the of 268, no. 9, of 25, pp. 6610-6614 1993 63 1993 ~s.a ... · site of aspirin and related...

THE JOURNAL OF BIOLOGICAL CHEMISTRY 63 1993 by The American Society for Biochemistry and Molecular Biology. Inc. Vol. 268, No. 9, Issue of March 25, pp. 6610-6614 1993

Printed in ~ s . A .

Differential Inhibition of Prostaglandin Endoperoxide Synthase (Cyclooxygenase) Isozymes by Aspirin and Other Non-steroidal Anti- inflammatory Drugs*

(Received for publication, November 9, 1992)

Elizabeth A. Meade, William L. Smith, and David L. DeWittS From the Department of Bwchemktry, Michigan State University, East Laming, Michigan 48824

Murine prostaglandin endoperoxide (PGH) synthase- 1 and PGH synthase-2 expressed in cos-1 cells were found to be differentially sensitive to inhibition by common nonsteroidal anti-inflammatory drugs (NSAIDs). Aspirin completely inhibited bis-oxygenation of arachidonate by PGH synthase-1; in contrast, aspirin-treated PGH synthase-2 metabolized arachidonate primarily to 16-hydroxyeicosatetraenoic acid (15-HETE) instead of PGH2. IDao values were determined for a panel of common NSAIDs by measuring instantaneous inhibition of cyclooxygenase activity using an oxygen electrode. Among common NSAIDs tested, indomethacin, sulindac sulfide, and piroxicam preferentially inhibited PGH synthase-1; ibuprofen, flurbiprofen, and meclofenamate inhibited both enzymes with comparable potencies; and 6-methoxy-2- naphthylacetic acid preferentially inhibited PGH synthase-2. These results demonstrate that the two PGH synthases are pharmacologically distinct and indicate that it may be possible to develop isozyme-specific cyclooxygenase inhibitors useful both for anti-inflammatory therapy and for delineating between the biological roles of the PGH synthase isozymes.

There are two isozymes of prostaglandin endoperoxide (PGH)’ synthase that catalyze the first committed step in prostaglandin synthesis: the conversion of arachidonate to prostaglandin Hz (PGH,). PGH synthase-1 (PGHS-1) was initially purified and cloned from sheep vesicular glands (1- 6) and is constitutively expressed in most tissues (7) and in blood platelets (8). PGHS-1 is probably involved in the production of prostaglandins involved in cellular “housekeeping” functions, such as coordinating the actions of circulating hormones (9) and regulating vascular homeostasis. PGH synthase-2 (PGHS-2), which shares about 62% amino acid identity with PGHS-1, is expressed only following cell activation (10, 11). The biosynthesis of PGHS-2 is stimulated by serum, growth factors, and phorbol esters in fibroblasts (7, 12), by

* This work was supported by National Institutes of Health Grants GM40713 (to D. L. D.), DK42509 (to W. L. S.), DK22042 (to W. L. S.): and Training Grant HL07404 (to E. A. M.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertise- ment” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$ To whom correspondence should be addressed 510 Biochemistry Bldg., Dept. of Biochemistry, Michigan State University, East Lan- sing, MI 48824. Tel.: 517-353-5284; Fax: 517-353-9334.

The abbreviations used are: PGH, prostaglandin H; 15-HETE, 15-hydro~y-(Z,Z,Z,E)-5,8,11,13-eicosatetraenoic acid; NSAID, nonsteroidal anti-inflammatory drug; PBS, phosphate-buffered saline; PGHS-1, prostaglandin H synthase isozyme-1; PGHS-2, prostaglandin H synthase isozyme-2; 6-MNA, 6-methoxy-2-naphthylacetic acid.

chorionic gonadotrophin in ovarian follicles (13, 14), and by lipopolysaccharide in monocyte/macrophages (15, 16). The observations that expression of PGHS-2 is stimulated by mediators of inflammation and in immune cells and that expression is inhibited by glucocorticoids (17-19) suggest that PGHS-2 may produce prostanoids involved in inflammation and/or mitogenesis.

The cyclooxygenase activity of PGH synthase is the target site of aspirin and related nonsteroidal anti-inflammatory drugs (NSAIDs) (20, 21); most NSAIDs, including aspirin, are competitive inhibitors of PGHS-1 (22). We set out to determine if these agents also inhibit PGHS-2 and, in addition, if there is any selectivity among common NSAIDs toward one or the other isozymes.

MATERIALS AND METHODS

Expression Vectors-The murine PGHS-1 and PGHS-2 were expressed individually in cos-1 cells as described previously (23). Two plasmid constructions were used for these experiments: pSVT7- PGHS-1, in which the murine PGHS-1 cDNA was subcloned into the SV40-based expression vector pSVT7, and pSVL-PGHS-2, in which the murine PGHS-2 cDNA was subcloned into the SV40-based expression vector pSVL. Each parent vector was chosen empirically, based on the efficiency of expression of the individual cDNAs.

Cyclooxygenase Assays-Microsomal membranes prepared from transfected cells were used to assay for cyclooxygenase activity and inhibition by NSAIDs as described previously (241, with the exception that assays were conducted with 10 p~ arachidonate. Specific activities for the PGHS-1 and PGHS-2 averaged 27 k 13 and 23 k 6 pmol of Oz/min/mg of microsomal protein, respectively. Membrane preparations were adjusted to approximately 200 pmol of 02/min of activity/ml, and a standard 10 pmol 02/min activity was used for each assay for IDw determinations. Because many of the NSAIDs tested exhibit complex non-Michaelis-Menten inhibition kinetics when preincubated with PGHS-1 (22, 25), we chose to compare the IDW values for instantaneous inhibition of the cyclooxygenase activity of the two isozymes (22). Representative compounds were chosen from the most common chemical families of NSAIDs (26). Inhibitor concentrations used in our tests ranged from 0.01 to 1000 pM, de- pending on the level of inhibition and the solubility of the compounds.

Indomethacin, acetaminophen, sulindac sulfide, meclofenamate, and aspirin were purchased from Sigma. The S-isomer of ibuprofen was purchased from Aldrich. Piroxicam was a gift of Dr. Thomas Carty at Pfizer. 6-Methoxy-2-naphthylacetic acid (6-MNA) was from SmithKline Beecham. Flurbiprofen was resuspended in 0.1 M Tris- C1 buffer containing 1 mM phenol (Tris-phenol); piroxicam was resuspended in acetone; all other NSAIDs were resuspended in ethanol. Inhibitors were added to reaction mixtures just prior to the addition of enzyme. Vehicle controls were performed for each inhibitor. All inhibitors were fully soluble in the assay buffer at the concentrations used.

Product Characterization-For product characterization studies, aspirin (100 p ~ ) was added 40 h following transfection to the media of cos-1 cells (5 X IO6 cells) expressing either PGHS-1 or PGHS-2, or to sham-transfected cells (no DNA). Following a 10-40-min incubation, the cells were washed, harvested, and resuspended in PBS containing 25 p~ [“Clarachidonate (53 mci/mmol) for 15 min. Radioactive products were then extracted and separated by thin layer

6610

Aspirin Alters the Oxygenase Activity of PGH Synthase-2 661 1

chromatography as described previously (27). Following visualization by autoradiography, prostaglandin formation was quantitated by densitometry on a Bioimage Visage 110 image analyzer.

For mass spectroscopy analysis, cells transfected with the PGHS- 2 expression vector were treated with aspirin for 40 min, washed with PBS, then harvested and incubated with unlabeled arachidonic acid in PBS or with PBS alone. The products of these reactions were extracted and separated by thin layer chromatography (27). Addi- tional transfected cells were also incubated with [1-"C]arachidonic acid, and the products from these reactions were applied in separate lanes to allow for localization of the unknown compound by autoradiography. The silica gel corresponding to the RF of the unknown was separately scraped from lanes spotted with products obtained from cells incubated with and without arachidonate. An authentic 15- hydroxyeicosatetraenoic acid (15-HETE) control (Cayman) was also prepared. Both samples and the 15-HETE were processed in parallel. The silica gel was extracted twice with 2 ml of methanol, which was then pooled and dried under nitrogen. The residues were resuspended in ethyl acetate and extracted twice with 1 ml of water. The ethyl acetate was evaporated under a stream of nitrogen. The extracted products and control samples were treated with bi(trimethysily1)- trifluoroacetamide containing 1% trimethylchlorosilane for several hours for conversion to trimethysilyl-ether/ester derivatives. The derivatized samples were analyzed on a Joel JMS AX505H magnetic sector gas chromatograph-mass spectrometer using a DBS-MS, 30- meter capillary column (0.32 mm, inner diameter; 0.25-pm coating) at the Michigan State University Mass Spectroscopy Center. The chromatography profile was from 150-325 "C at 5 "C/min. Electron ionization mode was employed with an ionization current of 100 mA.

RESULTS AND DISCUSSION

cos-1 cell populations were transfected with the plasmid expression constructs for PGHS-1 (pSVT7-PGHS-1) or for PGHS-2 (pSVL-PGHS-2). Microsomal membranes were then prepared from these transfected cells and used for oxygenase assays to determine kinetic constants for the two murine PGH synthase isozymes. Initial measurements of cyclooxygenase activity with varying arachidonate concentrations es- tablished that the K , values of PGHS-1 and PGHS-2 for arachidonic acid were essentially the same: 3.0 and 2.5 p ~ , respectively. We used 10 p~ arachidonate for all subsequent oxygenase assays to determine the IDbo values for cyclooxygenase inhibition by various nonsteroidal anti-inflammatory drugs (NSAIDs); this substrate concentration was high enough to give near-maximal oxygenase activity, but low enough to permit detection of inhibition by some of the less water-soluble inhibitors.

The NSAIDs tested in this study could be loosely grouped into three categories based on their relative abilities to in- stantaneously inhibit the oxygenase activity of the PGHS-1 and PGHS-2 enzymes (Table I). In the first group were meclofenamate and the propionic acid derivatives ibuprofen and flurbiprofen, three drugs that inhibited PGHS-1 and PGHS-2 with comparable potencies (Table I). Differences between the ID,, values toward PGHS-1 and PGHS-2 for this group of NSAIDs ranged from about a 2-fold preference for PGHS-2 for ibuprofen to about a 7-fold preference for PGHS- 1 for meclofenamate. The substrate analog docosahexanoic acid also inhibited both isozymes equally. A second group of compounds included NSAIDs that were about 10-30-fold better inhibitors of PGHS-1 than PGHS-2. Included in this group were piroxicam and two structurally related members of the acetic acid family of NSAIDs, indomethacin and sulindac sulfide. The magnitudes of the differences in IDb0 values for this group are illustrated by the dose-response curves for indomethacin shown in Fig. 1. At 20 p~ indomethacin, PGHS-1 was completely inhibited, whereas PGHS-2 retained about 80% of its activity. Of the NSAIDs tested, only a single compound, 6-methoxy-2-naphthylacetic acid, the active metabolite of nabumetone (Relafen") (28), was found to have a significant selectivity for inhibition of PGHS-2. 6"NA was about a 7-fold better inhibitor of PGHS-2 than PGHS-1.

TABLE I Inhibition of murine PGH synthase isozymes by non-steroidal anti-

inflammatory drugs Values presented represent the range of values obtained from

individual experiments, and not the statistical variance of a single measurement.

Specificity Inhibitor PGHS-1 PGHS-2 ID, (PGHS-1) ID, for ID, for ID, (PGHS-2)"

~~~~~~~~~~~

WM P M

Equipotent Flurbiprofen 0.46-0.50 2.1-3.4 5.7 (S)-Ibuprofen 8.9-14 7.2-8.2 0.67 Meclofenamic 2.0-2.5 13-18 6.92

Docosahexae- 11 13-17 1.36 acid

noic acid PGHS,-l Piroxicam 9.0-24 70-240 9.54

Indomethacin 4.9-8.1 130-160 22.3 Sulindac sulfideb 0.3-0.5 11-14 30.9

PGHSmu-2 6-MNA' 200-280 15-55 0.14 a Ratio of average IDm values.

The active metabolite of sulindac. 6-Methoxy-2-naphthyl acetic acid, the active metabolite of na-

bumetone (Relafen@).

100 * I

0 1 0 - 9 1 0 - 8 10-7 O O O O I 0001

[lndornelhactn]. M

FIG. 1. Dose-response curve for inhibition by indomethacin of murine PGHS-1 and PGHS-2 cyclooxygenase (COX) activities. Cyclooxygenase activities of microsomes from cos-1 cells expressing murine PGHS-1 or PGHS-2 were measured in the pres- ence of 10 p~ arachidonate as describedpreviously (24). The maximal activities in the absence of inhibitor (100%) for PGHS-1 and PGHS- 2 were 23 and 10.9 nmol of arachidonate consumed/min/mg of microsomal protein, respectively.

Acetaminophen, a commonly used analgesic and antipyretic drug that lacks the anti-inflammatory properties common to true NSAIDs (29), was also tested; however, at concentrations as high as 100 pM, acetaminophen had no detectable effects on the oxygenase activities of either PGHS-1 or PGHS-2 (data not shown).

IDao values for each NSAID were determined from microsomes prepared from at least two different transfections of cos-1 cells. Each ID60 value was calculated from approximately 10 measurements made, at minimum, in duplicate (see Fig. 1); most of the IDm values agreed closely between separate experiments. The largest variations in IDso values, about 3- fold, were observed for piroxicam and 6-MNA, but the variances in these values are less than the 5-fold variances in K,,, values previously reported for the sheep vesicular gland PGHS-1 (30), values that were also determined by oxygen electrode measurements. Transient expression of the murine PGH synthase enzymes is, at present, the only method available for obtaining sufficient quantities of both PGH isozymes from any single species to allow direct kinetic measurements comparing relative PGHS-1 and PGHS-2 inhibition. Al- though PGHS-1 is expressed at high levels in vesicular gland, no cell lines or tissues are known that express high levels of only PGHS-2. That the seven NSAIDs tested show a unique

6612 Aspirin Alters the Oxygenase Activity of PGH Synthase-2

PGHS-2

60

, , , , ,

10

0 0 1 0 2 0 3 0

T h e (min)

FIG. 2. Time course for inhibition of murine PGHS-1 and PCHS-2 cyclooxygenase activities by aspirin. Microsomal membrane preparations from cos-1 cells expressing either murine PGHS-1 or PGHS-2 were incubated at 37 “C with or without freshly prepared aspirin (final concentration = 200 phi). At the indicated times aliquots were removed and assayed for oxygenase activity (24). Remaining activity is presented as a percentage of the control activity (no aspirin) for each time point.

pattern of relative inhibition for PGHS-1 and PGHS-2 dem- onstrates that these isozymes are pharmacologically distinct. Furthermore, these comparisons of instantaneous inhibitor IDm values for different classes of NSAIDs should be useful in defining those inhibitor structures important for interac- tion with each isozyme, and potentially this information can be exploited to identify additional drugs or to guide the synthesis of new compounds with even greater selectivity. It is unlikely, however, that a direct correlation will be found between selectivity for instantaneous inhibition and i n vivo selectivity for all of the NSAIDs tested. For instance, indomethacin appears to selectively inhibit murine PGHS-1 activity, but this drug has been shown to be a time-dependent “irreversible” inhibitor of ovine PGHS-1 (22, 25). We have likewise found that indomethacin is a time-dependent irreversible inhibitor of the two mouse isozymes, as preincubation with this drug leads to rapid, near-complete inhibition of oxygenase activity of both murine PGHS-1 and PGHS-2 at indomethacin concentrations as much as 10-fold lower than the IDso values for instantaneous inhibition.’ It thus seems likely that indomethacin will have less selectivity in vivo than would be predicted from the IDm values for instantaneous inhibition. However, it is also likely that reversible competitive inhibitors that are structurally related to indomethacin (e.g. sulindac sulfide) will possess in vivo inhibitor characteristics similar to their instantaneous characteristics, al- though this can only be confirmed when appropriate in vivo cell or tissue systems are developed.

Studies employing purified PGHS-1 obtained from sheep vesicular gland have demonstrated that aspirin inhibits the cyclooxygenase activity of this enzyme via irreversible acetylation of the “active site” serine residue at position 530 (24). In accord with these previous findings, we found that aspirin treatment resulted in a time-dependent loss of the oxygenase activity of murine PGHS-1 (Fig. 2). The half-life ( t l l z ) of microsomal murine PGHS-1 oxygenase activity in the pres- ence of 200 pM aspirin at 37 “C was 25 min, nearly identical to the tllz for aspirin inactivation of ovine PGHS-1 (24). Because the amino acid sequences surrounding the active site serines are highly conserved between PGHS-1 and PGHS-2, we anticipated that aspirin would acetylate and inhibit PGHS-2, as it does PGHS-1. Surprisingly, treatment of PGHS-2 with aspirin failed to significantly inhibit the initial rate of oxygenation of arachidonate (Fig. 2). Following a 30-

E. A. Meade, unpublished results. Flurbiprofen was also found to exhibit time-dependent inhibition of the PGHS-2 enzyme.

min incubation with 200 I.IM aspirin, the initial rate of oxygen consumption by PGHS-2 remained at greater than 90% of the control value. One possible explanation for this anomalous observation was suggested by Holtzman et al. (31), who reported recently that aspirin treatment altered the oxygenase activity of an unidentified novel ovine cyclooxygenase, result- ing in the formation of 15-hydroxy-5,8,11,13-eicosatetraenoic acid (15-HETE) instead of the normal product, PGH,. There- fore, we examined the products formed from [14C]arachidonate by aspirin-treated PGHS-2. Initially, the reaction products were examined by thin layer chromatography. As ex- pected, aspirin treatment of cos-1 cells expressing PGHS-1 completely inhibited formation of prostaglandins (Fig. 3). Aspirin treatment of cells expressing PGHS-2 resulted in about a 90% inhibition of prostaglandin synthesis, but there was a corresponding increase in the synthesis of a new product, which co-chromatographed with 15-HETE (Fig. 3A). Isolation of this product following thin layer chromatography and subsequent gas chromatography-mass spectroscopy confirmed that this product was 15-HETE. The compound had the same retention time and the identical mass spectrum as the trimethysilyl-ester/trimethysilyl-ether derivative of authentic 15-HETE, including a base ion of m/z 225 and ions of m/z 374 (M - 90, loss of MeaSiOH), 449 (M - 15, loss of methyl), and 464 (MI. We have not determined the stereo- chemical configuration of 15-HETE formed from aspirin- treated murine PGHS-2; however, 15-HETE produced following aspirin treatment of cultured ovine epithelial cells is predominantly (3:l) in the 15R conformation (31).

Treatment of transfected cos-1 cells in culture with 100 p~ aspirin resulted in a time-dependent inhibition of prostaglandin production from [Wlarachidonate in cells expressing either PGHS-1 or PGHS-2. The t1I2 for inhibition was approximately 10 min in each case (Fig. 3); inactivation of PGHS-1 by aspirin was complete by 20 min, whereas cos-1 cells expressing PGHS-2 were able to synthesize prostaglandins at about 10% of the untreated control levels, even after a 40-min aspirin treatment (Fig. 3). In contrast, incubation of whole transfected cos-1 cells expressing PGHS-2 for 40 min with either indomethacin or flurbiprofen, two NSAIDs that, like aspirin, have been shown to cause a time-dependent inhibition of ovine PGHS-1 (22, 25), did not result in 15- HETE production, only in inhibition of PGH, synthesis (data not shown). The time dependence of the aspirin effect on 15- HETE production and the lack of 15-HETE production following treatment with other time-dependent inhibitors of PGHS-2 suggests that aspirin-stimulated 15-HETE production by PGHS-2 results from a unique modification of the enzyme by aspirin, presumably acetylation of PGHS-2.

Obviously, care must be taken when extrapolating our present findings in an animal model system to the effects of NSAIDs in humans. However, one would anticipate that the inhibitor sensitivities of the mouse and the human PGH synthase isozymes will prove to be similar because there is about 90% amino acid identity between human and mouse PGHS-1 and between the human and mouse PGHS-2; in contrast, the human PGHS-1 and PGHS-2 and mouse PGHS- 1 and PGHS-2 share only slightly more than 60% amino acid identity. Additional support for the generality of our inhibitor data comes from recent work in this laboratory with human PGHS-2. Aspirin treatment of this latter enzyme also results in production of 15-HETE.3

Our present knowledge does not allow us to reconcile the specificities of various NSAIDs found in these experiments with their anti-inflammatory profiles and the side effects associated with their use. The i n vivo efficiencies and speci-

E. A. Meade, unpublished results.

Aspirin Alters the Oxygenase Activity of PGH Synthase-2 6613

A SHAM PGHS-1 PGHS-2

0 4 0 0 1 0 2 0 30 4 0 0 1 0 2 0 30 4 0 mi"

PGE - PGF2 :; V-

-15.HETE

/ ".." -. 100 ;c

80 7 w / PGHS-1 P m d u

J /PGHS-2 P r o d W

0 5 1 0 1 5 20 25 30 35 40 o 5 1 0 15 20 25 30 35 4 0 Time of aspirin pre-incubation (min) Time of aspirin pre-incubation Win)

FIG. 3. Time course for inhibition of prostaglandin formation by aspirin. Aspirin (100 PM) was added to the medium of cos-1 cells 40 h following transfection with either PGHS-1- or PGHS-2-containing expression vector or following sham transfection (with no DNA). Following incubation with aspirin for the indicated times, cells (5 X IO6) were washed and harvested, and the cell suspension was incubated with 25 PM [1-"Clarachidonic acid (53 mCi/mmol) for 15 min. Radioactive products were then extracted and separated by thin layer chromatography as described previously (27). Following visualization by autoradiography, prostaglandin formation was quantitated by densitometry on a Bioimage Visage 110 image analyzer. Values plotted in panels B and C are the ratios of the integrated optical density of each metabolite, or arachidonate, relative to the integrated optical density for all products plus unconverted arachidonate. Several PGH synthase-dependent, arachidonate-derived compounds were formed during these incubations that were not identified and are not labeled in the figure. A , autoradiogram demonstrating the time-dependent loss of prostaglandin formation by murine PGHS-1 and PGHS-2 and the time-dependent increase in 15-HETE production by PGHS-2 following aspirin treatment. B, time course of aspirin inhibition of prostaglandin formation by murine PGHS-1. C, time course of aspirin inhibition of prostaglandin formation and increase in 15-HETE formation by murine PGHS-2.

ficities of NSAIDs are dependent not only on their inhibitor characteristics, but also on factors such as plasma half-life and tissue partitioning (26). Nabumetone, the parent compound of 6-MNA, has been found to be an effective anti- inflammatory agent with a low incidence of gastrointestinal side effects (23), and this could reflect in part its selectivity for PGHS-2. However, in general, any selectivity suggested by our in uitro measurements of inhibitor IDso values should be interpreted cautiously until such time as the in uiuo selectivity can be confirmed.

Aspirin is unique in that it is the only NSAID known to covalently modify PGHS-1 (32). Our finding that aspirin alters the enzymatic activity of PGHS-2 raises a new set of questions about the pharmacological properties of this drug. Recent studies have shown that aspirin used prophylactically may reduce the incidence of cardiovascular disease (33). The beneficial effects of aspirin have been ascribed to its anti- thrombotic effects mediated by irreversible inhibition of platelet PGHS-1 (34). However, the effects of aspirin could be more complex. 15-HETE could affect the function of any of a number of cells and tissues involved in homeostasis and inflammation, including platelets, endothelial cells, smooth muscle, monocytes, and granulocytes. For example, 15-HETE has been reported to inhibit production of pro-inflammatory

leukotrienes B,, C4, and D4 (35,36). Our results, together with the recent findings correlating the frequency of ingestion of aspirin with a reduced risk of development of colon cancer (37), suggest that a more extensive examination of the biological activities of 15-HETE is warranted.

The evidence to date suggests that the constitutively expressed PGHS-1 produces prostaglandins that are involved in housekeeping functions, whereas PGHS-2 is expressed only in activated cells, often in response to mediators of inflammation. We have shown that among currently available NSAIDs, there are those that exhibit some selectivity toward each PGHS isozyme. These results suggest that it will be possible to identify NSAIDs or design new drugs with isozyme specificity. I t is certainly possible that an isozyme-specific NSAID might spare cytoprotective prostaglandin synthesis in the stomach but reduce pro-inflammatory prostaglandin synthesis, thereby reducing or eliminating dyspepsia and ulcer formation, the most common untoward side effects of NSAIDs available currently (26). The availability of isozyme-specific inhibitors would also be valuable in delineating the biological role of these two isozymes.

Acknowledgments-We thank Beverly Chamberlin, who conducted the GC-MS analysis, Dr. Charles C. Sweeley for advice and help in

6614 Aspirin Alters the Oxygenase Activity of PGH Synthase-2

interpreting mass spectra, and Dr. Ron Spangler at SmithKline Beecham for kindly providing 6-MNA.

REFERENCES 1. Hemler, M., Lands, W. E. M., and Smith, W. L. (1976) J. BioL Chem. 261 ,

5575-5581 2. Mi amoto, T., Ogino, N., Yamamoto, S., and Hayaishi, 0. (1976) J. BioL

3. Van Der Ouderaa, F. J., Buytenhek, M., Nugteren, D. H., and van Dorp,

4. DeWitt, D. L., and Smith, &. i. (1988) Pmc. NatL Acad. Sci. U. S. A. 86,

. . . . . - -

8hem 261,2629-2636

D. A. (1976) Biochim. Bio h s. Acta 572,29-42

5. Merlie, J. P., Fagan, D., Mudd, J., and Needleman, P. (19%) J. Biol. Chem.

6. Yokoyama, C., and Tanabe, T. (1989) Bhhem. Biophys. Res. Commun. 263,3550-3553

166,888-894 7. Simmons, D. L., Xie, W., Chipman, J. G., and Evett, G. E. (1991) in

Pmstuglandins, Leukotrienes, Lipoxins, and PAF (Bailey, J. M., ed) pp.

8. Funk, C. D., Funk, L. B., Kennedy, M. E., Pong, A. S., and Fitzgerald, G. 67-78, Plenum Press, New York

A. (1991) FASEB J. 5,2304-2312 9. Smith, W. L. (1989) Biochem J. 269,315-324

1212-1416

10. Fletcher, B. S., Kujubu, D. A., Perrin, D. M., and Herschman, H. R. (1992)

11. Xie, W., Chipman, J. G Robertson, D. L. Erikson, R. L., and Simmons,

12. Ku'ubu, D. A., Fletcher, B. S., Varnum, B. C., Lim, R. W., and Herschman,

13. Sirois, J., and Richarda, J. S. (1992) J. BioL Chem. 267,6382-6388 14. Sirois. J.. Simmons. D. L.. and Richards. J. 11992) J. BioL Chem. 267.

J. BioL Chem. 267,4338-4344

D. L. (1991) Proc. NalL Acad. Sci. U. S. A. 88,2692-2696

k. R. (1991) J. BwL Chern. 266,12866-12872

15.

16.

17.

115j36-11592 .

(1992) J. Bwl. Chem. 267,14547-14550

C. E. (1992) Biochem. Biophys. Res. Commun. 187,1123-1127

7994

. . .

O'Sullivan, G. M., Chilton, F. H., Huggins, E. M., Jr., and McCall, C. E.

O'Sullivan, M. G., Huggins, E. M., Meade, E. A., DeWitt, D. L., and McCall,

Kujubu, D. A., and Herschman, H. R. (1992) J. BWL Chem. 2 6 7 , 7991-

18.

19.

20. 21.

22.

23.

24.

25.

26. 27. 28.

29.

30.

31.

32.

33.

34.

35. 36.

37.

Masferrer, J. L., Seibert, K. Zweifel, B., and Needleman, P. (1992) P m .

Sebaldt, R. J., Sheller, J. R Oates, J. A., and FitzGerald, G. A. (1990) P m .

Vane J. R and Bottin R. (1987) FASEB J. 1,89-96 SmitA, W.'L., DeWitt, b . L., Shimokawa, T., Kraemer, S. A., and Meade,

Rome, L. H., and Lands, W. E. M. (1975) Pmc. Natl. Acad. Sci. U. S. A. 72,4863-4865

Meade, E. A., Smith, W. L., and DeWitt, D. L. (1993) in Prostaglandins, Thromboxanes, Leukotrienes, and Related Compoundp: U p (1993) (Wolfe, L. S., and Ford-Hutchinson, A. W., eds Elsevler ctence Pub-

Natl. Acad. Sci. U. S. A. 89,3917-3921

Natl. Acad. Sci. U. S. A. 87,6974-6978

E. A. (1990) Stroke 2 1 , IV24-IV28

DeWitt, D. L., El-Harith, E. A., Kraemer, s. A., Andrews, M. J., Yao, E. liahers, Amsterdam, in press

F., Armstrong, R. L., and Smith, W. L. (1990) J. Biol. Chem. 266,5192- 51 Wl

K;i-6.lacz, R. J., and Lands, W. E. M. (1985) J. Bwl. Chem. 2 6 0 , 12572-

Brooks, P. M., and Day, R. 0. (1991) N. E 1. J. Med. 324,1716-1725 Smith, W. L., and Lands, W. E. M. (1972)%chemrstry 11,32763282 Haddock, R. E., Jeffery, D. J., Lloyd, J. A., and Thawley, A. R. (1984)

Xenobiotica 14,327-337 Flower R. J., and Vane J. R. (1974) in Prostaglandin Synthetase Inhibitors

(Robinson H. J. and Vane J. R., eds) p. 9-,18, Raven Press, New York Smith, W. L:, and Marnett, i. J. (1991) kkhrm. Bwphys. Acta 1083, 1-

17 Holtzman, M. J., Turk, J., and Shornick, L. P. (1992) J. Biol. Chem. 267,

Roth, G. R., Slok, C. J., and Ozols, J. (1980) J. BioL Chem. 2 6 6 , 1301-

Willard, J., Lange, R. A., and Hillis, L. D. (1992) N. EngL J. Med. 327.

Oates, J. A,, FitzGerald, G. A., Branch, R. A., Jackson, E. K., Knapp, H.

Goetzl, E. J. (1981) Bwchem. Bu, hys. Res. Commun 101,344-350 Vanderhoek, J. Y., Bryant, R. and Bailey, J. M. (1980) J. Biol. Chem.

Thun, M. J., Namboodiri, M. M., and Heath, C. W. (1991) N. Engl. J. Med.

12578

21438-21445,

1304

175-181

R., and Roberts, L. J. (1988) N. Engl. J. Med. 319,689-698

266,10064-10066

326,1593-1596

the of 268, no. 9, of 25, pp. 6610-6614 1993 63 1993 ~s.a ... · site of aspirin and related...

Documents