Search

Point mutations that trigger ligand-independent proteolysis of the Notch1 ectodomain occur frequently in human T-cell acute lymphoblastic leukemia (T-ALL) but are rare in murine T-ALL, suggesting that other mechanisms account for Notch1 activation in murine tumors. Here we show that most murine T-ALLs harbor Notch1 deletions that fall into 2 types, both leading to ligand-independent Notch1 activation. Type 1 deletions remove exon 1 and the proximal promoter, appear to be RAG-mediated, and are associated with mRNA transcripts that initiate from 3′ regions of Notch1. In line with the RAG dependency of these rearrangements, RAG2 binds to the 5′ end of Notch1 in normal thymocytes near the deletion breakpoints. Type 2 deletions remove sequences between exon 1 and exons 26 to 28 of Notch1, appear to be RAG-independent, and are associated with transcripts in which exon 1 is spliced out of frame to 3′ Notch1 exons. Translation of both types of transcripts initiates at a conserved methionine residue, M1727, which lies within the Notch1 transmembrane domain. Polypeptides initiating at M1727 insert into membranes and are subject to constitutive cleavage by γ-secretase. Thus, like human T-ALL, murine T-ALL is often associated with acquired mutations that cause ligand-independent Notch1 activation.

PU.1 is a hematopoietic-specific transcriptional activator that is absolutely required for the differentiation of B lymphocytes and myeloid-lineage cells. Although PU.1 is also expressed by early erythroid progenitor cells, its role in erythropoiesis, if any, is unknown. To investigate the relevance of PU.1 in erythropoiesis, we produced a line of PU.1-deficient mice carrying a green fluorescent protein reporter at this locus. We report here that PU.1 is tightly regulated during differentiation—it is expressed at low levels in erythroid progenitor cells and down-regulated upon terminal differentiation. Strikingly, PU.1-deficient fetal erythroid progenitors lose their self-renewal capacity and undergo proliferation arrest, premature differentiation, and apoptosis. In adult mice lacking one PU.1 allele, similar defects are detected following stress-induced erythropoiesis. These studies identify PU.1 as a novel and critical regulator of erythropoiesis and highlight the versatility of this transcription factor in promoting or preventing differentiation depending on the hematopoietic lineage.

The Ikaros gene encodes a zinc finger transcription factor that is selectively expressed by all hematopoietic cells. Although Ikaros is required for lymphocyte differentiation, its role in the myeloid lineage is unclear. We show here that Ikaros expression is temporally regulated during neutrophil differentiation: Ikaros is primarily expressed at immature stages and significantly less so in mature neutrophils. Furthermore IkL/L mice, harboring a hypomorphic mutation at the Ikaros locus, exhibit several defects during neutrophil differentiation. (1) IkL/L fetal livers contain high numbers of neutrophil lineage cells, and this increase is reflected in the number of GM-CSF-dependent progenitor cells. (2) The migratory potential and survival of neutrophil progenitors is altered in vitro. (3) Expression of the Gr-1 marker is delayed and repressed. In contrast, neutrophil function appears normal. These data demonstrate that Ikaros regulates early neutrophil differentiation but is dispensable in mature neutrophils.

The Notch pathway is frequently activated in T-cell acute lymphoblastic leukemias (T-ALLs). Of the Notch receptors, Notch1 is a recurrent target of gain-of-function mutations and Notch3 is expressed in all T-ALLs, but it is currently unclear how these receptors contribute to T-cell transformation in vivo. We investigated the role of Notch1 and Notch3 in T-ALL progression by a genetic approach, in mice bearing a knockdown mutation in the Ikaros gene that spontaneously develop Notch-dependent T-ALL. While deletion of Notch3 has little effect, T cell–specific deletion of floxed Notch1 promoter/exon 1 sequences significantly accelerates leukemogenesis. Notch1-deleted tumors lack surface Notch1 but express γ-secretase–cleaved intracellular Notch1 proteins. In addition, these tumors accumulate high levels of truncated Notch1 transcripts that are caused by aberrant transcription from cryptic initiation sites in the 3′ part of the gene. Deletion of the floxed sequences directly reprograms the Notch1 locus to begin transcription from these 3′ promoters and is accompanied by an epigenetic reorganization of the Notch1 locus that is consistent with transcriptional activation. Further, spontaneous deletion of 5′ Notch1 sequences occurs in approximately 75% of Ikaros-deficient T-ALLs. These results reveal a novel mechanism for the oncogenic activation of the Notch1 gene after deletion of its main promoter.

Understanding how multipotent cells commit to each of their terminal fate potentials is an important aspect of stem cell biology. In adult murine hematopoiesis, HSCs with long-term self-renewal potential reside within the Lin −Sca-1+c-Kit+ (LSK) fraction having CD34−, Thy1lo, and Flt3/Flk2−phenotypes. The LSK cells having CD34+, Thy1−, and/or Flt3+ phenotypes are capable of multi-lineage reconstitution only for a short-term, and therefore should contain multipotent progenitors (MPPs). In terms of developmental steps downstream of MPPs, there has been a controversy. The existence of prospectively isolatable progenitors capable of generating only myeloerythroid cells (common myeloid progenitor: CMP) or only lymphocytes (common lymphoid progenitors: CLP) outside the LSK fraction suggests that the first commitment step after the MPP stage is the strict bifurcation of lymphoid vs. myeloid pathway. Recently, however, several studies have suggested that the lineage commitment could occur within the LSK fraction, preceding the proposed bifurcation of myeloid and lymphoid pathways. For example, MPPs expressing Flt3 at a high level retained granulocytes/monocytes (GM) but not megakaryocyte/erythrocyte (MegE) potential together with lymphoid potential, suggesting the existence of common progenitor for GM and lymphoid lineages within the LSK fraction. Based on this data, the coupled loss of self-renewal activity and MegE potential in the early HSC commitment has been proposed. How can we reconcile the controversy in the early hematopoietic lineage map? By utilizing mice having GATA-1 or PU.1 transcriptional reporters, we here present a high-resolution map containing new lineage-restricted progenitor populations within the MPP population of CD34+ LSK phenotype. Initial upregulation of GATA-1 and PU.1 occurred independently at the MPP stage. The GATA-1+ MPP displayed potent myeloerythroid potential without giving rise to lymphocytes, whereas the PU.1+ MPP showed granulocyte/monocyte/lymphoid-restricted progenitor activity without megakaryocyte/erythroid differentiation. Furthermore, GATA-1+ and PU.1+ MPPs possessed huge expansion potential, and differentiated into the original CMPs and CLPs, respectively. Thus, the reciprocal activation of GATA-1 and PU.1 primarily organizes hematopoietic lineage fate decision to form the earliest hematopoietic branchpoint that comprises new, isolatable myeloerythroid and myelolymphoid progenitor populations.

Understanding how multipotent cells commit to each of their terminal fate potentials is an important aspect of stem cell biology. Hematopoietic stem cells (HSCs) of Lin −Sca-1+c-Kit+ (LSK) phenotype have been purified, which were further divided into CD34-long-term and CD34+ short-term (ST)-HSCs. The existence of phenotypically isolatable common lymphoid progenitors (CLPs) and common myeloid progenitors (CMPs) downstream of ST-HSCs suggests that the first commitment step after the HSC stage is the bifurcation of lymphoid vs. myeloid pathway. Recent studies, however, suggest that the loss of MegE potential could be an early event in HSC stage. For example, LSK cells activating RAG-1 or Flt-3 expression retained granulocyte/monocyte (GM) but not megakaryocyte/erythrocyte (MegE) potential together with lymphoid potential, suggesting the existence of common progenitor for GM and lymphoid lineages. Here we report that a fraction of ST-LSK cells expressing high levels of PU.1, a transcription factor necessary for GM and lymphoid development, represents GM/lymphoid bipotent progenitors. In mice harboring knock-in GFP reporter for PU.1, LSK cells were divided into GFP high and GFP low subpopulations. Although PU.1low LSK cells were multipotent, PU.1high LSK cells differentiated only into GM and lymphoid cells in vitro and in vivo. We also found that single PU.1high LSK cells differentiated into GM, T and B cells in vivo. These data formally prove the existence of the third major early progenitor population activating PU.1 at a high level, the granulocyte/monocyte/lymphoid progenitor (GMLP). The existence of prospectively isolatable GMLPs strongly suggests that HSCs sequentially lose MegE then GM potential during their lymphoid commitment.