Role of plasma membrane coenzyme Q on the regulation of apoptosis.

Abstract

Serum withdrawal is a model to study the mechanisms involved in the induction of apoptosis caused by mild oxidative stress. Apoptosis induced by growth factors removal was prevented by the external addition of antioxidants such as ascorbate, alpha-tocopherol and coenzyme Q (CoQ). CoQ is a lipophilic antioxidant which prevents oxidative stress and participates in the regeneration of alpha-tocopherol and ascorbate in the plasma membrane. We have found an inverse relationship between CoQ content in plasma membrane and lipid peroxidation rates in leukaemic cells. CoQ10 addition to serum-free culture media prevented both lipid peroxidation and cell death. Also, CoQ10 addition decreased ceramide release after serum withdrawal by inhibition of magnesium-dependent plasma membrane neutral-sphingomyelinase. Moreover, CoQ10 addition partially blocked activation of CPP32/caspase-3. These results suggest CoQ of the plasma membrane as a regulator of initiation phase of oxidative stress-mediated serum withdrawal-induced apoptosis.

Full text

Page 1

BioFactors 9 (1999) 171–177
IOS Press
171
Role of plasma membrane coenzyme Q
on the regulation of apoptosis
G. López-Llucha, M.P. Barrosob, S.F. Martínb, D.J.M. Fernández-Ayalab, C. Gómez-Díazb,
J.M. Villalbaband P. Navasa,∗
aLaboratorio Andaluz de Biología, Universidad Pablo de Olavide, 41013 Sevilla, Spain
bDepartamento de Biología Celular, Facultad de Ciencias, Universidad de Córdoba, Córdoba, Spain
Abstract.Serumwithdrawal isamodel tostudy themechanisms involved intheinduction of apoptosis caused by mildoxidative
stress. Apoptosis induced by growth factors removal was prevented by the external addition of antioxidants such as ascorbate,
α-tocopherol and coenzyme Q (CoQ). CoQ is a lipophilic antioxidant which prevents oxidative stress and participates in the
regeneration of α-tocopherol and ascorbate in the plasma membrane. We have found an inverse relationship between CoQ
content in plasma membrane and lipid peroxidation rates in leukaemic cells. CoQ10 addition to serum-free culture media
prevented both lipid peroxidation and cell death. Also, CoQ10 addition decreased ceramide release after serum withdrawal by
inhibition of magnesium-dependent plasma membrane neutral-sphingomyelinase. Moreover, CoQ10addition partially blocked
activation of CPP32/caspase-3. Theseresultssuggest CoQof theplasmamembrane asaregulator of initiationphase of oxidative
stress-mediated serum withdrawal-induced apoptosis.
1. Introduction
Plasma membrane contains an electron transport chain involved in the maintenance of an antioxi-
dant system (see Navarro et al. this volume), involving ascorbate, α-tocopherol and CoQ. Ascorbate is
a first order antioxidant which scavenges lipid peroxidation-initiating radicals [10]. In addition, ascor-
bate reduces membrane bound α-tocopheroxyl radicals produced by the oxidation of α-tocopherol after
scavenging of lipid-peroxyl radicals thus restoring the antioxidant capacity of this compound [10,12].
CoQ is a lipid-soluble antioxidant which can be considered as the central molecule in plasma membrane
protection against lipid peroxidation because it is directly reduced by cytochrome b5reductase [31,36],
and maintains both ascorbate and α-tocopherol in their reduced state [10,12,25].
Serum withdrawal causes a mild oxidative stress to cells in culture leading to membrane damage
and cell death [17]. Serum removal induces ceramide release to the cytosol by the activation of Mg2+-
dependent neutral sphingomyelinase (N-SMase) [21]. Ceramides are able to induce cell death after its
intracellular accumulation by activating proteases of the caspase family [18,27]. Thus, ceramide accu-
mulation appears as a key component in the stress response pathway triggered by serum removal [14].
Further, cellular accumulation of antioxidants can prevent apoptosis induced by growth factors depriva-
tion [3,13,32].
We show here the protection caused by antioxidants such as ascorbate, α-tocopherol and CoQ against
apoptosis induced by serum removal. Also, we demonstrate a correlation between the content of CoQ at
the plasma membrane and the extinction of lipid peroxidation. Further, plasma membrane CoQ inhibits
*Correspondence to: Prof. P. Navas, Laboratorio Andaluz de Biología, Universidad Pablo de Olavide, Carretera de Utrera
Km. 1, 41013 Sevilla, Spain.
0951-6433/99/$8.00  1999 – IOS Press. All rights reserved

Page 2

172
G. López-Lluch et al. / Role of CoQ in apoptosis
neutral sphingomyelinase and, as a consequence, prevents ceramide release and caspase-3 activation.
CoQ, as central part of the antioxidant system of the plasma membrane, regulates the initiation phase of
apoptosis induced by growth factor withdrawal.
2. Material and methods
2.1. Cell lines and cultures
HL-60 (myelocytic), K562 (erithroblastic), and CEM (T-lymphoid) cells, were cultured in RPMI-
1640 medium (Sigma, Spain) supplemented with 10% fetal calf serum (FCS) (PAA Labor, Austria),
100 U/ml penicillin, 100 µg/ml streptomycin and 2 mM glutamine (Sigma, Spain). Daudi (B-lymphoid)
cells were cultured in the same medium supplemented with 20% FCS. Mitochondrial deficient ρocells
were generated by culturing HL-60 cells in the presence of 50 ng/ml ethidium bromide (Sigma, Spain)
for 5–6 weeks as described elsewhere [8] and cultured in the same medium as HL-60 cells but with the
addition of 1 mM pyruvate and 50 ng/ml uridine (Sigma, Spain). Prior to serum withdrawal experiments,
cells were harvested and washed twice in serum-free medium. Viability was assessed by trypan blue dye
exclusion method.
2.2. Plasma membrane isolation and CoQ quantification
Microsomes were obtained from cell homogenates and plasma membrane vesicles were then isolated
by the two-phase partition method [1]. Membranes were resuspended in 50 mM Tris/HCl, pH 7.6 con-
taining 10% glycerol, 1 mM PMSF and 1 mM DTT, and stored under liquid nitrogen or at −86◦C until
needed. Purity was checked by marker enzyme analysis [1]. CoQ quantification was performed as de-
scribed previously [2]. CoQ determination was carried out by HPLC separation with UV monitoring at
275 nm. CoQ was quantified by integration of peak areas and comparison with external standards, and
referred to membrane protein. The procedure recovered more than 98% CoQ.
2.3. Lipid peroxidation measurements
Cells were preincubated with 45 µM cis-parinaric acid (Molecular Probes, USA) in 0.1 M Tris/HCl,
pH 7.5, for 5 min as described [2]. Fluorescence measurements were carried out at 37◦C with a stirred
cuvette in a SFM 25 spectrofluorometer (Kontron, Germany). Excitation wavelength was 334 nm and
emission wavelength, 375 nm.
2.4. Apoptosis determination
Apoptosis was detected by the labeling of 3?OHends of DNAbreaks (Apotag detection system, Oncor,
USA) as specified by the manufacturer, or by cell cycle analysis using flow cytometry after fixation of
cells in 70% ethanol and labeling with PI staining buffer (PBS pH 7.5, 0.1 mM EDTA, 50 U/ml DNAse
free-RNAse A and 50 µg/ml PI (Sigma, Spain). Cell cycle analysis was performed in a FACScan and
data analyzed using a Lysis II software (Becton–Dickinson, USA). Apoptotic cells were determined by
the presence of subG0/1 population.

Page 3

G. López-Lluch et al. / Role of CoQ in apoptosis
173
2.5. Ceramide and SMase activity determination
Ceramide production was determined as previously described [2,21]. Cells were harvested at indicated
times of incubation and lipid extracted as described by Bligh and Dyer [5]. After drying under nitrogen
and resuspension in chloroform, lipids were used for phosphate measurements and diacylglycerol kinase
assay as described [29]. Ceramide-1-phosphate generated in the reaction was resolved by TLC using
chloroform/acetone/methanol/acetic acid/water (10:4:3:2:1) as solvent. Ceramide-1-phosphate spots
were quantified with an automatic TLC-lineal analyzer (Berthold, Germany), using external standards,
and normalized to phosphate as indicator of total phospholipids.
N-SMase was measured by the method of Okazaki et al. [28]. Thirty nM final concentration of a
mixture of cold and [methyl-14C]-sphingomyelin (Amersham Iberica, Spain) was used as substrate for
the SMase. Enzymatic hydrolysis of [methyl-14C]-sphingomyelin by SMase was quantified using a liq-
uid scintillation counter (Beckman, USA). SMase activity was expressed as nmol sphingomyelin hy-
drolyzed/mg protein/hour.
2.6. Caspase activity determination
Cell lysates were prepared by five cycles of freezing and thawing and vortex in extraction buffer
(50 mM PIPES, pH 7.4; 50 mM KCl, 5 mM EGTA, 2 mM MgCl2, 1 mM DTT; 20 mM cytochalasin
B; 1 mM PMSF; 1 µg/ml leupeptin, 1 µg/ml pepstatin, 50 µg/ml antipain and 10 µg/ml chymopapain).
After centrifugation at 12,000 × g for 15 min at 4◦C, 50 µg of protein from extract was diluted in assay
buffer (100 mM HEPES-KOH buffer, pH 7.5; 10% sucrose, 0.1% CHAPS, 0.1 mg/ml ovoalbumin, and
10 mM DTT) and CPP32 specific substrate, Ac-DEVD-pNa (Bachem, Switzerland) was added. Activity
was developed in 96 well plates at 37◦C and p-nitroanilide (pNa) release was determined spectrometri-
cally by using a multiwell plate reader at 405 nm. Activity is expressed in units (1 unit = 1 pmol pNa
released/30 min/mg protein) as was determined by Enari et al. [9].
3. Results and discussion
Thereduced form of CoQhas been proposed to act as achain-breaking antioxidant in the plasma mem-
brane [11]. This role depends on the existence of a recycling system to reduce the oxidized CoQ [25], and
also on its capacity to regenerate α-tocopherol [22]. Two enzymes, the cytochrome b5reductase and DT-
diaphorase have been shown to act as CoQ-reductases [4,24,36] which are important in the maintenance
of appropriate levels of antioxidants in the plasma membrane [7,37].
Cells in culture undergo controlled rates of lipid peroxidation involved in their surviving equilibrium.
As is show in Fig. 1, we found a clear inverse correlation between the content of CoQ in the plasma
membrane and the rates of lipid peroxidation in different leukemic cell lines when submitted to oxidative
stress. The higher rate of peroxidation showed by K562 cells corresponded to the lower content of CoQ.
Also differences were found between ρoHL-60 cells and parental HL-60 cells, the former showing lower
rates of lipid peroxidation and a higher content of CoQ. Accordingly, serum deprivation induces lower
rates of both apoptosis and lipid peroxidation in ρoHL-60 than in HL-60 cells [2]. Moreover, after serum
deprivation the content of CoQ in plasma membrane was increased in ρoHL-60 cells than in parental
HL-60 cells [2]. Lipid peroxidation induced in vitro in egg phosphatidylcholine liposomes enriched with
CoQ was prevented by the addition of purified cytochrome b5reductase from plasma membrane [35].

Page 4

174
G. López-Lluch et al. / Role of CoQ in apoptosis
Fig. 1. Relationship between CoQ and lipid peroxidation in leukemic cells in culture. HL-60 (a), ρoHL-60 (2), Daudi ("), and
K562 (Q) cells were cultured in 10% FCS-supplemented culture media. Cells were harvested and plasma membrane obtained
as described in “Materials and methods”. Samples were divided in two fractions and both CoQ and lipid peroxidation rates
were determined as described in “Material and methods”. Data represent the mean ± SD from four separate experiments.
Thus, CoQ seems to play a central role in antioxidant protection against lipid peroxidation in plasma
membrane.
Bcl-2 regulates apoptosis induced by several factors including serum withdrawal [26]. Bcl-2 has been
proposed to function as antioxidant protecting cells from oxidative injury induced by most apoptotic
factors [16] although this could be due to an indirect effect through the inhibition of cytochrome c release
from mitochondria and the subsequent production of reactive oxygen species from the mitochondrial
electronic chain. In order to test the role of Bcl-2 in the protective effect of antioxidants such as CoQ,
ascorbate and α-tocopherol in apoptosis, we studied their effect in HL-60 cells, expressing Bcl-2, and
Daudi cells, which donot express this protein. Externally added antioxidants prevented apoptosis induced
by serum withdrawal in both cell lines (Fig. 2). In both cases, ascorbate, α-tocopherol and reduced CoQ10
prevented apoptosis induced after 48 h of serum withdrawal. Addition of oxidized CoQ10resulted in a
less but significant protective effect in both cell lines. Bcl-2 is located in different endomembranes but
not in the plasma membrane [16], where the antioxidant system integrated by ascorbate, α-tocopherol
and CoQ maintains its protective functions [12,22,24,36,37]. Plasma membrane performs then as the
first barrier for the protection against oxidative stress generated from the outer side of cells. Prevention
of apoptosis by antioxidants was then independent of the presence of Bcl-2 in cells, suggesting they act
upstream Bcl-2 in apoptosis pathway.
To determine how CoQ10prevents apoptosis induced by serum withdrawal in leukemic cells, we stud-
ied the release of ceramide, a known factor in apoptosis induced by several agents including serum
deprivation [21]. By using CEM cells, a lymphoblastic cell line with moderate expression of Bcl-2, we
determined the activity of N-SMase, ceramide release and activation of caspase-3/CPP32, one of the
main executioners of apoptosis (Table 1). After 48 h of culture in serum free medium, CEM cells showed
a high rate of apoptosis, which was 25% prevented by CoQ10addition. N-SMase activity increased 20
fold in serum-free cultures after 1 h of incubation respect to 10% FCS-cultures. CoQ10addition partially

Page 5

G. López-Lluch et al. / Role of CoQ in apoptosis
175
Fig. 2. Effect of antioxidants on apoptosis induced by serum withdrawal. Cells were grown in the absence of serum or sup-
plemented with 0.2 mM ascorbate (Asc), 40 µM α-tocopherol (α-Toc), 40 µM oxidized CoQ10 (ox-CoQ) or 40 µM reduced
CoQ10 (red-CoQ). Control cells were cultured in the presence of 10% FCS. (A) HL-60 cells. (B) Daudi cells. Apoptosis was
detected by using Apotag detection kit. *Significant differences vs. serum-free cultures without antioxidant supplement from
four different experiments, p ? 0.05.
Table 1
Role of CoQ10on the activation of apoptosis in serum-deprived CEM cells
Culture
conditions
10% FCS
0% FCS
+ CoQ10
Apoptosis was determined by flow cytometry after 48 h of incubation. N-SMase and
ceramide were determined after 1 h of incubation. CPP32 activity was determined after
2 h of incubation. Data represent the mean (n = 3), and *significant differences vs.
culture in absence of serum, p ? 0.05. SD was less than 10% in each case.
Apoptosis
(%)
5
80
60*
N-SMaseCeramide
(pmol/nmol Pi)
45
90
46*
CPP32
(units)
1200
4100
2450*
(nmol SM /mg protein/h)
2.0
40.5
27.3*
inhibited N-SMase activation. Ceramide release was also inhibited by CoQ10addition to serum-free
cultures. The role of ceramide in apoptotic program has been related to the activation of CPP32 [14].
CPP32 activity was increased in serum-free cultured cells and the addition of CoQ10inhibited about
40% of this increase. Thus, the presence of CoQ10in serum-free cultures partially avoided the activation
of apoptotic machinery through inhibition of SMase pathway. Ceramide quantification must be taken as
qualitative instead of quantitative since the diacylglycerol kinase assay used is currently under discus-
sion [33].
To date, the mechanism of SMase activation after serum withdrawal is not known, but must be re-
lated to the oxidative stress and lipid peroxidation produced in cultured cells by serum withdrawal [2].
Phospholipase A2(PLA2) is activated by hydroperoxides in plasma membrane [15,30]. Also, a dramatic
increase in Ca2+-independent PLA2activity is induced by oxidative damage in vivo [23], possibly to
increase metabolism of fatty acid hydroperoxides [34]. Moreover, PLA2has been related to activation
of N-SMase in HL-60 cells and murine fibroblasts [19,20]. Thus, CoQ10and other antioxidants added