Conjugated Linoleic Acid Suppresses NF-κB Activation and IL-12 Production in Dendritic Cells through ERK-Mediated IL-10 Induction¹

It is well established that some classes of dietary polyunsaturated fatty acids (PUFA)⁴ can suppress immune responses and studies focusing on n-3 PUFA, naturally abundant in marine oils, have demonstrated that they can suppress T cell activation and attenuate cytokine production from macrophages (1, 2). Furthermore, feeding fish oils has been shown to ameliorate the symptoms of certain autoimmune diseases in animal models (3). More recently, it has been reported that conjugated linoleic acid (CLA) has immunomodulatory properties (4). CLA is a heterogenous group of positional and geometric isomers of conjugated dienoic derivatives of linoleic acid, an n-6 PUFA. The cis-9, trans-11 CLA isomer (c9, t11-CLA) is the naturally occurring dietary form of CLA and is found in the lipid fraction of meat, milk, and dairy products. Lower levels of other isomers, including trans-10, cis-12 CLA (t10, c12-CLA), are also present in food sources (4). CLA suppresses proinflammatory cytokine production from a variety of cell types, and studies in animal models have shown that feeding CLA has beneficial effects on inflammatory-induced growth suppression, endotoxin-induced anorexia, and mucosal damage in experimental colitis (5, 6). These data suggest that this fatty acid has considerable potential as an immunomodulatory molecule in the prevention or treatment of inflammatory and immune-mediated diseases. However its role in modulating innate immune responses, in particular dendritic cell (DC) activation, has received little attention.

DC play a critical role in directing adaptive immune responses, including Th cell responses. Activation of DC by inflammatory stimuli, such as LPS, induces maturation and homing to lymph nodes where they present Ag to naive T cells. The differentiation of naive CD4⁺ T cells into Th1, Th2, or regulatory T (Tr) cells is determined in part by the maturation status of the DC, but also by the cytokines produced by DC and other cells of the innate immune system. Production of IL-12, in response to inflammatory stimuli promotes the induction of Th1 cells. In contrast, products of helminth parasites have been shown to activate DC, resulting in the production of IL-4 and induction of Th2 cells (7, 8). Furthermore, other pathogen-derived molecules, such as filamentous hemagglutinin from Bordetella pertussis, enhance IL-10 production by DC and promote the induction of type 1 Tr (Tr1) cells (9). The pivotal role of IL-4, IL-10, and IL-12 in directing Th cell responses suggests that modulating their production by DC can affect quantitatively and qualitatively the adaptive immune response. Indeed, molecules that enhance IL-10 and suppress IL-12 production by DC have been shown to promote the induction of Th2 and Tr1 cells (9, 10).

Expression of IL-12, a heterodimeric cytokine comprising p35 and p40 subunits, is regulated at the level of transcription, and NF-κB activation is a key factor involved in this process. Indeed inhibition of this transcription factor results in suppression of IL-12 production by APCs (11). Treatment of such cells with a variety of proinflammatory agents, such as LPS, results in phosphorylation, ubiquitination, and degradation of the IκB. The degradation of IκB proteins by IκB kinase results in liberation of NF-κB from IκB and subsequent translocation of NF-κB to the nucleus (12). There is evidence to suggest that certain fatty acids may mediate their immunomodulatory function through effects on NF-κB. For example, eicosapentaenoic acid (EPA) decreases NF-κB activation in LPS-stimulated macrophages (13) and docosahexaenoic acid (DHA) inhibits its activation in Jurkat T cells (14). However, the mechanism of inhibition of NF-κB activation by fatty acids is poorly understood. Studies have indicated that IL-10 can reduce degradation and/or phosphorylation of IκB proteins and inhibit NF-κB translocation into the nucleus (12, 15). Furthermore, IL-10 is involved in negative autocrine regulation of IL-12 production in DC. Therefore, IL-10-mediated inhibition of NF-κB activation may be one mechanism by which immunomodulatory molecules exert their suppressive effect on IL-12 production by activated DC.

The MAPK are important for regulation of survival, maturation, and cytokine secretion by DC. Activation of ERK has been shown to promote IL-10 production and mediate negative feedback regulation of IL-12 production (16). Moreover, the inhibition of ERK activation in DC results in suppression of IL-10 and enhancement of IL-12 production (17). These data suggest that ERK phosphorylation may play an important role in the regulation of two key cytokines involved in the induction of Th cell responses. Although n-3 PUFA have been shown to inhibit phosphorylation of ERK (18), recently it has been demonstrated that CLA isomers can enhance its activation (19).

In the present study we tested the hypothesis that the PUFA, c9, t11-CLA, can enhance anti-inflammatory cytokines by DC and modulate production of key cytokines associated with direction of Th cell responses. We report that CLA suppresses IL-12 production by LPS-stimulated DC by an IL-10-dependent mechanism. We provide evidence for a role in ERK activation in this IL-10-mediated suppression. Furthermore, these IL-10-mediated effects are mediated through inhibition of NF-κB activation. Given the importance of DC in directing Th cell responses, these results may explain some of the immunomodulatory properties of CLA.

BALB/c mice were purchased from Harlan (U.K.) and were used at 10–14 wk of age. Animals were maintained according to the regulations of the European Union and the Irish Department of Health. c9, t11-CLA was purchased from Alexis Chemicals (Cayman Chemicals). Escherichia coli LPS (serotype 127:B8) and the ERK inhibitor, U0126, were purchased from Alexis Biochemicals. The c9, t11-CLA isomer and the ERK inhibitor were dissolved in sterile DMSO (Sigma-Aldrich) and stored at −20°C in the dark.

Bone marrow-derived immature DC were prepared by culturing bone marrow cells obtained from the femurs and tibia of mice in RPMI 1640 medium with 5% FCS supplemented with 10% supernatant from a GM-CSF-expressing cell line (J558-GM-CSF). The cells were cultured at 37°C for 3 days, and the supernatant was carefully removed without disturbing the cell monolayer and replaced with fresh medium with 10% GM-CSF cell supernatant. On day 7 of culture, cells were collected, counted, and used for assays. For experiments, either DMSO (vehicle control) or c9, t11-CLA was added to the cells on day 1 or 5 of culture.

On day 7 of culture, after 48 h of treatment with either DMSO or c9, t11-CLA, DC (1 × 10⁶/ml) were cultured in 24-well plates with LPS (100 ng/ml) or medium alone for 0–24 h. In certain experiments, DC were incubated in the presence or absence of a neutralizing anti-IL-10 Ab (10 μg/ml) or the specific ERK inhibitor U0126 (5 μM). At the end of the relevant incubation periods, supernatants were removed and IL-10, IL-12p40, and IL-12p70 concentrations were measured using DuoSet ELISA kits from R&D Systems according to manufacturer’s instructions. After removal of the supernatants the cells were harvested and TRIreagent (Molecular Research Center) was added. Total RNA was extracted from the cells according to the manufacturer’s instructions. Two micrograms of RNA was DNase treated and reverse transcribed in a 25-μl reaction containing 1× reverse transcription buffer (50 mmol/L Tris-HCl, 75 mmol/L KCl, 3 mmol/L MgCl₂, 10 mmol/L DTT), 500 ng of random primers, 10 mmol/L dNTPs, and 200 U of Moloney murine leukemia virus reverse transcriptase. mRNA expression was quantified by real-time PCR on an ABI 7700 Sequence Detection System (PerkinElmer Applied Biosystems). A negative control was processed from RNA without Moloney murine leukemia virus reverse transcriptase. TaqMan real-time PCR was performed for IL-10, IL-12p40, and IL-12p35 using PreDeveloped Assay Reagent kits. Each 25-μl reaction contained 20 ng of cDNA, 2× TaqMan Universal PCR Mastermix, forward and reverse primers, and TaqMan probe. All reactions were performed in duplicate using the following conditions: 50°C for 2 min and 95°C for 10 min followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. For each sample, results were normalized by dividing the amount of target gene by the amount of GAPDH and expressed relative to the vehicle control treatment (DMSO) for the non-LPS-stimulated cells.

DC were treated with either DMSO or c9, t11-CLA from day 1 of culture and then cultured in 24-well plates (1 × 10⁶ cells) with LPS (100 ng/ml) or medium alone for 24 h. Cells were then washed and used for immunofluorescence analysis. The expression of IL-10R on DC was assessed using an anti-mouse CD11c (hamster IgG1; Caltag Laboratories), IL-10R (rat IgG1; BD Biosciences), and appropriately labeled isotype-matched Abs, which acted as controls. After incubation for 30 min at 4°C, cells were washed and immunofluorescence analysis was performed on a FACScan (BD Biosciences) using CellQuest software

DC were treated with either DMSO or c9, t11-CLA for 48 h and then cultured DC (3 × 10⁶ cells) in 6-well plates with LPS (100 ng/ml) or medium alone for 0–5 h. After this time, supernatants were removed and whole cell extracts were prepared by lysis of the cells in 100 μl of 1× sample buffer (62.5 mM Tris-HCl (pH 6.8), 10% glycerol, 2% w/v SDS, 0.1% w/v bromphenol blue, 50 mM DTT). Samples were then sonicated using a Vibra Cell sonicator for 10 s at a pulse of 1.5 and amplitude of 40, boiled at 100°C for 5 min, and stored at −80°C for Western blot analysis.

After treatment with either DMSO or c9, t11-CLA, DC (3 × 10⁶ cells) were cultured in 6-well plates with LPS (100 ng/ml) or medium alone for 0–5 h. In certain experiments, DC were incubated in the presence or absence of a neutralizing anti-IL-10 Ab (10 μg/ml). Supernatants were removed and cells were washed in ice-cold PBS and lysed in 10 mmol/L HEPES (pH 7.9), 1.5 mmol/L MgCl₂, 10 mmol/L KCl, 0.5 mmol/L PMSF, 0.5 mmol/L DTT, and 0.1% v/v Nonidet P-40. After centrifugation, the supernatant was removed and stored at −80°C as the cytosolic extract. To prepare the nuclear extract, the pellet was resuspended in 20 mmol/L HEPES (pH 7.9), 1.5 mmol/L MgCl₂, 420 mmol/L NaCl, 0.2 mmol/L EDTA, 25% glycerol, and 0.5 mmol/L PMSF, incubated on ice for at least 15 min with repeated vortexing, and then centrifuged. To the supernatant, 10 mmol/L HEPES (pH 7.9), 50 mmol/L KCl, 0.2 mmol/L EDTA, 20% glycerol, 0.5 mmol/L DTT, 0.5 mmol/L PMSF was added and samples were mixed well and stored at −80°C as the nuclear extract. The concentration of protein in the nuclear and cytoplasmic samples was quantified by Bradford assay (Bio-Rad). Samples were used for Western blot analysis, NF-κB activity assay, and EMSAs.

Proteins and prestained protein markers (precision plus protein standards; Bio-Rad) were separated by SDS-PAGE and blotted onto 0.45 μM Biotrace polyvinylidene fluoride transfer membrane (Pall Life Sciences). Membranes were blocked using 10% nonfat dried milk in PBS with 0.05% Tween (PBS-T) and incubated overnight at 4°C with either anti-phosphoERK (pERK; Santa Cruz Biotechnology; 1:750), anti-ERK (Santa Cruz Biotechnology; 1:1000), anti-NF-κBp65 (Santa Cruz Biotechnology; 1:500), or anti-IκBα (Santa Cruz Biotechnology; 1:500) Abs. Membranes were washed with PBS-T and incubated for 2 h at room temperature with either peroxidase-conjugated anti-mouse IgG or anti-rabbit IgG (Sigma-Aldrich; 1:1000). After further washing, protein complexes were visualized with supersignal (Pierce). Membranes were exposed to film for 1–10 min and processed using an Agfa x-ray processor. Protein bands were quantified using the GeneSnap acquisition and GeneTools analysis software (GeneGenius Gel Documentation and Analysis System; Syngene). Where required, membranes were stripped by incubating them in stripping buffer (Tris (pH 6.8), 2-ME, SDS) for 30 min at 50°C before probing with a subsequent Ab. Levels of pERK were normalized to total ERK and expressed in arbitrary units. NF-κBp65 and IκBα were either expressed as arbitrary units or normalized to controls and expressed as fold change relative to control.

Two micrograms of nuclear protein was incubated for 30 min at room temperature with 10,000 cpm ³²P-labeled oligonucleotide probe (PerkinElmer Life Sciences), 2 μg of poly(dI:dC) (Amersham-Pharmacia), 100 mmol/L Tris (pH 7.5), 500 mmol/L NaCl, 40% glycerol (w/v), 5 mmol/L EDTA, 5 mmol/L DTT, and 1 g/L nuclease-free BSA. The probes used in the EMSA contained the consensus binding sequence for NF-κB 5′-AGT TGA GGG GAC TTT CCC AGG C-3′, which had been previously labeled with [³²P]ATP by T4 polynucleoide kinase (Promega). DNA-protein complexes were resolved on a nondenaturing 4% (v/v) polyacrylamide gels and run for ∼1.75 h at 140 V in 0.5× Tris borate buffer. The gels were dried, exposed to film overnight, and developed using an Agfa x-ray processor.

NF-κB activity was determined by using a TransAM NF-κB Chemi assay (Active Motif) that measured NF-κBp65:DNA binding. The assay was conducted according to the manufacturer’s instructions. Briefly, 0.5 μg of nuclear protein was incubated in a 96-well plate containing the NF-κB consensus site (5′-GGGACTTTCC-3′) for 1 h. After washing, a primary Ab that recognizes an epitope on p65, which is accessible only when NF-κB is activated and bound to its target DNA, was added for 1 h. After a further washing step, a peroxidase-conjugated secondary Ab was added for 1 h and then a chemiluminescent substrate was applied to give a readout that could be quantified by luminescence. Results are normalized to controls and expressed as fold change relative to control.

One-way ANOVA was used to determine significant differences between conditions. When this indicated significance (p < 0.05), post hoc Student-Newmann-Keul test analysis was used to determine which conditions were significantly different from each other. Student’s unpaired t test was used for analysis when two conditions were compared. There was no significant difference between cells alone and DMSO (vehicle control)-treated cells; therefore, DMSO was used as the reference treatment.

Because IL-12 is critical in the induction of Th1 cells and IL-10 can enhance differentiation of Tr1 cells and has also been implicated in inhibiting Th1 responses (9), we examined the possibility that c9, t11-CLA could modulate the transcription and production of these cytokines from DC. Incubation of DC with c9, t11-CLA for 48 h did not induce detectable concentrations of IL-10, IL-12p70, or IL-12p40 (Fig. 1). Activation of DC with LPS resulted in significant production of IL-10, IL-12p70, and IL-12p40, detectable 6 (p < 0.05; p < 0.001), 12 (p < 0.001), and 24 h (p < 0.001) after treatment. The production of LPS-induced IL-10 was significantly enhanced (p < 0.001) at all three time points examined after pretreatment of the DC with c9, t11-CLA. In contrast, pretreatment of DC with c9, t11-CLA suppressed LPS-induced IL-12p70 (p < 0.001) and IL-12p40 (p < 0.001) at 12 and 24 h. Similar results were seen with another isomer of CLA, t10, c12-CLA and with the n-3 fatty acids EPA and DHA and not with the saturated fatty acid, stearic acid (data not shown). Treatment of DC with c9, t11-CLA had no effect on TNF-α production (data not shown). To elucidate whether c9, t11-CLA exerted its effects at the level of transcription, we also examined expression of mRNA for these cytokines. Although transcription of IL-12p35 (Fig. 2,B) and IL-12p40 (Fig. 2,C) were not modulated in resting cells, c9, t11-CLA enhanced IL-10 mRNA expression, although this increase did not reach statistical significance (Fig. 2,A). LPS stimulation of DC significantly increased transcription of IL-10 mRNA, IL-12p35 mRNA, and IL-12p40 mRNA at a number of time points over the 48-h period examined (p < 0.05; p < 0.01; p < 0.001). DC pretreated with c9, t11-CLA had significantly enhanced transcription of IL-10 at 2 and 6 h (p < 0.05; p < 0.01). Conversely, c9, t11-CLA pretreatment significantly suppressed LPS-induced IL-12p35, particularly at 12 h (Fig. 2,B; p < 0.05). LPS-induced transcription of IL-12p40 was also suppressed by c9, t11-CLA pretreatment at 6, 12, 18, and 24 h (Fig. 2 C; p < 0.05).

Given our data demonstrating that the c9, t11-CLA-induced enhancement of IL-10 transcription and protein production was detectable before the suppression of IL-12, we examined whether the suppression of IL-12 by c9, t11-CLA was mediated by IL-10. LPS stimulation of c9, t11-CLA-treated DC was conducted in the presence and absence of a neutralizing anti-IL-10 Ab and IL-12p70 and IL-10 mRNA and protein production were measured. Consistent with the data shown in Fig. 1, c9, t11-CLA significantly suppressed LPS-induced IL-12p70 protein production at 12 h (Fig. 3,A; p < 0.001). This suppression was significantly reversed by addition of anti-IL-10 Ab (p < 0.01). Similarly, LPS-induced IL-12p35 mRNA expression (Fig. 3,B) was suppressed by c9, t11-CLA, and significantly reversed in the presence of the anti-IL-10 Ab at 6 (p < 0.001) and 12 h (p < 0.001). Consistent with the previous data, c9, t11-CLA significantly enhanced LPS-induced IL-10 production at 6 (Fig. 3,C; p < 0.05) and 12 h (p < 0.001) and IL-10 mRNA expression at 6 h (Fig. 3 D; p < 0.05).

Having demonstrated enhanced production of IL-10 by DC, we examined the possibility that c9, t11-CLA may also modulate IL-10R expression on these cells. Incubation of bone marrow-derived DC with c9, t11-CLA increased the percentage of DC expressing IL-10R (48.8% in DMSO-treated DC vs 68.7% in CLA-treated DC) and enhanced the level of expression on CD11c⁺ DC (Fig. 4, top panel). Stimulation of DC with LPS decreased IL-10R expression, and treatment with CLA partially reversed this effect (Fig. 4, bottom panel).

Having demonstrated that c9, t11-CLA enhances IL-10 mRNA expression in immature DC and significantly increases LPS-induced IL-10 production, we examined whether ERK activation was involved in this enhancement and the subsequent suppression of IL-12 production. To address this question, we first determined whether c9, t11-CLA could modulate ERK phosphorylation in both resting DC and after stimulation with LPS at both early and later time points. Activation of ERK was significantly increased in resting DC after treatment with c9, t11-CLA treatment (Fig. 5, A and B; p < 0.01; p < 0.001). After LPS stimulation, c9, t11-CLA significantly enhanced phosphorylation of ERK at 5, 10, and 20 min (Fig. 5,A; p < 0.001; p < 0.05). This increase in ERK activation in c9, t11-CLA-treated DC was sustained at 1, 2, and 5 h (Fig. 5,B; p < 0.05; p < 0.01). We then tested whether the effect of c9, t11-CLA on IL-10 production and IL-12 suppression was dependent on ERK activation by using the specific ERK inhibitor, U0126. Consistent with the data in Fig. 1, c9, t11-CLA enhanced LPS-induced IL-10 production (Fig. 5,C; p < 0.05) and suppressed IL-12p70 production (Fig. 5,D; p < 0.001). Inhibition of ERK activation by U0126 resulted in an almost total abrogation of IL-10 production in both DMSO- and c9, t11-CLA-treated DC (Fig. 5,C; p < 0.001), demonstrating that ERK activation is necessary for IL-10 production in DC. Furthermore, the suppression of LPS-induced IL-12p70 production by c9, t11-CLA was significantly reversed in the presence of U0126 (Fig. 5 D; p < 0.001), indicating that ERK activation and subsequent production of IL-10 is necessary for c9, t11-CLA-induced IL-12p70 suppression.

Having demonstrated that c9, t11-CLA suppresses LPS-induced IL-12 mRNA expression, and that this effect was mediated in part through IL-10 production, we examined the role of NF-κB. Given that translocation of NF-κBp65 into the nucleus and degradation of IκBα are two events involved in NF-κB activation, we examined the effects of c9, t11-CLA on NF-κB:DNA binding and on NF-κBp65 and IκBα levels in the cytosol and the nucleus. c9, t11-CLA significantly decreased NF-κB activity in resting DC (Fig. 6,A; p < 0.001) and treatment of DC with c9, t11-CLA significantly decreased LPS-induced NF-κB activity at 2 (p < 0.001) and 5 h (p < 0.001). c9, t11-CLA had no effect on cytosolic or nuclear NF-κBp65 levels in resting cells (Fig. 6, B and C). A kinetic study showed that c9, t11-CLA significantly increased cytosolic NF-κBp65 (Fig. 6,B; p < 0.05) and decreased nuclear NF-κBp65 levels (Fig. 6,C; p < 0.01) at 5 h. These data suggest that c9, t11-CLA delays LPS-stimulated nuclear NF-κBp65 translocation for NF-κB activation. c9, t11-CLA also marginally increased nuclear IκBα expression in resting DC (Fig. 6,E). After LPS stimulation, c9, t11-CLA increased IκBα in the cytosol from 0.5 h and significantly at 1 (Fig. 6,D; p < 0.01) and 5 h (p < 0.01). IκBα was also significantly increased after LPS stimulation by c9, t11-CLA in the nucleus at 0.5 (Fig. 6 E; p < 0.001), 1, 2, and 5 h (p < 0.05). These results indicate that c9, t11-CLA prevents the degradation of IκBα in both resting cells and after LPS stimulation. These data, together with the delay in nuclear NF-κBp65 translocation, may explain how c9, t11-CLA decreases NF-κB activation.

Because the suppression of IL-12 by c9, t11-CLA is IL-10 dependent and IL-10 can inhibit NF-κB activity, we examined whether the effect of CLA on NF-κB activity was mediated by IL-10. c9, t11-CLA-treated DC were stimulated with LPS in the presence and absence of a neutralizing anti-IL-10 Ab for 5 h, a time point at which c9, t11-CLA modulated expression of both NF-κBp65 and IκBα (shown in Fig. 6). c9, t11-CLA significantly decreased LPS-induced NF-κB activity (Fig. 7,A; p < 0.05) that was significantly reversed in the presence of an anti-IL-10 Ab (p < 0.01). These results were confirmed by EMSA (Fig. 7,B). NF-κBp65 and IκBα levels were assessed in the cytosolic and the nuclear fractions by Western blotting using specific Abs. Consistent with the data in Fig. 6, Western blot analysis revealed that c9, t11-CLA significantly increased cytosolic NF-κBp65 (Fig. 7,C; p < 0.01) and decreased nuclear NF-κBp65 (Fig. 7,D; p < 0.05) when assessed 5 h after LPS stimulation. These effects were significantly reversed in the presence of the anti-IL-10 Ab (p < 0.01; p < 0.05). Furthermore, c9, t11-CLA significantly increased cytosolic and nuclear IκBα levels (Fig. 7, E and F; p < 0.05) after LPS stimulation, and these effects were significantly reversed in the presence of an anti-IL-10 Ab (p < 0.05). These data suggest that the IL-10-dependent suppression of LPS-induced IL-12 by c9, t11-CLA is mediated by inhibition of NF-κB activation.

The significant findings of this study are that the PUFA, c9, t11-CLA, suppresses LPS-induced IL-12 production in DC and that this suppression is dependent on activation of ERK, enhanced IL-10 production, and inhibition of NF-κB activation. This is the first study to demonstrate that c9, t11-CLA can modulate transcription and production of key cytokines by DC involved in controlling inflammatory responses and directing Th cell responses, and may explain certain of its immunomodulatory properties.

The effects of PUFA on cytokine production by immune cells, such as T cells and macrophages are well documented. For example, EPA reduces production of IL-2 and IL-13 from T cells (2) and CLA suppresses TNF-α, IL-1β, and IL-6 production in IFN-γ-stimulated RAW macrophages (20). In this study, we demonstrate that the c9, t11 isomer of CLA suppresses LPS-induced IL-12p40 and IL-12p70 production by murine DC. We also found similar suppressive effects with EPA, DHA, and with another CLA isomer, t10, c12-CLA (data not shown). IL-12, produced by innate immune cells, is a key cytokine in the development of Th1 responses and has been implicated in the pathogenesis of infectious, inflammatory, and autoimmune diseases, such as inflammatory bowel disease (21, 22). Our findings are consistent with studies demonstrating that PUFA can inhibit Th1 responses. Specifically, EPA and DHA reduce circulating and splenic IL-12p70 and IL-12p40 production in mice after infection with Listeria monocytogenes (23) and decrease adaptive Th1-driven responses in a murine sensitization model (24). Furthermore, a recent report demonstrated that EPA suppresses IL-12p40 and TNF-α concentrations in human DC (25). Interestingly, we found that c9, t11-CLA had no effect on TNF-α production (data not shown). This suggests that individual PUFA exert their effects on specific cytokines and, therefore, may have distinct mechanisms of action. The selective suppression of IL-12 we have observed in DC suggests the potential use of c9, t11-CLA in Th1-mediated diseases. Furthermore, the effects of c9, t11-CLA on IL-12p40 suggest that it may also modulate production of IL-23, which shares the p40 subunit with IL-12 and is involved in the maintenance of Th1 responses (26).

IL-10 is a regulatory cytokine with anti-inflammatory properties, potently inhibiting the capacity of innate immune cells to produce inflammatory mediators such as IL-12, TNF-α, and IL-1 (27). We report that c9, t11-CLA enhanced LPS-induced IL-10 transcription and production by DC. Furthermore, c9, t11-CLA enhanced IL-10R expression on both resting and LPS-stimulated DC. This enhanced IL-10R expression would allow the DC to have an increased response to the enhanced levels of IL-10 being secreted by these cells. These data indicate a possible role for IL-10 in mediating the effects of CLA. PUFA have previously been shown to enhance IL-10; treatment of mice with EPA results in increased production of IL-10 by whole blood cells and T cells (28, 29) and also enhances IL-10 levels in the brain in response to gamma irradiation (30). A recent study reported that EPA had no effect on IL-10 production from human DC (25). However, we found EPA, DHA, and the t10, c12 isomer of CLA significantly enhanced IL-10 production by murine DC (data not shown). This provides further evidence of different mechanisms of action of individual PUFA or differential effects on human and murine cells.

The present study is the first to demonstrate that c9, t11-CLA can enhance LPS-induced IL-10 in any cell type. In addition, our data demonstrate for the first time that suppression of IL-12 by a PUFA is at least in part mediated through IL-10 induction. Our findings are consistent with studies on a number of pathogen-derived molecules, including filamentous hemagglutinin from B. pertussis, which suppresses IL-12 production by DC through an IL-10-dependent mechanism (9). Interestingly suppression of IL-12 by the t10, c12-CLA isomer was not IL-10 dependent (data not shown). This strengthens the argument that individual PUFA may use different mechanisms through which they exert their immunosuppressive effects. IL-12 production by DC can also be down-regulated by a number of other factors including PGE₂ (31). Although c9, t11-CLA enhances production of PGE₂ in DC, it is not involved in IL-12 suppression, as inhibition of PGE₂ using specific inhibitors did not restore IL-12 production in LPS-stimulated DC (C. E. Loscher, K. H. Mills, and H. M. Roche, unpublished observation). This is consistent with other studies, which have demonstrated that the inhibitory effects of PUFA are not dependent on production of PGE₂ (25, 32); however, the involvement of other prostaglandins has not yet been investigated.

We also demonstrated c9, t11-CLA induction of IL-10 and suppression of IL-12 at the transcriptional level. Furthermore, IL-10 appears to exert its suppressive effect on IL-12 through inhibition of IL-12 transcription. This suggested the involvement of transcription factors, particularly NF-κB, which is a key factor involved in production of IL-12. Previous studies have demonstrated that specific inhibitors of NF-κB suppress IL-12 production (33) and anti-inflammatory molecules such as the glucocorticoid dexamethasone inhibit IL-12 production in LPS-stimulated monocytic cells by inhibiting NF-κB activity (34). Moreover, IL-10 has recently been shown to exert its anti-inflammatory effects through inhibition of NF-κB. Specifically, inhibition of NF-κB activation in DC is associated with suppressed IκB kinase activity (15), and IL-10 can delay translocation of NF-κBp65 into the nucleus in monocytic cells (12). The present study clearly demonstrates that c9, t11-CLA suppresses activation of NF-κB in both resting and LPS-stimulated DC. Analysis of cytosolic and nuclear NF-κBp65 and IκBα revealed that this suppression was associated with inhibition of IκBα degradation and delay of NF-κBp65 translocation into the nucleus. This is consistent with a recent study, which demonstrated that capsaicin, an anti-inflammatory nutrient, can inhibit IκBα degradation after LPS stimulation in peritoneal macrophages (35). Although few studies have reported the effects of CLA on NF-κB, our findings are consistent with a limited number of studies that have suggested that other PUFA can down-regulate NF-κB activity. Treatment of murine RAW macrophages with a PUFA lipid emulsion resulted in reduced LPS-stimulated NF-κB activity and IκB phosphorylation (36). Recently, EPA was shown to have no effect on NF-κB in human DC and T cells (2, 25). In contrast, it significantly inhibits NF-κB activation in monocytic THP-1 cells, endothelial cells and Jurkat T cells (13, 14, 37). These conflicting data again indicate that individual PUFA may have distinct effects on different transcription factors or work in a cell-specific manner. Although these studies suggest that PUFA exert their effects through inhibition of transcription factors, clear evidence of the mechanism involved is lacking. We report that the inhibitory effect of c9, t11-CLA on NF-κB was reversed in the presence of anti-IL-10 Abs indicating that this is an IL-10-dependent mechanism. Furthermore, we clearly demonstrate that IL-10 is directly involved in inhibition of IκBα degradation and NF-κBp65 translocation into the nucleus. Although the ability of IL-10 to inhibit NF-κB activation has been previously reported, this is the first study to demonstrate a role for IL-10 in suppression of NF-κB activity by any PUFA; however, whether this mechanism is unique to c9, t11-CLA is yet to be defined.

The role of ERK in DC function remains unclear, but recent studies have highlighted its importance in the initiation of responses and in cell survival (38). This study demonstrates that culture with c9, t11-CLA results in significant activation of ERK in resting DC and enhances ERK activation in LPS-stimulated DC. Although the effects of c9, t11-CLA on LPS-induced ERK phosphorylation may be additive, it is still clear that treatment of DC with c9, t11-CLA results in early and sustained enhancement of ERK activation, which is important given the role of ERK in production of IL-10. The few attempts to address the role of PUFA on MAPK have demonstrated that EPA and DHA decrease phosphorylation of ERK in T cells and can selectively modulate JNK (2, 18). However, EPA has no effect on ERK or JNK activation in human DC (25). In contrast, our findings are consistent with a recent study, which demonstrated that CLA enhanced ERK activation in stromal vascular cells, which was associated with enhanced production of IL-6 and IL-8 (19). These observations provide further evidence for cell-specific and fatty acid-specific effects of individual PUFA.

The specific role of ERK in IL-10 production remains to be elucidated; however, it is clear from some studies that ERK activation is important for IL-10 production in DC (16). Furthermore, studies using specific ERK inhibitors have shown that ERK activation suppresses LPS-mediated IL-12 production in both macrophages (39) and DC (17). We report that ERK activation is essential for IL-10 production in DC, as inhibition of ERK using a specific inhibitor completely abrogated IL-10 production. Furthermore, inhibition of ERK activation also reverses the c9, t11-CLA-induced suppression of IL-12 in LPS-stimulated DC. These data clearly demonstrate that ERK-mediated induction of IL-10 is involved in the suppression of IL-12 production by murine DC after treatment with c9, t11-CLA. This conclusion is consistent with a report that demonstrated that Leishmania lipophosphoglycan exerts its inhibitory effects on IL-12 production by macrophages through activation of ERK (39) and inhibition of ERK using specific inhibitors, results in enhancement of IL-12 production by DC (17). The mechanism of ERK induction in DC has not yet been fully defined nor has the mechanism by which PUFA modulate its activation. It has been suggested that PGJ₂ can enhance ERK activation (40). This may be important in determining the mechanism of action of PUFA given that they readily induce production of a variety of prostaglandins from cells, including PGJ₂. Although we have ruled out the involvement of PGE₂ in suppression of IL-12 by c9, t11-CLA, it may be important to examine the involvement of other prostaglandins in mediating this effect.

Our findings demonstrate that c9, t11-CLA can suppress IL-12 and enhance IL-10 production by DC, a cytokine profile that has been associated with suppression of Th1 and enhancement of Tr cell responses. It has previously been reported that PUFA promotes activation-induced cell death in T cells that are polarized toward the Th1 phenotype in vitro (41). Furthermore, CLA has been shown to protect against inflammatory bowel disease, a disease mediated by Th1 cells (5). Our study demonstrates that c9, t11-CLA can act at the level of the DC, inhibiting cytokines that promote Th1 cells and enhancing anti-inflammatory cytokines that promote Th2 or Tr cells, and thereby provides a possible mechanism whereby it may be capable of protecting against Th1-mediated diseases. This is the first study to provide clear evidence of a mechanism whereby PUFA may exert their anti-inflammatory effects.

The authors have no financial conflict of interest.