Suppression of Leukotriene Formation in RBL-2H3 Cells That Overexpressed Phospholipid Hydroperoxide Glutathione Peroxidase*

The overexpression of phospholipid hydroperoxide glutathione peroxidase (PHGPx) by RBL-2H3 cells was used as the basis for an investigation of the effects of PHGPx on the formation of leukotrienes. The rates of production of leukotriene C4 (LTC4) and leukotriene B4(LTB4) in cells that overexpressed PHGPx were 8 times lower than those in a control line of cells. The reduction in rates of production of leukotrienes apparently resulted from the increase in the PHGPx activity since control rates of formation of leukotrienes could be achieved in PHGPx-overexpressing cells upon inhibition of PHGPx activity by diethyl malate. The conversion of radioactively labeled arachidonic acid to intermediates in the lipoxygenase pathway, such as 5- hydroxyeicosatetraenoic acid (5-HETE), LTC4, and LTB4, was strongly inhibited in PHGPx-overexpressing cells that had been prelabeled with [14C]arachidonic acid. PHGPx apparently inactivated the 5-lipoxygenase that catalyzed the conversion of arachidonic acid to 5-hydroperoxyeicosatetraenoic acid (5-HPETE) since 5-HPETE is a common precursor of 5-HETE, LTC4, and LTB4. The rates of formation of LTC4 and LTB4 in PHGPx-overexpressing cells returned to control rates upon the addition of a small amount of 12-HPETE. Flow cytometric analysis revealed that the rapid burst of formation of lipid hydroperoxides induced by A23187 was suppressed in PHGPx-overexpressing cells as compared with the control lines of cells. Subcellular fractionation analysis showed that the amount of PHGPx associated with nuclear fractions from PHGPx-overexpressing cells was 3.5 times higher than that from the control line of cells. These results indicate that PHGPx might be involved in inactivation of 5-lipoxygenase via reductions in levels of the fatty acid hydroperoxides that are required for the full activation of 5-lipoxygenase. Thus, in addition to its role as an antioxidant enzyme, PHGPx appears to have a novel function as a modulator of the production of leukotrienes.

The overexpression of phospholipid hydroperoxide glutathione peroxidase (PHGPx) by RBL-2H3 cells was used as the basis for an investigation of the effects of PHGPx on the formation of leukotrienes. The rates of production of leukotriene C 4 (LTC 4

) and leukotriene B 4 (LTB 4 ) in cells that overexpressed PHGPx were 8 times lower than those in a control line of cells. The reduction in rates of production of leukotrienes apparently resulted from the increase in the PHGPx activity since control rates of formation of leukotrienes could be achieved in PHGPx-overexpressing cells upon inhibition of PHGPx activity by diethyl malate. The conversion of radioactively labeled arachidonic acid to intermediates in the lipoxygenase pathway, such as 5hydroxyeicosatetraenoic acid (5-HETE), LTC 4 , and LTB 4 , was strongly inhibited in PHGPx-overexpressing cells that had been prelabeled with [ 14 C]arachidonic acid. PHGPx apparently inactivated the 5-lipoxygenase that catalyzed the conversion of arachidonic acid to 5-hydroperoxyeicosatetraenoic acid (5-HPETE) since 5-HPETE is a common precursor of 5-HETE, LTC 4 , and LTB 4 . The rates of formation of LTC 4 and LTB 4 in
PHGPx-overexpressing cells returned to control rates upon the addition of a small amount of 12-HPETE. Flow cytometric analysis revealed that the rapid burst of formation of lipid hydroperoxides induced by A23187 was suppressed in PHGPx-overexpressing cells as compared with the control lines of cells. Subcellular fractionation analysis showed that the amount of PHGPx associated with nuclear fractions from PHGPx-overexpressing cells was 3.5 times higher than that from the control line of cells. These results indicate that PHGPx might be involved in inactivation of 5-lipoxygenase via reductions in levels of the fatty acid hydroperoxides that are required for the full activation of 5-lipoxygenase. Thus, in addition to its role as an antioxidant enzyme, PHGPx appears to have a novel function as a modulator of the production of leukotrienes.
Leukotrienes are important mediators both in host defense mechanisms and in inflammatory disease states since they have potent effects on cell migration, muscle contraction, vascular permeability, and the release of lysosomal enzymes (1)(2)(3). Leukotrienes can be formed in mast cells, granulocytes, and monocytes/macrophages in response to extracellular stimulation. The biosynthesis of leukotrienes is initiated by the release of arachidonic acid from membrane phospholipids (4). The liberated arachidonic acid is oxidized to hydroperoxyeicosatetraenoic acid (5-HPETE) 1 and subsequent dehydration yields an unstable epoxide intermediate, leukotriene A 4 (LTA 4 ). These sequential reactions are catalyzed by 5-lipoxygenase that is the primary enzymatic regulator of the biosynthesis of leukotrienes (5). LTA 4 is further converted to a variety of leukotrienes, such as LTB 4 and LTC 4 . The nuclear membrane is the site of leukotriene synthesis, and cytosolic phospholipase A 2 and 5-lipoxygenase are translocated to this site from the cytosol after the activation of cells (6,7). The activity of 5-lipoxygenase is regulated by several factors, such as Ca 2ϩ , ATP, and 5-lipoxygenase-activating protein (FLAP) (8 -12). There is evidence that fatty acid hydroperoxides might also activate 5-lipoxygenase (13,14). A requirement for fatty acid hydroperoxides in the activation of lipoxygenase suggests that the activity of 5-lipoxygenase might be regulated by its own product, namely 5-HPETE (15). Intracellular 5-HPETE can be reduced to 5-hydroxyeicosatetraenoic acid (5-HETE) by lipid peroxidases such as glutathione peroxidase (GPx). However, the lipid peroxidase responsible for the reduction of 5-HPETE remains to be defined. This reduction seems to be a key reaction in the control of the production of leukotrienes since it can determine the level of 5-HPETE, which is utilized as an activator of 5-lipoxygenase and as a precursor of LTA 4 .
GPx reduces cellular lipid hydroperoxides. Two types of selenium-dependent GPx are widely distributed in various tissues. Cytosolic glutathione peroxidase (cGPx, GPx1) is predominantly present in the cytosol of tissues, and it reduces fatty acid hydroperoxides and H 2 O 2 at the expense of glutathione (16). Phospholipid hydroperoxide glutathione peroxidase (PHGPx, GPx4) is localized in both the cytosol and the membrane fraction (17). PHGPx can directly reduce the phospholipid hydroperoxides, fatty acid hydroperoxides (18,19), and cholesterol hydroperoxides (20) that are produced in peroxidized membranes and oxidized lipoproteins. Although direct evidence for the involvement of GPx in the reduction of 5-HPETE, required for the activation of 5-lipoxygenase, is still lacking, it has been shown that impairment of GPx activity by selenium deficiency or glutathione depletion leads to enhanced activity of 5-lipoxygenase in rat polymorphonuclear leukocytes, rat basophile leukemia cells (RBL-1) (21), B lymphocytes (22,23) and human granulocytes (24,25). These results suggest that glutathione peroxidase might regulate the activity of 5-lipoxygenase in the cellular level. The question now arises as to whether cGPx or PHGPx might be responsible for the regulation of the formation of leukotrienes. We are interested in the possible role of PHGPx in the regulation of leukotriene synthesis since PHGPx is unique as an isozyme of GPx, being able to interact with the nuclear membrane in which the synthesis of leukotrienes occurs.
Stable transfectants that overexpress PHGPx provide useful model systems with which to attempt to clarify the ability of PHGPx to modulate leukotriene synthesis. We reported previously the cloning of a cDNA for rat PHGPx (26,27) and the establishment of PHGPx-overexpressing stable transfectants of rat basophile leukemia 2H3 (RBL-2H3) cells (28). PHGPxoverexpressing cells exhibit resistance to cell death that is due to oxidative damage. RBL-2H3 cells are a neoplastic cell line from rats. They are derived from basophiles, which are among the cells that have been used most extensively in studies of 5-lipoxygenase. RBL-2H3 cells produce LTC 4 and LTB 4 predominantly in response to stimulation by calcium ionophores (29,30).
In the present study, the PHGPx-overexpressing RBL-2H3 cells were used to examine the effects of PHGPx on the formation of leukotrienes. We showed that PHGPx significantly suppressed the production of leukotrienes via a reduction in the level of intracellular lipid hydroperoxides, which was due to suppression of the activity of 5-lipoxygenase.
Cell Cultures-RBL-2H3 cells that overexpressed PHGPx were generated by transfection with pSR␣-PHGPx and pSV2neo as described previously (28). In brief, fragments of a rat cDNA for PHGPx, namely the insert in pRPHGPx2 (761 base pairs) (26), were subcloned into pSR␣ as the expression vector (32). RBL-2H3 cells were harvested by trypsinization and were resuspended in the Dulbecco's modified essential medium that contained 20 mM HEPES, 2 mM glutamine, penicillin (100 units/ml), and streptomycin (100 units/ml) at a concentration of 4 ϫ 10 7 cells/ml. The suspension of cells (0.25 ml) was transferred to an electroporation cuvette (0.4-cm gap; Bio-Rad) with a total of 20 g of linearized DNA, which consisted of 18 g of SR␣-PHGPx and 2 g of pSV2neo (33). The latter plasmid was used to confer resistance to G418 (Geneticin; Life Technologies, Inc.). A charge of 250 V at 500 microfarads was applied at room temperature with a Gene Pulser II (Bio-Rad), and after a 10-min recovery period, the culture of the cells was reinitiated. Selection with G418 (1 mg/ml) was initiated after 24 h, and cells were subsequently exposed to 1 mg/ml G418 for 2 weeks. Individual G418-resistant colonies were isolated with cloning cylinders. G418resistant cells were analyzed by immunoprecipitation with antibodies against PHGPx (anti-PHGPx) to determine the level of expression of PHGPx 10 days after transfection. Control lines of cells were prepared by transfection with pSV2neo and pSR␣ that did not contain an insert of PHGPx cDNA.
Immunoblot Analysis-Cell homogenates and nuclear fractions were fractionated by SDS-PAGE on 12.5% acrylamide gels and transferred to PVDF membrane filter (Millipore Co., Bedford, MA) at 50 V for 150 min in 25 mM Tris, 192 mM glycine, 10% (w/v) methanol at 4°C in a protein transfer system (Bio-Rad), as described previously (35). Each PVDF membrane with blotted protein was blocked by incubation with 3% (w/v) skim milk in 10 mM Tris-HCl (pH 7.4) that contained 150 mM NaCl and 0.1% Tween 20 (TBS-T) for 1 h. The PVDF membrane was then incubated with antiserum against cPLA 2 , 5-lipoxygenase, and FLAP that had been diluted with TBS-T at an appropriate concentration for 2 h. Then the PVDF membrane was incubated for 1 h with horseradish peroxidase-conjugated goat antibodies against rabbit IgG or against mouse IgG (Zymed, South San Francisco, CA). The binding of antibodies to the antigen on the PVDF membrane was detected with an enhanced chemiluminescence Western blotting analysis system (Amersham Corp.).
Assays of Enzymatic Activities-RBL-2H3 cells (2 ϫ 10 7 cells) were sonicated in 1 ml of 10 mM Tris-HCl buffer (pH 7.4) that included 5 mg/ml leupeptin and 17 mg/ml PMSF. The homogenate was centrifuged at 10,000 ϫ g for 10 min at 4°C, and the supernatant was used for assays of enzymatic activity. PHGPx activity was measured as described previously (28). The reaction mixture contained, in a final volume of 0.8 ml, 0.1 M Tris-HCl (pH 7.4), 5 mM EDTA, 1.5 mM sodium azide, 3 mM glutathione, 0.25 mM NADPH, 1 unit of glutathione reductase, 0.1% Triton X-100, and 10 M phosphatidylcholine hydroperoxide (PCOOH), which had been prepared from egg yolk phosphatidylcholine as described previously (36). The reaction was started by the addition of PCOOH at 37°C after preincubation for 10 min. The rate of the reaction was measured by following the decrease in the absorbance at 340 nm over a 10-min period. The reaction was stopped by the addition of methanol, and the total lipids were extracted by the method of Bligh and Dyer (37). The lipid extract was dried under a stream of nitrogen gas, and the residue was dissolved in 0.05 ml of methanol. PCOOH without reduction by PHGPx was quantitated by the methylene blue method (38). The decrease in the amounts of PCOOH during the incubation was calculated by subtracting the amounts of non-reduced PCOOH from the original amount of PCOOH. cGPx activity was measured spectrophotometrically by monitoring the decrease in absorbance at 340 nm with 0.25 mM hydrogen peroxide as substrate (39). Glutathione S-transferase activity was measured spectrophotometrically by monitoring the formation of the conjugate of reduced glutathione and 1-chloro-2,4-dinitrobenzene at 340 nm (40). Catalase activity was determined as described by Clairborne (41) with 10 mM hydrogen peroxide as the substrate.
Liberation of [ 14 C]Arachidonic Acid from A23187-stimulated Cells-RBL-2H3 cells (4 ϫ 10 4 cells) were grown in 24-well plates for 3 days. Cells then were incubated with 0.1 Ci of [ 14 C]arachidonic acid (60 -100 Ci/mmol; NEN Life Science Products) for 24 h at 37°C. Labeled cells were washed three times with PBS and then preincubated with PBS that contained 1 mM CaCl 2 for 10 min. The cells that had been labeled with [ 14 C]arachidonic acid were stimulated with 5 M A23187 for 15 min. After centrifugation of samples to remove small amounts of dislodged cells (20 s at 16,000 ϫ g), supernatants were transferred to scintillation vials, and radioactivity was determined by scintillation counting (LS5000TD counter; Beckman Instruments, Fullerton, CA). Release of radiolabeled arachidonic acid was expressed as the ratio of the radioactivity of arachidonic acid released into the culture medium to the total radioactivity of cells multiplied by 100 to give a percentage. 4 from Exogenously Added LTA 4 -LTA 4 was prepared by the hydrolysis of the methyl ester of LTA 4 with 0.02 M lithium hydroxide at room temperature in darkness described previously (42,43). Cells were incubated in the presence of 1 M LTA 4 for 10 min. The amounts of LTC 4 and LTB 4 in the supernatant were quantitated by EIAs.

The Formation of LTC 4 and LTB
The Formation of LTC 4

, LTB 4 , and 5-HETE from Exogenous [ 14 C]
Arachidonic Acid-RBL-2H3 cells (5 ϫ 10 6 cells) were incubated with 0.1 Ci of [ 14 C]arachidonic acid for 24 h at 37°C. Labeled cells were stimulated with A23187 for 10 min in the presence of 1 mM CaCl 2 as described above. After the incubation, supernatants were obtained by centrifugation, and 3 ml of ethyl acetate (pH 3.0), supplemented with 0.1 ml of 0.09 M HCl, were added to each supernatant for the extraction of metabolites. Non-labeled LTC 4 (100 ng), LTB 4 (100 ng), and 5-HETE (200 ng), as carriers, and prostaglandin B 2 (PGB 2 ; 50 ng) as an internal standard, were added before the extraction. The supernatant was collected and injected into an HPLC system equipped with a reverse-phase column (LiChrosorb RP-18; 240 ϫ 0.4 mm inner diameter; Merck, Darmstadt, Germany) for the separation of LTC 4 , LTB 4 , and 5-HETE. The mobile phase was a mixture of acetonitrile, methanol, water, and acetic acid (350:150:250:1, v/v). The flow rate was 1.0 ml/min with monitoring of 280 nm for LTC 4 , LTB 4 , and PGB 2 and at 235 nm for 5-HETE. Each 0.5 ml of fraction was collected for measurement of radioactivity by scintillation counting.
Flow Cytometric Analysis of Intracellular Peroxides-To assess levels of intracellular peroxides, we performed flow cytometric analysis using an oxidation-sensitive fluorescent probe, 5,6-carboxy-2Ј,7Ј-dichlorofluorescein-diacetate (DCFH-DA). After preincubation with 12-HPETE and/or A23187, cells were washed with PBS and incubated with 2.5 M DCFH-DA in PBS for 30 min. DCFH-DA was deacylated to the nonfluorescent compound 2Ј,7Ј-dichlorofluorescein (DCFH) within the cells, and DCFH was oxidized to the fluorescent compound 2Ј,7Ј-dichlorofluorescein by a variety of peroxides (44). The fluorescent intensity of dichlorofluorescein in the cells was analyzed with a flow cytometer (EPICS R Elite Flow cytometer; Coulter, Hialeah, FL).
Subcellular Fractionation of Cells-[ 75 Se]Sodium selenite-labeled cells were fractionated by the previously described method with slight modification (45). Cells were labeled with 140 nCi/ml [ 75 Se]sodium selenite for 96 h, and confluent cells in 225-cm 2 culture flask were washed 3 times with phosphate-buffered saline and harvested with the aid of a rubber policeman in phosphate-buffered saline. The cell suspension was centrifuged at 800 ϫ g for 10 min at room temperature. The cell pellet was suspended in sucrose buffer (0.25 M sucrose, 1 mM EDTA, 3 mM imidazole, and 0.1% (v/v) ethanol (pH 7.2)) and centrifuged at 1,000 ϫ g for 10 min at 4°C. Pellets of cells were resuspended in the sucrose buffer at approximately 1 ϫ 10 8 cells/ml. Leupeptin, antipain, chymostatin, and pepstatin A were added at a final concentration of 10 g/ml each; PMSF was also added at a final concentration of 100 g/ml. Cells were homogenized with a Teflon/glass Potter-Elvehjem homogenizer. The nuclear fraction (the pellet) and a postnuclear fraction (the supernatant) were prepared by centrifugation at 2,700 ϫ g for 10 min. The nuclear fraction was suspended in 200 l of the sucrose buffer. Mitochondrial, microsomal, and cytosolic fractions from a postnuclear fraction were obtained by differential centrifugation according to the method of de Duve et al. (45). The distribution of each subcellular fraction was judged by standard enzymatic measurements, cytochrome c oxidase (mitochondrial marker), NADPH-cytochrome c reductase (microsomal marker), and lactate dehydrogenase (cytosolic marker) as reported previously (27,46). The distribution of histone H1 as nuclear marker in subcellular fractions was determined by the immunoblotting with anti-histone H1 polyclonal antibody (47). Antibody bound was detected by 125 I-protein A. The nuclear, mitochondrial, and microsomal fractions were solubilized with 0.4% Triton X-100 in phosphate-buffered saline for 2 h at 4°C and centrifuged at 100,000 ϫ g for 1 h at 4°C. The supernatant diluted with the same volume of PBS was immunoprecipitated with anti-PHGPx as described previously (28). Immunoprecipitated proteins were separated by SDS-PAGE (12.5% polyacrylamide) under non-reducing conditions. Gels were stained and dried for autoradiography. The distribution of PHGPx was calculated from scanning densitometry with a Bio-Imaging Analyzer (BAS2000; Fuji Film, Tokyo).
Quantitation of Proteins-Concentrations of protein were determined with the BCA protein assay reagent (Pierce) with bovine serum albumin as the standard.
Statistical Analysis-All data for which n Ն 3 are expressed as mean values Ϯ S.D.

Expression of Selenoproteins and the Activities of Antioxidant
Enzymes in PHGPx-overexpressing RBL-2H3 Cells-The distribution of selenoproteins was determined in RBL-2H3 cells that had been labeled with [ 75 Se]sodium selenite for 4 days. Fig. 1 shows the electrophoretic profiles of 75 Se-labeled proteins from control cells and PHGPx-overexpressing RBL-2H3 cells. Three major selenoproteins were detected, with estimated molecular masses of 57, 25, and 14 kDa respectively. The band of a 20-kDa protein that corresponded to PHGPx was quite faint in the control cells. The selenoproteins of 20 and 25 kDa were identified as cGPx and PHGPx, respectively, by immunoprecipitation with corresponding antibodies (data not shown). These results indicate that the level of PHGPx was quite low in the control lines of cells as compared with that of cGPx. PHGPx was extensively expressed in all lines of cells transfected with the cDNA for PHGPx, whereas the levels of expressions of the other three major selenoproteins in PHGPx-overexpressing cells were similar to those in the control lines of cells.
The specific activities of antioxidant enzymes were determined in the PHGPx-overexpressing cells and control lines of cells as shown in Table I. The activity of PHGPx in the PHGPxoverexpressing cells was three times that in the control lines of cells. No significant differences in the respective activities of cGPx, glutathione S-transferase, and catalase between the PHGPx-overexpressing cells and the control lines of cells were observed.
Production of Leukotrienes in the PHGPx-overexpressing RBL-2H3 Cells-Levels of leukotrienes were determined in the control and PHGPx-overexpressing cells after exposure to A23187 for 10 min by enzyme immunosorbent assays (Fig. 2). Stimulated RBL-2H3 cells predominantly produced leukotriene C 4 (LTC 4 ) and leukotriene B 4 (LTB 4 ), which reached levels of 2 and 0.2 ng/10 7 cells, respectively ( Fig. 2A). Overexpression of PHGPx in RBL2H3 cells led to a dramatic decrease in the production of leukotrienes (Fig. 2B). Levels of both leukotrienes produced in PHGPx-overexpressing cells were approximately 8 times lower than those in the control line of cells.
We examined the effects of diethyl malate (DEM) on the production of leukotrienes to estimate whether or not the suppression of leukotriene production had resulted from overexpression of PHGPx activity (Table II). DEM inhibits the activity of glutathione-dependent peroxidases, such as cGPx and PHGPx, by lowering the level of glutathione in the cells. DEM caused a decrease in the level of glutathione to about 2% that in non-treated RBL-2H3 cells under the present conditions (data not shown). The formation of leukotrienes was monitored in the control line of cells and in PHGPx-overexpressing cells that had been pretreated with DEM for 3 h. The activity of cGPx, which was extensively expressed in the control line of cells, was significantly suppressed by DEM under our assay conditions. However, the total amounts of LTC 4 and LTB 4 in the control line of cells were not altered in the case of DEM-treated control cells even though the activity of cGPx was significantly depressed. These results indicate that cGPx is not involved in the regulation of leukotriene production. In contrast to those in the control line of cells, rates of formation of LTC 4 and LTB 4 were dramatically enhanced in PHGPx-overexpressing cells after treatment with DEM. The production of leukotrienes recovered to 80% that in the control line of cells after inhibition of PHGPx activity by DEM. These results indicate that the induction of PHGPx activity led to a reduction in the rate of production of leukotrienes in RBL-2H3 cells.
Depletion of glutathione also inhibits LTC 4 synthetase, which catalyzes the synthesis of LTC 4 via the addition of glutathione to LTA 4 . The rate of formation of LTC 4 was significantly reduced in the control cells by treatment with DEM, and the accumulated LTA 4 was preferentially utilized for the syn-thesis of LTB 4 by LTA 4 hydrolase. The amount of LTB 4 in DEM-treated control cells was approximately 7 times that in non-treated control cells, in which the production of LTC 4 was markedly inhibited by the treatment with DEM (Table II).
Inhibition of 5-Lipoxygenase Activity in PHGPx-overexpressing Cells-The rates of production of individual metabolites in the 5-lipoxygenase pathway, such as free arachidonic acid, 5-HETE, LTC 4 , and LTB 4 , were determined in control line of cells and in PHGPx-overexpressing cells that had been prelabeled with [ 14 C]arachidonic acid to determine the step at which inhibition occurs in the 5-lipoxygenase pathway in PHGPxoverexpressing cells (Fig. 3 and 4). A23187 induced the activation of phospholipase A 2 and provoked the liberation of free arachidonic acid from membrane phospholipids (Fig. 3A). No significant difference in the release of arachidonic acid was observed between the control line of cells and the PHGPxoverexpressing cells. Levels of cytosolic phospholipase A 2 (cPLA 2 ) were determined by immunoblot analysis with anti-cPLA 2 (Fig. 3B). The levels of cPLA 2 in PHGPxoverexpressing cells were almost the same as that in the control line of cells. The rates of formation of 5-HETE, LTC 4 , and LTB 4 originated from [ 14 C]arachidonic acid were determined by reverse-phase HPLC (Fig. 4). Peaks I and II contained LTC 4 and LTB 4 , respectively. Peak III contained 5-HETE that had been produced by the reduction of 5-HPETE. The formation of three 5-lipoxygenase metabolites was facilitated in the control line of cells after the challenge with A23187. By contrast, the rate of formation of LTC 4 and LTB 4 from radioactive arachidonic acid was significantly reduced in the PHGPx-overexpressing cells. The rate of production of 5-HETE, formed by the reduction of 5-HPETE, was also reduced in the PHGPxoverexpressing cells, although the capacity for the reduction of 5-HPETE to 5-HETE was clearly elevated in the PHGPxoverexpressing cells. Immunoblot analysis with anti 5-lipoxygenase and 5-lipoxygenase activating protein (FLAP) revealed that levels of 5-lipoxygenase and FLAP were the same in PHGPx-overexpressing cells and in the control line of cells (Fig.  5B). These results indicated that activity of 5-lipoxygenase was inhibited in the PHGPx-overexpressing cells and, as a result, the production of 5-HPETE, the common precursor of LTC 4 , LTB 4 and 5-HETE, was significantly suppressed.
To determine the effects of PHGPx on the activities of LTC 4 synthase and LTA 4    Activity in PHGPx-overexpressing Cells-We determined the effects of hydroperoxides on the production of leukotrienes in an attempt to estimate whether the inhibition of the activity of 5-lipoxygenase was due to insufficient levels of hydroperoxides in PHGPx-overexpressing cells (Fig. 5). Levels of LTC 4 in A23187-stimulated cells were determined in the presence of various concentrations of 12-HPETE, which is not utilized as a substrate for the synthesis of LTC 4 . The rate of formation of LTC 4 in the PHGPx-overexpressing cells was significantly increased by the addition of 12-HPETE, and it returned to the rate in the control line of cells at a concentration of 12-HPETE of 5 ng (Fig. 5B). By contrast, 12-HPETE had no effect in the control line of cells (Fig. 5A). 12-HETE, a reduced form of 12-HPETE, failed to restore the production of LTC 4 in PHGPxoverexpressing cells.

. Production of leukotrienes C 4 and B 4 in control lines of cells and PHGPx-overexpressing cells. Control lines of cells (A) and
The levels of hydroperoxides were determined in the control line of cells and in the PHGPx-overexpressing cells by flow cytometric analysis after incorporation by cells of 2Ј,7-dichlorofluorescein diacetate (DCFH-DA), a hydroperoxide-sensitive fluorescent dye (Fig. 6). The basal fluorescence intensity of DCFH in PHGPx-overexpressing cells was lower than that observed in the control line of cells (Fig. 6A). A23187 provoked the production of intracellular hydroperoxides, and the intensities of peaks of fluorescence increased. The increase in fluorescence intensity in A23187-stimulated control cells was larger than that in similarly stimulated PHGPx-overexpressing cells (Fig. 6B). These results indicate that the levels of hydroperoxides in PHGPx-overexpressing cells were much lower than those in the control line of cells without stimulation and also after stimulation with A23187. The levels of hydroperoxide in PHGPx-overexpressing cells after the addition to the medium of 10 ng of 12-HPETE were the same as the basal level in the control line of cells (Fig. 6C). The intensity of fluorescence after stimulation with A23187 of PHGPx-overexpressing cells that had been pretreated with 12-HPETE was the same as that in the control line of cells after stimulation with A23187 (Fig. 6D). These results indicated that the increase in levels of intracellular hydroperoxides, caused by the addition of 12-HPETE, in PHGPx-overexpressing cells had dramatically increased the rate of formation of intracellular hydroperoxides.
Subcellular Localization of PHGPx in PHGPx-overexpressing Cells-To investigate the amounts of PHGPx in nuclear fraction of the control line of cells and PHGPx-overexpressing cells, cells prelabeled with [ 75 Se]sodium selenite for 4 days were fractionated into their organelle by the centrifugation, solubilized, and immunoprecipitated with anti-PHGPx. PHGPx was mainly enriched both in the mitochondrial and the nuclear fraction in the control line of cells (Fig. 7, A and C). In PHGPxoverexpressing cells, the amounts of PHGPx were significantly enhanced in nuclear, microsomal, and cytosolic fractions (Fig.  7, B and D). The amount of PHGPx associated with nuclear fractions from PHGPx-overexpressing cells was 3.5 times higher than that from the control line of cells (C and D).   trienes (29,30). The present study demonstrated that the overexpression of PHGPx in RBL2H3 cells caused a dramatic suppression of the production of leukotrienes in response to stimulation by A23187 (Fig. 2). The production of individual metabolites of 5-lipoxygenase was examined in PHGPx-overexpressing cells that had been preincubated with [ 14 C]arachidonic acid by reverse-phase HPLC to define the step in the 5-lipoxygenase pathway where inhibition occurs (Fig. 4A). The rates of production of radioactive 5-HETE, LTC 4 , and LTB 4 in PHGPx-overexpressing cells were much lower than in the control line of cells, whereas the rates of release of [ 14 C]arachidonic acid were unchanged (Fig. 3). No esterification of 5-HETE in phospholipid was also detected in PHGPx-overexpressing cells (data not shown). Suppression of the production of 5-HETE in PHGPx-overexpressing cells indicated that inhibi-tion of the synthesis of leukotrienes was not due to acceleration of the reduction of 5-HPETE to 5-HETE but was, rather, due to the inhibition of the production of 5-HPETE from arachidonic acid by 5-lipoxygenase. Levels of 5-lipoxygenase did not change in PHGPx-overexpressing cells (Fig. 4B), an indication that the activity of 5-lipoxygenase was suppressed in such cells.
The question remains as to how PHGPx down-regulates the activity of 5-lipoxygenase. The results obtained with PHGPxoverexpressing cells clearly indicate that PHGPx is intimately involved in the suppression of the activity of 5-lipoxygenase. Previous reports suggest that lipid hydroperoxides have dual effects on the activity of 5-lipoxygenase. In studies of enzyme kinetics, a small amount of HPETE activated 5-lipoxygenase (48), and a relatively high concentration (1.75 M) of HPETE caused inactivation (49). Although the precise mechanisms of the activation and inactivation of 5-lipoxygenase have not been elucidated, it seems likely that the state of the iron in 5-lipoxygenase might modulate the activity of the enzyme. When lipoxygenase is inactive, the enzyme contains a single non-heme iron that is in the ferrous oxidation state (Fe 2ϩ ). The active form of lipoxygenase contains iron in the ferric oxidation state (Fe 3ϩ ) (50,51). Thus, the conversion of iron to the ferric from the ferrous oxidation state is necessary for the activation of lipoxygenase (52)(53)(54)(55). Therefore, the activity of lipoxygenase might be regulated by a small amount of hydroperoxy lipids, acting as essential activators of the enzyme. In the present study, preincubation of PHGPx-overexpressing cells with 12-HPETE led to increased production of leukotrienes, whereas 12-HPETE did not affect the production of leukotrienes in the control line of cells (Fig. 5). Our results indicate that PHGPx is primarily responsible for the reduction of lipid hydroperoxides and thereby cause the down-regulation of 5-lipoxygenase in RBL-2H3 cells.
Flow cytometric analysis revealed that the basal level of intracellular hydroperoxides was significantly reduced in PHGPx-overexpressing cells (Fig. 6). A rapid and transient increase in levels of intracellular hydroperoxides induced by stimulation with A23187 was observed. The increase in fluorescence intensity was probably due to the production of intracellular lipid hydroperoxides, including 5-HPETE produced by 5-lipoxygenase. The increase in the levels of intracellular lipid hydroperoxides induced by A23187 was suppressed by overexpression of PHGPx. This effect by PHGPx was abolished by the addition of a small amount of 12-HPETE, which probably restored the production of LTC 4 . Thus, small variations in the levels of intracellular hydroperoxides can dramatically modulate the production of leukotrienes, and PHGPx seems to be able to regulate the activity of 5-lipoxygenase by modulating the levels of intracellular lipid hydroperoxides.
It is assumed that levels of lipid hydroperoxides are controlled by two kinds of intracellular GPx isozyme. Previous studies suggested that GPx might modulate the activity of 5-lipoxygenase because depletors of glutathione such as 1-chloro-2,4-dinitrobenzene or diamide induce the enhanced production of leukotrienes in human polymorphonuclear leukocytes (24) and B lymphocytes (22,23). To evaluate whether it is cGPx or PHGPx that has such an effect, we examined the effects of an inhibitor of GPx on the production of leukotrienes in the control line of cells and in PHGPx-overexpressing cells (Table II). The production of LTC 4 and LTB 4 in the control line of cells was not influenced by the treatment with DEM, even though the activity of cGPx, as the predominant GPx in the control line of cells, was inhibited by the depletion of glutathione. By contrast, the production of leukotrienes was significantly increased in the PHGPx-overexpressing cells when PHGPx activity was inhibited by DEM. These results indicate that PHGPx was involved in the inhibition of the activity of 5-lipoxygenase, whereas cGPx did not affect this activity.
PHGPx is located in the cytosol and membrane fraction, and cGPx is known to be present predominantly in the cytosol. Godeas et al. (56) demonstrated the specific localization of PHGPx in nuclear and mitochondrial fractions of rat testis. PHGPx has the highest ability to reduce lipid hydroperoxides in membranes of the four known isozymes of GPx (57). We showed subcellular localization of PHGPx in the control line of cells and PHGPx-overexpressing cells (Fig. 7). Control line of RBL2H3 cells contained the small amounts of PHGPx in the cytosol, whereas the large amounts of PHGPx are localized in the mitochondria and nucleus. In PHGPx-overexpressing cells, the amounts of PHGPx were significantly enhanced in nuclear, microsomal, and cytosolic fractions. The amount of PHGPx in nuclear fractions from PHGPx-overexpressing cells was 3.5 times higher than that from the control line of cells (Fig. 7). The synthesis of leukotrienes occurs at the nuclear membrane where cytosolic phospholipase A 2 and 5-lipoxygenase have been translocated from the cytosol (6,7). The expression of PHGPx in the nucleus might be critical for modulation of the trace amounts of lipid hydroperoxide required for the activation of 5-lipoxygenase. Overexpression of PHGPx did not affect the translocation of 5-lipoxygenase and cytosolic phospholipase A 2 from the cytosol to the nucleus in RBL-2H3 cells (data not FIG. 7. Subcellular localization of PHGPx in the control line of cells and PHGPx-overexpressing cells. The control line of cells (S2: A and C) and PHGPx-overexpressing cells (L28: B and D) were metabolically labeled with 75 Se (0.14 mCi/ml) for 4 days. The 75 Se-labeled cells were fractionated by differential centrifugation into nuclear fraction (N), mitochondrial fraction (M), microsomal fraction (Mc), and cytosolic fraction (S). The distributions of marker proteins in individual subfractionation were determined by enzymatic activity and immunoblot analysis (A and B). Lactate dehydrogenase (LDH), cytochrome c oxidase (Cyt Ox), and NADPH-cytochrome c reductase (Cyt Red) were measured as marker enzymes of cytosol, mitochondria, and microsome, respectively. Distribution of histone H1 (Histon) as marker protein of nucleus was calculated from scanning densitometry by a Bio-Imaging Analyzer as described in the text. Distribution of PHGPx was determined by immunoprecipitation analysis with anti-PHGPx (C and D). Relative radioactivity of PHGPx was calculated from scanning densitometry by a Bio-Imaging Analyzer.
shown). These results indicate that inactivation of 5-lipoxygenase was due to high level expression of PHGPx in the nucleus.
Metabolites of arachidonic acid seem to play an important physiological role in the signal transduction system that includes activation of NFB (58) and AP-1 (59), as well as in the regulation of cell survival and apoptosis (60,61). Furthermore, LTB 4 is not only a powerful chemotactic factor through LTB 4 receptor in inflammation (62) but is also an activator of the transcription factor PPAR␣ (the peroxisome proliferator-activated receptor ␣) (63). Thus, metabolites of lipoxygenase have a variety of biological activities in addition to their functions as chemical mediators. The present study suggests that PHGPx might be involved in the regulation of cellular functions and signal transduction through modulation of the production of leukotrienes.