Desensitization of Platelet-Derived Growth Factor Receptor-β by Oxidized Lipids in Vascular Cells and Atherosclerotic Lesions
Prevention by Aldehyde Scavengers
The platelet-derived growth factor receptor-β (PDGFRβ) signaling pathway regulates smooth muscle cell (SMC) migration and proliferation and plays a role in the vascular wall response to injury. Oxidized low-density lipoprotein (oxLDL) in atherosclerotic lesions can activate the PDGFRβ pathway, but the long-term effects of oxLDL on PDGFRβ function are not well understood. We found that oxLDL induced a dual effect on PDGFRβ signaling. Initial activation of the PDGFR was followed by desensitization of the receptor. PDGFRβ desensitization was not attributable to PDGFRβ degradation or changes in localization to the caveolae but instead resulted from decreased PDGF binding and inhibition of PDGFRβ tyrosine kinase activity. This inhibition was associated with formation of (4HNE)– and acrolein–PDGFRβ adducts and was mimicked by preincubation of cells with 4HNE. These PDGFRβ adducts were also detected in aortae of apolipoprotein-deficient mice and hypercholesterolemic rabbits and in human carotid plaques. The aldehyde scavengers DNPH and Hydralazine prevented both oxLDL- and 4HNE-induced structural modification and PDGFRβ signaling dysfunction in cells and in vivo. OxLDL inhibition of PDGF signaling may contribute to defective SMC proliferation and decrease the stability of a vulnerable plaque.
During atherogenesis, the formation of fibroatheroma plaques involves a complex sequence of events, including endothelial activation, transendothelial migration of mononuclear cells, lipoprotein oxidation, lipid accumulation in macrophagic cells, smooth muscle cell (SMC) migration and proliferation, and local inflammatory response.1,2
Low-density lipoproteins (LDLs) are atherogenic3 after undergoing oxidative modifications.4 Oxidized LDLs (oxLDLs), present in atherosclerotic areas, exert various biological effects potentially involved in atherogenesis, including lipid accumulation in macrophages (foam cells), changes in gene expression of adhesion molecules, cytokines, growth factors and coagulation proteins, cell migration, cell proliferation, apoptosis, and local inflammatory/immune response.4,5
SMC migration and proliferation play a critical role in the formation of fibrous cap of atherosclerotic plaques, whereas toxic events are potentially involved in endothelial cell injury, necrotic core formation, and plaque rupture.1,2,6 Excessive SMC proliferation and extracellular matrix biosynthesis may lead to occlusive lesions.1 In contrast, the “thin cap fibroatheroma” containing inflammatory cells and a lipid-rich core is generally a vulnerable plaque prone to atherothrombotic events.7–9
Proliferation of SMC is regulated by numerous growth factors and cytokines8,10 and by oxLDL.11,12 Platelet-derived growth factor (PDGF) plays a major role in the fibroproliferative response during atherogenesis and restenosis.1,13 PDGF-induced signaling is mediated by PDGF receptor-α (PDGFRα) and PDGFRβ, which belong to the large family of receptor tyrosine kinases.14 PDGFRα, which binds A and B PDGF isoforms, is involved in SMC hypertrophy. PDGFRβ, which binds only PDGF B-chain, is implicated in the migration and proliferation of SMCs.14,16 On ligand binding, the PDGFRβ tyrosine kinase is activated and creates phosphotyrosine residues that act as docking sites for Src homology 2–containing proteins, including Src, phospholipase C-γ1, phosphatidylinositol-3′-kinase, SH2 domain containing rynosine phosphatase protein tyrosine phosphatase, GTPase-activating protein of p21ras, Nck, Grb2, and Shc.14–16
We recently reported that PDGFRβ, like epidermal growth factor receptor (EGFR), acts as a sensor for both oxidized lipids and oxidative stress and that 4-hydroxynonenal (4HNE) contained in mildly oxidized LDL induces both adduct formation and activation of these receptor tyrosine kinases.17,18 Because oxLDL and PDGF are present in atherosclerotic lesions,1,2 we evaluated their respective contribution to SMC proliferation. Preliminary experiments showed unexpected results and led us to investigate this paradoxical effect.
The reported data show that after moderate but long-lasting PDGFRβ activation, oxLDL (as well as 4HNE) induced HNE–PDGFRβ adduct formation associated with desensitization of PDGFRβ. This was characterized by a loss of PDGFRβ signaling and mitogenic effect induced by PDGF. Moreover, HNE–PDGFRβ adducts were observed in atherosclerotic plaques of humans and animal models and was prevented by aldehyde scavenger treatment.
Materials and Methods
An extensive Materials and Methods section is available in the online supplement at http://circres.ahajournals.org.
Chemicals and Biological Reagents
For chemicals and biological reagents, see the online supplement.
Rabbit arterial SMCs were grown and starved under the conditions used previously.18 For extensive information, see the online supplement.
Animal Models: Apolipoprotein E–Deficient Mice and Rabbits
Apolipoprotein E–deficient (apoE−/−) and wild-type mice were from Transgenic Alliance (IFFA Credo; Charles River; Les Oncins, France). Mice, housed under specific pathogen-free conditions, were fed a standard diet.
New Zealand white male rabbits (2.5 kg) were randomly assigned to regular (n=4) and hypercholesterolemic (2% cholesterol for 12 weeks) diets, associated (n=6) or not (n=6) to Hydralazine treatment (10 mg/kg per day for 12 weeks). After euthanasia (by pentobarbital; 50 mg/kg), thoracic aortas were removed, cleaned of excess adventicial tissues, opened longitudinally to obtain an en face preparation, and photographed (Nikon Coolpix 995). Fatty streaks were stained with Sudan IV. The extent of atherosclerosis was assessed using computer-aided planimetry (Cyberview 3.0 software), and the lesion areas covering the aortic surface were quantitated.
Experiments were performed according to French accreditation for laboratory animal care.
Western Blot Analysis and Determination of PDGFRβ Free Amino Groups
Cells or tissue samples were lysed, PDGFRβ was immunoprecipitated, and Western blots were performed under conditions described previously.17,18 The free amino group content of PDGFRβ was evaluated after immunoprecipitation using [3H]NSP ([3H]N-succinimidyl propionate; Amersham) labeling.17
Detergent-Free Purification of Caveolae/Raft Membrane Fractions
Lipid raft-enriched domains were purified using a modified carbonate method.19 For extensive information see the online supplement.
Evaluation of DNA Synthesis, Cell Proliferation, and Viability
DNA synthesis was evaluated by [3H]thymidine incorporation as described.11 Alternatively, proliferating cells were detected by immunocytochemistry using anti–proliferating-cell nuclear antigen (PCNA) antibody and visualized using a Zeiss LSM 510 fluorescence confocal microscope.20 Cell viability was evaluated by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay.20
Cell surface PDGFR determination was performed by flow cytofluometry according to Okumura.21 For extensive information see the online supplement.
[125I]PDGF binding was performed according to Escargueil-Blanc et al.18 Cells incubated±oxLDL (75 μg/mL; 24 hours) were incubated with [125I]PDGF (70 000 cpm/mL; 30 pmol/L), then washed in PBS containing 0.5% BSA, and the cell-associated radioactivity was determined (Minaxi; Packard). Nonspecific binding was determined on the basis of excess unlabeled PDGF (5 nmol/L). Cells were washed twice in PBS containing 0.5% BSA and once in PBS alone, and the radioactivity was counted (Minaxi Gamma Packard Counter).
Determination of Enzyme Activities
Proteasome activity and PDGFRβ tyrosine kinase activities were determined using conditions described previously.18,22
Frozen sections of aortic sinus from apoE−/− and wild-type mice were labeled by antiacrolein antibody,23 and aortas from cholesterol-fed rabbits, treated or not by Hydralazine, were labeled by anti–4-HNE monoclonal antibody and revealed by a peroxidase-conjugated secondary antibody (immunoperoxidase kit; Dako; StrepABComplex/HRP Duet). For each detection, a control was done by omitting the primary antibody to determine the nonspecific binding. Nuclei were counterstained with hematoxylin.
Data are given as means±SEM. Estimates of statistical significance were performed by ANOVA (Tukey test; SigmaStat software). Values of P<0.05 were considered significant.
Preincubation of Cells With oxLDL Inhibits the Mitogenic Effect of PDGF and PDGFRβ Signaling
Mitogenic concentration (75 μg apoB/mL) of oxLDL moderately stimulated DNA synthesis, whereas, as expected, PDGF was a more potent mitogen (Figure 1A, top). However, unexpectedly, in cells preincubated with nontoxic oxLDL concentrations, the mitogenic response to PDGF was strongly reduced. Inhibition of DNA synthesis induced by oxLDL pretreatment was not attributable to cell loss, as assessed by MTT assay (Figure 1A, bottom) but rather to cell cycle inhibition, as shown by the nuclear expression of PCNA, a marker of cell cycle progression. PCNA labeling was intense in the nucleus of PDGF-treated cells but not in cells incubated with oxLDL or oxLDL+PDGF (Figure 1B). Interestingly, preincubation of cells with oxLDL did not inhibit the mitogenic effect of FCS (Figure 1A), indicating that under our conditions, all the mitogenic pathways were not blocked by oxLDL preincubation. We thus hypothesized that oxLDL could inhibit an early step of PDGF signaling.
The mitogenic effect of oxLDL was associated with PDGFRβ autophosphorylation,18 which was sustained for 16 to 20 hours and then declined toward the baseline at 24 hours (Figure 1C). In contrast, native LDL did not trigger PDGFRβ activation up to 4 hours (Figure 1C). Interestingly, in cells preincubated with oxLDL for 24 hours, addition of PDGF did not induce PDGFRβ activation nor extracellular signal-regulated kinase 1/2 (ERK1/2) signaling (Figure 1D and 1E). Of note, ultraviolet (UV)-oxLDL and cell oxLDL gave similar results (Figure 1D and 1E). We preferred to use UV-oxLDL because cell–oxLDL preparations may contain potentially interfering bioactive compounds (released from endothelial cell vein-304 cells).
These data suggest that preincubating cells with oxLDL induces a decrease of PDGF-induced PDGFRβ activation. We next investigated the mechanism of this inhibitory effect.
A Role for 4HNE in the OxLDL-Induced PDGFR Inhibition
In contrast to native LDL, oxLDL induced aldehyde–protein adduct formation, as shown by Western blots labeled with anti-4HNE and antiacrolein antibodies, and by determining free amino groups content (Figure 2 A through 2C).
Preincubation of cells with 4HNE mimicked the effect of oxLDL, as assessed by the loss of [3H]NSP-reactive free amino groups (Figure 2B) and the formation of 4HNE–PDGFR adducts (Figure 2C). These 4HNE-induced structural modifications (Figure 2C) were associated with a loss of PDGF-induced PDGFRβ activation and subsequent decrease of both ERK1/2 activation and DNA synthesis (Figure 2D through 2F). Acrolein gave similar results on PDGFRβ signaling (data not shown) but was not used in the following experiments because of its volatility (making it difficult to perform a dose-effect study) and toxicity (potentially interfering in signaling studies).
These data support the hypothesis that 4HNE and acrolein mediate, at least in part, the structural modification of PDGFRβ induced by oxLDL.
Because cumene hydroperoxide (CHP) has been shown to inhibit cell proliferation, similarly to 4HNE,24 we investigated whether CHP could also inactivate PDGFRβ. Cells were treated with the relatively hydrosoluble CHP (used at 50 μmol/L, a subtoxic concentration). CHP treatment induced a significant decrease in PDGF-induced PDGFRβ activation (Figure 2G) with no detectable level of cellular lipid peroxidation or thiobarbituric acid reactive substances (data not shown), similarly to oxLDL under the used nontoxic conditions (oxLDLs induce the peroxidation of cellular lipids only at toxic concentration).
Mechanisms of oxLDL-Induced PDGFRβ Dysfunction
Because PDGF stimulation is known to induce subcellular relocation of PDGFRβ and its degradation by proteasome or lysosomes,25,26 we investigated whether oxLDLs were able to induce these quantitative or topological changes. OxLDL stimulated proteasome activity (Figure 3A), but neither oxLDL nor 4HNE induced loss of total cellular PDGFRβ (Figure 3B). Then, we evaluated the level of PDGFRβ exposed at the cell surface by flow cytometry of nonpermeabilized cells labeled with anti-PDGFRβ antibody. After 24-hour preincubation with oxLDL or 4HNE, cell surface level of PDGFRβ was decreased by 19±6% and 15±4%, respectively (Figure 3C), whereas, as expected,24 the cell surface level of PDGFRβ was dramatically decreased in cells stimulated by PDGF (Figure 3C).
Because PDGF signaling may also depend on cholesterol-rich microdomains (rafts and caveolae),27 and extensively, oxLDLs have been shown to alter microdomains integrity,28 we investigated whether the location of PDGFRβ in caveolae/raft microdomains was altered in cells preincubated with mildly oxidized LDL. As shown in Figure 3D, oxLDL induced no major change in the distribution of PDGFRβ between raft and nonraft membranes.
Altogether, these data suggest that part of the inhibitory effect of oxLDL or 4HNE results from a decrease of PDGFRβ located at the cell surface.
Because this only partial internalization of PDGFRβ probably does not account for the complete inhibition of PDGFRβ by oxLDL, we investigated whether oxLDL and 4HNE altered the intrinsic functional properties of the receptor. Preincubation of cells with oxLDL or 4HNE reduced the binding of PDGF by 46% and 27%, respectively (Figure 3E). The tyrosine kinase activity of PDGFRβ was more strongly inhibited by 82% and 71%, respectively (Figure 3F), thus explaining the severe reduction of PDGF-induced PDGFRβ activation elicited by oxLDL or 4HNE.
DNPH and Hydralazine Prevent the oxLDL-Induced PDGFRβ Dysfunction In Vitro
Because oxLDL-induced PDGFRβ dysfunction results, at least in part, from the formation of 4HNE–PDGFRβ adducts, we hypothesized that aldehyde scavengers may prevent both protein modification and subsequent dysfunction. For this purpose, we used the prototypic hydrazine compound DNPH, which can be used on cultured cells but not in vivo because of its toxicity, and Hydralazine, another hydrazine compound, compatible with in vivo studies.
As reported in Figure 4 A and 4B, DNPH and Hydralazine prevented oxLDL-induced 4HNE–PDGFRβ adduct formation and rescued in part PDGF-induced PDGFRβ activation. Concomitantly, DNPH prevented oxLDL-induced inhibition of the mitogenic effect of PDGF. In contrast, the protective effect of Hydralazine on the mitogenic effect of PDGF could not be tested because Hydralazine by itself inhibited the mitogenic effect of PDGF through an unknown mechanism (data not shown).
PDGFRβ Modification in Atherosclerotic Lesions: Protection by Hydralazine
We next evaluated whether 4HNE– or acrolein–PDGFRβ adducts were detected in atherosclerotic areas and whether aldehyde scavengers were able to prevent PDGFRβ modification.
As shown in Figure 5 A and 5B, PDGFRβ modification was clearly visualized by antiacrolein and by anti-4HNE antibodies on Western blots of aortic atherosclerotic lesions from apoE−/− mice and in human carotid advanced plaques. In contrast, normal arteries were not labeled. It may be noted that atherosclerotic subendothelial areas of aortic sinus from apoE−/− mice were stained by antiacrolein antibodies, in contrast to aortic sinus of control mice (Figure 5A).
Cholesterol-fed rabbits exhibited hypercholesterolemia (23±4 g/L) and large fatty streaks in aorta (supplemental Figure IA, available in the online supplement; additional data available in the online supplement), whereas chow diet–fed rabbits have low cholesterol levels (1.1±0.2 g/L) and no fatty streak in aortas (data not shown). As reported in Western blots of Figure 6A, high levels of 4HNE–PDGFRβ adducts were detected in aortic atherosclerotic lesions of hypercholesterolemic rabbits, whereas PDGFRβ was only very slightly labeled by anti-4HNE antibody in normocholesterolemic controls. It may be noted that PDGFRβ was also labeled by antiubiquitin antibody in atherosclerotic lesions but not in control aortas. Hydralazine was effective in preventing both 4HNE– and ubiquitin–PDGFRβ adduct formation in aortas of hypercholesterolemic rabbits (Figure 6A). It may be noted that the extension of atherosclerotic plaques was lesser in Hydralazine-treated cholesterol-fed rabbits than in the untreated cholesterol fed rabbits, as shown by en face aorta photographs (supplemental Figure IA; additional data available in the online supplement) and by quantitative evaluation of atherosclerotic lesion area (supplemental Figure IB; additional data available in the online supplement). In the aortic intima of cholesterol-fed rabbits, fatty streaks areas contained areas of 4HNE–protein adduct formation (Figure 6B, arrows). In contrast, anti-4HNE labeling was much lower in the intima of aorta of Hydralazine-treated cholesterol-fed rabbits (Figure 6B, arrows).
The reported data show that: (1) long-term incubation of SMCs with mildly oxidized LDL reduced PDGF-induced PDGFRβ activation and subsequent mitogenic effect; (2) oxLDL- and 4HNE-induced structural modification and desensitization of PDGFRβ resulted mainly from inhibition of PDGF binding and PDGFRβ–tyrosine kinase; and (3) PDGFRβ downregulation was prevented in part by aldehyde scavengers, such as DNPH and Hydralazine, both in cell culture and in vivo in animal models.
We recently reported that short-term incubation of cells with oxLDL elicited the activation of EGFR17 and PDGFRβ18 by at least two mechanisms: one involving 4HNE–protein adduct formation and the second involving the generation of reactive oxygen species and subsequent PTPase inhibition.18 This partial PTPase inhibition may explain in part the sustained (20 hours) tyrosine phosphorylation of PDGFRβ observed during incubation with oxLDL. However, after 24-hour incubation with oxLDL, PDGFRβ was dephosphorylated but was insensitive to PDGF stimulation. Moreover, preliminary experiments showed that the mitogenic effect of oxLDL and of PDGF were not additive, which led us to investigate the mechanisms by which oxLDLs inhibit PDGF-induced PDGFRβ activation.
A classical mechanism of PDGFRβ downregulation after PDGF activation consists in the internalization of the complex ligand-receptor into endosomes.25 Then the complex dissociates and the receptor recycles to the cell surface, or the complex is degraded on fusion of endosomes with lysosomes.15 Alternatively, PDGFRβ may undergo ubiquitination and degradation in the proteasome.26 Because oxLDL-induced PDGFRβ downregulation occurred after a moderate but sustained activation of PDGFRβ,18 and because oxLDL induced proteasome activation, we investigated whether oxLDLs were able to trigger PDGFRβ degradation. Under our conditions, oxLDL induced no change in the whole level of PDGFRβ, whereas, as expected,15 PDGFRβ level was dramatically reduced after 8-hour stimulation by PDGF. This is consistent with the fact that oxLDL-induced PDGFRβ activation does not involve autocrine secretion of PDGF.18 In the same way, the inhibitory effect of oxLDL cannot be explained by the minor internalization of cell surface PDGFRβ.
Because PDGFRβ signaling is dependent on caveolae-associated signalosome27 and because extensively oxLDL can extract cholesterol from caveolae,28 thereby altering caveolae structure and potentially changing PDGFRβ localization, we hypothesized that oxLDL may induce PDGFRβ relocalization and subsequent function alteration. Because we observed no major change in PDGFRβ distribution between raft and nonraft membranes, it was therefore excluded that the severe inhibition of PDGFRβ induced by oxLDL may result from degradation or internalization of the receptor. This led us to hypothesize that peroxidation products of oxLDL (4HNE or acrolein) might alter the structure and function of the receptor.
4HNE and acrolein are lipid peroxidation products of polyunsaturated fatty acids, formed during the oxidation of LDL or membranes.29–31 These aldehydes can react with thiols and amino groups of cell proteins and apoB, thereby altering their structure and functional properties.30 Incubation of cells with oxLDL elicits the formation of 4HNE–cell protein adducts,17,18,23 and experiments with [3H]4HNE-labeled oxLDL have shown that a small part of [3H]4HNE of oxLDL is transferred to PDGFRβ (and other cell proteins).18 How 4HNE transfers from oxLDL to cell proteins is only partly understood because a larger part of 4HNE is covalently bound to amino groups of oxLDL and because the minor part of free 4HNE is probably dissolved in the phospholipid phase.30 The transfer of 4HNE from oxLDL to cellular proteins is energy dependent (inhibited at 4°C) and blocked by proton pump inhibitors,18 thus suggesting an active metabolic process.
Moreover, we examined whether oxLDL may trigger oxidative stress18,32 and thereby oxidize cellular lipids and generate 4HNE.31 Under nontoxic conditions used here, oxLDL induced no detectable peroxidation of cellular lipids (thus probably generating not 4HNE). This was confirmed by using CHP, which can induce DNA synthesis inhibition24 and cellular lipid peroxidation.33 In our experimental system, CHP (used at nontoxic concentration) was able to inhibit PDGFRβ, independently of any peroxidation of cellular lipids (thus independently of 4HNE formation). It is therefore unlikely that in our experiments, 4HNE may result from cellular lipid peroxidation.
Another crucial question was understanding whether oxLDL-induced PDGFRβ inactivation was causally related to 4HNE–PDGFRβ adducts formation. This was not a trivial question because our previous reports demonstrated that short incubation of PDGFRβ or EGFR with 4HNE induced a moderate but sustained activation of these receptors.17,18 Our data suggest that incubation of cells with oxLDL triggers a biphasic effect, the initial activation (sustained for 16 hours) being followed by a late inhibition, associating dephosphorylation of PDGFRβ with a marked resistance to PDGF stimulation. These effects were mimicked by preincubation of cells with 4HNE, suggesting that the formation of 4HNE–PDGFRβ adducts mediates the insensitivity of PDGFRβ to PDGF. The mechanism of this inhibitory effect was relatively complex because it resulted from decreased binding of PDGF and inhibition of PDGFRβ tyrosine kinase activity. These data are consistent with previous reports indicating that extensive 4HNE adduct formation is often associated with a loss of function of several enzymes,31 probably by altering the protein conformation or the active site.
The causal relationship between aldehyde–PDGFRβ adduct formation and PDGFRβ inhibition induced by oxLDL was supported by the protective effect of aldehydes scavengers DNPH and Hydralazine, which prevented concomitantly adduct formation and receptor inhibition.
4HNE– and acrolein–PDGFRβ adducts were found in aortic atherosclerotic lesions of two animal models, cholesterol-fed rabbits and apoE−/− mice, suggesting that PDGFRβ modification by lipid peroxidation products occurs in vivo, like modification of other proteins (apoB) visualized by antibodies recognizing oxidation-specific epitopes in atherosclerotic plaques.34,35 Oxidation-specific modifications of PDGFRβ were associated with increased ubiquitination of the receptor, in agreement with that observed in cultured cells treated with oxLDL.22 This demonstrates that modified PDGFRβ is detected and labeled by the ubiquitination system in vivo and confirms that the observed modification is not an artifactual postmortem modification.
From a (patho)physiological point of view, PDGFRβ is implicated in angiogenesis, neointima formation, fibrous cap thickening, and restenosis.1,13 Atherosclerotic plaques displaying thick fibrous cap are generally stable and may lead to occlusive lesions, whereas plaques with thin fibrous cap and lipid-rich core (containing oxidized lipids) are generally unstable, prone to rupture and initiate frequently atherothrombotic events.8,9
The presence of modified and ubiquitinated PDGFRβ in atherosclerotic lipid-rich lesions suggests that PDGFRβ could be locally downregulated and refractory to PDGF and could participate in forming a thin fibrous cap, characteristic of vulnerable lipid-rich plaques.8,9 Moreover, oxidized lipids can trigger local inflammatory response,36 metalloproteinase overexpression,37 and apoptosis,6,38 events converging toward triggering plaque disruption and subsequent atherothrombotic events. Hydralazine that acts in vitro as a carbonyl scavenger like DNPH is also active in vivo in preventing aldehyde–PDGFRβ adduct formation. However, it may be noted that Hydralazine is a well-known antihypertensive agent and is able to inhibit oxidizing enzymes such as NADPH oxidase and cytochrome P450.39 The antihypertensive effect is probably not involved in the inhibition of 4HNE–PDGFRβ adduct formation because we did not observe any major change in arterial blood pressure in Hydralazine-treated rabbits. However, it cannot be excluded that the protective effect of Hydralazine may be mediated by the inhibitory effect on oxidant enzymes because oxidative stress is thought to play a role in atherogenesis by inducing endothelial dysfunction, inflammatory response40 and LDL oxidation.4 All these mechanisms of action of Hydralazine converge to prevent the formation of aldehyde–protein and aldehyde–PDGFRβ adducts in vivo, thus suggesting that it may prevent the PDGFRβ dysfunction and subsequent events potentially involved in plaque formation and instability.
This work was supported by grants from INSERM, Université Paul Sabatier-Toulouse III to U-466.
Original received August 31, 2005; revision received February 16, 2006; accepted February 23, 2006.
Berk BC. Vascular smooth muscle growth: autocrine growth mechanisms. Physiol Rev. 2001; 81: 999–1030.
Auge N, Garcia V, Maupas-Schwalm F, Levade T, Salvayre R, Negre-Salvayre A. Oxidized LDL-induced smooth muscle cell proliferation involves the EGF receptor/PI-3 kinase/Akt and the sphingolipid signaling pathways. Arterioscler Thromb Vasc Biol. 2002; 22: 1990–1995.
Sano H, Sudo T, Yokode M, Murayama T, Kataoka H, Takakura N, Nishikawa S, Nishikawa SI, Kita T. Functional blockade of PDGF receptor-beta but not of receptor-alpha prevents vascular smooth muscle cell accumulation in fibrous cap lesions in apolipoprotein E-deficient mice. Circulation. 2001; 103: 2955–2960.
Claesson-Welsh L. PDGF signals. J Biol Chem. 1994; 268: 32023–32026.
Suc I, Meilhac O, Lajoie-Mazenc I, Vandaele J, Jurgens G, Salvayre R, Negre-Salvayre A. Activation of EGF Receptor by oxidized LDL. FASEB J. 1998; 12: 665–671.
Escargueil-Blanc I, Salvayre R, Arnal JF, Parthasarathy S, Nègre-Salvayre A. Mildly oxidized LDL induce activation of PDGF β-receptor pathway. Circulation. 2001; 104: 1814–1821.
Song KS, Li S, Okamoto T, Quilliam LA, Sargiacomo M, Lisanti MP. Co-purification and direct interaction of Ras with caveolin, an integral membrane protein of caveolae microdomains. Detergent-free purification of caveolae microdomains. J Biol Chem. 1996; 271: 9690–9697.
Vindis C, Elbaz M, Escargueil-Blanc I, Auge N, Heniquez A, Thiers JC, Negre-Salvayre A, Salvayre R. Two distinct calcium-dependent mitochondrial pathways are involved in oxidized LDL-induced apoptosis. Arterioscler Thromb Vasc Biol. 2005; 25: 639–645.
Vieira O, Escargueil-Blanc I, Jürgens G, Borner C, Almeida L, Salvayre R, Negre-Salvayre A. Oxidized LDL alter the activity of the ubiquitin-proteasome pathway. Potential role in oxidized LDL-induced apoptosis. FASEB J. 2000; 14: 532–542.
Uchida K, Kanematsu M, Sakai K, Matsuda T, Hattori N, Mizuno Y, Suzuki D, Miyata T, Noguchi N, Niki E, Osawa T. Protein bound acrolein: potential markers for oxidative stress. Proc Natl Acad Sci U S A. 1998; 95: 4882–4887.
Sorkin A, Westermark B, Heldin CH, Claesson-Welsh L. Effect of receptor kinase inactivation on the rate of internalization and degradation of PDGF and the PDGF beta-receptor. J Cell Biol. 1991; 112: 469–478.
Blair A, Shaul PW, Yuhanna IS, Conrad PA, Smart EJ. Oxidized low density lipoprotein displaces endothelial nitric-oxide synthase (eNOS) from plasmalemmal caveolae and impairs eNOS activation. J Biol Chem. 1999; 274: 32512–32519.
Uchida K, Szweda LI, Chae HZ, Stadtman ER. Immunochemical detection of 4-hydroxynonenal protein adducts in oxidized hepatocytes. Proc Natl Acad Sci U S A. 1993; 90: 8742–8746.
Robbesyn F, Garcia V, Auge N, Vieira O, Frisach MF, Salvayre R, Negre-Salvayre A. HDL counterbalance the proinflammatory effect of oxidized LDL by inhibiting intracellular reactive oxygen species rise, proteasome activation, and subsequent NF-kappaB activation in smooth muscle cells. FASEB J. 2003; 17: 743–745.
Hogberg J, Moldeus P, Arborgh B, O’Brien PJ, Orrenius S. The consequences of lipid peroxidation in isolated hepatocytes. Eur J Biochem. 1975; 15: 457–462.
Rosenfeld ME, Palinski W, Yla-Herttuala S, Butler S, Witztum JL. Distribution of oxidation specific lipid-protein adducts and apolipoprotein B in atherosclerotic lesions of varying severity from WHHL rabbits. Arteriosclerosis. 1990; 10: 336–349.
Furnkranz A, Schober A, Bochkov VN, Bashtrykov P, Kronke G, Kadl A, Binder BR, Weber C, Leitinger N. Oxidized phospholipids trigger atherogenic inflammation in murine arteries. Arterioscler Thromb Vasc Biol. 2005; 25: 633–638.