International Journal of Biological Macromolecules 77 (2015) 193–202
Contents lists available at ScienceDirect
International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac
Interaction of three new tetradentates Schiff bases containing N2 O2 donor atoms with calf thymus DNA Davood Ajloo a,∗ , Sajede Shabanpanah a , Bita Shafaatian a , Maryam Ghadamgahi a , Yasin Alipour a , Taghi Lashgarbolouki b , Ali Akbar Saboury c a
School of Chemistry, Damghan University, Damghan, Iran School of Biology, Damghan University, Damghan, Iran c Institute of Biochemistry and Biophysics, University of Tehran, Tehran, Iran b
a r t i c l e
i n f o
Article history: Received 27 November 2014 Received in revised form 3 March 2015 Accepted 10 March 2015 Available online 19 March 2015 Keywords: Calf-thymus DNA Schiff-base Spectroscopy
a b s t r a c t Interaction of 1,3-bis(2-hydroxy-benzylidene)-urea (H2L1), 1,3-bis(2-hydroxy-3-methoxybenzylidene)-urea (H2L2) and 1,3-bis(2-hydroxy-3-methoxy-benzylidene)-urea nickel(II) (NiL2) with calf-thymus DNA were investigated by UV–vis absorption, fluorescence emission and circular dichroism (CD) spectroscopy as well as cyclic voltammetry, viscosity measurements, molecular docking and molecular dynamics simulation. Binding constants were determined using UV–vis absorption and fluorescence spectra. The results indicated that studied Schiff-bases bind to DNA in the intercalative mode in which the metal derivative is more effective than non metals. Their interaction trend is further determined by molecular dynamics (MD) simulation. MD results showed that Ni derivative reduces oligonucleotide intermolecular hydrogen bond and increases solvent accessible surface area more than other compounds. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Schiff-bases are the important class of compounds in both medicinal and pharmaceutical fields [1,2]. In recent years, Schiffbase metal complexes have also found important applications in the biological field [3–7]. Schiff-bases play an important role in bioinorganic chemistry as they exhibit remarkable biological activity. On the other hand, nucleic acids have important influence in biological systems and carry out a broad range of biological functions. DNA is the primary intracellular target of anticancer drugs and so the interactions between small molecules and DNA causes DNA damage in cancer cells, blocking their division and resulting in cell death [8]. Therefore, the interaction of metal complexes with DNA has attracted much attention [9,10]. Among the transition metal ions, zinc and nickel are biologically active [11,12]. The knowledge of intracellular speciation of nickel is essential for deeper understanding of the toxicity mechanism of this carcinogenic metal [13,14]. However, the interaction of drug molecules with DNA has become an active research area [15]. Since DNA is the intracellular target for a wide range of anticancer and antibiotic drugs [16–18], generally, there are three interaction modes
∗ Corresponding author. Tel.: +98 0232 5233051–6; fax: +98 0232 5235713. E-mail addresses: [email protected], [email protected] (D. Ajloo). http://dx.doi.org/10.1016/j.ijbiomac.2015.03.016 0141-8130/© 2015 Elsevier B.V. All rights reserved.
between small molecules and DNA: (i) intercalative binding that small molecules intercalate into the base pairs of nucleic acids; (ii) groove binding in which the small molecules bound on nucleic acids in major or minor groove; (iii) long-range assembly on the molecular surfaces of nucleic acids [19]. The intercalative binding is stronger than other two binding modes because the surface of intercalative molecule is sandwiched between the aromatic, heterocyclic base pairs of DNA [20,21]. Tremendous interest has been attracted to interactions between transition metal complexes of morin and nucleic acids due to potential applications of the metal complexes as anticancer drugs or as complexes with other biological functions [8,22]. These studies are also important to understand the toxicity of drugs containing metal ions [23–25]. CT-DNA is a polymer of alternate sugar phosphate sequence with high polymerized skeleton. The investigation of drug–DNA interaction is important for understanding the molecular mechanism of drug action and for the design of specific DNA targeted drug. DNA binding is the critical step for DNA activity. To design effective chemotherapeutic agents and better anticancer drugs, it is essential to explore the interactions of drug with DNA. Interaction of a few ligands with DNA were investigated by different experimental methods such as UV–vis, fluorescence, CD, calorimetry, CV and computational methods such as MD, docking and QSAR techniques were investigated [26–31].
194
D. Ajloo et al. / International Journal of Biological Macromolecules 77 (2015) 193–202
In the present study, the interaction of new tetradentate Schiffbases containing N2 O2 donor atoms and its nickel complex with calf thymus DNA were investigated using various spectroscopy methods, molecular dynamics simulation and docking calculations.
experiments were performed in the wavelength range, 200–600 nm in a constant concentration of studied SBs and titrated by varying concentration of DNA ([DNA]/[compounds] = 0.00, 0.02, 0.04, 0.06, 0.08, 0.10, 0.12, 0.14, 0.16, 0.18 and 0.20) in (10 mM Tris–HCl, pH = 7.4). These solutions were incubated for 5 min.
2. Experimental 2.3. Thermal stability
2.1. Materials All materials were purchased from Sigma–Aldrich Company. All experiments involving interaction of the complexes with DNA were carried out in buffer (10 mM Tris–HCl, pH = 7.4). The DNA concentration per nucleotide was determined by absorption spectroscopy using the molar absorption coefficient (6600 M−1 cm−1 ) at 260 nm. The structure of studied Schiff base (SB)s which synthesized by inorganic chemistry laboratory of Damghan University were shown in Fig. 1. 2.2. UV–vis spectroscopy measurement UV–vis spectra were recorded on a Perkin-Elmer UV-Vis spectrophotometer model Lambda 25. Absorption titration
DNA thermal stability was investigated by indicated absorbance value versus temperature. The absorbance at 260 nm was scanned from 35 to 85 ◦ C at 5 ◦ C per min scan rate and fixed concentration ratio ([SB]/[DNA] = 0.5). The melting temperature Tm , was defined as the mid-point of transition temperature. 2.4. Fluorescence measurements Emission spectra were recorded on a Jasco spectrofluorometer model FP6200 luminescence spectrometer at 298 K. To compare quantitatively the affinity of the cited compounds to DNA, the average binding constants, Kb , were obtained by fluorescence spectroscopy [9]. Fluorescence measurements were performed on fixed amounts of SBs in the presence of different amounts of DNA. Ethidium bromide (EB) is a fluorescent probe for DNA structure which has been employed in examination of the ligand binding mode to DNA. The excitation wavelengths for 5 × 10−4 M H2L1, H2L2 and NiL2 were 276, 484 and 514 nm, respectively and average of fluorescence emission intensity was monitored for H2L1, H2L2 and NiL2 as 250–600 nm, 470–550 nm and 500–550 nm, respectively. 2.5. Viscosity measurement Viscosity was measured by an Ostwald viscometer immersed in a thermostatic water-bath maintained to 25.0 ◦ C. In all samples, the DNA concentration was kept constant (1 × 10−8 M). Calculation was carried out using Eq. (1); 0 = =
tDNA − t0 t0
t − t0 t0
(1a) (1b)
where 0 and are the viscosity of DNA in the absence and presence of SB, t0 , tDNA and t are the observed flow time of buffer, DNA and DNA containing solution upon the addition of SBs, respectively [32]. The values of (/0 )1/3 versus the concentration ratio of SB to DNA were plotted. 2.6. Circular dichroism Cyclic voltammetry measurements were done using Potentiostat/Galvanostat Autolab. The CD spectra are quite sensitive to the changes in the secondary structure of nucleic acids, which any conformational modification of DNA provoked by its interaction with SBs. CD spectrum of each sample was scanned in the ranges of 220–320 nm. The concentration of DNA was 1.0 × 10−3 M. A CD spectrum was generated by subtracting the CD spectrum of the native DNA and mixture of DNA-SB from the CD spectrum of the buffer and buffer-SB solutions. 2.7. Cyclic voltammetry
Fig. 1. Chemical structure of three studied SBs.
Circular dichroism spectra were recorded by AVIV circular dichroism spectrophotometer model 215. Cyclic voltammetry measurements were carried out by glassy carbon as working electrode, a platinum wire as an auxiliary electrode, and
D. Ajloo et al. / International Journal of Biological Macromolecules 77 (2015) 193–202
195
Ag/AgCl/3 M KCl as a reference electrode. Cyclic voltammograms of the 5 × 10−5 M SBs at 25 ◦ C were obtained in 60 mV s−1 scan rate. 2.8. Molecular docking AutoDock 3.0.5 program was used to obtain the energetic and binding site for the interaction of the titled Schiff-bases with DNA. Docking simulation was done in a box with dimension of 52 × 100 × 80 points and the spacing of 0.375 A˚ [33–35]. Maximum number of generations and energy evaluation were set to 27,000
Fig. 3. (a) Variation of maximum absorbance and (b) derivative of maximum absorbance for 5 × 10−5 M of CT-DNA in different temperatures in the absence (䊉) ), H2L2 ( ) and NiL2 ( ) SBs. and presence of 1 × 10−4 M H2L1 (
and 2.5 × 105 , respectively. For each run, 250 docking energies were reported with the initial population of 250 individuals. 2.9. Molecular dynamics simulation
Fig. 2. Absorption spectra of 1 × 10−4 M of (a) H2L1, (b) H2L2 and (c) NiL2 in Tris–HCl buffer upon addition of 0–200 L of calf-thymus DNA (1 × 10−3 M). [DNA]/[SB] = 0.00, 0.02, 0.04, 0.06, 0.08, 0.10, 0.12, 0.14, 0.16, 0.18 and 0.20. Arrows show the absorbance changing upon increasing DNA concentrations. Inset: plots of [DNA]/(εa − εf ) versus [DNA] for the titration of compounds with the DNA.
The structures of three Schiff-bases were drawn using Hyperchem 7 [36] software. Force field parameters and geometries of ligands were generated using PRODRG2 server [37] and they were changed to be adaptable with selected force field. The parameters of Ni metal were also inserted manually with parameters calculated from Gaussian software. The MD simulations were performed by the AMBER [38] force field and TIPP4 model was used for water molecules. A twin range cutoff was used for longer-range interactions: 0.9 nm for van der Waals interactions and 0.9 nm for electrostatic interactions. The PME [39] was used for calculating long-range interaction. The starting structure of DNA was constructed based on the X-ray crystal structure of it (PDB ID: 453D). A cubic simulation box of the volume 315 nm3 was made and 8 molecules of Schiff-bases were placed randomly in this box respectively. Then water molecules were randomly added into the simulation box and initial configurations were minimized using steepest descent algorithm with 5000 integration step and the system was
196
D. Ajloo et al. / International Journal of Biological Macromolecules 77 (2015) 193–202
Hypochromism and hyperchromism are both spectral feature of DNA concerning changes in its double helix structure. The UV–vis titration spectra of the SBs was shown in Fig. 2. This figure shows the absorption spectra in the absence and presence of calf thymus DNA. The addition of calf thymus DNA to solutions of compounds exhibited hyperchromicity which might be ascribed to the fact that the SBs could uncoil the helix structure of DNA and made more bases embedding in DNA [41]. Also hyperchromism may probably be due to dissociation of aggregated ligand or external contact with DNA. A similar hyperchromism has been observed for the Soret bands of certain porphyrins when they interact with DNA [42] that shows the compounds could bind to DNA by intercalation. The change in the absorbance when the binding sites on DNA are occupied by ligand, A was determined by
Fig. 4. The fluorescence spectra of 5 × 10−4 M (a) H2L1, (b) H2L2 and (c) NiL2 complex in the presence of DNA. Inset: plot log((F − F0 )/F0 ) versus log[Q] for titration of SBs by increasing concentration of CT-DNA.
equilibrated for 30 ns at constant pressure (1 atm) and temperature (300 K) using the Parilleno-Rahman procedure. The resulting system of MD models contains DNA and Schiff-base. All MD simulations were carried out using the GROMACS 4.5.4 package [40]. The calculations were performed using 5 quad core parallel computers including 40 processor units. The computer applied the Rocks cluster networking and Centos operating systems and the repeated trajectory showed similar result. The 8 molecules of SB were used in the 5145, 5132 and 5111 water molecules for H2L1, H2L2 and NiL2, respectively. 3. Results and discussion 3.1. UV–vis measurement Electronic absorption spectroscopy is an effective method to examine the binding mode of DNA with metal complexes.
Fig. 5. The emission spectra of 60 M DNA in the presence of different concentration ratio, [EB]/[DNA], and 0.25 mM of (a) H2L1, (b) H2L2 Ligands and (c) NiL2 complex in Tris–HCl buffer. Inset: Fluorescence Scatchard plots for binding of EB to DNA in ), 0.25 mM ( ) and 0.3 mM ( ) of (a) H2L1, (b) the presence of 0.2 mM ( H2L2 Ligands and (c) NiL2 complex.
D. Ajloo et al. / International Journal of Biological Macromolecules 77 (2015) 193–202
2 NiL2 H2L2 H2L1
1.8
(η/ηo)1/3
1.6 1.4 1.2 1 0.8 0
0.02
0.04
0.06
[Schiff base]/[DNA] Fig. 6. Effect of increasing amounts of 5 × 10−4 M of the (a) H2L1, (b) H2L2 and (c) NiL2 on the relative viscosity of 10−3 M calf-thymus DNA at 25.0 ◦ C.
A = |ADNA-SBs − ADNA − ASBs | and the spectral overlap of CT-DNA and SBs was eliminated. The average binding constant of the compound-DNA was calculated by Eq. (2) [43]: [DNA] [DNA] 1 = + (εb − εf ) K(εb − εf ) (εa − εf )
(2)
where [DNA] is the base pair concentration, εa , εb and εf are apparent, bound and free absorption coefficient, respectively. Binding constant, Kb , was obtained by dividing the slope to intercept of the [DNA]/(εa − εf ) versus [DNA] plot. The value of average binding constant for H2L1, H2L2 and NiL2 was obtained as 1.4 × 104 , 2.5 × 104 and 3.3 × 104 M−1 , respectively. 3.2. Thermal stability of DNA Structure of DNA is stable due to presence of intermolecular hydrogen bond. These stability and hydrogen bond will be
diminished by increasing temperature. By increasing the temperature, the helix structure is dissociated to single strand. The absorption data versus temperature was monitored in Fig. 3. In the presence of complex, by increasing temperature, hydrogen bond will be broken and subsequently the ligand diffuses between two strands. On other words, SB acts same as adhesive that connect the two strands stronger than hydrogen bond between base pairs. Whatever the interaction between reagent and docking site become stronger, the stability increases. Strength of interaction can be found from docking energy. As it comes in the next section docking energy is a good criterion for this purpose. The Tm value of DNA in the absence of complexes is recorded at 64.5 ◦ C. Values of Tm for DNA in the presence of H2L1, H2L2 and NiL2, were obtained about 64.5, 70.0 and 70.0 ◦ C, respectively. Therefore, by addition of cited SBs, Tm value of DNA increases and confirms the intercalation mode of the SB interaction. 3.3. Fluorescence results To further investigate the interaction mode between the SB and CT-DNA, fluorescence titration experiments was performed in two different ways: (a) titration of SB by DNA and (b) titration of DNA-EB by SBs. In the case (a) if it is assumed that there are similar and independent binding sites, in the biomolecule, the binding constant (Kb ) and the number of binding sites (n) can be determined according to the method described by [44]: log
F − F 0
F0
(a)
-2 237
7 μM H2L2 8
257
277
297
317
337
14 μM H2L2 21 μM H2L2
[Ɵ] mdeg
[Ɵ] mdeg
21 μM H2L1
(3)
(b)
14 μM H2L1
3
= log Kb + n log[Q]
Here, F0 and F are the fluorescence intensities of the fluorophore in the absence and presence of different concentrations of CTDNA, respectively. By plotting the (log(F − F0 )/F0 ) versus log[Q] (Fig. 4), the values of Kb and n were found to be, 1.5 × 103 , 1.6 × 103 ,
7 μM H2L1 8
197
3
-2 237
257
277
297
317
337
-7
-7 -12
-12
Wavelength (nm)
Wavelength (nm)
15 (C)
21 μM NiL2
3 -2 237
257
277
297
317
-7
337
DNA NiL2 H2L2 H2L1
(d)
10
14 μM NiL2
[Ɵ] mdeg
[Ɵ] mdeg
8
7 μM NiL2
5 0 -5
233
253
273
293
313
333
-10
-12
-15
Wavelength (nm)
Wavelength (nm)
Fig. 7. CD spectra of 5 × 10−5 M of CT-DNA in the absence and presence of 7, 14, 21 M of (a) H2L1, (b) H2L2 and (c) NiL2. (d) Comparison between CD spectra of DNA in the absence and presence of three SBs.
198
D. Ajloo et al. / International Journal of Biological Macromolecules 77 (2015) 193–202
1.8 × 103 and 0.902, 1.704, 0.783 for H2L1, H2L2 and NiL2, respectively. The data confirmed more and stronger interaction of the NiL2 complex with DNA. The emission spectra for titration of DNA-EB by SBs were illustrated in Fig. 5 in which the lowest spectrum is related to EB. By addition of DNA, the emission intensity of the EB was enhanced to the maximum peak. Then, emission intensity of the DNA–EB system decreases by the increase of SB concentration, which indicated that SB could displace EB from the DNA–EB system. In order to find the binding mode, the Scatchard plots were used. Therefore, rc /Cf values were obtained according to the following equations and plotted against r for all SBs bound to DNA. r = Kb (n − r) = Kb n − Kb r cf cb = ct [(F − F0 )/(Fmax − F0 )]
(4)
cf = ct − cb where ct , cf and cb are the total, free and bound SB concentration, respectively. F is the observed fluorescence emission intensity at given DNA concentration, F0 is the fluorescence intensity in the absence of DNA, and Fmax is the fluorescence intensity of the totally bound SB. The n is the binding site number that is 0.83, 0.51 and 0.26 for H2L1, H2L2 and NiL2, respectively, and Kb values are 0.65 × 104 , 0.96 × 104 and 1.01 × 104 . By drawing the fluorescence Scatchard plots (Fig. 5), we may achieve four behaviors for the binding of complex to DNA. In this experiment, by data analyzing, type D behavior was observed. Noncompetitive inhibition of EB binding (type D behavior) produces a Scatchard plot in which the slope is almost constant in the presence of increasing amounts of SB. In this kind of binding curve, the SB is attached to DNA via the groove/covalent or electrostatic interactions and sterically prevents the binding of EB to the latter [45]. 3.4. Viscosity results In general, interactions between the complexes and DNA were further clarified by viscosity measurements. Effect of the SBs on the viscosity of DNA at 25 ◦ C was shown in Fig. 6. The viscosity of DNA enhances by increasing in the ratio of SBs to DNA, so that the variation is significantly higher in the presence of Ni complex. In other words, all the three SBs can intercalate between DNA base pairs, that leading an extension in the helix, and thus increase the viscosity of DNA. Furthermore, the results indicated the intercalative binding mode. Also the nickel complex can intercalate more strongly than other SBs that causing increases in the flow viscosity. 3.5. Circular dichroism results Circular dichroism spectroscopy gives us useful information on how the conformation of DNA is influenced by the binding of the metal complex to DNA. Effect of SB on DNA secondary structure was studied by keeping the concentration of CT-DNA constant while varying the concentration of SBs in 10 mM Tris buffer. The CD spectra of DNA taken after incubation of the SBs with CT-DNA are shown in Fig. 7. The observed CD spectrum of calf-thymus DNA consists of a positive band at 274 nm due to base stacking and a negative band at 245 nm due to helicity. While the groove binding and electrostatic interaction of small molecules with DNA shows little or no perturbations on the base stacking and helicity bands, intercalation enhances the intensities of both bands [46,47]. Also the results confirmed probability of DNA structure in right-handed B form (Fig. 7) [48]. The comparison between CD spectra of DNA in the presence of SBs were shown in Fig. 7d. It shows that nickel complex intercalate more than other SBs.
Fig. 8. Cyclic voltammograms of 5 × 10−5 M of (a) H2L1, (b) H2L2 Ligands and (c) NiL2 Complex in the absence and presence of different concentrations of DNA 10−3 M in10 mM Tris–HCl (pH = 7.4) at 60 mV s−1 scan rate.
3.6. Electrochemical results The application of electrochemical methods to the study of metal complex intercalation and coordination of transition metal complexes to DNA provides a useful complement to the previously used methods of investigation, such as UV–vis spectroscopy [49]. Fig. 8 shows the electrochemical measurements in the absence and presence of CT-DNA. It can be seen that in all samples the cathode peak current gradually decreased due to the addition of DNA. The decrease in current may be attributed to the diffusion of the SBs bound to the large, slowly diffusing DNA molecule. Also the decreases in the peak currents are ascribed to the stronger binding between the SBs and DNA. Moreover, during DNA addition the cathodic peak potential showed small positive shift [50]. These positive shifts are confirming exciton
D. Ajloo et al. / International Journal of Biological Macromolecules 77 (2015) 193–202
199
Fig. 9. (a) Structure of DNA taken from protein data bank (PDB: 453D) and distribution of all cluster ranks for H2L2. (b) The most negative docking site for three SBs in the minor groove, (c) expanded view of the most negative docking site, another view of location H2L1 (d), H2L2 (e), NiL2 (f) in minor groove. The hydrogen bonds between SBs and DNA were shown by green dashed line.
type interaction and intercalation in SB in which the decreased value was more in NiL2 than the other SBs. Probably, system of NiL2 in DNA has more binding affinity than other compounds.
Table 1 Docking energies for a few cluster ranks obtained by Autodock. H2L1
3.7. Docking results Autodock software calculates docking energies and finds the most probable binding sites. After preparation of ligand and DNA input files, docking calculation is performed to search and find the best binding site. In each run, 250 values for docking free energies were obtained and sorted based on increasing in energies. Some of 250 sites which have nearly equal range of docking energies, form a cluster based on their RMSD values. The first twelve clusters, including the lowest and mean docking energies were listed in Table 1. Results show that trend of negative binding energy is H2L1 < H2L2 < Ni2L2. Fig. 9a shows the electrostatic surface structure of DNA taken from protein data bank (PDB 453D) and distribution of docking
1 2 3 4 5 6 7 8 9 10 11 12
H2L2
NiL2
LDE
MDE
NSC
LDE
MDE
NSC
LDE
MDE
NSC
−8.70 −8.65 −8.48 −8.32 −8.31 −8.01 −7.88 −7.75 −7.70 −7.55 −8.70 −8.65
−8.70 −8.65 −8.48 −8.32 −8.31 −8.01 −7.88 −7.75 −7.70 −7.55 −8.70 −8.65
1 1 1 1 1 1 1 1 1 1 1 1
−9.00 −8.90 −8.70 −8.62 −8.58 −8.57 −8.53 −8.52 −8.51 −8.45 −9.00 −8.90
−9.00 −8.90 −8.37 −8.62 −8.58 −8.34 −8.43 −8.52 −8.47 −8.45 −9.00 −8.90
1 1 4 1 1 2 2 1 3 1 1 1
−9.15 −9.14 −9.06 −8.95 −8.93 −8.72 −8.64 −8.63 −8.56 −8.52 −8.47 −8.45
−9.15 −9.04 −8.95 −8.86 −8.88 −8.72 −8.64 −8.63 −8.53 −8.42 −8.47 −8.41
2 10 14 8 5 1 1 1 10 4 1 38
LDE, lowest docking energy; MDE, mean docking energy; NSC, number of sites in a cluster.
200
D. Ajloo et al. / International Journal of Biological Macromolecules 77 (2015) 193–202
Table 2 Docking energies for a few cluster ranks of DNA (3FT6) obtained by Autodock. Rank
1 2 3 4 5 6 7
H2L1
H2L2
NiL2
LDE
MDE
NSC
LDE
MDE
NSC
LDE
MDE
NSC
−6.90 −6.87 −6.73 −6.72 −6.30 −5.97 −5.95
−6.50 −6.40 −6.49 −6.27 −6.26 −5.97 −5.95
7 9 9 21 2 1 1
−7.30 −7.26 −7.19 −7.15 −6.99 −6.90 −6.84
−6.96 −7.11 −6.39 −6.75 −6.99 −6.88 −6.79
12 9 7 4 1 4 2
−7.43 −7.40 −7.36 −7.33 −7.31 −7.30 −7.28
−7.43 −7.39 −7.36 −7.33 −7.31 −7.30 −7.28
45 44 6 1 2 1 1
sites for all cluster ranks belong to H2L2. As we see the probability of distribution in the minor groove is higher than other sites. Fig. 9b shows docking sites of the most negative docking energies for three SBs that superimposed on each other. Expanded view of docking site in stick form as well as the labels of the nearest nucleoside was also shown in Fig. 9c. Fig. 9d–f shows another view of three SBs in the minor groove that the hydrogen bonds were shown by dashed green lines. The H2L1 forms three hydrogen bonds as (DRG1 OAH. . .DA17:N3, DRG1 HAC. . .:DA18:O4 and DRG1 HAT. . .DT8:O2), H2L2 one hydrogen bond (DG10:N2 . . .:DRG1:HAO) and NiL2 dose not form any hydrogen bond. On the other hand, another DNA structure was taken from PDB bank (3FT6) in which the ligand had been intercalated between base pairs. Then the ligand was removed and docking calculation was done with SBs. The new results were listed in Table 2 and related docking sites shown in Fig. 10. Fig. 10a–e shows the structure of
Fig. 11. Calculated (a) RMSD, (b) hydrogen bond (H-bond) and (c) accessible surface area (ASA) by molecular dynamics for DNA in the presence of three SBs.
3FT6, the binding sites of the first seven docking ranks, the most negative binding sites in Table 2 for H2L1, H2L2 and NiL2, respectively. All SBs, specially NiL2, intercalate between base pairs and trend of the most negative docking energies is H2L1 < H2L2 < NiL2. 3.8. Molecular dynamic simulation
Fig. 10. Structure of the DNA fragment (3FT6) in the absence (a) and presence of seven docking cluster rank of (b) H2L1, (c) H2L2, (d) NiL2 and the first negative cluster ranks for three SBs (e), respectively.
Molecular dynamics simulation was performed on DNA (PDB code: 453D) during the 30 ns in the presence of three Schiff-bases by GROMACS 5.3.1 software. Schiff-base can interact with DNA and change its structural parameters such as hydrogen bond, surface area and radius of gyration. By addition of Schiff-base, the number of hydrogen bond in DNA decreases. Accessible surface area also varies by addition of Schiff-base. The structure information such as total accessible solvent surface area (SASA) and hydrogen bonding between DNA oligonucleotides, RMSD and RDF were obtained at 300 K. The root mean square deviation (RMSD) of system in the presence of three Schiff-bases derivatives was obtained. Fig. 11a shows the RMSD curves of DNA-Schiff-base system in the 30 ns
D. Ajloo et al. / International Journal of Biological Macromolecules 77 (2015) 193–202
201
Fig. 12. Calculated RDF, g(r), for SBs (a) around DNA and (b) around each other in three studied systems.
simulation. It shows that the system gets a stable state after about 4 ns. The figure also shows DNA has more structural changes (RMSD) in the presence of Ni derivative than other SBs that is compatible with experimental results.
Fig. 14. Variation of distances between the SB nearest molecule of (DRG28) and terminal base pairs (a) during 30 ns, (b) selected first 1 ns and (c) average of part (b).
Fig. 13. Snapshot molecular picture was taken from system in the presence of 8 H2L1 (a) before, (b) after simulation and (c) expanded view of part (b).
Fig. 11b compares intermolecular hydrogen bond of DNA in the presence of H2L1, H2L2 and NiL2, in 30 ns. This figure shows that intermolecular hydrogen bond of DNA has decreased more in the presence of NiL2 and the decreasing trend is: NiL2 > H2L2 > H2L1 respectively. These results indicated more reduction of intermolecular hydrogen bond by NiL2 and more potent interaction of NiL2 to DNA. Fig. 11c shows the solvent accessible surface area of DNA in 30 ns calculation in the presence of SBs. This figure shows the increase of surface area in the presence of NiL2, H2L2 and H2L1, respectively. This result is in good accordance with reduction of intermolecular hydrogen bond between DNA. It proves that the DNA structure has been changed more due to more interaction with NiL2. The radial distribution function, g(r), of SBs around CT-DNA vs. distance was shown in Fig. 12a and radial distribution function of NiL2, H2L2 and H2L1 around each other is depicted in Fig. 12b. It can be seen that RDF (NiL2, H2L2 and H2L1-CT-DNA) is more for NiL2 > H2L2 > H2L1 and the same trend is observed for g(r) NiL2, H2L2 and H2L1 around themselves. It can be seen in Fig. 12b, increasing trend of g(r) is NiL2 > H2L2 > H2L1, which is due to the self-aggregation of free Schiff-bases along the DNA surface. These results are indicating the
202
D. Ajloo et al. / International Journal of Biological Macromolecules 77 (2015) 193–202
formation of the larger aggregates of Schiff-base on the DNA surface which confirms above experimental results. Snapshot molecular picture was taken from system in the presence of 8 molecules of H2L2 before and after simulation and results are shown in Fig. 13. It shows that one of the molecules of H2L2 get to terminal base pair and orient parallel to them. In order to observe the intercalation process, higher simulation time, about micro second, or simulated annealing molecular dynamics need to be carried out. The later case is in consideration by our group. Distance between cited molecule (DRG28) and terminal base pairs are calculated and plotted in Fig. 14a. It seems that the DRG28 reach to the terminal residues at initial times. So the initial times were expanded to better seeing the distance (Fig. 14b) and average of distance (Fig. 14c) variations. 4. Conclusion The DNA-binding properties of three SBs were examined by absorption, fluorescence and CD spectra as well as CV and viscosity measurements. Experimental results indicated that the SBs can bind to DNA via intercalation mode. These compounds increase the thermal stability, Tm , viscosity, surface area and decrease the hydrogen bond. In comparison between three Schiff bases nickel complex has more affinity for binding to DNA and so has higher effect on DNA structural changes. Results of computational docking energies and structural parameters are compatible with experimental data. Acknowledgement Financial support of Damghan University is gratefully acknowledged. References [1] M.S. Karthikeyan, D.J. Parsad, B. Poojary, K.S. Bhat, B.S. Holla, N.S. Kumari, Bioorg. Med. Chem. 14 (2006) 7482–7489. [2] K. Singh, M.S. Barwa, P. Tyagi, Eur. J. Med. Chem. 41 (2006) 147–153. [3] H.X. Yu, J.F. Ma, G.H. Xu, S.L. Li, J. Yang, Y.Y. Liu, Y.X. Cheng, J. Organomet. Chem. 691 (2006) 3531–3539. [4] G. Turan-Zitouni, Z.A. Kaplanciki, M.T. Yildiz, P. Chevallet, D. Kaya, Eur. J. Med. Chem. 40 (2005) 607–613. [5] S. Lei, H.M. Ge, S.H. Tan, H.Q. Li, Y.C. Song, H.L. Zhu, R.X. Tan, Eur. J. Med. Chem. 42 (2007) 558–564. [6] X. Zhao, P.P.-F. Lee, Y.K. Yan, C.K. Chu, J. Inorg. Biochem. 101 (2007) 321–328. [7] G. Cerchiaro, K. Aquilano, G. Filomeni, G. Rotilio, M.R. Ciriolo, A.M. Ferreira, J. Inorg. Biochem. 99 (2005) 1433–1440. [8] Z.H. Xu, F.J. Chen, P.X. Xi, X.H. Liu, Z.Z. Zeng, J. Photochem. Photobiol. A: Chem. 196 (2008) 98–102. [9] G.F. Qi, Z.Y. Yang, B.D. Wang, Transit. Met. Chem. 32 (2007) 233–239.
[10] S. Moradi, D. Ajloo, T. Lashkarbolouki, R. Alizadeh, A.A. Saboury, Monatsh. Chem. 144 (2013) 1499–1505. [11] D.S. Auld, Biometals 14 (2001) 271–313. [12] D. Ajloo, E. Taghizadeh, A.A. Saboury, E. Bazyari, K. Mahnam, Int. J. Biol. Macromol. 43 (2008) 151–158. [13] W. Bal, K.S. Kasprzak, Toxicol. Lett. 127 (2002) 55–62. [14] K.S. Kasprzak, W. Bal, A.A. Karaczyn, J. Environ. Monit. 5 (2003) 183–187. [15] S. Rauf, J.J. Gooding, K. Akhtar, M.A. Ghauri, M. Rahman, M.A. Anwar, A.M. Khalid, J. Pharm. Biomed. Anal. 37 (2005) 205–217. [16] L.H. Hurley, Nat. Rev. Cancer 2 (2002) 188–200. [17] H. Mansouri-Torshizi, M. Islami-Moghaddam, A. Divsalar, A.A. Saboury, Bioorg. Med. Chem. 16 (2008) 9616–9627. [18] B.K. Sahoo, K.S. Ghosh, R. Bera, S. Dasgupta, Chem. Phys. 351 (2008) 163–169. [19] G. Zhang, J. Guo, J. Pan, X. Chen, J. Wang, J. Mol. Struct. 923 (2009) 114–119. [20] L.S. Ling, Z.K. He, F. Chen, Y.E. Zeng, Talanta 59 (2003) 269–275. [21] X.Q. He, Q.Y. Lin, R.D. Hu, X.H. Lu, Spectrochim. Acta A 68 (2007) 184–190. [22] M.M. Hassanien, K.S. Abou-E.I-Sherbini, Talanta 68 (2006) 1550–1559. [23] F. Cui, R. Huo, G.X. Hui, X. Lv, J.G. Jin, X.W. Zhang, J. Mol. Struct. 1001 (2011) 104–110. [24] C. Tong, Z. Hu, J. Wu, J. Fluoresc. 20 (2010) 261–267. [25] N. Shahabadi, S. Kashanian, F. Darabi, Eur. J. Med. Chem. 45 (2010) 4239–4245. [26] D. Ajloo, M. Sangian, M. Ghadamghahi, M. Evini, A.A. Saboury, Int. J. Biol. Macromol. 55 (2013) 47–61. [27] Z. Tingting, B. Shuyun, W. Yu, W. Tianjiao, P. Bo, G. Tingting, Spectrochim. Acta A 132 (2014) 198–204. [28] G. Zhang, L. Wang, X. Zhou, Y. Li, J. Agric. Food Chem. 62 (2014) 991–1000. [29] G. Zhang, Y. Ma, Food Chem. 141 (2013) 41–47. [30] Z. Tian, Z. Zhao, F. Zang, Y. Wang, C. Wang, J. Photochem. Photobiol. B 138 (2014) 202–210. [31] Y. Li, G. Zhang, J. Pan, Y. Zhang, Sens. Actuators B: Chem. 191 (2014) 464–472. [32] C.Z. Huang, Y.F. Li, P. Feng, Talanta 55 (2001) 321–328. [33] G.M. Morris, D.S. Goodsell, R.S. Halliday, R. Huey, W.E. Hart, R.K. Belew, A.J. Olson, J. Comput. Chem. 19 (1998) 1639–1662. [34] G.M. Morris, R. Huey, W. Lindstrom, M.F. Sanner, R.K. Belew, D.S. Goodsell, A.J. Olson, J. Comput. Chem. 30 (2009) 2785–2791. [35] B.D. Bursulaya, M. Totrov, R. Abagyan, C.L. Brooks, J. Comput. Aided Mol. Des. 17 (2003) 755–763. [36] HyperChem (TM) Professional 7.51; Hypercube, Inc., 1115 NW 4th Street, Gainesville, Florida, USA. [37] A.W. Schüttelkopf, D.M.F. van Aalten, Acta Crystallogr. D60 (2004) 1355–1363. [38] W.D. Cornell, P. Cieplak, C.I. Bayly, I.R. Gould, K.M. Merz Jr., D.M. Ferguson, D.C. Spellmeyer, T. Fox, J.W. Caldwell, P.A. Kollman, J. Am. Chem. Soc. 117 (1995) 5179–5197. [39] P. Ewald, Ann. Phys. 369 (1921) 253–287. [40] D. van der Spoel, E. Lindahl, B. Hess, G. Groenhof, A.E. Mark, H.J.C. Berendsen, J. Comp. Chem. 26 (2005) 1701–1718. [41] Z.S. Yang, Y.L. Wang, G.C. Zhao, Anal. Sci. 20 (2004) 1127–1131. [42] J.A. Ewing, I.D. Kuntz, J. Comput. Chem. 18 (1997) 1175–1189. [43] D.E. Schmechel, D.M. Crothers, Biopolymers 10 (1971) 465–480. [44] E. Froehlich, J.S. Mandeville, C.J. Jennings, R. Sedaghat-Herati, H.A. Tajmir-Riahi, J. Phys. Chem. B 113 (2009) 6986–6993. [45] Y. Li, Y. Yang, Inorg. Chim. Acta 362 (2009) 4823–4831. [46] Y. Sun, S. Bi, D. Song, C. Qiao, D. Mu, H. Zhang, Sens. Actuators B 129 (2008) 799–810. [47] M. Nejat-Dehkordi, A.K. Bordbar, M. Ayatolahi-Mehrgardi, V. Mirkhani, J. Fluoresc. 21 (2011) 1649–1658. [48] Y. Zhang, J. Pan, G. Zhang, X. Zhou, Sens. Actuators B: Chem. 206 (2015) 630–639. [49] D. Agudelo, P. Bourassa, G. Bérubé, H.A. Tajmir-Riahi, Int. J. Biol. Macromol. 66 (2014) 144–150. [50] S. Ramakrishnan, M. Palaniandavar, J. Chem. Sci. 117 (2005) 179–186.
Contents lists available at ScienceDirect
International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac
Interaction of three new tetradentates Schiff bases containing N2 O2 donor atoms with calf thymus DNA Davood Ajloo a,∗ , Sajede Shabanpanah a , Bita Shafaatian a , Maryam Ghadamgahi a , Yasin Alipour a , Taghi Lashgarbolouki b , Ali Akbar Saboury c a
School of Chemistry, Damghan University, Damghan, Iran School of Biology, Damghan University, Damghan, Iran c Institute of Biochemistry and Biophysics, University of Tehran, Tehran, Iran b
a r t i c l e
i n f o
Article history: Received 27 November 2014 Received in revised form 3 March 2015 Accepted 10 March 2015 Available online 19 March 2015 Keywords: Calf-thymus DNA Schiff-base Spectroscopy
a b s t r a c t Interaction of 1,3-bis(2-hydroxy-benzylidene)-urea (H2L1), 1,3-bis(2-hydroxy-3-methoxybenzylidene)-urea (H2L2) and 1,3-bis(2-hydroxy-3-methoxy-benzylidene)-urea nickel(II) (NiL2) with calf-thymus DNA were investigated by UV–vis absorption, fluorescence emission and circular dichroism (CD) spectroscopy as well as cyclic voltammetry, viscosity measurements, molecular docking and molecular dynamics simulation. Binding constants were determined using UV–vis absorption and fluorescence spectra. The results indicated that studied Schiff-bases bind to DNA in the intercalative mode in which the metal derivative is more effective than non metals. Their interaction trend is further determined by molecular dynamics (MD) simulation. MD results showed that Ni derivative reduces oligonucleotide intermolecular hydrogen bond and increases solvent accessible surface area more than other compounds. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Schiff-bases are the important class of compounds in both medicinal and pharmaceutical fields [1,2]. In recent years, Schiffbase metal complexes have also found important applications in the biological field [3–7]. Schiff-bases play an important role in bioinorganic chemistry as they exhibit remarkable biological activity. On the other hand, nucleic acids have important influence in biological systems and carry out a broad range of biological functions. DNA is the primary intracellular target of anticancer drugs and so the interactions between small molecules and DNA causes DNA damage in cancer cells, blocking their division and resulting in cell death [8]. Therefore, the interaction of metal complexes with DNA has attracted much attention [9,10]. Among the transition metal ions, zinc and nickel are biologically active [11,12]. The knowledge of intracellular speciation of nickel is essential for deeper understanding of the toxicity mechanism of this carcinogenic metal [13,14]. However, the interaction of drug molecules with DNA has become an active research area [15]. Since DNA is the intracellular target for a wide range of anticancer and antibiotic drugs [16–18], generally, there are three interaction modes
∗ Corresponding author. Tel.: +98 0232 5233051–6; fax: +98 0232 5235713. E-mail addresses: [email protected], [email protected] (D. Ajloo). http://dx.doi.org/10.1016/j.ijbiomac.2015.03.016 0141-8130/© 2015 Elsevier B.V. All rights reserved.
between small molecules and DNA: (i) intercalative binding that small molecules intercalate into the base pairs of nucleic acids; (ii) groove binding in which the small molecules bound on nucleic acids in major or minor groove; (iii) long-range assembly on the molecular surfaces of nucleic acids [19]. The intercalative binding is stronger than other two binding modes because the surface of intercalative molecule is sandwiched between the aromatic, heterocyclic base pairs of DNA [20,21]. Tremendous interest has been attracted to interactions between transition metal complexes of morin and nucleic acids due to potential applications of the metal complexes as anticancer drugs or as complexes with other biological functions [8,22]. These studies are also important to understand the toxicity of drugs containing metal ions [23–25]. CT-DNA is a polymer of alternate sugar phosphate sequence with high polymerized skeleton. The investigation of drug–DNA interaction is important for understanding the molecular mechanism of drug action and for the design of specific DNA targeted drug. DNA binding is the critical step for DNA activity. To design effective chemotherapeutic agents and better anticancer drugs, it is essential to explore the interactions of drug with DNA. Interaction of a few ligands with DNA were investigated by different experimental methods such as UV–vis, fluorescence, CD, calorimetry, CV and computational methods such as MD, docking and QSAR techniques were investigated [26–31].
194
D. Ajloo et al. / International Journal of Biological Macromolecules 77 (2015) 193–202
In the present study, the interaction of new tetradentate Schiffbases containing N2 O2 donor atoms and its nickel complex with calf thymus DNA were investigated using various spectroscopy methods, molecular dynamics simulation and docking calculations.
experiments were performed in the wavelength range, 200–600 nm in a constant concentration of studied SBs and titrated by varying concentration of DNA ([DNA]/[compounds] = 0.00, 0.02, 0.04, 0.06, 0.08, 0.10, 0.12, 0.14, 0.16, 0.18 and 0.20) in (10 mM Tris–HCl, pH = 7.4). These solutions were incubated for 5 min.
2. Experimental 2.3. Thermal stability
2.1. Materials All materials were purchased from Sigma–Aldrich Company. All experiments involving interaction of the complexes with DNA were carried out in buffer (10 mM Tris–HCl, pH = 7.4). The DNA concentration per nucleotide was determined by absorption spectroscopy using the molar absorption coefficient (6600 M−1 cm−1 ) at 260 nm. The structure of studied Schiff base (SB)s which synthesized by inorganic chemistry laboratory of Damghan University were shown in Fig. 1. 2.2. UV–vis spectroscopy measurement UV–vis spectra were recorded on a Perkin-Elmer UV-Vis spectrophotometer model Lambda 25. Absorption titration
DNA thermal stability was investigated by indicated absorbance value versus temperature. The absorbance at 260 nm was scanned from 35 to 85 ◦ C at 5 ◦ C per min scan rate and fixed concentration ratio ([SB]/[DNA] = 0.5). The melting temperature Tm , was defined as the mid-point of transition temperature. 2.4. Fluorescence measurements Emission spectra were recorded on a Jasco spectrofluorometer model FP6200 luminescence spectrometer at 298 K. To compare quantitatively the affinity of the cited compounds to DNA, the average binding constants, Kb , were obtained by fluorescence spectroscopy [9]. Fluorescence measurements were performed on fixed amounts of SBs in the presence of different amounts of DNA. Ethidium bromide (EB) is a fluorescent probe for DNA structure which has been employed in examination of the ligand binding mode to DNA. The excitation wavelengths for 5 × 10−4 M H2L1, H2L2 and NiL2 were 276, 484 and 514 nm, respectively and average of fluorescence emission intensity was monitored for H2L1, H2L2 and NiL2 as 250–600 nm, 470–550 nm and 500–550 nm, respectively. 2.5. Viscosity measurement Viscosity was measured by an Ostwald viscometer immersed in a thermostatic water-bath maintained to 25.0 ◦ C. In all samples, the DNA concentration was kept constant (1 × 10−8 M). Calculation was carried out using Eq. (1); 0 = =
tDNA − t0 t0
t − t0 t0
(1a) (1b)
where 0 and are the viscosity of DNA in the absence and presence of SB, t0 , tDNA and t are the observed flow time of buffer, DNA and DNA containing solution upon the addition of SBs, respectively [32]. The values of (/0 )1/3 versus the concentration ratio of SB to DNA were plotted. 2.6. Circular dichroism Cyclic voltammetry measurements were done using Potentiostat/Galvanostat Autolab. The CD spectra are quite sensitive to the changes in the secondary structure of nucleic acids, which any conformational modification of DNA provoked by its interaction with SBs. CD spectrum of each sample was scanned in the ranges of 220–320 nm. The concentration of DNA was 1.0 × 10−3 M. A CD spectrum was generated by subtracting the CD spectrum of the native DNA and mixture of DNA-SB from the CD spectrum of the buffer and buffer-SB solutions. 2.7. Cyclic voltammetry
Fig. 1. Chemical structure of three studied SBs.
Circular dichroism spectra were recorded by AVIV circular dichroism spectrophotometer model 215. Cyclic voltammetry measurements were carried out by glassy carbon as working electrode, a platinum wire as an auxiliary electrode, and
D. Ajloo et al. / International Journal of Biological Macromolecules 77 (2015) 193–202
195
Ag/AgCl/3 M KCl as a reference electrode. Cyclic voltammograms of the 5 × 10−5 M SBs at 25 ◦ C were obtained in 60 mV s−1 scan rate. 2.8. Molecular docking AutoDock 3.0.5 program was used to obtain the energetic and binding site for the interaction of the titled Schiff-bases with DNA. Docking simulation was done in a box with dimension of 52 × 100 × 80 points and the spacing of 0.375 A˚ [33–35]. Maximum number of generations and energy evaluation were set to 27,000
Fig. 3. (a) Variation of maximum absorbance and (b) derivative of maximum absorbance for 5 × 10−5 M of CT-DNA in different temperatures in the absence (䊉) ), H2L2 ( ) and NiL2 ( ) SBs. and presence of 1 × 10−4 M H2L1 (
and 2.5 × 105 , respectively. For each run, 250 docking energies were reported with the initial population of 250 individuals. 2.9. Molecular dynamics simulation
Fig. 2. Absorption spectra of 1 × 10−4 M of (a) H2L1, (b) H2L2 and (c) NiL2 in Tris–HCl buffer upon addition of 0–200 L of calf-thymus DNA (1 × 10−3 M). [DNA]/[SB] = 0.00, 0.02, 0.04, 0.06, 0.08, 0.10, 0.12, 0.14, 0.16, 0.18 and 0.20. Arrows show the absorbance changing upon increasing DNA concentrations. Inset: plots of [DNA]/(εa − εf ) versus [DNA] for the titration of compounds with the DNA.
The structures of three Schiff-bases were drawn using Hyperchem 7 [36] software. Force field parameters and geometries of ligands were generated using PRODRG2 server [37] and they were changed to be adaptable with selected force field. The parameters of Ni metal were also inserted manually with parameters calculated from Gaussian software. The MD simulations were performed by the AMBER [38] force field and TIPP4 model was used for water molecules. A twin range cutoff was used for longer-range interactions: 0.9 nm for van der Waals interactions and 0.9 nm for electrostatic interactions. The PME [39] was used for calculating long-range interaction. The starting structure of DNA was constructed based on the X-ray crystal structure of it (PDB ID: 453D). A cubic simulation box of the volume 315 nm3 was made and 8 molecules of Schiff-bases were placed randomly in this box respectively. Then water molecules were randomly added into the simulation box and initial configurations were minimized using steepest descent algorithm with 5000 integration step and the system was
196
D. Ajloo et al. / International Journal of Biological Macromolecules 77 (2015) 193–202
Hypochromism and hyperchromism are both spectral feature of DNA concerning changes in its double helix structure. The UV–vis titration spectra of the SBs was shown in Fig. 2. This figure shows the absorption spectra in the absence and presence of calf thymus DNA. The addition of calf thymus DNA to solutions of compounds exhibited hyperchromicity which might be ascribed to the fact that the SBs could uncoil the helix structure of DNA and made more bases embedding in DNA [41]. Also hyperchromism may probably be due to dissociation of aggregated ligand or external contact with DNA. A similar hyperchromism has been observed for the Soret bands of certain porphyrins when they interact with DNA [42] that shows the compounds could bind to DNA by intercalation. The change in the absorbance when the binding sites on DNA are occupied by ligand, A was determined by
Fig. 4. The fluorescence spectra of 5 × 10−4 M (a) H2L1, (b) H2L2 and (c) NiL2 complex in the presence of DNA. Inset: plot log((F − F0 )/F0 ) versus log[Q] for titration of SBs by increasing concentration of CT-DNA.
equilibrated for 30 ns at constant pressure (1 atm) and temperature (300 K) using the Parilleno-Rahman procedure. The resulting system of MD models contains DNA and Schiff-base. All MD simulations were carried out using the GROMACS 4.5.4 package [40]. The calculations were performed using 5 quad core parallel computers including 40 processor units. The computer applied the Rocks cluster networking and Centos operating systems and the repeated trajectory showed similar result. The 8 molecules of SB were used in the 5145, 5132 and 5111 water molecules for H2L1, H2L2 and NiL2, respectively. 3. Results and discussion 3.1. UV–vis measurement Electronic absorption spectroscopy is an effective method to examine the binding mode of DNA with metal complexes.
Fig. 5. The emission spectra of 60 M DNA in the presence of different concentration ratio, [EB]/[DNA], and 0.25 mM of (a) H2L1, (b) H2L2 Ligands and (c) NiL2 complex in Tris–HCl buffer. Inset: Fluorescence Scatchard plots for binding of EB to DNA in ), 0.25 mM ( ) and 0.3 mM ( ) of (a) H2L1, (b) the presence of 0.2 mM ( H2L2 Ligands and (c) NiL2 complex.
D. Ajloo et al. / International Journal of Biological Macromolecules 77 (2015) 193–202
2 NiL2 H2L2 H2L1
1.8
(η/ηo)1/3
1.6 1.4 1.2 1 0.8 0
0.02
0.04
0.06
[Schiff base]/[DNA] Fig. 6. Effect of increasing amounts of 5 × 10−4 M of the (a) H2L1, (b) H2L2 and (c) NiL2 on the relative viscosity of 10−3 M calf-thymus DNA at 25.0 ◦ C.
A = |ADNA-SBs − ADNA − ASBs | and the spectral overlap of CT-DNA and SBs was eliminated. The average binding constant of the compound-DNA was calculated by Eq. (2) [43]: [DNA] [DNA] 1 = + (εb − εf ) K(εb − εf ) (εa − εf )
(2)
where [DNA] is the base pair concentration, εa , εb and εf are apparent, bound and free absorption coefficient, respectively. Binding constant, Kb , was obtained by dividing the slope to intercept of the [DNA]/(εa − εf ) versus [DNA] plot. The value of average binding constant for H2L1, H2L2 and NiL2 was obtained as 1.4 × 104 , 2.5 × 104 and 3.3 × 104 M−1 , respectively. 3.2. Thermal stability of DNA Structure of DNA is stable due to presence of intermolecular hydrogen bond. These stability and hydrogen bond will be
diminished by increasing temperature. By increasing the temperature, the helix structure is dissociated to single strand. The absorption data versus temperature was monitored in Fig. 3. In the presence of complex, by increasing temperature, hydrogen bond will be broken and subsequently the ligand diffuses between two strands. On other words, SB acts same as adhesive that connect the two strands stronger than hydrogen bond between base pairs. Whatever the interaction between reagent and docking site become stronger, the stability increases. Strength of interaction can be found from docking energy. As it comes in the next section docking energy is a good criterion for this purpose. The Tm value of DNA in the absence of complexes is recorded at 64.5 ◦ C. Values of Tm for DNA in the presence of H2L1, H2L2 and NiL2, were obtained about 64.5, 70.0 and 70.0 ◦ C, respectively. Therefore, by addition of cited SBs, Tm value of DNA increases and confirms the intercalation mode of the SB interaction. 3.3. Fluorescence results To further investigate the interaction mode between the SB and CT-DNA, fluorescence titration experiments was performed in two different ways: (a) titration of SB by DNA and (b) titration of DNA-EB by SBs. In the case (a) if it is assumed that there are similar and independent binding sites, in the biomolecule, the binding constant (Kb ) and the number of binding sites (n) can be determined according to the method described by [44]: log
F − F 0
F0
(a)
-2 237
7 μM H2L2 8
257
277
297
317
337
14 μM H2L2 21 μM H2L2
[Ɵ] mdeg
[Ɵ] mdeg
21 μM H2L1
(3)
(b)
14 μM H2L1
3
= log Kb + n log[Q]
Here, F0 and F are the fluorescence intensities of the fluorophore in the absence and presence of different concentrations of CTDNA, respectively. By plotting the (log(F − F0 )/F0 ) versus log[Q] (Fig. 4), the values of Kb and n were found to be, 1.5 × 103 , 1.6 × 103 ,
7 μM H2L1 8
197
3
-2 237
257
277
297
317
337
-7
-7 -12
-12
Wavelength (nm)
Wavelength (nm)
15 (C)
21 μM NiL2
3 -2 237
257
277
297
317
-7
337
DNA NiL2 H2L2 H2L1
(d)
10
14 μM NiL2
[Ɵ] mdeg
[Ɵ] mdeg
8
7 μM NiL2
5 0 -5
233
253
273
293
313
333
-10
-12
-15
Wavelength (nm)
Wavelength (nm)
Fig. 7. CD spectra of 5 × 10−5 M of CT-DNA in the absence and presence of 7, 14, 21 M of (a) H2L1, (b) H2L2 and (c) NiL2. (d) Comparison between CD spectra of DNA in the absence and presence of three SBs.
198
D. Ajloo et al. / International Journal of Biological Macromolecules 77 (2015) 193–202
1.8 × 103 and 0.902, 1.704, 0.783 for H2L1, H2L2 and NiL2, respectively. The data confirmed more and stronger interaction of the NiL2 complex with DNA. The emission spectra for titration of DNA-EB by SBs were illustrated in Fig. 5 in which the lowest spectrum is related to EB. By addition of DNA, the emission intensity of the EB was enhanced to the maximum peak. Then, emission intensity of the DNA–EB system decreases by the increase of SB concentration, which indicated that SB could displace EB from the DNA–EB system. In order to find the binding mode, the Scatchard plots were used. Therefore, rc /Cf values were obtained according to the following equations and plotted against r for all SBs bound to DNA. r = Kb (n − r) = Kb n − Kb r cf cb = ct [(F − F0 )/(Fmax − F0 )]
(4)
cf = ct − cb where ct , cf and cb are the total, free and bound SB concentration, respectively. F is the observed fluorescence emission intensity at given DNA concentration, F0 is the fluorescence intensity in the absence of DNA, and Fmax is the fluorescence intensity of the totally bound SB. The n is the binding site number that is 0.83, 0.51 and 0.26 for H2L1, H2L2 and NiL2, respectively, and Kb values are 0.65 × 104 , 0.96 × 104 and 1.01 × 104 . By drawing the fluorescence Scatchard plots (Fig. 5), we may achieve four behaviors for the binding of complex to DNA. In this experiment, by data analyzing, type D behavior was observed. Noncompetitive inhibition of EB binding (type D behavior) produces a Scatchard plot in which the slope is almost constant in the presence of increasing amounts of SB. In this kind of binding curve, the SB is attached to DNA via the groove/covalent or electrostatic interactions and sterically prevents the binding of EB to the latter [45]. 3.4. Viscosity results In general, interactions between the complexes and DNA were further clarified by viscosity measurements. Effect of the SBs on the viscosity of DNA at 25 ◦ C was shown in Fig. 6. The viscosity of DNA enhances by increasing in the ratio of SBs to DNA, so that the variation is significantly higher in the presence of Ni complex. In other words, all the three SBs can intercalate between DNA base pairs, that leading an extension in the helix, and thus increase the viscosity of DNA. Furthermore, the results indicated the intercalative binding mode. Also the nickel complex can intercalate more strongly than other SBs that causing increases in the flow viscosity. 3.5. Circular dichroism results Circular dichroism spectroscopy gives us useful information on how the conformation of DNA is influenced by the binding of the metal complex to DNA. Effect of SB on DNA secondary structure was studied by keeping the concentration of CT-DNA constant while varying the concentration of SBs in 10 mM Tris buffer. The CD spectra of DNA taken after incubation of the SBs with CT-DNA are shown in Fig. 7. The observed CD spectrum of calf-thymus DNA consists of a positive band at 274 nm due to base stacking and a negative band at 245 nm due to helicity. While the groove binding and electrostatic interaction of small molecules with DNA shows little or no perturbations on the base stacking and helicity bands, intercalation enhances the intensities of both bands [46,47]. Also the results confirmed probability of DNA structure in right-handed B form (Fig. 7) [48]. The comparison between CD spectra of DNA in the presence of SBs were shown in Fig. 7d. It shows that nickel complex intercalate more than other SBs.
Fig. 8. Cyclic voltammograms of 5 × 10−5 M of (a) H2L1, (b) H2L2 Ligands and (c) NiL2 Complex in the absence and presence of different concentrations of DNA 10−3 M in10 mM Tris–HCl (pH = 7.4) at 60 mV s−1 scan rate.
3.6. Electrochemical results The application of electrochemical methods to the study of metal complex intercalation and coordination of transition metal complexes to DNA provides a useful complement to the previously used methods of investigation, such as UV–vis spectroscopy [49]. Fig. 8 shows the electrochemical measurements in the absence and presence of CT-DNA. It can be seen that in all samples the cathode peak current gradually decreased due to the addition of DNA. The decrease in current may be attributed to the diffusion of the SBs bound to the large, slowly diffusing DNA molecule. Also the decreases in the peak currents are ascribed to the stronger binding between the SBs and DNA. Moreover, during DNA addition the cathodic peak potential showed small positive shift [50]. These positive shifts are confirming exciton
D. Ajloo et al. / International Journal of Biological Macromolecules 77 (2015) 193–202
199
Fig. 9. (a) Structure of DNA taken from protein data bank (PDB: 453D) and distribution of all cluster ranks for H2L2. (b) The most negative docking site for three SBs in the minor groove, (c) expanded view of the most negative docking site, another view of location H2L1 (d), H2L2 (e), NiL2 (f) in minor groove. The hydrogen bonds between SBs and DNA were shown by green dashed line.
type interaction and intercalation in SB in which the decreased value was more in NiL2 than the other SBs. Probably, system of NiL2 in DNA has more binding affinity than other compounds.
Table 1 Docking energies for a few cluster ranks obtained by Autodock. H2L1
3.7. Docking results Autodock software calculates docking energies and finds the most probable binding sites. After preparation of ligand and DNA input files, docking calculation is performed to search and find the best binding site. In each run, 250 values for docking free energies were obtained and sorted based on increasing in energies. Some of 250 sites which have nearly equal range of docking energies, form a cluster based on their RMSD values. The first twelve clusters, including the lowest and mean docking energies were listed in Table 1. Results show that trend of negative binding energy is H2L1 < H2L2 < Ni2L2. Fig. 9a shows the electrostatic surface structure of DNA taken from protein data bank (PDB 453D) and distribution of docking
1 2 3 4 5 6 7 8 9 10 11 12
H2L2
NiL2
LDE
MDE
NSC
LDE
MDE
NSC
LDE
MDE
NSC
−8.70 −8.65 −8.48 −8.32 −8.31 −8.01 −7.88 −7.75 −7.70 −7.55 −8.70 −8.65
−8.70 −8.65 −8.48 −8.32 −8.31 −8.01 −7.88 −7.75 −7.70 −7.55 −8.70 −8.65
1 1 1 1 1 1 1 1 1 1 1 1
−9.00 −8.90 −8.70 −8.62 −8.58 −8.57 −8.53 −8.52 −8.51 −8.45 −9.00 −8.90
−9.00 −8.90 −8.37 −8.62 −8.58 −8.34 −8.43 −8.52 −8.47 −8.45 −9.00 −8.90
1 1 4 1 1 2 2 1 3 1 1 1
−9.15 −9.14 −9.06 −8.95 −8.93 −8.72 −8.64 −8.63 −8.56 −8.52 −8.47 −8.45
−9.15 −9.04 −8.95 −8.86 −8.88 −8.72 −8.64 −8.63 −8.53 −8.42 −8.47 −8.41
2 10 14 8 5 1 1 1 10 4 1 38
LDE, lowest docking energy; MDE, mean docking energy; NSC, number of sites in a cluster.
200
D. Ajloo et al. / International Journal of Biological Macromolecules 77 (2015) 193–202
Table 2 Docking energies for a few cluster ranks of DNA (3FT6) obtained by Autodock. Rank
1 2 3 4 5 6 7
H2L1
H2L2
NiL2
LDE
MDE
NSC
LDE
MDE
NSC
LDE
MDE
NSC
−6.90 −6.87 −6.73 −6.72 −6.30 −5.97 −5.95
−6.50 −6.40 −6.49 −6.27 −6.26 −5.97 −5.95
7 9 9 21 2 1 1
−7.30 −7.26 −7.19 −7.15 −6.99 −6.90 −6.84
−6.96 −7.11 −6.39 −6.75 −6.99 −6.88 −6.79
12 9 7 4 1 4 2
−7.43 −7.40 −7.36 −7.33 −7.31 −7.30 −7.28
−7.43 −7.39 −7.36 −7.33 −7.31 −7.30 −7.28
45 44 6 1 2 1 1
sites for all cluster ranks belong to H2L2. As we see the probability of distribution in the minor groove is higher than other sites. Fig. 9b shows docking sites of the most negative docking energies for three SBs that superimposed on each other. Expanded view of docking site in stick form as well as the labels of the nearest nucleoside was also shown in Fig. 9c. Fig. 9d–f shows another view of three SBs in the minor groove that the hydrogen bonds were shown by dashed green lines. The H2L1 forms three hydrogen bonds as (DRG1 OAH. . .DA17:N3, DRG1 HAC. . .:DA18:O4 and DRG1 HAT. . .DT8:O2), H2L2 one hydrogen bond (DG10:N2 . . .:DRG1:HAO) and NiL2 dose not form any hydrogen bond. On the other hand, another DNA structure was taken from PDB bank (3FT6) in which the ligand had been intercalated between base pairs. Then the ligand was removed and docking calculation was done with SBs. The new results were listed in Table 2 and related docking sites shown in Fig. 10. Fig. 10a–e shows the structure of
Fig. 11. Calculated (a) RMSD, (b) hydrogen bond (H-bond) and (c) accessible surface area (ASA) by molecular dynamics for DNA in the presence of three SBs.
3FT6, the binding sites of the first seven docking ranks, the most negative binding sites in Table 2 for H2L1, H2L2 and NiL2, respectively. All SBs, specially NiL2, intercalate between base pairs and trend of the most negative docking energies is H2L1 < H2L2 < NiL2. 3.8. Molecular dynamic simulation
Fig. 10. Structure of the DNA fragment (3FT6) in the absence (a) and presence of seven docking cluster rank of (b) H2L1, (c) H2L2, (d) NiL2 and the first negative cluster ranks for three SBs (e), respectively.
Molecular dynamics simulation was performed on DNA (PDB code: 453D) during the 30 ns in the presence of three Schiff-bases by GROMACS 5.3.1 software. Schiff-base can interact with DNA and change its structural parameters such as hydrogen bond, surface area and radius of gyration. By addition of Schiff-base, the number of hydrogen bond in DNA decreases. Accessible surface area also varies by addition of Schiff-base. The structure information such as total accessible solvent surface area (SASA) and hydrogen bonding between DNA oligonucleotides, RMSD and RDF were obtained at 300 K. The root mean square deviation (RMSD) of system in the presence of three Schiff-bases derivatives was obtained. Fig. 11a shows the RMSD curves of DNA-Schiff-base system in the 30 ns
D. Ajloo et al. / International Journal of Biological Macromolecules 77 (2015) 193–202
201
Fig. 12. Calculated RDF, g(r), for SBs (a) around DNA and (b) around each other in three studied systems.
simulation. It shows that the system gets a stable state after about 4 ns. The figure also shows DNA has more structural changes (RMSD) in the presence of Ni derivative than other SBs that is compatible with experimental results.
Fig. 14. Variation of distances between the SB nearest molecule of (DRG28) and terminal base pairs (a) during 30 ns, (b) selected first 1 ns and (c) average of part (b).
Fig. 13. Snapshot molecular picture was taken from system in the presence of 8 H2L1 (a) before, (b) after simulation and (c) expanded view of part (b).
Fig. 11b compares intermolecular hydrogen bond of DNA in the presence of H2L1, H2L2 and NiL2, in 30 ns. This figure shows that intermolecular hydrogen bond of DNA has decreased more in the presence of NiL2 and the decreasing trend is: NiL2 > H2L2 > H2L1 respectively. These results indicated more reduction of intermolecular hydrogen bond by NiL2 and more potent interaction of NiL2 to DNA. Fig. 11c shows the solvent accessible surface area of DNA in 30 ns calculation in the presence of SBs. This figure shows the increase of surface area in the presence of NiL2, H2L2 and H2L1, respectively. This result is in good accordance with reduction of intermolecular hydrogen bond between DNA. It proves that the DNA structure has been changed more due to more interaction with NiL2. The radial distribution function, g(r), of SBs around CT-DNA vs. distance was shown in Fig. 12a and radial distribution function of NiL2, H2L2 and H2L1 around each other is depicted in Fig. 12b. It can be seen that RDF (NiL2, H2L2 and H2L1-CT-DNA) is more for NiL2 > H2L2 > H2L1 and the same trend is observed for g(r) NiL2, H2L2 and H2L1 around themselves. It can be seen in Fig. 12b, increasing trend of g(r) is NiL2 > H2L2 > H2L1, which is due to the self-aggregation of free Schiff-bases along the DNA surface. These results are indicating the
202
D. Ajloo et al. / International Journal of Biological Macromolecules 77 (2015) 193–202
formation of the larger aggregates of Schiff-base on the DNA surface which confirms above experimental results. Snapshot molecular picture was taken from system in the presence of 8 molecules of H2L2 before and after simulation and results are shown in Fig. 13. It shows that one of the molecules of H2L2 get to terminal base pair and orient parallel to them. In order to observe the intercalation process, higher simulation time, about micro second, or simulated annealing molecular dynamics need to be carried out. The later case is in consideration by our group. Distance between cited molecule (DRG28) and terminal base pairs are calculated and plotted in Fig. 14a. It seems that the DRG28 reach to the terminal residues at initial times. So the initial times were expanded to better seeing the distance (Fig. 14b) and average of distance (Fig. 14c) variations. 4. Conclusion The DNA-binding properties of three SBs were examined by absorption, fluorescence and CD spectra as well as CV and viscosity measurements. Experimental results indicated that the SBs can bind to DNA via intercalation mode. These compounds increase the thermal stability, Tm , viscosity, surface area and decrease the hydrogen bond. In comparison between three Schiff bases nickel complex has more affinity for binding to DNA and so has higher effect on DNA structural changes. Results of computational docking energies and structural parameters are compatible with experimental data. Acknowledgement Financial support of Damghan University is gratefully acknowledged. References [1] M.S. Karthikeyan, D.J. Parsad, B. Poojary, K.S. Bhat, B.S. Holla, N.S. Kumari, Bioorg. Med. Chem. 14 (2006) 7482–7489. [2] K. Singh, M.S. Barwa, P. Tyagi, Eur. J. Med. Chem. 41 (2006) 147–153. [3] H.X. Yu, J.F. Ma, G.H. Xu, S.L. Li, J. Yang, Y.Y. Liu, Y.X. Cheng, J. Organomet. Chem. 691 (2006) 3531–3539. [4] G. Turan-Zitouni, Z.A. Kaplanciki, M.T. Yildiz, P. Chevallet, D. Kaya, Eur. J. Med. Chem. 40 (2005) 607–613. [5] S. Lei, H.M. Ge, S.H. Tan, H.Q. Li, Y.C. Song, H.L. Zhu, R.X. Tan, Eur. J. Med. Chem. 42 (2007) 558–564. [6] X. Zhao, P.P.-F. Lee, Y.K. Yan, C.K. Chu, J. Inorg. Biochem. 101 (2007) 321–328. [7] G. Cerchiaro, K. Aquilano, G. Filomeni, G. Rotilio, M.R. Ciriolo, A.M. Ferreira, J. Inorg. Biochem. 99 (2005) 1433–1440. [8] Z.H. Xu, F.J. Chen, P.X. Xi, X.H. Liu, Z.Z. Zeng, J. Photochem. Photobiol. A: Chem. 196 (2008) 98–102. [9] G.F. Qi, Z.Y. Yang, B.D. Wang, Transit. Met. Chem. 32 (2007) 233–239.
[10] S. Moradi, D. Ajloo, T. Lashkarbolouki, R. Alizadeh, A.A. Saboury, Monatsh. Chem. 144 (2013) 1499–1505. [11] D.S. Auld, Biometals 14 (2001) 271–313. [12] D. Ajloo, E. Taghizadeh, A.A. Saboury, E. Bazyari, K. Mahnam, Int. J. Biol. Macromol. 43 (2008) 151–158. [13] W. Bal, K.S. Kasprzak, Toxicol. Lett. 127 (2002) 55–62. [14] K.S. Kasprzak, W. Bal, A.A. Karaczyn, J. Environ. Monit. 5 (2003) 183–187. [15] S. Rauf, J.J. Gooding, K. Akhtar, M.A. Ghauri, M. Rahman, M.A. Anwar, A.M. Khalid, J. Pharm. Biomed. Anal. 37 (2005) 205–217. [16] L.H. Hurley, Nat. Rev. Cancer 2 (2002) 188–200. [17] H. Mansouri-Torshizi, M. Islami-Moghaddam, A. Divsalar, A.A. Saboury, Bioorg. Med. Chem. 16 (2008) 9616–9627. [18] B.K. Sahoo, K.S. Ghosh, R. Bera, S. Dasgupta, Chem. Phys. 351 (2008) 163–169. [19] G. Zhang, J. Guo, J. Pan, X. Chen, J. Wang, J. Mol. Struct. 923 (2009) 114–119. [20] L.S. Ling, Z.K. He, F. Chen, Y.E. Zeng, Talanta 59 (2003) 269–275. [21] X.Q. He, Q.Y. Lin, R.D. Hu, X.H. Lu, Spectrochim. Acta A 68 (2007) 184–190. [22] M.M. Hassanien, K.S. Abou-E.I-Sherbini, Talanta 68 (2006) 1550–1559. [23] F. Cui, R. Huo, G.X. Hui, X. Lv, J.G. Jin, X.W. Zhang, J. Mol. Struct. 1001 (2011) 104–110. [24] C. Tong, Z. Hu, J. Wu, J. Fluoresc. 20 (2010) 261–267. [25] N. Shahabadi, S. Kashanian, F. Darabi, Eur. J. Med. Chem. 45 (2010) 4239–4245. [26] D. Ajloo, M. Sangian, M. Ghadamghahi, M. Evini, A.A. Saboury, Int. J. Biol. Macromol. 55 (2013) 47–61. [27] Z. Tingting, B. Shuyun, W. Yu, W. Tianjiao, P. Bo, G. Tingting, Spectrochim. Acta A 132 (2014) 198–204. [28] G. Zhang, L. Wang, X. Zhou, Y. Li, J. Agric. Food Chem. 62 (2014) 991–1000. [29] G. Zhang, Y. Ma, Food Chem. 141 (2013) 41–47. [30] Z. Tian, Z. Zhao, F. Zang, Y. Wang, C. Wang, J. Photochem. Photobiol. B 138 (2014) 202–210. [31] Y. Li, G. Zhang, J. Pan, Y. Zhang, Sens. Actuators B: Chem. 191 (2014) 464–472. [32] C.Z. Huang, Y.F. Li, P. Feng, Talanta 55 (2001) 321–328. [33] G.M. Morris, D.S. Goodsell, R.S. Halliday, R. Huey, W.E. Hart, R.K. Belew, A.J. Olson, J. Comput. Chem. 19 (1998) 1639–1662. [34] G.M. Morris, R. Huey, W. Lindstrom, M.F. Sanner, R.K. Belew, D.S. Goodsell, A.J. Olson, J. Comput. Chem. 30 (2009) 2785–2791. [35] B.D. Bursulaya, M. Totrov, R. Abagyan, C.L. Brooks, J. Comput. Aided Mol. Des. 17 (2003) 755–763. [36] HyperChem (TM) Professional 7.51; Hypercube, Inc., 1115 NW 4th Street, Gainesville, Florida, USA. [37] A.W. Schüttelkopf, D.M.F. van Aalten, Acta Crystallogr. D60 (2004) 1355–1363. [38] W.D. Cornell, P. Cieplak, C.I. Bayly, I.R. Gould, K.M. Merz Jr., D.M. Ferguson, D.C. Spellmeyer, T. Fox, J.W. Caldwell, P.A. Kollman, J. Am. Chem. Soc. 117 (1995) 5179–5197. [39] P. Ewald, Ann. Phys. 369 (1921) 253–287. [40] D. van der Spoel, E. Lindahl, B. Hess, G. Groenhof, A.E. Mark, H.J.C. Berendsen, J. Comp. Chem. 26 (2005) 1701–1718. [41] Z.S. Yang, Y.L. Wang, G.C. Zhao, Anal. Sci. 20 (2004) 1127–1131. [42] J.A. Ewing, I.D. Kuntz, J. Comput. Chem. 18 (1997) 1175–1189. [43] D.E. Schmechel, D.M. Crothers, Biopolymers 10 (1971) 465–480. [44] E. Froehlich, J.S. Mandeville, C.J. Jennings, R. Sedaghat-Herati, H.A. Tajmir-Riahi, J. Phys. Chem. B 113 (2009) 6986–6993. [45] Y. Li, Y. Yang, Inorg. Chim. Acta 362 (2009) 4823–4831. [46] Y. Sun, S. Bi, D. Song, C. Qiao, D. Mu, H. Zhang, Sens. Actuators B 129 (2008) 799–810. [47] M. Nejat-Dehkordi, A.K. Bordbar, M. Ayatolahi-Mehrgardi, V. Mirkhani, J. Fluoresc. 21 (2011) 1649–1658. [48] Y. Zhang, J. Pan, G. Zhang, X. Zhou, Sens. Actuators B: Chem. 206 (2015) 630–639. [49] D. Agudelo, P. Bourassa, G. Bérubé, H.A. Tajmir-Riahi, Int. J. Biol. Macromol. 66 (2014) 144–150. [50] S. Ramakrishnan, M. Palaniandavar, J. Chem. Sci. 117 (2005) 179–186.
