NFATc1 affects mouse splenic B cell function by controlling the calcineurin--NFAT signaling network

Abstract

By studying mice in which the Nfatc1 gene was inactivated in bone marrow, spleen, or germinal center B cells, we show that NFATc1 supports the proliferation and suppresses the activation-induced cell death of splenic B cells upon B cell receptor (BCR) stimulation. BCR triggering leads to expression of NFATc1/αA, a short isoform of NFATc1, in splenic B cells. NFATc1 ablation impaired Ig class switch to IgG3 induced by T cell-independent type II antigens, as well as IgG3(+) plasmablast formation. Mice bearing NFATc1(-/-) B cells harbor twofold more interleukin 10-producing B cells. NFATc1(-/-) B cells suppress the synthesis of interferon-γ by T cells in vitro, and these mice exhibit a mild clinical course of experimental autoimmune encephalomyelitis. In large part, the defective functions of NFATc1(-/-) B cells are caused by decreased BCR-induced Ca(2+) flux and calcineurin (Cn) activation. By affecting CD22, Rcan1, CnA, and NFATc1/αA expression, NFATc1 controls the Ca(2+)-dependent Cn-NFAT signaling network and, thereby, the fate of splenic B cells upon BCR stimulation.

Figure 1.

NFATc1 induction in naive mouse splenic B cells. (A and B) Freshly prepared splenic B cells were stimulated with 10 µg/ml α-IgM or 10 µg/ml LPS for 0–48 h. Whole-cell protein was fractionated on 8% polyacrylamide gels and immunoblotted with either the mAb 7A6 recognizing all NFATc1 isoforms (A) or an antibody raised against the N-terminal α-peptide of NFATc1 (B). In lane 1 of panel B, protein from HEK 293 cells transfected with an NFATc1/αA expression plasmid was fractionated. NS, nonspecific band. (C) Splenic B cells from mice expressing an Nfatc1-Egfp reporter transgene were either left unstimulated or stimulated for 16 or 24 h. GFP fluorescence was determined by flow cytometry. (D and E) Splenic B cells from WT, Btk-deficient, or SLP-65-deficient mice were left unstimulated (Unst.) or were stimulated with LPS or α-IgM for 24 h, and whole protein extracts were immunoblotted using the NFATc1-specific mAb 7A6. ERK2 and β-actin, loading controls. In A-C, one typical experiment from more than three assays is shown; in D and E, one blot from two similar blots is presented.

Figure 2.

NFATc1 is required for optimal proliferation and survival of splenic B cells. (A) CFSE-labeled splenic B cells from a WT and mb1-cre x Nfatc1^flx/flx mouse were incubated for 72 h in the absence (Unstim.) or presence of 10 µg/ml LPS, 10 µg/ml α-IgM, or 2 µg/ml α-CD40. CFSE dilution was measured by flow cytometry. (B) Splenic B cells from WT, mb1-cre x Nfatc1^flx/flx, and Nfatc2^−/− mice were incubated for 72 h; cells were stained with PI and Annexin V and analyzed by flow cytometry. (C) WT and mb1-cre x Nfatc1^flx/flx (KO) splenic B cells were stimulated with α-IgM, as indicated. Gene expression was analyzed by DNA microarray. Key at the bottom indicates the fold changes in gene expression (log₂ values). (D) Induction of 10 NFATc1 target genes by α-IgM stimulation in WT and NFATc1^−/− splenic B cells. Error bars show ± the SEM. (E) ChIP assay showing the in vivo binding of NFATc1 in splenocytes to the promoters of Il2, Rcan1, Spp1, Fasl, Pdcd1, and Tnfsf14 genes upon stimulation for 16 h with α-IgM or T+I. Lanes 1–3, ChIP assays using splenocytes that were left unstimulated (-) or stimulated by T+I or α-IgM for 16 h. ChIP assays were also performed using an antibody directed against GST protein. As input control, 1% of chromatin of all assays was used. In A and B, one typical experiment is shown from more than three assays; in C and D, mean values from three individual DNA microarray analyses are presented, and in E one typical ChIP assay from three assays is shown.

Figure 3.

NFATc1 controls Ca²⁺/Cn signaling in splenic B cells. (A) DNA microarray assay showing 10 genes whose expression was enhanced in B cells from mb1-cre x Nfatc1^flx/flx mice compared with WT B cells. Mean values of 3 individual assays are shown. (B) Manipulation of Ca²⁺/Cn signaling by adding 10 mM CaCl₂ or 100 ng/ml CsA to proliferation assays. Splenic B cells from WT or mb1-cre x Nfatc1^flx/flx mice were stained with CFSE and treated with 10 µg/ml α-IgM in the absence or presence of Ca²⁺ or CsA in X-vivo medium or 1:1 X-vivo+RPMI medium that contains less Ca²⁺. Numbers indicate the percentage of cycling cells. One typical set of more than three experiments is shown. (C) Ca²⁺ release and influx in B cells from WT and mb1-cre x Nfatc1^flx/flx (KO) mice. One typical experiment from more than three assays is shown. (D) Effect of NFATc1 ablation on CD22 expression. (left) Semiquantitative RT-PCR assay showing an increase in CD22 RNA in NFATc1^−/− splenic B cells. One typical assay from three individual assays is shown. (right) Flow cytometric detection of CD22 expression on WT and Cd23-cre x Nfatc1^flx/flx splenic B cells left unstimulated or stimulated with LPS or α-IgM for 2 h. Two more assays with B cells from two mice each were performed and gave similar results. (E) CnA assays. (top) ChIP assay for the binding of NFATc1 to the Ppp3cb and Il2 promoter in vivo. Splenocytes were left unstimulated (−) or stimulated with α-IgM for 16 h. As indicated, either an NFATc1-specific or an unrelated antibody (α-GST) was used. As input control, 1% of chromatin of all assays was used. (middle) Immunoblot using whole cellular protein from B cells and a CnA α/β-specific antibody. Below the blot, Ponceau red staining is shown as loading control. (bottom) Mean values of Cn activity from three assays are shown (P ≥ 0.05 between WT and mb1-cre x Nfatc1^flx/flx cells). (F) Effect of NFATc1 ablation on NFATc2 expression. (top) Decrease in nuclear NFATc2 in NFATc1^−/− B cells. Splenic B cells were treated with α-IgM and α-CD40 for 24 h, and for the last 30 min, the cells were also treated with T+I. Splenic B cells stained with an NFATc2-specific antibody. (right) Nuclear staining of 100 cells measured by confocal microscopy is compiled. (bottom) Semiquantitative RT-PCR assay. Error bars show ± the SEM.

Figure 4.

NFATc1 deficiency results in a defect in IgG3 class switch by the TI-type II antigen NP-Ficoll. (A) IgM and IgG3 levels in sera of WT, Cd23-cre, and Aicda-cre x Nfatc1^flx/flx mice after immunization with the TI-type II antigen NP-Ficoll. Each symbol represents one mouse. (B) Appearance of circular IgG3 and IgG1 transcripts in WT and NFATc1^−/− splenic B cells after in vitro culture for 2 d in the presence of LPS. (C) ELISA showing a decrease in IgG3 production of B cells from Cd23-cre x Nfatc1^flx/flx mice upon culture for 6 d. (D) Decreased IgG⁺ plasmablast formation of B cells from Cd23-cre x Nfatc1^flx/flx mice upon 3 d in culture. Flow cytometry of CD19⁺-gated B cells. In the columns data from three assays are compiled. (E and F) Diminished proliferation (E) and increased AICD (F) of CD138⁺ plasmablasts generated from B cells of Cd23-cre x Nfatc1^flx/flx mice upon 3 d culture. (G) Strong Nfatc1-Egfp reporter gene expression in plasmablast-like B cells upon immunization of mice with NP-Ficoll. Flow cytometry of splenic B cells showing GFP and IgM+IgG3, PNA⁺FAS, CD138, or CD69 expression, respectively. In E–G, one typical experiment from three assays is shown. Error bars show ± the SEM.

Figure 5.

TNP-Ficoll antigen does not bind to NFATc1^−/− MZB cells, but forms large aggregates in spleen. 150 µg TNP-Ficoll was injected i.v., and 1 h later splenic B cells were prepared for flow cytometry (A) or spleens were taken for histochemical staining (B–D). In A, binding of antigen to MZB cells was determined by staining with APC-labeled α-TNP-Ficoll. In B–D, histochemical stainings were performed using α-TNP-Ficoll, α-IgD, α-CD169/MOMA-1, and α-IgM. Typical results from three injection experiments are shown.

Figure 6.

Ablation of NFATc1 in B cells impairs their T cell help. (A and B) MLC. In A, 5 × 10⁴ B cells from C57/BL6 WT or Cd23-cre x Nfatc1^flx/flx mice were incubated with 5 × 10⁴ CFSE-labeled CD4⁺ T cells from a BALB/c WT mouse. After 3 d, proliferation of T cells was measured. In B, B and T cells were cultured as in A and IL-2 secretion was measured by ELISA. In A, one typical assay out of three is shown; in B mean values of two assays are compiled. (C and D) Splenic B cells from WT and Cd23-cre x Nfatc1^flx/flx mice were treated overnight with 10 µg/ml α-IgM and 1 µM (C) or 10 µM OVA_323-339 peptide (D) followed by incubation with CFSE-stained 5 × 10⁴ T cells from an OTII tg mouse (Barnden et al., 1998) for 3 d. In C, one typical assay out of three is shown. In D, mean values of three experiments are presented.

Figure 7.

NFATc1 ablation leads to an increase in IL-10 production and amelioration of the clinical course of EAE. (A) Increased CD19⁺CD5⁺CD1d^high B cell number in spleen freshly prepared from Cd23-cre X Nfatc^flx/flx mice, compared with WT mice. Six individual mice were investigated, and mean values are shown. (B) IL-10 production of unstimulated CD5⁺CD1d^high NFATc1^−/− B cells or cells treated with α-IgM or LPS (10 µg/ml each) for 24 h and with T+I for the last 8 h. Cells were stained with antibodies directed against CD1d and CD5, and against IL-10 for intracellular flow cytometry. Shown are the results of one (out of five) typical experiment of IL-10 stains of CD5⁺CD1d^high B cells. (C) Increase in IL-10–producing CD138⁺CD19⁺ plasmablasts after 3 d in culture with LPS (and T+I for the last 6 h) in preparations of splenocytes from mice bearing Cd23-cre x Nfatc1^flx/flx B cells. One typical assay out of three is shown. (D) IL-10 production by unstimulated WT and NFATc1^−/− splenic B cells and cells stimulated by α-IgM or LPS for 72 h in culture. IL-10 was measured in ELISA, and mean values of three assays are shown. (E) MLC. Decreased IFN-γ production by CD4⁺ T cells incubated for 48 h in the presence of 0.5 µg/ml α-CD3 and, during the last 6 h, with T+I, without B cells or with unstimulated B cells or B cells prestimulated by α-IgM or LPS for 24 h from WT or mb1-cre x Nfatc1^flx/flx mice. As indicated, 5 ng/ml recombinant IL-10 and/or 1 µg/ml of IL-10–specific antibody JES5-2A5 (Lampropoulou et al., 2008) were added to the cultures. One typical set out of three assays is shown. (F) EAE. By injection of MOG_35-55 peptide in complete Freund’s adjuvant, and pertussis toxin at day 2 (Pierau et al., 2009), EAE was induced in mb1-cre x Nfatc1^flx/flx and WT mice as indicated. The data show the mean clinical score + SEM from 3 independent experiments, with n = 9 (WT) and n = 14 (mb1-cre x Nfatc1^flx/flx) mice.