Phorbol ester induces cultured endothelial cells to invade a fibrin matrix in the presence of fibrinolytic inhibitors

Alternative adhesion sites in human fibrinogen for vascular endothelial cells

Fibrinogen mediates endothelial cell adhesion, spreading, and angiogenesis through integrin alphavbeta3. Previous studies by several investigators have suggested that the Arg-Gly-Asp (RGD) site at position 572-574 on the alpha chain of human fibrinogen can bind to alphavbeta3. However, this RGD sequence is absent in fibrinogen from most other species, including bovine, hamster, monkey, mouse, pig, and rat fibrinogen. In these species, an RGD site exists at the equivalent of position alpha252-254, which has the sequence RGG in humans. In addition, the role of an integrin binding site on the gamma chain at position 400-411 has been an issue of controversy. In the present studies, recombinant fibrinogen molecules with mutations in the potential endothelial cell binding sites have been used to test the role of these sites directly. The results show that the RGD at alpha572-574 is the primary adhesion site, and that the gamma chain site plays no significant role. Human and bovine plasma fibrinogens were also assayed for their ability to support adhesion of human and bovine vascular endothelial cells. The results show that although the two types of fibrinogen have RGD sequences at widely divergent sites, there is no significant difference in their ability to support endothelial cell adhesion. Furthermore, a chimeric human fibrinogen molecule with an RGD sequence at the bovine site, position alpha252-254, also supported adhesion. These results indicate that an RGD site in human fibrinogen at either position alpha252-254 or position alpha572-574 can mediate endothelial cell adhesion.

A microcarrier-based cocultivation system for the investigation of factors and cells involved in angiogenesis in three-dimensional fibrin matrices in vitro

Angiogenesis in situ includes coordinated interactions of various microvascular cell types, i.e., endothelial cells, pericytes and perivascular fibroblasts. To study the cellular interactions of microvascular cells in vitro, we have developed a microcarrier-based cocultivation system. The technical details of this method include seeding of endothelial cells on unstained cytodex-3 microcarriers and seeding of pericytes, fibroblasts or vascular smooth muscle cells on microcarriers which have been labeled by trypan blue staining. A mixture of both unstained and trypan blue-stained microcarriers was subsequently embedded in a three-dimensional fibrin clot. The growth characteristics of each cell type could be conveniently observed since the majority of cells left their supporting microcarriers in a horizontal direction to migrate into the transparent fibrin matrix. As differently stained microcarriers were randomly arranged in the fibrin matrix, the characteristic patterns of the microcarriers allowed location of particular points of interest at different developmental stages, facilitating the observation of cellular growth over the course of time. One further advantage of this microcarrier-based system is the possibility of reliably quantifying capillary growth by determination of average numbers of capillary-like formations per microcarrier. Thus, this model allows convenient evaluation of the effects of non-endothelial cells on angiogenesis in vitro. By using this coculture system, we demonstrate that endothelial capillary-like structures in vitro do not become stabilized by contacting vascular smooth muscle cells or pericytes during the initial stages of capillary formation.

Fibrin II induces endothelial cell capillary tube formation

We studied the formation of capillary tubes by endothelial cells which were sandwiched between two fibrin gels under serum-free conditions. After formation of the overlying fibrin gel, the endothelial cell monolayer rearranged into an extensive net of capillary tubes. Tube formation was apparent at 5 h and was fully developed by 24 h. The capillary tubes were vacuolated, and both intracellular and intercellular lumina were present. Maximal tube formation was observed with fibrin II (which lacks both fibrinopeptide A and B), minimal tube formation with fibrin I (which lacks only fibrinopeptide A), and complete absence of tube formation with fibrin 325 (which lacks the NH2-terminal beta 15-42 sequence, in addition to fibrinopeptides A and B). The inability of fibrin 325 to stimulate capillary tube formation supports the idea that beta 15-42 plays an important role in this process, and its importance was confirmed by the finding that exogenous soluble beta 15-42 inhibited fibrin II-induced capillary tube formation. This effect was specific for fibrin, since beta 15-42 did not inhibit tube formation by endothelial cells sandwiched between collagen gels. The interaction of the apical surface of the endothelial cell with the overlying fibrin II gel, as opposed to the underlying fibrin gel upon which the cells were seeded, was necessary for capillary tube formation. These studies suggest that the beta 15-42 sequence of fibrin interacts with a component of the apical cell surface and that this interaction plays a fundamental role in the induction of endothelial capillary tube formation.

Transcriptional and posttranscriptional regulation of urokinase-type plasminogen activator expression in endothelial cells by basic fibroblast growth factor

The mechanism of induction of urokinase-type plasminogen activator (uPA) by basic fibroblast growth factor (bFGF) was explored in fetal bovine aortic endothelial GM 7373 cells. A three- to four-fold increase in the steady-state levels of uPA mRNA was observed after 6 h of incubation of the cell cultures with bFGF. Accordingly, nuclear run-on experiments showed a 2-2.4-fold increase in the rate of uPA gene transcription during the first 4 h of treatment with the growth factor. bFGF did not affect uPA mRNA stability, as evaluated by chase experiments with the mRNA synthesis inhibitor actinomycin D. Upregulation of uPA mRNA was followed by a delayed increase in uPA protein synthesis paralleled by an increase in secreted and cell-associated uPA activity. Twelve h were required before accumulated uPA mRNA was translated into the corresponding protein. During this time interval, the continuous presence of biologically active bFGF in the extracellular environment represented an absolute requirement for uPA mRNA translation. Substitution of residues Lys-27, Lys-30, and Arg-31 to glutamine residues in the bFGF molecule resulted in a mutant (M1Q-bFGF) that caused uPA mRNA accumulation in the absence of a significant increase in cell-associated uPA activity. M1Q-bFGF also induced an increase in cell-associated uPA activity only when added to the cell cultures in the presence of soluble heparin. These results provide evidence that bFGF can affect uPA expression in endothelial GM 7373 cells both at transcriptional and posttranscriptional translational levels. They also show the possibility to dissociate upregulation of uPA mRNA from upregulation of uPA activity by mutagenesis of the bFGF molecule.

Angiogenesis: models and modulators

Angiogenesis in vivo is distinguished by four stages: subsequent to the transduction of signals to differentiate, stage 1 is defined as an altered proteolytic balance of the cell allowing it to digest through the surrounding matrix. These committed cells then proliferate (stage 2), and migrate (stage 3) to form aligned cords of cells. The final stage is the development of vessel patency (stage 4), generated by a coalescing of intracellular vacuoles. Subsequently, these structures anastamose and the initial flow of blood through the new vessel completes the process. We present and discuss how the available models most closely represent phases of in vivo angiogenesis. The enhancement of angiogenesis by hyaluronic acid fragments, transforming growth factor beta, tumor necrosis factor alpha, angiogenin, okadaic acid, fibroblast growth factor, interleukin 8, vascular endothelial growth factor, haptoglobin, and gangliosides, and the inhibition of the process by hyaluronic acid, estrogen metabolites, genestein, heparin, cyclosporin A, placental RNase inhibitor, steroids, collagen synthesis inhibitors, thrombospondin, fumagellin, and protamine are also discussed.

Decreased levels of alpha 1(I) procollagen mRNA in dermal fibroblasts grown on fibrin gels and in response to fibrinopeptide B

We have investigated human neonatal fibroblast synthetic activity in response to fibrin substrates and components of fibrin formation and degradation. Greater than threefold downregulation of procollagen mRNA levels was seen 24 hours after fibroblasts were grown on fibrin gels as compared to tissue culture plastic. This downregulation occurred in both reptilase-generated fibrin (retention of fibrinopeptide B) and thrombin-generated fibrin (loss of both fibrinopeptide A and B). However, fibroblasts grown on fibrin retained their capacity to respond to the stimulatory action of transforming growth factor (TGF)-beta 1. Fibroblasts seeded on reptilase-generated fibrin displayed an abnormal morphology manifested by dendritic appearance and cell rounding, while fibroblast attachment was enhanced by 30% on thrombin-generated fibrin substrate (P < 0.02). Fibrinopeptides A and B, which are generated during fibrin formation, increased and decreased procollagen mRNA levels, respectively. Tissue plasminogen activator (t-PA) increased procollagen mRNA and TGF-beta 1 levels as early as 6 hours after cells were grown on tissue culture plastic, but this stimulation did not occur in cells cultured on a fibrin substrate. We conclude that alpha 1(I) procollagen mRNA levels in cultures of human dermal fibroblasts are consistently down-regulated by a fibrin substrate and are directly and profoundly influenced by complex interactions between components involved in the formation and removal of fibrin.

Angiogenin promotes invasiveness of cultured endothelial cells by stimulation of cell-associated proteolytic activities

Angiogenin, a potent inducer of neovascularization in the chicken chorioallantoic membrane and rabbit cornea, promotes endothelial cell invasion of Matrigel basement membrane. A transformed bovine aortic endothelial cell line, GM 7373, is 5 times more invasive when cultured in the presence of 1 microgram of bovine angiogenin per ml than in its absence. A polyclonal anti-angiogenin antibody and alpha 2-antiplasmin neutralize the effect of angiogenin, but an angiogenin-binding protein (actin) does not. Further, this concentration of angiogenin induces a 14-fold increase in the cell-associated proteolytic activity of cultured endothelial cells, determined with a tissue-type plasminogen activator-specific peptide as the substrate. In addition, cells cultured on a three-dimensional fibrin gel in the presence of angiogenin are 3 times more capable of dissolving the gel and forming focal defects in the underlying matrix. The results indicate that angiogenin can enhance the ability of endothelial cells to digest extracellular matrix components and degrade basement membrane, thereby facilitating cell invasion and migration. Binding of angiogenin to its cell-surface binding protein (actin) followed by dissociation of the angiogenin-actin complex from the cell surface and subsequent activation of tissue-type plasminogen activator/plasmin are likely steps involved in the processes of endothelial cell invasion and angiogenesis.

Tumor cells secrete an angiogenic factor that stimulates basic fibroblast growth factor and urokinase expression in vascular endothelial cells

Culture medium conditioned by human SK-Hep1 hepatoma cells or mouse S180 sarcoma cells rapidly up-regulates endothelial cell expression of basic fibroblast growth factor (bFGF) and induces formation of capillary-like structures by vascular endothelial cells grown on three-dimensional fibrin gels (in vitro angiogenesis). Incubation of endothelial cells with the tumor cell-conditioned media also results in increased expression of urokinase plasminogen activator (uPA), a key component of the proteolytic system required for cell invasion and capillary formation. Although the tumor cell-conditioned media contain no bFGF, addition of anti-recombinant bFGF IgG abolishes the up-regulation of uPA and blocks in vitro angiogenesis. This indicates that both the increase in uPA production and formation of capillary-like structures are mediated by endogenous bFGF expressed by the endothelial cells. Both the bFGF/uPA-inducing activity and the angiogenic activity of SK-Hep1 cell-conditioned medium copurify with a relatively acid-resistant peptide that has moderate affinity for heparin and M(r) < 18 kDa > 3.5 kDa. Known cytokines with similar biochemical features do not possess the same biological activity. These findings indicate that angiogenesis can be mediated by endothelial cell bFGF through an autocrine mechanism and that the bFGF-inducing peptide may represent a novel tumor-derived angiogenic factor that modulates in endothelial cells the concerted expression of cytokines and proteolytic enzymes required for capillary formation.

Modulation of bovine microvascular endothelial cell proteolytic properties by inhibitors of angiogenesis

A tightly controlled increase in extracellular proteolysis, restricted both in time and space, is an important component of the angiogenic process, while anti-proteolysis is effective in inhibiting angiogenesis. By focussing on the plasminogen activator (PA)-plasmin system, the objective of the present studies was to assess whether previously described inhibitors of angiogenesis modify bovine microvascular endothelial cell proteolytic properties. We demonstrate that although synthetic angiostatic steroids (U-24067 and U-42129), heparin, suramin, interferon alpha-2a, and retinoic acid are all inhibitors of in vitro angiogenesis, each of these agents has distinct effects on the plasminogen-dependent proteolytic system. Specifically, angiostatic steroids and interferon alpha-2a reduce urokinase-type PA (u-PA) and PA inhibitor-1 activity, while heparin and retinoic acid increase u-PA activity. Suramin reduces cell-associated u-PA activity and greatly increases PAI-1 production at doses which induce monolayer disruption. These findings demonstrate that a spectrum of alterations in extracellular proteolysis is associated with anti-angiogenesis, and that anti-angiogenesis and anti-proteolysis are not necessarily correlated. A reduction in extracellular proteolysis would be expected to reduce invasion, whereas an increase in proteolysis might modulate the activity of inhibitory cytokines, which in turn could reduce endothelial cell proliferation and migration and inhibit angiogenesis. The spectrum of effects on different elements of the PA system observed in response to the agents assessed suggests that the role of modulations in extracellular proteolytic activity in anti-angiogenesis is likely to be varied and complex.

Quantitative characterization of cell invasion in vitro: formulation and validation of a mathematical model of the collagen gel invasion assay

An in vitro assay proposed to systematically characterize and compare cell invasion under different conditions is the collagen gel invasion assay where cells, initially seeded onto the surface of a type I collagen gel, penetrate the surface and migrate within the gel over time. Using simplifying assumptions about cell transport across the gel surface and migration within the gel, we formulate and solve a mathematical model of this assay which predicts the resulting cell distribution based on three phenomenological parameters characterizing the ability of cells to penetrate the gel surface interface, migrate randomly within the gel, and return to the gel surface. An index of cell invasiveness is defined based on these parameters that reflects the overall ability of cells to transport across the gel surface interface, that is, invade the gel. Cell concentration profiles predicted by the model correspond well to measured profiles for murine melanoma cells invading gels supplemented with extracellular matrix proteins fibronectin and type IV collagen as well as unsupplemented gels, allowing these parameters to be estimated by a nonlinear regression fit of the model solution to the measured profiles. Our analysis suggests that type IV collagen and fibronectin primarily modulate cell transport across the gel surface interface rather than migration within the gel. Further, we validate the key model assumptions and obtain independent, direct estimates of model parameters by time-lapse video microscopy and digital image analysis of cell penetration of the gel surface and migration within the gel during the assay.

Regulation of in vitro capillary tube formation by anti-integrin antibodies

Human endothelial cells are induced to form an anastomosing network of capillary tubes on a gel of collagen I in the presence of PMA. We show here that the addition of mAbs, AK7, or RMAC11 directed to the alpha chain of the major collagen receptor on endothelial cells, the integrin alpha 2 beta 1, enhance the number, length, and width of capillary tubes formed by endothelial cells derived from umbilical vein or neonatal foreskins. The anti-alpha 2 beta 1 antibodies maintained the endothelial cells in a rounded morphology and inhibited both their attachment to and proliferation on collagen but not on fibronectin, laminin, or gelatin matrices. Furthermore, RMAC11 promoted tube formation in collagen gels of increased density which in the absence of RMAC11 did not allow tube formation. Neither RMAC11 or AK7 enhanced capillary formation in the absence of PMA. Lumen structure and size were also altered by antibody RMAC11. In the absence of antibody the majority of lumina were formed intracellularly from single cells, but in the presence of RMAC11, multiple cells were involved and the lumen size was correspondingly increased. Endothelial cells were also induced to undergo capillary formation in fibrin gels after PMA stimulation. The addition of anti-alpha v beta 3 antibodies promoted tube formation in fibrin gels and inhibited EC adhesion to and proliferation on a fibrinogen matrix. The enhancement of capillary formation by the anti-integrin antibodies was matrix specific; that is, anti-alpha v beta 3 antibodies only enhanced tube formation on fibrin gels and not on collagen gels while anti-alpha v beta 1 antibodies only enhanced tubes on collagen and not on fibrin gels. Thus we postulate that changes in the adhesive nature of endothelial cells for their extracellular matrix can profoundly effect their function. Anti-integrin antibodies which inhibit cell-matrix interactions convert endothelial cells from a proliferative phenotype towards differentiation which results in enhanced capillary tube formation.

Regulation of the endothelial cell urokinase-type plasminogen activator receptor. Evidence for cyclic AMP-dependent and protein kinase C-dependent pathways

Binding of urokinase-type plasminogen activator (u-PA) to specific receptors (u-PAR) on the surface of endothelial cells contributes to the regulation of plasmin-dependent processes such as fibrinolysis and angiogenesis. We studied the effect of raising intracellular levels of cyclic AMP (cAMP) and/or activating protein kinase C on the expression of u-PAR in cultured human umbilical vein endothelial cells (HUVEC). Incubation of HUVEC with forskolin stimulated a time- and concentration-dependent increase in the expression of u-PAR, measured both by an increase in the specific binding of radiolabeled single-chain u-PA (scu-PA) and by increased binding of anti-u-PAR antibodies. Maximal increase in u-PAR expression (81 +/- 11% above control, n = 11) was not associated with a change in receptor affinity for scu-PA when HUVEC were incubated for 20 hours at 37 degrees C with 50 microM forskolin. Receptor induction by forskolin was inhibited when HUVEC were preincubated with deoxyadenosine monophosphate (DAM), an inhibitor of adenylyl cyclase. A similar increase in receptor expression (128 +/- 27% above control, n = 3) was induced by the cAMP analogue 8-bromoadenosine 3':5'-cyclic monophosphate (50 mM). Forskolin induced an approximately twofold increase in the expression of a single approximately 1.4-kb u-PAR messenger RNA (mRNA) transcript within 2 hours. Phorbol myristate acetate (PMA) also stimulated a time- and concentration-dependent increase in specific scu-PA binding. The maximal increase in u-PAR expression (254 +/- 27% above control, n = 11) was observed when HUVEC were preincubated with 10 nM PMA for 20 hours. Induction of u-PAR by PMA was inhibited when HUVEC were preincubated with either cycloheximide or H7 but was unaffected by DAM. u-PAR induced by PMA showed a reduced affinity for scu-PA (Kd, 14 +/- 2 nM versus 3.6 +/- 0.6 nM, p < 0.001; n = 8). PMA stimulation for 20 hours resulted in a sixfold increase in a single approximately 1.4-kb u-PAR mRNA transcript, with increased levels detectable within 30 minutes. Coincubation of HUVEC with optimal concentrations of forskolin and PMA for 20 hours produced a fully additive increase in u-PAR expression at both the mRNA and protein levels. These data suggest that both cAMP-dependent and protein kinase C-dependent protein kinase pathways may independently regulate u-PAR expression in human endothelial cells.

Role of laminin in endothelial cell recognition and differentiation

The vascular endothelium normally is maintained in a quiescent state, but under certain conditions it is induced to undergo marked changes in behavior and form new vascular structures. A complex interaction among various growth and differentiation factors and the extracellular milieu regulates this behavior. One series of signals affecting endothelial behavior is provided by laminin, a major structural protein of basement membrane. These signals have been studied using Matrigel, a reconstituted basement membrane preparation from the murine Englebreth-Holm-Swarm sarcoma, in an in vitro assay of endothelial cell differentiation. Three biologically-active sequences from the laminin molecule have been evaluated. Synthetic peptides that include the sequences -RGD-, -YIGSR-, and -SIKVAV- mediate, respectively, cell binding to Matrigel, alterations in cell morphology, and induction of migration and collagenase activity. Preliminary data indicate that observations made with this system may be relevant to endothelial function in vivo. Endothelial cell differentiation on Matrigel may thus be a useful in vitro model for the study of certain steps in angiogenesis.

Interaction of endothelial cells with a laminin A chain peptide (SIKVAV) in vitro and induction of angiogenic behavior in vivo

Endothelial cells are known to bind to laminin, and two peptides derived from the laminin A (CTFALRGDNP) and B1 (CDPGYIGSR) chains block the capillary-like tube formation on a laminin-rich basement membrane matrix, Matrigel. In the present study, we have used various in vitro and in vivo assays to investigate the angiogenic-biologic effects of a third active site in the laminin A chain, CSRARKQAASIKVAVSADR (designated PA22-2) on endothelial cells. The SIKVAV-containing peptide was as active as the YIGSR-containing peptide for endothelial cell attachment but was less active than either the RGD-containing peptide or intact laminin. Endothelial cells seeded on this peptide appeared fibroblastic with many extended processes, unlike the normal cobblestone morphology observed on tissue culture plastic. In addition, in contrast to normal tube formation on Matrigel, short irregular structures formed, some of which penetrated the matrix and sprouting was more apparent. Analysis of endothelial cell conditioned media of cells cultured in the presence of this peptide indicated degradation of the Matrigel and zymograms demonstrated active collagenase IV (gelatinase) at 68 and 62 Kd. A murine in vivo angiogenesis assay and the chick yolk sac/chorioallantoic membrane assays with the peptide demonstrated increased endothelial cell mobilization, capillary branching, and vessel formation. These data suggest that the -SIKVAV-site may play an important role in initiating branching and formation of new capillaries from the parent vessels, a behavior that is observed in vivo in response to tumor growth or in the normal vascular response to injury.

In vivo patterns of expression of urokinase and its inhibitor PAI-1 suggest a concerted role in regulating physiological angiogenesis

To evaluate the role of plasminogen activators (PAs) in physiological angiogenesis, we have investigated the in vivo patterns of expression of urokinase-type PA (uPA) and PA-inhibitor type 1 (PAI-1) during neovascularization of ovarian follicles, the corpus luteum, and the maternal decidua. Using in situ hybridization, we detected uPA mRNA in the ovary along the route of capillary extension, originating at the existing ovarian vasculature, extending toward growing follicles, and terminating at the newly formed capillary sheaths surrounding each growing follicle. Following ovulation, uPA mRNA was expressed in capillary sprouts within the developing corpus luteum. During the process of decidual neovascularization, uPA expression was detected in endothelial cell cords traversing the maternal decidua in the direction of the newly implanted embryo. uPA mRNA was not detected in endothelial cells upon completion of neovascularization, suggesting that uPA expression is a part of the angiogenic response. During in vitro "angiogenesis" of cultured aortic explants, uPA was expressed in capillary sprouts but not in underlying endothelial cell sheets, suggesting that the expression of uPA depends on the histological context of the endothelial cell. Interestingly, during corpus luteum development and decidual neovascularization, and in aortic explants, PAI-1 expression was preferentially activated in cells in the vicinity of uPA-expressing capillary-like structures. These findings suggest a functional interplay between uPA- and PAI-1-expressing cells and support the idea that natural PA inhibitors function during angiogenesis to protect neovascularized tissues from excessive proteolysis.

Plasminogen activator inhibitor-1 is induced in migrating endothelial cells

It has been proposed that a finely tuned protease-anti-protease equilibrium must be maintained during processes of cell migration in order to limit extracellular proteolysis to the close proximity of the cell surface, and thereby to prevent excessive extracellular matrix degradation. We have previously shown that urokinase-type plasminogen activator (u-PA) activity is induced in microvascular endothelial cells migrating from the edges of a wounded monolayer in vitro (Pepper et al., J. Cell Biol., 105:2535-2541, 1987). By Northern analysis, we now demonstrate that plasminogen activator inhibitor 1 (PAI-1) mRNA is increased in multiple-wounded monolayers of bovine microvascular (BME) or aortic (BAE) endothelial cells, with a maximal 7- and 9-fold increase 4 h after wounding, respectively. By in situ hybridization, we demonstrate that the increase in PAI-1 mRNA is localized to cells at the edge of a wounded BME or BAE cell monolayer. The increase in PAI-1 mRNA observed in BME cells is independent of cell division and is inhibited by antibodies to basic fibroblast growth factor (bFGF), suggesting that PAI-1 induction in migrating cells is mediated by the autocrine activity of bFGF. Taken together with our previous observations on the induction of u-PA, these results support the hypothesis that the proteolytic balance in the pericellular environment of migrating cells is regulated through the concomitant production of proteases and protease inhibitors.

Basic fibroblast growth factor-induced activation of latent transforming growth factor beta in endothelial cells: regulation of plasminogen activator activity

Exposure of bovine aortic or capillary endothelial cells to basic FGF (bFGF) for 1 h resulted in an approximately sixfold increase in plasminogen activator (PA) activity by 18 h that returned nearly to basal levels by 36 h. We hypothesized that the decrease in PA activity following bFGF stimulation was mediated by transforming growth factor beta (TGF-beta) formed from its inactive precursor. Conditioned medium collected from endothelial cells 36 h after a 1-h exposure to bFGF, but not control medium, inhibited basal levels of PA activity when transferred to confluent monolayers of bovine aortic endothelial cells. Antibody to TGF-beta neutralized the inhibitory activity of this conditioned medium, indicating that the medium contained active TGF-beta. Northern blot analysis and quantitation of acid activatable latent TGF-beta in conditioned medium demonstrated that bFGF exposure did not increase the amount of transcription or secretion of latent TGF-beta by the endothelial cells. Both aprotinin, an inhibitor of plasmin, and anti-urokinase type PA IgG blocked the generation of active TGF-beta in cultures exposed to bFGF. These results demonstrated that plasmin generated by uPA activity is required for the activation of latent TGF-beta in endothelial cell cultures treated with bFGF. Activation of TGF-beta by endothelial cells exposed to bFGF appears to limit both the degree and duration of PA stimulation. Thus, in bFGF-stimulated endothelial cell cultures, PA levels are controlled by a negative feedback loop: PA, whose expression is stimulated by bFGF, contributes to the formation of TGF-beta, which in turn opposes the effects of bFGF by limiting PA synthesis and activity. These studies suggest a role for TGF-beta in reversing the invasive stage of angiogenesis and contributing to the formation of quiescent capillaries.

Protein kinase C regulates endothelial cell tube formation on basement membrane matrix, Matrigel

Human umbilical vein endothelial cells differentiate within 12 h to form capillary-like networks of tube structures when the cells are plated on Matrigel, a mixture of basement membrane proteins. Nothing is known about the intracellular signaling events involved in this differentiation. As a first step to define the process, we investigated the possible role of protein kinase C activation by beta-phorbol 12-myristate 13-acetate (PMA) in regulating the formation of the tube structures. In this model, PMA increased tube formation several-fold in a dose-dependent manner with half-maximum stimulation of tube formation at approximately 5 nM PMA. In the absence of serum, essentially little or no tubes were formed on Matrigel unless PMA was added to the medium. Only active phorbol analogs increased tube formation, while the protein kinase C inhibitor, H-7, blocked tube formation. The protein kinase C activators and inhibitors were effective only when added at or just after plating of the cells and did not affect already formed tubes. This study suggests that protein kinase C is involved in the early events of in vitro endothelial cell tube formation on Matrigel.

Intracellular mechanisms involved in basement membrane induced blood vessel differentiation in vitro

The extracellular matrix, particularly basement membranes, plays an important role in angiogenesis (blood vessel formation). Previous work has demonstrated that a basement membranelike substrate (Matrigel) induces human umbilical vein endothelial cells to rapidly form vessel-like tubes (Kubota, et al., 1988; Grant et al., 1989b); however, the precise mechanism of tube formation is unclear. Using this in vitro model, we have investigated morphologic changes occurring during tube formation and the cytoskeletal and protein synthesis requirements of this process. Electron microscopy showed that endothelial cells attach to the Matrigel surface, align, and form cylindrical structures that contain a lumen and polarized cytoplasmic organelles. The cytoskeleton is reorganized into bundles of actin filaments oriented along the axis of the tubes and is located at the periphery of the cells. The addition of colchicine or cytochalasin D blocked tube formation, indicating that both microfilaments and microtubules are involved in this process. Cycloheximide blocked tube formation by 100%, indicating that the process also required protein synthesis. In particular, collagen synthesis seems to be required for tube formation because cis-hydroxyproline inhibited tube formation, whereas either the presence of ascorbic acid or the addition of exogenous collagen IV to the Matrigel increased tube formation. Our results indicate that endothelial cell attachment to Matrigel induces the reorganization of the cytoskeleton and elicits the synthesis of specific proteins required for the differentiated phenotype of the cells.

Chondrocytes inhibit endothelial sprout formation in vitro: evidence for involvement of a transforming growth factor-beta

Using a quantitative in vitro model of spontaneous endothelial sprout formation, we have attempted to define physiological inhibitors of angiogenesis from hyaline cartilage, a tissue whose antiangiogenic properties have been well described. The model consists of embedding bovine microvascular endothelial cell aggregates into fibrin or collagen gels, which results in the formation of radially growing sprouts. When chondrocytes derived from the permanent cartilagenous region of the chick embryo sternum are cocultured with the endothelial cell aggregates, sprout formation is markedly inhibited. Addition of anti-TGF-beta antibodies to the cocultures significantly reduced the inhibitory effect of chondrocytes on sprout formation. Chondrocyte-conditioned medium or exogenously added TGF-beta 1 have a similar albeit transient inhibitory effect. Depletion of TGF-beta from chondrocyte conditioned medium with anti-TGF-beta antibodies and solid-phase protein-A significantly decreases the inhibition of sprout formation. These results demonstrate that a chondrocyte-derived TGF-beta-like molecule inhibits capillary sprout formation in vitro and suggest that the antiangiogenic properties of cartilage may at least in part, be mediated by TGF-beta.

Angiogenesis in arteries: review

The inner parts of the walls of large blood vessels do not normally contain intrinsic vasculature. In pathologic conditions such as arteriosclerosis or thrombosis, angiogenesis occurs, and may have significant clinical consequences. This review attempts to relate the little that is known about the factors specific to vascular walls which regulate angiogenesis to more general knowledge of the phenomenon.

Proteolytic balance and capillary morphogenesis

Angiogenesis is the process by which new capillary blood vessels are formed from preexisting vessels. A number of components of this morphogenetic process, including endothelial cell invasion and capillary lumen formation, are believed to be dependent on tightly controlled proteolytic degradation of the extracellular matrix. The critical importance of an appropriate balance between proteases and protease inhibitors in these processes is suggested by two sets of observations. Firstly, that extracellular matrix invasion and capillary lumen formation are inhibited in the presence of an excess of protease inhibitors. Secondly, that when unchecked by protease inhibitors, excessive proteolysis is incompatible with normal capillary morphogenesis. These results clearly suggest that a precisely regulated proteolytic balance is necessary for normal capillary morphogenesis.

Increased proteolytic activity is responsible for the aberrant morphogenetic behavior of endothelial cells expressing the middle T oncogene

Expression of the polyoma virus middle T (mT) oncogene in vivo is associated with a profound subversion of normal vascular development, which results in the formation of endothelial tumors (hemangiomas). In an attempt to understand the molecular mechanisms responsible for this phenomenon, we have investigated, in an in vitro system, the morphogenetic properties of endothelial cells expressing this oncogene. mT-expressing endothelioma (End) cells grown within fibrin gels formed large hemangioma-like cystic structures. All End cell lines examined expressed high levels of fibrinolytic activity resulting from increased production of urokinase-type plasminogen activator and decreased production of plasminogen activator inhibitors. Neutralization of excess proteolytic activity by exogenously added serine protease inhibitors corrected the aberrant in vitro behavior of End cells and allowed the formation of capillary-like tubules. These results suggest that tightly controlled proteolytic activity is essential for vascular morphogenesis and that physiological protease inhibitors play an important regulatory role in angiogenesis.

Transforming growth factor-beta 1 modulates basic fibroblast growth factor-induced proteolytic and angiogenic properties of endothelial cells in vitro

Tightly controlled proteolytic degradation of the extracellular matrix by invading microvascular endothelial cells is believed to be a necessary component of the angiogenic process. We have previously demonstrated the induction of plasminogen activators (PAs) in bovine microvascular endothelial (BME) cells by three agents that induce angiogenesis in vitro: basic FGF (bFGF), PMA, and sodium orthovanadate. Surprisingly, we find that these agents also induce plasminogen activator inhibitor-1 (PAI-1) activity and mRNA in BME cells. We also find that transforming growth factor-beta 1 (TGF-beta 1), which in vitro modulates a number of endothelial cell functions relevant to angiogenesis, also increases both PAI-1 and urokinase-type PA (u-PA) mRNA. Thus, production of both proteases and protease inhibitors is increased by angiogenic agents and TGF-beta 1. However, the kinetics and amplitude of PAI-1 and u-PA mRNA induction by these agents are strikingly different. We have used the ratio of u-PA:PAI-1 mRNA levels as an indicator of proteolytic balance. This ratio is tilted towards enhanced proteolysis in response to bFGF, towards antiproteolysis in response to TGF-beta 1, and is similar to that in untreated cultures when the two agents are added simultaneously. Using an in vitro angiogenesis assay in three-dimensional fibrin gels, we find that TGF-beta 1 inhibits the bFGF-induced formation of tube-like structures, resulting in the formation of solid endothelial cell cords within the superficial parts of the gel. These results suggest that a net positive proteolytic balance is required for capillary lumen formation. A novel perspective is provided on the relationship between extracellular matrix invasion, lumen formation, and net proteolytic balance, thereby reflecting the interplay between angiogenesis-modulating cytokines such as bFGF and TGF-beta 1.

Bimodal relationship between invasion of the amniotic membrane and plasminogen activator activity

Three human tumor cell lines, Bowes' melanoma, HT1080 and Osmond cells, were characterized for their ability to invade the amniotic membrane and their production of plasminogen activator. Bowes' melanoma cells, which release large amounts of tissue plasminogen activator (tPA), were poorly invasive on the amniotic membrane. The addition of plasmin inhibitors, anti-tPA antibody or tissue inhibitor of metalloproteinase (TIMP) to the amnion assay enhanced invasiveness. The depletion of plasminogen from the growth medium also enhanced the degree of invasiveness. Similarly, HT1080 cells, which produce high levels of urokinase-type plasminogen activator (uPA), were poorly invasive under standard conditions but invasion was enhanced by plasmin inhibitors or anti-uPA antibodies. Conversely, Osmond cells, which produce low levels of uPA, were very invasive on the amniotic membrane. Invasion by these cells was blocked by the addition of plasmin inhibitors or anti-uPA antibodies to the amnion assay. These results suggest that invasion requires only a minimum level of PA activity and that, as PA production exceeds this optimal level, the degree of invasion decreases. We propose that high levels of plasmin, generated by the tPA or uPA secreted by the cells, may cause uncontrolled matrix degradation and interrupt the interaction of cells and matrix in the early stages of invasion. The inhibition of excessive plasmin activity may stabilize and increase cell matrix contacts and result in an enhancement of invasion.

Plasminogen activator activity of normal and retinoic acid-treated post-implantation embryos

Urokinase-type plasminogen activator (uPA) has been implicated in cellular migration accompanying different biological phenomena including organogenesis. An increase in uPA activity was observed in mouse post-implantation embryos during the early organogenesis period. Since we have previously shown that all-trans retinoic acid (RA) prevented the induction of uPA in mouse peritoneal macrophages, we have now assessed whether teratogenic doses of this agent could also interfere with uPA activity in mouse embryo in vitro and in vivo. Post-implantation embryos (8.5 days) were incubated for up to 24 hr with micromolar concentration of RA resulting in the occurrence of malformations. No significant difference in uPA activity was found between control and treated embryos. Likewise, uPA activity was not altered in embryos explanted on day 9.5 from dams treated 24 hr before with a teratogenic dose of RA. This study indicates that the teratogenic activity of RA is not caused by an inhibition of the induction of uPA in embryos.

Factors affecting the migration and growth of endothelial cells from microvessels of bovine retina

We have developed a retinal microvessel culture system which permits study of the initial events of endothelial cell activation and migration during the angiogenic response. Enzyme digest experiments indicate that Type IV collagen is the major basement membrane component regulating migration and growth of endothelial cells. Following removal of basement membrane collagen, further cell migration and proliferation require a suitable substrate. Laminin, fibronectin and fibrin(ogen) provide excellent substrates for endothelial cell outgrowths while Type I collagen, even if prepared as a three-dimensional gel, or Type IV collagen fails to promote typical cell growth. In contrast to fibrin and fibronectin, plasmin was a poor substrate for cell outgrowth and it is suggested that cell-associated protease activity may exert a regulatory role over endothelial cell-matrix interactions during cell migration.

Correlation of cell migration, cell invasion, receptor number, proteinase production, and basic fibroblast growth factor levels in endothelial cells

The levels of endogenous basic fibroblast growth factor (bFGF) in seven clones of cultured bovine capillary endothelial (BCE) cells were assayed, and their relation to cell morphology, bFGF receptor number, cell migration, amniotic membrane invasivity, and proteinase levels were studied. Immunoblotting experiments with anti-bFGF IgG demonstrated that cells from these clones contained different amounts of bFGF. The cells containing high levels of bFGF had a spindle or elongated appearance at confluence and a low number of high affinity receptors for bFGF. The cells containing low levels of bFGF had a cobblestone-like appearance and a higher number of high affinity receptors. When exposed to 10 ng/ml bFGF, cells containing a low level of bFGF took on an elongated appearance with a crisscross pattern similar to that seen with the high producer bFGF cells. The endogenous bFGF levels of the BCE cell clones correlated with the extent of cell migration after wounding of a monolayer and the degree of invasion of the human amniotic membrane. Cells from the clone with the highest endogenous bFGF level migrated well, invaded the amnion membrane without the addition of exogenous bFGF, and were relatively unaffected by the addition of bFGF. Cells from the clone containing the lowest level of bFGF did not migrate or invade under normal conditions. However, the addition of bFGF to the culture medium strongly enhanced both of these processes. The inclusion of anti-bFGF IgG in the media suppressed cell migration and invasion. The plasminogen activator (PA) activities of cell lysates of the clones, assayed by the 125I-fibrin plate technique, indicated that the PA levels did not correlate with the bFGF levels. Metalloproteinase activities in the conditioned medium, assayed by gelatin zymography, correlated with the endogenous bFGF levels, suggesting that the degree of expression of metalloproteinases might be critical for cell migration and invasion. These data suggest that endogenous bFGF may have an important role for migration and invasion of BCE cells during neovascularization via the induction and/or activation of specific metalloproteinases.

Fibronectin and fibrinolysis are not required for fibrin gel contraction by human skin fibroblasts

Human skin fibroblasts contracted fibrin gels in a time- and cell-dependent manner. Under optimal conditions, gel contraction amounted to more than 50% in 2 hr. Fibronectin did not promote contraction, and fibrinolysis was not required for contraction, although gels contracted without serum or aprotinin were lysed. Before contraction, fibrin was present in loosely packed, randomly organized fibrils. After contraction, the fibrils were more densely packed and aligned in the plane of cell spreading. Cycloheximide treatment of fibroblasts inhibited gel contraction in serum-free medium but not in serum-containing medium. Fibronectin could not substitute for serum in overcoming the cycloheximide effect. Binding sites for fibrin were distributed randomly over the cells' surfaces based on electron microscopic observations. Often small groups of fibrils were localized in indentations at the cell surface. Finally, peptides containing the arg-gly-asp-ser sequence inhibited gel contraction.

In vitro angiogenesis on the human amniotic membrane: requirement for basic fibroblast growth factor-induced proteinases

The role of basic fibroblast growth factor-(bFGF) induced proteinases in basement membrane (BM) invasion by bovine capillary endothelial (BCE) cells was studied using a quantitative in vitro assay previously described (Mignatti et al., 1986). 125I-iododeoxyuridine-labeled BCE cells were grown for 72 h on the human amnion BM, and cell invasion was determined by measuring the radioactivity associated with the tissue after removal of the noninvasive cell layer. BCE cells were noninvasive under normal conditions. Addition of human bFGF to either the BM or to the stromal aspect of the amnion induced BCE cell invasion with a dose-dependent response. This effect was maximal in the presence of 70 ng/ml bFGF, and was inhibited by anti-FGF antibody. Transforming growth factor beta, as well as plasmin inhibitors and anti-tissue type plasminogen activator antibody inhibited BCE cell invasion. The tissue inhibitor of metalloproteinases, 1-10 phenanthroline, anti-type IV and anti-interstitial collagenase antibodies had the same effect. On the contrary, anti-stromelysin antibody and Eglin, an inhibitor of elastase, were ineffective. The results obtained show that both the plasminogen activator-plasmin system and specific collagenases are involved in the invasive process occurring during angiogenesis.

Phorbol ester stimulates macrophage invasion of fibrin matrices

Macrophages migrate through a fibrin-rich extracellular matrix in chronic inflammation, wound healing, and other pathophysiological processes. To investigate the factors that might influence the ability of mononuclear phagocytes to invade fibrin matrices, we cultured macrophage-like P388D1 cells as well as resident and thioglycollate-elicited mouse peritoneal macrophages on three-dimensional fibrin gels, and we examined the effect of agents known to stimulate a variety of macrophage functions, including the production of fibrinolytic enzymes. Cells grown on fibrin gels under control conditions, as well as cells treated with either bacterial lipopolysaccharide or concanavalin A, remained confined to the gel surface. In contrast, the tumor promoter 4 beta-phorbol 12-myristate 13-acetate (PMA) induced both P388D1 cells and peritoneal macrophages to invade the underlying fibrin matrix. The invasive behavior of PMA-treated P388D1 cells was not affected by protease inhibitors of various specificities. These results demonstrate that certain exogenous signals can profoundly modify the ability of macrophages to migrate through fibrin matrices.

Urokinase-type plasminogen activator is induced in migrating capillary endothelial cells

Cellular migration is an essential component of invasive biological processes, many of which have been correlated with an increase in plasminogen activator production. Endothelial cell migration occurs in vivo during repair of vascular lesions and angiogenesis, and can be induced in vitro by wounding a confluent monolayer of cells. By combining the wounded monolayer model with a substrate overlay technique, we show that cells migrating from the edges of an experimental wound display an increase in urokinase-type plasminogen activator (uPA) activity, and that this activity reverts to background levels upon cessation of movement, when the wound has closed. Our results demonstrate a direct temporal relationship between endothelial cell migration and uPA activity, and suggest that induction of uPA activity is a component of the migratory process.