| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Research Communications |
INSERM U-466 and Department of Biochemistry and Molecular Biology, IFR-31, CHU Rangueil, Toulouse, France
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: oxidized LDL • atherosclerosis • DNA ladder • cytosolic calcium concentration • antisense oligonucleotides
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Low density lipoproteins (LDL)2 play a major physiological role in delivering cholesterol to peripheral cells, but are also involved in the genesis of atherosclerosis (3) after undergoing oxidative modifications 4-6) . LDL oxidation is mediated in vitro by cultured vascular cells and is thought to occur in vivo, as suggested by the presence of oxidized LDL in atherosclerotic areas (reviewed in ref 6 ). Among the wide variety of their biological properties, oxidized LDL (oxLDL) exhibit a dramatic cytotoxic effect to cultured cells, which is potentially involved in necrotic core formation, endothelial cell injury, and plaque rupture 7-9) .
Toxic cell injury induces a complex sequence of events leading to cell death (10) . Two types of cell death, necrosis and apoptosis, have been discriminated on the basis of morphological studies (11) . Apoptosis, or programmed cell death, is characterized by DNA fragmentation, alterations of nucleus morphology (chromatin condensation and nucleus fragmentation), organelle relocalization, and cell fragmentation without leakage of cytosolic macromolecules (11) . Primary necrosis (or oncosis) is characterized by cellular swelling, organelles alterations, rupture of plasma membrane, and finally cell lysis and leakage of the cellular components (11) . During toxic cell injury, disruption of Ca2+ homeostasis seems to play a critical role by triggering activation of calcium-dependent degradative enzymes, resulting in irreversible damage of cellular components leading to cell death (12 , 13 ). Calcium-dependent cytoplasmic proteases are involved in cytoskeletal alterations and blebbing of the plasma membrane 12-14) , whereas nuclear calcium-dependent endonucleases are implicated in DNA cleavage 15-17) . The toxic effect of oxidized LDL is mediated by an intense and sustained rise of cytosolic calcium (14 , 18 ), which in turn activates calcium-dependent enzymes involved in cellular events leading to necrosis or/and apoptosis of lymphoid and macrophagic cells (14 , 19 ) and to massive apoptosis of endothelial cells (20) .
Bcl-2, located predominantly in the outer mitochondrial membrane, the endoplasmic reticulum, and the nuclear membrane, is a prototypic antiapoptotic protein, but its efficacy depends on the cause and mechanism of cell death (21) . Several mechanisms have been proposed to explain the antiapoptotic effect of Bcl-2, namely, antioxidant-like effect (22) , inhibition of release of calcium by the endoplasmic reticulum (23) , and, more recently, prevention of caspase activation, possibly through inhibition of mitochondrial cytochrome c release (24 , 25 ). Bcl-2 is able to inhibit in part apoptosis of monocytic cell lines triggered by oxysterols (which mediate, at least in part, the cytotoxicity of oxidized LDL) (26 , 27 ); however, its protective effect against necrosis (and against the overall cell death) has not been evaluated. This led us to investigate carefully this hypothesis (i.e., potential protective role of Bcl-2 against the toxicity of oxidized LDL).
We report here that 1) toxic concentrations of oxLDL induce apoptosis of cells expressing low levels of Bcl-2 protein and induce primary necrosis of cells expressing high Bcl-2 levels; 2) both apoptosis and necrosis induced by oxLDL are calcium dependent; 3) Bcl-2 overexpression does not really prevent cell death triggered by oxLDL, but rather induces a shift from apoptosis toward necrosis; Bcl-2 inhibits the apoptotic pathway by acting downstream from the calcium peak, thus without inhibiting necrosis. Finally, the potential relevance of these data is discussed in relation to atherogenesis and coronary heart disease.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-32P]dCTP (800 Ci/mmol) from Isotopchim
(Ganagobie-Peyruis, France), trypan blue dye, bovine serum albumin,
4',6-diamidino-2-phenylindole (DAPI), agarose, were from Sigma (St.
Louis, Mo.), RPMI 1640 containing Glutamax, fetal calf serum (FCS) heat
inactivated at 56°C for 1 h, penicillin, streptomycin from
Gibco (Cergy-Pontoise, France), Hydragel from Sebia (Issy, France), 100
bp DNA ladder and Klenow polymerase from Promega (Charbonnières,
France), G-Nome kit from Bio-101 (La Jolla, Calif.),
acrylamide-4X/bisacrylamide-2X solution from Bioprobe (Montreuil,
France), SYTO-13 and propidium iodide from Molecular Probes (Eugene,
Oreg.), and other chemicals from Merck (Darmstadt, Germany), Sigma, or
Prolabo (Paris).
Cell culture
HL60/B and HL60/Neo cell lines, derived from the myeloid cell
line HL60 by stable transfection by Bcl-2- or Neo-containing retrovirus
vectors (30)
, were a generous gift from Dr. J. P.
Jaffrezou (CJF INSERM-9503, Toulouse). ECV-304 human umbilical vein
endothelial cells (CRL-1998) were obtained from ATCC (Rockville, Md.).
Cells were grown in RPMI 1640 with Glutamax supplemented with 10% heat-inactivated FCS, 100 U/ml penicillin, and 100 µg/ml streptomycin. This medium was replaced by a serum-free medium 24 h before LDL incorporation.
Peripheral blood lymphocytes were prepared from heparinized venous blood from healthy donors by centrifugation (1500 x g for 20 min) on Histopaque (Sigma) and by removing adherent cells (for 2 h at 37°C) under the conditions used previously (28) . The lymphocyte preparations contained ~75–85% T cells, 5–10% B cells, 10–15% NK cells, and less than 0.5% monocytes. They were maintained up to 72 h for lymphocytes in RPMI 1640 medium. The standard medium was usually replaced by a serum-free medium 24 h before the experiments.
Antisense (AS) oligonucleotide experiments were performed according to Kitada et al. (29) : lymphocytes were incubated for 24 h with 10 µM phosphorothioate oligonucleotides (18-mers), 5'-TCTCCCAGCGTGCCCAT-3' (antisense to the first 6 codons of Bcl-2), or 5'-TGCACTCACGCTCGGCCT-3' (scrambled (SC) oligonucleotide, as negative control) (Eurogentec).
LDL isolation and oxidation
LDL from human pooled sera were isolated by ultracentrifugation
under the conditions used previously (14
,
20
), dialyzed against phosphate-buffered saline (PBS)
containing 100 µmol/l EDTA, sterilized on 0.2 µm Millipore membrane
and stored under nitrogen at 4°C (up to 3 wk). LDL relative
electrophoretic mobility (REM) was evaluated on Hydragel (Sebia).
Apoprotein B was determined by immunonephelometry.
Mildly oxidized LDL were obtained by (UV + copper/EDTA)-mediated
oxidation as previously described (20)
: briefly, LDL
solution [2 mg apoprotein B (apoB)/ml, containing 2 µmol/l
CuSO4] was irradiated for 2 h, as a thin film (5 mm)
in an open beaker placed 10 cm under the UV-C source (HNS 30W OFR Osram
UV-C tube, 
max 254 nm, 0.5 mW/cm2 determined using a
Scientech thermopile Model 360001). At the end of the irradiation,
aliquots were taken up for analyses and oxidized LDL (200 µg apoB/ml
under standard conditions or at the indicated concentration) were
immediately incorporated in the culture medium.
The level of LDL oxidation was evaluated by monitoring the formation of thiobarbituric acid reactive substances (TBARS) according to Yagi (31) , the lipid hydroperoxide content as reported (32) , and REM as indicated above. Under the standard conditions used here, mildly oxLDL contained 32–48 nmol lipid hydroperoxide/mg apoB, and 3.6–6.5 nmol TBARS/mg apoB. The relative electrophoretic mobility was 1.1 ± 0.1.
Labeling of oxLDL with DiIC18 and determination of LDL uptake by
cells
OxLDL were fluorescently labeled with DiIC18 according to Via
and Smith (33)
, reisolated by ultracentrifugation,
dialyzed, and sterilized again as indicated above. DiIC18-labeled LDL
were added to the culture medium (100 µg apoB/ml) and incubated with
cells for the times indicated. Cells were then washed twice with PBS.
Afterward cells were homogenized by sonication in 1 ml of distilled
water and an aliquot was used to extract DiIC18 and to read the
cell-associated fluorescence (Jobin-Yvon JY-3C spectrofluorometer;
excitation 545 nm, emission 568 nm).
Evaluation of necrosis and apoptosis
The percentage of morphologically apoptotic cells was
counted after staining by May-Grünwald-Giemsa or DAPI, as
previously used (14)
. DNA fragmentation was visualized in
situ by the TUNEL (terminal transferase-mediated dUTP-biotin nick end
labeling) procedure (34)
, using the terminal transferase
(TdT) kit of Boehringer Mannheim (Germany). Briefly, after
centrifugation on glass slides (using cytocentrifuge, Heraeus Labofuge
GL), cells were fixed in 3% buffered paraformaldehyde. Endogenous
peroxidases were inactivated by 2% H2O2; after
rinsing, 150 µl of TdT (0.3 U/µl) and biotinylated dUTP in TdT
buffer (150 mmol/l K cacodylate/25 mmol/l Tris-HCl pH 6.6, 0.25 mg/ml
bovine serum albumin), and 2 mmol/l CoCl2 were added for
1 h at 37°C. After stopping the reaction and rinsing four times
as indicated in ref 20,
the slides were covered by 150 µl of
Extra-avidin Peroxidase (Sigma) diluted 1/15 in water and incubated for
30 min at 37°C, washed twice, and stained with 1 mg/ml DAB for 5 min
at 37°C. The positive control was treated by DNAse I (1 µg/ml from
Sigma, for 10 min) before being processed through the TUNEL procedure.
Necrosis (characterized by cell membrane disruption) was evaluated by the trypan blue exclusion test (14) and by determining lactate dehydrogenase (LDH) released into the culture medium (Roche assay kit, MA kit 10).
Alternatively, necrosis and apoptosis were evaluated concomitantly on intact cultured cells (grown in 24-multiwell plates), after fluorescent staining by using two vital fluorescent dyes—0.6 µM SYTO-13 (a permeant DNA intercalating green-colored probe) and 15 µM propidium iodide (a nonpermeant intercalating orange probe)—and counted by using an inverted fluorescence microscope (Fluovert FU, Leitz). Normal nuclei exhibited a loose chromatin colored in green by SYTO; apoptotic nuclei exhibited condensed green-colored chromatin and/or fragmentation (postapoptotic necrosis being characterized by nuclei exhibiting the same apoptotic morphological features, but orange colored); necrotic cells exhibited orange-colored nuclei with loose chromatin and were also stained by trypan blue.
Determination of chromatin fragments and visualization of DNA
ladder
Chromatin fragments were determined by the procedure of McConkey
et al. (15)
. Briefly, cells were allowed to lyse for 15
min in 1 ml lysis buffer (5 g/l triton X-100 and 20 mmol/l EDTA, 5
mmol/l Tris pH 8.0), then ultracentrifuged for 20 min at 27,000 x
g to separate the chromatin pellet from cleavage products.
The pellet (resuspended in 1 ml of 10 mmol/l Tris-HCl pH 8.0 buffer
containing 1 mmol/l EDTA) and the supernatant were assayed for DNA
determination by the fluorometric DAPI procedure according to
Kapuscinski et al. (35)
.
To visualize DNA laddering, DNA was extracted using the G-Nome kit and labeled according to the procedure of Rösl (36) : 0.5–1 µg DNA was treated with 5 U of Klenow polymerase using 0.5 µCi of [32P]dCTP in a 10 mmol/l Tris/HCl pH 7.5 buffer containing 5 mmol/l MgCl2. After labeling, DNA was precipitated three times by 2.5 mol/l ammonium acetate/isopropanol, resuspended in 10 mmol/l Tris/ HCl pH 7.5 containing 1 mmol/l EDTA, used for electrophoresis (on 1.8% agarose gel), and revealed by autoradiography. Alternatively, DNA fragmentation was visualized by ethidium bromide fluorescent staining on electrophoresis gel.
Western blot analysis
Bcl-2 (and ß-actin used as control) was detected by
Western blotting (37)
under conditions used previously
(20)
. Briefly, cells were prepared in lysis buffer (20
mmol/l Tris-HCl pH 7.4 containing 50 mmol/l NaCl, 50 mmol/l NaF, 20
mmol/l Na pyrophosphate, 5 mmol/l EDTA, 1 mmol/l Na orthovanadate, 100
µg/ml phenyl-methyl-sulfonyl fluoride, 1 µg/ml leupeptin, 10 mg/ml
triton-X100, 1 mg/ml Na dodecylsulfate), and sonicated (MSE
Soniprep-150) for 5 s and centrifuged at 13,000 x
g for 10 min at 4°C. An aliquot (5 µg and 30 µg as
protein for lymphocytes and HL-60) of the supernatant was used for
separation on 12% sodium dodecyl sulfate polyacrylamide gel
electrophoresis (100 V for 90 min). After overnight blocking with 2%
non-fat milk in 0.15 mol/l PBS (pH 7.4) containing 0.05% Tween
(PBS-Tween-20) at 4°C, the membrane was incubated for 90 min at room
temperature with anti-Bcl-2 monoclonal antibody (Dako; dilution 1:200)
or anti-ß-actin (Sigma; dilution 1:1000). After washing in PBS-Tween,
the membrane was incubated with the secondary antibody (antimouse sheep
immunoglobulin G conjugated with horseradish peroxidase; dilution
1:1000) for 1 h at room temperature and developed by ECL
(Amersham).
Determination of cytosolic calcium concentration
[Ca2+]i
[Ca2+]i was determined by using
the permeant calcium probes fura-2/AM. Briefly, cells were incubated
for 15 min at 37°C in culture medium buffered with 20 mmol/l Hepes
and containing 0.5% bovine serum albumin and fura-2/AM (5 µmol/l).
After dilution and incubation for 45 min, cells were washed twice in
PBS and their fluorescence was recorded.
[Ca2+]i determination was performed at the
dual excitation wavelength of 340 and 380 nm and emission at 510 nm.
[Ca2+]i was calculated by the ratio method
(38)
. Statistical significance was estimated by the
Student's t test.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To evaluate the influence of Bcl-2 on cell death triggered by oxLDL, we used various model systems in which the level of Bcl-2 protein was altered either by antisense treatment of lymphocytes (expressing high initial Bcl-2 level) or by overexpression of Bcl-2 protein in HL-60 cells (expressing low level of Bcl-2).
Effect of Bcl-2 antisense oligonucleotides on oxLDL-mediated cell
death
The basal level of Bcl-2 protein is relatively high in peripheral
blood lymphocytes (PBL). Bcl-2 level reduced by AS oligonucleotides
whereas SC oligonucleotides did not induce any modification of Bcl-2
protein levels (Fig. 1A
).
|
The toxicity of oxLDL to lymphocytes was time dependent (Fig. 1B ). Under the conditions used (200 µg apoB/ml of oxLDL), cell death became obvious at 16 h, necrosis prevailing clearly over apoptosis (Fig. 1B ) since, after 24 h incubation with 200 µg apoB/ml of oxLDL, 54 ± 5% of lymphocytes underwent necrosis (trypan blue staining) and 21 ± 3% apoptosis (morphological apoptosis by MGG staining). Similar results were obtained with LDH release and chromatin fragmentation as necrosis and apoptosis markers, respectively (data not shown). The toxic effect was also dose dependent; under the conditions used, the toxic effect was apparent with concentrations of oxLDL higher than 100–120 µg apoB/ml (data not shown). Moreover, the concomitant staining by SYTO and propidium iodide showed that, at 16 h, the major part of necrotic cells (orange-colored with loose chromatin) did not exhibit any nuclear features of apoptosis, suggesting that a large portion of dying cells underwent primary necrosis (around 2/3) and the remaining minor part (1/3) underwent apoptosis. At 24 h and later, apoptotic cells are progressively stained by trypan blue, thus suggesting the occurrence of postapoptotic necrosis (data not shown).
When the Bcl-2 level of lymphocytes was reduced by AS oligonucleotide treatment (10 µmol/l) (Fig. 1A ), oxLDL (200 µg apoB/ml) induced concomitantly a drop in necrosis and a rise in apoptosis, but the total toxicity (necrosis plus apoptosis) was approximately similar (Fig. 1C, D ). In contrast, the level of necrosis and apoptosis was quite similar in SC-treated and untreated lymphocytes (Fig .1C, D). The other (as used above) methods for evaluating necrosis and apoptosis yielded similar results.
The results obtained by morphological methods are consistent with the data of DNA laddering (Fig. 1E ). The higher radiolabeling of the second line AS + oxLDL (in comparison to that of oxLDL alone) suggests that AS treatment promotes apoptosis since, under the conditions used (each lane of electrophoresis containing the same amount of cellular DNA and being treated by the same dCTP-Klenow polymerase procedure), the intensity of radiolabeling is related to the number DNA strand breaks and laddering indicated apoptotic internucleosomal breaks.
These data show that, in blood lymphocytes, the Bcl-2 level influences the ratio necrosis/apoptosis induced by toxic contractions of oxLDL, but does not change their overall toxicity.
OxLDL-induced cell death in HL60/B and HL60
To strengthen the above conclusions, we used another model
system consisting in HL60-derived cell lines expressing different
levels of the Bcl-2 protein. As assessed by Western blot and
immunofluorescence (Fig. 2A
), the (parental) HL60 cell line expresses low Bcl-2 level,
whereas HL60/B cells (transduced with Bcl-2) express relatively high
Bcl-2 levels.
|
Both HL60 and HL60/B cell lines were susceptible to the toxic effect of oxLDL. HL60/B cells were more susceptible than HL60 since, after 24 h incubation with oxLDL (200 µg apoB/ml), the overall toxicity (necrosis + apoptosis) was around 80–90% for HL60/B vs. 50–60% for HL60 (Fig. 2B, C ). The major part of HL60/B dead cells exhibited the characteristic features of primary necrosis, whereas HL60 cells underwent mainly apoptosis (Fig. 2B, C ). Both morphological (illustrated in Fig. 2D ) and biochemical (Fig. 2B, C, E ) tests used concomitantly (trypan blue staining and LDH release for necrosis, Fig. 2B ; morphological apoptosis, TUNEL labeling and chromatin fragmentation for apoptosis, Fig. 2C, D ) were consistent with the above conclusion. This was also supported by DNA electrophoresis (Fig. 2E ): oxLDL induced the characteristic apoptotic DNA laddering in HL60 cells, and a typical DNA smear (consistent with random DNA degradation occurring during necrosis) in HL60/B.
These data strongly suggest that overexpression of Bcl-2 1) reduces the rate of oxLDL-induced apoptosis, but raises concomitantly the rate of necrosis; 2) does not prevent oxLDL-induced cell death. Therefore, Bcl-2 levels seems to regulate a balance between oxLDL-induced necrosis and apoptosis. The mechanism of this balance in oxLDL-induced cell death is unknown.
Uptake of oxLDL by HL60/B and control HL60
Since the type of cell death may hypothetically be directed by the
level of toxic agent and the toxic effect is related to the level of
oxLDL taken up by cells (39)
, we compared the uptake of
oxLDL by HL60/B and HL60 cells in order to investigate whether the
difference of toxicity of oxLDL to HL60/B and HL60 may result from
different uptake rates. As shown in Fig. 3
, the uptake rates of oxLDL by HL60/B and HL60 were quite similar
and, as expected, were not related to the level of Bcl-2 protein.
Therefore, the difference of susceptibility of HL60/B and HL60 cannot
be explained by a difference in the rate of LDL uptake.
|
Role of calcium in necrosis and apoptosis in cells with variable
levels of Bcl-2 protein
As Bcl-2 overexpression was associated with a decrease of
apoptosis and an increase of necrosis induced by oxLDL, it was
hypothesized that oxLDL may trigger signaling pathways leading on the
one hand to necrosis and on the other hand to apoptosis. As the toxic
effect of oxLDL is mediated by a sustained calcium rise
(14
, 20
) and Bcl-2 has been shown to repress
apoptosis by regulating endoplasmic reticulum-associated
Ca2+ fluxes (23)
, we investigated whether the
level of Bcl-2 protein may influence the sustained calcium signal
triggered by oxLDL. For this purpose, we used PBL expressing high Bcl-2
level and undergoing oxLDL-induced necrosis and ECV-304 EC expressing
very low Bcl-2 level and undergoing oxLDL-induced apoptosis
(20)
. HL60-derived cell lines cannot be used because
calcium-depleted medium was highly toxic to this cell lines.
As shown in Fig. 4 , comparison between cells expressing low (ECV-304 EC) and high Bcl-2 levels showed that the oxLDL-induced calcium rise is independent of Bcl-2 levels. The calcium chelator EGTA and the calcium channel blocker nisoldipine were able to block both the calcium rise and cell death triggered by oxLDL (Fig. 4A–D ) in both cells undergoing necrosis (lymphocytes) and those undergoing apoptosis (ECV-304). Conversely, the calcium ionophore A23187 elicited a progressive calcium rise (data not shown) in the two cell lines and triggered necrosis of lymphocytes and apoptosis of ECV-304 cells (Fig. 4E, F ). As expected, the toxic effect of A23187 was blocked by the calcium chelator EGTA, which impedes the calcium entry (Fig. 4E, F ), but was not inhibited by nisoldipine (data not shown).
|
These data strongly suggest that (at least in the cellular model system used here) the sustained calcium peak (evoked by oxLDL), is a common trigger for necrosis and apoptosis and that Bcl-2 acts downstream from the calcium peak by inhibiting only the apoptotic pathway (without blocking the necrosis process) (Fig. 4G ).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Besides the classical methods used for evaluating necrosis and apoptosis, preliminary experiments allowed us to define new morphological methods by using concomitantly two fluorescent DNA intercalating probes (the classical orange-colored nonpermeant propidium iodide and the green-colored SYTO-13 permeant fluorescent DNA probe). Four conditions of fluorescently labeled cells were discriminated on the basis of morphological criteria and of the working definition of necrosis (increased permeability of the plasma membrane leading to orange colored staining by propidium iodide) and apoptosis (chromatin condensation, nucleus fragmentation): 1) orange-colored nucleus with near normal morphology (i.e., loose chromatin without any apoptotic feature) indicates primary necrosis or oncosis); 2) green-colored nucleus with morphological features of apoptosis indicates apoptosis; 3) orange-colored nucleus with morphological features of apoptosis indicates that the cell is affected by both apoptosis and alteration of the plasma membrane, and is regarded as postapoptotic necrosis (or secondary necrosis); 4) green nucleus with normal morphology indicates a normal (living) cells. As this method includes nontoxic vital dyes, it can be used to observe the time course of toxic events in the cell culture. Moreover, it allows one to define the state of each individual cell (in contrast to biochemical methods, which provide mean information).
Evaluation of apoptosis is probably underestimated by counting morphological apoptotic cells, because this method ignores cells engaged in early steps of the apoptotic process (when the morphological changes of the nucleus—i.e., chromatin margination and moderate condensation—are visible only by electron microscopy) and cells completely disrupted into small fragments (apoptotic bodies), characteristic of late steps of apoptosis.
In apoptosis-prone cell populations—e.g., HL-60 and ECV-304 EC (20) —apoptosis began at 12–14 h, then rose progressively. At 16–20 h, only a few apoptotic cells were stained by trypan blue (or propidium iodide), but many apoptotic cells were stained by trypan blue at 24–30 h, which suggests that apoptosis was followed by postapoptotic necrosis. In our experimental model system, DNA laddering agreed with morphological data and TUNEL labeling (although it has been recently reported that internucleosomal DNA cleavage is not strictly specific to apoptosis and may occur in the early steps of necrosis) (17) .
The data reported suggest that 1) oxLDL elicited mainly apoptosis in cells expressing low levels of Bcl-2 protein; and 2) Bcl-2 overexpression did not really prevent oxLDL-induced cell death (but rather potentiated the toxic effect of oxLDL in HL60/B), because the decrease of apoptosis is counterbalanced by a dramatic increased of necrosis. Our results are (in part) consistent with the protective effect of Bcl-2 against oxysterol-induced apoptosis (26 , 27 ), but the balance between apoptosis and necrosis induced by oxLDL has never been reported (probably because only apoptosis was evaluated in the above-quoted studies; 26, 27). The lack of effect of Bcl-2 against oxLDL-induced necrosis is consistent with the observation that Bcl-2 is not a universal inhibitor of cell death (21 , 42 ).
Moreover, together these data suggest that oxLDL are able to trigger cell death through various pathways: one, which is Bcl-2 sensitive (26 , 27 ), leads to apoptosis and is possibly mediated by CPP-32 (43) ; the other is Bcl-2 insensitive and leads to necrosis (14) . The sustained and intense calcium peak evoked by oxLDL, which is a common trigger of necrosis and apoptosis induced by oxLDL, activates these two separate death pathways. This is consistent with the data obtained in other model systems since high calcium levels may trigger both necrosis (possibly through calcium-dependent phospholipases and proteases) (12-14 , 44 ) and apoptosis (possibly mediated by calcium-dependent proteases, which in turn may induce caspase activation) (44) .
In the experimental model used here, high Bcl-2 levels inhibited apoptosis but did not prevent the calcium peak triggered by oxLDL. Therefore, the antiapoptotic effect of Bcl-2 cannot be explained by a regulatory activity of calcium homeostasis (23) , but rather by the inhibition of a late step of the apoptotic signaling (for instance, by inhibiting cytochrome c release and caspase 3 activation; 24, 25). The calcium peak is also a trigger for necrosis (this paper and ref 14 ) and Bcl-2 overexpression does not inhibit the calcium peak evoked by oxLDL, which led us to hypothesize that, in cells overexpressing Bcl-2, the oxLDL-induced calcium peak (not blocked by Bcl-2) activates both cell death pathways, but only necrosis occurs because the apoptotic pathway is blocked (downstream the calcium peak, as illustrated in Fig. 4 ).
Experiments using the calcium chelator EGTA and the calcium ionophore A23187 were designed in order to evaluate more precisely the role of the oxLDL-induced calcium rise in the cell death pathways and distinguish it from the effect of the other intracellular signaling pathways activated by oxLDL (18 , 45-49 ). These experiments support the hypothesis that the oxLDL-induced calcium rise is a common trigger of both cell death pathways since EGTA inhibits, at least in part, both oxLDL-induced apoptosis and necrosis, and A23187 mimics the toxic effect of oxLDL, i.e., induces apoptosis in cells expressing low level of Bcl-2 and necrosis in cells expressing high Bcl-2 levels.
Finally, the following pathophysiological scenario may be suggested. OxLDL activate various intracellular signaling pathways (18 , 45-49 ), some culminating (after a yet unknown signaling cascade) in a sustained calcium rise (14 , 18 ). This calcium rise constitutes a lethal hit (12 , 13 , 44 ) that acts by activating both necrosis and apoptosis pathways. In cells expressing low Bcl-2 levels, both pathways seem to be active since cells undergo apoptosis, followed relatively soon by postapoptotic necrosis. In cells expressing high Bcl-2 levels, only the necrosis pathway is working (since the apoptotic pathway being inhibited by Bcl-2). This may explain why high Bcl-2 levels do not actually increase the resistance of cells against the oxLDL-induced toxicity, but regulate the balance between necrosis and apoptosis.
From a pathophysiological point of view, when cells are prone to dye, it is likely that apoptosis is less harmful than necrosis since, in vivo, apoptotic cells are rapidly cleared by phagocytic cells and do not induce local inflammation, whereas necrosis is a proinflammatory event. For instance, during acute pancreatitis, the severity of this inflammatory disease involving massive cell death has been related to induction of necrosis, whereas apoptosis could minimize the severity of inflammatory processes (50) .
The (patho)physiological role of Bcl-2 in normal and atherosclerotic vascular wall is only poorly understood. In addition to proteins of the Bcl-2 family (26 , 27 , 51 , 52 ), various cellular factors (such as superoxide dismutase, p53) (53) may be involved in regulating apoptosis of vascular cells implicated in remodeling of the vascular wall and the susceptibility of cells to oxLDL-induced apoptosis. A prominent role of proteins of Bcl-2 family is supported by in vivo experiments showing that inhibition of Bcl-x expression induce apoptosis and regression of vascular disease (54) . This suggests that antiapoptotic members of the Bcl-2 family may be atherogenic.
Our data allow us to propose a plausible cellular mechanism and pathophysiological scenario to explain this atherogenic effect of Bcl-2. High Bcl-2 levels in vascular cells may participate in two atherogenic processes: 1) by inhibiting apoptosis, high Bcl-2 levels may favor the intimal accumulation of smooth muscle cells (the proliferation of which is also stimulated by growth factors and low levels of oxLDL) (2 , 46 ); and 2) by shifting the balance from apoptosis toward necrosis (induced by toxic concentrations of oxLDL), high Bcl-2 levels may favor inflammatory events, promote formation or enlargement of the necrotic core of atherosclerotic plaques, and therefore enhance the risk of plaque rupture and thrombosis.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Abbreviations: apoB, apoprotein B; AS, antisense;
[Ca2+]i, cytosolic calcium concentration;
DAPI, 4',6-diamidino-2-phenylindole; LDL, low density lipoproteins;
oxLDL, mildly oxidized LDL; FCS, fetal calf serum; TBARS,
thiobarbituric acid-reactive substances; PBL, peripheral blood
lymphocytes; PBS, phosphate-buffered saline; LDH, lactate
dehydrogenase; REM, relative electrophoretic mobility; SC, scrambled;
TdT, terminal transferase; TUNEL, terminal transferase-mediated
dUTP-biotin nick end labeling.
Received for publication September 23, 1998.
Revision received November 4, 1998.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
-32P]dCTP
incorporation by Klenow polymerase. M, 100 bp marker (Promega).
A, B, D) Mean ±SEM of three separate
experiments. C, E, F) Representative experiments.
