Antioxidants inhibit interleukin-1-induced cyclooxygenase and nitric-oxide synthase expression in rat mesangial cells. Evidence for post-transcriptional regulation.

Glomerular mesangial cells produce reactive oxygen intermediates when stimulated by interleukin-1 (IL-1) or tumor necrosis factor. Recent observations suggest that reactive oxygen intermediates may play a role in IL-1 and tumor necrosis factor signaling and may up-regulate gene expression. We therefore evaluated the effects of antioxidants on IL-1β-induced cyclooxygenase-2 (Cox-2) and inducible nitric-oxide synthase (iNOS) expression in rat mesangial cells. The oxidant scavenger, pyrrolidine dithiocarbamate (PDTC), inhibited iNOS expression at the transcriptional level, since PDTC abolished iNOS mRNA accumulation. In contrast, PDTC inhibited Cox-2 expression at the post-transcriptional level, since PDTC did not affect IL-1β-induced Cox-2 mRNA levels but inhibited Cox-2 protein expression and prostaglandin E production. Another antioxidant, rotenone, which inhibits reactive oxygen intermediate production by inhibiting the mitochondrial electron transport system, did not inhibit IL-1β-induced iNOS and Cox-2 mRNA expression but inhibited iNOS and Cox-2 protein expression, suggesting a post-transcriptional target for the inhibition of iNOS and Cox-2 expression induced by IL-1β. These results suggest that not only transcriptional regulation but also post-transcriptional mechanisms are involved in redox-sensitive inhibition of cytokine induced Cox-2 and iNOS expression. These results suggest a novel approach for intervention in cytokine-mediated inflammatory processes.

Interleukin-1 (IL-1) 1 is a cytokine which mediates a variety of processes in host defense, such as inflammation and the cellular response to injury (1). During glomerular inflammation, cytokines from infiltrating macrophages and activated mesangial cells may act to sustain and promote glomerular damage. We have previously demonstrated that IL-1␤ induces cyclooxygenase-2 (Cox-2) and the inducible nitric-oxide synthase (iNOS) with increases in proinflammatory mediators, PGE 2 (2) and NO (3), in rat mesangial cells. The molecular signaling mechanisms by which IL-1␤ induces Cox-2 and iNOS includes transcriptional activation of these genes to produce increased levels of mRNA species which are "unstable." This mRNA is then translated into protein and degraded. These intracellular events are therefore potentially subject to regulation at the transcriptional or post-transcriptional level. Furthermore, the factors which control message stability and translational efficiency are not well understood.
Mesangial cells produce reactive oxygen intermediates (ROI) with stimulation by endotoxin and cytokines, including IL-1 and tumor necrosis factor (4). ROI are produced during various electron-transfer reactions. When generated in excess, ROI can damage cells by peroxidizing lipids and disrupting proteins and nucleic acids. However, ROI may exert signaling functions and regulate gene expression at moderate concentrations (5)(6)(7)(8)(9)(10). During glomerular inflammation, ROI from activated mesangial cells may act as signaling molecules. We have therefore evaluated the mechanisms by which some antioxidants can influence the expression of the proinflammatory genes, Cox-2 and iNOS when they are up-regulated by the cytokine IL-1␤.

EXPERIMENTAL PROCEDURES
Materials-Humam IL-1␤ (100 half-maximal units/ng) and restriction enzymes were purchased from Boehringer Mannheim. Murine cDNA probes ligated in BlueScript SK Ϫ for Cox-1 (pBS-Cox-1) and Cox-2 (pBS-Cox-2) were generous gifts of Dr. Karen Seibert, Monsanto Co. (St. Louis, MO). Pyrrolidine dithiocarbamate (PDTC) and rotenone were from Sigma. Polyclonal antibody against murine Cox-2 was from Cayman Chemical, Ann Arbor, MI. Polyclonal antibody against murine iNOS was a generous gift of Dr. Thomas Misko, Monsanto Co. Doublestranded oligonucleotides specific for the B sequence and the affinity purified rabbit polyclonal antibody against p50 NF-B protein were from Santa Cruz Biotechnology, Santa Cruz, CA.
Cell Culture and Treatment-Primary mesangial cell cultures were prepared from male Sprague-Dawley rats as described previously (2). Cells were grown in RPMI 1640 medium supplemented with 15% (v/v) heat-inactivated fetal calf serum, 0.3 IU/ml insulin, 100 units/ml penicillin, 100 g/ml streptomycin, 250 g/ml amphotericin B, and 15 mM HEPES. Cells were used at passages between 2 and 6.
The membrane was hybridized with radiolabeled cDNA probes at 42°C for 18 -24 h. The membrane was washed twice at room temperature for 5 min in 2 ϫ SSC (1 ϫ SSC is 150 mM NaCl and 15 mM sodium citrate, pH 7.0), twice at 60°C for 30 min in 2 ϫ SSC ϩ 0.1% sodium dodecyl sulfate, and autoradiographed with an intensifying screen at Ϫ80°C overnight. To control for variability in the loaded quantity of RNA, all membranes were probed with glyceraldehyde-3-phosphate dehydrogenase cDNA to determine the steady state levels of glyceraldehyde-3phosphate dehydrogenase gene-related sequences and used to normalize the mRNA for Cox-1, Cox-2, and iNOS.
Western Blot Analysis-Confluent cells grown in 25-cm 2 flasks were preincubated with antioxidants for 1 h in RPMI 1640 containing 5% (v/v) fetal calf serum then stimulated with IL-1␤ for indicated periods. Cells were washed twice with ice-cold phosphate-buffered saline, and lysed in 1 ml of ice-cold extraction buffer (50 mM Tris-HCl, pH 7.4, 1% Nonidet P-40, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 5 g/ml leupeptin, 5 g/ml pepstatin, 5 g/ml aprotinin). After 30 min of incubation at 4°C with rocking, cells were scraped and centrifuged at 10,000 ϫ g for 10 min at 4°C. The supernatant was collected and protein content was determined using a microbicinchoninic acid assay (Sigma). Cell lysate was mixed with Laemmli buffer and boiled for 5 min. Equal amounts of protein (30 -60 g/lane) was electrophoresed in 8 -10% SDS-polyacrylamide gels. Proteins were transferred to polyvinylidene difluoride membranes (Immobilon-P TM , Millipore). The membranes were saturated with 5% fat-free dry milk in Tris-buffered saline (50 mM Tris, pH 8.0, 150 mM NaCl) with 0.1% (v/v) Tween 20 (TBS-T) overnight at 4°C. The membranes were then incubated with purified polyclonal rabbit IgG antibody against murine Cox-2 or iNOS at 1:1,000 dilution in the above solution for 1 h at room temperature. Blots were washed four times (15 min each) in TBS-T. Blots were further incubated for 1 h at room temperature with the goat anti-rabbit IgG antibody coupled to horseradish peroxidase (Amersham) at 1:2,500 dilution in TBS-T, followed by four washes (15 min each) in TBS-T before visualization. The enhanced chemiluminescence (ECL) kit (Amersham) was used for detection and exposed to Hyperfilm MP (Amersham).
Measurement of Nitrite and PGE2-The stable metabolite of nitric oxide, nitrite, in the medium was measured by the Griess reaction as described previously (11). PGE 2 in the media was determined by stable isotope gas chromatography-mass spectrometry as described previously (2). Nitrite and PGE 2 production were corrected for protein as determined by the microbicinchoninic acid assay. Data were expressed as the mean Ϯ S.E.
Electrophoretic Mobility Shift Assay-Confluent mesangial cells in 75-cm 2 flasks were treated with IL-1␤ and/or pharmacological agents for the indicated periods. Cells were gently washed twice with ice-cold phosphate-buffered saline and scraped into 1 ml of ice-cold hypotonic lysis buffer (10 mM HEPES-KOH, pH 7.9, 10 mM KCl, 1.5 mM MgCl 2 , 0.5 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride, 5 g/ml aprotinin, 5 g/ml pepstatin, 5 g/ml leupeptin, 0.6% Nonidet P-40). The cells were allowed to swell on ice for 10 min, and vortexed vigorously for 10 s. The nuclei were collected by centrifugation at 14,000 ϫ g for 10 s. The nuclear pellet was resuspended in 100 l of high salt extraction buffer (20 mM HEPES-KOH, pH 7.9, 0.42 M NaCl, 1.2 mM MgCl 2 , 0.5 mM dithiothreitol, 0.2 mM EDTA, 25% glycerol, 0.5 mM phenylmethylsulfonyl fluoride, 5 g/ml aprotinin, 5 g/ml pepstatin, 5 g/ml leupeptin) and incubated for 20 min on ice with shaking. The nuclear extract was centrifuged for 5 min at 4°C, and the supernatant was aliquoted and stored at Ϫ80°C. Protein concentration was determined by microbicinchoninic acid assay (Sigma). A 22-mer double-stranded oligonucleotide probe containing NF-B sequences was end-labeled with [ 32 P]ATP. The B sequence used was, forward 5Ј-AGTTGAGGGGACTTTCCCAG-GC-3Ј and the complement 3Ј-TCAACTCCCCTGAAAGGGTCC-5Ј. 10 g of nuclear protein was incubated for 20 min on ice with radiolabeled oligonucleotide probes (2.5-5.0 ϫ 10 4 cpm) in a 25-l reaction buffer containing 2 g of poly(dI-dC), 10 mM Tris-HCl, pH 7.5, 100 mM NaCl, 1 mM EDTA, 1 mM dithiothreitol, 1 g/l bovine serum albumin, 10% (v/v) glycerol. Nucleoprotein-oligonucleotide complexes were resolved by electrophoresis in a 5% nondenaturing polyacrylamide gel (acrylamide/bisacrylamide at 40:1) in 0.3 ϫ TBE buffer at 150 V for 2 h at 4°C. The gel was dried and autoradiographed with an intensifying screen at Ϫ80°C for 1-5 h. The specificity of the DNA-protein complex was confirmed by competition with unlabeled B sequence and by the supershift of the retarded band by the addition of antibody against the p50 NF-B protein to the incubation mixture.
In Viro Translation-In vitro translation was carried out with a rabbit reticulate lysate kit (Express Translation Kit) purchased from Stratagene (La Jolla, CA) as per the manufacturer's instructions. Translation was carried out using [ 35 S]methionine (Ͼ1000 Ci/mM) and control RNA supplied with the kit generated a 15-kDa translation product. Translation was carried out in absence and presence of PDTC and rotenone.

PDTC Inhibits IL-1␤-induced iNOS Expression at the Transcriptional Level-
The effect of an oxidant scavenger, PDTC, on IL-1␤-induced NO production and iNOS expression was determined. PDTC inhibited nitrite production and iNOS protein expression in a dose-dependent manner (Fig. 1A and Fig.  2A). Since iNOS mRNA expression peaked at 12 h after IL-1␤ stimulation, the effect of PDTC on iNOS mRNA was determined at 12 h after IL-1␤ stimulation. 100 M PDTC abolished IL-1␤-induced iNOS mRNA expression (Fig. 2B). This data suggest that PDTC inhibits iNOS expression at the transcriptional level. Since PDTC has been reported to inhibit NF-B activation in some cell types (8) and the promoter for the iNOS gene contains the B binding consensus (12), the effect of PDTC on IL-1␤-induced NF-B activation was determined. Somewhat surprisingly PDTC failed to inhibit IL-1␤-induced NF-B activation in rat mesangial cells at concentrations that inhibit NO and iNOS mRNA expression (Fig. 3), suggesting that PDTC may block the activation of iNOS gene by some other FIG. 1. Effect of PDTC on nitrite (A) and PGE 2 production (B). Cells were pretreated with PDTC for 1 h and then stimulated with IL-1␤ (50 units/ml) for 24 h. The stable metabolite of NO, nitrite, in the medium was measured by the Griess reaction. PGE 2 in the medium was determined by gas chromatography-mass spectrometry. E, basal; å, IL-1.

FIG. 2. Effect of PDTC on iNOS protein (A) and iNOS mRNA expression (B).
Cells were pretreated with PDTC for 1 h and then stimulated with IL-1␤ (50 units/ml). Cells were harvested at 24 h for Western blot analysis (A) and at 12 h for Northern blot analysis (B), respectively. GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

mechanisms.
PDTC Inhibits IL-1␤-induced Cox-2 Expression at the Posttranscriptional Level-The effect of PDTC on IL-1␤-induced PGE 2 production and Cox expression was determined. PDTC inhibited IL-1␤-induced PGE 2 production in a dose-dependent manner (Fig. 1B). PDTC did not affect the catalytic activity of Cox from microsomal fractions of rabbit medullary collecting duct (data not shown), indicating that the effect of PDTC is not because of the inhibition of catalytic activity of the Cox enzyme. Since Cox-2 mRNA expression peaked at 3 h after IL-1␤ stimulation, the effect of PDTC on Cox-2 mRNA was determined at 3 h after IL-1␤ stimulation. PDTC did not affect IL-1␤-induced Cox-2 mRNA expression (Fig. 4A). In contrast, PDTC inhibited IL-1␤-induced Cox-2 protein expression in a dose-dependent manner (Fig. 4B). These data suggest that post-transcriptional events are involved in the inhibition of IL-1␤ induced Cox-2 expression by PDTC.
Effect of Rotenone on Cox-2 and iNOS Expression-The effect of rotenone, an inhibitor for mitochondrial electron transport system, was evaluated, since it has been suggested that ROI produced by the mitochondrial electron transport system may mediate tumor necrosis factor and IL-1 signaling (10). Rotenone inhibited IL-1␤-induced nitrite and PGE 2 production in a dose-dependent manner (Fig. 5). Rotenone did not affect the catalytic activity of Cox assessed in microsomal fractions of rabbit medullary collecting duct (data not shown). Rotenone did not inhibit IL-1␤-induced Cox-2 mRNA (Fig. 6A) and iNOS mRNA expression (Fig. 7A) but inhibited IL-1␤-induced Cox-2 ( Fig. 6B) and iNOS protein expression (Fig. 7B). This data suggests that rotenone inhibited the Cox-2 and iNOS expression at a post-transcriptional level.
Effect of PDTC and Rotenone as General Translational Inhibitors-To determine whether PDTC and rotenone were nonspecific inhibitors of translation, we performed in vitro translation experiments using a rabbit reticulolysate system in the presence and absence of different concentrations of drugs used in Northern and Western analyses above. blots were then performed for Cox-2 and iNOS. Fig. 9 shows the quantitative densitometry of a mean of two such experiments on the time course of protein expression for Cox-2, Fig. 9A, and iNOS, Fig. 9B. It clearly shows that these drugs simply reduce the magnitude of the increase with no effect on the time to peak expression. DISCUSSION In this study, we have demonstrated that two mechanistically distinct antioxidants, PDTC and rotenone, inhibit Cox-2 protein expression and PGE 2 production. Inhibition of Cox-2 is likely to be unrelated to alterations in its gene transcription, because neither compounds significantly reduced the level of the Cox-2 gene transcripts. Neither of these compounds directly inhibited the catalytic activity of Cox enzyme in crude microsomal preparations in vitro. These data suggest that these two antioxidants inhibit the Cox-2 expression at a posttranscriptional level.
Recent observations have demonstrated that proinflammatory cytokines, IL-1 and tumor necrosis factor, increase the production of ROI in mesangial cells (4). The mitochondrial electron transport system is one of the major sources for cellular ROI generation. Schultz-Oschoff et al. (10) showed that tumor necrosis factor ␣-induced cytotoxicity and NF-B activation were abolished by the mitochondrial electron transport system inhibitor, rotenone. Changes in redox status have been reported to modulate the activation of transcription factors, such as NF-B (7-10) and AP-1 (5,6). Thus it is proposed that these cytokines may mediate their effects in part via ROI. However, several lines of evidence suggest that the redox status of the cell can affect post-transcriptional events in the cell (13)(14)(15).
Redox status has been shown to regulate mRNA stability and modulate translation in a cell free system (13)(14)(15). Our data suggests that similar post-transcriptional events might be involved in the regulation of Cox-2 expression in vivo. Cox-2 and iNOS mRNA have "AUUUA" motifs in their 3Ј-untranslated regions (16,17). This AU-rich element has been considered to be a mRNA instability determinant. We have shown that IL-1␤ stabilizes the Cox-2 message by phosphorylation of cytosolic factors which bind to the AUUUA-rich 3Ј-untranslated region in rat mesangial cells (18). Furthermore, some of these AU-rich motif binding factors are known to be redoxsensitive (15). However, changes in stability of Cox-2 mRNA may not account for the effect of the antioxidants PDTC and rotenone, since the steady state levels of Cox-2 transcripts were not different in the control cells (IL-1␤ treated) and the antioxidant-treated (IL-1␤ plus antioxidant) cells. Inhibition of translation is more consistent with our observations, since the antioxidants did not inhibit Cox-2 mRNA expression but inhibited Cox-2 protein expression. Eukaryotic translation is regulated by many eukaryotic initiation factors and RNA binding proteins. One of the eukaryotic initiation factors, eukaryotic initiation factor-2, changes its function with redox status as well as phosphorylation (14). Redox status is also known to regulate the RNA-protein interaction of the iron-responsive element binding protein, which binds to the 5Ј-untranslated region of ferritin mRNA and 3Ј-untranslated region of the transferrin receptor mRNA. These cellular events control the translation of ferritin mRNA and stability of the transferrin receptor mRNA (13). Thus the change in redox status might regulate directly or indirectly some eukaryotic initiation factors and/or RNA binding proteins which regulate the translation and/or stability of Cox-2 mRNA. In the case of iNOS, PDTC and rotenone appeared to inhibit IL-1 induced iNOS expression by the different mechanisms. PDTC inhibited iNOS expression at the transcriptional level, since PDTC inhibited iNOS mRNA accumulation. However, we cannot exclude the possibility that PDTC also affects translational efficiency of the iNOS gene since the drug produced a reduction in iNOS mRNA. It has been demonstrated that PDTC inhibits NF-B activation in many cell types and that the promoter for the iNOS gene has a B binding consensus. However, consistent with the recent observation by Rovin (19), our data demonstrates that PDTC does not inhibit IL-1␤-induced NF-B activation in mesangial cells at concentrations which inhibit PGE 2 and NO formation. Thus PDTC appeared to inhibit the iNOS mRNA expression by some other unknown mechanisms. In contrast to PDTC, rotenone inhibited iNOS expression at a post-transcriptional level, since it did not inhibit iNOS mRNA expression but suppressed iNOS protein expression.
In summary, we have shown that two mechanistically different antioxidants, PDTC and rotenone, inhibit IL-1␤-induced Cox-2 and iNOS expression in rat mesangial cells. Our data suggests that a change in cellular redox status may influence gene expression at multiple levels which include transcription and post-transcriptional events. Neither PDTC nor rotenone appeared to be general inhibitors of translation ( Fig. 8) but their effects appeared to be restricted to a subset of messages which include Cox-2 and iNOS. Furthermore, the effects of these agents was simply to reduce the magnitude of the increase of protein expression with an effect on the time to peak expression. These results may suggest a potentially novel mechanistic approach to therapeutically intervene and downregulate the biologic effects of the proinflammatory genes, iNOS and Cox-2, in glomerular inflammation. The observation that the 3Ј-untranslated region of many "unstable" messages carry motifs that may regulate translational efficiency by reversible binding of cytosolic or nuclear factors raise the possibility that some antioxidants may influence, either directly or indirectly, these RNA-protein interactions.