Oxidized LDLs alter the activity of the ubiquitin-

proteasome pathway: potential role in oxidized LDL-

induced apoptosis
OTÍLIA VIEIRA,*,† ISABELLE ESCARGUEIL-BLANC,* GÜNTHER JÜRGENS,‡
CHRISTOPH BORNER,§ LEONOR ALMEIDA,† ROBERT SALVAYRE,*,1 AND
ANNE NÈGRE-SALVAYRE*,1
*INSERM U.466, Biochemistry Department, University Paul Sabatier, Toulouse, France; †Laboratorio
de Bioquimica, Faculdade de Farmacia and Centro de Neurociências, Universidade de Coimbra,
3000 Coimbra, Portugal; ‡Institute of Medical Biochemistry, Karl-Franzen Universität Graz, Austria;
and §Institute of Biochemistry, University of Fribourg, Switzerland

ABSTRACT Oxidized low-density lipoproteins (ox-
LDL) play a role in the genesis of atherosclerosis.
OxLDL are able to induce apoptosis of vascular
cells, which is potentially involved in the formation
of the necrotic center of atherosclerotic lesions,
plaque rupture, and subsequent thrombotic events.
Because oxLDL may induce structural modifications
of cell protein and altered proteins may impair cell
viability, the present work aimed to evaluate the
extent of protein alterations, the degradation of
modified proteins through the ubiquitin-proteasome
system (a major degradative pathway for altered and
oxidatively modified proteins) and their role during
apoptosis induced by oxLDL. This paper reports the
following: 1) oxLDL induce derivatization of cell
proteins by 4-hydroxynonenal (4-HNE) and ubiquiti-
nation. 2) Toxic concentrations of oxLDL elicit a
biphasic effect on proteasome activity. An early and
transient activation of endogenous proteolysis is
followed rapidly by a subsequent decay (resulting
probably from the 26S proteasome inhibition) and
followed later by the inhibition of the 20S protea-
some (as assessed by inhibition of sLLVY-MCA hy-
drolysis). 3) Specific inhibitors of proteasome (lac-
tacystin and proteasome inhibitor I) potentiated
considerably the toxicity of oxLDL (nontoxic doses
of oxLDL became severely toxic). The defect of the
ubiquitination pathway (in temperature-sensitive mu-
tants) also potentiated the toxicity of oxLDL. This
suggests that the ubiquitin-proteasome pathway plays
a role in the cellular defenses against oxLDL-in-
duced toxicity. 4) Dinitrophenylhydrazine (DNPH),
an aldehyde reagent, prevented both the oxLDL-
induced derivatization of cell proteins and subse-
quent cytotoxicity. Altogether, the reported data
suggest that both derivatization of cell proteins (by
4-HNE and other oxidized lipids) and inhibition of
the proteasome pathway are involved in the mecha-
nism of oxLDL-induced apoptosis.—Vieira, O., Es-
cargueil-Blanc, I., Jürgens, G., Borner, C., Almeida,
L., Salvayre, R., Nègre-Salvayre, A. Oxidized LDL
alter the activity of the ubiquitin-proteasome path-
way: potential role in oxidized LDL-induced apopto-
sis. FASEB J. 14, 532–542 (2000)
Key Words: oxidized LDL z 4-hydroxynonenal z proteasome
z ubiquitin z apoptosis
Atherosclerosis and subsequent vascular dis-
eases are the first cause of morbidity and mortality in
Western countries. Atherosclerotic lesions associate
to variable degrees with accumulation of lipid laden
macrophagic cells, proliferating smooth muscle
cells, fibrosis, and necrotic areas in the subendothe-
lial space of the arterial wall (1, 2). Low-density
lipoproteins (LDL) play an important role in athero-
genesis (3) and are thought to become atherogenic
after undergoing oxidative modifications (4). LDL
oxidation is mediated by free radicals generated by
vascular cultured cells, transition metals, and heme
proteins such as ferrylmyoglobin (5, 8). During the
initial steps of the oxidative process, antioxidants are
consumed and lipid peroxidation begins to rise,
leading to the formation of mildly oxidized LDL
(characterized by relatively low levels of lipid peroxi-
dation products without or only minor changes in
apoB). Later, progression of the oxidative process
leads to the formation of extensively oxidized LDL
(containing high levels of lipid peroxidation prod-
ucts and severe apoB alterations), which is detected
in atherosclerotic plaques (4, 9).
Oxidized low-density lipoproteins (oxLDL) ex-
hibit a wide spectrum of biological properties and
are able to induce events potentially involved in
atherogenesis, such as monocytes chemotaxis,
1
Correspondence: INSERM U-466 and Biochimie, CHU
Rangueil, 1 Avenue Jean Poulhès, 31403 Toulouse Cedex 04,
France. E-mail: anesalv@rangueil.inserm.fr or salvayre@
rangueil.inserm.fr

532 0892-6638/00/0014-0532/$02.25 © FASEB

foam cells formation, endothelial dysfunction,
smooth muscle cell proliferation, and cytotoxicity
to cultured cells (4 –7, 10). Lipid peroxidation
products (e.g., oxysterols and aldehydes) con-
tained in oxLDL are able to elicit apoptosis or
necrosis of cultured cells (13, 14) through a
calcium-dependent pathway (14, 15). The primary
cellular targets and the precise mechanisms of
toxicity are only poorly understood.
OxLDL contain MDA, 4-HNE, hexanal, and other
aldehydes (6) able to form apoB- and cell proteins-
adducts that are detected in atherosclerotic areas (9,
16 –18). Derivatization may induce dramatic changes
in the functional properties of proteins; for instance,
apoB modifications alter LDL metabolism (4, 8) and
cell protein derivatization may lead to a loss (19) or
gain of function (20 –22), which may deregulate cell
functions. These events may play a significant role in
atherosclerosis and other pathophysiological pro-
cesses (such as inflammatory reactions, aging, and
related pathologies) (23, 24).
Cellular defense systems against oxidized proteins
consist in proteolytic degradation (25, 26) through a
multicatalytic proteinase complex (proteasome), ei-
ther dependent or independent of the ubiquitin
system (25–31). The 26S proteasome complex is
constituted by a central 20S proteolytic core (20S
proteasome) associated with two regulatory subcom-
plexes, termed PA700 or 19S and PA28 (or 11S
regulator) (26, 27). Ubiquitination of proteins is
performed by the ATP-dependent ubiquitin system,
and polyubiquitinated proteins are targeted to the
26S proteasome for degradation (28 –31). The ubiq-
uitin-dependent proteolytic pathway (26S protea-
some) is involved in the continuous turnover of
regulatory proteins and in the selective degradation
of misfolded and denatured proteins. It is thought to
play a major role in regulating numerous cell pro-
cesses (signal transduction, cell cycle progression,
transcription, endocytosis, apoptosis) and ‘detoxify-
ing’ altered proteins (25–31). Misfolded or ubiq-
uitin-tagged, or moderately oxidized or otherwise
structurally altered, proteins are valuable substrates
for the proteasome, but heavily oxidized proteins are
no longer degradable, accumulate, and may become
toxic (25).
As we have recently shown, 4-HNE contained in
oxLDL was able to derivatize the EGF receptor, and
we hypothesized that such a derivatization may mod-
ify the structure of a number of cell proteins, impair
their functions, and finally alter cell viability.
We report here that incubation of cells with ox-
LDL induces derivatization of cell proteins by 4-HNE
(4-HNE-protein adducts), ubiquitination of cell pro-
teins, and (de)regulation of proteasome activity (i.e.,
early activation followed by late inhibition). Inhibi-
tion of proteasome reduces the toxicity threshold of
UBIQUITIN-PROTEASOME PATHWAY IN oxLDL-INDUCED APOPTOSIS
oxLDLs and potentiates their toxic effect to ECV-304
cells, thus suggesting that, in these cells, accumula-
tion of altered proteins is involved in oxidized LDL-
induced apoptosis and that proteasome activity may
be involved in cellular defenses against oxidized
LDL-induced apoptosis.
MATERIALS AND METHODS
Chemicals and reagents
[3H] N-succinimidyl propionate (99.0 Ci/mmol), [35S]methi-
onine/cysteine mixture (ProMix [35S] cell labeling kit,
.1000 Ci/mmol), and ECL reagent were obtained from
Amersham (Les Ulis, France); N-succinyl-L-leucyl-L-leucyl-L-
valyl-Ltyrosine-7-amido-4-methyl coumarin (sLLVY-MCA)
from Bachem (Voisins-le-Bretonneux, France); Lactacystin
(Lc) and Proteasome Inhibitor I (PSI) from Calbiochem
(Nottingham, UK); anti-ubiquitin (polyclonal) antibody,
horse heart myoglobin, hydrogen peroxide, LLnL (N-acetyl-
Leucyl-leucyl-norleucinal), and buthionine sulfoximine from
Sigma (St. Louis, Mo.); and RPMI 1640, L-glutamine, penicil-
lin, and streptomycin from Life Technologies (Cergy-Pon-
toise, France). All other chemicals (grade for biochemical
analyses) were purchased from Merck (Darmstadt, Germany),
Sigma, or Prolabo (Paris). Before use, metmyoglobin was
purified by dialysis against phosphate buffered saline (PBS)
pH 7.4 containing 50 mM DTPA and Chelex-100. Stock
metmyoglobin and H2O2 solutions were standardized using
ε632 nm 5 2.1 mM21zcm21 and ε240 nm 5 43.6 mM21zcm21,
respectively.
LDL isolation and oxidation
LDL were isolated from human pooled and heat inactivated
(1 h at 56°C) sera by ultracentrifugation, under the previously
described conditions (16, 32), sterilized on 0.2 mM Millipore
membrane, and stored at 4°C under nitrogen (up to 4 wk).
The electrophoretic mobility of LDL was evaluated on aga-
rose gel (Hydragel®; Sebia, Paris), and apoB was determined
by immunonephelometry under the previously used condi-
tions (15, 16).
LDLs were oxidized by incubation with metmyoglobin/
H2O2 (18/27 mM) in PBS, at 37°C for 2 h, under the
previously used conditions (32). The level of LDL oxidation
was evaluated by monitoring lipid hydroperoxides (33), thio-
barbituric reactive substances (34), and 4-HNE content (35).
Lipid peroxidation levels of mildly oxLDL used here
ranged between 52 and 68 nmol lipid hydroperoxides/mg
apoB, between 5 and 8 nmol TBARS/mg apoB, and between
12 and 15 nmol 4-HNE/mg apoB, without major modifica-
tions of apoB (32).
Cell culture
ECV-304 human endothelial cell line (CRL-1998; ATCC,
Rockville, Md.) were grown under the previously described
conditions (15, 32). Briefly, all passages were made using a
splitting ratio 1:4. Cells were seeded (two 105 cells/ml) in six
multiwell plates or in falcons (Nunc, Roskilde, Denmark) and
grown in RPMI 1640 medium containing Glutamax® supple-
mented with 10% heat inactivated fetal calf serum, 100 U/ml
penicillin, and 100 mg/ml streptomycin (in 5% CO2, at
37°C). After starved in serum-free medium for 24 h before
LDL addition, subconfluent cell cultures were incubated with
533
oxLDL or native LDL, and inhibitors (PSI dissolved in
ethanol and DNPH in DMSO— each solvent used at 0.1%
final—were added at time 0, just before LDL addition) under
the conditions indicated below.
H38 –5 and ts20 fibroblasts were a generous gift of Dr. H. L.
Ozer (36). ts20 is a Balb/C 3T3 clone A31 fibroblast cell line
that is temperature-sensitive for the E1 ubiquitin activating
enzyme (i.e., E1 activity and ubiquitination are drastically
decreased at 39°C). H38 –5 is an E1-transfected (corrected)
ts20 derivative that expresses E1 and ubiquitinates at all
temperature. ts20-pMV12 and ts20Bcl-2#7 cell lines (which
exhibit a similar decrease in ubiquitination at 39°C as the
parental ts20 cell line) were generated in the laboratory of Dr.
C. Borner (37) by retroviral transduction of the pMV12 hygro
plasmid or the pMV12 hygro plasmid containing the mouse
Bcl-2 cDNA, respectively. The ts20Bcl-2#7 cell line overex-
presses Bcl-2 5- to 10-fold over controls. Cells were grown at 34
or 39°C as described previously (36, 37). Subconfluent cells
were starved in serum-free RPMI 1640 for 24 h at 34°C before
LDL were added, and the temperature was shifted to 39°C for
18 h.
In situ proteolysis measurements and in vitro determination
of proteasome activity
The degradation of cellular proteins was determined under
the conditions described by Grüne et al. (38). Briefly, ECV-
304, grown in six multiwell plates, were preincubated with a
[35S]methionine/cysteine mixture (0.5 mCi/ml) in methi-
onine-free MEM culture medium for 2 h (short-lived pro-
teins) or 16 h (long-lived proteins). Then, short-lived– and
long-lived–labeled cells were chased in standard RPMI 1640
medium containing 10 mmol/l of unlabelled methionine for
10 min and 2 h, respectively; then, cells were incubated with
oxLDL or native LDL, washed twice in PBS, scrapped off and
pelleted by centrifugation at 1,500g for 10 min. Cell proteins
were then precipitated by 10% trichloroacetic acid (TCA) for
30 min at 4°C and after centrifugation (15,000g for 10 min),
the radioactivity of TCA-soluble and TCA-precipitable frac-
tions (precipitate dissolved in 50 ml of NaOH 1N) was
counted by liquid scintillation counting (Aquasafe®, Packard
Tricarb 4530, Downers Grove, Ill.).
The in vitro activity of the 20S proteasome was determined
according to Grüne et al. (38). Cells were harvested, pelleted,
resuspended in PBS containing 0.1% Triton X-100 and 0.5
mM dithiotreitol, homogenized (sonication, two runs of 5 s,
Bransonic sonicator) and used immediately for determining
the enzymatic activity. The assay mixture contained 50 ml of
buffer (50 mM Tris-HCl pH 7.8, 20 mM KCl, 5 mM MgCl2,
and 0.1 mM DTT), 250 mM of sLLVY-MCA, and 50 ml of cell
lysates (15 mg of proteins). After 30 min at 37°C, the reaction
was stopped by adding 1 ml of 0.2 M glycine buffer pH 10 and
the fluorescence of the liberated 7-amino-4 methylcoumarin
was measured (spectrofluorometer Jobin-Yvon, excitation 365
nm, emission 460 nm). An aliquot of the cell homogenate was
used for protein determination using the biscinchoninic
reagent.
Western-blots experiments
ECV-304 cells, treated or not by oxLDL, were washed,
scrapped off in PBS, centrifuged (2,000g for 5 min at 4°C)
and lysed in solubilizing buffer (50 mM Tris pH 7.4, 250 mM
NaCl, 5 mM EDTA, 1 mM Na3VO4, 10 mM Na pyrophos-
phate, 160 mM NaF, 2.5 mM PMSF, 10 mM leupeptin, 2 mM
pepstatin, 10 mg/ml aprotinin, 1% triton X-100) for 30 min
on ice. Fifty micrograms of cell lysates were resolved by
electrophoresis in a 7.5% SDS-polyacrylamide gel, transferred
534 Vol. 14 March 2000 The FASEB Journal
onto nitrocellulose membranes (Hybond-C, Amersham),
probed with an anti-ubiquitin antibody (Sigma) or an anti-4-
HNE-protein antibody (K5– 4412), and revealed by ECL sys-
tem (Amersham) using a peroxidase-coupled secondary anti-
body, as previously used (22).
Determination of free amino group content in cell proteins
The free amino group content of cell proteins was evaluated
on cells homogenates using the amino-reactive probe [3H]N-
succinimidyl propionate ([3H]NSP) (39), under the previ-
ously used conditions (22). Briefly, after incubation, cells
were washed three times in PBS and homogenized in 0.5 M
borate buffer pH 8.5 (sonication, two runs of 5 s). An aliquot
of the cell suspension was saved for protein determination.
Cell homogenates were let to react with 10 mCi of [3H]NSP
(Amersham, 99.0 Ci/mmol) in 0.5 M borate buffer pH 8.5,
for 15 min, in an ice bath (22). Then, cell proteins were
precipitated by TCA and the radioactivity counted as above
indicated.
Determination of cytotoxicity and apoptosis
The whole cytotoxicity was evaluated by using the MTT test
(40). The number of morphologically apoptotic or/and
necrotic cells was evaluated concomitantly on intact cultured
cells (grown in six multiwell plates) according to the fluores-
cent double-staining we recently described (41). Briefly, cells
were incubated with two vital fluorescent dyes, 0.6 mM
SYTO-13 (a permeant DNA intercalating green-colored
probe) and 15 mM propidium iodide (a nonpermeant inter-
calating orange probe) and counted by using an inverted
fluorescence microscope (Fluovert FU; Leitz, Rockleigh,
N. J.). Normal nuclei exhibited a loose chromatin colored in
green by SYTO; apoptotic nuclei exhibited condensed green-
colored chromatin and/or fragmentation (postapoptotic ne-
crosis being characterized by nuclei exhibiting the same
apoptotic morphological features but orange-colored); ne-
crotic cells exhibited orange-colored nuclei with loose chro-
matin. It may be noted that necrotic cells (orange-colored by
propidium iodide) were generally stained by trypan blue.
Alternatively, the morphology was also examined after May-
Grünwald-Giemsa staining, as previously used (15).
Biochemical methods were also used in order to evaluate
the level of apoptosis and necrosis in the whole cell popula-
tion. Chromatin fragmentation, evaluated by the procedure
of McConkey et al. (42), and lactate deshydrogenase released
into the culture medium (Roche assay kit, MA kit 10), were
determined under the previously described conditions (14,
15). The results were generally consistent with morphological
counts (the data presented here were selected in order to
avoid redundancy).
RESULTS
oxLDL induce derivatization and ubiquitination of
cellular proteins
In ECV-304 cells, oxLDL induced a progressive
time- and dose-dependent decay of the level of
free [3H]NSP-reactive amino groups, whereas na-
tive LDL did not (Fig. 1A, B). As shown by western
blots revealed by anti-4-HNE antibody, the loss of
[3H]NSP-reactive amino groups induced by ox-
VIEIRA ET AL.
Figure 1. In situ modifications of cell proteins induced by oxLDL in ECV-304 cells. A) Time course of [3H]NSP-reactive amino
groups in cells incubated with 200 mg apoB/ml of native LDL (empty circles) or oxLDL (filled circles) or 1 mM of 4-HNE (filled
squares). B) Effect of increasing concentrations of oxLDL (filled circles) or native LDL (empty circles), incubated for 5 h with
the cells, on the level of [3H]NSP-reactive amino groups. In panels A and B, the results are expressed as percent of the initial
value (control); means 6 se of three separate experiments. C, D) Detection of 4-HNE-adducts (C) or ubiquitinated proteins (D)
in lysates of cells incubated for the indicated time without (Co, control) or with 200 mg/ml of oxLDL (oxL) or native (natL)
or FeMb or 1 mM 4-HNE. Western-blot experiments were done using anti-4-HNE-protein or anti-ubiquitin or anti-b-actin (as
control) antibodies, used under the conditions indicated in Materials and Methods.

LDL was associated with derivatization of cell
proteins by 4-HNE, a lipid peroxidation derivative
able to react with free amino groups of proteins
(Fig. 1C). This derivatization was mimicked by
oxLDL lipid extracts (data not shown) and by
4-HNE (1 mM) (Fig. 1A, C).
In the same time, an increased level of high-
molecular-mass (HMM) ubiquitin-protein conju-
gates (HMM . 200 kDa) was observed in cells
incubated with oxLDL but not in cells incubated
with native LDL or FeMb (used alone) (Fig. 1D).
Thus, in cells treated by oxLDL, some proteins are
probably recognized as structurally abnormal and
tagged with ubiquitin probably for subsequent deg-
radation through the ubiquitin-dependent proteo-
lytic pathway. This led us to evaluate the proteasome
activity in cells incubated with oxLDL.
Biphasic effect of toxic concentrations of oxLDL
on the proteasome activity
OxLDL induced a time- and dose-dependent tran-
sient activation of in situ (i.e., in intact living cell)
intracellular proteolysis of both short-lived and long-
lived [35S]-radiolabelled proteins (Figs. 2A, B). In
contrast, native (nonoxidized) LDL did not. Toxic
concentrations of oxLDL (200 mg apoB/ml) evoked
a rapid and transient peak of in situ proteolysis
(maximum between 1 and 3 h) followed by a return
to the ground level at 5–7 h. It may be noted that
lower oxLDL concentration (50 mg apoB/ml) also
induced a transient peak of intracellular proteolysis
that occurred later (maximum between 5 and 7 h)
and then return slowly to the basal level (at 15 h)
(Fig. 2A).

UBIQUITIN-PROTEASOME PATHWAY IN oxLDL-INDUCED APOPTOSIS 535


Figure 2. Effect of oxLDL on the proteasome activity (i.e., in
situ proteolysis and in vitro hydrolysis of sLLVY-MCA) in
ECV-304 cells. A) Time course of short-lived (squares) and
long-lived (triangles) protein degradation in cells prelabeled
with [35S]methionine/cysteine for 2 h (short-lived) or 16 h
(long-lived) and incubated with 200 or 50 mg/ml of OxLDL
(filled and empty symbols, respectively). B) Effect of increas-
ing concentrations of oxLDL or native LDL (filled and empty
symbols, respectively) (incubation time 5 h) on short-lived
(squares) and long-lived (triangles) protein degradation. C)
Time course of in vitro sLLVY-MCA hydrolysis by lysates from
cells incubated for the indicated time with 200 mg/ml of
oxLDL or native LDL (filled circles and empty squares,
respectively). As a comparison, the in situ proteolysis, trig-
gered under the same conditions (i.e., by 200 mg/ml of
oxLDL), is indicated by the dashed line. D) Effect of protea-
some inhibitors, LLnL (5 mM) and PSI (5 mM) on the in situ
proteolysis (black bars) and on the in vitro hydrolysis of
LLVY-MCA (hatched bars) (both activities being determined
at 3 h under standard conditions). Means 6 se of at least four
separate experiments.
Toxic concentrations of oxLDL (200 mg apoB/ml)
induced also a biphasic effect on the in vitro hydro-
lysis of sLLVY-MCA (a fluorogenic synthetic peptide
substrate for 20S proteasome) (Fig. 2C), but the time
course was different from that of in situ proteolysis.
The in vitro hydrolysis of sLLVY-MCA began to rise at
2 h, then reached a plateau (sustained for 3– 4 h),
and finally decreased toward the baseline (back at
15 h).
Despite different time courses, both in situ proteoly-
sis and in vitro hydrolysis of sLLVY-MCA resulted very
probably from proteasome activity, as assessed by inhi-
bition by the cell permeant inhibitors of proteasome,
LLnL (5 mM), and PSI (5 mM) (Fig. 2D).
It may be noted that intracellular accumulation of
altered or derivatized proteins is obvious as early as
5 h (Fig. 1) and persisted for the duration of the
536 Vol. 14 March 2000 The FASEB Journal
experiment (18 h) (i.e., after the decay of in situ
proteolysis) (Fig. 2). This suggests that accumulation
of derivatized and ubiquitinated proteins may result
from alteration of the proteasome activity.
Toxic concentrations of oxLDL induced both the
accumulation of altered cellular proteins and apo-
ptosis in ECV-304 cells. Because accumulation of
oxidized (or otherwise altered) proteins is poten-
tially toxic to the cell (24, 25) and proteasome is
involved in the degradation of altered proteins, this
led us to investigate whether proteasome inhibition
and apoptosis induced by oxLDL were causally re-
lated or parallel unrelated events. This was investi-
gated by using proteasome inhibitors and ubiquiti-
nation-deficient cells.
Proteasome inhibition potentiates the oxLDL-
induced toxicity
The concentrations of proteasome inhibitors used
here were not toxic per se (over the 24 h of the
experiments) but were effective in inhibiting protea-
some activity [i.e., inhibiting both the early peak of
oxLDL-induced autoproteolysis (data not shown)
and sLLVY-MCA hydrolysis (Fig. 2D)]. This concen-
tration of proteasome inhibitors potentiated both
the accumulation of ubiquitinated proteins and the
oxLDL-induced toxicity. Proteasome inhibitors po-
tentiated rapidly the accumulation of ubiquitinated
proteins (data not shown) probably because of the
inhibition of oxLDL-induced early peak of protea-
some activation.
The potentiation of the oxLDL-induced toxicity by
proteasome inhibitors was obvious at low (non- or
slightly toxic) oxLDL concentrations, because, in the
presence of inhibitors, the toxic effect of oxLDL to
ECV-304 cells occurred faster and at lower concen-
trations (Figs. 3A, B). For instance, when cells were
incubated with low oxLDL concentration (50 mg
apoB/ml) in the absence of proteasome inhibitor,
no significant toxicity was observed (Figs. 3B, C),
thus suggesting that the toxic threshold dose of
oxLDL (43) was not reached. In contrast, when the
same experiment was performed in the presence of
proteasome inhibitors, cells underwent apoptosis
(Fig. 3C). This suggests that proteasome inhibition
lowered the threshold of oxLDL concentration in-
ducing the cytotoxicity.
Proteasome inhibitors did not alter the type of cell
death induced by oxLDL (Fig. 3D–G), because ECV-
304 EC killed by low (not toxic per se) oxLDL
concentrations in the presence of proteasome inhib-
itors exhibited the characteristic features of apopto-
sis, quite similarly to cells killed by toxic concentra-
tions of oxLDL (15).
These data suggest that proteasome activity may
participate in the cellular defenses against oxLDL-
VIEIRA ET AL.
Figure 3. Proteasome inhibitors potentiate the effect of oxLDL on ECV-304 cells. A–G) Effect of proteasome inhibitors, LLnL
(5 mM), Lc (5 mM), and PSI (5 mM) on the whole toxicity (A, B) and apoptosis (C–G) induced by oxLDL. A) Time course of
toxicity using 200 mg apoB/ml of LDL or native LDL (empty circles); B) Dose dependence of the toxic effect (determined at
24 h) induced by increasing concentrations of oxLDL (filled symbols) in the absence (filled circles) or presence of proteasome
inhibitors LLnL, Lc, and PSI (filled squares and triangles, respectively). C) Apoptosis evaluated by morphological counting after
18 h incubation with low (nontoxic) concentration (50 mg apoB/ml) of oxLDL in the presence or absence of proteasome
inhibitors LLnL (5 mM), Lc (5 mM), and PSI (5 mM). Means 6 se of three separate experiments. D–G) Double fluorescent
staining of cell nuclei (by SYTO-13 and propidium iodide) discriminating normal cells and cells undergoing apoptosis (A),
primary necrosis and postapoptotic necrosis (AN). E) Untreated cells. E–G) Cells incubated for 18 h with 50 mg apoB/ml of
oxLDL in the absence (E) or presence (F) of 5 mM PSI, or with PSI alone (G). (*P,0.01).

induced toxicity and, conversely, that inhibition of
endogenous proteolysis lowered the toxic threshold
dose of oxLDL.
It may be noted that the oxLDL toxicity and its
potentiation by proteasome inhibitors is not restricted
to endothelial cells. In rabbit arterial smooth muscle
cells, after 24 h incubation with 200 mg apoB/ml of
oxLDL in the presence or absence of 10 mM PSI, the
viability was 33 6 4% and 78 6 8%, respectively. In the
U937 monocytic cell line, after 24 h incubation with
100 mg apoB/ml of oxLDL in the presence or absence
of 10 nM PSI (this cell line is very susceptible to PSI),
the viability was 30 6 5% and 60 6 7%, respectively (as
assessed by the trypan blue test).
Defective ubiquitination potentiates oxLDL-
induced apoptosis
As the degradation of altered proteins by protea-
some may be ubiquitin-dependent or -independent
(25), we investigated whether the ubiquitin pathway
is involved in the cellular defenses against oxLDL
toxicity by using genetically engineered cells (de-
rived from the E1-thermo-sensitive ts20 cell line). As
previously reported (36, 37), ubiquitination was ef-
fective at 34°C in the three cell lines, and at 39°C in
the E1-transduced H38 –5, but was defective at 39°C
in ts20pMV12 and ts20Bcl-2#7 cell lines (both in the
presence or absence of oxLDL; Fig. 4A).
The relative susceptibility of the three cell lines to
the toxic effect of oxLDL was compared at 34 and
39°C. The toxicity of oxLDL (200 mg apoB/ml) to
H38 –5 cells (in which ubiquitination is always work-
ing) was quite similar at 34 and 39°C. In contrast,
both ts20pMV12 and ts20Bcl-2#7 cell lines were
much more susceptible to the toxic effect of oxLDL
at 39°C (defective ubiquitination) than at 34°C (ef-
fective ubiquitination) (Fig. 4B). These data suggest
that the ubiquitin pathway plays a role in the cellular
defenses against oxLDL-induced toxicity.

UBIQUITIN-PROTEASOME PATHWAY IN oxLDL-INDUCED APOPTOSIS 537


Figure 4. Influence of ubiquitination on the cytotoxicity of
oxLDL in ts20 engineered cell lines ts20pMV12. H 38 –5 and
ts20Bcl-2#7 (ubiquitination is effective in the three cell lines
at 34 and 39°C in the H38 –5 cell line, whereas, it is defective
at 39°C in ts20pMV12 and ts20Bcl-2#7) (37). A) Western blot
showing the ubiquination levels of the three cell lines incu-
bated at 39°C for 5 h in the absence or presence of oxLDL
(200 mg apoB/ml). B) Evaluation of the cytotoxicity of oxLDL
(200 mg apoB/ml for 18 h) at 34 and 39°C.
It may be noted that the Bcl-2 overexpressing
ts20Bcl-2#7 was used in order to examine whether, in
this model system (ubiquitin-defective), the oxLDL-
538 Vol. 14 March 2000 The FASEB Journal
induced apoptosis was counterbalanced by Bcl-2
overexpression or not. At 39°C, the ubiquitination-
defective two cell lines ts20pMV12 (not expressing
Bcl-2) and ts20Bcl-2#7 (overexpressing Bcl-2) were
similarly susceptible to the toxic effect of oxLDL
[over the time of the experiment (i.e., 18 –24 h)],
thus suggesting that Bcl-2 is not effective in prevent-
ing oxLDL-induced cell death. In contrast, in agree-
ment with Monney et al. (37), overexpression of
Bcl-2 in ts20Bcl-2#7 increased the resistance of these
cells against heat shock-induced cell death (data not
shown).
DNPH prevents cell protein derivatization and
oxLDL-induced cytotoxicity
As it resulted implicitly from the above reported
data, that lipid peroxidation of oxLDL (e.g., 4-HNE)
generates cell protein adducts that are potentially
involved in the toxicity, we hypothesized that inhibi-
tion of cell protein derivatization should inhibit the
toxic effect of oxLDL. Dinitrophenylhydrazine, a
well-known reagent of lipid peroxidation-derived al-
dehydes, was used in order to scavenge oxLDL-
generated aldehydic compounds. As expected,
DNPH reduced both the level of 4-HNE-derivatized
proteins (Figs. 5A) and, at a lesser extent, the level of
the whole derivatization of cell proteins (evaluated
by [3H]NSP-reactive free amino groups) (Fig. 5B).
Concomitantly, DNPH (used at nontoxic concentra-
tion) was also able to inhibit the toxic effect induced
by oxLDL (Fig. 5C), thus supporting the hypothesis
that cell protein derivatization induced by oxLDL is
involved in their toxicity.
DISCUSSION
OxLDL toxicity may potentially be involved in the
genesis of the necrotic core and of complicated
atherosclerotic plaques prone to plaque rupture and
thrombosis. The mechanism and type of cell death
occurring in atherosclerotic areas may be of impor-
tance, because (in principle) apoptotic cells are
rapidly engulfed and cleared, whereas necrotic cell
debris may trigger a local inflammatory response. It
is therefore of interest to better understand the
cellular mechanisms regulating the susceptibility of
cells to oxLDL cytotoxicity.
To our knowledge, this is the first study linking the
toxicity induced by oxLDL with cell protein alter-
ations and proteasome inhibition. Conversely, it is
also suggested that the ubiquitin-proteasome path-
way plays a role in the cellular defenses against the
toxic effect of oxLDL.
The reported data show that lipid peroxidation
derivatives (among them 4-HNE) contained in ox-
VIEIRA ET AL.
Figure 5. DNPH prevents cell protein derivatization and cytotoxicity induced by oxLDL on ECV-304 cells. Cultured cells were
incubated with or without 200 mg apoB/ml oxLDL in the absence (vehicle only) or presence of 100 mM DNPH (dissolved in
DMSO). A, B) 4-HNE-adducts were evaluated at 16 and 24 h on western-blot probed with anti-4-HNE antibody (A), and the
whole derivatization was determined at 24 h by titration of free reactive-amino groups by [3H]NSP (B), under the conditions
of Fig. 1. C) Cytotoxicity was evaluated at 24 h by the MTT test. Means 6 se of three separate experiments.

Figure 6. Role of the ubiquitin-proteasome pathway on cell survival (black). Sites of action and potential interferences of toxic
concentrations of oxLDL (red), proteasome inhibitors, lactacystin and PSI (blue), and ubiquitination defect (temperature
sensitive cells) (green). The colored arrows indicate the observed variations and their relationship with oxLDL or/and
inhibitors.

UBIQUITIN-PROTEASOME PATHWAY IN oxLDL-INDUCED APOPTOSIS 539

LDL are able to derivatize cell proteins, in agree-
ment with the observation of Rosenfeld et al. (18) in
macrophagic cells. Until now, the possible interac-
tions between lipid peroxidation end products con-
tained in oxLDL and cell proteins have been only
poorly investigated in contrast to apoB modifica-
tions, which are well documented (see reviews in refs
4, 6). Our data suggest that 4-HNE, a major aldehy-
dic lipid peroxidation derivative able to form ad-
ducts with lysine, histidine, or cysteine of proteins
(16, 17), seems to play a major role in oxLDL-
induced derivatization of cell proteins. But, it is not
excluded that other reactive compounds of oxLDL
[e.g., malondialdehyde, fatty acid peroxides (9)],
may also derivatize cell proteins.
The dramatic accumulation of ubiquitinated pro-
teins may result from two additional mechanisms.
During the early time of incubation (up to 3 h with
toxic concentrations of oxLDL), the rise of ubiquiti-
nation cannot result from a defect in the de-ubiquiti-
nation process (by the 19S complex) because the
whole activity of the 26S proteasome is high and may
result from an activation of the ubiquitination path-
way. This may be because of the oxLDL-induced
protein derivatization, given that 4-HNE is able to
induce both protein derivatization and proteasome
activation (44). After 5 h, the inhibition of the 26S
proteasome may constitute an additional mechanism
of accumulation of ubiquitinated proteins (Fig. 6).
OxLDLs affect, in a biphasic manner, the protea-
some activity by inducing an early transient activa-
tion of proteasome followed by a sustained decay.
The early and transient peak (between 1 and 3 h
with 200 mg apoB/ml of oxLDL) of proteolysis
resulted from proteasome activation, as assessed by
the use of proteasome inhibitors (whereas the basal
proteolysis was probably proteasome-independent
because it was not affected by proteasome inhibi-
tors). This early proteasome activation may be sub-
sequent to oxLDL-induced structural alterations and
ubiquitination of cellular proteins, which are valu-
able substrates for the proteasome (23–25). Activa-
tion of endogenous proteolysis and of sLLVY-MCA
hydrolysis exhibited different time courses (the rise
of sLLVY-MCA hydrolysis beginning 2 h after that of
autoproteolysis). The activation of in vitro degrada-
tion of sLLVY-MCA synthetic substrate (cannot be
attributed to substrate generation) may result from
activation of the 20S proteasomal core by oxLDL.
OxLDL may act either 1) by generating a cellular
oxidative stress (47), which is known to activate the
20S proteasome (and subsequently, NFkB) (48, 49),
or 2) by activating protein kinases (22, 51, 52), some
of them being potentially able to phosphorylate and
activate (50) the PA28 proteasome activator (which
can in turn stimulate synthetic substrate hydrolysis by
the 20S proteasome).
540 Vol. 14 March 2000 The FASEB Journal
This transient proteasome activation was followed
by a decay of proteasomal proteolysis toward the
baseline. This decay resulted neither from a sub-
strate (modified proteins) depletion, because high
levels of derivatized proteins and ubiquitinated-
HMM still persisted up to 18 h, nor from an irrevers-
ible inhibition of the 20S proteasome activity, be-
cause proteasome was able to hydrolyze in vitro the
fluorogenic substrate sLLVY-MCA up to 10 –12 h.
Altogether, these data led us to hypothesize a two-
step mechanism of inhibition affecting first the 19S
and later the 20S. The rapid decay of endogenous
proteolysis may be a result of inhibition of the 19S
(thus inhibiting de-ubiquitination and degradation
of ubiquitinated proteins), because 4-HNE cross-
linked proteins are resistant to proteolysis and are
able to inhibit the 26S proteasome (45) and because
Reinheckel et al. (46) reported that the 26S protea-
some is less resistant to H2O2-induced oxidative
stress than the 20S proteolytic core. The second step
[i.e., inhibition of the 20S core (involved in sLLVY-
MCA hydrolysis)] may be a result of the progressing
intracellular oxidative stress induced by oxLDL (47;
Fig. 6). At this stage, when the proteasome is com-
pletely inhibited, cells are rapidly dying. This led us
to investigate the relationship between proteasome
inhibition and the oxLDL-induced cytotoxicity, be-
cause, according to the model systems, proteasome
may participate in the apoptotic process or prevent it
(26 –31).
The reported data strongly suggest that the active
proteasome plays a role in the cellular defenses
against the oxLDL-induced cytotoxicity, and, con-
versely, that proteasome inhibition may be involved
in the mechanism of cytotoxicity.
The protective effect of proteasome was (at least in
part) dependent on ubiquitination, because at 39°C
the ubiquitin-defective ts20pMV12 and ts20Bcl-2#7
cells were more susceptible to the toxic effect of
oxLDL than the ubiquitin-expressing H38 –5 cells. It
is noteworthy that the higher susceptibility of the
ubiquitin-defective cells seems to be independent of
heat-shock toxicity because in ts20pMV12, the time
courses of the oxLDL-induced and heat shock-in-
duced toxicity are different (beginning at 12–15 and
after 24 h, respectively); in ts20Bcl-2#7 cells, the heat
shock-induced apoptosis is blocked by Bcl-2 overex-
pression (36, 37) in contrast to oxLDL-induced
apoptosis. The inefficiency of Bcl-2 in preventing the
oxLDL-induced toxicity to ts20Bcl-2#7 cells is quite
consistent with our recently reported data, which
show that Bcl-2 overexpression alters only the bal-
ance between apoptosis and necrosis but does not
prevent cell death triggered by oxLDL (41). Alto-
gether, these data strongly suggest that ubiquitina-
tion plays a role in the cellular resistance against the
toxicity of oxLDL.
VIEIRA ET AL.
 Potentiation of the oxLDL-induced toxicity by
proteasome inhibitors may involve various mecha-
nisms. The ubiquitin-proteasome pathway is prob-
ably involved in cellular defenses against oxLDL by
degrading altered proteins generated by oxLDL
[e.g., proteins derivatized by 4-HNE (this paper
and ref 22) or other lipid peroxidation products].
Therefore, the inhibition of the ubiquitin-protea-
some pathway (by specific proteasome inhibitors,
by genetic alterations or by oxLDL) results in the
accumulation of structurally and functionally al-
tered cell proteins that impair cell functions and
viability (31, 53). Moreover, proteasome inhibition
may also impair cell cycle regulation and alter cell
viability through the defect of its other housekeep-
ing functions involved in the regulation of pro-
teins that play a role in the balance between cell
death and survival (29 –31). For instance, because
proteasome is involved in the degradation of p53
and of IkB (and NFkB activation), proteasome
inhibition may lead to accumulate the proapop-
totic p53 protein and block the activation of NFkB
[a survival factor (54)] (29 –31; Fig. 6).
In conclusion, the reported data may be summa-
rized by the following hypothetical scenario: Alde-
hydes (and possibly other compounds transported
by oxLDL) are able to derivatize cell proteins (this
paper and ref 22) and alter cell signaling (22, 51,
52). At low (nontoxic) concentrations of oxLDL, the
proteasome acts as a cellular defense system by
degrading these structurally (e.g., derivatized) and
functionally altered proteins. At high (toxic) concen-
trations of oxLDL, after a transient activation, the
proteasome activity is inhibited (the 26S complex at
5 h and the 20S proteolytic core at 14 h) by oxLDL,
thus leading to the accumulation of altered cell
proteins. The subsequent cell injury becomes irre-
versible and leads to cell death. This role of protea-
some in the oxLDL-induced toxicity seems to be a
relatively general mechanism, because proteasome
inhibition potentiates the effect of oxLDL in various
cell types present in atherosclerotic areas (e.g., en-
dothelial, monocytic, smooth muscle cells, and
fibroblasts).
The authors wish to thank Dr. H. L. Ozer for genetically
engineered cells (ts20 and H38 –5 cell lines); G. Ledinski,
C. Mora, J. Dumoulin, and M. F. Frisach for technical
assistance; and SNCF laboratory for providing human
serum. This work was supported by grants from INSERM,
University Paul Sabatier Toulouse-3, European Community
(Biomed-2 CA BMH4-CT98 –3191) to U-466; from PRAXIS
XXI Program (project 2/2.1/Qui/371/94) to L.A.; from
Austrian Research Council, special research center Bi-
omembranes project F00710 to G.J.; from the Swiss Na-
tional Science Foundation to C.B. O.V. was recipient of a
fellowship from PRAXIS XXI (BD/5493/95) and INSERM,
and I.E.-B. from VML.
UBIQUITIN-PROTEASOME PATHWAY IN oxLDL-INDUCED APOPTOSIS
REFERENCES
1. Stary, H. C. (1990) The sequence of cell and matrix changes in
atherosclerotic lesions of coronary arteries in the first forty years
of life. Eur. Heart J. 11, 3–19
2. Ross, R. (1993) The pathogenesis of atherosclerosis: a perspec-
tive for the 1990s. Nature (London) 362, 801– 809
3. Goldstein, J. L., and Brown, M. S. (1997) The low density
lipoprotein pathway and its relation to atherosclerosis. Annu.
Rev. Biochem. 46, 897–930
4. Witztum, J. L., and Steinberg, D. (1991) Role of oxidized LDL in
atherogenesis. J. Clin. Invest. 88, 1785–1792
5. Steinbrecher, U. P., Zhang, H., and Lougheed, M. (1990) Role
of oxidatively modified LDL in atherosclerosis. Free Rad. Biol.
Med. 9, 155–168
6. Esterbauer, H., Dieber-Rotheneder, M., Waeg, M., Striegl, G.,
and Jürgens, G. (1990) Biochemical, structural and functional
properties of oxidized LDL. Chem. Res. Toxicol. 3, 77–92
7. Chait A., and Heinecke J. W. (1994) Lipoprotein modification:
cellular mechanisms. Curr. Opin. Lipidol. 5, 365–370
8. Dee, G., Rice-Evans, C., Obeyesekera, S., Meraji, S., Jacobs, M.,
and Bruckdorfer, K. R. (1991) The modulation of ferrylmyoglo-
bin formation and its oxidative effects on LDL by nitric oxide.
FEBS Lett. 294, 38 – 42
9. Haberland, M. E., Fong, D., and Cheng, L. (1988) Malondial-
dehyde altered protein occurs in atheroma of Watanabe herita-
ble hyperlipidemic rabbits. Science 241, 215–218
10. Penn, M. S, and Chisolm, G. M. (1994) Oxidized lipoproteins,
altered cell function and atherosclerosis. Atherosclerosis 108
(Suppl.), S21–S29
11. Henricksen, T., Evensen, S. A., and Carler, B. (1979) Injury to
human endothelial cells in culture induced by LDL. Scand.
J. Clin. Lab. Invest. 39, 361–368
12. Hessler, H., Robertson, A. L., and Chisolm, G. M. (1979)
LDL-induced cytotoxicity and its inhibition by HDL in human
vascular smooth muscle cells and endothelial cells in culture.
Atherosclerosis 32, 213–229
13. Dimmeler, S., Haendeler, J., Galle, J., and Zeiher, A. M. (1997)
Oxidized low-density lipoprotein induces apoptosis of human
endothelial cells by activation of CPP32-like proteases. A mech-
anistic clue to the ‘response to injury’ hypothesis. Circulation 95,
1760 –1763
14. Escargueil-Blanc, I., Salvayre, R., and Nègre-Salvayre, A. (1994)
Necrosis and apoptosis induced by oxidized LDL occurs
through two calcium-dependent pathways in lymphoblastoid
cells. FASEB J. 8, 1075–1080
15. Escargueil-Blanc, I., Meilhac, O., Pieraggi, M. T., Arnal, J. F.,
Salvayre, R., and Nègre-Salvayre, A. (1997) Oxidized LDL
induce massive apoptosis of cultured human endothelial cells
through a calcium-dependent pathway. Arterioscler. Thromb. Vasc.
Biol. 17, 331–339
16. Esterbauer, H., Schaur, R. J., and Zollner, H. (1991) Chemistry
and biochemistry of 4-hydroxynonenal, malonaldehyde and
related aldehydes. Free Radic. Biol. Med. 11, 81–128
17. Uchida, K., Itakura, K., Kawakishi, S., Hiai, H., Toyokuni, S., and
Stadtman, E. R. (1995) Characterization of epitopes recognized
by 4-hydroxy-2-nonenal specific antibodies. Arch. Biochem. Bio-
phys. 324, 241–248
18. Rosenfeld, M. E., Khoo, J. C., Miller, E., Parthasarathy, S.,
Palinski, W., and Witztum, J. L. (1991) Macrophage-derived
foam cells freshly isolated from rabbit atherosclerotic lesions
degrade modified lipoproteins, promote oxidation of LDL, and
contain oxidation-specific lipid-protein adducts. J. Clin. Invest.
87, 90 –99
19. Szweda, L. I., Uchida, K., Tsai, L., and Stadtman, E. R. (1993)
Inactivation of glucose-6-phosphate dehydrogenase by 4-hy-
droxy-2-nonenal: selective modification of an active-site lysine.
J. Biol. Chem. 268, 3342–3347
20. Rossi, M. A., Fidale, F., Garramone, A., Esterbauer, H., and
Dianzani, M. U. (1990) Effect of 4-hydroxylalkenals on hepatic
phosphatidylinositol-4,5-bisphosphate-phospholipase C. Biochem.
Pharmacol. 39, 1715–1719
21. Natarajan, V., Scribner, W. M., and Taher, M. M. (1993)
4-Hydroxynonenal, a metabolite of lipid peroxidation, activates
phospholipase D in vascular endothelial cells. Free Radic. Biol.
Med. 15, 365–375
541
22. Suc, I., Meilhac, O., Lajoie-Mazenc, I., Vandaele, J., Jürgens, G.,
Salvayre, R., and Negre-Salvayre, A. (1998) Activation of EGF
receptor by oxidized LDL. FASEB J. 12, 665– 671
23. Dean, R. T., Fu, S., Stocker, R., and Davies, M. J. (1997)
Biochemistry and pathology of radical-mediated protein oxida-
tion. Biochem. J. 324, 1–18
24. Berlett, B. S., and Stadtman, E. R. (1997) Protein oxidation in
aging, disease and oxidative stress. J. Biol. Chem., 272, 20313–
20316
25. Grune, T., Reinheckel, T., and Davies, K. J. A. (1997) Degrada-
tion of oxidized proteins in mammalian cells. FASEB J. 11,
526 –534
26. Hilt, W., and Wolf, D. H. (1996) Proteasomes: destruction as a
programme. Trends Biochem. Sci. 21, 96 –102
27. Tanaka, K. (1998) Molecular biology of the proteasome. Bio-
chem. Biophys. Res. Comm. 247, 537–541
28. Pickart, C. (1997) Targeting of substrates to the 26S protea-
some. FASEB J. 11, 1055–1066
29. Hochstrasser, M. (1996) Ubiquitin-dependent protein degrada-
tion. Annu. Rev. Genet. 30, 405– 439
30. Haas, A. L., and Siepmann, T. J. (1997) Pathways of ubiquitin
conjugation. FASEB J. 11, 1257–1268
31. Hershko, A., and Ciechanover, A. (1998) The ubiquitin system.
Annu. Rev. Biochem. 67, 425– 479
32. Vieira, O., Escargueil-Blanc, I., Meilhac, O., Basile, J. P., Laran-
jinha, J., Almeida, L., Salvayre, R., and Negre-Salvayre, A. (1998)
Effect of dietary phenolic compounds on apoptosis of human
cultured endothelial cells induced by oxidized LDL. Br. J.
Pharmacol. 123, 565–573
33. Wolff, S. P. (1994) Ferrous ion oxidation in the presence of
ferric ion indicator xylenol orange for measurement of hy-
droperoxides. Methods Enzymol. 233, 182–189
34. Yagi, K. (1987) Lipids peroxides and human diseases. Chem.
Phys. Lipids 45, 337–351
35. Esterbauer, H., Zollner, H., and Lang, J. (1985) Metabolism of
the lipid peroxidation product 4-hydroxynonenal by isolated
hepatocytes and by liver cytosolic fractions. Biochem J. 228,
363–373
36. Chowdary, D. R., Dermody, J. J., Jha, K. K., and Ozer, H. L.
(1994) Accumulation of p53 in a mutant cell line defective in
the ubiquitin pathway. Mol. Cell Biol. 14, 1997–2003
37. Monney, L., Otter, I., Olivier, R., Ozer, H. L., Haas, A. L.,
Omura, S., and Borner, C. (1998) Defects in the ubiquitin
pathway induce caspase-independent apoptosis blocked by
Bcl-2. J. Biol. Chem. 273, 6121– 6131
38. Grune, T., Reinheckel, T., and Davies, K. J. A. (1996) Degrada-
tion of oxidized proteins in K562 human hematopoietic cells by
proteasome. J. Biol. Chem. 271, 15504 –15509
39. Staros, J. V., Wright, R. W., and Swingle, D. M. (1986) Enhance-
ment by N-hydroxysulfo-succinimide of water-soluble carbodi-
imide-mediated coupling reactions. Anal. Biochem. 156, 220 –222
40. Price, P., and McMillan, T. J. (1989) Use of the tetrazolium assay
in measuring the response of human tumor cells to ionizing
radiations. Cancer Res. 50, 1392–1396
41. Meilhac, O., Escargueil-Blanc, I., Thiers, J. C., Salvayre, R., and
Nègre-Salvayre, A. (1999) Bcl-2 alters the balance between
542 Vol. 14 March 2000 The FASEB Journal
apoptosis and necrosis, but does not prevent cell death induced
by oxidized low density lipoproteins. FASEB J. 13, 485– 494
42. McConkey, D. J., Hatrzell, P., Nicotera, P., and Orrenius, S.
(1989) Calcium-activated DNA fragmentation kills immature
thymocytes. FASEB J. 3, 1843–1849
43. Nègre-Salvayre, A., Lopez, M., Levade, T., Pierraggi, M. T.,
Dousset, N., Douste-Blazy, L., and Salvayre, R. (1990) UV-
treated LDL as a model system for the study of the biological
effects of lipid peroxides on cultured cells. 2. Uptake and
cytotoxicity of UV-peroxidized LDL on lymphoid cell lines.
Biochim. Biophys. Acta 1045, 224 –232
44. Grune, T., Reinheckel, T., Joshi, M., and Davies, K. J. (1995)
Proteolysis in cultured liver epithelial cells during oxidative
stress: role of the multicatalytic proteinase complex, protea-
some. J. Biol. Chem. 270, 2344 –2351
45. Friguet, B., Stadtman, E. R., and Szweda, L. I. (1994) Modifica-
tion of glucose-6-phosphate dehydrogenase by 4-hydroxy-2-non-
enal: formation of cross-linked protein that inhibits the multi-
catalytic protease. J. Biol. Chem. 269, 21639 –21643
46. Reinheckel, T., Sitte, N., Ullrich, O., Kuckelkorn, U., Davies,
K. J., and Grune, T. (1998) Comparative resistance of the 20S
and 26S proteasome to oxidative stress. Biochem J. 335, 637– 642
47. Schmitt, A., Nègre-Salvayre, A., Delchambre, J., and Salvayre, R.
(1995) a-tocopherol and rutin prevent the glutathione and ATP
depletion induced by oxidized LDL in cultured endothelial
cells. Brit. J. Pharmacol. 16, 1985–1990
48. Strack, P. R., Waxman, L., and Fagan, J. M. (1996) Activation of
the multicatalytic endopeptidase by oxidants: effects on enzyme
structure. Biochemistry 35, 7142–7149
49. Brand, K., Eisele, T., Kreusel, U., Page, M., Page, S., Haas, M.,
Gerling, A., Kaltschmidt, C., Neumann, F. J., Mackman, N.,
Baeurele, P. A., Walli, A. K., and Neumeier, D. (1997) Dysregu-
lation of monocytic nuclear factor-kappa B by oxidized low-
density lipoprotein. Arterioscler. Thromb. Vasc. Biol. 17, 1901–1909
50. Li, N., Lerea, K. M., and Etlinger, J. D. (1996) Phosphorylation
of the proteasome activator PA28 is required for proteasome
activation. Biochem. Biophys. Res. Commun. 225, 855– 860
51. Deigner. H. P., and Claus, R. (1996) Stimulation of mitogen
activated protein kinase by LDL and oxLDL in human U-937
macrophage-like cells. FEBS Lett. 385, 149 –153
52. Auge, N., Escargueil-Blanc, I., Lajoie-Mazenc, I., Suc, I., An-
drieu-Abadie, N., Pieraggi, M. T., Chatelut, M., Thiers, J. C.,
Jaffrezou, J. P., Laurent, G., Levade, T., Negre-Salvayre, A., and
Salvayre, R. (1998) Potential role for ceramide in mitogen-
activated protein kinase activation and proliferation of vascular
smooth muscle cells induced by oxidized low density lipopro-
tein. J. Biol. Chem. 273, 12893–12900
53. Fenteany, G., and Schreiber, S. L. (1998) Lactacystin, protea-
some function, and cell fate. J. Biol. Chem. 273, 8545– 8548
54. Van Antwerp, D. J., Martin, S. J., Kafri, T., Green, D. R., and
Verma, I. M. (1996) Suppression of TNF-alpha-induced apopto-
sis by NF-kappaB. Science 274, 787–789
Received for publication January 15, 1999.
Revised for publication September 24, 1999.
VIEIRA ET AL.