B cell peripheral tolerance is promoted by cathepsin B protease

Abstract

B cells that bind soluble autoantigens receive chronic signaling via the B cell receptor (signal-1) in the absence of strong costimulatory signals (signal-2), and this leads to their elimination in peripheral tissues. The factors determining the extent of soluble autoantigen-binding B cell elimination are not fully understood. Here we demonstrate that the elimination of B cells chronically exposed to signal-1 is promoted by cathepsin B (Ctsb). Using hen egg lysozyme-specific (HEL-specific) immunoglobulin transgenic (MD4) B cells and mice harboring circulating HEL, we found improved survival and increased proliferation of HEL-binding B cells in Ctsb-deficient mice. Bone marrow chimera experiments established that both hematopoietic and nonhematopoietic sources of Ctsb were sufficient to promote peripheral B cell deletion. The depletion of CD4⁺ T cells overcame the survival and growth advantage provided by Ctsb deficiency, as did blocking CD40L or removing CD40 from the chronically antigen-engaged B cells. Thus, we suggest that Ctsb acts extracellularly to reduce soluble autoantigen-binding B cell survival and that its actions restrain CD40L-dependent pro-survival effects. These findings identify a role for cell-extrinsic protease activity in establishing a peripheral self-tolerance checkpoint.

Keywords: B lymphocyte; CD40; cathepsin; self-tolerance; signal-1.

Fig. 1.

Ctsb Promotes Elimination of HEL-Binding B Cells. (A) Ctsb activity in assay buffer or spleen interstitial fluid of control (Ctsb^+/+ or Ctsb^+/−) or Ctsb^−/− (CtsB KO) mice. Data show results from two separate experiments. (B) Schematic of CTV-labeled MD4 B cell adoptive transfer into control or Ctsb-deficient mice, followed by HEL treatment. (C) Frequencies of transferred MD4 B cells in spleens of control or Ctsb-deficient recipients 3 d after saline (n = 16 control, n = 9 KO mice) or HEL treatment (n = 26 control, n = 30 KO mice). (D–F) MFI (Top) and representative histogram plot (Bottom) of CD23 (D), ICAM1 (E), and IgMa (F) on transferred MD4 B cells 3 d after saline (n = 11 control, n = 5 KO) or HEL treatment (n = 22 control, n = 24 KO). (G) MFI of HyHEL9 on MD4 B cells incubated with 1:10 or 1:50 dilutions of sera from control or Ctsb-deficient mice 3 d after saline (n = 7 control, n = 5 KO) or HEL treatment (n = 16 control, n = 17 KO). (H) Percentage of divided transferred MD4 B cells (Top) or representative histogram plot of CTV (Bottom) 3 d after saline (n = 8 control, n = 4 KO) or HEL treatment (n = 16 control, n = 17 KO). (I) Frequencies of transferred MD4 B cells in spleens of control ML5⁺ (n = 8) or Ctsb-deficient ML5⁺ (n = 7) recipients 3 d after MD4 B cell adoptive transfer. Control ML5⁻ (n = 3) mice used as deletion control. (J) Frequencies of transferred MD4 B cells in spleens of bone marrow (BM) chimeras 3 days after saline or HEL treatment. The Ctsb genotype of donor BM and recipient mice used to generate the BM chimeras is indicated. In graphs, each data point indicates an individual mouse and lines indicate means. Error bars represent SDs. I is representative of three experiments. Statistical significance for A–I was determined by unpaired t test. NS, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Statistical significance for J was determined by ordinary one-way ANOVA, ***, P = 0.0007.

Fig. 2.

Follicular Exclusion of HEL-Engaged B Cells Is Unaffected by Ctsb Deficiency. (A and B) Immunofluorescence for MD4 GFP B cells (GFP, green) transferred into HEL-treated control (A) or Ctsb-deficient (B) mice stained to detect endogenous B cell follicles (IgD, blue; CD21/35, white) and the T cell zone (CD4, red). Sections were prepared one day after HEL treatment. Two example images are shown and are representative of multiple cross-sections from at least three mice of each type. (Scale bar, 200 µm.) (C) Quantification of proportion of MD4 GFP B cells at the T zone–follicle (T-B) interface one day after HEL treatment. Each data point represents an individual follicle (n = 24 control, n = 25 KO) from sections prepared from at least three mice of each genotype. Lines indicate means, and error bars represent SDs. Statistical significance was determined by unpaired t test. NS, not significant.

Fig. 3.

Depletion of CD4⁺ T Cells Overcomes the Effect of Ctsb Deficiency on HEL-Binding B Cells. (A) Schematic of CTV-labeled MD4 B cell adoptive transfer into control or Ctsb-deficient mice, followed by GK1.5 CD4⁺ T cell depletion and HEL treatment. (B) Frequencies of transferred MD4 B cells in spleens of control or Ctsb-deficient recipients 3 d after saline (n = 6 control), HEL (n = 7 control, n = 9 KO), or HEL with GK1.5 treatment (n = 6 control, n = 9 KO). (C–E) MFI (Top) and representative histogram plot (Bottom) of CD23 (C), ICAM1 (D), and IgMa (E) on transferred MD4 B cells 3 d after saline (n = 5 control), HEL (n = 6 control, n = 5 KO), or HEL with GK1.5 treatment (n = 6 control, n = 5 KO). (F) Percentage of divided transferred MD4 B cells (Top) or representative histogram plot of CTV (bottom) 3 d after saline (n = 3 control, n = 2 KO), HEL (n = 6 control, n = 6 KO), or HEL with GK1.5 treatment (n = 6 control, n = 5 KO). Each data point indicates an individual mouse and lines indicate means. Error bars represent SDs. C–F are representative of three experiments. Statistical significance for B–F was determined by unpaired t test. NS, not significant; *P < 0.05; **P < 0.01, ***P < 0.001.

Fig. 4.

Enhanced HEL-Binding B Cell Persistence in Ctsb-deficient Hosts Depends on CD40L and CD40. (A) Frequencies of transferred MD4 B cells in spleens of control or Ctsb-deficient recipients 3 d after saline, HEL, or HEL with MR1 CD40L-blocking treatment. (B and C) MFI of CD23 (B) and ICAM1 (C) on transferred MD4 B cells 1.5 d after saline (n = 4 control, n = 4 KO), HEL (n = 8 control, n = 8 KO), or HEL with MR1 treatment (n = 9 control, n = 9 KO). (D) Percentage of divided transferred MD4 B cells 3 d after saline (n = 3 control, n = 4 KO), HEL (n = 6 control, n = 5 KO), or HEL with MR1 treatment (n = 6 control, n = 6 KO). (E) CD40L transcript abundance in naïve CD4⁺ T cells isolated from spleen and lymph node tissues harvested from control (n = 6) or Ctsb^−/− (n = 3) mice. CD40L and hypoxanthine phosphoribosyltransferase (HPRT) transcripts were quantitated by real-time PCR. Data show results from two separate experiments. (F) MFI of CD40L on purified CD4⁺ T cells from control (n = 7) or Ctsb-deficient (n = 8) mice incubated without (no stim) or with (PMA/iono) PMA and ionomycin for 2 h at 37 °C. (G) MFI of CD40L on purified CD4⁺ T cells from control (n = 27) or Ctsb-deficient (n = 26) mice kept on ice or incubated in a dilute culture for 2 h at 37 °C. (H) Frequencies of WT or CD40-deficient MD4 B cells in transferred splenocytes in spleens of control or Ctsb-deficient recipients 3 d after saline (n = 2 control, n = 2 KO) or HEL treatment (n = 6 control, n = 6 KO). (I and J) Frequencies of transferred MD4 B cells amongst total cells (I) and of MD4 B cells having a germinal center (GC) phenotype (J) in spleens of unimmunized mice (n = 3) or mice immunized with SRBC-HEL^2x (n = 4 control, n = 4 KO). (K) Frequencies of undivided MD4 B cells or MD4 B cells having undergone five or more divisions in spleens of control (n = 4) or Ctsb-deficient mice (n = 4) 5 d after SRBC-HEL^2x immunization. (L) Representative histogram plot of CTV labeling of transferred MD4 B cells in indicated recipients 5 d after SRBC-HEL^2x immunization. Each data point indicates an individual mouse and lines indicate means. Error bars represent SDs. A–D and H are representative of three experiments. I–L are representative of two experiments. Statistical significance for A–K was determined by unpaired t test. NS, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.