Bcl11a is essential for lymphoid development and negatively regulates p53

Abstract

Transcription factors play important roles in lymphopoiesis. We have previously demonstrated that Bcl11a is essential for normal lymphocyte development in the mouse embryo. We report here that, in the adult mouse, Bcl11a is expressed in most hematopoietic cells and is highly enriched in B cells, early T cell progenitors, common lymphoid progenitors (CLPs), and hematopoietic stem cells (HSCs). In the adult mouse, Bcl11a deletion causes apoptosis in early B cells and CLPs and completely abolishes the lymphoid development potential of HSCs to B, T, and NK cells. Myeloid development, in contrast, is not obviously affected by the loss of Bcl11a. Bcl11a regulates expression of Bcl2, Bcl2-xL, and Mdm2, which inhibits p53 activities. Overexpression of Bcl2 and Mdm2, or p53 deficiency, rescues both lethality and proliferative defects in Bcl11a-deficient early B cells and enables the mutant CLPs to differentiate to lymphocytes. Bcl11a is therefore essential for lymphopoiesis and negatively regulates p53 activities. Deletion of Bcl11a may represent a new approach for generating a mouse model that completely lacks an adaptive immune system.

Figure 1.

Dynamic expression patterns of Bcl11a in hematopoiesis. (A) Schematic diagram of the Bcl11a-eGFP reporter allele. The eGFP reporter cassette flanked by two F3 sites is introduced to the 3′UTR region of Bcl11a, 8 bp after the stop codon TAG. (B) Flow cytometry tracking Bcl11a expression using the Bcl11a^eGFP/eGFP reporter mice. Cell surface markers for defining these cells are described in Table S1. (C) Expression of Bcl11a (eGFP⁺) and Bcl11b (Tdtomato⁺) in double negative (DN) thymocytes was measured by flow cytometry. DN thymocytes are identified as described in Fig. S1 D. (D) qRT-PCR analysis of Bcl11a expression in sorted hematopoietic populations. Data represent mean values of three independent biological replicates and all values are normalized to the expression of the Gapdh gene. Error bars indicate the SD. Thy, thymus; PB, peripheral blood; M, macrophages; G, granulocytes; Mk, megakaryocytes; B, BM CD19⁺ B cells; T, spleen CD3⁺T cells. In all flow cytometry assays, at least four mice were analyzed for each cell type in independent experiments.

Figure 2.

Profound defects of lymphoid development in Bcl11a-deficient adult mouse. B and T cells at different developmental stages were analyzed in the BM or thymus 1 wk after the mice were injected with Tam (n = 4/genotype, and the experiment was repeated 4 times). (A) BM B cells. Numbers refer to percentages of B cells in total BM-nucleated cells. (B) Cellularity of BM B cells in mice of the indicated genotypes. Cells were collected from two femurs of each mouse. (C) Thymocytes of various development stages. Lin⁻ thymocytes were further analyzed by expression of c-Kit and CD25 for DN thymocytes. (D) Cellularity of DN thymocytes in mice of indicated genotypes. (E) Analysis of LSKs, LMPPs, CLPs, and Ly6d⁺CLPs. Numbers refer to percentages of progenitors in total BM-nucleated cells. (F) Total cell numbers. +/+, CreERT2; Bcl11a^+/+. +/flox, CreERT2; Bcl11a^+/flox. flox/flox, CreERT2; Bcl11a^flox/flox. *, P < 0.05; **, P < 0.01.

Figure 3.

Severe defects in B, T, and NK cell precursors and lymphoid progenitor compartments in recipient mice engrafted with Lin⁻ BM cells (CD45.1⁻). Lethally irradiated recipient mice (CD45.1⁺) engrafted with BM cells of indicated genotypes were treated with Tam and analyzed 4 wk later (n = 4/group, and the experiment was repeated 4 times). Panels show total cell numbers (bar graphs) and/or flow cytometry dot plots of donor BM B cells (A), donor thymocytes (B), donor BM NK cells (C and D), and donor lymphoid progenitors (E). Numbers in the flow cytometry plots refer to percentages of gated cells in total donor nucleated BM cells or thymocytes. *, P < 0.05; **, P < 0.01.

Figure 4.

Apoptosis in Bcl11a-deficient B cell precursors could be rescued by exogenous Bcl2. (A) Apoptosis was measured by flow cytometry in Bcl11a-deficient early B cells in the BM. Numbers in histograms are percentages of the indicated gates. (B) Western blot analysis of Bcl11a protein at indicated time points after Tam treatment. (C) Apoptosis in pro–B cells (IMDM supplemented with 50 ng/ml of IL-7, Flt3L, and SCF) after Bcl11a deletion detected by Annexin V and 7-AAD. Numbers in the flow cytometry plots refer to percentages of gated cells. (D) 1 million cultured pro–B cells were treated with Tam for 4 d, and live cells were enumerated at indicated time points. (E) Detection of intracellular Bcl2 in cultured pro–B cells at indicated time points after Tam treatment. (F) Apoptosis in cultured pro–B cells infected with retrovirus-expressing Bcl2. GFP⁺ cells (infected) were gated for analysis. An empty retroviral vector expressing GFP was used as the control. (G) qRT-PCR analysis of key lymphoid genes in cultured Bcl11a-deficient pro–B cells. Data represent mean values of three independent biological replicates and all values are normalized to Gapdh expression. Error bars indicate the SD. (H) Two-way hierarchical clustering of expression of 100 important lymphoid genes (Table S4) in cultured pro–B cells. Expression of these genes in wild-type DN3 thymocytes was used as an expression control. Scale indicates the log2 value of normalized signal level. For A, C, D, E, F, and G, data represent at least three independent experiments.

Figure 5.

Genetic interaction of Bcl11a and p53 in lymphoid development. (A) Cell cycle of cultured pro–B cells infected with retrovirus expressing Bcl2 and/or Mdm2 was analyzed by Edu/7-AAD staining. (B) 3 d after Tam treatment, 1 million pro–B cells were cultured and counted daily for viability. (C) qRT-PCR analysis of p21, p27, and p57 in Bcl11a-deficient pro–B cells infected with the Bcl2-expressing retrovirus. (D and E) Presence of pro–B, pre–B, and immature B cells (D), or lymphoid progenitors (E), in BM of Dflox/flox mice 5 d after Tam treatment. Numbers refer to percentages of gated cells in total donor nucleated BM cells. Data in A, B, C, D, and E represent at least three independent experiments. *, P < 0.05; **, P < 0.01.

Figure 6.

p53 deficiency did not rescue long-term B cell development or lymphoid progenitor defects. Lethally irradiated recipient mice (CD45.1⁺) were transplanted with Lin⁻ BM cells (CD45.1⁻) and were treated with Tam to delete Bcl11a. Recipients were subsequently analyzed 4 wk later (n = 4/genotype). (A) Total cell numbers of donor BM B cells of indicated genotypes. (B) Total numbers of donor LSKs, LMPPs, and CLPs. (C) Freshly sorted CLPs (5,000) from mice of indicated genotypes 3 d after Tam treatment were cultured on OP9 stromal cells in the presence of 10 ng/ml of IL-7 and Flt3L for 14 d. A–C represent at least three independent experiments. *, P < 0.05; **, P < 0.01.

Figure 7.

Loss of lymphoid development potential in LSKs of flox/flox and Dflox/flox mice. Flow cytometric–sorted LSKs from Tam-treated mice were transplanted to recipients with helper BM cells. Donor cells (CD45.1⁻) were analyzed 8 wk after transplantation. (A) Total numbers of donor LSK, HSC, LMPP, CLP and ETP of indicated genotypes. (B) Populations of B, T, and NK cells analyzed by flow cytometry. Numbers indicate percentages of lymphocytes in total BM-nucleated cells. (C) Apoptosis in Bcl11a-deficient stem cells or progenitors (red line) analyzed by Annexin V staining. Apoptosis in these cells were rescued by p53 deficiency (green line). BM cells were harvested and stained 4 d after Tam treatment. (D and E) qRT-PCR analysis of cell cycle, lymphoid, and myeloid genes in freshly sorted HSCs. (F) qRT-PCR analysis of lymphoid and myeloid genes in LMPPs. Data represent mean values of three independent biological replicates and all values are normalized to Gapdh expression. Error bars indicate the SD. A–F represent at least three independent experiments. *, P < 0.05; **, P < 0.01.

Figure 8.

Analysis of Bcl11a binding sites in mouse pro–B cells. (A) qPCR validation of Bcl11a binding at the Mdm2, Mdm4, and Bcl2 loci in ChIP assay. Potential Bcl11a binding sites (red boxes) at the Mdm2 and Mdm4 loci, predicted as GGCCGG-containing sequences. Black bars indicate the qPCR-amplified regions of the validated sites. Potential Bcl11a binding sites (arrows) at the mouse Bcl2 locus, predicted as homologous regions to BCL11A-binding sites at the human BCL2 locus (ENCODE project), are analyzed by ChIP assay. The fold-enrichments of the amplified regions are (from 5′ to 3′) are as follows: 1.31 ± 0.40, 1.69 ± 0.32, 5.32 ± 0.92 (red arrow), 1.87 ± 0.001, 2.37 ± 0.60, 2.80 ± 0.43, 1.32 ± 0.21, 2.12 ± 0.44, 1.88 ± 0.21, 2.29 ± 0.40, 3.00 ± 0.16, and 2.68 ± 0.72. (B) Distribution of Bcl11a-binding sites in the genome from ChIP-Seq analysis. The 10-kb region upstream from transcription start site is defined as Proximal Promoter, and the 10-kb region downstream from transcription stop site is defined as downstream. (C) qPCR validation of Bcl11a binding at the genomic loci of Bcl-xL, p21, Pax5, and E2A. Genomic DNA pull-down using IgG was used as a control. A–C represent three independent experiments. *, P < 0.05; **, P < 0.01.