Search

Histamine, leukotriene C4, IL-4, and IL-13 are major mediators of allergy and asthma. They are all formed by basophils and are released in particularly large quantities after stimulation with IL-3. Here we show that supernatants of activated mast cells or IL-3 qualitatively change the makeup of granules of human basophils by inducing de novo synthesis of granzyme B (GzmB), without induction of other granule proteins expressed by cytotoxic lymphocytes (granzyme A, perforin). This bioactivity of IL-3 is not shared by other cytokines known to regulate the function of basophils or lymphocytes. The IL-3 effect is restricted to basophil granulocytes as no constitutive or inducible expression of GzmB is detected in eosinophils or neutrophils. GzmB is induced within 6 to 24 hours, sorted into the granule compartment, and released by exocytosis upon IgE-dependent and -independent activation. In vitro, there is a close parallelism between GzmB, IL-13, and leukotriene C4 production. In vivo, granzyme B, but not the lymphoid granule marker granzyme A, is released 18 hours after allergen challenge of asthmatic patients in strong correlation with interleukin-13. Our study demonstrates an unexpected plasticity of the granule composition of mature basophils and suggests a role of granzyme B as a novel mediator of allergic diseases.

We tested the hypothesis that the ratio between the activating and inhibitory Fcgamma receptor type II (FcgammaRII) in neutrophils determines their responsiveness to immune complexes. We measured mRNA levels of FcgammaRII isoforms and observed differences in the ratio of FcgammaRIIa to FcgammaRIIb2 mRNA in granulocytes of 50 white and 10 black healthy volunteers, and found 4 discrete groups of ratios (ie, 4:1; 3:1, 2:1, or 1:1). The response to either dimeric IgG or aggregated IgG (aIgG) was assessed. Up-regulation of CD11b on the surface as well as the elastase release was significantly more pronounced in neutrophils with a high FcgammaRIIa/FcgammaRIIb2 mRNA ratio of 4:1 compared with a 2:1 or 1:1 ratio. Individual ratios as well as the functional responsiveness of neutrophils were constant over time, as was tested over 12 months. Neutrophil stimulation with various agents in vitro did not alter the FcgammaRIIa/FcgammaRIIb2 mRNA ratio in the neutrophils of these donors, in clear contrast to the findings in their mononuclear cells. We found a strong association between the 2B.4 haplotype of the FCGR2B promoter with increased transcriptional activity in individuals with 1:1 ratios and the more common low-expression 2B.1 haplotype in individuals with FcgammaRIIa/FcgammaRIIb2 mRNA ratios of 2:1, 3:1, or 4:1.

Despite the rapid and widespread integration of chimeric CD20 monoclonal antibody (mAb), rituximab, into the management of non-Hodgkin lymphoma, its efficacy remains variable and often modest when used as a single agent. To develop more potent reagents, human immunoglobulin transgenic mice were used to generate a panel of immunoglobulin G1kappa (IgG1kappa) CD20 mAbs. All reagents bound strongly to CD20(+) cells and recruited mononuclear cells for the lysis of malignant B cells. However, 2 mAbs, 2F2 and 7D8, were exceptionally active in complement-dependent cytotoxicity (CDC), being able to lyse a range of rituximab-resistant targets, such as CD20-low chronic lymphocytic leukemia (CLL), in the presence of human plasma or unfractionated blood. Further analysis showed that 2F2 and 7D8, like rituximab, redistributed CD20 into Triton X-100-insoluble regions of the plasma membrane, but that they had markedly slower off-rates. To determine whether off-rate influenced CDC, a non-complement activating F(ab')(2) antihuman kappa reagent was used. This reagent markedly slowed the off-rate of rituximab and increased its CDC activity to that of 2F2 and 7D8. Thus, with increasing evidence that mAb therapeutic activity in vivo depends on complement activation, these new CD20 reagents with their slow off-rates and increased potency in CDC hold considerable promise for improved clinical activity.

SERPINB6 (PI6) is a member of the intracellular serine protease inhibitors (serpins). Previous studies showed that SERPINB6 is localized mainly in the cytoplasm of endothelial cells, some epithelial cells, monocytes, and neutrophils. In these cells SERPINB6 is thought to prevent cellular damage by scavenging leaking lysosomal proteases. We show here, using novel, well-defined monoclonal antibodies, that SERPINB6 is abundantly expressed by mast cells in all organs and by the human mast cell line HMC-1. Gel filtration experiments revealed that the latter cells contain a high-molecular-weight form of SERPINB6, which consists of sodium dodecyl sulfate (SDS)-stable complexes of this inhibitor with monomeric beta-tryptase. Expression of SERPINB6 by mast cells was compared with those of tryptase and CD117 (c-kit) in biopsies from patients with different forms of mast cell disease. In all cases the lesional mast cells expressed SERPINB6, and, in diffuse cutaneous mastocytosis and mastocytoma, SERPINB6 was expressed by a substantially higher number of mast cells when compared with tryptase. In conclusion, SERPINB6 is abundantly expressed by normal mast cells and by mast cells in mastocytoma lesions. We suggest that in mast cells, SERPINB6 serves to regulate the activity of endogenous beta-tryptase in the cytoplasm.

P38 mitogen-activated protein kinase (MAPK) is an important component of intracellular signaling cascades that initiate various inflammatory cellular responses. To determine the role of p38 MAPK in the procoagulant response to lipopolysaccharide (LPS), 24 healthy subjects were exposed to an intravenous dose of LPS (4 ng/kg), preceded 3 hours earlier by orally administered 600 or 50 mg BIRB 796 BS (a specific p38 MAPK inhibitor), or placebo. The 600-mg dose of BIRB 796 BS strongly inhibited LPS-induced coagulation activation, as measured by plasma concentrations of the prothrombin fragment F1 + 2. BIRB 796 BS also dose dependently attenuated the activation and subsequent inhibition of the fibrinolytic system (plasma tissue-type plasminogen activator, plasmin-alpha2-antiplasmin complexes, and plasminogen activator inhibitor type 1) and endothelial cell activation (plasma soluble E-selectin and von Willebrand factor). Activation of p38 MAPK plays an important role in the procoagulant and endothelial cell response after in vivo exposure to LPS.

In vitro studies suggest that resistance to chemotherapy-induced apoptosis might explain poor response to therapy in fatal cases. Actual execution of apoptosis depends on proper functioning of effector caspases, particularly caspase 3, and on the expression levels of apoptosis-regulating proteins, including Bcl-2 and the recently identified granzyme B- specific protease inhibitor 9 (PI9). Thus, high levels of caspase 3 activation should reflect proper functioning of the apoptosis pathways, resulting in chemotherapy-sensitive neoplastic cells and a favorable prognosis. We tested this hypothesis by quantifying numbers of tumor cells positive for active caspase 3, Bcl-2, and PI9, respectively, in pretreatment biopsies of systemic anaplastic large cell lymphoma (ALCL) patients and by comparing these numbers with clinical outcome. Activation of caspase 3 in more than 5% of the tumor cells was strongly correlated with a highly favorable outcome. High numbers of Bcl-2- and PI9-positive tumor cells were found to predict unfavorable prognosis. This prognostic effect was strongly related to anaplastic lymphoma kinase (ALK) status: ALK-positive ALCL had significantly higher levels of active caspase 3, while high expression of the antiapoptotic proteins Bcl-2 and PI9 was almost completely restricted to ALK-negative cases. In conclusion, high numbers of active caspase 3-positive tumor cells predict a highly favorable prognosis in systemic ALCL patients. Poor prognosis is strongly related to high numbers of Bcl-2- and PI9-positive neoplastic cells. These data support the notion that a favorable response to chemotherapy depends on an intact apoptosis cascade. Moreover, these data indicate that differences in prognosis between ALK-positive and ALK-negative ALCL might be explained by differences in expression of apoptosis-inhibiting proteins.

In tumor cells, the serine protease granzyme B is the primary mediator of apoptosis induced by cytotoxic T lymphocytes (CTLs)/natural killer (NK) cells. The human intracellular serpin proteinase inhibitor 9 (PI9) is the only known human protein able to inhibit the proteolytic activity of granzyme B. When present in the cytoplasm of T lymphocytes, PI9 is thought to protect CTLs against apoptosis induced by their own misdirected granzyme B. Based on the speculation that tumors may also express PI9 to escape CTL/NK cell surveillance, immunohistochemical studies on the expression of PI9 in various lymphomas were performed. Ninety-two cases of T-cell non-Hodgkin lymphoma (NHL), 75 cases of B-cell NHL, and 57 cases of Hodgkin lymphomas were stained with a PI9-specific monoclonal antibody. In T-cell NHL, highest PI9 expression was found in the extranodal T-cell NHL. In nearly 90% of enteropathy-type T-cell NHLs and 80% of NK/T-cell, nasal-type lymphomas, the majority of the tumor cells expressed PI9. In nodal T-anaplastic large cell lymphomas and peripheral T-cell lymphomas (not otherwise specified), PI9 expression occurred less frequently. In B-cell NHL, PI9 expression was associated with high-grade malignancy; 43% of diffuse large B-cell lymphomas showed PI9(+) tumor cells. Finally, PI9 expression was also found in 10% of Hodgkin lymphomas. This is the first report describing the expression of the granzyme B inhibitor PI9 in human neoplastic cells in vivo. Expression of this inhibitor is yet another mechanism used by tumor cells to escape their elimination by cytotoxic lymphocytes.

Recently, it has been postulated that the beneficial effect of intravenous immunoglobulins (IVIGs) in antibody-mediated autoimmune disorders is based on accelerated catabolism of autoantibodies. In the current study, in vivo experiments were performed with mice in which autoantibody production was mimicked by continuous infusion of monoclonal antibodies. In this model, a single dose of IVIG reduced the plasma concentrations of the infused immunoglobulin (Ig)G1 monoclonal antibody (mAb) by approximately 40% after 3 days, whereas the concentration of an IgA mAb was not affected. To extrapolate these findings to humans, a computational model for IgG clearance was established that accurately predicted the time course and magnitude of the decrease in IgG plasma levels observed in mice. Adapted for humans, this model predicted a gradually occurring decrease in autoantibody levels after IVIG administration (2 g/kg), with a maximum reduction of approximately 25% after 3 to 4 weeks and a continued decrease of several months. In conclusion, a single high dose of IVIG induces a relatively small but long-lasting reduction of autoantibody levels by accelerated IgG clearance. This mechanism has clinical relevance in the sense that it can fully explain, as the sole mechanism, the gradual decrease in autoantibody levels observed in several patient studies. However, in some clinical studies, larger or more rapid effects have been observed that cannot be explained by accelerated clearance. Hence, IVIG can also reduce autoantibody levels through mechanisms such as down-regulation of antibody production or neutralization by anti-idiotypic antibodies.

The clinical benefit of intravenous immunoglobulin (IVIG) preparations in the treatment of immune thrombocytopenic purpura (ITP) is supposed to be mediated by blockade of Fc gamma receptor--bearing phagocytes. In 2 experimental models for ITP, it is shown that the therapeutic efficacy of IVIG preparations is related to the IgG dimer content present in these preparations. A rat monoclonal antibody (mAb; MWReg30) directed to the murine platelet-specific integrin alpha(IIb)beta(3) (gpIIb/IIIa) was administered intraperitoneally either as bolus injection or continuous infusion. With bolus injection, the circulating platelet count dropped to almost zero within 3 hours. Pretreatment with cobra venom factor did not affect platelet depletion, whereas pretreatment with anti-Fc gamma RII/III mAb 2.4G2 or IVIG greatly reduced platelet clearance. With continuous infusion, platelet numbers reached a steady state after 4 days, at approximately 25% of control. This reduction in platelets was, however, not observed in mice deficient for the FcR gamma-chain, lacking Fc gamma RI, Fc gamma RIII, and Fc gamma RIII(-/-) mice. Infusion of a single dose of IVIG with a high IgG dimer content on the 4th day--ie, mimicking therapeutic administration--resulted in a platelet increase for several days. IVIG predominantly consisting of monomeric IgG had no effect on platelet numbers. In conclusion, continuous infusion of MWReg30 induces thrombocytopenia in mice by enhancing Fc gamma receptor--mediated clearance of platelets. In this model, it is shown that IgG dimers present in IVIG preparations are responsible for the increase in platelet counts. (Blood. 2001;98:1095-1099)

Previously, we observed in a rat model that intravenous administration of intramuscular immunoglobulin preparations induced a long-lasting hypotension, which appeared to be associated with the presence of IgG polymers and dimers in the preparations, but unrelated to complement activation. We found evidence that this hypotensive response is mediated by platelet-activating factor (PAF) produced by macrophages. In this study, we compared the vasoactive effects of 16 intravenous immunoglobulin (IVIG) products from 10 different manufacturers, in anesthetized rats. Eight of the IVIG preparations showed no hypotensive effects (less than 15% decrease), whereas the other 8 had relatively strong effects (15%-50% decrease). The hypotensive effects correlated with the IgG dimer content of the preparations. Pretreatment of the rats with recombinant PAF acetylhydrolase completely prevented the hypotensive reaction on IVIG infusion, and administration after the onset of hypotension resulted in normalization of the blood pressure. We also observed PAF production on in vitro incubation of human neutrophils with IVIG, which could be blocked by anti-Fcgamma receptor antibodies. This indicates that induction of PAF generation may also occur in a human system. Our findings support the hypothesis that the clinical side effects of IVIG in patients may be caused by macrophage and neutrophil activation through interaction of IgG dimers with Fcgamma receptors. Because phagocyte activation may also lead to the release of other inflammatory mediators, recombinant PAF acetylhydrolase (rPAF-AH) provides a useful tool to determine whether PAF plays a role in the clinical side effects of IVIG. If so, rPAF-AH can be used for the treatment of those adverse reactions. (Blood. 2000;95:1856-1861)

Inhibition of the tissue factor pathway has been shown to attenuate the activation of coagulation and to prevent death in a gram-negative bacteremia primate model of sepsis. It has been suggested that tissue factor influences inflammatory cascades other than the coagulation system. The authors sought to determine the effects of 2 different doses of recombinant tissue factor pathway inhibitor (TFPI) on endotoxin-induced coagulant, fibrinolytic, and cytokine responses in healthy humans. Two groups, each consisting of 8 healthy men, were studied in a double-blind, randomized, placebo-controlled crossover study. Subjects were studied on 2 different occasions. They received a bolus intravenous injection of 4 ng/kg endotoxin, which was followed by a 6-hour continuous infusion of TFPI or placebo. Eight subjects received 0.05 mg/kg per hour TFPI after a bolus of 0.0125 mg/kg (low-dose group), and 8 subjects received 0.2 mg/kg per hour after a bolus of 0.05 mg/kg (high-dose group). Endotoxin injection induced the activation of coagulation, the activation and subsequent inhibition of fibrinolysis, and the release of proinflammatory and antiinflammatory cytokines. TFPI infusion induced a dose-dependent attenuation of thrombin generation, as measured by plasma F1 + 2 and thrombin-antithrombin complexes, with a complete blockade of coagulation activation after high-dose TFPI. Endotoxin-induced changes in the fibrinolytic system and cytokine levels were not altered by either low-dose or high-dose TFPI. The authors concluded that TFPI effectively and dose-dependently attenuates the endotoxin-induced coagulation activation in humans without influencing the fibrinolytic and cytokine response. (Blood. 2000;95:1124-1129)

Plasma-derived antithrombin III (ATIII) prevents the lethal effects of Escherichia coli infusion in baboons, but the mechanisms behind this effect are not clear. In the present study, we evaluated the effects of recombinant human ATIII (rhATIII) on the clinical course and the inflammatory cytokine and coagulation responses in baboons challenged with lethal dose of E coli. Animals in the treatment group (n = 5) received high doses of rhATIII starting 1 hour before an E coli challenge. Those in the control group were administered saline. Survival was significantly improved in the treatment group (P =.002). Both groups had similar hemodynamic responses to E coli challenge but different coagulation and inflammatory responses. The rhATIII group had an accelerated increase of thrombin-ATIII complexes and significantly less fibrinogen consumption compared to controls. In addition, the rhATIII group had much less severe thrombotic pathology on autopsy and virtually no fibrinolytic response to E coli challenge. Furthermore, the rhATIII group had a significantly attenuated inflammatory response as evidenced by marked reduction of the release of various cytokines. We conclude that the early administration of high doses of rhATIII improves the outcome in baboons lethally challenged with E coli, probably due to the combined anticoagulation and anti-inflammatory effects of this therapy. (Blood. 2000;95:1117-1123)

Granzyme A (GrA) and B (GrB) together with perforin are the main constituents of cytotoxic granules of cytotoxic T lymphocytes (CTLs) and natural killer (NK) cells. The cytotoxic proteins are released to deliver a lethal hit during contact between the CTL or NK cell and target cell. With the use of an enzyme-linked immunosorbent assay for antigenic levels, we showed in a recent study that plasma of patients with activated CTLs and NK cells contain elevated levels of extracellular GrA. In this study, we determined the form and proteolytic capacity of this extracellular GrA detected in plasma. With the use of various assays, we show that part of the extracellular GrA circulates in the mature conformation and is bound to proteoglycans that protect it against inactivation by protease inhibitors, such as antithrombin III and alpha-2-macroglobulin, whereas another part of GrA circulates as a complex with antithrombin III. Finally, with the use of a novel assay for active GrA, we demonstrate that some plasma samples with high levels of extracellular GrA contain active GrA. These results suggest that various forms of extracellular GrA occur in vivo and that the regulation of GrA activity may be modified by proteoglycans. These data support the notion that granzymes may exert extracellular functions distant from the site of CTL or NK cell interaction with their target cells. (Blood. 2000;95:1465-1472)

Patients with meningococcal sepsis generally suffer from disseminated intravascular coagulation (DIC). The aim of this study was to address whether these patients have elevated numbers of circulating microparticles that contribute to the development of DIC. Plasma samples from 5 survivors, 2 nonsurvivors, and 5 healthy volunteers were analyzed for the presence of microparticles by flow cytometry. Ongoing coagulation activation in vivo was quantified by enzyme-linked immunosorbent assay of plasma prothrombin fragment F(1 + 2), and procoagulant properties of microparticles in vitro were estimated by thrombin-generation assay. On admission, all patients had increased numbers of microparticles originating from platelets or granulocytes when compared with controls (P =.004 and P =.008, respectively). Patients had elevated levels of F(1 + 2) (P =.004), and their microparticles supported thrombin generation more strongly in vitro (P =.003) than those of controls. Plasma from the patient with the most fulminant disease course and severe DIC contained microparticles that expressed both CD14 and tissue factor, and these microparticles demonstrated extreme thrombin generation in vitro. We conclude that patients with meningococcal sepsis have elevated numbers of circulating microparticles that are procoagulant. These findings may suggest a novel therapeutic approach to combat clinical conditions with excessive coagulation activation.

Contact system activation, in vitro, is triggered by activation of factor XII (FXII) on binding to an activator, such as negatively charged surfaces. A putative surface-binding site of FXII has been located within the amino acid residues 1-28 by identifying the epitope recognized by a monoclonal antibody (MoAb), B7C9, which inhibits kaolin-induced clotting activity. To further elucidate the role of the amino terminal binding site in the regulation of FXII activation, we have characterized a FXII recombinant protein (rFXII-triangle up19) deleted of the amino acid residues 3-19, which are encoded by the second exon of FXII gene. A plasmid encoding for rFXII-triangle up19 was constructed and expressed in HepG2 cells by using vaccinia virus. Purified rFXII-triangle up19 migrated as a single band of Mr 77,000 on sodium dodecyl sulfate (SDS)-polyacrylamide gel, did not bind to MoAb B7C9 immobilized on Protein A-Sepharose, thus confirming that it lacked the epitope for this MoAb, and had no amidolytic activity towards the chromogenic substrate S-2302 in the absence of activator. rFXII-triangle up19 specific clotting activity was lower (44%) than that of native FXII. The activation rate of rFXII-triangle up19 by kallikrein in the absence of dextran sulfate was about four times higher than that of full-length FXII and was increased in the presence of dextran sulfate. However, rFXII-triangle up19 underwent autoactivation in the presence of dextran sulfate. Labeled rFXII-triangle up19 bound to kaolin, which binding was equally well inhibited by either, rFXII-triangle up19 or full-length FXII (IC50 = 7.2 +/- 2.2 nmol/L for both proteins). Accordingly, a synthetic peptide corresponding to FXII amino acid residues 3-19 did not inhibit the binding of labeled full-length FXII to kaolin. rFXII-triangle up19 generated a similar amount of FXIIa- and kallikrein-C1-inhibitor complexes in FXII-deficient plasma in the presence of kaolin, as did full-length FXII; but generated less factor XIa-C1-inhibitor complexes (50%) than full-length FXII. This impaired factor XI activation by rFXII-triangle up19a was also observed in a purified system and was independent of the presence of high molecular weight kininogen. Furthermore, the synthetic peptide 3-19, preincubated with factor XI, inhibited up to 30% activation of factor XI both in the purified system as well as in plasma. These results together indicate that amino acid residues 3-19 of FXII are involved in the activation of factor XI and do not contribute to the binding of FXII to negatively charged surfaces.

Evidence of factor XI (FXI) activation in vivo is scarce. In addition, it remains uncertain whether thrombin, factor XIIa (FXIIa), or perhaps another protease is responsible for FXI conversion. We investigated the activation of FXI in eight healthy volunteers after infusion of a low dose of endotoxin (4 ng/kg of body weight). Activation of prekallikrein FXII, FXI, and prothrombin was measured with sensitive enzyme-linked immunosorbent assays (ELISAs), and FXI activation was measured with a novel enzyme capture assay that detects noncomplexed FXIa. Activation of FXI was apparent with a significant plasma peak level of noncomplexed FXIa of 10 to 11 pmol/L at 1 and 2 hours after endotoxin infusion, followed by a gradual increase in FXIa-FXIa inhibitor complexes, measured in the ELISAs, with a summit of 11 to 15 pmol/L at 6 and 24 hours, respectively. In accordance with previous studies, thrombin generation was detected 1 hour after endotoxin infusion to become maximal after 3 to 4 hours. In contrast, we did not find any evidence of contact activation, because markers of activation of prekallikrein and FXII remained undetectable. From the FXIa data a theoretical model was constructed which suggested that inhibition of FXIa does not take place in the plasma compartment, but is localized on a surface. These data provide the first evidence for FXI activation in low-grade endotoxemia and suggest that FXI is activated independently of FXII., (Copyright 1998 by The American Society of Hematology)

Human eosinophils perform several functions dependent on phospholipase A2 (PLA2) activity, most notably the synthesis of platelet-activating factor (PAF) and leukotriene C4 (LTC4). Several forms of PLA2 have been identified in mammalian cells. In the present study, the 14-kD, secretory form of PLA2 was detected in human eosinophils by immunocytochemical staining with the specific monoclonal antibody (MoAb) 4A1. In contrast, preparations of neutrophils, monocytes, lymphocytes, and basophils did not show detectable staining. With two MoAbs in a sandwich enzyme-linked immunosorbent assay (ELISA), large amounts of sPLA2 were detected in lysates of eosinophils, that were 20-fold to 100-fold higher than in the other circulating leukocytes (ie, neutrophils, basophils, monocytes, and lymphocytes). In addition, with a commercially available sPLA2 activity assay kit, we were able to show high activity of sPLA2 in human eosinophils relative to neutrophils. Investigations at the ultrastructural level showed that sPLA2 in eosinophils is mainly located in specific granules. Immunoelectron microscopy also visualized sPLA2 within phagosomes after addition of opsonized particles to the eosinophils. However, sPLA2 was not detected in the cell-free supernatants of activated eosinophils, in contrast to eosinophil-cationic protein (ECP), which colocalizes with sPLA2 in resting eosinophils. These findings warrant further studies into the role of sPLA2 in eosinophil function.

Involvement of the contact system of coagulation in the pathogenesis of various inflammatory diseases is suggested by reduced plasma levels of factor XII (Hageman factor) and prekallikrein generally considered to result from activation of the contact system. However, in many of these diseases patients develop an acute-phase response and, therefore, an alternative explanation for the decreased levels of factor XII could be the downregulation of factor XII gene expression in the liver as described for negative acute-phase proteins. We report here that interleukin-6 (IL-6), the principal cytokine mediating the synthesis of most acute-phase proteins in the liver, downregulates the production of factor XII by the human hepatoma cell line HepG2 by up to 75%. The decrease in protein secretion correlated with an equivalent decrease of factor XII mRNA likely indicating a pretranslational control of factor XII gene expression by IL-6. Downregulation of factor XII production by IL-6 in vitro parallelled that of transthyretin, a known negative acute-phase protein. Moreover, we show that, in patients developing an acute-phase response after immunotherapy with IL-2, plasma levels of factor XII correlate (r = .76, P < .0001) with those of transthyretin. Taken together, these results suggest that factor XII behaves as a negative acute-phase protein.

To determine the role of tumor necrosis factor (TNF) in lipopolysaccharide (LPS)-induced inflammation, 12 healthy subjects received an intravenous injection with LPS (2 ng/kg) preceded by infusion of either a recombinant human dimeric TNF receptor type II-IgG fusion protein (TNFR:Fc; 6 mg/m2; n = 6) or vehicle (n = 6) from -30 minutes to directly before LPS injection. LPS elicited a transient increase in plasma TNF activity, peaking after 1.5 hours (219 +/- 42 pg/mL; P < .05). Infusion of TNFR:Fc completely neutralized endogenous TNF activity. LPS administration was associated with an early activation of fibrinolysis (plasma concentrations of tissue-type plasminogen activator, plasminogen activator activity, and plasmin-alpha2-antiplasmin complexes), followed by inhibition (plasma plasminogen activator inhibitor type I), changes that were completely prevented by TNFR:Fc. By contrast, TNFR:Fc did not influence LPS-induced activation of coagulation (plasma levels of prothrombin fragment F1 + 2 and thrombin-antithrombin III complexes). TNFR:Fc strongly inhibited endothelial cell activation (plasma levels of soluble E-selectin), modestly reduced neutrophil responses (neutrophilia and plasma concentrations of elastase-alpha1-antitrypsin complexes and lactoferrin), but did not affect the release of secretory phospholipase A2 or lipopolysaccharide-binding protein (P > .05). Infusion of TNFR:Fc only (without LPS) in another 6 normal subjects did not induce any inflammatory response. These data indicate that TNF is involved in only some inflammatory responses to intravenous LPS in humans.