A porous CuO nanowire-based signal amplification immunosensor for the detection of carcinoembryonic antigens

Abstract

A novel electrochemical immunosensor was developed for carcinoembryonic antigens (CEA) based on gold nanoparticle load carbon nanotubes (CNTs–AuNPs) as an immunosensor platform and porous CuO nanowire supported ferrocene (pCuOw@Fc) as signal amplification labels. The pCuOw were prepared by a simple decomposition of the Cu(OH)₂ precursor. The products were characterized using transmission electron microscopy (TEM), X-ray diffraction (XRD) and Fourier transform infrared spectroscopy (FT-IR). The electrochemical performance of the resulting immunosensor was investigated by cyclic voltammetry (CV), and the electrochemical immunosensor showed enhanced electrochemical performance toward the detection of CEA with a range from 0.005 ng mL⁻¹ to 80 ng mL⁻¹ and a detection limit of 0.0008 ng mL⁻¹ (S/N = 3). The proposed immunosensor was successfully used in determining the CEA in real samples and holds great potential for the sensitive electrochemical biosensing of other analytes.

---

1. Introduction

Carcinoembryonic antigens (CEA) in human serum or tissue, a common cancer marker, are widely used in clinical screening and diagnosis, because their levels provide essential information for clinical research and diagnosis of some types of cancer such as liver, colon, breast and colorectal cancer.^1–3 Normally, the levels of CEA are uncommonly lower in colon tissue of healthy adults,⁴ however, the levels of CEA correlated with cancer incidents may be suspected if the levels of CEA significantly increase. Thus, developing a reliable, sensitive and precise analytical method for the detection of CEA is highly desirable in diagnosing and treating carcinomas. Recently, various immune-based assay methods and strategies have been proposed for the sensitive evaluation of CEA, such as chemiluminescence assays,^5,6 enzyme-linked immunosorbent assay,⁷ mass spectrometric immunoassays,⁸ and electrophoretic.⁹ Nevertheless, some of those methods suffer from drawbacks such as time-consuming and complex operation procedure. Therefore, the improvements of assay methods are required for developing simple, rapid, no radiation and sensitive to detect CEA.

Recently, much attention has focus on the sandwich-type electrochemical immunoassay because of its fast analytical time, high specificity and sensitivity.^10–12 Ferrocene (Fc) and its derivatives, an excellent probes have high redox activity and used in electrochemical systems as a protein biomarker or redox active species for the detection of biomolecules detection, because of their good reversibility, regeneration at low potential, and generation of stable redox states.^13–17 Recently, many ferrocene-based electrochemical sensors have been fabricated and applied.^18,19 However, in most of these sensors, ferrocene derivatives were noncovalently linked or adsorption on the surface of electrode, the stability of such sensor is poor because ferrocene derivatives easily leak from electrode surface. Compared with noncovalently linked or adsorption, covalently linking ferrocene derivatives on the electrode surface has fascinating advantages. In particular, the stability of electrochemical sensor has been enhanced.

To obtain excellent analytical performance of electrochemical sensor, a variety of methods have been developed by using several nanomaterials to obtain signal amplification on account of their excellent catalytic performance.²⁰ Semiconductor nanomaterials have aroused strong interests due to their excellent electronic and optical properties. CuO, a well-known p-type metal oxide semiconductor with a narrow band-gap, are the most widely used in the fabrication of optical and photovoltaic devices, heterogeneous catalysis, magnetic storage media and lithium ion electrode materials²¹ due to its low production cost, high stability and easy availability, as well as could act as catalysts of the reduction of hydrogen peroxide (H₂O₂) and oxidation of glucose^22–24 because it possess the excellent electrocatalytic activity. Up to now, various nanostructured CuO materials such as nanoflowers,^25,26 nanorods,²⁷ nanospheres,²⁸ nanosheets,²⁹ nanofilms,³⁰ hollow microspheres,³¹ nanowire,^32,33 nanoporous^34,35 and nanofibres³⁶ have been synthesized to prepare various sensors. The synthesis of the various CuO nanostructures is very important for a number of applications. The investigation on the morphology of CuO nanomaterial has attracted great attention over the past decades for the great influence on the performance of CuO, while porous CuO have been extensively investigated due to excellent electrochemical performance. Wang et al. reported a porous CuO nanowire as the anode of rechargeable Na-ion batteries. The as-prepared porous CuO nanowires exhibit a Brunauer–Emmett–Teller (BET) surface area is six times larger than that of bulk CuO. This result that Na-ion batteries shows excellent performance.³⁷ However, as far as we know, no work has been reported for the electrochemical performance of pCuOw in electrochemical immunosensors.

In recent years, CNTs have attracted tremendous attention duo to its unique advantages, such as good biocompatibility and specific electronic properties.^38,39 Meanwhile, AuNPs are used increasingly in many electrochemical applications because of its good electrode conductivity to facilitate the electron transfer and biocompatibility.⁴⁰ The CNTs surface decorations via AuNPs have been performed by a simple chemical method.

Herein, we present a novel electrochemical immunosensor for the detection of CEA by using CNTs–AuNPs as immunosensor platform and pCuOw@Fc/AuNPs as signal amplification probe. The strategy overcomes the leak of ferrocenecarboxylic acid from electrode surface. The pCuOw as carrier has a large specific surface area and could hasten the decomposition of H₂O₂ with amplified signal for the detection of CEA in the presence of H₂O₂. Thus, the established strategy provided a potential consideration for the design of electrochemical immunosensor in the detection of other cancer markers.

2. Experimental

2.1. Apparatus and reagents

This section was provided in ESI.†

2.2. Preparation of the pCuOw@Fc/AuNPs–Ab₂ labels

This preparation section was provided in ESI.†

2.3. Immunosensor fabrication

Fabrication before, the glassy carbon electrode (GCE, 4 mm diameter) was polished down to mirror-like with 0.5 and 0.05 μm alumina slurry respectively, and then was successively sonicated in ethanol and doubly distilled water respectively, allowed to dry at room temperature.

First, 10 μL CNTs–AuNPs (1 mg mL⁻¹) were added on the electrode surface and dried. Secondly, Ab₁ was immobilized on CNTs–AuNPs modified electrode surface, and then blocking nonspecific binding sites of the electrode with BSA. Then, the modified electrode was incubated in different concentrations of CEA solution for 35 min at room temperature. As a sandwich format, the anti-CEA/CNTs–AuNPs/GCE electrode was immersed in the pCuOw@Fc/AuNPs–Ab₂ for immune-reaction and then was washed. The obtained immunosensor was stored at 4 °C. The synthesis process of signal probe and the schematic diagram of a stepwise assembled procedure of the immunosensor were shown in Scheme 1.

Scheme 1 Preparation procedure (A) of Ab₂ bioconjugates (pCuOw@Fc/AuNPs–Ab₂) and schematic illustration (B) of stepwise electrochemical immunosensor fabrication process and signal amplification.

2.4. Measurements and characterization

Based on sandwich-type for the detection of CEA, the resultant pCuOw@Fc/AuNPs–Ab₂/anti-CEA/CNTs–AuNPs/GCE electrode was performed in PBS (pH 7.4) solution at room temperature. The CV of the different concentrations of CEA was recorded in PBS (pH 7.4) solution containing 0.1 M KCl.

3. Results and discussion

3.1. Characterization of pCuOw

The morphology of the pCuOw was examined using TEM. Fig. 1A presents typical slow magnification TEM of the pCuOw, it was disorganized and consist of a large amount of 1 dimensional structure about 0.5–1 micrometers in length. There was a gap between nanoparticles vaguely. The high magnification TEM shown in Fig. 1B, and further reveals that these 1D structures were wires with widths in the range of 10–20 nm, the nanowire consists of many interconnected nanoparticles with sizes of ∼20 nm and accumulated pores. The XRD pattern of the Cu(OH)₂ precursor and the pCuOw were illustrated in Fig. 1C. Diffraction peaks were observed with 2θ of 16.9°, 24.0°, 34.3°, 36°, 38.3°, 39.9°, 53.5°, 55.4°, 63.3° and 65° (curve b), and these peaks are attributable to the (020), (021), (002), (111), (041/022), (130), (150/132), (061/023), (200/062/043) and (152) planes of orthorhombic crystal Cu(OH)₂ (JCPDS file no. 13-420). When the Cu(OH)₂ precursor was calcined in air for 4 h, the pure pCuOw product phase was obtained with 2θ of 32.5°, 35.5°, 38.7°, 46.6°, 48.8°, 53.4°, 58.3°, 61.6°, 66.3°, and 68.0° (curve a). These peaks were assigned to the (−110), (002/−111), (111/200), (−112), (−202), (020), (202), (−113), (022/−311) and (113/−202) planes of monoclinic CuO (JCPDS file no. 45-937). No impurity peaks of other copper oxides or precursor Cu(OH)₂ were observed after calcined, indicating the pure structure of the products.

Fig. 1 TEM (A and B) of porous CuO nanowires, XRD patterns (C) of pCuOw (a) and Cu(OH)₂ nanowires (b), FT-IR spectrum (D) of pCuOw (a), pCuOw–NH₂ (b) and pCuOw–Fc (c).

The FT-IR of the pCuOw and the modified pCuOw were displayed in Fig. 1D. The broad peak between 3200 cm⁻¹ and 3600 cm⁻¹ and the peak at 1630 cm⁻¹ were observed for all samples, due to O–H stretching and deformation vibrations of weakly bound water. The peak at 585 cm⁻¹ was a characteristic peak of CuO stretching (curve a). When the pCuOw was modified with APTES, there are additional peaks appeared (curve b). The newly-presented peaks at 2870 cm⁻¹ and 2934 cm⁻¹ were attributed to stretching vibrations of the –CH₂– group of aminopropyl, the broad peak at 3438 cm⁻¹ corresponds to the amino functional groups of the APTES that partly overlapped with the O–H stretching vibration. Due to the triethoxy of APTES has a reaction with pCuOw to form –Cu–O–Si–CH₂CH₂CH₂–NH₂, which verifies the aminopropyl of APTES has connected to the pCuOw. After pCuOw–NH₂ reacted with Fc–COOH, the characteristic peak at 1400 cm⁻¹ and 1645 cm⁻¹ were observed (curve c), which were the result of bonding between the –NH₂ of pCuOw–NH₂ surface and –COOH of Fc–COOH, as well as the absorption bands at about 480 cm⁻¹ (Fe-Cp) appeared. This indicated the successful formation of pCuOw–Fc.

3.2. Characterization of electrochemical behavior

The electrochemical behavior of each step modified electrode process were carried out in 0.2 M PBS (pH 7.4) solution containing 5.0 mM [Fe(CN)₆]^4−/3− and 0.1 M KCl by CV in the potential range from −0.4 to 0.8 V. As shown in Fig. 2A, a couple of reversible redox peak were observed at a bare GCE electrode (curve a), due to a one-electron process of [Fe(CN)₆]^3−/4− probe. The peak current is higher (curve b) in compared with bare GCE when the bare electrode was modified with CNTs–AuNPs. Because CNTs–AuNPs have a larger surface and could to promotes electron transfer from [Fe(CN)₆]^3−/4− to the electrode surface. However, the peak current response of the immunosensor was decreased (curve c) when anti-CEA was immobilized on the CNTs–AuNPs/GCE electrode surface. The reason was that a large of anti-CEA was immobilized on the surface of CNTs–AuNPs/GCE electrode surface, and the protein formed a block layer on the surface of modified electrode to block the electron transfer of [Fe(CN)₆]^3−/4− probe. Then, the peak current response further decreased in turn (curve d and e), after the modified electrode was incubated in BSA and CEA in turn. This phenomenon resulted from the electron inert feature of BSA and CEA, blocking the electron transfer of [Fe(CN)₆]^3−/4− at the electrode surface. Electrochemical impedance spectroscopy (EIS) was also an important tool for monitoring the impedance changes of modified electrodes surface. To further monitor the surface conditions of the modified electrodes, EIS has been employed to characterize the surface properties of different modified electrodes (Fig. S2†). The results which were good consistent with those obtained through CV measurements demonstrated the successful fabricated process of the electrochemical immunosensor.

Fig. 2 CVs (A) of different modified electrodes in 5.0 mM [Fe(CN)₆]^3−/4− containing 0.1 M KCl: (a) bare GCE electrode, (b) CNTs–AuNPs/GCE, (c) anti-CEA/CNTs–AuNPs/GCE, (d) BSA/anti-CEA/CNTs–AuNPs/GCE, (e) CEA/BSA/anti-CEA/CNTs–AuNPs/GCE, scan rate of 50 mV s⁻¹. CVs (B–E) responses of different immunosensor without (a) and with (b) H₂O₂ in pH 7.4 PBS towards CEA.

3.3. Comparison of different immunosensor

In this study, pCuOw@Fc/AuNPs and CNTs–AuNPs were employed as signal label and platform, respectively. Here, the mechanism of the electrochemical response could be described as follows:^41,42

1/2H₂O₂ + Cu₂O → OH⁻ + CuO (1)

CuO + Fc⁺ → Cu₂O + Fc (2)

Fc → Fc⁺ + e⁻ (3)

Here, it was achieved as the following steps: (i) when CEA were captured on the Ab₁-immobilized the GCE electrode surface, the pCuOw@Fc/AuNPs–anti-CEA signal label could specific recognized with CEA. So pCuOw and Fc were introduced on the anti-CEA/CNTs–AuNPs/GCE modified electrode surface. (ii) pCuOw could catalyze the reduction reaction of Fc⁺ by H₂O₂. During this progress, Fc⁺ was reduced to Fc. (iii) The Fc was oxidized back to Fc⁺ by a rapid reaction involving the loss of one electrons on the electrode surface, at the same time, the oxidation peak currents response were recorded by CVs.

Signal amplification is critical to obtain high sensitivity and low detection limit for the sandwich-type electrochemical immunosensor. To further to verify the effect of each individual component of the proposed immunosensor to electrochemical properties, AuNPs instead of CNTs–AuNPs was used to fabricate the immunosensor platform, Fc/AuNPs and pCuOw@Fc instead of pCuOw@Fc/AuNPs were used to fabricate the signal probes for the detection of CEA (Fig. 2). The same batch immunosensor were incubated in 1 ng mL⁻¹ CEA, then, the prepared electrochemical immunosensor was carried out in 0.2 M PBS solution containing 0.1 M KCl with the absence and presence of H₂O₂. The curve (a) and curve (b) in Fig. 2 shown CVs before and after the addition of 2.5 mM H₂O₂, respectively. As shown in Fig. 2, the immunosensor exhibited larger peak current using CNTs–AuNPs as platform (Fig. 2E) than using AuNPs (Fig. 2C). The reason might be attributed to the fact that CNTs have excellent electrical conductivity. Furthermore, the use of pCuOw@Fc/Ab₂ bioconjugates (Fig. 2D) as probe offered greater peak current shift than that obtained with Fc/AuNPs–Ab₂ bioconjugates (Fig. 2B) as label. The experiment result confirmed that the pCuOw exhibits efficient catalysis towards H₂O₂, illustrating the good peroxidase-like catalytic activity, and have a larger specific surface area that could immobilized a lot of Fc. More inspiringly, it can be found that the use of pCuOw@Fc/AuNPs bioconjugates as probe offers highest current shift (Fig. 2E) than those obtained with other two labeled probe. Suggesting the excellent catalytically amplification of the proposed bioconjugates. Accordingly, pCuOw@Fc/AuNPs–Ab₂ bioconjugates was chosen as a trace for immunosensor to obtain signal amplification.

3.4. Performance analysis

The detection capability of the proposed immunosensor was tested using various concentrations of CEA under optimal experimental conditions by CV (Fig. S3†). The peak currents of the immunosensor increased with the increasing of CEA concentration in the range from 0.005 ng mL⁻¹ to 80 ng mL⁻¹ (inset of Fig. 3A). It was noted that a linear relationship between the CV peak currents and the logarithm of CEA concentration range from 0.005 ng mL⁻¹ to 80 ng mL⁻¹ and the detection limit was 0.0008 ng mL⁻¹ (S/N = 3) with the linear regression equation: i = 13.50 + 3.23 log C_CEA (ng mL⁻¹) (R² = 0.997) (Fig. 3). The detection range of this method was much wider than that of the previously reported CEA assays and the detection limit was also much lower than that of the previously reported. The results were shown in Table 1.

Fig. 3 CVs (A) of the proposed immunosensor after incubation with increasing CEA concentration from 0.005 ng mL⁻¹ to 80 ng mL⁻¹ (inset) and calibration curves for detection of CEA in 0.2 M PBS, CVs (B) of the immunoassay with various signal probes, (a) pCuOw@Fc/AuNPs–Ab₂ (b) pCuOw@Fc/Ab₂ (c) Fc/AuNPs–Ab₂ toward various concentrations of CEA, using CNTs–AuNPs as the substrate.

Table 1 Comparison of analytical properties of different immunosensors

Immunosensor	Linear range (ng mL^−1)	Detection limit (ng mL^−1)	Reference
AuNPs–HRP–Ab2/CEA/Ab1–HPDMPA–ITO	0.01–80	0.00236	3
Ab/AuNPs/(PB–NPs/rGO–MWCNTs)n/GCE	0.2–1, 1–40	0.06	43
Anti-CEA/AuNPs/Thi@AuNPs/GCE	0.01–100	0.003	44
CEA/HRP/HRP–anti-CEA/Au–GN/GCE	0.1–80	0.04	45
pCuOw@Fc–AuNPs–Ab2/CEA/Ab1–CNTs–AuNPs/GCE	0.005–80	0.0008	This work

---

In order to verify the effect of each individual component of the signal probes, Fc/AuNPs–Ab₂ and pCuOw@Fc/Ab₂ were used to fabricate the probes for the detection of CEA (Fig. 3B). The proposed electrochemical immunosensor using pCuOw@Fc/AuNPs–Ab₂ as signal probe exhibited higher sensitivity and wider linear than the other two signal probes. The reason for this might be that the pCuOw have a larger specific surface area and could hasten the decomposition of H₂O₂ with amplified signal.

3.5. Specificity, reproducibility and stability of the immunosensor

The specificity of electrochemical immunosensor was evaluated through examining its responses to common interfering, such as AFP, BSA, HRP and DA (Fig. 4). Compared with its response to 1 ng mL⁻¹ of CEA, very weak peak current change (less than 3.2%) was observed upon addition of any of the interfering at a concentration of 100 ng mL⁻¹. This indicates that the electrochemical immunosensor have good specificity. In order to investigate the reproducibility of the proposed immunosensor, a series of five modified electrodes were prepared for the detection the same concentration of CEA (1 ng mL⁻¹) and the relative standard deviation (RSD) was 3.7%. To test the stability of the immunosensor, the modified electrode stored at 4 °C. After two weeks, the modified electrode still remains bioactive and the response signal at the electrode has no significant change (less than 4.8%). This indicates that the immunosensor have good stability.

Fig. 4 Specificity of the immunosensor to 1 ng mL⁻¹ CEA or 100 ng mL⁻¹ interferer and 1 ng mL⁻¹ CEA + 100 ng mL⁻¹ interferer.

3.6. Detection performance in serum samples

To further investigate the feasibility and potential practical application of the proposed electrochemical immunosensor, the detection performance of immunosensor were examined by standard addition methods. The assay results were given in Table S1.† The recoveries were from 98.1% to 103.5% and the RSD was from 2.24% to 4.08%. These results indicated that this proposed immunosensor was suitable for the analysis of real samples in clinical diagnosis.

4. Conclusions

In this work, the pCuOw were prepared by a simple decomposition of Cu(OH)₂ precursor and were composed of CuO nanoparticles (∼20 nm). A novel electrochemical immunosensor was developed for the detection of CEA by using pCuOw@Fc as signal label and CNTs–AuNPs as immunosensor platform. The electrochemical immunosensor exhibited excellent electrochemical properties, which were mostly ascribed to high specific surface area and excellent performance of pCuOw combined with conductivity and biocompatibility of CNTs–AuNPs. Therefore, the proposed method may present a promising strategy to construct high-performance electrochemical immunosensor for biomolecules analysis in the future.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (No. 21065009), the Key Project of Chinese Ministry of Education (No. 2010250) and Bingtuan Innovation Team in Key Areas (No. 2015BD003).

Notes and references

1. W. Lu, X. Cao, L. Tao, J. Ge, J. Dong and W. Qian, Biosens. Bioelectron., 2014, 57, 219–225.
2. T. Iwazawa, T. Kanoh, S. Matsui and T. Monden, Lung Cancer, 2000, 29, 254.
3. J. Miao, X. Wang, L. Lu, P. Zhu, C. Mao, H. Zhao, Y. Song and J. Shen, Biosens. Bioelectron., 2014, 58, 9–16.
4. N. Zamcheck and E. W. Martin, Cancer, 2006, 47, 1620–1627.
5. X. Y. Yang, Y. S. Guo, S. Bi and S. S. Zhang, Biosens. Bioelectron., 2009, 24, 2707–2711.
6. S. Bi, H. Zhou and S. Zhang, Biosens. Bioelectron., 2009, 24, 2961–2966.
7. R. Kurita, K. Arai, K. Nakamoto, D. Kato and O. Niwa, Anal. Chem., 2010, 82, 1692–1697.
8. O. Trenchevska, E. Kamcheva and D. Nedelkov, J. Proteome Res., 2010, 9, 5969–5973.
9. C. L. Hou and A. E. Herr, Anal. Chem., 2010, 82, 3343–3351.
10. D. Tang, R. Niessner and D. Knopp, Biosens. Bioelectron., 2009, 24, 2125–2130.
11. J. Gao, Z. Guo, F. Su, L. Gao, X. Pang, W. Cao, B. Du and Q. Wei, Biosens. Bioelectron., 2015, 63, 465–471.
12. S. Krishnan, V. Mani, D. Wasalathanthri, C. V. Kumar and J. F. Rusling, Angew. Chem., Int. Ed., 2011, 50, 1175–1178.
13. T. Li and M. Yang, Sens. Actuators, B, 2011, 158, 361–365.
14. J. Liu, S. Tian, L. Tiefenauer, P. E. Nielsen and W. Knoll, Anal. Chem., 2005, 77, 2756–2761.
15. G. Wang, X. Gang, X. Zhou, G. Zhang, H. Huang, X. Zhang and L. Wang, Talanta, 2013, 103, 75–80.
16. M. A. Sowole and H. B. Kraatz, Analyst, 2012, 137, 1120–1124.
17. S. Martić, S. Beheshti, M. K. Rains and H. B. Kraatz, Analyst, 2012, 137, 2042–2046.
18. K. Deng, J. Zhou and X. Li, Electrochim. Acta, 2013, 95, 18–23.
19. F. Qu, Y. Zhang, A. Rasooly and M. Yang, Anal. Chem., 2014, 86, 973–976.
20. X. Ren, T. Yan, Y. Zhang, D. Wu, H. Ma, H. Li, B. Du and Q. Wei, Biosens. Bioelectron., 2014, 58, 345–350.
21. S. Reddy, B. E. K. Swamy and H. Jayadevappa, Electrochim. Acta, 2012, 61, 78–86.
22. L. Zhang, H. Li, Y. Ni, J. Li, K. Liao and G. Zhao, Electrochem. Commun., 2009, 11, 812–815.
23. Y. Zhang, Y. Liu, L. Su, Z. Zhang, D. Huo, C. Hou and Y. Lei, Sens. Actuators, B, 2014, 191, 86–93.
24. J. Zhang, J. Ma, S. Zhang, W. Wang and Z. Chen, Sens. Actuators, B, 2015, 211, 385–391.
25. D. Volanti, D. Keyson, L. Cavalcante, A. Simoes, M. Joya, E. Longo, J. Varela, P. Pizani and A. Souza, J. Alloys Compd., 2008, 459, 537–542.
26. S. Zaman, M. H. Asif, A. Zainelabdin, G. Amin, O. Nur and M. Willander, J. Electroanal. Chem., 2011, 662, 421–425.
27. C. Yang, X. T. Su, F. Xiao, J. K. Jian and J. D. Wang, Sens. Actuators, B, 2011, 158, 299–303.
28. S. A. Khayyat, S. G. Ansari and A. Umar, J. Nanosci. Nanotechnol., 2014, 14, 3569–3574.
29. L. Tian and B. Liu, Appl. Surf. Sci., 2013, 283, 947–953.
30. Y. H. Choi, D. H. Kim, S. H. Hong and K. S. Hong, Sens. Actuators, B, 2013, 178, 395–403.
31. H. Zhang, Q. Zhu, Y. Zhang, Y. Wang, L. Zhao and B. Yu, Adv. Funct. Mater., 2007, 17, 2766–2771.
32. S. Z. Deng, V. Tjoa, H. M. Fan, H. R. Tan, D. C. Sayle, M. Olivo, S. Mhaisalkar, J. Wei and C. H. Sow, J. Am. Chem. Soc., 2012, 134, 4905–4917.
33. P. Zhang, L. Zhang, G. Zhao and F. Feng, Microchim. Acta, 2012, 176, 411–417.
34. T. G. S. Babu and T. Ramachandran, Electrochim. Acta, 2010, 55, 1612–1618.
35. S. Cherevko and C. H. Chung, Talanta, 2010, 80, 1371–1377.
36. W. Wang, L. Zhang, S. Tong, X. Li and W. Song, Biosens. Bioelectron., 2009, 25, 708–714.
37. L. Wang, K. Zhang, Z. Hu, W. Duan, F. Cheng and J. Chen, Nano Res., 2014, 7, 199–208.
38. T. Belin and F. Epron, Mat. Sci. Eng. B, 2005, 119, 105–118.
39. J. Huang, X. Xing, X. Zhang, X. He, Q. Lin, W. Lian and H. Zhu, Food Res. Int., 2011, 44, 276–281.
40. X. Feng, R. Li, Y. Ma, Q. Fan and W. Huang, Synth. Met., 2011, 161, 1940–1945.
41. M. J. Song, S. W. Hwang and D. Whang, Talanta, 2010, 80, 1648–1652.
42. K. Shang, X. Wang, B. Sun, Z. Cheng and S. Ai, Biosens. Bioelectron., 2013, 45, 40–45.
43. D. Feng, X. Lu, X. Dong, Y. Ling and Y. Zhang, Microchim. Acta, 2013, 180, 767–774.
44. X. Sun and Z. Ma, Anal. Chim. Acta, 2013, 780, 95–100.
45. L. Zhu, L. Xu, N. Jia, B. Huang, L. Tan, S. Yang and S. Yao, Talanta, 2013, 116, 809–815.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra26828a

---