Cytosolic phospholipase A2 induction and prostaglandin E2 release by interleukin-1β via the myeloid differentiation factor 88–dependent pathway and cooperation of p300, Akt, and NF-κB activity in human rheumatoid arthritis synovial fibroblasts

Authors


Abstract

Objective

Cytosolic phospholipase A2 (cPLA2) is a rate-limiting enzyme that plays a critical role in the biosynthesis of eicosanoids. The aim of this study was to investigate the mechanisms underlying interleukin-1β (IL-1β)–induced cPLA2 expression in human rheumatoid arthritis synovial fibroblasts (RASFs).

Methods

Synovial tissue was obtained from patients with RA who were undergoing joint replacement surgery. In a mouse model of IL-1β–mediated inflammatory arthritis, neutrophil infiltration, bone erosion, and cPLA2 expression in ankle synovium were analyzed by immunohistochemistry. IL-1β–induced cPLA2 expression was determined by Western blotting, real-time polymerase chain reaction, and gene promoter assay using pharmacologic inhibitors and transfection with short hairpin RNAs or small interfering RNAs. The recruitment of NF-κB and p300 to the cPLA2 promoter was determined by chromatin immunoprecipitation assay. Prostaglandin E2 (PGE2) biosynthesis was evaluated by enzyme-linked immunosorbent assay.

Results

IL-1β–induced cPLA2 expression and PGE2 release were mediated through a myeloid differentiation factor 88 (MyD88)/c-Src–dependent matrix metalloproteinase (MMP)/heparin-binding epidermal growth factor (HB-EGF) cascade linking to transactivation of the EGF receptor (EGFR)/phosphatidylinositol 3-kinase (PI 3-kinase)/Akt, p300, and NF-κB p65 pathways. IL-1β also stimulated Akt phosphorylation and nuclear translocation. Activation of Akt eventually led to the acetylation of histone residues by phosphorylation and recruitment of p300 and enhanced its histone acetyltransferase activity on the NF-κB elements of the cPLA2 promoter. IL-1β–induced NF-κB transcriptional activity was mediated through a PI 3-kinase/Akt–dependent cascade. Up-regulation of cPLA2 by IL-1β increased PGE2 biosynthesis in RASFs.

Conclusion

IL-1β–induced cPLA2 expression is mediated through activation of the MyD88/c-Src, MMP/HB-EGF, EGFR/PI 3-kinase/Akt, p300, and NF-κB pathways. These results provide insights into the mechanisms underlying IL-1β–enhanced joint inflammatory responses in RA and may inspire new targeted therapeutic approaches.

Rheumatoid arthritis (RA) is a chronic autoimmune inflammatory disease characterized by thickening of the synovial lining layers and predominantly inflammatory responses in joint tissues. In the affected joints of patients with RA, inflammatory cells, including immune cells, infiltrate and produce many proinflammatory cytokines, such as interleukin-1β (IL-1β) and tumor necrosis factor α (TNFα), in the synovial fluid (1, 2). Additionally, TNFα and IL-1 are predominant targets in the treatment of RA due to activation of the MAP kinase (MAPK) pathway and transcription factor NF-κB, which induces numerous proinflammatory proteins, including cytosolic phospholipase A2 (cPLA2) and cyclooxygenase 2 (COX-2), leading to the production of various inflammation mediators such as prostaglandin E2 (PGE2) (3–5). PGE2 is the most widely produced prostanoid in the human body and contributes to pain and swelling during inflammation through induction of hyperalgesia and vasodilatation. PGE2 production by cultured RA synovial fibroblasts (RASFs) is induced by IL-1β, mediated through NF-κB– and MAPK-dependent induction of cPLA2 and COX-2 (2, 3). A recent study indicated that COX-2–selective inhibitors might increase the risk of heart attack, thrombosis, and stroke through an increase in thromboxane synthesis (6).

The PLA2 enzymes have been implicated in diverse pathologic conditions, including inflammation (7). Among these PLA2 enzymes, cPLA2 is an important rate-limiting enzyme that catalyzes arachidonic acid release and initiates the multistage process of biosynthesis of eicosanoids such as COX-2–derived PGs (7). The expression of cPLA2 is increased by several proinflammatory cytokines, including TNFα and IL-1β. IL-1β has been shown to induce cPLA2 expression through MAPK-dependent NF-κB activation in various inflammatory diseases (3, 8). Epidermal growth factor receptor (EGFR) transactivation and phosphatidylinositol 3-kinase (PI 3-kinase)/Akt have also been shown to regulate cPLA2 expression by TNFα (9). Moreover, the cPLA2 promoter contains NF-κB binding sites, which are regulated by cytokines through various signaling mechanisms (10). Direct inhibition of NF-κB reduces the severity of arthritis in animal models, confirming the importance of these signaling pathways in inflammatory arthritis (11). However, little is known about the mechanisms underlying IL-1β–induced cPLA2 expression in RASFs.

Transcription coactivators such as p300, containing histone acetyltransferase (HAT), act as signal integrators in facilitating the expression of inflammatory genes (12). The activity of HAT is tightly regulated by various stimuli, including TNFα and IL-1β (9, 12, 13). Moreover, previous studies have shown that PI 3-kinase/Akt stimulates p300 phosphorylation and subsequent NF-κB activation (14). In A549 cells, TNFα has been shown to increase the acetylation of histones H3 and H4 and the assembly of RNA polymerase II on the intercellular adhesion molecule 1 (ICAM-1) promoter via PI 3-kinase/Akt (15). However, whether IL-1β–induced cPLA2 expression is mediated through PI 3-kinase/Akt and p300 activation in RASFs remained unknown.

In the present study, we performed experiments to investigate the mechanisms of IL-1β–induced cPLA2 expression and PGE2 release in RASFs. We observed that IL-1β–induced cPLA2 expression and PGE2 biosynthesis were mediated through activation of the myeloid differentiation factor 88 (MyD88)/c-Src, matrix metalloproteinase (MMP)/heparin-binding EGF (HB-EGF)/EGFR/PI 3-kinase/Akt, p300, and NF-κB pathways.

MATERIALS AND METHODS

Materials.

Anti–phospho–c-Src, anti–phospho-EGFR, anti–phospho-Akt, and anti–phospho-IκBα antibodies were obtained from Cell Signaling Technology. Antibodies against cPLA2, p65, Akt, and phospho-p300 were obtained from Santa Cruz Biotechnology. Anti–acetyl histone H3 and H4 antibodies were obtained from Upstate Biotechnology. PP1, AG-1478, LY294002, GM-6001, GR343, SH-5, and Bay 11-7082 were obtained from Biomol. Anti–HB-EGF antibody was obtained from R&D Systems. Enzymes and other chemicals were obtained from Sigma.

Primary synovial cell culture.

Synovial tissue samples were obtained from patients with RA who underwent knee or hip surgery. All patients with RA fulfilled the 1987 American College of Rheumatology criteria for the classification of RA (16). Informed consent was obtained from all patients, and the experimental protocol was approved by the Institutional Review Board of Chang Gung Memorial Hospital. RASFs and human synoviocytes from ScienCell Research Laboratories were cultured as previously described (17). Experiments were performed using cells from passages 3–6.

BALB/c mice.

BALB/c mice ages 6–8 weeks were purchased from the National Laboratory Animal Centre (Taipei, Taiwan). Mice were maintained according to the Guidelines of the Animal Care Committee of Chang Gung University and the National Institutes of Health Guide for the Care and Use of Laboratory Animals. BALB/c mice were anesthetized by intraperitoneal injection of 200 μl of pentobarbital sodium (5 mg/ml). Mice were given a single dose of AG-1478 (10 μM), Bay 11-7082 (1 μM), or phosphate buffered saline (PBS) (as a control) 1 hour before intraarticular treatment with IL-1β (30 μg/kg) and were killed after 24 hours. To examine the cellular expression and localization of cPLA2 protein, immunohistochemical staining was performed on serial sections of the ankle joints, which were deparaffinized, rehydrated, and washed with PBS. Nonspecific binding was blocked by preincubation with PBS containing 5 mg/ml bovine serum albumin, for 1 hour at room temperature. The first section was incubated with an anti-cPLA2 antibody at 37°C for 1 hour and then with horseradish peroxidase–conjugated anti-rabbit IgG antibodies at room temperature for 1 hour. Bound antibodies were detected by incubation with 0.5 mg/ml of 3,3′-diaminobenzidine/0.01% hydrogen peroxide in 100 mM Tris HCl buffer, as chromogen (Vector). The second section was incubated with an antivimentin antibody for the positive localization and identification of SFs and observed using an optical microscope.

Western blot analysis.

RASFs were plated onto 12-well culture plates and shifted to serum-free Dulbecco's modified Eagle's medium (DMEM)–Ham's F-12 for 24 hours. Cells were incubated with different concentrations of IL-1β for the indicated time intervals. The cell lysates were analyzed by Western blotting, using an anti-cPLA2 antibody as previously described (18).

Transient transfection.

The plasmids encoding short hairpin RNA (shRNA) of EGFR, p110, p85, c-Src, and Akt were graciously provided by Dr. C. P. Tseng (Chang Gung University). The NF-κB promoter luciferase reporter construct (pκB-Luc) was from Clontech. All plasmids were prepared using Qiagen plasmid DNA preparation kits. RASFs were transfected with 1 μg/well of shRNA or small interfering RNAs (siRNA), using GeneJammer transfection reagent, for 3 hours. One milliliter of DMEM–Ham's F-12 containing 10% fetal bovine serum was added, followed by incubation for an additional 20 hours. The cells were washed twice with PBS and maintained in serum-free DMEM–Ham's F-12 for 12 hours before treatment with IL-1β. The transfection efficiency (∼60%) was determined by transfection with an EGFR plasmid.

Quantitative real-time reverse transcription–polymerase chain reaction (PCR).

RNA was extracted using TRIzol, and first-strand complementary DNA (cDNA) synthesis was performed with 1 μg of total RNA using SuperScript II Reverse Transcriptase (Invitrogen) according to the manufacturer's protocols. The primers and probes used for real-time PCR of human cPLA2α and GAPDH were obtained from Applied Biosystems. Each PCR was performed using 100 ng of cDNA, PCR Master Mix, and premade TaqMan gene expression assay components containing a FAM reporter dye at the 5′ end of the TaqMan probe and a nonfluorescent quencher at the 3′ end of the probe. Human GAPDH was used as a control to verify the quality of the cDNA template. Real-time PCR was performed and analyzed using a StepOnePlus quantitative PCR instrument (Applied Biosystems).

Isolation of the cell fraction.

Cells were harvested, sonicated for 5 seconds with a Misonix sonicator (output 1.5), and centrifuged at 8,000 revolutions per minute for 15 minutes at 4°C. The pellet was collected as the nuclear fraction. The supernatant was centrifuged at 14,000 rpm for 60 minutes at 4°C to yield the pellet (membrane fraction) and the supernatant (cytosolic fraction).

Immunofluorescence staining.

Cells were plated on 6-well culture plates with coverslips, shifted to serum-free DMEM–Ham's F-12 for 24 hours, and then incubated with IL-1β. Cells were fixed, permeabilized, and stained using an anti-p65 or anti-Akt antibody as previously described (19). The images were observed under an Axiovert 200 M fluorescence microscope (Zeiss).

Promoter luciferase assay.

For construction of the cPLA2 promoter luciferase plasmid, a human cPLA2α promoter region (∼1,674 bp) was PCR-amplified from human genomic DNA and inserted between the luciferase gene and the SV40 late poly(A) signal coding regions of luciferase plasmid pGL3 as the wild-type cPLA2α promoter plasmids. The cPLA2α promoter region was amplified by conventional PCR using the following primers: forward 5′-GGGGTACCAGAACGAACATGCCCTGCAGTATAGA-3′ and reverse 5′-GGAAGCTTGCTGACTTTAAGCAGCGAGG-3′. The DNA fragments were directly subcloned into pGL3 using Kpn I and Hind III. The vector sequence was confirmed by DNA sequencing and amplified using Qiagen plasmid DNA preparation kits. The activity of cPLA2-Luc was determined as previously described (12), using a luciferase assay system (Promega). Firefly luciferase activity was standardized to β-galactosidase activity.

Coimmunoprecipitation assay.

Cell lysates containing 1 mg of protein were incubated with 2 μg of anti-MyD88 antibody at 4°C for 24 hours, and then 10 μl of 50% protein A–agarose beads was added and mixed at 4°C for 24 hours. The immunoprecipitates were collected and washed thrice with lysis buffer without Triton X-100. Laemmli buffer (5×) was added, subjected to electrophoresis on 12% sodium dodecyl–polyacrylamide gels, and then blotted using an anti–c-Src or anti-MyD88 antibody.

Chromatin immunoprecipitation (ChIP) assay.

The ChIP assay was performed as previously described, with modifications (9, 18). Briefly, RASFs were crosslinked with 1% formaldehyde and washed 3 times with ice-cold PBS containing 1 mM phenylmethylsulfonyl fluoride and 1% aprotinin. Soluble chromatin was prepared using a ChIP assay kit (Upstate), and immunoprecipitation was performed using an anti–NF-κB p65, anti-p300, anti–acetyl histone H3 or H4, or IgG antibody. Following washes and elution, precipitates were heated to reverse crosslinking of DNA and protein. DNA fragments were purified by phenol–chloroform extraction and ethanol precipitation. The purified DNA was subjected to PCR amplification using the primers specific for the region containing the NF-κB binding sites present in the cPLA2 promoter (sense primer 5′-GAGACGGAGTCTCGCTCTGT-3′; antisense primer 5′-GTGGCTCACGCCTGTAATCC-3′). PCR fragments were analyzed on 2% agarose in 1× Tris−acetate−EDTA gel containing ethidium bromide.

Measurement of PGE2 release.

Cells were treated with IL-1β for the indicated time intervals. The media were collected, and PGE2 was assayed using a PGE2 enzyme immunoassay kit (Cayman Chemical).

Statistical analysis.

All of the data were estimated using GraphPad Prism software. Data are expressed as the mean ± SEM and were analyzed with one-way analysis of variance, with Bonferroni adjustment for multiple comparisons. P values less than 0.05 were considered significant.

RESULTS

IL-1β induces cPLA2 expression and PGE2 generation.

Human RASFs were characterized by immunofluorescence staining using an antibody specific for a fibroblast protein, vimentin, or an endothelial cell marker (von Willebrand factor) as a negative control (20). As shown in Figure 1A, nearly 100% of the primary cultured fibroblasts were found on synovial tissues, so that, in essence, the characteristics of human RASFs were closely similar to those of normal human synoviocytes purchased from ScienCell Research Laboratories.

Figure 1.

Interleukin-1β (IL-1β) induces cytosolic phospholipase A2 (cPLA2) expression and prostaglandin E2 (PGE2) release in rheumatoid arthritis synovial fibroblasts (RASFs). A, Primary cultured human RASFs and normal human synoviocytes were identified by immunofluorescence double staining using an antibody specific for the fibroblast protein marker (vimentin) or an endothelial cell marker (von Willebrand factor [vWF]) as a negative control. DAPI was used to stain the nucleus. B and C, RASFs and human synoviocytes, respectively, were incubated with various concentrations of IL-1β for the indicated times. The cell lysates were analyzed by Western blotting using an anti-cPLA2 or anti-GAPDH (control) antibody. D, Cells were incubated with IL-1β (30 ng/ml) for the indicated times. Cytosolic PLA2 mRNA was analyzed by quantitative real-time polymerase chain reaction. E, Cells were pretreated with or without AACOCF3 (AA; 3 μM) for 1 hour and then incubated with IL-1β (30 ng/ml) for the indicated time intervals or for 16 hours. The media were collected to determine the levels of PGE2. Values in B, D, and E are the mean ± SEM of 4 independent experiments. ∗ = P < 0.05 versus cells incubated with vehicle; # = P < 0.01 versus cells incubated with vehicle; Ψ = P < 0.05.

To determine the effects of IL-1β on cPLA2 gene expression and PGE2 biosynthesis, RASFs were incubated with various concentrations of IL-1β for various time periods. As shown in Figure 1B, IL-1β increased cPLA2 protein expression, in a time- and concentration-dependent manner. A significant increase was observed within 16 hours, and a maximal response was reached within 16–24 hours. Similar to the results observed with RASFs, IL-1β also induced cPLA2 expression in human synoviocytes (Figure 1C). However, the responses in human synoviocytes were less than those in RASFs. These results demonstrated that IL-1β could induce cPLA2 expression, with a similar trend in human synoviocytes and RASFs.

To examine whether IL-1β induces cPLA2 messenger RNA (mRNA) expression in RASFs, quantitative real-time PCR was performed. IL-1β induced cPLA2 mRNA accumulation in a time-dependent manner, with a maximal response within 6 hours (Figure 1D). Next, we determined the levels of PGE2 biosynthesis as a parameter of cPLA2 activity. We observed that IL-1β induced PGE2 synthesis, with a maximal response within 16 hours. Moreover, pretreatment of RASFs with AACOCF3 (3 μM), a cPLA2 activity inhibitor, significantly reduced IL-1β–induced PGE2 production (Figure 1E), suggesting the specificity of cPLA2 for the increase in IL-1β–evoked PGE2.

Involvement of MyD88/c-Src in IL-1β–induced cPLA2 expression.

The Toll/IL-1 receptor domain, MyD88 recruitment, and subsequent downstream signaling are critical to the production of mediators of inflammation (21). When we examined whether MyD88 participated in IL-1β–induced responses, we observed that transfection of RASFs with MyD88 siRNA significantly reduced MyD88 protein expression and IL-1β–induced cPLA2 expression (Figure 2A). Moreover, MyD88 is directly associated with kinase activity and tyrosine phosphorylation of c-Src (22).

Figure 2.

IL-1β induces cPLA2 expression via myeloid differentiation factor 88 (MyD88)–dependent c-Src phosphorylation. A, Cells were transfected with MyD88 or scrambled (scrb) small interfering RNA (siRNA) and then incubated with IL-1β for 16 hours. B, Cells were pretreated with PP1 for 1 hour or transfected with c-Src short hairpin RNA (shRNA) or empty vector and then incubated with IL-1β for 16 hours. C, Cells were pretreated with PP1 (10 μM) for 1 hour and then stimulated with IL-1β (30 ng/ml) for the indicated times. D, Cells were transfected with scrambled or MyD88 siRNA and then incubated with IL-1β (30 ng/ml) for the indicated times. Cell lysates were analyzed by Western blotting, using the indicated antibodies. E, Cells were stimulated with IL-1β (30 ng/ml) for the indicated times. Cell lysates were immunoprecipitated (IP) using an anti–c-Src or anti-MyD88 antibody, and the immunoprecipitates were analyzed by Western blotting using an anti–c-Src or anti-MyD88 antibody. Data in B and C are expressed as the mean of 3 independent experiments. # = P < 0.01 versus cells exposed to IL-1β alone. IB = immunoblotting (see Figure 1 for other definitions).

Thus, to assess the role of c-Src in IL-1β–induced responses, RASFs were pretreated with a c-Src inhibitor (PP1) or were transfected with c-Src shRNA. As shown in Figure 2B, pretreatment with an inhibitor of c-Src (PP1) inhibited IL-1β–induced cPLA2 expression, in a concentration-dependent manner. Moreover, transfection with c-Src shRNA reduced IL-1β–induced cPLA2 expression. In addition, as shown in Figure 2C, IL-1β stimulated c-Src phosphorylation in a time-dependent manner, which was attenuated by pretreatment with PP1. To further determine whether c-Src phosphorylation is dependent on MyD88 signaling, cells were transfected with scrambled or MyD88 siRNA and then incubated with IL-1β. Transfection with MyD88 siRNA attenuated IL-1β–stimulated c-Src phosphorylation (Figure 2D).

We also investigated the relationship between MyD88 and c-Src in IL-1β–induced responses. Cell lysates prepared from IL-1β–treated cells were immunoprecipitated using an anti–c-Src or anti-MyD88 antibody. The results showed that the expression of MyD88- or c-Src–associated protein was increased, in a time-dependent manner, by IL-1β in a c-Src– or MyD88-immunoprecipitated complex (Figure 2E), suggesting that IL-1β–induced cPLA2 expression is mediated through an MyD88-dependent c-Src pathway.

Involvement of EGFR transactivation in IL-1β–induced cPLA2 expression.

Thrombin has been shown to stimulate c-Src–dependent EGFR transactivation via an MMP/HB-EGF process (23). To determine whether the MMP/HB-EGF process is also involved in IL-1β–induced responses, cells were pretreated with a broad-spectrum MMP inhibitor (GM-6001) or an HB-EGF neutralizing antibody. As shown in Figure 3A, IL-1β–induced cPLA2 expression was reduced by pretreatment with GM-6001 or an HB-EGF neutralizing antibody. To determine whether EGFR was involved in these responses, cells were pretreated with an EGFR inhibitor (AG-1478) or were transfected with EGFR shRNA. As shown in Figure 3B, pretreatment with AG-1478 or transfection with EGFR shRNA significantly attenuated cPLA2 induction by IL-1β.

Figure 3.

IL-1β–induced cPLA2 expression is mediated via matrix metalloproteinase/heparin-binding epidermal growth factor (HB-EGF)–dependent transactivation of EGF receptor (EGFR)/phosphatidylinositol 3-kinase (PI 3-kinase)/Akt. A–E, Cell lysates were analyzed by Western blotting using the indicated antibodies. A, Cells were pretreated with GM-6001 or 10 μg/ml HB-EGF neutralizing antibody (nAb) for 1 hour and incubated with IL-1β for 16 hours. B, Cells were pretreated with AG-1478 for 1 hour or transfected with EGFR short hairpin RNA (shRNA) and incubated with IL-1β for 16 hours. C, Cells were pretreated with HB-EGF nAb (10 μg/ml), PP1 (10 μM), GM-6001 (GM; 10 μM), CRM197 (CRM; 1 μg/ml), or AG-1478 (AG; 10 μM) for 1 hour and stimulated with IL-1β for the indicated times or for 5 minutes. D, Cells were pretreated with LY294002 (LY) for 1 hour or transfected with PI 3-kinase subunit p110 or subunit p85 shRNA and incubated with IL-1β for 16 hours. E, Cells were pretreated with HB-EGF nAb, PP1, GM, CRM, AG, LY (30 μM), or SH-5 (1 μM) for 1 hour or transfected with Akt shRNA and incubated with IL-1β for the indicated times or for 10 minutes and 6 hours. F, Cytosolic PLA2 mRNA expression was analyzed by quantitative real-time polymerase chain reaction. Values are the mean ± SEM of 3 independent experiments. # = P < 0.01 versus cells exposed to IL-1β alone. See Figure 1 for other definitions.

To further confirm whether IL-1β transactivates EGFR via an MMP/HB-EGF shedding process, cells were pretreated with an HB-EGF neutralizing antibody, PP1, GM-6001, CRM197, or AG-1478 and stimulated with IL-1β. As shown in Figure 3C, IL-1β stimulated time-dependent EGFR phosphorylation, with a maximal response observed within 5 minutes, which was attenuated by pretreatment with HB-EGF neutralizing antibody and the inhibitors of c-Src (PP1), MMPs (GM-6001), HB-EGF (CRM197), and EGFR (AG-1478), suggesting that the c-Src–dependent MMP/HB-EGF cascade is involved in IL-1β–stimulated EGFR transactivation and cPLA2 expression.

PI 3-kinase and Akt have been shown to be downstream components of EGFR that participate in various cellular functions (9, 24). Therefore, we investigated the role of PI 3-kinase/Akt in IL-1β–mediated responses. Pretreatment with a PI 3-kinase inhibitor (LY294002) inhibited IL-1β–induced cPLA2 expression, in a concentration-dependent manner (Figure 3D). To further confirm that PI 3-kinase is involved in IL-1β–induced cPLA2 expression, cells were transfected with either p110 or p85 shRNA. Transfection with either p110 or p85 shRNA significantly reduced IL-1β–induced cPLA2 expression, suggesting that PI 3-kinase is essential for IL-1β–induced cPLA2 expression.

To determine whether IL-1β–induced responses are mediated through Akt phosphorylation, cells were pretreated with HB-EGF neutralizing antibody, PP1, GM-6001, CRM197, AG-1478, LY294002, or SH-5. As shown in Figure 3E, IL-1β stimulated Akt phosphorylation in a time-dependent manner, which was attenuated by pretreatment with HB-EGF neutralizing antibody, PP1, GM-6001, CRM197, AG-1478, LY294002, or SH-5. Transfection with Akt shRNA also reduced Akt protein expression and IL-1β–induced cPLA2 expression (Figure 3E). Additionally, pretreatment with these inhibitors attenuated IL-1β–induced cPLA2 mRNA expression in RASFs (Figure 3F). These results indicated that in RASFs, IL-1β–induced cPLA2 expression is mediated through a c-Src–dependent MMP/HB-EGF/EGFR/PI 3-kinase/Akt cascade.

Requirement for NF-κB in IL-1β–induced cPLA2 expression.

Proinflammatory stimuli induce a rapid and transient translocation of NF-κB into the nucleus, where it activates transcription of several inflammatory genes such as cPLA2 (25, 26). Thus, we investigated whether NF-κB is involved in IL-1β–induced cPLA2 expression. Pretreatment with an NF-κB inhibitor (Bay 11-7082) attenuated IL-1β–induced cPLA2 protein and mRNA expression (Figure 4A), suggesting that NF-κB is required for IL-1β–induced cPLA2 expression. IL-1β also time-dependently stimulated phosphorylation of IκBα and translocation of p65 NF-κB (Figure 4B). There was significant phosphorylation of IκBα within 3–5 minutes and, following initial nuclear translocation of p65 NF-κB, within 10–60 minutes. Pretreatment with AG-1478, LY294002, or SH-5 had no effect on IL-1β–induced NF-κB translocation (Figure 4B).

Figure 4.

NF-κB is essential for IL-1β–induced cPLA2 expression. A, RASFs were pretreated with Bay 11-7082 (Bay) for 1 hour and incubated with IL-1β for 16 hours (cPLA2 protein expression) or 6 hours (mRNA expression). B, RASFs were pretreated with AG-1478 (AG; 10 μM), LY294002 (LY; 30 μM), or SH-5 (5 μM) for 1 hour and stimulated with IL-1β for the indicated times or for 30 minutes. The nuclear and cytosolic extracts were prepared. Cell lysates were analyzed by Western blotting using the indicated antibodies, with lamin A used as an internal control. C and D, RASFs were transfected with an NF-κB promoter gene (pκB-Luc), pretreated without (C) or with (D) 10 μM PP1, 10 μM GM-6001 (GM), 1 μg/ml CRM197 (CRM), 10 μM AG-1478 (AG), 30 μM LY294002 (LY), or 1 μM SH-5 for 1 hour, and incubated with IL-1β for the indicated times (C) or for 4 hours (D). Luciferase activity was analyzed. Bars show the mean ± SEM of 3 independent experiments. ∗ = P < 0.01; # = P < 0.01 versus cells exposed to IL-1β alone. E, Immunohistochemical staining for cPLA2 and vimentin and hematoxylin and eosin (H&E) staining in serial sections of ankle joints from phosphate buffered saline–treated mice (sham; a–d), IL-1β–injected mice (IL-1β; e–l), AG-1478–pretreated mice (AG/IL-1β; m–p), and Bay 11-7082–pretreated mice (Bay/IL-1β; q–t) are shown. Microscopic observation of the joints of IL-1β–treated mice showed erosive destruction and neutrophil infiltration in the articular cavity (arrows in e and i, respectively), vimentin in the synovial membrane of ankle joints (arrow in f), and synovial fibroblasts overlapping with cPLA2 expression (arrows in g, k, h, and l). b = bone; c = cartilage; ss = synovial surface; js = joint space (see Figure 1 for other definitions).

To further determine whether IL-1β regulated NF-κB transcriptional activity through these signaling molecules, transfection with a promoter reporter plasmid containing the NF-κB binding site (pκB-Luc) was performed. As shown in Figures 4C and D, IL-1β increased NF-κB transcriptional activity, with a maximal response within 4 hours and a sustained response over 6 hours, which was attenuated by pretreatment with PP1, GM-6001, CRM197, AG-1478, LY294002, and SH-5, suggesting that IL-1β induced NF-κB transcriptional activity via a c-Src–dependent MMP/HB-EGF/EGFR/PI 3-kinase/Akt pathway.

Confirming these results in vivo, IL-1β–treated mouse ankle joints showed marked synovial and periarticular inflammation, with articular erosion (Figure 4E, part e) and invasion of inflammatory cells, including lymphocytes and neutrophils (Figure 4E, part i). In contrast, administration of AG-1478 or Bay 11-7082 significantly inhibited IL-1β–induced infiltration of these cells in articular joints (Figure 4E, parts m and q). Moreover, in IL-1β–treated mice, the synovial layer strongly expressed cPLA2 (Figure 4E, parts g and h), which was reduced by pretreatment with AG-1478 or Bay 11-7082 (Figure 4E, parts o and s). These data demonstrated that IL-1β–induced cPLA2 expression occurs in vitro and in vivo and is dependent on EGFR and NF-κB.

IL-1β induces cPLA2 expression via Akt-dependent p300 activation.

Nuclear translocation of activated Akt regulates the downstream targets involved in up-regulation of several genes (27, 28). We observed that IL-1β stimulated the nuclear translocation of phosphorylated Akt, which was attenuated by pretreatment with LY294002 or SH-5 (data not shown). Activation of Akt by TNFα also promotes the transcriptional activity of p300 and leads to the acetylation of histone residues (9, 15). The transcription coactivator p300/CREB binding protein plays a central role in coordinating multiple signal-development events with the transcription apparatus, leading to gene expression by diverse stimuli (29). Thus, to examine the effect of p300 on IL-1β–induced cPLA2 expression, an anti–phospho-p300 (Ser89) antibody was used.

As shown in Figure 5A, IL-1β stimulated time-dependent phosphorylation of p300, which was inhibited by pretreatment with the inhibitor of PI 3-kinase (LY294002) or the inhibitor of Akt (SH-5), suggesting that IL-1β stimulates p300 phosphorylation via a PI 3-kinase/Akt cascade. To confirm whether Akt is associated with p300, the cell nuclear fraction was immunoprecipitated using an anti-Akt or anti-p300 antibody. As shown in Figures 5B and C, IL-1β stimulated the association between Akt and p300, which was blocked by pretreatment with LY294002 or SH-5. Moreover, transfection with p300 siRNA reduced p300 protein expression and IL-1β–induced cPLA2 expression (Figure 5D). These results indicated that PI 3-kinase/Akt-dependent p300 phosphorylation was involved in IL-1β–induced cPLA2 expression.

Figure 5.

IL-1β induces cPLA2 expression in RASFs via phosphatidylinositol 3-kinase/Akt–dependent p300 phosphorylation. A, Cells were pretreated without (control) or with 30 μM LY294002 (LY) and 1 μM SH-5 for 1 hour and incubated with IL-1β for the indicated times. B, Nuclear extracts were immunoprecipiated (IP) using an anti-p300 (top) or anti-Akt (bottom) antibody after IL-1β stimulation for the indicated times. C, Cells were pretreated with 30 μM LY294002 or 1 μM SH-5 for 1 hour and incubated with IL-1β for 60 minutes. The nuclear extracts were immunoprecipitated and analyzed by Western blotting using an anti-Akt or anti-p300 antibody. D, Cells were transfected with p300 or scrambled small interfering (siRNA) and incubated with IL-1β for 16 hours. Cell lysates were analyzed by Western blotting using the indicated antibodies. Values are the mean of 3 independent experiments. ∗ = P < 0.01 versus cells exposed to IL-1β alone. See Figure 1 for other definitions.

Role of c-Src–dependent EGFR transactivation in p65 and p300 activation, cPLA2 promoter activity, and PGE2 release.

The promoter region of cPLA2 contains NF-κB binding sites for gene expression (30). P300 and CREB binding protein are highly homologous coactivators that promote gene transcription by bridging between DNA-binding transcription factors and the basal transcription machinery, by providing a scaffold for integrating transcription factors, and by modifying transcription factors and chromatin through acetylation (30–32). Therefore, we investigated the role of PI 3-kinase/Akt in the recruitment of p300 and NF-κB induced by IL-1β. Chromatin was immunoprecipitated using an anti-p65, anti-p300, or anti–acetyl histone H3 or H4 antibody, and the region (−1,243 to −983 bp) of the NF-κB binding site within cPLA2 promoter (Figure 6A) was amplified by PCR. As shown in Figure 6A, IL-1β stimulated recruitment of p65 and p300 and the enrichment of acetylated H3 and H4 within the cPLA2 promoter region, in a time-dependent manner. Recruitment and activation of p300 were attenuated by pretreatment with LY294002, SH-5, or a p300 inhibitor (GR343) (Figure 6B), indicating that IL-1β stimulated the recruitment and activation of p300 via the PI 3-kinase/Akt cascade. IL-1β–stimulated recruitment of NF-κB p65 was also blocked by pretreatment with PP1, CM6001, CRM197, AG-1478, LY294002, and SH-5 (Figure 6C). The results demonstrated that IL-1β–stimulated recruitment of NF-κB p65 was mediated through c-Src–dependent transactivation of the EGFR/PI 3-kinase/Akt pathway.

Figure 6.

IL-1β–mediates responses via a c-Src–dependent epidermal growth factor receptor (EGFR)/phosphatidylinositol 3-kinase (PI 3-kinase [PI3K])/Akt cascade. A and B, RASFs were pretreated without or with, respectively, 30 μM LY294002 (LY), 1 μM SH-5, or 1 μM GR343 for 1 hour and incubated with IL-1β for the indicated times or for 2 hours. C, Cells were pretreated with PP1, GM-6001 (GM), CRM197 (CRM), AG-1478 (AG), LY294002 (LY), or SH-5 for 1 hour and incubated with IL-1β for 4 hours. Chromatin was immunoprecipitated (IP) using an anti-p65, anti-p300, anti–acetylated histone H3 or H4, or IgG antibody. The associated cPLA2 promoter DNA was amplified by polymerase chain reaction (PCR). Input represents PCR products from chromatin pellets before immunoprecipitation. D, RASFs were transfected with a cPLA2-Luc reporter gene, pretreated without or with PP1, GM-6001, CRM197, AG-1478, LY294002, SH-5, Bay 11-7082 (Bay), or GR343 for 1 hour, and incubated with IL-1β (30 ng/ml) for the indicated times (inset) or for 6 hours. Promoter luciferase activity was analyzed. E, The PGE2 levels in the media were analyzed using a PGE2 enzyme immunoassay kit. Bars show the mean ± SEM of 3 separate experiments. ∗ = P < 0.05; # = P < 0.01 versus cells exposed to IL-1β alone. F, Schematic diagram showing that IL-1β induces cPLA2 expression via the myeloid differentiation factor 88 (MyD88), c-Src, matrix metalloproteinase (MMP)/heparin-binding EGF (HB-EGF), EGFR/PI 3-kinase/Akt pathways and the recruitment of Akt, NF-κB, and p300 to the cPLA2 promoter region and induces gene expression in RASFs. See Figure 1 for other definitions.

Moreover, we constructed a human cPLA2 promoter luciferase reporter plasmid (pcPLA2-Luc) to investigate the effect of IL-1β on the regulation of cPLA2 promoter activity. The data showed that IL-1β increased cPLA2 promoter activity in a time-dependent manner (Figure 6D), and this was attenuated by pretreatment with PP1, GM-6001, CRM197, AG-1478, LY294002, SH-5, and Bay 11-7082 (Figure 6D), indicating that cPLA2 promoter activity was up-regulated by IL-1β through the c-Src–dependent MMP/HB-EGF/EGFR/PI 3-kinase/Akt/p300 and NF-κB pathways. Pretreatment with these inhibitors also attenuated IL-1β–induced PGE2 synthesis (Figure 6E), suggesting that the c-Src–dependent transactivation of EGFR/PI 3-kinase/Akt/p300 and NF-κB participates in IL-1β–induced cPLA2 expression and PGE2 production in RASFs.

DISCUSSION

Cytosolic PLA2 has been shown to play a key role in RA pathogenesis initiated by IL-1β (33). The kinetics study of the induction of cPLA2 suggests that up-regulation of this enzyme contributes to PGE2 production, which is associated with disease severity (2, 3). It has been proposed that IL-1β induces cPLA2 expression and PGE2 biosynthesis in RASFs (2). Here, we demonstrated that IL-1β induces cPLA2 expression, leading to PGE2 production in RASFs and human synoviocytes, which was attenuated by the inhibitors of c-Src, MMPs, HB-EGF, EGFR, PI 3-kinase, Akt, NF-κB, and p300 HAT or transfection with their respective shRNA or siRNA. Furthermore, activation of NF-κB transcriptional activity was repressed by these inhibitors. However, inhibition of PI 3-kinase/Akt had no effect on NF-κB translocation. We observed that activation of PI 3-kinase/Akt induced NF-κB binding and transcriptional activity through the regulation of p300 HAT activity. These results suggest that activation of MyD88, c-Src, MMP/HB-EGF, EGFR/PI 3-kinase/Akt, p300, and NF-κB by IL-1β is crucial for cPLA2 up-regulation and PGE2 biosynthesis in RASFs.

The MyD88 adaptor molecule plays a crucial role in IL-1–induced NF-κB activation (1). Recent studies suggested that c-Src mediates hypoxia-induced IL-6 production via an MyD88-dependent pathway (21). Here, we showed that transfection with MyD88 siRNA or Src shRNA reduced IL-1β–induced cPLA2 expression. Furthermore, IL-1β–induced cPLA2 expression was mediated through c-Src phosphorylation. This observation was confirmed by transfection with MyD88 siRNA, which attenuated IL-1β–induced cPLA2 expression, suggesting that IL-1β–induced responses were mediated through MyD88/c-Src–dependent signaling. We further demonstrated that IL-1β–induced cPLA2 expression was mediated by the formation of a MyD88/c-Src complex. Although the specific protein–protein interactions between MyD88 and c-Src are not known, our results are the first to show a novel role of MyD88/c-Src complex formation in IL-1β–induced responses in RASFs. Future studies are planned to determine which domains of MyD88 and c-Src are involved in protein–protein interactions induced by IL-1β.

G protein–coupled receptor (GPCR)–mediated PI 3-kinase/Akt activation is critically dependent on the tyrosine kinase activity of EGFR via a c-Src–dependent mechanism in various cell types (23, 34). Transactivation of EGFR may be attributable to the release of EGFR ligands such as HB-EGF, which is regulated by MMPs (34, 35). Although EGFR transactivation by GPCR agonists has been well studied, little was known about the mechanisms regulated by IL-1β in RASFs. Our data show that IL-1β induced cPLA2 expression via an EGFR/PI 3-kinase/Akt signaling cascade. Moreover, we observed that IL-1β–induced EGFR transactivation was mediated through MMPs and HB-EGF in addition to c-Src, since the activation of EGFR and Akt by IL-1β was attenuated by pretreatment with GM-6001, CRM197, or anti–HB-EGF antibodies, consistent with reports indicating that MMPs are responsible for HB-EGF shedding upon GPCR stimulation (36–38). These results are consistent with those of previous studies indicating that IL-1β–induced ICAM-1 expression in A549 cells and MMP-9 expression in brain astrocytes are mediated through c-Src–dependent transactivation of EGFR or platelet-derived growth factor receptor (12, 39).

Inflammatory responses during exposure to cytokines are highly dependent on activation of NF-κB, which plays an important role in the expression of inflammatory genes (10, 40, 41). Activation of NF-κB is regulated by various signaling cascades, including the IKK signalosome (42, 43). IKK phosphorylates IκBα by several stimuli, resulting in its ubiquitination and subsequent proteasomal degradation (44). In this study, we observed that the increase in NF-κB p65 translocation correlated with the phosphorylation of IκBα in the cytosol of RASFs treated with IL-1β. The translocation of NF-κB p65 following IL-1β exposure was inhibited by Bay 11-7082, consistent with the finding that IL-1β–induced cPLA2 expression is mediated through NF-κB activation in canine tracheal smooth muscle cells (18).

In addition to IL-1β mediating induction of cPLA2 by MAPK-dependent NF-κB activation in various inflammatory diseases, EGFR transactivation and PI 3-kinase/Akt have been shown to activate NF-κB and lead to IL-1β–induced cPLA2 expression (9, 45, 46). Here, we observed that pretreatment with AG-1478, LY294002, or SH-5 attenuated IL-1β–induced NF-κB promoter activation but had no effect on NF-κB translocation in RASFs. These results are consistent with reports that the inhibition of PI 3-kinase/Akt had no effect on TNFα- or IL-1–induced IκBα degradation and NF-κB translocation in various cell types (15, 45, 46). We further demonstrated that IL-1β–induced cPLA2 expression was mediated through c-Src, MMP/HB-EGF, EGFR/PI 3-kinase/Akt, and NF-κB signaling, consistent with our recent studies indicating that IL-1β–induced ICAM-1 expression or thrombin-regulated COX-2 expression was mediated via similar signaling mechanisms including EGFR transactivation (39). However, IL-1β–induced NF-κB transcriptional activity in RASFs was different from that described in previous studies (23, 39).

TNFα has been shown to increase the acetylation of histones H3 and H4 and the assembly of RNA polymerase II on the ICAM-1 promoter via Akt-dependent phosphorylation of p300 in A549 cells (15). In RASFs, we also demonstrated that IL-1β–induced NF-κB transcriptional activity was mediated through Akt-dependent recruitment and activation of p300 HAT and augmented NF-κB–mediated cPLA2 transcription.

Because release of free arachidonic acid by cPLA2 represents the rate-limiting step for PG synthesis, coordination between cPLA2 and COX-2 is a prerequisite for PGE2 formation during inflammation. In this study, we demonstrated that IL-1β induced PGE2 production in a cPLA2-dependent manner in RASFs, by using a selective cPLA2 inhibitor (AACOCF3). This result is consistent with those of previous studies indicating that cPLA2-derived PGE2 is important for the integrity and function of vascular smooth muscle cells by up-regulation of COX-2 (47, 48).

In conclusion, we have demonstrated that IL-1β induces cPLA2 expression via the MyD88/c-Src, MMP/HB-EGF, EGFR/PI 3-kinase/Akt cascade, linking to activation of NF-κB and p300 HAT activity, which results in up-regulation of the cPLA2 gene in RASFs. Based on reports in the literature and our findings, Figure 6F depicts a model for the molecular mechanisms underlying IL-1β–induced cPLA2 expression and PGE2 release in human RASFs. The mechanisms by which IL-1β induces cPLA2 expression and PGE2 release in RASFs provide an important link between the pathogenesis and risk of RA and may inspire development of new targeted therapeutic approaches.

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Yang had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Chi, Luo, Hsieh, Hsiao, Chen, Yang.

Acquisition of data. Chi, Lee, Hsiao, Chen, Yang.

Analysis and interpretation of data. Chi, Luo, Hsieh, Lee, Hsiao, Yang.

Ancillary