Reciprocal Modulation of Toll-like Receptor-4 Signaling Pathways Involving MyD88 and Phosphatidylinositol 3-Kinase/AKT by Saturated and Polyunsaturated Fatty Acids*

Toll-like receptor-4 (TLR4) can be activated by nonbacterial agonists, including saturated fatty acids. However, downstream signaling pathways activated by nonbacterial agonists are not known. Thus, we determined the downstream signaling pathways derived from saturated fatty acid-induced TLR4 activation. Saturated fatty acid (lauric acid)-induced NFκB activation was inhibited by a dominant-negative mutant of TLR4, MyD88, IRAK-1, TRAF6, or IκBα in macrophages (RAW264.7) and 293T cells transfected with TLR4 and MD2. Lauric acid induced the transient phosphorylation of AKT. LY294002, dominant-negative (DN) phosphatidylinositol 3-kinase (PI3K), or AKT(DN) inhibited NFκB activation, p65 transactivation, and cyclooxygenase-2 (COX-2) expression induced by lauric acid or constitutively active (CA) TLR4. AKT(DN) blocked MyD88-induced NFκB activation, suggesting that AKT is a MyD88-dependent downstream signaling component of TLR4. AKT(CA) was sufficient to induce NFκB activation and COX-2 expression. These results demonstrate that NFκB activation and COX-2 expression induced by lauric acid are at least partly mediated through the TLR4/PI3K/AKT signaling pathway. In contrast, docosahexaenoic acid (DHA) inhibited the phosphorylation of AKT induced by lipopolysaccharide or lauric acid. DHA also suppressed NFκB activation induced by TLR4(CA), but not MyD88(CA) or AKT(CA), suggesting that the molecular targets of DHA are signaling components upstream of MyD88 and AKT. Together, these results suggest that saturated and polyunsaturated fatty acids reciprocally modulate the activation of TLR4 and its downstream signaling pathways involving MyD88/IRAK/TRAF6 and PI3K/AKT and further suggest the possibility that TLR4-mediated target gene expression and cellular responses are also differentially modulated by saturated and unsaturated fatty acids.

TLRs are type I transmembrane receptors characterized by the presence of extracellular leucine-rich repeat motifs and a cytoplasmic Toll/interleukin-1 receptor (TIR) homology domain, which is required for the activation of downstream signaling pathways leading to the activation of nuclear factor-B (NFB) (21). Myeloid differentiation factor-88 (MyD88) is known as an immediate downstream adaptor molecule that interacts directly with the TIR domain of TLRs (22,23). MyD88 recruits interleukin-1 receptor-associated kinase (IRAK) and tumor necrosis factor receptor-associated factor-6 (TRAF6), leading to activation of NFB and mitogenactivated protein kinases (MAPKs) (24,25). Activation of NFB leads to the expression of target genes, including cyclooxygenase-2 (COX-2) and cytokines. TIR domain-containing adaptor protein (TIRAP)/MyD88 adaptor-like (Mal) is another adaptor molecule cooperating with MyD88, leading to activation of IRAK-1 and NFB (26 -28). TIR domaincontaining adaptor inducing interferon-␤ (TRIF)/TIR domain-containing adaptor molecule-1 (TICAM-1) has been reported as another adaptor molecule responsible for the MyD88-independent signaling pathway derived from TLR3, leading to the activation of interferon regulatory factor-3 and the expression of interferon-␤ (29 -31). Thus, individual TLR agonist can activate different downstream signaling pathways, leading to differential target gene expression and cellular responses.
Phosphatidylinositol 3-kinase (PI3K) and AKT have been implicated in TLR4 and TLR2 signaling pathways. LPS, a TLR4 agonist, activates PI3K, resulting in the phosphorylation of AKT, a downstream target of PI3K (32,33). It has been demonstrated that TLR2 can associate with Rac1 and PI3K, leading to the activation of AKT (34). AKT, also known as protein kinase B, is a serine/threonine kinase that is activated in response to cytokines and growth factors (35,36). AKT is activated via phosphorylation by PI3K and further phosphorylates its downstream signaling molecules, including glycogen synthase kinase-3, BAD, and caspase-9 (37)(38)(39). AKT has been shown to induce p65 phosphorylation, resulting in enhanced NFB transactivation (40 -44). However, it has not been clearly understood how PI3K/AKT is linked to TLR4 and what the role of PI3K/AKT is in activating downstream signaling pathways of TLR4.
Lipid A, which possesses most of the biological activities of LPS, is acylated with saturated fatty acids. Removal of these acylated saturated fatty acids from lipid A not only results in complete loss of endotoxic activity, but also makes lipid A act as an antagonist to native lipid A (45,46). Lipid A containing unsaturated fatty acids is also known to be nontoxic or to act as an antagonist against endotoxin (47,48). These results suggest that the fatty acids acylated in lipid A play a critical role in ligand recognition and receptor activation for TLR4. Results from our previous studies demonstrated that saturated fatty acids induce NFB activation and the expression of COX-2, but that unsaturated fatty acids inhibit LPS-induced NFB activation and the expression of COX-2 and other inflammatory markers in a murine monocytic cell line (RAW264.7) by modulating the activation of TLR4 (18).
However, the signaling pathways leading to NFB activation by saturated fatty acids are not known. Identifying signaling pathways activated by different TLR agonists can help us to understand the mechanisms by which each agonist regulates the expression of target genes and consequent cellular responses. Therefore, in this study, we determined whether TLR4-derived downstream signaling pathways are modulated by saturated and polyunsaturated fatty acids.
FIG. 1. Lauric acid (C12:0)-induced NFB activation is inhibited by a dominant-negative TLR4 mutant in macrophages (RAW264.7) and 293T cells transfected with TLR4 and MD2. A, RAW264.7 cells were transfected with the (NFB) 2 -luciferase (Luc) reporter plasmid and the expression plasmid of TLR4(DN) or a corresponding vector. Cells were further treated with LPS (200 ng/ml) or lauric acid (75 M) for 18 h. B and C, 293T cells were transfected with the (NFB) 2 -luciferase reporter plasmid and the expression plasmid of wild-type TLR4, MD2, CD14, or TLR4(DN) as indicated. Cells were further treated with LPS (100 ng/ml) or lauric acid (50 M) for 18 h. Cell lysates were prepared, and luciferase and ␤-galactosidase enzyme activities were measured as described under "Experimental Procedures." Relative luciferase activity (RLA) was determined by normalization to ␤-galactosidase activity. Data are representative of more than three independent experiments. Values are means Ϯ S.E. (n ϭ 3).
The expression plasmid of MD2 was obtained from Kensuke Miyake (Saga Medical School, Saga, Japan). CD14 was provided by Richard J. Ulevitch (Scripps Research Institute). A constitutively active form of AKT (myristoylated AKT) and wild-type AKT were provided by Michael Weber (University of Virginia Health Sciences Center). A constitutively active form of MyD88 (MyD88(⌬Toll)) and the dominant-negative mutant MyD88(⌬DD) were kindly provided by Jurg Tschopp (University of Lausanne, Lausanne, Switzerland). Constitutively active chimeric CD4-TLR4 was obtained from C. A. Janeway, Jr. (Yale University, New Cell Culture-RAW264.7 cells (a murine monocytic cell line, ATCC TIB-71) and 293T cells (human embryonic kidney cells; provided by Sam Lee, Beth Israel Hospital, Boston, MA) were cultured in Dulbecco's modified Eagle's medium containing 10% (v/v) heat-inactivated fetal bovine serum (Intergen Co.), 100 units/ml penicillin, and 100 g/ml streptomycin (Invitrogen) at 37°C in a 5% CO 2 and air environment.
Transient Transfection and Luciferase Assay-These were performed as described in our previous studies (10,18,51). Briefly, RAW264.7 or 293T cells were plated in 24-well plates (1.5 ϫ 10 5 cells/well) and cotransfected with a luciferase plasmid containing either the (NFB) 2binding site or the murine COX-2 promoter (Ϫ3.2 kb) and with the HSP70-␤-galactosidase plasmid as an internal control using SuperFect transfection reagent (QIAGEN Inc.) according to the manufacturer's instructions. Various expression plasmids or corresponding empty vector plasmids for signaling components were cotransfected. The total amount of transfected plasmids was equalized by supplementing with the corresponding empty vector to eliminate the experimental error from transfection itself. Luciferase and ␤-galactosidase enzyme activities were determined using the luciferase assay system and the ␤-galactosidase enzyme system (Promega, Madison, WI) according to the manufacturer's instructions. Luciferase activity was normalized to ␤-galactosidase activity.
Immunoblotting-This was performed as previously described (52,53). Briefly, solubilized proteins were subjected to 8% SDS-PAGE and transferred to a polyvinylidene difluoride membrane. The membrane was blocked in 20 mM Tris-HCl, 137 mM NaCl, and 0.05% (v/v) Tween 20 (pH 7.6) containing 5% nonfat dried milk (Carnation). The membrane was immunoblotted with primary antibody for 1-24 h, followed by secondary antibody coupled to horseradish peroxidase (Amersham Biosciences) for 1 h. The membrane was exposed on an x-ray film (Eastman Kodak Co.) using ECL Western blot detection reagents (Amersham Biosciences). To reprobe with different antibodies, the membrane was stripped in stripping buffer (54) at 56°C for 1 h. inhibited by a dominant-negative mutant of TLR4 (Fig. 1A). In this study, we further determined whether lauric acid can activate ectopically expressed TLR4 in 293T cells which do not express endogenous TLR4. The expression of TLR4 alone in 293T cells was not sufficient to induce NFB activation by lauric acid (Fig. 1B). MD2 and CD14 are known to participate in the recognition of LPS by the TLR4 complex (55). Cotransfection of TLR4 and MD2 was sufficient to induce both lauric acid-and LPS-induced NFB activation in 293T cells (Fig. 1B), suggesting that MD2 is required for the activation of TLR4 by saturated fatty acid and LPS. In contrast, cotransfection of CD14 with TLR4 in the absence of MD2 was not sufficient to confer lauric acid and LPS responsiveness to TLR4. However, the overexpression of CD14 with TLR4 plus MD2 resulted in a slight potentiation of NFB activation by lauric acid or LPS as compared with TLR4 plus MD2 in 293T cells (Fig. 1B). A dominant-negative mutant of TLR4 inhibited lauric acid-in-duced NFB activation in 293T cells transfected with TLR4 and MD2 (Fig. 1C). These results demonstrate that lauric acid induces the activation of endogenous or ectopically expressed TLR4 in the presence of MD2.

NFB Activation Induced by Saturated Fatty Acid
Saturated Fatty Acid-induced NFB Activation Is Mediated through the MyD88/IRAK/TRAF6 Signaling Pathway-MyD88 is an immediate adaptor molecule recruited by activated TLR4, leading to the activation of signaling cascades, including IRAK-1, TRAF6, NIK, and IB kinase (3). The activation of IB kinase-␤ results in the phosphorylation of IB␣ at Ser 32 and Ser 36 and the consequent degradation of IB␣. This degradation leads to the nuclear translocation and DNA binding of NFB (56). We determined whether lauric acid stimulates the MyD88-dependent signaling pathways. Dominant-negative MyD88 inhibited lauric acid-or LPS-induced NFB activation in both RAW264.7 cells ( Fig. 2A)  activation was also inhibited by a dominant-negative IRAK-1, TRAF6, NIK, or IB␣ in 293T cells transfected with TLR4 and MD2 (Fig. 2, C and D). A dominant-negative mutant of TIRAP, another adaptor molecule of TLR4 that cooperates with MyD88dependent signaling, also inhibited lauric acid-induced NFB activation (Fig. 2B). These results demonstrate that lauric acid-induced NFB activation is mediated through MyD88-dependent signaling pathways and that MyD88, TIRAP, IRAK-1, and TRAF6 are common downstream signaling components shared by both saturated fatty acid and LPS.
The activation of TLR4 also leads to the activation of MAPKs (3). Lauric acid induced not only the degradation of IB␣, but also the transient phosphorylation of c-Jun N-terminal kinase (JNK) and extracellular signal-regulation kinase (ERK) in RAW264.7 cells (Fig. 3). These results demonstrate that lauric acid induces the activation of MAPK signaling pathways as well as NFB activation.
Saturated Fatty Acid-induced NFB Activation Is Also Mediated through the PI3K/AKT Pathway-LPS, a TLR4 agonist, is known to activate PI3K and AKT, a downstream target of PI3K (32). AKT is known to regulate the activation of NFB (40 -44). However, it has not been known how PI3K/AKT is involved in TLR4 signaling. We determined whether the activation of NFB induced by TLR4 and saturated fatty acid is mediated through the PI3K/AKT signaling pathway. LY294002, a PI3K inhibitor, and a dominant-negative mutant of PI3K (p85(DN)) or AKT (AKT(DN)) suppressed NFB activation induced by LPS in RAW264.7 cells (Fig. 4, A and B). NFB activation induced by constitutively active (CA) TLR4 (⌬TLR4(CA)) was also inhibited by p85(DN) or AKT(DN) (Fig.  4C). AKT(DN) blocked MyD88-induced NFB activation (Fig.  4D), suggesting that AKT is the downstream signaling component of MyD88. AKT(CA) alone was sufficient to induce NFB activation (Fig. 4E). These results suggest that PI3K/AKT is the downstream signaling pathway activated by TLR4 and that AKT activation is required and sufficient for NFB activation mediated through TLR4.
Lauric acid-induced NFB activation was also inhibited by LY294002, p85(DN) or AKT(DN) in both RAW264.7 and 293T cells transfected with TLR4 and MD2 (Fig. 5). These results suggest that NFB activation induced by saturated fatty acid is mediated through TLR4 and the PI3K/AKT signaling pathway. Indeed, lauric acid induced the rapid and transient phosphorylation of AKT, followed by the phosphorylation of glycogen synthase kinase-3, the substrate of AKT in RAW264.7 cells (Fig. 6A), demonstrating the activation of PI3K and AKT by lauric acid. The activation of AKT induces the phosphorylation and transactivation of p65, a subunit of NFB, leading to the activation of NFB (40 -44). Thus, we determined whether lauric acid induces the phosphorylation of p65 by immunoblotting. In addition, the transactivation of p65 was determined using the p65/Gal4 plasmid containing p65 fused to the DNA-binding domain of the Gal4 transcription factor and the plasmid containing the Gal4responsive element-luciferase reporter gene. Lauric acid induced the phosphorylation of p65 and also increased the transactivation of p65 (Fig. 6, B and C). This transactivation was inhibited by LY294002 and p85(DN) or AKT(DN) (Fig. 6C). These results suggest that NFB activation induced by lauric acid is at least partly mediated through AKT activation, leading to p65 transactivation. Together, these results suggest that lauric acid activates TLR4, leading to the activation of both the MyD88/IRAK/TRAF6/ NIK/NFB and MyD88/PI3K/AKT/NFB pathways.

COX-2 Expression by Saturated Fatty Acid Is Mediated through the TLR4/PI3K/AKT Pathway in Macrophages-
COX-2 was one of the target genes that were induced by saturated fatty acid through the activation of TLR4 in RAW264.7 cells (18). The increase in COX-2 expression induced by lauric acid started at 4 h (Fig. 7A) and continued at least until 18 h (data not shown). We determined whether saturated fatty acidinduced COX-2 expression is mediated through the PI3K/AKT signaling pathway. p85(DN) or AKT(DN) inhibited COX-2 expression induced by ⌬TLR4(CA) (Fig. 7B) or lauric acid (Fig.  7C). LY294002 also suppressed both COX-2 promoter activity and COX-2 protein expression induced by lauric acid (Fig. 7, D  and E). The expression of AKT(CA) resulted in COX-2 expression (Fig. 7F). These results suggest that AKT activation is required and sufficient for COX-2 expression mediated through TLR4 activation and that COX-2 expression induced by saturated fatty acid is at least partly mediated through the PI3K/ AKT signaling pathway.
Docosahexaenoic Acid (DHA) Inhibits the Phosphorylation of AKT and the Activation of NFB Induced by the Activation of TLR4 -The results from our previous studies showed that DHA, an n-3 polyunsaturated fatty acid, inhibits COX-2 expression induced by TLR4 (18). In this study, we determined whether DHA inhibits the activation of PI3K/AKT and NFB induced by TLR4 activation. DHA suppressed the phosphorylation of AKT induced by LPS or lauric acid (Fig. 8A), demon- strating that DHA inhibits the activation of the PI3K/AKT pathway derived from TLR4 activation. DHA also inhibited NFB activation induced by TLR4(CA), but not MyD88(CA) or Myr-AKT(CA) (Fig. 8, B-D). These results suggest that the molecular targets of unsaturated fatty acids are not the downstream signaling components of TLR4, but the TLR itself or its associated molecules.

DISCUSSION
Results from our previous studies showed that saturated fatty acids induce, but unsaturated fatty acids inhibit, the activation of TLR4 as determined by NFB activation and COX-2 expression (18,51). These results suggest the novel role of dietary fatty acids as a modulator of TLR4 signaling. However, the downstream signaling pathways through which the fatty acids modulate NFB activation and target gene expression have not been determined. Different TLR agonists may induce the activation of different downstream signaling pathways, leading to the diverse array of target gene expression and cellular responses. Therefore, in this study, we determined how the TLR4 signaling pathways are regulated by saturated and unsaturated fatty acids. Tentative downstream signaling pathways derived from TLR4 activation that are modulated by fatty acids are illustrated in Fig. 9. The results in this study demonstrate that saturated fatty acid and LPS share the common signaling pathways of TLR4 involving the MyD88, IRAK-1, TRAF6, and PI3K/AKT pathways, leading to NFB activation and target gene expression, including COX-2. These results reinforce the possibility that both saturated fatty acids and LPS interact with the same receptor complex. Furthermore, these results provide an important clue to elucidate the molecular mechanisms by which LPS and saturated fatty acids induce target gene expression and the consequent cellular responses through the TLR4 complex. Our results also establish that saturated and polyunsaturated fatty acids reciprocally modulate the TLR4 signaling pathways; saturated fatty acid induces phosphorylation of AKT and activation of NFB through TLR4 activation, whereas unsaturated fatty acid inhibits TLR4-induced AKT phosphorylation and NFB activation.
The activation of NFB is known to be regulated by at least two different mechanisms: the phosphorylation and consequent degradation of the inhibitor protein IB␣ and the phosphorylation of p65, a subunit of NFB, in a cooperative manner (40, 56 -58). IB␣ is phosphorylated by IB kinase activated through the MyD88/IRAK/TRAF6 pathway and subsequently degraded, resulting in nuclear translocation and DNA binding of NFB. p65 is phosphorylated by serine kinases, including protein kinase A (Ser 276 ) and IB kinase (Ser 536 ) (42,43). The phosphorylation of p65 is critical for the interaction with the coactivator proteins p300 and CBP to promote the transcriptional activity of NFB (43,59). PI3K/AKT signaling has been suggested as one of the regulators for NFB activation induced by inflammatory stimuli such as tumor necrosis factor, interleukin-1, and LPS (32,40,60,61). Several studies demonstrated that AKT can enhance the transcriptional activity of NFB through the phosphorylation of p65 independent of IB␣ degradation (40 -44). Our results demonstrated that, in addition to the degradation of IB␣ (Fig. 3), lauric acid induces the serine phosphorylation of p65 at Ser 529 and the transactivation of p65 (Fig. 6). p65 transactivation was associated with AKT activation induced by saturated fatty acid (Fig. 6). This is consistent with a previous report that the activation of p65 by AKT is accompanied by the phosphorylation of p65 at Ser 529 (43). Because the magnitude of IB␣ degradation by lauric acid was mild, it is possible that the phosphorylation of p65 may be a major mechanism by which saturated fatty acid promotes NFB activation. Together, these results suggest that the full activation of NFB by saturated fatty acids requires both IB␣ degradation via the MyD88/IRAK/TRAF6 pathway and p65 transactivation by the PI3K/AKT pathway.
Synthetic bacterial lipopeptides (TLR2 agonist) and CpG DNA (TLR9 agonist) can neither activate NFB nor induce the expression of inflammatory cytokines in macrophages derived from MyD88-deficient mice (62)(63)(64). However, unlike TLR2 and TLR9 agonists, LPS can still induce the activation of NFB and MAPKs, the production of interleukin-18, and the maturation of dendritic cells in a delayed fashion in MyD88-deficient cells (64 -67). These results suggest that TLR4 is coupled to an additional signaling pathway that is independent of MyD88. TIRAP/Mal has been identified as an adaptor molecule associated with TLR4, leading to the activation of IRAK-1 and NFB (26,27). It has been suggested that TIRAP may mediate the MyD88-independent signaling pathway. However, it was recently demonstrated that LPS-induced expression of interferon-inducible genes such as IP-10, GARG-16, and IRG-1 was still observed in macrophages derived from MyD88 and TIRAP double-knockout mice (28). These results suggest that TIRAP cooperates with MyD88 and participates in the MyD88-dependent signaling pathway and that there may be additional adap-tors responsible for the MyD88-independent signaling pathway. Recently, TRIF/TICAM was reported to mediate the MyD88-independent signaling pathway of TLR3 (30,31). Our results show that both MyD88 and TIRAP are involved in the signaling pathways activated by saturated fatty acid. It remains to be determined whether saturated fatty acid can also activate MyD88-independent signaling components like TRIF.
The PI3K/AKT pathway was suggested as one of the downstream signaling pathways of TLRs. Wortmannin, a PI3K inhibitor, inhibits NFB activation and cytokine production by CpG DNA, a TLR9 agonist (68). The activation of NFB and AP-1 induced by LPS, a TLR4 agonist, is also inhibited by wortmannin (69). B lymphocytes deficient in PI3K regulatory units (p85␣, p55␣, and p50␣) fail to respond to LPS (70). Arbibe et al. (34) demonstrated that TLR2 stimulation by Staphylococcus aureus induces the recruitment of PI3K and the activation of AKT, leading to NFB transactivation. These results demonstrate the role of PI3K in TLR signaling pathways. However, unlike TLR2, TLR4 does not have a PI3K-binding site. Therefore, it has not been clear how TLR4 activation leads to PI3K/AKT activation and whether TLR4-mediated AKT activation is MyD88-dependent or -independent. Our results show that AKT(DN) inhibited NFB activation induced by MyD88(CA) (Fig. 4D). MyD88 has a binding motif for p85, a regulatory unit of PI3K. Recently, Ojaniemi et al. (71) reported that PI3K can be associated with MyD88 in response to LPS in mouse macrophages. Together, these results suggest that PI3K/AKT is the downstream signaling component of the MyD88-dependent TLR4 signaling pathway.
The recognition of LPS is mainly mediated through the interaction with three proteins, TLR4, CD14, and MD2. LPS binds to the serum protein LPS-binding protein, and the LPS⅐LPS-binding protein complex can be recognized by CD14. CD14, a cell-surface 55-kDa glycoprotein with a glycosylphosphatidylinositol membrane anchor, is present as a soluble form in blood or as a membrane-bound form in myeloid lineage cells (72,73). Because CD14 does not have a cytosolic domain, it has been suggested that CD14 interacts with other receptors to transduce the LPS signal into cells. Recently, TLR4 was identified as an LPS receptor (10,74,75), and it has been further shown that LPS can induce physical proximity between CD14 and TLR4 using fluorescence resonance energy transfer techniques (76). MD2 is a 20 -30-kDa glycoprotein that binds to the extracellular domain of TLR4 (55). TLR4 and MD2 provide greater specificity for ligands and more efficient responsiveness to LPS (55). A recent study also showed that LPS is in close proximity to TLR4 in the presence of CD14 and MD2 (77). Our results show that cotransfection of TLR4 with MD2 was sufficient to induce NFB activation induced not only by LPS, but also by lauric acid in 293T cells, suggesting that MD2 is required for both saturated fatty acid-and LPS-induced TLR4 signaling. Although cotransfection of CD14 with TLR4 in the absence of MD2 did not confer the responsiveness of LPS or lauric acid to TLR4, the cotransfection of CD14 with TLR4 plus MD2 resulted in a slight potentiation of NFB activation by lauric acid or LPS compared with TLR4 plus MD2 in 293T cells. The serum in the culture medium contains soluble CD14, and both soluble CD14 and membrane-bound CD14 can function to confer LPS responsiveness (72,78). Therefore, our experimental conditions do not allow us to conclude whether CD14 is required for saturated fatty acid-induced TLR4 activation. We cannot exclude the possibility that CD14 also contributes to the activation of TLR4 by saturated fatty acid. Together, our results demonstrate that the responsiveness of lauric acid to the combination of TLR4, MD2, and CD14 is similar to that of LPS, suggesting that both saturated fatty acid and LPS activate the signaling pathways through the same receptor complex. The results that unsaturated fatty acids inhibit NFB activation induced by TLR4, but not the downstream signaling components, suggest that the inhibitory target of unsaturated fatty acids is TLR4 or its associated molecules. However, the detailed molecular mechanism by which the fatty acids modulate TLR activation is not known. The molecular mechanism by which LPS interacts with the TLR4 complex is also not well understood, although it has been shown that LPS stimulation leads to close proximity between TLR4, CD14, and MD2 (77). Whether saturated fatty acids induce and unsaturated fatty acids inhibit the physical proximity between TLR4, CD14, and MD2 to modulate TLR signaling needs to be determined in the future studies.
COX-2 is overexpressed in sites of inflammation and in many types of tumors and is involved in tumorigenesis and angiogenesis (50, 79 -82). Our results demonstrating that saturated fatty acid induces, but DHA inhibits, NFB activation and COX-2 expression suggest the possibility that the risk of tumors and inflammatory diseases can be differentially modulated by types of dietary fatty acids. Together, these results suggest the possibility that TLR-mediated target gene expression and cellular responses may be differentially modulated by saturated and unsaturated fatty acids through the reciprocal modulation of TLR signaling pathways.