High-affinity, neutralizing antibodies to SARS-CoV-2 can be made without T follicular helper cells

Abstract

T follicular helper (T_FH) cells are the conventional drivers of protective, germinal center (GC)–based antiviral antibody responses. However, loss of T_FH cells and GCs has been observed in patients with severe COVID-19. As T cell–B cell interactions and immunoglobulin class switching still occur in these patients, noncanonical pathways of antibody production may be operative during SARS-CoV-2 infection. We found that both T_FH-dependent and -independent antibodies were induced against SARS-CoV-2 infection, SARS-CoV-2 vaccination, and influenza A virus infection. Although T_FH-independent antibodies to SARS-CoV-2 had evidence of reduced somatic hypermutation, they were still high affinity, durable, and reactive against diverse spike-derived epitopes and were capable of neutralizing both homologous SARS-CoV-2 and the B.1.351 (beta) variant of concern. We found by epitope mapping and B cell receptor sequencing that T_FH cells focused the B cell response, and therefore, in the absence of T_FH cells, a more diverse clonal repertoire was maintained. These data support an alternative pathway for the induction of B cell responses during viral infection that enables effective, neutralizing antibody production to complement traditional GC-derived antibodies that might compensate for GCs damaged by viral inflammation.

Figure 1:. SARS-CoV-2 infection and vaccination induce Tfh-dependent and -independent antibodies.

(A) Schematic of experimental design for assessing serum antibody responses against SARS-CoV-2 in AAV-hACE2 mice. Mice were infected intranasally with SARS-CoV-2 (isolate USA-WA1/2020) two weeks following intratracheal administration of AAV-hACE2. Sera were collected 14 days post infection (dpi) for quantification of antibody titers by ELISA. (B) Spike (S)-specific total IgG, IgG1, IgG2b, IgG2c, and IgG3 antibody titers in sera from control (Bcl6^fl/fl; blue symbol), Tfh-deficient (Bcl6^fl/flCd4^Cre; magenta symbol), or CD4⁺ T cell-deficient (Ciita^−/−; green symbol) mice at 14 dpi with SARS-CoV-2. (C) Schematic of experimental design for assessing serum antibody responses against SARS-CoV-2 in K18-hACE2 mice. Mice were infected with SARS-CoV-2 (isolate USA-WA1/2020) by intranasal administration. Sera were collected at 14 dpi for quantification of antibody titers by ELISA. (D) S-specific total IgG, IgG1, IgG2b, IgG2c, and IgG3 antibody titers in sera from K18-hACE2 Bcl6^fl/fl (blue) or K18-hACE2 Bcl6^fl/flCd4^Cre (magenta) mice at 14 dpi with SARS-CoV-2. (E) Schematic of experimental design for assessing serum antibody responses against SARS-CoV-2 mRNA vaccination. Mice were vaccinated intramuscularly with Moderna mRNA-1273 or Pfizer-BioNTech BNT162b2 mRNA vaccine. Sera were collected 14 days post vaccination (dpv) for quantification of antibody titers by ELISA. (F) S-specific total IgG, IgG1, IgG2b, IgG2c, and IgG3 antibody titers in sera from Bcl6^fl/fl (blue), Bcl6^fl/flCd4^Cre (magenta), or Ciita^−/− (green) mice at 14 dpv with mRNA vaccine. LOD, limit of detection of the assay. Statistical significance was assessed by one-way ANOVA with Tukey’s test, or Welch’s t-test with Bonferroni multiple hypothesis correction when sample variances were 0 (B and F); two-tailed unpaired t-test or Welch’s t-test, based on the F test for unequal variance (D). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. ns, not significant. Data are expressed as mean ± standard error of mean (SEM) log₁₀ arbitrary units (AU). Each symbol represents an individual mouse. Data are aggregated from at least two independent experiments with a total of seven to ten mice per condition.

Figure 2:. Influenza virus infection induces Tfh-dependent and -independent antibodies.

(A) Schematic of experimental design for assessing serum antibody responses against influenza virus. Mice were infected with mouse-adapted influenza virus (PR8) by intranasal administration. Sera were collected at 14 dpi for quantification of antibody titers by ELISA. (B and C) PR8-specific total IgG (B) and IgM (C) antibody titers in sera from Bcl6^fl/fl (blue), Bcl6^fl/flCd4^Cre (magenta), or Ciita^−/− (green) mice at 14 dpi with PR8. (D) PR8-specific IgG1, IgG2b, IgG2c, and IgG3 antibody titers at 14 dpi with PR8. LOD, limit of detection of the assay. Statistical significance was assessed by one-way ANOVA with Tukey’s test or Dunnett’s test, or Welch’s t-test with Bonferroni multiple hypothesis correction when sample variances were 0 (B to D). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. ns, not significant. Data are expressed as mean ± SEM log₁₀ arbitrary units (AU). Each symbol represents an individual mouse. Data are aggregated from two independent experiments with a total of six to eight mice per condition.

Figure 3:. Durable, high-affinity antibodies to SARS-CoV-2 can be made in the absence of Tfh cells.

(A) Affinity index of serum IgG antibodies against SARS-CoV-2 S (left) or RBD (right), calculated as relative ELISA signal after urea wash. Sera are from AAV-hACE2 Bcl6^fl/fl (blue), Bcl6^fl/flCd4^Cre (magenta), or Ciita^−/− (green) mice infected with SARS-CoV-2 at 14 dpi. Fold changes relative to Bcl6^fl/fl mice are annotated. (B) Affinity index of serum IgG antibodies against SARS-CoV-2 S (left) or RBD (right). Sera are from K18-hACE2 Bcl6^fl/fl (blue) or K18-hACE2 Bcl6^fl/flCd4^Cre (magenta) mice infected with SARS-CoV-2 at 14 dpi. Fold changes relative to K18-hACE2 Bcl6^fl/fl mice are annotated. (C) Affinity index of serum IgG antibodies against PR8 (left) or HA (right), calculated as relative ELISA signal after urea wash. Sera are from Bcl6^fl/fl (blue), Bcl6^fl/flCd4^Cre (magenta), or Ciita^−/− (green) mice infected with PR8 at 14 dpi. Fold changes relative to Bcl6^fl/fl mice are annotated. (D) Longitudinal dynamics of S-specific IgG antibodies up to 84 dpi, in sera from K18-hACE2 Bcl6^fl/fl or K18-hACE2 Bcl6^fl/flCd4^Cre mice infected with SARS-CoV-2. (E) Change in S-specific IgG antibody titer per day between 28 and 84 dpi, as determined by linear regression (left). Relative titers of S-specific IgG at 84 dpi, compared to S-specific IgG at 28 dpi (right). (F) ELISpot quantification of S-specific IgG antibody-secreting cells (ASC) in the bone marrow of K18-hACE2 Bcl6^fl/fl or K18-hACE2 Bcl6^fl/flCd4^Cre mice at 85 dpi with SARS-CoV-2. S-specific IgG ASC per femur + tibia (left), per 10⁷ bone marrow (BM) cells (middle), or per 10⁵ CD138⁺ bone marrow plasma cells (BMPC) (right). Dotted line indicates background signal from naïve mice. (G) Total number of BM cells (left) and CD138⁺ BMPC (right) in femur + tibia of K18-hACE2 Bcl6^fl/fl or K18-hACE2 Bcl6^fl/flCd4^Cre mice at 85 dpi with SARS-CoV-2. Statistical significance was assessed by one-way ANOVA with Dunnett’s test, or Welch’s t-test with Bonferroni multiple hypothesis correction when sample variances were 0 (A and C); two-tailed unpaired t-test or Welch’s t-test, based on the F test for unequal variance (B, E, G); two-tailed Mann–Whitney test (F). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. ns, not significant. Data are expressed as mean ± SEM. Each symbol in (A to C and E to G) represents an individual mouse. Each symbol in (D) represents the mean of nine (Bcl6^fl/fl) and seven (Bcl6^fl/flCd4^Cre) mice. Data are aggregated from at least two independent experiments with a total of five to fifteen mice per condition.

Figure 4:. Tfh-deficient mice demonstrate similar V gene usage but impaired mutation selection compared to Tfh-sufficient mice.

(A) Pie charts detailing the relative proportions of B cell clones identified in K18-hACE2 Bcl6^fl/fl (top row) or K18-hACE2 Bcl6^fl/flCd4^Cre (bottom row) mice infected with SARS-CoV-2 (n = 4 mice per condition from one experiment). The 10 largest clones in each sample are shown individually, with all other clones grouped together. The total number of clones in each sample is annotated above each pie chart. (B) Tukey boxplots detailing Shannon entropy (left) and Simpson’s diversity index (right) in Bcl6^fl/fl or Bcl6^fl/flCd4^Cre mice. Statistical significance was assessed by two-tailed unpaired Welch’s t-test. Each symbol represents an individual mouse BCR repertoire. (C) Tukey boxplots detailing V gene usage frequencies, calculated on the level of total BCR sequences in each sample. V genes with ≥ 0.02 usage frequency in ≥ 2 samples are shown here. Statistical significance was assessed by two-tailed unpaired Mann-Whitney test. *P < 0.05. ns, not significant. (D) Alignment of human IGHV3-53, IGHV1-58, IGHV3-30-3 segments to the 3 murine V segments shown in (C) with the highest homology to each of the human V segments. The percentage sequence identity relative to the corresponding human V gene (bolded) is indicated on the left. The FWR and CDR regions within the V segments are annotated below. (E) Violin plots of nucleotide mutation frequencies in IgG-predominant B cell clones from Bcl6^fl/fl or Bcl6^fl/flCd4^Cre mice. Red lines indicate median nucleotide mutation frequency. Statistical significance was assessed by two-tailed unpaired Mann-Whitney test. (F) Quantification of the percentage of IgG-predominant B cell clones with either 1+ or 2+ mutations in Bcl6^fl/fl or Bcl6^fl/flCd4^Cre mice. Statistical significance was assessed by two-tailed unpaired Welch’s t-test. Data are expressed as mean ± SEM. (G) Probability density function curves of selection strength in IgG-predominant B cell clones from Bcl6^fl/fl or Bcl6^fl/flCd4^Cre mice, as determined by BASELINe analysis. The density curves are separated by region within the V segment: CDRs (top) or FWRs (bottom). Selection was estimated using BASELINe focused test. Statistical differences between the probability density functions were assessed by a two-sample significance test with numerical integration.

Figure 5:. Tfh cells focus the antibody repertoire but are dispensable for broad coverage of SARS-CoV-2 epitopes.

(A) Sera from infected AAV-hACE2 Bcl6^fl/fl or Bcl6^fl/flCd4^Cre mice (n = 5 mice per condition from two independent experiments) were assayed for reactivity against a bacterial display library of overlapping linear epitopes tiling the SARS-CoV-2 proteome. (B) Tukey boxplots detailing Shannon entropy (left), Simpson’s diversity index (middle), and repertoire focusing index (right) in Bcl6^fl/fl or Bcl6^fl/flCd4^Cre mice. Statistical significance was assessed by two-tailed unpaired Welch’s t-test. Each symbol represents an individual mouse. (C) Volcano plot of differentially enriched linear epitopes in Bcl6^fl/fl vs Bcl6^fl/flCd4^Cre mice. Epitopes that are significantly enriched in Bcl6^fl/fl sera are colored in blue, while epitopes that are enriched in Bcl6^fl/flCd4^Cre sera are colored in magenta. Point labels describe the amino acids within the indicated SARS-CoV-2 protein from which the epitope was derived. Statistical significance was assessed by two-tailed DESeq2 Wald test, with Benjamini-Hochberg multiple hypothesis correction (adjusted p < 0.05). (D) Heatmap of differentially enriched linear epitopes in Bcl6^fl/fl vs Bcl6^fl/flCd4^Cre mice, from (C). Data are expressed as log₂-transformed normalized counts. The position of each epitope in relation to known protein domains is annotated. (E) Scatter plots of relative enrichment scores for linear epitopes derived from each SARS-CoV-2 protein, comparing Bcl6^fl/fl vs Bcl6^fl/flCd4^Cre sera. Data are expressed as average z-scores (log₂ normalized counts were scaled to z-scores within each sample, then averaged across Bcl6^fl/fl or Bcl6^fl/flCd4^Cre mice). The gray dashed line demarcates equivalence between Bcl6^fl/fl and Bcl6^fl/flCd4^Cre, while the red line denotes the linear regression model with 95% confidence intervals shaded in pink. Epitopes that were identified as differentially enriched are annotated as in (C). The number of epitopes derived from each SARS-CoV-2 protein is annotated in the top left.

Figure 6:. RBD-specific antibodies are generated in the absence of Tfh cells.

(A) Top: differential enrichment scores for spike (S)-derived linear epitopes, such that epitopes that are comparatively enriched in Bcl6^fl/fl mice are represented by positive values (blue), whereas epitopes that are relatively enriched in Bcl6^fl/flCd4^Cre mice are indicated by negative values (magenta). Linear epitopes are each indicated by their center position within the S protein. Bottom: annotation tracks detailing S protein domains, pan-human coronavirus (hCoV) conservation scores, and mutations found in select SARS-CoV-2 variants of concern. SP, signal peptide. NTD, N-terminal domain. RBD, receptor-binding domain. RBM, receptor-binding motif. FP1, fusion peptide 1. FP2, fusion peptide 2. HR1, heptad repeat 1. HR2, heptad repeat 2. TM, transmembrane domain. CP, cytoplasmic domain. (B) Relative enrichment scores for linear epitopes derived from RBD, comparing Bcl6^fl/fl vs Bcl6^fl/flCd4^Cre sera. Data are expressed as average z-scores in Bcl6^fl/fl or Bcl6^fl/flCd4^Cre mice, with SEM error bars and loess regression lines. The RBM is annotated below with a gray box. (C) Ratio of RBD-specific to S-specific IgG antibodies in sera from Bcl6^fl/fl or Bcl6^fl/flCd4^Cre mice at 14 dpi with SARS-CoV-2. Statistical significance was assessed by two-tailed unpaired t-test. ns, not significant. Data are expressed as mean ± SEM. Each symbol represents an individual mouse. Data are aggregated from three independent experiments with a total of eight to ten mice per condition.

Figure 7:. Tfh-dependent and -independent antibodies demonstrate similar neutralization potency against homologous SARS-CoV-2 as well as the B.1.351 variant of concern.

(A) Left: neutralizing titers (inverse of half maximal inhibitory concentration; NT50) of sera from AAV-hACE2 Bcl6^fl/fl (blue) or Bcl6^fl/flCd4^Cre (magenta) mice against VSV particles pseudotyped with homologous S protein (USA-WA1/2020, the same isolate used to infect the mice). Right: neutralization potency indices (NT50 divided by S-specific total IgG) of sera from Bcl6^fl/fl or Bcl6^fl/flCd4^Cre mice against USA-WA1/2020 pseudovirus. Samples with NT50 at the limit of detection (LOD) were assigned a neutralization potency index of 0.002. (B) Left: neutralizing titers of sera from AAV-hACE2 Bcl6^fl/fl (blue) or Bcl6^fl/flCd4^Cre (magenta) mice against VSV particles pseudotyped with variant S protein (B.1.351). Right: neutralization potency indices of sera from Bcl6^fl/fl or Bcl6^fl/flCd4^Cre mice against B.1.351 pseudovirus. Samples with NT50 at the LOD were assigned a neutralization potency index of 0.0002. (C) Matched comparison of neutralizing titers against homologous (USA-WA1/2020) vs variant (B.1.351) pseudovirus. Fold changes are indicated above each group. (D) Schematic of experimental design for assessing in vivo neutralization potency of sera. (E and F) Viral burden in lungs from K18-hACE2 mice infected with SARS-CoV-2 at 4 dpi. Mice were treated with serum from naïve, infected AAV-hACE2 Bcl6^fl/fl, or infected AAV-hACE2 Bcl6^fl/flCd4^Cre mice. Bcl6^fl/fl sera was given undiluted or diluted (dil.) 7- to 9-fold to match the S-specific IgG titer of Bcl6^fl/flCd4^Cre sera in each experiment. Data are expressed as log₁₀ plaque forming units (PFU) per lung lobe by plaque assay (E) or log₁₀ N1 gene copy number by qPCR, normalized to Actb (F). Statistical significance was assessed by two-tailed Mann–Whitney test (A and B), two-tailed Wilcoxon signed-rank test (C), or one-way ANOVA with Tukey’s test (E and F). **P < 0.01; ***P < 0.001; ****P < 0.0001. ns, not significant. Data are expressed as mean ± SEM. Each symbol represents an individual mouse. Data are aggregated from at least two independent experiments with a total of eight to eighteen mice per condition.