An irreversible tissue inhibitor of collagenase in human amniotic fluid: Characterization and separation from fibronectin

Soluble fibronectin isolated from human plasma and amniotic fluid by gelatin-Sepharose affinity chromatography was tested for inhibitory activity against specific collagenase secreted by human and rabbit fibroblasts. The fibronectin preparation derived from plasma showed little inhibition, but the one derived from amniotic fluid contained potent inhibitory activity against collagenase. This activity was separated from fibronectin on a DE-52 cellulose column and did not cross-react with antibodies to fibronectin. The inhitor was a glycoprotein that was partially purified from amniotic fluid by concanavalin A-Sepharose affinity chromatography. Inhibition was irreversible and enzyme activity was not recovered after reaction with latent or activated collagenase by either trypsin or organomercurial treatment.     

Further characterization of porcine plasma fibronectin which contains fucosylated carbohydrate chains

Porcine plasma fibronectin and its functional four fragments produced by cathepsin B digestion were examined for biological, immunochemical and biochemical properties. Native fibronectin, 150-kDa and 130-kDa fragments exhibited similar cell attachment-promoting activity to each other. In an Ouchterlony double immunodiffusion system, these three polypeptides formed a precipitin line with anti-fibronectin antiserum, while the 50-kDA and 30-kDa fragments did not. The 150-kDa and 130-kDa fragments contained free sulfhydryl(s). The glycopeptide fractions were prepared by pronase digestion of porcine and human plasma fibronectin, and radiolabeled with [¹⁴C]acetic anhydride. The results of affinity chromatography with concanavalin A and lentil lectin immobilized on agarose indicated that the porcine glycopeptide fraction was different from the human fraction in that a larger part (58%) of the former was bound to lentil lectin. About 90% of this lentil lectin-reactive glycopeptides lost this reactivity upon α-L-fucosidase digestion. The glycopeptide fractions were also prepared from three carbohydrate-containing domains. Less than 30% of the radioactivity of the glycopeptide fractions of 150-kDa and 130-kDa fragments was retained on the lentil lectin-agarose, while about 90% of that from the 50-kDa fragment was retained. These results indicate that porcine plasma fibronectin has characteristics very similar to those of human plasma fibronectin and others, but is unique in that it contains fucosylated carbohydrate chains which unevenly distribute through functional domains.     

Structure of the carbohydrate units of human amniotic fluid fibronectin

Human amniotic fluid fibronectin was found to contain three types of carbohydrates: complex-type N-glycosidic glycans, lactosaminoglycans, and O-glycosidic glycans. The structures of the complex-type glycans were established by carbohydrate and methylation analysis, Smith degradation, sequential exoglycosidase treatments, lectin chromatography, and DEAE-Sephadex chromatography. Lactosaminoglycans were analyzed by fast atom bombardment mass spectrometry, and the O-glycosidically-linked oligosaccharides by gas-liquid chromatography-mass spectrometry and high-pressure liquid chromatography. The results show that amniotic fluid fibronectin contains 2 mol of biantennary and 2-3 mol of triantennary, complex-type N-glycosidic glycans. Unlike the N-glycosidic glycans of human adult plasma fibronectin, which contain only traces of fucose and are completely sialylated, the glycans from amniotic fluid fibronectin are fucosylated and only partially sialylated. The complex-type N-glycosidic glycans present in amniotic fluid fibronectin also include a fractional amount (0.1 mol) of glycans with a polylactosaminyl structure. In addition, 4 mol of O-glycosidic oligosaccharides, which have not previously been described in fibronectins, were found in amniotic fluid fibronectin. The major oligosaccharides in this fraction have the structures Gal beta 1----3GalNAcol, NeuNAc alpha 2----3Gal beta 1----3GalNAcol and NeuNAc alpha 2----3Gal beta 1----3(NeuNAc alpha 2----6)GalNAcol. O-glycosidically linked oligosaccharides were also detected in human adult plasma fibronectin but in smaller amounts than in amniotic fluid fibronectin. These results show that amniotic fluid fibronectin differs from plasma fibronectin with regard to the number of glycans attached to the polypeptide and that the glycans present in these two fibronectins differ in structure.     

Amniotic fluid fibronectin. Characterization and synthesis by cells in culture

A glycoprotein immunologically related to plasma cold-insoluble globulin (CIG) and fetal skin fibroblast fibronectin has been purified from second-trimester human amniotic fluid. This protein (amniotic fluid fibronectin) migrated more slowly than CIG on sodium dodecyl sulfate gel electrophoresis and showed greater polydispersity which could result, at least in part, from heterogeneity in glycosylation. Cloned human amniotic fluid epithelioid and fibroblastic cells synthesized and secreted a protein with similar properties into the culture medium. Fibronectin was shown to be associated with the pericellular and extracellular matrix of cultured amniotic fluid cells by immunofluorescence, lactoperoxidase-catalyzed iodination, and labeling with ferritin-conjugated antibodies. The kinetics of secretion of the protein were consistent with its role as a matrix protein. We anticipate that amniotic fluid fibronectin will prove to be the same protein which elsewhere in the body is incorporated into connective tissues and basement membranes. Amniotic fluid could, therefore, serve as a convenient source of in vivo synthesized fibronectin for biological and structural studies.     

Biochemical evidence for the local synthesis and retention of fibronectin in the tissue of human placenta

1. Human chorionic tissues were incubated with [14C]leucine and/or [3H]glucosamine, and fibronectin synthesis was examined. 2. Radio-labeled fibronectin was detected in the tissue fraction of the incubation mixture, but not in the medium fraction, indicating that fibronectin is synthesized and retained in the tissue. 3. The glycopeptides derived from 3H-labeled fibronectin showed the lectin-binding characteristics similar to those from unlabeled placenta fibronectin, but different from those of plasma fibronectin.     

Developmental changes in the glycosylation and binding properties of human fibronectins. Characterization of the glycan structures and ligand binding of human fibronectins from adult plasma, cord blood and amniotic fluid

Fibronectins from human adult plasma, fetal plasma and from amniotic fluid obtained during early and late gestation were compared with respect to (i) their reactivity with lectins, (ii) their binding to the physiological ligands gelatin and heparin, and (iii) the role of the carbohydrate residues in the binding to these two ligands. The two fibronectin isoforms displayed distinct developmental differences in both glycosylation and binding properties: (i) Proportions of tri/tetraantennary complex glycans compared to the fraction of biantennary structures, as inferred from the reactivity with concanavalin A, were highest in amniotic fluid fibronectin from late pregnancy, lower in amniotic fluid fibronectin from early gestation, and even lower in fetal and adult plasma fibronectins. Likewise, fucose (alpha 1-6) linked to the innermost N-acetylglucosamine of the chitobiosyl core, defined by reactivity with Lens culinaris agglutinin (LCA), was present primarily in amniotic fluid fibronectin, and decreased in content during gestation from the 2nd. to the 3rd. trimenon. Both fetal and adult plasma fibronectins were only weakly reactive with LCA, indicating a low content of (alpha 1-6) linked fucose residues. After prior treatment with sialidase, both plasma and amniotic fluid fibronectins strongly reacted with erythrocyte phytohaemagglutinin (E-PHA), indicating that both fibronectin isoforms contain bisecting (beta 1-4) N-acetylglucosamine residues. Amniotic fluid fibronectins showed much greater reactivity than adult and fetal plasma fibronectins with wheat germ agglutinin; binding of this lectin to amnion fluid fibronectins was not decreased by desialylation indicating the presence of poly(N-acetyllactosamine) units. Whereas amniotic fluid fibronectins were strongly reactive with peanut agglutinin, neither adult nor fetal plasma fibronectins did bind to this lectin unless after prior desialylation. Hence, both fibronectin isoforms contain O-glycan residues that are fully sialylated in fetal and adult plasma fibronectins, but only partly sialylated in amniotic fluid fibronectins. According to these differences, glycosylation of plasma and amniotic fluid fibronectins is under developmental regulation. (ii) Amniotic fluid fibronectins had a significantly lower binding activity for both heparin and gelatin than plasma fibronectins. Moreover, amnion fibronectin from late gestation displayed a significantly lower binding to these two ligands than amnion fibronectin from early gestation. Fetal plasma fibronectins had a lower binding activity for gelatin than adult plasma fibronectin. (iii) Treatment of fibronectins with sialidase, fucosidase and removal of N-glycans with endoglycosidases H and F did not affect binding to gelatin and heparin, indicating that the interaction of plasma and amnion fibronectin with these two ligands is not influenced by their oligosaccharide moieties.     

Binding of bovine follicular fluid glycosaminoglycans to fibronectin, laminin and low-density lipoproteins

Interactions of bovine follicular fluid glycosaminoglycans (GAGs) with extracellular matrix (ECM) components fibronectin and laminin and with low-density lipoproteins (LDL) were examined using affinity chromatography. Glycosaminoglycans from small (diameter less than 5 mm) and large (diameter 11-20 mm) follicles were isolated from follicular fluid. The dermatan sulphate or heparan sulphate from small or large follicles was applied to Fn-, Lm- or LDL-Sepharose columns. Portions of each fraction of the bound or unbound GAG were then subjected to gel filtration h.p.l.c. for quantification. The binding interaction between dermatan sulphate and fibronectin was significantly greater than between heparan sulphate and fibronectin (P less than 0.05); the binding interaction between GAGs from small follicles and fibronectin was significantly greater than between GAGs from large follicles (P less than 0.05). The binding interaction between GAGs from small follicles and laminin was significantly greater than for GAGs from large follicles (P less than 0.05). Dermatan sulphate from small follicles bound to fibronectin (42%), laminin (36%) and LDL (14%) and that from large follicles bound to fibronectin (14%), laminin (23%) and LDL (14%). Heparan sulphate from small follicles bound to fibronectin (17%), laminin (15%) and that from large follicles bound to fibronectin (13%), laminin (10%) and LDL (6%). These results suggest that dermatan sulphate, but not heparan sulphate, from follicles at different stages of development exhibit a varied ability to interact with components of the ECM. Both substances bound to LDL comparably in small amounts.     

Isolation and characterization of human placenta fibronectin

Fibronectin was isolated from human placenta tissues and compared with human plasma fibronectin. Placenta and plasma fibronectins had similar amino acid compositions, immunological properties, and cell attachment-promoting activities, but differed in apparent molecular weight on sodium dodecyl sulfate-polyacrylamide gel electrophoresis, which could be accounted for at least partly by the difference in carbohydrate composition. Unlike plasma fibronectin, placenta fibronectin failed to form a precipitin line with concanavalin A in a double diffusion system. The non- or low-reactivity of placenta fibronectin with this lectin was also demonstrated by affinity chromatography with concanavalin A-agarose, in which more than 90% of the radiolabeled glycopeptides derived from placenta fibronectin was not retained on the gel. The two fibronectins also differed in the reactivity with Lens culinaris agglutinin of their glycopeptide fractions. These data indicate that placenta and plasma fibronectins are different in their carbohydrate structures and, therefore, suggest the presence of a tissue- or cell-specific mechanism for processing the carbohydrates of this glycoprotein.     

Plasma fibronectin binds glucose

Equilibrium dialysis studies demonstrated that plasma fibronectin bound D-glucose with moderate affinity. The binding of glucose by plasma fibronectin caused the dissoclation of plasma fibronectin-gelatin complexes. Glucose and gelatin did not compete for the same binding sites on plasma fibronectin. The glucose-caused dissociation of plasma fibronectin from plasma fibronectin-gelatinized horse erythrocyte complexes destroyed the potent hemagglutination activity of these complexes against trypsinized, formalinized sheep erythrocytes.     

Human placental (fetal) fibronectin: increased glycosylation and higher protease resistance than plasma fibronectin. Presence of polylactosamine glycopeptides and properties of a 44-kilodalton chymotryptic collagen-binding domain: difference from human plasma fibronectin

Human placental fibronectin was isolated from fresh term placenta by urea extraction and purified by gelatin affinity chromatography. A 44-kDa chymotryptic fragment, also purified by gelatin affinity chromatography, gave a broad, diffuse band on polyacrylamide gel electrophoresis, whereas the analogous 43-kDa fragment from human plasma fibronectin migrated as a defined, narrow band. Upon extended treatment with endo-beta-galactosidase from Escherichia freundii, the 44-kDa chymotryptic gelatin-binding fragment from placental fibronectin changed its behavior on gel electrophoresis and migrated as a narrower, more defined band. The carbohydrates on human placental fibronectin contained a large percentage of polylactosamine structures, part of which occurred on the gelatin-binding fragment, comprising almost twice as much carbohydrate as plasma fibronectin. NH2-terminal amino acid sequence analysis of the chymotryptic gelatin-binding fragments from both fibronectins showed the first 21 residues to be identical. Tryptic and chymotryptic peptide maps of the gelatin-binding fragment from placental fibronectin, however, showed differences including several protease-resistant domains not found in the analogous fragment from plasma fibronectin. Intact placental fibronectin contains 20,000 Da of carbohydrate, whereas plasma fibronectin contains 11,000 Da. Placental fibronectin is more protease-resistant than plasma fibronectin, possibly due to the additional carbohydrate. Polyclonal antibodies against either fibronectin completely cross-react with amniotic fluid fibronectin, placental fibronectin, and plasma fibronectin upon Ouchterlony immunodiffusion. Human fibronectins of putatively the same polypeptide structure are, therefore, glycosylated in a dramatically different fashion, depending on the tissue of expression. If the patterns of glycosylation comprise the only difference in the glycoprotein, this may confer the characteristic protease resistance found for each of the fibronectins.     

Binding specificity of a baby hamster kidney lectin for H type I and II chains, polylactosamine glycans, and appropriately glycosylated forms of laminin and fibronectin.

The carbohydrate binding specificity of Mr = 30,000 lectin (CBP30) from baby hamster kidney (BHK) cells has been studied by inhibition of binding of the radiolabeled lectin to asialofetuin-Sepharose using model oligosaccharides and glycopeptides. CBP30 binds type I or II Gal beta(1----3(4))GlcNAc chains but not Gal(beta 1----3)GalNAc. The inhibitory potency of straight chain polylactosamine structures or complex-type branched glycans is increased in proportion to the number of Gal(beta 1----3(4)) units present. Fucosylation or sialylation of terminal galactose residues or further substitution by (alpha 1----3)-linked galactose or N-acetylgalactosamine does not affect binding whereas substitution of the penultimate N-acetylglucosamine residue drastically reduces binding. Thus, blood group A, H type I or H type II structures, shows high affinity whereas Lex, Lea, and Leb structures bind poorly. CBP30 binds to murine Engelbreth-Holm-Swarm (EHS) tumor laminin and human amniotic fluid fibronectin but not human plasma fibronectin. Binding involves polylactosamine glycans as well as tri- and tetraantennary complex-type glycans present in EHS laminin and amniotic fluid fibronectin but absent in plasma fibronectin. Proteolytic fragments of EHS laminin (E1X/Nd, P1, E8, and E3) bind CBP30, but only fragment E8 supports attachment and spreading of BHK cells. BHK cell adhesion to EHS laminin or fragment E8 was not disturbed by CBP30-specific antibodies, but at relatively high concentrations (45 micrograms/ml) CBP30 inhibited spreading and partially attachment of cells on laminin.     

Comparative studies on amniotic fluid and plasma fibronectins.

Human fibronectin was isolated from second-trimester amniotic fluid, from amniotic fluid obtained at term and from adult plasma. The amniotic-fluid fibronectins had a slightly higher apparent molecular weight on sodium dodecyl sulphate/polyacrylamide-gel electrophoresis than the plasma fibronectin. Early- and late-amniotic-fluid fibronectin had 9.5 and 9.6% carbohydrate respectively, whereas plasma fibronectin had 5.8%. The amniotic-fluid fibronectins had similar mannose and sialic acid contents to plasma fibronectin, but greater amounts of glucosamine, galactosamine, galactose and fucose. There were no detectable differences in the amino-acid composition of amniotic-fluid and plasma fibronectins, and the patterns of peptides obtained after tryptic digestion of fibronectin from the two sources showed extensive similarities. Fibronectins from plasma and amniotic fluid were equally active in promoting cell attachment and were immunologically indistinguishable. These results show that fibronectin from amniotic fluid is more heavily glycosylated than plasma fibronectin or previously analysed fibronectins from cultured fibroblasts. The observed differences in glycosylation may be related to cell type and/or stage of development.     

Electron spin resonance spin label studies of plasma fibronectin: effect of temperature

Plasma fibronectin was chemically modified by 4-maleimido-2,2,6,6-tetramethylpiperidinooxyl (maleimide spin label). Only the free sulfhydryl groups of plasma fibronectin were modified by the label under the experimental conditions. The ESR spectrum of spin-labeled fibronectin showed that the sites of labeling were highly immobilized, suggesting that the sulfhydryl groups of the protein are in small, confined environments. The conversion of the strongly immobilized ESR spectrum into a weakly immobilized one was observed when the spin-labeled protein was heated from 30 to 60 degrees C, indicating the thermal unfolding of the protein molecules. The midpoint temperature for the thermal unfolding of plasma fibronectin is about 50 degrees C. The results suggest that plasma fibronectin is stable to about 40 degrees C and starts unfolding above this temperature. The rotational correlation time estimated from the ESR spectrum of spin-labeled fibronectin at 21 degrees C was about 2.0 X 10(-8) s. The rotational correlation time calculated from the Stokes-Einstein equation, assuming a rigid globular configuration for fibronectin with a Stokes radius of 10 nm, was about 7.8 X 10(-7) s. The differences in rotational correlation time by a factor of 39 between experimental and calculated values do not support a globular configuration for plasma fibronectin.     

Characterization of porcine plasma fibronectin and its fragmentation by porcine liver cathepsin B

Fibronectin was isolated from porcine plasma by affinity chromatography with gelatin-linked Sepharose 4B. Porcine fibronectin had a chemical composition similar to those of human and other fibronectins and reacted with antiserum raised against human fibronectin. It showed hemagglutination activity with trypsin-treated rabbit erythrocytes, though the activity was far less than that of human fibronectin. Porcine plasma fibronectin consisted of two subunit chains of about 230,000-daltons linked by disulfide bonds(s). Limited proteolysis of this protein with porcine liver cathepsin B yielded five major fragments which were investigated by affinity chromatography with gelatin- and heparin-linked Sepharose 4B. One fragment (Mr = 50,000) was bound to gelatin but not to heparin, while the remaining four were bound to heparin but not to gelatin, suggesting that plasma fibronectin takes a discrete domain structure with respect to interaction with these two macromolecules. The three larger heparin-binding fragments, Mr = 175,000, 150,000, and 130,000 were eluted with different concentrations of a mixture of NaCl and urea from the heparin-column, suggesting that they have different interactions with heparin, the 130,000-dalton fragment being the one with the strongest interaction. After reduction with 2-mercaptoethanol, the 175,000-dalton fragment was converted to the 150,000-dalton region fragment, which, together with the unchanged 150,000-dalton fragment, appeared to be equivalent in amount to the 130,000-dalton fragment. This finding suggests that the 150,000- and 130,000-dalton fragments may have originated from different subunit chains. Since the 175,000-dalton fragment was not produced by cathepsin B digestion of fibronectin which had been treated with plasmin, it was concluded that the 175,000-dalton fragment contained interchain disulfide bond(s) which had linked the native subunit chains. These results suggest that porcine plasma fibronectin has non-identical subunit chains composed of domains which differ in interaction with heparin and in susceptibility to cathepsin B.     

The interaction of plasma fibronectin with fibroblastic cells in suspension

We have examined the interaction of soluble plasma [3H]fibronectin with fibroblastic cells in suspension. Fibronectin labeled by reductive methylation binds to baby hamster kidney cells in serum-free medium in a time-dependent manner at 4, 22, and 37 degrees C, with half-maximal binding occurring in 12-15 min at 22 degrees C. The binding is saturable and reversible. At least 90% of the cell-associated fibronectin is external to the plasma membrane, as judged by trypsin susceptibility of the bound radioactivity. Scatchard analysis of the concentration dependence of binding indicates the presence of a single class of binding sites, even at low input concentrations of fibronectin. There are approximately 5 +/- 1 X 10(5) sites/cell with an apparent dissociation constant of 8.0 +/- 0.5 X 10(-7) M; thus, the binding of soluble fibronectin to these cells is of moderate affinity. This putative fibroblast fibronectin receptor is resistant to trypsin in the presence of physiological concentrations of divalent cations but is susceptible to trypsin in the presence of 5 mM EDTA. Binding of 0.1 mg/ml [3H]fibronectin is 60-80% inhibited by 8 mg/ml unlabeled fibronectin and 95% inhibited by 1 mg/ml purified 75-kDa fibronectin cell-binding domain, but is unaffected by 1 mg/ml 44-kDa collagen-binding domain or 5 mg/ml ovalbumin. The binding parameters determined in this study further define the fibroblast cell-surface fibronectin receptor.     

Binding of fibronectin by the acute phase reactant C-reactive protein

Following tissue injury, the concentration of C-reactive protein (CRP) is known to increase in plasma rapidly, while that of fibronectin often decreases. We now report that CRP immobilized onto polystyrene surfaces binds soluble plasma fibronectin (Kd = 1.5 X 10(-8) M). The binding of fibronectin by CRP was relatively sensitive to ionic conditions, being maximal at physiological NaCl concentrations. A decrease of pH from neutral to 5-6 greatly enhanced the binding of fibronectin by CRP. Ca2+ ions at greater than 1 mM inhibited binding. No binding was observed between fibronectin and CRP in soluble phase. CRP was found also to bind fibrinogen, which competed with fibronectin for CRP-binding sites. This was shown to explain why fibronectin was effectively bound from serum but not from plasma by immobilized CRP. The amount of CRP immobilized was critical in binding fibronectin; a too dense molecular layer of CRP inhibited the binding, as did the postsaturation of free surfaces with albumin, which itself was not bound by CRP. Soluble fibronectin agglutinated CRP-coated latex particles. Most or all of the CRP-binding activity in the fibronectin molecule was localized to the 120-140-kilodalton fragment, which also contains cell-binding and heparin-binding domains of fibronectin. The results provide a link between acute phase response and tissue repair.     

Comparison of the structures of human fibronectin and plasma cold-insoluble globulin

Human amniotic fluid fibronectin and plasma fibronectin (cold-incoluble globulin) are indistinguishable both immunologically and by amino acid composition. Cyanogen bromide and tryptic peptides also suggest substantial structural homology. However, carbohydrate analysis has demonstrated additional saccharides in fibronectin and an overall increase in carbohydrate content relative to coldinsoluble globulin. Furthermore, limited proteolytic cleavage of the two proteins indicates differences in primary structure or in conformation. Using affinity-purified antibodies to cold-insoluble globulin, a glucosamine-labeled pronaseresistant component, probably proteoglycan, was found to coprecipitate with fibronectin, suggesting an association between these two macromolecules in the connective tissue matrix.     

Heat-induced fragmentation of human plasma fibronectin

Human plasma fibronectin was found to undergo fragmentation during heat-denaturation, leading to artifacts in SDS-polyacrylamide gel electrophoretic analyses. Electrophoretic patterns of heated samples showed a progressive decrease in intact fibronectin chains (225 kDa) which coincided with the appearance of increasing amounts of numerous smaller components having molecular weights ranging from 10 000 to 200 000. The fragmentation was temperature-dependent, being undetectable after 2 h at 60 degrees C, but detectable after 30 min at 70 degrees C or as little as 2 min at 100 degrees C. After 2 h at 100 degrees C, the intact monomer was no longer visible. Neither mercaptoethanol nor SDS was required for fragmentation. Sterile filtration or pretreatment with inhibitors of proteolytic enzymes had no effect. Treatment with amines did not diminish the degradation, indicating that the process differs from heat-fragmentation of alpha 2-macroglobulin and complement proteins, which occurs at a reactive internal thiolester bond. Fibronectin fragmentation was highly pH-dependent, being markedly accelerated under acidic conditions, suggesting that autolytic cleavage of the peptide chain at acid-labile aspartyl bonds was responsible for this phenomenon.     

The secondary structure of human plasma fibronectin: conformational changes induced by acidic pH and elevated temperatures; a circular dichroic study

The secondary structure of native human plasma fibronectin, based on circular dichroic spectra, has been estimated to contain 79% beta sheet and 21% beta turn structures (Osterlund, E., Eronen, I., Osterlund, K. and Vuento, M. (1985) Biochemistry 24, 2661-2667). In this work changes in the secondary structure of the protein molecule are followed as a function of different temperatures and pH values by using circular dichroic spectroscopy in far- and near-ultraviolet regions. Conformational changes are reversible when raising the temperature quickly to 55 degrees C, and then cooling slowly to 20 degrees C. A few percent of alpha-helix is apparent, when the temperature is raised to 58.5 degrees C, but only about 9% random coil is formed, when the temperature is raised up to 70 degrees C. The largest conformational change is taking place, when fibronectin samples are heated from 57 to 58.5 degrees C. According to this study more than 90% of the secondary structure of the fibronectin molecule is preserved throughout the whole temperature range studied from 20 to 70 degrees C, and this is a fact even at pH as low as 3.0, when samples are fresh and not denatured by preparative procedures.     

Plasma fibronectin: three steps to purification and stability.

Large amounts of soluble fibronectin were easily purified from cryoprecipitated or fresh citrated human blood plasma by a three-step combination of gelatin and heparin-cellufine affinity chromatography. The elution conditions were optimized to obtain a homogeneous fraction on SDS-PAGE and Western blot under reducing condition. No proteolytic activities were detected by zymography at acidic or neutral pH. Furthermore, the fibronectin preparation was stable over time up to 456 h at 37 degrees C in the presence of calcium, zinc, or mercury. This preparation of very stable fibronectin, called highly purified fibronectin (hpFN), gave a yield of 7.00 +/- 0.77 mg of fibronectin per gram of cryoprecipitated plasma and 0.16 mg of fibronectin per milliliter of fresh citrated, giving a yield of 32 to 53% (from presumed fibronectin concentration). This preparation may be useful for cellular tests and interaction analysis.