Regulation of cell adhesion to collagen via β1 integrins is dependent on interactions of filamin A with vimentin and protein kinase C epsilon

Cell adhesion and spreading on collagen, which are essential processes for development and wound healing in mammals, are mediated by β1 integrins and the actin and intermediate filament cytoskeletons. The mechanisms by which these separate cytoskeletal systems interact to regulate β1 integrins and cell spreading are poorly defined. We previously reported that the actin cross-linking protein filamin A binds the intermediate filament protein vimentin and that these two proteins co-regulate cell spreading. Here we used deletional mutants of filamin A to define filamin A-vimentin interactions and the subsequent phosphorylation and re-distribution of vimentin during cell spreading on collagen. Imaging of fixed and live cell preparations showed that phosphorylated vimentin is translocated to the cell membrane during spreading. Knockdown of filamin A inhibited cell spreading and the phosphorylation and re-distribution of vimentin. Knockdown of filamin A and/or vimentin reduced the cell surface expression and activation of β1 integrins, as indicated by immunoblotting of plasma membrane-associated proteins and shear force assays. In vitro pull-down assays using filamin A mutants showed that both vimentin and protein kinase C? bind to repeats 1-8 of filamin A. Reconstitution of filamin-A-deficient cells with full-length filamin A or filamin A repeats 1-8 restored cell spreading, vimentin phosphorylation, and the cell surface expression of β1 integrins. We conclude that the binding of filamin A to vimentin and protein kinase Cε is an essential regulatory step for the trafficking and activation of β1 integrins and cell spreading on collagen.     

Role of integrins in regulation of collagen phagocytosis by human fibroblasts

Phagocytosis of collagen fibrils by fibroblasts is an important pathway for degradation of extracellular matrix in mature connective tissues. To study regulatory mechanisms in phagocytosis, 2-μm fluorescent beads coated with either collagen (COL) or bovine serum albumin (BSA) were incubated with human gingival fibroblasts in vitro. For these studies single cell suspensions were prepared by trypsinization, and bead internalization and collagen receptor expression were assessed by flow cytometry. After 3-h incubations, up to 8-fold more cells internalized COL beads than BSA-coated beads. Increased collagen coating concentration was associated with elevated proportions of cells that internalized COL beads, and was observed also in the presence of competing fibronectin-coated beads. The number of beads per cell and the percent of phagocytic cells increased proportionally with higher bead loadings. At > 4 beads per cell a maximum of ∼︁80% of cells were phagocytic. Cells reacted with mAbs against the α1, α2, and α3 integrin subunits were, respectively, 5%, 98% and 93% positively stained above background controls. All cells that internalized COL beads exhibited α2 staining but there were large proportions of phagocytic cells that were not stained for α1. In unfixed cells, bead internalization caused an immediate reduction of surface staining of membrane-bound α2 by ∼︁55% which returned to control levels within 3 h, indicating that cell-surface α2 was internalized by phagocytosis. Preincubation of cells with up to 8 COL beads per cell reduced the proportion of phagocytic cells and the number of internalized beads after a second COL bead incubation 4 h later. To assess the relationship between the percent of phagocytic cells and α2 integrin levels, serum starvation and cycloheximide experiments were conducted. Compared to controls, serum starvation for 24 h induced a 3.2-fold increase of cells internalizing COL beads but did not alter α2 staining levels. In contrast, 3 h cycloheximide treatment reduced α2 staining to 60% of control levels and this treatment also inhibited COL bead internalization. GRGDTP peptide as well as mAbs against the α1 and α2 subunits significantly reduced internalization of COL beads by 1.8 to 2.6-fold, whereas GRGESP peptide and α3 mAb exerted no effect. Internalization of BSA beads was not affected by any of these treatments. Collectively, these data indicate that the α2 integrin, along with other, as yet unidentified components, is likely involved in COL bead internalization. The α2 integrin subunit is rapidly recycled or synthesized following a phagocytic load. In contrast, the α1 integrin is not directly required for phagocytosis but may regulate the internalization step. © 1996 Wiley-Liss, Inc.     

Regulation of stretch-activated intracellular calcium transients by actin filaments.

Stretch activation of cation-permeable channels may be an important proximal sensory mechanism in mechanotransduction. As actin filaments may mediate cellular responses to changes of the mechanical properties of the substrate and regulate stretch-induced calcium transients, we examined the role of actin filaments and substrate flexibility in modulating the amplitude of stretch-activated intracellular calcium transients. Human gingival fibroblasts were subjected to mechanical stretch through integrins by magnetic force acting on collagen-coated ferric oxide beads. Intracellular calcium concentration was measured in fura-2-loaded cells by ratio fluorimetry. Cytochalasin D-treatment greatly increased (3-fold) the amplitude of stretch-activated calcium transients in well-spread cells grown on glass coverslips while phalloidin, colchicine or taxol exerted no signficant effects, indicating that actin filaments but not microtubules modulate stretch-activated calcium transients. In freshly plated cells with rounded shapes and poorly developed cortical actin filaments, stretch-induced calcium transients were of 3-fold higher amplitude than well-spread cells plated for 6-24 hrs and with well developed actin filaments. Cells plated on soft collagen-polyacrylamide gels showed round morphology but exhibited <50% of the response to stretch of well-spread cells on inflexible gels. Notably, cells on soft gels showed very heavy phalloidin staining for cortical actin filaments compared with cells on more inflexible surfaces which showed only light staining for cortical actin. While cell shape may have some effect on responsiveness to mechanical stretch, the rigidity of the cell membrane mediated by the extensive cortical actin network appears to be a central determinant in the regulation of stretch-induced calcium signals.     

A low-serum medium with BSA and ferrous sulfate for the production of tissue plasminogen activator by human embryo lung cells

A low-serum medium containing bovine serum albumin (BSA) was investigated with respect to the growth of and tissue plasminogen activator (TPA) production by human embryo lung (HEL) cells on microcarrier beads and in collagen gel. BSA and ferrous sulfate were chosen as substitutes for fetal calf serum (FCS) through a simple screening test involving many substances. The growth promoting effects of BSA and ferrous sulfate were independent of each other and from the FCS concentration. Though BSA inhibited initial cell attachment to the carrier surface, it did promote the growth of cells attached to microcarrier beads. Cells grown on microcarrier beads in the low-serum medium containing BSA, ferrous sulfate and 3% FCS produced an amount of TPA similar to that produced by ones grown in the 10% FCS medium. Although cells on the dish surface did not grow at all on serum-free media containing BSA and ferrous sulfate, cells in the collagen gel were able to grow slightly on the serum-free medium. Cells grown on the low-serum medium in collagen gel produced more TPA over a long period than those in the microcarrier beads using the low-serum medium. The optimum concentration of proteose peptone in the TPA production medium for the collagen gel culture was similar to that for the dish surface culture.     

Protein tyrosine phosphatase alpha regulates cell detachment and cell death profiles induced by nitric oxide donors in the A431 human carcinoma cell line

We investigated the role of protein tyrosine phosphatase-alpha (PTPα) expression in the cell death profile of the A431 human carcinoma cell line that was induced by cytotoxic concentrations of the nitric oxide (NO) donors sodium nitroprusside (SNP) and 3,3-bis-(aminoethyl)-1-hydroxy-2-oxo-1-triazene (NOC-18). Both NO donors promoted extensive cell detachment in A431 parental cells as compared to the detachment observed for A431 cells that ectopically expressed PTPα (A431 (A27B(PTPα)) cells). The NO-induced cell death characteristics for both cell lines were examined. After incubation for 10 hours with 2.0 mM SNP, attached or detached A431 cells underwent apoptosis. Cells were highly positive for Annexin-V, featured increased cleavage of procaspase-8, activation of downstream caspase-3, and activation of poly-ADP-ribose polymerase 1 (PARP-1). In contrast, exposure of A431 (A27B(PTPα)) cells to 2.0 mM SNP produced an increase in the release of lactate dehydrogenase and enhanced incorporation of propidium iodide. In addition, A431 (A27B(PTPα)) cells showed partial inhibition of the activities of caspase-8, caspase-3, and PARP-1 upon detachment and cell death induced by SNP treatment. Results indicate that necrotic cell damage was induced, characterized by cellular swelling and lysis. We conclude from these results that PTPα regulates the A431 tumor cell death profile mediated by NO donors. Expression of PTPα or its absence may determine the occurrence of NO-induced cell death with necrotic or apoptotic features, respectively.     

Filamin A regulates cell spreading and survival via beta1 integrins

Cell spreading and exploration of topographically complex substrates require tightly-regulated interactions between extracellular matrix receptors and the cytoskeleton, but the molecular determinants of these interactions are not defined. We examined whether the actin-binding proteins cortactin, vinculin and filamin A are involved in the formation of the earliest extensions of cells spreading over collagen or poly-L-lysine-coated smooth and beaded substrates. Spreading of human gingival fibroblasts was substantially reduced on beaded or poly-L-lysine-coated substrates. Filamin A, vinculin and cortactin were found in cell extensions on smooth collagen. HEK-293 cells also spread rapidly on smooth collagen and formed numerous cell extensions enriched with filamin A. Knockdown of filamin A in HEK-293 cells by short hairpin RNA reduced spreading and the number of cell extensions. Blocking beta1 integrin function significantly reduced cell spreading and localization of filamin A to cell extensions. Conversely, filamin A-knockdown reduced beta1 integrin-collagen binding as measured by 12G10 antibody, suggesting co-dependence between filamin A and beta1 integrin functions. TUNEL staining showed higher percentages of apoptosis after filamin A-knockdown in spreading cells. Chelation of [Ca2+]i with BAPTA/AM reduced spreading of wild-type and filamin A-knockdown cells, however wild-type cells showed recruitment of filamin A to the subcortex, indicating independent roles of filamin A and [Ca2+]i in cell spreading. We conclude that filamin A integrates with beta1 integrins to mediate cell spreading and prevent apoptosis.     

Morphological and functional differentiation of cryopreserved lactating bovine mammary cells cultured on floating collagen gels

Cryopreserved bovine mammary epithelial cells prepared from lactating mammary tissue synthesize and secrete the milk proteins alpha_s1-casein, lactoferrin (Lf), and alpha-lactalhumin during in vitro culture on collagen gels in serum-free medium. Each milk protein is differently regulated by detachment and thickness of the collagen substratum, fetal calf scrum, and prolactin in the medium. Collagen detachment did not modulate lactoferrin secretion but strongly induced casein secretion, with detachment on day 6 (after formation of cell sheets) inducing casein secretion to 3 μg/ml medium, which was 2–3-fold higher than for cells on collagen detached on day 2 (prior to cell spreading to form sheets), and ten-fold higher than for cells grown on collagen not detached. Alpha-lactalbumin secretion was also induced, but only to low levels, in cells grown on detached but not on attached collagen. Cells grown on thin collagen gels secreted lower levels of lactoferrin and casein compared to cells on thick collagen. Lactoferrin but not casein secretion was increased in cells grown in the presence of fetal calf serum. Casein but not lactoferrin secretion was completely dependent on prolactin. Cells grown serum-free on collagen gels detached on day 6 of culture showed a polarized epithelial cell layer with high differentiation evidenced by the apical microvilli, tight junctions, and fat droplets surrounded by casein-containing secretory vesicles. An underlying layer of myoepithelial-like cells was also evident. These studies show for eryopreserved primary bovine mammary cells prepared from lactating mammary tissue the induction of highly differentiated and polarized cell morphology and ultrastructure with concomitant induction of the secretion of casein, lactoferrin. and alpha-lactalbumin in vitro, and that the non-coordinate regulation of milk protein secretion by substratum, prolactin, and serum likely involves alternate routing and control of secretion pathways for casein and lactoferrin.     

Release of cell fragments by invading melanoma cells

Tumor cell invasion requires coordinated cell adhesion to an extracellular matrix (ECM) substrate at the leading edge and concomitant detachment at the cell rear. Known detachment mechanisms include the slow sliding of focal contacts, the detachment of adhesion receptors by affinity and avidity regulation, as well as the shedding of adhesion receptors, most notably integrins. In highly invasive melanoma cells migrating within 3D collagen matrices, beta1 integrins and CD44 are released upon retraction of the trailing edge, together with ripping-off complete cell fragments to become deposited along the migration trail of remodeled matrix. Cell fragments reach a size up to 12 microm in diameter, contain cytoplasm and occasionally polymerized actin enclosed by intact cell membrane including surface beta1 integrins, but do not include nuclear material. The release of cell fragments was migration dependent, as impairment of motility by a blocking anti-beta1 integrin antibody also blocked cell particle release. Invasion-associated deposition of cell fragments combines the secretory-type release of vesicles with a physical mechanism of rear retraction and migration efficiency. The deposition of cell fragments may further represent a disregulated detachment strategy with implications for neoplastic cell behavior, such as the paracrine effects on neighbor cells or a negative impact on immune effector cells.     

Herpes simplex virus 1 blocks caspase-3-independent and caspase-dependent pathways to cell death

Earlier reports have shown that herpes simplex virus 1 (HSV-1) mutants induce programmed cell death and that wild-type HSV blocks the execution of the cell death program triggered by viral gene products, by the effectors of the immune system such as the Fas and tumor necrosis factor pathways, or by nonspecific stress agents such as either osmotic shock induced by sorbitol or thermal shock. A report from this laboratory showed that caspase inhibitors do not block DNA fragmentation induced by infection with the HSV-1 d120 mutant. To identify the events in programmed cell death induced and blocked by HSV-1, we examined cells infected with wild-type virus or the d120 mutant or cells infected and exposed to sorbitol. We report that: (i) the HSV-1 d120 mutant induced apoptosis by a caspase-3-independent pathway inasmuch as caspase 3 was not activated and DNA fragmentation was not blocked by caspase inhibitors even though the virus caused cytochrome c release and depolarization of the inner mitochondrial membrane. (ii) Cells infected with wild-type HSV-1 exhibited none of the manifestations associated with programmed cell death assayed in these studies. (iii) Uninfected cells exposed to osmotic shock succumbed to caspase-dependent apoptosis inasmuch as cytochrome c was released, the inner mitochondrial potential was lost, caspase-3 was activated, and chromosomal DNA was fragmented. (iv) Although caspase-3 was activated in cells infected with wild-type HSV-1 and exposed to sorbitol, cytochrome c outflow, depolarization of the inner mitochondrial membrane, and DNA fragmentation were blocked. We conclude that although d120 induces apoptosis by a caspase-3-independent pathway, the wild-type virus blocks apoptosis induced by this pathway and also blocks the caspase-dependent pathway induced by osmotic shock. The block in the caspase-dependent pathway may occur downstream of caspase-3 activation.     

Force-induced apoptosis mediated by the Rac/Pak/p38 signalling pathway is regulated by filamin A

Cells in mechanically challenged environments cope with high-amplitude exogenous forces that can lead to cell death, but the mechanisms that mediate force-induced apoptosis and the identity of mechanoprotective cellular factors are not defined. We assessed apoptosis in NIH 3T3 and HEK (human embryonic kidney)-293 cells exposed to tensile forces applied through β1-integrins. Apoptosis was mediated by Rac-dependent activation of p38α. Depletion of Pak1 (p21-activated kinase 1), a downstream effector of Rac, prevented force-induced p38 activation and apoptosis. Rac was recruited to sites of force transfer by filamin A, which inhibited force-induced apoptosis mediated by Rac and p38α. We conclude that, in response to tensile force, filamin A regulates Rac-dependent signals, which induce apoptosis through Pak1 and p38.     

Modality of Cell Death Induced by Foscan®-Based Photodynamic Treatment in Human Colon Adenocarcinoma Cell Line HT29

Apoptosis induced by photodynamic therapy (PDT) is considered to be an important factor defining the treatment outcome. Nevertheless, the relevance of apoptotic events in overall cell death should be established for every given photosensitizer. The present study addresses the contribution of Foscan-(meta-tetra(hydroxyphenyl)chlorine; mTHPC) photosensitized apoptosis in overall cell death in a model of cultured HT29 adenocarcinoma cells. Early events of cell death were assessed by the evaluation of mitochondrial response to mTHPC-mediated PDT, cytochrome c release and membrane depolarization. Apoptosis was measured through the activity of caspase-3 and the binding of the fluorescent conjugate Ca2+-dependent protein Annexin-V on membrane externalized phosphatidylserine at 2, 4, and 24 h post-PDT. Immediately after mTHPC-PDT, from 28 to 57% cells exhibited cytochrome c release concomitantly with mitochondrial membrane depolarization for light doses inducing more than 90% overall cell death. The maximum of caspase-3 activation (12-fold more than control) was reached 24 h after irradiation at fluence inducing 90% cell death (LD(90)). The corresponding measurement of apoptotic cells (12% of Annexin-V bound cells) confirmed the mild and delayed apoptotic response of HT29 cells to mTHPC-PDT.     

Limited proteolysis of filamin is catalyzed by caspase-3 in U937 and Jurkat cells

Members of the caspase family have been implicated as key mediators of apoptosis in mammalian cells. However, few of their substrates are known to have physiological significance in the apoptotic process. We focused our screening for caspase substrates on cytoskeletal proteins. We found that an actin binding protein, filamin, was cleaved from 280 kDa to 170, 150, and 120 kDa major N-terminal fragments, and 135, 120, and 110 kDa major C-terminal fragments when apoptosis was induced by etoposide in both the human monoblastic leukemia cell line U937, and the human T lymphoblastic cell line Jurkat. The cleavage of filamin was blocked by a cell permeable inhibitor of caspase-3-like protease, Ac-DEVD-cho, but not by an inhibitor of caspase-1-like protease, Ac-YVAD-cho. These results suggest that filamin is cleaved by a caspase-3-like protease. To examine whether caspase-3 cleaves filamin in vitro, we prepared a recombinant active form of caspase-3 directly using a Pichia pastoris overexpression system. When we applied recombinant active caspase-3 to the cell lysate of U937 and Jurkat cells, filamin was cleaved into the same fragments seen in apoptosis-induced cells in vivo. Platelet filamin was also cleaved directly from 280 kDa to 170, 150, and 120 kDa N-terminal fragments, and the cleavage pattern was the same as observed in apoptotic human cells in vivo. These results suggest that filamin is an in vivo substrate of caspase-3.     

Interaction of the pacemaker channel HCN1 with filamin A

Pacemaker channels are encoded by the HCN gene family and are responsible for a variety of cellular functions including control of spontaneous activity in cardiac myocytes and control of excitability in different types of neurons. Some of these functions require specific membrane localization. Although several voltage-gated channels are known to interact with intracellular proteins exerting auxiliary functions, no cytoplasmic proteins have been found so far to modulate HCN channels. Through the use of a yeast two-hybrid technique, here we showed that filamin A interacts with HCN1, an HCN isoform widely expressed in the brain, but not with HCN2 or HCN4. Filamin A is a cytoplasmic scaffold protein with actin-binding domains whose main function is to link transmembrane proteins to the actin cytoskeleton. Using several HCN1 C-terminal constructs, we identified a filamin A-interacting region of 22 amino acids located downstream from the cyclic nucleotide-binding domain; this region is not conserved in HCN2, HCN3, or HCN4. We also verified by immunoprecipitation from bovine brain that the filamin A-HCN1 interaction is functional in vivo. In filamin A-expressing cells (filamin+), HCN1 (but not HCN4) channels were expressed in hot spots, whereas they were evenly distributed on the membrane of cells lacking filamin A (filamin-) indicating that interaction with filamin A affects membrane localization. Also, in filamin- cells the gating kinetics of HCN1 were strongly accelerated relative to filamin+ cells. The interaction with filamin A may contribute to localizing HCN1 channels to specific neuronal areas and to modulating channel activity.     

Intercellular transfer of apoptotic signals via electrofusion

We determined whether cells that are induced to undergo anoikis by matrix detachment can initiate apoptosis in healthy cells following electroporation-induced fusion. Separate populations of MDCK cells undergoing anoikis and stained with FITC-annexin or viable MDCK cells that were labeled with spectrally discrete fluorescent beads were electroporated. Cells were analyzed by flow cytometry for enumeration of viable cells with beads, apoptotic cells or fused cells. Electroporation promoted a 49-fold increase of the percentage of viable cells that had fused with apoptotic cells. Apoptotic cell-viable cell fusions were 8-fold more likely to not attach to cell culture plastic and 2.3-fold less likely to proliferate after 24hr incubation than viable cell fusion controls. These data demonstrate that apoptotic signals can be transferred between cells by electrofusion, possibly suggesting a novel investigative approach for optimizing targeted cell deletion in cancer treatment.     

Combined treatment with <Emphasis Type="SmallCaps">l</Emphasis>-carnitine and a pan-caspase inhibitor effectively reverses amiodarone-induced injury in cultured human lung epithelial cells

Amiodarone is an effective class III antiarrhythmic drug, however, the pulmonary toxicity is one of the most life-threatening complications of its use. The present study was designed to determine the mechanisms underlying pulmonary toxicity of amiodarone. In cultured human lung epithelial cells A549, amiodarone caused cell injury characterized by mitochondrial membrane depolarization, ATP depletion, enhanced propidium iodide (PI) uptake and increase in the number of Annexin-V positive cells, although the population of PI-stained cells appeared earlier and was not identical to that of Annexin-V stained cells, suggesting that the apoptosis and necrosis appeared in different cells. The apoptosis was accompanied with the activation of caspase-2, -3 and -8 but not caspase-9, and reversed by these caspase inhibitors. However, the caspase inhibitors had no influence on mitochondrial membrane potential or PI uptake after exposure of A549 cells to amiodarone. In contrast, mitochondrial cofactors such as L-carnitine and acetyl-l-carnitine attenuated mitochondrial membrane depolarization, abrogated cellular ATP depletion and reversed PI uptake without affecting Annexin-V positive cells. These finding suggest that different intracellular events operate to cause apoptosis and necrosis after exposure of pulmonary epithelial cells to amiodarone.     

Deregulation of Collagen Phagocytosis in Aging Human Fibroblasts: Effects of Integrin Expression and Cell Cycle

Intracellular degradation of collagen by phagocytosis in fibroblasts is essential for physiological remodeling of the extracellular matrix in a wide variety of connective tissues but imbalances between degradation and synthesis can lead to loss of tissue collagen. As aging is associated with loss of dermal and periodontal collagen and with increased lysomomal enzyme content in fibroblasts, we examined the regulation of collagen phagocytosis by integrin expression and the cell cycle in anin vitrofibroblast aging model. Two different fibroblast lines (CL1; CL2) at the fourth subculture were passaged up to replicative senescence to model aging processesin vitro.Cells were incubated with collagen-coated or BSA-coated green fluorescent beads for 3 h to assess α₂β₁-integrin-mediated or nonspecific phagocytosis, respectively. Single-cell suspensions were stained with DAPI and sulforhodamine 101 to separate cycling G₁and noncycling G₀cells. Staining for α₂-integrin, bead internalization, and bivariate analyses of DNA/protein content were measured by three-color flow cytometry. Serum deprivation was used to induce increases in the proportion of G₀cells. For G₁cells, the proportion of collagen phagocytic cells was >50% for all passages and collagen beads were internalized >5-fold more frequently than BSA beads. In contrast, G₀cells with diploid DNA content but low protein content exhibited greatly reduced phagocytic capacity (<10% of cells internalized collagen or BSA beads), the number of beads per cells was 4-fold less, and α₂integrin expression was very low compared to G₁cells. The proportion of collagen phagocytic cells and the proportion of α₂-integrin-positive cells increased with transit through the cell cycle. At higher passage numbers mean cell volume and cytoplasmic granularity were reduced 30% but at replicative senescence cells with large surface area and subdiploid DNA predominated. The proportion of collagen and BSA phagocytic G₁cells increased 1.5- and 5-fold, respectively, and the number of beads per cell increased <3-fold. However, surface α₂-integrin staining remained unchanged. These data indicate that the collagen and nonspecific internalization pathways were greatly upregulated, independent of cell cycle phase, and that cellular agingin vitrostrongly influences the specificity and rate of phagocytic processes in fibroblasts. We suggest that age-related loss of collagen in connective tissues undergoing turnover may be a manifestation of a deregulated increase of collagen phagocytosis in which the net loss of degraded collagen exceeds new synthesis.     

A novel subcellular collagen organization process visualized by total internal reflection fluorescence microscopy

The alpha1beta1 and alpha2beta1 integrins belong to a family of cell-surface molecules involved in structural contacts and signal-transduction events across the cell membrane. Employing two-dimensional substrates coated with fluorescently labeled type I collagen, we have discovered a novel subcellular matrix remodeling event that is particular to cells that express the fibrillar collagen receptor alpha2beta1. Cells expressing alpha1beta1 also perform this collagen organization process, but less proficiently. This work will provide a basis for subsequent studies of cell-mediated collagen fibril assembly.     

Neovascular responses induced by cultured aortic endothelial cells

Neovascularization was studied in the chorioallantoic membrane of the chick embryo after implantation of bovine aortic endothelial and smooth muscle cells, Swiss and BALB/c 3T3 cells and human diploid fibroblasts cultured separately on microcarrier beads. Quantitative analysis of neovascularization indicated a 3 1/2-fold increase in the number of blood vessels responding to endothelial cells while smooth muscle cells induced a twofold increase when compared to the response of beads without cells. Skin fibroblasts and Swiss 3T3 cells did not elicit a comparable response. The marked angiogenic response induced by endothelial cells was characterized by a 137% increase in total vessel length and a 35% increase in average vessel area when compared to controls. Two of the properties required for an angiogenesis factor--stimulation of cellular migration and proliferation--can also be demonstrated using endothelial cell-conditioned medium in cell culture systems. Medium from cultured bovine aortic endothelium stimulates DNA synthesis, proliferation, and migration of smooth muscle cells. In addition, conditioned media from both endothelial cells and smooth muscle cells produced an angiogenic response in the chorioallantoic membrane assay, which was comparable to that produced by intact cells growing on microcarrier beads. Similar responses were not evident with medium conditioned by other cell types. These results indicate the potential importance of endothelial cells and endothelial cell products in regulating blood vessel growth.