The Role of Arginine 120 of Human Prostaglandin Endoperoxide H Synthase-2 in the Interaction with Fatty Acid Substrates and Inhibitors*

Arg-120 is located near the mouth of the hydrophobic channel that forms the cyclooxygenase active site of prostaglandin endoperoxide H synthases (PGHSs)-1 and -2. Replacement of Arg-120 of ovine PGHS-1 with a glutamine increases the apparentK m of PGHS-1 for arachidonate by 1,000-fold (Bhattacharyya, D. K., Lecomte, M., Rieke, C. J., Garavito, R. M., and Smith, W. L. (1996) J. Biol. Chem. 271, 2179–2184). This and other evidence indicate that the guanido group of Arg-120 forms an ionic bond with the carboxylate group of arachidonate and that this interaction is an important contributor to the overall strength of arachidonate binding to PGHS-1. In contrast, we report here that R120Q human PGHS-2 (hPGHS-2) and native hPGHS-2 have very similar kinetic properties, but R120L hPGHS-2 catalyzes the oxygenation of arachidonate inefficiently. Our data indicate that the guanido group of Arg-120 of hPGHS-2 interacts with arachidonate through a hydrogen bond rather than an ionic bond and that this interaction is much less important for arachidonate binding to PGHS-2 than to PGHS-1. The K m values of PGHS-1 and -2 for arachidonate are the same, and all but one of the core residues of the active sites of the two isozymes are identical. Thus, the results of our studies of Arg-120 of PGHS-1 and -2 imply that interactions involved in the binding of arachidonate to PGHS-1 and -2 are quite different and that residues within the hydrophobic cyclooxygenase channel must contribute more significantly to arachidonate binding to PGHS-2 than to PGHS-1. As observed previously with R120Q PGHS-1, flurbiprofen was an ineffective inhibitor of R120Q hPGHS-2. PGHS-2-specific inhibitors including NS398, DuP-697, and SC58125 had IC50 values for the R120Q mutant that were up to 1,000-fold less than those observed for native hPGHS-2; thus, the positively charged guanido group of Arg-120 interferes with the binding of these compounds. NS398 did not cause time-dependent inhibition of R120Q hPGHS-2, whereas DuP-697 and SC58125 were time-dependent inhibitors. Thus, Arg-120 is important for the time-dependent inhibition of hPGHS-2 by NS398 but not by DuP-697 or SC58125.

Prostaglandin endoperoxide H synthases (PGHSs) 1 catalyze the committed step in the formation of prostanoids, the conversion of arachidonic acid to prostaglandin (PG) H 2 (1). There are two PGHS isozymes. PGHS-1 is expressed constitutively in most tissues, and PGH 2 formed via PGHS-1 is involved in modulating and coordinating intercellular, intratissue responses to fluctuations in the levels of circulating hormones. PGHS-2 is expressed inducibly in response to cytokines (2)(3)(4)(5), growth factors (6 -8), v-src (9), and tumor promoters (10), and products formed through the action of this isoform appear to play a role in cell growth and differentiation (11). Both PGHS-1 and PGHS-2 have similar kinetic and structural properties (1). Both isoforms catalyze two distinct reactions: (a) a cyclooxygenase reaction involving the bis-oxygenation of arachidonic acid to yield PGG 2 , and (b) a peroxidase reaction involving a twoelectron reduction of the 15-hydroperoxyl group of PGG 2 to form PGH 2 . PGH 2 , in turn, is further metabolized to the individual prostaglandins, prostacyclins, and thromboxanes by the appropriate synthases (1). PGHSs are the therapeutic targets of nonsteroidal anti-inflammatory drugs (NSAIDs) such as aspirin and ibuprofen, and as such, they are the focus of research on mechanisms of catalysis and inhibition.
Examination of the crystal structure of ovine PGHS-1 (oPGHS-1) complexed with flurbiprofen established that Arg-120 is positioned near the mouth of the hydrophobic tunnel that forms the cyclooxygenase active site and suggested that this residue interacts with the carboxylate groups of various nonsteroidal anti-inflammatory drugs and fatty acid substrates (12,13). Kinetic analyses of Arg-120 mutants of oPGHS-1 have been consistent with this hypothesis and indicate that an ionic interaction between the carboxylate group of arachidonate and the guanido group of Arg-120 is a major determinant of substrate and inhibitor binding to this isozyme (14,15). The homologous Arg-120 (Arg-106) 2 residue in human PGHS-2 (hPGHS-2) and murine PGHS-2 (mPGHS-2) is also postulated to interact with fatty acid substrates and carboxylic acid-containing inhibitors in a manner analogous to that observed for Arg-120 of oPGHS-1 (16 -18). To determine if this is the case, we replaced Arg-120 of hPGHS-2 with glutamine and leucine and examined the kinetic properties of the mutant enzymes with a variety of fatty acid substrates and inhibitors. Our data suggest that the Arg-120 residue of PGHS-2 forms a hydrogen bond and not an ionic linkage with the carboxyl group of arachi-* This work was supported in part by Program Project Grant P01 GM57323 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 2 Arg-106 of hPGHS-2 is homologous to Arg-120 of oPGHS-1. The numbering of Arg-106 in hPGHS-2 is based on numbering the aminoterminal methionine of the signal peptide as residue number 1. In this study, we number homologous residues of PGHS-2 in parallel with oPGHS-1; accordingly, Arg-106 of hPGHS-2 is numbered as Arg-120. donate, that the interaction of arachidonate with Arg-120 is quantitatively much less important for substrate binding to PGHS-2 than to PGHS-1; this, in turn, implies that hydrophobic residues in the core of the cyclooxygenase active site of PGHS-2 are relatively more important for substrate binding than is the case for PGHS-1. In examining the role of Arg-120 in the interaction of PGHS-2 with PGHS-2-specific inhibitors, this residue was found to be important in time-dependent inhibition by some, but not all, inhibitors. We interpret our data in the context of a model for substrate and inhibitor binding to hPGHS-2.
Preparation of Mutants by Site-directed Mutagenesis-Mutants of hPGHS-2 mutants were prepared using a Bio-Rad Muta-Gene kit essentially as described previously (19). M13mp19-hPGHS-2, which contains the coding region of hPGHS-2, was mutated following instructions supplied by the manufacturer. The single-stranded M13 phage samples were sequenced using the dideoxy method to identify mutants (20). The 1.8-kilobase insert from the replicative form of M13mp19-PGHS-2 containing the desired mutation was isolated by SalI digestion and subcloned into the SalI site of pOSML (19). The correct orientation was confirmed by restriction mapping with PstI. The double-stranded pOSML-hPGHS-2 mutant plasmids were sequenced to confirm that the mutations were present in the plasmids used for transfections. Plasmids were purified by CsCl gradient ultracentrifugation as described previously. The oligonucleotide primers used to prepare the mutants were 5Ј-GTGTTGACATCCCAGTCACATTTGATT-3Ј for R120Q and 5Ј -GTGTTGACATCCCTATCACATTTG-3Ј for R120L.
Transfections and Preparation of Cell Microsomes-COS-1 cells (ATCC CRL-1650) were grown to near confluence in Dulbecco's modified Eagle's medium containing 8% calf serum and 2% fetal calf serum in a water-saturated 5% CO 2 atmosphere. Approximately 16 h before the transfections, the cells were subcultured 1:2. The COS-1 cells were transfected using the DEAE-dextran/chloroquine method as described previously (19,21). Forty h after transfection, the cells were harvested using a rubber policeman to scrape them into 3 ml of ice-cold phosphatebuffered saline and collected by centrifugation at 1,000 ϫ g for 5 min. Transfected cells from 20 -50 tissue culture plates (100 mm) 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 ϫ g for 10 min at 4°C, and the resulting supernatants were centrifuged at 200,000 ϫ g for 50 min in a Beckman SW50.1 rotor to yield microsomal membranes. These membranes were resuspended by homogenization 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 Lowry method, using bovine serum albumin as the standard (22).
Cyclooxygenase and Peroxidase Assays-Cyclooxygenase assays were performed at 37°C using a Yellow Springs Instruments Model 5300 oxygen electrode. A typical assay contained 3 ml of 0.1 M Tris-Cl, pH 8.0, with 1 mM phenol, 100 M arachidonate, and 85 g of bovine hemoglobin (as a source of heme). Reactions were typically initiated by the addition of enzyme to the chamber, and initial rates of oxygen consumption were determined. To measure instantaneous inhibition, NSAIDs were added at the appropriate concentration to the assay chamber before the addition of enzyme. For measurements of time-dependent inhibition, NSAIDs, at the appropriate concentrations, were preincubated with 250 g of microsomal protein at 37°C for various times, reactions were initiated by the addition of the enzyme/inhibitor sample to the assay chamber, and initial rates of oxygen consumption were determined. For studies of substrate specificity, concentrations of 100 M 20:5n-3, 18:2n-6, 20:3n-6, or 20:2n-6 were used. K m determinations were performed for 20:4n-6, 20:5n-3, and 18:2n-6 using varying concentrations of these substrates.
Western Transfer Blotting-Microsomal membranes were resolved by one-dimensional SDS-polyacrylamide gel electrophoresis and transferred electrophoretically to 0.45-mm nitrocellulose membranes (Schleicher & Schuell). Membranes were blocked for 0.5-12 h in 3% (w/v) dry milk and 0.1% (v/v) Tween 20/Tris-buffered saline, followed by an incubation with a 1:3,000 dilution of an affinity-purified antibody to the 18-amino acid cassette unique to PGHS-2 (25) in 1% dry milk and 0.1% Tween 20/Tris-buffered saline for 2 h at 25°C. Membranes were washed and incubated with a 1:1,000 dilution of a goat anti-rabbit IgG horseradish peroxidase for 1 h. The membrane was washed and incubated with Amersham Pharmacia Biotech ECL reagents and exposed to Kodak XAR film for chemiluminescence.
Analyses of Prostaglandin and Hydroxy Acid Products-Microsomal membrane suspensions (250 g of protein for native and R120Q hPGHS-2 and 1 mg of protein for R120L hPGHS-2) were incubated with 100 M [1-14 C]arachidonic acid for 10 min in 0.1 M Tris-HCl, pH 8.0, containing 1 M hematin and 1 mM phenol. The reactions were stopped by the addition of 1.4 ml of chloroform:methanol (1:1, v/v) and centrifuged at 3,000 rpm for 5 min. The supernatants were transferred to clean tubes, and 0.6 ml of chloroform and 0.32 ml of 0.88% formic acid were added. After centrifugation, the organic layer was removed and dried under N 2 . The dried samples were redissolved in 50 l of chloroform and spotted on Silica Gel 60 thin layer chromatography plates. Plates were developed twice in benzene:dioxane:acetic acid:formic acid (82:14:1:1, v/v/v/v) and exposed to XAR-5 film for 48 h to visualize the products. Prostaglandin synthesis by hPGHS-2 and the R120 mutant enzymes was quantified by densitometry using a Molecular Dynamics Storm 820 PhosphorImager with ImageQuant software.
The relative amounts of 11-and 15-HETE were also quantified by reverse phase-HPLC. Native or mutant oPGHS-1 (1 mg of microsomal protein) was reacted with 100 M arachidonic acid for 30 min at 37°C. The products were collected as described above, dried under N 2 , and resuspended in HPLC buffer (0.1% aqueous acetic acid/acetonitrile containing 0.1% acetic acid; 1:1, v/v). 15-and 11-HETEs were separated by reverse phase-HPLC using a C-18 column purchased from Vydac Co. and detected with a Waters Model 600 HPLC equipped with a 990 photo diode array detector set to 234 nm. The strong component of the mobile phase was 0.1% aqueous acetic acid, and the eluting solvent was acetonitrile containing 0.1% acetic acid. The flow rate was 1 ml/min. The following elution profile was used: 0 -30 min, 30% acetonitrile; 30 -100 min, 50% acetonitrile; 100 -125 min, 75% acetonitrile; and 125-130 min, 100% acetonitrile. The retention times for 15-HETE and 11-HETE were 36 and 38 min, respectively. The relative amounts of 11-HETE and 15-HETE were determined by measuring the area associated with each peak.

RESULTS
Kinetic Properties of hPGHS-2, R120Q hPGHS-2, and R120L hPGHS-2-hPGHS-2, R120Q hPGHS-2, and R120L hPGHS-2 were expressed transiently in COS-1 cells. As shown in Fig. 1, quantitative Western transfer blotting established that hPGHS-2, R120Q hPGHS-2, and R120L hPGHS-2 were present at essentially identical levels in microsomes prepared from the transfected COS-1 cells and that there was no immunoreactive hPGHS-2 in sham-transfected cells. Table I provides a tabulation of the results of the product analyses derived from the quantitation of radio thin layer chromatograms of the products formed upon incubation of hPGHS-2, R120Q hPGHS-2, and R120L hPGHS-2 with [1-14 C]arachidonic acid. The native and mutant enzymes formed a similar mixture of prostanoid products, but both R120Q hPGHS-2 and R120L hPGHS-2 formed about 1.5 times more 11-HETE as the native enzyme 3 ; when the 11-HETE and 15-HETE products were separated and quantified by reverse phase-HPLC, the ratios of 11-HETE to 15-HETE were determined to be 0.79 for native hPGHS-2 and 1.5 for both R120Q hPGHS-2 and R120L hPGHS-2 (i.e. approximately the same ratios found using radio thin layer chromatography (Table I)). Also summarized in Table I are kinetic data for the native and mutant enzymes. hPGHS-2 and R120Q hPGHS-2 have virtually identical kinetic properties with arachidonate as substrate; moreover, the peroxidase activities of these mutants are quite similar. In contrast, R120L hPGHS-2 has considerably less peroxidase and cyclooxygenase activity than does the native enzyme. The R120Q mutant has a t1 ⁄2 for suicide inactivation of 22 s compared with 15 s for native hPGHS-2. R120L hPGHS-2 was also found to undergo suicide inactivation, but a t1 ⁄2 was not determined for this mutant.
Effects of Inhibitors on hPGHS-2 and R120Q hPGHS-2-Flurbiprofen causes a time-dependent inhibition of both PGHS-1 and PGHS-2 (21,24,26). To determine if Arg-120 of hPGHS-2 plays a role similar to that of Arg-120 in oPGHS-1 in inhibition by flurbiprofen (14,15,27,28), we characterized both instantaneous and time-dependent inhibition of hPGHS-2 and the R120Q hPGHS-2 mutant (Table III). Native hPGHS-2 had an IC 50 value for flurbiprofen of 1 M and underwent time-dependent inhibition with a t1 ⁄2 of less than 1 min. In contrast, the R120Q mutant had an IC 50 value of 0.5 mM and did not undergo time-dependent inhibition during a 20-min incubation with 0.5 mM flurbiprofen (data not shown).  1. Western transfer blotting of native hPGHS-2, R120Q hPGHS-2, and R120L hPGHS-2. COS-1 cells were sham-transfected or transfected with native hPGHS-2, R120Q hPGHS-2, or R120L hPGHS-2. Microsomal samples were prepared, and 50 g of protein from each of the indicated transfections were subjected to Western transfer blotting analyses as described in the text using an anti-peptide antibody directed against the unique 18-amino acid cassette located near the carboxyl terminus of PGHS-2. The relative chemiluminescence intensities were as follows: native hPGHS-2, 100%; R120Q hPGHS-2, 105%; and R120L hPGHS-2, 98%. mPGHS-2, murine PGHS-2 standard. b Peroxidase activities were determined as described in the text using H 2 O 2 and N,N,NЈ,NЈ-tetramethyl-p-phenylenediamine as oxidizing and reducing cosubstrates, respectively. c Determined by quantitative densitometry of autoradiographs of thin layer chromatograms of products formed using 100 M [1-14 C]arachidonate as described in the text. PG products include all those derived from PGH 2 including HHTre, PGD 2 , PGE 2 , and PGF 2 ␣. d Estimated from quantitative densitometry from radio thin layer chromatographic assays. e ND, not determined.

TABLE III Instantaneous and time-dependent inhibition of native hPGHS-2 and
R120Q hPGHS-2 As described in detail in the text, IC 50 values for instantaneous inhibition were determined by using different concentrations of inhibitor and by measuring the cyclooxygenase rates immediately upon the addition of aliquots of microsomal enzyme preparations to assay mixtures containing substrates (100 M arachidonate and O 2 ) and inhibitor; t 1/2 values for time-dependent inhibition were obtained by preincubating the microsomal enzyme at 37°C with the indicated concentration of inhibitor and measuring the inhibition observed upon the addition of the preincubation mixture to an assay chamber containing the substrates (21). Experiments with flurbiprofen were performed twice, and those with other inhibitors were performed three times with similar results. All measurements of cyclooxygenase rates were performed in duplicate or triplicate. Several PGHS-2-specific inhibitors have been shown to cause time-dependent inhibition of PGHS-2 but simple competitive inhibition of PGHS-1 (29 -31). We next examined the effect of changing Arg-120 to a glutamine on the interactions with several of these inhibitors (Fig. 2 and Table III). Native hPGHS-2 had an IC 50 value with NS398 of 50 M and exhibited time-dependent inhibition. The R120Q mutant had a 5-fold lower IC 50 value (10 M) but, interestingly, did not undergo time-dependent inhibition (Fig. 2A). The R120Q hPGHS-2 mutant had dramatically reduced IC 50 values with both SC58125 and DuP-697 (Table III). There was a somewhat decreased t1 ⁄2 for timedependent inhibition with DuP-697 ( Fig. 2B; Table III) although both the native and mutant enzymes underwent timedependent inhibition by DuP-697. Both the native and mutant enzymes also underwent time-dependent inhibition with SC58125 ( Fig. 2C; Table III).

DISCUSSION
Effect of Arg-120 Mutations of hPGHS-2 on Substrate Binding and Oxygenation-Previous crystallographic and mutagenesis studies of oPGHS-1 have indicated that ionic bonds between the guanido group of Arg-120 and the carboxylate groups of arachidonic acid and a variety of NSAIDs are the major determinants of the affinity of binding of substrate and inhibitors to the cyclooxygenase active site of this isozyme (12,14,15). The carboxylate groups of flurbiprofen and other NSAIDs are located within ionic bonding distance of Arg-120 in oPGHS-1 crystal structures (12). Replacing Arg-120 of oPGHS-1 with a neutral glutamine residue results in about a 1,000-fold increase in the K m for arachidonate, a 1,000-fold increase in the IC 50 value for flurbiprofen, and a loss in the ability of flurbiprofen to cause time-dependent inhibition (14). In contrast, neutral isosteric substitutions of a variety of residues lining the hydrophobic cyclooxygenase channel of oPGHS-1 affect substrate positioning and the efficiency of cyclization but, in general, have relatively minor effects (Ͻ10-fold) on the K m of arachidonate for the cyclooxygenase active site (32,33). 4 In the experiments reported here, we examined the effects of mutations of the homologous Arg-120 residue of hPGHS-2 on the interactions of substrates and inhibitors with this inducible isozyme. In marked contrast to the results obtained in studies of oPGHS-1, the R120Q hPGHS-2 mutant had essentially the same kinetic properties as native hPGHS-2 when arachidonate was used as the substrate; moreover, both the native and mutant enzymes formed the same products in approximately similar proportions. In contrast, R120L hPGHS-2 oxygenated arachidonate relatively inefficiently. These results indicate that there is hydrogen bonding, not ionic bonding, between the carboxylate group of arachidonate and the guanido group of Arg-120 in native PGHS-2, that a hydrogen bonding interaction can occur through the amide group of glutamine in the R120Q mutant, and, accordingly, that other favorable interactions within the cyclooxygenase active site of PGHS-2 must compensate for the lack of ionic bonding of arachidonate in PGHS-2 versus PGHS-1 (Fig. 3). Other results that are consistent with this interpretation include the following: (a) there is a large decrease in cyclooxygenase catalytic efficiency associated with the mutation of Arg-120 to glutamate (34,37) or leucine in PGHS-2; (b) the crystal structure of mouse apo-PGHS-2 with arachidonate bound in the cyclooxygenase site suggests that the carboxylate group is in close proximity to Arg-120 (35), and 4  this is also likely to be the case with the R120Q hPGHS-2 mutant (Fig. 3); (c) anandamide (arachidonylethanolamide), which lacks a carboxylate group, is a substrate for hPGHS-2 (36); (d) mutation of Tyr-355, which neighbors Arg-120, to a phenylalanine has little effect on the k cat /K m value for arachidonate oxygenation (36); and (e) the hydrophobic tunnel of the active site of PGHS-2 is larger (i.e. more open and accommodating) than that of PGHS-1 (1,12,16,17).
Comparisons of the crystal structures of PGHS-1 and PGHS-2 provide no definitive explanation for the differing behavior of Arg-120 between the two isozymes. No polar residue other than Arg-120 in PGHS-2 could ligand to the carboxylate of arachidonic acid. However, one interesting feature is seen; although the relatively hydrophobic environment around Arg-120 is preserved between the two isozymes, Arg-120 is involved in an extended electrostatic/hydrogen bond network with Glu-524 and Arg-513 in PGHS-2 (16,17). This network could also involve the carboxylate group of arachidonic acid when it binds to the cyclooxygenase site (Fig. 3). However, no such network can exist in PGHS-1 because residue 513 is a histidine, and this much shorter residue is too distant from Glu-524 to interact. The differing electrostatic and hydrogen bonding environments experienced by Arg-120 in PGHS-1 and PGHS-2 could explain the observed differences in arachidonic acid binding between the two isoforms. However, an understanding of the structural details underlying our kinetic results will require direct crystallographic observations of arachidonate binding in both isoforms and in mutant enzymes.
Collectively, the results of various studies on Arg-120 in PGHS-1 and PGHS-2 clearly show that the interactions involved in the binding of arachidonate and other fatty acids are qualitatively and quantitatively different between the isozymes. In PGHS-1, the ionic character of Arg-120 contributes very significantly to the binding of arachidonate; in PGHS-2, a neutral substitution at this site does not disturb arachidonate binding. Because the K m values of PGHS-1 and -2 for arachidonate are the same, other interactions between PGHS-2 and arachidonate must compensate for the weaker interactions between Arg-120 and arachidonate in PGHS-2. One explanation is that the hydrophobic residues lining the cyclooxygenase channel contribute much more significantly to arachidonate binding in PGHS-2 than in PGHS-1. Nonetheless, because all but one of the core residues of the active sites in the two isozymes are identical, the enhancement of hydrophobic interactions in PGHS-2 must be subtle.
It should be noted that there is a small difference in the ratio of the products formed with the R120Q hPGHS-2. 11-HETE is produced in greater abundance by this mutant (and by R120L hPGHS-2) than by the native enzyme. Thus, although Arg-120 appears not to contribute as importantly to the binding affinity of arachidonate for PGHS-2 versus PGHS-1, this residue does appear to help position this fatty acid appropriately in PGHS-2 to favor net bis-oxygenation over mono-oxygenation. The importance of this type of interaction in substrate positioning has also been shown with human 15-lipoxygenase, where substitution of the arginine residue thought to interact with the carboxylate group of arachidonic acid with a leucine appears to result in a slight change in the conformation of arachidonate such that the position of oxygen insertion is altered (38).
The oxygenation of both dihomo-␥-linolenic acid and linoleic acid proceeded efficiently with the R120Q hPGHS-2 mutant. Again, this suggests that there is a hydrogen bonding interaction between the carboxylate groups of these latter two fatty acids and Arg-120 in native hPGHS-2. In contrast to the results obtained with arachidonate, dihomo-␥-linolenate and linoleate, eicosapentaenoic acid had markedly increased K m values with R120Q hPGHS-2, and 11,14-eicosadienoic acid was not oxygenated by the R120Q mutant. This implies that Arg-120 is essential for the oxygenation of 20:5n-3 and 20:2n-6. In the case of 20:5n-3, Arg-120 appears to be important for positioning this fatty acid for catalysis but not for binding (i.e. despite having a much higher K m with R120Q hPGHS-2, 20:5n-3 was an effective inhibitor of the oxygenation of arachidonate by this mutant, indicating that 20:5n-3 does bind effectively to R120Q hPGHS-2). In a broader sense, these findings with various fatty acids imply that each fatty acid substrate, like different inhibitors, depends on a unique set of interactions within the cyclooxygenase active site of hPGHS-2 for binding and oxygenation.
Effect of Arg-120 Mutations of hPGHS-2 on Inhibitor Binding-Flurbiprofen is representative of the 2-phenylpropionic acid class of NSAIDs that are relatively nonspecific for PGHS isozymes. Flurbiprofen causes a time-dependent inhibition of both PGHS-1 and -2 (21,26). Flurbiprofen appears to bind much less tightly to R120Q hPGHS-2 than to the native enzyme, as evidenced by a 500-fold increase in the IC 50 value with arachidonate as the substrate. Furthermore, even at high concentrations, flurbiprofen failed to cause a time-dependent inhibition of R120Q hPGHS-2. Thus, as observed with oPGHS-1 (14), the Arg-120 group of hPGHS-2 is necessary for both efficient binding and time-dependent inhibition of hPGHS-2 by flurbiprofen.
The substitution of glutamine for Arg-120 resulted in changes in inhibitor profiles with PGHS-2-specific inhibitors. In all cases, the IC 50 values for the inhibitors were actually decreased. For DuP-697 and SC58125, the changes were marked (on the order of 1,000-fold). This would suggest that the guanido group of Arg-120 actually interferes with the binding of these inhibitors. The other notable finding was that NS398 failed to cause time-dependent inhibition of R120Q hPGHS-2. With DuP-697 and SC58125, the sulfonamide group of the inhibitor is located in the side pocket neighboring Val-509 (16,17), whereas with NS398, the sulfonamide group neighbors Arg-120. 5 Thus, NS398, like flurbiprofen, depends on an interaction with Arg-120 for time-dependent inhibition to occur.