The search for Fc receptors on human tissues and human endothelial cells in culture

Intravenous immunoglobulins for therapeutic use contain anti-idiotypes against xenophile antibodies and prolong discordant graft survival

Xenotransplantation between discordant species leads to a graft survival of a few minutes, due to binding of natural antibodies to the xenogeneic endothelial cells, complement activation, and endothelial cell activation. Polyclonal human immunoglobulins for intravenous use (IVIg) from normal donors have been proven effective in a variety of antibody-mediated disorders and contain anti-idiotypic antibodies directed against a number of disease-associated and natural antibodies. We have shown that administration of IVIg delays rejection of a guinea pig heart to a rat. We demonstrate herein that IVIg can inhibit the binding of xenoreactive rat IgG antibodies to guinea pig endothelial cells. This inhibition is likely due to the presence, among IVIg, of anti-idiotypic antibodies as F(ab')2 fragments of IVIg were as effective as whole IVIg. In addition, natural anti-endothelium rat antibodies were retained on a column of F(ab')2 fragments of IVIg coupled to Sepharose. The degree of inhibition of binding of IgG natural antibodies correlated with the survival of the xenograft when IVIg was administered prior to transplantation. Thus IVIg prolong xenograft survival through idiotypic-anti-idiotypic interactions with natural xenoreactive antibodies of the IgG isotype.

Differential patterns of reaction of human natural antibodies to pig hepatocytes and vascular endothelium

We have recently conducted a series of experiments to characterize the pattern of reaction of human natural antibodies (NA) with individual pig liver cells. Pooled normal human serum (PHS) was incubated with cultured pig hepatocytes (HEP), aortic endothelial cells (AEC), and portal endothelial cells (PEC), and the reaction of NA to different cell types was measured by antibody-mediated cytotoxic (MTT assay), antibody binding (ELISA), and flow cytometric analysis. The human NA displayed a differential pattern of binding with hepatocytes exhibiting a more limited expression of xenoantigen expression than either aortic or portal endothelial cells. These differences in reaction patterns were also noted for Western blot analysis of individual cell membrane extracts. Preincubation of the pig cells with anti-pig MHC antibodies did not inhibit the binding of human IgM natural antibodies to the pig cells. Comparison of the pattern of NA absorption following the use of bioartificial liver support in patients with acute hepatic failure demonstrated limited ability of pig hepatocytes to absorb substantial amounts of NA. These studies indicate that pig hepatocytes are less vulnerable to NA cytotoxicity than pig vascular endothelial cells and that pig vascular endothelial cells express xenoantigens that are unique and not found on hepatocytes.

The relationship of anti-endothelial cell antibodies to anti-phospholipid antibodies in patients with giant cell arteritis and/or polymyalgia rheumatica

Sera from patients with giant cell arteritis and/or polymyalgia rheumatica were tested for the presence of IgG, IgM and IgA antibody to endothelial cells (AEC), cardiolipin (ACL) and phosphatidylethanolamine (APE) using enzyme-linked immunosorbent assays. There were strong correlations between ACL and APE, but also between AEC and ACL IgM (p < 0.02) and between AEC and APE IgA (p < 0.003). Inhibition of AEC binding was achieved by absorption onto EC, but ACL and APE binding was also significantly reduced. In contrast, the binding of AEC antibody could not be inhibited by incubation with CL. Our data suggest that AEC constitute a heterogeneous population of autoantibodies.

Production of soluble ICAM-1 from human endothelial cells induced by IL-1 beta and TNF-alpha

The present study was designed to establish the effects of cytokines on soluble ICAM-1 (sICAM-1) production by human endothelial cells (EC) and ICAM-1 expression on these cells and the effects of purified sICAM-1 on lymphocyte-EC adhesion. Expression of ICAM-1 and production of sICAM-1 were measured by a specific ELISA method. ICAM-1 expression was enhanced by IL-1 beta, TNF-alpha, and most effectively by IFN-gamma. IL-4, IL-6, M-CSF, or GM-CSF showed no effects on ICAM-1 expression. IL-4 (100 units/ml) or IL-6(100 units/ml) abolished the enhancing effect of IL-1 beta, while TNF-alpha (1, 10, 100 units/ml) synergized with IL-1 beta to promote ICAM-1 expression in EC. In contrast with the transient increase of cell-associated ICAM-1 expression after activation by IL-1 beta, which peaked 40 h poststimulation and declined thereafter, sICAM-1 continued to accumulate in culture supernatants even after 48 h poststimulation in IL-1 beta-stimulated EC. IL-1 beta treatment resulted in an increase in adhesion. sICAM-1, purified from cell-free supernatants obtained after a 48-h culture of EC in IL-1 beta by affinity chromatography using monoclonal ICAM-1 antibody coupled to Sepharose beads, significantly inhibited lymphocyte EC adhesion. Preincubation of lymphocytes with conditioned medium of EC cultured with 100 units/ml IL-1 beta for 48 h, which contained a considerable amount of sICAM-1, resulted in a significant inhibition of lymphocyte adhesion to IL-1 beta-stimulated EC. These results suggest that there is a cumulative increase in sICAM-1 concentration in the vicinity of cytokine-stimulated EC and that this sICAM-1 modulates ICAM-1-mediated cell to cell interaction.

The effects of cytokines on metalloproteinase inhibitors (TIMP) and collagenase production by human chondrocytes and TIMP production by synovial cells and endothelial cells

It has been suggested that IL-1 produces cartilage matrix degradation by metalloproteinases such as collagenase, and that such degradation is regulated by metalloproteinase inhibitors (TIMP). Therefore, the balance between collagenase and TIMP is an important factor for tissue destruction in inflammatory joints. In the present study the effects of cytokines on collagenase and TIMP production in chondrocytes as well as the effects of cytokines on TIMP production in connective tissue cells were studied. IL-1 beta inhibited TIMP production in endothelial cells while enhancing TIMP production in synovial cells and chondrocytes. In addition, tumour necrosis factor-alpha (TNF-alpha) significantly inhibited and IL-6 significantly enhanced TIMP production in endothelial cells, synovial cells and chondrocytes. In the chondrocyte supernatant, collagenase activity/TIMP ratio was significantly elevated by the addition of either IL-1 beta or TNF-alpha to the cells, whereas the ratio was significantly decreased by IL-6. These results suggest that the cytokine effects on TIMP production are different among the different cell types, and that either IL-1 beta or TNF-alpha induce cartilage matrix degradation by disrupting the collagenase/TIMP balance, while, on the other hand, IL-6 protects the tissue through an opposite effect.

Anti-endothelial cell antibodies in inflammatory bowel disease

Antibodies to endothelial cells may be involved in the pathogenesis of vasculitic disorders. In view of recent evidence implicating intestinal vascular injury in the pathogenesis of inflammatory bowel disease, we have sought anti-endothelial cell antibodies in affected patients, examined their relationship to vascular injury, and tested their ability to mediate endothelial cell cytotoxicity in vitro. Anti-endothelial cell antibody levels were elevated in ulcerative colitis (P < 0.0001) and Crohn's disease (P < 0.05) compared with healthy controls. In ulcerative colitis, anti-endothelial cell antibody levels were related to disease activity and correlated with circulating levels of von Willebrand factor (r = 0.58, P < 0.01), a marker of vascular injury. Anti-endothelial cell antibodies, however, were not directly cytotoxic to endothelial cells in vitro. These data indicate, for the first time, an association between anti-endothelial cell antibody levels and vascular injury in vivo and suggest that they may be important in the pathogenesis of inflammatory bowel disease, particularly ulcerative colitis.

Complement enhances the elimination of soluble aggregates of IgG by rat liver endothelial cells in vivo

In the present study, we have investigated the role of complement (C) and possible C receptors present on rat liver endothelial cells (EC) in the clearance and tissue distribution of soluble aggregates of IgG (AIgG). To study the effect of elimination of AIgG by EC in vivo, Kupffer cell (KC)-depleted rats were used, with or without an intact C system (These rats will be referred to throughout this report as EC-rats.) In EC-rats with an intact C system, clearance of AIgG (2000-3000 kDa, 20-27 IgG molecules/aggregate) occurred in a biphasic manner with a first T 1/2 (T1) of 9.4 +/- 2.3 min and a second T 1/2 (T2) of 44.7 +/- 16.0 min. In EC-rats without an intact C system [cobra venom factor (COVF)-treated group], clearance of AIgG was significantly delayed with a T1 of 25.3 +/- 9.9 min (p < 0.005) and a T2 of 124.5 +/- 18.4 min (p < 0.001). There were less degradation products of AIgG in the circulation in EC-rats treated with COVF as compared to EC-rats with an intact C system. Eight minutes after injection, 27.5 +/- 11.6% of the injected AIgG was found in the livers of EC-rats while 15.1 +/- 3.2% was found in the livers of the COVF-treated group. Double immunofluorescence studies indicated that AIgG in the liver was associated with EC in the rats with an intact C system. Clear deposits of C3 and lesser amounts of C1q accompanied the deposition of AIgG. In COVF-treated EC-rats, AIgG together with C1q was also associated with EC but no detectable C3 was seen. These data suggest clearance of AIgG via Fc and C receptors present on EC in vivo.

Lung capillary endothelial cells produce and secrete urokinase-type plasminogen activator

Bovine lung capillary endothelial cells (BLuEC) were isolated, and their ability to produce plasminogen activator (PA) in vitro was demonstrated. BLuEC secreted more than 10 times as much as urokinase-type PA (u-PA) as did bovine aortic, hepatic capillary and adrenal capillary endothelial cells, and lung fibroblasts. BLuEC secreted u-PA on both sides of the cell layer, the luminal surface, and the basic surface attached to the basement membrane. u-PA mRNA was detected in BLuEC by Northern blotting, but not in endothelial cells from other tissues and fibroblasts. These results suggest that BLuEC may contribute not only to the patency of lung vessels but also to the maintenance of alveolar functions through the production and secretion of u-PA.

Kupffer cell depletion in vivo results in preferential elimination of IgG aggregates and immune complexes via specific Fc receptors on rat liver endothelial cells

In the present study we have investigated the clearance kinetics and tissue distribution of monomeric (m) IgG and soluble aggregates of IgG (AIgG) and immune complexes (IC) in normal and Kupffer cell (KC) depleted rats. In normal rats, clearance of mIgG occurred in a biphasic manner with a first half-life (T1/2) (T1) of 36.3 +/- 6.3 min and a second T1/2 (T2) of 168.4 +/- 4.7 min. AIgG composed of 20-27 IgG molecules per aggregate were cleared significantly faster than mIgG with a T1 of 2.5 +/- 0.1 min and a T2 of 32.5 +/- 5.6 min. KC depletion did not have a significant effect on the clearance rate of mIgG (T1: 33.4 +/- 8.9 min; T2; 159.5 +/- 12.5 min), while clearance of AIgG was delayed significantly with T1 4.8 +/- 0.7 min and T2 41.2 +/- 3.2 min. Eight minutes after injection, 77% of AIgG was found in the liver in normal rats while 62% was found in the liver of KC-depleted rats. Double immunofluorescence studies indicated that AIgG in the liver was associated with KC and endothelial cells (EC) in normal rats. In KC-depleted rats, AIgG was strongly associated with EC. A similar staining pattern was observed when IgG-immune IC were administered. The clearance of AIgG in KC-depleted rats was inhibited fully by pre-administration of high concentrations of IgG but not by pretreatment with IgA. asialofetuin (ASFe) or ovalbumin (OVA). Aggregated F(ab')2IgG was cleared with a comparable rate to mIgG from the circulation, again suggesting Fc gamma receptor-mediated elimination of AIgG by EC. There was a reduced degradation of AIgG in rats depleted of KC as compared with normal rats. These data suggest binding and degradation of AIgG by EC in vivo.

Presence of antibodies against endothelial cells in the sera of patients with episodic angioedema and hypereosinophilia

We reported three additional cases of a newly described syndrome called episodic angioedema with hypereosinophilia. In order to investigate its pathophysiological mechanisms, four parameters were concurrently investigated, including blood eosinophil density, serum chemoattractant activity, serum major basic protein (MBP) levels and the presence of anti-endothelial cell antibodies. Distribution of eosinophils through a metrizamide density gradient showed a preferential sedimentation of blood eosinophils in intermediate layers, clearly different from the hypodense cells (low-density layers) identified in a group of seven patients with idiopathic hypereosinophilic syndrome (HES). In two of the three patients with cyclic angioedema, a chemotactic activity towards eosinophils was detected in the serum (30 +/- 6 and 42 +/- 12 eosinophils per high-power field; P less than 0.05 compared with a control group). Serum MBP levels were at 1524, 619 and 1200 pg/ml. All three patients had circulating anti-endothelial cell antibodies, predominantly of the IgG isotype, in contrast to controls (P less than 0.01) or to patients with HES (P less than 0.01). Specificity of the antibody for endothelial cells was demonstrated in the three patients studied by the absence of binding to various blood cells, including monocytes, lymphocytes, eosinophils and platelets. In one case (patient 2), the levels of anti-endothelial cell antibodies, as well as the serum chemoattractant activity to eosinophils varied according to the successive acute phases of the disease. Although further investigations are needed to clarify the exact pathophysiology of this syndrome, and especially the possible participation of the anti-endothelial cell antibodies in the cutaneous lesions, these data suggest that angioedema observed in this syndrome could result from the combined effects of activated eosinophils and of immunologically induced endothelial lesions.

C1q binding to human vascular smooth muscle cells mediates immune complex deposition and superoxide generation

Evidence was obtained for the binding of C1q to the membrane of cultured vascular smooth muscle cells derived from human umbilical cord veins. C1q was fixed to the cell membrane at 4 degrees C, whereas it was ingested into the cytoplasm, as a cytoplasmic inclusion, when tested at 37 degrees C. The addition of C1q in advance inhibited the subsequent binding of C1q. Neither fibronectin nor laminin was detected on the cell membrane. Aggregated IgG bound to vascular smooth muscle cells in the case of preincubation with C1q at 4 degrees C, whereas aggregated IgG did not bind to the cells in the absence of C1q. The addition of C1q molecules to the cells in suspension enhanced superoxide generation by vascular smooth muscle cells. There was no effect of C1q on superoxide generation by the cells in monolayer. These results suggest that C1q binds on the membrane of vascular smooth muscle cells via its specific receptor that mediates immune complex binding to the cells and superoxide generation. These properties elucidate the mechanisms by which circulating immune complexes deposit in the vascular wall, and subsequent degradation of tissue components surrounding vascular smooth muscle cells occurs through oxidative burst of the cells.

A modified method for application of indirect immunofluorescent staining for factor VIII/vWF to capillary endothelia

Immunofluorescent staining for factor VIII/von Willebrand factor (VIII/vWF) as applied to capillary endothelia has sources of false negativity and false positivity. False negativity occurs because capillary endothelia are not as rich as endothelia of larger vessels in this factor. To overcome this problem we enriched liver endothelium by preincubation with the factor. The source of false positivity is the presence of Fc receptors on freshly isolated capillary endothelia. These receptors may bind antibody to this factor through the Fc portion of the molecule. To overcome this problem we preincubated the cells with IgG not against VIII/vWF (irrelevant antibody) to cover Fc receptors. The results presented indicate that considerable improvement in the immunofluorescent staining of capillary endothelia can be achieved by application of our method.

In situ demonstration of Fc gamma-receptors in human chronic marginal and apical periodontitis

The distribution of receptors for the Fc portion of IgG (Fc gamma R) in chronic marginal and apical periodontitis were studied by using a monoclonal antibody against placental Fc gamma R and soluble immune complexes as indicators. Cryostat sections were used in indirect immunofluorescence technique. Fc gamma R were detected on varying numbers of cells in the inflammatory cell infiltrates, on endothelial cells of certain vessels and in fibrous tissue. In chronic marginal periodontitis Fc gamma R were also observed on cells within the oral gingival epithelium (OGE) and the pocket epithelium (PE). There was a distinct fluorescence in stratum spinosum and occasionally in stratum basale of OGE and in the coronal portion of the PE. Fc gamma R on Langerhans cells could not be demonstrated. In apical periodontitis Fc gamma R were also detected on cells within the epithelium. In some cases epithelium in periapical cysts was positive. Soluble immune complexes bound to morphologically similar, but fewer cells compared to the monoclonal antibody against Fc gamma R. The results indicate that Fc gamma R are generally expressed on cells in inflamed tissue. Thus, this presence of Fc gamma R on certain specialized cells such as endothelial cells and keratinocytes, may endow these cells with functions previously thought to be restricted to cells of the lymphoreticular system.

Phagocytosis of Staphylococcus aureus by cultured bovine aortic endothelial cells: model for postadherence events in endovascular infections

We examined the interaction of Staphylococcus aureus with cultured bovine aortic endothelial cells as a model for the initial events in the pathogenesis of endovascular infections. Confluent monolayers of cultured endothelial cells were incubated with S. aureus. Cell-associated bacteria were measured by washing away nonadherent organisms, disrupting the monolayers, and performing quantitative cultures. Phagocytosis was differentiated from adherence by treating the cells with lysostaphin; approximately 60% of cell-associated bacteria was found to be intracellular. Phagocytosis could be blocked by using cytochalasin B, which interferes with microfilament function. Addition of fibronectin resulted in a 63% increase in adherence of S. aureus to the endothelial cells but did not increase ingestion. Transmission electron microscopy demonstrated a sequence of events similar to that which occurs during ingestion by professional phagocytes, including: adherence of bacteria to the endothelial cell; formation and elongation of surface extensions of the endothelial cell to surround the adherent bacteria; and complete enclosure within apparent phagosomes. Phagocytosis of bacteria by endothelial cells, followed by intracellular persistence, may be an important postadherence event in the pathogenesis and pathophysiology of endovascular infections.

Serum alkaline ribonuclease derived from vascular endothelial cells is raised in patients with rheumatoid vasculitis

We investigated the mechanisms of the marked increase of alkaline ribonuclease (RNase) in the sera of patients with rheumatoid arthritis associated with vasculitis. Among various tissues examined, blood vessels contained higher levels of alkaline RNase than acid RNase. Cultured human endothelial cells contained significantly higher amounts of alkaline RNase than acid RNase. In contrast, acid RNase was predominant in most other tissues and cells. Endothelial cells cultured with sera from patients with vasculitis released alkaline RNase into the extracellular medium. The phosphocellulose chromatographic profile of these sera differed from that of sera from healthy subjects. These results imply that the alkaline RNase in sera of patients with vasculitis is derived from blood vessels, probably from endothelial cells.

Immunohistochemical demonstration of FC receptors in rat tissues using immune complexes as ligand

To demonstrate the presence and localization of Fc receptors, rat liver cryostat sections were incubated with heterologous and autologous immune complexes (ICx) and immunoglobulin (Ig) aggregates. Binding was demonstrated using the immunoperoxidase technique. Autologous and heterologous ICx as well as aggregates from human and rat Ig appeared to bind to the sinusoidal wall. ICx bind in preference to aggregates. Monomeric Ig and aggregated Ig from swine and rabbit did not bind. The results demonstrated that ICx and rat and human Ig aggregates were bound via an Fc receptor. This Fc receptor was still intact in livers from carbontetra chloride and galactosamine treated rats. The receptor could also be demonstrated on spleen macrophages and on kidney interstitial cells. This method turned out to be an useful functional histochemical method to localize Fc receptors and to demonstrate their affinity and species specificity in tissues.

Studies on the mechanism of binding and uptake of immune complexes by various cell types of rat liver in vivo

Soluble heterologous immune complexes (IC) were used to study the mechanism of IC binding to rat liver in vivo. Binding of IC to the various cell types of the liver, endothelial cells, hepatocytes, and Kupffer cells, was only inhibited by aggregated swine immunoglobulins. Binding was not inhibited by the absence of complement components. Intravenous injection of asialoglycoproteins, to block the galactose receptor, could not prevent IC binding. We conclude that Fc receptors play an important role in the binding of soluble heterologous IC to hepatocytes, endothelial cells, and Kupffer cells.

Human vascular smooth muscle cells and endothelial cells lack catalase activity and are susceptible to hydrogen peroxide

51Cr release as lytic and cell detachment as nonlytic injury were employed to estimate neutrophil-mediated injury of cultured human vascular smooth muscle cells and endothelial cells. The reagents hydrogen peroxide or hypoxanthine-xanthine oxidase produced dose-dependent killing and nonlytic cell detachment, which were specifically inhibited by catalase but not by superoxide dismutase. The concentration of hydrogen peroxide or xanthine oxidase to induce cell detachment was less than lytic dose, suggesting that cell detachment was a much more sensitive assay of injury. Neutrophil-mediated cell lysis averaged 15% at most and was mostly dependent on hydrogen peroxide, while neutrophil-mediated cell detachment was nearly 100% and its dependency on hydrogen peroxide varied from 46% to 60%. These results suggest that vascular smooth muscle cells and endothelial cells in neutrophil-mediated events are destroyed by a hydrogen peroxide-dependent process, mainly via a nonlytic cell detachment mechanism. There was no striking difference of sensitivity to hydrogen peroxide between vascular smooth muscle cells and endothelial cells. Vascular smooth muscle cells and endothelial cells contained fairly high concentrations of superoxide dismutase, but not catalase, activity. The sensitivity of these cells to hydrogen peroxide but not to superoxide may arise from the fact that these cells lack intracellular catalase activity. The injury of vascular cells, which constitute important components of blood vessels, may lead to vascular injury and subsequent tissue damage.

Fc receptor--more answers, more questions

Fc receptors, belonging to the most important surface structures of a number of cells participating in the immune processes, have been intensely studied during the past decade. The present review summarizes the contemporary knowledge of the specificity and heterogeneity of Fc receptors and of factors influencing their expression, and includes some views on their function. In addition, it mentions their relationship to other cell surface structures, expression of Fc receptors during ontogeny of the organism and in certain diseases. Finally, data concerning the isolation and biochemical characterization of the Fc receptor molecule are presented.

Sinusoidal endothelial cells from normal guinea pig liver: isolation, culture and characterization

Guinea pig nonparenchymal hepatic cells were isolated by enzymatic digestion and subsequent separation on a 17.5% metrizamide gradient. Endothelial cell and Kupffer cell-enriched fractions were separated by centrifugal elutriation. Viability of both cell fractions was approximately 80%. Endothelial cells were cultured on a substratum of guinea pig liver collagen and 1% gelatin (1:1). Freshly isolated and cultured sinusoidal endothelial cells contained Factor VIII R:antigen, angiotensin I converting enzyme activity, and they synthesized prostaglandins characteristic of other endothelial cells. Sieve plates were identified in both freshly isolated and cultured cells. Fresh endothelial cells and Kupffer cells formed Fc receptor-mediated rosettes with IgG-opsonized sheep red blood cells, but cultured endothelial cells did not. Only Kupffer cells demonstrated Fc and C3 receptor-mediated phagocytosis. These methods for isolating and culturing sinusoidal endothelial cells should permit further functional assessment of endothelial cells and their interrelationship with other sinusoidal lining cells.

Presence of complement-fixing anti-endothelial cell antibodies in systemic lupus erythematosus

Vasculitis in systemic lupus erythematosus (SLE) is associated with the deposition of IgG and complement in blood vessel walls. However, it is not known whether immune injury to endothelial cells is a part of this process. Therefore, we used a solid phase radioimmunoassay to study the ability of IgG from normal human sera and sera from patients with SLE to bind to endothelial cells. In this assay, cultured human umbilical venous endothelial cells were sequentially incubated with normal or SLE sera, goat anti-human IgG, and 125I-labeled staphylococcal protein A (*SPA). After exposure to normal sera, 2.5 +/- 0.5% (mean +/- SD) of the added *SPA bound to the cells, whereas after exposure to SLE sera 13.8 +/- 7.6% of the added *SPA bound to these cells. This difference in binding was highly significant (P less than 0.001). Binding was partially reduced when SLE sera were preincubated with B-lymphocytes or monocytes, but not after exposure to erythrocytes, platelets, or T lymphocytes. Incubation of endothelial cells with the 7S fraction of SLE sera or with the F(ab')2 fragment of SLE-IgG resulted in the deposition of greater than 80% as much IgG as was deposited on endothelial cells by whole serum. However, since higher molecular weight fractions (greater than 7S) of SLE sera were also active, we tested the capacity of endothelial cells to bind IgG complexes. Endothelial cells bound heat-aggregated IgG (HA-IgG) in a saturable manner at one log concentration below the binding of normal monomeric IgG. Binding of HA-IgG to endothelial cells was markedly enhanced by preincubation with a serum source of complement. Both HA-IgG and SLE-IgG also bound to freshly obtained endothelial cells in suspension, as detected by automated fluorescence flow cytometry. Binding of SLE-IgG and HA-IgG to endothelium initiated complement activation, deposition of the third component of complement, and disruption of the monolayer. In addition, SLE-IgG and HA-IgG caused endothelial cells to secrete prostacyclin and caused the adherence of platelets, confirmed by scanning electron microscopy. These studies demonstrate that IgG anti-endothelial antibodies are present in the sera of patients with active SLE. These sera may also contain IgG complexes that are capable of binding to endothelial cells. The association of IgG and complement with endothelial cells may initiate vascular injury in SLE and other human disorders.

Cellular cooperation in endothelial cell thromboplastin synthesis

Endothelial cells from human umbilical veins produce a procoagulant identified as thromboplastin (tissue factor, factor III) when stimulated with the phorbol ester 12-0-tetradecanoyl-phorbol-13-acetate (TPA), phytohaemagglutinin (PHA) or endotoxin, Inducible thromboplastin synthesis (i.e. synthesis of the protein component of thromboplastin, apoprotein III) was totally inhibited by cycloheximide and actinomycin D, indicating that de novo protein and RNA syntheses are necessary. Serum enhanced the induced apoprotein synthesis. Of the total thromboplastin activity in homogenates of stimulated endothelial cells, about 50--70% was available on the cell surface for interaction with other coagulation factors, inactivation by trypsin and neutralization with antiserum against apoprotein III. Induced synthesis of thromboplastin in endothelial cells was 2--7-fold enhanced by the presence of several other cell types in optimal ratio 4--10 cells per endothelial cell. Some of these cell types were themselves thromboplstin producers (U-937, U-937-4), some were not inducible (lymphocytes, granulocytes and the lymphoblast lines Daudi and Molt 4). This enhancing effect was also seen with cell-free culture supernatants, but these were generally somewhat less effective than the intact cells. Supernatants derived from cells cultured in the presence of TPA, PHA or endotoxin were in most cases more effective than supernatants from unstimulated cells.

Immune thrombocytopenic purpura and pregnancy

Neonatal thrombocytopenia is a potentially life-threatening complication of immune thrombocytopenic purpura (ITP). We followed 23 pregnant women who had either a history of ITP (11 women) or clinically active disease (12 women) to delineate the factors responsible for neonatal thrombocytopenia. No relation was observed between maternal and neonatal platelet counts (P greater than 0.5). Eleven women delivered thrombocytopenic children; antiplatelet antibodies were detectable in each mother, including five who were in clinical remission at delivery. The level of platelet-associated IgG in the mothers did not identify the neonates at risk for thrombocytopenia (P greater than 0.05). However, the level of maternal circulating antiplatelet antibody correlated with both the presence and the extent of neonatal thrombocytopenia (P less than 0.005). A discrepancy between maternal platelet count and maternal antibody level may be especially notable in mothers treated with steroids or splenectomy. Monitoring the level of circulating antiplatelet antibody may help in identifying and managing pregnant women with ITP at risk of delivering neonates with serious thrombocytopenia.