High-Density Lipoproteins Prevent the Oxidized Low-Density Lipoprotein–Induced Endothelial Growth Factor Receptor Activation and Subsequent
http://www.100md.com
动脉硬化血栓血管生物学 2005年第6期
From INSERM U-466 and Biochimie IFR-31 (F.R., N.A., C.V., A.-V.C., A.N.-S., R.S.), Faculty of Medicine, University Paul Sabatier, Toulouse, France; INSERM U-563 (R.B.), Centre de Physiopathologie de Toulouse-Purpan, Department Lipoproteines et Mediateurs Lipidiques, IFR-30, Toulouse, France.
Correspondence to R. Salvayre or A. Negre-Salvayre, Biochimie, INSERM U466, IFR-31, CHU Rangueil, 1, avenue Jean Poulhès, TSA-50032, 31059 Toulouse Cedex 9, France. E-mail salvayre@toulouse.inserm.fr or anesalv@toulouse.inserm.fr
Abstract
Objectives— The atherogenic oxidized low-density lipoprotein (oxLDL) induces the formation of carbonyl-protein adducts and activates the endothelial growth factor receptor (EGFR) signaling pathway, which is now regarded as a central element for signal transduction. We aimed to investigate whether and by which mechanism the anti-atherogenic high-density lipoprotein (HDL) prevents these effects of oxLDL.
Methods and Results— In vascular cultured cells, HDL and apolipoprotein A-I inhibit oxLDL-induced EGFR activation and subsequent signaling by acting through 2 separate mechanisms. First, HDL, like the aldehyde scavenger dinitrophenyl hydrazine, prevented the formation of oxLDL-induced carbonyl–protein adducts and 4-hydroxynonenal (HNE)–EGFR adducts. Secondly, HDL enhanced the cellular antioxidant defenses by preventing (through a scavenger receptor class B-1 (SR-BI)–dependent mechanism) the increase of intracellular reactive oxygen species (ROS) and subsequent EGFR activation triggered by oxLDL or H2O2. A pharmacological approach suggests that this protective effect of HDL is independent of cellular glutathione level and glutathione peroxidase activity, but it requires catalase activity. Finally, we report that oxLDL upregulates both membrane type 1 (MT1)-matrix metalloproteinase-1 (MT1-MMP) and MMP-2 through an EGFR-dependent mechanism and that HDL inhibits these events.
Conclusions— HDLs block in vitro oxLDL-induced EGFR signaling and subsequent MMP-2 activation by inhibiting carbonyl adducts formation and cellular oxidative stress. These effects of HDL may participate to reduce cell activation, excessive remodeling, and alteration of the vascular wall.
Oxidized LDLs induce EGFR activation and subsequent MMP-2 activation. HDLs inhibit these events by 2 separate mechanisms, ie, by blocking carbonyl–protein adduct formation and by inhibiting the oxLDL-induced and H2O2-induced intracellular ROS increase, through a catalase-dependent process. This may contribute to reduce cell activation, excessive remodeling, and vascular wall alteration.
Key Words: atherosclerosis ? endothelial growth factors ? lipoproteins
Introduction
Low-density lipoproteins (LDLs) are thought to become atherogenic after undergoing oxidative modifications.1–3 Oxidized LDLs (oxLDLs) trigger cell signaling pathways involved in cellular responses, such as proliferation, inflammation, and apoptosis, which occur during atherogenesis.4–7 In contrast, high-density lipoproteins (HDLs) are anti-atherogenic and cardioprotective.8 Besides their role in the reverse transport of cholesterol, 9 HDL inhibits LDL oxidation and cell signaling mediated by oxLDL, and counters several adverse biological effects, such as cytotoxicity and inflammatory response triggered by cytokines, oxLDL, or oxidants.4,10–12 The molecular mechanism of this protective effect of HDL is only partly understood.
The endothelial growth factor receptor (EGFR) family consists of 4 members of tyrosine kinase receptors, EGFR (or ErbB1), HER2 (or ErbB2/neu), HER3 (or ErbB3), and ErbB4. EGFR is activated by binding of specific peptide ligands (EGF, transforming growth factor-, amphiregulin, neu differentiation factor [NDF], neuregulins, heregulins, heparin-binding epidermal growth factor [HB-EGF], betacellulin, and epiregulin);13–15 nonspecific and/or stress stimuli, such as ultraviolet and gamma radiations, alkylating agents signaling,16,17 oxidized lipids,18 fatty acids,19 and oxidants;20 and transactivation by several signaling pathways.21,22 For instance, cross-talk between G-protein–coupled receptor (for instance, angiotensin receptor-1 in vascular cells) and EGFR is mediated by cell signaling involving G-proteins, calcium, protein kinase C, Src, and metalloproteinases (such as a disintegrin and metalloprotease [ADAM]) that cleave pro-EGFR ligands, thereby liberating agonists of EGFR.21,22 Through these mechanisms, EGFR is a start point of a classical tyrosine kinase receptor signaling cascade, as well as a switch point of a wide cellular communication network.21,22 Therefore, EGFR acts as "a central element for signal transduction" and regulation of several cellular functions, such as cell growth, differentiation, motility, survival, and death,21 which also play a pivotal role in various pathophysiological processes. In addition to its prominent role in the development of several cancers,21 EGFR has been recently implicated in vascular pathophysiological processes associated with excessive remodeling and atherosclerosis.22,23 Among various biological effects, EGFR could play a role in cell migration,20–22 mitogenic signaling,23 and NF-B activation,24,25 and induces the upregulation of metalloproteinases (MMPs).26
MMPs are zinc proteases that are expressed in atherosclerotic plaques, where they may contribute to vascular remodeling and plaque disruption.27–29 MMPs are regulated through gene expression and activation of latent MMPs by cleavage of the N-terminal prosegment.30 oxLDL and inflammatory cytokines enhance the expression and activity of several MMPs, such as MMP-1, MMP-9 (gelatinase B), and membrane type 1 (MT1)-MMP (that cleaves and activates MMP-2) in vascular cells.27,31,32
The reported data show that HDLs are able to modulate the oxLDL-induced EGFR activation and subsequent upregulation of MMP-2 in vascular smooth muscle cells (SMCs).
Methods
A detailed Methods section is reported as an online-only data supplement (available at http://atvb.ahajournals.org).
Results
HDL Inhibits oxLDL-Induced EGFR Activation
Incubation of cells with oxLDL induced mild tyrosine-phosphorylation of EGFR that was inhibited by HDL (Figure 1A and 1B). Because similar data were obtained on vascular SMCs and endothelial cell vein (ECV)-304, we used, alternatively, ECV-304 cells when high EGFR level was required. Native LDL triggered no (or only minor) EGFR activation (Figure 1A). Under the used conditions, EGFR activation began at relatively low concentrations of oxLDL (25 to 50 μg apolipoprotein B/mL), was dose-dependent, and was completely inhibited by HDL in a dose-dependent manner (Figure 1B). Of note, 200 μg apolipoprotein B/mL (UV+Cu) oxLDL were not toxic over the 5 hours of the experiments (apoptosis becoming manifest only after 12-hour incubation). oxLDL-induced EGFR activation was independent of the apoptotic effect because it occurred at low, nontoxic, concentrations of (UV+Cu) oxLDL (25 to 50 μg apolipoprotein B/mL; UV) oxLDL (which are less toxic than [UV+Cu] oxLDL) induced also EGFR activation (data not shown).
Figure 1. HDLs inhibit the oxLDL-induced EGFR tyrosine phosphorylation and subsequent signaling. Cells (SMCs in A left, B, C; and ECV in A right and D) were preincubated with or without HDL (200 μg apolipoprotein A/mL, unless otherwise indicated) for 18 hours, before incubation with oxLDL (200 μg apolipoprotein B/mL), for 5 hours, and used for Western blots and labeled using the indicated antibodies. EGF (10 nmol/L for 15 minutes) was used as maximal positive control. A, EGFR tyrosine phosphorylation in SMCs incubated with oxLDL or native (nonoxidized) LDL (200 μg apolipoprotein B/mL) for 5 hours, after preincubation with HDLs (200 μg apoA/mL), when indicated (+). EGFR was immunoprecipitated (IP) by anti-EGFR antibody before Western blotting revealed by anti-phosphotyrosine (anti-PY) or anti-EGFR antibodies. B, Left panel shows the oxLDL-induced EGFR activation is dose-dependent (0 to 200 μg apolipoprotein B/mL of oxLDL) and is inhibited by HDLs (200 μg apolipoprotein A/mL). B, Right panel shows inhibitory effect of increasing HDL concentrations (0 to 200 μg apolipoprotein B/mL) on the oxLDL-induced EGFR activation. C, HDLs inhibit both the early (1-hour) and late (5-hour) components of oxLDL-induced EGFR activation (experimental conditions as in A). D, HDLs inhibit signaling subsequent to oxLDL-induced EGFR activation. Western blot of proteins coimmunoprecipitation (same experimental conditions as in A). Representative data of 3 to 5 experiments.
Because 2 phases of oxLDL-induced EGFR activation can be discriminated by their antioxidant susceptibility,19 we investigated whether HDL were effective on each component. Preincubation of cells with HDL prevented the early (1-hour) and late (5-hour) phases of oxLDL-induced EGFR activation (Figure 1C) and the subsequent signaling (Figure 1D).
The Protective Effect of HDL Is Mediated by Apolipoprotein A-I and SR-B1
The oxLDL-induced EGFR activation was inhibited by delipidated HDL apolipoprotein fraction and by apolipoprotein A-I (Figure 2A). Serum albumin, which is a frequent contaminant of HDL fractions (10% to 25% of total HDL-associated proteins) but is only minor in LDL fractions (<1% of apolipoprotein B content), did not prevent oxLDL-induced EGFR activation. The inhibitory effect of HDL was lost when apolipoprotein A-I was altered by HClO-mediated oxidation (containing 3.8±0.4 nmol thiobarbituric reactive substances [TBARS]/μg apolipoprotein A). In contrast, UV-oxidized HDL (2.7±0.3 nmol TBARS/μg apolipoprotein A) that exhibited only moderately altered apolipoprotein A-I were effective to prevent oxLDL-induced EGFR activation (Figure 2B). The protective effect of HDL was abolished by co-incubation of cells with HDL and anti-SRB1 antibodies (Figure 2C). This suggests that this effect of HDL requires the interaction between HDL and SRB1. Finally, paraoxonases were not implicated in this protective effect, because HDL fractions with very low paraoxonase activity (prepared in the presence of EDTA 100 μmol/L) and HDL fractions with high paraoxonase activity (prepared without EDTA) exhibited the same inhibitory effect (Figure 2D). These data strongly suggest that the protective effect of HDL against oxLDL-induced EGFR activation is mainly mediated through the interaction of apolipoprotein A-I with SRB1.
Figure 2. The protective effect of HDL is mediated by apolipoprotein A-I and SRB1. A, Protective effect of HDL compared with that of delipidated HDL apolipoproteins (apo), apoA-I (200 μg apoA/mL), and serum albumin (SA) (200 μg/mL). B, Left panel shows effect of native (nonoxidized [n]) HDL compared with that of UV-oxidized HDL (UV) and HClO-oxidized HDL (H) (200 μg apoA/mL). B, Right panel shows SDS-PAGE of apolipoproteins of native HDL, UV-oxidized HDL, HClO-oxidized HDL, and apoA-I fraction. C, Anti–SRB1 antibody (2 μg/mL) prevents the protective effect of HDLs. Cells were preincubated with or without HDL (200 μg apoA/mL) and/or anti–SRB1 antibody for 18 hours, before incubation with oxLDL, and Western blotting, as in Figure 1A. D, Inhibition of oxLDL-induced EGFR activation by HDL is independent of paraoxonase activity. HDLs, dialyzed against saline solution without (wo) or with (w) 100 μmol/L EDTA, were used to evaluate the protective effect (upper panels) and the paraoxonase activity (lower panel).
HDL Prevent the Formation of oxLDL-Induced Carbonyl–Protein Adducts and HNE–EGFR Adducts
oxLDL-induced EGFR activation results from HNE–EGFR adduct formation and reactive oxygen species (ROS) generation.18,19 This led to examining whether HDL may prevent the formation of HNE-EGFR adducts and other oxidized lipid–protein adducts. As shown in Figure 3A, HNE–EGFR adducts and EGFR activation induced by oxLDL were inhibited by HDL. More generally, HDL prevented the oxLDL-induced formation of carbonyl–protein adducts and the loss of [3H]NSP-reactive amino groups of cell proteins induced by oxLDL (Figure 3B and 3C). This effect was partly mimicked by the aldehyde scavenger dinitrophenyl hydrazine that prevented completely the formation of oxLDL-induced HNE–EGFR adducts and inhibited in part the oxLDL-induced EGFR activation (Figure 3A, right panel). Interestingly, Trolox, a hydrosoluble analog of tocopherol, induced a strong inhibition of EGFR autophosphorylation, but reduced only partly the level of HNE–EGFR adducts (Figure 3A). These data suggest that HDL are able to inhibit the oxLDL-induced EGFR activation (dinitrophenyl hydrazine-sensitive) mediated by HNE, and led us to hypothesize that HDL could also inhibit a second mechanism (Trolox-sensitive) involved in oxLDL-induced EGFR activation.
Figure 3. HDLs block the formation of HNE proteins and HNE–EGFR adducts induced by oxLDL. A, HDLs inhibit HNE–EGFR adduct formation. Cells were treated with or without HDL (200 μg apoA/mL, preincubation for 18 hours) and oxLDL (200 μg apoB/mL, incubation for 1 hour or 5 hours). Western blots were labeled using polyclonal anti-HNE protein, anti-PY, and anti-EGFR antibodies. Representative data of 4 experiments. B and C, HDLs prevent the oxLDL-induced formation of carbonyl adducts (B) and the decrease in free reactive amino groups (C) of cell proteins. Cells were preincubated with or without HDL (200 μg apoA/mL, 18 hours), then treated with oxLDL (200 μg apoB/mL, 5 hours) or with EGF (10 nmol/L, 15 minutes). Free amino groups determined by [3H]succinimidyl propionate ([3H]NSP) fixation, and carbonyls were expressed as percent of the value of controls. Mean±SEM of 4 experiments. *P<0.05.
HDL Inhibits oxLDL-Induced ROS Increase and Subsequent EGFR Activation
Because HDL inhibit an early increase of oxLDL-induced cellular ROS,33 and because ROS reinforce EGFR activation,16 we investigated whether HDL blocked the oxLDL-induced EGFR activation through a ROS-dependent mechanism. As reported in Figure 4, the oxLDL-induced cellular ROS increase (monitored by the fluorescence of oxidized dihydrorhodamine 123 [DHR] probe) was inhibited by HDL (Figure 4A) in a dose-dependent manner (Figure 4B). This inhibitory (protective) effect was lost with HClO-oxidized HDL, but not with UV-oxidized HDL (Figure 4C). Because apolipoprotein A-I is degraded in HClO-oxidized HDL, but is less altered in UV-oxidized HDL, it is suggested that this inhibitory effect of HDL requires the integrity of apolipoprotein A-I. Interestingly, preincubation of cells with anti-SRB1 antibody abrogated the protective effect of HDL (Figure 4C), thus suggesting that this effect is mediated through HDL/SRB1 interaction. Because EGFR activation by ROS and oxLDL results in part from the inactivation of phosphotyrosine phosphatases (PTPases),17,19,34 we examined whether HDLs were able to prevent the oxLDL-induced inhibition of PTPases. Under preincubation and coincubation conditions, HDL prevented effectively the oxLDL-induced inhibition of PTPases (Figure 4D). This protection of PTPases explains, at least in part, the protective effect of HDL against the oxLDL-induced EGFR activation. Interestingly, under "preincubation only" experimental conditions (cells were preincubated with HDL for 18 hours, then HDLs were removed just before oxLDL addition), HDLs were also able to block the oxLDL-induced ROS increase (Figure 4A) and EGFR activation (Figure 4E). Because, under "preincubation only" conditions, HDL were not in contact with oxLDL, we hypothesized that the persisting effect of HDL could result from HDL–cell contact, which may modulate intracellular ROS production or degradation.
Figure 4. HDLs inhibit the oxLDL-induced intracellular ROS increase, subsequent PTPase inactivation, and EGFR activation. A, Time course of the intracellular oxLDL-induced ROS increase and inhibition by HDLs. Cells were preincubated with (white circles) or without (black symbols) HDL (200 μg apoA/mL) for 18 hours. Then, oxLDLs (200 μg apoB/mL) were added to the preincubation medium (ie, preincubation and coincubation in the presence of HDL, white triangles) or after change of culture medium (ie, preincubation with HDL followed by incubation with only oxLDL, white squares). At the indicated time, cellular ROS were estimated by DHR fluorescence determination. B, Dose-dependence of the inhibitory effect of HDLs. Cells were preincubated for 18 hours with the indicated concentration of HDLs, then oxLDLs (200 μg apoB/mL) were added for 5 hours. ROS were determined as in (A). C, Inhibition of oxLDL-induced ROS increase requires intact apoAI and is blocked by anti–SRB1 antibody. Experiments were performed under the conditions of Figure 2B and 2C, using native HDLs (n), UV-oxidized HDLs (UV), HClO-oxidized HDLs (H), and anti–SRB1 antibody, when indicated. D, HDLs prevent the oxLDL-induced PTPase inactivation. Cells were preincubated with HDLs (200 μg apoA/mL) for 18 hours, then incubated with oxLDLs (200 μg apoB/mL) for the indicated time. PTPase activity in untreated controls (100%) ranged between 123 and 145 pmol/min per mg cell protein. E, HDLs prevent the intracellular oxLDL-induced EGFR activation under "preincubation only" and under "preincubation and coincubation" conditions used in (A). Representative of 3 experiments. A, B, and D, The results of 4 experiments are normalized (controls as 100%) and expressed as mean±SEM. *P<0.05.
HDLs Prevent the oxLDL-Induced H2O2 Increase Through a Catalase-Dependent Mechanism
As shown in Figure I (available online at http://atvb.ahajournals.org), HDL prevented both the intracellular ROS increase (Figure IA) and EGFR activation triggered by exogenous H2O2 (Figure IB), thus suggesting that HDLs stimulate the cellular degradation of H2O2. The degradation of oxLDL-induced ROS stimulated by HDLs was (partly) blocked by aminotriazole (a catalase inhibitor), but not by buthionine sulfoximine (a glutathione-depleting agent), and by mercaptosuccinate (a glutathione peroxidase inhibitor; Figure IC and ID). This suggests that HDL-induced ROS degradation is mediated by catalase. Subsequently, inhibition of oxLDL-induced EGFR activation by HDL was also catalase-dependent, but independent of glutathione and glutathione peroxidase (Figure IE). These data suggest that oxLDL-induced EGFR activation is mediated in part by H2O2 and that its inhibition by HDL is partly dependent on catalase activity.
oxLDLs Induce the EGFR-Mediated Upregulation of MMP-2 Inhibition by HDLs
MMPs play a critical role in extracellular matrix degradation, tissue remodeling, aneurysm formation, and plaque rupture.27–30 MMP-2 is implicated in SMC mitogenic signaling35 and intimal hyperplasia.36 The expression of MMPs is regulated by pro-inflammatory molecules and oxidized lipids.27,31,32 Moreover, EGFR is involved in the regulation of MMPs expression.26 Because MT1–MMP, which activates MMP-2 (gelatinase A), is expressed in atherosclerotic lesions and is upregulated by oxLDL,32 we investigated whether MMP-2 was also regulated and activated by oxLDL, and whether HDLs were able to prevent the increase of MT1–MMP and MMP-2. oxLDL (100 μg apolipoprotein B/mL) stimulated MMP-2 expression at 16 hours and MMP-9 expression at 36 hours (Figure 5A). We also observed the processing of pro-MT1–MMP in its active form, MT1–MMP (Figure 5B), in agreement with Rajavashisth et al.32 The upregulation of MMP-2 expression at 16 hours was associated with an increase of the activity of MMP released in the culture medium (Figure 5C). Pro–MMP-2 upregulation, MT1–MMP activation, and subsequent release of active MMP-2 were strongly inhibited by HDL and by AG-1478, a specific inhibitor of EGFR (Figure 5A to 5C). In contrast, the upregulation of pro–MMP-9 expression was only partly inhibited by HDL (Figure 5A and 5B) and was poorly inhibited by AG-1478. This suggests a role for EGFR in the activation of MT1–MMP and MMP-2. Altogether, these results indicate that the activation of SMCs by oxLDL enhances the expression of MMP-2 and the release of active MMP-2 through an EGFR-dependent mechanism, and that HDLs are effective in blocking these events.
Figure 5. HDLs prevent the oxLDL-induced upregulation of MMP-2 released in the culture medium. A, Expression of MMP-2 and MMP-9 in SMCs preincubated with or without HDLs (200 μg apoA/mL) in 1% fetal calf serum RPMI-1640 for 18 hours, then oxLDL (100 μg apoB/mL) were added for the indicated time. B, Expression and/or activation of MMP-9, MT1—MMP, and MMP-2 in SMCs preincubated with or without HDLs (200 μg apoA/mL for 18 hours) or AG-1478 (10 μmol/L for 2 hours), then oxLDLs (100 μg apoB/mL) were added for 16 hours, under the conditions of (A). C, The culture media were concentrated and used for fluorometric determination of MMP activity using DNP-pro-Leu-Gly-Leu-Trp-Ala-D-Arg-NH2 as substrate. A, Representative of 3 experiments. C, Mean±SEM of 4 experiments. *P<0.05.
Discussion
The major finding of the present study is that in vascular SMCs, HDLs are able to prevent HNE–protein adduct formation, oxidative stress, oxLDL-induced EGFR activation, and the subsequent upregulation and activation of MMP-2 (Figure II, available online only at http://atvb.ahajournals.org). oxLDL and oxidants are able to trigger a moderate and sustained activation of EGFR15–19,22 through (at least) 2 mechanisms, 4-HNE–EGFR adduct formation and intracellular ROS generation.18,19 EGFR is now considered as a convergence point in the complex signaling network that regulates cellular functions (such as cell growth, differentiation, motility, survival/death) and thereby plays a pivotal role in various pathophysiological processes.13–15 Besides its obvious role in cancer progression or metastasis,21 the role of EGFR in vascular biology, angiogenesis, and atherosclerosis is being recognized.3,20,23
Our data show, for the first time to our knowledge, that HDLs can prevent oxLDL-induced EGFR activation by inhibiting both the 4-HNE–EGFR adduct formation and the ROS-dependent mechanism (PTPase inactivation) involved in the sustained oxLDL-induced activation of EGFR. HDLs prevent HNE–protein adduct formation by acting probably through 2 mechanisms. First, HDLs are able to react with HNE37,38 and act like the aldehyde scavenger dinitrophenyl hydrazine (Figure 3). But, this "scavenger" activity of HDL required the presence of cells, because HDLs coincubated with oxLDL in cell-free medium did not scavenge the oxLDL-associated HNE (data not shown) and because the transfer of HNE from oxLDL to cell proteins is blocked in part by metabolic inhibitors.39 Moreover, HDLs, by acting through an antioxidant mechanism,33 inhibit the oxidative stress induced by oxLDL or H2O2 (Figures 4 and 5 ), thereby reducing the formation of peroxidation derivatives of cellular lipids and, subsequently, of aldehyde–protein adducts. Because lipid peroxidation derivatives are present in atherosclerotic lesions and are thought to play a role in atherogenesis,39,40–42 the inhibitory effect on the aldehyde–protein adduct formation might explain in part their anti-atherogenic effect. In our experimental model system, paraoxonase activity is not required for this "scavenger" activity of HDL, because HDL preparations with high or low paraoxonase activity and apolipoprotein A-I fraction with no detectable paraoxonase activity exhibited similar protective properties. Of course, our observation does not exclude that paraoxonases may act in vivo by other mechanisms.10
A second mechanism of action of HDL results from the inhibition of the increase of oxLDL-induced cellular ROS. This may result either from the inhibition of ROS biosynthesis or from increased ROS degradation. The latter hypothesis is supported by the fact that HDLs enhance H2O2 catabolism in cells incubated with exogenous H2O2. Moreover, this HDL-induced H2O2 degradation requires a cellular process, because in "preincubation only" experiments, HDL increase the resistance of cells against exogenous oxidative stress and because HDLs induce no significant H2O2 degradation in cell-free medium. This protective effect of HDL did not require glutathione and glutathione peroxidase activity, but was partly dependent on catalase activity, because the catalase inhibitor aminotriazole abrogated in part the protective effect of HDL. To date, the molecular mechanism by which HDL promotes the catalase-mediated H2O2 degradation remains unknown (under the used conditions, HDL did not induce any upregulation of the total cellular catalase activity; data not shown).
oxLDL and MMP are colocalized in atherosclerotic plaques and are thought to play a crucial role in vascular remodeling, pathogenesis of atherosclerosis, and plaque instability.27–29 Because several MMPs are upregulated by oxLDL in vascular cells,27,31,32 and by H2O243,44 or EGFR activation45 in cancer cells, this led us to investigate whether MMP upregulation was mediated through EGFR transactivation and whether HDLs were able to counter this MMP upregulation. oxLDL (16-hour incubation) induced the expression of MMP-2, whereas, at this time, MMP-9 was only poorly expressed. Interestingly, at the same time, we observed, in agreement with Rajavashisth,32 an increase of the active (processed) form of MT1–MMP, which cleaves and activates MMP-2. oxLDL-induced expression and activation of MT1–MMP and MMP-2 were coordinately regulated through EGFR transactivation, in agreement with Menashi et al.26 Both the oxLDL-induced expression and activation of MT1–MMP and MMP-2 were strongly inhibited by HDL (Figure II). In contrast, the regulation of the oxLDL-induced expression of MMP-9 (peaking at 36 hours) was not dependent on EGFR signaling and was only partly inhibited by HDL. From a pathophysiological point of view, this could be of importance in the pathogenesis of atherosclerosis, because MMPs are overexpressed in atherosclerotic lesions and could act in the atherosclerosis process.27,29 Active MMP-2 may degrade basement membrane collagen type IV and may favor the local desquamation of endothelial cells by lysing contacts with the underlying extracellular matrix, thereby participating in plaque erosion.27 Moreover, MMP-2 (with MMP-9 and urokinase plasminogen activator) may also participate in the excessive proteolysis of extracellular matrix and outward remodeling occurring in aneurysm.27 Finally, besides its role in extracellular matrix degradation, MMP-2 activation is implicated in SMC migration and proliferation, as shown in experimental neointimal hyperplasia36 and in oxLDL-induced proliferation mediated by the sphingomyelin/ceramide/sphingosine-1 phosphate pathway.35
The inhibitory effect of HDLs described here acts in concert with other protective effects. For instance, the same concentration of HDLs prevents the toxic effect of oxLDL.46 However, oxLDL-induced EGFR activation and toxicity are not causally related because EGFR activation was triggered by low nontoxic, as well as at high toxic, concentrations of oxLDL, and because both events are dissociated by inhibitors.47 All these protective effects of HDL may converge to inhibit "inflammatory" cell signaling triggered by oxLDL and, finally, to protect the integrity of the arterial wall. It is not excluded that the reported results, obtained under in vitro experimental conditions, may also play a role in atherogenesis, because accompanying events, such as 4-HNE–protein adduct formation,8,39 and MMP activation occur in atherosclerotic areas,27–29 and because HDLs are able to slow atherosclerotic lesion formation.4,11,12 However, the role of oxLDL-induced EGFR activation and subsequent MMP-2 activation in the atherogenic process remain to be evaluated in vivo.
Acknowledgments
This work was supported by grants from INSERM and University Paul Sabatier. The authors thank C. Mora and B. Bocquet for the technical assistance. F. Robbesyn was recipient of a fellowship from "Fondation pour la Recherche Médicale."
References
Brown MS, Goldstein JL. Lipoprotein metabolism in the macrophage: implication for cholesterol deposition in atherosclerosis. Annu Rev Biochem. 1983; 52: 223–261.
Lusis AJ. Atherosclerosis. Nature. 2000; 407: 233–241.
Ross R. Atherosclerosis - an inflammatory disease. N Engl J Med. 1999; 340: 115–126.
Witztum JL, Steinberg D. Role of oxidized LDL in atherogenesis. J Clin Invest. 1991; 88: 1785–1792.
Berliner JA, Heinecke JW. The role of oxidized lipoproteins in atherogenesis. Free Radic Biol Med. 1996; 20: 707–727.
Hajjar DP, Haberland ME. Lipoprotein trafficking in vascular cells. Molecular Trojan horses and cellular saboteurs. J Biol Chem. 1997; 272: 22975–22978.
Chisolm GM, Steinberg D. The oxidative modification hypothesis of atherogenesis: an overview. Free Radic Biol Med. 2000; 28: 1815–1826.
Assmann G, Nofer JR. Atheroprotective effects of HDL. Annu Rev Med. 2003; 54: 321–341.
Fielding CJ, Fielding PE. Molecular physiology of reverse cholesterol transport. J Lipid Res. 1995; 36: 211–228.
Navab M, Berliner JA, Subbanagounder G, Hama S, Lusis AJ, Van Lenten BJ, Vora D, Fogelman AM. HDL and the inflammatory response induced by LDL-derived oxidized phospholipids. Arterioscler Thromb Vasc Biol. 2001; 21: 481–488.
Libby P. Managing the risk of atherosclerosis: the role of HDL. Am J Cardiol. 2001; 88: 3N–8N.
Calabresi L, Gomaraschi M, Franceschini G. Endothelial protection by HDL: from bench to bedside. Arterioscler Thromb Vasc Biol. 2003; 23: 1724–1731.
Schlessinger J. Cell signaling by receptor tyrosine kinases. Cell. 2000; 103: 211–225.
Jorissen RN, Walker F, Pouliot N, Garrett TP, Ward CW, Burgess AW. EGF receptor: mechanisms of activation and signalling. Exp Cell Res. 2003; 284: 31–53.
Moghal N, Sternberg PW. Multiple positive and negative regulators of signaling by the EGF-receptor. Curr Opin Cell Biol. 1999; 11: 190–196.
Gamou S, Shimizu N. Hydrogen peroxide preferentially enhances the tyrosine phosphorylation of EGF receptor. FEBS Lett. 1995; 357: 161–164.
Knebel A, Rahmsdorf HJ, Ullrich A, Herrlich P. Dephosphorylation of receptor tyrosine kinases as target of regulation by radiation, oxidants or alkylating agents. EMBO J. 1996; 15: 5314–5325.
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.
Vacaresse N, Vieira O, Robbesyn F, Jurgens G, Salvayre R, Negre-Salvayre A. Phenolic antioxidants trolox and caffeic acid modulate the oxidized LDL-induced EGF-receptor activation. Br J Pharmacol. 2001; 132: 1777–1788.
Frank GD, Eguchi S. Activation of tyrosine kinases by reactive oxygen species in vascular smooth muscle cells: significance and involvement of EGF receptor transactivation by angiotensin II. Antioxid Redox Signal. 2003; 5: 771–780.
Prenzel N, Fischer OM, Streit S, Hart S, Ullrich A. The EGF receptor family as a central element for cellular signal transduction and diversification. Endocr Relat Cancer. 2001; 8: 11–31.
Kalmes A, Daum G, Clowes AW. EGFR transactivation in the regulation of SMC function. Ann N Y Acad Sci. 2001; 947: 42–54.
Berk BC. Vascular smooth muscle growth: autocrine growth mechanisms. Physiol Rev. 2001; 81: 999–1030.
Habib AA, Chatterjee S, Park SK, Ratan RR, Lefebvre S, Vartanian T. The EGF receptor engages receptor interacting protein and NF-kB-inducing kinase to activate NF-kB. J Biol Chem. 2001; 276: 8865–8874.
Monaco C, Andreakos E, Kiriakidis S, Mauri C, Bicknell C, Foxwell B, Cheshire N, Paleolog E, Feldmann M. Canonical pathway of NF-kB activation selectively regulates proinflammatory and prothrombotic responses in human atherosclerosis. Proc Natl Acad Sci U S A. 2004; 101: 5634–5639.
Menashi S, Serova M, Ma L, Vignot S, Mourah S, Calvo F. Regulation of extracellular matrix metalloproteinase inducer and matrix metalloproteinase expression by amphiregulin in transformed human breast epithelial cells. Cancer Res. 2003; 63: 7575–7580.
Galis ZS, Khatri JJ. Matrix metalloproteinases in vascular remodeling and atherogenesis: the good, the bad, and the ugly. Circ Res. 2002; 90: 251–262.
Lijnen HR. Extracellular proteolysis in the development and progression of atherosclerosis. Biochem Soc Trans. 2002; 30: 163–167.
Ikeda U, Shimada K. Matrix metalloproteinases and coronary artery diseases. Clin Cardiol. 2003; 26: 55–59.
Visse R, Nagase H. Matrix metalloproteinases and tissue inhibitors of metalloproteinases: structure, function, and biochemistry. Circ Res. 2003; 92: 827–839.
Huang Y, Mironova M, Lopes-Virella MF. Oxidized LDL stimulates MMP-1 expression in vascular endothelial cells. Arterioscler Thromb Vasc Biol. 1999; 19: 2640–2647.
Rajavashisth TB, Liao JK, Galis ZS, Tripathi S, Laufs U, Tripathi J, Chai NN, Xu XP, Jovinge S, Shah PK, Libby P. Inflammatory cytokines and oxidized LDL increase endothelial expression of MT1-MMP. J Biol Chem. 1999; 274: 11924–11929.
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. FASEB J. 2003; 17: 743–745.
Sullivan SG, Chiu DT, Errasfa M, Wang JM, Qi JS, Stern A. Effects of H2O2 on protein tyrosine phosphatase activity in HER14 cells. Free Radic Biol Med. 1994; 16: 399–403.
Auge N, Maupas-Schwalm F, Elbaz M, Thiers JC, Waysbort A, Itohara S, Krell HW, Salvayre R, Negre-Salvayre A. Role for matrix metalloproteinase-2 in oxidized LDL-induced activation of the sphingomyelin/ceramide pathway and smooth muscle cell proliferation. Circulation. 2004; 110: 571–578.
Kuzuya M, Kanda S, Sasaki T, Tamaya-Mori N, Cheng XW, Itoh T, Itohara S, Iguchi A. Deficiency of gelatinase A suppresses smooth muscle cell invasion and development of experimental intimal hyperplasia. Circulation. 2003; 108: 1375–1381.
McCall MR, Tang JY, Bielicki JK, Forte TM. Inhibition of lecithin-cholesterol acyltransferase and modification of HDL apolipoproteins by aldehydes. Arterioscler Thromb Vasc Biol. 1995; 15: 1599–1606.
Sangvanich P, Bharti Mackness B, Gaskell SJ. Paul Durrington P, Mackness M. The effect of HDL on the formation of lipid/protein conjugates during in vitro oxidation of LDL. Biochem Biophys Res Commun. 2003; 300: 501–506.
Escargueil-Blanc I, Salvayre R, Vacaresse N, Jürgens G, Darblade B, Arnal J, Parthasarathy S, Negre-Salvayre A. Mildly oxidized LDL induce activation of PDGF ?-receptor pathway. Circulation. 2001; 104: 1814–1821.
Baynes JW, Thorpe SR. Glycoxidation and lipoxidation in atherogenesis. Free Radic Biol Med. 2000; 28: 1708–1716.
Napoli C, de Nigris F, Palinski W. Multiple role of reactive oxygen species in the arterial wall. J Cell Biochem. 2001; 82: 674–682.
Uchida K. 4-Hydroxy-2-nonenal: a product and mediator of oxidative stress. Prog Lipid Res. 2003; 42: 318–343.
Belkhiri A, Richards C, Whaley M, McQueen SA, Orr FW. Increased expression of activated MMP-2 by human endothelial cells after sublethal H2O2 exposure. Lab Invest. 1997; 77: 533–539.
Yoon SO, Park SJ, Yoon SY, Yun CH, Chung AS. Sustained production of H2O2 activates pro-MMP-2 through receptor tyrosine kinases/phosphatidylinositol 3-kinase/NF-kB pathway. J Biol Chem. 2002; 277: 30271–30282.
Kheradmand F, Rishi K, Werb Z. Signaling through the EGF receptor controls lung morphogenesis in part by regulating MT1-MMP-mediated activation of gelatinase A/MMP2. J Cell Sci. 2002; 115: 839–848.
Suc I, Escargueil-Blanc I, Troly M, Salvayre R, Negre-Salvayre A. HDL and ApoA prevent cell death of endothelial cells induced by oxidized LDL. Arterioscler Thromb Vasc Biol. 1997; 17: 2158–2166.
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.(Fanny Robbesyn; Nathalie )
Correspondence to R. Salvayre or A. Negre-Salvayre, Biochimie, INSERM U466, IFR-31, CHU Rangueil, 1, avenue Jean Poulhès, TSA-50032, 31059 Toulouse Cedex 9, France. E-mail salvayre@toulouse.inserm.fr or anesalv@toulouse.inserm.fr
Abstract
Objectives— The atherogenic oxidized low-density lipoprotein (oxLDL) induces the formation of carbonyl-protein adducts and activates the endothelial growth factor receptor (EGFR) signaling pathway, which is now regarded as a central element for signal transduction. We aimed to investigate whether and by which mechanism the anti-atherogenic high-density lipoprotein (HDL) prevents these effects of oxLDL.
Methods and Results— In vascular cultured cells, HDL and apolipoprotein A-I inhibit oxLDL-induced EGFR activation and subsequent signaling by acting through 2 separate mechanisms. First, HDL, like the aldehyde scavenger dinitrophenyl hydrazine, prevented the formation of oxLDL-induced carbonyl–protein adducts and 4-hydroxynonenal (HNE)–EGFR adducts. Secondly, HDL enhanced the cellular antioxidant defenses by preventing (through a scavenger receptor class B-1 (SR-BI)–dependent mechanism) the increase of intracellular reactive oxygen species (ROS) and subsequent EGFR activation triggered by oxLDL or H2O2. A pharmacological approach suggests that this protective effect of HDL is independent of cellular glutathione level and glutathione peroxidase activity, but it requires catalase activity. Finally, we report that oxLDL upregulates both membrane type 1 (MT1)-matrix metalloproteinase-1 (MT1-MMP) and MMP-2 through an EGFR-dependent mechanism and that HDL inhibits these events.
Conclusions— HDLs block in vitro oxLDL-induced EGFR signaling and subsequent MMP-2 activation by inhibiting carbonyl adducts formation and cellular oxidative stress. These effects of HDL may participate to reduce cell activation, excessive remodeling, and alteration of the vascular wall.
Oxidized LDLs induce EGFR activation and subsequent MMP-2 activation. HDLs inhibit these events by 2 separate mechanisms, ie, by blocking carbonyl–protein adduct formation and by inhibiting the oxLDL-induced and H2O2-induced intracellular ROS increase, through a catalase-dependent process. This may contribute to reduce cell activation, excessive remodeling, and vascular wall alteration.
Key Words: atherosclerosis ? endothelial growth factors ? lipoproteins
Introduction
Low-density lipoproteins (LDLs) are thought to become atherogenic after undergoing oxidative modifications.1–3 Oxidized LDLs (oxLDLs) trigger cell signaling pathways involved in cellular responses, such as proliferation, inflammation, and apoptosis, which occur during atherogenesis.4–7 In contrast, high-density lipoproteins (HDLs) are anti-atherogenic and cardioprotective.8 Besides their role in the reverse transport of cholesterol, 9 HDL inhibits LDL oxidation and cell signaling mediated by oxLDL, and counters several adverse biological effects, such as cytotoxicity and inflammatory response triggered by cytokines, oxLDL, or oxidants.4,10–12 The molecular mechanism of this protective effect of HDL is only partly understood.
The endothelial growth factor receptor (EGFR) family consists of 4 members of tyrosine kinase receptors, EGFR (or ErbB1), HER2 (or ErbB2/neu), HER3 (or ErbB3), and ErbB4. EGFR is activated by binding of specific peptide ligands (EGF, transforming growth factor-, amphiregulin, neu differentiation factor [NDF], neuregulins, heregulins, heparin-binding epidermal growth factor [HB-EGF], betacellulin, and epiregulin);13–15 nonspecific and/or stress stimuli, such as ultraviolet and gamma radiations, alkylating agents signaling,16,17 oxidized lipids,18 fatty acids,19 and oxidants;20 and transactivation by several signaling pathways.21,22 For instance, cross-talk between G-protein–coupled receptor (for instance, angiotensin receptor-1 in vascular cells) and EGFR is mediated by cell signaling involving G-proteins, calcium, protein kinase C, Src, and metalloproteinases (such as a disintegrin and metalloprotease [ADAM]) that cleave pro-EGFR ligands, thereby liberating agonists of EGFR.21,22 Through these mechanisms, EGFR is a start point of a classical tyrosine kinase receptor signaling cascade, as well as a switch point of a wide cellular communication network.21,22 Therefore, EGFR acts as "a central element for signal transduction" and regulation of several cellular functions, such as cell growth, differentiation, motility, survival, and death,21 which also play a pivotal role in various pathophysiological processes. In addition to its prominent role in the development of several cancers,21 EGFR has been recently implicated in vascular pathophysiological processes associated with excessive remodeling and atherosclerosis.22,23 Among various biological effects, EGFR could play a role in cell migration,20–22 mitogenic signaling,23 and NF-B activation,24,25 and induces the upregulation of metalloproteinases (MMPs).26
MMPs are zinc proteases that are expressed in atherosclerotic plaques, where they may contribute to vascular remodeling and plaque disruption.27–29 MMPs are regulated through gene expression and activation of latent MMPs by cleavage of the N-terminal prosegment.30 oxLDL and inflammatory cytokines enhance the expression and activity of several MMPs, such as MMP-1, MMP-9 (gelatinase B), and membrane type 1 (MT1)-MMP (that cleaves and activates MMP-2) in vascular cells.27,31,32
The reported data show that HDLs are able to modulate the oxLDL-induced EGFR activation and subsequent upregulation of MMP-2 in vascular smooth muscle cells (SMCs).
Methods
A detailed Methods section is reported as an online-only data supplement (available at http://atvb.ahajournals.org).
Results
HDL Inhibits oxLDL-Induced EGFR Activation
Incubation of cells with oxLDL induced mild tyrosine-phosphorylation of EGFR that was inhibited by HDL (Figure 1A and 1B). Because similar data were obtained on vascular SMCs and endothelial cell vein (ECV)-304, we used, alternatively, ECV-304 cells when high EGFR level was required. Native LDL triggered no (or only minor) EGFR activation (Figure 1A). Under the used conditions, EGFR activation began at relatively low concentrations of oxLDL (25 to 50 μg apolipoprotein B/mL), was dose-dependent, and was completely inhibited by HDL in a dose-dependent manner (Figure 1B). Of note, 200 μg apolipoprotein B/mL (UV+Cu) oxLDL were not toxic over the 5 hours of the experiments (apoptosis becoming manifest only after 12-hour incubation). oxLDL-induced EGFR activation was independent of the apoptotic effect because it occurred at low, nontoxic, concentrations of (UV+Cu) oxLDL (25 to 50 μg apolipoprotein B/mL; UV) oxLDL (which are less toxic than [UV+Cu] oxLDL) induced also EGFR activation (data not shown).
Figure 1. HDLs inhibit the oxLDL-induced EGFR tyrosine phosphorylation and subsequent signaling. Cells (SMCs in A left, B, C; and ECV in A right and D) were preincubated with or without HDL (200 μg apolipoprotein A/mL, unless otherwise indicated) for 18 hours, before incubation with oxLDL (200 μg apolipoprotein B/mL), for 5 hours, and used for Western blots and labeled using the indicated antibodies. EGF (10 nmol/L for 15 minutes) was used as maximal positive control. A, EGFR tyrosine phosphorylation in SMCs incubated with oxLDL or native (nonoxidized) LDL (200 μg apolipoprotein B/mL) for 5 hours, after preincubation with HDLs (200 μg apoA/mL), when indicated (+). EGFR was immunoprecipitated (IP) by anti-EGFR antibody before Western blotting revealed by anti-phosphotyrosine (anti-PY) or anti-EGFR antibodies. B, Left panel shows the oxLDL-induced EGFR activation is dose-dependent (0 to 200 μg apolipoprotein B/mL of oxLDL) and is inhibited by HDLs (200 μg apolipoprotein A/mL). B, Right panel shows inhibitory effect of increasing HDL concentrations (0 to 200 μg apolipoprotein B/mL) on the oxLDL-induced EGFR activation. C, HDLs inhibit both the early (1-hour) and late (5-hour) components of oxLDL-induced EGFR activation (experimental conditions as in A). D, HDLs inhibit signaling subsequent to oxLDL-induced EGFR activation. Western blot of proteins coimmunoprecipitation (same experimental conditions as in A). Representative data of 3 to 5 experiments.
Because 2 phases of oxLDL-induced EGFR activation can be discriminated by their antioxidant susceptibility,19 we investigated whether HDL were effective on each component. Preincubation of cells with HDL prevented the early (1-hour) and late (5-hour) phases of oxLDL-induced EGFR activation (Figure 1C) and the subsequent signaling (Figure 1D).
The Protective Effect of HDL Is Mediated by Apolipoprotein A-I and SR-B1
The oxLDL-induced EGFR activation was inhibited by delipidated HDL apolipoprotein fraction and by apolipoprotein A-I (Figure 2A). Serum albumin, which is a frequent contaminant of HDL fractions (10% to 25% of total HDL-associated proteins) but is only minor in LDL fractions (<1% of apolipoprotein B content), did not prevent oxLDL-induced EGFR activation. The inhibitory effect of HDL was lost when apolipoprotein A-I was altered by HClO-mediated oxidation (containing 3.8±0.4 nmol thiobarbituric reactive substances [TBARS]/μg apolipoprotein A). In contrast, UV-oxidized HDL (2.7±0.3 nmol TBARS/μg apolipoprotein A) that exhibited only moderately altered apolipoprotein A-I were effective to prevent oxLDL-induced EGFR activation (Figure 2B). The protective effect of HDL was abolished by co-incubation of cells with HDL and anti-SRB1 antibodies (Figure 2C). This suggests that this effect of HDL requires the interaction between HDL and SRB1. Finally, paraoxonases were not implicated in this protective effect, because HDL fractions with very low paraoxonase activity (prepared in the presence of EDTA 100 μmol/L) and HDL fractions with high paraoxonase activity (prepared without EDTA) exhibited the same inhibitory effect (Figure 2D). These data strongly suggest that the protective effect of HDL against oxLDL-induced EGFR activation is mainly mediated through the interaction of apolipoprotein A-I with SRB1.
Figure 2. The protective effect of HDL is mediated by apolipoprotein A-I and SRB1. A, Protective effect of HDL compared with that of delipidated HDL apolipoproteins (apo), apoA-I (200 μg apoA/mL), and serum albumin (SA) (200 μg/mL). B, Left panel shows effect of native (nonoxidized [n]) HDL compared with that of UV-oxidized HDL (UV) and HClO-oxidized HDL (H) (200 μg apoA/mL). B, Right panel shows SDS-PAGE of apolipoproteins of native HDL, UV-oxidized HDL, HClO-oxidized HDL, and apoA-I fraction. C, Anti–SRB1 antibody (2 μg/mL) prevents the protective effect of HDLs. Cells were preincubated with or without HDL (200 μg apoA/mL) and/or anti–SRB1 antibody for 18 hours, before incubation with oxLDL, and Western blotting, as in Figure 1A. D, Inhibition of oxLDL-induced EGFR activation by HDL is independent of paraoxonase activity. HDLs, dialyzed against saline solution without (wo) or with (w) 100 μmol/L EDTA, were used to evaluate the protective effect (upper panels) and the paraoxonase activity (lower panel).
HDL Prevent the Formation of oxLDL-Induced Carbonyl–Protein Adducts and HNE–EGFR Adducts
oxLDL-induced EGFR activation results from HNE–EGFR adduct formation and reactive oxygen species (ROS) generation.18,19 This led to examining whether HDL may prevent the formation of HNE-EGFR adducts and other oxidized lipid–protein adducts. As shown in Figure 3A, HNE–EGFR adducts and EGFR activation induced by oxLDL were inhibited by HDL. More generally, HDL prevented the oxLDL-induced formation of carbonyl–protein adducts and the loss of [3H]NSP-reactive amino groups of cell proteins induced by oxLDL (Figure 3B and 3C). This effect was partly mimicked by the aldehyde scavenger dinitrophenyl hydrazine that prevented completely the formation of oxLDL-induced HNE–EGFR adducts and inhibited in part the oxLDL-induced EGFR activation (Figure 3A, right panel). Interestingly, Trolox, a hydrosoluble analog of tocopherol, induced a strong inhibition of EGFR autophosphorylation, but reduced only partly the level of HNE–EGFR adducts (Figure 3A). These data suggest that HDL are able to inhibit the oxLDL-induced EGFR activation (dinitrophenyl hydrazine-sensitive) mediated by HNE, and led us to hypothesize that HDL could also inhibit a second mechanism (Trolox-sensitive) involved in oxLDL-induced EGFR activation.
Figure 3. HDLs block the formation of HNE proteins and HNE–EGFR adducts induced by oxLDL. A, HDLs inhibit HNE–EGFR adduct formation. Cells were treated with or without HDL (200 μg apoA/mL, preincubation for 18 hours) and oxLDL (200 μg apoB/mL, incubation for 1 hour or 5 hours). Western blots were labeled using polyclonal anti-HNE protein, anti-PY, and anti-EGFR antibodies. Representative data of 4 experiments. B and C, HDLs prevent the oxLDL-induced formation of carbonyl adducts (B) and the decrease in free reactive amino groups (C) of cell proteins. Cells were preincubated with or without HDL (200 μg apoA/mL, 18 hours), then treated with oxLDL (200 μg apoB/mL, 5 hours) or with EGF (10 nmol/L, 15 minutes). Free amino groups determined by [3H]succinimidyl propionate ([3H]NSP) fixation, and carbonyls were expressed as percent of the value of controls. Mean±SEM of 4 experiments. *P<0.05.
HDL Inhibits oxLDL-Induced ROS Increase and Subsequent EGFR Activation
Because HDL inhibit an early increase of oxLDL-induced cellular ROS,33 and because ROS reinforce EGFR activation,16 we investigated whether HDL blocked the oxLDL-induced EGFR activation through a ROS-dependent mechanism. As reported in Figure 4, the oxLDL-induced cellular ROS increase (monitored by the fluorescence of oxidized dihydrorhodamine 123 [DHR] probe) was inhibited by HDL (Figure 4A) in a dose-dependent manner (Figure 4B). This inhibitory (protective) effect was lost with HClO-oxidized HDL, but not with UV-oxidized HDL (Figure 4C). Because apolipoprotein A-I is degraded in HClO-oxidized HDL, but is less altered in UV-oxidized HDL, it is suggested that this inhibitory effect of HDL requires the integrity of apolipoprotein A-I. Interestingly, preincubation of cells with anti-SRB1 antibody abrogated the protective effect of HDL (Figure 4C), thus suggesting that this effect is mediated through HDL/SRB1 interaction. Because EGFR activation by ROS and oxLDL results in part from the inactivation of phosphotyrosine phosphatases (PTPases),17,19,34 we examined whether HDLs were able to prevent the oxLDL-induced inhibition of PTPases. Under preincubation and coincubation conditions, HDL prevented effectively the oxLDL-induced inhibition of PTPases (Figure 4D). This protection of PTPases explains, at least in part, the protective effect of HDL against the oxLDL-induced EGFR activation. Interestingly, under "preincubation only" experimental conditions (cells were preincubated with HDL for 18 hours, then HDLs were removed just before oxLDL addition), HDLs were also able to block the oxLDL-induced ROS increase (Figure 4A) and EGFR activation (Figure 4E). Because, under "preincubation only" conditions, HDL were not in contact with oxLDL, we hypothesized that the persisting effect of HDL could result from HDL–cell contact, which may modulate intracellular ROS production or degradation.
Figure 4. HDLs inhibit the oxLDL-induced intracellular ROS increase, subsequent PTPase inactivation, and EGFR activation. A, Time course of the intracellular oxLDL-induced ROS increase and inhibition by HDLs. Cells were preincubated with (white circles) or without (black symbols) HDL (200 μg apoA/mL) for 18 hours. Then, oxLDLs (200 μg apoB/mL) were added to the preincubation medium (ie, preincubation and coincubation in the presence of HDL, white triangles) or after change of culture medium (ie, preincubation with HDL followed by incubation with only oxLDL, white squares). At the indicated time, cellular ROS were estimated by DHR fluorescence determination. B, Dose-dependence of the inhibitory effect of HDLs. Cells were preincubated for 18 hours with the indicated concentration of HDLs, then oxLDLs (200 μg apoB/mL) were added for 5 hours. ROS were determined as in (A). C, Inhibition of oxLDL-induced ROS increase requires intact apoAI and is blocked by anti–SRB1 antibody. Experiments were performed under the conditions of Figure 2B and 2C, using native HDLs (n), UV-oxidized HDLs (UV), HClO-oxidized HDLs (H), and anti–SRB1 antibody, when indicated. D, HDLs prevent the oxLDL-induced PTPase inactivation. Cells were preincubated with HDLs (200 μg apoA/mL) for 18 hours, then incubated with oxLDLs (200 μg apoB/mL) for the indicated time. PTPase activity in untreated controls (100%) ranged between 123 and 145 pmol/min per mg cell protein. E, HDLs prevent the intracellular oxLDL-induced EGFR activation under "preincubation only" and under "preincubation and coincubation" conditions used in (A). Representative of 3 experiments. A, B, and D, The results of 4 experiments are normalized (controls as 100%) and expressed as mean±SEM. *P<0.05.
HDLs Prevent the oxLDL-Induced H2O2 Increase Through a Catalase-Dependent Mechanism
As shown in Figure I (available online at http://atvb.ahajournals.org), HDL prevented both the intracellular ROS increase (Figure IA) and EGFR activation triggered by exogenous H2O2 (Figure IB), thus suggesting that HDLs stimulate the cellular degradation of H2O2. The degradation of oxLDL-induced ROS stimulated by HDLs was (partly) blocked by aminotriazole (a catalase inhibitor), but not by buthionine sulfoximine (a glutathione-depleting agent), and by mercaptosuccinate (a glutathione peroxidase inhibitor; Figure IC and ID). This suggests that HDL-induced ROS degradation is mediated by catalase. Subsequently, inhibition of oxLDL-induced EGFR activation by HDL was also catalase-dependent, but independent of glutathione and glutathione peroxidase (Figure IE). These data suggest that oxLDL-induced EGFR activation is mediated in part by H2O2 and that its inhibition by HDL is partly dependent on catalase activity.
oxLDLs Induce the EGFR-Mediated Upregulation of MMP-2 Inhibition by HDLs
MMPs play a critical role in extracellular matrix degradation, tissue remodeling, aneurysm formation, and plaque rupture.27–30 MMP-2 is implicated in SMC mitogenic signaling35 and intimal hyperplasia.36 The expression of MMPs is regulated by pro-inflammatory molecules and oxidized lipids.27,31,32 Moreover, EGFR is involved in the regulation of MMPs expression.26 Because MT1–MMP, which activates MMP-2 (gelatinase A), is expressed in atherosclerotic lesions and is upregulated by oxLDL,32 we investigated whether MMP-2 was also regulated and activated by oxLDL, and whether HDLs were able to prevent the increase of MT1–MMP and MMP-2. oxLDL (100 μg apolipoprotein B/mL) stimulated MMP-2 expression at 16 hours and MMP-9 expression at 36 hours (Figure 5A). We also observed the processing of pro-MT1–MMP in its active form, MT1–MMP (Figure 5B), in agreement with Rajavashisth et al.32 The upregulation of MMP-2 expression at 16 hours was associated with an increase of the activity of MMP released in the culture medium (Figure 5C). Pro–MMP-2 upregulation, MT1–MMP activation, and subsequent release of active MMP-2 were strongly inhibited by HDL and by AG-1478, a specific inhibitor of EGFR (Figure 5A to 5C). In contrast, the upregulation of pro–MMP-9 expression was only partly inhibited by HDL (Figure 5A and 5B) and was poorly inhibited by AG-1478. This suggests a role for EGFR in the activation of MT1–MMP and MMP-2. Altogether, these results indicate that the activation of SMCs by oxLDL enhances the expression of MMP-2 and the release of active MMP-2 through an EGFR-dependent mechanism, and that HDLs are effective in blocking these events.
Figure 5. HDLs prevent the oxLDL-induced upregulation of MMP-2 released in the culture medium. A, Expression of MMP-2 and MMP-9 in SMCs preincubated with or without HDLs (200 μg apoA/mL) in 1% fetal calf serum RPMI-1640 for 18 hours, then oxLDL (100 μg apoB/mL) were added for the indicated time. B, Expression and/or activation of MMP-9, MT1—MMP, and MMP-2 in SMCs preincubated with or without HDLs (200 μg apoA/mL for 18 hours) or AG-1478 (10 μmol/L for 2 hours), then oxLDLs (100 μg apoB/mL) were added for 16 hours, under the conditions of (A). C, The culture media were concentrated and used for fluorometric determination of MMP activity using DNP-pro-Leu-Gly-Leu-Trp-Ala-D-Arg-NH2 as substrate. A, Representative of 3 experiments. C, Mean±SEM of 4 experiments. *P<0.05.
Discussion
The major finding of the present study is that in vascular SMCs, HDLs are able to prevent HNE–protein adduct formation, oxidative stress, oxLDL-induced EGFR activation, and the subsequent upregulation and activation of MMP-2 (Figure II, available online only at http://atvb.ahajournals.org). oxLDL and oxidants are able to trigger a moderate and sustained activation of EGFR15–19,22 through (at least) 2 mechanisms, 4-HNE–EGFR adduct formation and intracellular ROS generation.18,19 EGFR is now considered as a convergence point in the complex signaling network that regulates cellular functions (such as cell growth, differentiation, motility, survival/death) and thereby plays a pivotal role in various pathophysiological processes.13–15 Besides its obvious role in cancer progression or metastasis,21 the role of EGFR in vascular biology, angiogenesis, and atherosclerosis is being recognized.3,20,23
Our data show, for the first time to our knowledge, that HDLs can prevent oxLDL-induced EGFR activation by inhibiting both the 4-HNE–EGFR adduct formation and the ROS-dependent mechanism (PTPase inactivation) involved in the sustained oxLDL-induced activation of EGFR. HDLs prevent HNE–protein adduct formation by acting probably through 2 mechanisms. First, HDLs are able to react with HNE37,38 and act like the aldehyde scavenger dinitrophenyl hydrazine (Figure 3). But, this "scavenger" activity of HDL required the presence of cells, because HDLs coincubated with oxLDL in cell-free medium did not scavenge the oxLDL-associated HNE (data not shown) and because the transfer of HNE from oxLDL to cell proteins is blocked in part by metabolic inhibitors.39 Moreover, HDLs, by acting through an antioxidant mechanism,33 inhibit the oxidative stress induced by oxLDL or H2O2 (Figures 4 and 5 ), thereby reducing the formation of peroxidation derivatives of cellular lipids and, subsequently, of aldehyde–protein adducts. Because lipid peroxidation derivatives are present in atherosclerotic lesions and are thought to play a role in atherogenesis,39,40–42 the inhibitory effect on the aldehyde–protein adduct formation might explain in part their anti-atherogenic effect. In our experimental model system, paraoxonase activity is not required for this "scavenger" activity of HDL, because HDL preparations with high or low paraoxonase activity and apolipoprotein A-I fraction with no detectable paraoxonase activity exhibited similar protective properties. Of course, our observation does not exclude that paraoxonases may act in vivo by other mechanisms.10
A second mechanism of action of HDL results from the inhibition of the increase of oxLDL-induced cellular ROS. This may result either from the inhibition of ROS biosynthesis or from increased ROS degradation. The latter hypothesis is supported by the fact that HDLs enhance H2O2 catabolism in cells incubated with exogenous H2O2. Moreover, this HDL-induced H2O2 degradation requires a cellular process, because in "preincubation only" experiments, HDL increase the resistance of cells against exogenous oxidative stress and because HDLs induce no significant H2O2 degradation in cell-free medium. This protective effect of HDL did not require glutathione and glutathione peroxidase activity, but was partly dependent on catalase activity, because the catalase inhibitor aminotriazole abrogated in part the protective effect of HDL. To date, the molecular mechanism by which HDL promotes the catalase-mediated H2O2 degradation remains unknown (under the used conditions, HDL did not induce any upregulation of the total cellular catalase activity; data not shown).
oxLDL and MMP are colocalized in atherosclerotic plaques and are thought to play a crucial role in vascular remodeling, pathogenesis of atherosclerosis, and plaque instability.27–29 Because several MMPs are upregulated by oxLDL in vascular cells,27,31,32 and by H2O243,44 or EGFR activation45 in cancer cells, this led us to investigate whether MMP upregulation was mediated through EGFR transactivation and whether HDLs were able to counter this MMP upregulation. oxLDL (16-hour incubation) induced the expression of MMP-2, whereas, at this time, MMP-9 was only poorly expressed. Interestingly, at the same time, we observed, in agreement with Rajavashisth,32 an increase of the active (processed) form of MT1–MMP, which cleaves and activates MMP-2. oxLDL-induced expression and activation of MT1–MMP and MMP-2 were coordinately regulated through EGFR transactivation, in agreement with Menashi et al.26 Both the oxLDL-induced expression and activation of MT1–MMP and MMP-2 were strongly inhibited by HDL (Figure II). In contrast, the regulation of the oxLDL-induced expression of MMP-9 (peaking at 36 hours) was not dependent on EGFR signaling and was only partly inhibited by HDL. From a pathophysiological point of view, this could be of importance in the pathogenesis of atherosclerosis, because MMPs are overexpressed in atherosclerotic lesions and could act in the atherosclerosis process.27,29 Active MMP-2 may degrade basement membrane collagen type IV and may favor the local desquamation of endothelial cells by lysing contacts with the underlying extracellular matrix, thereby participating in plaque erosion.27 Moreover, MMP-2 (with MMP-9 and urokinase plasminogen activator) may also participate in the excessive proteolysis of extracellular matrix and outward remodeling occurring in aneurysm.27 Finally, besides its role in extracellular matrix degradation, MMP-2 activation is implicated in SMC migration and proliferation, as shown in experimental neointimal hyperplasia36 and in oxLDL-induced proliferation mediated by the sphingomyelin/ceramide/sphingosine-1 phosphate pathway.35
The inhibitory effect of HDLs described here acts in concert with other protective effects. For instance, the same concentration of HDLs prevents the toxic effect of oxLDL.46 However, oxLDL-induced EGFR activation and toxicity are not causally related because EGFR activation was triggered by low nontoxic, as well as at high toxic, concentrations of oxLDL, and because both events are dissociated by inhibitors.47 All these protective effects of HDL may converge to inhibit "inflammatory" cell signaling triggered by oxLDL and, finally, to protect the integrity of the arterial wall. It is not excluded that the reported results, obtained under in vitro experimental conditions, may also play a role in atherogenesis, because accompanying events, such as 4-HNE–protein adduct formation,8,39 and MMP activation occur in atherosclerotic areas,27–29 and because HDLs are able to slow atherosclerotic lesion formation.4,11,12 However, the role of oxLDL-induced EGFR activation and subsequent MMP-2 activation in the atherogenic process remain to be evaluated in vivo.
Acknowledgments
This work was supported by grants from INSERM and University Paul Sabatier. The authors thank C. Mora and B. Bocquet for the technical assistance. F. Robbesyn was recipient of a fellowship from "Fondation pour la Recherche Médicale."
References
Brown MS, Goldstein JL. Lipoprotein metabolism in the macrophage: implication for cholesterol deposition in atherosclerosis. Annu Rev Biochem. 1983; 52: 223–261.
Lusis AJ. Atherosclerosis. Nature. 2000; 407: 233–241.
Ross R. Atherosclerosis - an inflammatory disease. N Engl J Med. 1999; 340: 115–126.
Witztum JL, Steinberg D. Role of oxidized LDL in atherogenesis. J Clin Invest. 1991; 88: 1785–1792.
Berliner JA, Heinecke JW. The role of oxidized lipoproteins in atherogenesis. Free Radic Biol Med. 1996; 20: 707–727.
Hajjar DP, Haberland ME. Lipoprotein trafficking in vascular cells. Molecular Trojan horses and cellular saboteurs. J Biol Chem. 1997; 272: 22975–22978.
Chisolm GM, Steinberg D. The oxidative modification hypothesis of atherogenesis: an overview. Free Radic Biol Med. 2000; 28: 1815–1826.
Assmann G, Nofer JR. Atheroprotective effects of HDL. Annu Rev Med. 2003; 54: 321–341.
Fielding CJ, Fielding PE. Molecular physiology of reverse cholesterol transport. J Lipid Res. 1995; 36: 211–228.
Navab M, Berliner JA, Subbanagounder G, Hama S, Lusis AJ, Van Lenten BJ, Vora D, Fogelman AM. HDL and the inflammatory response induced by LDL-derived oxidized phospholipids. Arterioscler Thromb Vasc Biol. 2001; 21: 481–488.
Libby P. Managing the risk of atherosclerosis: the role of HDL. Am J Cardiol. 2001; 88: 3N–8N.
Calabresi L, Gomaraschi M, Franceschini G. Endothelial protection by HDL: from bench to bedside. Arterioscler Thromb Vasc Biol. 2003; 23: 1724–1731.
Schlessinger J. Cell signaling by receptor tyrosine kinases. Cell. 2000; 103: 211–225.
Jorissen RN, Walker F, Pouliot N, Garrett TP, Ward CW, Burgess AW. EGF receptor: mechanisms of activation and signalling. Exp Cell Res. 2003; 284: 31–53.
Moghal N, Sternberg PW. Multiple positive and negative regulators of signaling by the EGF-receptor. Curr Opin Cell Biol. 1999; 11: 190–196.
Gamou S, Shimizu N. Hydrogen peroxide preferentially enhances the tyrosine phosphorylation of EGF receptor. FEBS Lett. 1995; 357: 161–164.
Knebel A, Rahmsdorf HJ, Ullrich A, Herrlich P. Dephosphorylation of receptor tyrosine kinases as target of regulation by radiation, oxidants or alkylating agents. EMBO J. 1996; 15: 5314–5325.
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.
Vacaresse N, Vieira O, Robbesyn F, Jurgens G, Salvayre R, Negre-Salvayre A. Phenolic antioxidants trolox and caffeic acid modulate the oxidized LDL-induced EGF-receptor activation. Br J Pharmacol. 2001; 132: 1777–1788.
Frank GD, Eguchi S. Activation of tyrosine kinases by reactive oxygen species in vascular smooth muscle cells: significance and involvement of EGF receptor transactivation by angiotensin II. Antioxid Redox Signal. 2003; 5: 771–780.
Prenzel N, Fischer OM, Streit S, Hart S, Ullrich A. The EGF receptor family as a central element for cellular signal transduction and diversification. Endocr Relat Cancer. 2001; 8: 11–31.
Kalmes A, Daum G, Clowes AW. EGFR transactivation in the regulation of SMC function. Ann N Y Acad Sci. 2001; 947: 42–54.
Berk BC. Vascular smooth muscle growth: autocrine growth mechanisms. Physiol Rev. 2001; 81: 999–1030.
Habib AA, Chatterjee S, Park SK, Ratan RR, Lefebvre S, Vartanian T. The EGF receptor engages receptor interacting protein and NF-kB-inducing kinase to activate NF-kB. J Biol Chem. 2001; 276: 8865–8874.
Monaco C, Andreakos E, Kiriakidis S, Mauri C, Bicknell C, Foxwell B, Cheshire N, Paleolog E, Feldmann M. Canonical pathway of NF-kB activation selectively regulates proinflammatory and prothrombotic responses in human atherosclerosis. Proc Natl Acad Sci U S A. 2004; 101: 5634–5639.
Menashi S, Serova M, Ma L, Vignot S, Mourah S, Calvo F. Regulation of extracellular matrix metalloproteinase inducer and matrix metalloproteinase expression by amphiregulin in transformed human breast epithelial cells. Cancer Res. 2003; 63: 7575–7580.
Galis ZS, Khatri JJ. Matrix metalloproteinases in vascular remodeling and atherogenesis: the good, the bad, and the ugly. Circ Res. 2002; 90: 251–262.
Lijnen HR. Extracellular proteolysis in the development and progression of atherosclerosis. Biochem Soc Trans. 2002; 30: 163–167.
Ikeda U, Shimada K. Matrix metalloproteinases and coronary artery diseases. Clin Cardiol. 2003; 26: 55–59.
Visse R, Nagase H. Matrix metalloproteinases and tissue inhibitors of metalloproteinases: structure, function, and biochemistry. Circ Res. 2003; 92: 827–839.
Huang Y, Mironova M, Lopes-Virella MF. Oxidized LDL stimulates MMP-1 expression in vascular endothelial cells. Arterioscler Thromb Vasc Biol. 1999; 19: 2640–2647.
Rajavashisth TB, Liao JK, Galis ZS, Tripathi S, Laufs U, Tripathi J, Chai NN, Xu XP, Jovinge S, Shah PK, Libby P. Inflammatory cytokines and oxidized LDL increase endothelial expression of MT1-MMP. J Biol Chem. 1999; 274: 11924–11929.
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. FASEB J. 2003; 17: 743–745.
Sullivan SG, Chiu DT, Errasfa M, Wang JM, Qi JS, Stern A. Effects of H2O2 on protein tyrosine phosphatase activity in HER14 cells. Free Radic Biol Med. 1994; 16: 399–403.
Auge N, Maupas-Schwalm F, Elbaz M, Thiers JC, Waysbort A, Itohara S, Krell HW, Salvayre R, Negre-Salvayre A. Role for matrix metalloproteinase-2 in oxidized LDL-induced activation of the sphingomyelin/ceramide pathway and smooth muscle cell proliferation. Circulation. 2004; 110: 571–578.
Kuzuya M, Kanda S, Sasaki T, Tamaya-Mori N, Cheng XW, Itoh T, Itohara S, Iguchi A. Deficiency of gelatinase A suppresses smooth muscle cell invasion and development of experimental intimal hyperplasia. Circulation. 2003; 108: 1375–1381.
McCall MR, Tang JY, Bielicki JK, Forte TM. Inhibition of lecithin-cholesterol acyltransferase and modification of HDL apolipoproteins by aldehydes. Arterioscler Thromb Vasc Biol. 1995; 15: 1599–1606.
Sangvanich P, Bharti Mackness B, Gaskell SJ. Paul Durrington P, Mackness M. The effect of HDL on the formation of lipid/protein conjugates during in vitro oxidation of LDL. Biochem Biophys Res Commun. 2003; 300: 501–506.
Escargueil-Blanc I, Salvayre R, Vacaresse N, Jürgens G, Darblade B, Arnal J, Parthasarathy S, Negre-Salvayre A. Mildly oxidized LDL induce activation of PDGF ?-receptor pathway. Circulation. 2001; 104: 1814–1821.
Baynes JW, Thorpe SR. Glycoxidation and lipoxidation in atherogenesis. Free Radic Biol Med. 2000; 28: 1708–1716.
Napoli C, de Nigris F, Palinski W. Multiple role of reactive oxygen species in the arterial wall. J Cell Biochem. 2001; 82: 674–682.
Uchida K. 4-Hydroxy-2-nonenal: a product and mediator of oxidative stress. Prog Lipid Res. 2003; 42: 318–343.
Belkhiri A, Richards C, Whaley M, McQueen SA, Orr FW. Increased expression of activated MMP-2 by human endothelial cells after sublethal H2O2 exposure. Lab Invest. 1997; 77: 533–539.
Yoon SO, Park SJ, Yoon SY, Yun CH, Chung AS. Sustained production of H2O2 activates pro-MMP-2 through receptor tyrosine kinases/phosphatidylinositol 3-kinase/NF-kB pathway. J Biol Chem. 2002; 277: 30271–30282.
Kheradmand F, Rishi K, Werb Z. Signaling through the EGF receptor controls lung morphogenesis in part by regulating MT1-MMP-mediated activation of gelatinase A/MMP2. J Cell Sci. 2002; 115: 839–848.
Suc I, Escargueil-Blanc I, Troly M, Salvayre R, Negre-Salvayre A. HDL and ApoA prevent cell death of endothelial cells induced by oxidized LDL. Arterioscler Thromb Vasc Biol. 1997; 17: 2158–2166.
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.(Fanny Robbesyn; Nathalie )