B cells from knock-in mice expressing broadly neutralizing HIV antibody b12 carry an innocuous B cell receptor responsive to HIV vaccine candidates

Abstract

Broadly neutralizing Abs against HIV protect from infection, but their routine elicitation by vaccination has not been achieved. To generate small animal models to test vaccine candidates, we have generated targeted transgenic ("knock-in") mice expressing, in the physiological Ig H and L chain loci, two well-studied broadly neutralizing Abs: 4E10, which interacts with the membrane proximal external region of gp41, and b12, which binds to the CD4 binding site on gp120. 4E10HL mice are described in the companion article (Doyle-Cooper et al., J. Immunol. 191: 3186-3191). In this article, we describe b12 mice. B cells in b12HL mice, in contrast to the case in 4E10 mice, were abundant and essentially monoclonal, retaining the b12 specificity. In cell culture, b12HL B cells responded avidly to HIV envelope gp140 trimers and to BCR ligands. Upon transfer to wild-type recipients, b12HL B cells responded robustly to vaccination with gp140 trimers. Vaccinated b12H mice, although generating abundant precursors and Abs with affinity for Env, were unable to rapidly generate neutralizing Abs, highlighting the importance of developing Ag forms that better focus responses to neutralizing epitopes. The b12HL and b12H mice should be useful in optimizing HIV vaccine candidates to elicit a neutralizing response while avoiding nonprotective specificities.

Figure 1. H-chain allelic exclusion in b12H knock-in B cells

A,B, Flow cytometry analysis of H-chain allelic exclusion in b12H B cells. b12H mice (IgH^b) were crossed to IgH^a/a mice (“b12H/a”) to allow identification of cells carrying H-chains encoded by the endogenous allele and compared to B6 (b/b) or (B6 x IgH^a)F1 (a/b) cells. A, Plot shows expression of IgM^a and IgM^b on B220⁺-gated splenocytes from the indicated strains. B, Statistical analysis of IgM^b frequency among eight b12H/a and five a/b mice. ****, p < 0.0001.

Figure 2. B cell generation and Env binding by b12 knock-in B cells

A–C, Comparison of B cell generation in bone marrow (BM), spleen (SP) and lymph nodes (LN) of b12HL, b12H, and b12L mice. A, Flow cytometry analysis of IgM and B220 staining in BM, SP, and LN of the indicated strains. Plots shown were gated on TCRβ negative lymphocytes. Top panels indicate with boxes B220^intIgM⁻ immature B cells, B220^intIgM⁺ immature B cells, and B220^hi recirculating B cells, respectively. Middle and bottom panels box B220⁺ B cells. B, Statistical analysis of BM immature vs recirculating B cell numbers from wild type (WT), b12HL and b12H mice. *, p< 0.05; **, p< 0.005; ***, p< 0.0005. C, Analysis of B220+ cell numbers in SP and LN of 9 WT, 6 b12HL and 8 b12H mice. D,E, Binding of soluble Env (JRFL gp140 trimer) by B cells from the indicated strains and tissues. Each data point in E gives the value obtained from an individual mouse.

Figure 3. Analysis of normal serum Ig levels and HIV reactivity in b12H and HL mice

A,B, Statistical analysis of serum IgM and IgG levels in unmanipulated mice. Data from 10 WT, 6 b12HL, 10 b12H and 4 b12L mice. C, Env (JRFL)-specific IgM and IgG concentrations from 6 b12HL and 6 b12H mice. b12 IgM or IgG1 was used to standardize the assay. D, Analysis of neutralization activity against JRFL pseudovirus of normal sera from 4 b12HL and 4 b12H mice. Shown are mean values and SEM. Results were derived from at least three independent experiments except for the serum analyses which were obtained in one experiment.

Figure 4. Maturation status of Env-binding B cells in b12HL and b12H mice

A,B, Flow cytometry analysis of splenic B cell maturation in b12HL and b12H mice. A, Analysis of CD23 and CD21 expression. Plots shown were gated on B220⁺ (WT) or JRFL⁺B220⁺ (b12HL and b12H). Boxes define CD21⁺CD23⁺ follicular B cells and CD21⁺⁺ marginal zone B cells. B, CD93 expression levels on b12HL and b12H CD23⁺CD21⁺ B cells (Upper box from A). C, Calcium mobilization analysis of gated splenic B cells of the indicated genotypes upon stimulation with anti-Igκ or soluble Env trimers. Data shown in A and B are representative of at least 6 mice analyzed in three experiments. Data in C are representative of three experiments.

Figure 5. Analysis of b12HL B cell function in mixed chimeras with wild type cells

Irradiated Rag1^−/− mice were reconstituted with a mixture of b12HL and WT BM at a 1:1 ratio. At six weeks post transfer, reconstitution was assessed and immunization initiated. A, Analysis of peripheral blood taken at 6 weeks for the presence of Env-binding (b12HL) and non-binding B cells by gating on TCRβ⁻ lymphocytes and staining for B220 and YU2 gp120 binding. B, CD93 density on B220⁺YU2⁻ and B220⁺YU2+ cells. C,D, Serum IgM and IgG responses to Env immunization at various weeks post priming. Chimeric mice were immunized with 40 μg JRFL gp140 in Sigma Adjuvant and boosted three weeks later with antigen in saline. Data are from a total of eight mice as indicated carried out in one experiment.

Figure 6. Antibody response of b12H mice to soluble Env trimers

b12H and WT mice were challenged with soluble Env JRFL gp140 trimers and antibody responses measured using the JRFL gp140, YU2 gp120 or 2bodx_43 as the capture reagent in ELISA. The immunization protocol was as in Figure 3, with 4 mice/group. A, Ability of naive b12H B cells to bind to biotinylated YU2 gp120 and 2bodx_43. B,C, JRFL gp140-specific IgM and IgG response. D,E, IgG responses cross-reactive to YU2 gp140 or 2bodx_43. F, Neutralization of JRFL pseudotyped virus by immune sera from four indicated mice. Non-immunized b12HL sera were used as positive control. G,H, Analysis of IgG1 hybridomas from b12H mice. G, ELISA analysis of Env binding by selected hybridoma proteins to plate-bound JRFL gp140. H, Flow cytometry analysis of the binding of the indicated IgG1 hybridoma proteins to 293 cells transfected with full-length JRFL Env (cleaved, lower panels) or JRFL carrying a mutated furin cleavage site (uncleaved, upper panels).