An ultrasensitive HRP labeled competitive aptasensor for oxytetracycline detection based on grapheme oxide–polyaniline composites as the signal amplifiers

Abstract

In this work, we explored an amplification strategy which was based on graphene oxide–polyaniline (GO–PANI) and horseradish peroxidase (HRP) to construct the competitive aptasensor for ultrasensitive detection of oxytetracycline (OTC). In the protocol, the GO–PANI film was immobilized on the surface of the electrodes. Then, gold nanoparticles (AuNPs) were electrodeposited on the electrode surface using a constant potential stripping technique. The selected aptamer which had high affinity and specificity for OTC was used as a capture probe and labeled with HRP. The linear response to OTC concentration of the developed aptasensor was in the range of 4.0 × 10⁻⁶ mg L⁻¹ to 1.0 mg L⁻¹. The detection limit (LOD) of 2.3 × 10⁻⁶ mg L⁻¹ was obtained (S/N = 3). In addition to good repeatability and stability, the proposed aptamer sensor also showed the advantages of low background current and high sensitivity to examine OTC in real samples.

---

Introduction

Tetracyclines (TCs) are broad-spectrum antibiotics that have been widely used in agriculture operations and human medicine for the treatment of infectious diseases. They operate very effectively at low concentration and can be completely metabolized from the body only a short residence time later. However, excessive use of TCs may cause a serious threat to human health like allergies^1,2 and bacterial resistance.³ Besides, the widespread use of TCs could lead to TCs residues in food animals, such as meat,^4,5 milk,⁶ and eggs,⁷ which can damage the liver and kidneys and present potential risks to public health.^8,9 Furthermore, TCs residues can enhance the drug-resistance of bacteria and cause environmental pollution.¹⁰

Due to its outstanding antibacterial properties,^11,12 oxytetracycline (OTC) is the most useful member among the TCs. The WHO has announced that the maximum residual value of OTC is 0.1 mg L⁻¹ in drugs and human food.¹³ In recent years, the detection methods of OTC in foods of animal and biological samples analysis have been reported in the literatures, including high performance liquid chromatography,^14,15 fluorescence,^16,17 mass spectrometry¹⁸ and so on. But these methods are often disadvantage of time-consuming, expensive and low specificity. There has been reported electrochemical aptasensor using interdigitated array electrode chip for oxytetracycline detection,¹⁹ the detection range of which was unsatisfactory. Establishing an efficient and accurate method for the detection of OTC residue is of great importance in food safety.

Graphene oxide (GO) is stripped from graphite oxide with –OH, –COOH functional groups on the basal planes and at the edges of the sheets, which has been widely used in the supercapacitors and electrorheology.^20–23 Because of its cost-effective and redox reversibility, polyaniline (PANI) has been studied for a long time as conducting polymer.^24,25 Studies have reported that the electrical conductivity of the graphene oxide–polyaniline (GO–PANI) composites is obviously higher than that of pure GO or PANI.²⁶ The GO–PANI is with high controllable conductivity and good electrochemical properties for signal amplification.

Since screened through the systematic evolution of ligands by exponential enrichment (SELEX) process from random RNA or DNA libraries, a large number of aptamers have been selected for all kinds of analytes in the past few years. These DNA or RNA molecules have unprecedented advantages such as high selectivity and affinity toward their targets, chemical stability, synthetic convenience, easy modification by functional group in biosensor fabrication.^27–29 Aptamers are used in developing biosensors for detecting different types of analytes including small biomolecules,^30–32 drugs,³³ proteins^34–36 and so on. They are chemically stable, accurate and reproducible via chemical production and easily modified. Aptamers are easily designed to undergo significant conformational changes upon target binding, which offer high flexibility in biosensor fabrication.^37,38

In this work, we designed the competitive sensor based on GO–PANI composites and horseradish peroxidase (HRP) for signal amplification. The complete antigen oxytetracycline-bovine serum albumin (OTC-BSA) was fixed on the electrodes. Herein, BSA, as a carrier for OTC, can adsorb effectively the surface of gold nanoparticles by free amino groups. Thus, the OTC samples competed with the OTC-BSA on the electrodes to bind horseradish peroxidase labeled aptamer (HRP–Apt). The more OTC samples were added, the less horseradish peroxidase (HRP) was combined with OTC-BSA on the electrodes. This caused the current decreased which indicated the quantity of OTC samples. The content of OTC in the samples was inversely proportional to the signal. Our competition-based aptasensor has a major advantage in its high specificity, which is suitable for complex samples. In our study, it is also successfully applied to OTC analysis in complicated samples.

Experimental section

Materials and apparatus

The sequence of aptamer was purchased from Shanghai Sangon Biotechnology Co. Ltd. (Shanghai, China) as follows: 5′-biotin-C₃-GGA ATT CGC TAG CAC GTT GAC GCT GGT GCC CGG TTG TGG TGC GAG TGT TGT GTG GAT CCG AGC TCC ACG TG-3′. HRP–Apt conjugates were consisted of the biotinylated aptamer and streptavidin (SA) labeled HRP. In a word, the biotinylated aptamer was fully linked to SA-horseradish peroxidase (SA-HRP) for overnight at room temperature (RT) by the strong specific binding (K ∼ 10⁻¹⁵ M) between SA and biotin.³⁹ After the reaction, the final product was stored at 4 °C before using. Bovine serum albumin (BSA) and SA-HRP were obtained from Sigma-Aldrich (St. Louis, MO, USA). OTC dehydrate, chlortetracycline hydrochloride and tetracycline hydrochloride were all purchased from Aladdin Chemistry Co. Ltd. (Shanghai, China) Aniline (99%) was purchased from Beijing Chemical Technology Co., Ltd. (Beijing, China). The complete antigen of oxytetracycline-BSA (OTC-BSA) was purchased from Mai-Cang Inc. (Shanghai, China) which was synthesized by BSA as matrix. All chemicals used were analytical grade. Ultrapure water obtained from a Milli-Q (18.25 MΩ cm) water purification system was used throughout the experiments.

CV experiments were carried out with a CHI 660D electrochemical workstation (Chenhua Inc., Shanghai, China). Scanning electron microscope (SEM) images and transmission electron microscope (TEM) were obtained respectively using field emission SEM (ZEISS, Germany) and JEOL-1400 (Japan). All electrochemical experiments were performed with a conventional three-electrode system that the modified GCE as the working electrode, a platinum wire as auxiliary electrode and an Ag/AgCl (sat. KCl) as reference electrode.

Preparation of GO

Firstly, GO was prepared from graphite powder by a classical Hummers and Offenman's method.⁴⁰ In short, graphite power was joined in the sulfuric acid (98%) solution containing NaNO₃ and KMnO₄. Then H₂O₂ was rather slowly added to the mixture. The reaction mixture was filtered after stabilized for several hours. The residue was washed with HCl aqueous solution and ultrapure water, respectively. Precipitate dried at a constant temperature box at 60 °C. Then GO was obtained.

Synthesis of GO–PANI composites

In the presence of GO, the GO–PANI composites were synthesized through an in situ polymerization process of aniline monomers. The obtained GO (25 mg) was dissolved in 50 mL of 1 M H₂SO₄ with ultrasonic dispersion for 1 h. The purpose was to make the layers of GO stripping dispersed into GO monolayer. 200 μL of aniline monomer was added into the GO solution and sonicated for 30 min to form a stable mixture of GO and aniline. Next, 0.2281 g (NH₄)₂S₄O₈ was poured into the solution and ultrasonic for 2 h at 20 °C. When the color of the solution came into green, the GO–PANI composites were synthesized successfully. After isolated by centrifugation and rinsed with ultrapure water for several times, the obtained polymers was dried at 60 °C in vacuum. For further using in the experiment, the prepared GO–PANI composites were dispersed in dimethylformamide with ultrasonication for 1 h to get a homogenous suspension (1 mg mL⁻¹).

Fabrication of the aptasensor

Scheme 1 showed the procedure for the fabrication of the aptasensor. The GCEs were polished with 0.3 and 0.05 μm alumina slurries, respectively. Then they were ultrasonically treated in ethanol and ultrapure water for 5 min. Next, the electrodes were electrochemically cleaned in 0.5 M H₂SO₄ by potential scanning between −1 V and +1 V for 20 cycles, followed by washing them with ultrapure water.

Scheme 1 Schematic representation of aptasensor based on GO–PANI composites as the signal amplifiers for the competitive detection of OTC.

The prepared GO–PANI (1 mg mL⁻¹) composites were dropped on the surface of electrodes via electrostatic adsorption force and dried at RT for 4 h. After washed with ultrapure water for 6 times, the electrodes were immersed into HAuCl₄ (1 mg mL⁻¹) solution. Under stirring, gold nanoparticles (AuNPs) were electrodeposited on the surface of electrodes via current–time curve technique at −0.2 V for 120 s. After dried, the OTC-BSA was dropped on the electrode surface based on the binding between the amino group of BSA and AuNPs. OTC-BSA was synthesized by OTC as the target and BSA as carrier. The electrodes were incubated overnight at 4 °C and then rinsed with PBS. Next, the electrodes were covered with the solution of BSA (0.1%) in PBS for 2 h to block non-specific binding sites and washed again with PBS. Finally, the mixture of horseradish peroxidase labeled aptamer (HRP–Apt) and OTC samples were added onto the electrode surface at the same time. The OTC in the samples competed with those of OTC-BSA on the electrodes to bind aptamer. At last, the competitive aptasensor was fabricated successfully.

CV measurements

The CV measurements were carried out in PBS containing the mixture of HQ and H₂O₂ over a potential range from −0.4 to +0.4 V at a scan rate of 50 mV s⁻¹. The experiments were performed thoroughly at RT.

Results and discussions

Charge transport properties of the composites

PANI has attracted considerable interest because of its high capacitive characteristic, good electrochemical activity, and ease of synthesis. GO, a chemically modified grapheme sheet, not only owns the two-dimensional structure, but possesses high surface area, excellent mechanical properties, and conductivity. Incorporation of PANI in GO can promote the composite in its higher electrochemical capacitance and charge–discharge cyclic stability. AuNPs, which are modified on the surfaces of the electrode by abundant amino groups of PANI, can lead to the enhancement of electrical conductivity. In addition, the complete antigen (OTC-BSA) can be combined on the surfaces of AuNPs by utilizing the action of Au–NH₂ bond. Therefore, the utilizing of AuNPs–GO–PANI can improve significantly the performance in electrochemical signal transduction. In this study, we used CV to monitor the electrons transmission procedure of the modified electrodes. The working solution was 0.1 M PBS (pH 7.4) containing K₃[Fe(CN)₆]. As shown in Fig. 1A, with the bare GCE electrode, a pair of well-defined current peaks appears at 0.17 mV and 0.26 mV, a typical redox peak range of K₃[Fe(CN)₆] (curve a). With the electrode modified by the GO–PANI composites, there were significantly increased peak current in the CV curves (curve b). Moreover, a stronger CVs signal was observed after the electrodeposition of AuNPs on the electrodes, indicating AuNPs adsorbed on the surface of the GO–PANI composites (curve c). After incubation with OTC-BSA, the peak current decreased obviously because of the hindering of electronic transfer by the biological macromolecules (curve d). Additionally, we performed a control experiment in which PANI and GO are analyzed respectively. When the electrode is coated with GO, there is remarkably decreased peak current of redox peak, which is likely attributed to the carboxyl groups with negative charges on GO surface (curve e). A further control experiment has been performed using PANI in place of the GO–PANI composites. It is observed that the peak current of the GO–PANI composites is obviously higher than that of PANI (curve f). These findings give clear evidence that the GO–PANI composites improve the performance in electrochemical signal transduction and significantly facilitate the electron transfer between the electrodes and the working solution.

Fig. 1 (A) CVs obtained for the bare GCE (a), GO–PANI/GCE (b), AuNPs/GO–PANI/GCE (c), BSA-OTC/AuNPs/GO–PANI/GCE (d), GO/GCE (e), PANI/GCE (f); Measurements were performed in PBS containing of 5 mM of K₃[Fe(CN)₆] and 0.2 M KCl. (B) CVs of aptasensor obtained in 10 mM PBS (PH 7.4, 140 mM NaCl, 5 mM KCl, 1 mM MgCl₂, 1 mM CaCl₂) containing 9 mM HQ and 1 mM H₂O₂ before (a) and after (b) reaction with 1 × 10⁻¹ mg L⁻¹ of OTC solution and 1 μM HRP–Apt solution. Scan rate is 50 mV s⁻¹.

Responses characteristics of GO–PANI composites-based competition assay

Fig. 1B shows typical CV responses using the developed technique. In the absence of OTC, the CV curves showed a strong current peak at −0.1 mV, a typical reduction peak of HQ, indicating the HRP–Apt probe reacted with BSA-OTC immobilized on the electrode (curve a). In contrast, an obviously decreased peak current of redox peak was observed in the presence of 1 × 10⁻¹ mg L⁻¹ OTC solution, which suggested that the HRP–Apt probe bound to target OTC (curve b). Thus, the developed GO–PANI composites-based competition assay method offered a possibility for highly selective detection of the target OTC.

Electrochemical behaviors

Fig. 2 showed the CVs of the GCE/GO–PANI/AuNPs/OTC/Apt–HRP electrode in PBS buffer containing H₂O₂ at different scan rates from 10 mV s⁻¹ to 200 mV s⁻¹. Each voltammogram in Fig. 2 showed a series of well-defined reduction peaks. The cathodic peak currents of HQ at the modified electrodes were linearly proportional to the square root of the scan rates in the range of 10–200 mV s⁻¹, which showed a characteristic of a diffusion-controlled electrochemical process.

Fig. 2 Cyclic voltammograms of aptasensor obtained in 10 mM PBS (PH 7.4, 140 mM NaCl, 5 mM KCl, 1 mM MgCl₂, 1 mM CaCl₂) containing 9 mM HQ and 1 mM H₂O₂ after incubation with 1 × 10⁻¹ mg L⁻¹ OTC solution and 1 μM HRP–Apt solution at different scan rates: 10 mV s⁻¹ (a), 50 mV s⁻¹ (b), 100 mV s⁻¹ (c), 150 mV s⁻¹ (d), 200 mV s⁻¹ (e). (Inset: plot of peak currents vs. scan rate).

Characteristics of morphology

The morphology and microstructure of all samples were investigated by SEM technique. Fig. 3a clearly illustrated the surfaces of the composites were very rough and the most of GO surface was covered with worm-like structure or tube-like particles. The result indicated that the GO–PANI composites were synthesized as expected. After the HAuCl₄ reduction on the electrodes, dense gold nanoparticles were observed adsorbed on the surface of the GO–PANI composites (Fig. 3b). These results demonstrated that the modification of the electrodes was successful.

Fig. 3 SEM images of the GO–PANI composites solution deposited on the electrode (a) and the AuNPs/GO–PANI composites solution deposited on the electrode (b).

Fig. 4 showed the TEM pictures of the GO, PANI and PANI–GO composite. We observed from the TEM images that GO displayed an ultra-thin paperlike morphology and PANI exhibited typical nanofibers with large length-to-diameter ratios (Fig. 4a and b). After growing PANI from the surface of GO, PANI nanofibers appeared to surround the GO nanosheets (Fig. 4c). Therefore, GO–PANI composites were synthesized successfully and the result was consistent with SEM.

Fig. 4 TEM images of the GO (a), the pure PANI (b) and the GO–PANI composites (c).

Optimization of the reaction time and concentration of the HAuCl₄ solution

The quantity and thickness of Au monolayer were directly decided by the time of HAuCl₄ electrochemical reduction. Then they affected the subsequent fixed quantity of OTC and the current response, which further affected the performance of the sensor. If the reduction time exceeded a certain value, it would likely make Au layer become thick and hinder the electronic transmission, then, affect the response of the sensor. The time of reduction was selected in working solution containing K₃[Fe(CN)₆] by CV technique. As shown in Fig. 5A, the current response reached a maximum at 120 s. So we considered the optimal time for HAuCl₄ reduction as 120 s. The effect of HAuCl₄ concentration was also studied and the results were shown in Fig. 5B. The current responses were larger with the increasing of HAuCl₄ concentration till 1 mg mL⁻¹ and remained nearly unchanged for larger concentration. Thus, the HAuCl₄ concentration of 1 mg mL⁻¹ was selected as an optimal condition.

Fig. 5 (A) Effect of electrochemical reduction time of the HAuCl₄ solution on the CV peak current of aptasensor. (B) Effect of the concentration of HAuCl₄ solution on the CV peak current of aptasensor. (C) Effect of the concentration of OTC-BSA solution on the CV peak current of aptasensor. (D) Effect of incubation time for surface hybridization reaction with 1 × 10⁻¹ mg L⁻¹ OTC on the CV peak current of aptasensor. (E) Effect of the incubation buffer with different PH on the CV peak current of aptasensor.

Optimization of the concentration of OTC-BSA

The insulative of antigen prevents the spread of the redox probe to the electrode surface. We used CV technique in the working solution containing K₃[Fe(CN)₆] to explore the fixed OTC-BSA concentration on the influence of current signal. The results were shown in Fig. 5C. The current decreased with the increasing of modified OTC-BSA concentration until 0.01 mg mL⁻¹. And from 0.01 to 0.1 mg mL⁻¹, there was no obvious current change. The reason was that the antigen concentration in the binding sites had reached saturated, further increasing concentration will not affect the change of the current signal. Therefore, the 0.01 mg L⁻¹ of modified OTC-BSA was chosen as the best concentration.

Optimization of incubation time

The OTC immobilized on the electrodes competed with those under test to bind HRP–Apt. The strong binding force was based on the specific recognition between aptamer and OTC. And the incubation time of HRP–Apt was vital for the performance of the sensor. The effect of the competitive time between the two kinds of OTC was investigated by CV technique in PBS solution containing the mixture of HQ and H₂O₂. As shown in Fig. 5D, with the increase of binding time, current response value increased among 10 to 30 min. This showed that with the increase of time, a growing number of HRP was fixed to the electrode surface. And the catalytic reduction effect of H₂O₂ became better along with current signal changing stronger. On the contrary, within the scope of the 30 to 80 min, current response value decreased gradually. This may be because the OTC to be detected carried off more HRP with the increase of time, leading to the decrement of HRP amount on the electrodes. Then it affected the weak current response. Therefore, 30 min was selected for optimal incubation time.

Optimization of the pH of the working solution

The pH of working solution was also a very important influence for the performance of the aptasensor. Different pH showed obvious change of current response. In our experiment, the pH in the range of 6 to 8 was discussed via the CV technology. The results were shown in Fig. 5E, the current responses significantly increased from pH 6 to 7, but decreased from 7 to 8. So we chose 7 as the optimal pH of working solution.

Calibration curve of aptasensor

To analyze the performance of the aptasensor, we measured the current response after the aptasensor immersed in different concentrations (2 × 10⁻⁶ to 1 × 10³ mg L⁻¹) of the OTC under test by the CV. The parameters were set as follows: voltage range of −0.4 to +0.4 V and the scan rate of 0.05 V s⁻¹. As shown in Fig. 6, the linear regression equation was I (10⁻⁶ A) = 15.4095 − 4.2350 × log C (C is the concentration of OTC (mg L⁻¹)) with a correlation coefficient of 0.9957. The linear range of prepared sensor was from 4 × 10⁻⁶ mg L⁻¹ to 1 mg L⁻¹. And the detection limit was 2.3 × 10⁻⁶ mg L⁻¹ at a signal-to-noise ratio of 3 (S/N = 3), which was better than that of previous report.⁴¹

Fig. 6 Calibration curve of CVs peak currents for different OTC concentrations. Error bars are standard deviations across three repetitive experiments.

The selectivity and reproducibility of the aptasensor

The specificity of the aptasensor also played an important role in the analysis of the samples. To evaluate the selectivity of the sensor, we chose chlortetracycline and tetracycline as interfering substance based on the similar structure and type of interaction. As shown in Fig. 7, it was revealed that these interferences had little effect on our OTC assay via HRP labeled aptamer. This might attribute that although the aptamer was immobilized on the electrodes, it still retained the specific recognition with OTC. The results showed that the prepared aptasensor exhibited good selectivity for OTC.

Fig. 7 CVs of aptasensor obtained for different sample solution containing 1 × 10⁻¹ mg L⁻¹ OTC (a), 1 mg L⁻¹ chlortetracycline (b), 1 mg L⁻¹ tetracycline (c), 1 × 10⁻¹ mg L⁻¹ OTC and 1 mg L⁻¹ chlortetracycline (d), 1 × 10⁻¹ mg L⁻¹ OTC and 1 mg L⁻¹ tetracycline (e), 1 × 10⁻¹ mg L⁻¹ OTC, 1 mg L⁻¹ chlortetracycline and 1 mg L⁻¹ tetracycline (f).

In order to detect the repeatability of sensors, we selected five electrodes with the same assembly step to test 1 × 10⁻¹ mg L⁻¹ OTC, respectively. The relative standard deviation was 4% for five independent determinations. The experiment results showed that the fabricated sensor presented good reproducibility. Additionally, the stability of the sensor was investigated in a period of 14 days. The aptasensor was stored in the refrigerator at 4 °C for the detection of 1 × 10⁻¹ mg L⁻¹ OTC. Two weeks later, it still retained 92% of the original current response. The results demonstrated that the aptasensor had good stability.

The detection of real samples

In order to value the actual application of the aptasensor, the recovery rates of different concentrations of OTC in honey were detected. The pretreatment of the sample was according to the following steps. Firstly, honey was diluted 10 times with PBS buffer, and then centrifuged 30 min under the speed of 16 000 rpm at 4 °C. The purpose was to remove the fat and protein. After adding into the samples with known concentrations of OTC respectively, recovery rates were tested through our sensor. The results were shown in Table 1. The recoveries were ranged from 97.2% to 101%, which indicated that developed aptasensor had good recovery and practicability.

Table 1 OTC analysis in real spiked samples

Samples	Added (mg L^−1)	Total found (mg L^−1)	Recovery (%)
Honey	1.00 × 10^−1	9.83 × 10^−2	98.3%
	1.00 × 10^−1	9.72 × 10^−2	97.2%
	1.00 × 10^−1	1.01 × 10^−1	101%

---

Conclusions

In this study, we investigated a competitive aptasensor for oxytetracycline detection based on GO–PANI composites and HRP as the signal amplifiers. The sensor was decorated by GO–PANI composites, AuNPs and HRP. The modification steps of the electrodes were characterized by CV and SEM technology which proved the successful fabrication of the aptasensor. It was shown that our sensor had good selectivity from the interfering substance. And it also exhibited the advantages of wide linear range, low limit of detection, acceptable reproducibility and stability. This aptasensor has been successfully applied to the detection of OTC in actual samples.

Acknowledgements

This work was supported by the National Natural Science Foundation of the People's Republic of China (no. 31171700 and 31101296), the National High Technology Research and Development Program of China (National 863 Program of China) (no. 2012AA101604), the Natural Science Foundation of Shandong Province (no. ZR2010DQ025) and the Shandong Province Higher Educational Science and Technology Program (no. J10LB14).

References

1. Y. K. Lv, L. M. Wang, L. Yang, C. X. Zhao and H. W. Sun, J. Chromatogr. A, 2012, 1227, 48–53.
2. X. Q. Yang, C. X. Yang and X. P. Yan, J. Chromatogr. A, 2013, 1304, 28–33.
3. M. Guarddon, J. Miranda, J. Rodríguez, B. Vázquez, A. Cepeda and C. Franco, Int. J. Food Microbiol., 2011, 146, 284–289.
4. H. J. Koo and G. J. Woo, Int. J. Food Microbiol., 2011, 145, 407–413.
5. A. Shalaby, N. A. Salama, S. Abou-Raya, W. H. Emam and F. Mehaya, Food Chem., 2011, 124, 1660–1666.
6. N. Vragović, D. Bažulić and B. Njari, Food Chem. Toxicol., 2011, 49, 352–355.
7. A. B. SN, E. AE and M. IY, Asian Pac. J. Trop. Med., 2013, 6, 718–722.
8. H. Yu, H. Mu and Y. M. Hu, J. Pharm. Anal., 2012, 1, 76–81.
9. B. Deng, Q. X. Xu, H. Lu, L. Ye and Y. Z. Wang, Food Chem., 2012, 134, 2350–2354.
10. Y. X. Guo, L. Meng, Y. H. Zhang, W. Tang, W. F. Zhang, Y. Xia, F. G. Ban, N. P. Wu and S. S. Zhang, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2013, 942–943, 151–157.
11. C. Brentnall, Z. Cheng, Q. McKellar and P. Lees, Res. Vet. Sci., 2013, 94, 687–694.
12. W. R. Chen and C. H. Huang, Environ. Pollut., 2011, 159, 1092–1100.
13. J. Li, F. Jiang, Y. Li and Z. Chen, Biosens. Bioelectron., 2011, 26, 2097–2101.
14. L. Norambuena, N. Gras and S. Contreras, Mar. Pollut. Bull., 2013, 73, 154–160.
15. I. Pérez-Silva, J. A. Rodríguez, M. T. Ramírez-Silva and M. E. Páez-Hernández, Anal. Chim. Acta, 2012, 718, 42–46.
16. G. Chen, G. Liu and F. Qin, Food Chem., 2011, 127, 264–269.
17. K. Škrášková, L. H. M. L. M. Santos, D. Šatínský, A. Pena, M. C. B. S. M. Montenegro, P. Solich and L. Nováková, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2013, 927, 201–208.
18. G. Kahsay, F. Shraim, P. Villatte, J. Rotger, C. Cassus-Coussère, A. Van Schepdael, J. Hoogmartens and E. Adams, J. Pharm. Biomed. Anal., 2013, 75, 199–206.
19. Y. S. Kim, J. H. Niazi and M. B. Gu, Anal. Chim. Acta, 2009, 634, 250–254.
20. Y. Bao, J. Song, Y. Mao, D. Han, F. Yang, L. Niu and A. Ivaska, Electroanalysis, 2011, 23, 878–884.
21. S. Chen, J. Zhu and X. Wang, J. Phys. Chem. C, 2010, 114, 11829–11834.
22. S. Li, K. Tu, C. Lin and C. Chen, ACS Nano, 2010, 4, 3169–3174.
23. W. L. Zhang, B. J. Park and H. J. Choi, Chem. Commun., 2010, 46, 5596–5598.
24. J. S. Do and S. H. Wang, Sens. Actuators, B, 2013, 185, 39–46.
25. A. A. Khan and L. Paquiza, Synth. Met., 2011, 161, 899–905.
26. J. Xu, K. Wang, S. Z. Zu, B. H. Han and Z. Wei, ACS Nano, 2010, 4, 5019–5026.
27. J. Chen, Z. Fang, J. Liu and L. Zeng, Food Control, 2012, 25, 555–560.
28. K. Guo, Y. Wang, H. Chen, J. Ji, S. Zhang, J. Kong and B. Liu, Electrochem. Commun., 2011, 13, 707–710.
29. L. Li, H. Zhao, Z. Chen, X. Mu and L. Guo, Biosens. Bioelectron., 2011, 30, 261–266.
30. L. Huang, X. Yang, C. Qi, X. Niu, C. Zhao, X. Zhao, D. Shangguan and Y. Yang, Anal. Chim. Acta, 2013, 787, 203–210.
31. Z. Lin, H. Huang, Y. Xu, X. Gao, B. Qiu, X. Chen and G. Chen, Talanta, 2012, 103, 371–374.
32. X. Zhu, Y. Zhang, W. Yang, Q. Liu, Z. Lin, B. Qiu and G. Chen, Anal. Chim. Acta, 2011, 684, 121–125.
33. Y. Li, J. Bao, M. Han, Z. Dai and H. Wang, Biosens. Bioelectron., 2011, 26, 3531–3535.
34. J. Ashley and S. F. Li, Biosens. Bioelectron., 2013, 48, 126–131.
35. Y. Chai, D. Tian and H. Cui, Anal. Chim. Acta, 2012, 715, 86–92.
36. L. Xie, L. You and X. Cao, Spectrochim. Acta, Part A, 2013, 109, 110–115.
37. M. Li, X. Zhou, S. Guo and N. Wu, Biosens. Bioelectron., 2013, 43, 69–74.
38. X. Zhang, Y. Xu, Y. Zhao and W. Song, Biosens. Bioelectron., 2013, 39, 338–341.
39. M. J. Waner and D. P. Mascotti, J. Biochem. Biophys. Methods, 2008, 70, 873–877.
40. W. S. Hummers Jr and R. E. Offenman, J. Am. Chem. Soc., 1958, 80, 1339.
41. F. T. Moreira, A. H. Kamel, J. R. Guerreiro and M. G. F. Sales, Biosens. Bioelectron., 2010, 26, 566–574.

---