Search

The cyclooxygenase (COX) reaction of prostaglandin (PG) biosynthesis begins with the highly specific oxygenation of arachidonic acid in the 11R configuration and ends with a 15S oxygenation to form PGG2. To obtain new insights into the mechanisms of stereocontrol of oxygenation, we mutated active site residues of human COX-2 that have potential contacts with C-11 of the reacting substrate. Although the 11R oxygenation was not perturbed, changing Leu-384 (into Phe, Trp), Trp-387 (Phe, Tyr), Phe-518 (Ile, Trp, Tyr), and Gly-526 (Ala, Ser, Thr, Val) impaired or abrogated PGG2 synthesis, and typically 11R-HETE was the main product formed. The Gly-526 and Leu-384 mutants formed, in addition, three novel products identified by LC-MS, NMR, and circular dichroism as 8,9-11,12-diepoxy-13R-(or 15R)-hydro(pero)xy derivatives of arachidonic acid. Mechanistically, we propose these arise from a free radical intermediate in which a C-8 carbon radical displaces the 9,11-endoperoxide O-O bond to yield an 8,9-11,12-diepoxide that is finally oxygenated stereospecifically in the 13R or 15R configuration. Formation of these novel products signals an arrest in the normal course of prostaglandin synthesis just prior to closing of the 5-membered carbon ring, and points to a crucial role for Leu-384 and Gly-526 in the correct positioning of the reacting fatty acid intermediate. Some of the Gly-526 and Leu-384 mutants catalyzed both formation of PGG2 (with the normal 15S configuration) and the 13R- or 15R-oxygenated diepoxides. This result suggests that oxygenation specificity can be determined by the orientation of the reacting fatty acid radical and is not a predetermined outcome based solely on the structure of the cyclooxygenase active site.

Acetaminophen has antipyretic and analgesic properties yet differs from the nonsteroidal antiinflammatory drugs and inhibitors of prostaglandin H synthase (PGHS)-2 by exhibiting little effect on platelets or inflammation. We find parallel selectivity at a cellular level; acetaminophen inhibits PGHS activity with an IC(50) of 4.3 microM in interleukin (IL)-1 alpha-stimulated human umbilical vein endothelial cells, in contrast with an IC(50) of 1,870 microM for the platelet, with 2 microM arachidonic acid as substrate. This difference is not caused by isoform selectivity, because acetaminophen inhibits purified ovine PGHS-1 and murine recombinant PGHS-2 equally. We explored the hypothesis that this difference in cellular responsiveness results from antagonism of the reductant action of acetaminophen on the PGHSs by cellular peroxides. Increasing the peroxide product of the PGHS-cyclooxygenase, prostaglandin G(2) (PGG(2)), by elevating the concentration of either enzyme or substrate reverses the inhibitory action of acetaminophen, as does the addition of PGG(2) itself. 12-Hydroperoxyeicosatetraenoic acid (0.3 microM), a major product of the platelet, completely reverses the action of acetaminophen on PGHS-1. Inhibition of PGHS activity by acetaminophen in human umbilical vein endothelial cells is abrogated by t-butyl hydroperoxide. Together these findings support the hypothesis that the clinical action of acetaminophen is mediated by inhibition of PGHS activity, and that hydroperoxide concentration contributes to its cellular selectivity.

Prostaglandin H synthase (PGHS) is the primary enzyme responsible for the biosynthesis of prostaglandins and thromboxanes. Of the two isoenzymes of PGHS, PGHS-1 is constitutively expressed and PGHS-2 is inducible by mitogens or other inflammatory stimuli. Constitutive expression of PGHS-2 in neoplastic tissues has been implicated in carcinogenesis. Resveratrol, a lignan, was recently shown to be an anticarcinogen that selectively inhibits PGHS-1. In vitro experiments to resolve these seemingly paradoxical observations revealed that resveratrol is not only an inhibitor of PGHS-1 but also is an activator of PGHS-2. Resveratrol non-competitively inhibited PGHS-1 with a K1 of 26 +/- 2 microM but enhanced the PGHS-2 activity nearly twofold. Additionally, resveratrol did not serve as a reducing co-substrate for the peroxidase activities of either enzyme despite being an easily oxidizable phenolic compound. Resveratrol inhibited the peroxidase activity of PGHS-1 (IC50 = 15 microM) better than that of PGHS-2 (IC50 = > 200 microM). Inhibition of the perxidase activity but not the cyclooxygenase activity of PGHS-2 resulted in the production of PGG2 from arachidonic acid. A plausible relationship between these observation and the anticarcinogenic activity of resveratrol is discussed.

Prostaglandin H synthase (PGHS) is a heme protein that catalyzes both the cyclooxygenase and peroxidase reactions needed to produce prostaglandins G2 and H2 from arachidonic acid. Replacement of the heme group by mangano protoporphyrin IX largely preserves the cyclooxygenase activity, but lowers the steady-state peroxidase activity by 25-fold. Thus, mangano protoporphyrin IX serves as a useful tool to evaluate the function of the heme in PGHS. A detailed kinetic analysis of the peroxidase reaction using 15-hydroperoxyeicosatetraenoic acid (15-HPETE), EtOOH, and other peroxides as substrates has been carried out to compare the characteristics of PGHS reconstituted with mangano protoporphyrin IX (Mn-PGHS) to those of the native heme enzyme (Fe-PGHS). The rate constant describing the reaction of Mn-PGHS with 15-HPETE to form the oxidized, Mn(IV) intermediate with absorption at 420 nm, exhibits saturable behavior as the 15-HPETE concentration is raised from 10 to 400 microM. This is most likely due to the presence of a second, earlier intermediate between the resting enzyme and the Mn(IV) species. Measurements at high substrate concentrations permitted resolution of the absorbance spectra of the two oxidized Mn-PGHS intermediates. The spectrum of the initial intermediate, assigned to a Mn(V) species, had a line shape similar to that of the later intermediate, assigned to a Mn(IV) species, suggesting that a porphyrin pi-cation radical is not generated in the peroxidase reaction of Mn-PGHS. The rate constant estimated for the formation of the earlier intermediate with 15-HPETE is 1.0 x 10(6) M-1 s-1 (20 degrees C, pH 7.3). A rate constant of 400 +/- 100 s-1 was estimated for the second step in the reaction. Thus, Mn-PGHS reacts considerably more slowly than Fe-PGHS with 15-HPETE to form the first high-valent intermediate, but the two enzymes appear to follow a similar overall reaction mechanism for generation of oxidized intermediates. The difference in rate constants explains the observed lower steady-state peroxidase activity of Mn-PGHS compared with Fe-PGHS.

Prostaglandin H synthase catalyzes the formation of prostaglandin (PG) G2 from arachidonic acid (cyclooxygenase activity), and also the reduction of PGG2 to PGH2 (peroxidase activity). The ability of the pure synthase to accumulate the hydroperoxide, PGG2, under conditions allowing the concurrent function of both catalytic activities was investigated. The peroxidase velocity was continuously determined from the absorbance increases at 611 nm that accompanied oxidation of a peroxidase cosubstrate, N,N,N',N'-tetramethylphenylenediamine, and PGG2 concentrations were calculated from the peroxidase velocities and the peroxidase Vmax and Km values. Cyclooxygenase velocities were then calculated from the changes in PGG2. Parallel reactions monitored by the use of radiolabelled arachidonate or with a polarographic oxygen electrode were used to confirm the calculated PGG2 levels and the cyclooxygenase velocities. The concentration of PGG2 was found to follow a transient course as the reaction of the synthase progressed, rapidly rising to a maximum of 0.7 microM in the first 10 s, and then declining slowly, reaching 0.1 microM after 60 s. The maximal level of PGG2 achieved during the reaction was constant at about 0.7 microM with higher amounts of added cyclooxygenase capacity (0.3-0.6 microM PGG2/s) but was only about 0.4 microM when the added cyclooxygenase capacity was 0.1 microM PGG2/s. The peroxidase was found to lose only 30% of its activity after 90 s, a point where the cyclooxygenase was almost completely inactive. These results support the concept of a burst of catalytic action from the cyclooxygenase and a reactive, more sustained, catalytic action from the peroxidase during the reaction of the synthase with arachidonic acid.

The rate of acetylation of fatty acid cyclooxygenase (prostaglandin synthase, EC 1.14.99.1) by [acetyl-3H]-aspirin was measured in microsomes from human aortas and coronary arteries and intact and disrupted human platelets. We also measured the inhibition by aspirin of prostacyclin generation from exogenous arachidonic acid in shredded human aorta. Cyclooxygenase in human aorta and coronary artery microsomes is approximately 1/250th as sensitive to aspirin as enzyme in intact platelets, and 1/60th as sensitive to aspirin as enzyme measured in a platelet microsomal preparation. On the basis of the in vitro data presented, we predict that small oral doses of aspirin are sufficient to inhibit platelet prostaglandin production but are not sufficient to substantially affect aorta or coronary artery prostaglandin production.