Role of the vascular endothelial growth factor isoforms in retinal angiogenesis and DiGeorge syndrome.

The aim of this study was to characterize the specific role of the various vascular endothelial growth factor (VEGF) isoforms in different aspects of blood vessel formation: vessel outgrowth, arterial and venous differentiation, and vascular remodeling and patterning. Although the role of VEGF in the early stages of vascular assembly has been studied extensively, its role in the maturation stage, involving vascular remodeling and patterning, as well as in the establishment of arteries and veins, remains enigmatic. The three major VEGF isoforms are known to differ in their solubility (VEGF120 is freely soluble and VEGF188 is completely matrix-bound, while VEGF164 has intermediate properties) and receptor binding properties (VEGF164 does and VEGF120 does not bind to neuropilin-1 (Nrp-1)), but the specific biological function of these VEGF isoforms is largely unknown. To study the differential function of the VEGF isoforms in these particular aspects of vascular development, three different transgenic mice were generated: VEGF(120/120), VEGF(164/164) and VEGF(188/188), that express only VEGF120, VEGF164 or VEGF188, respectively. Postnatal blood vessel formation was studied in the retina, which is an excellent organ to study angiogenesis because of the unique structural properties of the retinal vascular bed. Subsequently, the cardiac outflow tract and pharyngeal arch system were analyzed during embryogenesis, since vascular remodeling and patterning play a crucial role in the establishment of the mature configuration of these vascular structures. Both vascular systems are of major clinical relevance. Retinal neovascularization, the major cause of blindness, is a complication of a variety of common eye diseases, including diabetic retinopathy, age-related macular degeneration, retinopathy of prematurity, and vascular occlusions. Abnormal remodeling of the pharyngeal arch system and the cardiac outflow tract, on the other hand, results in life-threatening congenital cardiovascular defects, and occurs in association with craniofacial, thymic and parathyroid defects in DiGeorge syndrome (DGS) that affects 1/4000 live births. Eigthy to ninety percent of DGS-affected individuals are heterozygous a micro-deletion of chromosome 22q111, but the search for causal genes in the remaining 10-20% of patients with DGS remains ungoing. Moreover, the variable penetrance and severity of this syndrome suggests the contribution of additional modifier genes outside this chromosomal region. Finally, deletions of chromosome 22q11 only encounter for 15% of all cases of conotruncal defects, and the other gene(s) involved in the pathogenesis of these conotruncal defects in remain to be identified. The identification and characterization of these additional causal and modifier genes is an important goal for the future. Extensive investigation of the various aspects of vascular development in the VEGF isoform specific mice, using the retina as a model, revealed that vascular development was normal in VEGF(164/164) mice (that only express VEGF164), indicating that this isoform contains all necessary information for normal (arterial and venous) outgrowth, remodeling and patterning of blood vessels. In contrast, VEGF(120/120) mice exhibited pronounced vascular defects, with impaired venous and severely defective arterial vascular development in the retina. VEGF(188/188) mice had normal venous development, but aborted retinal arterial outgrowth. Dramatically reduced retinal vascular outgrowth in the mice that exclusively express the soluble VEGF120 isoform indicates that the longer isoforms are crucial for the establisment of a VEGF-gradient that guides the endothelial cells towards the periphery of the retina. Moreover, mice that lack the VEGF164 isoform exhibit impaired arterial outgrowth despite normal arterial and venous differentiation. This observation provides evidence for the recent theory that arterial and venous differentiation is predetermined in endothelial cells rather than established after initial outgrowth of undifferentiated vessels, and dedicates a novel role to VEGF164 in outgrowth of arterially differentiated endothelial cells. Predominant arterial expression of Nrp-1 implies that VEGF164 may mediate arterial outgrowth via Nrp-1. Half of the VEGF(120/120) neonates die within a few hours after birth because of conotruncal defects that are typically observed in DiGeorge syndrome. Further analysis revealed that these mice also exhibit aortic arch anomalies, a cleft palate, micrognathia, as well as absent and/or ectopic parathyroid glands and thymus. Thus, absence of the VEGF164 isoform in mice causes the entire spectrum of characteristic lifethreatening cardiovascular malformations and craniofacial, thymic and parathyroid defects of DGS, while mice expressing only the VEGF164-isoform appear normal. VEGF164 expression consistently colocalized with its receptor, Nrp-1 at DGS predilection sites, suggesting that both provide critical guidance or differentiation cues for vascular remodeling. In the VEGF164 deficient mice, no neural crest cell migration or differentation defects were detected. Moreover, the DGS phenotype was most prominent in mice with severe vascularization defects, possibly indicating that derailed signaling by vascular growth factors may be more important than originally anticipated, and suggesting a vascular etiology underlying DGS. This vascular hypothesis implies that DGS may be primarily a vascular phenotype, irrespective of the involvement of neural crest cells. Taken together, our observations indicate that, while the distinct VEGF isoforms are redundant for initial vessel assembly and growth, they differ greatly in providing critical spatial guidance cues for vascular remodeling and, perhaps also for differentiation of neural crest cell-derived tissues. These data implicate that the VEGF164-isoform may represent a candidate disease effector or modifier in the pathogenesis of congenital cardiovascular malformations in general, and of DGS in particular. Finally, I would like to conclude with the recent words of Dr. D. Srivastava: "Discovery of the causes of complex genetic traits, such as congenital heart defects, has been difficult. However, the observation that secondary factors, be they genetic or environmental, may contribute to DiGeorge syndrome provides hope for the treatment and prevention of congenital heart defects. While prospects for gene therapy remain in the distant future, knowledge of the genetic pathways regulating cardiogenesis should lead to some of the secondary factors that may be modulated during the period of embryonic heart development. Given the rapid pace of discovery and the ever-increasing tools available to scientists and clinicians, the hope of translating genetic information regarding heart formation into tangible benefits for families with congenital heart defects has never been brighter".