B cell-intrinsic deficiency of the Wiskott-Aldrich syndrome protein (WASp) causes severe abnormalities of the peripheral B-cell compartment in mice

Abstract

Wiskott Aldrich syndrome (WAS) is caused by mutations in the WAS gene that encodes for a protein (WASp) involved in cytoskeleton organization in hematopoietic cells. Several distinctive abnormalities of T, B, and natural killer lymphocytes; dendritic cells; and phagocytes have been found in WASp-deficient patients and mice; however, the in vivo consequence of WASp deficiency within individual blood cell lineages has not been definitively evaluated. By conditional gene deletion we have generated mice with selective deficiency of WASp in the B-cell lineage (B/WcKO mice). We show that this is sufficient to cause a severe reduction of marginal zone B cells and inability to respond to type II T-independent Ags, thereby recapitulating phenotypic features of complete WASp deficiency. In addition, B/WcKO mice showed prominent signs of B-cell dysregulation, as indicated by an increase in serum IgM levels, expansion of germinal center B cells and plasma cells, and elevated autoantibody production. These findings are accompanied by hyperproliferation of WASp-deficient follicular and germinal center B cells in heterozygous B/WcKO mice in vivo and excessive differentiation of WASp-deficient B cells into class-switched plasmablasts in vitro, suggesting that WASp-dependent B cell-intrinsic mechanisms critically contribute to WAS-associated autoimmunity.

Figure 1

Generation of a conditional mouse model with selective lack of WASp in B lymphocytes (B/WcKO mice). (A) Schematic representation of the targeting strategy. A Was allele with exons 2-6 flanked by loxP sites and a neomycin-resistance (Neo) cassette flanked by frt sites was introduced in embryonic stem cells by homologous recombination (targeting construct). Correct targeting was verified by PCR using appropriate primers. Was locus-targeted mice were bred to transgenic mice expressing the FLP1-recombinase under the Rosa 26Sor promoter, leading to excision of the Neo cassette and allowing generation of Was^fl/fl mice. (B) Representative flow cytometric analysis of expression of CD19 and WASp in peripheral blood lymphocytes of mice with the indicated genotype.

Figure 2

Proportion of B-lineage cell types in the BM and spleen. (A) Different B-cell subsets in the BM of WT, B/WcKO, and WKO mice were analyzed by flow cytometry with the indicated cell surface markers. Results from individual mice are shown. Statistical significance was assessed with 1-way ANOVA and Bonferroni posttest analysis. **P < .01 and ***P < .001. (B) Different B-cell subsets in the spleen of WT, B/WcKO, and WKO mice were analyzed by flow cytometry with the indicated cell surface markers. Data from individual mice are shown. Statistical significance was assessed with 1-way ANOVA and Bonferroni posttest analysis. *P < .05, **P < .01, and ****P < .0001.

Figure 3

Reduction of MZ B cells and relative increase of GCs in the spleen of B/WcKO mice. (A) Splenic sections of WT, B/WcKO, and WKO mice were stained for B220 (blue), MOMA (green), and CD1d (red). MZ B cells (in purple) surround MOMA⁺ metallophilic macrophages in the spleen of WT mice but are severely depleted in B/WcKO and WKO mice. Scale bars, 150 μm. Eight mice per group were examined in 2 different experiments. Representative images are shown. (B) GC formation was assessed by staining for B220 (blue), PNA (red), and MOMA (green). Scale bars, 150 μm. Eight mice per group were examined in 2 different experiments. Representative images are shown. (C) The mean area occupied by follicles (FO; left panel) and by GCs (middle panel) is shown and was quantitated on immunohistologic slides as described in “Immunofluoresecence.” In addition, the GC/FO ratio of the respective areas is reported in the right panel. Values from individual spleen sections are shown. Significance was measured with 1-way ANOVA and Bonferroni posttest analysis. *P < .05.

Figure 4

Analysis of dysgammaglobulinemia in B/WcKO mice. Serum from 6- to 8-week-old (A) or 6-month-old (B) naive WT, B/WcKO, and WKO mice was tested for total serum IgM, IgG, IgA, and IgE concentrations by ELISA. Values of individual mice are shown. Bars indicate mean ± SEM. Significance was assessed with the Student t test. *P < .05, ***P < .001, and ****P < .0001.

Figure 5

Normal follicular, but impaired MZ B-cell responses in B/WcKO mice. (A) WT, B/WcKO, or WKO mice were immunized intraperitoneally with 100 μg of TNP-KLH and challenged with 25 μg of TNP-KLH 14 days later. Seven days after TNP-challenge, TNP-specific IgG responses (total TNP-specific IgG response, left panel; high-affinity TNP-specific IgG response, right panel) were measured by ELISA. Preimmune serum from WT mice was used as a control. Results are reported as mean ± SEM of 5 mice per group. The differences were not significant with 2-way ANOVA and Bonferroni posttest analysis, P > .05. (B) WT, B/WcKO, and WKO mice were immunized subcutaneously with noncapsulated PnWCV twice 2 weeks apart. PnWCV-specific IgG responses were measured by ELISA in sera collected 1 week after the second immunization. Preimmune serum from WT mice was used as a control. Mean ± SEM values of 5 mice per group are shown. The differences were not significant with 2-way ANOVA and Bonferroni posttest analysis. P > .05. (C) WT, WKO, and B/WcKO mice were immunized intravenously with 1 μg of Pneumovax23 vaccine. At the indicated times, serum was tested by ELISA for pneumococcal capsular polysaccharide-specific IgM (left) and IgG (right) Abs. Mean ± SEM values of 5 mice per group are shown. Significance was assessed by 2-way ANOVA and Bonferroni posttest analysis. ****P < .0001. (D) WT, WKO, and B/WcKO mice were infected with 2 × 10⁸ PFU UV-inactivated VSV. At the indicated times, serum was tested in a VSV neutralization assay for total neutralizing Ig. Mean ± SEM values of 5 mice per group are shown. Significances were assessed by 2-way ANOVA and Bonferroni posttest analysis. **P < .01.

Figure 6

Increased spontaneous production of anti-TNP and anti-phosphocholine IgM Abs in B/WcKO mice. (A-B) Serum from naive WT, B/WcKO, and WKO mice was tested for IgM binding to TNP (A) and phosphocholine (B). Mean ± SEM values of 5 mice per group are shown. Significance was assessed by 2-way ANOVA and Bonferroni posttest analysis. ***P < .001 and ****P < .0001. (C) The proportion of NP-specific CD19⁺ splenic B cells in WT, WKO, and B/WcKO mice was analyzed by FACS by staining for cells reactive to PE-labeled NP hapten. Differences were not significant (NS) with 1-way ANOVA and Bonferroni posttest analysis. (D) The relative proportion of peritoneal CD19⁺ CD5⁺ B1 cells among peritoneal CD19⁺ B cells was assessed by flow cytometry in WT, WKO, and B/WcKO mice. The mean ± SEM for 4 mice per group is shown. Differences were not significant (NS) with 1-way ANOVA and Bonferroni posttest analysis. (E) Het-B/WcKO female mice were tested for the proportion of WASp⁺ versus WASp⁻ cells within follicular B cell, MZ, GC, and PC B-cell compartments. The mean ± SEM values of 4-8 mice per group are shown. Statistical significance was assessed by 2-way ANOVA and Bonferroni posttest analysis. ****P < .0001. (F) Splenocytes of WT, B/WcKO, and WKO mice were stimulated in vitro with CpG (ODN 1826, 1.25μM final concentration). Five days later, generation of intracytoplasmic IgG⁺ (icIgG^hi) plasmablasts was measured by flow cytometry. Bars indicate the mean ± SEM of icIgG^hi plasmablasts generated in vitro from individual mice. Statistical significance was assessed by 1-way ANOVA and Bonferroni posttest analysis. *P < .05 and **P < .01. (G) HetB/WcKO mice were injected intraperitoneally with 2 mg of BrdU. Twelve hours later, BrdU incorporation in WASp⁺ versus WASp⁻ fractions of follicular B cells (left) and GC B cells (right) was measured by BrdU-specific Ab. Data of individual mice are shown. Statistical significance was assessed by Student t test. *P < .05 and **P < .01. (H) Apoptosis of GC B cells was assessed by annexin V labeling gating on CD19⁺ FAS⁺ GL7⁺ GC B cells from WT, B/WcKO, and WKO mice.

Figure 7

Increased autoantibody production in B/WcKO mice. (A-C) Sera of 6-month-old WT, B/WcKO, and WKO mice were tested by ELISA for IgG Abs specific for anti-ssDNA (A), anti-dsDNA (B), and chromatin (C). Data of individual mice are shown. Significance was calculated with the Mann-Whitney test. *P < .05 and **P < .01. (D) An autoantibody array chip was used to detect autoantibodies of IgM (left) or IgG (right) isotype against 73 different autoantigens (listed on the right). Sera from 6 WT mice (blue bar), 9 B/WcKO mice (green bar), and 7 WKO mice (gray bar) were tested. A serum from a lupus-prone MRL/NZM mouse (first row) served as a positive control. A red dot in the array indicates a 4-fold increase of autoantibody titer compared with average values in control sera.