3D coupling of fibronectin fibril arrangement with topology of ventral plasma membrane

Interaction of integrins with extracellular matrices is essential for cell adhesion to substrata. Ventral surfaces of fibroblasts adhering to flat substrata are not flat but have uneven 3D topology. However, spatial relationship between the topology of the ventral cell surface and arrangement of extracellular matrix fibrils remains unclear. Here, we report a novel and simple method based on total internal reflection fluorescence microscopy to quantify the distance between the ventral plasma membrane and the glass substratum. We observe that the distance varies from < 25 nm at focal adhesions to 40-50 nm at close contacts and > 80 nm in other regions. Furthermore, by applying this novel method, we show that fibronectin fibrils are also separated from the substratum in regions where the ventral cell surface-substratum distance is > 80 nm. Our results reveal that fibronectin fibrils are not simply adsorbed to the glass substratum but follow the ventral cell surface topology.     

Focal adhesion sites and the removal of substratum-bound fibronectin

Fibronectin was not removed from the substratum beneath focal adhesion sites when fibroblasts spread in serum-free medium on adsorbed fibronectin substrata, or when fibroblasts spread in serum-containing medium on covalently cross-linked fibronectin substrata. Under these conditions, there was colocalization between 140-kD fibronectin receptors and focal adhesion sites. It was concluded that removal of adsorbed fibronectin from beneath focal adhesion sites was a mechanical process that required serum. The effect of serum was nonspecific since serum could be replaced by equivalent concentrations of serum albumin, ovalbumin, or gamma globulins. Quantitative measurements indicated that the presence of proteins in the incubation medium weakens the interaction of fibronectin with the substratum, thereby allowing the adsorbed protein to be removed from the substratum at sites of high stress. After removing fibronectin from the substratum, cells reorganized this material into patches and fibrils beneath cells, and the reorganized fibronectin colocalized with fibronectin receptors. Some of the patches of fibronectin were phagocytosed. The fibronectin fibrils were observed to be in register with actin filament bundles and sometimes translocated to the upper cell surfaces. It is proposed that removal of fibronectin from beneath focal adhesion sites is an example of how cells can modify their extracellular matrices through contractile activity.     

Fibronectin fibril formation involves cell interactions with two fibronectin domains

Fibronectin fragments and domain-specific antibodies have been used to study the mechanism by which cells reorganize exogenous fibronectin substrata into fibrils. Fibroblasts prevented from protein synthesis, and hence not secreting endogenous fibronectin or other matrix components, reorganized exogenous fibronectin substrata into arrays resembling the matrix of normally cultured cells. Cells also formed fibrils from substrata containing mixtures of cell- and either of two different heparin-binding fibronectin fragments but not from either fragment alone. The gelatin-binding fragment alone or in conjunction with the cell-binding fragment did not promote fibril formation. Antibodies recognizing cell- and either heparin- or the gelatin-binding domains labeled fibrils formed by cells under normal culture conditions or when a substratum of intact fibronectin was used as the sole exogenous source. However, only antibodies recognizing the cell- or either heparin-binding fragment reduced fibrillogenesis from intact fibronectin substrates when added during cell spreading. These data suggest that formation of fibronectin fibrils can occur at the cell surface and that membrane components recognizing the cell- and the heparin-binding domains in fibronectin may cooperate in the assembly process     

Ambystoma maculatum gastrulae have an oriented, fibronectin-containing extracellular matrix.

During early development of the urodele Ambystoma maculatum, the appearance and distribution of fibronectin-containing fibrillar extracellular materials were studied by immunocytochemistry. Fibronectin (FN) first appears in the early blastula (stage 7) as thin punctate fibrils on the cell surface concentrated in the marginal zone. In late blastula (stage 9), thin fibrils are found throughout the blastocoel roof. Early gastrulae (stage 10) have numerous fibrils and multifibrillar strands concentrated in the dorsal lip region and oriented preferentially along a line parallel to the dorsal lip-animal pole axis. There is a striking increase in the amount of FN fibrils during the rest of gastrulation. This FN-containing network can be transferred to plastic substrata with preservation of the preferential orientation observed in vivo. Dorsal marginal zone explants placed on such conditioned substrata show polarized outgrowth toward the animal pole region of conditioned areas when placed on the dorsal lip side or the ventral marginal zone side of conditioned substrata. This outgrowth occurs symmetrically on bovine plasma FN-coated substrata, is prevented by Fab' fragments of antibodies to FN but fails to occur on laminin coated substrata. When migrating mesodermal cells from early gastrulae are cultured on substrata conditioned by deposition of the fibrillar matrix, these cells exhibit striking contact inhibition of locomotion, a phenomenon that may explain dispersal of migrating mesodermal cells across the blastocoel roof. When leading edges of mesodermal cells collide, cells abruptly change direction. When leading edges collide with trailing edges, the trailing edges detach from the substratum and cells move apart in the direction of the leading edge.     

Adhesive responses of fibroblast and neuroblastoma cells to substrata coated with polyvalent or monoclonal antibody to fibronectin

Both polyvalent and hybridoma-produced antibodies to fibronectin (Fn) were used to ‘map’ the immunoaccessible subsets of cell surface fibronectin on virus-transformed murine fibroblast SVT2 and rat neuroblastoma B104 cells. As one approach to this end, attachment and spreading responses of cells were measured on tissue culture substrata coated with antibody or with plasma fibronectin to compare their adhesive responses. Both SVT2 and B104 cells adhere poorly to polyvalent anti-Fn-coated substrata over short time intervals, but within several hours changes occur which permit cells to attach and spread as well on anti-Fn as on Fn (post-adsorption of the anti-Fn with Fn also generates a maximal response). This adhesive response could be completely prevented by predigesting the cells with Flavobacterium heparanase, but not with chondroitinase ABC, indicating that the cell surface Fn responsible for antibody-mediated adhesion is associated with heparan sulfate proteoglycans on the cell surface. The compositions of the substratum-attached material (left bound after EGTA-mediated detachment of cells) from cells attaching to anti-Fn or Fn were analysed by SDS-PAGE and found to be identical within the same cell type for the two different substrata. Three hybridoma-produced antibodies, which recognize different determinants on Fn, generated different adhesive responses for SVT2 or B104 cells when adsorbed to the substratum. SVT2 cells adhered well to antibody no. 32-coated substrata but poorly to antibodies 92 or 136; on the other hand, B104 cells responded similarly to all three antibodies over short times of attachment but much better to no. 32 after a several hour incubation. These experiments indicate that (1) much of the cell surface fibronectin is complexed with heparan sulfate proteoglycan and is initially inaccessible to bind to polyvalent antibody on the substratum to promote adhesion; (2) the surface of neuroblastoma cells contains a fibronectin-like molecule which is important in their substratum adhesion; and (3) monoclonal antibodies are valuable tools in ‘mapping’ the orientation of cell surface molecules like fibronectin by measuring adhesive responses to antibody-coated substrata.     

Fibroblast adhesion to RGDS shows novel features compared with fibronectin

As previously shown by others, the fibroblast attachment and spreading activity of fibronectin is mimicked by a short peptide (RGDS or longer) from the cell binding domain. Normal rat kidney fibroblasts showed similar attachment kinetics on either peptide GRGDSC or bovine plasma fibronectin and binding to either substratum was inhibited by peptide alone. We now demonstrate, however, considerable differences in biological activity between peptide and fibronectin. In particular, cells developed novel adhesion structures on peptide-coated substrata. Interference reflection microscopy showed a predominance of small round dark grey/black patches of adherent membrane ("spots") with relatively few focal adhesions, which occurred only at the outermost cell margins in contrast to their distribution in cells spread on fibronectin. The spots were resistant to detergent extraction and stained less strongly or not at all for vinculin. Electron microscopy in vertical thin section showed that the ventral surface of the cell was characterized by "point-contacts", corresponding in size to the spot structures seen by interference reflection microscopy, and which were only occasionally associated with microfilaments. Cells also required a higher substratum loading of peptide than fibronectin to promote spreading and proceeded to spread less rapidly and to a lesser extent, developing very few and extremely fine actin cables.     

Adhesion sites of murine fibroblasts on cold insoluble globulin-adsorbed substrata

The attachment and detachment behavior of three mouse fibroblast cell lines adhering to plastic tissue culture substrata coated with the serum protein cold-insoluble globulin (CIg) resembles that seen on the usual serumcoated substrata. The transformed cell line SVT2 spreads more extensively on the CIg-coated than on the serum-coated substratum, while the nontransformed Balb/c 3T3 line and concanavalin A-selected “revertant” of SVT2 are equally well spread on both substrata. In all three cases, immunofluorescence microscopy using antibodies to CIg suggests that the cells are more tightly apposed to the CIg-coated substratum than to the serum-coated substratum. Substrate-attached material (SAM), which contains cell-substratum adhesion sites and which is left after EGTA-mediated detachment of cells, is enriched for cell surface fibronectin and glycosaminoglycans (GAG). When cells are seeded onto CIg-coated substrata rather than serum-coated substrata, there is an increased deposition of GAG but a comparable deposition of cellular proteins. The protein distribution of the two types of SAM are identical as analyzed by sodium dodecyl sulfate/polyacrylamide gel electrophoresis, including fibronectin content. This indicates that substratum-bound CIg cannot functionally substitute for cell surface fibronectin in these adhesion sites. Analysis of the GAG deposited on CIg-coated substrata reveals that hyaluronate and the chondroitins are increased to a much greater extent than heparan sulfate; however, the ratio of hyaluronate to the various chondroitin species is invariant. These data provide further evidence that hyaluronate and the chondroitins are deposited in adhesion sites in well-defined stoichiometric proportions, possibly as supramolecular complexes, and that CIg may mediate adhesion of cells in the serum layer by binding to GAG-containing proteoglycans.     

Dorsal-ventral differentiation in Simonsiella and other aspects of its morphology and ultrastructure

The morphology and ultrastructure of the aerobic, Gram-negative multicellular-filamentous bacteria of the genus Simonsiella were investigated by scanning and transmission electron microscopy. The flat, ribbon-shaped, multicellular filaments show dorsal-ventral differentiation with respect to their orientations to solid substrata. The dorsal surface, orientated away from the substrate, is convex and possesses an unstructured capsule. The ventral surface, on which the organisms adhere and glide, is concave and has an extracellular layer with fibrils extending at right angles from the cell wall. The cytoplasm in the ventral region contains a proliferation of intracytoplasmic membranes and few ribosomes in comparison to the cytoplasm in other parts of the cell. Centripetal cell wall formation is asymmetrical and commences preferentially in the ventral region. Quantitative differences in morphology and cytology exist among selected Simonsiella strains. Functional aspects of this dorsalventral differentiation are discussed with respect to the colonization and adherence of Simonsiella to mucosal squamous epithelial cells in its ecological habitat, the oral cavities of warm-blooded vertebrates.List of Abbreviations SEM scanning electron microscope - TEM transmission electron microscope     

Regulation of integrin-mediated adhesion at focal contacts by cyclic AMP

Cyclic AMP (cAMP) elevation causes diverse types of cultured cells to round partially and develop arborized cell processes. Renal glomerular mesangial cells are smooth, muscle-like cells and in culture contain abundant actin microfilament cables that insert into substratum focal contacts. cAMP elevation causes adhesion loss, microfilament cable fragmentation, and shape change in cultured mesangial cells. We investigated the roles of the classical vitronectin (αVβ3 integrin) and fibronectin (α5β1 integrin) receptors in these changes. Mesangial cells on vitronectin-rich substrata contained microfilament cables that terminated in focal contacts that stained with antibodies to vitronectin receptor. cAMP elevation caused loss of focal contact and associated vitronectin receptor. Both fibronectin and its receptor stained in a fibrillary pattern at the cell surface under control conditions but appeared aggregated along the cell processes after cAMP elevation. This suggested that cAMP elevation caused loss of adhesion mediated by vitronectin receptor but not by fibronectin receptor. We plated cells onto fibronectin-coated slides to test the effect of ligand immobilization on the cellular response to cAMP. On fibronectin-coated slides fibronectin receptor was observed in peripheral focal contacts where actin filaments terminated, as seen with vitronectin receptor on vitronectin-coated substrata, and in abundant linear arrays distributed along microfilaments as well. Substratum contacts mediated by fibronectin receptor along the length of actin filaments have been termed fibronexus contacts. After cAMP elevation, microfilaments fragmented and fibronectin receptor disappeared from peripheral focal contacts, but the more central contacts along residual microfilament fragments appeared intact. Also, substratum adhesion was maintained after cAMP elevation on fibronectin—but not on vitronectincoated surfaces. Although other types of extracellular matrix receptors may also be involved, our observations suggest that cAMP regulates adhesion at focal contacts but not at fibronexus-type extracellular matrix contacts. © 1993 Wiley-Liss, Inc.     

Fibronectin visualized by scanning electron microscopy immunocytochemistry on the substratum for cell migration in Xenopus laevis gastrulae

In amphibian gastrulae, scanning electron microscopy (SEM) has shown the presence of a network of extracellular fibrils on the inner aspect of the ectoderm layer, which serves as the substratum for migration by the presumptive mesoderm cells. In vitro experiments have shown that the fibril network promotes attachment and migration by mesoderm cells, and probably guides the migration by contact guidance. Filopodia of the migrating cells showed preferential attachment to the fibrils. Use of a colloidal gold probe for SEM immunocytochemistry has shown that fibrils observed by SEM contain fibronectin, probably as a major component. This provides direct evidence that the extracellular matrix containing fibronectin provides the substratum and guides cell migration in morphogenetic movement.     

Visualization of cellular focal contacts using a monoclonal antibody to 80 kD serum protein adsorbed on the substratum

We have obtained a monoclonal antibody to 80 kD protein of calf serum; this protein easily and uniformly adsorbs on glass from serum-containing media. Indirect immunofluorescence staining of chick and mouse embryo fibroblasts cultured in the presence of calf serum, fixed with formaldehyde and permeabilized with Triton X-100, revealed black non-fluorescent strips and dots under the ventral cell surface, whereas all other parts of the substratum under and between cells were highly fluorescent. The distribution of non-fluorescent regions coincided with the distributed of focal contacts of cells with the substratum, revealed by interference reflection microscopy, as well as with the distribution of vinculin-containing plaques. The dark regions were also associated with the ends of microfilament bundles revealed by immunofluorescence with an anti-actin antibody. Thus, non-fluorescent regions seen after anti-80 kD staining are parts of the substratum under the focal contacts. Visualization of focal contacts with anti-80 kD provides very contrasting and high resolution pictures. Evidence is presented that 80 kD protein is adsorbed to glass in the areas of focal contacts, but the antibodies used for staining cannot penetrate these contacts.     

Collagen modulates cell shape and cytoskeleton of embryonic corneal and fibroma fibroblasts: Distribution of actin, α-actinin,and myosin

In the embryo, fibroblasts migrating through extracellular matrices (ECM) are generally elongate in shape, exhibiting a leading pseudopodium with filopodial extensions, and a trailing cell process. Little is known about the mechanism of movement of embryonic cells in ECM, for studies of fibroblast locomotion in the past have been largely confined to observations of flattened cells grown on planar substrata. We confirm here that embryonic avian corneal fibroblasts migrating within hydrated collagen gels in vitro have the bipolar morphology of fibroblasts in vivo, and we show for the first time that highly flattened gerbil fibroma fibroblasts, grown as cell lines on planar substrata, can also respond to hydrated collagen gels by becoming elongate in shape. We demonstrate that the collagen-mediated change in cell shape is accompanied by dramatic rearrangement of the actin, α-actinin, and myosin components of the cytoskeleton. By immunofluorescence, the stress fibers of the flattened corneal fibroblasts grown on glass are seen to stain with antiactin, anti-α-actinin, and antimyosin, as has been reported for fibroma and other fibroblasts grown on glass. Stress fibers, adhesion plaques, and ruffles do not develop when the corneal or fibroma fibroblast is grown in ECM; these features seem to be a response to strong attachment of the cell underside to a planar substratum. When the fibroblasts are grown in ECM, antimyosin staining is distributed diffusely through the cytoplasm. Antiactin and anti-α-actinin stain the microfilamentous cell cortex strongly. We suggest that locomotion of the fibroblast in ECM is accompanied by adhesion of the cell to the collagen fibrils and may involve an interaction of the myosin-rich cytosol with the actin-rich filamentous cell cortex. Interestingly, the numerous filopodia that characterize the tips of motile pseudopodia of cells in ECM are very rich in actin and α-actinin, but seem to lack myosin; if filopodia use myosin to move, the interaction must be at a distance. Soluble collagen does not convert flattened fibroblasts on planar substrata to bipolar cells. Thus, the effect of collagen on the fibroblast cytoskeleton seems to depend on the presence of collagen fibrils in a gel surrounding the cell.     

Wettability of substrata controls cell-substrate and cell-cell adhesions

The maintenance of endothelial cell (EC) monolayer architecture requires stable adhesions not only between neighboring cells but also between cells and the extracellular matrix. While the influence of biomaterials surface wettability on cell-substratum adhesion is rather well studied, its impact on cell-cell cohesion has not been extensively investigated. In the present study a model system consisting of hydrophilic and hydrophobic glass pre-coated with fibronectin and fibrinogen was used to study the influence of surface wettability on both types of cell adhesions. It was demonstrated that the substrate wettability controls the adhesion and cytoskeletal organization of endothelial cells, which has an impact on the subsequent ability of cells to establish stable cell-cell cohesions. These effects were related to the accessibility of specific domains of the adsorbed proteins. While the hydrophobic substratum promoted cell-cell cohesion, on hydrophilic substrata cell-substrate adhesion was dominant. In addition, evidence for an influence of surface wettability on the cross talk between integrins and cadherins was found.     

Proteolytic activity of specialized surface protrusions formed at rosette contact sites of transformed cells

Surface protrusions at the leading edge of a moving cell that make contact with the surrounding extracellular matrix (ECM) are its main motor for locomotion and invasion. Chicken embryonic fibroblasts transformed by Rous sarcoma virus (RSV-CEF) form specialized membrane rosette-shaped contact sites on planar substrata as shown by interference reflection microscopy (IRM). Such activity is lacking in normal cells. These rosette contacts are more labile than other adhesion sites, such as focal and close contacts. Ultrastructural studies demonstrate that rosettes are sites at which membrane protrusions from the ventral cell surface contact the substratum. These protrusions are filled with meshworks of microfilaments and contain the pp60src oncogene product, actin, vinculin, and alpha-actinin. However, unlike focal contacts, at the rosettes these proteins interact to extend a highly motile membrane. Rosettes have the biological activity of degrading ECM components, as demonstrated by (1) local degradation of fibronectin substrata at sites of rosette contacts, but not focal and close contacts; (2) localization of putative antiprotease antibody at sites of rosette contacts, but not at focal an close contacts; and (3) local disruption of fibronectin matrix at sites of protrusive activity seen by transmission electron microscopy (TEM). In addition, formation of the rosette contact is insensitive to the ionophore monensin, and to inhibitors of proteolytic enzymes, while local fibronectin degradation at rosette contacts is inhibited by inhibitors of metalloproteases, 1,10-phenanthroline and NP-20. I consider these membrane protrusions of the rosette contacts in RSV-transformed cells specialized structural entities--invadopodia--that are involved in the local degradation of the ECM.     

Nitrate reductase and the membrane composition of pleiotropic chlorate resistant mutants of Escherichia coli K-12.

Attachment of rat hepatocytes to collagen but not to fibronectin substrata was efficiently inhibited by antibodies against the hepatocyte surface. Further analyses of this inhibition suggested that hepatocyte attachment to collagen involves cell surface antigens which are not identical to membrane bound fibronectin or collagen.Rabbit antibodies against rat fibronectin inhibited hepatocyte attachment to rat fibronectin but not to collagen or rabbit fibronectin. After plasmin digestion of fibronectin, peptides were isolated that lacked affinity for collagen but could serve as a substratum for hepatocyte attachment. These results suggested that attachment to fibronectin does not involve membrane bound fibronectin or collagen.     

Modulation of protein synthesis and secretion by substratum in primary cultures of rat hepatocytes

Hepatocytes isolated by perfusion of adult rat liver and cultured on substrata consisting of one or more of the major components of the liver biomatrix (fibronectin, laminin, type IV collagen) have been examined for the synthesis of defined proteins. Under these conditions, tyrosine amino transferase, a marker of hepatocyte function, is maintained at similar levels in response to dexamethasone over 5 days in culture on each substratum, and total cellular protein synthesis remains constant. By contrast, there is a rapid decrease in synthesis and secretion of albumin and a 3-7-fold increase in synthesis and secretion of alpha-fetoprotein which are most marked on a laminin substratum, but least evident on type IV collagen, and an increased synthesis of fibronectin and type IV collagen. The newly synthesized matrix proteins are present in the cell layer as well as in cell secretions. The enhanced synthesis of fibronectin is less in cells seeded onto a fibronectin substratum than on laminin or type IV collagen substrata, and its synthesis by hepatocytes seeded onto a mixed substratum of laminin and fibronectin is down-regulated by fibronectin in a dose-related manner. Similarly, type IV collagen synthesis is less when the cells are seeded on the homologous matrix protein substratum than on heterologous substrata. These results indicate that hepatocytes cultured in serum-free medium on substrata composed of components of the liver biomatrix maintain certain functions of the differentiated state (tyrosine amino transferase), lose others (albumin secretion) and switch to increased synthesis of matrix components as well as fetal markers such as alpha-fetoprotein. The magnitude of these effects depends on the substratum on which the hepatocytes are cultured.     

A Role for Topographic Cues in the Organization of Collagenous Matrix by Corneal Fibroblasts and Stem Cells

Human corneal fibroblasts (HCF) and corneal stromal stem cells (CSSC) each secrete and organize a thick stroma-like extracellular matrix in response to different substrata, but neither cell type organizes matrix on tissue-culture polystyrene. This study compared cell differentiation and extracellular matrix secreted by these two cell types when they were cultured on identical substrata, polycarbonate Transwell filters. After 4 weeks in culture, both cell types upregulated expression of genes marking differentiated keratocytes (KERA, CHST6, AQP1, B3GNT7). Absolute expression levels of these genes and secretion of keratan sulfate proteoglycans were significantly greater in CSSC than HCF. Both cultures produced extensive extracellular matrix of aligned collagen fibrils types I and V, exhibiting cornea-like lamellar structure. Unlike HCF, CSSC produced little matrix in the presence of serum. Construct thickness and collagen organization was enhanced by TGF-ß3. Scanning electron microscopic examination of the polycarbonate membrane revealed shallow parallel grooves with spacing of 200–300 nm, similar to the topography of aligned nanofiber substratum which we previously showed to induce matrix organization by CSSC. These results demonstrate that both corneal fibroblasts and stromal stem cells respond to a specific pattern of topographical cues by secreting highly organized extracellular matrix typical of corneal stroma. The data also suggest that the potential for matrix secretion and organization may not be directly related to the expression of molecular markers used to identify differentiated keratocytes.     

Contact formation by fibroblasts adhering to heparan sulfate-binding substrata (fibronectin or platelet factor 4)

The process of cell-substratum adhesion of BALB/c 3T3 fibroblasts on fibronectin (FN)-coated substrata was compared with that of cells adhering to substrata coated with the heparan sulfate (HS)-binding protein, platelet factor four (PF4). FN has binding domains for HS and an unidentified cell surface receptor, whereas PF4 binds to only HS on the surface of the cell. The attachment and early spreading sequences of cells on either substratum were similar as shown by scanning electron microscopy (SEM). Within 2 h of spreading, cells on FN developed typical fibroblastic morphologies, whereas those on PF4 lacked polygonal orientations and formed numerous broadly spread lamellae. Interference reflection microscopic analysis indicated that PF4-adherent cells formed only close adhesive contacts, whereas FN-adherent cells formed both close contacts and tight focal contacts. Cells on either substratum responded to Ca2+ chelation with EGTA by rounding up, but remained adherent to the substratum by relatively EGTA-resistant regions of the cell's undersurface, demonstrating that cell surface HS by binding to an appropriate substratum is capable of initiating a Ca2+-dependent spreading response. The EGTA-resistant substratum-attached material on PF4 was morphologically similar to that on FN, the latter of which was derived from both tight focal contacts and discrete specializations within certain close contacts. These studies show that heparan sulfate proteoglycans on the surface of these cells can participate in the formation of close contact adhesions by binding to an appropriate substratum and suggest that sub-specializations within close contact adhesions may evolve into tight focal contacts by the participation of an unidentified cell surface receptor which binds specifically to fibronectin but not to PF4. In addition, the functional role of FN in tight focal contact formation is demonstrated.     

Ultrastructural and tissue-culture studies on the role of fibronectin,collagen and glycosaminoglycans in the migration of neural crest cells in the fowl embryo

Summary The initial migration of neural crest (NC) cells into cell-free space was studied by transmission electron microscopy at trunk levels of fowl embryos, some of which were fixed in the presence of ruthenium red. Migrating NC cells occurred in zones which contained fewer ruthenium-red stained 15–40 nm diameter granules than other regions. The ruthenium-red stained granules were linked by similarly stained thin ( 3 nm diameter) microfibrils. The granules resemble proteoglycan and the microfibrils may be hyaluronate. NC cells contacted thicker ( 10 nm diameter) fibrils and interstitial bodies, which did not require ruthenium red for visualization. Cytoplasmic microfilaments were sometimes aligned at the point of contact with the extracellular fibrils, which may be fibronectin and collagen.Phase-contrast time-lapse videotaping and scanning electron microscopy showed that NC cells of the fowl embryo in vitro migrated earlier and more extensively on glass coated with fibronectin-rich fibrous material and adsorbed fibronectin molecules than on glass coated with collagen type I (fibres and adsorbed molecules). NC cells became completely enmeshed in fibronectin-rich fibres, but generally remained on the surface of collagen-fibre gels. When given a choice, NC cells strongly preferred fibronectin coatings to plain glass, and plain glass to dried collagen gels. NC cells showed a slight preference for plain glass over glass to which collagen was adsorbed. Addition to the culture medium of hyaluronate (initial conc. 20 mg/ml), chondroitin (5 mg/ml) and fully sulphated chondroitin sulphate and dermatan sulphate (up to 10 mg/ml) did not drastically alter NC cell migration on fibronectin-rich fibrous substrates. However, partially desulphated chondroitin sulphate (5mg/ml) strongly retarded the migration of NC cells.The in vivo and in vitro studies suggest that fibronectin may dictate the pathways of NC cell migration by acting as a highly preferred physical substrate. However, the utilization of these pathways may be reduced by the presence of proteoglycans bearing undersulphated chondroitin sulphate.Abbreviations NC neural crest - ECM extracellular material - GAG glycosaminoglycan - FN fibronectin - CIG cold insoluble globulin - TEM transmission electron microscopy - SEM scanning electron microscopy - DMEM-H HEPES buffered Dulbecco's modified Eagle's medium - FCS foetal calf serum - CEE chick embryo extract - SDS-PAGE sodium dodecyl sulphate-polyacrylamide gel electrophoresis - PBS phosphate-buffered saline