Monoclonal antibody pairs against SARS-CoV-2 for rapid antigen test development

Abstract

Background: The focus on laboratory-based diagnosis of coronavirus disease 2019 (COVID-19) warrants alternative public health tools such as rapid antigen tests. While there are a number of commercially available antigen tests to detect severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), all cross-react with the genetically similar SARS-CoV-1 or require an instrument for results interpretation.

Methodology/principal findings: We developed and validated rapid antigen tests that use pairs of murine-derived monoclonal antibodies (mAbs), along with gold nanoparticles, to detect SARS-CoV-2 with or without cross-reaction to SARS-CoV-1 and other coronaviruses. In this development, we demonstrate a robust antibody screening methodology for the selection of mAb pairs that can recognize SARS-CoV-2 spike (S) and nucleocapsid (N) proteins. Linear epitope mapping of the mAbs helped elucidate SARS-CoV-2 S and N interactions in lateral flow chromatography. A candidate rapid antigen test for SARS-CoV-2 N was validated using nasal swab specimens that were confirmed positive or negative by quantitative reverse-transcription polymerase chain reaction (RT-PCR). Test results were image-captured using a mobile phone and normalized signal pixel intensities were calculated; signal intensities were inversely correlated to RT-PCR cycle threshold (Ct) value.

Conclusion/significance: Overall, our results suggest that the rapid antigen test is optimized to detect SARS-CoV-2 N during the acute phase of COVID-19. The rapid antigen tests developed in this study are alternative tools for wide scale public health surveillance of COVID-19.

Fig 1. Binding of SARS-CoV-2 S1 and N mAbs. mAb clones were harvested from the hybridomas of SARS-CoV-2 spike subunit (S1) or nucleocapsid (N) immunized mice.

(A) ELISA using hybridoma supernatants from the lymph nodes of mouse EB017 or the spleen of mouse EB024, both of which were infected with SARS-CoV-2 S1. Hybridoma supernatants were tested with S1 from SARS-CoV-2, SARS-CoV-1, MERS, NL63, 229E, HKU1, and OC43 to evaluate cross-reactivity. B) ELISA using hybridoma supernatants from the lymph nodes of mouse EB025 infected with SARS-CoV-2 N. Hybridoma supernatants were tested with S1 from SARS-CoV-2 and SARS-CoV-1 to evaluate cross-reactivity. Cross rection for both ELISAs is measured by % clones that demonstrated a positive reaction, or an OD₅₀ greater than or equal to 5 times the standard deviation of the blank signal intensity.

Fig 2. Limit of detections of mAb combinations.

Serially diluted recombinant SARS-CoV-2 spike (S1) or nucleoprotein (N) were run on dipsticks with pairs of (A) SARS-CoV-2 S1 mAbs or (B) SARS-CoV-2 N mAbs. mAbs were either conjugated to the nanoparticle or applied onto the nitrocellulose membrane to create a sandwich immunoassay if antigen binding was present. (C) The SARS-CoV-2 N rapid test composed of mAbs 1 and 453 was selected based on its lowest limit of detection to evaluate performance using a panel of contrived SARS-CoV-2 nasal swab specimens. Nasal swab specimens with known viral loads (VL), nucleocapsid (N) protein concentrations, and cycle threshold (Ct) values were allowed to react with the rapid test for 15 minutes. C, control. T, test.

Fig 3. SARS-CoV-2 S1 and N protein alignment and linear epitope mapping of the mAbs.

(A) Clustal Omega-generated amino acid sequence alignment of SARS-CoV-2 S1 and SARS-CoV-1 S1. The epitope recognized by each mAb is indicated by the highlighted boxes, the position of the epitope is noted. SD1/2, subdomain 1/2. RBD, receptor binding domain. The amino acids representing the RBD are in bold. (B) Clustal Omega-generated amino acid sequence alignment of SARS-CoV-2 N and SARS-CoV-1 N. The epitope recognized by each mAb is indicated by the highlighted boxes, the position of the epitope is noted. CTD, C terminal domain. NTD, N terminal domain. *, single, fully conserved residue;:, conservation between groups of strongly similar properties (i.g., those scoring >0.5 in the Gonnet PAM 250 matrix); •, conservation between groups of weakly similar properties (i.e., those scoring ≤0.5 in the Gonnet PAM 250 matrix).

Fig 4. SARS-CoV-2 S1 and N interactions in lateral flow immunochromatography. mAbs were either conjugated to gold nanoparticles or adsorbed onto the nitrocellulose membrane.

Different combinations were tested to distinguish between specific and cross-reactive detection and indicate differential epitope binding. SARS-CoV-1/2 S1 or SARS-CoV-1/2 N, at final concentrations of 125 ng/ml, were allowed to react with the rapid antigen test. (A) mAb pairs 46–349 and 46–124 specifically recognize SARS-CoV-2 S1, whereas mAb pair 124–349 recognize both SARS-CoV-2 S1 and SARS-CoV-1 S1. (B) Ligand binding is not observed in rapid tests composed of mAb pairs 192–267, 267–453, and 453–192. However, ligand binding occurs with mAb combinations 1–453, 1–267, 1–329+453, and 329–192; these combinations detect both SARS-CoV-1 and SARS-CoV-2 N. C, control. T, test.

Fig 5. SARS-CoV-2 N rapid antigen test performance.

(A) Nasal swab specimen (50 μl) collected from qRT-PCR confirmed positive patients (n = 13) and negative patients (n = 10) were allowed to react with a SARS-CoV-2 N rapid antigen test composed of mAbs 1–453. Ct, Cycle threshold value. C, control. T, test. (B) Receiver operator characteristic (ROC) curve of SARS-CoV-2 N rapid antigen test. Test performance is demonstrated in terms of true positive rate (sensitivity) versus false positive rate (1-specificity). AUC, area under the curve. CI, confidence interval. (C) Normalized signal intensity was compared to the qRT-PCR-derived sample cycle threshold (Ct) value. Points are plotted with a linear regression line of y = -6.378x + 21 and p < 0.0001 and R² = 0.847.