Metal Complex Derivatives of Azole : a Study on Their Synthesis , Characterization , and Antibacterial and Antifungal Activities

Four new zinc complex derivatives of azoles and ligands were synthesized and isolated as white air-stable solids and characterized by elemental analyses, thermogravimetric analysis (TGA), infrared spectroscopy, nuclear magnetic resonance (NMR) and mass spectra. The elemental analysis, theoretical calculations and NMR show that the complexes likely have a 1:1 (M:L) stoichiometry and tetrahedral geometry. To evaluate the biological activity of the complexes and to discuss the role of metal ions and structural properties, the ligands and their metal complexes have been studied. Their antimicrobial activity was determined in vitro by agar-well diffusion and broth microdilution against nine bacterial strains and seven fungal strains with clinical relevance. In vitro assays showed that the complexes exhibited moderate antibacterial and/or antifungal activities. The antimicrobial activity was found to be more active for the metal complexes than the ligands. The metal complexes that contained copper and cobalt, respectively, displayed notable antibacterial and antifungal effects against all the tested bacterial strains. The minimum inhibitory concentration 50 (MIC50) values were in the range 2454-0.7 μg mL. Metal complexes were more effective at inhibiting bacteria than fungi. The results could provide a high-potential solution for antimicrobial growth resistance, for both bacteria and fungi.


Introduction
Microorganisms, such as bacteria, molds and yeasts that are associated with the surrounding environment, may be related to the development of severe infections in humans. 1 Bacteria such as Staphylococcus aureus, Escherichia coli and Pseudomonas aeruginosa are the most common pathogens associated with hospital and community-acquired infections, and they are responsible for diseases such as urinary tract infections, pneumonia and bloodstream infections. 2 Fungal infections are also responsible for high rates of mortality and morbidity, primarily in immunosuppressed patients. 3Besides providing resistance against some fungal and bacterial strains, a limited number of compounds are available for the treatment of fungal and bacterial infections. 4In this context, azole derivatives have displayed antifungal and antibacterial activity.This activity is related to their ability to bind readily to enzymes and receptors in biological systems; in fungi, azoles bind to the heme cofactor that is located in the active site of cytochrome P450 14α-demethylase, and they inhibit the conversion pathway from lanosterol to ergosterol, an essential step in the synthesis of the fungal cell membrane.][18][19][20] The increased antimicrobial activity is likely related to azole's better solubility, bioavailability and interaction with DNA (deoxyribonucleic acid) through intermolecular associations. 19,200][21][22][23] With the aim of developing new antifungal and antibacterial compounds, we report the synthesis, characterization, and antibacterial and antifungal activities of metal complex derivatives of azoles.Four new zinc complexes were tested along with another two additional and previously reported complexes that contain copper and cobalt.Their structural properties in terms of donor groups, size, and metallic ions along with their antibacterial and antifungal activities are discussed, together with the interpretation of possible mechanisms that might be responsible for inhibition.A detailed characterization of each complex is provided.

Synthesis of ligands
Synthesis of 1,3-bis(indazol-1-ylmethyl)benzene (7)   In a Schlenk tube equipped with a reflux condenser, 1,3-bis(bromomethyl)benzene (1.20 mmol; 317.9 mg), 1H-indazole (2.61 mmol; 308.5 mg) and toluene (15 mL) were added and the mixture was heated and refluxed for 72 h.The solvent was evaporated, water was added to the solid residue, and then it was neutralized with a 10% sodium bicarbonate solution.The aqueous mixture was extracted with 3-10 mL volumes of dichloromethane.The organic layer was dried with magnesium sulfate, and the solution was concentrated and cooled at −5 °C for 24 h.The pure compound was obtained by crystallization from dichloromethane-pentane and washed with n-pentane and n-hexane.Yield 324.

Synthesis of dichloro[1,3-bis(indazol-1-ylmethyl)benzene-NN]zinc(II) (14)
A solution of 1,3-bis(indazol-1-ylmethyl)benzene (7) (0.181 mmol; 61.2 mg) in 5mL of ethanol was added to a solution of zinc(II) chloride (0.177 mmol; 24.2 mg) in 2 mL of ethanol.A white solid precipitate was produced immediately, but the reaction mixture was stirred for 30 min (rt).The white solid was filtered off and washed with ethanol and diethyl ether.Yield: 37.0 mg (44%); mp: > 300 °C; IR (KBr) ν / cm - Fusarium solani (FMR 4591), were used during the in vitro susceptibility test and were obtained from the Laboratory of Mycology and Phytopathology, at Universidad de los Andes, Bogotá D.C., Colombia, with the exception of the Fusarium species, which were kindly provided by the Microbiology Unit of the Universitat Rovira i Virgili.Bacteria were subcultured on nutrient agar (NA) (Oxoid, UK), yeasts were cultured on Sabouraud dextrose agar (SDA) (Oxoid) and molds were grown on potato dextrose agar (PDA) (Oxoid) at 37 °C for yeast and bacteria, and at 25 °C for molds to assess the purity and viability of the isolates.All the strains used here were previously characterized, both genotypically and phenotypically.

Antibacterial and antifungal activities of metal salts, ligands and their metal complexes
All the metal salts, ligands and their respective metal(II) complexes were tested against the strains by agar-well diffusion method and broth microdilution method.8][29] Metal complexes, free ligands and metal salts were dissolved in dimethyl sulfoxide (DMSO) to a final concentration of 100 mM.Bacterial and fungal inoculum containing approximately 10 6 colony-forming units (CFU mL -1 ) or conidia were spread on the surface of Mueller-Hinton agar plates with a sterile cotton swab.Wells (5 mm in diameter) were cut in the media with a sterile metallic tube, and 20 μL of each sample were placed in each well.
The plates were incubated at 37 °C (bacteria and yeasts) or 25 °C (molds) for 24 to 48 h.DMSO (1%) was selected as the negative control to determine if it played any participating role in the biological screening.Gentamicin (24 μg mL -1 ) and fluconazole (15 μg mL -1 ) were used as the standard reference drugs.The antibacterial and antifungal activity was determined by measuring the diameter of the inhibition zone around each well.The results were calculated as the means of three replicates.
1][32] Inocula containing approximately 10 4 CFU or conidia mL -1 were incubated with serial dilutions of the compounds in Luria-Bertani broth for bacteria or RPMI 1640 medium for fungi in 96-well plates.The plates were incubated at 37 °C for 6, 24 and 48 h for bacteria, yeasts and molds, respectively.After the incubation time, the absorbance at 600 nm for bacteria and 630 nm for fungi were read in a microtiter plate spectrophotometer.Gentamicin (5 μg mL -1 ) and fluconazole (3 μg mL -1 ) were used as the standard reference drugs.

Minimum inhibitory concentration (MIC)
Compounds exhibiting high antibacterial or antifungal activities were selected for MIC studies.1][32] Serial dilutions were performed from 2454 to 0.7 μg mL -1 to determinate the concentration of the compound that inhibited 50% of the microorganism's growth.

Results and Discussion
Schemes 2-4 show the azole ligands and complexes under study.

Synthesis
The synthesis of (3) was performed according to the method described by Juliá et al. 33 with some modifications.We found that the synthesis was better when the mixture reaction was initiated with the triazol salt.For this reason, a triazol sodium derivative was obtained and then added to the reaction mixture together with tetrabutylammonium bromide and CH 2 Cl 2 .The ligand (7) was synthesized by the reaction between 1H-indazole and 1,3-bis(bromomethyl) benzene in toluene.The deprotonation of the 1H-indazole with KOH or NaOH did not produce good results because it yielded an unidentified yellow oil.
We reported the reaction of 2,6-bis(bromomethyl) pyridine with 1H-indazole.When the reaction was performed with KOH, it produced a mixture of two regioisomers at a low yield (10%), which corresponded to the N1 and N2 disubstituted indazoles.The electron pair that is located at N2 becomes more reactive than the corresponding one at N1. 34,35 Compound (7) was isolated as a white solid with a 79% yield by crystallization from dichloromethane-pentane.The ligand was characterized by NMR using DEPT-135 for the identification of the methylene group at 56.06 ppm.The 13 C and 1 H chemical shifts were assigned with the aid of a heteronuclear single quantum coherence (HSQC) experiment.The ligands (2-7) exhibit important chelating properties, as evidenced by the immediate precipitation of the complexes when the ligand is mixed with Zn 2+ salt to form complexes (9-14).All the zinc complexes are non-hygroscopic, white solids, and they are air-stable at room temperature.These complexes show very low solubility in organic solvents but they are soluble in dimethylformamide (DMF) and DMSO.

Elemental analysis
The spectroscopy, analytical data and theoretical calculations reported here are consistent with the proposed formulation.Table 1 shows the elemental analysis, melting point, and ligand and complex colors.These results allowed us to propose the 1:1 (M:L) relative rate in the complexes (9-16) and their general form, M(L)(Cl) 2 .Complex (8) had a 2:1 (M:L) relative rate, resulting in an M(L) 2 (Cl) 2 type.
Ligands (2-7) coordinate in a bidentate manner to the metals and (1) coordinates in a monodentate manner.7][38] Melting points above 190 °C for these complexes reveal their high stability.

Infrared spectroscopy
A vibrational analysis in which ligands were compared with their complexes showed that the coordination of the ligand to the metal causes an absence or shift in the bands to higher wave numbers.Those results indicate the greater rigidity of the chemical bonds in the complex (Table 2).
The zone between 3200 and 2600 cm -1 in (1) (Scheme 2) shows intensive peaks from the intermolecular N−H hydrogen bonds in the free ligand.In complex (8), this hydrogen interaction is not possible because the metal is coordinated via N2 from the pyrazole.The shift to a longer frequency in N−H is conferred by the increased bond strength from the loss of the hydrogen interaction.The presence of the N−H band indicates the coordination of the metal through the N2 of the azole. 39The formation of ligands (2-7) is confirmed by the complete absence of N−H stretching in the corresponding azole.In (9) (Scheme 3), the metal-ligand interaction results in the absence of the N−C band (1353 cm -1 ) assigned to the pyrazoles in the free ligand, indicating coordination.The aliphatic asymmetric C−H vibration in free ligand (3) (3115 cm -1 ) appears to shift, and in complex (10), it was observed at 3120 cm -1 . 18,40t was also possible to observe the shift and the decreased intensity of the band at 1494 cm -1 as assigned to the N−C and C=C of the pyridine ring in (5), which appear in (12)  at 1500 cm -1 . 41,42This shift indicates a moderate interaction between the metal center and the pyridinic nitrogen of the ligand (Scheme 2).In the case of ( 6) and ( 13), the band that assigned C=C to the aromatic toluene ring was found in both the ligands and complexes; it indicates that there was no Zn-toluene interaction.In addition to complexes (8-14), weak absorption bands in the 450-550 cm -1 ranges could be assigned to Zn−N vibration, 18,43 or Zn−Cl vibration if we consider the range described by Socrates. 42

Thermal analysis
The stages of decomposition, temperature ranges and decomposition products as well as the weight loss percentages of the complexes are given in Table 3.All the results presented in the Table 3 are proposals of the probable mass losses because the detection was not possible.Figure 1 shows the TGA curve of complexes ( 12) and (14) as a representative example.All the TGA and derivative thermogravimetric analysis (DTG) are shown in the Supplementary Information section.All the complexes studied here showed degradation by steps with corresponding losses of neutral ligand fragments.In most cases, the complexes were very stable, resulting in a substantial percentage of residual mass after heating to 700 °C.Complex (10) is especially stable, with a percent loss of only 29.0%.By contrast, compound (12)  exhibits near-complete decomposition, leaving a residual mass of only 3.1%.
In addition, the probable loss of one of the ligands for compound (8) is observed (3,5-dimethylpirazole) and also two HCl molecules at 330 °C, corresponding to 52.02% (calculated 51.45%).The losses of HCl in similar complexes were previously reported by Szécsényi et al. 44 Finally the decomposition of the other ligand, leaving a residual mass of 11.5%.
An initial loss of 4.47% is observed in complex ( 9), which could correspond to the elimination of the methane molecule from the compound, possibly with a methylene bridge connecting the two pyrazoles (calculated mass 4.42%).Finally, at approximately 421 °C, a loss of another organic residue is produced (74.02%), leaving only the metal (calculated 79.92%).Compound (10) only presents slight decomposition, losing a fragment that perhaps is related to a methylene group (4.47%), HCl (13.07%) and nitrogen (12.61%).The values that were calculated for these losses are 4.89, 13.38 and 11.87%, respectively.The TG and DTG curves of the complex ( 12) and ( 13) ligand fragment losses are observed, but there was never a complete degradation of the complexes to stabilize the metal with one of the azoles.However, compound ( 12) was the only one that showed a degradation of nearly 100%, and it even lost about 10% of the total metal mass.In this complex, 3 stages of decomposition were observed.The first was treated from 300-396 °C and experienced a mass loss of 26.2%, which corresponded probably to the decomposition of the ligand that released a benzotriazole fragment and a methylene bridge (calculated mass, 27.4%).Subsequently, it lost 70.7% of its total mass from 396-700 °C, indicating a multistage decomposition without the generation of a stable intermediary from the remaining fragments (calculated mass, 72.6%).

Molecular modeling
Computational calculations were performed to propose a probable structure for complexes (12) and ( 13), because these two complexes have different coordination positions with the metal.After seeing the results of the elemental analysis, it is possible to propose a 1:1 stoichiometric relation between the ligand and metal to start the calculation.The results for (12) are presented in Figure 2.These calculations show that complex (12c) is the more favorable structure with the lowest energy of the three possibilities.In this structure, (5) stabilizes the metal as a bidentate ligand in a tetrahedral structure by using the pyridinic nitrogen and the N2 of the benzotriazole moiety.Compounds 12a, 12b and 12c have lower energy compared with the free ligand, Zn 2+ and separated chloride anion (−4296.3699eV).Compound 12c is energetically favored over 12a because the tetrahedral coordination structure is the most stable for this type of complex, and it is also favored over 12b for showing less steric impediment.
Finally, the vibrational frequencies presented in 12c as the minimum potential energy surfaces are positive.In the FTIR spectra, it was possible to find some experimental results that support the proposed structure.A similar analysis was performed in (13) because it is possible to obtain different structures of this complex.Different possibilities were studied, and they are shown in Figure 3.In this case, 13a is more stable than 13b in a vacuum.The total energy of free ligand 6, Zn 2+ and Cl − is −4403.0886eV.The two structures are more stable than the free compounds, and structure 13a is favored because it has fewer steric impediments in comparison with 13b.With this information, it was possible to propose 13a as the probable structure of ( 13).

Biological activity
The biological activity of the metal salts, ligands and metal complexes were evaluated against all the strains in three replicates.Combinations of ligands and metal salts did not exhibit antimicrobial activity.The in vitro biological screening results are summarized in Table 4. Figure 4 shows the inhibition zones of the complexes (8, 15 and 16) as a representative example.

Antibacterial activity
The antibacterial activity of the ligand and its complexes were studied against Staphylococcus aureus, methicillin-resistant Staphylococcus aureus (MRSA), Enterococcus faecalis, Bacillus cereus, Escherichia coli, Pseudomonas aeruginosa, Enterobacter aerogenes, Salmonella typhimurium and Shigella flexneri.The metal salts, ligands, metal complexes, standard gentamicin drug and DMSO solvent control were screened separately for their antibacterial activity at concentrations of 100 mM, except the standard drug, which was tested at a concentration of 24 μg mL -1 (Figures 5 and 6).These effects were dosedependent.The antibacterial results suggested that all the synthesized metal complexes were effective in at least one of the strains used here, including the four new zinc complexes (11-14).This finding is likely related to the better solubility, bioavailability, and interaction with DNA through intermolecular associations. 23,45Also, it is possible that an increase in the lipophilicity of the complexes reduces the permeability barrier of the cells and slows the normal cellular processes of the microorganisms, resulting in increased antimicrobial activity. 36The tested gram-positive strain was susceptible to all the metal complexes (Figure 5).However, some of the compounds in this series (9) were not effective against any gram-negative bacteria (Figure 6).Other studies have shown that the pyrazoles in complex (9) inhibit the DNA gyrase and topoisomerase IV in bacteria but are more efficient against gram-positive than gramnegative bacteria. 36The presence of different metal ions in the complexes varied with the activity.In compounds (9, 15 and 16), only changes in the metal ion or the presence of copper and cobalt showed inhibitory activity against all bacterial strains.However, complex (9) only inhibited four strains of bacteria, namely those of S. aureus, MRSA, B. cereus, and S. flexneri.Compounds (10) and (13) were inactive against E. faecalis but were effective against the remaining bacteria strains, likely because of their abilities to diffuse into the single-cell membrane of the grampositive bacteria, as opposed to the gram-negative bacteria, which have two membranes.The outer membrane of gram-negative bacteria possesses a negative charge, which confers resistance to hydrophobic compounds and therefore makes the membrane more stable, with greater integrity. 46  The difference between the new zinc complexes lies in the ligand.In complexes ( 12) and ( 13), the benzotriazole in their structures showed increased activity.Benzotriazole derivatives have been shown to be able to induce the alteration of mitochondrial membranes in microorganisms. 47s the data analysis further indicated, complex ( 14) was very similar in its activity to complex (11), indicating that the more extended conjugation of the indazole rings on the metal center does not increase the antibacterial activity.Complexes ( 9) and ( 11) differ because of the presence of toluene or methylene in the center (Scheme 3).However, the activity in (11) was effective against eight strains, but for complex (9), it was only effective against four (Figure 5).This finding is most likely explained by the presence of  The concentration of each compound was similar to that used in Figure 5 and Table 4.
toluene in the molecule.Toluene is known to cause an increase in the membrane fluidity and therefore causes an increase in its permeability. 48,49In the case of complexes ( 8) and ( 9), (8) had activity against all the strains and (9) only showed activity against four.1][52] These bonds favor interactions with DNA through intermolecular associations as it bridges hydrogen, which in turn results in increased antibacterial activity. 12,14tifungal activity The antifungal activities of the ligand and its complexes were studied against C. albicans, C. krusei, C. parapsilosis, A. flavus, A. fumigatus, F. oxysporum, and F. solani.The metal salts, ligands, metal complexes, the standard drug fluconazole, and a DMSO solvent control were screened separately for their antibacterial activities at concentrations of 100 mM, except the standard drug, which was tested at 15 μg mL -1 (Figure 7).As reported in bacteria, the inhibition effects were dose-dependent.
Fungal species were more resistant to treatments with the new complexes.However, complex (12) showed activity against three fungi strains.This compound contains a benzotriazole group that is uncoordinated from zinc, and it can induce the alteration of mitochondrial membranes in the microorganisms or may inhibit some of the many steps leading to ergosterol synthesis. 14Compound (9) only inhibited A. fumigatus.Candida albicans and C. krusei have both shown resistance to azoles, 14 but they were inhibited by complexes (12, 15 and 16).Mishra et al. 53 reported that cobalt, copper, and zinc azole complexes showed biological activity against Aspergillus niger and A. flavus, and these complexes disturb the cell respiration process and thus block protein synthesis.This blockage in turn restricts the growth of the organisms.Candida krusei, Fusarium spp., and Aspergillus spp.were not inhibited by the standard drug, that is, fluconazole.However, A. fumigatus and C. krusei were highly inhibited by compound (15).Fusarium spp.were not inhibited by any complex.Fusarium species are well known to be highly resistant to a variety of antifungals, including azoles, particularly fluconazole, 54 and these strains are also capable of resisting high concentrations of metals such as zinc. 55,56he complexes that contained copper and cobalt showed more antifungal activity than the zinc complexes.The tested fungi appear to be zinc-tolerant.Zinc is an essential nutrient for fungi, but it is toxic at high concentrations.Therefore, the intracellular concentration of this metal is tightly regulated by zinc transporters, which could be removing zinc from the fungal cells. 57,58n general, complex (15) showed the highest antifungal effects and inhibited the growth of all tested Candida species and also A. fumigatus.Singh et al. 16 reported  4.
that the small size of Cu 2+ relative to that of other metals such as Zn 2+ and Cu 2+ reduces polarity and increases the lipophilicity of the fungal membrane, interrupting normal cellular processes and enhancing the antifungal activity of Cu 2+ complexes.

Minimum inhibitory concentration (MIC 50 )
The MIC 50 values suggest that the antimicrobial activity of complexes (8-16) were species-dependent (Table 2).The results showed low antifungal and antibacterial activities against all the evaluated strains as compared relative to that of the standard drugs gentamicin and fluconazole.The MIC 50 values of the complexes ranged from 2454-0.7 μg mL -1 and were slightly higher for gram-negative than gram-positive bacteria.Gram-positive bacteria had an MIC 50 of 26 to 2454 μg mL -1 , gram-negative bacteria had an MIC 50 of 45 to 2454 μg mL -1 , and fungi had an MIC 50 of 13 to 847 μg mL -1 .In general, S. aureus and S. flexneri were the more susceptible strains.
The highest activity of the complexes against fungi was observed for complex (16) against C. krusei, with an MIC 50 of 13 μg mL -1 , and for bacteria complex (15)  against S. aureus, MRSA and E. faecalis, with an MIC 50 value of 26 μg mL -1 .Fungal species were the most resistant, particularly the Fusarium and Aspergillus species.However, our data showed that complexes (15)  and ( 16) possess antifungal activity against C. albicans, C. parapsilosis, and the fluconazole-resistant strain of C. krusei.Complexes ( 9) and ( 15) were efficient at inhibiting the fluconazole-resistant strain of A. fumigatus.
It is clear that pyrazole, imidazole and indazole derivatives combined with metals have strong antibacterial activity against gram-positive and gram-negative bacteria, including MRSA, and also against fluconazole-resistant fungal species such as C. krusei and A. fumigatus.All the tested bacterial species were inhibited by metal complexes (8-16), and Candida albicans and C. krusei, which were both resistant to the azoles, were inhibited by complexes (12, 15 and 16). A. fumigatus was efficiently inhibited by complexes (9 and 15).

Conclusions
In summary, we have synthesized and characterized a series of air-stable complex derivatives of azole ligands.The ligand 1,3-bis(indazol-1-ylmethyl)benzene (7) and the zinc complexes (11-14) are new.The synthesized azole derivatives (2-7) act as bidentate ligands to coordinate to the metal center.Ligand bonding to metal ions was confirmed by elemental analyses, FTIR spectroscopy studies, TGA and molecular modeling.In vitro assays showed that the  complexes exhibited moderate antibacterial and antifungal activities.
The complex that contains copper was active against all the test strains, except Fusarium spp.and A. flavus.We found that the metal complexes have higher antimicrobial activity, and the ligands themselves have no effect.Metal complexes were more effective at inhibiting bacteria than fungi.These antimicrobial results indicated that structural factors such as the donor groups, sizes and the metallic ions in the complexes could significantly affect their antimicrobial activities.Thus, the agent with a broad-spectrum nature can become a new antimicrobial agent, after further biological studies.Further work in this direction is in progress.

Figure 6 .
Figure 6.Comparison of antibacterial activities.SD = standard drug; gentamicin 24 μg mL -1 .The antibacterial screening of all the synthesized compounds was performed against gram-negative Escherichia coli, Pseudomonas aeruginosa, Enterobacter aerogenes, Salmonella typhimurium and Shigella flexneri.The concentration of each compound was similar to that used in Figure5and Table4.

Figure 7 .
Figure 7.Comparison of antifungal activities.SD = standard drug; fluconazole, 15 μg mL -1 .The antifungal screening of all the synthesized compounds was performed against C. albicans, C. krusei, C. parapsilosis, A. flavus and A. fumigatus.The concentration of each compound was similar to that used in Figure 5 and Table4.

Fungi
drug, gentamicin for bacteria, fluconazole for fungi.

Table 1 .
Analytical and physical data of the ligands and complexes

Table 2 .
Infrared spectral bands for ligands and their complexes vw: very weak; w: weak; m: medium; s: strong; vs: very strong.