PDGF-BB modulates endothelial proliferation and angiogenesis in vitro via PDGF beta-receptors.

To delineate potential angiogenic roles of platelet-derived growth factor (PDGF), we have investigated PDGF and its receptors on bovine aortic endothelial cells that exhibit spontaneous angiogenesis in vitro (angiogenic endothelial cells). Initiation of cord/tube formation by angiogenic endothelial cells required bovine or human serum. Neutralization of PDGF-BB in human serum with a monoclonal anti-PDGF-BB antibody reduced cord/tube formation by 37 +/- 10%, whereas neutralizing anti-PDGF-AA and an IgG isotype-matched control antibody had no effect. DNA synthesis in response to PDGF-BB increased as the cords and tubes developed; furthermore, PDGF-BB induced the incorporation of BrdU in the nuclei of cells associated with these structures. PDGF beta-receptor (PDGF-beta) mRNA increased concomitantly with cord/tube formation, and PDGFR-beta were specifically localized by immunocytochemistry to developing and mature cords and tubes. However, PDGFR-beta transcripts and protein were undetectable in nonangiogenic endothelial cells, and PDGF alpha-receptor mRNA was not expressed in either endothelial cell strain. In contrast to nonangiogenic endothelial cells, angiogenic endothelial cells did not express the PDGF B-chain, the required ligand for the PDGFR-beta. We conclude that (a) PDGF-BB can contribute to angiogenesis in vitro, (b) PDGFR-beta are specific for cord/tube-forming endothelial cells and mediate endothelial proliferation and cord/tube formation, and (c) in angiogenic and nonangiogenic endothelial cells, the expression of PDGFR-beta and PDGF B-chain is inversely correlated. We therefore suggest that paracrine PDGF might amplify angiogenesis via direct action on endothelially expressed PDGFR-beta.

Elf-1 is a transcriptional regulator of the Tie2 gene during vascular development.

Vascular development requires the tightly coordinated expression of several growth factors and their receptors. Among these are the Tie1 and Tie2 receptors, which are almost exclusively endothelial cell-specific. The critical transcriptional regulators of vascular-specific gene expression remain largely unknown. The Ets factors are a family of evolutionarily conserved transcription factors that regulate genes involved in cellular growth and differentiation. We have recently shown that the Ets factor NERF is a strong transactivator of the Tie1 and Tie2 genes. To extend these studies, we have begun to identify the Ets factors that are expressed in developing blood vessels of the chicken chorioallantoic membrane (CAM), a highly vascular embryonic network. RNA was extracted from microdissected CAM blood vessels, and reverse transcriptase-polymerase chain reaction was performed using oligonucleotides encoding conserved amino acids within the Ets domain. One of the polymerase chain reaction fragments was subcloned and identified as the chicken homologue of the Ets factor ELF-1, cELF-1. ELF-1 is most closely related to the Ets factor NERF. In situ hybridization and immunohistochemistry demonstrate that cELF-1 is enriched in developing chicken blood vessels. cELF-1 is also a strong transactivator of the Tie1 and Tie2 genes and can bind to conserved Ets sites within the promoters of these genes. A complex of similar size forms when gel shifts are performed with cellular extracts derived from the CAM blood vessels, which is recognized by an antibody against cELF-1. In summary, ELF-1 belongs to a subset of Ets factors that regulate vascular-specific gene expression during blood vessel development.

Transcriptional activity of the alpha 1(I)-collagen promoter is correlated with the formation of capillary-like structures by endothelial cells in vitro.

Bovine aortic endothelial (BAE) cells spontaneously form structures in vitro that resemble capillary-like cords or tubes. This process is associated with changes in the expression of certain extracellular matrix proteins that include type I collagen. BAE cells exhibiting angiogenesis in vitro were transfected with plasmids containing either chloramphenicol acetyltransferase or human growth hormone genes directed by promoter sequences from the human alpha 1(I)-collagen gene. Immunostaining for chloramphenicol acetyltransferase demonstrated that collagen promoter activity was restricted to cells involved in the formation of endothelial cords. In comparison to transfected monolayers of BAE cells, the transcriptional activity of the alpha 1(I)-collagen promoter increased by 7-fold in cultures undergoing angiogenesis in vitro. The selective ability of angiogenic endothelium to utilize the alpha 1(I)-collagen promoter is consistent with previous studies showing high levels of alpha 1(I)-collagen mRNA in BAE cells actively engaged in the formation of tubes (Iruela-Arispe, L., Hasselaar, P., and Sage, H. (1991) Lab. Invest. 64, 174-186). We conclude that transcriptional activation of the alpha 1(I)-collagen gene is closely linked to the morphologic alterations in cellular phenotype that accompany the transition of quiescent endothelial monolayers to the angiogenic state.

Thrombospondin-1 suppresses wound healing and granulation tissue formation in the skin of transgenic mice.

The function of the endogenous angiogenesis inhibitor thrombospondin-1 (TSP-1) in tissue repair has remained controversial. We established transgenic mice with targeted overexpression of TSP-1 in the skin, using a keratin 14 expression cassette. TSP-1 transgenic mice were healthy and fertile, and did not show any major abnormalities of normal skin vascularity, cutaneous vascular architecture, or microvascular permeability. However, healing of full-thickness skin wounds was greatly delayed in TSP-1 transgenic mice and was associated with reduced granulation tissue formation and highly diminished wound angiogenesis. Moreover, TSP-1 potently inhibited fibroblast migration in vivo and in vitro. These findings demonstrate that TSP-1 preferentially interfered with wound healing-associated angiogenesis, rather than with the angiogenesis associated with normal development and skin homeostasis, and suggest that therapeutic application of angiogenesis inhibitors might potentially be associated with impaired wound vascularization and tissue repair.

Expression of thrombospondins by endothelial cells. Injury is correlated with TSP-1.

The thrombospondins (TSP-1, -2, and -3) comprise a family of proteins that are homologous at the carboxy terminus but have unique sequences at the amino terminus that might be correlated with the regulation of cell behavior. To investigate the expression of TSP-1, -2, and -3 in endothelial cells, we examined developing murine blood vessels and human atherosclerotic plaques by in situ hybridization. The expression of TSP-1 was also characterized in cultured bovine aortic endothelial cells. Expression of TSP-2 was seen in the dorsal aorta as early as embryonic day 10; TSP-1 was not detected in endothelial cells until later stages, and TSP-3 was not apparent in the vasculature. In atherosclerotic specimens, TSP-1 mRNA was detected in many intraplaque microvessels and in the endothelium lining the atheromatous plaque; TSP-2 was absent from these regions. Cultured bovine aortic endothelial cells did not transcribe TSP-2 mRNA at detectable levels. There were high steady-state levels of TSP-1 mRNA in subconfluent bovine aortic endothelial cells before confluence and at the wound edge after injury of the cell monolayer, with maximal expression of TSP-1 in cultures at a time during which approximately 35% of the cells were in S phase. As the majority of these cells subsequently undergo mitosis, these data are consistent with TSP-1 as an inhibitor of endothelial cell proliferation that functions in G1. These results support the conclusion that, despite sequence homology, the TSPs have distinct functions in vascular biology.

Altered glomerular extracellular matrix synthesis in experimental membranous nephropathy.

Chronic progressive membranous nephropathy (MN) in humans is characterized by thickening of the glomerular basement membrane (GBM) with formation of spikes which contain laminin and other extracellular matrix (ECM) proteins. We have utilized two models of MN in the rat (active and passive Heymann nephritis, AICN, PHN) to define the sequential changes in composition of GBM as they relate to changes in glomerular gene expression for ECM components, altered permeability and morphological changes. Renal biopsies obtained during the course of AICN and PHN were immunostained for various ECM proteins and total glomerular RNA was hybridized with cDNA probes specific for laminin B2-chain, s-laminin, and types I and IV collagen. In addition, the ability of anti-glomerular epithelial cell (GEC) antibody and complement on rat GEC in culture to induce laminin release or laminin and s-laminin mRNA expression was determined. The results demonstrate that at weeks 12, 16, and 20 of AICN, immunostaining for laminin, s-laminin, fibronectin, entactin, and heparan sulfate proteoglycan increased in the GBM in a spike-like pattern. Concomitantly, glomerular mRNA levels of laminin B2-chain and of s-laminin increased. Type IV collagen protein and gene expression remained unchanged or decreased. No glomerular immunostaining for type I collagen occurred during AICN despite increased expression of mRNA for this collagen type. In contrast to AICN, in PHN no pronounced changes of the glomerular ECM occurred, except for transient expression of type I collagen mRNA in whole glomerular RNA and type I collagen protein the GEC cytoplasm. Stimulation of GEC in culture with anti-GEC antibody and complement also failed to induce transcription of laminin or s-laminin mRNA or the release of laminin protein. These findings suggest that the polyantigenic expansion of GBM which occurs in chronic experimental MN may be stimulated by factors different from the C5b-9 mediated processes that cause the initial proteinuria.

Participation of glomerular endothelial cells in the capillary repair of glomerulonephritis.

In many glomerular diseases severe injury to the mesangium may occur, leading to matrix dissolution and damage to the glomerular capillaries. Although the destruction of glomerular architecture may lead to permanent injury, in some cases spontaneous recovery occurs. The mechanisms that mediate this recovery are unknown. In this study we provide evidence for glomerular capillary repair (angiogenesis) in the adult injured glomerulus. Injection of anti-Thy 1 antibody into rats results in severe mesangiolysis with capillary ballooning, microaneurysm formation, and loss of endothelial cells in addition to mesangial cells. Although mesangial proliferation is a major response to injury, proliferation of endothelial cells also can be documented from days 2 to 14 in association with repair of the capillaries. The endothelial cell proliferation peaks on days 2 and 7, when it is seven- to ninefold greater than normal. Many of the endothelial cells display morphological features of angiogenesis. The initial wave of endothelial cell proliferation can be reduced by 40% with neutralizing anti-basic fibroblast growth factor antibodies (P < 0.001). The later glomerular endothelial cell proliferation is associated with upregulated expression of vascular permeability factor/endothelial cell growth factor (VPF/VEGF) and an increase of flk, a VPF/VEGF receptor. Although PDGF is expressed in this model, anti-PDGF antibody treatment did not affect the endothelial cell proliferative response. In summary, glomerular endothelial cells have an active role in the glomerular response to injury. Glomeruli are capable of healing microaneurysms, and the mechanism involves basic fibroblast growth factor- and VPF/VEGF-mediated endothelial proliferative responses.

Inhibition of angiogenesis and breast cancer xenograft tumor growth by VEGI, a novel cytokine of the TNF superfamily.

Recently, we reported a novel protein of the tumor necrosis factor (TNF) superfamily, named vascular endothelial cell growth inhibitor (VEGI), which is expressed predominantly in endothelial cells. When a secreted form of this new protein was overexpressed in mouse colon cancer cells, the growth of tumors formed by these cells in black mice was inhibited. We now report that recombinant VEGI inhibits the proliferation of endothelial cells but not that of other types of cells examined. The protein also inhibits formation of capillary-like structures by endothelial cells in collagen gels, and the growth of capillaries into collagen gels placed on the chick chorioallantoic membrane. The anticancer potential of VEGI was examined in a breast cancer xenograft tumor model in which the cancer cells were co-injected with Chinese hamster ovary cells overexpressing a secreted form of the protein. The co-injection resulted in potent inhibition of xenograft tumor growth. Our findings are consistent with the view that VEGI is an endothelial cell-specific negative regulator of angiogenesis.

Tissue factor expression during human and mouse development.

In the adult organism the cellular distribution of tissue factor (TF) expression corresponds to biological boundary layers forming a hemostatic barrier ready to activate blood coagulation after tissue injury. Whether TF expression might also play a role in development is unknown. To determine the significance of TF in ontogenesis, we examined the pattern of TF expression in mouse development and compared it with the distribution of TF in human post-implantation embryos and fetuses of corresponding gestational age. At early embryonic periods of murine (6.5 and 7.5 pc) and human (stage 5) development, there was strong expression of TF in both ectodermal and entodermal cells. In situ hybridization and immunohistochemistry demonstrated that TF mRNA and protein were expressed widely in epithelial areas with high levels of morphogenic activity during organogenesis. Staining for TF was seen during ontogenetic development in tissues such as epidermis, myocardium, bronchial epithelium, and hepatocytes, which express TF in the adult organism. Surprisingly, during renal development and in adults, expression of TF differed between humans and mice. In humans, maturing stage glomeruli were stained for TF whereas in mice, TF was absent from glomeruli but was present in the epithelia of tubular segments. In neuroepithelial cells, there was a substantial expression of TF. Moreover, there was robust TF expression in tissues such as skeletal muscle and pancreas, which do not express it in the adult. In contrast, expression of the physiological ligand for TF, factor VII, was not detectable during early stages of human embryogenesis using immunohistochemistry. The temporal and spatial pattern of TF expression during murine and human development supports the contention that TF serves as an important morphogenic factor during embryogenesis.