We describe a specific fibrinogen-hepatocyte interaction. Rabbit 125I-labeled fibrinogen (125I-FGN) was incubated at 4 degrees C with suspensions of rabbit hepatocytes (approximately 1 X 10(6) cells/ml). Bound ligand was separated from free by centrifugation of cells through oil and quantitated by gamma-scintillation counting. Specific binding, determined by subtraction of nonspecific binding in the presence of 8 mM EDTA from total binding in the presence of 2 mM CaCl2, required 3 h to plateau and represented approximately 70% of total binding. Specific binding was calcium-dependent and was negligible in buffer containing 2 mM MgCl2. Half-maximal saturation occurred at approximately 30 nM 125I-FGN with approximately 480,000 molecules/cell at saturation. Dilution experiments revealed comparable affinities for labeled and unlabeled fibrinogen. Total binding was irreversible as determined by addition of excess unlabeled fibrinogen or EDTA. Specific binding of 25 nM 125I-FGN was inhibited, in a concentration-dependent fashion, by unlabeled fibrinogen or fibrinogen fragment D95 (Mr = 95,000), but not by fibrinogen fragment E or Arg-Gly-Asp-containing peptides. Unlabeled fibrinogen (3.1 microM) completely abolished specific binding, whereas greater than 80% inhibition was achieved with 10 microM fragment D95. Sodium dodecyl sulfate polyacrylamide gel electrophoresis and autoradiography of 125I-FGN bound in the presence of calcium demonstrated disappearance of A alpha chains with formation of products of Mr greater than 200,000; EDTA or unlabeled fibrinogen prevented fibrinogen processing. These data describe a unique fibrinogen-hepatocyte interaction which differs considerably from the platelet-fibrinogen interaction, especially with regard to the processing of the fibrinogen molecule.

Human hepatoma cell (HepG2) or rabbit hepatocyte monolayers were incubated with [35S]methionine in presence or absence of tunicamycin, a potent inhibitor of asparagine-linked glycosylation. The 35S-labeled nonglycosylated and control fibrinogens purified from the media were used to evaluate the influence of the oligosaccharide on the catabolic properties of this glycoprotein. Plasmin, pronase, cathepsin D or cathepsin B each degraded the nonglycosylated and control fibrinogens similarly, as evidenced by the release of trichloroacetic acid-soluble radioactivity and by SDS-polyacrylamide gel electrophoresis and autoradiography of plasmic digests. Nonglycosylated and control fibrin clots also showed no differences in susceptibility to plasmic digestion. The two forms of fibrinogen demonstrated the same plasma half-life in rabbits. These data indicate that the oligosaccharide does not influence the proteolytic stability or the in vivo plasma survival of fibrinogen, and suggest that other biochemical determinants may influence the catabolic properties of this molecule.

We investigated the effects of the branched-chain amino acids--valine, leucine and isoleucine--or their keto analogs, the branched-chain keto acids--alpha-ketoisovaleric acid, alpha-ketoisocaproic acid and alpha-keto-beta-methylvaleric acid--on protein synthesis and secretion by monolayers of rabbit hepatocytes incubated with [35S] methionine in pulse-chase and steady-state experiments. The branched-chain amino acids (2.0 mM or 1.0 mM), in the presence or absence of insulin (2 X 10(-4) IU per dish) and in both types of experiments, reduced the trichloroacetic acid-precipitable 35S-protein secreted into the medium. The branched-chain keto acids (2.0 mM or 1.0 mM) had a stimulatory effect on secreted trichloroacetic acid-precipitable 35S-protein which was observed only by the pulse-chase technique in the presence of insulin. Immunoaffinity chromatography of medium demonstrated a slight inhibition by branched-chain amino acids and a slight stimulation by branched-chain keto acids on secretion of 35S-albumin and no effect of either treatment on secretion of 35S-fibrinogen. ELISA analysis of total (i.e., 35S-labeled and unlabeled) secreted albumin revealed an inhibitory effect of the branched-chain amino acids in both pulse-chase and steady-state experiments, and a small stimulatory effect, in steady-state experiments, of the branched-chain keto acids; both effects were insulin-dependent. Total secreted fibrinogen, under steady-state conditions, was increased by the branched-chain keto acids in the presence of insulin, while transferrin production was unaffected by any treatment.(ABSTRACT TRUNCATED AT 250 WORDS)