Assembly and activation of the intrinsic fibrinolytic pathway on the surface of human endothelial cells in culture.

Factor XII has long been implicated in the intrinsic pathway of fibrinolysis, but the mechanism by which it triggers plasminogen activation and targets fibrinolysis has not been established. In the present study, the assembly and function of activated Factor XII (F.XIIa), prourokinase (pro-u-PA), high molecular weight kininogen (H-kininogen), and prekallikrein on human umbilical vein endothelial cells (HUVEC) was investigated. 125I-prekallikrein was shown to bind to HUVEC via receptor-bound H-kininogen in the presence of 50 microM ZnCl2. After the addition of F.XIIa, 78% of the 125I-prekallikrein initially bound to HUVEC was converted to 125I-kallikrein. However, only 6% of the HUVEC-bound 125I-pro-u-PA was thereby activated. This discrepancy was shown to be related to rapid dissociation (> 50% within 15 min) of prekallikrein/kallikrein, but not pro-u-PA, from HUVEC. Increasing the level of cell-bound kallikrein increased the portion of cell-bound pro-u-PA activated, indicating that their co-localization was important for this pathway. Finally, F.XIIa was shown to trigger plasminogen activation on HUVEC via this pathway. This assembly of reactants on the endothelium suggests a mechanism whereby local fibrinolysis may be triggered by blood coagulation.

The influence of glycosylation on the catalytic and fibrinolytic properties of pro-urokinase.

We previously found that human pro-UK expressed in Escherichia coli is more active in fibrinolysis than recombinant human pro-UK obtained from mammalian cell culture media. To determine whether this difference is related to the lack of glycosylation of the E. coli product, we compared the activity of E. coli-derived pro-UK [(-)pro-UK] with that of a glycosylated pro-UK [(+)pro-UK] and of a mutant of pro-UK missing the glycosylation site at Asn-302 [(-)(302)pro-UK]. The latter two pro-UKs were obtained by expression of the human gene in a mammalian cell. The nonglycosylated pro-UKs were activated by plasmin more efficiently (approximately 2-fold) and were more active in clot lysis (1.5-fold) than the (+)pro-UK. Similarly, the nonglycosylated two-chain derivatives (UKs) were more active against plasminogen and were more rapidly inactivated by plasma inhibitors than the (+)UK. These findings indicate that glycosylation at Asn-302 influences the activity of pro-UK/UK and could be the major factor responsible for the enhanced activity of E. coli-derived pro-UK.

The effect of the carboxy-terminal lysine of urokinase on the catalysis of plasminogen activation.

When single-chain pro-UK is activated by plasmin or kallikrein, the Lys158-Ile159 bond is cleaved, leaving a C-terminal lysine on the A-chain (Lys-UK). Two-chain, high molecular weight urokinase (UK) purified from urine, however, has been shown to contain a phenylalanine residue as the C-terminal of the A-chain (Phe-UK). Since C-terminal lysine residues have a strong binding affinity for plasminogen that may promote its activation, we undertook kinetic studies comparing plasminogen activation by Lys- and Phe-UK. A two-stage method was employed in order to minimize factors known to interfere with plasminogen activation and plasmin determination. The Lys-UK was prepared by plasmin activation of pro-UK purified from human fetal kidney cell culture medium. The Phe-UK was prepared by carboxypeptidase B (CpB) treatment of Lys-UK. Removal of the C-terminal lysine of Lys-UK by CpB produced small but significant increases in the Michaelis constants for the activation of both Glu- and Lys-plasminogen. The apparent Michaelis constants for Glu-plasminogen activation by Lys- and Phe-UK were 3.7 microM +/- .36 microM and 5.9 microM +/- .70 microM, respectively and the Michaelis constants for Lys-plasminogen activation by Lys- and Phe-UK were 5.4 microM +/- .72 microM and 15.2 microM +/- 1.4 microM, respectively. The catalytic efficiency (kcat/Km) of Lys-UK was approximately 2-fold greater than that of Phe-UK for the activation of either Glu- or Lys-plasminogen. When the fibrinolytic activities of Lys- and Phe-UK were compared in a plasma milieu no significant differences were detected. In conclusion, the findings indicate that the C terminal lysine on the A-chain of UK significantly promotes the catalysis of plasminogen in a purified system. However, the higher catalytic efficiency of Lys-UK was not found to induce significant acceleration of clot lysis at pharmacological concentrations in plasma.

Chylomicron remnant-vitamin A metabolism by the human hepatoma cell line HepG2.

The binding and metabolism of [3H]vitamin A-containing chylomicron (CM) remnants by the human hepatoma cell line HepG2 were studied. Mesenteric lymph chylomicrons were collected from [3H]retinol-fed rats and incubated with lipoprotein lipase to obtain CM remnants. At 4 degrees C, specific CM remnant binding was inhibited by an excess of unlabeled CM remnants. Specific binding predominated at low concentrations and approached saturation while total binding continued to increase over an extensive concentration range (0.45-32 microgram triglyceride/ml). CM remnant uptake at 37 degrees C was greater than that of CM and at least 70 times more efficient than the pinocytosis of sucrose. CM remnant binding increased with the extent of lipolysis. Addition of human apolipoprotein E enhanced both CM remnant and CM binding. After internalization, HepG2 cells hydrolyzed CM remnant-[3H]retinyl esters, and radiolabeled metabolites accumulated. As a function of the concentration of [3H]retinoid initially bound to cells, retinol and retinyl esters accumulated as the major cell-associated metabolites. In contrast, retinol was the major metabolite in the medium only at low retinoid concentrations; other more polar metabolites accumulated at higher concentrations (greater than 110 pmol retinoid/mg cell protein). The accumulation in the medium of labeled metabolites derived from CM remnant-retinoid was reduced when cells were preincubated in unlabeled retinol-supplemented media. The specific activity of retinol in the medium indicated that CM remnant-vitamin A had mixed with the cellular store prior to its secretion as retinol. These results indicate that HepG2 cells internalize CM remnants in part by specific binding sites, and that the metabolism of CM remnant-retinoids by the HepG2 cell involves retinyl ester hydrolysis and the secretion of retinol and other more polar metabolites. These processes were regulated in part by the concentration of retinoid delivered by the CM remnant and by the initial retinoid content of the cell.

Production and secretion of retinol-binding protein by a human hepatoma cell line, HepG2.

Retinol-binding protein (RBP) that is synthesized and secreted by the human hepatoma cell HepG2 has been measured using a sensitive radioimmunoassay in which RBP in media and hepatoma cell sonicates reacts identically to human serum RBP. RBP was synthesized and secreted when cells were grown in retinol-depleted as well as retinol-containing media. However, immunoreactive transthyretin (prealbumin) could not be detected in concentrated HepG2 medium. RBP secretion and accumulation per mg of cell protein could be modulated by the concentration of fetal calf serum in the growth medium: secreted RBP equaled 782 +/- 123 ng/mg of cell protein per 8 hr after preincubation with 10% fetal calf serum versus 555 +/- 86 ng/mg per 8 hr in the absence of serum, whereas RBP in cell sonicates decreased only slightly. When HepG2 cells were cultured for two or more passages in medium containing fetal calf serum depleted of retinol by ultraviolet irradiation, the amounts of RBP in the cells and released to the medium were both significantly increased. When vitamin A (90% as retinyl esters) in the form of chylomicron remnants was presented to cells, there was a significant, dose-dependent redistribution of RBP from cells to medium, both in cells grown in normal fetal calf serum and in retinol-depleted serum. These data indicate that the secretion of RBP by HepG2 can occur constitutively in the absence of retinol, but that secretion can be enhanced and regulated by retinol delivered by the chylomicron remnant.

Thrombin stimulation of platelets induces plasminogen activation mediated by endogenous urokinase-type plasminogen activator.

Gene knockout mice studies indicate that urokinase-type plasminogen activator (u-PA) is importantly involved in fibrinolysis, but its physiologic mechanism of action remains poorly understood. We postulated that platelets may be involved in this mechanism, as they carry a novel receptor for u-PA and a portion of the single-chain u-PA (scu-PA) intrinsic to blood is tightly associated with platelets. Therefore, plasminogen activation by platelet-associated u-PA was studied. When washed platelets were incubated with plasminogen, no plasmin was generated as detected by plasmin synthetic substrate (S2403) hydrolysis; however, after the addition of thrombin, but not other agonists, platelet-dependent plasminogen activation occurred. Plasminogen activation was surface-related, being inhibited by blocking platelet fibrinogen receptors or by preventing plasminogen binding to the thrombin-activated platelet surface. U-PA was identified as the only plasminogen activator responsible and enrichment of platelets with exogenous scu-PA significantly augmented plasminogen activation. These findings appeared paradoxical because thrombin inactivates scu-PA. Indeed, zymograms showed inactivation of scu-PA during the first hour of incubation with even the lowest dose of thrombin used (1 u/mL). However, this was followed by a thrombin dose-dependent (1 to 10 u/mL) partial return of u-PA activity. Reactivation of u-PA was not due to the direct action of thrombin, but required platelets and was found to be related to a platelet lysosomal thiol protease, consistent with cathepsin C. In conclusion, a new pathway of plasminogen activation by platelet-associated endogenous or exogenous scu-PA was demonstrated, which is specifically triggered by thrombin activation of platelets. These findings may help explain u-PA-mediated physiological fibrinolysis and have implications for therapeutic thrombolysis with scu-PA.

Effect of dietary cholesterol and alloxan-diabetes on tissue cholesterol and apolipoprotein E mRNA levels in the rabbit.

Alloxan-diabetic rabbits develop hypercholesterolemia and hypertriglyceridemia in response to cholesterol feeding. To determine whether alterations in apolipoprotein composition of plasma lipoproteins were due to changes in apolipoprotein gene expression, we measured the steady state apoE mRNA levels in several tissues from both control and diabetic rabbits. Control rabbits were fed either chow or chow plus 1.5% cholesterol (chow-fed or cholesterol-fed groups) and diabetic rabbits were fed either chow or chow plus 0.5% cholesterol for dietary periods of 5, 21, and 42 days. The tissues examined were liver, small intestine, brain, adrenals, and aorta. ApoE mRNA levels were measured by Northern and dot blot analysis with a human apoE cDNA probe. In control rabbits fed either chow or cholesterol diets for up to 42 days, the steady state apoE mRNA levels remained relatively constant in all of the tissues examined. In contrast, in alloxan-diabetic rabbits fed a 0.5% cholesterol diet, apoE mRNA was reduced in liver, brain, and adrenals (46 +/- 19%, 56 +/- 5%, and 39 +/- 18% of chow-fed control, respectively), but showed little change in the aorta (91 +/- 22% of chow-fed control). Despite a similar increase in plasma cholesterol, the cholesterol content of the liver and adrenals of cholesterol-fed diabetic rabbits were 20% and 50%, respectively, of that of the liver and adrenals in cholesterol-fed control rabbits. The result that apoE mRNA levels and tissue cholesterol content are altered in the diabetic cholesterol-fed rabbit suggests that insulin deficiency in the rabbit may influence apoE gene expression and tissue cholesterol homeostasis.

Apolipoprotein gene expression in the rabbit: abundance, size, and distribution of apolipoprotein mRNA species in different tissues.

We have used human apolipoprotein cDNAs as hybridization probes to study the relative abundance and distribution of apolipoprotein mRNAs in rabbit tissues by RNA blotting analysis. The tissues surveyed included liver, intestine, lung, pancreas, spleen, stomach, skeletal muscle, testis, heart, kidney, adrenal, aorta, and brain. We found that liver is the sole or major site of synthesis of apoA-II, apoA-IV, apoB, apoC-I, apoC-II, apoC-III, and apoE, and the intestine is a major site of synthesis of apoA-I, apoA-IV, and apoB. Minor sites of apolipoprotein mRNA synthesis were as follows: apoA-I, liver and skeletal muscle; apoA-IV, spleen and lung; apoB, kidney; apoC-II and apoC-III, intestine. ApoE mRNA was detected in all tissues surveyed with the exception of skeletal muscle. Sites with moderate apoE mRNA (10% of the liver value) were lung, brain, spleen, stomach, and testis. All rabbit mRNAs had forms with sizes comparable to their human counterparts. In addition, hybridization of hepatic and intestinal RNA with human apoA-IV and apoB probes produced a second hybridization band of approximately 2.4 and 8 kb, respectively. Similarly, hybridization of rabbit intestinal RNA with human apoC-II produced a hybridization band of 1.8 kb. The 8 kb apoB mRNA form may correspond to the apoB-48 mRNA, whereas the apoA-IV- and apoC-II-related mRNA species have not been described previously. This study provides a comprehensive survey of the sites of apolipoprotein gene expression and shows numerous differences in both the abundance and the tissue distribution of several apolipoprotein mRNAs between rabbit and human tissues. These findings and the observation of potentially new apolipoprotein mRNA species are important for our understanding of the cis and trans acting factors that confer tissue specificity as well as factors that regulate the expression of apolipoprotein genes in different mammalian species.