Abstract
Abstract 907
The hematopoietic system is a highly regulated cellular hierarchy, responsible for the day-to-day production of mature blood cells which can be divided in two major lineages, myeloid and lymphoid. Hematopoietic stem cells (HSCs) have the unique ability to give rise to all hematopoietic cell types, by first generating lineage-commited progenitors which in turn will produce terminally differentiated cells. HSCs are characterized by their extensive self-renewal and differentiation capacities. While in mice the mechanisms underlying early HSC differentiation and lineage determination are well understood at the molecular level, very few transcription factors regulating lineage decisions have been identified in human hematopoiesis. Our group has recently established a novel cell sorting strategy for human HSCs and early lineage committed progenitors (Doulatov et al., Nature Immunology, 2010; Notta et al., Science, 2011) which uncovered the existence of a novel human multilymphoid progenitor (MLP). MLPs give rise to all lymphoid cell types, as well as dendritic cells and monocytic cells.
Here we report a comprehensive analysis of gene expression at each developmental stage of the early human hematopoietic hierarchy, ranging from the long-term repopulating stem cells to lineage-restricted progenitors through multipotent progenitors such as MLP, CMP (common myeloid progenitor), GMP (granulocyte-monocyte progenitor) and MEP (megakaryocyte-erythroid progenitor). We show that hematopoietic specification is defined by a small number of global gene expression clusters that correspond to major biological lineages and that lineage programs in committed progenitors are paired with HSC-shared priming programs. HSCs display most extensive priming along the lympho-myeloid branch (MLP). In contrast early progenitors of the megakaryocytic/erythrocytic lineage form a distinct cluster, highly enriched for cell cycle genes.
To identify regulators of each major developmental transition, we computationally extracted population-specific gene-sets (“signatures”). We then integrated transcription factor expression data and enrichment of transcription factors binding sites in the promoters of each “signature” to obtain a map of transcriptional regulators in the context of the developmental hierarchy. Based on this model, we selected more than 15 candidate genes for functional validation. We chose genes predicted to act either on lymphoid (MLP), myeloid (MLP, CMP) or erythroid (MEP) commitment. Among these, we investigated the function of BCL11a, a C2H2 zinc finger transcriptional repressor, which expression is primed in HSCs then peaks in the newly discovered MLP population, indicating a putative role in lymphocyte specification. Consistent with this hypothesis, BCL11a has been implicated in the development of B cell progenitors in mouse. When BCL11a was knocked down in cord blood derived hematopoietic stem cells and early progenitors, we observed reduced formation of cells committed to the B cell fate both in vitro and in an in vivo xenograft assay. BCL11a knock-down resulted in a partial block of B cell maturation at the proB to preB cell transition, that was accompanied by a decrease in the key B cell maturation transcription factor, Pax5. These preliminary results suggest that BCL11a directs B cell specification in human and that our genome-wide strategy not only provides a valuable resource for the hematology community but also allows identification of key regulators of early human lineage commitment.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.