Detection of Alpha Fetoprotein Based on AIEgen Nanosphere Labeled Aptamer Combined with Sandwich Structure of Magnetic Gold Nanocomposites

As a biomarker, alpha-fetoprotein (AFP) is valuable for detecting some tumors in men, non-pregnant women, and children. However, the detection sensitivity in some methods needs to be improved. Therefore, developing a simple, reliable, and sensitive detection method for AFP is important for non-malignant diseases. An aptamer binding was developed based on aggregation-induced emission luminogen (AIEgen) nanosphere labeled with Fe3O4@MPTMS@AuNPs. AFP was detected with a sandwich structure of AuNPs magnetic composite particles. An aggregation-induced emission (AIE) molecule and polystyrene (PS) nanosphere complex were assembled, enhancing the fluorescence and improving the sensitivity of detection. The limit of detection (LOD) was at a given level of 1.429 pg/mL, which can best be achieved in serum samples. Finally, the results obtained showed the complex to be promising in practical applications.


Introduction
Alpha-fetoprotein (AFP), the most commonly used protein biomarker, has attracted great attention recently. As a biomarker, AFP is valuable for detecting some tumors in men, non-pregnant women, and children. In addition, in high-risk but asymptomatic populations, AFP detection indicators are vital for identifying early curable tumors and reducing disease-related mortality in a cost-effective manner. The development of AFP detection technology with high sensitivity is necessary for the early detection of some cancers and the clinical detection of AFP [1][2][3][4]. To this end, a variety of methods for detecting AFP have been developed, such as enzyme-linked immunosorbent assay [5], radioimmunoassay [6], fluorescence immunoassay [7], electrochemiluminescence [8], Raman spectroscopy [9], electrochemical immunosensing [10], and so on. However, they possess the shortcomings of non-specific binding and are unsuitable for high-throughput analysis [11]. Antibodies are associated with poor reproducibility and instability, which limits their broad application. For example, some small molecules cannot react and antibodies are unstable in extreme environments. Furthermore, immunoassays can produce false-positive or false-negative results [12]. An aptamer is a single-stranded RNA or DNA molecule selected in vitro from the nucleic acid molecular library by systematic evolution of ligands by exponential enrichment (SELEX) to specifically combine targets with high affinity (nucleic acids, small molecules, proteins, etc.) [13]. Aptamers are flexible, repeatable, easy to fix and regenerate, and show no difference between batches, and have been widely used in the sensor field [14]. Since aptamers exhibit many beneficial properties, aptamer-based biosensor systems have been developed to analyze various classes of detectors. Therefore, developing produce TEM images at 200 KV. A vacuum freeze-dryer (FD-1A-50) was used to freeze-dry the samples at a given temperature of 53 • C with a vacuum at 20 Pa. Fourier transform infrared spectroscopy (FT-IR) was obtained from Perkin Elmer Inc using KBr to treat the sample. A PHS-25 digital viscometer was used to adjust the pH of the solution. The data obtained in the experiment were processed using Origin 8.0.

Preparation of Fe 3 O 4 @MPTMS@AuNPs
The AuNPs in this experiment were prepared using the classical Frens method [26]; the particle size was about 12 nm. A sol-gel method was used to synthesize Fe 3 O 4 @MPTMS@AuNPs according to Mohebbi et al. [27]. The synthesized gold nanoparticles were stored at a standby temperature of 4 • C. MPTMS, ultrapure water, pure ethanol, and hydrochloric acid were thoroughly mixed with a magnetic stirrer at a molar concentration ratio of 4:20:50:0.1 for 3 h. The solution was then placed in the chamber for 24 h and ultrasonically mixed with Fe 3 O 4 in a volume ratio of 1:1, 2:1 and 4:1 for 30 min, washed many times, and lyophilized to obtain the solid Fe 3 O 4 @MPTMS. The sulfhydryl (-SH) was the surface of Fe 3 O 4 @MPTMS. This can further bind to Au NPs. After mixing the prepared gold nanoparticles with ethanol for 5 min, 0.1 g Fe 3 O 4 @MPTMS was added and stirred for 30 min. After washing and drying 3 times, Fe 3 O 4 @MPTMS@AuNPs were prepared into a solution of 2 mg/mL and homogenized by sonication for 5 min.

Preparation of AIEgen
An amount of 100 mg polystyrene nanospheres was dissolved in 10 mL of ultrapure water (containing 0.25% SDS) to obtain a uniformly dispersed solution with ultrasound. A quantity of 0.5 mL of DSAI (mass fraction 10%) was dissolved in CH 2 Cl 2 solution. After sonication for 1 min, it was stirred at 40 • C, and CH 2 Cl 2 evaporated by rotation. The obtained AIEgen nanospheres were washed with ultrapure water 3 times to remove SDS, centrifuged at 8000 rpm for 10 min and washed with ethanol 3 times until fluorescence was not observed in the supernatant. Finally, the solid was dried using a freeze-dryer. A certain amount of solid was dissolved in ultrapure water to prepare a solution with a final concentration of 2 mg/mL and ultrasonic homogenization was applied [28].

Preparation of Carboxyl Functionalized AIEgen Nanospheres
Amounts of 0.6 g NaOH and 0.5 g chloroacetic acid were added to 5 mL of AIEgene nanosphere solution with a size of 50 nm, respectively. The solution was sonicated in a water bath for 2 h, then neutralized by NaOH and further purified by centrifugation at 8000 rpm for 10 min and washed 3 times. The solution was dried to obtain a carboxylfunctionalized AIEgene nanosphere.

Preparation of AIEgen Nanosphere Labeled Aptamer (AIEgen Aptamer)
AIEgene nanospheres were weighed and prepared as a solution with a 2 mg/mL final concentration. The solution was reacted with 30 µL (50 nM) aptamer 2 with amino (-NH 2 ) functional group at 4 • C for 12 h and was subsequently centrifuged at 8000 rpm for 10 min to remove the excess aptamer 2 and dispersed in the buffer.

Sensor Fabrication Processing
An aptamer that was based on AIEgen nanosphere labelling Fe 3 O 4 @MPTMS @AuNPs was developed in this study. AFP was detected using a sandwich-structure aptamer fluorescence sensor constructed with a combination of AuNPs and magnetic composites ( Figure 1). Aptamer 1, with a final concentration of 15 nM, was mixed with 10 µg/mL Fe 3 O 4 @MPTMS@AuNPs. The mixture was incubated for 12 h in PBS (10 mM, pH 7.4) buffer with a total concentration of 500 µL and adsorbed with a magnet. Different concentrations of AFP were subsequently added and incubated at room temperature for 30 min. Consequently, AFP was explicitly bound to aptamer 1 and remained on the surface of Fe 3 O 4 @MPTMS @AuNPs, which prompted the addition of 8 µg/mL AIEgen aptamer 2, incubated at room temperature for 30 min. Fe3O4@MPTMS@AuNPs. The mixture was incubated for 12 h in PBS (10 mM, pH 7.4) buffer with a total concentration of 500 μL and adsorbed with a magnet. Different concentrations of AFP were subsequently added and incubated at room temperature for 30 min. Consequently, AFP was explicitly bound to aptamer 1 and remained on the surface of Fe3O4@MPTMS @AuNPs, which prompted the addition of 8 μg/mL AIEgen aptamer 2, incubated at room temperature for 30 min.

Sensor Detection Principle
The chemical binding between sulfhydryl (-SH) at the end of aptamer 1 and Fe3O4@MPTMS@AuNPs occurred on the surface of AuNPs with aptamer 2. The unbound AIEgen aptamer 2 was removed by a magnet. At this juncture, aptamer 1, AFP, and AIEgen aptamer 2 constituted a sandwich structure. AIE molecules and PS nanospheres were assembled into a complex state. The spatial structure inside polymer nanoparticles can limit the rotation of AIE molecules. AIEgen was assembled into nanoparticles that can emit fluorescence. If AFP does not remain on the surface of Fe3O4@MPTMS @AuNPs, AIEgen aptamer 2 should be quenched by Fe3O4@MPTMS @AuNPs. AIEgen aptamer 2 was far from the surface of Fe3O4@MPTMS@AuNPs based on FRET between Fe3O4@MPTMS@AuNPs and AIEgen aptamer 2 after adding AFP. Then an AIEgen fluorescent ball emitted the fluorescence. After the same elution, the fluorescence intensity was measured at this time. The measured values of the reaction system using the multifunctional enzyme-labelling instrument were processed by Origin 8.0, and the value of F/F0-1 was obtained. The value of F/F0-1 is the ratio of fluorescence intensity, where F0 and F, respectively, represent the fluorescence intensity of the fluorescence sensing system at 550 nm before and after the addition of AFP.

Characterization of Fe3O4@MPTMS @ AuNPs and AIEgen Nanospheres
In the process of combining Fe3O4 with MPTMS, if the amount of MPTMS is too low, the effective binding rate of Fe3O4 with AuNPs will be reduced; if the amount of MPTMS is too high, it will make the MPTMS in the oil state difficult to clean. Therefore, we optimized the combination ratio of Fe3O4 and MPTMS based on literature reports. It can be intuitively observed from Figure 2 that the AuNPs nanoparticles were bound to the surface of Fe3O4 through the action of MPTMS. However, it can be seen from Figure 2A that some AuNPs in the composite system were not fully combined on the surface of magnetic nanoparticles. Figure 2C shows that AuNPs were unevenly distributed on the surface of Fe3O4 particles. In Figure 2B, AuNPs were distributed on the surface of each magnetic nanosphere. Therefore, a volume ratio of Fe3O4 to MPTMS of 1:2 was chosen to prepare

Sensor Detection Principle
The chemical binding between sulfhydryl (-SH) at the end of aptamer 1 and Fe 3 O 4 @MPTMS@AuNPs occurred on the surface of AuNPs with aptamer 2. The unbound AIEgen aptamer 2 was removed by a magnet. At this juncture, aptamer 1, AFP, and AIEgen aptamer 2 constituted a sandwich structure. AIE molecules and PS nanospheres were assembled into a complex state. The spatial structure inside polymer nanoparticles can limit the rotation of AIE molecules. AIEgen was assembled into nanoparticles that can emit fluorescence. If AFP does not remain on the surface of Fe 3 O 4 @MPTMS @AuNPs, AIEgen aptamer 2 should be quenched by Fe 3 O 4 @MPTMS @AuNPs. AIEgen aptamer 2 was far from the surface of Fe 3 O 4 @MPTMS@AuNPs based on FRET between Fe 3 O 4 @MPTMS@AuNPs and AIEgen aptamer 2 after adding AFP. Then an AIEgen fluorescent ball emitted the fluorescence. After the same elution, the fluorescence intensity was measured at this time.
The measured values of the reaction system using the multifunctional enzyme-labelling instrument were processed by Origin 8.0, and the value of F/F 0 -1 was obtained. The value of F/F 0 -1 is the ratio of fluorescence intensity, where F 0 and F, respectively, represent the fluorescence intensity of the fluorescence sensing system at 550 nm before and after the addition of AFP.

Characterization of Fe 3 O 4 @MPTMS @ AuNPs and AIEgen Nanospheres
In the process of combining Fe 3 O 4 with MPTMS, if the amount of MPTMS is too low, the effective binding rate of Fe 3 O 4 with AuNPs will be reduced; if the amount of MPTMS is too high, it will make the MPTMS in the oil state difficult to clean. Therefore, we optimized the combination ratio of Fe 3 O 4 and MPTMS based on literature reports. It can be intuitively observed from Figure 2 that the AuNPs nanoparticles were bound to the surface of Fe 3 O 4 through the action of MPTMS. However, it can be seen from Figure 2A that some AuNPs in the composite system were not fully combined on the surface of magnetic nanoparticles. Figure 2C shows that AuNPs were unevenly distributed on the surface of Fe 3 O 4 particles. In Figure 2B, AuNPs were distributed on the surface of each magnetic nanosphere. Therefore, a volume ratio of Fe 3 O 4 to MPTMS of 1:2 was chosen to prepare subsequent composite nanomaterials. Figure 2C shows Fe 3 O 4 @MPTMS @AuNPs. A comparative image of the AuNPs solution and the solution after the action of the magnet shows that the magnet was able to completely adsorb the composite nanomaterials on the bottle wall and make the solution clear.  Figure 2C shows Fe3O4@MPTMS @AuNPs. A co parative image of the AuNPs solution and the solution after the action of the mag shows that the magnet was able to completely adsorb the composite nanomaterials on bottle wall and make the solution clear.  Figure 3A,B show the TEM images of PS nanospheres and AIEgen nanospheres, spectively. Figure 3C compares the fluorescence spectra of AIEgen (DSAI) and AIE PS assembled into polymers. The fluorescence value was enhanced when AIEgen was sembled into PS nanospheres. Figure 3D shows the FT-IR spectrum of carboxyl functi alization of AIEgen. The peak at 1606 cm −1 represented the formation of carboxyl fu tional groups, and there was a strong, wide peak at 1073 cm −1 .   Figure 3A,B show the TEM images of PS nanospheres and AIEgen nanospheres, respectively. Figure 3C compares the fluorescence spectra of AIEgen (DSAI) and AIE and PS assembled into polymers. The fluorescence value was enhanced when AIEgen was assembled into PS nanospheres. Figure 3D shows the FT-IR spectrum of carboxyl functionalization of AIEgen. The peak at 1606 cm −1 represented the formation of carboxyl functional groups, and there was a strong, wide peak at 1073 cm −1 . subsequent composite nanomaterials. Figure 2C shows Fe3O4@MPTMS @AuNPs. A comparative image of the AuNPs solution and the solution after the action of the magnet shows that the magnet was able to completely adsorb the composite nanomaterials on the bottle wall and make the solution clear.  Figure 3A,B show the TEM images of PS nanospheres and AIEgen nanospheres, respectively. Figure 3C compares the fluorescence spectra of AIEgen (DSAI) and AIE and PS assembled into polymers. The fluorescence value was enhanced when AIEgen was assembled into PS nanospheres. Figure 3D shows the FT-IR spectrum of carboxyl functionalization of AIEgen. The peak at 1606 cm −1 represented the formation of carboxyl functional groups, and there was a strong, wide peak at 1073 cm −1 .

Optimization of Experimental Conditions
The experimental conditions were optimized in this study to achieve the best detection effect of AFP on the sensing platform, including the concentrations of Fe 3 O 4 @MPTMS@ AuNPs, aptamer 1, and AIEgen aptamer 2.

Optimization of Fe 3 O 4 @MPTMS @ AuNPs Concentration
The influence of AuNPs on the reaction system was optimized. Different concentrations of Fe 3 O 4 @MPTMS@AuNPs (6, 8, 10, 12 and 14 µg/mL) were added to the buffer system containing 15 nM aptamer 1. The fluorescence intensity of the system was measured before and after adding AFP, and the change in fluorescence intensity at 550 nm was compared. The results are shown in Figure 4. The results showed that F/F 0 -1 gradually increased with the low concentration of Fe 3 O 4 @MPTMS@AuNPs nanocomposites. When its concentration was 10 µg/mL, F/F 0 -1 reached a maximum value. When the concentration of nanomaterials continued to increase, F/F 0 -1 gradually decreased. Therefore, the study used 10 µg/mL Fe 3 O 4 @MPTMS@AuNPs complex as the optimal concentration.

Optimization of Experimental Conditions
The experimental conditions were optimized in this study to achieve the best detection effect of AFP on the sensing platform, including the concentrations of Fe3O4@MPTMS@ AuNPs, aptamer 1, and AIEgen aptamer 2.

Optimization of Fe3O4@MPTMS @ AuNPs Concentration
The influence of AuNPs on the reaction system was optimized. Different concentrations of Fe3O4@MPTMS@AuNPs (6, 8, 10, 12 and 14 µg/mL) were added to the buffer system containing 15 nM aptamer 1. The fluorescence intensity of the system was measured before and after adding AFP, and the change in fluorescence intensity at 550 nm was compared. The results are shown in Figure 4. The results showed that F/F0-1 gradually increased with the low concentration of Fe3O4@MPTMS@AuNPs nanocomposites. When its concentration was 10 µg/mL, F/F0-1 reached a maximum value. When the concentration of nanomaterials continued to increase, F/F0-1 gradually decreased. Therefore, the study used 10 µg/mL Fe3O4@MPTMS@AuNPs complex as the optimal concentration.

Optimization of Aptamer 1 Concentration
The concentration of aptamer 1 was optimized. Amounts of 5, 10, 15, 20, and 25 nM aptamer 1 were added to the reaction system. Very low concentrations of aptamer 1 in the system affected Fe3O4@MPTMS@AuNPs binding and immobilization for AFP. However, it was not necessary at very high concentrations of aptamer 1. Figure 5 shows the highest F/F0-1 caused by aptamer 1 with a concentration of 15 nM under the same conditions. Consequently, aptamer 1, with a concentration of 15 nM, was chosen as the optimal concentration in this study.

Optimization of Aptamer 1 Concentration
The concentration of aptamer 1 was optimized. Amounts of 5, 10, 15, 20, and 25 nM aptamer 1 were added to the reaction system. Very low concentrations of aptamer 1 in the system affected Fe 3 O 4 @MPTMS@AuNPs binding and immobilization for AFP. However, it was not necessary at very high concentrations of aptamer 1. Figure 5 shows the highest F/F 0 -1 caused by aptamer 1 with a concentration of 15 nM under the same conditions. Consequently, aptamer 1, with a concentration of 15 nM, was chosen as the optimal concentration in this study.

Optimization of AIEgen Aptamer 2 Concentration
The influence of AIEgen aptamer 2 on the fluorescence sensor system was further explored and optimized. The sandwich structure system added different concentrations (4, 6, 8, 10, and 12 µg/mL) of AIEgen aptamer 2. Figure 6 shows that, with gradual increase in AIEgen aptamer 2 concentration, F/F0-1 gradually increased. When the AIEgen aptamer

Optimization of AIEgen Aptamer 2 Concentration
The influence of AIEgen aptamer 2 on the fluorescence sensor system was further explored and optimized. The sandwich structure system added different concentrations (4,6,8,10, and 12 µg/mL) of AIEgen aptamer 2. Figure 6 shows that, with gradual increase in AIEgen aptamer 2 concentration, F/F 0 -1 gradually increased. When the AIEgen aptamer 2 concentration increased to 8 µg/mL, F/F 0 -1 reached a maximum value. Moreover, the subsequent fluorescence change value gradually decreased with increase in AIEgen aptamer 2 concentration. However, if the concentration of aptamer 2 was more than 8 µg/mL, it was unnecessary. A higher concentration of aptamer 2 was able to increase the intensity of the fluorescence background. Nevertheless, the effect of detection was not improved. F/F 0 -1 did not increase with a higher concentration of aptamer 2, prompting the selection of 8 µg/mL as the optimal concentration.

Optimization of AIEgen Aptamer 2 Concentration
The influence of AIEgen aptamer 2 on the fluorescence sensor system was further explored and optimized. The sandwich structure system added different concentrations (4, 6, 8, 10, and 12 µg/mL) of AIEgen aptamer 2. Figure 6 shows that, with gradual increase in AIEgen aptamer 2 concentration, F/F0-1 gradually increased. When the AIEgen aptamer 2 concentration increased to 8 µg/mL, F/F0-1 reached a maximum value. Moreover, the subsequent fluorescence change value gradually decreased with increase in AIEgen aptamer 2 concentration. However, if the concentration of aptamer 2 was more than 8 µg/mL, it was unnecessary. A higher concentration of aptamer 2 was able to increase the intensity of the fluorescence background. Nevertheless, the effect of detection was not improved. F/F0-1 did not increase with a higher concentration of aptamer 2, prompting the selection of 8 µg/mL as the optimal concentration.

Sensitivity Test
Different concentrations of AFP (0.005, 0.01, 0.08, 0.1, 0.5, 0.8, 5, 8, and 10 ng/mL) were added to the solution containing 10 µg/mL Fe 3 O 4 @MPTMS in the sensing system of AuNPs; amounts of 15 nM aptamer 1 and 8 µg/mL AIEgen aptamer 2 were mixed evenly and eluted after incubation for 30 min at room temperature. The changes in fluorescence were recorded and calculated under different concentrations of AFP. The results are shown in Figure 7. Higher concentration of AFP can bind to more aptamer with DSAI. Therefore, the fluorescence intensity value gradually increased with increase in AFP concentration. The AFP concentration was in the range of 0.005-0.1 ng/mL as shown in Figure 7B; F/F 0 -1 exhibited an apparent linear relationship with AFP concentration; the linear regression equation was y = 3.082 × C[AFP] + 1.148, R 2 = 0.99. As the AFP concentration increased, the fluorescence intensity increased relatively slowly. We calculated LOD based on signal-tonoise (LOD = 3S/N), where S is the standard deviation of the eleven blank measurements (without AFP), and N is the slope of the fluorescence intensity of DSAI relative to the AFP concentration. Based on 3S/N, the detection limit was 1.429 pg/mL. As shown in Table 1, this sensor had a lower LOD and an excellent linear range. It was better than some reported sensors. The limit of quantification (LOQ, ten times of standard deviation of the blank signal/slope) was 4.873 pg/mL. increased, the fluorescence intensity increased relatively slowly. We calculated LOD based on signal-to-noise (LOD = 3S/N), where S is the standard deviation of the eleven blank measurements (without AFP), and N is the slope of the fluorescence intensity of DSAI relative to the AFP concentration. Based on 3S/N, the detection limit was 1.429 pg/mL. As shown in Table 1, this sensor had a lower LOD and an excellent linear range. It was better than some reported sensors. The limit of quantification (LOQ, ten times of standard deviation of the blank signal/slope) was 4.873 pg/mL.

Selective Detection
Several substances similar to AFP, such as BSA, CEA, HSA, IgG, and thrombin, were introduced into the sensing system for interference experiments. Under the optimal conditions, AFP and other analogues with the same concentration (0.5 ng/mL) were added. The corresponding fluorescence intensity value and F/F0-1 were recorded. The result is

Selective Detection
Several substances similar to AFP, such as BSA, CEA, HSA, IgG, and thrombin, were introduced into the sensing system for interference experiments. Under the optimal conditions, AFP and other analogues with the same concentration (0.5 ng/mL) were added. The corresponding fluorescence intensity value and F/F 0 -1 were recorded. The result is shown in Figure 8. The same substance concentration was added under the same conditions. The fluorescence intensity of the sensor could be distinguished from others. This showed that the AFP detection using the sensor had good specificity. shown in Figure 8. The same substance concentration was added under the same conditions. The fluorescence intensity of the sensor could be distinguished from others. This showed that the AFP detection using the sensor had good specificity.

Serum Sample Analysis
To prove the clinical stability of the sensor, we replaced the reaction system with a sensing platform based on serum. The serum was bought from Tianhang Biotechnology Co., Ltd. (Huzhou, China). The obtained information from AFP added in serum was de-

Serum Sample Analysis
To prove the clinical stability of the sensor, we replaced the reaction system with a sensing platform based on serum. The serum was bought from Tianhang Biotechnology Co., Ltd. (Huzhou, China). The obtained information from AFP added in serum was detected using the biosensors. Three different concentrations of AFP (0.01 ng/mL, 0.08 ng/mL and 0.1 ng/mL) were selected in the linear range for standard recovery testing in normal serum. Three repeated experiments were carried out in each group. As shown in Table 2, the recoveries of the three groups of samples were 96.69%, 103.18%, and 105.224%, respectively. After calculation, the relative standard deviation was 1.856%-4.484%, which met the practical application requirements. A series of measurements from the batch resulted in a relative standard deviation (RSD) of 4.484%, demonstrating that the sensor results were repeatable and reproducible.

Conclusions
In this study, an aptamer based on AIEgen nanosphere labeled with Fe 3 O 4 @MPTMS@AuNPs was developed. The detection of alpha-fetoprotein by the sandwich structure of AuNPs magnetic composite particles was carried out. The unbound AIEgen aptamer 2 was removed by a magnet. AuNPs with Fe 3 O 4 formed a core-shell composite structure, which provided a functional platform for magnetic nanomaterials and increased the binding with aptamers. At the same time, the magnetic Fe 3 O 4 was obviously separated from the buffer system by the action of the magnet, which enabled a straightforward experimental operation. Therefore, this can specifically detect AFP. The AIE molecules and PS nanospheres were assembled into a complex. The spatial structure inside the polymer nanoparticles limited the rotation of the AIE molecules. AIEgen was assembled into nanoparticles and emitted fluorescence after aptamer 2 was bound to AFP, which caused aptamer 2 to be far from the surface of Fe 3 O 4 @MPTMS@AuNPs based on FRET because Fe 3 O 4 @MPTMS@AuNPs was able to quench the fluorescence of AIEgen. This enhanced the fluorescence intensity and improved the sensitivity of detection. Moreover, the detection limit using this method was 1.429 pg/mL, which was lower than LOD for most reported methods.

Conflicts of Interest:
The authors declare no conflict of interest.