Differential effects of oxidized LDL on apolipoprotein AI and B synthesis
in HepG2 cells
Emmanuel Bourdona,1, Nadine Loreaua, Laurent Lagrosta, Jean Davignonb,
Lise Bernierb, Denis Blachea,⁎
aINSERM U498, Dijon, France;–Faculté de Médecine, Université de Bourgogne, 21079 Dijon, France
bHyperlipidemia and Atherosclerosis Research Group, Clinical Research Institute of Montreal, Montreal, Quebec, Canada
Received 20 December 2005; revised 5 May 2006; accepted 23 May 2006
Available online 3 June 2006
Oxidized low-density lipoproteins (Ox-LDL) are key elements in atherogenesis. Apolipoprotein AI (apoAI) is an active component of the
antiatherogenic high-density lipoproteins (HDL). In contrast, plasma apolipoprotein B (apoB), the main component of LDL, is highly correlated
with coronary risk. Our results, obtained in HepG2 cells, show that Ox-LDL, unlike native LDL, leads to opposite effects on apoB and apoAI,
namely a decrease in apoAI and an increase in apoB secretion as evaluated by [3H]leucine incorporation and specific immunoprecipitation.
Parallel pulse–chase studies show that Ox-LDL impaired apoB degradation, whereas apoAI degradation was increased and mRNA levels were
decreased. We also found that enhanced lipid biosynthesis of both triglycerides and cholesterol esters was involved in the Ox-LDL-induced
increase in apoB secretion. Our data suggest that the increase in apoB and decrease in apoAI secretion may in part contribute to the known
atherogenicity of Ox-LDL through an elevated LDL/HDL ratio, a strong predictor of coronary risk in patients.
© 2006 Elsevier Inc. All rights reserved.
Keywords: LDL oxidation; Lipid biosynthesis; Atherosclerosis; ApoAI; ApoB; LDL/HDL ratio; HepG2 cells; Free radicals
Population studies carried out in atherosclerotic patients
have largely established that a positive continuous linear
relationship exists between both serum low-density lipoprotein
(LDL)-cholesterol and apolipoprotein B100 (apoB) and the
extent of cardiovascular disease . LDL particles interact
with cells through LDL receptors and apoB works as a ligand
in this process . ApoB synthesis occurs in the liver and it is
involved in triglyceride-rich lipoprotein assembly . It is
well known that apoB overproduction occurs in diabetes,
insulin resistance, and coronary artery disease. High levels of
apoB-containing lipoproteins may be the result of an increased
production and/or a diminished catabolism. The resulting
increased residence time of LDL may increase its oxidiz-
ability. Accumulating data from biochemical, animal, and
epidemiological studies strongly support the hypothesis that
oxidative modification of LDL plays a crucial and causative
role in the pathogenesis of atherosclerosis [4–6].
Conversely, serum high-density lipoprotein (HDL)-choles-
terol and apoAI levels are negatively correlated with coronary
heart disease [7,8]. It is now clearly demonstrated that one of the
main protective effects of high HDL concentrations is related to
its involvement in reverse cholesterol transport, a system by
Free Radical Biology & Medicine 41 (2006) 786–796
Abbreviations: ACAT, acyl-CoA:cholesterol acyltransferase; BCA, bicin-
choninic acid; CE, cholesteryl ester; CVD, cardiovascular disease; DMEM,
Dulbecco’s minimum essential medium; EDTA, ethylenediamine tetraacetate;
N-LDL, Ac-LDL, and Ox-LDL, native, acetylated, and oxidized low-density
lipoproteins, respectively; EPA, eicosapentaenoic acid; FCS, fetal calf serum;
GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GC, gas chromatogra-
phy; LDH, D-lactic dehydrogenase; LPC, lysophosphatidylcholine; LPDS,
lipoprotein-depleted serum; Mops, 3-(N-morpholino)propanesulfonic acid;
PBS, phosphate-buffered saline; PIM, protease inhibitor mix; SDS–PAGE,
sodium dodecyl sulfate–polyacrylamide gel electrophoresis; SSPE buffer, 150
mM sodium chloride, 10 mM sodium hydrogen phosphate, 1 mM EDTA, pH
7.4; TBARS, thiobarbituric acid-reactive substances; TG, triglycerides; TLC,
⁎Corresponding author. Fax: +33 380 39 3300.
E-mail address: firstname.lastname@example.org (D. Blache).
1Present address: LBGM, Université de la Réunion, BP 7151, 97715 Saint
Denis, Ile de La Réunion, France.
0891-5849/$ - see front matter © 2006 Elsevier Inc. All rights reserved.
which excess cholesterol is transported from peripheral tissues
by HDL to the liver for excretion [8,9]. ApoAI, as a major
component of HDL, is central to this process. Although the
intestine can also produce significant amounts of this
apolipoprotein, the liver is the main production site for apoAI.
Abundant data indicate that overexpression of apoAI results in
elevated HDL levels in mice and rabbits and inhibition of
atherosclerotic lesions. In addition to its major role in reverse
cholesterol transport, HDL has been reported to inhibit cell-
mediated LDL modifications  and to reduce cellular uptake
and degradation of native and oxidized LDL . HDL is also
capable of protecting against LDL peroxidation in vitro and in
An elevated LDL/HDL ratio is recognized as being a strong
predictor of coronary risk in patients. Low plasma levels of
HDL may be the result of abnormal apoAI biosynthesis and/or
increased degradation. In this context, it has been demonstrated
that in several pathological conditions oxidant stress alters LDL
and HDL levels through apoB and apoAI plasma metabolism,
respectively. Such a situation characterized by both increased
LDL oxidizability and occurrence of plasma oxidized LDL (Ox-
LDL) has been particularly encountered in hypothyroidism,
metabolic syndrome, and diabetes, perturbations largely known
and in diabetes, data from apoAI kinetics studies, conducted in
animal models and in humans, have established whether the
conditions occur due to a reduced production of apoAI, an
increased degradation, or both .
Consequently, the purpose of the present study was to
examine whether Ox-LDL would influence the synthesis and
secretion of apoAI and apoB using HepG2 cells. In previous
studies, we established that the liver synthesis of albumin was
strongly decreased by Ox-LDL . We found that Ox-LDL
drastically reduced the synthesis and secretion of apoAI,
whereas opposite effects were found for apoB. By means of
pulse–chase studies and Northern blot analyses and lipid
biosynthesis, we provide evidence indicating that Ox-LDL-
mediated differential effects on apoAI and apoB synthesis are
explained by different mechanisms. ApoB synthesis is mainly
driven by cholesterol supply, whereas reduced apoAI synthesis
may be linked to oxidation.
HepG2 cells were grown in a CO2incubator (5% CO2, 95%
air) in 75-cm2flasks (Polylabo, France). Cultures were
maintained in 20 ml Dulbecco’s minimum essential medium
(DMEM; Gibco) containing 10% fetal calf serum (FCS), 1.25%
L-glutamine, and 2% penicillin/streptomycin. Some 5 or 6 days
before each experiment, approximately 1,000,000 cells were
seeded in a six-well tissue cluster (Costar) in 3 ml DMEM
containing 10% FCS. Incubations were performed in 10%
lipoprotein-deficient FCS (LPDS) and with the indicated
amounts of LDL preparations or various agents. Cell viability
was assayed using lactate dehydrogenase (LDH) released into
the medium (Sigma Procedure 500) and expressed as the
percentage of total LDH activity in lysed cells.
LDL preparation and modification
LDLs (1019–1055 g/ml) were isolated by sequential
ultracentrifugation using KBr (Beckman centrifuge) of pooled
plasma from normolipidemic subjects . After dialysis
against phosphate-buffered saline (PBS), pH 7.4, LDLs were
assayed for protein content by the BCA method . Ox-LDLs
were obtained by incubating 100 μg/ml protein LDL with
25 μM CuSO4at 37°C overnight, which resulted in a medium
level of oxidation measured as conjugated dienes (absorbance at
234 nm in the ranges 0.5–0.7 and 0.3–0.5 for Ox-LDL and N-
LDL, respectively). Ac-LDLs were prepared from LDLs by the
addition of acetic anhydride in the presence of cold saturated
sodium acetate . In order to prevent LDL oxidation,
acetylation was carried out in argon-enriched capped vials.
After extensive dialysis against PBS, pH 7.4, LDLs were
filtered and sterilized by passing through 0.22-μm Millipore
membranes. LDL oxidation and integrity were evaluated by
determining (1) the content of thiobarbituric acid reactive
substances (TBARS), (2) lipid peroxides using an iodine
reagent , and (3) the chromophore fluorescence at 430 nm
after excitation at 355 nm . Agarose gel electrophoresis
(0.5%) was performed with a Beckman’s Paragon lipoprotein
electrophoresis kit .
Sterols and oxysterols were obtained from Sigma or
Steraloids except for 7α- and 7β-cholesterol, which were
synthesized as in our previous paper . When necessary
(purity <95%), the oxysterols were repurified by thin-layer
chromatography (TLC) on silica gel-impregnated plates
(Merck) with hexane/ethyl acetate (70/30, v/v) and purity was
checked by gas chromatography . Total lipid extracts were
carried out according to Folch et al.  and analyzed according
to our previous gas chromatographic technique . Phos-
pholipids and lysophosphatidylcholine (LPC) were analyzed
by HPLC with a light-scattering detector according to Blache
et al. .
Studies on secretion of apoAI and apoB by HepG2 cells
Dose–response studies of modified LDLs on apoAI and
apoB secretions were performed by incubating HepG2 cells
with various amounts of LDLs (0 to 200 μg/ml) at 37°C for
14 h. The effects of incubation time on apoAI or apoB
secretions were examined in the presence of 150 μg/ml LDLs
for varying times (4 to 96 h). At the end of the incubations,
culture media were removed from flasks and cell monolayers
washed with PBS and collected for DNA measurement .
Because of exogenous addition of apoB-containing LDL and
obvious interference, apoB secretion was evaluated by
incubation in leucine-poor DMEM containing 10% LPDS and
787E. Bourdon et al. / Free Radical Biology & Medicine 41 (2006) 786–796
5 μCi/ml [3H]leucine (Amersham) followed by immunoprecip-
itation with anti-h-apoB antiserum provided by Boehringer
(Cat. Nos. 726478 and 726494). For consistency and easy
comparison, apoAI production was assessed with the same
protocol using immunoprecipitation with anti-h-apoAI antise-
rum. In brief, medium was first precleared by 1-h incubation
with 100 μl of washed protein A–Sepharose (10 mg/ml;
Pharmacia). After centrifugation, 100 μl of the medium was
added to 200 μl PBS and 100 μl anti-h-apoAI or apoB
antiserum (1:80 and 1:20 dilution, respectively). After a 1-
h incubation on ice, 100 μl of protein A–Sepharose (10 mg/ml)
was added to Net buffer (30 mmol/L Tris–HCl, 150 mmol/L
NaCl, 0.5% (v/v) Nonidet P-40, and 2 mmol/L ethylenediamine
tetraacetate (EDTA), pH 8). The precipitated apoprotein
immunocomplexes were isolated by centrifugation. After four
washes in Net buffer, pellets were dissolved in glycine buffer
(0.1 M, pH 3) and an aliquot was used for radioactivity
measurement. The incorporation of [3H]leucine into apoAI and
apoB was expressed as disintegrations per minute per
microgram cellular DNA. The specificity of this immunopre-
cipitation technique was examined after extensive washings of
the cells and by analyzing apoprotein immunocomplexes by
12% SDS–PAGE for apoAI and with enzyme-linked immuno-
sorbent assay (ELISA) for apoB100. Total secreted proteins
were assayed by radioactivity counting after trichloroacetic acid
precipitation on Whatman filters as previously described .
Quantification of apoAI in the medium was done by a
noncompetitive polyclonal ELISA developed in our laboratory
. Briefly, an affinity-purified goat polyclonal anti-human
apoAI antibody (Biodesign, Saco, MA, USA) with no cross-
reaction with other apolipoproteins was used as the capture
antibody. Detection was done with a goat anti-human apoAI
conjugated to horseradish peroxidase. Quantification was done
on controls calibrated on values obtained from human plasma
In the pulse–chase protocol, and for each sample, two six-
well plates (plate A for pulse-labeling and plate B for the chase)
were seeded with HepG2 cells according to Kempen et al. .
The cells were grown until they reached confluence and each
plate was washed with PBS. DMEM with 10% LPDS
containing 200 μg/ml N-LDL, Ac-LDL, or Ox-LDL was
added and incubation was continued for 14 h. The medium was
then removed and cells were incubated for 15 min with leucine-
depleted DMEM, 15 min being the time required for the
synthesis of the long apoB molecule . Thereafter, leucine-
poor medium containing 5 μCi/ml [3H]leucine was added for
pulse-labeling for a 15-min incubation period. After pulse-
labeling of plate A, the medium was removed, and the cells
were washed with cold PBS. After being pelleted, the cells were
mixed with 1 ml lysis buffer containing 100 μl 10× concentrated
protease-inhibitor mix (PIM; containing 186 mg EDTA, 35 mg
phenylmethylsulfonyl fluoride, 1 mg antipain, and 1 mg
pepstatin A per 50 ml) and 250 μl fivefold concentrated PBS
containing 47.25 g NaCl, 7 g Na2HPO4, 1.08 g KH2PO4, 5 ml
Triton X-100, and 2.5 g deoxycholate. After centrifugation for
5 min at 4000g, the supernatants were kept frozen at −20°C for
later analysis. After pulse-labeling of plate B, the hot medium
was removed and the cells were washed with DMEM and then
incubated for 2 h with DMEM 10% LPDS containing 200 μg/
ml N-LDL, Ac-LDL, or Ox-LDL (chase incubation). The cells
were then mixed and lysed with PBS with PIM as described
above. The medium was collected and centrifuged (5 min at
4000g), and 900 μl of the medium was mixed with 100 μl 10×
PIM. After pretreatment with protein A–Sepharose, the media
and cell extracts were subjected to immunoprecipitation as
described above. Radioactivity of washed cells obtained after
pulse was referred as “pulse cells” and radioactivity of
supernatants and cells obtained after chase was referred as
“chase totalQ. For better understanding, chase was also
expressed as a percentage of pulse (ratio of chase total to
pulse cells as a percentage).
Quantification of mRNA of apoAI and apoB in HepG2 cells
Total cellular RNA was isolated with Trizol reagent
(Pharmacia). After analysis of RNA integrity by agarose
electrophoresis, Northern blots were performed as follows:
20 μg total RNA in 7 μl diethylpyrocarbonate water was mixed
with 26 μl sample mix containing 3 μl 10× 3-(N-morpholino)
propanesulfonic acid (Mops), 6 μl formaldehyde, and 17 μl
formamide. After 15 min at 55°C, samples were loaded with
10 μl formaldehyde loading buffer. Electrophoresis was
performed in a 1% agarose gel containing 10× Mops and
formaldehyde (15 and 27 ml, respectively, in 108 ml deionized
water). Transfer to nylon membrane (0.45 μm; Amersham) was
performed by capillarity overnight. After washing and UV-
fixation (Stratagene), membranes were dried and stored at room
temperature before analysis. An 872-bp fragment (nt 1579–
2455) from human apoAI [28,29] and a 490-bp fragment (nt
3724–4224) from human GAPDH  were amplified from
HepG2 cells by RT-PCR and ligated into pGem-T vectors
(Promega). A 959-bp cDNA insert in the EcoRI site of pBR322
 was used as probe for apoB.
Probes were created by labeling DNA fragments with
[α-32P]dCTP using a random priming kit (Boehringer). After
incubation for 3 h in prehybridization buffer (50% formamide,
5× Denhardt’s, 5× SSPE, 0.1% SDS, and 100 μg/ml salmon
sperm), membranes were incubated overnight with about
1 × 106cpm/ml boiled probe. Finally, the membranes were
washed two times in 6× SSPE/0.5% SDS at room temperature
for 15 min,twotimes in 1× SSPE/0.5% SDS at 37°C for 15 min,
and once in 0.1× SSPE/0.1% SDS at 65°C. After film exposure
(Kodak X-OMat) and development, radioactive signals were
quantified by using Bio1D software (Vilbert–Lourmat).
GAPDH signals were used to normalize the RNA applied in
each lane. Data are presented as percentage change from control
incubations, namely without LDLs.
Northern blot data were confirmed by real-time PCR assays.
Six hundred nanograms of RNAwas transcribed using M-MLV
reverse transcriptase from Invitrogen (Burlington, ON, Canada)
and the resulting cDNA was amplified by real-time PCR on an
788 E. Bourdon et al. / Free Radical Biology & Medicine 41 (2006) 786–796
MX3500 (Stratagene) using the SYBR green master mix
purchased from Qiagen (Mississauga, ON, Canada). All primers
were designed using Primer3 software and were chosen to span
an intron–exon border in order to avoid contamination with
genomic DNA. Primers were purchased from Invitrogen.
Analysis of the PCR data was done with the Mx3500 software.
ApoAI and apoB mRNA levels were normalized with
ribosomal protein S14.
Biosynthesis of triglycerides and cholesteryl ester in HepG2
Confluent HepG2 cells in six-well plates were incubated
with 150 μg/ml N-LDL, Ac-LDL, or Ox-LDL in DMEM 10%
LPDS containing U-14C-labeled sodium acetate (3 μCi/ml;
Amersham; sp act 56 mCi/mmol). After 14 or 20 h of
incubation, the medium was removed, cells were washed with
PBS, and lipids were extracted from HepG2 monolayers in the
plastic well with hexane/isopropanol (3/2, v/v) for 30 min at
room temperature . Organic phases were evaporated under
nitrogen, and after addition of 20 μg nonradiolabeled standards
(Sigma), lipid extracts were applied to silica TLC plates
(Supelco) and separated with hexane/diethyl ether/formic acid
(80/20/2, v/v/v). Areas containing triglycerides (TG) and
cholesteryl esters (CE) were visualized with I2 vapor and
transferred into vials for radioactivity counting. Cell proteins
were dissolved in 0.1 N NaOH and assayed by the BCA method
(Pierce) after the cells had been extracted for lipids. In some
instances, eicosapentaenoic acid (EPA) and ACAT inhibitor
compound Sandoz 58-035 (through the courtesy of Dr. P.
Pfister, Novartis) were used to inhibit TG and CE syntheses,
Data are expressed as means ± SD of at least three
experiments performed in triplicate. The main effects of Ox-
LDL on apoAI and apoB secretions were evaluated with Prism
(GraphPad Software, Inc., San Diego CA, USA) using one-way
analysis of variance (ANOVA) followed by the Tukey or
Newman–Keuls posttest for multiple comparisons. Correlations
were calculated using Pearson’s rank test.
Effects of oxidized LDL on apoAI and apoB secretion rates
We confirmed our previous data  indicating that liver-
specific proteins were best secreted into the medium when cells
reached confluence . We incubated HepG2 cells with
150 μg/ml N-LDL, Ac-LDL, or Ox-LDL for a time-course
study of their effects on apoAI and apoB secretion into the
medium. Fig. 1A illustrates that, compared to N-LDL, Ox-LDL
induced a decrease in apoAI secreted into the medium after 24 h
of incubation (from 120 to 89 dpm/μg DNA, −25%, p < 0.01 vs
N-LDL). This Ox-LDL-induced decrease in apoAI secretion
remained significantly constant (−25 to −18%, p < 0.01)
whatever the incubation time. Ac-LDL at 150 μg/ml has no
effect on apoAI secretion compared with N-LDL. It is
noteworthy that, as indicated by the linearity of the apoAI
secretion throughout the study period, Ox-LDL resulted in a
significantly lower production rate (p < 0.0001) in comparison
with N-LDL and Ac-LDL. The effects of various concentrations
of LDL on apoAI secretion are shown in Fig. 1B. Compared
with corresponding N-LDL concentrations, Ox-LDL resulted in
significant decreases in apoAI secretion of 95.4 to 60.8, 103.5 to
60.5, and 100.4 to 50.2 dpm/μg DNA, which represented 36,
42, and 50% decreases with concentrations of 100, 150, or
200 μg/ml, respectively.
The results of time-course studies, illustrated in Fig. 2A,
indicate that incubation with 150 μg/ml Ox-LDL led to a
significant increase (p < 0.01) in apoB secreted into the medium
compared with N-LDL (137.7 to 213.5 (+55%), 611.8 to 971
Fig. 1. Effects of various LDL preparations on apoAI secretion analyzed by
[3H]leucine incorporation.(A) Kineticsof apoAI secretionby HepG2cellsin the
presence of various LDL preparations. Confluent cells were incubated for
varying times (25 to 96 h) with 150 μg/ml N-LDL, Ac-LDL, or Ox-LDL in
leucine-poor 10% LPDS DMEM containing [3H]leucine. ApoAI secretion was
evaluated by specific immunoprecipitation and radioactivity measurement as
described under Methods. Statistical significance (n = 4) was assessed using
one-way ANOVA, followed by Tukey’s test:++p < 0.01,+++p < 0.001 vs
N-LDL. (B) Cells were incubated for 14 h without (Control) or with 100, 150, or
200 μg/ml preparations of N-LDL, Ac-LDL, and Ox-LDL in leucine-poor 10%
LPDS DMEM containing [3H]leucine. Statistical significance (n = 4) was
assessed using one-way ANOVA, followed by Tukey’s test: *p < 0.05 vs
Control;+p < 0.05 vs native LDL.
789E. Bourdon et al. / Free Radical Biology & Medicine 41 (2006) 786–796
(+59%), and 984 to 1774 (+80%), after 4, 19, and 28 h of
incubation, respectively). Results of 14-h incubations with
various LDL preparations on apoB secretion (Fig. 2B) showed
that a stimulating effect of Ox-LDL was obtained with 100, 150,
and 200 μg/ml (+29, +59, and +64%, respectively). Under the
same incubation conditions, stimulation induced in the presence
of Ac-LDL did not reach the levels obtained with Ox-LDL but
rather resulted in a slight decrease.
No significant changes in total protein synthesis, as
measured by trichloroacetic acid precipitation of HepG2 cells
incubated with the various LDL preparations after [3H]leucine
incorporation, were observed (Fig. 3). These data confirm those
obtained with LDH release indicating no change between N-
LDL and Ox-LDL. This demonstrates clearly that the Ox-LDL-
mediated effects on apolipoprotein secretions were specific and
not the result of some general toxic effect affecting protein
synthesis and cell viability.
Oxidation parameters of lipoprotein preparations
To further understand the difference between the various
LDL preparations used, oxidation parameters were analyzed
(Table 1). Our data confirm that both Ac-LDLs and copper-
oxidized LDLs have enhanced electrophoretic mobility, relative
to N-LDLs. Ox-LDLs also have significantly higher conjugated
dienes, lipoperoxides, and TBARS than Ac-LDLs and N-LDLs.
The level of the chromophore (λex355 nm, λem430 nm) which
resulted from the reaction of aldehydic lipid peroxidation
products with free amino groups of apolipoprotein B  was
strongly enhanced in Ox-LDL. The lipid composition, summa-
rized in Table 2, indicated that no significant change in Ac-LDL
occurred compared to N-LDL, whereas the unsaturation index
of Ox-LDL phospholipids markedly diminished. The total
phospholipids expressed as PC were decreased, whereas a
significant increase in LPC was observed in Ox-LDL (more
than 15 times). These changes were also associated with a
significant decrease in total cholesterol and a marked appear-
ance of various oxysterols. In particular, as assessed by GC, we
found that 7β-OH, 25-OH, 7-Keto, α-epoxy-cholesterol, and
cholestanetriol appeared in Ox-LDL.
Effects of Ox-LDL on synthesis and turnover of apoAI and
To assess whether Ox-LDL directly affects the rate of protein
synthesis and/or intracellular protein turnover, pulse–chase
experiments were conducted. The cells were first incubated for
14 h with or without modified LDLs, washed, and pulse-labeled
for 15 min with [3H]leucine (pulse); the label was then removed
and the cells were either directly processed for synthesized
Fig. 2. Effects of various LDL preparations on apoB secretion analyzed by
[3H]leucine incorporation. (A) Kinetics of apoB secretion by HepG2 cells in the
presence of various LDL preparations. Confluent cells were incubated for
varying times (4 to 28 h) with 150 μg/ml preparations of N-LDL, Ac-LDL, or
Ox-LDL in leucine-poor 10% LPDS DMEM containing [3H]leucine. ApoB
secretion was evaluated by immunoprecipitation. Statistical significance (n = 4)
was assessed using one-way ANOVA, followed by Tukey’s test:+p < 0.05,
++p < 0.01 vs N-LDL. (B) Cells were incubated for 14 h without (Control) or
with 100, 150, or 200 μg/ml preparations of N-LDL, Ac-LDL, and Ox-LDL in
leucine-poor 10% LPDS DMEM containing [3H]leucine. ApoB secretion was
evaluated by specific immunoprecipitation and radioactivity measurement as
described under Methods. Statistical significance (n = 4) was assessed using
one-way ANOVA, followed by Tukey’s test: *p < 0.05; ***p < 0.001 vs
Control;++p < 0.05;+++p < 0.001 vs N-LDL.
Fig. 3. Effects of various LDL preparations on total secreted protein by HepG2
cells and cell viability. Cells were incubated without (Control) and with 100 or
200 μg/ml preparations of N-LDL, Ac-LDL, and Ox-LDL for 14 h in leucine-
poor 10% LPDS DMEM containing 5 μCi/ml [3H]leucine. Total secreted
proteins were precipitated using trichloroacetic acid and counted for
radioactivity. The inset shows the absence of effects of the native LDL (dashed
line) and Ox-LDL (solid line) on cytotoxicity as assessed by LDH release assay
expressed as total LDH activity of lysed cells.
790E. Bourdon et al. / Free Radical Biology & Medicine 41 (2006) 786–796
proteins or further incubated for 2 h in fresh medium without
label (chase). As shown in Table 3, incorporation of label into
apoAI during the pulse was lower in cells preincubated with
Ox-LDL than in cells incubated with N-LDL (from 7800 to
6700 dpm, −14%, p < 0.05). This loss was accounted for by a
decrease in apoAI synthesis. After the chase, the labeled apoAI
in the medium was lower when cells were preincubated with
Ox-LDL (from 9800 to 6600 dpm, −35%, p < 0.01 vs N-LDL).
Incubation with Ac-LDL resulted in levels of apoAI similar to
those of control and incubation with N-LDL. These data
indicate that the effects of N-LDL and Ox-LDL on apoAI
metabolism differ significantly. With N-LDL, degradation does
not seem to contribute to a large extent to the resulting apoAI
concentration. Conversely, in the presence of Ox-LDL, both its
synthesis is reduced and its degradation is enhanced as assessed
by the significant decrease in the chase/pulse ratio from 132 to
94% (down to 71%, p < 0.01 vs N-LDL). ApoAI secreted into
the medium was measured by ELISA. We found that incubation
with 200 μg/ml Ox-LDL for 24 h reduced apoAI mass in the
medium by 43%, from 3.56 ± 0.39 to 2.02 ± 0.17 ng/mg protein.
These results are in agreement with the pulse experiment.
not affected by preincubation with N-LDL or modified LDLs.
chase was noted in the presence of Ox-LDL (3800 to 6600 dpm,
+74%, p < 0.01 vs N-LDL). For apoB, the low ratio between
chase and pulse (22% for the control, Table 3) indicates that a
major part of apoB synthesized during the pulse has been
intracellularly degraded during the chase. These results are in
agreement with most studies on apoB synthesis, which have
established that intracellular degradation plays the major
ratio from 130 to 222% (by 70%, p < 0.01), clearly explains the
N-LDL. We could not measure the apoB mass into the medium
because of interference of the added LDL with our ELISA.
To explore the mechanisms by which Ox-LDLs alter apoAI
and apoB synthesis, we studied their effects on the levels of
apoAI and apoB mRNA by Northern blot analysis. After a 14-
hincubation,apoAImRNAlevels were significantlyreducedby
compared with N-LDL, both p < 0.01; Figs. 4A and 4B). Under
the same conditions, Ac-LDL led to a decrease especially at
higher concentration (200 μg/ml). In contrast, Ox-LDL did not
not shown). These results were also confirmed by semiquanti-
tative RT-PCR. We found that Ox-LDL induced a significant
decrease in mRNA coding for apoAI compared to N-LDL
(Control 100; N-LDL, Ac-LDL, and Ox-LDL,116.0, 112.0, and
88.4, respectively; p < 0.01), whereas, as expected, no change
was observed for apoB (Control 100; N-LDL, Ac-LDL, and Ox-
LDL, 100.1, 99.9, and 98.8, respectively; p = NS,).
Electrophoretic mobility and oxidation parameters in various LDL preparations
N-LDL (n = 3) Ac-LDL (n = 3)
4.84 ± 0.10⁎
0.109 ± 0.002⁎
Ox-LDL (n = 3)
3.46 ± 0.12⁎
0.390 ± 0.004⁎
0.17 ± 0.002
0.40 ± 0.30 0.97 ± 0.05 16.3 ± 1.53⁎
2.8 ± 1.3 21.3 ± 0.3⁎
361 ± 1.1⁎
Measurements were performed as described under Methods. REM, relative
electrophoretic mobility; TBARS, thiobarbituric acid-reactive substances.
Statistical significance between means was assessed using one-way ANOVA.
⁎p < 0.001 vs N-LDL.
Lipid composition in various LDL preparations
N-LDL (n = 6)Ac-LDL (n = 3) Ox-LDL (n = 6)
1249 ± 64⁎
624 ± 46⁎
220 ± 13⁎
2812 ± 188⁎
Total FA (nmol/mg)
Chol (μmol/mg protein)
1625 ± 46
832 ± 54
14 ± 6
3850 ± 127
1598 ± 51
824 ± 61
21 ± 8
3620 ± 176
Total OS (nmol/mg protein)
9.9 ± 0.3
13.7 ± 0.2
4.5 ± 0.1
10.1 ± 0.6
0.4 ± 0.3
14.2 ± 0.6
6.1 ± 0.5
13.8 ± 0.3⁎
46.4 ± 2.2⁎
33.2 ± 1.8⁎
42.6 ± 3.6⁎
31.3 ± 2.3⁎
6.4 ± 2.1⁎
Measurements were performed using procedures described under Methods.
Statistical significance between means was assessed using one-way ANOVA.
⁎p < 0.001 vs N-LDL or AC-LDL.
Synthesis, intracellular turnover, and secretion of apoAI and B in HepG2 as
assessed by pulse–chase experiment with [3H]leucine incorporation
8.1 ± 1.2
4.8 ± 0.9
7.8 ± 0.9
3.3 ± 0.8
7.6 ± 1.0
3.7 ± 0.3
6.7 ± 0.9a
3.1 ± 1.5
9.0 ± 1.9
1.0 ± 0.4
9.8 ± 2.2
3.8 ± 0.3c
7.4 ± 1.5
0.7 ± 0.2d
6.4 ± 1.2a,b
6.6 ± 0.5b,c
Chase (as % of pulse)
121 ± 18
22 ± 11
132 ± 12
130 ± 13c
98 ± 27
19 ± 2d
94 ± 19b,e
222 ± 23b,c
The experiments were performed in two six-well plates (one for pulse, one for
chase) as described under Methods. Results (means ± SD, n = 6) are expressed
as dpm. Pulse cells, radioactivity of washed cells obtained after pulse; Chase
total, radioactivity of supernatants and cells obtained after chase. Chase is also
expressed as a percentage of pulse (ratio of Chase total to Pulse cells in %).
aStatistically different from control: p < 0.05.
bStatistically different from values for N-LDL: p < 0.01.
cStatistically different from control: p < 0.001.
dStatistically different from values for N-LDL: p < 0.001.
eStatistically different from control: p < 0.01.
791 E. Bourdon et al. / Free Radical Biology & Medicine 41 (2006) 786–796
Influence of Ox-LDL on triglyceride and cholesteryl ester
It is well knownthat apoB secretion by liver cells is related to
lipid biosynthesis . Additional experiments were designed to
determine whether the various LDL preparations might exert
their effects through changes in lipid synthesis. Compared with
N-LDL, overnight incubation with Ac-LDL decreased CE in
HepG2 cells with almost no change in TG (Table 4). By
contrast, Ox-LDL drastically increased the syntheses and the
intracellular mass of both TG and CE.
To further study the relationship between Ox-LDL-induced
enhancement of lipid biosynthesis and apoB production, we
tested the effects of inhibitors of TG and CE biosynthesis. The
results are illustrated in Table 5 and further confirm our findings
(Fig. 2B and Table 4) indicating that Ox-LDL leads to an
increase in TG and CE concentrations in parallel with an
increase in apoB secretion (103%, p < 0.05 vs N-LDL). A
known inhibitor of TG synthesis, EPA , leads to a 36%
reduction in apoB in the presence of Ox-LDL along with
significant lowering of TG and CE synthesis. The decrease in
TG and CE biosynthesis induced by Ox-LDL was 30 and 48%,
respectively, when the ACAT inhibitor Sandoz 58035 was
added, and the corresponding apoB synthesis decreased from
203.5 to 156.3%. When both ACAT inhibitor and EPA were
added to the incubation medium, the effects of Ox-LDL on
apoB, TG, and CE synthesis were significantly reduced by 56,
60, and 50%, respectively, compared with N-LDL (Table 5).
These results mean that lipid biosynthesis is involved in the Ox-
LDL-mediated stimulation of apoB secretion. Fig. 5 shows
correlation studies indicating that apoB production is signifi-
cantly correlated with both CE and TG synthesis (r = 0.615,
p < 0.019, and r = 0.895, p < 0.0001, respectively).
ApoAI and apoB are important markers of atherosclerotic
risk. Oxidation processes are increasingly thought to be
Fig. 4. Effects of various LDL preparations on apoAI mRNA expression. (A)
Incubations were carried out as for Fig. 2. Aliquots of total RNA (approx 20 μg)
isolated from HepG2 cells were electrophoresed and Northern blot analysis was
performed using corresponding cDNA probes specific for apoAI or GAPDH as
described under Methods. (B) Quantitative analysis was done by scanning of
blots,andGAPDHsignalswereusedtonormalizethe RNAapplied ineach lane.
Data are presented as percentage change from Control incubations, namely
without LDLs. Statistical significance (n = 3) was assessed using one-way
ANOVA, followed by Tukey’s test:+p < 0.05;++p < 0.01 vs N-LDL.
Effects of various LDLs on lipid biosynthesis and mass in HepG2 cells
Triglycerides Cholesteryl esters
218 ± 31
237 ± 11
266 ± 23b,c
446 ± 17a,d
37.3 ± 3.2
42.6 ± 5.1
44.1 ± 3.8
61.8 ± 4.2a,c
10.3 ± 7.0
26.5 ± 1.5a
13.2 ± 7.3d
38.9 ± 1.6a,d
9.6 ± 1.8
14.8 ± 2.2
10.1 ± 0.9d
36.6 ± 2.4a,d
Lipid biosynthesis and net mass in HepG2 cells were quantified after thin-layer
chromatography fractionation as detailed under Methods after cell incubation
with or without (Control) 150 μg protein/ml of LDLs for 14 h in DMEM 10%
LPDS. No significant change in unesterified cholesterol content was observed
(mean range: 12.4–14.8 μg/mg cell protein). Data represent means ± SD of 3 to
10 separate experiments.
aDifferent from value for control: p < 0.001.
bDifferent from value for control: p < 0.01.
cDifferent from value for native LDL: p < 0.01.
dDifferent from value for native LDL: p < 0.001.
Effects of EPA and Sandoz 58035 on Ox-LDL-stimulated triglyceride,
cholesteryl ester, and apoB biosynthesis
Ox-LDL + EPA
Ox-LDL + Sdz
Ox-LDL + EPA + Sdz
100 ± 4.6
188.2 ± 7.2a
84.5 ± 14.5c
133.0 ± 11.5b,e
76.6 ± 25.2e
100 ± 5.7
146.8 ± 6.0a
130.9 ± 9.9b
76.3 ± 6.6b,c
73.8 ± 16.0e
100 ± 33
203.5 ± 23.6b
131.9 ± 36.9d
156.3 ± 24.2d
89.7 ± 15.2c
HepG2 cells were incubated with 150 μg protein/ml of N-LDL, Ac-LDL, or Ox-
LDL in the presence or absence of 0.5 mmol/L EPA and/or 4 μg/ml Sandoz
58035 (Sdz) for 20 h. Triglyceride (TG) and cholesteryl ester (CE) biosynthesis
was quantified in lipid extracts by separation using thin-layer chromatography
after cell incubation in DMEM 10% LPDS containing U14C-labeled sodium
acetate as detailed under Methods. Immunoprecipitation of apoB in supernatants
was conducted as described, after incubations in DMEM containing 5 μCi/ml
[3H]leucine. Values represent the data of the various incubations (means ± SD of
three separate dishes, each in quadruplicate) expressed as a percentage of the
data obtained with N-LDL. Absolute values for N-LDL (dpm/mg protein): TG,
162,849 ± 7491; CE, 17,531 ± 999; apoB, 1223 ± 433.
aStatistically different from N-LDL: p < 0.001.
bStatistically different from N-LDL: p < 0.01.
cStatistically different from Ox-LDL: p < 0.001.
dStatistically different from Ox-LDL: p < 0.05.
eStatistically different from Ox-LDL: p < 0.01.
792E. Bourdon et al. / Free Radical Biology & Medicine 41 (2006) 786–796
involved in atherogenesis. The purpose of the present study was
to examine the influence of Ox-LDL on apoAI and apoB
secretion by liver cells. Using HepG2 cells and by comparison
with N-LDL, we show that Ox-LDL induced a decrease in
apoAI and an increase in apoB secretion. Using pulse–chase
studies and Northern blot analyses, we demonstrate that
different mechanisms are responsible for these effects. The
Ox-LDL-induced reduction in apoAI production is the result of
both translational and transcriptional steps, whereas for apoB,
posttranscriptional steps are the only regulatory steps affected.
We also report that cellular lipid biosynthesis is a key
determinant for the enhancement of apoB synthesis by Ox-
LDL. These results bring in vitro data that help give a better
understanding of the clinical alteration of the apoAI and apoB
levels that govern the HDL to LDL ratio, which are all
biomarkers for CVD.
In humans, LDL-associated apoB levels (among the multiple
risk factors for atherosclerosis) are highly predictive of coronary
heart disease, one of the most prevalent causes of morbidity and
mortality in Western countries . The reasons for overpro-
duction of hepatic apoB, found in diabetes and coronary artery
disease patients, are not fully understood. Several human
experiments using [13C]leucine infusion proposed that the
primary defect might be a reduced LDL catabolic rate with
some, or no, change in LDL synthesis . On the other hand,
high HDL-apoAI levels have been associated with a reduced
number of cardiovascular events . Low HDL-cholesterol
and high LDL-cholesterol are often associated with alteration of
the redox status. This is characterized by circulating Ox-LDL
particles, which may exert their deleterious effects related to
atherothrombosis [5,39]. Although the precise localization of
the oxidation process is unknown, the presence of Ox-LDL has
been demonstrated in human atherosclerotic lesions, and
antibodies for Ox-LDL have been detected in plasma. Recently,
we have documented that Ox-LDL are found in type 2 diabetics
in relation to an increase in phospholipid transfer protein
activity . Treatment of hypercholesterolemic animals with
antioxidants, such as probucol, has been shown to reduce
atherosclerotic lesions [41,42], thus supporting the concept that
LDL oxidation is directly involved in atherogenesis. Although it
has been demonstrated that antioxidant therapy improved
antioxidant defense , solid data indicating that it reduces
atherosclerosis in humans is still lacking. In addition to
promoting foam cell and scavenger receptor expression, Ox-
LDL have other biological properties potentially related to the
atherogenic process, including cytotoxicity and modulation of
vascular reactivity . Ox-LDL may induce expression of
monocyte adhesion molecules, tissue factor, macrophage
chemotactic protein-1 (MCP-1), and macrophage-colony stim-
ulating factor in endothelial cells . To our knowledge, this is
the first study addressing the effects of copper-oxidized LDL on
the biosynthesis of apoB and apoAI.
Several studies have suggested that hepatic apoB production
is related to lipid biosynthesis and lipoprotein assembly
[3,34,46,47]. Data from various approaches in both primary
liver cells and the HepG2 cell line indicate that apoB
metabolism is regulated through degradation rather than its
biosynthesis . This idea is further confirmed by our pulse–
chase study results. [3H]Leucine incorporated into apoB during
the pulse was not altered by preincubation with N-LDL or
modified LDL. In contrast, a large increase in apoB-containing
lipoproteins in the medium after the chase was noted in the
presence of Ox-LDL (+74%, p < 0.01 vs N-LDL). The low ratio
between chase and pulse (22% for the control LPDS, Table 3)
indicates that a major part of the apoB newly synthesized during
the pulse has been degraded during the chase. Furthermore, Ox-
LDL-induced increase in apoB secretion was not associated
with an increase in apoB mRNA. Therefore, a reduction in apoB
degradation (by 74%, p < 0.01) clearly explains the increased
levels of apoB in the presence of Ox-LDL compared to N-LDL.
Under our conditions, Ox-LDL resulted in increases in both TG
and CE (by 88 and 47%, respectively) which are correlated with
apoB synthesis. These data are directly in line with the results of
Dashti et al. , who reported a positive relation between
short-term incubations with 1% ethanol and apoB production
rate in HepG2 as well as an increase in CE and TG synthesis. In
addition, we found that incubation with both an ACATinhibitor
and EPA (known TG synthesis inhibitor [35,48]) results not
only in an inhibition of Ox-LDL-induced TG and CE synthesis
but also in a significant reduction of apoB synthesis (Table 3).
These results indicate that lipid biosynthesis, especially TG
Fig. 5. Plots of correlation between apoB synthesis and lipid syntheses. The
correlation coefficient r and the probability value were determined using
absolute individual values of Tables 2 and 3 with Pearson’s rank test.
793E. Bourdon et al. / Free Radical Biology & Medicine 41 (2006) 786–796
synthesis, is deeply involved in the Ox-LDL-mediated stimu-
lation of apoB secretion. Several studies have emphasized the
role of cholesterol esterification and led to the development of
ACAT inhibitors . A putative stimulation of ACAT activity
would be in favor of a role for oxidized lipids such as oxysterols
. It was recently reported that lipid oxidation and oxidant
stress regulated liver apoB degradation . One limitation of
our work was that it was difficult to generate in vitro and hence
to study the exactly modified LDL that hepatocytes would be
exposed to in vivo. In the present study, we generated Ox-LDL
by copper-mediated oxidation. It was demonstrated that this
procedure modifies both the lipid and the protein moieties of
LDL. We have analyzed the lipid composition and oxidation
parameters of the various LDL preparations (Tables 1 and 2)
and these data confirm that both Ac-LDL and copper-oxidized-
LDL have enhanced electrophoretic mobility relative to N-
LDL. Conjugated dienes, lipoperoxides, and TBARS are
significantly higher in Ox-LDL than Ac-LDL and N-LDL.
Regarding lipid composition, no significant change in Ac-LDL
occurred compared to N-LDL, whereas the unsaturation index
of Ox-LDL phospholipids markedly diminished. Furthermore,
the total phospholipids were decreased in Ox-LDL and there
was a drastic increase in lysophospholipids. These changes
were also associated with a marked appearance of various
oxysterols. The different effects of Ac-LDL were probably
because the modification was directed only to LDL apoB.
Therefore, our results are strongly in favor of effects ascribable
to the lipid moiety. Although apoB modifications are important
for binding and degradation of native LDL through the LDL
receptor, there is increasing evidence suggesting that oxidized
lipids are responsible for most of the biological effects of Ox-
LDL [17,51,52]. It is well recognized that lipoproteins—
especially oxidized LDL—can modulate signal transduction
and gene expression in many cells . These effects can be
reproduced by in vitro incubation with the lipid moiety of Ox-
LDL or lipid oxidation products . We have recently
demonstrated that Ox-LDL may affect the antioxidant defense
by down-regulating serum albumin expression in HepG2 cells,
a proatherogenic effect that can be mimicked by oxysterols and
lysophosphatidylcholine . Oxidized phospholipids also
seem to be lipid constituents responsible for the Ox-LDL-
induced overexpression of MCP-1 and adhesion molecules in
endothelial cells [54,55].
In the present work, we show that Ox-LDL-mediated effects
on apoAI and apoB secretion occur through different mechan-
isms. These effects seem to be specific and independent of Ox-
LDL-induced cytotoxic effects as no significant variations in
total secreted proteins or LDH release were observed (Fig. 3) as
in our previous work . Pulse–chase experiments have
shown that these Ox-LDL-mediated effects can be explained by
both an enhanced intracellular degradation and an impaired
synthesis. These effects were not observed with either N-LDL
or Ac-LDL. A significant decrease in mRNA levels confirms
that transcriptional and posttranscriptional steps might be
altered, resulting in an Ox-LDL-induced decrease in apoAI
secretion. From our data, we could not discriminate between
decreased transcription and impaired mRNA stability.
Part of the changes we observed in lipid synthesis might
be explained by different lipid delivery by LDL to the
hepatocyte. We have not yet measured the relative uptake of
the various LDLs under our conditions. However, data from
Krieger’s group have indicated that the class B scavenger
receptor CD36, which could be present in hepatocytes and
HepG2 under certain inflammatory conditions, binds Ac-
LDL, Ox-LDL, and particularly lysine-modified proteins .
If this were the case, Ox-LDL and Ac-LDL would give
similar results. In the present study we found almost no effect
of Ac-LDL or even a decrease in apoB synthesis compared to
N-LDL (Fig. 2A and Table 3). Therefore, differences in LDL
uptake cannot fully explain our present findings. We have not
investigated the precise compounds responsible for these
effects further. However, as demonstrated by Dashti et al.
, who reported that apoAI secretion was increased by
ethanol-stimulated lipid synthesis, we can reasonably suggest
that, compared with N-LDL, lipid oxidation products present
in Ox-LDL might be responsible for these effects. This is in
line with data indicating that active lipid oxidation products
are present in oxidized lipoproteins and with our previous
data demonstrating that oxysterols and lysophospholipids
reproduced the Ox-LDL-mediated decrease in albumin
secretion by HepG2 .
It is tempting to extrapolate our findings to in vivo situations.
Clearly, elevated HDL levels are associated with a reduced
atherosclerotic risk in humans and animals . That is why
one of the strategies to lower the incidence of atherosclerosis is
to increase HDL levels, especially in patients bearing a low
plasma HDL-cholesterol. The mechanisms by which HDL acts
to inhibit the development of atherosclerosis are unknown. The
principal structural protein of HDL is apoAI and HDL
cholesterol levels are highly correlated with plasma apoAI.
Importantly, studies using transgenic mice overexpressing the
human apoAI gene show increased HDL cholesterol levels and
a reduced atherosclerosis susceptibility. There are at least three
existing hypotheses to explain the role of HDL and apoAI in
protecting against the development of atherosclerotic lesions.
First, involvement of HDL and apoAI in reverse cholesterol
transport has been well documented. Second, direct effects of
HDL and apoAI on the vessel wall and on LDL oxidation have
been described. Various enzymes such as platelet-activating
factor–acetyl hydrolase and paraoxonase, conferring “antiox-
idative” properties, are mainly distributed in HDL fractions. The
impact of Ox-LDL on these agents needs to be investigated
further. Third, an inverse relationship between HDL and apoB-
containing lipoproteins has been proposed as a key parameter to
characterize atherosclerotic risk. In view of the important
contribution of the liver to the overall apoAI pool, it is
interesting to propose that the opposite effects of Ox-LDL on
apoB and apoAI secretion described here might contribute to a
better understanding of HDL protective function.
In conclusion, Ox-LDL induced significant but opposite
dose-dependent effects on apoB and apoAI secretion. These
effects seem related to the oxidation process as N-LDL and
Ac-LDL did not show similar effects. The mechanisms
responsible for the action of Ox-LDL are different for each
794 E. Bourdon et al. / Free Radical Biology & Medicine 41 (2006) 786–796
apoprotein. Stimulation of lipid biosynthesis is related to Ox-
LDL-induced apoB increase, whereas a reduction in mRNA is
responsible for the reduced concentration of apoAI. It has
been largely documented that an increase in apoB and a
decrease in apoAI are directly related to the secretion of LDL
and HDL, respectively. It is evident that further studies are
needed to confirm these in vitro effects and especially to
obtain in vivo confirmation. However, these data point to new
effects induced by oxidative stress mediated via oxidation of
LDL particles, a clinical situation largely described in
atherosclerotic patients. This stress may lead to an elevated
LDL/HDL ratio, one of the strongest predictors of myocardial
This work was supported by the Institut National de la Santé
et de la Recherche Médicale (INSERM), Conseil Régional de
Bourgogne, Université de Bourgogne. The excellent technical
help of Ann Chamberland, Lucie Boulet, Michel Tremblay, and
Jacques Lavigne is greatly acknowledged. E.B. was supported
by a fellowship from the Ministère de l’Education Nationale, de
l’Enseignement Supérieure, et de la Recherche.
 Assmann, G.; Cullen, P.; Jossa, F.; Lewis, B.; Mancini, M. International
Task Force for Prevention of Coronary Heart Disease. Reducing the risk—
The scientific background to primary and secondary prevention of
coronary heart disease, a worldwide view. Arterioscler. Thromb. Vasc.
Biol. 19:1819–1824; 1999.
 Hussain, M. M.; Strickland, D. K.; Bakillah, A. The mammalian low-
density lipoprotein receptor family. Annu. Rev. Nutr. 19:141–172;
A. A.; Reuben, M. A.; Bondjers, G. Apolipoprotein B: structure, biosyn-
 Steinberg, D. Role of oxidized LDL and antioxidants in atherosclerosis.
Adv. Exp. Med. Biol. 369:39–48; 1995.
 Steinberg, D.; Parthasarathy, S.; Carew, T. E.; Khoo, J. C.; Witztum, J. L.
Beyond cholesterol: modifications of low-density lipoprotein that increase
its atherogenicity. N. Engl. J. Med. 320:915–924; 1989.
 Esterbauer, H.; Gebicki, J.; Puhl, H.; Jürgens, G. The role of lipid
peroxidation and antioxidants in oxidative modification of LDL. Free
Radic. Biol. Med. 13:341–390; 1992.
 De Backer, G.; De Bacquer, D.; Kornitzer, M. Epidemiological aspects of
high density lipoprotein cholesterol. Atherosclerosis 137:S1–S6; 1998.
 Martinez, L. O.; Jacquet, S.; Terce, F.; Collet, X.; Perret, B.; Barbaras, R.
New insight on the molecular mechanisms of high-density lipoprotein
cellular interactions. Cell Mol. Life Sci 61:2343–2360; 2004.
 Johnson, W. J.; Mahlberg, F. H.; Rothblat, G. H.; Phillips, M. C.
Cholesterol transport between cells and high-density lipoproteins. Bio-
chim. Biophys. Acta Lipids Lipid Metab. 1085:273–298; 1991.
 Parthasarathy, S.; Barnett, A. H.; Fong, L. G. High-density lipoprotein
inhibits the oxidative modification of low-density lipoprotein. Biochim.
Biophys. Acta 1044:275–283; 1990.
 Kauser, K.; Rubanyi, G. M. Potential cellular signaling mechanisms
mediating upregulation of endothelial nitric oxide production by estrogen.
J. Vasc. Res. 34:229–236; 1997.
 Mackness, M. I.; Durrington, P. N. HDL, its enzymes and its potential to
influence lipid peroxidation. Atherosclerosis 115:243–253; 1995.
 Cappola, A. R.; Ladenson, P. W. Hypothyroidism and atherosclerosis.
J. Clin. Endocrinol. Metab. 88:2438–2444; 2003.
 Packard, C. J. Understanding coronary heart disease as a consequence of
defective regulation of apolipoprotein B metabolism. Curr. Opin. Lipidol.
 Ginsberg, H. N. Insulin resistance and cardiovascular disease. J. Clin.
Invest 106:453–458; 2000.
 Duvillard, L.; Pont, F.; Florentin, E.; Gambert, P.; Verges, B. Inefficiency
of insulin therapy to correct apolipoprotein A-I metabolic abnormalities in
non-insulin-dependent diabetes mellitus. Atherosclerosis 152:229–237;
 Bourdon, E.; Loreau, N.; Davignon, J.; Bernier, L.; Blache, D.
Involvement of oxysterols and lysophosphatidylcholine in the oxidized
LDL-induced impairment of serum albumin synthesis in HepG2 cells.
Arterioscler. Thromb. Vasc. Biol. 20:2643–2650; 2000.
 Hatch, F. T. Practical methods for plasma lipoprotein analysis. Adv. Lipid
Res. 6:1–68; 1968.
 Smith, P. K.; Krohn, R. I.; Hermanson, G. T.; Mallia, A. K.; Gartner, F. H.;
Provenzano, M. D.; Fujimoto, E. K.; Goeke, N. M.; Olson, B. J.; Klenk,
D. C. Measurement of protein using bicinchoninic acid. Anal. Biochem.
 Basu, S. K.; Goldstein, J. L.; Anderson, R. G. W.; Brown, M. S.
Degradation of cationized low density lipoprotein and regulation of
cholesterol metabolism in homozygous familial hypercholesterolemia
fibroblasts. Proc. Natl. Acad. Sci. USA 73:3178–3182; 1976.
 El-Saadani, M.; Esterbauer, H.; El-Sayed, M.; Goher, M.; Nassar, A. Y.;
Jürgens, G. A spectrophotometric assay for lipid peroxides in serum
lipoproteins using a commercially available reagent. J. Lipid Res.
 Blache, D.; Rodriguez, C.; Davignon, J. Pro-oxidant effects of 7-
hydroperoxycholest-5-en-3β-ol on the copper-initiated oxidation of low
density lipoprotein. FEBS Lett. 357:135–139; 1995.
 Blache, D.; Durand, P.; Girodon, F.; Gesquière, L.; Loreau, N.
Determination of sterols, oxysterols and fatty acids of phospholipids in
cells and lipoproteins: a one sample method. J. Am. Oil Chem. Soc.
 Folch, J.; Lees, M.; Sloane-Stanley, G. H. A simple method for the
isolation and purification of total lipids from animal tissues. J. Biol. Chem.
 Hill, T. B.; Whatley, S. A simple, rapid microassay for DNA. FEBS Lett.
 Fredenrich, A.; Giroux, L. M.; Tremblay, M.; Krimbou, L.; Davignon, J.;
Cohn, J. S. Plasma lipoprotein distribution of apoC-III in normolipidemic
and hypertriglyceridemic subjects: comparison of the apoC-III to apoE
ratio in different lipoprotein fractions. J. Lipid Res. 38:1421–1432; 1997.
 Kempen, H. J.; Imbach, A. P.; Giller, T.; Neumann, W. J.; Hennes, U.;
Nakada, N. Secretion of apolipoproteins A-I and B by HepG2 cells:
regulation by substrates and metabolic inhibitors. J. Lipid Res.
 Shoulders, C. C.; Kornblihtt, A. R.; Munro, B. S.; Barralle, F. Gene
structure of human apolipoprotein A-I. Nucleic Acids Res. 11:2827–2837;
 Karathanasis, S. K.; Zannis, V. I.; Breslow, J. L. Isolation and
characterization of the human apolipoprotein A-I gene. Proc. Natl.
Acad. Sci. USA 89:6147–6151; 1983.
 Tso, J. Y.; Sun, X.-H.; Kao, T.; Reece, K. S.; Wu, R. Isolation and charac-
terization of rat and human glyceraldehyde-3-phosphate dehydrogenase
cDNAs: genomic complexity and molecular evolution of the gene. Nucleic
Acids Res. 13:2485–2502; 1985.
 Darnfors, C.; Nilsson, J.; Protter, A. A.; Carlson, P.; Talmud, P. J.;
Humphries, S. E.; Whalstrom, J.; Wiklund, O.; Bjurssell, G. RFLPs for the
human apolipoprotein B gene: HincII and PvuII. Nucleic Acids Res.
 Hara, A.; Radin, N. S. Lipid extraction of tissues with a low-toxicity
solvent. Anal. Biochem. 90:220–226; 1978.
 Knowles, B. B.; Howe, C. C.; Aden, D. P. Human hepatocellular
carcinoma cell lines secrete the major plasma proteins and hepatitis B
surface antigen. Science 209:497–499; 1980.
795E. Bourdon et al. / Free Radical Biology & Medicine 41 (2006) 786–796
 Boström, K.; Wettesten, M.; Boren, J.; Bondjers, G.; Wiklund, O.;
Olofsson, S. -O. Pulse–chase studies of the synthesis and intracellular
transport of apolipoprotein B100 in HepG2 cells. J. Biol. Chem.
 Harris, W. S.; Windsor, S. L.; Caspermeyer, J. J. Modification of lipid-
related atherosclerosis risk factors by omega3 fatty acid ethyl esters in
hypertriglyceridemic patients. J. Nutr. Biochem. 4:706–712; 1993.
 Rader, D. J.; Hoeg, J. M.; Brewer H. B., Jr. Quantitation of plasma
apolipoproteins in the primaryand secondarypreventionof coronaryartery
disease. Ann. Intern. Med. 120:1012–1025; 1994.
 Duvillard, L.; Pont, F.; Florentin, E.; Galland-Jos, C.; Gambert, P.; Verges,
B. Metabolic abnormalities of apolipoprotein B-containing lipoproteins in
non-insulin-dependent diabetes: a stable isotope kinetic study. Eur. J. Clin.
Invest. 30:685–694; 2000.
 Barter, P. J.; Rye, K. A. High density lipoproteins and coronary heart
disease. Atherosclerosis 121:1–12; 1996.
 Holvoet, P.; Collen, D. Oxidized lipoproteins in atherosclerosis and
thrombosis. FASEB J. 8:1279–1284; 1994.
 Schneider, M.; Verges, B.; Klein, A.; Miller, E. R.; Deckert, V.;
Desrumaux, C.; Masson, D.; Gambert, P.; Brun, J. M.; Fruchart-Najib, J.;
Blache, D.; Witztum, J. L.; Lagrost, L. Alterations in plasma vitamin E
distribution in type 2 diabetic patients with elevated plasma phospholipid
transfer protein activity. Diabetes 53:2633–2639; 2004.
H.; Kawai, C. Probucol prevents the progression of atherosclerosis in
Watanabe heritable hyperlipidemic rabbit, an animal model for familial
hypercholesterolemia.Proc. Natl.Acad.Sci. USA84:5928–5931;1987.
 Carew, T. E.; Schwenke, D. C.; Steinberg, D. Antiatherogenic effect of
probucol unrelated to its hypocholesterolemic effect: evidence that
antioxidants in vivo can selectively inhibit low density lipoprotein
degradation in macrophage-rich fatty streaks and slow the progression of
atherosclerosis in the Watanabe heritable hyperlipidemic rabbit. Proc.
Natl. Acad. Sci. USA 84:7725–7729; 1987.
 Girodon, F.; Blache, D.; Monget, A. L.; Lombart, M.; Brunet-Lecompte,
P.; Arnaud, J.; Richard, M. J.; Galan, P. Effect of two year supplementation
with low dose antioxidant vitamins and/or minerals in elderly subjects on
levels of nutrients and on antioxidant defense parameters. J. Am. Coll.
Nutr. 16:357–365; 1997.
 Berliner, J. A.; Heinecke, J. W. The role of oxidized lipoproteins in
atherogenesis. Free Radic. Biol. Med. 20:707–727; 1996.
 Cominacini, L.; Garbin, U.; Pasini, A. F.; Davoli, A.; Campagnola, M.;
Contessi, G. B.; Pastorino, A. M.; Lo Cascio, V. Antioxidants inhibit the
expression of intercellular cell adhesion molecule-1 and vascular cell
adhesion molecule-1 induced by oxidized LDL on human umbilical vein
endothelial cells. Free Radic. Biol. Med. 22:117–127; 1997.
 Brown, A. M.; Wiggins, D.; Gibbons, G. F. Manipulation of cholesterol
and cholesteryl ester synthesis has multiple effects on the metabolism of
apolipoprotein B and the secretion of very-low-density lipoprotein by
primary hepatocyte cultures. Biochim. Biophys. Acta Mol. Cell Biol. Lipids
 Dashti, N.; Franklin, F. A.; Abrahamson, D. R. Effect of ethanol on the
synthesis and secretion of apoA-I- and apoB-containing lipoproteins in
HepG2 cells. J. Lipid Res. 37:810–824; 1996.
 Pan, M.; Cederbaum, A. I.; Zhang, Y. L.; Ginsberg, H. N.; Williams, K. J.;
Fisher, E. A. Lipid peroxidation and oxidant stress regulate hepatic
apolipoprotein B degradation and VLDL production. J. Clin. Invest.
 Burnett, J. R.; Wilcox, L. J.; Huff, M. W. Acyl coenzyme A: cholesterol
acyltransferase inhibition and hepatic apolipoprotein B secretion. Clin.
Chim. Acta 286:231–242; 1999.
 Cheng, D.; Chang, C. C. Y.; Qu, X.; Chang, T. Y. Activation of acyl-
coenzyme A:cholesterol acyltransferase by cholesterol or by oxysterol in a
cell-free system. J. Biol. Chem. 270:685–695; 1995.
 Nagy, L.; Tontonoz, P.; Alvarez, J. G. A.; Chen, H.; Evans, R. M. Oxidized
LDL regulates macrophage gene expression through ligand activation of
PPAR gamma. Cell 93:229–240; 1998.
 Gesquière, L.; Loreau, N.; Minnich, A.; Davignon, J.; Blache, D.
Oxidative stress leads to cholesterol accumulation in vascular smooth
muscle cells. Free Radic. Biol. Med. 27:134–145; 1999.
 Lusis, A. J.; Navab, M. Lipoprotein oxidation and gene expression in the
artery wall: new opportunities for pharmacologic intervention in
atherosclerosis. Biochem. Pharmacol. 46:2119–2126; 1993.
 Schmitt, A.; Nègre-Salvayre, A.; Troly, M.; Valdiguié, P.; Salvayre, R.
Phospholipid hydrolysis of mildly oxidized LDL reduces their cytotoxicity
to cultured endothelial cells: potential protective role against atherogen-
esis. Biochim. Biophys. Acta Lipids Lipid Metab. 1256:284–292; 1995.
 Berliner, J.; Leitinger, N.; Watson, A.; Huber, J.; Fogelman, A.; Navab, M.
Oxidized lipids in atherogenesis: formation, destruction and action.
Thromb. Haemostasis 78:195–199; 1997.
 Acton, S.; Scherer, P. E.; Lodish, H. F.; Krieger, M. Expression cloning of
SR-BI, a CD36-related class B scavenger receptor. J. Biol. Chem.
 Plump, A. S.; Scott, C. J.; Breslow, J. L. Human apolipoprotein A-I gene
expression increases high density lipoprotein and suppresses atheroscle-
rosis in the apolipoprotein E- deficient mouse. Proc. Natl. Acad. Sci. USA
796 E. Bourdon et al. / Free Radical Biology & Medicine 41 (2006) 786–796