Sensitive detection of total anti-Spike antibodies and isotype switching in asymptomatic and symptomatic individuals with COVID-19

Abstract

Early detection of infection is crucial to limit the spread of coronavirus disease 2019 (COVID-19). Here we develop a flow cytometry-based assay to detect severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike (S) protein antibodies in individuals with COVID-19. The assay detects specific immunoglobulin M (IgM), IgA, and IgG in individuals with COVID-19 and also acquisition of all IgG subclasses, with IgG1 being the most dominant. The antibody response is significantly higher at a later stage of infection. Furthermore, asymptomatic individuals with COVID-19 also develop specific IgM, IgA, and IgG, with IgG1 being the most dominant subclass. Although the antibody levels are lower in asymptomatic infection, the assay is highly sensitive and detects 97% of asymptomatic infections. These findings demonstrate that the assay can be used for serological analysis of symptomatic and asymptomatic infections, which may otherwise remain undetected.

Keywords: COVID-19; IgG subclasses; S protein; SARS-CoV-2; antibodies; asymptomatic; serological; symptomatic.

Graphical abstract

Figure 1

Specific antibodies against full-length S protein Plasma samples were collected from individuals with COVID-19 (n = 81) at time points with a median of 5 days post-illness onset (pio, n = 34), 10 days pio (n = 59), and 23 days pio (n = 66). (A–C) Samples were screened at 1:100 dilution for specific (A) IgM, (B) IgA, and (C) IgG against full-length SARS-CoV2 S protein expressed on the surface of HEK293T cells. Control samples included plasma samples from recovered SARS individuals (recovered SARS, n = 20), plasma samples from healthy donors (healthy, n = 22), and sera from seasonal human CoV-infected individuals (seasonal CoV, n = 20). (D–F) The proportions of individuals having a positive (D) IgM, (E) IgA, and (F) IgG response were analyzed. Data are shown as mean ± SD of two independent experiments, with dotted lines indicating mean + 3 SDs of healthy donors. An isotype response was defined as positive by SFB assay when binding was more than mean + 3 SDs of the healthy controls. Statistical analysis was carried out using Kruskal-Wallis tests followed by post hoc Dunn’s multiple comparisons tests. The p values for comparisons between the different time points are shown; ∗p ≤ 0.05, ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001, ∗∗∗∗p ≤ 0.0001.

Figure 2

Specific IgG subclasses against full-length S protein (A–D) Plasma samples collected from 81 individuals with COVID-19 at time points with a median of 5 days pio (n = 34), 10 days pio (n = 59), and 23 days pio (n = 66) were further screened for IgG subclasses, (A) IgG1, (B) IgG2, (C) IgG3, and (D) IgG4. (E–G) The four IgG subclass responses at a median of (E) 5 days pio, (F) 10 days pio, and (G) 23 days pio were plotted. (H–J) The proportions of individuals having a positive response at time points of (H) a median of 5 days, (I) a median of 10 days, and (J) a median of 23 days pio were analyzed. Data are shown as mean ± SD of two independent experiments, with dotted lines indicating mean + 3 SDs of healthy donors. An isotype response was defined as positive by SFB assay when the binding was more than mean + 3 SDs of the healthy controls. Statistical analysis was carried out using Kruskal-Wallis tests followed by post hoc Dunn’s multiple comparisons tests. For (A)–(D), only p values for comparisons between the different time points are shown; ∗p ≤ 0.05, ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001, ∗∗∗∗p ≤ 0.0001.

Figure 3

S protein-specific antibody profile in pre/asymptomatic infections (A–G) Plasma samples were collected from individuals with suspected COVID-19 infection or randomly from the general population under the Singapore Infectious Disease Act. Samples were screened at 1:100 dilution for specific (A) IgM, (B) IgA, (C) IgG, (D) IgG1, (E) IgG2, (F) IgG3, and (G) IgG4 against full-length SARS-CoV2 S protein expressed on the surface of HEK293T cells. A total of 109 samples were analyzed and included sera from PCR-positive and symptomatic individuals (PCR-positive, symptomatic; n = 16), sera from PCR-positive and no-symptom individuals (PCR-positive, none; n = 34), sera from PCR-positive individuals and individuals with unknown symptom status individuals (PCR-positive, unknown; n = 11), sera from PCR-negative individuals and individuals with unknown symptom status (PCR-negative, unknown; n = 13), sera from PCR-negative individuals/PCR not done and individuals with no symptoms (PCR-negative/not done [ND], none; n = 20), and sera from individuals with unknown PCR status and unknown symptom status (PCR unknown, unknown; n = 15). Statistical analysis was carried out using Kruskal-Wallis tests followed by post hoc Dunn’s multiple comparisons tests. The p values for comparisons of PCR-positive, symptomatic, and other groups are shown; ∗p ≤ 0.05, ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001, ∗∗∗∗p ≤ 0.0001.

Figure 4

Comparison of total IgG and IgG1 responses and comparison of the SFB assay and other serological assays (A–F) The total IgG response was compared with the IgG1 responses for all 109 samples from the NPHL: (A) PCR-positive and symptomatic infections (PCR-positive, symptomatic; n = 16), (B) PCR-positive and no-symptom infections (PCR-positive, none; n = 34), (C) PCR-positive and unknown symptom status infections (PCR-positive, unknown; n = 11), (D) PCR-negative and unknown symptom status infections (PCR-negative, unknown; n = 13), (E) PCR-negative/PCR not done and no-symptom infections (PCR-negative/ND, none; n = 20), and (F) PCR unknown and unknown symptom status infections (PCR unknown, unknown; n = 15). Statistical analysis was carried out using Wilcoxon matched-pairs signed-rank test. The p values for comparisons between the total IgG and IgG1 responses are shown; ∗p ≤ 0.05, ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001, ∗∗∗∗p ≤ 0.0001. Data are shown as mean ± SD of two independent experiments, with dotted lines indicating mean + 3 SDs of healthy donors. (G) Proportion of samples that were positive by serological tests: SFB assay, cPass, and RDT. An isotype response was defined as positive by SFB assay when the binding was more than mean + 3 SDs of the healthy controls. Samples that had a positive response for any one of the seven isotypes (IgM, IgA, IgG, IgG1, IgG2, IgG3, and/or IgG4) were defined as positive by the SFB assay. Data are shown as mean SD of two independent experiments, with dotted lines indicating mean + 3 SDs of healthy donors. +, positive; −, negative; symp, symptomatic; none, no symptoms.