A simple lateral flow biosensor for the rapid detection of copper(II) ions based on click chemistry

Abstract

Copper(II) ions (Cu²⁺) at a high concentration are harmful to human health. Herein a simple and enzyme-free lateral flow biosensor for the rapid detection of Cu²⁺ based on copper(I) ion (Cu⁺)-catalyzed click chemistry has been constructed for the first time. In the presence of sodium ascorbate, Cu²⁺ was reduced to Cu⁺, which could catalyze the cycloaddition between azide-DNA and alkyne/biotin-DNA in aqueous solution. The ligated DNA product could then be immobilized onto the test zone of the lateral flow biosensor to form a red band which could be unambiguously read by the naked eye. Taking advantage of the optical properties of gold nanoparticles (AuNPs) and high efficiency and selectivity of Cu⁺-catalyzed click chemistry, this assay enabled the visual detection of Cu²⁺ as low as 100 nM with excellent specificity. In comparison with conventional methods, this biosensor is more simple to operate and more cost-effective to use, and therefore has great potential in point-of-care diagnosis and environmental monitoring.

---

Introduction

As an essential micronutrient element for human life, Cu²⁺ has been recognized as a necessary cofactor or structural component of numerous enzymes needed in metabolic processes. However, Cu²⁺ at a high intracellular concentration can cause adverse health effects, such as liver damage, gastrointestinal disturbance, and neurodegenerative diseases including Alzheimer, Menke and Wilson diseases.^1,2 The U.S. Environmental Protection Agency (EPA) set 20 μM as the maximum contamination concentration for Cu²⁺ in drinking water.³ To monitor the Cu²⁺ pollution in the environment, sensitive and specific analytical methods for Cu²⁺ are imperative.

Current standard methods for Cu²⁺ determination include graphite furnace atomic absorption spectrometry (AAS)^4,5 and inductive coupled plasma atomic emission spectroscopy (ICP-AES).⁶ Although these methods are commonly used to determine the trace metals, they require expensive instruments and experienced operators, which limits their wider application.

The advent of new methods and technologies, such as colorimetric or fluorescence assays based on Cu²⁺-dependent DNA-cleaving DNAzyme (Cu²⁺-specific DNAzyme),^7,8 Cu²⁺-dependent DNA ligation enzyme,^9,10 quantum dots,^11–13 or gold nanoclusters,^14–17 provide effective tools for Cu²⁺ detection. Peng and co-workers developed a fluorescent microarray using Cu²⁺-specific DNAzyme, resulting a high sensitivity of 10 nM for Cu²⁺ detection.¹⁸ Li and co-workers reconstructed Cu²⁺-specific DNA-cleaving DNAzyme with an intramolecular stem-loop structure to achieve a label-free and visual detection method for aqueous Cu²⁺ with a detection limit of 3.9 nM.¹⁹ Zhang and co-workers prepared novel luminescent gold nanoclusters for the detection of Cu²⁺ with a sensitivity of 0.9 μM within 10 minutes.²⁰

Although these methods are effective, the formation of stable triplex-stranded DNA structure of Cu²⁺-dependent DNA-cleaving DNAzyme needs procedures of denaturation and renaturation, and the use of fluorophore labeled oligonucleotides is not only expensive but also increases the complexity of the operation.

Considerable efforts have been devoted to the development of easy-to-operate and low-cost technologies for Cu²⁺ detection. In recent years, researchers have shown great interests in Cu⁺-catalyzed click chemistry due to its high efficiency and selectivity.^21–23 It refers to the [3 + 2] cycloaddition reaction between an azide group and an alkyne group at room temperature in aqueous solution, which finally result in the formation of a 5-membered triazole ring. The source of Cu⁺ in the click chemistry can be generated from the reduction of Cu²⁺ in the presence of sodium ascorbate.²⁴

As a gold-nanoparticle based assay, lateral flow biosensor has a user friendly format which minimizes the requirement for bulky and costly apparatus and eliminates complex analysis procedures. The detection result can be obtained visually by observing the color intensity of the red band on the test zone. Our group and others have successfully developed strip biosensors for the detection of proteins, nucleic acids, cells, and small molecules.^25–28

In this study, we presented a lateral flow biosensor for Cu²⁺ detection based on Cu⁺-catalyzed click chemistry for the first time (Scheme 1). This biosensor has the following features: (1) easy to use and operate, (2) no requirement for bulky and costly apparatus, (3) high sensitivity to detect a minimum of 100 nM of Cu²⁺, (4) excellent selectivity to other metal ions.

Scheme 1 Principle of the strip biosensor for Cu²⁺ detection. (A) The formation of the ligation product of azide-DNA and alkyne/biotin-DNA based on Cu⁺-promoted click chemistry. (B) Schematic illustration of the structure of lateral flow biosensor.

Experimental

Reagents and materials

Sodium ascorbate was purchased from Shanghai Aladdin Reagent Co., Ltd (Shanghai, China). Ammonium peroxodisulfate (AP), urea, copper(II) chloride dihydrate, potassium chloride, and other metal ions were purchased from Guangzhou Chemical Reagent Factory (Guangzhou, China). Streptavidin (SA), SYBR Green II, bovine serum albumin (BSA), and human serum were purchased from Sigma-Aldrich (St. Louis, USA). Phosphate Buffer Solution (PBS) was purchased from Genetimes Technology, Inc. (Shanghai, China). Tris-base, boric acid, ethylene diamine tetracetic acid (EDTA), and Triton X-100 were purchased from Guangzhou WhiGa Technology Ltd (Guangzhou, China). Nitrocellulose membrane was purchased from Sartorius (Goettingen, Germany). Fiberglass and absorbent paper were purchased from Shanghai Kinbio (Shanghai, China). Oligonucleotides purified by HPLC were synthesized by Shanghai Sangon Biotechnology Co., Ltd (Shanghai, China), and the sequences of the oligonucleotides were listed in Table S1.† All buffer solutions used in this study were prepared in our lab. Other chemicals were purchased from standard commercial sources and were of analytical grade purity.

PAGE gels were visualized via fluorescence detection using a Tanon 1600 automatic digital gel image analysis system (Shanghai, China). Biojet HM 3030 dispenser, the Guillotine cutting module ZQ 4200, and portable strip reader DT1030 were all purchased from Shanghai Kinbio Tech. Co., Ltd (Shanghai, China).

Polyacrylamide gel electrophoresis (PAGE) analysis of the ligation product of click chemistry

Cu²⁺ with various concentrations were added into the mixture containing azide-DNA and alkyne/biotin-DNA (each 10 μM) and ascorbate (600 μM), with the final Cu²⁺ concentration of 0, 20 μM, 200 μM, 2 mM, respectively. After incubation for 4 h at room temperature (RT), the reaction mixture was separated by a 15% PAGE. The electrophoresis was carried out in 0.5 Tris–borate–EDTA (TBE) buffer (90 mM Tris, 90 mM boric acid, and 10 mM EDTA, pH 8.0) at 80 V for 2 h. After electrophoresis, the gel was stained by 10× SYBR Green II for 20 min in 1× TBE.

Preparation of AuNPs and streptavidin modified gold nanoparticle (AuNP–SA) conjugates

AuNPs with an average diameter of 15 nm were prepared using the citrate reduction method. Briefly, 4 mL of 1% trisodium citrate was added to 100 mL of a rapidly stirred and boiling HAuCl₄ solution (0.01%) in a 500 mL round bottom flask. After turning red, the solution was boiled for 10 additional minutes then cooled to room temperature with gentle stirring. The resulting AuNP solution was stored at 4 °C and used for preparation of AuNP–SA conjugates. To prepare AuNP–SA conjugates, 10 μg (46 μL) SA and 4 μL 0.1 M K₂CO₃ were added to 1 mL AuNP solution and the mixture was shaken gently at room temperature for 1 h. BSA was added into the SA-coated AuNPs (1050 μL) to 1%. The solution was shaken at room temperature for 1 h. Particles were centrifuged (12 × 10³ rpm, 20 min) and rinsed three times with rinsing buffer (20 mM Na₃PO₄, 5% BSA, 0.25% Tween-20, 10% sucrose, and 0.1% NaN₃) to remove any unbound SA. The red pellet was re-suspended in 50 μL of rinsing buffer and then stored at 4 °C until use. The UV absorption spectrum and the TEM photographs of AuNPs are shown in Fig. S1.†

Construction of lateral flow biosensor

The LFB was assembled as the schematic diagram shown in Scheme 1B. It consists of a sample pad, a conjugate pad, a nitrocellulose membrane, and an absorbent pad. The sample pad (17 mm × 30 cm) was prepared by soaking a glass fiber pad in sample pad buffer (pH 8.0) (1% Triton, 1% BSA, 2% glucose, and 50 mmol L⁻¹ boric acid). The pad was then dried and stored in a low-humidity chamber at room temperature. The conjugate pad (8 mm × 30 cm) was prepared by dispensing AuNP–SA solution (10 μL cm⁻¹) onto the fiberglass using a dispenser (Shanghai Kinbio, Shanghai, China). The pad was dried at room temperature for 12 hours and stored at room temperature in low-humidity until use. 30 μL of 100 μM control zone DNA and 30 μL of 100 μM test zone DNA were dispensed onto the nitrocellulose (NC) membrane (25 mm × 30 cm, capillary rate: 140 ± 40 s, thickness: 145 ± 20 μm) simultaneously to form a control zone and a test zone with a lateral flow dispenser, respectively. The distance between the test zone and the control zone was approximately 5 mm. The dispensed DNA was crosslinked to the NC membrane by exposing to ultraviolet light for 15 minutes. The membrane was then dried and stored at room temperature in low-humidity until use. Finally, the four components of the lateral flow biosensor were assembled on plastic adhesive backing (60 mm × 30 cm). Each part overlapped by 2 mm to ensure solution migration through the strip during the assay. The plate was finally cut into 0.4 cm wide strips using a paper cutter (Programmed high speed cutter, Shanghai Kinbio, Shanghai, China).

Detection of Cu²⁺ using the lateral flow biosensor

Azide-DNA (10 μM, 2 μL) and alkyne/biotin-DNA (10 μM, 2 μL) were mixed at a mole ratio of 1 : 1 in PBS buffer (14 μL) at a final concentration of 1 μM, then Cu²⁺ with various concentrations (2 μL) were added. Control experiment (in the absence of Cu²⁺) was conducted at the same condition. For the promotion on the cycloaddition reaction between azide and alkyne group, sodium ascorbate was added to each reaction at a final optimal concentration of 600 μM (Fig. S2†). After shaken gently at RT for 45 min, 20 μL of the mixture and 40 μL of loading buffer (4× SSC containing 0.1% Triton X-100) were loaded onto the sample pad successively. The red bands on the test zone and control zone was observed by the naked eye in 15 minutes, and their optical intensities of the red bands on the test zone were recorded using the strip reader.

Results and discussion

The principle of this strip biosensor for Cu²⁺ detection

The structure of the strip biosensor was shown in Scheme 1B, it consists of four parts: a sample pad, a conjugated pad, a nitrocellulose (NC) membrane, and an absorbent pad. Two oligonucleotides (3 and 4), which are complementary to azide-DNA and alkyne/biotin-DNA (1 and 2), were immobilized onto the NC membrane simultaneously to form the test zone and control zone, respectively. In the presence of Cu²⁺ and sodium ascorbate, azide-DNA and alkyne/biotin-DNA were ligated together in aqueous solution through the formation of a triazole ring between the two reacting groups (Scheme 1A). After Cu⁺-promoted click chemistry, the solution containing ligation product was loaded onto the sample pad, and it began to flow along the long axis of the strip due to the capillary effect. Lateral flow system possesses a fundamental property in utilizing capillary flow to transfer and separate molecules in the sample liquid through the porous network without any external equipment. AuNP–SA on the conjugate pad could react with the biotin-ligation product to form the AuNP-ligation product complexes. The complexes continued to migrate to the test zone and were immobilized through the formation of the double stranded DNA (dsDNA) between the azide-DNA and its complementary sequence. The excess complexes migrated to the control zone and were captured by the DNA which is complementary to the alkyne/biotin-DNA. As a result, two red bands can be observed by the naked eye, and the red color intensity on the test zone was proportional to the Cu²⁺ concentration. In the absence of Cu²⁺, there was only one red band on the control zone, which indicated that the strip biosensor functioned correctly. Qualitative analysis was performed by observing the color change on the test zone, and semi-quantitative analysis was performed by recording the optical intensity of the red band on the test line (peak areas) using a portable strip reader.

PAGE analysis of the ligation product of click chemistry

The ligation product of azide-DNA and alkyne/biotin-DNA was initially confirmed by 15% PAGE. As shown in Fig. 1, only one band containing azide-DNA (18 bases) and alkyne/biotin-DNA (17 bases) could be observed in lane 4, indicating no ligation reaction was occurred without Cu²⁺. However, a new product band appeared after addition of Cu²⁺, and the brightness of the band reduced with decreasing Cu²⁺ concentration (lane 1: 2 mM; lane 2: 200 μM; lane 3: 20 μM; lane 4: 0 μM). These results indicated that azide-DNA and alkyne/biotin-DNA were ligated together with the aid of Cu⁺-promoted azide–alkyne cycloaddition.

Fig. 1 PAGE analysis of the ligation product in the presence of different concentrations of Cu²⁺ (lane 1: 2 mM; lane 2: 200 μM; lane 3: 20 μM; lane 4: 0 μM).

Optimization of the experimental parameters

The experimental parameters (composition of loading buffer, the volume of concentrated AuNP–SA solution, the concentration of azide-DNA and alkyne/biotin-DNA, reaction time of click chemistry, and the concentrations of sodium ascorbate) that affect the sensitivity and reproducibility of the biosensor were optimized. The optimizations were performed by varying one experimental condition and keeping other parameters constant.

In order to investigate the effect of the composition of loading buffer on the optical intensity of the red band on the test zone, five types of loading buffer (PBS, 4× SSC, 4× SSC containing 0.05%, 0.1% and 0.2% Triton X-100, respectively) were used in the presence of 20 μM Cu²⁺. As shown in Fig. 2A, the optical intensity of the red band was the highest when using 4× SSC containing 0.1% Triton X-100 as the loading buffer. This buffer was then chosen as the optimal buffer and used in the following experiments.

Fig. 2 The optical intensities of the red bands on the test zone corresponding to (A) five types of loading buffer (PBS, 4× SSC, 4× SSC containing 0.05%, 0.1% and 0.2% Triton X-100), (B) different volumes (1, 2, 3, 4, and 5 μL) of AuNP–SA solution, (C) different concentrations of azide-DNA and alkyne/biotin-DNA (0.1, 0.2, 0.6, 1, 1.5, and 2 μM), and (D) different click time (15 min, 30 min, 45 min, 60 min, 90 min, and 120 min). The error bars represent the standard deviation of three independent measurements.

The volume of concentrated AuNP–SA solution on the conjugated pad also has an effect on the optical intensity of the red band. From Fig. 2B, we can see that in the presence of 20 μM Cu²⁺, the intensity of the red bands increases gradually as the volume of AuNP–SA solution increases from 1 to 5 μL. However, the intensity of red band does not increase significantly when the volume of AuNP–SA solution increases up to 4 μL or higher. Their corresponding optical intensities were recorded using the portable strip reader. Therefore, 4 μL was chosen as the optimal volume of AuNP–SA solution.

The unreacted azide-DNA (DNA 1) and alkyne/biotin-DNA (DNA 2) in the solution can hybridize with DNA 3 and DNA 4, which in turn reduce the amount of AuNP-ligation product on the test zone and control zone, respectively. Hence, the concentration of azide-DNA and alkyne/biotin-DNA used in the click chemistry should be taken into account for the parameter optimization. As illustrated in Fig. 2C, it was found that the intensity increased with the increase of DNA concentrations from 0.1 μM to 0.6 μM, further increasing the DNA concentration caused a significant decrease of optical intensity in the presence of 20 μM Cu²⁺. Thus, 0.6 μM was the optimal concentration of azide-DNA and alkyne/biotin-DNA.

The optical intensity of the red band on the test zone directly depended on the yield of AuNP-ligation product, in turn, corresponded to the click time of Cu⁺-promoted cycloaddition. Fig. 2D shows the effect of click time on the optical intensity on the test zone. Although the intensity increased with increasing the click time from 15 min to 2 h in the presence of 20 μM Cu²⁺, a red band can be observed obviously on the test zone when the click time is 45 min, which can reach the EPA requirement for Cu²⁺ detection in drinking water (20 μM). As a result, 45 min was used as the click time to reduce the whole experimental time.

In the present study, the ligation efficiency depends on the catalytic efficiency of Cu⁺-promoted click chemistry, which ultimately depends on the reduction rate of Cu²⁺ into Cu⁺. The reduction process is complex and listed as follows:

Ascorbate acid + 2Cu²⁺ → dehydroascorbic acid + 2Cu⁺ + 2H⁺ (1)

2Cu⁺ + 2O₂ → 2Cu²⁺ + 2O₂⁻ (2)

2O₂⁻ + 2H⁺ → H₂O₂ + O₂ (3)

Cu²⁺ + H₂O₂ → ˙OH (hydroxyl radicals) + OH⁻ + Cu⁺ (4)

It is apparent that the sodium ascorbate plays an important role as reducing reagent in accelerating the reaction rate. Fig. 3 shows the effect of different concentrations of sodium ascorbate on the optical intensity on the test zone. The intensity improved with increasing the sodium ascorbate concentrations from 200 μM to 600 μM, further increasing the sodium ascorbate concentration caused little increase of optical intensity in the presence of 20 μM Cu²⁺. Thus, 600 μM was the optimal concentration of sodium ascorbate.

Fig. 3 The optical intensities of the red bands on the test zone corresponding to different concentrations of sodium ascorbate (200, 400, 600, 800, and 1000 μM). The error bars represent the standard deviation of three independent measurements.

Detection of Cu²⁺ under optimal experimental conditions

To evaluate the sensitivity and dynamic range of the strip biosensor for Cu²⁺ detection, Cu²⁺ with various concentrations (1 mM, 200 μM, 20 μM, 1 μM, 100 nM, 50 nM, 0) was used under optimal experimental conditions. The typical images and optical response of red band on the test zone in the presence of different concentrations of Cu²⁺ were presented in Fig. 4. As shown in Fig. 4A, the optical intensities of the red bands on the test zones increased with the increment of Cu²⁺ concentration, and no red band was observed on the test zone in the absence of Cu²⁺ (negative control, bottom). Meanwhile, it is easy to distinguish the presence of red band using 100 nM Cu²⁺ from the absence of red band using 0 nM Cu²⁺ on the test zone by the naked eye. Therefore, the limit of detection (LOD) for visual detection was set to 100 nM. Calibration curve in Fig. 4C (inset) shows that the optical intensities of the red bands are proportional to the logarithm of Cu²⁺ concentration in the range of 50 nM to 200 μM with a linear equation of optical intensity = 265.82x − 177.3 (R² = 0.9735).

Fig. 4 Sensitivity of the biosensor for Cu²⁺ detection. (A) Typical photo images and (B) the corresponding optical intensities of the strip biosensor in the presence of different concentrations of Cu²⁺. (C) Plots of the optical intensities of red bands on the test zone vs. different concentrations of Cu²⁺. Inset: calibration curve of the optical intensities of red bands on the test zone vs. the log value of Cu²⁺ concentration. The error bars represent the standard deviation of three independent measurements.

Although the detection limit of this method is higher than some of the fluorescent-based assays,²³ it is comparable with or lower than previously reported AuNP-based assays for Cu²⁺ detection,²² and can meet the EPA requirement for Cu²⁺ detection in drinking water (20 μM) with the advantages of being simple, rapid and cost-effective.

Specificity of the lateral flow biosensor for Cu²⁺ detection

To test the specificity of this assay for Cu²⁺ detection, the reaction solutions containing different metal ions, such as Hg²⁺, Pb²⁺, Co²⁺, Mg²⁺, Fe²⁺, Fe³⁺, Zn²⁺, Ba²⁺, and Mn²⁺ were applied to lateral flow biosensor under the same experimental condition, respectively. The images of lateral flow biosensor recorded by the strip reader are shown in Fig. 5. The red band on the test zone can be observed obviously when using mixtures of 20 μM Cu²⁺ and other ions (Hg²⁺, Pb²⁺, Co²⁺, Mg²⁺, Fe²⁺, Fe³⁺, Zn²⁺, Ba²⁺, and Mn²⁺), whereas no red band can be observed on the test zone of the lateral flow biosensor in the presence of 200 μM other metal ions (Hg²⁺, Pb²⁺, Co²⁺, Mg²⁺, Fe²⁺, Fe³⁺, Zn²⁺, Ba²⁺, and Mn²⁺) separately, indicating the high specificity of this biosensor for Cu²⁺ detection.

Fig. 5 Specificity of the biosensor for Cu²⁺ detection. (A) Photo images and optical intensities of the red band on the test zone in the presence of 20 μM Cu²⁺ mix (Cu²⁺, Hg²⁺, Pb²⁺, Co²⁺, Mg²⁺, Fe²⁺, Fe³⁺, Zn²⁺, Ba²⁺, and Mn²⁺) and other metal ions (Hg²⁺, Pb²⁺, Co²⁺, Mg²⁺, Fe²⁺, Fe³⁺, Zn²⁺, Ba²⁺, and Mn²⁺) separately. The photo was taken within 20 minutes after sample loading.

The recovery rates, repeatability and reproducibility of the strip biosensor for Cu²⁺ detection

The practical application of the strip biosensor was demonstrated by applying it to detect Cu²⁺ in tap water and human serum samples. Recovery experiments were performed by spiking different amounts of Cu²⁺ into tap water and 50% human serum (Table S2, ESI†). Acceptable recovery rates in tap water (between 91.3% and 109.3%) and human serum (between 95.5% and 104.2%) for Cu²⁺ detection were obtained, which confirmed that the proposed biosensor was able to detect Cu²⁺ in water and human serum samples with little interference.

The assessment of repeatability and reproducibility was done by calculating the coefficient of variations for 15 repeated measurements of 20 μM Cu²⁺ done by two researchers at one week interval. The coefficient of variations (CV) are lower than 5% which show good repeatability and reproducibility (Table S3, ESI†).

Conclusions

In conclusion, we have successfully constructed a lateral flow biosensor for the rapid detection of Cu²⁺ based on Cu⁺-promoted click chemistry with high sensitivity and selectivity. This method does not need bulky and costly apparatus, at the same time, the LOD for visual Cu²⁺ detection is 100 nM, which is lower than the EPA limit of Cu²⁺ in drinking water (20 μM). In comparison with previously reported Cu⁺-catalyzed click chemistry based methods for Cu²⁺ detection, this biosensor shows the advantages of being simple, rapid and cost-effective, and therefore has great potential in point-of-care diagnosis and environmental monitoring.

Acknowledgements

Financial support was provided by Key Deployment Project of the Chinese Academy of Sciences (KSZD-EW-Z-021-1-4), Strategic Cooperation Project of the Chinese Academy of Sciences and Guangdong Province (2012B091100267) and (2014B020212011).

Notes and references

1. W. Si, J. Liu, L. Cai, H. Jiang, C. Zheng, X. He, J. Wang and X. Zhang, Bull. Environ. Contam. Toxicol., 2015, 94, 214–219.
2. A. Saffari, N. Daher, M. M. Shafer, J. J. Schauer and C. Sioutas, Environ. Sci. Technol., 2014, 48, 7576–7583.
3. B. Nath, P. Chaudhuri and G. Birch, Ecotoxicol. Environ. Saf., 2014, 107, 284–290.
4. M. S. Luz, A. N. Nascimento and P. V. Oliveira, Talanta, 2013, 115, 409–413.
5. T. Ren, L. J. Zhao, B. S. Sun and R. G. Zhong, J. Environ. Qual., 2013, 42, 1752–1762.
6. D. J. Zhang, L. N. Jiang, Z. J. Zhang, Y. Shao and C. X. Li, Guang pu xue yu guang pu fen xi = Guang pu, 2011, 31, 1935–1938.
7. B. C. Yin, B. C. Ye, W. Tan, H. Wang and C. C. Xie, J. Am. Chem. Soc., 2009, 131, 14624–14625.
8. J. L. He, S. L. Zhu, P. Wu, P. P. Li, T. Li and Z. Cao, Biosens. Bioelectron., 2014, 60, 112–117.
9. J. Liu and Y. Lu, Chem. Commun., 2007, 4872–4874.
10. F. Wang, R. Orbach and I. Willner, Chemistry, 2012, 18, 16030–16036.
11. C. Guo, J. Wang, J. Cheng and Z. Dai, Biosens. Bioelectron., 2012, 36, 69–74.
12. P. Wang, X. Ma, M. Su, Q. Hao, J. Lei and H. Ju, Chem. Commun., 2012, 48, 10216–10218.
13. L. H. Jin and C. S. Han, Anal. Chem., 2014, 86, 7209–7213.
14. H. H. Deng, L. N. Zhang, S. B. He, A. L. Liu, G. W. Li, X. H. Lin, X. H. Xia and W. Chen, Biosens. Bioelectron., 2014, 65, 397–403.
15. H. Ding, C. Liang, K. Sun, H. Wang, J. K. Hiltunen, Z. Chen and J. Shen, Biosens. Bioelectron., 2014, 59, 216–220.
16. N. Zhang, Y. Si, Z. Sun, L. Chen, R. Li, Y. Qiao and H. Wang, Anal. Chem., 2014, 86, 11714–11721.
17. Y. W. Lin, C. C. Huang and H. T. Chang, Analyst, 2011, 136, 863–871.
18. P. Zuo, B. C. Yin and B. C. Ye, Biosens. Bioelectron., 2009, 25, 935–939.
19. H. Li, X. X. Huang, Y. Cai, H. J. Xiao, Q. F. Zhang and D. M. Kong, PLoS One, 2013, 8, e73012.
20. M. Zhang, H. N. Le, X. Q. Jiang, S. M. Guo, H. J. Yu and B. C. Ye, Talanta, 2013, 117, 399–404.
21. Y. Zhou, S. Wang, K. Zhang and X. Jiang, Angew. Chem., Int. Ed. Engl., 2008, 47, 7454–7456.
22. Q. Shen, W. Li, S. Tang, Y. Hu, Z. Nie, Y. Huang and S. Yao, Biosens. Bioelectron., 2013, 41, 663–668.
23. C. Ge, Q. Luo, D. Wang, S. Zhao, X. Liang, L. Yu, X. Xing and L. Zeng, Anal. Chem., 2014, 86, 6387–6392.
24. K. Zhu, Y. Zhang, S. He, W. Chen, J. Shen, Z. Wang and X. Jiang, Anal. Chem., 2012, 84, 4267–4270.
25. H. Xu, X. Mao, Q. Zeng, S. Wang, A. N. Kawde and G. Liu, Anal. Chem., 2009, 81, 669–675.
26. Z. Fang, J. Huang, P. Lie, Z. Xiao, C. Ouyang, Q. Wu, Y. Wu, G. Liu and L. Zeng, Chem. Commun., 2010, 46, 9043–9045.
27. J. Chen, Z. Fang, P. Lie and L. Zeng, Anal. Chem., 2012, 84, 6321–6325.
28. Z. Xiao, P. Lie, Z. Fang, L. Yu, J. Chen, J. Liu, C. Ge, X. Zhou and L. Zeng, Chem. Commun., 2012, 48, 8547–8549.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra11752f

---