Systematic analysis of human antibody response to ebolavirus glycoprotein shows high prevalence of neutralizing public clonotypes

Abstract

Understanding the human antibody response to emerging viral pathogens is key to epidemic preparedness. As the size of the B cell response to a pathogenic-virus-protective antigen is poorly defined, we perform deep paired heavy- and light-chain sequencing in Ebola virus glycoprotein (EBOV-GP)-specific memory B cells, allowing analysis of the ebolavirus-specific antibody repertoire both genetically and functionally. This approach facilitates investigation of the molecular and genetic basis for the evolution of cross-reactive antibodies by elucidating germline-encoded properties of antibodies to EBOV and identification of the overlap between antibodies in the memory B cell and serum repertoire. We identify 73 public clonotypes of EBOV, 20% of which encode antibodies with neutralization activity and capacity to protect mice in vivo. This comprehensive analysis of the public and private antibody repertoire provides insight into the molecular basis of the humoral immune response to EBOV GP, which informs the design of vaccines and improved therapeutics.

Keywords: CP: Immunology; Ebola; adaptive immunity; antibodies; human; monoclonal; repertoire.

Figure 1.. Identification and diversity of EBOV-GP-specific memory B cells

(A) Schematic of sample processing to identify and sort memory B cells. (B) Schematic of flow cytometric staining to identify EBOV-GP-specific B cells. (C) Gating for lymphocytes, singlets, and live cells using DAPI, followed by class-switched B cells. Cells were stained with anti-CD19 antibody conjugated to phycoerythrin (PE) and anti-IgM and anti-IgD conjugated to fluorescein isothiocyanate (FITC). EBOV GP was biotinylated and conjugated to streptavidin allophycocyanin (APC). Fluorescence-activated cell sorting (FACS) isolation of class-switched B cells (CD19⁺IgM⁻IgD⁻) specific to the EBOV GP (antigen-APC) in a donor that has not previously been exposed to EBOV (left) or the convalescent donor (right) is shown. (D) Diversity metrics calculated for the EBOV-GP-specific repertoire (purple) compared with the non-antigen-specific repertoire (gray). The first plot shows species richness and the second shows Chao diversity. The sample depth on the x axis indicates the number of sequences, and the unique species on the y axis indicates the number of clonal families. Additional diversity metrics were calculated, including Shannon entropy and Simpson index. (E) Variable gene usage in the heavy-chain (top) or light-chain (bottom) repertoire from sequencing. The number of sequences using each gene was calculated and normalized to a percentage using the total number of sequences as 100%. Purple dots indicate the EBOV-GP-specific repertoire, gray dots indicate the non-antigen-specific repertoire.

Figure 2.. Analysis of the clonally expanded EBOV-GP-specific repertoire

(A) Schematic of identification of clonally expanded families and selection of one clone per family. (B) Graph showing the distribution of clonal families. After clustering was completed, clusters were ordered from the largest to the smallest cluster and then plotted in that order as the percentage of the EBOV-GP-specific repertoire (purple) and non-antigen-specific repertoire (gray). (C) Network plot of the clonally expanded repertoire with the functional characteristics of each clonal family plotted and schematic showing how calculations were derived to construct a network diagram. Reactivity to each glycoprotein in ELISA is denoted by different colors. Different shades of each color indicate neutralizing capacity, with the darker dots indicating neutralizing antibodies for VSV-EBOV and the lighter dots indicating non-neutralizing antibodies. Different shades within each color represent whether the antibody preferably bound to the intact GP (GP_ecto) or the cleaved GP (GP_cl) or if it bound well to both.

Figure 3.. Characteristics of plasma antibodies and unmutated common ancestors of clonally expanded antibodies

(A) Venn diagram detailing the identification of antibodies present in both plasma and the memory B cell repertoire. (B) Characteristics of antibodies present in both the plasma and the memory B cell repertoire that originated from clonally expanded families. The number of antibodies present in the clonal family is shown in the first column, followed by blue color indicating binding reactivity in ELISA to the different GPs, followed by binding to cleaved or intact GP, followed by neutralization for VSV-EBOV indicated in green. The empty box in the cleaved/intact preferential binding column indicates equivalent binding in that assay. Experiments were performed in biological duplicate, and the compilation of replicates is shown. (C) Antibodies and their inferred unmutated common ancestors (UCAs) with the functional profile of each antibody. Gene use is listed in the first column followed by antibody names. Antibodies are listed with the mutated version of the antibody on the top row and the UCA version on the bottom. The blue boxes indicate binding in ELISA to the different GPs. The gray boxes indicate percentage blocking in competition-binding ELISA against biotinylated EBOV-515, a base-region-specific reference antibody, or against the glycan cap reference antibody 13C6. The green boxes indicate neutralization for VSV-EBOV, -BDBV, or -SUDV. The numbers inside the boxes indicate the IC₅₀ value for each antibody. Experiments were performed in biological duplicates and technical triplicates with similar results. A biological replicate from a single experiment is shown. (D) Neutralization curves of UCA antibodies (dotted lines) that retained cross-reactive neutralization and their mutated counterparts (solid lines) against VSV-EBOV, -BDBV, or -SUDV. Experiments were performed in biological duplicates and technical triplicates with similar results. A biological replicate from a single experiment is shown. (E) Maximum-likelihood phylogenetic tree of the EBOV-879 lineage. The inferred UCA is indicated in the orange circle on the heavy-chain and light-chain trees. Blue lines indicate antibody sequences that were found in paired chain sequencing; black lines indicate sequences that were found in unpaired chain bulk sequencing that clustered with the clonal family. (F) Maximum-likelihood phylogenetic tree of the EBOV-872 lineage. The inferred UCA is indicated in the orange circle on the heavy-chain and light-chain trees. Blue lines indicate those antibodies found in paired sequencing; black lines are bulk sequences that clustered with the clonal family. (G) Maximum-likelihood phylogenetic tree of the EBOV-967 lineage. The inferred UCA is indicated in the orange circle on the heavy-chain and light-chain trees. Blue lines indicate those antibodies found in paired sequencing; black lines are bulk sequences that clustered with the clonal family. (H) Maximum-likelihood phylogenetic tree of the EBOV-591 lineage. The inferred UCA is indicated in the orange circle on the heavy-chain and light-chain trees. Blue lines indicate those antibodies found in paired sequencing; black lines are bulk sequences that clustered with the clonal family.

Figure 4.. Identification of public clonotypes

(A) Clonal overlap between vaccinated (green) and convalescent (blue) donors. Numbers inside the first outer circle indicate the number of sequences that were identified as public clonotypes from the respective donor. The light gray color shows the distribution of and the median CDR3 length. The dark gray color shows the distribution and median number of somatic mutations of the public antibodies from that donor. (B) Heavy- and light-chain variable gene usage combinations for all public clonotypes identified. Numbers inside boxes indicate the number of public clonotypes using that gene combination. Public clonotype groups using highly used genes are listed on the right. Blue indicates binding to GPs in ELISA, purple indicates binding via cell-surface GP display assay, pink indicates non-neutralizing, and green indicates neutralizing. (C) Functional profile of each of the 73 public clonotypes after expression and functional testing of the 294 public clonotype antibodies and inferring a functional profile for each of the public clonotype groups. Blue indicates binding to GPs in ELISA, purple indicates binding in a cell-surface GP display assay, pink indicates non-neutralizing, and green indicates neutralizing. Experiments were performed in biological duplicates. A compilation of the average of all experiments is shown.

Figure 5.. Properties of neutralizing public clonotypes

(A) Table showing all 15 neutralizing public clonotypes. The first column identifies the public clonotype group number, the second column details the variable gene usage. The third column indicates clone name, with all the public clonotype antibodies in the group indicated in white and the germline revertant version of that group’s antibody in yellow. Blue boxes indicate ELISA binding. Gray indicates percentage blocking in a competition-binding assay. Green indicates neutralization for VSV-EBOV, -BDBV, or -SUDV, with the IC₅₀ values written inside the boxes. Clone name highlighted in yellow at the bottom row of each section is the germline revertant version of that public clonotype and its respective functionality. (B) Authentic virus neutralization curve for a representative antibody of each public clonotype group. (C) Neutralization curves from antibodies that retained cross-reactive neutralization at the germline level. Group 2.23 antibodies to VSV-EBOV and -BDBV are shown on the left and group 2.22 antibodies to VSV-EBOV and -BDBV are on the right. Dotted lines in each graph indicate the germline revertant antibody curve. Solid lines in varying colors indicate the matured versions of the antibodies in that public clonotype group.

Figure 6.. Epitopes targeted by neutralizing public clonotypes

(A) Competition-binding ELISA results for antibodies within each public clonotype group in competition with one another. Unlabeled blocking antibodies applied to the GP antigen first are listed across the top of each grid, while the biotinylated antibodies that were added to the antigen-coated wells second are indicated on the left. The number under each box represents the percentage of uncompeted binding of the biotinylated antibody in the presence of the indicated competed antibody. The experiment was performed in biological duplicates and technical triplicates with similar results. A biological replicate from a single experiment is shown. (B) Negative-stain EM of EBOV GP in complex with Fab forms of different antibodies. Three-dimensional reconstructions are shown. The Fab (blue) is docked to a trimer of the EBOV GP (gray). (C) Critical binding residues for EBOV-852, EBOV-598, EBOV-786, EBOV-709, and EBOV-823 as determined by loss of binding in alanine-scanning GP mutagenesis studies. Residues where alanine caused a loss of binding for antibodies are indicated in green.

Figure 7.. Public clonotypes infer protection in vivo

(A) Mice (n = 5) were treated i.p. with 100 μg (~5 mg/kg) of an individual antibody per mouse on day 1 post-challenge with EBOV. mAb DENV 2D22 (specific to dengue virus) served as a negative control. Mice were monitored twice daily from day 0 to day 14 post-challenge for survival and monitored daily from day 15 to day 28 as described previously. (B) Authentic virus neutralization curves for the two most potent cross-reactive neutralizing antibodies, EBOV-786 (blue) and EBOV-852 (red), for EBOV or BDBV. (C) Mice (n = 5) were treated i.p. with either 100 μg (~5 mg/kg) of individual antibody per mouse for EBOV challenge (left) or 500 μg (~25 mg/kg) of an individual antibody per mouse for BDBV challenge (right) on day 1 post-challenge. Human antibody DENV 2D22 (specific to dengue virus) served as a negative control. Mice were monitored twice daily from day 0 to day 14 post-challenge for survival and monitored daily from day 15 to day 28 as described previously. EBOV-785 is in blue, and EBOV-852 is in red.