The Role of Neuropilin in Vascular and Tumor Biology

Summary

Neuropilin1 (NRP1) and NRP2 are related transmembrane receptors that function as mediators of neuronal guidance and angiogenesis. NRPs bind members of the class 3 semaphorin family, regulators of neuronal guidance, and of the vascular endothelial growth factor (VEGF) family of angiogenesis factors. There is substantial evidence that NRPs serve as mediators of developmental and tumor angiogenesis. NRPs are expressed in endothelial cells (EC) and bind VEGF₁₆₅. NRP1 is a coreceptor for VEGF receptor2 (VEGFR2) that enhances the binding of VEGF₁₆₅ to VEGFR2- and VEGF₁₆₅-mediated chemotaxis. NRP1 expression is regulated in EC by tumor necrosis factor α, the transcription factors dHAND and Ets1, and vascular injury. During avian blood vessel development NRP1 is expressed only in arteries whereas NRP2 is expressed in veins. Transgenic mouse models demonstrate that NRP1 plays a critical role in embryonic vascular development. Overexpression of NRP1 results in the formation of excess capillaries and hemorrhaging. NRP1 knockouts have defects in yolk sac, embryo and neuronal vascularization, and in development of large vessels in the heart. Tumor cells express NRPs and bind VEGF₁₆₅. NRP1 upregulation is positively correlated with the progression of various tumors. Overexpression of NRP1 in rat tumor cells results in enlarged tumors and substantially enhanced tumor angiogenesis. On the other hand, soluble NRP1 (sNRP1) is an antagonist of tumor angiogenesis. Semaphorin 3A binds to EC and tumor cells. It also inhibits EC motility and capillary sprouting in vitro. VEGF₁₆₅ and Sema3A are competitive nhibitors for NRP1-mediated functions in EC and neurons. These results suggest that NRP1 is a novel regulator of the vascular system.

Introduction

Neuropilins (NRPs) are mediators of neuronal guidance and angiogenesis.¹⁶ NRP1, a 130-140 kDa cell-surface glycoprotein, was first identified in developing nervous tissue.⁷⁹ It is a highly conserved type 1 membrane protein. Subsequently a second gene, NRP2, was identified that shared a similar structure.¹⁰ Despite a 45-50% structural homology, NRP1 and NRP2 differ considerably in their biological properties (Table 1).

Table

Comparison of NRP1 and NRP2.

In the nervous system NRP expression is localized to axons as opposed to the somata of neurons. It is expressed in axons of particular neuron classes and at stages when axons are actively growing to form neuronal connections. These original observations suggested that NRP is involved in growth, nerve fiber fasciculation and neuronal guidance (see chapter 2). Subsequently, it was discovered that NRP expression is not confined to the developing embryonic nervous system. It is also expressed in the following: developing heart, vasculature, and limb;¹¹ many adult tissues such as the heart, placenta, lung, kidney and epidermis,^12–14 and uterine glandular epithelium;¹⁵ and many cell types such as endothelial cells (EC),¹¹ tumor cells,¹² neural crest cells,¹⁶ osteoblasts,^17,18 marrow stromal cells,¹⁹ human mesangial cells,²⁰ neuroendocrine cells (NRP2)²¹ and glomerular epithelial cells.²² These expression patterns suggest that NRPs have physiological roles well beyond mediating neuronal guidance. Expression of NRPs in EC and tumor cells will be described in detail below.

NRPs are highly conserved among vertebrate species. The homology between NRP1 and NRP2 is 45%.¹⁰ The primary structure of NRPs contains a relatively large extracellular domain of about 860 amino acids, a transmembrane domain and a relatively short cytoplasmic domain of 40 amino acids.^7,8 The extracellular domain in turn is composed of five subdomains, each of which is thought to be involved in molecular and/or cellular interactions. These subdomains are referred to as a1, a2, b1, b2, and c. The a1a2 and b1b2 are tandem repeats which are involved in ligand binding. The c domain is responsible for homo or heterodimerization of NRP1 and NRP2. The function of the short cytoplasmic domain, the most highly conserved domain, with over 90% homology, is not clear. However, a PDZ domain-containing protein has been isolated using the two-yeast hybrid system that interacts with the C-terminal three amino acids of NRP1 (SEACOOH).²³

In addition to full-length NRPs, some cell types also express truncated NRP isoforms.^13,14 These proteins contain the a1a2 and b1b2 subdomains, but lack the c, transmembrane and cytoplasmic domains. These naturally occurring 60-90 kDa proteins are soluble and released by cells. Several soluble NRPs (sNRP) have been cloned, three sNRP1s and one sNRP2. The sNRP molecules are produced by premature truncation within introns and as a result are characterized by having intron-derived sequences, nucleotides and amino acids, at their C-termini.

Neuropilin Expression in Endothelial Cells

There have been a number of reports demonstrating that blood vessel EC express NRPs; for example, umbilical vein EC, aortic EC and capillary EC.^12,48 Expression in vivo has been demonstrated, for example, in the heart,¹¹ coronary blood vessels,⁴⁹ glomerular capillaries,⁵⁰ and bone capillaries.^17,18 Vascular smooth muscle cells (vSMC) also express NRP1, which suggests a possible role in EC/vSMC interactions.⁵¹

NRPs are not ubiquitously expressed in EC. Previous reports have shown differential growth factor/receptor expression patterns on blood vessels; for example, expression of ephrinB2 on arteries and of EphB receptors on veins.⁵² Two recent reports demonstrated that there are differential embryonic blood vessel expression patterns for NRP1 and NRP2 as well.^53,54 In the avian vascular system NRP1 and NRP2 are both expressed in blood islands which are the earliest vascular structures. However, once arteries and veins differentiate, NRP1 is expressed exclusively in arteries and in mesenchyme surrounding developing arteries,^53,54 and NRP2 is expressed only in veins.⁵⁴ Similar expression patterns were detected throughout chick and quail development. Using quail arterial EC grafted onto chick embryos, EC expressing specific markers colonized both the arteries and veins of the chick embryo,⁵³ suggesting that expression of NRPs in EC is insufficient to determine the fate of these cells. In mice, a recent report has shown that in the retina, NRP1 is predominantly expressed in arterioles, suggesting there may be similar NRP expression pattern in mammals as in birds.⁵⁵

Regulation of Neuropilin Expression in Blood Vessels

NRP expression is regulated by cytokines and transcription factors. For example, tumor necrosis factor α (TNFα1 upregulates in a dose- and time-dependent manner the expression and the function of VEGFR2, as well as the expression of NRP1 in human EC.⁵⁶

The basic helix-loop-helix transcription factor, dHAND/Hand2, is expressed in the developing vascular mesenchyme and derivative vSMC. Targeted deletion of the dHAND gene in mice revealed severe defects of embryonic and yolk sac vascular development by E9.5.⁵⁷ In the dHAND mouse knockout, EC appear to be normal. The vascular mesenchymal cells migrated appropriately but failed to make contact with vascular EC and did not differentiate into vSMC. In a subtractive hybridization screen for genes comparing wild-type and dHAND-null hearts, NRP1 was found to be downregulated in dHAND mutants. At E9.5 the expression of dHAND and NRP1 in wild-type mice overlapped in the developing vasculature, for example, in aorta, aortic arch arteries and yolk sac. In the dHAND-null mice NRP1 expression was severely downregulated, specifically in blood vessels that expressed dHAND. These results suggest that dHAND is required for normal cardiovascular development and that it regulates angiogenesis, possibly through a NRP1-dependent mechanism.

Whereas dHAND regulates NRP1 expression in vSMC, another transcription factor, Ets1, induces NRP1 expression in EC. Ets1 is expressed in EC during angiogenesis and is induced by angiogenesis factors including VEGF.⁵⁸ Ets1 was transiently overexpressed in human umbilical vein EC (HUVEC) and potential downstream targets of Ets1 were analyzed by cDNA microarray analysis.⁵⁹ NRP1 was one of several angiogenesis-related genes induced by Ets1. In contrast, dominant-negative Ets1 decreased the levels of NRP1 mRNA and protein. Since TNFα increases the expression of both NRP1⁵⁶ and Ets1⁶⁰ in EC, it is possible that TNFα induces NRP1 expression via Ets1.

Neuropilin and Angiogenesis

NRP1 appears to be a coreceptor of VEGFR2 in cultured EC.¹² When coexpressed in cells with VEGFR2, NRP1 enhances the binding of VEGF₁₆₅ to VEGFR2 and VEGF₁₆₅-mediated chemotaxis. Conversely, inhibition of VEGF₁₆₅ binding to NRP1 inhibits its binding to VEGFR2 and its mitogenic activity for EC.

There is ample evidence from transgenic mouse studies that NRPs mediate angiogenesis, both normal and pathological. The first hint to this effect was a transgenic mouse study in which NRP1 was overexpressed.¹¹ In wild-type mice, NRP1 is expressed in the cardiovascular system, nervous system and limbs at particular developmental stages. The transgenics overexpressing NRP1 were embryonic lethal and displayed several morphological abnormalities. Besides ectopic sprouting and defasciculation of nerve fibers, there was an abnormal vascular phenotype that included excess capillaries and blood vessels, dilation of blood vessels, hemorrhaging and malformed hearts. The chimeric embryos usually appeared redder than their normal counterparts, suggesting that blood vessels were leaky which was possibly due to the enhanced vascular permeability activity of VEGF₁₆₅. Extra digit formation was also noted. These abnormalities occurred in the organs in which NRP1 was expressed in normal development. It was concluded that expression of NRP1 was essential not only for neuronal development but also development of the cardiovascular system and limbs.

Knockout studies have been very useful in determining the physiological role of NRPs in angiogenesis. In the initial study it was demonstrated that NRP1-deficient mutant mice were embryonic lethal between E12.5 to E13.5 and had, for example, severe abnormalities in the trajectory of efferent fibers of the peripheral nervous system.⁶¹ Interestingly, it was mentioned but not demonstrated that the embryo died due to cardiovascular defects. A follow-up study analyzed cardiovascular defects in depth.⁶² The NRP1 mutant mouse embryos exhibited defects in yolk sac, embryo and neuronal vascularization, and in development of large vessels in the heart. In yolk sacs and embryos the vascular network of large and small vessels was disorganized, the capillary networks were sparse, and normal branching did not occur. In the central nervous system (CNS) capillary invasion into the CNS was delayed for more than 1 day and the capillary networks that were in the CNS were disorganized and had degenerated. In the cardiovascular system the mutant embryos showed abnormal development, such as agenesis of the branchial arch-related great vessels and dorsal aorta and transposition of the aortic arch. For example, the most frequent variant was the absence of the left 4th branchial arch artery. The development of heart outflow tracts was also disturbed and separation of the truncus arteriosus was incomplete (persistent truncus arteriosus).

On the other hand, two reports on NRP2 knockouts did not report any abnormal vascular phenotype.^63,64 Unlike the NRP1 knockouts, NRP2 mutant mice were viable into adulthood. NRP2 was required for the organization and fasciculation of cranial nerves and spinal nerves and for Sema3F activity, but possible effects on the cardiovascular system were not described.

Double knockouts in which both NRP1 and NRP2 were targeted (NRP1^//NRP2^/) have also been generated.⁶⁵ These mice died in utero at E8. Their yolk sacs showed an absence of branching arteries and veins, the absence of a capillary bed and the presence of large avascular spaces between the blood vessels. The embryos had large avascular regions in the head and trunk, and blood vessel sprouts that were not connected. These double NRP1/NRP2 knockout mice had an even more severely abnormal vascular phenotype than either NRP1 or NRP2 single knockouts. Their abnormal vascular phenotype resembled those of VEGF and VEGFR2 knockouts. These results suggest that NRPs are early genes in embryonic vessel development and that both NRP1 and NRP2 are involved in normal blood vessel development.

NRP1 knockout embryos have been used to analyze NRP1-dependent vascular function in vitro as well as in vivo. Cultured wild-type paraaortic splanchnopleural mesoderm (PSp) explants supported vasculogenesis and angiogenesis whereas PSp explants derived from NRP1^/ mice had defects in capillary sprouting in vitro, consistent with the impaired vascular sprouting demonstrated in vivo in the CNS and cardiovascular system.⁶⁶ A soluble NRP1 (sNRP1), corresponding to the a1a2, b1b2 and c extracellular domains of NRP1, inhibited capillary sprouting in the cultured wild-type PSp explants. In contrast, an sNRP1 dimer produced by fusion with the Fc part of human IgG, enhanced vascular development in wild-type explants and rescued the defective vascular phenotype of mutant NRP1^/ explants. Furthermore, sNRPFc dimer, when injected into pregnant mice, reversed and rescued the NRP1^/ embryo phenotype. sNRP1 monomers have been shown to bind VEGF₁₆₅ and inhibit VEGF mitogenic activity for EC.¹³ Whereas an sNRP1 monomer appears to sequester VEGF₁₆₅ and inhibit its activity, sNRP1 dimer appears to deliver VEGF₁₆₅ to EC VEGFR2, thereby promoting angiogenesis and vasculogenesis.

Semaphorins were first described as mediators of neuronal guidance acting via NRPs^10,25 but they may also be mediators of EC activity. Sema3A binds to aortic EC and inhibits the motility of EC only if they express NRP1.⁴⁸ Sema3A also inhibits the capillary sprouting of EC from rat aortic ring segments in an in vitro angiogenesis assay. VEGF₁₆₅ and Sema3A are competitive inhibitors in EC motility, ligand binding and dorsal root ganglia collapse assays, suggesting possible overlapping binding sites.⁴⁸ VEGF₁₆₅ and Sema3A are also antagonists in neuronal survival/apoptosis assays.⁶⁷ VEGF₁₆₅ interacts with neuronal NRP1 and is a survival factor for neurons, such as hippocampal neurons and motor neurons,^68–70 whereas Sema 3A induces neuronal apoptosis.^67,71 These results suggest that a balance of semaphorins and VEGF₁₆₅ can modulate the migration, apoptosis/survival and proliferation of neurons and EC through shared receptors.

Tumor Cell Neuropilin

Many tumor cell types express NRP1 and NRP2 and bind VEGF₁₆₅. The first report was that of VEGF₁₆₅ binding to PC3 prostate and MDAMB231 breast carcinoma cells.¹² VEGF₁₆₅ binds to NRP1 in these tumor cell types with a Kd of approximately 2 × 10¹⁰ M, with about 12 × 10⁵ receptors per cell. NRPs are the only VEGF receptors expressed by these tumor cells so that any VEGF₁₆₅ activity for tumor cells is mediated by NRPs. Subsequently a number of tumor cell types have been shown to express NRP1 and/or NRP2 in vitro and in vivo. These include prostate carcinoma,⁷² melanoma,⁷³ astrocytoma,⁷⁴ osteosarcoma⁷⁵ and rat pituitary tumors.⁷⁶ In several clinical studies NRP1 and NRP2 expression was correlated with increased aggressiveness, malignancy or hypervascularity. For example, NRP1 was upregulated in primary sporadic prostate tumors at different clinical stages as determined by quantitative RT-PCR.⁷² The correlation between NRP1 overexpression with advanced disease and a high Gleason grade (a morphological measure of prostate cancer progression) suggested that NRP1 overexpression might be a marker of aggressiveness. The expression pattern of NRP1 and VEGF by human astrocytoma cell lines and specimens was closely correlated and associated with malignant astrocytomas.⁷⁴ Osteosarcoma, a malignant bone tumor characterized by hypervascularity, expressed NRP2 (24 out of 30 specimens) and the NRP2-positive tumors showed both a significantly increased vascularity and a significantly poorer prognosis than those without NRP2.⁷⁵

Besides these correlational studies, the function of NRP1 in tumor cells had been analyzed more directly. NRP1 was overexpressed in Dunning rat prostate carcinoma AT2.1 cells using a tetracycline-inducible promoter.⁷⁷ Concomitant with increased NRP1 expression in response to a tetracycline homologue, doxycycline (Dox), AT2.1 cell migration was enhanced, and VEGF₁₆₅ binding was increased 3 to 4 fold in vitro. However, induction of NRP1 did not affect tumor cell proliferation. When rats injected with AT2.1/NRP1 tumor cells were fed Dox, NRP1 synthesis was induced in vivo, and tumor size was increased 2.5 to 7-fold, in a three- to four-week period, compared to control. The larger tumors with induced NRP1 expression were characterized by markedly increased microvessel density, increased proliferating EC, dilated blood vessels and notably less tumor cell apoptosis compared to noninduced controls. It was concluded that NRP1 expression results in enlarged tumors associated with substantially enhanced tumor angiogenesis.

On the other hand, sNRP1 is a tumor antagonist.¹³ Tumors of rat prostate carcinoma cells overexpressing recombinant sNRP1 in vivo were characterized by extensive hemorrhage, damaged vessels and apoptotic tumor cells. Since sNRP1 inhibits ¹²⁵I-VEGF₁₆₅ binding to EC and VEGF₁₆₅-induced tyrosine phosphorylation of VEGFR2 in EC in vitro, this tumor phenotype may be due to VEGF₁₆₅ withdrawal and lack of bioavailability. Withdrawal of VEGF₁₆₅ from tumors using a Tet-off system has previously been shown to result in vascular damage, EC apoptosis, hemorrhage and extensive tumor necrosis.⁷⁸

NRPs may also be involved in tumor cell survival.⁷⁹ Suppression of VEGF expression in metastatic breast carcinoma MDA-MB-231 cells in vitro induced apoptosis. These effects were probably NRP-dependent since NRP1 and NRP2 are the only VEGF receptors expressed in these cells. Furthermore, VEGF₁₆₅ enhanced breast carcinoma cell survival but VEGF₁₂₁, the isoform which lacks the ability to bind to NRP1 did not, implicating a role for NRP1 in tumor cell survival.

Expression of several class 3 semaphorins has been studied extensively in lung cancer. The progression of small cell lung cancer correlates with a deletion in 3p21, and a loss of semaphorin expression; in particular, Sema3B and Sema3F.^80–82 In non-small cell lung carcinomas, low levels of Sema3F expression correlated with higher stages of disease.⁸³ Sema3B and Sema3F were transfected into lung cancer NCIH1299 cells, which do not express either gene.⁸⁴ Colony formation of H1299 cells was reduced by 90% after transfection with wild-type Sema3B as compared with the control vector. A 30-40% reduction in colony formation was seen after the transfection of Sema3F or Sema3B variants carrying single amino acid missense mutations that had been associated with lung cancer. H1299 cells transfected with wild-type, but not mutant Sema3B, underwent apoptosis. Lung cancers (n = 34) always expressed NRP1, and most of these expressed NRP2. In a very recent report, ovarian tumor cells overexpressing Sema3B exhibited diminished tumorigenicity in mice. Taken together, these results suggest that Sema3B and Sema3A are functional tumor suppressor genes.⁸⁵

On the other hand, Sema3C mRNA is overexpressed severalfold in metastatic lung tumors as determined by differential display and Northern blot analysis of lung tumor cell lines.⁸⁶ Thus, class 3 semaphorins are involved in tumor progression and metastasis, both as inhibitors and promoters.

Vascular Injury

NRPs are induced following injury in several model systems; for example, cerebral artery occlusion, cerebral ischemia, hind limb ischemia and retinal vascularization. A recurring pattern is that NRPs are highly expressed in the developing embryo as compared with the normal adult, but are induced following injury or ischemia. Several diseases characterized by increased angiogenesis, such as diabetic retinopathy and rheumatoid arthritis, show NRP1 upregulation. The first demonstration that NRP1 is induced following injury was in regenerating Xenopus optic nerves.⁸⁷ In embryos NRP1 was expressed in retinal ganglion cells, maximal at stages 41-43, and then decreased as the tadpole developed. After stage 50 NRP1 expression was almost nil. When the tadpole optic nerves were crushed and prompted to regenerate, however, NRP protein reappeared in the optic nerve fibers, being maximal at the second and third week after the optic nerve crush, and then declined thereafter.

Ischemia upregulates NRP expression. In the adult mouse, ischemic brain NRP1 mRNA expression was significantly upregulated as early as two hours and persisted at least 28 days after focal cerebral ischemia.⁸⁸ Acute upregulation of NRP1 mRNA was primarily localized to the ischemic neurons but there was also a marked increase in NRP1 expression in EC of cerebral blood vessels at the border and in the core of the ischemic lesion seven days after ischemia. NRP1 expression persisted on these vessels for at least 28 days after ischemia. Activated astrocytes also exhibited NRP1 immunoreactivity during 7 to 28 days of ischemia. Double immunofluorescent staining showed colocalization of NRP1 and VEGF to cerebral blood vessels and activated astrocytes. These results suggest that in addition to its role in axonal growth, upregulation of NRP1 may contribute to neovascular formation in the adult ischemic brain.

In another mouse ischemia model system very little NRP2 expression was observed in normal blood vessels after birth (Takashima et. al, unpublished). However, it was possible to induce NRP2 expression in blood vessels in response to ischemia in a hind limb model in which occlusion of the femoral artery by ligation resulted in the sprouting of new vessels (Fig. 1). Prior to injury or in a sham operation without ligation, NRP2 was not expressed in the femoral artery. However, after one week, NRP2 expression was clearly seen in the sprouting vessels at the edge of the ligated artery. By two weeks NRP2 expression was more prominent and was detected in EC in the vascular wall of newly developed mid-sized arteries. These expression profiles indicate that NRP2 is expressed primarily in the embryo and extraembryonic tissue, whereas expression in the adult is atypical and occurs only when induced, for example, by ischemia.

Figure 1

NRP2LacZ expression in an 8-10-week adult ischemic hindlimb. Ischemia was induced by ligation of the femoral artery as previously described. Left: Two weeks after a sham operation without ligation of the femoral artery, there was very little if any NRP2 (more...)

NRP1 expression is also induced in retinal neovascularization. A model of retinopathy of prematurity (ROP) was produced by ischemia induced ocular neovascularization. Postnatal day 7 mice were exposed to 75% oxygen for five days and then returned to room air for five days.⁸⁹ Retinal neovascularization was visualized by injection of fluorescein-dextran. Expression of NRP1 and VEGFR2 mRNAs was colocalized in the area of neovascularization. In addition, expression of VEGFR2 and NRP1 was restricted to neovascularized vessels of the retina from ROP mice. The restricted expression of VEGFR2 and NRP1 on neovascularized vessels suggests that these molecules may play important roles in retinal neovascularization.

In a clinical study, fibrovascular tissues were obtained at vitrectomy from 22 cases with proliferative diabetic retinopathy.⁹⁰ RT-PCR analysis demonstrated the expression of VEGF receptors VEGFR1, VEGFR2 and NRP1 in 12, 14 and 14 of 22 cases, respectively. Notably, VEGFR2 and NRP1 were simultaneously expressed in the identical 14 tissues. The vascular density of fibrovascular tissues as determined by immunohistochemistry for CD34, an EC marker, was significantly higher in cases with the expression of VEGFR2 and NRP1 versus those without their expression. VEGFR1 expression had no such relationship with the vascular density. It was concluded that coexpression of VEGFR2 and NRP1 may facilitate fibrovascular proliferation in diabetic retinopathy.

Perspectives and Future Directions

There is substantial evidence, based on cell culture and transgenic mouse studies which indicates that NRPs are novel and significant regulators of blood vessel development. In embryonic development NRP expression occurs early, in the blood islands of the yolk sac. NRP expression is required for the normal branching and organization of large vessels and capillaries in the developing yolk sac and embryo. In the developing heart NRP1 is required for normal formation of the large arteries. In the adult, NRP expression is generally reduced but it is strongly upregulated in blood vessels in response to vascular injury. How NRP expression in EC affects cellular function is not clear. One can speculate that the ability of NRPs to bind VEGF and class 3 semaphorins to EC must play a role and that in the absence of NRPs, VEGF and semaphorin activity is compromised. Since both VEGF and Sema3A are involved in forming networks of blood vessels and neurons, respectively, these processes might be adversely affected by a lack of NRP expression.

There are two NRPs, NRP1 and NRP2, and they probably are responsible for some nonoverlapping functions (Table 1). For example, the abnormal vascular phenotype is much more severe in NRP1-deficient mice than in NRP2-deficient mice. During development, NRP1 is expressed by arteries and NRP2 by veins. There are some differences in NRP1 and NRP2 ligand interactions. For example, NRP1 is activated by Sema3A and NRP2 by Sema3F. PlGF2 binds NRP1 but not NRP2, whereas VEGF₁₄₅ binds NRP2 but not NRP1. Thus, arteries and veins may bind different ligands and thereby may be subjected to different signals. Differential interactions of NRP1 and NRP2 with VEGFR1 and/or VEGFR2 might also contribute to different signaling pathways in arteries and veins.

Tumor cells are among the highest expressers of NRPs and as a consequence bind VEGF, typically in the absence of other VEGF receptors. The significance of direct VEGF binding to tumor cells is unknown but might involve enhancement of tumor cell migration and survival. NRP1 overexpression in tumor cells enhances tumor angiogenesis whereas sNRP1 suppresses it. The reason for this might be VEGF bioavailability. Full-length NRP is membrane anchored and might be expected to concentrate VEGF on the cell surface, thereby signaling neighboring EC or the tumor cells themselves. On the other hand, sNRPs are soluble and may sequester VEGF away from the cell surface, thereby inducing cell apoptosis. Tumor cells express both full-length NRP and sNRP, thus, a balance of these two types of NRP might contribute to the level of tumor angiogenesis.

Future Directions include: 1) determination of the mechanisms by which NRPs regulate angiogenesis, for example, whether NRP is involved in ligand signaling, directly or as coreceptors for VEGF RTKs, Plexins or other receptors; 2) identification of upstream regulators of NRP1 function. So far TNFα, and the transcription factors dHAND and Ets1 have been implicated; 3) identification of target genes and proteins downstream of NRP; 4) determination of novel VEGF and semaphorin biological functions given that many different cell types besides neuronal cells and EC express NRP and bind these two families of ligands; and 5) determination of whether there are any NRP clinical applications. Possible tumor antagonists include sNRP which induces tumor cell apoptosis, Sema3A, which blocks in vitro angiogenesis, and Sema3B and Sema3F which may have tumor suppressor activity.

Acknowledgments

This article was supported by NIH grants CA37392 and CA44548 (MK) and a grant from the Erenst Schering Research Foundation in Berlin, Germany (RM). We thank Alexandra Grady for preparation of the manuscript.

References

1.

Fujisawa H, Kitsukawa T. Receptors for collapsin/semaphorins. Curr Opin Neurobiol. 1998;8:587–592. [PubMed: 9811625]

2.

Neufeld G, Cohen T, Gengrinovitch S. et al. Vascular endothelial growth factor (VEGF) and its receptors. FASEB J. 1999;13:9–22. [PubMed: 9872925]

3.

Shima D, Mailhos C. Vascular development biology: getting nervous. Curr Opin Genet Dev. 2000;10:536–542. [PubMed: 10980432]

4.

Raper JA. Semaphorins and their receptors in vertebrates and invertebrates. Curr Opin Neurobiol. 2000;10:88–94. [PubMed: 10679438]

5.

Miao HQ, Klagsbrun M. Neuropilin is a mediator of angiogenesis. Cancer Metastasis Rev. 2000;19:29–37. [PubMed: 11191060]

6.

Soker S. Neuropilin in the midst of cell migration and retraction. Int J Biochem Cell Biol. 2001;33:433–437. [PubMed: 11312112]

7.

Takagi S, Kasuya Y, Shimizu M. et al. Expression of a cell adhesion molecule, neuropilin, in the developing chick nervous system. Dev Biol. 1995;170:207–222. [PubMed: 7601310]

8.

Kawakami A, Kitsukawa T, Takagi S. et al. Developmentally regulated expression of a cell surface protein, neuropilin, in the mouse nervous system. J Neurobiol. 1996;29:1–17. [PubMed: 8748368]

9.

Fujisawa H, Kitsukawa T, Kawakami A. et al. Roles of a neuronal cellsurface molecule, neuropilin, in nerve fiber fasciculation and guidance. Cell Tissue Res. 1997;290:465–470. [PubMed: 9321711]

10.

Kolodkin AL, Levengood DV, Rowe EG. et al. Neuropilin is a semaphorin III receptor. Cell. 1997;90:753–762. [PubMed: 9288754]

11.

Kitsukawa T, Shimono A, Kawakami A. et al. Overexpression of a membrane protein, neuropilin, in chimeric mice causes anomalies in the cardiovascular system, nervous system and limbs. Development. 1995;121:4309–4318. [PubMed: 8575331]

12.

Soker S, Takashima S, Miao HQ. et al. Neuropilin-1 is expressed by endothelial and tumor cells as an isoformspecific receptor for vascular endothelial growth factor. Cell. 1998;92:735–745. [PubMed: 9529250]

13.

Gagnon ML, Bielenberg DR, Gechtman Z. et al. Identification of a natural soluble neuropilin1 that binds vascular endothelial growth factor: In vivo expression and antitumor activity. Proc Natl Acad Sci USA. 2000;97:2573–2578. [PMC free article: PMC15970] [PubMed: 10688880]

14.

Rossignol M, Gagnon ML, Klagsbrun M. Genomic organization of human neuropilin1 and neuropilin2 genes: identification and distribution of splice variants and soluble isoforms. Genomics. 2000;70:211–222. [PubMed: 11112349]

15.

Pavelock K, Braas K, Ouafik L. et al. Differential expression and regulation of the vascular endothelial growth factor receptors neuropilin 1 and neuropilin 2 in rat uterus. Endocrinology. 2001;142:613–622. [PubMed: 11159832]

16.

Eickholt BJ, Mackenzie SL, Graham A. et al. Evidence for collapsin 1 functioning in the control of neural crest migration in both trunk and hindbrain regions. Development. 1999;126:2181–2189. [PubMed: 10207143]

17.

Deckers MM, Karperien M, van der Bent C. et al. Expression of vascular endothelial growth factors and their receptors during osteoblast differentiation. Endocrinology. 2000;141:1667–1674. [PubMed: 10803575]

18.

Harper J, Gerstenfeld LC, Klagsbrun M. Neuropilin 1 expression in osteogenic cells: down-regulation during differentiation of osteoblasts into osteocytes. J Cell Biochem. 2001;81:82–92. [PubMed: 11180399]

19.

Tordjman R, Ortega N, Coulombel L. et al. Neuropilin 1 is expressed on bone marrow stromal cells: a novel interaction with hematopoietic cells? Blood. 1999;94:2301–2309. [PubMed: 10498602]

20.

Thomas S, Vanuystel J, Gruden G. et al. Vascular endothelial growth factor receptors in human mesangium in vitro and in glomerular disease. J Am Soc Nephrol. 2000;11:1236–1243. [PubMed: 10864579]

21.

Cohen T, Gluzman-Poltorak Z, Brodzky A. et al. Neuroendocrine cells along the digestive tract express neuropilin2. Biochem Biophys Res Commun. 2001;284:395–403. [PubMed: 11394892]

22.

Harper SJ, Xing CY, Whittle C. et al. Expression of neuropilin 1 by human glomerular epithelial cells in vitro and in vivo. Clin Sci. 2001;101:439–446. [PubMed: 11566082]

23.

Cai H, Reed RR. Cloning and characterization of neuropilin1interacting protein: a PSD95/Dlg/ZO1 domaincontaining protein that interacts with the cytoplasmic domain of neuropilin1. J Neurosci. 1999;19:6519–6527. [PMC free article: PMC6782790] [PubMed: 10414980]

24.

Rossignol M, Beggs AH, Pierce EA. et al. Human neuropilin 1 and neuropilin 2 map to 10p12 and 2q34, respectively. Genomics. 1999;57:459–460. [PubMed: 10329017]

25.

He Z, TessierLavigne M. Neuropilin is a receptor for the axonal chemorepellent Semaphorin III. Cell. 1997;90:739–751. [PubMed: 9288753]

26.

Chen H, Chedotal A, He Z. et al. Neuropilin2, a novel member of the neuropilin family, is a high affinity receptor for the semaphorins Sema E and Sema IV but not Sema III. Neuron. 1997;19:547–559. [PubMed: 9331348]

27.

Chen H, He Z, Bagri A. et al. Semaphorinneuropilin interactions underlying sympathetic axon responses to class III semaphorins. Neuron. 1998;21:1283–1290. [PubMed: 9883722]

28.

Takahashi T, Nakamura F, Jin Z. et al. Semaphorins A and E act as antagonists of neuropilin1 and agonists of neuropilin2 receptors. Nat Neurosci. 1998;1:487–493. [PubMed: 10196546]

29.

Giger RJ, Urquhart ER, Gillespie SK. et al. Neuropilin 2 is a receptor for semaphorin IV: insight into the structural basis of receptor function and specificity. Neuron. 1998;21:1079–1092. [PubMed: 9856463]

30.

Luo Y, Raible D, Raper JA. Collapsin: a protein in brain that induces the collapse and paralysis of neuronal growth cones. Cell. 1993;75:217–227. [PubMed: 8402908]

31.

Marin O, Yaron A, Bargri A. et al. Sorting of striatal and cortical interneurons regulated by semaphorin-neuropilin interactions. Science. 2001;293:872–875. [PubMed: 11486090]

32.

Winberg ML, Noordermeer JN, Tamagnone L. et al. Plexin A is a neuronal semaphorin receptor that controls axon guidance. Cell. 1998;95:903–916. [PubMed: 9875845]

33.

Tamagnone L, Artigiani S, Chen H. et al. Plexins are a large family of receptors for transmembrane, secreted, and GPI-anchored semaphorins in vertebrates. Cell. 1999;99:71–80. [PubMed: 10520995]

34.

Tamagnone L, Comoglio PM. Signaling by semaphorin receptors: cell guidance and beyond. Trends Cell Biol. 2000;10:377–383. [PubMed: 10932095]

35.

Castellani V, Chedotal A, Schachne M. et al. Analysis of the L1-deficient mouse phenotype reveals crosstalk between Sema3A and L1 signaling pathways in axonal guidance. Neuron. 2000;27:237–249. [PubMed: 10985345]

36.

Ferrara N. Molecular and biological properties of vascular endothelial growth factor. J Mol Med. 1999;77:527–543. [PubMed: 10494799]

37.

Dvorak HF, Nagy JA, Feng D. et al. Vascular permeability factor/vascular endothelial growth factor and the significance of microvascular hyperpermeability in angiogenesis. Curr Top Microbiol Immunol. 1999;237:97–132. [PubMed: 9893348]

38.

Soker S, Fidder H, Neufeld G. et al. Characterization of novel vascular endothelial growth factor (VEGF) receptors on tumor cells that bind VEGF165 via its exon 7-encoded domain. J Biol Chem. 1996;271:5761–5767. [PubMed: 8621443]

39.

Gluzman-Poltorak Z, Cohen T, Herzog Y. et al. Neuropilin-2 is a receptor for the vascular endothelial growth factor (VEGF) forms VEGF145 and VEGF165. J Biol Chem. 2000;275:29922. [PubMed: 10748121]

40.

Migdal M, Huppertz B, Tessler S. et al. Neuropilin-1 is a placenta growth factor-2 receptor. J Biol Chem. 1998;273:22272–22278. [PubMed: 9712843]

41.

Makinen T, Olofsson B, Karpanen T. et al. Differential binding of vascular endothelial growth factor B splice and proteolytic isoforms to neuropilin-1. J Biol Chem. 1999;274:21217–21222. [PubMed: 10409677]

42.

Wise LM, Veikkola T, Mercer AA. et al. Vascular endothelial growth factor (VEGF)like protein from orf virus NZ2 binds to VEGFR2 and neuropilin-1. Proc Natl Acad Sci USA. 1999;96:3071–3076. [PMC free article: PMC15896] [PubMed: 10077638]

43.

Savory LJ, Stacker SA, Fleming SB. et al. Viral vascular endothelial growth factor plays a critical role in orf virus infection. J Virol. 2000;74:10699–10706. [PMC free article: PMC110944] [PubMed: 11044114]

44.

Karkkainen MJ, Saaristo A, Jussila L. et al. A model for gene therapy of human hereditary lymphedema. Proc Natl Acad Sci USA. 2001;98:12677–12682. [PMC free article: PMC60113] [PubMed: 11592985]

45.

Fuh G, Garcia KC, de Vos AM. The interaction of neuropilin-1 with vascular endothelial growth factor and its receptor flt1. J Biol Chem. 2000;275:26690–26695. [PubMed: 10842181]

46.

Whitaker GB, Limberg BJ, Rosenbaum JS. Vascular endothelial growth factor receptor-2 and neuropilin-1 form a receptor complex that is responsible for the differential signaling potency of VEGF(165) and VEGF(121). J Biol Chem. 2001;276:25520–25531. [PubMed: 11333271]

47.

Gluzman-Poltorak Z, Cohen T, Shibuya M. et al. Vascular endothelial growth factor receptor-1 and neuropilin-2 form complexes. J Biol Chem. 2001;276:18688–18694. [PubMed: 11278319]

48.

Miao HQ, Soker S, Feiner L. et al. Neuropilin-1 mediates collapsin-1/semaphorin III inhibition of endothelial cell motility: functional competition of collapsin-1 and vascular endothelial growth factor-165. J Cell Biol. 1999;146:233–242. [PMC free article: PMC2199727] [PubMed: 10402473]

49.

Partanen TA, Makinen T, Arola J. et al. Endothelial growth factor receptors in human fetal heart. Circulation. 1999;100:583–586. [PubMed: 10441093]

50.

Robert B, Zhao X, Abrahamson DR. Coexpression of neuropilin-1, Flk1, and VEGF(164) in developing and mature mouse kidney glomeruli. Am J Physiol Renal Physiol. 2000;279:F275–F282. [PubMed: 10919846]

51.

Ishida A, Murray J, Saito Y. et al. Expression of vascular endothelial growth factor receptors in smooth muscle cells. J Cell Physiol. 2001;188:359–368. [PubMed: 11473363]

52.

Adams RH, Wilkinson GA, Weiss C. et al. Roles of ephrin B ligands and EphB recceptors in cardiovascular development: demarcation of arterial/venous domains, vascular morphogenesis, and sprouting angiogenesis. Genes Dev. 1999;13:295–306. [PMC free article: PMC316426] [PubMed: 9990854]

53.

Moyon D, Pardanaud L, Yuan L. et al. Plasticity of endothelial cells during arterialvenous differentiation in the avian embryo. Development. 2001;128:3359–3370. [PubMed: 11546752]

54.

Herzog Y, Kalcheim C, Kahane N. et al. Differential expression of neuropilin-1 and neuropilin-2 in arteries and veins. Mech Dev. 2001;109:115–119. [PubMed: 11677062]

55.

Stalmans I, Ng YS, Rohan R. et al. Arteriolar and venular patterning in retinas of mice selectively expressing VEGF isoforms. J Clin Invest. 2002;109:327–336. [PMC free article: PMC150858] [PubMed: 11827992]

56.

Giraudo E, Primo L, Audero E. et al. Tumor necrosis factor alpha regulates expression of vascular endothelial growth factor receptor2 and of its coreceptor neuropilin1 in human vascular endothelial cells. J Biol Chem. 1998;273:22128–22135. [PubMed: 9705358]

57.

Yamagishi H, Olson EN, Srivastava D. The basic helix-loop-helix transcription factor, dHAND, is required for vascular development. J Clin Invest. 2000;105:261–270. [PMC free article: PMC377450] [PubMed: 10675351]

58.

Sato Y, Abe M, Tanaka K. et al. Signal transduction and transcriptional regulation of angiogenesis. Adv Exp Med Biol. 2000;476:109–115. [PubMed: 10949659]

59.

Teruyama K, Abe M, Nakano T. et al. Neurophilin1 is a downstream target of transcription factor Ets1 in human umbilical vein endothelial cells. FEBS Lett. 2001;504:1–4. [PubMed: 11522285]

60.

Wernert N, Raes MB, Lassalle P. et al. cets1 protooncogene is a transcription factor expressed in endothelial cells during tumor vascularization and other forms of angiogenesis in humans. Am J Pathol. 1992;140:119–127. [PMC free article: PMC1886238] [PubMed: 1370594]

61.

Kitsukawa T, Shimizu M, Sanbo M. et al. Neuropilin-semaphorin III/D-mediated chemorepulsive signals play a crucial role in peripheral nerve projection in mice. Neuron. 1997;19:995–1005. [PubMed: 9390514]

62.

Kawasaki T, Kitsukawa T, Bekku Y. et al. A requirement for neuropilin1 in embryonic vessel formation. Development. 1999;126:4895–4902. [PubMed: 10518505]

63.

Giger RJ, Cloutier JF, Sahay A. et al. Neuropilin-2 is required in vivo for selective axon guidance responses to secreted semaphorins. Neuron. 2000;25:29–41. [PubMed: 10707970]

64.

Chen H, Bagri A, Zupicich JA. et al. Neuropilin-2 regulates the development of selective cranial and sensory nerves and hippocampal mossy fiber projections. Neuron. 2000;25:43–56. [PubMed: 10707971]

65.

Takashima S, Kitakaze M, Asakura M. et al. Targeting of both mouse neuropilin-1 and neuropilin-2 genes severely impairs developmental yolk sac and embryonic angiogenesis. Proc Natl Acad Sci USA. 2002;99:3657–3662. [PMC free article: PMC122579] [PubMed: 11891274]

66.

Yamada Y, Takakura N, Yasue H. et al. Exogenous clustered neuropilin 1 enhances vasculogenesis and angiogenesis. Blood. 2001;97:1671–1678. [PubMed: 11238106]

67.

Bagnard D, Vaillant C, Khuth ST. et al. Semaphorin 3A vascular endothelial growth factor165 balance mediates migration and apoptosis of neural progenitor cells by the recruitment of shared receptor. J Neurosci. 2001;21:3332–3341. [PMC free article: PMC6762465] [PubMed: 11331362]

68.

Sondell M, Sundler F, Kanje M. Vascular endothelial growth factor is a neurotrophic factor which stimulates axonal outgrowth through the flk1 receptor. Eur J Neurosci. 2000;12:4243–4254. [PubMed: 11122336]

69.

Jin KL, Mao XO, Greenberg DA. Vascular endothelial growth factor: direct neuroprotective effect in in vitro ischemia. Proc Natl Acad Sci USA. 2000;97:10242–10247. [PMC free article: PMC27841] [PubMed: 10963684]

70.

Oosthuyse B, Moons L, Storkebaum E. et al. Deletion of the hypoxiaresponse element in the vascular endothelial growth factor promoter causes motor neuron degeneration. Nat Genet. 2001;28:131–138. [PubMed: 11381259]

71.

Shirvan A, Shina R, Ziv I. et al. Induction of neuronal apoptosis by Semaphorin3Aderived peptide. Brain Res Mol Brain Res. 2000;83:81–93. [PubMed: 11072098]

72.

Latil A, Bieche I, Pesche S. et al. VEGF overexpression in clinically localized prostate tumors and neuropilin1 overexpression in metastatic forms. Int J Cancer. 2000;89:167–171. [PubMed: 10754495]

73.

Lacal PM, Failla CM, Pagani E. et al. Human melanoma cells secrete and respond to placenta growth factor and vascular endothelial growth factor. J Invest Dermatol. 2000;115:1000–1007. [PubMed: 11121133]

74.

Ding H, Wu X, Roncari L. et al. Expression and regulation of neuropilin1 in human astrocytomas. Int J Cancer. 2000;88:584–592. [PubMed: 11058875]

75.

Handa A, Tokunaga T, Tsuchida T. et al. Neuropilin-2 expression affects the increased vascularization and is a prognostic factor in osteosarcoma. Int J Oncol. 2000;17:291–295. [PubMed: 10891538]

76.

Banerjee SK, Zoubine MN, Tran TM. et al. Overexpression of vascular endothelial growth factor164 and its coreceptor neuropilin-1 in estrogen-induced rat pituitary tumors and GH3 rat pituitary tumor cells. Int J Oncol. 2000;16:253–260. [PubMed: 10639567]

77.

Miao HQ, Lee P, Lin H. et al. Neuropilin-1 expression by tumor cells promotes tumor angiogenesis and progression. FASEB J. 2000;14:2532–9. [PubMed: 11099472]

78.

Benjamin LE, Keshet E. Conditional switching of vascular endothelial growth factor (VEGF) expression in tumors: induction of endothelial cell shedding and regression of hemangioblastoma-like vessels by VEGF withdrawal. Proc Natl Acad Sci USA. 1997;94:8761–8766. [PMC free article: PMC23118] [PubMed: 9238051]

79.

Bachelder R, Crago A, Chung J. et al. Vascular endothelial growth factor is an autocrine survival factor for neuropilinexpressing breast carcinoma cells. Cancer Res. 2001;61:5736–5740. [PubMed: 11479209]

80.

Xiang RH, Hensel CH, Garcia DK. et al. Isolation of the human semaphorin III/F gene (SEMA3F) at chromosome 3p21, a region deleted in lung cancer. Genomics. 1996;32:39–48. [PubMed: 8786119]

81.

Roche J, Boldog F, Robinson M. et al. Distinct 3p21.3 deletions in lung cancer and identification of a new human semaphorin. Oncogene. 1996;12:1289–1297. [PubMed: 8649831]

82.

Sekido Y, Bader S, Latif F. et al. Human semaphorins A(V) and IV reside in the 3p21.3 small cell lung cancer deletion region and demonstrate distinct expression patterns. Proc Natl Acad Sci USA. 1996;93:4120–4125. [PMC free article: PMC39497] [PubMed: 8633026]

83.

Brambilla E, Constantin B, Drabkin H. et al. Semaphorin SEMA3F localization in malignant human lung and cell lines: A suggested role in cell adhesion and cell migration. Am J Pathol. 2000;156:939–950. [PMC free article: PMC1876858] [PubMed: 10702410]

84.

Tomizawa Y, Sekido Y, Kondo M. et al. Inhibition of lung cancer cell growth and induction of apoptosis after reexpression of 3p21.3 candidate tumor suppressor gene SEMA3B. Proc Natl Acad Sci USA. 2001;98:13954–13959. [PMC free article: PMC61148] [PubMed: 11717452]

85.

Tse C, Xiang RH, Bracht T. et al. Human semaphorin 3B (SEMA3B) located at chromosome 3p21.3 suppresses tumor formation in an adenocarcinoma cell line. Cancer Res. 2002;62:542–546. [PubMed: 11809707]

86.

Martin-Satue M, Blanco J. Identification of semaphorin E gene expression in metastatic human lung adenocarcinoma cells by mRNA differential display. J Surg Oncol. 1999;72:18–23. [PubMed: 10477871]

87.

Fujisawa H, Takagi S, Hirata T. Growth-associated expression of a membrane protein, neuropilin, in Xenopus optic nerve fibers. Dev Neurosci. 1995;17:343–349. [PubMed: 8829923]

88.

Zhang ZG, Tsang W, Zhang L. et al. Upregulation of neuropilin1 in neovasculature after focal cerebral ischemia in the adult rat. J Cereb Blood Flow Metab. 2001;21:541–549. [PubMed: 11333364]

89.

Ishihama H, Ohbayashi M, Kurosawa N. et al. Colocalization of neuropilin1 and Flk1 in retinal neovascularization in a mouse model of retinopathy. Invest Ophthalmol Vis Sci. 2001;42:1172–1178. [PubMed: 11328724]

90.

Ishida S, Shinoda K, Kawashima S. et al. Coexpression of VEGF receptors VEGFR2 and neuropilin1 in proliferative diabetic retinopathy. Invest Ophthalmol Vis Sci. 2000;41:1649–1656. [PubMed: 10845581]

91.

Murohara T, Asahara T, Silver M. et al. Nitric oxide synthase modulates angiogenesis in response to tissue ischemia. J Clin Invest. 1998;101:2567–2578. [PMC free article: PMC508846] [PubMed: 9616228]