A label-free amperometric immunosensor for the detection of carcinoembryonic antigen based on novel magnetic carbon and gold nanocomposites

Abstract

In this work, a novel label-free electrochemical immunosensor was developed for the quantitative detection of carcinoembryonic antigen (CEA). To this end, gold nanoparticle (Au NP) functionalized magnetic multi-walled carbon nanotubes (MWCNTs–Fe₃O₄) were prepared and applied to lead ion (Pb²⁺) adsorption. Because of the synergetic effect present in Pb²⁺@Au@MWCNTs–Fe₃O₄, they show better electrocatalytic activity towards the reduction of hydrogen peroxide (H₂O₂) than MWCNTs, MWCNTs–Fe₃O₄ or Au@MWCNTs–Fe₃O₄. Featuring a large specific surface area, good biocompatibility and excellent electrical conductivity, Pb²⁺@Au@MWCNTs–Fe₃O₄ were used as transducing materials to achieve efficient conjugation of capture antibodies and signal amplification of the proposed immunosensor. Cyclic voltammetry and the amperometric i–t technique were used to record the change of electrochemical signal when the electrodes were modified with different concentrations of CEA. Under optimal conditions, the label-free immunosensor exhibited a wide linear range from 5 fg mL⁻¹ to 50 ng mL⁻¹ with a low detection limit of 1.7 fg mL⁻¹ for CEA. The proposed immunosensor displays high electrochemical performance with good reproducibility, selectivity and stability, and has great potential in clinical and diagnostic applications.

---

1. Introduction

Cancer has been regarded as one of the major causes of mortality worldwide and its effective treatment still has not been reported.¹ In clinical analysis, cancer patients can be detected when the concentration of tumor markers is enhanced to a certain level in serum.² Hence, it is necessary to achieve sensitive, fast, and accurate assays for the monitoring of tumor markers for effective early diagnosis and therapy of cancer.^1,3,4 Carcinoembryonic antigen (CEA) is a member of a family of cell surface glycoproteins that are produced in excess in essentially all human colon carcinomas and in a high proportion of carcinomas at many other sites.^5,6 As a widely used tumor maker of human colon carcinomas, the quantitative detection of CEA in serum is valuable in diagnosis or clinical management.⁶

In the past few years, a variety of methods have been applied to the adetection of tumor markers, such as electrochemical immunosensors,^7,8 enzyme-linked immunosorbent assay,^9,10 ECL immunosensor,¹¹ and electrochemiluminescence immunoassay.^11–14 Comparatively, electrochemical immunosensors have attracted widespread attention due to their high sensitivity, and rapid, highly selective and simple operation.^15,16 Among all of the electrochemical immunosensors, label-free electrochemical immunosensors have recently emerged as a novel assay to detect proteins due to their simple preparation, greater cost effectiveness and good activity conservation of antibodies or antigens.^15,17

In order to improve the sensitivity of electrochemical analysis, a variety of nanomaterials have been used to fabricate immunosensors to achieve signal amplification, such as carbon nanotubes,^18,19 metal nanoparticles,^20–22 and metal oxides.²³ Among these tested materials, multi-walled carbon nanotubes (MWCNTs) have gained the most attention due to their large specific surface area, excellent conductivity^24,25 and good biocompatibility.^26,27 Fe₃O₄ has a great auxiliary catalytic activity towards the reduction of hydrogen peroxide (H₂O₂), according to previous reports.²⁸ Simultaneously, Fe₃O₄ nanoparticles can promote electron transfer, thus providing a better choice for the fabrication of electrochemical immunosensors.¹⁵ Metal NPs dispersed on an oxide support often display higher catalytic activity than such NPs as single components, which is due to the synergetic effect occurring at the interface between the metal and oxide support.^28–30 Recent studies declared that Au NPs deposited on a metal-oxide support showed high catalytic activity for CO oxidation, even though Au NPs are chemically inert.²⁸ The surface functionalization of MWCNTs would provide an avenue for further chemical modification, such as ion adsorption.^31,32 As a type of protein containing amino groups (–NH₂), primary antibodies (Ab₁) can be effectively conjugated onto the Au@MWCNTs–Fe₃O₄ by the interaction between the Au NPs and –NH₂ groups on the antibodies to construct Au–N.^6,33,34 Au@MWCNTs–Fe₃O₄ have a large surface area, high conductivity and exceptional adsorption capability for lead ions (Pb²⁺) due to the synergetic effect, which is applied to signal amplification. After adsorbing Pb²⁺, the redox process of Pb²⁺ could further promote the redox process of H₂O₂, which is applied to signal amplification. The signal amplification strategy, using the synergetic effect present in gold nanoparticle-functionalized magnetic multi-walled carbon nanotubes loaded with lead ions (Pb²⁺@Au@MWCNTs–Fe₃O₄), can further increase the electron transfer efficiency on an electrode surface and the reaction efficiency of the nanocomposite toward H₂O₂ reduction to improve the detection sensitivity of the immunosensor.^15,28

In this research, an innovative label-free electrochemical immunosensor was designed to achieve quantitative detection of CEA. The synergetic effect existing in the Pb²⁺@Au@MWCNT–Fe₃O₄ system, could not only improve the electron transfer efficiency but also enhance the effective immobilization of Ab₁. The proposed immunosensor provides a useful technology for the quantitative detection of CEA in human serum, and shows the advantages of wide linear range, low detection limit, good reproducibility and selectivity, as well as acceptable stability. The results of electrochemical studies suggested that the proposed immunosensor possessed great performance for CEA determination and provided great potential for application to the analysis of other low-abundance proteins.

2. Material and methods

2.1. Apparatus and reagents

A CHI760D electrochemical workstation was used in the entire process of electrochemical measurement (Shanghai CH Instruments Co, China). Scanning electron microscopy (SEM) images were obtained using a Quanta FEG250 field emission environmental SEM (FEI, United States). Energy Dispersive X-ray (EDX) spectra were recorded using a JEOL JSM-6700F microscope (Japan). Fourier transform infrared spectroscopy (FTIR) spectrum was obtained from VERTEX 70 spectrometer (Bruker, Germany). For AC impedance measurements, a frequency range of 0.1 kHz to 100 Hz and an AC voltage amplitude of 5 mV were used.

CEA antibody and antigen were purchased from Beijing Dingguochangsheng Biotechnology Co. Ltd China. Bovine serum albumin (BSA, 96–99%) was purchased from Sigma reagent Co., Ltd (St. Louis, MO, USA). Multi-walled carbon nanotubes (MWCNTs) were purchased from Alfa Aesar Co., Ltd (Shanghai, China). HAuCl₄·4H₂O was obtained from Sinopharm Chemical Reagent Shanghai Co., Ltd, China. FeCl₃·6H₂O was purchased from Damao Chemical Reagent Co., Ltd, Tianjin, China. K₃Fe(CN)₆ was purchased from Sinopharm Chemical Reagent Co., Ltd (Beijing, China). Phosphate buffered saline (PBS, pH = 7.4) was prepared as an electrolyte in the electrochemical measurements. The Ab₁ solution was prepared as follows: 1 mg of CEA antibody was dissolved in PBS (pH = 7.4, 10 mL) to obtain Ab₁ stock solutions. Subsequently, it was diluted to the required concentration (10 μg mL⁻¹). Ultrapure water (18.25 MΩ cm, 24 °C) was used throughout the study. All other reagents are at analytical grade. All solutions were stored at 4 °C for frequent usage.

2.2. Preparation of the amino-functionalized MWCNTs–Fe₃O₄

To remove all the metal oxide, MWCNTs (0.5 g) were added into a mixture of 3 M HNO₃ and 2 M H₂SO₄ (3 : 1, v/v), which was kept under ultrasonic conditions at 40 °C for 3 h before cooling down to room temperature. Then it was washed to neutrality, and dried at room temperature.

MWCNTs–Fe₃O₄ were synthesized according to an established protocol.³⁵ In brief, FeCl₃·6H₂O (0.7 g) and MWCNTs (0.2 g) were dissolved in ethylene glycol (37.5 mL) to form a clear solution. Subsequently, sodium acetate (NaAc, 1.8 g) was added and dissolved under ultrasonic conditions. The mixture was stirred vigorously for 30 min and then transferred to a Teflon-lined stainless steel autoclave. The autoclave was heated to 200 °C and maintained at this temperature for 16 h, and finally cooled down to room temperature. Ultra-pure water (50 mL) was used to clean the composite and the resultant solid was separated from the liquid mixture using a strong magnet.

MWCNTs–Fe₃O₄ (0.1 g) and 3-aminopropyl triethoxysilane (0.1 mL) were dissolved in anhydrous ethanol (10 mL) and heated to 70 °C for 1.5 h. Subsequently, the solid was separated by magnetic separation and washed with anhydrous ethanol. Then, the desired amino-functionalized MWCNTs–Fe₃O₄ were stored at 50 °C.

2.3. Preparation of Au@MWCNTs–Fe₃O₄

The preparation of the Au NPs was carried out through the reduction of AuCl₄⁻ ions by sodium citrate.³⁶ In brief, sodium citrate (1.5 mL, 10 mg mL⁻¹) was added to an aqueous solution (100 mL) containing HAuCl₄ (1 wt%, 1 mL). Then, the mixture was heated to reflux and kept under these conditions for 15 min. After cooling down, the mixture was stored at 4 °C, which provided an Au NP solution that was wine-red.

Then, the prepared amino-functionalized MWCNTs–Fe₃O₄ (10 mg) were added to the Au NP (20 mL) solution. The suspension was stirred for 12 h. The Au NPs could bind with all of the amino groups on the surface of the amino-functionalized MWCNTs–Fe₃O₄. The sediment was dried and the obtained powder was designated as Au@MWCNTs–Fe₃O₄.

2.4. Preparation of Pb²⁺@Au@MWCNTs–Fe₃O₄

Au@MWCNTs–Fe₃O₄ (4 mg) were dispersed into a lead nitrate solution (4 mL, 1 mg mL⁻¹). The solution was oscillated for 24 hours to achieve the goal that Pb²⁺ was adsorbed as much as possible onto the Au@MWCNTs–Fe₃O₄. The Pb²⁺@Au@MWCNTs–Fe₃O₄ were obtained for further use after magnetic separation. The Pb²⁺@Au@MWCNT–Fe₃O₄ solution was prepared as follows: a certain amount of Pb²⁺@Au@MWCNTs–Fe₃O₄ was dispersed in ultrapure water to obtain the required concentration. Fig. 1A shows the preparation procedure of the Pb²⁺@Au@MWCNTs–Fe₃O₄.

Fig. 1 (A) The preparation procedure of the Pb²⁺@Au@MWCNTs–Fe₃O₄; (B) the fabrication process of the label-free electrochemical immunosensor.

2.5. Fabrication of the immunosensor

Fig. 1B shows the fabrication process of the label-free electrochemical immunosensor. A bare glassy carbon electrode (GCE) was polished repeatedly using alumina powder and thoroughly washed. The pretreated bare glassy carbon electrode (GCE) was coated with Pb²⁺@Au@MWCNTs–Fe₃O₄ (2 mg mL⁻¹, 6 mL) and dried under atmospheric temperature. Following that, the resultant electrode was incubated with Ab₁ (10 μg mL⁻¹, 6 μL) and dried at 4 °C. After storing at 4 °C for drying, the nonspecific binding sites for CEA were blocked using 3 μL of 1 wt% bovine serum albumin (BSA) at room temperature for 1 h. Subsequently, the electrode was washed thoroughly with PBS (pH = 7.4). The fabricated immunosensor could recognize different concentrations of CEA (5 fg mL⁻¹ to 50 ng mL⁻¹, 6 mL) based on immunoreaction at room temperature for 1 h. The proposed immunosensor was stored at 4 °C for further usage.

2.6. Detection of CEA

All electrochemical measurements were carried out with a conventional three-electrode system using a GCE (4 mm in diameter) as a working electrode, a platinum wire as a counter electrode and a saturated calomel electrode (SCE) as a reference electrode. PBS at pH = 7.4 was used for all the electrochemical measurements. All cyclic voltammetry (CV) experiments were performed in the conventional electrochemical cell by scanning the potential from −1.0 V to 1.0 V. An amperometric i–t curve was used to record the amperometric responses at −0.4 V. After the current became stable under stirring, 5 mM H₂O₂ was added into the PBS and the current change was recorded.

3. Results and discussion

3.1. Morphology of MWCNT–Fe₃O₄ and Au@MWCNT–Fe₃O₄ composites

As can be seen from the SEM image (Fig. 2A), the untreated carbon nanotubes had a long thin tubular shape and were irregularly agglomerated. After magnetization, there were a lot of nearly monodisperse microspheres of around 120 nm in size uniformly attached on the surface of the MWCNTs, proving the morphology change before and after Fe₃O₄ loading (Fig. 2B). The EDX spectrum (Fig. 2D) confirms the presence of Fe₃O₄ in the sample after magnetization. This also proved that the magnetic field has a significant effect on the dispersion (Fig. 2F). Furthermore, FTIR spectroscopy was used to verify each step of the MWCNT functionalization. As shown in Fig. 2G, there was a peak at 3420 cm⁻¹ corresponding to the hydroxyl group stretching vibration, and two peaks at 1631 and 1050 cm⁻¹ corresponding to the carboxyl³⁷ and carbonyl⁸ stretching vibrations. The absorption at 1431 cm⁻¹ was due to the COO–Fe bond.³⁸ A peak (in spectrum b) at 573 cm⁻¹ was typical of a Fe–O stretching vibration³⁹ of the prepared MWCNT@Fe₃O₄ composites. All of these are evidence for the synthesized material. When the Au NPs were loaded onto the MWCNTs–Fe₃O₄, the surface morphology was greatly altered. Many light particles around 20 nm in size were loaded onto the MWCNTs–Fe₃O₄ by the Au NPs and the –NH₂ groups on the surface of the amino-functionalized MWCNTs–Fe₃O₄ to construct Au–N (Fig. 2C). The EDX analysis of the Au@MWCNT–Fe₃O₄ sample confirms the presence of C, Fe, Au and O elements (Fig. 2E).

Fig. 2 The SEM images of the MWCNTs (A), MWCNTs–Fe₃O₄ (B), and Au@MWCNTs–Fe₃O₄ (C); EDX spectrum of the MWCNTs–Fe₃O₄ (D); EDX spectrum of the Au@MWCNTs–Fe₃O₄ (E); comparison of MWCNT–Fe₃O₄ solution in the absence (a) and presence (b) of a magnet (F); FTIR spectra of (a) MWCNTs, (b) MWCNTs–Fe₃O₄ (G).

3.2. Comparison of the electron transfer ability of the different materials

For label-free immunosensors, the sensitivity is very dependent on the transducing material. In order to verify that Pb²⁺@Au@MWCNTs–Fe₃O₄ provide a superior electrochemical performance, Au NPs, MWCNTs, MWCNTs–Fe₃O₄, Au@MWCNTs–Fe₃O₄ and Pb²⁺@Au@MWCNTs–Fe₃O₄ were loaded onto the GCE surface, respectively, to test the electrocatalytic performance for H₂O₂ reduction (Fig. 3A). Tested using an amperometric i–t curve, a weak signal was detected when Au NPs (curve a) and MWCNTs (curve b) were loaded onto the electrode. Fe₃O₄ has a great auxiliary catalytic activity towards H₂O₂ reduction.²⁸ When MWCNTs–Fe₃O₄ were used to modify the bare GCE, a much larger current response was observed (curve c). The electrochemical signal was further increased (curve d) when Au@MWCNTs–Fe₃O₄ were loaded onto the electrode. As expected, the immunosensor using Pb²⁺@Au@MWCNTs–Fe₃O₄ to modify the bare GCE displayed the highest current change (curve e). These results suggest that Pb²⁺ and Au NPs promote multiple signal amplification toward the reduction of H₂O₂ as an analytical signal. The Pb²⁺@Au@MWCNTs–Fe₃O₄ possess excellent electrochemical performance to improve the sensitivity of the proposed immunosensor.

Fig. 3 (A) Amperometic response of the immunosensors with different materials: (a) Au NPs, (b) MWCNTs, (c) MWCNTs–Fe₃O₄, (d) Au@MWCNTs–Fe₃O₄, and (e) Pb²⁺@Au@MWCNTs–Fe₃O₄; (B) CV of the immunosensor using Pb²⁺@Au@MWCNTs–Fe₃O₄ as the transducing material in PBS at pH = 7.4 before (curve b) and after (curve c) the addition of 5 mM H₂O₂. For comparison purposes, a bare GCE was scanned in 0.1 mg mL⁻¹ of Pb²⁺ from −1.0 V to 1.0 V (curve a).

CV was used to further prove the successful adsorption of Pb²⁺, as shown in Fig. 3B. An oxidation peak potential around −0.25 V was found when a bare GCE was scanned in a 0.1 mg mL⁻¹ of Pb²⁺ solution (curve a). The oxidation peak was also found around −0.25 V when the electrode was scanned (curve b) in PBS (pH = 7.4) using Pb²⁺@Au@MWCNTs–Fe₃O₄ as the transducing material. From the comparison, it was obvious that Pb²⁺ was adsorbed successfully. Subsequently, the immunosensor using Pb²⁺@Au@MWCNTs–Fe₃O₄ as the transducing material was scanned in PBS (pH = 7.4) with the addition of 5 mM H₂O₂. After the addition of H₂O₂, a dramatic increase in the reduction current (curve c) was observed at the same potential, which indicates that the Pb²⁺@Au@MWCNTs–Fe₃O₄ have a good electrocatalytic performance towards the reduction of H₂O₂.

3.3. Optimization of experimental conditions

In order to obtain a better electrochemical signal, experimental conditions including pH and the concentration of Pb²⁺@Au@MWCNTs–Fe₃O₄ were optimized. Firstly, the pH of the PBS has a great influence on the electrochemical properties of the immunosensors. As shown in Fig. 4A, the current signal increases with the variation of pH from 5.6 to 7.4, and then decreases with the variation of pH from 7.4 to 8.7. The pH value of 7.4 presents the largest electrochemical signal. The experimental results show that the optimal current signal was achieved at pH 7.4. Therefore, PBS at pH 7.4 was used as an electrolyte for electrochemical tests.

Fig. 4 The optimization of experimental conditions with pH (A), and Pb²⁺@Au@MWCNTs–Fe₃O₄ concentration (B); error bars = RSD (n = 5).

The concentration of the Pb²⁺@Au@MWCNTs–Fe₃O₄ loaded onto the surface of the electrode has important implications in the response of the electrochemical sensor. The amperometric i–t method was used to investigate the electrochemical signal response of different concentrations of Pb²⁺@Au@MWCNTs–Fe₃O₄. As seen in Fig. 4B, with the increase in the concentration from 0.5 mg mL⁻¹ to 2.0 mg mL⁻¹, the current responses first increased, and then decreased with a further concentration increase from 2.0 to 3.5 mg mL⁻¹. The increase in Pb²⁺@Au@MWCNTs–Fe₃O₄ film thickness may lead to an increase in interface electron transfer resistance. Therefore, the concentration of 2.0 mg mL⁻¹ was used as the optimal concentration in this study.

3.4. Characterization of the immunosensor

Electrochemical impedance spectroscopy (EIS) is regarded as an effective method to probe the process of a modified electrode surface.^40,41 A typical impedance spectrum includes a semicircle portion and a straight line portion. The semicircle portion represents the electron-transfer-limited process, which can be observed in the higher region. The linear portion represents the diffusion-limited process at lower frequencies. The semicircle diameter equals the electron-transfer resistance.⁴¹

The Nyquist plots of electrochemical impedance spectroscopy were recorded from 0.1 to 10⁵ Hz at 0.24 V in a solution containing 0.1 M KCl and 2.5 mmol L⁻¹ Fe (CN)₆³⁻/Fe(CN)₆⁴⁻. As shown in Fig. 5A, the bare GCE exhibited a very small semicircle diameter (curve a), suggesting a diffusion-limiting step of the electrochemical process. It can be seen that the semicircle is much smaller (curve b) than that of bare GCE when Pb²⁺@Au@MWCNTs–Fe₃O₄ were loaded on the surface of the GCE. The reason for this observation is that Pb²⁺@Au@MWCNTs–Fe₃O₄ are an excellent electrically conducting material, which could accelerate the electron transfer and make electron transfer easier. Then, after incubation with Ab₁, the resistance was significantly increased, demonstrating that Ab₁ was immobilized on the electrode successfully and blocked the electron transfer between the base solution and electrode (curve c). Similarly, when the BSA was loaded onto the electrode surface, the resistance was significantly increased (curve d). Another possible reason is that the modified protein molecules on the surface of the electrode greatly block the transfer of electrons. Additionally, resistance further increased with the addition of CEA (curve e), because the added molecules resist the electron-transfer kinetics of the redox probe at the electrode interface. As a result, we can conclude that the biosensor has been fabricated successfully.

Fig. 5 (A) Nyquist plots of the AC impedance for each immobilization step, recorded from 1 to 10⁵ Hz, of bare GCE (a), Pb²⁺@Au@MWCNTs–Fe₃O₄/GCE (b), Pb²⁺@Au@MWCNTs–Fe₃O₄@Ab₁/GCE (c), BSA/Pb²⁺@Au@MWCNTs–Fe₃O₄@Ab₁/GCE (d), and CEA/BSA/Pb²⁺@Au@MWCNTs–Fe₃O₄@Ab₁/GCE (e) in PBS at pH = 7.4 containing 0.1 M KCl and 2.5 mM Fe(CN)₆³⁻/Fe(CN)₆⁴⁻; (B) amperometric response of the immunosensor with varying concentration of CEA at −0.4 V (5 mM H₂O₂): (a) 5 fg mL⁻¹, (b) 50 fg mL⁻¹, (c) 500 fg mL⁻¹, (d) 5 pg mL⁻¹, (e) 50 pg mL⁻¹, (f) 500 pg mL⁻¹, (g) 5 ng mL⁻¹, (h) 50 ng mL⁻¹, (i) 100 ng mL⁻¹; (C) calibration curve of the immunosensor towards different concentrations of CEA. Error bars = RSD (n = 5).

Under the optimal conditions, a label-free electrochemical immunosensor using Pb²⁺@Au@MWCNTs–Fe₃O₄ as the transducing material was applied to the detection of different concentrations of CEA using an amperometric i–t curve in pH 7.4 PBS at −0.4 V. The relationship between the amperometric response towards the reduction of H₂O₂ and CEA concentration is shown in Fig. 5B. As can be seen (Fig. 5C), a linear relationship between the amperometric response and the logarithmic values of CEA concentration was observed within the range of 0.005 pg mL⁻¹–50 ng mL⁻¹, with a detection limit of 1.7 fg mL⁻¹. The regression equation of the calibration curve is I = −2.714 log(C/5) + 4.219, with a correlation coefficient of 0.9960 at a signal to noise ratio (S/N) of 3. The low detection limit might be attributed to the Pb²⁺@Au@MWCNTs–Fe₃O₄ conjugated to amounts of capture antibodies and the synergetic effect present in Pb²⁺@Au@MWCNTs–Fe₃O₄ that favors electron transfer. They could greatly increase the response to H₂O₂, broaden the scope of testing and lead to higher sensitivity.

3.5. Comparison of different methods

Compared with previously reported methods for the detection of CEA, this specially designed label-free immunosensor has a wider linear range and lower detection limit, as is shown in Table S1.† In this work, Pb²⁺@Au@MWCNTs–Fe₃O₄ could not only immobilize the antibodies but also produce electrochemical signals. Consequently, high sensitivity is one of the advantages of this designed immunosensor.

3.6. Reproducibility, selectivity, and stability of the immunosensor

To evaluate the reproducibility of the immunosensor, five electrodes were prepared for the detection of 5 ng mL⁻¹ of CEA. An amperometric i–t curve was used to record the electrochemical signal in PBS at pH 7.4, with a concentration of Pb²⁺@Au@MWCNTs–Fe₃O₄ of 2.0 mg mL⁻¹. The relative standard deviation (RSD) of the measurements for the five electrodes was 3.4% (Fig. 6A). The results suggested acceptable reproducibility and precision of the proposed immunoassay.

Fig. 6 (A) Amperometric change response of the biosensor to different electrodes treated in same way; (B) current responses of the immunosensor to 0.5 ng mL⁻¹ CEA (1), 0.5 ng mL⁻¹ CEA + 50 ng mL⁻¹ human IgG (2), 0.5 ng mL⁻¹ CEA + 50 ng mL⁻¹ vitamin C (3), 0.5 ng mL⁻¹ CEA + 50 ng mL⁻¹ glucose (4), 0.5 ng mL⁻¹ CEA + 50 ng mL⁻¹ BSA (5). Error bars = RSD (n = 5).

To investigate the selectivity of the proposed immunosensor, interference studies were performed using human IgG (HIgG), vitamin C, glucose and BSA. The 0.5 ng mL⁻¹ CEA solutions containing 50 ng mL⁻¹ of interfering substances were measured using the proposed immunosensor and the results are shown in Fig. 6B. The results showed that the current variation due to the interfering substances was 4.4%, indicating that the selectivity of the immunosensor was acceptable.

The stability of immunosensors is also an important factor in their applications. The stability of the immunosensor was investigated by checking the current responses periodically. The immunosensor was stored in pH 7.4 PBS at 4 °C when not in use. It could be found that current responses to the same concentration of CEA showed no apparent change compared to the freshly prepared immunosensor, which was used to directly detect the same concentration of CEA without being stored, suggesting that the stability of the immunosensor was also acceptable. The reproducibility, selectivity and stability of this immunosensor were acceptable, and thus it is suitable for the determination of CEA in real samples.

3.7. Application of the immunosensor in serum sample

In order to evaluate the feasibility of the proposed immunosensor, it was used to detect the recoveries of different concentrations of CEA in human serum samples by standard addition methods. As shown in Table S2,† the recovery of CEA was from 97.5% to 102.2% and the relative standard deviation (RSD) was in the range of 3.90% to 4.38%. This shows that the proposed immunoassay methodology could be clinically applied to the detection of CEA concentrations in serum samples.

4. Conclusions

A novel electrochemical immunosensor for the sensitive detection of CEA has been developed based on Pb²⁺@Au@MWCNTs–Fe₃O₄ as a signal amplifier. To provide a high-performance electrochemical immunosensor, Pb²⁺@Au@MWCNTs–Fe₃O₄ was immobilized on the electrode, which can increase the surface area to capture a larger amount of Ab₁ as well as improve the electronic transmission rate. The proposed immunosensor has a linear response to increasing concentration of CEA, from 5 fg mL⁻¹ to 50 ng mL⁻¹. The proposed immunosensor is also characterized by a low detection limit, good reproducibility, excellent selectivity and stability. This proposed strategy might have a promising application in clinical immunoassays for other biomolecules.

Acknowledgements

This study was supported by the National Natural Science Foundation of China (no. 21375047, 21377046, 21405095) and the Project of Shandong Province Higher Educational Science and Technology Program (no. J14LC09), and QW thanks the Special Foundation for Taishan Scholar Professorship of Shandong Province and UJN (no. ts20130937). All of the authors express their deep thanks.

References

1. W. Jiang, R. Yuan, Y. Chai, L. Mao and H. Su, A novel electrochemical immunoassay based on diazotization-coupled functionalized bioconjugates as trace labels for ultrasensitive detection of carcinoembryonic antigen, Biosens. Bioelectron., 2011, 26(5), 2786–2790.
2. R. Goldman, H. W. Ressom, R. S. Varghese, L. Goldman, G. Bascug, C. A. Loffredo, M. Abdel-Hamid, I. Gouda, S. Ezzat and Z. Kyselova, Detection of hepatocellular carcinoma using glycomic analysis, Clin. Cancer Res., 2009, 15(5), 1808–1813.
3. M. Ferrari, Cancer nanotechnology: opportunities and challenges, Nat. Rev. Cancer, 2005, 5(3), 161–171.
4. J. D. Wulfkuhle, L. A. Liotta and E. F. Petricoin, Proteomic applications for the early detection of cancer, Nat. Rev. Cancer, 2003, 3(4), 267–275.
5. S. Benchimol, A. Fuks, S. Jothy, N. Beauchemin, K. Shirota and C. P. Stanners, Carcinoembryonic antigen, a human tumor marker, functions as an intercellular adhesion molecule, Cell, 1989, 57(2), 327–334.
6. Y. Wang, X. Li, W. Cao, Y. Li, H. Li, B. Du and Q. Wei, Ultrasensitive sandwich-type electrochemical immunosensor based on a novel signal amplification strategy using highly loaded toluidine blue/gold nanoparticles decorated KIT-6/carboxymethyl chitosan/ionic liquids as signal labels, Biosens. Bioelectron., 2014, 61, 618–624.
7. S. Liu, Q. Lin, X. Zhang, X. He, X. Xing, W. Lian and J. Huang, Electrochemical immunosensor for salbutamol detection based on CS–Fe₃O₄–PAMAM–GNPs nanocomposites and HRP–MWCNTs–Ab bioconjugates for signal amplification, Sens. Actuators, B, 2011, 156(1), 71–78.
8. H. Zarei, H. Ghourchian, K. Eskandari and M. Zeinali, Magnetic nanocomposite of anti-human IgG/COOH–multiwalled carbon nanotubes/Fe₃O₄ as a platform for electrochemical immunoassay, Anal. Biochem., 2012, 421(2), 446–453.
9. S. Zhang, J. Yang and J. Lin, 3,3′-Diaminobenzidine (DAB)–H₂O₂–HRP voltammetric enzyme-linked immunoassay for the detection of carcinoembryonic antigen, Bioelectrochemistry, 2008, 72(1), 47–52.
10. R. Yuan, Y. Zhuo, Y. Chai, Y. Zhang and A. Sun, Determination of carcinoembryonic antigen using a novel amperometric enzyme–electrode based on layer-by-layer assembly of gold nanoparticles and thionine, Sci. China, Ser. B: Chem., 2007, 50(1), 97–104.
11. B. Wu, C. Hu, X. Hu, H. Cao, C. Huang, H. Shen and N. Jia, Sensitive ECL immunosensor for detection of retinol-binding protein based on double-assisted signal amplification strategy of multiwalled carbon nanotubes and Ru(bpy)₃²⁺ doped mesoporous silica nanospheres, Biosens. Bioelectron., 2013, 50, 300–304.
12. Y. Zhuo, G. Gui, Y. Chai, N. Liao, K. Xiao and R. Yuan, Sandwich-format electrochemiluminescence assays for tumor marker based on PAMAM dendrimer–L-cysteine–hollow gold nanosphere nanocomposites, Biosens. Bioelectron., 2014, 53, 459–464.
13. X. Feng, N. Gan, J. Zhou, T. Li, Y. Cao, F. Hu, H. Yu and Q. Jiang, A novel dual-template molecularly imprinted electrochemiluminescence immunosensor array using Ru(bpy)₃²⁺–silica@poly-L-lysine–Au composite nanoparticles as labels for near-simultaneous detection of tumor markers, Electrochim. Acta, 2014, 139, 127–136.
14. S. Wang, Y. Zhang, J. Yu, X. Song, S. Ge and M. Yan, Application of indium tin oxide device in gold-coated magnetic iron solid support enhanced electrochemiluminescent immunosensor for determination of carcinoma embryonic antigen, Sens. Actuators, B, 2012, 171–172, 891–898.
15. S. Yu, Q. Wei, B. Du, D. Wu, H. Li, L. Yan, H. Ma and Y. Zhang, Label-free immunosensor for the detection of kanamycin using Ag@Fe₃O₄ nanoparticles and thionine mixed graphene sheet, Biosens. Bioelectron., 2013, 48(0), 224–229.
16. P. He, Z. Wang, L. Zhang and W. Yang, Development of a label-free electrochemical immunosensor based on carbon nanotube for rapid determination of clenbuterol, Food Chem., 2009, 112(3), 707–714.
17. G. Marchesini, J. Buijs, W. Haasnoot, D. Hooijerink, O. Jansson and M. Nielen, Nanoscale affinity chip interface for coupling inhibition SPR immunosensor screening with nano-LC TOF MS, Anal. Chem., 2008, 80(4), 1159–1168.
18. J. Zhang, J. Lei, C. Xu, L. Ding and H. Ju, Carbon nanohorn sensitized electrochemical immunosensor for rapid detection of microcystin-LR, Anal. Chem., 2010, 82(3), 1117–1122.
19. B. Kavosi, A. Salimi, R. Hallaj and K. Amani, A highly sensitive prostate-specific antigen immunosensor based on gold nanoparticles/PAMAM dendrimer loaded on MWCNTS/chitosan/ionic liquid nanocomposite, Biosens. Bioelectron., 2014, 52, 20–28.
20. M. Lan, C. Chen, Q. Zhou, Y. Teng, H. Zhao and X. Niu, Voltammetric detection of microcystis genus specific-sequence with disposable screen-printed electrode modified with gold nanoparticles, Adv. Mater. Lett., 2010, 1, 217–224.
21. H. Wang, J. Wu, J. Li, Y. Ding, G. Shen and R. Yu, Nanogold particle-enhanced oriented adsorption of antibody fragments for immunosensing platforms, Biosens. Bioelectron., 2005, 20(11), 2210–2217.
22. S. Zhang, J. Xia and X. Li, Electrochemical biosensor for detection of adenosine based on structure-switching aptamer and amplification with reporter probe DNA modified Au nanoparticles, Anal. Chem., 2008, 80(22), 8382–8388.
23. A. A. Ansari, A. Kaushik, P. Solanki and B. Malhotra, Sol–gel derived nanoporous cerium oxide film for application to cholesterol biosensor, Electrochem. Commun., 2008, 10(9), 1246–1249.
24. J. Tang, D. Tang, B. Su, J. Huang, B. Qiu and G. Chen, Enzyme-free electrochemical immunoassay with catalytic reduction of p-nitrophenol and recycling of p-aminophenol using gold nanoparticles-coated carbon nanotubes as nanocatalysts, Biosens. Bioelectron., 2011, 26(7), 3219–3226.
25. Y. Xiang, Y. Zhang, X. Qian, Y. Chai, J. Wang and R. Yuan, Ultrasensitive aptamer-based protein detection via a dual amplified biocatalytic strategy, Biosens. Bioelectron., 2010, 25(11), 2539–2542.
26. J. Chłopek, B. Czajkowska, B. Szaraniec, E. Frackowiak, K. Szostak and F. Beguin, In vitro studies of carbon nanotubes biocompatibility, Carbon, 2006, 44(6), 1106–1111.
27. A. O. Lobo, E. F. Antunes, M. B. Palma, C. Pacheco-Soares, V. J. Trava-Airoldi and E. J. Corat, Biocompatibility of multi-walled carbon nanotubes grown on titanium and silicon surfaces, Mater. Sci. Eng., C, 2008, 28(4), 532–538.
28. Q. Wei, Z. Xiang, J. He, G. Wang, H. Li, Z. Qian and M. Yang, Dumbbell-like Au–Fe₃O₄ nanoparticles as label for the preparation of electrochemical immunosensors, Biosens. Bioelectron., 2010, 26(2), 627–631.
29. C. Wang, C. Xu, H. Zeng and S. Sun, Recent Progress in Syntheses and Applications of Dumbbell-like Nanoparticles, Adv. Mater., 2009, 21(30), 3045–3052.
30. S. Empedocles and M. Bawendi, Quantum-confined stark effect in single CdSe nanocrystallite quantum dots, Science, 1997, 278(5346), 2114–2117.
31. S. S. Wong, A. T. Woolley, E. Joselevich, C. L. Cheung and C. M. Lieber, Covalently-Functionalized Single-Walled Carbon Nanotube Probe Tips for Chemical Force Microscopy, J. Am. Chem. Soc., 1998, 120(33), 8557–8558.
32. H. Niu, R. Yuan, Y. Chai, L. Mao, H. Liu and Y. Cao, Highly amplified electrochemiluminescence of peroxydisulfate using bienzyme functionalized palladium nanoparticles as labels for ultrasensitive immunoassay, Biosens. Bioelectron., 2013, 39(1), 296–299.
33. J. Han, Y. Zhuo, Y. Chai, Y. Yu, N. Liao and R. Yuan, Electrochemical immunoassay for thyroxine detection using cascade catalysis as signal amplified enhancer and multi-functionalized magnetic graphene sphere as signal tag, Anal. Chim. Acta, 2013, 790(0), 24–30.
34. Z.-D. Gao, F.-F. Guan, C.-Y. Li, H.-F. Liu and Y.-Y. Song, Signal-amplified platform for electrochemical immunosensor based on TiO₂ nanotube arrays using a HRP tagged antibody–Au nanoparticles as probe, Biosens. Bioelectron., 2013, 41, 771–775.
35. G. Morales-Cid, A. Fekete, B. M. Simonet, R. Lehmann, S. Cárdenas, X. Zhang, M. Valcárcel and P. Schmitt-Kopplin, In situ synthesis of magnetic multiwalled carbon nanotube composites for the clean-up of (fluoro) quinolones from human plasma prior to ultrahigh pressure liquid chromatography analysis, Anal. Chem., 2010, 82(7), 2743–2752.
36. G. Frens, Controlled nucleation for the regulation of the particle size in monodisperse gold suspensions, Nature, 1973, 241(105), 20–22.
37. K. V. Singh, R. R. Pandey, X. Wang, R. Lake, C. S. Ozkan, K. Wang and M. Ozkan, Covalent functionalization of single walled carbon nanotubes with peptide nucleic acid: nanocomponents for molecular level electronics, Carbon, 2006, 44(9), 1730–1739.
38. R. Y. Hong, T. T. Pan and H. Z. Li, Microwave synthesis of magnetic Fe₃O₄ nanoparticles used as a precursor of nanocomposites and ferrofluids, J. Magn. Magn. Mater., 2006, 303(1), 60–68.
39. S. Zhang, B. Du, H. Li, X. Xin, H. Ma, D. Wu, L. Yan and Q. Wei, Metal ions-based immunosensor for simultaneous determination of estradiol and diethylstilbestrol, Biosens. Bioelectron., 2014, 52(0), 225–231.
40. Q. Zhang, I. Lee, J. Ge, F. Zaera and Y. Yin, Surface-Protected Etching of Mesoporous Oxide Shells for the Stabilization of Metal Nanocatalysts, Adv. Funct. Mater., 2010, 20(14), 2201–2214.
41. A. Guo, Y. Li, W. Cao, X. Meng, D. Wu, Q. Wei and B. Du, An electrochemical immunosensor for ultrasensitive detection of carbohydrate antigen 199 based on Au@CuxOS yolk–shell nanostructures with porous shells as labels, Biosens. Bioelectron., 2015, 63, 39–46.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra16569a

---