Conversely,inhibitors of lipid oxidation and treat-ments that decrease the unsaturation ratio of fatty acids decrease kinetics of Bax-induced death.. Manipulation of the unsaturation deg
Trang 1Bax-induced cell death in yeast depends on mitochondrial lipid
oxidation
Muriel Priault1,*, Jean-Jacques Bessoule2, Angela Grelaud-Coq1, Nadine Camougrand1
and Ste´phen Manon1
1
UMR5095 C.N.R.S./Universite´ de Bordeaux 2, Bordeaux, France;2UMR5544 C.N.R.S./Universite´ de Bordeaux 2, Bordeaux, France
The oxidant function of pro-apoptotic protein Bax was
investigated through heterologous expression in yeast
Direct measurements of fatty acid content show that
Bax-expression induces oxidation of mitochondrial lipids This
effect is prevented by the coexpression of Bcl-xL The
oxi-dation actually could be followed on isolated mitochondria
as respiration-induced peroxidation of polyunsaturated
cis-parinaric acid and on whole cells as the increase in the
amount of thiobarbituric acid-reactive products Treatments
that increase the unsaturation ratio of lipids,making them
more sensitive to oxidation,increase kinetics of Bax-induced
death Conversely,inhibitors of lipid oxidation and
treat-ments that decrease the unsaturation ratio of fatty acids
decrease kinetics of Bax-induced death Taken together,
these results show that Bax-induced mitochondrial lipid
oxidation is relevant to Bax-induced cell death Conversely, lipid oxidation is poorly related to the massive Bax-induced superoxide and hydrogen peroxide accumulation,which occurs at the same time,as chemical or enzymatic scavenging
of ROS does not prevent lipid oxidation nor has any effects
on kinetics of Bax-induced cell death Whatever the origin of mitochondrial lipid oxidation,these data show that it rep-resents a major step in the cascade of events leading to Bax-induced cell death These results are discussed in the light of the role of lipid oxidation both in mammalian apoptosis and
in other forms of cell death in other organisms
Keywords: Bax; apoptosis; ROS; lipid oxidation; yeast mitochondria
Apoptosis is a cellular death program involved in
homeo-stasis and development of tissues Numerous inducers
including growth factors deprivation,extracellular death
signal or subnecrotic chemical or physical damages activate
this program Apoptosis has been well documented in a
wide variety of cellular types and a common set of events
underlies the whole process Transcriptional activation of
pro-apoptotic genes and repression of antiapoptotic genes
change the balance ratio between pro- and antiapoptotic
proteins among which Bcl-2 family members play a key role
(reviewed in [1,2]) This group of proteins was identified on
the basis of sequence homologies of domains called BH-1–4
(Bcl-2 Homology domain) Three-dimensional structures of
three members of the family (namely Bcl-2,Bcl-xLand Bax)
show remarkable conservation although they exhibit
opposite functions [3–5] A third group of Bcl-2 family
members gathers proteins sharing sequence homology
restricted to only the BH3 domain (e.g Bid,Bad) These
proteins do not have pro-apoptotic activity by themselves but do potentiate the pro-apoptotic activity of Bax and other pro-apoptotic proteins (reviewed in [1])
The main action site of Bcl-2 family members is mito-chondria It is now widely accepted that,following apoptosis-induction,the pro-apoptotic protein Bax is translocated from the cytosol to the outer mitochondrial membrane and induces the release of several proteins localized in the intermembrane space to the cytosol,where they exert their pro-apoptotic activity The molecular mechanism underlying this process is still a matter of debate (reviewed in [2]) and may actually differ,depending on models Anti-apoptotic proteins,such as Bcl-2,prevent this release
Besides their effects on the outer mitochondrial mem-brane,Bcl-2 family members were shown to act as pro/ antioxidant proteins Apoptosis is accompanied by an oxidative burst,with an increase of the intracellular concentration of reactive oxygen species (reviewed in [6,7]) Mitochondrial respiratory chain is the main producer
of reactive oxygen species (ROS) and mitochondrial dysfunctions are known to increase the intracellular con-centration of namely superoxide and peroxide ions [8] In addition to unselective deleterious effects on cellular com-ponents,ROS may exhibit more specific roles in the apoptotic process: the intracellular redox state has been shown to modulate directly permeability transition pore opening [9],to regulate cell cycle checkpoints [10] by acting possibly on p53 transactivation [11,12] and to modulate caspase-3 activity [13–15]
In addition to ROS accumulation,other oxidative path-ways may play a role in apoptosis Namely,lipoxygenases
Correspondence to S Manon,IBGC/CNRS,
1 rue Camille Saint-Sae¨ns,F-33077 Bordeaux cedex,France.
Fax: 33 (0)5 56 99 90 51,Tel.: 33 (0)5 56 99 90 45,
E-mail: stephen.manon@ibgc.u-bordeaux2.fr
Abbreviations: C 18:1 ,oleic acid; C 18:2 ,linoleic acid; H 2 -DCFDA,
dihydro-dichlorofluorescein diacetate; DHE,dihydroethidium; ROS,
reactive oxygen species; SOD,superoxide dismutase.
*Present address: De´partement de Biologie Cellulaire,Sciences III,
Quai Ernest-Ansermet 30,CH-1211 Gene`ve 4,Switzerland
(Received 11 June 2002,revised 21 August 2002,
accepted 6 September 2002)
Trang 2activation has been identified as a possible positive
modu-lator of apoptosis [16–18] A common regulation between a
lipoxygenase-activating protein and Bcl-2 family member
Bcl-xLhas also been reported [19] As the critical function of
Bcl-2 family members occurs at the mitochondrial level,
lipoxygenase-induced oxidation of mitochondrial lipids
could also be a regulatory step of apoptosis
Yeast has already been a powerful tool to evaluate the
functions of Bcl-2 family members Heterologous Bcl-2
expression in yeast has been shown to prevent deleterious
effects caused by the inactivation of ROS-scavenging
enzymes superoxide dismutases [20] Conversely,hydrogen
peroxide-treatment of yeast induces similar phenotypes as
heterologous Bax expression [21] Bax expressed in yeast is
translocated to the outer mitochondrial membrane [22,23]
and induces cell death [24] accompanied by apoptotic
characteristics such as cytochrome c release
[25],phospha-tidylserine exposure and DNA fragmentation [21,26] and
late maintenance of plasma membrane properties [27] It has
been used as a tool to demonstrate that Bax is able to induce
these apoptotic characteristics through a process that does
not involve any transition permeability of the inner
membrane [28],nor the voltage-dependent anion channel
[28,29], or the adenine nucleotides carrier [29–31] Results
obtained with yeast support the hypothesis that Bax acts on mitochondria by creating a de novo giant channel in the outer mitochondrial membrane,independent from the voltage-dependent anion channel [23]
The availability of different mutants and the manipula-tion of metabolic condimanipula-tions prompted the use of yeast to evaluate the participation of a number of factors in Bax-induced cell death In the present report,we show that Bax expression induces lipid oxidation Manipulation of the unsaturation degree of fatty acids shows that this oxidation
is involved in Bax-induced cell death Although ROS accumulation occurs at the same time as lipid oxidation,the two phenomena are poorly related and,opposite to lipid oxidation,preventing ROS accumulation does not have any positive effect on Bax-induced cell death
E X P E R I M E N T A L P R O C E D U R E S
Strains, constructions and cultures Strains are listed in Table 1 Constructions for Bax and
Bcl-xL expression have been described previously [24,31] Briefly,human Bax gene carrying a C-terminal c-myc tag was placed under the control of a galactose-inducible
Table 1 Genotypes and phenotypes of strains used in this study.
W303–1B/50 Mat a, ade1, his3, leu2, trp1, ura3
rho 0
Wild-type rho 0
pCM189(URA3)-bax-cmyc
Bax-expression (tet-promoter)
pCM184(TRP1)-bax-cmyc
Bax-expression (tet-promoter) WtB1-rho 0 Mat a, ade1, his3, leu2, trp1, ura3
rho 0
pCM189(URA3)-bax-cmyc
Bax-expression (tet-promoter)
WtB2-SOD1 Mat a, ade1, his3, leu2, trp1, ura3
pCM184(TRP1)-bax-cmyc pYES2/GS(URA3)-SOD1-V5-his6
Bax-expression (tet-promoter) Cytosolic SOD-overexpression WtB2-SOD2 Mat a, ade1, his3, leu2, trp1, ura3
pCM184(TRP1)-bax-cmyc pYES2/GS(URA3)-SOD2-V5-his6
Bax-expression (tet-promoter) Mitochondrial SOD-overexpression WtB2-CTT1 Mat a, ade1, his3, leu2, trp1, ura3
pCM184(TRP1)-bax-cmyc pYES2/GS(URA3)-CTT1-V5-his6
Bax-expression (tet-promoter) Cytosolic Catalase-overexpression WtB2-CTA1 Mat a, ade1, his3, leu2, trp1, ura3
pCM184(TRP1)-bax-cmyc pYES2/GS(URA3)-CTA1-V5-his6
Bax-expression (tet-promoter) Perox Catalase-overexpression
sod1::URA3
Cytosolic SOD-deficient BY4742-DSOD2 Mat a, his3, leu2, lys2, ura3
sod2::kanMX4
Mitochondrial SOD-deficient
GAL1/10-bax-cmyc/LEU2
Bax-expression (Gal-promoter)
pDP83(URA3)-GAL1/10-Bcl-x L
Bcl-x L -expression (Gal-promoter)
GAL1/10-bax-cmyc/LEU2 pDP83(URA3)-GAL1/10-Bcl-x
Bax-expression (Gal-promoter) Bcl-x L -expression (Gal-promoter)
Trang 3GAL1/GAL10 promoter (strong expression conditions) or a
doxycyclin-repressed tet-off promoter (mild expression
conditions) Dcls1, Dsod2, Dctt1 and Dcta1 strains were
obtained from Euroscarf (Frankfurt,Germany) Dsod1
strain was a gift from Dr Vale´rie Prouzet
SOD1, SOD2 and CTT1 genes were amplified by PCR
using the oligonucleotides pairs 5¢-CCCCAATTGATATC
TATACCTCC-3¢ and 5¢-CTTCAGAGGTTACCAGCA
TCGA-3¢, 5¢-CAGGCAAGAAAGATATCGCGC-3¢ and
5¢-ATTAGTTGGTGACCAATGACACC-3¢, 5¢-CCTCT
ATTCCAGATATCAATCTTGT-3¢ and 5¢-CAAGTCTT
GGGTTAACCTTCAAG-3¢,respectively The fragments
contained an EcoRV and a BstEII site for further cloning
between the PvuII/BstEII sites of the pYES2/GS plasmid,in
frame with the sequences of V5 and His6 tags at the
C-terminal end of the proteins (Invitrogen,The
Nether-lands) The same construction with CTA1 gene was directly
purchased from Invitrogen The pYES2/GS plasmid allows
the inducible expression of the proteins under the control of
a GAL1/GAL10 promoter All constructions were
con-trolled by PCR and protein expression was checked by
Western blots Proteins were separated on 12.5% SDS/
PAGE and transferred onto poly(vinylidene difluoride)
(PVDF) membranes (Problott,Perkin-Elmer) The
mem-branes were washed in NaCl/Pi(10 mMsodium phosphate
pH 7.2,137 mM NaCl) containing 0.1% Tween-20 and
saturated with 5% milk powder in NaCl/Pi/Tween-20
Primary antihuman Bax rabbit polyclonal antibody N20
(Santa-Cruz,USA) and anti-V5 tag mouse monoclonal
antibody (Invitrogen,The Netherlands) were used at
1 : 2000 and 1 : 5000 dilutions,respectively,and secondary
goat anti-rabbit IgG and goat anti-mouse IgG antibodies
coupled to horse radish peroxidase (Jackson Laboratories,
USA) were used at a 1 : 5000 dilution,and peroxidase
activity was revealed by an Enhanced
Chemio-Lumines-cence kit (Amersham,UK)
Cells were grown in a synthetic medium Yeast Nitrogen
Base (Difco,USA) supplemented with 0.1% KH2PO4,
0.5% (NH4)2SO4and a mixture of all aminoacids except
tryptophan,plus adenine (Drop-Mix) The carbon source
was 2% lactate,giving a strict respiratory metabolism,
except for the Dsod1 strain which is unable to grow on
lactate,and for which 2% glucose was used Plasmids used
throughout the study carry URA3 or TRP1 genes and uracil
and tryptophan are thus used as selection markers For cells
transformed by a tet-off/h-Bax construct,10 lgÆmL)1
doxycyclin was added to prevent Bax expression under
repressive conditions Cells were grown aerobically in
Erlenmeyers (air volume/medium volume¼ 5),at 28 C
under rotary shaking (300 r.p.m.) For experiments with
dioctylphtalate,a homemade minimal medium was used,
according to [32],and dioctylphtalate dissolved in ethanol
was added at a final concentration of 100 lM
Bax induction was achieved by adding 1% galactose to
lactate-supplemented cultures (GAL1/GAL10 promoter),or
by washing the cells three times in water to remove
doxycyclin (tet-off promoter)
Spheroplasts and mitochondria preparation
Spheroplasts and mitochondria were isolated from
zymol-yase-treated cells as described previously [33,34] Protein
concentration was measured using the biuret method
Bax-induced lethality Strains were grown under repressive conditions and then transferred under inductive conditions At different times, aliquots of 200 cells were plated on YPD medium (1% Yeast Extract,1% Bacto-Peptone,2% glucose) added with doxycyclin (10 lgÆmL)1) The number of growing colonies was counted after a 3-day-incubation at 28C
Lipids extraction and fatty acids analysis Spheroplasts or mitochondria were suspended in methanol
at 10 mg proteinsÆmL)1and lipids were extracted twice by a chloroform/methanol/water (2 : 1 : 1,v/v/v) mixture The organic phases were pooled,evaporated to dryness and resolubilized in 50 lL chloroform/methanol (2 : 1,v/v) Aliquots were mixed with 900 lL methanol containing 2.5% H2SO4 and 10 lg C17:0 as an external standard Transesterification was carried out at 80C for 1 h and, after cooling on ice,methyl ester fatty acids were extracted with 1 mL hexane Separation of methyl ester fatty acids was perfomed by gas chromatography (Hewlett-Packard
5890 series II) on a 15-m· 0.53 mm Carbowax column (Alltech,USA) The oven temperature is programmed for
1 min at 160C,followed by a 20 C min)1ramp up to
190C,a 5 C min)1ramp up to 210C and a final 5 min
at 210C Methyl ester fatty acids were identified and quantified by comparison of their retention time with those
of standards
Polar lipids were resolved by one-dimensional thin layer chromatography on 10· 10 cm silica gel plates (Merck, Germany) using the solvent system as described [35] Lipids were located by spraying the plates with a solution of 0.001% primuline in 80% acetone and vizualized under UV light The silica gel zones corresponding to individual lipids were scrapped and methyl ester fatty acids were prepared as above
Measurements of ROS and lipid oxidation Aliquots of cells (5· 106cellsÆmL)1) were incubated in the dark at 28C in the presence of 20 lM dihydroethidium (DHE) (Sigma,USA) Following oxidation by ROS, dihydroethidium is converted to ethidium,which binds to DNA with a fluorescence increase at 565 nm (excitation at
495 nm) Conditions were adjusted on a strain inactivated for SOD1 gene,thus producing high amounts of ROS
F565nmincreases for 45 min,reaches a plateau and eventu-ally decreases,probably corresponding to ethidium-induced death All further experiments were carried out after a 45-minute incubation
Aliquots of cells (5· 106cellsÆmL)1) were incubated in the dark at 28C in the presence of 100 lMH2-DCFDA (dihydro-dichlorofluorescein diacetate) (Molecular Probes, The Netherlands) This nonfluorescent compound enters the cell and is converted to highly fluorescent and nonper-meant DCF by esterases and chemical oxidation by ROS
F525nm(excitation at 495 nm) increases for 20 min and then reaches a plateau All further experiments were carried out after a 45-minute incubation
Aliquots of mitochondria (1 mg proteinÆmL)1) were incubated in the dark at 4C in the presence of 10 lM cis-parinaric acid (Molecular Probes,The Netherlands)
Trang 4added as a complex with BSA (0.5 mg of cis-parinaric acid
dissolved in 25 lL dimethylsulfoxide were added to 50 mg
BSA dissolved in 1 mL of a 10-mM sodium phosphate
buffer (pH 7.2) containing 137 mMNaCl) These conditions
allow the integration of the probe into lipid bilayer [36]
After a 2-h incubation,measurements were carried out at
28C (excitation at 324 nm,emission at 420 nm) Ethanol
(20 mM) was added as a respiratory substrate
Fluorescence measurements were performed using a
Xenius spectrofluorometer (Safas,Monaco)
The extent of lipid oxidation was measured according to
[37] One millilitre of cell suspension (5· 107 cells) was
harvested,washed and added to an equal volume of a
mixture of 15% trichloroacetic acid,0.375% thiobarbituric
acid in 0.25MHCl After a 15-minute incubation at 90C,
samples were centrifuged and the absorbance of the
supernatant was measured at 535 nm and the amounts of
thiobarbituric acid reactive products were determined using
a standard curve built with 1,1,3,3-tetramethoxypropane
solutions treated under the same conditions
Superoxide dismutase and catalase activities
Cells (100 mL of a culture at A550nm¼ 4) were washed
twice and then broken by vigorous shaking for 4 min with
an equal volume of glass beads in a 10-mM potassium
phosphate buffer (pH 7.2) containing 0.6Mmannitol and a
mixture of proteases inhibitors (Complete EDTA-freeTM,
Boehringer,Germany) Centrifugation (900 g,10 min)
allows the elimination of pelleted unbroken cells and nuclei
A second centrifugation (12 000 g,15 min) allows the
recovery of the pellet containing most mitochondria and
peroxisomes,containing Sod2p and Cta1p A third
centri-fugation (105 000 g,30 min) allows the elimination of the
microsomal pellet and the recovery of the cytosolic
super-natant containing Sod1p and Ctt1p
Superoxide dismutase (SOD) was measured by a reverse
titration method in a double-beam/double wavelength
spectrophotometer in dual mode or directly after separating
proteins by nondenaturating polyacrylamide gel
electro-phoresis
The reaction mixture for SOD activity was a 50-mM
sodium carbonate buffer (pH 10.2) containing 0.01%
Triton X-100,0.1 mM EDTA,0.1 mM xanthine,25 lM
nitro-blue tetrazolium and the protein extract or bovine
erythrocytes SOD (Boehringer,Germany) The reaction
was started by adding buttermilk xanthine oxidase (Sigma,
USA) and nitro-blue tetrazolium oxidation was followed as
the absorbance difference between 560 and 500 nm The
amount of xanthine oxidase was chosen so that the variation
in the absorbance difference in the absence of SOD activity
is approximately 2· 10)2absorbance unitsÆmin)1and may
change depending on commercial preparations Under these
conditions,the variation of absorbance difference is linear
for at least one minute The addition of SOD prevents the
oxidation of nitro-blue tetrazolium by superoxide ions
produced in the xanthine oxidase reaction A titration curve
of the activity was first built with commercial bovine
erythrocytes SOD so that the activity of SOD in proteic
extract could be calculated
SOD activity was also qualitatively estimated after
separ-ating the proteins by nondenatursepar-ating polyacrylamide gel
electrophoresis The gel (0.75 mm thick,10% acrylamide)
was soaked for 5 min under low-light conditions in a 36-mM potassium phosphate buffer (pH 7.8) containing 1.5 mgÆmL)1nitro-blue tetrazolium and a further 10 min in the same buffer containing 28 mM TEMED and 28 lM riboflavine After a brief washing in water,the gel was revealed on a light table
Catalase activity was measured on the same proteic extracts Proteins were suspended in a 10-mM potassium phosphate buffer (pH 7.2) containing 0.01% Triton X-100 and dispatched in the cuvettes of a double-beam/double wavelength spectrophotometer in Split mode A concentra-tion range of hydrogen peroxide (1–50 mM) was added in the sample cuvette and the absorbance difference between the cuvettes was measured at 240 nm Titration curves with bovine heart catalase (Sigma,USA) were built to calculate specific activity in the extracts All spectrophotometric measurements were performed with a DW2000 double beam/double wavelength spectrophotometer (Aminco, USA)
R E S U L T S
Bax-expression induces alterations of mitochondrial lipids
In yeast,as in mammalian cells,unsaturated fatty acids of phospholipids are primary targets during oxidative stress (reviewed in [38]) Also,the unsaturation index of fatty acids
is strongly dependent on culture conditions [39] and was shown to modulate several stress responses,such as heavy metals-induced stress [40] We thus investigated a possible involvement of lipid oxidation in Bax-induced cell death Phospholipids were extracted from both whole cells and isolated mitochondria and fatty acid amount and compo-sition were measured When grown on a standard medium, yeast cells only contain monounsaturated fatty acids and not polyunsaturated fatty acids Hydroperoxidation of monounsaturated fatty acids such as C18:1 results in the rupture of the hydrocarbon chain,leading to the appearance
of lower molecular weight compounds (C8to C11) including alkanes,alkenes and aldehydes [41] We thus expected that lipid oxidation would result in the disappearance of monounsaturated fatty acids without an increase in satur-ated fatty acids
The total amount of fatty acids in whole cells was not significantly affected (data not shown) but the amount of fatty acids in isolated mitochondria was markedly decreased
by Bax expression (Fig 1A) Also,a strong decrease of the unsaturated/saturated fatty acids ratio was measured on isolated mitochondria but not on whole cells (Fig 1B) This effect is time-dependent,reaching a maximum after 4 h,in correlation to Bax-induced cell death kinetics ([25] and Fig 9) These observations suggest that Bax-expression is accompanied by mitochondrial lipid oxidation
Co-expression of antiapoptotic proteins of the Bcl-2 family,such as Bcl-xL,was shown to prevent Bax-induced release of cytochrome c and cell death [25] Expression of Bcl-xLalone had a slight increasing effect on mitochondrial fatty acids unsaturation ratio and,most importantly,fully prevented the effect of Bax (Fig 2)
To see whether a particular class of phospholipid was affected,they were separated by thin layer chromatography and individual fatty acids amount and composition were
Trang 5determined It appeared that the alteration of fatty acids
amount (Fig 3A) and unsaturation index (Fig 3B) affected
all lipids,but was slightly more marked for cardiolipin
Although Bax-expression in yeast results in a localization in
the outer mitochondrial membrane and not in the inner
mitochondrial membrane [23],the fact that inner membrane
cardiolipin is altered may suggest that Bax,as in
mamma-lian mitochondria,localizes preferably at the contact sites
between both membranes (reviewed in [42]),where
cardio-lipin plays a crucial role (reviewed in [43])
To investigate if the alteration of mitochondrial
unsatur-ated fatty acids was actually linked to an oxidation,we first
used the probe cis-parinaric acid This fluorescent
poly-unsaturated fatty acid can be integrated in vivo or in vitro
into biological membranes,where it supports the same
alterations as native lipids [43–46] Namely,peroxidation of
this probe induces a decrease of its emission fluorescence
Fig 1 Effect of Bax-expression on mitochondrial fatty acids amounts
and unsaturation index HT444/Bi strain was grown in a
lactate-sup-plemented medium and then added with 1% galactose to achieve
Bax-expression Cells were harvested and spheroplasts and mitochondria
were isolated Lipids were extracted and analyzed as indicated in the
methods section (A) Quantification of the different fatty acids in
mitochondria isolated from control (hatched bars) and 4-h
Bax-expressing (white bars) cells (B) Evolution of the unsaturation index
following Bax-expression of fatty acids in whole spheroplasts (h) and
isolated mitochondria (n) Data are representative of three
inde-pendent experiments.
Fig 2 Effect of Bcl-x L -expression on mitochondrial fatty acids unsat-uration index Strains HT444 (control),HT444/Bi (Bax),HMP1
(Bcl-x L ) and MP1 (Bax/Bcl-x L ) were grown in a lactate-supplemented medium and then added with 1% galactose to achieve Bax and/or Bcl-x L expressions After 4 h,cells were harvested and mitochondrial lipids were analyzed as in Fig 1.
Fig 3 Effect of Bax expression on fatty acid amounts and unsaturation index in the different classes of mitochondrial phospholipid Growth and expression conditions (4 h) are similar to Fig 1 Phospholipids were extracted from isolated mitochondria and fatty acids were analyzed as indicated in the methods section (A) Fatty acid amounts in the dif-ferent phospholipids extracted from mitochondria of control (hatched bars) and Bax-expressing (white bars) cells (B) Unsaturation index of fatty acids in the different phospholipids extracted from mitochondria
of control (hatched bars) and Bax-expressing (white bars) cells Data are representative of two independent experiments.
Trang 6intensity Isolated mitochondria were incubated in the
presence of cis-parinaric acid to allow its integration in
membranes [36] A respiratory substrate (ethanol) was then
added and the fluorescence of cis-parinaric acid was
monitored A slow decrease of fluorescence following
ethanol-driven respiration was observed in wild-type
mito-chondria (Fig 4),showing that normal respiratory chain
activity only induces a marginal oxidation of lipids The rate
of this decrease was strongly accelerated in mitochondria
isolated from Bax-expressing cells (Fig 4) This supports
the hypothesis that Bax strongly sensitizes mitochondrial
lipids to respiration-induced oxidation
The amount of thiobarbituric-acid reactive species is an
indication of the amount of lipid peroxidation products [37]
As shown by Riely et al [41],the oxidation of
monoun-saturated fatty acids does not normally lead to the formation
of thiobarbituric acid-reactive products and no formation of
such products could be observed in Bax-expressing cells
(Fig 5) The amount of polyunsaturated fatty acids can be
increased by the addition of C18:2,which represses the
expression of D9-acyl-coenzyme A desaturase Ole1p,thus
allowing the incorporation of exogenous fatty acids in
phospholipids [47] Under these conditions,Bax-expression
resulted in a strong increase in the amount of thiobarbituric acid-reactive products (Fig 5),supporting further the
hypothesis that Bax actually induces lipid peroxidation
Mitochondrial lipid oxidation is involved in cell death kinetics
Chatterjee et al [32] showed that the unsaturation index of yeast fatty acids could be artificially manipulated by adding dioctylphtalate to the cultures Commercial Yeast Nitrogen Base from Difco contains this contaminating product that induces a high degree of unsaturation Bax-expressing cells were therefore grown in a homemade minimal medium containing or not dioctylphtalate We first verified that the addition of dioctylphtalate did not have significant effect on cells growth in lactate-supplemented medium (doubling times of 290 ± 20 and 280 ± 20 min (n¼ 5) in the absence and in the presence of 100 lM dioctylphtalate, respectively) as already shown by Chatterjee et al [32] for growth in glucose-supplemented medium In the absence of Bax expression,dioctylphtalate did not induce any cell death nor cytochrome c release (data not shown) From these data,dioctylphtalate had no obvious effects on mitochondrial metabolism or on cells viability and thus could be assayed on Bax effects
The effect of dioctylphtalate on mitochondrial lipid oxidation was measured Chatterjee et al [32] reported that,on glucose-grown cells,dioctylphtalate induced an increase of unsaturation index of mitochondrial lipids from 0.75 to 3.30 We found a similar increase on lactate-grown cells (from 1.54 ± 0.02–6.85 ± 0.06; n¼ 3),reaching a value close to that measured in Yeast Nitrogen Base medium (7.05 ± 0.05; n¼ 3) Kinetics of Bax-induced cell death varied accordingly to the extent of Bax-induced mitochondrial lipid oxidation In the minimal medium without dioctylphtalate,Bax-induced cell death kinetics was much slower than in Yeast Nitrogen Base medium (Fig 6) The addition of dioctylphtalate restored cell death kinetics comparable to experiments in Yeast Nitrogen Base medium, supporting a role of lipids unsaturation index,and thus oxidation sensitivity,in Bax-induced cell death
Fig 4 Respiration-induced cis-parinaric acid oxidation in isolated
mitochondria Mitochondria were isolated from control cells and
GAL1/GAL10-driven Bax-expressing cells cis-Parinaric acid was
incorporated in mitochondrial membranes,and fluorescence
meas-urements were carried out as indicated in the methods section (A)
Time-course of a typical experiment (B) Variation of fluorescence
during the first minute after ethanol addition: average from four
dif-ferent mitochondria preparations for each strain ± SD.
Fig 5 Amount of thiobarbituric acid-reactive species HT444/Bi strain grown in lactate-supplemented medium in the absence or in the pres-ence of 100 l M C 18:2 until mid-exponential growth phase Bax-induc-tion was achieved by adding 1% galactose After 2 h,cells were harvested,washed and resuspended in water The amount of thio-barbituric acid-reactive products was determined as described in the methods section.
Trang 7The amount of unsaturated fatty acids was also increased
by adding C18:1 or C18:2,which repress the expression of
D9-acyl-coenzyme A desaturase Ole1p,thus allowing the
incorporation of exogenous fatty acids in phospholipids [47]
Bax expression was achieved in the presence of these fatty
acids (Fig 7A): kinetics of cell death significantly increased
in the presence of C18:2as compared to C18:1and control
The effects of known inhibitors of lipid oxidation,namely
a-tocopherol and resveratrol were also assayed: both
compounds significantly decreased kinetics of Bax-induced
cell death (Fig 7B) Taken together,these results strongly
support the hypothesis that Bax-induced lipid oxidation is
directly related to Bax-killing efficiency
Bax-induced lipid peroxidation is poorly related
to Bax-induced ROS production
Madeo et al [21] previously reported that mouse Bax
expression in yeast induced a production of ROS,measured
with fluorescent probes DHE and H2-DCFDA Similar
results were observed with human Bax expression under the
control of the strong promoter GAL1/10 (Fig 8A)
How-ever,under these conditions,yeast supports a massive
cytochrome c relocalization [25],which is likely to interfere
with the response of these probes,namely H2-DCFDA,to
ROS [48] The experiments were therefore reproduced in
yeast cells expressing Bax under the control of the
low-strength promoter tet-off,under conditions where no
massive cytochrome c relocalization is observed [31]
Interestingly,with a system allowing the expression of
Bax under the control of a low-strength promoter,a
difference was observed in the response of the two probes:
no oxidation of DHE could be observed under this
condition (Fig 8B) Although both probes are oxidizable
by any type of ROS,it has been reported that DHE is more
sensitive to superoxide ion whereas H-DCFDA is more
sensitive to hydrogen peroxide [49] The difference between the responses of the two probes to Bax-expression may indicate that hydrogen peroxide is accumulated more dramatically than superoxide ion The response of DHE under conditions where cytochrome c is massively relocal-ized to the cytosol probably reflects a secondary accumu-lation of superoxide ion following the inhibition of mitochondrial electron transfer after the Bc1 complex [25] Measurements of ROS-scavenging activities superoxide dismutases and catalases were performed on whole extracts from cells expressing Bax under the control of GAL1/10 promoter Superoxide dismutase activity was decreased by less than 10% (± 5% SD; n¼ 4) and catalase activity was decreased by 55% (± 20% SD, n¼ 4) in Bax-expressing cells This confirms that the accumulation of superoxide ion depicted when Bax is strongly expressed does not result
Fig 7 Effect of unsaturated fatty acids and of inhibitors of lipids oxi-dation on Bax-induced cell death (A) Wtb1 cells were grown in YNB medium supplemented with lactate in the presence of doxycyclin,and added or not with 100 l M C 18:1 or 100 l M C 18:2 At t ¼ 0,cells were washed and resuspended in the same medium with doxycyclin After 2
or 4 h,the number of colony-forming cells was determined as indicated
in the methods section (B) Wtb1 cells were grown in YNB medium supplemented with lactate in the presence of doxycyclin At t ¼ 0,cells were washed and resuspended in the same medium in the absence or in the presence of 200 l M a-tocopherol or 100 l M resveratrol After 6 or
14 h,the number of colony-forming cells was determined as indicated
in the methods section.
Fig 6 Effect of fatty acid-unsaturation-inducer dioctylphtalate on
h-Bax-induced cell death kinetics WtB1 cells were grown in a
com-mercial Yeast Nitrogen Base medium (Difco, h),or in a home-made
minimal medium in the absence (j) or in the presence (.) of 100 l M
dioctylphtalate,all three supplemented with lactate as a carbon source
and in the presence of doxycyclin (Bax-repression) At t ¼ 0,cells were
washed and transferred in the same media in the absence of doxycyclin
(Bax-expression) The number of colony-forming cells was determined
as indicated in the methods section.
Trang 8from an inhibition of superoxide dismutase but,more likely,
from the alteration of the redox state of quinones following
the release of cytochrome c
Is ROS accumulation responsible for lipid oxidation? To
answer this question,ROS accumulation was scavenged by
adding the chemical reducer Tiron to the culture medium:
under this condition,oxidation of H2-DCFDA disappeared
(Fig 9A),but lipid oxidation still occured (Fig 9B)
sug-gesting that lipid oxidation is not a direct consequence of
ROS accumulation Consequently,Tiron did not have any
protecting effect on Bax-induced cell death kinetics
(Fig 9C,D) Identical results were obtained by using
N-acetylcysteine instead of Tiron (data not shown)
In order to confirm this lack of protection of ROS scavenging against Bax effects,individual overexpressions
of the cytosolic form of yeast superoxide dismutase (Sod1p) and the two yeast catalases (Cta1p and Ctt1p) were achieved Overexpression of the mitochondrial form of superoxide dismutase (Sod2p) led to a dramatic decrease of cell growth,even in the absence of Bax (data not shown) The other enzymes were overexpressed under the form
of hexahistidine/V5-C-terminal-tagged active proteins (Fig 10A,B) Overexpression of any of the three enzymes did not have any positive effect on Bax-induced cell death kinetics (Fig 10C,D)
D I S C U S S I O N
Data presented in this paper show that heterologous expression of Bax alone activates lipid oxidation,and that preventing lipid oxidation has a significant positive effect on Bax-induced cell death Although an indirect effect of ROS cannot be completely ruled out,it is noteworthy that direct modulation of lipid oxidation has marked effects on the kinetics of Bax-induced cell death,while modulation of ROS accumulation has not
Bax-induced ROS-accumulation Numerous reports have demonstrated the occurrence of oxidative stress as a side effect in apoptosis As mitochon-dria are the central effector in apoptosis,it has been suggested that alterations of the respiratory chain may be responsible for a dramatic increase of intracellular ROS concentrations,thus having deleterious effects on biological constituents Release of cytochrome c obviously leads to an increase of the reduced state of ubiquinone,thus favoring the reduction of molecular oxygen to superoxide ion by the Bc1 complex [50,51] Some reports showed that ROS scavenging induced through overproduction of the mitoch-ondrial isoform of SOD,might counteract apoptosis,at least partially [52,53] Conversely, mitochondrial SOD deficiency induces apoptosis [54]
Yeast has been a useful tool to test these hypotheses Madeo et al have observed phenotypic similarities between Bax-expressing yeast cells and H2O2-treated cells [21] and that a caspase-like activity is involved in yeast response to
H2O2-treatment [55] It has been reported that plant antimicrobial protein osmotin induced a RAS2-dependent
apoptosis-like stress response in yeast,prevented by chemical reducer N-acetylcysteine [56] Longo et al [20] have shown that overexpression of antiapoptotic protein Bcl-2 protected yeast cells against oxidative stress induced
by SOD1 and SOD2 inactivation: the antioxidant function
of Bcl-2,already shown in mammalian cells [57,58],could therefore also be depicted in a heterologous model Mitochondrial ROS production also appears to be involved
in yeast ageing [59]
Bax-induced lipid oxidation The present study evidences an additional oxidant function for Bax on mitochondrial lipids This effect is prevented by the coexpression of antiapoptotic Bcl-xL,suggesting that it
is relevant to an actual function of Bax in apoptosis Different treatments and conditions allowed modulating the
Fig 8 Measurements of ROS production by Bax-expressing cells.
HT444/Bi (A) or WtB1 (B) were grown aerobically under
Bax-repressive conditions in lactate-supplemented medium (see methods).
At t ¼ 0,cells were transferred under Bax-inductive conditions At the
indicated times,an aliquot of 1 mL was taken out,diluted in the same
medium to 5 · 10 6 cellsÆmL)1,and incubated with 20 l M
dihydro-ethidium (h) or 100 l M H 2 -DCFDA (n) in the dark for 45 min.
Emission spectra were acquired between 500 and 600 nm (excitation at
495 nm) for both fluorochromes Y-axis is the ratio of the maximal
emission intensity (at 565 nm for ethidium/DNA complex and 525 nm
for DCF) at time t over the value at t ¼ 0,thus representing the
increase ratio of fluorescence associated to Bax expression Data are
representative of five experiments for each curve.
Trang 9unsaturation ratio of fatty acids and,as
expected,Bax-killing effect was modulated by these treatments In
addition,inhibitors of lipid oxidation,actually slowed down
Bax-induced cell death
Numerous reports have evidenced a role for fatty
acid oxidation in apoptosis (recently reviewed in [60])
Although it may occur as a final consequence of ROS
production [61],enzymatic peroxidation by 5-lipoxygenase and cyclooxygenase has often been demonstrated (reviewed
in [17,61]) enlightning the role of polyunsaturated arachi-donic acid
The first crucial observation reported in the present paper
is that,in the absence of a mammalian apoptotic network, Bax alone is able to activate lipid oxidation The origin of
Fig 10 Effect of overexpression of scavenging enzymes on Bax-effects (A) Control of the overexpression of active Sod1p-V5-his6: (1) Dsod1 strain; (2) Dsod2 strain; (3) Wtb2-SOD1 grown in glucose + doxycyclin; (4) Wtb2-SOD1 grown in galactose + doxycyclin; (5) Wtb2-SOD1 grown in glucose ) doxycyclin; (6) Wtb2-SOD1 grown in galactose ) doxycyclin (A) SOD activities revealed on nondenaturing polyacrylamide gel; (B) Western-blots with an anti-Bax antibody: (C) Western-blots with an anti-V5-tag antibody (B) Control of the overexpression of active Cta1p-V5-his6 and Ctt1p-V5-Cta1p-V5-his6: Wtb2-CTA1 or Wtb2-CTT1 were grown on YNB medium supplemented with glucose or galactose,as indicated,and in the presence of doxycyclin Cells were then washed and transferred in the same medium in the absence of doxycyclin Catalase activity was measured in cells extracts after 12 h (C) Effect of overexpression of Sod1p on Bax-induced cell-death Wtb2 (j) or Wtb2-SOD1 (m) were grown in YNB supplemented with galactose in the presence of doxycyclin At t ¼ 0,cells were washed and transferred in the same medium in the absence of doxycyclin At the indicated times,aliquots of 200 cells were plated on YPD + doxycyclin and the number of growing colonies was counted after
48 h (D) Effect of overexpression of Cta1p or Ctt1p on Bax-induced cell-death Wtb2 (j),Wtb2-CTA1 (.) or Wtb2-CTT1 (r) survival ratio was measured as in (C).
Fig 9 Effects of reducer Tiron on Bax-effects (A) ROS production by HT444/Bi following Bax expression was measured as in Fig 8 in the absence (black bars) or in the presence (white bars) of 5 m M Tiron (B) Unsaturation ratio of mitochondrial fatty acids was measured as in Fig 1 on HT444/Bi following or not Bax expression,
in the absence or in the presence of 5 m M
Tiron (C) Survival ratio of HT444/Bi following Bax expression in the absence (h)
or in the presence (s) of 5 m M Tiron (D) Survival ratio of WtB1 following Bax expression in the absence (h) or in the presence (s) of 5 m M Tiron.
Trang 10this lipid oxidation can be discussed Chemical reducers
Tiron and N-acetylcysteine scavenged ROS but did not
prevent lipid oxidation,suggesting that the two effects are
independent However,one cannot rule out the possibility
that a low concentration of ROS produced even in the
presence of those reducers is still enough to induce
mitochondrial lipid oxidation
The second crucial observation is that modulation of
Bax-induced lipid oxidation,by acting on the unsaturation
ratio of fatty acids or with inhibitors of oxidation such as
resveratrol,has significant consequences on the rate of
Bax-induced yeast cell death Conversely,chemical or enzymatic
scavenging of ROS did not have any positive effect on
Bax-induced cell death From these results,whatever the primary
origin of mitochondrial lipid oxidation (activation of
lipid-oxidizing enzymes or secondary consequence of ROS
production),it is an important step in the cascade of events
leading to Bax-induced cell death
It should be noted that a similar role of lipid oxidation
was drawn for the reaction of programmed cell death
involved in plant resistance to pathogens and was related to
the presence of a selective lipoxygenase [62]
A role for enzymes catalyzing lipid oxidation might
therefore be a general process,not only in apoptosis,but
also in other forms of programmed cell death,including in
yeast [63]
A C K N O W L E D G E M E N T S
This work was supported by grants from the Centre National de la
Recherche Scientifique,the Association pour la Recherche contre le
Cancer,the Conseil Re´gional d’Aquitaine and the Universite´ de
Bordeaux 2 and a fellowship from the Ministe`re de la Recherche et de la
Technologie (to M.P.).
R E F E R E N C E S
1 Gross,A.,McDonnell,J.M & Korsmeyer,S.J (1999) BCL-2
family members and the mitochondria in apoptosis Genes Dev 13,
1899–1911.
2 Antonsson,B & Martinou,J.C (2000) The Bcl-2 protein family.
Exp Cell Res 256,50–57.
3 Muchmore,S.W.,Sattler,M.,Liang,H.,Meadows,R.P.,Harlan,
J.E.,Yoon,H.S.,Nettesheim,D.G.,Chang,B.S.,Thompson,
C.B.,Wong,S.L.,Ng,S.L & Fesik,F.W (1996) X-ray and NMR
structure of human Bcl-xL,an inhibitor of programmed cell death.
Nature 381,335–341.
4 Suzuki,M.,Youle,R.J & Tjandra,N (2000) Structure of Bax:
coregulation of dimer formation and intracellular localization.
Cell 103,645–654.
5 Petros,A.M.,Medek,A.,Nettesheim,D.G.,Kim,D.H.,Yoon,
H.S.,Swift,K.,Matayoshi,E.D.,Oltersdorf,T & Fesik,F.W.
(2001) Solution structure of the antiapoptotic protein bcl-2 Proc.
Natl Acad Sci USA 98,3012–3017.
6 Cai,J & Jones,D.P (1999) Mitochondrial redox signaling during
apoptosis J Bioenerg Biomemb 31,327–334.
7 Simon,H.U.,Haj-Heyia,A & Levi-Schaffer,F (2000) Role of
reactive oxygen species (ROS) in apoptosis induction Apoptosis 5,
415–418.
8 Esposito,L.A.,Melov,S.,Panov,A.,Cottrell,B.A & Wallace,
D.C (1999) Mitochondrial disease in mouse results in increased
oxidative stress Proc Natl Acad Sci USA 96,4820–4825.
9 Scorrano,L.,Petronilli,V & Bernardi,P (1997) On the voltage
dependence of the mitochondrial permeability transition pore A
critical appraisal J Biol Chem 272,12295–12299.
10 Shackelford,R.E.,Kaufmann,W.K & Paules,R.S (2000) Oxi-dative stress and cell cycle checkpoint function Free Radic Biol Med 28,1387–1404.
11 Li,P.F.,Dietz,R & Von Harsdorf,R (1999) p53 regulates mitochondrial membrane potential through reactive oxygen spe-cies and induces cytochrome c-independent apoptosis blocked by Bcl-2 EMBO J 18,6027–6036.
12 Huang,C.,Zhang,Z.,Ding,M.,Li,J.YeJ.,Leonard,S.S.,Shen, H.M.,Butterworth,L.,Lu,Y.,Costa,M.,Rojanasakul,Y., Castranova,V.,Vallyathan,V & Shi,X (2000) Vanadate induces p53 transactivation through hydrogen peroxide and causes apoptosis J Biol Chem 275,32516–32522.
13 Baker,A.,Santos,B.D & Powis,G (2000) Redox control of caspase-3 activity by thioredoxin and other reduced proteins Biochem Biophys Res Commun 268,78–81.
14 Park,H.S.,Huh,S.H.,Kim,Y.,Shim,J.,Lee,S.H.,Park,I.S., Jung,Y.K.,Kim,I.Y & Choi,E.J (2000) Selenite negatively regulates caspase-3 through a redox mechanism J Biol Chem 275,8487–8491.
15 Marini,M.,Frabetti,F.,Canaider,S.,Dini,L.,Falcieri,E & Poirier,G.G (2001) Modulation of caspase-3 activity by zinc ions and by the cell redox state Exp Cell Res 266,323–332.
16 Maccarrone,M.,Ranalli,M.,Bellincampi,L.,Salucci,M.L., Sabatini,S.,Melino,G & Finazzi-Agro,A (2000) Activation of different lipoxygenase isozymes induces apoptosis in Human Erythroleukemia and Neuroblastoma cells Biochem Biophys Res Commun 272,345–350.
17 Maccarrone,M.,Melino,G & Finazzi-Agro,A (2001) Lipox-ygenases and their involvement in programmed cell death.Cell Death Differ 8,776–784.
18 Gu,J.,Liu,Y.,Wen,Y.,Natarajan,R.,Lanting,L & Nadler,J.L (2001) Evidence that increased 12-lipoxygenase activity induces apoptosis in fibroblasts J Cell Physiol 186,357–365.
19 Biswal,S.S.,Datta,K & Kehrer,J.P (2001) Association between bcl-x(L) and 5-lipoxygenase activating protein (FLAP) levels in IL-3-dependent FL5.12 cells Toxicology 160,97–103.
20 Longo,V.D.,Ellerby,L.M.,Bredesen,D.E.,Valentine,J.S & Gralla,E.B (1997) Human Bcl-2 reverses survival defects in yeast lacking superoxide dismutase and delays death of wild-type yeast.
J Cell Biol 137,1581–1588.
21 Madeo,F.,Fro¨hlich, E., Ligr, M., Grey, M., Sigrist, S.J., Wolf, D.H & Fro¨hlich,K.U (1999) Oxygen stress: a regulator of apoptosis in yeast J Cell Biol 145,747–757.
22 Priault,M.,Camougrand,N.,Chaudhuri,B & Manon,S (1999) Role of the C-terminal domain of Bax and Bcl-XL in their loca-lization and function in yeast cells FEBS Lett 443,225–228.
23 Pavlov,E.V.,Priault,M.,Pietkiewicz,D.,Cheng,E.H.Y.,An-tonsson,B.,Manon,S.,Korsmeyer,S.J.,Mannella,C.A & Kinnally,K.W (2001) A novel,high conductance channel of mitochondria linked to apoptosis in mammalian cells and Bax expression in yeast J Cell Biol 155,725–732.
24 Greenhalf,W.,Stephan,C & Chaudhuri,B (1996) Role of mitochondria and C-terminal membrane anchor of Bcl-2 in Bax induced growth arrest and mortality in Saccharomyces cerevisiae FEBS Lett 380,169–175.
25 Manon,S.,Chaudhuri,B & Gue´rin,M (1997) Release of cyto-chrome c and decrease of cytocyto-chrome c oxidase in Bax-expressing yeast cells,and prevention of these effects by coexpression of Bcl-xL FEBS Lett 415,29–32.
26 Ligr,M.,Madeo,F.,Fro¨hlich,E.,Hilt,W.,Fro¨hlich,K.U & Wolf,D.H (1998) Mammalian Bax triggers apoptotic changes in yeast.FEBS Lett 438,61–65.
27 Marza,E.,Camougrand,N & Manon,S (2002) Bax expression protects yeast plasma membrane against ethanol-induced per-meabilization FEBS Lett 521,47–52.
28 Priault,M.,Chaudhuri,B.,Clow,A.,Camougrand,N & Manon,
S (1999) Investigation of bax-induced release of cytochrome c