Biochemical and immunochemical comparison of fibronectin and polynectin from porcine plasma

[Human plasma fibronectin. Comparison of methods for preparation of a concentrate for therapeutic use from different sources]

Three methods, successive precipitations, affinity chromatography on immobilized gelatin and immunoaffinity chromatography with monoclonal anti-fibronectin antibodies were optimized and compared in order to be used for large scale preparation of human plasma fibronectin (Fn). The functional properties of the various Fn preparations were investigated by means of two assays: quantitation of the gelatin-binding activity by ELISA and quantitation of the Fn-mediated attachment of fibroblasts on plastic. Functional alterations of the purified Fn were observed when it was isolated by successive precipitations. Both chromatographic methods provide a rapid and convenient way for isolation of pure and functional Fn. Mass production of monoclonal antibodies is too expensive and legislative requirements for the therapeutic use of monoclonal antibodies are limiting factors for the choice of immunopurification as large scale isolation procedure. Plasma Fn can be isolated from different sources: fresh frozen plasma, cryoprecipitate supernatant or by-products from factor VIII preparation. When gelatin-Sepharose chromatography is performed under optimized conditions, fibronectins isolated from these sources show similar properties. Large scale purifications of Fn from a by-product of factor VIII preparation were performed either by gelatin affinity chromatography or by successive precipitations. These two purification methods can be easily scaled-up since the data obtained closely correlate with analytical results. The chromatographic method supplies a higher purified (98 vs 75%) and functional (95 vs 50%) material when compared with successive precipitations. Yield is also higher (50 vs 26%). The starting material undergoes viral inactivation and the affinity purified Fn, sterile, atoxic, apyrogen, which can be freeze-dried without additives fulfils all requirements for an injectable product.

Isolation and characterization of chondroitin sulfate proteoglycans from porcine thoracic aorta

A chondroitin sulfate proteoglycan fraction was prepared from the 3 M MgCl2 extract of porcine aortas by DEAE-cellulose chromatography, followed by gel filtration through Sepharose CL-4B. Affinity chromatography of the fraction with antithrombin III-agarose yielded two chondroitin sulfate proteoglycans of a non-binding (proteoglycan IA) and binding (proteoglycan IB) nature. Proteoglycans IA and IB were different from each other in molecular size, in proportion of the protein relative to the polysaccharide portion, and in size of the chondroitin sulfate chain. They were also distinguished immunochemically. These data indicate that the intima-media of the aorta contains at least two distinct species of chondroitin sulfate proteoglycan.

An enzyme-linked immunoassay for direct measurement of the gelatin-binding capacity of human plasma fibronectin

A new solid-phase enzyme immunoassay measuring the gelatin-binding capacity of plasma fibronectin has been developed. This assay is based on the direct and high-affinity interaction between fibronectin and gelatin coated to polyvinyl chloride plates. The amount of fibronectin bound to gelatin is then measured by sequential incubation with a specific rabbit anti-human fibronectin antiserum, with horseradish peroxidase-conjugated goat anti-rabbit IgG antibodies and with substrate. The final degradation of the substrate is read at 492-650 nm in an ELISA processor. The assay allows the accurate detection of fibronectin concentrations ranging from 1 to 20 micrograms/ml, is inhibited by the addition of gelatin to plasma, is highly reproducible (interplate CV less than 10%), requires 100 microliter of plasma only and has been fully automated. Significant linear correlations were noted between total antigenic fibronectin (measured by laser nephelometry) and fibronectin gelatin-binding capacity in plasma from 310 blood donors. Both parameters were higher in men than in women and significantly increased according to age. Dissociation between immunoreactive fibronectin and fibronectin gelatin-binding capacity was observed in two polytraumatized patients. This enzyme immunoassay therefore provides a new method to investigate functional alterations of the gelatin-binding domain of fibronectin in various pathological conditions.

A weaker gelatin-binding affinity and increased glycosylation of amniotic fluid fibronectin than plasma fibronectin

Human amniotic fluid fibronectin had different carbohydrate moieties from plasma fibronectin. Nearly 90% of glycopeptides released from amniotic fluid fibronectin was not bound by concanavalin A-Sepharose, whereas 75% of glycopeptides from plasma fibronectin was bound. Amniotic fluid fibronectin showed a significantly lower gelatin-binding affinity than plasma fibronectin at 25 degrees C. When the incubation temperature was lowered to 4 degrees C, no significant difference in this activity was found. Cell-attachment promoting activity of the two fibronectins was not significantly different.

Characterization of fibronectin from fetal human plasma in comparison with adult plasma fibronectin

Fibronectin was purified from fetal human plasma and characterized in comparison with fibronectin from adult human plasma. Fetal plasma fibronectin had an amino-acid composition, immunological properties, and cell-attachment-promoting activity similar to those of adult plasma fibronectin. However, fetal plasma fibronectin was shown to have a distinct carbohydrate composition which is characterized by the presence of fucose. To ascertain the differences in the carbohydrate moiety, 14C-labeled glycopeptides were prepared and sequentially analyzed with columns of immobilized concanavalin A and lentil lectin. Glycopeptides from fetal plasma fibronectin contained a population of glycopeptides which bound to both lectin gels. Almost all of the glycopeptides in this population lost their ability to bind to lentil lectin upon fucosidase digestion, indicating that fetal plasma fibronectin possesses a substantial amount of fucosylated biantennary glycans. In contrast, glycopeptides from adult plasma fibronectin practically lacked such glycopeptides. Another difference observed was that fetal plasma fibronectin had a larger amount of concanavalin A-unbound glycopeptides than did adult plasma fibronectin. These results indicate the presence of age-related variation in glycosylation of human plasma fibronectin.