Induction of COX-2 and PGE₂ biosynthesis by IL-1β is mediated by PKC and mitogen-activated protein kinases in murine astrocytes

Abstract

1. Interleukin-1 (IL-1) is an important mediator of immunoinflammatory responses in the brain. In the present study, we examined whether prostaglandin E₂ (PGE₂) production after IL-1β stimulation is dependent upon activation of protein kinases in astroglial cells.

2. Astrocyte cultures stimulated with IL-1β or the phorbol ester, PMA significantly increased PGE₂ secretion. The stimulatory action of IL-1β on PGE₂ production was totally abolished by NS-398, a specific inhibitor of cyclo-oxygenase-2 activity, as well as by the protein synthesis inhibitor cycloheximide, and the glucocorticoid dexamethasone. Furthermore, IL-1β induced the expression of COX-2 mRNA. This occurred early at 2 h, with a maximum at 4 h and declined at 12 h. IL-1 β treatment also induced the expression of COX-2 protein as determined by immunoblot analysis. In that case the expression of the protein remained high at least up to 12 h.

3. Treatment of cells with protein kinase C inhibitors (H-7, bisindolylmaleimide and calphostin C) inhibited IL-1β stimulation of PGE₂. In addition, PKC-depleted astrocyte cultures by overnight treatment with PMA no longer responded to PMA or IL-1. The ablation of the effects of PMA and IL-1β on PGE₂ production, likely results from down-regulation of phorbol ester sensitive-PKC isoenzymes. Immunoblot analysis demonstrated the translocation of the conventional isoform cPKC-α from cytosol to membrane following treatment with IL-1β.

4. In addition, IL-1β treatment led to activation of extracellular signal-regulated kinase (ERK1/2) and p38 subgroups of MAP kinases in astroglial cells. Interestingly, the inhibition of ERK kinase with PD 98059, as well as the inhibition of p38 MAPK with SB 203580, prevented IL-1β-induced PGE₂ release.

5. ERK1/2 activation by IL-1β was sensitive to inhibition by the PKC inhibitor bisindolylmaleimide suggesting that ERK phosphorylation is a downstream signal of PKC activation.

6. These results suggest key roles for PKC as well as for ERK1/2 and p38 MAP kinase cascades in the biosynthesis of PGE₂, likely by regulating the induction of cyclo-oxygenase-2, in IL-1β-stimulated astroglial cells.

Introduction

The pro-inflammatory cytokine, interleukin-1 (IL-1) is secreted by a variety of cells in response to infection, activated lymphocyte products, microbial toxins, and inflammatory stimuli (Dinarello, 1996). Besides the classical role of IL-1 as immune mediator, this cytokine has numerous effects in the central nervous system (CNS) (Berkenbosh et al., 1987; Hansen & Krueger, 1997; Rothwell, 1991; Hopkins & Rothwell, 1995). IL-1β has been shown to be produced in the CNS in response to a number of stimuli, such as peripheral lipopolysaccharide administration (Layé et al., 1994), traumatic brain injury (Fan et al., 1995) and brain viral infection (Marquette et al., 1996; Lledó et al., 1999). Under pathological situations involving CNS inflammation, IL-1 is mainly secreted by activated microglia/macrophages (Giulian et al., 1986; Benveniste, 1992), while astrocytes appear to be its main target as suggested by the presence of IL-1 receptors in their surface (Ban et al., 1993). In astroglial cells, IL-1 induces the expression of other cytokines, such as interleukin-6 and tumour necrosis factor-α, as well as other inflammatory mediators that may be implicated in the CNS response to injury (Merrill & Benveniste, 1996).

Arachidonic acid metabolites, such as prostaglandins are closely involved in the inflammatory responses following brain injury and bacterial or viral infections. High levels of prostaglandins have been measured in multiple sclerosis, AIDS-associated dementia, and other neurodegenerative disorders (Bolton et al., 1986; Fretland et al., 1992; Griffin et al., 1994). Cyclo-oxygenase is the rate-limiting enzyme in prostanoids biosynthesis and exists as a widely constitutive isoform (COX-1), responsible for physiological levels of prostaglandins and as an inducible COX-2 encoded by an immediate early gene, rapidly induced after pro-inflammatory stimuli. COX-2 is expressed by inflammatory cells and is responsible for the high levels of prostaglandins present in acute and chronic inflammation, being inhibited by glucocorticoid hormones (O'Banion et al., 1992; Herschman 1996). Interestingly, systemic LPS and IL-1β injections have been shown to stimulate production of COX-2 within the mouse (Breder & Saper, 1996) and rat (Cao et al., 1996; Lacroix & Rivest, 1998) brain microvaculature. In addition, COX-2 expression is increased in frontal cortex of Alzheimer's disease brains (Pasinetti & Aisen, 1998). During the last years, the physiological role of COX-2 has been a topic of much interest (Vane et al., 1998).

In vitro studies have revealed the capacity of astrocytes to release prostaglandins and express mRNA COX-2 in response to IL-1β (Hartung et al., 1989; Katsuura et al., 1989; O'Banion et al., 1996). However, the intracellular signalling mechanisms triggered by IL-1β and responsible for prostaglandin production are not completely defined. Protein kinase C (PKC) represents a family of closely related serine/threonine kinases that play a key role in different cellular signal transduction pathways (Nishizuka, 1986; 1992). A number of studies have demonstrated that phorbol esters, direct activators of PKC, stimulate the synthesis of prostaglandins (Crofford et al., 1994). However, in mesangial cells, prostaglandin biosynthesis induced by IL-1β seems to be independent on PKC activation (Conquer et al., 1992). Activation of mitogen-activated protein kinases (MAPK) cascade is one of the most rapid cellular responses to different stimuli, including IL-1 (Wilmer et al., 1997; O Neill & Greene 1998). Moreover, the cascade of MAPKs represent a point of convergence in intracellular signalling activated by protein Tyr kinases, by G-protein coupled receptors and by PKC. Several studies have linked prostaglandin biosynthetic pathway which mediate inflammatory responses, with activation of MAPK signalling cascade (Lin et al., 1993; Sanghera et al., 1996; Hambleton et al., 1996; Niiro et al., 1998). In mammalian cells, three subgroups of MAPKs have been detected and included the extracellular signal-regulated kinase, p42/44 (ERK1/2), the C-jun amino terminal kinases (JNKs) and p38 MAPKs (Cano & Mahadevan, 1995). These distinct sets of MAPKs can be activated by a variety of extracellular stimuli, and individual MAPK kinases may be implicated in the expression of a number of pro-inflammatory molecules depending on the cell type (Pouliot et al., 1997).

In the current study, we described that IL-1β induced the expression of mRNA COX-2 and the protein in astrocytes. In addition, we analysed pharmacologically the possible participation of PKC activation and MAPKs cascade in IL-1β-induced PGE₂ production by mouse astrocyte cultures.

Methods

The following reagents were obtained from the indicated suppliers: Dulbecco's modified Eagle medium, foetal calf serum (FCS), penicillin/streptomycin mix from Gibco-BRL (Barcelona, Spain); recombinant mouse interleukin-1β (IL-1β) from Genzyme (Cambridge, MA, U.S.A.); monoclonal anti-GFAP, lipopolysaccharide (LPS) (from E. Coli 026:B6), H-7 1-(5-isoquinolinylsulphonyl)-2-)-2-methylpiperazine), 12-O-tetradecanoylphorbol 13 acetate (TPA); 4-α phorbol 12-myristate 13 acetate (4-αPMA), actinomycine D and cycloheximide from Sigma (St. Louis, MO, U.S.A.); bisindolylmaleimide I, NS-398, calphostin-C, PD 98059, SB 203580, from Calbiochem (La Jolla, CA, U.S.A.); PGE₂ enzymeimmunoassay system BIOTRAK, Hybond ECL-nitrocellulose membrane and ECL Western blotting detection reagents from Amersham Pharmacia Biotech (London, U.K.); culture flasks and dishes were from Falcon (Franklin Lakes, NJ, U.S.A.); Affinity-purified rabbit anti-phospho p42/44 and anti-phospho p38 were from New England Biolabs (Beverly, MA, U.S.A.); rabbit polyclonal anti-PKC-α was from Santa Cruz Biotechnology (Santa Cruz, CA, U.S.A.); COX-2 antibody from Cayman Chemicals (MI), Mac-1 antibody from Serotec (Oxford, U.K.) and the secondary antibody peroxidase-conjugated goat anti-rabbit IgG was from Jackson Immuno Research Laboratories (West Grove, PA, U.S.A.). Secondary antibodies for immuno-fluorescence were from Southern Biotechnology (Birmingham, AL, U.S.A.). All other reagents were obtained from standard suppliers.

Astrocyte cultures

Primary astrocyte cultures were generated from the cerebral cortex of 1-day-old neonatal mice (Balb/c Cajal Institute, Madrid, Spain) as described by McCarthy & de Vellis (1980) with our modifications (Molina-Holgado et al., 1995). Cells were plated on 75 cm² flasks in Dulbecco's modified Eagle medium (DMEM) containing 10% heat-inactivated foetal calf serum plus 50 units ml⁻¹ penicillin and 50 μg ml⁻¹ streptomycin (growth medium). After 10 days, the cultures were shaken for 18 h at 240 r.p.m. at 37°C to remove microglia and oligodendrocyte progenitors. The astrocyte monolayers were then trypsinized and plated at an approximate density of 30,000 cells cm⁻² (6 or 24 cm well dishes and 90 cm² culture plates) and subcultured for 3–4 days before experiments. The cultures were immunocytochemically characterized by specific markers for different cell types. More than 95% of the cells reacted positively with the monoclonal anti Glial Fibrilar Acidic Protein (GFAP), a marker of astrocytes and less than 5% were A2B5 oligodendrocyte progenitors or Mac-1 positive microglia. Cultures with more than 5% of cells positive for A2B5 or Mac-1 were discarded.

Astrocyte stimulation

Cells were stimulated in 1 ml of growth media containing mouse recombinant IL-1β, LPS or TPA for different periods of time up to 24 h, and supernatants collected for PGE₂ determination. The protein synthesis inhibitor, cycloheximide, the COX-2 blocker NS-398 and the protein kinase inhibitors used were prepared as 1000 times stock solutions in DMSO and added 60 min before stimulation to the wells. Control cells were only exposed to equivalent amounts of DMSO.

For PKC or MAPK immunoblotting cells were serum starved for 24 h and then stimulated for the indicated times with mouse recombinant IL-1β or pharmacological agents as described in the Figure legends.

Western blot analysis

After culture in serum-free medium for 24 h, quiescent astrocytes were stimulated with test agents for the indicated times. Cells were washed with ice-cold PBS and lysed in 150 μl of TBS, pH 7.6, containing 10% glycerol, 1% NP-40, 0.5 mM EDTA, 1 mM PMSF Western blot, 50 μg ml⁻¹ leupeptin, 10 μg ml⁻¹ aprotinin, 5 mM benzamidine, 100 mM sodium ortovanate, 2 mM NaF and 5 mM DTT. Lysates were added 5× Laemmli sample buffer before boiling for 4 min. Then, equal amounts of protein (20 μg) were resolved on 10% SDS-polyacrylamide gel electrophoresis and electroblotted at 4°C for 1 h to ECL nitrocellulose. The membranes were blocked for 1 h at room temperature in 5% (w v⁻¹) dry skim milk (Sveltese, Nestlé) in TBS (pH 7.6) plus 0.1% Tween-20 (TTBS) and then blots rinsed in TTBS. The membranes were incubated overnight with activation state-specific antibodies for the MAPKs (rabbit anti-phospho p42/44 and p38, New England BioLabs) at 1 : 1000 dilution or specific COX-2 antibody (Cayman Chemicals, MI, U.S.A.) at 1 : 2000 dilution. After extensive washing in 5% milk TTBS solution, blots were incubated with peroxidase-conjugated goat anti-rabbit IgG1 at 1 : 10,000 dilution. Blots were washed again before developing with an enhanced chemiluminiscence (ECL) kit.

To determine changes in the subcellular distribution of PKC isoenzymes, the cells were fractionated into cytosol and membrane fractions according to the following protocol. Astrocytes grown into 100 cm² were stimulated with IL-1β for different times and harvested in 1 ml of ice-cold homogenization buffer (20 mM K-HEPES, pH 7.5, 5 mM EDTA, 5 mM EGTA, 1 mM PMSF, 50 μg ml⁻¹ leupeptin, 10 μg ml⁻¹ aprotinin, 5 mM benzamidine, 100 mM sodium ortovanate, 2 mM NaF and 5 mM DTT) and disrupted with a teflon homogenizer. The cytosolic fraction was obtained as supernatant after centrifugation at 100,000×g for 60 min. To prepare a membrane fraction, the pellets were resuspended in 400 μl of the same buffer plus 1% Triton X-100 and collected after centrifugation at 100,000×g for 30 min. Lysates (20 μg) were resolved on 10% SDS–PAGE and immunoblotted with rabbit polyclonal anti-PKC-α (1 : 6000) overnight at 4°C as described above.

RT–PCR analysis of COX-2

Astrocytes were plated in 35 mm culture dishes and stimulated with or without IL-1β (10 ng ml⁻¹) for different time periods. The cells were washed with PBS, and total RNA was isolated by the guanidinium isothiocyanate/phenol/chloroform method (Chomczynski & Sacchi, 1987). RNA concentration was quantified spectrophotometrically and the isolated RNA was treated with DNase to digest any contaminant genomic DNA. RT–PCR was performed in one step using Titan™ one tube RT–PCR system according to the manufacturer's instructions (Roche Molecular Biochemicals). RT–PCR amplification was carried out with 2 μg of RNA using the primer pair 5′-CCATGTCAAAACCGTGGTGAATG-3′ and 5′-ATGGGAGTTGGGCAGTCATAG-3′ (Nogawa et al., 1997). The steps of amplification were 94°C for 30 s, 55°C for 45 s, 68°C for 45 s for 30 cycles. PCR products were resolved on 2% agarose gel with 0.5 μg ml⁻¹ of ethidium bromide. Only one PCR product of the expected size was obtained and COX-2 transcript was identified as a 374 bp band. The mouse glyceraldehyde 3 phosphate dehydrogenase (GAPDH) gene (a houskeeping gene) was used as an internal standard gene.

Measurement of PGE₂

The cell-free supernatants were collected and stored at −70°C until measurement of PGE₂. The PGE₂ concentration in the culture medium was determined by a PGE₂ enzyme-immunoassay system (BIOTRAK, Amersham Pharmacia Biotech, U.K.). PGE₂ standard curve range from 2.5 to 320 pg well⁻¹. The sensitivity of the assay was 2 pg well⁻¹. The cross-reactivity with PGE₁ is 7.0%, with PGF₂α 0.04% and 6-Keto-PGF₁ α <0.1 and with Arachidonic acid <0.001%. The within assay precision was <7.6% and between assay was <14%.

The evaluation of the protein content in the cultures was determined by a BCA protein assay reagent (Pierce, IL, U.S.A.) with bovine serum albumin as the standard.

Statistical analysis

The data are presented as mean±s.e.mean values of n independent determinations, and were triplicated within each experiment. Comparisons were analysed by using one-way analysis of variance (ANOVA) followed by the posteriori Student-Newman-Keuls' t-test. A value of P<0.05 was considered significant.

Results

IL-1β stimulates prostaglandin biosynthesis and induces the expression of COX-2 mRNA and protein in mouse cultured astrocytes

Treatment of astrocyte cultures with IL-1β (10 ng ml⁻¹) caused a time-dependent increase in PGE₂ release, that was nearly 2 fold and more than 3 fold increased in comparison to control cultures at 8 and 12 h respectively (Figure 1). Exposure of the cells to IL-1β also induced the expression of mRNA COX-2 as well as COX-2 protein (Figure 2A–C). The kinetic profile of the expression of COX-2 mRNA by RT–PCR analysis revealed a time-dependent induction of COX-2 mRNA by IL-1β. COX-2 mRNA was detected as early as 2 h, peaked at 4 h and then, started to decline by 12 h (Figure 2A). The earliest induction of COX-2 protein occurred at 2 h and remained elevated up to 12 h (Figure 2B).

Figure 1.

Time course of IL-1β (10 ng ml⁻¹) induced release of PGE₂ from cultured mouse astrocytes in the presence or not of dexamethasone (1 μM). Supernatants were collected at the indicated times. Levels of PGE₂ in the medium were determined by enzyme-immunoassay system. Data are mean±s.e.mean (bars) values for three to four independent experiments in triplicate. Statistical significance: ^*P<0.001 compared with non-treated and dexamethasone treated cells after the corresponding incubation period.

Figure 2.

(A) Time-course of COX-2 mRNA expression in IL-1β (10 ng ml⁻¹)-cultured astrocytes. Total RNA was extracted as described in Methods and levels of mRNA for COX-2 and GAPDH were determined by RT–PCR. Representative results of three independent experiments. (B) Time course of COX-2 protein expression in IL-1β (10 ng ml⁻¹)-stimulated astrocytes. Cell lysates were used to determine COX-2 expression by Western blot analysis. Proteins were analysed by SDS-polyacrylamide gel electrophoresis and immunoblotting techniques using specific COX-2 antibody. Equal amount of proteins (20 μg) were loaded in all lanes. Bands were visualized by the ECL method. Representative results of three independent experiments. (C) Integrated band densities were obtained by scanning using a densitometer. Plotted means±s.e.mean of three independents experiments.

Dexamethasone (1 μM) prevented IL-1β-induced PGE₂ release (Figure 1), while both PMA (100 nM) and LPS (1 μg ml⁻¹) increased PGE₂ secretion after 24 h treatment (Figure 3). As expected, the inactive phorbol 4-αPMA (100 nM) failed to stimulate PGE₂ release. Figure 3 also shows that 2 μM of NS-398, a specific inhibitor of COX-2, (Futaki et al., 1993), totally abolished the stimulatory action of IL-1β on PGE₂ secretion, clearly indicating that prostaglandin production is due to COX-2 activity. In addition, PGE₂ production was completely prevented by the protein synthesis inhibitor cycloheximide (10 μg ml⁻¹), as well as by the transcription inhibitor actinomycine D (1 μg ml⁻¹) confirming the requirement for the de novo protein synthesis.

Figure 3.

Interleukin-1β (10 ng ml⁻¹), PMA (100 nM) and LPS (1 μg ml⁻¹) stimulate the production of PGE₂ in murine astrocytes. Supernatants were collected after 24 h stimulation. Pre-treatment (60 min before) with the COX-2 inhibitor, NS-398 (2 μM), or the inhibitor of the synthesis protein, cycloheximide (Cyclo) (10 μg ml⁻¹) or the transcription inhibitor actinomycine-D (1 μg ml⁻¹) abrogates the stimulatory effect of IL-1β on PGE₂ secretion. The data are presented as the mean±s.e.mean (vertical lines) of four independent determinations in triplicate. Statistical significance: ^*P<0.0001 vs control; NS-398+IL-1β, Cyclo+IL-1β and Acti+IL-1β.

Signal transduction pathways involved in IL-1β-increased prostaglandin production

Involvement of PKC and MAP kinases

The ability of PMA to increase PGE₂ production in astrocytes, suggests that PKC may be involved in the action of IL-1β. To test this hypothesis, astrocytes were pre-treated with various kinase inhibitors at the adequate doses prior to stimulation with IL-1β (10 ng ml⁻¹). As shown in Figure 4, the compound, H7 (10 μM), or bisindolylmaleimide (2 μM), inhibitors of PKA and PKC suppressed PGE₂ accumulation by 52.4 and 51.8%, respectively. The more specific PKC inhibitor, calphostin C (1 μM), also blocked (64.5%) IL-1β-stimulated PGE₂ production. Since MAP kinase pathways have been shown to be activated by IL-1β treatment in a number of cells, we also used specific inhibitors of p38 and p42/44 (ERK1/2) MAP kinase cascade. As shown in Figure 4, PGE₂ production was inhibited by 77% in IL-1β-stimulated astrocytes exposed to the specific inhibitor of ERK kinase (MEK), PD 98059 at the dose of 15 μM, known to inhibit ERK phosphorylation (Alessi et al., 1995). This dose of PD 98059 did not induce cell death during the 24 h treatment, as examined by the MTT reduction assay (data not shown). In addition, IL-1β-induced-PGE₂ release was blunted by 15 μM SB 203580 (79.4%), a specific inhibitor of p38 MAP kinase (Badger et al., 1996), without affecting cellular viability. These results indicate the involvement of both, PKC and MAPKs in IL-1β-induced PGE₂ biosynthesis by astroglial cells.

Figure 4.

The protein kinase C inhibitors (10 μM H-7, 2 μM BIS and 1 μM Caphostin C) as well as MAPKinase inhibitors (15 μM PD 98059, 15 μM SB 203580) blocked IL-1β-induced PGE₂ production. Inhibitors were added 60 min before IL-1β (10 ng ml⁻¹) stimulation and supernatants were collected 24 h after for PGE₂ determination by enzyme-immunoassay. Values are the means±s.e.mean (vertical bars) of four independent experiments in triplicate. Statistical significance: ^*P<0.0001 vs IL-1β alone; ⁺P<0.0001 vs control.

To further confirm a role of PKC on PGE₂ biosynthesis, astrocyte cultures were depleted of PKC by overnight treatment with PMA (100 nM), then IL-1β (10 ng ml⁻¹) or PMA (10 nM) were added for 4 h and the amount of PGE₂ in the culture supernatants measured. As shown in Figure 5, PKC-depleted cells no longer responded to PMA or IL-1β, suggesting the requirement for a conventional PKC isoform. A potential candidate is cPKC-α, which is present in astrocytes at high level. Therefore, astrocyte cultures were treated with IL-1β for different periods of time and cell extracts were analysed for cPKC-β activation. Immunoblot analysis (Figure 6) showed that IL-1β (10 ng ml⁻¹) caused translocation of cPKC-α from cytosol to the membrane fraction. This effect occurred rapidly, after 5 min, was maximal after 30 min, and declined after 6 h treatment. As expected, TPA treatment also caused the translocation of PKC-α from cytosol to the membrane fraction (data not shown).

Figure 5.

Depletion of PKC by overnight PMA pre-treatment prevents PGE₂ secretion by IL-1β or PMA. Astrocytes were treated with 100 nM PMA overnight and then were stimulated with IL-1β (10 ng ml⁻¹) or PMA (10 nM) for 4 h to collect supernatants for PGE₂ determination by enzyme-immunoassay. Results are means ±s.e.mean of four independent experiments. ^*P<0.001 vs control.

Figure 6.

Subcellular distribution of cPKC-α in response to activation with IL-1β. Cultured astrocytes were stimulated with 10 ng ml⁻¹ IL-1β, for 0, 5, 30 min, 1 h, or 6 h, and fractionated into soluble cytosolic or membrane fractions as described in Methods. Then, cytosolic and membrane fractions were immunoblotted with antibody against PKC-α. The experiments were repeated at least four times with similar results. Bands were visualized by the ECL method. Densitometric values were plotted, and values represent the means±s.e.mean of four experiments. ^•P<0.01; ^••P<0.001 vs basal.

MAPKs activation in astrocyte cultures by IL-1β was assessed by Western blot analysis, using specific antibodies for the phosphorylated forms of p38 and ERK1/2. Figure 7 shows the time course of IL-1β activation of p38 MAPK. IL-1β (10 ng ml⁻¹) elicited a rapid (detectable at 5 min) Thr/Tyr phosphorylation of p38 MAPK that increased with time, reaching a maximum at 30 min–1 h, followed by a gradual decline reaching basal levels by 5 h. Next, we examined whether the p42/44 MAPK pathway was also involved in IL-1β signalling in astrocytes. Figure 8 shows the time course activation of p42 and p44 MAPKs in response to IL-1β (10 ng ml⁻¹). IL-1β induced a rapid (5 min) and prolonged p42/44 MAPKs activation (over 5 h in the case of p44 MAPK), peaking at 30 min. Besides, IL-1β-induced phosphorylation of p42/44 MAPK was significantly inhibited by prior treatment (30 min) with the PKC inhibitor, bisindolylmaleimide (Figure 9), suggesting that IL-1β induced p42/44 MAPK activation may be the downstream signal of PKC activation. Collectively, these results suggest that both, activation of PKC and then, the phosphorylation of p42/44 and p38 MAPKs are required for the production of PGE₂ by IL-1β in murine astrocytes.

Figure 7.

Time-dependent activation of p38 MAPK by IL-1β in astrocytes. Astrocyte cultures were treated with 10 ng ml⁻¹ IL-1β for various time periods. Whole-cell lysates were prepared and subjected to immunoblot analysis using antibodies specific for the activated (phosphorylated) forms of p38 MAPK as described in Methods. Parallel blots were assayed using the antibodies recognizing the total p38 MAPK protein. The experiments were repeated at least four times with similar results. Bands were visualized by the ECL Method. Densitometric values were plotted, and values represent the means ±s.e.mean of four experiments. ^•P<0.01; ^••P<0.001 vs basal.

Figure 8.

Time-dependent activation of p42/44 MAPKs by IL-1β in astrocytes. Astrocyte cultures were treated with 10 ng ml⁻¹ IL-1β for various time periods. Whole-cell lysates were prepared and subjected to immunoblot analysis using antibodies specific for the activated (phosphorylated) forms of p42/44 MAPKs as described in Methods. Parallel blots were assayed using the antibodies recognizing the total p42/44 MAPK proteins. The experiments were repeated at least four times with similar results. Bands were visualized by the ECL method. Densitometric values were plotted, and values represent the means ±s.e.mean of four experiments. ^•P<0.05; ^••P<0.001 vs corresponding basal. ^*P<0.001 vs corresponding basal.

Figure 9.

The inhibition of PKC activity by bisindolylmaleimide (2 μM) reduced the phosphorylation of p42/44 MAPK induced by IL-1β (10 ng ml⁻¹). Cells were treated with bisindolylmaleimide (Bis) 30 min prior to stimulation with IL-1β. Whole cell lysates were prepared and subjected to immunoblot analysis using a specific antibody for the phosphorylated form of p42/44 MAPK. Representative results of three independent experiments with similar results. Bands were visualized by the ECL method. Densitometric values were plotted, and values represent the means±s.e.mean of three experiments. p42 MAPK: ^•P<0.05 vs corresponding basal; ^○○P<0.001 vs corresponding IL-1β. p44 MAPK: ^*P<0.001 vs corresponding basal; ^⋆P<0.001 vs corresponding IL-1β.

Discussion

The pro-inflammatory cytokine IL-1β enhanced PGE₂ production in mouse astrocyte cultures, in agreement with previous studies (Hartung et al., 1989; Katsuura et al., 1989; Gerbicke-Haerter, et al., 1991). Our study demonstrated that IL-1β-mediated PGE₂ release is a consequence of COX-2 induction, which is expressed by inflammatory agents and down-regulated by glucocorticoid hormones (O'Banion et al., 1992; Herschman, 1996). Thus, IL-1β treatment induced COX-2 mRNA expression as measured by RT–PCR, and COX-2 protein as shown by Western blot analysis; while the stimulatory action of IL-1β in PGE₂ release was abolished by the COX-2 selective inhibitor, NS-398, and by the glucocorticoid, dexamethasone. Furthermore, IL-1β-induced PGE₂ release was found to depend on the de novo transcription and protein synthesis since both actinomycine D and cycloheximide blocked the response.

However, the signalling pathways implicated in IL-1β-induced COX-2 expression and, thus, PGE₂ biosynthesis are not well understood. Cellular responses to IL-1β include a cascade of protein phosphorylation which transmits signals from cell surface to the nucleus and that ultimately regulates gene expression (O'Neill & Greene, 1998). The PKC family of serine/threonine kinases have been implicated in the signalling pathway of cell surface receptors, including those used by IL-1β in several cell types. Consistent with this, activation of PKC has been previously involved in IL-1-regulated astrocyte function (Aloisi et al., 1994; Norris et al., 1994; Ballestas & Benveniste, 1995). We observed a stimulatory effect of PMA on PGE₂ production, suggesting a positive role of PKC, in agreement with other studies (Hartung et al., 1989). Further evidence for the involvement of PKC was obtained using H-7 and bisindolylmaleimide, since these protein kinases inhibitors significantly reduced PGE₂ production in response to IL-1β. In addition, a more specific PKC inhibitor, calphostin C, at the dose of 1 μM almost completely blocked PGE₂ responses to IL-1β. Since calphostin C interacts with the regulatory domain of PKC and competes with the DAG phorbol ester binding site located in the C1 domain (Hug & Sarre, 1993; Hofman, 1997), our results suggest the possible involvement of a conventional PKC isoform in the signal transduction leading to the increased PGE₂ release by IL-1β. Consistent with this, down-regulation of PKC abolished the stimulatory effect of PMA, but also that of IL-1β. Moreover, Western blot analysis showed that IL-1β induced translocation of the conventional isoform, PKC-α, from cytosol to the membrane fraction indicating its activation. Nevertheless, because of lack of using selective inhibitors for PKC-α (Hofmann, 1997) in the present work, the possibility of involvement of other PKC isoenzymes in COX-2 expression after IL-1β treatment cannot be discounted.

Cellular prostaglandin production depends on cyclo-oxygenase activity as well as the availability of arachidonic acid substrate, which is controlled by enzymes such as phospholipase A₂ (PLA₂). Previous studies (Oka & Arita, 1991) have shown that enhanced PGE₂ production by IL-1β is subsequent to PLA₂ activation in rat astrocytes. Furthermore, activation of PKC stimulates the release of arachidonic acid due to increased PLA₂ activity which also implicated MAPK activation (Nam et al., 1995). Thus, it is possible that enhanced PGE₂ release in PMA-stimulated astrocytes also involved PLA₂ activation.

Recent work has linked activation of the prostaglandin generative pathway with the MAPK pathway (Guan et al., 1998; Niiro et al., 1998). Moreover, one of the signalling mechanisms triggered by IL-1β includes the activation of MAPKs, a family of serine/threonine kinases activated by dual phosphorylation of Thr and Tyr within a Thr-X-Tyr motif (Kyriakis & Avruch, 1996; Guan et al., 1997). The kinase cascade involving MAPKs appears to have a fundamental role in integrating multiple intracellular signals activated by Tyr kinases, but also by PKC (Blumer & Johnson 1994). Therefore, we investigated whether IL-1β stimulation was able to activate the ERK1/2 and p38 subgroups of MAPKs and whether such activation could be related to the ability of IL-1β to activate PKC and induce the biosynthesis of PGE₂. Our results indicate for the first time that IL-1β activates ERK1/2 and p38 MAPK in astroglial cells. The activation occurs rapidly at 5 min and with maximum at 30 min, indicating that the activation of MAPK cascade is an early event in astrocytes in response to IL-1β stimulation. An important question was therefore whether IL-1β induced activation of MAPKs was related to PGE₂ production. To this end, treatment of astrocytes with the specific inhibitors of MEK and p38 MAPK (PD 98059 or SB 203580) dramatically decreased PGE₂ release by IL-1β, suggesting key roles for the kinases in this event. Support for this notion was provided by Guan et al. (1997) showing that IL-1β-mediates PGE₂ production and COX-2 expression with concomitant activation of p38 MAPK-mediated signalling pathways in human mesangial cells. Our study presents further evidence that the intracellular-signalling pathway activated by IL-1β involves p42/44 MAPK and requires PKC activation.

In conclusion, we propose that the phosphorylation of MAPKs might constitute the signalling pathway through which an activation of PKC by IL-1β leads to increased PGE₂ biosynthesis in astroglial cells. This pathway in astrocytes provides further evidence for the role of the MAPKs cascade in the mediation of prostaglandin biosynthesis and as a potential target for modulation of the inflammatory response.

Abbreviations

CNS

central nervous system

DMEM

Dulbecco's modified Eagle Medium

DMSO

dimethyl sulphoxide

DTT

dithiothreitol

ERK

extracellular signal regulated kinase

FCS

foetal calf serum

GAPDH

glyceraldehyde 3-phosphate dehydrogenase

GFPA

glial fibrilar protein acidic

H-7

1-(5-isoquinolinyl-sulphonyl)-2-methylpiperazine

HEPES

4-(2-hydroxyethyl)-1-piperazineethanesulphonic acid

IL-1

interleukin-1

LPS

lipopolysacharide

MAPK

mitogen activated protein kinase

MTT

3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide

PAGE

polyacrylamide gel electrophoresis

PGE₂

prostaglandin E₂

PKA

protein kinase A

PKC

protein kinase C

PMA

phorbol 12-myristate 13-acetate

4-α PMA

4-α phorbol 12-myristate 13 acetate

PCR

polymerase chain reaction

RT

reverse transcription

SDS

sodium dodecyl sulphate

Thr

threonine

TPA

12-O-tetradecanoylphorbol 13 acetate

Tyr

tyrosine