A Novel β-Oxa Polyunsaturated Fatty Acid Downregulates the Activation of the IκB Kinase/Nuclear Factor κB Pathway, Inhibits Expression of Endothelial Cell Adhesion Molecules, and Depresses Inflammation
Several novel polyunsaturated fatty acids (PUFAs) that contain either an oxygen or sulfur atom in the β-position were found to exhibit more selective antiinflammatory properties than their natural PUFA counterparts. One of these, β-oxa-23:4n-6, unlike natural PUFAs, lacked ability to stimulate oxygen radical production in neutrophils but caused marked inhibition of agonist-induced upregulation of leukocyte adhesion to cultured human umbilical vein endothelial cells (HUVEC) and E-selectin, intercellular adhesion molecule-1, and vascular cell adhesion molecule-1 expression. In addition, β-oxa-23:4n-6 inhibited acute and chronic inflammatory responses in mice as well as the upregulation of adhesion molecule expression in arterial endothelium. This action of β-oxa-23:4n-6 required a functional 12- but not 5-lipoxygenase or cyclooxygenases, consistent with its metabolism via the 12-lipoxygenase pathway. Whereas β-oxa-23:4n-6 did not affect the activation of mitogen-activated protein kinases by tumor necrosis factor, activation of the IκB kinase/nuclear factor κB pathway was selectively inhibited. These novel PUFAs could form the basis for a potential new class of pharmaceuticals for treating inflammatory diseases, including atherosclerosis.
Arachidonic acid (20:4n-6), either as a free fatty acid or through its metabolites, stimulates responses in a variety of cell types.1–7 For example, in neutrophils, 20:4n-6 stimulates the respiratory burst, degranulation, adhesion, and expression of CD11b/CD18.8 These effects of 20:4n-6 can be amplified by cytokines such as tumor necrosis factor (TNF).9 It has been established that 20:4n-6 also stimulates the activities of intracellular signaling molecules such as protein kinase C and the mitogen-activated protein kinases, extracellular signal-regulated kinase (ERK) 1/2, c-Jun N-terminal kinase (JNK), and p38,8,10,11 and modulates the activity of the rho family of small G proteins.12 Metabolites of 20:4n-6 such as leukotriene B4 (LTB4)1,8 and 5-oxo-6,8,11,14(E,Z,Z,Z)-eicosatetraenoic acid2 are also well-defined stimulators of cellular function. Thus it is not surprising that strategies have been developed to either target fatty acids and their metabolites or use these to treat a range of acute and chronic inflammatory diseases. For example, the dietary n-3 polyunsaturated fatty acids (PUFAs) eicosapentaenoic acid (20:5n-3) and docosahexaenoic acid (22:6n-3) have been reported to be effective in treating allergic and autoimmune inflammatory diseases.13 This is attributable to the ability of n-3 PUFAs to compete against 20:4n-6 for metabolism by the lipoxygenases and cyclooxygenases and their products being markedly less proinflammatory than those derived from 20:4n-6. There is however, a lack of confidence in the use of n-3 PUFAs in treating chronic inflammatory diseases. Whereas some issues can be resolved through more research into the actions of n-3 PUFAs, others cannot. For example, n-3 PUFAs are able to stimulate oxygen radical production by neutrophils and monocytes.8
Accordingly, we have synthesized14,15 long-chain PUFAs with either an oxygen atom (β-oxa derivatives; β-oxa-23:4n-6, β-oxa-21:3n-6, β-oxa-21:3n-3) or a sulfur atom in the β position (β-thia derivatives; β-thia-23:4n-6, β-thia-21:3n-6, β-thia-21:3n-3). The objective of this synthesis was to make compounds that are PUFA based, with many of the properties of PUFAs, such as absorption following oral administration and esterification into membrane phospholipids, but with more selective biological activities, skewed toward antiinflammatory effects.
Materials and Methods
Natural fatty acids were from Sigma Chemical Co (St Louis, Mo). The β-oxa and β-thia compounds were synthesized as described.14,15 β-oxa-23:4n-6 methyl ester was generated by treating β-oxa-23:4n-6 with diazomethane in diethyl ether; β-oxa-23:0 was prepared by hydrogenation of β-oxa-23:4n-6 in the presence of platinum oxide16 and 15- and 18-monohydroperoxy-β-oxa-23:4n-6 by incubating β-oxa-23:4n-6 with porcine leukocyte 12-lipoxygenase and soybean 15-lipoxygenase (Cayman Chemical, Ann Arbor, Mich), respectively. The 15- and 18-monohydroxy-β-oxa-23:4n-6 were obtained by reduction with sodium borohydride and purified by silicic acid column chromatography.16 Purity of the products was assessed by thin-layer chromatography (diethyl ether/hexane/acetic acid=60:40:1). The lipid zones were visualized under UV light with dichlorofluorescein (0.2%) in ethanol and identified by comparison of retention factors with those of similarly structured analogs. No other monohydroxylated materials were evident.
Fatty acids and derivatives (at least 98% purity) were dissolved in ethanol (0.1% final, vol/vol) (in vitro), dipalmitoylphosphatidylcholine (DPC),4 or DSMO (7%) (in vivo), which did not affect cellular functions at these concentrations.
Lipofectin, Opti-modified Eagle’s medium (OPTI-MEM), and oligonucleotides corresponding to the consensus nuclear factor κB (NF-κB) element, and T4 polynucleotide kinase were obtained from Invitrogen (Mt Waverley, Vic, Australia) and New England Biolabs (Ipswich, MA), respectively.
Neutrophil and Monocyte Adhesion to Human Umbilical Vein Endothelial Cells
Preparation of human umbilical vein endothelial cells (HUVEC), neutrophils, and monocytes and adhesion of leukocytes to HUVEC were performed as described.16
Measurement of Endothelial Cell Adhesion Molecules In Vitro and In Vivo
HUVEC were treated with a stimulus for up to 24 hours. Expression of E-selectin, intercellular adhesion molecule (ICAM)-1, and vascular cell adhesion molecule (VCAM)-1 was determined by ELISA or as mRNA using a slot blot technique.16
To investigate the effect of β-oxa-23:4n-6 on lipopolysaccharide (LPS)-induced expression of E-selectin in vivo, BALB/c mice (Gillies Plains Animal Resource Center, Gillies Plains, SA, Australia) were treated intravenously with the fatty acid or vehicle, and 2 hours later injected intraperitoneally with LPS (50 μg). After 5 hours, the aortas encompassing the ascending aorta through to the bifurcation at the common iliac arteries were isolated, each cut into 2 pieces of equal length, minced, and fixed in 0.25% glutaraldehyde. For each aorta, the half comprising the aortic arch was incubated with a monoclonal antibody to mouse E-selectin, whereas the other half was incubated with isotype matched control (BD Biosciences). This was followed by incubation with an horseradish peroxidase-conjugated secondary antibody and then ABTS (ELISA). All procedures involving animals, including maintenance, were approved by the Women’s and Children’s Hospital Animal Ethics Committee.
Analysis of Lipoxygenase Products
Lipids were extracted (3× 100 mL of diethyl ether) from the HUVEC culture medium (acidified with 6 mL of 1 mol/L citric acid/100 mL medium), and oxygenated fatty acid derivatives were characterized by electrospray mass spectrometry. Briefly, electrospray ionization tandem mass spectra (ESI-MS-MS) were recorded on a PE SCIEX API3000 spectrometer, operating in a negative ion mode at an ion spray voltage of 4.2 kV, capillary temperature of 100°C. Declustering potential (DP), focusing potential (FP), entrance potential (EP), collision energy (CE), and collision cell exit potential (CXP) were maintained at −61, −300, −10, −30, and −15 V, respectively.
Preparation of Cell Lysates and Nuclear Fractions
Transfection and Luciferase Assays
HUVEC were washed (OPTI-MEM) and transiently cotransfected with an NF-κB-luc vector (1.6 μg/well)18 and pRL-null vector (50 ng) using Lipofectin. After 5 hours, RPMI-1640 plus human group AB serum (20%, vol/vol) was added and the cells incubated for 48 hours. The cells were pretreated with PUFAs for 1 hour and stimulated with TNF for 24 hour. After lysis, luciferase activity was assayed using a dual luciferase assay kit (Promega Corp, Annandale, Australia).
Proteins (50 μg) were fractionated by SDS-PAGE (12% gel), transferred to nitrocellulose, and probed with an anti-IκBα (lysate samples) or anti-NF-κB p65 (nuclear fractions) antibody, respectively (Santa Cruz Biotechnology, Santa Cruz, Calif). Immunocomplexes were detected by enhanced chemiluminescence.11
IκB Kinase Assay
IκB kinase (IKKβ) activity was assayed after immunoprecipitation with anti-IKKβ (M-280) antibody (Santa Cruz Biotechnology) using GST-IκBα (residues 5 to 55) as a substrate.19
p38, ERK, and JNK Assays
ERK and p38 were immunoprecipitated with anti-ERK2 and anti-p38 antibody, respectively (Santa Cruz Biotechnology) and the activity determined as described.10,11 JNK activity was determined in a solid-phase assay using GST c-Jun (1–79) as a substrate.10,11
Electrophoretic Mobility Shift Assays
The double-stranded NF-κB probe (5′-AGTTGAGGGGACTTT- CCCAGGC-3′) was labeled using T4 polynucleotide kinase according the instructions of the manufacturer. All subsequent steps were as described,20 with the exception that the gels were analyzed using an Instant Imager (Packard Instrument).
The effects of β-oxa-23:4n-6 on in vivo inflammatory responses was measured as a delayed type hypersensitivity (DTH) reaction and LPS-induced influx of neutrophils and mononuclear cells in the peritoneal cavity in BALB/c mice. For DTH experiments, mice were injected subcutaneously with 100 μL of 10% hematocrit sheep red blood cells (SRBC) and challenged with the antigen (25 μL of 40%) in the hindfoot pad 6 days later, and the degree of foot pad swelling was measured 48 hours later. One hour before challenge, mice were given 10 mg/kg body weight of the PUFAs in 7% DMSO or the same amount of vehicle, intraperitoneally. For peritoneal inflammation, mice were given β-oxa-23:4n-6 (40 mg/kg) intravenously and, 6 hours later, were injected with LPS (50 μg) intraperitoneally. At 24 and 72 hours, the animals were euthanized and cellular infiltrates examined in Giemsa-stained smears.
Effects on Neutrophil Respiratory Burst
Compared with 22:6n-3, β-oxa-23:4n-6 had little or no effect on neutrophil oxygen radical production (chemiluminescence response) (Figure 1).
Effects on Neutrophil and Monocyte Adherence to Activated HUVEC
HUVEC, pretreated with the β-oxa PUFAs, showed reduced ability to be stimulated by TNF-α for enhanced neutrophil adhesion (Figure 2). In contrast, the naturally occurring 20:4n-6, octadecadienoic acid (linoleic acid, 18:2n-6), and 22:6n-3 did not affect this response. Similar results were seen irrespective of whether we presented the fatty acids in ethanol (0.1% vol/vol final concentration) or as mixed fatty acid-DPC micelles. The cells remained viable under these experimental conditions as judged by trypan blue exclusion and lack of [51Cr] chromate release. Furthermore, the engineered PUFAs did not affect DNA synthesis, glucose metabolism, and G3PDH mRNA expression in HUVEC (data not presented). β-Oxa-23:4n-6 caused the greatest suppression of neutrophil adhesion to TNF-stimulated HUVEC (Figure 2) and was therefore examined further.
The magnitude of the suppressive effect of β-oxa-23:4n-6 was dependent on pretreatment time and concentration with significant effects observed with a pretreatment time of 1 hour and a concentration of ≥5 μmol/L (data not presented). β-Oxa-23:4n-6 also inhibited PMA- or LPS-stimulated increase in neutrophil adhesion to HUVEC (data not presented). Interestingly, β-oxa-23:4n-6 did not affect the ability of TNF to stimulate the adherence of neutrophils to plasma-coated surfaces (not shown), indicating that the effect was solely on HUVEC.
Similarly, β-oxa-23:4n-6 inhibited the ability of TNF to stimulate HUVEC to bind monocytes. The cells were preincubated with β-oxa-23:4n-6 for 1 hour, TNF was added and monocyte adherence was determined 16 hours later. Adherence of monocytes was also inhibited by 22:6n-3. Results (mean±SEM) from 3 experiments (percentage of vehicle control) demonstrate that β-oxa-23:4n-6 and 22:6n-3 (20 μmol/L) caused a similar degree of inhibition (control, 100±11; β-oxa-23:4n-6, 60±7; 22:6n-3, 57±13.6; P<0.05).
Effects of Derivatives of β-oxa-23:4n-6
Derivatization of β-oxa-23:4n-6 to methylated, saturated, and 18-monohydroxy and hydroperoxy forms abolished its inhibitory effect on neutrophil adhesion to TNF-stimulated HUVEC (Figure 3), demonstrating not only the specificity of the parent molecule but also that its structure is critical for activity.
Effects on TNF-Induced Expression of Adhesion Molecules
The inhibitory effect of β-oxa-23:4n-6 on leukocyte adhesion was consistent with its suppression of TNF-induced expression of E-selectin, ICAM-1, and VCAM-1 on HUVEC. Maximum effect was observed after 4, 6, and 12 hours of cytokine treatment for E-selectin, ICAM-1, and VCAM-1, respectively, after which there was recovery, especially for E-selectin and ICAM-1 (Figure 4). This confirms that β-oxa-23:4n-6 did not affect HUVEC viability. β-oxa-23:4n-6 inhibited the expression of the adhesion molecules in a concentration-dependent manner (data not presented), which corresponded with the levels required to reduce neutrophil adherence. In contrast, 20:4n-6 did not alter the expression of adhesion molecules (Figure 4). β-oxa-23:4n-6 also inhibited the ability of TNF, LPS, or PMA to upregulate the expression of E-selectin mRNA by approximately 50% compared with vehicle-treated cells (data not presented).
In Vivo Activity of β-oxa-23:4n-6
β-oxa-23:4n-6 also inhibited the DTH reaction. Mice sensitized with SRBC were inhibited in their ability to manifest this reaction if given an injection of β-oxa-23:4n-6 1 hour before antigen challenge (Figure 5A). This illustrates an effect on chronic inflammation probably through the inhibition of T-cell and monocyte binding to the endothelium. When an acute inflammatory reaction (24 hours) was induced by intraperitoneal injection of LPS, β-oxa-23:4n-6 inhibited neutrophil influx (Figure 5A). A similar inhibition of chronic inflammation occurred for the mononuclear cell infiltrate after 72 hours (Figure 5A), with no selectivity toward monocytes or T lymphocytes. β-oxa-23:4n-6 also inhibited E-selectin expression in the aortic endothelium of LPS-treated mice (Figure 5B).
Metabolism of β-oxa-23:4n-6
Incubation of HUVEC with β-oxa-23:4n-6 for 60 minutes resulted in the formation of oxygenated products. Analysis by electrospray tandem mass spectrometry indicated a molecular ion at m/z 363 (M-1) (expected for monohydroxylated analogs of β-oxa-23:4n-6). Fragmentation of this ion produced 2 daughter ions at m/z 262 and 222, corresponding to loss of a C6H13O and C9H17O fragment, resulting from C17–C18 and C14–C15 bond cleavage, respectively (Figure 6c). These fragments suggest the presence of a monohydroxyl group at carbon 18 and 15, respectively, because they match up with the fragments formed by incubating β-oxa-23:4n-6 with purified 12-lipoxygenase (Figure 6a) and 15-lipoxygenase (Figure 6b). The lipoxygenase positional isomer specificity is determined by the carbon chain length from the carboxyl end of PUFAs. Because β-oxa-23:4n-6 has 3 additional atoms in its backbone compared with 20:4n-6, we conclude that the 15- and 12-lipoxygenases produced the 18- and 15-monohydroxylated derivatives, respectively. A fragment at m/z 210 was also present. This could have been produced from either 14,15-epoxy β-oxa-23:4n-6, generated by cytochrome P450 dependent 11,12-epoxygenase, 14-hydroxy β-oxa-23:4n-6, generated by “11-lipoxygenase” or from a cyclooxygenase-generated 14-hydroxy metabolite. None of these fragments was detected in the absence of β-oxa-23:4n-6 (Figure 7C).
Role of the 12-LO
Pretreatment of HUVEC with nordihydroguaiaretic acid (NDGA), a lipoxygenase/epoxygenase inhibitor with some selectivity for 5-lipoxygenase, markedly suppressed the ability of β-oxa-23:4n-6 to inhibit TNF-stimulated expression of E-selectin (Figure 7a). This effect was also seen with baicalein, an inhibitor with selectivity for the 12-lipoxygenase. In contrast, MK886, an inhibitor of the 5-lipoxygenase activating protein; indomethacin, a cyclooxygenase inhibitor; vitamin E, an antioxidant; and diluent (control) were without any effects (Figure 7a). These data suggest that a product(s) of β-oxa-23:4n-6 formed by the 12-lipoxygenase is important for its biological activity. Interestingly, baicalein predominantly suppressed the formation of 15-monohydroxy β-oxa-23:4n-6 in HUVEC (comparing Figure 7b with 7c). Although baicalein also reduced the fragments at m/z 262 and 210, the effect on m/z 222 was greater by a factor of 2 to 3. In a separate set of experiments, we demonstrated that NDGA reduced the fragments at m/z 262, 222, and 210 by approximately 44%, 25%, and 70% (Figure 7f), respectively, compared with control (Figure 7e). Interestingly, when we pretreated a proportion of these cells with indomethacin, only the fragment at m/z 210 was reduced (Figure 7G). These data imply that the 12-lipoxygenase was involved in facilitating the inhibitory action of β-oxa-23:4n-6, whereas the 15-lipoxygenase was unlikely to be involved because the 18-monohydroxy/hydroperoxy derivatives of β-oxa-23:4n-6 were inactive (Figure 3). Our data exclude the involvement of the 5-lipoxygenase in the action of β-oxa-23:4n-6. A role for the product that formed the fragment at m/z 210 can also be excluded because indomethacin did not alter the response to β-oxa-23:4n-6. The lack of effect of indomethacin on the formation of m/z 222 fragment further supports a role for the 12-lipoxygenase. These results provide evidence that HUVEC converted β-oxa-23:4n-6 to monohydroxylated derivatives by enzymatic, rather than autooxidation, processes.
Effects on Signaling Molecules Activated by TNF
p38 and JNK MAP kinases and the NF-κB pathway are known to be involved in regulating TNF-stimulated adhesion molecule expression in HUVEC.21–23 Although pretreatment of HUVEC with β-oxa-23:4n-6 did not affect the ability of TNF to stimulate p38, ERK, and JNK (data not presented), the β-oxa PUFAs caused a selective inhibition of the NF-κB pathway (Figure 8). Pretreatment of HUVEC with β-oxa-23:4n-6 caused a marked inhibition of TNF-stimulated degradation of IκBα (>92%), a marker of activation of the NF-κB pathway (Figure 8a). TNF-stimulated IκB degradation was also inhibited by 22:6n-3 (&50% inhibition) (Figure 8a). Consistent with this, β-oxa-23:4n-6 inhibited TNF-stimulated nuclear translocation of NF-κB (p65), as seen by Western blotting or electrophoretic mobility shift assay (EMSA) (data not shown). To confirm that transcription via the NF-κB site was blocked, we transiently cotransfected HUVEC with an NF-κB promoter firefly luciferase construct and a vector carrying the renilla luciferase gene under the control of the thymidine kinase promoter. The data in Figure 8b demonstrate that β-oxa-23:4n-6 and 22:6n-3 nearly totally blocked the expression of firefly luciferase. We also examined whether β-oxa-23:4n-6 and 22:6n-3 inhibited the ability of TNF to stimulate the activity of IκB kinase (IKK) as this has not been demonstrated. Both PUFAs caused a near total inhibition of the activation of IKKβ by TNF (Figure 8c). These data demonstrate that β-oxa-23:4n-6 and 22:6n-3 can inhibit the IKK/NF-κB pathway but not MAP kinases in HUVEC. The lesser effect of 22:6n-3 on IκBα degradation was likely to have been due a nonoptimized time being tested.
The data demonstrate that by placing an oxygen or sulfur atom in the β position of a PUFA, molecules can be generated that differ in their biological activities from the natural PUFAs. Not only did the engineered PUFAs have greatly reduced ability to stimulate the neutrophil respiratory burst, but they also inhibited neutrophil adherence to activated HUVEC, whereas 20:4n-6 and 22:6n-3 did not. This is surprising because 22:6n-3 inhibits monocyte adherence to activated endothelial cells21,24 (and our data). The most active novel PUFA was β-oxa-23:4n-6, which was more active than β-thia 23:4n-6. This illustrates that PUFAs bearing the same structural elements can vary dramatically in activity depending on whether they contain an oxygen or sulfur atom in the β position. The biological properties of the novel PUFAs, in particular β-oxa-23:4n-6, are similar to those of the 15-lipoxygenase product of 20:4n-6, 15-hydroperoxyeicosatetraenoic acid (15-HPETE). For example, 15-HPETE is very weak at stimulating neutrophil superoxide production,8 but it strongly inhibits TNF production by monocytes6 and the upregulation of adhesion molecule expression in HUVEC.16 Our data demonstrate that it is possible to synthesize PUFA-based molecules with more selective biological activities, possibly skewed toward antiinflammatory properties, akin to 15-HPETE.
The in vitro effects of β-oxa-23:4n-6 is clearly reflected in vivo because the PUFAs not only inhibited the influx of neutrophils into the peritoneal cavity in LPS-treated mice but also the subsequent mononuclear infiltrate. β-Oxa-23:4n-6 also inhibited the DTH response. We have previously found 22:6n-3 to be very weak at inhibiting this response,25 suggesting that β-oxa-23:4n-6 would make a better antiinflammatory agent than 22:6n-3. These in vivo effects of β-oxa-23:4n-6 are most likely to be attributable to its ability to inhibit ligand-induced upregulation of E-selectin, ICAM-1, and VCAM-1 expression, as these adhesion molecules are essential for the recruitment of leukocytes from the blood.26 Indeed, β-oxa-23:4n-6 inhibited LPS-induced upregulation of E-selectin in the aorta, consistent with its effect on LPS-induced upregulation of E-selectin on HUVEC. Whereas the suppressive effect of β-oxa-23:4n-6 on cytokine-stimulated upregulation of E-selectin and ICAM-1 expression was transient, its effect on VCAM-1 expression was sustained, lasting for more than 24 hours. This is likely to account for the ability of β-oxa-23:4n-6 to also inhibit mononuclear infiltrate when this was assessed at 72 hours. Administration of β-oxa-23:4n-6 to mice by gavage once daily (100 mg/kg body weight) for 3 days resulted in the esterification of the PUFAs in lipids of various tissues, including liver, kidneys, lungs, heart, spleen, brain, adipose tissue, and blood (data not presented). These mice maintained normal body weight and retained normal liver and kidney function.
Monohydroxylated products of β-oxa-23:4n-6 were formed by HUVEC. Three of these were provisionally identified as the 14-monohydroxy-β-oxa-23:4n-6, 15-monohydroxy-β-oxa-23:4n-6, and 18-monohydroxy-β-oxa-23:4n-7. Analysis by liquid chromatography coupled with tandem mass spectrometry would have provided additional evidence for the formation of these products, but this was not possible because of the low levels of the products being formed. A role of the 12-lipoxygenase in the action of β-oxa-23:4n-6 is supported by the finding that baicalein, a 12-lipoxygenase inhibitor, not only reduced the inhibitory effect of β-oxa-23:4n-6 on TNF-stimulated E-selectin expression but relatively less of the fragment at m/z 222 was formed in the presence of the inhibitor, compared with the fragments at m/z 210 and 262. In contrast, the 5-lipoxygenase and cyclooxygenases were not required. Because only small amounts of the 12-lipoxygenase-derived compound were produced by HUVEC, the compound responsible could be very potent.
Interestingly, NDGA and indomethacin, specific for the lipoxygenases/epoxygenases and cyclooxygenases, respectively,27,28 reduced the fragment at m/z 210. This suggests that an 11-lipoxygenase, a cyclooxygenase, or the cytochrome P450–dependent 11,12-epoxygenase was involved in generating the 14-hydroxylated metabolite. Although an NDGA-sensitive 11-lipoxygenase has been reported in mollusk and hydra,29,30 its presence in mammalian cells has not been reported. Because indomethacin was more effective at suppressing the fragment at m/z 210 than NDGA, we favor the view that a cyclooxygenase was responsible for generating 14-hydroxy-β-oxa-23:4n-6.
Although β-oxa-23:4n-6 did not inhibit TNF-stimulated activation of the MAP kinases, it blocked the NF-κB pathway. The NF-κB pathway was also inhibited by 22:6n-3. β-oxa-23:4n-6 and 22:6n-3 were equally effective at inhibiting IKKβ activity and NF-κB-mediated transcription. Interestingly, whereas β-oxa-23:4n-6 inhibited the adherence of neutrophils to activated HUVEC, 22:6n-3 did not, although both PUFAs inhibited the adherence of monocytes to HUVEC by similar degrees. Thus, adherence of neutrophils to HUVEC is not strictly dependent on the state of activation of the NF-κB pathway. Consistent with this, expression of ICAM-1 that mediates neutrophil adherence is least dependent on NF-κB. The PUFA-insensitive component of monocytes adherence to HUVEC, despite the near complete inhibition of the NF-κB pathway, suggests the involvement of other signaling molecules such as p38 and JNK that are insensitive to inhibition by β-oxa-23:4n-6.
How β-oxa-23:4n-6 inhibited IKKβ is not known. Fatty acids, including n-3 PUFAs, and some of their hydroxylated metabolites, are known to stimulate the peroxisomal proliferator-activated receptor α (PPARα).31 Recently, oxidized 20:5n-3 has been reported to activate PPARα32 and inhibit NF-κB activation in a PPARα-dependent manner.33 However, oxidized 20:5n-3 and β-oxa-23:4n-6 differed in their effects on NF-κB activation. Whereas β-oxa-23:4n-6 inhibited TNF-stimulated activation of IKKβ and IκBα degradation, oxidized 20:5n-3 did not inhibit cytokine-stimulated phosphorylation and ubiquitination of IκBα.33 It is therefore unlikely that β-oxa-23:4n-6 and hydroxylated derivatives acted via PPARα.
Although natural n-3 PUFAs and their metabolic products have received widespread interest, it is well appreciated that their broad spectrum of activities limits their usefulness, either as biological tools or inhibitors of inflammation.5,8 The selective action of β-oxa PUFAs on different intracellular signaling molecules permits the selective targeting of particular cell types such that endothelial cells are more sensitive to β-oxa-23:4n-6 than β-oxa-21:3n-3 and vice versa for T cells.25 By making simple structural changes, molecules with desired specificity and biological stability can be obtained to target processes responsible for pathophysiology, including those in atherogenesis, with a goal for developing novel pharmaceuticals.
Sources of Funding
This study received financial support from the National Heart Foundation of Australia, The Women’s and Children’s Hospital Research Foundation, and the UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases.
Original received October 5, 2005; resubmission received March 31, 2006; revised resubmission received May 17, 2006; accepted May 26, 2006.
Samuelsson B. From studies of biochemical mechanisms to novel biological mediators: prostaglandin, endoperoxides, thromboxanes and leukotrienes. In: Les Prix Nobel, The Nobel Prizes, Nobel Prizes: Presentations, Biographies and Lectures. Stockholm, Sweden: Almquist & Wiksell; 1982: 153–174.
Graber R, Sumida C, Nunez EA. Fatty acids and cell signal transduction. J Lipid Mediat Cell Signal. 1994; 2: 91–116.
Ferrante JV, Huang ZH, Nandoskar M, Hii CST, Robinson BS, Rathjen DA, Poulos A, Morris CP, Ferrante A. Altered responses of human macrophages to lipopolysaccharide by hydroperoxy eicosatetraenoic acid, hydroxy eicosatetraenoic acid and arachidonic acid: inhibition of TNF production. J Clin Invest. 1997; 99: 1445–1452.
Ferrante A, Goh DHB, Harvey DP, Robinson BS, Hii CST, Bates EJ, Hardy SJ, Johnson DW, Poulos A. Neutrophil migration inhibitory properties of polyunsaturated fatty acids: the role of fatty acid structure, metabolism and possible second messenger systems. J Clin Invest. 1994; 93: 1063–1070.
Ferrante A, Hii CST, Costabile M. Regulation of neutrophil functions by long chain fatty acids. In: Dimitry Gabrilovich, ed. The Neutrophils: New Outlook for Old Cells. 2nd ed. London, UK: Imperial College Press; 2005: 201–228.
Hii CST, Ferrante A, Edwards YS, Huang ZH, Hartfield PJ, Rathjen DA, Poulos A, Murray AW. Activation of mitogen-activated protein kinase by arachidonic acid in rat liver epithelial WB cells by a protein kinase C-dependent mechanism. J Biol Chem. 1995; 270: 4201–4204.
Hii CST, Huang ZH, Bilney A, Costabile M, Murray AW, Rathjen DA, Der CJ, Ferrante A. Stimulation of p38 phosphorylation and activity by arachidonic acid in HeLa cells, HL60 promyelocytic leukemic cells, and human neutrophils. J Biol Chem. 1998; 273: 19277–19282.
Chuang TH, Bohl BP, Bokoch GM. Biologically active lipids are regulators of Rac. GDI complexation. J Biol Chem. 1993; 15:268: 26206–26211.
Pitt MJ, Easton CJ, Moody CJ, Ferrante A, Poulos A, Rathjen DA. Synthesis of polyunsaturated β-oxa fatty acid via rhodium mediated carbenoid insertion. Synthesis. 1997; 1240–1242.
Huang ZH, Bates EJ, Ferrante JV, Hii CS, Poulos A, Robinson BS, Ferrante A. Inhibition of stimuli-induced endothelial cell intercellular adhesion molecule -1, E-selectin and vascular cellular adhesion molecule-1 expression by arachidonic acid and its hydroxy- and hydroperoxy-derivatives Circ Res. 1997; 80: 149–158.
Andrews NC, Faller DV. A rapid micropreparation technique for extraction of DNA-binding proteins from limiting numbers of mammalian cells. Nucl Acids Res. 1992; 19: 2499.
Lee FS, Peters RT, Dang LC, Maniatis T. MEKKI activates both IκB kinase α and IκB kinase β. Proc Natl Acad Sci U S A. 1998; 95: 9319–9324.
Nutchey BK, Kaplan JS, Dwivedi PP, Omdhal JL, Ferrante A, May BK, Hii CS. Molecular action of 1,25-dihydroxyvitamin D3 and phorbol ester on the activation of the rat cytochrome P450C24 (CYP24) promoter: role of MAP kinase activities and identification of an important transcription factor binding site. Biochem J. 2005; 389: 753–762.
De Caterina R, Cybulsky MI, 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. 1994; 14: 1929–1936.
Read MA, Whitley MZ, Gupta S, Pierce JW, Best J, Davis RJ, Collins T. Tumor necrosis factor alpha-induced E-selectin expression is activated by the nuclear factor-kappaB and c-JUN N-terminal kinase/p38 mitogen-activated protein kinase pathways. J Biol Chem. 1997; 272: 2753–2761.
Min W, Pober JS. TNF initiates E-selectin transcription in human endothelial cells through parallel TRAF-NF-kappa B and TRAF-RAC/CDC42-JNK-c-Jun/ATF2 pathways. J Immunol. 1997; 159: 3508–3518.
Weber C, Erl E, Pietsch A, Danesch L, Weber P. Docosahexaenoic acid selectively attenuates induction of vascular cell adhesion molecule-1 and subsequent monocytic cell adhesion to human endothelial cells stimulated by tumor necrosis fact-α. Arterioscler Thromb Vasc Biol. 1995; 15: 622–628.
Costabile M, Hii CS, Robinson BS, Rathjen DA, Pitt M, Easton C, Miller RC, Poulos A, Murray AW, Ferrante A. A novel long chain polyunsaturated fatty acid, beta-Oxa 21:3n-3, inhibits T lymphocyte proliferation, cytokine production, delayed-type hypersensitivity, and carrageenan-induced paw reaction and selectively targets intracellular signals. J Immunol. 2001; 167: 3980–3987.
Hong SH, Avis I, Vos MD, Martinez A, Treston AM, Mulshine JL. Relationship of arachidonic acid metabolizing enzyme expression in epithelial cancer cell lines to the growth effect of selective biochemical inhibitors. Cancer Res. 1999; 59: 2223–2228.
Di Margo V, Gianfrani C, De Petrocellis L, Milone A, Cimino G. Polyunsaturated-fatty-acid oxidation in Hydra: regioselectivity, substrate-dependent enantioselectivity and possible biological role. Biochem J. 1994; 300: 501–507.
Marx N, Duez H, Fruchart JC, Staels B. Peroxisome proliferator-activated receptors and atherogenesis: regulators of gene expression in vascular cells. Circ Res. 2004; 94: 1168–1178.
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 alpha. Blood. 2002; 100: 1340–1346.
Mishra A, Chaudhary A, Sethi S. Oxidized omega-3 fatty acids inhibit NF-kappaB activation via a PPARalpha-dependent pathway. Arterioscler Thromb Vasc Biol. 2004; 24: 1621–1627.