Oxidized Omega-3 Fatty Acids Inhibit NF-B Activation Via a PPAR-Dependent Pathway
http://www.100md.com
《动脉硬化血栓血管生物学》
From the Department of Pathology, University of Iowa Hospitals and Clinics, Iowa City, Iowa.
ABSTRACT
Objective— The aim of this study was to determine the effects of oxidized versus native omega-3 fatty acids on the endothelial expression of chemokines MCP-1 and IL-8, and, if effective in inhibiting chemokine expression, to determine the mechanism for the inhibition of chemokine expression.
Methods and Results— Using enzyme-linked immunosorbent assays, we show that oxidized EPA and DHA but not unoxidized EPA or DHA inhibit cytokine-induced endothelial expression of monocyte chemoattractant protein (MCP)-1 and, to a lesser extent, IL-8. In electrophoretic mobility shift assays, oxidized EPA but not unoxidized EPA potently inhibited cytokine-induced activation of endothelial nuclear factor-B (NF-B). Using Western blot analyses, we show that the inhibition of NF-B activation was not caused by prevention of phosphorylation of IB because oxidized EPA did not inhibit cytokine-induced phosphorylation and ubiquination of IB. Furthermore, oxidized EPA inhibited NF-B activation in endothelial cells derived from wild-type mice but had no inhibitory effects on NF-B activation in endothelial cells derived from peroxisome proliferator-activated receptor (PPAR)-deficient mice, indicating that oxidized EPA requires PPAR for its inhibitory effects on NF-B.
Conclusions— These studies show that the antiinflammatory effects of fish oil may result from the inhibitory effects of oxidized omega-3 fatty acids on NF-B activation via a PPAR-dependent pathway.
Key Words: monocyte chemoattractant protein-1 ? oxidized omega-3 fatty acids ? oxidized eicosapentaenoic acid ? nuclear factor-B ? PPARa
Introduction
Consumption of marine fish oil has been reported to improve the prognosis of several chronic inflammatory diseases characterized by leukocyte accumulation and leukocyte-mediated tissue injury, including atherosclerosis, IgA nephropathy, inflammatory bowel disease, rheumatoid arthritis, etc.1–4 These beneficial effects of fish oil have been associated with the omega-3 polyunsaturated fatty acids (PUFAs), eicosapentaenoic acid (EPA), and docosahexaenoic acid (DHA), which are abundant in marine fish oil.
EPA and DHA are highly polyunsaturated and easily undergo auto-oxidation.5,6 In fact, it is very difficult to avoid the oxidation of these very labile fatty acids. More importantly, in vivo, a large increase in tissue and plasma accumulation of both omega-3 fatty acids and fatty acid oxidation products is noted in subjects consuming fish oil, even after addition of antioxidant supplements to the diet.7–10 This suggests the possibility that oxidized omega-3 fatty acids may be an important component of the observed antiinflammatory effects of fish oil. Indeed, our previous studies have shown that oxidized EPA, and not unoxidized EPA, potently inhibits leukocyte–endothelial interactions, both in vitro and in vivo, through a peroxisome proliferator-activated receptor (PPAR)-dependent mechanism.11,12
One of the early events in inflammation is the upregulation of endothelial chemokines, monocyte chemoattractant protein-1 (MCP-1) and IL-8, in response to proinflammatory cytokines such as tumor necrosis factor (TNF)- and IL-1. MCP-1 and IL-8 in turn promote leukocyte chemotaxis, adhesion, and transendothelial migration,13,14 and neutralization of MCP-1 has been shown to attenuate in vivo injury arising from inflammatory mechanisms.15–17 The aim of this study was to determine the effects of oxidized versus native omega-3 fatty acids on the endothelial expression of chemokines MCP-1 and IL-8, and, if effective in inhibiting chemokine expression, to determine the mechanism for the inhibition of chemokine expression.
Methods
Preparation of Fatty Acids
EPA, DHA, and arachidonic acid were purchased from Cayman Chemicals. They were relatively unoxidized as assessed by thiobarbituric acid-reactive substances assay using malondialdehyde as a standard; this assay is a measure of fatty acid oxidation.18 The fatty acids were stored as a 100-mmol/L stock in 100% ethanol under nitrogen to ensure minimal oxidation. For oxidation, 3.3 mmol/L EPA, DHA, or arachidonic acid was prepared in phosphate-buffered saline (PBS; pH 4.9) and the sample was incubated at 37°C for 16 hours. Gas chromatography mass spectrometry of the samples revealed a single peak in the unoxidized (native) EPA sample, which was consistent with EPA, whereas oxidized EPA had <5% of the unoxidized EPA and a number of additional peaks that likely correspond to EPA oxidation products (data not shown). Using thiobarbituric acid-reactive substances assay, the aldehyde content of oxidized EPA was 1.78 μmol/L malondialdehye in 100 μmol/L EPA.11 For treatment of cells, oxidized or unoxidized EPA, DHA, or arachidonic acid was diluted in medium containing 20% fetal calf serum to final concentrations between 10 and 100 μmol/L. The vehicle control was media containing PBS.
Enzyme-Linked Immunosorbent Assays for MCP-1 and IL-8
Confluent human umbilical vein endothelial cells (HUVEC) and human microvessel endothelial cells (Clonetics) in 96-well plates were incubated with vehicle, unoxidized, or oxidized EPA, DHA, or arachidonic acid at varying concentrations of 10, 25, 50, 75, and 100 μmol/L for 1 hour. The cells were washed with PBS and fresh medium was added before stimulation with hIL-1? (10 U/mL) or hTNF (10 ng/mL) for 5 hours. Aliquots of the medium were removed for analysis of MCP-1 and IL-8 by enzyme-linked immunosorbent assay (ELISA) using matched antibodies as previously described.19,20 Antibody binding was visualized with horseradish peroxidase-conjugated streptavidin and tetra methyl benzidine liquid substrate system.
Results are expressed as MCP-1 and IL-8 (pg/mL) in the culture medium using values determined on a standard curve. TNF-induced expression of MCP-1 and IL-8 was taken as 100%, and results are also expressed as percent decrease compared with TNF-induced expression of MCP-1 and IL-8.
RNAse Protection Assays
HUVEC were grown to confluence in 100-mm dishes and were pretreated for 1 hour with vehicle control or unoxidized (100 μmol/L) or oxidized EPA (100 μmol/L) in standard growth medium. The HUVEC were washed and the cells were stimulated with hTNF (10 ng/mL) for 3 hours. RNA extraction was performed by guanidine thiocyanate (RNAzol) method of Chomczynski and Sacchi.21 Twenty μg of RNA was subjected to multiprobe RPA system using probes for multiple human chemokine related RNA molecules, including MCP-1 and IL-8, and GAPDH as described in the manufacturer protocols (Riboquant; Pharmingen).
Isolation and Culture of Murine Aortic Endothelial Cells
We isolated endothelial cells from thoracic aortas of 4-week-old 129SV wild-type and PPAR–/– mice.22 Briefly, thoracic aortas were removed from anesthetized mice and rinsed 3times with PBS and placed in endothelial cell growth medium (M199, 20% FBS, 25 μg/mL Gentamycin, 5 mg/50 mL endothelial mitogen, 50 μg/mL, heparin, 2 mmol/L L-glutamine 2 mmol/L). The periadventitial fat around the aortas was carefully removed. The aortas were cut open longitudinally and cut into 4 to 6 pieces of 1- to 2-mm2 and placed endothelial cell side-down in a fibronectin-coated 100-mm culture plate. Initially only a small amount of media was placed around the explants to prevent them from floating. After 12 hours, 8 mL of medium was added. Cells began to expand from the edges of the explant between days 2 to 5. Medium was changed every 2 to 3 days. On day 7, the explants were removed. After 2 to 4 days of explant removal, the cells were confluent and were easily subcultured from passage 2 to 5. These cells express cytokine-inducible mouse adhesion receptors such as vascular cell adhesion molecule-1 and intercellular adhesion molecule-1 (data not shown). Oxidized and unoxidized EPA had no effect on the viability of these cells as determined by tetrazolium salt 3--2-diphenyltetrazolium bromide (MTT) assays (data not shown). For electrophoresis mobility shift assays, the endothelial cells were pretreated for 1 hour with vehicle control or unoxidized (100 μmol/L) or oxidized EPA (100 μmol/L) before stimulation with mouse TNF (10 ng/mL) for 2 hours. Nuclear extracts were then prepared and used in electrophoretic mobility shift assays.
Preparation of Nuclear and Cytoplasmic Extracts
HUVEC were grown to confluence in gelatin-coated 100-mm dishes. HUVEC were pretreated for 1 hour with vehicle control or unoxidized (100 μmol/L) or oxidized EPA (100 μmol/L) in standard growth medium before stimulation with TNF (10 ng/mL) for 0,15, and 60 minutes. Nuclear and cytoplasmic extracts were prepared.23 The cytoplasmic extract was used in Western blot analyses while the nuclear extract was used in the electrophoretic gel shift assays. Protein concentration was determined by Bio-Rad DC protein assay.
Electrophoretic Mobility Gel Shift Assays
Nuclear extracts from HUVEC pretreated with vehicle control or unoxidized or oxidized EPA in standard growth medium before stimulation with TNF (10 ng/mL) for 60 minutes was used in the gel shift assays. The double-stranded oligonucleotide containing the consensus sequence for the NF-B (5'-AGT TGA GGG GAC TTT CCC AGG C-3') and AP-1 (5'-TTC CGG CTG ACT CAT CAA GCG -3') was end-labeled by incubating the oligonucleotide with 32P-labeled ATP and T4 polynucleotide kinase at 37°C for 10 minutes according to standard protocols (Promega).24 For binding reactions, nuclear extracts (10 μg) of HUVEC or mice endothelial cells were incubated in 10 to 15 μL of total reaction volume with 20 μL of binding buffer and 32P-labeled NF-B or AP-1 oligonucleotides for 20 minutes at room temperature. Samples were electrophoresed on a 5% nondenaturing acrylamide gel. The gels were dried and autoradiographed.
Polyclonal antisera against NF-B subunit p65 was used to determine the subunit composition of gel-shifted NF-B complexes; 1 μg of antibody was added to the binding reaction 20 minutes before addition of labeled oligonucleotide probe. Incubation of the nuclear extract from TNF-treated cells with excess cold NF-B or AP-1 oligonucleotide was used to confirm the specificity of AP-1 binding activity.
Western Blot Analyses
Cytoplasmic extracts were resolved by SDS-PAGE on 10% polyacrylamide gels and transferred to polyvinylidene fluoride membranes (Bio-Rad) in 25 mmol/L Tris, pH 8.3, 192 mmol/L glycine, and 20% methanol. Membranes were blocked overnight at 4°C in blocking solution (PBS with 10% milk and 0.2% Tween-20). After blocking the membranes, the primary antibodies to IkB, p50, p65, and ?-actin (Santa Cruz Biotechnology) or phospho-IkB (Cell signaling) were applied according to the manufacturers directions for 1 hour at room temperature.25 Horseradish peroxidase-conjugated secondary antibodies, goat antirabbit IgG, and goat antimouse IgG (Santa Cruz Biotechnology) were used. After each antibody application, blots were washed 3 times in TBS containing 0.2% Tween-20. Antibody complexes were visualized with the use of chemiluminescence.
Statistical Analysis
Data are presented as average±SEM. Statistical significance was assessed by unpaired Student t test.
Results
Oxidized EPA Inhibits Cytokine-Induced Endothelial Expression of MCP-1
We used ELISA methods to determine the effect of oxidized and unoxidized EPA on cytokine-induced endothelial expression of MCP-1 and IL-8. As expected, treatment of HUVECs with TNF and IL-1? results in a 6- to 8-fold increase in MCP-1 and IL-8 expression. Pretreatment of HUVECs with oxidized EPA for 1 hour significantly inhibited TNF and IL-1?–induced expression of MCP-1 whereas incubation with unoxidized EPA had little effect (Figure 1A). The inhibition was dose-dependent; 75 and 100 μmol/L oxidized EPA reduced MCP-1 expression to levels close to those seen for unstimulated HUVECs. When HUVECs were pretreated with 50 and 25 μmol/L oxidized EPA, there was a 30% and a 15% decrease in MCP-1 expression when compared with TNF-treated cells, and a 31% and a 24% decrease when compared with IL-1?–treated cells, respectively. The effect of oxidized EPA on cytokine-induced IL-8 expression was less marked. There was a modest but significant decrease in TNF-induced and IL-1?–induced endothelial expression of IL-8 when the HUVECs were pretreated with 100 μmol/L oxidized EPA, whereas incubation with unoxidized EPA had little effect (Figure 1B). At concentrations of 75 μmol/L and lower, oxidized EPA had no significant effect in inhibiting TNF-induced and IL-1?–induced IL-8 expression. Instead, there was mild increase in the expression of IL-8 when HUVEC were pretreated with lower concentrations of oxidized EPA.
Figure 1. Oxidized EPA inhibits cytokine-induced endothelial MCP-1 and, to a lesser extent, IL-8 expression. HUVECs were pretreated for 1 hour with vehicle (Veh) or unoxidized (100 μmol/L, UnoxEPA), or oxidized EPA (OxEPA) at indicated concentrations before stimulation with TNF (white) or IL-1? (black) for 5 hours. Aliquots of the culture medium were analyzed for (A) MCP-1 and (B) IL-8 by ELISA techniques (n=3). *P<0.0001 and ** P<0.005 compared with EPA/TNF and Veh/TNF. C, HUVECs were pretreated for 1 hour with vehicle, UnoxEPA (100 μmol/L), or OxEPA (100 μmol/L) before stimulation with TNF for 3 hours. Total RNA was extracted and mRNA for MCP-1 and IL-8 was assayed using the multiprobe RPA system.
Similar results were also noted when human microvessel endothelial cells were used instead of HUVECs (Table). We also compared the effect of EPA, DHA, and arachidonic acid on TNF-induced endothelial MCP-1 and IL-8 expression (Table I, available online at http://atvb.ahajournals.org).
Effect of Unoxidized and Oxidized EPA on Expression of Cytokine-Induced Human Microvessel Endothelial Cell MCP-1 and IL-8
Using RNAse protection assays, we determined whether oxidized EPA inhibits cytokine-induced expression of endothelial MCP-1 and IL-8 at the transcriptional level. Pretreatment of HUVECs with oxidized EPA for 1 hour before stimulation with TNF resulted in almost complete absence of the MCP-1 RNA transcripts and a milder decrease in IL-8 mRNA. Unoxidized EPA had no effect in decreasing either MCP-1 or IL-8 mRNA levels (Figure 1C). When the expression of MCP-1 and IL-8 was normalized to GAPDH expression, oxidized EPA pretreatment resulted in a 60% decrease in MCP-1 expression whereas there was only marginal decrease (10%) in IL-8 expression when compared with TNF-treated endothelial cells.
Oxidized EPA Inhibits Cytokine-Induced NF-B Activation but Has no Effect on AP-1
The expression of MCP-1 and IL-8 is regulated through transcription factors NF-B and AP-1.26,27 Studies have shown that although NF-B activation is required for optimal activation of MCP-1, AP-1 activation can bypass the NF-B–mediated IL-8 gene transcription.28
We studied the effect of oxidized versus unoxidized EPA on cytokine-induced activation of NF-B by performing gel shift assays on nuclear extracts of endothelial cells (Figure 2A). Treatment of HUVECs with TNF for 60 minutes results in activation of NF-B as indicated by the gel shifts. Pretreatment of HUVECs with oxidized EPA for 1 hour almost completely inhibited TNF-induced activation of NF-B whereas incubation with unoxidized EPA had little effect. The p65 component of NF-B was confirmed in supershift assays using antibodies to the p65 subunit of NF-B. Next, we studied the effect of oxidized versus unoxidized EPA on TNF-induced activation of AP-1 (Figure 2B). Treatment of HUVECs with TNF for 1 hour results in activation of AP-1 as indicated by the gel shifts. Pretreatment of HUVECs with oxidized EPA or unoxidized EPA for 1 hour had no significant inhibitory effect on TNF-induced activation of AP-1. Incubation of nuclear extract from TNF-treated cells with excess cold AP-1 oligonucleotide was used to confirm the specificity of AP-1 binding activity.
Figure 2. Oxidized EPA inhibits TNF-induced activation of NF-B but has no significant inhibitory effect on AP-1 activation. Nuclear extracts were prepared from HUVEC-treated with vehicle or unoxidized (UnoxEPA, 100 μmol/L) or oxidized EPA (OxEPA, 100 μmol/L) before stimulation with TNF for 1 hour. The nuclear extracts were coincubated with 32P-labeled (A) NF-B oligonucleotide or (B) AP-1 oligonucleotide, and electrophoretic mobility shift assays were performed. In (A), nuclear extracts were preincubated with antibodies to p65 or excess unlabelled NF-B (cold) as indicated. In (B) (lane 5), nuclear extracts were preincubated with excess unlabelled AP-1 before incubation with 32P-labeled AP-1 oligonucleotide (n=3).
Oxidized EPA Promotes Cytoplasmic Retention of p50 and p65 Subunits of NF-B After Cytokine-Induced Activation
Proinflammatory cytokines activate IB kinase complex that phosphorylate NF-B inhibitor, IB, resulting in its conjugation with ubiquitin and subsequent degradation. The free p50/p65 subunits of the NF-B complex then translocate to the nucleus and induce target genes.29
To determine whether oxidized EPA prevented the nuclear translocation of p50 and p65 subunits of NF-B, we performed Western blot analyses using antibodies to p50 (Figure 3A) and p65 (Figure 3B) on the cytosolic extracts of endothelial cells. Baseline levels of p50 and p65 are present in the cytoplasm before stimulation of HUVECs with TNF. Stimulation with TNF for 15 minutes results in almost complete disappearance of p50 and p65 subunits from the cytoplasm. At 60 minutes, both p50 and p65 reappear in the cytoplasm. After pretreatment of HUVECs with oxidized EPA, p50 and p65 are present at 15 and 60 minutes, indicating cytoplasmic retention of the subunits at both time points after stimulation with TNF. Unoxidized EPA had no effect in preventing the nuclear translocation of p50 and p65 subunits, and at 15 minutes of treatment with TNF virtually no cytoplasmic p50 or p65 was identified.
Figure 3. Oxidized EPA inhibits TNF-induced nuclear translocation of p50 and p65. HUVEC were treated with vehicle or unoxidized (UnoxEPA, 100 μmol/L) or oxidized EPA (OxEPA, 100 μmol/L) for 1 hour before stimulation with TNF for varying time points as indicated. Cytoplasmic extracts were prepared and the proteins were separated by SDS-PAGE, transferred to polyvinylidene fluoride membrane, and probed with antibodies to p50, p65, and ?-actin.
Oxidized EPA Does not Inhibit Cytokine-Induced Phosphorylation and Ubiquination of IBa
To determine whether the mechanism of oxidized EPA-mediated inhibition of NF-B was via inhibition of phosphorylation of IB, we performed Western blot analyses on cytoplasmic extracts of HUVECs using antibodies to IB (Figure 4A) and phosphorylated IB (Figure 4B). At baseline, IB was noted in the cytoplasm although no phosphorylated IB was detected. After stimulation with TNF for 15 minutes, minimal IB was detectable in the cytoplasm whereas a significant amount of phosphorylated IB was noted. After 60 minutes, IB reappeared in the cytoplasm even though phosphorylated IB was still present. Similar results were noted in HUVECs pretreated with oxidized EPA; after 15 minutes of stimulation with TNF, there was complete absence of IB in the cytoplasm. This correlated with appearance of phosphorylated IB. At 60 minutes, even though phosphorylated IB was still present, IB reappeared in the cytoplasm (although the expression was slightly decreased). These findings are similar to HUVECs with no oxidized EPA pretreatment, indicating that oxidized EPA does not prevent phosphorylation and ubiquination of IB, suggesting a different mechanism for the inhibition of nuclear translocation of p50/p65 subunits.
Figure 4. Oxidized EPA does not inhibit TNF-induced phosphorylation and ubiquination of IB. HUVEC were treated with vehicle or unoxidized (UnoxEPA, 100 μmol/L) or oxidized EPA (OxEPA, 100 μmol/L) for 1 hour before stimulation with TNF for varying time points as indicated. Cytoplasmic extracts were prepared and the proteins were separated by SDS-PAGE, transferred to polyvinylidene fluoride membrane, and probed with antibodies to (A) IB and (B) phosphorylated-IB.
Oxidized EPA Does not Inhibit Cytokine-Induced Activation of NF-B in PPAR-Deficient Endothelial Cells
Our previous studies have shown that oxidized EPA is a potent activator of PPAR.11 To determine whether PPAR is required for the inhibitory effects of oxidized EPA on NF-B activation, we isolated and grew endothelial cells from wild-type and PPAR–/– mice thoracic aortas. Treatment of wild-type endothelial cells with TNF for 2 hours resulted in activation of NF-B as indicated by the gel shifts. Pretreatment of wild-type endothelial cells with oxidized EPA for 1 hour almost completely inhibited TNF-induced activation of NF-B, whereas incubation with unoxidized EPA had little effect (Figure 5A). Endothelial cells from PPAR–/– mice showed TNF-induced activation of NF-B. However, oxidized EPA pretreatment had no significant inhibitory effect on TNF induced activation of NF-B in PPAR–/– endothelial cells (Figure 5B).
Figure 5. Oxidized EPA inhibits TNF-induced activation of NF-B wild-type endothelial cells but has no inhibitory effect in endothelial cells derived from PPAR–/– mice. Nuclear extracts were prepared from thoracic aortic endothelial cells derived from (A) wild-type and (B) PPAR–/– mice treated with vehicle or unoxidized EPA (EPA, 100 μmol/L) or oxidized EPA (OxEPA, 100 μmol/L) before stimulation with TNF for 2 hours. Nuclear extracts were preincubated with antibodies to p65 as indicated. The nuclear extracts were coincubated with 32P-labeled NF-B oligonucleotide, and electrophoresis mobility shift assays were performed (n=3).
Discussion
Omega-3 fatty acids in fish oil has been reported to improve the prognosis of several chronic inflammatory diseases characterized by leukocyte accumulation, including atherosclerosis, inflammatory bowel disease, rheumatoid arthritis, psoriasis, etc.1–4
Omega-3 fatty acids, such as EPA and DHA, are highly polyunsaturated and readily undergo oxidation at ambient and subambient temperatures, even in the absence of exogenous oxidizing reagents.5,6 In view of the ease with which omega-3 PUFA spontaneously oxidize and in vivo data suggesting extensive accumulation of oxidation products after fish oil consumption, we investigated the possibility that oxidized omega-3 fatty acids may be an important component of the observed antiinflammatory effects of fish oil. In our previous studies, we showed that oxidized EPA and not unoxidized EPA pretreatment of HUVEC inhibits leukocyte adhesion to cytokine-stimulated HUVEC, and this effect is mediated through a PPAR-dependent pathway.11,12 In our present studies, we extend these observations and show that oxidized EPA is also effective in inhibiting cytokine-induced endothelial chemokine expression, particularly MCP-1, and propose a mechanism for these antiinflammatory effects.
The fact that oxidation of the omega-3 fatty acids is required for the aforementioned antiinflammatory effects is also pointed out by other studies. Nohe et al and De Caterina et al have shown that prolonged incubation of endothelial cells with native omega-3 fatty acids (26 hours of total exposure, 6 hours before and 20 hours after TNF stimulation) results in inhibition of cytokine and adhesion molecule expression, whereas shorter incubation periods (6 hours) has no effect.30–32 This suggests that oxidation of omega-3 fatty acids takes place during the long incubation period, and that the oxidation product(s) and not the native omega-3 fatty acids are most likely responsible for the inhibition of leukocyte–endothelial interactions.
To ascertain that the inhibition of chemokine expression was not caused by the cytotoxic effects of oxidized EPA, we performed MTT assays, which showed that oxidized EPA in the doses used for our experiments had no effect on the viability of endothelial cells (data not shown). Also, oxidized EPA did not have any effect on the constitutively expressed surface proteins such as von Willebrand factor, endoglin, and human leukocyte antigen class I molecules.11,12
The expression of MCP-1 and IL-8 is regulated through activation of NF-B. NF-B activation appears to be necessary for the induction of chemokine genes, and deletion of NF-B binding sites results in an inability to induce these genes.26 In these studies, we show that oxidized EPA and not unoxidized EPA inhibits cytokine-induced activation of NF-B and promotes cytosolic retention of the p50 and p65 subunits. Hence, the oxidized EPA-mediated inhibition of MCP-1 and IL-8 in cytokine-stimulated endothelial cells could be explained by an inhibitory effect of oxidized EPA on NF-B activity.
The difference in the extent of inhibition of MCP-1 and IL-8 expression by oxidized EPA is most likely caused by the differential effects of oxidized EPA on endothelial NF-B and AP-1 activation. Oxidized EPA almost completely inhibits endothelial NF-B activation, whereas it had minimal effects in preventing AP-1 activation. Because AP-1 activation alone can result in IL-8 expression,28 oxidized EPA had only a mild inhibitory effect on IL-8 expression.
What is the mechanism for the oxidized EPA-mediated inhibition of cytokine-induced NF-B activation? We first hypothesized that oxidized EPA inhibits phosphorylation of IB by inhibiting the IKK (kinase) complex. However, it is unlikely that oxidized EPA inhibits IKK (kinase) activity because phosphorylation and ubiquination of IkB is noted in oxidized EPA-treated endothelial cells. In fact, when normalized to actin controls, oxidized EPA pretreatment before TNF stimulation for 60 minutes resulted in 30% more p-IkB when compared with TNF-treated cells. Also, it unlikely that oxidized EPA induces IkB expression because after 15 minutes of cytokine stimulation, no IkB is noted in the cytoplasm of cells pretreated with oxidized EPA. After 60 minutes, IB starts to appear in the cytoplasm of oxidized EPA pretreated cells and its concentration is, in fact, somewhat less than the TNF-treated cells. Thus, it is likely that oxidized EPA does not prevent NF-B activation by increasing the expression of IB.
In our previous studies we noted that oxidized EPA is a potent activator of PPAR and that PPAR is needed for the inhibitory effects of oxidized EPA on leukocyte–endothelial interactions,12 which lead us to hypothesize that oxidized EPA might inhibit cytokine-induced NF-B activation through a PPAR-dependent pathway. Here, we show that although oxidized EPA inhibits cytokine-induced NF-B activation in wild-type cells, it has no inhibitory effects in PPAR-deficient endothelial cells, suggesting that oxidized EPA mediates its inhibitory effects on NF-B through PPAR. The PPAR-mediated inhibitory effects of oxidized EPA on NF-B activation are possibly through direct interactions of PPAR with the p50/p65 subunits. Delerive et al (1999) have shown, using glutathione S-transferase pull-down experiments, that after fibrate (PPAR agonist) treatment, PPAR physically interacts with p65 in vitro.33
Taken together with our previous studies,11,12 we show that auto-oxidation of omega-3 fatty acids results in the generation of oxidized compounds with potent antiinflammatory properties that inhibit proinflammatory responses such as leukocyte adhesion receptor and chemokine expression. The central theme for the antiinflammatory effects of oxidized omega-3 fatty acids is likely through inhibition of NF-B via a PPAR-dependent pathway. The oxidation of omega-3 fatty acids is likely to occur in areas of active inflammation caused by increased expression of oxidative enzymes (eg, NADPH oxidase, myeloperoxidase, cyclooxygenase, lipoxygenase) and the generation of reactive oxygen species in these areas, which are capable of oxidizing PUFAs. The identification of these products could result in potent, low-toxicity, proinflammatory response inhibitors with potent PPAR agonist and anti-NF-B properties for the treatment of inflammatory diseases.
Acknowledgments
We thank Dr. Lynn Stoll for assistance with ELISA assays and Joel Carl for assistance with the figures. We acknowledge support (Scientist Development Grant) from the American Heart Association (S.S.)
References
Kromhout D, Bosschieter EB, de Lezenne Coulander C. The inverse relation between fish consumption and 20-year mortality from coronary heart disease. N Engl J Med. 1985; 312: 1205–1209.
Belluzzi A, Brignola C, Campieri M, Pera A, Boschi S, Miglioli M. Effect of and enteric coated fish oil preparation on relapses in Crohn’s disease. N Engl J Med. 1996; 334: 1557–1560.
Donadio JV Jr., Bergstralh EJ, Offord KP, Spencer DC, Holley KE. A controlled trial of fish oil in IgA nephropathy. N Engl J Med. 1994; 331: 1194–1199.
Volker D, Fitzgerald P, Major G, Garg M. Efficacy of fish oil concentrate in the treatment of rheumatoid arthritis. J Rheumatol. 2000; 27: 2343–2346.
Chan HWS. The mechanism of autoxidation. Autoxidation of unsaturated lipids. Academic Press: UK; 1987.
Porter NA, Caldwell SE, Mills KA. Mechanisms of free radical oxidation of unsaturated lipids. Lipids. 1995; 30: 277–290.
Gonzalez MJ, Gray JI, Schemmel RA, Dugan L Jr, Welsch CW. Lipid peroxidation products are elevated in fish oil diets even in the presence of added antioxidants. J Nutr. 1992; 122: 2190–2195.
Wander RC, Du S-H. Oxidation of plasma proteins is not increased after supplementation with eicosapentaenoic and docosahexaenoic acids. Am J Clin Nutr. 2000; 72: 731–737.
Meydani M, Natiello F, Goldin B, Free N, Woods M, Schaefer E, Blumberg JB, Gorbach SL. Effect of long-term fish oil supplementation on vitamin E status and lipid peroxidation in women. J Nutr. 1991; 121: 484–91.
Higdon JV, Liu J, Du SH, Morrow JD, Ames BN, Wander RC. Supplementation of postmenopausal women with fish oil rich in eicosapentaenoic acid and docosahexaenoic acid is not associated with greater in vivo lipid peroxidation compared with oils rich in oleate and linoleate as assessed by plasma malondialdehyde and F2-isoprostanes Am J Clin Nutr. 2000; 72: 714–722.
Sethi S, Eastman AY, Eaton JW. Inhibition of phagocyte-endothelium interactions by oxidized fatty acids: a natural anti-inflammatory mechanism? J Lab Clin Med. 1996; 128: 27–38.
Sethi S, Ziouzenkova O, Ni H, Wagner DD, Plutzky J, Mayadas TN. Oxidized omega-3 fatty acids in fish oil inhibit leukocyte-endothelial interactions through activation of PPAR Blood. 2002; 100: 1340–1346.
Gerszten RE, Garcia-Zepeda EA, Lim YC, Yoshida M, Ding HA, Gimbrone MA Jr, Luster AD, Luscinskas FW, Rosenzweig A. MCP-1 and IL-8 trigger firm adhesion of monocytes to vascular endothelium under flow conditions. Nature. 1999; 398: 718–723.
Maus U, Henning S, Wenschuh H, Mayer K, Seeger W, Lohmeyer J. Role of endothelial MCP-1 in monocyte adhesion to inflamed human endothelium under physiological flow Am J Physiol. 2002; 283: H2584–H2591.
Shimizu H, Maruyama S, Yuzawa Y, Kato T, Miki Y, Suzuki S, Sato W, Morita Y, Maruyama H, Egashira K, Matsuo S. Anti-monocyte chemoattractant protein-1 gene therapy attenuates renal injury induced by protein-overload proteinuria. J Am Soc Nephrol. 2003; 14: 1496–505.
Egashira K. Molecular mechanisms mediating inflammation in vascular disease: special reference to monocyte chemoattractant protein-1. Hypertension. 2003; 41: 834–41.
Dawson J, Miltz W, Mir AK, Wiessner C. Targeting monocyte chemoattractant protein-1 signalling in disease. Exper Opin Ther Targets. 2003; 7: 35–48.
Halliwell B, Gutteridge JM. Free radicals in biology and medicine, 3rd ed. New York, NY: Oxford University Press; 1993.
Denning GM, Wollenweber LA, Railsback MA, Cox CD, Stoll LL, Britigan BE. Pseudomonas Pyocyanin increases interleukin-8 expression by human airway epithelial cells. Infect Immun. 1998; 66: 5777–5784.
Rice JB, Stoll LL, Li W-G, Denning GM, Weydert J, Charipar E, Richenbacher WE, Miller FJ Jr, Weintraub NL. Low-Level Endotoxin Induces Potent Inflammatory Activation of Human Blood Vessels: Inhibition by Statins. Arterioscler Thromb Vasc Biol. 2003; 23: 1576–1582.
Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem. 1987; 162: 156–9.
Huang H, McIntosh J, Hoyt DG. An efficient, nonenzymatic method for isolation and culture of murine aortic endothelial cells and their response to inflammatory stimuli. In Vitro Cell Dev Biol Anim. 2003; 39: 43–50.
Dignam JD, Lebovitz RM, Roeder RG. Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei. Nucleic Acids Res. 1983; 11: 1475–1489.
Canty TG Jr., Boyle EM Jr, Farr A, Morgan EN, Verrier ED, Pohlman TH. Oxidative stress induces NF-kappaB nuclear translocation without degradation of IkappaB. Circulation. 1999; 100: 361–364.
Zandi E, Chen Y, Karin M. Direct phosphorylation of IB by IKK and IKK?: discrimination between free and NF-B-bound substrate. Science. 1998; 281: 1360–1363.
Martin T, Cardarelli PM, Parry GC, Felts KA, Cobb RR. Cytokine induction of monocyte chemoattractant protein-1 gene expression in human endothelial cells depends on the cooperative action of NF-kappa B and AP-1. Eur J Immunol. 1997; 27: 1091–1097.
Roebuck KA. Regulation of Interleukin-8 gene expression. J Interferon Cytokine Res. 1999; 19: 429–438.
Hipp MS, Urbich C, Mayer P, Wischhusen J, Weller M, Kracht M, Spyridopoulos I. Proteosome inhibition leads to NF-kB-independent IL-8 transactivation in human endothelial cells through induction of AP-1. Eur J Immunol. 2002; 32: 2208–2217.
Lenard MJ, Baltimore D. NF-kB: a pleiotropic mediator of inducible and tissue specific gene control. Cell. 1988; 58: 227–229.
Nohe B, Johannes T, Dieterich HJ, Mayer K, Merfels M, Grimminger F, Seeger W. Antiinflammatory effects of omega-3 fatty acids vary at different stages of inflammation. Am J Physiol. 2003; 285: H2248–H2250.
De Caterina R, Bernini W, Carluccio MA, Liao JK, Libby P. Structural requirements for inhibition of cytokine-induced endothelial activation by unsaturated fatty acids. J Lipid Res. 1998; 39: 1062–1070.
De Caterina R CM, Clinton SK, Gimbrone MA Jr, Libby P. The omega-3 fatty acid docosahexaenoate reduces cytokine-induced expression of proatherogenic and proinflammatory proteins in human endothelial cells. Arterioscler Thromb Vasc Biol. 1994; 14: 1829–1836.
Delerive P, De Bosscher K, Besnard S, Vanden Berghe W, Peters JM, Gonzalez FJ, Fruchart J-C, Tedgui A, Haegeman G, Staels B. Peroxisome proliferator-activated receptor alpha negatively regulates the vascular inflammatory gene response by negative cross-talk with transcription factors NF-kappa B and AP-1. J Biol Chem. 1999; 274: 32048–32054.(Archana Mishra; Ashok Cha)
ABSTRACT
Objective— The aim of this study was to determine the effects of oxidized versus native omega-3 fatty acids on the endothelial expression of chemokines MCP-1 and IL-8, and, if effective in inhibiting chemokine expression, to determine the mechanism for the inhibition of chemokine expression.
Methods and Results— Using enzyme-linked immunosorbent assays, we show that oxidized EPA and DHA but not unoxidized EPA or DHA inhibit cytokine-induced endothelial expression of monocyte chemoattractant protein (MCP)-1 and, to a lesser extent, IL-8. In electrophoretic mobility shift assays, oxidized EPA but not unoxidized EPA potently inhibited cytokine-induced activation of endothelial nuclear factor-B (NF-B). Using Western blot analyses, we show that the inhibition of NF-B activation was not caused by prevention of phosphorylation of IB because oxidized EPA did not inhibit cytokine-induced phosphorylation and ubiquination of IB. Furthermore, oxidized EPA inhibited NF-B activation in endothelial cells derived from wild-type mice but had no inhibitory effects on NF-B activation in endothelial cells derived from peroxisome proliferator-activated receptor (PPAR)-deficient mice, indicating that oxidized EPA requires PPAR for its inhibitory effects on NF-B.
Conclusions— These studies show that the antiinflammatory effects of fish oil may result from the inhibitory effects of oxidized omega-3 fatty acids on NF-B activation via a PPAR-dependent pathway.
Key Words: monocyte chemoattractant protein-1 ? oxidized omega-3 fatty acids ? oxidized eicosapentaenoic acid ? nuclear factor-B ? PPARa
Introduction
Consumption of marine fish oil has been reported to improve the prognosis of several chronic inflammatory diseases characterized by leukocyte accumulation and leukocyte-mediated tissue injury, including atherosclerosis, IgA nephropathy, inflammatory bowel disease, rheumatoid arthritis, etc.1–4 These beneficial effects of fish oil have been associated with the omega-3 polyunsaturated fatty acids (PUFAs), eicosapentaenoic acid (EPA), and docosahexaenoic acid (DHA), which are abundant in marine fish oil.
EPA and DHA are highly polyunsaturated and easily undergo auto-oxidation.5,6 In fact, it is very difficult to avoid the oxidation of these very labile fatty acids. More importantly, in vivo, a large increase in tissue and plasma accumulation of both omega-3 fatty acids and fatty acid oxidation products is noted in subjects consuming fish oil, even after addition of antioxidant supplements to the diet.7–10 This suggests the possibility that oxidized omega-3 fatty acids may be an important component of the observed antiinflammatory effects of fish oil. Indeed, our previous studies have shown that oxidized EPA, and not unoxidized EPA, potently inhibits leukocyte–endothelial interactions, both in vitro and in vivo, through a peroxisome proliferator-activated receptor (PPAR)-dependent mechanism.11,12
One of the early events in inflammation is the upregulation of endothelial chemokines, monocyte chemoattractant protein-1 (MCP-1) and IL-8, in response to proinflammatory cytokines such as tumor necrosis factor (TNF)- and IL-1. MCP-1 and IL-8 in turn promote leukocyte chemotaxis, adhesion, and transendothelial migration,13,14 and neutralization of MCP-1 has been shown to attenuate in vivo injury arising from inflammatory mechanisms.15–17 The aim of this study was to determine the effects of oxidized versus native omega-3 fatty acids on the endothelial expression of chemokines MCP-1 and IL-8, and, if effective in inhibiting chemokine expression, to determine the mechanism for the inhibition of chemokine expression.
Methods
Preparation of Fatty Acids
EPA, DHA, and arachidonic acid were purchased from Cayman Chemicals. They were relatively unoxidized as assessed by thiobarbituric acid-reactive substances assay using malondialdehyde as a standard; this assay is a measure of fatty acid oxidation.18 The fatty acids were stored as a 100-mmol/L stock in 100% ethanol under nitrogen to ensure minimal oxidation. For oxidation, 3.3 mmol/L EPA, DHA, or arachidonic acid was prepared in phosphate-buffered saline (PBS; pH 4.9) and the sample was incubated at 37°C for 16 hours. Gas chromatography mass spectrometry of the samples revealed a single peak in the unoxidized (native) EPA sample, which was consistent with EPA, whereas oxidized EPA had <5% of the unoxidized EPA and a number of additional peaks that likely correspond to EPA oxidation products (data not shown). Using thiobarbituric acid-reactive substances assay, the aldehyde content of oxidized EPA was 1.78 μmol/L malondialdehye in 100 μmol/L EPA.11 For treatment of cells, oxidized or unoxidized EPA, DHA, or arachidonic acid was diluted in medium containing 20% fetal calf serum to final concentrations between 10 and 100 μmol/L. The vehicle control was media containing PBS.
Enzyme-Linked Immunosorbent Assays for MCP-1 and IL-8
Confluent human umbilical vein endothelial cells (HUVEC) and human microvessel endothelial cells (Clonetics) in 96-well plates were incubated with vehicle, unoxidized, or oxidized EPA, DHA, or arachidonic acid at varying concentrations of 10, 25, 50, 75, and 100 μmol/L for 1 hour. The cells were washed with PBS and fresh medium was added before stimulation with hIL-1? (10 U/mL) or hTNF (10 ng/mL) for 5 hours. Aliquots of the medium were removed for analysis of MCP-1 and IL-8 by enzyme-linked immunosorbent assay (ELISA) using matched antibodies as previously described.19,20 Antibody binding was visualized with horseradish peroxidase-conjugated streptavidin and tetra methyl benzidine liquid substrate system.
Results are expressed as MCP-1 and IL-8 (pg/mL) in the culture medium using values determined on a standard curve. TNF-induced expression of MCP-1 and IL-8 was taken as 100%, and results are also expressed as percent decrease compared with TNF-induced expression of MCP-1 and IL-8.
RNAse Protection Assays
HUVEC were grown to confluence in 100-mm dishes and were pretreated for 1 hour with vehicle control or unoxidized (100 μmol/L) or oxidized EPA (100 μmol/L) in standard growth medium. The HUVEC were washed and the cells were stimulated with hTNF (10 ng/mL) for 3 hours. RNA extraction was performed by guanidine thiocyanate (RNAzol) method of Chomczynski and Sacchi.21 Twenty μg of RNA was subjected to multiprobe RPA system using probes for multiple human chemokine related RNA molecules, including MCP-1 and IL-8, and GAPDH as described in the manufacturer protocols (Riboquant; Pharmingen).
Isolation and Culture of Murine Aortic Endothelial Cells
We isolated endothelial cells from thoracic aortas of 4-week-old 129SV wild-type and PPAR–/– mice.22 Briefly, thoracic aortas were removed from anesthetized mice and rinsed 3times with PBS and placed in endothelial cell growth medium (M199, 20% FBS, 25 μg/mL Gentamycin, 5 mg/50 mL endothelial mitogen, 50 μg/mL, heparin, 2 mmol/L L-glutamine 2 mmol/L). The periadventitial fat around the aortas was carefully removed. The aortas were cut open longitudinally and cut into 4 to 6 pieces of 1- to 2-mm2 and placed endothelial cell side-down in a fibronectin-coated 100-mm culture plate. Initially only a small amount of media was placed around the explants to prevent them from floating. After 12 hours, 8 mL of medium was added. Cells began to expand from the edges of the explant between days 2 to 5. Medium was changed every 2 to 3 days. On day 7, the explants were removed. After 2 to 4 days of explant removal, the cells were confluent and were easily subcultured from passage 2 to 5. These cells express cytokine-inducible mouse adhesion receptors such as vascular cell adhesion molecule-1 and intercellular adhesion molecule-1 (data not shown). Oxidized and unoxidized EPA had no effect on the viability of these cells as determined by tetrazolium salt 3--2-diphenyltetrazolium bromide (MTT) assays (data not shown). For electrophoresis mobility shift assays, the endothelial cells were pretreated for 1 hour with vehicle control or unoxidized (100 μmol/L) or oxidized EPA (100 μmol/L) before stimulation with mouse TNF (10 ng/mL) for 2 hours. Nuclear extracts were then prepared and used in electrophoretic mobility shift assays.
Preparation of Nuclear and Cytoplasmic Extracts
HUVEC were grown to confluence in gelatin-coated 100-mm dishes. HUVEC were pretreated for 1 hour with vehicle control or unoxidized (100 μmol/L) or oxidized EPA (100 μmol/L) in standard growth medium before stimulation with TNF (10 ng/mL) for 0,15, and 60 minutes. Nuclear and cytoplasmic extracts were prepared.23 The cytoplasmic extract was used in Western blot analyses while the nuclear extract was used in the electrophoretic gel shift assays. Protein concentration was determined by Bio-Rad DC protein assay.
Electrophoretic Mobility Gel Shift Assays
Nuclear extracts from HUVEC pretreated with vehicle control or unoxidized or oxidized EPA in standard growth medium before stimulation with TNF (10 ng/mL) for 60 minutes was used in the gel shift assays. The double-stranded oligonucleotide containing the consensus sequence for the NF-B (5'-AGT TGA GGG GAC TTT CCC AGG C-3') and AP-1 (5'-TTC CGG CTG ACT CAT CAA GCG -3') was end-labeled by incubating the oligonucleotide with 32P-labeled ATP and T4 polynucleotide kinase at 37°C for 10 minutes according to standard protocols (Promega).24 For binding reactions, nuclear extracts (10 μg) of HUVEC or mice endothelial cells were incubated in 10 to 15 μL of total reaction volume with 20 μL of binding buffer and 32P-labeled NF-B or AP-1 oligonucleotides for 20 minutes at room temperature. Samples were electrophoresed on a 5% nondenaturing acrylamide gel. The gels were dried and autoradiographed.
Polyclonal antisera against NF-B subunit p65 was used to determine the subunit composition of gel-shifted NF-B complexes; 1 μg of antibody was added to the binding reaction 20 minutes before addition of labeled oligonucleotide probe. Incubation of the nuclear extract from TNF-treated cells with excess cold NF-B or AP-1 oligonucleotide was used to confirm the specificity of AP-1 binding activity.
Western Blot Analyses
Cytoplasmic extracts were resolved by SDS-PAGE on 10% polyacrylamide gels and transferred to polyvinylidene fluoride membranes (Bio-Rad) in 25 mmol/L Tris, pH 8.3, 192 mmol/L glycine, and 20% methanol. Membranes were blocked overnight at 4°C in blocking solution (PBS with 10% milk and 0.2% Tween-20). After blocking the membranes, the primary antibodies to IkB, p50, p65, and ?-actin (Santa Cruz Biotechnology) or phospho-IkB (Cell signaling) were applied according to the manufacturers directions for 1 hour at room temperature.25 Horseradish peroxidase-conjugated secondary antibodies, goat antirabbit IgG, and goat antimouse IgG (Santa Cruz Biotechnology) were used. After each antibody application, blots were washed 3 times in TBS containing 0.2% Tween-20. Antibody complexes were visualized with the use of chemiluminescence.
Statistical Analysis
Data are presented as average±SEM. Statistical significance was assessed by unpaired Student t test.
Results
Oxidized EPA Inhibits Cytokine-Induced Endothelial Expression of MCP-1
We used ELISA methods to determine the effect of oxidized and unoxidized EPA on cytokine-induced endothelial expression of MCP-1 and IL-8. As expected, treatment of HUVECs with TNF and IL-1? results in a 6- to 8-fold increase in MCP-1 and IL-8 expression. Pretreatment of HUVECs with oxidized EPA for 1 hour significantly inhibited TNF and IL-1?–induced expression of MCP-1 whereas incubation with unoxidized EPA had little effect (Figure 1A). The inhibition was dose-dependent; 75 and 100 μmol/L oxidized EPA reduced MCP-1 expression to levels close to those seen for unstimulated HUVECs. When HUVECs were pretreated with 50 and 25 μmol/L oxidized EPA, there was a 30% and a 15% decrease in MCP-1 expression when compared with TNF-treated cells, and a 31% and a 24% decrease when compared with IL-1?–treated cells, respectively. The effect of oxidized EPA on cytokine-induced IL-8 expression was less marked. There was a modest but significant decrease in TNF-induced and IL-1?–induced endothelial expression of IL-8 when the HUVECs were pretreated with 100 μmol/L oxidized EPA, whereas incubation with unoxidized EPA had little effect (Figure 1B). At concentrations of 75 μmol/L and lower, oxidized EPA had no significant effect in inhibiting TNF-induced and IL-1?–induced IL-8 expression. Instead, there was mild increase in the expression of IL-8 when HUVEC were pretreated with lower concentrations of oxidized EPA.
Figure 1. Oxidized EPA inhibits cytokine-induced endothelial MCP-1 and, to a lesser extent, IL-8 expression. HUVECs were pretreated for 1 hour with vehicle (Veh) or unoxidized (100 μmol/L, UnoxEPA), or oxidized EPA (OxEPA) at indicated concentrations before stimulation with TNF (white) or IL-1? (black) for 5 hours. Aliquots of the culture medium were analyzed for (A) MCP-1 and (B) IL-8 by ELISA techniques (n=3). *P<0.0001 and ** P<0.005 compared with EPA/TNF and Veh/TNF. C, HUVECs were pretreated for 1 hour with vehicle, UnoxEPA (100 μmol/L), or OxEPA (100 μmol/L) before stimulation with TNF for 3 hours. Total RNA was extracted and mRNA for MCP-1 and IL-8 was assayed using the multiprobe RPA system.
Similar results were also noted when human microvessel endothelial cells were used instead of HUVECs (Table). We also compared the effect of EPA, DHA, and arachidonic acid on TNF-induced endothelial MCP-1 and IL-8 expression (Table I, available online at http://atvb.ahajournals.org).
Effect of Unoxidized and Oxidized EPA on Expression of Cytokine-Induced Human Microvessel Endothelial Cell MCP-1 and IL-8
Using RNAse protection assays, we determined whether oxidized EPA inhibits cytokine-induced expression of endothelial MCP-1 and IL-8 at the transcriptional level. Pretreatment of HUVECs with oxidized EPA for 1 hour before stimulation with TNF resulted in almost complete absence of the MCP-1 RNA transcripts and a milder decrease in IL-8 mRNA. Unoxidized EPA had no effect in decreasing either MCP-1 or IL-8 mRNA levels (Figure 1C). When the expression of MCP-1 and IL-8 was normalized to GAPDH expression, oxidized EPA pretreatment resulted in a 60% decrease in MCP-1 expression whereas there was only marginal decrease (10%) in IL-8 expression when compared with TNF-treated endothelial cells.
Oxidized EPA Inhibits Cytokine-Induced NF-B Activation but Has no Effect on AP-1
The expression of MCP-1 and IL-8 is regulated through transcription factors NF-B and AP-1.26,27 Studies have shown that although NF-B activation is required for optimal activation of MCP-1, AP-1 activation can bypass the NF-B–mediated IL-8 gene transcription.28
We studied the effect of oxidized versus unoxidized EPA on cytokine-induced activation of NF-B by performing gel shift assays on nuclear extracts of endothelial cells (Figure 2A). Treatment of HUVECs with TNF for 60 minutes results in activation of NF-B as indicated by the gel shifts. Pretreatment of HUVECs with oxidized EPA for 1 hour almost completely inhibited TNF-induced activation of NF-B whereas incubation with unoxidized EPA had little effect. The p65 component of NF-B was confirmed in supershift assays using antibodies to the p65 subunit of NF-B. Next, we studied the effect of oxidized versus unoxidized EPA on TNF-induced activation of AP-1 (Figure 2B). Treatment of HUVECs with TNF for 1 hour results in activation of AP-1 as indicated by the gel shifts. Pretreatment of HUVECs with oxidized EPA or unoxidized EPA for 1 hour had no significant inhibitory effect on TNF-induced activation of AP-1. Incubation of nuclear extract from TNF-treated cells with excess cold AP-1 oligonucleotide was used to confirm the specificity of AP-1 binding activity.
Figure 2. Oxidized EPA inhibits TNF-induced activation of NF-B but has no significant inhibitory effect on AP-1 activation. Nuclear extracts were prepared from HUVEC-treated with vehicle or unoxidized (UnoxEPA, 100 μmol/L) or oxidized EPA (OxEPA, 100 μmol/L) before stimulation with TNF for 1 hour. The nuclear extracts were coincubated with 32P-labeled (A) NF-B oligonucleotide or (B) AP-1 oligonucleotide, and electrophoretic mobility shift assays were performed. In (A), nuclear extracts were preincubated with antibodies to p65 or excess unlabelled NF-B (cold) as indicated. In (B) (lane 5), nuclear extracts were preincubated with excess unlabelled AP-1 before incubation with 32P-labeled AP-1 oligonucleotide (n=3).
Oxidized EPA Promotes Cytoplasmic Retention of p50 and p65 Subunits of NF-B After Cytokine-Induced Activation
Proinflammatory cytokines activate IB kinase complex that phosphorylate NF-B inhibitor, IB, resulting in its conjugation with ubiquitin and subsequent degradation. The free p50/p65 subunits of the NF-B complex then translocate to the nucleus and induce target genes.29
To determine whether oxidized EPA prevented the nuclear translocation of p50 and p65 subunits of NF-B, we performed Western blot analyses using antibodies to p50 (Figure 3A) and p65 (Figure 3B) on the cytosolic extracts of endothelial cells. Baseline levels of p50 and p65 are present in the cytoplasm before stimulation of HUVECs with TNF. Stimulation with TNF for 15 minutes results in almost complete disappearance of p50 and p65 subunits from the cytoplasm. At 60 minutes, both p50 and p65 reappear in the cytoplasm. After pretreatment of HUVECs with oxidized EPA, p50 and p65 are present at 15 and 60 minutes, indicating cytoplasmic retention of the subunits at both time points after stimulation with TNF. Unoxidized EPA had no effect in preventing the nuclear translocation of p50 and p65 subunits, and at 15 minutes of treatment with TNF virtually no cytoplasmic p50 or p65 was identified.
Figure 3. Oxidized EPA inhibits TNF-induced nuclear translocation of p50 and p65. HUVEC were treated with vehicle or unoxidized (UnoxEPA, 100 μmol/L) or oxidized EPA (OxEPA, 100 μmol/L) for 1 hour before stimulation with TNF for varying time points as indicated. Cytoplasmic extracts were prepared and the proteins were separated by SDS-PAGE, transferred to polyvinylidene fluoride membrane, and probed with antibodies to p50, p65, and ?-actin.
Oxidized EPA Does not Inhibit Cytokine-Induced Phosphorylation and Ubiquination of IBa
To determine whether the mechanism of oxidized EPA-mediated inhibition of NF-B was via inhibition of phosphorylation of IB, we performed Western blot analyses on cytoplasmic extracts of HUVECs using antibodies to IB (Figure 4A) and phosphorylated IB (Figure 4B). At baseline, IB was noted in the cytoplasm although no phosphorylated IB was detected. After stimulation with TNF for 15 minutes, minimal IB was detectable in the cytoplasm whereas a significant amount of phosphorylated IB was noted. After 60 minutes, IB reappeared in the cytoplasm even though phosphorylated IB was still present. Similar results were noted in HUVECs pretreated with oxidized EPA; after 15 minutes of stimulation with TNF, there was complete absence of IB in the cytoplasm. This correlated with appearance of phosphorylated IB. At 60 minutes, even though phosphorylated IB was still present, IB reappeared in the cytoplasm (although the expression was slightly decreased). These findings are similar to HUVECs with no oxidized EPA pretreatment, indicating that oxidized EPA does not prevent phosphorylation and ubiquination of IB, suggesting a different mechanism for the inhibition of nuclear translocation of p50/p65 subunits.
Figure 4. Oxidized EPA does not inhibit TNF-induced phosphorylation and ubiquination of IB. HUVEC were treated with vehicle or unoxidized (UnoxEPA, 100 μmol/L) or oxidized EPA (OxEPA, 100 μmol/L) for 1 hour before stimulation with TNF for varying time points as indicated. Cytoplasmic extracts were prepared and the proteins were separated by SDS-PAGE, transferred to polyvinylidene fluoride membrane, and probed with antibodies to (A) IB and (B) phosphorylated-IB.
Oxidized EPA Does not Inhibit Cytokine-Induced Activation of NF-B in PPAR-Deficient Endothelial Cells
Our previous studies have shown that oxidized EPA is a potent activator of PPAR.11 To determine whether PPAR is required for the inhibitory effects of oxidized EPA on NF-B activation, we isolated and grew endothelial cells from wild-type and PPAR–/– mice thoracic aortas. Treatment of wild-type endothelial cells with TNF for 2 hours resulted in activation of NF-B as indicated by the gel shifts. Pretreatment of wild-type endothelial cells with oxidized EPA for 1 hour almost completely inhibited TNF-induced activation of NF-B, whereas incubation with unoxidized EPA had little effect (Figure 5A). Endothelial cells from PPAR–/– mice showed TNF-induced activation of NF-B. However, oxidized EPA pretreatment had no significant inhibitory effect on TNF induced activation of NF-B in PPAR–/– endothelial cells (Figure 5B).
Figure 5. Oxidized EPA inhibits TNF-induced activation of NF-B wild-type endothelial cells but has no inhibitory effect in endothelial cells derived from PPAR–/– mice. Nuclear extracts were prepared from thoracic aortic endothelial cells derived from (A) wild-type and (B) PPAR–/– mice treated with vehicle or unoxidized EPA (EPA, 100 μmol/L) or oxidized EPA (OxEPA, 100 μmol/L) before stimulation with TNF for 2 hours. Nuclear extracts were preincubated with antibodies to p65 as indicated. The nuclear extracts were coincubated with 32P-labeled NF-B oligonucleotide, and electrophoresis mobility shift assays were performed (n=3).
Discussion
Omega-3 fatty acids in fish oil has been reported to improve the prognosis of several chronic inflammatory diseases characterized by leukocyte accumulation, including atherosclerosis, inflammatory bowel disease, rheumatoid arthritis, psoriasis, etc.1–4
Omega-3 fatty acids, such as EPA and DHA, are highly polyunsaturated and readily undergo oxidation at ambient and subambient temperatures, even in the absence of exogenous oxidizing reagents.5,6 In view of the ease with which omega-3 PUFA spontaneously oxidize and in vivo data suggesting extensive accumulation of oxidation products after fish oil consumption, we investigated the possibility that oxidized omega-3 fatty acids may be an important component of the observed antiinflammatory effects of fish oil. In our previous studies, we showed that oxidized EPA and not unoxidized EPA pretreatment of HUVEC inhibits leukocyte adhesion to cytokine-stimulated HUVEC, and this effect is mediated through a PPAR-dependent pathway.11,12 In our present studies, we extend these observations and show that oxidized EPA is also effective in inhibiting cytokine-induced endothelial chemokine expression, particularly MCP-1, and propose a mechanism for these antiinflammatory effects.
The fact that oxidation of the omega-3 fatty acids is required for the aforementioned antiinflammatory effects is also pointed out by other studies. Nohe et al and De Caterina et al have shown that prolonged incubation of endothelial cells with native omega-3 fatty acids (26 hours of total exposure, 6 hours before and 20 hours after TNF stimulation) results in inhibition of cytokine and adhesion molecule expression, whereas shorter incubation periods (6 hours) has no effect.30–32 This suggests that oxidation of omega-3 fatty acids takes place during the long incubation period, and that the oxidation product(s) and not the native omega-3 fatty acids are most likely responsible for the inhibition of leukocyte–endothelial interactions.
To ascertain that the inhibition of chemokine expression was not caused by the cytotoxic effects of oxidized EPA, we performed MTT assays, which showed that oxidized EPA in the doses used for our experiments had no effect on the viability of endothelial cells (data not shown). Also, oxidized EPA did not have any effect on the constitutively expressed surface proteins such as von Willebrand factor, endoglin, and human leukocyte antigen class I molecules.11,12
The expression of MCP-1 and IL-8 is regulated through activation of NF-B. NF-B activation appears to be necessary for the induction of chemokine genes, and deletion of NF-B binding sites results in an inability to induce these genes.26 In these studies, we show that oxidized EPA and not unoxidized EPA inhibits cytokine-induced activation of NF-B and promotes cytosolic retention of the p50 and p65 subunits. Hence, the oxidized EPA-mediated inhibition of MCP-1 and IL-8 in cytokine-stimulated endothelial cells could be explained by an inhibitory effect of oxidized EPA on NF-B activity.
The difference in the extent of inhibition of MCP-1 and IL-8 expression by oxidized EPA is most likely caused by the differential effects of oxidized EPA on endothelial NF-B and AP-1 activation. Oxidized EPA almost completely inhibits endothelial NF-B activation, whereas it had minimal effects in preventing AP-1 activation. Because AP-1 activation alone can result in IL-8 expression,28 oxidized EPA had only a mild inhibitory effect on IL-8 expression.
What is the mechanism for the oxidized EPA-mediated inhibition of cytokine-induced NF-B activation? We first hypothesized that oxidized EPA inhibits phosphorylation of IB by inhibiting the IKK (kinase) complex. However, it is unlikely that oxidized EPA inhibits IKK (kinase) activity because phosphorylation and ubiquination of IkB is noted in oxidized EPA-treated endothelial cells. In fact, when normalized to actin controls, oxidized EPA pretreatment before TNF stimulation for 60 minutes resulted in 30% more p-IkB when compared with TNF-treated cells. Also, it unlikely that oxidized EPA induces IkB expression because after 15 minutes of cytokine stimulation, no IkB is noted in the cytoplasm of cells pretreated with oxidized EPA. After 60 minutes, IB starts to appear in the cytoplasm of oxidized EPA pretreated cells and its concentration is, in fact, somewhat less than the TNF-treated cells. Thus, it is likely that oxidized EPA does not prevent NF-B activation by increasing the expression of IB.
In our previous studies we noted that oxidized EPA is a potent activator of PPAR and that PPAR is needed for the inhibitory effects of oxidized EPA on leukocyte–endothelial interactions,12 which lead us to hypothesize that oxidized EPA might inhibit cytokine-induced NF-B activation through a PPAR-dependent pathway. Here, we show that although oxidized EPA inhibits cytokine-induced NF-B activation in wild-type cells, it has no inhibitory effects in PPAR-deficient endothelial cells, suggesting that oxidized EPA mediates its inhibitory effects on NF-B through PPAR. The PPAR-mediated inhibitory effects of oxidized EPA on NF-B activation are possibly through direct interactions of PPAR with the p50/p65 subunits. Delerive et al (1999) have shown, using glutathione S-transferase pull-down experiments, that after fibrate (PPAR agonist) treatment, PPAR physically interacts with p65 in vitro.33
Taken together with our previous studies,11,12 we show that auto-oxidation of omega-3 fatty acids results in the generation of oxidized compounds with potent antiinflammatory properties that inhibit proinflammatory responses such as leukocyte adhesion receptor and chemokine expression. The central theme for the antiinflammatory effects of oxidized omega-3 fatty acids is likely through inhibition of NF-B via a PPAR-dependent pathway. The oxidation of omega-3 fatty acids is likely to occur in areas of active inflammation caused by increased expression of oxidative enzymes (eg, NADPH oxidase, myeloperoxidase, cyclooxygenase, lipoxygenase) and the generation of reactive oxygen species in these areas, which are capable of oxidizing PUFAs. The identification of these products could result in potent, low-toxicity, proinflammatory response inhibitors with potent PPAR agonist and anti-NF-B properties for the treatment of inflammatory diseases.
Acknowledgments
We thank Dr. Lynn Stoll for assistance with ELISA assays and Joel Carl for assistance with the figures. We acknowledge support (Scientist Development Grant) from the American Heart Association (S.S.)
References
Kromhout D, Bosschieter EB, de Lezenne Coulander C. The inverse relation between fish consumption and 20-year mortality from coronary heart disease. N Engl J Med. 1985; 312: 1205–1209.
Belluzzi A, Brignola C, Campieri M, Pera A, Boschi S, Miglioli M. Effect of and enteric coated fish oil preparation on relapses in Crohn’s disease. N Engl J Med. 1996; 334: 1557–1560.
Donadio JV Jr., Bergstralh EJ, Offord KP, Spencer DC, Holley KE. A controlled trial of fish oil in IgA nephropathy. N Engl J Med. 1994; 331: 1194–1199.
Volker D, Fitzgerald P, Major G, Garg M. Efficacy of fish oil concentrate in the treatment of rheumatoid arthritis. J Rheumatol. 2000; 27: 2343–2346.
Chan HWS. The mechanism of autoxidation. Autoxidation of unsaturated lipids. Academic Press: UK; 1987.
Porter NA, Caldwell SE, Mills KA. Mechanisms of free radical oxidation of unsaturated lipids. Lipids. 1995; 30: 277–290.
Gonzalez MJ, Gray JI, Schemmel RA, Dugan L Jr, Welsch CW. Lipid peroxidation products are elevated in fish oil diets even in the presence of added antioxidants. J Nutr. 1992; 122: 2190–2195.
Wander RC, Du S-H. Oxidation of plasma proteins is not increased after supplementation with eicosapentaenoic and docosahexaenoic acids. Am J Clin Nutr. 2000; 72: 731–737.
Meydani M, Natiello F, Goldin B, Free N, Woods M, Schaefer E, Blumberg JB, Gorbach SL. Effect of long-term fish oil supplementation on vitamin E status and lipid peroxidation in women. J Nutr. 1991; 121: 484–91.
Higdon JV, Liu J, Du SH, Morrow JD, Ames BN, Wander RC. Supplementation of postmenopausal women with fish oil rich in eicosapentaenoic acid and docosahexaenoic acid is not associated with greater in vivo lipid peroxidation compared with oils rich in oleate and linoleate as assessed by plasma malondialdehyde and F2-isoprostanes Am J Clin Nutr. 2000; 72: 714–722.
Sethi S, Eastman AY, Eaton JW. Inhibition of phagocyte-endothelium interactions by oxidized fatty acids: a natural anti-inflammatory mechanism? J Lab Clin Med. 1996; 128: 27–38.
Sethi S, Ziouzenkova O, Ni H, Wagner DD, Plutzky J, Mayadas TN. Oxidized omega-3 fatty acids in fish oil inhibit leukocyte-endothelial interactions through activation of PPAR Blood. 2002; 100: 1340–1346.
Gerszten RE, Garcia-Zepeda EA, Lim YC, Yoshida M, Ding HA, Gimbrone MA Jr, Luster AD, Luscinskas FW, Rosenzweig A. MCP-1 and IL-8 trigger firm adhesion of monocytes to vascular endothelium under flow conditions. Nature. 1999; 398: 718–723.
Maus U, Henning S, Wenschuh H, Mayer K, Seeger W, Lohmeyer J. Role of endothelial MCP-1 in monocyte adhesion to inflamed human endothelium under physiological flow Am J Physiol. 2002; 283: H2584–H2591.
Shimizu H, Maruyama S, Yuzawa Y, Kato T, Miki Y, Suzuki S, Sato W, Morita Y, Maruyama H, Egashira K, Matsuo S. Anti-monocyte chemoattractant protein-1 gene therapy attenuates renal injury induced by protein-overload proteinuria. J Am Soc Nephrol. 2003; 14: 1496–505.
Egashira K. Molecular mechanisms mediating inflammation in vascular disease: special reference to monocyte chemoattractant protein-1. Hypertension. 2003; 41: 834–41.
Dawson J, Miltz W, Mir AK, Wiessner C. Targeting monocyte chemoattractant protein-1 signalling in disease. Exper Opin Ther Targets. 2003; 7: 35–48.
Halliwell B, Gutteridge JM. Free radicals in biology and medicine, 3rd ed. New York, NY: Oxford University Press; 1993.
Denning GM, Wollenweber LA, Railsback MA, Cox CD, Stoll LL, Britigan BE. Pseudomonas Pyocyanin increases interleukin-8 expression by human airway epithelial cells. Infect Immun. 1998; 66: 5777–5784.
Rice JB, Stoll LL, Li W-G, Denning GM, Weydert J, Charipar E, Richenbacher WE, Miller FJ Jr, Weintraub NL. Low-Level Endotoxin Induces Potent Inflammatory Activation of Human Blood Vessels: Inhibition by Statins. Arterioscler Thromb Vasc Biol. 2003; 23: 1576–1582.
Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem. 1987; 162: 156–9.
Huang H, McIntosh J, Hoyt DG. An efficient, nonenzymatic method for isolation and culture of murine aortic endothelial cells and their response to inflammatory stimuli. In Vitro Cell Dev Biol Anim. 2003; 39: 43–50.
Dignam JD, Lebovitz RM, Roeder RG. Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei. Nucleic Acids Res. 1983; 11: 1475–1489.
Canty TG Jr., Boyle EM Jr, Farr A, Morgan EN, Verrier ED, Pohlman TH. Oxidative stress induces NF-kappaB nuclear translocation without degradation of IkappaB. Circulation. 1999; 100: 361–364.
Zandi E, Chen Y, Karin M. Direct phosphorylation of IB by IKK and IKK?: discrimination between free and NF-B-bound substrate. Science. 1998; 281: 1360–1363.
Martin T, Cardarelli PM, Parry GC, Felts KA, Cobb RR. Cytokine induction of monocyte chemoattractant protein-1 gene expression in human endothelial cells depends on the cooperative action of NF-kappa B and AP-1. Eur J Immunol. 1997; 27: 1091–1097.
Roebuck KA. Regulation of Interleukin-8 gene expression. J Interferon Cytokine Res. 1999; 19: 429–438.
Hipp MS, Urbich C, Mayer P, Wischhusen J, Weller M, Kracht M, Spyridopoulos I. Proteosome inhibition leads to NF-kB-independent IL-8 transactivation in human endothelial cells through induction of AP-1. Eur J Immunol. 2002; 32: 2208–2217.
Lenard MJ, Baltimore D. NF-kB: a pleiotropic mediator of inducible and tissue specific gene control. Cell. 1988; 58: 227–229.
Nohe B, Johannes T, Dieterich HJ, Mayer K, Merfels M, Grimminger F, Seeger W. Antiinflammatory effects of omega-3 fatty acids vary at different stages of inflammation. Am J Physiol. 2003; 285: H2248–H2250.
De Caterina R, Bernini W, Carluccio MA, Liao JK, Libby P. Structural requirements for inhibition of cytokine-induced endothelial activation by unsaturated fatty acids. J Lipid Res. 1998; 39: 1062–1070.
De Caterina R CM, Clinton SK, Gimbrone MA Jr, Libby P. The omega-3 fatty acid docosahexaenoate reduces cytokine-induced expression of proatherogenic and proinflammatory proteins in human endothelial cells. Arterioscler Thromb Vasc Biol. 1994; 14: 1829–1836.
Delerive P, De Bosscher K, Besnard S, Vanden Berghe W, Peters JM, Gonzalez FJ, Fruchart J-C, Tedgui A, Haegeman G, Staels B. Peroxisome proliferator-activated receptor alpha negatively regulates the vascular inflammatory gene response by negative cross-talk with transcription factors NF-kappa B and AP-1. J Biol Chem. 1999; 274: 32048–32054.(Archana Mishra; Ashok Cha)