Journal of Inorganic Biochemistry 145 (2015) 94–100
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry journal homepage: www.elsevier.com/locate/jinorgbio
Synthesis, structural studies and biological activity of new Cu(II) complexes with acetyl derivatives of 7-hydroxy-4-methylcoumarin Marcin T. Klepka a,⁎, Aleksandra Drzewiecka-Antonik a, Anna Wolskaa, Paweł Rejmak a, Kinga Ostrowskab, Elżbieta Hejchman b, Hanna Kruszewskac, Agnieszka Czajkowskad, Izabela Młynarczuk-Biały d, Wiesława Ference a
Institute of Physics, Polish Academy of Sciences, Al. Lotnikow 32/46 PL-02668, Warsaw, Poland Faculty of Pharmacy, Medical University of Warsaw, Banacha 1, PL-02097, Warsaw, Poland c National Medicines Institute, Chelmska 30/34, PL-00725, Warsaw, Poland d Centre of Biostructure Research, Medical University of Warsaw, Chalubinskiego 5, PL-02004, Warsaw, Poland e Faculty of Chemistry, Maria Curie-Sklodowska University, Sq. Maria Curie-Sklodowska 2, PL-20031, Lublin, Poland b
a r t i c l e
i n f o
Article history: Received 5 November 2014 Received in revised form 14 January 2015 Accepted 15 January 2015 Available online 23 January 2015 Keywords: Cu(II) complex Coumarin derivative 7-hydroxycoumarin XAFS DFT Electrochemical synthesis
a b s t r a c t The new Cu(II) complexes with 6-acetyl-7-hydroxy-4-methylcoumarin (HL1) and 8-acetyl-7-hydroxy4-methylcoumarin (HL2) have been obtained by the electrochemical method. The density functional theory calculations and X-ray absorption spectroscopy techniques have been used to geometrically describe a series of new compounds. The studies have been focused on the coordination mode of the hydroxy ligands to the metallic centre. The complexes, Cu(HL1) 2 and Cu(HL2) 2 ⋅ 0.5H 2O, have flat square geometry with oxygen atoms in the first coordination sphere. Two bidentate anionic coumarins are bonded to the metal cation via the acetyl and deprotonated hydroxyl O atoms. Biological activity, including microbiological and cytotoxic, has been evaluated and found to be enhanced in comparison with the parent ligands. Moreover, the Cu(II) complex with 8-acetyl-7-hydroxy-4-methylcoumarin shows similar antifungal activity as commercially used fluconazole. © 2015 Elsevier Inc. All rights reserved.
1. Introduction The natural as well as synthetic coumarins, therein hydroxycoumarins, have a large spectrum of biological activity. Such derivatives proved usefulness as anti-coagulants [1], antibacterial agents [2], antifungal agents [3], biological inhibitors [4], chemotherapeutics [5] and as bio-analytical reagents [6]. It has been found out that coordination of metal ions to therapeutic agents (such as simple coumarins) can improve their efficacy and accelerate bioactivity. In many cases such metal complexes are more potent and less toxic compared to the parent drug. Therefore, among others, also biologically active metal complexes of coumarin based ligands are being widely investigated. Among them, triorganotin(IV) derivatives of umbelliferone (7hydroxycoumarin) have shown good antimicrobial activity against Staphylococcus aureus, Bacillus subtilis, Candida albicans and Microsporum gypseum, and this activity was slightly enhanced upon adduct formation with 1,10-phenanthroline. Creaven et al. have investigated the antimicrobial activity of a number of coumarin complexes ⁎ Corresponding author. E-mail address: [email protected] (M.T. Klepka).
http://dx.doi.org/10.1016/j.jinorgbio.2015.01.006 0162-0134/© 2015 Elsevier Inc. All rights reserved.
with silver(I), copper(II) and manganese(II) ions. For example, the Cu (II) complexes exhibit antifungal activity against a clinical strain of C. albicans comparable to that of the commercially available antifungal drugs, i.e. ketoconazole and Amphotericin B [7]. This paper is focused on the synthesis and characterization of copper complexes of two hydroxyligands: HL1 and HL2 (Fig. 1). These ligands have acetyl group attached at two different positions, C6 and C8, to the rigid coumarin ring. The electrochemical method was applied for the synthesis of the complexes. In order to determine the geometry of metal-ligand interactions X-ray crystallography is usually used. However, this technique requires compound in the crystal form, which is not always possible to obtain. Since that was the case for the investigated complexes, the X-ray absorption spectroscopy (XAS) was applied. The great advantage of XAS is that it can be used for crystal as well as amorphous materials at different states: solid, liquid or gaseous. XAS provides information about the local atomic order, coordination number, kind of atoms, oxidation state, relative disorder and even angles between central atom and near neighbours [8–10]. Data obtained from XAS combined with the density functional theory (DFT) calculations allowed building models of the complexes. Additionally to
M.T. Klepka et al. / Journal of Inorganic Biochemistry 145 (2015) 94–100
R1
CH3
5
6
4
2
7
HO 3
R1
3
8
R2
O 1
O 2
R2
HL1 C(=O)CH3 H HL2 H
C(=O)CH3
95
reference compounds (two copper oxides, Cu2O and CuO) were also measured. The quantitative analysis of EXAFS spectra of compounds 1–2 was performed as follows: k2 weighted χ(k) data were Fourier transformed in the k range 2.4–11 for 1 and 2.5–10 for 2. The Fourier's back transformation ranges were from 1 to 2 Å for both complexes. For the XAS analysis, the Athena and Artemis programs included in the IFEFFIT package [14] were applied. 2.5. DFT calculations
Fig. 1. Molecular structure of ligands.
geometrical characterization of complexes also the biological activity tests, including microbiological and cytotoxic studies, were performed. 2. Experimental 2.1. Instrumentation The elemental analysis was performed using the Perkin-Elmer 2400 CHN elemental analyzers. The thermal stability and decomposition process of complexes were studied in air using the Setsys 16/18 (Setaram) TG, DTG and DSC instruments. The experiments were carried under air flow in the temperature range of 293–823 K at a heating rate of 5 K/min. The initial masses of samples of complexes used for measurements changed from 4.59 mg to 4.56 mg for 1 and 2.01 mg to 1.96 mg for 2. Samples of these compounds were heated in Al2O3 crucibles. The IR spectra were gathered between 400 and 4000 cm−1 in the ATR mode on a Thermo Scientific Nicolet iS5 spectrometer. 2.2. Synthesis of ligands The initial reagents were bought from Sigma-Aldrich. The synthesis of 6-acetyl-7-hydroxy-4-methylcoumarin (HL1) and 8-acetyl-7-hydroxy-4-methylcoumarin (HL2) was performed as described in literature [11,12]. 2.3. Synthesis of complexes The copper complexes, 1–2, were obtained by an electrochemical synthesis [13]. For the complexation reaction the 7-hydroxycoumarins, HL1 and HL2, were used as ligands. The copper plate was used as a metal anode and a platinum wire was the cathode. A 96% ethanol solution of HL1 and HL2 containing about 80 mg of tetrabutylammonium perchlorate (electrolyte) was electrolyzed for 2 h (see Table 1). After filtration of the resulting solution, the powder product was purified by washing thoroughly with water. 2.4. X-ray absorption spectroscopy The K-edges of Cu for the complexes 1 and 2 were measured at the beamline I811 at MAX-lab in Lund, Sweden. Both XANES (X-ray absorption near edge structure) and EXAFS (extended X-ray absorption fine structure) spectra were recorded in the transmission mode. To confirm the oxidation state of metal in the complexes, the XANES spectra of the
The DFT calculations were performed using Turbomole 6.5 code [15]. Relying on reasonable assumption, that weak interactions keeping molecular solids together should not affect strongly the properties of molecules, structural models consisting of single molecule were used. Such models were successfully applied in elucidation of structural and spectroscopic properties of metal complexes with coumarin derivatives [16]. Geometry optimization was carried out using gradient Perdew– Burke–Ernzerhof (PBE) [17] and hybrid PBE0 (i.e. PBE with 25% of Hartree–Fock exchange) [18] exchange-correlation functionals with def2-TZVPP basis set [19,20]. The convergence criteria for total energy and gradient norm were 10−6 Eh and 10−4 Eh/bohr, respectively. Harmonic vibrational analysis was performed at PBE level to verify, if true minimum of energy was achieved and support quantitative analysis of FT-IR spectra. 2.6. Biological activity In order to check the biological activity of complexes 1–2 microbiological and cytotoxic tests were carried out and afterwards compared with the parent ligands (HL1–HL2) described before [21]. 2.6.1. Microbiological assays The following microbial strains were chosen from American Type Culture Collection (ATCC): bacteria Gram-positive strains: Micrococcus luteus (9341, 10240), B. subtilis 6633, Bacillus cereus 11778, S. aureus (6538, 6538P), bacteria Gram-negative strains: Escherichia coli (8739, 10536), Pseudomonas aeruginosa (15442, 9027), fungi (yeast strains): C. albicans (10231, 2091), Candida parapsilosis 22019, Zygosaccharomyces rouxii 28253, Saccharomyces cerevisiae 9763 and a mould strain Aspergillus brasiliensis 16404. The MRSA hospital strain 573/12 (methicillin resistant S. aureus) isolated from peripheral blood was obtained from MUW Museum Collection. According to disc-diffusion Kirby-Bauer method, the 573/12 strain is resistant to: penicillin, erythromycin, amikacin, clindamycin, and ciprofloxacin. 2.6.2. Antimicrobial activity — preliminary test In the preliminary tests antimicrobial activity was determined by a modified cylinder-plate method [22]. The inhibition of bacterial growth was observed as a halo around the cylinder containing the tested compound. The size of inhibition zone reflected an antimicrobial activity of the examined compounds. 2.6.3. Minimum inhibitory concentration (MIC) For the compound which showed some activity against any of the tested strains an MIC was determined based on M7-A6 method [23]. The lowest concentration of tested compound, which totally inhibited
Table 1 Experimental conditions for the synthesis of complexes 1–2. Complex
Amount of solved copper [g]
Ligand
Amount of HL [g]
Voltage⁎ [V]
Colour of complex
Elemental analysis (C, H calc⁎⁎/found %)
1 2
0.0415 0.0515
HL1 HL2
0.0814 0.0814
29 18
green green
57.89, 3.64/57.22, 3.72 56.86, 3.85/56.78, 3.75
⁎ Voltage to produce a current of 10 mA. ⁎⁎ Anal. calc. for Cu(HL1)2 and Cu(HL2)2∙0.5 H2O.
96
M.T. Klepka et al. / Journal of Inorganic Biochemistry 145 (2015) 94–100
Table 2 The characteristic IR experimental frequencies ν (cm-1) of carbonyl groups in ligands (HL1–HL2) and complexes (1–2), along with calculated harmonic DFT/PBE frequencies ω(cm-1).
HL1 HL2 1 2
ν(C = O)
ω(C = O)DFT/PBE
1725coumarin, (1550–1675)acetyl 1730coumarin, (1550–1650)acetyl 1737coumarin, (1550–1625)acetyl 1730coumarin, (1550–1640)acetyl
1759coumarin, (1565–1635)acetyl 1761coumarin, (1550–1625)acetyl 1753coumarin, (1550–1590)acetyl 1760coumarin, (1535–1585)acetyl
growth of the examined strain, was taken as MIC value. The negative control was performed with the solution of 10% DMSO (in 0.08 M phosphate buffer, pH 7.0). All tests were performed in triplicate. 2.7. Cytotoxicity and antiviral assays 2.7.1. Reagents Synthesized substances were dissolved in DMSO (Sigma-Aldrich) at 20 mM and kept in 4 °C prior to use. 2.7.2. Cell culture The human prostate cancer cells DU145 and NIH3T3 mouse fibroblasts, both from ATCC, were maintained in humidified atmosphere containing 5% CO2 at 37 °C. DU145 and NIH3T3 cells were cultured in RPMI and DMEM media, respectively, both supplemented with stable glutamate and with 10% foetal calf serum (FCS) as well as with antibiotic antimycotic (Sigma-Aldrich). Cells were passaged every 2–3 days. 2.7.3. Cytotoxicity assays For the analysis of cytostatic/cytotoxic effects induced by new compounds, resarium based PrestoBlue assay (Invitrogen) was performed according to manufacturer's instructions [24].
Fig. 3. Fourier transformed experimental EXAFS oscillations for complexes 1 (panel a) and 2 (panel b) (black line) and fitted first coordination sphere (red dot line). As an inset EXAFS oscillations together with fitted first coordination sphere are presented.
3. Results and discussion Both electrochemically synthesized compounds were obtained as green powders. Complex 1 was found to be anhydrous, whereas 2 contains approximately half of the water molecule per complex which was confirmed by thermal analysis. The mass loss of dehydration process calculated from thermogravimetric (TG) curve (323–448 K) is equal to 2.4% for 2 which indicates that the complex 2 released approximately half of the water molecule in one step. The dehydration process is
connected with the endothermic effect (differential scanning calorimetry (DSC) curve). In the second step of decomposition (in the range of 473–823 K) these two complexes eliminate gradually the organic ligands with the exothermic effect on DSC curve and ultimately form the CuO. For both cases, the elemental analysis determined metal to ligand ratio as 1:2 [Cu(HL1)2 (1) and Cu(HL2)2∙0.5H2O (2)]. 3.1. FTIR spectroscopy The complexation reactions were monitored by infrared spectroscopy. The IR analysis was focused on the stretching frequencies of carbonyl groups, which contribute to the most intense bands and are quite well separated from other molecular motions. These characteristic bands observed for the parent ligands and their complexes are presented in Table 2. The comparison of the IR spectra of the complexes (1–2) with the spectra of their ligands (HL1–HL2) reveals that: (i) the peaks corresponding to the (C = O)coumarin stretching are intact or even blue shifted in complex, which precludes their bonding to Cu, (ii) the peaks corresponding to the (C = O) acetyl stretching are red shifted, which indicates the coordination to Cu Table 3 EXAFS parameters of Cu(II) complexes: cation—O distance—R, number of atoms in a coordination sphere—N, Debye–Waller factor—σ2 and R-factor.
Fig. 2. XANES Cu K-edge spectra of the metal complexes 1 and 2 together with reference oxides Cu2O and CuO.
Complex
Bond
R [Å]
N
σ2 [Å2]
R-factor
1 2
Cu–O Cu–O
1.93(4) 1.94(1)
3.9(2) 4.0(1)
0.004(1) 0.004(1)
0.001 0.001
M.T. Klepka et al. / Journal of Inorganic Biochemistry 145 (2015) 94–100
97
Fig. 4. The relative DFT/PBE0 energy of conformers of complex 2 with respect to torsion angle defined by 4 chelating O atoms. 0 and 180° torsions correspond to C2h and C2v structures, respectively.
through the carbonyl O atom of acetyl group. (C = O)acetyl vibrations generally correspond to lower frequencies due to the conjugation to phenol ring and, for pristine ligand, due to the participation in the intramolecular hydrogen bond. DFT calculations revealed also that (C = O)acetyl vibration component is present in several normal modes of molecules, thus the range of infrared frequencies with possible (C = O)acetyl mode contribution is given in Table 2. The assignment of acetyl and coumarin (C = O) modes is consistent with the reported carbonyl stretching frequencies in 2-acetylphenol (1646 cm− 1 and below) [25] and in 2-pyrone (1732 cm− 1) [26]. The fingerprint regions have also been analysed. The bands present in the spectra of the ligands are observed at different positions than the respective bands in the IR spectra of their metal complexes. These data are analogous to the spectra of similar series of Cu(II) complexes in which two bidentate ligands are coordinated to the metal centre via O atoms from the acetyl and hydroxyl groups substituted on the rigid benzo[b]furan system [27]. The main difference between these two series of complexes is the lack of a broad band at 3300–3600 cm1 region in the IR spectra of coumarin derivatives (1–2). This band (observed for benzofuran complexes of copper) is responsible for OH stretching in water molecules. Its absence agrees with the data obtained from thermal analysis.
Fig. 5. DFT optimized molecular structures of complexes 1 (panel a) and 2 (panel b).
energy positions of the Cu K-edge, evaluated by comparison of position of the first derivative maximum, in compounds 1–2 and CuO are similar,
3.2. XAFS analysis of complexes XANES, due its sensitivity to the chemistry of the absorbing atom, was used to determine the oxidation state and local geometry around the Cu cation in the investigated compounds. The complexes were synthesized by an electrochemical method (from metallic copper), therefore, it is important to verify the copper oxidation state in new compounds. Thus, the energy positions of copper edges in the complexes were compared with the reference oxides: Cu2O and CuO. The
Table 4 Selected calculated bond distances Cu–O/C (Å) and valence angles Cu–O–C (°) of complexes 1–2. PBE
Cu–O3 Cu–Oacetyl Cu–C7 Cu–Cacetyl Cu–C6/C8 Cu–O3–C7 Cu–Oacetyl–Cacetyl
PBE0
1
2
1
2
1.908 1.962 2.893 2.925 3.325 128.25 128.77
1.902 1.943 2.899 2.944 3.362 129.44 131.91
1.881 1.944 2.865 2.900 3.297 128.98 129.36
1.877 1.925 2.870 2.918 3.333 130.04 132.50
Fig. 6. Experimental and calculated Cu K-edge XANES spectra of complexes 1 (panel a) and 2 (panel b).
98
M.T. Klepka et al. / Journal of Inorganic Biochemistry 145 (2015) 94–100
Table 5 Antibacterial activity (in mg/mL) of tested complexes. Strains Micrococcus luteus ATCC 9341 Bacillus cereus ATCC 11178 Bacillus subtilis ATCC 6633 Staphylococcus epidermidis ATCC 12228 Staphylococcus aureus ATCC 6538 Staphylococcus aureus ATCC 6538P Enterococcus hirae ATCC 10541 Escherichia coli ATCC 8739 Pseudomonas aeruginosa ATCC 9027 Pseudomonas aeruginosa ATCC 15442 Staphylococcus aureus MRSA 573/12
Table 6 Antifungal activity (in mg/mL) of tested compounds.
1
2 1.25
0.01875
0.625
0.0375
2.5
0.075
2.5
0.15
N10
0.625
N10
0.3
N10
0.15
N10
N10
N10
N10
N10
N10
N10
Strains
1
2
Candida albicans ATCC 10231 Candida albicans ATCC 2091 Candida parapsilosis ATCC 22019 Aspergillus brasiliensis ATCC 16404 Zygosacharomyces rouxi ATCC 28253 Saccharomyces cerevisiae ATCC 9763
N10
0.075
N10
0.0375
N10
0.075
N10
0.15
N10
0.01875
N10
0.0375
1.93 and 1.90 Å at PBE and PBE0 levels, respectively, which is in a good agreement with EXAFS data (see Table 3). It should be stressed that using nonplanar structures as initial points in geometry optimization resulted in convergence to nearly planar local minima of energy. Conformational analysis (Fig. 4) proves that structures with pseudotetrahedral Cu(II) coordination are significantly less stable in respect to flat species. Formally, Cu(II) ion can be coordinated by two 2-acetylphenol type ligand in two ways, with resulting complex with the inversion centre or without. For planar molecules these both possible isomers would correspond to C2h and C2v point groups, respectively. The DFT studies showed that both isomers have the same energetic stability. Taking into account that X-ray crystal structure analysis on similar Cu(II) complexes reported centrosymmetric type of coordination [27–30], such type of coordination was considered as the most probable and used as an initial model for XANES data analysis (see Fig. 5). The issue of apparent preference of centrosymmetric isomer in nature is under further theoretical investigation by our group.
0.3
which implies that analysed complexes contain mostly copper in the +2 oxidation state (Fig. 2). Moreover, a similar local density of unoccupied Cu 4p states, represented by near edge absorption spectra, for both complexes is observed. This suggests a similar metal coordination environment for analysed complexes of hydroxycoumarin. The fitting of EXAFS spectra was performed in an analogous way for compounds 1 and 2 (Fig. 3). For both complexes four O atoms were found in the first coordination sphere (Table 3). The Cu–O bond lengths are around 1.94 Å. Any additional copper ion was not identified in the first nor farther coordination spheres, which proves that the multinuclear copper complexes are not formed.
3.4. XANES — full potential multiple scattering 3.3. DFT calculations
Structural models from DFT calculations were used as the input structures for full potential multiple scattering (FPMS) XANES calculations. Several attempts were made to calculate Cu K-edge XANES spectra for these DFT structures. In the first step standard approach using muffin tin potential was applied. However, this type of potential does not work properly for the flat geometry, which is the case for the considered Cu(II) complexes. In order to perform reliable calculations the FPMS code developed by Hatada and coworkers [31,32] was used. The real part of Hedin–Lundqvist potential (HL) was employed [33]. The energy dependent broadening due to the inelastic scattering was taken into account for FPMS spectra by MXAN program [34,35]. Comparison of the calculated and experimental spectra for complexes 1 and 2 is
The results of FTIR, XAFS spectroscopy, elemental analysis of compounds 1 and 2 as well as previously reported structures of Cu(II) complexes with ligands containing 2-acetylphenol ring [27] pointed out towards the structural models with Cu(II) being coordinated by two coumarin ligands in planar, bidentate mode. The proposed molecules of complexes 1–2 consist of two HL anions chelating to the metal through the O atom from deprotonated hydroxyl group and O atom from the acetyl substituent. The results of geometry optimization of such models are shown in Fig. 4 and summarized in Table 4. Calculated mean Cu–O average distances, for both analysed complexes, are about
Table 7 Comparison of antibacterial activities (MIC values in mg/mL) between ligands and complexes. Strains
HL1
1
HL2
2
Micrococcus luteus ATCC 9341 Bacillus cereus ATCC 11178 Bacillus subtilis ATCC 6633 Staphylococcus aureus ATCC 6538 Staphylococcus aureus ATCC 6538P Escherichia coli ATCC 8739
45.8
1.25
68.7
0.01875
4.6
0.625
22.9
0.0375
22.9
2.5
45.8
0.075
45.8
N10
68.7
0.625
–
N10
–
0.3
–
N10
–
N10
M.T. Klepka et al. / Journal of Inorganic Biochemistry 145 (2015) 94–100
99
to the our best knowledge, no such activity was reported for complexes with Cu(II) being chelated by 2-acetylphenol type ligand prior to our previous work [27]. Proposed mechanisms for antitumour activities of Cu complexes include DNA binding or topoisomerase inhibition. Compound 2 induces similar cell growth inhibition, although NIH3T3 cells are slightly more susceptible than DU145 cells. 4. Conclusions
Fig. 7. Cytostatic/cytotoxic activity of complexes 1 (panel a) and 2 (panel b) against DU145 and NIH3T3.
shown in Fig. 6. It is clearly seen that shape of the XANES spectra is reproduced which points out that considered model reproduces the Cu(II) coordination geometry of the electrochemically synthesized complexes. 3.5. Antimicrobial activity The MIC values counted on mmol of examined compound are presented in Tables 5 and 6. Complexes inhibited growth of Grampositive bacterial strains. Complex 2 was also active against tested fungal strains. The MIC concentration of 2 against fungal strains was between 0.0187 and 0.075 mg/mL, similar to MIC of fluconazole which is between 0.0125 and 0.256 mg/mL for different fungal strains [36]. Activity of both complexes increased in comparison witch parent ligands, see Table 7 [21]. 3.6. Antitumour activity For estimation of tumour-cell selectivity of examined compounds, viability assays were performed in tumourigenic DU145 cell line and in non-tumourigenic one, NIH3T3. Cells were treated as described in the experimental part of the paper. Both synthesized agents induced dose-dependent cytostatic/cytotoxic effects in examined cells. Prostate tumour cells are slightly more susceptible towards complex 1 than the non-tumourigenic mouse fibroblasts, this effect is most pronounced at 200 and 500 μM concentrations (Fig. 7). Cytotoxic activity was reported for many Cu(II) κ2-O, O complexes with planar environment of central ion [37]. However, up
Two new copper complexes with the acetyl derivatives of hydroxycoumarin as the O-donor ligands have been electrochemically synthesized, structurally described and biologically tested. Initial characterization including elemental and thermal analysis confirmed that complexes were formed with metal to ligand ratio as 1:2. The FTIR spectroscopy indicated the coordination to Cu(II) cation through the carbonyl O atom of the acetyl group. These information were used as starting point in EXAFS analysis. The EXAFS fitting parameters, correlated with DFT calculations indicated that centrosymmetric type of coordination, with four O atoms at average distance 1.94 Å, is formed. The XANES data confirmed that two bidentate ligands are coordinated to the metal centre via O atoms from the acetyl and hydroxyl groups substituted on the coumarin system. The first coordination sphere consists of four O-atoms and there is an indication that the central atom has a slightly distorted planar square geometry. Biological activity against Gram positive and negative bacteria: M. luteus, B. cereus, B. subtilis, and S. aureus, has significantly increased in comparison to the parent ligands (see Table 7). Furthermore, compound 2 showed activity against fungal strains, C. albicans, C. parapsilosis, A. brasiliensis, Zygosacharomyces rouxi, and S. cerevisiae. Complex 1 did not show activity against these strains (see Table 6). On the other hand, comparing viability tests, the compound 1 is more selective for tumour cells than 2. Complex with acetyl group at C6 position (1) shows activity against Gram-positive, Gram-negative bacteria and has better antitumour properties than 2. The compound containing the acetyl group at C8 position (2) is additionally active against fungal strains but has weaker antitumour properties. Observed differences in activity indicated that position of acetyl group modulates complex biological properties. Farther studies are planned on complexes with acetyl group at different positions. From application point of view it is required to have detailed geometric data, which are necessary to understand the mode of action of such complexes. In this paper it was demonstrated that XAS and computational techniques enable structural characterization in lack of crystallographic data. Acknowledgements Experimental research was funded from the Polish National Science Centre (Grant No. UMO-2012/07/D/ST5/02251). The synchrotron experiment was partially supported by the Baltic Science Link project coordinated by the Swedish Research Council, VR. Authors would like to acknowledge Dr. K. Hatada for valuable discussion and help in starting FPMS calculations. This research was supported in part by PL-Grid Infrastructure. P. Rejmak acknowledges EAgLE project no. 316014 for the financial support. References [1] J.W. Suttie, Clin. Cardiol. 13 (VI) (1990) 16–18. [2] A.H. Bedair, N.A. El-Hady, M.S. Abd El-Latif, A.H. Fakery, A.M. El-Agrody, Il Farmaco 55 (2000) 708–714. [3] T. Patonay, G. Litkei, R. Bognar, J. Eredi, C. Miszti, Pharmazie 39 (1984) 86–91. [4] C. Gnerre, M. Catto, F. Leonetti, P. Weber, P.A. Carrupt, C. Altomare, A. Carotti, B.J. Testa, Med. Chem. 43 (2000) 4747–4758. [5] D.A. Egan, P. James, D. Cooke, R. O'Kennedy, Cancer Lett. 118 (1997) 201–211. [6] M. Jiménez, J.J. Mateo, R. Mateo, J. Chromatogr. A 870 (2000) 473–481.
100
M.T. Klepka et al. / Journal of Inorganic Biochemistry 145 (2015) 94–100
[7] B.S. Creaven, D.A. Egan, D. Karcz, K. Kavanagh, M. McCann, M. Mahon, A. Noble, B. Thati, M. Walsh, J. Inorg. Biochem. 103 (2009) 1196–1203. [8] M.T. Klepka, A. Drzewiecka, A. Wolska, W. Ferenc, Chem. Phys. Lett. 553 (2012) 59–63. [9] M.T. Klepka, K. Lawniczak-Jablonska, A. Wolska, A. Slawska-Waniewska, C.A. Rodrigues, A. Debrassi, C. Bordini, Chem. Phys. Lett. 501 (2011) 523. [10] A. Drzewiecka, A.E. Koziol, M.T. Klepka, A. Wolska, S.B. Jimenez-Pulido, T. Lis, K. Ostrowska, M. Struga, Chem. Phys. Lett. 559 (2013) 41–45. [11] V.S. Moskvina, A.A. Ishchenko, V.P. Khilya, Chem. Nat. Compd. 48 (2012) 751–756. [12] A. Kumar, R.A. Maurya, S. Sharma, P. Ahmad, A.B. Singh, G. Bhatia, A.K. Srivastava, Bioorg. Med. Chem. Lett. 19 (2009) 6447–6451. [13] C. Oldham, D.G. Tuck, J. Chem. Educ. 59 (1982) 420–421. [14] B. Ravel, M. Newville, J. Synchrotron Radiat. 12 (2005) 537–541. [15] R. Ahlrichs, M. Bär, M. Häser, H. Horn, C. Kölmel, Chem. Phys. Lett. 162 (1989) 165–169. [16] I. Georgieva, Tz Mihaylov, N. Trendafilova, J. Inorg. Biochem. 135 (2014) 100–112. [17] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77 (1996) 3865–3868. [18] J.P. Perdew, M. Ernzerhof, K. Burke, J. Chem. Phys. 105 (1996) 9982–9985. [19] F. Weigend, M. Häser, H. Patzelt, R. Ahlrichs, Chem. Phys. Lett. 294 (1998) 143–152. [20] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 7 (2005) 3297–3305. [21] J. Trykowska Konc, E. Hejchman, H. Kruszewska, I. Wolska, D. Maciejewska, Eur. J. Med. Chem. 46 (2011) 2252–2263. [22] European Pharmacopoeia 5.0. Microbial assay of antibiotics 2.7.2, 2013. 188.
[23] Clinical and Laboratory Standards Institute, Approved standard M7–A9, Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically, 9th ed., 2012. [24] M. Boncler, M. Rozalski, U. Krajewska, A. Podsedek, C. Watala, J. Pharmacol. Toxicol. Methods 69 (2014) 9–16. [25] J. Palomar, J.L.G. De Paz, J. Catalán, Chem. Phys. 246 (1999) 167–208. [26] J. Seixas de Melo, G. Quinteiro, J. Pina, S. Breda, R. Fausto, J. Mol. Struct. 565 (2001) 59–67. [27] A. Drzewiecka, A.E. Koziol, M.T. Klepka, A. Wolska, H. Przybylinska, S.B. JimenezPulido, K. Ostrowska, M. Struga, J. Kossakowski, T. Lis, Polyhedron 43 (2012) 71–80. [28] H.M. Ali, S.N.A. Halim, S.W. Ng, Acta Crystallogr. E 61 (2005) m1429–m1430. [29] S. Rayati, M.R. Yaftian, H.R. Khavasi, Acta Crystallogr. E 63 (2007) m898–m899. [30] G. Mao, X. Nong, S.H. Zhang, Acta Crystallogr. E 65 (2009) m1208–m1208. [31] K. Hatada, K. Hayakawa, M. Benfatto, C.R. Natoli, Phys. Rev. B 76 (2007) 0601202(R)4 pp. [32] K. Hatada, K. Hayakawa, M. Benfatto, C.R. Natoli, J. Phys. Condens. Matter 22 (2010) 185501–185524 - 24 pp. [33] L. Hedin, B.I. Lundqvist, J. Phys. C Solid State Phys. 4 (1971) 2064–2083. [34] M. Benfatto, S. Della Longa, J. Synchrotron Radiat. 8 (2001) 1087–1094. [35] M. Benfatto, S. Della Longa, C.R. Natoli, J. Synchrotron Radiat. 10 (2003) 51–53. [36] A. Kucers, The use of antibiotics. A clinical review of antibacterial, antifungal and antiviral drugs, Butterworth-heinemann, Oxford, 1997. [37] C. Santini, M. Pellei, V. Gandin, M. Porchia, F. Tisato, C. Marzano, Chem. Rev. 114 (2014) 815–862.
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry journal homepage: www.elsevier.com/locate/jinorgbio
Synthesis, structural studies and biological activity of new Cu(II) complexes with acetyl derivatives of 7-hydroxy-4-methylcoumarin Marcin T. Klepka a,⁎, Aleksandra Drzewiecka-Antonik a, Anna Wolskaa, Paweł Rejmak a, Kinga Ostrowskab, Elżbieta Hejchman b, Hanna Kruszewskac, Agnieszka Czajkowskad, Izabela Młynarczuk-Biały d, Wiesława Ference a
Institute of Physics, Polish Academy of Sciences, Al. Lotnikow 32/46 PL-02668, Warsaw, Poland Faculty of Pharmacy, Medical University of Warsaw, Banacha 1, PL-02097, Warsaw, Poland c National Medicines Institute, Chelmska 30/34, PL-00725, Warsaw, Poland d Centre of Biostructure Research, Medical University of Warsaw, Chalubinskiego 5, PL-02004, Warsaw, Poland e Faculty of Chemistry, Maria Curie-Sklodowska University, Sq. Maria Curie-Sklodowska 2, PL-20031, Lublin, Poland b
a r t i c l e
i n f o
Article history: Received 5 November 2014 Received in revised form 14 January 2015 Accepted 15 January 2015 Available online 23 January 2015 Keywords: Cu(II) complex Coumarin derivative 7-hydroxycoumarin XAFS DFT Electrochemical synthesis
a b s t r a c t The new Cu(II) complexes with 6-acetyl-7-hydroxy-4-methylcoumarin (HL1) and 8-acetyl-7-hydroxy4-methylcoumarin (HL2) have been obtained by the electrochemical method. The density functional theory calculations and X-ray absorption spectroscopy techniques have been used to geometrically describe a series of new compounds. The studies have been focused on the coordination mode of the hydroxy ligands to the metallic centre. The complexes, Cu(HL1) 2 and Cu(HL2) 2 ⋅ 0.5H 2O, have flat square geometry with oxygen atoms in the first coordination sphere. Two bidentate anionic coumarins are bonded to the metal cation via the acetyl and deprotonated hydroxyl O atoms. Biological activity, including microbiological and cytotoxic, has been evaluated and found to be enhanced in comparison with the parent ligands. Moreover, the Cu(II) complex with 8-acetyl-7-hydroxy-4-methylcoumarin shows similar antifungal activity as commercially used fluconazole. © 2015 Elsevier Inc. All rights reserved.
1. Introduction The natural as well as synthetic coumarins, therein hydroxycoumarins, have a large spectrum of biological activity. Such derivatives proved usefulness as anti-coagulants [1], antibacterial agents [2], antifungal agents [3], biological inhibitors [4], chemotherapeutics [5] and as bio-analytical reagents [6]. It has been found out that coordination of metal ions to therapeutic agents (such as simple coumarins) can improve their efficacy and accelerate bioactivity. In many cases such metal complexes are more potent and less toxic compared to the parent drug. Therefore, among others, also biologically active metal complexes of coumarin based ligands are being widely investigated. Among them, triorganotin(IV) derivatives of umbelliferone (7hydroxycoumarin) have shown good antimicrobial activity against Staphylococcus aureus, Bacillus subtilis, Candida albicans and Microsporum gypseum, and this activity was slightly enhanced upon adduct formation with 1,10-phenanthroline. Creaven et al. have investigated the antimicrobial activity of a number of coumarin complexes ⁎ Corresponding author. E-mail address: [email protected] (M.T. Klepka).
http://dx.doi.org/10.1016/j.jinorgbio.2015.01.006 0162-0134/© 2015 Elsevier Inc. All rights reserved.
with silver(I), copper(II) and manganese(II) ions. For example, the Cu (II) complexes exhibit antifungal activity against a clinical strain of C. albicans comparable to that of the commercially available antifungal drugs, i.e. ketoconazole and Amphotericin B [7]. This paper is focused on the synthesis and characterization of copper complexes of two hydroxyligands: HL1 and HL2 (Fig. 1). These ligands have acetyl group attached at two different positions, C6 and C8, to the rigid coumarin ring. The electrochemical method was applied for the synthesis of the complexes. In order to determine the geometry of metal-ligand interactions X-ray crystallography is usually used. However, this technique requires compound in the crystal form, which is not always possible to obtain. Since that was the case for the investigated complexes, the X-ray absorption spectroscopy (XAS) was applied. The great advantage of XAS is that it can be used for crystal as well as amorphous materials at different states: solid, liquid or gaseous. XAS provides information about the local atomic order, coordination number, kind of atoms, oxidation state, relative disorder and even angles between central atom and near neighbours [8–10]. Data obtained from XAS combined with the density functional theory (DFT) calculations allowed building models of the complexes. Additionally to
M.T. Klepka et al. / Journal of Inorganic Biochemistry 145 (2015) 94–100
R1
CH3
5
6
4
2
7
HO 3
R1
3
8
R2
O 1
O 2
R2
HL1 C(=O)CH3 H HL2 H
C(=O)CH3
95
reference compounds (two copper oxides, Cu2O and CuO) were also measured. The quantitative analysis of EXAFS spectra of compounds 1–2 was performed as follows: k2 weighted χ(k) data were Fourier transformed in the k range 2.4–11 for 1 and 2.5–10 for 2. The Fourier's back transformation ranges were from 1 to 2 Å for both complexes. For the XAS analysis, the Athena and Artemis programs included in the IFEFFIT package [14] were applied. 2.5. DFT calculations
Fig. 1. Molecular structure of ligands.
geometrical characterization of complexes also the biological activity tests, including microbiological and cytotoxic studies, were performed. 2. Experimental 2.1. Instrumentation The elemental analysis was performed using the Perkin-Elmer 2400 CHN elemental analyzers. The thermal stability and decomposition process of complexes were studied in air using the Setsys 16/18 (Setaram) TG, DTG and DSC instruments. The experiments were carried under air flow in the temperature range of 293–823 K at a heating rate of 5 K/min. The initial masses of samples of complexes used for measurements changed from 4.59 mg to 4.56 mg for 1 and 2.01 mg to 1.96 mg for 2. Samples of these compounds were heated in Al2O3 crucibles. The IR spectra were gathered between 400 and 4000 cm−1 in the ATR mode on a Thermo Scientific Nicolet iS5 spectrometer. 2.2. Synthesis of ligands The initial reagents were bought from Sigma-Aldrich. The synthesis of 6-acetyl-7-hydroxy-4-methylcoumarin (HL1) and 8-acetyl-7-hydroxy-4-methylcoumarin (HL2) was performed as described in literature [11,12]. 2.3. Synthesis of complexes The copper complexes, 1–2, were obtained by an electrochemical synthesis [13]. For the complexation reaction the 7-hydroxycoumarins, HL1 and HL2, were used as ligands. The copper plate was used as a metal anode and a platinum wire was the cathode. A 96% ethanol solution of HL1 and HL2 containing about 80 mg of tetrabutylammonium perchlorate (electrolyte) was electrolyzed for 2 h (see Table 1). After filtration of the resulting solution, the powder product was purified by washing thoroughly with water. 2.4. X-ray absorption spectroscopy The K-edges of Cu for the complexes 1 and 2 were measured at the beamline I811 at MAX-lab in Lund, Sweden. Both XANES (X-ray absorption near edge structure) and EXAFS (extended X-ray absorption fine structure) spectra were recorded in the transmission mode. To confirm the oxidation state of metal in the complexes, the XANES spectra of the
The DFT calculations were performed using Turbomole 6.5 code [15]. Relying on reasonable assumption, that weak interactions keeping molecular solids together should not affect strongly the properties of molecules, structural models consisting of single molecule were used. Such models were successfully applied in elucidation of structural and spectroscopic properties of metal complexes with coumarin derivatives [16]. Geometry optimization was carried out using gradient Perdew– Burke–Ernzerhof (PBE) [17] and hybrid PBE0 (i.e. PBE with 25% of Hartree–Fock exchange) [18] exchange-correlation functionals with def2-TZVPP basis set [19,20]. The convergence criteria for total energy and gradient norm were 10−6 Eh and 10−4 Eh/bohr, respectively. Harmonic vibrational analysis was performed at PBE level to verify, if true minimum of energy was achieved and support quantitative analysis of FT-IR spectra. 2.6. Biological activity In order to check the biological activity of complexes 1–2 microbiological and cytotoxic tests were carried out and afterwards compared with the parent ligands (HL1–HL2) described before [21]. 2.6.1. Microbiological assays The following microbial strains were chosen from American Type Culture Collection (ATCC): bacteria Gram-positive strains: Micrococcus luteus (9341, 10240), B. subtilis 6633, Bacillus cereus 11778, S. aureus (6538, 6538P), bacteria Gram-negative strains: Escherichia coli (8739, 10536), Pseudomonas aeruginosa (15442, 9027), fungi (yeast strains): C. albicans (10231, 2091), Candida parapsilosis 22019, Zygosaccharomyces rouxii 28253, Saccharomyces cerevisiae 9763 and a mould strain Aspergillus brasiliensis 16404. The MRSA hospital strain 573/12 (methicillin resistant S. aureus) isolated from peripheral blood was obtained from MUW Museum Collection. According to disc-diffusion Kirby-Bauer method, the 573/12 strain is resistant to: penicillin, erythromycin, amikacin, clindamycin, and ciprofloxacin. 2.6.2. Antimicrobial activity — preliminary test In the preliminary tests antimicrobial activity was determined by a modified cylinder-plate method [22]. The inhibition of bacterial growth was observed as a halo around the cylinder containing the tested compound. The size of inhibition zone reflected an antimicrobial activity of the examined compounds. 2.6.3. Minimum inhibitory concentration (MIC) For the compound which showed some activity against any of the tested strains an MIC was determined based on M7-A6 method [23]. The lowest concentration of tested compound, which totally inhibited
Table 1 Experimental conditions for the synthesis of complexes 1–2. Complex
Amount of solved copper [g]
Ligand
Amount of HL [g]
Voltage⁎ [V]
Colour of complex
Elemental analysis (C, H calc⁎⁎/found %)
1 2
0.0415 0.0515
HL1 HL2
0.0814 0.0814
29 18
green green
57.89, 3.64/57.22, 3.72 56.86, 3.85/56.78, 3.75
⁎ Voltage to produce a current of 10 mA. ⁎⁎ Anal. calc. for Cu(HL1)2 and Cu(HL2)2∙0.5 H2O.
96
M.T. Klepka et al. / Journal of Inorganic Biochemistry 145 (2015) 94–100
Table 2 The characteristic IR experimental frequencies ν (cm-1) of carbonyl groups in ligands (HL1–HL2) and complexes (1–2), along with calculated harmonic DFT/PBE frequencies ω(cm-1).
HL1 HL2 1 2
ν(C = O)
ω(C = O)DFT/PBE
1725coumarin, (1550–1675)acetyl 1730coumarin, (1550–1650)acetyl 1737coumarin, (1550–1625)acetyl 1730coumarin, (1550–1640)acetyl
1759coumarin, (1565–1635)acetyl 1761coumarin, (1550–1625)acetyl 1753coumarin, (1550–1590)acetyl 1760coumarin, (1535–1585)acetyl
growth of the examined strain, was taken as MIC value. The negative control was performed with the solution of 10% DMSO (in 0.08 M phosphate buffer, pH 7.0). All tests were performed in triplicate. 2.7. Cytotoxicity and antiviral assays 2.7.1. Reagents Synthesized substances were dissolved in DMSO (Sigma-Aldrich) at 20 mM and kept in 4 °C prior to use. 2.7.2. Cell culture The human prostate cancer cells DU145 and NIH3T3 mouse fibroblasts, both from ATCC, were maintained in humidified atmosphere containing 5% CO2 at 37 °C. DU145 and NIH3T3 cells were cultured in RPMI and DMEM media, respectively, both supplemented with stable glutamate and with 10% foetal calf serum (FCS) as well as with antibiotic antimycotic (Sigma-Aldrich). Cells were passaged every 2–3 days. 2.7.3. Cytotoxicity assays For the analysis of cytostatic/cytotoxic effects induced by new compounds, resarium based PrestoBlue assay (Invitrogen) was performed according to manufacturer's instructions [24].
Fig. 3. Fourier transformed experimental EXAFS oscillations for complexes 1 (panel a) and 2 (panel b) (black line) and fitted first coordination sphere (red dot line). As an inset EXAFS oscillations together with fitted first coordination sphere are presented.
3. Results and discussion Both electrochemically synthesized compounds were obtained as green powders. Complex 1 was found to be anhydrous, whereas 2 contains approximately half of the water molecule per complex which was confirmed by thermal analysis. The mass loss of dehydration process calculated from thermogravimetric (TG) curve (323–448 K) is equal to 2.4% for 2 which indicates that the complex 2 released approximately half of the water molecule in one step. The dehydration process is
connected with the endothermic effect (differential scanning calorimetry (DSC) curve). In the second step of decomposition (in the range of 473–823 K) these two complexes eliminate gradually the organic ligands with the exothermic effect on DSC curve and ultimately form the CuO. For both cases, the elemental analysis determined metal to ligand ratio as 1:2 [Cu(HL1)2 (1) and Cu(HL2)2∙0.5H2O (2)]. 3.1. FTIR spectroscopy The complexation reactions were monitored by infrared spectroscopy. The IR analysis was focused on the stretching frequencies of carbonyl groups, which contribute to the most intense bands and are quite well separated from other molecular motions. These characteristic bands observed for the parent ligands and their complexes are presented in Table 2. The comparison of the IR spectra of the complexes (1–2) with the spectra of their ligands (HL1–HL2) reveals that: (i) the peaks corresponding to the (C = O)coumarin stretching are intact or even blue shifted in complex, which precludes their bonding to Cu, (ii) the peaks corresponding to the (C = O) acetyl stretching are red shifted, which indicates the coordination to Cu Table 3 EXAFS parameters of Cu(II) complexes: cation—O distance—R, number of atoms in a coordination sphere—N, Debye–Waller factor—σ2 and R-factor.
Fig. 2. XANES Cu K-edge spectra of the metal complexes 1 and 2 together with reference oxides Cu2O and CuO.
Complex
Bond
R [Å]
N
σ2 [Å2]
R-factor
1 2
Cu–O Cu–O
1.93(4) 1.94(1)
3.9(2) 4.0(1)
0.004(1) 0.004(1)
0.001 0.001
M.T. Klepka et al. / Journal of Inorganic Biochemistry 145 (2015) 94–100
97
Fig. 4. The relative DFT/PBE0 energy of conformers of complex 2 with respect to torsion angle defined by 4 chelating O atoms. 0 and 180° torsions correspond to C2h and C2v structures, respectively.
through the carbonyl O atom of acetyl group. (C = O)acetyl vibrations generally correspond to lower frequencies due to the conjugation to phenol ring and, for pristine ligand, due to the participation in the intramolecular hydrogen bond. DFT calculations revealed also that (C = O)acetyl vibration component is present in several normal modes of molecules, thus the range of infrared frequencies with possible (C = O)acetyl mode contribution is given in Table 2. The assignment of acetyl and coumarin (C = O) modes is consistent with the reported carbonyl stretching frequencies in 2-acetylphenol (1646 cm− 1 and below) [25] and in 2-pyrone (1732 cm− 1) [26]. The fingerprint regions have also been analysed. The bands present in the spectra of the ligands are observed at different positions than the respective bands in the IR spectra of their metal complexes. These data are analogous to the spectra of similar series of Cu(II) complexes in which two bidentate ligands are coordinated to the metal centre via O atoms from the acetyl and hydroxyl groups substituted on the rigid benzo[b]furan system [27]. The main difference between these two series of complexes is the lack of a broad band at 3300–3600 cm1 region in the IR spectra of coumarin derivatives (1–2). This band (observed for benzofuran complexes of copper) is responsible for OH stretching in water molecules. Its absence agrees with the data obtained from thermal analysis.
Fig. 5. DFT optimized molecular structures of complexes 1 (panel a) and 2 (panel b).
energy positions of the Cu K-edge, evaluated by comparison of position of the first derivative maximum, in compounds 1–2 and CuO are similar,
3.2. XAFS analysis of complexes XANES, due its sensitivity to the chemistry of the absorbing atom, was used to determine the oxidation state and local geometry around the Cu cation in the investigated compounds. The complexes were synthesized by an electrochemical method (from metallic copper), therefore, it is important to verify the copper oxidation state in new compounds. Thus, the energy positions of copper edges in the complexes were compared with the reference oxides: Cu2O and CuO. The
Table 4 Selected calculated bond distances Cu–O/C (Å) and valence angles Cu–O–C (°) of complexes 1–2. PBE
Cu–O3 Cu–Oacetyl Cu–C7 Cu–Cacetyl Cu–C6/C8 Cu–O3–C7 Cu–Oacetyl–Cacetyl
PBE0
1
2
1
2
1.908 1.962 2.893 2.925 3.325 128.25 128.77
1.902 1.943 2.899 2.944 3.362 129.44 131.91
1.881 1.944 2.865 2.900 3.297 128.98 129.36
1.877 1.925 2.870 2.918 3.333 130.04 132.50
Fig. 6. Experimental and calculated Cu K-edge XANES spectra of complexes 1 (panel a) and 2 (panel b).
98
M.T. Klepka et al. / Journal of Inorganic Biochemistry 145 (2015) 94–100
Table 5 Antibacterial activity (in mg/mL) of tested complexes. Strains Micrococcus luteus ATCC 9341 Bacillus cereus ATCC 11178 Bacillus subtilis ATCC 6633 Staphylococcus epidermidis ATCC 12228 Staphylococcus aureus ATCC 6538 Staphylococcus aureus ATCC 6538P Enterococcus hirae ATCC 10541 Escherichia coli ATCC 8739 Pseudomonas aeruginosa ATCC 9027 Pseudomonas aeruginosa ATCC 15442 Staphylococcus aureus MRSA 573/12
Table 6 Antifungal activity (in mg/mL) of tested compounds.
1
2 1.25
0.01875
0.625
0.0375
2.5
0.075
2.5
0.15
N10
0.625
N10
0.3
N10
0.15
N10
N10
N10
N10
N10
N10
N10
Strains
1
2
Candida albicans ATCC 10231 Candida albicans ATCC 2091 Candida parapsilosis ATCC 22019 Aspergillus brasiliensis ATCC 16404 Zygosacharomyces rouxi ATCC 28253 Saccharomyces cerevisiae ATCC 9763
N10
0.075
N10
0.0375
N10
0.075
N10
0.15
N10
0.01875
N10
0.0375
1.93 and 1.90 Å at PBE and PBE0 levels, respectively, which is in a good agreement with EXAFS data (see Table 3). It should be stressed that using nonplanar structures as initial points in geometry optimization resulted in convergence to nearly planar local minima of energy. Conformational analysis (Fig. 4) proves that structures with pseudotetrahedral Cu(II) coordination are significantly less stable in respect to flat species. Formally, Cu(II) ion can be coordinated by two 2-acetylphenol type ligand in two ways, with resulting complex with the inversion centre or without. For planar molecules these both possible isomers would correspond to C2h and C2v point groups, respectively. The DFT studies showed that both isomers have the same energetic stability. Taking into account that X-ray crystal structure analysis on similar Cu(II) complexes reported centrosymmetric type of coordination [27–30], such type of coordination was considered as the most probable and used as an initial model for XANES data analysis (see Fig. 5). The issue of apparent preference of centrosymmetric isomer in nature is under further theoretical investigation by our group.
0.3
which implies that analysed complexes contain mostly copper in the +2 oxidation state (Fig. 2). Moreover, a similar local density of unoccupied Cu 4p states, represented by near edge absorption spectra, for both complexes is observed. This suggests a similar metal coordination environment for analysed complexes of hydroxycoumarin. The fitting of EXAFS spectra was performed in an analogous way for compounds 1 and 2 (Fig. 3). For both complexes four O atoms were found in the first coordination sphere (Table 3). The Cu–O bond lengths are around 1.94 Å. Any additional copper ion was not identified in the first nor farther coordination spheres, which proves that the multinuclear copper complexes are not formed.
3.4. XANES — full potential multiple scattering 3.3. DFT calculations
Structural models from DFT calculations were used as the input structures for full potential multiple scattering (FPMS) XANES calculations. Several attempts were made to calculate Cu K-edge XANES spectra for these DFT structures. In the first step standard approach using muffin tin potential was applied. However, this type of potential does not work properly for the flat geometry, which is the case for the considered Cu(II) complexes. In order to perform reliable calculations the FPMS code developed by Hatada and coworkers [31,32] was used. The real part of Hedin–Lundqvist potential (HL) was employed [33]. The energy dependent broadening due to the inelastic scattering was taken into account for FPMS spectra by MXAN program [34,35]. Comparison of the calculated and experimental spectra for complexes 1 and 2 is
The results of FTIR, XAFS spectroscopy, elemental analysis of compounds 1 and 2 as well as previously reported structures of Cu(II) complexes with ligands containing 2-acetylphenol ring [27] pointed out towards the structural models with Cu(II) being coordinated by two coumarin ligands in planar, bidentate mode. The proposed molecules of complexes 1–2 consist of two HL anions chelating to the metal through the O atom from deprotonated hydroxyl group and O atom from the acetyl substituent. The results of geometry optimization of such models are shown in Fig. 4 and summarized in Table 4. Calculated mean Cu–O average distances, for both analysed complexes, are about
Table 7 Comparison of antibacterial activities (MIC values in mg/mL) between ligands and complexes. Strains
HL1
1
HL2
2
Micrococcus luteus ATCC 9341 Bacillus cereus ATCC 11178 Bacillus subtilis ATCC 6633 Staphylococcus aureus ATCC 6538 Staphylococcus aureus ATCC 6538P Escherichia coli ATCC 8739
45.8
1.25
68.7
0.01875
4.6
0.625
22.9
0.0375
22.9
2.5
45.8
0.075
45.8
N10
68.7
0.625
–
N10
–
0.3
–
N10
–
N10
M.T. Klepka et al. / Journal of Inorganic Biochemistry 145 (2015) 94–100
99
to the our best knowledge, no such activity was reported for complexes with Cu(II) being chelated by 2-acetylphenol type ligand prior to our previous work [27]. Proposed mechanisms for antitumour activities of Cu complexes include DNA binding or topoisomerase inhibition. Compound 2 induces similar cell growth inhibition, although NIH3T3 cells are slightly more susceptible than DU145 cells. 4. Conclusions
Fig. 7. Cytostatic/cytotoxic activity of complexes 1 (panel a) and 2 (panel b) against DU145 and NIH3T3.
shown in Fig. 6. It is clearly seen that shape of the XANES spectra is reproduced which points out that considered model reproduces the Cu(II) coordination geometry of the electrochemically synthesized complexes. 3.5. Antimicrobial activity The MIC values counted on mmol of examined compound are presented in Tables 5 and 6. Complexes inhibited growth of Grampositive bacterial strains. Complex 2 was also active against tested fungal strains. The MIC concentration of 2 against fungal strains was between 0.0187 and 0.075 mg/mL, similar to MIC of fluconazole which is between 0.0125 and 0.256 mg/mL for different fungal strains [36]. Activity of both complexes increased in comparison witch parent ligands, see Table 7 [21]. 3.6. Antitumour activity For estimation of tumour-cell selectivity of examined compounds, viability assays were performed in tumourigenic DU145 cell line and in non-tumourigenic one, NIH3T3. Cells were treated as described in the experimental part of the paper. Both synthesized agents induced dose-dependent cytostatic/cytotoxic effects in examined cells. Prostate tumour cells are slightly more susceptible towards complex 1 than the non-tumourigenic mouse fibroblasts, this effect is most pronounced at 200 and 500 μM concentrations (Fig. 7). Cytotoxic activity was reported for many Cu(II) κ2-O, O complexes with planar environment of central ion [37]. However, up
Two new copper complexes with the acetyl derivatives of hydroxycoumarin as the O-donor ligands have been electrochemically synthesized, structurally described and biologically tested. Initial characterization including elemental and thermal analysis confirmed that complexes were formed with metal to ligand ratio as 1:2. The FTIR spectroscopy indicated the coordination to Cu(II) cation through the carbonyl O atom of the acetyl group. These information were used as starting point in EXAFS analysis. The EXAFS fitting parameters, correlated with DFT calculations indicated that centrosymmetric type of coordination, with four O atoms at average distance 1.94 Å, is formed. The XANES data confirmed that two bidentate ligands are coordinated to the metal centre via O atoms from the acetyl and hydroxyl groups substituted on the coumarin system. The first coordination sphere consists of four O-atoms and there is an indication that the central atom has a slightly distorted planar square geometry. Biological activity against Gram positive and negative bacteria: M. luteus, B. cereus, B. subtilis, and S. aureus, has significantly increased in comparison to the parent ligands (see Table 7). Furthermore, compound 2 showed activity against fungal strains, C. albicans, C. parapsilosis, A. brasiliensis, Zygosacharomyces rouxi, and S. cerevisiae. Complex 1 did not show activity against these strains (see Table 6). On the other hand, comparing viability tests, the compound 1 is more selective for tumour cells than 2. Complex with acetyl group at C6 position (1) shows activity against Gram-positive, Gram-negative bacteria and has better antitumour properties than 2. The compound containing the acetyl group at C8 position (2) is additionally active against fungal strains but has weaker antitumour properties. Observed differences in activity indicated that position of acetyl group modulates complex biological properties. Farther studies are planned on complexes with acetyl group at different positions. From application point of view it is required to have detailed geometric data, which are necessary to understand the mode of action of such complexes. In this paper it was demonstrated that XAS and computational techniques enable structural characterization in lack of crystallographic data. Acknowledgements Experimental research was funded from the Polish National Science Centre (Grant No. UMO-2012/07/D/ST5/02251). The synchrotron experiment was partially supported by the Baltic Science Link project coordinated by the Swedish Research Council, VR. Authors would like to acknowledge Dr. K. Hatada for valuable discussion and help in starting FPMS calculations. This research was supported in part by PL-Grid Infrastructure. P. Rejmak acknowledges EAgLE project no. 316014 for the financial support. References [1] J.W. Suttie, Clin. Cardiol. 13 (VI) (1990) 16–18. [2] A.H. Bedair, N.A. El-Hady, M.S. Abd El-Latif, A.H. Fakery, A.M. El-Agrody, Il Farmaco 55 (2000) 708–714. [3] T. Patonay, G. Litkei, R. Bognar, J. Eredi, C. Miszti, Pharmazie 39 (1984) 86–91. [4] C. Gnerre, M. Catto, F. Leonetti, P. Weber, P.A. Carrupt, C. Altomare, A. Carotti, B.J. Testa, Med. Chem. 43 (2000) 4747–4758. [5] D.A. Egan, P. James, D. Cooke, R. O'Kennedy, Cancer Lett. 118 (1997) 201–211. [6] M. Jiménez, J.J. Mateo, R. Mateo, J. Chromatogr. A 870 (2000) 473–481.
100
M.T. Klepka et al. / Journal of Inorganic Biochemistry 145 (2015) 94–100
[7] B.S. Creaven, D.A. Egan, D. Karcz, K. Kavanagh, M. McCann, M. Mahon, A. Noble, B. Thati, M. Walsh, J. Inorg. Biochem. 103 (2009) 1196–1203. [8] M.T. Klepka, A. Drzewiecka, A. Wolska, W. Ferenc, Chem. Phys. Lett. 553 (2012) 59–63. [9] M.T. Klepka, K. Lawniczak-Jablonska, A. Wolska, A. Slawska-Waniewska, C.A. Rodrigues, A. Debrassi, C. Bordini, Chem. Phys. Lett. 501 (2011) 523. [10] A. Drzewiecka, A.E. Koziol, M.T. Klepka, A. Wolska, S.B. Jimenez-Pulido, T. Lis, K. Ostrowska, M. Struga, Chem. Phys. Lett. 559 (2013) 41–45. [11] V.S. Moskvina, A.A. Ishchenko, V.P. Khilya, Chem. Nat. Compd. 48 (2012) 751–756. [12] A. Kumar, R.A. Maurya, S. Sharma, P. Ahmad, A.B. Singh, G. Bhatia, A.K. Srivastava, Bioorg. Med. Chem. Lett. 19 (2009) 6447–6451. [13] C. Oldham, D.G. Tuck, J. Chem. Educ. 59 (1982) 420–421. [14] B. Ravel, M. Newville, J. Synchrotron Radiat. 12 (2005) 537–541. [15] R. Ahlrichs, M. Bär, M. Häser, H. Horn, C. Kölmel, Chem. Phys. Lett. 162 (1989) 165–169. [16] I. Georgieva, Tz Mihaylov, N. Trendafilova, J. Inorg. Biochem. 135 (2014) 100–112. [17] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77 (1996) 3865–3868. [18] J.P. Perdew, M. Ernzerhof, K. Burke, J. Chem. Phys. 105 (1996) 9982–9985. [19] F. Weigend, M. Häser, H. Patzelt, R. Ahlrichs, Chem. Phys. Lett. 294 (1998) 143–152. [20] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 7 (2005) 3297–3305. [21] J. Trykowska Konc, E. Hejchman, H. Kruszewska, I. Wolska, D. Maciejewska, Eur. J. Med. Chem. 46 (2011) 2252–2263. [22] European Pharmacopoeia 5.0. Microbial assay of antibiotics 2.7.2, 2013. 188.
[23] Clinical and Laboratory Standards Institute, Approved standard M7–A9, Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically, 9th ed., 2012. [24] M. Boncler, M. Rozalski, U. Krajewska, A. Podsedek, C. Watala, J. Pharmacol. Toxicol. Methods 69 (2014) 9–16. [25] J. Palomar, J.L.G. De Paz, J. Catalán, Chem. Phys. 246 (1999) 167–208. [26] J. Seixas de Melo, G. Quinteiro, J. Pina, S. Breda, R. Fausto, J. Mol. Struct. 565 (2001) 59–67. [27] A. Drzewiecka, A.E. Koziol, M.T. Klepka, A. Wolska, H. Przybylinska, S.B. JimenezPulido, K. Ostrowska, M. Struga, J. Kossakowski, T. Lis, Polyhedron 43 (2012) 71–80. [28] H.M. Ali, S.N.A. Halim, S.W. Ng, Acta Crystallogr. E 61 (2005) m1429–m1430. [29] S. Rayati, M.R. Yaftian, H.R. Khavasi, Acta Crystallogr. E 63 (2007) m898–m899. [30] G. Mao, X. Nong, S.H. Zhang, Acta Crystallogr. E 65 (2009) m1208–m1208. [31] K. Hatada, K. Hayakawa, M. Benfatto, C.R. Natoli, Phys. Rev. B 76 (2007) 0601202(R)4 pp. [32] K. Hatada, K. Hayakawa, M. Benfatto, C.R. Natoli, J. Phys. Condens. Matter 22 (2010) 185501–185524 - 24 pp. [33] L. Hedin, B.I. Lundqvist, J. Phys. C Solid State Phys. 4 (1971) 2064–2083. [34] M. Benfatto, S. Della Longa, J. Synchrotron Radiat. 8 (2001) 1087–1094. [35] M. Benfatto, S. Della Longa, C.R. Natoli, J. Synchrotron Radiat. 10 (2003) 51–53. [36] A. Kucers, The use of antibiotics. A clinical review of antibacterial, antifungal and antiviral drugs, Butterworth-heinemann, Oxford, 1997. [37] C. Santini, M. Pellei, V. Gandin, M. Porchia, F. Tisato, C. Marzano, Chem. Rev. 114 (2014) 815–862.
