Diverse priming outcomes under conditions of very rare precursor B cells

Abstract

Rare naive B cells have special pathogen-recognition features that enable outsized contributions to protective immunity but infrequently participate in immune responses. We investigatee how germline-targeting vaccine delivery and adjuvant selection affect priming of exceptionally rare BG18-like HIV broadly neutralizing antibody-precursor B cells (<1-in-50 million) in non-human primates. Only escalating dose (ED) priming immunization using the saponin adjuvant SMNP elicited detectable BG18-like cells in germinal centers (GCs) compared with other conditions. All groups had strong GC responses, but only ED+SMNP and bolus+SMNP induced BG18-like memory B cells in >50% of animals. One group had vaccine-specific GC responses equivalent to ED+SMNP but scarce BG18-like B cells. Following homologous boosting, BG18-like memory B cells were present in a bolus priming group but with lower somatic hypermutation and affinities than ED+SMNP. This outcome inversely associated with post-prime antibody titers, suggesting antibody feedback significantly influences rare precursor B cell responses. Thus, antigen and inflammatory stimuli extensively impact priming and affinity maturation of rare B cells.

Keywords: BG18; HIV bnAbs; HIV vaccines; adjuvants; bnAb precursors; escalating dose; germline-targeting; naive B cells.

Figure 1:. Study schematic showing immunization groups and sampling timepoints

Figure 2:. ED+SMNP prime larger GC responses.

A) Flow cytometry gating of B_GC cells (CD38⁻CD71⁺), B) frequency of total B cells (CD20⁺), and C) cumulative frequencies post-prime (weeks 3–9) based on area-under-the-curve (AUC) of B. Full gating shown in Figure S4A. D) Flow cytometry gating of GC-T_FH cells (PD-1^HiCXCR5⁺), E) frequency of total CD4⁺ cells, and F) cumulative frequencies post-prime (AUC of E). G) Flow cytometry gating of antigen-specific B_GC cells (N332-GT5-AF647⁺N332-GT5-BV421⁺;N332-GT5⁺⁺), H) frequency of total B cells, and I) cumulative frequencies post-prime (AUC of H). J) Flow cytometry gating of epitope-specific B_GC cells (N332-GT5-AF647⁺N332-GT5-BV421⁺N332-GT5KO-PE⁻), K) frequency of total B cells, and I) cumulative frequencies post-prime (AUC of K). Triangles (B,E,H,K) represent immunizations. Mean and SEM (B,E) or geometric mean and SD (H,K) are plotted in longitudinal figures. Median is plotted in all per animal figures. Statistical significance was tested using unpaired two-tailed Mann-Whitney tests as described in Methods, NS: p>0.1. Gray regions (H,K) represent B_GC LOD. See also Figure S4.

Figure 3:. ED+SMNP leads to a larger, more diverse, composition of BCRs 3-weeks post-prime.

A) Frequency of BG18 type I B_GC from week 3 among total B cells. Table indicates total and number of BG18 type I. Definitions shown in Figure S5A. B) Clonal richness of the B_GC BCRs plotted as Chao1 index. C) Clonal abundance curves for each group, and D) cumulative abundance of the top-5 clones per animal. E) Ig isotype distribution from each group. (A,B,D) lines represent medians. Statistical significance was tested using unpaired two-tailed Mann-Whitney tests as described in Methods. In A) multiple Fisher’s exact tests were used to compare each group to G1, NS: p>0.1. See also Figure S5.

Figure 4:. Vaccine-elicited T cell and serum IgG responses.

A) Flow cytometry gating of AIM assay performed at week 2 using N332-GT5;(Env) overlapping peptide pools or a DMSO control (unstimulated). Full gating shown in Figure S6A, and B) AIM⁺(CD40L⁺OX40⁺) T cell frequency of total CD4⁺ T cells. C) Flow cytometry gating of AIM⁺cT_FH, and D) frequency of total CD4⁺cT_FH cells. Red dots are Env+. E) Flow cytometry gating of AIM⁺ICS⁺(CD40L⁺IL21⁺), and F) frequency of total CD4⁺ T cells. G) Flow cytometry gating of AIM⁺ICS⁺(CD69⁺IFNg⁺), and H) frequency of total CD8⁺ T cells. I) Longitudinal area-under-the-curve (AUC) of antigen-specific serum IgG, and J) AUC at week 10 per animal. K) Longitudinal AUC of epitope-specific serum IgG, and L) AUC at week 10 per animal. Mean and SD plotted in (I,K), all others are median. Statistical significance was tested using unpaired two-tailed Mann-Whitney tests as described in Methods, NS: p>0.1. Asterixis in (B,D,F,H) indicate statistical significance compared to the pre-immunizations samples: NS>0.05, *<0.05 ,**<0.01, ***<0.001, ****<0.0001. See also Figure S6 and S7.

Figure 5:. Antigen-specific B_mem responses.

A) Flow cytometry gating of antigen-specific B_mem, and B) longitudinal frequency of total B cells. Full gating shown in Figure S8A. C) Flow cytometry gating of epitope-specific B_mem, and D) longitudinal frequency of total B cells. E) Week 10 and 12 antigen-specific B_mem frequency by animal, and F) fold change from week 10 to 12. G) Week 10 and 12 epitope-specific B_mem frequency by animal, and H) fold change from week 10 to 12. I) Frequency of BG18 type I B_mem among total B cells at week 12. Table indicates total and number of BG18 type I. J) Clonal richness of the B_mem from week 12. K) Clonal abundance curves for each group, and L) cumulative abundance of the top-5 clones per animal. M) Frequency of IGHD3–41 usage in IgM⁺ naive B cells, plotted by genotype. Lines represent medians. Statistical significance was tested using unpaired two-tailed Mann-Whitney tests as described in Methods, NS: p>0.1. Gray regions (B,D,E,G) represent B_mem LOD. See also Figures S7,S8,S9.

Figure 6:. Level of antigen-specific circulating IgG predict boost outcomes

A) AUC of antigen-specific serum IgG at week 12, and B) fold change from week 10 to 12. C) AUC of epitope-specific serum IgG at week 12, and D) fold change from week 10 to 12. E) Antigen-specific IgG fold change and week 10 IgG AUC correlation. F) Antigen-specific B_mem fold change from week 10 to 12 and week 10 IgG AUC correlation. G) Number of antigen-specific BM-B_PC per million BM cells at weeks 10 and 16, and H) fold change from week 10 to 16. r and p-values (E,F) are from pearson correlation analysis. Statistical significance was tested using unpaired two-tailed Mann-Whitney tests as described in Methods, NS: p>0.1. See also Figure S7 and S10.

Figure 7:. ED+SMNP leads to more SHM and higher affinity for boosting candidates.

A) Number of BG18 type I clones and IGHV genes used per animal. B) Frequency of IGHV gene usage by BG18 type I clones. C) Heavy-chain nucleotide mutations per cell. Total and BG18 type I BCRs are shown with the number of sequences, median mutations, and percent of mutated sequences. D) Binding affinities (Apparent K_D) for a subset of BG18 type I BCRs to N332-GT5 and three potential boosting immunogens. Number and percent of binders are listed (G1: n = 8, G2: n = 14). Dotted line represents the LOD for this assay. Statistical significance (C,D) was tested between groups using unpaired two-tailed Mann-Whitney tests. See also Figure S11.