MINIREVIEW Brain angiogenesis in developmental and pathological processes: therapeutic aspects of vascular endothelial growth factor Masabumi Shibuya 1,2 1 Department of Molecular Oncology, Tokyo Medical and Dental University, Japan 2 Jobu University, Takasaki, Japan Introduction The central nervous system (CNS) is a complex of well-vascularized tissues through which oxygen and nutrition are supplied to the brain via the carotid artery. Actually, cells such as neurons and glial cells in the CNS require a fresh supply of blood to function. In embryogenesis, the formation of primitive blood vessels from progenitors, hemangioblasts ⁄ angioblasts, is dependent on the vascular endothelial growth fac- tor ⁄ vascular endothelial growth factor receptor (VEGF ⁄ VEGFR) system [1,2], and the further devel- opment of blood vessels in various tissues and organs, including the brain, is regulated by the VEGF system in combination with other signaling systems such as the angiopoietin–Tie, ephrin–Eph, Delta–Notch sys- tems, and the Wnt pathway. Furthermore, the blood vessel network in the CNS has a unique stabilizing system at the postnatal to adult stages known as the blood–brain barrier (BBB) Keywords macrophage; malignant glioma; motor neuron; tumor angiogenesis; vascular hyperpermeability; VEGF-A; VEGF-B; VEGF-E; VEGFR-1; VEGFR-2 Correspondence M. Shibuya, Department of Molecular Oncology, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8519, Japan Fax: +81 3 5803 0125 Tel: +81 3 5803 5086 E-mail: shibuya@ims.u-tokyo.ac.jp (Received 19 February 2009, revised 26 May 2009, accepted 15 June 2009) doi:10.1111/j.1742-4658.2009.07175.x The angiogenic process in the central nervous system (CNS) is basically regulated by typical angiogenic signaling systems such as vascular endothe- lial growth factor (VEGF)–VEGF receptors and angiopoietin–Tie recep- tors. In addition to regular endothelial–pericyte interaction, the CNS vasculature has a unique system of cell to cell communication between endothelial cells and astrocytes which is known as the blood–brain barrier. Among the pathological conditions of the CNS vascular network, stroke is a major disease in which the supply of blood is decreased. Pro-angiogenic therapy using natural VEGF-A has so far been unsuccessful, indicating the possible need for a new approach related to upstream or downstream regu- lators involved in the VEGF-signaling pathway, or alternate VEGF family members. By contrast, a pathological increase in the blood supply in the CNS is seen in brain tumors, in particular malignant gliomas. In phase II clinical trials, anti-VEGF therapies have been shown to suppress tumor growth and improve survival rates to some extent. However, tumor inva- sion and the distant metastasis of gliomas can occur following anti-angio- genic therapy. Further studies are needed to obtain safer clinical outcomes by developing new strategies with combination therapy using known anti- angiogenic drugs or by developing unique medicines specifically targeting the blood vessels in brain tumors. Abbreviations BBB, blood–brain barrier; CNS, central nervous system; EC, endothelial cell; FGF, fibroblast growth factor; HIF, hypoxia-inducible factor; PlGF, placenta growth factor; VEGF, vascular endothelial growth factor; VEGFR, vascular endothelial growth factor receptor; VHL, von Hippel–Lindau. 4636 FEBS Journal 276 (2009) 4636–4643 ª 2009 The Author Journal compilation ª 2009 FEBS [3]. The BBB mainly consists of a strong interaction between vascular endothelial cells and astrocytes, and the tight junctions of vascular endothelial cells (ECs) in the BBB are well organized with claudins and ZO- proteins through a decrease in angiogenesis signaling and an increase in the stability of ECs by the AKAP12 ⁄ SSeCKS ⁄ Gravin and other systems [4]. Because the VEGF–VEGFR system is central to angiogenesis in almost all stages of life, we briefly introduce it here (Fig. 1), followed by discussion of two pathological conditions of angiogenesis in the brain, stroke and brain tumors, along with possible therapeutic strategies. VEGF and VEGFRs The VEGF family The VEGF family includes VEGF-A (also called VEGF), placenta growth factor (PlGF), VEGF-B, -C, -D and -E, and Trimeresurus flavoviridis, snake-venom VEGF, although the latter two proteins are not encoded in the human genome (Table 1) [1]. VEGF-A is most important for vasculogenesis as well as angio- genesis in both embryogenesis and adulthood, and functions by binding with two tyrosine kinase recep- tors, VEGFR-1 (Flt-1) and VEGFR-2 (KDR ⁄ Flk-1), Fig. 1. The VEGF–VEGFR system and the inhibitors of various signaling steps. Five VEGF family members and three VEGFRs are encoded in the human genome. In addi- tion, VEGF-E encoded in the Orf virus binds and activates only VEGFR-2. Anti-(VEGF-A) IgG and aptamer, as well as VEGFR tyrosine kinase inhibitors (multikinase inhibitors) have been approved for the treatment of various solid tumors. Anti-(VEGF-A) IgG was effec- tive in phase II clinical trials for glioblastoma multiforme. Other inhibitors or methods to suppress VEGF–VEGFR system such as VEGF–Trap, anti-VEGFR neutralizing IgG and immunotherapy against VEGFRs are under- going clinical trials. Table 1. Activities of vascular endothelial growth factor (VEGF) family members. PlGF, placenta growth factor; T.F. svVEGF, Trimeresurus flavoviridis, snake-venom vascular endothelial growth factor. Binding activity to Biological activity Ligands VEGFR-1 VEGFR-2 VEGFR-3 Angiogenesis Permeability a VEGF-A +++ ++ ) +++ +++ PlGF +++ ))+ ⁄ ) + VEGF-B +++ ))+ ⁄ ) + VEGF-C ) + ++ (lymphangiogenesis) ) VEGF-D ) + ++ (lymphangiogenesis) ) VEGF-E ) ++ ) +++ ++ T.f. svVEGF +++ + ) + +++ a Vascular permeability activity detected by Miles assay (at acute phase: within 15 min). M. Shibuya Therapeutic aspects of VEGF in brain diseases FEBS Journal 276 (2009) 4636–4643 ª 2009 The Author Journal compilation ª 2009 FEBS 4637 and neuropilin-1 [2]. VEGF-A has two major biologi- cal roles, in angiogenesis and vascular permeability. Angiogenesis VEGF-A stimulates endothelial proliferation directly via the activation of VEGFR-2 tyrosine kinase. VEG- FR-2 tyrosine kinase has a strong kinase activity simi- lar to epidermal growth factor receptor (EGFR), but the signaling pathway towards proliferation differs from that in EGFR. VEGFR-2 activates the phospho- lipase Cc–protein kinase C–Raf–mitogen-activated protein kinase pathway via a phosphorylated tyrosine residue at position 1175 of the receptor, and stimulates EC proliferation [5]. In adults, particularly under path- ological conditions, VEGFR-1 also contributes to angiogenesis indirectly via the recruitment of mono- cyte ⁄ macrophage lineage cells which secrete various angiogenic factors [6]. In addition, VEGFR-1 expressed in ECs generates mitotic and survival sig- nals, although much less intensely than VEGFR-2. Vascular permeability VEGF-A stimulates the vascular leakage of fluids from blood vessels in both an acute and a chronic manner. Although the molecular basis of the signaling pathway for vascular hyperpermeability within the cell is not fully understood, both types of hyperpermeability depend strongly on the simultaneous activation of two receptors, VEGFR-1 and VEGFR-2. VEGF-E, a viral genome-encoded VEGF-like protein, activates only VEGFR-2, resulting in angiogenesis without severe vascular hyperpermeability or tissue edema [7,8]. By contrast, T. flavoviridis, snake-venom VEGF, encoded in the genome of the venomous ‘Habu’ snake, activates VEGFR-1 strongly and VEGFR-2 weakly, resulting in high vascular permeability with only minor angiogenic activity [9]. Expression of VEGF and VEGFRs VEGF-A, as well as VEGFR-1 and VEGFR-2, is highly expressed in brain tissue in the early to middle stages of embryogenesis, and gradually decreases at the perinatal to postnatal stages. However, expression of VEGF-A and its two receptors is upregulated in brain tumor tissue [10,11]. The VEGF–VEGFR system is generally used as a paracrine system in vivo. Neuro- nal cells, astrocytes, tumor cells and bone-marrow- derived cells secrete VEGF-A, whereas VEGFRs are specifically expressed in vascular ECs. VEGFR-1 is also expressed in macrophages. Surprisingly however, Lee et al.[12] recently reported that ECs do express a low level of VEGF-A, which contributes in part to EC survival in an autocrine manner. Neuronal cells also express a low level of VEGFR under certain condi- tions, such as post injury, as discussed later. VEGFR-3 is expressed in lymph endothelial cells. VEGF-C, as well as VEGF-D, binds and activates this receptor, resulting in the proliferation and migration of lymph ECs and lymphangiogenesis [13]. Angiogenesis in brain diseases Brain stroke Stroke is induced through: (a) the obstruction of mid- sized to large blood vessels, or (b) massive bleeding from mid-sized to large vessels in the brain. These lesions result in severe ischemia of neurons and astro- cytes around and downstream of the lesions, eventually inducing necrotic cell death. Several risk factors including aging, hypertension, diabetes and atheroscle- rosis have been described, but explain only about half of the causes of stroke, suggesting that unknown mechanisms are also involved in the onset [3]. Increased vascular density surrounding the ‘stroke’ area has been observed after stroke, and such an increase in blood flow may rescue the ischemic and still viable region of the brain called the ‘penumbra’ [14]. Therefore, degree of angiogenesis appears to correlate with rate of recovery from stroke. A variety of angiogenic factors such as VEGF, fibroblast growth factor (FGF) and platelet-derived growth factor are secreted from neuronal cells, astro- cytes and inflammatory cells, including macrophages that have infiltrated the stroke area. Zhang et al. [15] have shown that the intravenous administration of VEGF-A within 2 days after stroke induces angiogene- sis in the penumbra, and contributes to a recovery in neuron function from the ischemic events. Administration of VEGF-A into the brain after stroke may be effective for recovery. However, VEGF- A not only has pro-angiogenic activity, but also increases vascular permeability, and increases in tissue fluid and edema in the brain may be dangerous because the volume of the brain tissue is tightly limited by the cranial bone. This limitation is unique among tissues in the body, and the risk of edema to brain functions should be carefully considered. Brain edema may increase pressure in the cranial cavity and brain tissue, resulting in suppression of blood flow in the vessel network. Therefore, it is not easy to control the administration of VEGF-A, i.e. timing, dosage, dura- tion and combination with other factors ⁄ medicines. Therapeutic aspects of VEGF in brain diseases M. Shibuya 4638 FEBS Journal 276 (2009) 4636–4643 ª 2009 The Author Journal compilation ª 2009 FEBS In terms of vascular permeability, a VEGF family member, VEGF-E, appears to have the attractive char- acteristic of binding and activating only VEGFR-2, inducing a relatively strong pro-angiogenic signal. Sev- eral reports have indicated that VEGF-E, but not VEGF-A, has marked angiogenic activity without causing severe edema or an inflammatory response in vivo in transgenic mouse models as well as a hind- limb ischemia model [7,8,16]. This seems reasonable because vascular permeability is known to be induced after the simultaneous activa- tion of VEGFR-1 and VEGFR-2. Furthermore, inflammation after VEGF-A therapy could be explained by a strong recruitment of macrophage line- age cells via VEGF-A because macrophages express VEGFR-1, and VEGFR-1-dependent signaling pro- motes the migration of macrophages. Thus, VEGF-E might be safer than VEGF-A in terms of dosage and duration of administration (Fig. 2). Because the VEGF-E gene was originally found in a proangiogenic sheep ⁄ goat (sometimes human)-oriented parapox virus, ‘Orf virus’, and does not exist in the human genome, ‘humanization’ of this protein to decrease its possible antigenicity is needed. Such a trial has been already carried out successfully [8]. Other factors unrelated to VEGF, including angio- poietin or its modified molecule Comp-Ang1, FGF and hepatocyte growth factor may also improve the supply of blood into ischemic areas after stroke. In addition, the transcription factor PGC-1a was recently reported to have angiogenic activity via upregulation of VEGF gene expression independent of the hypoxia-inducible factor (HIF) system [17]. Further study is needed to clarify which factor is most benefi- cial for the recovery from brain ischemia. Motor neuron degeneration In 2001, Oosthuyse et al. [18] reported that deletion of the hypoxia-response element of the VEGF-A gene promoter and a reduction in VEGF-A expression cause the degeneration of motor neurons. This study raised the possibility that the VEGF and motor neuron systems interact closely. Furthermore, Sun et al. [19] found that VEGF-B, a member of the VEGF family, has neuroprotective activity. VEGF-B knockout mice showed increased severity after cerebral ischemic injury. However, it was not clear whether the effect of VEGF-B is direct or indirect, for example, via the promotion of pericyte activity. Poesen et al. [20] have studied this theme exten- sively, and found that: (a) VEGF-B is dispensable for the survival of motor neurons in healthy mice; (b) however, among mutant SOD1-overexpressing trans- genic mice, a model for amyotrophic lateral sclerosis, VEGF-B ) ⁄ ) mice showed faster motor neuron degeneration than VEGF-B + ⁄ ) or + ⁄ + mice; (c) the VEGF-B receptor, VEGFR-1 (Flt-1), is expressed in astrocytes and motor neurons after injury, and using VEGFR-1 (flt-1) TK) ⁄ ) mice, which are defi- cient in VEGFR-1 signaling [21], they showed that the VEGFR-1 expressed in motor neurons mediates the neuroprotective effect of VEGF-B. In addition, the administration of VEGF-B was reported to increase the survival rate of amyotrophic lateral scle- rosis rats (Fig. 3). Taken together, these results sug- gest that the VEGF-B–VEGFR-1 system (and maybe also the PlGF–VEGFR-1 system) is motor neuron protective in vivo, and may be a therapeutic target for diseases involving the degeneration of motor neurons. VEGF-A also binds and activates VEGFR-1, and so could be another candidate for the treatment of motor neuron degeneration. However, as discussed previ- ously, VEGF-A simultaneously activates VEGFR-1 and VEGFR-2 on vascular endothelial cells, resulting in strong hyperpermeability and brain edema. In this regard, VEGF-B or a molecule with similar activity like PlGF is expected to be a potential therapeutic tool for this disease. Brain tumors – malignant glioma Major malignant tumors in the brain include high- grade astrocytoma and glioblastoma multiforme. These Fig. 2. The VEGFR-2-specific ligand VEGF-E may have a broad range of therapeutic uses with less edema. VEGF-A activates both VEGFR-1 and VEGFR-2, resulting in angiogenesis and vascular per- meability. Therefore, the transfer of VEGF-A to ischemic tissue such as brain stroke areas can easily induce tissue edema. An inflammatory response may also be elevated via recruitment of VEGFR-1-expressing macrophages. By contrast, VEGF-E and its humanized version efficiently induced angiogenesis without severe edema or inflammation. The safety of VEGF-E appears greater than that of VEGF-A. M. Shibuya Therapeutic aspects of VEGF in brain diseases FEBS Journal 276 (2009) 4636–4643 ª 2009 The Author Journal compilation ª 2009 FEBS 4639 tumors have a relatively high incidence and are signifi- cantly invasive and metastatic within the CNS in the late stages. The origins of both tumors are thought to be glial cells, thus, these tumors appear to be very sim- ilar, and have been designated as a single entity, malig- nant glioma. Because malignant glioma is a highly vascularized tumor and its vascular density has been reported to correlate with a poor clinical prognosis, it is focused on here. Malignant glioma cells show a loss of function in tumor suppressor genes such as PTEN and p53, and the activation of oncogenes such as gene amplification of EGFR in either the wild-type or dominant active form [22,23]. Some malignant gliomas also show c-myc activation, but the dominant active mutant form of Ras is less frequent. Gene amplification and activation of EGFR occur in more than one third of cases, the highest incidence among human tumors. Malignant glioma cells secrete a variety of angiogenic factors such as VEGF and basic FGF [24]. VEGF is considered to have a major role in angiogenesis, as suggested in other solid tumors like colon carcinoma and breast carcinoma. The molecular basis for the upregulation of VEGF gene expression in gliomas has at least four mechanisms. (a) A hypoxia ⁄ HIF-related mechanism because of a low oxygen concentration in growing malignant glioma tissues. (b) Another involves oncogenes, particularly the EGFR signaling pathway, which stimulates VEGF gene expression via a HIF- independent mechanism. (c) It has been reported that the FoxM1B transcription factor is upregulated in glioblastoma multiforme, but not in low-grade astro- cytoma, and stimulates VEGF expression independent of HIF [25]. (d) In addition to these mechanisms, Ido et al. recently reported that HuR protein is upregulated in glioblastoma multiforme under hypoxia [26]. HuR functions to suppress the post-transcriptional degrada- tion of VEGF-A mRNA under hypoxia, contributing to a further increase in VEGF levels. Accumulating evidence indicates that the VEGF and VEGFR system plays a major role in tumor angiogen- esis in malignant glioma, similar to most other solid tumors. VEGF-A activates both VEGFR-1 and VEG- FR-2, but these two receptors differ biochemically. The affinity of VEGFR-1 for VEGF-A is extremely high (K d = 1–10 pm), 10-fold that of VEGFR-2. However, the tyrosine kinase activity of VEGFR-1 is one order of magnitude lower than that of VEGFR-2 which is as strong as other typical tyrosine kinase receptors like EGFR. An important question is how tightly the signaling from each receptor is linked to tumor angiogenesis and the growth of malignant glioma in vivo. VEGFR-2 is specifically expressed in vascular endothelial cells, and directly transduces most of the mitotic signal towards ECs, resulting in angiogenesis. However, VEGFR-1 is expressed not only in vascular endothelial cells, but also in monocyte ⁄ macrophage lineage cells. To clarify the role of VEGFR-1 signaling in angiogenesis and tumor growth in glioma, Kerber et al. [27] recently studied the growth rate of intracranially transplanted glioma cells in bone marrow-transplanted mice. They used two systems, irradiated wild-type mice carrying wild-type bone marrow cells, and irradiated wild-type mice carrying VEGFR-1 (Flt-1) TK) ⁄ ) bone marrow cells. VEGFR-1 TK) ⁄ ) mouse cells are deficient in sig- naling from VEGFR-1 because of a lack of the tyro- sine kinase domain [21]. They used three cell types, the original glioma cells, VEGF-A-overexpressing glioma cells and PlGF-overexpressing glioma cells. Remark- ably, all three gliomas showed a significant decrease in growth in vivo ( 30–50% decrease) in mice carrying VEGFR-1 TK) ⁄ ) bone marrow cells compared with mice carrying wild-type cells. In parallel with the decrease in tumor volume, the total number of tumor vessels, vessel density and number of infiltrating mac- rophage lineage cells were significantly reduced in mice carrying VEGFR-1 TK) ⁄ ) bone marrow cells. These results strongly suggest that, in this model of cranial glioma, nearly half of all tumor growth is dependent on VEGFR-1 signaling, possibly on bone marrow- derived VEGFR-1-expressing macrophages. These macrophages may act as pro-angiogenic and Fig. 3. The VEGF-B–VEGFR-1 system expressed on motor neurons acts to stop degeneration. VEGFR-1 is expressed on motor neu- rons, and its ligand VEGF-B activates the receptor to generate a survival signal. In VEGF-B) ⁄ ) or VEGFR-1 TK() ⁄ )) condition, motor neuron showed severe degeneration [20]. Thus, the VEGFR-1 path- way activated either by VEGF-B or by PlGF may be useful for pro- tecting motor neurons under certain conditions such as amyotrophic lateral sclerosis. Therapeutic aspects of VEGF in brain diseases M. Shibuya 4640 FEBS Journal 276 (2009) 4636–4643 ª 2009 The Author Journal compilation ª 2009 FEBS pro-tumorigenic cells similar to the tumor-associated macrophages reported by several groups [28]. Glioma metastasis and VEGF Intracranial invasion and metastasis are major prob- lems in the prognosis of malignant glioma. However, their molecular basis is largely unknown. Several possi- bilities can be considered, including: (a) the intravascu- lar migration of glioma cells into blood vessels and their transfer to distant areas in the brain via blood flow; (b) the rapid migration of tumor cells outside blood vessels; and (c) the migration of tumor cells independent of the vascular network, but via other brain-specific structures such as neuronal fibers ⁄ axon bundles. Under physiological conditions, glial cells and vascular endothelial cells have cross-contacts, and establish the BBB. It is of interest whether such a glial cell–EC contact system is partly used for the rapid migration of tumor cells through the vessel network. The VEGF–VEGFR system is now widely accepted as a major factor in a variety of solid tumors, as strongly suggested to be the case in malignant glioma also. Based on the results of phase III studies [29], bev- acizumab, a humanized monoclonal anti-VEGF-A neutralizing IgG, has been approved in many countries for the treatment of colorectal cancer, lung cancer (non-small cell, nonepithelial type) and breast cancer. Furthermore, orally available small molecules, solafe- nib and sunitinib, which inhibit a variety of tyrosine kinases including VEGFRs, have been approved for the treatment of renal cell cancer and liver cancer. These anti-angiogenic drugs have significantly improved the disease-free survival rate and total sur- vival rate of cancer patients via at least two mecha- nisms, (a) blocking of tumor angiogenesis and (b) normalization of tumor vessels, although some adverse effects have been observed [30]. Bevacizumab in com- bination with cytotoxic agents such as irinotecan, and other anti-angiogenic drugs such as VEGF-Trap have recently been reported to be beneficial for the suppres- sion of tumor growth and for longer survival in malig- nant glioma patients in phase II clinical trials [31–33]. However, a few reports suggest that glioma cells might have acquired invasive and metastatic phenotypes via the co-option of tumor cells with the vascular network and via other mechanisms [30,34]. Under hypoxic conditions or at poor nutrition, some tumors have been reported to become resistant, being less apoptotic, and more invasive. Our recent studies also indicate that, in the malignant melanoma model, tumor cells show a spheroid-like structure and become more resistant to hypoxia–low nutrition double stress, resulting in an aggressive phenotype in terms of inva- sion and metastasis in vivo [35]. Late in its clinical course, malignant glioma is known to show extensive cell migration, and become more invasive and metastatic. Such metastasis within the brain is inoperable, and thus lethal. In this regard, anti-angiogenic therapy should also be studied exten- sively in animal models to optimize the suppression of tumor growth as well as block the nerve dysfunction caused by the tumor mass without making the tumor more invasive or aggressive. To this end, cellular responses of malignant glioma to anti-angiogenic stress (hypoxia and low nutrition) should be clarified, and combinations of anti-angiogenic drugs and inhibitors to suppress aggressiveness induced by anti-angiogenic stress need to be considered. Other brain tumors In other brain tumors, hemangioblastoma is relatively rare ( 3% of all tumors in the CNS), but occurs in von Hippel–Lindau (VHL) patients. This tumor occurs not in the cerebrum, but in limited areas such as the retina, cerebellum, brainstem and spinal cord. Heman- gioblastoma in VHL patients might be sensitive to anti-VEGF–VEGFR therapy because VEGF-A is thought to be abnormally upregulated because of con- stitutive activation of the HIF pathway, similar to VHL-deficient renal cell cancer. Treatment of a retinal hemangioblastoma patient with SU5416, a VEGFR- specific inhibitor, was effective in recovering visual functions [36], suggesting that the above strategy may work. Comparative studies with tumors in the brain and other organs in terms of the molecular mechanism for tumor angiogenesis are also important to obtain a strategy to block the angiogenic pathway. Conclusion and perspectives Major brain diseases, i.e. ischemic diseases and brain tumors such as malignant glioma, are closely linked to the blood vessel system in the CNS. Therefore, thera- peutic strategies in the near future will be directly related to the artificial manipulation of vessel struc- tures and functions via pro- or anti-angiogenic agents. The basic regulators of blood vessels in the CNS appear to be VEGF–VEGFR, angiopoietin–Tie and BBB-related factors, but the molecular basis of these signaling pathways is not fully understood. More stud- ies on these pathways are needed in a CNS-specific manner. In addition, the mechanism behind vascular permeability and the formation of edema in brain tissue needs to be clarified to obtain a strategy with M. Shibuya Therapeutic aspects of VEGF in brain diseases FEBS Journal 276 (2009) 4636–4643 ª 2009 The Author Journal compilation ª 2009 FEBS 4641 which to rapidly and efficiently suppress vascular leaks during the clinical course of brain diseases or during the treatment of brain ischemia using pro-angiogenic medicine. Acknowledgements This work was supported by Grant-in-Aid Special Pro- ject Research on Cancer-Bioscience 17014020 from the Ministry of Education, Culture, Sports, Science and Technology of Japan, a grant from the program ‘Research for the Future’ of the Japan Society for the Promotion of Science, and the program ‘Promotion of Fundamental Research in Health Science’ of the Organi- zation for Pharmaceutical Safety and Research. In May 2009, the FDA in the USA approved bevacizumab for the treatment of patients with relapsed glioblastoma. References 1 Ferrara N, Gerber HP & LeCouter J (2003) The biology of VEGF and its receptors. Nat Med 9, 669–679. 2 Shibuya M (2008) Vascular endothelial growth factor- dependent and -independent regulation of angiogenesis. BMB Rep 41, 278–286. 3 de Almodovar CR, Zacchigna S & Carmeliet P (2008) Angiogenesis in the central nervous system. In Angiogenesis: An Integrative Approach from Science to Medicine (Figg W & Folkmann J, eds) pp. 489–504. Springer, New York, NY. 4 Lee SW, Kim WJ, Choi YK, Song HS, Son MJ, Gelman IH, Kim YJ & Kim KW (2003) SSeCKS regulates angiogenesis and tight junction formation in blood–brain barrier. Nat Med 9, 900–906. 5 Sakurai Y, Ohgimoto K, Kataoka Y, Yoshida N & Shibuya M (2005) Essential role of Flk-1 (vascular endothelial growth factor receptor-2) tyrosine residue- 1173 in vasculogenesis in mice. Proc Natl Acad Sci USA 102, 1076–1081. 6 Sawano A, Iwai S, Sakurai Y, Ito M, Shitara K, Nakahata T & Shibuya M (2001) Vascular endothelial growth factor receptor-1 (Flt-1) is a novel cell surface marker for the lineage of monocytes–macrophages in humans. Blood 97, 785–791. 7 Kiba A, Sagara H, Hara T & Shibuya M (2003) VEG- FR-2-specific ligand VEGF-E induces non-edematous hyper-vascularization in mice. Biochem Biophys Res Commun 301, 371–377. 8 Zheng Y, Murakami M, Takahashi H, Yamauchi M, Kiba A, Yamaguchi S, Yabana N, Alitalo K & Shibuya M (2006) Chimeric VEGF-E NZ7 ⁄ PlGF promotes angio- genesis via VEGFR-2 without significant enhancement of vascular permeability and inflammation. Arterioscler Thromb Vasc Biol 26, 2019–2026. 9 Takahashi H, Hattori S, Iwamatsu A, Takizawa H & Shibuya M (2004) A novel snake venom vascular endo- thelial growth factor (VEGF) predominantly induces vascular permeability through preferential signaling via VEGF receptor-1. J Biol Chem 279, 46304–46314. 10 Plate KH, Breier G, Weich HA, Mennel HD & Risau W (1994) Vascular endothelial growth factor and gli- oma angiogenesis: coordinate induction of VEGF recep- tors, distribution of VEGF protein and possible in vivo regulatory mechanisms. Int J Cancer 59, 520–529. 11 Kargiotis O, Rao JS & Kyritsis AP (2006) Mechanisms of angiogenesis in gliomas. J Neurooncol 78, 281–293. 12 Lee S, Chen TT, Barber CL, Jordan MC, Murdock J, Desai S, Ferrara N, Nagy A, Roos KP & Iruela-Arispe ML (2007) Autocrine VEGF signaling is required for vascular homeostasis. Cell 130, 691–703. 13 Alitalo K & Carmeliet P (2002) Molecular mechanisms of lymphangiogenesis in health and disease. Cancer Cell 1, 219–227. 14 Heiss WD, Sobesky J & Hesselmann V (2004) Identify- ing thresholds for penumbra and irreversible tissue damage. Stroke 35 (11 Suppl 1), 2671–2674. 15 Zhang ZG, Zhang L, Jiang Q, Zhang R, Davies K, Powers C, Bruggen N & Chopp M (2000) VEGF enhances angiogenesis and promotes blood–brain barrier leakage in the ischemic brain. J Clin Invest 106, 829–838. 16 Inoue N, Kondo T, Kobayashi K, Aoki M, Numaguchi Y, Shibuya M & Murohara T (2007) Therapeutic angio- genesis using novel vascular endothelial growth factor- E ⁄ human placental growth factor chimera genes. Arterioscler Thromb Vasc Biol 27, 99–105. 17 Arany Z, Foo SY, Ma Y, Ruas JL, Bommi-Reddy A, Girnun G, Cooper M, Laznik D, Chinsomboon J, Rangwala SM et al. (2008) HIF-independent regulation of VEGF and angiogenesis by the transcriptional coac- tivator PGC-1alpha. Nature 451, 1008–1012. 18 Oosthuyse B, Moons L, Storkebaum E, Beck H, Nuyens D, Brusselmans K, Van Dorpe J, Hellings P, Gorselink M, Heymans S et al. (2001) Deletion of the hypoxia-response element in the vascular endothelial growth factor promoter causes motor neuron degenera- tion. Nat Genet 28, 131–138. 19 Sun Y, Jin K, Childs JT, Xie L, Mao XO & Greenberg DA (2006) Vascular endothelial growth factor-B (VEG- FB) stimulates neurogenesis: evidence from knockout mice and growth factor administration. Dev Biol 289, 329–335. 20 Poesen K, Lambrechts D, Van Damme P, Dhondt J, Bender F, Frank N, Bogaert E, Claes B, Heylen L, Verheyen A et al. (2008) Novel role for VEGF-recep- tor-1 and its ligand VEGF-B in motor neuron degenera- tion. J Neurosci 28, 10451–10459. 21 Hiratsuka S, Minowa O, Kuno J, Noda T & Shibuya M (1998) Flt-1 lacking the tyrosine kinase domain is Therapeutic aspects of VEGF in brain diseases M. Shibuya 4642 FEBS Journal 276 (2009) 4636–4643 ª 2009 The Author Journal compilation ª 2009 FEBS sufficient for normal development and angiogenesis in mice. Proc Natl Acad Sci USA 95, 9349–9354. 22 Fueyo J, Gomez-Manzano C, Yung WKA & Kyritsis AP (1998) The functional role of tumor suppressor genes in gliomas: clues for future therapeutic strategies. Neurology 51, 1250–1255. 23 Prigent SA, Nagane M, Lin H, Huvar I, Boss GR, Feramisco JR, Cavenee WK & Huang HS (1996) Enhanced tumorigenic behaviour of glioblastoma cells expressing a truncated epidermal growth factor receptor is mediated through the Ras–Shc–Grb2 pathway. J Biol Chem 271, 25639–25645. 24 Saleh M, Stacker SA & Wilks AF (1996) Inhibition of growth of C6glioma cells in vivo by expression of anti- sense vascular endothelial growth factor sequence. Cancer Res 5, 393–401. 25 Zhang Y, Zhang N, Dai B, Liu M, Sawaya R, Xie K & Huang S (2008) FoxM1B transcriptionally regulates vascular endothelial growth factor expression and pro- motes the angiogenesis and growth of glioma cells. Cancer Res 68, 8733–8742. 26 Ido K, Nakagawa T, Sakuma T, Takeuchi H, Sato K & Kubota T (2008) Expression of vascular endothelial growth factor-A and mRNA stability factor HuR in human astrocytic tumors. Neuropathology 28, 604–611. 27 Kerber M, Reiss Y, Wickersheim A, Jugold M, Kiessling F, Heil M, Tchaikovski V, Waltenberger J, Shibuya M, Plate KH et al. (2008) Flt-1 signaling in macrophages promotes glioma growth in vivo. Cancer Res 68, 7342–7351. 28 Lin EY & Pollard JW (2007) Tumor-associated macro- phages press the angiogenic switch in breast cancer. Cancer Res 67, 5064–5066. 29 Hurwitz H, Fehrenbacher L, Novotny W, Cartwright T, Hainsworth J, Heim W, Berlin J, Baron A, Griffing S, Holmgren E et al. (2004) Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 350, 2335. 30 Jain RK, Tong RT & Munn LL (2007) Effect of vascu- lar normalization by antiangiogenic therapy on intersti- tial hypertension, peritumor edema, and lymphatic metastasis: insights from a mathematical model. Cancer Res 67, 2729. 31 Vredenburgh JJ, Desjardins A, Herndon JE 2nd, Dowell JM, Reardon DA, Quinn JA, Rich JN, Sathornsumetee S, Gururangan S, Wagner M et al. (2007) Phase II trial of bevacizumab and irinotecan in recurrent malignant glioma. Clin Cancer Res 13, 1253. 32 Norden AD, Young GS, Setayesh K, Muzikansky A, Klufas R, Ross GL, Ciampa AS, Ebbeling LG, Levy B, Drappatz J et al. (2008) Bevacizumab for recurrent malignant gliomas: efficacy, toxicity, and patterns of recurrence. Neurology 70, 779–787. 33 Zuniga RM, Torcuator R, Jain R, Anderson J, Doyle T, Ellika S, Schultz L & Mikkelsen T (2009) Efficacy, safety and patterns of response and recurrence in patients with recurrent high-grade gliomas treated with bevacizumab plus irinotecan. J Neurooncol 91, 329–336. 34 Rubenstein JL, Kim J, Ozawa T, Zhang M, Westphal M, Deen DF & Shuman MA (2000) Anti-VEGF anti- body treatment of glioblastoma prolongs survival but results in increased vascular cooption. Neoplasia 2, 306–314. 35 Aiello LP, George DJ, Cahill MT, Wong JS, Cavaller- ano J, Hannah AL & Kaelin WG Jr (2002) Rapid and durable recovery of visual function in a patient with von Hippel–Lindau syndrome after systemic therapy with vascular endothelial growth factor receptor inhibi- tor su5416. Ophthalmology 109, 1745–1751. 36 Osawa T, Muramatsu M, Watanabe M & Shibuya M (2009) Hypoxia and low nutrition double stress induces aggressiveness in a murine model of melanoma. Cancer Sci 100, 844–851. M. Shibuya Therapeutic aspects of VEGF in brain diseases FEBS Journal 276 (2009) 4636–4643 ª 2009 The Author Journal compilation ª 2009 FEBS 4643 . MINIREVIEW Brain angiogenesis in developmental and pathological processes: therapeutic aspects of vascular endothelial growth factor Masabumi. (1994) Vascular endothelial growth factor and gli- oma angiogenesis: coordinate induction of VEGF recep- tors, distribution of VEGF protein and possible in