Synthesis and characterization of 5-amino-2-((3-hydroxy-4-((3-hydroxyphenyl) phenyl) diazenyl) phenol and its Cu(II) complex – a strategy toward developing azo complexes for reduction of cytotoxicity

Abstract

A major drawback of azo compounds is their associated toxicity, often carcinogenic, which is related to the reduction of the azo bond. This study intends to re-investigate this behavior by studying 5-amino-2-((3-hydroxy-4-((3-hydroxyphenyl) phenyl) diazenyl) phenol (AHPD), a compound containing two azo bonds. Interaction of AHPD and its dimeric Cu(II) complex with bacterial strains Escherichia coli and Staphylococcus aureus revealed the complex was less toxic. Reductive cleavage of the azo bond in AHPD and the complex followed using cytochrome c reductase (a model azo-reductase) as well as azo-reductase enzymes obtained from bacterial cell extracts. Degradation of the azo bond was less in the complex allowing us to correlate the observed cytotoxicity. Cyclic voltammetry on AHPD and the complex support observations of enzyme assay experiments. These were particularly useful in realizing the formation of amines as an outcome of the reductive cleavage of azo bonds in AHPD that could not be identified through an enzyme assay. Results suggest that complex formation of azo compounds could be a means to control the formation of amines responsible for cytotoxicity. Studies carried out on bacterial cells for mere simplicity bear significance for multicellular organisms and could be important for human beings involved with the preparation and utilization of azo dyes.

• 5-amino-2-((3-hydroxy-4-((3-hydroxyphenyl) phenyl) diazenyl) phenol
• Mass spectroscopy
• Azo-bond cleavage
• Enzyme assay
• Cyclic voltammetry

1. Introduction

The azo bond enjoys a unique position in chemistry. If one were to consider a single functional moiety, it would be difficult to find another having such diverse applications [1–5]. The total number of compounds containing the azo bond is very large, not to speak of its impact on society. From a synthetic point of view these compounds are potential ligands in co-ordination chemistry [6–8]. Although interesting synthetic aspects and industrial applications of azo dyes were discussed, very few have reported their interaction with living systems [9–12]. Azo compounds are toxic [13,14], even carcinogenic, which gives rise to questions on their use [13,15]. Some efforts were made to identify the reasons of toxicity [13–15] with most researchers considering reductive cleavage of the azo bond to primary amines as being largely responsible [14–16].

Toxicity of azo compounds or of the amines they form is also an outcome of their interaction with deoxyribonucleic acid leading to disruption in replication, transcription or protein synthesis [14,16–18]. However, no single mechanism is thought to be responsible for azo toxicity and the general belief is that it is a consequence of several events [14,19–22]. Although complexes of azo compounds were prepared, few have ventured further to see if complex formation was able to modulate the associated toxicity [23]. It would definitely be better if modified forms of azo compounds (for example their complexes) could show similar utility and yet possess reduced toxicity. This could protect the work force involved in preparing such compounds and the end user [13,15,16]. This is important for we cannot do away with a lot of these compounds immediately, as they are extremely useful and also a part of our daily routine. Be it as food colorants, colorants in medicine and cosmetics, or as laboratory chemicals (either in routine analysis or in research for staining biomolecules) azo compounds have a place of their own. The cause for concern is the human element at the end, who unknowingly is under threat of established toxicity, even carcinogenicity.

5-amino-2-((3-hydroxy-4-((3-hydroxyphenyl) phenyl) diazenyl) phenol (AHPD) with two azo bonds in its structure was prepared since it represents similar azo compounds used industrially. A Cu(II) complex was prepared and a comparative study on the reductive cleavage of the azo bond and action on bacterial cells was carried out.

2. Experimental

2.1 Materials

3-aminophenol (E. Merck, India) was used to prepare AHPD. of analytical grade (E. Merck, India) was used to prepare the complex. Cytochrome c reductase isolated from porcine heart was purchased from Sigma-Aldrich, USA. NADPH was purchased from Sisco Research Laboratories, India. Sodium nitrate of analytical grade (E. Merck, India) was used to maintain the ionic strength of the medium. Tetra butyl ammonium bromide (E. Merck, India) was used as the supporting electrolyte during cyclic voltammetry experiments performed in dimethyl formamide (E. Merck, India). Dimethyl sulfoxide used as solvent was obtained from E. Merck, India. Otherwise, all experiments were performed in triple distilled water.

2.1.1 Synthesis of AHPD

Two solutions of 3-amino-phenol (0.55 g, 5 mmol) were prepared in 100 mL triple distilled water. To one solution, was added. Both were kept in ice for 25 min. The solution of 3-aminophenol not having Na₂CO₃ was treated with a cold acidified solution of NaNO₂ (0.5 g in 100 mL, 10 mmol) to diazotize it. This was then coupled with the other cold solution of 3-aminophenol treated with . A red dye formed which was recrystallized from ethanol. Yield: 80%. Anal. Calcd for (%): C, 61.89; H, 4.30; N, 20.06. Found: C, 62.24; H, 4.18; N, 20.33.

2.1.2 Synthesis of

(0.06 g, 0.25 mmol) was dissolved in 25 mL water and mixed with a solution of AHPD (0.0875 g, 0.25 mmol) in 25 mL ethanol. The mixture was refluxed for 2 h resulting in a reddish-black color. 10% sodium acetate solution was added and the mixture was refluxed for a further 2 h when a chocolate-colored compound was formed. The precipitate was filtered, washed with hot water to remove traces of Cu(II) and sodium acetate. It was re-crystallized from ethanol. Single crystals were not obtained which is why X-ray diffraction data could not be given. However, effort is on to obtain suitable single crystals. Yield: 70%. Anal. Calcd for (%): C, 51.06; H, 3.62; N, 14.89. Found: C, 51.41; H, 3.88; N, 14.65. Cu(II) present in the complex was estimated using standard procedure [24].

2.1.3 Bacterial strains

Staphylococcus aureus (MTCC 96) and Escherechia coli (MTCC 739) were used for antimicrobial activity. All strains were maintained on a nutrient agar (NA) plate and stored at 4°C. Thereafter, a single colony was picked and transferred to Muller Hinton broth and incubated for 4 h at 37°C prior to the test. Cell extracts from these strains containing live cells were used in the enzyme assay for azo-reductase activity [25,26].

2.2 Methods

2.2.1 Preparation of test inoculum

Test organisms were cultured in fresh NA plates at 37°C for 18–24 h. Thereafter, these were sub-cultured by colony suspension method in Müeller Hinton nutrient broth at 37°C for approximately 4 h at 100 rpm. All test inoculums contained  mL⁻¹. Density of all inoculums was adjusted to that of 0.5 McFarland standard.

2.2.2 Determination of MIC for AHPD and on bacterial cells by micro-plate broth dilution assay

Antibacterial screening was carried out by serial micro-plate broth dilution method to determine minimum inhibitory concentration (MIC). In the case of the micro-broth dilution technique, 5.0 μL of each bacterial suspension was added to the wells of a 96-well microtiter plate containing 50 μL of twofold serially diluted AHPD and . Control wells were prepared with culture medium containing 5% dimethyl sulfoxide (DMSO). Plates were covered and incubated (100 rpm, 37°C) for 18 h. MIC of the compounds was determined with the addition of 20 μL of a 0.2 mg mL⁻¹ p-iodonitrotetrazolium chloride (INT) and incubated in the dark at 37°C for 30 min. Viable bacteria produced a pink coloration by reducing INT to INT-formazan while colorless wells indicated no bacterial growth [27], MIC being the lowest concentration of each compound where no viability was observed after 24 h.

2.2.3 Determination of zone of inhibition by disk diffusion method

For the disk diffusion assay, 100 μL of each bacterial suspension was uniformly spread on a solid NA medium in a Petri dish. Four sterile paper disks (6 mm in diameter; Himedia) were placed on the surface of each agar plate and impregnated with 10 μL AHPD or . Plates were incubated for 18 h. Disks impregnated with 5.0% DMSO served as negative control while those with ampicillin served as positive control [27]. Inhibitory compounds create zones of inhibition in the grown microbial region. Assessment of antibacterial activity was based on the measurement of the diameter of the zone of inhibition formed around the well. The simple character of the agar-diffusion assay, its wide range for effective concentrations and compatibility with many organisms allow it to be extensively used as an efficient technique.

2.2.4 Enzyme assay for reduction of the azo bond using cytochrome c reductase

Cytochrome c reductase was dissolved in triple distilled water. Enzyme activity of the prepared stock solution of cytochrome c reductase was 1000 UL⁻¹. 0.0045 gram of NADPH was dissolved in 1 mL triple distilled water. The assay used was based on methods previously reported in the literature that used cytochrome c reductase as a model azo-reductase showing catalytic activity in the reduction of azo dyes to primary amines [10–14]. NADPH was the reducing substrate while AHPD and were electron acceptors. For AHPD, 15 μL of 10⁻³ M solution in DMSO was taken in a quartz cuvette to which 665 μL phosphate buffer was added while in case of 25 μL was taken from a stock of  M in DMSO and 655 μL phosphate buffer was added. 240 μL 0.5 M NaCl, 70 μL NADPH and 10 μL cytochrome c reductase were thereafter added to the solutions. The final assay solution (1.0 mL) was 10 U L⁻¹ in cytochrome c reductase and 444 μM in NADPH. Compound concentrations were 15 μM (AHPD) and 50 μM (). The cuvette was inverted to mix and monitored by ultraviolet–visible (UV–Vis) spectroscopy against buffer–DMSO blank. Loss in absorbance of the azo group followed at 440 nm for AHPD and 445 nm for for 90 min. Spectra for each solution were taken every 5 min.

2.2.5 Enzyme assay for in vitro reduction of the azo bond using bacterial cell extracts

Azo-reductase present in bacterial cells catalyzes the reduction of the azo bond to primary amines [14,16,28–32]. Cell extracts from E. coli and S. aureus were taken as sources of azo-reductase [28,29]. Reduction of the azo bond was monitored at the of each compound. NADPH was the reducing substrate while compounds AHPD or were electron acceptors. For AHPD, 15 μL of 10⁻³ M solution in DMSO was taken in a quartz cuvette to which 495 μL phosphate buffer was added while in case of 25 μL was used from a stock of  M in DMSO to which 485 μL phosphate buffer was added. 120 μL 0.5 M NaCl and 70 μL NADPH were then added. To initiate the reaction 300 μL cell extract containing live cells was added to the mixture in the cuvette. Final assay solutions were 1.0 mL having 444 μM NADPH and 15 μM AHPD or 50 μM respectively. Contents of the cuvette were mixed and a UV–Vis spectrum was recorded against a buffer–DMSO blank. Loss in absorbance owing to the reduction of the azo bond in the case of AHPD followed at 430 nm using cell extract obtained from S. aureus and at 436 nm for cell extract obtained from E. coli. For , reduction followed at 450 nm using cell extract from S. aureus and at 443 nm using cell extract from E. coli. All assays were done for 90 min. Spectra of each solution were taken every 2 min for the first 30 min after which they were taken every 5 min. The entire assay was carried out under de-aerated (argon saturated) conditions. The enzyme and other components of the assay were very carefully degassed separately. After this was done, using a syringe the enzyme was added to the other constituents of the reaction mixture that was already present in a degassed condition in the cuvette. Thus the reaction was initiated. Concentrations used for AHPD and were different since we wanted to have the same value for the initial absorbance for both compounds during the assay for better comparison. This provided a visible comparison of the spectra that could tell the difference in the change in absorbance of the compounds during the assay.

2.2.6 Cyclic voltammetry

Cyclic voltammetry was performed using an EG&G Potentiostat Model 263A, Princeton Applied Research using power suite software for electrochemistry. Experiments were done using the conventional three-electrode system at 298 K. A glassy carbon electrode of surface area 0.1257 cm² served as the working electrode. Ag/AgCl, KCl (saturated) was the reference electrode while a platinum wire was the counter-electrode. Electrochemical measurements were performed in a 5 mL electrochemical cell. Since experiments were done in pure dimethyl formamide, TBAB (tetra butyl ammonium bromide) was used as the electrolyte. Before each experiment solutions were de-aerated using high-purity argon for a minimum of 25 min.

2.3 Instruments used for different analyses

UV–Vis spectra were recorded on a Jasco J-630 Spectrophotometer, JASCO, Japan. Fourier transform infrared spectroscopy of solid samples (in KBr pellets) was recorded on a PerkinElmer RX-I spectrophotometer. Thermo gravimetric-differential thermal analysis was done on a Mettler Toledo thermo-gravimetric analysis (TGA)/SDTA 851 thermal analyzer. ¹H NMR of AHPD was recorded on a Bruker Avance 300 NMR spectrometer. Elemental analysis was carried out on a 2400 SERIES II CHN Analyser, PerkinElmer. Mass spectra were recorded on Micromass Q-T of micro™, Waters Corporation. Electron paramagnetic resonance (EPR) was recorded on a JEOL JES-FA 200 ESR Spectrophotometer.

3. Results and discussion

3.1 Characterization of the complex

3.1.1 ¹H NMR of AHPD

The ¹H NMR (Figure S1) of AHPD shows peaks at 14.39 ppm (1H, s, intramolecularly H-bonded phenolic ‒OH), 9.66 ppm (1H, s, intramolecularly H-bonded phenolic ‒OH), 7.35–7.22 ppm (4H, m, aromatic ‒CH ortho to ‒N˭N‒), 7.05 ppm (2H, s, aromatic ‒CH), 6.62–6.69 ppm (1H, m, aromatic ‒CH), 6.61 ppm (2H, s, aromatic ‒CH), 6.31–6.28 ppm (1H, m, aromatic ‒CH), 5.91 ppm (1H, s, phenolic ‒OH).

3.1.2 Electronic spectra of AHPD and Cu(II) complex

Spectra for AHPD and the Cu(II) complex were recorded in DMSO. AHPD showed a band in the region 410–470 nm with a peak at 439 nm while in case of the complex the spectra underwent a slight red shift to show a broad band at 420–465 nm (Figure S2).

3.1.3 IR spectra of AHPD and Cu(II) complex

The IR spectrum of AHPD (Figure S3) shows bands at 3431.28 cm⁻¹, 3344.15 cm⁻¹, 3225.81 cm⁻¹ owing to stretching of phenolic ‒OH with an evidence for intramolecular hydrogen bonding [33]. Peaks at 1651 cm⁻¹ and 1600 cm⁻¹ were characteristic for stretching of a C‒C linked to an azo group. 1508.66 cm⁻¹ was characteristic for the azo bond N˭N [33]. Weak peaks at 1464 cm⁻¹, 1404 cm⁻¹ and 1310 cm⁻¹ were due to C‒H bending, C‒N stretching and C‒C stretching (aromatic), respectively. Bands at 1234 cm⁻¹ and 1152 cm⁻¹ were observed for O‒H deformation (bending) and C‒O stretching combinations. In case of the IR spectrum for the complex (Figure S4) the region 3500 cm⁻¹ to 3200 cm⁻¹ was completely different with characteristic peaks due to phenolic ‒OH (Figure S3) no more seen. The regions for N˭N as well those for C‒C and C‒N stretching were strikingly different in the complex, which indicates the participation of one of the azo bonds in complex formation. New peaks appeared at 1622 cm⁻¹ and 1412 cm⁻¹, characteristic of and . Magnitude of splitting of was 210 cm⁻¹ which was observed for bridging acetates [33].

3.1.4 Thermal analysis

Thermal curves were obtained for AHPD and (Figures 1and S5, respectively). Figure 1 shows the complex was stable up to 119.5°C. An initial drop in the thermal curve corresponds to reduction in weight owing to loss of two OH groups, a phenyl unit and a three carbon fragment of another phenyl ring. At ∼248.15°C, reduction in weight occurred, attributed to loss of two terminal hydroxyphenyl units. At ∼288.9°C there was loss of two amino groups and 4 H. At slightly elevated temperatures (greater than 327.8°C) the weight corresponds to a product formed following loss of one acetate unit. At 468.2°C, the second acetate and a molecule of nitrogen depart.

Figure 1. Thermogravimetric data for [].

3.1.5 Analysis of the mass spectrum of

The mass spectrum of the complex (Figure 2) was analyzed considering the formula to be . Molecular ion peaks at m/z=940 (⁶³Cu) and m/z=944 (⁶⁵Cu) were not obtained. Instead, peaks at m/z=924.62 (⁶³Cu) and 925.50 (⁶⁵Cu) were detected, following loss of one ‒NH₂ from the molecular ion. The mass spectrum showed peaks at m/z=892.53, 893.58, 896.6 and 897.6 attributed to structures resulting from either the loss of one ‒OH and two ‒NH₂ units or one ‒NH₂ and two ‒OH units from the molecular ion. Theoretical values for these fragments were m/z=894.0 (⁶³Cu), 898.0 (⁶⁵Cu), 893.0 (⁶³Cu) and 897.0 (⁶⁵Cu). Loss of an acetate and a hydroxyl unit from the molecular ion provided fragments m/z=867.0 (⁶³Cu) and 869.0 (⁶⁵Cu), respectively. Loss of two hydroxyphenyl units and two coordinating oxygens (O⁻) should result in a fragment having a molecular mass of 726.0 with two units of positive charge. Loss of 6 H from the said species results in m/z=360.0. The fragment at m/z=359.3 obtained in the mass spectrum corresponds to this species. Loss of both the acetates from the complex should result in fragments having m/z=411.0 (⁶³Cu) and 413.0 (⁶⁵Cu), respectively. An experimental peak was found at m/z=413.34. Loss of one hydroxyphenyl and two ‒NH₂ units from the molecular ion should give a fragment having a molecular mass of 818. Further loss of two oxygens (as O⁻) coordinated to Cu(II) results in a species having a mass of 786.0 but with two units of positive charge. Thus, theoretical m/z for this species would be 393.0. An experimental peak at m/z=393.36 in the mass spectrum attests to this species.

Figure 2. Mass spectrum of [].

Indications for a dimeric nature of the complex, i.e. with acetate bridges was obtained from IR spectroscopy. This was supported by thermogravimetry and elemental analysis. Mass spectroscopy actually confirmed the dimeric nature due to the good correlation for possible fragmentations.

3.1.6 EPR spectrum of

The EPR spectrum at X-band frequency for powdered taken over 0–800 mT at 77 K shows three signals at 28 mT (Hz₁), 323 mT (H₋₂) and 612 mT (Hz₂) (Figure S6). The spin Hamiltonian for the triplet state of dimeric copper(II) compounds can be represented as [34] where D and E are zero-field splitting parameters, β is the Bohr magneton, while x, y and z denote the principal axes coordinate system fixed with respect to Cu‒Cu bond. Three resonance fields observed for such bridged complexes and their g-values (g_‖ and g_⊥) are related by the following equations. where and . g-values were calculated from the EPR spectrum and were found to be 2.23 (g_‖) and 2.24 (g_⊥).

The above analysis reveals a dimeric complex of copper (II) where copper atoms are bridged by acetate groups and each is connected to bidentate AHPD units where each tetra-coordinated Cu (II) has a chromophore. The ligand AHPD uses one phenolic ‒OH and one N (of an azo group) for coordinating the metal ion while other two oxygens come from bridging acetates.

3.2 Antimicrobial activity of AHPD and

3.2.1 Zone of inhibition

Results in Figure 3 and Table 1 suggest that AHPD showed better antibacterial activity than on bacterial cells (S. aureus and E. coli). A difference in cytotoxicity of AHPD and was understood from the zone of inhibition data that revealed AHPD was almost twice as effective in imparting toxicity to cells. Hence, complex formation reduced toxicity by approximately one half the value obtained with AHPD alone.

Table 1.  Zone of inhibition study using AHPD and on bacterial cells.

		Zone of inhibition (mm)
Compounds	Concentration (M)	Gram-positive S. aureus	Gram-negative E. coli
AHPD	60 μgm disc^−1	12.0±0.2	15.0±0.1
	60 μgm disc^−1	7.0±0.1	8.0±0.1
Ampicillin	2.5 μgm disc^−1	15.0±0.1	18.0±0.1

Figure 3. Zone of inhibition induced by different compounds on S. aureus.

3.2.2 Minimum inhibitory concentration

MIC for on S. aureus was 0.125 mg mL⁻¹ (132.9 μM) while on E. coli it was 0.5 mg mL⁻¹ (531.9 μM). For AHPD, MIC on S. aureus was 0.03125 mg mL⁻¹ (89.5 μM) and on E. coli it was 0.0625 mg mL⁻¹ (179.1 μM). The standard drug ampicillin provided an MIC value of 0.015 mg mL⁻¹ (42.9 μM). Therefore, MIC values obtained by employing the serial dilution technique on S. aureus and E. coli revealed AHPD was more effective. As a part of this study both bacteria were treated with aqueous Cu(II) (a solution of copper acetate in water) and an inert complex of Cu(II) like Cu(II)-EDTA that was earlier dissolved in DMSO. While MIC for aqueous Cu(II) was >162 μM for both bacterial cells, for Cu(II)-EDTA it was 81 μM for S. aureus and 162 μM for E. coli. From control experiments it was realized that the ability of aqueous Cu(II) to cause an inhibition to the growth of bacterial cells was lost when complexed with AHPD. For this reason, MIC values for the complex were much higher than aqueous Cu(II) for both bacterial cells.

3.3 Reductive cleavage of the azo bond by enzyme assay

3.3.1 NADPH-cytochrome c reductase assay on AHPD and

Experimental results showed enzyme-assisted reduction of the azo bond was only slightly higher in the case of AHPD than when the compounds were subjected to assay for the same period of time. Figure 4(a) and 4(b) show gradual decrease in absorbance for AHPD and , respectively. It is relevant to mention here that in AHPD there are two azo bonds while in the complex there are four (two each from two AHPD units). Of the four azo bonds two are bound to Cu(II) while the other two are free. Therefore, if we consider AHPD and the complex as participants in the same enzyme assay both actually possess two free azo bonds. For this reason results were almost similar. One has to realize here that the change in absorbance for the complex should have been much greater had all four azo bonds undergone reduction [14–16]. The fact that the complex was similar to AHPD with regard to reductive cleavage of the azo bond in the enzyme assay served as good proof that complex formation was able to check the reduction of those azo bonds bound to Cu(II).

Figure 4. Plot of absorbance of (a) AHPD and (b) in the presence of NADPH and cytochrome c reductase in phosphate buffer medium (pH ∼ 7.4) containing 0.12 M NaCl for time t=0 to t=90 min at 300 K in an enzymatic assay monitoring the gradual reduction of the azo bond. The two spectra indicate a gradual decrease in absorbance at (a) 440 nm for AHPD and (b) 445 nm for . [NADPH]=0.00032 gm ml⁻¹; cytochrome c reductase=10 U L⁻¹; (a) [AHPD]=15 μM; (b)  μM.

3.3.2 NADPH – azo-reductase (bacterial cell extracts) assay on AHPD and

Several investigators have worked on the reductive cleavage of azo compounds choosing those whose industrial applications were well established [18,19]. In some of these pioneering works, reductive fission of the azo bond was shown using NADPH as the electron donor [14,16,28–32]. We performed an enzyme assay using bacterial cell extracts from gram-positive S. aureus and gram-negative E. coli [28,29]. The study revealed reductive fission of the azo bond in the presence of azo-reductase present in bacterial cell extracts was only slightly higher for AHPD than (Figure 5).

Figure 5. Plots show the change in absorbance of (a) AHPD and (b) in the presence of NADPH and bacterial cell extract containing live cells of S. aureus in phosphate buffer (pH ∼ 7.4) and 0.06 M NaCl. Spectra were taken for time t=0 to t=90 min at 305 K for the enzyme assay that monitored the gradual reduction of the azo bond. Spectra indicate decrease in absorbance at 430 nm for (a) AHPD and at 450 nm for (b) . 300 μL S. aureus solution was added to a total volume of 2000 μL. [NADPH]=0.00032 gm mL⁻¹; (a) [AHPD]=15 μM and (b) [ μM.

As mentioned earlier, a much greater change in absorbance was expected for the ruptures of the azo bonds in the complex since the number of such bonds was exactly double that in AHPD. In the enzyme assay using bacterial cell extracts also, the change in absorbance for the complex resembled AHPD, serving as further evidence that the two azo bonds bound to Cu(II) in the complex did not undergo reductive cleavage. Bacterial azo-reductase activity and the breakdown of an azo bond to primary amines being well established justify our findings [14,16,28–32].

Figure 5(a) and 5(b) show reductive degradation of the azo bond for AHPD and , respectively, in the presence of azo-reductase obtained from the cell extract of S. aureus. However, the formation of amines could not be followed directly making use of UV–Vis spectroscopy since other constituents of the enzyme assay in the cuvette interfered. Therefore, the change in absorbance corresponding to azo-bond reduction in the compounds could only tell us about the relative difference in aromatic amine formation leading us to believe that amines were formed to a lesser extent for the complex. This manifests with decreased cytotoxicity of the complex on bacterial cells [14,16,28–32]. Although the steric factor for the complex interacting with bacterial cells is not ruled out, decreased formation of toxic amines in the case of the complex was considered the main reason for this.

3.4 Electrochemical studies on AHPD and its Cu(II) complex

Cyclic voltammetry on AHPD was performed in pure dimethyl formamide. Two reduction peaks at −0.824 V ( and −1.375 V ( were obtained owing to the reduction of the azo bond (Figure 6, curve a). Peaks obtained for AHPD were in good agreement with azo reduction data reported earlier [35–37]. The plot of the cathodic peak current (I_pc) with square root of the scan rate (v^1/2) was linear, indicating a diffusion-controlled process (Figure S7). This solution of AHPD was then subjected to reduction at constant potential (−1.3 V) for 15 min. Following such reduction, a cyclic voltammogram taken for the same solution showed a decrease in peak current at both potentials (Figure 6) indicating a rupture of azo bonds on being subjected to reduction at constant potential. While discussing reductive cleavage of the azo bond by enzyme assay it was mentioned that amines were generated as a consequence of reduction. However, the formation of amines could not be followed with the help of UV–Vis spectroscopy owing to certain limitations. Cyclic voltammetry helped us to get over this situation and using this technique, formation of amines was detected. It is reported in the literature that amines undergo oxidation in the potential range 0.6–1.2 V [38]. We verified this by performing experiments with 2-, 3- and 4-amino phenols.

Figure 6. Cyclic voltammogram of (a) pure AHPD and (b) AHPD after being subjected to reduction at a constant potential of−1.3 V for 15 min recorded using a glassy carbon electrode at a scan rate of 100 mV s² in dimethyl formamide having 0.1 (M) TBAB as supporting electrolyte. [AHPD]=1000 μM.

The AHPD molecule having one amino group in its structure showed a peak at 1.17 V (Figure 7a). When cyclic voltammograms for AHPD were taken after it was reduced at constant potential (−1.3 V) for 15 and 45 min, respectively, we found that the peak current for the oxidation of amines increased (Figure 7b and c). This showed clearly that amines were formed as a consequence of the reduction of the azo bonds. Although voltammogram (b) in Figure 7 did not show any shift in potential compared to voltammogram (a), voltammogram (c) showed a slight shift toward higher potential accompanied by peak broadening. Voltammogram (c) was obtained after AHPD was reduced at constant potential (−1.3 V) for 45 min resulting in the presence of a mixture of amines in solution. If AHPD undergoes reduction, one molecule of 3-amino phenol and two molecules of 2, 5 diamino phenol are formed. As already mentioned AHPD itself contains an amine group. Hence peak broadening for voltammogram (c) of Figure 7 could be attributed to the presence of a mixture of amines in solution. It seems apparent from Figure 7 that the increase in current at 1.20 V for (b) and 1.31 V for (c) was not very high compared to that obtained for pure AHPD at 1.17 V (a). However, that is not the case since AHPD's contribution to the peak at 1.20 V for (b) and 1.31 V for (c) would have decreased considerably following its reduction. Even then the current showing 697 μA for voltammogram (b) and 700 μA for (c) were higher than 645 μA recorded for (a). This meant amines were formed as a consequence of the reduction of AHPD and contributed to the peaks for voltammograms (b) and (c), respectively (Figure 7). For example, for voltammogram (c) of Figure 7 the contribution from the amine of AHPD would be very small since AHPD was subjected to reduction for 45 min at −1.3 V and the major contributor to the current were amines formed from the breakdown of AHPD. Hence, the change in current obtained from the oxidation of amines helps us to demonstrate that they were formed following the reduction of azo bonds of AHPD.

Figure 7. Cyclic voltammogram of (a) pure AHPD and AHPD after being reduced at a constant potential of−1.3 V for 15 min (b) and 45 min (c) showing the oxidation of amines in solution that were recorded using a glassy carbon electrode at a scan rate of 100 mV s² in dimethyl formamide having 0.1 (M) TBAB as supporting electrolyte. [AHPD]=1000 μM. The arrow indicates with increase in time provided for the reduction of AHPD at−1.3 V current for amine oxidation increased.

We demonstrated through enzyme assay experiments that complex formation had a retarding effect on the reduction of the azo bond. To substantiate this with cyclic voltammetry, experiments on in pure dimethyl formamide were carried out. Both the first and the second reduction peaks shifted to more negative potentials. While the first peak appeared at −1.08 V (E_pc1), the second (E_pc2) was obtained at −1.39 V. This negative shift in potential for indicated it was difficult to reduce the azo bond in the complex (Figure 8) [23]. It was also observed that for in spite of having two AHPD units reduction peak current for the complex (∼2 μA for a 500 μM solution) was smaller than that obtained for AHPD (∼10 μA for a 1000 μM solution) indicating the disinclination of azo bonds in AHPD units in the complex to get reduced (Figures 6a and 8).

Figure 8. Cyclic voltammogram of recorded on a glassy carbon electrode at a scan rate of 25 mV s² in dimethyl formamide having 0.1 (M) TBAB as supporting electrolyte.  μM.

4. Conclusion

The interaction of gram-positive S. aureus and gram-negative E. coli with AHPD and provides interesting observations. The MIC and zone of inhibition studies revealed AHPD was more effective. Studies on the reductive fission of the azo bond helped to correlate cytotoxic action of the compounds on bacterial cells to the generation of amines. It was concluded reduction of the azo bond in case of the complex was significantly lower than AHPD, the chosen azo compound, suggesting complex formation affects such reductions. This could be important for modifying azo compounds such that their toxicity was somewhat controlled and could have a significance with regard to applications of azo compounds. Although findings of this study was based on results obtained using bacterial cells they should be true for higher organisms also.

Supplemental data

Supplemental data for this article can be accessed at http://dx.doi.org/10.1080/2164232X.2014.883287.

Nomenclature

:

=

A dimeric complex of Cu(II) with AHPD having bridging acetates

Acknowledgements

D.G. wishes to thank the UGC, New Delhi, for a project fellowship. S.D. is grateful to Prof. Samiran Mitra of the Department of Chemistry, Jadavpur University, for kindly providing the TGA data of the compounds. D.G. remains grateful to Mr Piyal Das for his help in enzyme assay experiments.

Funding

This work was funded by the University Grants Commission, New Delhi, in the form of a Major Research Project [39-749/2010(SR)] to S.D. Funding from the UGC is gratefully acknowledged.