Syx, a RhoA Guanine Exchange Factor, Is Essential for Angiogenesis In Vivo
Rho GTPases play an important and versatile role in several biological processes. In this study, we identified the zebrafish ortholog of the mammalian Rho A guanine exchange factor, synectin-binding guanine exchange factor (Syx), and determined its in vivo function in the zebrafish and the mouse. We found that Syx is expressed specifically in the vasculature of these organisms. Loss-of-function studies in the zebrafish and mouse point to a specific role for Syx in angiogenic sprouting in the developing vascular bed. Importantly, vasculogenesis and angioblast differentiation steps were unaffected in syx knockdown zebrafish embryos, and the vascular sprouting defects were partially rescued by the mouse ortholog. Syx knockdown in vitro impairs vascular endothelial growth factor-A–induced endothelial cell migration and angiogenesis. We have also uncovered a potential mechanism of endothelial sprout guidance in which angiomotin, a component of endothelial cell junctions, plays an additive role with Syx in directing endothelial sprouts. These results identify Syx as an essential contributor to angiogenesis in vivo.
Two distinct processes, vasculogenesis and angiogenesis, characterize vascular development in vertebrates.1,2 During vasculogenesis, vessels form de novo from endothelial cell (EC) precursors, or angioblasts, that coalesce at the midline to form lumenized tubes. Once primary vessels have formed, ECs sprout from preexisting vasculature to form secondary vessels via angiogenesis. ECs proliferate, migrate, and differentiate to form mature vasculature. Directed cell migration is necessary for angiogenesis and is governed in part by Rho-family GTPases, including RhoA, Cdc42, and Rac1.3 Activation of Rho GTPases at the cell periphery by Rho guanine exchange factors (GEFs)4 leads to remodeling of the actin cytoskeleton and consequently to cell migration.5
Recently, a novel Rho-GEF, synectin-binding RhoA exchange factor (Syx), also named GEF720, PLEKHG5, and Tech,6 was identified by yeast 2-hybrid analysis using Synectin as bait7 (M. Simons, personal communication). The mouse Syx ortholog is expressed as 2 splice variants that differ by only 2 C-terminal amino acid residues.8 Full-length Syx1 contains a PDZ-binding (postsynaptic density 95, disk large, zona occludens-1) motif9 required for the interaction of Syx1 with Synectin and localization to the plasma membrane. The shorter variant, Syx2, lacks the PDZ motif and is diffusely distributed in the cytoplasm. Syx1 augments EC migration and tube formation, whereas Syx2 does not.8 The first report on Syx localized it to band 1p36, in the distal region of human chromosome 1.6 This chromosomal region is rearranged in several types of cancer, suggesting that Syx may be involved in malignant transformation. The same report detected high expression levels of Syx in the brain and the heart, both highly vascularized organs.6 More recently, a missense mutation in Syx was linked with a degenerative motor neuron disease.10 Given the well-known similarities between signaling pathways in the nervous and vascular systems,11 the expression and functions of Syx in both system is not surprising. The most recent annotations of the human genome classified Syx as belonging to a 7-member family of GEFs, PLEKHG5 (pleckstrin homology domain containing, family G [with Rho-GEF domain] member 5). The functions of most members of this family are still unknown.
To investigate the function of Syx in vertebrate vascular development, we identified the Syx ortholog in zebrafish and performed whole mount in situ hybridization (ISH). Syx is expressed in the dorsal aorta (DA) and intersomitic vessels (ISVs). Syx vascular expression is recapitulated in mammals where the mouse Syx protein is observed in the DA and coronary vessels. Both gain- and loss-of-function analyses in zebrafish demonstrated specific defects in ISV sprouting. Injection of mouse syx mRNA partially rescued these defects, suggesting evolutionary conservation of Syx function. Furthermore, we found that Syx and angiomotin, a regulator of EC migration,12–14 act additively during angiogenesis. These results implicate for the first time a Rho-GEF that functions specifically in vertebrate angiogenesis in vivo.
Materials and Methods
Zebrafish Stocks and Reagents
Zebrafish were maintained at 28.5°C15 under Medical College of Wisconsin guidelines (protocol no. 312-06-2). Mating was carried out at 28.5°C, and embryos were staged according to established protocols.16 Morpholino (MO)1 and MO2 were designed by Gene Tools (Philomath, Ore) to target a syx intron–exon boundary and the syx ATG start codon, respectively. MO1, AGCTGTTTCTGTGTGGCCTGCTGA; MO2, CATGCCTTCGCCAATAGAACATCGT. Splice site nucleotides are italicized. The Amot MO sequence has been published previously.14
Whole Mount ISH
Wild-type (WT) embryos were grown in 0.003% phenylthiourea until the desired stage, fixed overnight at 4°C in 4% paraformaldehyde, dechorionated, and stored in 100% methanol at −20°C until use. Whole mount ISH was performed as described17 using etsrp, fli, flk, flt4, grl, and syx probes. Digoxigenin-labeled sense and antisense syx probes were transcribed from a XmnI-linearized vector containing 541 bp of syx using T7 and SP6 RNA polymerases, respectively. Zebrafish syx cDNA used to make the RNA probes was obtained from Open Biosystems (Clone: 65260512, Huntsville, Ala).
For information regarding Tg (fli: EGFP)clom39−/+ zebrafish and Cre/loxP syx−/− mouse generation, refer to the online data supplement, available at http://circres.ahajournals.org.
We used WT and syx knockout mice fixed at end diastole. The mice were anesthetized by ketamine/xylazine, heparinized, and euthanized by perfusion with citric KCl. After a PBS wash, mice were fixed by perfusion with 4% paraformaldehyde. A MV-132 (Flow-Tech) contrast medium was infused into the coronary system via the aorta or into the renal arteries. The hearts were imaged by a General Electric eXplore Locus SP apparatus at a resolution of 13 μm, and the vasculature was reconstructed by Microview (GE) software.
Two zebrafish National Center for Biotechnology Information sequences (accession numbers XM_686228.1 and XM_686228.2, encoding 1143-aa and 858-aa proteins, respectively) match mouse and human Syx, which are 1073 and 1091 aa, respectively. Amino acids 14 to 858 of the shorter zebrafish sequence (Figure IB in the online data supplement) are identical to amino acids 297 to 1141 of the longer sequence (supplemental Figure IA), suggesting that alternative start sites generate 2 proteins with different N termini. We have also detected multiple isoforms of syx by RT-PCR (data not shown). The 1143-aa Syx protein contains a (Dbl-homology (DH) (supplemental Figure IA, black arrows) and pleckstrin homology (PH) domains (supplemental Figure IA, gray arrows), and a PDZ-binding motif (ASEV) at its C terminus (Figure 1A). Its amino acid sequence is 60% and 59% homologous to mouse (supplemental Figure IA) and human (data not shown) orthologs, respectively. Human syx is located on chromosome 1, which is syntenic to linkage group 11, where zebrafish syx is located. Based on synteny and conserved domain structure, we conclude that the zebrafish syx gene reported here is the ortholog for human syx.
Syx Is Expressed in the Developing Zebrafish and Mouse Vasculature
To determine the spatial and temporal expression of syx, we performed whole mount ISH on zebrafish embryos. A syx antisense digoxigenin-labeled RNA probe was generated from an expressed sequence tag (Open Biosystems, Clone ID 65260512), which spans 541 bp of the 5′ end of the longer syx transcript and 203 bp of the shorter transcript. Syx is expressed maternally in 4-cell-stage embryos (supplemental Figure IIA and IIB). Expression continues in blastomeres at 3 hours postfertilization (hpf) (supplemental Figure IIC) and is ubiquitous in 12-somite embryos (supplemental Figure IID). At 18 somites (supplemental Figure IIE), syx expression is fairly ubiquitous and is observed along the midline (supplemental Figure IIF). Starting at 23 to 24 hpf (Figure 1B and 1C), syx transcripts are restricted to axial vessels (Figure 1B, black arrow). At 24 to 26 hpf (Figure 1D and 1E), syx expression is noted in the ISVs (Figure 1E, white asterisks). We also checked for Syx expression in mammalian tissue by probing cryosections of embryonic day (E)12.5 mouse embryos with anti-Syx (a custom-made antibody) (Figure 1F) and anti-platelet endothelial cell adhesion molecule (PECAM)-1/CD31 (Figure 1H), an EC marker. We found that Syx is expressed in the endothelium of the aorta, coronary vessels, and endocardium of the atria and left ventricle (Figure 1F and 1G). Syx was also detected in the ascending aorta (data not shown) but not in the vena cava, suggesting that Syx is an artery-specific protein. Both zebrafish and mouse share Syx aorta expression.
The endocardial expression of Syx in the mouse prompted us to investigate whether syx is differentially expressed in the cardiovascular genetic mutant cloche (clo) in zebrafish,18 which lacks the endocardial layer. Real time PCR analysis for syx showed that syx levels are reduced in clo−/− embryos compared to WT embryos (supplemental Figure IIG). We injected syx mRNA into Tg(fli1: EGFP)clom39−/− embryos where the Tg(fli1: EGFP) line has been crossed into the clom39+/− background and observed for rescue of clo defects. Syx mRNA–injected embryos from Tg(fli1: EGFP)clom39+/− mating pairs did not show significant differences in Mendelian ratios for edema or gain of the vascular marker fli1-EGFP compared to embryos from uninjected mating pairs (supplemental Figure IIH), suggesting that syx expression may be indirectly affected in the cloche mutant.
Knockdown of syx Perturbs ISV Development in the Zebrafish
We designed 2 MOs to knockdown (KD) syx transcripts in zebrafish. MO efficacy is shown in supplemental Figure III and discussed in the text of the online data supplement. The effect of syx KD on zebrafish vascular development was analyzed in transgenic fish carrying an endothelial-specific promoter of vascular endothelial growth factor receptor-2 (vegfr2 or flk)19 or friend leukemia integration factor-1 (fli1),20 which drive expression of green fluorescent protein (GFP) specifically in the vasculature. Transgenic embryos were injected with 8 ng of MO1 or MO2 at 1-cell stage, and vessel development was monitored via fluorescence microscopy. By 24 hpf, uninjected embryos showed ISVs that migrated dorsally between somites (Figure 2A, asterisk). In MO1-injected embryos, the ISVs were truncated, and the leading front of the tip cell was blunt (Figure 2C, asterisk). Occasionally, 1 or 2 ISVs were completely absent (Figure 2C). At 28 hpf, the growth of ISVs in MO1-injected embryos remained stagnant (Figure 2D, asterisk), whereas those of uninjected embryos continued to grow to form the dorsal longitudinal anastomotic vessel (Figure 2B, asterisk). A second MO targeting the ATG translational start site, MO2 (8 ng), also caused growth arrest of ISVs (Figure 2E and 2F, asterisk). Quantification of MO-injected embryos at 24 hpf (Figure 2G) revealed that 62% of Tg(fli1: EGFP) and 42% of Tg(flk: G-RCFP) MO1-injected embryos displayed ISV defects compared to only 10% and 7% of their age-matched uninjected siblings, respectively. Thirty percent of MO2-injected Tg(fli1: EGFP) embryos had ISV growth defects (Figure 2G), suggesting some variability in MO efficacy, which is expected because MO1 targeted both forms of the syx transcript, whereas MO2 targeted only the long form.
To determine if gastrulation delays were caused by MO injection, we performed ISH with the gastrulation marker myod in MO1-injected (supplemental Figure IIIE) and uninjected (supplemental Figure IIID) embryos at 24 hpf. We observed no difference in myod expression patterns between samples, suggesting that the vascular phenotype caused by MO1 injection did not result from delayed gastrulation.
To investigate whether vascular defects originated at a developmental stage that preceded ISV sprouting, we performed ISH with etsrp, an early angioblast marker.21 At 14 hpf (Figure 3A through 3D) or 18 hpf (Figure 3E through 3H), all of the sample groups displayed normal etsrp expression pattern. To confirm the phenotypes observed in Tg(fli1: EGFP) embryos, we performed ISH with the fli probe at 24 hpf in uninjected (Figure 3I) and MO-injected (Figure 3J and 3K) embryos. We observed a darker fli staining of the DA in MO1-injected embryos, similar to the more intense GFP signal in the Tg(fli1: EGFP) zebrafish (Figure 2C), suggesting the presence of more number of ECs. In MO1-injected embryos at 24 hpf, secondary fli+ ISVs sprouting from the DA appeared defective and were truncated (Figure 3J, white asterisk) or misguided (Figure 3J, white arrowhead). Some ISV ends were marked by darker staining in both MO1- and MO2-injected embryos similar to the swollen ends of GFP-expressing ISVs (asterisks in Figure 3J and 3K). Quantification at 24 hpf (Figure 3M) showed that 11% of the MO1-injected embryos had no more than 5 ISVs, whereas 42% had truncated ISVs. In MO2-injected embryos, 2% had no ISVs, 5% had less than five ISVs, and 44% had truncated ISVs. Similar results were observed in MO-injected flk+ embryos at 22 hpf (supplemental Figure IIIF). The most common ISV defect was growth arrest rather than complete absence, suggesting that syx primarily affected ISV directional migration and not the initial sprouting from the DA.
Syx Gain of Function Also Affects ISV Development
To determine whether syx gain-of-function can complement syx loss-of-function, we injected capped mouse syx mRNA and performed ISH using etsrp, flk, and fli probes. At 14 hpf, etsrp-probed syx RNA–injected (Figure 3D) embryos appeared normal, showing angioblast patterning comparable to uninjected and the MO1- and MO2-injected embryos. Similarly, at 18 hpf, we observed no vascular defects in flk expression (data not shown). In syx mRNA–injected fli ISH embryos, vascular defects were observed at 24 hpf where ISVs had blunt ends (Figure 3L, asterisk), mimicking the phenotype of the MO-injected embryos at the same time point. These result suggest that gain- and loss-of-function phenotypes are not complementary but similar.
syx Does Not Affect Artery Versus Vein Specification
In 24 hpf syx KD (Figure 3J and 3K) or syx mRNA–injected (Figure 3L) embryo, we observed an intense blue staining in the DA upon fli ISH. In particular, the fli+-stained region in syx mRNA–injected embryos expanded into both the DA and posterior cardinal vein (Figure 3L). Therefore, we checked the arterial marker gridlock (grl)22 at 24 hpf and the venous marker flt423 at 30 hpf to determine whether artery versus vein differentiation is affected in syx KD embryos. ISH with grl and flt4 antisense probes showed no difference in staining between MO or RNA-injected embryos and uninjected embryos (data not shown), suggesting that syx is not required for angioblast differentiation. Collectively, our ISH results confirm the transgenic embryo data indicating that syx plays an exclusive role in directing ISV sprout growth from the DA.
Vascular Defects Are Partially Rescued by syx Gene Complementation
To demonstrate MO specificity, we coinjected Tg(fli1: EGFP) embryos with syx sense mRNA and MO1 and compared them to embryos injected with MO1 alone. Initially, we performed rescue experiments with 100 pg of syx mRNA and observed no rescue in MO1-injected embryos (data not shown). Subsequently, we increased the dose of mRNA to 150 or 200 pg and observed a concomitant reduction in ISV defects in coinjected embryos when compared to MO1-injected embryos (Figure 3N). However, we were only able to partially rescue the phenotype with the 2 mRNA doses (12% and 15% rescue, respectively). When the amount of mRNA was further increased to 300 pg, the number of rescued embryos was reduced to 6%, suggesting a saturation effect. These results indicate that ISV sprouting requires a precise level of syx expression. Because the injection of mouse syx mRNA partially rescued the endogenous syx KD effects in zebrafish, we conclude that syx function is conserved across vertebrate species.
Disruption of syx Expression in the Mouse Causes Angiogenic Defects
To study Syx function in mammals, Cre loxP syx−/− mice were generated and crossed with germ-line Cre mice, producing a global disruption of syx expression. Syx−/− mice were viable up to at least 12 months and appeared grossly normal. Because syx is expressed in the cardiac and vascular tissue in the mouse (Figure 1F), and syx KD zebrafish shows defective ISVs, we imaged the coronary and kidney arterial systems by microcomputed tomography (μCT) of WT and syx−/− mice. Both the coronary (Figure 4A) and kidney (Figure 4B) arterial systems of the syx−/− mice were sparser than those in the WT mouse. On close examination, these systems appeared to be deficient in small diameter vessels, but the major coronary and kidney arteries were intact. This defect is analogous to the syx KD zebrafish, where the DA was intact but ISVs were truncated (Figure 2D and 2F). To quantify the coronary arterial defect, we used μCT images to measure the total volume of the contrast medium infused into the coronary arteries of the WT and syx−/− mice. We found that the ratio between the volumes of the syx−/− and WT coronary arterial systems was 0.75±0.11 (n=5, P=0.002).
Because the μCT technique cannot image capillaries due to the viscosity of the contrast medium, we probed sections of WT and syx−/− myocardia by histochemistry. The images clearly showed that the density of capillaries in the syx−/− was significantly lower than in the WT myocardium (Figure 4C). Together with the μCT images, these results demonstrate that the syx−/− mouse harbors a vascular defect affecting multiple organs. The major arteries were intact, but the growth of secondary arteries and capillaries was defective. We conclude that the defect is specific to the angiogenic stage of arterial development, without affecting arteriogenesis, similar to the vascular defect in zebrafish treated with syx MOs. The similarity of secondary vessel defects in the zebrafish and the mouse suggests that the function of syx in angiogenesis is highly conserved throughout evolution.
Angiomotin and syx Function Additively During ISV Sprouting
The recently published angiomotin (amot) KD phenotype in zebrafish14 is strikingly similar to that of syx. To determine whether amot and syx regulate ISV sprouting additively, we coinjected half-doses of MOs for each gene (4 ng each) or each MO alone (4 ng or 8 ng) into Tg(flk: GRCFP) fish and compared the images of the vascular system in each sample. The ISVs of uninjected embryos followed the somite borders (Figure 5A), but the ISVs of embryos injected with 4 ng of syx MO (Figure 5B) or amot MO (Figure 5C) injected embryos were stunted. When amot and syx MOs were coinjected at 4 ng each, truncation of ISV sprouts (Figure 5D, white asterisk) was more severe than those in embryos injected with a single MO. In some cases, the ISVs in embryos coinjected by amot and syx MOs were completely absent (Figure 5E, white asterisk). Injection of syx or amot MO (4 ng) alone resulted in approximately half (23% and 28%, respectively) as many ISV-defective embryos (Figure 5I) as injection of double the dose (8 ng) of each MO (60% and 45%, respectively; Figure 5I). In coinjected (4 ng syx MO +4 ng amot MO) samples, the number of embryos with defective ISVs doubled compared to embryos injected by 4 ng of a single type of MO (50%, Figure 5I). High-power images of ISVs in uninjected embryos at 28 hpf revealed elongated filopodia protruding out of their tip cells (Figure 5F), whereas the tip cells of ISVs in embryos injected by either syx MO (Figure 5G) or by both syx and amot MOs (Figure 5H) were round and devoid of filopodia. Given the known ISV defect caused by robo4 KD in zebrafish,17 we sought to determine the effect of coinjecting syx and robo4 MOs. We did not observe an additive effect between robo4 and syx KDs (data not shown), suggesting that Syx and Robo4 do not share the same signaling pathway.
Angiomotin is thought to be involved in regulating EC migration12 similar to Syx8 and is known to bind multi-PDZ domain protein 1 (MUPP1), a large scaffold protein containing 13 PDZ domains.24 We were able to coimmunoprecipitate MUPP1 and Syx1 but not Syx2 in ECs expressing each splice isoform fused to yellow fluorescent protein (supplemental Figure IVA). This result shows that Syx1 binds MUPP1 via its PDZ binding motif. The zebrafish ortholog of MUPP1 is likely, therefore, to functionally couple Amot and Syx1, accounting for the genetic interaction we observed between the 2 proteins.
Syx KD Impairs EC Migration and Angiogenesis in Response to Vascular Endothelial Growth Factor-A
Vascular endothelial growth factor (VEGF)-A is a major agonist of angiogenesis.25 Because the vascular defects in the syx−/− mouse and in syx KD zebrafish were angiogenic in nature, we tested the effect of silencing syx expression on EC vessel formation, namely VEGF-A165-induced migration and in vitro tube formation. Transfection by syx small interfering (si)RNA significantly depleted Syx and reduced EC migration under basal conditions and in response to VEGF-A165 (Figure 6A). The tubular network formed by ECs transfected with syx siRNA, either with or without VEGF-A165, was sparser than the network formed by cells transfected with control siRNA, indicating that Syx depletion impaired angiogenesis (Figure 6B). However, ECs transfected by syx siRNA did not show a difference in invasive behavior compared with control siRNA-transfected ECs to either VEGF or serum stimulus indicating no function for Syx in invasion (supplemental Figure IVB). These results indicate that Syx has an essential role in specific steps of angiogenesis such as EC migration, in agreement with our previous results.8
This study identifies the function of a novel Rho-GEF, Syx, in vascular development in vivo. We report three important findings. First, syx is specifically expressed in the vasculature of teleosts and mammals. Second, gain- and loss-of-function analyses show that syx plays a specific role in orchestrating directional migration of endothelial sprouts during vertebrate angiogenesis. Third, a novel signaling axis between syx and amot necessary for proper patterning of angiogenic vessels has been uncovered.
Vessel patterning in vertebrates is a complex process that involves the formation of primary axial vessels by vasculogenesis and the sprouting of secondary vessels by angiogenesis. This study focuses on the angiogenic mechanism of ISV sprouting. Syx has been identified as a novel RhoA-specific GEF,7 which contributes to EC migration in vitro.8 This function of Syx is consistent with previously published reports suggesting an important role for Rho GTPases in axon and endothelial tip cell guidance.26,27
We have extended the EC expression profile of syx to mouse and zebrafish vessels. In both species, syx is expressed in the vasculature and in zebrafish the expression is almost exclusive to the DA and emerging sprouts. Because we did not detect Syx expression in the zebrafish posterior cardinal vein and in the mouse vena cava, it is likely that the Syx is specifically involved in arterial angiogenesis.
In the mouse, global disruption of the syx gene was accompanied by a significant reduction in the density of secondary arteries and of capillaries in the heart. We observed a similar reduction in the density of secondary arteries of the kidneys. Although we have not characterized additional vascular beds, we suspect that the angiogenic defect in the syx−/− mouse is likely to extend to all tissues and organs. Given the pivotal function of Syx in EC migration,8 it is surprising that vascular defects in the syx−/− mouse and the syx KD zebrafish are not severe. In vascular development, isoform-specific function of signaling molecules is frequently confined to specific developmental stages. For example, in zebrafish, the KDRa isoform of the VEGF-A receptor is specifically involved in arterial but not in vein development, whereas KDRb is dispensable for vascular development.28 Therefore, functional redundancy across isoforms may partly explain this issue. Syx itself has multiple splice variants and moreover in humans and the mouse, Syx is 1 isoform of a 7-member family of RhoA GEFs. Therefore, multiple backups exist in the system; in which case, syx must be tightly regulated.
Two lines of evidence suggest that maintaining optimal levels of syx is critical for directional migration of ISV sprouts. First, both KD and overexpression of syx resulted in similar phenotypes. In both experiments, ISVs had sprouting defects, suggesting that ISV growth requires a strict balance of guidance cues. When there is either a deficiency or overabundance of guidance molecules, the ISVs stall until a proper migratory path is established. These results mimic those observed for robo4,17 although we did not find evidence for a genetic interaction with syx. Second, in our complementation experiments, we injected mouse syx mRNA together with the syx MO1 to demonstrate MO specificity and evolutionary conservation of syx function. Because 100 pg of syx mRNA did not rescue MO-induced vascular defects, we increased the dose of mRNA to 150 and 200 pg and saw a proportional increase in rescued embryos. However, when the dose of mRNA was increased to 300 pg, the percentage of rescued embryos declined. These results also suggest that syx levels have to be maintained within a particular range to support proper vascular patterning.
A recent study demonstrated that amot KD caused ISV defects phenotypically similar to that of syx.14 In the present study, we found that syx and amot function additively to regulate ISV sprouting. The genetic interaction between amot and syx is accounted for, at least in part, by the coupling of the corresponding proteins via the adaptor protein MUPP1 (elsewhere24 and supplemental Figure IV). Both Amot and Syx are involved in mediating EC migration in response to VEGF (14 and Figure 6A), further substantiating the functional cooperation between these proteins. The inhibitory effect of Syx depletion on in vitro EC migration and tube formation in response to VEGF-A provides a potential molecular mechanism to explain the angiogenic defects in the syx−/− mouse and the syx KD zebrafish. The genetic interaction between Amot and Syx implies that the role of Syx in VEGF-A–induced angiogenesis is part of a more complex scenario involving additional signaling pathways. The binding of Syx to MUPP1, a component of tight junctions,24,29 and its presence in these junctions (data not shown) suggest a putative functional link between the association of Syx with tight junctions and its involvement in cell migration.
Sources of Funding
R.R. is a recipient of a National Cancer Institute Scholar Award. This study was supported by NIH grant HL67960 (to A.H.), an American Heart Association Scientist Development Grant (M.L.), and seed funds from the Children’s Research Institute at the Medical College of Wisconsin (to R.R.).
↵*Both authors contributed equally to this work.
Original received May 6, 2008; resubmission received June 12, 2008; revised resubmission received July 30, 2008; accepted August 18, 2008.
Risau W. Differentiation of endothelium. FASEB J. 1995; 9: 926–933.
Lamalice L, Le Boeuf F, Huot J. Endothelial cell migration during angiogenesis. Circ Res. 2007; 100: 782–794.
Liu M, Horowitz A. A PDZ-binding motif as a critical determinant of Rho guanine exchange factor function and cell phenotype. Mol Biol Cell. 2006; 17: 1880–1887.
Hung AY, Sheng M. PDZ domains: structural modules for protein complex assembly. J Biol Chem. 2002; 277: 5699–5702.
Maystadt I, Rezsohazy R, Barkats M, Duque S, Vannuffel P, Remacle S, Lambert B, Najimi M, Sokal E, Munnich A, Viollet L, Verellen-Dumoulin C. The nuclear factor kappaB-activator gene PLEKHG5 is mutated in a form of autosomal recessive lower motor neuron disease with childhood onset. Am J Hum Genet. 2007; 81: 67–76.
Bratt A, Birot O, Sinha I, Veitonmaki N, Aase K, Ernkvist M, Holmgren L. Angiomotin regulates endothelial cell-cell junctions and cell motility. J Biol Chem. 2005; 280: 34859–34869.
Aase K, Ernkvist M, Ebarasi L, Jakobsson L, Majumdar A, Yi C, Birot O, Ming Y, Kvanta A, Edholm D, Aspenstrom P, Kissil J, Claesson-Welsh L, Shimono A, Holmgren L. Angiomotin regulates endothelial cell migration during embryonic angiogenesis. Genes Dev. 2007; 21: 2055–2068.
Westerfield M. The Zebrafish Book. 4 ed. Eugene, Ore: University of Oregon Press; 2000.
Bedell VM, Yeo SY, Park KW, Chung J, Seth P, Shivalingappa V, Zhao J, Obara T, Sukhatme VP, Drummond IA, Li DY, Ramchandran R. roundabout4 is essential for angiogenesis in vivo. Proc Natl Acad Sci U S A. 2005; 102: 6373–6378.
Stainier DY, Weinstein BM, Detrich HW III, Zon LI, Fishman MC. Cloche, an early acting zebrafish gene, is required by both the endothelial and hematopoietic lineages. Development. 1995; 121: 3141–3150.
Cross LM, Cook MA, Lin S, Chen JN, Rubinstein AL. Rapid analysis of angiogenesis drugs in a live fluorescent zebrafish assay. Arterioscler Thromb Vasc Biol. 2003; 23: 911–912.
Zhong TP, Rosenberg M, Mohideen MA, Weinstein B, Fishman MC. gridlock, an HLH gene required for assembly of the aorta in zebrafish. Science. 2000; 287: 1820–1824.
Lawson ND, Scheer N, Pham VN, Kim CH, Chitnis AB, Campos-Ortega JA, Weinstein BM. Notch signaling is required for arterial-venous differentiation during embryonic vascular development. Development. 2001; 128: 3675–3683.
Kaur S, Castellone MD, Bedell VM, Konar M, Gutkind JS, Ramchandran R. Robo4 signaling in endothelial cells implies attraction guidance mechanisms. J Biol Chem. 2006; 281: 11347–11356.
Covassin LD, Villefranc JA, Kacergis MC, Weinstein BM, Lawson ND. Distinct genetic interactions between multiple Vegf receptors are required for development of different blood vessel types in zebrafish. Proc Natl Acad Sci U S A. 2006; 103: 6554–6559.