Cyclooxygenase (COX), known to exist in two isoforms, COX-1 and COX-2, is a key enzyme in prostaglandin synthesis and the target for most nonsteroidal anti-inflammatory drugs. In this study, we show that human T lymphocytes express the COX-2 isoenzyme. COX-2 mRNA and protein were induced in both Jurkat and purified T cells stimulated by TCR/CD3 or PMA activation. COX-2 mRNA was induced very early after activation and superinduced by protein synthesis inhibitors, whereas it was inhibited by the immunosuppressive drug cyclosporin A, identifying it as an early T cell activation gene. Interestingly, treatment with COX-2-specific inhibitors such as NS398 or Celecoxib severely diminished early and late events of T cell activation, including CD25 and CD71 cell surface expression, IL-2, TNF-α, and IFN-γ production and cell proliferation, but not the expression of CD69, an immediate early gene. COX-2 inhibitors also abolished induced transcription of reporter genes driven by IL-2 and TNF-α promoters. Moreover, induced transcription from NF-κB- and NF-AT-dependent enhancers was also inhibited. These results may have important implications in anti-inflammatory therapy and open a new field on COX-2-selective nonsteroidal anti-inflammatory drugs as modulators of the immune activation.

Activation through the TCR/CD3 complex triggers a variety of intracellular signals, resulting in a program of gene expression that culminates in cell proliferation and acquisition of effector functions (1, 2, 3). As a consequence, T cells produce a variety of cytokines, including IL-2, IL-3, IL-4, IL-5, IL-6, IL-10, TNF-α, IFN-γ, and GM-CSF. The coordinated production of these cytokines is crucial for the regulation of an immune response because they control the proliferation, differentiation, and function of lymphoid and nonlymphoid cells. On the other hand, activated T cells play an important role in the pathogenesis of inflammatory diseases, initiating and sustaining inflammation (4, 5). Proinflammatory cytokines secreted by T cells activate the inflammatory activity of macrophages and granulocytes. As a consequence, a new wave of inflammatory mediators, including cytokines such as TNF-α, IL-1β, eicosanoids, reactive oxygen intermediates, and nitric oxide, is produced that further augment inflammation (6).

Eicosanoids, including PGs, thromboxanes, and leukotrienes, are metabolites of the unsaturated fatty acid arachidonic acid (AA),3 synthesized and secreted by most human tissues and cell types (7, 8). Among the cells of the immune system, it is generally accepted that AA metabolites are mainly produced by accessory cells (9, 10). On the other hand, it is well known that eicosanoids play a key role in the regulation of both humoral and cell-mediated immunity, modulating cytokine and Ig production as well as T cell proliferation and activation (11, 12, 13, 14).

Prostaglandin H endoperoxide synthase or cyclooxygenase (COX) catalyzes the two-step conversion of AA to PGH2, the first reaction required for the biosynthesis of PGs and thromboxanes. At least two isoforms of the enzyme are expressed in mammalian tissues, COX-1 and COX-2. COX-1 is constitutively expressed in most mammalian tissues and is thought to be involved in homeostatic prostanoid biosynthesis (15, 16). In contrast, COX-2 is induced by various proinflammatory agents, including cytokines and mitogens (15, 17). COX-2 is thought to be the predominant isoform involved in the inflammatory response. Accordingly, the ability of nonsteroidal anti-inflammatory drugs (NSAIDs) to inhibit COX-2 activity may explain their therapeutic effects as anti-inflammatory drugs, whereas inhibition of COX-1 activity may account for some of their unwanted side effects (16, 18). Therefore, most of the new research on anti-inflammatory drugs has been aimed at targeting the COX-2-inducible production of PGs. Newly developed drugs that have high selectivity against COX-2, such as NS398, Celecoxib, or Meloxicam, have been proved to be potent anti-inflammatory compounds without causing gastric toxicity (19, 20, 21). However, there is growing evidence that some NSAIDs may have additional immunomodulatory properties not apparently related to the inhibition of PG synthesis. Thus, salicylates and aspirin, besides inhibiting COX-1, have also been described as inhibitors of the activation of the NF-κB as well as NF-κB-dependent gene expression (22). Tenidap and Tepoxalin, both cyclooxygenase and lipoxygenase inhibitors, suppress proliferation and IL-2 or IFN-γ expression in activated T cells (23, 24, 25). Besides, several NSAIDs, including Indomethacin, Ibuprofen, and Fenoprofen, have been recently shown to act as agonists of the transcription factor, peroxisome proliferator-activated receptor (PPAR)-γ, inhibiting PMA-induced cytokine synthesis in human peripheral blood monocytes (26, 27).

All of these observations led us to investigate the expression, regulation, and functional role of COX-1 and COX-2 during human T cell activation. Our results clearly show that upon TCR/CD3 triggering, COX-2 mRNA and protein expression were rapidly induced in T cells. COX-2-selective inhibitors such as NS398 and Celecoxib, but not Indomethacin, a nonselective NSAID, negatively regulated proliferation, cell surface expression of activation markers, and cytokine production by activated T cells. These effects could be attributable to down-regulation of the transcription of these genes through inhibition of transcription factors such as NF-κB and NF-AT. Thus, the results presented in this study provide the first evidence that COX-2 is transcriptionally up-regulated in T cells and that it behaves as early inducible gene involved in the T cell activation process. Our results suggest that COX-2-selective NSAIDs are immunosuppressive drugs and could have important applications in anti-inflammatory therapy.

The human Jurkat T leukemia cell line was grown in complete RPMI 1640 medium supplemented with 10% FCS. Purified human T lymphocytes were obtained from buffy coats of healthy donors by Ficoll-Hypaque (Pharmacia Fine Chemicals, Uppsala, Sweden) centrifugation (28). The PBL fraction was plated and adherent cells were removed. Purified T cells were obtained by passing the nonadherent population through a nylon fiber wool column, as previously described (28). The purity of the population, detected by flow cytometry, was always greater than 95% CD3+ cells.

Activation of human T lymphocytes through the TCR/CD3 complex and the CD28 receptor was conducted by adding the purified T cells to plates coated with anti-CD3 Ab (2 μg/ml), followed by subsequent addition of anti-CD28 Ab (250 ng/ml) (Immunokontact, Bioggio, Switzerland). Jurkat cells were stimulated by PMA (Sigma, St. Louis, MO), A23187 ionophore (Sigma), or immobilized anti-CD3 Ab (28), as indicated. Cycloheximide (CHX; 5 μg/ml) or actinomycin D (ActD; 5 μg/ml) was added to cells 45 min, and cyclosporin A (CsA; 100 ng/ml) (Sandoz, East Hanover, NJ), NS398 (10–100 μM) (Cayman Chemical, Ann Arbor, MI), Celecoxib (10–100 μM) (provided by Laboratorios Dr. Esteve, Barcelona, Spain), or Indomethacin (10–100 μM) (Sigma) 1 h prior to activation. None of the agents affect the viability of the cells at the concentrations used, as confirmed by the trypan blue dye exclusion test.

Total RNA was prepared from Jurkat or purified human T cells by the TRIzol reagent RNA protocol (Life Technologies, Paisley, U.K.). Total RNA (1 μg) was reverse transcribed into cDNA and used for PCR amplification with either human COX-1, COX-2, CD69, or GAPDH-specific primers by the RNA PCR Core Kit (Perkin-Elmer, Norwalk, CT), as previously described (29). The sequences of the sense and antisense primers for CD69 were: sense, 5′-GCACCATGTTCAGAACAAGC-3′; antisense, 5′-TGAAGGGTCCTTCCAAGTTC-3′. Briefly, the PCR was amplified by 20–35 repeat denaturation cycles at 94°C for 1 min, annealing at 60°C for 1 min, and extension at 72°C for 1 min. Amplified cDNAs were separated by agarose gel electrophoresis and bands visualized by ethidium bromide staining. For quantitation of amplified cDNA, dot-blot analysis was conducted with human COX-1, COX-2, CD69, and GAPDH oligonucleotide radioactive probes complementary to internal sequences of the cDNA amplified. Radioactivity in the COX bands or dots was quantified by a phosphorimager and normalized with respect to the GAPDH values from parallel samples.

Poly(A)+ mRNA was purified by affinity chromatography on oligo(dT) cellulose. A total of 2 μg of poly(A)+ mRNA was separated on formaldehyde agarose gels and blotted onto nylon membranes according to standard protocols (30). Human COX-1, COX-2, and GAPDH probes corresponding to the cDNA fragments amplified by RT-PCR were labeled by the random priming method (30). After hybridization and washes as previously described (31), the blots were then exposed to x-ray film for autoradiography.

Cells were disrupted in ice-cold lysis buffer (50 mM Tris-HCl (pH 8), 10 mM EDTA, 50 mM NaCl, 1% Nonidet P-40, 0.1% SDS, 10 μg/ml leupeptin, 10 μg/ml aprotinin, and 1 mM PMSF). Solubilized extracts (50 μg) were separated by SDS-PAGE on 10% polyacrylamide gel, and electrophoretically transferred to nitrocellulose filters. After blocking for 2 h with 5% nonfat dried milk in TBST (Tris-buffered saline containing 0.1% Tween-20), the membranes were incubated overnight at 4°C with monoclonal mouse anti-COX-2 (1:250) (Transduction Laboratories, Lexington, KY) in blocking buffer. The filters were washed and incubated for 1 h with horse anti-mouse IgG secondary Ab linked to horseradish peroxidase (Pierce, Rockford, IL) at 1/15,000 dilution, and the stained bands were visualized by the SuperSignal Substrate detection system (Pierce, Rockford, IL).

Purified T lymphocytes or Jurkat cells were stimulated with anti-CD3 plus anti-CD28 Abs or PMA plus Ion, respectively. After 24-h incubation, cells were washed twice with ice-cold PBS and disrupted by sonication in ice-cold 100 mM Tris-HCl, pH 7.8. COX-2 enzymatic activity was determined with 100 μg of protein from cell sonicates in 0.2 ml vol of assay buffer (100 mM Tris-HCl (pH 7.8), 5 mM tryptophan, and 5 mM reduced glutathione). Samples were incubated at 37°C for 15 min in the presence of an excess arachidonic acid (100 μM). The reaction was stopped by boiling, and samples were centrifuged at 10,000 × g for 10 min. Concentrations of PGs were measured by a prostaglandin screen colorimetric assay kit (Cayman Chemical).

Purified T cells (1 × 105 cells/100 μl RPMI plus 2% FCS) were grown in the absence or presence of different concentrations of the indicated compounds. Cells were incubated for 36 h in a 37°C and 5% CO2 atmosphere. Cell proliferation was estimated by measuring [3H]thymidine (New England Nuclear, Boston, MA) incorporation to DNA during the last 24 h of culture. Cells were harvested through glass fiber filter paper using a multiwell cell harvester (Scatron, Lier, Norway). The amount of radioactivity incorporated was estimated in a liquid scintillation beta counter. All of the experiments were conducted in triplicate cultures.

Cell surface expression in T cell cultures was evaluated by direct flow cytometry, as previously described (28). T cells were stimulated with anti-CD3 plus anti-CD28 Abs for 24 h in the presence or absence of NSAIDs, as indicated. Surface expression of the CD69, CD25, and CD71 Ags was analyzed by flow cytometry using FITC-conjugated Abs (Becton Dickinson, San Jose, CA) in a FACScan flow cytometer (Becton Dickinson). A minimum of 5000 cells per point was analyzed. The final percentage of positive cells was obtained by subtracting the values of negative control X63-FITC-conjugated Ab from those obtained with the specific Ab.

The concentration of IL-2, TNF-α, and IFN-γ was quantified using specific ELISA in supernatants of the same cell cultures used for cell surface analysis, harvested 24 h after activation. Commercially available kits were used according to manufacturer’s instructions (IL-2, R&D Systems, Minneapolis, MN; TNF-α, Bender MedSystems, Vienna, Austria; IFN-γ, Endogen, Cambridge, MA). Cytokine concentration was assayed in duplicate.

Transcriptional activity was measured using reporter gene assays in transiently transfected Jurkat cells. The pκBF-Luc (NF-κB-LUC) plasmid includes a trimer of the NF-κB motif of the H-2Kb gene upstream of the TK promoter (32). The plasmid TNF-α-LUC contains a region 1311 bp upstream from the transcriptional initiation site of human TNF-α promoter (33). The reporter constructs, NF-AT-LUC, containing three tandem copies of the NF-AT binding site fused to the IL-2 minimal promoter, and the IL-2-LUC plasmid, containing the region spanning from −326 to +45 of the human IL-2 promoter, have been described previously (34). Both were a generous gift of Dr. G. Crabtree (Stanford Medical School, Stanford, CA). The plasmid TK-LUC contains the herpes simplex I thymidine kinase promoter upstream the luciferase gene (35).

Jurkat cells were transfected with Lipofectin, as recommended by the manufacturer (Life Technologies). After transfection, cells were then treated with different stimuli as indicated for another 4–6 h, harvested by centrifugation, and lysed. Luciferase activity was measured in a luminometer following the instructions given in a luciferase kit assay (Promega, Madison, WI). Data are represented in relative luciferase units (RLU) or fold induction (observed experimental RLU/basal RLU in absence of any stimulus).

To analyze COX expression in T cells, RNA from Jurkat cells or purified T lymphocytes was subjected to Northern blot and RT-PCR analysis. As shown in Fig. 1 A, very low levels of COX-1 and COX-2 mRNAs were detected by RT-PCR in unstimulated Jurkat T cells. However, exposure of cells to the phorbol ester PMA plus the Ca2+ ionophore A23187 (Ion), a pharmacologic treatment known to mimic TCR/CD3 T cell activation (36), led to a substantial induction of COX-2 mRNA. PMA by itself was able to induce 2–3-fold COX-2 mRNA, whereas treatment with Ion alone did not increase COX-2 expression, suggesting the involvement of protein kinase C activation in COX-2 induction. A weak but significant induction of COX-2 mRNA expression was also observed after stimulation with immobilized anti-CD3. In contrast, COX-1 mRNA expression did not change significantly after any of these treatments.

FIGURE 1.

Expression of COX-1 and COX-2 mRNAs in T cells. Total RNA (1 μg) from Jurkat T cells (A) or purified human T lymphocytes (B) was analyzed by RT-PCR to measure the COX-1, COX-2, and GAPDH mRNA levels. An aliquot of the amplified DNA was separated on an agarose gel and stained with ethidium bromide for qualitative comparison. Northern blot of poly(A)+ from Jurkat (C) or T lymphocytes (D) hybridized with cDNA probes for COX-2 and GAPDH. Cells were cultured in absence of stimuli (CONT) or treated with PMA (15 ng/ml), Ion (1 μM), PMA plus Ion, immobilized anti-CD3 Ab (2 μg/ml), or combinations of PMA plus anti-CD3 and anti-CD3 plus anti-CD28 (250 ng/ml), as indicated. CsA (100 ng/ml) was added 1 h before stimulation.

FIGURE 1.

Expression of COX-1 and COX-2 mRNAs in T cells. Total RNA (1 μg) from Jurkat T cells (A) or purified human T lymphocytes (B) was analyzed by RT-PCR to measure the COX-1, COX-2, and GAPDH mRNA levels. An aliquot of the amplified DNA was separated on an agarose gel and stained with ethidium bromide for qualitative comparison. Northern blot of poly(A)+ from Jurkat (C) or T lymphocytes (D) hybridized with cDNA probes for COX-2 and GAPDH. Cells were cultured in absence of stimuli (CONT) or treated with PMA (15 ng/ml), Ion (1 μM), PMA plus Ion, immobilized anti-CD3 Ab (2 μg/ml), or combinations of PMA plus anti-CD3 and anti-CD3 plus anti-CD28 (250 ng/ml), as indicated. CsA (100 ng/ml) was added 1 h before stimulation.

Close modal

To determine whether the results observed with the Jurkat cell line were physiologically relevant and applicable to human peripheral T lymphocytes, we examined COX mRNA expression by RT-PCR in human T lymphocytes isolated from blood of healthy donors (Fig. 1,B). T lymphocytes were activated with pharmacological agents or with immobilized anti-CD3 alone or combined with soluble CD28 Abs. This latter treatment provides a costimulatory signal that induces a substantial increase in T cell activation (37). COX-1 mRNA was also observed in resting T lymphocytes, but its expression did not change appreciably after stimulation. In contrast, COX-2 mRNA expression increased after activation with PMA alone or in combination with Ion. Treatment with immobilized anti-CD3 Abs alone also produced a strong induction that was not further augmented by costimulation with PMA or anti-CD28 Abs. The lower COX-2 mRNA response to immobilized anti-CD3 in Jurkat cells compared with human T lymphocytes was probably due to the low level of CD3 expression in our Jurkat cloned line (data not shown). These results were confirmed by Northern blot analysis. Thus, COX-1 and COX-2 steady-state levels of mRNA expression both in Jurkat and T cells were below detection by Northern blotting, a less sensitive technique. In agreement with the RT-PCR data, PMA + Ion treatment increases several fold COX-2 mRNA expression both in Jurkat and T cells. Human T lymphocytes stimulated with anti-CD3 plus anti-CD28 increased as well COX-2 mRNA expression (Fig. 1, C and D). Together, these data indicate that T cell activation induces COX-2 expression without affecting COX-1 levels.

Treatment with CsA, a potent immunosuppressor that inhibits T cell activation and proliferation (38, 39), completely abolished COX-2 mRNA induction in human T lymphocytes stimulated with anti-CD3 plus anti-CD28 (Fig. 2,A) or PMA plus Ion (not shown) without affecting basal levels of COX-1 transcripts. Similar results were observed in Jurkat T cells, in which CsA inhibited the weak PMA induction, whereas PMA + Ion induction was reduced to PMA-induced levels (Fig. 2,B). Results obtained by Northern blot analysis were in agreement with those observed by RT-PCR (Fig. 1, C and D).

FIGURE 2.

Inhibition of COX-2 mRNA induction in activated T cells by CsA. Total RNA from purified human T lymphocytes (A) or Jurkat cells (B) treated with anti-CD3 (2 μg/ml) plus anti-CD28 (250 ng/ml) or PMA (15 ng/ml) plus Ion (1 μM), respectively, was used in the RT-PCR assay to determine the levels of COX-1, COX-2, and GAPDH mRNAs. CsA (100 ng/ml) was added, where indicated, 1 h before stimulation.

FIGURE 2.

Inhibition of COX-2 mRNA induction in activated T cells by CsA. Total RNA from purified human T lymphocytes (A) or Jurkat cells (B) treated with anti-CD3 (2 μg/ml) plus anti-CD28 (250 ng/ml) or PMA (15 ng/ml) plus Ion (1 μM), respectively, was used in the RT-PCR assay to determine the levels of COX-1, COX-2, and GAPDH mRNAs. CsA (100 ng/ml) was added, where indicated, 1 h before stimulation.

Close modal

Kinetic experiments using total RNA from purified human T cells treated with anti-CD3 plus anti-CD28 Abs were conducted to determine the time course of induction of COX-2 transcripts in these cells. To better quantify mRNA levels, aliquots from the PCR were obtained at different cycles of amplification and dot blotted, and the filters were hybridized with a radioactive oligonucleotide probe complementary to an internal sequence of the amplified cDNA fragments for COX-1, COX-2, CD69, and GAPDH. Dot-blot radioactivity was quantified and normalized to GAPDH values. COX-2 mRNA were detected as early as 1 h, reaching maximal levels 6 h after activation, and remained elevated for several hours. COX-1 expression did not change significantly, maintaining low levels of expression through the entire period measured (Fig. 3,A). CD69 mRNA expression, an immediate early gene in T cells, was induced with a similar kinetic. Experiments performed in Jurkat cells stimulated with PMA plus Ion showed a similar pattern of COX-2 mRNA induction. Again, COX-1 expression remained unaffected, maintaining the basal levels through the entire period studied (Fig. 3 B).

FIGURE 3.

Kinetics of COX-2 mRNA expression in T cells. Human T lymphocytes (A) or Jurkat cells (B) were treated with anti-CD3 (2 μg/ml) plus anti-CD28 (250 ng/ml) or PMA (15 ng/ml) plus A23187 (Ion) (1 μM), respectively, for the times indicated. RT-PCR was performed as described in Materials and Methods. Amplified cDNA was subjected to dot-blot analysis and hybridized with internal oligonucleotide probes for COX-1, COX-2, CD69, and GAPDH. Radioactivity was quantified with a phosphor imager and normalized considering GAPDH values. Results of the scanning of the upper panels are shown in the graphics below as the ratio of arbitrary densitometric units of COX-2 (○) or COX-1 (•) mRNA to GAPDH mRNA.

FIGURE 3.

Kinetics of COX-2 mRNA expression in T cells. Human T lymphocytes (A) or Jurkat cells (B) were treated with anti-CD3 (2 μg/ml) plus anti-CD28 (250 ng/ml) or PMA (15 ng/ml) plus A23187 (Ion) (1 μM), respectively, for the times indicated. RT-PCR was performed as described in Materials and Methods. Amplified cDNA was subjected to dot-blot analysis and hybridized with internal oligonucleotide probes for COX-1, COX-2, CD69, and GAPDH. Radioactivity was quantified with a phosphor imager and normalized considering GAPDH values. Results of the scanning of the upper panels are shown in the graphics below as the ratio of arbitrary densitometric units of COX-2 (○) or COX-1 (•) mRNA to GAPDH mRNA.

Close modal

To determine whether COX-2 induction was dependent on new protein or RNA synthesis, Jurkat T cells were treated with CHX, an inhibitor of translation, and ActD, an inhibitor of transcription. COX-2 mRNA expression induced by PMA plus Ion was completely suppressed by ActD (Fig. 4), indicating that the increase in mRNA levels occurs mainly at the transcriptional level, requiring new RNA synthesis. CHX stimulated the steady state levels of COX-2 mRNA in Jurkat cells. In addition, CHX was able to superinduce the stimulation of COX-2 transcription by PMA plus Ion. This pattern of expression, independent of the requirement for new protein synthesis, identifies COX-2 as a characteristic inducible early T cell activation gene (40, 41).

FIGURE 4.

Effects of CHX and ActD on COX-2 mRNA expression. Jurkat T cells were treated with CHX (5 μg/ml) or ActD (5 μg/ml) 45 min before stimulation with PMA (15 ng/ml) plus Ion (1 μM) for 2 and 6 h. Total RNA was purified, and COX-1, COX-2, and GAPDH mRNA levels were detected by RT-PCR analysis.

FIGURE 4.

Effects of CHX and ActD on COX-2 mRNA expression. Jurkat T cells were treated with CHX (5 μg/ml) or ActD (5 μg/ml) 45 min before stimulation with PMA (15 ng/ml) plus Ion (1 μM) for 2 and 6 h. Total RNA was purified, and COX-1, COX-2, and GAPDH mRNA levels were detected by RT-PCR analysis.

Close modal

To address whether COX-2 mRNA induction was paralleled by COX-2 protein increase, Western blot analysis, using specific Abs against COX-2 protein, was performed with extracts of T cells treated with anti-CD3 plus anti-CD28 and Jurkat cells stimulated with PMA plus Ion. COX 2 protein levels increased after T cell activation, showing a pattern of induction similar to that of its mRNA, although delayed. The COX-2 protein appeared later (6 h), reaching maximal levels at 12–24 h (Fig. 5,A). A similar pattern of expression was observed with Jurkat T cells. Treatment with CsA completely abrogated COX-2 protein induction in these cells (Fig. 5 B).

FIGURE 5.

COX-2 protein expression in human T cells. Protein extracts from purified human T lymphocytes (A) or Jurkat cells (B) treated with anti-CD3 plus anti-CD28 Abs and PMA plus Ion, respectively, for the times indicated (h) were separated by SDS-PAGE. Sample labeled as 24-C contains protein extracts pretreated with CsA (100 ng/ml). COX-2 protein levels were analyzed by Western blot analysis using anti-COX-2 mAb, as described in Materials and Methods. A band of similar molecular mass (∼70 kDa) is detected by this Ab in the control lane (labeled COX-2) that contains purified ovine COX-2.

FIGURE 5.

COX-2 protein expression in human T cells. Protein extracts from purified human T lymphocytes (A) or Jurkat cells (B) treated with anti-CD3 plus anti-CD28 Abs and PMA plus Ion, respectively, for the times indicated (h) were separated by SDS-PAGE. Sample labeled as 24-C contains protein extracts pretreated with CsA (100 ng/ml). COX-2 protein levels were analyzed by Western blot analysis using anti-COX-2 mAb, as described in Materials and Methods. A band of similar molecular mass (∼70 kDa) is detected by this Ab in the control lane (labeled COX-2) that contains purified ovine COX-2.

Close modal

To determine whether the induced expression of COX-2 after T cell activation results in increased cyclooxygenase activity, cell sonicates from both activated and normal T cells were incubated with AA as substrate. PMA plus Ion or anti-CD3 plus anti-CD28 treatment substantially increased PG production both in Jurkat and T cells (Table I).

Table I.

COX activity in Jurkat cells and T lymphocytes after PMA plus Ion or anti-CD3 + anti-CD28 treatment

Cell TypeTreatmentaPGs (ng/107 cells)b
Jurkat cells Control 0.13 ± 0.02c 
 PMA+ Ion 2.44 ± 0.18 
T lymphocytes Control 0.32 ± 0.01 
 Anti-CD3+ anti-CD28 8.88 ± 1.52 
Cell TypeTreatmentaPGs (ng/107 cells)b
Jurkat cells Control 0.13 ± 0.02c 
 PMA+ Ion 2.44 ± 0.18 
T lymphocytes Control 0.32 ± 0.01 
 Anti-CD3+ anti-CD28 8.88 ± 1.52 
a

Cells were incubated for 24 h with PMA + Ion or anti-CD3 + anti-CD28 Abs as described in Materials and Methods.

b

PG production in cell sonicates after 15-min incubation with AA measured by an standard EIA assay as described in Materials and Methods.

c

Mean ± SD.

To address the physiologic role of COX-2 induction in T cells, we studied the influence of NSAIDs, as COX inhibitors, on several T cell activation events. Human T cells were activated by anti-CD3 plus anti-CD28, and cell proliferation was measured in the presence of Indomethacin, a classical COX inhibitor being 60 times more active against COX-1 than COX-2 (42) or NS398, a newly described COX-2-selective inhibitor being 1000-fold more efficient for COX-2 inhibition (20, 43). Treatment with NS398 diminished in a concentration-dependent manner anti-CD3 plus anti-CD28-induced T cell proliferation (an average of 70–80% at 100 μM in the several experiments performed with different blood donors) (Table II). However, it failed to affect either basal proliferation or cell viability, as measured by trypan blue dye exclusion test (not shown). In contrast, Indomethacin at similar doses did not affect T cell proliferation.

Table II.

Effect of COX-2 inhibitors on anti-CD3 plus anti-CD28 induced T lymphocyte proliferation, CD69, CD71, and CD25 cell surface expression

ProliferationaCD69bCD71CD25
Control 1,389 ± 55 4 ± 2 (6) 5 ± 3 (11) 3 ± 1 (1) 
Anti-CD3+ anti-CD28 12,111 ± 1,930 64 ± 7 (104) 36 ± 3 (35) 48 ± 12 (44) 
Anti-CD3+ anti-CD28+ indomethacin (10 μM) 14,702 ± 3,796 63 ± 7 (92) 32 ± 4 (28) 34 ± 7 (29) 
Anti-CD3+ anti-CD28+ indomethacin (100 μM) 12,241 ± 4,334 66 ± 8 (106) 30 ± 4 (23) 38 ± 7 (28) 
Anti-CD3+ anti-CD28+ NS398 (10 μM) 9,598 ± 1,969 60 ± 6 (90) 30 ± 3 (27) 34 ± 4 (29) 
Anti-CD3+ anti-CD28+ NS398 (100 μM) 3,912 ± 961 53 ± 8 (85) 22 ± 3 (10)* 26 ± 6 (14)** 
ProliferationaCD69bCD71CD25
Control 1,389 ± 55 4 ± 2 (6) 5 ± 3 (11) 3 ± 1 (1) 
Anti-CD3+ anti-CD28 12,111 ± 1,930 64 ± 7 (104) 36 ± 3 (35) 48 ± 12 (44) 
Anti-CD3+ anti-CD28+ indomethacin (10 μM) 14,702 ± 3,796 63 ± 7 (92) 32 ± 4 (28) 34 ± 7 (29) 
Anti-CD3+ anti-CD28+ indomethacin (100 μM) 12,241 ± 4,334 66 ± 8 (106) 30 ± 4 (23) 38 ± 7 (28) 
Anti-CD3+ anti-CD28+ NS398 (10 μM) 9,598 ± 1,969 60 ± 6 (90) 30 ± 3 (27) 34 ± 4 (29) 
Anti-CD3+ anti-CD28+ NS398 (100 μM) 3,912 ± 961 53 ± 8 (85) 22 ± 3 (10)* 26 ± 6 (14)** 
a

[3H]Thymidine uptake was measured during the last 24 h of culture. Results are expressed as means ± SD cpm of triplicate measures. Experiments were repeated four times using T cells from different donors with reproducible results.

b

CD Ag surface expression was determined by FACS analysis as described in Materials and Methods. Arithmetic means ± SEM of the percentage of positive cells of four independent experiments are shown. The number in parentheses corresponds to the arithmetic mean of the mean fluorescent intensity (MFI) of positive cells.

c

, p < 0.05; and ∗∗, 0.05 < p < 0.1, as compared to anti-CD3 + anti-CD28 alone (Mann-Whitney U test).

T cell activation involves the induction of several cell surface molecules such as CD69 (a very early activation Ag not requiring new gene expression), CD25, or IL-2R α-chain (an early gene, requiring protein synthesis), and CD71 or transferrin receptor (a late gene) (1). Thus, the interference with T cell activation can be monitored through changes in the expression of markers of differentiation and proliferation in the T cell surface. Anti-CD3 plus anti-CD28 treatment was able to induce cell surface expression of the IL-2R α-chain (CD25). This induction was significantly inhibited by NS398 at 100 μM (an average of 70–80% considering together both the percentage of positive cells and the mean fluorescence intensity). Transferrin receptor expression (CD71), a marker of cell proliferation, was also inhibited by NS398 to a similar extent. Although Indomethacin at the highest concentrations used showed a slight inhibition of CD25 and CD71, this effect was not statistically significant. Meanwhile, COX-2-selective inhibition by NS398 did not significantly affect CD69 expression, further discarding nonspecific effects of this compound (Table II).

As a consequence of cell activation, T cells produce a variety of lymphokines such as IL-2, IL-5, IL-10, TNF-α, and IFN-γ. The coordinated production of these lymphokines is crucial for regulation of the inflammatory response. We evaluated the immunomodulatory actions of COX inhibitors on production of lymphokines by activated T cells cultured in the presence or absence of Indomethacin or NS398. As shown in Fig. 6, NS398, even at 10 μM, strongly inhibited IFN-γ production by activated T lymphocytes. It also substantially inhibited TNF-α, and IL-2 production. However, Indomethacin did not markedly alter either IL-2, TNF-α, or IFN-γ production upon activation.

FIGURE 6.

Effect of COX inhibitors on cytokine production by activated human T cells. Purified human T lymphocytes stimulated with anti-CD3 (2 μg/ml) plus anti-CD28 (250 ng/ml) Abs were treated with the indicated concentrations (μM) of Indomethacin or NS398. Levels of TNF-α, IL-2, and IFN-γ cytokines released into the medium were measured by ELISA. Results are the means of two determinations, each conducted in duplicate. Experiments were repeated with T cells from four different healthy donors, with reproducible results.

FIGURE 6.

Effect of COX inhibitors on cytokine production by activated human T cells. Purified human T lymphocytes stimulated with anti-CD3 (2 μg/ml) plus anti-CD28 (250 ng/ml) Abs were treated with the indicated concentrations (μM) of Indomethacin or NS398. Levels of TNF-α, IL-2, and IFN-γ cytokines released into the medium were measured by ELISA. Results are the means of two determinations, each conducted in duplicate. Experiments were repeated with T cells from four different healthy donors, with reproducible results.

Close modal

To evaluate whether the effect of COX-2-selective NSAIDs on IL-2 or TNF-α expression was at the transcriptional level, we investigated the regulation of their promoters in Jurkat cells transiently transfected with TNF-α or IL-2 promoter reporter plasmids TNF-α-LUC and IL-2-LUC. After transfection, cells were activated with PMA plus Ion in the presence or absence of COX inhibitors, and tested for luciferase activity. COX-2-selective inhibitors NS398 and Celecoxib efficiently inhibited PMA plus Ion-stimulated transcription from both of those reporters in a dose-dependent manner (Fig. 7,A). In contrast, Indomethacin only slightly affected IL-2 promoter-dependent transcription at 100 μM. The effect of COX-2 inhibitors was not due to an interference with the transcriptional or translational machinery or with the in vitro activity of the luciferase enzyme, as the basal luciferase activity in cells transfected with a plasmid containing the luciferase gene cloned downstream of the TK promoter was not affected substantially by any of these drugs (Fig. 7 B).

FIGURE 7.

Effect of COX inhibition on TNF-α and IL-2 promoter activity. A, Jurkat T cells transfected with TNF-α and IL-2 promoter luciferase reporter plasmids. B, Jurkat T cells transfected with the TK-LUC plasmid. Cells were preincubated for 1 h with vehicle (control), Indomethacin, NS398, or Celecoxib at the concentrations indicated (μM), before stimulation with PMA (15 ng/ml) plus Ion (1 μM) for 4–6 h. Results are the means of two determinations expressed as fold induction (observed experimental RLU/basal RLU in absence of any stimulus) or RLU ± SD. Results are representative of two independent experiments.

FIGURE 7.

Effect of COX inhibition on TNF-α and IL-2 promoter activity. A, Jurkat T cells transfected with TNF-α and IL-2 promoter luciferase reporter plasmids. B, Jurkat T cells transfected with the TK-LUC plasmid. Cells were preincubated for 1 h with vehicle (control), Indomethacin, NS398, or Celecoxib at the concentrations indicated (μM), before stimulation with PMA (15 ng/ml) plus Ion (1 μM) for 4–6 h. Results are the means of two determinations expressed as fold induction (observed experimental RLU/basal RLU in absence of any stimulus) or RLU ± SD. Results are representative of two independent experiments.

Close modal

Transcriptional activation of TNF-α, IL-2, and IFN-γ depends on the coordinate interactions among several transcription factors, including members of the NF-AT and NF-κB families. Thus, we evaluated the effect of NSAIDs on the transcriptional activity of those factors, by using reporter genes under the control of minimal promoters containing binding sites for each of them. Activation by PMA plus Ion increased the activity of these promoters. We found that NS398 and Celecoxib, but not Indomethacin, effectively inhibited each of these promoters in a dose-dependent manner (Fig. 8). Thus, NS398 and Celecoxib treatment results in an inhibitory effect of transcription factors that play general roles in the regulation of inflammatory responses in most cell types. These observations suggest that COX-2-selective NSAIDs might inhibit the expression of genes that became up-regulated during T cell activation and differentiation.

FIGURE 8.

Regulation of transcription factor-mediated transactivation by COX-2 inhibitors. Jurkat T cells were transiently transfected with luciferase reporter plasmids NF-κB-LUC or NF-AT-LUC, as described in Materials and Methods. Cells were preincubated for 1 h with Indomethacin, NS398, or Celecoxib at the concentrations indicated (μM), before stimulation with PMA (15 ng/ml) plus Ion (1 μM) for 4–6 h. Data shown are the means of replicate determinations from a representative experiment normalized to the nonstimulated samples. Similar results were obtained from three experiments, each one conducted in duplicate.

FIGURE 8.

Regulation of transcription factor-mediated transactivation by COX-2 inhibitors. Jurkat T cells were transiently transfected with luciferase reporter plasmids NF-κB-LUC or NF-AT-LUC, as described in Materials and Methods. Cells were preincubated for 1 h with Indomethacin, NS398, or Celecoxib at the concentrations indicated (μM), before stimulation with PMA (15 ng/ml) plus Ion (1 μM) for 4–6 h. Data shown are the means of replicate determinations from a representative experiment normalized to the nonstimulated samples. Similar results were obtained from three experiments, each one conducted in duplicate.

Close modal

In the present work, we have shown that basal low levels of COX-1 and COX-2 mRNAs can be detected in resting human T lymphocytes. However, while COX-2 mRNA and protein expression were markedly increased after treatment with several stimuli mimicking TCR/CD3 T cell activation, COX-1 expression was not significantly affected by any of these treatments. Despite the controversy regarding the ability of human T lymphocytes to generate eicosanoids (10, 44, 45), to our knowledge, no studies have yet analyzed COX expression in this cell type. The present study clearly demonstrates that the COX-2 isoenzyme is expressed in T cells and up-regulated on activation. Increased COX-2 mRNA levels lead to increased COX-2 protein and cyclooxygenase activity both in Jurkat and T lymphocytes.

Inducible expression of COX-2 in response to growth factors, tumor promoters, or cytokines has been previously and extensively described in several cell types, including fibroblasts, endothelial cells, and macrophages (15, 17), but not, to our knowledge, in T cells. The COX-1 isoform appears to be maintained at constant levels in virtually all of these cells, although stimulation with growth factors can result in a moderate increase (15, 17). On the other hand, COX-2 gene transcription after T cell activation fulfills all of the criteria for an immediate early T cell activation gene: it is rapidly inducible, independent on new protein synthesis, superinducible by CHX, and dependent on new RNA synthesis. The COX-2 gene has been previously defined as an inducible immediate early gene in other cell types (40, 46).

In addition, COX-2 induction was completely abolished by CsA, a potent immunosuppressive drug that strongly inhibits the transcription of early T cell activation genes, including a number of cytokine genes such as IL-2, TNF-α, IFN-γ, GM-CSF, as well as c-myc and various related protooncogenes of the src family (38, 39, 47). As shown by Martin, et al. (48), CsA also effectively inhibits COX-2 mRNA induction by IL-1 or TNF in rat mesangial cells.

To investigate the role of COX-2 on T cell activation events, we treated T cells with NSAIDs. Thus, NS398, a highly selective COX-2 inhibitor (43), but not Indomethacin, a preferential COX-1 NSAID (42), decreased T cell proliferation. Although it is generally accepted that Indomethacin is a preferential but not selective inhibitor of COX-1, the doses required to inhibit COX-2 vary with the assay conditions (49). Thus, doses between 10 and 20 μM Indomethacin have been reported to act efficiently as COX-1-specific inhibitor (50, 51). The action of NS398 was accompanied by a decrease in IL-2R α-chain (CD25) and transferrin receptor (CD71) membrane surface expression. This inhibition was presumably not an overall action unspecifically affecting surface membrane expression or T cell activation, as the induction of other cell surface Ags expressed in activated T cells such as CD69 was not affected. NS398 inhibited efficiently IL-2, TNF-α, and IFN-γ production. Similar results were obtained with other chemically unrelated COX-2-specific inhibitors such as Celecoxib (19) (data not shown). The small inhibitory effect of Indomethacin observed at the highest dose in some experiments may reflect their ability to partially inhibit COX-2 enzymatic activity at that concentration. These data support the hypothesis that highly COX-2-selective NSAIDs are immunosuppressive agents. They inhibit not only T cell proliferation and CD25 and CD71 expression, but also IL-2, TNF-α, and IFN-γ transcription.

We believe that both the expression of COX-2 and the effect of COX-2-specific drugs in our system cannot be ascribed to the possible contamination by monocytes/macrophages of our purified T cell populations for several reasons: 1) the effects observed on normal T cells can be mimicked in Jurkat T cells; 2) COX-2 induction (mrRNA and protein) occurs rapidly in response to stimuli that specifically activate T lymphocytes through the TCR/CD3 complex and were blocked by CsA. Moreover, if the effects were at the level of macrophages, their COX-2 would presumably produce PGE2 (10) that in turn would decrease CD25 expression, IL-2, TNF-α, and IFN-γ production and cell proliferation (11, 12, 13, 14). Therefore, blocking of COX-2-produced PGE2 by specific inhibitors will produce opposite effects to those we observed, resulting in an enhancement and not a decrease of T cell activation as we found. Thus, other actions may account for COX-2 physiologic functions in the T cell activation process.

In this regard, it is worth noting that some COX-2 inhibitors have been shown to possess immunosuppressive effects and to inhibit cytokine synthesis in other cell types in addition to their anti-inflammatory properties. Thus, it is well known that glucocorticoids, widely used as immunosuppressive and potent anti-inflammatory drugs, also inhibit the induction of COX-2 expression in a variety of cell lines and in response to different stimuli (52, 53). Tepoxalin, a cyclooxygenase and lipoxygenase inhibitor, blocked neutrophil infiltration by diminishing up-regulation of adhesion molecules E-selectin and Mac-1 and inhibited IL-2, IL-6, IL-8, and IFN-γ induction in T cells (25, 54, 55). Moreover, a new class of cytokine-suppressive anti-inflammatory drugs (CSAIDs) have been defined as sharing both COX- and cytokine-inhibiting properties in LPS-stimulated human monocytes (56, 57). Other NSAIDs, such as Tenidap, Ibuprofen, and Naproxen, suppressed the proliferative response of T cells to IL-2 (23). Furthermore, Tenidap inhibited IFN-γ production and mRNA expression in T lymphocytes (24). Collectively, these data suggest that the anti-inflammatory actions of NSAIDs will not be only restricted to the inhibition of COX-dependent PGE2 synthesis and support the hypothesis of the immunomodulatory role of NSAIDs.

Activation of T cells through the TCR/CD3 complex triggers a program of gene expression that can be divided into three phases: immediate events (independent on new protein synthesis), early events (dependent on protein synthesis and preceding cell division), and late events (occurring after cell proliferation) (1). In response to TCR/CD3-transduced signals, T cells activate a number of immediate early activation genes, such as c-fos, c-myc, and members of the src family (1, 47). The production of these genes is presumably involved in controlling and initiating the cascade of transcriptional responses that lead to complete activation. Our results indicate that COX-2 belongs to this category of immediate early genes. Therefore, by analogy with other immediate early genes, COX-2 activation may participate in the cascade of events occurring after T cell activation. Because COX-2 and CD69 share the characteristic of immediate early genes, the lack of effect of COX-2 inhibitors on CD69 expression reinforces the fact that COX-2 actions specifically occur downstream in the early and late but not in the immediate early phase of T cell activation.

The mechanism of action of COX-2-inhibitory drugs appears to be related to nuclear transcriptional activation. COX-2-specific inhibitors NS398 and Celecoxib effectively inhibited PMA plus Ion-stimulated transcription from these promoters. IL-2, TNF-α, and IFN-γ activation take place through activation of transcription factors, including members of the NF-AT and NF-κB families (33, 58, 59). COX-2 inhibition also led to down-regulation of NF-AT- and NF-κB-driven transcription. These effects were specific because transcription, translation, or activity of a TK-driven luciferase reporter was not altered by any of these drugs. These observations suggest some general effects of COX-2 inhibitors on the activity of TCR/CD3-driven signals through pathways involving Ca2+/calcineurin or NF-κB. It has been shown previously that NF-AT and NF-κB activation are sensitive to CsA (60, 61). This suggests that calcineurin acts as a common upstream signaling molecule. Moreover, our results indicate that COX-2 induction is also sensitive to CsA, also suggesting the requirement for calcineurin. However, COX-2 inhibition blocks NF-AT and NF-κB activation, placing COX-2 actions upstream, not downstream of NF-AT. This apparent paradox reflects the fact that T cell activation as well as NF-κB and NF-AT activation are biphasic (60, 61). Both factors are initially present in an inactive form in the cytoplasm of resting T cells, and rapidly translocate to the nucleus upon activation. Thus, it is impossible that COX-2 actions will be involved in this initial phase because COX-2 protein is not present at this moment. Rather, COX-2 actions will most likely occur at the second phase of T cell activation. In support of this theory is the fact that CD69, a NF-κB-dependent immediate early gene, was not affected by COX-2 inhibition. Both pathways share activation by numerous kinases and phosphatases that might be targets of COX-2 inhibition in this second phase. Together, these data emphasize the role of COX-2 in the T cell signaling activation process.

On the other hand, our results may provide an explanation at the molecular level for previous results with COX inhibitors. Thus, it has been shown that aspirin and sodium salicylate inhibit NF-κB-dependent activation in Jurkat cells, thus decreasing dependent gene induction (22). Interestingly, although both aspirin and sodium salicylate are more effective COX-1 inhibitors, the doses used in these experiments (1–10 mM) were high enough to also inhibit COX-2 activity (42). Tepoxalin, a COX-2 inhibitor, also inhibits NF-κB function and activation (55). Taken together, several structurally unrelated COX-2 inhibitors, NS-398, Celecoxib, Tepoxalin, as well as Aspirin or salicylate at high doses produce the same effects on T cells: blockade of transcription factor NF-κB-dependent induction. The most likely explanation is that all of these effects derive from their inhibition of COX-2 activity. COX-2 produces PGs within or in the nucleus envelope, whereas COX-1 products are only located at the endoplasmic reticulum (62). New undescribed COX-2 metabolites (or simply previously described ones undetected because of their nuclear localization), produced in the nucleus of activated T cells, may further regulate gene transcription or other nuclear events during cell activation.

However, other mechanisms cannot be discarded. Evidence of a mechanism involving gene regulation by eicosanoids through nuclear receptors of the PPAR family has been described (26, 27, 63). Thus, PGs of the J series regulate gene expression of proinflammatory genes at the promoter level in macrophages by binding to PPAR-γ transcription factor (64). In agreement with a role of COX-2 metabolites in gene transcription is the fact that COX-2 overexpression controls mitogenesis, apoptosis, carcinogenesis, and even angiogenesis at the transcription level in other cell types (65, 66, 67), further supporting an important role of COX-2 in many important aspects of cell activation. Alternatively, COX might play an important role through its peroxidase activity on T lymphocyte function and activation, because transcription factors with a key role in this process such as NF-κB can be activated by COX-1, apparently using intracellular reactive oxygen intermediates as second messengers (68). It has been shown recently that aspirin and salicylate inhibit NF-κB activation through binding to the IκB kinase β (IKκ-β) (69). Thus, it is possible that new targets for NS398 and Celecoxib may account for some of their actions in addition to COX-2 enzymatic inhibition. Together, these data suggest that COX-2 actions may not only be exerted through synthesis of classical PGs, and new mechanisms of action that play a key role in controlling activation processes remain to be elucidated.

In summary, this study provides the first evidence that transcriptional up-regulation of COX-2 isoenzyme occurs after T cell activation, and suggests functional implications of COX-2 activity in this process. More importantly, our results suggest that NSAIDs with COX-2 selectivity may be attractive anti-inflammatory drugs, as they possess additional anti-inflammatory properties to their selective down-regulation of PG production, such as regulation of the immune activation and the production of proinflammatory cytokines by T cells. This may have profound implications for NSAIDs therapy and the development of new COX-2-specific inhibitors.

We thank Dr. J. M. Redondo for the critical reading of the manuscript, Laboratorios Dr. ESTEVE for the supplying of reagents, and Maria Chorro for her excellent technical assistance.

1

This work was supported by grants from Laboratorios Dr. Esteve, Dirección General de Investigación Científica y Técnica of Spain, Comunidad Autónoma de Madrid, and Fundación Ramón Areces.

3

Abbreviations used in this paper: AA, arachidonic acid; ActD, actinomycin D; CHX, cycloheximide; COX, cyclooxygenase; CsA, cyclosporin A; Ion, calcium ionophore A23187; LUC, luciferase; NSAID, nonsteroidal anti-inflammatory drugs; PPAR, peroxisome proliferator-activated factor; RLU, relative luciferase unit.

1
Ullman, K. S., J. P. Northrop, C. L. Verweij, G. R. Crabtree.
1990
. Transmission of signals from the T lymphocyte antigen receptor to the genes responsible for cell proliferation and immune function: the missing link.
Annu. Rev. Immunol.
8
:
421
2
Weiss, A., D. R. Littman.
1994
. Signal transduction by lymphocyte antigen receptors.
Cell
76
:
263
3
Altman, A., K. M. Coggeshall, T. Mustelin.
1990
. Molecular events mediating T cell activation.
Adv. Immunol.
48
:
227
4
Harris, E. D., Jr.
1990
. Rheumatoid arthritis. pathophysiology and implications for therapy.
N. Engl. J. Med.
322
:
1277
5
Feldmann, M., F. Brennan, R. N. Maini.
1996
. Role of cytokines in rheumatoid arthritis.
Annu. Rev. Immunol.
14
:
397
6
Moulton, P. J..
1996
. Inflammatory joint disease: the role of cytokines, cyclooxygenases and reactive oxygen species.
Br. J. Biomed. Sci.
53
:
317
7
Samuelson, B., M. Goldyne, E. Granström, M. Hamberg, S. Hammarström, C. Malmsten.
1978
. Prostaglandins and thromboxanes.
Annu. Rev. Biochem.
47
:
997
8
Hammarström, S..
1983
. Leukotrienes.
Annu. Rev. Biochem.
52
:
355
9
Kurland, J. I., R. Bockman.
1978
. Prostaglandin E production by human blood monocytes and mouse peritoneal macrophages.
J. Exp. Med.
147
:
952
10
Kennedy, M. S., J. D. Stobo, M. E. Goldyne.
1980
. In vitro synthesis of prostaglandins and related lipids by populations of human peripheral blood mononuclear cells.
Prostaglandins
20
:
135
11
Goodwin, J. S..
1989
. Immunomodulation by eicosanoids and anti-inflammatory drugs.
Curr. Opin. Immunol.
3
:
295
12
Della Bella, S., M. Molteni, S. Compasso, C. Zulian, M. Vanoli, R. Scorza.
1997
. Differential effects of cyclo-oxygenase pathway metabolites on cytokine production by T lymphocytes.
Prostaglandins Leukotrienes Essent. Fatty Acids
56
:
177
13
Kumar, G. S., U. N. Das.
1994
. Effect of prostaglandins and their precursor on the proliferation of human lymphocytes and their secretion of tumor necrosis factor and various interleukins,.
Prostaglandins Leukotrienes Essent. Fatty Acids
50
:
331
14
Phipps, R. P., S. H. Stein, R. L. Roper.
1991
. A new view of prostaglandin E regulation of the immune response.
Immunol. Today
12
:
349
15
Smith, W. L., D. L. DeWitt.
1996
. Prostaglandin endoperoxide H synthase-1 and -2.
Adv. Immunol.
62
:
167
16
Griswold, D. E., J. L. Adams.
1996
. Constitutive cyclooxygenase (Cox-1) and inducible cyclooxygenase (Cox-2): rationale for selective inhibition and progress to date.
Med. Res. Rev.
16
:
181
17
Herschman, H. R..
1996
. Prostaglandin synthase 2.
Biochim. Biophys. Acta
1299
:
125
18
Jouzeau, J.-Y., B. Terlain, A. Abid, E. Nédélec, P. Netter.
1997
. Cyclo-oxygenase isoenzymes: how recent findings affect thinking about nonsteroidal anti-inflammatory drugs.
Drugs
53
:
563
19
Penning, T. D., J. J. Talley, S. R. Bertenshaw, J. S. Carter, P. W. Collins, S. Docter, M. J. Graneto, L. F. Lee, J. W. Malecha, J. M. Miyashiro, et al
1997
. Synthesis and biological evaluation of the 1,5-diarylpyrazole class of cyclooxygenase-2 inhibitors: identification of 4-[5-(4-methylphenyl)-3-(trifluoromethyl)-1H-pyrazol-1-yl]benze nesulfonamide (SC-58635, celecoxib).
J. Med. Chem.
40
:
1347
20
Futaki, N., I. Arai, Y. Hamasaka, S. Takahashi, S. Higuchi, S. Otomo.
1993
. Selective inhibition of NS-398 on prostanoid production in inflamed tissue in rat carrageenan-air-pouch inflammation.
J. Pharm. Pharmacol.
45
:
753
21
Engelhardt, G., D. Homma, K. Schlegel, R. Utzmann, C. Schnitzler.
1995
. Anti-inflammatory, analgesic, antipyretic and related properties of meloxicam, a new non-steroidal anti-inflammatory agent with favorable gastrointestinal tolerance.
Inflamm. Res.
44
:
423
22
Koop, E., S. Ghosh.
1994
. Inhibition of NF-κB by sodium salycilate and aspirin.
Science
265
:
956
23
Hall, V. C., R. E. Wolf.
1997
. Effects of tenidap and nonsteroidal antiinflammatory drugs on the response of cultured human T cells to interleukin 2 in rheumatoid arthritis.
J. Rheumatol.
24
:
1467
24
Dolhain, R. J. E. M., P. DeKuiper, C. L. Yerweis, J. M. A. Penders, F. C. Breedveld, B. A. C. Dijkmans, A. M. M. Miltenburg.
1995
. Tenidap, but not nonsteroidal anti-inflammatory drugs, inhibits T-cell proliferation and cytokine induction.
Scand. J. Immunol.
42
:
686
25
Zhou, L., D. Ritchie, E. Y. Wang, A. G. Barbone, D. Argentier, C. Y. Lau.
1994
. Tepoxalin, a novel immunosuppressive agent with a different mechanism of action from cyclosporin A.
J. Immunol.
153
:
5026
26
Jiang, C., A. T. Ting, B. Seed.
1998
. PPAR-γ agonists inhibit production of monocyte inflammatory cytokines.
Nature
391
:
82
27
Lehman, J. M., J. M. Lenhard, B. B. Olier, G. M. Ringold, S. A. Kliewer.
1997
. Peroxisome proliferator-activated receptors α and γ are activated by indomethacin and other non-steroidal anti-inflammatory drugs.
J. Biol. Chem.
272
:
3406
28
Pimentel-Muiños, F. X., M. A. Muñoz-Fernández, M. Fresno.
1994
. Control of T lymphocyte activation and interleukin-2 receptor expression by endogenously secreted lymphokines.
J. Immunol.
152
:
5714
29
Iñiguez, M., J. Pablos, P. Carreira, F. Cabré, J. Gomez-Reino.
1998
. Detection of Cox-1 and Cox-2 isoforms in synovial fluid cells from inflammatory joint diseases.
Br. J. Rheumatol.
37
:
773
30
Sambrook, J., E. Fritsch, T. Maniatis.
1989
.
Molecular Cloning. A Laboratory Manual
Cold Spring Harbor Laboratory Press, Cold Spring Harbor.
31
Morte, B., M. Iñiguez, P. Lorenzo, J. Bernal.
1997
. Thyroid hormone-regulated expression of RC3/neurogranin in the immortalized hypothalamic cell line GT1-7.
J. Neurochem.
69
:
902
32
Yano, O., J. Kanellopoulos, M. Kieran, O. Le Bail, A. Israël, P. Kourilsky.
1987
. Purification of κBF1, a common factor binding to both H-2 and β2-microglobulin enhancers.
EMBO J.
6
:
3317
33
Rhoades, K. L., S. H. Golub, J. Economou.
1992
. The regulation of the human tumor necrosis factor α promoter region in macrophage, T cell, and B cell lines.
J. Biol. Chem.
267
:
22102
34
Durand, D. B., S. J.-P., R. Bush, R. E. Peplogle, R. Belagaje, and G. R. Crabtree. 1988. Characterization of antigen receptor response elements within the interleukin-2-enhancer. Mol. Cell. Biol. 8:1715.
35
Nordeen, S. K..
1988
. Luciferase reporter gene vectors for analysis of promoter and enhancers.
Biotechniques
6
:
454
36
Truneh, A., F. Albert, P. Golstein, A. M. Schmitt-Verhulst.
1985
. Early steps of lymphocyte activation bypassed by synergy between calcium ionophores and phorbol ester.
Nature
313
:
318
37
Linsley, P. S., J. A. Ledbetter.
1993
. The role of the CD28 receptor during T cell responses to antigen.
Annu. Rev. Immunol.
11
:
191
38
Quesniaux, V. F. J..
1993
. Immunosuppressants: tools to investigate the physiological role of cytokines.
Bioessays
15
:
731
39
Schreiber, S. L., G. R. Crabtree.
1992
. The mechanism of action of cyclosporin A and FK506.
Immunol. Today
13
:
136
40
Ryseck, R., C. Raynoschek, K. MacDonald-Bravo, K. Dorfman, R. Bravo.
1992
. Identification of an immediate early gene, PGHS-b, whose protein product has prostaglandin synthase/cyclooxygenase activity.
Cell Growth Differ.
3
:
443
41
Herschmann, H. R..
1991
. Primary response genes induced by growth factors and tumor promoters.
Annu. Rev. Biochem.
60
:
281
42
Mitchell, J. A., P. Akarasereenont, C. Thiemermann, R. J. Flower, J. R. Vane.
1993
. Selectivity of nonsteroidal antiinflammatory drugs as inhibitors of constitutive and inducible cyclooxygenase.
Proc. Natl. Acad. Sci. USA
90
:
11693
43
Futaki, N., S. Takahashi, M. Yokoyama, I. Arai, S. Higuchi, S. Otomo.
1994
. NS-398, a new anti-inflammatory agent, selectively inhibits prostaglandin G/H synthase/cyclooxygenase (COX-2) activity in vitro.
Prostaglandins
47
:
55
44
Goldyne, M. E..
1989
. Eicosanoid metabolism by lymphocytes: do all human nucleated cells generate eicosanoids?.
Pharmacol. Res.
21
:
241
45
Goldyne, M. E..
1988
. Lymphocytes and arachidonic acid metabolism.
Prog. Allergy
44
:
140
46
Maier, J. A., T. Hla, T. Maciag.
1990
. Cyclooxygenase is an immediate-early gene induced by interleukin-1 in human endothelial cells.
J. Biol. Chem.
265
:
10805
47
Furue, M., S. I. Katz, Y. Kawakami, T. Kawakami.
1990
. Coordinate expression of src family protooncogenes in T cell activation and its modulation by cyclosporine.
J. Immunol.
144
:
736
48
Martin, M., D. Neumann, T. Hoff, K. Resch, D. L. DeWitt, S. M. Goppelt.
1994
. Interleukin-1-induced cyclooxygenase 2 expression is suppressed by cyclosporin A in rat mesangial cells.
Kidney Int.
45
:
150
49
Griswold, D., J. Adams.
1996
. Constitutive cyclooxygenase (Cox-1) and inducible cyclooxygenase (Cox-2): rationale for selective inhibition and progress to date.
Med. Res. Rev.
16
:
181
50
Cadroy, Y., D. Dupouy, B. Boneu.
1998
. Arachidonic acid enhances the tissue factor expression of mononuclear cells by the cyclo-oxygenase-1 pathway: beneficial effect of n-3 fatty acids.
J. Immunol.
160
:
6145
51
Meade, E., W. Smith, D. DeWitt.
1993
. Differential inhibition of prostaglandin endoperoxide synthase (cyclooxygenase) isozymes by aspirin and other non-steroidal antiinflammatory drugs.
J. Biol. Chem.
268
:
6610
52
Masferrer, J. L., B. S. Zweifel, K. Seibert, P. Needleman.
1990
. Selective regulation of cellular cyclooxygenase by dexamethasone and endotoxin in mice.
J. Clin. Invest.
86
:
1375
53
O’Banion, M. K., V. D. Winn, D. A. Young.
1992
. cDNA cloning and functional activity of a glucocorticoid-regulated inflammatory cyclooxygenase.
Proc. Natl. Acad. Sci. USA
89
:
4888
54
Zhou, L., B. L. Pope, E. Chourmouzis, L. W. P. Fung, C. Y. Lau.
1996
. Tepoxalin blocks neutrophil migration into cutaneous inflammatory sites by inhibiting Mac-1 and E-selectin expression.
Eur. J. Immunol.
26
:
120
55
Kazmi, S. M. I., R. K. Plante, V. Visconti, G. R. Taylor, L. Zhou, C. Y. Lau.
1995
. Suppression of NFκB activation and NFκB-dependent gene expression by Tepoxalin, a dual inhibitor of cyclooxygenase and 5-lipoxygenase.
J. Cell. Biochem.
57
:
299
56
Griswold, D. E., L. M. Hillegass, J. J. Breton, K. M. Esser, J. L. Adams.
1993
. Differentiation in vivo of classical non-steroidal antiinflammatory drugs from cytokine suppressive antiinflammatory drugs and other pharmacological classes using mouse tumor necrosis factor α production.
Drugs Exp. Clin. Res.
19
:
243
57
Pouliot, M., J. Baillargeon, J. C. Lee, L. G. Cleland, M. J. James.
1997
. Inhibition of prostaglandin endoperoxide synthase-2 expression in stimulated human monocytes by inhibitors of p38 mitogen-activated protein kinase.
J. Immunol.
158
:
4930
58
Jain, J., C. Lom, A. Rao.
1995
. Transcriptional regulation of the IL-2 gene.
Curr. Opin. Immunol.
7
:
333
59
Penix, L., W. M. Weaver, Y. Pang, H. A. Young, C. B. Wilson.
1993
. Two essential regulatory elements in the human interferon γ promoter confer activation specific expression in T cells.
J. Exp. Med.
178
:
1483
60
Rao, A., C. Luo, P. Hogan.
1997
. Transcription factors of the NFAT family: regulation and function.
Annu. Rev. Immunol.
15
:
707
61
Pimentel-Muiños, F., J. Mazana, M. Fresno.
1995
. Biphasic control of nuclear factor-κB activation by the T cell receptor complex: role of tumor necrosis factor α.
Eur. J. Immunol.
25
:
179
62
Morita, I., M. Schindler, M. K. Regier, J. C. Otto, T. Hori, D. L. DeWitt, W. L. Smith.
1995
. Different intracellular locations for prostaglandin endoperoxide H synthases-1 and -2.
J. Biol. Chem.
270
:
10902
63
Devchand, P. R., H. Keller, J. M. Peters, M. Vazquez, F. J. González, W. Wahli.
1996
. The PPAR-α-leukotriene B4 pathway to inflammation control.
Nature
384
:
39
64
Ricote, M., A. C. Li, T. M. Willson, C. J. Kelly, C. K. Glass.
1998
. The peroxisome proliferator-activated receptor-γ is a negative regulator of macrophage activation.
Nature
391
:
79
65
DuBois, R. N., J. Shao, M. Tsujii, H. Sheng, R. D. Beauchamp.
1996
. G1 delay in cells overexpressing prostaglandin endoperoxide synthase-2.
Cancer Res.
56
:
733
66
Tsuji, M., S. Kawano, S. Tsuji, H. Sawaoka, M. Hori, R. DuBois.
1998
. Cyclooxygenase regulates angiogenesis induced by colon cancer cells.
Cell
93
:
705
67
Tsujii, M., R. N. DuBois.
1995
. Alterations in cellular adhesion and apoptosis in epithelial cells overexpressing prostaglandin endoperoxide synthase 2.
Cell
83
:
493
68
Munroe, D. G., E. Y. Wang, J. P. MacIntyre, S. S. Tam, D. H. Lee, G. R. Taylor, L. Zhou, R. K. Plante, S. M. Kazmi, P. A. Bauerle.
1995
. Novel intracellular signalling function of prostaglandin H synthase-1 in NF-κB activation.
J. Inflamm.
45
:
260
69
Yin, M.-J., Y. Yamamoto, R. Gaynor.
1998
. The anti-inflammatory agents aspirin and salycilate inhibit the activity of IκB kinase-β.
Nature
396
:
77