Preexisting and de novo humoral immunity to SARS-CoV-2 in humans

Abstract

Zoonotic introduction of novel coronaviruses may encounter preexisting immunity in humans. Using diverse assays for antibodies recognizing SARS-CoV-2 proteins, we detected preexisting humoral immunity. SARS-CoV-2 spike glycoprotein (S)-reactive antibodies were detectable using a flow cytometry-based method in SARS-CoV-2-uninfected individuals and were particularly prevalent in children and adolescents. They were predominantly of the immunoglobulin G (IgG) class and targeted the S2 subunit. By contrast, SARS-CoV-2 infection induced higher titers of SARS-CoV-2 S-reactive IgG antibodies targeting both the S1 and S2 subunits, and concomitant IgM and IgA antibodies, lasting throughout the observation period. SARS-CoV-2-uninfected donor sera exhibited specific neutralizing activity against SARS-CoV-2 and SARS-CoV-2 S pseudotypes. Distinguishing preexisting and de novo immunity will be critical for our understanding of susceptibility to and the natural course of SARS-CoV-2 infection.

Fig. 1. Flow cytometric detection and specificity of antibodies reactive with SARS-CoV-2 S.

(A) Detection of IgG, IgA, and IgM in five individuals from each indicated group. IgM levels are indicated by a heatmap. (B to D) Inhibition of SARS-CoV-2 S binding of sera from SARS-CoV-2–infected (SARS-CoV-2⁺, n = 10) or SARS-CoV-2–uninfected (SARS-CoV-2⁻ HCoV⁺, n = 6) patients by soluble S1 or S2. (B) Flow cytometry profile of one representative patient per group. (C) Mean frequency of positive cells. *P = 0.015; **P = 0.006, one-way analysis of variance (ANOVA) on ranks. (D) Mean staining intensity [mean fluorescence intensity (MFI) of sample as a percentage of negative control MFI]. In (C) and (D), dots represent individual samples from one of three similar experiments.

Fig. 2. Prevalence of SARS-CoV-2 S–cross-reactive antibodies detected by different methods.

(A) Flow cytometry and ELISA results for each sample in cohorts A and C to E listed in table S1. (B) Flow cytometry and ELISA results for serum samples from SARS-CoV-2–uninfected pregnant women. (C to E) SARS-CoV-2 S–cross-reactive antibodies in healthy children and adolescents. (C) Representative flow cytometry profiles of seronegative donors (Negative) or COVID-19 patients (Positive) and of SARS-CoV-2–uninfected adolescents with SARS-CoV-2 cross-reactive antibodies. (D) Frequency of cells stained with all three antibody classes (IgG⁺IgM⁺IgA⁺) or only with IgG (IgG⁺) ranked by their IgG⁺IgM⁺IgA⁺ frequency. The dashed line denotes the assay sensitivity cutoff. (E) Flow cytometry and ELISA results for each sample. (F) Prevalence of SARS-CoV-2 S–cross-reactive antibodies in the indicated age groups (line) and frequency of cells that stained only with IgG (dots) in all samples for which the date of birth was known. The heatmaps in (A), (B), and (E) represent the quartile values above each assay’s technical cutoff.

Fig. 3. Neutralization of SARS-CoV-2 S pseudotypes and authentic SARS-CoV-2 by SARS-CoV-2–infected and –uninfected patient sera.

(A) Inhibition of transduction efficiency of SARS-CoV-2 S or VSVg pseudotypes by adult COVID-19 patients who seroconverted (SARS-CoV-2⁺ Adults Ab⁺) or not (SARS-CoV-2⁺ Adults Ab⁻) and SARS-CoV-2–uninfected adult donors (SARS-CoV-2⁻ Adults Ab⁺) or children and adolescent donors (SARS-CoV-2⁻ Children/Adolescents Ab⁺) with SARS-CoV-2 S–binding antibodies. Each line is an individual serum sample. (B) Authentic SARS-CoV-2 neutralization titers of sera from the same donors as in (A), as well as SARS-CoV-2–uninfected donors without SARS-CoV-2 S–binding antibodies (Ab⁻). Dots represent individual samples. *P = 0.037; **P = 0.014; ns, not significant by one-way ANOVA on ranks.

Fig. 4. Mapping of cross-reactive epitopes in SARS-CoV-2 S.

(A) Signal intensity for each overlapping peptide along the length of SARS-CoV-2 S covered in the peptide arrays using pooled sera with (Ab⁺) or without (Ab⁻) flow cytometry–detectable SARS-CoV-2 S–reactive antibodies. Differentially recognized peaks are boxed. (B) Alignment of the amino acid sequences of SARS-CoV-2 and HCoV S glycoproteins. Boxes indicate predicted core epitopes. (C) Mapping of predicted epitopes targeted on the trimeric SARS-CoV-2 spike. The S1 (blue) and S2 (pink) subunits of one monomer are colored. Epitopes are shown for one monomer; the circled dashed line represent the membrane proximal region not present in the structure. (D) Left: Reactivity with the S glycoproteins of each HCoV of the indicated sera with (Ab⁺) or without (Ab⁻) flow cytometry–detectable SARS-CoV-2 S–reactive antibodies as determined by flow cytometry. Each column is an individual sample. Rows depict the staining for each antibody class. Right: Correlation coefficients between percentages of IgG staining for SARS-CoV-2 S and IgG, IgM, and IgA staining for each HCoV S glycoprotein.