Research ArticleCORONAVIRUS

Enhanced SARS-CoV-2 neutralization by dimeric IgA

See allHide authors and affiliations

Science Translational Medicine  20 Jan 2021:
Vol. 13, Issue 577, eabf1555
DOI: 10.1126/scitranslmed.abf1555
  • Fig. 1 Patients with COVID-19 have plasma IgA antibodies that recognize SARS-CoV-2 RBD.

    (A) ELISAs were used to measure plasma IgA reactivity to the SARS-CoV-2 RBD. The graph shows optical density units at 450 nm (OD) and reciprocal plasma dilutions. Negative controls in black; individuals 21, 47, and 96 in blue, orange, and green lines and arrowheads, respectively (11). (B) The normalized area under the curve (AUC) values for 8 controls and each of 149 individuals in the cohort were plotted. Horizontal bar indicates mean values. Black dots indicate the individuals that are 2 SDs over the mean of controls, and gray dots represent the individuals below the same parameter. (C) The duration of symptoms in days was plotted against normalized AUC for plasma IgA binding to RBD. (D) Subjective symptom severity was plotted against the normalized AUC for IgA binding to RBD. (E) Symptom onset to time of sample collection in days was plotted against normalized AUC for plasma IgA anti-RBD. (F) Age was plotted against normalized AUC for plasma IgA anti-RBD. (G) Normalized AUC of plasma anti-RBD IgG ELISA plotted against the normalized AUC for plasma IgA anti-RBD. (H) The normalized AUC of anti-RBD IgA ELISA was plotted for males (n = 83) and females (n = 66). (I) The normalized AUC of anti-RBD IgA ELISA was plotted for outpatients (n = 138) and hospitalized (n = 11) individuals. (J) The normalized AUC of plasma anti-RBD IgA ELISA for all cases (n = 111) and contacts (n = 38) in the cohort was plotted. (K and L) The normalized AUC of anti-RBD IgA (K) or IgG (L) ELISA for patients with gastrointestinal (GI) symptoms (n = 32) and without GI symptoms (n = 117) was plotted. The r and P values for the correlations in (C) to (G) were determined by two-tailed Spearman’s correlations. For (H) to (L), horizontal bars indicate median values. Statistical significance was determined using a two-tailed Mann-Whitney U test.

  • Fig. 2 SARS-CoV-2 pseudovirus is neutralized by purified IgA and IgG.

    Neutralization activity of plasma-purified IgG and IgA from 99 participants was measured in cell lysates of HT1080ACE2cl.14 cells 48 hours after infection with pNL4-3ΔEnv-nanoluc–based SARS-CoV-2 pseudovirus. (A and B) The normalized AUC for plasma IgG (A) or IgA (B) anti-RBD ELISA was plotted against purified IgG (A) or IgA (B) pseudovirus neutralization 1/IC50 values. Individuals 21, 47, and 96 are indicated with blue, orange, and green arrowheads, respectively. (C and D) Published plasma neutralizing titer 50 (NT50) values (11) were plotted against purified IgG (C) or IgA (D) pseudovirus neutralization 1/IC50 values. (E) Purified IgA pseudovirus neutralization IC50 values were plotted against purified IgG pseudovirus neutralization IC50 values. (F) Purified IgA and IgG pseudovirus neutralization IC50 values were compared. (G) The plot depicts the ratio of pseudovirus neutralization IC50 values of purified IgG to IgA (n = 95). The r and P values in (A) to (E) and (G) were determined by two-tailed Spearman’s correlations. In (F), P values were determined by two-tailed Mann-Whitney U tests, and horizontal bars indicate median values. Samples for which purified IgA IC50 values could not be detected are not plotted, resulting in n = 95 for (B) and (D) to (G).

  • Fig. 3 Clinical parameters correlate with plasma IgA or IgG pseudovirus neutralization ability.

    (A and B) Symptom severity was plotted against purified IgG (A) and IgA (B) pseudovirus neutralization 1/IC50 values. Individuals 21, 47, and 96 are indicated with blue, orange, and green arrowheads, respectively. (C and D) Symptom onset to time of sample collection in days was plotted against purified plasma IgG (C) and IgA (D) pseudovirus neutralization 1/IC50 values. (E and F) Purified plasma IgG (E) and IgA (F) pseudovirus neutralizing IC50 values were compared for all outpatient (n = 90) and hospitalized (n = 9) participants in the cohort. (G and H) Purified IgG (G) and IgA (H) pseudovirus neutralization IC50 values for patients were compared for patients with GI symptoms (n = 21) and without GI symptoms (n = 74). The r and P values in (A) to (D) were determined by two-tailed Spearman’s correlations. In (E) to (H), P values were determined by two-tailed Mann-Whitney U tests, and horizontal bars indicate median values.

  • Fig. 4 Characterization of monoclonal anti–SARS-CoV-2 RBD IgM, IgG, and IgA antibodies.

    (A) B cells producing IgM, IgG, and IgA from three individuals—COV21, COV47, and COV96—were analyzed, and clonality was evaluated. The number in the inner circle indicates the number of sequences analyzed for the individual denoted. Pie slice size is proportional to the number of clonally related sequences. Colored pie slices indicate clones or singlets that share the same IGHV and IGLV genes and have highly similar CDR3s across isotypes. Gray indicates clones that are not shared. White indicates singlets that are not shared. The right side circos plots show the relationship between antibodies of different isotypes that share the same IGH V(D)J and IGL VJ genes and have highly similar CDR3s. Purple, green, and gray lines connect related clones, clones and singles, and singles to each other, respectively. (B) Circos plot shows sequences from all three individuals with clonal relationships depicted as in (A). (C) Sample sequence alignment for antibodies of different isotypes isolated from individual COV47 that display the same IGH V(D)J and IGL VJ genes and highly similar CDR3s. Amino acid differences in CDR3s to the reference sequence (bold) are indicated in red, and dots represent identical amino acids.

  • Fig. 5 Monoclonal IgA and IgM antibodies bind and neutralize SARS-CoV-2 in vitro.

    (A) Pseudovirus IC50 neutralization values for IgA and IgM monoclonal antibodies and IgG monoclonal antibodies from the same individuals (11). Antibodies with IC50 < 1000 ng/ml are shown. Orange lines indicate geometric mean. (B) RBD binding was quantified by biolayer interferometry. (C) The binding of a second antibody (Ab2) to preformed first antibody (Ab1)–RBD complexes was also quantified by biolayer interferometry. Dotted line denotes where Ab1 and Ab2 are the same, and Ab2 is color-coded as indicated. We tested selected IgA antibodies against preformed complexes representing antibodies that bind to different structurally determined epitopes (40) from class 1, 2, 3, or 4 (C144-, C121-, C135-, or CR3022-RBD, respectively).

  • Fig. 6 IgA dimers bind to RBD and neutralize authentic SARS-CoV-2 more potently than monomers in vitro.

    (A) Graphs depict binding affinity measurements of individual IgA monomers and corresponding dimers. (B) The dissociation constant (KD) values of monomers and dimers were compared. Horizontal lines indicate mean (n = 7). (C and D) Normalized relative luminescence values (RLU) for cell lysates of 293TACE2 cells after infection with SARS-CoV-2 pseudovirus (C) or normalized percentage of SARS-CoV-2–positive Vero E6 cells 48 hours after infection with SARS-CoV-2 authentic virus (D). Values obtained in the absence of antibody are plotted at x = 0.1 to be visible on a log scale in the presence of increasing concentrations of indicated monoclonal antibodies in their monomeric or dimeric form. Four-parameter nonlinear regression curve fits of normalized data are shown. (E) IC90 values were compared between monomer to dimers after normalization to number of antibody binding sites. For (B), Student’s t test was used.

Supplementary Materials

  • stm.sciencemag.org/cgi/content/full/13/577/eabf1555/DC1

    Fig. S1. Age, duration of symptoms, and sex do not correlate with plasma IgA or IgG neutralization ability.

    Fig. S2. Analysis of antibody somatic hypermutation.

    Fig. S3. Analysis of antibody CDR3 length.

    Fig. S4. Analysis of antibody CDR3 hydrophobicity.

    Fig. S5. Frequency distributions of human V genes.

    Fig. S6. Binding and neutralizing activity of anti–SARS-CoV-2 RBD IgA and IgM monomers.

    Fig. S7. Biolayer interferometry experiment.

    Fig. S8. Purification of dimeric IgA by size exclusion chromatography.

    Fig. S9. Neutralizing activity of monoclonal monomeric and dimeric IgA antibodies.

    Table S1. Inhibitory concentrations of monoclonal antibodies from isotype shared clones.

    Table S2. Inhibitory concentrations of monoclonal IgA monomers and dimers.

    Data file S1. Sequences of anti–SARS-CoV-2 antibodies.

    Data file S2. Sequences of antibodies from isotype shared clones.

    Data file S3. Sequences of cloned recombinant antibodies.

    Data file S4. Effective and inhibitory concentrations of monoclonal antibodies.

    Data file S5. Primers.

    Data file S6. Raw data for graphs with n < 20.

  • The PDF file includes:

    • Fig. S1. Age, duration of symptoms, and sex do not correlate with plasma IgA or IgG neutralization ability.
    • Fig. S2. Analysis of antibody somatic hypermutation.
    • Fig. S3. Analysis of antibody CDR3 length.
    • Fig. S4. Analysis of antibody CDR3 hydrophobicity.
    • Fig. S5. Frequency distributions of human V genes.
    • Fig. S6. Binding and neutralizing activity of anti–SARS-CoV-2 RBD IgA and IgM monomers.
    • Fig. S7. Biolayer interferometry experiment.
    • Fig. S8. Purification of dimeric IgA by size exclusion chromatography.
    • Fig. S9. Neutralizing activity of monoclonal monomeric and dimeric IgA antibodies.
    • Table S1. Inhibitory concentrations of monoclonal antibodies from isotype shared clones.
    • Table S2. Inhibitory concentrations of monoclonal IgA monomers and dimers.

    [Download PDF]

    Other Supplementary Material for this manuscript includes the following:

    • Data file S1 (Microsoft Excel format). Sequences of anti–SARS-CoV-2 antibodies.
    • Data file S2 (Microsoft Excel format). Sequences of antibodies from isotype shared clones.
    • Data file S3 (Microsoft Excel format). Sequences of cloned recombinant antibodies.
    • Data file S4 (Microsoft Excel format). Effective and inhibitory concentrations of monoclonal antibodies.
    • Data file S5 (Microsoft Excel format). Primers.
    • Data file S6 (Microsoft Excel format). Raw data for graphs with n < 20.
  • The PDF file includes:

    • Fig. S1. Age, duration of symptoms, and sex do not correlate with plasma IgA or IgG neutralization ability.
    • Fig. S2. Analysis of antibody somatic hypermutation.
    • Fig. S3. Analysis of antibody CDR3 length.
    • Fig. S4. Analysis of antibody CDR3 hydrophobicity.
    • Fig. S5. Frequency distributions of human V genes.
    • Fig. S6. Binding and neutralizing activity of anti-SARS-CoV-2 RBD IgA and IgM monomers.
    • Fig. S7. Biolayer interferometry experiment.
    • Fig. S8. Purification of Dimeric IgA by Size Exclusion Chromatography.
    • Fig. S9. Neutralizing activity of monoclonal monomeric and dimeric IgA antibodies.
    • Table S1. Inhibitory concentrations of monoclonal antibodies from isotype shared clones.
    • Table S2. Inhibitory concentrations of monoclonal IgA monomers and dimers.

    [Download PDF]

    Other Supplementary Material for this manuscript includes the following:

Stay Connected to Science Translational Medicine

Navigate This Article