Prostanoid synthesis by human umbilical artery

A discrepancy between published values of PGI2 production by human umbilical artery in vitro measured by platelet bioassay, compared with values of 6-oxo-PGF1 alpha by radioimmunoassay, raised the possibility that another anti-aggregatory prostanoid was produced by this tissue. To test this hypothesis, umbilical artery rings were incubated in buffer and PGI2 determined by platelet bioassay and by a more specific radioimmunoassay based on comparison of 6-oxo-PGF1 alpha in hydrolysed and non-hydrolysed samples. 6-oxo-PGF1 alpha, PGF2 alpha and TXB2 were also measured by gas chromatography negative ion chemical ionisation mass spectrometry. PGI2 concentrations by radioimmunoassay and bioassay were significantly correlated (r = 0.92, p less than 0.01). There was no difference between them, disproving the presence of an additional antiaggregatory substance. PGI2 production determined by bioassay (mean 1.21 ng/mg wet weight/h, range 0.59-1.53 ng/mg/h) differed from previously reported values (range 70-325 ng/mg/h). 6-oxo-PGF1 alpha concentrations were confirmed by gas chromatography negative ion chemical ionisation mass spectrometry. Previous determinations of PGI2 production by this tissue overestimated it by approximately 100 times.     

6-oxo-PGF(1 alpha)and 8-epi-PGF(2 alpha)in the arterial wall layers of various species: a comparison between intact and atherosclerotic areas

PGI(2)and 8-epi-prostaglandin(PG)F(2 alpha)are antagonizing compounds. For both a key role in vascular pathology has been hypothesized. The isoprostane 8-epi-PGF(2 alpha)and the stable derivative of PGI(2), 6-oxo-PGF(1 alpha)were determined immunologically in the arterial wall of various species including humans. Human arterial tissue contained the highest amounts of 8-epi-PGF(2 alpha)and synthesized the lowest PGI(2). A significant negative correlation between 8-epi-PGF(2 alpha)and 6-oxo-PGF(1 alpha)was observed. Atherosclerotic segments showed significantly higher 8-epi-PGF(2 alpha)and lower 6-oxo-PGF(1 alpha). 8-epi-PGF(2 alpha)in the intima was higher than in the media, the highest amounts being found in foam-cell rich areas. Synthetic (activated) smooth muscle cells were associated with an enhanced 8-epi-PGF(2 alpha)as well as 6-oxo-PGF(1 alpha). Tissue samples derived from smokers contained more 8-epi-PGF(2 alpha)and produced less PGI(2). The by far highest 8-epi-PGF(2 alpha)/6-oxo-PGF(1 alpha)ratio was found in foam cell rich areas. Similar findings were obtained in rabbit and in minipig arteries. The total 8-epi-PGF(2 alpha)/6-oxo-PGF(1 alpha)ratio is low in normal tissue, increases significantly in an active atherosclerotic process and seems to be even further increased in an inactive atherosclerotic process. These findings are providing an information on the extent of oxidation injury at various sites of different types of atherosclerotic process.     

Transepithelial prostacyclin gradient in isolated canine trachea

Cyclooxygenase products of arachidonic acid, potential modulators of airway smooth muscle, have recently been described in bronchoalveolar lavage from canine lungs. To evaluate the possibility that airway epithelium represents a barrier to movement of prostacyclin (PGI2), an important bronchodilator synthesized by isolated airway, we measured the concentrations of 6-oxoprostaglandin F1 alpha (6-oxo-PGF1 alpha), the stable degradation product of PGI2, on the mucosal and serosal sides of isolated canine tracheal segments (CTS) mounted in Ussing chambers. 6-oxo-PGF1 alpha was measured by radioimmunoassay after purification by high-performance liquid chromatography. The concentration of 6-oxo-PGF1 alpha was significantly higher on the serosal than the mucosal side of CTS (1,262 +/- 252 vs. 390 +/- 168 pg.min-1.g-1, n = 8, P less than 0.05). A significant correlation was present between 6-oxo-PGF1 alpha measured on both sides of each CTS (r = 0.778, n = 26, P less than 0.01). 6-oxo-PGF1 alpha production from CTS stripped of mucosa was significantly greater than from isolated mucosa. Radiochromatograms obtained after incubation with [3H]arachidonic acid and calcium ionophore A23187 confirmed PGI2 as the predominant cyclooxygenase product of the submucosa, whereas the mucosa produced only small amounts of PGI2 in proportion to other cyclooxygenase products. PGI2 (10(-8) to 10(-6) M) applied to the mucosal surface of closed tracheal segments precontracted with histamine resulted in no significant relaxation, whereas serosal application showed a concentration-dependent effect. Radiolabeled 6-oxo-PGF1 alpha did not cross the isolated epithelium.(ABSTRACT TRUNCATED AT 250 WORDS)     

Urinary levels of 2,3-dinor-6-oxo-PGF1 alpha: a reliable index of the production of PGI2 in the spontaneously hypertensive rat

The urinary levels of 2,3-dinor-6-oxo-PGF1 alpha (PGI2-M), a major metabolite of PGI2, are determined by the balance between the amount of PGI2 synthesized and the extent of its further metabolic oxidation. The purpose of the present study was to determine if the urinary excretion of PGI2-M can be used as a reliable index of the in vivo production of PGI2 in both normal Wistar-Kyoto (WKY) and spontaneously hypertensive rats (SHR). This involved the exclusion of differences in metabolism between these two strains of rats. In order to do so, we monitored the urinary excretion of PGI2-M during paired intravenous infusions of 6-oxo-PGF1 alpha (the stable product of the spontaneous hydrolysis of PGI2) in conscious, unrestrained SHR and WKY rats aged 12-15 weeks, in doses ranging from 250 to 700 ng. In one experiment, PGI2 was infused instead of 6-oxo-PGF1 alpha. The results of these experiments indicate that SHR and WKY rats are equal with regard to the transformation of 6-oxo-PGF1 alpha and PGI2 into PGI2-M. For both groups, there is a good correlation between the amount of 6-oxo-PGF1 alpha infused and the amount of PGI2-M excreted in urine. These observations confirm the validity of using the urinary levels of 2,3-dinor-6-oxo-PGF1 alpha as an index of PGI2 production in both WKY and SHR. In addition, they support the conclusions drawn from our previous studies, namely that SHR do not produce more PGI2 than WKY rats in vivo, contrary to the situation prevailing in vitro.     

The effects of prostacyclin (PGI2) on tissues which detect prostaglandins (PG'S).

The effects of prostacyclin (PGI2) and its stable metabolite 6-oxo-PGF1alpha on various bioassay tissues are compared with those of PGE2 and PGF2alpha, using the cascade superfusion method. On vascular smooth muscle, PGI2 caused relaxation of all tissues tested except the rabbit aorta. PGE2 relaxed rabbit coeliac and mesenteric artery but contracted bovine coronary artery and had no effect on rabbit aorta. 6-oxo-PGF1alpha was ineffective at the concentrations tested. On gastro-intestinal smooth muscle, PGI2 contracted strips of rat and hamster stomach and the chick rectum. It was less potent than PGE2 or PGF2alpha. None of these substances contracted the cat terminal ileum. 6-oxo-PGF1alpha was inactive on these tissues at the doses tested. PGI2 was less active than PGE2 or PGF2alpha in contracting guinea-pig trachea and rat uterus; 6-oxo-PGF1alpha was active only on the rat uterus. Thus, PGI2 can be distinguished from the other stable prostaglandins using the cascade method of superfusion.     

Measurement of 6-oxo-PGF1 alpha in human plasma using gas chromatography-mass spectrometry.

The plasma concentration of the prostacyclin (PGI2) hydration product 6-oxo-PGF1 alpha has been assayed by stable isotope dilution GC-MS in six normal volunteers infused with increasing doses of PGI2 intravenously. The predosing levels of 6-oxo-PGF1 alpha ranged between 114 and 266 pg/ml. Infusion of PGI2 increased 6-oxo-PGF1 alpha concentration in plasma but the increments were lower than expected suggesting less conversion of the PGI2 to 6-oxo-PGF1 alpha at high infusion rates.     

Thromboxane synthase inhibition: implications for prostaglandin endoperoxide metabolism. II. Testing the 'redirection hypothesis' in an acute intravenous challenge model

In the preceding paper we described the characterisation of an acute intravenous challenge model for the evaluation of the effects of thromboxane synthase inhibition (TXSI) on eicosanoid metabolism. Herein we describe the biochemical pharmacology of two TXSI and aspirin in this model. Both TXSI caused significant inhibition of plasma TXB2 in vivo without elevation of 6-oxo-PGF1 alpha levels. Similar results were obtained when combined levels of 6-oxo-PGF1 alpha,13,14 dihydro 6-oxo-PGF1 alpha,13,14 dihydro 6,15-dioxo-PGF1 alpha and 6-oxo-PGE1 were measured as an index of PGI2 biosynthesis (PGI2m). Thus no evidence of in vivo redirection of PGH2 to PGI2 was found. Ex vivo experiments performed in serum gave an apparent stimulation of immunoreactive 6-oxo-PGF1 alpha following TXSI but RPHPLC analysis of extracted serum showed that this stimulation was accounted for by increase in a product co-eluting with [3H]PGF2 alpha. The implications of these findings in relation to TXSI and receptor antagonists are discussed.     

Prostacyclin and thromboxane A2 in thrombotic thrombocytopenic purpura.

A study was conducted to find whether a deficiency in prostacyclin (prostaglandin I2; PGI2) is implicated in the pathogenesis of thrombotic thrombocytopenic purpura. Plasma samples from two patients with the disease before treatment and from 22 healthy controls were therefore assayed for concentrations of 6-oxo-PGF1 alpha and thromboxane B2, the stable metabolites of PGI2 and thromboxane A2, respectively. Neither of the patients responded to treatment, which in one case included an infusion of PGI2. Both patients had normal concentrations of 6-oxo-PGF1 alpha and thromboxane B2, thus implying that circulating amounts of PGI2 and thromboxane A2 were also normal. These findings suggest that 6-oxo-PGF1 alpha may be detectable in normal amounts in thrombotic thrombocytopenic purpura and that the condition need not be associated with a high concentration of thromboxane A2.     

Activity of prostaglandin biosynthetic pathways in rat pancreatic islets

Isolated pancreatic islets of the rat were either prelabeled with [3H]arachidonic acid, or were incubated over the short term with the concomitant addition of radiolabeled arachidonic acid and a stimulatory concentration of glucose (17mM) for prostaglandin (PG) analysis. In prelabeled islets, radiolabel in 6-keto-PGF1 alpha, PGE2, and 15-keto-13,14-dihydro-PGF2 alpha increased in response to a 5 min glucose (17mM) challenge. In islets not prelabeled with arachidonic acid, label incorporation in 6-keto-PGF1 alpha increased, whereas label in PGE2 decreased during a 5 min glucose stimulation; after 30-45 min of glucose stimulation labeled PGE levels increased compared to control (2.8mM glucose) levels. Enhanced labelling of PGF2 alpha was not detected in glucose-stimulated islets prelabeled or not. Isotope dilution with endogenous arachidonic acid probably occurs early in the stimulus response in islets not prelabeled. D-Galactose (17mM) or 2-deoxyglucose (17mM) did not alter PG production. Indomethacin inhibited islet PG turnover and potentiated glucose-stimulated insulin release. Islets also converted the endoperoxide [3H]PGH2 to 6-keto-PGF1 alpha, PGF2 alpha, PGE2 and PGD2, in a time-dependent manner and in proportions similar to arachidonic acid-derived PGs. In dispersed islet cells, the calcium ionophore ionomycin, but not glucose, enhanced the production of labeled PGs from arachidonic acid. Insulin release paralleled PG production in dispersed cells, however, indomethacin did not inhibit ionomycin-stimulated insulin release, suggesting that PG synthesis was not required for secretion. In confirmation of islet PGI2 turnover indicated by 6-keto-PGF1 alpha production, islet cell PGI2-like products inhibited platelet aggregation induced by ADP. These results suggest that biosynthesis of specific PGs early in the glucose secretion response may play a modulatory role in islet hormone secretion, and that different pools of cellular arachidonic acid may contribute to PG biosynthesis in the microenvironment of the islet.     

Thromboxane synthase inhibition: implications for prostaglandin endoperoxide metabolism. I. Characterisation of an acute intravenous challenge model to measure prostaglandin endoperoxide metabolism

It has been proposed that thromboxane synthase inhibition (TXSI) may be a useful form of anti-thrombotic therapy and that this is due, in part, to redirection of PGH2 metabolism in favour of PGI2, a potent vasodilator and anti-platelet agent. While redirection has been observed ex vivo there are conflicting reports of its occurrence in vivo. We now describe the characterisation of an acute intravenous challenge model using thrombin, collagen, arachidonic acid (AA) and PGH2 for the study of PGH2 metabolism. Following challenge, plasma concentrations of TXB2, 6-oxo-PGF1 alpha, alleged metabolites of PGI2 (PGI2m) and PGE2 were measured by radioimmunoassay (RIA). Thrombin and collagen challenge resulted in a dose-related increase in plasma TXB2 while AA and PGH2, in addition, elevated 6-oxo-PGF1 alpha and PGI2m. Injection of PGH2 elevated 6-oxo-PGF1 alpha, PGI2m, TXB2 and PGE2 levels. Experimental conditions were defined such that challenge with thrombin (40 NIH units kg-1), collagen (100 micrograms kg-1), AA (1 mg kg-1) and PGH2 (5 micrograms kg-1) and measurement of eicosanoids 0.5 min following challenge were optimal for detection of redirection of PGH2 metabolism in vivo. The identity of immunoreactive TXB2 and 6-oxo-PGF1 alpha was further supported by experiments in which the extracted immunoreactive eicosanoids co-eluted with authentic [3H]standards when subject to reverse phase high performance liquid chromatography (RPHPLC). Evidence is also presented that the levels of plasma eicosanoids measured in this model reflect in vivo biosynthesis.     

Transforming growth factor alpha stimulates prostacyclin production by cultured human vascular endothelial cells more potently than epidermal growth factor

Transforming growth factor alpha (TGF alpha) induces dose- and time-dependent stimulation of prostacyclin (PGI2) production by cultured human umbilical vein endothelial cells. The lowest stimulatory concentration of TGF alpha was 0.1 ng/ml and the maximal response, a 2.7-fold rise, was obtained with 10 ng/ml. The stimulation, which lasted at least 24 h, was blocked by cycloheximide and by indomethacin. TGF alpha induced PGI2 production at 10-100 times lower concentrations than did epidermal growth factor (EGF), although in stimulating endothelial cell growth the two factors were equipotent. This is the first demonstration that TGF alpha enhances PGI2 production by human cells. Moreover, this is the first evidence that it acts as both an agonist (growth) and a superagonist (PGI2 production) of EGF in the same cell type. I suggest that this phenomenon may be involved with the angiogenic activity of TGF alpha.     

Bronchoconstrictor and antibronchoconstrictor properties of inhaled prostacyclin in asthma

Prostacyclin (PGI2) is generated in appreciable amounts during allergic reactions in human lung tissue. To define its activity on human airways we have studied the effects of doubling concentrations of inhaled PGI2 and its hydrolysis product 6-oxoprostaglandin F1 alpha (6-oxo-PGF1 alpha) on specific airway conductance (sGaw), maximum expiratory flow at 30% vital capacity (Vmax30), forced expiratory volume in 1 s (FEV1), and static lung volumes in subjects with mild allergic asthma. In a second study the effect of inhaled PGI2 on bronchoconstriction provoked by increasing concentrations of inhaled prostaglandin (PG) D2 and methacholine was observed. Inhalation of PGI2 up to a concentration of 500 micrograms/ml had no significant effect on sGaw but produced a concentration-related decrease in FEV1 and Vmax30 in all subjects. In two of four subjects inhalation of PGI2 also increased residual volume and decreased vital capacity but had no effect on total lung capacity. PGI2, but not 6-oxo-PGF1 alpha, protected against bronchoconstriction provoked by either PGD2 or methacholine whether airway caliber was measured as sGaw, FEV1, or Vmax30. The apparent disparity between the bronchoconstrictor and antibronchoconstrictor effects of PGI2 might be explained by its potent vasodilator effect in causing airway narrowing through mucosal engorgement and reducing the spasmogenic effects of other inhaled mediators by increasing their clearance from the airways.     

Signal for term parturition is of trophoblast and therefore of fetal origin

Up to now we know, that cytokines are key intermediates in the mechanisms, responsible for intrauterine activation in case of intra amniotic infection. The aim of our study was to investigate the role of cytokine- and prostaglandin production in normal term labor. Release of Il-6, Il-1β, TNF-, PGE₂, PGF₂ was monitored in vaginal secretions originating from uterine cavity, cervix and vagina during normal course of labor. Cells from fetal membranes, decidua and villous trophoblast, obtained from placentas of patients after spontaneous delivery (n = 12), or without labor, after elective cesarean section (n = 12), were cultured, in order to identify cytokine and prostaglandin producing cells. In all cases, term labor and parturition was associated with strongly elevated cytokine- and prostaglandin concentrations in cervical secretions. Cell culture experiments clearly demonstrated, that cells from villous trophoblast, cultured after spontaneous delivery produced significantly more cytokines and prostaglandins than cells form villous trophoblast, cultured after elective cesarean section. Cells from fetal membranes also produced more Il-6 and PGE2 after labor. In contrast to that, cells from decidua produced similar amounts of cytokines and prostaglandins before and after spontaneuos term labor. Therefore we conclude, that the signal for term labor and delivery is of trophoblast and so of fetal origin.     

Invasive depth of extravillous trophoblast correlates with cellular phenotype: a comparison of intra- and extrauterine implantation sites

During intrauterine human placentation, extravillous trophoblast invades uterine tissues starting with proliferating stem cells at the basement membrane of anchoring villi. Transition to the postproliferative invasive phenotype takes place several cell layers distant. Here we show that in intrauterine pregnancies invasive trophoblast comprises three cellular phenotypes: a. Small spindle-shaped trophoblast cells are found along the whole invasive pathway throughout pregnancy. They are embedded in little heterogeneous extracellular matrix but expose only fibronectin receptors (integrins alpha5beta1, alphavbeta3/5), resulting in a partial integrin-matrix mismatch. b. Large polygonal trophoblast cells are rare in early pregnancy but increase in number towards term. They secrete ample heterogeneous extracellular matrix and expose integrins specifically matching the opposing matrix molecules (integrins alpha6beta4, alpha5beta1). c. Multinucleated giant cells in all stages of pregnancy form a kind of peripheral shell of trophoblast.In contrast to intrauterine pregnancies, in viable tubal pregnancies, Mib-1 expression indicating proliferation, extends deeply into the invasive pathway. Trophoblast cells of the invasive pathway mostly belong to the small spindle-shaped phenotype and secrete little extracellular matrix, mainly fibronectins. At the transition to the second cellular layer of cell columns expression of integrin alpha6beta4 switches to expression of alpha5beta1 and alphavbeta3/5. Viable tubal pregnancies are characterised by a broad overlap of proliferative with invasive phenotype as well as a general integrin-matrix mismatch. The differences in proliferation patterns, cellular phenotype and matrix-integrin co-localisation may well explain the increase of invasiveness of normal extravillous trophoblast from term intrauterine via early intrauterine to viable tubal pregnancies.     

6-Oxo-PGF1alpha and 8-epi-PGF2alpha in human atherosclerotic vascular tissue.

Isoprostanes are a new family of compounds generated by the free radical catalyzed action on arachidonic acid. Formed during oxidation they have been claimed to be a reliable indicator of in vivo oxidation injury. We assessed the amount of 8-epi-PGF2alpha in human surgical specimens as compared to PGI2 (via its stable metabolite 6-oxo-PGF1alpha), the major compound generated by vascular tissue. 8-epi-PGF2alpha is low in normal vascular tissue as is the 8-epi-PGF2alpha/6-oxo-PGF1alpha ratio. The vessels of smokers in general exhibited an increased 8-epi-PGF2alpha (r=0.82) and a decreased 6-oxo-PGF1alpha (r=0.71). The 8-epi-PGF2alpha/6-oxo-PGF1alpha ratio is, not significantly, increased in vessels derived from hyperlipidemics and hypertensives. These findings indicate that lipid peroxidation occurs within the human arterial wall as evidenced by 8-epi-PGF2alpha, probably further decreasing the synthesis of PGI2 and promoting atherogenic mechanisms.     

Effects of long-term supplementation of eicosapentanoic and docosahexanoic acid on the 2-, 3-series prostacyclin production by endothelial cells.

We studied the effects of polyunsaturated fatty, acids such as arachidonic acid [20:4 (n-6)], eicosapentanoic acid [EPA, 20:5 (n-3)], and docosahexanoic acid [DHA, 22:6 (n-3)] on the changes of lipid profiles and prostacyclin production by cultured bovine aortic endothelial cells. The amounts of 6-keto-prostaglandin F1alpha(6-keto-PGF1alpha) and delta17-6-keto-PGF1alpha, non-enzymatic metabolites of prostacyclin (PGI2 and PGI3) in culture medium were measured by gas chromatography/selected ion monitoring. Endothelial cells were supplemented for five passages with arachidonic acid, EPA, or DHA, and the fatty acids of cell lipids and prostacyclin production in cultured medium were quantified. From the fatty acid analysis, the amounts of docosapentaenoic acid [22:5 (n-3)] were significantly increased in EPA-grown cells. In DHA-grown cells, the amounts of EPA were slightly increased compared to control cells. These cells produced similar amounts of PGI2 as the controls, but larger amounts of PGI3 under basal conditions. These findings suggest that EPA, docosapentaenoic acid, and DHA are interconverted to each other, and anti-aggregatory effects of EPA or DHA may be partially due to the stimulation of prostacyclin formation in endothelial cells.     

Influence of prostacyclin and two metabolites on the contractility of cultured rat heart cells

Heart cells were cultured from newborn rats, and the contractile activity (CA) and beating frequency (BF) were recorded using an electrooptical technique. Myocardial cells were found to be highly sensitive to Prostacyclin (PGI2) since a 10(-11) M concentration increased the BF and CA. Increasing the concentration (2.7 x 10(-10) to 2.7 x 10(-8) M) resulted in a dose-dependent decrease in CA and BF. The stable product of the non-enzymatic degradation of PGI2 (6 Keto PGF1 alpha) was found to be completely ineffective, and the stable product of the enzymatic PGI2 metabolism (6 Keto PGE1) exerted only a dose-dependent (10(-6) to 10(-5) M) positive inotropic effect. PGI2 was also effective in the presence of serum instead of culture medium but the decrease in CA was less marked than in culture medium, probably due to protein-binding of the drug. When the CA was decreased by PGI2, perfusion with the intracellular calcium-releasing and phosphodiesterase inhibiting agent, caffeine, reversed the PGI2-induced negative inotropic effect. These results suggest that PGI2 participates in the regulation of the heart cell contractility. Its metabolite 6 Keto PGEI could also influence heart cell contractility but higher concentrations are needed. Moreover myocardial intracellular calcium availability seems to be influenced by PGI2.     

Effect of acetylsalicylic acid on prostacyclin production in trophoblast

In attempt to elucidate whether acetylsalicylic acid (ASA) has an in vivo effect on prostacyclin (PGI2)-like activity released from trophoblast we have evaluated PGI2-like activity in pregnant women scheduled for pregnancy termination after ASA ingestion. Following subjects were studied: Group I: 7 healthy pregnant women who were treated with 1.5 g ASA for two days; Group II: 18 control pregnant women who received placebo for two days. Trophoblast specimens were obtained by legal abortions; PGI2-like activity in trophoblast was measured by the method of Moncada. In normal pregnant women (8-10 weeks gestation) treated with ASA the mean PGI2-like activity of trophoblast significantly decreased compared to the controls. These data indicate that treatment with ASA of early pregnant women might have a harmful effect on trophoblast and the problem should be further explored before allowing the administration of cyclooxygenase inhibiting drugs during early pregnancy.     

Prostacyclin synthesis in irradiated endothelial cells cultured from bovine aorta

Confluent monolayers of bovine aortic endothelial cells were examined 2-72 h after exposure to 0.5-5.0 Gy of 60Co gamma-rays. Accumulation of prostacyclin [PGI2, measured as 6-ketoprostaglandin F1 alpha (6-keto-PGF1 alpha)] in the culture media and PGI2 production stimulated by exogenous arachidonate were correlated with cell detachment and release of lactate dehydrogenase (LDH) activity. Platelet adherence to irradiated and control monolayers also was studied. There were simultaneous time- and dose-dependent increases in cell detachment and in the titers of 6-keto-PGF1 alpha and LDH activity in the culture medium. These changes were evident between 4 and 8 h after 5 Gy or at 24 h after 0.5 Gy. Four hours after 5 Gy, both adherent and detached endothelial cells showed a twofold increase in PGI2 production during a 15-min incubation with arachidonate (10 microM). However, by 72 h this increase was less significant. The accumulation of 6-keto-PGF1 alpha appeared to be related to cell destruction, but radiation also stimulated PGI2 synthesis independent of cell detachment. There was an increased platelet interaction with irradiated monolayers, as a result of platelet adherence to subendothelial matrix exposed after cell detachment. However, irradiation did not alter the nonadherent property of the endothelial cell surface toward platelets.