INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

PROSTAGLANDINS (PGs) play important roles in normal body functions as well as in pathophysiological events. In inflammation, the production of PGs is associated with local and systemic symptoms of fever, pain, and edema (12, 14, 26, 40, 48). Therefore, the mechanisms that control PG biosynthesis during inflammation may dictate the extent and duration of inflammatory events.

PGs are synthesized from free arachidonic acid (AA) via one of two isoforms of cyclooxygenase (COX, also referred to as PGH₂-synthase), COX-1 and COX-2 (11, 22, 28, 35, 58). COX-1 is constitutively expressed in most tissues and is associated with the production of PGs that play important roles in normal body functions. COX-2 on the other hand, is transiently expressed in response to inflammatory challenge and is associated with pathological production of PGs (4, 19, 36). AA, which is the major substrate for COX, is stored in membrane phospholipids and released on stimulation by the action of phospholipases. Thus availability of free AA is a major limiting factor in PG biosynthesis (42, 47). De novo COX enzyme synthesis is another important mechanism controlling PG production, and COX-2 expression has been shown to be induced in response to various proinflammatory stimuli (20).

In addition, cellular cofactors may also regulate PG biosynthesis. COX activity is dependent on low levels of hydroperoxide (23, 27, 33) or peroxynitrite (25) for activation. Reduction of hydroperoxide tone by glutathione S-peroxidase (GSSPx) and glutathione (GSH) in vitro inhibited both COX-1 and -2 (24). Because COX-1 was ten times more sensitive to the inhibitory effect of GSH/GSSPx, it was suggested that this mechanism may be responsible for the shift in PG synthesis from COX-1 to COX-2 in inflammation (24). GSH may also have a direct regulatory effect on COX activity. It was recently demonstrated that the product profile of purified COX-1 and COX-2 shifted from PGE₂ to 12-hydroxyheptadecatrienoic acid (12-HHT) in the presence of millimolar concentrations of GSH (7). 12-HHT was probably formed as an alternate COX product from a cleavage of the 8,9 and 11,12 carbon-carbon bonds and the release of malondialdehyde (46). The sites of action for GSSPx and GSH on the COX metabolic pathway are illustrated in Fig. 1.

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Enzymatic conversion of arachidonic acid (AA) to prostaglandins (PGs) via cyclooxygenase (COX) and the sites of inhibition of this pathway by glutathione S-peroxidase (GSSPx) (23) and reduced glutathione (GSH) (7). 12-HHT, 12-hydroxyheptadecatrienoic acid.

Monosodium urate crystals are a naturally occurring proinflammatory agent associated with the manifestations of the rheumatic disorder, gout. This disease is characterized by occasional periods of severe pain and inflammatory swelling that, if untreated, resolve spontaneously within a few days (56). Although urate crystals are known to be the etiologic agent for gout, and their accumulation seems to be the major pathophysiological event leading to inflammatory episodes, additional factors influence the inflammatory response. Urate crystal deposition in joints may occur without associated inflammation (45), and inflammatory attacks may remit although urate crystals are still present in the joints (55). PGs appear to play a prominent role in the onset of gout attacks because nonsteroidal anti-inflammatory drugs, which inhibit PG formation, are considered the drugs of choice for the treatment of acute attacks (1). In a previous study we used a mouse model of urate crystal-induced peritonitis to investigate the role of PG in the onset and the duration of gouty inflammatory attacks. We found that, although urate crystals triggered an initial burst of eicosanoid products, at later times additional AA metabolism was attenuated in a fashion independent of substrate concentration (32).

In the present study, the mechanism for time-dependent attenuation of PG production in urate crystal-induced peritonitis was investigated. It is demonstrated that AA metabolism after urate crystal administration is shifted from PGs to 12-HHT and that this shift is associated with elevation of intracellular and extracellular GSH. Urate crystal administration triggered the expression of COX-2, indicating that the attenuation of PG synthesis was not correlated with enzyme levels but rather with the inhibition of COX-2 activity by GSH.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Materials. PGE₂, deuterated PGE₂ (PGE₂-d4), PGE₁, PGD₂, PGF₂, 6-keto-PGF₁, thromboxane B₂, 12-hydroxyeicosatetraenoic acid (12-HETE), 12-HHT, leukotriene B₄, and AA were purchased from Cayman Chemical (Ann Arbor, MI). Absolute ethanol and petroleum ether were obtained from Fisher (Pittsburgh, PA). Methyl formate was obtained from Eastman Kodak (Rochester, NY). Autosampler vials containing 100-µl glass inserts were purchased from Hewlett Packard (Palo Alto, CA). GSH, metaphosphoric acid, zymosan, L-buthionine-[S,R]-sulfoximine (BSO), and Trypan blue were obtained from Sigma Chemical (St. Louis, MO).

Solutions. Urate crystals were prepared by a modification of the method of McCarty and Faires (37). Briefly, 1.68 g of uric acid (Sigma U-2625) was dissolved in 400 ml of endotoxin-free water containing 4 ml of 2.5 N NaOH. This uric acid solution was sterilized by passage through a 0.22-µm filter and stirred overnight. The resulting urate crystals were washed and resuspended in Dulbecco's phosphate-buffered saline (PBS) at 5 mg/ml. The AA standard solution (200 µg/ml) was prepared by diluting 10 mg/ml AA stock solution (in ethanol) with H₂O and sonicating before use. The zymosan standard solution (1 mg/ml) was freshly prepared by suspending zymosan in heated PBS (60°C) and sonicating until no aggregates were observed.

Animals. Female Swiss-Webster mice (Charles River, Portage, MI), 20-27 g, were housed in the Monsanto Animal Resource Facility, five in a cage in a temperature-controlled room (24 ± 1°C) and fed a commercial diet with water ad libitum. The room was illuminated from 6:00 AM to 6:00 PM.

In vivo experiments. To address the effect of urate crystals on subsequent AA metabolism, mice were initially injected intraperitoneally with urate crystals (2.5 mg) or zymosan (0.5 mg). After 4 or 6 h, AA (100 µg) was injected intraperitoneally and the mice were euthanized after an additional 10 min. When the GSH synthesis inhibitor BSO was used, the drug was administered to mice in the drinking water (10 mM) 12 h before urate crystal administration and by intraperitoneal injection (100 mg/kg) 4 h before urate crystal administration. The peritoneal cavities were washed with 4 ml of cold PBS and massaged for 20 s, and 4 ml of lavage fluid per mouse was collected into polypropylene tubes on ice. Lavage samples were centrifuged 1,200 g for 10 min, and 1 ml of the supernatants was subjected to solid-phase extraction for eicosanoid analysis by electrospray and tandem mass spectrometry (ES-MS/MS). To assess the effect of urate crystal on the mRNA or protein levels of COX-1 and -2, mice were injected intraperitoneally with urate crystals (2.5 mg). After 1, 2, 4, or 6 h the mice were asphyxiated, lavages were collected as described above, and the cell pellets were kept at 80°C until analyzed.

In vitro experiments. Cells were harvested from untreated mice, washed three times in cold PBS, and incubated with GSH for 10 min at 37°C. Thereafter, 10 µg/ml AA was added and the cells were incubated with the substrate for an additional 10 min. The incubation was terminated by centrifugation at 1,200 g, 4°C, and the cell-free supernatants were extracted by solid phase and analyzed by ES-MS/MS. Each experiment was performed with three mice and repeated three times.

Solid-phase extraction. One nanogram of PGE₂-d4 was spiked into each sample of 1.0 ml of the biological preparations, and solid-phase extraction was performed according to a published procedure (49). In brief, the C₁₈ cartridges (Bond Elute C₁₈; Analytichem, Harbor City, CA) were preconditioned with 1.5 ml of methanol and 1.5 ml of H₂O. After the samples were applied, the cartridges were successively washed with 1.5 ml H₂O and 1.5 ml petroleum ether. The eicosanoids were eluted in 1.5 ml methyl formate. The methyl formate residue was evaporated under a dry N₂ stream, and the samples were resuspended in 100 µl ethanol. The samples then were transferred to autosampler vials and stored under nitrogen at 80°C until analysis.

Eicosanoid analysis by ES-MS/MS. Quantitation of different eicosanoid species and AA was performed by ES-MS/MS as described previously (31, 32). Briefly, eicosanoids were quantified by delivering 5-µl injections of the standards and the samples to the mass spectrometer using a Hewlett Packard 1090 autoinjector (Hewlett Packard, Palo Alto, CA). Injections were made every 2 min. The mass spectrometer was operated in the "multiple reaction monitoring" (MRM) mode. The ion spray interface was maintained at 4.5 kV. The argon target gas thickness was maintained at 2.3 × 10¹⁴ atoms/cm², and collision energies were adjusted to 30 eV. The MRM experiments were performed by setting the first quadrupole to pass the (M-H) of a selected eicosanoid while simultaneously setting the third quadrupole to pass a selected fragment ion of that eicosanoid. The mass resolution of the first quadrupole was set to unit mass resolution, and the mass resolution of the third quadrupole was adjusted so that the product ion widths were 2- to 3-µm wide to gain added sensitivity in the MRM experiments. To calculate the quantities of each eicosanoid species in the experimental samples, a standard mixture of commercial eicosanoids and AA was prepared and injected at elevated concentrations (1.56-100 ng/ml) before the experimental samples. The standard curve was calculated from the linear regression of the peak area of each species. PGE₂-d4 was added as an internal standard in these experiments to allow correction for differences in extraction efficiency.

Nuclease protection assay. mRNA was isolated from fresh lavage cells using a Totally RNA Kit (Ambion, Austin, TX) according to the manufacturer's instructions. COX expression was analyzed by nuclease protection assay using a ribonuclease protection assay II kit, (Ambion). Briefly, 5-µg samples of total RNA were hybridized with ³²P-labeled antisense RNA corresponding to mouse COX-1 and COX-2 as described previously (52) or commercial glyceraldehyde-3-phosphate dehydrogenase (GAPDH; Ambion). The RNA hybrids were digested with ribonucleases A and T1 according to manufacturer's instruction. All samples were analyzed in duplicate. Protected RNAs were separated by electrophoresis in 7.5 M urea/8% acrylamide sequencing gels. Gels were dried and exposed to Kodak X-OMAT film, in the presence of an intensifier screen for 4-5 days at 80°C. Relative expression levels were determined using a Phosphorimager (Molecular Dynamics, Sunnyvale CA).

Western blot. Peritoneal lavage cells were collected at various times after intraperitoneal administration of urate crystals and suspended in solubilization buffer composed of 0.5% deoxycholate and 1% Nonidet P-40 dissolved in PBS + 1× of Complete protease inhibitor cocktail (Boehringer Mannheim, Indianapolis, IN). After three cycles of five to eight pulses, duty cycle, 15%, the samples were centrifuged once to sediment residual crystals and the protein was determined by Bradford Assay (Bio-Rad, Richmond, CA). Fifty micrograms of sample protein was subjected to 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis using Tris-glycine buffer (Bio-Rad). The proteins were electroblotted onto nitrocellulose membranes using Mini-trans blot cell (Bio-Rad, Hercules, CA). The membranes were blocked for 1.5 h in TBS-T buffer (150 mM NaCl, 0.05% Tween 20, pH 8) + 5% nonfat dry milk. After the blocking solution was discarded, the membranes were then incubated with murine COX-1 or COX-2 antibodies (36) at 1:1,000 dilution in blocking solution, overnight, at 4°C. The membranes were then washed three times with TBS-T buffer and incubated with goat anti-rabbit immunoglobulin G conjugated to horseradish peroxidase (Boehringer Mannheim) at a dilution of 1:3,000 in blocking solution. After 45 min of incubation, the membranes were washed three times in TBS-T buffer and immunodetection was performed with the enhanced chemoluminescence (ECL)-Western blotting detection reagents (Amersham, Buckinghamshire, UK). The membranes were exposed to Hyperfilm-ECL film (Amersham).

GSH assay. Peritoneal lavages were collected and centrifuged at 1,500 revolutions/min for 10 min at 4°C. One-half milliliter from the supernatant was collected for extracellular GSH analysis. The cell pellets were resuspended in 0.5 ml of 0.15 M NH₄Cl and 17 mM tris(hydroxymethyl)aminomethane, pH 7.2, to lyse residual red blood cells. After three washes in PBS, the cells were counted and viability was assessed by Trypan blue exclusion. One million cells from each experiment were sonicated by eight cycles of 2-s pulse, treated with 5% metaphosphoric acid, and centrifuged at 2,000 g to sediment proteins. The GSH assay was performed using a GSH-400 colorimetric kit (Oxis, Portland, OR) according to the manufacturer instructions except that the total assay volume was 200 µl and the analysis was performed in a 96-well plate and read by an Emax plate reader (Molecular Devices, Menlo Park, CA) at 405 nm.

Statistics. All values in this study are given as means ± SE. Comparison between groups were made by one-way analysis of variance and t-test.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

In previous experiments we found that metabolism of AA to PGs and 12-HETE was attenuated for at least 4 h after the intraperitoneal administration of urate crystals (32). To further evaluate this effect, mice were injected first with either urate crystals or zymosan, followed by the injection of exogenous AA at different time intervals. Lavage fluid was collected 10 min after AA administration and analyzed for AA metabolites. A representative ES-MS/MS tracing for 6-keto-PGF₁ and 12-HHT is shown in Fig. 2A. Consistent with previous results, the metabolism of AA to 6-keto-PGF₁, PGE₂, and 12-HETE was reduced by 90, 65, and 80%, respectively, after urate crystal injection (Fig. 2B). In contrast, levels of 12-HHT were markedly elevated after AA administration in the presence of urate crystals (Fig. 2C). Pretreatment with zymosan resulted in a moderate reduction of 6-keto-PGF₁ and an elevation of PGE₂. Zymosan did not significantly affect 12-HETE and 12-HHT production.

View larger version (34K): [in this window] [in a new window]   View larger version (31K): [in this window] [in a new window]   View larger version (25K): [in this window] [in a new window]   Fig. 2.   Effect of intraperitoneal injection of urate crystals or zymosan (Zym) on subsequent AA metabolism. Mice (3 per group) were treated with either urate crystals or zymosan for indicated time and then injected intraperitoneally with AA (100 µg). After 10 min, lavages were collected and products were analyzed for eicosanoid species by electrospray and tandem mass spectrometry (ES-MS/MS) as described in MATERIALS AND METHODS. A: representative ES-MS/MS tracing of 6-keto-PGF1 (top) and 12-HHT (bottom). Each peak corresponds to 1 injection/experimental sample or standard. B: calculated values for 6-keto-PGF1, PGE2, and 12-hydroxyeicosatetraenoic acid (12-HETE) from 2 separate experiments. Levels are expressed as ng/lavage ± SE, n = 6. C: calculated value for 12-HHT from 2 separate experiments. Levels are expressed as means ± SE, n = 6. Significance of * P < 0.05 and ** P < 0.001 relative to AA-treated mice was determined by one-way analysis of variance.

The elevation in 12-HHT levels relative to PGs is consistent with the product profile of COX enzymes in the presence of reduced GSH (7). To determine whether administration of urate crystals alters peritoneal GSH levels, lavages were collected as a function of time after urate crystal injection and analyzed for extracellular and intracellular GSH. As shown in Fig. 3, GSH levels were elevated approximately fivefold in response to urate crystals in both cell-free lavages (Fig. 3A) and peritoneal cell lysates (Fig. 3B). Zymosan induced a twofold elevation of extracellular GSH but did not affect intracellular GSH levels. The intracellular GSH levels of 2.2 nmol/10⁶ cells measured in peritoneal macrophages from the untreated animals (Fig. 2B) are similar to the intracellular GSH levels of human monocytes (51). The calculated intracellular GSH concentration based on mean cell volume of 470 fl (29) was 4.5 mM for untreated cells and 20 mM after urate crystal administration. These values are within the physiological range of GSH (39, 41). To verify that these levels of GSH can alter AA metabolism in peritoneal macrophages, lavage cells were harvested from untreated mice and exposed to 10 µg/ml AA in the presence of exogenous GSH. As shown in Fig. 4A, 10 and 20 mM GSH significantly inhibited PGE₂ and 6-keto-PGF₁ while augmenting 12-HHT production. GSH also inhibited the production of 12-HETE (Fig. 4B).

View larger version (33K): [in this window] [in a new window]   View larger version (30K): [in this window] [in a new window]   Fig. 3.   Effect of intraperitoneal injection of urate crystals on extracellular (A) or intracellular (B) GSH levels in the peritoneal lavage cells. Lavages were harvested at indicated times after urate crystal administration, and GSH was measured in the supernatant or the cell pellet as described in MATERIALS AND METHODS. GSH are expressed as means ± SE, n = 6. Significance of * P < 0.05 and ** P < 0.001 relative to untreated mice was determined by one-way analysis of variance.

View larger version (28K): [in this window] [in a new window]   View larger version (29K): [in this window] [in a new window]   Fig. 4.   Effect of in vitro incubation of peritoneal macrophages with GSH on 6-keto-PGF1, PGE2, and 12-HETE (A) or 12-HHT (B) synthesis. Cells were harvested from untreated mice and incubated with GSH for 10 min. Thereafter, 10 µg/ml AA was added and the cells were incubated with the substrate for an additional 10 min. Cell-free supernatants were collected, and products were measured by ES-MS. Levels are expressed as means ± SE, n = 4.

To further establish the role of GSH in modulating PG production in this model, the -glutamylcysteine synthetase inhibitor BSO (16) was administered to mice in the drinking water (10 mM) and by intraperitoneal injection (100 mg/kg) 4 h before urate crystal administration. BSO treatment partially reduced the elevation of the extracellular GSH induced by urate crystals (Fig. 5C). In the same experiment, BSO treatment partially reduced 12-HHT production (Fig. 5B) and partially restored conversion of AA to PGs (Fig. 5A). These results indicate that the BSO treatment was not sufficient to fully deplete the GSH pools but are consistent with a role for GSH in attenuation of PG synthesis in this model.

View larger version (21K): [in this window] [in a new window]   View larger version (20K): [in this window] [in a new window]   View larger version (28K): [in this window] [in a new window]   Fig. 5.   Effect of intraperitoneal injection of urate crystals on the subsequent AA metabolism in absence or presence of L-buthionine-[S,R]-sulfoximine (BSO). Mice were treated with urate crystals or zymosan for 4 h and than injected with AA (10 µg) for 10 min. BSO (10 mM) was administered to the mice in the drinking water and by intraperitoneal injection (100 mg/kg) 4 h before urate crystal administration. Lavages were collected, and 6-keto-PGF1, PGE2, and 12-HETE (A) or 12-HHT (B) were measured by ES-MS. Levels are expressed as mean GSH, n = 6. Significance of * P < 0.05 and ** P < 0.001 between urate and BSO + urate mice was determined by one-way analysis of variance.

Urate crystals induce changes in the expression levels of COX-1 and COX-2. The elevation of intracellular GSH levels after urate crystals administration suggests one possible mechanism by which PG synthesis can be attenuated. However, it was still possible that urate crystals affected protein expression and, therefore, reduced COX protein levels following urate crystal injection. To determine whether urate crystals exert an effect on COX expression, total RNA was harvested from peritoneal lavage cells collected at various times after urate crystal injection, and the steady-state mRNA levels for COX-1 and COX-2 were measured by nuclease protection. As shown in Fig. 6A, COX-1 mRNA was readily detectable in lavage cells from untreated animals, but declined rapidly after urate crystal treatment. In contrast, COX-2 mRNA, which was undetectable in untreated control animals, was markedly induced by urate crystal administration, peaking at 2 h after injection (Fig. 6B). GAPDH mRNA levels were also measured to normalize between samples. GAPDH expression appeared to decrease initially at 1 h, but returned to control levels by 6 h (Fig. 6C). Western blot analysis of COX-1 and COX-2 protein using isoform-specific antibodies showed a corresponding decrease in COX-1 (Fig. 7, top) and increase in COX-2 (Fig. 7, bottom) protein levels, consistent with observed expression of mRNA. In a separate experiment, COX-1 and COX-2 expression was measured in peritoneal lavage cells harvested from untreated animals, which were incubated with urate crystals in vitro. In vitro treatment of a fixed population of cells with urate crystals failed to reduce COX-1 or GAPDH, but resulted in induction of COX-2 message (data not shown).

View larger version (28K): [in this window] [in a new window]   View larger version (29K): [in this window] [in a new window]   View larger version (33K): [in this window] [in a new window]   Fig. 6.   Effect of intraperitoneal injection of urate crystals on COX-1 (A), COX-2 (B), and glyceraldehyde-6-phosphate dehydrogenase (GAPDH) (C) mRNA levels in peritoneal lavage cells. Cells were harvested at indicated times after urate crystal administration, and mRNA was assessed by nuclease protection analysis. Bottom panels show gel autoradiogram, and top panels show relative changes in signal intensity. Data represent 1 of 3 separate experiments with similar results.

View larger version (42K): [in this window] [in a new window]   Fig. 7.   Effect of intraperitoneal injection of urate crystals on expression of COX-1 (top) and COX-2 protein (bottom) in peritoneal lavage cells. Cells were harvested at indicated times after urate crystal administration, and COX proteins were assessed by Western blot analysis. Data represent 1 of 2 separate experiments with similar results.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

The extent and duration of PG production that occurs under inflammatory conditions is dependent on the inflammatory stimulus and the target organ. Therefore, diverse mechanisms are likely to be involved in the regulation of PG biosynthesis in various inflammatory conditions. Some of the known factors that have been shown to influence PG production are phospholipase activity, the induced synthesis of COX-2, and the effects of hydroperoxide or peroxynitrate within the microenvironment of the cell. Recently, the effects of hydroperoxide tone (23, 24, 33) and GSH levels (7) on COX activity have been postulated on the basis of in vitro studies with recombinant COX enzymes. Our study provides evidence that GSH can attenuate PG biosynthesis in vivo. This conclusion is based on the following observations: 1) urate crystal administration caused a sustained shift in the eicosanoid profile from PG to 12-HHT (Fig. 2), 2) urate crystal administration elevated both the intracellular and extracellular GSH levels in peritoneal lavages (Fig. 3), and 3) physiological concentrations of GSH (39, 41) shifted AA metabolism of isolated peritoneal macrophages from PG to 12-HHT in a concentration-dependent manner (Fig. 4A).

As shown in Fig. 2B and previously reported (32), another proinflammatory agent, zymosan, did not inhibit the overall COX activity, although the product profile was changed, as 6-keto-PGF₁ levels were reduced and PGE₂ levels were elevated with time. Zymosan administration resulted in less than twofold elevation of the extracellular GSH pool but had insignificant effect on the intracellular pool (Fig. 3). These data further demonstrate that the inhibition of the COX pathway and GSH elevation is a unique response to urate crystals and is not a common phenomenon shared by other means of inducing inflammation.

GSH is synthesized in virtually all animals by the sequential action of -glutamylcysteine synthetase and GSH synthetase (38). An inhibitor of -glutamylcysteine synthetase, BSO (15, 16), has been used in vivo (38) to assess the protective role of GSH in different pathological states and has also been used experimentally in humans as an antitumor therapy (13, 44). In adult mice, BSO failed to fully deplete GSH levels, due to an alternative GSH recycling pathway that involves ascorbic acid (34, 38). This was probably the reason why BSO had only a limited effect on GSH levels and did not completely reverse the attenuation of PG biosynthesis induced by urate crystals (Fig. 5).

In addition to the effects urate crystals exert on PG biosynthesis, we have also observed a reduction of the in vivo formation of 12-HETE (Ref. 32 and Fig. 2A). Likewise, direct treatment with GSH inhibited the formation of 12-HETE in isolated peritoneal macrophages (Fig. 4B). Interestingly, GSH has also been shown to negatively regulate 12-lipoxygenase activity (17, 53). Inhibition of a second, independent enzymatic pathway in this model further supports the conclusion that, in response to urate crystals, the peritoneal concentration of GSH is elevated to levels that are capable of altering the enzymatic activity of enzymes that metabolize AA.

Intraperitoneal administration of urate crystals resulted in dramatic changes in COX-1 and COX-2 expression. The expression of COX-1 mRNA and protein in the lavage cells from untreated animals was reduced to virtually undetectable levels within 1 h of urate crystal administration and recovered only slightly after 6 h (Figs. 6A and 7, top). COX-2 on the other hand, was undetected in control cells but was dramatically induced after urate crystal administration. GAPDH mRNA levels (which were measured as a means of standardization of relative mRNA amounts) also appeared to decline in response to urate crystal administration but returned to control levels within 6 h. The rapid decline in both COX-1 and GAPDH mRNA that occurred in vivo, but was not reproduced on in vitro treatment of lavage cells, suggests that urate crystal treatment induces a switch in the peritoneal cell population and that infiltrating mononuclear cells express less COX-1 than the residential peritoneal macrophages. Such a change in peritoneal cell population from peritoneal macrophages to blood monocytes was reported earlier after intraperitoneal administration of Listeria monocytogenes (57). The effect of urate crystals on COX-1 expression was not further evaluated in this study.

The administration of urate crystals in this model resulted in the de novo synthesis of COX-2, which follows a time course similar to that seen in other inflammation models (36, 52). However, in other models the induction of COX-2 is closely paralleled by coordinate elevation in PG levels. Therefore, the urate crystal-induced inflammation in the mouse appears to be unique, in that the presence of GSH dramatically alters the activity of the induced COX-2 enzyme, resulting in the inhibition of PG production. In summary, our results indicate that elevated GSH levels inhibit PG production in this model by shifting COX product profile from PGs to 12-HHT and provide in vivo evidence for the role of GSH in the regulation PG biosynthesis.

Perspectives

GSH is the most abundant nonprotein thiol in mammalian cells and plays an important role in detoxification of xenobiotic compounds and protection from oxidative damage. Reduced GSH levels have been associated with various pathological states, including arthritis (30), lung disorders (41), cystic fibrosis (50), liver toxicity (3), cardiovascular diseases (54), and ischemia (21). It is generally accepted that the beneficial effects of GSH are due to protection from reactive oxygen radicals (6, 10). Our study provides in vivo evidence that GSH may also limit PG synthesis by a direct interaction with COX enzymes, potentially limiting PG production in various inflammatory states. It is possible that the reduced GSH levels observed in different pathophysiological disorders may correlate to elevated PGs and that these PGs further contribute to the pathological manifestation.