The synthesis, characterization, DNA/protein interaction, molecular docking and catecholase activity of two Co(II) complexes constructed from the aroylhydrazone ligand

Article abstract of DOI:10.1016/j.molstruc.2019.127229

Two Co(II) complexes, Co(BBH)phenMeCN (1) and Co(BSH)phenMeCN (2) (HBBH = benzophenone benzoyl hydrazone, HBSH = benzophenone salicylylhydrazone ligand and phen = 1,10-phenanthroline) were prepared and characterized by single X-ray crystallography. The investigation of the complexes binding with herring sperm DNA (HS-DNA) revealed that stronger binding interaction exists between the complexes and HS-DNA compared to the corresponding free ligands. And both complexes can bind to DNA through intercalation of the phenyl rings between adjacent base pairs in the double helix. Meanwhile, bovine serum albumin (BSA) binding studies reveal the complexes can effectively interact with BSA and change the secondary structure of BSA. Superoxide dismutase (SOD) mimic activity of the complexes was investigated and the results showed that complex 2 showed a stronger radical scavenging potency against O₂ ^.- radical with an IC₅₀ value of 16.10 μM. The catecholase-like activity of both complexes has been investigated with 3,5-di-tert-butylcatechol (3,5-DTBC) as the substrate in methanol solution under completely aerobic conditions. The catalytic reaction follows Michaelis-Menten enzymatic reaction kinetics with turnover numbers (k_cat) of 34.14 h^?1 and 30.36 h^?1 for complex 1 and complex 2, respectively.

Full text of DOI:10.1016/j.molstruc.2019.127229

Journal Pre-proof
The synthesis, characterization, DNA/protein interaction, molecular docking and
catecholase activity of two Co(II) complexes constructed from the aroylhydrazone
ligand
Yun Li, Yueqin Li, Xiaohui Liu, Yong Yang, Dong Lin, Qinwei Gao
PII:
S0022-2860(19)31338-9
DOI:
https://doi.org/10.1016/j.molstruc.2019.127229
MOLSTR 127229
Reference:
To appear in: Journal of Molecular Structure
Received Date: 31 August 2019
Revised Date: 8 October 2019
Accepted Date: 13 October 2019
Please cite this article as: Y. Li, Y. Li, X. Liu, Y. Yang, D. Lin, Q. Gao, The synthesis, characterization,
DNA/protein interaction, molecular docking and catecholase activity of two Co(II) complexes constructed
from the aroylhydrazone ligand, Journal of Molecular Structure (2019), doi: https://doi.org/10.1016/
j.molstruc.2019.127229.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
©
2019 Published by Elsevier B.V.

The synthesis, characterization, DNA/protein interaction,
molecular docking and catecholase activity of two Co(II)
complexes constructed from the aroylhydrazone ligand
a
*a,b
a
a
a
*a,b
Yun Li , Yueqin Li , Xiaohui Liu , Yong Yang , Dong Lin , Qinwei Gao
a
College of Chemical Engineering, Jiangsu Key Lab for the Chemistry & Utilization
of Agricultural and Forest Biomass, Nanjing Forestry University, Nanjing 210037,
China
b
Co-Innovation Center of Efficient Processing and Utilization of Forest Resources,
Nanjing Forestry University, Nanjing 210037, China
Corresponding author email: yueqinli@njfu.edu.cn, gqw@njfu.edu.cn
Abstract
Two Co(II) complexes, Co(BBH)phenMeCN (1) and Co(BSH)phenMeCN (2)
(HBBH
=
benzophenone benzoyl hydrazone, HBSH
=
benzophenone
salicylylhydrazone ligand and phen = 1,10-phenanthroline) were prepared and
characterized by single X-ray crystallography. The investigation of the complexes
binding with herring sperm DNA (HS-DNA) revealed that stronger binding
interaction exists between the complexes and HS-DNA compared to the
corresponding free ligands. And both complexes can bind to DNA through
intercalation of the phenyl rings between adjacent base pairs in the double helix.
Meanwhile, bovine serum albumin (BSA) binding studies reveal the complexes can
effectively interact with BSA and change the secondary structure of BSA. Superoxide
1

dismutase (SOD) mimic activity of the complexes was investigated and the results
.
-
showed that complex 2 showed a stronger radical scavenging potency against O2
radical with an IC50 value of 16.10 ꢀM. The catecholase-like activity of both
complexes has been investigated with 3,5-di-tert-butylcatechol (3,5-DTBC) as the
substrate in methanol solution under completely aerobic conditions. The catalytic
reaction follows Michaelis-Menten enzymatic reaction kinetics with turnover numbers
-1
-1
(
k ) of 34.14 h and 30.36 h for complex 1 and complex 2, respectively.
cat
Keywords: Cobalt(II) complexes; DNA binding; BSA interaction; SOD mimic
activity; catecholase-like activity
2

1
. Introduction
Cobalt complexes of a Schiff base ligand containing mixed donors have attracted
much attention because they have shown diverse coordination sphere and promising
anticancer activity [1, 2]. In particular, many cobalt complexes bearing typical Schiff
bases, such as hydrazones, semicarbazones, and thiosemicarbazones, as well as their
derivative analogs, have been reported with the most structurally characterized
showing a wide range of biological applications, such as antibacterial, antioxidant,
antiproliferative, and antitumor, etc [3-13]. These Schiff base compounds give not
only new complexes with important pharmacological properties but also exhibit
interesting and varying coordination modes in their complexes. More significantly,
various hydrazone compounds are obtained as biologically active compounds and
their corresponding metal complexes are more active candidates for applications in
biology and medicine. Recently, a number of attempts have been made to obtain a
wide variety of cobalt complexes containing polydentate hydrazone ligands and
investigate the biological activity of the new complexes. For example, cobalt complex
incorporating
acetohydrazide ligand was synthesized and showed impressive antimicrobial activity
14]. Cobalt-hydrazone complexes bearing phenyl-2-pyridylketon-2-quinolinhyrazone,
with
N'-((2Z,
3E)-3-(hydroxyimino)butan-2-ylidene)-2-phenyl
[
methyl-2-pyridylketone-2-quinolinhyrazone,
pyridine-2-aldehyde-2-quinolylhydrazone have proved to act as more potent
anticancer agents against colon carcinoma cells and breast carcinoma cells than the
corresponding hydrazone ligands and the metal salt [15]. Cobalt complexes bearing
3

2
-benzoylpyridine-4-methoxybenzhydrazone were also investigated to have better
biological activity than the metal free ligand against the bacterial and fungal species
16]. Govindarajan et al. also reported water-soluble cobalt(II) and cobalt(III)
[
complexes supported by aroylhydrazone ligands [17], which exhibited strong DNA
and protein binding activities and effective cytotoxicity against human cervix
epithelia carcinoma cells and breast carcinoma cells.
Besides, the cobalt complexes with non-copper centre is mimicking an enzyme with a
dicopper active site, also showing promising catalytic oxidation performance [18].
Cobalt complexes have been regarded as very efficient and selective catalysts towards
different organic transformations, such as o-diphenols (catechols) to the
corresponding quinones and phenoxazinone synthase [18]. In this respect, Cobalt
complexes as catalyst has received considerable attraction to reveal mechanistic
aspect, catalytic activity, product formation selectivity, structure-reactivity
relationship and so on [19-21]. Recent reports with structurally characterized catalyst
and catalytic turnover show that phenoxo containing Schiff base ligands are good to
provide efficient catechol oxidation due to the relatively more hard nature of cobalt
being stabilized by the phenoxo/hydroxo type O-donors [22, 23]. However, cobalt
complexes containing phenoxo hydrazone and phenanthroline mixed donors as
biomimetic models of catechol oxidase have not been extensively studied.
The biological and pharmacological properties of metal complexes not only depend
on the properties of the metal centre but also affect by molecular structure of the
ligands. In regard to structurally characterized cobalt-Schiff base complexes, cobalt(II)
4

complexes are the most studied because of their aqueous stability, availability, and
ease of synthesis. The substituent of the aroylhydrazone moiety has shown significant
impact on the biological activity of the complexes [24, 25]. In this regard, we have
recently reported cobalt(II) complexes derived from 2-acetonaphthone benzoyl
hydrazone and 2-acetonaphthone salicylylhydrazone Schiff bases and their biological
properties such as DNA binding, protein binding, antioxidative and cytotoxic activity
[
6, 7]. However, there are limited studies about the structural and biological behaviors
of transition metal complexes derived from benzophenone benzoyl hydrazone ligand
HBBH) and benzophenone salicylylhydrazone ligand (HBSH) [26]. No data is
(
available regarding biological activities of cobalt(II) complex based on HBBH and
HBSH ligands. This aroused our interest in the preparation of new cobalt complexes
with a view toward evaluating the interaction behaviors of these compounds with
DNA, protein, and to explore their catalytic activity in oxidation reactions. Herein,
mononuclear cobalt complexes, namely Co(BBH) PhenMeCN (1) and
2
Co(BSH) PhenMeCN (2), were synthesized and fully characterized. The interaction
2
of complexes with DNA and albumin protein has been studied, which showed
promising results with very high affinity towards DNA and protein. Moreover, two
Co(II) complexes were investigated for possible catechol like activity and particularly,
complex 1 has shown very high activity towards the catalytic oxidation of
3
,5-di-tert-butylcatechol (3,5-DTBC).
5

2
2
. Experimental
.1. Materials and instruments
The hydrazone ligands, benzophenone benzoyl hydrazone ligand (HBBH) and
benzophenone salicylylhydrazone ligand (HBSH) were synthesized according to our
previous work [26]. Herring sperm DNA (HS-DNA), tris (hydroxymethyl)
aminomethane hydrochloride (Tris-HCl), ethidium bromide (EB, chemical structure
shown in Fig. S1), 1,10-phenanthroline (phen), bovine serum albumin (BSA), and
3
,5-di-tert-butylcatechol (3,5-DTBC) were purchased from Aladdin Industrial
(Shanghai, China). Other chemicals and solvents were obtained from commercial
sources as reagent grade and used without further purification.
Melting points were determined with WRS-1B melting point instrument with 1
o
C/min gradient, and the results were uncorrected. FT-IR spectra on KBr pellets were
conducted on a PerkinElmer Spectrum Two spectrophotometer within the range of
-1
4
00 - 4000 cm . Thermogravimetric analysis (TGA) was perfromed using TA
Q5000IR in a ceramic crucible under dynamic nitrogen atmosphere (150 mL/min)
o
o
from laboratory temperature to 800 C (gradient 10 C/min). The ultraviolet-visible
(UV-vis) absorption spectrum was undertaken by a Varian Cary 300 UV-vis
spectrophotometer. Fluorescence spectroscopy was performed by a PerkinElmer
LS-55 luminescence spectrometer at room temperature.
6

2
.2. Synthesis of the complexes
Synthesis of Co(BBH) PhenMeCN (1). HBBH (43.94 mg，0.05 mmol) was dissolved
2
in methanol (4 mL) with the presence of triethylamine (15 ꢀL), to which a solution of
CoCl ·6H O (11.9 mg，0.05 mmol) with 3 mL of MeCN was added dropwise. The
2
2
o
reacting solution was then stirred at 60 C for 3 h. After cooled, filtered and kept the
filtrate under slow evaporation for about a week, a reddish-brown crystalline solid
was observed. The isolated crystals was washed carefully with cold ethanol, and dried
o
in a desiccator over night. Yield: 62%. Melting point: 235.0 C. FTIR (KBr disks,
-1
cm ): 1588, 1514 (-C=N-N=C-), 1361(C-O), 1022 (N-N). ESI-MS (MeOH) (Fig. S2):
+
m/z = 838.25 (calculated m/z = 838.24 for [Co(BBH) Phen +H] ). UV-vis, λ (nm)
2
max
-1
-1
(
ε, M cm ) (Tris-HCl buffer): 231 (67000), 270 (59440), 294 (56000).
Synthesis of Co(BSH) PhenMeCN (2). The synthetic procedure was similar to that of
2
o
-1
complex 1. Yield: 45%. Melting point: 299.4 C. FTIR (KBr disks, cm ): 1590, 1530
(-C=N-N=C-), 1368 (C-O), 1023(N-N). ESI-MS (MeOH) (Fig. S2): m/z = 689.15
+
-1
-1
(
calculated m/z = 689.16 for [Co(BSH) ] ). UV-vis, λ (nm) (ε, M cm ) (Tris-HCl
2
max
buffer): 231 (57680), 270 (42760), 297 (40270), 349 (28520).
X
X
O
N
N
N
N
N
N
O
Phen, MeOH/MeCN, CoCl 6H O
2
2
H
Co
o
Et N, 60 C
X
O
3
N
N
MeCN
X= H Co(BBH) PhenMeCN (1)
2
OH Co(BSH) PhenMeCN (2)
2
7

Scheme 1 The synthetic route of the cobalt(II) complexes.
2
.3. X-ray diffraction data
Single-crystal X-ray diffraction test for complexes were performed on an Oxford
Diffraction Xcalibur Eos Gemini diffractometer at 298 K using graphite
monochromated Mo-Kα radiation (λ = 0.71069 Å). Data processing was undertaken
with CrysAlisPro. The structures were solved by direct methods and refined on F2 by
full-matrix least-squares procedures using the SHELXL-97 programs. All
non-hydrogen atoms were refined anisotropically and the hydrogen atoms were
included in idealized position. The crystallographic data along with
structure-refinement parameters for the complexes are listed in Table 1.
Table 1 Crystallographic data and refinement parameters.
complex 1
complex 2
Formula
Formula weight
Crystal system
Space group
a (Å)
C H CoN O ·C H N
C H CoN O ·C H N
52
38
6
2
2
3
52 38
6
4
2
3
879.87
910.87
Monoclinic
Monoclinic
P2 /c
P2 /c
1
1
17.7361 (9)
13.6452 (5)
19.6822 (9)
107.393 (5)
4545.5 (4)
4
17.7789 (16)
13.5188 (12)
19.6911 (15)
106.437 (9)°
4539.3 (7)
4
b (Å)
c (Å)
β (°)
3
V (Å )
Z value
F(000)
1832
1892
Crystal size (mm)
0.25 × 0.2 × 0.18
1.286
0.28 × 0.18 × 0.15
1.333
−
3
D (Mg m )
x
8

λ (Mo-Kα) (Å)
θ (°)
0.71073
3.9-25.0
0.43
0.71073
3.1-24.8°
0.43
-1
ꢀ
(Mo-Kα) (mm )
Temperature(K)
298
298
Rint
0.043
0.060
2
.4. DNA binding experiments
The HS-DNA stock solution with a mass concentration of 200 ꢀg/mL was prepared in
Tris-HCl stock solution (5 mM Tris-HCl/50 mM NaCl, pH=7.4). The ratio of UV
absorbance at 260 nm and 280 nm (A /A ) is 1.85, which indicates the HS-DNA
260 280
sample was sufficiently purified and essentially protein-free [27]. The molar
-4
concentration of the prepared HS-DNA solution was found to be 3.72 × 10 mol/L,
which was examined through considering the DNA bases molar extinction coefficient
-1
-1
qualed 6600 M cm at 260 nm [28]. The study compound was dissolved in
DMSO-buffer (5% DMSO and 95% Tris-HCl buffer) mixture for all experiments
unless mentioned otherwise. In the UV-Vis titration experiments, the absorbance of
the study compound (25 ꢀM) was first identified. Then the DNA (0-5 equivalents)
was gradually added to both the compound solution and the reference solution.
Absorption spectra were undertaken when the interaction get equilibration. In the EB
fluorescence displacement experiment, DNA was pretreated with EB at saturation
binding level of [DNA] = [EB] = 50 ꢀM, then a certain amount of the ligands (20-200
ꢀ
M) and the complexes (10-100 ꢀM) was added to the EB-DNA system followed by
incubation for 2 h in the dark. Fluorescence emission spectra were measured with λex=
9

5
45 nm at fixed excitation and emission slit width. Viscosity tests were conducted by
o
an Ubbelohde viscometer immersed in a thermostated bath at 37.0 ± 0.1 C. The
HS-DNA solution was of 50 ꢀM and the study compound was added with increasing
concentrations from 5 ꢀM to 25 ꢀM to get the [compound]/[DNA] ratios range from
0
.10 to 0.50 with 0.10 intervals. Flow time was measured by a digital stopwatch. Each
test was run three times to obtain an average calculated flow time. The viscosity value,
η, was obtained from the flow time of the DNA-containing solutions (t) duly
corrected for that of the free buffer (t ): η = (t - t )/t [29]. Relative specific viscosity
0
0
0
1
/3
(
η/η0) was plotted against [compound]/[DNA], where η was the specific viscosity of
DNA in the presence of compound and η was the specific viscosity of DNA alone.
0
2
.4. BSA binding experiments
The interactions with BSA were studied by fluorescence quenching experiment. BSA
-4
solution (1 ꢀM) was made in PBS buffer (50 mM, pH 7.2). The study compound (10
M) was prepared with the same method in DNA binding study. In the assay, aliquots
of the stock solution of the study compound were successively added into a 2.5 mL of
BSA solution by using a micropipette. The fluorescence emission spectra were
undertaken when the interaction get equilibration, in which emission spectra were
scanned at λ = 280 nm and monitored the emission intensity changes at λ = 343 nm.
ex
em
Synchronous fluorescence spectra were also performed by simultaneously scanning
with the wavelength interval ∆λ equalled 15 and 60 nm. For the three-dimensional
fluorescence tests, the excitation wavelength was set to range from 200 to 360 nm at
1
0

1
0 nm gradients and the emission wavelength was over a range of 240-440 nm. For
the UV-vis titration experiment, a 3 mL solution of BSA (10 ꢀM) in PBS buffer was
titrated by successive additions of the stock solutions of the study compound (2-10
ꢀ
M). Aliquots of the study compound with the same concentration were added to the
reference cuvette to overcome the compound’s absorbance.
2
.5. Molecular docking
Molecular docking molecular docking was performed using the Autodock 4.2
software [30] and the Lamarckian genetic algorithm [31]. The crystal structures of
DNA (PDB ID: 1Z3F) and BSA (PDB ID: 4F5S) were obtained from the protein data
bank (http://www.rcsb.org/) at a solution of 1.60 and 2.47 A, respectively. Before
docking with 1Z3F, the heteroatoms and ellipticine molecule in 1Z3F were removed
to expose the intercalation gap, and polar hydrogen atoms as well as Kollman charges
were added to the molecule. The coordination sphere of the complex was generated
from its X-ray crystal structure by converting the CIF file to the PDB format using the
Mercury software (http://www.ccdc.cam.ac.uk/). The receptor and “the ligand”
(complexes) files were prepared using AutoDock Tools. In the docking analysis, the
binding site was assigned to include the entire DNA fragment, which was enclosed in
3
a box with number of grid points in 52 × 48 × 48 Å and a grid spacing of 0.375 Å.
While in the docking with BSA, the blind docking was firstly run to identify the
possible binding sites by scanning the entire surface of the target (4F5S) and selects
the location with the highest binding affinity. After finding the best location by the
1
1

blind docking, the focus docking was performed on the best location. The size of the
3
grid map used for the docking of the complexes was 70 × 70 × 70 Å , respectively.
The PyMOL (The PyMOL Molecular Graphics System, Version 1.3, Schrödinger,
LLC) package was used to produce molecular images and animations.
2
.6. Superoxide dismutase (SOD) mimic activity
-
In the superoxide scavenging assay, the superoxide anion (O ˙ ) was generated by in
2
vitro nonenzymatic system, namely MET/VitB /NBT system [32, 33]. The amount of
2
-
O ˙ could be determined spectrophotometrically by NBT photo reduction method.
2
Solution of MET, NBT and VitB were made in PBS buffer (pH= 7.8) in the dark. The
2
tested compounds were prepared in N,N-dimethylformamide (DMF). The reaction
mixture contained MET (10 mM), VitB (3.3 mM), NBT (46 mM), a phosphate buffer
2
(67 mM, pH = 7.8) and the tested compound of different concentration. After
o
incubated at 30 C with illuminating by a fluorescent lamp for 10 min, absorbance is
examined at 560 nm. In the scavenging assays, the absorbance of the sample was Ai.
The solution without the tested compound was set as the control and its absorbance
was A . All experiments were done in triplicate and the results were expressed as an
c
average of three analyses. Concentrations of the tested compounds were adjusted to
get around 50% activities. The scavenging effect was calculated as follows:
scavenging effect (%) = [(A - A )/A ] ×100.
c
i
c
1
2

2
.7. Catecholase activity study
1
00 equiv of 3, 5-di-tertbutylcatechol (3, 5-DTBC) in methanol as substrate were
-4
added to 10 M solutions of complex in methanol under aerobic condition at room
temperature. UV-vis absorption spectrum was recorded at a regular time intervals of 3
min in the range of 325-550 nm. The dependence of reaction rate on substrate
concentration and various kinetic parameters were obtained by separate treatment of
-4
1
0 M solution of complex with 10, 30, 50, 70, 90 and 100 equivalent of 3, 5-DTBC
substrate, and monitoring the upsurge in absorbance at 402 nm (the peak
corresponding to the quinone band maxima) as a function of time.
3
3
. Results and discussion
.1. Structural characterization
The structures of the synthetic complexes were characterized firstly by FTIR spectra.
Fig. S3 presents the characteristic peaks of both complexes and comparison with the
free hydrazone ligands. In the free HBBH ligand, the bands at 3031, 1686 and 1671
-1
cm can be attributed to the stretching vibration of N-H, C=O and C=N groups of the
amide [34], respectively. Upon the complexation reaction, these bands are absent, and
-1
new bands appeared at 1588, 1514 and 1361 cm due to the stretching vibration of the
-
conjugated C=N-N=C-O moiety in the complex 1 [3], suggesting the coordination via
the enolate form and the azomethine nitrogen chelated to the metal center. While
compared the FTIR spectrum of free HBSH with that of the complex 2, similar results
1
3

can be observed. Specifically, broad medium intensity band appeared at 3200-2600
-1
cm region result from phenolic OH and the probably intramolecular hydrogen bonds
(O-H···O) [35].
The molecular structure of the complexes was further confirmed by single crystal
X-ray diffraction. The ORTEP drawing of both complexes was presented in Fig. 1.
Both complexes are monomeric octahedral crystals of cobalt ion, which are
crystallized in the identical monoclinic crystal systems. Two monoanionic bidentate
donors are coordinated to cobalt(II) through the azomethine nitrogens (N1 and N3)
and the ketoamide oxygen (O1 and O2) forming two five-membered chelate rings.
The two N donor atoms (N5 and N6) of 1,10-phenanthroline fulfill the other
coordination sites of octahedral geometry. Specifically, hydrogen bonds cannot be
observed in the complex 1, however, the salicylic moieties induces two intramolecular
o
hydrogen bonds through O3-H3⋯N2 (2.557 Å, 141.99 ) and O4-H4⋯N4 (2.544 Å,
o
1
44.42 ). In addition, nonbonded short contacts, including intramolecular π-π
interaction (involving the C15-C20 ring and the C47-C51 ring with centroid-centroid
contacts of 3.559 Å) and some intermolecular C-H⋯π interactions are found within
the packing diagram of complex 1, as shown in Fig. 1b. Similar nonbonded short
contacts are also found within the packing diagram of complex 2, including π-π
interactions between the C9-C13 benzene ring and C48-C52 phen ring
(centroid-centroid distance = 3.571 Å) and the C27-C22 benzene ring and C41-C45
phen ring (centroid-centroid distance = 4.186 Å), as well as some more C-H⋯π
interactions, as shown in Fig. 1d. The comparison for typical bond length and angle
1
4

of the two complexes were summarized in Table 2. Particularly, the C1-N2 distance
1.286 Å, Table 2) reveals that the bond is significantly shorter than the N-C single
(
one and is a character of the C=N double bond in the amide group [36], also
confirming the coordination via the enolate form. What’s more, according to the
chelate bite angles of the central cobalt atom (see Table 2), it can be concluded that
the cobalt ion in complex 2 lies in a more distorted octahedron coordination sphere.
Fig. 1 ORTEP drawing of complex 1 (a) and complex 2 (c) omitted the
non-coordinated solvent molecule MeCN for clarity and the intramolecular π-π
stacking interactions with centroid-centroid contacts, H-bonds and intermolecular
1
5

C-H⋯π interactions within the single unit cell of complex 1 (b) and complex 2 (d).
Table 2 Comparisons of bond distances and bond angles for two complexes.
Bond lengths (Å)
Co1-O1
Complex 1
2.0160(15)
2.0068(16)
2.1708(18)
2.2069(18)
2.2265(18)
2.114(2)
Complex 2
2.012(3)
2.017(3)
2.209(4)
2.188(4)
2.120(3)
2.222(3)
1.399(4)
1.286(5)
1.317(6)
Co1-O2
Co1-N1
Co1-N3
Co1-N5
Co1-N6
N1-N2
1.408(2)
C1-N1
1.286(3)
N2-C14
1.308(3)
Angles (°)
N1-Co1-N3
O1-Co1-O2
O1-Co1-N5
O2-Co1-N6
N5-Co1-N6
N1-Co1-O1
N1-Co1-O2
162.77(7)
104.02(7)
165.45(7)
163.47(7)
75.14(8)
76.71(6)
90.62(7)
160.76(13)
104.07(12)
162.91(13)
166.46(13)
75.71(14)
76.80(13)
94.30(13)
3
.2. Thermal stability of the cobalt complexes
The prepared cobalt complexes were subjected to TGA scans to evaluate their thermal
stability. Fig. S4 presents the TGA curves of the cobalt complexes under nitrogen.
o
Only one stage of degradation at around 300 C was noticed for the complex 1. The
absence of the weight loss signal of the non-coordinated solvent MeCN may be due to
its evaporation under the house vacuum drying. For the complex 2, the weight loss of
1
6

about 4.7% is present at stage of 100-160 °C, which can be attributed to the loss of the
MeCN (calculated value of 4.5%). Then, a similar change occurred in the TGA curve
for major weight loss of the complex 2 at around 300 °C. The residue of pyrolysis of
both complexes is about 25%, corresponding to cobalt oxide Co O (calculated value
3
4
of 28% for complex 1 and 26% for complex 2).
3
.3. DNA binding interaction
The DNA binding ability of the metal complexes is a vital aspect to consider for
assessing their suitability as anticancer agents. Understanding the binding mode and
the binding strength are of paramount importance for describing the efficacy of the
potential drug. Therefore, the interaction between the new complexes and herring
sperm DNA were studied by using UV absorption, emission titration and viscosity
measurements. The comparative binding ability of the parent hydrazone ligands was
also investigated. The biological activity of metallodrugs is related to its stability in
solution. In this respect, the stability of the tested compounds was estimated in
phosphate buffer-DMSO (99:1) mixture and in Tris-HCl-DMSO (99:1) mixture by
recording their electronic spectra over a period of two hours, as shown in Fig. S5. The
obtained results indicate that they are stable in aqueous media where all the in vitro
biological screening was carried out.
Fig. 2 showed the absorption spectral characteristics of the hydrazone liagnds and the
corresponding complexes in the absence and presence of HS-DNA. The absorption
1
spectra of free HL showed a sharp peak at 230 nm and an intense absorption band at
1
7

3
00 nm, corresponding to the π-π* transition of the benzenoid moiety. However, an
2
extra shoulder peak can be observed at 350 nm for the HL , which is due to the
intraligand π-π* transition. For the cobalt complexes 1 and 2, the above-mentioned
bands existed at the similar position indicating the ligands participated in coordination
with the cobalt(II) ion. The new appearance of the band at around 270 nm is attributed
to the absorbance of the Phen due to the π-π* transition. With subsequent addition of
DNA, the visible bands in the UV-vis spectra showed significantly decreased with
1
2
hypochromism up to 18.7%, 10.9%, 17.9% and 24.4% at 300 ꢁnm for HL , HL ,
complex 1 and complex 2, respectively. In order to determine the magnitude of the
binding strength of the investigated compounds with HS-DNA, the intrinsic binding
constant K values were calculated from the plot of [DNA]/[ε ꢁ-ε ] vs. [DNA], as
b
a
f
shown in the inserted figures. The K value from the ratio of the slope (1/(ε -ε )) to the
b
b
f
4
-1
3
-1
4
y-intercept (1/[K (ε -ε )]) is found to be 4.14ꢁ ×ꢁ10 ꢁM , 8.33ꢁꢁ×ꢁ10 ꢁM , 5.82×ꢁ10 ꢁ
b
b
f
-1
4
-1
1
2
M and 6.79ꢁ×ꢁ10 ꢁM for the ligand HL , ligand HL , complex 1 and complex 2
6
respectively. In comparison with K value for classical intercalator EB (K = 1.4×10
b
b
-1
M ) [37], all tested complexes exhibited lower binding constants. However, they are
comparable to some well-established DNA intercalative cobalt complexes with
4
5
acylhydrazone ligands (K = 10 ~10 ) [3, 38, 39], this signifies that the ligands and
b
complexes may bind with DNA either via classical intercalative mode or via partial
intercalative mode [40].
1
8

(
a)
1
(b)
Complex 1
-2
0
0
0
0
0
0
0
.6 HL
-10
-4
1
.5
.0
-20
-30
-40
.5
.4
.3
.2
.1
.0
-6
-8
-10
1
0
50
100
[
DNA] (µM)
0
50
100
[DNA] (µM)
0.5
0.0
2
00 250 300 350 400 450 500
200 250 300 350 400 450 500
Wavelength (nm)
Wavelength (nm)
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
-
40
60
1.5 Complex 2
-2
(c)
2
(d)
HL
-4
-
-6
-80
-8
1.0
-100
-10
0
50
[Q] (
100
0
50
100
µ
M)
[
DNA] (µM)
0.5
0.0
200 250 300 350 400 450 500
200 250 300 350 400 450 500
Wavelength (nm)
Wavelength (nm)
1
2
Fig. 2 UV-vis spectra of HL , HL and the cobalt(II) complexes in the absence and
presence of HS-DNA. Inset: plot of [DNA]/(ε -ε ) vs. [DNA].
b
f
EB is a very good intercalator, and it can show strong emission intensity in the
presence of DNA, due to its intercalation between the adjacent DNA base pairs. It has
been previously reported that the fluorescence intensity of EB-DNA system could be
quenched by addition of the metal complexes as quenchers, indicating the competition
between the complexes and EB in binding to DNA that proved the intercalation of
complexes to the base pairs of DNA [41]. Therefore, the EB displacement assay was
1
2
conducted to quantitatively estimate the binding strength of HL , HL and the
corresponding cobalt(II) complexes with HS-DNA. Fig. 3 showed the emission
spectra of EB bound to DNA as a function of added the compound concentration.
From the figure, the typical fluorescence peak of the EB-DNA system at 600 nm was
1
9

significantly decreased upon addition of the tested compounds, with the
hypochromicity up to 38.8%, 26.4%, 65.0% and 64.0% of the initial fluorescence
1
2
intensity for the HL , HL , complex 1 and complex 2, respectively. These
observations demonstrate that both the ligands and the complexes can serve as the
quencher and replace a certain amount of EB molecules from the saturated binding
EB-DNA system. The quenching efficiency for each compound can be analyzed
according to the classical Stern-Volmer equation: I /I = K [Q] + 1, in which I and I
0
q
0
are indicative of the fluorescence intensity in the absence and presence of quencher
and K represents the quenching constant. The K value from slope of the I /I vs. [Q]
q
q
0
3
-1
3
-1
4
-1
4
-1
plot is obtained as 3.83×10 M , 1.63×10 M , 1.83×10 M and 1.81×10 M for
1
2
H L , H L , complex 1 and complex 2, respectively. Additionally, the apparent
2
2
binding constant (K ) can be determined from the equation: K [complex] = K [EB]
app
app
EB
[
42], in which [complex] is equal to the value at a 50% quenching of the EB-DNA
7
-1
emission intensity, KEB = 1.0 × 10 M and [EB] = 1.3 ꢀΜ [43]. The K values is
app
5
-1
4
-1
5
-1
6
-1
found to be 1.68×10 M , 8.19×10 M , 9.15×10 M and 9.09×10 M
1
2
corresponding to H L , H L , complex 1 and complex 2, respectively. The magnitude
2
2
of Kapp values reveals a stronger affinity of the cobalt(II) complexes binding to
HS-DNA than the parent hydrazone ligands. For some of the metal complexes of
hyzrazone ligands containing Phen as ancillary ligand, the reported binding constant
3
4
-1
values lie in the 10 to 10 M range [39, 44-47], thus the new complexes have shown
very similar binding strength.
2
0

7
6
5
00
00
00
(a)
1
(b)
Complex 1
6
5
4
3
2
1
00 HL
1.6
.4
.2
.0
2.5
.0
1.5
.0
2
1
00
1
1
00
400
300
1
0
50
Q] (
100
0
50
100
[Q] (
150
200
[
µ
M)
00
µM)
2
00
00
0
00
1
00
0
5
50
600
650
700
550
600
650
700
Wavelength (nm)
Wavelength (nm)
7
6
5
00
00
00
1
.4
.3
(
c) 600
2
(d)
3.0
2.5
2.0
1.5
1.0
Complex 2
HL
1
5
4
3
2
1
00
1.2
1
.1
.0
00
00
00
00
1
0
50
100
Q] (
150
200
400
0
50
Q] (
100
[
µ
M)
[
µM)
300
200
100
0
0
5
50
600
650
700
550
600
650
700
Wavelength (nm)
Wavelength (nm)
Fig. 3 The emission spectra of the DNA-EB system in the presence of the cobalt(II)
complexes. [DNA]ꢁ = [EB]ꢁ =ꢁ 25ꢁ µM, [ligand]ꢁ=ꢁ25-200ꢁ µM, [complex]=25-100 µM.
The arrow shows the emission intensity variation upon increasing complex
concentration.
Studies of conventional intercalation mode have stressed that the intercalator can
lengthen the DNA helix as DNA base pairs are separated to accommodate the binding
compound at intercalation sites and thus cause significant enhancement in DNA
viscosity [48]. In contrast, a partial and/or non-classical intercalation that binds
exclusively in the DNA grooves may lead to a bend (or kink) of the DNA helix,
causing less pronounced or no change in DNA viscosity [49]. To clarify further
interaction mode between DNA and complexes, viscosity measurements were carried
out by varying the concentration of the added compounds to DNA. The values of
2
1

1
/3
relative specific viscosity (η/η0) versus the molecular ratio of [compound]/[DNA]
are plotted in Fig. 4. It was observed that viscosity of HS-DNA gradually increased on
addition of increasing amount of the tested compounds, which is distinctive feature of
intercalative mode of binding. These results are in accordance with those obtained in
UV-Vis spectroscopic study.
3
3
2
2
1
1
0
0
.5
.0
.5
.0
.5
.0
.5
.0
Complex 1
1
HL
Complex 2
2
HL
0
.2
0.4
0.6
0.8
1.0
[Complex]/[DNA]
1
2
Fig. 4 Effect of the addition of increasing amounts of HL , HL , complex 1 and
complex 2 on the relative viscosity of HS-DNA (100 mM), in Tris-HCl buffer (pH =
o
7
.2) at 37 C.
To get further insight of the binding nature of the first ligands and the two complexes
to DNA, molecular docking analysis was done to validate the affinity and binding
orientation of the complex to DNA. In the docking analysis, the best results were
obtained when the conformation was ranked at the lowest free binding energy. The
binding energy values are summarized in Table S1. The binding free energy value of
1
2
HL , HL , complex 1 and complex 2 was found separately at -7.74, -7.81, -8.79, -8.33
kcal/mol, which is indicative of the stronger DNA binding affinity of complexes than
the hydrazone ligands. The binding free energy values for both the complexes are
2
2

close that of ellipticine (ꢂGbinding = -8.71 kcal/mol), suggesting that both the
complexes have a comparable DNA intercalating ability with ellipticine [50]. From
the docking pose presented in Fig. 5, both liagnds and complexes interaction mainly
through intercalating the phenyl rings from the hydazone moiety in an orientation
parallel between adjacent base pairs in the double helix. It should be noted that no
1
2
H-bond was found in the DNA docking of the ligand HL , however, the ligand HL
and complexes were able to form H-bond. A hydrogen bond was found between the
2
phenolic hydroxy group of HL and the O4ʹ atom of A: DG2 residue (d = 2.800 Å).
Similarly, two hydrogen bonds are observed within O1 atom to H3 and H21 atoms of
A: DG6 (d = 1.783 Å and 2.184 Å) for the complex 1 docking with DNA, while one
hydrogen bond was found between O3 atom of complex 2 and H7 atom A: DC2 (d =
1
.689 Å). Thus, hydrogen bonding interactions play significant role in stabilizing of
the complexes in the intercalations of DNA. These results are consistent with the
DNA-binding affinity obtained from the above experimental observations. The
binding nature of the investigated cobalt(II) complexes is coincided with the reported
dicopper(II)
complexes
deriving
from
N-(2-hydroxy-5-methylphenyl)-Nʹ-
[
3-(dimethylamino)propyl]oxamide and dinuclear copper(II) complex derived from
2
-acetyl pyridine and thiosemicarbazide, they were also docked into the intercalation
gap of the 1Z3F hexamer [51-53].
2
3

1
2
Fig. 5 Docking pose of HL , HL , complex 1 and complex 2 with DNA hexamer
d(CGATCG)2.
3
.4. BSA binding studies
The structure of bovine serum albumin (BSA) is homology with human serum
albumin. Therefore, BSA is widely chosen as a model protein in the study of
protein-small molecule interactions [54]. The investigation on the interaction of the
new cobalt(II) complexes and BSA is expected to be helpful for explaining the
transportation of the drugs in vivo. Typically, BSA shows a feature of intrinsic
fluorescence due to the aromatic amino acids, tryptophan, tyrosine, and phenylalanine.
Considering their quantum yields, tyrosine and tryptophan are used experimentally.
Tryptophan and tyrosine are hydrophobic and hydrophilic aromatic amino acids
respectively; they are high sensitive to solvent polarity and the local environment. As
a result, the intrinsic fluorescence intensity of the BSA solution could change when
2
4

the organometallic drug interacts with BSA. When the BSA solution is excited at 280
nm, both amino acids tyrosine and tryptophan are excited [55]. Therefore, the intrinsic
fluorescence of BSA in the absence and presence of two complexes was recorded by
excited at λ = 280 nm, as presented in Fig. 6a and Fig. 6b. As we see, the intensity
ex
of characteristic fluorescence emission band at 346 nm decreased gradually upon
titration of the new complexes. The hypochromism for the complex 1 is 24.8% with a
blue shift of 3 nm in the emission wavelength, while for the complex 2, the
hypochromism is higher up to 43.7% and also accompanied with a typical blue shift
of 4 nm. The blue shift in the emission wavelength occurred during the fluorescence
quenching process suggests the possible changes in the secondary structure of BSA
[
56]. The quenching strength was estimated by the Stern-Volmer equation [65]:
I /I = 1+ K [Q]= 1+k τ [Q]
0
SV
q 0
Where I and I represent the fluorescence intensity in the absence and presence of the
0
title complexes, respectively. KSV refers to the Stern-Volmer quenching constant, [Q]
is denoted as the concentration of the complexes, k is the quenching rate constant of
q
-8
the biological macromolecule and τ is the life time of the fluorophore (10 s). A good
0
linear relationship of the plot of I /I vs. [Q] was obtained as inserted the upper right of
0
5
Fig. 6a and Fig. 6b. Therefore, the KSV from the slope was found to be 1.60 × 10 and
5
-1
3
.50 × 10 M for the complex 1 and complex 2, respectively. The k values (k =
q q
1
3
13
-1 -1
K /τ ) was calculated as 1.60 × 10 and 3.50 × 10 M s , which largely overrides
SV
0
1
0
-1 -1
the highest reported value for a diffusion-controlled process (2.0×10 M s ) of
biological macromolecules. These data provide evidence that the binding of complex
2
5

to BSA is caused by a predominantly static quenching procedure, in which a specific
BSA-complex] formed when the study compound interact with BSA [57].
[
Additionally, the binding constant (K ) and number of binding sites (n) were
bin
calculated using the Scatchard equation [58]:
log[(I -I)/I] = logK + nlog[Q]
0
bin
From the intercept of the double logarithm regression curve of log[(I -I)/I] vs. log[Q]
0
4
4
-1
(
Fig. 6c), the values of Kbin were obtained as 1.48×10 and 1.84×10 M for complex
1
and complex 2, respectively, which is indicative of efficient interaction with the
protein for getting transported and distributed through the organism, but also released
once it reaches its target [58]. The value of n was found to be 0.8165 and 0.7699,
respectively, which is close to 1, thus suggests that there is only a single binding site
per protein [59]. This observation is in accordance with the aforementioned data about
the involvement of a mechanism of static quenching.
5
00
00
5
4
3
2
1
00
00
00
00
00
0
(
a)
(b)
1.8
Complex 2
Complex1
1
.3
1.6
1.4
1.2
4
1.2
1
.1
.0
300
200
100
0
1
1.0
0.0
0
.0
0.5
1.0
1.5
2.0
0.5
1.0
1.5
2.0
[Q] (µ
M)
Q (µM)
300
350
400
450
500
300
350
400
450
500
Wavelength (nm)
Wavelength (nm)
2
6

(c)
Complex 1
Complex 2
Fitted data of complex 1
Fitted data of complex 2
-
-
-
-
-
-
0.2
0.4
0.6
0.8
1.0
1.2
-
6.8 -6.6 -6.4 -6.2 -6.0 -5.8 -5.6
Log [Q]
Fig. 6 Emission spectra of BSA (1.0 ꢀM; λex=280nm) as a function of concentration
of the complex 1 (a) and complex 2 (b) (0-2.0 ꢀM). Arrow indicates the effect of
metal complexes on the fluorescence emission of BSA. Insert: Stern-Volmer plot of
the fluorescence titration data corresponding to the complex. (c) Plot of log [(I -I)/I]
0
vs. log [Q].
UV-vis absorption spectra were employed to further probe the above-mentioned static
quenching process based on the variation in the nature of BSA. As shown in Fig. 7,
there are two characteristic bands observed for BSA. The first broad absorption band
appeared around 226 nm is assignable to the π-π* transition of the characteristic
polypeptide backbone structure C=O [60], and the second weak band around 278 nm
is related to the n-π* transition of aromatic amino acids: tryptophan (Trp), tyrosine
(Tyr) and phenylalanine (Phe) [60]. As the test complex is added to BSA solution, the
maximum absorbance peak at 226 nm shows an insignificant red shift accompanied
by a decrease in the intensity. These spectral changes may arise from the disturbance
of the microenvironment around the polypeptide caused by the binding of the
complex with BSA [60]. The subtle variation at 278 nm after addition of the
complexes also suggests more aromatic acid residues were buried in a hydrophobic
2
7

pocket. These results further suggest that the microenvironment of the three aromatic
acid residues and the tertiary structure of BSA were altered due to the interactions
between the complex and BSA. The outcomes also lend support for the static
interaction process.
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0
.5
.4
Complex 1
Complex 2
3
2
2
1
1
0
0
.0
.5
.0
.5
.0
.5
.0
0.3
0.2
0
0.3
0.2
0.1
0.0
0
.1
.0
0
260 280 300 320
Wavelength (nm)
260
280
300
320
Wavelength (nm)
2
00 220 240 260 280 300 320 340
Wavelength (nm)
200 220 240 260 280 300 320 340
Wavelength (nm)
Fig. 7 UV-vis absorption spectra of BSA (10 ꢀM) as a function of concentration of the
complexes (0-10 ꢀM). Arrow indicates the effect of complexes on the absorbance of
BSA.
In considering the microenvironmental and conformational (tertiary structure)
changes around aromatic chromophores in BSA monitored by the fluorescence and
UV-vis titration experiments, synchronous and three dimensional (3D) fluorescence
spectra of BSA with the absence and presence of the complexes were also carried out
to determine the subtle structural changes. In the synchronous fluorescence studies,
wavelength intervals (ꢂλ = λ -λ ) can provide information about the types of
em ex
chromophore. The spectra provide the characteristics of the Tyr residues when ꢂλ =
1
5ꢁnm, whereas suggest information on the Trp residues when ꢂλ=60ꢁnm. The shift of
the emission wavelength indicates a change in the polarity of their surroundings [61].
2
8

The influence of both complexes on the synchronous fluorescence spectra of BSA are
illustrated in Fig. 8. It is obvious that the intensity of emission maxima for both Tyr
and Trp residues gradually decreased along with the addition of complexes, which is
evidence of conformational changes. When ꢂλ = 15 nm, the hypochromicity at 286
nm was of 23.6% and 29.4% for complex 1 and complex 2, respectively. When ꢂλ =
6
0 nm, the hypochromicity of 23.7% and 33.1% at 280 nm along with noticeable blue
shift of 2 nm was observed. The minor shifts in the synchronous fluorescence spectra
of Trp residues declare an increase in hydrophobicity of the Trp residues and decrease
in their polarity.
(b)
(a)
Complex1 ∆λ=15 nm
Complex 2 ∆λ=15 nm
200
150
100
1
50
00
1
50
50
0
0
2
60
280
300
320
340
260
280
300
320
340
Wavelength (nm)
Wavelength (nm)
(
c)
(d)
Complex 2 ∆λ=60 nm
Complex 1 ∆λ=60 nm
4
3
2
1
00
400
300
200
100
0
00
00
00
0
260
280
300
320
260
280
300
320
Wavelength (nm)
Wavelength (nm)
Fig. 8 Synchronous fluorescence spectra of BSA (1 ꢀM) and titration with complex 1
and complex 2 (0-2 ꢀM) at ꢂλ = 15 nm (a, b) and ꢂλ = 60 nm (c, d).
Apart from synchronous fluorescence measurements, 3D fluorescence spectroscopy is
2
9