Binding of factors IX and IXa to cultured vascular endothelial cells.

Factor IX and its activated form IXa have been found to bind to confluent cultured bovine aortic and human umbilical vein endothelial cells. Binding of bovine factors IX and IXa to the bovine endothelial cells was saturable and specific and reached a plateau in 75 min at 4 degrees C and 30 min at 37 degrees C. Binding was half-maximal at a total factor IX or IXa concentration of 2.3 +/- 0.2 nM. At 4 degrees C, a maximum of 42 fmol of tritiated factor IX or IXa bound to 10(6) cells (an average of 20,000 molecules per cell). The binding of tritiated factor IX or IXa was inhibited by excess unlabeled factor IX or IXa but not by factor X, prothrombin, or thrombin. Competition studies indicated that factors IX and IXa interacted with the same site. Binding was reversible, with 50% of the specifically bound factor IX or IXa eluted in 40 min by a 400-fold excess of unlabeled protein. Specific binding required Ca2+ with half-maximal binding at 1.2 mM CaCl2. Factor IXa bound to the cells was tested for procoagulant activity in a clotting assay with factor IX-deficient plasma, cephalin, and CaCl2. Cell-bound factor IXa was at least 3-fold more active than was factor IXa in solution. The retention of procoagulant activity by cell surface-bound factor IXa provides a mechanism for the localization of clot-promoting activity.     

Detection of an unusually stable fibrinolytic inhibitor produced by bovine endothelial cells.

Fibrin/agar films were prepared and used to detect plasminogen activators produced by cultured bovine aortic endothelial cells (fibrin autography). One preparation of fibrin underwent spontaneous lysis upon incubation at 37 degrees C. This lysis was prevented by antibodies to tissue-type plasminogen activator but not by antibodies to urokinase. Conditioned medium from the confluent endothelial cells was fractionated by polyacrylamide gel electrophoresis in the presence of NaDodSO4. The gels were analyzed on indicator films prepared with the spontaneously lysing fibrin (reverse fibrin autography). Unexpectedly, as the opaque fibrin film cleared, a distinct lysis-resistant zone appeared in the indicator gel at a region corresponding to Mr 55,000. Experiments were devised to determine whether the lysis-resistant zone in the indicator film reflected the presence of a cellular inhibitor in the polyacrylamide gel. The corresponding region was excised from a polyacrylamide gel, extracted with buffer, and tested directly for antifibrinolytic activity by the 125I-labeled fibrin plate method. Urokinase-mediated fibrinolytic activity was inhibited by the gel extract in a dose-dependent manner indicating the presence of such an inhibitor. Inhibitor activity was detected in Triton X-100 extracts of washed monolayers and in conditioned medium, where it accumulated with time. The endothelial cell inhibitor not only survived exposure to NaDodSO4 but also was active after incubation at pH 12 or treatment with 5% (vol/vol) 2-mercaptoethanol, 6 M urea, 4 M guanidine hydrochloride, or 1 M acetic acid. Considerable activity also remained after heating at 100 degrees C for 30 min. These results indicate that cultured bovine aortic endothelial cells synthesize and secrete a previously undetected, unusually stable fibrinolytic inhibitor of Mr 55,000. Reverse fibrin autography offers a convenient approach for studying such molecules.     

Inhibition of in vitro angiogenesis by platelet factor-4-derived peptides and mechanism of action.

In this study, we examined in detail the interaction of platelet factor-4 (PF-4) with fibroblast growth factor-2 (FGF-2) and vascular endothelial growth factor (VEGF) and the effect of PF-4-derived synthetic peptides. We show that a peptide between amino acids 47 and 70 that contains the heparin-binding lysine-rich site inhibits FGF-2 or VEGF function. This is based on the following observations: PF-4 peptide 47-70 inhibited FGF-2 or VEGF binding to endothelial cells; it inhibited FGF-2 or VEGF binding to FGFRs or VEGFRs in heparan sulfate-deficient CHO cells transfected with FGFR1 (CHOFGFR1) or VEGFR2 (CHOmVEGFR2) cDNA; it blocked proliferation or tube formation in three-dimensional angiogenesis assays; and, finally, it competed with the direct association of (125)I-PF-4 with FGF-2 or VEGF, respectively, and inhibited heparin-induced FGF-2 dimerization. A shorter C-terminal peptide (peptide 58-70), which still contained the heparin-binding lysin-rich site, had no effect. Peptide 17-58, which is located in the central part of the molecule, although it does not inhibit FGF-2 or VEGF binding or biologic activity in endothelial cells, inhibited heparin-dependent binding of (125)I-FGF-2 or (125)I-VEGF to CHOmFGFR1 or CHOmVEGFR2 cells, respectively. Shorter peptides (peptides 34-58 and 47-58) did not show any of these effects.     

Receptor distribution and the endothelial uptake of transcobalamin II in liver cell suspensions

To determine the nature of binding of transcobalamin II (TC-II) to liver cells, we covalently coupled purified holo-TC-II to submicron latex minibeads using glutaraldehyde. Incubation of the probe with liver cell suspensions at 4 degrees C led to its binding by endothelial cells but not by hepatocytes or Kupffer cells, as visualized by scanning electron microscopy. At 37 degrees C, the probe was internalized by the endothelium through a system of coated pits and vesicles as shown by transmission electron microscopy. Inhibition studies by pre-incubation with excess native TC-II demonstrated the specificity of binding. Fractionation of these cell suspensions on metrizamide gradients yielded large cell (hepatocyte-rich) and small cell (endothelium-rich) fractions. The binding of the minibead probe occurred again exclusively on endothelial cells in the small cell fraction. 125I-labeled holo-TC-II also bound to the small cell but not to the large cell fraction. Binding was saturable (Ka, 0.225 X 10(9) mol/L-1) and receptor number was calculated to be 1.33 X 10(3) per cell. Time-dependent incubation of 125I-labeled TC-II with the endothelium-rich fraction led to its uptake, reaching a steady-state plateau at 4 degrees C. At 37 degrees C, however, the initial uptake was followed by gradual release of the label into the medium. We conclude that in the liver, holo-TC-II binds initially to endothelium, where it is internalized and is subsequently released probably to the interstitial space. Thus, the endothelium may play a fundamental role in the regulation of the uptake of TC-II by the liver.     

Processing and release of insulin and insulin-like growth factor I by macro- and microvascular endothelial cells

Insulin and insulin-like growth factor I (IGF-I) processing by macro- and microvascular endothelial cells was investigated. Specific binding of insulin and IGF-I on the capillary endothelial cells derived from rat fat pads was 4 +/- 0.5% (+/- SE) and 4.3 +/- 0.3%/mg protein, respectively, in contrast to bovine aortic endothelial cells, which bound 9.3 +/- 0.3% IGF-I/mg protein. Both binding and processing of insulin and IGF-I were time and temperature dependent in macro- and microvascular endothelial cells. After 30 min at 37 C, between 40-50% of the bound IGF-I and insulin were internalized in both capillary and aortic endothelial cells, whereas 20-25% insulin and 15-20% IGF-I internalization were observed at 15 C. Less than 20% internalization was observed for both insulin and IGF-I at 4 C. Cellular inhibitors of hormone processing, such as chloroquine and monensin, enhanced cell-associated insulin at 37 C on the bovine aortic endothelial cells from 4.7% to 10.4 +/- 1% and 9.9 +/- 2% mg protein, respectively, at 60 min. Similarly, chloroquine and monensin increased the amount of [125I]IGF-I associated with aortic endothelial cells from 4.3 +/- 0.2% to 5.5 +/- 0.3% and 6.2 +/- 0.7%/mg protein, respectively. Chloroquine and monensin increased [125I]insulin associated with rat capillary endothelial cells from a control of 2.9 +/- 0.1% to 4.0 +/- 0.2% and 3.8% +/- 0.37%, respectively. No effect of chloroquine and monensin was observed on [125I]IGF-I binding to rat capillary endothelial cells. Leupeptin, a lysosomal protease inhibitor, did not affect insulin or IGF-I binding in either cell type. The internalized insulin and IGF-I were both rapidly released, with 70-80% of both hormones being detected in the medium by 120 min. The released hormones were mostly intact (greater than 80-90%), as assessed by trichloroacetic acid precipitability, gel filtration, and immunoprecipitation. Both insulin and IGF-I induced corresponding down-regulation of their receptors, as shown by a 66 +/- 7% decrease in insulin binding in the capillary endothelial cells and a 72 +/- 1% and 58 +/- 1% decrease in IGF-I binding in the aortic and capillary endothelial cells, respectively. Thus, macro- and microvascular endothelial cells bind and process insulin and IGF-I by degradative and nondegradative pathways. The predominance of the nondegradative pathway for the processing of insulin and IGF-I and the modulation of their receptors by physiological hormone concentrations suggested that endothelial cells may regulate the access of insulin and IGF-I to their target cells.     

Binding and intracellular degradation of atrial natriuretic factor by cultured vascular smooth muscle cells

Binding studies were performed on vascular smooth muscle cells (VSMC) from the rat aorta, using 125I-atrial natriuretic factor (Ser99-Tyr126) (ANF (Ser99-Tyr126] as the ligand. Kinetic studies at 37 degrees C indicated a rapid onset of binding with a maximum total binding of 25% being reached by 60 min. Crosslinking experiments demonstrated that ANF bound to a 120 kDa and a 60 kDa protein with the former dissociating into the 60 kDa species in presence of beta-mercaptoethanol. Of the total radioactivity bound, 15% represented internalized material. Analysis of the medium after different incubation periods revealed a 42% degradation of 125I-ANF by 120 min. At 4 degrees C, no internalization of 125I-ANF was observed. However, surface binding occurred, albeit at a much slower rate, and not reaching a maximum even at the end of 3 h. No degraded material was detected in the extracellular medium even after a 2-h incubation. Chloroquine (100 microM) and monensin (10 microM) significantly increased the cell-associated radioactivity, causing a 2- to 3-fold elevation of internalized material and a 1.5- to 2-fold rise in the surface-bound ligand. Both lysosomotropic agents also inhibited ANF degradation by 70-80%. Kinetic of the intracellular labeled material was analyzed: within 5-10 min it reaches a maximum level and it decreases rapidly. In presence of monensin the intracellular signal was amplified and the decay was minimized. The intracellular material was found to be mostly bound to a 60 kDa protein. These studies suggest an intracellular degradation of ANF, probably in the lysosomal compartment, following receptor-mediated endocytosis.     

Cell-type-specific receptors for alpha-fetoprotein in a mouse T-lymphoma cell line.

Binding and uptake of alpha-fetoprotein (AFP) by mouse T-lymphoma YAC-1 cells exhibited the characteristics of receptor-mediated endocytosis. The binding saturation curve obtained by incubating YAC-1 cells at 4 degrees C with 125I-labeled AFP at different concentrations (50 ng/ml to 2.5 mg/ml) showed three saturation plateaus. AFP binding was inhibited by unlabeled mouse, rat, or bovine AFP and, to a lesser extent, by rat or bovine serum albumin. No significant competition was observed with transferrin, alpha 2-macroglobulin, IgG, or ovalbumin. Scatchard analysis suggested the presence of three types of receptor sites with a Kd of 2.2 X 10(-9) M (approximately equal to 700 sites per cell), 8.6 X 10(-7) M (approximately equal to 210,000 sites per cell), and 5.7 X 10(-6) M (approximately equal to 910,000 sites per cell), respectively. At 37 degrees C, AFP was rapidly internalized and could be localized in the cytoplasm after incubation of cells with fluoresceinated AFP. After a short residence time, AFP was released undegraded from the cells. Normal adult thymocytes and T lymphocytes, which are counterparts of YAC-1 cells, did not show any significant uptake of AFP. On the other hand, a small subpopulation of fetal and newborn thymocytes was labeled by fluoresceinated AFP.     

High density lipoprotein inhibits low density lipoprotein binding and uptake by bovine aortic endothelial cells

The antiatherogenic effect of high density lipoprotein (HDL) has been attributed to either an inhibition of cholesterol uptake or to reversed cholesterol transport from peripheral cells. In order to determine whether HDL competitively blocks receptor-mediated low density lipoprotein (LDL) binding and uptake, bovine aortic endothelial cells (BAECs) were cultured in Dulbecco's modified Eagles Medium (DMEM) containing 10% LDL-free fetal bovine serum, and incubated with 125I-LDL in concentrations of either 10 or 25 micrograms protein/mL. Varying amounts of HDL (0-200 micrograms/mL) were added to the media. Following a twenty-four hour incubation period at 37 degrees C, 125I-LDL binding and uptake were measured. At the lower concentration of 125I-LDL, which represents high-affinity receptor binding, there was no significant difference in either binding or uptake within the range of HDL concentrations studied. At the higher concentration of LDL, however, there was a marked inhibition of 125I-LDL binding (p less than .006) and uptake (p less than, 001; ANOVA), which did not saturate at the highest HDL concentrations used. These data suggest that HDL does not influence high-affinity, receptor-mediated binding and uptake of LDL but that its effect is seen at a concentration of LDL representing nonspecific binding. The lack of saturation at increasing concentrations of HDL also indicates that HDL-receptor interaction is not essential for the effect.     

An acid-stable insulin-like growth factor (IGF)-binding protein from pig serum inhibits binding of IGF-I and IGF-II to vascular endothelial cells

The effects of a porcine insulin-like growth factor (IGF)-binding protein on binding of IGF-I and IGF-II to porcine aortic endothelial cells (PAEC) were determined. Binding of 125I-labelled IGF-I and -II to IGF receptors was inhibited by IGF-binding protein. IGF-binding protein inhibited binding of IGF-I and -II in a dose-dependent manner with half-maximal inhibition occurring at 5.43 and 108 micrograms/l respectively. A 125I-labelled IGF-I--IGF-binding protein complex, formed by incubating 125I-labelled IGF-I with IGF-binding protein overnight at 4 degrees C, did not effectively bind to endothelial IGF receptors. Addition of IGF-binding protein to PAEC previously incubated with IGF-I caused a marked dissociation of bound IGF-I (47% dissociation within 12h). These results indicate that the acid-stable IGF-binding protein which appears to be a part of the 150 kDa GH-dependent binding protein, blocks binding of IGF-I and -II by the IGF receptors and appears to exhibit a higher affinity for IGF-I than the endothelial type-I IGF receptor. The ramifications of this latter point with respect to transfer of circulating IGFs (bound to their IGF-binding proteins) across the vascular endothelium are not clear.     

Oxidative stress impairs insulin internalization in endothelial cells in vitro

Aims/hypothesis. Because oxidative stress has been suggested to be a significant contributing factor in the development of endothelial dysfunction and insulin resistance, we investigated whether reactive oxygen species contribute to insulin resistance by impairing insulin uptake through an effect on endothelial insulin receptor function. Methods. Following a 2-h pro-oxidant challenge with xanthine oxidase, we examined the temporal pattern of insulin processing in the human umbilical endothelial cell line Ea.Hy926 and bovine aortic endothelial cells equilibrated with [¹²⁵I]-insulin. Insulin receptor mRNA concentrations were analysed by RT-PCR and insulin receptor tyrosine phosphorylation and protein concentrations were estimated by western blotting. Results. Xanthine oxidase exposure resulted in a major reduction in total insulin receptor-mediated [¹²⁵I]-insulin internalization over a 1-h period in both Ea.Hy926 and bovine aortic endothelial cells. After 15 min, untreated bovine aortic endothelial cells internalized fivefold more cell-bound [¹²⁵I]-insulin than pro-oxidant treated cells. The [¹²⁵I]-insulin disappeared from the cell surface at a similar rate in both pro-oxidant and untreated cells, with relatively more [¹²⁵I]-insulin being released into the medium in pro-oxidant treated cells. Although xanthine oxidase reduced insulin receptor mRNA and protein concentrations, cell surface insulin binding capacity was not affected. Following 5 min insulin exposure, insulin receptor auto-phosphorylation was considerably reduced in cells challenged with xanthine oxidase for 2 h, which could be important for insulin receptor activation and internalization. Conclusion/interpretation. Oxidative stress impairs insulin endocytosis in both arterial and venous endothelial cell lines. This was not a consequence of modified insulin binding capacity but could involve insufficient insulin receptor activation. [Diabetologia (2001) 44: 605–613] Received: 2 October 2000 and in revised from: 18 December 2000     

Toxic action of advanced glycation end products on cultured retinal capillary pericytes and endothelial cells: relevance to diabetic retinopathy

Summary The toxic effects of advanced glycation end products (AGEs) on bovine retinal capillary pericytes (BRP) and endothelial cells (BREC) were studied. AGE-modified bovine serum albumin (AGE-BSA) was toxic to BRP. At a concentration of 500 μg/ml it reduced the BRP number to 48 ± 3 % (p < 0.05) of untreated controls, as determined by cell counting with haemocytometer. AGE-BSA was also toxic to bovine aortic endothelial cells (BAEC) reducing cell number to 84 ± 3.1 % of untreated controls. Under similar conditions, low concentrations (62.5 μg/ml) of AGE-BSA were mitogenic to BREC increasing the cell proliferation to 156 ± 11 % (p < 0.05) above that of untreated controls. At a higher dose of 500 μg/ml AGE-BSA decreased the proliferation of BREC to 85 ± 6 % of untreated controls. Immunoblot analysis demonstrated that BRP and BREC express the p60 AGE-receptor. Retinal capillary bed from the human also stained positively for the p60 AGE-receptor. Addition of 0.25 μg/ml of p60 AGE-receptor antibody was able to block the effects of AGE-BSA on BRP and BREC. The level of binding of [¹²⁵I]-labelled AGE-BSA to the cell surface was small but significant among the three cell types. There was also an increase in the internalized pool of radioligand in BRP and BREC but this was very much lower than in BAEC. In all the cell types the internalized pool of [¹²⁵I]-labelled AGE-BSA was much larger than the amount associated with the cell surface. Degradation products were not detected in the media over the 24-h incubation of the cells with [¹²⁵I]AGE-BSA. The binding of [¹²⁵I]-labelled AGE-BSA to the cell surface was prevented by the addition of p60 AGE-receptor. These results suggest that the interaction of AGE-modified proteins with the membrane-bound AGE-receptor may play an important role in the pathogenesis of diabetic retinopathy. [Diabetologia (1997) 40: 156–164] Received: 23 August 1996; and in revised form: 29 October 1996     

Specific binding sites for insulin in the rat pituitary gland

Studies were carried out in order to characterize specific insulin binding sites in the rat pituitary gland. Binding of labeled insulin by pituitary microsomes reached equilibrium after 4 h at 4 degrees C and remained stable over 16 h; at 25 degrees C the plateau was reached in 20 min. Equilibrium binding data analysis of competitive displacement of bound 125I-iodo insulin by unlabeled insulin yielded a non-linear Scatchard plot. At 25 degrees C the Kd for the high affinity component was 2.8 +/- 0.1 X 10(-9) M and the receptor concentration was 260 +/- 80 fmol/mg of microsomal protein. A Kd value of 4.6 +/- 0.4 X 10(-8) M and a binding capacity of 800 +/- 200 fmol/mg microsomal protein were obtained for the low affinity sites. Insulin binding to microsomes was enhanced 2.7 times by increasing the ionic strength of the incubation medium with 2 M NaCl, and was abolished when the microsome preparation was preincubated with trypsin prior to binding measurements. Other hormones, such as bovine thyrotropin, ovine follitropin, human somatotropin and ovine prolactin did not interact with the insulin receptor. Proinsulin displaced the labeled hormone in direct proportion to its insulin-like biological activity.     

Insulin receptors in human endothelial cells: identification and characterization.

Specific binding of 125I-insulin was found in cultured human endothelial cells obtained from human umbilical veins. The binding reaction was rapid and reversible, demonstrated receptor site-site interactions of the negatively cooperative type, and was dependent on the temperature, pH and duration of incubation. At 21 degrees C, steady-state conditions of binding occurred in 90 minutes, the pH optimum was 7.8 and less than 10% of the labeled hormone was degraded. Binding of tracer amounts of 125I-insulin was inhibited by concentrations of unlabeled insulin as low as 0.2 ng/ml and 50% inhibition was obtained at 2-5 ng/ml of unlabeled insulin. Unlabeled porcine insulin, porcine proinsulin and desoctapeptide insulin inhibited the binding of 125I-porcine insulin in direct proportion to their biological potencies, whereas to the cells was inhibited by antibodies against insulin receptors. We conclude that human endothelial cells possess specific receptors for insulin whose physio-chemical properties are similar to those of insulin receptors in other tissues.     

Insulin binding and degradation in vascular endothelial cells: modulation by cell growth and culture organization

The interaction of insulin with the vascular endothelium and its modulation by cell growth and culture organization was studied using bovine aortic endothelial cells in monolayer cultures. Three types of cultures were investigated: 1) confluent nondividing cultures, organized and differentiated as the in vivo tissue; 2) subconfluent, not yet organized cell cultures, representing proliferating endothelium; and 3) endothelial cell cultures modified to lose their property of contact inhibition, growing in multiple layers. All three types of cultures exhibited specific binding of 125I-insulin to high and low affinity cell surface receptor sites, and were capable of degrading 125I-insulin. Preexposure of the cultures to insulin resulted in a time dependent reduction in the availability of cell surface receptors (down-regulation). Insulin binding per cell was 2.4-fold and 10-fold higher in the subconfluent and modified cultures, respectively, as compared to the contact-inhibited confluent cultures. Similarly, the rate of insulin degradation was higher in the subconfluent and modified cultures (2.3-fold and 20-fold, respectively). Subconfluent cultures were more sensitive than confluent cultures to the down-regulatory effect of insulin. They exhibited a 60% decrease in insulin binding as compared to a 40% decrease in confluent cultures after preexposure to 50 ng/ml insulin. The increase in insulin binding and degradation in growing endothelial cells suggests a role for the hormone in the regulation of endothelial cell growth, e.g. in response to injury. This was further supported by the observation of a dose-dependent stimulation of [3H]thymidine incorporation into sparse, serum-starved endothelial cells by physiological concentrations of the hormone.     

Binding and internalization of transforming growth factor-beta 1 by human hepatoma cells: evidence for receptor recycling.

Cellular processing of 125I-labeled transforming growth factor-beta 1 was investigated in the human hepatoma cell lines Hep G2 and Hep 3B. Binding of 125I-transforming growth factor-beta 1 to cell surface receptors was specific, saturable and calcium-independent. Both cell lines exhibited a single class of high-affinity (Kd = 2.2 x 10(-10) mol/L) binding sites (4.5 x 10(3) for the Hep G2 cell; 1.5 x 10(3) for the Hep 3B cell) for both human and porcine transforming growth factor-beta 1. Binding was temperature dependent, time dependent and pH dependent. Cell-bound 125I-transforming growth factor-beta 1 was removed by brief exposure to acidic medium (pH less than 4) but was converted into an acid-resistant state rapidly after shifting the cells to 37 degrees C. Spontaneous dissociation of bound ligand over a 6 hr period at 4 degrees C was less than 10%. Disuccinimidyl suberate was used to covalently label 125I-transforming growth factor-beta 1 to cell-surface binding sites. Labeling of the ligand/receptor complexes was inhibited by unlabeled transforming growth factor-beta 1 but was unaffected by other growth factors. The radiolabeled complexes showed approximate molecular weights of 280,000, 85,000 and 65,000 when run on reducing sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Cell-bound 125I-transforming growth factor-beta 1 was internalized and degraded at 37 degrees C, and the products were released into the medium as trichloroacetic acid-nonprecipitable radioactivity. The lysosomotropic base chloroquine and the carboxylic ionphore monensin inhibited degradation and release of 125I-labeled products from the cells.(ABSTRACT TRUNCATED AT 250 WORDS)     

Processing of insulin-like growth factors I and II by capillary and large vessel endothelial cells

Cultured endothelial cells from bovine capillaries and pulmonary arteries were incubated with highly purified [125I]insulin-like growth factor I ([125I]IGF-I), [125I]IGF-II, or [125I]insulin for periods up to 120 min, the cells were washed, and the cell-bound radioactivity was allowed to dissociate from the cells into fresh incubation medium. For insulin, 85-95% of the 125I dissociated from cells in 15 min, with 75% dissociating by 5 min. The 125I material released into the medium during the first 5 min of dissociation was entirely intact insulin, while the material released in the next 10 min was 80% intact insulin. For the [125I]IGFs, several differences were observed. First, after 5 min of dissociation, only 30-45% of the cell-bound 125I was released into the medium. For IGF-I, this rapidly dissociating material was entirely intact peptide, whereas for IGF-II, up to 55% of the dissociated radioactivity was degraded peptide. Second, during the next 10 min of dissociation, an additional 20% of the 125I was released from the cells. For IGF-I, this was 85% intact peptide; however, for IGF-II, this dissociated fraction contained as little as 25% intact peptide. Third, for both IGFs, after 15 min of dissociation, 40-65% of the initial cell-associated 125I remained within the endothelial cells; after the 15-min dissociation period, 95% of the remaining internalized 125I was intact IGF-I, whereas for IGF-II, the internalized 125I was 90% intact IGF-II after short periods of association (less than 15 min) and progressively decreased to 65% intact peptide after 60 min of association. We conclude that in addition to having separate surface receptors for insulin, IGF-I, and IGF-II, endothelial cells also process each hormone by distinct pathways.     

Prothrombin is activated on vascular endothelial cells by factor Xa and calcium.

Vascular endothelial cells derived from adult bovine aorta (ABAE) treated with factor Xa and calcium were found to activate prothrombin. In contrast, nonvascular cells (human foreskin fibroblasts, bovine corneal endothelial cells, or human fetal lung cells) had either no or very little effect on prothrombin activation. In the presence of 6 X 10(5) ABAE cells, 20 ng of factor Xa converted 90 micrograms of prothrombin into 80 units of thrombin after 45 min at 37 degrees C. Exogenous factor V was not required for prothrombin activation, but thrombin generation was enhanced 2- to 4-fold by the addition of factor V (500-2,500 ng/ml). Treatment of ABAE cells with anti-bovine factor V IgG markedly inhibited prothrombin activation by factor Xa and calcium. In cells grown in serum-free medium for 3 months, the amount of factor V activity was equivalent to that found in cells grown with serum, which suggests that these cells probably synthesize factor V. Sparse ABAE cells increased prothrombin activation by factor Xa 6-fold compared to activation in confluent cells. Although previous thrombin treatment of ABAE cells did not enhance prothrombin activation, addition of dansyl arginine-4-ethyl piperidine amide markedly inhibited activation of 125I-labeled prothrombin by factor Xa, indicating that thrombin formation is necessary for optimal prothrombin activation. These data indicate that aortic endothelium may provide a physiologically important surface for activation of prothrombin as well as a mechanism for optimal formation of clots at sites of vascular injury.     

Uptake and degradation of low density lipoproteins (LDL) by confluent, contact-inhibited bovine and human endothelial cells exposed to physiological concentrations of LDL.

Metabolism of low density lipoproteins (LDL) was studied in cultures of endothelial cells derived from bovine aorta or heart and from human umbilical veins. At low LDL concentrations nonconfluent cultures of bovine endothelial cells catabolized more LDL protein than contact-inhibited confluent cultures but this difference was reduced at high LDL concentrations. Nonconfluent human endothelial cells displayed also a higher rate of LDL degradation than their contact-inhibited counterparts, but this difference was less pronounced than in the bovine cells. Bovine endothelial cells grown in the presence of fibroblast growth factor metabolized less LDL than those cultured without fibroblast growth factor (FGF), but this difference was not consistent in the human endothelial cells. The data presented provide evidence that contact-inhibited confluent human endothelial cells are capable of catabolizing LDL when exposed to physiological concentrations of this lipoprotein.     

Receptor-mediated endocytosis and degradative processing of growth hormone by rat adipocytes in primary culture

At 37 degrees C, cultured rat adipocytes bound [125I]human GH ([125I]hGH) rapidly, with binding being detectable within 1 min of incubation. The bound [125I]hGH was then internalized (within 10 min) and accumulated in the cell interior until a steady state was reached (by 60 min). At this time, where the rates of GH internalization, processing, and release are equivalent, 55% of total cell-associated [125I]hGH was intracellular. Internalization of [125I]hGH by acutely isolated (noncultured) adipocytes was preceded by a 20-min lag phase indicative of a temporary postbinding defect. The lag phase was not seen with cultured adipocytes. After preloading of [125I]hGH into the cell interior, cultured cells rapidly released [125I]hGH (t1/2 = 20-30 min) into the extracellular medium as both intact (25%) and degraded (75%) GH. The release of intact vs. degraded GH was distinguishable on the basis of kinetics and temperature dependence. In order to determine when internalized [125I]hGH entered a catabolic compartment, cultured adipocytes were incubated with [125I]hGH and the composition of intracellular GH was determined as a function of time. All [125I]hGH internalized during the first 20 min was intact. Between 20 and 30 min some of the internalized [125I]hGH entered a catabolic compartment and degradation products began accumulating within the adipocytes. Release of degraded [125I]hGH from cultured adipocytes began at 60 min. The processing of GH through the complete degradative pathway (binding, internalization, degradation, release) required a period of 1 h at 37 degrees C.     

Dialyzable factor in human serum of platelet origin stimulates endothelial cell replication and growth

Porcine aortic endothelial cells were isolated and maintained in Dulbecco's modified Eagle's medium (DME medium)/10% citrate-treated human plasma. They were stimulated by DME medium/10% human serum to grow from a density of 10,100 ± 500 per well to a final density of 83,000 ± 1,800 per well over a 9-day period. On the other hand these cells grew poorly (11% increase) in DME medium/10% human platelet-poor plasma prepared without chelating agents and containing platelet factor 4 at 18 ng/ml by radioimmunoassay. Dialysis of the human serum (M_r cutoff, 3,500) eliminated all the stimulatory activity. The activity recovered from the dialysate stimulated growth when added to endothelial cultures in conjunction with either dialyzed serum or platelet-poor plasma alone. The dialyzable factor could be obtained directly from platelets; both acetic acid extracts and boiled NaCl extracts stimulated porcine aortic endothelial cell replication. Gel filtration chromatography on Sephadex G-15 showed that the endothelial growth factor had a molecular weight of 700. Partially purified material induced a concentration-dependent stimulation of porcine aortic endothelial cell replication in the presence of DME medium alone; however, simultaneous incubation with platelet-poor plasma resulted in a much greater response. Fibroblast growth factor isolated from bovine brain was found to be mitogenic only in the presence of nondialyzed serum or of the dialyzable factor together with plasma. In the absence of this serum factor, fibroblast growth factor had no effect. We conclude that human serum contains a potent endothelial cell mitogen of platelet origin. Human plasma that is devoid of platelet content does not stimulate endothelial cell growth. This growth factor may be an important stimulant of the endothelial cell response to vascular wall injury.