Placental growth factor neutralising antibodies give limited anti-angiogenic effects in an in vitro organotypic angiogenesis model

Vascular Endothelial Growth Factor Receptor (VEGFR) mediated signalling drives angiogenesis. This is predominantly attributed to the activity of VEGFR-2 following binding of VEGF-A. Whether other members of the VEGFR and ligand families such as VEGFR-1 and its ligand Placental Growth Factor (PlGF) can also contribute to developmental and pathological angiogenesis is less clear. We explored the function of PlGF in VEGF-A dependent angiogenesis using an in vitro co-culture assay in which endothelial cells are cultured on a fibroblast feeder layer. In the presence of 2% FS MCDB media (containing limited growth factors) in vitro endothelial tube formation is driven by endogenous angiogenic stimuli which are produced by the fibroblast and endothelial cells. Under these conditions independent sequestration of either free VEGF-A or PlGF with polyclonal and monoclonal antibodies inhibited tube formation suggesting that both ligands are required to drive an angiogenic response. Endothelial tube formation could only be driven within this assay by the addition of exogenous VEGF-A, VEGF-E or VEGF-A/PlGF heterodimer, but not by PlGF alone, implying that activation of either VEGFR-2/VEGFR-1 heterodimers or VEGFR-2 homodimers were responsible for eliciting an angiogenic response directly, but not VEGFR-1 homodimers. In contrast to results obtained with an endogenous angiogenic drive, sequestration of PlGF did not affect endothelial tube formation when the assay was driven by 1 ng/ml exogenous VEGF-A. These data suggest that although neutralising PlGF can be shown to reduce endothelial tube formation in vitro, this effect is only observed under restricted culture conditions and is influenced by VEGF-A. Such data questions whether neutralising PlGF would have a therapeutic benefit in vivo in the presence of pathological concentrations of VEGF-A.     

Collagen type 1 retards tube formation by human microvascular endothelial cells in a fibrin matrix

Angiogenesis, or the formation of new microvessels, is often encountered in pathological situations. A fibrinous exudate can often act as a temporary matrix for the ingrowth of these new microvessels. This matrix consists mainly of fibrin, but is mingled with other plasma components and interstitial collagen fibres. In vitro, capillary-like tube formation can be mimicked by exposing human microvascular endothelial cells (hMVECs), seeded on top of a three-dimensional fibrin matrix, to an angiogenic growth factor (e.g. fibroblast growth factor (FGF)-2) and the cytokine tumour necrosis factor (TNF)-. Plasmin activity is required in this process. We investigated whether the angiogenic potential of hMVECs was altered by the presence of collagen. The addition of type I collagen to fibrin matrices dose-dependently inhibited tube-formation. Tube-formation in these fibrin/collagen matrices by hMVECs required matrix metalloprotease (MMP) activity, as well as plasmin activity. On a pure collagen type I matrix, hMVECs were not able to form tube-like structures in the matrix but formed sprouts. This sprouting required MMP activity and was, in contrast to the tube-like structures in a fibrin matrix, not influenced by hypoxia. These data indicate that the interaction between endothelial cells and different matrix components is of importance for the angiogenic potential of these cells.     

Involvement of VEGFR-2 (kdr/flk-1) but not VEGFR-1 (flt-1) in VEGF-A and VEGF-C-induced tube formation by human microvascular endothelial cells in fibrin matrices in vitro

Different forms of vascular endothelial growth factor (VEGF) and their cellular receptors (VEGFR) are associated with angiogenesis, as demonstrated by the lethality of VEGF-A, VEGFR-1 or VEGFR-2 knockout mice. Here we have used an in vitro angiogenesis model, consisting of human microvascular endothelial cells (hMVEC) cultured on three-dimensional (3D) fibrin matrices to investigate the roles of VEGFR-1 and VEGFR-2 in the process of VEGF-A and VEGF-C-induced tube formation. Soluble VEGFR-1 completely inhibited the tube formation induced by the combination of VEGF-A and TNFα (VEGF-A/TNFα). This inhibition was not observed when tube formation was induced by VEGF-C/TNFα or bFGF/TNFα. Blocking monoclonal antibodies specific for VEGFR-2, but not antibodies specifically blocking VEGFR-1, were able to inhibit the VEGF-A/TNFα-induced as well as the VEGF-C/TNFα-induced tube formation in vitro. PlGF-2, which interacts only with VEGFR-1, neither induced tube formation in combination with TNFα, nor inhibited or stimulated by itself the VEGF-A/TNFα-induced tube formation in vitro. These data indicate that VEGF-A or VEGF-C activation of the VEGFR-2, and not of VEGFR-1, is involved in the formation of capillary-like tubular structures of hMVEC in 3D fibrin matrices used as a model of repair-associated or pathological angiogenesis in vitro. This revised version was published online in June 2006 with corrections to the Cover Date.     

Membrane-type matrix metalloproteinase-mediated angiogenesis in a fibrin-collagen matrix

Adult angiogenesis, associated with pathologic conditions, is often accompanied by the formation of a fibrinous exudate. This temporary matrix consists mainly of fibrin but is intermingled with plasma proteins and collagen fibers. The formation of capillary structures in a fibrinous matrix in vivo was mimicked by an in vitro model, in which human microvascular endothelial cells (hMVECs) seeded on top of a fibrin-10% collagen matrix form capillarylike tubular structures after stimulation with basic fibroblast growth factor/tumor necrosis factor alpha (bFGF/TNF-alpha) or vascular endothelial growth factor (VEGF)/TNF-alpha. In the fibrin-collagen matrix the metalloproteinase inhibitor BB94 inhibited tubule formation by 70% to 80%. Simultaneous inhibition of plasmin and metalloproteinases by aprotinin and BB94 caused a nearly complete inhibition of tubule formation. Adenoviral transduction of tissue inhibitor of metalloproteinases 1 (TIMP-1) and TIMP-3 into endothelial cells revealed that TIMP-3 markedly inhibited angiogenesis, whereas TIMP-1 had only a minor effect. Immunohistochemical analysis showed the presence of matrix metalloproteinase 1 (MMP-1), MMP-2, and membrane-type 1 (MT1)-MMP, whereas MMP-9 was absent. The endothelial production of these MMPs was confirmed by antigen assays and real-time polymerase chain reaction (PCR). MT1-MMP mRNA was markedly increased in endothelial cells under conditions that induced tubular structures. The presence of MMP-1, MMP-2, and MT1-MMP was also demonstrated in vivo in the newly formed vessels of a recanalized arterial mural thrombus. These data suggest that MMPs, in particular MT-MMPs, play a pivotal role in the formation of capillarylike tubular structures in a collagen-containing fibrin matrix in vitro and may be involved in angiogenesis in a fibrinous exudate in vivo.     

Activation of plasminogen into plasmin at the surface of endothelial microparticles: a mechanism that modulates angiogenic properties of endothelial progenitor cells in vitro

The regulation of plasmin generation on cell surfaces is of critical importance in the control of vascular homeostasis. Cell-derived microparticles participate in the dissemination of biological activities. However, their capacity to promote plasmin generation has not been documented. In this study, we show that endothelial microparticles (EMPs) from tumor necrosis factor alpha (TNFalpha)-stimulated endothelial cells served as a surface for the generation of plasmin. The generation of plasmin involved expression of urokinase-type plasminogen activator (uPA) and its receptor (uPAR) at the surface of EMPs and was further increased by their ability to bind exogenous uPA on uPAR. Plasminogen was activated at the surface of EMPs in a dose-dependent, saturable, and specific manner as indicated by the inhibition of plasmin formation by epsilon-amino-caproic acid (epsilon-ACA) and carboxypeptidase B. EMP-induced plasmin generation affects tube formation mediated by endothelial progenitor cells. However, low amounts of EMPs increased tube formation, whereas higher concentrations inhibited it. Prevention of these effects by inhibitors of either uPA or plasmin underscore the key role of EMP-induced plasmin generation. In conclusion, we demonstrated that EMPs act as vectors supporting efficient plasmin generation and dissemination, a new pathway in the regulation of endothelial proteolytic activities with potential involvement in inflammation, angiogenesis, and atherosclerosis.     

Structure and function of placental growth factor

Placental growth factor (PlGF) belongs to the same family as the vascular endothelial growth factor A (VEGF-A). Recent gene inactivation studies in mice have demonstrated that loss of PlGF does not affect development, reproduction, or normal postnatal life. However, the mice show significantly impaired angiogenesis and arteriogenesis during pathological conditions such as ischemia and tumor formation, conditions in which the expression of VEGF-A is normally increased. Mice expressing a truncated form of the specific receptor for PlGF, the VEGF receptor 1 (VEGFR-1), show impaired angiogenesis similar to that observed in Plgf(-/-)mice. These data suggest a pivotal role for PlGF and VEGFR-1 in regulating VEGF-A-dependent angiogenesis under pathological conditions. VEGF-A has been utilized for the therapeutic stimulation of new blood vessels in ischemic hearts and limbs, with controversial results from the initial clinical experience. This review discusses the possibility of using the PlGF/VEGFR-1 pathway as an alternative target for angiogenic therapy.     

Factor XIIIa supports microvascular endothelial cell adhesion and inhibits capillary tube formation in fibrin

Coagulation factor XIIIa is a transglutaminase that catalyzes covalent cross-link formation in fibrin clots. In this report, we demonstrate that factor XIIIa also mediates adhesion of endothelial cells and inhibits capillary tube formation in fibrin. The adhesive activity of factor XIIIa was not dependent on the transglutaminase activity, and did not involve the factor XIIIb-subunits. The adhesion was inhibited by 99% using a combination of monoclonal antibodies directed against integrin alpha(v)beta(3) and beta(1)-containing integrins, and was dependent on Mg(2+) or Mn(2+). Soluble factor XIIIa also bound to endothelial cells in solution, as detected by flow cytometry. In addition, factor XIIIa inhibited endothelial cell capillary tube formation in fibrin in a dose-dependent manner. Furthermore, the extent of inhibition differed in 2 types of fibrin. The addition of 10 to 100 microg/mL factor XIIIa produced a dose-dependent reduction in capillary tube formation of 60% to 100% in gammaA/gammaA fibrin, but only a 10% to 37% decrease in gammaA/gamma' fibrin. These results show that factor XIIIa supports endothelial cell adhesion in an integrin-dependent manner and inhibits capillary tube formation. (Blood. 2000;95:2586-2592)     

Hypoxia induces angiogenic factors in brain microvascular endothelial cells

Hypoxia is increasingly recognized as an important contributing factor to the development of brain diseases such as Alzheimer's disease (AD). In the periphery, hypoxia is a powerful regulator of angiogenesis. However, vascular endothelial cells are remarkably heterogeneous and little is known about how brain endothelial cells respond to hypoxic challenge. The objective of this study is to characterize the effect of hypoxic challenge on the angiogenic response of cultured brain-derived microvascular endothelial cells. Brain endothelial cell cultures were initiated from isolated rat brain microvessels and subjected to hypoxia (1% O(2)) for various time periods. The results showed that hypoxia induced rapid (≤ 0.5h) expression of hypoxia-inducible factor 1α (HIF-1α) and that cell viability, assessed by MTT assay, was unaffected within the first 8h. Examination of brain endothelial cell cultures for pro- and anti-angiogenic proteins by western blot, RT-PCR and ELISA revealed that within 0.5 to 2h of hypoxia levels of vascular endothelial growth factor and endothelin-1 mRNA and protein were elevated. The expression of heme oxygenase-1 also increased but only after 8h of hypoxia. In contrast, similar hypoxia exposure evoked a decrease in endothelial nitric oxide synthase and thrombospondin-2 levels. Exposure of brain endothelial cell cultures to hypoxia resulted in a significant (p<0.001) decrease (94%) in tube length, an in vitro index of angiogenesis, compared to control cultures. The data indicate that, despite a shift toward a pro-angiogenic phenotype, hypoxia inhibited vessel formation in brain endothelial cells. These results suggest that in brain endothelial cells expression of angiogenic factors is not sufficient for the development of new vessels. Further work is needed to determine what factors/conditions prevent hypoxia-induced angiogenic changes from culminating in the formation of new brain blood vessels and what role this may play in the pathologic changes observed in AD and other diseases characterized by cerebral hypoxia.     

An Engineered Heparin-Binding Form of VEGF-E (hbVEGF-E). Biological effects in vitro and mobilizatiion of precursor cells

Vascular endothelial growth factor (VEGF-A) is the founding member of a family of angiogenic proteins with various binding abilities to three cognate VEGF receptors. Previously, a gene encoding from the genome of parapox orf virus (OV) with about 25% amino acid identity to mammalian VEGF-A was named VEGF-E and shown to bind and specifically activate the vascular endothelial growth factor receptor VEGFR-2 (KDR/flk-1). Here, we have generated a novel heparin-binding form of VEGF-E by introducing the heparin-domain of the human VEGF-A(165) splice variant into the viral VEGF-E protein. Recombinant heparin-binding VEGF-E (hbVEGF-E) is shown to stimulate proliferation and sprout formation of macro- and microvascular endothelial cells to a similar extent as the parental OV-VEGF-E but fails to activate peripheral mononuclear cells. However, hbVEGF-E is more potent in binding competition assays with primary human endothelial cells when compared to the OV-VEGF-E. This can be explained by our finding that binding of hbVEGF-E but not of parental OV-VEGF-E to the VEGFR-2 is strongly increased by the addition of neuropilin-1 (NP-1), a cognate co-receptor for VEGF-A. The engineered hbVEGF-E was compared with the VEGFR-1 selective and also heparin-binding form of placenta growth factor (PlGF-2) in vivo. Both heparin-binding homologues induced mobilization of endothelial progenitor cells from the bone marrow and gave rise to similar colony numbers of myeloic cells in a colony-forming assay. These findings suggest that both VEGFR-1 and VEGFR-2 are involved in stem cell mobilization.     

Contribution of Bcl-2, but not Bcl-xL and Bax, to antiapoptotic actions of hepatocyte growth factor in hypoxia-conditioned human endothelial cells

Angiogenic growth factors play important roles in angiogenic responses, such as vasculogenesis and angiogenesis in response to hypoxia. A novel angiogenic growth factor, hepatocyte growth factor (HGF), has been reported to inhibit endothelial cell death. However, its molecular mechanisms are largely unknown. Thus, we studied (1) the effects of HGF on hypoxia-induced endothelial apoptosis and (2) the molecular mechanisms of the antiapoptotic actions of HGF in endothelial cells. Severe hypoxia increased the cell death rate in human aortic endothelial cells, whereas HGF significantly attenuated cell death. In addition, hypoxic treatment resulted in a significant increase in apoptotic cells, whereas HGF could attenuate apoptosis, accompanied by attenuation of the increase in caspase-3-like activity (P<0.01). Of importance, HGF significantly increased Bcl-2, an inhibitor of apoptosis, in a dose-dependent manner under normoxic and hypoxic conditions (P<0.01), whereas hypoxic conditions resulted in a significant decrease in Bcl-2. In contrast, HGF failed to affect Bcl-xL, which is also well known as an inhibitor of apoptosis under both normoxic and hypoxic conditions, whereas Bcl-xL was significantly decreased in endothelial cells exposed to hypoxia (P<0.01). No significant change in Bax, a promoter of apoptosis, was also observed in endothelial cells under hypoxia, whereas HGF did not affect BAX: Overall, this study demonstrated that HGF prevented endothelial cell death induced by hypoxia through its antiapoptotic action. The antiapoptotic mechanisms of HGF in hypoxia-induced endothelial cell death largely depend on Bcl-2, but not Bcl-xL and BAX:     

Dependence of human vascular cell surface proteolysis on expression of the urokinase receptor

To delineate the role of binding of urokinase type plasminogen activator (uPA) to its receptor (uPAR) in the local generation of plasmin by endothelium, we transfected spontaneously transformed immortalized human vascular endothelial cells that express high levels of uPA but low levels of uPAR with human uPAR complementary DNA. Compared with nontransfected cell, the stably transformed clonal cell line exhibited (a) a > 10-fold increase in steady-state uPAR mRNA levels documented with Northern blot analysis (n = 3), (b) a 2.8-fold increase in cell surface expression of uPAR protein quantified by enzyme linked immunosorbent assay (n = 3), (c) a 2.9-fold increase in specific binding of radiolabeled single chain uPA (n = 4), and (d) markedly increased matrix adhesion. The participation of uPAR in cell surface proteolysis was apparent based on a 3.0-fold increase in cell associated plasmin activity (n = 3) and a 2.3-fold increase in lysis of noncrosslinked fibrin clots (n = 5). Thus, local generation of plasmin and consequent degradation of fibrin are likely to be promoted by cell surface localization of uPA by uPAR in cellular constituents of the vessel wall. Furthermore, genetic engineering of endothelium to enhance expression of uPAR may confer resistance to thrombosis or restenosis associated with endovascular stents.     

VEGF-mediated endothelial P-selectin translocation: role of VEGF receptors and endogenous PAF synthesis

The acute increase in vascular permeability produced by vascular endothelial growth factor (VEGF-A(165)) requires activation of endothelial Flk-1 receptors (VEGFR-2) and stimulation of platelet-activating factor (PAF) synthesis. Like PAF, VEGF-A(165) promotes translocation of P-selectin to the endothelial cell (EC) surface. However, the mechanisms involved remain unknown. By treating human umbilical vein endothelial cells (HUVECs) with VEGF analogs, we show that activation of VEGFR-1 or VEGFR-2 or both induced a rapid and transient translocation of endothelial P-selectin and neutrophil adhesion to activated ECs. The effects mediated by VEGF-A(165) and VEGF-A(121) (VEGFR-1/VEGFR-2 agonists) were blocked by a selective VEGFR-2 inhibitor, SU1498. VEGF-A(165) was twice as potent as VEGF-A(121), which can be explained by the binding capacity of VEGF-A(165) to its coreceptor neuropilin-1 (NRP-1). Indeed, treatment with NRP-1 antagonist (GST-Ex7) reduced the effect of VEGF-A(165) to the levels observed upon stimulation with VEGF-A(121). Finally, the use of selective PAF receptor antagonists reduced VEGF-A(165)-mediated P-selectin translocation. Together, these data show that maximal P-selectin translocation and subsequent neutrophil adhesion was mediated by VEGF-A(165) on the activation of VEGFR-2/NRP-1 complex and required PAF synthesis.     

New functions of the fibrinolytic system in bone marrow cell-derived angiogenesis

Angiogenesis is a process by which new blood vessels form from preexisting vasculature. This process includes differentiation of angioblasts into endothelial cells with the help of secreted angiogenic factors released from cells such as bone marrow (BM)-derived cells. The fibrinolytic factor plasmin, which is a serine protease, has been shown to promote endothelial cell migration either directly, by degrading matrix proteins such as fibrin, or indirectly, by converting matrix-bound angiogenic growth factors into a soluble form. Plasmin can also activate other pericellular proteases such as matrix metalloproteinases (MMPs). Recent studies indicate that plasmin can additionally alter cellular adhesion and migration. We showed that factors of the fibrinolytic pathway can recruit BM-derived hematopoietic cells into ischemic/hypoxic tissues by altering the activation status of MMPs. These BM-derived cells can function as accessory cells that promote angiogenesis by releasing angiogenic signals. This review will discuss recent data regarding the role of the fibrinolytic system in controlling myeloid cell-driven angiogenesis. We propose that plasmin/plasminogen may be a potential target not only for development of effective angiogenic therapeutic strategies for the treatment of cancer, but also for development of strategies to promote ischemic tissue regeneration.     

A proangiogenic peptide derived from vascular endothelial growth factor receptor-1 acts through alpha5beta1 integrin

Vascular endothelial growth factor receptor-1 (VEGFR-1) is a tyrosine kinase receptor for growth factors of the VEGF family. Endothelial cells express a membrane-bound and a soluble variant of this protein, the latter being mainly considered as a negative regulator of VEGF-A signaling. We previously reported that the soluble form is deposited in the extracellular matrix produced by endothelial cells in culture and is able to promote cell adhesion and migration through binding to alpha5beta1 integrin. In this study, we demonstrate that the Ig-like domain II of VEGFR-1, which contains the binding determinants for the growth factors, is involved in the interaction with alpha5beta1 integrin. To identify domain regions involved in integrin binding, we designed 12 peptides putatively mimicking the domain II surface and tested their ability to inhibit alpha5beta1-mediated endothelial cell adhesion to soluble VEGFR-1 and directly support cell adhesion. One peptide endowed with both these properties was identified and shown to inhibit endothelial cell migration toward soluble VEGFR-1 as well. This peptide directly binds alpha5beta1 integrin, but not VEGF-A, inducing endothelial cell tubule formation in vitro and neoangiogenesis in vivo. Alanine scanning mutagenesis of the peptide defined which residues were responsible for its biologic activity and integrin binding.