Cooperative interaction of Angiopoietin-like proteins 1 and 2 in zebrafish vascular development

Abstract

Angiopoietin-like protein (Angptl) 1 and Angptl2, which are considered orphan ligands, are highly homologous, particularly in the fibrinogen-like domain containing the putative receptor binding site. This similarity suggests potentially cooperative functions between the two proteins. In this report, the function of Angptl1 and Angptl2 is analyzed by using morpholino antisense technology in zebrafish. Knockdown of both Angptl1 and Angptl2 produced severe vascular defects due to increased apoptosis of endothelial cells at the sprouting stage. In vitro studies showed that Angptl1 and Angptl2 have antiapoptotic activities through the phosphatidylinositol 3-kinase/Akt pathway, and coinjection of constitutively active Akt/protein kinase B mRNA rescued impaired vascular development seen in double knockdown embryos. These results provide a physiological demonstration of the cooperative interaction of Angptl1 and Angptl2 in endothelial cells through phosphatidylinositol 3-kinase/Akt mediated antiapoptotic activities.

Fig. 1.

Vascular defects observed in Angptl1 and Angptl2 double knockdown embryos. (A) Representative photographs of simple microscopic observation of 72 hpf embryos St, Angptl1MO, Angptl2MO, and DMO groups. The lower four photographs are high magnifications of the upper four. A closed arrowhead indicates pericardial effusion seen especially in DMO groups (Scale bar: 400 μm.). (B) Representative photographs of angiography at 48 hpf. The lower four photographs are high magnifications of the upper four. Open arrowheads indicate the absence of blood flow in ISVs and dorsal-longitudinal anastomosus vessels. (C) Representative photographs of whole-mount ISH of fli-1 at 48 hpf. Filled arrowheads indicate absence of ISVs and dorsal-longitudinal anastomosus vessels in DMO.

Fig. 2.

Impaired sprouting of ISVs and elevated apoptotic cells around the DA are seen in DMO embryos. (A) Representative photographs of whole-mount ISH of flk-1 at 28 hpf in St and DMO embryos. High magnification images of head portions indicated by dotted grids in the upper left are shown in the lower left. Normal vasculature in head portion is seen in both St and DMO. High magnification images in the trunks indicated by closed grids in the upper left are shown in the right images. Filled arrowheads indicate ISVs with impaired sprouting in DMO. (B) Representative photographs of whole-mount ISH of fli-1 at 12 hpf and flk-1 at 18 hpf in St and DMO embryos. Normal vasculogenesis occurs in DMO-injected embryos. (C) Representative photographs of whole-mount TUNEL assay at 28 hpf. High magnification images of the trunk indicated by dotted grids in Upper are shown in Lower. Open arrowheads indicate increased apoptotic cells in the area of major trunk vessels in DMO. (D) A sagittal section of DMO embryos stained with TUNEL. Open arrowheads indicate apoptotic cells around the DA, and open arrows indicate naturally occurring apoptosis in the neural tube. NC, notochord. (Scale bar: 50 μm.). (E) Representative photographs of somites examined by whole-mount ISH of myoD at 12 hpf, simple microscopic observation at 18 hpf, hematoxylin/eosin staining of sagittal sections at 72 hpf, and whole-mount ISH of vegf at 24 hpf. Somitogenesis and vegf expression are not impaired in DMO embryos. (Scale bar: 100 μm.).

Fig. 3.

Angptl1 and Angptl2 bind to endothelial cells, and possess antiapoptotic activity through the PI3-K/Akt pathway. (A) Western blotting analysis with an anti-FLAG antibody with (Left) or without (Right) 2-mercaptoethanol (2ME). (B and C) Binding of COMP-Angptl1-FLAG and Angptl2-FLAG to HUVECs. The FITC intensity indicates cells bound by FLAG-tagged proteins. With both COMP-Angptl1 (Left) and Angptl2 (Right), intensities increased in a dose-dependent manner with saturation at ≈2 μg/ml. (D) Phosphorylation assay of ERK1/2 and Akt. HUVECs were treated with COMP-Angptl1 or -Angptl2 proteins. Immunoblotting was performed with anti-pERK1/2 or anti-pAkt antibody. Total amounts of ERK1/2 or Akt proteins were monitored by reprobing membranes with anti-ERK1/2 antibody and anti-Akt antibody. (E) TUNEL assay. HUVECs were cultured in the presence of either vehicle (PBS) or 0.5 μg/ml of COMP-Angptl1 or -Angptl2. In some experiments, cells were incubated with DMSO or 5 μg/ml PD980059 or LY294002. Data shown is the average of 6 fields each in six independent experiments (n = 36). ^*, P < 0.03 (compared with PBS in each group).

Fig. 4.

Coinjection of myristoylated Akt rescues vascular defects in DMO embryos. (A) Representative photographs of whole-mount ISH of flk-1 at 28 hpf, whole-mount TUNEL assay at 28 hpf, angiography at 48 hpf, and ISH of fli-1 at 48 hpf. Fifty picograms of EGFP mRNA or myr-Akt mRNA was coinjected into DMO (2 ng each of Angptl1mo and Angptl2mo). As a control, we coinjected 50 pg of EGFP mRNA with 4 ng of the standard morpholino (St plus EGFP). Rescued blood vessels are indicated by filled arrowheads and open arrowheads. (B) Percentages of embryos with impaired sprouting of ISVs in five independent experiments (n = 31-42 in each experiment) ^*, P < 0.01. (C) EGFP activity in the EGFP coinjected group (Left) and the myr-Akt coinjected group (Right) at 24 hpf and 48 hpf. Open arrowheads indicate abundant EGFP expression in the trunk of an embryo at 24 hpf.