Biol Cell (1991) 72, 275-278


© Elsevier, Paris

L e t t e r to the e d i t o r

Phenotypic modulations of human umbilical vein endothelial cells and human dermal fibroblasts using two angiogenic assays A n d r e a s B i k f a l v i * , E l i s a b e t h M C r a m e r , Dani~le T e n z a , G 6 r a r d T o b e l e m I N S E R M U 150, [email protected] Lariboisi~)re, 6, rue Guy-Patin, 75010 Paris, France (Received 22 April 1991; accepted 1i July 1991)

Summary - 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 fihroblasts 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 result 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. in vitro angiogenesis I Matdgel I cord formation

Angiogenesis is defined as the formation of new blood vessels. Different angiogenic assays have been established in order to study this phenomenon in rive [1, 5, 10] and in vitro [3, 13, 14-18, 20]. In vitro angiogenic assays consist of culturing endothelial cells on collagen gels [14-16], fibrin gels [17], fibronectin [8] or on Matrigel [6]. Threedimensional collagen gels have helped to define several morphological and biochemical events occurring in anglogenesis and were used to establish the angiogenic potential of basic fibroblast growth factor (bFGF), 1 2 - 0 tetradecanoylphorbol 13-acetate (TPA) and transforming growth factor-p (TGF-~) [13, 14-17, 20]. Matrige! is prepared from Engelbreth-Hoim-Swarm (EHS) murme sarcoma [11]. It is a complex mixture of basement membrane proteins including laminin, collagen type IV, endactin/nidogen and proteoheparan sulfate, but it contains also growth factors such as epidermal growth factor (EGF) [11, 22]. When endothelial cells are layered on the top of this gel, rapid cord formation occurs. In this report we have compared the phenotypic modifications of human dermal fibroblasts and HUVE cells using this two angiogenic assays either on Matrigel or on collagen gels. When human dermal fibroblasts were layered on the top of Matrigel, rapid cord formation was observed (fig la). The cells became rapidly attached and shape change occurred. The cells became elongated and cords of adjacent cells were formed within 2-3 h. After 12 h, the whole structure was formed by a network of several cellular aggregates joined to each other by cellular cords. Cord formation in human dermal fibroblasts was similar to that observed for HUVE cells (fig lb). Examination of the hematoxylin stained cells at high magnification showed that the cords were composed of adjacent elongated cells which seemed not to possess a lumen (data not shown). Indirect ira* Present address and correspondence and reprints: A Bikfalvi, Department of Cell Biology, New York University Medical Center, 550 first Avenue, New York, NY 10016, USA

munofluorescence in HUVE cells using an antibody against von Willebrand Factor (VWF) antigen, demonstrated a granular pattern which suggested that the inner space of the cellular cords was completely filled with cytoplasm (fig 2a). Granules were distributed in the perinuclear region but were also seen in the cellular extensions. This suggested that clustering of VWF was not observed in HUVE cells cultured on Matrigei. Immunofluorescence with anti-vimentin antibody showed in HUVE cells (fig 2b) and in fibroblasts (not shown) a vimentin network extending from the nucleus to the plasma membrane which entirely filled the cytoplasm. Cross-sections of cells cultured on Matrigel were subsequently processed for analysis by light microscopy (fig 3). Cells were detected on the surface of Matrigel but also sometimes more deeply in the gel. HUVE cells were devoid of any lumen. Some of them were found to form "invaginations" into Matrigel (fig 3b). Several HUVE cells bear vacuole-like spaces (fig 3c). Human dermal fibroblasts never exhibited a lumen or a vacuole-like space (fig 3d). These results indicate that human dermal fibroblasts and HUVE cells plated on the top of Matrigel formed cord-like structures which did not contain lumens. Using Matrigel, Kubota et ai [12] and Grant et al [6] have shown that laminin plays a crucial role in angiogenesis. Laminin induces angiogenesis by a double signal, RGD and YIGSR located respectively on A and B chain of laminin. These results, however, must be interpreted with caution in the light of the present study. HUVE cells plated on Matrigel underwent phenotypic modifications which are not restricted to endothelial cells. These modifications were rapid and were not accompanied by major invasion of the cells into the gel. Therefore, this system explores remodeling of the cellular phenotype involving certainly the cytoskeleton with minimal invasion. These events certainly underlie capillary morphogenesis and are necessary for the appearance of an angiogenic phenotype. They are, however, not sufficient for accounting for all






i .

the steps involved in angiogenesis. Besides endothelial cells, Matrigel has been reported to induce cord formation in malignant fibrosarcoma cells, but not in normal human fibroblasts [4]. This could be due to a more activated state in fibrosarcoma cells, since these cells express an activated N-ras gene [2, 19]. However, these data are in apparent contradiction with our study in which normal human dermal fibroblasts were capable of forming cellular cords. Therefore, considerable differences might exist in the phenotypic modifications induced by Matrigel in fibre.. blasts cultured from different sources. Moreover, Matrigel has been reported to stimulate differentiation of Sertoli cells [7] and mouse kidney cells [22] when they are included in the gel. Thus, Matrigel is capable of inducing phenotypic modifications in non-transformed cell types other than endothelial cells and fibroblasts when they are included into the gel. When human dermal fibroblasts were layered on the top


Fig 1. Effect of Matrigel on the phenotype of human dermal fibroblasts and HUVE cells. Human umbilical vein endothelial cells (HUVE cells) were isolated and cultured according to Jaffe et al [9]. The cells were cultured in medium 199 containing 10% of human serum and were passaged in a split ratio of 1: 2. Human dermal flbroblasts were isolated by the explant technique from the skin of healthy donors. The cells were cultured in RPMI containing 15% fetal calf serum and passaged in a split ration of 1: 3. Passages 2 and 3 were used throughout the study for HUVE cells and passages 6 to 10 for human derman fibroblasts. Matrigel was prepared from EHS tumor by hight salt and urea extraction according to Kleinman et al [11] or purchased from Collaborative Research (Bedford, MA) and stored at -200C until use. Matrigel was then layered at 4"C into wells of 24 multiwell plates (200 ~i/well) and the plates were maintained at 37°C for 10 rain. Confluent monolayers of HUVE cells or dermal fibroblasts were then trypsinized and layered on top of the gel at a density of 60 000 cells/well in medium 199 or RPMI respectively containing 10~/0FCS. After 2-3 h, cords formatioa occurred. At the end of the assay (24 h), the cells were fixed and processed for light microscopy, a.fibroblast, b. HUVE cells.

of three-dimensional collagen gels (Collagen type 1, Collaborative Research, Bedford, MA) prepared according to Montesano et al [14], invasion into the gel was observed (data not shown), invasion into the gel was rapid and occurred spontaneously in absence of any "angiogenic factor", as previously reported [21]. However, a complete network formed by anastomosing cords was not observed. Invasion of fibroblasts into collagen gels could be stimulated by basic fibroblast growth factor (bFGF, British Biotechnology, Oxford, UK) and was also prevented by inhibition of protein synthesis with cycloheximide (10 t~g/ml) or by 1,10 phenanthroline (5 × l0 "s M)(Sigma, St Louis, USA) an inhibitor of metalloprotease, as has been described for bovine capillary endothelial cells [14]. Twelve-O-tetradecanoylphorbol 13-acetate (100 ng/ ml)(TPA)(Sigma, St Louis, USA) only had a small, but not significant effect on the invasion by this cell type. Cross-section of invading fibroblasts showed that migra-

Phenotypic modulations of human cells •







i i i!!il


Fig 2. Indirect immunofiuorescence of HUVE cells and fibroblasts cultured on Matrigel using anti-VWF or anti-vimentin antibodies. Cells cultured on cover-slips precoated with Matrigel, were fixed 20 s in absolute methanol. After washing with phosphate-buffered saline (PBS) pH 7.4, blocking was performed with PBS containing 2% BSA. The cells were subsequently incubated for 2 h with anti-yon Willebrand Factor antibody (Sanbio, Uden, the Netherlands) or antivimentin antibody (Sanbio, Uden, the Netherlands)in PBS-BSA at a dilution of 1/200. After 3 washes in PBS, the cells were incubated for 30 rain with rhodamin conjugated antirabbit lgG or antimouse lgG (Sanbio, Uden, the Netherlands) and washed again 3 times in PBS. FIuLorescence was examined with a Leitz immunofluorescence microscope (Wetzlar, Germany). a. antiVWF antibody, b. antivimentin antibody.

® d

) :i-'

Fig 3. Cross sections of HUVE cells (a-e) and fibroblasts (d) cultured on Matrigel and analysis by light microscopy. Cells cultured on cover-slips on which Matrigel was layered, were fixed in 1.5 glutaraldehyde in 0.1 M PBS (pH 7.4) for 2 h, postfixed in 1070osmic acid and embed~ed in Epon. Section cut transversely to the plane of the gel on an LKB ultramicrotome and stained with 1070toluidine blue for light microscopy or stained with uranyi acetate followed by Reynold's lead citrate for electron microscopy (Philips EM 300 electron microscope), a. Filled cords. b. "lnvagination" of cells in the gel. c. Vacuole-like structure, d. Fibroblasts.


A Bikfalviet al

tion into collagen gels was not accompanied by lumen formation. In opposite, HUVE cells dit not spontaneously i~vade collagen gels. Nevertheless, invasion and a network of anastomosing cords with lumen formation could be induced by TPA in that cell type as it has been previously described [14, 16]. In summary, our results demonstrate that cord formation Matrigel is not specific for endothelial cells and that cells cultured on the top of this type of gel mimic only partially the angiogenic phenomenon. Untransformed cells such as normal human dermal fibroblasts also undergo these kinds of "angiogenic modifications". Result with this angiogenic system must be interpreted with caution. Models using three-dimensional collagen gels are perhaps more meaningful to describe angiogenesis.

Acknowledgments This work has been supported by a grant from the Association de la recherche pour le Cancer (ARC). AB was a recipient of a fellowship from de Fondation pour la Recherche M~dicaie. The authors thanks Dr D Hicks for criticallyreading the manuscript.

References 1 Ausprunk DH, Knighton DR, Folkman J (1974) Differentiation of vascular endothelium in the chicken chorioallantois: a structural and autoradiographic study. Dev Bio138, 237-248 2 Brown R, Marshall CJ, Pennie SG, Hall A (1984) Mechan. ism of activation of an N-ras gene in the human fibrosarcoma cell line HT 1080, EMBO J 3, 551-556 3 Folkman J, Haudenschild C (1980) Angiogenesis in vitro. Nature 32, 551-556 4 Fridman R, Lacal JC, Reich R, Bonfil D, Ahn C (1990) Differential effects of phorbol ester on the in vitro invasive. hess of malignant and non-maiignan human fibroblast cells. J Cell Phy$iol 142, 55-60 5 GimbroneMA, Cotran R, Leapoman SB, Folkman J (1974) Tumor growth and neovascularization: an experimental model using the rabbit cornea. J Natl Cancer lnst 52, 413-419 6 Grant D, Tashiro KI, Segui-Real B, Yamada Y, Martin G, Kleinman H (1989) Two different laminin domains mediate the differentiation of human endothelial cells into capillarylike structures, Cell 58, 433-437 7 HadleyMA, Byers SW, Suarez-Quian CA, Kleinman HK, Dym M (1985) Extracellular matrix regulates Sertoli cell

differentiation, testicular cord formation and germ cell development. J Cell Biol 10I, 1511- 1522 8 IngberD, Folkman J (1989) Mechanochemical switching between growth and differentiation during fibroblast growth factor-stimulated angiogenesis in vitro. Role of extracellufar matrix. J Cell Biol 109, 317-331 9 Jaffee E, Nachman RL, Becker CG, Minick CR (1976) Culture of human endothelial cells derived from umbilical veins. J Clin Invest 52, 2745-2756 I0 Klagsbrun M, Kneighton D, Folkman J (1976)Tumor angiogenesis in cells grown in tissue culture. Cancer Res 36, 110-114 11 Kleinman H, McGarveyML, Hassel JR, Star VL, cannon FB, Laurie GW, Martin G (1986) Basement membrane complex with biological activity. Biochemistry 25, 312-318 12 Kubota Y, Kleinman HK, Martin G, LawleyT (1988) Role of laminin and basement membrane in the morphological differentiation of human endothelial cells into capillary-like structures. J Cell Biol 107, 1589-1598 13 Madri J, Pratt BM, Tucker A (1989) Phenotypic modulation of endothelial cells by transforming growth factor beta depends upon the composition of the extracellular matrix. J Cell Biol 106, 1375-1384 14 Montesano R, Orci L (1985) Tumor promoting phorbol ester induces angiogenesis in vitro 42, 469-487 15 Montesano R, Vassalli JD, Baird A, Guillemin R, Orci L (1986) Basic fibroblast growth factor induces angiogenesis in vitro. Proc Nail Acad Sci USA, 83, 7297-7301 16 Montesano R, Orci L (1987) Phorbol ester induces anglogenesis in vitro from large vessel endothelial cells. J Cell Physiol 130, 284-291 17 Montesano R, Pepper MS, Vassalli JD, Orci L (1987) Phorbol ester induces cultured endothelial cells to invade a fibrin matrix in presence of fibrinolytic inhibitors. J Cell Physiol 132, 460-466 18 Montesano R, Pepper MS, Belin D, Vassaili JD, Orci L 0987) Induction of angiogenesis in vitro by vanadate, an inhibitor of phosphotyrosine phosphatase. J Cell Physiol 134, 460-566 19 Paterson H, Reeves B, Brown R, Hall A, Furth M, Bos J, Jones P, Marshall C 0987) Activated N-ras controls the transformed phenotype of HT 1080 human fibrosarcoma cells. Cell $1,803-812 20 Pepper MS, Belin D, Montesano R, Orci L, Vassilli JD (1990) Transforming growth factor-induced proteolytic and angiogenic properties of endothelial cells in vitro. J Cell Biol 111,743-755 21 Schor SL (1980) Cell proliferation and migration on collagen substrata. J Cell 8ci 41, 159-175 22 Taub M, Wang Y, SzesnyMT, Kleinman HK (1990) Epidermal growth factor or transforming growth factor alpha is required for kidney tubulogenesis in matrigel cultured in serum free medium. Proc Natl Acad $ci USA 87, 4002-4006

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 modi...
828KB Sizes 0 Downloads 0 Views