Loss of the BH3-only protein Bmf impairs B cell homeostasis and accelerates gamma irradiation-induced thymic lymphoma development

Abstract

Members of the Bcl-2 protein family play crucial roles in the maintenance of tissue homeostasis by regulating apoptosis in response to developmental cues or exogenous stress. Proapoptotic BH3-only members of the Bcl-2 family are essential for initiation of cell death, and they function by activating the proapoptotic Bcl-2 family members Bax and/or Bak, either directly or indirectly through binding to prosurvival Bcl-2 family members. Bax and Bak then elicit the downstream events in apoptosis signaling. Mammals have at least eight BH3-only proteins and they are activated in a stimulus-specific, as well as a cell type-specific, manner. We have generated mice lacking the BH3-only protein Bcl-2-modifying factor (Bmf) to investigate its role in cell death signaling. Our studies reveal that Bmf is dispensable for embryonic development and certain forms of stress-induced apoptosis, including loss of cell attachment (anoikis) or UV irradiation. Remarkably, loss of Bmf protected lymphocytes against apoptosis induced by glucocorticoids or histone deacetylase inhibition. Moreover, bmf(-/-) mice develop a B cell-restricted lymphadenopathy caused by the abnormal resistance of these cells to a range of apoptotic stimuli. Finally, Bmf-deficiency accelerated the development of gamma irradiation-induced thymic lymphomas. Our results demonstrate that Bmf plays a critical role in apoptosis signaling and can function as a tumor suppressor.

Figure 1.

Generation of bmf^−/− mice by homologous recombination. (A) Schematic representation of the bmf gene locus on mouse chromosome 2. Exons 3 and 4, as well as the neomycin selection marker, were flanked by loxP elements in the targeting construct. Exons 3 and 4, as well as the neomycin selection marker cassette, were removed in vivo by cre-mediated deletion using Ubi-cre deleter mice (dashed lines). The location of the 5′ and 3′ probes used for Southern blot analysis are indicated as black bars. (B) Southern blot analysis of liver-derived DNA from WT, bmf^+/−, and bmf^−/− mice using 5′ and 3′ external probes. (C) Analysis of bmf mRNA expression in MEFs by Northern blot analysis. (D) Bmf protein expression analysis by Western blotting in extracts from thymus, spleen, or lymph nodes derived from WT and both strains of Bmf-deficient mice, which had been generated from independent ES cell clones.

Figure 2.

Expression analysis of Bmf and Bim in different leukocyte subsets. (A) Cells of the indicated differentiation stages were isolated from WT mice by cell surface marker staining and FACS sorting. Cells were lysed, and proteins were size fractioned by SDS-PAGE and transferred onto nitrocellulose membranes. Protein expression was evaluated using monoclonal antibodies recognizing mouse Bmf, Bim, or ERK (loading control). Membranes were stripped before subsequent reprobing with anti-Bim– and -ERK–specific antibodies, respectively. (B) Total thymocytes and CD19⁺ B cells from WT and bmf^−/− mice were analyzed for expression of Bim, Bad, and Puma. Reprobing with an anti-ERK–specific antibody served as loading control. (C) B cell subsets of the indicated differentiation stages were isolated from WT or Bmf-deficient mice by cell surface marker staining and FACS sorting and processed as in A.

Figure 3.

Bmf-deficient thymocytes are abnormally resistant to certain apoptotic stimuli. (A) Total thymocytes from mice of the indicated genotypes were put in culture without further treatment (spontaneous death; A) or exposed to the DNA-damaging drug VP16 (etoposide; B), dexamethasone (C), or the HDAC inhibitor SAHA (D). Cell death was monitored by Annexin V-FITC/PI staining and flow cytometric analysis at the indicated time points. The extent of apoptosis induced specifically by different stimuli was calculated by the following equation: (induced apoptosis − spontaneous cell death)/(100 − spontaneous cell death). Means ± the SEM from four independent experiments and n = 5 animals per genotype are shown. Significant differences in cell death induction were observed between WT and Bmf-deficient cells treated with VP16 (P = 0.0096), dexamethasone (P = 0.0048), or SAHA (P = 0.0089) using ANOVA analysis. Thymocytes from WT and Bmf-deficient mice were evaluated for caspase-3 activation in response to dexamethasone (E) or HDAC inhibition by SAHA (F) using an antibody specifically recognizing the proteolytically generated active p17 fragment of caspase-3. Filters were reprobed using antibodies specific for Bmf or β-actin (loading control). Reactivity of SAHA was confirmed using an antibody recognizing acetylated histone H4 (Lys12) that was showing highest reactivity after 6 h. 8–10-wk-old female mice of the indicated genotypes were injected with graded doses of dexamethasone, and the percentages (G) and number (H) of surviving CD4⁺8⁺ thymocytes were assessed 20 h later by cell counting of single-cell suspensions and flow cytometric analysis of CD4 and CD8 cell surface marker expression. Bars represent the mean ± the SEM of 8 WT, 8 bmf^−/−, and ≥3 bim^−/− mice per treatment and ≥3 independent experiments. Using analysis of variance, significant differences (*) in the numbers of surviving thymocytes were observed between WT, bmf^−/−, and bim^−/− mice using 125 μg dexamethasone/mouse (P < 0.04) and between WT and bim^−/− mice using 500 μg dexamethasone (P = 0.007).

Figure 4.

Loss of Bmf causes B cell hyperplasia. (A) Flow cytometric analysis of spleen-derived single-cell suspensions from WT, bmf^−/−, and bim^−/− mice using antibodies against the T cell marker Thy1 and the B cell marker B220. Representative dot blots of spleen cells stained with different B cell maturation markers to identify different stages of B cell development: T1 transitional B cells (sIgM^highCD21⁺); T2 transitional B cells (CD21⁺CD23⁺sIgM^high); mature follicular (FO) B cells (CD21⁺CD23⁺); marginal zone (MZ) B cells(CD21^highCD21⁻); and plasma cells (PC; B220⁺CD138⁺). (bottom) Quantification of B cell subsets in the spleens derived from T, bmf^−/−, and bim^−/− mice (n = 4) gathered in two independent experiments. Error bars represent the means ± the SEM. Using the unpaired Student's t test, significant differences between WT and bmf^−/− or bim^−/− mice were observed in T1 cells (P < 0.039), T2 cells (P < 0.0019), and FO cells (P < 0.01). The number of plasma cells was only significantly elevated in Bim-deficient mice (WT vs. bim^−/−; P < 0.03).

Figure 5.

Mild hypergammaglobulinemia in untreated Bmf-deficient mice. Ig-titers in the sera from 8–12 wk or 12-mo-old WT and bmf^−/− mice were quantified by ELISA. Dilutions were predetermined to produce absorbance readings in the linear range. Total Ig 1:80.000 (A), IgM 1:40.000 (B), IgG1 1:40.000 (C), IgG2a 1:4000 (D), IgG2b 1:80.000 (E), IgG3 1:40.000 (F), and IgA 1:40.000 (G). Box plots represent values from n = 4 WT and 8 bmf^−/− mice (8–10 wk), and 8 WT and 8 bmf^−/− aged mice. Significant differences were only observed in the total Ig levels between young WT and Bmf-deficient mice using Student's t test (P < 0.0123). (H) Autoantibodies to dsDNA were quantified in the sera from 8–12-wk- or 12-mo-old mice by ELISA using ds calf thymus DNA for coating (serum dilution 1:100). Box plots represent values from n = 6 WT and 6 bmf^−/− mice (8–10 wk) and 5 WT and 8 bmf^−/− aged mice.

Figure 6.

Impaired cell death responses in bmf^−/− pre–B cells. FACS-sorted pre–B cells (B220⁺sIgM⁻CD43⁻) from mice of the indicated genotypes were put in culture without further treatment (A; spontaneous death) or exposed to dexamethasone (B), the DNA-damaging drug VP16 (C), or the HDAC-inhibitor SAHA (D). Cell death was monitored, and the extent of apoptosis induced specifically by different stimuli was calculated as indicated in Fig. 3. Means ± the SEM from four independent experiments and n = 5 animals per genotype are shown. Significant differences in cell death induction were observed between WT and Bmf-deficient cells treated with dexamethasone (P = 0.0001) or SAHA (P = 0.004) using ANOVA analysis. Mice of the indicated genotypes were injected with graded doses of dexamethasone and the percentage (E) and total number (F) of surviving pre–B cells was assessed 20 h later by cell counting of bone marrow single-cell suspensions and flow cytometric analysis of pre–B cell surface marker expression (B220⁺sIgM⁻CD43⁻). Bars represent the means ± the SEM of ≥3 animals per genotype/dose (three independent experiments). Significant differences in vivo were observed between the number of surviving WT and Bmf-deficient pre B-cells in mice treated with 125 μg dexamethasone (P = 0.01).

Figure 7.

Bmf-deficiency promotes abnormally enhanced lymphocyte accumulation through a cell autonomous mechanism. Ly5.1 congenic mice were reconstituted 6 h after lethal irradiation (10 Gy) using a 50:50 mixture of WT Ly5.1 + WT Ly5.2 or WT Ly5.1 + bmf^−/−Ly5.2 bone marrow cells (total 4 × 10⁶ cells/mouse). Reconstitution was monitored over time by evaluating the percentage of donor cells in peripheral blood by FACS analysis using a Ly5.2-specific antibody, either alone (A) or in combination with the T cell markers CD4 (B), CD8 (C), or the B cell marker specific antibody CD19 (D). 21 wk after reconstitution, animals were killed and the percentage of donor-derived lymphocytes was assessed in thymus (E) and spleen (F) using the cell surface marker–specific antibodies indicated.

Figure 8.

Loss of Bmf enhances γ irradiation induced thymic lymphoma development. Mice of the indicated genotypes (7–8 wk of age) were exposed to whole body irradiation using 2.5 Gy. Animals were killed 20 h later, and thymocyte numbers were evaluated by FACS analysis of cell surface marker expression (A) and cell counting to calculate the total number of thymocytes (B) and CD4⁺8⁺ immature thymocytes (C). Bars represent means ± the SEM of 3 WT and 5 Bmf-deficient animals and two independent experiments. Significant differences in depletion of total thymocytes (P = 0.0044) and CD4⁺8⁺ thymocytes (P = 0.003) were confirmed using ANOVA analysis. Cohorts of WT (n = 16), bmf^−/− (n = 8), and p53^+/− (n = 9) mice were exposed to a fractionated γ irradiation protocol (4 × 1.75 Gy in weekly intervals, starting at 4 wk of age) and monitored for the development of thymic lymphomas over time. (B) Kaplan-Meier analysis of tumor-free survival of mice of the indicated genotypes. Log rank (Mantle-Cox) analysis was used to calculate differences between WT and p53^+/− (P = 0.027) and WT and bmf^−/− mice (P = 0.009).