Eicosapentaenoic acid inhibits interleukin-6 production in interleukin-1β-stimulated C6 glioma cells through peroxisome proliferator-activated receptor-gamma

ARTICLE IN PRESS Prostaglandins, Leukotrienes and Essential Fatty Acids 79 (2008) 59–65

Contents lists available at ScienceDirect

Prostaglandins, Leukotrienes and Essential Fatty Acids journal homepage: www.elsevier.com/locate/plefa

Eicosapentaenoic acid inhibits interleukin-6 production in interleukin-1b-stimulated C6 glioma cells through peroxisome proliferator-activated receptor-gamma Akiko Kawashima, Tsuyoshi Harada , Kazunori Imada, Takashi Yano, Kiyoshi Mizuguchi Development Research, Pharmaceutical Research Center, Mochida Pharmaceutical Company Limited, 722 Uenohara, Jimba, Gotemba, Shizuoka 412-8524, Japan

a r t i c l e in fo

abstract

Article history: Received 18 May 2008 Received in revised form 23 July 2008 Accepted 30 July 2008

Epidemiological studies suggest that intake of omega-3 polyunsaturated fatty acids improves neurological disorders such as Alzheimer’s disease which exhibit inflammatory pathology. We therefore investigated the anti-inflammatory effects of eicosapentaenoic acid (EPA) on interleukin (IL)-1bstimulated C6 glioma cells. In the present study, EPA inhibited pro-inflammatory cytokine IL-6 production, a characteristic of certain neurodegenerative disorders, in IL-1b-stimulated C6 glioma cells in dose-dependent fashion. EPA down-regulated the expression of IL-6 at mRNA level, indicating that the effect of EPA occurs at the transcriptional level. In addition, peroxisome proliferator-activated receptor (PPAR) g antagonists abolished the inhibitory effect of EPA on IL-1b-induced IL-6 production, whereas PPARa antagonist did not block the inhibitory effect of EPA. EPA might thus contribute to the regulation of pro-inflammatory cytokine production in astrocytes through interaction with PPARg. Among the PPARg ligands tested in this study, ciglitazone, a synthetic agonist of PPARg, effectively inhibited IL-6 production, but while neither rosiglitazone nor 15-deoxy-D12,14-prostaglandin J2 did. These findings indicate that the coordination of PPAR gamma ligands is important in inhibiting the production of IL-6 in C6 glioma cells. & 2008 Elsevier Ltd. All rights reserved.

1. Introduction Omega-3 polyunsaturated fatty acids (PUFAs) such as eicosapentaenoic acid (20:5n-3, EPA) and docosahexaenoic acid (22:6n-3, DHA) are known to have anti-inflammatory effects. It has been reported that dietary n-3 PUFA supplementation rich in EPA decreases production of pro-inflammatory cytokines such as interleukin (IL)-1b, IL-6, and tumor necrosis factor (TNF)-a in both healthy subjects and patients with rheumatoid arthritis [1,2]. Consistent with these anti-inflammatory effects, n-3 PUFAs have

Abbreviations: AA, arachidonic acid; AP, activator protein; BSA, bovine serum albumin; C/EBP, CCAAT/enhancer binding protein; CNS, central nervous system; COX, cyclooxygenase; CREB, CAMP-responsive element-binding protein; DHA, docosahexaenoic acid; DMEM, Dulbecco’s modified Eagle’s medium; DMSO, dimethyl sulphoxide; DPBS, Dulbecco’s phosphate-buffered saline; ELISA, enzyme-linked immunosorbent assay; EPA, eicosapentaenoic acid; FBS, fetal bovine serum; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; IL, interleukin; NF-kB, nuclear factor-kB; OA, oleic acid; 15d-PGJ2, 15-deoxyD12,14-prostaglandin J2; PPAR, peroxisome proliferator-activated receptor; PUFAs, polyunsaturated fatty acids; RT-PCR, reverse transcription-polymerase chain reaction; SA, stearic acid; STAT, signal transducer and activator of transcription; TNF, tumor necrosis factor  Corresponding author. Tel.: +81 550 88 2449; fax: +81 550 89 8070. E-mail address: [email protected] (T. Harada). 0952-3278/$ - see front matter & 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.plefa.2008.07.002

been reported to improve the prognosis of several chronic inflammatory diseases including rheumatoid arthritis [3,4], IgA nephropathy [5], inflammatory bowel disease, asthma, and psoriasis [4]. Inflammatory processes also play critical roles in the neurodegenerative disorders. An increased number of activated astrocytes and microglia as well as pro-inflammatory cytokines produced by these cells are one of the hallmarks of the inflammatory pathology of neurodegenerative diseases [6]. Indeed, IL-6 immunoreactivity has been demonstrated in amyloid plaques, the neuropathological hallmark of Alzheimer’s disease, in brains of patients [7]. Suppression of increase in inflammatory reactions of glial cells including IL-6 production has thus been proposed as a molecular target for therapeutic intervention in these diseases [8,9]. EPA has been reported to have anti-inflammatory neuroprotective properties in the hippocampus in an animal model of Alzheimer’s disease, attenuating the amyloid-b-induced increase in IL-1b [10]. This EPA-associated suppression of IL-1b concentration in cultured rat hippocampal microglial cells has been suggested to be mediated by peroxisome proliferator-activated receptor (PPAR) g, a member of the nuclear receptor superfamily of ligand-dependent transcription factors which plays a critical

ARTICLE IN PRESS 60

A. Kawashima et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 79 (2008) 59–65

role in the regulation of differentiation, lipid and glucose homeostasis, and inflammatory responses [10,11]. A series of PUFAs including EPA and eicosanoid derivatives such as 15-deoxy-D12,14prostaglandin J2 (15d-PGJ2), a cyclooxygenase (COX) product of arachidonic acid (20:4n-6, AA), are known to be natural ligands for PPARg [12]. Nonetheless, the inhibitory effects of EPA on astrocyte reactions, in particular cytokine production, through PPARg signaling remain poorly understood. The present study was therefore performed to compare the anti-inflammatory effects of EPA in astrocytes with those of various types of fatty acids and PPARg ligands, and to examine the mechanisms by which EPA reduces inflammatory responses. For this purpose, we evaluated the effects of EPA on pro-inflammatory cytokine production in rat C6 glioma cells, which have the property of glial cells and express PPARg and have been used for the studies of IL-6 production as a model of inflammatory neurodegenerative disorders [13,14].

2. Materials and methods 2.1. Materials EPA, oleic acid (OA), and AA (all sodium salts) were purchased from Nu-chek Prep (Elysian, MN, USA). Stearic acid (SA), DHA, and MK-886 (all sodium salts) were purchased from Sigma-Aldrich (St. Louis, MO, USA). T0070907, GW9662, ciglitazone, and 15d-PGJ2 were purchased from Cayman Chemical (Ann Arbor, MI, USA). Rosiglitazone was purchased from Alexis (Lausen, Switzerland). The C6 glioma cell line was obtained from the American-Type Culture Collection (Manassas, VA, USA). Recombinant rat IL-1b was obtained from R&D Systems (Minneapolis, MN, USA). IL-1b was dissolved directly in DMEM at a concentration of 10 mg/mL. SA, PPARg antagonists, PPARa antagonist, and synthetic PPARg agonists were dissolved in dimethyl sulphoxide (DMSO, SigmaAldrich). 15D-PGJ2 (a solution in methyl acetate) was dried under a stream of nitrogen and resolved in DMSO at a concentration of 20 mM. 2.2. C6 glioma cell culture C6 glioma cells were maintained in a humidified atmosphere of 5% CO2–95% air at 37 1C in Dulbecco’s modified Eagle’s medium (DMEM, Invitrogen, Carlsbad, CA, USA) supplemented with 10% heat-inactivated fetal bovine serum (FBS, Moregate Biotech, Bulimba, Australia) and antibiotics (100 mg/mL streptomycin, 100 IU/mL penicillin). Cells were removed from the tissue culture flasks with 0.05% trypsin-EDTA (Sigma-Aldrich) in Dulbecco’s phosphate-buffered saline (DPBS, Sigma-Aldrich), resuspended in DMEM supplemented with 10% FBS and antibiotics, and dispersed into 24-well tissue culture plates (Costar, Corning, Lowell, MA, USA) at a density of 5  105 cells/well (for the IL-6 release studies) or six-well tissue culture plates (Costar) at a density of 2.5  106 cells/well [for real-time reverse transcription-polymerase chain reaction (RT-PCR) studies]. The cells in the plates were allowed to adhere for 24 h, rinsed with serum-free DMEM, and then incubated for 24 h with serum-free DMEM. Then fresh solutions of the fatty acids dissolved in fatty acid-free bovine serum albumin (BSA, Merck, Darmstadt, Germany) and serum-free DMEM at a concentration of 50 mM (unless otherwise indicated) in the presence of IL-1b at 50 ng/mL were added to the cells for 24 h. Control conditions consisted of cells treated with identical concentrations of BSA and IL-1b in serum-free DMEM. In some experiments, C6 cells were simultaneously exposed to PPAR

antagonist in the presence of IL-1b and EPA. For the IL-6 release studies, conditioned media was removed and centrifuged at 12,000 rpm for 5 min. The supernatant was collected and stored at 80 1C pending analysis for IL-6. For the real-time RT-PCR studies, total RNA was immediately isolated from the cells. EPA and other fatty acids stock solutions in 5% BSA/DMEM were diluted in DMEM under an argon atmosphere so that the final concentration of BSA became 0.5%. In the case of SA, PPARg antagonists, PPARa antagonist, and synthetic PPARg agonists, final concentrations of DMSO were 0.2%, 0.1%, 0.1%, and 0.1%, respectively. 2.3. Enzyme-linked immunosorbent assay (ELISA) IL-6 levels were measured with an ELISA kit (Invitrogen). Experiments were performed according to the manufacturer’s instructions. 2.4. Cell viability assay Cell viability was monitored using an assay kit (CellTiter-GloTM luminescent cell viability assay, Promega, Madison, WI, USA). Procedures were performed according to the manufacturer’s instructions. In brief, after the medium for the IL-6 release studies was removed, DMEM and CellTiter-GloTM reagent were added to the cells. After mixing for 2 min on a shaker to induce cell lysis, cells were incubated at room temperature for 10 min in the dark to stabilize luminescence signal. The luminescent signal, which was proportional to the amount of adenosine 50 -triphosphate present in viable cells, was measured with a luminometer (Wallac 1420 ARVO SX Multilabel Counter, Perkin-Elmer, Waltham, MA, USA). 2.5. Cell lysate preparation In a separate experiment, after whole medium (500 mL) was removed for the IL-6 release studies, cells were washed twice with DPBS and suspended in 500 mL of ice-cold lysis buffer that contained 50 mM Tris–HCl (pH 7.5) (Sigma-Aldrich) and 0.1% Triton X-100 (Sigma-Aldrich) supplemented with complete protease inhibitor cocktail (Roche, Basel, Switzerland), followed by high-speed centrifugation at 13,000 rpm for 5 min at 4 1C. The supernatant was taken as the cell lysate for IL-6 measurement. 2.6. Isolation of total RNA and quantitative real-time RT-PCR Total RNA was isolated from C6 glioma cells using TRIzol reagent (Invitrogen) and an RNeasy Mini Kit (QIAGEN, Hilden, Germany) according to the manufacturer’s instructions. Total RNA (4 mg) from each sample was converted to cDNA using the SuperScript III first-strand synthesis system (Invitrogen) according to the manufacturer’s instructions. cDNA (0.25 mg) was mixed with 0.1 mL of 20 mM primers for IL-6 or glyceraldehyde-3phosphate dehydrogenase (GAPDH) and 12.5 mL of SYBR Green PCR master mix (Applied Biosystems, Foster City, CA, USA). The volume was made up to 25 mL with diethyl pyrocarbonate-treated water. The primer sequences were as follows: IL-6-F, ACTATGAGGTCTACTCGGCAAACC; IL-6-R, CACAGTGAGGAATGTCCACAAAC; GAPDH-F, GCTACACTGAGGACCAGGTTGTCT; GAPDH-R, CCCAGCATCAAAGGTGGAA. Quantitative real-time RT-PCR was performed on an ABI PRISM 7000 Sequence Detection System (Applied Biosystems) with the following thermal cycler conditions: 1 cycle at 50 1C for 2 min and 95 1C for 10 min, followed by 40 cycles of denaturation at 95 1C for 15 s and annealing at 60 1C for 60 s. The rat GAPDH gene was used as an internal control to evaluate relative expression of IL-6.

ARTICLE IN PRESS A. Kawashima et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 79 (2008) 59–65

2.7. Statistical analysis Comparisons between two groups were performed using Student’s t-test or Aspin–Welch t-test. Comparisons between more than two groups were performed using the Dunnett

61

test. SAS statistical programs were used (Version 9.1.3, SAS Institute Japan, Tokyo, Japan). Differences with Po0.05 were considered significant. Unless noted otherwise, values are expressed as the mean7SD of groups consisting of three to four observations.

250 *

IL-6 (pg/mL)

200

***

150

*** 100 50

Cell viability {Luminescence (x105 relative light units)}

ND 0 IL-1β EPA (μM) 0

+ 0

+ 12.5

+ 25

+ 50

+ 100

+ 0

+ 12.5

+ 25

+ 50

+ 100

25

20

15

10

5

0 IL-1β EPA (μM) 0 500

IL-6 (pg/mL)

400 *** ***

300

***

200 100 ND 0 IL-1β FA

-

+ -

+ SA

+ OA

+ AA

+ EPA

+ DHA

Fig. 1. Effects of EPA on IL-1b-stimulated IL-6 production in and viability of C6 glioma cells, and comparison of the effects of different types of fatty acids (50 mM) on IL-1bstimulated IL-6 production. C6 glioma cells were treated with IL-1b at 50 ng/mL in the absence or presence of EPA (A and B) or the indicated fatty acid (C) for 24 h. (A and C) IL-6 concentrations in cell culture supernatants were determined by ELISA. (B) Cell viability was examined by CellTiter-GloTM luminescent cell viability assay. Significant differences from the value for cells treated with IL-1b in the absence of fatty acids are marked, *Po0.05, ***Po0.001 (Dunnett test, n ¼ 4). ND: not detected.

ARTICLE IN PRESS 62

A. Kawashima et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 79 (2008) 59–65

3. Results

350 3.1. EPA attenuates IL-6 production in IL-1b-stimulated C6 glioma cells

300 250 IL-6 (pg/mL)

We examined the levels of IL-6 in IL-1b-stimulated C6 glioma cell culture supernatants. Exposure to IL-1b (50 ng/mL) for 24 h induced IL-6 production (218.4713.7 pg/mL, Fig. 1A). EPA at concentrations of 25, 50, and 100 mM significantly and dosedependently inhibited IL-1b-induced IL-6 production (189.57 12.0, 134.3715.8 and 111.6712.5 pg/mL, respectively). Since EPA has been reported to induce tumor cell apoptosis [15], we examined the effects of EPA on the viability of C6 glioma cells. To determine viability, cells were cultured with EPA at final concentrations of 12.5, 25, 50 and 100 mM for 24 h, and then CellTiter-GloTM luminescent cell viability assay was carried out. Cell viability did not decrease in the presence of EPA (Fig. 1B). Cells treated with IL-1b in the absence of EPA were used as the control; control conditions were found not to affect cell viability. These findings suggest that the inhibitory effects of EPA on IL-6 production induced by IL-1b were probably not toxic effects of EPA.

Extracellular

*** 200

***

150 100 50 ND 0 IL-1β

-

+

+

+

EPA (μM) 0

0

50

100

1.8

Intracellular †††

1.6 *

3.3. EPA suppresses both extracellular and intracellular IL-6 protein levels Since it was uncertain whether the inhibitory effects of EPA on IL-1b-stimulated IL-6 production were due to the inhibitory effects of EPA on the step of IL-6 secretion, we next examined the levels of IL-6 in IL-1b-stimulated C6 glioma cell lysates as well as cell culture supernatants. Exposure to IL-1b (50 ng/mL) for 24 h significantly increased extracellular as well as intracellular IL-6 levels (Fig. 2A and B). EPA at concentrations of 50 and 100 mM significantly decreased both extracellular and intracellular IL-1b-induced IL-6 protein levels. These findings suggest that the inhibitory effects of EPA on IL-6 production induced by IL-1b were probably not inhibitory effects of EPA on IL-6 secretion from the cytoplasm into medium. 3.4. EPA suppresses IL-6 mRNA level We examined the levels of IL-6 mRNA in IL-1b-stimulated C6 glioma cells. Exposure to IL-1b (50 ng/mL) for 24 h significantly induced IL-6 mRNA (Fig. 2C). EPA at concentrations of 50 and 100 mM significantly and dose-dependently inhibited IL-1binduced IL-6 mRNA production. These findings indicate that the

IL-6 (pg/μg protein)

In order to examine the effects of other fatty acids on IL-1b-stimulated IL-6 production, we examined the levels of IL-6 in IL-1b-stimulated C6 glioma cells treated with either the saturated fatty acid SA, monounsaturated fatty acid OA, n-6 PUFA AA, or n-3 PUFAs EPA and DHA at concentrations of 50 mM. AA, EPA, and DHA each significantly inhibited IL-1b-induced IL-6 production (317.7715.7, 283.7715.5 and 315.4711.5 pg/mL, respectively, compared to 464.7730.4 pg/mL for control) (Fig. 1C). On the other hand, SA and OA did not inhibit this production. Cell viability did not decrease in the presence of fatty acids at concentrations of 50 mM (data not shown). In addition, the vehicle (0.2% DMSO), which was used in the case of SA, did not significantly alter cell viability (data not shown). These findings suggest that EPA inhibited IL-1b-induced IL-6 production most effectively among the fatty acids tested.

**

1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 -

+

+

+

EPA (μM) 0

0

50

100

IL-1β

25

†

20 IL-6/GAPDH mRNA

3.2. n-3 PUFAs and n-6 PUFA attenuate IL-6 production in IL-1b-stimulated C6 glioma cells

15 * 10

**

5

0 IL-1β EPA (μM) 0

+ 0

+ 50

+ 100

Fig. 2. Effects of EPA on extracellular and intracellular IL-6 levels and IL-6 mRNA level in IL-1b-stimulated C6 glioma cells. C6 glioma cells were treated with IL-1b at 50 ng/mL in the absence or presence of EPA for 24 h. IL-6 concentrations in cell culture supernatants (A) or in cell lysate (B) were determined by ELISA. IL-6 mRNA expression (C) was examined by quantitative real-time RT-PCR. mRNA level of IL-6 was normalized to that of GAPDH. ySignificantly different from the value for cells without IL-1b and EPA, yPo0.05, yyyPo0.001 [Student’s t-test, n ¼ 4 (B), Aspin–Welch t-test, n ¼ 3 (C)]. Significant differences from the value for cells treated with IL-1b in the absence of EPA are marked, *Po0.05, **Po0.01, ***Po0.001 [Dunnett test, n ¼ 4 (A and B), n ¼ 3 (C)]. ND: not detected.

ARTICLE IN PRESS A. Kawashima et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 79 (2008) 59–65

63

600

IL-6 (pg/mL)

500 400 ** 300 *** 200 100 ND 0 IL-1β EPA antagonist (μM) -

+ -

+ + -

+ + 5

+ + 10 T0070907

+ + 20

900 800

IL-6 (pg/mL)

700 600 500 **

***

*** ***

400

***

300 200 100 ND 0 IL-1β

antagonist (μM) EPA

+ -

+ + -

+ + 10

+ + 25 GW9662

+ + 50

+ + 10

+ + 25

+ + 50

MK-886

Fig. 3. Effects of PPARg antagonists and PPARa antagonist on the inhibitory effect of EPA in IL-1b-stimulated C6 glioma cells. C6 glioma cells were treated with IL-1b at 50 ng/mL in the absence or presence of 50 mM EPA for 24 h. Some cultures were incubated with T0070907 (A), GW9662, or MK-886 (B) in the presence of 50 ng/mL IL-1b and 50 mM EPA. IL-6 concentrations in cell culture supernatants were determined by ELISA. Significant differences from the value for cells treated with IL-1b in the absence of EPA are marked, ** Po0.01, *** Po0.001 (Dunnett test, n ¼ 4). ND: not detected.

effect of EPA on IL-1b-stimulated IL-6 release may involve reduction of transcription of IL-6 mRNA in C6 glioma cells.

effect of EPA. These findings suggest that EPA exerts antiinflammatory effects through a PPARg-dependent pathway.

3.5. PPARg-specific antagonists block the inhibitory effect of EPA 3.6. EPA and ciglitazone inhibit IL-1b-stimulated IL-6 production It has been reported that EPA is a ligand of PPARg and PPARa [16]. We examined whether the effects of EPA on C6 glioma cells were dependent on PPARg and PPARa. Cells were incubated with PPARg-specific antagonists T0070907 or GW9662 or the PPARaspecific antagonist MK-886 in the presence of IL-1b (50 ng/mL) and EPA (50 mM), and levels of IL-6 in cell culture supernatants were measured. As can be seen in Fig. 3A and B, incubation of cultures with PPARg-specific antagonists, T0070907 or GW9662, significantly blocked the inhibitory effect of EPA on IL-6 production induced by IL-1b, whereas incubation of cultures with the PPARa-specific antagonist MK-886 did not block the inhibitory

Finally, in order to examine the effects of other PPARg ligands on IL-1b-stimulated IL-6 production, we examined the level of IL-6 in IL-1b-stimulated C6 glioma cell culture supernatants. Ciglitazone at concentrations of 1, 5 and 20 mM significantly and dose-dependently inhibited IL-1b-induced IL-6 production (Fig. 4). On the other hand, rosiglitazone was ineffective in inhibiting this production up to a concentration of 20 mM. In addition, 15d-PGJ2 stimulated IL-1b-induced IL-6 production in dose-dependent fashion. Cell viability did not decrease in the presence of PPARg ligands at concentrations of 20 mM (data not

ARTICLE IN PRESS 64

A. Kawashima et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 79 (2008) 59–65

700 *** 600 **

IL-6 (pg/mL)

500 400 *

300

*

**

200 ***

100 0

ND

IL-1β ligand (μM) -

+ -

+ 50 EPA

+ 1

+ 5

+ 20

Rosiglitazone

+ 1

+ 5

+ 20

Ciglitazone

+ 1

+ 5

+ 20

15d-PGJ2

Fig. 4. Effects of PPARg ligands on IL-1b-stimulated IL-6 production in C6 glioma cells. C6 glioma cells were treated with IL-1b at 50 ng/mL in the absence or presence of the indicated PPARg ligand for 24 h. IL-6 concentrations in cell culture supernatants were determined by ELISA. Significant differences from the value for cells treated with IL-1b in the absence of PPARg ligands are marked, *Po0.05, **Po0.01, ***Po0.001 (Dunnett test, n ¼ 4). ND: not detected.

shown). In addition, vehicle (0.1% DMSO) alone did not significantly alter cell viability (data not shown).

4. Discussion and conclusions In the present study, we demonstrated that EPA inhibited IL-1b-induced IL-6 production in C6 glioma cells. Previous studies of n-3 PUFAs have suggested that EPA might elicit a pleiotropic anti-inflammatory effect through modulation of several biological cascades, e.g. by PPARg activation [17,18], PPARa activation [19], nuclear factor-kB (NF-kB) signaling inhibition [20], reduction of prostaglandin E2 and leukotriene B4 production [1], or destabilization of surface presentation of lipopolysaccharide receptors [21]. Among these pathways, we have confirmed that PPARgspecific antagonists significantly block the inhibitory effect of EPA on IL-6 production in C6 glioma cells, whereas a PPARa antagonist did not block the inhibitory effect of EPA, suggesting that the antiinflammatory effect of EPA in astrocytes may be in a PPARgdependent manner. This is in agreement with previous suggestions that EPA increased PPARg expression in rat hippocampus [22], and that EPA attenuated amyloid-b-induced IL-1b production in PPARg-dependent fashion in cultured rat hippocampal microglia [23]. The present study also demonstrated that PUFAs (EPA, DHA, and AA), whether of n-3 or n-6 series, reduced IL-1b-induced IL-6 production, while a saturated fatty acid (SA) and a monounsaturated fatty acid (OA) were ineffective in doing so. These differences in effects of fatty acids on inflammation might be explained in part by their binding affinity to PPARg. Among the various fatty acids, PUFAs (such as AA, EPA, and DHA) interact most efficiently with PPARg, while monounsaturated fatty acids do so only weakly [16,24]. In addition, saturated fatty acids such as SA do not bind to PPARg. This ligand selectivity of PPARg might contribute to the anti-inflammatory activities of n-3 PUFAs and n-6 PUFA, and to the ineffectiveness of saturated and monounsaturated fatty acids in C6 glioma cells. Although PUFAs have binding affinity for PPARg, whether of n-3 or n-6 series, EPA might have stronger anti-inflammatory activity than other PUFAs in the central nervous system (CNS), since it can reduce

pro-inflammatory biological AA metabolites most effectively by competing with AA as a substrate for COX and lipoxygenase enzymes as well as by inhibiting phospholipase A2 activity [1,25,26]. We also found that, among the PPARg ligands tested in this study, EPA and ciglitazone inhibited IL-1b-stimulated IL-6 production in C6 glioma cells effectively, while rosiglitazone and 15d-PGJ2 did not. Since the efficacy of the PPARg synthetic ligands ciglitazone and rosiglitazone in inhibiting IL-6 production is not related to their reported binding affinities for PPARg [27], the antiinflammatory activity of liganded PPARg observed in the present study appears to be mediated by a signaling pathway, which does not require peroxisome proliferator response element-binding transcriptional activity. In fact, recent studies have shown that the anti-inflammatory effects of PPARg are mediated by transrepression, a form of negative transcriptional control of key transcription factors such as NF-kB at the non-genomic level [28]. In addition to NF-kB, three other transcription factor-binding sites exist in the IL-6 promoter, i.e., cAMP-responsive element-binding protein (CREB), CCAAT/enhancer binding protein (C/EBP), and activator protein (AP)-1 [29], and it has been reported that not only regulation of NF-kB but also regulation of C/EBP by PPARg is responsible for IL-1b-stimulated IL-6 induction [30]. It is thus possible that the anti-inflammatory effects of PPARg in astrocytes observed in the present study are mediated in part by suppression of the transcription factors, NF-kB and C/EBP. The molecular pathway by which PPARg represses transcriptional activation of the IL-6 gene in astrocytes is not yet fully understood. However, the reason for the functional diversity of effects of PPARg ligands on IL-1b-stimulated IL-6 production observed in the present study might involve the existence of distinct molecular pathways between PPARg ligands for negative regulation of transcription factors. Indeed, it has been reported that two structurally distinct PPARg ligands, 15d-PGJ2 and troglitazone, could activate PPARg to negatively regulate the IL-6-activated signal transducer and activator of transcription (STAT) 3 through distinct molecular pathways [31]. In brief, 15dPGJ2 inactivates STAT3 by enhancing the direct physiological protein–protein association between PPARg and IL-6-activated STAT3, whereas troglitazone inactivates STAT3 by inducing

ARTICLE IN PRESS A. Kawashima et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 79 (2008) 59–65

redistribution of co-repressor silencing mediator of retinoid and thyroid hormone receptors from PPARg to IL-6-activated STAT3. Three transrepression models, the tethering model, squelching model, and sumoylation model, have been proposed for nongenomic inhibition by PPARg of the transcription of inflammatory genes through antagonism of the activity of NF-kB [28]. For example, the nonpathogenic gut bacterium Bacteroides thetaiotaomicron has been reported to attenuate expression of proinflammatory cytokines such as TNF, IL-8, and COX-2 through the tethering pathway, which involves physical interaction of the liganded PPARg with NF-kB [32]. In this transrepression, tethering of NF-kB promotes translocation of the PPARg from the nucleus into the cytoplasm, resulting in nuclear export of p65 (the transcriptional regulatory subunit of NF-kB) and the subsequent anti-inflammatory effect. By contrast, rosiglitazone has been reported to negatively regulate the NF-kB target gene, inducible nitric oxide synthase, through the co-factor squelching pathway [33] or ligand-dependent sumoylation pathway [34]. It is thus possible that EPA as well as ciglitazone transrepress IL-b-induced IL-6 expression through the tethering pathway in C6 glioma cells, unlike the inhibitory pathway of rosiglitazone. Consistent with this, structurally distinct PPARg ligands have the potential to induce different conformational changes of PPARg. In brief, it has been found that rosiglitazone and pioglitazone induced distinct conformational changes in the co-activator-binding surface of PPARg, resulting in distinct effects on the interaction of PPARg with PPARg co-activator-1 [35]. Therefore, different conformational changes of PPARg might result in the diversity of posttranslational modifications such as sumoylation. Although further studies are needed to elucidate the molecular mechanisms responsible for the functional diversity of PPARg ligands for IL-b-induced IL-6 production in the CNS, the differences in inflammatory response between EPA and the other PPARg ligands might depend on the structural conformation of ligand–PPARg complexes. In conclusion, EPA protects against IL-1b-induced IL-6 production in C6 glioma cells through a PPARg-dependent pathway. EPA and ciglitazone may thus exhibit anti-inflammatory effects in the neurodegenerative diseases by suppressing the production of the cytokine through ligand-specific interaction with PPARg. Acknowledgements We would like to thank Mr. Yuji Akahori, Ms. Yuriko Okazaki, Ms. Yuri Kikuchi, and Mr. Ryo Saito for their skillful assistance. References [1] M.J. James, R.A. Gibson, L.G. Cleland, Dietary polyunsaturated fatty acids and inflammatory mediator production, Am. J. Clin. Nutr. 71 (2000) 343S–348S. [2] S.N. Meydani, S. Endres, M.M. Woods, et al., Oral (n-3) fatty acid supplementation suppresses cytokine production and lymphocyte proliferation: comparison between young and older women, J. Nutr. 121 (1991) 547–555. [3] J.M. Kremer, D.A. Lawrence, W. Jubiz, et al., Dietary fish oil and olive oil supplementation in patients with rheumatoid arthritis. Clinical and immunologic effects, Arthritis Rheum. 33 (1990) 810–820. [4] A.P. Simopoulos, Omega-3 fatty acids in inflammation and autoimmune diseases, J. Am. Coll. Nutr. 21 (2002) 495–505. [5] J.V. Donadio, J.P. Grande, The role of fish oil/omega-3 fatty acids in the treatment of IgA nephropathy, Semin. Nephrol. 24 (2004) 225–243. [6] A. Simi, N. Tsakiri, P. Wang, N.J. Rothwell, Interleukin-1 and inflammatory neurodegeneration, Biochem. Soc. Trans. 35 (2007) 1122–1126. [7] M. Huell, S. Strauss, B. Volk, M. Berger, J. Bauer, Interleukin-6 is present in early stages of plaque formation and is restricted to the brains of Alzheimer’s disease patients, Acta Neuropathol. 89 (1995) 544–551. [8] R.A. Gadient, U.H. Otten, Interleukin-6 (IL-6)—a molecule with both beneficial and destructive potentials, Prog. Neurobiol. 52 (1997) 379–390. [9] N.J. Van Wagoner, E.N. Benveniste, Interleukin-6 expression and regulation in astrocytes, J. Neuroimmunol. 100 (1999) 124–139.

65

[10] A.M. Lynch, D.J. Loane, A.M. Minogue, et al., Eicosapentaenoic acid confers neuroprotection in the amyloid-beta challenged aged hippocampus, Neurobiol. Aging 28 (2007) 845–855. [11] T.M. Willson, P.J. Brown, D.D. Sternbach, B.R. Henke, The PPARs: from orphan receptors to drug discovery, J. Med. Chem. 43 (2000) 527–550. [12] R. Kapadia, J.H. Yi, R. Vemuganti, Mechanisms of anti-inflammatory and neuroprotective actions of PPAR-gamma agonists, Front. Biosci. 13 (2008) 1813–1826. [13] J.M. Perez-Ortiz, P. Tranque, C.F. Vaquero, et al., Glitazones differentially regulate primary astrocyte and glioma cell survival. Involvement of reactive oxygen species and peroxisome proliferator-activated receptor-gamma, J. Biol. Chem. 279 (2004) 8976–8985. [14] J.W. Zumwalt, B.J. Thunstrom, B.L. Spangelo, Interleukin-1beta and catecholamines synergistically stimulate interleukin-6 release from rat C6 glioma cells in vitro: a potential role for lysophosphatidylcholine, Endocrinology 140 (1999) 888–896. [15] T. Shirota, S. Haji, M. Yamasaki, et al., Apoptosis in human pancreatic cancer cells induced by eicosapentaenoic acid, Nutrition 21 (2005) 1010–1017. [16] G. Krey, O. Braissant, F. L’Horset, et al., Fatty acids, eicosanoids, and hypolipidemic agents identified as ligands of peroxisome proliferatoractivated receptors by coactivator-dependent receptor ligand assay, Mol. Endocrinol. 11 (1997) 779–791. [17] C.K. Combs, D.E. Johnson, J.C. Karlo, S.B. Cannady, G.E. Landreth, Inflammatory mechanisms in Alzheimer’s disease: inhibition of beta-amyloid-stimulated proinflammatory responses and neurotoxicity by PPARgamma agonists, J. Neurosci. 20 (2000) 558–567. [18] G. Zhao, T.D. Etherton, K.R. Martin, et al., Anti-inflammatory effects of polyunsaturated fatty acids in THP-1 cells, Biochem. Biophys. Res. Commun. 336 (2005) 909–917. [19] B. Staels, W. Koenig, A. Habib, et al., Activation of human aortic smoothmuscle cells is inhibited by PPARalpha but not by PPARgamma activators, Nature 393 (1998) 790–793. [20] Y. Zhao, S. Joshi-Barve, S. Barve, L.H. Chen, Eicosapentaenoic acid prevents LPS-induced TNF-alpha expression by preventing NF-kappaB activation, J. Am. Coll. Nutr. 23 (2004) 71–78. [21] V. De Smedt-Peyrusse, F. Sargueil, A. Moranis, H. Harizi, S. Mongrand, S. Laye, Docosahexaenoic acid prevents LPS-induced cytokine production in microglial cells by inhibiting LPS receptor presentation but not its membrane subdomain localization, J. Neurochem. (2008). [22] T. Kavanagh, P.E. Lonergan, M.A. Lynch, Eicosapentaenoic acid and gammalinolenic acid increase hippocampal concentrations of IL-4 and IL-10 and abrogate lipopolysaccharide-induced inhibition of long-term potentiation, Prostaglandins Leukot. Essent. Fatty Acids 70 (2004) 391–397. [23] A.M. Minogue, A.M. Lynch, D.J. Loane, C.E. Herron, M.A. Lynch, Modulation of amyloid-beta-induced and age-associated changes in rat hippocampus by eicosapentaenoic acid, J. Neurochem. 103 (2007) 914–926. [24] H.E. Xu, M.H. Lambert, V.G. Montana, et al., Molecular recognition of fatty acids by peroxisome proliferator-activated receptors, Mol. Cell 3 (1999) 397–403. [25] M.J. Finnen, C.R. Lovell, Purification and characterisation of phospholipase A2 from human epidermis, Biochem. Soc. Trans. 19 (1991) 91S. [26] C. Song, X. Li, Z. Kang, Y. Kadotomi, Omega-3 fatty acid ethyl-eicosapentaenoate attenuates IL-1beta-induced changes in dopamine and metabolites in the shell of the nucleus accumbens: involved with PLA2 activity and corticosterone secretion, Neuropsychopharmacology 32 (2007) 736–744. [27] A. Bernardo, L. Minghetti, PPAR-gamma agonists as regulators of microglial activation and brain inflammation, Curr. Pharm. Des. 12 (2006) 93–109. [28] J.N. Feige, L. Gelman, L. Michalik, B. Desvergne, W. Wahli, From molecular action to physiological outputs: peroxisome proliferator-activated receptors are nuclear receptors at the crossroads of key cellular functions, Prog. Lipid Res. 45 (2006) 120–159. [29] W. Vanden Berghe, K. De Bosscher, E. Boone, S. Plaisance, G. Haegeman, The nuclear factor-kappaB engages CBP/p300 and histone acetyltransferase activity for transcriptional activation of the interleukin-6 gene promoter, J. Biol. Chem. 274 (1999) 32091–32098. [30] Y. Takata, Y. Kitami, Z.H. Yang, M. Nakamura, T. Okura, K. Hiwada, Vascular inflammation is negatively autoregulated by interaction between CCAAT/ enhancer-binding protein-delta and peroxisome proliferator-activated receptor-gamma, Circ. Res. 91 (2002) 427–433. [31] L.H. Wang, X.Y. Yang, X. Zhang, et al., Transcriptional inactivation of STAT3 by PPARgamma suppresses IL-6-responsive multiple myeloma cells, Immunity 20 (2004) 205–218. [32] D. Kelly, J.I. Campbell, T.P. King, et al., Commensal anaerobic gut bacteria attenuate inflammation by regulating nuclear-cytoplasmic shuttling of PPARgamma and RelA, Nat. Immunol. 5 (2004) 104–112. [33] M. Li, G. Pascual, C.K. Glass, Peroxisome proliferator-activated receptor gamma-dependent repression of the inducible nitric oxide synthase gene, Mol. Cell Biol. 20 (2000) 4699–4707. [34] G. Pascual, A.L. Fong, S. Ogawa, et al., A SUMOylation-dependent pathway mediates transrepression of inflammatory response genes by PPAR-gamma, Nature 437 (2005) 759–763. [35] Y. Wu, W.W. Chin, Y. Wang, T.P. Burris, Ligand and coactivator identity determines the requirement of the charge clamp for coactivation of the peroxisome proliferator-activated receptor gamma, J. Biol. Chem. 278 (2003) 8637–8644.