Inhibition of cyclooxygenase activity and platelet aggregation by ...

THEJOURNAL OF BIOLOGICAL CHEMISTRY

I!! 1986

Val. 261, Nu. 32, Issue of November 15, pp. 15334-15338,1986 Printed in U.S.A.

by The American Society of Biological Chemists, Inc

Inhibition of Cyclooxygenase Activity and Platelet Aggregation by Epoxyeicosatrienoic Acids INFLUENCE OF STEREOCHEMISTRY* (Received for publication, July 1, 1986)

F. A. Fitzpatrick, M. D. Ennis, M. E. Baze, M. A. Wynalda, J. E. McGee, and W. F. Liggett From LiDids Research 7234-25-12.Pharmaceutical Research and Development, The Upjohn Company, Kalamaloo, Michigan 49007 ~~

Certain epoxyeicosatrienoic acids (EETs) that were of renal Na’-K+-ATPase (lo), and mobilization of microsomal not cyclooxygenase substrates were effectivecycloox- Ca2+(11).These effects vary according to the localization of ygenase inhibitors. Both (f)-14,15-cis-EET and (2)- the epoxide group (9); however, the detailed mechanisms of 8,g-cis-EET inhibited purified enzyme at concentra- EET action are uncertain. Interactions with other enzymes ; was ineffec- in the arachidonic acid cascade are likely in viewof the tions from 1 to 50 p ~ (+)-11,12-cis-EET . the case of structural similarities between arachidonic acid itself and its tive at concentrations below 100 p ~ For l4,15-cis-EET, only the (14R,15S)-stereoisomer was epoxygenase metabolites (12-14). Therefore, we examined the active.Otherisomersincluding(14S,15R)-cis-EET, influence of several EET isomers on cyclooxygenase-depend(14R,IBR)-truns-EET, (14S,lBS)-truns-EET, and the ent metabolism of arachidonic acid. Our results indicate that erythro and threo vicinal 14,15-diols were inactive. In certain, but not all, EETs, inhibit both isolated enzyme prepaddition to their effects on isolated enzyme preparations, cyclooxygenase activity in plateletsuspensions, arations and cyclooxygenase metabolism in platelets stereosreflected by thromboxane B2 formation, was also in- pecifically. Unlike their stereospecific effect on cyclooxygenhibited by (14R,lBS)-cis-EET and(.+)-8,9-cis-EET but ase, several EET isomers inhibited arachidonic acid-induced not by the otherisomers. Thus potency and stereospec- platelet aggregation without evident stereospecificity and without any associated inhibition of cyclooxygenase activity. ificity requirements were maintained cyclooxygenfor ase within intact platelets. EXPERIMENTALPROCEDURES Unlike the stereospecific inhibitionof the cyclooxygenase enzyme, platelet aggregation induced by araMaterials-Bovine hematin (Sigma); cyclooxygenase enzyme chidonic acid was inhibitedby all EET isomersat con- (67,000 units/mg, >90% pure; Oxford Biomedical Research); arachidonic acid (NuChek Prep); carrier-free ortho [32P]phosphoricacid, centrations from 1 to 10 p M with no evident stereo[3H]PGE2(120 Ci/mmol), [ 3 H ] T ~ B(150 2 Ci/mmol), and cyclic [3H] specificity.Inhibition of aggregationwasnotuniformly associated with inhibition of thromboxane B2 AMP assay kits (New England Nuclear); PGE,, prostacyclin-sodium, and U-46619 (The Upjohn Co.); and water, hexane, ethyl formation; ordinarily, these two parameters correlate TxB,, acetate, acetonitrile, and methanol, all distilled in glass (Burdick & closely. This dissociation was not maintained for anJackson Laboratories Inc.), were used as received. Ram seminal other biochemical process involved in platelet activa- vesicle acetone-pentane powders containing cyclooxygenase activity tion. For instance, there was a uniform correlation were prepared as described (15). between inhibition of phosphorylation of a 40-kDa Preparation of Epoxyeicosatrienoic Acid Isomers-Racemic mixplatelet protein and inhibition of aggregation. Our re- tures of (+)-5,6-cis-EET, (+)-8,9-cis-EET,(&)-11,12-&-EET, and sults suggest that effects of EET may originate from (+)-14,15-cis-EET were prepared by epoxidation of arachidonic acid either stereospecific or nonspecific mechanisms. Defi- (1 mmol) with 1 eq of m-chloroperbenzoic acid (16). These four nition of such mechanisms may beimportant to appre- epoxides were isolated in one fraction by chromatography over silicic acid (200 g) eluted with hexane/ethyl acetate (9:l v/v) at 3.0 ml/min. ciate anyphysiological relevance of these substances. Individual regioisomers were subsequently isolated by reversed-phase chromatography on a Cla column (250 X 10 mm) eluted with acetonitrile/water (80:20, v/v) at 2.0 ml/min. The identity of the racemic isomers was established by chromatographic and mass spectrometric Most biologically active eicosanoids originate from the cy- comparisons with standards. Geometric and stereochemical isomers clooxygenase or lipoxygenase pathways of arachidonic acid of 14,15-EET including (14R,15S)-cis-EET, (14S,lSR)-cis-EET, metabolism (1).Recently, arachidonic acid epoxides produced (14S,15S)-trans-EET, and (14R,lBR)-trans-EET were synthesized as by the cytochrome P-450 monooxygenase system, termed the described (17). Assay--Ram seminal vesicle acetone-pentane powepoxygenase pathway, have also been recognized as biologi- derCyclooxygenose (0.75-1.5 mg/ml) or purified cyclooxygenase (100-200 units) was cally active substances (2-6). The pharmacological traits of activated for 1min at 37 “C in 0.1 M Tris, pH 7.8 (2.0 ml), containing cis-epoxyeicosatrienoic acids (EETs)’ include stimulation of 0.5 mM phenol. The enzyme was then incubated with hematin (1p ~ ) peptide hormone release from endocrine cells (7-9), inhibition and EET isomers (0-200 p ~ for ) 2 min. Enzymatic reactions were initiated by addition of arachidonic acid (100-300 nmol/30 p1 of 0.01 * The costs of publication of this article were defrayed in part by N NaOH). Cyclooxygenase-catalyzed incorporation of oxygen into the payment of page charges. This article must therefore be hereby arachidonic acid was determined polarographically (18). Reaction marked “advertisement” in accordance with 18 U.S.C. Section 1734 rates refer to theinitial, linear phase of oxygen consumption. Product formation was monitored by radioimmunoassay for PGE, (19). solely to indicate this fact. Isolation of Human Platelets-Human whole blood (225 ml) colThe abbreviations used are: EETs, epoxyeicosatrienoic acids; TxB,, thromboxane B,; PGE,, prostaglandin E,; HPLC, high per- lected in sterile 3.8% (w/v) sodium citrate (25 ml) was centrifuged at formance liquid chromatography; Hepes, 4-(2-hydroxyethyl)-l-piper- 200 X g for 20 min, andthe platelet-rich plasma was isolated. azineethanesulfonic acid; EGTA, [ethylenebis(oxyethylenenitrilo)] Suspensions containing humanwashed platelets (6-8 X lo9platelets/ ml) in Hanks’ balanced salt solution with 0.05 M Hepes buffer, pH 7, tetraacetic acid.

15334

Inhibit Epoxygenase Metabolites 0.8 mM M$+, 1.5 mM Ca'+ were prepared as described (20). Platelet Aggregation-Washed platelet suspensions (1.2 ml) were incubated with E E T isomers (0-120 nmol) for 2 min a t 37 "C. These suspensions were then transferred to an aggregometer cuvette containingarachidonic acid(1.2pg) toinitiateplatelet aggregation. Aggregation was monitored as described (21). Samples (0.20 ml) were withdrawn a t specified times for determination of TxB2 by radioimmunoassay (22). TxB2 biosynthesis reflects the cyclooxygenase metabolism by intact platelets. In certain experiments, samples (1.0 ml) were analyzed by reversed-phase HPLC to determine their 12-hydroxyeicosatetraenoic acid content (23) which reflects their 12-lipoxygenase activity. The effects of E E T isomers on platelet cyclic AMP were determined by incubating platelet suspensions ( 2 ml) for 2 min at 37 "C with the respective isomers and then quenching with20 mM EGTA (0.5 mi). Platelets were lysed by sonication; the supernatant fluid was isolated, lyophilized, and reconstituted in 0.2 ml of water. The cyclic AMP contentwas determined by radioimmunoassay using an assay kit from New England Nuclear. Platelet ProteinPhosphorylation-The platelet pellet isolated from 100 ml of platelet-rich plasma by centrifugation a t 2000 X g for 15 min was resuspendedin 50 ml of Hanks'balancedsaltsolution without Ca'+ or M$+, plus 5% (v/v) platelet-poor plasma. Cells were labeled with 2.5 mCi of ortho [Y2P]phosphoricacid for 3 h at 37 "C. Excess "'PO, was removed by washing the platelet suspension twice with 50 ml of Ca2+/Mg2"free Hanks' balanced saltsolution. Platelets were then resuspended (4.5 X 10' cells/ml) in Hanks' balanced salt solution with 1.4 mM Caz+ and 0.8 mM M$+. The effect of EETs (0.1-30 p ~ on) aggregation initiated by arachidonic acid (1 rg/ml) or U-46619 ( 1 pg/ml) was performed as described above. T o determine their correspondingeffects on protein phosphorylation,a sample (100 pl) was withdrawn at 15s and quenched by incubation for 10 min a t 100 "Cin 25 pl of Laemmli (24) sample buffer, 5 X concentrate. Proteins were separated by electrophoresis on 10% sodium dodecyl sulfate-polyacrylamide (24)and visualized by staining with 0.1% (w/ v) Coomassie Brilliant Blue in 50% (v/v) methanol, 10% (v/v) acetic acid. Protein phosphorylation was visualized by autoradiography on Kodak ARO x-ray film. Instrumental Analysis-Aggregations were monitored with a dualchannel aggregometer (Payton Associates, Buffalo, NY). Dissolved oxygenwas monitored with an oxygenelectrode and a model 930 oxygraph (Yellow SpringsInstrument Co.).A Varian Model500 HPLC with a variable wavelength UV detector wasused for high performance liquid chromatography.

Cyclooxygenase

15335

100Y I

4

-

-

@=

z 80P I-

n . -

z 3

600

0 w

0

40z 0

ztL s

-

-EET

20-

-

4

-

'

'

I

5I ' 1 ' '101

1

,

1

20 50 I.~MINHIBITOR

1

I

1

100

200

,

,

I 500

FIG. 1. Effect of EET isomers on cyclooxygenase metabolism of arachidonic acid. Suspensions of an acetone-pentanepowder preparation of ram seminal vesicles (0.75 mg/ml) in 0.10 M Tris, pH 8.5, containing 0.13 mM phenol were preincubated with (+)-14,15cis-EET (O), (+)-8,9-cis-EET (a), and (f)-cis-11,12-EET (A) for 2 min at 37 "C.Arachidonicacid (3 X M ) was added to initiate the enzymaticreaction. The initialrate of oxygen incorporation was measured polarographically. Values represent the mean f S.E. ( n = 4-8). The effect of a conventional inhibitor, ibuprofen, is depicted for comparative purposes. T

RESULTS

Inhibition of Cyclooxygenase-Two E E T regioisomers inhibited arachidonic acid metabolism by the cyclooxygenase enzyme ina dose-dependent manner. Concentrationsrequired for 50% inhibition of oxygen consumption by an acetonepentane powder enzyme preparationwere 30 W M for (+)-14,15cis-EET and 50 W M for (+)-8,9-cis-EET, compared t o greater than 200 W M for (+_)-11,12-cis-EET. Unlike (+)-5,6-cis-EET, theseisomers were notsubstrates for the cyclooxygenase enzyme. The potency of (&)-14,15-cis-EET surpassed thatof nonsteroidal anti-inflammatory agents, such as ibuprofen or aspirin, which reduced oxygen consumption by 50% at concentrations of 100 W M (Fig. 1). Plots of velocity-' uersus inhibitor concentrationwere nonlinear; however, extrapolation of the linearregion corresponding to low inhibitor concentrations allowed an estimation of K , = 25 pM for (+)-14,15-cis-EET and K,= 45 W M for (&)8,g-cis-EET. The potency of (+)-11,12-cis-EET was insufficient to estimate a valuefor Kifrom theDixon plot. Following incubation at 37 "C for 5 min, recoveries of 50 W M (&)-14,15cis-EET, (+)-11,12-cis-EET, and (+)-8,9-cis-EET from acetone-pentane powder cyclooxygenase preparations were 99 & 2, 96 & 1, and 65 4%, respectively (mean + S.D., n = 4). In the latter case, the lower recovery was due to formation of vicinal 8,9-dihydroxyeicosatrienoicacid during the incubation. Effect of Stereoisomerism and Geometric Isomerism on Cyclooxygenase Inhibition-For the case typified by (+)-14,15_+

I

I

0

1

I

I

3

I

I 10

I , , I ,

1

I

30

I

1 100

I , , , ,

Micromolar

FIG. 2. Effect of 14,15-epoxyeicosatrienoicacid isomers on cyclooxygenase metabolism of arachidonic acid. Purified cyclooxygenase (200 units/ml) was incubated for 0.5 min a t 37 "C with 100 PM arachidonic acid. Samples were removed, and their PGE, content was determined byradioimmunoassay. ET, eicosatrienoic acid.

cis-EET, only the cis- and not the trans-epoxide inhibited PGE, formation and oxygen consumption by a purified cyclooxygenase preparation. Furthermore, this effect was stereospecific: only the dextrorotary isomer (14R,15S)-cis-EET was active. At concentrations below 100 WM, neither (14S,lSR)-cis-EET,(+)-14,15-trans-EET,(14R,15R)-transEET, and (14S,15S)-trans-EET nor the erythro and threo vicinal 14,15-diol hydration products inhibited cyclooxygenase activity (Fig. 2). Inhibition of Cellular Cyclooxygenase Actiuity-Enzymatic

15336

Epoxygenase Inhibit Metabolites

inhibition by cis-EET isomers was not restricted to isolated cyclooxygenase preparations. Cyclooxygenase activity in platelets, reflected by TxB2 formation, was also reduced by (14R,15S)-cis-EET in the concentration range 0-13 ~ L M(Fig. 3). Aggregation of human washed platelets initiated by arachidonic acid (1 pg/ml) was also inhibited (Fig. 3, inset). Potency and structure-activity relationships evident for the isolated cyclooxygenase enzyme also applied to platelet suspensions. TxBz formation was unaltered by (14S,15R)-cisEET, (14S,15S)-truns-EET, (14R,ER)-trans-EET, (f)14,15trans-EET, threo/erythro vicinal 14,15-diol, and (*)-11,12cis-EET at concentrations below 50 ~ L M(Table I). Inhibition was evident only with (14R,15S)-cis-EET and (f)-8,9-cisEET. Inhibition of arachidonic acid metabolism in platelets by (f)-14,15-cis-EET was confined tothe cyclooxygenase enzyme; the 12-lipoxygenase enzyme in platelets was unaffected. Platelet suspensions preincubatedwith 0,3, 10,20,33, and 50 p~ (f)-14,15-cis-EETproduced 7.4 f 0.2, 7.5 f 0.2, 7.5 f 0.1, 7.3 +- 0.1, 7.1 f 0.1, and 6.7 f 0.2 nmol of 12hydroxyeicosatetraenoic acid, respectively, following incubation with 25 ~ L Marachidonic acid at 37 "C for 5 min. Inhibition of Platelet Aggregation: Luck of StereospecificityIn contrastto the stereospecific inhibition of the cyclooxygenase enzyme, platelet aggregation induced by arachidonic acid was inhibited by all EET isomers at concentrations from 1 to 10 p~ with no evidentstereospecificity. Cellular thromboxane biosynthesis and platelet aggregation were dissociated in several instances. For example, aggregation was inhibited by 1-10 p~ (14S,lEiS)-truns-EET, (14R,lSR)-trans-EET, (14S,15R)-cis-EET, (+)-14,15-truns-EET and (?)-11,12cis-EET; yet TxB, biosynthesis in the presence of 30-70 p~ concentrations of these compounds remained indistinguishable from control values (Fig. 4). This dissociation was not maintained for another biochemical process involved in platelet activation.For instance, there was a uniform correlation between inhibition of aggregation and inhibition of phosphorylation of a 40-kilodalton protein by all EETs regardless of their regiochemical, geometric, or stereochemical form (Fig. 5). The inhibition of aggregation and protein phosphorylation was not due to an EET-dependent increase in platelet intracellular cyclic AMP. The cyclic AMP levels in platelets incubated for 2 min at 37 "C with ) statistically indistindifferent EET isomers (30-100 p ~ were guishable from control values (Table 11).

Cyclooxygenase TABLE I

Effect of EET isomers on cyclooxygenaseactiuity in platelets Compound

*

TxB2 (mean S.E., n 4-10)

(+.)-ll,lP-cis-EET Control 50 p M 75 pM

163 -t 14 142 f 13 128 +. 15

(+.)-8,9-&-EET Control 10 pM 25 p M 75 p M

294 +. 8 129 f 7" 97 f 12" 46 f 2'

(+.)-14,15-~i~-EET Control 13p M 20 p M 27 p M

306 f 12 258 f 7" 118 f 2" 86 f 9"

(+.)-14,15-tr~ns-EET Control 35 pM 70 p M

306 & 12 318 & 10 318 & 10

(14S,15R)-~is-EET Control 40 pM 80 JLM 100 JLM

204 f 8 170 f 13 201 & 16 238 t- 19

(14R,15R)-tran~-EET Control 100 #M

214 f 10 214 4- 15

(14S,15S)-trans-EET Control 20 p M

182 f 8 196 5 22

14,15-dihydroxy-ET* Control 35 p M 70 #M

206 f 13 225 =t 27 198 =t 21

=

" p < 0.01. ET. eicosatrienoic acid.

0 0

1 2 3 MINUTES

MINUTES

0.6 1.2 3.3 13 MICROMOLAR (14R,lSS)-cis-€ET

FIG. 3. Effect of (14R,15S)-cis-EET on platelet thromboxane B, biosynthesis and aggregation. Human washed platelet suspensions (1.2 ml) were incubated at 37 "C for 2 min with (14R,15S)-cis-EET prior to addition of arachidonic acid (1.2 pg). Samples (200 pl) were withdrawn after 3 min for determination of their TxB, content by radioimmunoassay.

FIG. 4. Effect of EET isomers on platelet aggregation and thromboxane Bz biosynthesis. Human washed platelet suspensions (1.2 ml) were incubated at 37 "C for 2 min with (+)-11,12-cisEET, (+)-14,15-trans-EET, (14S,15R)-cis-EET, and (14RJ5R)trans-EET prior to addition of arachidonic acid (1.2 pg). Samples were withdrawn after 3 min for determination of their TxB, content by radioimmunoassay.

EpoxygenaseInhibit Metabolites

Cyclooxygenase

15337

other than those previously reported (7-11). First, certain, but notall, EET isomers inhibited the isolated cyclooxygenase (14S.15A)(14S,15S)enzyme in vitro and cyclooxygenase in intact platelets. For cis-€ET trans-EET the case typified by 14,15-EET, it is noteworthy that this inhibition was stereospecific. Furthermore, the stereospecific effect of (14R,15S)-cis-EET conforms to a pattern establishing the dextrorotary isomer as more potent than the levorotary isomer of an enantiomeric pair (25, 26). Little is known about the biosynthesis of these compounds in vivo; however, epoxidation of arachidonic acid by microsomal cytochrome P450 in vitro proceeds stereoselectively. For 14,15-cis-EET, the predominant form (80%) corresponds to the active species: 1 2 (14R,15S)-~is-EET(27). MINUTES Differences in their interaction with the cyclooxygenase enzyme may be one mechanism to account for differences in the pharmacological profiles of EET isomers (7, 11). For example, (+)-5,6-cis-EET is a substrate (12-14), (14RJ5S)cis-EET and (&)-8,9-cis-EET are inhibitors, and (+)-11,12cis-EET has no effect at concentrations that arelikely from either pharmacological or physiological perspectives. It is uncertain if 14,15-cis-EET and 8,g-cis-EET occur as intact compounds in vivo at concentrations sufficient to inhibit the cyclooxygenase in view of the limited evidence of their biosynthesis (28) and their tendency toward hydration or con0 0.1 1 3 o l jugation (29,30). However, their potency in vitro exceeds that VM for clinically useful cyclooxygenase inhibitors such as aspirin FIG. 5. Effect of 14,15-EET isomers on platelet aggregation or ibuprofen. Few other naturally occurring metabolites or and protein phosphorylation.Human washed plateletsuspensions isomers of arachidonic acid have been described as cyclooxy(1.0 ml)radiolabeledwith [3ZP]phosphate werepreincubatedwith of genase inhibitors (31). 0.1-3.0 p~ EETisomers at 37 "C for 2 minpriortoaddition Experiments to determine the influence of EET isomers on arachidonic acid (1pglml). After 15 s, samples were withdrawn for analysis of the protein phosphorylation pattern by sodium dodecyl cyclooxygenaseactivity in plateletsrevealed another property sulfate gel electrophoresis. of these compounds: inhibition of platelet aggregation. In TABLEI1 contrast to their stereospecific inhibition of cyclooxygenase, the inhibition of platelet aggregation occurred with all EET Effect of EET isomers on CAMP content in platelets isomers regardless of their regiochemical, geometric, or sterCAMP (mean ? Sample S.D.) eochemical features. Ordinarily, TxB2 formationcorrelates well with platelet aggregation; however, several EET isomers prnolj1P platelets inhibited aggregation without a corresponding reduction in 5.5 Control f 1.2 TxB2 levels. We are unaware of other examples where the PGI2, 3 p M 84.1 f 2.9" (14R,15S)-&-EET dissociation between TxB2 biosynthesis and platelet aggre30 p M 6.0 f 1.7 gation is so prominent. It is unlikely that theantiaggregatory 6.9 f 2.3 100 p M effect of the EET isomers is due to anelevation of intracellular (14S,15R)-cis-EET cyclic AMP because such increases also reduce TxB2 forma30 p M 6.9 f 0.9 tion (32, 33), and we observed that concentrations of EET 100 p M 8.0 f 2.1 10-20-fold greater than those thatinhibited aggregation still (14R,15R)-trans-EET 30 p M 6.7 -t 0.8 had no capacityto reduce TxB2 levels or toraise intracellular 6.8 f 0.8 100 p M cyclic AMP. (14S,15S)-trans-EET The dissociation between platelet aggregation and related 30 pM 6.1 f 0.3 biochemical processes, such as TxB2formation, was limited. 100 pM 4.8 f For instance, phosphorylation of a 40-kDa platelet protein is (,)-ll,lf-cis-EET another event that correlates with platelet activation (34). 4.6 f 30 pM 5.0 f 100 pM Unlike the dissociation between TxB2levels and aggregation, "Values represent the mean f S.D. from three to five separate there was a uniform association between the effect of various experiments. For platelets incubated with prostacyclin (PGIz),there EET on phosphorylation and aggregation. This suggests that was a statistically significant increase(p < 0.001) in their cyclic AMP EETs may be useful probes to dissect the separate compocontent. For platelets treated with EETs, the cyclic AMP content nents involved in stimulus-response coupling in platelets. was indistinguishable from the control. In summary, our results indicate that effects of EET may The EETisomers also inhibitedaggregation of platelet-rich originate from stereospecific or nonspecific mechanisms. Defplasma; however, the amounts required were greater than inition of such mechanisms may allow a determination of those necessary toinhibit aggregation of washed platelet which, if any, traits of the EET arephysiologically relevant. suspensions. For (+)-14,15-cis-EET, (+)-14,15-truns-EET, Acknowledgment-Thanks are due to G. L. Bundy for providing (+)-11,12-cis-EET and (&)-8,9-cis-EET,concentration-deracemic epoxyeicosatrienoicacids. pendent inhibition was evident between 50 and 200 p ~ .

""-

DISCUSSION

Our results indicate that epoxygenase metabolites of arachidonic acid have enzymatic and pharmacological attributes

REFERENCES 1. Needleman, P.,Turk, J., Jakschik, B., Morrison, A., and Lefkowith, J. (1986)Annu. Reu. Biochem. 55,69-102

15338

Epoxygenase Metabolites Inhibit Cyclooxygenase

2. Capdevila,J., Marnett,L., Chacos, N., Prough, R., and Estabrook, R. (1982) Proc. Natl. Acad. Sci. U. S. A. 79, 767-770 3. Chacos, N., Falck, J . R., Wixstrom, C., and Capdevila, J. (1982) Biochem. Biophys. Res. Commun. 104,916-922 4. Oliw, E., Guengerich, F. P., and Oates, J. A. (1982)J. Bioi. Chem. 257,3771-3781 5. Morrison, A., and Pascoe, N. (1981) Proc. Natl. Acad. Sei. U. S. A. 78, 7375-7378 6. Ferreri, N., Schwartzman, M., Ibraham, N., Chande, P., and McGiff, J. (1984) J. Pharmncol. Exp. Ther. 231,441-448 7. Capdevila, J., Chacos, N., Falck, J. R., Manna, S., Negro-Vilar, A., and Ojeda, S. R. (1983) Endocrinology 113, 421-423 8. Snyder, G., Capdevila, J., Chacos, N., Manna, S., and Falck, J. R. (1983) Proc. Natl. Acad. Sci. U. S. A. 80,3504-3507 9. Falck, J. R., Manna, S., Moltz, J., Chacos, N., and Capdevila, J. (1983) Biochem. Biophys. Res. Commun. 114, 743-749 10. Schwartzrnan, M., Ferreri, N., Carroll, M., Songu-Mize, E., and McGiff, J. (1985) Nature 3 1 4 , 620-622 11. Kutsky, P., Falck, J. R., Weiss, G., Manna, S., Chacos, N., and Capdevila, J . (1983) Prostaglandins 26, 13-21 12. Oliw, E. H. (1984) J. Bid. Chem. 2 5 9 , 2716-2721 13. Oliw, E. (1984) FEBS Lett. 172, 279-283 14. Oliw, E. (1984) Biochim. Biophys. Acta 795, 384-391 15. Wallach, D., and Daniels, E. (1971) Biochim. Biophys. Acta231, 445-457 16. Chung, S.-K., and Scott, A. (1974) Tetrahedron Lett. 35, 30233024 17. Ennis, M. D., and Baze, M. E. (1986) Tetrahedron Lett., in press

18. Smith, W., and Lands, W. E. M. (1972) Biochemistry 11,32763285 19. Fitzpatrick, F., and Bundy, G. (1978) Proc. Natl. Acad. Sci. U. S. A. 75, 2689-2693 20. Vargas, J., Radomski, M., and Moncada, S. (1982)Prostaglandins 23,929-945 21. Born, G. V. R. (1962) Nature 194,927-929 22. Fitzpatrick, F. (1982) Methods Enzymol. 86, 286-297 23. Sun, F., and McGuire, J. (1984) Biochim. Biophys. Acta794,5664

24. Laemmli, U. K. (1970) Nature 227,680-685 25. Kulmacz, R. J., and Lands, W. E. M. (1985) J. Biol. Chem. 260, 12572-12578 26. Takeguchi, C., and Sih, C. J. (1972) Prostaglandins 2, 169-184 27. Falck, d . , Manna, S., Jacobson, H., Estabrook, R., Chacos, N., and Capdevila, J. (1984) J.Am. Chem. SOC.106,3334-3336 28. Capdevila, J., Pramanik, B., Napoli, J., Manna, S., and Falck, J. (1984)Arch. Biochem. Biophys. 231,511-517 29. Chacos, N., Capdevila, J., Falck, J. R., Manna, S., Martin-Wixstrom, C., Gill, S., Hammock, B., and Estabrook, R. (1983) Arch. Biochem. Biophys. 223,639-648 30. Spearman, M., Prough, R., Estabrook, R., Falck, J., Manna, S., Liebman, K., Murphy, R., and Capdevila, J. (1985) Arch. Bwchem. Biophys. 242,225-230 31. Nugteren, D. (1970) Biochim. Biophys. Acta210, 171-176 32. Malmsten, C., Granstrom, E., and Samuelsson, B. (1976) Biochem. Bwphys. Res. Commun. 68,569-576 33. Fitzpatrick, F., and Gorman, R. (1979) Biochim. Biophys. Acta 582944-58 34. Lyons, R., Stanford, N., and Majerus, P. (1975) J. Clin. Invest. 56,924-936