Lipid composition of cultured human endothelial cells

Primary cultures of endothelial cell monolayers (ECM) obtained from umbilical cord veins were grown in a medium containing fetal calf serum (FCS) and L-glutamine. Endothelial cells (EC) were harvested after fourth or fifth day of seeding. Lipid extracts were separated into neutral lipids and phospholipid fractions by thin layer chromatography (TLC). The fractions were analyzed by gas liquid chromatography (GLC) for fatty acid distribution. Free cholesterol was the major component of neutral lipids but the percentage was lower in EC than in platelets. Cholesterol ester (ChE) and free fatty acids (FFA) were found in equal percentages and higher than that observed in platelets. Triglycerides comprised 2.6 % of the total lipids. Palmitic, stearic and oleic acids were the main fatty acids of EC neutral lipids. Phosphatidyl choline (PC) was the largest (36.3 %) phospholipid fraction. Phosphatidyl ethanolamine (PE), and the combined phosphatidyl serine and phosphatidyl inositol fractions (PS + PI) were present in equal percentages (25.3 and 25.6 %). Each phospholipid fraction had a characteristic fatty acid pattern. Long chain polyunsaturated FA were present in high concentrations in all fractions. Arachidonic acid (AA) ranged from 4.0 – 18.3 % whereas linoleic acid and di-homo-γ-linolenic acid ranged from 1.9 – 7.3 % and 2.3 – 5.5 % respectively, in the various PL-fractions. EC examined immediately after collection from the umbilical cords, had significantly higher concentrations of linoleic and arachidonic acids in the total phospholipids than EC grown in culture medium for 4–5 days.     

The hemostatic agent ethamsylate enhances P-selectin membrane expression in human platelets and cultured endothelial cells

Ethamsylate possesses antihemorrhagic properties, but whether or not it directly activates blood platelets is unclear. Here we investigated the platelet activation potential of ethamsylate, by measuring membrane P-selectin expression with flow cytometry in human whole blood and also by immunofluorescence imaging of isolated human platelets. Moreover, we measured membrane P-selectin expression in the SV40-transformed aortic rat endothelial cell line (SVAREC) and ¹⁴C-ethamsylate membrane binding and/or uptake in platelets and endothelial cells. Whole blood flow cytometry showed a modest, but statistically significant increase by ethamsylate in the percentage of platelets expressing P-selectin (from 2% to 4–5%, p<0.05). Immunofluorescence showed a sizable (39%) and significant (p<0.01) enhancement of P-selectin expression at the lowest concentration of ethamsylate tested (1 μM), with maximal enhancement of P-selectin expression (75–90%) at 10 μM ethamsylate. Similar results were obtained in SVAREC endothelial cells. ¹⁴C-ethamsylate specifically bound to platelets and endothelial cell membranes, without significant uptake into the cell interior. In conclusion, ethamsylate enhances membrane P-selectin expression in human platelets and in cultured endothelial cells. Ethamsylate specifically binds to some protein receptor in platelet and endothelial cell membranes, receptor which can signal for membrane P-selectin expression. These results support the view that ethamsylate acts on the first step of hemostasis, by improving platelet adhesiveness and restoring capillary resistance.     

Cyclic AMP increases thrombomodulin expression on membrane surface of cultured human umbilical vein endothelial cells.

Dibutyryl-cyclic AMP (Bt2cAMP; final concentration 1-5 mM) or beraprost sodium (synthetic prostacyclin, 100 nM) enhanced the expression of thrombomodulin (TM; an anticoagulant factor of endothelial cells) on the membrane surface of cultured human umbilical vein endothelial cells up to 1.4 times over the control within 9 hrs after the treatment, while the expression fell below the control level at 12 hrs and thereafter. 8-Bromo-cAMP (final concentration 1-5 mM) or 3-isobutyl-1-methylxanthine (IBMX; an inhibitor of phosphodiesterase; final concentration 10-1000 microM) enhanced the expression of TM on the cell surface at 12 hrs after the treatment. The enhancement of TM expression caused by Bt2cAMP was inhibited by incubation with phorbol 12-myristate 13-acetate. These results suggest that cAMP stimulates expression of TM in the endothelial cells.     

The effect of thrombin on fibronectin in cultured human endothelial cells

Cultures of human endothelial cells (EC) incubated for periods up to 24 h with highly purified thrombin (2 NIH u/ml) contained considerably less cell-associated fibronectin fibrils than corresponding controls. The loss of fibronectin fibrils was evident after 4 h and was accompanied by a 2-3 fold increase in the concentration of fibronectin in the incubation medium. Hirudin inhibited the effects of thrombin. Thrombin also induced characteristic shape changes of EC. These shape changes were reversible within a 4-6 h period and could not be reinvoked by new additions of thrombin. Thus, structural refractoriness to thrombin coincided temporally with a period when EC-associated fibronectin fibrils were markedly reduced.     

The effects of intact platelets on cultured human endothelial cells

Effects of intact platelets on DNA synthesis, replication and the morphology of cultured human endothelial cells were studied. When either primary or first passage endothelial cells were seeded at a low density, the cells formed small clusters with large intercellular spaces. When intact platelets were added to these cultures, the cells appeared to migrate from these clusters and dispersed themselves more evenly over the culture surface. In addition, intact platelets were found to enhance the repair of a scratched defect produced in confluent primary cultures. Cells that were cultured with platelets also appeared to be changed from their ususal polygonal shape to a more elongated or bipolar form. This effect appeared at a platelet count of about 10,000 platelet/μl of medium and after 12 hours of culture. Maximal effects were seen at a platelet density of 100,000/μl and after 24 hours of cultivation. DNA synthesis and cell replication were not essential for this effect. However, if these cultures were incubated for an additional 48 hours, enhancement of DNA synthesis and the labelling index was found, but a significant increase in cell number was not yet apparent. If however, these cells were cultured for an additional 72 hours then a significant increase in cell growth was noted. Cells disrupted by either sonification or freeze-thawing were without effect. Treating the platelet with either neuraminidase or trypsin did not eliminate this effect. This effect appeared to be limited to intact platelets and not to leukocytes or erythrocytes. Examination of the platelet-treated cells by electron microscopy revealed a small number of cells that contained platelets within their cytoplasm. These observations may provide a mechanism by which platelets help to maintain and support the integrity of the vessel wall.     

Glycosphingolipid composition of primary cultured human brain microvascular endothelial cells

Glycosphingolipid (GSL) antigens have been considered to be involved in the pathogenesis of autoimmune neurologic disorders including multiple sclerosis. To establish the GSL pattern specific for endothelial cells forming blood-brain barrier (BBB), we established a method to yield sufficient quantities of highly purified human brain microvascular endothelial cells (HBMECs) and compared their GSL composition to that of human umbilical cord vein endothelial cells (HUVECs), as the representative of endothelial cells not forming BBB. The major gangliosides were GM3 and sialyl paragloboside (LM1), and the major neutral GSLs were lactosylceramide (LacCer), globotriaosylceramide (Gb3), and globoside (Gb4). Trace amounts of GM1, GD1a, GD1b, GT1b, and sulfoglucuronosyl paragloboside (SGPG) could be detected by the high performance thin layer chromatography-overlay method. SGPG was detected only at a nonconfluent state in an amount almost 1/30 that of in nonconfluent HUVECs. Conversely, GM3 and LM1 increased significantly after confluency. The amount of Gb3 in HBMECs was almost as twice that in HUVECs. The significance of these differences in GSL content between HBMECs and HUVECs and between confluent and nonconfluent states is obscure. It might be related, however, to the defense mechanism at the BBB and to the susceptibility of the central nervous system in some disorders that target cell surface GSL, such as hemolytic uremic syndrome.     

Inflammatory mediators stimulate granulocyte adherence to cultured human endothelial cells

The factors that regulate the adherence of granulocytes to the endothelium under normal conditions, and in states of inflammatory vascular injury, are largely unknown. We found that treatment of primary, confluent monolayers of human umbilical vein endothelial cells with inflammatory mediators (zymosan-activated plasma, as a source of C5a fragments, and N-formylmethionyl-leucyl-phenylalanine) stimulated granulocyte adherence to the monolayers. The augmented adherence was dose-, time-, and temperature-related and demonstrated kinetics characteristic of the adherence of single cells with increased affinity for monolayer cultures. Adherence stimulated by the inflammatory mediators was not prevented by washing the monolayers after treatment with the mediator. Human albumin diminished both spontaneous and stimulated adherence, although the relative increase in adherence stimulated by the inflammatory mediator was unchanged. Calcium and magnesium were required for stimulated adherence. These data suggest that inflammatory mediators may bind to, or directly alter, human endothelial cells resulting in enhanced granulocyte adherence, and define characteristics of this endothelial cell-granulocyte interaction.     

Binding of heparin on the surface of cultured human endothelial cells

³H-labelled heparin was shown to bind to the surface of cultured human endothelial cells. The binding was time dependent, saturable, reversible and trypsin sensitive. The binding capacity corresponds to about 10⁶ molecules of heparin per cell. Labelled material released from the cells retained its ability to bind to antithrombin III. Binding of heparin to endothelial cells as well as its release is suggested to be of importance in the pharmacokinetics of heparin and to influence the non-thrombogenic properties of the endothelial lining during heparin therapy.     

Interaction of 3H-defibrotide with cultured human umbilical vein endothelial cells

Defibrotide is a profibrinolytic and antithrombotic drug which seems to modulate endothelial cell function. In this study, a method for radioactive labeling of the drug and its interaction with cultured endothelial cells is proposed. 3H-Acetic anhydride was used to label defibrotide. Endothelial cells obtained by collagenase treatment of human umbilical cord veins were cultured in 24-welled plastic culture dishes. Binding experiments were carried out by incubating cell cultures with media containing various concentrations of labeled defibrotide. Our results showed that labeled defibrotide has a KL value of 4.2 micrograms/ml for endothelial cells. Although the presence of a specific transporter is possible, the high molecular weight of the fraction used suggests that the interaction is binding to a specific receptor.     

Voltage-dependent membrane currents of cultured human neurofibromatosis type 2 Schwann cells

Previous experimental observations indicate that inhibition of voltage-dependent K⁺ currents suppresses proliferation of normal Schwann cells. In the present study we tested the opposite relationship, i.e., whether Schwann cells from tumors with abnormally high rates of proliferation would have an increase in membrane K⁺ currents. Whole-cell membrane currents were studied in cultured cells from schwannomas of two neurofibromatosis type 2 (NF2) patients (n = 53), one patient with a sporadic schwannoma (n = 22), and two control subjects (n = 41). Five different types of voltage-dependent membrane currents were found in all of the Schwann cells tested. Membrane depolarization activated outward K⁺ and Cl⁻ currents; quinidine was found to block the K⁺ current (IC₅₀ ≈ 1 μM), and NPPB reduced the Cl⁻ current. Ba²⁺-sensitive inward rectifier K⁺ currents, fast Na⁺ currents, and a transient, inactivating K⁺ current were less frequently observed. On average, NF2 cells were found to have statistically significant higher membrane potential and larger non-inactivating K⁺ outward current as compared to controls. Electrophysiological parameters of Schwann cells from a sporadic schwannoma showed a tendency for larger outward currents; however, the difference did not reach statistical significance. Together the data support the suggestion of a possible link between K⁺ outward current and proliferation of Schwann cells. GLIA 24:313–322, 1998. © 1998 Wiley-Liss, Inc.     

Uremic medium disturbs the hemostatic balance of cultured human endothelial cells.

We have investigated the ability of serum from uremic patients to modify the thrombogenic properties of the endothelium. The effects of uremic medium on the morphology of endothelial cells (ECs), and their resistance to flow was analyzed. The influence of uremic media on the reactivity of the extracellular matrix (ECM) generated by ECs towards normal platelets was evaluated in a parallel-plate perfusion chamber. Exposure of ECs to uremic medium resulted in abnormal cell morphology and signs of an accelerated growth. Detachment of ECs exposed to circulating blood was increased when cells had been grown with media supplemented with uremic serum (21% vs. 14% non exposed). Platelet deposition was significantly elevated on ECMs generated in the presence of uremic media (uremicECMs) (p<0.01 vs. control studies). Effects of uremic serum were not observed at short incubation periods (5 h) but were evident after 24 or 72 h of incubation. Northern blot analysis revealed increased expression of tissue factor (TF) mRNA in ECs exposed to uremic conditions. Immunocytochemical methods detected an augmented expression of TF antigen on uremic ECMs. Incubation of ECMs with an antibody to human tissue factor prevented the increase in platelet deposition observed in uremic ECMs, suggesting that the presence of TF in ECM could be responsible for the enhanced platelet deposition. Results from our study indicate that uremic medium impairs the antithrombotic functions of cultured endothelial cells.     

Elastase-like activity in cultured aortic endothelial cells

Cultured porcine aortic endothelial cells were studied for cellular and secreted elastase activity. We describe an activity hydrolyzing the synthetic elastase substrate, succinyl(alanine)3 nitroanilide, but not elastin, which was shown to be membrane located and was not secreted to the culture medium. A different neutral proteinase activity degrading insoluble elastin was demonstrated in the culture medium following its fractionation by gel filtration high performance liquid chromatography (HPLC). Since no elastinolytic activity could be directly detected in the conditioned medium, it is likely that the chromatographic separation removed an endogenous inhibitor.     

Effects of thrombin on the integrity of monolayers of cultured human endothelial cells

⁵¹Cr-prelabelled endothelial cells (EC) in confluent monolayers were incubated in RPMI 1640 + foetal calf serum 20% (v/v) to which purified thrombin was added. Thrombin (≧ 0.1 NIH U/ml) significantly accelerated ⁵¹Cr-release and caused extensive but reversible cell “contraction”. Thrombin-exposed EC reacted to a new dose of thrombin with no appreciable shape change, but ⁵¹Cr-efflux was again accelerated. EC exposed to thrombin pretreated with N-bromosuccinimide (modifying the macromolecular site) or phenylmethylsulfonyl fluoride (blocking the serine site) retained normal morphology and did not leak excess amounts of ⁵¹Cr. Antithrombin III also inhibited the effect of thrombin. Pretreatment of EC with either indomethacin, aspirin, sulfinpyrazone, pronase or neuraminidase did not influence the effect of subsequent thrombin exposure.     

Epidermal growth factor stimulates prostacyclin production by cultured human vascular endothelial cells

Epidermal growth factor (EGF) stimulated prostacyclin (PGI2) production by cultured human umbilical vein endothelial cells, as measured by radioimmunoassay of its stable metabolite 6-keto-prostaglandin F1 alpha. This effect of EGF was dose-dependent, the lowest stimulatory concentration of EGF was 1.0 ng/ml and 100 ng/ml caused a 2.7 +/- 0.3 (mean +/- SEM) fold increase in the PGI2 synthesis. The stimulation appeared at 3-6 h of incubation and lasted at least 24 h. It was suppressed by EGF antibodies and blocked by protein synthesis inhibitor cycloheximide. Cells preincubated 12 h with EGF released also higher amounts of PGI2 when incubated with thrombin for 5 min. It is concluded that EGF liberated from platelets during aggregation may prevent local thrombogenesis and atherogenesis by stimulating the release of the antiaggregatory, vasodilatory PGI2 from vascular endothelial cells.     

Immunoelectrophoretic studies of prekallikrein in human plasma

Using a rabbit anti-human prekallikrein antibody crossed immunoelectrophoresis (CIE) and Laurell rocket antigen determinations were done in plasmas of subjects with Fletcher (prekallikrein, PKA), Fitzgerald (high molecular weight kininogen), Hageman (XII), and PTA (XI) deficiencies as well as in patients with activation of coagulation (intravascular coagulation syndromes). Abnormal CIE patterns were seen in the Fletcher and Fitzgerald deficient plasmas and also in some of the patients with intravascular coagulation. $\text{In}$$\text{vitro}$ studies of plasma treated with thrombin, plasmin, and contact activating agents indicated that abnormal CIE patterns and increased PKA antigen levels were indicative of activation of the Hageman factor dependent pathway and not the result of plasma clotting by thrombin. $\text{In}$$\text{vivo}$ activation of the Hageman factor dependent pathway frequently results in an abnormal CIE and a low PKA antigen level.     

Hirudin stimulates prostacyclin but not endothelin-1 production in cultured human vascular endothelial cells

To study the effect of hirudin on endothelial cell prostacyclin (PGI₂) and endothelin-1 (ET-1) production, we cultured human umbilical vein endothelial cells (HUVECs), stimulated them with 0.00001–10 kU/l of hirudin for 12–24 hours, and measured by radioimmunoassays the concentrations of 6-ketoprostaglandinF_{1 alfa} (6-keto, a metabolite of PGI₂) and ET-1 in the incubation medium. In incubation medium containing 10% serum hirudin stimulated PGI₂-production dose-dependently. The lowest stimulatory hirudin concentration was 0.001 kU/l, which increased the concentration of 6-keto by 10.8±4.4% (mean±S.E) (p<0.01). The greatest stimulation rate (28.6±6.2 %, p<0.001) was obtained with the highest hirudin concentration (10 kU/l), when the culture medium contained 10% human serum. The PGI₂-stimulating activity was exaggerated in the absence of serum, when 1 kU/l of hirudin increased PGI₂-production by 59.7±6.2% (p<0.001, N=14). Stimulation of PGI₂ appeared after 12 hour incubation. Hirudin had no effect on the conversion of exogenous arachidonic acid to 6-keto or on the production of ET-1. We thus conclude that hirudin stimulates PGI₂-production through de novo protein synthesis. Stimulation of PGI₂-production by hirudin may contribute to its antithrombotic activity, since PGI₂ favours vasodilatation and attenuates platelet aggregation.     

15 deoxy delta12,14 PGJ2 induces procoagulant activity in cultured human endothelial cells

15 deoxy delta12,14 PGJ2 (15d-PGJ2), a high affinity ligand of peroxisome proliferator-activated receptor gamma (PPARgamma) has been proposed to act as a negative feedback regulator of the inflammatory response. We investigated the effect of 15d-PGJ2 on the anticoagulant property of endothelial cells. 15d-PGJ2 stimulated a moderate but sustained increase in tissue factor (TF) activity in HUVECs and EA.hy926 cells while causing a partial loss of thrombomodulin (TM) activity. When cells were co-treated with 15d-PGJ2 and TNF-alpha, the subsequent elevation of TF activity was synergistically increased over that of cells treated with TNF-alpha alone and the decline of TF activity after 24 h was less marked than TNF-alpha alone. The induction of TF by 15d-PGJ2 alone and in combination with TNF-alpha was reduced in the presence of PD 98059, suggesting the participation of the MEK/ERK pathway. The thiazolidinedione PPARgamma agonist ciglitazone had no effect on TF levels but reduced the expression of endothelial protein C receptor. The ability of 15d-PGJ2 to enhance a procoagulant phenotype arising from TNF-alpha suggests a pro-inflammatory role for the prostaglandin.