Anti-Spike Antibody Response to Natural Infection with SARS-CoV-2 and Its Activity against Emerging Variants

Abstract

The outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has substantially affected human health globally. Spike-specific antibody response plays a major role in protection against SARS-CoV-2 infection. Here, we examined serological anti-spike antibody and memory B cell responses in adults with acute SARS-CoV-2 infection. Twenty-five adult patients were enrolled between January and September 2020, and 21 (84%) had a detectable spike-binding antibody response in serum on day 21 ± 8 (6 to 33) after the onset of illness. Among those with positive spike-binding antibody response, 19 (90%) had a positive hemagglutination titer and 15 (71%) had angiotensin-converting enzyme 2 (ACE2)-blocking serological activities. Follow-up serum samples collected 11 ± 1 (7 to 15) months after infection exhibited an average of 2.6 ± 1.0 (1.0 to 3.5)-fold reduction in the spike-binding antibody response. Moreover, convalescent and follow-up serum samples showed 83 ± 82 (15 to 306)- and 165 ± 167 (12 to 456)-fold reductions in the neutralization activity against the Omicron variant, respectively. Upon acute infection, spike-specific memory B cell responses were elicited, with an average frequency of 1.3% ± 1.2% of peripheral B cells on day 19 ± 7 (6 to 33) after the onset of illness. IgM memory B cells were predominantly induced. Patients with fever and pneumonia showed significantly stronger spike-binding, ACE2-blocking antibody, and memory B cell responses. In conclusion, spike-specific antibody response elicited upon acute SARS-CoV-2 infection may wane over time and be compromised by the emergence of viral variants. IMPORTANCE As spike protein-specific antibody responses play a major role in protection against SARS-CoV-2, we examined spike-binding and ACE2-blocking antibody responses in SARS-CoV-2 infection at different time points. We found robust responses following acute infection, which waned approximately 11 months after infection. Patients with fever and pneumonia showed significantly stronger spike-binding, ACE2-blocking antibody, and memory B cell responses. In particular, spike-specific antibody response in the convalescent and follow-up serum samples was substantially affected by emerging variants, especially Beta and Omicron variants. These results warrant continued surveillance of spike-specific antibody responses to natural infections and highlight the importance of maintaining functional anti-spike antibodies through immunization.

Keywords: COVID-19; Omicron; SARS-CoV-2; memory B cell; neutralization antibody.

FIG 1

Anti-spike antibody response to natural infection with SARS-CoV-2. (A) Spike-binding, RBD-binding, hemagglutination, and ACE2-blocking activity of serum samples from COVID-19 adult patients (n = 25). Each point represents peak serological activity for each patient, and the red line represents mean ± standard error of mean. Spike- and RBD-binding activities were measured by flow cytometry-based binding assay, and the activity was determined by the percentage of antibody-bound spike (or RBD)-expressing cells of serum minus the percentage of antibody-bound cells of PBS control. Healthy adult serum samples collected in 2017 were included as serum control. Spike (or RBD)-binding activity that was above two standard deviations plus the mean of healthy adult serum samples (dashed line) was positive. Anti-flu H3 and anti-SARS-CoV-2 spike human monoclonal antibodies were included as controls in the binding assay. For hemagglutination (HAT) assay, a serological titer of 1:20 or more was considered to be positive. Samples with a reciprocal titer of less than 20 were assigned a value of 10. The ACE2-blocking titer was expressed as the reciprocal of the serum dilution, giving 50% inhibition of signal compared to maximum signal. The original, undiluted serum that failed to inhibit ACE2-RBD interaction was scored as negative. (B) Spike- and RBD-binding serological activities during the course of COVID-19 in 25 patients. There were 5, 11, 12, and 20 serum samples collected in the 1st, 2nd, 3rd, and over the 4th week of illness, respectively. The mean ± standard error of mean of days after onset for each group was indicated. The significance between serological activities at different time points was determined using one-way ANOVA with post hoc Tukey’s test. *, P < 0.05; **, P < 0.01; ns, not significant. (C) Magnitude and kinetics of serological response for four COVID-19 pneumonia patients. (D) Magnitude and kinetics of serological response for 10 COVID-19 patients with mild upper respiratory tract illness. RBD, receptor-binding domain; D, day after onset.

FIG 2

Anti-spike memory B cell response to natural infection with SARS-CoV-2. (A) Enzyme-linked immunosorbent spot showing SARS-CoV-2 spike-specific IgG, IgM, and IgA memory B cell in the peripheral blood upon natural infection. Circulating total IgG, IgM, and IgA memory B cells were also measured. A total of 200,000 cultured cells were added into each of the anti-spike B cell wells, and 5,000 cultured cells were added into each of the total B cell wells. Donor 1 (day 27) was a 55-year-old woman with severe COVID-19 with pneumonia. Donor 6 (day 21) was a 43-year-old man with moderate COVID-19 and pneumonia. Donor 8 (day 18) was a 22-year-old woman with mild COVID-19 and a running nose. (B) Frequency of spike-specific IgG, IgM, and IgA memory B cells in the peripheral blood (n = 20). The frequency is defined as the percentage of spike-specific IgG (IgM or IgA) B cells in the total IgG (IgM or IgA) B cells. Each point represents memory B cell frequency for each patient, and the red line represents mean ± standard error of mean. One-way ANOVA with post hoc Tukey’s test was used to compare the difference among groups. *, P < 0.05; **, P < 0.01. ns, not significant. (C) Relationship of spike-specific memory B cell frequency (IgG + IgM + IgA) and spike-binding and RBD-binding serological activity among COVID-19 patients. Linear regression was used to model the relationship between two variables. MBC, memory B cell.

FIG 3

Comparison of anti-spike antibody and B cell responses between patients with (n = 5) and without (n = 20) pneumonia (A) and patients with (n = 14) and without (n = 11) fever (B). Each point represents antibody or B cell response for each patient, and the red line represents mean ± standard error of mean. Unpaired two-tailed t test was used to compare the difference between two groups. *, P < 0.05; **, P < 0.01; ***, P < 0.001. ns, not significant. HAT, hemagglutination titer.

FIG 4

Effect of SARS-CoV-2 variants on anti-spike antibody response to natural infection. Neutralization activities of convalescent-phase serum samples (n = 12, left) and follow-up serum samples (n = 8, right) against wild type, Beta variant, Delta variant, and Omicron variant pseudoviruses. All serum samples were tested with a starting dilution of 1:10 (cutoff, dashed line), and those that failed to neutralize virus at the starting dilution were recorded as half-maximal inhibitory concentration (IC₅₀) titer 5. Each point represents IC₅₀ titer for each sample, and the red line represents mean ± standard error of mean. One-way ANOVA with post hoc Tukey’s test was used to compare the difference among groups. *, P < 0.05; ****, P < 0.0001; ns, not significant.