= . . swmv... 2. . . I . “dizz‘aauu fimuhu. : :35 «U- udJPHI suammuh... :1; Lu... :me 1". “$4.43 a... 1.. .flra u . .49. . . “Wu-kn fl: . 6 :61) 2. :mm. anon—"u . .amuhm‘hfi. f V . wfimwwwfiuflfl . . um“ 3.3:...3 “2&0: .l ‘I . .. , . . .. {fit- .. .u. . . 33.3.3. 4%. mm as. .._ .52.! $2.53. 2 t... n . 53h}. I nima. m? “65.”... V la...“ . 3:151 _ a. .infihwng » .flV . 3...... . Izmir #339 .3... wfifiafifr . ,1 1... O (.1 13:32 2.3833». {mmfi v 01125.". w‘ a. .n 43.4., :1: . > so asked: is .t E. . .. e. i. to: ‘3’...) .2: 4 919%...» .l‘.‘ .F-o. .Elflg r... . ... . firhmwm ls: . (11...? 3.... . . g i! .3 3.1.1; gaitéfa? o..!....¢.......l... stiflfifdgflw s. 9.1 v! . nu. Ecol. a? J? ”twang III-Pi] wmwmfl .yiaTNl. .1 I: 3...... . Slat}. lid-”.3. I. 1:... .s 193 $93.71.... 5:. it...) 3.5.5.35 5-3571}! . 53.... v. :93 A .0. 2:5 r .iflviz ii...) .5, :1). .1 Eli: “151.54.. knfififlhn I. . . bur-n... : x. .5 [.575 .t..h.u.v.twu0w‘ v.9. -. . VII-iiiir. . «$\t.ti.||ll. . Silasriu .41.- f... I. ltfl. . V A. . f7 .0 . \1. .1 a? s A. .. a... . 7.. THESlS llllllllllllllllllllllllllllllllJllllllHllllllllllllll 302060 4132 This is to certify that the thesis entitled Studies involving the cyclooxygenase active sites of prostaglandin endoperoxide H synthase-l and -2 presented by Caroline Jill Rieke has been accepted towards fulfillment of the requirements for MS degree in Biochemistry Wt... gm Major professor Date /2’//3,/§5 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 11/00 chlRC/‘DmDquS-p.“ STUDIES INVOLVING THE CYCLOOXYGENASE ACTIVE SITES OF PROSTAGLANDIN ENDOPEROXIDE H SYNTHASE -1 AND —2 By Caroline Jill Rieke A THESIS Submitted to Michigan State University In partial fulfillment of the requirement For the degree of MASTER OF SCIENCE Department of Biochemistry 1 999 ABSTRACT STUDIES INVOLVING THE CYCLOOXYGNEASE ACTIVE SITES OF PROSTAGLANDIN ENDOPEROXIDE H SYNTHASE-1 AND -2 By Caroline Jill Rieke PGHS-1 and -2 catalyze the conversion of arachidonic acid to prostaglandin H2 via two activities, a cyclooxygenase and a peroxidase activity. Two residues at the mouth of the cyclooxygenase site, Arg120 and Tyr355, are in a position to interact with the substrate, arachidonic acid, and with non-steroidal anti- inflammatory drugs (NSAIDs). The role of these two residues in fatty acid substrate binding and oxygenation and in NSAID interactions was investigated using site-directed mutagenesis. We determined that Arg120 is necessary in both PGHS-1 and -2 for interaction with carboxylate containing NSAIDs. In addition, we showed that the Arg120 of the two isozymes interacts differently with arachidonic acid. A R1200 mutation of oPGHS-1 greatly affected catalysis while a R1200 mutation in hPGHS-2 had no affect on catalysis. We also determined that Tyr355 plays a role in stereoselectivity of the S-isomer of the 2-phenylpropionic acid class of NSAIDs and that it is involved in binding and oxygenation of fatty acid substrates. TABLE OF CONTENT LIST OF TABLES iii LIST OF FIGURES iv LIST OF ABBREVIATIONS vi CHAPTER 1: LITERATURE REVIEW Introduction 1 Prostanoid Biosynthesis 3 Primary Structure 5 Regulation of PGHSs 6 Biochemical and structural properties of PGHSs Crystal structure of PGHSs 13 Active site of PGHSs 17 Reactions of PGHSs 17 Substrates of PGHSs 21 Inhibition of PGHSs 23 Thesis overview 25 CHAPTER 2: STUDIES TO DETERMINE THE ROLE OF ARGIZO oPGHS-1 AND ARG12O hPGHS-2 IN SUBSTRATE AND INHIBITOR INTERACTIONS Introduction 26 Materials and Methods 32 Results Discussion 53 CHAPTER 3: STUDIES TO DETERMINE THE ROLE OF TYR355 oPGHS-1 IN SUBSTRATE AND INHIBITOR INTERACTIONS Introduction 57 Materials and Method 60 Results 64 Discussion 71 iii CONCLUSION REFERENCES iv 74 76 LIST OF TABLES Table 1. Kinetic properties of hPGHS-2 and R1200 hPGHS-2 Table2. Comparison of substrate specificity of hPGHS-2 and R1200 hPGHS-2 Table 3. Comparison of inhibition of hPGHS-2 and R1200 hPGHS-2 by non- steroidal anti-inflammatory drugs Table 4. Comparison of Kinetic properties of oPGHS-1 and Y355 mutant oPGHSs Table 5. Substrate Specificities of native oPGHS-1 and Y355 mutant oPGHS-1s 40 46 52 66 68 Fig Fig Fig Fig Fig: FIgL LIST OF FIGURES Figure 1. The Arachidonate cascade 2 Figure 2. Biosynthetic pathway for prostanoid synthesis 4 Figure 3. Comparison of deduced amino acid sequences of various PGHSs 8 Figure 4. Crystal structure of ovine PGHS-1 and murine PGHS-2 12 Figure 5. Kabsch Sander representations of ovine PGHS-1 and murine PGHS-2 showing similarity in secondary structure 16 Figure 6. Model of the active site for ovine PGHS-1 18 Figure 7. The Ruf Model 20 Figure 8. Pathway for formation of the 1-2- and 3-series prostaglandins 22 Figure 9. Model of active site of oPGHS-Iand R1200 oPGHS-1 with arachidonic acid modeled into the cyclooxygenase active site 29 Figure 10. Model of active site of mPGHS-2 and R1200 mPGHS-2 with arachidonic acid modeled into the cyclooxygenase active site 31 Figure 11. Time course for inhibition of R1200 oPGHS-1 by flurbiprofen 38 Figure 12. Western blot analysis of hPGHS-2 and R120 mutant PGH synthases 41 Figure 13. Product characterization of native and R120 mutant PGH synthases 42 Figure 14 Time-dependent inhibition of hPGHS-2 and R1200 hPGHS-2 by flurbiprofen 45 Figure 15. Structures of PGHS-2 specific inhibitors tested with hPGHS-2 and R1200 hPGHS—2 47 vi Figure 16. Figure 17. Figure 18. Figure 19. Figure 20. Figure 21 Figure 22. Time-dependent inhibition of hPGHS-2 and R1200 hPGHS-2 by NS-398 49 Time-dependent Inhibition of hPGHS-2 and R1200 hPGHS-2 by Dup-697 50 Time-dependent inhibition of PGHS-2 and R1200 hPGHS-2 by SC-58125 51 Model of active site of oPGHS-1 and Y355F oPGHS-1 with arachidonic acid modeled into the cyclooxygenase site 59 Western blot of oPGHS-1 and Y355A and Y355L mutants PGH synthases 63 Product characterization of hPGHS-2 and Y355 mutant PGHSs 67 R vs S Ibuprofen discrimination by native oPGHS-l and Y355 mutant oPGHS-1s 69 vii 20:4 n-6 20:3 n-6 20:5 n-3 20:2 n-6 18:2 n-6 1 1-HET E 15-HETE 12-HHT 15- HPETE BSA COX CPLA2 DMEM EGF leo Km NSAID PBS PGHS. oPGHS-1 hPGHS-1 POX SDS-PAGE SPLA2 TBS TLC Vmax ABBREVIATIONS arachidonic acid (AA) dihomo-y—Iinolenic acid eicosapentanoic acid (EPA) eicosadianoic acid .linoleic acid 1 1 -hydroxy-5,8,12,14-eicosatetraenoic acid 15-hydroxy-5,8,1 1 ,13-eicosatetraenoic acid 12-hydroxy-5,8,1 O-heptadecatrienoic acid 15-hydroperoxy-5,8,1 1 ,13-eicosatetraenoic acid bovine serum albumin cyclooxygenase cytosolic, 85 kDa phospholipase A2 Dulbecco’s modifies Eagle medium epidermal growth factor concentration for 50% inhibition Michaelis—Menton constant non-steroidal anti-inflammatory drug phosphate-buffered saline prostaglandin endoperoxide H synthase ovine prostaglandin synthase isozyme 1 human prostaglandin synthase isozyme 2 peroxidase sodium dodecyl sulfate polyacrylamide gel electrophoresis secretory phospolipase A2 tris-buffered saline thin layer chromotagraphy maximum velocity of a reaction viii PFC anc PGI DIOs Dies CHAPTER 1 LITERATURE REVIEW Introduction Arachidonic acid, a 20 carbon polyunsaturated essential fatty acid, is the precursor for the formation of the eicosanoids. The eicosanoids. which consist of the prostanoids, Ieukotrienes and epoxyecosatetraenoic acids (EETs), are produced via the actions of the enzymes of the arachidonic acid cascade (Figure 1). There are three major pathways within this cascade, with each being named for the central enzyme of the pathway. The prostanoids, which consist of the prostaglandins, prostacyclins, and thromboxanes are produced via the cyclo- oxygenase pathway. The Ieukotrienes are produced through the actions of lipoxygenases, and the EETs are produced through the actions of P450 epoxygenases. The Iipoxygenase and epoxygenase pathways are reviewed in Ref. 1. Prostaglandin endoperoxide H synthase (PGHS) is the enzyme which is responsible for catalyzing the committed step in the formation of prostanoid products. This enzyme has two enzymatic functions, a cyclooxygenase (COX) and a peroxidase (POX) activity. These activities convert arachidonic acid to PGH2 which, in turn, can be further metabolized to the individual prostaglandins, prostacyclins, and thromboxanes by the appropriate synthases (2). These latter prostanoids are then able to act in an autocrine and/or paracrine fashion /\/\/‘\/'\/‘\/'\/\/C°°" ARACl-IDOMC ACID 20 PGH SYNTHASE o 5,12,AND15- 0 P450 2 (CYCLoovaENASE) 2 urovaENAsm 2 EPOXYGENASE PROSTAGIANDINS (PCS) LEUKOTRIENES (LTs) EPOXY ACIIB (EETs) THROMBOXANES (TXS) HYDROXY ACIDS (HETEs) HYDROXY ACIDS LIPOXINS (LXs) DIHYDROXY ACIIB Figure 1. The arachidonate cascade. Arachidonic acid is metabolized by three different enzymes systems: the cyclooxygenase, the lipoxygenase and the P450 epoxygenase pathways through the actions of G-protein linked receptors with the subsequent generation of second messengers such as cAMP or Ca++ in order to mediate their physiologic responses (3.4). Prostaglandins are found in all animal tissues, though not necessarily in all cells of each tissue. They play a broad role in basic physiological events. The prostanoids have been shown to play a role in vascular homeostasis, platelet aggregation, renal water homeostasis, parturition and ovulation (5,6) and in the inflammatory responses to many different stimuli (2). In addition, they have been proposed to have a role in the development of colon cancer (7,8) and Alzheimer's disease (9). PGHS is the therapeutic target of non-steroidal anti-inflammatory drugs (NSAIDs) such as aspirin and ibuprofen (2) and as such is the focus of research concerning the mechanisms of catalysis and inhibition. Prostanoid Biosynthesis Synthesis of the prostanoid products can be broken down into three steps: 1) Stimulus-activated release of arachidonic acid; 2) conversion of arachidonic acid to PGHz; 3) cell specific conversion of PGH2 to a biologically active prostanoid (Figure 2). Following an extracellular stimulus, arachidonic acid is released from membrane phospholipids through the actions of phospholipases. Two classes of phospholipases have been shown to be involved in arachidonic acid release HORMONE ( ceu. menus mosmoupm (PI,PC,PE). / TRIGLYCERIDE. on ACTIVATION CHOLESTEROL esren I i LipAsas) I .....___.. ._.. _. coon ARACHIDONIC acre CYCLOOXYGENASE PGH SYNTHASE 0,- ~/"V\/ coon Penomoase 0,, ~/-vv°°°" \ _—.l PGH: OH HO‘ / \m mm SYNTHASES :2 . ' p6.2 ””2 bu / ‘ \ maVWV o- - 0" ./-v\/°°°" b ’\==/\/\coou HO‘ ‘ ' 'OH Figure 2. Biosynthetic pathway for prostanoid synthesis in vitro (10,11,12,13). These are cytoplasmic PLA2 (cPLAg) and secretory PLA2 (SPLAz). The 85 kDa CPLAz is regulated by low concentrations of Ca“(1-10 pM) and phosphorylation and is relatively specific for the arachidonate group located at the sn-2 position of phospholipids (14,15). The 14 kDa SPLAz requires high concentrations of Ca++ (1 mM) and is not specific for any particular phospholipid head group nor for the acyl group at the sn-2 position (16). Once the arachidonic acid is released, it is then converted to PGH2 through the actions of PGHSs. The conversion of arachidonic acid to PGH2 is initiated by the removal of the 13 pro- § hydrogen and 2 molecules of oxygen are added at the 11 and 15 positions to give PGGz. The hydroperoxide at the C-15 position is subsequently reduced to an alcohol to give PGH2 (17). PGH2 then can undergo conversion either non- enzymatically or enzymatically to give the individual prostaglandin products (4). Primary Structure Two isozymes of PGHS have been described, and are encoded by separate genes. The first isozyme, PGHS-1 was initially purified from sheep vesicular gland (18,19) and the cDNA cloned (20). cDNAs for PGHS-1 have since been cloned from murine (21), human (22) and rat sources (23). There is approximately 90% sequence identity among the PGHS-1 enzymes. In the early 19905 a second isozyme, PGHS-2 ,was discovered as a v-src inducible gene product in chicken fibroblasts (24) and a phorbol ester inducible immediate early gene product in murine 3T3 cells (25). PGHS-2 cDNAs from human (26,28), rat (23) and ovine (27) sources have also been cloned. Within a species PGHS-1 and PGHS-2 have 60% sequence identity, with all residues identified as important for catalysis being conserved (Figure 3). The most notable differences between the two isozymes are at the amino terminus and at the carboxyl terminus. The signal peptide at the amino terminus of PGHS-1 contains a series of hydrophobic amino acids not present in the signal peptide of PGHS-2, while the carboxyl terminus of PGHS-2 has a unique 18 amino acid “cassette”. The function of this “cassette” has not yet been identified, although it has been postulated that it may have a role in targeting, in protein degradation or possibly in protein-protein interactions. Regulation of PGHSs The two isozymes of PGHS are regulated differently. PGHS-1 is normally present in many tissues, while PGHS-2 is usually undetectable in most tissues (29).The exceptions to this are brain (30) and kidney (31). Upon stimulation with growth factors, cytokines or mitogens (32,28,33,34), PGHS-2 mRNA and protein levels are dramatically upregulated. In addition, PGHS-2 is down-regulated by dexamethasone (34). Because PGHS-1 is typically present under normal conditions it has been proposed that this isozyme is involved in synthesizing prostanoids involved in “housekeeping“ functions such as regulation of stomach acid secretion and renal homeostasis. lnpart because PGHS-2 is expressed by cells of the immune system, this isozyme is believed to be involved in inflam- Figure 3. Comparison of the deduced amino acid sequences of various PGHSs. The amino acid sequences of various PGH synthases 1 and 2 are aligned for comparison. Numbering corresponds to oPGHS-1, starting with Met 1 of the deduced sequence. Chick—2 Human-2 House-2 House-1 Human-1 Sheep-1 Chick-2 Human-2 House-2 House-1 Human-1 Sheep-1 Chick-2 fiuman-Z Mouse-2 House-1 Human-1 Sheep-1 Chick-2 Human-2 Mouse-2 House-1 Human-1 Sheep-1 Chick-2 Human-2 Mouse—2 mouse-1 Human-1 Sheep-1 Chick-2 Human-2 Mouse-2 House-1 Human-1 Sheep—1 Chick-2 Human-2 Mouse-2 Hausa-1 Human-1 Sheep-1 Chick-2 Eamon-2 House-2 House-1 Human-1 Sheep-1 HLLPCALLAALL-——— AAOIAANPCCSLPCQNRGECHTTGFDRYECD “LARALLLCAVL ------ ALSBTANPCCSHPCQNRGVCHSVGFDQYKCD KLIRAVLLCAAL—— =GLSQAANPCCSNPCQNRGECHSTGFDQYKCD HSRRSLSLHFPLLLLLLLPPTPSVLLADPGVPSPVNPCCYYPCQNQGVCVRFGLDNYQCD KSR-SLLLR -PPLP-VLLADPGAPTPVNPCCYYPCQHQGICVRFGLDRYQCD FLLLLLLL MSRQSISLRPPLLLLLL-SPSP-VPSADPGAPAPVNPCCYYPCQHQGICVRFGLDRYQCD 10 20 30 40 SO CTRTGYYGENCTTPEFFTWLKLLIKPTPNTVHYILTHFKGVWNIINNSPFLRDTIHRYVL CTRTGFYGENCSTPEFLTRIKFLLKPTPNTVHYILTBFKGFWNVVNNIPFLRNAIHSYVL CTRTGFYGENCTTPEFLTRLKLLLKPTPNTVHYILTHFKGVWNIVNNIPFLRSLTMKYVL CTRTGYSGPNCTIPEIWTWLRNSLRPSPSFTHFLLTHGYWLWEFVNAT-FIREVLMRLVL CTRTGYSGPNCTIPGLWTWLRNSLRPSPSFTHFLLTHGRWFWEFVNAT-FIREHLMLLVL CTRTGYSGPNCTIPEIWTWLRTTLRPSPSFIHFLLTHGRWLWDFVNAT-FIRDTLHRLVL 7O 80 90 100 110 TSRSHLIDSPPTYNSDYSYKSWEAYSNLSYYTRSLPPVCHDCPTPHGVXGKKELPDSKLI TSRSHLIDSPPTYNADYGYKSWEAFSNLSYYTRALPPVPDDCPTPLGVKGKKQLPDSNEI TSRSYLIDSPPTYNVHYGYKSWEAFSKLSYYTRALPPVADDCPTPMGVKGNKELPDSKEV TVRSNLIPSPPTYNSAHDYISHESFSNVSYYTRILPSVPKDCPTPHGTKGKKQLPDVQLL TVRSNLIPSPPTYNSAHDYISWESFSNVSYYTRILPSVPKDCPTPHGTKGKKQLPDAQLL TVRSNLIPSPPTYNIAHDYISWESFSHVSYYfggLPSVPRDEESPHGTKGngngAEFL 120 130 140 VEKFLLRRKFIPDPQGTNVMFTFFAQHFTBQFFKTDHKRGPGFTKAYGHGVDLNHIYGET VGKLLLRRKFIPDPQGSNNHFAFFAQHFTEQFFKTDHKRGPAFTNGLGHGVDLNHIYGET LEKVLLRREFIPDPQGSNHHFAPFAQHFTHQFFKTDHKRGPGFTRCLGHGVDLNHIYGET AQQLLLRREFIPAPQCTNILFAFFAQHFTBQFFKTSGKMGPCFTKALGHGVDLGHIYGDN ARRFLLRRKFIPDPQGTNLHFAFFAQHFTBQFFXTSGKMGPGFTKALGHGVDLCHIYGDN SRRELLRRXFIPDPQGTNLMFAFFAQHFTBQFFKTSGKMGPGFTKALGHGVDLGHIYCDN 180 190 200 210 220 230 LERQLKLRLRKDGKLKYQHIDGEHYPPTVKDTQAEHIYPPHVPEHLQFSVGQEVFGLVPG LARQRKIRLFKDGKMKYQIIDGEHYPPTVKDTQAEHIYPPQVPEHLRFAVGQEVFGLVPG LDRQHKLRLFKDGKLKYQVIGGEVYPPTVKDTQVEHIYPPHIPENLQFAVGQEVFGLVPG LERQYHLRLFKDGKLKYQVLDGEVYPPSVEQASVLMRYPPGVPPERQHAVGQEVFGLLPG LERQYQLRLFKDGKLKYQVLDGEHYPPSVEEAPVLHHYPRGIPPQSQHAVGQEVFGLLPG LERSYQLRLPKDGKLKYQMLNGEVYPPSVEEAPVLHHYPRGIPPQSQHAVG3£VFGLLPG 24 250 260 270 280 90 LHMYATIWLREENRVCDVLKQEHPEWDDEQLFQTTRLILIGETIKIVIDDYVQHLSGYHF LMMYATIWLREHNRVCDVLKQEHPEWGDEQLFQTSRLILIGETIKIVIDDYVQHLSGYHF LHHYATIWLREaNRVCDILKQEHPEWGDEQLFQTSRLILIGETIKIVIDDYVQHLSGYHF LMLFSTIWLREHNRVCDLLKEBHPTWDDEQLFQTTRLILICETIKIVIEEYVQHLSGYFL LMLYATLWLREHNRVCDLLKAEHPTWGDEQLFQTTRLILIGETIKIVIEEYVQQLSGYFL LMLYATIHLREBNRVCDLLKAEHPTWGDEQLFgTARLILIGETIKIVIEEYVQgLSGYFL 300 310 320 3 0 340 35 KLKFDPELLFNQRFQYQNRIAAEFNTLYHWKPLLPDTFQIHNQEYTFQQFLYNNSIHLEH KLKFDPELLFNKQFQYQNRIAAEFNTLYHWEPLLPDTFQINDQKYNYQQFIYNNSILLEH KLKFDPELLFNQQFQYQNRIASEFNTLYHWHPLLPDTFNIEDQEYSFKQFLYNNSILLEH QLKFDPELLFRAQFQYRNRIAHEFNHLYHWBPLHPNSFQVGSQEYSYEQFLFNTSHLVDY QLKFDPELLFGVQFQYRNRIATEFNHLYHWHPLMPDSFKVGSQEYSYEQFLFNTSHLVDY QLKFDPELLFGAQFQYRNRIAMEFNQLYHWFPLHPDSFRVGP8DYSYEQFLFNTSMLVDY 360 370 380 390 4 0 410 GLSHMVKSSKRQIAGRVAGGKNVPAAVQKVAKASIDQSRQHRYQSLNEYRKRFHLKPFKS GITQFVESFTRQIAGRVAGGRNVPPAVQKVSQASIDQSRQMKYQSFNEYRKRFMLKPYES GLTQFVESFTRQIAGRVAGGRNVPIAVQAVAKASIDQSREHXYQSLNEYRKRFSLKPYTS GVEALVDAFSRQRAGRIGGGRNFDYKVLHVAVDVIKESREHRLQPFNEYRKRFGLKPYTS GVEALVDAFSRQIAGRIGGGRNHDHHILHVAVDVIRESREMRLQPFNEYRKRFGMKPYTS GVEALVDAFSRQPAGRIGGGRNIDHHILHVAVDVIKESRVLRL PFNEYRKRFGHKPYTS 420 430 440 450 46 470 FEELTGEKEMAAELEELYGDIDAHELYPGLLVEKPRPGAIPGETHVEIGAPFSLKGLHGN FEELTGEKEMSAELEALYGDIDAVELYPALLVEKPRPDAIFGETHVEVGAPFSLKGLMGN FEELTGEKEMAAELKALYSDIDVMELYPALLVEKPRPDAIFGETHVELGAPFSLKGLHGN FQELTGEKEMAAELEELYGDIDALEFYPGLLLEKCQPNSIFGESHIEHGAPFSLKGLLGN FQELVGEKEHAAELEELYGDIDALEFYPGLLLEKCHPNSIFGESHIEIGAPFSLKGLLGN FgELTGEKEMAAELEELYGDIDALEFYPGLLLEKCHPNSIFGESMIEMGAPFSLKGLLGN 80 490 500 $10 520 530 TICSPEYWKPSTFGGKVGFEIINTASLQSLICNNVKGSPFTAFHVLNPEPTETATIUVSI VICSPAYWKPSTFGGEVGFQIINTASIQSLICNNVKGCPFTSFSVPDPELIKTVTIIISB PICSPQYWKPSTFGGEVGFKIINTASIQSLICNNVKGCPFTSFNVQDPQPTKTATINA‘I PICSPEYWKPSTFGGDVGFNLVNTASLKKLVCLflTKTCPYVSFRVPDYPGDDGSVIV--- PICSPEYWKPSTFGGEVGFNIVXTATLKKLVCLNTKTCPYVSFRVPDASQDDGPAVE--- PICSPEYWKASTFGGEVGFNLVKTATLKKLVCLNTXTCPYVSFHVPDPRQEDRPGVE--- 0 S90 540 550 $60 570 IIPIIBDINPILLLKEQSAEL ORSGLDDINPTVLLKERSTEL SIBRLDDIRPTVLIKRRSTEL RRSTEL RPSTEL — -- RPPTEL 600 Figure 3. mation. Studies with knockout mice have shown that the two knockouts have distinct phenotypes PGHS-1 knockout mice appear to develop normally. These mice have no stomach pathologies and exhibit less indomethacin-induced gastric ulceration than wild type mice (35). PGHS-2 knockout mice exhibit abnormal renal development and die early from renal failure (36). Biochemical and Structural Properties of PGHSs PGHSs are classified as integral membrane proteins based on observations that detergents are necessary to solubilize the enzymes from the membrane (19). They are localized to the luminal side of the endoplasmic reticulum and the nuclear envelope (37,38). While it was initially shown, by immunofluorescent staining, that the two isozymes were differentially localized to the two membranes (39), more recent immunogold staining has shown that the two proteins appear to be localized equally between the two membrane systems (40). Initially it was predicted, based on hydrophobicity analysis (41) and trysin cleavage patterns (42), that PGHSs contained one or more transmembrane regions. More recently with the appearance of the crystal structure it has been proposed that the proteins are inserted into one layer of the lipid bilayer via a novel membrane binding domain (43,44). This membrane binding domain consists of four amphipathic helices with the sidechains of the hydrophobic residues inserted into the membrane. Very recently another protein, squalene cyclase, has been shown to contain a similar membrane binding domain (45). PGHS-1 has a MW of 72 kDa on SDS-PAGE (19) while PGHS-2 appears to be a doublet of 72 and 74 kDa (46). This is in contrast to the theoretical molecular weight of 66 kDa and 67 kDa for the two isozymes, respectively (20,46). This difference is due to the addition of several carbohydrate moieties. Both isozymes contain 3 N- glycosylation sites with PGHS-2 having an additional site, which is glycosylated in approximately 50% of the protein, located at the carboxyl terminal (42). PGHSs have three distinct folding domains including an EGF homolgy domain, a membrane binding domain and a large globular catalytic domain (Figure 4). Between the two isozymes the EGF domains exhibit 49% identity, the membrane binding domains show 38% identity and the catalytic domains 70% identity. PGHSs function as homodimers and are heme containing proteins, with one molecule of heme per monomer (18,19). PGHSs have been shown to exist as homodimers by sedimentation analysis (19) and crosslinking studies (46). The crystal structure supports this as the enzyme was crystallized in homodimer form with the dimer interface containing interactions between the EGF domain subunits. 10 Figure 4. Crystal structures of ovine PGHS-1 and murlne PGHS-2 . The three folding domains are colored as listed: EGF domain dark blue, membrane binding domain yellow and catalytic domain green. A. oPGHS-1 B. mPGHS-2 11 \', ‘\ _‘ "‘ ~~-‘/" (. . {swims} 5'; (:36 “if, (~41 \‘v‘: “I" :‘ ‘Vd ‘i'.’ V214 fir)? «.3 'Y o \ "r” ”v ":9 2 ' "o ‘7‘“ H.“ Sp“ _ I‘, 35.: _ . (t Figure 4 Crystal Structure of PGHSs Several years ago the crystal structure of oPGHS-1 complexed with the NSAID, flurbiprofen, was solved (43). The enzyme has a large helical content with little B-sheet. As mentioned previously, the enzyme can be divided into three distinct folding regions; an EGF domain, a membrane binding domain and a catalytic domain. The catalytic domain is a large globular structure which contains both cyclooxygenase and peroxidase active sites. The cyclooxygenase active site is a long hydrophobic channel which opens at the outer surface of the membrane binding domain. The peroxidase active site is a shallow cleft containing the Fe- protophorphyrin IX prosthetic group and is structurally related to the peroxidase site of myeloperoxidase. With the crystal structure it became possible to make hypotheses regarding the function of various active site amino acids. Several amino acids located at the mouth of the cyclooxygenase channel form a hydrogen bonding network involved in binding the carboxylate group of flurbiprofen. These amino acids include Arg120, Tyr355, and Glu524 and it is proposed that in addition to their interaction with NSAIDs they would also interact with the carboxylate group of the substrate. In 1996 the crystal stmcture of PGHS-2 was solved by two groups (47,48). This allowed a direct comparison of the structure of the two isozymes (Figure 5). In general, the two structures are superimposible with almost all residues in the active site being conserved. One notable difference is the substitution of Val for lie at residue 523 in PGHS-2. This results in a slightly larger active site for the 13 PGHS-2 isozyme and possibly explains some of the differences that the two isozymes exhibit with respect to interaction of the various NSAIDs and substrates. Another difference is the substitution of Arg for His at residue 513. This substitution has been postulated to be involved in an alternative hydrogen bonding network in PGHS-2 and interaction with some PGHS-2 specific inhibitors. Luong et al (47) comparison of the crystal structures of hPGHS-2 with two different inhibitors bound in the cyclooxygenase site has provided evidence for flexiblity at the mouth of the active site. In the “closed conformation” Arg120, Tyr355 and Glu524 participate in a hydrogen bonding network at the bottom of the channel similar to that seen in the oPGHS-1 structure, while in the “open conformation” the Arg120 interacts with the backbone carbonyl oxygens of Glu524 and the side chain of Glu524 interacts with Arg513 on the other side of the NSAID binding site. Kurumball et al (48) crystallized the murine PGHS-2 without inhibitor and with several different NSAIDs in the active site, providing evidence for some of the interactions associated with PGHS-2 specific inhibition. Overall they found the structure of the free enzyme to be similar to that of the complexes with inhibitors. They postulated a role for Ar9513 involving an interaction with the sulphonamide group of the PGHS-2 specific inhibitor, 80-558. 14 Figure 5. Kabsch Sander representations of ovine PGHS-1 and murlne PGHS-2 showing similarity in secondary structure. Helices are shown in red, B-sheets are shown in yellow. A. oPGHS-1 B. mPGHS-2 15 Figure 5 Active Site of PGHSs As mentioned previously, all residues that have been shown to be important for catalysis are conserved between the two isozymes. Figure 6 is a representation of the active site of oPGHS-1. Tyr385 is the residue that has been postulated to be the source of the tyrosyl radical that removes the pro-S hydrogen from arachidonic acid (49). Arg120 is located at the mouth of the active site and is postulated to be involved in substrate and inhibitor interactibns (43). Ser530 is the residue that is acetylated by aspirin (50). His338 and 207 are the proximal and distal ligands of the Fe, respectively (43). Tyr355 is postulated to be involved in stereoselectivity of inhibitors of the 2-phenylpropionic acid class (43). Reactions of PGHSs PGHSs have two distinct catalytic activities including a cyclooxygenase component which converts arachidonic acid to the hydroperoxide, PGGZ, and a peroxidase component which catalyzes the two electron reduction of P662 to produce PGH2. The enzymes have been shown to have an absolute requirement for a hydroperoxide (52). Recently it has been demonstrated that the hydroperoxide requirement for PGHS-2 is lower than that for PGHS-1 (53). These catalytic activies have been shown structurally and functionally to occur at distinct but neighboring sites (54). Initial studies on the mechanistic properties of 17 1118207 / ) @momm D Figure 6. Model of the active site for ovine PGHS-1 Residues shown are conserved between the two isozymes and are described in the text. 18 the enzyme were performed using oPGHS-1 but subsequent studies have demonstrated that the reaction mechanism and kinetic properties of the two isozymes are similar (54,55). The cyclooxygenase reaction is initiated by the removal of the 13 pro-S hydrogen of arachidonic acid (57). This is the rate limiting step in the reaction and formation of an arachidonyl radical has been demonstrated (58,59). It is proposed that a protein radical is generated that can remove the 13-pro S hydrogen resulting in the formation of the arachidonyl radical. Formation of a tyrosyl radical has been demonstrated using EPR (60). In addition, it has been shown by site directed mutagenesis that Tyr385 is necessary for COX activity (50). This has lead to the proposal that a Tyr385 radical is responsible for the abstraction of the hydrogen from arachidonate. Subsequent rearrangement and addition of two molecules of oxygen leads to bis-oxygenation at carbons 11 and 15. The resulting P662 diffuses from the cyclooxygenase site to the peroxidase site where the two electron reduction of P662 to the corresponding alcohol, PGH2, takes place. Ruf and coworkers developed a model to explain the relationship between the cyclooxygenase and peroxidase reactions of PGHSs (60) (Figure 7). During the peroxidase reaction a two electron oxidation of the heme group of PGHS by a hydroperoxide yields a peroxidase spectral intermediate l with an oxyferryl form of iron (Fe IV) and a protoporphorin radical cation. The oxidized heme group oxidizes a neighboring tyrosine residue to yield a spectral lnterrnediate Il having an oxyferryl Fe IV and a tyrosyl radical. This lnterrnediate II can then cycle back 19 - u .. mot-€301 Tm " W—\ 0 J mar-‘33! Tm 31v:- ”- mom '1'" : suicioe «Activator! Figure 7. The Hut Model. Model for peroxide-dependent activation of cyclooxygenase activity of PGH synthase via formation of an intermediate tyrosyl radical ( 60). 20 CY affl the to lnterrnediate I or undergo suicide inactivation. The catalytic rate of the two enzymes has been shown to be similar and both isozymes of PGHS have been shown to undergo suicide inactivation (56). The mechanism underlying the suicide inactivation has not yet been determined. Substates of PGHSs The PGHS isozymes have been shown to be capable of oxygenating a variety of 18- and 20- carbon polyunsaturated fatty acids. Both isozymes show a preference for arachidonate (61). The Km for arachidonic acid for both enzymes is 5 pM. PGHSs use arachidonic acid as the substate for the formation of the 2- series prostaglandins. The 1-series of prostaglandins are formed from 8,11,14- eicosatetrienoate while the 3-series prostaglandins are formed from 5,8,11,14,17-eicosapentaenoate (EPA). n-3 and n-9 Polyunsaturated fatty acids are poor substrates for the enzymes, but are effective inhibitors of the oxygenation reaction (62, 63). 22:6 n-3 is a competitive inhibitor for both isozymes without being a substrate for either (62). In addition, PGHSs can use the 18 carbon fatty acids (18:2 and 18:3) for the production of hydroxy fatty acids (Figure 8). Little is known about the roles of the individual residues, within the cyclooxygenase site, in determining substrate specificity for the two isozymes. Laneuville et al demonstrated that the two human isozymes show different affinities for the various fatty acid substrates with PGHS-2 being more flexible in the use of the various substrates (61 ). 21 Arachidonic acid (20:4, n-6) CDOH CECI/VJZ/ Dihorno-yl'nolenic acid (20:3, n-6) Ecxapmtaemic acid (20:5. "-3) — (IDOl-I _ COOH l I , 1-series Prostaglandins 2-series Prostagandins 3-series Prostagland'ns Figure 8. Formation of the 1-, 2- and 3-series prostaglandins PGHSs use arachidonic acid as the substate for the formation of the 2-series prostaglandins. The 1-series of prostaglandins are formed from 8,11,14- eicosatetrienoate while the 3-series prostaglandins are formed from 5,8,1 1 ,14,17-eicosapentaenoate (EPA). 22 Inhibition of PGHS Nonsteroidal antiinflammatory drugs (NSAIDs) directly inhibit the cyclooxygenase reaction of PGHSs by competing with arachidonic acid for binding at the cyclooxygenase active site. The interaction of most NSAIDs with PGHSs follow a two-step sequence (64): E +I H EIH El* The first step is rapid and reversible and leads to competitive inhibition of the enzyme. The subsequent step is conversion of the initial enzyme-inhibitor complex to one in which the inhibitor is more tightly bound and functionally irreversible. NSAIDs can generally be classified into three groups based on their modes of action. Class I NSAIDs, of which ibuprofen and naproxen are examples, show simple competitive inhibition. Class II NSAIDs, flurbiprofen and indomethacin, show a competitive time-dependent slowly reversible interaction with the enyzme. The mechanism for the ability of the class II NSAIDs to form a semistable El* complex is not yet understood. It has been demonstrated that all of the non-selective, time-dependent inhibitors contain a carboxylic acid moiety and at least one halogen substituent (64). Previously it had been demonstrated that substitution of a methyl ester for the carboxylate group of these inhibitors resulted in a lost of time-dependent inhibition (64). Class III NSAle, aspirin and related acylsalicylates, show a competitive, time dependent, irreversible inhibition (covalent modification). Aspirin binds to PGHS with a low affinity and upon 23 binding transfers its acetyl group from the salicylate to a serine residue (8530) of the enzyme (21 ). Generally, all commonly used NSAIDs have been shown to inhibit both PGHS-1 and PGHS-2, although with some differences in affinity as shown by differences in the 1050 values between the two isozymes (56,55,65). In addition, it has been shown that while aspirin abolishes the cyclooxygenase activity of PGHS-1, it causes PGHS-2 to produce 15R-HET E upon acetylation of the serine residue (8516) (56,66). Because of the many side effects associated with the NSAle, particularly gastric irritation, much effort has gone into producing compounds specific for the second isozyme of PGHS. Recently several PGHS-2 specific inhibitors have been produced. These inhibitors show a time dependent, slowly reversible inhibition with PGHS-2 but a simple competitive inhibition with PGHS-1 (67,68). In addition, in animal models of inflammation, PGHS-2 specific inhibitors are able to reduce the pain and swelling associated with inflammation without the concomitant gastric irritation (69). The mechanism underlying this selectivity is not yet understood. All of these inhibitors contain an arylsulfur group rather than a carboxylate group. In addition mutagenesis studies have shown that by exchanging several PGHS-2 residues for those found in PGHS-1 the time dependence can be abolished. In particular, substitution of lie for Val at residue 523 causes PGHS-2 to exhibit PGHS-1 kinetics with some of the PGHS-2 specific inhibitors ( 70). Another residue which 24 has been demonstrated to play a role in determining selectivity for PGHS-2 specific inhibitors is His513 (71 ). Based on the crystal structure of PGHS-1 it has been postulated that Arg120 plays a critical role in binding of the substrate, arachidonic acid, and in binding carboxylate-containing NSAle such as flurbiprofen. Another hypothesis that was proposed based on the position of Tyr355 was that this latter residue may play a role in the selectivity toward stereoisomers of NSAIDs of the 2-phenyl- propionic acid class. In order to test these hypotheses, mutants of Arg120 and Tyr355 were prepared, and functional studies performed on these mutant PGHSs. In addition, mutations of the homologous Arg120 residue of PGHS-2 (Arg106) were made and experiments performed to compare the role of the Arg120 residue between the two isozymes. The research presented in this thesis focuses on functional and mechanistic studies involving the residues Arg120 of oPGHS-1 and hPGHS-2 and Tyr355 of oPGHS-1 at the mouth of the active sites of the two PGHS isozymes. Chapter 2 presents investigations involving Arg120, of both oPGHS-1 and hPGHS-2, in interactions with substrate and inhibitors and compares the role of this residue in the two isozymes. Chapter 3 looks at the role of Tyf355 in interaction with substrates and in NSAID stereospecificity. 25 CHAPTER 2 STUDIES TO DETERMINE THE ROLE OF ARG120 oPGHS-1 AND ARG120 hPGHS-2 IN SUBSTRATE AND INHIBITOR INTERACTIONS Introduction Examination of the oPGHS-1 crystal structure complexed with flurbiprofen suggests that Arg120 is positioned at the mouth of the cyclooxygenase active site such that this residue interacts with the carboxylate group of fatty acid substrates and inhibitors (43). To test this hypothesis several mutations were made at this position and the kinetic properties of the mutant enzymes examined. Initial characterization of R120K, R1200 and R120E mutant enzymes supported the hypothesis. Mutations of Arg120 increased the Km for arachidonic acid up to greater than 500 fold and decreased cyclooxygenase activity (72). Further experiments were performed to examine the effect of the R1200 mutation on time dependent inhibition by flurbiprofen. It was postulated that the homologous Arg120 in hPGHS-2 ( Arg106) may also interact with fatty acid substrates and carboxylic acid containing inhibitors in a similar way. To determine if this is the case two mutations were made at this position and the kinetic properties of the mutant enzymes were examined using a variety of fatty acid substrates. In addition, flurbiprofen and several PGHS-2 specific inhibitors were tested to determine if Arg120 is involved in determining the selectivity of PGHS-2 for specific inhibitors. 26 Figure 9 shows the active sites of native oPGHS-1 and R1200 oPGHS-1 with arachidonic acid modeled into the cyclooxygenase site. Figure 10 shows mPGHS-2 and R1200 mPGHS-2 with arachidonic acid modeled into the cyclooxygenase site. 27 Figure 9. Model of active site of oPGHS-1 and R1200 oPGHS—1 with arachidonic acid modeled into the cyclooxygenase site. . Coordinates were obtained from the Protein Data Bank and molecular modeling was performed using InsightII.Arachidonic acid is shown in light blue, the heme is red and residue 120 is dark blue A. Native oPGHS-l B. R1200 oPGHS-1 28 lS-l will ales we: Monte: red alt ‘ 0120 M55 Figure 9. 29 Figure 10. Model of active site of mPGHS—2 and R1200 mPGHS-2 with arachidonic acid modeled into the cyclooxygenase active site. . Coordinates were obtained from the Protein Data Bank and molecular modeling was performed using lnsightll.Arachidonic acid is shown in light blue, the heme is red and residue 120 is dark blue A. Native mPGHS-2 B. R1200 mPGHS-2 3O with ales was / Q120 Figure 10 31 Materials and methods Materials- DMEM was from Life Technologies, Inc. Fetal calf serum and calf serum were from Hyclone. Chloroquine, bovine hemoglobin, DEAE-dextran and penicillin G were from Sigma.CsC| from Boehringer Mannheim. 20:4 n-6, 18:2 n- 6 20:5 n-3, 20:2 n-6 and 20:3 n-6 were from Cayman Chemical Co. [1340]- arachidonic acid (50-55 mCi/mmole) was from DuPont NEN. All other reagents were purchased from common commercial sources. Prepaiation of Mutants by Site Directed Mutaqenesis- The R1200 mutant of PGHS-1 was prepared as previously described (72). Briefly, M13mp19-oPGHS- 1, which contains the coding region of oPGHS-1, was mutated using a Bio-Rad Muta-Gene kit following manufacterers instructions. The phage samples were sequenced to identify the mutant. The 2.3 kb insert from the replicative form of M13mp19-oPGHS-1 containing the desired mutation was isolated after digestion with SalI and subcloned into the Sall site of pSVT7. The correct orientation of the insert was determined by restriction digestion with Pst I. Plasmids used for transfection were purified by 0501 density gradient ultracentrifugation. The hPGHS-2 mutants were prepared using a Bio Rad Muta-Gene kit. M13mp19- hPGHS-2, which contains the coding region of PGHS-2, was mutated following manufacterer’s instructions. The single stranded M13 phage samples were sequences using the dideoxy method to identify mutants (74). The 1.8 kb insert, from the replicative form of M13mp19-PGHS-2, containing the desired 32 mutation, was isolated by Sal I digestion and subcloned into the Sal1 site of pOSML (55). The correct orienation was confirmed by restriction digestion with Pst I. The double stranded pOSML-hPGHS-2 was sequenced to confirm that the mutation was still present. Plasmids used in transfection were purified by CsCl gradient ultracentrufigation. The oligonucleotide primers used to prepare the mutants are as listed: oPGHS-1 R1200 5’-GTACTCACAGTGG_AA;TCCAACCT'I’ATCCC-3” hPGHS-2 R1200 5‘-GTGTI'GACATCCQAQTCACAT'ITGATT-3’ hPGHS-2 R120L 5’ -GTGTI'GACATCCQ_IATCACATTI'G-3’ Transfection- Cos -1 cells (ATCC CRL-1650) were grown to near confluency in DMEM containing 8% calf serum and 2% fetal calf serum in a water-saturated 5% 002 atmosphere. Approximately 16 hr prior to transfection the cells were subcultured 1:2. The cos -1 cells were transfected using the DEAE- dextran/chloroquine method as previously described (21). Forty hours post transfection the cells were harvested by scraping in 3 ml of ice-cold phosphate buffered saline (PBS) with a rubber policeman and collected by centrifugation at 1200 rpm for 5 mins. Pregaration of Microsome - Pelleted cells from 20-50 plates were resuspended in 4-5 ml of ice-cold 0.1 M Tris-Cl, pH 7.4 and disrupted by sonication The sonicated cells were centrifuged at 10,000 rpm for 10 mins at 4° C and the resulting supematents centrifuged at 45,000 rpm for 50 mins in a Beckman SW50.1 swinging rotor to give a microsomal membrane fraction. Microsomal 33 membranes were resuspended by homogenation in a volume of 0.1 M Tris-Cl, pH 7.4 sufficient to yield a final protein concentration of about 5 mg/ml. Protein concentrations were determined by a modified Lowrey method, using BSA as a standard (75). Cyclooygenase Assay - Cyclooxygenase assays were carried out at 37° C using a Yellow Springs Instruments Model 5300 oxygen electrode. A typical assay contained 3 ml 0.1 M Tris-Cl, pH 8.0 with 1 mM phenol, 100 pM arachidonate and 25 pg of hemoglobin (as a source of heme). Reactions were initiated by the addition of enzyme to the chamber and initial rates of oxygen consumption determined. For instantaneous inhibition, NSAIDs, at the apprOpriate concentration, were added to the assay chamber prior to the addition of the enzyme. For time- dependent inhibition studies, NSAIDs, at an appropriate concentration were pre- incubated with the enzyme ( 250 pg) at 37° C for various times; reactions were then initiated by the addition of the enzyme/inhibitor complex to the assay chamber, and initial rates of oxygen consumption determined. For studies of substrate specificity 100 uM 20:5 n-3, 18:2 n-6, 20:3 n-6, or 20:2 n-6 were used. Km determinations were performed for 20:4 n-6, 20:5 n-3 and 18:2 n-6 using varying concentrations of these substrate. Peroxidase assays- Peroxidase assays on microsmal protein were performed spectrophotometrically using a Perkins-Elmer model 552A Double Beam UVNIS 34 spectrophotometer, measuring the oxidation of 3,3,3’,3’- tetramethyl- phenylenediamine (TMPD) at 611 nm (78). Western transfer blotting- Microsomal membranes were resolved by one- dimensionsal SOS-PAGE and transfered electrophoretically to 0.45 pm nitrocellulose membranes (Schleicher & Schuell). Membranes were blocked 0.5- 12 hr in 3% (w/v) dry milk, 0.1 % (v/v) Tween-20/Tris buffered saline (TBS) followed by an incubation with a 1:3000 dilution of the affinity purified antibody (Dinah) in 1% dry milk, 0.1% Tween-20/T BS for 2 hours at 25° C. Membranes were washed and incubated with a 1:1000 dilution of a goat anti-rabbit lgG horseradish peroxidase for 1 hour. The membrane was washed and incubated with Amersham ECL reagents and exposed to Kodak XAR film for chemiluminescence. Product A_nalvsis by Thin Layer Chromotaggglly- Microsomal membrane fractions ( 250 pg) were incubated with [1 -”C]-arachidonic acid for 10 mins. Reactions were stopped with the addition of 1.4 ml CCl4zMe (1:1) and centrifuged at 3,000 rpm for 5 min. Supematants were transfered to clean tubes and 0.6 ml CCI4 and 0.32 ml 0.88% formic acid added. After centrifugation the organic layer was removed and dried under N2. The dried samples were redissolved in 50 pl of CCI4 and spotted on Silca Gel 60 thin layer chromatography plates. Plates were developed twice in benzene: dioxane: acetic acid: formic acid ( 82:14:1z1) and exposed to XAR-5 film for 48 hours to allow visualization of the products. 35 Prostaglandin synthesis by hPGHS-2 and the R120 mutant enzymes was quantified by densitometry using a Molecular Dynamics Storm 820 phosphoimager with lmageOuant software. 36 Results Native oPGHS-1 and the R1200 mutant (R1200 oPGHS-1) were expressed transiently in cos -1 cells and microsomal membrane fractions prepared from these cells were assayed for cyclooxygenase and peroxidase activity. Initial characterization had determined that the native and mutant enzymes were expressed at comparable levels in cos-1 cells and that the various prostaglandin products were produced in the same proportions by both enzymes (72). The R1200 mutant enzyme had 3- 5% of the cyclooxygenase activity of the native, with 100 pM arachidonic acid, but retained 60% of the peroxidase activity. The Km values for archidonic acid were determined for the native oPGHS-1 and the R1200 mutant. The native enzyme had a Km of 4 pM while the value for the R1200 mutant was approximately 1000 fold greater (3300 pM). Flurbiprofen causes a time dependent inhibition of oPGHS-l but the methyl ester of flurbiprofen does not (64). The following experiments were performed to determine if flurbiprofen inhibits R1200 oPGHS-1 time-dependently. First the ICso values for “instantaneous” inhibition was determined. The values were 5 pM and 1 mM for native PGHS-1 and R1200 oPGHS-1 respectively. To determine if the R1200 enzyme undergoes time-dependent inhibition with flurbiprofen the enzyme was incubated with 1 mM flurbiprofen for varying lengths of time. No time dependent inhibition of the mutant enzyme was seen even with incubation periods of up to 20 minutes (Figure 11). 37 - w/oflurbiprofen 1oo- w/flurbiprofen PERCENT INITIAL ACTIVITY TIME (min) Figure 11. Time course for inhibition of R1200 oPGHS-1 by flurbiprofen Microsomes prepared from cos-1 cells expressing R1200 oPGHS-1 were incubated for the indicated times at 370 C with or without 10'3 M flurbiprofen and assayed for cyclooxygenase activity using an oxygen electrode. The experiment shown in this figure was performed three times with similar results. 38 To determine if the R1200 hPGHS-2 enzyme shows similar kinetic properties native hPGHS-2 and R120 hPGHS-2 mutants were transiently expressed in cos- 1 cells, and microsomal membrane fractions prepared from the transfected cells. Levels of protein expression were determined by Western transfer blotting and quantified by densitometery using Bio-Rad Molecular Analysis software. hPGHS-2, R120L hPGHS-2 and R1200 hPGHS-2 were expressed at similar levels (Figure12). The cyclooxygenase activity, measured using an oxygen electrode and 100 pM arachidonic acid as substrate, indicated that the R1200 mutant had the same amount of cyclooxygenase activity as the native enzyme but that the R120L mutant had no detectable cyclooxygenase activity. The R1200 mutant had greater than 100% of the peroxidase activity of the native enzyme while the R120L enzyme has 17% of native hPGHS-2 peroxidase activity. The Km value for arachidonic acid for the R1200 mutant was determined to be 10 pM compared to 8 pM for the native enzyme (Table 1). Assays using [1-‘4C]-arachidonic acid and 1000 pg of membrane protein revealed that the R120L mutant did have approximately 20% of the activity of the native hPGHS-2. Both the R1200 and the R120L mutant enzymes produced the same products as the hPGHS-2 enzyme although in slightly different proportions (Figure 13). The most obvious difference was in the production of monohydroxy fatty acids (HETE’s). Both the R1200 and R120L mutants produced twice as much 11- HETE as the native hPGHS-2, 26% vs 13% respectively. The R1200 mutant had a tug for suicide inactivation of 22 sec. 39 Table 1. Kinetic properties of hPGHS—2 and R120 mutant PGH synthases Cos-1 cells were transfected with plasmid encoding native hPGHS-2 or R120 mutant hPGHSs, and microsomal membrane fractions prepared as described in the text.. Cyclooxygenase activity was determined by oxygen electrode and peroxidase activity determined spectroscopically, as described in the materials and methods section. Enzyme Percent COX Activity Percent POX Activity Km AA (pM) hPGHS-2 1 00% 1 00% 8 R1 200 1 00% 1 32% 1 0 R1 20L 0% 1 7% ND 4O 3 v (D I (P to INVHS Z'SHOdII 003 It] 103 I-I:I Figure 12. Western blot analysis of hPGHS-2 and R120 mutant PGH synthases. Cos-1 cells were sham-transfected or transfected with native hPGHS-2, R1200 hPGHS-2, or R120L hPGHS-2. Microsomal fractions were prepared and 50 pg of protein was subjected to Western transfer blotting as described in the material and methods section. 41 15-HETE 1 1 -HETE 12-HHT WVHS Z'SH‘DdlI 090 I8 "190 I8 Figure 13. Product characterization of native and R120 mutant PGH synthases Microsomes containing sham, native or mutant hPGHS-2 protein were incubated with [1-1 C]-arichadonic acid for 10 minutes and the reaction products separated by thin layer chromatography as described in the text 42 compared with 15 sec for native hPGHS-2. Using radio thin layer chromatography it was also determined that the R120L enzyme does undergo inactivation, because there was no change in the total product formation between 5 and 10 min for either the native enzyme or the R120L mutant; however a tug was not determined for the R120L hPGHS-2. To examine the role of Arg120 in the binding and oxygenation of fatty acid substrates other than arachidonate 100 pM 20:5 n-3, 20:3 n-6, 20:2 n-6 or 18:2 n-6 were tested as substrates for the cyclooxygenase reaction with both the native hPGHS-2 and R1200 hPGHS-2. The rate for each substrate was compared with that observed with 100 pM arachidonic acid. The native enzyme using 20:5 n-3 as the substrate had 38% of the COX activity found with arachidonic acid. In contrast, the R1200 mutant had only 3% of the activity observed with arachidonic acid. With linoleic acid the native enzyme had 50-60% of the activity as did the R1200 enzyme. When 20:3 n-6 was used as the substrate the native enzyme exhibited 100% of the activity. The R1200 enzyme was also able to use 20:3 n-6 as a substrate, although not quite as well, having 71% of the activity. 20:2 n-6 was not used efficiently by either enzyme; the native enzyme had only 9% activity while the mutant enzyme had no detectable activity. The Km values for 20:5 n-3 and 18:2 n-6 were determined for native hPGHS-2 and R1200 hPGHS-2. For 20:5 n-3 the values were approximately 2 pM and 2000 pM respectively. With 18:2 n-6 the Km was approximately 22 pM for the R1200 mutant, which is within the range previously reported for native hPGHS-2 (61). The data are summarized in Table 2. It has been previously demonstrated 43 that the addition of 15-HPETE to the reaction when 20:5 n-3 is used as the substrate leads to an increase in Vmax, for PGHS-1 but not for PGHS-2. To determine if this is also true of the R1200 mutant 10 pM 15-HPETE was added to the cyclooxygenase assay mixture; no significant change in Vmax was observed. Flurbiprofen causes time-dependent inhibition of native hPGHS-2 similiar to that of oPGHS-1. To determine if the Arg120 in hPGHS-2 plays a similiar role in inhibition as that of Arg120 in oPGHS-1 “instantaneous” and “timedependenf' inhibition experiments were performed. Native hPGHS-2 had an 1050 value for flurbiprofen of 1 pM and unden/vent time-dependent inhibition with a tug of less than 1 minute. This is in contrast to the R1200 mutant which had a I050 value of 0.5 mM and did not undergo time-dependent inhibition during a 20 min incubation with 0.5 mM flurbiprofen (Figure 14). Several PGHS-2 specific inhibitors have been shown to cause time- dependent inhibition with PGHS-2 but simple competitive inhibition of PGHS-1. We examined the role of Arg120 in the interactions with several of these inhibitors. Figure 15 shows the structures of the compounds used. Native hPGHS-2 had an leo value with NS-398 of 50 pM and exhibited time-dependent inhibition. The R1200 mutant had a 5 fold lower IC50 value (10pM) but did not undergo time-dependent inhibition (Figure 16). On the other hand, the mutant enzyme showed an increased IC50 value and a decreased t1/2 for timedependent 44 d ‘ J t + native hPGHS-Z E + R1200 hPGHS-Z I- 0 < 100 T .1 S I: 1' .z. a I- .. E 50 o 'r s . .. a it o 1 I l l o 5 10 15 20 TIME (min) Figure 14. Time-dependent Inhibition of hPGHS-2 and R1200 hPGHS-2 by flurbiprofen Microsomes prepared from cos-1 cells transfected with hPGHS-2 or R1200 hPGHS-2 were incubated with 500 pM flurbiprofen for the indicated times and assayed for initial cyclooxygenase activity as described. Measurements at each time point were done in duplicate or triplicate and the experiment was repeated several times with similar results. 45 on $2 #8 mm od #2. $8 8 o-.. 3: oz oz So oz oz oz #9 oz m... Now oz oz as: oz oz oz #5 oz o-.. mom So seem: 0% 88 to: some some 3 m-.. mom 3 .8: $8: 2 2 $8. <08: o 9: sow ex}; 32> £38: 99 Ease. l eves; I 52> 318: 29 33 ex 08E moron: mzfizm mEEmoom 46 .co:m:cooc8 29:38 some .2 28:9: .0 6625:: c_ 885:8 who; 958938.). 6.9 9: E 25:88 on 86:33 20m 35 :o moo:m:coocoo mote? ago: 80:86 comes 3 85.52% 9o; 33:8 68595226 .9 829, ex as 35> uozoo; 08E 23 «-mzooz 6. 3.058% 22.85 .o 8:338 .u use. Dup-697 O H H30 —s u / \ O 8 Br F H SH'—CF3 I O F N I \N / CF3 SC-58125 NO 0 2 O || I O NS-398 Figure 15. Structures of PGHS-2 specific inhibitors tested with hPGHS—2 and R1200 hPGHS-Z 47 inhibition with Dup-697 (Figure 17). 80-58125 has an aryl sulphur group similar to that of Dup-697, and it was anticipated that SC-58125 would exhibit a similiar inhibition profile with the R1060 enzyme. This appeared to be the case. Native hPGHS-2 was not inhibited “instantaneously” by SC-58125 while the R1200 enzyme has a leo value of 2.5 pM. Both enzymes undenlvent time-dependent inhibition ( Figure 18 ). The data from the inhibitor studies is summarized in Table 3. 48 150 '- t + native hPGHS-Z S + R1200 hPGHS-Z F: 0 < 100 T .l < .l. t T .2. ~11.- l- 3 E 50 g T |.I.| l T 0- :2 o I I I I 0 5 10 15 20 TIME (min) Figure 16. Time-dependent inhibition of hPGHS-2 and R1200 hPGHS-2 by NS-398 Microsomes prepared from cos-1 cells transfected with hPGHS-2 or R1200 hPGHS-2 were incubated with 1 pM N8398 for the indicated times and assayed for initial cyclooxygenase activity as described. The experiment was repeated several times with similar results. 49 150- E —+— nativehPGHS-Z B ,_ —I— RIZOQhPGI-IS-z j 100 S T .— .L E I- 50- 2 I.I.I O n: II‘. o J . 0 5 10 15 TIME (min) Figure 17. Time-dependent inhibition of hPGHS-2 and R1200 hPGHS-2 by Dup-697 Microsomes prepared from cos-1 cells transfected with hPGHS-2 or R1200 hPGHS-2 were incubated with 1pM Dup-697 for the indicated tiimes and assayed for initial cyclooxygenase activity. Measurements for each time point were performed in duplicate and the experiment was performed several times with similar results. 50 150-1 E -O— native hPGHS-Z 5 .. + R1200 hPGHs-z < 100 .I S. t: E I- 50: 2 Ill 3 T m - ‘- o. o .- 4 0 5 10 15 20 25 TIME (min) Figure 18. Time-dependent Inhibition of hPGHS-z and R1200 hPGHS—2 by 80-58125 Microsomes prepared from cos-1 cells transfected with hPGHS-2 or R1200 hPGHS-2 were incubated with 10pM 8058125 for the indicated times and assayed for cyclooxygenase activity, The experiment was repeated at least two times with similar results. 51 Table 3. Instantanous and time-dependent Inhibition of native hPGHS-2 and R1200 hPGHS-Z. Summary of results of inhibition studies described in detail in the text. INHIBITOR ENZYME hPGHS-Z R1200 ICso I112 Icso ttlz flurbiprofen 1 pM <1mln @ 0.5 mM 0.5 mM > 20 min @0 .5 li NS-398 50 pM 2 min 6 1pM 10 pM > 20 min @1 pM Dup-697 0.5 mM 4 mln @ 1pM 0.5 pM 1 min 6 1 pM SC-58125 >1 mM < 2min O 10pM 2.5 pM < 2 mln O 10 pM 52 Discussion In these experiments the role of the Arg120 residue of oPGHS-1 and the Arg120 residue of hPGHS-2, in the binding of fatty acid substrate and nonsteroidal anti-inflammatory drugs (NSAIDs) in the cyclooxygenase active site, was examined. The results support the hypothesis that Arg120 is an important residue in the binding of the carboxylate group of arachidonic acid and inhibitors. Replacing this arginine with a neutral glutamine residue allowed the enzyme to retain a small amount of the cyclooxygenase active and about 60% of the peroxidase activity, but the Km value for arachidonic acid was increased approximately 1000 fold. This is in contrast to the results obtained with the R1200 hPGHS-2, which had no significant changes in kinetic properties with arachidonic acid as the substrate. However, replacing this Arg residue with a smaller hydrophobic residue, Ieucine, resulted in a significant loss of cyclooxygenase activity, suggesting that, in hPGHS-2, while a hydrogen bonding interaction can replace an ionic interaction at this position some interaction is necessary in order for cyclooxygenase activity with arachidonic acid. Previously, it has been demonstrated that there are some differences in substrate specificity, between the two PGHS isozymes, which have been attributed to a larger cyclooxygenase active site in PGHS-2 (61). These results suggest that some of these differences in specificity may result from a need for different stabilizing interactions at the mouth of the cyclooxygenase active site. 53 With the majority of the other substrates tested (18:2 n-6, 20:3 n-6, 20:2 n-6) it appears that the hydrogen bond can substitute for the ionic interaction. However, when 20:5 n-3 is used as a substrate it appears that the ionic interaction is extremely important as is shown by the dramatic decrease in cyclooxygenase activity and increase in Km by the R1200 mutant with this substrate. One possible explanation for this could be that with the loss of the ionic interaction the fatty acid is not positioned correctly in the active site. Because the 20:5 n-3 is not in an optimal position to begin with, the loss of this stabilizing interaction could lead to the fatty acid not remaining in the active site in a conformation in which the n-8 hydrogen can be abstracted, as efficiently as by the native enzyme. For arachidonic acid the loss of the ionic interaction does not appear to affect the kinetic factors to any appreciable degree. This probably is due to the carboxylate group of the fatty acid being positioned in such a way that it is still able to form a hydrogen bond interaction with the glutamine. In addition, there may be other interactions with the protein that assist in stabilizing the fatty acid such as one with Ar9513. The corresponding residue is a histidine in oPGHS-1. When 20:5 n-3 is the substrate, it is likely that the carboxylate group is not in proximity of the glutamine to form a hydrogen bonding network or that the hydrogen bonding is not sufficient to hold the fatty acid in the active site. Additionally, the fatty acid may be poorly positioned to interact with other potentially stabilizing residues. With the loss of the stabilizing interaction at the mouth of the active site, the fatty acid could then migrate up into the active site 54 or down through the floor of the active site, so that it is no longer in position for interaction with Tyr385. There is a difference in the ratio of the products formed with the R1200; 11- HETE is produced in greater proportions of the products than with the native enzyme. This provides evidence for the position of the arachidonic acid being different in the mutant enzyme. The importance of this type of interaction has also been shown with the lipoxygenase enzyme, where substitution of the the arginine that interacts with the carboxylate group of arachidonic acid by a Ieucine results in a change in proportion of products (77). NSAle can be grouped into two general categories based on their ability to cause time-dependent vs simple competitive inhibition. Flurbiprofen causes a time-dependent inhibition, however the methyl ester of flurbiprofen is a simple competitive inhibitor indicating that an ionic interaction with oPGHS-I is important for time-dependent inhibition. The results with the R1200 mutant are consistant with this concept. R1200 oPGHS-1 did not undergo time-dependent inhibition with flurbiprofen even at high concentrations of the inhibitor. In order for an inhibitor to cause time dependent inhibition the enzyme/inhibitor complex must undergo a secondary conformational rearrangement. It appears that one interaction that is critical for time-dependent inhibition is an ionic interaction involving the carboxylate group of the inhibitor and the guanido group of Arg120. As with PGHS-1, the arginine group is necessary for the time-dependent inhibition of hPGHS-2 by carboxylic acid inhibitors as demonstrated by the lack of time-dependent inhibition of the R1200 hPGHS-2 mutant with flurbiprofen. 55 The substitution of glutamine for Arg120 resulted in some changes in inhibitor profiles with PGHS-2 specific inhibitors. Specifically, the R1200 mutant did not undergo time-dependent inhibition by NS-398; however this mutant did undergo time-dependent inhibition with both Dup-697 and SC-58125 although with a reduced t1/2. 56 CHAPTER 3 STUDIES TO DETERMINE THE ROLE OF TYR355 OF oPGHS-1 IN SUBSTRATE AND INHIBITOR INTERACTIONS Introduction Examination of the crystal structure of oPGHS-l complexed with flurbiprofen suggests that Tyr355 is involved in the hydrogen bonding network at the mouth of the cyclooxygenase active site that positions flurbiprofen in this site. In addition, the bulky hydroxyl group of this tyrosine residue places a constriction on one side of the cyclooxygenase channel that may determine the stereo- selectivity of the enzyme for the S -isomer of 2-phenylpropionic acid inhibitors (43). This hypothesis was tested with the Y355F oPGHS-1 mutant and the mutant enzyme did show a decreased stereoselectivity towards ibuprofen (72). To further determine the interaction of Tyr355 with this class of inhibitors the Y355A and Y355L oPGHS-1 mutants were prepared and the kinetic properties examined. In addition, further characterization of the role of Tyr355 on substrate and inhibitor interactions were performed, to examine the effect of Tyr355 on the binding and oxygenation of various fatty acid substrates and its role in time- dependent inhibition. Figure 19 shows the active site of native oPGHS-1 and Y355F oPGHS-1 with arachidonic acid modeled into the cyclooxygenase site. 57 Figure 19. Model of active site of oPGHS-1 and Y355F oPGHS-1 with arachidonic acid modeled Into the cyclooxygenase site. Coordinates were obtained from the Protein Data Bank and molecular modeling was performed using lnsightll. The arachidonic acid is shown in light blue, the heme in red and residue 355 in green A. native oPGHS-1 B. Y355F oPGHS-1 58 F355 59 Figure 19 Materials and Methods Materials- DMEM was from Life Technologies, Inc. Fetal calf serum and calf serum were from Hyclone. Chloroquine, bovine hemoglobin, DEAE-dextran and penicillin G were from Sigma.20:4 n-6, 18:2 n-6, 20:5 n-3, 20:2 n-6, 20:3 n-6 were from Cayman Chemical Co. [1-“Cj-arachidonic acid (50-55 mCi mmole ) was from NEN. CsCI was from Boehringer Mannheim. All other reagents were purchased from common commercial sources Preparation of oPGHS—1 Mutants by Site-directed Mutagenesis- The Y355F mutant was prepared as described previously (77). The Y355A and Y355L mutants were prepared starting with M13mp19-oPGHS-1, which contains the coding region of ovine PGHS-1, using a Bio-Rad kit as described in the manufacterers directions. The oligonucleotide primers used to prepare each mutant are as follows: Y355L: 5’- CTGAGCGGCQTCTTCCTGCAGCTC- 3’ Y355A: 5’- CTGAGCGGCGCCTTCCTGCACTC - 3’ Phage samples were sequenced using the dideoxy method (74) to identify mutants. The 2.3 kb insert from the replicative form of M13mp19-oPGHS-1 containing the desired mutation was isolated after digestion with Sal l and subcloned into the Sall site of pSVT7. The correct orientation of the insert was confirmed by restriction digestion with Pst l. Plasmids used for transfection were purified by CsCI density gradient ultracentrifugation. 60 Transfection- Cos-1 cells (ATCC CRL-1650) were grown in DMEM containing 8% calf serum and 2% fetal calf serum until near confluence and transfected with plasmid containing DNA for native oPGHS-1, Y355F oPGHS-l, Y355L oPGHS-1 and Y355A oPGHS-1 as described in Materials and Methods in Chapter 2. Forty hours post transfection the cells were hawested by scraping in 3 ml of ice-cold phosphate buffered saline (PBS) with a rubber policeman and collected by centrifugation at 1200 rpm for 5 min. Preparation of Microsomes- Pelleted cells from 20-50 plates were resuspended in 4-5 ml of ice-cold 0.1 M Tris-Cl, pH 7.4, and microsmes prepared as described in Material and Methods in Chapter 2. Protein concentrations were determined by a modified Lowrey method, using BSA as a standarad (75). Western transfer blotting- Microsomal membrane fracions were resolved by 10% SDS-PAGE and transferred electrophorectically to 0.45 pm nitrocellulose membranes (Schleicher & Schuell). The membranes were blocked for 30 min in 3% (w/v) dry milk, 0.1 % (v/v) Tween-20/ Tris buffered saline (TBS) and incubated for 2 hrs with a 1:10,000 dilution of an affinity purified antibody (COX 2). The membranes were washed and incubated with a 1:1000 dilution of a goat anti-rabbit IgG-horseradish peroxidase conjugate for 1 hour. Membranes were washed and incubated for 1 min with Amersham ECL Western Blotting detection 61 reagents, blot dried and exposed to XAR-5 film. Expression levels were quantified by Bio Rad Molecular Analysis software. Cyclooxygenase Assays- Cyclooxygenase assays were carried out at 37° C using a Yellow Springs Instruments Model 5300 oxygen electrode as described in Materials and Methods in Chapter 2. For determination of the stereoselectivity of the mutant enzymes either R or S— ibuprofen, at the appropriate concentration, was added to the assay chamber prior to the addition of the enzyme. For time- dependent inhibition studies, flurbiprofen at an appropriate concentration was pre-incubated with the enzyme (250 pg) at 37° C for various times; reactions were then initiated by the addition of the enzyme/inhibitor complex to the assay chamber, and initial rates of oxygen consumption determined. For substrate studies 100 pM 20:5 n-3, 18:2 n-6, 20:3 n-6, or 20:2 n-6 were used. Km determinations were done for 20:4 n-6 using varying concentrations of the substrate. Peroxidase Assavs- The peroxidase activity of microsomal membrane preparations was measure spectrophotometrically at 436 nm using guaicol and H202 as the cosubstrates. The assay mixture contained 0.1 M Tris-Cl, pH7.2, 5.6 mM guaiacol, 1pM hematin and 50—100 pg microsomal protein in a total volume of 0.3 ml. Reactions were initiated by the addition of hydrogen peroxide and the product ( 3,3’-dimethoxydopheno-4,4’-quinone) formation detected at 436 nm. 62 Product Analvsis bv Thin Layer Chromgtagraghy- Microsomal membrane fractions (250 pg) were incubated with [1-14C]-arachidonic acid for 10 mins. Reactions were stopped with the addition of 1.4 ml CCI4zMeOH (1:1) and centrifuged at 3,000 rpm for 5 min. Supernatants were transfered to clean tubes and 0.6 ml 0014 and 0.32 ml 0.88% formic acid added. After centrifugation the organic layer was removed and dried under N2. The dried samples were resuspended in 50 pl of CCI4 and spotted on Silca Gel 60 thin layer chromatography plates. Plates were developed twice in benzene: dioxane: acetic acid: formic acid ( 82:14:1z1) and exposed to XAR-5 film for 48 hours to allow visualization of the products. 63 Resuus Native oPGHS-1 and Y355 oPGHS-1 mutants were transiently expressed in cos-1 cells and microsomal membrane fractions prepared. Levels of expression were determined by Western transfer blotting and densitometry (Figure 20). The level of expression for the Y355F mutant was previously determined to be similar to that of the native oPGHS-1 (50). The mutant enzymes were expressed at 92- 100 % of the native oPGHS-1. Cyclooxygenase and peroxidase activities were measured, and the Km value for arachidonic acid for each mutant was determined. The Y355F enzyme had 20% of the cyclooxygenase activity of the native oPGHS-1 while the Y355A and Y355L enzymes each had approximately 10 % of native activity. All the mutant enzymes had substantial peroxidase activity compared with the native oPGHS-1. The kinetic data is summarized in Table 4. The Y355L enzyme had a similar Km value for arachidonic acid as the native enzyme, while both the Y355F and Y355A mutants had values about five fold higher. Product characterization by radio thin layer chromatography revealed that all the enzymes formed the same products in similar proportions (Figure 21). To examine the role of Tyr 355 in binding of other fatty acid substrates 100 pM 20:5 n-3, 20:3 n-6, 18:2 n-6 and 20:2 n-6 were used as substrates for the cyclooxygenase reaction with the native oPGHS-1 and Y355F oPGHS-1. The Y355F mutant exhibited 20% of the cyclooxygenase activity of the native enzyme with 20:4 n-6. Relative to the native oPGHS-1 the Y355F mutant had 32 % 64 oPGHS-1 standard oPGHS-1 native Sham Y355A Y355L Figure 20. Western blot of oPGHS-1 and Y355A and Y355L mutants PGH synthases Cos-1 cells were sham-transfected or transfected with native oPGHS-1, Y355A oPGHS-1 or Y355L oPGHS-1. Microsomal fractions were prepared and 5 pg of protein was subjected to Western transfer blotting as described in the text. 65 Table 4. Comparison of Kinetic properties of oPGHS-1 and Y355 mutant oPGHSs Cos-1 cells were transfected with plasmid encoding native oPGHS—1 or Y355 mutant oPGHS-1s, and microsomal membrane fractions prepared as described in the text.. Cyclooxygenase activity was determined by oxygen electrode and peroxidase activity determined spectroscopically, as described in the text. ENZYME % COX ACTIVITY % POX ACTIVITY Km AA (mM) oPGHS-1 100% 100 4.2 Y355F 19 :I: 4% 130 18 Y355L 10 :I: 3% 82 :I: 24% 8 Y355A 10 :t 4% 84 :I: 26% 22 66 1 5-H ETE 1 1-I'I ETE PG D2 PGE2 PGan AA oPGHS—1 Y355A Y355F Y355L Figure 21. Product characterization of oPGHS—1 and Y355 mutant PGH synthases. Microsomes containing native or Y355 mutant oPGHS-1 protein were incubated with [1 -“C]-arachidonic acid and the reaction products separated by thin layer chromatography as described in the text. 67 cyclooxygenase activity with 20:3 n-6, 32 % with 20:5 n-3 ( in the presence of 10 pM 15-HPET E) and no detectable COX activity with either 18:2 n-6 or 20:2 n-6. The data are summarized in Table 5. To examine the role of Y355 in the stereospecificity of oPGHS-1 for the S - isomer of the 2-phenylpropionic inhibitors, IC50 values for the two stereoisomers of ibuprofen were determined and the R/S ratios calculated for each of the Y355 mutant enzymes and native oPGHS-1 (Figure 22). The Y355F mutant showed decreased discrimination for the Fi isomer of ibuprofen. The Y355L and the Y355A mutants were inhibited to a lesser extent by both ibuprofen isomers and there was no appreciable discrimination for H vs S -ibuprofen. The IC50 for flurbiprofen for the Y355F mutant was determined to be 50 pM as compared to 1pM for the native oPGHS-1. For the time-dependent inhibition experiments this concentration (50 pM) of flurbiprofen was used. The native oPGHS-1 had a tug of less than 1 min at this concentration, in contrast the Y355F mutant did not lose any activity until after a 10 min incubation, but lost all cyclooxygenase activity during a 20 min incubation. 68 Table 5. Substrate Specificities of native oPGHS-1 and Y355 mutant oPGHS-1s. Vm,x values for cyclooxygenase activity were determined using 100pM fatty acid substrates, as described in the text. Measurements for each substrate were repeated in duplicate. The experiment was performed at least twice with similar results. SUBSTRATE % ACTIVITY ENZYME (100 11M) (Oz/moi of fatty acid) = —_—II==I oPGHS-1 20:4 n-6 100 s: 2 % 18:2 n-6 30 a: 5 % 20:3 n-6 70 z 3 % 20:2 n-6 5 :I: 2 % 20:5 n-3 + 10pM HPETE 15% Y355F 20:4 n-6 100:7 % 18:2 n-6 0% 20:3 n-6 1131:1096 20:2 n-6 0% 20:5 n-3 + 10pM HPETE 24% 69 401 2 Q 30: A 1.. < z . E E 204 E’.’ o (D a: ,0- A A .A o T oPGHS-1 Y355F Y355L Y355A ENZYME Figure 22. R vs S Ibuprofen discrimination by native oPGHS-1 and Y355 mutant oPGHS-1s ICso values for the two stereoisomers of ibuprofen were determined by oxygen electrode and the R/S ratios calculated for each of the Y355 mutant enzymes and native oPGHS-1. 7O Discussion Changing Tyr355 to Phe, Leu, or Ala caused the oPGHS-1 to have a decreased Vmax and an increased Km for arachidonic acid. The Km values for the Y355F and Y355A mutant enzymes were about five fold higher than that of native oPGHS-1 and the cyclooxygenase activities were about five and ten fold lower, respectively. The Km value for the Y355L mutant was about the same as that of native oPGHS-1 but the cyclooxygenase activity was ten fold lower. These results, interpreted in the context of the oPGHS-1 crystal structure and previous studies of the neighboring Arg120 group, suggest that Tyr355 is involved in an interaction with the carboxylate group of arachidonic acid and elimination of the phenolic hydroxyl group of Tyr355 leads to a decrease in the effeciency of oxygenation of arachidonic acid presumably due to the loss of proper substrate positioning for catalysis. The Y355F mutant was able to utilize both 100 pM 20:3 n-6 and 20:5 n-3 as substrates albeit less efficiently than native oPGHS-1, but it was unable to oxygenate either 18:2 n-6 or 20:2 n-6. Thus, the hydroxyl group of Tyr355 most likely participates in a hydrogen bonding network involving all the carboxylate groups of all fatty acid substrates. In the case of 20:5 n-3 the relative increase in activity as compared with 20:4 n-6 can be explained as loss of a hinderince. By examining stereoviews of the active site it appears that with the removal of the phenolic hydroxyl group the mouth of the active site is less constricted allowing the less flexible fatty acid better access and thus increasing the potential for catalysis. Y355F oPGHS-1 does not oxygenate 18:2 n-6 and 20:2 n-6. Linoleic acid is oxygenated relatively efficiently 71 by native oPGHS-I. This implies that the phenolic hydroxyl group of Tyr355 is very important for binding and oxygenation for 18:2 n-6. In the case of 20:2 n-6, it appears that the absence of double bonds at C-5 and C-8 cause this substrate to be used relatively poorly by native oPGHS-1; with the loss of the Tyr355 interaction it is not used at all. Further studies will need to be done to examine why these fatty acids are not used as substrates with the Y355F enzyme; if it is decreased affinity for the fatty acid or if it is fatty acid positioning. Experiments will need to be performed to determine if 18:2 n-6 and 20:2 n-6 can inhibit the oxygenation of 20:4 n-6. Experiments to examine the stereoselectivity of the ibuprofen isomers demonstrate that removal of the hydroxyl group allows the R isomer to be a relatively more efficient inhibitor. Examination of stereoviews of the Ala and Leu mutant enzymes show that the substitution of the smaller groups leave a large space at the mouth of the cyclooxygenase active site. While it is possible that some rearrangements takes place among neighboring residues to compensate, our results suggest that it is not sufficient to provide stabilization of the NSAID as evidenced by the decrease in inhibition by the S -ibuprofen. Recently it was shown that the analogous residue (Y341) in hPGHS-2 is involved in time dependent inhibition by flurbiprofen. (76). So at al found that the Y355F hPGHS-2 reached maximal inhibition more rapidly than the native PGHS-2. In order to determine if this was also the case with oPGHS-1, the Y355F enzyme was incubated with flurbiprofen for varying amounts of time. The results of two experiments showed the opposite results. With the Y355F oPGHS- 72 1 enzyme the leo was 50 pM. a 50 fold increase over that for the native oPGHS-1, but the enzyme had a greatly increased t1/2 for time-dependent inhibition. It appears that for oPGHS-1 the interaction of Tyr355 plays a more important role in time-dependent inhibition with carboxylate containing inhibitors than does the corresponding Tyr355 in hPGHS-2. The two notable differences between the active sites of the isomers are He 523 to Val and His 513 to Arg.. It would be interesting to look at double mutants of Y355F with either a l523V or a H513R mutant to see if time dependent inhibition by flurbiprofen can be restored. 73 CONCLUSION From examination of the crystal structure of oPGHS-1 complexed with flurbiprofen it appears that Arg120 and Tyr355 are in a position to interact with the carboxylate group of fatty acid substrates and non-steroidal antiinflammatory drugs. To determine the role of Arg120 we used site-directed mutagenesis to replace Arg120 with glutamine. The R1200 oPGHS-1 had a decreased catalytic rate and a greatly increased Km for arachidonic acid. In addition, it was no longer inhibited by flurbiprofen. Thus we conclude that Argl20 is necessary for arachidonic acid binding and oxygenation, and for inhibition by carboxylate containing inhibitors such as flurbiprofen. Two different isoforrns of PGHS, PGHS-1 and PGHS-2, have been identified, and while they have similar primary sequences and overall structures, they demonstrate subtle differences in substrate specificity and inhibition by NSAIDS. The homologous Arg120 residue in PGHS-2 was examined to determine if it interacts with fatty acid substrates and NSAle in a similar manner. We used site-directed mutagenesis to replace hPGHS-2 R120 with glutamine and Ieucine. We determined that ArgI20 plays a similar role in binding carboxylate containing inhibitors because a R1200 hPGHS-2 was no longer inhibited by flurbiprofen. However hPGHS-2 R1200 interacts with fatty acid substrates in a different manner than oPGHS-1 Arg120 because the PGHS-2 R1200 mutant retained all of the cyclooxygenase activity and had no change in the Km for arachidonic acid. Changing the arginine to a Ieucine did result in a substantial decrease in cyclooxygenase activity. Further experiments to examine the nature of this 74 interaction will need to be performed. Examination of the role of hPGHS-2 Arg120 in fatty acid binding and oxygenation of fatty acid substrates revealed that the guinido group of this Arg120 is important for binding and oxygenation with 20:5 n-3, because the R1200 hPGHS-2 showed a significant decrease in cyclooxygenase activity and a greatly elevated Km for 20:5 n-3. Preliminary studies performed to examine if 20:5 n-3 is able to act as an inhibitor for 20:4 n-6 indicate that 20:5 n-3 is able to enter the cyclooxygenase site, as it can inhibit catalysis of 20:4 n-6, but is not able to undergo catalysis as demonstrated by the increase in K.... Experiments to examine the role of Arg120 in the specificity for several PGHS-2 selective inhibitors were performed. The R1200 hPGHS-2 lost time dependent inhibition with NS-398, but not with Dup-697 or SC-58125 indicating that the guinido group of arginine is necessary for interaction with the sulphanimide group of NS-398 for time-dependent inhibition. Site-directed mutagenesis was used to replace Tyr355 with phenylalanine,leucine and alanine to determine the role of Tyr355 in binding and oxygenation of fatty acid substrates and in interactions with inhibitors. PGHSs exhibit selectivity for the S-isomer of NSAIDs. From the position of Tyr355 it was postulated that the hydroxyl group of Tyr355 is involved in this selectivity. We found that the Y355F oPGHS-1 showed a decreased selectivity against the Fi- isomer of ibuprofen , but the Y355L oPGHS-1 and Y355A oPGHS-1 lost inhibition with both isomers of ibuprofen indicating that some interaction is necessary at this position for inhibition. Additionally, the Y355F oPGHS-I showed a increased 75 t1/2 for time-dependent inhibition with flurbiprofen indicating that this interaction is important for time-dependent inhibition, but not necessary. Tyr355 was shown to be involved in fatty acid substrate binding and oxygenation because the Y355F oPGHS-l demonstrated no detectable cyclooxygenase activity with 18: n-6 and 20:2 n-6. This implies that the phenolic hydroxyl group of Tyr355 is important for positioning the 18:2 n-6 in the cyclooxygenase site. Further studies will need to be done to examine why these fatty acids are not used as substrates with the Y355F enzyme. 76 10. 11. 12. 13. 14. 15. References Fitzpatrick, FA. and Smith W.L. (1996) The eicosanoids: cyclooxygenase, lipoxygenase, and epoxygenase pathways. In Biochemistry of Lipids,Lipoproteins and Membranes, Vance and Vance, Eds. pp. 283-311 Elsevier Science B. V. Smith, W.L. and Mamett, L.J. (1991) Biochem. Biophys. Acta. 1083: 1-17. Smith, W.L. (1989) Biochem. J 259: 315-324. Smith, W.L. (1992) Am. J. Physiol. 263: F181-F191. Oates, J.A., Fitzgerald, G.A., Branch, R.A., Jackson, E.K., Knapp, HR. and Roberts,L.J. (1988) N. Engl. J. Med. 319: 689-698. Oates, J.A., Fitzgerald, G.A., Branch, R.A., Jackson, E.K., Knapp, HR. and Roberts,L.J. (1988) N. Engl. J. Med. 319: 761-767. Eberhart, C.E., Coffey, R.J., Radhika, A., Giardiello, F.M. Ferrenbach,S. and DuBois, RN. (1994) Gastroenferology 107: 1183-1186. Kargman, S., O’Neill, G., Vickers, P., Evans, J., Mancini, J. and Jothy, S. (1995) Cancer Res. 55: 2556-2559. Stephenson, J. (1996) JAMA 275: 1389-1390. Channon, J.Y. and Leslie, CC. (1989) J. Biol. Chem. 265: 5409- 541 3. Roshak, A., Sathel, G. and Marshall, L. A. (1994) J. Biol. Chem. 269: 25999-26005. Reddy, S.T., Winstead, M.V., Tischfield, J.A. and Herschman, HR. (1997) J. Biol. Chem. 272: 13591 -1 3596. Kuwata, H., Nakatani, Y., Murakami, M. and Kudo, I. (1997) J. Bio/Chem. 273:1 733-1 740. Clark, J.D., Lin, L.L., Kriz, R.W., Ramesha, C.S., Sultzman, L.A., Lin, A.Y., Milona, N. and Knoft,J.L. (1991) Cal/65: 1043-1051. Gronich, J.H., Bonventre, JV. and Nemenoff, RA. (1990) Biochem. J. 271: 37-43. 77 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. Kudo,l., Murakami, M., Hara, S. and lnoue, K. (1993) Biochem. Biophys. Acta. 1170: 217-231. Smith, W.L., Mamett, L.J. and DeWitt, D.L. (1991) J. Pharmac. Ther. 1 991: 153-179 Hemler, M., Lands, W. E. M. and Smith, W. L. (1976) J. Biol. Chem. 251: 5575-5579 Van der Ouderan, F. J., Buytenhek, M., Nugteren, DH. and Van Drop, D. A. (1977) Biochem. Biophys. Acta. 487: 315-331 Dewitt, D. L. and Smith, W. L. (1988) Proc. Natl. Sci. USA 85: 1412-1416 Dewitt, D. L., El-Harith. E. A., Kraemer, S. A., Andrews, M. J., Yao, E. F., Armstrong, R. L. and Smith, W. L. (1990) J. Biol. Chem. 265: 5192-5198 Funk, C. 0., Funk, L. B. Kennedy, M. E., Pong, A. S. and Fitzgerald, G. A. (1991 ) FASEB J. 5: 2304-2312 Feng, L., Sun, W., Xia, Y., Tang, W. W., Chanmugam, P., Soyoola, E., Wilson, C. B. and Hwang, D. (1993) Arch. Biochem. Biophys. 307: 361- 368 Xie, W., Chipman, J. G., Robertson, D. L., Erickson, R. L. and Simmons, D. L. (1991) Proc. Natl. Acad. Sci. USA 88: 2692-2696 Kujubu, D. A., Fletcher, B. S., Vamum, B. C., Lim, R. W. and Herschman, H. R. (1991) J. Biol. Chem. 266: 12866-12872 Hla, T. and Neilson, K. (1992) Proc. Natl. Acad, Sci. USA 87: 7384-7388 Zhang, V., O’Sullivan, M., Hussain, H., Roswit, W. T. and Holtzman (1996) Biochem Biophys Res Commun. 227: 499-506. Jones, D.A., Carlton. D.P., Mclntryre, T.M., Zimmerman, GA. and Prescott, SM. (1993) J. Biol. Chem. 268: 9049-9054. Herschman, HR. (1994) Cancer and Metastasism Reviews 13: 241 -256. Breder, C.D., DeWitt, D, and Kraig, RP. (1995) J. Comp. Neurol. 355: 296-315. Harris, R.C., McKanna, J.A., Akai, Y., Jacobson, H.R., Dubois, RN. and Breyer, MD. (1994) J. Clin. Invest. 94:2504-250. 78 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 45. 46. 47. Morris, J.K. and Richards, J. (1993) Endocrinology 133: 770-779. Ristimaki, A., Garfinkel, S., Wessendorf, J., Maciag, T. and Hla, T. (1994) J. Biol. Chem. 269: 11769-11775. Evett, G.E., Xie, W., Chipman, J.G., Robertson, D.L. and Simmons, D.L. (1993) Arch. Biochem. Biophys. 306: 169-177. Langenbach, R., Morham, S.G., Tiano, H.F., Loftin, C.D., Ghanayem, B.|., Chulada, P.C., Mahler, J.F., Lee, C.A., Goulding, E.H., Kluckman,K.D., Kim, HS. and Smithies,O. (1995) Cell 83: 483-492. Morham, S.G., Langenbach, R., Loftin, C.D., Tiano, H.F., Vouloumanos, N., Jennette, J.C., Mahler, J.F., Kluckman, K.D., Ledford, A., Lee, CA and Smithies, O. (1995) Cell 83: 473-482 Regier, M.K., DeWitt, D.L., Schindler, MS, and Smith, W.L. (1993), Arch. Biochem. Biophys. 301: 439-444 Otto, JC. and Smith, W.L (1994) J. Biol. Chem. 269: 19868-19875. Morita, l., Schindler, M., Regier, M.K., Otto, J.C., Hori, T., DeWitt, D.L and Smith, W.L. (1995) J. Biol. Chem. 270: 10902-10908 Spencer, A.G., Woods, J.W., Arakawa,T., Singer, LI. and Smith W.L. (1998) J. Biol. Chem. 273: 9886-9893 Merlie, J.P., Fagan, D., Mudd, J. and Needleman, P. (1988) J. Biol. Chem. 263: 3550-3553. Otto, J.C., DeWitt, D.L. and Smith, W.L. (1993) J. Biol. Chem. 268: 18234-18242. Picot, D., Loll, P.J. and Garavito, RM. (1994) Nature 367: 243-249. Otto, JC. and Smith, W.L. (1996) J. Biol. Chem. 271: 9906-9910. Wendt, K.U., Poralla, K. and Schulz, GE. (1997) Science 277: 1811- 1815. Siros, J. and Richards, J.S. (1992) J. Biol. Chem. 267: 6382-6388. Roth, G.J,. Siok, OJ, and Ozols, J. (1980) J. Biol. Chem. 255: 1301- 1304. 79 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. Luong, C., Miller, A., Barnett, J., Chow, J., Ramesha, C. and Browner, MP. (1996) Nature Struct. Biol. 3: 927-933. Kurumball, R.G., Stevens, A.M., Gierse, J.K., McDonald, J.J., Stegeman, R.A., Pak, J.Y., Gildehaus, D., Miyashire, J.M.,Penning, T.D., Seibert, K., lsakson, PC. and Stallings, WC. (1996) Nature 384: 644-648. Shimokawa,T., Kulmacz, R.J., DeWitt, D.L., and Smith, W.L (1990) J. Biol. Chem. 265: 20073-20076. Roth, G.J., Machuga, ET. and Ozolis, J. (1983) Biochemistry 22: 4672- 4675. Hemler, ME. and Lands, W.E.M. (1980) J. Biol. Chem. 255: 6253-6261. Kulmacz, R.J. and Wang, L. (1995) J. Biol. Chem. 270: 24019-24023. Marshall, P.J. and Kulmacz, R.J. (1988) Arch. Biochem. Biophys. 266: 162-170. Laneuville, O., Breuer,D.K., DeWitt, D.L., Hla,T., Funk, CD. and Smith, W.L. (1994) J. Pharrnacol. Exp. Ther. 271: 927-934. Meade, E.A., Smith, W.L. and DeWitt, D.L. (1992) J. Biol. Chem. 268:6610-6614. Hamburg, M. and Samuelsson,B. (1967) J. Biol. Chem. 240: 5336-43. Eling, T.E., Glasgow, W.C., Curtis, J.F., Hubbard, WC. and Handler (1991) J. Biol. Chem. 266: 12348-12355. Schreiber, J., Eling, TE and Mason, RP. (1986) Arch. Biochem. Biophys. 249: 126-136. Dietz, R., Nastainczky, W. and Ruf, H.H. (1988) Eur. J. Biochem. 171: 321-328. Laneuville, O., Breuer, D.K., Xu, N., Huang, Z.H., Gage, D.A., Watson, J.T., Lagarde, M., DeWitt, D.L. and Smith, W.L. (1995) J.Biol. Chem. 270: 1 9330-1 9336. Lands, W.E.M., LeTeIlier, P.R., Rome, L.H. and Vanderhoek, J.Y. (1973) Adv. Biosci. 9: 15-28. Culp, B.R., Titus, B.G. and Lands, W.E.M. (1979) Prostaglandins and Medicine 3: 269-278. 80 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. Rome, L.H., and Lands, W.E.M. (1975) Proc. Nat. Acad. Sci. USA 72:4863-4865. Mitchell, J.A., Akarasereenont, P., Thiemennann, 0., Flower, R.J. and Vane, JR. (1994) Proc. Natl. Acad. Sci, USA 90: 11693-11697. Lecomte, M., Laneuville, O., Ji, C., DeWitt, D.L. and Smith, W.L. (1994) J. Bio. Chem. 269: 13207-13215. Copeland,R.A., Williams, J.M., Giannara, J., Numberg, S., Covington, M., Pinto, 0., Pick, S. and Trzaskos, J.M. (1994) Proc. Natl. Acad. Sci. USA 91: 11202-11206. Quellet, M. and Percival, MD. (1995) Biochem. J. 306: 247-251. Masterrer, J.L., Zweifel, B.S., Manning, P.T., Hauser,S.D., Leary, K.M., Smith, W.G., lsakson, PC. and Seibert,K. (1994) Proc. Natl. Acad. Sci. USA 91 : 3228-3232. Guo, 0., Wang,L.,Ruan, K. and Kulmacz, R.J. (1996) J. Biol. Chem. 271: 19134-19139. Wong, E., Bayly, C., Waterman, H.L., Riendeau, D. and Mancini, J.A. (1997) J. Biol. Chem. 272: 9280-9286. Bhattacharyya, D.K., Lecomte, M., Rieke, C.J., Garavito, R.M.,and Smith, W.L. (1996), J. Bio. Chem. 271: 2179-2184 Mancini, J.A., Riendeau, D., Falgueyret, J., Vickers, P.J., and O’Neill, GP. (1995) J. Bio. Chem. 270: 29372-29377 Sanger, F., Nicklen, S. and Coulson, AR. (1977) Proc. Natl. Acad. Sci. USA 74:5463-5467. Markwell, M.A.K., Haas, S.M., Bieber, LL. and Tolbert, NE. (1976) Anal. Biochem. 87: 206-210. So, 0., Scarafia, L.E., Mak, A.Y., Callan, OH, and Swinney, DC. (1998). J. Bio. Chem. 273:5801-5807 Gan, Q., Browner, M.F., Sloane, D.L. and Sigal, E. (1996) J. Bio. Chem. 271 : 25412-25418 Van der Ouderaa, F.J.G. and Buytenhek, M. (1982) Methods in Enzymology 86: 60-68. 81