Biochemical and Biophysical Research Communications 307 (2003) 477–482

www.elsevier.com/locate/ybbrc

BBRC

Anoxia–reoxygenation-induced cytochrome c and cardiolipinrelease from rat brain mitochondria

Christophe Morin,* Roland Zini, and Jean-Paul Tillement

D�eepartement de Pharmacologie, Facult�ee de M�eedecine de Paris XII, 8 rue du G�een�eeral Sarrail, F-94010 Cr�eeteil, France

Received 5 June 2003

Abstract

Rat brain mitochondria were successively submitted to anoxia and reoxygenation. The main mitochondrial functions were as-

sessed at different reoxygenation times. Although the respiratory control ratio decreased, the activity for each one of the enzymes

participating in the respiratory chain was not affected. However, during reoxygenation, mitochondrial membrane lipoperoxidation

quickly increased and was proportional to the decrease seen in membrane fluidity. Under the same conditions, cytochrome c and

cardiolipin were released from mitochondria and their rate of release increased with reoxygenation time. The release of cytochrome c

and cardiolipin was followed by the collapse of the membrane potential and it was not inhibited by cyclosporin A. Addition of the

antioxidant a-tocopherol abolished all these reoxygenation-induced changes. These data indicate that, in this model, reoxygenationpromotes the uncoupling of respiratory chain, and cytochrome c and cardiolipin releases. These events are not related to the

membrane potential collapse but to an oxidative stress.

� 2003 Elsevier Inc. All rights reserved.

Keywords: Brain mitochondria; Anoxia–reoxygenation; Respiratory control ratio; Cytochrome c; Cardiolipin; Membrane fluidity

Anoxia induces cellular damages promoted by the

defect of oxygen, glucose, and other nutriments. They

mainly involve decrease of ATP synthesis and acidosis

resulting from anaerobic glycolysis [1–3]. When anoxia

is followed by reperfusion, the quick restoration of

blood flow, and the accompanying excess of oxygen

supply promote other specific damages due to an excessof reactive oxygen species (ROS) generation [3,4].

As both cellular ATP synthesis and ROS generation

occur in mitochondria, one should be able to reproduce

the generation of early damages of anoxia-reoxygen-

ation using isolated mitochondria. Indeed, we recently

succeeded in reproducing the cellular anoxia-reoxygen-

ation damages in isolated rat brain mitochondria [5,6],

thus setting up a model to examine in detail the eventsoccurring during this process.

In this work, we report on the different events oc-

curring after reoxygenation, namely the modifications of

oxidative phosphorylation, mitochondrial membrane

* Corresponding author. Fax: +33-1-49-81-35-94.

E-mail address: ch.morin@univ-paris12.fr (C. Morin).

0006-291X/03/$ - see front matter � 2003 Elsevier Inc. All rights reserved.

doi:10.1016/S0006-291X(03)01203-8

fluidity, and lipoperoxidation. Furthermore, we also

investigated the effect of reperfusion on the mitochon-

drial permeability transition pore (MPTP) and the re-

lease of factors promoting apoptotic and necrotic cell

death [7].

Materials and methods

Chemicals. Sucrose, EGTA, DD-mannitol, malate, glutamate, ADP,

bovine serum albumin, a-tocopherol, and cyclosporin A were pur-

chased from Sigma Chemical. KCl, MgCl2, Tris base, and KH2PO4were purchased from Merck.

Isolation of brain mitochondria. Mitochondria were purified by

Percoll density gradient centrifugation according to the method de-

scribed by Sims [8], modified as we previously reported [5,6]. In brief,

mitochondria were extracted from the homogenate of two rat fore-

brains (Wistar male, 250–300 g; Janvier, Le Genest Saint Isle, France).

The homogenate was centrifuged at 2000g for 3min to remove cell

debris and nuclei. The supernatant was preserved and the pellet was

homogenized in 20ml of isolation buffer (Tris–HCl 20mM, sucrose

250mM, KCl 40mM, EGTA 2mM, and bovine serum albumin 1mg/

ml, pH 7.2, at 4 �C) and centrifuged at 2000g for 3min. Both super-natants were then mixed and centrifuged at 12,000g for 10min. The

pellet was resuspended in 20ml of 15% Percoll and 3ml fractions of

Fig. 1. (A) Typical oxygen electrode traces obtained in the anoxia-

reoxygenation model. Oxygen uptake by rat brain mitochondria

(0.2mg/ml) was measured at 37 �C, in the presence of 10mM gluta-

mate and 10mM malate as substrates, with a Clark-type electrode.

Mitochondria were submitted to 5min of anoxia and after 0.2, 2.5, or

5min of reoxygenation different parameters were measured as de-

scribed in Materials and methods: respiratory parameters (RCR, state

3, and state 4), lipoperoxidation, membrane fluidity, release of cyto-

chrome c, and release of CL. Membrane potential was studied after

5min of anoxia and during at least 15min of reoxygenation. (B) State

3, state 4, and respiratory control ratio (RCR) of brain mitochondria

after 5min of anoxia and at different times of reoxygenation. Oxygen

uptake by rat brain mitochondria (0.2mg/ml) was determined, at 37 �Cafter 5min of anoxia followed by 0.2, 2.5, or 5min of reoxygenation,

with a Clark-type electrode. RCR was calculated as the ratio of oxygen

uptake in the presence of 0.2mM ADP (state 3) to oxygen uptake in

the presence of 10mM glutamate and 10mM malate (state 4). State 3

( ), state 4 ( ), and RCR ( ) were expressed in percent with regard to

control mitochondria (100%), which were not submitted to anoxia.

Basal rates: state 3, 115� 4.2 nmolO2/min/mg proteins; state 4,

38.1� 11.9 nmolO2/min/mg proteins, RCR, 3.1� 0.3. The data shownare means� SEM (bars) from six experiments performed in triplicate.

**p < 0:01 versus control mitochondria without anoxia.

478 C. Morin et al. / Biochemical and Biophysical Research Communications 307 (2003) 477–482

this pellet were laid on two preformed layers consisting of 3.5ml of

23% Percoll and 3.5ml of 40% Percoll. The gradient was centrifuged

for 5min at 30,700g. The fraction accumulating near the interface of

the two lower layers was collected and slowly diluted 1:4 with isolation

buffer. The mixture was centrifuged twice at 12,000g for 10min, pro-

ducing a pellet which was resuspended in 300ll of respiratory buffer(mannitol 300mM, KH2PO4 10mM, KCl 10mM, and MgCl2 5mM,

pH 7.2) at 4 �C. This preparation provided sufficient mitochondriawith a high respiratory control ratio (RCR) for multiple experiments

to be performed in a single afternoon. Protein concentration of the

mitochondrial suspension was determined by the method of Lowry

et al. [9].

The animal protocol was approved by the French agency regarding

animal experimentation (authorization No. 00748).

In vitro anoxia-reoxygenation model. Oxygen uptake was deter-

mined using a Clark-type microelectrode (Hansatech, UK). Anaerobic

conditions were applied by placing 200lg of cerebral mitochondria,which consume the oxygen content in 200ll of respiratory buffer into aclosed incubation chamber at 37 �C [6] (Fig. 1A). Malate/glutamate

(10/10mM) were used as substrates and the anaerobic conditions were

obtained following one pulse of ADP (0.2mM) in 2min. Anoxia was

kept for 5min and followed by 0.2, 2.5, or 5min reoxygenation in-

duced by adding 300ll of respiratory buffer. After the reoxygenation,the oxygen concentration again became stable after 0.2min where the

different measures were carried out such as membrane potential be-

cause it was technically impossible to take samples at the end of the

anoxia to measure the different parameters. At the selected time of

reoxygenation (0.2, 2.5, or 5min), the polarographic chamber was

closed, and to initiate state 3 respiratory activity, 0.2mM ADP was

added to the cuvette. When all ADP was converted to ATP, the state 4

was measured. The following parameters were determined: the respi-

ratory rates calculated as nanomoles of O2 per min and per mg of

mitochondrial protein, the respiratory control ratio (RCR) expressed

as the ratio of state 3/state 4 oxygen consumption respiratory activities.

Controls were treated under the same conditions but without the

anoxia. The effect of a-tocopherol (1mM) added before the anoxia wastested in separate experiments.

The activity of each complex of the respiratory chain after anoxia-

reoxygenation was measured as we previously described [6].

The changes in fluorescence anisotropy (r) of mitochondrial

membranes labeling with 1,6-diphenyl-1,3,5-hexatriene (DPH) were

measured as previously described [5,6]. The hematoporphyrin (HP)

labeling was realized by the addition of 3 lM HP into stirred mito-

chondria suspension (1mg/ml). The mixture was incubated for 2min

before measuring anisotropy. HP incorporated very rapidly into mi-

tochondria, whereas DPH required much longer incubation time (>1 h

30). Then DPH was excited at 340 nm and its fluorescence was detected

at 460 nm whereas HP was excited at 520 nm and its fluorescence was

detected at 626 with a Perkin–Elmer LS 50B spectrophotofluorimeter

[10].

Following anoxia, the mitochondrial suspension, taken at different

times of reoxygenation, was quickly centrifuged in order to eliminate

mitochondria. The amount of cytochrome c release in the reaction

buffer was measured using a Quantikine M Rat/Mouse Cytochrome c

immunoassay (R&D Systems, GB).

The peroxidation of mitochondrial membrane lipids by ROS gen-

erated from the anoxia and after different times of reoxygenation was

determined by measuring the malondialdehyde formation in the mi-

tochondrial suspension formed in the presence of thiobarbituric acid

(1%) in trichloroacetic acid (3%) [11]. The obtained complex was

quantified using a Perkin–Elmer LS 50B spectrophotofluorimeter (Ex:

485 nm; Em: 553 nm).

The release of cardiolipin (CL) at different times of reoxygenation

was determined using an assay which is based upon the peculiar

properties of the fluorescent dye nonyl acridine orange (NAO) [12,13].

Two molecules of NAO bind with high affinity to one single CL

molecule, forming NAO dimers [12,13]. Thus, NAO (2.5lM) was

added to the mitochondrial suspension at various times of reoxygen-

ation. After centrifugation of this suspension the release of NAO–CL

complex in the supernatant was determined by fluorescence spectros-

copy (Ex: 317 nm; Em: 640 nm) as described by Mileykovskaya et al.

[14]. A standard curve of CL measured under the same conditions was

linear over the concentration range tested.

The membrane potential (Dw) of mitochondria during reoxygen-ation (open chamber; Fig. 1A) was measured by rhodamine 123

(0.3 lM) labeling (Ex: 503 nm; Em: 527 nm) [15]. The uncoupler car-bonyl cyanide m-chlorophenylhydrazone (CCCP; 10 lM) was used toobserve the complete Dw collapse of control mitochondria after

immediate addition.

C. Morin et al. / Biochemical and Biophysical Research Communications 307 (2003) 477–482 479

Statistical analysis. For each experiment, mean values were com-

pared by a one-way analysis of variance (ANOVA) followed by

Dunnett’s test.

Fig. 2. (A) Anisotropy of DPH-labeled mitochondria and MDA pro-

duction during reoxygenation. After 5min of anoxia, mitochondria

(0.2mg/ml) submitted to 0, 2.5, or 5min of reoxygenation were incu-

bated for 2 h at 4 �C with DPH ( ) or ( ) in the presence of thio-

barbituric acid (1%) in trichloroacetic acid (3%) for MDA production

as reported in Materials and methods. The fluorescence of DPH (Ex:

340 nm; Em: 460 nm) or MDA (Ex: 485 nm; Em: 553 nm) was mea-

sured with a Perkin–Elmer LS 50B spectrophotofluorimeter. a-To-copherol (1mM) was added before anoxia. The data shown are

means�SEM (bars) of six experiments, each performed in triplicate.

**p < 0:01 versus control mitochondria without anoxia. (B) Release of

cytochrome c and CL by cerebral mitochondria during reoxygenation.

After 5min of anoxia, the release of cytochrome c ( ) and CL ( ) by

rat brain mitochondria was measured at 0, 2.5, and 5min of reoxy-

genation as described in Materials and methods. a-Tocopherol (1mM)was added before anoxia. Results are expressed as means� SEM of six

experiments, each performed in triplicate and compared to those of the

cytochrome c or CL released by mitochondria without anoxia

(**p < 0:01).

Results

Early alterations of oxygen utilisation

RCR decreased at the beginning of reoxygenation

and proceeded with time, reaching 50% of initial value,

at 5min of reoxygenation. It was roughly proportional

to the period of oxygen exposure (Fig. 1B). These

changes might result from the combination of twoalterations, the increase of the oxygen consumption

(state 4) and the decrease of oxidative phosphorylation

(state 3). During the observation time, none of the en-

zymatic activities of the respiratory chain complexes

were modified (data not shown).

In the presence of a-tocopherol (1mM), after 5min ofanoxia and 5min of reoxygenation, the state 4 only in-

creased by 4% whereas the state 3 and the RCR onlydecreased by 8%. These results were not statistically

different from those of control mitochondria that were

not submitted to anoxia. Moreover, a-tocopherol didnot modify mitochondria not submitted to anoxia.

Thus, all the alterations of respiratory chain functions

induced by anoxia-reoxygenation were prevented by

a-tocopherol.

Early alteration of mitochondrial membranes

A continuous increase of MDA production, i.e.,

lipoperoxidation of mitochondrial membranes, was ob-

served during the reoxygenation process (Fig. 2A). A

simultaneous decrease of membrane fluidity occurred as

assessed by the increase in DPH fluorescence of an agent

probing hydrophobic regions, i.e., lipid phases of

membranes (Fig. 2A). In contrast, HP which probespolar, i.e., protein regions of the membrane was not

modified.

All these membrane alterations induced during

reoxygenation were fully prevented by a-tocopherol.

Early releases of cytochrome c and CL

At the beginning of reoxygenation, the burst of ROS

induces an important release of cytochrome c relatedwith the release of CL (Fig. 2B). These releases are lower

during the following few minutes of reoxygenation. The

amount of cytochrome c released during the 5min of

reoxygenation was about 10% of the total mitochondrial

cytochrome c content.

The fluorophore NAO has been reported to tightly

bind to the phospholipid CL. In our experimental con-

ditions, NAO did not fluoresce in the absence of CL and

the standard curve of CL binding was linear over the

concentrations tested. In addition, the NAO signal was

not dependent on the collapse of Dw as described by

Jacobson et al. [16]. Thus in our experimental condi-tions, the fluorescence of NAO was related to the pres-

ence of CL.

The amount of CL released was high, from 30.8 to

63.7 pmol/mg proteins between the beginning and the

end of the observation of the reoxygenation process

(Fig. 2B). The release of both cytochrome c and CL was

fully inhibited by a-tocopherol (1mM).

480 C. Morin et al. / Biochemical and Biophysical Research Communications 307 (2003) 477–482

Membrane potential (Dw) collapse during reoxygenation

A Dw collapse was observed both in control and

anoxia-reoxygenated mitochondria. The magnitude of

Dw collapse was identical to the complete depolarizationinduced by the respiratory chain uncoupler CCCP. Due

to our experimental conditions of fluorescence mea-

surement, this Dw collapse was caused by an anoxia.

Discussion

Following anoxia, the reoxygenation process induces

a cascade of successive alterations that increase the

severity of brain damages [3,4].

In order to elucidate the mechanisms of mitochon-

drial dysfunctions induced during reoxygenation, wehave developed an in vitro brain ischemia model using

isolated mitochondria submitted to a period of 5min

anoxia followed by different times of reoxygenation.

The present data demonstrate that all the early

functional alterations of mitochondria were due to an

excess of ROS generation. The reperfusion process, as

previously shown in this model [5], induced a burst of

O�2 and a production of H2O2 by the flavin mononu-

cleotide (FMN) group of complex I [17] and mainly by

the ubiquinone–cytochrome b-c1 region of complex III

[18]. This burst of ROS could be related with the MDA

generation at 0.2min of reoxygenation. The complete

prevention of alterations by a-tocopherol confirmed thisobservation. During the first 5min of reoxygenation, the

ROS attack was mainly directed to membrane struc-

tures, whereas enzymatic activities of respiratory chaincomplexes were not modified. More precisely, the MDA

generation and the increase of DPH fluorescence show

that the ROS attack was targeted to lipid components of

mitochondrial membranes [19]. The lack of HP fluo-

rescence modification indicated that at this step of the

reoxygenation process, protein regions of the mem-

branes were not modified.

The first consequence of these membrane alterationswas the decrease of oxidative phosphorylation combined

with the increase of oxygen consumption of state 4 that

reduced RCR by 50%. These alterations were very

similar to those seen in in vivo focal ischemia induced by

middle cerebral artery occlusion [3,20]. At 2 h of ische-

mia, respiratory activity was reduced by 45–60% in focal

tissue. This suggested that ROS, acting on membrane

lipids, uncoupled the respiratory chain. The observedincrease of anisotropy and the lipoperoxidation of mi-

tochondrial membranes suggested a modification of the

membrane fluidity, inducing either an increase of proton

transfer across the inner membrane [21] or a modifica-

tion of the electron shuttle into the inner membrane [22].

During reoxygenation, cytochrome c and CL were

simultaneously and quickly released. CL release by

brain mitochondria is a new observation, which may beput together with the uncoupling effect. Recent data

suggested a regulatory role of CL in the oxidative

phosphorylation process [23]. Indeed, mitochondria

isolated from the Saccharomyces cerevisiae cardiolipin

synthase-null mutant, thus without CL, exhibited in-

creased basal metabolic activity (state 4) suggesting that

CL might regulate the oxidative phosphorylation pro-

cess in a negative manner [24]. If so, CL release mayincrease the state 4 of oxygen consumption. Moreover,

as CL is a phospholipid exclusively found in the mito-

chondrial inner membrane, its release assesses the al-

teration of mitochondrial membrane, notably by the

modification of the membrane fluidity [25]. This release

of CL, which was not inhibited by CsA (data not

shown), did not involve the opening of MPTP. As cy-

tochrome c is bound to CL [26], release of the former islikely to occur. In agreement with this, it has been also

reported that changes in CL are due to oxidative dam-

age [27] and can trigger the release of cytochrome c from

mitochondria [28]. However, additional experiments are

needed to confirm this hypothesis. At present, it is not

clear whether the released CL is oxidized or not [29] and

which pool of cytochrome c is released, the “loosely

bound” by electrostatic interactions or the “tightlybound” by hydrophobic interactions [30].

The loss of mitochondrial cytochrome c may be

classically explained by two mechanisms, either a Ca2þ-

dependent process that involves the opening of the

MPTP, followed by the matrix swelling and the rupture

of the outer mitochondrial membrane [31], or a Ca2þ-

independent mechanism, remaining unclear, but prob-

ably governed by different members of Bcl-2 family ofproteins [32] and prevailing in the brain [33]. At present,

we consider the latter mechanism to be involved in the

anoxia-reoxygenation experiments reported herein, be-

cause of the absence of Ca2þ in the buffers used and the

inability of CsA to restore RCR, inhibit cytochrome c

release (data not shown), and prevent the collapse of the

membrane potential. Moreover, the use of isolated mi-

tochondria, obviously not subjected to any nuclearDNA activation, suggests a direct effect of ROS on cy-

tochrome c release that does not require the MPTP.

Considering the timing of the events, the RCR de-

crease, the MDA generation, the DPH fluorescence in-

crease, and the cytochrome c and CL release are not the

consequence of the Dw collapse. Indeed these events

occurred whereas the Dw was maintained during 8min

after the beginning of the reoxygenation. The anoxiainduced an uncoupling of oxidative phosphorylation

and thus an increase of oxygen consumption of mito-

chondria (Fig. 1A). This explains why the Dw collapse

was observed 110 s before the Dw collapse observed withcontrol mitochondria (Fig. 3). Consequently, these dif-

ferent alterations observed on mitochondria are likely

due to a direct effect of ROS on inner mitochondrial

Fig. 3. The membrane potential of brain mitochondria during reoxy-

genation. The membrane potential (Dw) of mitochondria during

reoxygenation was measured using rhodamine 123 (0.3lM; Ex:503 nm; Em: 527nm). (a) Control mitochondria without anoxia (open

chamber); (b) mitochondria submitted to 5min of anoxia; (c) control

mitochondria in the presence of CsA (2lM); (d) mitochondria sub-mitted to anoxia in the presence of CsA (2lM), and (e) Dw collapse ofcontrol mitochondria after the immediate addition of CCCP (10lM;see the arrow in the figure). Similar results were obtained in six sepa-

rate experiments.

C. Morin et al. / Biochemical and Biophysical Research Communications 307 (2003) 477–482 481

membrane. Thus, using malate/glutamate as substrates,

changes in Dw and the ROS generation by brain mito-

chondria are independent events [34].

In conclusion, our mitochondrial model of anoxia-

reoxygenation reproduces in vitro the main steps of an

oxidative stress in the brain. It confirms that ROS pro-

duced during the reoxygenation are responsible for theperoxidation of the lipid membrane, the rigidification of

the mitochondrial membranes, the loss of cytochrome c,

and shown for the first time in brain mitochondria, the

release of CL. This results in prominent changes in the

brain mitochondrial respiratory chain activity which it-

self induces an increase of the free radical production

leading to a vicious cycle. Interestingly, this model fo-

cuses exclusively on mitochondrial damages and disso-ciates alterations due to ROS during the reoxygenation

(early events) from the MPTP opening (latest events).

Lastly, it gives an opportunity for the study of the lone

steps of ROS-induced mitochondrial damages.

Acknowledgments

This work was supported by grants from the Minist�eere de

l’Education Nationale (DRED EA 427). We wish to thank Dr. Si-

mon N. (University of Marseille, France) for helpful discussions and

Pr. Papadopoulos V. (University of Georgetown, USA) for review-

ing the manuscript.

References

[1] R.L. Macdonald, M. Stoodley, Pathophysiology of cerebral

ischemia, Neurol. Med. Chir. 38 (1998) 1–11.

[2] C.N. Oliver, P.E. Starke-Reed, E.R. Stadtman, G.J. Liu, J.M.

Carney, R.A. Floyd, Oxidative damage to brain proteins, loss of

glutamine synthetase activity, and production of free radicals

during ischemia/reperfusion-induced injury to gerbil brain, Proc.

Natl. Acad. Sci. USA 87 (1990) 5144–5147.

[3] G. Fiskum, A.N. Murphy, M.F. Beal, Mitochondria in neurode-

generation: acute ischemia and chronic neurodegenerative dis-

eases, J. Cereb. Blood Flow Metab. 19 (1999) 351–369.

[4] C.A. Piantadosi, J. Zhang, Mitochondrial generation of reactive

oxygen species after brain ischemia in the rat, Stroke 27 (1996)

327–332.

[5] R. Zini, C. Morin, A. Bertelli, A.A.E. Bertelli, J.P. Tillement,

Resveratrol-induced limitation of dysfunction of mitochondria

isolated from rat brain in an anoxia-reoxygenation model, Life

Sci. 71 (2002) 3091–3108.

[6] C. Morin, R. Zini, N. Simon, J.P. Tillement, Dehydroepiandros-

terone and a-estradiol limit the functional alterations of rat brainmitochondria submitted to different experimental stresses, Neu-

roscience 115 (2002) 415–424.

[7] N.R. Sims, M.F. Anderson, Mitochondrial contributions to tissue

damage in stroke, Neurochem. Int. 40 (2002) 511–526.

[8] N.R. Sims, Rapid isolation of metabolically active mitochondria

from rat brain and subregions using Percoll density gradient

centrifugation, J. Neurochem. 55 (1990) 698–707.

[9] O.H. Lowry, N.J. Resebrough, A.J. Farr, R.J. Randall, Protein

measurement with the Folin phenol reagent, J. Biol. Chem. 193

(1951) 265–275.

[10] F. Ricchelli, S. Gobbo, G. Moreno, C. Salet, Changes of fluidity

of mitochondrial membranes induced by the permeability transi-

tion, Biochemistry 38 (1999) 9295–9300.

[11] R. Zini, C. Morin, A. Bertelli, A.A.E. Bertelli, J.P. Tillement,

Effects of resveratrol on the rat brain respiratory chain, Drugs

Exp. Clin. Res. XXV (1999) 87–97.

[12] J.M. Petit, A. Maftah, M.H. Ratinaud, R. Julien, 10N-nonyl

acridine orange interacts with cardiolipin and allows the quanti-

fication of this phospholipid in isolated mitochondria, Eur. J.

Biochem. 209 (1992) 267–273.

[13] J.M. Petit, O. Huet, P.F. Gallet, A. Maftah, M.H. Ratinaud, R.

Julien, Direct analysis and significance of cardiolipin transverse

distribution in mitochondrial inner membranes, Eur. J. Biochem.

220 (1994) 871–879.

[14] E. Mileykovskaya, W. Dowhan, R.L. Birke, D. Zheng, L.

Lutterodt, T.H. Haines, Cardiolipin binds nonyl acridine orange

by aggregating the dye at exposed hydrophobic domains on

bilayer surfaces, FEBS Lett. 507 (2001) 187–190.

[15] A. Elimadi, A. Settaf, D. Morin, R. Sapena, F. Lamchouri, Y.

Cherrah, J.P. Tillement, Trimetazidine counteracts the hepatic

injury associated with ischemia–reperfusion by mitochondrial

function, J. Pharmcol. Exp. Ther. 286 (1998) 23–28.

[16] J. Jacobson, M.R. Duchen, S.J. Heales, Intracellular distribution

of the flurescent dye nonyl acridine orange responds to the

mitochondrial membrane potential: implications for assays of

cardiolipin and mitochondrial mass, J. Neurochem. 82 (2002)

224–233.

[17] Y. Liu, G. Fiskum, D. Schubert, Generation of reactive oxygen

species by the mitochondrial electron transport chain, J. Neuro-

chem. 80 (2002) 780–787.

[18] D. Han, E. Williams, E. Cadenas, Mitochondrial respiratory

chain-dependent generation of superoxide anion and its release

into the intermembrane space, Biochem. J. 353 (2001) 411–416.

[19] S.P. Gabbita, R. Subramaniam, F. Allouch, J.M. Carne, D.A.

Butterfield, Effects of mitochondrial respiratory stimulation on

membrane lipids and proteins: an electron paramagnetic reso-

nance investigation, Biochim. Biophys. Acta 1372 (1998) 163–173.

[20] S. Kuroda, K.I. Katsura, R. Tsuchidate, B.K. Siesjo, Secondary

bioenergetic failure after transient focal ischaemia is due to

mitochondrial injury, Acta Physiol. Scand. 156 (1996) 149–150.

482 C. Morin et al. / Biochemical and Biophysical Research Communications 307 (2003) 477–482

[21] V.P. Skulachev, Uncoupling: new approaches to an old problem

of bioenergetics, Biochim. Biophys. Acta 1363 (1998) 100–124.

[22] G. Lenaz, Role of mitochondria in oxidative stress and ageing,

Biochim. Biophys. Acta 1366 (1998) 53–67.

[23] T.H. Haines, N.A. Dencher, Cardiolipin: a proton trap for

oxidative phosphorylation, FEBS Lett. 528 (2002) 35–39.

[24] V. Koshkin, M.L. Greenberg, Cardiolipin prevents rate-depen-

dent uncoupling and provides osmotic stability in yeast mito-

chondria, Biochem. J. 364 (2002) 317–322.

[25] A. Halestrap, The regulation of the matrix volume of mammalian

mitochondria in vivo and in vitro and its role in the control of

mitochondrial metabolism, Biochim. Biophys. Acta 976 (1989)

355–382.

[26] R.A. Demel, W. Jordi, H. Lambrechts, H. Van Damme, R.

Hovius, B. De Kruijff, Differential interactions of apo- and

holocytochrome c with acidic membrane lipids in model systems

and the implications for their import into mitochondria, J. Biol.

Chem. 264 (1989) 3988–3997.

[27] R.A. Kirkland, R.M. Adibhatla, J.F. Hatcher, J.L. Franklin, Loss

of cardiolipin and mitochondria during programmed neuronal

death: evidence of a role for lipid peroxidation and autophagy,

Neuroscience 115 (2002) 587–602.

[28] K. Nomura, H. Imai, T. Koumura, T. Kobayashi, Y. Nakagawa,

Mitochondrial phospholipid hydroperoxide glutathione peroxi-

dase inhibits the release of cytochrome c from mitochondria by

suppressing the peroxidation of cardiolipin in hypoglycaemia-

induced apoptosis, Biochem. J. 351 (2000) 183–193.

[29] G. Petrosillo, F.M. Ruggiero, M. Pistolese, G. Paradies, Reactive

oxygen species generated from the mitochondrial electron trans-

port chain induce cytochrome c dissociation from beef-heart

submitochondrial particles via cardiolipin peroxidation. Possible

role in the apoptosis, FEBS Lett. 509 (2001) 435–438.

[30] M. Ott, J.D. Robertson, V. Gogvadze, B. Zhivotovsky, S.

Orrenius, Cytochrome c release from mitochondria proceeds

by a two step process, Proc. Natl. Acad. Sci. USA 99 (2002)

1259–1263.

[31] P.X. Petit, M. Goubern, P. Diolez, S.A. Susin, N. Zamzami, G.

Kroemer, Disruption of the outer mitochondrial membrane as a

result of large amplitude swelling: the impact of irreversible

permeability transition, FEBS Lett. 426 (1998) 111–116.

[32] V. Gogvadze, J.D. Robertson, B. Zhivotovsky, S. Orrenius,

Cytochrome c release occurs Ca2þ-dependent via and Ca2þ-

independent mechanisms that are regulated by Bax, J. Biol. Chem.

276 (2001) 19066–19071.

[33] A. Andreyev, G. Fiskum, Calcium induced release of mitochon-

drial cytochrome c by different mechanisms selective for brain

versus liver, Cell Death Differ. 6 (1999) 825–832.

[34] T.V. Voytakova, I.J. Reynolds, Dwm-dependent and -independent

production of reactive oxygen species by rat brain mitochondria,

J. Neurochem. 79 (2001) 266–277.