A Label-Free Immunosensor Based on Graphene Oxide/Fe3O4/Prussian Blue Nanocomposites for the Electrochemical Determination of HBsAg

In this article, a highly sensitive label-free immunosensor based on a graphene oxide (GO)/Fe3O4/Prussian blue (PB) nanocomposite modified electrode was developed for the determination of human hepatitis B surface antigen (HBsAg). In this electrochemical immunoassay system, PB was used as a redox probe, while GO/Fe3O4/PB nanocomposites and AuNPs were prepared and coated on screen-printed electrodes to enhance the detection sensitivity and to immobilize the hepatitis B surface antibody (HBsAb). The immunosensor was fabricated based on the principle that the decrease in peak currents of PB is proportional to the concentration of HBsAg captured on the modified immunosensor. The experimental results revealed that the immunosensor exhibited a sensitive response to HBsAg in the range of 0.5 pg·mL−1 to 200 ng·mL−1, and with a low detection limit of 0.166 pg·mL−1 (S/N = 3). Furthermore, the proposed immunosensor was used to detect several clinical serum samples with acceptable results, and it also showed good reproducibility, selectivity and stability, which may have a promising potential application in clinical immunoassays.


Introduction
Recently, the hepatitis B virus (HBV) has become a leading cause of death worldwide [1]. Obviously, the earlier the diagnosis, the more treatment options are available and the greater the possibility of healing. Therefore, the highly sensitive detection of viruses in human serums is considered to be a key point in the processes of treatment of HBV [2,3]. Furthermore, once you have been infected with chronic hepatitis B and this is not been arrested in a timely manner and effectively, it will further evolve into acute hepatitis, chronic hepatitis and cirrhosis, and will even lead to serious complications such as liver cancer, eventually leading to death [4]. Therefore, the prevention of hepatitis has gradually become an important public health problem all over the world [5].
HBsAg is one of the main markers of HBV infection, and it usually occurs after being infected for 1-2 weeks [6]. Many methods have been exploited to detect HBsAg in the past few decades [7], including time-resolved fluoroimmunossay (TRFIA), electro-chemiluminescence immunoassay (ECLIA), gold immunochromatography assay (GICA), ELISA, etc. [8][9][10]. In particular, the ELISA test has since been seen as the "gold standard" for comparison against all newly developed immunoassays and immunosensors [11]. However, the methods above usually need expensive instruments, skilled operation and strict culture conditions, but have low sensitivity and a narrow range [12]. In contrast with this, as an innovation for conventional determination methods, electrochemical immunosensors are a kind

Apparatus
All electrochemical measurements were carried out on a CHI660D electrochemical Workstation (Chenhua Instrument Co, Shanghai, China). Constant temperature Biochemical Culture was carried out on a BSD-100 (Shanghaibo xunshiye Company, Shanghai, China). A conventional three-electrode system was used for all electrochemical measurements and a bare SPE or modified SPE was served as the measurement electrode. The pH measurements were carried out on a PHS-3E (Shanghai INESA Scientific Instrument CO. Ltd., Shanghai, China). Centrifugal processes were all accomplished by using the high-speed freezing centrifuge TGL-20M (Hunan Xiangyi Development Co. Ltd., Yiyang, China).

Preparation of Fe 3 O 4 /GO Nanocomposites
Fe 3 O 4 /GO nanocomposites were synthesized based on the previous reported method with a little modification [43,44]. Briefly, 15 mg GO was ultrasonic dispersed in 75 mL (CH 2 OH 2 ) 2 for 2.5 h. Afterwards, 0.81 g FeCl 3 was added to the above GO aqueous solution and dissolved thoroughly. Then, 1.23 g CH 3 COONa was added to the mixture and stirred vigorously for 30 min, until the color changed from black to yellow brown. The obtained mixture was moved into a Teflon-lined stainless-steel autoclave and heated to 180 • C for 11 h. After it was cooled to room temperature, the obtained Fe 3 O 4 /GO nanocomposites were isolated in the magnetic field and washed several times with ethanol and ultrapure water. Finally, the products were dried in a vacuum at 60 • C for 12 h.

Preparation of PB/Fe 3 O 4 /GO Nanocomposites
A total of 2 mg Fe 3 O 4 /GO nanocomposites were dispersed homogeneously in 2 mL ultrapure water by sonication. Then, 2 mL of the above aqueous solution was added into a 2 mL aqueous solution (pH 1.5, adjusted with HCl) containing 15 mmol L −1 K 3 Fe (CN) 6 and 15 mmol L −1 FeCl 3 ·6H 2 O. After vigorously stirring for 5 h, the color changed from yellow brown to dark cyan, which indicated that GO/Fe 3 O 4 nanocomposites were completely synthesized. The final mixture was separated by a magnet and washed several times and then dispersed in 2 mL ultrapure water.

Fabrication of the Immunosensor
In total, 1.5 µL GO/Fe 3 O 4 /PB of the nanocomposites (1 mg/mL) were dropped onto the surface of the SPE and dried at a constant temperature in a biochemical incubator (25 • C). The modified SPE was immersed in 50 µL (0.5 mM) HAuCl 4 , which was performed by using the electrochemical workstation to complete the electrodeposition system at a potential of −0.5 V for 180 s. A total of 5.0 µL HBsAb (0.25 mg/mL) was dropped onto the modified electrode and stored at 4 • C for 12 h. For the further fabrication of the immunosensor, it was blocked through incubation in 10 µL 1% BSA for 30 min to Biosensors 2020, 10, 24 4 of 12 avoid possible nonspecific adsorption. After each step, the fabricated SPE was thoroughly cleaned with PBS and dried at room temperature prior to use.

Electrochemical Measurements
In total, 5 µL standard HBsAg solution at different concentrations was dipped on the proposed immunosensor and incubated at 25 • C for 40 min, followed by washing with PBS buffer. After that, the electrochemical measurements were performed in 50 µL PBS (0.01 M, PH = 7.4) by cyclic voltammetry (CV), differential pulse voltammetry (DPV) and electrochemical impedance spectroscopy (EIS). The change in the peak currents was proportional to the concentrations of HBsAg captured on the modified electrode immunosensor. Figure 1 shows the fabrication procedure of the immunosensor. For the determination of HBsAg, GO/Fe 3 O 4 nanocomposites were synthesized in the first step. As illustrated in Figure 1, Fe 3 O 4 NPs were grown on the GO surface through the solvothermal method, while PB were attached onto the GO/Fe 3 O 4 surface by an in situ reduction method. A stable PB/Fe 3 O 4 /GO composite with electrocatalytic activity was first coated on the carbon working electrode, while PB was used as redox probe. Subsequently, AuNPs were attached onto the modified electrode by electrodeposition, which offered an interface for HBsAb immobilization.

Electrochemical Measurements
In total, 5 μL standard HBsAg solution at different concentrations was dipped on the proposed immunosensor and incubated at 25 °C for 40 min, followed by washing with PBS buffer. After that, the electrochemical measurements were performed in 50 μL PBS (0.01 M, PH = 7.4) by cyclic voltammetry (CV), differential pulse voltammetry (DPV) and electrochemical impedance spectroscopy (EIS). The change in the peak currents was proportional to the concentrations of HBsAg captured on the modified electrode immunosensor. Figure 1 shows the fabrication procedure of the immunosensor. For the determination of HBsAg, GO/Fe3O4 nanocomposites were synthesized in the first step. As illustrated in Figure 1, Fe3O4 NPs were grown on the GO surface through the solvothermal method, while PB were attached onto the GO/Fe3O4 surface by an in situ reduction method. A stable PB/Fe3O4/GO composite with electrocatalytic activity was first coated on the carbon working electrode, while PB was used as redox probe. Subsequently, AuNPs were attached onto the modified electrode by electrodeposition, which offered an interface for HBsAb immobilization. CV is an effective and convenient technique for probing the fabrication process of the modified electrode surface. Here, CV was used to further characterize the stepwise assembly process of the modified electrode. As shown in Figure 2A, no obvious redox peaks could be observed at the bare electrode (curve a). Subsequently, on the GO/Fe3O4/PB modified electrode, a couple of clear and symmetric redox peak at −0.048 V and −0.071 V (curve b) could be observed. The background current of the GO/Fe3O4/PB modified electrode was greater than the bare electrode, exhibiting the efficient redox activity and excellent conductivity of the GO/Fe3O4/PB nanocomposite. After HAuCl4 was electrodeposited onto the GO/Fe3O4/PB composite film, the peak current was further increased, which was because AuNPs could increase the electron-transfer efficiency (curve c). Furthermore, a large amount of AuNPs had strong adsorption capacity to protein and can bond with HBsAb molecule to form AuNPs-Ab nanocomposites. After the modified electrode was incubated with HBsAb, there was an obvious decrease of the current response (curve d), suggesting that HBsAb was successfully immobilized on the electrode. Then, the peak currents further decreased as BSA was employed to block possible remaining active sites (curve e). Particularly, the peak current gradually decreased after the immunosensors were incubated in HBsAg solution as the immunocomplex was formed (curve f). After this, HBsAb molecules were combined with the HBsAg molecules, which acted as the insulating protein layers on the electrode retarding the electron transfer. Especially, the decrease of the peak current was related to the amount of HBsAg captured on the modified electrode surface. CV is an effective and convenient technique for probing the fabrication process of the modified electrode surface. Here, CV was used to further characterize the stepwise assembly process of the modified electrode. As shown in Figure 2A, no obvious redox peaks could be observed at the bare electrode (curve a). Subsequently, on the GO/Fe 3 O 4 /PB modified electrode, a couple of clear and symmetric redox peak at −0.048 V and −0.071 V (curve b) could be observed. The background current of the GO/Fe 3 O 4 /PB modified electrode was greater than the bare electrode, exhibiting the efficient redox activity and excellent conductivity of the GO/Fe 3 O 4 /PB nanocomposite. After HAuCl 4 was electrodeposited onto the GO/Fe 3 O 4 /PB composite film, the peak current was further increased, which was because AuNPs could increase the electron-transfer efficiency (curve c). Furthermore, a large amount of AuNPs had strong adsorption capacity to protein and can bond with HBsAb molecule to form AuNPs-Ab nanocomposites. After the modified electrode was incubated with HBsAb, there was an obvious decrease of the current response (curve d), suggesting that HBsAb was successfully immobilized on the electrode. Then, the peak currents further decreased as BSA was employed to block possible remaining active sites (curve e). Particularly, the peak current gradually decreased after the immunosensors were incubated in HBsAg solution as the immunocomplex was formed (curve f). After this, HBsAb molecules were combined with the HBsAg molecules, which acted as the insulating protein layers on the electrode retarding the electron transfer. Especially, the decrease of the peak current was related to the amount of HBsAg captured on the modified electrode surface. As EIS can effectively probe the electron transfer kinetics at the electrode surface, it was also used here to characterize the stepwise assembly of the immunosensor. The impedance spectra included a semicircle portion and a linear portion. At higher frequencies, the semicircle portion corresponded to the electron-transfer limited process, and at lower frequencies, the linear portion represented the diffusion-limited process. Figure 2B shows the EIS of the stepwise modification processes performed in 5.0 mM potassium ferricyanide. Compared with the bare SPE (curve a), a greatly lower resistance was obtained after GO/Fe3O4/PB was modified on the electrode, implying that GO/Fe3O4/PB nanocomposites have excellent electric conduction and could accelerate the electron transfer in to some degree (curve b). Furthermore, the GO/Fe3O4/PB @AuNPs showed a much lower resistance than the previous example, indicating that the electrodeposited AuNPs were highly beneficial to the electron transfer (curve c). In contrast, after HBsAb was immobilized on GO/Fe3O4/PB@AuNPs, the resistance obviously increased (curve d), suggesting that HBsAb was successfully immobilized on the surface and blocked the electron transport between the redox probe and electrode. Then the resistance increased again when BSA was immobilized onto the immunosensor (curve e). Finally, when HBsAg interacted with the proposed immunosensor, an increase of resistance was observed (curve f), suggesting the effective specific recognition between antibodies and antigens. This was consistent with the fact that the hydrophobic layer of protein insulates the conductive support and hinders the interfacial electron transfer. These results were in accordance with the CV measurements, demonstrating the successful construction of a GO/Fe3O4/PBbased immunosensor for HBsAg detection.

Electrochemical Characterization of the Immunosensor
Valuable information involving the electrochemical mechanism can usually be obtained from the relationship between the peak current and scan rate. Here, the CVs of the modified immunosensors in 0.01M PBS (pH 7.4) at different scan rates are shown in Figure 3. It can be observed that both the anodic and cathodic peak currents were directly proportional to the scan rates in the range of 10-200 mV s −1 . Additionally, linear relationships with good correlation coefficients were observed between the peak currents and scan rates, suggesting that the electrochemical reaction on the modified electrode is an adsorption-controlled redox process. As EIS can effectively probe the electron transfer kinetics at the electrode surface, it was also used here to characterize the stepwise assembly of the immunosensor. The impedance spectra included a semicircle portion and a linear portion. At higher frequencies, the semicircle portion corresponded to the electron-transfer limited process, and at lower frequencies, the linear portion represented the diffusion-limited process. Figure 2B shows the EIS of the stepwise modification processes performed in 5.0 mM potassium ferricyanide. Compared with the bare SPE (curve a), a greatly lower resistance was obtained after GO/Fe 3 O 4 /PB was modified on the electrode, implying that GO/Fe 3 O 4 /PB nanocomposites have excellent electric conduction and could accelerate the electron transfer in to some degree (curve b). Furthermore, the GO/Fe 3 O 4 /PB @AuNPs showed a much lower resistance than the previous example, indicating that the electrodeposited AuNPs were highly beneficial to the electron transfer (curve c). In contrast, after HBsAb was immobilized on GO/Fe 3 O 4 /PB@AuNPs, the resistance obviously increased (curve d), suggesting that HBsAb was successfully immobilized on the surface and blocked the electron transport between the redox probe and electrode. Then the resistance increased again when BSA was immobilized onto the immunosensor (curve e). Finally, when HBsAg interacted with the proposed immunosensor, an increase of resistance was observed (curve f), suggesting the effective specific recognition between antibodies and antigens. This was consistent with the fact that the hydrophobic layer of protein insulates the conductive support and hinders the interfacial electron transfer. These results were in accordance with the CV measurements, demonstrating the successful construction of a GO/Fe 3 O 4 /PB-based immunosensor for HBsAg detection.
Valuable information involving the electrochemical mechanism can usually be obtained from the relationship between the peak current and scan rate. Here, the CVs of the modified immunosensors in 0.01M PBS (pH 7.4) at different scan rates are shown in Figure 3. It can be observed that both the anodic and cathodic peak currents were directly proportional to the scan rates in the range of 10-200 mV s −1 . Additionally, linear relationships with good correlation coefficients were observed between the peak currents and scan rates, suggesting that the electrochemical reaction on the modified electrode is an adsorption-controlled redox process. Biosensors 2020, 10, x FOR PEER REVIEW 6 of 12  Figure 4A-D shows the typical scanning electron microscope (SEM) images of the bare SPE, as well as the GO/Fe3O4, GO/Fe3O4/PB and GO/Fe3O4/PB @AuNPs modified SPEs. It can be seen from Figure 4A that the SPE shows a relatively smooth surface with a few small flack-like projections. Figure 4B shows regular, compact pellets combined with lamellar folds, indicating the successful immobilization of GO/Fe3O4, and the scraggy surface can provide a large surface area for loading nanoparticles. In Figure 4C, the complicated structure of the modified SPE was observed to contain dense fine particle, indicating the successful formation of the GO/Fe3O4/PB nanocomposite film. The SEM analysis ( Figure 4D) shows a good amount of granular matter, which indicated that a large number of AuNPs were immobilized onto the surface of the modified electrode. Because the structure gave a high specific surface area, it facilitates antibody attachment. Energy Dispersive Spectrometer (EDS) characterization was employed to further analyze the nanocomposites. As shown in Figure 5A, the fully scanned spectra demonstrated the existence of C   Figure 4A that the SPE shows a relatively smooth surface with a few small flack-like projections. Figure 4B shows regular, compact pellets combined with lamellar folds, indicating the successful immobilization of GO/Fe 3 O 4 , and the scraggy surface can provide a large surface area for loading nanoparticles. In Figure 4C, the complicated structure of the modified SPE was observed to contain dense fine particle, indicating the successful formation of the GO/Fe 3 O 4 /PB nanocomposite film. The SEM analysis ( Figure 4D) shows a good amount of granular matter, which indicated that a large number of AuNPs were immobilized onto the surface of the modified electrode. Because the structure gave a high specific surface area, it facilitates antibody attachment.

Characterization of the Modified SPEs
Biosensors 2020, 10, x FOR PEER REVIEW 6 of 12  Figure 4A-D shows the typical scanning electron microscope (SEM) images of the bare SPE, as well as the GO/Fe3O4, GO/Fe3O4/PB and GO/Fe3O4/PB @AuNPs modified SPEs. It can be seen from Figure 4A that the SPE shows a relatively smooth surface with a few small flack-like projections. Figure 4B shows regular, compact pellets combined with lamellar folds, indicating the successful immobilization of GO/Fe3O4, and the scraggy surface can provide a large surface area for loading nanoparticles. In Figure 4C, the complicated structure of the modified SPE was observed to contain dense fine particle, indicating the successful formation of the GO/Fe3O4/PB nanocomposite film. The SEM analysis ( Figure 4D) shows a good amount of granular matter, which indicated that a large number of AuNPs were immobilized onto the surface of the modified electrode. Because the structure gave a high specific surface area, it facilitates antibody attachment. Energy Dispersive Spectrometer (EDS) characterization was employed to further analyze the nanocomposites. As shown in Figure 5A, the fully scanned spectra demonstrated the existence of C Energy Dispersive Spectrometer (EDS) characterization was employed to further analyze the nanocomposites. As shown in Figure 5A, the fully scanned spectra demonstrated the existence of C and O elements in bare SPE. After the deposition of GO/Fe 3 O 4 nanocomposites, two sharp signals related to Fe at about 0.71 KeV and 6.39 KeV could be observed (curve b). In curve c, the signals of Cl and K appeared, which came from PB nanoparticles. As shown in Figure 5D, two signals appear in the image at about 2.15 KeV and 9.7 KeV, which are clearly indicating that AuNPs attached onto the surface of the GO/Fe 3 O 4 /PB nanocomposites.

Characterization of the Modified SPEs
Biosensors 2020, 10, x FOR PEER REVIEW 7 of 12 and O elements in bare SPE. After the deposition of GO/Fe3O4 nanocomposites, two sharp signals related to Fe at about 0.71 KeV and 6.39 KeV could be observed (curve b). In curve c, the signals of Cl and K appeared, which came from PB nanoparticles. As shown in Figure 5D, two signals appear in the image at about 2.15 KeV and 9.7 KeV, which are clearly indicating that AuNPs attached onto the surface of the GO/Fe3O4/PB nanocomposites.

Optimization of Experimental Condition
To provide an optimal electrochemical experimental environment, factors include including the incubation time and concentration of HBsAb, which may affect the performance of the immunosensor, should be optimized.
AuNPs have been widely applied in the construction of electrochemical biosensors, due to their excellent biocompatibility and conductivity. Therefore, it is essential to study the effect of AuNPs for the enhancement of charge transfer. The working potential of −1-0.1 V was selected for the electrodeposition of HAuCl4. As presented in Figure 6A, the peak current increased from −1 V to −0.5 V, achieving a maximum at −0.5 V, indicating that the optimal work potential of electrodeposition was −0.5 V.

Optimization of Experimental Condition
To provide an optimal electrochemical experimental environment, factors include including the incubation time and concentration of HBsAb, which may affect the performance of the immunosensor, should be optimized.
AuNPs have been widely applied in the construction of electrochemical biosensors, due to their excellent biocompatibility and conductivity. Therefore, it is essential to study the effect of AuNPs for the enhancement of charge transfer. The working potential of −1-0.1 V was selected for the electrodeposition of HAuCl 4 . As presented in Figure 6A, the peak current increased from −1 V to −0.5 V, achieving a maximum at −0.5 V, indicating that the optimal work potential of electrodeposition was −0.5 V.
Biosensors 2020, 10, x FOR PEER REVIEW 7 of 12 and O elements in bare SPE. After the deposition of GO/Fe3O4 nanocomposites, two sharp signals related to Fe at about 0.71 KeV and 6.39 KeV could be observed (curve b). In curve c, the signals of Cl and K appeared, which came from PB nanoparticles. As shown in Figure 5D, two signals appear in the image at about 2.15 KeV and 9.7 KeV, which are clearly indicating that AuNPs attached onto the surface of the GO/Fe3O4/PB nanocomposites.

Optimization of Experimental Condition
To provide an optimal electrochemical experimental environment, factors include including the incubation time and concentration of HBsAb, which may affect the performance of the immunosensor, should be optimized.
AuNPs have been widely applied in the construction of electrochemical biosensors, due to their excellent biocompatibility and conductivity. Therefore, it is essential to study the effect of AuNPs for the enhancement of charge transfer. The working potential of −1-0.1 V was selected for the electrodeposition of HAuCl4. As presented in Figure 6A, the peak current increased from −1 V to −0.5 V, achieving a maximum at −0.5 V, indicating that the optimal work potential of electrodeposition was −0.5 V.  The quantity of immobilized HBsAb is a crucial parameter in the construction of immunosensors. As shown in Figure 6B, the reduction peak current decreased with the increased HBsAb concentration, and the downtrend approximately leveled off when the concentration increased from 250 to 500 ng/mL. Therefore, 250 ng/mL was selected as the optimum incubation concentration for HBsAb in following experiments.
Incubation time also plays an important role in analyzing the performance of the immunosensor. In this experiment, the incubation time was investigated in the range of 20-60 min. As shown in Figure 6C, the results showed that the response current of the immunosensor rapidly decreased when the incubation time was increased from 20 to 40 min. When incubation time was extended to 60 min, the current response become steady. Therefore, 40 min was chosen as the best incubation time for all the immunoassay.

Selectivity, Stability and Repeatability of the Immunosensor
To characterize the specificity of the immunosensor, the effect of possible interferences that might impact the determination of target analytes was investigated. DPV responses of the proposed immunosensor to 100 ng/mL of HBsAg containing different interferences, such as HSA, AFP, CEA, TC and LC, were assayed. Compared with the current response obtained by HBsAg only, variations from the interferents were less than 7%. The result indicated that the interference can be neglected, and the immunosensor has good selectivity for HBsAg ( Figure 7A).
Biosensors 2020, 10, x FOR PEER REVIEW 8 of 12 The quantity of immobilized HBsAb is a crucial parameter in the construction of immunosensors. As shown in Figure 6B, the reduction peak current decreased with the increased HBsAb concentration, and the downtrend approximately leveled off when the concentration increased from 250 to 500 ng/mL. Therefore, 250 ng/mL was selected as the optimum incubation concentration for HBsAb in following experiments.
Incubation time also plays an important role in analyzing the performance of the immunosensor. In this experiment, the incubation time was investigated in the range of 20-60 min. As shown in Figure 6C, the results showed that the response current of the immunosensor rapidly decreased when the incubation time was increased from 20 to 40 min. When incubation time was extended to 60 min, the current response become steady. Therefore, 40 min was chosen as the best incubation time for all the immunoassay.

Selectivity, Stability and Repeatability of the Immunosensor
To characterize the specificity of the immunosensor, the effect of possible interferences that might impact the determination of target analytes was investigated. DPV responses of the proposed immunosensor to 100 ng/mL of HBsAg containing different interferences, such as HSA, AFP, CEA, TC and LC, were assayed. Compared with the current response obtained by HBsAg only, variations from the interferents were less than 7%. The result indicated that the interference can be neglected, and the immunosensor has good selectivity for HBsAg ( Figure 7A). The stability of the immunosensor is a crucial factor in actual application and storage. The longtime stability of the immunosensor was studied by keeping the fabricated immunosensor in a refrigerator at 4 °C for 30 days when not in use. The peak current of the immunosensor was measured every five days in the first half of the month, and the current response still retained about 93% of the initial peak current, demonstrating that the immunosensor had good stability ( Figure 7B).
Besides, the repeatability of the proposed immunosensor was studied. Five different immunosensors modified with the same procedures were evaluated with 100 ng/mL of HBsAg. The relative standard deviation (RSD) of the inter-assay was 2.8%, indicating that the immunosensor possessed good reproducibility.

Analytical Performance
Under optimized detection conditions, the DPV responses of the immunosensor to different concentrations of HBsAg were obtained. Figure 8A shows that the peak currents of DPV decreased The stability of the immunosensor is a crucial factor in actual application and storage. The long-time stability of the immunosensor was studied by keeping the fabricated immunosensor in a refrigerator at 4 • C for 30 days when not in use. The peak current of the immunosensor was measured every five days in the first half of the month, and the current response still retained about 93% of the initial peak current, demonstrating that the immunosensor had good stability ( Figure 7B).
Besides, the repeatability of the proposed immunosensor was studied. Five different immunosensors modified with the same procedures were evaluated with 100 ng/mL of HBsAg. The relative standard deviation (RSD) of the inter-assay was 2.8%, indicating that the immunosensor possessed good reproducibility.

Analytical Performance
Under optimized detection conditions, the DPV responses of the immunosensor to different concentrations of HBsAg were obtained. Figure 8A shows that the peak currents of DPV decreased with an increased HBsAg concentration. The reason for this was that the formed immunocomplex on the electrode surface acted as an inert block layer, which hindered the electron transfer toward the electrode surface. As shown in Figure 8B, a linear relationship between the peak currents and the logarithmic values of HBsAg concentration was obtained in the range of 0.5 pg/mL to 200 ng/mL. The regression equation was y = 62.32−8.54x, with a correlation coefficient of 0.9843 and detection limit of 0.166 pg/mL (S/N = 3). The analytical performance of the immunoassay has been compared with the performances of other HBsAg immunoassays reported (Table 1), and the proposed immunosensor showed a widely linear range and a low detection limit.
Biosensors 2020, 10, x FOR PEER REVIEW 9 of 12 with an increased HBsAg concentration. The reason for this was that the formed immunocomplex on the electrode surface acted as an inert block layer, which hindered the electron transfer toward the electrode surface. As shown in Figure 8B, a linear relationship between the peak currents and the logarithmic values of HBsAg concentration was obtained in the range of 0.5 pg/mL to 200 ng/mL. The regression equation was y = 62.32−8.54x, with a correlation coefficient of 0.9843 and detection limit of 0.166 pg/mL (S/N = 3). The analytical performance of the immunoassay has been compared with the performances of other HBsAg immunoassays reported (Table 1), and the proposed immunosensor showed a widely linear range and a low detection limit.

Analysis of Real Samples
In order to investigate the reliability and accuracy of the label-free electrochemical immunosensor, five human serum samples were measured. The content of HBsAg in the serum samples was detected by the proposed immunosensor according to the relationship between the current response and HBsAg concentration. The obtained results were compared with those obtained by ECLIA, which was provided by the affiliated hospital of Guilin Medical College. As shown in Table 2, the relative errors between the two methods ranged from −2.83% to 14%, indicating that the fabricated immunosensor was suitable for real sample analysis. Therefore, the proposed immunosensor could be effectively applied in the quantitative detection of HBsAg in human serums and would have potential application in clinical diagnostics.

Analysis of Real Samples
In order to investigate the reliability and accuracy of the label-free electrochemical immunosensor, five human serum samples were measured. The content of HBsAg in the serum samples was detected by the proposed immunosensor according to the relationship between the current response and HBsAg concentration. The obtained results were compared with those obtained by ECLIA, which was provided by the affiliated hospital of Guilin Medical College. As shown in Table 2, the relative errors between the two methods ranged from −2.83% to 14%, indicating that the fabricated immunosensor was suitable for real sample analysis. Therefore, the proposed immunosensor could be effectively applied in the quantitative detection of HBsAg in human serums and would have potential application in clinical diagnostics.

Conclusions
A novel, simple and label-free electrochemical immunosensor was developed for selective and sensitive detection of HBsAg. Thus, GO/Fe 3 O 4 /PB nanocomposites coated on SPE not only served as substrate materials for promoting electron transfer, but also acted as the electrochemical redox mediator. AuNPs that attached onto the modified electrode were used for HBsAb adsorption and further signal amplification. The proposed immunosensor showed excellent performance in the detection of HBsAg with a wide linear range, low detection limit, good biocompatibility, good selectivity and long-term stability. In summary, an ultrasensitive electrochemical immunosensor was developed for the detection of HBsAg. The simple and cost-effective sensing strategy provides a new promising platform for the design of a highly sensitive detection method, showing potential application for clinical immunoassays.
Author Contributions: Conceptualization, methodology, writing-original draft preparation and validation, S.W.; formal analysis, investigation, H.X.; resources, data curation and writing-review and editing, L.C.; project administration and funding acquisition, Z.C. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.