105 M−1 b
)
1
2
3
4
376
378
382
380
376.8
379.6
384.0
381.0
0.8
1.6
2.0
1.0
9.6
4.4
4.9
5.6
4.5
2.6
3.2
3.8
aH = [(Afree − Abound)/Afree] × 100%.
bError limit: ꢀ2%.
it possesses large hydrophobic contacts and π–π stacking of
the DNA bases, while the other complexes act as moderate
binders. In particular, the observed binding constant values
are found to be much higher than that of the effective antican-
cer drug cisplatin (3.20 ꢀ 0.15 × 104 M−1 [35]
)
and less than
that of ethidium bromide (1.4 × 106 M−1).[36]
3.6.2 | Electrochemical behaviour
The electrochemical method is an important tool in
examining the type and approach of DNA binding with metal
complexes. Cyclic voltammetry was chosen to understand the
electrochemical properties of the complexes. Cyclic
voltammetric studies of the complexes were conducted in
DMSO (10−3 M) at a sweep rate of 0.5 V s−1 in the
potential range + 2 to −2 V. The characteristic cyclic
voltammograms of complex 1 are depicted in Figure 3 and
quasi‐reversible redox couple for each complex has been
analysed upon the addition of CT DNA and shifts of E1/2
and ΔEp are presented in Table 3.
The incremental addition of CT DNA leads to a substan-
tial drop in the voltammetric current of the redox wave with a
slight shift in E1/2 to positive potential. The drop in current
FIGURE 2 Absorption spectra of complex 1 in the absence (dashed
curve) and presence (solid curves) of CT DNA in 5 mM Tris–HCl/
50 mM NaCl (pH = 7.2) at 25 °C. Arrow indicates the change in
absorbance with increasing DNA concentration
FIGURE 3 Cyclic voltammograms of complex 1 in 5 mM Tris–HCl/
50 mM NaCl (pH = 7.2) at 25 °C in presence of increasing amounts of
CT DNA

PAULPANDIYAN ET AL.
7 of 11
TABLE 3 Electrochemical parameters for interaction of CT DNA
with complexes
E1/2 (V)a
ΔEp (V)b
Complex
Free
Bound
Free
Bound
Ipa/Ipc
1
2
3
4
0.248
−0.287
−0.512
0.567
0.356
−0.161
0.623
1.325
1.566
1.002
1.497
1.426
1.784
2.106
1.506
0.91
0.89
0.84
0.66
0.778
aE1/2 = (Epa + Epc)/2.
bΔEp = Epa − Epc
.
intensity can be elucidated in terms of an equilibrium mixture
of free and DNA‐bound complexes at the electrode sur-
face.[37] This condition reveals that the complex stabilizes
the duplex (GC pairs) by an intercalating mode. Evidently,
the decline of the voltammetric currents in the presence of
DNA may be accounted for by the hindered diffusion of the
metal complex bound to CT DNA. Both cathodic and anodic
peaks show positive or negative shift which specifies the
intercalation of complex to DNA base pairs.[38]
FIGURE 4 Effect of increasing amounts of ethidium bromide (EB)
and complexes 1–4 on the relative viscosity of CT DNA. Plot of
relative viscosity (η/η0)1/3 versus [complex]/[DNA]
3.6.4 | Agarose gel electrophoresis
The agarose gel pattern of plasmid DNA is accounted for by
two distinct conformations, one is supercoiled DNA (Form I)
and the other is nicked circular DNA (Form II). Altogether,
the relaxed form of plasmid DNA is formed due to the cleav-
age of the DNA strands (nicking of DNA), known as the
nicked circular form, which migrates very slowly in agarose
gel. The linear form of DNA (Form III) is generated to
migrate between supercoiled and nicked circular forms, if
both strands are cleaved.[41] Agarose gel electrophoresis can
differentiate all three different plasmid DNA conformations
during analysis.
3.6.3 | Viscosity measurements
To further investigate the binding nature of the complexes
with DNA, viscosity was measured after the incubation of
DNA solutions with the complexes. Hydrodynamic
measurement which differs with respect to length variation
(i.e. viscosity and sedimentation) is considered as the least
ambiguous as well as the essential test for the binding model
in solution, when crystallographic data are unavailable.[39] To
explore the binding mode of complexes with CT DNA,
change in viscosity of DNA at room temperature was mea-
sured on varying the complex concentration. A typical
intercalative mode causes a major enhancement in viscosity
of DNA solution due to the appreciable separation of base
pairs at intercalation sites, leading to an overall increase in
DNA length. In contrast, complexes which bind absolutely
in the DNA grooves by partial and/or non‐classical intercala-
tion, under identical conditions, generally cause less of a
change (positive or negative) or no change in viscosity of
DNA solution.[40] The effects of all the synthesized com-
plexes on the viscosity of DNA at 30 ꢀ 0.1 °C are shown in
Figure 4. Viscosity measurement results evidently show that
each complex can intercalate between adjacent DNA base
pairs, initiating an extension in the helix and increasing the
viscosity of DNA. When the concentration of the complex
is increased, the complexes can intercalate strongly, leading
to a larger increase in viscosity of DNA. Therefore, based
on spectroscopic studies along with viscosity measurements,
it can be perceived that these complexes bind with the DNA
via intercalation mode.
3.6.5 | Absence of activators
We deduce that complex 1 shows binding tendency with CT
DNA exclusively. The cleavage activity is examined using
supercoiled pUC18 DNA. It is incubated with different con-
centrations (control, 20, 40, 60, 80 and 100 μM) of complex
1 in Tris–borate–EDTA buffer at pH = 7.2 for 3 h, free from
any activating agents. With increasing concentration of 1, the
ratio of Form I reduces gradually and there is partial transfor-
mation to Form II with concurrent increase in the intensity of
the latter form, whereas the production of Form III also
increases at 100 μM (Figure 5, lane 6). Thus, in the absence
of activators, the concentration‐dependent cleavage of
pUC18 DNA is observed in the case of complex 1 as shown
in Figure 5.
3.6.6 | Presence of activators
Moreover, DNA cleavage was evaluated by detecting the
transformation of supercoiled to linear form in the presence

8 of 11
PAULPANDIYAN ET AL.
The above OH free radicals are employed to oxidize the
deoxyribose moiety as well as the hydrolytic cleavage of
sugar phosphate backbone.[42,43] Further studies in this direc-
tion would assist to envisage the exact mechanistic pathway
in cleavage reactions involved in radical scavengers and reac-
tive oxygen species of metal complexes with plasmid DNA.
3.7 | Antimicrobial activity
FIGURE 5 Gel electrophoresis pattern showing cleavage of pUC18
DNA (30 μM) by complex 1 at 35 °C after 2 h of incubation. Lane 1:
DNA control; lane 2: 20 μM of 1 + DNA; lane 3: 40 μM of 1 + DNA;
lane 4: 60 μM of 1 + DNA; lane 5: 80 μM of 1 + DNA; lane 6: 100 μM
of 1 + DNA
The antimicrobial activity of the synthesized compounds
towards biological systems is an intriguing aspect in current
research. The in vitro antimicrobial activity of the synthe-
sized free ligand and its metal complexes was tested against
various microorganisms, namely two Gram‐positive bacteria
(Staphylococcus aureus and Bacillus subtilis), three Gram‐
negative bacteria (Escherichia coli, Klebsiella pneumoniae
and Salmonella typhi) and five fungi (Aspergillus niger,
Fusarium solani, Curvularia lunata, Rhizoctonia bataticola
and Candida albicans), using the broth dilution method. Flu-
conazole and kanamycin were chosen as standards for fungi
and bacteria, respectively. The minimum inhibitory concen-
tration (MIC) values against the strains of bacteria and fungi
are shown in Figures 7 and 8, respectively.
Tables 4 and 5 summarize the MIC values of the com-
plexes under investigation. The data for the antibacterial
and antifungal activity reveal that the metal chelates exhibit
potential activity as compared to free ligand against the same
microorganisms due to the greater lipophilic nature of the
complexes, under identical experimental conditions.[44] This
tends to suggest that the chelation could facilitate the ability
of a complex to cross a cell membrane according to Tweedy’s
chelation theory.[45] Chelation significantly reduces the polar-
ity of the metal ion due to the partial sharing of its positive
charge with donor groups and possible electron
delocalization over the whole chelate ring. Such type of
chelation could improve the lipophilic properties of the
of activators. Figure 6 represents the cleavage of pUC18
DNA stimulated by the metal complexes in the presence of
H2O2 as oxidant. Control experiments with DNA itself failed
to indicate any considerable cleavage of pUC18 DNA, even
upon long exposure period (lane 1). The ratio of supercoiled
form (Form I) reduced constantly and partly transformed to
nicked form (Form II), and the intensity of the Form II band
increased and transformed to Form III by the complexes even
at a concentration of 20 μM. Each complex tends to exhibit a
pronounced cleavage activity in the presence of H2O2. This
may be responsible for the formation of hydroxyl free radi-
cals. It is manifested that complex 1 possesses high ability
to cleave the supercoiled plasmid DNA as compared to the
other complexes. These observations imply that DNA alters
its conformation due to the binding of the metal complexes.
The generation of a hydroxyl radical upon reaction of metal
complex with oxidant may be described as follows:
M(II) + e− → M(I)
−
M(I) + O2 → M(II) + O2
2O2− + 2H+ → H2O2 + O2
M(I) + H2O2 → M(II) + OH− + OH•
O2− + H2O2 → O2 + OH− + OH•
FIGURE 6 Gel electrophoresis pattern showing cleavage of pUC18
DNA (30 μM) treated with L and its metal complexes 1–4 (20 μM) at
35 °C after 2 h of incubation. Lane 1: DNA control; lane 2: DNA + L+
H2O2; lane 3: DNA + 1 + H2O2; lane 4: DNA + 2 + H2O2; lane 5:
DNA + 3 + H2O2; lane 6: DNA + 4 + H2O2
FIGURE 7 Antibacterial activity of synthesized compounds against
various pathogens

PAULPANDIYAN ET AL.
9 of 11
activity of the copper complex is deduced from the fact that
increasing the size of the metal ion decreases the polarization
which is further explained on the basis of chelation theory.[45]
The variation in the efficiency of the various complexes
against diverse organisms depends on the impermeability of
the cells of the microbes or on the variation in ribosome of
the microbial cells.[46]
3.8 | In Vitro cytotoxicity
The promising results observed from the earlier studies con-
cerned with this work revealing the potential to act as cancer
chemotherapeutic agents prompted us to evaluate the in vitro
cytotoxic activity of the synthesized compounds. The in vitro
anticancer activity was evaluated against a couple of cancer
cell lines, MCF‐7 and HeLa, and a couple of non‐cancerous
cell lines, NHDF and HEK, using MTT assay with cisplatin
as standard (positive control). The cell lines were treated with
the synthesized compounds at a range of concentrations
(0–100 μM) for 48 h. The results were analysed by means
of cell inhibition expressed as IC50 values, which are listed in
Table 6.
FIGURE 8 Antifungal activity of synthesized compounds against
various pathogens
TABLE 4 Minimum inhibitory concentration (MIC) of synthesized
compounds against growth of bacteria
MIC (× 104 μM); SEM = ꢀ1.5
S.
B.
E.
K.
S.
Compound aureus
subtilis
coli
pneumoniae
typhi
Complexes 1–4 show higher cytotoxic activity against
cell lines HeLa and MCF‐7. Among the complexes studied,
complex 1 has significantly stronger inhibitor effect on the
proliferation of HeLa and MCF‐7 cells than cisplatin under
similar experimental conditions. This result suggests that
complex 1 is a potential cytotoxic agent against cancer cells
HeLa and MCF‐7 for which the IC50 values are 18 and
16 μM, respectively. It is found that the anticancer activity
of complex 1 is better than that of cisplatin. In particular,
the results of in vitro cytotoxic activity indicate that the IC50
values of the synthesized complexes against non‐cancerous
human cell lines (NHDF and HEK) are found to be above
72 μM which indicates that complexes 1–4 are very specific
towards cancer cell lines only. These results are displayed in
Figure 9. Further in vitro investigations are underway in order
to understand the mechanism of the antiproliferative activity
of these complexes.
L
15.8
3.2
5.2
6.5
4.5
1.8
16.5
3.5
5.8
6.2
4.2
1.6
16.2
3.8
5.6
6.8
4.5
2.2
17.2
4.2
5.2
2.2
4.7
2.8
15.2
4.0
5.4
6.4
4.2
2.6
1
2
3
4
Kanamycina
aKanamycin used as standard.
TABLE 5 Minimum inhibitory concentration (MIC) of synthesized
compounds against growth of fungi
MIC (× 104 μM); SEM = ꢀ2
F.
A.
niger
C.
lunata
R.
C.
Compound
solani
bataticola albicans
L
25.4
7.6
9.2
9.5
8.5
2.5
24.6
7.2
27.5
8.5
25.8
8.2
27.2
7.5
TABLE 6 Cytotoxicity of complexes 1–4 against various human cell
lines
1
2
9.8
9.6
9.4
9.5
IC50 (μM)a
3
10.6
8.2
10.2
8.6
11.5
8.8
11.2
8.4
Compound
HeLa
MCF‐7
NHDF
HEK
4
Fluconazolea
2.8
2.4
3.2
3.4
1
18 ꢀ 0.6
30 ꢀ 1.1
26 ꢀ 0.8
24 ꢀ 0.8
22 ꢀ 1.2
16 ꢀ 0.8
38 ꢀ 1.2
32 ꢀ 1.2
28 ꢀ 0.8
26 ꢀ 1.2
72 ꢀ 0.8
86 ꢀ 1.0
92 ꢀ 0.6
78 ꢀ 0.8
76 ꢀ 1.1
>100
>100
>100
>100
>100
aFluconazole used as standard.
2
3
4
central metal atom, which consequently favours its perme-
ation through the lipid layer of the cell membrane. In partic-
ular, complex 1 shows a higher antifungal activity and
antibacterial activity than the other complexes. This higher
Cisplatin (standard)
aIC50 = concentration of drug required to inhibit growth of 50% of cancer cells
(data are mean ꢀ SD of three replicates each).

10 of 11
PAULPANDIYAN ET AL.
facilities. IIT Bombay (SAIF) and IIT Chennai, India are
gratefully acknowledged for providing instrumental facilities.
REFERENCES
[1] S. Kasibhatla, B. Tseng, Mol. Cancer Ther. 2003, 2, 573.
[2] P. Kahlem, B. Dörken, C. A. Schmitt, J. Clin. Invest. 2004,
113, 169.
[3] L. Kelland, Nat. Rev. Cancer 2007, 7, 573.
[4] M. J. Hannon, Pure Appl. Chem. 2007, 79, 2243.
[5] S. P. Fricker, Dalton Trans. 2007, (43), 4903.
[6] E. R. Jamieson, S. J. Lippard, Chem. Rev. 1999, 99, 2467.
FIGURE 9 Trends in the inhibition of various human cell lines
against synthesized complexes 1–4
[7] P. Sathyadevi, P. Krishnamoorthy, R. R. Butorac, A. H. Cowley,
N. S. P. Bhuvanesh, N. Dharmaraj, Dalton Trans. 2011, 40, 9690.
[8] A. G. Majouga, M. I. Zvereva, M. P. Rubtsova, D. A. Skvortsov,
A. V. Mironov, D. M. Azhibek, O. O. Krasnovskaya, V. M.
Gerasimov, A. V. Udina, N. I. Vorozhtsov, E. K. Beloglazkina,
L. Agron, L. V. Mikhina, A. V. Tretyakova, N. V. Zyk, N. S.
Zefirov, A. V. Kabanov, O. A. Dontsova, J. Med. Chem. 2014,
57, 6252.
4 | CONCLUSIONS
The Knoevenagel condensate Schiff base (L) and its transi-
tion metal(II) complexes (1–4) have been synthesized and
characterized using analytical and spectral methods. Based
on the physicochemical analysis, these complexes adopt
square planar geometry. The binding propensity of DNA
towards the complexes has been studied using absorption
spectrophotometric titration, cyclic voltammetry and viscos-
ity measurements, which suggest that the complexes bind
with CT DNA through the intercalation mode. The binding
constant values disclose that the DNA binding affinity of
the complexes lies in the order 1 > 4 > 3 > 2. Fascinat-
ingly, complex 1 exhibited potential antibacterial and anti-
fungal activity against various microbial strains and
proved to have better antimicrobial activity than the free
ligand and other complexes. Gel electrophoresis experiment
shows that the presence of hydrogen peroxide enhances the
DNA cleavage to a significant extent. It has also been
found that concentration plays a vital role in increasing
the cleavage activity. Moreover, all the complexes show
considerable cytotoxic activity against human cancer
cell lines MCF‐7 and HeLa but less toxicity against non‐
cancerous human cell lines NHDF and HEK. In particular,
complex 1 shows excellent cytotoxic activity against the
two human tumour cell lines. All encouraging chemical
and biological findings indicate that complex 1 is a suitable
candidate for potential application as an anticancer drug.
Further studies are needed to assess the in vivo pharmaco-
logical properties and to elucidate the actual mechanism of
the biological activity.
[9] C. R. Kowol, R. Trondl, V. B. Arion, M. A. Jakupec, I.
Lichtscheidl, B. K. Keppler, Dalton Trans. 2010, 39, 704.
[10] W. Al Zoubi, Y. Gun Ko, J. Organometal. Chem. 2016, 822, 173.
[11] B. Neto, A. Lapis, Molecules 2009, 14, 1725.
[12] P. R. Inamdar, R. Chauhan, J. Abraham, A. Sheela, Inorg. Chem.
Commun. 2016, 67, 67.
[13] S. Chan, W. T. Wong, Coord. Chem. Rev. 1995, 138, 219.
[14] G. Prativel, J. Bernadou, B. Meunier, Adv. Inorg. Chem. 1998,
45, 251.
[15] Y. Xiong, L. N. Ji, Coord. Chem. Rev. 1996, 185, 711.
[16] V. Rajendiran, M. Murali, E. Suresh, K. Somasundaram, M.
Palaniandavar, Dalton Trans. 2008, (1), 148.
[17] S. Srinivasan, G. Rajagopal, P. R. Athappan, Transition Met. Chem.
2001, 26, 588.
[18] G. Menozzi, L. Mosti, L. Merello, A. Piana, U. Armani, M. Ghia,
M. Angiola, F. Mattioli, Il Farmaco 2000, 55, 219.
[19] A. M. Asiri, S. A. Khan, Molecules 2010, 15, 6850.
[20] N. Raman, R. Jeyamurugan, R. Senthilkumar, B. Rajkapoor, S. G.
Franzblau, Eur. J. Med. Chem. 2010, 45, 5438.
[21] S. Sumathi, P. Tharmaraj, C. D. Sheela, C. Anitha, Spectrochim.
Acta A 2012, 97, 377.
[22] N. M. EI‐Metwally, I. M. Gabr, A. M. Shallaby, J. Coord. Chem.
2005, 58, 1154.
[23] R. Prabhakaran, P. Kalaivani, R. Jayakumar, M. Zeller, A. D.
Hunter, S. V. Renukadevi, E. Ramachandran, K. Natarajan,
Metallomics 2011, 3, 42.
[24] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coor-
dination Compounds, Wiley‐Interscience, New York 1986.
ACKNOWLEDGEMENTS
[25] A. B. P. Lever, Inorganic Electronic Spectroscopy, Elsevier, New
York 1968.
The authors express their heartfelt thanks to the College Man-
aging Board, Principal and Head of the Department of Chem-
istry, VHNSN College for providing necessary research
[26] A. A. Nejo, G. A. Kolawole, A. O. Nejo, J. Coord. Chem. 2010,
63, 4398.

PAULPANDIYAN ET AL.
11 of 11
[27] S. Srinivasan, G. Rajagopal, P. R. Athappan, Transition Met. Chem.
2001, 26, 588.
[41] V. Thamilarasan, A. Jayamani, N. Sengottuvelan, Eur. J. Med.
Chem. 2015, 89, 266.
[28] B. J. Hathaway, A. A. G. Tomlinson, Coord. Chem. Rev. 1970, 6, 1.
[42] N. Raman, N. Pravin, Eur. J. Med. Chem. 2014, 80, 57.
[29] S. Belaid, A. Landreau, S. Djebbar, O. Benali‐Baitich, G. Bouet, J.
P. Bouchara, J. Inorg. Biochem. 2008, 102, 63.
[43] R. Paulpandiyan, N. Raman, Appl. Organometal. Chem. 2016,
30, 531.
[30] V. G. Vaidyanathan, B. U. Nair, Eur. J. Inorg. Chem. 2004, 9, 1840.
[44] W. Al Zoubi, A. Ali Salih Al‐Hamdani, M. Kaseem, Appl.
Organometal. Chem. 2016, 30, 810.
[31] F. Dimiza, A. N. Papadopoulos, V. Tangoulis, V. Psycharis, C. P.
Raptopoulou, D. P. Kessissoglou, G. Psomas, Dalton Trans.
2010, 39, 4517.
[45] B. G. Tweedy, Phytopathology 1964, 55, 910.
[46] M. Kurtoğlu, M. M. Dağdelen, S. Toroğlu, Transition Met. Chem.
2006, 31, 382.
[32] E. C. Long, J. K. Barton, Acc. Chem. Res. 1990, 23, 271.
[33] A. M. Pyle, J. P. Rehmann, R. Meshoyrer, C. V. Kumar, N. J. Turro,
J. K. Barton, J. Am. Chem. Soc. 1989, 111, 3053.
SUPPORTING INFORMATION
[34] J. K. Barton, A. Danishefsky, J. M. Goldberg, J. Am. Chem. Soc.
1984, 106, 2172.
Additional Supporting Information may be found online in
the supporting information tab for this article.
[35] F. Arjmand, F. Sayeed, D. Muddassir, J. Photochem. Photobiol. B
2011, 103, 166.
[36] P. R. Inamdar, A. Sheela, J. Photochem. Photobiol. B 2016,159, 133.
[37] S. Tabassum, S. Parveen, F. Arjmand, Acta Biomater. 2005, 1, 677.
How to cite this article: Paulpandiyan R, Arunadevi
A, Raman N. Role of Knoevenagel condensate
pyrazolone derivative Schiff base ligated transition
metal complexes in biological assay and cytotoxic
efficacy. Appl Organometal Chem. 2017;e3792.
https://doi.org/10.1002/aoc.3792
[38] L. Z. Li, C. Zhao, T. Xu, H. W. Ji, Y. H. Yu, G. Q. Guo, H. Chao,
J. Inorg. Biochem. 2005, 99, 1076.
[39] P. Arthi, S. Shobana, P. Srinivasan, D. Prabhu, C. Arulvasu, A.
Kalilur Rahiman, J. Photochem. Photobiol. B 2015, 153, 247.
[40] S. Satyanarayana, J. C. Dabrowiak, J. B. Chaires, Biochemistry
1992, 31, 9319.