- Email: [email protected]
Different Suicide Inactivation Processes for the Peroxidase and Cyclooxygenase Activities of Prostaglandin Endoperoxide H Synthase-1
Biochemical and Biophysical Research Communications 289, 869 – 875 (2001) doi:10.1006/bbrc.2001.6071, available online at http://www.idealibrary.com on
Different Suicide Inactivation Processes for the Peroxidase and Cyclooxygenase Activities of Prostaglandin Endoperoxide H Synthase-1 Inseok Song,* Terry M. Ball,† and William L. Smith† ,1 †Department of Biochemistry and Molecular Biology, Michigan State University, East Lansing, Michigan; and *Department of Life Science, University of Seoul, Seoul, Korea
Received November 5, 2001
Prostaglandin endoperoxide H synthases (PGHSs)-1 and -2 have a cyclooxygenase (COX) activity involved in forming prostaglandin G 2 (PGG 2) from arachidonic acid and an associated peroxidase (POX) activity that reduces PGG 2 to PGH 2. Suicide inactivation processes are observed for both POX and COX reactions. Here we report COX reaction conditions for PGHS-1 under which complete COX inactivation occurs but with >60% retention of POX activity. The rates of POX inactivation were compared for native oPGHS-1 versus Y385F oPGHS-1, a mutant that cannot form the Tyr385 radical of COX Intermediate II; the rates were the same for both native and Y385F oPGHS-1. Our data indicate that a COX Intermediate II/acyl or product complex is the precursor in COX inactivation. However, another species, probably an Intermediate II-like species but with a radical centered on a tyrosine other than Tyr385, is the immediate precursor for POX inactivation. © 2001 Elsevier Science Key Words: arachidonic acid; linoleic acid; aspirin; hydroperoxide; flurbiprofen; Compound I; Compound II; peroxidase; cyclooxygenase.
The committed step in prostaglandin synthesis is catalyzed by prostaglandin endoperoxide H synthases (PGHS)-1 and -2 (1, 2). PGHSs catalyze the conversion of arachidonic acid (AA), two molecules of O 2 and two electrons to PGH 2. This involves two separate reactions—a cyclooxygenase (COX) reaction in which AA is oxygenated to yield PGG 2 and a peroxidase (POX) reaction in which the 15-hydroperoxyl group of Abbreviations used: PGHSs, prostaglandin endoperoxide H synthases; PG, prostaglandin; oPGHS-1, ovine PGHS-1; EtOOH, ethyl hydrogen peroxide; TMPD, N,N,N⬘,N⬘-tetramethylphenylenediamine; COX, cyclooxygenase; POX, peroxidase; AA, arachidonic acid. 1 To whom correspondence should be addressed at 513 Biochemistry Building, Department of Biochemistry and Molecular Biology, Michigan State University, East Lansing, MI 48824. Fax: (517) 3539334. E-mail: [email protected].
PGG 2 is reduced to PGH 2. PGHS-1 and -2 are closely related structurally and enzymologically. However, the expression of each isozyme is regulated differently and each isozyme subserves different physiological functions (reviewed in (1, 3)). The COX and POX activities of PGHSs are functionally interdependent. The mechanism of catalysis involves an initial peroxide-dependent oxidation of the heme group at the POX active site of PGHS (4) to yield an oxyferryl heme radical cation (Compound I) (5, 6). Compound I can undergo two sequential one electron reductions involving exogenous electron donors to produce Compound II and then native heme; or alternatively, Compound I can undergo an intramolecular one electron reduction by a protein tyrosine residue to form a COX Intermediate II that has an oxyferryl heme and an associated tyrosyl radical (7, 8). The tyrosyl radical of Intermediate II-Tyr385—abstracts the 13 proS hydrogen from AA yielding an AA radical (9 –12), the first definable chemical species unique to the COX reaction. Both the COX and POX activities of PGHSs undergo rapid first order suicide inactivation reactions. Suicide inactivation places an upper limit on the capacities of cells to synthesize prostaglandins and thromboxanes. The t 1/2 for the POX activity of PGHS in the presence of a peroxide is 2–20 s (13). POX inactivation has been proposed to involve a sequence of POX Compound I to COX Intermediate II to Intermediate III to suicide-inactivated enzyme (13, 14). The COX also exhibits a first order loss of activity when PGHS is incubated with fatty acid substrates (1, 15, 16). Although these processes have usually been measured under different sets of experimental conditions, the rates of POX and COX suicide inactivation are comparable (1, 13, 16). The chemical mechanism(s) of suicide inactivation is not known. Suicide inactivation of native enzyme does not appear to involve either heme oxidation (13) or covalent modification of the enzyme by an activated form of substrate fatty acid (17, 18). It seems likely that it is a
869
0006-291X/01 $35.00 © 2001 Elsevier Science All rights reserved.
Vol. 289, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Different rates of inactivation of cyclooxygenase and peroxidase activities of oPGHS-1. COX and POX activities were monitored as described under Experimental Procedures. (A) oPGHS-1 (210 units of COX activity) was reconstituted with hematin (5 M final concentration) and added to a 3-ml COX assay mixture containing 2 M EtOOH, 200 M TMPD, 5 M hematin, and 110 M AA. O 2 consumption was monitored at 23°C using an O 2 electrode. The initial rate was 65 nmol hydroperoxide (PGG 2 )/min. After 4 min when the rate had dropped to 0.9 nmol hydroperoxide/min, EtOOH (150 nmol) and then AA (330 nmol) were added. (B) Reactions were initiated as in A. After 4 min when, again, the rate had dropped to 0.9 nmol hydroperoxide/min, EtOOH (150 nmol) and then oPGHS-1 (210 units reconstituted with 5 M heme) were added. (C) In a parallel experiment O 2 consumption was monitored as in A and B. Four minutes after the initial addition of enzyme, 1 ml of the COX assay mixture was removed and placed in a spectrophotometer cuvette and POX activity (TMPD oxidation) was measured upon addition of EtOOH (final concentration 50 M); a control experiment was performed in the absence of AA. All of the experiments depicted in this figure were performed a minimum of three times with similar results.
free radical-mediated process leading to intramolecular protein cross-linking. POX and COX inactivation have many common characteristics and, quite reasonably, have been assumed to arise from the same COX Intermediate II species (13, 14). However, the rate of COX inactivation is slightly different for different substrates (15, 16) whereas the rate of POX suicide inactivation appears to be independent of the nature of the oxidizing peroxide substrate (13, 14). These differences led us to determine if we could distinguish between the POX and COX suicide inactivation processes. EXPERIMENTAL PROCEDURES COX and POX assay procedures. Rates of O 2 uptake were measured using a YSI Model 5300 oxygen electrode. Reaction mixtures contained 3 ml of 0.1 M TrisHCl, pH 8.0, containing 110 M fatty
acid substrate (Cayman Chemical Co., Ann Arbor, MI), 200 M N,N,N⬘,N⬘-tetramethylphenylenediamine (TMPD) and 1 M ethyl hydrogen peroxide (EtOOH) equilibrated at 23°C. Typically, 210 units of highly purified oPGHS-1 (19) (i.e., with specific activities of ⱖ10,000 units/mg protein) were equilibrated with 5 M hematin for 3 min and used to initiate the COX reactions. One unit of COX activity is defined as that amount of enzyme that will catalyze the uptake of 1 nmol of O 2 per min at 37°C in an assay mixture containing 0.1 M TrisHCl, pH 8.0, 100 M AA, 5 M hematin and 1 mM phenol. Velocity measurements were obtained at 1-s intervals using Dasy Lab Data Processing software (DASYTEC, Amherst, NH). Corrections for damping of electrode responses were made using the protocol of Wei et al. (20). To determine t 1/ 2 values, data of initial rates versus time were first transferred to an Excel worksheet and graphed using an XY scatter plot. This graph was fitted to an exponential plot and a resulting exponential equation was used for the t 1/ 2 determination; t 1/ 2 ⫽ 0.693/k where k is the exponential constant. POX assays were performed essentially as reported previously (19) using 1 ml of 0.1 M TrisHCl, pH 8.0, containing 5 M hematin, 200 M TMPD and 50 M EtOOH equilibrated at 23°C and an ⫽
870
Vol. 289, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS purified as described previously for native histidine-tagged oPGHS-1 (22). POX assays were performed as described above but with 100 M EtOOH.
TABLE 1
Effect of Peroxidase Inactivation on Cyclooxygenase Activity EtOOH (nmol)
Peroxidase activity remaining (%)
Cyclooxygenase activity remaining (%)
0 25 50 75 100
100 87 ⫾ 3.5 54 ⫾ 0.5 28 ⫾ 1.5 12 ⫾ 5.0
100 61 ⫾ 1.0 55 ⫾ 8.5 30 ⫾ 6.5 17 ⫾ 1.5
Note. Experiments were performed as described in the text. oPGHS-1 (70 units of COX activity) was preincubated for 5 min in a spectrophotometer cuvette containing 1 ml of 0.1 M TrisHCl, pH 8.0, 5 M hematin and 200 M TMPD. The indicated amounts of EtOOH were added, and the sample was incubated at 23° for 5 min (by which time no further peroxide reduction was occurring). The samples were then assayed for POX or COX activity. To measure POX activity 50 nmol EtOOH plus sufficient TMPD to restore the TMPD concentration to 200 mM was added; to measure COX activity 110 nmol of AA plus sufficient TMPD to restore the TMPD concentration to 200 mM was added. In each case the rate of TMPD oxidation was monitored at 611 nm. Values represent the mean ⫾ SD from two independent experiments.
12,200 L mol ⫺1 cm ⫺1 for TMPD oxidation (21); the reactions were initiated by adding purified oPGHS-1, histidine-tagged oPGHS-1 or histidine-tagged Y385F oPGHS-1. In some experiments involving COX assays, 1 ml of the COX assay mixture was removed and placed in a spectrophotometer cuvette and TMPD oxidation was measured upon addition of EtOOH (final concentration 50 M). One unit of POX activity is defined as that amount of activity that will catalyze the reduction of 1 nmol of EtOOH at 23°C in 1 min in a reaction mixture containing 1 ml of 0.1 M TrisHCl, pH 8.0, 5 M hematin, 100 M EtOOH and 4.5 mM guaiacol. Preparation, expression, and assay of histidine-tagged native and Y385F oPGHS-1. Site-directed mutagenesis of histidine-tagged oPGHS-1 in pFastbac (22) was performed using the QuickChange Site-Directed Mutagenesis Kit (Stratagene) according to the protocol of the manufacturer with the oligonucleotide primers 5⬘-GGAGTTCAACCAGCTGTTCCACTGGCACCCGCTCA-3⬘ (sense) and 5⬘-TGAGCGGGTGCCAGTGGAACAGCTGGTTGAACTCC-3⬘ (antisense), which code for amino acids 380 –390 (with one nucleotide on either side) of oPGHS-1. The DNA sequence of the mutant was confirmed by automated fluorescent DNA sequencing performed by the MSU DNA sequencing facility. Recombinant baculovirus particles were isolated using the Bac-to-Bac baculovirus expression system (Life Technologies, Inc.). The mutant enzyme was expressed and
POX inactivation experiments. Purified oPGHS-1 (70 units of COX activity) was preincubated for 5 min in a spectrophotometer cuvette containing 1 ml of 0.1 M TrisHCl, pH 8.0, 5 M hematin and 200 M TMPD and in some cases 5 M flurbiprofen. Varying amounts of EtOOH (0 –100 nmol) were added, and the sample was incubated at 23°C for 5 min to inactivate varying amounts of POX activity. The samples were then assayed for POX or COX activity. To measure POX activity 50 nmol EtOOH plus sufficient TMPD to restore the TMPD concentration to 200 mM was added; to measure COX activity 110 nmol of AA plus sufficient TMPD to restore the TMPD concentration to 200 mM was added. In each case the rate of TMPD oxidation was monitored at 611 nm.
RESULTS Cyclooxygenase inactivation without peroxidase inactivation. In the experiments depicted in Figs. 1A and 1B COX activity was assayed by monitoring O 2 uptake at 23°C following the addition of heme-reconstituted oPGHS-1 to an assay chamber containing AA and O 2 as COX substrates, TMPD as a POX reducing cosubstrate and a small amount of EtOOH to initiate COX catalysis. In each case, the initial COX rate was 65 nmol of hydroperoxide/min, and the rate dropped to less than 0.9 nmol of hydroperoxide/min within 4 min of adding enzyme to the assay mixture. At that time approximately 27 nmol of PGG 2 had been formed. The loss of cycloxygenase activity exhibited first order kinetics, and the t 1/ 2 was determined to be 16 s in both Figs. 1A and 1B. Addition of either EtOOH or AA after the COX rate had dropped to near zero did not cause a resumption of O 2 consumption (Fig. 1A); however, when fresh enzyme was added, O 2 uptake resumed, again at a rate of 65 nmol of hydroperoxide/min, and with a t 1/ 2 for COX inactivation of 17 s (Fig. 1B). Thus, under the conditions used in these experiments the COX activity of oPGHS-1 underwent suicide inactivation, and no appreciable COX activity remained after a 4-min incubation. In a parallel experiment O 2 consumption was monitored exactly as in Figs. 1A and 1B, but 4 min after the initial addition of enzyme, 1 ml of the assay mixture was quickly transferred to a spectrophotometer cu-
TABLE 2
Cyclooxygenase Suicide Inactivation of oPGHS⫺ with Different Fatty Acid Substrates Fatty acid
Initial cyclooxygenase rate (nmol hydroperoxide/min)
Cyclooxygenase half-life (s)
Total product (nmol of hydroperoxide)
AA (20:4n-6) Linoleic (18:2n-6) Eicosapentaenoic (20:5n-3) Eicosadienoic (20:2n-6)
65 ⫾ 4.7 26 ⫾ 0.96 16 ⫾ 0.87 42 ⫾ 3.5
16 ⫾ 0.85 45 ⫾ 12 15 ⫾ 0.07 17 ⫾ 3.3
28 ⫾ 3.3 31 ⫾ 11.3 9.5 ⫾ 4.7 16 ⫾ 8.8
Note. Initial rates of COX activity, COX half-lives (t 1/ 2 values), and amounts of product were determined as described in the legend to Fig. 1 and represent the mean ⫾ SD from a minimum of three independent measurements. 871
Vol. 289, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Experiments parallel to those depicted for AA in Fig. 1 were also performed using linoleic acid as the COX substrate and yielded similar results. Collectively, our data indicate that a major portion (ⱖ60%) of the POX activity of PGHS-1 can be retained following complete inactivation of the COX activity.
FIG. 2. Peroxidase inactivation with native and Y385F oPGHS-1. Histidine-tagged native and Y385F oPGHS-1 were solubilized and partially prepared as described under Experimental Procedures. Specific COX and POX activities of native and Y385F oPGHS-1 were measured as described under Experimental Procedures; note that the standardized conditions used to determine these specific activities are different from those used in the experiments depicted in Fig. 1. The specific COX and POX activities of native oPGHS-1 were 12,500 and 12,700 units/mg, respectively; these values indicate that the enzyme is about 40% pure. The specific POX activity of Y385F oPGHS-1 isolated using an identical protocol was 3970 units/mg. (A) Histidine-tagged Y385F oPGHS-1 (12 units of POX activity) was added to a POX reaction mixture containing EtOOH and guaiacol as the oxidizing and reducing cosubstrates, respectively, and activity was monitored by measuring the changes in absorbance at 426 nm as described in the text; EtOOH (100 nmol) and fresh enzyme (12 POX units) were added at the times indicated. (B) Native oPGHS-1 or Y385F oPGHS-1 (80 POX units) was added to standard POX assay mixtures and oxidation of guaiacol was monitored spectrophotometrically as described in the text.
vette, 50 nmol of EtOOH was added and POX activity was measured (Fig. 1C). An initial POX rate of 26 nmol hydroperoxide reduced/min was observed. Thus, POX activity remained following the loss of COX activity. A control experiment like those depicted in Figs. 1A and 1B was performed in the absence of AA, and a 1-ml aliquot of this sample was also assayed for POX activity (Fig. 1C). Comparison of the POX rates of the two samples (i.e., with versus without AA during the first 4 min) established that 68% of the POX activity remained after COX inactivation in the presence of AA.
Effect of peroxidase inactivation on cyclooxygenase activity. Different amounts of the POX activity of oPGHS-1 were inactivated by incubating the enzyme with 0 –100 M EtOOH for 5 min, and the amount of residual COX activity was determined (Table 1). There was a close correspondence between the percentage of POX activity inactivated and the amount of COX activity that was lost. This result was not unexpected because activation of the COX is directly dependent on the functioning of the POX, and only those enzyme molecules that have POX activity would be expected to have COX activity. However, these results do indicate that a relatively “pure” POX inactivation, with no appreciable additional, accompanying COX inactivation, occurs when oPGHS-1 is incubated with EtOOH and TMPD and without AA. A parallel experiment similar to that presented in Table 1 was performed in which the POX activity of oPGHS-1 that had been pretreated with 5 M flurbiprofen was partially inactivated with 25– 40 nmol of EtOOH. The presence of flurbiprofen in the sample causes complete, time dependent inhibition of COX activity under these conditions (23) but did not affect the degree of either POX or COX inactivation caused by incubating the enzyme with EtOOH and TMPD in the absence of flurbiprofen (data not shown). This is an important result because it indicates that simple occupancy of the COX site under conditions in which a Tyr385 radical is generated (24) does not lead to a loss of COX activity beyond that caused by the loss of POX activity. This is consistent with the concept that COX inactivation requires substrate oxygenation and not simply occupancy of the COX active site. Effects of different fatty acids on cyclooxygenase and peroxidase suicide inactivation. We analyzed the rates of COX suicide inactivation with several different fatty acid substrates under the same experimental conditions used for the experiments depicted in Fig. 1 with AA. The results are summarized in Table 2. It should be noted that the reaction mixtures were supplemented initially with 2 M EtOOH and that under these conditions neither the rate nor extent of the reactions was limited by the availability of initiator hydroperoxide. The rates of COX suicide inactivation with different fatty acids were substantially different from one another. For example, the rate of inactivation with linoleic acid was only one third of the rate observed with AA. Half-lives for COX inactivation with linoleate and AA were also measured over the first 30 s of the oxygenase reaction in the presence of different concentrations of EtOOH (2, 10, and 20 M). Within
872
Vol. 289, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. Cyclooxygenase and peroxidase catalysis and inactivation. A model depicting the cycles of COX and POX catalysis and suicide inactivation (1, 14). AA, arachidonic acid; PPIX, protoporphorin IX; ROOH, an alkyl hydroperoxide (e.g., PGG 2); ROH, an alkyl alcohol.
experimental error the initial rates were the same and the t 1/ 2 values were the same at each of these starting peroxide concentrations. These results indicate that the number of productive turnovers of which the COX is capable differs significantly among different substrates. Related and similar results have been reported by others in comparing AA with 5,8,11,14,17eicosapentaenoic acid (16). 2 To determine if there is an effect of COX fatty acid substrates on POX inactivation, POX activities were measured in the presence of no fatty acid, 110 M AA or 110 M linoleate under reaction conditions used to measure COX inactivation in Figs. 1A–1C (but with 50 M EtOOH instead of 2 M EtOOH). The t 1/ 2 values for POX inactivation were 27, 36 and 34 s with no fatty acid, AA and linoleate, respectively. Thus, COX fatty acid substrates have a relatively modest effect (ⱕ33%) on the rate of POX inactivation. Clearly, this contrasts with what is observed with COX inactivation where the inactivation rates differ by more than twofold with linoleate versus AA (Table 2). When POX inactivation is measured in the presence of AA, PGG 2 is generated within seconds (via COX activity), and this PGG 2 becomes a substrate for the POX along with the added EtOOH. The fact that POX inactivation appears to be independent of the nature of the peroxide (e.g., PGG 2 generated when the POX reaction is measured in the 2
Richard J. Kulmacz, personal communication.
presence of AA does not substantially affect the rate of POX inactivation) is consistent with previous reports (13, 14). Additionally and as noted above, COX inhibitors such as flurbiprofen, which, like fatty acid substrates occupy the COX active site, have little effect on the rates of POX inactivation (14). Finally, it should be noted that when assayed measured under identical conditions in the presence of 110 M AA and 50 M EtOOH, the POX half-life (36 s) is more than twice the COX half life (16 s). Peroxidase inactivation with native versus Y385F oPGHS-1. It is often assumed that both COX and POX inactivation processes occur through a common COX Intermediate II (1, 13, 14). Intermediate II is an oxoferryl heme with an associated Tyr385 radical. Tyr385 is required for COX catalysis but not POX catalysis (9). We prepared and purified a hexahistidine tagged recombinant version of Y385F oPGHS-1 and compared the POX activity and rate of inactivation of this mutant to that of purified hexahistidine-tagged recombinant oPGHS-1 (Fig. 2). The POX specific activity of the purified Y385F mutant was 31% of native oPGHS-1. Both enzymes underwent suicide inactivation as depicted for Y385F oPGHS-1 in Fig. 2A. When equal amounts of native and Y385 oPGHS-1 POX activities were used to determine the rates of loss of POX activity, the rates were essentially the same (t 1/ 2 values of 13 and 15 s for native and
873
Vol. 289, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Y385F oPGHS-1, respectively; Fig. 2B). 3 These results establish that POX inactivation need not proceed via COX Intermediate II because Y385F oPGHS-1 cannot form this intermediate. DISCUSSION We have identified reactions conditions under which the COX activity of PGHS-1 undergoes complete inactivation but where about two thirds of the starting POX activity is retained. Thus, there are distinct suicide inactivation processes for the COX activities and the POX activities of PGHS-1. Consistent with this concept is the finding that the rates of COX inactivation vary by a factor of 3 for different fatty acid substrates whereas the POX inactivation rates are not greatly affected by the presence of fatty acids. We also found that a Y385F oPGHS-1 mutant that cannot form COX Intermediate II undergoes POX suicide inactivation and at the same rate as native oPGHS-1. This latter result strongly suggests that the COX active site Tyr385 is not involved directly in POX suicide inactivation. In contrast, because COX inactivation requires catalytic turnover of the COX, we presume that Tyr385 is required for COX inactivation. A model that accounts for our present results and previous studies of PGHS catalysis and suicide inactivation is presented in Fig. 3. Kinetic and spectral evidence have established that POX inactivation proceeds from an intermediate with UV/VIS spectral characteristics of a POX Compound II-like species that, in turn, converts to a heme spectral intermediate dubbed Compound III (13, 14). A tyrosyl radical is present at the time of POX inactivation (13) and a tyrosyl radical is formed even when the Y385F oPGHS-1 mutant enzyme is incubated with peroxides (25, 26). Therefore, it seems likely that the Compound II-like intermediate of POX inactivation is a COX Intermediate II-like species in that it has an oxyferryl heme and an associated protein tyrosyl radical, but that the tyrosyl radical is localized to a residue other than Tyr385. In Fig. 3 we have depicted this intermediate as Intermediate IIA; as indicated, the identity of the tyrosine residue bearing the radical remains to be determined. COX suicide inactivation, like POX inactivation is a first order process kinetically (15, 27) and has a certain probability of occurring as the enzyme traverses the COX cycle during oxygenation of a fatty acid substrate. That is, inactivation apparently starts with one of the intermediates in COX catalysis. Of the five unique intermediates depicted in Fig. 3, two involve organic substrate radicals and three involve protein Tyr385
radicals. It is not known which of the five intermediates is the starting point for inactivation, but it seems most likely that it is either the fatty acyl radical/ Tyr385 complex, the fatty acyl hydroperoxide/Tyr385 complex or the complex of product with Intermediate II. It seems less likely that Intermediate II itself or the complex in which fatty acid substrate is bound to Intermediate II is the most proximal starting point for inactivation. Our reason for discounting these latter possibilities is that in these cases relatively more COX inactivation than POX inactivation would be expected to occur when the enzyme is incubated with a peroxide either with or without a time-dependent COX inhibitor (e.g., flurbiprofen), and this is not the case (Table 1). However, it is not clear which of the three other, subsequent intermediates is the most immediate precursor of COX inactivation. The rate of COX inactivation is related to the structure of the substrate fatty acid (Table 2), but this dependence on substrate could either be because of the nature of the substrate radical or the influence of the structure of the substrate or oxygenated product on the reactivity of a Tyr385 radical-containing intermediate. Covalently coupling does occur when AA is incubated with PGHS-1 but only at one thirtieth of the rate of COX suicide inactivation (17, 18). Thus, if an intermediate fatty acid radical or fatty acid peroxyl radical does participate in suicide inactivation, it would be expected to do so as an initiator of a protein free radical reaction by abstracting a hydrogen atom from the side chain of a COX active site amino acid. This would yield a protein radical with the potential to react with another radical species, most likely an O 2 diradical. A product radical (e.g., PGG 2 radical) complexed with Tyr385 or the Tyr385 radical complexed with an oxygenated product (e.g., PGG 2) could also initiate this type of reaction. In summary, we have demonstrated that the processes of COX and POX inactivation are different, and we have identified experimental conditions for defining further the chemistries underlying these processes. ACKNOWLEDGMENT This work was supported in part by P01 GM57323 from the National Institutes of Health.
REFERENCES
3 The POX half-life values in these experiments were determined under different experimental conditions from those described earlier in the text.
874
1. Smith, W. L., DeWitt, D. L., and Garavito, R. M. (2000) Cyclooxygenases: Structural, cellular and molecular biology. Annu. Rev. Biochem. 69, 149 –182. 2. Marnett, L. J., Rowlinson, S. W., Goodwin, D. C., Kalgutkar, A. S., and Lanzo, C. A. (1999) Arachidonic acid oxygenation by COX-1 and COX-2. J. Biol. Chem. 274, 22903–22906. 3. Smith, W. L., and Langenbach, R. (2001) Why there are two cyclooxygenases? J. Clin. Invest. 107. 4. Kulmacz, R. J., and Wang, L. H. (1995) Comparison of hydroper-
Vol. 289, No. 4, 2001
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
oxide initiator requirements for the cyclooxygenase activities of prostaglandin H synthase-1 and -2. J. Biol. Chem. 270, 24019 – 24023. Lambeir, A. M., Markey, C. M., Dunford, H. B., and Marnett, L. J. (1985) Spectral properties of the higher oxidation states of prostaglandin H synthase. J. Biol. Chem. 260, 14894 –14896. Dietz, R., Nastainczyk, W., and Ruf, H. H. (1988) Higher oxidation states of prostaglandin H synthase. Rapid electronic spectroscopy detected two spectral intermediates during the peroxidase reaction with prostaglandin G2. Eur. J. Biochem. 171, 321–328. Karthein, R., Dietz, R., Nastainczyk, W., and Ruf, H. H. (1988) Higher oxidation states of prostaglandin H synthase. EPR study of a transient tyrosyl radical in the enzyme during the peroxidase reaction. Eur. J. Biochem. 171, 313–320. Tsai, A. L., Palmer, G., and Kulmacz, R. J. (1992) Prostaglandin H synthase: Kinetics of tyrosyl radical formation and of cyclooxygenase catalysis. J. Biol. Chem. 267, 17753–17759. Shimokawa, T., Kulmacz, R. J., DeWitt, D. L., and Smith, W. L. (1990) Tyrosine 385 of prostaglandin endoperoxide synthase is required for cyclooxygenase catalysis. J. Biol. Chem. 265, 20073– 20076. Hamberg, M., and Samuelsson, B. (1967) On the mechanism of the biosynthesis of prostaglandins E-1 and F-1-alpha. J. Biol. Chem. 242, 5336 –5343. Tsai, A., Kulmacz, R. J., and Palmer, G. (1995) Spectroscopic evidence for reaction of prostaglandin H synthase-1 tyrosyl radical with arachidonic acid. J. Biol. Chem. 270, 10503–10508. Schneider, C., and Brash, A. R. (2000) Stereospecificity of hydrogen abstraction in the conversion of arachidonic acid to 15RHETE by aspirin-treated cyclooxygenase-2: Implications for the alignment of substrate in the active site. J. Biol. Chem. 275, 4743– 4746. Wu, G., Wei, C., Kulmacz, R. J., Osawa, Y., and Tsai, A.-L. (1999) A mechanistic study of self-inactivation of the peroxidase activity in prostaglandin H synthase-1. J. Biol. Chem. 274, 9231– 9237. Wu, G., Vuletich, J. L., Kulmacz, R. J., Osawa, Y., and Tsai, A.-L. (2001) Peroxidase self-inactivation in prostaglandin H synthase-1 pretreated with cyclooxygenase inhibitors or substituted with mangano protoporphyrin IX. J. Biol. Chem. 276. Smith, W. L., and Lands, W. E. M. (1972) Oxygenation of polyunsaturated fatty acids during prostaglandin biosynthesis by sheep vesicular glands. Biochemistry 11, 3276 –3282. Kulmacz, R. J., Pendleton, R. B., and Lands, W. E. M. (1994)
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
875
Interaction between peroxidase and cyclooxygenase activities in prostaglandin-endoperoxide synthase. Interpretation of reaction kinetics. J. Biol. Chem. 269, 5527–5536. Kulmacz, R. J. (1987) Attachment of substrate metabolite to prostaglandin H synthase upon reaction with arachidonic acid. Biochem. Biophys. Res. Commun. 148, 539 –545. Lecomte, M., Lecocq, R., Dumont, J. E., and Boeynaems, J.-M. (1990) Covalent binding of arachidonic acid metabolites to human platelet proteins. J. Biol. Chem. 265, 5178 –5187. Thuresson, E. D., Lakkides, K. M., and Smith, W. L. (2000) Different catalytically competent arrangements of arachidonic acid within the cyclooxygenase active site of prostaglandin endoperoxide H synthase-1 lead to the formation of different oxygenated products. J. Biol. Chem. 275, 8501– 8507. Wei, C., Kulmacz, R. J., and Tsai, A. L. (1995) Comparison of branched-chain and tightly coupled reaction mechanisms for prostaglandin H synthase. Biochemistry 34, 8499 – 8512. Kulmacz, R. J. (1987) Prostaglandin G2 levels during reaction of prostaglandin H synthase with arachidonic acid. Prostaglandins 34, 225–240. Smith, T., Leipprandt, J., and DeWitt, D. L. (2000) Purification and characterization of the human recombinant histidine-tagged prostaglandin endoperoxide H synthases-1 and -2. Arch. Biochem. Biophys. 375, 195–2000. Smith, T., McCracken, J., Shin, Y. K., and DeWitt, D. (2000) Arachidonic acid and nonsteroidal anti-inflammatory drugs induce conformational changes in the human prostaglandin endoperoxide H2 synthase-2 (cyclooxygenase-2). J. Biol. Chem. 275, 40407– 40415. Goodwin, D. C., Gunther, M. R., Hsi, L. C., Crews, B. C., Eling, T. E., Mason, R. P., and Marnett, L. J. (1998) Nitric oxide trapping of tyrosyl radicals generated during prostaglandin endoperoxide synthase turnover. Detection of the radical derivative of tyrosine 385. J. Biol. Chem. 273, 8903– 8909. Tsai, A., Hsi, L. C., Kulmacz, R. J., Palmer, G., and Smith, W. L. (1994) Characterization of the tyrosyl radicals in ovine prostaglandin H synthase-1 by isotope replacement and site-directed mutagenesis. J. Biol. Chem. 269, 5085–5091. Hsi, L. C., Hoganson, C. W., Babcock, G. T., Garavito, R. M., and Smith, W. L. (1995) An examination of the source of the tyrosyl radical in ovine prostaglandin endoperoxide synthase-1. Biochem. Biophys. Res. Commun. 207, 652– 660. Hemler, M. E., and Lands, W. E. M. (1980) Evidence for a peroxide-initiated free radical mechanism of prostaglandin biosynthesis. J. Biol. Chem. 255, 6253– 6261.
Different Suicide Inactivation Processes for the Peroxidase and Cyclooxygenase Activities of Prostaglandin Endoperoxide H Synthase-1 Inseok Song,* Terry M. Ball,† and William L. Smith† ,1 †Department of Biochemistry and Molecular Biology, Michigan State University, East Lansing, Michigan; and *Department of Life Science, University of Seoul, Seoul, Korea
Received November 5, 2001
Prostaglandin endoperoxide H synthases (PGHSs)-1 and -2 have a cyclooxygenase (COX) activity involved in forming prostaglandin G 2 (PGG 2) from arachidonic acid and an associated peroxidase (POX) activity that reduces PGG 2 to PGH 2. Suicide inactivation processes are observed for both POX and COX reactions. Here we report COX reaction conditions for PGHS-1 under which complete COX inactivation occurs but with >60% retention of POX activity. The rates of POX inactivation were compared for native oPGHS-1 versus Y385F oPGHS-1, a mutant that cannot form the Tyr385 radical of COX Intermediate II; the rates were the same for both native and Y385F oPGHS-1. Our data indicate that a COX Intermediate II/acyl or product complex is the precursor in COX inactivation. However, another species, probably an Intermediate II-like species but with a radical centered on a tyrosine other than Tyr385, is the immediate precursor for POX inactivation. © 2001 Elsevier Science Key Words: arachidonic acid; linoleic acid; aspirin; hydroperoxide; flurbiprofen; Compound I; Compound II; peroxidase; cyclooxygenase.
The committed step in prostaglandin synthesis is catalyzed by prostaglandin endoperoxide H synthases (PGHS)-1 and -2 (1, 2). PGHSs catalyze the conversion of arachidonic acid (AA), two molecules of O 2 and two electrons to PGH 2. This involves two separate reactions—a cyclooxygenase (COX) reaction in which AA is oxygenated to yield PGG 2 and a peroxidase (POX) reaction in which the 15-hydroperoxyl group of Abbreviations used: PGHSs, prostaglandin endoperoxide H synthases; PG, prostaglandin; oPGHS-1, ovine PGHS-1; EtOOH, ethyl hydrogen peroxide; TMPD, N,N,N⬘,N⬘-tetramethylphenylenediamine; COX, cyclooxygenase; POX, peroxidase; AA, arachidonic acid. 1 To whom correspondence should be addressed at 513 Biochemistry Building, Department of Biochemistry and Molecular Biology, Michigan State University, East Lansing, MI 48824. Fax: (517) 3539334. E-mail: [email protected].
PGG 2 is reduced to PGH 2. PGHS-1 and -2 are closely related structurally and enzymologically. However, the expression of each isozyme is regulated differently and each isozyme subserves different physiological functions (reviewed in (1, 3)). The COX and POX activities of PGHSs are functionally interdependent. The mechanism of catalysis involves an initial peroxide-dependent oxidation of the heme group at the POX active site of PGHS (4) to yield an oxyferryl heme radical cation (Compound I) (5, 6). Compound I can undergo two sequential one electron reductions involving exogenous electron donors to produce Compound II and then native heme; or alternatively, Compound I can undergo an intramolecular one electron reduction by a protein tyrosine residue to form a COX Intermediate II that has an oxyferryl heme and an associated tyrosyl radical (7, 8). The tyrosyl radical of Intermediate II-Tyr385—abstracts the 13 proS hydrogen from AA yielding an AA radical (9 –12), the first definable chemical species unique to the COX reaction. Both the COX and POX activities of PGHSs undergo rapid first order suicide inactivation reactions. Suicide inactivation places an upper limit on the capacities of cells to synthesize prostaglandins and thromboxanes. The t 1/2 for the POX activity of PGHS in the presence of a peroxide is 2–20 s (13). POX inactivation has been proposed to involve a sequence of POX Compound I to COX Intermediate II to Intermediate III to suicide-inactivated enzyme (13, 14). The COX also exhibits a first order loss of activity when PGHS is incubated with fatty acid substrates (1, 15, 16). Although these processes have usually been measured under different sets of experimental conditions, the rates of POX and COX suicide inactivation are comparable (1, 13, 16). The chemical mechanism(s) of suicide inactivation is not known. Suicide inactivation of native enzyme does not appear to involve either heme oxidation (13) or covalent modification of the enzyme by an activated form of substrate fatty acid (17, 18). It seems likely that it is a
869
0006-291X/01 $35.00 © 2001 Elsevier Science All rights reserved.
Vol. 289, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Different rates of inactivation of cyclooxygenase and peroxidase activities of oPGHS-1. COX and POX activities were monitored as described under Experimental Procedures. (A) oPGHS-1 (210 units of COX activity) was reconstituted with hematin (5 M final concentration) and added to a 3-ml COX assay mixture containing 2 M EtOOH, 200 M TMPD, 5 M hematin, and 110 M AA. O 2 consumption was monitored at 23°C using an O 2 electrode. The initial rate was 65 nmol hydroperoxide (PGG 2 )/min. After 4 min when the rate had dropped to 0.9 nmol hydroperoxide/min, EtOOH (150 nmol) and then AA (330 nmol) were added. (B) Reactions were initiated as in A. After 4 min when, again, the rate had dropped to 0.9 nmol hydroperoxide/min, EtOOH (150 nmol) and then oPGHS-1 (210 units reconstituted with 5 M heme) were added. (C) In a parallel experiment O 2 consumption was monitored as in A and B. Four minutes after the initial addition of enzyme, 1 ml of the COX assay mixture was removed and placed in a spectrophotometer cuvette and POX activity (TMPD oxidation) was measured upon addition of EtOOH (final concentration 50 M); a control experiment was performed in the absence of AA. All of the experiments depicted in this figure were performed a minimum of three times with similar results.
free radical-mediated process leading to intramolecular protein cross-linking. POX and COX inactivation have many common characteristics and, quite reasonably, have been assumed to arise from the same COX Intermediate II species (13, 14). However, the rate of COX inactivation is slightly different for different substrates (15, 16) whereas the rate of POX suicide inactivation appears to be independent of the nature of the oxidizing peroxide substrate (13, 14). These differences led us to determine if we could distinguish between the POX and COX suicide inactivation processes. EXPERIMENTAL PROCEDURES COX and POX assay procedures. Rates of O 2 uptake were measured using a YSI Model 5300 oxygen electrode. Reaction mixtures contained 3 ml of 0.1 M TrisHCl, pH 8.0, containing 110 M fatty
acid substrate (Cayman Chemical Co., Ann Arbor, MI), 200 M N,N,N⬘,N⬘-tetramethylphenylenediamine (TMPD) and 1 M ethyl hydrogen peroxide (EtOOH) equilibrated at 23°C. Typically, 210 units of highly purified oPGHS-1 (19) (i.e., with specific activities of ⱖ10,000 units/mg protein) were equilibrated with 5 M hematin for 3 min and used to initiate the COX reactions. One unit of COX activity is defined as that amount of enzyme that will catalyze the uptake of 1 nmol of O 2 per min at 37°C in an assay mixture containing 0.1 M TrisHCl, pH 8.0, 100 M AA, 5 M hematin and 1 mM phenol. Velocity measurements were obtained at 1-s intervals using Dasy Lab Data Processing software (DASYTEC, Amherst, NH). Corrections for damping of electrode responses were made using the protocol of Wei et al. (20). To determine t 1/ 2 values, data of initial rates versus time were first transferred to an Excel worksheet and graphed using an XY scatter plot. This graph was fitted to an exponential plot and a resulting exponential equation was used for the t 1/ 2 determination; t 1/ 2 ⫽ 0.693/k where k is the exponential constant. POX assays were performed essentially as reported previously (19) using 1 ml of 0.1 M TrisHCl, pH 8.0, containing 5 M hematin, 200 M TMPD and 50 M EtOOH equilibrated at 23°C and an ⫽
870
Vol. 289, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS purified as described previously for native histidine-tagged oPGHS-1 (22). POX assays were performed as described above but with 100 M EtOOH.
TABLE 1
Effect of Peroxidase Inactivation on Cyclooxygenase Activity EtOOH (nmol)
Peroxidase activity remaining (%)
Cyclooxygenase activity remaining (%)
0 25 50 75 100
100 87 ⫾ 3.5 54 ⫾ 0.5 28 ⫾ 1.5 12 ⫾ 5.0
100 61 ⫾ 1.0 55 ⫾ 8.5 30 ⫾ 6.5 17 ⫾ 1.5
Note. Experiments were performed as described in the text. oPGHS-1 (70 units of COX activity) was preincubated for 5 min in a spectrophotometer cuvette containing 1 ml of 0.1 M TrisHCl, pH 8.0, 5 M hematin and 200 M TMPD. The indicated amounts of EtOOH were added, and the sample was incubated at 23° for 5 min (by which time no further peroxide reduction was occurring). The samples were then assayed for POX or COX activity. To measure POX activity 50 nmol EtOOH plus sufficient TMPD to restore the TMPD concentration to 200 mM was added; to measure COX activity 110 nmol of AA plus sufficient TMPD to restore the TMPD concentration to 200 mM was added. In each case the rate of TMPD oxidation was monitored at 611 nm. Values represent the mean ⫾ SD from two independent experiments.
12,200 L mol ⫺1 cm ⫺1 for TMPD oxidation (21); the reactions were initiated by adding purified oPGHS-1, histidine-tagged oPGHS-1 or histidine-tagged Y385F oPGHS-1. In some experiments involving COX assays, 1 ml of the COX assay mixture was removed and placed in a spectrophotometer cuvette and TMPD oxidation was measured upon addition of EtOOH (final concentration 50 M). One unit of POX activity is defined as that amount of activity that will catalyze the reduction of 1 nmol of EtOOH at 23°C in 1 min in a reaction mixture containing 1 ml of 0.1 M TrisHCl, pH 8.0, 5 M hematin, 100 M EtOOH and 4.5 mM guaiacol. Preparation, expression, and assay of histidine-tagged native and Y385F oPGHS-1. Site-directed mutagenesis of histidine-tagged oPGHS-1 in pFastbac (22) was performed using the QuickChange Site-Directed Mutagenesis Kit (Stratagene) according to the protocol of the manufacturer with the oligonucleotide primers 5⬘-GGAGTTCAACCAGCTGTTCCACTGGCACCCGCTCA-3⬘ (sense) and 5⬘-TGAGCGGGTGCCAGTGGAACAGCTGGTTGAACTCC-3⬘ (antisense), which code for amino acids 380 –390 (with one nucleotide on either side) of oPGHS-1. The DNA sequence of the mutant was confirmed by automated fluorescent DNA sequencing performed by the MSU DNA sequencing facility. Recombinant baculovirus particles were isolated using the Bac-to-Bac baculovirus expression system (Life Technologies, Inc.). The mutant enzyme was expressed and
POX inactivation experiments. Purified oPGHS-1 (70 units of COX activity) was preincubated for 5 min in a spectrophotometer cuvette containing 1 ml of 0.1 M TrisHCl, pH 8.0, 5 M hematin and 200 M TMPD and in some cases 5 M flurbiprofen. Varying amounts of EtOOH (0 –100 nmol) were added, and the sample was incubated at 23°C for 5 min to inactivate varying amounts of POX activity. The samples were then assayed for POX or COX activity. To measure POX activity 50 nmol EtOOH plus sufficient TMPD to restore the TMPD concentration to 200 mM was added; to measure COX activity 110 nmol of AA plus sufficient TMPD to restore the TMPD concentration to 200 mM was added. In each case the rate of TMPD oxidation was monitored at 611 nm.
RESULTS Cyclooxygenase inactivation without peroxidase inactivation. In the experiments depicted in Figs. 1A and 1B COX activity was assayed by monitoring O 2 uptake at 23°C following the addition of heme-reconstituted oPGHS-1 to an assay chamber containing AA and O 2 as COX substrates, TMPD as a POX reducing cosubstrate and a small amount of EtOOH to initiate COX catalysis. In each case, the initial COX rate was 65 nmol of hydroperoxide/min, and the rate dropped to less than 0.9 nmol of hydroperoxide/min within 4 min of adding enzyme to the assay mixture. At that time approximately 27 nmol of PGG 2 had been formed. The loss of cycloxygenase activity exhibited first order kinetics, and the t 1/ 2 was determined to be 16 s in both Figs. 1A and 1B. Addition of either EtOOH or AA after the COX rate had dropped to near zero did not cause a resumption of O 2 consumption (Fig. 1A); however, when fresh enzyme was added, O 2 uptake resumed, again at a rate of 65 nmol of hydroperoxide/min, and with a t 1/ 2 for COX inactivation of 17 s (Fig. 1B). Thus, under the conditions used in these experiments the COX activity of oPGHS-1 underwent suicide inactivation, and no appreciable COX activity remained after a 4-min incubation. In a parallel experiment O 2 consumption was monitored exactly as in Figs. 1A and 1B, but 4 min after the initial addition of enzyme, 1 ml of the assay mixture was quickly transferred to a spectrophotometer cu-
TABLE 2
Cyclooxygenase Suicide Inactivation of oPGHS⫺ with Different Fatty Acid Substrates Fatty acid
Initial cyclooxygenase rate (nmol hydroperoxide/min)
Cyclooxygenase half-life (s)
Total product (nmol of hydroperoxide)
AA (20:4n-6) Linoleic (18:2n-6) Eicosapentaenoic (20:5n-3) Eicosadienoic (20:2n-6)
65 ⫾ 4.7 26 ⫾ 0.96 16 ⫾ 0.87 42 ⫾ 3.5
16 ⫾ 0.85 45 ⫾ 12 15 ⫾ 0.07 17 ⫾ 3.3
28 ⫾ 3.3 31 ⫾ 11.3 9.5 ⫾ 4.7 16 ⫾ 8.8
Note. Initial rates of COX activity, COX half-lives (t 1/ 2 values), and amounts of product were determined as described in the legend to Fig. 1 and represent the mean ⫾ SD from a minimum of three independent measurements. 871
Vol. 289, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Experiments parallel to those depicted for AA in Fig. 1 were also performed using linoleic acid as the COX substrate and yielded similar results. Collectively, our data indicate that a major portion (ⱖ60%) of the POX activity of PGHS-1 can be retained following complete inactivation of the COX activity.
FIG. 2. Peroxidase inactivation with native and Y385F oPGHS-1. Histidine-tagged native and Y385F oPGHS-1 were solubilized and partially prepared as described under Experimental Procedures. Specific COX and POX activities of native and Y385F oPGHS-1 were measured as described under Experimental Procedures; note that the standardized conditions used to determine these specific activities are different from those used in the experiments depicted in Fig. 1. The specific COX and POX activities of native oPGHS-1 were 12,500 and 12,700 units/mg, respectively; these values indicate that the enzyme is about 40% pure. The specific POX activity of Y385F oPGHS-1 isolated using an identical protocol was 3970 units/mg. (A) Histidine-tagged Y385F oPGHS-1 (12 units of POX activity) was added to a POX reaction mixture containing EtOOH and guaiacol as the oxidizing and reducing cosubstrates, respectively, and activity was monitored by measuring the changes in absorbance at 426 nm as described in the text; EtOOH (100 nmol) and fresh enzyme (12 POX units) were added at the times indicated. (B) Native oPGHS-1 or Y385F oPGHS-1 (80 POX units) was added to standard POX assay mixtures and oxidation of guaiacol was monitored spectrophotometrically as described in the text.
vette, 50 nmol of EtOOH was added and POX activity was measured (Fig. 1C). An initial POX rate of 26 nmol hydroperoxide reduced/min was observed. Thus, POX activity remained following the loss of COX activity. A control experiment like those depicted in Figs. 1A and 1B was performed in the absence of AA, and a 1-ml aliquot of this sample was also assayed for POX activity (Fig. 1C). Comparison of the POX rates of the two samples (i.e., with versus without AA during the first 4 min) established that 68% of the POX activity remained after COX inactivation in the presence of AA.
Effect of peroxidase inactivation on cyclooxygenase activity. Different amounts of the POX activity of oPGHS-1 were inactivated by incubating the enzyme with 0 –100 M EtOOH for 5 min, and the amount of residual COX activity was determined (Table 1). There was a close correspondence between the percentage of POX activity inactivated and the amount of COX activity that was lost. This result was not unexpected because activation of the COX is directly dependent on the functioning of the POX, and only those enzyme molecules that have POX activity would be expected to have COX activity. However, these results do indicate that a relatively “pure” POX inactivation, with no appreciable additional, accompanying COX inactivation, occurs when oPGHS-1 is incubated with EtOOH and TMPD and without AA. A parallel experiment similar to that presented in Table 1 was performed in which the POX activity of oPGHS-1 that had been pretreated with 5 M flurbiprofen was partially inactivated with 25– 40 nmol of EtOOH. The presence of flurbiprofen in the sample causes complete, time dependent inhibition of COX activity under these conditions (23) but did not affect the degree of either POX or COX inactivation caused by incubating the enzyme with EtOOH and TMPD in the absence of flurbiprofen (data not shown). This is an important result because it indicates that simple occupancy of the COX site under conditions in which a Tyr385 radical is generated (24) does not lead to a loss of COX activity beyond that caused by the loss of POX activity. This is consistent with the concept that COX inactivation requires substrate oxygenation and not simply occupancy of the COX active site. Effects of different fatty acids on cyclooxygenase and peroxidase suicide inactivation. We analyzed the rates of COX suicide inactivation with several different fatty acid substrates under the same experimental conditions used for the experiments depicted in Fig. 1 with AA. The results are summarized in Table 2. It should be noted that the reaction mixtures were supplemented initially with 2 M EtOOH and that under these conditions neither the rate nor extent of the reactions was limited by the availability of initiator hydroperoxide. The rates of COX suicide inactivation with different fatty acids were substantially different from one another. For example, the rate of inactivation with linoleic acid was only one third of the rate observed with AA. Half-lives for COX inactivation with linoleate and AA were also measured over the first 30 s of the oxygenase reaction in the presence of different concentrations of EtOOH (2, 10, and 20 M). Within
872
Vol. 289, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. Cyclooxygenase and peroxidase catalysis and inactivation. A model depicting the cycles of COX and POX catalysis and suicide inactivation (1, 14). AA, arachidonic acid; PPIX, protoporphorin IX; ROOH, an alkyl hydroperoxide (e.g., PGG 2); ROH, an alkyl alcohol.
experimental error the initial rates were the same and the t 1/ 2 values were the same at each of these starting peroxide concentrations. These results indicate that the number of productive turnovers of which the COX is capable differs significantly among different substrates. Related and similar results have been reported by others in comparing AA with 5,8,11,14,17eicosapentaenoic acid (16). 2 To determine if there is an effect of COX fatty acid substrates on POX inactivation, POX activities were measured in the presence of no fatty acid, 110 M AA or 110 M linoleate under reaction conditions used to measure COX inactivation in Figs. 1A–1C (but with 50 M EtOOH instead of 2 M EtOOH). The t 1/ 2 values for POX inactivation were 27, 36 and 34 s with no fatty acid, AA and linoleate, respectively. Thus, COX fatty acid substrates have a relatively modest effect (ⱕ33%) on the rate of POX inactivation. Clearly, this contrasts with what is observed with COX inactivation where the inactivation rates differ by more than twofold with linoleate versus AA (Table 2). When POX inactivation is measured in the presence of AA, PGG 2 is generated within seconds (via COX activity), and this PGG 2 becomes a substrate for the POX along with the added EtOOH. The fact that POX inactivation appears to be independent of the nature of the peroxide (e.g., PGG 2 generated when the POX reaction is measured in the 2
Richard J. Kulmacz, personal communication.
presence of AA does not substantially affect the rate of POX inactivation) is consistent with previous reports (13, 14). Additionally and as noted above, COX inhibitors such as flurbiprofen, which, like fatty acid substrates occupy the COX active site, have little effect on the rates of POX inactivation (14). Finally, it should be noted that when assayed measured under identical conditions in the presence of 110 M AA and 50 M EtOOH, the POX half-life (36 s) is more than twice the COX half life (16 s). Peroxidase inactivation with native versus Y385F oPGHS-1. It is often assumed that both COX and POX inactivation processes occur through a common COX Intermediate II (1, 13, 14). Intermediate II is an oxoferryl heme with an associated Tyr385 radical. Tyr385 is required for COX catalysis but not POX catalysis (9). We prepared and purified a hexahistidine tagged recombinant version of Y385F oPGHS-1 and compared the POX activity and rate of inactivation of this mutant to that of purified hexahistidine-tagged recombinant oPGHS-1 (Fig. 2). The POX specific activity of the purified Y385F mutant was 31% of native oPGHS-1. Both enzymes underwent suicide inactivation as depicted for Y385F oPGHS-1 in Fig. 2A. When equal amounts of native and Y385 oPGHS-1 POX activities were used to determine the rates of loss of POX activity, the rates were essentially the same (t 1/ 2 values of 13 and 15 s for native and
873
Vol. 289, No. 4, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Y385F oPGHS-1, respectively; Fig. 2B). 3 These results establish that POX inactivation need not proceed via COX Intermediate II because Y385F oPGHS-1 cannot form this intermediate. DISCUSSION We have identified reactions conditions under which the COX activity of PGHS-1 undergoes complete inactivation but where about two thirds of the starting POX activity is retained. Thus, there are distinct suicide inactivation processes for the COX activities and the POX activities of PGHS-1. Consistent with this concept is the finding that the rates of COX inactivation vary by a factor of 3 for different fatty acid substrates whereas the POX inactivation rates are not greatly affected by the presence of fatty acids. We also found that a Y385F oPGHS-1 mutant that cannot form COX Intermediate II undergoes POX suicide inactivation and at the same rate as native oPGHS-1. This latter result strongly suggests that the COX active site Tyr385 is not involved directly in POX suicide inactivation. In contrast, because COX inactivation requires catalytic turnover of the COX, we presume that Tyr385 is required for COX inactivation. A model that accounts for our present results and previous studies of PGHS catalysis and suicide inactivation is presented in Fig. 3. Kinetic and spectral evidence have established that POX inactivation proceeds from an intermediate with UV/VIS spectral characteristics of a POX Compound II-like species that, in turn, converts to a heme spectral intermediate dubbed Compound III (13, 14). A tyrosyl radical is present at the time of POX inactivation (13) and a tyrosyl radical is formed even when the Y385F oPGHS-1 mutant enzyme is incubated with peroxides (25, 26). Therefore, it seems likely that the Compound II-like intermediate of POX inactivation is a COX Intermediate II-like species in that it has an oxyferryl heme and an associated protein tyrosyl radical, but that the tyrosyl radical is localized to a residue other than Tyr385. In Fig. 3 we have depicted this intermediate as Intermediate IIA; as indicated, the identity of the tyrosine residue bearing the radical remains to be determined. COX suicide inactivation, like POX inactivation is a first order process kinetically (15, 27) and has a certain probability of occurring as the enzyme traverses the COX cycle during oxygenation of a fatty acid substrate. That is, inactivation apparently starts with one of the intermediates in COX catalysis. Of the five unique intermediates depicted in Fig. 3, two involve organic substrate radicals and three involve protein Tyr385
radicals. It is not known which of the five intermediates is the starting point for inactivation, but it seems most likely that it is either the fatty acyl radical/ Tyr385 complex, the fatty acyl hydroperoxide/Tyr385 complex or the complex of product with Intermediate II. It seems less likely that Intermediate II itself or the complex in which fatty acid substrate is bound to Intermediate II is the most proximal starting point for inactivation. Our reason for discounting these latter possibilities is that in these cases relatively more COX inactivation than POX inactivation would be expected to occur when the enzyme is incubated with a peroxide either with or without a time-dependent COX inhibitor (e.g., flurbiprofen), and this is not the case (Table 1). However, it is not clear which of the three other, subsequent intermediates is the most immediate precursor of COX inactivation. The rate of COX inactivation is related to the structure of the substrate fatty acid (Table 2), but this dependence on substrate could either be because of the nature of the substrate radical or the influence of the structure of the substrate or oxygenated product on the reactivity of a Tyr385 radical-containing intermediate. Covalently coupling does occur when AA is incubated with PGHS-1 but only at one thirtieth of the rate of COX suicide inactivation (17, 18). Thus, if an intermediate fatty acid radical or fatty acid peroxyl radical does participate in suicide inactivation, it would be expected to do so as an initiator of a protein free radical reaction by abstracting a hydrogen atom from the side chain of a COX active site amino acid. This would yield a protein radical with the potential to react with another radical species, most likely an O 2 diradical. A product radical (e.g., PGG 2 radical) complexed with Tyr385 or the Tyr385 radical complexed with an oxygenated product (e.g., PGG 2) could also initiate this type of reaction. In summary, we have demonstrated that the processes of COX and POX inactivation are different, and we have identified experimental conditions for defining further the chemistries underlying these processes. ACKNOWLEDGMENT This work was supported in part by P01 GM57323 from the National Institutes of Health.
REFERENCES
3 The POX half-life values in these experiments were determined under different experimental conditions from those described earlier in the text.
874
1. Smith, W. L., DeWitt, D. L., and Garavito, R. M. (2000) Cyclooxygenases: Structural, cellular and molecular biology. Annu. Rev. Biochem. 69, 149 –182. 2. Marnett, L. J., Rowlinson, S. W., Goodwin, D. C., Kalgutkar, A. S., and Lanzo, C. A. (1999) Arachidonic acid oxygenation by COX-1 and COX-2. J. Biol. Chem. 274, 22903–22906. 3. Smith, W. L., and Langenbach, R. (2001) Why there are two cyclooxygenases? J. Clin. Invest. 107. 4. Kulmacz, R. J., and Wang, L. H. (1995) Comparison of hydroper-
Vol. 289, No. 4, 2001
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
oxide initiator requirements for the cyclooxygenase activities of prostaglandin H synthase-1 and -2. J. Biol. Chem. 270, 24019 – 24023. Lambeir, A. M., Markey, C. M., Dunford, H. B., and Marnett, L. J. (1985) Spectral properties of the higher oxidation states of prostaglandin H synthase. J. Biol. Chem. 260, 14894 –14896. Dietz, R., Nastainczyk, W., and Ruf, H. H. (1988) Higher oxidation states of prostaglandin H synthase. Rapid electronic spectroscopy detected two spectral intermediates during the peroxidase reaction with prostaglandin G2. Eur. J. Biochem. 171, 321–328. Karthein, R., Dietz, R., Nastainczyk, W., and Ruf, H. H. (1988) Higher oxidation states of prostaglandin H synthase. EPR study of a transient tyrosyl radical in the enzyme during the peroxidase reaction. Eur. J. Biochem. 171, 313–320. Tsai, A. L., Palmer, G., and Kulmacz, R. J. (1992) Prostaglandin H synthase: Kinetics of tyrosyl radical formation and of cyclooxygenase catalysis. J. Biol. Chem. 267, 17753–17759. Shimokawa, T., Kulmacz, R. J., DeWitt, D. L., and Smith, W. L. (1990) Tyrosine 385 of prostaglandin endoperoxide synthase is required for cyclooxygenase catalysis. J. Biol. Chem. 265, 20073– 20076. Hamberg, M., and Samuelsson, B. (1967) On the mechanism of the biosynthesis of prostaglandins E-1 and F-1-alpha. J. Biol. Chem. 242, 5336 –5343. Tsai, A., Kulmacz, R. J., and Palmer, G. (1995) Spectroscopic evidence for reaction of prostaglandin H synthase-1 tyrosyl radical with arachidonic acid. J. Biol. Chem. 270, 10503–10508. Schneider, C., and Brash, A. R. (2000) Stereospecificity of hydrogen abstraction in the conversion of arachidonic acid to 15RHETE by aspirin-treated cyclooxygenase-2: Implications for the alignment of substrate in the active site. J. Biol. Chem. 275, 4743– 4746. Wu, G., Wei, C., Kulmacz, R. J., Osawa, Y., and Tsai, A.-L. (1999) A mechanistic study of self-inactivation of the peroxidase activity in prostaglandin H synthase-1. J. Biol. Chem. 274, 9231– 9237. Wu, G., Vuletich, J. L., Kulmacz, R. J., Osawa, Y., and Tsai, A.-L. (2001) Peroxidase self-inactivation in prostaglandin H synthase-1 pretreated with cyclooxygenase inhibitors or substituted with mangano protoporphyrin IX. J. Biol. Chem. 276. Smith, W. L., and Lands, W. E. M. (1972) Oxygenation of polyunsaturated fatty acids during prostaglandin biosynthesis by sheep vesicular glands. Biochemistry 11, 3276 –3282. Kulmacz, R. J., Pendleton, R. B., and Lands, W. E. M. (1994)
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
875
Interaction between peroxidase and cyclooxygenase activities in prostaglandin-endoperoxide synthase. Interpretation of reaction kinetics. J. Biol. Chem. 269, 5527–5536. Kulmacz, R. J. (1987) Attachment of substrate metabolite to prostaglandin H synthase upon reaction with arachidonic acid. Biochem. Biophys. Res. Commun. 148, 539 –545. Lecomte, M., Lecocq, R., Dumont, J. E., and Boeynaems, J.-M. (1990) Covalent binding of arachidonic acid metabolites to human platelet proteins. J. Biol. Chem. 265, 5178 –5187. Thuresson, E. D., Lakkides, K. M., and Smith, W. L. (2000) Different catalytically competent arrangements of arachidonic acid within the cyclooxygenase active site of prostaglandin endoperoxide H synthase-1 lead to the formation of different oxygenated products. J. Biol. Chem. 275, 8501– 8507. Wei, C., Kulmacz, R. J., and Tsai, A. L. (1995) Comparison of branched-chain and tightly coupled reaction mechanisms for prostaglandin H synthase. Biochemistry 34, 8499 – 8512. Kulmacz, R. J. (1987) Prostaglandin G2 levels during reaction of prostaglandin H synthase with arachidonic acid. Prostaglandins 34, 225–240. Smith, T., Leipprandt, J., and DeWitt, D. L. (2000) Purification and characterization of the human recombinant histidine-tagged prostaglandin endoperoxide H synthases-1 and -2. Arch. Biochem. Biophys. 375, 195–2000. Smith, T., McCracken, J., Shin, Y. K., and DeWitt, D. (2000) Arachidonic acid and nonsteroidal anti-inflammatory drugs induce conformational changes in the human prostaglandin endoperoxide H2 synthase-2 (cyclooxygenase-2). J. Biol. Chem. 275, 40407– 40415. Goodwin, D. C., Gunther, M. R., Hsi, L. C., Crews, B. C., Eling, T. E., Mason, R. P., and Marnett, L. J. (1998) Nitric oxide trapping of tyrosyl radicals generated during prostaglandin endoperoxide synthase turnover. Detection of the radical derivative of tyrosine 385. J. Biol. Chem. 273, 8903– 8909. Tsai, A., Hsi, L. C., Kulmacz, R. J., Palmer, G., and Smith, W. L. (1994) Characterization of the tyrosyl radicals in ovine prostaglandin H synthase-1 by isotope replacement and site-directed mutagenesis. J. Biol. Chem. 269, 5085–5091. Hsi, L. C., Hoganson, C. W., Babcock, G. T., Garavito, R. M., and Smith, W. L. (1995) An examination of the source of the tyrosyl radical in ovine prostaglandin endoperoxide synthase-1. Biochem. Biophys. Res. Commun. 207, 652– 660. Hemler, M. E., and Lands, W. E. M. (1980) Evidence for a peroxide-initiated free radical mechanism of prostaglandin biosynthesis. J. Biol. Chem. 255, 6253– 6261.