HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) All Versions of this Article: 19/1/121 04-1991fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Suchting, S. Articles by Bicknell, R. Search for Related Content PubMed PubMed Citation Articles by Suchting, S. Articles by Bicknell, R.

Soluble Robo4 receptor inhibits in vivo angiogenesis and endothelial cell migration

Molecular Angiogenesis Laboratory, Cancer Research UK, Weatherall Institute of Molecular Medicine, John Radcliffe Hospital, University of Oxford, Oxford, UK

¹ Correspondence: E-mail: roy.bicknell{at}cancer.org.uk

SPECIFIC AIMS

Roundabout receptors are molecular guidance molecules that function by interaction with Slit proteins to regulate axon guidance, neuronal migration, and leukocyte chemotaxis. We recently isolated a novel roundabout protein, Robo4, which is restricted in expression to endothelial cells in vitro and sites of angiogenesis in vivo. The aim of this study was to use the soluble exrtracellular domain of Robo4 as a probe of function in angiogenesis and endothelial biology. We further aimed to establish whether Robo4, in common with other Robo family members, functions as a Slit receptor.

PRINCIPAL FINDINGS

1. Soluble Robo4 receptor inhibits in vivo and in vitro angiogenesis
A soluble chimeric receptor (Robo4Fc) comprised of the Robo4 extracellular domain fused to the Fc region of human immunoglobulin was tested in the mouse subcutaneous sponge angiogenesis model. A piece of inert polyether sponge is implanted s.c. in the mouse flank and injected on alternate days with bFGF to stimulate angiogenesis and cellular ingrowth. Coinjection of Robo4Fc into sponges resulted in less cellular ingrowth and an absence of large vessels vs. injection of bFGF alone (Fig. 1 A). Counting of vessel numbers showed that Robo4Fc-treated sponges had significantly fewer vessels than bFGF-treated ones (Fig. 1A ). To confirm our results, we tested the effect of Robo4Fc on sprouting in the rat aortic ring assay. Rings were embedded in matrigel and cultured in full medium for several days to stimulate sprouting. Sprouts start to form by days 3–4, with a complex 3-dimensional network emerging by day 5. Addition of Robo4Fc to the culture medium significantly inhibited sprout formation compared with controls (Fig. 1B ). After replenishment of cultures with fresh medium (free from Robo4Fc), sprouting occurred after a lag of 2–3 days (Fig. 1B ), indicating rings remained viable in the presence of soluble Robo4. To examine the effect of Robo4Fc on HUVEC migration, we used a modified Boyden chamber assay system. Cells were stimulated to migrate from the upper to lower chambers by addition of 10 ng/mL VEGF or 10 ng/mL bFGF to the lower chambers. Soluble Robo4 but not control NABAFc protein had a potent inhibitory effect on HUVEC migration when added to the lower chamber (Fig. 1C ). Replacement of the Robo4Fc with 5% FCS stimulated migration showing that the cells remain viable. We next determined whether soluble Robo4 would affect the normal growth of endothelial cells. Addition of Robo4Fc to HUVEC cultures inhibited cell growth compared with controls; the inhibition was dose dependent from 1.25 µM to 40 nM (Fig. 1D ). To test the specificity of the inhibitory growth effect, we treated cultures of MCF-7 and MRC-5 cells with Robo4Fc. Soluble Robo4 had no effect on cell growth in either cell line (Fig. 1E ). Cell cycle analysis of HUVEC was performed by propidium-iodide staining followed by flow cytometry of cells after treatment with Robo4Fc (Fig. 1F ). HUVEC treated with Robo4Fc accumulated in the G1 phase at 16 h. Whereas control and IgG-treated cells reached confluence and exited the cell cycle by 96 h, Robo4Fc-treated cells remained subconfluent and continued to cycle, but at a slower rate. These results suggest that Robo4Fc may cause a G1/S phase cell cycle arrest. Treatment with Robo4Fc did not increase HUVEC apoptosis as assessed by increased Annexin-V staining (% apoptotic cells: control 1.15%±0.19; IgG treatment 1.01%±0.11; Robo4Fc treatment 1.20%±0.14).

View larger version (49K): [in this window] [in a new window]   Figure 1. A) Inhibition of angiogenesis by Robo4Fc. Polyether sponges were s.c. implanted into C57 black mice and injected with 10 ng bFGF or 10 ng bFGF + 10 µg Robo4Fc. After 21 days sponges were harvested, sectioned, and stained with H&E (left panel). Vessel counts (right) showed that Robo4Fc treatment resulted in significantly fewer vessels than control. Images obtained at 20x. The vessels in sponges from 4 mice/group were counted by 2 independent observers (*P<0.001, Student’s t test). B) Inhibition of sprouting from rat aortic ring by Robo4Fc. 1–2 mm sections of rat aortic ring were cultured in matrigel in the presence of 1.25 µM Robo4Fc or control human IgG. Left: after 5 days culture control and IgG-treated rings showed numerous sprouts. In contrast, Robo4Fc-treated rings had no or very few sprouts. Replenishing Robo4Fc-treated rings with media free from Robo4Fc resulted in regrowth of sprouts over the next 3 days (replenished media). Right: quantification of sprouting on a scale from 0 (no sprouts) to 4 (profuse sprouting). Robo4Fc-treated rings show significantly lower levels of sprouting than controls. Images were taken at 4x. Sprouts quantified by 3 independent observers (*P<0.002, ANOVA, P<0.001 Kruskall-Wallis test). Bars, SD. C) HUVEC were placed in modified Boyden chambers and exposed to 10 ng/mL VEGF or 10 ng/mL bFGF in the presence of Robo4Fc or control NABAFc protein (both 1.25 µM). NABAFc is a synthetic fusion protein comprised of the N1, A1, and A2 domains of biliary glycoprotein, the B3 domain of carcinoembryonic antigen protein, and the Fc region of immunoglobulin. After 22 h cells were stained with fluorescent dye Calcein-AM; cell migration was assessed by microscopy (left) and quantification using a fluorescence plate reader (right). Wells treated with Robo4Fc showed fewer migrating cells than untreated wells. Replacement of the Robo4Fc with media containing 5% FCS stimulated migration after another 16 h (far right) shows that cells remained viable. Images taken at 10x. *P< 0.001, Student’s t test; bars = SD. D) Inhibition of HUVEC growth by Robo4Fc. 5 x104 HUVEC/well were seeded into 6-well plates and cultured in complete media in the presence of Robo4Fc or control IgG. Left: treatment with 1.25 µM Robo4Fc for 96 h caused inhibition of HUVEC growth vs. untreated or control human IgG-treated cells. Right: inhibition of cell growth by Robo4Fc was dose dependent over a range of 1.25 µM–40 nM; shown at 96 h (*P <0.05, Student’s t test) Bars, SD. E) Robo4Fc does not inhibit growth of MCF-7 or MRC-5 cells. 5 x 104 MCF-7 or MRC-5 cells were seeded into 6-well plates and cultured in full media in the presence of 1.25 µM Robo4Fc or control IgG. Treatment with Robo4Fc over 12 days had no effect on cell growth. Bars, SD. F) Robo4Fc causes accumulation of cells in the G1/S phase as assessed by PI staining and FACS analysis. All experiments were performed at least 3 times.

2. Robo4 does not interact with known Slit proteins
Roundabout receptors function via interaction with Slit family ligands. Slit proteins therefore represented likely candidates for the Robo4 ligand, with Robo4Fc acting by interfering with Slit-mediated effects on endothelial cell function. To determine if Slit proteins interact with Robo4, we examined the ability of Robo4 to bind Slit1, Slit2 and Slit3 by coimmunoprecipitation and BiaCore analysis. Slit proteins tagged with myc epitope and soluble Fc-tagged Robo1 or Robo4 extracellular domains were expressed in COS-7 cells. We found that Slit3 was not secreted from cells in our system and therefore used cell lysates for Slit3 experiments. Concentrated preparations of Robo1 or Robo4 conditioned media were mixed with Slit1 or Slit2 salt washes or Slit3 cell lysate and immunoprecipitated with Protein A beads. Immunoprecipitated complexes were analyzed by immunoblot using anti-myc antibody. Robo1 was able to immunoprecipitate all three Slit proteins whereas Robo4 was unable to immunoprecipitate any Slit protein. To confirm these results, we used BiaCore analysis. Pure protein preparations of Fc-tagged Robo1 or Robo4 proteins were covalently immobilized to chips using standard amine coupling. Robo1 interacted with salt washes containing Slit1 or Slit2 and with cell lysate containing Slit3. The relative response unit (RU) values for these interactions were: Slit1, 435 RU; Slit2, 426 RU; and Slit3, 458 RU. Under identical experimental conditions no interaction was detected between Robo4 and salt washes containing Slit1 or Slit2, or with lysates containing Slit3. Relative RU values for these interactions were Slit1, 2 RU; Slit2, -2 RU; and Slit3, 7 RU. Robo4 did interact with a monoclonal antibody raised against its first Ig domain (491 RU), indicating that Robo4 was attached to the chip and free to associate with ligand.

CONCLUSIONS AND SIGNIFICANCE

During angiogenesis, new endothelial-lined blood vessels sprout from the sides and ends of preexisting vessels and grow into areas requiring oxygenation and nutrient exchange. Roundabouts (Robos) are large transmembrane receptor proteins that bind to Slit ligands to induce a repulsive signal during axon guidance and neuronal migration. Although first characterized in neuronal development, Slit-Robo interactions are now implicated in diverse processes, including endothelial migration and leukocyte chemotaxis, indicating they may comprise a conserved migration/guidance pathway functioning in many cell types. Robo4 was first identified by us as an endothelial-specific member of the roundabout family and later shown to be differentially expressed in activin-like kinase-1-deficient mice, which exhibit impaired angiogenesis during development. In silico and in vitro expression analysis indicates that Robo4 is highly endothelial specific; its expression in vivo appears to be restricted to areas of developmental and tumor angiogenesis, with no expression detected in normal adult vessels.

In this work we show that the soluble extracellular domain of Robo4 has potent inhibitory effects on endothelial cell function and on angiogenesis in vivo. To account for these effects, we hypothesized that the soluble receptor would likely be blocking Slit-mediated effects on endothelial cell function. A role for Slit2 in promoting endothelial cell migration and tube formation in vitro and angiogenesis in vivo has been reported. Our results, however, suggest that neither Slit1, Slit2, nor Slit3 is able to interact with Robo4. The failure to detect an interaction between Robo4 and Slits by BiaCore analysis even though Robo1 interacted with Slits in this system argues against Robo4 being a Slit receptor. The absence of binding by Robo4 to Slits may reflect the substantial structural differences between Robo4 and other roundabouts in the extracellular region. Whereas archetypal Robo has an extracellular domain comprised of five Ig domains and three FN domains, Robo4 has only two Ig domains and two fibronectin domains.

Controversy remains concerning the role of Slit in endothelial cell function. Conflicting reports have suggested that Slit2 is inhibitory to migration of endothelial cells, or acted as a chemoattractant to endothelial cells and to 293 cells overexpressing Robo1. These studies and ours raise several fundamental questions: 1) what effect does Slit have on endothelial cell function? 2) through which Robo receptor is Slit acting? and 3) do Robo1 and Robo4 functionally interact? Our results suggest that Robo4 is activated by a Slit-independent pathway (Fig. 2 ). The identity of the Robo4 ligand and the mechanism of Robo4Fc action remain to be determined.

View larger version (42K): [in this window] [in a new window]   Figure 2. Robo4Fc may inhibit endothelial cell migration and angiogenesis by blocking the effects of an unidentified ligand on Robo4 function. Slit2 activates Robo1, but not Robo4, to increase endothelial cell migration and angiogenesis. Robo4 may directly or indirectly modulate Robo1 function.

This article has been cited by other articles:

				M. O'Rourke, C. Ward, J. Worthington, J. McKenna, A. Valentine, T. Robson, D. G. Hirst, and S. R. McKeown Evaluation of the Antiangiogenic Potential of AQ4N Clin. Cancer Res., March 1, 2008; 14(5): 1502 - 1509. [Abstract] [Full Text] [PDF]

				Y. Okada, K. Yano, E. Jin, N. Funahashi, M. Kitayama, T. Doi, K. Spokes, D. L. Beeler, S.-C. Shih, H. Okada, et al. A Three-Kilobase Fragment of the Human Robo4 Promoter Directs Cell Type-Specific Expression in Endothelium Circ. Res., June 22, 2007; 100(12): 1712 - 1722. [Abstract] [Full Text] [PDF]

				C. Fouquet, T. Di Meglio, L. Ma, T. Kawasaki, H. Long, T. Hirata, M. Tessier-Lavigne, A. Chedotal, and K. T. Nguyen-Ba-Charvet Robo1 and Robo2 Control the Development of the Lateral Olfactory Tract J. Neurosci., March 14, 2007; 27(11): 3037 - 3045. [Abstract] [Full Text] [PDF]

				S.-A. Hussain, M. Piper, N. Fukuhara, L. Strochlic, G. Cho, J. A. Howitt, Y. Ahmed, A. K. Powell, J. E. Turnbull, C. E. Holt, et al. A Molecular Mechanism for the Heparan Sulfate Dependence of Slit-Robo Signaling J. Biol. Chem., December 22, 2006; 281(51): 39693 - 39698. [Abstract] [Full Text] [PDF]

				M. Fujiwara, M. Ghazizadeh, and O. Kawanami Potential role of the Slit/Robo signal pathway in angiogenesis Vascular Medicine, May 1, 2006; 11(2): 69 - 74. [Abstract] [PDF]

				S. Kaur, M. D. Castellone, V. M. Bedell, M. Konar, J. S. Gutkind, and R. Ramchandran Robo4 Signaling in Endothelial Cells Implies Attraction Guidance Mechanisms J. Biol. Chem., April 21, 2006; 281(16): 11347 - 11356. [Abstract] [Full Text] [PDF]

				R. H. Adams Nerve Cell Signposts in the Blood Vessel Roadmap Circ. Res., March 3, 2006; 98(4): 440 - 442. [Full Text] [PDF]

				N. S. Patel, J.-L. Li, D. Generali, R. Poulsom, D. W. Cranston, and A. L. Harris Up-regulation of Delta-like 4 Ligand in Human Tumor Vasculature and the Role of Basal Expression in Endothelial Cell Function Cancer Res., October 1, 2005; 65(19): 8690 - 8697. [Abstract] [Full Text] [PDF]

				S. Ghenea, J. R. Boudreau, N. P. Lague, and I. D. Chin-Sang The VAB-1 Eph receptor tyrosine kinase and SAX-3/Robo neuronal receptors function together during C. elegans embryonic morphogenesis Development, August 15, 2005; 132(16): 3679 - 3690. [Abstract] [Full Text] [PDF]

				V. M. Bedell, S.-Y. Yeo, K. W. Park, J. Chung, P. Seth, V. Shivalingappa, J. Zhao, T. Obara, V. P. Sukhatme, I. A. Drummond, et al. roundabout4 is essential for angiogenesis in vivo PNAS, May 3, 2005; 102(18): 6373 - 6378. [Abstract] [Full Text] [PDF]

				A. Eichmann, T. Makinen, and K. Alitalo Neural guidance molecules regulate vascular remodeling and vessel navigation Genes & Dev., May 1, 2005; 19(9): 1013 - 1021. [Abstract] [Full Text] [PDF]

This Article Full Text (PDF) All Versions of this Article: 19/1/121 04-1991fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Suchting, S. Articles by Bicknell, R. Search for Related Content PubMed PubMed Citation Articles by Suchting, S. Articles by Bicknell, R.