A new bio-active asymmetric-Schiff base: synthesis and evaluation of calf thymus DNA interaction, topoisomerase IIα inhibition, in vitro antiproliferative activity, SEM analysis and molecular docking studies

Abstract In this paper, the asymmetric-Schiff base 2-(4-(2-hydroxybenzylideneamino)benzylideneamino)benzoic acid (SB-2) was newly synthesized and characterized by various spectroscopic methods. The interaction of SB-2 with calf thymus DNA was investigated by UV–vis, fluorescence spectroscopy and molecular docking methods. It was determined that SB-2 effectively binds to DNA via the intercalation mode. DNA electrophoretic mobility experiments displayed that topoisomerase IIα could not cleave pBR322 plasmid DNA in the presence of SB-2, confirming that the Schiff base acts as a topo II suppressor. In the molecular docking studies, SB-2 was found to show an affinity for both the DNA-topoisomerase IIα complex and the DNA. In vitro antiproliferative activity of SB-2 was screened against HT-29 (colorectal) and HeLa (cervical) human tumor cell lines by MTT assay. SB-2 diminished the cell viability in a concentration- and incubation time-dependent manner. The ability of SB-2 to measure DNA damage in tumor cells was evaluated with cytokinesis-block micronucleus assay after incubation 24 h and 48 h. Light and scanning electron microscopy experiments of tumor cells demonstrated an incubation time-dependent increase in the proportion of apoptotic cells (nuclear condensation and apoptotic bodies) suggesting that autophagy and apoptosis play a role in the death of cells. Based on the obtained results, it may be considered that SB-2 is a candidate for DNA-targeting antitumor drug. Communicated by Ramaswamy H. Sarma GRAPHICAL ABSTRACT


Introduction
Cancer is a class of multigenic disease that has become the major cause of worldwide deaths, often characterized by a variety of genetic and epigenetic changes and treated by inhibiting proliferation of cancer cells through inhibition of DNA replication (Nartop et al., 2008). The targeting of DNA in cancer cells has been identified as the primary intracellular target for anticancer drug design and remains one of the most promising biological receptors for the development of chemotherapeutic agents (G€ ung€ or & G€ urkan, 2010).
Despite the great success of cisplatin and its derivatives, their use has been restricted due to severe systemic toxicity, intrinsic drug resistance (due to off-target binding of the drug) and high cost problems (G€ ung€ or, 2017;G€ ung€ or & G€ urkan, 2019;Nartop et al., 2017). Such DNA-targeting drugs act by covalently binding to nucleobases, but have very low selectivity (G€ ung€ or & G€ urkan, 2010).Therefore, in order to develop more effective drugs, new DNA targeted drugs with high selectivity, low toxicity and pharmacokinetic profiles need to be developed that can produce damage to DNA.
Asymmetric-Schiff bases of the formula R-N@CH-Ar-N@CH-R can be synthesized employing a two-step procedure (G€ ung€ or, 2017;G€ ung€ or & G€ urkan, 2010;Nartop et al., 2008). In this procedure, the nitro group of the nitro-Schiff base has been reduced into amino group with sodium dithionite (Na 2 S 2 O 4 ) and the new imine bond is formed upon addition of aromatic aldehyde. In literature, their antimicrobial (G€ ung€ or & G€ urkan, 2019;Nartop et al., 2017;Nartop & € O g€ utc€ u, 2020;Nartop & € O g€ utç€ u, 2020) and antimutagenic  properties have been investigated, but information on their DNA binding, topoisomerase II inhibition activity and the cytotoxicity is still rare.
Based on the mentioned properties for Schiff bases, we focused to develop a new Schiff base as a DNA-targeted anticancer drug candidate. The asymmetric-Schiff base (SB-2) was synthesized for the first time via a two-step method starting from anthranilic acid. Here, the interaction of the compound with DNA was tested by UV-vis, fluorescence spectroscopy and molecular docking studies. The ability of SB-2 to inhibit topoisomerase IIa enzyme was evaluated by utilizing gel electrophoresis and molecular docking studies. The Schiff base was analyzed for anticancer activity against HT-29 and HeLa cancer cell line by MTT assay. The mechanism of its action was investigated through light and scanning electron microscopies.
Melting points were determined on Barnstead Electrothermal BI 9200 (Gazi University). Elemental analysis was obtained on LECO CHNS-932 analyzer (Ankara University). FT-IR spectra were measured in the range of 4000-400 cm À1 on Thermo Scientific Nicolet iD5 ATR spectrometer (Gazi University). 1 H and 13 C NMR spectra were taken in DMSO-d 6 at room temperature on a Bruker Ultrashield 300 MHz (Gazi University). The chemical shifts were in units of ppm relative to TMS. The mass spectrum was recorded on a Waters 2695 Alliance Micromass ZQ LC/MS mass spectrometer (Ankara University) using the negative ion electrospray ionization modus (ESI). UV-visible spectra were performed at room temperature using Analytik Jena UV-200 spectrophotometer (Gazi University). Fluorescence spectra were recorded at room temperature by using Hitachi F-4500 fluorescence spectrophotometer (Gazi University).

UV-visible spectroscopy
The stock solution of calf thymus (CT) DNA in buffer (5 mM Tris-HCl/50 mM NaCl, pH 7.2) gave a ratio by microplate reader (Epoch, BioTek) at 260 and 280 nm of calculated 1.8-1.9, showing that DNA was enough free of protein (Marmur, 1961). DNA concentration was determined by absorption spectroscopy with the molar extinction coefficient of 6600 M À1 cm À1 at 260 nm (Gultneh et al., 1999). The absorption spectra of SB-2 (35 lM) were taken with increasing amounts of CT-DNA (5 to 120 lM) in references buffer solution by UV-visible spectrophotometer (PG Instruments, T60) . To quantitatively illustrate the result, the absorption data were analyzed to judge the binding constant (Kb), which can be determined from the following the Wolfe-Shimer equation (1) (Pyle et al., 1989;Thompson & Orvig, 2006;Wolfe et al., 1987):  (Borowska et al., 2015;Jeyalakshmi et al., 2014).
In equation (2) (Pravin et al., 2016;Shahabadi et al., 2009), A Free stands the absorbance intensity of the Schiff base in the free state, while A Bound stands the absorbance intensity of the Schiff base after the DNA is added at the maximum concentration.

Fluorescence spectroscopy
This study was performed in order to investigate whether SB-2 can compete with ethidium bromide (EB) for the DNA intercalating sites by displacing it from its ethidium bromide bound CT-DNA. The EB-DNA complex was prepared by adding 20 mM EB and 26 mM CT-DNA in buffer solution (5 mM Tris-HCl/50 mM NaCl, pH 7.2). The intercalating effect of SB-2 was tested by adding a certain amount of a solution of SB-2 step by step into a solution of the EB-DNA complex. For each addition, the sample was shaken and allowed to stand for 10 min. Then, the emission spectra were recorded at excitation wavelength (k ex ) of 540 nm (Tserkezidou et al., 2016).
The values of the Stern-Volmer quenching constant (K sv , in M À1 ) have been calculated according to the linear Stern-Volmer equation (equation (3)) and the plots Io/I vs.
[Q]. This equation is the following (Varna et al., 2018;Zhao et al., 1998): where I 0 is the fluorescence intensity in the absence of SB-2, I is the fluorescence intensity in the presence of SB-2, and [Q] is the concentration of the SB-2.

Topoisomerase IIa inhibition study
DNA topoisomerase IIa inhibitory activity was monitored by relaxation of supercoiled pBR322 DNA in the presence of SB-2. For the relaxation assay, the reaction mixture containing 200 ng of supercoiled pBR322 plasmid DNA (Thermo Scientific) and 1 unit of human topo IIa (TopoGEN, USA) was incubated in the absence/presence of the Schiff base (500-3500 mM) in the assay buffer (0.5 M Tris-HCl, pH 8.0 containing 1.5 M NaCl, 0.1 M MgCl 2 , 5 mM dithiothreitol, 1 mM ATP) for 30 min at 37 C. And the reaction in a final volume of 20 mL was quenched by adding 3 mL of 7 mM EDTA. The reaction mixtures were examined on 1% agarose gel at 25 V for 4 h with TAE as the running buffer. After electrophoresis, the gel was dyed for 30 min in a hydrous solution of ethidium bromide (0.5 mg/mL). DNA bands were visualized by transillumination with UV light, and the supercoiled DNA was quantified using the Gel Documentation System (Bio-Rad, Image Lab TM Software Program, Version 5.2.1) (Jun et al., 2011;Xue et al., 2012;Zeglis et al., 2011). One unit of topo II activity was described as the amount of the enzyme required to relax 50% of the supercoiled pBR322 DNA under the standard test condition (Arjmand et al., 2011).

Molecular docking studies
AutoDockTools4.2 and AutoGrid4 (Morris et al., 2009) were used to prepare ligand and receptor structures for docking and AutoDock4 (Forli et al., 2016;Morris et al., 2009) was used to perform docking between SB-2, the DNA-human topoII complex (PDB ID: 5GWK) (Karatas et al., 2021;Wang et al., 2017) and the DNA (PDB ID: 1BNA) (Drew et al., 1981). Human Topo-II-DNA complex with Etoposide (EVP) and DNA was downloaded from Protein Data Bank with 3.15 Å and 1.90 Å resolutions respectively. EVP was isolated from the structure and then docked with the complex to validate the method. SB-2 structure prepared and cleaned for 3 D visualization with MarvinSketch (Marvin, 2020

MTT assay
MTT assay was utilized for testifying the cell viability of SB-2 on HT-29 and HeLa tumor cells with respect to standard procedure (Mosmann, 1983). Cells were seeded into a 96well plate at a density of 1 Â 10 4 cells/well and treated of the different concentrations (50,100,150,200,250, 500 and 1000 mM) with the Schiff base for 24 h, 48 h and 72 h incubation at 37 C (5% CO 2 ). Schiff base was dissolved in DMSO, and diluted with culture medium to the final assay concentrations (0.5% v/v DMSO). As controls, cells were treated with DMEM (negative control) and DMEM-0.5% DMSO (positive control). In this assay, cells were plated in 96-well plates at a density of 1 Â 10 4 cells/well (200 lL/well). The cells were incubated at 37 C in a 5% CO 2 incubator for 24, 48 and 72 h, followed by the addition of 20 lL of MTT solution (5 mg/mL in PBS) to each well and incubated for 4 h at 37 C. After incubation, the media with MTT was removed from each wells, and the formazan crystals formed were solubilized adding 200 lL of DMSO at room temperature. Formation of formazan was measured at 570 nm by a microplate reader (Epoch, Biotek, Winooski, VT, USA) and the absorbance was correlated with cell number. The antiproliferative effect was evaluated by comparing the viability of treated sample with the untreated control. The IC 50 values were determined by plotting the percentage viability versus concentration on a linear graph and reading off the control (Rubner et al., 2010;Scheffler et al., 2010). By means of the equation (4): Viability % ¼ ðMean absorbance in samplesÞ Mean absorbance in control groups Â 100 (4)

Cytokinesis-block micronucleus (CBMN) assay
CBMN test was done according to the original procedure (Fenech, 2007) with marginal modifications for adaptation to HT-29 and HeLa cells. HT-29 and HeLa cells were seeded into 6-well tissue culture plates (SPL Life Sciences) at 5 Â 10 5 cells/well (2.0 mL/well) and treated with SB-2 (1000 mM) for 24 h and 48 h incubation at 37 C. Instantly after incubation, the solution of the Schiff base was changed with fresh media containing cytochalasin-b (6 lg/ml) for another 27 h to inhibit cell division after mitosis. Then, the cells were washed twice with PBS and separated with trypsin-EDTA. Following by centrifugation, the cells were resuspended in medium and then fixed with methanol and stained with 10% Giemsa for 20 min for nuclei. Cells were examined under a a light microscope (Leica) for micronuclei (MN) at 1000 Â magnification with respect to prior reports (Fenech, 2007) and MN were scored in 500 binucleated cells (BNC) for 1000 lM concentration for every test. In addition, cells were then analysed to determine microscopically changes in cell morphology compared to controls. Negative control (cells without treatment) was included in each experiment.
For the analysis of cell-cycle progression, 500 cells per treatment were scored for the presence of one, two or more than two nuclei and the cytokinesis-block proliferation index (CBPI) was computed as follows (equation (5)): where 1 N is number of cells with one nucleus, 2 N with two nuclei, >2 N with more than two nuclei and TC is the number of cells analyzed (Fenech, 2000;Peng et al., 2015;Pepe et al., 2013;Titenko-Holland et al., 1997). Cytotoxicity was appreciated by cytostasis% and the replication index (RI) using the equations (6-7): where t and c are treated and control samples, respectively ( € Ozyurt et al., 2013;Peng et al., 2015).

Scanning electron microscopy (SEM) study
To prepare the cells for SEM, HT-29 and HeLa cell lines were seeded into 6-well tissue culture plates at 5 Â 10 5 cells/well (2.0 mL/well) and treated with SB-2 (1000 mM) for 48 h and 72 h incubation at 37 C (5% CO 2 ). Immediately after the in vitro treatment, the cells were fixed in 2.5% glutaraldehyde solution at 4 C overnight. Then, the cells were gently washed in sodium phosphate buffer (pH 7.2) for 10 minutes at 4 C. This step was repeated two times. Washed cells were dehydrated in ascending series of ethanol. After critical point drying with CO 2 (Polaron, CPD 701 Critical Point Dryer), samples were coated with gold in a Polaron SC502 sputter coater. The coated samples were examined with Jeol JSM-6060LV SEM at an accelerating voltage at 10 kV. Photographs were taken from randomly selected areas ( € Ozyurt et al., 2013;Suludere et al., 2009).

Statistical analysis
For the cell viability study, six replicates were utilized in all experiments and values are stated as means ± standard (SD) deviation. One way ANOVA test was utilized to compare of mean values between groups. To assess the difference between the groups, in the cytotoxicity test, Dunnett C and Games-Howell tests were done for 24 and 72 hours on the HT-29 cell line and for 24, 48 and 72 hours on the HeLa cell line. Tukey and Bonferroni post hoc tests were done on the HT-29 cell line for 48 hours and multiple comparisons were made. P values of less than 0.05 were considered to be statistically significant. Data were analysed using an SPSS statistical software package version 22.

Characterization
As depicted in Scheme 1, the nitro-Schiff base SB-1 was synthesized by condensation of 4-nitro-benzaldehyde with 2amino-benzoic acid at 1:1 mol ratios. The novel asymmetric-Schiff base SB-2 was prepared via selective reduction of nitro group of SB-1, and later the condensation reaction with salicylaldehyde. The structures of the compounds were identified by using FT-IR, 1 H NMR, 13 C NMR, and UV-vis spectral studies. The newly Schiff base SB-2 was also characterized by elemental analysis and LC-MS spectrum.
The main absorption bands for the compounds were listed in an experimental section (Figures S1-2). In the infrared spectra of SB-1, the strong sharp band at 3333 cm À1 was assigned to the (O-H) stretching vibration frequency of the carboxyl group. The strong broad band in the region of 3570-3100 cm À1 was due to the (O-H) stretching vibration of the phenolic and the carboxylic groups of SB-2, which form (O-HÁÁÁN) type of intramolecular hydrogen bonds with imine group. The band at 1710 cm À1 was attributed to the (C@O) of the carboxyl group for SB-1. For SB-2, (C@O) stretching frequency appeared at 1660 cm À1 . The imine (C@N) stretching band was observed at about 1654 cm À1 and 1607 cm À1 for SB-1 and SB-2, respectively. This result proved that Schiff bases exist as the phenol-imine tautomer in the solid state (Scheme 2). The absorption bands at 1610/ 1487 cm À1 and 1580/1455 cm À1 were attributed to the aromatic (C@C) frequencies. The bands in the region of 3078-2850 cm À1 belonged to the aromatic and the iminic (C-H) stretching vibrations. The strong bands at 1287/ 1188 cm À1 and 1148 cm À1 were corresponded to the carboxylic and the phenolic (C-O) stretchings, respectively. In addition, the bands at 1524 cm À1 and 1350 cm À1 were assigned to (ONO) as and (ONO) s stretchings for SB-1, respectively. 1 H and 13 C NMR spectra ( Figures S3-4) of Schiff bases were taken in DMSO-d 6 solution, and the spectral data were given in an experimental section. In 1 H NMR spectrum of SB-1, the carboxylic OH proton signal was seen at d 10.15 ppm. The imine (CH@N) proton signal was observed at d 8.45 ppm as a doublet. The coupling constant value 3 J ( NH,H ) ¼ 6.5 Hz on imine proton confirmed that the carboxylic group is engaged in an intramolecular hydrogen bond with the imine nitrogen atom. In 1 H NMR spectrum of SB-2, the phenolic OH proton gave two singlet at d 13.00 ppm and d 13.25 ppm (G€ ung€ or & G€ urkan, 2010) ( Figure S5), which are the typical chemical shift of a medium strong intramolecular hydrogen bond with the imine nitrogen atom. The signal at d 10.75 ppm was due to the shift of the carboxylic OH proton involved in the intramolecular hydrogen bonding. The singlet peaks at d 8.95 ppm and d 9.70 ppm were assigned to different imine protons confirming that the phenol-imine tautomer predominates in DMSO solution.
In 13 C NMR spectrum ( Figure S4) of SB-1, the peaks at d 164.0 ppm and d 170.5 ppm belonged to the imine (C@N) and the carboxyl (COOH) carbon atoms, respectively. For SB-2 (Figure 1), the phenolic, the iminic and the carboxylic carbons were observed at d 151.5 ppm, d 161.1 ppm and d 162.2 ppm, respectively. The resonance within the range of d 114-136 ppm was attributed to phenyl carbon atoms.
The electron impact mass spectrum ( Figure S6) of SB-2 was recorded at 70 eV. The liquid chromatography-mass spectral pattern of this Schiff base displayed the highest intensity fragment at m/z 212 (M À 132, 100%) which corresponds the loss of (C 8 H 6 NO) radical from the molecule.
The electronic absorption spectra ( Figure S7) of the Schiff bases were recorded in DMSO solution and data were presented in an experimental section. The absorption band at 339 nm and 331 nm was attributed to p!p Ã transition of the imine and phenolic chromophore. The spectrum of SB-2 Scheme 2. The phenol-keto tautomeric equilibria in the asymmetric-Schiff base SB-2.  exhibited a weak shoulder at 426 nm additionally, which is assigned to n!p Ã transition of carbonyl chromophore of the carboxyl group. When SB-1 solution was acidified, the band at 339 nm exhibited small bathochromic shifts (Dk max ¼ 3-4 nm for acetic acid; Dk max ¼ 2-3 nm for hydrochloric acid) indicating the protonation of imino nitrogen atom. Also, the absorbance intensity decreased gradually ( Figure  S8a-b). Upon addition of base solution, this band shifted to shorter wavelength (308 nm) due to the deprotonation of acidic COOH proton ( Figure S8c). Adding of water did not cause any noticeable changes in its spectrum ( Figure S8d). This result suggested that SB-1 does not undergo hydrolysis in DMSO solution at different volume ratios of water. For SB-2, the band at 331 nm gave small shifts and the shoulder at 426 nm disappeared when acid or base was added to each of the solution in DMSO ( Figure S9a-b). In contrast, no significant change was observed in the spectrum of this Schiff base upon addition of water ( Figure S9c).

UV-visible spectroscopic study
DNA is one of the most important biological targets of antitumor drugs. For this purpose, DNA binding affinity of SB-2 was studied by UV-vis spectroscopy. The absorption spectrum of SB-2 displayed two maxima and a shoulder at 227, 330 and 426 nm in the absence of calf thymus DNA (CT-DNA). As given in Figure 2, the bands gave no shifts with the progressive addition of DNA to the Schiff base, while a new band was seen at 275 nm. Upon gradually increasing the concentration of DNA, the absorbance intensity of all the bands increased (hyperchromism). After adding of 40 lM concentration of DNA, the bands at 227 and 275 nm overlapped and converted to a maximum, and this new band red-shifted to 295 nm (bathochromism) in each step. It is known that a weaker binding to DNA does not generally lead to a shift in wavelength (Arslantas & Agirtas, 2017). Therefore, it was suggested that SB-2 is strongly bound to DNA. The obtained results indicated that SB-2 may bind to DNA via intercalative mode. This was related to the existence of two imine groups of phenol tautomer in the structure which is directly bound to DNA via intercalation mode. On the other hand, free carboxyl and hydroxyl groups may also increase DNA interaction through the forming hydrogen bonding. The binding interaction between SB-2 and DNA may result characteristic hyperchromic effect in the absorption spectrum, which is well supported in literature (Li et al., 1996;Shahabadi et al., 2011). The hyperchromicity feature of intercalation mode may be explained by that the double helix DNA is damaged after interaction with Schiff base via an intercalation including a powerful pi-pi stacking between chromophores of molecule and the base pairs of DNA duplex (Shi et al., 2006).
To research on the intensity and the mode of the interaction between SB-2 and CT-DNA, the intrinsic binding constant (K b ) was determined from the plot of [DNA]/ea À ef  versus of variant concentration of DNA. The K b value was computed to be 1.5 Â 10 5 M À1 (Table 1). This Kb value was found to be similar to classical intercalator such as ethidium bromide (3.3 Â 105 M) (Sedighipoor et al., 2018) and be higher than those of some reported Schiff bases with CT-DNA. For example, a Schiff base was synthesized by 2,4-dihydroxybenzaldehyde and benzoylhydrazide whose K b value with DNA was 5.2 Â 10 3 M À1 (Aboafia et al., 2018). A Schiff base was synthesized by 3,5-di-tert-butyl-2-hydroxybenzaldehyde and 4,4 0 -methylenedicyclohexan-amine whose K b value was 0.94 Â 10 5 M À1 (Ali et al., 2020). A new symmetrical-Schiff base was synthesized by 2-hydroxybenzaldehyde and polyamine (2,2 0 -((1,4-diazepane-1,4-diyl)bis(methylene))dianiline) whose K b value was 2.5 Â 10 4 M À1 (Keypour et al., 2019). Schiff bases synthesized by pyridine-2-carboxaldehyde with 4-hydroxyaniline and amino-3,4-dimethylisoxazole whose K b values were 4.06 Â 10 4 M À1 and 4.13 Â 10 4 M À1 (Al-Khathami et al., 2019). According to the values of K b mentioned above, these Schiff bases were considered to interact with DNA as an intercalating ligand. This means that SB-2 can bind to CT-DNA through intercalation mode. Hence, it shows that the affinity of SB-2 to DNA is more stronger than that of the above Schiff bases. The free energy (DG ¼ ÀRTInK b ) of SB-2-DNA interaction was calculated to be À29.53 kJ mol À1 .

Fluorescence quenching study
The intercalative binding mode between the CT-DNA and SB-2 can be substantiated by fluorescence quenching study. This study was carried out using emission spectral titrations. Ethidium bromide (EB) is a typical DNA intercalator (Varna et al., 2018). EB is a weak fluorescent compound, but EB bound CT-DNA exhibits an intense emission band at 592-593 nm upon excitation with k ex ¼ 540 nm. This is explained by binding to DNA base pairs in an intercalation of planar EB-phenanthridine ring. When another intercalating agents are added into EB-CT-DNA solution, these agents compete with EB bound DNA and displace the EB from DNA (Tserkezidou et al., 2016). This causes the quenching of the fluorescence/emission at 592-593 nm.
In this work, the emission spectra of EB-CT-DNA complex solution were recorded in the absence and presence of increasing concentrations of SB-2. As seen in Figure 3, the emission intensity of EB-DNA at 590 nm decreased gradually upon addition of SB-2. The extent of the observed quenching may reveal high EB displacing ability of SB-2 which is in competition to EB when binding to DNA and may indicate indirectly the existence of intercalation of the complexes to CT-DNA. The relative binding constant of SB-2 was  determined by Stern-Volmer equation (Mariappan et al., 2012;Ramesh et al., 2020). K sv is the Stern-Volmer quenching constant which is calculated from the slope to intercept of plot I 0 /I versus [Q]. K sv value of SB-2 was found to be 2.42 Â 10 4 M À1 . This K sv value was determined to be higher than some reported Schiff base complexes. For example, the K sv values of Ni(II) and Co(II) metal complexes of 6-aminobenzothiazole Schiff base were reported as 6.11 Â 10 3 M À1 and 1.12 Â 10 3 M À1 , respectively (Vamsikrishna et al., 2020). These findings support that SB-2 exhibit high fluorescence quenching activity and binds to CT-DNA by intercalation which are consistent with UV-vis spectral results.

Topoisomerase IIa inhibition activity
DNA topoisomerases are one of the major molecular targets of antitumor drugs because of the essential role of these enzymes in quickly proliferating tumor cells that triggers a series of cellular events, inducing apoptosis and eventually causing cell death (Jarvinen & Liu, 2006;Sandhaus et al., 2016). Particularly, topoisomerase II inhibitors are among the most effective anticancer drugs currently used in clinical application (Guerrant et al., 2012). Thus, the inhibition effect of SB-2 on the activity of human topoisomerase IIa by agarose gel electrophoresis (pBR322 DNA) were investigated.
In this study, each type of DNA shows a different behavior of the enzyme: relaxed DNA reveals that the enzyme's isomerase activity remains intact; supercoiled DNA suggests that the enzyme's action was inhibited (Suzuki & Uyeda, 2002). To understand the inhibition effect of SB-2 on topoisomerase IIa function, the procedure reported by Lee et al (Lee et al., 1989) was used. Gel mobility assay was performed by incubating pBR322 plasmid DNA and Topo II enzyme with increasing concentration of SB-2 (500-3500 lM). As presented in Figure 4, lane 1 was assigned to control DNA, which contains a mixture of main supercoiled DNA (Form I) and a small amount of nicked open circular form (Form II) (Venugopal et al., 2019). Lane 2 was assigned to the plasmid DNA treated with Topo II. Lanes 3-7 were related to the interaction between the plasmid DNA and Topo II in presence of SB-2. The supercoiled DNA was relaxed completely and turned into nicked form in case of Topo II (Lane 2). Upon incubation of Topo II with SB-2, the enzyme showed no remarkable DNA cleaving ability. Topo II could not cause cleavage of the supercoiled and nicked forms, and no linear bands were detected within the concentration range of 500-3500 mM. Its catalytic activity was inhibited by increasing amounts of SB-2. A significant reduction was observed in DNA relaxation at > 2000 lM concentration. Especially, level of nicked form was quenched at 3500 lM concentration which was ascertained by the appearance of supercoiled form only (Lane 7). Especially at 3000 and 3500 lM concentrations of SB-2, a significant reduction in the DNA relaxation was sighted compared to other concentrations. Therefore, it can be said that SB-2 shows better human topo IIa inhibition activity at these concentrations. In addition, this study was conducted to understand the mechanism of action of SB-2, to determine whether this compound could act as topoisomerase II poisons. If a compound acts as a Topoisomerase suppressor, the cleavage cycle catalysed by topoisomerase is blocked by the compound and supercoiled DNA will stay intact. In other words, Topo II catalytic inhibitors interact non-covalently with enzyme to block enzymatic activity during the catalytic cycle, so, change their normal functioning resulting in cell death (Guerrant et al., 2012). Contrarily, if the compound behaves as a topoisomerase poison, takes place DNA cleavage by binding to the Topo-DNA complex covalently, but the continuity of the recovery of the DNA strand will be blocked because of the formation of a stable cleavage enzyme-inhibitor-DNA complex (Konkolova et al., 2018). In our study, no linear bands were detected at 500-3500 lM concentrations of SB-2. These results suggested that SB-2 behaves as a Topo IIa suppressor, not poison. Similarly, heterobimetallic Cu(II)-Zn(II) and Zn(II)-Sn(II) complexes of a Schiff base derived from salicylaldehyde and 5aminosalicylic acid were categorised under catalytic Topo II inhibitors dependent on reduction in the DNA relaxation at 25 lM concentration (Arjmand et al., 2012). On the other hand, the comparison of SB-2 with classical topoisomerase inhibitor doxorubician (Suzuki & Uyeda, 2002) displayed that SB-2 has little inhibition effect on the cleavage pattern of Topo IIa.

Molecular docking studies
In order to determine the observed Topo IIa inhibition study with SB-2, molecular docking studies were performed to understand the binding mode of SB-2 with human-DNA-Topo II complex. To validate method and investigate the possible binding modes, EVP docked with human Topoisomerase-II-DNA complex (PDB ID: 5GWK). As given in Figure 5, AutoDock4 software was able to dock EVP. Docking scores were found to be 0.78 RMSD value and À10.9 kcal/ mol binding energy. The high negative binding energy indicates the greater binding with the targeted molecule. The ligand interactions were calculated by Discovery Studio 2021 of docked conformation and original conformation. Figure 6 displayed that EVP interacts with human Topo-II-DNA complex with two alkyl and pi-alkyl interactions at MET766 and ARG487, three amide-pi stacked and pi-pi stacked interactions at ARG487 and DG13. EVP can form four hydrogen bonds with the amino acid residues (ASP463, GLY462 and DT9). Docking results indicated that alkyl and pi-alkyl interactions at MET762 and PRO803, hydrogen bonds at DG10 and DC8 may also be important in order to human Topo-II inhibition.
Topoisomerase II enzymes are known to relax both strands of DNA duplex. According to the catalytic mechanistic pathway, they use tyrosine residues as an active site to attack the phosphodiester backbone of the DNA, which generate a staggered double-strand break (Afsan et al., 2020). An important feature of this mechanism is that while the DNA strand is broken, the enzyme is bind to it non-covalently. However, this reversibly binding interaction is reduced or inhibited selectively by Topo II inhibitors (Jaswal et al., 2020).
In order to determine the binding free energy, binding mode and mechanism of antiproliferative activity, SB-2 docked predefined binding site of human Topo-II. The binding model between SB-2 and human Topo-II-DNA complex was illustrated in Figure 7. Analyses of binding modes of SB-2 indicated that it has six attractive charges, pi-cation and pi-anion interactions with ARG487, DT9, DC8, DG13 and DA12. The imine and carboxyl groups of SB-2 have a position to interact with DC8 and DT9 by forming a hydrogen bond. SB-2 also effectively binds to DG13, DA12 and DC8 with these groups through four pi-pi interactions both stacked and T-shaped. The binding free energy of the Schiff base was found to be À7.68 kcal/mol. As a result, SB-2 was found to show affinity for the DNA-topoisomerase IIa complex. In addition, it has been determined that SB-2 interacts with the enzyme in a non-covalently way, thereby blocking the enzymatic activity. These results therefore confirmed that SB-2 behaves as a Topo IIa suppressor and the results of the topoisomerase IIa inhibition activity study.

Molecular docking with DNA
SB-2 was also docked with B-DNA (PDB ID: 1BNA) in order to confirm and elucidate the observed spectroscopic results and to check the intercalation pattern. UV-vis and fluorescence spectroscopic results indicated that SB-2 bind to DNA via the intercalation mode. In order to verify these results, molecular docking calculations were employed. SB-2 was docked with B-DNA (PDB ID: 1BNA). The binding pocket between SB-2 and DNA are presented in Figure 8. Docking studies revealed that SB-2 has binding affinity towards human Topo-II from an important residue for inhibition, ARG487. SB-2 also interacts with DNA cleaved by human Topo-II. This is due to the electrostatic interaction (pi-anion, pi-cation, attractive charges) between the Schiff base and DNA bases. SB-2 also forms seven hydrogen bonding with DA6, DT8, DC9, DA18, DT19 and DT20, an attractive charge interaction with DT20. The effective stacking between the imine, carboxyl and hydroxyl groups of the Schiff base and DNA base leads to an increase in the efficiency of these interactions. The binding energy was found to be À7.49 kcal/ mol. Literature reveals that the forces maintaining the stability of DNA-intercalator complex include van der Waals, hydrogen bonding, hydrophobic charge transfer and electrostatic complementarity (Baginski et al., 1997;Proudfoot et al. 2001).

Antiproliferative activity study
In vitro antiproliferative activity of SB-2 on the growth of human cancer cell lines (HT-29 and HeLa) was evaluated by using a MTT-dye reduction procedure. Schiff base (50-1000 mM) was applied to the cells at different concentrations and for different incubation times (24-72 h). The results were compared to the solvent control (untreated cell in DMSO).
On the basis of Figure 9, it was determined that SB-2 has an antiproliferative effect in the all dose range used. The cell viability decreased on increment in the concentrations of the Schiff base and long-term incubation time. SB-2 displayed the best cytotoxicity against HT-29 cells among the other  cancer cell lines. For HT-29 cell lines, the cell-death value was found to be within the range of 35 À 56%, 36 À 56% and 36 À 96% after 24 h, 48 h and 72 h exposure time. SB-2 exhibited the highest antiproliferative effect against HT-29 cell at concentrations of 500 mM and 1000 mM after 72 h, and it achieved to kill 94% and 96% percentages of cells. In the concentration range of 50-1000 mM, SB-2 killed 15 À 44%, 28 À 48%, 28 À 65% percentages of HeLa cell lines after incubation 24 h, 48 h and 72 h, respectively.
As listed in Table 2, the 50% inhibition concentration (IC 50 ) of SB-2 was calculated to be 662 mM (40 lg/mL) and 1040 (62 lg/mL) for HT-29 and HeLa cell lines after 24 h treatment. For selected tumor cells, IC 50 value of doxorubicin was reported as 0.58 mg/mL and < 0.068 mg/mL after 24 h incubation, respectively assessed MTT assay (Kraicheva et al., 2012). Cisplatin's IC 50 value was recorded as 70 lM (Stockert et al., 2014) and 12.2 lM (Reytman et al., 2012) on HT-29 and HeLa cells. In comparison to standard antitumor drugs, it was suggested that SB-2 is more cytotoxic than doxorubicin and less cytotoxic than cisplatin on selected tumor cells.
According to pharmacological studies of Schiff bases, it has been noted that free Schiff bases have no antiproliferative effect against tumor cells ( € Ozdemir et al., 2020) or their antitumor activity is lower than metal complexes ( € Ozdemir et al., 2020). From these results, it may be said that SB-2 has potential to inhibit HT-29 and HeLa cells, and it can be categorized as high cytotoxic on HT-29 cell lines for 72 h incubation time especially.

Cytokinesis-block micronucleus (CBMN) assay
Cytokinesis-blocked micronucleus (CBMN) assay is one of the most commonly used cytogenetic methods for measuring DNA damage in human or mammalian cells (Fenech, 2000;2007). This assay is a multi-endpoint assay that assesses DNA damage endpoints (in the form of binucleated cells with micronuclei (MNBNC) as well as other cellular incidents (such as necrosis, apoptosis, and cell proliferation) simultaneously. The cytokinesis-block proliferation index (CBPI) and the replication index (RI) are utilized to forecast the cytotoxicity. The CBPI and the proportion of binucleated cells are useful parameters for comparing the mitogenic response of lymphocytes and cytostatic effects of agents examined in the assay (Fenech, 2000). A decrease in the CBPI reflects an inhibition in the cell cycle and reduction in the proliferation capacity of cell. The RI is a criterion used to determine the effect of the agent applied on cells on DNA replication (Fenech, 2000).
The capability of SB-2 to induce chromosome and genome damage was investigated by using CBMN assay. After incubation 24 h and 48 h, the frequency of micronuclei in binucleated cells (MNBNC) was not increased in HT-29 cells   treated with SB-2. While the micronucleus frequency 1.5-fold increased on HT-29 cells for 48 h, it 1.28-fold increased on HeLa cells for 24 h. According to this, it was suggested that SB-2 is more effective against HT-29 cancer cell. This was also supported by the results of the antitumor activity study. Comparison with the control showed that CBPI value decreased after 48 hours of application time on HT-29 cell lines. Thus, SB-2 showed more antiproliferative activity against HT-29 cells for 48 h compared to 24 h. For HeLa cell lines, a decrease in CBPI was observed at both 24 h and 48 h. The decrease was found to be higher at 24 h indicating the higher inhibition and antiproliferative effect in the cell cycle. Although a reduce in the RI was measured at 1000 lM concentration on HT-29 (except for 24 h) and HeLa cells, it stayed above 55% as submitted in the OECD guideline (OECD, 2010). Results are given in Table 3.

Microscopic analysis of cancer cells
Schiff base was applied to the HT-29 and HeLa cells for 24 h and 48 h intervals at 1000 lM concentration to determine whether the cytotoxic effect of SB-2 was due to apoptosis. Light and electron microscope (SEM) was examined.
As depicted in Figures 10 and 11, light microscopic analysis of HT-29 and HeLa cells showed that the increasing incubation times of SB-2 caused an increase in micronuclei, nucleoplasmic bridges and nuclear buds, formation of cytoplasmic vacuoles and the nuclear buds were being engulfed by these vacuoles, which is bio-markers of genotoxic damage and chromosomal instability (Fenech et al., 2011). In particular, these increases were observed in both cancer cells intensively during 48 hours of application. As shown in Figures  10b and 11b nuclei with condensed chromatin and apoptotic bodies of various sizes containing good-protected, however compressed cytoplasmic organelles and/or nuclear fragments, which are the typical characteristics of apoptosis, were sighted in treated HT-29 and HeLa cells. Multinucleation and apoptotic bodies, which are characteristic of programmed cell death (apoptosis), were also detected in treatment groups. At some studies, similar apoptotic morphological changes have been observed in apoptotic death of cancer cells (Barros et al., 2011;Hernandez et al., 2011;Larsson et al., 2010;Liu et al., 2011;Rao et al., 2011;Vidhya & Devaraj, 2011). The searches for anticancer drugs are aimed at the fact that cancer cells replicate more rapidly than normal cells, and the vast majority of the currently used drugs cause DNA damage, thereby interrupting cell division and subsequently causing cell death (Bailon-Moscoso et al., 2014;Venkatesan et al., 2012).
It was also observed that following treatment with SB-2, formation of cytoplasmic vacuoles was observed in many cells, and the nuclear buds were being swallowed by these vacuoles, suggesting autophagy (Figures 10d and 11e-f). In most studies, autophagic vacuoles were encountered in the treated cancer cell lines (Symonds et al., 2013;Zhivotovsky & Kroemer, 2004). As one of the three types of programmed cell death, autophagy, a catabolic process for the degradation and recycling of organelles and macromolecules which can be activated during stress conditions was assigned as an significant point at the tumor control process (Guan et al., 2016;Shinojima et al., 2007;Xia et al., 2019;Xiao et al., 2013). The recent years, a thriving number of research proof that autophagy may perform a binary role in tumors (Thongrakard et al., 2014). On the one hand, autophagic degradation of amino acids, free fatty acids and nucleotides increase cell survival. On the other hand, it is also a significant mechanism of cancer cell death (White, 2012). Liu et al. reported that curcumin was importantly inhibit the proliferation of the human lung adenocarcinoma A549 cells . It was also reported that curcumin shows anticancer activity through activation of autophagy and apoptosis.
SEM analysis of HT-29 and HeLa cells showed that the increasing incubation times caused an increase in apoptotic cells. At SB-2 treated cells for 48 and 72 hours (IC 50 39-15 lg/mL and 52-23 lg/mL, respectively), distinct morphological changes were observed corresponding to typical apoptotic morphologies. These were characterized by blebbing of the plasma membrane followed by the formation of apoptotic bodies, including cell surface microvilli reduction and decreased number of cytoplasmic extensions (Figures 12  and 13). At some studies, similar morphological changes were detected in apoptotic death of HT-29 (Liang et al., 2018) and HeLa (Wahab et al., 2009) cells. The phenotype of cancer cell death induced by the Schiff base which could largely be related to DNA damage as indicating by micronuclei. As a result, it can be said that SB-2 indicates antitumor activity through activation of autophagy and apoptosis.

Conclusion
In summary, a new Schiff base SB-2 was synthesized and characterized. The obtained in vitro pharmacological screening results of this Schiff base exhibited that: i. SB-2 could strongly interact with CT-DNA via intercalation with high DNA binding affinity based upon UV-vis and fluorescence spectral, and molecular docking studies. ii. SB-2 inhibited DNA cleavage activity of topoisomerase IIa enzyme at a concentration of 500-3500 lM indicating that it acts as a suppressor. Also, in the molecular docking studies, SB-2 was found to show an affinity for the DNA-topoisomerase IIa complex. iii. SB-2 showed promising anticancer efficacy towards HT-29 and HeLa tumor cell lines. iv. The apoptosis or autophagy of tumor cells by SB-2 might largely be related to DNA damage, because of the potential failure of DNA repair machinery.
According to our results, it may be suggested that the new Schiff base SB-2 has potential to be an anticarcinogenic agent with DNA targeting.