To be or not to be, unraveling molecular mechanisms for lineage decisions in developing blood cells

Abstract: All hematopoietic cells originate from hematopoietic stem cells (HSC) residing in the BM. The process of differentiation, through which HSCs generate progenitors and subsequently mature blood cells has been extensively studied but many of the regulatory mechanisms involved remains elusive. The classical model for hematopoiesis has been established based on the identification of common myeloid precursors and common lymphoid precursors, suggesting that the first lineage commitment step results in the strict separation between the lymphoid and myeloid lineages. However, by subfractionation of the LIN-SCA1+KIT+ bone marrow population, containing all HSCs in the mouse, based on expression of CD34 and FLT3, three functionally distinct populations were identified. CD34- cells contained the HSCs, CD34+FLT3- short term repopulating cells and CD34+ FLT3+/hi cells with combined lymhpoid and GM potential but, only a small fraction (3%) displaying MkE potential in vitro. In agreement with this MkE associated genes (Gata1, Epor, Vwf etc) were found to be down-regulated in CD34+FLT3hi cells while lymphoid associated genes (Il7r, Rag1, Tdt, sterile IgH transcripts) were found to be up-regulated both at the population and single cell level. Based on this we proposed a new model for hematopoietic development were the first restriction point involves the separation of MkE and lymphoid potential while retaining GM potential. In order to investigate the continued development of B-lymphoid cells we investigated the functional role of the transcription factor EBF1, known to be of critical importance for B cell development. By means of comparativ and correlativ analysis of microarray data from cell-lines we, in addition to the previously known EBF1 target genes, identified Vpreb3, CD19, Ceacam1 and CD53 as EBF1 target genes. Furthermore, by transplantations of EBF1 deficient fetal liver cells, we were able to show EBF1 to be dispensable for the generation of CLPs but that lack of EBF1 expression resulted in diminished IgH DJ recombination and reduced expression of B-lineage associated genes already at this early stage of development. Using a hCD25 lambda5-promoter transgene as a marker for EBF1 activity a small CLP subpopulation (~6%) could be identified. This population demonstrated diminished T-cell potential, increased B-cell potential, as well as up-regulation of Vpreb, lambda5, B29, Mb1 and importantly Pax5. Taken together this suggest that B-lineage commitment occurs already at the level of the CLP.