Highly specific monoclonal antibodies and epitope identification against SARS-CoV-2 nucleocapsid protein for antigen detection tests

Abstract

The ongoing coronavirus disease 2019 (COVID-19) pandemic is a major global public health concern. Although rapid point-of-care testing for detecting viral antigen is important for management of the outbreak, the current antigen tests are less sensitive than nucleic acid testing. In our current study, we produce monoclonal antibodies (mAbs) that exclusively react with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and exhibit no cross-reactivity with other human coronaviruses, including SARS-CoV. Molecular modeling suggests that the mAbs bind to epitopes present on the exterior surface of the nucleocapsid, making them suitable for detecting SARS-CoV-2 in clinical samples. We further select the optimal pair of anti-SARS-CoV-2 nucleocapsid protein (NP) mAbs using ELISA and then use this mAb pair to develop immunochromatographic assay augmented with silver amplification technology. Our mAbs recognize the variants of concern (501Y.V1-V3) that are currently in circulation. Because of their high performance, the mAbs of this study can serve as good candidates for developing antigen detection kits for COVID-19.

Keywords: COVID-19; SARS-CoV-2; monoclonal antibody; nucleoprotein; point-of-care testing.

Graphical abstract

Figure 1

Production of high-affinity and specific mAb against SARS-CoV-2 NP using wheat germ cell-free synthesized antigen (A) Schematic diagram of hybridoma cells production to generate anti-SARS-CoV-2-NP mAb. Purified protein was injected into BALB/c mice. After 4 weeks, lymphocytes from immunized mice were fused with myeloma cells, and 144 hybridoma cells were established. (B) Of the 144 clones, the 12 that exhibited high reactivity to antigen proteins, as revealed by indirect ELISA, AlphaScreen, and Bio-layer interferometry, were selected for further investigation (two technical replicates). Red line indicates cutoff line of screening; S/N = 10 for ELISA, S/N = 40 for AlphaScreen, and K_D = 1.0 × 10⁹. k_on and k_off values for each antibody clones to the ΔN-NP antigen estimated by OctetRED96 instrument using hybridoma supernatant are indicated as dots on the two-dimensional plot. (C) Specificity screening of mAbs. FLAG-glutathione S-transferase (GST)-tagged NPs derived from several human coronaviruses were produced in the wheat germ extract system. Reactivity of generated mAbs was validated by immunoblot analysis using either anti-FLAG or the indicated antibodies. Three clones specifically detect SARS-CoV-2 NP and were selected (representative data of two technical replicates). (D) Isotype of selected mAbs. (E) Schematic diagram of SARS-CoV-2-NP domain architecture and epitopes of antibodies. CTD, C-terminal domain; LKR, flexible linker region; NTD, N-terminal domain. Positions of epitopes in a structural model of whole-length dimer-forming NP, constructed by homology modeling using partial structures of SARS-CoV-2 NP are shown (PDB: 6yun; PDB: 6m3m). Epitope localizations of each mAb on molecular surface are highlighted in different colors. Critical residues for the specificity of mAb were indicated by arrows.

Figure 2

Binding affinity of developed mAbs and the application for viral NP antigen detection (A) Affinity measurement of selected monoclonal antibodies on the Octet RED96 instrument. Association and dissociation of each mAb to full-length NP at various concentrations (50, 25, 12.5, 6.25, 3.13, 1.56, and 0.78 nM) was evaluated using anti-mouse IgG capture (AMC) sensor (two technical replicates). (B) Immunoblot analysis of mock or SARS-CoV-2-infected VeroE6/TMPRSS2 cell lysates (representative image of two technical replicates). (C) Immunofluorescence analysis (representative image of two technical replicates). VeroE6/TMPRSS2 cells were infected or mock infected with SARS-CoV-2. After 24 h, cells were fixed and then stained with mAbs (hybridoma supernatant; red) and DAPI (blue). (D) Paraffin-embedded lung biopsy specimen from a case of COVID-19 was examined for immunohistochemical detection of SARS-CoV2 using our antibody. The positive signals (arrowheads) are seen in bronchiolar epithelial cells (top), pneumocytes (middle), and endothelial cell (bottom).

Figure 3

Development of antigen-capture ELISA (A) Schematic representation of sandwich ELISA. Each mAb was labeled with horseradish peroxidase (HRP) and subjected to ELISA analysis. 9 pairs of antibodies were tested. (B) Determination of the optimal combination of capturing and detection mAbs. S/N ratios for antigen detection by each of the 9 combinations were calculated in the presence of 2 ng/mL antigen versus blank (two technical replicates). (C) Detection limit of ELISA. Serially diluted recombinant NP and inactivated SARS-CoV-2 were subjected to ELISA. Graph data are presented as mean ± SD (six technical replicates). The error bars represent SD. The detection limits of both recombinant NP protein and SARS-CoV-2 were determined according to the cutoff value, which was calculated by the formula (average +3SD). (D) Specificity of ELISA. Simulated specimens positive for indicated viruses prepared by adding recombinant protein or common respiratory viruses to pooled COVID-19 RT-PCR-negative specimens were analyzed by ELISA. “Virus” indicates inactivated virus (at least 10⁶ copies/mL for HCoV-229E and HCoV-OC43 or at least 10⁵ TCID₅₀/mL [50% tissue culture infectious dose] for other viruses). “Recombinant protein” indicates recombinant NP antigen of human coronavirus (200 ng/mL). Graph data are presented as mean ± SD (three technical replicates). HRV, human rhinovirus; IFAV, influenza A virus; IFBV, influenza B virus; RSV, respiratory syncytial virus. (E) Detection of variant of concerns strains by ELISA. All the three major variants were detected efficiently. Graph data are presented as mean ± SD (three technical replicates). (F) Clinical performance of ELISA assay for nasal swab samples from RT-PCR negative (n = 72) and positive (n = 72). Boxplots of index values at ELISA assay were depicted. The p value was calculated using Welch’s t test (two-tailed). (G) Receiver operating characteristic curves for antigen-capture ELISA. (H and I) Sensitivity and specificity of ELISA results according to real-time PCR cycle threshold (Ct) values group based on NIID-N2 primer set. (J) Relation between RNA copy number in PCR-positive specimens and reactivity of antigen-capture ELISA (n = 72; Spearman’s correlation).

Figure 4

Immunochromatographic test using developed monoclonal antibodies with silver amplification technology (A) Schematic diagram of lateral flow immunoassay with silver amplification technology. The antigen in the sample dropped into the device flows on the cellulose membrane together with the colloidal gold-labeled anti-SARS-CoV-2 NP antibody, and when captured by the membrane-immobilized capture antibody, it develops color and appears as a single band. Adherence of silver ions to the surface of a catalytic gold nanoparticle causes electrons to reduce the silver atoms, leading to the size enhancement followed by 1,000-fold improvement in visibility. (B and C) Size differences in SEM images (B) and naked eye visualized bands (C) with and without silver amplification. (D) Representative test result for positive and negative. (E) Specificity of LFIA. Simulated specimens positive for indicated viruses prepared by adding recombinant protein or common respiratory viruses to pooled COVID-19 RT-PCR-negative specimens were analyzed by LFIA. Virus indicates inactivated virus (at least 10⁶ copies/mL for HCoV-229E and HCoV-OC43 or at least 10⁵ TCID₅₀/mL for other viruses). “Protein” indicates recombinant NP antigen of corresponding virus (200 ng/mL). NT, not tested (two technical replicates). (F and G) Detection limit of LFIA. Recombinant protein (F) and inactivated SARS-CoV-2 (G) were subjected to LFIA analysis. Detection limit using SARS-CoV-2 was compared with indicated antigen detection kits. + and − indicate positive and negative detection, respectively (two technical replicates). (H) Detection of variant strains of SARS-CoV-2. Virus indicates inactivated virus (less than 1.5 × 10⁶ copies/mL). Protein indicates recombinant NP antigen from each strain (less than 100 pg/mL; two technical replicates). (I) Sensitivity and specificity of indicated SARS-CoV-2 antigen-detection kits in PCR-positive (n = 45) and negative (n = 63 for this study; n = 45 for the others) specimens in nasopharyngeal swabs.