Endothelial cell-specific chemotaxis receptor (ecscr) promotes angioblast migration during vasculogenesis and enhances VEGF receptor sensitivity

Abstract

Endothelial cell-specific chemotaxis receptor (ECSCR) is a cell surface protein expressed by blood endothelial cells with roles in endothelial cell migration and signal transduction. We investigated the function of ecscr in the development of the zebrafish vasculature. Zebrafish ecscr is expressed in angioblasts and in axial vessels during angioblast migration and vasculogenesis. Morpholino-directed ecscr knockdown resulted in defective angioblast migration in the posterior lateral plate mesoderm, a process known to depend on vascular endothelial-derived growth factor (VEGF). In cultured cells, transfected ECSCR localized to actin-rich membrane protrusions, colocalizing with kinase insert domain protein receptor (KDR)/VEGF receptor 2 in these regions. ECSCR-silenced cells show reduced VEGF-induced phosphorylation of KDR but not of FMS-like tyrosine kinase 1 (FLT1)/VEGF receptor 1. Finally, chemical inhibition of VEGF receptor activity in zebrafish resulted in angioblast deficiencies that partially overlap with those seen in ecscr morphants. We propose that ecscr promotes migration of zebrafish angioblasts by enhancing endothelial kdr sensitivity to VEGF.

Figure 1

Expression of zebrafish ECSCR. (A) qPCR measurement of ecscr transcript levels, normalized to beta-actin transcripts, during various stages of zebrafish development. Primer design is indicated in supplemental Figure 1. (B) Normalized qPCR of kdrl transcripts. (C) Dorsal view, with posterior up, of a 10-som embryo processed for ISH for the ets transcription factor etsrp showing the distribution of angioblasts in the posterior LPM. Scale bars for panels C through E, G, and I represent 200 μm. (D-I) Whole-mount ISH for ecscr message in zebrafish of the indicated stages. (D) 10-som, showing expression on premigratory angioblasts similar to etsrp⁺ cells. (E) Lateral view at 16 som. (F) Oblique view near the tail at 16 som, showing faint hybridization to migrating angioblasts (). (G) Lateral view at 18 som. (H) Dorsal view at 18 som. Hybridization is detected at the midline vascular cord. (I) Lateral view at 24 hpf. (J-K) Methacrylate sections of whole-mount in situs showing ecscr transcripts localized to midline vascular structures. (J) 18 som. (K) Higher magnification view of a 24-hpf section. point to DA and PCV. Scale bar represents 10 μm.

Figure 2

Knockdown of ECSCR2 results in aberrant positioning of migrating angioblasts. (A-G) Dorsal views of the posterior of control MO (cMO)–injected or ecscr morpholino (MO1, MO2) zebrafish embryos of the indicate stages subjected to whole-mount ISH for etsrp. (A-C) 8 som. Angioblasts are detected in apparently normal numbers and position in morpholino-injected embryos. Scale bar represents 200 μm. (D-G) 14 som. cMO-injected embryos show etsrp⁺ angioblasts arriving at the midline () and convergence of the premigratory rows toward the midline, whereas anti-ecscr morphants show a greater gap between angioblast rows and few midline angioblasts. MO2 morphants showed a greater range of morpholino phenotypes, with more severely affected morphants (G) showing an irregular row of angioblasts with occasional angioblasts positioned away from the midline ( in panel G). (H-J) Lateral view of 16-som embryos. cMO-injected embryos and the bulk of MO1- and MO2-injected embryos show a consolidated midline vascular cord (H), whereas approximately 20% of MO1 and MO2 morphants (I-J) show increased numbers of premigratory angioblasts lateral and dorsal to the consolidating vascular cord. Results at each embryonic stage are representative of at least 20 embryos from 2 independent injections.

Figure 3

Transfected ECSCR and endogenous KDR colocalize at the edge of cultured endothelial cells. Human ECSCR cDNA was generated with a C-terminal V5 tag and overexpressed in HUVECs. (A) Triple label confocal micrograph comparing transfected ECSCR (green) to phalloidin (red) and DAPI (4,6 diamidino-2-phenylindole; blue). Transfected ECSCR is enriched in actin-rich membrane protrusions. Scale represents 40 μm. (B-D) Higher magnification view of a different cell labeled with anti-V5 (top), phalloidin (center), and ARPC2, a component of the Arp2/3 complex (bottom). ECSCR localizes to areas of active membrane protrusion. (E-G) Comparison of ECSCR with KDR. Confocal antitag (panel D and red signal in panel F) and antiendogenous KDR (panel E and green signal in panel F) cytochemistry. The punctate vesicular pattern of KDR overlaps with ECSCR at convex membrane protrusions where KDR approaches the membrane (seen as yellow), but not away from the cell edge. Scale bar represents 10 μm. Results were replicated in 3 independent experiments.

Figure 4

Knockdown of ECSCR in HUVECs results in reduced migration, apoptosis, and proliferation. (A) HUVECs and HPAECs were transfected with ECSCR siRNA or control lacZ siRNA and assayed for transmigration activity in Boyden chambers. ECSCR knockdown cells showed reduced transmigration activity relative to controls in response to 10% fetal bovine serum (HUVECs: 38% of control; HPAECs: 57% of control; P < .05 for both, Student paired t test). (B) ECSCR siRNA or control lacZ siRNA-transfected HUVEC transmigration activity in response to purified VEGF (25 ng/mL). (ECSCR knockdown, 117% of unstimulated vs 157% of unstimulated for control knockdown; P < .05.) (C) Knockdown cells were contact inhibited then analyzed for annexin translocation by flow cytometry. ECSCR knockdown resulted in strongly reduced apoptosis compared with lacZ controls (11% annexin positive vs 36% in controls; P < .01). (D) HUVECs were transfected with siRNA constructs, contact inhibited, serum starved, and then released from inhibition in the presence or absence of VEGF. Proliferation was assayed by incorporation of [³H]thymidine. ECSCR knockdown resulted in greatly decreased VEGF-stimulated thymidine incorporation (104% of unstimulated vs 281% in control cells; P < .001). Results are representative of 3 independent experiments.

Figure 5

Knockdown of ECSCR in HUVECs results in reduced VEGF-induced phosphorylation of VEGF receptor 2/KDR and downstream PLCβ3, but not tyrosine phosphorylation of VEGF receptor 1/FLT1. (A) HUVECs transfected with the indicated siRNAs were serum starved overnight and stimulated with 25 ng/mL VEGF for the indicated periods. Total lysates were blotted using the indicated phosphoepitope-specific antisera. ECSCR knockdown cells show strongly reduced tyrosine phosphorylation of KDR on multiple epitopes. (B) Total lysates as described in panel A were blotted using anti-PLCβ3 phosphoepitopes. Activating phosphorylation of serines 537 and 1105 was reduced in ECSCR knockdown cells. (C) FLT1 was immune precipitated from unstimulated and VEGF-stimulated cells (5 minutes), and immune precipitates were blotted for phosphotyrosine. Tyrosine-phosphorylated FLT1 was present at similar levels in VEGF-stimulated control lacZ and ECSCR knockdown cells. Results were observed in at least 2 independent experiments.

Figure 6

Inhibition of VEGF receptors in zebrafish results in reduced etsrp⁺ angioblast numbers including reduced numbers at the midline. Embryos were treated from 10 to 16 hpf with vehicle control DMSO (A) or VEGF receptor inhibitors ZM323881 (1 μm; B) or SU4516 (10 μm; C), and angioblast positions were analyzed by whole-mount ISH for etsrp. Inhibitor-treated embryos show reduced angioblast numbers, and strongly reduced etsrp⁺ cells at the midline. (D) ecscr morphants show disrupted anterior vascular cord. Each result is representative of more than 20 embryos from 2 independent experiments. Scale bar in panel A represents 200 μm.