Redundancy Between Ets-1 and Ets-2 during Mouse Embryogenesis: Regulators of Angiogenesis?
Ets family transcription factors, characterized by a conserved Ets DNA binding domain, encompass more than thirty vertebrate members. Ets family members regulate important biological processes, including cell proliferation, differentiation, and apoptosis, and these factors can also contribute to human diseases, in particular to cancer. How specificity of target gene regulation is achieved within this large gene family is a question of general biological interest. One mechanism through which specificity can be achieved is via modification of specific family members by signal transduction pathways. For example, phosphorylation of the ets-1 and ets-2 by ras -dependent pathways in cultured cells leads to persistent expression of target genes, and ets-1 and ets-2 are phosphorylated at a conserved residue (Threonine 38 and Threonine 72, respectively) by the well-characterized ras -effector pathway, the Raf/MAPK pathway. Our lab has taken a genetic approach to characterize the in vivo role of ets-2 phosphorylation in mice. Published work has demonstrated that ets-2 knockout mice are embryonic lethal due to an extra-embryonic defect, while ets-1 knockout mice are viable and fertile. Recently a mouse "knockin" model has been created by Bob Oshima's lab in which the ets-2 phosphorylation site at threonine 72 is mutated to alanine (T72A). Surprisingly, mice homozygous for the ets-2 T72A allele were viable and fertile with no obvious abnormality. Because ets-1 and ets-2 are highly related genes, the absence of a phenotype in Ets-2 T72A/T72A mice could be caused by genetic redundancy of the two genes. To test this hypothesis in vivo , we combined the ets-2 T72A allele and the ets-1 knockout allele.
The double mutant mice die beginning at embryonic day 11.5 probably due to defects in angiogenesis. These mice with ets-1 and et-2 gene mutations have large, diluted blood vessels and lack the characteristic branching of the vascular system that occurs in control embryos, or in embryos having mutations in ets-1 or ets-2 genes alone.

We have observed no defects in placental development in these mutant embryos. Compared to controls, GFP-tagged endothelial cells isolated by FACS from the double mutant embryos expressed lower levels of Ets target genes, such as in particular anti-apoptotic genes like bcl-X, bcl-2, and c-IAP-2, as well as the extracellular protease MMP9.
Because anti-apoptotic genes were downregulated, apoptosis in endothelial cells in ets-1/ets-2 mutant mice was studied using TUNEL assays. A 6-fold increase in apoptosis was observed in the double mutant mice. Remarkably, apoptosis was restricted to PECAM-positive endothelia, even though all cells types in the mutant mice harbored the mutated ets-alleles. Using a conditional "floxxed" ets-2 allele created by our group in combination with Cre that is endothelial cell specific (tie-2 Cre), the same vascular branching phenotype and the same level of increased apoptosis was observed. These results indicate that ets-1 and ets-2 act to coordinate expression of involved with cell motility (MMP9) with genes that prevent cell death (Bcl-x, Bcl-2, c-IAP-2) in an endothelial cell autonomous manner. Coordination of expression of these genes is critical for angiogenesis.
The goal of these projects is to determine the relevance of ets-1 and ets-2 in angiogenesis during development and tumorigenesis.