Toll-Like Receptor 4–Dependent and –Independent Cytokine Secretion Induced by Minimally Oxidized Low-Density Lipoprotein in Macrophages
http://www.100md.com
动脉硬化血栓血管生物学 2005年第6期
From the Division of Endocrinology and Metabolism (Y.I.M., D.S.W., A.B., S.B., J.L.W.), Department of Medicine, and Veterans Administration (S.V.), San Diego Healthcare System and Department of Pathology and Medicine, University of California, San Diego.
Correspondence to Yury I. Miller, MD, PhD, University of California, San Diego, 9500 Gilman Dr, La Jolla, CA 92093-0682. E-mail yumiller@ucsd.edu
Abstract
Objective— Innate immune responses to oxidized low-density lipoprotein LDL (LDL) regulate the development of atherosclerosis. We demonstrated previously that an early form of oxidized LDL, minimally modified LDL (mmLDL), triggers cytoskeletal rearrangements in macrophages via CD14 and Toll-like receptor 4 (TLR4)/MD-2. Because lipopolysaccharide (LPS) activation of TLR4 leads to proinflammatory gene expression, in this study, we asked whether mmLDL also induced proinflammatory signaling.
Methods and Results— We studied cytokine secretion and signaling in J774 and primary peritoneal macrophages stimulated with mmLDL, which was prepared by incubating LDL with cells expressing human 15-lipoxygenase. MmLDL stimulated robust phosphoinositide 3-kinase (PI3K) activation, and Akt and extracellular signal-regulated kinase 1/2 (ERK1/2) phosphorylation, which exceeded that induced by LPS. On the other hand, although mmLDL induced nuclear factor B (NF-B) p65 translocation to the nucleus, there was no detectable NF-B activation. However, mmLDL induced early mRNA and protein expression of the cytokines MIP-2, MCP-1, tumor necrosis factor-, and interleukin-6. Chemokine MIP-2 but not MCP-1 secretion depended on TLR4/MyD88, ERK1/2, and PI3K signaling. In turn, TLR4 regulated phosphorylation of ERK1/2 but not of Akt, suggesting that mmLDL-induced PI3K activation is TLR4 independent.
Conclusions— In macrophages, mmLDL activates TLR4-dependent and -independent signaling pathways, resulting in secretion of proinflammatory cytokines. These results provide new insights into the inflammatory origins of atherosclerosis.
Minimally oxidized LDL (mmLDL) stimulated PI3K activation, Akt, and ERK1/2 phosphorylation, and induced mRNA and protein expression of MIP-2, MCP-1, TNF-, and IL-6 in macrophages. MIP-2 but not MCP-1 secretion depended on TLR4/MyD88, ERK1/2, and PI3K signaling. Thus, mmLDL activates TLR4-dependent and independent proinflammatory signaling in macrophages.
Key Words: macrophage ? oxidized low-density lipoprotein ? toll-like receptor 4 ? phosphoinositide 3-kinase ? cytokines
Introduction
Toll-like receptor 4 (TLR4) is a key signaling receptor of innate immunity because it senses the presence of bacterial lipopolysaccharide (LPS) associated with CD14 or MD-2 and initiates a proinflammatory signaling cascade. The classical mechanism by which LPS-induced TLR4 signaling occurs includes activation of mitogen-activated protein (MAP) kinases and nuclear factor B (NF-B), resulting in secretion of proinflammatory cytokines such as tumor necrosis factor- (TNF-), interleukin-1? (IL-1?), IL-6, and others.1 Moreover, a fundamental biological importance of TLR4 was demonstrated recently by the discovery that TLR4-deficient mice of various strains have less body fat and stronger bones than wild-type mice.2
See page 1085
Atherosclerosis is a chronic inflammatory disease of the vascular wall, and various forms of oxidized low-density lipoprotein (LDL) are key proinflammatory agents present during all stages of atherogenesis.3–5 We demonstrated recently that a very early form of oxidized LDL, minimally modified LDL (mmLDL), binds to the LPS receptor CD14 and stimulates macrophage spreading and actin polymerization.6 Furthermore, the mmLDL-induced macrophage spreading inhibits phagocytosis of apoptotic cells but promotes uptake of oxidized LDL by macrophages, events that should be proatherogenic. Remarkably, the mmLDL-induced cytoskeletal changes depend on TLR4/MD-2.6 Indeed, deficiency of TLR4 or myeloid differentiation factor-88 (MyD88), an adaptor molecule downstream from TLR4 and other Toll-like receptors, significantly reduces diet-induced atherosclerosis in the apolipoprotein E–/– mouse model.7,8 Moreover, human epidemiological data demonstrate that an Asp299Gly TLR4 polymorphism, which attenuates receptor signaling, is associated with a decreased risk of atherosclerosis and acute coronary events.9,10 Because mmLDL represents an early form of oxidized LDL found in vivo in the artery wall,11 and because the mmLDL-induced changes in cytoskeleton depend on TLR4, it is potentially highly relevant and important to elucidate the mechanisms by which mmLDL interactions with TLR4 affect cellular signaling and thus may impact atherogenesis.
In addition to effects on macrophage spreading and phagocytosis, we wondered whether incubation of mmLDL with macrophages would also result in the activation of MAP kinases and NF-B and in cytokine production. Berliner et al have suggested that oxidation products of 1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphocholine (Ox-PAPC), such as 1-palmitoyl-2-(5-oxovaleroyl)-sn-glycero-3-phosphocholine (POVPC), found in oxidized LDL and in mmLDL, induce secretion of chemokines IL-8 and monocyte chemoattractant protein-1 (MCP-1) by human endothelial cells and monocytes.12,13 They have also shown that the endothelial IL-8 production stimulated by Ox-PAPC depends on TLR4.13 Ox-PAPC activates MAP kinases but not NF-B in endothelial cells.14 Paradoxically, Ox-PAPC has been reported to inhibit LPS-induced activation in endothelial cells, and when administered in vivo to mice, to protect from LPS-induced tissue damage.15,16
In this study, we analyze intracellular events and cytokine production in mouse macrophages initiated by mmLDL stimulation (in comparison with native LDL [nLDL] and oxidized LDL). We found that part of but not all macrophage responses to mmLDL depended on TLR4/MyD88 and that the mmLDL-induced signaling differed from the LPS-induced signaling. These findings provide new insights into the potential role of early forms of oxidized LDL in modulating inflammatory events in atherosclerosis.
Methods
An extended Methods section is available online at http://atvb.ahajournals.org.
Cells
Resident peritoneal macrophages were harvested from 8- to 10-week-old female mice of wild-type (C57BL/6), TLR4–/–, or MyD88–/–. The TLR4–/– and MyD88–/– mice were kindly provided by Dr Shizuo Akira from Research Institute for Microbial Diseases, Osaka University, Japan.17,18 (We also thank Drs Eyal Raz and Peter Tobias for providing knockout mice.) Murine macrophage-like cell line J774A.1, a murine fibroblast cell line stably overexpressing human 15-lipoxygenase (15-LO)19 and Chinese hamster ovary (CHO) cell lines, constitutively expressing an NF-B reporter together with TLR4 or TLR4/MD-2,20 were maintained as described previously.6
LDL Isolation and Modification
LDL was isolated from plasma of normolipidemic donors by sequential ultracentrifugation.21 Contamination of nLDL and modified LDL preparations by endotoxin (LPS) was assessed with a LAL kit (BioWhittaker). LDL preparations with LPS >50 pg/mg protein were discarded. For extensive oxidation, the LDL (1 mg/mL) was incubated with 10 μmol/L CuSO4 for 18 hours at 37°C.
Minimally Modified (Oxidized) LDL
To produce mmLDL, we incubated 50 μg/mL LDL in serum-free DMEM for 18 hours with a murine fibroblast cell line overexpressing 15-LO.19 We documented previously that this procedure generates mmLDL (ie, it binds to nLDL receptors but not to scavenger receptors).6,19,22–24 MmLDL contains early lipid peroxidation products,22,24 but it does not contain any measurable thiobarbituric acid reactive substances or EO6-reactive phospholipid oxidation products above that of nLDL.6 The mmLDL modification appeared to be very reproducible, and a successful modification was documented in a biological assay in which mmLDL induced spreading of J774 macrophages in cell culture.6,24
Cell Signaling and Cytokine Assays
Four different assays were used to assess NF-B translocation and activation: p65 nuclear translocation, p65-DNA binding, NF-B reporter gene expression, and p65 phosphorylation (please see supplemental material for details).
Phosphorylation of signaling proteins was assessed by Western blot. Phosphoinositide 3-kinase (PI3K) activity was measured in cell lysates and immunoprecipitated with an antiphosphotyrosine antibody, as described25 in a reaction with phosphatidylinositol in the presence of [-32P]ATP.
Real-time polymerase chain reaction (PCR) of reverse-transcribed total RNA was performed using an Applied Biosystems ABI Prism 7700 Sequence Detection System, as performed in Genomic Core, Center for AIDS Research, University of California, San Diego.
Cytokine ELISAs were performed according to standard antibody manufacturer protocols, except that chemiluminescent detection was used, as described previously from our laboratory.26
Results
MmLDL-Induced Cell Signaling
Our model of early forms of oxidized LDL, mmLDL, which is produced by incubation with 15-LO–expressing fibroblasts in DMEM media, contains mostly early lipid oxidation products, such as fatty acid hydroperoxides and hydroxides.22,24 MmLDL does not bind EO624 and thus likely has few if any of the bioactive oxidized phospholipids found in Ox-PAPC. In contrast, Ox-PAPC and the lipid moieties of oxidized LDL bind CD36 and EO6.27–29 Because of these differences and because mmLDL but not Ox-PAPC binds CD14,6,13 we investigated intracellular signaling pathways, TLR4-dependent and -independent, stimulated by mmLDL.
We examined NF-B activation in mmLDL-stimulated macrophages by 4 independent techniques. First, we analyzed NF-B p65 translocation from the cytosol to the nucleus of J774 cells by immunocytochemistry. As expected, nuclear localization of p65 was observed in the LPS-treated macrophages (Figure 1A). Remarkably, a majority of the cells treated with mmLDL also showed clear nuclear localization of p65. In contrast, p65 in control macrophages and in the cells treated with either nLDL or oxidized LDL predominantly located in the cytosol. As we reported previously, mmLDL and, to a lesser degree, LPS, induced macrophage spreading, which was also observed in this experiment (note red staining for F-actin in Figure 1A). However, nuclear p65 localization was not a function of spreading, because in the relatively spread oxidized LDL–treated cells, p65 was still cytosolic (see Figure 4).
Figure 1. MmLDL induces p65 nuclear translocation but not NF-B activation. A, J774 cells were activated with 50 μg/mL nLDL, mmLDL, or oxidized LDL, or 10 ng/mL LPS for 15 minutes, fixed, and stained for p65 (green), nucleic acid (blue), and F-actin (red). Each treatment is presented by 2 images of the same cell. On the left, nucleus (blue) is contrasted by actin cytoskeleton (red), and on the right, p65 cellular localization is shown in green. B, Nuclear extracts were made from J774 cells after 15 minutes of application of 50 μg/mL nLDL, mmLDL, or oxidized LDL, or 100 ng/mL LPS. Binding of NF-B p65 to a specific DNA sequence was assessed by a TransAM ELISA assay, which yields quantitative results that parallel those of a gel-shift assay (see Methods). C, CHO cell lines bearing inducible membrane CD25 under the transcriptional control of a human E-selectin promoter containing NF-B binding sites, and stably transfected with either TLR4 or TLR4/MD-220 were treated with 50 μg/mL mmLDL or 100 ng/mL LPS for 24 hours, and at the end of incubation, analyzed for CD25 surface expression by fluorescence-activated cell sorter (FACS). Note that LPS only induces CD25 in CHO cells transfected with TLR4/MD-2.
Figure 4. Cytokine protein secretion by wild-type (WT), TLR4–/– and MyD88–/– peritoneal macrophages in response to mmLDL. Resident peritoneal macrophages were stimulated for 6 hours with 50 μg/mL mmLDL or media alone. At the end of incubation, culture media were collected and analyzed by ELISA for the presence of MIP-2 (A), MCP-1 (B; note the Y scale break), TNF- (C), and IL-6 (D). *P<0.005. KO indicates knockout.
NF-B nuclear localization is necessary but not sufficient for its activity. Therefore, we directly tested p65 binding to DNA using an ELISA-based immunoassay, which is reported to be more sensitive than traditional gel-shift assays.30 However, no p65-DNA binding was observed in the nuclear extracts of the cells treated with mmLDL, whereas the LPS-treated cells showed the expected p65-DNA binding (Figure 1B). To confirm this result, we performed a third assay in which CHO cells, stably transfected with an NF-B reporter (surface CD25 expression; see Methods) and either TLR4 or TLR4/MD-2, were treated with mmLDL or LPS. In accordance with the p65-DNA binding results, mmLDL failed to stimulate any increase in NF-B activity, as measured by CD25 surface expression, whereas LPS did (Figure 1C). In yet another assay, mmLDL did not stimulate p65 phosphorylation (Figure 2B). Thus, the mmLDL-induced NF-B nuclear translocation was not associated with detectable NF-B activation.
Figure 2. PI3K activation and protein phosphorylation in response to mmLDL and LPS. A, J774 cells were activated with 100 ng/mL LPS or 50 μg/mL mmLDL or oxidized LDL for 15 minutes. At the end of incubation, cell lysates were prepared to analyze PI3K activity and protein phosphorylation as described in Methods. B, J774 cells were incubated with 100 ng/mL LPS or 50 μg/mL mmLDL for indicated times, and the cell lysates were analyzed for protein phosphorylation. Because prolonged treatments of up to 1 hour could have affected protein expression, here we also analyzed total p65 and Akt proteins. C, Resident peritoneal macrophages from wild-type or TLR4–/– mice were incubated with or without 50 μg/mL mmLDL for 15 minutes, and the cell lysates were analyzed for protein phosphorylation and total protein expression. D, J774 cells were pretreated for 30 minutes with 100 nmol/L wortmannin (Wortm) or 10 μmol/L LY294002, and then incubated for additional 15 minutes with 50 μg/mL mmLDL.
As we reported previously, mmLDL induces strong PI3K activation in mouse macrophages.6,24 In the current study, we found that the degree of activation of PI3K by mmLDL was even stronger than that achieved by 100 ng/mL of LPS (Figure 2A). Accordingly, the PI3K target Akt was phosphorylated extensively. The mmLDL-induced Akt phosphorylation was 6- to 8-fold stronger than LPS-induced phosphorylation, and oxidized LDL induced no Akt phosphorylation. Further downstream, Akt phosphorylated GSK-3? (Figure 2A). MmLDL stimulated profound extracellular signal-regulated kinase 1/2 (ERK1/2) phosphorylation, whereas the LPS effect on ERK1/2 was rather modest (Figure 2B).
To further compare mmLDL and LPS stimulation, we analyzed the time course of phosphorylation of relevant signaling proteins in J774 macrophages (Figure 2B). The mmLDL-stimulated phosphorylation of ERK1/2, Akt, and, to some extent, of JNK, sustained for the whole 1 hour of incubation. There was no detectable phosphorylation of p65 by mmLDL, confirming the lack of NF-B activation. In contrast, LPS induced strong p65 phosphorylation, which peaked at 10 minutes, and evident phosphorylation of MAP kinases, ERK1/2, p38, and JNK, which peaked at 30 minutes, but no noticeable Akt phosphorylation (Figure 2B).
ERK1/2 but not PI3K Activation Depends on TLR4
To identify which of the above signaling events induced by mmLDL depend on TLR4, we tested phosphorylation of signaling proteins in the lysates of macrophages from wild-type and TLR4–/– mice. ERK1/2 phosphorylation was clearly dependent on TLR4. On the other hand, mmLDL-induced phosphorylation of either Ser473 or Thr308 of Akt was similar in wild-type and in TLR4–/– macrophages (Figure 2C), suggesting that PI3K and downstream Akt signaling are TLR4 independent.
PI3K Activation by mmLDL Does Not Silence NF-B
Next, we tested the hypothesis that a strong PI3K activation by mmLDL might be responsible for the inhibition of NF-B activation. Indeed, Guha and Mackman31 proposed that the LPS activation of PI3K/Akt in human monocytic cells downregulates NF-B transactivation via several mechanisms, including inactivation of GSK-3? via Akt-dependent phosphorylation. As expected, inhibition of PI3K by either wortmannin or LY294002 abolished mmLDL-induced Akt and GSK-3? phosphorylation (Figure 2D). It appeared that ERK phosphorylation also depended to some extent on PI3K activity (Figure 2D). However, preserving GSK-3? in its active (nonphosphorylated) state during the stimulation with mmLDL did not result in any detectable increase in p65 phosphorylation (Figure 2D). Further, wortmannin and LY294002 treatments failed to increase p65-DNA binding in the nuclear extracts of mmLDL-treated J774 cells (data not shown).
MmLDL-Induced Cytokine Expression
As shown in Figures 1 and 2, we observed that mmLDL induced p65 nuclear translocation but were unable to detect NF-B activation. In addition, mmLDL initiated PI3K and ERK1/2 signaling. Thus, next we examined whether these signaling events resulted in macrophage expression of proinflammatory cytokines. We tested 2 of the most common LPS-stimulated genes, TNF- and IL-6, as well as 2 chemokines, MIP-2 (a mouse analog of human IL-8) and MCP-1, which are stimulated not only by LPS but also by Ox-PAPC.12 In addition, we measured interferon-? (IFN-?) because its expression is Toll/IL-1R domain-containing adapter inducing IFN-? (TRIF) dependent and MyD88 independent.1,32
MmLDL stimulation of J774 macrophages produced early, within 30 to 60 minutes, expression of mRNA for macrophage inflammatory protein-2 (MIP-2), MCP-1, TNF-, and IL-6. This was not seen with nLDL or oxidized LDL (Figure 3). The mRNA expression for these cytokines decreased to basal levels by 3 hours after stimulation. There was no significant difference in the IFN-? expression induced by mmLDL, nLDL, or oxidized LDL. These data are consistent with mmLDL stimulation of the MyD88 but not of the TRIF pathway.1,32 Although mmLDL induced a significant, up to 25-fold, increase in cytokine mRNAs, LPS (10 ng/mL) induced much stronger mRNA expression of MIP-2, TNF-, and IL-6, but not of MCP-1 (Figure 3A through 3F). In addition, the LPS-induced IL-6 expression did not peak at 1 hour as it did with mmLDL stimulation, but increased steadily for 6 hours (Figure 1F).
Figure 3. Time course of cytokine mRNA expression in J774 cells. J774 macrophages were stimulated with 50 μg/mL of nLDL, mmLDL, or oxidized LDL (A through E), or 10 ng/mL LPS (F). At the indicated time points, the reactions were stopped and total RNA was isolated, reverse-transcribed, and subjected to real-time PCR with the primers specific for cytokines and GAPDH. The data are normalized to the GAPDH levels in each sample and presented as a fold increase compared with the 0 time point. Note that A through E show individual cytokines stimulated by various modified LDL, whereas F contains data for all measured cytokines stimulated by LPS.
MIP-2, MCP-1, and TNF- protein secretion, as measured in the J774 culture media by ELISA after a 6-hour incubation with nLDL, mmLDL, or oxidized LDL, was consistent with the mRNA results, except that there was no IL-6 protein secretion induced by mmLDL (supplemental Figure I, available online at http://atvb.ahajournals.org), even after 24 hours of incubation (data not shown). In addition to the data presented, we also measured IL-12 and IL-1? production by J774 cells but were unable to detect any mmLDL stimulation of these cytokines (data not shown).
MmLDL-Induced MIP-2 Secretion Depends on TLR4/MyD88 and PI3K
The cytokine production induced by mmLDL in wild-type resident peritoneal macrophages was similar to that in J774 cells (Figure 4). Of note is a dramatic increase (up to 10 000 pg/mL) in the MCP-1 production by mouse macrophages induced by mmLDL (Figure 4B, scale break). To test the role of TLR4 pathway in mediating mmLDL-induced cytokine expression, we used macrophages harvested from TLR4–/– and MyD88–/– mice. We found that the secretion of only MIP-2 was clearly dependent on TLR4/MyD88 (Figure 4A). In contrast, MCP-1 induction was clearly independent of TLR4/MyD88, and surprisingly, the minute TNF- and IL-6 production induced by mmLDL was also independent of TLR4/MyD88 (Figure 4B through 4D).
Inhibition of PI3K by wortmannin or LY294002 completely blocked MIP-2 and TNF- secretion (supplemental Figure IIA and IIC, available online at http://atvb.ahajournals.org), although they did not significantly change MCP-1 production (supplemental Figure IIB). Similarly, U0126, through its inhibition of MAP kinase kinase and accordingly ERK1/2, also effectively blocked MIP-2 and TNF- (supplemental Figure IIA and IIC). These results suggest that PI3K and ERK positively regulate MIP-2 and TNF-.
Discussion
In the current study, we analyzed intracellular signaling events and cytokine secretion by mouse macrophages induced by mmLDL. MmLDL stimulated early mRNA and protein expression of MIP-2, MCP-1, TNF-, and IL-6 (Figures 3 and 4; Figure I), which were regulated differentially by TLR4/MyD88, PI3K/Akt, and ERK1/2 pathways (Figure 4; Figure II). In contrast to LPS, the mmLDL stimulation showed a very different pattern of intracellular signaling, with stronger PI3K/Akt and ERK1/2 activation and undetectable activation of NF-B (Figures 1 and 2).
There is now substantial evidence that human 15-LO (and its mouse analog 12/15-LO) can mediate the oxidation of LDL,19,22,33,34 that it has a proatherogenic role in murine models of atherosclerosis,35–37 and that it is expressed in the intima of human atherosclerotic lesions.34,38 Because in this study we aimed at modeling a likely early form of modified LDL found in the intima, we produced our mmLDL by incubating LDL with 15-LO–expressing cells. It is likely that early forms of oxidized LDL may be generated by other mechanisms as well, such as by myeloperoxidase28,39 or even 5-LO.40
Special attention was paid to ensure that all the LDL preparations used in this work were free from biologically relevant levels of LPS contamination. In all experiments, there was <2.5 pg/mL of LPS in the incubations used, concentrations below those reported to have effects in murine cells. Indeed, the fact that the mmLDL effects differed substantially from the effects of nLDL and oxidized LDL, even though every experiment was controlled by preparations made from the same LDL pool, further argues against a role of LPS contamination to explain the mmLDL effects. In addition, mmLDL induced very different patterns of cell signaling than did LPS (at 100 ng/mL). These arguments support our conclusion that the biological effects of mmLDL demonstrated in this article are not attributable to LPS contamination.
MmLDL interaction with CD14 and TLR4/MD-26 and the mmLDL-induced early cytokine expression in macrophages (Figures 3 and 4; Figure I) imply involvement of the TLR4/MyD88-dependent pathway. Indeed, using macrophages from TLR4–/– and MyD88–/– mice, we demonstrated that secretion of 1 of the 4 cytokines tested, MIP-2, was clearly dependent on TLR4/MyD88 (Figure 3A). MIP-2 is a mouse analog of human chemokine IL-8, and Ox-PAPC–induced expression of IL-8 by endothelial cells has also been demonstrated to be dependent on TLR4, although another glycosylphosphatidylinositol-anchored receptor, not CD14, was suggested to mediate this Ox-PAPC signaling in endothelial cells.13 At this point, a mechanism for the robust stimulation of MCP-1 secretion by mmLDL (Figure 4) remains elusive.
Compared with the cytokine secretion induced by LPS, cytokine secretion induced by mmLDL was rather weak, with the exception of MCP-1. This in vitro cytokine expression pattern might better correspond to what one would find in a low-grade but chronic inflammatory state in vivo, as opposed to an acute, LPS-type, inflammatory response. Strong mmLDL-induced expression of the chemoattractants MCP-1 and MIP-2 also supports the known central role of such chemoattractants in murine atherosclerosis, in which leukocyte recruitment to the artery wall is a rate-limiting event.41
The experiments in this and our previous studies6,24 suggest that PI3K activation is a central signaling event in the macrophage stimulation by mmLDL. Indeed, mmLDL induced by far the strongest PI3K activation and Akt phosphorylation than any other ligand tested, including LPS (Figure 2A). A putative PI3K p85-binding motif has been found in MyD88, and LPS stimulation results in formation of a PI3K–MyD88 complex.42 PI3K in endothelial cell has also been shown to mediate Ox-PAPC–induced monocyte adhesion, although it is not clear whether this is a TLR4-mediated process.43 In our experiments, mmLDL-induced Akt phosphorylation (and thus, PI3K activation) did not depend on the presence of TLR4 in macrophages, suggesting that mmLDL activates PI3K via a TLR4-independent signaling pathway.
Our initial hypothesis was that the robust PI3K activation by mmLDL would interfere with NF-B transactivation because of Akt-dependent GSK-3? phosphorylation (inactivation).31 However, PI3K inhibition resulted in blocking MIP-2 and TNF- secretion (Figure II). Indeed, in addition to GSK-3? phosphorylation, Akt has also been demonstrated to interact with and activate IB kinase, thereby leading to NF-B–dependent gene expression.44 In addition, PI3K activates the ERK1/2 pathway,45 which stimulates serum response element–dependent gene expression. Coordinate activation by PI3K of Akt and TLR4-dependent ERK1/2 pathways might constitute the mechanisms of mmLDL-induced MIP-2 secretion.
In conclusion, the present study demonstrates that mmLDL activation of macrophages induces a unique cross-talk between TLR4- and PI3K-signaling cascades. Because this makes the mmLDL stimulation proinflammatory and complimentary to the LPS-induced signaling, the results of the present study provide new insights into the inflammatory origins of atherosclerosis.
Acknowledgments
This work was supported by the UC Tobacco-Related Disease grant 12KT-0104 (Y.I.M.), a scholars grant from Medicine Education and Research Foundation (Y.I.M.), National Institutes of Health grants HL56989 to La Jolla SCOR in Molecular Medicine and Atherosclerosis (J.L.W., Y.I.M, A.B. and S.B.), PO1GM37696 (S.V.), DK62025 (D.S.W.), and by the Medical Research Service of the Department of Veterans Affairs (S.V.).
References
Akira S. Toll-like receptor signaling. J Biol Chem. 2003; 278: 38105–38108.
Johnson GB, Riggs BL, Platt JL. A genetic basis for the "Adonis" phenotype of low adiposity and strong bones. FASEB J. 2004; 18: 1282–1284.
Binder CJ, Chang MK, Shaw PX, Miller YI, Hartvigsen K, Dewan A, Witztum JL. Innate and acquired immunity in atherogenesis. Nat Med. 2002; 8: 1218–1226.
Miller YI, Chang MK, Binder CJ, Shaw PX, Witztum JL. Oxidized low-density lipoprotein and innate immune receptors. Curr Opin Lipidol. 2003; 14: 437–445.
Libby P, Ridker PM, Maseri A. Inflammation and atherosclerosis. Circulation. 2002; 105: 1135.
Miller YI, Viriyakosol S, Binder CJ, Feramisco JR, Kirkland TN, Witztum JL. Minimally modified LDL binds to CD14, induces macrophage spreading via TLR4/MD-2, and inhibits phagocytosis of apoptotic cells. J Biol Chem. 2003; 278: 1561–1568.
Bjorkbacka H, Kunjathoor VV, Moore KJ, Koehn S, Ordija CM, Lee MA, Means T, Halmen K, Luster AD, Golenbock DT, Freeman MW. Reduced atherosclerosis in MyD88-null mice links elevated serum cholesterol levels to activation of innate immunity signaling pathways. Nat Med. 2004; 10: 416–421.
Michelsen KS, Wong MH, Shah PK, Zhang W, Yano J, Doherty TM, Akira S, Rajavashisth TB, Arditi M. Lack of Toll-like receptor 4 or myeloid differentiation factor 88 reduces atherosclerosis and alters plaque phenotype in mice deficient in apolipoprotein E. Proc Natl Acad Sci U S A. 2004; 101: 10679–10684.
Kiechl S, Lorenz E, Reindl M, Wiedermann CJ, Oberhollenzer F, Bonora E, Willeit J, Schwartz DA. Toll-like receptor 4 polymorphisms and atherogenesis. N Engl J Med. 2002; 347: 185–192.
Ameziane N, Beillat T, Verpillat P, Chollet-Martin S, Aumont MC, Seknadji P, Lamotte M, Lebret D, Ollivier V, de Prost D. Association of the Toll-like receptor 4 gene Asp299Gly polymorphism with acute coronary events. Arterioscler Thromb Vasc Biol. 2003; 23: e61–e64.
Navab M, Ananthramaiah GM, Reddy ST, Van Lenten BJ, Ansell BJ, Fonarow GC, Vahabzadeh K, Hama S, Hough G, Kamranpour N, Berliner JA, Lusis AJ, Fogelman AM. Thematic review series: the pathogenesis of atherosclerosis the oxidation hypothesis of atherogenesis: the role of oxidized phospholipids and HDL. J Lipid Res. 2004; 45: 993–1007.
Berliner JA, Subbanagounder G, Leitinger N, Watson AD, Vora D. Evidence for a role of phospholipid oxidation products in atherogenesis. Trends Cardiovasc Med. 2001; 11: 142–147.
Walton KA, Hsieh X, Gharavi N, Wang S, Wang G, Yeh M, Cole AL, Berliner JA. Receptors involved in the Ox-PAPC-mediated synthesis of IL-8: A role for toll-like receptor 4 and a GPI anchored protein. J Biol Chem. 2003; 278: 29661–29666.
Leitinger N. Oxidized phospholipids as modulators of inflammation in atherosclerosis. Curr Opin Lipidol. 2003; 14: 421–430.
Bochkov VN, Kadl A, Huber J, Gruber F, Binder BR, Leitinger N. Protective role of phospholipid oxidation products in endotoxin-induced tissue damage. Nature. 2002; 419: 77–81.
Walton KA, Cole AL, Yeh M, Subbanagounder G, Krutzik SR, Modlin RL, Lucas RM, Nakai J, Smart EJ, Vora DK, Berliner JA. Specific phospholipid oxidation products inhibit ligand activation of Toll-like receptors 4 and 2. Arterioscler Thromb Vasc Biol. 2003; 23: 1197–1203.
Hoshino K, Takeuchi O, Kawai T, Sanjo H, Ogawa T, Takeda Y, Takeda K, Akira S. Cutting edge: Toll-like receptor 4 (TLR4)-deficient mice are hyporesponsive to lipopolysaccharide: evidence for TLR4 as the Lps gene product. J Immunol. 1999; 162: 3749–3752.
Kawai T, Adachi O, Ogawa T, Takeda K, Akira S. Unresponsiveness of MyD88-deficient mice to endotoxin. Immunity. 1999; 11: 115–122.
Benz DJ, Mol M, Ezaki M, Mori-Ito N, Zelaan I, Miyanohara A, Friedmann T, Parthasarathy S, Steinberg D, Witztum JL. Enhanced levels of lipoperoxides in low-density lipoprotein incubated with murine fibroblast expressing high levels of human 15-lipoxygenase. J Biol Chem. 1995; 270: 5191–5197.
Viriyakosol S, Tobias PS, Kitchens RL, Kirkland TN. MD-2 binds to bacterial lipopolysaccharide. J Biol Chem. 2001; 276: 38044–38051.
Havel RJ, Bragdon JH, Eder HA. The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J Clin Invest. 1955; 34: 1345–1353.
Ezaki M, Witztum JL, Steinberg D. Lipoperoxides in LDL incubated with fibroblasts that overexpress 15-lipoxygenase. J Lipid Res. 1995; 36: 1996–2004.
Sigari F, Lee C, Witztum JL, Reaven PD. Fibroblasts that overexpress 15-lipoxygenase generate bioactive and minimally modified LDL. Arterioscler Thromb Vasc Biol. 1997; 17: 3639–3645.
Miller YI, Worrall DS, Funk CD, Feramisco JR, Witztum JL. Actin polymerization in macrophages in response to oxidized LDL and apoptotic cells: role of 12/15-lipoxygenase and phosphoinositide 3-kinase. Mol Biol Cell. 2003; 14: 4196–4206.
Kruszynska YT, Worrall DS, Ofrecio J, Frias JP, Macaraeg G, Olefsky JM. Fatty acid-induced insulin resistance: decreased muscle PI3K activation but unchanged Akt phosphorylation. J Clin Endocrinol Metab. 2002; 87: 226–234.
Binder CJ, H?rkk? S, Dewan A, Chang MK, Kieu EP, Goodyear CS, Shaw PX, Palinski W, Witztum JL, Silverman GJ. Pneumococcal vaccination decreases atherosclerotic lesion formation: molecular mimicry between Streptococcus pneumoniae and oxidized LDL. Nat Med. 2003; 9: 736–743.
Boullier A, Gillotte KL, H?rkk? S, Green SR, Friedman P, Dennis EA, Witztum JL, Steinberg D, Quehenberger O. The binding of oxidized low-density lipoprotein to mouse CD36 is mediated in part by oxidized phospholipids that are associated with both the lipid and protein moieties of the lipoprotein. J Biol Chem. 2000; 275: 9163–9169.
Podrez EA, Poliakov E, Shen Z, Zhang R, Deng Y, Sun M, Finton PJ, Shan L, Gugiu B, Fox PL, Hoff HF, Salomon RG, Hazen SL. Identification of a novel family of oxidized phospholipids that serve as ligands for the macrophage scavenger receptor CD36. J Biol Chem. 2002; 277: 38503–38516.
Subbanagounder G, Wong JW, Lee H, Faull KF, Miller E, Witztum JL, Berliner JA. Epoxyisoprostane and epoxycyclopentenone phospholipids regulate monocyte chemotactic protein-1 and interleukin-8 synthesis. Formation of these oxidized phospholipids in response to interleukin-1beta. J Biol Chem. 2002; 277: 7271–7281.
Renard P, Ernest I, Houbion A, Art M, Le Calvez H, Raes M, Remacle J. Development of a sensitive multi-well colorimetric assay for active NFkappaB. Nucleic Acids Res. 2001; 29: E21.
Guha M, Mackman N. The phosphatidylinositol 3-kinase-Akt pathway limits lipopolysaccharide activation of signaling pathways and expression of inflammatory mediators in human monocytic cells. J Biol Chem. 2002; 277: 32124–32132.
Fitzgerald KA, Rowe DC, Barnes BJ, Caffrey DR, Visintin A, Latz E, Monks B, Pitha PM, Golenbock DT. LPS-TLR4 Signaling to IRF-3/7 and NF-{kappa}B involves the Toll adapters TRAM and TRIF. J Exp Med. 2003; 198: 1043–1055.
Rankin SM, Parthasarathy S, Steinberg D. Evidence for a dominant role of lipoxygenase(s) in the oxidation of LDL by mouse peritoneal macrophages. J Lipid Res. 1991; 32: 449–456.
Rydberg EK, Krettek A, Ullstrom C, Ekstrom K, Svensson PA, Carlsson LMS, Jonsson-Rylander AC, Hansson GI, McPheat W, Wiklund O, Ohlsson BG, Hulten LM. Hypoxia Increases LDL Oxidation and Expression of 15-Lipoxygenase-2 in Human Macrophages. Arterioscler Thromb Vasc Biol. 2004; 24: 2040–2045.
Cyrus T, Pratico D, Zhao L, Witztum JL, Rader DJ, Rokach J, FitzGerald GA, Funk CD. Absence of 12/15-lipoxygenase expression decreases lipid peroxidation and atherogenesis in apolipoprotein E–deficient mice. Circulation. 2001; 103: 2277–2282.
George J, Afek A, Shaish A, Levkovitz H, Bloom N, Cyrus T, Zhao L, Funk CD, Sigal E, Harats D. 12/15-Lipoxygenase gene disruption attenuates atherogenesis in LDL receptor-deficient mice. Circulation. 2001; 104: 1646–1650.
Huo Y, Zhao L, Hyman MC, Shashkin P, Harry BL, Burcin T, Forlow SB, Stark MA, Smith DF, Clarke S, Srinivasan S, Hedrick CC, Pratico D, Witztum JL, Nadler JL, Funk CD, Ley K. Critical role of macrophage 12/15-lipoxygenase for atherosclerosis in apolipoprotein E–deficient mice. Circulation. 2004; 110: 2024–2031.
Yl?-Herttuala S, Rosenfeld ME, Parthasarathy S, Sigal E, Sarkioja T, Witztum JL, Steinberg D. Gene expression in macrophage-rich human atherosclerotic lesions. 15-lipoxygenase and acetyl low-density lipoprotein receptor messenger RNA colocalize with oxidation specific lipid-protein adducts. J Clin Invest. 1991; 87: 1146–1152.
Heinecke JW. Oxidative stress: new approaches to diagnosis and prognosis in atherosclerosis. Am J Cardiol. 2003; 91: 12A–16A.
Zhao L, Funk CD. Lipoxygenase pathways in atherogenesis. Trends in Cardiovasc Med. 2004; 14: 191–195.
Boring L, Gosling J, Cleary M, Charo IF. Decreased lesion formation in CCR2–/– mice reveals a role for chemokines in the initiation of atherosclerosis. Nature. 1998; 394: 894–897.
Ojaniemi M, Glumoff V, Harju K, Liljeroos M, Vuori K, Hallman M. Phosphatidylinositol 3-kinase is involved in Toll-like receptor 4-mediated cytokine expression in mouse macrophages. Eur J Immunol. 2003; 33: 597–605.
Cole AL, Subbanagounder G, Mukhopadhyay S, Berliner JA, Vora DK. Oxidized phospholipid-induced endothelial cell/monocyte interaction is mediated by a cAMP-dependent R-Ras/PI3-kinase pathway. Arterioscler Thromb Vasc Biol. 2003; 23: 1384–1390.
Ozes ON, Mayo LD, Gustin JA, Pfeffer SR, Pfeffer LM, Donner DB. NF-kappaB activation by tumor necrosis factor requires the Akt serine-threonine kinase. Nature. 1999; 401: 82–85.
Gingery A, Bradley E, Shaw A, Oursler MJ. Phosphatidylinositol 3-kinase coordinately activates the MEK/ERK and AKT/NFkappaB pathways to maintain osteoclast survival. J Cell Biochem. 2003; 89: 165–179.(Yury I. Miller; Suganya V)
Correspondence to Yury I. Miller, MD, PhD, University of California, San Diego, 9500 Gilman Dr, La Jolla, CA 92093-0682. E-mail yumiller@ucsd.edu
Abstract
Objective— Innate immune responses to oxidized low-density lipoprotein LDL (LDL) regulate the development of atherosclerosis. We demonstrated previously that an early form of oxidized LDL, minimally modified LDL (mmLDL), triggers cytoskeletal rearrangements in macrophages via CD14 and Toll-like receptor 4 (TLR4)/MD-2. Because lipopolysaccharide (LPS) activation of TLR4 leads to proinflammatory gene expression, in this study, we asked whether mmLDL also induced proinflammatory signaling.
Methods and Results— We studied cytokine secretion and signaling in J774 and primary peritoneal macrophages stimulated with mmLDL, which was prepared by incubating LDL with cells expressing human 15-lipoxygenase. MmLDL stimulated robust phosphoinositide 3-kinase (PI3K) activation, and Akt and extracellular signal-regulated kinase 1/2 (ERK1/2) phosphorylation, which exceeded that induced by LPS. On the other hand, although mmLDL induced nuclear factor B (NF-B) p65 translocation to the nucleus, there was no detectable NF-B activation. However, mmLDL induced early mRNA and protein expression of the cytokines MIP-2, MCP-1, tumor necrosis factor-, and interleukin-6. Chemokine MIP-2 but not MCP-1 secretion depended on TLR4/MyD88, ERK1/2, and PI3K signaling. In turn, TLR4 regulated phosphorylation of ERK1/2 but not of Akt, suggesting that mmLDL-induced PI3K activation is TLR4 independent.
Conclusions— In macrophages, mmLDL activates TLR4-dependent and -independent signaling pathways, resulting in secretion of proinflammatory cytokines. These results provide new insights into the inflammatory origins of atherosclerosis.
Minimally oxidized LDL (mmLDL) stimulated PI3K activation, Akt, and ERK1/2 phosphorylation, and induced mRNA and protein expression of MIP-2, MCP-1, TNF-, and IL-6 in macrophages. MIP-2 but not MCP-1 secretion depended on TLR4/MyD88, ERK1/2, and PI3K signaling. Thus, mmLDL activates TLR4-dependent and independent proinflammatory signaling in macrophages.
Key Words: macrophage ? oxidized low-density lipoprotein ? toll-like receptor 4 ? phosphoinositide 3-kinase ? cytokines
Introduction
Toll-like receptor 4 (TLR4) is a key signaling receptor of innate immunity because it senses the presence of bacterial lipopolysaccharide (LPS) associated with CD14 or MD-2 and initiates a proinflammatory signaling cascade. The classical mechanism by which LPS-induced TLR4 signaling occurs includes activation of mitogen-activated protein (MAP) kinases and nuclear factor B (NF-B), resulting in secretion of proinflammatory cytokines such as tumor necrosis factor- (TNF-), interleukin-1? (IL-1?), IL-6, and others.1 Moreover, a fundamental biological importance of TLR4 was demonstrated recently by the discovery that TLR4-deficient mice of various strains have less body fat and stronger bones than wild-type mice.2
See page 1085
Atherosclerosis is a chronic inflammatory disease of the vascular wall, and various forms of oxidized low-density lipoprotein (LDL) are key proinflammatory agents present during all stages of atherogenesis.3–5 We demonstrated recently that a very early form of oxidized LDL, minimally modified LDL (mmLDL), binds to the LPS receptor CD14 and stimulates macrophage spreading and actin polymerization.6 Furthermore, the mmLDL-induced macrophage spreading inhibits phagocytosis of apoptotic cells but promotes uptake of oxidized LDL by macrophages, events that should be proatherogenic. Remarkably, the mmLDL-induced cytoskeletal changes depend on TLR4/MD-2.6 Indeed, deficiency of TLR4 or myeloid differentiation factor-88 (MyD88), an adaptor molecule downstream from TLR4 and other Toll-like receptors, significantly reduces diet-induced atherosclerosis in the apolipoprotein E–/– mouse model.7,8 Moreover, human epidemiological data demonstrate that an Asp299Gly TLR4 polymorphism, which attenuates receptor signaling, is associated with a decreased risk of atherosclerosis and acute coronary events.9,10 Because mmLDL represents an early form of oxidized LDL found in vivo in the artery wall,11 and because the mmLDL-induced changes in cytoskeleton depend on TLR4, it is potentially highly relevant and important to elucidate the mechanisms by which mmLDL interactions with TLR4 affect cellular signaling and thus may impact atherogenesis.
In addition to effects on macrophage spreading and phagocytosis, we wondered whether incubation of mmLDL with macrophages would also result in the activation of MAP kinases and NF-B and in cytokine production. Berliner et al have suggested that oxidation products of 1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphocholine (Ox-PAPC), such as 1-palmitoyl-2-(5-oxovaleroyl)-sn-glycero-3-phosphocholine (POVPC), found in oxidized LDL and in mmLDL, induce secretion of chemokines IL-8 and monocyte chemoattractant protein-1 (MCP-1) by human endothelial cells and monocytes.12,13 They have also shown that the endothelial IL-8 production stimulated by Ox-PAPC depends on TLR4.13 Ox-PAPC activates MAP kinases but not NF-B in endothelial cells.14 Paradoxically, Ox-PAPC has been reported to inhibit LPS-induced activation in endothelial cells, and when administered in vivo to mice, to protect from LPS-induced tissue damage.15,16
In this study, we analyze intracellular events and cytokine production in mouse macrophages initiated by mmLDL stimulation (in comparison with native LDL [nLDL] and oxidized LDL). We found that part of but not all macrophage responses to mmLDL depended on TLR4/MyD88 and that the mmLDL-induced signaling differed from the LPS-induced signaling. These findings provide new insights into the potential role of early forms of oxidized LDL in modulating inflammatory events in atherosclerosis.
Methods
An extended Methods section is available online at http://atvb.ahajournals.org.
Cells
Resident peritoneal macrophages were harvested from 8- to 10-week-old female mice of wild-type (C57BL/6), TLR4–/–, or MyD88–/–. The TLR4–/– and MyD88–/– mice were kindly provided by Dr Shizuo Akira from Research Institute for Microbial Diseases, Osaka University, Japan.17,18 (We also thank Drs Eyal Raz and Peter Tobias for providing knockout mice.) Murine macrophage-like cell line J774A.1, a murine fibroblast cell line stably overexpressing human 15-lipoxygenase (15-LO)19 and Chinese hamster ovary (CHO) cell lines, constitutively expressing an NF-B reporter together with TLR4 or TLR4/MD-2,20 were maintained as described previously.6
LDL Isolation and Modification
LDL was isolated from plasma of normolipidemic donors by sequential ultracentrifugation.21 Contamination of nLDL and modified LDL preparations by endotoxin (LPS) was assessed with a LAL kit (BioWhittaker). LDL preparations with LPS >50 pg/mg protein were discarded. For extensive oxidation, the LDL (1 mg/mL) was incubated with 10 μmol/L CuSO4 for 18 hours at 37°C.
Minimally Modified (Oxidized) LDL
To produce mmLDL, we incubated 50 μg/mL LDL in serum-free DMEM for 18 hours with a murine fibroblast cell line overexpressing 15-LO.19 We documented previously that this procedure generates mmLDL (ie, it binds to nLDL receptors but not to scavenger receptors).6,19,22–24 MmLDL contains early lipid peroxidation products,22,24 but it does not contain any measurable thiobarbituric acid reactive substances or EO6-reactive phospholipid oxidation products above that of nLDL.6 The mmLDL modification appeared to be very reproducible, and a successful modification was documented in a biological assay in which mmLDL induced spreading of J774 macrophages in cell culture.6,24
Cell Signaling and Cytokine Assays
Four different assays were used to assess NF-B translocation and activation: p65 nuclear translocation, p65-DNA binding, NF-B reporter gene expression, and p65 phosphorylation (please see supplemental material for details).
Phosphorylation of signaling proteins was assessed by Western blot. Phosphoinositide 3-kinase (PI3K) activity was measured in cell lysates and immunoprecipitated with an antiphosphotyrosine antibody, as described25 in a reaction with phosphatidylinositol in the presence of [-32P]ATP.
Real-time polymerase chain reaction (PCR) of reverse-transcribed total RNA was performed using an Applied Biosystems ABI Prism 7700 Sequence Detection System, as performed in Genomic Core, Center for AIDS Research, University of California, San Diego.
Cytokine ELISAs were performed according to standard antibody manufacturer protocols, except that chemiluminescent detection was used, as described previously from our laboratory.26
Results
MmLDL-Induced Cell Signaling
Our model of early forms of oxidized LDL, mmLDL, which is produced by incubation with 15-LO–expressing fibroblasts in DMEM media, contains mostly early lipid oxidation products, such as fatty acid hydroperoxides and hydroxides.22,24 MmLDL does not bind EO624 and thus likely has few if any of the bioactive oxidized phospholipids found in Ox-PAPC. In contrast, Ox-PAPC and the lipid moieties of oxidized LDL bind CD36 and EO6.27–29 Because of these differences and because mmLDL but not Ox-PAPC binds CD14,6,13 we investigated intracellular signaling pathways, TLR4-dependent and -independent, stimulated by mmLDL.
We examined NF-B activation in mmLDL-stimulated macrophages by 4 independent techniques. First, we analyzed NF-B p65 translocation from the cytosol to the nucleus of J774 cells by immunocytochemistry. As expected, nuclear localization of p65 was observed in the LPS-treated macrophages (Figure 1A). Remarkably, a majority of the cells treated with mmLDL also showed clear nuclear localization of p65. In contrast, p65 in control macrophages and in the cells treated with either nLDL or oxidized LDL predominantly located in the cytosol. As we reported previously, mmLDL and, to a lesser degree, LPS, induced macrophage spreading, which was also observed in this experiment (note red staining for F-actin in Figure 1A). However, nuclear p65 localization was not a function of spreading, because in the relatively spread oxidized LDL–treated cells, p65 was still cytosolic (see Figure 4).
Figure 1. MmLDL induces p65 nuclear translocation but not NF-B activation. A, J774 cells were activated with 50 μg/mL nLDL, mmLDL, or oxidized LDL, or 10 ng/mL LPS for 15 minutes, fixed, and stained for p65 (green), nucleic acid (blue), and F-actin (red). Each treatment is presented by 2 images of the same cell. On the left, nucleus (blue) is contrasted by actin cytoskeleton (red), and on the right, p65 cellular localization is shown in green. B, Nuclear extracts were made from J774 cells after 15 minutes of application of 50 μg/mL nLDL, mmLDL, or oxidized LDL, or 100 ng/mL LPS. Binding of NF-B p65 to a specific DNA sequence was assessed by a TransAM ELISA assay, which yields quantitative results that parallel those of a gel-shift assay (see Methods). C, CHO cell lines bearing inducible membrane CD25 under the transcriptional control of a human E-selectin promoter containing NF-B binding sites, and stably transfected with either TLR4 or TLR4/MD-220 were treated with 50 μg/mL mmLDL or 100 ng/mL LPS for 24 hours, and at the end of incubation, analyzed for CD25 surface expression by fluorescence-activated cell sorter (FACS). Note that LPS only induces CD25 in CHO cells transfected with TLR4/MD-2.
Figure 4. Cytokine protein secretion by wild-type (WT), TLR4–/– and MyD88–/– peritoneal macrophages in response to mmLDL. Resident peritoneal macrophages were stimulated for 6 hours with 50 μg/mL mmLDL or media alone. At the end of incubation, culture media were collected and analyzed by ELISA for the presence of MIP-2 (A), MCP-1 (B; note the Y scale break), TNF- (C), and IL-6 (D). *P<0.005. KO indicates knockout.
NF-B nuclear localization is necessary but not sufficient for its activity. Therefore, we directly tested p65 binding to DNA using an ELISA-based immunoassay, which is reported to be more sensitive than traditional gel-shift assays.30 However, no p65-DNA binding was observed in the nuclear extracts of the cells treated with mmLDL, whereas the LPS-treated cells showed the expected p65-DNA binding (Figure 1B). To confirm this result, we performed a third assay in which CHO cells, stably transfected with an NF-B reporter (surface CD25 expression; see Methods) and either TLR4 or TLR4/MD-2, were treated with mmLDL or LPS. In accordance with the p65-DNA binding results, mmLDL failed to stimulate any increase in NF-B activity, as measured by CD25 surface expression, whereas LPS did (Figure 1C). In yet another assay, mmLDL did not stimulate p65 phosphorylation (Figure 2B). Thus, the mmLDL-induced NF-B nuclear translocation was not associated with detectable NF-B activation.
Figure 2. PI3K activation and protein phosphorylation in response to mmLDL and LPS. A, J774 cells were activated with 100 ng/mL LPS or 50 μg/mL mmLDL or oxidized LDL for 15 minutes. At the end of incubation, cell lysates were prepared to analyze PI3K activity and protein phosphorylation as described in Methods. B, J774 cells were incubated with 100 ng/mL LPS or 50 μg/mL mmLDL for indicated times, and the cell lysates were analyzed for protein phosphorylation. Because prolonged treatments of up to 1 hour could have affected protein expression, here we also analyzed total p65 and Akt proteins. C, Resident peritoneal macrophages from wild-type or TLR4–/– mice were incubated with or without 50 μg/mL mmLDL for 15 minutes, and the cell lysates were analyzed for protein phosphorylation and total protein expression. D, J774 cells were pretreated for 30 minutes with 100 nmol/L wortmannin (Wortm) or 10 μmol/L LY294002, and then incubated for additional 15 minutes with 50 μg/mL mmLDL.
As we reported previously, mmLDL induces strong PI3K activation in mouse macrophages.6,24 In the current study, we found that the degree of activation of PI3K by mmLDL was even stronger than that achieved by 100 ng/mL of LPS (Figure 2A). Accordingly, the PI3K target Akt was phosphorylated extensively. The mmLDL-induced Akt phosphorylation was 6- to 8-fold stronger than LPS-induced phosphorylation, and oxidized LDL induced no Akt phosphorylation. Further downstream, Akt phosphorylated GSK-3? (Figure 2A). MmLDL stimulated profound extracellular signal-regulated kinase 1/2 (ERK1/2) phosphorylation, whereas the LPS effect on ERK1/2 was rather modest (Figure 2B).
To further compare mmLDL and LPS stimulation, we analyzed the time course of phosphorylation of relevant signaling proteins in J774 macrophages (Figure 2B). The mmLDL-stimulated phosphorylation of ERK1/2, Akt, and, to some extent, of JNK, sustained for the whole 1 hour of incubation. There was no detectable phosphorylation of p65 by mmLDL, confirming the lack of NF-B activation. In contrast, LPS induced strong p65 phosphorylation, which peaked at 10 minutes, and evident phosphorylation of MAP kinases, ERK1/2, p38, and JNK, which peaked at 30 minutes, but no noticeable Akt phosphorylation (Figure 2B).
ERK1/2 but not PI3K Activation Depends on TLR4
To identify which of the above signaling events induced by mmLDL depend on TLR4, we tested phosphorylation of signaling proteins in the lysates of macrophages from wild-type and TLR4–/– mice. ERK1/2 phosphorylation was clearly dependent on TLR4. On the other hand, mmLDL-induced phosphorylation of either Ser473 or Thr308 of Akt was similar in wild-type and in TLR4–/– macrophages (Figure 2C), suggesting that PI3K and downstream Akt signaling are TLR4 independent.
PI3K Activation by mmLDL Does Not Silence NF-B
Next, we tested the hypothesis that a strong PI3K activation by mmLDL might be responsible for the inhibition of NF-B activation. Indeed, Guha and Mackman31 proposed that the LPS activation of PI3K/Akt in human monocytic cells downregulates NF-B transactivation via several mechanisms, including inactivation of GSK-3? via Akt-dependent phosphorylation. As expected, inhibition of PI3K by either wortmannin or LY294002 abolished mmLDL-induced Akt and GSK-3? phosphorylation (Figure 2D). It appeared that ERK phosphorylation also depended to some extent on PI3K activity (Figure 2D). However, preserving GSK-3? in its active (nonphosphorylated) state during the stimulation with mmLDL did not result in any detectable increase in p65 phosphorylation (Figure 2D). Further, wortmannin and LY294002 treatments failed to increase p65-DNA binding in the nuclear extracts of mmLDL-treated J774 cells (data not shown).
MmLDL-Induced Cytokine Expression
As shown in Figures 1 and 2, we observed that mmLDL induced p65 nuclear translocation but were unable to detect NF-B activation. In addition, mmLDL initiated PI3K and ERK1/2 signaling. Thus, next we examined whether these signaling events resulted in macrophage expression of proinflammatory cytokines. We tested 2 of the most common LPS-stimulated genes, TNF- and IL-6, as well as 2 chemokines, MIP-2 (a mouse analog of human IL-8) and MCP-1, which are stimulated not only by LPS but also by Ox-PAPC.12 In addition, we measured interferon-? (IFN-?) because its expression is Toll/IL-1R domain-containing adapter inducing IFN-? (TRIF) dependent and MyD88 independent.1,32
MmLDL stimulation of J774 macrophages produced early, within 30 to 60 minutes, expression of mRNA for macrophage inflammatory protein-2 (MIP-2), MCP-1, TNF-, and IL-6. This was not seen with nLDL or oxidized LDL (Figure 3). The mRNA expression for these cytokines decreased to basal levels by 3 hours after stimulation. There was no significant difference in the IFN-? expression induced by mmLDL, nLDL, or oxidized LDL. These data are consistent with mmLDL stimulation of the MyD88 but not of the TRIF pathway.1,32 Although mmLDL induced a significant, up to 25-fold, increase in cytokine mRNAs, LPS (10 ng/mL) induced much stronger mRNA expression of MIP-2, TNF-, and IL-6, but not of MCP-1 (Figure 3A through 3F). In addition, the LPS-induced IL-6 expression did not peak at 1 hour as it did with mmLDL stimulation, but increased steadily for 6 hours (Figure 1F).
Figure 3. Time course of cytokine mRNA expression in J774 cells. J774 macrophages were stimulated with 50 μg/mL of nLDL, mmLDL, or oxidized LDL (A through E), or 10 ng/mL LPS (F). At the indicated time points, the reactions were stopped and total RNA was isolated, reverse-transcribed, and subjected to real-time PCR with the primers specific for cytokines and GAPDH. The data are normalized to the GAPDH levels in each sample and presented as a fold increase compared with the 0 time point. Note that A through E show individual cytokines stimulated by various modified LDL, whereas F contains data for all measured cytokines stimulated by LPS.
MIP-2, MCP-1, and TNF- protein secretion, as measured in the J774 culture media by ELISA after a 6-hour incubation with nLDL, mmLDL, or oxidized LDL, was consistent with the mRNA results, except that there was no IL-6 protein secretion induced by mmLDL (supplemental Figure I, available online at http://atvb.ahajournals.org), even after 24 hours of incubation (data not shown). In addition to the data presented, we also measured IL-12 and IL-1? production by J774 cells but were unable to detect any mmLDL stimulation of these cytokines (data not shown).
MmLDL-Induced MIP-2 Secretion Depends on TLR4/MyD88 and PI3K
The cytokine production induced by mmLDL in wild-type resident peritoneal macrophages was similar to that in J774 cells (Figure 4). Of note is a dramatic increase (up to 10 000 pg/mL) in the MCP-1 production by mouse macrophages induced by mmLDL (Figure 4B, scale break). To test the role of TLR4 pathway in mediating mmLDL-induced cytokine expression, we used macrophages harvested from TLR4–/– and MyD88–/– mice. We found that the secretion of only MIP-2 was clearly dependent on TLR4/MyD88 (Figure 4A). In contrast, MCP-1 induction was clearly independent of TLR4/MyD88, and surprisingly, the minute TNF- and IL-6 production induced by mmLDL was also independent of TLR4/MyD88 (Figure 4B through 4D).
Inhibition of PI3K by wortmannin or LY294002 completely blocked MIP-2 and TNF- secretion (supplemental Figure IIA and IIC, available online at http://atvb.ahajournals.org), although they did not significantly change MCP-1 production (supplemental Figure IIB). Similarly, U0126, through its inhibition of MAP kinase kinase and accordingly ERK1/2, also effectively blocked MIP-2 and TNF- (supplemental Figure IIA and IIC). These results suggest that PI3K and ERK positively regulate MIP-2 and TNF-.
Discussion
In the current study, we analyzed intracellular signaling events and cytokine secretion by mouse macrophages induced by mmLDL. MmLDL stimulated early mRNA and protein expression of MIP-2, MCP-1, TNF-, and IL-6 (Figures 3 and 4; Figure I), which were regulated differentially by TLR4/MyD88, PI3K/Akt, and ERK1/2 pathways (Figure 4; Figure II). In contrast to LPS, the mmLDL stimulation showed a very different pattern of intracellular signaling, with stronger PI3K/Akt and ERK1/2 activation and undetectable activation of NF-B (Figures 1 and 2).
There is now substantial evidence that human 15-LO (and its mouse analog 12/15-LO) can mediate the oxidation of LDL,19,22,33,34 that it has a proatherogenic role in murine models of atherosclerosis,35–37 and that it is expressed in the intima of human atherosclerotic lesions.34,38 Because in this study we aimed at modeling a likely early form of modified LDL found in the intima, we produced our mmLDL by incubating LDL with 15-LO–expressing cells. It is likely that early forms of oxidized LDL may be generated by other mechanisms as well, such as by myeloperoxidase28,39 or even 5-LO.40
Special attention was paid to ensure that all the LDL preparations used in this work were free from biologically relevant levels of LPS contamination. In all experiments, there was <2.5 pg/mL of LPS in the incubations used, concentrations below those reported to have effects in murine cells. Indeed, the fact that the mmLDL effects differed substantially from the effects of nLDL and oxidized LDL, even though every experiment was controlled by preparations made from the same LDL pool, further argues against a role of LPS contamination to explain the mmLDL effects. In addition, mmLDL induced very different patterns of cell signaling than did LPS (at 100 ng/mL). These arguments support our conclusion that the biological effects of mmLDL demonstrated in this article are not attributable to LPS contamination.
MmLDL interaction with CD14 and TLR4/MD-26 and the mmLDL-induced early cytokine expression in macrophages (Figures 3 and 4; Figure I) imply involvement of the TLR4/MyD88-dependent pathway. Indeed, using macrophages from TLR4–/– and MyD88–/– mice, we demonstrated that secretion of 1 of the 4 cytokines tested, MIP-2, was clearly dependent on TLR4/MyD88 (Figure 3A). MIP-2 is a mouse analog of human chemokine IL-8, and Ox-PAPC–induced expression of IL-8 by endothelial cells has also been demonstrated to be dependent on TLR4, although another glycosylphosphatidylinositol-anchored receptor, not CD14, was suggested to mediate this Ox-PAPC signaling in endothelial cells.13 At this point, a mechanism for the robust stimulation of MCP-1 secretion by mmLDL (Figure 4) remains elusive.
Compared with the cytokine secretion induced by LPS, cytokine secretion induced by mmLDL was rather weak, with the exception of MCP-1. This in vitro cytokine expression pattern might better correspond to what one would find in a low-grade but chronic inflammatory state in vivo, as opposed to an acute, LPS-type, inflammatory response. Strong mmLDL-induced expression of the chemoattractants MCP-1 and MIP-2 also supports the known central role of such chemoattractants in murine atherosclerosis, in which leukocyte recruitment to the artery wall is a rate-limiting event.41
The experiments in this and our previous studies6,24 suggest that PI3K activation is a central signaling event in the macrophage stimulation by mmLDL. Indeed, mmLDL induced by far the strongest PI3K activation and Akt phosphorylation than any other ligand tested, including LPS (Figure 2A). A putative PI3K p85-binding motif has been found in MyD88, and LPS stimulation results in formation of a PI3K–MyD88 complex.42 PI3K in endothelial cell has also been shown to mediate Ox-PAPC–induced monocyte adhesion, although it is not clear whether this is a TLR4-mediated process.43 In our experiments, mmLDL-induced Akt phosphorylation (and thus, PI3K activation) did not depend on the presence of TLR4 in macrophages, suggesting that mmLDL activates PI3K via a TLR4-independent signaling pathway.
Our initial hypothesis was that the robust PI3K activation by mmLDL would interfere with NF-B transactivation because of Akt-dependent GSK-3? phosphorylation (inactivation).31 However, PI3K inhibition resulted in blocking MIP-2 and TNF- secretion (Figure II). Indeed, in addition to GSK-3? phosphorylation, Akt has also been demonstrated to interact with and activate IB kinase, thereby leading to NF-B–dependent gene expression.44 In addition, PI3K activates the ERK1/2 pathway,45 which stimulates serum response element–dependent gene expression. Coordinate activation by PI3K of Akt and TLR4-dependent ERK1/2 pathways might constitute the mechanisms of mmLDL-induced MIP-2 secretion.
In conclusion, the present study demonstrates that mmLDL activation of macrophages induces a unique cross-talk between TLR4- and PI3K-signaling cascades. Because this makes the mmLDL stimulation proinflammatory and complimentary to the LPS-induced signaling, the results of the present study provide new insights into the inflammatory origins of atherosclerosis.
Acknowledgments
This work was supported by the UC Tobacco-Related Disease grant 12KT-0104 (Y.I.M.), a scholars grant from Medicine Education and Research Foundation (Y.I.M.), National Institutes of Health grants HL56989 to La Jolla SCOR in Molecular Medicine and Atherosclerosis (J.L.W., Y.I.M, A.B. and S.B.), PO1GM37696 (S.V.), DK62025 (D.S.W.), and by the Medical Research Service of the Department of Veterans Affairs (S.V.).
References
Akira S. Toll-like receptor signaling. J Biol Chem. 2003; 278: 38105–38108.
Johnson GB, Riggs BL, Platt JL. A genetic basis for the "Adonis" phenotype of low adiposity and strong bones. FASEB J. 2004; 18: 1282–1284.
Binder CJ, Chang MK, Shaw PX, Miller YI, Hartvigsen K, Dewan A, Witztum JL. Innate and acquired immunity in atherogenesis. Nat Med. 2002; 8: 1218–1226.
Miller YI, Chang MK, Binder CJ, Shaw PX, Witztum JL. Oxidized low-density lipoprotein and innate immune receptors. Curr Opin Lipidol. 2003; 14: 437–445.
Libby P, Ridker PM, Maseri A. Inflammation and atherosclerosis. Circulation. 2002; 105: 1135.
Miller YI, Viriyakosol S, Binder CJ, Feramisco JR, Kirkland TN, Witztum JL. Minimally modified LDL binds to CD14, induces macrophage spreading via TLR4/MD-2, and inhibits phagocytosis of apoptotic cells. J Biol Chem. 2003; 278: 1561–1568.
Bjorkbacka H, Kunjathoor VV, Moore KJ, Koehn S, Ordija CM, Lee MA, Means T, Halmen K, Luster AD, Golenbock DT, Freeman MW. Reduced atherosclerosis in MyD88-null mice links elevated serum cholesterol levels to activation of innate immunity signaling pathways. Nat Med. 2004; 10: 416–421.
Michelsen KS, Wong MH, Shah PK, Zhang W, Yano J, Doherty TM, Akira S, Rajavashisth TB, Arditi M. Lack of Toll-like receptor 4 or myeloid differentiation factor 88 reduces atherosclerosis and alters plaque phenotype in mice deficient in apolipoprotein E. Proc Natl Acad Sci U S A. 2004; 101: 10679–10684.
Kiechl S, Lorenz E, Reindl M, Wiedermann CJ, Oberhollenzer F, Bonora E, Willeit J, Schwartz DA. Toll-like receptor 4 polymorphisms and atherogenesis. N Engl J Med. 2002; 347: 185–192.
Ameziane N, Beillat T, Verpillat P, Chollet-Martin S, Aumont MC, Seknadji P, Lamotte M, Lebret D, Ollivier V, de Prost D. Association of the Toll-like receptor 4 gene Asp299Gly polymorphism with acute coronary events. Arterioscler Thromb Vasc Biol. 2003; 23: e61–e64.
Navab M, Ananthramaiah GM, Reddy ST, Van Lenten BJ, Ansell BJ, Fonarow GC, Vahabzadeh K, Hama S, Hough G, Kamranpour N, Berliner JA, Lusis AJ, Fogelman AM. Thematic review series: the pathogenesis of atherosclerosis the oxidation hypothesis of atherogenesis: the role of oxidized phospholipids and HDL. J Lipid Res. 2004; 45: 993–1007.
Berliner JA, Subbanagounder G, Leitinger N, Watson AD, Vora D. Evidence for a role of phospholipid oxidation products in atherogenesis. Trends Cardiovasc Med. 2001; 11: 142–147.
Walton KA, Hsieh X, Gharavi N, Wang S, Wang G, Yeh M, Cole AL, Berliner JA. Receptors involved in the Ox-PAPC-mediated synthesis of IL-8: A role for toll-like receptor 4 and a GPI anchored protein. J Biol Chem. 2003; 278: 29661–29666.
Leitinger N. Oxidized phospholipids as modulators of inflammation in atherosclerosis. Curr Opin Lipidol. 2003; 14: 421–430.
Bochkov VN, Kadl A, Huber J, Gruber F, Binder BR, Leitinger N. Protective role of phospholipid oxidation products in endotoxin-induced tissue damage. Nature. 2002; 419: 77–81.
Walton KA, Cole AL, Yeh M, Subbanagounder G, Krutzik SR, Modlin RL, Lucas RM, Nakai J, Smart EJ, Vora DK, Berliner JA. Specific phospholipid oxidation products inhibit ligand activation of Toll-like receptors 4 and 2. Arterioscler Thromb Vasc Biol. 2003; 23: 1197–1203.
Hoshino K, Takeuchi O, Kawai T, Sanjo H, Ogawa T, Takeda Y, Takeda K, Akira S. Cutting edge: Toll-like receptor 4 (TLR4)-deficient mice are hyporesponsive to lipopolysaccharide: evidence for TLR4 as the Lps gene product. J Immunol. 1999; 162: 3749–3752.
Kawai T, Adachi O, Ogawa T, Takeda K, Akira S. Unresponsiveness of MyD88-deficient mice to endotoxin. Immunity. 1999; 11: 115–122.
Benz DJ, Mol M, Ezaki M, Mori-Ito N, Zelaan I, Miyanohara A, Friedmann T, Parthasarathy S, Steinberg D, Witztum JL. Enhanced levels of lipoperoxides in low-density lipoprotein incubated with murine fibroblast expressing high levels of human 15-lipoxygenase. J Biol Chem. 1995; 270: 5191–5197.
Viriyakosol S, Tobias PS, Kitchens RL, Kirkland TN. MD-2 binds to bacterial lipopolysaccharide. J Biol Chem. 2001; 276: 38044–38051.
Havel RJ, Bragdon JH, Eder HA. The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J Clin Invest. 1955; 34: 1345–1353.
Ezaki M, Witztum JL, Steinberg D. Lipoperoxides in LDL incubated with fibroblasts that overexpress 15-lipoxygenase. J Lipid Res. 1995; 36: 1996–2004.
Sigari F, Lee C, Witztum JL, Reaven PD. Fibroblasts that overexpress 15-lipoxygenase generate bioactive and minimally modified LDL. Arterioscler Thromb Vasc Biol. 1997; 17: 3639–3645.
Miller YI, Worrall DS, Funk CD, Feramisco JR, Witztum JL. Actin polymerization in macrophages in response to oxidized LDL and apoptotic cells: role of 12/15-lipoxygenase and phosphoinositide 3-kinase. Mol Biol Cell. 2003; 14: 4196–4206.
Kruszynska YT, Worrall DS, Ofrecio J, Frias JP, Macaraeg G, Olefsky JM. Fatty acid-induced insulin resistance: decreased muscle PI3K activation but unchanged Akt phosphorylation. J Clin Endocrinol Metab. 2002; 87: 226–234.
Binder CJ, H?rkk? S, Dewan A, Chang MK, Kieu EP, Goodyear CS, Shaw PX, Palinski W, Witztum JL, Silverman GJ. Pneumococcal vaccination decreases atherosclerotic lesion formation: molecular mimicry between Streptococcus pneumoniae and oxidized LDL. Nat Med. 2003; 9: 736–743.
Boullier A, Gillotte KL, H?rkk? S, Green SR, Friedman P, Dennis EA, Witztum JL, Steinberg D, Quehenberger O. The binding of oxidized low-density lipoprotein to mouse CD36 is mediated in part by oxidized phospholipids that are associated with both the lipid and protein moieties of the lipoprotein. J Biol Chem. 2000; 275: 9163–9169.
Podrez EA, Poliakov E, Shen Z, Zhang R, Deng Y, Sun M, Finton PJ, Shan L, Gugiu B, Fox PL, Hoff HF, Salomon RG, Hazen SL. Identification of a novel family of oxidized phospholipids that serve as ligands for the macrophage scavenger receptor CD36. J Biol Chem. 2002; 277: 38503–38516.
Subbanagounder G, Wong JW, Lee H, Faull KF, Miller E, Witztum JL, Berliner JA. Epoxyisoprostane and epoxycyclopentenone phospholipids regulate monocyte chemotactic protein-1 and interleukin-8 synthesis. Formation of these oxidized phospholipids in response to interleukin-1beta. J Biol Chem. 2002; 277: 7271–7281.
Renard P, Ernest I, Houbion A, Art M, Le Calvez H, Raes M, Remacle J. Development of a sensitive multi-well colorimetric assay for active NFkappaB. Nucleic Acids Res. 2001; 29: E21.
Guha M, Mackman N. The phosphatidylinositol 3-kinase-Akt pathway limits lipopolysaccharide activation of signaling pathways and expression of inflammatory mediators in human monocytic cells. J Biol Chem. 2002; 277: 32124–32132.
Fitzgerald KA, Rowe DC, Barnes BJ, Caffrey DR, Visintin A, Latz E, Monks B, Pitha PM, Golenbock DT. LPS-TLR4 Signaling to IRF-3/7 and NF-{kappa}B involves the Toll adapters TRAM and TRIF. J Exp Med. 2003; 198: 1043–1055.
Rankin SM, Parthasarathy S, Steinberg D. Evidence for a dominant role of lipoxygenase(s) in the oxidation of LDL by mouse peritoneal macrophages. J Lipid Res. 1991; 32: 449–456.
Rydberg EK, Krettek A, Ullstrom C, Ekstrom K, Svensson PA, Carlsson LMS, Jonsson-Rylander AC, Hansson GI, McPheat W, Wiklund O, Ohlsson BG, Hulten LM. Hypoxia Increases LDL Oxidation and Expression of 15-Lipoxygenase-2 in Human Macrophages. Arterioscler Thromb Vasc Biol. 2004; 24: 2040–2045.
Cyrus T, Pratico D, Zhao L, Witztum JL, Rader DJ, Rokach J, FitzGerald GA, Funk CD. Absence of 12/15-lipoxygenase expression decreases lipid peroxidation and atherogenesis in apolipoprotein E–deficient mice. Circulation. 2001; 103: 2277–2282.
George J, Afek A, Shaish A, Levkovitz H, Bloom N, Cyrus T, Zhao L, Funk CD, Sigal E, Harats D. 12/15-Lipoxygenase gene disruption attenuates atherogenesis in LDL receptor-deficient mice. Circulation. 2001; 104: 1646–1650.
Huo Y, Zhao L, Hyman MC, Shashkin P, Harry BL, Burcin T, Forlow SB, Stark MA, Smith DF, Clarke S, Srinivasan S, Hedrick CC, Pratico D, Witztum JL, Nadler JL, Funk CD, Ley K. Critical role of macrophage 12/15-lipoxygenase for atherosclerosis in apolipoprotein E–deficient mice. Circulation. 2004; 110: 2024–2031.
Yl?-Herttuala S, Rosenfeld ME, Parthasarathy S, Sigal E, Sarkioja T, Witztum JL, Steinberg D. Gene expression in macrophage-rich human atherosclerotic lesions. 15-lipoxygenase and acetyl low-density lipoprotein receptor messenger RNA colocalize with oxidation specific lipid-protein adducts. J Clin Invest. 1991; 87: 1146–1152.
Heinecke JW. Oxidative stress: new approaches to diagnosis and prognosis in atherosclerosis. Am J Cardiol. 2003; 91: 12A–16A.
Zhao L, Funk CD. Lipoxygenase pathways in atherogenesis. Trends in Cardiovasc Med. 2004; 14: 191–195.
Boring L, Gosling J, Cleary M, Charo IF. Decreased lesion formation in CCR2–/– mice reveals a role for chemokines in the initiation of atherosclerosis. Nature. 1998; 394: 894–897.
Ojaniemi M, Glumoff V, Harju K, Liljeroos M, Vuori K, Hallman M. Phosphatidylinositol 3-kinase is involved in Toll-like receptor 4-mediated cytokine expression in mouse macrophages. Eur J Immunol. 2003; 33: 597–605.
Cole AL, Subbanagounder G, Mukhopadhyay S, Berliner JA, Vora DK. Oxidized phospholipid-induced endothelial cell/monocyte interaction is mediated by a cAMP-dependent R-Ras/PI3-kinase pathway. Arterioscler Thromb Vasc Biol. 2003; 23: 1384–1390.
Ozes ON, Mayo LD, Gustin JA, Pfeffer SR, Pfeffer LM, Donner DB. NF-kappaB activation by tumor necrosis factor requires the Akt serine-threonine kinase. Nature. 1999; 401: 82–85.
Gingery A, Bradley E, Shaw A, Oursler MJ. Phosphatidylinositol 3-kinase coordinately activates the MEK/ERK and AKT/NFkappaB pathways to maintain osteoclast survival. J Cell Biochem. 2003; 89: 165–179.(Yury I. Miller; Suganya V)