Critical roles for c-Myb in lymphoid priming and early B-cell development

Abstract

c-Myb is a transcription factor with functions in many hematopoietic lineages. c-Myb-deficient mice display reduced numbers of B cells; however, it is unknown what role c-Myb plays in B lymphopoiesis because no critical target genes have been identified in the B-cell lineage. We demonstrate that conditional deletion of c-Myb in B-cell progenitors completely abolishes B-cell development. c-Myb is required for lymphoid progenitors to respond to the cytokines interleukin-7 and thymic stromal lymphopoietin; in the absence of sufficient c-Myb activity, mice display a B lymphopenia that closely resembles that observed in interleukin-7 receptor alpha-deficient animals. Analysis of the multipotent progenitor compartment indicates that c-Myb is also required for up-regulation of multiple lymphoid-associated genes, including Il7r, and for the subsequent development of the common lymphoid progenitor population. These data show that c-Myb plays a critical role in the regulatory pathways governing lymphoid specification and early B-cell differentiation.

Figure 1

c-Myb is essential for B-cell development. (A-B) Flow cytometry of bone marrow of 3- to 4-week-old c-Myb^+/+, c-Myb^Plt4/+, c-Myb^Plt4/Plt4, and c-Myb^Δmb1/Δmb1 mice. Numbers adjacent to the boxes indicate proportion of (A) total B cells (B220⁺CD19⁺) and (B) immature (B220^loIgM⁺) and mature (B220^hiIgM⁺) B cells among all live cells (mean ± SD of 4-10 mice per genotype). (C) Flow cytometry of density-fractionated and lineage-depleted bone marrow of 6- to 9-week-old c-Myb^+/+, c-Myb^Plt4/+, c-Myb^Plt4/Plt4, and c-Myb^Δmb1/Δmb1 mice. CLPs were identified as Sca-1⁺Thy1⁻ c-kit^int IL-7Rα⁺; the gating strategy is available in supplemental Figures 1 and 3. Numbers in plots indicate the proportion of cells in each gated area (mean ± SD of 5-6 mice per genotype). (D-E) Flow cytometry of bone marrow of 3- to 4-week-old c-Myb^+/+, c-Myb^Plt4/+, c-Myb^Plt4/Plt4, and c-Myb^Δmb1/Δmb1 mice. (D) Pre-pro-B cells were identified as CD19⁻ CD11c⁻ NK1.1⁻ B220⁺CD43^lo; the gating strategy is available in supplemental Figures 1 and 3. Numbers in plots indicate the proportion of cells in each gated area (mean ± SD of 4-10 mice per genotype). (E) The numbers above the boxes in the top plots indicate the proportion of pro-B cells (B220⁺c-kit⁺) among all live cells. The numbers above the boxes in the bottom plots indicate the proportion of committed B cells (CD19⁺) among B220⁺c-kit⁺ cells. Data are the mean ± SD of 4 to 10 mice per genotype. *P < .05; **P < .01; ***P < .001 in test group vs c-Myb^+/+ cells.

Figure 2

The B-cell defect in c-Myb^Plt4/Plt4 mice is more severe in older mice. (A) Flow cytometry of adult (8-10 weeks old) c-Myb^+/+, c-Myb^Plt4/+, and c-Myb^Plt4/Plt4 mice. Top indicates total B cells; bottom, pro-B cells. Populations were identified as described in Figure 1. Plots are representative of 3 mice per genotype. (B) Flow cytometry of fetal liver cells from e16.5 embryos. Numbers in the plots represent the proportion of cells within the gated population expressed as the mean ± SD of 7 to 21 embryos per genotype. (Top) Numbers above the boxes indicate percentages of total B cells (B220⁺CD19⁺) among all live cells. (Bottom) Numbers above the boxes indicate percentages of pro-B cells (CD19⁺c-kit⁺) among all live cells. *P < .05; **P < .01; ***P < .001 in test group vs c-Myb^+/+ cells.

Figure 3

c-Myb is required for lymphoid progenitors to respond to IL-7 and TSLP. (A) Limiting dilution analysis of LSK cells isolated from the bone marrow of 6- to 8-week-old c-Myb^+/+ and c-Myb^Plt4/Plt4 mice. Cultures were performed on OP9 or OP9-DL1 stromal cells in the presence of IL-7 and Flt3L. *P < .025. Data represent the mean ± SEM of 5 samples per genotype. (B) Limiting dilution analysis of CLPs isolated from the bone marrow of 6- to 9-week-old c-Myb^+/+ and c-Myb^Δmb1/Δmb1 mice. Cultures were performed on OP9 or OP9-DL1 stromal cells in the presence of IL-7 and Flt3L. The difference in B-cell precursor frequency was not significant (P > .025), probably because of incomplete deletion of the floxed allele; see panel C. Data represent the mean ± SEM of 2 to 3 samples per genotype. (C) PCR genotyping of B-cell colonies generated from c-Myb^Δmb1/Δmb1 CLPs. CLPs were isolated from c-Myb^fl/fl or c-Myb^Δ/fl mice carrying the mb1^Cre allele. After 8 days of culture on OP9 stromal cells, individual B-cell colonies were harvested, and DNA was extracted for genotyping at the c-Myb locus. The deleted (Δ), floxed (fl), and wild type (+, derived from the OP9 cells) alleles are indicated. Data are representative of 26 colonies analyzed. Sizes in base pairs are indicated on the right. (D) Limiting dilution analysis of pro-B cells (B220⁺CD19⁺c-kit⁺) isolated from the fetal liver of c-Myb^+/+, c-Myb^Plt4/+, and c-Myb^Plt4/Plt4 e15.5 to e16.5 embryos. Cultures were performed on OP9 stromal cells in the presence of either IL-7 or TSLP. *P < .025; **P < .005; ***P < .001. Data represent the mean ± SEM of 2 to 8 embryos per genotype. Only 2 small colonies were observed from c-Myb^Plt4/Plt4 pro-B cells stimulated with IL-7; hence, the number in the graph represents an overestimate of precursor frequency. The precursor frequency in panels A, B and D was defined as the inverse of the number of cells per well at which more than 37% of the wells lacked a colony. (E) Intracellular flow cytometry for phosphorylated STAT5 (pSTAT5) after IL-7 stimulation of c-Myb^+/+ and c-Myb^Plt4/Plt4 bone marrow. Plots are gated on B220⁺CD43⁺ pro-B cells and are representative of 4 mice per genotype. The red line indicates cells without IL-7; the blue line indicates cells stimulated with IL-7.

Figure 4

Analysis of genes differentially expressed in c-Myb^Plt4/Plt4 pro-B cells. (A) Semiquantitative RT-PCR of gene expression in pro-B cells (B220⁺CD19⁺c-kit⁺) isolated from the bone marrow of 3- to 4-week-old c-Myb^+/+ and c-Myb^Plt4/Plt4 mice. Wedges indicate serial dilution of cDNA. Data are representative of 2 experiments. (B) Flow cytometry of Sca-1 (Ly6a) surface expression on pro-B cells from 3-week-old mice of the indicated genotypes. Plots are representative of 3 mice per genotype. (C) Quantitative real-time PCR of gene expression in pro-B cells (B220⁺CD19⁺c-kit⁺) isolated from the bone marrow of 3- to 4-week-old c-Myb^+/+ and c-Myb^Plt4/Plt4 mice. Expression is normalized to Hprt1 and is presented as relative to that in one of the wild-type samples, set as 1. *P < .05 in test group vs c-Myb^+/+ cells. Data represent the mean ± SEM of 3 to 6 samples of each genotype.

Figure 5

c-Myb is required for normal expression of IL-7Rα on lymphoid progenitors. (A) Flow cytometry of IL-7Rα surface expression on pro-B cells from 3- to 4-week-old mice. The upper panel is gated on B220⁺CD19⁺c-kit⁺ cells; the lower panel is gated on B220⁺c-kit⁺ cells. Plots are representative of 3 to 9 mice per genotype. (B) Flow cytometry of splenocytes from c-Myb^+/+ and c-Myb^Plt4/Plt4 mice. Numbers adjacent to the boxes indicate percentages of CD4⁺ and CD8⁺ T cells among all live cells. Dot plots are representative of 3 mice per genotype. Note that the relative increase in CD8⁺ and CD4⁺ cells is due to the loss of B cells in c-Myb^Plt4/Plt4 mice. (C) Flow cytometry of IL-7Rα surface expression on splenic T cells from c-Myb^+/+ and c-Myb^Plt4/Plt4 mice, gated for CD8⁺ or CD4⁺ cells as indicated. Plots are representative of 3 mice per genotype. (D) Flow cytometry of density-fractionated and lineage-depleted bone marrow of 5- to 9-week-old c-Myb^+/+ and c-Myb^Plt4/Plt4 mice on a Rag1^gfp/+ background. Cells at the CLP stage of development were identified as Sca-1⁺c-kit^int GFP⁺; numbers in plots indicate the percentage of cells in each gated area. (E) Flow cytometry of IL-7Rα surface expression on Sca-1⁺c-kit^int GFP⁺ cells. Plots are representative of 4 to 5 mice per genotype.

Figure 6

c-Myb^Plt4/Plt4 LSKs display reduced expression of lymphoid genes. (A) Heat map depicting expression of 27 lymphoid-specific genes that display altered expression in c-Myb^Plt4/Plt4 LSK samples (P < .1). (Left) Expression pattern of these genes in progenitor populations. Pre-GM indicates pre–granulocyte-macrophage progenitor; Mk progenitor, megakaryocyte progenitor; pre–CFU-E, erythroid progenitor pre–colony forming unit; pre-MegE, pre–megakaryocyte-erythroid progenitor. (Right) Expression pattern in c-Myb^+/+ and c-Myb^Plt4/Plt4 LSK samples. Red squares denote up-regulated genes, blue squares denote down-regulated genes, with scale denoting z score, ie, standard deviation from mean expression. (B) Flow cytometry of density-fractionated and lineage-depleted bone marrow of 6- to 9-week-old c-Myb^+/+ and c-Myb^Plt4/Plt4 mice. Upper plots display the proportion of Sca-1⁺c-kit^hi cells among lineage⁻ cells (mean ± SD of 5-6 mice per genotype). Lower plots display the percentage of CD34⁺Flt3⁺ cells within the LSK fraction. Plots are representative of 5 to 6 mice per genotype. (C) Flow cytometry of density-fractionated and lineage-depleted bone marrow of 5- to 7-week-old c-Myb^+/+ and c-Myb^Plt4/Plt4 mice on a Rag1^gfp/+ background. Early lymphoid progenitors were identified as CD27⁺ GFP⁺ cells within the LSK fraction, identified as in panel B. Numbers in plots indicate the proportion of cells in the gated area. Data are the mean ± SD of 6 to 8 mice per genotype. (D) Graph depicting the percentage of LMPPs within the LSK fraction of c-Myb^+/+ and c-Myb^Plt4/Plt4 bone marrow. *P < .05 in test group vs c-Myb^+/+ cells. Data represent the mean ± SEM of 5 to 6 mice per genotype.

Figure 7

IL-7Rα cannot rescue B-cell development from c-Myb deficient fetal liver cells. (A) Generation of CD19⁺ B cells from retrovirally infected c-Myb^Plt4/Plt4 fetal liver cells. Fetal liver cells from e14.5 embryos were depleted of Ter119⁺ cells and transduced with control retrovirus or retrovirus encoding c-Myb, Ebf1, Bcl2, constitutively active STAT5A(1*6), or IL-7Rα; all retroviruses included an IRES GFP to identify infected cells. Retrovirally infected cells were grown on OP9 stromal cells for 8 days in the presence of IL-7 and Flt3L. Numbers above the boxes indicate percentages of B cells (CD19⁺) among GFP⁺ cells. Plots are representative of 6 independent experiments. (B) Quantitation of CD19⁺ B cells generated from c-Myb^Plt4/Plt4 fetal liver cells transduced with retrovirus as in panel A. The number of GFP⁺ CD19⁺ cells was counted after 4, 6, and 8 days of culture. Data represent the mean ± SEM of 3 independent experiments. (C) Flow cytometry of IL-7Rα surface expression on B cells generated from c-Myb^Plt4/Plt4 fetal liver cells transduced with control, c-Myb, or Ebf1 retrovirus. Plots are gated on GFP⁺ CD19⁺ cells. Data are representative of 6 independent experiments. (D-E) Generation of CD19⁺ B cells from retrovirally infected (D) c-Myb^+/+ or (E) c-Myb^Δmb1/Δmb1 CLPs. Cells were transduced and cultured as in panel A. Plots are representative of 2 independent experiments.