Platelet-activating factor enhances interleukin-6 production by alveolar macrophages.

The production of the cytokine interleukin-6 (IL-6) by rat alveolar macrophages (AMs) was analyzed after their stimulation with muramyl dipeptide (1 microgram/ml), in the presence of graded concentrations of platelet-activating factor (PAF). Significantly enhanced production of IL-6 was observed at 10(-10) to 10(-8) mol/L PAF, with peak effect at 10(-10) mol/L. This enhancement was blocked by three structurally unrelated specific PAF receptor antagonists BN 52021, WEB 2170, and CV 3988. The biologically inactive PAF precursor/metabolite, lyso-PAF, and the enantiomer enantio-PAF failed to induce significant enhancement in IL-6 production. In parallel, addition of PAF to AM triggered leukotriene B4 (LTB4) release. Inhibition of 5-lipoxygenase pathway by AA-861 or MK 886 inhibited the PAF-induced augmentation of both IL-6 and LTB4 production, suggesting an implication of endogenous leukotrienes in this mechanism. Furthermore, addition of exogenous LTB4 to AMs could augment their IL-6 production, with peak activity at 10(-12) mol/L LTB4, and reverse the inhibitory effects of 5-lipoxygenase inhibitors. Taken together, these observations suggest that PAF can modulate lung immune and inflammatory responses by enhancing IL-6 production and that this activity may be dependent on secondary 5-lipoxygenase metabolites. This may have clinical relevance in PAF-mediated events in the lung, such as the cellular components of late-phase asthma.     

Differentiation-dependent modulation of TNF production by PAF in human HL-60 myeloid leukemia cells.

Platelet-activating factor (PAF) can augment tumor necrosis factor (TNF) production by human monocytes in a bimodal manner, with two peaks of activation at picomolar and micromolar concentrations. These peaks are partially associated with monocyte subsets presenting different characteristics in terms of size, density, phenotypic markers, and [Ca2+]i mobilization responses. In the present study, we used the human promyelocytic leukemia cell line HL-60, at various times during differentiation with 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] toward the monocyte lineage, in order to study the relation of cell differentiation to responsiveness to PAF in terms of cytokine production. TNF production was induced by pretreatment with interferon gamma for 24 h and treatment with muramyl dipeptide. Although detectable TNF was produced by 4 day-differentiated cells, no effect was seen with PAF (10(-16)-10(-6) M) at this or earlier stages. In contrast, 5 day-differentiated cells had a comparable baseline production of TNF but responded with a 2.5-fold increase to PAF with a single peak, maximal at 10(-8) M. Moreover, 6 day- or 7 day-differentiated HL-60 cells showed a further increase in TNF production in response to PAF, and the response was bimodal, similar to that of the less dense subset of monocytes, with peaks at 10(-14) and 10(-7) M PAF. In parallel, undifferentiated HL-60 failed to respond to PAF in terms of [Ca2+]i mobilization. The earliest responsiveness to PAF (10(-7) M) was observed by 4 days of treatment with 1,25(OH)2D3, and by day 7 the response to PAF became bimodal (10(-14) and 10(-7) M). These results indicate that myeloid cells acquire, during maturation toward the monocyte lineage, a progressive responsiveness to PAF in terms of [Ca2+]i mobilization and enhanced cytokine production, and they suggest that the heterogeneity in responses to PAF observed in normal monocytes may be related to their stage of differentiation or maturation.     

Immune regulation by platelet-activating factor: II. Mediation of suppression by cytokine-stimulated endothelial cells in vitro

Human umbilical vein endothelial cells (EC) can respond to endotoxin or to the inflammatory cytokines tumor necrosis factor (TNF) and interleukin 1 (IL-1) by producing platelet-activating factor (PAF). When EC were preexposed to TNF-alpha (25 U/ml) for 1 h, and then washed, their subsequent coculture with peripheral blood mononuclear cells (PBMC) resulted in suppressed proliferative response of the latter to the mitogen Con A (P less than 0.05). This effect was completely reversed by the concomitant use of the PAF receptor antagonist BN 52021 (0.1 mM). Preexposure of EC to IL-1 beta (0.5 U/ml) induced similar effects, but IL-1 and TNF were not additive. Removal of monocytes from the PBMC population abolished the effects. On the other hand, coculture of monocytes with cytokine-preexposed EC resulted in significant induction of suppressor activity on lymphocyte proliferation. Our data indicate that EC, preexposed to inflammatory cytokines, can modulate lymphocyte functions via the production of PAF and its action on monocytes.     

Effect of long-term treatment with platelet-activating factor on cytokine production by rat spleen cells

The long-term in vivo effect of platelet-activating factor (PAF) production of interleukin-1 and -2 (IL-1, IL-2) was investigated. Alzet infusion minipumps loaded with PAF or solvent were placed under the back skin of Sprague-Dawley rats and connected to the jugular vein. Lymphocytes from animals having received 1, 4.5 or 9 micrograms PAF/7 days showed an increased capacity to produce IL-1 and IL-2. In contrast, splenocytes from rats receiving 28 micrograms PAF/7 days exhibited decreased capacity to produce IL-1, whereas IL-2 was unaffected. The decrease in IL-1 synthesis induced by 28 micrograms PAF and the increase in IL-2 production evoked by 1 microgram PAF were not observed in rats treated daily with the PAF antagonist, BN 52021. Thus, PAF appears to play a role in the regulation of the immune response. The reversal of the effect of PAF by BN 52021 indicates that the mediator is acting via specific binding sites similar to those reported on other cell types. These data also suggest that PAF antagonists may be used as immunomodulatory drugs.     

Platelet-activating factor modulates interleukin-6 production by mouse fibroblasts

We compared the effects of platelet-activating factor (PAF), interleukin-1 beta (IL-1 beta) and polyriboinositic-polyribocytidylic acid (poly-I:C) on IL-6 production by confluent L929 mouse fibroblasts. At concentrations above 1 microM, PAF dose-dependently enhanced IL-6 production; at 5 microM PAF this increase (72.7 +/- 19.9 U/ml) was higher than that evoked by 100 U/ml IL-1 beta (5.7 +/- 0.4 U/ml) or 50 micrograms/ml poly-I:C (39.3 +/- 6.7 U/ml). The IL-6 production induced by 5 microM PAF was not inhibited by addition of the specific PAF antagonist BN 52021 (10 microM) to the incubation medium. These results demonstrate that, as this is the case for IL-1, PAF also modulates IL-6 production.     

Stimulation of tumour necrosis factor release by cytotoxic analogues of platelet-activating factor.

The capacity of cytotoxic analogues of platelet-activating factor (PAF) to stimulate tumour necrosis factor-alpha (TNF-alpha) synthesis and release by human monocytes was determined. Cell-associated TNF-alpha was quantified by protein immunoblotting and released TNF-alpha was quantified by cytotoxicity bioassay. Picomolar concentrations of methoxyPAF, SDZ 62-759, SDZ 68-826, SDZ 62-434 and SRI 62-834 induced a two- to fivefold increase in cell-associated and released TNF-alpha. These compounds were as potent as PAF for stimulating monocytes. In contrast, they lacked direct platelet-activating activity and inhibited platelet aggregation induced by PAF selectivity. The analogues inhibited PAF binding to platelets but not to monocytes. The PAF binding antagonists kadsurenone, BN52021 and WEB2086 inhibited TNF-alpha release induced by 10(-11) M PAF or methoxyPAF by a maximum of only 30-60% whereas they inhibited platelet aggregation by 10(-8) M PAF completely. Monocyte receptors for methoxyPAF were evaluated. Scatchard analysis of [3H]methoxyPAF binding to monocytes revealed large numbers of relatively low affinity receptors (Kd = 5.9 +/- 0.5 x 10(-7) M; 9.1 +/- 4.2 x 10(7) sites/monocyte). These values do not correspond to binding constants of monocyte receptors for PAF and do not account for monocyte activation by picomolar concentrations of methoxyPAF. Cytotoxic analogues of PAF stimulate TNF-alpha synthesis and release but they do not stimulate monocytes by interacting with PAF receptors.     

Control of human T cell proliferation by platelet-activating factor

Platelet-activating factor, 1-O-alkyl-2-acetyl-sn-glycero-3-phosphocholine (PAF), is a membrane phospholipid with immunomodulatory functions. We studied the influence of PAF on mitogen-stimulated proliferation of human peripheral blood mononuclear cells (PBMC), purified T cells and T cell subsets. High concentrations of PAF suppressed the proliferation of all cell populations studied (44% mean inhibition with 5 microM PAF and 78% inhibition with 10 microM PAF). In contrast, the deacetylated metabolite of PAF, lyso-PAF, had no effect on proliferation at micromolar concentrations. Lower, and presumably physiologically relevant, concentrations of PAF (10(-14) to 10(-8) M) stimulated a small increase in the proliferation of unfractionated T cells. When T cells were fractionated into CD4+ and CD8+ subsets, a difference in sensitivity to PAF was observed. PAF stimulated a modest, yet statistically significant, increase in the proliferation of CD4+ T cells at concentrations ranging from 10(-14) to 10(-10) M, while either having no effect or inhibiting the proliferation of CD8+ cells across the entire concentration range. Addition of indomethacin to the cultures further enhanced CD4+ proliferation, possible due to the blockade of PAF-induced PGE2 production by monocytes. The PAF receptor antagonist BN 52021 did not block the PAF effects in this system, and the PAF receptor antagonist SRI 63-675 caused a dose dependent inhibition of T cell subset proliferation. These findings suggest that while high concentrations of PAF suppress T cell proliferation, low concentrations selectively stimulate proliferation of the CD4+ subset, an effect which is partially counteracted by PAF-induced monocyte PGE2 production.     

Inhibition of platelet-activating factor (PAF)-induced chemotaxis and PAF binding to human eosinophils and neutrophils by the specific ginkgolide-derived PAF antagonist, BN 52021

We have investigated the effects of BN 52021 (a specific PAF antagonist derived from Ginkgo biloba) on PAF-induced human eosinophil and neutrophil chemotaxis. In response to an optimal concentration of PAF (10(-6) mol/L), the drug was significantly more potent (p less than 0.001) in inhibiting eosinophil as compared to neutrophil locomotion. These inhibitory effects were observed in a dose-dependent manner with a concentration of drug required to produce 50% inhibition of 7.0 (+/- 2.2) X 10(-6) mol/L and 2.3 (+/- 0.2) X 10(-5) mol/L for eosinophils and neutrophils, respectively. Sodium cromoglycate, nedocromil sodium, salbutamol, and dexamethasone (preincubated with cells up to 6 hours) had no effect over a wide dose range (10(-3) to 10(-9) mol/L). BN 52021 was significantly more effective in inhibiting chemotaxis when the cells were preincubated with the compound for up to 1 hour before commencement of the locomotion assay, whereas washing the cells completely abolished this effect. Inhibition by BN 52021 was specific for PAF in that it had no effect on chemotaxis induced by either leukotriene B4, N-formyl-methionyl-leucyl-phenylalanine, or a purified human mononuclear cell-derived neutrophil chemotactic factor. BN 52021 also inhibited the specific binding of [3H]-PAF (10(-8) mol/L) to eosinophils and neutrophils in a concentration-dependent fashion with a concentration of drug required to produce 50% inhibition of 1.5 (+/- 0.3) X 10(-6) mol/L and 9.1 (+/- 2.5) X 10(-7) mol/L, respectively. These results suggest that BN 52021 has potential as an anti-inflammatory agent in conditions associated with PAF-induced accumulation of neutrophils and eosinophils.     

IL-10 stimulates production of platelet-activating factor by monocytes of patients with active systemic lupus erythematosus (SLE)

IL-10 displays modulatory properties on the synthesis of platelet-activating factor (PAF), a potent inflammatory mediator of vascular injury. Despite the fact that IL-10 is considered to be an anti-inflammatory cytokine, IL-10 levels correlate with disease activity in SLE. Moreover, in SLE IL-10 is unable to exert its immunosuppressive and anti-inflammatory effects. We have investigated the ability of IL-10 to stimulate PAF production from monocytes of SLE patients. Spontaneous and IL-10-stimulated PAF production by peripheral blood monocytes was measured in active (n = 13) and inactive (n = 14) SLE patients and in 15 normal control subjects. We observed that monocytes derived from patients with active SLE, but not from controls or inactive SLE, spontaneously produced significant amounts of PAF. Moreover, IL-10 enhanced the synthesis of PAF from monocytes of active SLE patients only. IL-10-induced PAF production correlated with the severity of the disease and with the extent of proteinuria. These results indicate that IL-10 only stimulates the synthesis of PAF from monocytes of SLE patients when immunologically active, suggesting that IL-10 may possess a paradoxical proinflammatory effect in SLE by promoting the production of PAF, a secondary mediator of inflammation.     

Effect of the platelet-activating factor antagonist BN 52021 on human natural killer cell cytotoxicity

The possible role of platelet-activating factor (PAF) in natural killer (NK) cell cytotoxicity was investigated by examining the effect of the PAF antagonist BN 52021 in NK cytotoxicity towards 51Cr-labelled K 562 target cells. When BN 52021 (30-120 microM) was added during the assay, a dose-dependent inhibition of NK activity was observed. The inhibition of cytotoxicity by BN 52021 was not due to an alteration of the binding of lymphocytes to K 562 cells. When lymphocytes were preincubated with BN 52021 (60 microM) for 60 min before the target cells were added, the inhibitory effect of the drug was similar to that observed when it was added at the start of the reaction. Inhibition was more pronounced when the target cells were pretreated for 60 min before the start of the assay. BN 52021 (60 microM) also inhibited gamma interferon induced NK activity. These studies provide indirect evidence that NK cells can generate PAF and that this mediator is involved in cytotoxic processes.     

Effects of the platelet activating factor antagonists BN 52021 and BN 50730 on antigen-induced bronchial hyperresponsiveness and eosinophil infiltration in lung from sensitized guinea-pigs

The involvement of platelet activating factor (PAF) in antigen-induced bronchial hyperresponsiveness was investigated by the use of the PAF antagonists BN 52021 and BN 50730, in a guinea-pig model where sensitization and challenge were performed by aerosol. Male Hartley guinea-pigs were sensitized by two aerosol exposures at 48 hr intervals to a 0.9% NaCl solution (saline) containing 2 mg/ml ovalbumin for 30 min. Fifteen to 20 days later, guinea-pigs were challenged by exposure to five successive aerosols of increasing concentrations of ovalbumin (OA) or respectively, 10 microg/ml, 100 microg/ml, 1 mg/ml, 5 mg/ml and 10 mg/ml for 15 min each, or saline alone. Three to four hr and 18-24 hr after the aerosol challenge the guinea-pigs were prepared for recording of bronchopulmonary response and aerosol administrations were then generated with an ultrasonic nebulizer. The bronchopulmonary responses induced by successive 1-min aerosol bursts of acetylcholine (ACh) was assessed. As compared with saline-challenged guinea-pigs, an enhanced bronchopulmonary response to aerosol administration of cumulative doses of ACh was observed, 3-4 hr and 18-24 hr post-ovalbumin challenge. When the sensitized guinea-pigs were pretreated 1 hr before ovalbumin exposure with BN 52021 or BN 50730 (25 mg/kg, per os), a significant inhibition of the increase in the bronchopulmonary response to ACh was observed, both at 3-4 hr and 18-24 hr. Furthermore, when guinea-pigs were treated 3-4 hr after the ovalbumin exposure with BN 52021 or BN 50730, a significant inhibition of the hyperresponsiveness to ACh was recorded at 18-24 hr. A marked accumulation of eosinophils in the peribronchial regions was observed on histological preparations of lung specimens collected 4 hr or 24 hr after ovalbumin exposure. Pretreatment of the guinea-pigs by BN 50730 or BN 52021 did not modify the eosinophil accumulation in the peribronchial area. No significant difference in the number of eosinophils collected in the bronchoalveolar lavage fluid is observed, 24 hr post-ovalbumin challenge, under the pretreatment with BN 52021 or BN 50730. Pretreatment of guinea-pigs by BN 50730 or BN 52021 significantly reduced the PAF-induced (100 microg/ml) increase in eosinophil number in the peribronchial area. By contrast, they did not inhibit the eosinophilia induced by aerosol administration of LTB4 (5 microg/ml). These results suggest that the bronchial hyperresponsiveness observed in this study is associated with eosinophil accumulation in the lung. The potent inhibition of the bronchial hyperresponsiveness by the two unrelated antagonists of PAF suggests that the lipid mediator is involved in its triggering and duration, but not in the eosinophil infiltration.     

Protection of platelet-activating factor-induced acute renal failure by BN 52021.

Although platelet-activating factor (PAF) is a well-known mediator in experimental shock, its precise role in acute renal failure remains to be defined. Male Wistar rats (209 +/- 12 g) were placed in individual metabolic cages for 24 h before i.v. injection of PAF (2-6 micrograms/kg). Injection of 6 micrograms/kg PAF proved lethal and use of such a high dose was thus discontinued. Administration of 2 micrograms/kg and 4 micrograms/kg PAF resulted in a fall of glomerular filtration rate (GFR) associated with a reduction in urinary flow rate (UFR). Rats pretreated with BN 52021 (25 mg/kg p.o.) exhibited values of GFR similar to that of the control group, but not after 4 micrograms/kg PAF. In addition, in the group of BN 52021 pretreated rats and injected with 2 micrograms/kg or 4 micrograms/kg PAF, UFR was not significantly different from that of the control group at 24 h. Examination by electron microscopy revealed the presence of platelets in the glomeruli, as well as loss of fixed anionic charges in PAF injected rats. The presence of these platelets was not observed in rats treated with BN 52021 and injected with PAF. No changes in GFR and UFR were observed at 6 h or 24 h in vehicle or BN 52021 treated rats. Thus, BN 52021 which affords protection against acute renal failure induced by PAF may be of therapeutic value in other types of kidney disease in which this mediator is active.     

Protection of platelet-activating factor-induced acute renal failure by BN 52021

Although platelet-activating factor (PAF) is a well-known mediator in experimental shock, its precise role in acute renal failure remains to be defined. Male Wistar rats (209 +/- 12 g) were placed in individual metabolic cages for 24 h before i.v. injection of PAF (2-6 micrograms/kg). Injection of 6 micrograms/kg PAF proved lethal and use of such a high dose was thus discontinued. Administration of 2 micrograms/kg and 4 micrograms/kg PAF resulted in a fall of glomerular filtration rate (GFR) associated with a reduction in urinary flow rate (UFR). Rats pretreated with BN 52021 (25 mg/kg p.o.) exhibited values of GFR similar to that of the control group, but not after 4 micrograms/kg PAF. In addition, in the group of BN 52021 pretreated rats and injected with 2 micrograms/kg or 4 micrograms/kg PAF, UFR was not significantly different from that of the control group at 24 h. Examination by electron microscopy revealed the presence of platelets in the glomeruli, as well as loss of fixed anionic charges in PAF injected rats. The presence of these platelets was not observed in rats treated with BN 52021 and injected with PAF. No changes in GFR and UFR were observed at 6 h or 24 h in vehicle or BN 52021 treated rats. Thus, BN 52021 which affords protection against acute renal failure induced by PAF may be of therapeutic value in other types of kidney disease in which this mediator is active.     

The effect of the platelet-activating factor antagonist, BN 52021, on human natural killer cell-mediated cytotoxicity.

The influence of the platelet-activating factor (PAF) antagonist, BN 52021, on human natural killer (NK) cell cytotoxicity against K 562 target cells was determined. Cytotoxicity was measured by a short-term (4 hr) 51Cr-release assay. The cytotoxicity was significantly reduced in the presence of PAF antagonist at concentrations from 30 to 120 microM. This reduction of killing was not due to the impairment of binding of effector cells to target cells. Pretreatment of K 562 target cells with the PAF antagonist led to a greater inhibition of NK cell cytotoxicity compared with that observed when the effector cells were preincubated with BN 52021. Thus, the inhibition of cytotoxicity appears to be due to an effect of BN 52021 on target cells rather than on lymphocytes. Furthermore, the increase in NK activity induced by interferon was less pronounced when BN 52021 was added in the incubation medium. The natural cytotoxicity of platelet-depleted or large granular lymphocyte-enriched effector cell populations was inhibited by the PAF antagonist in a similar manner. The effect of BN 52021 appears to be related to its specific PAF antagonistic activity since a similar action on NK cells was noted with two other structurally unrelated PAF antagonists, BN 52111 and WEB 2086. In contrast, Ginkgolide J (BN 52024), which is structurally related to BN 52021 but lacks PAF antagonistic activity, was ineffective in inhibiting NK cell cytotoxicity. Finally, synthetic PAF induces a dose-dependent cytotoxic action on K 562 cells and this effect of the autacoid is inhibited by BN 52021. These observations provide indirect evidence that PAF could play a role in the mechanism(s) of NK cytotoxity.     

Modulatory effect of interleukin-10 on the production of platelet-activating factor and superoxide anions by human leucocytes.

We observed that human monocytes (MO) and polymorphonuclear neutrophils (PMN) stimulated by lipopolysaccharide (LPS) produce platelet-activating factor (PAF) in a pattern characterized by an early and a delayed peak of synthesis. The early peak of PAF synthesis was due to a direct stimulation of these cells through mCD14 receptor as it was inhibited by anti-CD14 monoclonal antibody. The delayed and sustained peak of PAF synthesis was dependent on protein synthesis and cytokine production as shown by the inhibitory effect of cycloheximide on both MO and PMN, and of anti-tumour necrosis factor-alpha (anti-TNF-alpha) and of anti-interleukin-8 (anti-IL-8) neutralizing antibodies on MO and PMN respectively. IL-10 completely prevented this second, cytokine-dependent peak of PAF synthesis. In contrast, IL-10 markedly enhanced the first peak of PAF synthesis both in MO and PMN. Moreover, IL-10 was shown to modulate the production of superoxide anions (O2-) on both MO and PMN. As suggested by previous studies, IL-10 inhibited the delayed production of O2-. In the present study, we observed that IL-10 directly stimulated an early production of O2-. In addition, IL-10 enhanced the synthesis of O2- by MO and PMN challenged with LPS. The IL-10-induced O2- production was dependent, at least in part, from its effect on PAF synthesis, as it was inhibited by the PAF receptor antagonist WEB 2170. These results suggest that IL-10 may upregulate the early synthesis of PAF and O2- triggered by direct LPS stimulation, whereas it may downregulate the delayed production of these mediators.     

Effects of PAF-acether on electrophysiological response of isolated retina

Results of experiments performed on rat isolated retina indicate that platelet-activating factor (PAF) is able to inhibit the functional response of the retina electroretinogram (ERG) recorded in response to a brief light flash. In the presence of PAF, the ERG b-wave amplitude decreases according to a dose-dependent (2.10(-11) M; 2.10(-9) M; 2.10(-7) M) process. This effect is partially inhibited by the simultaneous administration of a Ginkgo biloba extract (GBE, 10 mg/l) or Ginkgolide B (BN 52021, 2.10(-5)M). The authors interpret these results with reference to the main mechanism of the membrane signal triggered by PAF, namely the activation of phosphatidylinositol cycle with the formation of inositol-triphosphate, the inhibition of the light-induced response of the retina by administration of inositol-triphosphate, and the antagonistic effect of GBE and BN 52021 on specific PAF-receptors demonstrated on other models. Thus specific PAF-receptors may exist at the level of the retina, which suggests that they are also present in the brain.     

Effects of the platelet-activating factor antagonist BN 52021 on anti-islet cytotoxicity of mononuclear cells and serum from type 1 (insulin-dependent) diabetic patients

The effect of platelet-activating factor (PAF) antagonist BN 52021 (0.06-2.5 mM) on the cytotoxic activity of mononuclear cells (MNC) from newly diagnosed type 1 diabetic patients against 51Cr-labeled Langerhans islets from neonatal rats was investigated in a 6-hour cytotoxicity test. A dose-dependent inhibition of anti-islet cytotoxicity by BN 52021 was observed. The suppression of the islet lysis was significant at a concentration of 0.6 mM BN 52021. During a 4-day cell culture, BN 52021 had no inhibitory effect on the antigen-mediated triggering of immunocytes with anti-islet cytotoxicity. The results suggest that the drug is only effective during immunocytolytic reactions of MNC against pancreatic islets. A PAF-independent action of BN 52021 can not be excluded at present.     

Thrombin-treated endothelium primes neutrophil functions: inhibition by platelet-activating factor receptor antagonists

We have shown previously that fluid phase platelet-activating factor (PAF) can enhance or "prime" polymorphonuclear (PMN) responses to subsequent stimulation with agonists such as formyl-methionine-leucine-phenylalanine (FMLP). Since thrombin induces PAF production in endothelial cells, we tested whether this thrombin-provoked endothelial PAF primes responses of marginated PMNs. Monolayers of human umbilical vein endothelial cells were exposed to either thrombin (0.5-5.0 units/ml) or buffer for up to 5 min and then PMNs were layered on top of the endothelial cells. After a further 5 min incubation, the PMNs were stimulated with a suboptimal concentration of FMLP (10(-7) M), and their superoxide production, elastase release, adhesion to endothelium, and capacity to cause endothelial cell lysis and detachment were assessed. Thrombin pretreatment significantly enhanced each of these FMLP-stimulated neutrophil responses. The extent of this enhancement correlated with both the dose and duration of thrombin treatment of endothelial cells and also the duration of PMN incubation with thrombin-exposed endothelium. Evidence that the augmentation was due to endothelial-derived PAF was obtained as follows: (1) thrombin induced [3H]acetate incorporation into endothelial PAF (assayed in lipid extracts); (2) antithrombin III conjointly inhibited this [3H]acetate uptake and prevented the priming effect of thrombin-treated endothelium on PMN responses; and (3) the PAF receptor antagonist BN52021, when preincubated with PMNs, also effectively blocked the enhancement of PMN responses. We conclude that thrombin stimulation of endothelial cells initiates a sequence of events culminating in the production of PAF--a membrane phospholipid capable of priming marginated PMNs. We suggest that this coagulation-fostered endothelial/PMN interaction may underlie a paracrine response that may potentiate PMN-mediated endothelial injury during sepsis and other thrombin-generating disorders.