Transforming growth factor beta increases cell surface binding and assembly of exogenous (plasma) fibronectin by normal human fibroblasts.

Transforming growth factor beta (TGF-beta) enhances the cell surface binding of 125I-fibronectin by cultured human fibroblasts. The effect of TGF-beta on cell surface binding was maximal after 2 h of exposure to TFG-beta and did not require epidermal growth factor or protein synthesis. The enhancement was dose dependent and was found with the 125I-labeled 70-kilodalton amino-terminal fragment of fibronectin as well as with 125I-fibronectin. Treatment of cultures with TGF-beta for 6 h resulted in a threefold increase in the estimated number of fibronectin binding sites. The increase in number of binding sites was accompanied by an increased accumulation of labeled fibronectin in detergent-insoluble extracellular matrix. The effect of TGF-beta was biphasic; after 6 h of exposure, less labeled fibronectin bound to treated cultures than to control cultures. Exposure of cells to TGF-beta for greater than 6 h caused a two- to threefold increase in the accumulation of cellular fibronectin in culture medium as detected by a quantitative enzyme-linked immunosorbent assay. The second phase of the biphasic effect and the increase in soluble cellular fibronectin were blocked by cycloheximide. Immunofluorescence staining of fibroblast cultures with antifibronectin revealed that TGF-beta caused a striking increase in fibronectin fibrils. The 70-kilodalton amino-terminal fragment of fibronectin, which blocks incorporation of fibronectin into extracellular matrix, blocked anchorage-independent growth of NRK-49F cells in the presence of epidermal growth factor. Our results show that an increase in the binding and rate of assembly of exogenous fibronectin is an early event preceding the increase in expression of extracellular matrix proteins. Such an early increase in cell surface binding of exogenous fibronectin may be a mechanism whereby TGF-beta can modify extracellular matrix characteristics rapidly after tissue injury or during embryonic morphogenesis.     

Extracellular matrix assembly of cell-derived and plasma-derived fibronectins by substrate-attached fibroblasts

Using a previously described model system for the incorporation of plasma fibronectin into the extracellular matrix (McKeown-Longo, P.J. and Mosher, D.F., 1985. J. Cell Biol., 100:364-374), we compared the binding of cell-derived and plasma-derived fibronectins to human fibroblast cell layers. Binding was measured in time course experiments using metabolically labeled cell-derived, iodinated cell-derived, and iodinated plasma-derived fibronectins. The kinetics of matrix assembly of cell- and plasma-derived fibronectins were the same. Competitive binding curves using intact fibronectin or the 70-kD amino-terminal fragment of fibronectin suggested that cell surface binding sites have equal affinity for cell- and plasma-derived fibronectins. Iodinated fibronectins did not bind to isolated matrices containing collagen type I, fibronectin, and thrombospondin. These results suggest that fibroblasts do not distinguish between cell-derived and plasma-derived fibronectins when assembling exogenous fibronectin into extracellular matrix.     

Role of fibronectin as a growth factor for fibroblasts

Fibroblast replication is regulated by exogenous signals provided by growth factors, mediators that interact with the target cell surface and signal the cell to proliferate. A useful model of growth regulation, the "dual control model," suggests that growth factors can be grouped either as competence factors or as progression factors, and that optimal replication of fibroblasts requires the presence of both types of growth factors. Although most growth factors are soluble mediators, recent studies have demonstrated that, for some cell types, the extracellular matrix can replace the requirement for a competence factor. Since fibronectin is an important constituent of the extracellular matrix that interacts with specific domains on the fibroblast surface, we examined the ability of fibronectin to act as a competence factor to promote the growth of human diploid fibroblasts. To accomplish this, fibronectins purified from two sources, human plasma and human alveolar macrophages, were tested for their ability to (a) stimulate fibroblast replication in serum-free medium containing characterized progression factors (insulin or alveolar macrophage-derived growth factor); (b) provide a growth-promoting signal early in G1. Fibronectin stimulated fibroblast replication in a dose-dependent manner in the presence of a fixed dose of a progression factor. Conversely, fibronectin conferred on previously unresponsive fibroblasts the ability to replicate in a dose-dependent manner when cultured with increasing amounts of a progression factor. Moreover, fibronectin signaled growth-arrested fibroblasts to traverse G1 approximately 4 h closer to S phase. No differences were observed in the ability of plasma or macrophage fibronectins to provide a competence signal for fibroblast replication. Since fibronectin is a major component of the extracellular matrix, these observations suggest that it may provide at least one of the signals by which the matrix conveys the "competence" that permits fibroblasts to replicate in the presence of an appropriate progression signal.     

Immunomorphological research on the formation of the extracellular matrix in epithelial cell cultures

Formation of extracellular matrix structures in cultures of rat liver epithelial nontransformed cell line IAR2 was studied with antisera to fibronectin, laminin and type IV collagen by immunofluorescence and immunoelectron microscopy of platinum replicas. Fibronectin formed peripheral spots of variable size some of which outlined free cell edges, as well as fibrils located towards the center of single cells or of cellular islands. Similarly distributed structures were seen in isolated matrices. Codistribution of fibronectin and actin was observed only for the peripheral line of fibronectin spots and marginal circular actin bundle. Basement membrane components. laminin and type IV collagen, formed mainly spots of variable size predominantly beneath the cell or each cell in an island. Occasional fibrils were seen also. Essentially the same results were obtained by immunofluorescence and immunogold electron microscopy. Cytochalasin D treated cells displayed spots of both fibronectin and laminin. The relevance of previously postulated receptor-mediated assembly of extracellular matrix structures to the epithelial cells is discussed.     

Immunological characterization of a major transformation-sensitive fibroblast cell surface glycoprotein. Localization, redistribution, and role in cell shape

The major cell surface glycoprotein of chick embryo fibroblasts, cellular fibronectin (formerly known as CSP or LETS protein), was purified and used to produce monospecific antisera. After affinity purification, the anti-fibronectin was used to investigate fibronectin's localization, its transfer from intracellular to extracellular pools, its antibody-induced redistribution on the cell surface, and its role in cell shape. Anti-fibronectin localizes to extracellular fibrils located under and between sparse cells, and to a dense matrix that surrounds confluent cells. Cellular fibronectin is also present in granular intracytoplasmic structures containing newly synthesized fibronectin before secretion. This intracellular staining disappears 2 h after treatment with cycloheximide or puromycin, and returns after removal of these protein synthesis inhibitors. In pulse-chase experiments using cycloheximide, fibronectin was sequentially transferred from the intracellular to the fibrillar extracellular forms. Transformation of chick fibroblasts results in decreases in both extracellular and intracellular fibronectin, and in altered cell shape. Treatment of untransformed chick fibroblasts with anti-fibronectin results in rapid (30 min) alteration to a rounder cell shape resembling that of many transformed cells. These rapid shape changes are followed by a slow, antibody-induced redistribution of fibronectin to supranuclear caplike structures. This "capping" is inhibited by metabolic inhibitors. Reconstitution of cell surface fibronectin onto transformed cells restores a more normal fibroblastic phenotype. The reconstituted fibronectin on these cells organizes into fibrillar patterns similar to those of untransformed cells. As with untransformed cells, treatment of these reconstituted cells with anti-fibronectin also results in cell rounding and "capping" of fibronectin.     

Different substrates influence the expression of intermediate filaments and the deposition of basement membrane proteins

Summary A primary culture of serous cystadenocarcinoma of the ovary was used to study the expression of intermediate filament proteins and the deposition of basal lamina proteins. It was found that cells grown on type I and IV collagens or in collagen gels failed to express vimentin, which was readily demonstrable in cultures of the same cells grown on plastic or glass. Furthermore cells grown in collagen gels formed colonies demonstrating a cystic architecture Unlike what is commonly observed on glass or plastic where laminin and fibronectin are deposited as disorganized fibrils in the extracellular space, in or on collagen these proteins appear solely at the interface between the epithelial cells and matrix. The results suggest that the extracellular matrix influences the cytoskeletal organization of the intermediate filaments and determines cell polarity. They confirm that collagen substrates permit epithelial cell cultures to progress toward a more differentiated state. Supported by grants from the Italian Assciation for Cancer Research (AIRC).     

The structure of fibronectin and its role in cellular adhesion

Fibronectin is a large, adhesive glycoprotein which is found in a number of locations, most notably on cell surfaces, in extracellular matrixes, and in blood. Fibronectin has been detected in all vertebrates tested and in many invertebrates. Its presence in sponges is significant because this suggests that fibronectin may have appeared very early in evolution, possibly with the most primitive multicellular organisms. Cellular and plasma fibronectins have many striking similarities. However, the locations of the polypcptide chain differences between these two proteins indicate that plasma fibronectin cannot be derived from cellular fibronectin by means of simple post-translational proteolysis. Instead, these different types of fibronectin may be products of different genes or of differentially spliced messenger RNA molecules. Amniotic fluid fibronectin is possibly a third form of the protein. Cellular and plasma fibronectins are composed of at least six protcaseresistant domains which contain specific binding sites for actin, gelatin, heparin, Staphylococcus aureus, transglutarninase, fibrin, DNA, and a cell surface receptor. The relative locations of these domains have been mapped in the primary structure of fibronectin. The cell surface receptor for fibronectin has not been positively identified, but may be a glycoprotein, a glycolipid, or a complex of the two. Although cell-substratum adhesion is mediated by fibronectin, the locations of the areas of closest approach of the cell to the substratum (the adhesion plaques) and fibronectin are not coincident under conditions of active cell growth. Under conditions of cell growth arrest in low scrum concentrations, some fibronectin may become localized at the adhesion plaques. Models describing the domain structure of fibronectin and the molecular organization of the adhesion plaque area are presented.     

Factor XIIIa-mediated cross-linking of fibronectin in fibroblast cell layers. Cross-linking of cellular and plasma fibronectin and of amino-terminal fibronectin fragments

Factor XIIIa cross-links plasma fibronectin as it is being assembled into the extracellular matrix of cultured human skin fibroblasts (Barry, E. L. R., and Mosher, D. F. (1988) J. Biol. Chem. 262, 10464-10469). We have further characterized this process. Fibroblasts were metabolically labeled with proline in the presence or absence of ascorbate and Factor XIIIa. Endogenous fibronectin in the extracellular matrix was cross-linked by Factor XIIIa. There was no evidence for cross-linking of collagenous proteins. Fibro-blast cell layers were incubated with iodinated 27-kDa heparin-binding or 70-kDa collagen- and heparin-binding amino-terminal fibronectin fragments. Factor XIIa cross-linked the fragments into high molecular weight aggregates. The amounts of cross-linked fragments reaches a steady state after 1 to 2 h, whereas intact fibronectin continues to be cross-linked for 24 h. When fibroblast cell layers were pulsed with iodinated fibronectin or amino-terminal fragments and Factor XIIIa was included in the chase media, the high molecular weight aggregates were formed in a step-wise manner. The smallest cross-linking steps were to high molecular weight extracellular matrix molecules forming approximately 270-, 300-, and 440-kDa complexes for the 27-kDa fragment, 70-kDa fragment, and intact fibronectin, respectively. When iodinated fibronectin was bound to fibroblast cell layers and chased into the matrix pool in the absence of Factor XIIIa, it could also be cross-linked into high molecular weight complexes when Factor XIIIa was added to the media. These results, therefore, indicate that both cellular and plasma fibronectin and amino-terminal fragments are cross-linked specifically by Factor XIIIa, that the cross-linking is probably to other fibronectin molecules rather than to collagenous proteins, and that both assembling and assembled fibronectin are substrates for Factor XIIIa.     

Factor XIII cross-linking of fibronectin at cellular matrix assembly sites

We describe the effect of activated Factor XIII (Factor XIIIa, plasma transglutaminase) on the incorporation of plasma fibronectin into extracellular matrix by cultured human fibroblasts. In the absence of added Factor XIIIa, fibronectin binds to cultured fibroblast cell layers and is assembled into disulfide-bonded multimers of the extracellular matrix. When Factor XIIIa was included in the binding medium of skin fibroblasts, accumulation of 125I-fibronectin in the deoxycholate-insoluble matrix was increased. Fibronectin accumulating in the cell layer was cross-linked into nonreducible high molecular weight aggregates. The 70-kDa amino-terminal fragment of fibronectin inhibited the binding and cross-linking of 125I-fibronectin to cell layers, whereas fibrinogen had little effect. When 125I-fibronectin was incubated with isolated matrices or with cell layers pretreated with cytochalasin B, it did not bind and could not be cross-linked by Factor XIIIa into the matrix. HT-1080 human fibrosarcoma cells bound exogenous fibronectin following treatment with dexamethasone; Factor XIIIa cross-linked the bound fibronectin and caused its efficient transfer to the deoxycholate-insoluble matrix. These results indicate that exogenous fibronectin is susceptible to Factor XIIIa-catalyzed cross-linking at cellular sites of matrix assembly. Thus, Factor XIIIa-mediated fibronectin cross-linking complements disulfide-bonded multimer formation in the stabilization of assembling fibronectin molecules and thus enhances the formation of extracellular matrix.     

Differential effects of soluble and immobilized fibronectins on aortic endothelial cell proliferation and attachment

Summary We studied the effects of soluble and immobilized forms of plasma fibronectin on bovine aortic endothelial cell (AEC) proliferation and attachment. Soluble fibronectin stimulated AEC growth at 10 μg/ml, but at higher concentrations of soluble fibronectin AEC growth was progressively inhibited. The growth rates of arterial smooth muscle cells (ASMC) and dermal fibroblasts (DF) were not altered by soluble fibronectin concentrations of 10 to 100 μg/ml. Plasma fibronectin, immobilized by attachment to culture dish surfaces, had no significant effects on the proliferation of any of the cell types examined. The attachment rates of AEC were decreased in the presence of 50 μg/ml soluble fibronectin. Immobilized fibronetin increased the rate of AEC attachment, but had no significant effects on ASMC or DF attachment; however, 12 h after plating there was nearly 100% attachment in all groups, whether or not fibronectin was present in the system. That soluble and immobilized fibronectins elicit disparate cellular responses is consistent with published reports of different cell surface receptors for different forms of the protein; in this manner, cells enmeshed in an interstitial matrix containing immobilized fibronectin could still respond to soluble fibronectin in the extracellular milieu. These studies were supported in part by grant EY-0229 from the National Institutes of Health, Bethesda, MD.     

Fibroblast cellular and plasma fibronectins are similar but not identical

Fibronectins are multimeric, adhesive glycoproteins present on cell surfaces and circulating in blood. Cellular fibronectin produced by fibroblasts in vitro and fibronectin isolated from plasma are known to be very similar immunologically and biochemically. We investigated whether or not they are identifical. Purified chicken and human cell-surface fibronectins are 150-fold more active in hemagglutination of fixed erythrocytes than plasma fibronectins. Cell-surface fibronectin is also 50-fold more active in restoring a more normal morphology to transformed cells originally missing the protein. However, in two other assays that measure cell attachment to collagen and cell spreading, cell-surface and plasma fibronectins have identical specific activities. In sodium dodecyl sulfate polyacrylamide gels, the subunits of human and chicken plasma fibronectins have significantly smaller apparent subunit molecular weights than cellular fibronectins present on cell surfaces or secreted into culture media. These differences are also present in a characteristic large subfragment of both forms of fibronectin after limited proteolysis by trypsin. We conclude that by both biological and biochemical criteria, cellular and plasma fibronectins are similar but not identical.     

Effect of retinyl acetate on the assembly of the fibronectin extracellular matrix and the processing of the fibronectin receptor β subunit of confluent C3H/10T1/2 mouse embryo fibroblasts

The mouse embryo fibroblast cell line, C3H/10T1/2, synthesized and deposited a large amount of fibronectin especially in the pericellular matrix. Confluent cultures of these cells cultured in the presence of 0.3 μg/ml of retinyl acetate released cell surface fibronectin and the extracellular matrix fibronectin fibrils were disorganized. The immunoblot analysis demonstrated that the number of the fibronectin receptor was decreased in the prolonged culturing of retinyl acetate-treated cells. Immunoprecipitation of ³⁵S-methionine pulse-chase labeled cell extracts by antifibronectin receptor antibody indicated that about one-half of the pre-β subunit was processed and converted to the mature form in control cells, and only about one-fourth of the pre-β subunit was processed in the retinyl acetate-treated confluent cells. 1-deoxymannojirimycin (MNJ), which is an inhibitor of oligosaccharide processing, induced disorganization of the extracellular matrix fibronectin assembly similar to that observed with retinyl acetate. The results of this study suggest that a mechanism of action of retinyl acetate is inhibition of the glycosylation during processing of the fibronectin receptor, a step necessary for fibronectin binding and for assembly of the extracellular matrix. © 1993 Wiley-Liss, Inc.     

Matrix assembly sites for exogenous fibronectin are decreased on human fibroblasts after treatment with agents which increase intracellular cAMP

Studies with cultured fibroblasts have shown that plasma as well as cellular fibronectin can be organized into fibrillar structures and that this organization is mediated by sites at the cell surface. Treatment of human skin fibroblasts with cholera toxin resulted in a prompt decrease in the number of binding sites for 125I-labeled plasma fibronectin and a 125I-labeled 70-kDa amino-terminal fragment of fibronectin. This decrease was accompanied by less incorporation of labeled fibronectin into deoxycholate-insoluble extracellular matrix. Binding of 125I-fibronectin was also decreased in cultures treated with epinephrine, isoproterenol, or forskolin. These results, therefore, indicate that G proteins and the adenylate cyclase system are involved in regulation of fibronectin matrix assembly sites may be one mechanism whereby hormones or growth factors can modify extracellular matrix characteristics.     

Conditioned medium from endothelial cell cultures can restore the normal phenotypic expression of vascular endothelium maintained in vitro in the absence of fibroblast growth factor

Bovine vascular endothelial cells continuously maintained and grown in the presence of FGF adopt at confluence the configuration of a cell monolayer composed of contact-inhibited cells which do not overgrow each other and which are highly flattened and closely apposed. Such cultures exhibit structural and morphological characteristics similar to those observed with their in vivo counterparts. These include the production of an extracellular matrix consisting mostly of basement membrane collagen and fibronectin localized exclusively beneath the cell monolayer, but not on top of it, as well as a nonthrombogenic, blood-compatible apical cell surface. Removal of fibroblast growth factor (FGF) from adult bovine aortic endothelial cell (ABAE) cultures results within three passages in the loss by the cells of their characteristic contact-inhibited morphology. The cells, which during their logarithmic growth phase divide with a greatly increased doubling time, become larger and more elongated. Confluent cultures, instead of adopting the morphology of a contact inhibited cell monolayer, are now composed of overgrowing cells. Parallel with the morphological alterations taking place within the culture, the cells also lose the polarity of cell surfaces characteristics of the vascular endothelium. Formation of an extracellular matrix composed primarily of fibronectin and collagen types I, III, and IV is observed on both the apical and basal cell surfaces. Platelets which previously did not bind to the apical cell surface now become capable of binding to it. CSP-60, a major cell surface protein present in highly confluent and contact-inhibited vascular endothelial cell cultures, can no longer be detected. Exposure of confluent endothelial cell cultures, maintained in the absence of FGF to medium conditioned by cells which had been grown in the presence of FGF, but maintained in its absence upon reaching confluence led, within four to eight days, to a reversion of the altered phenotype. This medium has little or no mitogenic activity and retains a full activity in the absence of serum or after depletion of its fibronectin content by affinity chromatography on a gelatin-Sepharose column. Cultures which were previously composed of cells growing in multiple layers reorganized into a single cell monolayer composed of closely apposed and highly flattened cells. The cultures thereby regained the contact-inhibited morphology characteristic of the vascular endothelium. Concomitant with this cellular reorganization, the extracellular matrix disappeared from the apical cell surface, the cells regained their nonthrombogenic properties, and CSP-60 reappeared as one of the major cell surface proteins. These results suggest that vascular endothelial cells secrete a soluble factor(s) which can restore the normal morphology and function lost following removal of FGF from the medium. Such a factor(s) may be involved in maintaining the differentiated state of the vascular endothelium.     

Rapid isolation of metaphase chromosomes containing high molecular weight DNA

Normal rat kidney cells were cultured in medium supplemented with normal fetal bovine serum (FBS) or FBS depleted of fibronectin. The cell surface fibronectin of these cultures was visualized by indirect immunofluorescence using species-specific antisera for either rat fibronectin or bovine fibronectin. Anti-rat-fibronectin revealed fibrillar structures on the cells grown in either normal medium or fibronectin-depleted medium. Anti-bovine fibronectin revealed similar fibrillar networks, but only on the cells grown in medium containing bovine fibronectin. Staining in each case was abolished by absorption with the homologous antigen. It appears that exogenous fibronectin was incorporated into the same structures as endogenous fibronectin. This finding suggests that circulating fibronectin may serve as a building block for the assembly of extracellular matrix, possibly by cells which are incapable of synthesizing it.     

Induction of fibronectin matrix assembly in human fibrosarcoma cells by dexamethasone

Previous studies have suggested that the assembly of fibronectin into the extracellular matrix of cultured fibroblasts is mediated by specific matrix assembly receptors that recognize a binding site in the amino terminus of the fibronectin molecule (McKeown-Longo, P.J., and D.F. Mosher, 1985, J. Cell Biol., 100:364-374). In the presence of dexamethasone, human fibrosarcoma cells (HT-1080) acquired the ability to specifically bind exogenous plasma fibronectin and incorporate it into a detergent-insoluble extracellular matrix. Dexamethasone-induced fibronectin binding to HT-1080 cells was time dependent, dose dependent, and inhibited by cycloheximide. Saturation binding curves indicated that dexamethasone induced the appearance of 7.7 X 10(4) matrix assembly receptors per cell. The induced receptors exhibited a dissociation constant (KD) for soluble fibronectin of 5.0 X 10(-8) M. In parallel experiments, normal fibroblasts exhibited 4.1 X 10(5) receptors (KD = 5.3 X 10(-8) M) per cell. In the presence of cycloheximide, the induced fibronectin-binding activity on HT-1080 cells returned to uninduced levels within 12 h. In contrast, fibronectin-binding activity on normal fibroblasts was stable in the presence of cycloheximide for up to 54 h. The first-order rate constant (Kt = 2.07 X 10(-4) min-1) for the transfer of receptor-bound fibronectin to extracellular matrix was four- to fivefold less than that for normal fibroblasts (Kt = 1.32 X 10(-3) min-1). Lactoperoxidase-catalyzed iodination of HT-1080 monolayers indicated that a 48,000-mol-wt cell surface protein was enhanced with dexamethasone. The results from these experiments suggest that dexamethasone induces functional matrix assembly receptors on the surface of HT-1080 cells; however, the rate of incorporation of fibronectin into the matrix is much slower than that of normal fibroblasts.     

Novel Form of Fibronectin from Zebrafish Mediates Infectious Hematopoietic Necrosis Virus Infection

The presence of a novel form of zebrafish fibronectin (FN2) on the cell surface increased the cell's susceptibility to infection by infectious hematopoietic necrosis virus (IHNV). Unlike other fibronectins, FN2 possesses a truncated structure and accumulates on the cell surface instead of in the extracellular matrix. Fish embryo cells expressing recombinant FN2 were more susceptible to IHNV infection, with a greater percentage of cells exhibiting cytopathic effect (CPE) compared to nontransfected control cells. Incubation of nontransfected cells with soluble recombinant FN2 increased IHNV infection, as measured by plaque assay. The number of plaques increased in correlation with the amount of protein added and the length of time that cells were incubated with the protein. Incubation of IHNV with soluble FN2 before addition to cells also increased infection. FN2 immobilized on the culture surface inhibited IHNV infection. The results indicate that FN2 present on the cell surface is able to mediate IHNV attachment and cell entry.     

Plasma fibronectin is synthesized and secreted by hepatocytes

Primary cultures of hepatocytes of rats and hamsters were established and examined for the synthesis and secretion of fibronectin. Hepatocytes of both species secreted fibronectin as a soluble dimeric protein which could be purified by its affinity for gelatin and using specific antisera. Plasma and cellular fibronectins could be clearly resolved on two-dimensional gels. In both species, the majority of the fibronectin secreted by hepatocytes was of the plasma type, as shown by analyses on one- and two-dimensional gels. The secretion of plasma fibronectin increased with time in culture, both in absolute terms and relative to the secretion of albumin. Even during the first day of culture, the secretion of fibronectin relative to that of albumin appeared to be sufficient to account for the relative levels of these two proteins in plasma. Hepatocytes of both species secreted preferentially the chain of plasma fibronectin with higher apparent molecular weight, although the faster migrating chain was also secreted. In addition, hamster hepatocytes cultured for 2 or more days appeared to secrete a cellular form of fibronectin. Possible origins for the different chain types of cellular and plasma fibronectins are discussed.     

A truncated form of fibronectin is expressed in fish and mammals.

Fibronectin is a highly conserved and well-characterized extracellular matrix protein that plays a crucial role in many cellular processes. In amphibians, birds and mammals, multiple isoforms of fibronectin are generated by differential RNA splicing at three exons. Previously, we identified a novel form of fibronectin (FN2) in zebrafish that possesses a truncated structure and is generated by a unique RNA splicing pattern. Unlike other fibronectins, FN2 exists as a monomer and localizes on the cell surface. Here we show that this novel fibronectin is evolutionarily conserved, found in multiple species of fish and in mice and humans. The unusual structure, unique cellular distribution and conserved pattern of expression indicate that FN2 may have a novel function in normal cellular processes.     

Two-dimensional peptide mapping of fibronectins from bovine aortic endothelial cells and bovine plasma

We have made a comparison between plasma and endothelial cell fibronectin, since these cells are in intimate contact with plasma $\text{in vivo}$. Cellular and secreted fibronectins were purified from cloned lines of adult bovine aortic endothelial cells, and compared to purified bovine plasma fibronectin by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and two-dimensional peptide mapping. When unreduced, all three fibronectins migrated on gels as single bands with M_r 440,000. After reduction, cellular and secreted fibronectins migrated on gels as single bands with M_r 220,000, but plasma fibronectin migrated as two bands with M_r 220,000 and 210,000. All three fibronectins, including the two subunits of plasma fibronectin, had identical structures by peptide mapping analysis.