roundabout4 is essential for angiogenesis in vivo

Victoria M. Bedell*, Sang-Yeob Yeo†, Kye Won Park‡, Jeffrey Chung§, Pankaj Seth§, Venkatesha Shivalingappa§, Jinhua Zhao¶, Tomoko Obara¶, Vikas P. Sukhatme§, Iain A. Drummond¶, Dean Y. Li‡, and Ramani Ramchandran*ʈ

*Laboratory of Pathology, National Cancer Institute, National Institutes of Health, 9610 Medical Center Drive, Suite 320, Rockville, MD 20850; †Laboratory of Molecular Genetics, Unit on Vertebrate Neural Development, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD 20892; ‡Cardiology Division, University of Utah, Salt Lake City, UT 84132; §Department of Medicine and Center for Study of the Tumor Microenvironment, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA 02215; and ¶Renal Unit, Massachusetts General Hospital, Harvard Medical School, Boston, MA 02114

Edited by Kathryn V. Anderson, Sloan–Kettering Institute, New York, NY, and approved March 21, 2005 (received for review November 8, 2004) Stereotypical patterns of vascular and neuronal networks suggest In summary, these findings demonstrate that robo signaling has a that specific genetic programs tightly control path determination critical role in embryonic angiogenesis. and, consequently, angiogenesis and axon-guidance mechanisms. Our study focuses on one member of the roundabout family of Experimental Procedures receptors, which traditionally mediate repulsion from the midline. Stocks and Reagents. Zebrafish were grown and main- Here, we characterize a fourth member of this family, roundabout4 tained at 28.5°C (13) under National Cancer Institute guidelines (), which is the predominant roundabout (robo) that is (animal protocol no. LP-020). Mating was routinely carried out expressed in embryonic zebrafish vasculature. Gene knockdown at 28.5°C, and the embryos were staged according to established and overexpression approaches show that robo4 is essential for protocols (14). Zygogen (Atlanta) provided the Tg(vegfr2: G- coordinated symmetric and directed sprouting of intersomitic ves- RCFP)y10 (green reef coral fluorescent protein) fish (15). Flk, sels and provide mechanistic insights into this process. Also, human cdh5, znp-1, and acetylated tubulin markers were obtained from robo4 gene functionally compensates for loss of robo4 gene Nathan Bahary (University of Pittsburgh, Pittsburgh), J. J. function, suggesting evolutionary conservation. This article re- Essner (Discovery Genomics, Minneapolis), the Developmental ports an endothelial-specific function for a robo gene in verte- Studies Hybridoma Bank (Iowa City, IA), and Sigma, respec- BIOLOGY brates in vivo. tively. and cDNAs were obtained from C. Beattie (Ohio State University, Columbus) and C. B. Chien (University DEVELOPMENTAL ͉ endothelial cell ͉ zebrafish of Utah, Salt Lake City), respectively. Based on the intron–exon boundary sequences, robo4 MO phosphorodiamidate oligonu- ascular and neural networks established by angiogenesis and cleotides were designed by Gene Tools (Carvalis, OR), and the Vneurogenesis are essential for the regulation of physiological sequences are as follows: MO1, TTTTTTAGCGTACCTAT- processes. Whether the processes of blood-vessel and peripheral- GAGCAGTT; MO2, TATTATGGATTACCTGAGCTTT- nerve formation are intimately associated at a mechanistic level is TGC; mismatch MO1 (msMO1), TTTTTTAcCcTACgTAT- a subject of active investigation (1). Recently, description of cell- GAcCAcTT; and msMO2, TATTATcGATTAgCTcAGgTT- surface molecules that are shared by neuronal and endothelial cells TTcC. Splice-site sequences are shown in italics, and mismatch has suggested that this association may be more prevalent than nucleotides are indicated by lowercase letters. previously thought. Neuropilin (2), ephrin (3), and plexin (4) are a Molecular Biology and MO͞RNA Injections. To determine efficacy, few examples of molecules that are implicated in both processes. ␮ Recently, roundabout (robo), a class of neural guidance receptors reverse transcription was performed on 0.7–1.0 g of total RNA isolated from WT-, MO-, and msMO-injected embryos at 24 h after that bind ligands have joined this group (5). slit–robo signaling ␮ mediates axonal repulsion (6) and inhibition of leukocyte migration fertilization (hpf) by using oligo(dT) primers (1 g). The following (7). In vertebrate systems, three robo receptor family members were cycling parameters were used for PCR: 34 cycles of 94°C for 2 min, identified, all with prominent neural expression (8, 9). More 94°C for 30 sec, 58°C for 30 sec, and 72°C for 1 min; and 72°C for recently, a fourth member of the robo gene family, roundabout4 5 min. The following primers were used for MO-efficacy experi- (robo4) was identified (10). Compared with the canonical robo ments: F, AGCCCAGAAGAATGACTCTGGAG; and R, TGTT- structure, robo4 is smaller in that it possesses only two of the five TGATGATGAGATCTTCAC. Microinjections of one-cell stage Ig and two of the three fibronectin domains present in the extra- zebrafish embryo with RNA or MO were carried out as described (13). MOs were reconstituted in nuclease-free water to a stock cellular component of robo1, robo2, and (11). robo4 has been ͞ described as endothelial-specific, and it binds to slit and inhibits the concentration of 2 mM (16 ng nl). Appropriate dilutions were made in 5ϫ injection dye (100 mM Hepes͞1 M KCl͞1% phenol migration of heterologous cells that express robo4 and primary red), and Ϸ2–3 nl of MOs (8–14 ng) were injected at the one-cell endothelial cells (12). Although studies of robo4 in endothelial cells stage. For rescue experiments, 150 pg of hrobo4 capped RNA suggest that robo signaling is important for regulating endothelial transcribed by SP6 RNA polymerase from a NotI linearized vector , there are no reports that document the role of robo containing the full-length hrobo4 cDNA was injected with MO. For signaling in vascular development in vivo. Here, we cloned a RNA overexpression experiments, we routinely injected 150 pg per zebrafish (Danio rerio) ortholog of human robo4 (hrobo4) and embryo. Full-length zebrafish robo4 and yellow fluorescent protein studied its expression and role in vascular development by gene (YFP) constructs were linearized with NotI and ScaI, respectively, knockdown and overexpression approaches. In contrast to the and transcribed with T7 polymerase to make sense RNA. endothelial-specific expression of murine robo4, zebrafish robo4 is expressed in both endothelial and neural tissues with vascular expression seen in angioblasts, the dorsal aorta (DA), the posterior This paper was submitted directly (Track II) to the PNAS office. cardinal vein, and intersomitic vessels (ISV). Morpholino (MO) Abbreviations: robo4, roundabout4; robo, roundabout; hrobo4, human robo4; DA, dorsal knockdown of robo4 and overexpression of zebrafish robo4 results aorta; MO, morpholino; hpf, hours after fertilization; ISV, intersomitic vessels; msMO, in a vascular phenotype with asynchronous ISV sprouting, resulting mismatch MO; som, somite(s); TM, transmembrane; YFP, yellow fluorescent protein. in either complete lack of, or misdirected, ISV. The hrobo4 gene ʈTo whom correspondence should be addressed. E-mail: [email protected]. functionally rescued the vascular defects induced by loss of robo4. © 2005 by The National Academy of Sciences of the USA

www.pnas.org͞cgi͞doi͞10.1073͞pnas.0408318102 PNAS ͉ May 3, 2005 ͉ vol. 102 ͉ no. 18 ͉ 6373–6378 Downloaded by guest on September 27, 2021 Fig. 1. Expression patterns of robo4 in development. A montage of robo4 whole-mount in situ staining is shown, with 16 (A), 19 (B), 20 (C), 23 (D), 24 (E), and 20 (F) som. (G) High power of F.(H) High power of I.(I) A 22-som telencephalon (t), hindbrain (hb), and neural tube (nt). J is 28–29 hpf. Arrowhead in G indicates expression in angioblasts; asterisk in H shows ISV expression. robo4 (K), G-RCFP (L), DAPI (M), and merge (N) show two-color staining of frozen 20-␮m anterior trunk sections of vegfr2-G-RCFP 25-hpf embryos. The arrowhead and asterisk in K–N indicate ISV and axial vessels, respectively. Images were taken with a ϫ20 (A–E), ϫ10 (F, I, and J), ϫ40 (G and H), and ϫ20 (K–N) lens with ϫ2.5 zoom.

In Situ Hybridization and Immunostaining. WT embryos were grown (P Ͻ 0.001). In the rescue experiment, when the MO1͞2-treated in 0.003% phenylthiourea until the desired stage, fixed overnight in embryos with defective vasculature were compared with msMO1͞2 4% paraformaldehyde (PFA)͞PBS at 4°C, dechorionated, and and MO1͞2ϩhrobo4 embryos, a significant difference (P Ͻ 0.001) stored in 100% methanol until use. Whole-mount in situ hybrid- was noted. However, there was no significant difference between ization was carried out as described in ref. 16. Antisense probes msMO1͞2 and MO1͞2 ϩhrobo4 (P Ͼ 0.05). In the overexpression were generated with T3 RNA polymerase by using NotI linearized experiment, there was also a significant difference between the vector containing 1.8 kb of robo4 fragment. Hybridized embryos number of defective WT versus robo4 (P Ͻ 0.001) embryos and were photographed as described (16). For all double-staining robo4 versus hrobo4 (P Ͻ 0.001) embryos. protocols, immunostaining was performed first with acetylated tubulin primary mAb (1:25) or znp-1 (neuronal specific marker), Results followed by in situ hybridization as described (16). Two-color We cloned the zebrafish ortholog of hrobo4 gene by extracting immunostaining for G-RCFP and in situ hybridization for robo4 nucleotide sequences from the Sanger Institute zebrafish shotgun transcript was performed on frozen sections from 25-hpf Tg(vegfr2: genome sequences (D. rerio Sequencing Group; available at ftp:͞͞ G-RCFP)y10 embryos. Embryos were fixed overnight in 4% PFA ftp.sanger.ac.uk͞pub͞zebrafish) that matched hrobo4 amino acid with 4% sucrose in PBS, rinsed in 1ϫ PBS containing 0.1% Tween sequence. The zebrafish robo4 ORF encodes a protein of 1,134 aa 20, and embedded initially in 2.5% low-melt agarose with 5% with a single transmembrane (TM) domain and shares 30% amino sucrose͞PBS, followed by overnight embedding in 30% sucrose͞ acid homology to the human gene (Fig. 5A, which is published as PBS at 4°C. The agar mold was transferred to a cryomold and supporting information on the PNAS web site). Structurally, the frozen under liquid N2, and cryosectioning was carried out to zebrafish robo4 protein contains three Ig domains in the extracel- generate 20-␮m sections. Slides were immunostained first by using lular region, compared with two Ig domains for human and mouse primary rabbit anti-GFP Ab, followed by a secondary biotin- robo4 proteins. In comparison with other members of the robo conjugated anti-rabbit Ab, which was detected by using a strepta- family in zebrafish, robo4 is smaller and contains three of five Ig, vidin–Alexa Fluor 488 probe (green). After immunostaining, in situ two of three fibronectin, and two of four conserved cytoplasmic hybridization for robo4 transcript was performed by using antisense domains (Fig. 5B). digoxigen RNA probes as described previously. For in situ robo4 signal, the alkalinephosphatase-conjugated antidigoxigenin was Expression of robo4 in Zebrafish. Whole-mount in situ hybridization detected by using 2-hydroxy-3-naphthoic acid-2Ј-phenylanilide using robo4 antisense RNA probes revealed the spatial and tem- phosphate (HNPP) fluorescent solution substrate (red) (Roche) poral expression of robo4 during zebrafish embryonic development. and DAPI for nuclear staining. Notochord robo4 expression begins as early as 8 hpf (data not shown) and continues from 16 somites (som) (Fig. 1A)to23som Time-Lapse Analysis. MO-injected embryos were allowed to develop (Fig. 1D). Earliest vascular expression of robo4 appears in angio- overnight to 19 hpf, and age-matched embryos were embedded in blasts ventral to notochord at 19 or 20 som (Fig. 1 B and C) and a glass-bottom culture dish (MatTek, Ashland, MA) with 1% continues in ISV in 23 or 24 som (Fig. 1 D and E) before receding low-melt agarose and tricaine (0.16 mg͞ml) added to prevent at 29 hpf (Fig. 1 J). robo4 expression overlaps between notochord movement during imaging. Images of the vasculature were captured and vascular tissue during a 4-h period, starting at 19 som (Fig. 1B) with a ϫ10 objective mounted on an inverted fluorescent micro- to 23 som (Fig. 1D). The progressive rostral–caudal loss of robo4 scope using the Delta Vision microscopy system (Applied Preci- expression in the notochord (Fig. 1 A–D) correlates with robo4 sion, Issquah, WA). We took 35 z-scan images (2 ␮m per slice; total, expression in adjacent sprouting ISV (compare Fig. 1 G and H). 70 ␮m) every 5 min for 12 h and converted into one projection by During this period, robo4 neural-tube expression clearly persists SOFTWORX software and scaled by using METAMORPH software. (Fig. 1 B–J) and is expressed in other regions of the embryo, such as the developing CNS regions (namely, telencephalon and hind- Statistical Analysis. A two-way ␹2 test was performed to determine brain regions) (Fig. 1I). In situ hybridization for robo4 transcript the P value of injected samples. The knockdown experiment (Fig. 1K, arrowhead) and immunostaining for vascular marker flk showed a significant difference between msMO1͞2 and MO1͞2 (Fig. 1L, arrowhead) in serial frozen sections of Tg(vegfr2-GRCFP)

6374 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0408318102 Bedell et al. Downloaded by guest on September 27, 2021 25-hpf embryos reveal that robo4 expression overlaps with flk ISV (Fig. 1N, arrowhead) and colocalizes with flkϩ arterial and venous endothelial cells (Fig. 1N, asterisk).

Role of robo4 Gene in Zebrafish Vascular Development. Based on in situ vascular expression pattern, we hypothesized that robo4 has a role in ISV sprout formation in the trunk region of the developing zebrafish vasculature. To test this hypothesis, we investigated the function of robo4 gene during zebrafish embryonic development by using MO antisense oligonucleotides directed against robo4 exon splice donor sites to disrupt mRNA splicing (17, 18). We designed two splice MOs, which target different regions of the robo4 gene. The genomic structure of robo4 contains 18 exons of which splice MO1 targets exon 9, whereas splice MO2 targets exon 10 putative TM domains in robo4 (Fig. 2A). Efficacy of splice MOs was con- firmed by RT-PCR across the flanking exons of the targeted splice donor sites, which revealed three bands in MO1 (Fig. 2A, lane 2, upper gel) and one band in MO2-injected embryos (Fig. 2A, lane 2, lower gel) when compared with control uninjected (Fig. 2A, lane 1) or msMO1-injected (Fig. 2A, lane 3, upper gel) or msMO2-injected (Fig. 2A, lane 3, lower gel) embryos. Sequencing the 1.5-kb band (gray arrowhead, Fig. 2B) in MO1 lane revealed that part of exon 9 was deleted and the intron was retained, which resulted in an out-of-frame protein. A similar out-of-frame protein was predicted from the lower 1.2-kb band (black arrow, Fig. 2B). The 1.1-kb band (gray and dotted arrows) revealed deletions of

exon 10 in MO1 and MO2 lanes predicting proteins lacking TM BIOLOGY

domains (Fig. 2B). Bands amplified in the nontargeted regions DEVELOPMENTAL showed identical sequence similar to common bands in WT and msMO lanes (Ϸ1.3 kb, black arrowhead). Initially, we examined expression of flk, which defines ISV and axial vessels in the trunk region. MO1-injected embryos show absence of flkϩ ISV (Fig. 2D), whereas MO2-injected (Fig. 2E) embryos show thinner flkϩ vessels traversing a linear path when compared with control uninjected 22- to 24-som embryos (Fig. 2C). Quantification of these defects in Fig. 2H shows that most MO1- injected embryos (65%) display a lack of ISV, whereas a few MO1-injected embryos (33%) display either partial absence or abnormal shape of ISV, and 40% of MO2-injected embryos display a partial absence of ISV. We also stained the MO-injected embryos for either znp-1 (motor neuron) (compare Fig. 2 C with D and E) or acetylated tubulin (most neurons) (Fig. 2 F and G), and preliminary observation at low resolution shows only subtle changes in axonogenesis when compared with controls that will not be discussed further. These results suggest that ISV were apparently missing from their usual location in the trunk. To determine whether this apparent absence of vessels was due to a failure to sprout or, alternatively, Fig. 2. Splice MO targeting and endothelial marker analysis of MO-injected from misguided growth of vessels, we examined more closely at embryos. (A) The exon–intron genomic structure from exon 7–12 is shown. vessel outgrowth by using live time-lapse analysis of fluorescent Splice MO1 and MO2 target the predicted TM domains. Total RNA was y10 extracted from uninjected (lane 1), MO (lane 2; MO1, upper gel; MO2, lower vasculature in Tg(vegfr2: G-RCFP) transgenic fish line (15) gel), and msMO (lane 3; msMO1, upper gel; msMO2, lower gel) embryos at 24 carrying a 6.5-kb flk-promoter fragment driving G-RCFP expres- hpf, and RT-PCR was performed. Black, gray, and dotted arrows indicate the sion in angioblasts and mature vasculature. MO2 was used because alternative transcripts that were generated in MO-targeted embryos (lane 2), it induced only one alternative spliced product (Fig. 2A). Movies the black arrowhead indicates a common untargeted bands in lanes 1 and 3, from MO2-injected embryos show that flk vessels are either stunted and the gray arrowhead points to 1.5-kb band in lane 2. E, exon. (B) Structural or misrouted when compared with msMO2 or WT embryos (see details of predicted protein products generated from aberrant splicing are Movie 1, which is published as supporting information on the PNAS shown. Hexagons, pentagons, white bar, and black bars represent IgG, FN, TM, web site). In WT and msMO2-injected embryos, most ISV extend and CC domains, respectively. Double staining with flk digoxigenin RNA (blue) slightly rostrally and then caudally after crossing the transverse and Ab to znp-1 (C–E) (brown) or acetylated tubulin (F and G) in the trunk region of 22–24 som is shown. (C–G) Magnified images of the trunk region, myoseptum, potentially following the chevron-like contours of the showing WT (C and F), MO1-injected (8 ng) (D and G), and MO2-injected (8 ng) som, before reaching the dorsal–lateral surface, where tubes from (E) embryos. Asterisk indicates location of ISV. (H) Quantification of embryos adjacent ISV fuse to form the DLAV (dorsal longitudinal anasto- with ISV defects. WT (I), MO1 (J), and msMO1 (K) show in situ staining for cdh5 motic vessels). In MO2-injected embryos, the chevron-shaped marker at 24 hpf. Anterior is to the left. trajectory is replaced by a much straighter extension of the vessels. Also, some ISV never initiate a sprout, whereas other ISV regress after an initial attempt. This linear route is reminiscent of the generally emerging from DA are asynchronous in MO2-injected MO2-injected flk in situ stained embryos (Fig. 2E) in which embryos. We stained the MO-injected embryos with a vascular flk-stained ISV show loss of directionality. Also, primary sprouts cdh5 (ve-cadherin) marker, which is also affected in MO1-injected

Bedell et al. PNAS ͉ May 3, 2005 ͉ vol. 102 ͉ no. 18 ͉ 6375 Downloaded by guest on September 27, 2021 in inhibition of ISV sprouting, we investigated whether robo4 overexpression would induce more ISV sprouts. We injected capped mRNA for full-length zebrafish robo4 and YFP in Tg(vegfr2: G-RCFP)y10 embryos. Z-stack images of the trunk region of 22- to 24-som transgenic embryos show that primary sprouts from the DA develop normally and reach the dorsal–lateral surface in uninjected (Fig. 3A) and YFP-injected (Fig. 3B) embryos. Surprisingly, robo4 RNA-injected embryos (Fig. 3C), show missing ISV sprouts (as- terisk). Fig. 3D shows the quantitation of this phenotype, where the robo4 RNA-injected embryos display a 4-fold increase in ISV defects when compared with WT.

ϩ Functional Rescue of robo4 by Gene Complementation. To confirm Fig. 3. robo4 overexpression results in misdirected or truncated flk ISV from the specificity effect of MOs and investigate whether robo4 function DA. WT (A), YFP (B), and robo4 (C) RNA-injected embryos at 22–24 som are shown. (D) Quantitation of overexpression is shown for WT (n ϭ 36), YFP (n ϭ was conserved during evolution, we sought to determine whether 29), and robo4 (n ϭ 28). Anterior is to the left. hrobo4 gene would rescue ISV defects induced by robo4 knock- down. We injected MOs alone or in combination with hrobo4 RNA y10 (Fig. 2J) embryos when compared with WT (Fig. 2I) and msMO1- in Tg(vegfr2: G-RCFP) embryos and checked for ISV defects at injected (Fig. 2K) embryos. 22–24 som and 48 hpf. Static images of the MO-injected embryos (Fig. 4 A, D, G, and J) confirm the findings shown in Movie 1 that ϩ Overexpression of robo4 Gene Affects ISV Formation. Because gene- the flk ISV are absent, misdirected, and asymmetrical. The knockdown experiments with MOs suggest that loss of robo4 results MO1-injected (Fig. 4A) and MO2-injected (Fig. 4G) embryos show

Fig. 4. robo4 gene knockdown affects flkϩ ISV and rescue of defects by hrobo4. Splice MOs alone (12 ng) or in combination with hrobo4 RNA (150 pg) were microinjected into Tg(vegfr2: G-RCFP)y10 zebrafish embryos at the one-cell stage, and images were taken at 22–24 som and 48 hpf. From left to right, images are MO, msMO, and MOϩhrobo4-injected embryos. A–C and G–I show embryos at 22–24 som, whereas D–F and J–L show embryos at 48 hpf. (A and D) MO1 (n ϭ 31). (G and J) MO2 (n ϭ 28). (B and E) msMO1 (n ϭ 31). (H and K) msMO2 (n ϭ 54). (C and F) MO1ϩhrobo4 (n ϭ 39). (I and L) MO2ϩhrobo4 (n ϭ 46). (M) Quantitation for rescue experiments with hrobo4 (A–L). (N) Quantitation of embryos injected for WT (n ϭ 124), MO2 (n ϭ 119), MO2ϩrobo1 (n ϭ 82), MO2ϩrobo2 (n ϭ 115), and MO2ϩrobo4 (n ϭ 91). (O) Quantitation comparison of overexpression experiment for WT (n ϭ 118), hrobo4 (n ϭ 71), and robo4 (n ϭ 91) RNA-injected (150 pg) embryos. A defective embryo was counted as three or more missing ISV. Anterior is to the left.

6376 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0408318102 Bedell et al. Downloaded by guest on September 27, 2021 missing ISV at 22–24 som, whereas the msMO (Fig. 4 B and H) analysis of robo4 MO transgenic embryos suggest that ISV show ISV growing synchronously at 22–24 som, similar to embryos outgrowth is misdirected in these embryos and, when vessels injected with both MO and hrobo4 (Fig. 4C, MO1ϩhrobo4;Fig.4I, sprout in the wrong direction, vessel formation is often aborted. MO2ϩhrobo4). At 48 hpf, MO-injected embryos show ISV that The presence of premature misdirected vessel sprouts in robo4 have terminated midway (Fig. 4D, MO1; Fig. 4J, MO2), and in some MO embryos (Fig. 4D) also indicates that robo4 acts to restrict cases, the sprouts from DA are asymmetric and misdirected when vessel growth to the proper path. In addition to directing the path compared with msMO (Fig. 4E, msMO1; Fig. 4K, msMO2) and of ISV sprouts, robo4 has a role in the timing of initiation of ISV rescue (Fig. 4F, MO1ϩhrobo4;Fig.4L, MO2ϩhrobo4) embryos. sprouts along the anterior–posterior axis of the DA. Vessel Quantification of the hrobo4 rescue experiments shows that Ϸ10– sprouting in WT embryos occurs in a coordinated anterior-to- 20% of embryos coinjected with MO and hrobo4 are defective when posterior progression, whereas in robo4 MO embryos, sprouting compared with 80% with MO alone (Fig. 4M). Comparing the is often observed in posterior intersomitic regions of the aorta numbers of rescue (Fig. 4M) with in situ hybridization analysis (Fig. before sprouts have formed in more anterior positions (Movie 2H), we note that increasing the dose of MO from 8–12 ng in 1). Together, these results suggest that robo4 acts not only to time-lapse analysis resulted in an increase in the percentage of restrict aortic sprouting in a spatial manner but also that it acts embryos displaying vascular defects. Further, we have also injected to temporally restrict sprouting and coordinate ISV formation. other robo family members (namely, robo1 and robo2;Fig.4N), Vascular, like neural, guidance is a complex repertoire of both which do not rescue the vessel defects, suggesting specificity of the attraction and repulsion mechanism mediated by cell surface vascular function for robo4. However, when we injected zebrafish receptors on endothelial cells and ligands in milieu. Members of robo4 along with MO, we noticed no rescue (Fig. 4N). This lack of guidance molecule family (namely, ephrin, semaphorin, , and rescue suggests that removing endogenous robo4 and providing slit) work cooperatively with cognate receptors Eph (22), plexin zebrafish robo4 results in an exacerbation of vascular guidance (23), unc5b (24), and robo (25), respectively, to guide axonal growth defects, possibly by removal of both ligand and receptor. To confirm cone pathfinding and vascular development (26). The conventional inherent differences between zebrafish and hrobo4 gene function- thinking for robo is that they are negative regulators of cell ality, we injected full-length RNA for both genes and compared the migration. The role of slit in repulsive-guidance mechanisms have percentage of embryos that display vascular defects. As expected, been documented in growth cones through robo (19) and in the zebrafish robo4 RNA showed a 4-fold increase in ISV-defective endothelial cells through robo4 (12). Recently, attraction mecha-

embryos, as opposed to hrobo4 RNA, which showed a 2-fold nisms have also been reported for slit in relation to tumor vascu- BIOLOGY

increase at the same concentration when compared with uninjected lature through robo1 (27) and in tracheal branches by DEVELOPMENTAL embryos (Fig. 4O). robo2 (28). Thus, slit–robo signaling can mediate both attraction and repulsion mechanisms. Discussion In this study, blocking a presumptive negative regulator results in Our work demonstrates that robo4 is essential for angiogenesis and no sprouts, as opposed to the expected increase in sprouts, which provides evidence that demonstrates robo4 function in vascular suggests that this conventional paradigm needs to be revisited. development in vivo. First, robo4 is expressed in both the vascular Here, blocking robo4 leads to failure of sprouting from endothe- and neural systems of zebrafish, with staggered expression in lium. Two possibilities could explain our findings. The first possi- notochord and angiogenic vessels. Second, knockdown of endog- bility is a nonproductive sprouting mechanism, in which the endo- enous robo4 expression by using two distinct splicing MOs during thelial cells without directional cues retract and extend with no clear embryogenesis results in temporal and spatial disruption of em- decisions, resulting in the ultimate decision to retract. The second bryonic vascular development. Third, vascular defect induced by possibility is that, besides repulsion mechanisms, robo also mediate loss of robo4 expression is rescued by hrobo4 gene. Together, these attraction mechanisms through ligands other than slit in which case studies provide evidence that the robo family of guidance receptors removal of attraction signal results in collapse of vessels. In either have a critical role in vascular development in vertebrates. case, vascular guidance by robo is mediated by ligand–receptor The role of robo as a repulsive guidance receptor in the neural interactions and that robo4ϩ endothelial cells are actively guided to system has been well documented in Drosophila (19). Previous the target location by signals from surrounding milieu. studies of robo in mesenchymal tissues (20) and in leukocyte We propose that, during initiation of ISV sprouts from DA, migration (7) have hinted at nonneural functions. Recently, slit2– ligand-mediated robo4–robo4 dimerization in angioblasts provides robo2 signaling has been implicated in kidney induction (21) during the directional cue for sprouting, and loss of robo4 here results in mouse development. The identification of robo4 (10) and its in vitro a confused angioblast cell, with many directions to sprout and often functional analysis (12) has implied vascular function for robo genes sprouting in the wrong direction, which results in eventual regres- during vertebrate development. Our work here demonstrates in vivo sion. In vitro evidence suggests that robo4–robo4 interaction occurs that robo signaling is essential for vascular structures. through the cytoplasmic domains and this interaction is indepen- Zebrafish embryonic vasculature primarily consists of axial (DA dent of slit binding (D.Y.L. and R.R., unpublished data). Whether and posterior cardinal vein) and ISV. Vessel development com- this interaction has functional significance in vasculature similar to mences as early as 12 hpf from angioblasts that arise in lateral robo–robo interactions in neurons (29) has not been determined. mesoderm cells and migrate toward the midline to the future site For ligand-mediated signaling events through robo, slit ligands of axial vessels by 20 hpf. robo4 vascular expression begins in the are the obvious choice. Three important facts question slit as ligands angioblasts at 19 som and overlaps initially with notochord robo4 for robo4. First, extracellular region of robo4 is significantly differ- expression for a period of 4 h. A concomitant down-regulation of ent from other robo genes. Second, our unpublished biochemical robo4 expression in notochord and up-regulation in ISV angiogenic experiments suggest that robo4–robo4 dimerization occurs inde- sprouts occurs in a rostral–caudal direction and continues until the pendently of slit. Third, data from various groups provide conflict- sprouts have traversed the midline, which suggests a fundamental ing evidence for slit2 binding to robo4 (12, 30). Further, slit2 has regulatory function for robo4 in ISV sprouting and pathfinding. been proposed as an attractant for robo1 expressing endothelial To investigate robo4 function in coordinating ISV sprouting cells (27) and a repellant for robo4 expressing endothelial cells (12). and directionality from DA, we pursued a gene-knockdown In zebrafish, of the three slit genes, som expression for slit1a is seen approach. Gene-knockdown analysis shows that axial vessels from 10 som stage until 24 hpf in tail som (31), and for slit2 at 16 have formed in these embryos, suggesting a redundant role for hpf (16). slit3 is not expressed in som but is observed in motor robo4 in vasculogenesis. However, ISV arising from DA were neurons in the ventral spinal chord of 26 hpf embryos (16). So far missing from their usual location in the trunk region. Time-lapse none of the slit expression correlate temporally and spatially to the

Bedell et al. PNAS ͉ May 3, 2005 ͉ vol. 102 ͉ no. 18 ͉ 6377 Downloaded by guest on September 27, 2021 vicinity of robo4 expression in vessels. slit1a expression appears the cently, similar results in which overexpression and knockdown closest in som at 20 hpf (31) but is already in posterior som when experiments show identical phenotype have been reported for vilse, robo4 expression is beginning to appear in angioblasts (Fig. 2 A and a GTPase-activating protein that links the robo signaling to rac in B). It is not known whether multiple slit mediate redundant regulating midline axonal repulsion (32, 33), suggesting that for functions through robo4 or ligands other than slit mediate robo4 robo in general, too much or too little leads to defects in signaling signaling. Interestingly, the extracellular domain of hrobo4 ex- outputs. pressed as a Fc fusion protein was reported (30) to inhibit endo- In summary, our results show that robo4 has an essential role in thelial migration and angiogenesis in vitro. angiogenesis and guides endothelial cells to their targets during Because overexpression and underexpression experiments show angiogenesis analogous to robo1, robo2, and robo3 during axonal similar phenotype, we reasoned that a fundamental difference in guidance. This study provides a glimpse into functional role for a structure and function for human and zebrafish robo4 would robo molecule in vascular structures in a vertebrate in vivo. The explain why hrobo4 RNA rescues the endogenous knockdown mechanism underlying robo4 function in vasculature, its signifi- phenotype. Functional RNA overexpression comparisons between cance in pathological angiogenesis, and its functional conservation zebrafish and hrobo4 show that zebrafish robo4 display a higher in mammalian development remain to be determined. percentage of ISV-defective embryos (Fig. 4O). This increase suggests that extracellular Ig domain differences between human and zebrafish robo4 accounts for differences in binding affinities to We thank James McNally and Tatiana Karpova from the National Cancer Institute Imaging Core Facility; members of the David Roberts putative ligands that mediate vascular function. In experiments with laboratory at the National Cancer Institute for their valuable scientific zebrafish robo4 RNA and MO, more embryos displayed vascular input; Raymond Stock and Charles River Laboratories members for defects compared with MO alone, suggesting that removing en- maintaining our fish stocks; and Ajay Chitnis of the National Institute of dogenous robo4 and absorbing putative ligand by injected robo4 Child Health and Human Development for valuable inputs and critical RNA creates a directionless environment for the sprouts, resulting comments on the manuscript. V.M.B. is a recipient of the National in regression (Fig. 4N). Furthermore, other members of the robo Cancer Institute Cancer Research Training Award fellowship. R.R. is a family (namely, robo1 and robo2) do not rescue robo4 knockdown recipient of a National Cancer Institute Scholar grant and is supported phenotypes, suggesting vascular-specific function for robo4. Re- by the intramural program at the National Cancer Institute.

1. Mukouyama, Y. S., Shin, D., Britsch, S., Taniguchi, M. & Anderson, D. J. 20. Greenberg, J. M., Thompson, F. Y., Brooks, S. K., Shannon, J. M. & Akeson, (2002) Cell 109, 693–705. A. L. (2004) Dev. Dyn. 230, 350–360. 2. Neufeld, G., Cohen, T., Shraga, N., Lange, T., Kessler, O. & Herzog, Y. (2002) 21. Grieshammer, U., Le, M., Plump, A. S., Wang, F., Tessier-Lavigne, M. & Trends Cardiovasc. Med. 12, 13–19. Martin, G. R. (2004) Dev. Cell 6, 709–717. 3. Gale, N. W. & Yancopoulos, G. D. (1999) Genes Dev. 13, 1055–1066. 22. Wilkinson, D. G. (2000) Int. Rev. Cytol. 196, 177–244. 4. Fiore, R. & Puschel, A. W. (2003) Front. Biosci. 8, S484–S499. 23. Torres-Vazquez, J., Gitler, A. D., Fraser, S. D., Berk, J. D., Van, N. P., 5. Brose, K., Bland, K. S., Wang, K. H., Arnott, D., Henzel, W., Goodman, C. S., Fishman, M. C., Childs, S., Epstein, J. A. & Weinstein, B. M. (2004) Dev. Cell Tessier-Lavigne, M. & Kidd, T. (1999) Cell 96, 795–806. 7, 117–123. 6. Brose, K. & Tessier-Lavigne, M. (2000) Curr. Opin. Neurobiol. 10, 95–102. 24. Lu, X., le Noble, F., Yuan, L., Jiang, Q., de Lafarge, B., Sugiyama, D., Breant, 7. Wu, J. Y., Feng, L., Park, H. T., Havlioglu, N., Wen, L., Tang, H., Bacon, K. B., C., Claes, F., De Smet, F., Thomas, J. L., et al. (2004) Nature 432, 179–186. Jiang, Z., Zhang, X. & Rao, Y. (2001) Nature 410, 948–952. 25. Kidd, T., Brose, K., Mitchell, K. J., Fetter, R. D., Tessier-Lavigne, M., 221, 8. Lee, J. S., Ray, R. & Chien, C. B. (2001) Dev. Dyn. 216–230. Goodman, C. S. & Tear, G. (1998) Cell 92, 205–215. 9. Challa, A. K., Beattie, C. E. & Seeger, M. A. (2001) Mech. Dev. 101, 249–253. 26. Dickson, B. J. (2002) Science 298, 1959–1964. 10. Huminiecki, L., Gorn, M., Suchting, S., Poulsom, R. & Bicknell, R. (2002) 27. Wang, B., Xiao, Y., Ding, B. B., Zhang, N., Yuan, X., Gui, L., Qian, K. X., Genomics 79, 547–552. Duan, S., Chen, Z., Rao, Y. & Geng, J. G. (2003) Cancer Cell 4, 19–29. 11. Huminiecki, L. & Bicknell, R. (2000) Genome. Res. 10, 1796–1806. 28. Englund, C., Steneberg, P., Falileeva, L., Xylourgidis, N. & Samakovlis, C. 12. Park, K. W., Morrison, C. M., Sorensen, L. K., Jones, C. A., Rao, Y., Chien, (2002) Development (Cambridge, U.K.) 129, 4941–4951. C. B., Wu, J. Y., Urness, L. D. & Li, D. Y. (2003) Dev. Biol. 261, 251–267. 13. Westerfield, M. (2000) The Zebrafish Book (Univ. of Oregon Press, Eugene). 29. Hivert, B., Liu, Z., Chuang, C. Y., Doherty, P. & Sundaresan, V. (2002) Mol. 14. Kimmel, C. B., Ballard, W. W., Kimmel, S. R., Ullmann, B. & Schilling, T. F. Cell. Neurosci. 21, 534–545. (1995) Dev. Dyn. 203, 253–310. 30. Suchting, S., Heal, P., Tahtis, K., Stewart, L. M. & Bicknell, R. (2005) FASEB 15. Cross, L. M., Cook, M. A., Lin, S., Chen, J. N. & Rubinstein, A. L. (2003) J. 19, 121–123. Arterioscler. Thromb. Vasc. Biol. 23, 911–912. 31. Hutson, L. D., Jurynec, M. J., Yeo, S. Y., Okamoto, H. & Chien, C. B. (2003) 16. Yeo, S. Y., Little, M. H., Yamada, T., Miyashita, T., Halloran, M. C., Kuwada, Dev. Dyn. 228, 358–369. J. Y., Huh, T. L. & Okamoto, H. (2001) Dev. Biol. 230, 1–17. 32. Hu, H., Li, M., Labrador, J. P., McEwen, J., Lai, E. C., Goodman, C. S. & 17. Ekker, S. C. (2000) Yeast 17, 302–306. Bashaw, G. J. (2005) Proc. Natl. Acad. Sci. USA 102, 4613–4618. 18. Draper, B. W., Morcos, P. A. & Kimmel, C. B. (2001) Genesis 30, 154–156. 33. Lundstrom, A., Gallio, M., Englund, C., Steneberg, P., Hemphala, J., Aspen- 19. Simpson, J. H., Kidd, T., Bland, K. S. & Goodman, C. S. (2000) Neuron 28, strom, P., Keleman, K., Falileeva, L., Dickson, B. J. & Samakovlis, C. (2004) 753–766. Genes Dev. 18, 2161–2171.

6378 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0408318102 Bedell et al. Downloaded by guest on September 27, 2021