The transcriptional coactivator Maml1 is required for Notch2-mediated marginal zone B-cell development

Abstract

Signaling mediated by various Notch receptors and their ligands regulates diverse biological processes, including lymphoid cell fate decisions. Notch1 is required during T-cell development, while Notch2 and the Notch ligand Delta-like1 control marginal zone B (MZB) cell development. We previously determined that Mastermind-like (MAML) transcriptional coactivators are required for Notchinduced transcription by forming ternary nuclear complexes with Notch and the transcription factor CSL. The 3 MAML family members (MAML1-MAML3) are collectively essential for Notch activity in vivo, but whether individual MAMLs contribute to the specificity of Notch functions is unknown. Here, we addressed this question by studying lymphopoiesis in the absence of the Maml1 gene. Since Maml1(-/-) mice suffered perinatal lethality, hematopoietic chimeras were generated with Maml1(-/-), Maml1(+/-), or wild-type fetal liver progenitors. Maml1 deficiency minimally affected T-cell development, but was required for the development of MZB cells, similar to the phenotype of Notch2 deficiency. Moreover, the number of MZB cells correlated with Maml1 gene dosage. Since all 3 Maml genes were expressed in MZB cells and their precursors, these results suggest that Maml1 is specifically required for Notch2 signaling in MZB cells.

Figure 1

Maml genes are differentially expressed in hematopoietic and lymphoid progenitor subsets. Quantitative RT-PCR was performed on cDNA isolated from multipotent BM progenitors, thymocyte subsets, BM B lineage progenitors, and splenic B-cell subsets from C57BL/6 mice fractioned by flow cytometry. Maml1-Maml3 transcript levels were normalized using Hprt expression. The Maml1/Hprt ratio in LSK progenitors was normalized to 1. The measurements were performed in triplicates. Results are shown as means plus or minus SD. (A) Relative expression of Maml1-Maml3 in multipotent BM progenitors (LSK subset) and thymocyte subsets. LSK indicates Lin⁻Sca-1⁺c-kit⁺; ETP, early T lineage progenitors (Lin⁻c-Kit⁺CD25⁻); DN2, CD4⁻CD8⁻ double-negative 2 (Lin⁻c-Kit⁺CD25⁺); DN3, double-negative 3 (Lin⁻c-Kit⁺CD25⁻); and DP, CD4⁺CD8⁺ double-positive thymocytes. (B) Maml expression in BM B-cell progenitor subsets. CLP indicates common lymphoid progenitors (Lin⁻Sca-1^loc-kit^loIL-7Rα⁺Flt3⁺); ProB, pro-B cells (B220⁺CD43⁺AA4.1⁺CD19⁺sIgM⁻); PreB, pre-B cells (B220⁺CD43⁻AA4.1⁺sIgM⁻); and ImmB, immature B cells (B220⁺CD43⁻AA4.1⁺sIgM⁺). (C) Maml expression in splenic B-cell subsets. Tr1-Tr3 indicates successive stages of immature transitional B cells (Tr1, B220⁺AA4.1⁺sIgM^hiCD23⁻; Tr2, B220⁺AA4.1⁺sIgM^hiCD23⁺; Tr3, B220⁺AA4.1⁺sIgM^intCD23⁺); FoB, follicular B cells (most abundant mature B-cell type in the spleen; B220⁺AA4⁻sIgM^intCD23⁺); MZP B cells, marginal zone B cell precursors (B220⁺AA4.1⁻sIgM^hiCD21^hiCD23⁺); MZB, marginal zone B cells (B220⁺AA4.1⁻sIgM^hiCD21^hiCD23^lo).

Figure 2

Normal T-cell development and decreased numbers of MZB cells in Maml1^+/− mice. (A) Flow cytometric analysis showed a similar percentage of CD4⁺CD8⁺ double-positive and CD4⁺ or CD8⁺ single-positive cells in the thymus of Maml1^+/− mice compared with wild-type littermates. Numbers indicate the percentage of cells in each quadrant. (B) Flow cytometric analysis of splenic B-cell subsets showing a reduced percentage of B220⁺CD21^hiCD23^lo MZB cells in Maml1^+/− compared with wild-type spleens. Numbers indicate the percentage of cells in the rectangular box identifying MZB cells. (C) Data from multiple mice were compiled, indicating significant Maml1 dosage-dependent decrease in MZB cells. Each triangle represents a data point for a single mouse (Maml1^+/+, ▴, n = 5; Maml1^+/−, ▵, n = 5).

Figure 3

Characterization of splenic B-cell subsets in fetal liver hematopoietic chimeras reveals an absolute dosage-dependent requirement for Maml1 in MZB development. (A) Representative contour plots showing that MZB precursors (MZP B) and MZB populations were absent in Maml1^−/− fetal liver chimera, while reduced to about half of normal in Maml1^+/− fetal liver chimera compared with the wild-type controls. Host-derived CD45.1⁺ cells were excluded from the analysis. (B) Compilation of data collected from all the fetal liver chimeras that were analyzed (Maml1^+/+, n = 9; Maml1^+/−, n = 21; Maml1^−/−, n = 9). MZB cells were essentially absent in Maml1^−/− chimeras and decreased in numbers in Maml1^+/− heterozygous chimeras. Differences were highly statistically significant (P < .01; Student t test).

Figure 4

Preserved development of αβ lineage T cells in the absence of Maml1. (A) Flow cytometric analysis showed a similar percentage of CD4⁺CD8⁺ double-positive and CD4⁺ or CD8⁺ single-positive thymocytes in recipients of Maml1^−/− compared with Maml1^+/− progenitors. Numbers indicate the percentage of cells in each quadrant. Host-derived CD45.1⁺ cells were excluded from the analysis. (B) Normal distribution of Lin⁻ thymocyte progenitor subsets, as defined using c-Kit and CD25 expression, in recipients of Maml1^−/− compared with Maml1^+/− progenitors. Numbers indicate the percentage of cells in each quadrant. Host-derived CD45.1⁺ cells were excluded from the analysis. (C) Absolute number of donor-derived double-positive thymocytes in all the fetal liver chimeras that were analyzed (Maml1^+/−, ▵, n = 21; Maml1^−/−, ▴, n = 9). The trend for decreased numbers of double-positive thymocytes in Maml1^−/− compared with Maml1^+/− progenitors was not statistically significant (P = .14; Student t test).

Figure 5

Increased numbers of γδ lineage T cells in the absence of Maml1. (A) Flow cytometric analysis showed an increased percentage of TCRγ⁺ cells among CD3⁺ thymocytes in recipients of Maml1^−/− compared with Maml1^+/− progenitors (representative examples). Numbers indicate the percentage of cells in each box. Host-derived CD45.1⁺ cells were excluded from the analysis. (B) Absolute number of donor-derived CD3⁺ γδ T cells in the thymus of all the fetal liver chimeras that were analyzed (Maml1^+/−, ▵, n = 21; Maml1^−/−, ▴, n = 9). The increased number of γδ T cells in Maml1^−/− compared with Maml1^+/− recipients was statistically significant (P < .05; Student t test).