Synthesis, structural and thermal properties of the

Journal of Molecular Liquids 230 (2017) 482–495

Contents lists available at ScienceDirect

Journal of Molecular Liquids journal homepage: www.elsevier.com/locate/molliq

Synthesis, structural and thermal properties of the hexapyrrolidinocyclotriphosphazenes-based protic molten salts: Antiproliferative effects against HT29, HeLa, and C6 cancer cell lines Hüseyin Akbaş a, Ahmet Karadağ a,⁎, Ali Aydın b, Ali Destegül a, Zeynel Kılıç c a b c

Department of Chemistry, Gaziosmanpaşa University, 60250 Tokat, Turkey Department of Molecular Biology, Chemistry, Gaziosmanpaşa University, 60250 Tokat, Turkey Department of Chemistry, Ankara University, 06100 Ankara, Turkey

a r t i c l e

i n f o

Article history: Received 31 March 2016 Received in revised form 10 January 2017 Accepted 18 January 2017 Available online 21 January 2017 Keywords: Cyclotriphosphazenes Protic ionic liquids Molten salts Antiproliferative effect

a b s t r a c t Novel three protic ionic liquids or protic molten salts (PILs) (1a, 1b, and 1c) were obtained from the reactions of hexapyrrolidinocyclotriphosphazenes, [N3P3(NC4H8)6] PYR (1), with the gentisic (2,5-dihydroxy benzoic), decanoic and boric acids. The structures of PILs were determined by elemental analyses, FTIR and 1H, 13C{1H}, 31 1 P{ H} NMR techniques. The thermal properties of PYR (1) and PILs were described using thermogravimetric analysis (TGA). The results obtained from TGA indicate that the onset temperatures of PYR (1), 1a, 1b, and 1c are 303.38, 256.82, 166.81, and 287.35 °C, respectively. Binding interaction of these protic ionic liquids with calf thymus (CT-DNA) and bovine serum albumin (BSA) were evaluated by UV–vis spectrophotometry, fluorescence spectroscopy techniques, and electrophoresis measurements. Biological properties of the compounds were evaluated by using in vitro techniques towards three cancer cell lines (HT29, HeLa, and C6) and one non-cancer cell line, namely Vero. The IC50 data exhibitions that 1b and 1c are the most effective antiproliferative agents against HT29, C6 cells and HeLa cells, respectively. The cytotoxic effects of the compounds were detected to be in a safe level at 75 or 100 μg/mL. The analysis of the DNA topoisomerase I relaxing activity indicates that 1a and 1b inhibit topoisomerase I which regulate topological states of DNA strands during the cell process. The apoptotic potential of the compounds at the single cell level indicates that they may inhibit cell proliferation by inducing apoptosis. Immunohistochemistry staining analysis displays that these compounds significantly decreases the level of Bcl-2 in HeLa and HT29 cells while increasing the accumulation of P53. Overall, the potent antiproliferative action, low cytotoxic effect, good solubility in a physiological medium and micro molar range dosage of these compounds reveals that they are likely to be the good drug candidates for pharmacological trials. © 2017 Elsevier B.V. All rights reserved.

1. Introduction Chlorocyclophosphazenes, (NPCl2)n (n:3–40), representing a significant class of the inorganic ring systems have continued to attract considerable interest for a variety of applications [1]. Hexachlorocyclotriphosphazene, (NPCl2)3, which is an important starting compound for the preparation of numerous cyclotriphosphazene derivatives, is one of the most important compounds of these groups [2]. Cyclotriphosphazene-based compounds exhibit some interesting physical and chemical properties as materials of ionic liquids [3–6], lubricants [7,8], chemosensors [9], liquid crystals [10–12], Li-ion batteries [13], OLEDs [14,15], flame retardants [16–21] and biomaterials [22,23]. The development of bioactive cyclotriphosphazenes is currently also considered a promising approach in the search for new effective drug candidates for the therapy ⁎ Corresponding author. E-mail address: [email protected] (A. Karadağ).

http://dx.doi.org/10.1016/j.molliq.2017.01.067 0167-7322/© 2017 Elsevier B.V. All rights reserved.

of various diseases, especially cancer, microbial, and fungal [24–29]. Recent works revealed that the cyclotriphosphazenes display significant anticancer and cytotoxic activities towards cancer cells related to prevalent cancer diseases such as HT-29 (human colon adenocarcinoma), Hep2 (Human epidermoid larynx carcinoma), HeLa (human cervical adenocarcinoma), LNCaP and PC-3 (human prostatic carcinoma), MCF7 (human breast cancer), HL-60 (human promyelocytic leukemia), A549 (human lung adenocarcinoma) and HCV29T (urinary bladder cancer) cancer cells [30–34]. For example, in a study, aziridine-crown substituted cyclotriphosphazenes creating fractures in the DNA chain caused inhibition on the growth of cancer cells [35]. In another study, the Cu+2 complex of fully phenoxysubstituted cyclotetraphosphazene showed oxidative cleavage activity on the DNA chain [36]. Also, studies on the pharmacological activity of protic ionic liquids (PILs) of cyclotriphosphazenes are very important. The chemistry of PILs of cyclotriphosphazenes became largely alluring because of their wide range of pharmacological applications that provides a diverse variety of PILs with different features [37]. However,

H. Akbaş et al. / Journal of Molecular Liquids 230 (2017) 482–495

483

the pharmacological properties of the PILs should be further developed on the basis of Lipinski's rule-of-five enough to enter clinical practice. We therefore exerted the rule of 5 design strategy for the new PILs of hexapyrrolidinocyclotriphosphazenes, PYR (1), and so very effective newly synthesized molecules were obtained to increase the possibility of their use as chemotherapeutic agents in the treatment of various diseases. The results we acquired from our previous studies have developed our interest in this promising cyclotriphosphazenes, and this paper showed a new class of cyclotriphosphazenium salts, denoted as 1a, 1b, and 1c, which were tested for their pharmacological properties on human cervical cancer (HeLa), human colon cancer (HT29), rat glioma (C6) and African green monkey kidney (Vero) cell lines. The DNA/BSA binding abilities of 1a, 1b, and 1c were examined by UV–vis spectrophotometry, fluorescence spectroscopy techniques, and electrophoresis measurements. On the other hand, the PILs, one of the oldest groups of ionic liquids started with the ethanolammonium nitrate, reported in 1888 by Gabriel and followed with the ethylammonium nitrate, reported in 1914 by Walden [38]. These types of amine-based ionic liquids were synthesized in 2005 by Bicak [39] and in 2013 by Karadag et al. [40]. However, the cyclotriphosphazene-based PILs with bulky organic acids were obtained by our group for the first time [37]. This paper also reports the preparation of the PILs (1a, 1b, and 1c) of hexapyrrolidinocyclotriphosphazene PYR (1) (Scheme I) and the characterizations of these compounds using elemental analyses, FTIR, 1H, 13C {1H} and 31P {1H} NMR data.

obtained on the DSC instrument (Perkin Elmer JADE model). The FTIR spectra of PILs were recorded on a Jasco FT/IR-430 spectrometer in KBr discs and reported in cm−1 units. 1H,13C{1H} and 31P{1H} NMR spectra were recorded on a Bruker DPX FT-NMR (600 MHz) spectrometer (SiMe4 as an internal standard for 1H and 85% H3PO4 as an external standard for 31P NMR), operating at 600, 150 and 243 MHz. The spectrometer was equipped with a 5 mm PABBO BB inverse-gradient probe and standard Bruker pulse program was used [42]. The TG and DTG curves were obtained using a PYRIS Diamond TG/DTA apparatus in a dynamic nitrogen atmosphere (heating rate: 10 °C/min, ceramic crucibles, mass ~ 10 mg and temperature range 35–1150 °C).

2. Experimental

2.3.1. Data for compound 1a Yield: 0.56 g (79%). M.p.: 216 °C. Anal. Calc. for C31H54N9O4P3: C, 52.46; H, 7.67; N, 17.76 Found: C, 52.75; H, 7.48; N, 17.68. FTIR (KBr, cm−1): 3051 (C\\H arom.asym.), 3028 (C\\H arom.sym.), 1228, 1205 (P_N), 2629 (NH+), 1588, 1485 (COO−). NMR δH (600 MHz, CDCl3, ppm, numberings of protons are given in Scheme I): 1.77 [(m, 24H, NCH2CH2(pyrr)], 3.06 (m, 24H, NCH2), 7.12 (d,1H, H3,4JHH: 3.1 Hz), 6.61 (dd,1H, H5,3JHH: 8.6 Hz, 4JHH: 3.1 Hz), 6.46 (d,1H, H6,3JHH: 8.6 Hz), 2.51 (m, 1H, NH, 2JPNH: 6.7 Hz), 8.52 (broad, OH, 2H) ; NMR δC (150 MHz, CDCl3, ppm, numberings of carbons are given in Scheme I): 26.87 [NCH2CH2(pyrr)], 46.82 [NCH2(pyrr)], 172.08 (C1), 119.90 (C5),116.61 (C6), 148.49 (C4), 120.00 (C2),116.46 (C3), 155.82 (C7)

2.1. Materials The free based, PYR (1), was obtained according to reported procedures [41]. All other chemicals obtained from commercial sources were used without further purification. 2.2. Measurements The melting points (Tm) of PYR (1) and PILs (1a–1c) were determined on a Barnstead Electrothermal 9100. The melting points were

2.3. General procedure for the synthesis of compounds 1a–1c A solution of PILs (1a–1c) containing 2,5-dihdyroxy benzoate, decanoate or dihydrogen borate was prepared by the dropwise addition of gentisic, decanoic or boric acid (1.0 mmol), respectively, in dry tetrahydrofuran (10 mL) to a solution of hexapyrrolidinocyclotriphosphazene PYR (1) (1.0 mmol) dispersion in dry tetrahydrofuran (30 mL). The mixture was refluxed for 15 h. Afterwards, the solvent was evaporated under reduced pressure and the oily crude product was crystallized from n-hexane-THF.

Scheme I. The syntheses of the PILs with PYR (1) and gentisic, decanoic and boric acids.

484


2.3.2. Data for compound 1b Yield: 0.53 g (73%). M.p.: 199 °C. Anal. Calc. for C34H68N9O2P3: C, 56.10; H, 9.42; N, 17.32. Found: C, 56.48; H, 9.33; N, 17.54. FTIR (KBr, cm− 1): 2953, 2848 (C\\H aliph.), 1244, 1203 (P_N), 2608 (NH+), 1596, 1406 (COO−). NMR δH (600 MHz, CDCl3, ppm, numberings of protons are given in Scheme I): 1.71 [(m, 24H, NCH2CH2(pyrr))], 3.00 (m, 24H, NCH2(pyrr)), 2.19 (m, 2H, H2), 1.50 (m, 2H, H8), 1.26 (m, 14H, H3–H9), 0.87 (m,3H, H10), 2.52 (broad, 1H, NH); NMR δC (150 MHz, CDCl3, ppm, numberings of carbons are given in Scheme I): 26.40 [NCH2CH2(pyrr)], 46.29 [NCH2(pyrr)], 174.98 (C1) 34.13 (C2), 31.74 (C3), 29.35 (C4), 29.22 (C5), 29.13 (C6), 29.02 (C7), 24.97 (C8), 22.57 (C9), 14.41 (C10). 2.3.3. Data for compound 1c Yield: 0.47 g (76%). M.p.: 220 °C. Anal. Calc. for C24H51BN9O3P3: C, 46.68; H, 8.33; N, 20.42. Found: C, 47.30; H, 8.63; N, 20.14. FTIR (KBr, cm− 1): 2960, 2865 (C\\H aliph.), 1233, 1183 (P_N), 2629 (NH+), 1457, 1009 (B\\O). NMR δH (600 MHz, CDCl3, ppm): 1.76 [(m, 24H, NCH2CH2(pyrr))], 3.03 (m, 24H, NCH2(pyrr)), 3.30 (s, 2H, O\\H), 2.48 (broad, 1H, NH); NMR δC (150 MHz, CDCl3, ppm): 26.42 [NCH2CH2(pyrr)], 46.42 [NCH2(pyrr)]. 2.4. Antiproliferative effects The antiproliferative effects of 1a–1c were determined by the cell culture, cell proliferation assay (CPA), calculation of IC50 and % inhibition, cytotoxic activity assay, terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay, DNA topoisomerase I inhibition assay, cell imaging, Immunohistochemistry, DNA/BSA binding gel electrophoresis and statistical analysis studies, and the explanations were presented below. 2.4.1. Cell culture C6, HT29, HeLa, and Vero cell lines were maintained in Dulbecco's modified eagle's medium (DMEM, Sigma) supplemented with 10% (v/ v) fetal bovine serum (Sigma, Germany) and PenStrep solution (10,000 U/10 mg) (Sigma, Germany) (ATTC, American Type Culture Collection). At confluence, cells were detached from the flasks using 4 mL of trypsin-EDTA (Sigma, Germany) and centrifuged, and the cell pellet was resuspended in 4 mL of supplemented DMEM. 2.4.2. Cell proliferation assay (CPA) A cell suspension containing 5 × 103 cells in 100 μL was pipetted into the wells of 96-well cell culture plates (COSTAR, Corning, USA). The test compounds (1a, 1b, and 1c) and a positive control compound (cisplatin) were dissolved in sterile DMSO. The amount of DMSO was adjusted to 0.5% maximum. The cells were treated with test compounds and cis-platin at final concentrations of 25, 50, 100, 150, 200, 250, 375, and 500 μg/mL. Cell controls and solvent controls were treated with supplemented DMEM and sterile DMSO, respectively. The final volume of the wells was adjusted to 200 μL with supplemented DMEM. The cells were then incubated at 37 °C with 5% CO2 overnight. The antiproliferative activity of the compounds was determined using a BrdU Cell proliferation ELISA kit according to the manufacturer's protocol (Roche, USA) for a calorimetric immunoassay based on BrdU incorporation into the cellular DNA. Briefly, cells were exposed to BrdU labeling reagent for 4 h, followed by fixation in FixDenat solution for 30 min at room temperature. Then cells were cultured with a 1:100 dilution of anti-BrdUPOD for 1 h and 30 min at room temperature. Substrate solution was added to each well, and BrdU incorporation was measured at 450– 650 nm using a microplate reader (Rayto, China). Each experiment was repeated at least three times for each cell line. 2.4.3. Calculation of IC50 and % inhibition IC50 represents the concentration of an agent that is required for 50% inhibition in vitro. The half maximal inhibitory concentration

(IC50) of the test and control compounds was calculated using XLfit5 software (IDBS) and expressed in μg/mL at 95% confidence intervals. The CPA assay results were reported as the percent inhibition of the test and control substances. The percent inhibition was calculated according to the following formula: % inhibition = [1 − (Absorbance of Treatments/Absorbance of DMSO) × 100]. 2.4.4. Cytotoxic activity assay The cytotoxicity of 1a, 1b, and 1c on C6, HT29, HeLa, and Vero cells was determined using a Lactate Dehydrogenase (LDH) Cytotoxicity Detection Kit (Roche, USA) based on the measurement of LDH activity released from the cytosol of damaged cells into the supernatant according to the manufacturer's instructions; 5 × 103 cells in 100 μL were seeded into 96-well microtiter plates as triplicates and treated with varying concentrations (25, 50, 100, 150, 200, 250, 375, and 500 μg/mL.) of 1a, 1b, and 1c as described above at 37 °C with 5% CO2 overnight. LDH activity was determined by measuring absorbance at 492–630 nm using a microplate reader. 2.4.5. Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay In vitro detection of apoptosis was assessed using a TUNEL assay kit (Roche, Germany) according to the manufacturer's protocol. HT29 cell lines (30.000 cells/well) were placed in a poly-L-lysine covered chamber slide. The cells were treated with IC50 concentrations of 1a, 1b, and 1c and left for 24 h of incubation. There were two controls for this assay: one was a positive control that had DNase-1 treatment and the other was a negative control that had no terminal deoxynucleotidyl transferase (TdT). When the incubation time was over, the chamber was removed from the slide and washed with DPBS to remove the medium and unattached cells. All of the incubation and washing steps were done in a plastic jar. The slides were gently washed with DPBS, and for fixation 4% paraformaldehyde in DPBS at pH 7.4 was freshly prepared and added to the slides for 60 min at room temperature. Following incubation, the slides were washed twice with DPBS. The cells were blocked with freshly prepared 3% H2O2 in methanol for 10 min at room temperature. Following incubation, the slides were washed twice with DPBS. The cells were permeabilized by prechilled 0.1% Triton X-100 and freshly prepared 0.1% sodium citrate in water and then incubated for 2 min on ice. All the slides were washed with DPBS twice for 5 min each. At this point, in order to prepare a DNase I enzyme-treated positive control, 100 μL of DNase-I buffer was added to the slide, which was incubated at room temperature for 10 min. Fixative cells were transferred into a TUNEL reaction mixture (50 μL/section) containing TdT and fluoresceindUTP. Intracellular DNA fragments were then labeled by exposing the cells to TUNEL reaction mixture for 1 h at 37 °C in a humidified atmosphere and protected from light. After washing with DPBS twice, cells positive for apoptosis showed a green fluorescent signal and were visualized by a Leica fluorescent microscope (Leica DMIL LED fluo, Germany). 2.4.6. DNA topoisomerase I inhibition assay The DNA topoisomerase I inhibitory activities of 1a, 1b, and 1c were evaluated using a cell-free topoisomerase I assay kit (TopoGen, USA). The principle of the assay is to measure the conversion of supercoiled pHOT1 plasmid DNA to its relaxed form in the presence of DNA topoisomerase I alone and with test compounds. The supercoiled substrate (pHOT1 plasmid DNA) and its relaxed product can easily be distinguished in agarose gel, because the relaxed isomers migrate more slowly than the supercoiled isomer. In brief, 20 μL of reaction mixture containing 1 μL of plasmid pHOT1 DNA in relaxation buffer was incubated with 2 U recombinant human topoisomerase I enzyme in the presence of IC50 concentrations of 1a, 1b, and 1c, or camptothecin as positive control. The reactions were carried out at 37 °C for 30 min and then terminated by the addition of stop solution. After the termination, the sample was analyzed using 1% agarose gel at 4 V/cm for 60 min.


485

Table 1 31 1 P{ H} NMR parameters of PYR (1) and 1a–1c. Compound

δP

PYR (1) 1a 1b 1c

18.60 12.39 16.43 17.72

After electrophoresis, DNA bands were stained with ethidium bromide (EtdBr) (1 mg/mL) solution and photographed through a gel imaging system (UVP BioSpectrum, Germany). 2.4.7. Cell imaging Cells were seeded in 96-well plates at a density of 5000 cells per well and allowed 24 h for attachment. Using previously established IC50 doses of 1a, 1b, and 1c treatments was performed for 24 h, during which morphology changes were assessed by phase contrast microscopy. Images of vehicle (DMSO), 1a, 1b, and 1c treated cells were taken at the end of experimental period using a digital camera attached inverted microscope (Leica IL10, Germany). 2.4.8. Immunohistochemistry Immunohistochemistry (IHC) techniques used for to localize antigens changing expression level following 1a, 1b, and 1c treatments. Accordingly, HT29 and HeLa cell lines (15.000 cells/well) were placed in a poly-L-lysine covered chamber slide. The cells were treated with IC50 concentration of test compounds and left for 24 h of incubation. There was a negative control that had no test compounds. When the incubation time was over, the chamber was removed from the slide and washed with DPBS to remove the medium and unattached cells. All of the incubation and washing steps were done in a plastic jar. The slides were gently washed with DPBS, and for fixation 4% paraformaldehyde in DPBS at pH 7.4 was freshly prepared and added to the slides for 60 min at room temperature. Following incubation, the slides were washed twice with DPBS. Heat-induced epitope retrieval (HIER) was performed using Cell Conditioning 1 (CC1), and visualization was achieved with the Universal DAB Detection Kit, according to manufacturer's instructions. IHC was performed using Bcl-2 (mouse monoclonal, clone 124; Ventana), CK7 (mouse monoclonal, clone OVTL 12/30; Ventana), CK20 (mouse monoclonal, clone Ks20.8; Ventana), and P53 (mouse monoclonal, clone D07; Ventana) on the VENTANA Bench-Mark XT System. Briefly, sections were newly pretreated with CC1 Ventana reagent for 30 min at 95 °C. After pretreatment, sections were incubated with above mentioned primary antibody for 32 min at 37 °C. Reactions were revealed with an ultraView Universal DAB Detection Kit (Ventana, USA). The slides were counterstained with Hematoxylin II (Ventana) for 4 min and Bluing Reagent (Ventana) for 4 min and coverslips were applied by an automated coverslipper (Leica CV5030). For the HeLa and the HT29 cell lines, the number of positive and negative cells was counted in five zones. This procedure was repeated 3 × for each protein stained slide. The slides were scored staining intensity score rated as follows: no staining (0, no stained cells or b 5% positive cells), weak staining (1 +, 5–24% positive cells), moderate staining (2 +, 25–49% positive cells), and strong staining (3 +, N50% positive cells). A score of 2+ or 3+ was considered positive for relevant expression while a score of 0 or 1+ was considered negative. Fig. 1. The molecular structures drawn using ChemBioOffice of PILs (1a, 1b and 1c).

2.4.9. DNA/BSA binding and gel electrophoresis studies UV spectrophotometer was used to find the interaction of the compounds with CT-DNA and to calculate the binding constants (????). A CT-DNA solution was prepared by dissolving 2.5 mg CT-DNA in 10.0 mL Tris–HCl buffer (20 mM Tris–HCl, 20 mM NaCl at pH 7.0) and stored in the refrigerator. The concentration of CT-DNA was determined spectrophotometrically using the known Ɛ value of 6600 M−1 cm−1 at 260 nm. After dissolving the CT-DNA fibers in Tris–HCl buffer, the purity

of this solution was checked from the absorbance ratio A260/A280. The CT-DNA solution in the buffer displayed an A260/A280 ratio of 1.89, indicating that the DNA was sufficiently pure. The 1a, 1b and 1c were dissolved in DMSO and diluted with Tris–HCl buffer to obtain 25 μM concentrations. Test compounds in the solutions were incubated at 25 °C for about 30 min before measurements. The UV absorption

486


Fig. 2. TGA curves of PYR (1) and 1a–1c.

titrations were conducted by keeping the concentration of these compounds fixed while varying the CT-DNA concentrations (0–100 μM). Absorption spectra were recorded by using 1-cm-path quartz cuvettes at room temperature. To evaluate the interaction of the compounds with BSA was used UV–vis spectrophotometry and fluorescence spectroscopy. A BSA solution was prepared by dissolving 2.5 mg BSA in 10.0 mL in Tris–HCl buffer (5 mM Tris–HCl, 10 mM NaCl at pH 7.4) and stored in the refrigerator. The UV spectra of the BSA solutions (0–100 μM) in the presence of a fixed concentration of the PILs (25 μM) were scanned against the Tris– HCl buffer in the wavelength range from 250 to 320 nm. The fluorescence quenching of the BSA solutions (13 μM) in the absence and presence of the PILs (25–75 μM) was observed ~350 nm (λex = 280 nm). Ethidium bromide (EB) displacement experiments were performed by tracking alters in the fluorescence intensities of the EB-DNA solutions in the presence of increasing amounts of the test compounds. The fluorescence spectra of EB were measured using an excitation wavelength of 295 nm and the emission range was set between 200 and 600 nm. The spectra were analyzed according to the Stern–Volmer equation, I0/ I = 1 + KSV [Q], where I0 is the fluorescence intensity in the absence of quencher, I is the fluorescence intensity in the presence of quencher, KSV is the Stern–Volmer quenching constant, [Q] is the quencher concentration. KSV can be calculated from the slope of the plot of I0/I vs. [DNA]. The restriction enzyme inhibition assay was conducted to evaluate both specific or nonspecific binding and enzyme inhibition by PILs. Supercoiled pTOLT (10 μM) plasmid DNA was incubated with the PILs 1a, 1b or 1c (25 μM) and restriction enzymes KpnI and BamHI (2 units) at 37 °C in 50 mM Tris–HCl/18 mM NaCl buffer (pH 7.2) for 4 h. The digestion products were resolved by using 1.5% (wt/vol) agarose gels with ethidium bromide.

2.4.10. Statistical analysis The statistical significance of differences was determined by oneway analysis of variance (one-way ANOVA) tests. Post hoc analyses of group differences were performed using the Tukey test, and the levels of probability were noted. SPSS for Windows was used for the statistical analyses. The results are reported as the mean values ± SEM of three independent assays, and differences between groups were considered to be significant at P b 0.05. 3. Results and discussion 3.1. Syntheses and characterizations The PILs (1a–1c) were obtained from the reactions of PYR (1) with gentisic, decanoic and boric acid in tetrahydrofuran (Scheme I). The reaction yields were found to be in the range of 73–79%. 1a, 1b, and 1c are at solid, creamy and solid state, respectively at room temperature and the melting points (Tm) of those are 216, 199, and 220 °C, respectively, as it is in the case of PILs such as butylammonium BF4; Tm:198.2°, ammonium H2PO4; Tm: 193.3°and 1-ethyl-2-methylimidazolium chloride; Tm: 178 °C in the literature [43]. On the other hand, the onset temperatures of PILs (1a, 1b, and 1c) are 223, 167 and 286 °C, respectively. It is observed that the change in the states of PILs depending on the temperature rise is irregular (Fig. S1; S designates Supplementary data). It is possible to say that PILs might be used in some applications as solvents. All the phosphazenium PILs (1a–1c) were soluble in the common organic polar solvents, e.g., dichloromethane, chloroform, dimethylformamide and DMSO. On the other hand, they were also considerably soluble in water. Data obtained from the microanalyses, FTIR, 1H, 31P {1H} and 13C{1H} –NMR data were consistent with the proposed structures of the PILs. The 1H NMR, FTIR and elemental analysis

Table 2 Thermoanalytical data of PYR (1) and 1a–1c. Compound

Onset temperature; °C

Stage

Temperature range; °C

DTG(max)/°C

Mass loss (Obs.)/Δm %

C24H48N9P3 PYR (1) 555.62 g mol−1

300

C31H54N9O4P3 (1a) 709.74 g mol−1

223

1 2 1 2

C34H68N9O2P3 (1b) 727.88 g mol−1

167

C24H51BN9O3P3 (1c) 727.88 g mol−1

286

174–416 416–1150 190–387 387–587 642–1150 93–230 230–392 392–1150 99–396 396–1150

337 908 293 438 907 198 310 715 321 937

90.36 7.52 65.13 7.58 24.75 25.20 67.79 6.58 80.27 16.70

1 2 3 1 2

Total mass loss (Obs.)/Δm % 97.88

97.46

99.57 96.97


487

Fig. 3. Effects of 1a, 1b, 1c and cis-platin on the proliferation of HeLa, HT29, C6, and Vero cells. Percent inhibition was reported as mean values ± SEM of three independent assays (P b 0.05).

results indicate that all the compounds have mono protanated with the nitrogen of phosphazene ring. The 31P {1H} NMR data were listed in Table 1, and the spectra of the salts were illustrated in Figs. S2–S4. As expected, the δP shifts of the phosphazenium salts are smaller than that of free base PYR (1) [44]. The expected spin systems of the PILs would have been the AX2, but they are interpreted as a singlet for all the PILs. In the CDCI3 solution, the hydrogen ion may exchange between the nitrogen atoms of the phosphazene ring [37]. The chemical shifts, multiplicities, and the coupling constants are very useful for the interpretations of the 1H and 13C{1H} NMR signals of all the PILs (Fig. S5-S10). The pyrrolidino NCH2 protons for the compounds 1a, 1b and 1c were observed at 3.06, 2.99 and 3.03 ppm, respectively. The pyrrolidino NCH2CH2 protons for the compounds 1a, 1b and 1c were observed at 1.77, 1.71 and 1.76 ppm, respectively. In the 13C{1H} NMR spectra of the PILs, the carbonyl carbon atoms (C_O) for 1a and 1b were observed at 172.08 and 174.98 ppm, respectively. The pyrrolidino NCH2 and NCH2CH2 carbons for 1a, 1b and 1c were observed at 46.82, 46.29 and 46.42 ppm, and 26.87, 26.40 and 26.42 ppm respectively. While the aliphatic carbons for 1b were observed between 14.41 and 34.13 ppm, those for 1a were appeared between 116.46 and 155.82 ppm. The PILs (1a–1c) exhibit intense bands between 1243 and 1225 cm− 1 and 1203–1183 cm−1, related to the νP_N bonds of the phosphazene rings [45]. The characteristic absorption bands in the range of 2633–2608 cm−1 are attributed to the ν+ (NH) stretching vibrations for the PILs (Fig. S11). The PILs (1b, 1c) show two strong absorption bands between 1596 and 1586 and 1481–1405 cm−1, which are assigned to the asymmetric and symmetric stretching vibrations of the νCOO\\ bonds, respectively, indicating clearly the PIL formation. The IR and NMR data showed that the molecular structures of the PILs (1a– 1c) were in agreement with the structures depicted in Scheme I. On

the other hand, the cation-anion interaction in the PILs (1a–1c) was shown using ChemBioOffice 2008 Ultra version 11 program (Fig. 1). 3.2. Thermal studies The thermogravimetric analyses (TGA) of N3P3(NC4H8)6, PYR (1), and PILs (1a–1c) were performed in the temperature range from 35 °C up to 1150 °C under a nitrogen atmosphere at the heating rate of 10 °C/min. TG and DTG curves of PYR (1) and 1a–1c were depicted in Fig. 2 and Fig. S12, while significant thermoanalytical data of these compounds were summarized in Table 2. It is envisaged that the thermal decomposition mechanism of 1a–1c exhibits primarily the decomposition of 2,5-dihydroxy benzoate, decanoate and dihydrogen borate anions, respectively. Afterwards, the hexapyrrolidinocyclo triphosphazenium cation starts to decompose. The thermal decomposition of 1b is completely finished at 1150 °C, but the decompositions of 1, 1a and 1c are not completed at this temperature, and the analysis crucible is available a residue of about 2.56% (Table 2). As seen from Fig. 2 and Fig. S12, the thermal decomposition steps of 1a and 1b are slightly more complicated than those of PYR (1) and 1c. The 1a and 1b are mainly subjected to degradation in three steps and is also in Table 3 IC50 values and tumor specificity rate for 1a, 1b, 1c, and Cis-platin. Compounds

1a 1b 1c Cis-platin

IC50 (μM)

Tumor specificity

HeLa

HT29

C6

Vero

HeLa

HT29

C6

238.60 232.52 261.28 586.88

297.22 298.28 233.51 292.62

308.72 317.41 298.66 389.82

326.65 293.37 327.76 685.08

1.37 1.26 1.25 1.12

1.10 0.98 1.40 2.25

1.06 0.92 1.10 1.76

488


Fig. 4. Effects of PYR (1) and cis-platin on the proliferation of HeLa, HT29, C6, and Vero cells. Percent inhibition was reported as mean values ± SEM of three independent assays (P b 0.05).

two steps the decomposition of 1c. Compared to the onset temperature of PYR (1) and 1a–1c (Table 2), it appeared that the thermal stability of PYR (1) is greater the average of 66.38 °C than those of 1a–1c. The lower thermal stabilities of 1a–1c may be caused by weak electrostatic interaction because of steric hindrance between the voluminous cation of PYR (1) and anions (especially 1a and 1b). The thermal stabilities of 1a, 1b, and 1c according to the onset temperature tend to increase in the order of 1b, 1a, and 1c. It is estimated that the changes in thermal stabilities may be attributed to greater than that of decanoate anion of the interactions by hexapyrrolidinocyclotriphosphazenium cation of the 2,5-dihydroxy benzoate and dihydrogen borate anions (Table 2). 3.3. The antiproliferative property of the PILs The in vitro antiproliferative effects of three PILs using CPA against three cancer cell lines (HT29, HeLa, and C6) and one noncancerous cell line (Vero) were performed. The commercially used anticancer drug cis-platin was used as a reference compound. Data obtained from the proliferation assay, % inhibition information of the compounds was shown in Fig. 3. IC50 values and tumor specificity rate to be used in the following studies were determined by implementing ELISA BrdU assay and available at the Table 3. To achieve tumor specificity, we divided the sum of the IC50 values from normal cells (Vero) to the sum of the IC50 values of the cancerous cells (HT29, HeLa, and C6) and shown in Table 3. Compound 1a and 1c showed good selectivity for the cancer cell line over the non-cancerous cell line and the overall specificities of these compounds are similar to cis-platin. As shown in Fig. 3, each PIL significantly inhibited proliferation of HeLa, C6, and HT-29 (P b 0.05) cells as equivalent to the effect of the control compound cis-platin. A gradient from the % inhibition values of proliferation assay can be seen with PILs being more active as follows: PYR

(1) N 1c N 1a N 1b (Figs. 3 and 4). The results of the cell proliferation assay displayed that the PILs were nearly the same antiproliferative effect with cis-platin (Fig. 3) against the cell lines, indicating their anticancer potential, as in our previous studies [37]. 3.4. The cytotoxic activity of the PILs The cytotoxic activity of the PILs on HT29, HeLa, and C6 cancer cell lines and non-cancer Vero cells were performed using an LDH cytotoxicity assay kit by using varying concentrations of the test compounds (25, 50, 100, 150, 200, 250, 375, and 500 μg/mL). The results of the LDH assay are expressed in terms of % cytotoxicity values and presented in Fig. 5. We would determine by using this data, whether the effect of the BV was cytotoxic or cytostatic against the cells. The findings of % cytotoxicity delineate that treatment of cells with 200 μg/mL or higher concentrations of test compounds resulted in significant rupture of cell membrane integrity, but its 150 μg/mL or lower concentrations did not same effect on the cell membrane (Fig. 5). Thus, our compounds at 150 μg/mL or lower concentrations exhibited cytostatic properties, while they at 200 μg/mL or higher concentrations displayed cytotoxic to cells. Especially, HT29, C6 and Vero cells showed no cytotoxic situations in response to test compound treatment up to 200 μg/mL, which confirms that our compounds have no cytotoxicity in cells, but it exhibited potent antiproliferative activity (Fig. 5). According to another finding, cytotoxicity of each compound was closed to cytotoxicity of cisplatin, at their IC50 concentrations against any of the four cell lines (data not shown). According to another finding, the PYR (1) has relatively high cytotoxicity values indiscriminately against cell tested (Fig. 6). This high (about 2-fold) of cytotoxicity values was more than that observed for 1a with 30% to 40% cytotoxicity values, for 1b with 5% to 15%


Fig. 5. The cytotoxic activity of 1a, 1b, and 1c on HeLa, HT29, C6, and Vero cells. Percent cytotoxicity was reported as mean values ± SDs of three independent assays.

Fig. 6. The cytotoxic activity of PYR (1) on HeLa, HT29, C6, and Vero cells. Percent cytotoxicity was reported as mean values ± SDs of three independent assays.

489

490


Fig. 7. Fluorescence and phase-contrast images of the HT29 cancer cell line examined by TUNEL assay. TUNEL positive cell nuclei in brilliant green were observed under a fluorescence (1a, 1b, and 1c, NC, and PC) and phase-contrast microscope (1a′, 1b′, and 1c′, NC′, and PC′), (NC: Negative control, PC: Positive control). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)


491

cytotoxicity values, and for 1c with 15% to 20% cytotoxicity values at 100 μg/mL concentrations (Figs.s 5 and 6). The antiproliferative activity level of PYR (1) was more than those of PILs on the cell lines tested (Fig. 4). However, the cytotoxicity of the PILs was significantly lower than that of PYR (1) alone (P b 0.05) (Fig. 5). This means is that a safe concentration level for 1a, 1b, and 1c but not PYR (1), exists for their antiproliferative effects without significant cytotoxicity to cells. That is, to say that these PILs are better candidates for pharmacological researches.

3.5. The apoptotic potential of the PILs at the single cell level

Fig. 8. Inhibition of recombinant human topoisomerase I relaxation activity. A DNA unwinding assay was performed with 250 ng pHOT-1 supercoiled DNA, 2U TOP1 and IC50 concentrations of 1a, 1b, and 1c. The forms of DNA are denoted as I (Nicked DNA), II (Relaxed DNA), and III (Supercoiled DNA). Lane 1 represents the negative control (Supercoiled DNA + TOP1); lane 2 is the positive control (Supercoiled DNA + TOP1 + Camptothecin), and Lanes 3–5 represent test compounds over an IC50-concentration titration.

Further studies to understand action mechanism of antiproliferative properties of 1a, 1b, and 1c, a TUNEL assay were used to investigate whether test compounds induced inhibition of cell proliferation was associated with cell apoptosis. The TUNEL reaction would prefer to marks DNA strand breaks generated during apoptosis, in comparison to necrosis. This selection allows discrimination of apoptosis from necrosis and from primary DNA strand breaks induced by cytostatic agent. Also, TUNEL assay was able to make soundly the distinction between apoptotic cells and apoptotic bodies with surrounding cells. As illustrated in Fig. 7, test compound 1a, 1b, and 1c treated cells displayed a higher percentage of TUNEL-positive apoptotic cell nuclei (P b 0.05), indicating the nicked DNA, whereas the DPBS control was TUNEL-negative cell nuclei.

Fig. 9. The effect of 1a on the morphology of HeLa, HT29, C6, and Vero cells. Exponentially growing cells were incubated with IC50 concentrations of 1a at 37 °C overnight and visualized by digital camera attached inverted microscope (Leica IL10, Germany). DMSO treated cells as controls. All scales are 100 μm.

492


Fig. 10. Representative images of the cells examined by immunohistochemical staining for functional protein group (Bcl-2 and P53), and for marker protein group (CK7 and CK20). The specific signals are shown as brown staining. Bar is 100 μm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

However, 1c treated cells were found to provide a more intense light emission than 1a or 1b treated cells. For each concentration (25, 50, 100, 150, 200, 250, 375, and 500 μg/mL), the apoptotic index was determined by counting the percentage of TUNEL-positive cells from at least 100 nuclei. The apoptotic index was about N 25% at IC50 concentration. A significant alteration in the incidence of TUNEL-positive cell nuclei was determined ≥100 μg/mL concentrations compared with ≤75 μg/mL or lower concentrations. However, at ≥ 200 μg/mL concentrations were not found statistically significant (P N 0.05). 3.6. Analysis of inhibition of DNA topoisomerase I To investigate possible mechanisms of action of the test compounds was used DNA topoisomerase I assay in vitro. DNA topoisomerase I is a nuclear enzyme that regulate cellular mechanisms and that an important target of anticancer agents. As shown in Fig. 8, Lanes 4–5 reflect the results show that 1a and 1b compounds inhibited the activity of recombinant human DNA topoisomerase I as a positive control, camptothecin. This data therefore may indicate that 1a and 1b inhibit cell proliferation using the suppression of DNA topoisomerase I action during replication. 3.7. The effect of the PILs on the morphology of the cells The morphology of cells was investigated using phase-contrast microscopy. Morphological changes of 1a, 1b, and 1c were visualized in cells exposed to test compounds for 24 h. As shown in Fig. 9, Figs. S13 and S14, obvious morphological changes were observed in the treated cells in a concentration-dependent manner compared to the control cells. The morphology of compound PYR (1) treated cells at 100 μg/mL or higher concentration changed from a cell rounding to floating cells, an indication of cell death. All compounds were not able to inhibit growth of the cells in culture at 75 μg/mL and lower dosages. According to our observations, at 100 μg/mL and higher doses, the numbers of cells seem to lesser and separate from one another and they look small. Treatment with test compounds also depicted degradations on the cells, as the cell growth affected and cellular morphology was similar to that of apoptotic situations at high and medium concentrations. In addition, the compounds clearly showed characteristic apoptotic

changes in morphology like cell shrinkage and rounding, and other changes such as surface membrane changes, and detachment from the culture plate (Figs. S13 and S14). The cells lost their fibroblast-like appearance and clumped together because the compounds also inhibited the elongation and growth of the cells in culture. 3.8. IHC evaluation of slides treated by PILs Immunohistochemistry staining of the treated slides showed that these compounds decreased expression of Bcl-2 in treating cells while they caused increased the accumulation of P53 (Fig. 10 and Fig. S15) which emphasizes the apoptotic effects of these. The results also showed that these compound let noticeably reduced the expression of CK20 (a marker protein for HT29 cells) and CK7 (a marker protein for HeLa cells) in the cells. Cytokeratins releasing from proliferating or apoptotic cells may be implied change of cell characteristics which occurred due to decreasing mRNA levels. With decreasing expression of CK7 and CK20 can be associated with the reduced metastatic ability influencing intermediate filament (IF) proteins (Fig. 10). 3.9. DNA/BSA binding and gel electrophoresis studies PIL–DNA interactions can be observed by comparison of UV–visible absorption spectra of the free PIL and PIL–DNA adducts [46]. The binding constant (K) of 1a–1c with DNA can be determined according to Benesi–Hildebrand equation, A0/A − A0 = ƐG/ƐH–G − ƐG + ƐG/ƐH–G − ƐGx1/K[DNA], where K is the binding constant, A0 and A are the absorbances of 1a–1c and its adduct with DNA, respectively, and ƐG and ƐH– G are the absorption coefficients of the PILs and the PIL–DNA adduct, respectively [46]. The binding constant can be obtained from the intercept-to-slope ratios of A0/(A − A0) vs. 1/[DNA] plots. Fig. 11A represents the interaction of 1a–1c with CT-DNA. According to Benesi–Hildebrand equation, the plot of A0/(A − A0) vs. 1/[DNA] data was yielded the binding constant (K) which was 3.0 × 103 M−1 for 1a, 6.8 × 103 M− 1 for 1b, and 5.8 × 104 M− 1 for 1c (Fig. 11A). With the increase in CT-DNA concentration resulted in hyperchromic effect indicate a strong interaction between the 1a–1c and DNA. This hyperchromic effect on the spectra of the PIL-DNA adduct may be indicative of groove binding. Also, the absorption titration spectrum of


493

Fig. 11. (A) UV–visible absorption spectra of 25 μM 1a, 1b, and 1c in the absence (a) and presence of 25 μM (b), 50 μM (c), 75 μM (d) and 100 μM (e) DNA. Note: The direction of arrow demonstrates increasing concentrations of DNA. Inside graph is the plot of A0/(A − A0) vs. 1/[DNA] to find the binding constant of the PIL–DNA adduct. (B) Absorption spectra of 25 μM 1a, 1b, and 1c (a) in presence of different concentrations of BSA 25 μM (b), 50 μM (c), 75 μM (d), and 100 μM (e). Note: The direction of arrow demonstrates increasing concentrations of BSA.

the PILs upon addition of CT-DNA in a solution, the spectra of 1a and 1c displayed a negligible blue shift (∼ 2 nm) but 1b showed red shift (∼3 nm) (Fig. 11A). Isosbestic point near at ca. 258 nm was observed for the 1b. This type of spectral changes indicates the covalent binding of ionic protic liquids of cyclotriphosphazenes with DNA. The interaction of 1a–1c with BSA can be observed by comparison of UV–visible absorption spectra of the free PIL and PIL–BSA adduct. The absorption spectra of the BSA solutions (0–100 μM) in the absence and presence of the PILs (25 μM) are shown in Fig. 11B. The 1a caused an increase in the absorbance of BSA and exhibited a slightly blue shift, indicating a van der Waals contacts or hydrogen bonds during interaction with BSA. The 1b and 1c displayed a moderate red shift, while all PILs caused hyperchromic effect indicates the intercalative binding mode of molecules to DNA double helix. The florescence of EB increases in the presence of CT-DNA. However, another molecule intercalating into DNA give rise to a decrease in the fluorescence intensity of the EB-DNA and this reduction can be used to determine binding constant of relevant molecule. The emission spectra of EB bound to CT-DNA in the absence and presence of PILs are shown in Fig. 12A. Upon increasing concentration of CT-DNA the emission intensities of 1a–1c exhibited hypochromism indicating the strong stacking interaction between the PILs and base pairs of DNA. The extent of quenching of EB bound to CT-DNA by 1a–1c is shown the linear Stern–Volmer equation (I0/I = 1 + KSV [Q]), which provides further

evidence that the PILs bind to DNA [46]. The KSV value for 1a, 1b, and 1c are 8.2 × 103 M−1, 2.8 × 103 M−1, and 1.1 × 105 M−1, respectively. The interaction of 1a or 1b with CT-DNA is stronger than that of 1c. An understanding of the properties of the PIL interactions with bovine serum albumin that is the main protein in the circulatory system can provide more information about absorption, distribution, metabolism, and excretion. The fluorescence emission intensity of BSA when excited at 280 nm is mainly due to the presence of the two tryptophan residues. The fluorescence spectra of BSA were observed in the presence of increasing amounts of 1a–1c. As shown in Fig. 12B, the fluorescence intensity of BSA reduced regularly with increasing concentration of these PILs. This indicated that these PILs could bind to BSA strongly. Isosbestic point near at ca. 380 nm and 560 nm were observed for the 1a and 1b, respectively. The KSV value for 1a, 1b, and 1c are 1.1 × 105 M−1, 1.8 × 105 M−1, and 5.0 × 104 M−1, respectively. The interaction of 1c with BSA is stronger than those of 1a or 1b. After KpnI and BamHI digestion of pTOLT plasmid DNA, digestion products was identified by two DNA bands in the absence PILs (Lane 4), whereas in the presence of 1a–1c produced three bands (Lane 1–3) (Fig. 13) indicating incomplete DNA digestion. Treatment of KpnI and BamHI with these PILs inhibits the restriction endonucleases activity of these enzymes and new band attributed to the whole plasmid. The results indicated that these PILs probably bound to pTOLT plasmid DNA.

494


Fig. 12. (A) The emission spectra of EB-bound (a) DNA solutions in the absence and presence of increasing concentrations of 3 25 μM (e), 50 μM (d) and 75 μM (c). [EB] = 10.0 μM (b), [DNA] 50.0 μM. The arrows show the changes in intensity upon increasing amounts of 1a–1c. Inset shows the plots of emission intensity I0/I vs. [Q] (μM) for determining KSV. (B) Emission spectra of BSA (13 μM) in presence of 1a–1c (25–75 μM). The arrow shows the emission intensity changes upon increasing PIL concentration. Insets: Stern–Volmer plot of the fluorescence data.

4. Conclusions In conclusion, synthesis of three novel PILs (1a, 1b, and 1c) has been attempted and characterized by various spectroscopic methods and investigated for their biological activity in cultured cell lines. The binding of 1a, 1b or 1c to CT-DNA and BSA resulted in significant changes in

spectral characteristics. The PILs showed hyperchromic absorption spectra and interacted directly with CT-DNA through a groove binding mode. The PILs were tendency of a van der Waals contacts or hydrogen bonds during interaction with BSA. The fluorescence emission of these PILs was efficiently quenched, which indicated that PILs could interact directly with CT-DNA or BSA. We have demonstrated that PILs are potent anticancer drug candidates with low cytotoxic, strong apoptotic, and effective DNA topoisomerase inhibitory characteristics. These findings provide important preliminary data for the use of our compounds against cancer cell line and suggest a new opportunity for enhancing efficacy and reducing toxicity by optimizing with metals and offer suitable drug pipeline for further pharmacological testing. Consequently, we think that there is a great opportunity about new molecules built from aminocyclophosphazenes with better solubility in biological fluid and less toxicity. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.molliq.2017.01.067. References

Fig. 13. Inhibition of KpnI and BamHI restriction endonucleases activity. Following 4 h 37 °C digestion of the 14 μL with 10 U KpnI and BamHI, these digestion products were resolved with 1.5% agarose gel containing ethidium bromide. Lane 1: enzyme + DNA + 1a, Lane 2: enzyme + DNA + 1b, Lane 3: enzyme + DNA + 1c, Lane 4: Positive control (enzyme + DNA), Lane 5: Negative control (plasmid DNA + water); lane 6: DNA marker (1Kb).

[1] H.R. Allcock, Chemistry and Applications of Polyphosphazenes, Wiley New York, 2003 (Chapter 3). ISBN: 978-0-471-44371-1. [2] V. Chandrasekhar, P. Thilagar, B.M. Pandian, Coord. Chem. Rev. 251 (2007) 1045, http://dx.doi.org/10.1016/j.ccr.2006.07.005. [3] J. Choi, Y. Kang, D. Kima, Electrochim. Acta 89 (2013) 359–364, http://dx.doi.org/10. 1016/j.electacta.2012.11.083. [4] B.A. Omotowa, B.S. Phillips, J.S. Zabinski, J.M. Shreeve, Inorg. Chem. 43 (2004) 5466–5471, http://dx.doi.org/10.1021/ic049483o.

H. Akbaş et al. / Journal of Molecular Liquids 230 (2017) 482–495 [5] K.L. Gering, M.K. Harrup, H.W. Rollins, US Patent App. 13 (253) (2011) 707. [6] H.R. Allcock, M.L. Levin, P.E. Austin, Inorg. Chem. 25 (14) (1986) 2281–2288, http:// dx.doi.org/10.1021/ic00234a002. [7] X. Wang, A.Y.X. Tan, C.M. Cho, Q. Ye, C. He, R. Ji, H.Q. Xie, J.W.H. Tsai, J. Xu, Tribol. Int. 90 (2015) 257–262, http://dx.doi.org/10.1016/j.triboint.2015.04.018. [8] Z. Song, Y. Liang, M. Fan, F. Zhou, W. Liu, Tribol. Int. 70 (2014) 136–141, http://dx. doi.org/10.1016/j.triboint.2013.09.025. [9] G. Yenilmez Çiftçi, E. Şenkuytu, M. Bulut, M. Durmuş, J. Fluoresc. 25 (2015) 1819–1830, http://dx.doi.org/10.1007/s10895-015-1672-4. [10] R. Bao, M. Pan, J.J. Qiu, C.M. Liu, Chin. Chem. Lett. 21 (2010) 682–685, http://dx.doi. org/10.1016/j.cclet.2009.12.020. [11] J. Jimenez, A. Laguna, E. Gascon, J.A. Sanz, J.L. Serrano, J. Barbera, L. Oriol, Chem. Eur. J. 18 (2012) 16801–16814, http://dx.doi.org/10.1002/chem.201202748. [12] J. Jimenez, I. Pintre, E. Gascon, C. Sanchez-Somolinos, R. Alcala, E. Cavero, J.L. Serrano, L. Oriol, Macromol. Chem. Phys. 215 (2014) 1551–1562, http://dx.doi.org/10.1002/ macp.201400190. [13] S. Ahn, H.S. Kim, S. Yang, J.Y. Do, B.H. Kim, K. Kim, J. Electroceram. 23 (2009) 289–294, http://dx.doi.org/10.1007/s10832-008-9437-y. [14] T. Nishimoto, T. Yasuda, S.Y. Lee, R. Kondo, C. Adachi, Mater. Horiz. 1 (2014) 264–269, http://dx.doi.org/10.1039/C3MH00079F. [15] P. Schrogel, M. Hoping, W. Kowalsky, A. Hunze, G. Wagenblast, C. Lennartz, P. Strohriegl, Chem. Mater. 23 (2011) 4947–4953, http://dx.doi.org/10.1021/ cm201940f. [16] H. Liu, X. Wang, D. Wu, Thermochim. Acta 607 (2015) 60–73, http://dx.doi.org/10. 1016/j.tca.2015.04.002. [17] H. Liu, X. Wang, D. Wu, Polym. Degrad. Stab. 103 (2014) 96–112, http://dx.doi.org/ 10.1016/j.polymdegradstab.2013.02.008. [18] J. Wang, X. Su, Z. Mao, Polym. Degrad. Stab. 109 (2014) 154–161, http://dx.doi.org/ 10.1016/j.polymdegradstab.2014.07.010. [19] R. Yang, W. Hu, L. Xu, Y. Song, J. Li, Polym. Degrad. Stab. 122 (2015) 102–109, http:// dx.doi.org/10.1016/j.polymdegradstab.2015.10.007. [20] P. Jiang, X. Gu, S. Zhang, J. Sun, S. Wu, Q. Zhao, Phosphorus Sulfur Silicon 189 (2014) 1811–1822, http://dx.doi.org/10.1080/10426507.2014.902828. [21] J. Sun, X. Wang, D. Wu, ACS Appl. Mater. Interfaces 4 (2012) 4047–4061, http://dx. doi.org/10.1021/am300843c. [22] F. Chang, X. Huang, H. Wei, K. Chen, C. Shan, X. Tang, Mater. Lett. 125 (2014) 128–131, http://dx.doi.org/10.1016/j.matlet.2014.03.137. [23] H. Ozay, O. Ozay, Colloids Surf. A Physicochem. Eng. Asp. 450 (2014) 99–105, http:// dx.doi.org/10.1016/j.colsurfa.2014.03.022. [24] D. Erdener Çıralı, Z. Uyar, İ. Koyuncu, N. Hacıoğlu, Appl. Organomet. Chem. 29 (2015) 536–542, http://dx.doi.org/10.1002/aoc.3328. [25] Y. Tümer, L.Y. Koç, N. Asmafliz, Z. Kılıç, T. Hökelek, H. Soltanzade, L. Açık, M.L. Yola, A.O. Solak, J. Biol. Inorg. Chem. 20 (2015) 165–178, http://dx.doi.org/10.1007/ s00775-014-1223-5.

495

[26] N.S. Başterzi, S. Bilge Koçak, A. Okumuş, Z. Kılıç, T. Hökelek, Ö. Çelik, M. Türk, L.Y. Koç, L. Açık, B. Aydın, H. Dal, New J. Chem. 39 (2015) 8825–8839, http://dx.doi.org/10. 1039/C5NJ01530H. [27] E. Cil, M.A. Tanyildizi, F. Ozen, M. Boybay, M. Arslan, A.O. Gorgulu, Arch. Pharm. Chem. Life Sci. 345 (2012) 476–485, http://dx.doi.org/10.1002/ardp.201100412. [28] M. Işıklan, N. Asmafiliz, E.E. Özalp, E.E. İlter, Z. Kılıç, B. Çoşut, S. Yeşilot, A. Kılıç, A. Öztürk, T. Hökelek, L.Y. Koç Bilir, L. Açık, E. Akyüz, Inorg. Chem. 49 (2010) 7057–7071, http://dx.doi.org/10.1021/ic100781v. [29] N. Asmafiliz, Z. Kılıç, Z. Hayvalı, L. Açık, T. Hökelek, H. Dal, Y. Öner, Spectrochim. Acta A 86 (2012) 214–223, http://dx.doi.org/10.1016/j.saa.2011.10.027. [30] H. Akbaş, A. Okumuş, Z. Kılıç, T. Hökelek, Y. Süzen, L.Y. Koç, L. Açık, Z.B. Çelik, Eur. J. Med. Chem. 70 (2013) 294–307, http://dx.doi.org/10.1016/j.ejmech.2013.09.046. [31] T. Yıldırım, K. Bilgin, G. Yenilmez Çiftçi, E. Tanrıverdi Eçik, E. Şenkuytu, Y. Uludağ, L. Tomak, A. Kılıç, Eur. J. Med. Chem. 52 (2012) 213–220, http://dx.doi.org/10.1016/j. ejmech.2012.03.018. [32] A.O. Görgülü, K. Koran, F. Özen, S. Tekin, S. Sandal, J. Mol. Struct. 1087 (2015) 1–10, http://dx.doi.org/10.1016/j.molstruc.2015.01.033. [33] S.S. Machakanur, B.R. Patil, G.N. Naik, R.P. Bakale, S.W.A. Bligh, K.B. Gudasi, Inorg. Chim. Acta 421 (2014) 459–464, http://dx.doi.org/10.1016/j.ica.2014.06.040. [34] M. Siwy, D. Sek, B. Kaczmarczyk, J. Wietrzyk, A. Nasulewicz, A. Opolski, Anticancer Res. 27 (2007) 1553–1558. [35] K. Brandt, T.J. Bartczak, R. Kruszynski, I. Porwolik-Czomperlik, Inorg. Chim. Acta 322 (2001) 138–144, http://dx.doi.org/10.1016/S0020-1693(01)00557-6. [36] X. Zhu, Y. Liang, D. Zhang, L. Wang, Y. Ye, Y. Zhao, Phosphorus Sulfur Silicon Relat. Elem. 186 (2011) 281–286, http://dx.doi.org/10.1080/10426507.2010.496120. [37] H. Akbaş, A. Okumuş, A. Karadağ, Z. Kılıç, T. Hökelek, Y. Süzen, L.Y. Koç, L. Açık, B. Aydın, M. Türk, J. Therm. Anal. Calorim. 123 (2) (2016) 1627–1641, http://dx.doi. org/10.1007/s10973-015-5001-6. [38] P. Walden, Bull. Acad. Imper. Sci. St. Pétersbourg. 8 (6) (1914) 405–422. [39] N. Bicak, J. Mol. Liq. 116 (2005) 15–18, http://dx.doi.org/10.1016/j.molliq.2004.03.006. [40] A. Karadağ, A. Destegül, J. Mol. Liq. 177 (2013) 369–375, http://dx.doi.org/10.1016/j. molliq.2012.10.040. [41] H.R. Allcock, T.J. Fuller, K. Matsumura, Inorg. Chem. 21 (1982) 515–521, http://dx. doi.org/10.1021/ic00132a009. [42] Bruker program 1D WIN-NMR (release 6.0) and 2D WIN-NMR (release 6.1). [43] T.L. Greaves, C.J. Drummond, Chem. Rev. 108 (1) (2008) 206–237, http://dx.doi.org/ 10.1021/cr068040u. [44] Z. Kılıç, N. Gündüz, E. Kılıç, H. Nazır, Synth. React. Inorg. Met.-Org. Chem. 19 (5) (1989) 453–467, http://dx.doi.org/10.1080/00945718908048084. [45] G.A. Carriedo, F.G. Alonso, P.A. Gonzalez, J.R. Menendez, J. Raman Spectrosc. 29 (4) (1998) 327–330, http://dx.doi.org/10.1002/(SICI)1097-4555(199804)29:4b327:: AID-JRS245N3.0.CO;2-5. [46] M. Sirajuddin, S. Ali, A. Badshah, J. Photochem, Photochem. Photobiol. B Biol. 124 (2013) 1–19, http://dx.doi.org/10.1016/j.jphotobiol.2013.03.013.