Search

Fatty acid hydroperoxides (lipoxygenase products) are metabolized to allene oxides by a type of dehydrase that has been detected in plants, corals, and starfish oocytes. The allene oxides are unstable epoxide precursors of more complex products such as jasmonic acid, the plant growth hormone. Characterization of the dehydrase enzyme of flaxseed revealed that it is a 55-kilodalton hemoprotein. The spectral characteristics of this dehydrase revealed it to be a cytochrome P-450. It operates with the remarkable activity of greater than or equal to 1000 turnovers per second. The results establish a new catalytic activity for a cytochrome P-450 and illustrate the cooperation of different oxygenases in pathways of fatty acid metabolism.

One of the many changes induced by topical application of phorbol ester or calcium ionophore A23187 to mouse skin is the appearance of an enzymic activity which will convert arachidonic acid to its 8-hydroxyeicosatetraenoic acid metabolite (8-HETE) (Gschwendt, M., et al (1986) Carcinogenesis 7, 449-455). Induction of this activity is lower in strains of mice with a weak inflammatory response to TPA, and the 8-HETE may be involved in the inflammation or hyperplasia. To further characterize the activity, we first measured the chirality of the product; it is almost exclusively the 8DS)-hydroxy enantiomer (8S-HETE). The 8(S)-HETE is formed from octadeuterated arachidonic acid with complete retention of deuterium labels, indicating that a keto intermediate is not involved in the biosynthesis. Using arachidonic acids labeled with a prochiral tritium in either the 10DR or 10LS positions, we found that the biosynthesis of 8S-HETE is associated with the stereoselective abstraction of the 10DR hydrogen from the 10-carbon of the substrate. This stereoselective hydrogen removal conforms to the properties of an 8S-lipoxygenase. This is the only lipoxygenase known to catalyze solely 8S-oxygenation of arachidonic acid. The recent characterization of stereoselective biological effects for other HETEs serve as strong precedents to suggest that 8S-HETE has a specific role in the cellular tissue response to TPA.

We find that the reticulocyte lipoxygenase can oxygenate rat liver mitochondrial membranes, beef heart submitochondrial particles, rat liver endoplasmic membranes, and erythrocyte plasma membranes (inside-out and right side-out ghosts) without prior action of a phospholipase. After alkaline hydrolysis of the ester lipids, the main products were identified as 15S-hydro(pero)xy-5Z,8Z,11Z,13E-eicosatetr aenoic acid, 17S-hydro(pero)xy-4Z,7Z,10Z,13Z,15E, 19Z,-docosahexaenoic acid, 13S-hydro(pero)xy-9Z,11E-octadecadienoic acid, 9(S/R)-hydro(pero)xy-10E,12Z-octadecadienoic acid as well as the two all-E hydro(pero)xy octadecadienoic acid isomers. At low membrane concentrations (1 mg of protein/ml), the enzyme maintains a high stereospecificity for the S-configuration, but at higher concentrations (20 mg/ml), the products were virtually racemic. Addition of the antioxidant 2,6-ditert-butyl-p-cresol counteracted this tendency to lose stereospecificity. During these enzyme-catalyzed reactions, substantially more oxygen is consumed than can be accounted for as the hydro(pero)xy products. This discrepancy is due to secondary reactions which lead to the decomposition of the primary oxygenation products, the hydroperoxy lipids, and to oxidative modifications of membrane proteins. These data indicate that the reticulocyte lipoxygenase can oxygenate polyenoic fatty acids in various types of biological membrane and that the oxidative modifications are not restricted to the membrane lipids. The results are discussed in terms of the proposed role of the enzyme in the breakdown of mitochondria and other intracellular organelles during the maturation of red blood cells.

We tested a simple model which explains the singular or dual specificity of lipoxygenases. The dual specificity considered here is typified by the oxygenation of arachidonic acid by the reticulocyte lipoxygenase: two chiral products are formed (12S- and 15S-hydroperoxides, ratio approximately 1:9) via hydrogen abstraction from two separate methylene groups (C-10 and C-13). The rate-limiting step is known to involve this hydrogen abstraction, and we assumed that alignment of the methylenes with the hydrogen acceptor on the enzyme is critical in terms of reaction rate and positional specificity. Optimal alignment will be associated with a fast rate of reaction and formation of a single chiral product. A shift in position of the double bonds (and hence of the methylene groups) should be associated with a slower rate of reaction and formation of two chiral products; two methylenes are now able to react, although neither has perfect alignment. We tested this idea using two lipoxygenases and polyenoic fatty acids differing in the number and position of the double bonds. Optimal substrates for the soybean lipoxygenase had a doubly allylic methylene in the n-8 position, while the reticulocyte enzyme preferred substrates with a n-9 methylene. These substrates were converted to a single chiral product. With both enzymes, the other series of substrates reacted more slowly and were converted to two chiral products. We conclude that alignment of methylene groups of the substrate at the active site is a major determinant of the reaction rate and the singular or dual specificity of lipoxygenases.

One potential biosynthetic route to the prostaglandins involves the participation of lipoxygenase and allene oxide synthase enzymes, giving a hydroxylated allene oxide, which then might cyclize to form prostaglandin A or a close analogue. We have tested a model of this type of transformation using 8-hydroxy-15S-hydroperoxy eicosanoids as substrates for the dehydrase (allene oxide synthase) in flaxseed. Four of these substrates, each with a 9E,11Z,13E-conjugated triene, gave an observable rate of reaction. The two derived from eicosapentaenoic acid reacted more rapidly than the corresponding arachidonic acid analogues. Also, the 8S-hydroxy-15S-hydroperoxy diastereomers reacted more rapidly than their 8R-hydroxy analogues. Products were characterized by high pressure liquid chromatography, UV, gas chromatography-mass specrometry, 1H NMR, and CD. Reaction of the (8S)-hydroxy-(15S)-hydroperoxy-eicosapentaenoic acid gave two alpha-ketols [8S),15-dihydroxy-14-oxoeicosa-5Z,9E,11Z,17Z+ ++-tetraenoic acid and the corresponding 11E isomer in a 2:1 ratio), together with four prostaglandin A3 analogues which differed in the configurations of the side chains. Oxygen 18 labeling fully supported the intermediacy of an allene oxide in the biosynthesis. The corresponding "8R" substrate was converted to the enantiomers of these products plus three 13-hydroxy-14,15-epoxy derivatives. The arachidonate analogues formed the epoxy-hydroxy derivatives, the alpha-ketols, and two prostaglandin A2 analogues with trans configuration of the side chains. These results demonstrate (i) a feasible route of metabolism of lipoxygenase products to hydroxy allene oxide, (ii) the potential for the resulting allene oxide to cyclize to a prostaglandin A analogue, and (iii) the marked influence of the hydroxyl configuration of the rate of reaction and resulting profile of products. Some of these reactions may occur in a natural pathway of prostanoid biosynthesis in corals and other organisms.

A lipoxygenase has been found in the reticulocytes of all mammalian species tested so far (rabbit, rat, mouse, monkey, and humans); evidence from in vitro studies suggests that the lipid-peroxidizing effects of this enzyme could render the mitochondrion and other intracellular organelles prone to the proteolytic degradation which is a natural step in development of the reticulocyte to the mature red cell. In this study we sought evidence of an active lipoxygenase in vivo. A bleeding anemia was induced in rabbits, and in the course of the subsequent reticulocytosis the red cell membranes were examined for the presence of the characteristic lipoxygenase products of linoleic and arachidonic acids. Erythrocyte membranes from control collections contained only small amounts of hydroxy fatty acids (0.03-0.08% of the polyenoic fatty acids). In contrast, reticulocyte-enriched red cells contained up to 3.3% of the polyenoic acids as hydroxylated derivatives. The main hydroxy fatty acid in reticulocyte membranes was identified as 13-L(S)-hydroxy-9Z,11E-octadecadienoic acid. Small amounts of other hydroxy derivatives including 15-hydroxy-5,8,11,13-(Z,Z,Z,E)eicosatetraenoic acid were also detected. These products appeared about 3 days after development of reticulocytosis. The precise structures of the hydroxylated polyenoic fatty acids and the time course of their appearance strongly suggest that their formation is due to the intracellular action of the cell-specific reticulocyte lipoxygenase. These findings are the first evidence for an activity of this enzyme in vivo, and the results support the hypothesis that enzymic peroxidation of reticulocyte intracellular membranes is a step in preparation of the intracellular organelles for proteolytic degradation.

The consequences of inhibiting the metabolism of prostaglandin G2 to thromboxane A2 in man were studied by using an inhibitor of thromboxane synthase, 4-[2-(IH-imidazol-1-yl)ethoxy] benzoic acid hydrochloride (dazoxiben). Single doses of 25, 50, 100, and 200 mg of dazoxiben were administered to healthy volunteers at 2-wk intervals in a randomized, placebo-controlled, double-blind manner. Serum thromboxane B2 and aggregation studies in whole blood and platelet-rich plasma were measured before dosing and at 1, 4, 6, 8, and 24 h after dosing. Both serum thromboxane B2 and the platelet aggregation response to arachidonic acid (1.33 mM) were reversibly inhibited in a dose-dependent manner. Aggregation induced by 1-O-alkyl-2-acetyl-sn-glycero-3-phosphocholine (0.4 and 4.0 microM) in platelet-rich plasma as well as both aggregation and nucleotide release induced by collagen (95 micrograms/ml) in platelet-rich plasma and whole blood were unaltered by dazoxiben. Additional evidence for a platelet-inhibitory effect of the compound was a significant prolongation of the bleeding time at 1 h after administration of the highest dose (200 mg) of dazoxiben. Endogenous prostacyclin biosynthesis was assessed by measurement of the major urinary metabolite of prostacyclin, 2,3-dinor-6-keto-PGF1 alpha (PGI-M). PGI-M excretion was increased by dazoxiben; it rose a mean 2.4-fold from predosing control values at 0-6 h after administration of the highest dose studied (200 mg).

Recently, oocytes or eggs of two marine invertebrates have been found to metabolize arachidonic acid to specific monohydroxy products. These studies have prompted our examination of the oocytes of higher organisms. In the present study, oocytes of an amphibian, Xenopus laevis, were examined for their capacity to biosynthesize hydroxyeicosatetraenoic acids (HETEs) and related hydroxy fatty acids. Two hydroxyeicosanoids were formed during incubations of oocyte homogenates with [14C]arachidonic acid; their structures and stereochemistry were determined by high-pressure liquid chromatography, uv spectroscopy, and gas chromatography-mass spectrometry. The compounds were identified as 15(S)- and 12(S)-hydroxyeicosatetraenoic acids. The synthesis of the two HETEs was not blocked by a cyclooxygenase inhibitor, indomethacin (10 microM), or by prior exposure of the oocyte homogenates to carbon monoxide, an inhibitor of cytochrome P450. Furthermore, 12(S)- and 15(S)-hydroperoxyeicosatetraenoic acids were isolated from brief incubations of gel-filtered ammonium sulfate fraction of frog oocyte homogenates; isolation of the hydroperoxide is further support for the existence of 12(S)- and 15(S)-lipoxygenase activities in the oocytes of X. laevis. Other polyunsaturated acids, including C18.2, C18.3, C20.3, C20.5, and C22.6 were also substrates for the lipoxygenase, and in each case the major product was formed by omega 6 oxygenation.

Urinary excretion of 2, 3 dinor-6-keto-PGF1 alpha, a metabolite of prostacyclin (PGI2), was measured in 9 patients with Bartter's syndrome. The rate of excretion of this metabolite was normal in these patients during ingestion of both a normal and high dietary intake of potassium. This suggests that in Bartter's syndrome the rate of entry of PGI2 into the circulation is normal. Excessive systemic synthesis of PGI2 is therefore unlikely to be an explanation for either the vascular insensitivity to angiotensin II or the defect in platelet aggregation characteristic of the syndrome.

We investigated prostacyclin (PGI2) biosynthesis during pregnancy by measuring urinary excretion of 2,3-dinor-6-keto-prostaglandin F1 alpha (dinor) and 15-keto-13,14-dihydro-2,3-dinor-6-keto-prostaglandin F1 alpha (15 kd dinor) with the use of specific gas chromatography-mass spectrometry assays. Nine normotensive nonpregnant women, five normotensive women in the mid-trimester of pregnancy, eight normotensive women in the third trimester of pregnancy, and six women who developed hypertension during the third trimester provided 24-hour samples of urine. Normal pregnant women had a fivefold increase in urinary excretion of dinor in comparison to nonpregnant women (253 +/- 21 ng dinor/gm creatinine for controls vs. 1,224 +/- 110 and 1,127 +/- 152 for second and third trimesters) (mean +/- SEM). Pregnant subjects with hypertension had a significant (50%) reduction in urinary dinor excretion in comparison to normotensive pregnant subjects (561 +/- 105 ng dinor/gm creatinine). In subjects selected from each group, the ratio of dinor to 15 kd dinor remained constant. We conclude that PGI2 biosynthesis is increased during normal pregnancy, and that this increase is less in pregnancy-induced hypertension. This raises the possibility that PGI2 helps mediate hemodynamic changes during normal pregnancy, and that a relative decrease in production might be related to the pathogenesis of pregnancy-induced hypertension.

An investigation was undertaken to compare the responses of human neutrophils to the epoxide leukotriene A4 with those elicited by its stable product leukotriene B4 under identical IN VITRO conditions. LTA4 evokes neutrophil responses similar in nature to those induced by LTB4 but at much higher concentrations. Evidence suggests that LTA4 is important primarily for its role as an intermediate rather than for inherent activity.

5beta,7alpha-Dihydroxy-11-oxotetranor-prostane-1,16-dioic acid has been identified by gas chromatography-mass spectrometry as a urinary metabolite of [9beta-3H]prostaglandin F2alpha in the rat. This tetranor prostaglandin F derivative, which is the 5beta epimer of the major urinary metabolite of prostaglandin F2alpha, accounted for at least 2% of the total dose. Absence from the metabolite of tritium label at the C-5 position indicated the existence of a minor, previously unknown metabolic pathway by which prostaglandin Falpha derivatives may be converted by oxido-reduction into prostaglandins of Fbeta stereochemistry.

A new method was developed for analyzing the steric configuration of hydroxyeicosatetraenoates (HETEs) and other hydroxy fatty acids. Racemic HETE methyl esters were reacted with either benzoyl or naphthoyl chloride in pyridine and the resulting aromatic ester derivatives purified by reversed phase HPLC and subsequently chromatographed on a chiral stationary phase HPLC column [(R)-(-)-N-3,5-dinitrobenzoyl-alpha-phenylglycine)]. In contrast to the enantiomers of the underivatized HETE methyl esters which were only partially resolved, the enantiomers of their aromatic ester derivatives were completely separable on this chiral phase. Chiral HETEs can be retrieved from the aromatic derivatives by alkaline hydrolysis. Thus, this method has both analytical and preparative applications.

Lipoxygenases react with hydroperoxy fatty acids and catalyze dioxygenase or dehydrase (leukotriene A4 (LTA4) synthase) types of reactions. In the present investigation we studied the mechanism of reaction of the purified porcine leukocyte 12S-lipoxygenase with 15S-hydroperoxyeicosatetraenoic acid (15S-HPETE). Oxygen-18 labeling experiments with GC-MS analysis were used to distinguish dioxygenase and leukotriene synthase activities of the enzyme; 8S,15S-DiHPETE and 14R,15S-DiHPETE were formed by oxygenation, and a series of 8,15- and 14,15-diols were formed via enzymatic synthesis of 14,15-LTA4 and nonenzymatic hydrolysis of the epoxide. 10D-3H- and 10L-3H-labeled substrates were used to study the stereospecificity of the C-10 hydrogen abstraction in the synthesis of these products. Formation of 14,15-DiHPETE and 14,15-LTA4 was associated with stereoselective abstraction of hydrogen from the 10L position of 15S-HPETE. The same type of measurements on the 8S,15S-DiHPETE product indicated a variable (50-250%) retention of the 10L-3H label, and a consistent 90% retention of the 10D-3H. In contrast, the synthesis of 8S,15S-DiHPETE by the soybean lipoxygenase was associated with the expected stereoselective abstraction of the 10D hydrogen. It appears that the porcine leukocyte 12S-lipoxygenase synthesizes 8S,15S-DiHPETE by a different mechanism.

Prostacyclin has been implicated as a mediator of renin release, whereas angiotensin II evokes prostaglandin I2 (PGI2) release from both vascular and nonvascular tissues in vitro. The physiological significance of these observations was assessed by measurement of an index of endogenous prostacyclin biosynthesis in human volunteers during varied activation of the renin-angiotensin system secondary to manipulation of dietary sodium. Excretion of the major urinary metabolite of prostacyclin in man, 2,3-dinor-6-keto-PGF1 alpha (PGI-M), fell from 295 +/- 51 to 176 +/- 35 (+/- SEM) ng g creatinine-1 (P less than 0.01) in 10 normal subjects when sodium intake was decreased from 150 to 10 meq/day. In five patients with primary hyperaldosteronism, PGI-M fell from 199 +/- 34 ng g creatinine-1 preoperatively to 120 +/- 26 pg/mg creatinine-1 after removal of the adenoma. In such patients, the reduction in PGI-M was associated with a significant increase in PRA. Thus, in both normal subjects and patients with hyperaldosteronism, PGI-M excretion fell rather than increased with activation of the renin-angiotensin system. This suggests that systemic biosynthesis of PGI2 is unrelated to renin release and that angiotensin II is unlikely to stimulate endogenous prostacyclin biosynthesis under these conditions in man.

Plexaura homomalla is a rich natural source of prostaglandins and recent evidence suggest the prostaglandin biosynthesis could occur through a lipoxygenase pathway. We have investigated the metabolism of arachidonic acid in homogenates and acetone powders of the fresh frozen coral. The biosynthesis of natural prostaglandins was not detected. However, we find a prominent 8(R)-lipoxygenase pathway leading to an alpha-ketol, characterized by high pressure liquid chromatography, gas chromatography-mass spectrometry, and NMR as 8-hydroxy, 9-keto-eicosa-5Z, 11Z, 14Z-trienoic acid, and a prostaglandin A-like cyclopentenone identified as 9-oxo-[8, 12-cis]-prosta-5Z, 10, 14Z-trienoic acid. These reactions appear analogous to the transformation of linolenic acid hydroperoxide by "isomerase" and "cyclase" of corn and flaxseed. From analysis of the absolute configurations of the coral products, and from additional stable isotope labeling experiments in H218O and D2O, we deduce that both compounds arise via conversion of 8(R)-HPETE to an 8(R), 9-allene oxide, 8R,9-oxido-eicosa-5Z, 9, 11Z, 14Z-tetraenoic acid. This unstable intermediate undergoes hydrolysis to form the alpha-ketol or cyclization to give the cyclopentenone. Significantly, we find that the prostaglandin-like product is a racemic mixture of cis side chain enantiomers, pointing to its nonenzymatic origin from the allene oxide. The alpha-ketol is formed with partial racemization and inversion of configuration, also compatible with formation in a nonenzymatic reaction. We conclude that the isomerase and cyclase reactions may merely reflect nonenzymatic breakdown of the enzymatically formed allene oxide. The origin of the endogenous (chiral) prostaglandins of the coral may involve an allene oxide intermediate, although the potential for formation of racemic products presents an interesting dilemma regarding its relationship to the natural pathway of biosynthesis.

Leukocyte preparations convert the hydroperoxy icosatetraenoic acids 5(S)-HPETE and 15(S)-HPETE to the unstable leukotriene epoxides LTA4 and 14,15-LTA4. In several ways, the conversion of 5- or 15-HPETE to leukotriene epoxide bears a formal mechanistic resemblance to the reaction catalyzed by the 12-lipoxygenase in the conversion of arachidonic acid to 12(S)-HPETE. Points of similarity include enzymatic removal of a hydrogen at carbon 10, double bond isomerization, and formation of a new carbon-to-oxygen bond. In the case of 15(S)-HPETE, two 8,15- and an erythro-14,15-dihydroxy acid (8,15- and 14,15-DiHETEs), which result from incorporation of molecular oxygen into each hydroxyl group, are coproducts in the formation of 14,15-LTA4. These facts prompted us to test the hypothesis that the biosynthesis of 14,15-LTA4 and of 8,15- and 14,15-DiHETEs from 15(S)-HPETE occurs by a mechanism similar to that observed in lipoxygenase reactions. Based on the results presented here, we conclude that the biosynthesis of 14,15-LTA4 and of 8,15- and 14,15-DiHETEs from 15(S)-HPETE occurs via a common intermediate and that, moreover, the formation of these metabolites from 15(S)-HPETE is catalyzed by an enzyme with many mechanistic features in common with the 12-lipoxygenase.

Incubation of suspensions containing polymorphonuclear and eosinophilic leukocytes with arachidonic acid led to the formation of two pairs of diastereomeric 8,(15S)-dihydroxy-5,9,11,13-icosatetraenoic acids and two erythro-14,15-dihydroxy-5,8,10,12-icosatetraenoic acids. The structures were elucidated by ultraviolet spectroscopy and gas chromatography--mass spectrometric analysis of several derivatives of each compound, catalytic hydrogenation, oxidative ozonolysis with steric analysis of alcohols, and comparison to reference compounds prepared by chemical synthesis. Experiments with 18O2 and H218O indicated that in all six compounds the hydroxyl group at C-15 was derived from molecular oxygen. Two of the diastereomeric 8,15-dihydroxy acids incorporated H218O at C-8, while the other two 8,15-dihydroxy products (also diastereomers) and the 14,15-dihydroxy compounds (geometric isomers) incorporated 18O2 at C-8 and C-14, respectively, in addition to C-15. Two of the 8,15-dihydroxy acids are formed by reaction of water with an unstable allylic epoxide intermediate, (14S,15S)-oxido-5,8,10,12-icosatetraenoic acid; the two 14,15-dihydroxy acids are proposed to be formed by reaction of activated molecular oxygen with the same epoxide, which in turn originates via 15S oxygenation of arachidonic acid.

Healthy volunteers received 60 microgram of [8,10,10(-2)H3] PGF2alpha by intravenous infusion both before and during a course of treatment with indomethacin (200 mg/day). Excretion of deuterated 5alpha, 7alpha-dihydroxy-11-ketotetranor-prostane-1, 16-dioic acid in urine was quantified by GC-MS using a reverse stable isotope dilution procedure. Indomethacin was found to have no detectable effect on the metabolism of the labelled PGF2alpha whereas output of the endogenous metabolite was markedly reduced by the effect of the drug on prostaglandin biosynthesis.

The generation from arachidonic acid and purification of large quantities of a series of monohydroxy-eicosatetraenoic acids (HETEs) which differed only in the position of the hydroxyl group permitted an in vitro analysis of the relative effects of the HETEs on a variety of human neutrophil functions. All of the HETEs elicited maximal neutrophil chemotactic responses of comparable magnitude, but the chemotactic potencies exhibited a distinct rank order with 5-HETE greater than 8-HETE:9-HETE (85:15, w:w) greater than 11-HETE=12-L-HETE. Peak chemotactic responses were achieved at concentrations of 1 microgram/ml for 5-HETE, 5 microgram/ml for 8-HETE:9-HETE and 10 microgram/ml for 11-HETE and 12-L-HETE. In the absence of a concentration gradient, the HETEs were similar in potency with respect to the stimulation of neutrophil chemokinesis and the enhancement of the expression of neutrophil C3b receptors. At optimally chemotactic and chemokinetic concentrations, none of the HETEs stimulated the generation of superoxide by neutrophils, altered the expression of neutrophil IgG-Fc receptors, or evoked the release of lysosomal enzymes. Methyl esterification of 5-HETE and 12-L-HETE reduced the chemotactic activity to less than 12% of that of the parent compound. The HETE methyl esters competitively inhibited the chemotactic activity of the homologous free acids by approximately 50% at equimolar concentrations, without inhibiting the chemotactic activity of formyl-methionyl peptides or of chemotactic fragments of the fifth component of complement (C5fr). The stimulus specificity of the competitive inhibition of chemotaxis by HETE methyl esters and the functional selectivity of the HETEs as compared to the formyl-methionyl peptides and C5fr, which stimulate neutrophil oxidative metabolism and lysosomal enzyme release, suggest that HETEs activate human neutrophils by a unique mechanism.

The metabolism of arachidonic and linoleic acids by VX2 carcinoma tissue in vitro was determined. Prostaglandin E2 was the major metabolic product of arachidonic acid in the neoplastic tissue. Minor products accounting for 3- 8% of arachidonic acid metabolism were II-hydroxy-5, 8, 12, 14-eicosatetraenoic acid (II-HETE) and 15-hydroxy-5, 8, 11, 13-eicosatetraenoic acid (15-HETE). Linoleic acid was converted to a mixture of 9-hydroxy-10, 12-octadecadienoic acid (9-HODD) and 13-hydroxy-9, 11-octadecadienoic acid (13-HODD). The conversion of linoleic acid to monohydroxy C-18 fatty acids varied from 40-80% 9-HODD and 20-60% 13-HODD in tumor tissue harvested from different animals. The quantity of monohydroxy C-18 fatty acids biosynthesized by VX2 carcinoma tissue from endogenous linoleic acid equals or exceeds that of prostaglandin E2 biosynthesis from endogenous arachidonic acid. The presence of a hydroxyl group adjacent to a conjugated diene suggest that the monohydroxy C-18 and monohydroxy C-20 fatty acids were formed via the action of lipoxygenase-like enzymes. These lipoxygenase-like reactions are inhibited by indomethacin in a concentration-dependent fashion similar to the inhibition of prostaglandin E2 biosynthesis. The enzymes catalyzing the lipoxygenase-like reactions of linoleic and arachidonic acids are localized in the microsomal fraction of VX2 carcinoma tissue. These data suggest that the lipoxygenase-like reactions are catalyzed by fatty acid cyclooxygenase and that there are two major pathways of fatty acid cyclooxygenase metabolism of polyenoic fatty acids in the neoplastic tissue. One pathway involves the formation of prostaglandin E2 via cyclic endoperoxy intermediates. The second pathway involves the formation of monohydroxy C-18 fatty acids from linoleic acid via lipoxygenase-like reactions.

Elicited rat peritoneal mononuclear cells converted arachidonic acid to a new dihydroxy acid, 5(S), 15(S)-dihydroxy-6,13-trans-8,11-cis-eicosatetraenoic acid (5,15-DiHETE). In this system, the amount of 5,15-DiHETE formed was about 20% that of leukotriene B4. The structure of the compound was determined by ultraviolet and mass spectrometric analysis, and comparison to a reference compound prepared by incubation of synthetic 5(R,S)-hydroxy- or hydroperoxy-6-trans-8,11,14-cis-eicosatetraenoic acid (5(R,S)-HETE or HPETE) with soybean lipoxygenase (linoleate:oxygen oxidoreductase, EC 1.13.11.12). Cell incubations performed under an atmosphere of 18O2 demonstrated that both hydroxyl groups in the cell product derived from molecular oxygen and that the oxygen atoms were from different oxygen molecules. Steric analysis indicated that each hydroxyl group had the S-configuration. The structural data thus indicate that 5,15-DiHETE is formed by an enzymatic double oxygenation of arachidonic acid catalyzed by both C-5 and C-15 lipoxygenases. Incubations with with [3H 8]5 (S)-HETE and [3H8]15(S)-HETE revealed that both compounds could be converted to the product. When [3H8]5(S), 15(S)-DiHPETE was added to cells, the majority of the substrate was reduced to 5,15-DiHETE. Leukocytes obtained from three human donors with peripheral blood eosinophilia also synthesized 5,15-DiHETE. Formation of the compound occurred in both eosinophils and neutrophils from these donors.

Three distinct pairs of HETEs can be distinguished on the basis of their UV spectra. We used hydroxy-linoleates (hydroxy-octadeca-cis-trans-dienoates) as a base for comparisons; both the 9- and 13-hydroxy isomers have identical chromophores with lambda max near 234 nm. The presence of a double bond three carbons removed from the conjugated diene (the chromophore of 9- and 11-HETE) causes a shift in the observed lambda max to near 235 nm. A double bond beta to the chromophore (5- and 15-HETE) gives a further shift of 1.5 nm, giving a lambda max between 236-236.5 nm. With double bonds in both these positions (8- and 12-HETE), the lambda max is observed near 237 nm. It is apparent that the exact lambda max of the cis-trans diene chromophore is influenced in a consistent way by the adjacent methylene interrupted cis double bonds.

The rate of secretion of prostacyclin (PGI2) into the circulation of normal man was estimated by measurement of the 2,3-dinor-6-keto-PGF1 alpha (D) and 15-keto-13,14-dihydro-2,3-dinor-6-keto-PGF1 alpha (KDD) urinary metabolites of PGI2. Subjects received 6-h intravenous infusions of vehicle alone and PGI2 at 0.1, 0.4, and 2.0 ng/kg per min in random order. The fractional elimination of the metabolites was independent of the rate of PGI2 infusion. 6.8 +/- 0.3% of the infused PGI2 appeared as D and 4.1 +/- 0.4% as KDD. The regression of infused PGI2 upon the quantities of the two metabolites excreted in excess of control values permitted estimation of the rate of entry of endogenous PGI2 into the circulation corresponding to a given quantity of metabolite excreted. Using the quantities excreted in the 24 h from commencement of the infusions the estimated rates were 0.08 +/- 0.02 ng/kg per min from D and 0.10 +/- 0.03 from KDD. Studies with exogenous PGI2 suggest that infusion rates 2--4 ng/kg per min are required to achieve the threshold for inhibition of platelet function (ex vivo) in man. Although not precluding a role for PGI2 in local platelet-vessel wall interactions, the much lower estimates obtained in this study suggest that endogenous PGI2 is unlikely to act as a circulating antiplatelet agent in healthy man.

Soybean lipoxygenase was reacted with phosphatidylcholine (at pH 9, with 10 mM deoxycholate), and the oxygenation products were analyzed by high-pressure liquid chromatography, UV, gas chromatography-mass spectrometry (GC-MS), and NMR. The structures of the intact glycerolipid products were established by GC-MS of diglycerides recovered by phospholipase C hydrolysis and by proton NMR of the intact phosphatidylcholine. These analyses, together with analyses of the transesterified fatty acids, indicated that arachidonyl and linoleoyl moieties in the phosphatidylcholine were converted exclusively to the 15(S)-hydroperoxy-5(Z),8(Z),11(Z),13(E)-eicosatetraenoate and 13(S)-hydroperoxy-9(Z),11(E)-octadecadienoate analogues, respectively. Control experiments proved that the intact phospholipid (and not hydrolyzed/reesterified fatty acid) was the true substrate of the oxygenation reaction. Phosphatidylethanolamine and phosphatidylinositol lipids were also substrates for specific oxygenation by the soybean lipoxygenase. The results provide concrete evidence that fatty acids esterified in phospholipid can be subject to highly specific oxygenation by a lipoxygenase enzyme.

To assess the pharmacologic effects of aspirin on endogenous prostacyclin and thromboxane biosynthesis, 2,3-dinor-6-keto PGF1 alpha (PGI-M) and 2,3-dinor-thromboxane B2 (Tx-M) were measured in urine by mass spectrometry during continuing administration of aspirin. To define the relationship of aspirin intake to endogenous prostacyclin biosynthesis, sequential urines were initially collected in individuals prior to, during, and subsequent to administration of aspirin. Despite inter- and intra-individual variations, PGI-M excretion was significantly reduced by aspirin. However, full mass spectral identification confirmed continuing prostacyclin biosynthesis during aspirin therapy. Recovery of prostacyclin biosynthesis was incomplete 5 d after drug administration was discontinued. To relate aspirin intake to indices of thromboxane biosynthesis and platelet function, volunteers received 20 mg aspirin daily followed by 2,600 mg aspirin daily, each dose for 7 d in sequential weeks. Increasing aspirin dosage inhibited Tx-M excretion from 70 to 98% of pretreatment control values; platelet TxB2 formation from 4.9 to 0.5% and further inhibited platelet function. An extended study was performed to relate aspirin intake to both thromboxane and prostacyclin generation over a wide range of doses. Aspirin, in the range of 20 to 325 mg/d, resulted in a dose-dependent decline in both Tx-M and PGI-M excretion. At doses of 325-2,600 mg/d Tx-M excretion ranged from 5 to 3% of control values while PGI-M remained at 37-23% of control. 3 d after the last dose of aspirin (2,600 mg/d) mean Tx-M excretion had returned to 85% of control, whereas mean PGI-M remained at 40% of predosing values. Although the platelet aggregation response (Tmax) to ADP ex vivo was inhibited during administration of the lower doses of aspirin the aggregation response returned to control values during the final two weeks of aspirin administration (1,300 and 2,600 mg aspirin/d) despite continued inhibition of thromboxane biosynthesis. These results suggest that although chronic administration of aspirin results in inhibition of endogenous thromboxane and prostacyclin biosynthesis over a wide dose range, inhibition of thromboxane biosynthesis is more selective at 20 than at 2,600 mg aspirin/d. However, despite this, inhibition of platelet function is not maximal at the lower aspirin dosage. Doses of aspirin in excess of 80 mg/d resulted in substantial inhibition of endogenous prostacyclin biosynthesis. Thus, it is unlikely that any dose of aspirin can maximally inhibit thromboxane generation without also reducing endogenous prostacyclin biosynthesis. These results also indicate that recovery of endogenous prostacyclin biosynthesis is delayed following aspirin administration and that the usual effects of aspirin on platelet function ex vivo may be obscured during chronic aspirin administration in man.

Arachidonate 12-lipoxygenase purified from porcine leukocytes shows 14R-oxygenase and 14, 15-leukotriene A synthase activities with 15-hydroperoxy-arachidonic acid as substrate. The enzyme transformed 5, 15-dihydroperoxy-arachidonic acid to several compounds with a conjugated tetraene. A major product was identified as 5S, 14R, 15S-trihydroperoxy-6, 10, 12-trans-8-cis-eicosatetraenoic acid, which was reduced to 5S, 14R, 15S-8-cis-lipoxin B. A requirement of molecular oxygen and the results of H2(18)O experiments suggested that formation of the latter compound was attributed mostly to the 14R-oxygenase activity of the enzyme. There were several other minor products identified as lipoxin A and B isomers. They were produced presumably by hydrolysis of 14, 15-epoxy compound formed by the leukotriene A synthase activity of 12-lipoxygenase.

Prostacyclin is a potent vasodilator and platelet inhibitor produced by vascular endothelium. Endogenous production of prostacyclin under physiologic conditions is extremely low, far below the capacity of vascular tissue to generate this substance in response to stimulation in vitro. This may reflect a low frequency or intensity of stimulation of prostacyclin production. We postulated that if prostacyclin does act as an endogenous platelet-inhibitory agent, it should be produced in greater amounts in a clinical setting in which platelet-vascular interactions are likely to be increased. To test this hypothesis, we examined prostacyclin biosynthesis in patients with severe atherosclerosis and evidence of platelet activation in vivo. Excretion of 2,3-dinor-6-keto-prostaglandin F1 alpha, a major urinary prostacyclin metabolite, was significantly higher in 9 patients with severe atherosclerosis and evidence of platelet activation (251 to 1859 pg per milligram of creatinine) than in 54 healthy volunteers (45 to 219 pg per milligram of creatinine; P less than 0.001). This difference represented an alteration in biosynthesis rather than in metabolism, since the fractional conversion of infused prostacyclin to the dinor metabolite was identical in both groups. Prostacyclin production may be low in healthy persons because there is almost no stimulus for its production but enhanced in patients with severe atherosclerosis as a consequence of platelet interactions with endothelium or other vascular insults. These observations are compatible with a role for prostacyclin as a local regulator of platelet-vascular interactions.

cis-12-Oxophytodienoic acid (cis-12-oxo-PDA) is a C18 cyclopentenone formed from the 13-(S)-hydroperoxide of linolenic acid in flaxseed and other plant tissues. Although the structure of cis-12-oxo-PDA is well established, the absolute configuration of the side chains has not been determined. We have now measured this important parameter by two independent approaches. The CD spectrum of freshly prepared cis-12-oxo-PDA showed no deviations from base line--implying that the product is racemic. This conclusion was checked by a high-pressure liquid chromatography (HPLC) method capable of resolving the enantiomers; cis-12-oxo-PDA was reduced to two saturated hydroxy analogues which were each converted to (-)-menthoxycarbonyl diastereomers and analyzed by HPLC. Each epimer was resolved as two peaks of equal area, thus confirming that their cis-12-oxo-PDA parent is a racemic mixture, enantiomeric at the ring junctures. We propose that the biosynthesis of racemic cis-12-oxo-PDA proceeds by dehydration of the 13(S)-hydroperoxide to an allene oxide. A major fate of the allene oxide is hydrolysis to an alpha-ketol, which is always formed together with cis-12-oxo-PDA. The allene oxide also opens to a zwitterion, which undergoes charge delocalization to form a planar intermediate; this structure is the achiral precursor of the stable end product of pericyclic ring closure, viz., racemic cis-12-oxo-PDA.

The rabbit reticulocyte lipoxygenase is known to display an unusual facility for oxygenation of esterified polyunsaturated fatty acids, yet the precise structures of the products are not known. With free arachidonate as substrate the enzyme is known to catalyze 15S and 12S oxygenations, and demonstration of a facility for catalysis of these reactions on phospholipids would extend the potential scope of lipoxygenase reactions in cells. We elected to study in detail the reaction of the enzyme with a natural phospholipid, palmitoyl/arachidonoyl-phosphatidylcholine. We determined the nature of the products by initial isolation by RP-HPLC, followed by transesterification and identification of the oxygenated products by HPLC, uv, GC-MS, and steric analysis of hydroxyl configuration by HPLC. The major product was identified as a phosphatidylcholine in which the arachidonate component was converted to the 15(S)-hydroperoxy-eicosatetraenoate. A second oxygenated phospholipid was produced in smaller quantities (2-5% of the latter product) and identified as the 12(S)-oxygenated analog. These products were also identified after reaction of the reticulocyte lipoxygenase with human red cell membranes which were radiolabeled by preincubation with [3H]arachidonic acid. The finding of 12S oxygenation represents the first evidence that a lipoxygenase can control a reaction centered on the 10-carbon of an arachidonoyl phospholipid. This is an important precedent, because hydrogen abstraction from carbon-10 is a critical step in the lipoxygenase-catalyzed synthesis of 8- and 12-hydroperoxy-eicosatetraenoates (HPETEs) and for the conversion of 5- and 15-HPETEs to leukotrienes.

The pure reticulocyte lipoxygenase converts 15LS-hydroxy-5,8,11,13(Z,Z,Z,E)-icosatetraenoic acid (15LS-HETE) methyl ester to a complex mixture of products containing 5DS,14LR,15LS-trihydro(pero)xy-6E,++ +8Z,10E,12E-icosatetraenoate methyl ester (lipoxin B methyl ester), 5DS,15LS-DiH(P)ETE methyl ester and four 8,15LS-DiH(P)ETE methyl ester isomers [DiH(P)ETE = dihydro(pero)xy-icosatetraenoic acid]. After a short incubation period (15 min) 5DS,15LS-DiH(P)ETE methyl ester was found to be the main product, whereas after a 3-h incubation lipoxin B methyl ester was the predominant product. The reaction shows a remarkable stereoselectivity since only small amounts of other trihydroxy tetraenes are formed. Anaerobiosis, heat inactivation of the enzyme, or incubation in the presence of lipoxygenase inhibitors (icosatetraynoic acid, nordihydroguaiaretic acid) completely abolished the reaction. The complete steric structure of the major tetraene product (lipoxin B methyl ester) was established by ultraviolet spectroscopy, HPLC on four different types of columns, gas chromatography/mass spectrometry, gas/liquid chromatography of the ozonolysis fragments of the menthoxycarbonyl derivatives, and by 400-MHz 1H-NMR. Atmospheric oxygen was incorporated at carbon-5 and carbon-14 into the major product. 5DS,15LS-DiH(P)ETE methyl ester was shown to be an intermediate in the synthesis. Lipoxin B was also formed during the oxygenation of arachidonic acid, 15LS-HETE and 5DS,15LS-DiHETE. The results presented here indicate that lipoxin B can be formed by pure lipoxygenases via a sequential oxygenation of arachidonic acid or its hydro(pero)xy derivatives.

Specific evidence is presented of the release of prostaglandins from the colon in active ulcerative colitis. An increase in the levels of bioassayed prostaglandin-like activity in the stools and colo-rectal venous plasma from patients with active ulcerative colitis is described. Radioimmunoassay confirms the presence of prostaglandin E and prostaglandin F in the stools in colitis. Increased urinary levels of prostaglandin F metabolite occur in patients with active colitis and return to normal as the disease becomes quiescent. Sulphasalazine and its faecal metabolite, 5-aminosalicylic acid, were shown by an indirect method (reduction of the tone of the isolated rat fundus strip) to inhibit prostaglandin biosynthesis in vitro. In contrast, sulphasalazine was without effect on the urinary excretion of prostaglandin F metabolite in 7 healthy subjects. In 2 patients with colitis withdrawal of sulphasalazine was associated with increasing levels of stool prostaglandin-like activity and urinary prostaglandin F metabolite excretion. Indomethacin, given to 3 patients with chronically active ulcerative colitis, unresponsive to standard medical treatment, was associated with a decreased urinary excretion of prostaglandin F metabolite but was without clinical benefit. The possible mode of action of sulphasalazine as a prostaglandin inhibitor in colitis is discussed along with the potential use of other prostaglandin inhibitors.