In vitro effect of trichosanic acid, a major component of Trichosanthes japonica on platelet aggregation and arachidonic acid metabolism in human platelets

The in vitro effect of trichosanic acid (TCA; C18:3, omega-5), a major component of Trichosanthes japonica, on platelet aggregation and arachidonic acid (AA) metabolism in human platelets was studied. TCA dose-dependently suppressed platelet aggregation of platelet rich plasma and washed platelets. TCA decreased collagen (50 micrograms/ml)-stimulated production of thromboxane B2 (TXB2) and 12-hydroxyhepta-decatrienoic acid (HHT) in a dose-dependent manner, while that of 12-hydroxyeicosatetraenoic acid (12-HETE) was rather enhanced. The conversion of exogenously added [14C]AA to [14C]TXB2 and [14C]HHT in washed platelets was dose-dependently reduced by the addition of TCA, while that to [14C]12-HETE was increased. Similar observations were obtained when linolenic acid (LNA; C18:3, omega-3) was used. These results suggest that TCA may decrease TXA2 formation in platelets, probably due to the inhibition of cyclooxygenase pathway, and thereby reduce platelet aggregation.     

15-lipoxygenase metabolites of gamma-linolenic acid/eicosapentaenoic acid suppress growth and arachidonic acid metabolism in human prostatic adenocarcinoma cells: possible implications of dietary fatty acids

Although gammalinolenic acid (GLA) and eicosapentaenoic acid (EPA) have independently been reported to suppress growth of cancer cells, their relative potencies are unknown. To determine the possible attenuating efficacies of dietary GLA or EPA on prostate carcinogenesis, we hereby report the in vitro effects of GLA, EPA and their 15-lipoxygenase (15-LOX) metabolites: 15(S)-HETrE and 15(S)-HEPE, respectively, on growth and arachidonic acid (AA) metabolism in human androgen-dependent (LNCaP) and androgen-independent (PC-3) prostatic cancer cells in culture. Specifically, both cells were preincubated respectively with the above PUFAs. Growth was determined by [3H]thymidine uptake and AA metabolism by HPLC analysis of the extracted metabolites. Our data revealed increased biosynthesis of prostaglandin E2 (PGE2) and 5-hydroxyeicosatetraenoic acid (5(S)-HETE) by both cells. Preincubation of the cells with 15(S)-HETrE or 15(S)-HEPE more markedly inhibited cellular growth and AA metabolism when compared to precursor PUFAs. Notably, 15(S)-HETrE exerted the greatest inhibitory effects. These findings therefore imply that dietary GLA rather than EPA should better attenuate prostate carcinogenesis via its in vivo generation of 15(S)-HETrE, thus warranting exploration.     

Time-dependent inhibition of the cyclooxygenase pathway by 12-hydroperoxy-5,8,10,14-eicosatetraenoic acid

We examined effects of small dose (1 microM or less) of exogenous 12-hydroperoxy-5,8,10,14-eicosatetraenoic acid (12-HPETE) on the formation of cyclooxygenase products from exogenous arachidonic acid (AA) in washed human platelets. With a simultaneous addition of AA, 12-HPETE did not affect the formation of thromboxane (TX)B2 and 12-hydroxy-5,8,10-heptadecatrienoic acid (HHT). However, by being preincubated with platelets before an addition of AA, 0.1 microM or greater of 12-HPETE inhibited the formation of TXB2 and HHT dose-dependently. In addition, the inhibitory effect of 12-HPETE increased as the preincubation time was prolonged. These results suggest that 12-HPETE is a strong inhibitor for the cyclooxygenase pathway.     

Effect of tert-butyl hydroperoxide on cyclooxygenase and lipoxygenase metabolism of arachidonic acid in rabbit platelets

The effect of tert-butyl hydroperoxide (t-BOOH) on the formation of thromboxane (TX) B2, 12-hydroxy-5,8,10-heptadecatrienoic acid (HHT) and 12-hydroxy-5,8,10,14-eicosatetraenoic acid (12-HETE) from exogenous arachidonic acid (AA) in washed rabbit platelets was examined. t-BOOH enhanced TXB2 and HHT formation at concentrations of 8 microM and below, and at 50 microM it inhibited the formation, suggesting that platelet cyclooxygenase activity can be enhanced or inhibited by t-BOOH depending on the concentration. t-BOOH inhibited 12-HETE production in a dose-dependent manner. When the platelets were incubated with 12-hydroperoxy-5,8,10,14-eicosatetraenoic acid (12-HPETE) instead of AA, t-BOOH failed to inhibit the conversion of 12-HPETE to 12-HETE, indicating that the inhibition of 12-HETE formation by t-BOOH occurs at the lipoxygenase step. Studies utilizing indomethacin (a selective cyclooxygenase inhibitor) and desferrioxamine (an iron-chelating agent) revealed that the inhibitory effect of t-BOOH on the lipoxygenase is not mediated through the activation of the cyclooxygenase and that this effect of t-BOOH is due to the hydroperoxy moiety. These results suggest that hydroperoxides play an important role in the control of platelet cyclooxygenase and lipoxygenase activities.     

Covalent binding of arachidonic acid metabolites to human platelet proteins. Identification of prostaglandin H synthase as one of the modified substrates

The covalent modification of proteins by metabolites of arachidonic acid (AA) was investigated in human platelets. Following incubation of washed human platelets with radiolabeled AA, ethanol precipitation of the proteins, and lipid extraction by organic solvents, a small fraction of the radioactivity added (0.3%) was tightly bound to the protein pellet. A dozen labeled protein bands were resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Exhaustive hydrolysis of platelet proteins by proteases released an amphipathic radiolabeled material which had a chromatographic behavior similar to that of a known peptidolipid, leukotriene C4. These findings suggest a covalent nature for the observed binding. This binding was specific for AA since palmitate, myristate, or linoleate did not bind to a significant extent. It involved products of both cyclooxygenase and lipoxygenase pathways: it was indeed inhibited to a greater extent by eicosatetraynoic acid than by indomethacin. The protein-associated radioactivity was increased by the thromboxane synthase inhibitor dazoxiben. Indomethacin completely abolished this increase in binding, which could not be reproduced by exogenous prostaglandin (PG) E2, F2 alpha, or D2, and might thus involve PGG2 and/or PGH2. Diamide, an agent known to inhibit the reduction of 12-hydroperoxyeicosatetraenoic acid in platelets, produced an increase of the covalent binding, which was abolished by eicosatetraynoic acid but not by indomethacin: this suggests that the lipoxygenase product bound was 12-hydroperoxyeicosatetraenoic acid or a by-product. Dazoxiben and diamide produced distinct patterns of protein labeling after sodium dodecyl sulfate-polyacrylamide gel electrophoresis. One labeled band had a Mr of 70,000 as the PGH synthase monomer. Addition of AA at 17 microM enhanced the labeling of this band, while 100 microM was inhibitory. Labeling of this band was also induced by thrombin in prelabeled platelets. Two monoclonal antibodies against PGH synthase caused immune precipitation of a 70-kDa labeled protein in homogenates of [3H]AA-labeled platelets. PGH synthase, purified from ram seminal vesicles, was covalently modified after incubation with [3H]AA: this labeling was almost completely abolished by indomethacin. As much as 40% of platelet PGH synthase was covalently modified after incubation with 17 microM AA. It can be concluded that in intact platelets PGH synthase is covalently modified by an eicosanoid following incubation with exogenous AA or after AA mobilization from phospholipids by thrombin.     

An improved malondialdehyde assay for estimation of thromboxane synthase activity in washed human blood platelets

After stimulation of the washed human blood platelets by arachidonic acid (AA), the concurrent evaluations for formed malondialdehyde (MDA) measured by the common photometrical thiobarbituric acid (TBA) method, and for thromboxane B2 (TXB2) measured by gas chromatography, revealed that the formed MDA exceeded the amount of TXB2 on a molar base. However, MDA and TXB2 originating from thromboxane synthase activity should be produced in approximately equimolar amounts. By treatment of the stimulated platelet samples with stannous chloride it is possible to reduce all peroxidized products of AA which generate MDA otherwise during the TBA reaction and to estimate MDA and TXB2 in a ratio of nearly 1:1. The stannous chloride treatment does not destroy the MDA and does not influence the TBA reaction with MDA. Therefore the simple and quick TBA method can be used after stannous chloride treatment for estimation of thromboxane synthase activity in AA stimulated washed human platelets.     

Stimulation of lipoxin synthesis from leukotriene A4 by endogenously formed 12-hydroperoxyeicosatetraenoic acid in activated human platelets

Human platelets are devoid of 5-lipoxygenase activity but convert exogenous leukotriene A₄ (LTA₄) either by a specific LTC₄ synthase to leukotriene C₄ or via a 12-lipoxygenase mediated reaction to lipoxins. Unstimulated platelets mainly produced LTC₄, whereas only minor amounts of lipoxins were formed. Platelet activation with thrombin, collagen or ionophore A23187 increased the conversion of LTA₄ to lipoxins and decreased the leukotriene production. Maximal effects were observed after incubation with ionophore A23187, which induced synthesis of comparable amounts of lipoxins and cysteinyl leukotrienes (LTC₄, LTD₄ and LTE₄). Chelation of intra- and extracellular calcium with quin-2 and EDTA reversed the ionophore A23187-induced stimulation of lipoxin synthesis from LTA₄ and inhibited the formation of 12-hydroxyeicosatetraenoic acid (12-HETE) from endogenous substrate. However, calcium did not affect the 12-lipoxygenase activity in the 100 000 × g supernatant of sonicated platelet suspensions. Furthermore, the stimulatory effect on lipoxin formation induced by platelet agonists could be mimicked in intact platelets by the addition of low concentrations of arachidonic acid, 12-hydroperoxyeicosatetraenoic acid (12-HPETE) or 13-hydroperoxyoctadecadienoic acid (13-HPODE). The results indicate that the elevated lipoxin synthesis during platelet activation is due to stimulated 12-lipoxygenase activity induced by endogenously formed 12-HPETE.     

Inhibition of platelet arachidonic acid 12-lipoxygenase by acetylenic acid compounds

Eighteen acetylenic fatty acids were tested as inhibitors of human platelet arachidonic acid 12-lipoxygenase. 4,7,10,13-Eicosatetraynoic (4,7,10,13-ETYA) acid emerged as the most potent compound. Additional experiments have shown that 4,7,10,13-ETYA selectively blocked the 12-lipoxygenase in washed human platelets with lesser activity against the cyclooxygenase. The ID50 value for lipoxygenase was 7.8 microM in comparison with an ID50 of 100 microM for the cyclooxygenase. The commonly used inhibitor 5,8,11,14-eicosatetraynoic acid inhibited both enzymes with equal potency. It appears that 4,7,10,13-ETYA may be a valuable lead for selective modulation of the 12-lipoxygenase pathway in platelet or other target tissues.     

The metabolism of arachidonic and eicosapentaenoic acids in human neutrophils stimulated by A23187 and FMLP

A23187 stimulates the metabolism of endogenous as well as exogenous arachidonic acid (AA) and eicosapentaenolc acid (EPA) to their corresponding leukotrienes in human neutrophils. In contrast, conflicting results have been obtained concerning the effect of FMLP on the metabolism of these fatty acids. In the present study we compared the effect of A23187 and FMLP on the release and metabolism of these fatty acids in neutrophils. Stimulation of neutrophils with A23187, but not with FMLP, resulted in detectable levels of AA in the presence or absence of BW755C (a dual inhibitor of cyclooxygenase and lipoxygenase). The absolute amount of nonesterified AA in the extracts of neutrophils exposed to the agonist A23187 in the presence of BW755C was 20% higher than that obtained in the absence of BW755C, indicating that only a small fraction of the released AA was converted to lipoxygenase products. Furthermore, significant quantities of AA and EPA metabolites were detected only after treatment of neutrophils with A23187, but not with FMLP. Both A23187 and FMLP stimulated the conversion of exogenous EPA to 5-lipoxygenase products, with A23187 being somewhat more effective. In addition, significant differences were noted on the effect of EPA and DHA on the conversion of AA to its metabolites in A23187-stimulated neutrophils. Our results provide strong evidence that the amounts of eicosanoid precursors mobilized in response to FMLP are extremely small, if any, and this appears to be the likely explanation for the lack of eicosanoid detection by HPLC in FMLP-stimulated neutrophils.     

Manipulation of platelet aggregation by prostaglandins and their fatty acid precursors: pharmacological basis for a therapeutic approach

Addition of the one-, two- or three- series endoperoxide to human platelet-rich plasma tend to suppress aggregation, through the action of their respective non-enzymatic breakdown products PGE1, PGD2, or PGD3 all of which elevate cyclic AMP levels. On the other hand, these stable primary products do not arise in appreciable amounts from intrinsic endoperoxides generated from either endogenous or exogenous free fatty acids. 5,8,11,14,17-Eicosapentaenoic acid (EPA) suppresses arachidonic acid (5,8,11,14-eicosatetraenoic acid) conversion by cyclooxygenase (as well as lipoxygenase) to aggregatory metabolites in platelets. Exogenously added EPA was capable of inhibiting PRP aggregation induced either by exogenous or endogenous (released by ADP or collagen) arachidonate. The hypothetical combination of an EPA-rich diet and a thromboxane synthetase inhibitor might abolish production of the pro-aggregatory species, thromboxane A2, and enhance formation of the anti-aggregatory metabolite, prostacyclin. Whereas EPA is not detectably metabolized by platelets, dihomo-gamma-linolenic acid (8,11,14-eicosatrienoic acid) is primarily converted by cyclooxygenase and thromboxane synthetase into the inactive metabolite, 12-hydroxyheptadecadienoic (HHD) acid. Pretreatment of human platelet suspensions with the thromboxane synthetase inhibitor imidazole unmasks the aggregatory property of PGH1 and DLL which was partially compromised by the PGE1 formed. The combination of the thromboxane synthetase inhibitor and an adenylate cyclase inhibitor unmasks a complete irreversible aggregation by DLL or PGH1. The basis of a dietary strategy that replaces AA with DLL must rely on the production by the platelet of an inactive metabolite (HHD) rather than thromboxane A2.     

Monochloramine potently inhibits arachidonic acid metabolism in rat platelets

In the present study, the effects of hypochlorous acid (HOCl), monochloramine (NH(2)Cl), glutamine-chloramine (Glu-Cl) and taurine-chloramine (Tau-Cl) on the formation of 12-lipoxygenase (LOX) metabolite, 12-HETE, and cyclooxygenase (COX) metabolites, TXB(2), and 12-HHT, from exogenous arachidonic acid (AA) in rat platelets were examined. Rat platelets (4x10(8)/ml) were preincubated with drugs for 5min at 37 degrees C prior to the incubation with AA (40microM) for 2min at 37 degrees C. HOCl (50-250microM) showed an inhibition on the formation of LOX metabolite (12-HETE, 5-67% inhibition) and COX metabolites (TXB(2), 33-73% inhibition; 12-HHT, 27-74% inhibition). Although Tau-Cl and Glu-Cl up to 100microM were without effect on the formation of 12-HETE, TXB(2) and 12-HTT, NH(2)Cl showed a strong inhibition on the formation of all three metabolites (10-100microM NH(2)Cl, 12-HETE, 21-92% inhibition; TXB(2), 58-94% inhibition; 12-HHT, 36-92% inhibition). Methionine reversed a reduction of formation of LOX and COX metabolites induced by NH(2)Cl, and taurine restoring that induced by both NH(2)Cl and HOCl. These results suggest that NH(2)Cl is a more potent inhibitor of COX and LOX pathways in platelets than HOCl, and taurine and methionine can be modulators of NH(2)Cl-induced alterations in the COX and LOX pathways in vivo.     

Manipulation of platelet aggregation by prostaglandins and their fatty acid precursors: Pharmacological basis for a therapeutic approach

Addition of the one-, two- or three- series endoperoxide to human platelet-rich plasma tend to supress aggregation, through the action of their respective non-enzymatic breakdown products PGE₁, PGD₂, or PGD₃ all of which elevate cyclic AMP levels. On the other hand, these stable primary products do not arise in appreciable amounts from intrinsic endoperoxides generated from either endogenous or exogenous free fatty acids. 5,8,11,14,17-Eicosapentaenoic acid (EPA) suppresses arachidonic acid (5,8,11,14-eicosatetraenoic acid) conversion by cycloogygenase (as well as lipoxygenase) to aggregatory metabolites in platelets. Exogenously added EPA was capable of inhibiting PRP aggregation induced either by exogenous or endogenous (released by ADP or collagen) arachidonate. The hypothetical combination of an EPA-rich diet and a thromboxane synthetase inhibitor might abolish production of the pro-aggregatory species, thromboxane A₂, and enhance formation of the anti-aggregatory metabolite, prostacyclin.Whereas EPA is not detectably metabolized by platelets, dihomo-γ-linolenic acid (8,11,14,-eicosatrienoic acid) is primariley converted by cyclooxygenase and thromboxane synthetase into the inactive metabolite, 12-hydroxyheptadecadienoic (HHD) acid. Pretreatment of human platelet suspensions with the thromboxane synthetase inhibitor imidazole unmasks the aggregatory property of PGH₁ and DLL which was partially compromised by the PGE₁ formed. The combination of the thromboxane synthetase inhibitor and an adenylate cyclase inhibitor unmasks a complete irreversible aggregation by DLL or PGH₁. The basis of a dietary strategy that replaces AA with DLL must rely on the production by the platelet of an inactive metabolite (HHD) rather than thromboxane A₂.     

Inhibition of human leukocyte 5-lipoxygenase by 15-HPETE and related eicosanoids

The inhibition of human leukocyte 5-lipoxygenase by 15-hydroperoxyeicosatetraenoic acid and its chemical or enzymatic rearrangement products was investigated. 15-Hydroperoxyeicosatetraenoic acid was the most potent inhibitor tested. The inhibition was found to be time dependent and is not due to chemical or enzymatic decomposition products nor metabolism of 15-hydroperoxyeicosatetraenoic acid to 5,15-dihydroperoxyeicosatetraenoic acid.     

Eicosapentaenoic acid metabolism in human and rabbit anterior uvea

Eicosapentaenoic acid (EPA) metabolism into 3 series cyclooxygenase and 5 series lipoxygenase products was assessed in human and rabbit anterior uvea. Both tissues synthesized 3 series cyclooxygenase products such as delta17 6-keto-PGF1 (PGI3 metabolite), PGE3 alpha, PGE3, PGD3 and TxB3 (a stable product of TxA3) and lipoxygenase products 12-hydroxyeicosapentaenoic acid (HEPE), 5-HEPE and 5,12-diHEPE from 14C-EPA. EPA-derived cyclooxygenase product synthesis was considerably greater than the formation of lipoxygenase products from EPA in both tissues.     

A comparative study of eicosapentaenoic acid metabolism by human platelets in vivo and in vitro

During long-term dietary n-3 fatty acid supplementation, eicosapentaenoic acid (EPA) is not incorporated into phosphatidylinositol or -serine of human platelets in vivo and is not detectable in phosphatidic acid upon stimulation with thrombin. However, EPA is released from platelet phospholipids and metabolized to thromboxane B3 (TXB3). In contrast, in vitro, platelets incorporate [14C]EPA into phosphatidylinositol, whether they contain endogenous EPA in their cellular lipids or not. Following platelet stimulation, [14C]EPA appears in phosphatidic acid, as free fatty acid, and is transformed to TXB3. We conclude that the fatty acid compositions of platelet phospholipid subclasses are regulated with a high degree of specificity in vivo. Qualitative differences exist between in vivo and in vitro uptake of EPA into platelet phospholipid subclasses. After in vivo incorporation, EPA is released by action of a phospholipase A2.     

Optimization of an assay for studying the effects of agents on cyclooxygenase and lipoxygenase metabolism of arachidonic acid in washed human platelets

Before one can examine the effects of substances on the metabolism of arachidonic acid (AA) by the cyclooxygenase and lipoxygenase pathways, an assay system which allows one to detect increases or decreases in both pathways in needed. In order to develop such a system, we have examined nonaggregating washed human platelets (10(8) platelets/0.5 ml) incubated for various times with 2 microCi 3H-AA and increasing concentrations of AA. T/B2, HHT, 12-HETE, and AA were extracted and separated using reverse phase-HPLC. We first calculated the mass of AA products formed with 10(-7) to 10(-4) M AA and found that the cyclooxygenase was saturated with 10(-5) M AA whereas the lipoxygenase was not saturated with 10(-4) M AA. Cyclooxygenase products were more prevalent than 12-HETE below 10(-5) M AA, while lipoxygenase products predominated at 3 x 10(-5)-10(-4) M AA. Using 3 microM AA, which does not saturate the cyclooxygenase, we examined the effect of 0.25-10 minute incubation durations on the distribution of AA metabolites and found AA product formation to increase throughout this period without completely depleting the substrate. Since substrate depletion does not occur and further metabolism could be detected for both pathways with a 5 minute incubation with 3 microM AA, these incubation parameters were chosen in order to further test the assay system. Using these parameters, we found that 10(-4) M 5-hydroxytryptamine enhanced platelet 12-HETE formation and decreased T/B2 and HHT formation, thus demonstrating the capacity of this system to simultaneously detect changes in cyclooxygenase and lipoxygenase enzyme metabolism.(ABSTRACT TRUNCATED AT 250 WORDS)     

Uptake, release and metabolism of docosahexaenoic acid (DHA, c22:6 omega 3) in human platelets and neutrophils

Exogenous DHA is converted by human platelets to 14- and 11- HDHE and by human neutrophils mainly to 7- HDHE . Human platelets prelabeled with 14C-DHA, 14C-EPA and 14C-AA and stimulated with thrombin release and metabolize DHA only in trace amounts as compared to EPA and AA. 14C-DHA is incorporated into the 2-position of platelet phospholipids and occurs predominantly in phosphatidylethanolamine. DHA and EPA were also incorporated by dietary means into phospholipids of platelets and neutrophils. In resting platelets free DHA as well as free AA and EPA are not detectable. In platelets stimulated ex vivo with thrombin DHA is not significantly released which is in contrast to EPA and AA. After stimulation, 14- HDHE is found only in trace amounts as compared to 12-HETE and 12- HEPE . In DHA enriched neutrophils formation of HDHEs cannot be demonstrated after stimulation with ionophore A 23187. We conclude that even after dietary enrichment of DHA in phospholipids of platelets and neutrophils the level of free DHA and/or formation of HDHEs might be too low to substantially affect arachidonic acid metabolism and related functions of these cells.     

The platelet cyclooxygenase metabolite 12-L-hydroxy-5, 8, 10-hepta-decatrienoic acid (HHT) may modulate primary hemostasis by stimulating prostacyclin production

Although HHT accounts for approximately one third of the arachidonic acid (AA) metabolites produced by stimulated platelets, no well defined function has been attributed to this product. We report that HHT stimulates prostacyclin production by endothelial cells, and have identified the mechanism for this effect. In human umbilical venous endothelial cells, HHT (0.5 and 1 microM) stimulated prostacyclin (RIA for 6KPGF1 alpha) by 32 +/- 22% (1SD) and 42 +/- 38% (P less than 0.05 and less than 0.01). Similar changes were observed when the effect of HHT on exogenous [1-14C] AA metabolism in fetal bovine aortic endothelial cells (FBAECs) was studied. Kinetic analyses revealed that HHT affected vascular cyclooxygenase. HHT (1 microM) increased Vmax in test microsomes (706 +/- 21 pmol/mg/min, mean +/- 1SE) when compared to controls (529 +/- 20; P less than 0.02). No concomitant effect on Km was observed. A further effect of HHT on AA release from endothelial cell membrane phospholipids was noted. Prelabeling experiments revealed that HHT (1 microM) increased the ionophore stimulated release of AA from FBAECs (20952 +/- 555 cpm/well control mean +/- 1SE vs 25848 +/- 557 for paired HHT treated cells; P less than 0.05). The effect of HHT on platelet AA metabolism was next studied. Preincubation of washed platelets with HHT (1 microM) did not enhance thrombin or arachidonic acid induced platelet TXB2 formation. In platelets prelabelled with [1-14C]AA, HHT (1 microM) had no effect on AA release post thrombin stimulation. Conversion to cyclooxygenase metabolites was also not enhanced. HHT stimulates vascular prostacyclin without a concomitant effect on platelet AA metabolism. HHT may thus be an important local modulator of platelet plug formation.     

Conversion of arachidonic acid into 12-oxo derivatives in human platelets. A pathway possibly involving the heme-catalysed transformation of 12-hydroperoxy-eicosatetraenoic acid

Analysis of arachidonic acid metabolites in human platelets by reverse-phase HPLC with radioactivity and UV detection revealed, besides Thromboxane B2 (TXB2), 12-hydroxy-heptadecatrienoic acid (HHT) and 12-hydroxy-eicosatetraenoic acid (12-HETE) previously described, two peaks of unidentified material absorbing at 280 nm. This material was purified by straight-phase HPLC and characterized by UV spectroscopy and gas chromatography-mass spectrometry. Three carbonyl compounds were identified: 12-keto-5,8,10,14-eicosatetraenoic acid and two geometric isomers of 12-oxo-5,8,10-dodecatrienoic acid. In a 5 min incubation at 37 degrees C in the presence of 9 microM arachidonic acid, the yield was of 0.5 to 1% of added arachidonic acid for the ketonic compound and of 4 to 7% for the sum of the two isomeric fatty acid aldehydes in comparison to 10 to 13% and 25 to 28% for TXB2 and 12-HETE, respectively. Because the three compounds carry a carbonyl group at position 12, their relationship with the 12-lipoxygenase pathway was investigated. It was found that the three compounds were formed when 12-hydroperoxy-eicosatetraenoic acid (12-HPETE) was incubated with intact or heat denaturated platelets or hemoproteins, strongly suggesting that these carbonyl compounds are products of a heme-catalysed transformation of 12-HPETE.