Biochemical activities and electronic spectra of different cobalt phenanthroline complexes

2.1 Chemicals and instruments

Chemicals were Merck and Sigma products. Sodium carbonates, 1,10-phenanthroline (AO-A), cobalt dichloride hexahydrates (AO-B), cobalt sulfate heptahydrate (AO-C). All organic solvents used were of spectroscopic quality (>99.8%). The cell lines HEPG-2, MCF-7 and HCT-116 were purchased from ATCC from “Holding” company for biological products and vaccines (VACSERA), Cairo, Egypt.

FT-IR reflectance spectra were recorded by using Shimadzu Infrared Spectrophotometer (FT-IR-8300) in the range (4000–400) cm⁻¹. UV–Vis spectra of cobalt phenantholine complexes were determined on Shimadzu UV–Vis spectrophotometer.

2.2 Synthesis of complexes

The octahedral complexes (Fig. 1) were prepared according to the methods described in the literature with some modification:

Close

Fig. 1

Structures of cobalt phenanthroline complexes.

Export to PPT

CoCl₂·6H₂O (5.9 mmol) and phenanthroline (3 mmol) were mixed in 40 ml acetonitrile, a blue powder completely precipitated after 3 h, filtered, collected, washed with little acetonitrile and dried giving [Co(phen)₂Cl₂].1.5 CH₃CN (AO-2) (1.64 mmol) (28% yield).

[Co(phen)(H₂O)₄]SO₄·2H₂O (AO-3) was prepared by mixing CoSO₄·7H₂O (3.56 mmol), phenanthroline (3.56 mmol) in 20 ml water for 2 h, the obtained pink powder was collected by filtration, washed with water and dried to give AO-3 (2.03 mmol) (57% yield).

CoCl₂·6H₂O (2 mmol), phenanthroline (4 mmol) and sodium carbonate (4 mmol) were mixed in 20 ml water for 2 h, the precipitated orange powder was collected, washed with little water and ethanol and dried giving [Co(phen)₂CO₃].7H₂O (AO-5) (1.16 mmol) (58% yield).

The structures shown in Fig. 1 are redrawn based on reported structures obtained by single X-ray diffraction analysis (CIF files) of the crystals of complexes: [Co(phen)₂Cl₂].1.5 CH₃CN (AO-2) (Alan et al., 1997); [Co(phen)(H₂O)₄]SO₄·2H₂O (AO-3) (Zhu et al., 2002); [Co(phen)₂CO₃].7H₂O (AO-5) (Li et al., 2004).

2.3 Electronic spectral studies

The electronic spectra were recorded on a Shimadzu Spectrophotometer using 1 cm quartz cell at 25 °C. The electronic spectra of the solutions of complexes (concentration 1 × 10⁻⁵ to 1 × 10⁻⁶ mol/L) were investigated in water and organic solvents of various polarities: methanol, ethanol, acetonitrile, N,N-dimethylformamide (DMF), and dimethylsulfoxide (DMSO).

2.4 Antibacterial test

The antimicrobial evaluation of the tested ligands (AO-A, AO-B, AO-C) and their complexes (AO-2, AO-3, AO-5) were carried out via cup diffusion technique (Youssef et al., 2016).

Minimal inhibitory concentration (MIC) values of AO-A, AO-B, AO-C, AO-2, AO-3 and AO-5 compounds were studied according to micro well dilution method (Kavanagh, 1963).

2.5 Antioxidant activity

Estimation of antioxidant activities of Co complexes using DPPH: Radical scavenging activity of compounds AO-A, AO-B, AO-C, AO-2, AO-3 and AO-5 against stable 2, 2-diphenyl-1-picrylhidrazine (DPPH) was estimated by UV/visible light spectrophotometery according to (Mensor et al., 2001).

Determination of Superoxide Dismutase (SOD) like activity: AO-A, AO-B, AO-C, AO-2, AO-3 and AO-5 were assayed for superoxide dismutase enzyme SOD like activity using phenazenemethosulphate (PMS) to generate a superoxide anion radicals at pH = 8.3 (phosphate buffer) as reported (Bridges and Salin, 1981).

ABTS radical cation decolorization assay: ABTS [(2,2′-azino-bis-3-ethylbenzthiazoline-6-sulphonic acid) diammonium salt] forms a relatively steady free radical which is decolorized in its non-radical form. The evaluation of ABTS free radical scavenging activity was assessed spectrophotometrically as reported (Re et al., 1999).

2.6 Agarose gels electrophoresis

The AO-A, AO-B, AO-C, AO-2, AO-3 and AO-5 (20 μg) were added individually to 1 µg of the DNA isolated from E. coli strain W3110 (Youssef and Al-Omair, 2008). The DNA was visualized on a UV transilluminator after staining the agarose gels with ethidium bromide (0.5 µg/ml) (Sambrook et al., 1989).

2.7 Cytotoxicity assay

Doxorubicin was used as a reference anticancer drug for comparison. The cell lines stated above were utilized to evaluate the inhibitory effects of the ligands and metal salts (AO-A, AO-B, AO-C), and the complexes (AO-2, AO-3, AO-5) according to the reported method (Elghamry et al., 2017).

3.1 FTIR studies

The strong bands at 732.2 cm⁻¹ and 839.9 cm⁻¹ have been assigned to out of plane C—H deformation motion on the heterocyclic rings, and on the center ring respectively of phenanthroline AO-A. There is a decrease of ≈10 cm⁻¹ in the wavenumbers of the first band due to co-ordination of nitrogen atom. It appears at 722.2, 723.6 and 721.9 cm⁻¹ for complex AO-2, AO-3 and AO-5 respectively, Fig. 2. The band at 839.9 cm⁻¹ band is affected upon coordination in the opposite way. An increase in wavenumbers is observed at 852.2, 845.5 and 844.8 cm⁻¹ for complexes AO-2, AO-3 and AO-5 respectively. The bands in the region 1135.8 and 1215.8 cm⁻¹ are referred to in-plane hydrogen deformation motions or possibly ring vibrations. There has been a shift to 1142.6 and 1197.9 cm⁻¹ in AO-2 and to 1145.9 cm⁻¹ in AO-5.

Close

Fig. 2

FTIR spectra of [Co(phen)₂Cl₂].1.5 CH₃CN (AO-2), [Co(phen)(H₂O)₄]SO₄·2H₂O (AO-3) and [Co(phen)₂CO₃].7H₂O (AO-5).

Export to PPT

Most intense and characteristic bands in the vibrational spectrum of phenanthroline occur in the range 1400–1600 cm⁻¹. All of these bands involve C⚌C and C⚌N stretching vibrations, and upon co-ordination move to higher frequencies. The strong peak at 1416.8 cm⁻¹ due to in plane antisymmetric ring deformation involving C⚌C and C⚌N stretching modes, slightly increases upon coordination to 1421.4 AO-2, 1423.1 (AO-3) and 1424.5 cm⁻¹ (AO-5). While a prominent increase in wavenumber by 13–20 cm⁻¹ occurs at 1498.0 cm⁻¹ for the symmetric ring stretching mode peak to 1513.2 (AO-2), 1518.5 (AO-3) and 1510.9 cm⁻¹ (AO-5). This is expected as coordination takes place at nitrogen atom of the C⚌N (phenanthroline ring) groups to cobalt(II) ion, and the effect is transmitted throughout the region resulting into the readjustment of electron density (Mashaly et al., 2005).

The strong and very broad bands spread from 2600 to 3500 cm⁻¹ and centered at 3116.56 and 3069.87 cm⁻¹, are due to OH stretching vibrations of coordinating and crystalline H₂O molecules in AO-3 and of crystalline H₂O in AO-5 (El-Sherif et al., 2012; Masoud et al., 2012). In the lower wavenumber region, the weak bands observed for the three complexes at 423 cm⁻¹ have been assigned to ν(M—N) vibrations while the weak bands observed at 600.0 and 571.0 cm⁻¹ have been assigned to ν(M—O) of AO-3 and AO-5 respectively (Ali et al., 2010).

Conclusive evidence regarding bonding of two oxygen atoms of carbonate ligand to cobalt ions in AO-5 is the considerable change in IR peak position of carbonate to lower wave number upon complexation with metal ion. The O—C⚌O stretching vibrations at 1755 cm⁻¹ and 1440 cm⁻¹ of sodium carbonates are shifted to lower frequencies 1510.92 and 1333.5 cm⁻¹ in the complexes. Finally, the very strong peak at 1142.6 cm⁻¹ confirm the presence of sulfate counter ion in complex AO-3, Fig. 2 (Miller and Wilkins, 1952).

3.2 Antibacterial activities

The antimicrobial activities of the compounds AO-A, AO-B, AO-C, AO-2, AO-3 and AO-5 are tested against both Gram-negative and Gram-positive bacteria with similar concentration in DMSO. The inhibition zone against the growth of the microorganisms by the compounds AO-A, AO-B, AO-C, AO-2, AO-3 and AO-5 are given in Table 1, Fig. 3. It is confirmed that among the tested compounds AO-A, AO-C and AO-5 exhibit the most antimicrobial activities toward all microorganisms, and are little less than the reference antibiotic ampicillin. The tested compounds AO-B and AO-3 revealed good antimicrobial activity toward the examined microorganisms while compound AO-2 reveal the lowest antimicrobial activity all over the study.

Close

Fig. 3

The inhibition zone diameter (mm) of various compounds and reference ampicillin at the same concentration for different gram-negative and gram positive-bacteria.

Export to PPT

Comparison of minimum inhibitory concentration MICs (in μg/mL) of compounds and the standard drug ampicillin against susceptible Gram-negative and Gram-positive bacterial strains are represented in Table 2. It was found that compound AO-A, AO-C and AO-5 exhibited the highest inhibitory activity against all bacteria with MIC ranging from 15 to 35 μg/mL. Compounds AO-B and AO-3 had good inhibitory activity against all strains (MIC: 30–55 μg/mL). On the other hand, compound AO-2 has low inhibitory activity toward the tested bacterial strains (MIC: 55–70 μg/mL). This suggests that the antibacterial activity of compounds is dependent on the basic structure as well as the nature of the ligands and counterions of the formed complex. In general, it can be also inferred that the complexes have more inhibitory activity towards gram negative bacteria than gram positive bacteria.

MIC was also calculated in μmol/ml (Table 2 and Fig. 4) to compare antibacterial activity with respect to stoichiometric relation of moles for different complexes and their metal ion and ligands.

Close

Fig. 4

Histogram of MIC (μmol/ml) of compounds against bacteria species.

Export to PPT

It was found that MIC in (μmol/l) of the complexes are lower than the corresponding starting metal salts AO-B and AO-C. Indicating enhancement of antibacterial activity of complexes due to incorporation of phenanthroline ligand in their structures. AO-5 being the most potent antibacterial agent with MIC (μmol/l) value close to that of ampicillin. It was found that MIC in μmol/L for complex AO-5 for example is smaller than its corresponding phenanthroline ligand AO-A and metal ion salt cobalt chloride AO-B.

3.3 Antioxidant activity

The production of free radicals, reactive oxygen species (ROS) and reactive nitrogen species (RNS) is inevitable during oxidative metabolic process. High concentrations of these reactive molecules in the cells induce denaturation and deterioration of many cellular components. This results in cellular oxidative stress and consequently cancer.

DPPH free radical scavenging activity: In a typical antioxidant activity study, the radical scavenging activity (RSA) of compounds can be rapidly screened by evaluation of DPPH^• RSA. The compounds showed antiradical activity by inhibiting DPPH radical (Table 3). Most of analyzed compounds revealed high to moderate interaction with DPPH radical at a concentration of 2 μg/mL. The maximum antioxidant activity was found in compounds in the following order AO-A > AO-C > AO-5 with inhibitions greater than 50% and is comparable to standard vitamin C (78.6 ± 0.88) while other compounds showed moderate RSA in the order AO-B > AO-3 > AO-2, Table 3. The antioxidant activity of these compounds is caused by their electron donating ability to DPPH radical.

SOD mimetic scavenging catalytic activity: Based on DNA binding capability of the compounds AO-A, AO-B, AO-C, AO-2, AO-3 and AO-5 compounds, it is beneficial to study the antioxidant activity of these compounds. Superoxide radical anions (O₂^•−) generates dynamic free radicals that react with biomolecules and cause cell deterioration. In this colorimetric method, the reduced NBT to blue formazan was utilized as a marker of O2^•− generation and detected at 560 nm (Youssef et al., 2016). SOD - mimetic superoxide scavenging catalytic activity by the tested compounds was detected by their inhibition of reduction of NBT to formazan by decreasing the superoxide ion concentration. The tested compounds compete with NBT for oxidation of the generated superoxide ions. The tested compounds AO-A, AO-C and AO-5 exhibited the greatest antioxidant activities with % inhibition of 68.2 ± 0.71, 65.1 ± 0.66 and 63.7 ± 0.84 respectively. While compounds AO-B and AO-3 displayed good SOD like activity within % inhibition greater than 50% as shown in Table 3. Horse radish (82.4 ± 0.57) was evaluated as standard for comparison.

ABTS radical cation decolorization assay: The antioxidant activity is based on color intensity reduction of free radical cation at 734 nm from ABTS solution due to radical scavenging by antioxidant material. The importance of ABTS derived free radical method is that the reaction is stoichiometric and the color is stable for more than 1 h (Chyong et al., 2011). The results obtained in Table 3 with the control (77.3% ± 0.89) indicated that the antioxidant power of the tested compounds was high in accordance with DPPH and SOD results. The tested compounds AO-A, AO-C and AO-5 showed >50% inhibition of ABTS radical cation, while AO-B, AO-2 and AO-3 had weak scavenging activities with inhibition of ABTS less than 50%, Table 3.

Trends in antioxidant activity: AO5 showed greater antioxidant activity than other complexes. This can be explained by the ease of Co(II) in AO5 to undergo oxidation to Co(III) and providing an electron to the free radical and thus causing its reduction as observed in DPPH, SOD and ABTS radical scavenging catalytic activity. The produced [Co(III)(phen)₂CO₃] ⁺ complex is stable (Omer et al., 2010) and the antioxidant activity of AO5 can thus be explained by its behavior as a reducing agent according to the following reaction: [Co(II)(phen)₂CO₃] (AO5) ⇌ [Co(III)(phen)₂CO₃]⁺ + e –

3.4 DNA affinity and DNA cleavage ability

Agarose gel electrophoresis was used to study the ability of compounds to cleave genomic DNA as one possible mechanism of interaction. The tested compounds AO-A, AO-C, and AO-5 displayed DNA degradation effect in a concentration dependent manner due to their binding affinity to DNA. The cleavage of DNA by the compounds AO-A, AO-B, AO-C, AO-2, AO-3 and AO-5 was found to increase with increase in concentrations from 2, 4, 6 to 8 µM (Fig. 5). The results indicated that at 2 µM all compounds caused negligible degradation of DNA. Raising the concentration of compounds AO-A, AO-C and AO-5 to 4 µM showed noticeable degradation effect on DNA. The tested compounds AO-B and AO-3 caused an efficient cleavage ability of DNA at 6 µM concentration. Furthermore, all compounds AO-A, AO-B, AO-C, AO-2, AO-3 and AO-5 demonstrated a good cleavage power of DNA at a concentration of 8 µM (Fig. 5).

Close

Fig. 5

A figure showing the degradation effect of 2, 4, 6 and 8 µM of compounds AO-A, AO-B, AO-C, AO-2, AO-3 and AO-5 (Lanes 3–8) on the genomic DNA isolated from E. coli Lane1 E. coli DNA; lane 2 E. coli DNA + DMSO.

Export to PPT

Although we expected that Cobalt (II) phenanthroline aqua complex (AO3) (Fig. 1) to have the highest binding effect to DNA moieties, because its labile water ligands can be lost and provide an empty d- orbital in the Lewis acid cobalt (II) ion. The ion can bind to electron rich nucleic base moieties in DNA and causes its cleavage. However, since Cobalt (II) phenanthroline carbonates (AO5) (Fig. 1) showed the highest cleavage power of DNA, the cleavage mechanism can be due to proper geometry of two phenanthroline ligands in complex (AO5) which causes them to interact strongly with heterocylic aromatic ring of DNA through π–π interactions.

3.5 Cytotoxicity evaluation

Hepatocellular carcinoma (HEPG-2), Mammary gland (MCF-7) and Colorectal carcinoma (HCT-116) cells (Table 4) were used in cytotoxicity test to evaluate anti-proliferative activities of the ligands and metal salts (AO-A, AO-C, AO-B), and the complexes (AO-2, AO-3, AO-5). Doxorubicin was utilized as a standard anticancer agent since it gives a very strong anticancer activity. As illustrated in Table 4, phenanthroline ligand (AO-A) has high anticancer activities against cell lines HePG-2, HCT-116 and MCF-7 with IC50 values of 17.82 ± 0.5, 14.65 ± 0.7 and 19.94 ± 0.6 (µM), respectively. The cobalt sulfates salt AO-C and cobalt phenanthroline carbonate complex AO-5 exhibited moderate anticancer activities against all studied cell lines HePG-2, HCT-116 and MCF-7. The salt cobalt dichloride AO-B, cobalt (II) phenanthroline chloride acetonitrile (AO-2) and cobalt(II) phennthroline aqua sulfate (AO-3) displayed weak anticancer activities against all cell lines, Table 4. Although, it is expected that the complexes (AO2) and (AO3) to show noticeable cytotoxicity because they contain labile ligands chloride and water respectively. The loss of mobile ligand can poromote the formation of empty orbital in cobalt (II) that can bind to DNA nucleic bases. It could be that other factors impede biological activity of these two complexes such as dispersion and stability in physiological medium as well as cell penetration and targeting tumor cells. The activity can be improved by adding an agent that can noncovalently bond to complexes and help in maintaining their suspension and stability in cell media and increase their targeting and penetration ability of tumor cells. Cobalt phenanthroline aqua complex (AO5) shows a better cytotoxicity may be because it is more stable and more soluble in the tested medium and can thus reach easily and interact with the target DNA in the anticancer cell.

3.6 Solvent effect on electronic absorption spectra

The maximum absorption peak location Y ( λ max, nm) of phenanthroline cobalt complexes in a specific solvent (Fig. 6) has been estimated by the following multi-parameter equation (Eq. (1)) (Hassan et al., 2006):

Close

Fig. 6

UV Absorption spectra of [Co(phen)₂CO₃].7H₂O (AO-5) in different solvents. Absorbance vs. wavelength (nm).

Export to PPT

Equation (Eq. (1)) is solved for the intercept a0 and the coefficient ai using multiple regression techniques. Y is used as the dependent variable, while the solvent interaction mechanisms are taken as the independent variables Xi, and designated as E, K, M and N with equation (Eqs. (2)–(5)) listed in (Table 5). E (Eq. (2)) is the empirical solvent polarity and depends on both solute–solvent hydrogen bonding and dipolar interactions. K (Eq. (3)) is a measure of the polarity of the solvent and depends on the solvent dielectric constant D (Eq. (4)). While M (Eq. (5)) depends on the solvent refractive index n and estimates solute permanent dipole–solvent induced dipole interactions. N is a measure of permanent dipole–permanent dipole interactions. A multiple regression analysis has been performed using SPSS 23 software. In each case fits are obtained as a function of one parameter, two parameters, or three parameters. The results are presented in Tables 6–8.

All the cobalt (II) complexes studied contain phenanthroline ligand, their UV spectra were recorded in different solvents Fig. 6. The λmax at 269 nm for all three complexes shifts to longer wavelengths with increase in the values of parameters E and D i.e. increase in hydrogen bonding ability and polarity of the solvent. This is due to the more stabilization of the exited state than that of the ground state. Thus, the energy difference between excited and unexcited states is slightly reduced, resulting in a small red shift which suggests that the obtained absorption peaks are caused by π–π* electronic transition of phenanthroline ligand.

Based on the obtained values of correlation coefficients for E, K, M or N for the studied one parameter equation, there is strong dependence of the shift in λmax of AO-2 on M and to lesser extent on K for both complexes AO-2 and AO-3 (Table 6 and 7). This indicates that the refractive index of the solvent has greater effect on the absorption peak position than the solvent dielectric constant while λmax of AO-5 shows strong dependence on E (Table 8).

There is improvement in the fit in going from one parameter, two, three to four parameters equation since the multiple correlation coefficients MCC are getting more positive.

The MCC coefficients are good for AO-2 and AO-3 complexes while the coefficients for AO-5 are poor for one parameters equations. When using two parameters (E, N) equation the correlation obtained is 0.941, indicating the great influence of solvent empirical polarity and permanent dipoles-dipoles interactions on λ max. Acceptable correlation for AO-5 are observed only in going to three and four parameters equations. Tables 6–8 also lists the coefficients for the multi-parameters regression analysis when using three-parameter equations E, K, M or N parameters.

The value of a0 is the intercept. The solvent effects due to E, K, M or N parameters produce bathochromic shift for positive values of coefficients ai, and hypsochromic shift for negative values of the coefficients.