A tale of two lineages: the origins of blood and endothelium

Anthony R. Green

Comment on D'Souza et al, page 3862, and Schlaeger et al, page 3871

The origins of blood and its conduits are intimately intertwined. Two papers in this issue shed further light on the role of the SCL/TAL-1 transcription factors in hematopoietic and endothelial development.

Studies of embryonic stem cells originally identified the murine hemangioblast as a cell (the blast colony-forming cell [BL-CFC]) that generates colonies containing both hematopoietic and endothelial progeny.1 It was subsequently realized that the BL-CFC also gives rise to vascular smooth muscle,2 and similar BL-CFCs have recently been identified in the mouse embryo.3 As a consequence, a degree of semantic confusion has crept into the literature with some authors equating the hemangioblast with the BL-CFC and others restricting the terms to a bipotent cell with only hematopoietic and endothelial potential.

The BL-CFCs described so far appear to be the progenitors of yolk sac hematopoiesis since they can generate primitive erythropoiesis, together with some definitive lineages, but not lymphoid cells or hematopoietic stem cells (HSCs). In the para-aortic splanchnopleura the emergence of HSCs is also closely associated with the endothelium, and it therefore seems probable that this region contains a similar progenitor capable of generating both HSCs and endothelium. A developmental link between blood and the vasculature is highly conserved in evolution with cells analogous to the hemangioblast described in Drosophilia.4 In addition, mounting evidence suggests that bipotent progenitors of blood and endothelium are not only restricted to the embryo but also present in adult mice.5

The stem cell leukemia (SCL) gene encodes a basic helix-loop-helix (bHLH) protein, which is critical for blood and endothelial development. Mice lacking SCL failed to generate any hematopoietic cells, and although endothelial cells are formed, they are functionally abnormal. In contrast to its essential role in the formation of early hematopoietic cells including HSCs, SCL expression is not needed for the maintenance of adult HSCs, although it is required subsequently for normal erythroid and megakaryocyte differentiation.6,7 D'Souza and colleagues and Schlaeger and colleagues now shed light on the precise timing of SCL-dependent and -independent phases of hematopoietic development.

D'Souza and colleagues demonstrate that the vast majority of BL-CFCs from day-3 embryoid bodies do not express SCL but that SCL is switched on within the first 24 hours of colony formation from an individual progenitor. Moreover, they show that SCL-/- embryonic stem cells can give rise to BL-CFCs, that the cells produce only vascular smooth muscle in the absence of SCL, and that blood and endothelial development can be rescued by expression of exogenous SCL. The authors conclude that SCL is not needed for the formation of BL-CFCs (which they equate with hemangioblasts) but is required for hematopoietic commitment.

Schlaeger and colleagues investigate the time point at which HSC maintenance becomes SCL independent. Mice with a conditional SCL allele were bred with mice carrying a Tie2Cre transgene. The progeny (SCLfl/flTie2Cre) died by embryonic day 13.5 to 14.5 and exhibited defective erythroid and megakaryocyte differentiation but, importantly, contained long-term repopulating hematopoietic stem cells in the fetal liver. SCL is therefore essential for hematopoietic commitment during a narrow window of time that extends from the onset of SCL expression and ends before Tie2Cre-mediated inactivation of SCL. Although it was unclear precisely when the Tie2 regulatory elements become active in this transgenic line, these results are broadly consistent with a previous embryonic stem (ES) cell study, which also showed that SCL is required at an early stage of hematopoietic commitment, prior to vascular endothelial–cadherin (VE-cadherin) expression.8

The 2 papers in this issue underscore the central role played by SCL in the development of both blood and vasculature. Outstanding issues include the mechanisms by which SCL itself is regulated and the identity of key SCL target genes. ▪