A Au nanoparticle and polydopamine co-modified biosensor: A strategy for in situ and label-free surface plasmon resonance immunoassays

2.1 Reagents and instruments

Biotinylated goat anti-human IgG and hIgG were received from Beijing Biosynthesis Biotechnology Co., ltd. (China). Bovine serum albumin (BSA) was purchased from Beijing Solarbio Science & Technology Co., ltd. (China). Ovalbumin (OVA) was purchased from Shanghai Macklin Biochemical Co., ltd. (China). Dopamine hydrochloride used for PDA polymerization was purchased from Thermo Fisher Scientific (United States). Streptavidin (SA) and poly(allylamine hydrochloride) (PAH) were received from Sigma–Aldrich Co. (USA). Diiodomethane (CH₂I₂, n = 1.74) was received from J&K Scientific ltd. (China). Phosphate buffered saline (PBS, 10 mmol/L, pH = 7.2) and Tris-HCl buffer solution (10 mmol/L, pH = 8.5) were prepared with deionized water. OVA, BSA, goat anti-human IgG, and hIgG were dissolved in PBS (10 mmol/L, pH = 7.2). All the other reagents were of analytical grade and purchased from Beijing Chemical Works (China). A 0.5 mm thick quartz wafer was used as the substrate.

A 2WAJ Abbé refractometer (Shanghai CSOIF Co., ltd., China) was used to measure the refractive indices of NaCl aqueous solutions with different mass concentrations. The particle size distribution, TEM image and absorption spectra of the AuNPs solution before and after deposition on the surface of Au/PAH film were obtained on a Zetasizer Nano-ZS nano particle analyzer (Malvern Panalytical ltd., UK), a Tecnai G2 F30 transmission electron microscope (FEI, USA) and a BioMATE 3S UV–Vis spectrophotometer (Thermo Fisher Scientific Inc., USA), respectively. A scanning probe microscope (SPM) (SOLVER NEXT, NT-MDT Co., Russia) was used for characterization of the surface morphologies and topographies of biosensor chips, and 1 μm × 1 μm regions of the chips were scanned using tapping mode.

2.2 Construction of the SPR analyzer

A wavelength-modulated SPR device can provide a real-time curve of the wavelength-reflected light intensity relationship, and it can also meet the demand of miniaturization, imaging, multiple channels, multiparameter measurements, etc. (Liu et al., 2005). Fig. 1A displays the schematic diagram SPR analyzer used in this paper. The wavelength-modulated SPR analyzer works with a Kretschmann configuration to achieve the resonant condition by attenuated total reflection (ATR) (Juárez et al., 2006; Nizamov and Mirsky, 2011) in a triangular prism (45°/45°/90°, Scott N-SF6 glass, n = 1.799 at 632.8 nm). The SPR chip was placed on the prism, and a sample chamber was tightly attached to the middle of the SPR chip. To increase the coupling efficiency, diiodomethane (n = 1.74) was introduced between the prisms and the SPR chip as an index-matching liquid. Ultimately, the sample chamber, SPR chip and prism were sandwiched, and the three were placed on a precision rotating stage to adjust the optical path. The broadband light generated by a tungsten-halogen lamp (LS-1, Ocean Optics, USA) successively passed through a multimode quartz fiber, collimator and linear polarizer and then became p-polarized parallel light. Afterward, p-polarized parallel light entered the prism at an angle (θ = 13°) with respect to the prism surface normal, and the evanescent field caused by ATR penetrated through the glass/metal interface to excite the surface plasmon mode at the metal/medium interface of the SPR chip at a specific wavelength, which was named the resonance wavelength (λ_R). The total reflection beam output from another side of the prism was transmitted to a spectrometer (USB 2000+, Ocean Optics, USA) through another collimator and another quartz fiber, and the spectra were recorded for the measurement. A photograph of the SPR analyzer is shown in Fig. 1B.

Close

Fig. 1

Schematic diagram (A) and photograph (B) of the wavelength-modulated SPR analyzer.

Export to PPT

2.3 Fabrication of the SPR biosensor chip

Compared with the traditional Frens trisodium citrate reduction method (Frens, 1973), the microwave method for the synthesis of AuNPs is simple and efficient and can heat the material uniformly, evenly disperse the nanoparticles, control the particle size, and shorten the nucleation time (Xu et al., 2019). First, 100 mL of 0.01 % chloroauric acid aqueous solution was placed in a microwave oven and heated for 3 min at a power of 700 W to boil the solution. Subsequently, 3 mL of 1 % trisodium citrate aqueous solution was quickly added to the boiling solution and heated for 3 min at a power of 400 W, and the solution changed from colorless to deep red under boiling. After heating, the volume of solution was adjusted to 100 mL with distilled water to maintain the concentration of AuNPs.

The Au film SPR chips were prepared by successive sputtering of a 5 nm chromium layer and a 45 nm gold film on a 0.5 mm thick quartz wafer with area 15 mm × 15 mm, and then the SPR chip was washed with distilled water and ethanol to clean the Au film for the next modification. The method of nanoparticle film deposition followed the steps depicted in Fig. 2 as follows: the Au film SPR chip was immersed in 2 mg/mL positively charged polymer PAH for 5 min and then washed and dried (Bubniene et al., 2014). Subsequently, the chip was immersed into a solution containing negatively charged AuNP nanoparticles for 5 min; after deposition of the AuNPs, the Au/PAH/AuNP film SPR chip was washed and dried. In the process of biological modification, the SPR biosensor chip was then modified with a biological crosslinked PDA membrane by the surface functionalization method. The Au/PAH/AuNP film SPR chip obtained was immersed in a solution of dopamine (2 mg/mL of 10 mmol/L Tris buffer, pH = 8.5) for 10 min at room temperature, and dopamine was oxidatively self-polymerized to form a certain thickness of PDA membrane and coated the surface of the SPR chip. Then, the Au/PAH/AuNP/PDA film SPR chip was rinsed with deionized water to remove the large PDA particles. Next, 50 μg/mL SA solution was added to the surface of the Au/PAH/AuNP/PDA film of the SPR chip for 5 min, and SA was immobilized by covalent binding between the amino groups of SA and the quinone group of PDA. Then, the unbound SA was washed away with PBS. Subsequently, biotinylated goat anti-human IgG solution (0.2 mg/mL) was dropped on the surface for 0.5 h to connect the biotin group of the antibody to SA. After the reaction was completed, unbound antibodies were washed with PBS buffer, and the obtained SPR biosensor chip surface was blocked with 1 mg/mL BSA. Combining biotin-avidin technology with AuNPs’ excellent optical characteristics and large specific surface area, the fixed amount of biorecognition molecules can be amplified, which improves the sensitivity of SPR to detect biomolecules.

Close

Fig. 2

Schematic diagram of the SPR biosensor chip structure.

Export to PPT

2.4 Immunoassay and simulated samples measurements

Immunoassays were performed by analyzing a series of hIgG solutions with different concentrations using the SPR biosensor chip immobilized with goat anti-human IgG. The sensitivity of the sensor was evaluated by monitoring the _Δλ_R caused by the binding reaction of the antigen and antibody. First, the obtained SPR biosensor chip was fixed on the SPR analyzer, and the sample chamber was pressed on the chip tightly. The PBS solution was injected to stabilize the baseline, hIgG solution was added to the chamber for immunoassay, and real-time changes in SPR spectra were monitored. After 30 min, the sample solution was washed with PBS solution three times to remove unbound antigen and maintained a steady λ_R; ultimately, the λ_R shifts caused by the interactions between the antibody and antigen were determined. All experiments were performed at room temperature and conducted in triplicate to examine the reliability of the assays. To evaluate the biosensor specificity of hIgG, BSA and OVA were determined by the same procedure as was utilized to detect hIgG. After the SPR biosensor was washed with PBS buffer, BSA (1 mg/mL), OVA (1 mg/mL) and hIgG (5 μg/mL) were added to the sample chamber and incubated for 30 min. After the specificity test, the SPR biosensor was washed three times with PBS buffer to wash away the nonspecifically bound analytes, and the changes in the SPR spectra before and after the sample addition were recorded.

Although hIgG exists in blood, it should be noted that the SPR biosensor developed in this paper can be extended to the detection of other biomolecules such as various pathogens in the future. Therefore, it is necessary to evaluate the actual detection performance of the SPR biosensor for various possible complex simulated samples. Soil (1 g), lake water (1 mL) and rabbit serum (80 μL) were added to 160 μL of 100 μg/mL hIgG standards. Then, the mixture was diluted to 8 mL with PBS to obtain a final hIgG concentration of 2 μg/mL. Rabbit blood samples of 1 mL were collected form normal rabbits. The rabbit blood was then centrifuged at 1,000 g for 5 min at 4 °C. The upper layer of serum was collected in clean tubes and stored at −20 °C before further experiment. The lake water and rabbit serum samples were measured directly. The soil sample was centrifuged at 5000 g for 15 min, the supernatants were collected, and the recovery rate, relative standard deviation (RSD) and other indices of detection were analyzed and calculated.

2.5 Stability and regeneration

The PBS solution was injected into the sample chamber, and the changes in the SPR spectra with time were observed. The spectra were recorded every 10 min for a total of 2 h. When the immunoassay was finished, the biosensor chip was thoroughly cleaned separately with piranha solution (H₂SO₄:30 % H₂O₂ = 7:3 V/V) and distilled water and then dried. The results show that the Au film could be utilized repeatedly.

3.1 Topography and refractive index sensitivity of the Au film SPR chip

The surface morphology and topography of the as-prepared SPR biosensor chip were determined using a SOLVER NEXT scanning probe microscope. For comparison, AFM images of the Au film, Au/PAH/AuNP film and Au/PAH/AuNP/PDA film are also given in Fig. 3. Fig. 3A clearly shows that the surface of the Au film without any modification is very smooth, and its root-mean-square (RMS) roughness is 0.21 nm. When AuNPs were absorbed on the Au film through positive and negative charge interactions, particles were irregularly and densely distributed on the chip surface (Fig. 3B). Particles and interface analysis show that these particles have a size of ∼ 18 nm. Combined with the particle size analysis of AuNPs above, it shows that AuNPs are well distributed on the chip surface. As seen from Fig. 3C, after AuNPs and exposed areas on the chip surface coated with PDA, the number of particles on the chip surface decreased, and the morphology became smooth, indicating that dopamine polymerized in situ at each location on the chip surface, making the Au film modified by AuNP/PDA smooth compared to the Au film modified by AuNPs. When goat anti-human IgG was fixed by the biotin-avidin system, the surface of the chip became rougher, and the fixed antibodies were distributed between AuNPs (Fig. 3D). AFM characterization shows that this method successfully fixes goat anti-human IgG to the surface of the biosensor chip through the biotin-avidin system.

Close

Fig. 3

AFM image (1 μm × 1 μm) of the surfaces of Au film (A), Au/PAH/AuNP film (B), Au/PAH/AuNP/PDA film (C), and SPR.

Export to PPT

The refractive indices of NaCl aqueous solution with a weight concentration in the range of 1 wt% to 15 wt% were measured, and the corresponding refractive indices ranged from 1.335 to 1.356 (Supplementary Fig. S1). As shown in Fig. 4, when the refractive indices of the NaCl aqueous solutions to be measured in the sample cell gradually increased, the λ_R of the resonance spectra redshifted accordingly. After plotting the value of resonance wavelength shift (_Δλ_R) and the refractive indices of the NaCl aqueous solutions (n) to be plotted (Supplementary Fig. S2), the following linear relationship was obtained:

Close

Fig. 4

The resonance spectra of the Au film SPR sensor in NaCl solutions with varying concentrations.

Export to PPT

_Δλ_R = 4295n-5727.4 (R² = 0.997, N = 7).

The refractive index sensitivity of the Au film SPR chip was 4295 nm·RIU⁻¹, which indicated that the Au film SPR chip has high sensitivity (Zhang et al., 2012; Wang et al., 2021).

3.2 AuNPs and SPR biosensor chip

According to the particle size distribution and TEM image, the particle size of AuNPs prepared by the microwave method in this experiment was ∼ 18 nm (Supplementary Fig. S3 and Fig. S4). The UV–Vis absorption spectra show that about 366 μg of AuNPs were adsorbed on the surface, meanwhile the characteristic absorption peak is located at 520 nm (Supplementary Fig. S5), which is largely consistent with the literature reports (Ye et al., 2012; Joshi et al., 2013) and indicates that AuNPs prepared by the microwave method have a uniform particle size. The resonance spectra changes of the SPR biosensor chip during the modification process were measured, and the resonance spectra of the Au film SPR chip, Au/PAH/AuNP film, and Au/PAH/AuNP/PDA film on water were measured on the SPR analyzer. Fig. 5 displays SPR spectra measured at θ = -13° with the Au film, Au/PAH/AuNP film and Au/PAH/AuNP/PDA in water, from which the λ_R of the Au film was determined to be 668.71 nm and redshifted to 680.41 nm after AuNPs were absorbed to the Au film by positive and negative charge interactions with PAH, and the _Δλ_R was 11.7 nm. Such a large shift in λ_R indicated that AuNPs were successfully deposited on the Au film. After the chip was coated with PDA for 10 min, its redshifted _Δλ_R was 13.14 nm.

Close

Fig. 5

Changes in the resonance spectra of water on the SPR chip before and after modification with AuNPs/PAH and PDA at an angle of incidence θ = 13°.

Export to PPT

Subsequently, SA was coupled to the chip surface by the reaction of the amino group and PDA. SA with different concentrations were added to the surface of Au/PAH/AuNP/PDA film, and the optimal concentration of SA was 50 μg/mL (Table S1). When biotinylated goat anti-human IgG was conjugated on the surface of the SA-immobilized chip, the _Δλ_R resulting from the immobilization of the antibodies onto the chip surface due to avidin–biotin interaction was monitored. Fig. 6A shows the real-time change in resonance spectra of goat anti-human IgG coupled on the surface of Au/PAH/AuNP/PDA/SA film within 60 min. The extended λ_R gradually redshifted, indicating that the biotinylated goat anti-human IgG was loaded onto the chip surface by the biotin-avidin reaction. A real-time change graph of _Δλ_R was plotted and is shown in Fig. 6B. At the initial stage of immobilization, λ_R redshifted rapidly, indicating that the antibody was rapidly coupled to the surface of the Au/PAH/AuNP/PDA/SA film. As the coupling progressed, the λ_R redshift rate decreased significantly after 30 min, which indicated that the SA immobilized on the PDA surface had gradually bound and saturated with the antibody, and the antibody binding rate decreased gradually and finally reached an equilibrium state. The linear relationship of _Δλ_R with biotinylated goat anti-human IgG coupling time T was as follows:

Close

Fig. 6

Real-time change of SPR spectra when antibody immobilized on the sensor chip (A) and kinetic curve of goat anti.

Export to PPT

_Δλ_R = 0.963lnT + 7.938 (R² = 0. 995, P < 0.0001, N = 17).

The change in UV–Vis absorption spectra of biotinylated goat anti-human IgG solution before and after coupling on the surface of the Au/PAH/AuNPs/PDA/SA film was also determined. After antibody immobilization, the absorption peak at 280 nm decreased significantly (Supplementary Fig. S6), which indicated that the 3.2 μg of antibody bound to SA and immobilized on the surface of the SPR biosensor chip; thus, the concentration of residual antibody solution decreased.

3.3 Determination of hIgG

In this paper, the SPR biosensor response to hIgG solutions with different concentrations was recorded. Fig. S7 displays the change in the SPR spectra before and after the detection of a 1 μg/mL hIgG solution. The λ_R of the blank sample (PBS solution) was 717.94 nm; when the sample was replaced with a 1 μg/mL hIgG solution, λ_R gradually redshifted, and after 30 min, _Δλ_R was 1.2 nm. As shown in Fig. 7, the SPR biosensor shows a good response to hIgG in the concentration range of 50 ng/mL ∼ 50 μg/mL. It could be noted that the biosensor still responded with a concentration of hIgG as low as 50 ng/mL, and its detection limit (LOD) was 50 ng/mL. Moreover, when the hIgG concentration reached 20 μg/mL, the response intensity of the biosensor reached the maximum value. The SPR biosensor showed a linear response to hIgG in the concentration range of 50 ng/mL ∼ 10 μg/mL. The curve of the concentration of the hIgG solution (C_hIgG) and _Δλ_R is presented in Fig. 7 (insert graph), and the following relation was obtained:

Close

Fig. 7

Resonance wavelength shifts of the SPR biosensor by detecting diffrent concentrations of hIgG and calibration curve for hIgG detection by the SPR biosensor.

Export to PPT

_Δλ_R = 2.041 × 10^-4 C_hIgG + 1.229 (R² = 0.987, P < 0.00001, N = 7).

The practical value and analytical performance of the developed SPR biosensor were evaluated by simulating samples at the same hIgG concentration. The soil, lake water and rabbit serum samples containing 2 μg/mL hIgG were injected into the chamber, and the _Δλ_R was recorded. The measured values were obtained by the regression equation (Table 1). The designed biosensor met the requirements for analysis of the simulated samples above; its RSDs below 10 % and recovery rates (between 94.45 % and 108.45 %) were acceptable, indicating that the designed SPR biosensor is feasible for detecting complex biological samples.

3.4 Specificity, stability and regeneration of hIgG detection

Specificity capabilities were investigated by measuring the response of the SPR biosensor to different proteins. Fig. 8 shows the _Δλ_R before and after the sensor detected BSA, OVA and hIgG. The _Δλ_R of 5 μg/mL hIgG was 1.72 nm, while BSA and OVA had almost no influence on the detection of hIgG. This phenomenon is due to antigen–antibody specific binding between goat anti-human IgG and hIgG, indicating that the SPR biosensor can specifically detect target molecules.

Close

Fig. 8

Response of SPR biosensor to different proteins.

Export to PPT

Stability is one of the critical evaluation indicators of biosensors. The stability of the sensor’s output signal is related to the accuracy of the detection results. This article evaluated the stability of SPR biosensors over a period of time. Fig. S8 shows the superposition of all resonance spectra within 2 h. These results suggest that the λ_R of the PBS solution was virtually unchanged within 2 h, remaining at 712.1 nm. The output light intensity only fluctuated slightly at this wavelength. This result indicates that the biotin-avidin-immobilized goat anti-human IgG on the sensitive film of the sensor did not detach. That is, the sensitive film remained stable and could be utilized stably for a certain period of time.

The SPR biosensor chip was immersed in piranha solution to clean the Au film and then rinsed repeatedly with ethanol and distilled water. The SPR spectra of the cleaned SPR biosensor chip and the unused Au film SPR chip against water were measured at the same incident angle (Supplementary Fig. S9), and the results clearly showed that after regeneration, the λ_R of the tested chip was basically returned to the position of the unused Au film SPR chip. Then the regenerated Au film SPR chip can be modified to fabricate SPR biosensor chip, and the cycle can be repeated at least 5 times. This result indicates that the prepared SPR biosensor chip can be reused, lowering detection costs.

In brief, AuNPs and PDA were used as the interface modification materials of the sensor, the biotin-avidin system was used to increase the antibody loading, and a new method for functionalization of the SPR biosensor chip was developed. Compared with our previous work, the LOD of the SPR biosensor developed in this article is approximately 40-fold lower than that of the sensor modified with PDA film and then conjugated with antibody, and compared with other SPR sensors for hIgG in situ detection, the SPR biosensor exhibits comparable simplification and convenience, as well as satisfactory sensitivity (Table 2).