Activation of the alternative pathway of human complement by apoptotic human umbilical vein endothelial cells

Complement activation generally does not occur on homologous cells. We observed C3 deposition on cultured human umbilical vein endothelial cells (HUVEC) when those which had died of apoptosis were treated with human serum. The C3 deposition on apoptotic HUVEC required Mg2+ but not Ca2+. In addition, the incubation of apoptotic HUVEC with purified C3, B, and D in the presence of Mg2+ resulted in C3 deposition. These results indicated that the C3 deposition on apoptotic HUVEC is mediated by the activation of the alternative complement pathway. C3 contains an intrachain thioester bond in the alpha chain (110 kDa) and upon activation to C3b, binds with membrane molecules by forming an ester or amide bond. Western blotting of reduced C3b-membrane molecule complexes, isolated from serum-treated apoptotic HUVEC by immunoprecipitation with anti-C3, revealed the covalent binding of C3b to several membrane molecules. Most of the C3b-membrane molecule complexes were cleaved by hydroxylamine, suggesting covalent binding via an ester bond. The molecular mass of the major alpha chain fragment released by hydroxylamine treatment was not 105 kDa but 68 kDa, which corresponds to the alpha 1 fragment of iC3b. These results indicate that most of the C3b on HUVEC was cleaved at its alpha' chain to yield iC3b, which consists of three disulfide-linked polypeptide chains and is a ligand of the complement receptor type 3 (CR3) of phagocytes. These results suggest that apoptotic HUVEC can activate the alternative pathway of the homologous complement and that the complement is related to the clearance of apoptotic cells by phagocytes.     

Structure/function analyses of recombinant variants of human factor Xa: factor Xa incorporation into prothrombinase on the thrombin-activated platelet surface is not mimicked by synthetic phospholipid vesicles

This report describes the expression, purification, and characterization of a series of recombinant factor Xa variants bearing aspartate substitutions for each of the glutamate residues which normally undergo gamma-carboxylation. Factor X was expressed in human embryonic kidney cells and purified from conditioned media by immunoaffinity and hydroxylapatite chromatography. Factor X was activated with Russell's viper venom factor X activator, and single-chain unactivated factor X was removed from activated factor X by size-exclusion chromatography. Recombinant wild-type factor Xa had normal activity in a clotting assay, and mutants with aspartate substitutions for glas residues 16, 26, and 29 had no detectable clotting activity. In purified component assays, these gla variants had essentially no detectable activity in the prothrombinase complex assembled on synthetic phospholipid vesicles but had significant activity when the prothrombinase was assembled on thrombin-activated platelets. In addition, the gla 32 variant had normal activity in the platelet prothrombinase but diminished activity in prothrombinase assembled on synthetic PSPC vesicles. These differences were not accounted for by the total phospholipid composition of the thrombin-activated platelet membrane. We have produced fully active recombinant human factor Xa and demonstrated that gla residues 16, 26, and 29 are critical for normal activity of factor Xa. More importantly, this study provides an extensive characterization of macromolecular enzyme complex formation with gla variants of a vitamin K-dependent coagulation protein and provides evidence that prothrombinase complex assembly on thrombin-activated platelets is not equivalent to assembly on synthetic phospholipid vesicles. The data suggest that thrombin-activated platelets possess some element(s) (other than 30% phosphatidyl serine or factor Va), presumably either protein or phospholipid, that serves as a component of the factor Xa binding site.     

Inhibition of prothrombinase by human secretory phospholipase A2 involves binding to factor Xa

Human group II secretory phospholipase A2 (hsPLA2) exhibits significant anticoagulant activity that does not require its enzymatic activity. We examined which coagulation factor was targeted by hsPLA2 and analyzed which region of the protein may be involved in this inhibition. Prothrombin time coagulation assays indicated that hsPLA2 did not inhibit activated factor V (FVa) activity, whereas activated factor X (FXa) one-stage coagulation assays suggested that FXa was inhibited. The inhibitory effect of hsPLA2 on prothrombinase activity of FXa, FV, phospholipids, and Ca2+ complex was markedly enhanced upon preincubation of hsPLA2 with FXa but not with FV. Prothrombinase activity was also strongly inhibited by hsPLA2 in the absence of PL. High concentrations of FVa in the prothrombinase generation assay reversed the inhibitory effect of hsPLA2. By using isothermal titration calorimetry, we demonstrated that hsPLA2 binds to FXa in solution with a 1:1 stoichiometry and a Kd of 230 nM. By using surface plasmon resonance we determined the rate constants, kon and koff, of the FXa/hsPLA2 interaction and analyzed the Ca2+ effect on these constants. When preincubated with FXa, synthetic peptides comprising residues 51-74 and 51-62 of hsPLA2 inhibited prothrombinase assays, providing evidence that this part of the molecule, which shares similarities with a region of FVa that binds to FXa, is likely involved in the anticoagulant interaction of hsPLA2 with FXa. In conclusion, we propose that residues 51-62 of hsPLA2 bind to FXa at a FVa-binding site and that hsPLA2 decreases the prothrombinase generation by preventing FXa.FVa complex formation.     

TNF-alpha stimulates the biosynthesis of complement C3 and factor B by human umbilical cord vein endothelial cells

The effect of basic fibroblast growth factor (bFGF) on apoptosis in normal rat palatal fibroblasts and rat palatal scar fibroblasts was examined by the TUNEL method in order to clarify the mechanism of apoptosis induction in myofibroblasts during the scar formation process. A percentage of scar fibroblasts undergoing apoptosis was significantly higher than that of palatal fibroblasts when they were treated with bFGF succeeding to serum starvation. Palatal fibroblasts, phenotypically modulated into myofibroblasts by the pretreatment with transforming growth factor-beta 1 (TGF-beta 1), similarly showed a higher level of apoptosis induction by bFGF-treatment. TGF-beta 1 elevated protein and mRNA level of FGF receptor (FGFR) in palatal fibroblasts. Tyrosine autophosphorylation of FGFR upon stimulation by bFGF was significantly higher in scar fibroblasts than in normal palatal fibroblasts. These findings suggested that bFGF may be a potential stimulator of apoptosis in myofibroblasts during palatal scar formation and that FGFR may be responsible for this process.     

Effect of flupirtine on cell death of human umbilical vein endothelial cells induced by reactive oxygen species

Flupirtine (KATADOLON), known as a nonopiate centrally acting analgesic drug, was tested as to its potential to prevent apoptosis of human endothelial cells induced by reactive oxygen species (ROS). It was found that Flupirtine displayed no effect on viability and cell proliferation of human umbilical vein endothelial cells (HUVEC) up to a concentration of 10 microg/mL. Apoptosis, induced by ROS and generated by hypoxanthine/xanthine oxidase (EC 1.1.3.22) (HX/XOD) or t-butyl hydroperoxide, was reduced after preincubation with Flupirtine for 3 hr by 35% and 41%, respectively. The maximal cytoprotective effect against apoptosis was observed at a drug concentration of 1 to 3 microg/mL. Flow cytometric studies revealed that Flupirtine was able to decrease the number of necrotic cells as well as of apoptotic cells. Neither the simultaneous administration of Flupirtine with the apoptosis-inducing agent nor the preincubation of HUVEC with Flupirtine influenced the increase in the intracellular Ca2+ concentration [Ca2+]i caused by the production of ROS.     

The solution structure of human coagulation factor VIIa in its complex with tissue factor is similar to free factor VIIa: a study of a heterodimeric receptor-ligand complex by X-ray and neutron scattering and computational modeling

Factor VIIa (FVIIa) is a soluble four-domain plasma serine protease coagulation factor that forms a tight complex with the two extracellular domains of the transmembrane protein tissue factor in the initiating step of blood coagulation. To date, there is no crystal structure for free FVIIa. X-ray and neutron scattering data in solution for free FVIIa and the complex between FVIIa and soluble tissue factor (sTF) had been obtained for comparison with crystal structures of the FVIIa-sTF complex and of free factor IXa (FIXa). The solution structure of free FVIIa as derived from scattering data is consistent with the extended domain arrangement of FVIIa seen in the crystal structure of its complex with sTF, but is incompatible with the bent, less extended domain conformation seen in the FIXa crystal structure. The FVIIa scattering curve is also compatible with a subset of 317 possible extended structures derived from a constrained automated conformational search of 15 625 FVIIa domain models. Thus, the scattering data support extended domain models for FVIIa free in solution. Similar analyses showed that the solution scattering derived and crystal structures of the FVIIa-sTF complex were in good agreement. An automated constrained search for allowed structures for the complex in solution based on scattering curves showed that only a small family of compact models gave good agreement, namely those in which FVIIa and sTF interact closely over a large surface area. The general utility of this approach for structural analysis of heterodimeric complexes in solution is discussed. Analytical ultracentrifugation data and the modeling of these data were consistent with the scattering results. It is concluded that in solution FVIIa has an extended or elongated domain structure, which allows rapid interaction with sTF over a large surface area to form a high-affinity complex.     

Effects of flavonoids isolated from scutellariae radix on fibrinolytic system induced by trypsin in human umbilical vein endothelial cells

Studies on the effects of flavonoids isolated from the roots of Scutellaria baicalensis on the fibrinolytic system induced by trypsin in cultured human umbilical vein endothelial cells (HUVECs) showed that baicalein (1) strongly inhibited the reduction of t-PA production and the elevation of PAI-1 production induced by trypsin. The IC50 for PAI-1 production was 3.7 microM. In addition, wogonin (3), oroxylin A (5), skullcapflavone II (6), and 2',5,5',7-tetrahydroxy-6',8-dimethoxyflavone (7) inhibited the elevation of PAI-1 induced by trypsin, though less strongly; their IC50 were 105, 61, 110, and 88 microM, respectively. These findings suggest that baicalein prevents the thrombotic tendency induced by trypsin.     

Histamine induces nuclear factor of activated T cell-mediated transcription and cyclosporin A-sensitive interleukin-8 mRNA expression in human umbilical vein endothelial cells

Strong evidence indicates that virions of mammalian reoviruses undergo proteolytic processing by acid-dependent cellular proteinases as an essential step in productive infection. Proteolytic processing takes the form of a series of cleavages of outer-capsid proteins final sigma3 and mu1/mu1C. Previous studies showed an effect of both NH4Cl and E-64 on these cleavages, indicating that one or more of the acid-dependent cysteine proteinases in mammalian cells (cathepsins B and L, for example) is required; however, these studies did not address whether acid-dependent aspartic proteinases in those cells (cathepsin D, for example) may also be required. To determine the role of aspartic proteinases in reovirus entry, studies with pepstatin A, a specific inhibitor of aspartic proteinases, were performed. The results showed that pepstatin A neither blocks nor slows reovirus infection of L or MDCK cells. Experiments using ribonuclease A and other proteins as cleavable substrates showed that cathepsin-D-like proteinases from these cells are inhibited within the tested range of pepstatin A concentrations both in vitro and within living cells. In other experiments, virion-bound final sigma3 protein was shown to be a poor substrate for cleavage by cathepsin D in vitro, consistent with the findings with inhibitors. In sum, the data indicate that cathepsin-D-like aspartic proteinases provide little or no activity toward proteolytic events required for infection of L or MDCK cells with reovirus virions.     

The effect of methylmercury (CH3HgCl) on the production of endothelium-derived relaxing factor (EDRF) by cultured human umbilical vascular endothelial cells based on its anti-aggregatory effect on human platelets

The effect of methylmercury (CH3HgCl) on the production of endothelium-derived relaxing factor (EDRF) by cultured human umbilical vascular endothelial cells (HUVECs) based on its anti-aggregatory effect on human platelets was examined. HUVECs were harvested from umbilical veins by collagenase treatment. The platelet aggregation test was performed with cuvettes lined with HUVECs. Platelet aggregation induced by 0.05 units thrombin/ml was inhibited in the presence of HUVECs. This HUVEC-dependent anti-platelet aggregatory effect was enhanced by the addition of bradykinin (10 nmol/L), which stimulates the production of EDRF. Indomethacin (IND, 1 mumol/L) reduced the HUVEC-dependent anti-platelet aggregatory effect. The effect of NG-monomethyl-L-arginine L-NMMA, 100 mumol/L), an inhibitor of nitric oxide synthase (NOS) in endothelial cells, on HUVECs pretreated with IND showed almost complete platelet aggregation similar to results without HUVECs. The anti-platelet aggregatory effect of HUVECs pretreated with IND seemed to depend mainly on EDRF. Methylmercury (MeHg) (20-50 mumol/L) induced dose-dependent platelet aggregation in cuvettes, without HUVECs. Methylmercury (30 mumol/L) induced less platelet aggregation in the presence of HUVECs than in their absence. The degree of inhibitory effect by HUVECs on MeHg-induced platelet aggregation was reduced dose-dependently (30-50 mumol/L MeHg). Methylmercury-induced platelet aggregation at 50 mumol/L MeHg with or without HUVECs was similar. These findings suggest that this simple new experimental system is useful for assessing the production of EDRF by HUVECs, and show that MeHg inhibits the production of EDRF by HUVECs, which may be involved in the etiology of cardiovascular diseases such as hypertension and arteriosclerosis.     

Induction and release of manganese superoxide dismutase caused by tumor necrosis factor-alpha from mitochondria in human umbilical vein endothelial cells

The effects of tumor necrosis factor-alpha (TNF-alpha) on cultured human umbilical vein endothelial cells (EC) and five cancer cell lines, A549, ME180, A2780, KURAMOCHI, and Hela, were compared. While A549, A2780, KURAMOCHI, and Hela cells were fairly resistant to the cytolytic effects of TNF-alpha, ME180 cells were sensitive. EC were also less sensitive to TNF-alpha than ME180 cells as judged by the viability of individual cells and by the release of lactate dehydrogenase (LDH) into the medium. Manganese superoxide dismutase (Mn-SOD) was markedly induced by these cytokines in EC and in A549 cells but not in ME180 cells. The levels of Mn-SOD in the conditioned medium of EC were dramatically increased after stimulation with cytokines, whereas those in ME180 and A549 cells were relatively low. The amount of Mn-SOD released appears to be comparable to that from cells lysed by other means. Immunoblot analysis of Mn-SOD in the medium showed that the molecular mass of the immunoreactive protein was the same as mitochondrial Mn-SOD, indicating that no proteolysis had occurred. These data suggest that in vivo the TNF-alpha produced by cancer cells may induce Mn-SOD in vascular endothelial cells, resulting in release of a relatively large amount of this protein into the serum.     

The induction of cyclooxygenase-2 by interleukin-4 in human umbilical vein endothelial cells: a possible role in regulation of fetal vascular tone

In immunocytes from the mollusc Mytilus galloprovincialis, the major pathway followed by platelet-derived growth factor (PDGF)-AB and transforming growth factor (TGF)-beta1 in provoking the release of norepinephrine, epinephrine and dopamine into cell-free hemolymph (serum) is mediated by a corticotropin-releasing hormone-adrenocorticotropin hormone (CRH-ACTH) biogenic amine axis. This axis not only annulled the inhibiting properties of PDGF-AB, it even reversed the latter's effect, while the inducing effect of TGF-beta1 was amplified. These findings show that non-classical immune-neuroendocrine molecules, such as PDGF-AB and TGF-beta1, are involved in building stress response, using the same conserved mechanisms present from invertebrates to vertebrates.     

Differential antibiotic-induced endotoxin release and interleukin-6 production by human umbilical vein endothelial cells (HUVECs): amplification of the response by coincubation of HUVECs and blood cells

The activation of human vascular endothelial cells (ECs), as assessed by interleukin (IL)-6 production, was investigated in response to antibiotic-induced lipopolysaccharide release from Escherichia coli. Antibiotic-induced killing of E. coli resulted in the activation of human umbilical vein endothelial cells (HUVECs). Imipenem and meropenem induced faster killing of E. coli than ceftriaxone at 2 and 6 h. However, imipenem-induced bacterial killing resulted in significantly less IL-6 release compared with meropenem or ceftriaxone. When HUVECs were coincubated with diluted (4%) human blood, bacterial killing-induced total IL-6 release was significantly higher than that produced by HUVECs or by 4% blood cells alone. These observations suggest that antibiotic-induced lipopolysaccharide release activates both endothelial cells and monocytes and that the effect is significantly amplified when endothelial cells and monocytes are coincubated. Imipenem, which has a greater affinity for penicillin-binding protein (PBP) 2, induced less IL-6 release from blood and endothelial cells than did PBP 3-specific ceftriaxone and meropenem.     

Ability of reconstituted high density lipoproteins to inhibit cytokine-induced expression of vascular cell adhesion molecule-1 in human umbilical vein endothelial cells

Previous studies have shown that both high density lipoproteins (HDL) isolated from human plasma and reconstituted HDL (rHDL) are effective inhibitors of adhesion molecule expression in human endothelial cells. In this study rHDL have been used to investigate whether HDL particle shape, size, apolipoprotein composition or lipid composition are important determinants of the ability of HDL to inhibit the TNF-alpha induced expression of vascular cell adhesion molecule-1 (VCAM-1) in human umbilical vein endothelial cells (HUVECs). On the basis of these studies it is possible to draw several firm conclusions. i) Neither phospholipid-containing vesicles nor lipid-free apolipoprotein (apo) A-I inhibit VCAM-1 expression in HUVECs. ii) Simple discoidal complexes containing only phospholipid and apoA-I (discoidal (A-I)rHDL) are sufficient to inhibit the TNF-alpha-induced expression of VCAM-1 in HUVECs. iii) Spherical apoA-I-containing rHDL (spherical (A-I)rHDL) are superior to discoidal (A-I)rHDL as inhibitors. iv) The particle size of spherical (A-I)rHDL has no influence on the inhibition. v) Spherical rHDL that contain apoA-I are superior as inhibitors of VCAM-1 to those containing apoA-II when the rHDL preparations are equated for apolipoprotein molarity. However, when compared at equivalent particle molarities, this difference is no longer apparent. vi) Replacement of cholesteryl esters with triglyceride in the core of spherical (A-I)rHDL has no effect on the ability of these particles to inhibit VCAM-1 expression. From these results it is tempting to speculate that variations in inhibitory activity may contribute to the variations observed in the anti-atherogenicity of different HDL subpopulations.-Baker, P. W., K-A. Rye, J. R. Gamble, M. A. Vadas, and P. J. Barter. Ability of reconstituted high density lipoproteins to inhibit cytokine-induced expression of vascular cell adhesion molecule-1 in human umbilical vein endothelial cells.     

Autocrine regulation of endothelial exocytosis: von Willebrand factor release is induced by prostacyclin in cultured endothelial cells

OBJECTIVE: To find out whether recombinant human erythropoietin (r-HuEPO) given perioperatively has any effect on haemostatic activity in patients undergoing elective colorectal resection. DESIGN: A placebo-controlled double-blind study. SETTING: Odense university hospital, Denmark. SUBJECTS: 24 patients undergoing elective colorectal resection, 13 of whom were given r-HuEPO (Eprex) and 11 placebo. MAIN OUTCOME MEASURES: Concentrations of haemoglobin, tissue-type plasminogen activator and plasminogen activator inhibitor-1; activated partial thromboplastin time; prothrombin time; platelet and reticulocyte counts; blood loss; and transfusions. RESULTS: There was no significant change in fibrinolytic activity, prothrombin time, or activated prothrombin time in the treatment group. Platelet counts differed slightly but not significantly, being higher in the r-HuEPO group. There was a significant increase in reticulocyte counts in the r-HuEPO group. CONCLUSION: R-HuEPO given perioperatively significantly increased erythropoiesis in patients undergoing elective colorectal operations but had no influence on haemostatic activity.     

Selective migration of highly differentiated primed T cells, defined by low expression of CD45RB, across human umbilical vein endothelial cells: effects of viral infection on transmigration

Low expression of CD45RB on CD45RO+ T lymphocytes defines a subset of highly differentiated T lymphocytes that accumulate in vivo within the affected joints of patients with rheumatoid arthritis (RA). Although it is known that CD45RO+ T lymphocytes migrate to sites of inflammation in vivo, it is not clear whether within this subset the CD45RBlo cells are selectively recruited or develop in situ within the joint. Using a transwell system we show that a small proportion of resting T lymphocytes migrated across unactivated human umbilical vein endothelial cells (HUVEC). These migrating cells were CD45RO+ and enriched for low CD45RB expression. In addition, both the CD45RO+CD45RBlo subset and migrating cells expressed increased levels of beta 1 and beta 2 integrins and CD44. The percentage of CD45RO+CD45RBlo T lymphocytes was increased in the circulation of patients with acute Epstein-Barr virus (EBV) infection. These in vivo activated cells also expressed increased levels beta 1 and beta 2 integrins and CD44, and showed an enhanced rate of transmigration compared with resting T lymphocytes. Transmigration of T lymphocytes was increased using the chemokines RANTES and lymphotactin and the cytokine interleukin-15 (IL-15). In addition, infection of the HUVEC with cytomegalovirus (CMV) led to an enhanced movement of T lymphocytes. In all of these cases the selective migration of the CD45RBlo subset was maintained. Thus although the rate of T-lymphocyte transmigration could be influenced by a number factors, the CD45RO+CD45RBlo subset has a migratory advantage suggesting that more differentiated CD45RO+CD45RBlo T lymphocytes are selectively recruited to sites of inflammation.     

Dual pathways of tubular morphogenesis of vascular endothelial cells by human glioma cells: vascular endothelial growth factor/basic fibroblast growth factor and interleukin-8

In this study, we examined whether human glioma cells are angiogenic in a model using human microvascular endothelial cells, and also which factor is responsible for the glioma-dependent angiogenesis. Tubular morphogenesis in type I collagen gel by human microvascular endothelial cells was stimulated in the presence of 10 and 100 ng/ml of vascular endothelial growth factor (VEGF), 10 ng/ml basic fibroblast growth factor (bFGF) and 10 ng/ml of interleukin-8 (IL-8). Tube formation of the microvascular endothelial cells was assayed in the glioma cell lines IN157 and IN301, co-cultured using the double chamber method. IN301 cells had much higher levels of VEGF, bFGF and transforming growth factor-beta mRNA than IN157 cells, whereas the two had similar levels of transforming growth factor-alpha mRNA. By contrast, IN157 cells had much higher levels of IL-8 mRNA than IN301 cells. IN301-dependent tubular morphogenesis was inhibited by anti-VEGF or anti-bFGF antibody, and the inhibition was almost complete when anti-VEGF and anti-bFGF antibodies were present. On the other hand, IN157-dependent tubular morphogenesis was inhibited by anti-IL-8 antibody, but not by anti-VEGF or anti-bFGF antibodies. These findings demonstrated dual paracrine controls of tumor angiogenesis by human glioma cells. One is mediated through VEGF and/or bFGF, and the other, through IL-8.     

X-linked vacuolated myopathy: membrane attack complex deposition on the surface membrane of injured muscle fibers is not accompanied by S-protein

We have studied the expression of S-protein on the muscle from patients with X-linked vacuolated myopathy [characterized by the deposition of the complement C5b-9 membrane attack complex (MAC) over abnormal muscle fibers] and controls by immunocytochemistry and immunoblotting. No expression was detected on muscle from controls and patients with X-linked vacuolated myopathy. These findings suggest that S-protein does not render the MAC inactive in X-linked vacuolated myopathy. This situation may be due to the fact that the pathways of MAC activation and the expression of S-protein in X-linked vacuolated myopathy are different from the ones observed in ischemic and/or necrotic, or immune diseases. These results emphasize the role of the membrane complement regulatory proteins (i.e., CD59) in X-linked vacuolated myopathy.     

VCAM-1 is internalized by a clathrin-related pathway in human endothelial cells but its alpha 4 beta 1 integrin counter-receptor remains associated with the plasma membrane in human T lymphocytes

Lymphocyte extravasation involves a step(s) of de-adhesion to allow trans- and subendothelial migration in response to inflammatory signals. We show here that ligated VCAM-1 was rapidly internalized (t1/2 14.5 min) in ECV 304 endothelial cells and in TNF-alpha-primed human umbilical vein-derived endothelial cells (t1/2 11.2 min). The process required energy (ATP), intracellular Ca2+, an intact cytoskeletal network and active protein kinases. The internalization of VCAM-1 involved a clathrin-dependent pathway based on the observations that 1) it was inhibited in cells treated with lysosomotropic agents or with a hypertonic concentration of sucrose, and 2) internalized VCAM-1 colocalized with clathrin. In contrast, the cross-linked alpha 4 beta 1 integrin counter-receptor of VCAM-1 remained associated with the plasma membrane of purified peripheral T and Jurkat cells. Our results suggest a model where VCAM-1 would initially participate in the retention of T cells to the endothelium by binding alpha 4 beta 1 integrin. Lymphocyte de-adhesion would be facilitated as a result of the internalization of VCAM-1. The persistent cell surface expression of alpha 4 beta 1 integrin would allow the migrating T cells to interact with and receive signal(s) from its fibronectin ligand of the extracellular matrix.     

Thrombopoietin and its receptor, c-mpl, are constitutively expressed by mouse liver endothelial cells: evidence of thrombopoietin as a growth factor for liver endothelial cells

Present data suggest that the primary site of thrombopoietin (TPO) mRNA is the liver. Previously, we reported that specific murine liver endothelial cells (LEC-1) located in the hepatic sinusoids support in vitro megakaryocytopoiesis from murine hematopoietic stem cells suggesting that these cells may be a source of TPO. We report here that TPO and its receptor, c-mpl, are coexpressed on cloned LEC-1. Enzyme-linked immunosorbent assay (ELISA), biological assay, and flow cytometry studies confirmed the expression of both TPO and its receptor, respectively, at the protein level. TPO activity was enhanced in supernatants from LEC-1 treated with tumor necrosis factor (TNF)-alpha and gamma-interferon (INF). Our results show that TPO through its receptor stimulated the growth of LEC-1 in vitro. These observations establish LEC-1 as a novel source of TPO in the liver. To our knowledge, this is the first report that liver endothelial cells express both TPO and its receptor, c-mpl, and our findings indicate that this cytokine constitutes a growth factor for liver endothelial cells in vitro.