An impedimetric biosensor for COVID-19 serology test and modification of sensor performance via dielectrophoresis force

Abstract

Coronavirus disease 2019 (COVID-19) has caused significant global morbidity and mortality. The serology test that detects antibodies against the disease causative agent, the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has often neglected value in supporting immunization policies and therapeutic decision-making. The ELISA-based antibody test is time-consuming and bulky. This work described a gold micro-interdigitated electrodes (IDE) biosensor for COVID antibody detection based on Electrochemical Impedance Spectroscopy (EIS) responses. The IDE architecture allows easy surface modification with the viral structure protein, Spike (S) protein, in the gap of the electrode digits to selectively capture anti-S antibodies in buffer solutions or human sera. Two strategies were employed to resolve the low sensitivity issue of non-faradic impedimetric sensors and the sensor fouling phenomenon when using the serum. One uses secondary antibody-gold nanoparticle (AuNP) conjugates to further distinguish anti-S antibodies from the non-specific binding and obtain a more significant impedance change. The second strategy consists of increasing the concentration of target antibodies in the gap of IDEs by inducing an AC electrokinetic effect such as dielectrophoresis (DEP). AuNP and DEP methods reached a limit of detection of 200 ng/mL and 2 μg/mL, respectively using purified antibodies in buffer, while the DEP method achieved a faster testing time of only 30 min. Both strategies could qualitatively distinguish COVID-19 antibody-positive and -negative sera. Our work, especially the impedimetric detection of COVID-19 antibodies under the assistance of the DEP force presents a promising path toward rapid, point-of-care solutions for COVID-19 serology tests.

Keywords: COVID-19 serology test; Dielectrophoresis force; Gold interdigitated electrodes; Gold nanoparticles; Impedimetric biosensor.

Fig. 1

Experimental setup. (A) Basic components of the impedimetric sensing for COVID-19 serology test. From left to right are the IDE sensor chip with PDMS mask, detailed structure of each electrode and the chip-holder that connects the chip to the impedance analyzer or the functional generator. Differential interference contrast (DIC) microscopy images showing the edge of the electrode fingers (i) and the middle of the digit pattern (ii). Scale bar represents 10 μm. (B) Schematic representation of the COVID-19 Ab detection via the AuNP approach, which requires the immobilization of either the SARS-CoV-2 S1 subunit or the full-length S protein (S1/S) on the SiO2 substrate between the gold (Au) electrode digits. The anti-S1/S Abs in the positive serum sample will bind to the antigen (S1/S), resulting in the deposit of AuNPs when AuNP-conjugated anti-human IgG Abs are applied. The presence of AuNPs will cause measurable impedance changes. The negative serum may have some non-specific binding to S1/S, but no AuNP will be deposited under this scheme. (C) Schematic representation of the COVID-19 Ab detection via the DEP approach. Anti-S IgG Abs in the positive serum were selectively preconcentrated by the DEP force against other serum components. The binding of Anti-S IgG Abs (in the positive serum) to the sensor surface generates significant impedance change so that negative and positive sera can be separated.

Fig. 2

Sensitivity of the sensor. (A) The equivalent circuit of the IDE sensor. Impedance spectra (magnitude) at the frequency range from 1k Hz – 1000 kHz measured after each step of surface modification when 200 ng/mL (B), 40 ng/mL (C) and 0 ng/mL of anti-S1 Ab (D) were tested. (B-D) IgG: anti-SARS-CoV-2 S1 IgG mAb; full assembly: APTES + S1 + block + anti-SARS-CoV-2 S1 IgG + anti-human IgG conjugated AuNPs. (E) Percentage change of impedance magnitude (%ΔZ) between the AuNPs binding step and the Ab binding step measured at frequency range from 1k Hz – 1000 kHz for different concentrations of anti-S mAb. %ΔZ=(Z_AuNP-Z_mAb)/Z_mAb*100% where Z_AuNP, and Z_mAb represent the total impedance after AuNP and mAb step, respectively. (F) %ΔZ calculated at 5.76 kHz (pointed out by arrows in (B-D) and the dotted vertical line in (E)) for two experimental replicates. Error bars represent standard deviation. Statistics is determined by one-way ANOVA with Tukey's multiple comparisons test. *p<0.05.

Fig. 3

Evaluating the in-house made S protein and Ab-AuNP conjugates for sensor's clinical application. Impedance spectra (magnitude) at the frequency range from 1 kHz–1000 kHz measured after blocking step (blue), serum incubation step (red) and AuNPs binding step (black), respectively for a positive (A) and a negative serum (B). (C) %ΔZ between the AuNPs binding step and the serum step over all frequencies measured when using our in-house made AuNPs. %ΔZ=(Z_AuNP-Z_serum)/Z_serum*100%. (D) %ΔZ calculated at 2.3 kHz from (C) for two positive and two negative sera. Error bars represent standard deviation. Statistics is determined by one-way ANOVA with Tukey's multiple comparisons test. *p<0.05. (E-G) FESEM images for positive (Pos#8), negative (Neg#7) and blank (blnk) samples after the AuNP step when using the in-house made AuNPs. Yellow arrowheads point to the individual AuNPs. Blank samples had no serum added. The green scale bar represents 300 nm.

Fig. 4

The effect of the impedance measuring buffer. (A) The absolute impedance magnitude over frequencies from 1 kHz–1000 kHz of a typical EIS response before (Serum, red curve) and after reacting with secondary Ab-AuNP conjugates (AuNP, blue curve) for a COVID-19 positive serum sample. EIS was measured in the ELISA buffer. Gray arrows mark the frequency range where impedance is dominated by R_sol. (B) %ΔZ caused by the binding of Ab-AuNP conjugates over all frequencies tested when EIS responses were measured in the ELISA buffer. %ΔZ=(Z_AuNP-Z_serum)/Z_serum*100%. (C) %ΔZ calculated at 1 kHz from (A) for two positive (Pos#9, Pos#10) and two negative (Neg#8, Neg#9) sera. Error bars represent standard deviation. Statistics is determined by one-way ANOVA with Tukey's multiple comparisons test. *p<0.05. **p<0.01. (D-G) AFM characterization of the cleaned gap surface (D), the sample before applying the serum (E), the samples after applying negative serum and Ab-AuNP conjugates (F), and the sample after applying positive serum and Ab-AuNP conjugates (G). The gap area between the electrode digits were outlined by white dotted line.

Fig. 5

Ab detection with the assistance of DEP. (A-B) optimization of DEP conditions by calculating %ΔZ between the Ab binding step and its previous step (blocking) over all frequencies measured. %ΔZ = (Z_Serum-Z_blocking)/Z_blocking*100% where Z_serum and Z_block represent the total impedance after the serum step and its previous step (blocking), respectively. Four DEP conditions were tested: no DEP, DEP by applying AC voltage at 50 MHZ (f) with 0.05Vpp, 0.1Vpp and 0.2Vpp, respectively. The immunoreaction between total human IgG and anti-human IgG Abs (at 1 μg/ml) was used as a test. (B) %ΔZ calculated at the transition frequency for all conditions in (A). Statistics is determined by one-way ANOVA with Tukey's multiple comparisons test. *p<0.05. (C-D) The effect of DEP force induced by AC voltage at 50 MHZ, 0.1Vpp on the detection of COVID-19 mAbs (C) or serum samples with different dilutions (D) by measuring %ΔZ at the transition frequency. The IDE was treated with S protein for COVID-19 Ab detection. One positive (Pos#8) and one negative (Neg#7) serum were used. (1:4): serum was diluted four times. (1:16): serum was diluted 16 times. Error bars represent the standard deviation of three replicates. Statistics is determined by two-way ANOVA with Tukey's multiple comparisons test. *p<0.05, **p<0.01.