Electrochemical immunosensor based on carboxylated single‐walled carbon nanotube‐chitosan functional layer for the detection of cephalexin

Abstract In this study, a sensitive and selective electrochemical immunosensor for cephalexin (CEX) determination on a glassy carbon electrode (GCE) surface was modified by a carboxylated single‐walled carbon nanotubes/chitosan (SWNTs‐COOH/CS) composite. The SWNTs‐COOH/CS composite was used to enhance sensor performance and to enlarge the electrochemical response of CEX. The cephalosporin‐ovalbumin coupling (CEX‐OVA) was synthesized using the reactive ester method. The free CEX in solution could be effectively measured based on the competitive immunoreaction between CEX‐antibody and CEX. Under optimal conditions, the electrochemical immunosensor offered an excellent response for CEX. The linear range was 1–800 ng/ml, with a detection limit of 45.7 ng/ml (S/N = 3). This method was applied to determine CEX in six different samples and obtained the recovery range from 80.15% to 94.04%. These results indicated that the fabricated electrochemical immunosensor and sensing method are suitable for quantification of CEX in real samples. These have great potential for wider applications in environmental and agri‐food products industries.

with the stipulations during withdrawal periods, the cephalexin residues in animal foods were consequently becoming more severe. In particular, excessive residues in milk have been endangering human health and food safety issues, and affecting the export trade severely (Xie et al., 2009;Zhi, Meyer, Bedem, & Meusel, 2001). Studies have shown the phenomenon of over-standard use of cephalosporin antibiotics in livestock and poultry farms. Therefore, sensing the presence of CEX in food samples is of great importance not only for consumer exposure assessment on CEX but also for identifying the abuse of the antibiotic CEX.
Traditional detection methods include microbiological methods, spectrophotometry, high-performance liquid chromatography (HPLC), and high-performance liquid chromatography-mass spectrometry (HPLC-MS). However, these methods inevitably count on advanced apparatuses being available to deal with complicated operations, and it takes a long time to complete (Cazorla-Reyes, Romero-González, Frenich, Rodríguez Maresca, & Martínez Vidal, 2014;Han et al., 2012;Zhang et al., 2013). Therefore, it is necessary to develop simple, rapid, sensitive, and low-cost methods for sensing CEX residue in food products.
Electrochemical immunosensor combines specific ELISA with highly sensitive sensing technology, good selectivity, and fast analyzing ability (Gayathri, Mayuri, Sankaran, & Kumar, 2016;Jing et al., 2016;Zhang, Li, Ma, Zhang, & Zheng, 2016). In recent years, the rapid development of nanomaterials technologies have drawn much research attention. Single-walled carbon nanotubes, one of the nano-materials, were made of a grapheme monolayer which has good physical and chemical properties. Furthermore, it is considered to be a high-quality sensor signal (Cao, Zhang, Chai, & Yuan, 2013;Tong et al., 2016;Yu et al., 2015). Electrochemical immunosensor, single-walled carbon nanotubes (SWNTs) were introduced to enhance the ability of electron transfer, biocompatibility of the working electrode surface, and the sensitivity of the electrochemical immunosensor method (Zhang, Li, Wang, et al., 2016). Yang et al. (2016) using ultrasound to uniformly disperse single-walled carbon nanotubes in chitosan to form a composite membrane, modified the surface of the electrode to construct an electrochemical immunosensor. This was done to detect fumonisin B 1 , whereby the linearity is good, the minimum detection limit is 2 g/ml, which is much lower than the EU limit of 2-4 mg/L. Furthermore, recovery and reproducibility is better and can be applied to the actual sample subjected to fumonisin B 1 detection . The work done by Zhang et al. presents the carbon nanotubes modified on the electrode surface-immobilized antigen coated with nano-gold labeled secondary antibody.
The hydrochloric acid that occurs in electrical oxidation, resulted in changes in the signal to complete the microsystem. The linear range was 0.0025-5 μg/L, and the lowest detection limit was 1.68 ng/ml.
Both indicated good accuracy and reproducibility, and there a wide range of applications in environmental monitoring emerged (Zhang, Kang, Hao, Lu, & Cheng, 2015). The present study successfully produced an electrochemical immunosensor for the rapid detection of cephalexin in animal foods. This method was more sensitive than other methods and has good reproducibility and satisfactory results.

| Apparatus
The differential pulse voltammetry (DPV), and cyclic voltammetry (CV) measurements were carried out on CHI 660e electrochemistry workstation (Shanghai CH Instruments), connected to a computer.
The morphology and particle sizes of the samples were estimated from scanning electron microscope (SEM) (Hitachi 600). All experiments were performed with a conventional three-electrode system.
The modified glassy carbon electrode (GCE) was used as the working electrode, a platinum wire as the counter electrode, and Ag/AgCl (sat. KCl) as the reference electrode.

| Preparation of immunosensor
The electrode was first polished with 0.3 μm alumina, ultrasonically cleaned in sulfuric acid, absolute ethanol, deionized water, and dried at room temperature. CEX-OVA was prepared using the active ester method. SWNTs-COOH/CS (1 mg/ml) nano-composites were prepared using ultrasonication of SWNTs-COOH, chitosan, and 1% acetic acid solution (Zhang, Wang, Qiuwei, & Shuguo, 2016 About 10 μl of 1% bovine serum albumin (BSA) was added to the surface of CEX-OVA/SWNTs-COOH/CS/GCE and kept for 1 hr to block the unreacted active sites on the electrode surface. Finally, the prepared electrodes were thoroughly washed with PBS buffer to obtain CEX-OVA/SWNTs-COOH/CS/GCE, which was then stored in a refrigerator at 4°C for later use.
About 5 μl of PBS (pH 7.4)-diluted CEX in the range of 0-1,000 μg/ ml was evenly mixed with 5 ml of anti-CEX (diluted at 1:160). The mixture was added to the surface of CEX-OVA/SWNTs-COOH/CS/ GCE electrode and incubated at 37°C for 50 min to obtain anti-CEX/ CEX-OVA/SWNTs-COOH/CS/GCE. During the reaction, CEX-OVA fixed on the electrode competed with free CEX in the mixed solution to produce a certain amount of anti-CEX.
The electrode was washed with phosphate buffer solution (1× PBS, pH 7.4). About 10 μl of 160-fold diluted horseradish peroxidase was dropped vertically using a pipette gun to label goat anti-rabbit IgG(H + L) secondary antibody and then incubated at 37°C for 50 min. Then, the prepared buffer solution of 3 mM H 2 O 2 and 3 mM HQ (1:1, v/v) were used as the immune response buffer system, in which the working electrode was placed. The saturated calomel electrode was used as the reference electrode, and the platinum wire electrode was used as the counter electrode, which was determined using differential pulse voltammetry (DPV). Scheme 1 shows the detailed production of the sensor.

| Sample preparation
Milk-derived and meat products are purchased at local supermarkets. The dairy products were put accurately into the polystyrene centrifuge tube, and 35 ml acetonitrile was added. It was then mixed and subjected to ultrasonic extraction for 30 min. Then, it was cooled to room temperature, followed by adding 200 g/L zinc acetate 2.0 ml, acetonitrile constant volume to 50 ml, vortexed for 0.5 min, subjected to 15,000 g centrifugation for 5 min, collection of supernatant (10 ml), 15,000 g and then centrifuged for 5 min. The supernatant was collected into a 0.22 μm filter and diluted 10 times by PBS solution before use.
It is important to obtain a proper-sized meat sample and stir it well. Then, it is extracted for 15 min at room temperature with 1:1 (v:v) methanol/water (8:2), centrifuged at 2,400 g for 15 min and a supernatant is created through an 0.45 μm organic filter and diluted 20 times with a PBS solution before use.

| SEM characterization of SWNTs-COOH, CS and SWNTs-COOH/CS nano-composites
The morphologies and microstructures of the as-prepared different modified films were investigated by using SEM (Figure 1). The results in Figure 1a show that the structure of SWNTs-COOH film is a smooth sheet. Figure 1b illustrates the loose cluster structure of CS. Figure 1c depicts a good dispersion structure when the CS was embedded steadily and evenly on the surface of SWNTs-COOH.
This phenomenon suggests the successful adsorption of the CS on the SWNTs-COOH surface through carboxyl affinity. Furthermore, the SWNTs-COOH/CS can significantly increase the sensitivity of the sensor by providing greater biocompatibility and a specific surface area.

| XRD characterization of SWNTs-COOH/ CS structure
The structure of SWNTs-COOH/CS was characterized by XRD. The results are shown in Figure 2. In the spectra of the complexes, the 2θ characteristic peaks of CS appeared between 10° and 20°, and the two 2θ characteristic peaks of SWNTs appeared between 35° and 45°. This indicated that CS and SWNTs-COOH were well combined and the synthesis of SWNTs-COOH/CS was successful.

| Electrochemical characterization of the immunosensor
The

| Optimization of experimental conditions
In order to achieve excellent analytical capabilities, some of the key parameters of the test were optimized. As shown in Figure 4a, the peak current of the CV increased gradually, when SWNTs-COOH/CS was between 4 and 12 μl. This could possibly be caused by the excellent electron transfer ability of SWNTs-COOH/CS. The peak current reached the maximum value of (5.46 × 10 -4 A) when SWNTs-COOH/ CS was 12 μl. An excessive amount of SWNTs-COOH/CS gathering on the electrode surface will impede the transfer of electrons.
Therefore, 12 μl was used as the optimal amount of SWNTs in this work. As shown in Figure 4b, the peak current detected by DPV increases when the CEX-OVA concentration also increases. When the concentration reaches 200 μg/ml, the peak current tends to be stable, so the optimal working concentration of CEX-OVA is 200 μg/ ml. The concentration of anti-CEX in the incubation solution is also an important parameter in competitive format due to a fixed amount of CEX-OVA domain at the electrode interface and a competitive immune response to the binding of free CEX to anti-CEX in solution.
In order to obtain the optimal concentration of anti-CEX, the pre- The effect of HRP-anti-antibody dilution factor on the peak current of DPV is shown in Figure 4e, when HRP-anti-antibody is diluted 100 times, the peak current of DPV reaches the maximum and then gradually stabilizes. Therefore, 100 times is selected as the optimal dilution multiple.
The factors that influence the different scanning rates on

| Analytical characteristics for CEX determination
To evaluate the performance of the established sensor, the concentration of CEX was determined using the DPV with prepared CEX-OVA/SWNTs-COOH/CS/GCE. The calibration curve was linear in the range of 1 μg/ml to 800 μg/ml ( Figure 6). The limit of detection was 45.7 ng/ml (S/N = 3). Compared with some of the detection methods based on CEX, as shown in Table 1, it is indicated that the proposed method was more sensitive than most of those previously studied.  has a strong ability to penetrate the outer membrane and affinity with penicillin-binding protein (PBPs), the methoxy group increases steric hindrance and its polar group thiol group makes it more stable to β-lactamase (Gholivand, Torkashvand, & Malekzadeh, 2012).

| The specificity of the immunosensor
The substituent of CFX is connected to the methoxy group, and its stability is improved. The side chain of PER is replaced by the N-methyltetrazole side chain (the MTT side chain), and the substituent of PER is complex. Its stability increases. The substituent of ZOL is N-methyl thiadiazole side chain (MDT side chain) (Barbosa, Araujo, & Ferreira, 2012;Yang, Zhao, & Zeng, 2014), the steric hindrance increases. The structure of TAX, CFX, PER, and ZOL are quite different from CEX, so the cross-reaction rate is low.

| The stability of the immunosensor
To further ascertain the reproducibility of the experimental results, six different HRP-anti-antibody/anti-CEX/CEX-OVA/SWNTs-COOH/CS/GCEs were tested toward the oxidation of 50 μmol/L CEX. The peak currents obtained by the six independent electrodes showed RSD of 8.6%, confirming that the sensor is reproducible. The stability of the HRP-anti-antibody/anti-CEX/CEX-OVA/SWNTs-COOH/CS/GCE was also tested. After measurements, the modified electrode was stored at 4°C. As seen from Figure 9, the peak current intensity only decreased by 10.7% for the sensor after 6 days. The RSD (n = 5) for all these species was less than 11%. These results mentioned above indicated that the modified electrode possesses good reproducibility and good stability.

| Application to actual sample determination
Different tissue samples were treated according to the methods described in Section 2.4, as shown in Figure 10. There were some differences but only a little between the standard curve in working buffer and the adjusted standard curves which were built in two different sample extracts. It is indicated that the organic reagents in the diluted extract and the matrix effect exert little influence on the detection performance of the developed method. The adjusted standard curves were used for the real sample analysis in the following studies.
In order to validate the feasibility and sensibility of the developed immunosensor on detecting CEX, the content of CEX in milk samples, and meat samples were determined using the spiked recoveries F I G U R E 8 The molecular structure diagrams of CEX, CER, RAD, TAX, CFX, PER, and ZOL F I G U R E 9 Current value changes in 6 days method. The measurement results in Tables 1 and 2 indicated its good recoveries at three levels, ranging from 80.15% to 94.04%.
Therefore, after improvement, the method could be regarded as a potential alternative for quantitative determination of CEX.
The comparisons between this electrochemical immunosensor and other recently reported methods for CEX were shown in Table 3.
The LODs of these methods were ranging from 1.8 to 288 ng/ml, and our developed method exhibited a LOD of 45.7 ng/ml, which was lower than the MRLs of CEX in different food samples. So this method can meet the need for determination of CEX. In addition, the recoveries were better than most of other methods. This simple, reasonable, and economical method offers a new way for rapid and sensitive detection of CEX.

| CON CLUS IONS
In conclusion, this study described a sensitive, specific, and simple electrochemical immunosensor for detection of CEX by integrating SWNTs-COOH/CS with enzymatic signal readout. The developed approach embodied its acceptable fabrication reproducibility and a high level of specificity and sensitivity for the detection of CEX at levels of 1-800 μg/ml. The sensor might further furnish a versatile and forceful tool to guarantee food safety.

ACK N OWLED G M ENTS
The authors would like to express their acknowledgment for financial support by the Science and Technology Plan Project of Hebei

CO N FLI C T O F I NTE R E S T
None.

E TH I C A L A PPROVA L
This research did not include any human subjects and animal experiments.