DNA interaction studies of pyrazolone- and diimine-incorporated Mn(II), Co(II), Ni(II), Cu(II), and Zn(II) complexes: synthesis, spectroscopic characterization, and antimicrobial study

Abstract

Mn(II), Co(II), Ni(II), Cu(II), and Zn(II) mixed-ligand complexes were synthesized using 4-[(furan-2-ylmethylene)amino]-1,2-dihydro-1,5-dimethyl-2-phenyl-3H-pyrazol-3-one as the main ligand and 1,10-phenanthroline/2,2′-bipyridine as co-ligand(s). They were characterized by the usual analytical and spectral techniques. The data obtained reveal that the complexes adopt an octahedral geometry around the central metal ions. These complexes were found to be better antimicrobial agents than the free ligands. The DNA (CT) binding properties of the complexes were explored by UV–Vis, viscosity measurements, cyclic voltammetry, differential pulse voltammetry, and molecular docking studies. Binding constants for the above complexes were found to be on the order of 10⁴, indicating that most of the synthesized complexes are partial intercalators. The DNA-cleavage activities of the complexes were also assessed using supercoiled pUC19 DNA and gel electrophoresis, and the results revealed that the hydroxyl radical is likely to be the reactive species responsible for the cleavage of pUC19 DNA by the synthesized complexes.

Graphical abstract

.

Introduction

Schiff bases are very attractive and have achieved recognition for their roles in biological processes such as mutagenesis and carcinogenesis. In addition to this, in recent years, lots of research have been done on the coordination chemistry of transition metal complexes with Schiff base ligands in order to mimic the physical and chemical behavior seen in biological processes [1, 2]. The interactions of transition metal complexes with DNA have been the focus of recent investigations. Metal complexes that exhibit interactions with DNA have been studied with the aim of developing probes for nucleic acid structures and chemotherapy agents [3]. By changing the ligand environment, it is possible to study the DNA binding and cleavage abilities of metal complexes. Such studies are also important for determining the mechanism of metal ion toxicity [4, 5].

The transition metal complexes of 1,2-dihydro-1,5-dimethyl-2-phenyl-3H-pyrazol-3-one (4-aminoantipyrine) and its derivatives have been extensively examined due to their wide range of applications in biological, analytical, and therapeutic fields [6]. Furfural is an important nontoxic compound. It is used as a fungicide and nematicide [7]. Hence, we elected to use this heterocyclic compound when attempting to design a Schiff base ligand with 4-aminoantipyrine. This ligand is useful as it contains aromatic rings that are able to stack the bases in DNA, oxygen and nitrogen atoms which can establish hydrogen bonds with DNA, and there is the possibility that the metal complexes display a positive charge in order to interact electrostatically with the phosphate groups of DNA. In general, metal complexes that are capable of abstracting ribosyl hydrogen are expected to oxidize guanine or other nucleobases [8–10]. Current efforts are being made to design transition metal complexes that can be used as chemical nucleases in plasmid nicking or in direct strand scission. The practical use of transition metal complexes as chemical nucleases has also been documented [11, 12]. DNA binding, cleavage, and the antimicrobial activities of Mn(II), Co(II), Ni(II), Cu(II), and Zn(II) complexes that are derived from furfural with 4-aminoantipyrine and 1,10-phenanthroline (phen)/2,2′-bipyridine (bpy) are of interest to us in our studies, as discussed in this present paper.

Results and discussion

Synthetic routes for the preparation of different compounds are depicted in Scheme 1. The ligand L and the complexes were found to be stable in air. L is soluble in common organic solvents but the complexes are soluble only in DMF and DMSO. The elemental analysis data on L and the complexes are in good agreement with the presented formulae. Physicochemical and spectroscopic studies indicate the formation of monomeric complexes of stoichiometry [M(L)(phen/bpy)₂]Cl₂ [13].

Scheme 1

 

Infrared spectra

The IR spectra of the complexes were compared with the ligands in order to ascertain the changes that might have taken place. Although the Schiff base L and bpy/phen mixed-ligand complexes contain different types of C=N bonds arising from the bpy/phen ring and azomethine linkage, they are not well resolved in the spectra of the complexes [14, 15]. Furthermore, in the spectrum of L, although the azomethine –CH=N band is superimposed on that of the C=O group of the imide, it is only a weak band. The peaks that appear in the range 1,639–1,600 cm⁻¹ may be due to stretching of the azomethine –CH=N group, whereas the bands in the range 1,552–1541 cm⁻¹ could be due to the ν(C=N) stretches of bpy/phen. The ν(C=O) band from L at 1655 cm⁻¹ is shifted in frequency by ~10 cm⁻¹ on complexation [16, 17]. A shift in the frequency of the band due to HC=N cannot be observed because it is not well resolved in the infrared spectra of the ligand and the complexes. However, the HC=N present in L is also involved in the formation of a bond with the metal atom during complexation, as clearly confirmed by the NMR data. The change in the frequency of the C=N band by 10 cm⁻¹ indicates that bpy/phen rings are involved in the formation of the complex. Moreover, after coordination, the bands observed at around 881–892 cm⁻¹ in bpy, 847–871 cm⁻¹ in phen, 735–750 cm⁻¹ (C–H_def), and 1,471–1,480 cm⁻¹ and 1,315–1,325 cm⁻¹ (C–N stretching bands) prove that the bpy and phen rings are coordinated to the metal ion through nitrogen. The new bands observed in the regions 418–442 and 478–490 cm⁻¹ for all of the complexes were assigned to M–N and M–O stretching vibrations, respectively, confirming the metal coordination [18, 19].

Molar conductivity

The high molar conductivities of the complexes in 10⁻³ M DMF solution reveal that they are electrolytic in nature, implying that the chloride anion is not coordinated to the central metal ion. The presence of a counter chloride ion was confirmed by Volhard’s test.

Magnetic moments and electronic spectra

The free Schiff base ligand L exhibits two intense bands at 35,211 and 27,777 cm⁻¹ which are due to π–π* and n–π* transitions, respectively. In all metal complexes, the absorption bands at 31,055–34,602 and 27,778–24,937 cm⁻¹, which are due to π–π* and n–π* transitions, exhibit blueshifts or redshifts due to the coordination of the ligand with metal ions. The electronic absorption spectral data for the ligand and all of the complexes are given in Table 1. The electronic spectra of the complexes [Cu(L)(phen)₂]Cl₂ and [Cu(L)(bpy)₂]Cl₂ exhibit one broad band at 15,760 and 13,755 cm⁻¹, respectively, with a low-intensity hump assigned to the ² E _g → ² T _2g transition. The above data reveal that the Cu(II) complexes adopt a distorted octahedral geometry around the central metal ion. The observed magnetic moments of the complexes (1.93 and 1.82 BM, respectively) at room temperature indicate uncoupled mononuclear complexes of a magnetically diluted d ⁹ system with an s = 1/2 spin state of distorted octahedral geometry [20]. The monomeric nature of the complexes is further supported by microanalytical data and mass spectral data. The electronic spectra of the complexes [Co(L)(phen)₂]Cl₂ and [Co(L)(bpy)₂]Cl₂ show three broad bands in the visible region (10,813, 10,989, 27,778 and 10,111, 11,248, 24937 cm⁻¹, respectively), which were assigned to ⁴ T _1g(F)→ ⁴ T _2g(F), ⁴ T _1g(F)→ ⁴ A _2g(F), and ⁴ T _1g(F)→ ⁴ T _2g(P). These data reveal that the complexes have octahedral geometries. The observed magnetic moments of the cobalt(II) complexes at room temperature indicate that these complexes are monomeric. This is further supported by microanalytical data. The electronic spectra of the complexes [Ni(L)(phen)₂]Cl₂ and [Ni(L)(bpy)₂]Cl₂ show three low-intensity bands in the visible region around 10,977, 10,040, 27,732 and 11,248, 10,060, 25,125 cm⁻¹, which were assigned to ³ A _2g(F)→ ³ T _2g(F), ³ A _2g(F)→ ³ T _1g(F), and ³A_2g(F)→ ³ T _1g(P) transitions, respectively, suggesting an octahedral geometry around the Ni(II) ion. The observed magnetic moments of the Ni(II) complexes (3.09 and 3.14 BM, respectively) at room temperature indicate uncoupled mononuclear complexes of a diluted d ⁸ system with an s = 1 spin state of octahedral geometry [21, 22]. The monomeric nature of the nickel complexes is further proven by their microanalytical data. The electronic spectra of the complexes [Mn(L)(phen)₂]Cl₂ and [Mn(L)(bpy)₂]Cl₂ show four broad, low-intensity bands in the visible region at around 15,478, 24,331, 27,732, 29,700 and 14,705, 26,954, 27,397, 29,325 cm⁻¹, respectively, which were assigned to ⁶ A _1g → ⁴ T _1g, ⁶ A _1g → ⁴ T _2g(G), ⁶ A _1g → ⁴ E _g, ⁶ A _1g → ⁴ T _2g(D) transitions. The electronic spectral data suggest an octahedral geometry around the Mn(II) ion. The observed magnetic moments of the Mn(II) complexes (5.1–5.4 BM) at room temperature indicate that these are monomeric complexes with octahedral geometries [23]. This is further supported by microanalytical data. The electronic absorption spectra for the diamagnetic Zn(II) complexes show bands in the region 34,602–29,498 cm⁻¹ which were assigned to intraligand charge-transfer transitions [24].

Table 1 Electronic absorption spectral data for the synthesized compounds at 300 K

Nuclear magnetic resonance spectra

The ¹H NMR spectra of the ligand L and the zinc complexes were recorded in DMSO-d ₆. The ¹H NMR spectrum of the Schiff base ligand shows signals at δ = 2.4 and 3.1 ppm for C–CH₃ and N–CH₃ protons of the 4-aminoantipyrine moiety. It also shows signals at 6.63, 6.95, and 7.84 ppm for the furfuryl moiety and at 6.96, 7.35, and 7.38 ppm for the phenyl group of the 4-aminoantipyrine moiety. The azomethine proton (CH=N) is observed at 9.54 ppm. The ¹H NMR spectrum of the Schiff base ligand is given in the Electronic supplementary material (ESM). The signals for the azomethine proton (–CH=N) of the bpy and phen mixed-ligand zinc complexes are shifted upfield, to 9.34 and 9.41 ppm, respectively, compared to the free ligand (9.54 ppm), suggesting the coordination of the azomethine group with the metal ion. The zinc complex [Zn(L)(bpy)₂]Cl₂ (Fig. 1) shows additional resonance peaks at 7.67, 8.20, and 8.62 ppm, which were assigned to bpy protons. Similarly, the additional resonance peaks observed at 8.0, 8.25, 8.45, and 8.86 ppm in the spectrum of the complex [Zn(L)(phen)₂]Cl₂ (see the ESM) were assigned to phen protons. There is no appreciable change in the other signals of the complex [18].

Fig. 1

¹H NMR spectrum of the [Zn(L)(bpy)₂]Cl₂ complex in DMSO-d ₆ (hash and astereisk indicate peaks from the DMSO solvent and trace water, respectively)

The ¹³C NMR spectrum of L shows aromatic carbons at 112.06–143.19 ppm. The signals from the C=O and C=N carbons of ligand L observed at 159 and 163 ppm are shifted upfield to 153 and 160 ppm for the bpy mixed-ligand Zn(II) complex and to 152.9 and 159.5 ppm for the phen mixed-ligand Zn(II) complex (see the ESM). There are no appreciable changes in the other peaks.

Mass spectra

The FAB mass spectra of synthesized L and the [Mn(L)(bpy)₂]Cl₂ complexes were recorded and the molecular ion peaks obtained confirm the proposed formulae. The mass spectrum of L shows an M + 1 peak at m/z = 282 (91.71 %) corresponding to the (C₁₆H₁₅N₃O₂)⁺ ion. Also, the spectrum exhibits fragments at m/z = 214, 202, 188, 173, and 157 corresponding to [C₁₂H₁₂N₃O]⁺, [C₁₁H₁₁N₃O]⁺, [C₁₁H₁₁N₂O]⁺, [C₁₀H₈N₂O]⁺, and [C₁₀H₈N₂]⁺, respectively (Fig. 2a). The mass spectrum of [Mn(L)(bpy)₂]²⁺ shows peaks at m/z = 648 and 649 with abundances of 20 and 15 %, respectively. The peak at m/z = 648 may represent the molecular ion peak of the complex, and the other peak is an isotopic species (Fig. 2b). The strongest peaks (base peaks) at m/z = 281, 157, and 95 represent the stable species (C₁₆H₁₅N₃O₂)⁺, (C₁₀H₈N₂)⁺, and (C₅H₄NO)⁺, respectively. The m/z of the molecular ion peak confirms that the stoichiometry of the complexes is of the [ML(bpy)₂]Cl₂ type. This is further supported by the mass spectra of all the complexes.

Fig. 2

FAB mass spectra of ligand L (a) and the complex [Mn(L)(bpy)₂]Cl₂ (b)

The observed peaks are in good agreement with their empirical formulae as indicated by microanalytical data. Similarly, the DART-MS spectrum was recorded for the [Co(L)(phen)₂]Cl₂ complex (see the ESM). The strongest peak (base peak) at m/z = 181 represents the stable [C₁₂H₈N₂]⁺ moiety (i.e., the phen ligand). Moreover, the spectrum exhibits fragments at m/z = 702, 422, and 282 corresponding to [CoC₄₀H₃₁O₂N₇]⁺, [Co(C₁₂H₈N₂)₂]⁺, and [C₁₆H₁₅N₃O₂]⁺, respectively. These represent the [Co(L)(phen)₂]Cl₂ complex after the removal of the two chloride ions, the Co(phen)₂ moiety, and L, respectively. Similarly, the other fragments are also in good agreement with the proposed structure of the complex. Thus, the mass spectral data confirm the stoichiometry of the complex as [CoL(X)₂]Cl₂ (where L = ligand, X = phen). The mass spectral data of other complexes also support the above stoichiometry. This stoichiometry is further supported by the elemental data, which are in close agreement with the values calculated from molecular formulae assigned to these complexes. Thus, the mass spectral data reinforce the conclusion drawn from the analytical and conductance values.

Electron paramagnetic resonance spectra

To obtain further information about the stereochemistry and the site of the metal–ligand bonding, and to determine the magnetic interactions in the metal complexes, EPR studies of all the copper(II) complexes were carried out. The spectra of the complexes were recorded in powder and also in DMSO solution at 300 K (room temperature) and 77 K (liquid nitrogen temperature, LNT). However, the [Cu(L)(bpy)₂]Cl₂ complex in the frozen state shows four well-resolved peaks with low intensities in the low-field region and one intense peak in the high-field region (Fig. 3). The spin Hamiltonian parameters and bonding parameters were calculated and are summarized in Tables 2 and 3, respectively. The EPR spectrum of [Cu(L)(phen)₂]Cl₂ at both 300 and 77 K shows one intense absorption band in the high-field region, and is isotropic due to dumpling motion of the molecules (see the ESM) as well as dipolar exchange and unresolved hyperfine interaction in the solid state. The magnetic moment calculated from the equation μ _eff = g[s(s + 1)]^1/2 using the experimental g _iso value (2.14) agrees very well with the measured magnetic susceptibility range of 1.93 BM, indicating that the solid structure is retained in DMSO solution. The EPR spectrum of the copper complex at 77 K indicates a poorly resolved nitrogen super hyperfine structure (Fig. 3) in the perpendicular region due to the interaction of the odd Cu(II) electron with nitrogen atoms. The magnetic susceptibility value reveals that the copper complex has a magnetic moment of 1.93 BM, corresponding to one unpaired electron, indicating that the complex is mononuclear.

Fig. 3

The X-band EPR spectrum of the [Cu(L)(bpy)₂]Cl₂ complex in DMSO at 77 K

Table 2 The spin Hamiltonian parameters of copper complexes in DMSO solution at 300 and 77 K

Table 3 The EPR bonding parameters of copper complexes in DMSO solution

In the present case, Cu(II) complexes measured in DMSO at LNT gave the following values: g _║ = 2.2728, g _⊥ = 2.0665 for [Cu(L)(phen)₂]Cl₂; g _║ = 2.2721, g _⊥ = 2.0536 for [Cu(L)(bpy)₂]Cl₂. For both complexes, the g _║ and g _⊥ values were greater than 2.04, consistent with copper(II) in axial symmetry with all the principal axes aligned parallel. These g values indicate an elongated tetragonally distorted octahedral stereochemistry [25]. However, the observation g _║ > g _⊥ > g _e shows that the unpaired electron is in the d _x²–y² orbital of the Cu(II) ion [26]. The values of the g factors are used to determine the geometric parameter G, representing a measure of the exchange interaction between Cu(II) centers in a polycrystalline compound, using the formula G = (g _II − 2.0023)/(g _⊥ − 2.0023) [27]. If G < 4, there is an exchange interaction between the Cu(II) centers, while if G > 4, the exchange interaction is negligible. The present copper complexes have G values greater than 4 indicating exchange interaction is either absent or very little in the solid complexes.

The EPR parameters g _║, g _⊥, A _II and the d–d transition energies are used to evaluate the bonding parameters α ², β ², and γ ², which may be regarded as measures of the covalency of the in-plane σ bonds, in-plane π bonds, and out-of-plane π bonds. α ² can be calculated using the equation:

$$ {\alpha^2} = \left( {{A_\parallel }/P} \right) + \left( {{g_\parallel } - 2.0023} \right) + 3/7\left( {{g_\bot } - 2.0023} \right) + 0.04. $$

If α ² is 0.5, it indicates pure covalent bonding, while α ² = 1.0 suggests pure ionic bonding. In the present case, α ² is 0.6430 for [Cu(L)(phen)₂]Cl₂ and 0.6506 for [Cu(L)(bpy)₂]Cl₂, which indicate that both of these complexes have some covalent character. The out-of-plane π-bonding (γ ²) and in plane π-bonding (β ²) parameters were calculated using the following expressions:

$${\beta^2} = \left( {{g_\parallel } - 2.0023} \right)E/\left( { - 8\lambda {\alpha^2}} \right) $$

$$ {\gamma^2} = \left( {{g_\bot } - 2.0023} \right)E/\left( { - 2\gamma {\alpha^2}} \right). $$

In these equations, λ = −828 cm⁻¹ for the free metal ion, and the E values for the complexes [Cu(L)(phen)₂]Cl₂ and (Cu(L)(bpy)₂)Cl₂ are 15,760 and 13,755 cm⁻¹, respectively. The observed values for [Cu(L)(phen)₂]Cl₂ (β ² = 0.999 and γ ² = 0.944) and for [Cu(L)(bpy)₂]Cl₂ (β ² = 0.8629 and γ ² = 0.6498) indicate that there is an interaction in the out-of-plane π-bonding whereas the in-plane π-bonding is completely ionic. This is further confirmed by orbital reduction factors, which can be estimated using the following relations:

$$ {K_\parallel }^2 = \left[ {\left( {{g_\parallel } - 2.0023} \right)\Updelta E} \right]/8{\lambda_0} $$

$$ {K_\bot }^2 = \left[ {\left( {{g_\bot } - 2.0023} \right)\Updelta E} \right]/8{\lambda_0}, $$

where λ _ο is the spin orbit coupling constant for the copper(II) ion (−828 cm⁻¹). K _║ and K _⊥ are the parallel and perpendicular components of the orbital reduction factor (K). Significant information about the nature of bonding in the copper(II) complex can be derived from the relative magnitudes of K _║ and $k_{\perp}$. In the case of pure σ-bonding, K _║ ≈ $k_{\perp}$, whereas K _║ < $k_{\perp}$ implies considerable in-plane π-bonding, while K _║ > $k_{\perp}$ for out-of-plane π-bonding. For the present case, the observed order is K _║ (0.6426) > $k_{\perp}$ (0.1518) for [Cu(L)(phen)₂]Cl₂ and K _║ (0.5594) > $k_{\perp}$ (0.1057) for [Cu(L)(bpy)₂]Cl₂, implying a greater contribution from out-of-plane π-bonding than from in-plane π-bonding to metal–ligand π-bonding. Thus, the spectral data discussed above confirm the proposed structures of the ligand L and the complexes (Scheme 1).

DNA binding studies: electronic spectra

Adding increasing amounts of calf thymus (CT) DNA to all complexes results in a decrease in the molar absorption (hypochromism) of the π–π* absorption band as well as a redshift of a few nm (1–2 nm), indicating the binding of the complexes to DNA in different modes in different extents. The binding of an intercalative complex molecule to DNA has been well characterized to produce notable hypochromism and a significant redshift due to a strong stacking interaction between the aromatic chromophore of the ligand and DNA base pairs, with the degree of hypochromism and the size of the redshift linked to the strength of the intercalative interaction [28–30]. The magnitudes of the hypochromism and the redshift observed for all of the phen mixed-ligand complexes are higher than those seen for the bpy mixed-ligand complexes (shown in Table 4). The observed binding is typical of classical intercalators and partially intercalating metal complexes bound to CT DNA. A quantitative comparison of the DNA-binding affinities of the complexes was performed using the following equation:

$$ \left[ {\text{DNA}} \right]/\left( {{\varepsilon_{\text{a}}} - {\varepsilon_{\text{f}}}} \right) = \left[ {\text{DNA}} \right]/\left( {{\varepsilon_{\text{b}}} - {\varepsilon_{\text{f}}}} \right) + 1/{K_{\text{b}}}\left( {{\varepsilon_{\text{b}}} - {\varepsilon_{\text{f}}}} \right), $$

where [DNA] is the concentration of DNA in base pairs, ε _a is the apparent extinction coefficient obtained by calculating A _obs/[complex], ε _f corresponds to the extinction coefficient of the complex in its free form, and ε _b refers to the extinction coefficient of the complex in the bound form. A straight line is obtained when [DNA]/(ε _a – ε _f) versus [DNA] is plotted. The slope of the straight line is equivalent to 1/(ε _b – ε _f) and the y intercept is equivalent to 1/K _b(ε _b – ε _f). The value of K _b (the binding constant) was determined from the ratio of the slope to the intercept.

Table 4 Absorption spectral properties of the synthesized complexes with DNA

The intrinsic binding constants K _b for [M(L)(phen)₂]²⁺ are greater than those for [M(L)(bpy)₂]²⁺, suggesting that the diimine rather than the 4-(furfurylideneamino)antipyrine face of each complex is involved in DNA binding, and that coordinated phen and bpy rings are engaged in partial insertion in between the base pairs of DNA. The observed K _b values are much lower than those observed for typical classical intercalators (EthBr, K _b = 1.4 × 10⁶ M⁻¹ in 25 mM Tris–HCl/40 mM NaCl buffer, pH 7.2). Hence, it is obvious that the present complexes are involved in weaker intercalative interactions, in spite of the steric clash between the 4-(furfurylideneamino)antipyrine ligand and the DNA double helix. Generally, the mixed-ligand complex offers the variation in geometry, size, hydrophobicity, and hydrogen-bonding ability which are the major factors to determine the DNA binding affinity [31]. Here, the ligand is not expected to be flat with the phenyl group; it may be twisted out of the plane of phenanthroline. Further, its lack of planarity diminishes the favorability of the ligand for intercalation, while the diimine part of the complex enhances the stacking of the diimine with the DNA base pairs. Further, the small binding constant values may indicate surface aggregation and groove binding rather than any interaction with the DNA base pairs. It is also notable that compounds exhibiting low binding constants have been identified in the literature as intercalating agents [32]. The electronic absorption spectrum of [Mn(L)(bpy)₂]Cl₂ in the absence and presence of increasing amounts of DNA is given in Fig. 4, and the phen complex is shown in the ESM.

Fig. 4

Absorption spectra of [Mn(L)(bpy)₂]Cl₂ in the presence of increasing amounts of DNA ([complex] = 50 μM, [DNA] = 25–125 μM from top to bottom) in Tris–HCl buffer. The arrow indicates the change in absorbance upon increasing the DNA concentration

From Table 4, we can infer that the Ni(II) complexes interact with DNA to the greatest degree, with binding constants K _b of 4.79 × 10⁴ M⁻¹ and 9.52 × 10⁴ M⁻¹, respectively. These values are very distinctive of metal-based compounds that bind to DNA via intercalation. This is because the nickel complexes enhance the oxidation of guanine residues in DNA [33–35]. This has been experimentally proven through DNA cleavage studies, as described later in this manuscript. Moreover, this distinct mechanism causes the nickel complexes to tend to act as effective intercalators and artificial nucleases when compared to other members.

Viscosity measurements

Measuring the viscosity of DNA is regarded as the least ambiguous and most critical test of a DNA binding model in solution, and affords stronger arguments for an intercalative DNA binding mode [36, 37]. The viscosity of DNA is significantly enhanced by complete or partial intercalation of complexes into the DNA base-pair stack, but it is slightly disturbed by the electrostatic or covalent binding of molecules [38]. To investigate the nature of DNA binding of the mixed-ligand complexes, viscosity measurements of CT DNA were carried out by varying the concentrations of the added complexes. The changes in the relative specific viscosity of CT DNA in the presence and absence of the complex were plotted against 1/R (=[complex]/[DNA]).

A classical intercalation mode causes a significant increase in the viscosity of DNA due to an increase in base-pair separation at intercalation sites and hence an increase in the overall DNA length. In our study, the ability of each complex to increase the viscosity of DNA depended upon the diimine ligand. Phen insertion produces greater viscosity than insertion of the bpy ligand (as shown in the ESM). The results of this viscosity study agree well with the hypochromism and K _b values observed for the above complexes. Lengthening of the DNA duplex occurs upon the partial insertion of phen and the deeper insertion of bpy between the DNA base pairs.

Electrochemical studies

Cyclic and differential pulse voltammetric techniques are extremely useful for probing the nature and mode of the binding of metal complexes to DNA. Typical cyclic voltammograms obtained for the complex [Cu(L)(bpy)₂]Cl₂ in the absence and presence of varying amounts of DNA are shown in Fig. 5. The incremental addition of CT DNA to the complex causes the anodic and cathodic peak currents of the complex to decrease. This result shows that the complex stabilizes the duplex (GC pairs) by intercalation. The incremental addition of CT DNA to the complex causes a shift in the potential of the peaks in the cyclic voltammograms. The cathodic and anodic peaks both show positive or negative shifts, which indicates the intercalation of the complex into the base pairs of the DNA. If one of the peak shifts was positive and the other was negative, it would indicate that the complex intercalates into and electrostatically binds to the CT DNA, or that it breaks the secondary structure of the DNA.

Fig. 5

Cyclic voltammogram (scan rate of 10 mV/s) of [Cu(L)(bpy)₂]Cl₂ in the absence and presence of increasing amounts of DNA: 10, 20, 40, 100, 150, 200, 250 μM. Supporting electrolyte: 5 mM Tris–HCl + 50 mM NaCl in water (pH 7.2)

In the differential pulse voltammogram of the complex [Cu(L)(bpy)₂]Cl₂ obtained in the absence and the presence of varying amounts of DNA (in which a significant decrease in current intensity with increasing DNA concentration is observed; see Fig. 6), the shift in potential is related to the ratio of the binding constants by the following equation:

$$ {E_{\text{b}}} - {E_{\text{f}}} = 0.0591\,\log \left( {{K_{\left[ {\text{red}} \right]}}/{K_{\left[ {\text{ox}} \right]}}} \right), $$

where E _b and E _f are the peak potentials of the complex in its bound and free forms, respectively. In the present study, both the bpy and the phen mixed-ligand complexes exhibited one-electron transfer during the redox process, and the Ip _c/Ip _a value was less than unity, indicating that the reaction of the complex on the glassy carbon electrode surface was quasi-reversible. Other complexes—those including Co(II), Ni(II), Mn(II), and Zn(II)—show considerable shifts in both their cathodic and anodic peak potentials when CT DNA is added incrementally. Most of the complexes we synthesized exhibited both anodic and cathodic peak potential shifts which were either positive or negative, except for manganese complexes (Table 5). This indicates the mode of intercalation that occurs when the DNA binds with bpy/phen mixed-ligand Schiff base complexes. It is interesting to note that phen mixed-ligand complexes show a greater decrease in current intensity than bpy mixed-ligand complexes (Fig. 5 and the ESM). This observation suggests that the intercalative mode that occurs when DNA binds to phen mixed-ligand complexes is stronger than that involved when DNA binds to bpy mixed-ligand complexes. Another interesting aspect of our study is that manganese complexes exhibit positive and negative shifts in peak potential. This indicates that there is an intercalative and electrostatic binding mode between DNA and these complexes, or that they break up the secondary structure of DNA.

Fig. 6

Differential pulse voltammogram (scan rate of 10 mV/s) of 100 μM of the complex [Cu(L)(bpy)₂]Cl₂ in the absence and presence of increasing amounts of DNA: 10, 20, 40, 100, 150, 200, 250 μM. Supporting electrolyte: 5 mM Tris–HCl + 50 mM NaCl in water (pH 7.2)

Table 5 Electrochemical parameters for the interaction of DNA with metal(II) complexes

Antimicrobial activity

The investigated compounds were tested for antimicrobial activity against bacteria and fungi. The minimum inhibitory concentration (MIC) values of the investigated compounds are summarized in Table 6 and 7. A comparative study of L and the complexes (in terms of MIC values) indicates that the complexes exhibit higher antimicrobial activities than the free ligand. From the MIC values, it is apparent that the phen mixed-ligand complexes are more potent than the other investigated complexes, which can be explained via Overtone’s concept and Tweedy’s chelation theory [39, 40]. These complexes also disturb the respiration process of the cell and thus block the synthesis of proteins that restrict further growth of the organism. Furthermore, the mode of action of the compound may involve the formation of a hydrogen bond through the azomethine group with the active center of the cell, resulting in interference with the normal cell process, and the metal complexes are generally more active than the ligands because the metal complexes can serve as a vehicle for the activation of ligands—the principal cytotoxic species [41]. The 4-(furfurylideneamino)antipyrine Schiff base ligand may play an important role in this antimicrobial activity, as well as the presence of four imine groups brings the transformation reaction in biological systems. The activities of ligands and their complexes increase as the concentration increases, because it is a well-known fact that concentration plays a vital role in increasing the degree of inhibition. The results indicate that the complexes show more activity than the ligands against the same microorganisms under identical conditions (Tables 6, 7). These data are in good agreement with literature values [42].

Table 6 Minimum inhibitory concentrations (mg cm⁻³) of the synthesized compounds against the growth of four fungi

Table 7 Minimum inhibitory concentrations (mg cm⁻³) of the synthesized compounds against the growth of four bacteria

It is well known that chelation is not the only criterion for antibacterial activity. Some important factors such as the nature of the metal ion, the nature of the ligand, the coordinating sites, the geometry of the complex, the concentration, hydrophilicity, lipophilicity, and the presence of co-ligands have a considerable influence on the antibacterial activity. Steric and pharmacokinetic factors undoubtedly also play a decisive role in deciding the potency of an antimicrobial agent. The presence of lipophilic and polar substituents is expected to enhance antibacterial activity. Heterocyclic ligands with multifunctionality which have a greater chance of interacting with either nucleoside bases (even after complexation with a metal ion) or with biologically essential metal ions present in the biosystem may be promising candidates for bactericides since they always tend to interact (especially with a few enzymatic functional groups) in order to achieve higher coordination numbers [43]. Thus, the antibacterial property of metal complexes cannot be ascribed to chelation alone; it is a complicated blend of several contributions.

Cleavage of pUC19 DNA

The nuclease activity exhibited by certain transition metal complexes in the presence of hydrogen peroxide can be attributed to the participation of the hydroxyl radical in DNA cleavage. When nicked, the naturally occurring supercoiled circular form of pUC19 DNA (form I) gives rise to an open circular form (form II). Relatively fast migration is observed for form I, but rather slow migration for form II. In the present study, a DNA cleavage experiment was conducted at 35 °C in the presence and absence of H₂O₂. As can be seen from the results shown in Fig. 7a, the gel electrophoretic separation of plasmid pUC19 DNA occurs when treated with [ML(bpy)₂]Cl₂ complexes. Control experiments suggest that untreated DNA does not show any significant DNA cleavage (lane 1). However, in the presence of peroxide, most of the [ML(bpy)₂]Cl₂ and [ML(phen)₂]Cl₂ complexes convert the supercoiled plasmid DNA into a mixture of nicked (form II) and linear (form III) DNA (Fig. 7b and Fig. 7c). Further, in the presence of DMSO (4 mm³), [CuL(bpy)₂]Cl₂ + H₂O₂ does not show any DNA-cleavage activity (lane 4 in Fig. 7b). In this case, the cleavage is inhibited by free-radical scavengers (DMSO), implying that hydroxyl radicals mediate the cleavage. This provides clear evidence for the involvement of hydroxyl radicals in the DNA strand scission performed by M(II) complexes. Here, the toxicity of metal is expressed through the formation of hydroxyl radical, which involves redox process with metal center and H₂O₂. Moreover, the mechanism commences with the oxidation of deoxyribose units by the free radicals, leading to the fragmentation of the sugar-phosphate backbone. The loaded (L(phen/bpy)₂)M²⁺ complexes damage DNA more efficiently in the presence of an oxidant and form DNA adducts that migrate at different rates. This can be attributed to the formation of hydroxyl free radicals. The production of a hydroxyl free radical in the reaction between the complex and the oxidant may occur as shown below:

$$ \left( {{\text{L}}{{\left( {{\text{phen}}/{\text{bpy}}} \right)}_2}} \right){{\text{M}}^{2 + }} + {{\text{H}}_2}{{\text{O}}_2} \to \left( {{\text{L}}{{\left( {{\text{phen}}/{\text{bpy}}} \right)}_2}} \right){{\text{M}}^{3 + }} + {\text{OH}} \cdot + {\text{O}}{{\text{H}}^- } $$

These OH free radicals participate in the oxidation of the deoxyribose moiety, followed by the hydrolytic cleavage of the sugar-phosphate backbone. Here, the most significant shifts are observed for Co(II) and Ni(II) (Fig. 7b, lanes 7 and 5), whereas the DNA adducts with Cu(II) and Zn(II) complexes (lanes 3 and 8) migrate to a lesser extent. This difference in the migration patterns of the adducts points to a difference in the binding modes of these compounds compared to the other members of the series. Control experiments using DNA alone did not show any significant cleavage of pUC19 DNA, even after longer exposure times. Based on these results, we can conclude that the Ni(II) complexes (Fig. 7b, lane 5) loaded with H₂O₂ are most effective at cleaving DNA, and can thus be considered better artificial nucleases than the other members and the control. Hence, it is evident from the interactions of the complexes with CT DNA and pUC19 DNA that the cobalt and nickel complexes are more effective DNA-damaging agents than the other members. This is because the cobalt and nickel complexes enhance the oxidation of guanine residues in DNA [33–35]. This mechanism means that the cobalt and nickel complexes are more effective intercalators and artificial nucleases than the copper and zinc complexes.

Fig. 7

a The gel electrophoretic separation of plasmid pUC19 DNA treated with [M(L)(bpy)₂]Cl₂ complexes. Lane 1 DNA marker, lane 2 DNA + ligand, lane 3 DNA + [Ni(L)(bpy)₂]Cl₂, lane 4 DNA + [Co(L)(bpy)₂]Cl₂, lane 5 DNA + [Cu(L)(bpy)₂]Cl₂, lane 6 DNA + [Mn(L)(bpy)₂]Cl₂, lane 7 DNA + [Zn(L)(bpy)₂]Cl₂. b The gel electrophoretic separation of plasmid pUC19 DNA treated with [M(L)(bpy)₂]Cl₂ complexes. Lane 1 DNA marker, lane 2 DNA + ligand, lane 3 DNA + [Cu(L)(bpy)₂]Cl₂ + H₂O₂, lane 4 DNA + [Cu(L)(bpy)₂]Cl₂ + H₂O₂ + DMSO, lane 5 DNA + [Ni(L)(bpy)₂]Cl₂ + H₂O₂, lane 6 DNA + [Mn(L)(bpy)₂]Cl₂ + H₂O₂, lane 7 DNA + [Co(L)(bpy)₂]Cl₂ + H₂O₂, lane 8 DNA + [Zn(L)(bpy)₂]Cl₂ + H₂O₂. c The gel electrophoretic separation of plasmid pUC19 DNA treated with [M(L)(phen)₂]Cl₂ complexes. Lane 1 DNA marker, lane 2 DNA + [Cu(L)(phen)₂]Cl₂ + H₂O₂, lane 3 DNA + [Cu(L)(phen)₂]Cl₂, lane 4 DNA + [Co(L)(phen)₂]Cl₂ + H₂O₂, lane 5 DNA + [Ni(L)(phen)₂]Cl₂ + H₂O₂, lane 6 DNA + [Zn(L)(phen)₂]Cl₂ + H₂O₂, lane 7 DNA + [Mn(L)(phen)₂]Cl₂ + H₂O₂, lane 8 DNA + ligand + H₂O₂

Molecular docking analysis

The results of spectroscopic titration, viscosity, and cyclic voltammetry experiments suggest that Cu(II), Ni(II), Co(II), and Zn(II) complexes interact with DNA via intercalation. The binding constant values are in the range of 10⁴ per mole for all of the complexes, which indicates that these complexes may intercalate between the DNA base pairs. This is mainly due to effective stacking forces between the aromatic ligands and DNA bases. Docking experiments were performed in order to find the perferred binding site and orientation of the ligand inside the DNA. Docking experiments using the energy-minimized docked structures suggest that the best possible geometries of the metal complexes occur when they act as partial intercalators (as shown in Fig. 8). Structural analysis of the docked structures provides significant information on the binding patterns of these complexes. The binding energies of the docked metal complexes [CuL(phen)₂]Cl₂, [CuL(bpy)₂]Cl₂, [CoL(phen)₂]Cl₂, [CoL(bpy)₂]Cl₂, [NiL(phen)₂]Cl₂, [NiL(bpy)₂]Cl₂, [ZnL(phen)₂]Cl₂, [Zn(bpy)₂]Cl₂, [MnL(phen)₂]Cl₂, and [MnL(bpy)₂]Cl₂ are −557.32, −555.19, −557.33, −527.94, −547.01, −534.19, −528.81, −527.08, −530.04, and −529.36 kJ mol⁻¹, respectively, which correlate well with the corresponding experimental DNA binding values. It was found that the binding between the DNA and target molecules was stronger when the relative binding energy was more negative.

Fig. 8

Interaction of the complexes a [Cu(L)(bpy)₂]Cl₂ and b [Cu(L)(phen)₂]Cl₂ with d(CGCGAATTCGCG) strands of DNA via intercalation binding

Conclusion

The synthesized L and the complexes were characterized by microanalytical data, IR, UV–Vis, ¹H NMR, EPR, and mass spectra. The data show that the complexes are of type [M(L)(phen/bpy)₂]Cl₂. The UV–Vis, magnetic susceptibility, and EPR spectral data for the complexes suggest an octahedral geometry around the central metal ion. The complexes show high electrical conductance, indicating that these chelates are electrolytes. Their magnetic susceptibility values provide evidence for their monomeric nature. The planarity and flexible extended conjugation of the ligands have a strong effect on the DNA-binding and DNA-cleavage activities of these complexes. The intercalative binding of these complexes with DNA is supported by their electronic absorption spectra as well as cyclic voltammetry, difference pulse voltammetry, and viscometric studies. The planarity of the complexes increases their DNA-binding and DNA-cleavage activities. The results obtained from in vitro antifungal and antibacterial tests show that all of these complexes possess potent antimicrobial activities toward bacteria and fungi. It was found that the activities of the complexes are higher than that of the free ligand.

Experimental

All reagents and chemicals were procured from Merck (Darmstadt, Germany). Solvents used for electrochemical and spectroscopic studies were purified by standard procedures [44]. DNA was purchased from Bangalore Genei (Bangalore, India). Agarose (molecular biology grade) and ethidium bromide (EB) were obtained from Sigma (St. Louis, MO, USA). Tris(hydroxymethyl)aminomethane-HCl (Tris–HCl) buffer solution was prepared using deionized, doubly distilled water. Carbon, hydrogen, and nitrogen analyses of the complexes were carried out on a CHN analyzer (1108, Heraeus Carlo Erba, Hanau, Germany). The IR spectra (KBr discs) of the samples were recorded in the range of 400–4,000 cm⁻¹ on a PerkinElmer (Waltham, MA, USA) 783 series FTIR spectrophotometer. Electronic absorption spectra in the 200–1,100 nm range were obtained on a Shimadzu (Kyoto, Japan) UV-1601 spectrophotometer. N,N-dimethylformamide (DMF) was used as solvent when obtaining electronic spectra for both the ligand L and its complexes. The NMR spectra of L and the zinc complexes were recorded on a Bruker (Billerica, MA, USA) Avance DRX 300 FT-NMR spectrometer, using DMSO-d ₆ as solvent, at the University Science Instrumentation Centre (USIC), Madurai Kamaraj University, Madurai, India. Tetramethylsilane was used as the internal standard. Fast atom bombardment mass spectra of the complexes were recorded on a JEOL (Tokyo, Japan) SX 102/DA-6000 mass spectrometer/data system using argon/xenon (6 kV, 10 mA) as the FAB gas. X-band EPR spectra of the copper complexes were recorded at RT (300 K) and LNT (77 K) using DPPH as the g-marker.

The molar conductances of the complexes (10⁻³ M) in DMF were measured at room temperature with a Deep Vision (Tamilnadu, India) model 601 digital direct reading deluxe conductivity meter. Magnetic susceptibility measurements were carried out using the Gouy method at room temperature on powdered samples of the complexes. CuSO₄·5H₂O was used as calibrant. Electrochemical measurements were performed on a CHI (Austin, TX, USA) 620C electrochemical analyzer with a three-electrode system comprising a glassy carbon electrode as the working electrode, a platinum wire as the auxiliary electrode, and Ag/AgCl as the reference electrode. Solutions were deoxygenated by purging with N₂ prior to measurements. The metal content of each complex was determined according to a method described previously [45]. Chloride ion was determined gravimetrically as silver chloride [46]. The ligand L was synthesized by following the previously reported procedure [47]. The ligand was purified by column chromatography eluted with alcohol:DMF mixture and the purity was checked by TLC. The complexes were purified by repeated washing with ethanol.

Synthesis of [M(L)(phen)₂]Cl₂ and [M(L)(bpy)₂]Cl₂

The metal(II) complexes were prepared by mixing the appropriate molar quantities of the ligands and metal salts using the following procedure. An ethanolic solution of L (10 mmol) was heated under reflux with an ethanolic solution of the metal(II) chloride (10 mmol) for ca. 3 h. To the above mixture, an ethanolic solution of 1,10-phenanthroline/2,2′-bipyridine (20 mmol) was added, and the reflux was continued for ca. 1 h. The resulting solid product was filtered, washed with ethanol, and dried in vacuo.

[4-[(Furan-2-ylmethylene)amino]-1,2-dihydro-1,5-dimethyl-2-phenyl-3H-pyrazol-3-one]bis(1,10-phenanthroline)copper(II) chloride (C₄₀H₃₁Cl₂CuN₇O₂)

Yield: 62 %; IR (KBr): $\bar \nu$ = 1,647 (C=O), 1,601 (–CH=N), 1,545 (C=N) cm⁻¹; Λ_m = 157.50 S cm² mol⁻¹; μ _eff = 1.93 BM; UV–Vis (DMF): λ _max = 15,760, 33,898 cm⁻¹; MS: m/z = 741 (M⁺).

[4-[(Furan-2-ylmethylene)amino]-1,2-dihydro-1,5-dimethyl-2-phenyl-3H-pyrazol-3-one]bis(1,10-phenanthroline)cobalt(II) chloride (C₄₀H₃₁Cl₂CoN₇O₂)

Yield: 54 %; IR (KBr): $\bar \nu$ = 1,645 (C=O), 1,597 (–CH=N), 1,547 (C=N) cm⁻¹; Λ_m = 138.96 S cm² mol⁻¹; μ _eff = 4.81 BM; UV–Vis (DMF): λ _max = 27,778, 10,989, 10,813 cm⁻¹; MS: m/z = 736 (M⁺).

[4-[(Furan-2-ylmethylene)amino]-1,2-dihydro-1,5-dimethyl-2-phenyl-3H-pyrazol-3-one]bis(1,10-phenanthroline)nickel(II) chloride (C₄₀H₃₁Cl₂N₇NiO₂)

Yield: 59 %; IR (KBr): $\bar \nu$ = 1,647 (C=O), 1,599 (–CH=N), 1,545 (C=N) cm⁻¹; Λ_m = 131.47 S cm² mol⁻¹; μ _eff = 3.09 BM; UV–Vis (DMF): λ _max = 27,732, 10,040, 10,977 cm⁻¹; MS: m/z = 736 (M⁺).

[4-[(Furan-2-ylmethylene)amino]-1,2-dihydro-1,5-dimethyl-2-phenyl-3H-pyrazol-3-one]bis(1,10-phenanthroline)manganese(II) chloride (C₄₀H₃₁Cl₂MnN₇O₂)

Yield: 60 %; IR (KBr): $\bar \nu$ = 1,648 (C=O), 1,591 (–CH=N), 1,547 (C=N) cm⁻¹; Λ_m = 144.46 S cm² mol⁻¹; μ _eff = 5.10 BM; UV–Vis (DMF): λ _max = 29,700, 27,732, 24,331, 15,748 cm⁻¹; MS: m/z = 732 (M⁺).

[4-[(Furan-2-ylmethylene)amino]-1,2-dihydro-1,5-dimethyl-2-phenyl-3H-pyrazol-3-one]bis(1,10-phenanthroline)zinc(II) chloride (C₄₀H₃₁Cl₂N₇O₂Zn)

Yield: 57 %; IR (KBr): $\bar \nu$ = 1,646 (C=O), 1,597 (–CH=N), 1,548 (C=N) cm⁻¹; ¹H NMR (DMSO-d ₆): δ = 2.40 (s, 3H, C–CH₃), 3.10 (s, 3H, N–CH₃), 6.63, 6.95 (m, 2H, furfuryl), 6.96–7.38 (m, 5H, phenyl), 7.84 (m, 1H, furfuryl), 8.00, 8.25, 8.45, 8.86 (m, 8H, phen), 9.41 (s, CH=N) ppm; ¹³C NMR (DMSO-d ₆): δ = 9.7 (C–CH₃), 35.3 (N–CH₃), 112.2, 113.6, 116.2, 124.7, 125.5, 126.9, 128.8, 129.1, 134.4, 139.5, 139.8, 143.2, 145.1, 148.8, 151.7, 152.9 (C=O), 159.5 (HC=N) ppm; Λ_m = 136.86 S cm² mol⁻¹; μ _eff = 0 BM; MS: m/z = 743 (M⁺).

[4-[(Furan-2-ylmethylene)amino]-1,2-dihydro-1,5-dimethyl-2-phenyl-3H-pyrazol-3-one]bis(2,2′-bipyridine)copper(II) chloride (C₃₆H₃₁Cl₂CuN₇O₂)

Yield: 64 %; IR (KBr): $\bar \nu$ = 1,647 (C=O), 1,601 (–CH=N), 1,545 (C=N) cm⁻¹; Λ_m = 112.69 S cm² mol⁻¹; μ _eff = 1.82 BM; UV–Vis (DMF): λ _max = 31,055, 13,755 cm⁻¹; MS: m/z = 657 (M⁺).

[4-[(Furan-2-ylmethylene)amino]-1,2-dihydro-1,5-dimethyl-2-phenyl-3H-pyrazol-3-one]bis(2,2′-bipyridine)cobalt(II) chloride (C₃₆H₃₁Cl₂CoN₇O₂)

Yield: 61 %; IR (KBr): $\bar \nu$ = 1,648 (C=O), 1,599 (–CH=N), 1,545 (C=N) cm⁻¹; Λ_m = 132.45 S cm² mol⁻¹; μ _eff = 4.93 BM; UV–Vis (DMF): λ _max = 24,937, 11,248, 10,111 cm⁻¹; MS: m/z = 653 (M⁺).

[4-[(Furan-2-ylmethylene)amino]-1,2-dihydro-1,5-dimethyl-2-phenyl-3H-pyrazol-3-one]bis(2,2′-bipyridine)nickel(II) chloride (C₃₆H₃₁Cl₂N₇NiO₂)

Yield: 60 %; IR (KBr): $\bar \nu$ = 1,649 (C=O), 1,592 (–CH=N), 1,546 (C=N) cm⁻¹; Λ_m = 130.09 S cm² mol⁻¹; μ _eff = 3.14 BM; UV–Vis (DMF): λ _max = 25,125, 11,248, 10,060 cm⁻¹; MS: m/z = 652 (M⁺).

[4-[(Furan-2-ylmethylene)amino]-1,2-dihydro-1,5-dimethyl-2-phenyl-3H-pyrazol-3-one]bis(2,2′-bipyridine)manganese(II) chloride (C₃₆H₃₁Cl₂MnN₇O₂)

Yield: 58 %; IR (KBr): $\bar \nu$ = 1,647 (C=O), 1,602 (–CH=N), 1,543 (C=N) cm⁻¹; Λ_m = 152.70 S cm² mol⁻¹; μ _eff = 5.40 BM; UV–Vis (DMF): λ _max = 29,325, 27,397, 26,954, 14,705 cm⁻¹; MS: m/z = 649 (M⁺).

[4-[(Furan-2-ylmethylene)amino]-1,2-dihydro-1,5-dimethyl-2-phenyl-3H-pyrazol-3-one]bis(2,2′-bipyridine)zinc(II) chloride (C₃₆H₃₁Cl₂N₇O₂Zn)

Yield: 56 %; IR (KBr): $\bar \nu$ = 1,647 (C=O), 1,601 (–CH=N), 1,545 (C=N) cm⁻¹; ¹H NMR (DMSO-d ₆): δ = 2.40 (s, 3H, C–CH₃), 3.10 (s, 3H, N–CH₃), 6.63, 6.95 (m, 2H, furfuryl), 6.96–7.38 (m, 5H, phenyl), 7.84 (m, 1H, furfuryl), 7.67, 8.20, 8.62 (m, 8H, bpy), 9.34 (s, 1H, CH=N) ppm; ¹³C NMR (DMSO-d ₆): δ = 9.6 (C–CH₃), 35.1 (N–CH₃), 112.1, 113.3, 116.1, 124.6, 126.8, 129.0, 134.4, 139.9, 143.2, 144.9, 149.1, 151.6, 153.0 (C=O), 160.0 (HC=N) ppm; Λ_m = 123.18 S cm² mol⁻¹; μ _eff = 0 BM; MS: m/z = 659 (M⁺).

DNA-binding and DNA-cleavage experiments

All experiments involving the interaction of the complexes with calf thymus (CT) DNA were carried out in Tris–HCl buffer (50 mM Tris–HCl, pH 7.2) containing 5 % DMF at room temperature. A solution of CT DNA in the buffer gave a ratio of UV absorbances at 260 and 280 nm of about 1.89:1, indicating that the CT DNA was sufficiently free from protein [48]. The CT DNA concentration per nucleotide was determined by absorption spectroscopy using a molar absorption coefficient of 6,600 M⁻¹ cm⁻¹ at 260 nm [49].

An absorption titration experiment was performed by maintaining the concentration of the metal complex at 50 μM while varying the concentration of CT DNA within 40–400 μM. While measuring the absorption spectrum, equal quantities of CT DNA were added to both the complex solution and the reference solution to eliminate the absorbance of CT DNA itself. Using the absorption data, the intrinsic binding constant K _b was determined from a plot of [DNA]/(ε _a – ε _f) versus [DNA] and the equation

$$ \left[ {DNA} \right] \, / \, \left( {{\varepsilon_a} - {\varepsilon_f}} \right) = \left[ {DNA} \right]/\left( {{\varepsilon_b} - {\varepsilon_f}} \right) + {\left[ {{K_b}\left( {{\varepsilon_b} - {\varepsilon_f}} \right)} \right]^{ - 1}}, $$

(1)

where [DNA] is the concentration of CT DNA in base pairs. The apparent absorption coefficients ε _a, ε _f, and ε _b correspond to A _obs/[M], the extinction coefficient for the free metal(II) complex, and the extinction coefficient for the metal(II) complex in the fully bound form, respectively [50]. K _b is given by the ratio of the slope to the intercept.

Cyclic voltammetry and differential pulse voltammogram studies were performed on a CHI 620C electrochemical analyzer with a three-electrode system comprising glassy carbon as the working electrode, a platinum wire as the auxiliary electrode, and Ag/AgCl as the reference electrode. Solutions were deoxygenated by purging with N₂ prior to measurements.

Viscosity experiments were carried out on an Ostwald viscometer immersed in a thermostated water bath maintained at a constant temperature of 30.0 ± 0.1 °C. CT DNA samples of approximately 0.5 mM were prepared by sonicating in order to minimize complexities arising from CT DNA flexibility [51]. The flow time was measured with a digital stopwatch three times for each sample, and an average flow time was then calculated. Data are presented as (η/η ₀)^1/3 versus the concentration of the metal(II) complex, where η is the viscosity of the CT DNA solution in the presence of the complex, and η ₀ is the viscosity of the CT DNA solution in the absence of the complex. Viscosity values were calculated after correcting the flow time of the buffer alone (t ₀), η = (t – t ₀)/t ₀ [52].

The degree of cleavage of the supercoiled (SC) pUC19 DNA (33.3 μM, 0.2 μg) to its nicked circular (NC) form was determined by agarose gel electrophoresis in 5 mM Tris–HCl buffer (pH 7.2) containing 50 mM NaCl. After they had been incubated for 2 h at 37 °C in a dark chamber, the samples were added to a loading buffer containing 25 % bromophenol blue, 0.25 % xylene cyanol, and 30 % glycerol (3 mm³), and the solution was finally loaded onto 0.8 % agarose gel containing 1 μg/cm³ ethidium bromide. Electrophoresis was carried out in a dark chamber for 3 h at 50 V in Tris–HCl–EDTA buffer. Bands were visualized by UV light and photographed.

Antimicrobial activity

The in vitro biological screening effects of the investigated compounds were tested against the Gram-positive bacteria Staphylococcus aureus and Bacillus subtilis and the Gram-negative bacteria Escherichia coli and Klebsiella pneumoniae by the well diffusion method [53], using agar nutrient as the medium and streptomycin as the standard. The antifungal activities of the compounds against the fungi Aspergillus niger, Fusarium solani, Curvularia lunata, and Rhizoctonia bataicola (cultured on potato dextrose agar as medium, and with nystatin used as the standard) were evaluated by the well diffusion method. The stock solution (10⁻² M) was prepared by dissolving the compounds in DMSO, and the solutions were serially diluted in order to determine the minimum inhibitory concentration (MIC) values. In a typical procedure [54], a well was made in the agar medium inoculated with microorganisms. The well was filled with the test solution using a micropipette, and the plate was incubated for 24 h for bacteria and for 72 h for fungi at 35 °C. During this period, the test solution diffused and the growth of the inoculated microorganisms was affected. The inhibition zone developed, and the corresponding concentration was noted.

Molecular modeling studies

The interactions of the metal complexes with DNA were also studied by molecular modeling and docking. All calculations were performed in Open Eye with fast rigid exhaustive docking using the FRED docking software package. The main function employed for the calculations was ChemGauss 2. Prior to docking, the structures of the metal complexes were constructed and their geometries were optimized by the MM2 force field. The crystal structure of the complex of netropsin with B-DNA dodecamer d(CGCGAATTCGCG)₂ (NDB code GDLB05) was downloaded from the Protein Data Bank. During the docking analysis, the binding site of netropsin was used as the binding site for our metal complexes across the entire DNA molecule. The docked poses were generated by the exhaustive search and optimization step. FRED selects the single best pose from among a set of candidates. This pose is then scored and the score is used to rank ligands in the output hit list. The consensus structure step allows multiple scoring functions to vote for the best docked structure in a rank-by-vote approach.

Statistical analysis

All of the experiments conducted in this work were done in triplicate, and all of the data are reported here as the mean ± standard deviation.