Growth of brain microvessel endothelial cells on collagen gels: Applications to the study of blood-brain barrier physiology and CNS inflammation

Summary Brain microvessel endothelial cells (BMEC) exhibit the tendency to migrate through 3.0-vm pore semipermeable inserts and establish monolayers on both apical and basal filter surfaces. This can potentially lead to complications in accurately assessing a wide variety of physiologic parameters uniquely associated with these cells. To avoid this problem, we have explored growing BMEC on Transwell filters coated with hydrated collagen gels. BMEC seeded on such gels grow as a monolayer until confluency, but do not invade the subendothelial collagen matrix or the underlying support filter. Furthermore, BMEC grown in this manner exhibit biochemical, morphologic, and electrophysiologic properties reflective of the endothelial cells that comprise the blood-brain barrier in vivo. Although the collagen gel acts as an impenetrable barrier to BMEC, and thus ensures the growth of only a single layer of cells, it nevertheless can be infiltrated by monocytes that have been stimulated by a chemotaxin to undergo diapedesis. Thus, growing BMEC on collagen gel-coated Transwells has broad applications for the in vitro study of both blood-brain barrier physiology as well as the mechanisms underlying central nervous system inflammation.     

Contraction of fibrillar type I collagen by endothelial cells: A study in vitro

The formation of microvascular sprouts during angiogenesis requires that endothelial cells move through an extracellular matrix. Endothelial cells that migrate in vitro generate forces of traction that compress (i.e., contract) and reorganize vicinial extracellular matrix, a process that might be important for angiogenic invasion and morphogenesis in vivo. To study potential relationships between traction and angiogenesis, we have measured the contraction of fibrillar type I collagen gels by endothelial cells in vitro. We found that the capacity of bovine aortic endothelial (BAE) cells to remodel type I collagen was similar to that of human dermal fibroblasts—a cell type that generates high levels of traction. Contraction of collagen by BAE cells was stimulated by fetal bovine serum, human plasma-derived serum, bovine serum albumin, and the angiogenic factors phorbol myristate acetate and basic fibroblast growth factor (bFGF). In contrast, fibronectin and immunoglobulin from bovine serum, several nonserum proteins, and polyvinyl pyrrolidone (a nonproteinaceous substitute for albumin in artificial plasma) were not stimulatory. Contraction of collagen by BAE cells was diminished by an inhibitor of metalloproteinases (1, 10-phenanthroline) at concentrations that were not obviously cytotoxic. Zymography of proteins secreted by BAE cells that had contracted collagen gels revealed matrix metalloproteinase 2. Subconfluent BAE cells that were migratory and proliferating were more effective contractors of collagen than were quiescent, confluent cells of the same strain. Moreover, bovine capillary endothelial cells contracted collagen gels to a greater degree than was seen with BAE cells. Collectively, our observations indicate that traction-driven reorganization of fibrillar type I collagen by endothelial cells is sensitive to different mediators, some of which, e.g., bFGF, are known regulators of angiogenesis in vivo. © 1996 Wiley-Liss, Inc.     

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

We have previously shown that the tumor promoter 4 beta-phorbol 12-myristate 13-acetate (PMA) induces capillary endothelial cells grown to confluency on the surface of three-dimensional collagen gels to invade the underlying matrix and to form capillary-like tubular structures, a phenomenon mimicking angiogenic processes that occur in vivo (Montesano and Orci: Cell, 42:469-477, 1985). Since angiogenesis frequently occurs within a fibrin-rich extracellular matrix, we have examined the ability of PMA-treated endothelial cells to invade fibrin gels. Control endothelial cells grown on fibrin gels formed a confluent monolayer on the gel surface and did not invade the underlying matrix. Treatment of the cultures with PMA resulted in a progressive lysis of the substrate without invasion of the fibrin matrix. However, if the cells were treated with PMA either in the presence of fibrinolytic inhibitors (Trasylol, epsilon-aminocaproic acid) or in the absence of detectable plasminogen, dissolution of the substrate was prevented, and the endothelial cells invaded the fibrin gel, forming vessel-like tubular structures similar to those previously observed with collagen gels. These results demonstrate that the invasive and morphogenetic events induced by PMA do not necessarily require an interaction between endothelial cells and collagen fibrils but can also occur with other biologically relevant substrata. They also suggest (1) that invasion may occur via a plasmin-independent mechanism and (2) that in vivo, neutralization of excess proteolytic activity may play an important permissive role in angiogenesis and other invasive processes by preventing uncontrolled matrix degradation.     

In vitro rapid organization of endothelial cells into capillary-like networks is promoted by collagen matrices

We have studied the behavior of cloned capillary endothelial cells grown inside a three dimensional collagen matrix. Cell monolayers established on the surface of collagen gels were covered with a second layer of collagen. This induced the monolayers of endothelial cells to reorganize into a network of branching and anastomosing capillary-like tubes. As seen by electron microscopy, the tubes were formed by at least two cells (in transverse sections) delimiting a narrow lumen. In addition, distinct basal lamina material was present between the abluminal face of the endothelial cells and the collagen matrix. These results showed that capillary endothelial cells have the capacity to form vessel-like structures with well-oriented cell polarity in vitro. They also suggest that an appropriate topological relationship of endothelial cells with collagen matrices, similar to that occurring in vivo, has an inducive role on the expression of this potential. This culture system provides a simple in vitro model for studying the factors involved in the formation of new blood vessels (angiogenesis).     

The relative magnitudes of endothelial force generation and matrix stiffness modulate capillary morphogenesis in vitro

When suspended in collagen gels, endothelial cells elongate and form capillary-like networks containing lumens. Human blood outgrowth endothelial cells (HBOEC) suspended in relatively rigid 3 mg/ml floating collagen gels, formed in vivo-like, thin, branched multi-cellular structures with small, thick-walled lumens, while human umbilical vein endothelial cells (HUVEC) formed fewer multi-cellular structures, had a spread appearance, and had larger lumens. HBOEC exert more traction on collagen gels than HUVEC as evidenced by greater contraction of floating gels. When the stiffness of floating gels was decreased by decreasing the collagen concentration from 3 to 1.5 mg/ml, HUVEC contracted gels more and formed thin, multi-cellular structures with small lumens, similar in appearance to HBOEC in floating 3 mg/ml gels. In contrast to floating gels, traction forces exerted by cells in mechanically constrained gels encounter considerable resistance. In constrained collagen gels (3 mg/ml), both cell types appeared spread, formed structures with fewer cells, had larger, thinner-walled lumens than in floating gels, and showed prominent actin stress fibers, not seen in floating gels. These results suggest that the relative magnitudes of cellular force generation and apparent matrix stiffness modulate capillary morphogenesis in vitro and that this balance may play a role in regulating angiogenesis in vivo.     

Induction of cysteine-rich motor neuron 1 mRNA expression in vascular endothelial cells

Cysteine-rich motor neuron 1 (CRIM1) is expressed in vascular endothelial cells and plays a crucial role in angiogenesis. In this study, we investigated the expression of CRIM1 mRNA in human umbilical vein endothelial cells (HUVECs). CRIM1 mRNA levels were not altered in vascular endothelial growth factor (VEGF)-stimulated monolayer HUVECs or in cells in collagen gels without VEGF. In contrast, the expression of CRIM1 mRNA was elevated in VEGF-stimulated cells in collagen gels. The increase in CRIM1 mRNA expression was observed even at 2 h when HUVECs did not form tubular structures in collagen gels. Extracellular signal-regulated kinase (Erk) 1/2, Akt and focal adhesion kinase (FAK) were activated by VEGF in HUVECs. The VEGF-induced expression of CRIM1 mRNA was significantly abrogated by PD98059 or PF562271, but was not affected by LY294002. These results demonstrate that CRIM1 is an early response gene in the presence of both angiogenic stimulation (VEGF) and environmental (extracellular matrix) factors, and Erk and FAK might be involved in the upregulation of CRIM1 mRNA expression in vascular endothelial cells.     

Suramin inhibits C6 glioma-induced angiogenesis in vitro

Aspects of tumor-induced angiogenesis in vitro were examined using an assay involving collagen gel invasion by a surface monolayer of bovine endothelial cells under the influence of serum free conditioned medium produced by C6 cells, an experimentally derived rat glial tumor cell line. The effects of the polyanionic compound suramin, known to interfere with growth factor/cell signaling on this process were evaluated. Collagen gel invasion was quantified by adding C6 conditioned medium with or without various doses of suramin to monolayers of bovine aortic endothelial cells grown on type I collagen gels in transwell inserts. Cultures were monitored with phase-contrast microscopy. After various periods of incubation collagen gels were fixed, embedded in epoxy resin, and 1-μm thick sections were stained with toluidine blue. Additional cultures were used to evaluate the effects of C6 conditioned medium and suramin on endothelial cell proliferation, and on chemotaxis through 8-μm pores. C6 glioma cell conditioned medium induced large vessel endothelial cells to sprout into the underlying collagen matrix and subsequently from networks of capillary like tubes. Conditioned medium was also chemotactic and mitogenic for these cells. The addition of suramin to C6 glioma conditioned medium prevents tube formation in collagen gels, and inhibits both endothelial cell proliferation and chemotaxis in a dose dependent manner. These results suggest that glial tumor cell conditioned medium induces angiongenesis in large vessel endothelial cells in vitro via mechanisms which are disrupted by suramin, most likely involving tumor-derived growth factor release and/or endothelium-mediated matrix proteolysis.     

Induction of Endothelial Cell Differentiation in Vitro by Fibroblast-Derived Soluble Factors

Recent studies have suggested that fibroblasts, widely distributed mesenchymal cells, not only function to sustain various organs and tissues as stroma cells but also act directly to regulate adjacent cell behavior including migration, proliferation, and differentiation. Since fibroproliferative diseases and lesions (fibroplasia) are accompanied by new capillary growth (angiogenesis), we hypothesized that fibroblasts may have direct effects on endothelial cell behavior, independent of the elaboration of extracellular matrix, that are relevant to complex process of angiogenesis. To test this hypothesis, bovine aortic endothelial cells were cocultured in collagen gels with human skin fibroblasts. This coculture system caused the endothelial cells to become spindle shaped and to organize into a capillary-like structure within the collagen gels. We found that fibroblast-conditioned medium (FCM) also induced endothelial cells initially to elongate and subsequently to organize into a capillary-like structure within collagen gels. While FCM had no significant effect on endothelial cell DNA synthesis, the soluble factor(s) in FCM increased endothelial cell motility in an in vitro wound assay and in a Boyden chamber assay. The chemoattractant(s) in FCM was alkaline (pH 9.0)—and acid (pH 3.0)—stable, relatively heat stable (stable at 60°C for 30 min, unstable at 98°C for 3 min), dithiothreitol (DTT)-sensitive, and bound to an anionic exchange resin (DEAE-cellulose). Another factor(s) stimulated endothelial cell reorganization into capillary-like structure both within a collagen gel and on a reconstituted basement membrane matrix, Matrigel. This factor(s) was alkaline (pH 9.0)—and acid (pH 3.0)—stable, heat (98°C for 3 min)stable, and DTT-sensitive and bound an anionic exchange resin (DEAE-cellulose). These in vitro results suggest that fibroblasts secrete soluble factors that can influence endothelial cell behaviors relevant to the angiogenesis process with possible implications for vascularization in fibroproliferative conditions.     

Three-dimensional collagen matrices induce delayed but sustained activation of gelatinase A in human endothelial cells via MT1-MMP

Gelatinase A, a member of the matrix metalloproteinase (MMP) family, plays an important role during angiogenesis. It is constitutively expressed by human endothelial cells as a latent enzyme and requires activation. Thrombin is the only described physiological inducer of gelatinase A in human endothelial cells. In this study, we investigated the mechanisms of gelatinase A activation by another physiological inducer, collagen. Endothelial cells were cultured on various ECM components for 24 h and the conditioned media were assessed for gelatinase A activity using gelatin zymography. The results demonstrated that type I collagen matrix specifically activates gelatinase A after 24 h in human umbilical vein and 48 h in neonatal foreskin endothelial cells. In contrast, thrombin activated gelatinase A after only 2 h. Activation by collagen was sustained over long periods of time in culture (96 h). Unlike thrombin-induced activation, collagen required active membrane type 1-MMP (MT1-MMP) on the endothelial cell surface to activate gelatinase A. In addition, collagen-induced activation of gelatinase A was inhibited by antibodies to the integrin receptor, alpha(2)beta(1), but not alpha(3)beta(1). Our findings, that collagen can provide long-term activation of gelatinase A are likely to be relevant to endothelial cell invasion during angiogenesis.     

Paracrine regulation of angiogenesis by different cell types in the aorta ring model

The development of blood vessels during angiogenesis is the result of paracrine interactions between tube-forming endothelial cells and angiogenic factor-producing nonendothelial cells. This process can be reproduced and studied under chemically defined culture conditions by culturing vascular explants in three-dimensional gels of extracellular matrix. Rings of rat or mouse aorta cultured in collagen, fibrin or basement membrane gels produce angiogenic outgrowths composed of a mixed population of endothelial cells and nonendothelial cells. Aortic angiogenesis is regulated by endogenous angiogenic factors, inflammatory cytokines, chemokines, extracellular matrix molecules, and proteolytic enzymes produced by cells of the vessel wall in response to the injury of the dissection procedure. In this paper, we review how macrophages, mural cells and fibroblasts regulate different stages of the angiogenic process, from the formation of immature endothelial sprouts to the reabsorption of the neovessels. We also describe how aortic cultures can be used to study interactions between angiogenic outgrowths and nonvascular cell types such as bone marrow macrophages, platelets or cancer cells. Morphologic, genetic and functional studies of this model have provided invaluable information on how vessels form, mature, interact with nonvascular cell types, and are eventually reabsorbed. Further analysis of the paracrine cross-talk between aortic endothelial and nonendothelial cells is likely to provide new insights into the angiogenic process and its key mechanisms.     

Fibronectin promotes the elongation of microvessels during angiogenesis in vitro

Fibronectin is a component of the extracellular matrix of developing microvessels whose role in angiogensis is poorly understood. This study evaluated the effect of plasma fibronectin on angiogenesis in serum-free collagen gel culture of rat aorta. Aortic explants embedded in collagen gels generated microvascular outgrowths. Fibronectin incorporated in the collagen gel promoted a selective dose-dependent elongation of the newly formed microvessels without stimulating vascular proliferation. The fibronectin-treated microvessels were longer due to a proportional increase in the number of microvascular cells. However, fibronectin had no effect on microvascular DNA synthesis and mitotic activity. Fibronectin stimulated microvascular length also in cultures in which mitotic activity was suppressed and angiogenesis was markedly reduced by pretreating the aortic explants with mitomycin C. The synthetic peptide Gly-Arg-Asp-(GRGDS), which competes for the binding of fibronectin to its cell receptors and inhibits the adhesion of endothelial cells to substrates, arrested the elongation of developing microvessels causing regression and inhibition of angiogenesis. Conversely, Gly-Arg-Glu-(GRGES), which lacks the RGD sequence, had no inhibitory effect. These data support the hypothesis that fibronectin promotes angiogenesis and suggest that developing microvessels elongate in response to fibronectin as a result of an adhesion-dependent migratory recruitment of endothelial cells that does not require increased cell proliferation. © 1993 Wiley-Liss, Inc.     

Angiogenic potential of microvessel fragments established in three-dimensional collagen gels

Summary During angiogenesis, the microvasculature displays both vessel remodeling and expansion under the control of both cellular and extracellular influences. We have evaluated the role of angiogenic and angiostatic molecules on angiogenesis in anin vitro model that more appropriately duplicates the cellular and extracellular components of this process. Freshly isolated microvessel fragments from rat adipose tissue (RFMF) were cultured within three-dimensional collagen I gels. These fragments were characterized at the time of isolation and were composed of vessel segments observed in the microvasculature of fatin situ (i.e., arterioles, venules, and capillaries). Fragments also exhibited characteristic ablumenally associated cells including smooth muscle cells and pericytes. Finally, fragments were encased in an extracellular matrix composed of collagen type IV and collagen type I/III. The elongation of microvascular elements was subsequently evaluated using morphologic and immunocytochemical techniques. The proliferation, migration, and elongation of cellular elements in microvessel fragments from rat adipose tissue was dependent on initial fragment density, matrix density, and required serum. Inclusion of endothelial cell growth factors to microvessel fragments from rat adipose tissue 3-D cultures resulted in the accelerated elongation of tube structures and the expression of von Willebrand factor in cells constituting these tubes. Molecules with reported angiostatic capacity (e.g., transforming growth factor and hydrocortisone) inhibited vessel tube elongation. In vitro methods have been developed to evaluate numerous mechanisms associated with angiogenesis, including endothelial cell proliferation, migration, and phenotypic modulation. Microvascular endothelial cell fragments described in this study represent anin vitro population of cells that accurately duplicate thein vivo microcirculatory elements of fat. The proliferation of cells and elongation of microvascular elements subsequently observed in three-dimensional cultures provides anin vitro model of angiogenesis. Microvascular formation in this system results from pre-existing microvessel fragments unlike tube formation observed when cultured endothelial cells are placed in three-dimensional gels. This form of tube formation from cultured endothelium is more characteristic of vasculogenesis. Thus, the formation of microvascular elements from microvessel fragments provides the opportunity to examine the mechanisms regulating angiogenesis in anin vitro system amenable to precise experimental manipulation.     

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.     

Signaling via fibroblast growth factor receptor-1 is dependent on extracellular matrix in capillary endothelial cell differentiation

Differentiation of endothelial cells, i.e., formation of a vessel lumen, is a prerequisite for angiogenesis. The underlying molecular mechanisms are ill defined. We have studied a brain capillary endothelial cell line (IBEC) established from H-2Kb-tsA58 transgenic mice. These cells form hollow tubes in three-dimensional type I collagen gels in response to fibroblast growth factor-2 (FGF-2). Culture of IBEC on collagen gels in the presence of FGF-2 protected cells from apoptosis and allowed tube formation (i.e., differentiation) but not growth of the cells. FGF-induced differentiation, but not cell survival, was inhibited by treatment of the cells with an anti-beta1-integrin IgG. Changes in integrin expression in the collagen-gel cultures could not be detected. Rather, cell-matrix interactions critical for endothelial cell differentiation were created during the culture, as indicated by the gradual increase in tyrosine phosphorylation of focal adhesion kinase in the collagen-gel cultures. Inclusion of laminin in the collagen gels led to FGF-2-independent formation of tube structures, but cells were not protected from apoptosis. These data indicate that FGF receptor-1 signal transduction in this cell model results in cell survival. Through mechanisms dependent on cell-matrix interactions, possibly involving the alpha3beta1-integrin and laminin produced by the collagen-cultured IBE cells, FGF stimulation also leads to differentiation of the cells.     

Regulation of retinal capillary cells by basic fibroblast growth factor, vascular endothelial growth factor, and hypoxia

Vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (bFGF) feature prominently in retinal neovascular diseases. Although the role of VEGF in retinal angiogenesis is well established, the importance of bFGF in this process requires further clarification. This study was undertaken to investigate the responses of retinal capillary cells (endothelial cells and pericytes) to bFGF under hypoxic conditions, as well as the potentially synergistic effects of bFGF and VEGF on the proliferation and cord formation of retinal endothelial cells. Cell proliferation was determined by cell number and by 3H-thymidine incorporation. Cord formation was assessed in three-dimensional gels of collagen type I. VEGF and bFGF increased 3H-thymidine incorporation by both cell types, an effect that was more pronounced in a hypoxic environment. Moreover, the proliferation of pericytes was stimulated to a greater extent by bFGF relative to VEGF. Endothelial migration in collagen gels, however, was induced more effectively by VEGF than by bFGF. A synergistic effect of VEGF and bFGF on cell invasion was observed in the collagen gel assay. VEGF and bFGF each augment proliferation of these cells, especially under hypoxia. We thus propose that these two cytokines have a synergistic effect at several stages of angiogenesis in the retina.     

Retinal capillary endothelial cells prefer different substrates for growth and migration

Recent studies have shown that the extracellular matrix modifies the behaviour of endothelial cells. We have studied the effects of extracellular matrix components on retinal capillary endothelial cell migration and proliferation. Bovine retinal capillary endothelial cells were selectively cultured from collagenase-digested microvessel fragments. In a filter system for the assessment of migration, endothelial cells responded to substrate-bound fibronectin but not to soluble fibronectin. Cell migration on collagen- or gelatin-coated filters was minimal, and these cells failed to adopt a spread morphology, remaining instead as round cells. Cell replication was quantified using a protein dye binding assay for adherent cells in 96 well plates. Serum was essential for growth irrespective of the substrate. Cells harvested from microvessel cultures proliferated more rapidly on collagen- and gelatin-coated plastic than on fibronectin and were unaffected by additions to the medium such as endothelial cell conditioned medium, whereas cells proliferating directly from the microvessels grew at a faster rate on fibronectin and also responded to conditioned medium supplement. When cultured on collagen gels, initial microvessel cells and harvested cells required surface fibronectin in order to adopt a cobblestone morphology. These results show that fibronectin is a requirement for bovine retinal capillary endothelial cell migration, but proliferation of these cells can be supported, with slight differences, by both fibronectin and collagen provided serum growth factors are present. These findings are relevant to the early phase of angiogenesis in which migration and proliferation of endothelial cells occurs.     

Phenotypic modulations of human umbilical vein endothelial cells and human dermal fibroblasts using two angiogenic assays.

Different angiogenic assays in vitro have helped to define various events underlying angiogenesis. In this report we have compared the phenotypic modifications of human umbilical vein endothelial cells (HUVE cells) and human dermal fibroblasts using Matrigel and collagen gels. Both HUVE cells and human dermal fibroblasts form a network of anastomosing cords that apparently resemble blood capillaries when grown on Matrigel. The whole network was formed by several cellular aggregates joined to each other by cellular cords. Lumen formation was not observed in this angiogenic system. In opposite, considerable differences between HUVE cells and human dermal fibroblasts were observed in the three-dimensional angiogenic assay on collagen gels described by Montesano et al [14]. These results indicate that data obtained with angiogenic systems using Matrigel must be interpreted with caution and that the assay described by Montesano et al [14], is more reliable to describe angiogenesis.     

Phenotypic modulation of endothelial cells by transforming growth factor-beta depends upon the composition and organization of the extracellular matrix

Transforming growth factor beta (TGF-beta) is angiogenic in vivo. In vitro, endothelial cell proliferation is inhibited by TGF-beta. We have correlated this inhibitory effect with an increase in cellular fibronectin synthesis and deposition in a two-dimensional culture system using specific matrix coatings. The inhibitory effect was mimicked by addition of soluble fibronectin to cultures. In contrast, TGF-beta was found to elicit the formation of tube-like structures (mimicking angiogenesis) when microvascular endothelial cells were grown in three-dimensional collagen gels. In this culture system TGF-beta elicited rapid extensive formation of complex, branching, tube-like structures, while cell proliferation was not inhibited. These data confirm and support the hypothesis that TGF-beta is angiogenic and may exert some of its effects through modulation of matrix synthesis and are consistent with the hypothesis that the organization of the extracellular environment influences cellular responses to this "panregulin."     

Hyaluronan synthase elevation in metastatic prostate carcinoma cells correlates with hyaluronan surface retention, a prerequisite for rapid adhesion to bone marrow endothelial cells

Bone marrow is the primary site of metastasis in patients with advanced stage prostate cancer. Prostate carcinoma cells metastasizing to bone must initially adhere to endothelial cells in the bone marrow sinusoids. In this report, we have modeled that interaction in vitro using two bone marrow endothelial cell (BMEC) lines and four prostate adenocarcinoma cell lines to investigate the adhesion mechanism. Highly metastatic PC3 and PC3M-LN4 cells were found to adhere rapidly and specifically (70-90%) to BMEC-1 and trHBMEC bone marrow endothelial cells, but not to human umbilical vein endothelial cells (15-25%). Specific adhesion to BMEC-1 and trHBMEC was dependent upon the presence of a hyaluronan (HA) pericellular matrix assembled on the prostate carcinoma cells. DU145 and LNCaP cells were only weakly adherent and retained no cell surface HA. Maximal BMEC adhesion and HA encapsulation were associated with high levels of HA synthesis by the prostate carcinoma cells. Up-regulation of HA synthase isoforms Has2 and Has3 relative to levels expressed by normal prostate corresponded to elevated HA synthesis and avid BMEC adhesion. These results support a model in which tumor cells with up-regulated HA synthase expression assemble a cell surface hyaluronan matrix that promotes adhesion to bone marrow endothelial cells. This interaction could contribute to preferential bone metastasis by prostate carcinoma cells.