Chapter 52
Slit-Robo Signaling in Ocular Angiogenesis
Haoyu Chen, Mingzhi Zhang, Shibo Tang, Nyall R. London, Dean Y. Li, and Kang Zhang
Abstract Slit-Robo signaling was firstly discovered as a major repellent pathway at the midline of the central nervous system. Intense investigation found that this pathway also plays an important role in other biological process including angiogenesis. Robo4 is the vascular endothelial cell specific member of Robo family. It was found that Slit-Robo signaling can inhibit endothelial cell migration, tube formation and vascular permeability. Slit-Robo signaling also plays an important role in embryonic and tumor angiogenesis. In animal model of ocular angiogenesis, addition of Slit inhibited laser induced choroidal neovascularization, oxygen induced retinopathy and VEGF induced retinal permeability in a Robo4 dependent manner. Recent data demonstrates that Robo1 and Robo4 form a heterodimer in endothelial cells, The role of this heterodimer in counteracting VEGF signaling is unknown. Further investigation is required to better understand Slit-Robo signaling and develop novel therapy for angiogenesis.
52.1 Ocular Angiogenesis
Angiogenesis, the growth of new blood vessels from pre-existing vessels, is an important biological process involved in several physiologic and pathological conditions. These include development of embryo vasculature, would healing, female reproductive cycling, tumor growth and metastasis, ischemic cardiovascular diseases, ocular disorders and so on. The eye is an optic transparent organ, which allows in vivo observation of retina and choroid with the assistant of some optic instruments. Therefore, the eye provides a valuable model to study angiogenesis
H. Chen (B)
Joint Shantou International Eye Center, Shantou University and the Chinese University of Hong Kong, North Dongxia Road, Shantou, 515041 China; Moran Eye Center, University of Utah, Salt Lake City, Utah 84112 USA
e-mail: drchenhaoyu@gmail.com
R.E. Anderson et al. (eds.), Retinal Degenerative Diseases, Advances in Experimental |
457 |
Medicine and Biology 664, DOI 10.1007/978-1-4419-1399-9_52,C Springer Science+Business Media, LLC 2010
458 |
H. Chen et al. |
(Campochiaro and Hackett 2003). Development of retinal vasculature is an example of physiological angiogenesis. Pathological angiogenesis or neovascularization may occur in several ocular tissues, such as cornea, iris, anterior chamber angle, retina and choroid, and usually result in severe vision impairment even blindness.
Age-related macular degeneration (AMD) is one of the leading causes of irreversible blindness in developed countries (Friedman et al. 2004). The prevalence of AMD in the developing world such as China is increasing as well (Zou et al. 2005). Among two clinical types, wet AMD usually results in more severe vision loss due to neovascularization of choroid by breakthrough of Bruch’s membrane to sub subretinal or sub retinal pigment epithelium space, hemorrhage and eventually fibrous formation. Diabetic retinopathy (DR) is the most common cause of legal blindness in working-age United States adults and accounts for 10% of new blindness at all ages (Varma et al. 2007). The most severe type of DR, proliferative diabetic retinopathy (PDR) is characterized by retinal neovascularization, eventually vitreoretinal hemorrhage, tractional retinal detachment and vision loss. Macular edema is a clinical syndrome related to retinal or choroidal neovascularization. It occurs when fluid and proteins accumulated at macular region secondary to the leakage of inner or outer blood-retina barrier.
The imbalance of angiogenic and anti-angiogenic factors is the major mechanism of angiogenesis. Vascular endothelial growth factor (VEGF) is the most potent angiogenic factor and anti-VEGF therapy has been successfully showed inhibiting tumor and macular degeneration (Rosenfeld et al. 2006). However, more and more novel signaling pathways have been discovered by intense investigation. These findings help us better understand the pathogenesis of angiogenesis and provide novel therapeutic targets.
52.2 Slit-Robo Signaling in Axon Guidance
The nerve fibers and blood vessels follow parallel routes in peripheral tissues. For example, in the retina, the relationships between retinal blood vessel and ganglion cell axons are accompanied by each other in a radial orientation (Gariano and Gardner 2005). More importantly, the pattern of growth of nerve fiber and blood vessel are similar too. During development, there is a special structure at the distal tip of neuron axon called growth cone, which can dynamically extend filopodia to explore repulsive or attractive guidance cues in the spatial environment, interpret these signals and direct the growth of axon. Similarly, there is specialized endothelial cell, termed tip cell, covers the distal end of vascular sprouts. Tip cells dynamically form filopodial extension and recognize signals at surrounding cells and the matrix environment (Adams 2006). Based on the structure and functional similarity, it is suggested that the signals controlling axon pathfinding may also control endothelial tip cell guidance and angiogenic sprouting of blood vessels. There are several examples of these pathways, including Ephrins-Eph receptors, Semaphorins-Neuropilins, Netrins-UNC5 and Slit-Robo.
The Slit and Robo gene was first identified from Drosophila melanogaster. There are three family members of Slit identified till now, Slit1, Slit2 and Slit3. The Slits
52 Slit-Robo Signaling in Ocular Angiogenesis |
459 |
are large, secreted multidomain glycoproteins including leucine-rich repeats, EGFlike repeats, laminin G domain, and C-terminal cysteine-rich knot (Rothberg et al. 1990). Robo is the receptor of Slit, and there are four Robo family members identified so far, Robo1, Robo2, Robo3 and Robo4. Robo is a single-pass transmembrane receptor with an extracellular region containing immunoglobulin (Ig) domains and fibronectin type III repeats, and an intracellular tail composed of conserved motifs (CC0, CC1, CC2, and CC3) (Kidd et al. 1998).
Slit-Robo signaling was first discovered as a major repellent pathway at the midline of the central nervous system. Activation of Slit-Robo signaling triggers changesin cytoskeletal structure within the growth cone and results in axon repulsion (Li et al. 1999). With extensive investigation, many other Robo-dependent Slit functions were discovered, including axon branching and migration of neurons, migration of leucocytes and inflammation, migration of endothelial cells, development of the lung and kidney, tumor angiogenesis and metastasis (Hohenester et al. 2006). We will discuss the role of Slit-Robo signaling in angiogenesis in detail below.
52.3 Slit-Robo Signaling in Angiogenesis
The endothelial cell specific Robo family member, called Robo4 or magic roundabout, was firstly identified using a bioinformatic strategy. Robo4 has similar domains in cytoplasmic region and transmembrane domain with other Robo members. However, there are only two IgG and two fibronectin domains in the ligand-binding region of Robo4, which is quite different from the five IgG and three fibronectin domains of the other Robo family members (Huminiecki et al. 2002). It was reported that Robo4 bind to Slit2 in immunoprecipitation and immunofluorescence, despite its unique extracellular domain structure (Park et al. 2003).
In vitro studies showed that Robo4 expressed specific in endothelial cells but not in neural, vascular smooth muscle cells or other types of cell (Huminiecki et al. 2002; Jones et al. 2008), while Robo1, Robo2, Robo3 did not express in human microvascular endothelial cell (HMVEC) (Park et al. 2003). In vivo, Robo4 expressed specifically in embryonic vasculature and predominantly in adult vasculature (Park et al. 2003). Further investigation revealed that Robo4 expression was limited to stalk cells but not tip cells in the developing retinal vasculature (Jones et al. 2008). The expression of Robo4 was upregulated by hypoxia in endothelial cells (Huminiecki et al. 2002) and mice retina (Jones et al. 2008). Robo4 is also over expressed in tumors endothelial cells compared to normal adult endothelial cells in numerous solid tumors (Huminiecki et al. 2002; Seth et al. 2005).
There is controversy on the function of Slit-Robo signaling in endothelial cells and angiogenesis. It was reported that the endothelial cell migration induced by VEGF or Human Embryonic Kidney 293 cells conditioned media was inhibit by Slit-myc conditioned media while the proliferation was not affected. The inhibitory
460 |
H. Chen et al. |
effect of Slit-myc conditioned media on HMVEC migration is blocked by depleting the conditioned media with either N-Robo1 or anti-myc antibody (Park et al. 2003). In another study, endothelial cell migration induced by VEGF, fibroblast growth factors (FGF) or fetal bovine serum was inhibited by transfection of Robo4 but not soluble Robo4. The migration effect of VEGF was also inhibited by external addition of Slit2 and rescued by addition of soluble Robo4 (Seth et al. 2005). It was also reported that recombinant N terminal cleavage of Slit2 prevented the platelet-derived growth factor (PDGF)-stimulated migration of vascular smooth muscle cells (Liu et al. 2006). Endothelial cell migration, tube formation and permeability induced by VEGF have been found to be inhibited by Slit2 and the inhibitory effects were lost in Robo4 knockout endothelial cells (Jones et al. 2008). These finding suggested that Slit-Robo signaling is an inhibitor of angiogenesis. However, it was also reported that Slit-Robo signaling has pro-angiogenic effects and blockage of Slit-Robo signaling can inhibit angiogenesis. Recombinant Slit2 protein was reported to attract endothelial cells and promote tube formation in a Robo1-dependent manner (Wang et al. 2003). It was also reported that soluble Robo4 inhibited VEGFand FGF-stimulated endothelial cell migration and endothelial proliferation in vitro, inhibited tube formation in the rat aortic ring assay Ex vivo and angiogenesis in the rodent subcutaneous sponge model in vivo (Suchting et al. 2005).
The role of Robo4 in embryonic angiogenesis was studied in zebrafish. Both overexpression and knockdown of zebrafish Robo4 resulted in asynchronous intersomitic vessel (ISV) sprouting, culminating in a reduction and misdirection of the intersomitic vessels. The vascular phenotype in Robo4 knockdown was rescued by human Robo4 expression (Bedell et al. 2005). In another report, injection of zebrafish Robo4 disrupted ISVs sprouting from dorsal aorta. Furthermore, angioblasts isolated from Robo4 embryos showed movement that resemble cells actively searching for guidance and continue rolling in a nondirectional manner (Kaur et al. 2008).
A role for Slit-Robo signaling in tumor angiogenesis has also been reported. Robo4 expression was increased on tumor vessels of brain, colon, breast, kidney, and bladder. Further histological studies found that the expression of Robo4 was upregulated in site of active angiogenesis (Huminiecki et al. 2002; Seth et al. 2005). Deletion or epigenetic modification of the Slit–Robo genes has been identified in the progression of numerous cancers (Dallol et al. 2002; Narayan et al. 2006). Blockade of Slit2 activity in tumor bearing mice through overexpression of RoboN or addition of a monoclonal antibody that blocks Robo1–Slit binding gave reduced tumor microvessel densities and tumor masses (Wang et al. 2003).
52.4 Slit-Robo Signaling in Ocular Angiogenesis
In the mouse retina, Robo4 is expressed specifically in vascular endothelial cells, which was detected by immunohistochemistry. While Slit2 expression was detected