Metronidazole Interaction with Cu2+ and Zn2+: Speciation Study in Aqueous Solution and Biological Activity Evaluation

Metronidazole (2-methyl-5-nitro-1H-imidazole-1-ethanol, MNZ) is a well-known and widely used drug for its excellent activity against various anaerobic bacteria and protozoa. The purpose of this study is to elucidate the ability of MNZ to form metal complexes with Cu2+ and Zn2+ and to demonstrate that complexation increases its bioactivity profile against different pathogenic microorganisms. The interaction of MNZ with Cu2+ and Zn2+ was investigated in NaCl aqueous solution under different conditions of temperature (15, 25, and 37 °C) and ionic strength (0.15, 0.5, and 1 mol L–1) by potentiometric and spectrophotometric titrations. The obtained speciation models include two species for the Cu2+-containing system, namely, CuL and CuL2, and three species for the Zn2+-containing system, namely, ZnLH, ZnL, and ZnLOH. The formation constants of the species were calculated and their dependence on temperature and ionic strength evaluated. Comparison of the sequestering ability of MNZ under physiological conditions revealed a capacity toward Cu2+ higher than that toward Zn2+. A simulation under the same conditions also showed a significant percentage of the Cu2+–MNZ species. The biological assessments highlighted that the complexation of MNZ with Cu2+ has a relevant impact on the potency of the drug against two Trypanosoma spp. (i.e., T. b. brucei and T. b. rhodesiense) and one gram-(−) bacterial species (i.e., Escherichia coli). It is noteworthy that the increased potency upon complexation with Cu2+ did not result in cytotoxicity against MRC-5 human fetal lung fibroblasts and primary peritoneal mouse macrophages.


INTRODUCTION
−10 It has also been shown to have excellent antifungal properties. 11Unfortunately, over the years, its overuse has led to the development of MNZ-resistant microorganisms. 12−14 Among the various methods reported in the literature to overcome the resistance issue, 12 complexation with transition metals has proven to be one of the strategies with the greatest potential to enhance the bioactivity of MNZ 15 and protect it from degradation by gastric juice. 16The presence of three potential donor atoms (the nitrogen atom of the imidazole ring and the oxygen atoms of the primary alcohol moiety and the nitro group) makes MNZ a good candidate for the coordination with several metal cations such as Mn 2+ , Fe 3+ , Co 2+ , Ni 2+ , Cu 2+ , Zn 2+ , Cd 2+ , Hg 2+ , and Ag + . 16,17In addition, it is noteworthy that Ru(I)-based MNZ metal complexes have shown good antitumor efficacy in the treatment of lung cancer. 18Therefore, the study of the interaction between MNZ and transitional metal cations is interesting to propose new therapeutic approaches.Zn 2+ and Cu 2+ are essential trace elements for all living organisms. 19,20Both are implicated in numerous biological activities, including modulation of enzymatic and catalytic activities in several oxidation− reduction processes. 16The structural role of Zn 2+ is critical in metalloproteins, which are important in transcription and replication processes and in nucleic acid repair. 21Zn 2+ is also involved in several enzymatic activities necessary for various physiological functions, including protection of sulfhydryl groups from oxidation and inhibition of oxidative processes leading to the formation of reactive oxygen species (ROS). 22u 2+ is an essential trace element that functions as a catalytic cofactor in numerous biological processes in both bacteria and humans. 20Furthermore, it has recently been pointed out that Cu + and Cu 2+ have a significant impact on antibiotic activity. 23−26 To date, there appear to be few studies on in vitro to in vivo translation, which is attributed to the lack of knowledge of the molecular mechanisms underlying Cu 2+ −antibiotic interactions under physiological conditions. 23So far, the exact mechanism for bacterial damage caused by Cu 2+ has not been fully explained.Therefore, the aim of this study is to verify the ability of MNZ (or L, which indicates "ligand") to form complexes with Cu 2+ and Zn 2+ ions and to demonstrate that complexation allows for increased drug bioactivity.The study of the interaction between MNZ and both metal cations was carried out by potentiometric and UV-spectroscopic titrations in NaCl aqueous solutions of Cu 2+ − and Zn 2+ −MNZ complexes under different temperatures, ionic strengths, and metal/ MNZ molar ratio conditions.Measurements performed at different background salt concentrations (NaCl) and temperature values allowed the study of the dependence of stability constants on ionic strength and temperature.The sequestering ability of MNZ toward Zn 2+ and Cu 2+ was calculated under physiological conditions by using the pL 0.5 parameter, the cologarithm of the ligand concentration that sequesters 50% of the metal cation in trace.Moreover, a simulation under real conditions of plasma pH, temperature, equilibria, and concentrations of the main inorganic constituents is also reported.In order to test the efficacy of Zn 2+ −MNZ and Cu 2+ −MNZ species, a standard in vitro screening was carried out against a wide panel of pathogenic microorganisms, including one gram-(−) bacterial strain (Escherichia coli), one gram-(+) bacterial strain (Staphylococcus aureus), two species of fungi (Candida albicans and Aspergillus f umigatus), and various protozoa (Trypanosoma cruzi, Trypanosoma brucei brucei, Trypanosoma brucei rhodesiense, and Leishmania infantum).In addition, in vitro cytotoxicity was evaluated on human fetal lung fibroblasts (MCR-5) and primary peritoneal mouse macrophages (PMM).

Materials.
The metal cation solutions were obtained by weighing and dissolving the corresponding salt, zinc(II) chloride monohydrate (Merk, puriss.),and copper(II) sulfate pentahydrate (Fluka, puriss.).All metal solutions were standardized by titration with EDTA (ethylenediamine tetraacetic acid disodium salt, Sigma-Aldrich, purity ≥ 99%).Metronidazole was purchased from Alfa-Aesar/Thermo Fisher (purity ≥ 97%).Solutions of hydrochloric acid and sodium hydroxide used for titrations were prepared by dilution of Fluka ampules and standardized with sodium carbonate (purity ≥ 99.5%, Sigma-Aldrich) and potassium biphthalate (purity ≥ 99.5%, Sigma-Aldrich), respectively.Solutions of sodium chloride were obtained by weighing the salt (Sigma-Aldrich, puriss.),previously dried in an oven at 110 °C for at least 2 h.Distilled water (conductivity <0.1 μS cm −1 ) and grade A glassware were employed for the preparation of all solutions.
2.2.Potentiometric Apparatus and Procedure.The potentiometric titrations were performed using a Metrohm model 809 Titrando potentiometer combined with a glass electrode.Each potentiometric system was connected to a PC and the experimental titration data were acquired by the Metrohm TIAMO 2.2 software, which allows control of emf stability, titrant delivery, and data acquisition.The estimated accuracy of the apparatus is ±0.15 and ±0.002 mL for emf and for readings of titrant volume, respectively.Each titration consists of additions of volumes of NaOH standard to 25 mL of the solution containing Cu 2+ or Zn 2+ , MNZ, HCl, and the supporting electrolyte NaCl.The experimental conditions of the titrations are reported in Table 1.Glass jacket thermostated cells were employed for the measurements performed under different conditions of temperature (15 ≤ t/°C ≤ 37) by bubbling pure N 2 in order to avoid CO 2 and O 2 inside the solutions.For each measurement, an independent titration of HCl with standard NaOH was performed to calculate the standard electrode potential E 0 and the pK w value under the same experimental ionic strength and temperature conditions.

UV−Vis Apparatus.
The spectrophotometric titrations were carried out with a Varian Cary 50 UV−vis spectrophotometer equipped with an optical fiber with a fixed 1 cm path length.The instrument is interfaced to a PC by Varian Cary WinUV software.By using a Metrohm glass electrode and a Metrohm-Titrando 809 potentiometer, the couple data of absorbance (Abs) and pH vs volume of titrant (mL) were recorded simultaneously for each titration point.All titrations were performed using a thermostated cell to keep the desired temperature under N 2 to remove CO 2 and O 2 from the solutions.The experimental conditions of the titrations are reported in Table 1.For each system, at least four measurements were performed in different conditions of the metal−MNZ ratio.The UV spectra were obtained by scanning from 225 to 400 nm, and a baseline containing only HCl, NaCl, and H 2 O was recorded to consider the contribution of the matrix.

Activity against S. aureus and E. coli.
The assays were performed at 37 °C in sterile 96-well microtiter plates, each well containing 10 μL of aqueous compound together with a bacterial/fungal/MRC-5 inoculum (190 μL; 5 × 10 3 CFU/mL).The compounds were tested at 64, 16, 4, 1, and 0.25 μ-mol L −1 .Bacterial/fungal/MRC-5 growth was compared to untreated-control wells (100% growth) and mediumcontrol wells (0% cell growth).After 3 days of incubation, 50 μL of resazurin was added in each well.After 30 min at 37 °C, the fluorescence was measured at λ ex = 550 nm and λ em = 590 nm.The results are expressed as % reduction in cell growth/ viability compared to control wells, and IC 50 values were determined.
2.5.Calculations.Experimental data of potentiometric titrations were processed by using BSTAC4 and STACO4 programs.The most reliable speciation model for each system was obtained taking into account several factors, namely, the statistical parameters standard and mean deviation, the formation percentages of the species, and the simplicity of the model.The LIANA program was used in order to calculate the parameters for the dependence of formation constant values on temperature and ionic strength.More details on software employed in the refinement of the experimental data are reported in ref 27.The HySS program was used to plot the speciation diagrams.S1 and S2).Several calculation tests were carried out in order to select the best speciation model for each system, taking into account principles such as model simplicity and improved fit. 29The results evidenced that the Cu 2+ −MNZ(L) speciation pattern includes two species, CuL and CuL 2 , and three species for the Zn 2+ −MNZ(L) system, namely, ZnL, ZnLH, and ZnLOH.Formation constant values of Cu 2+ −L and Zn 2+ −L at different temperatures and ionic strength values are reported in Table 2. Unlike the Zn 2+ − L system, the complex species generated by the interaction with Cu 2+ are characterized by higher stepwise formation constant values, which lead to greater stability of the species and higher formation fractions under the same experimental conditions.More in detail, a comparison between the speciation diagrams under two different temperature values (t = 15 and 37 °C) and I = 0.15 mol L −1 of both systems is shown in Figure 2. In addition, at t = 25 °C, the ZnLH species, present in the range of 2 ≤ pH ≤ 8, reaches its maximum formation fraction, corresponding to 0.2, in the range 3 ≤ pH ≤ 7 (see Figure 2a).A notable decrease of its formation fraction is evident at t = 37 °C.The two ZnL and ZnLOH species are formed in the ranges 6.5 ≤ pH ≤ 8.5 and 7 ≤ pH ≤ 10, respectively, and reach a maximum of 0.25 at pH = 7.6 (ZnL) and 0.95 at pH = 10 (ZnLOH).Under physiological conditions (t = 37 °C, I = 0.15 mol L −1 ), a slight increase of formation fraction and a shift of the maximum fraction toward lower pH are distinguishable for both species.As shown in Figure 2b, at t = 25 °C, CuL species is formed in the range of 4 ≤ pH ≤ 8 with a maximum of 0.4 at pH = 5.8, while CuL 2 , present in the range of 5 ≤ pH ≤ 10, reaches a maximum of 0.98 at pH = 10.Both species decrease their formation fractions at t = 37 °C, also shifting the maximum formation percentage toward slightly higher pH values.Both Cu 2+ −L and Zn 2+ −L systems were also analyzed by spectrophotometric titrations to confirm the values of the formation constants and the speciation models.The spectra acquired on the solutions containing Cu 2+ −L and Zn 2+ −L at selected pH in the spectral range 225 ≤ λ ≤ 400 nm are reported in Figure 3a,b.Both systems exhibit a hyperchromic effect as a result of the pH increase.The trend of the spectra referring to solutions containing L and Cu 2+ or Zn 2+ is very similar to that of solutions containing only the ligand, as reported in the ref 30.More in detail, an intense UV band with λ max = 319 nm was observed for free MNZ and for the solutions also containing the metal cations.Moreover, the spectra reported in Figure 3a,b clearly indicate that, under the  same concentration conditions, the absorbance of solutions containing L and Cu 2+ is lower than those containing L and Zn 2+ , confirming the greater extent of the interaction with Cu 2+ compared to that with Zn 2+ .
The trend of the molar extinction coefficients vs λ is shown in Figure 4a,b.CuL and CuL 2 species reach their maximum peak corresponding to 12500 L mol −1 cm −1 at λ = 320 nm and 24000 L mol −1 cm −1 at λ = 320 nm, respectively.
For data processing, the absorption of the main Cu 2+ hydrolytic species (CuOH + ) was taken into account, with an ε trend reported in Figure S1 of the Supporting Information.ZnL, ZnLH, and ZnLOH species showed a maximum peak at λ = 325 nm, corresponding to 18500 L mol −1 cm −1 for ZnLH species, and 10000 L mol −1 cm −1 for both ZnLOH and ZnL species.The processing of the spectrophotometric experimental data allowed confirmation of both speciation models resulting from the potentiometry and the formation constants of both species relating to the Cu 2+ −L system and of ZnLOH species.The latter is the only species of the Zn 2+ −L system that reaches significant formation percentages under the low concentration conditions employed in the spectrophotometric titrations.Therefore, the formation constants of the other two species, MLH and ML, have been considered in the calculations, and the values obtained by the potentiometry have been kept constant during the process of spectrophotometric data.A comparison between the formation constants gained by potentiometry and spectrophotometry is reported in Table 3. where log β T is the stability constant at a certain temperature (expressed in kelvin), while log β θ is the stability constant at t = 298.15K, and R is the universal gas constant expressed as 8.314 J K −1 mol −1 .The values of formation enthalpy changes of all of the species of Cu 2+ −L and Zn 2+ −L systems are collected in Table 4, together with entropy and free energy values.As well-known for electrostatic interactions, it is    expected that the entropic contributes to the free energy change should be higher than the one given by enthalpy changes, due to the orientation disorder given by the solvation water molecules.This, considering the parameter values referring to partial reactions, is verified for all the species, except for ZnLH.The dependence of the formation constants on ionic strength was studied by the following Debye-Huckel-type equation: 32,33

Dependence of Formation Constant
where β 0 is the protonation constant at infinite dilution, z* is the parameter that refers to the charges of the species involved in the formation equilibrium, i.e., ∑z 2 reagents − ∑z 2 products , and C is an empirical parameter that depends on the stoichiometric coefficients and charges.The formation constants extrapolated to I = 0 mol L −1 , together with the empirical parameter C are reported in Table 5.

Sequestering Ability.
The tendency of a ligand to complex a free metal cation in solution to form metal−ligand species is expressed as the sequestering ability.To describe the sequestering ability of a ligand toward a certain metal cation, however, the mere knowledge of formation constants is not enough as it is necessary to consider all the possible interactions that both ligand and metal cation may have with all the species present in solution.In order to define it quantitatively, the empirical parameter pL 0.5 was introduced, which represents the cologarithm of the ligand concentration necessary to sequester 50% of the metal cation in traces. 34,35he calculation is based on the following Boltzmann-type sigmoidal equation: where χ is the sum of molar fractions of the different complex species and pL is the cologarithm of the total ligand concentration.
A comparison between the sequestering ability of MNZ toward Zn 2 , Cu 2+ , and Ca 2+ (one of the main metal cations present in human body and previously studied in our recent work 30 ) was evaluated under physiological conditions.According to the calculated pL 0.5 values at t = 37 °C, I = 0.15 mol L −1 , and pH plasma mean value (pH = 7.4), the sequestering ability is significantly higher toward Cu 2+ (pL 0.5 = 4.3) than Zn 2+ (pL 0.5 = 3.1) and Ca 2+ (pL 0.5 = 1.9) (Figure 5).
3.4.Simulations under Plasma Conditions.The evaluation of the formation percentages of the complex species under physiological conditions can be an intriguing topic to pursue considering that drug complexation can significantly affect cytotoxicity.MNZ(L) with the most significant electrolytes and their respective equilibria (reported in Supporting Information, Table S3) was considered under plasma conditions (t = 37 °C, 36,37 As shown in Figure 6a,b, both species of the Cu 2+ −L system reach 10% under plasma conditions, while Zn 2+ −MNZ species are not significant. 3.5.Antimicrobial and Antiprotozoal Activity.Both Cu 2+ −L and Zn 2+ −L complexes were evaluated for their biological activity profiles against an exhaustive panel of pathogenic microorganisms.Specifically, the antimicrobial activity was assessed against the gram-(−) E. coli, the gram-(+) S. aureus, the yeast C. albicans, and the fungus A. f umigatus, whereas the antiprotozoal activity was evaluated against three Trypanosoma spp.(i.e., T. cruzi, T. b. brucei, and T. b. rhodesiense) and one Leishmania sp.(L.infantum).A range of reference compounds (tamoxifen, benznidazole, miltefosine, suramine, doxycycline, flucytosine, and miconazole) were included to demonstrate the validity of the antimicrobial assays (Supporting Information, Table S4).The noncomplexed drug MNZ was used as a main comparator.Salts (chlorides) of Cu 2+ and Zn 2+ , which can exert their own toxicities, were also used as positive controls.Cytotoxicity assays were performed on both MRC-5 and PMM cells (Table 6).As compared to the pure drug, the complexation with the Zn 2+ ion did neither affect potency nor selectivity as all IC 50 values remained above the highest in-test concentration.The interesting outcome of this study came from the complexation of MNZ with the Cu 2+ ion.In this latter case, the potency of the drug was increased at least 2.6−2.9-foldagainst the T. brucei spp.(Cu 2+ −L → IC 50 = 2.14 and 1.89 μg/mL, for T. b. brucei and T. b. rhodesiense, respectively) and at least 7.8-fold against E. coli (Cu 2+ −L → IC 50 = 0.72 μg/mL).More importantly, this increase in drug potency did not affect cytotoxicity against the two healthy reference cell lines.The activity of Cu 2+ −L against other microorganisms remained unaffected as compared to the reference drug (IC 50 > highest in-test concentration), while showing intriguing pathogenselectivity probably related to different transport mechanisms within cells.

CONCLUSION
An in-depth speciation study of Cu 2+ −L and Zn 2+ −L systems in aqueous solution was carried out in order to point out the thermodynamic behavior of the species and then evaluate their antimicrobial activity.By potentiometric titrations on aqueous solutions, the speciation models and corresponding formation constants of the formed species were determined, which showed the formation of the two species CuL and CuL 2 and ZnLH, ZnL, and ZnLOH.Both models and formation constants obtained were confirmed using spectrophotometric titrations.With the aim of getting a complete thermodynamic picture, the dependence of the formation constant values on the temperature and ionic strength was also investigated.Moreover, the formation percentages of these species were evaluated by simulating the real plasma conditions of temperature, pH, and ionic strength, and over 50 equilibria between the main electrolytes present in the plasma were considered.According to the biological screening, the L complexation with Cu 2+ has a significant impact on the potency of the drug against T. b. brucei, T. b. rhodesiense, and E. coli; on the contrary, the presence of Zn 2+ does not lead to enhanced activity.
Values on the Temperature and Ionic Strength.As is well-known, temperature and ionic strength have a significant impact on formation constant values.For this reason, determining constants under different ionic strengths is crucial to simulate real fluid conditions.Starting from the knowledge of the formation constant values at different temperatures, the enthalpy change values were calculated by using the following van't Hoff equation:31

Table 2 .
Experimental Formation Constants of Zn 2+ −MNZ(L) and Cu 2+ −MNZ(L) Species Obtained by Potentiometry a ≥95% of confidence interval.constants of Cu 2+ and Zn 2+ , as well as Cu−Cl − species formation constant values, were taken into account (see Supporting Information, Tables

Table 3 .
Comparison of Experimental Formation Constants of Zn 2+ −L and Cu 2+ −L Species Obtained by Spectrophotometry and Potentiometry at t = 25 °C and I = 0.15 mol L −1

Table 5 .
Formation Constants at Infinite Dilution and Parameters for the Dependence on Ionic Strength of Cu 2+ − MNZ(L) and Zn 2+ −MNZ(L) Species at t = 25 °C in NaCl

Table S1 :
Hydrolysis constants of Zn 2+ at different temperatures and ionic strength values; TableS2: Hydrolysis constants of Cu 2+ and formation constant values of Cu 2+ -Cl − species at different temperatures and ionic strength values; FigureS1: ε vs λ of CuOH + at t = 25 °C, I = 0.15 mol L −1 ; TableS3: Formation constant values of the species considered for simulations under plasma conditions; TableS4: List of reference drug and IC 50 values of the screening panel (PDF) ■ AUTHOR INFORMATION
a Trypanosoma cruzi c h l Primary peritoneal mouse macrophages cytotoxicity.