No abstract

No abstract

SummaryThe pro-inflammatory cytokine tumour necrosis factor-a (TNF-α) is able to alter the haemostatic balance of human umbilical vein endothelial cells (HUVECs) towards that of a procoagulant and anti-fibrinolytic state. Treatment of HUVECs in culture with human recombinant TNF-α (0.5-50 U/ml; 6 h) significantly increased total cell expression of tissue factor (TF) 10-fold from 40 mU/well to 400-500 mU/well. Levels of plasminogen activator inhibitor-1 (PAI-1) antigen secreted from HUVECs also increased up to 2-fold in concentration-dependent fashion following addition of TNF-α (10-100 U/ml; 24 h). TNF-α induced total and cell surface expression of TF on HUVECs was significantly inhibited when the cells were pre-incubated with interleukin-4 (IL-4; p <0.001). This effect was time and concentration dependent. Pretreatment of HUVECs with IL-4 for 4 h had no significant effect, but increasing inhibition of total TF expression occurred after 8 and 16 h pre-incubations. Treatment with IL-4 at 20 and 200 U/ml significantly inhibited cell surface TF responses induced by TNF-α, whereas a low concentration (0.2 U/ml) was without effect. In contrast, the production of PAI-1 from HUVECs stimulated by TNF-α (50 U/ml) was unaffected by the presence and/or prior incubation with 200 U/ml IL-4. Thus, IL-4 may regulate the pro-coagulant but not the antifibrinolytic effects of TNF-α at sites of vascular inflammation.

No abstract

Summary Collagen (10-40 μg kg−1), thrombin (1-10 units kg−1), adenosine diphosphate (ADP; 3-300 pμ kg−1), 1-0-hexadecyl Paf-acether and 1-0-octadecyl Paf-acether (1-300 ng kg−1) administered by bolus intravenous injection each caused dose-dependent thrombocytopoenia accompanied by marked hypotension in anaesthetized rabbits. Responses to ADP and the Paf-acether derivatives were transient in nature (3-8 min) whereas those induced by collagen and thrombin were always of longer duration (5-20 min) and frequently fatal at high doses. Responses to collagen, thrombin, and the Paf-acether derivatives were invariably accompanied by substantial, dose-related increases in plasma levels of thromboxane B2 in samples obtained 30 s after agonist administration, whereas following ADP, no change in plasma thromboxane B2 was detected at any dose level. Indomethacin (3.0 mg kg−1 by infusion) had no effect on responses to thrombin or Paf-acether, partially inhibited collagen-induced thrombocytopenia, and potentiated responses to ADP. In contrast, dazoxiben (10 mg kg−1 by infusion) partially but significantly inhibited responses to thrombin, whereas those induced by collagen, Paf-acether or ADP were unchanged. These results indicate that in this model of intravascular aggregation, whilst platelet responses to collagen and thrombin appear partially dependent on intact cyclic endoperoxide and thromboxane A2 synthetic capacity respectively, responses to ADP and Paf-acether are independent of arachidonate metabolism via cyclo-oxygenase despite measurably increased TXB2 formation in the latter case.