Quantitative determination and toxicity evaluation of 2,4-dichlorophenol using poly(eosin Y)/hydroxylated multi-walled carbon nanotubes modified electrode

This study aimed at developing simple, sensitive and rapid electrochemical approach to quantitatively determine and assess the toxicity of 2,4-dichlorophenol (2,4-DCP), a priority pollutant and has potential risk to public health through a novel poly(eosin Y, EY)/hydroxylated multi-walled carbon nanotubes composite modified electrode (PEY/MWNTs-OH/GCE). The distinct feature of this easy-fabricated electrode was the synergistic coupling effect between EY and MWNTs-OH that enabled a high electrocatalytic activity to 2,4-DCP. Under optimum conditions, the oxidation peak current enhanced linearly with concentration increasing from 0.005 to 0.1 μM and 0.2 to 40.0 μM, and revealed the detection limit of 1.5 nM. Moreover, the PEY/MWNTs-OH/GCE exhibited excellent electrocatalytic activity toward intracellular electroactive species. Two sensitive electrochemical signals ascribed to guanine/xanthine and adenine/hypoxanthine in human hepatoma (HepG2) cells were detected simultaneously. The sensor was successfully applied to evaluate the toxicity of 2,4-DCP to HepG2 cells. The IC50 values based on the two electrochemical signals are 201.07 and 252.83 μM, respectively. This study established a sensitive platform for the comprehensive evaluation of 2,4-DCP and posed a great potential to simplify environmental toxicity monitoring.

Scientific RepoRts | 6:38657 | DOI: 10.1038/srep38657 cathodic peak attributed to the reduction (EY 2− + e − → EY 3·− ) was observed at − 0.87 V in the 1 st cycle. The peak current decreases with increase in number of cycles. On the MWNTs-OH/GCE (Fig. 2B), the background current was much larger than that of GCE, indicating the higher surface area of MWNTs-OH. Two oxidation and reduction peaks were found at the potential of + 0.69 V, + 0.45 V, − 0.50 V, and − 0.87 V, respectively. During the one-step electrodeposition process, the EY molecules acted as an excellent electron acceptor, which exist mainly in the form of EY 2− and could be reduced to EY 3− by the cleavage of the C= O bond on the benzene ring (Fig. 3). The obtained EY 3− could rapidly combine with the electrode surface 44 . The peak current was virtually constant after 15 cycles, indicating that the polymerization reached saturation at the MWNTs-OH/GCE. The thickness of the film had a significant contribution to the property of the PEY/MWNTs-OH/GCE. Thus, the influences of the cycle number and scan rate of EY on the electrocatalytic performance of PEY/MWNTs-OH/GCE were investigated (the inset of Fig. 2B). The oxidation peak current of 2,4-DCP reached to the maximum after 15 scans at the scan rate of 50 mV s −1 . The peak current began to drop when the polymerizing cycles were more than 15. Thick films may prevent the electron transfer process and the oxidation process. Hence, the optimal electropolymerization cycles are selected as 15 cycles.
The surface morphologies of each layer were characterized by SEM. Randomly oriented MWNTs-OH was detected with interconnected tubular structures (Fig. 4A), which was the characteristic of carbon nanotubes 45 . After the introduction of EY molecules, the EY polymer was uniformly distributed over the MWNTs-OH (Fig. 4B). The homogeneous and uniform film with three-dimensional network structure was produced, demonstrating that the EY could be modified effectively on the surface with the electrodeposition method. The highly conjugated PEY/MWNTs-OH composite exhibited high surface area, and thus providing more sites for the accumulation of the target molecule.
Then the CVs of different electrodes were studied in 5.0 mM [Fe(CN) 6 ] 3−/4− containing 0.1 M KCl (Fig. 4C). The electrochemically active surface area can be obtained according to the following equation 46  where i p is the peak current, n is the number of electrons involved in the redox reaction of Fe(CN) 6 3−/4− (n = 1), A is the electroactive surface area, C is the reactant concentration, D is the diffusion coefficient (6.30 × 10 6 cm 2 s −1 ),  and v is the scan rate. The active surface areas for MWNTs-OH/GCE and PEY/MWNTs-OH/GCE were calculated as 0.079 cm 2 and 0.108 cm 2 , respectively, which further confirmed that the PEY film increased the surface area of the MWNTs-OH/GCE. EIS can provide details on the interfacial properties of the electrode interface during the fabrication process. The semicircular portion at higher frequencies represents the electron transfer-limited process, and the diameter corresponds to the electron transfer resistance (R ct ) 47 . Figure 4D shows the typical Nyquist plots recorded at frequencies ranging from 0.01 to 10 5 Hz in 0.1 M KCl containing 5.0 mM [Fe(CN) 6 ] 3− / 4− as the electrochemical redox probe. The inset shows the equivalent circuit model to fit the impedance data. R s , C, and W represented the solution resistance, pure capacitance, and Warburg impedance, respectively. The EIS at the bare GCE displayed a well-defined semicircle, with a huge interfacial R ct of 14.3 KΩ. The increase in the R ct of PEY/GCE suggested a successful modification of EY on the GCE surface. The R ct decreased dramatically to 12.1 KΩ after the introduction of MWNTs-OH on the GCE, confirming an excellent electron conducting ability of MWNTs-OH. The lower R ct of PEY/MWNTs-OH/GCE (9.2 KΩ) demonstrated the enhanced electron transfer reaction, which was mainly attributed to the synergistic effect between EY and MWNTs-OH.
Raman spectroscopy is a useful tool to obtain structural information of carbonaceous materials 48 . The D band corresponds to the disordered structural defects and the G band is due to the first-order scattering of the E 2g mode for sp 2 carbon lattice. The relative intensity ratio of D band to G band (I D /I G ) can be used to examine the disorder and defects. The decrease of I D /I G indicated the increase in the average size of sp 2 domains 49 . PEY/MWNTs-OH (Fig. 5A, curve b) exhibits the D band at 1357 cm −1 and the G band at 1580 cm −1 with intensities lower than that of MWNTs-OH (Fig. 5A, curve a). The I D /I G value of MWNTs-OH was estimated to be 1.09. In the case of PEY/ MWNTs-OH, the I D /I G value decreased to 0.78, which implied the successful formation of the PEY/MWNTs-OH hybrids and an efficient π -π interaction between them.
Voltammetric behavior of 2,4-DCP at the PEY/MWNTs-OH/GCE. The CVs of 2,4-DCP at different electrodes were investigated (Fig. 5B). No peak was observed at the GCE (curve a), which was consistent with the previous study 13 . A broad oxidation peak appeared at the PEY/GCE (curve b) but it was too weak to distinguish. In the case of MWNTs-OH/GCE, a peak at + 0.78 V was observed (curve c), while a well-defined anodic peak was obtained at + 0.77 on PEY/MWNTs-OH/GCE (curve d). There was no corresponding reduction peak in the inverse scan, which was the characteristic of an irreversible electrode process. The high background current of PEY/MWNTs-OH/GCE reflected its effective surface area. In addition, the peak current was 5.2 times that of PEY/GCE and 2.9 times that of MWNTs-OH/GCE. Considering that graphene (Gr) was also an ideal material which can be coupled with organic dye molecules through the electron cloud overlap, the PEY/MWNTs-OH/ GCE was compared with PEY/Gr/GCE (Fig. S1). The oxidation peak current of 2,4-DCP at PEY/MWNTs-OH/ GCE was 2.1 times as large as PEY/Gr/GCE. The above results were caused by two reasons. First, MWNTs-OH increased the electronic conductivity, and the well-distributed poly(eosin Y) film improved the surface area for high adsorptive capability for 2,4-DCP. Second, the synergistic effect between PEY and MWNTs-OH increased the electrocatalytic activity as well as promoted the electron-transfer rate.
The experimental parameters for the detection of 2,4-DCP using the PEY/MWNTs-OH/GCE were optimizated. The effects of pH on the oxidation peak current (I p ) and peak potential (E p ) of 2,4-DCP at the PEY/ MWNTs-OH/GCE were studied at the pH range of 2.0-8.0 (Fig. 6A). The I p reached to the maximum at pH 3.0 and declined obviously at pH 4.0. There was no significant change in peak current from pH 4.0 to 7.0, but the I p decreased sharply at the pH value of 8.0, which was related to the electrochemical mechanism of 2,4-DCP. The lower pH promoted the ionization process of 2,4-DCP at the initial stages of the electrochemical oxidation, in which some phenol hydroxyl radicals were generated 50 . Meanwhile, the E p of 2,4-DCP shifted linearly toward negative potential values with increasing the pH between 2.0 and 8.0 (inset of Fig. 6A). The linear regression equation was E p = − 0.0662 pH + 0.975 (R = 0.993). The slope (− 0.0662 V pH −1 ) revealed that equal numbers of electron and proton are involved in the reaction process of the 2,4-DCP 51 . Therefore, the optimum pH was selected as 3.0.
The effect of scan rate on the electrochemical characteristics of 2,4-DCP was then studied. The I p of 2,4-DCP increased with the scan rate in the range from 30 to 200 mV s −1 (Fig. 6B). The linear relationship between I p and the square root of the scan rate (v 1/2 ) can be expressed as I p (μ A) = 2.227 ν 1/2 (mVs −1 ) − 6.692 (R = 0.997), indicating that the electron transfer process was controlled by diffusion. Meantime, the E p shifted positively with the increasing in the scan rate. The E p changed linearly versus the natural logarithm of scan rate (ln v) with a linear regression equation of E p (V) = 0.0188 ln ν (mV s −1 ) + 0.703 (R = 0.996). According to Laviron's theory 52 , the relationship between E p and ln v in the irreversible electrode process could be described as where v is the scan rate, α is the electron transfer coefficient, k s is standard rate constant, and R, T, and F have their usual meanings. α is presumed as 0.4 in the irreversible electrode process 53 , and α n was easily calculated from the plot of E p versus ln v. The electron transfer number n is calculated to be 2. Thus, the oxidation process of 2,4-DCP on PEY/MWNTs-OH/GCE is a two-electron and two-proton process 5 .
Moreover, the peak current is related to the amount of phenols accumulated on the modified electrode 54 . The influence of accumulation time and potential was investigated. The peak current of 2,4-DCP increased noticeably with raised accumulation time within 250 s (Fig. 6C). When the accumulation time was longer than 250 s, the peak current dropped slightly, indicating the 2,4-DCP reached the level of saturation. Additionally, the influence of accumulation potential was investigated (Fig. 6D), in which the maximum current was obtained at − 0.  the PEY/MWNTs-OH/GCE electrode exhibited relatively wide linear range and low detection limit (Table 1), and thus was a promising sensor for the sensitive determination of 2,4-DCP.
The reproducibility was investigated by six measurements of 20.0 μ M 2,4-DCP with the same PEY/ MWNTs-OH/GCE. The relative standard deviation (RSD) was 3.74%. Additionally, six electrodes were examined at 20.0 μ M 2,4-DCP and the RSD was 4.77%. These results proved that the sensor possessed excellent reproducibility and repeatability.
Furthermore, the analytical reliability and application potential of PEY/MWNTs-OH/GCE was investigated to determine 2,4-DCP in real water samples obtained from Yitong River (Changchun, China). The recoveries ranged from 93.2% to 105.6% (Table 2) indicating the excellent reliability and applicability of PEY/MWNTs-OH/GCE.

Electrochemical behavior of HepG2 cells at the PEY/MWNTs-OH/GCE. The potential application
of the PEY/MWNTs-OH/GCE in HepG2 cell suspension (3.0 × 10 6 cells mL −1 ) was explored (Fig. 8A). No peak appeared on the bare GCE (curve a). A broad oxidation peak was observed at about + 0.65 V at the PEY/GCE (curve b). For MWNTs-OH/GCE (curve c), an oxidation peak was obtained at + 0.60 V, and an oxidation peak appeared at + 0.92 V, whereas, it was too weak to recognize. With the PEY/MWNTs-OH/GCE in HepG2 cell suspension (curve d), the background current was greater than other electrodes, suggesting the more effective surface area. Meanwhile, two well-defined oxidation peaks at + 0.59 V and + 0.90 V attributed to xanthine/guanine, and hypoxanthine/adenine 22,23,26 were observed. The oxidation potentials shifted to less positive ones and the maximum peak currents were obtained. These results implied that the PEY/MWNTs-OH film possessed unique electrocatalytic activities toward the purine bases in HepG2 cells, and could promote electron transfer reactions. The influence of accumulation condition on the electrochemical signal was investigated (Fig. 8B), and 420 s was chosen as the best accumulation time.
The two electrochemical signals on PEY/MWNTs-OH/GCE were applied to describe the growth of HepG2 cells (Fig. 9). The peak currents increased gradually with the culture time within 30 h owing to the proliferation of cells. Then the peak currents reduced substantially because of the cell death caused by the lack of nutrients. These results well corresponded with the cell counting method (inset of Fig. 9), implying that the PEY/MWNTs-OH/ GCE can be applied to monitor cell growth in real-time mode.    control groups (curves a and b), the electrochemical signal I (curve c) and signal II (curve d) of 2,4-DCP treated groups increased slowly until 24 h and then decreased ascribed to the toxicity of 2,4-DCP. Meanwhile, the peak currents of 2,4-DCP were lower than those of the control group, indicating the inhibitory effects of 2,4-DCP on HepG2 cells. A notable decrease in the electrochemical response was observed after treated by 2,4-DCP for 30 h, and the cytotoxicity reached to the maximum (inset of Fig. 10). Therefore, 30 h was chosen as the best 2,4-DCP-treated duration. The electrochemical behaviors of HepG2 cells after treated by 2,4,-DCP with different concentrations for 30 h were studied using PEY/MWNTs-OH/GCE (Fig. 11). 2,4,-DCP exhibited significant cytotoxicity with a concentration-dependent pattern (inset of Fig. 11). The cytotoxicity (Y) was linear to the logarithm of 2,4,-DCP concentration (X) with the linear regression equations of Y = 46.35 X − 56.76 (R = 0.990) for signal I and Y = 42.50 X − 52.12 (R = 0.993) for signal II. The IC 50 values based on the two electrochemical signals were 201.07 and 252.83 μM, respectively, reflecting that 2,4,-DCP had greater impacts on xanthine/guanine than those on hypoxanthine/adenine. It also confirmed the toxicity differences of 2,4-DCP to purine metabolism, which had a significance for the study of the cellular physiological process. 2,4,-DCP has been proved as a potential environmental endocrine disruptor and oxidative damage inducer 55 , and can change the antioxidant enzyme activities and induce oxidative stress, a mechanism responsible for DNA damage and inhibition of cell growth 56,57    Conclusion A novel electrochemical sensing platform was constructed in a simple and green way. The PEY/MWNTs-OH/ GCE significantly facilitated the electron transfer efficiency and possessed excellent electrocatalytic activity. Under the optimal condition, it exhibited a high sensitivity, wide linear range, excellent reproducibility and good stability toward 2,4-DCP. The easy-fabricated electrode developed was also employed in real water sample analysis and showed high accuracy. Moreover, the electrochemical behavior of HepG2 cells was investigated by the PEY/MWNTs-OH/GCE. The cytotoxicity of 2,4-DCP was successfully evaluated by this electrode. This study constituted a promising tool for 2,4-DCP analysis and toxicity evaluation, and revealed a simple sensitive electrochemical approach in the field of environmental monitoring and toxicology.

Experimental Section
Materials and chemicals. The MWNTs-OH (Nanjing XFNANO Materials TECH Co., Ltd.) was 20-40 nm in diameter with purity higher than 97%. The minimum essential medium (MEM), defined fetal bovine serum (FBS) (Gibco Co., USA), penicillin, streptomycin, trypsin (Sigma, Co., USA) were used for cell culture. Dimethyl formamide and eosin Y (EY) (J&K Chemical Ltd., China) were applied for electrode fabrication. Apparatus. The electrochemical measurements including cyclic voltammetry (CV) and differential pulse voltammetry (DPV) were performed on the CHI 760E electrochemical workstation (Shanghai CH Instruments, China). A three-electrode system was used, which consisted of a modified GCE as the working electrode, a saturated calomel electrode (SCE) as the reference electrode, and a platinum wire as the auxiliary electrode. The surface morphology of the nanocomposite was observed using a JEOL 6340F Scanning electron microscope (SEM). Electrochemical impedance spectroscopy (EIS) was performed using the PARSTAT 2273 potentiostat (Princeton applied research, USA). The Raman scattering spectra was obtained by the Jobin-Yvon HR 800 instrument with an Ar + laser source of 488 nm wavelength. The pH of the solution was measured with a PB-10 pH-meter (Sartorius Co., Germany).  (Fig. S2). The obtained electrode was termed as PEY/MWNTs-OH/GCE. Cells culture and treatment. The HepG2 cells (COBIOER Biosciences Co., Ltd.) were cultured in 60 mm cell culture dish in the minimum essential medium (MEM) that was supplemented with 10% FBS, penicillin (100 μ g mL −1 ), streptomycin (100 μ g mL −1 ), 1% non-essential amino acids, and 1% sodium pyruvate at 37 °C in a humidified 5% CO 2 . For the toxicity investigation, the growth medium was replaced with the medium containing 2,4-DCP. The in-situ cell collection was conducted as our previous study 23 . Briefly, the PBS was added to cells after the medium was removed, and then the mixture was heated in the water bath at 50 °C for 30 min to obtain the HepG2 cell suspension.

Preparation of the PEY/MWNTs
Electrochemical determination. For the electrochemical detection of 2,4-DCP, the CV was employed between + 0.1 and + 1.0 V with the scan rate of 50 mV s −1 . The DPV was performed with the following parameters: increment potential, 4 mV; pulse amplitude, 0.05 V; pulse width, 0.05 V; pulse period, 0.2 s; sample width, 50 ms. The in vitro toxicity of 2,4-DCP to HepG2 cells was investigated from 0.0 to + 1.1 V with the scan rate of 50 mV s −1 . After each measurement, the PEY/MWNTs-OH/GCE was scanned for five cycles between 0.0 and + 1.1 V in PBS and rinsed thoroughly with double-distilled water.
The MTT assay. HepG2 cells (1.2 × 10 4 cells mL −1 ) in medium alone (200 μ L) or the medium containing 2,4-DCP (200 μ L) were added to the 96-well microtitre plates. 20 μ L 5 mg mL −1 MTT was added to each well after the incubation at 37 °C for 30 h. The medium containing MTT was removed after 4 hours, and 150 μ L sodium dodecyl sulfate was added. The measurement was registered on an ELX800 Microplate Reader (BioTek Instruments, Inc., USA) and determined by the absorbance values at 490 nm.