Two Distinct Calcium-Dependent Mitochondrial Pathways Are Involved in Oxidized LDL-Induced Apoptosis
http://www.100md.com
动脉硬化血栓血管生物学 2005年第3期
From INSERM U-466 and the Department of Biochemistry and Molecular Biology, IFR-31, CHU Rangueil, Toulouse, France.
Correspondence to Prof R. Salvayre, Biochimie, Inserm U-466, IFR-31, CHU Rangueil 1, avenue Jean Poulhès, TSA-0032, 31059 Toulouse Cedex 9, France. E-mail salvayre@toulouse.inserm.fr
Abstract
Objective— Oxidized low-density lipoprotein (oxLDL)-induced apoptosis of vascular endothelial cells may contribute to plaque erosion and rupture. We aimed to clarify the relationship between the oxLDL-induced calcium signal and induction of apoptotic pathways.
Methods and Results— Apoptosis was evaluated by biochemical methods, including studies of enzyme activities, protein processing, release of proapoptotic factors, chromatin cleavage, and especially by morphological methods that evaluate apoptosis/necrosis by SYTO-13/propidium iodide fluorescent labeling. The oxLDL-induced sustained calcium rise activated 2 distinct calcium-dependent mitochondrial apoptotic pathways in human microvascular endothelial cells. OxLDLs induced calpain activation and subsequent Bid cleavage and cytochrome C release, which were blocked by calpeptin. Cyclosporin-A inhibited cytochrome C release, possibly by inhibiting the opening of the mitochondrial permeability transition pore (mPTP). Calcineurin, another cyclosporin-sensitive step, was not implicated, because oxLDLs inhibited calcineurin and FK-506 treatment was ineffective. Cytochrome C release in turn induced caspase-3 activation. In addition, oxLDLs triggered release and nuclear translocation of mitochondrial apoptosis-inducing factor through a mechanism dependent on calcium but independent of calpains, mPTP, and caspases.
Conclusions— OxLDL-induced apoptosis involves 2 distinct calcium-dependent pathways, the first mediated by calpain/mPTP/cytochrome C/caspase-3 and the second mediated by apoptosis-inducing factor, which is cyclosporin-insensitive and caspase-independent.
We investigated the potential relationship between calcium signaling, proteolytic cascade, and mitochondrial apoptotic pathways involved in oxidized low-density lipoprotein (oxLDL)-induced apoptosis. OxLDL-induced apoptosis was mediated by 2 distinct mitochondrial pathways, one involving calcium/calpain/mitochondrial permeability transition pore/cytochrome C/caspase-3 and the other calcium-dependent release of apoptosis inducing factor.
Key Words: calpain ? caspase ? mitochondria ? apoptosis-inducing factor ? oxidized low-density lipoprotein ? atherosclerosis
Introduction
Atherogenesis is characterized by lipid deposition, a chronic inflammatory response, and chronic wound healing processes.1,2 Apoptosis may play a role in endothelial cell lining defects, necrotic core formation, or plaque erosion and rupture.3–5 Among the variety of proapoptotic factors present in atherosclerotic plaques, oxidized low-density lipoproteins (oxLDLs) are thought to play a crucial role by concomitantly inducing lipid storage, local inflammation, and toxic events.4,5–8 OxLDLs trigger apoptosis or necrosis of cultured vascular cells7–9 and may therefore participate in vascular wall injury, plaque erosion/rupture, and subsequent athero-thrombotic events.4,5
The proapoptotic effects of oxLDLs are mediated through a complex sequence of signaling events that lead to activation of several caspase-dependent or -independent apoptotic pathways.8,9 Two separate caspase-dependent apoptotic pathways have been implicated in oxLDL-induced apoptosis.7–9 The extrinsic apoptotic pathway, mediated by death receptors, Fas, and/or tumor necrosis factor receptor (TNFR) and downstream by caspase-8/caspase-3, is involved in oxLDL-induced apoptosis in endothelial cells.8,10,11 However, a recent report contests this hypothesis.12 The intrinsic mitochondrial apoptotic pathway, involving bcl-2 family members, cytochrome C, and caspase-3, is especially activated by oxLDLs.8,13 Moreover, a sustained rise of cytosolic calcium plays a central role in oxLDL-induced apoptosis/necrosis,14,15 possibly by activating the calcium-dependent calpains and, subsequently, Bid cleavage and cytochrome C release.15 However, whereas Porn-Ares et al15 found that oxLDLs induce Bid truncation without any activation of caspase-3, Chen et al12 observed that oxLDLs did not cause Bid truncation but activated caspase-3 in agreement with Dimmeler et al.16 Furthermore, oxLDLs can activate a caspase-independent apoptotic pathway mediated by the release of apoptosis-inducing factor (AIF) from mitochondria.17
Because data in the literature on mechanisms implicated in oxLDL-induced apoptosis are controversial and because calcium can trigger multiple apoptotic pathways,18 we presently investigated the potential link between oxLDL-induced calcium signaling and the activation of mitochondrial apoptotic pathways.
Methods
Extensive Materials and Methods are available online at http://atvb.ahajournals.org.
Cell Culture
Human microvascular endothelial cells (HMEC-1) were starved in serum-free medium for 24 hours before LDL treatment.
LDL Isolation and Oxidation
LDLs were prepared as previously indicated.14 Under standard conditions, oxLDLs contained 71 to 104 nmol lipid hydroperoxide per mg apoB and 6.4 to 9.7 nmol of thiobarbituric acid-reactive substances (TBARS) per mg apoB.
Cytotoxicity, Necrosis, and Apoptosis
Cytotoxicity was evaluated by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide test14 and cell lysis (necrosis) by lactate dehydrogenase (LDH) release. Apoptotic and necrotic cells were counted by fluorescence microscopy after staining by SYTO-13 and propidium iodide.19,20
Chromatin Fragments and DNA Ladder
Chromatin fragments were determined by the procedure of McConkey et al.21 DNA laddering was visualized on agarose electrophoresis.
Cell Fractionation
Cytosol was separated from mitochondria by a method adapted from De Duve et al.22
Enzyme Activities
Enzyme activities were determined using fluorogenic substrates.23–25
Western Blot Analysis
Cell extracts were subjected to SDS-PAGE, as previously reported.26
Immunocytochemistry
After the indicated treatment, cells were fixed, incubated with anti-AIF antibody, and revealed with fluorescein isothiocyanate-conjugated secondary antibody.
Statistical Analysis
Data are given as mean±SEM, and statistical significance was evaluated by 1-way ANOVA (Tukey test, SigmaStat software).
Results
The toxicity of oxLDLs was time- and dose-dependent and correlated with the extent of LDL oxidation.14 OxLDLs obtained by UV+copper/EDTA oxidation or by cell-mediated oxidation exhibited similar toxicity to cultured cells.14,26 We used preferably (UV+copper/EDTA)-oxLDLs, because cell-mediated oxLDL preparations may contain mediators secreted by cells that may interfere with oxLDL toxicity.
Time Course of the Effects of oxLDLs on the Calcium-Dependent Mitochondrial Apoptotic Pathway and Apoptosis
Figure I (available online at http://atvb.ahajournals.org) shows the time course of calpain activation and -spectrin breakdown, Bid degradation, release of cytochrome C, and caspase-3 activation (Figure IA through ID). Cell death (Figure IE) occurred mainly by apoptosis, as shown by chromatin fragmentation, DNA laddering, and morphological changes visualized by SYTO-13/PI fluorescent staining (Figure IF through II). The level of primary necrosis was very low, as shown by low levels of LDH release (Figure IE) and by small numbers of PI-stained cells with the morphological features of primary necrosis (Figure IH and II).
Calcium/Calpain Mediates oxLDL-Induced Bid Cleavage, Cytochrome C Release, and Subsequent Caspase-3 Activation
When calcium in the culture medium was chelated by EGTA, under conditions blocking the rise in cytosolic calcium,14 oxLDL-induced calpain activation, cytochrome C release, caspase-3 activation, chromatin fragmentation, and toxicity were inhibited (Figure 1A through 1E). It may be noted that EGTA did not inhibit apoptosis induced by taxol, a potent microtubule-damaging agent, when used as positive control.
Figure 1. EGTA blocks oxLDL-induced activation of calpains, mitochondrial apoptotic pathway, caspase-3, and subsequent toxicity. Cells were incubated in the presence or absence of 0.4 mmol/L EGTA and with or without oxLDLs (200 μg apoB per mL) for 16 hours (A, C, and D), or at the indicated time (B), or for 24 hours (D and E). Cells lysates were used for fluorometric determination of calpain activity (A). Western blots of cytochrome C were performed on cytosolic fractions (B). Cell lysates were used for the fluorometric measurement of DEVDase and Western blots of caspase-3 (C) and chromatin fragmentation (D). Cell viability and necrosis (E) were evaluated at 24 hours using the McConkey procedure and 4,'6-diamidino-2-phenylindole fluorescent staining (D), by the MTT test, and by leakage of LDH in the culture medium (the basal LDH level was very low, 5% of the total cellular LDH; E). In A, C, D, and E, mean±SEM was of at least 4 experiments (*P<0.05). In B and C, Western blots are representative of 4 experiments.
To investigate whether calpain activation is involved in oxLDL-induced apoptosis, we used calpeptin, a selective inhibitor of calpains. Calpeptin inhibited oxLDL-induced calpain activation (Figure 2A) and prevented Bid cleavage, release of cytochrome C, caspase-3 activation (Figure 2B through 2D), and both chromatin fragmentation and cytotoxicity by oxLDLs (Figure 2E through 2F).
Figure 2. The calpain inhibitor calpeptin inhibits oxLDL-induced activation of calpains, apoptotic mitochondrial pathways, caspase-3 activation, and subsequent apoptotic events. Cells were treated with or without 2.5 μmol/L calpeptin, or 25 μmol/L zVAD-fmk, or 25 μmol/L zVDVAD-fmk, and added at time 0 with 200 μg apoB per mL oxLDLs. After 20 hours or after the indicated time of incubation, cells were used to evaluate calpain activity (A), Bid cleavage (B), and cytochrome C release in the cytosol (C). Chromatin fragmentation (D), cell viability (MTT test), and LDH leakage (index of necrosis; F) were evaluated at 24 hours. In A, D, E, and F mean±SEM was of 3 to 4 experiments (*P<0.05). In B and D, Western blots are representative of 4 experiments.
As Bid can theoretically be cleaved by both calpains and caspases,18 we used the multicaspase inhibitor zVAD-fmk and the caspase-2 inhibitor zVDVAD-fmk to investigate whether caspases were involved in oxLDL-induced Bid cleavage. As shown in Figure 2B, when used at effective concentrations, zVAD-fmk and zVDVAD-fmk did not prevent Bid cleavage, thus excluding any major role for caspases-8, -3, and -2.
These data strongly support the hypothesis that calcium and calcium-dependent calpains play a pivotal role in Bid cleavage and subsequent release of cytochrome C, which are caspase-independent.
A Cyclosporin A-Sensitive but FK-506–Insensitive Mechanism Mediates oxLDL-Induced Cytochrome C Release and Subsequent Activation of the Apoptotic Pathway
Activated Bid (truncated Bid, tBid) can elicit cytochrome C release by 2 mechanisms. tBid, through its BH3 domain, triggers the translocation of Bax/Bak that oligomerizes and renders the outer mitochondrial membrane permeable (cyclosporin-A-insensitive mechanism); alternatively, tBid, through its non-BH3 domain, triggers inner mitochondrial membrane remodeling and opening of the mitochondrial permeability transition pore (mPTP; cyclosporin-A-sensitive step).27–30
In HMEC-1, cyclosporin-A inhibited the release of cytochrome C (Figure 3A) and subsequent activation of caspase-3 (Figure 3B), thereby suggesting that cytochrome C release did not result from a direct permeabilization of the mitochondrial membrane by Bax/Bak but rather involved mPTP opening which is cyclosporin-A sensitive.
Figure 3. Cyclosporin-A, but not FK-506, inhibits oxLDL-induced apoptosis. Cells were treated with or without 10 μmol/L cyclosporin-A or 100 nmol/L FK-506 with or without oxLDLs added at time 0. After 20 hours (or the indicated time) of incubation, cells were harvested and used for Western blots of cytochrome C (A), for fluorometric determination of DEVDase and Western blot of caspase-3 (B), for calcineurin determination (C), and chromatin fragmentation (D). At 24 hours, apoptosis and necrosis was evaluated after staining by SYTO-13/PI (E). In A and B (bottom), data are representative of 4 experiments. In C through E, mean±SEM is representative of 3 to 4 separate experiments (*P<0.05).
However, cyclosporin-A may also inhibit calcineurin, a calcium/calmodulin-dependent proapoptotic protein phosphatase that activates the Bad-Bax/Bak pathway.18 To evaluate the possible role of calcineurin, we used FK-506, a specific inhibitor of calcineurin. FK-506, used at effective concentration to inhibit calcineurin (Figure 3C), did not prevent the oxLDL-induced release of cytochrome C (Figure 3A) nor the subsequent activation of caspase-3 (Figure 3B). Moreover, oxLDLs induced marked inhibition of calcineurin (Figure 3C), indicating that the calcineurin-mediated apoptotic pathway plays no major proapoptotic role under our experimental conditions. Finally, this is consistent with the protective effect of cyclosporin-A, whereas FK-506 did not prevent oxLDL-induced apoptosis (Figure 3D and 3E).
These data demonstrate that in our model system oxLDL-induced apoptosis (i) requires the activation of the calcium/calpain/Bid pathway and the release of cytochrome C, which is mediated by a cyclosporin-sensitive mechanism, possibly mPTP; (ii) does not require the calcium/calmodulin/calcineurin pathway; and (iii) is not mediated by a Bax/Bak oligomerization independent of mPTP.
Caspase Inhibitors Partly Prevent oxLDL-Induced Apoptosis
We next investigated whether the caspase-dependent pathway is the only apoptotic pathway activated by oxLDLs, using the caspase-3 inhibitor DEVD-CHO and the multicaspase inhibitor zVAD-fmk. These caspase inhibitors did not prevent the release of cytochrome C (Figure 4A), a finding consistent with the lack of effect of zVAD-fmk on the oxLDL-induced Bid cleavage (Figure 2B). As expected, DEVD-CHO and zVAD-fmk completely inhibited caspase-3 cleavage and activation (Figure 4B and 4C), but these inhibitors only partially blocked oxLDL-induced chromatin fragmentation and apoptosis (Figure 4D through 4F). It should be noted that cotreatment with calpeptin and zVAD-fmk exhibited the same protective effect as each inhibitor used alone (Figure 4F), thus suggesting that these inhibitors act on the same apoptotic pathway.
Figure 4. Caspase inhibitors block caspase-3 activation but only partly inhibit the oxLDL-induced apoptosis. Cells were treated with or without caspase inhibitors, 25 μmol/L of zVAD-fmk (ZV), or DEVD-CHO (DV) for 24 hours. At 20 hours, cells were harvested and used for Western blots of cytochrome C (cytosolic fraction; A) and of caspase-3 (lysate; B), fluorometric determination of DEVDase (C), and chromatin fragmentation (D). At 24 hours, cell death was evaluated by microscopy counting after SYTO-13/IP staining (E) and by the MTT test and LDH released in the culture medium (F). In A and B, data are representative of 4 experiments. In C through F, mean±SEM is representative of 4 experiments (*P<0.05).
Because calpeptin and cyclosporin-A, like caspase inhibitors, completely blocked caspase-3 activation but only partially inhibited chromatin fragmentation and cell death (Figures 2 and 3), this led us to examine the possibility that a calcium-dependent caspase-independent apoptotic mechanism might be implicated in oxLDL-induced apoptosis.
OxLDLs Trigger the Activation of AIF Through a Calcium-Dependent Pathway
Zhang et al17 recently reported that oxLDLs enhance the expression of AIF, and thereby promote caspase-independent apoptosis. Therefore, we investigated whether oxLDL-induced AIF release is mediated by the same mechanism as cytochrome C release.
Western blot analysis of the cytosolic fraction of AIF showed that oxLDLs induced an increase in AIF in the cytosolic fraction in a time- and dose-dependent manner (Figure 5A and 5B). Interestingly, oxLDL-induced release of AIF from mitochondria to cytosol was inhibited by EGTA but not by calpeptin, cyclosporin-A, or zVAD-fmk (Figure 5C). These data suggest that oxLDL-induced AIF release is calcium-dependent but independent of calpain activation, mPTP opening, and caspase activation.
Figure 5. OxLDLs induce AIF release through a calcium-dependent mechanism. Effect of signaling inhibitors is shown. Cells were incubated with oxLDLs (200 μg apoB per mL, under standard conditions) and with or without inhibitors added at time 0. After incubation, the cytosolic fraction was used for Western blots of AIF (A through C). In A, time course of AIF release induced by 200 μg apoB per mL. In B, dose-response of AIF release after incubation of cells for 20 hours with increasing concentration of oxLDLs. In C, cells were incubated for 20 hours with oxLDLs (200 μg apoB per mL) and inhibitors, 0.4 mmol/L EGTA (EG), 2.5 μmol/L calpeptin (Calp), 25 μmol/L of zVAD-fmk (ZV), and 10 μmol/L cyclosporin-A (CSA). In D, immunofluorescence of AIF (confocal fluorescence microscopy) of cells treated without or with oxLDLs and inhibitors used at the same concentrations as in Figure 5C. Representative of 4 experiments is shown.
In control cells, immunofluorescence staining revealed the localization of AIF in mitochondria and the absence of nuclear labeling (Figure 5D). In contrast, oxLDL-treated cells exhibited a partial relocation of AIF in the cytoplasm and the nucleus. Among the inhibitors that we used, only EGTA was able to block AIF release, whereas calpeptin, cyclosporin-A, and zVAD-fmk were ineffective.
We conclude that oxLDL-induced AIF release and nuclear translocation require calcium signaling but not calpain and mPTP activation.
Discussion
OxLDLs can trigger apoptosis or necrosis of cultured vascular cells8,9 and may therefore participate in vascular wall injury, necrotic core formation, plaque erosion/rupture, and in subsequent athero-thrombotic events.4,5 Several signaling pathways activated by oxLDLs could potentially trigger apoptosis or necrosis.7–9 However, the data on signaling and apoptotic pathways triggered by oxLDLs are rather scattered. In this study, we attempted to clarify the relationship between the calcium rise and the mitochondrial apoptotic pathways triggered by oxLDLs.
The data reported here provide new evidence indicating that oxLDLs induced 2 calcium-dependent mitochondrial apoptotic pathways; one mediated by cytochrome C, the other by AIF (Figure II, available online at http://atvb.ahajournals.org).
Calcium
The crucial role of calcium in oxLDL-induced apoptotic signaling was demonstrated by the use of EGTA (at a concentration that buffered extracellular calcium but that avoided the toxic effect of higher concentrations of the chelator). In our model, the oxLDL-induced rise in cytosolic calcium level was required to trigger the activation of the mitochondrial apoptotic pathway, because the calcium chelator EGTA blocked Bid cleavage, cytochrome C release, and subsequent apoptotic events. This finding is consistent with reports pointing to the role of calcium on the toxicity of oxLDLs14–16 or oxysterols31 and to the protective effect of calcium channel blockers31,32
Because high cytosolic calcium levels can theoretically activate several proapoptotic mechanisms mediated by calcium-dependent enzymes, mPTP opening, or calcium-induced osmotic swelling/rupture of mitochondrial membranes,18,27–29 it was of interest to identify the calcium-dependent apoptotic pathways leading to release cytochrome C in our cell system.
Calpains
The role of the calpains in apoptosis is somewhat controversial, because they have reported to be either proapoptotic or antiapoptotic.18 Our data suggest that calpains play a crucial role in the oxLDL-induced activation of the mitochondrial apoptotic pathway (Figure II). The data reported here reveal that calpains induce Bid cleavage and cytochrome C release and subsequent caspase-3 activation as assessed by the inhibitory effect of calpeptin. Our data also permitted us to exclude the hypothesis that calpains directly activate caspases,18 because the activation of executioner caspase-3 was mediated through the mitochondrial apoptotic pathway and was blocked by cyclosporin-A. Other potential candidate proteases, such as cathepsin B and proteasome, can also be excluded, because they are not calcium-dependent and because inhibitors of cathepsin B16 or proteasome20 are not effective in inhibiting the oxLDL-induced apoptosis.
Bid
Bid, a proapoptotic BH3-only member of the Bcl-2 family, can theoretically be activated by various proapoptotic pathways, including death receptors/caspase-8,18 caspase-2,18,33 and calcium/calpains.15 In our model system, oxLDL-induced Bid cleavage is mediated by calpains in agreement with P?rn-Ares et al15 but independent of caspase-8 or caspase-2, because it was inhibited by calpeptin but not by the multicaspase inhibitor zVAD-fmk and the caspase-2 inhibitor zVDVAD-fmk.
Bid cleavage results in formation of the active truncated form tBid that can promote release of cytochrome C. Several mechanisms have been proposed for the action of tBid, including the possibility (i) that tBid promotes the oligomerization of Bak and Bax, (ii) that tBid itself homo-oligomerizes, and (iii) that tBid induces inner mitochondrial membrane remodeling and mPTP opening.18,27–29,34,35 In our model system, tBid mediates the release of cytochrome C through cyclosporin-sensitive and calcium-dependent mechanisms in agreement with the data of Walter et al13 and P?rn-Ares et al,15 thereby suggesting a role for mPTP which is calcium-dependent and cyclosporin-sensitive.18,28
Cyclosporin Targets
Cyclosporin-A inhibits several proapoptotic signaling pathways that lead to mitochondrial cytochrome C release. The initial specific target of cyclosporin-A is the mPTP, which is formed by adenine nucleotide translocator, by cyclophylin D (the target of cyclosporin-A), by voltage dependent anion channel, and by the peripheral benzodiazepine receptor.27–29 Moreover, recent data suggest that cyclosporin-A can prevent the tBid-induced destabilization of the mitochondrial membrane.34 Another target of cyclosporin-A is the calcium/calmodulin-dependent serine/threonine protein phosphatase calcineurin that may activate the Bad/Bax/Bak pathway.18 Calcineurin plays no prominent role in oxLDL-induced apoptosis because the calcineurin inhibitor FK-506 did not prevent apoptosis, and because calcineurin was inhibited by oxLDL treatment. Our data are in contrast with the activation of calcineurin reported in oxysterol-induced apoptosis.31 This discrepancy could result from oxidized lipids contained in oxLDLs which can inhibit calcineurin and other protein phosphatases.36,37
Caspases
Under our experimental conditions, we observed that oxLDLs induced caspase-3 activation. Although caspases have been reported to be substrates for calpains,18 caspase-3 was cleaved and activated through the conventional cytochrome C pathway in our system, because it was inhibited by cyclosporin-A which acts downstream of calpains and blocks cytochrome C release. These data are in agreement with those of Dimmeler et al16 and Chen et al,12 but are in contrast to those of P?rn-Ares et al,15 who reported that oxLDLs cause ubiquitination and inactivation of caspase-3. This discrepancy may result from differences in the level of LDL oxidation because our oxLDL preparations are mildly oxidized (6 to 9 nmol TBARS per mg apoB), whereas those used by P?rn-Ares et al15 are extensively oxidized (25 to 45 nmol TBARS per mg apoB). The higher level of oxidized lipids in oxLDLs may explain both ubiquitination and enzyme inactivation, because highly oxidized LDL may induce cell protein modification and enzymes inactivation.20
Cell Death Receptor-Mediated Extrinsic Apoptotic Pathway
OxLDL-induced apoptosis has also been shown to be mediated by cell death receptors, including Fas and TNFR, and subsequently by caspase-8.9,11 Interestingly, in our model system, caspase-8 played no major role because (i) apoptosis was calcium dependent, whereas Fas and TNFR signaling pathways are calcium-independent; (ii) cleavage of Bid was only poorly sensitive to zVAD-fmk, a multicaspase inhibitor able to inhibit caspase-8; and (iii) a direct activation of caspase-3 by caspase-8 was also excluded, because caspase-3 activation was mediated through a mitochondrial cyclosporin-sensitive mechanism. This minor role (if any) of caspase-8 in HMEC-1 is in agreement with the data of Chen et al12 but in contrast with those of Sata and Walsh11 and Napoli et al,9 who reported a role for death receptors and caspase-8 in oxLDL-induced apoptosis. This discrepancy may result from differences in the expression of TNF or FasL in endothelial cell subtypes used by the authors.
To summarize, in our model system, the signaling apoptotic cascade triggered by oxLDLs involved a sustained calcium rise, calpain activation, Bid cleavage, and cyclosporin-sensitive release of mitochondrial cytochrome C, leading finally to the activation of apoptosome and executioner caspase-3 (Figure II).
AIF
Calpeptin and cyclosporin-A, 2 inhibitors of the oxLDL-induced cytochrome C-mediated apoptotic pathway, and the multicaspase inhibitor zVAD-fmk only partially blocked the oxLDL-induced apoptosis. It was therefore suggested that another caspase-independent apoptotic pathway was activated by oxLDLs. Zhang et al17 reported recently that oxLDLs enhance the expression of the 57-kDa flavoprotein AIF, but the mechanism of release of AIF from mitochondria is indeterminate. Our data show that AIF release is calcium-dependent, but that downstream from calcium, the pathways leading to release of AIF and cytochrome C are distinct because AIF release was not inhibited by calpeptin and cyclosporin-A in contrast to cytochrome C release and caspase-3 activation. Moreover, in our system, oxLDL-induced AIF release was not inhibited by zVAD-fmk and was independent of caspase activation, in agreement with the findings of Cande et al38 but in contrast with those of Arnoult et al.39 The mechanism by which calcium triggers AIF release is not firmly established, but poly(ADP-ribose) polymerase (PARP-1) may constitute such a link because calcium activates PARP-1 which in turn signals AIF release.40
A particularly intriguing question is why the cell has evolved to use 2 different apoptotic pathways induced by toxic agents, like oxLDLs, that trigger unregulated calcium signaling.18 The sustained rise of cytosolic calcium constitutes a lethal hit that activates both necrosis and apoptosis pathways.19 Necrotic cells release proinflammatory molecules potentially dangerous to surrounding cells, whereas apoptosis may prevent or reduce the local inflammation because, in vivo, apoptotic cells are rapidly cleared by phagocytes. It may therefore be speculated that redundant apoptotic pathways selected during the evolution of multicellular organisms to effectively eliminate abnormal, infected, or damaged cells, are also activated in pathophysiological conditions to prevent the necrotic cell destruction and the subsequent local inflammatory processes.
From a pathophysiological point of view, increased apoptosis has been documented in atherosclerotic areas, particularly in lesions associated with unstable angina and ruptured plaques.4 This suggests that apoptosis could play a critical role in plaque erosion and rupture and, subsequently, in athero-thrombotic events.3–5 Because oxLDLs are present in atherosclerotic areas and can induce apoptosis of cultured vascular cells,8,9 it has been hypothesized that oxLDLs and other toxic compounds, including Fas/FasL and inflammatory cytokines, could converge to trigger apoptosis during atherogenesis and plaque disruption. Our data demonstrate that inhibitors of the classical mitochondrial apoptotic pathway would confer only partial protection against oxLDL-induced apoptosis, because these inhibitors are not active on the AIF apoptotic pathway. Moreover, as the oxLDL-induced sustained calcium rise is a common trigger to necrosis and apoptosis,19 protective drugs should act upstream from the peak of cytosolic calcium. A better understanding of upstream proapoptotic signaling activated by oxLDLs should permit the modulation of apoptosis and the prevention of plaque instability.
Acknowledgments
This work was supported by INSERM, University Paul Sabatier. I.E.-B. was supported by a postdoctoral fellowship from the French Atherosclerosis Society and C.V. from INSERM. We thank Dr M.J. Chapman for careful reading of the manuscript.
References
Lusis AJ. Atherosclerosis. Nature. 2000; 407: 233–241.
Ross R. Atherosclerosis: an inflammatory disease. N Engl J Med. 1999; 340: 115–126.
Dimmeler S, Hermann C, Zeiher AM. Apoptosis of endothelial cells. Contribution to the pathophysiology of atherosclerosis? Eur Cytokine Netw. 1998; 9: 697–698.
Mallat Z, Tedgui A. Apoptosis in the vasculature: mechanisms and functional importance. Br J Pharmacol. 2000; 130: 947–962.
Geng YJ, Libby P. Progression of atheroma: a struggle between death and procreation. Arterioscler Thromb Vasc Biol. 2002; 22: 1370–1380.
Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witztum JL. Beyond cholesterol. Modification of LDL that increase its atherogenicity. N Engl J Med. 1989; 320: 915–924.
Colles SM, Maxson JM, Carlson SG, Chisolm GM. Oxidized LDL-induced injury and apoptosis in atherosclerosis. Potential roles for oxysterols. Trends Cardiovasc Med. 2001; 11: 131–138.
Salvayre R, Auge N, Benoist H, Negre-Salvayre A. Oxidized LDL-induced apoptosis. Biochim Biophys Acta. 2002; 1585: 213–221.
Napoli C. Oxidation of LDL, atherogenesis, and apoptosis. Ann N Y Acad Sci. 2003; 1010: 698–709.
Li D, Yang B, Mehta JL. Ox-LDL induces apoptosis in human coronary artery endothelial cells: role of PKC, PTK, bcl-2, and Fas. Am J Physiol. 1998; 275 (pt 2): H568–H576.
Sata M, Walsh K. Oxidized LDL activates fas-mediated endothelial cell apoptosis. J Clin Invest. 1998; 102: 1682–1689.
Chen J, Mehta JL, Haider N, Zhang X, Narula J, Li D. Role of caspases in Ox-LDL-induced apoptotic cascade in human coronary artery endothelial cells. Circ Res. 2004; 94: 370–376.
Walter DH, Haendeler J, Galle J, Zeiher AM, Dimmeler S. Cyclosporine-A inhibits apoptosis of human endothelial cells by preventing release of cytochrome C from mitochondria. Circulation. 1998; 98: 1153–1157.
Escargueil-Blanc I, Salvayre R, Nègre-Salvayre A. Necrosis and apoptosis induced by oxidized LDL occur through two calcium-dependent pathways in lymphoblastoid cells. FASEB J. 1994; 8: 1075–1080.
Porn-Ares MI, Saido TC, Andersson T, Ares MP. Oxidized LDL induces calpain-dependent cell death and ubiquitination of caspase 3 in HMEC-1 endothelial cells. Biochem J. 2003; 374: 403–411.
Dimmeler S, Haendeler J, Galle J, Zeiher AM. Oxidized low-density lipoprotein induces apoptosis of human endothelial cells by activation of CPP32-like proteases. A mechanistic clue to the ‘response to injury’ hypothesis. Circulation. 1997; 95: 1760–1763.
Zhang W, Li D, Mehta JL. Role of AIF in human coronary artery endothelial cell apoptosis. Am J Physiol Heart Circ Physiol. 2004; 286: H354–H358.
Orrenius S, Zhivotovsky B, Nicotera P. Regulation of cell death: the calcium-apoptosis link. Nat Rev Mol Cell Biol. 2003; 4: 552–565.
Meilhac O, Escargueil-Blanc I, Thiers JC, Salvayre R, Negre-Salvayre A. Bcl-2 alters the balance between apoptosis and necrosis, but does not prevent cell death induced by oxidized LDL. FASEB J. 1999; 13: 485–494.
Vieira O, Escargueil-Blanc I, Jurgens G, Borner C, Almeida L, Salvayre R, Negre-Salvayre A. Oxidized LD. Ls alter the activity of the ubiquitin-proteasome pathway: potential role in oxidized LDL-induced apoptosis. FASEB J. 2000; 14: 532–542.
McConkey DJ, Hartzell P, Nicotera P, Orrenius S. Calcium-activated DNA fragmentation kills immature thymocytes. FASEB J. 1989; 3: 1843–1849.
De Duve C, Pressman BC, Gianetto R, Wattiaux R, Appelmans F. Tissue fractionation studies. 6. Intracellular distribution patterns of enzymes in rat-liver tissue. Biochem J. 1955; 60: 604–609.
Chen M, He H, Zhan S, Krajewski S, Reed JC, Gottlieb RA. Bid is cleaved by calpain to an active fragment in vitro and during myocardial ischemia/reperfusion. J Biol Chem. 2001; 276: 30724–30728.
Padma S, Subramanyam C. Extracellular calcineurin: identification and quantitation in serum and amniotic fluid. Clin Biochem. 1999; 32: 491–494.
Garcia-Calvo M, Peterson EP, Rasper DM, Vaillancourt JP, Zamboni R, Nicholson DW, Thornberry NA. Purification and catalytic properties of human caspase family members. Cell Death Differ. 1999; 6: 362–369.
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.
Crompton M. The mitochondrial permeability transition pore and its role in cell death. Biochem J. 1999; 341: 233–249.
Kroemer G, Reed JC. Mitochondrial control of cell death. Nat Med. 2000; 6: 513–519.
Bernardi P, Petronilli V, Di lisa F, Forte M. A mitochondrial perspective on cell death. Trends Biochem Sci. 2001; 26: 112–117.
Newmeyer DD, Ferguson-Miller S. Mitochondria: releasing power for life and unleashing the machineries of death. Cell. 2003; 112: 481–490.
Rusinol AE, Yang L, Thewke D, Panini SR, Kramer MF, Sinensky MS. Isolation of a somatic cell mutant resistant to the induction of apoptosis by oxidized low density lipoprotein. J Biol Chem. 2000; 275: 7296–7303.
Negre-Salvayre A, Salvayre R. Protection by Ca2+ channel blockers (nifedipine, diltiazem and verapamil) against the toxicity of oxidized LDL to cultured lymphoid cells. Br J Pharmacol. 1992; 107: 738–744.
Guo Y, Srinivasula SM, Druilhe A, Fernandes-Alnemri T, Alnemri ES. Caspase-2 induces apoptosis by releasing proapoptotic proteins from mitochondria. J Biol Chem. 2002; 277: 13430–13437.
Henry-Mowatt J, Dive C, Martinou JC, James D. Role of mitochondrial membrane permeabilization in apoptosis and cancer. Oncogene. 2004; 23: 2850–2860.
Zhao Y, Ding WX, Qian T, Watkins S, Lemasters JJ, Yin XM. Bid activates multiple mitochondrial apoptotic mechanisms in primary hepatocytes after death receptor engagement. Gastroenterology. 2003; 125: 854–867.
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.
Rusnak F, Reiter T. Sensing electrons: protein phosphatase redox regulation. Trends Biochem Sci. 2000; 25: 527–529.
Cande C, Cecconi F, Dessen P, Kroemer G. Apoptosis-inducing factor (AIF): key to the conserved caspase-independent pathways of cell death? J Cell Sci. 2002; 115: 4727–4734.
Arnoult D, Parone P, Martinou JC, Antonsson B, Estaquier J, Ameisen JC. Mitochondrial release of apoptosis-inducing factor occurs downstream of cytochrome C release in response to several proapoptotic stimuli. J Cell Biol. 2002; 159: 923–929.
Cregan SP, Dawson VL, Slack RS. Role of AIF in caspase-dependent and caspase-independent cell death. Oncogene. 2004; 23: 2785–2796.(Cécile Vindis; Meyer Elba)
Correspondence to Prof R. Salvayre, Biochimie, Inserm U-466, IFR-31, CHU Rangueil 1, avenue Jean Poulhès, TSA-0032, 31059 Toulouse Cedex 9, France. E-mail salvayre@toulouse.inserm.fr
Abstract
Objective— Oxidized low-density lipoprotein (oxLDL)-induced apoptosis of vascular endothelial cells may contribute to plaque erosion and rupture. We aimed to clarify the relationship between the oxLDL-induced calcium signal and induction of apoptotic pathways.
Methods and Results— Apoptosis was evaluated by biochemical methods, including studies of enzyme activities, protein processing, release of proapoptotic factors, chromatin cleavage, and especially by morphological methods that evaluate apoptosis/necrosis by SYTO-13/propidium iodide fluorescent labeling. The oxLDL-induced sustained calcium rise activated 2 distinct calcium-dependent mitochondrial apoptotic pathways in human microvascular endothelial cells. OxLDLs induced calpain activation and subsequent Bid cleavage and cytochrome C release, which were blocked by calpeptin. Cyclosporin-A inhibited cytochrome C release, possibly by inhibiting the opening of the mitochondrial permeability transition pore (mPTP). Calcineurin, another cyclosporin-sensitive step, was not implicated, because oxLDLs inhibited calcineurin and FK-506 treatment was ineffective. Cytochrome C release in turn induced caspase-3 activation. In addition, oxLDLs triggered release and nuclear translocation of mitochondrial apoptosis-inducing factor through a mechanism dependent on calcium but independent of calpains, mPTP, and caspases.
Conclusions— OxLDL-induced apoptosis involves 2 distinct calcium-dependent pathways, the first mediated by calpain/mPTP/cytochrome C/caspase-3 and the second mediated by apoptosis-inducing factor, which is cyclosporin-insensitive and caspase-independent.
We investigated the potential relationship between calcium signaling, proteolytic cascade, and mitochondrial apoptotic pathways involved in oxidized low-density lipoprotein (oxLDL)-induced apoptosis. OxLDL-induced apoptosis was mediated by 2 distinct mitochondrial pathways, one involving calcium/calpain/mitochondrial permeability transition pore/cytochrome C/caspase-3 and the other calcium-dependent release of apoptosis inducing factor.
Key Words: calpain ? caspase ? mitochondria ? apoptosis-inducing factor ? oxidized low-density lipoprotein ? atherosclerosis
Introduction
Atherogenesis is characterized by lipid deposition, a chronic inflammatory response, and chronic wound healing processes.1,2 Apoptosis may play a role in endothelial cell lining defects, necrotic core formation, or plaque erosion and rupture.3–5 Among the variety of proapoptotic factors present in atherosclerotic plaques, oxidized low-density lipoproteins (oxLDLs) are thought to play a crucial role by concomitantly inducing lipid storage, local inflammation, and toxic events.4,5–8 OxLDLs trigger apoptosis or necrosis of cultured vascular cells7–9 and may therefore participate in vascular wall injury, plaque erosion/rupture, and subsequent athero-thrombotic events.4,5
The proapoptotic effects of oxLDLs are mediated through a complex sequence of signaling events that lead to activation of several caspase-dependent or -independent apoptotic pathways.8,9 Two separate caspase-dependent apoptotic pathways have been implicated in oxLDL-induced apoptosis.7–9 The extrinsic apoptotic pathway, mediated by death receptors, Fas, and/or tumor necrosis factor receptor (TNFR) and downstream by caspase-8/caspase-3, is involved in oxLDL-induced apoptosis in endothelial cells.8,10,11 However, a recent report contests this hypothesis.12 The intrinsic mitochondrial apoptotic pathway, involving bcl-2 family members, cytochrome C, and caspase-3, is especially activated by oxLDLs.8,13 Moreover, a sustained rise of cytosolic calcium plays a central role in oxLDL-induced apoptosis/necrosis,14,15 possibly by activating the calcium-dependent calpains and, subsequently, Bid cleavage and cytochrome C release.15 However, whereas Porn-Ares et al15 found that oxLDLs induce Bid truncation without any activation of caspase-3, Chen et al12 observed that oxLDLs did not cause Bid truncation but activated caspase-3 in agreement with Dimmeler et al.16 Furthermore, oxLDLs can activate a caspase-independent apoptotic pathway mediated by the release of apoptosis-inducing factor (AIF) from mitochondria.17
Because data in the literature on mechanisms implicated in oxLDL-induced apoptosis are controversial and because calcium can trigger multiple apoptotic pathways,18 we presently investigated the potential link between oxLDL-induced calcium signaling and the activation of mitochondrial apoptotic pathways.
Methods
Extensive Materials and Methods are available online at http://atvb.ahajournals.org.
Cell Culture
Human microvascular endothelial cells (HMEC-1) were starved in serum-free medium for 24 hours before LDL treatment.
LDL Isolation and Oxidation
LDLs were prepared as previously indicated.14 Under standard conditions, oxLDLs contained 71 to 104 nmol lipid hydroperoxide per mg apoB and 6.4 to 9.7 nmol of thiobarbituric acid-reactive substances (TBARS) per mg apoB.
Cytotoxicity, Necrosis, and Apoptosis
Cytotoxicity was evaluated by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide test14 and cell lysis (necrosis) by lactate dehydrogenase (LDH) release. Apoptotic and necrotic cells were counted by fluorescence microscopy after staining by SYTO-13 and propidium iodide.19,20
Chromatin Fragments and DNA Ladder
Chromatin fragments were determined by the procedure of McConkey et al.21 DNA laddering was visualized on agarose electrophoresis.
Cell Fractionation
Cytosol was separated from mitochondria by a method adapted from De Duve et al.22
Enzyme Activities
Enzyme activities were determined using fluorogenic substrates.23–25
Western Blot Analysis
Cell extracts were subjected to SDS-PAGE, as previously reported.26
Immunocytochemistry
After the indicated treatment, cells were fixed, incubated with anti-AIF antibody, and revealed with fluorescein isothiocyanate-conjugated secondary antibody.
Statistical Analysis
Data are given as mean±SEM, and statistical significance was evaluated by 1-way ANOVA (Tukey test, SigmaStat software).
Results
The toxicity of oxLDLs was time- and dose-dependent and correlated with the extent of LDL oxidation.14 OxLDLs obtained by UV+copper/EDTA oxidation or by cell-mediated oxidation exhibited similar toxicity to cultured cells.14,26 We used preferably (UV+copper/EDTA)-oxLDLs, because cell-mediated oxLDL preparations may contain mediators secreted by cells that may interfere with oxLDL toxicity.
Time Course of the Effects of oxLDLs on the Calcium-Dependent Mitochondrial Apoptotic Pathway and Apoptosis
Figure I (available online at http://atvb.ahajournals.org) shows the time course of calpain activation and -spectrin breakdown, Bid degradation, release of cytochrome C, and caspase-3 activation (Figure IA through ID). Cell death (Figure IE) occurred mainly by apoptosis, as shown by chromatin fragmentation, DNA laddering, and morphological changes visualized by SYTO-13/PI fluorescent staining (Figure IF through II). The level of primary necrosis was very low, as shown by low levels of LDH release (Figure IE) and by small numbers of PI-stained cells with the morphological features of primary necrosis (Figure IH and II).
Calcium/Calpain Mediates oxLDL-Induced Bid Cleavage, Cytochrome C Release, and Subsequent Caspase-3 Activation
When calcium in the culture medium was chelated by EGTA, under conditions blocking the rise in cytosolic calcium,14 oxLDL-induced calpain activation, cytochrome C release, caspase-3 activation, chromatin fragmentation, and toxicity were inhibited (Figure 1A through 1E). It may be noted that EGTA did not inhibit apoptosis induced by taxol, a potent microtubule-damaging agent, when used as positive control.
Figure 1. EGTA blocks oxLDL-induced activation of calpains, mitochondrial apoptotic pathway, caspase-3, and subsequent toxicity. Cells were incubated in the presence or absence of 0.4 mmol/L EGTA and with or without oxLDLs (200 μg apoB per mL) for 16 hours (A, C, and D), or at the indicated time (B), or for 24 hours (D and E). Cells lysates were used for fluorometric determination of calpain activity (A). Western blots of cytochrome C were performed on cytosolic fractions (B). Cell lysates were used for the fluorometric measurement of DEVDase and Western blots of caspase-3 (C) and chromatin fragmentation (D). Cell viability and necrosis (E) were evaluated at 24 hours using the McConkey procedure and 4,'6-diamidino-2-phenylindole fluorescent staining (D), by the MTT test, and by leakage of LDH in the culture medium (the basal LDH level was very low, 5% of the total cellular LDH; E). In A, C, D, and E, mean±SEM was of at least 4 experiments (*P<0.05). In B and C, Western blots are representative of 4 experiments.
To investigate whether calpain activation is involved in oxLDL-induced apoptosis, we used calpeptin, a selective inhibitor of calpains. Calpeptin inhibited oxLDL-induced calpain activation (Figure 2A) and prevented Bid cleavage, release of cytochrome C, caspase-3 activation (Figure 2B through 2D), and both chromatin fragmentation and cytotoxicity by oxLDLs (Figure 2E through 2F).
Figure 2. The calpain inhibitor calpeptin inhibits oxLDL-induced activation of calpains, apoptotic mitochondrial pathways, caspase-3 activation, and subsequent apoptotic events. Cells were treated with or without 2.5 μmol/L calpeptin, or 25 μmol/L zVAD-fmk, or 25 μmol/L zVDVAD-fmk, and added at time 0 with 200 μg apoB per mL oxLDLs. After 20 hours or after the indicated time of incubation, cells were used to evaluate calpain activity (A), Bid cleavage (B), and cytochrome C release in the cytosol (C). Chromatin fragmentation (D), cell viability (MTT test), and LDH leakage (index of necrosis; F) were evaluated at 24 hours. In A, D, E, and F mean±SEM was of 3 to 4 experiments (*P<0.05). In B and D, Western blots are representative of 4 experiments.
As Bid can theoretically be cleaved by both calpains and caspases,18 we used the multicaspase inhibitor zVAD-fmk and the caspase-2 inhibitor zVDVAD-fmk to investigate whether caspases were involved in oxLDL-induced Bid cleavage. As shown in Figure 2B, when used at effective concentrations, zVAD-fmk and zVDVAD-fmk did not prevent Bid cleavage, thus excluding any major role for caspases-8, -3, and -2.
These data strongly support the hypothesis that calcium and calcium-dependent calpains play a pivotal role in Bid cleavage and subsequent release of cytochrome C, which are caspase-independent.
A Cyclosporin A-Sensitive but FK-506–Insensitive Mechanism Mediates oxLDL-Induced Cytochrome C Release and Subsequent Activation of the Apoptotic Pathway
Activated Bid (truncated Bid, tBid) can elicit cytochrome C release by 2 mechanisms. tBid, through its BH3 domain, triggers the translocation of Bax/Bak that oligomerizes and renders the outer mitochondrial membrane permeable (cyclosporin-A-insensitive mechanism); alternatively, tBid, through its non-BH3 domain, triggers inner mitochondrial membrane remodeling and opening of the mitochondrial permeability transition pore (mPTP; cyclosporin-A-sensitive step).27–30
In HMEC-1, cyclosporin-A inhibited the release of cytochrome C (Figure 3A) and subsequent activation of caspase-3 (Figure 3B), thereby suggesting that cytochrome C release did not result from a direct permeabilization of the mitochondrial membrane by Bax/Bak but rather involved mPTP opening which is cyclosporin-A sensitive.
Figure 3. Cyclosporin-A, but not FK-506, inhibits oxLDL-induced apoptosis. Cells were treated with or without 10 μmol/L cyclosporin-A or 100 nmol/L FK-506 with or without oxLDLs added at time 0. After 20 hours (or the indicated time) of incubation, cells were harvested and used for Western blots of cytochrome C (A), for fluorometric determination of DEVDase and Western blot of caspase-3 (B), for calcineurin determination (C), and chromatin fragmentation (D). At 24 hours, apoptosis and necrosis was evaluated after staining by SYTO-13/PI (E). In A and B (bottom), data are representative of 4 experiments. In C through E, mean±SEM is representative of 3 to 4 separate experiments (*P<0.05).
However, cyclosporin-A may also inhibit calcineurin, a calcium/calmodulin-dependent proapoptotic protein phosphatase that activates the Bad-Bax/Bak pathway.18 To evaluate the possible role of calcineurin, we used FK-506, a specific inhibitor of calcineurin. FK-506, used at effective concentration to inhibit calcineurin (Figure 3C), did not prevent the oxLDL-induced release of cytochrome C (Figure 3A) nor the subsequent activation of caspase-3 (Figure 3B). Moreover, oxLDLs induced marked inhibition of calcineurin (Figure 3C), indicating that the calcineurin-mediated apoptotic pathway plays no major proapoptotic role under our experimental conditions. Finally, this is consistent with the protective effect of cyclosporin-A, whereas FK-506 did not prevent oxLDL-induced apoptosis (Figure 3D and 3E).
These data demonstrate that in our model system oxLDL-induced apoptosis (i) requires the activation of the calcium/calpain/Bid pathway and the release of cytochrome C, which is mediated by a cyclosporin-sensitive mechanism, possibly mPTP; (ii) does not require the calcium/calmodulin/calcineurin pathway; and (iii) is not mediated by a Bax/Bak oligomerization independent of mPTP.
Caspase Inhibitors Partly Prevent oxLDL-Induced Apoptosis
We next investigated whether the caspase-dependent pathway is the only apoptotic pathway activated by oxLDLs, using the caspase-3 inhibitor DEVD-CHO and the multicaspase inhibitor zVAD-fmk. These caspase inhibitors did not prevent the release of cytochrome C (Figure 4A), a finding consistent with the lack of effect of zVAD-fmk on the oxLDL-induced Bid cleavage (Figure 2B). As expected, DEVD-CHO and zVAD-fmk completely inhibited caspase-3 cleavage and activation (Figure 4B and 4C), but these inhibitors only partially blocked oxLDL-induced chromatin fragmentation and apoptosis (Figure 4D through 4F). It should be noted that cotreatment with calpeptin and zVAD-fmk exhibited the same protective effect as each inhibitor used alone (Figure 4F), thus suggesting that these inhibitors act on the same apoptotic pathway.
Figure 4. Caspase inhibitors block caspase-3 activation but only partly inhibit the oxLDL-induced apoptosis. Cells were treated with or without caspase inhibitors, 25 μmol/L of zVAD-fmk (ZV), or DEVD-CHO (DV) for 24 hours. At 20 hours, cells were harvested and used for Western blots of cytochrome C (cytosolic fraction; A) and of caspase-3 (lysate; B), fluorometric determination of DEVDase (C), and chromatin fragmentation (D). At 24 hours, cell death was evaluated by microscopy counting after SYTO-13/IP staining (E) and by the MTT test and LDH released in the culture medium (F). In A and B, data are representative of 4 experiments. In C through F, mean±SEM is representative of 4 experiments (*P<0.05).
Because calpeptin and cyclosporin-A, like caspase inhibitors, completely blocked caspase-3 activation but only partially inhibited chromatin fragmentation and cell death (Figures 2 and 3), this led us to examine the possibility that a calcium-dependent caspase-independent apoptotic mechanism might be implicated in oxLDL-induced apoptosis.
OxLDLs Trigger the Activation of AIF Through a Calcium-Dependent Pathway
Zhang et al17 recently reported that oxLDLs enhance the expression of AIF, and thereby promote caspase-independent apoptosis. Therefore, we investigated whether oxLDL-induced AIF release is mediated by the same mechanism as cytochrome C release.
Western blot analysis of the cytosolic fraction of AIF showed that oxLDLs induced an increase in AIF in the cytosolic fraction in a time- and dose-dependent manner (Figure 5A and 5B). Interestingly, oxLDL-induced release of AIF from mitochondria to cytosol was inhibited by EGTA but not by calpeptin, cyclosporin-A, or zVAD-fmk (Figure 5C). These data suggest that oxLDL-induced AIF release is calcium-dependent but independent of calpain activation, mPTP opening, and caspase activation.
Figure 5. OxLDLs induce AIF release through a calcium-dependent mechanism. Effect of signaling inhibitors is shown. Cells were incubated with oxLDLs (200 μg apoB per mL, under standard conditions) and with or without inhibitors added at time 0. After incubation, the cytosolic fraction was used for Western blots of AIF (A through C). In A, time course of AIF release induced by 200 μg apoB per mL. In B, dose-response of AIF release after incubation of cells for 20 hours with increasing concentration of oxLDLs. In C, cells were incubated for 20 hours with oxLDLs (200 μg apoB per mL) and inhibitors, 0.4 mmol/L EGTA (EG), 2.5 μmol/L calpeptin (Calp), 25 μmol/L of zVAD-fmk (ZV), and 10 μmol/L cyclosporin-A (CSA). In D, immunofluorescence of AIF (confocal fluorescence microscopy) of cells treated without or with oxLDLs and inhibitors used at the same concentrations as in Figure 5C. Representative of 4 experiments is shown.
In control cells, immunofluorescence staining revealed the localization of AIF in mitochondria and the absence of nuclear labeling (Figure 5D). In contrast, oxLDL-treated cells exhibited a partial relocation of AIF in the cytoplasm and the nucleus. Among the inhibitors that we used, only EGTA was able to block AIF release, whereas calpeptin, cyclosporin-A, and zVAD-fmk were ineffective.
We conclude that oxLDL-induced AIF release and nuclear translocation require calcium signaling but not calpain and mPTP activation.
Discussion
OxLDLs can trigger apoptosis or necrosis of cultured vascular cells8,9 and may therefore participate in vascular wall injury, necrotic core formation, plaque erosion/rupture, and in subsequent athero-thrombotic events.4,5 Several signaling pathways activated by oxLDLs could potentially trigger apoptosis or necrosis.7–9 However, the data on signaling and apoptotic pathways triggered by oxLDLs are rather scattered. In this study, we attempted to clarify the relationship between the calcium rise and the mitochondrial apoptotic pathways triggered by oxLDLs.
The data reported here provide new evidence indicating that oxLDLs induced 2 calcium-dependent mitochondrial apoptotic pathways; one mediated by cytochrome C, the other by AIF (Figure II, available online at http://atvb.ahajournals.org).
Calcium
The crucial role of calcium in oxLDL-induced apoptotic signaling was demonstrated by the use of EGTA (at a concentration that buffered extracellular calcium but that avoided the toxic effect of higher concentrations of the chelator). In our model, the oxLDL-induced rise in cytosolic calcium level was required to trigger the activation of the mitochondrial apoptotic pathway, because the calcium chelator EGTA blocked Bid cleavage, cytochrome C release, and subsequent apoptotic events. This finding is consistent with reports pointing to the role of calcium on the toxicity of oxLDLs14–16 or oxysterols31 and to the protective effect of calcium channel blockers31,32
Because high cytosolic calcium levels can theoretically activate several proapoptotic mechanisms mediated by calcium-dependent enzymes, mPTP opening, or calcium-induced osmotic swelling/rupture of mitochondrial membranes,18,27–29 it was of interest to identify the calcium-dependent apoptotic pathways leading to release cytochrome C in our cell system.
Calpains
The role of the calpains in apoptosis is somewhat controversial, because they have reported to be either proapoptotic or antiapoptotic.18 Our data suggest that calpains play a crucial role in the oxLDL-induced activation of the mitochondrial apoptotic pathway (Figure II). The data reported here reveal that calpains induce Bid cleavage and cytochrome C release and subsequent caspase-3 activation as assessed by the inhibitory effect of calpeptin. Our data also permitted us to exclude the hypothesis that calpains directly activate caspases,18 because the activation of executioner caspase-3 was mediated through the mitochondrial apoptotic pathway and was blocked by cyclosporin-A. Other potential candidate proteases, such as cathepsin B and proteasome, can also be excluded, because they are not calcium-dependent and because inhibitors of cathepsin B16 or proteasome20 are not effective in inhibiting the oxLDL-induced apoptosis.
Bid
Bid, a proapoptotic BH3-only member of the Bcl-2 family, can theoretically be activated by various proapoptotic pathways, including death receptors/caspase-8,18 caspase-2,18,33 and calcium/calpains.15 In our model system, oxLDL-induced Bid cleavage is mediated by calpains in agreement with P?rn-Ares et al15 but independent of caspase-8 or caspase-2, because it was inhibited by calpeptin but not by the multicaspase inhibitor zVAD-fmk and the caspase-2 inhibitor zVDVAD-fmk.
Bid cleavage results in formation of the active truncated form tBid that can promote release of cytochrome C. Several mechanisms have been proposed for the action of tBid, including the possibility (i) that tBid promotes the oligomerization of Bak and Bax, (ii) that tBid itself homo-oligomerizes, and (iii) that tBid induces inner mitochondrial membrane remodeling and mPTP opening.18,27–29,34,35 In our model system, tBid mediates the release of cytochrome C through cyclosporin-sensitive and calcium-dependent mechanisms in agreement with the data of Walter et al13 and P?rn-Ares et al,15 thereby suggesting a role for mPTP which is calcium-dependent and cyclosporin-sensitive.18,28
Cyclosporin Targets
Cyclosporin-A inhibits several proapoptotic signaling pathways that lead to mitochondrial cytochrome C release. The initial specific target of cyclosporin-A is the mPTP, which is formed by adenine nucleotide translocator, by cyclophylin D (the target of cyclosporin-A), by voltage dependent anion channel, and by the peripheral benzodiazepine receptor.27–29 Moreover, recent data suggest that cyclosporin-A can prevent the tBid-induced destabilization of the mitochondrial membrane.34 Another target of cyclosporin-A is the calcium/calmodulin-dependent serine/threonine protein phosphatase calcineurin that may activate the Bad/Bax/Bak pathway.18 Calcineurin plays no prominent role in oxLDL-induced apoptosis because the calcineurin inhibitor FK-506 did not prevent apoptosis, and because calcineurin was inhibited by oxLDL treatment. Our data are in contrast with the activation of calcineurin reported in oxysterol-induced apoptosis.31 This discrepancy could result from oxidized lipids contained in oxLDLs which can inhibit calcineurin and other protein phosphatases.36,37
Caspases
Under our experimental conditions, we observed that oxLDLs induced caspase-3 activation. Although caspases have been reported to be substrates for calpains,18 caspase-3 was cleaved and activated through the conventional cytochrome C pathway in our system, because it was inhibited by cyclosporin-A which acts downstream of calpains and blocks cytochrome C release. These data are in agreement with those of Dimmeler et al16 and Chen et al,12 but are in contrast to those of P?rn-Ares et al,15 who reported that oxLDLs cause ubiquitination and inactivation of caspase-3. This discrepancy may result from differences in the level of LDL oxidation because our oxLDL preparations are mildly oxidized (6 to 9 nmol TBARS per mg apoB), whereas those used by P?rn-Ares et al15 are extensively oxidized (25 to 45 nmol TBARS per mg apoB). The higher level of oxidized lipids in oxLDLs may explain both ubiquitination and enzyme inactivation, because highly oxidized LDL may induce cell protein modification and enzymes inactivation.20
Cell Death Receptor-Mediated Extrinsic Apoptotic Pathway
OxLDL-induced apoptosis has also been shown to be mediated by cell death receptors, including Fas and TNFR, and subsequently by caspase-8.9,11 Interestingly, in our model system, caspase-8 played no major role because (i) apoptosis was calcium dependent, whereas Fas and TNFR signaling pathways are calcium-independent; (ii) cleavage of Bid was only poorly sensitive to zVAD-fmk, a multicaspase inhibitor able to inhibit caspase-8; and (iii) a direct activation of caspase-3 by caspase-8 was also excluded, because caspase-3 activation was mediated through a mitochondrial cyclosporin-sensitive mechanism. This minor role (if any) of caspase-8 in HMEC-1 is in agreement with the data of Chen et al12 but in contrast with those of Sata and Walsh11 and Napoli et al,9 who reported a role for death receptors and caspase-8 in oxLDL-induced apoptosis. This discrepancy may result from differences in the expression of TNF or FasL in endothelial cell subtypes used by the authors.
To summarize, in our model system, the signaling apoptotic cascade triggered by oxLDLs involved a sustained calcium rise, calpain activation, Bid cleavage, and cyclosporin-sensitive release of mitochondrial cytochrome C, leading finally to the activation of apoptosome and executioner caspase-3 (Figure II).
AIF
Calpeptin and cyclosporin-A, 2 inhibitors of the oxLDL-induced cytochrome C-mediated apoptotic pathway, and the multicaspase inhibitor zVAD-fmk only partially blocked the oxLDL-induced apoptosis. It was therefore suggested that another caspase-independent apoptotic pathway was activated by oxLDLs. Zhang et al17 reported recently that oxLDLs enhance the expression of the 57-kDa flavoprotein AIF, but the mechanism of release of AIF from mitochondria is indeterminate. Our data show that AIF release is calcium-dependent, but that downstream from calcium, the pathways leading to release of AIF and cytochrome C are distinct because AIF release was not inhibited by calpeptin and cyclosporin-A in contrast to cytochrome C release and caspase-3 activation. Moreover, in our system, oxLDL-induced AIF release was not inhibited by zVAD-fmk and was independent of caspase activation, in agreement with the findings of Cande et al38 but in contrast with those of Arnoult et al.39 The mechanism by which calcium triggers AIF release is not firmly established, but poly(ADP-ribose) polymerase (PARP-1) may constitute such a link because calcium activates PARP-1 which in turn signals AIF release.40
A particularly intriguing question is why the cell has evolved to use 2 different apoptotic pathways induced by toxic agents, like oxLDLs, that trigger unregulated calcium signaling.18 The sustained rise of cytosolic calcium constitutes a lethal hit that activates both necrosis and apoptosis pathways.19 Necrotic cells release proinflammatory molecules potentially dangerous to surrounding cells, whereas apoptosis may prevent or reduce the local inflammation because, in vivo, apoptotic cells are rapidly cleared by phagocytes. It may therefore be speculated that redundant apoptotic pathways selected during the evolution of multicellular organisms to effectively eliminate abnormal, infected, or damaged cells, are also activated in pathophysiological conditions to prevent the necrotic cell destruction and the subsequent local inflammatory processes.
From a pathophysiological point of view, increased apoptosis has been documented in atherosclerotic areas, particularly in lesions associated with unstable angina and ruptured plaques.4 This suggests that apoptosis could play a critical role in plaque erosion and rupture and, subsequently, in athero-thrombotic events.3–5 Because oxLDLs are present in atherosclerotic areas and can induce apoptosis of cultured vascular cells,8,9 it has been hypothesized that oxLDLs and other toxic compounds, including Fas/FasL and inflammatory cytokines, could converge to trigger apoptosis during atherogenesis and plaque disruption. Our data demonstrate that inhibitors of the classical mitochondrial apoptotic pathway would confer only partial protection against oxLDL-induced apoptosis, because these inhibitors are not active on the AIF apoptotic pathway. Moreover, as the oxLDL-induced sustained calcium rise is a common trigger to necrosis and apoptosis,19 protective drugs should act upstream from the peak of cytosolic calcium. A better understanding of upstream proapoptotic signaling activated by oxLDLs should permit the modulation of apoptosis and the prevention of plaque instability.
Acknowledgments
This work was supported by INSERM, University Paul Sabatier. I.E.-B. was supported by a postdoctoral fellowship from the French Atherosclerosis Society and C.V. from INSERM. We thank Dr M.J. Chapman for careful reading of the manuscript.
References
Lusis AJ. Atherosclerosis. Nature. 2000; 407: 233–241.
Ross R. Atherosclerosis: an inflammatory disease. N Engl J Med. 1999; 340: 115–126.
Dimmeler S, Hermann C, Zeiher AM. Apoptosis of endothelial cells. Contribution to the pathophysiology of atherosclerosis? Eur Cytokine Netw. 1998; 9: 697–698.
Mallat Z, Tedgui A. Apoptosis in the vasculature: mechanisms and functional importance. Br J Pharmacol. 2000; 130: 947–962.
Geng YJ, Libby P. Progression of atheroma: a struggle between death and procreation. Arterioscler Thromb Vasc Biol. 2002; 22: 1370–1380.
Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witztum JL. Beyond cholesterol. Modification of LDL that increase its atherogenicity. N Engl J Med. 1989; 320: 915–924.
Colles SM, Maxson JM, Carlson SG, Chisolm GM. Oxidized LDL-induced injury and apoptosis in atherosclerosis. Potential roles for oxysterols. Trends Cardiovasc Med. 2001; 11: 131–138.
Salvayre R, Auge N, Benoist H, Negre-Salvayre A. Oxidized LDL-induced apoptosis. Biochim Biophys Acta. 2002; 1585: 213–221.
Napoli C. Oxidation of LDL, atherogenesis, and apoptosis. Ann N Y Acad Sci. 2003; 1010: 698–709.
Li D, Yang B, Mehta JL. Ox-LDL induces apoptosis in human coronary artery endothelial cells: role of PKC, PTK, bcl-2, and Fas. Am J Physiol. 1998; 275 (pt 2): H568–H576.
Sata M, Walsh K. Oxidized LDL activates fas-mediated endothelial cell apoptosis. J Clin Invest. 1998; 102: 1682–1689.
Chen J, Mehta JL, Haider N, Zhang X, Narula J, Li D. Role of caspases in Ox-LDL-induced apoptotic cascade in human coronary artery endothelial cells. Circ Res. 2004; 94: 370–376.
Walter DH, Haendeler J, Galle J, Zeiher AM, Dimmeler S. Cyclosporine-A inhibits apoptosis of human endothelial cells by preventing release of cytochrome C from mitochondria. Circulation. 1998; 98: 1153–1157.
Escargueil-Blanc I, Salvayre R, Nègre-Salvayre A. Necrosis and apoptosis induced by oxidized LDL occur through two calcium-dependent pathways in lymphoblastoid cells. FASEB J. 1994; 8: 1075–1080.
Porn-Ares MI, Saido TC, Andersson T, Ares MP. Oxidized LDL induces calpain-dependent cell death and ubiquitination of caspase 3 in HMEC-1 endothelial cells. Biochem J. 2003; 374: 403–411.
Dimmeler S, Haendeler J, Galle J, Zeiher AM. Oxidized low-density lipoprotein induces apoptosis of human endothelial cells by activation of CPP32-like proteases. A mechanistic clue to the ‘response to injury’ hypothesis. Circulation. 1997; 95: 1760–1763.
Zhang W, Li D, Mehta JL. Role of AIF in human coronary artery endothelial cell apoptosis. Am J Physiol Heart Circ Physiol. 2004; 286: H354–H358.
Orrenius S, Zhivotovsky B, Nicotera P. Regulation of cell death: the calcium-apoptosis link. Nat Rev Mol Cell Biol. 2003; 4: 552–565.
Meilhac O, Escargueil-Blanc I, Thiers JC, Salvayre R, Negre-Salvayre A. Bcl-2 alters the balance between apoptosis and necrosis, but does not prevent cell death induced by oxidized LDL. FASEB J. 1999; 13: 485–494.
Vieira O, Escargueil-Blanc I, Jurgens G, Borner C, Almeida L, Salvayre R, Negre-Salvayre A. Oxidized LD. Ls alter the activity of the ubiquitin-proteasome pathway: potential role in oxidized LDL-induced apoptosis. FASEB J. 2000; 14: 532–542.
McConkey DJ, Hartzell P, Nicotera P, Orrenius S. Calcium-activated DNA fragmentation kills immature thymocytes. FASEB J. 1989; 3: 1843–1849.
De Duve C, Pressman BC, Gianetto R, Wattiaux R, Appelmans F. Tissue fractionation studies. 6. Intracellular distribution patterns of enzymes in rat-liver tissue. Biochem J. 1955; 60: 604–609.
Chen M, He H, Zhan S, Krajewski S, Reed JC, Gottlieb RA. Bid is cleaved by calpain to an active fragment in vitro and during myocardial ischemia/reperfusion. J Biol Chem. 2001; 276: 30724–30728.
Padma S, Subramanyam C. Extracellular calcineurin: identification and quantitation in serum and amniotic fluid. Clin Biochem. 1999; 32: 491–494.
Garcia-Calvo M, Peterson EP, Rasper DM, Vaillancourt JP, Zamboni R, Nicholson DW, Thornberry NA. Purification and catalytic properties of human caspase family members. Cell Death Differ. 1999; 6: 362–369.
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.
Crompton M. The mitochondrial permeability transition pore and its role in cell death. Biochem J. 1999; 341: 233–249.
Kroemer G, Reed JC. Mitochondrial control of cell death. Nat Med. 2000; 6: 513–519.
Bernardi P, Petronilli V, Di lisa F, Forte M. A mitochondrial perspective on cell death. Trends Biochem Sci. 2001; 26: 112–117.
Newmeyer DD, Ferguson-Miller S. Mitochondria: releasing power for life and unleashing the machineries of death. Cell. 2003; 112: 481–490.
Rusinol AE, Yang L, Thewke D, Panini SR, Kramer MF, Sinensky MS. Isolation of a somatic cell mutant resistant to the induction of apoptosis by oxidized low density lipoprotein. J Biol Chem. 2000; 275: 7296–7303.
Negre-Salvayre A, Salvayre R. Protection by Ca2+ channel blockers (nifedipine, diltiazem and verapamil) against the toxicity of oxidized LDL to cultured lymphoid cells. Br J Pharmacol. 1992; 107: 738–744.
Guo Y, Srinivasula SM, Druilhe A, Fernandes-Alnemri T, Alnemri ES. Caspase-2 induces apoptosis by releasing proapoptotic proteins from mitochondria. J Biol Chem. 2002; 277: 13430–13437.
Henry-Mowatt J, Dive C, Martinou JC, James D. Role of mitochondrial membrane permeabilization in apoptosis and cancer. Oncogene. 2004; 23: 2850–2860.
Zhao Y, Ding WX, Qian T, Watkins S, Lemasters JJ, Yin XM. Bid activates multiple mitochondrial apoptotic mechanisms in primary hepatocytes after death receptor engagement. Gastroenterology. 2003; 125: 854–867.
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.
Rusnak F, Reiter T. Sensing electrons: protein phosphatase redox regulation. Trends Biochem Sci. 2000; 25: 527–529.
Cande C, Cecconi F, Dessen P, Kroemer G. Apoptosis-inducing factor (AIF): key to the conserved caspase-independent pathways of cell death? J Cell Sci. 2002; 115: 4727–4734.
Arnoult D, Parone P, Martinou JC, Antonsson B, Estaquier J, Ameisen JC. Mitochondrial release of apoptosis-inducing factor occurs downstream of cytochrome C release in response to several proapoptotic stimuli. J Cell Biol. 2002; 159: 923–929.
Cregan SP, Dawson VL, Slack RS. Role of AIF in caspase-dependent and caspase-independent cell death. Oncogene. 2004; 23: 2785–2796.(Cécile Vindis; Meyer Elba)