The Notch ligand DLL4 specifically marks human hematoendothelial progenitors and regulates their hematopoietic fate

Notch signaling is essential for definitive hematopoiesis, but its role in human embryonic hematopoiesis is largely unknown. We show that in hESCs the expression of the Notch ligand DLL4 is induced during hematopoietic differentiation. We found that DLL4 is only expressed in a sub-population of bipotent hematoendothelial progenitors (HEPs) and segregates their hematopoietic versus endothelial potential. We demonstrate at the clonal level and through transcriptome analyses that DLL4(high) HEPs are enriched in endothelial potential, whereas DLL4(low/-) HEPs are committed to the hematopoietic lineage, albeit both populations still contain bipotent cells. Moreover, DLL4 stimulation enhances hematopoietic differentiation of HEPs and increases the amount of clonogenic hematopoietic progenitors. Confocal microscopy analysis of whole differentiating embryoid bodies revealed that DLL4(high) HEPs are located close to DLL4(low/-) HEPs, and at the base of clusters of CD45+ cells, resembling intra-aortic hematopoietic clusters found in mouse embryos. We propose a model for human embryonic hematopoiesis in which DLL4(low/-) cells within hemogenic endothelium receive Notch-activating signals from DLL4(high) cells, resulting in an endothelial-to-hematopoietic transition and their differentiation into CD45+ hematopoietic cells.

Ligand-independent FLT3 activation does not cooperate with MLL-AF4 to immortalize/transform cord blood CD34+ cells

MLL-AF4 fusion is hallmark in high-risk infant pro-B-acute lymphoblastic leukemia (pro-B-ALL). Our limited understanding of MLL-AF4-mediated transformation reflects the absence of human models reproducing this leukemia. Hematopoietic stem/progenitor cells (HSPCs) constitute likely targets for transformation. We previously reported that MLL-AF4 enhanced hematopoietic engraftment and clonogenic potential in cord blood (CB)-derived CD34+ HSPCs but was not sufficient for leukemogenesis, suggesting that additional oncogenic lesions are required for MLL-AF4-mediated transformation. MLL-AF4+ pro-B-ALL display enormous levels of FLT3, and occasionally FLT3-activating mutations, thus representing a candidate cooperating event in MLL-AF4+ pro-B-ALL. We have explored whether FLT3.TKD (tyrosine kinase domain) mutation or increased expression of FLT3.WT (wild type) cooperates with MLL-AF4 to immortalize/transform CB-CD34+ HSPCs. In vivo, FLT3.TKD/FLT3.WT alone, or in combination with MLL-AF4, enhances hematopoietic repopulating function of CB-CD34+ HSPCs without impairing migration or hematopoietic differentiation. None of the animals transplanted with MLL-AF4+FLT3.TKD/WT-CD34+ HSPCs showed any sign of disease after 16 weeks. In vitro, enforced expression of FLT3.TKD/FLT3.WT conveys a transient overexpansion of MLL-AF4-expressing CD34+ HSPCs associated to higher proportion of cycling cells coupled to lower apoptotic levels, but does not augment clonogenic potential nor confer stable replating. Together, FLT3 activation does not suffice to immortalize/transform MLL-AF4-expressing CB-CD34+ HSPCs, suggesting the need of alternative (epi)-genetic cooperating oncogenic lesions.