The reduction of nitrobenzenes is a complex process, especially for those compunds bearing hydroxyl or amine groups in the ortho- or para- positions [21]. The electrochemical behavior of o-nitrophenol on GCE was explored with cyclic voltammetry. Figure 1 showed the cyclic voltammograms of 5.0×10⁻⁵ M o-nitrophenol in 0.10 M pH 4.5 acetate buffer under different electrochemical conditions. Figure 1A was obtained with cycling between − 0.80 V and + 1.20 V, initial potential − 0.30V. A very large irreversible cathodic peak (R₁) and an irreversible anodic peak (Ox₅) appeared at around E_pc = − 0.602 V and E_pa = + 0.998 V in the first cycle, respectively. The R₁ peak corresponded to a four electrons / four protons reaction, forming hydroxylamine [17, 22], typical of a nitro reduction in acidic aqueous solution, and decreased in subsequent cycles. Peak Ox₅ at E_pa= + 0.998 V, was related to the irreversible oxidation of the phenolic group and also decreased in following cycles due to the phenol polymerization at the electrode surface [18]. It was interesting to observe another irreversible anodic peak (Ox₄) at E_pa = + 0.335 V in the second and third cycle, which probably corresponded to condensation reactions expected concerning the oxidation of the reduced intermediates formed, generating electroactive azoxy derivatives [17, 23, 24].

However, two reversible waves could be distinguished if the potential ranged from − 0.80 V to + 0.40 V, with initial potential at − 0.30 V from negative to positive potentials scan (see Figure 1B). On the second and subsequent cycles, the voltammogram was characterized with the one reversible wave centered at − 0.173 V (Ox₂/R₂) and the other reversible peak centered at + 0.041 V (Ox₃/R₃). The irreversible reduction wave R₁, corresponding to a four electrons and four protons typical nitro reduction in weak acid aqueous solution, was present in the first scan, but the reversible wave Ox₂/R₂ corresponding to nitroso/hydroxylamine derivative, occurred and slowly increased in the second and the following scan, as predicted. It was similar to the behavior of this compound observed at a hanging mercury drop electrode [15, 21]. The latter reversible peak Ox₃/R₃ possibly corresponded to the pair azoxy/azo.

Repetitive differential pulse voltammograms for reduction of o-nitrophenol is shown in Figure 2. The R₂ and R₃ wave only appeared in the second and following cycles after the initial o-nitrophenol reduction, which was the same with what observed in the cyclic voltammogram. The reduction peak R₁ corresponding to the nitro group was decreased, while the reduction peak corresponding to the nitroso derivative R₂ and azoxy derivative R₃ were increased in subsequent scans. This behavior could be explained by reduction of o-nitrophenol leading to its consumption on the electrode surface and electrogenerating a hydroxylamine derivative that adsorbed onto the electrode surface and was oxidized to nitroso and other derivatives.

Figure 3 shows the differential pulse voltammogram for oxidation of o-nitrophenol starting at different potentials and deposition at a constant potential. It was confirmed that peak Ox₅ was due to the irreversible oxidation of the phenolic group, as only this group is able to undergo oxidation in o-nitrophenol (Figure 3, curve a), and that Ox₂, Ox₃ and Ox₄ were generated only after o-nitrophenol was reduced at − 0.60 V (Figure 3, curve b and c). It caused an increase in the peak Ox₂, Ox₃ and Ox₄ after applying a potential of − 0.6 V for 2 minters (Figure 3, curve c), which produced abundant hydroxylamine in this process.

In a similar way to what was observed for nitroaromatic compounds [16, 17], o-nitrophenol reduction was also pH dependent in weak acidic medium. The potential of reduction peak shifted to negative values with increasing pH. The variation of E_pc for the reduction of o-nitrophenol with pH showed the slop of 52 mV/pH (data not shown). This behavior indicated the same number of electrons and protons participated in the reduction of o-nitrophenol.

In order to investigate that electrochemical reduction of o-nitrophenol generates short-lived radical that interacts with DNA causing damage, differential pulse voltammetry was applied to detect DNA damage using a DNA-modified GCE. The blank experiment, i.e. oxidation at DNA-modified GCE without o-nitrophenol (Figure 6d), was performed in buffer solution to test the above hypothesis. There were almost no oxidation peaks of the DNA base, because nucleobases within native DNA are hidden inside the helix and stiffness of the structure keeps them far from the electrode surface [28]. Figure 6a shows the differential pulse voltammetry of o-nitrophenol oxidation using DNA-modified GCE. There was only an oxidation peak at + 0.982 V, which corresponded to the o-nitrophenol oxidation (Figure 6 curve a). However, the damage to DNA was detected if o-nitrophenol was reduced at a constant deposition potential E_d = − 0.60 V for 2 min (Figure 6, curve c). The appearance of two oxidative peak was attributed to guanine, E_pa = + 0.785 V and adenosine, E_pa = + 1.271 V [29]. In addition, the peak current at E_pa = + 0.982 V increased. It was interesting to observe DNA damage if the reduction of o-nitrophenol was performed in repetitive differential pulse scanning from 0 to − 0.80 V (10 scans) (Figure 6, curve b). This distinctly indicated that o-nitrophenol could damage to DNA after the o-nitro-phenol underwent reduction process. The radicals generated in reduction process of the o-nitrophenol interacted with DNA [22, 30]. The appearance of purine oxidation peak demonstrated clearly that the radicals caused distortion of DNA double helix and exposure of the bases.

The above experimental results provided some evidence that reduction of o-nitrophenol was a complex process. Its electrochemically reduction process underwent some intermediate product formation such as nitro radical anion, nitroso, hydroxylamine, etc. The radical formed by reduction of the nitro group could interact with dsDNA and cause dsDNA damage. After the o-nitrophenol was reduced, the detected oxidation current signal of the guanine and adenosine nucleobase supplied evidences of DNA damage. According to above experimental results and reference to [31], a possible mechanism of the o-nitrophenol redox and damage to dsDNA was proposed as follows:

Calf thymus DNA (sodium salt, type I) was purchased from Sigma and o-nitrophenol was received from Shanghai Chemical Reagent Company, respectively. They were used without further purification. Acetate buffer solution of 0.10 M (pH = 4.5) was prepared using analytical grade reagents. Double-distilled water was used throughout. All experiments were performed at room temperature.

Stock solution of 5.0×10⁻³ M o-nitrophenol was prepared in a 1:1 mixture of pH = 4.5 0.10 M acetate buffer and ethanol. The stock solution was diluted with 0.10 M acetate buffer according to the demand. The solution was saturated with N₂ by bubbling high purity nitrogen gas into solution for 15 min before experiment. Nitrogen gas was maintained over the solution by continuing with a flow of the pure gas during the voltammetric experiment.

All voltammetric measurements were carried out using a CHI440A electrochemical analyzer (Shanghai Chenhua Apparatus Corporation, China). The experiments were performed using a 5 mL one-compartment electrochemical cell with a three electrodes system, which was consisted of a glassy carbon electrode or DNA-modified glassy carbon electrode as working electrode, a platinum coil auxiliary electrode and a saturated calomel electrode (SCE) as reference electrode.

The DNA-modified glassy carbon electrode (Φ = 3 mm) was prepared as follows: It was polished using a piece of 1500 diamond paper, and then polished to mirror smoothness with about 0.05 um alumina water slurry on silk. Afterward, it was washed with absolute alcohol and double-distilled water in an ultrasonic bath to remove the adsorbates on the electrode. After electrode was dried in air, 20 μL of DNA solution (2.0 mg/mL) was dropped onto the surface of the clean glassy carbon electrode and solvent was evaporated at 4 °C for 12 hours. The modified electrode was immersed the double-distilled water for 15 min to remove the unadsorbed DNA before using.