Figure 1.
Flk1 expression precedes Etv2 expression during embryogenesis.
(A) Diagram of the transgenic construct used in this study. Numbers shown identify the genomic location relative to the translation start site of Etv2. CR-1 represents conserved region one and CR-2 represents conserved region two. (B) Schematic diagram of the embryonic axes and germ layers of a late streak stage embryo. Black lines indicate approximate levels of sections in C. (C) A series of transverse sections of a late streak stage embryo. Representative sections were stained with antibodies to Brachyury (Bry), Flk1 and EYFP (scale bar: 100 microns).
Figure 2.
Etv2 is coexpressed with Flk1.
(A–Q) Expression analysis of Etv2-EYFP and Flk1 in E7.0 to E9.5 embryos. Sections were stained with the GFP antibody (A, B, D, E, G–I, K, L, N, P; shown in green) or the Flk1 antibody (A, C, D, F–H, J, K, M, O, Q; shown in red). DAPI nuclear staining is shown in blue (N–Q). Yellow indicates overlap of green and red channels (A, D, G, H, K). A–C: No bud stage. D–F: Early-bud stage. G–M: Early head-fold stage. N–Q: E9.5. The boxed area in A is enlarged in B–C; the boxed area in D is enlarged in E–F; the boxed area in G is enlarged in H–J, and the boxed area in K is enlarged in L–M. Arrowheads in B, C point to progenitors of blood islands coexpressing Flk1 and EYFP. Arrows and arrowheads in H–M indicate endothelial lineages co-expressing Flk1 and EYFP. Asterisks in A, H–M indicate regions in which either gene is expressed alone. Arrowheads in P indicate developing intersomitic vessels. Structures are designated as follows (al: allantois, bi: blood island, da: dorsal aorta, h: heart, ec: endocardium, hf: head fold, nt: neural tube, ph: pharynx, s: somite). Bars in A, D, G, K, N, and P indicate 200 µm. Bars in C, F, J, and M indicate 50 µm.
Figure 3.
Etv2 mutant embryos phenocopy Flk1 null embryos.
(A–D) Whole mount β-galactosidase staining of Flk1 LacZ allele in a representative Etv2 wildtype, Flk1 heterozygous embryo (A); Etv2 mutant, Flk1 heterozygous embryo (B); Etv2 heterozygous, Flk1 null embryo (C); and Etv2 mutant, Flk1 null embryo (D). (E–H) Histological sections of β-galactosidase stained embryos counterstained with nuclear fast red in Etv2 wildtype, Flk1 heterozygous embryo (E); Etv2 mutant, Flk1 heterozygous embryo (F); Etv2 heterozygous, Flk1 null embryo (G); and Etv2 mutant, Flk1 null embryo (H). (I–J) FACS analysis of Flk1 stained Etv2 wildtype and mutant embryos. Representative FACS plots are shown (I) and data from all experiments are graphically represented (J) (SSC-W: side scatter- width; **p<0.01). Scale bars represent 200 µm. Arrowheads represent lateral plate mesoderm. Structures are designated as follows (h: heart; ec: endocardium; da: dorsal aorta; phv: primary heart vein).
Figure 4.
Etv2 is absent in Flk1 null embryos.
(A–C) FACS analysis of the Etv2-EYFP transgenic reporter crossed into the Flk1 wildtype and mutant backgrounds at E7.75 and E8.0. Representative FACS profiles are shown (A). (B–C) Data from all experiments are compiled graphically to show the percent of EYFP positive cells in Flk1 wildtype, heterozygous, and null embryos (B) and the mean fluorescent intensity of the EYFP positive cells (C). (D–E) Quantitative PCR for Flk1 (D) and Etv2 (E) was performed on RNA isolated from Flk1 wildtype and null embryos at E7.75, E8.0, and E9.0. (F) Quantitative PCR for Etv2 was performed on FACS isolated Etv2-EYFP positive cells. Each bar represents triplicate measurements from independent samples (single embryos). Asterisks indicate the following: **p<0.01, *p<0.05.
Figure 5.
VEGF activates the Etv2 promoter.
(A–B) Transcriptional assays using an empty luciferase vector or the 3.9 kb Etv2 luciferase reporter standardized to the CMV-Renilla Luciferase reporter. Flk1 cDNA or a balancer DNA plasmid is co-transfected and the cells are treated with or without VEGF (50 ng/ml) for 2–18 hours as indicated in the panel (A). Flk1 or a balancer DNA is cotransfected and cells are treated for 6 hours with varying amounts of VEGF (0–50 ng/ml) as indicated in the panel (B). (C–D) Transcriptional inhibition assays in which Flk1 cDNA and the 3.9 kb Etv2 Luciferase promoter are co-transfected followed by pretreatment of various doses of SB202190 (C), or GF109203X (D) and a 6 hour treatment of 50 ng/ml VEGF. Data were analyzed by two-way anova (A,C,D) or one-way anova (B). n.s.: not significant; *: p<0.05; ***: p<0.001; ****: p<0.0001.
Figure 6.
VEGF activation of the Etv2 promoter is dependent on a Creb binding site.
(A) A schematic of the 3.9 kb Etv2 luciferase reporter showing 3 evolutionary conserved Creb binding motifs (CRE1-3) and the mutations schematized in panel B. Numbers reflect genomic position relative to the translational start site of Etv2. (B) A schematic of the truncation and mutation strategy. × indicates a CG to AA mutation of either CRE1 or CRE2. Numbers indicate genomic position relative to the translational start site of Etv2. (C) Transcriptional assays using the constructs shown in B co-transfected with Flk1 cDNA and treated with or without VEGF. (D) Transcriptional assay using the 1 kb Etv2 luciferase cotransfected with increasing amounts of expression vector encoding constitutively phosphorylated Creb cDNA. Data were analyzed using two-way anova (C) or Kruskal-Wallis nonparametric t-test with Dunn’s post test comparison (D). n.s.: not significant; *: p<0.05; **: p<0.01; ****: p<0.0001.
Figure 7.
Creb1 binds to the CRE2 motif in the Etv2 promoter.
(A–B) Creb1 specifically interacts with the CRE2 motif in the Etv2 promoter as shown by ChIP assay. EBs were collected 3.5 days after initiating mesoderm differentiation. A region in the Gapdh gene was used as a negative control (Control) and 1% of the total chromatin DNA before the immunoprecipitation was used as a positive control (Input). The PCR products were run on a gel for direct visualization (A) and qPCR was performed for quantitation (B). (C) Creb1 binds CRE2 directly in vitro. Creb1 was synthesized in vitro. The oligonucleotides harboring CRE2 motif is labeled with 32P. Synthesized Creb1, non-radioactive oligonucleotides, and the antibodies were incubated with radioactive oligonucleotide probes as indicated in the figure. The interaction between Creb1 and the CRE2 motif was analyzed on a 4% TBE gel.