Label-free impedimetric immunosensor for point-of-care detection of COVID-19 antibodies

Abstract

The COVID-19 pandemic has posed enormous challenges for existing diagnostic tools to detect and monitor pathogens. Therefore, there is a need to develop point-of-care (POC) devices to perform fast, accurate, and accessible diagnostic methods to detect infections and monitor immune responses. Devices most amenable to miniaturization and suitable for POC applications are biosensors based on electrochemical detection. We have developed an impedimetric immunosensor based on an interdigitated microelectrode array (IMA) to detect and monitor SARS-CoV-2 antibodies in human serum. Conjugation chemistry was applied to functionalize and covalently immobilize the spike protein (S-protein) of SARS-CoV-2 on the surface of the IMA to serve as the recognition layer and specifically bind anti-spike antibodies. Antibodies bound to the S-proteins in the recognition layer result in an increase in capacitance and a consequent change in the impedance of the system. The impedimetric immunosensor is label-free and uses non-Faradaic impedance with low nonperturbing AC voltage for detection. The sensitivity of a capacitive immunosensor can be enhanced by simply tuning the ionic strength of the sample solution. The device exhibits an LOD of 0.4 BAU/ml, as determined from the standard curve using WHO IS for anti-SARS-CoV-2 immunoglobulins; this LOD is similar to the corresponding LODs reported for all validated and established commercial assays, which range from 0.41 to 4.81 BAU/ml. The proof-of-concept biosensor has been demonstrated to detect anti-spike antibodies in sera from patients infected with COVID-19 within 1 h. Photolithographically microfabricated interdigitated microelectrode array sensor chips & label-free impedimetric detection of COVID-19 antibody.

Keywords: Biosensors; NEMS.

Photolithographically microfabricated interdigitated microelectrode array sensor chips & label-free impedimetric detection of COVID-19 antibody.

Fig. 1. Schematic illustration of the surface functionalization of IMA and spike protein immobilization.

a Hydroxyl groups produced on the glass surface upon plasma exposure react with APTES to form SAM with surface -NH₂ groups, b MUOH forms SAM with surface -OH groups on the Au surface, and c MUA forms SAM with surface -COOH groups on the Au surface. The -NH₂ group reacts with SA to functionalize the surface with -COOH groups in b, c, which on subsequent application of EDC/NHS conjugation chemistry led to covalent immobilization of the S-proteins ().

Fig. 2. EIS characterization of the surface functionalization of IMA and spike protein immobilization.

a Bode plot of bare gold (black line) IMA functionalized with APTES (blue line), MUOH (red line), SA (purple dashed line), and spike protein immobilization (green dashed line). b Plot of the impedance magnitude (Z) at 13.2 Hz recorded after each step of IMA functionalization.

Fig. 3. Model equivalent circuit of biosensor and the interfacial capacitance representing each layer formed on surface functionalization and probe immobilization.

Schematic representation of a a simplified model equivalent circuit constituting a capacitive immunosensor and b interfacial capacitors formed by surface functionalization and S-protein immobilization of the IMA.

Scheme 1

Schematic rendering of the label-free biosensor non-Faradaic EIS detection of the selective binding of antibodies to the trimeric spike proteins immobilized on the recognition layer of the impedimetric immunosensor. Ready-to-use biosensor a obtained after surface functionalization and trimeric spike protein immobilization. Non-Faradaic EIS detection was accomplished first by measuring the impedance (Z_bgr) after blocking b with milk solution. In the second step, the impedance response (Z_Serum) due to the affinity-specific captured antibodies c was measured. The detection signal (ΔZ) was obtained from the difference between these two measurements, i.e., ΔZ = Z_Serum – Z_bgr. The sensors were thoroughly rinsed with 0.001× PBS pH 6.5 buffer after blocking (b) and after serum sample treatments (c) to minimize unintended contamination from the 1x PBS solution. All EIS measurements were conducted in 0.001× PBS solution at pH 6.5.

Fig. 4. Dependance of the biosensor sensitivity on the ionic strength of the sensing medium for Non-Faradaic EIS detection.

Non-Faradaic EIS detection of selective binding of antibodies to the S-proteins immobilized on the recognition layer of the impedimetric immunosensor comparing the effect of PBS concentration on the device sensitivity using MUOH- and APTES-modified IMA. (a, e)Bode plot of experimental EIS spectra before (black dash line) and after (redline) incubation with serum sample, (b, f) EIS spectra plotted in expanded scale to show the impedance change, (c, g) plot of ΔZ (ΔZ = Z_Serum– Z_bgr) versus frequency, () COVID-19 positive serum, () COVID-19 negative serum, () 1% milk blank samples, solid lines are drawn to guide the eyes, and (d, h) plot of the detection signal ΔZ_Rel % (ΔZ_Rel % = ΔZ/Z_bgr × 100).Plots (a–d) and (e–h) correspond to EIS measurements made in 1× PBS or 0.001× PBS solution at pH 6.5, respectively.

Fig. 5. Effect of spike protein coated gold electrode on the biosensor sensitivity.

a Impedance magnitudes recorded after different sequential stages of surface functionalization with APTES, MUA, SA, and S-protein immobilization. b As depicted schematically in Figure 1, Bode plot of the experimental EIS spectra before (black dashed line) and after (redline) incubation with the serum sample. c Magnified EIS spectra plotted to show the impedance change. d Plot of the detection signal ΔZ_Rel % (ΔZ_Rel % = ΔZ/Z_bgr × 100). All EIS measurements were performed in 0.001× PBS.

Fig. 6. Comparison of the experimental and simulated EIS data points obtained from the model equivalent circuit.

Experimental (black line) and simulated (red dashed line) EIS spectra for detecting the selective binding of antibodies to the S-proteins immobilized on IMA modified with MUOH and APTES plotted in a Bode and b Nyquist representations and c plot of the capacitance change ΔQ_Rel% obtained from the simulation.

Fig. 7. Standard calibration curve for the detection of anti-SARS-CoV-2 antibodies obtained using label-free non-Faradaic EIS biosensor and comparison of its performance with other biosensors.

Standard calibration curve for the detection of anti-SARS-CoV-2 immunoglobulin (IgG, IgA, IgM), which were experimentally determined in duplicate (n = 2) using different concentrations of WHO IS (NIBSC 20/136) antibody solutions. Linear fit (solid line) to the experimental data points (filled squares) yielded a straight line, ΔZ_Rel% = 2.1178 log [BAU/mL] + 3.7173 with R² = 0.9979, indicating that the highest concentration used, i.e., 100 BAU/mL still lied within the linear region of the impedance response (ΔZ_Rel % = ΔZ/Z_bgr × 100) to antibody concentration. The dotted line represents the LOD determined fromthe impedance response ΔZ_Rel% of the negative serum sample with no anti-SARS-CoV-2 immunoglobulin. For ease of representation, the bindingantibody unit BAU/mL is used to express the concentration of all immunoglobulins (IgG, IgA, IgM), and according to WHO/NIBSC both (IU/mL and BAU/mL) units are numerically identical. All EIS measurements were conducted in 0.001× PBS at pH 6.5.

Fig. 8. Instruments and accessories for experimental EIS data acquisition.

Custom built electrical contact pad, sensor chip, and sensor chip with a PDMS mask that contained eight matching wells (a). Experimental setup consisting of a Biologic SP-200 potentiostat and the associated computer control system for EIS measurements (b) of the affinity-bound target analytes on the sensor chip.

Fig. 9. Schematic depiction of lift-off process flow for microfabrication of gold interdigitated microelectrode array.

Standard lift-off microfabrication technique forfabrication of IMA (a): (1) borofloat glass wafer coated with hexamethyldisilazane, (2) UV light exposureof the bilayer coated wafer after spin coating and baking AZ1512 photoresist and LOR5B resist through a custom designed photomask using mask aligner, (3) after sequential development with AZ 1:1 and followed by MF 319 developers, (4) after metal deposition with 5 nm of a Cr adhesion layer and 60 nm gold, andfinally (5) after lift-off with remover-PG. Borofloat glass wafer withpatterned IMA electrode sensor chips obtained after lift-off (b), and image of interdigitated electrodes (c) with width, gap, andthickness of 4 µm, 2 µm, and 60 nm, respectively.