Antibodies from primary humoral responses modulate the recruitment of naive B cells during secondary responses

Abstract

Vaccines generate high-affinity antibodies by recruiting antigen-specific B cells to germinal centers (GCs), but the mechanisms governing the recruitment to GCs on secondary challenges remain unclear. Here, using preclinical SARS-CoV and HIV mouse models, we demonstrated that the antibodies elicited during primary humoral responses shaped the naive B cell recruitment to GCs during secondary exposures. The antibodies from primary responses could either enhance or, conversely, restrict the GC participation of naive B cells: broad-binding, low-affinity, and low-titer antibodies enhanced recruitment, whereas, by contrast, the high titers of high-affinity, mono-epitope-specific antibodies attenuated cognate naive B cell recruitment. Thus, the directionality and intensity of that effect was determined by antibody concentration, affinity, and epitope specificity. Circulating antibodies can, therefore, be important determinants of antigen immunogenicity. Future vaccines may need to overcome-or could, alternatively, leverage-the effects of circulating primary antibodies on subsequent naive B cell recruitment.

Keywords: BG18; HIV; RBD; SARS-CoV; antibody; germinal center; humoral immunity.

Graphical abstract

Figure 1

Previously elicited high-affinity antibodies restrict naive BG18^gH B cell responses (A) Schematic of evaluation of naive BH18^gH B cell responses to immunization by N332-GT2 NP in primed and unprimed recipients. (B) Representative FACS plots of gating strategy to quantify FAS⁺CD38⁻ GC B cells and CD45.2⁺ BG18^gH GC B cell responses versus endogenous CD45.1⁺ GC B cell responses in unprimed (left) and primed (right) recipients. (C and D) (C) GC cells as the percentage of total B cells and (D) CD45.2⁺ BG18^gH cells as the percentage of total GC B cells at experiment day 52 (10 days post-immunization [dpi]) in previously unprimed (gray) or primed (red) recipients. (E) Experimental design to evaluate naive BH18^gH B cell responses in primed and unprimed MD4 recipients. (F and G) (F) GC cells as the percentage of total B cells and (G) GFP⁺ BG18^gH cells as the percentage of total GC B cells at experiment day 52 (10 dpi) in previously unprimed (gray) or primed (red) MD4 BG18^gH recipients. (H) N332-GT2 trimer binding serum IgG of unprimed (gray) and primed (red) MD4 BG18^gH recipients at experiment day 42 (prior to immunization) and experiment day 52 (10 dpi). AUC = area under the ELISA binding curve. (I) Schematic of evaluation of the effect of serum IgG from previously immunized mice on naive BG18^gH recipients consisting of (1) isolation of serum from N332-GT2 immunized cohort and (2) administration of purified IgG to the naive BG18^gH recipients. (J and K) (J) GC cells as the percentage of total B cells and (K) CD45.2⁺ BG18^gH cells as the percentage of total GC B cells at 10 dpi in BG18^gH recipients receiving 200 μg of IgG isolated from 7- and 28-dpi serum. (L and M) (L) GC cells as the percentage of total B cells and (M) CD45.2⁺ BG18^gH cells as the percentage of total GC B cells at 10 dpi in BG18^gH recipients receiving escalating doses of IgG isolated from 28-dpi serum. p values were calculated by unpaired Student’s t test (C, D, F, and G) or ordinary one-way ANOVA with Dunnett’s multiple comparisons (K and M) (^∗p < 0.05; ^∗∗p < 0.01; ^∗∗∗∗p < 0.0001; ns, not significant). Figures represent data from one of at least two experiments with 3–5 mice per condition, with data presented as mean ± SD. See also Figures S1–S3.

Figure 2

Naive CR3022Ma but not CR3022Gl B cell responses are enhanced in primed mice (A) (Top) Nested pie chart of CR3022Gl HC (light green), murine HC (dark gray), CR3022Gl LC (light purple), and murine LC (light gray) sequences amplified from single-cell sorted B220+ naive B cells from two CR3022GL mice. n = sequence pairs amplified. (Bottom) As top, for CR3022Ma. (B) Representative FACS plot of binding of naive CR3022Gl and CR3022Ma and WT B cells to SARS-CoV-1 RBD probes. (C) Schematic of evaluation of the effect of previous exposures on naive CR3022Gl and CR3022Ma B cell responses. (D) Representative FACS plots of gating strategy to quantify GC B cell responses as the percentage of total B cells and the percentage of CD45.2⁺ (here CR3022Gl/Ma) GC B cells of the total GC B cell population. (E) (Left) GC cells as the percentage of total B cells and (left-center) CR3022Gl GC B cells as percentage of total B cells at experiment day 40 (10 dpi) in previously unprimed (gray) or primed (red) CR3022Gl recipients. (Right-center) GC cells as the percentage of total B cells and (right) CR3022Ma GC B cells as the percentage of total B cells at experiment day 40 (10 dpi) in previously unprimed (gray) or primed (red) CR3022Ma recipients. (F) Nested pie chart of CC12.1 HC (teal), murine HC (dark gray), CC12.1 LC (pink), and murine LC (light gray) sequences amplified from single-cell sorted B220+ naive B cells from 2 CC12.1 mice. Center number indicates sequence pairs amplified. (G) Representative FACS plot of binding of naive CC12.1 and WT B cells to SARS-CoV-2 RBD probes. (H) Schematic of experimental evaluation of the effects of previous exposures on naive CC12.1 B cell responses in primed and unprimed recipients. (I) Representative FACS plots of gating strategy to quantify GC B cell responses as the percentage of total B cells and the percentage of CD45.2+ (here CC12.1) GC B cells of the total GC B cell population. (J) (Left) GC cells as the percentage of total B cells and (right) CC12.1 GC B cells as the percentage of total B cells at experiment day 37 (7 dpi) in previously unprimed (gray) or primed (red) CC12.1 recipients. p values were calculated by unpaired Student’s t test (^∗∗p < 0.01; ^∗∗∗p < 0.001; ns, not significant). Figures represent data from one experiment with 4–5 mice per condition, with data presented as mean ± SD. See also Figure S4.

Figure 3

SARS-CoV spike responses engage multiple epitopes, whereas N332-GT2 responses are highly epitope focused (A) Summary of nsEMPEM analysis of human serum samples from four patients naturally infected with SARS-CoV-2. (B) Summary of nsEMPEM analysis of NHP serum samples from four subjects after immunization by NVX-CoV2373. (C) Schematic of experimental design to evaluate serum in WT and BG18^gH-recipient mice 14 and 42 days post N332-GT2 immunization. (D) N332-GT2 binding serum IgG of WT B6 (gray) and BG18^gH recipients (WT+BG18^gH, red) at 14 and 42 dpi. AUC = area under the ELISA binding curve. p values were calculated by unpaired Student’s t test (^∗∗∗∗p < 0.0001; ns, not significant). Data from one experiment with 6–7 mice per condition and presented as mean ± SD. (E) Summary of nsEMPEM analysis of pooled mouse serum samples from (D). Composite models represent polyclonal antibody targeting against the N332-GT2 trimer. (Purple) Fabs targeting the base of the trimer; (green) Fabs targeting the V1/V3 region. See also Figure S5.

Figure 4

High-affinity mAbs restrict BG18^gH responses in an epitope-dependent manner (A) Schematic of experiment to evaluate the effects of i.v. administration of monoclonal antibodies on naive BH18^gH B cell responses in WT recipients. (B and C) (B) GC cells as the percentage of total B cells and (C) CD45.2⁺ BG18^gH cells as the percentage of total GC B cells at 10 dpi in BG18^gH recipients receiving 30 μg of either high-affinity murine IgG1 BG18 mAb (BG18_d42.10, green) or high-affinity human IgG1 BG18 mAb (BG18_iGL0, red). (D) CD45.2⁺ BG18^gH cells as the percentage of total GC B cells at 10 dpi of BG18^gH recipients receiving escalating concentrations of the high-affinity BG18_iGL0 mAb. (E) CD45.2⁺ BG18^gH cells as the percentage of total GC B cells at 10 dpi of BG18^gH recipients receiving 10 μg of BG18 mAbs with increasing affinity for the N332_GT2 immunogen (BG18_Pre5 with K_D 1.3μM, BG18_Pre14 with K_D 93 nM, and BG18_iGL0 with K_D 4 nM). (F) Schematic of the HIV Env trimer and binding sites of the bnAbs used in (G)–(J). (G and H) (G) GC cells as the percentage of total B cells and (H) CD45.2⁺ BG18^gH cells as the percentage of total GC B cells at 10 dpi in BG18^gH recipients receiving 10 μg of the BG18-epitope binding mAbs BG18_iGL0 and PGT128. (I and J) (I) GC cells as the percentage of total B cells and (J) CD45.2⁺ BG18^gH cells as the percentage of total GC B cells at 10 dpi in BG18^gH recipients receiving 10 μg of antibodies binding other epitopes on the N332-GT2 immunogen. p values were calculated by ordinary one-way ANOVA with Dunnett’s multiple comparisons (^∗p < 0.05; ^∗∗p < 0.01; ^∗∗∗p < 0.001; ^∗∗∗∗p < 0.0001; ns, not significant). Figures represent data from one of at least two experiments with 3–5 mice per condition, with data presented as mean ± SD.

Figure 5

High-affinity antibody restricts VRC01-precursor responses (A) Schematic of experiment to evaluate naive CLK09 B cell responses in primed and unprimed recipients. (B and C) (B) GC cells as the percentage of total B cells and (C) CD45.2⁺ CLK09 cells as the percentage of total GC B cells at experiment day 52 (10 dpi) in previously unprimed (gray) or primed (red) recipients. (D) Representative FACS plots at experiment day 52 (10 dpi) of gating strategy to quantify eOD-GT8 binding in the endogenous CD45.1⁺ GC B cell population. (E) eOD-GT8 binding of endogenous CD45.1⁺ GC B cells at experiment day 52 (10 dpi) in unprimed and primed hosts. (F) Schematic of experiment to evaluate the effects of i.v. administration of 10 μg of CLK mAbs with increasing affinity for the eOD-GT8 immunogen (CLK31 with K_D 790 nM, CLK09 with K_D 350 nM, and CLK20 with K_D 13 nM) on naive CLK09 and CLK19 B cell responses in WT recipients (G–J). (G and H) (G) GC cells as the percentage of total B cells and (H) CD45.2⁺ CLK09 cells as the percentage of total GC B cells at 10 dpi of CLK09 recipients. (I and J) (I) GC cells as the percentage of total B cells and (J) CD45.2⁺ CLK19 cells as the percentage of total GC B cells at 10 dpi of CLK19 recipients. p values were calculated by unpaired Student’s t test (B, C, and E) or ordinary one-way ANOVA with Dunnett’s multiple comparisons (G–J) (^∗p < 0.05; ^∗∗p < 0.01; ^∗∗∗p < 0.001; ^∗∗∗∗p < 0.0001; ns, not significant). Figures represent data from one of at least two experiments with 3–5 mice per condition, with data presented as mean ± SD. See also Figure S6.

Figure 6

High-affinity antigen-specific antibody can restrict corresponding B cell populations from participating during co-evolution (A) Schematic of experiment to test the antibody-mediated restriction of immunogen-specific responses during co-evolution. (B) Representative FACS plots at 10 dpi of gating strategy to identify the effects of administration of no antibody, 10 μg of CLK09, or BG18_iGL0 mAb in BG18^gH-GFP/CLK09 double recipients immunized with either N332-GT2 or eOD-GT8 or both (C–E). (C and D) (C) GC cells as the percentage of total B cells and (D) CD45.2⁺ population as the percentage of total GC B cells at 10 dpi. (E) Distribution of the CD45.2+ GC B cell populations between BG18^gH-GFP (blue) and CLK09 (red) cells at 10 dpi. p values were calculated by unpaired Student’s t test (^∗∗p < 0.01; ^∗∗∗∗p < 0.0001). Figures represent data from one of at least two experiments with 2–5 mice per condition, with data presented as mean ± SD.

Figure 7

Excess antigen can overcome antibody-mediated restriction (A) Schematic of experiment to evaluate the antigen-mediated depletion of circulating antibodies. (B and C) (B) ELISA quantification of N332-GT2-specific human IgG and (C) N332-GT2-specifc murine IgG from mice receiving 10 μg BG18_iGL0 and subsequently immunized (blue) or left unimmunized (red) or mice that were immunized without receiving any mAb (black). p values were calculated by unpaired Student’s t tests comparing red versus blue (B) or blue versus black (C) groups at individual time points (^∗p < 0.05; ^∗∗p < 0.01; ^∗∗∗∗p < 0.0001). (D) Schematic of experiment to test the effects of multiple doses of antigen on naive CLK09 B cell responses in CLK09 recipients receiving a single dose or triple doses of eOD-GT8 immunogen after receiving either 10 μg CLK09 mAb (red) or no mAb (gray) (E and F). (E and F) (E) GC cells as the percentage of total B cells and (F) CD45.2⁺ CLK09 cells as the percentage of total GC B cells at 10 dpi of CLK09 recipients. Figures represent data from one of at least two experiments with 3–5 mice per condition, with data presented as mean ± SD. See also Figure S7.