Comment on Ema et al, page
New evidence shows that Flk1 (VEGFR-2) is broadly and transiently expressed in multipotent mesoderm during development, indicating that Flk1 expression does not uniquely define the hematovascular lineage.
Although expression of the VEGF receptor Flk1 (VEGFR-2) has been used extensively to define the vascular and hematovascular lineages, Ema and colleagues now provide convincing evidence that Flk1 expression is actually quite promiscuous within lineages derived from mesoderm. They show this in mouse embryos by analyzing a knock-in lacZ reporter allele with the drug selection cassette removed by recombination. Prior to recombination, this was the original knockout allele used by this group to define a critical function for Flk1 during vascular development.1 Removal of the drug selection cassette expanded the lacZ expression pattern significantly, and in several instances Ema et al were able to confirm expression of Flk1 protein in the same nonvascular lineages that express the reporter. However, despite the expanded expression, Flk1 is not required for development outside of the hematovascular compartment.
This finding is notable because Flk1 and its human ortholog VEGFR-2 have been used as markers of the vascular and/or the hematovascular lineage in numerous cases, and this finding reinforces the idea that it is dangerous to rely on any single marker to define a lineage or progenitor population. In fact, Flk1 expression was documented in the nervous system almost 10 years ago.2,3 This example of nonvascular Flk1 expression is intriguing but occurs long after the neural ectoderm has diverged from the hematovascular mesoderm. More recently, investigators working with ES cells showed transient Flk1 expression in a majority of cells between days 3 and 4 of differentiation.4 Yamashita and colleagues5 isolated these early Flk1-expressing cells from ES cell cultures and showed that they were multipotent mesoderm, giving rise to both smooth muscle-like and endothelial cells. However, the relevance of these findings to lineage relationships in vivo remained unclear. Now it appears that transient expression of Flk1 by multipotent mesoderm is a hallmark of both ES cell differentiation and development in vivo.
Ema and colleagues also show that some mature mesoderm derivatives, such as cardiomyocytes, express Flk1. In an attempt to separate endothelial from nonendothelial Flk1 expression, they deleted a portion of intron 1 implicated in endothelial Flk1 expression, but this alteration did not restrict Flk1 expression. Thus, the cis-acting sequences that selectively regulate Flk1 expression in vivo remain elusive. The authors conclude that levels of Flk1 expression may determine different lineages, because the vascular and hematopoietic lineages express much higher levels of Flk1 than do the nonvascular compartments, but this hypothesis has not yet been tested.
These results are intriguing in light of the intense interest in stem cells and progenitor cells. The relationships among different lineages and the plasticity of stem and progenitor cells are subjects of controversy in numerous instances, and marker expression has been used extensively to define both these populations and their relationships. Thus the relative promiscuity of Flk1, once thought to be a gold-standard hematovascular marker, indicates that recombination-based lineage tracing and functional assays are required to sort out the relationships and properties of developing cells. ▪
High-power magnification of Flk1-Pax3 double-positive cells. See the complete figure in the article beginning on page .
High-power magnification of Flk1-Pax3 double-positive cells. See the complete figure in the article beginning on page .
