FEBS Letters 582 (2008) 1519–1525

Cytochrome c is released from coupled mitochondria of yeast en routeto acetic acid-induced programmed cell death and can work as an

electron donor and a ROS scavenger

Sergio Giannattasioa,1, Anna Atlantea,1, Lucia Antonaccib, Nicoletta Guaragnellab,Paolo Lattanzioa, Salvatore Passarellab,*, Ersilia Marraa

a CNR, Istituto di Biomembrane e Bioenergetica, I-70126 Bari, and Trani (BA), Italyb Dipartimento di Scienze per la Salute, Universita del Molise, Via De Sanctis, I-86100 Campobasso, Italy

Received 2 January 2008; revised 15 March 2008; accepted 26 March 2008

Available online 7 April 2008

Edited by Vladimir Skulachev

Abstract To gain insight into the processes by which aceticacid-induced programmed cell death (AA-PCD) takes place inyeast, we investigated both cytochrome c release from yeastmitochondria and mitochondrial coupling over the time courseof AA-PCD. We show that the majority of cytochrome c releaseoccurs early in AA-PCD from intact coupled mitochondria whichundergo only gradual impairment. The released cytochrome ccan be reduced both by ascorbate and by superoxide anion andin turn be oxidized via cytochrome c oxidase, thus working bothas a ROS scavenger and a respiratory substrate. Late in AA-PCD, the released cytochrome c is degraded.� 2008 Federation of European Biochemical Societies. Pub-lished by Elsevier B.V. All rights reserved.

Keywords: PCD; Cytochrome c; Mitochondria; ROS;Saccharomyces cerevisiae

1. Introduction

The yeast Saccharomyces cerevisiae commits to pro-

grammed cell death (PCD) in response to different stimuli

[1,2] including acetic acid treatment (AA-PCD) [3,4]. The

mitochondrial pathway in yeast PCD has been recently re-

viewed [5]. Mitochondria play a major part in that they pro-

duce reactive oxygen species (ROS) and release different

proteins into the cytosol including cytochrome c (Cyt c)[6–

10]. Moreover, some mitochondrial functional alterations

such as decrease in cytochrome c oxidase (COX) activity

and in the amounts of COX II subunit and of cytochromes

a+a3 were also shown in AA-PCD [10]. Cyt c release was

shown to involve the ADP/ATP carrier for mitochondrial

outer membrane permeabilization [10,11]. Conversely, both

Abbreviations: AA-PCD, acetic acid-induced programmed cell death;Cyt c, cytochrome c; COX, cytochrome c oxidase; Fe3+-Cyt c,ferricytochrome c; KCN, potassium cyanide; PCD, programmed celldeath; RCI, respiratory control index; ROS, reactive oxygen species;TMPD, N,N,N 0,N0-tetramethyl-p-phenylenediamine

*Corresponding author. Fax: +39 0805443317.E-mail address: passarel@unimol.it (S. Passarella).

1These authors contributed equally to this paper.

0014-5793/$34.00 � 2008 Federation of European Biochemical Societies. Pu

doi:10.1016/j.febslet.2008.03.048

in mammalian and in plant apoptosis, it was shown that

cytochrome c is released early from intact coupled mito-

chondria in the absence of mitochondrial permeability tran-

sition opening [12–15]. Just as the role played by the

released Cyt c remains to be established, there is as yet no

information as to how Cyt c is released en route to AA-

PCD. On the other hand, since homologues of Apaf, the

mammalian apoptosome-associated factor, have not been

found and proteins of similar function have not been de-

scribed in yeast so far [5], it seems unlikely that the mecha-

nisms downstream of cytochrome c release are related to

formation of an apoptosome-like structure and subsequent

activation of the yeast caspase Yca1p, and it is necessary

to investigate other possible roles for released Cyt c in yeast

PCD.

We investigate here release of Cyt c as a function of time in

yeast en route to AA-PCD. We show that release occurs early

in AA-PCD from coupled mitochondria, that the mitochon-

dria only gradually become uncoupled, and that the released

Cyt c can both transfer reducing equivalents to the respiratory

chain and work as a ROS scavenger.

2. Materials and methods

2.1. Strains, media, acetic acid treatment and cell fractionationSaccharomyces cerevisiae strain used in this study was W303-1B

(MATa ade2 leu2 his3 trp1 ura3). Acetic acid treatment of exponen-tial phase (OD600 = 0.5–0.8) yeast cells and cell viability determina-tion was performed in rich medium (1% yeast extract and 2%Bacto-peptone) containing 2% dextrose as in Ref. [3]. Cell homoge-nates, mitochondria and cytosolic fractions were isolated from AA-PCD cells or cells without acetic acid treatment (control) essentiallyas described in Ref. [16]. Briefly, cells were incubated with zymolyase20T (MP Biomedicals) 2.5 mg/g of cell wet weight for 1 h at 30 �C in1.2 M sorbitol, 20 mM K2HPO4/KH2PO4, pH 7.4 (SIB) to obtainspheroplasts. Spheroplasts were harvested by centrifugation for5 min at 3000 · g, washed once with 1.2 M sorbitol and suspendedin an ice-cold buffer containing 0.6 M mannitol, 20 mM HEPES–KOH, pH 7.4 (MIB) (2 ml/g of starting cell wet weight). Spheroplastswere homogenized on ice by 30 strokes in a tight-fitting Douncehomogenizer. From this point on, all operations were carried outon ice. Homogenate was centrifuged for 5 min at 3000 · g. The super-natant was stored, and the pellet was re-homogenized as before andcentrifuged at 3000 · g. The supernatants were combined and usedimmediately for polarographic measurements. Alternatively, cellhomogenate was used for cytosolic and mitochondrial fraction prep-aration as described [16]. Protein content was determined accordingto Waddel and Hill [17].

blished by Elsevier B.V. All rights reserved.

1520 S. Giannattasio et al. / FEBS Letters 582 (2008) 1519–1525

2.2. ImmunoblottingTwenty microgram of either cytosolic or mitochondrial proteins

were loaded onto a 12% SDS–polyacrylamide gel, separated and trans-ferred to a polyvinylidene fluoride (PVDF) membrane which wasprobed with different antibodies: polyclonal anti-cytochrome c, kindlyprovided by R. Lill (Philipps-Universitat, Marburg), monoclonal anti-phosphoglycerate kinase (Pgk1p) (Molecular Probes) and polyclonalanti-acetohydroxyacid reductoisomerase (Ilv5p) kindly provided byR.A. Butow (University of Texas Southwestern Medical Center, Dal-las, USA). Immunoblot analysis was performed as in Ref. [13]. Densi-tometric values for immunoreactive bands were quantified using a GS-700 Imaging Densitometer (Bio-Rad Laboratories Inc.). Protein levelswere calculated as a percentage of control cells taken as 100 in arbi-trary units after normalization based on the amount of Pgk1p or Ilv5p,for cytosolic and mitochondrial fraction in each lane on the same filter.

2.3. Cytochrome c peroxidase assayCytochrome c peroxidase activity, a marker enzyme of the inter-

membrane space [16], was assayed in cell homogenates (about 2 mgprotein) according to Ref. [18]. Briefly, cell homogenate in 0.6 M man-nitol, 20 mM HEPES–KOH, pH 7.4 (MIB) was incubated in the pres-ence of 25 lM cytochrome c, previously reduced with a minimalamount of Na2S2O4, and absorbance decrease was measured at550 nm after 180 lM H2O2 addition. Relative enzyme activity was cal-culated as the percentage of total enzyme activity, taken as 100, mea-sured in the presence of 1% Triton X-100 to solubilize mitochondrialmembranes.

2.4. Polarographic measurements and cytochrome c assayFerricytochrome c (Fe3+-Cyt c) from bovine heart and biochemicals

were purchased from Sigma–Aldrich. O2 consumption was measuredat 25 �C in a thermostated water-jacketed glass vessel (final volumeequal to 1.5 ml) using cell homogenate (about 2 mg protein) in MIBcontaining 0.5 mM phenylmethylsulfonylfluoride (PMSF), by meansof a Gilson 5/6 oxygraph using a Clark electrode, as in Ref. [19]. In or-der to detect functional Cyt c released in the extramitochondrial phase,the capability of homogenate, containing Cyt c, to oxidize ascorbate

100

10

0

10

20

30

40

Control

Cyt c

Pgk1p

Control

0 30 60 90 150 200 250

Cyt c

Ilv5p

Ilv5p

time (min)

cyt

c pe

roxi

dase

acti

vity

(%

of

tota

l)

0 30 60 90 150 200 250

0 30 60 90 150 200

Control

cCyt c

c’Cyt c

mCyt c

AB

C

D

Fig. 1. Western blot analysis of the released cytochrome c. (A) Cytosolic andAA-PCD obtained at indicated times were analyzed by Western blotting anaamount of Cyt c, normalized with antibodies against Pgk1p or Ilv5p (cCmitochondrial one, normalized with antibody against Ilv5p (mCyt c, light greafter normalization, as a percentage of control cells, where control cells wergiven. Reported values are the means of triplicate measurements and represelevels in the control cells, measured by comparing the percentage of protein l5%. ANOVA and Bonferroni test: statistically significantly different with *P <in yeast en route to AA-PCD. The activity of Cyt c peroxidase in the cell homas % of the total activity (0.15 ± 0.08 mU/mg prot) measured after solubiliza

was checked as in Ref. [20]. The rationale for this approach is de-scribed in Section 3.

2.5. TUNEL assayAA-PCD and control cells (10 ml) were harvested after 30, 90 and

150 min incubation with or without acetic acid, respectively. Cells werefixed, digested with 750 lg/ml zymolyase 20T and transferred into pol-ylysine-coated wells of a microscope slide as in Ref. [21]. Each well wasincubated at room temperature with 50 ll phosphate buffer saline(PBS) containing 4% paraformaldehyde, 0.01 M NaOH for 60 min,then with 20 ll of 3% H2O2 in methanol for 10 min, washed withPBS and incubated with 20 ll fresh permeabilization solution (0.1%Triton X-100, 0.1% sodium citrate) for 2 min at 4 �C, and then with10 ll TUNEL reaction mixture (In Situ Cell Death Detection Kit,POD, Roche) for 60 min at 37 �C in a humidified container [22]. Slideswere rinsed with PBS, incubated with anti-fluorescein antibody conju-gated with peroxidase for 30 min at 37 �C, washed with PBS and incu-bated with 3,3 0-diaminobenzidine tetrahydrochloride (DAB EnhancedLiquid Substrate System for Immunohistochemistry, Sigma) for10 min at room temperature following instructions of the supplier.Colour development of the substrates was observed using an Axioplan2 microscope (Zeiss) equipped with a 100· objective. Digital images(200 cells per sample) were acquired with an Axiocam CCD camerausing Axio Vision software.

3. Results and discussion

Release of Cyt c from mitochondria of yeast en route to AA-

PCD was first determined by immunological assay (five exper-

iments with different yeast preparations) of the cytosolic frac-

tions from yeast cells at AA-PCD times up to 250 min after

acetic acid treatment by using a polyclonal anti-cytochrome

c antibody (Fig. 1A). In parallel, the purity of the cytosolic

fraction was assessed by using a monoclonal antibody against

102103

100

240

160

134

100

103108100

234

164

135

1022

100

5556

527290103100

0

40

80

120

160

200

240

time (min)

*

**

**

** **** **

*%

of

cont

rol

0 30

60 90 150 200 250 Control

(C) mitochondrial fractions from either control cells or cells undergoinglysis as described in Section 2. (B) The histogram reports the cytosolicyt c and c 0Cyt c, white and dark grey bars, respectively) and they bars), as quantified by densitometric scanning of the film, expressed,

e analyzed at the different times and to which a value of 100 has beenntatives of at least five different experiments. The variability of proteinevel from different control cells on the same filter, was always less than

0.05, **P < 0.001. (D) Integrity of the mitochondrial outer membraneogenate (�2 mg of protein) is reported as a function of AA-PCD time,tion of mitochondrial membranes with 1% Triton X-100.

S. Giannattasio et al. / FEBS Letters 582 (2008) 1519–1525 1521

Pgk1p and a polyclonal antibody against Ilv5p, cytosolic and

mitochondrial matrix protein markers, respectively. Probably

as a result of mitochondrial damage during isolation, a low

but constant amount of Ilv5p was found in the cytosolic frac-

tion of both control and AA-PCD cells at all times (Fig. 1A).

Even though a significant amount of Cyt c was found in the

cytosolic fraction from control cells, presumably arising from

mitochondrial damage during the fractionation procedure, a

very clear increase in released Cyt c was found in AA-PCD

cells from 60 up to 150 min (140 ± 20% and 134 ± 18% when

either Pgk1p or Ilv5p, respectively, was used to normalize

the amount of cytosolic protein loaded). The amount of Cyt

c detected declined thereafter (Fig. 1A and B). Consistently,

a decrease in the amount of mitochondrial Cyt c, as normal-

ized with Ilv5p, was revealed after 60 min with minimum at

150 min after which it remained constant up to 250 min

(Fig. 1B and C).

To confirm that en route to AA-PCD release of Cyt c occurs

from mitochondria with an intact outer membrane, we assayed

the activity of the cytochrome c peroxidase, a marker enzyme

Fig. 2. Oxygen uptake dependent on cytochrome c released from mitochondcells were incubated at 25 �C in 1.5 ml of MIB buffer in the presence of 3 lMadditions were made at the indicated concentrations: ascorbate (ASC; 5 mM),AA-PCD cells: homogenates from yeast cells taken at different times [(a) t =route to AA-PCD. (D) Cells taken at 30 min of AA-PCD were used, and Fe3+

AA-PCD were used, and Fe3+-Cyt c (5 lM) was added after ascorbate. NumO2/min/mg of cell protein from one representative experiment out of five ex

of the intermembrane space [16], in the cell homogenates as a

function of AA-PCD time (Fig. 1D). About 15 ± 5% of the to-

tal Cyt c peroxidase activity was detected in cell homogenates

from both control and AA-PCD cells at times up to 150 min.

At 200 min AA-PCD Cyt c peroxidase activity outside mito-

chondria was about 30 ± 3% of the total. This shows that dur-

ing the period of maximum release of Cyt c en route to AA-

PCD there was no time-dependent damage of mitochondria.

To confirm that the Cyt c released during AA-PCD was fully

functional, we resorted to a procedure [20] in which we mea-

sured the rates of oxygen consumption in the presence of

ascorbate (5 mM) by homogenates from control and yeast cells

at different times of AA-PCD (Fig. 2). Since ascorbate cannot

cross the outer mitochondrial membrane [23], oxygen uptake

can occur only in the presence of a reducing equivalent accep-

tor able in turn to reduce oxygen via COX, i.e. N,N,N 0,

N 0-tetramethyl-p-phenylenediamine (TMPD) (Fig. 2A) or

commercial Fe3+-Cyt c, a COX substrate (Fig. 2B); in both

cases externally added cyanide (1 mM) blocks oxygen uptake.

We noticed that no oxygen uptake was found in the control in

ria. (A, B) Control cells: homogenates (�2 mg of protein) from controlrotenone, 0.8 lM antimycin A, and 6 lM myxothiazole. The followingTMPD (0.2 mM) or Fe3+-Cyt c (5 lM) and cyanide (CN�; 1 mM). (C)

0 min; (b) t = 30 min; (c) t = 60 min; (d) t = 90 min; (e) t = 150 min] en-Cyt c (5 lM) was added after ascorbate. (E) Cells taken at 150 min of

bers along the traces are rates of oxygen uptake expressed as natom ofperiments with different yeast preparations.

1522 S. Giannattasio et al. / FEBS Letters 582 (2008) 1519–1525

the presence of ascorbate alone, showing both that mitochon-

dria were intact and that no release of Cyt c occurred. Very dif-

ferent results were obtained when ascorbate was added to

homogenates from cells en route to AA-PCD (Fig. 2C).

Whereas at 30 min of AA-PCD, the rate of oxygen uptake

was 4 natom of O2/min/mg cell protein (trace b), this increased

dramatically to about 50 natom of O2/min/mg cell protein

90 min AA-PCD (trace d) demonstrating that the Cyt c re-

leased at that time was in an active form. Interestingly, at

150 min AA-PCD, when maximum release of Cyt c occurs

(see Fig. 1A), oxygen uptake drastically decreased to about

4 natom of O2/min/mg (trace e). This could arise either from

loss of activity of the Cyt c or from damage to the mitochon-

dria. To resolve this point, we checked whether externally

added Fe3+-Cyt c, could increase oxygen uptake in the pres-

ence of ascorbate at 30 and 150 min AA-PCD (Fig. 2D). In

the former case, the initial ascorbate-dependent oxygen uptake

Cytosol

XANTHINE

URIC ACID

H O2+H

O2−.O2

XOD

Mitoc

x

SOD

0.02 nmol/mg

1 min

Abs

orba

nce

incr

ease

(55

0nm

)

XX

Fig. 3. ROS scavenging by released cytochrome c. Homogenates (�2 mg ofcontrol cells, were incubated at 25 �C in 1.5 ml of MIB in the presence of 3oxidase (1 e.u./ml) in the presence or absence of 1 mM cyanide. Cyt c reductiincrease of absorbance at 550 nm (e550 = 21 mM�1 cm�1 for Fe2+-Cyt c). SODanion-producing system used to reduce released Cyt c.

(4 natom of O2/min/mg) increased as a result of Fe3+-Cyt c

addition (25 natom of O2/min/mg), this showing that at this

AA-PCD time mitochondria could already oxidize external

Cyt c. In the latter case, the failure of added Fe3+-Cyt c to

stimulate oxygen uptake shows that mitochondria, in distinc-

tion from those of control yeast (Fig. 2B) and of 30 min AA-

PCD, were functionally incompetent at least at COX level.

In this regard, such a conclusion is in agreement with [11].

In a parallel experiment, we confirmed that the Cyt c released

at 150 min AA-PCD was still active by monitoring its ability to

be reduced by superoxide anion, thus working as a ROS scav-

enger, as in Ref. [24] (Fig. 3 and scheme therein). To do this,

we first confirmed the existence of oxidized Cyt c in the yeast

extracts using reduction by dithionite as in Ref. [25] (not

shown). Cell homogenate from either control yeast or from

yeast undergoing AA-PCD up to 200 min, was added with

rotenone, antimycin A plus myxothiazole used to block elec-

Cyt cOX

A550

½ O2

H O2

COX

Cyt cRED

hondrion

x

-C N

time (min)

Fe2+

-cyt

c (

pmol

/mg)

150 min

150 min + SOD

30 min

60 min

90 min

200 min/control

00

40

80

120

160

50 100 150

mim

momims

protein), from cells en route to AA-PCD at the indicated times or fromlM rotenone, 0.8 lM antimycin A, 6 lM myxothiazole, and xanthineon was started with 10 lM xanthine (XX) addition and measured as an

was added at 50 U concentration. The scheme depicts the superoxide

S. Giannattasio et al. / FEBS Letters 582 (2008) 1519–1525 1523

tron flow via Complex I and II respectively, and with a super-

oxide anion-producing system, consisting of 10 lM xanthine

plus 1 e.u./ml xanthine oxidase (XX + XOD) in the presence

of 1 mM potassium cyanide (KCN) used to block the COX.

OX

YG

EN

CO

NSU

MP

TIO

N

50 natom O2/ mg

1 min ADP ADP

CN_

SUCC

8

23

AA-PCD

Control

0

10

20

30

40

50

30 90 150time (min)

TU

NE

L p

osit

ive

cells

(%

)

2101501209060300

100

80

60

40

20

0

time (min)

3

2

1

% o

f su

rviv

al

RC

I

AA-PCD(150 min)

Control

8

9

SUCC

CN_

A

B

C

Fig. 4. Mitochondrial coupling and cell viability en route to AA-PCD.(A) Homogenates (�2 mg of protein) from control cells after 200 minincubation without acetic acid addition or 150 min AA-PCD cells wereincubated at 25 �C in 1.5 ml of MIB buffer. The following additionswere made at the indicated concentrations: succinate (SUCC; 5 mM),ADP (1 mM), cyanide (CN�; 1 mM). Numbers along the traces arerates of oxygen uptake expressed as natom of O2/min/mg of cellprotein. (B) Respiratory control index of AA-PCD (j) and control(u) cells as well as AA-PCD cell viability (d) at different times arereported. For details see text and Section 2. (C) Kinetics of theappearance of DNA strand breaks detected using the TUNEL assay inAA-PCD (black bars) and control (white bars) cells. A photomicro-graph of images of negative (control) and positive (AA-PCD) stainingof yeast cells at 150 min is shown.

No Cyt c reduction occurred in the cytosolic fraction obtained

from control cells. The amount of reduced Cyt c increased with

time from triggering of AA-PCD up to 150 min. It is important

to note that when KCN was omitted no reduction was found,

this showing that once the released Cyt c is reduced it is re-oxi-

dized by oxygen via COX at least up to 90 min AA-PCD (not

shown). We assume that the reduced Cyt c is oxidized at the

contact sites between the mitochondrial membranes which

are known to exist in yeast [26]. As a control, we checked that

very little reduction of Cyt c occurred in the cytosolic fraction

from cells after 150 min of AA-PCD incubated in the presence

of superoxide dismutase (SOD) (50 U), used as a ROS scaven-

ger. After 200 min AA-PCD the amount of reduced Cyt c re-

turned to control level. By plotting the amount of the

reduced Cyt c at the completion of the reaction against PCD

times, a sigmoidal-like curve was found (see inset in Fig. 3).

Having ascertained that release of Cyt c occurs from mito-

chondria which basically retain their cytochrome c peroxidase,

i.e. from mitochondria not undergoing outer membrane per-

meabilization and/or time-dependent damage, we investigated

whether and how mitochondrial coupling might change en

route to AA-PCD. To do this, we measured oxygen uptake

caused by 5 mM succinate in state 4 and 3 (i.e. in the absence

and presence of 1 mM ADP, respectively) as shown for control

cells and for cells at 150 min of AA-PCD in Fig. 4A. We then

plotted the respiratory control index (RCI), i.e. the ratio of the

rate of oxygen uptake in state 3 and 4, which mirrors mito-

chondrial coupling, as a function of AA-PCD time. The RCI

of the control cells stayed relatively constant (2.7–2.5), but that

of the cells undergoing PCD decreased with time being 2 at

90 min and 1 at 150 min of AA-PCD (Fig. 4B). Measurements

made in parallel of cell survival show that it was 40% of the

control at 90 min of AA-PCD, as in Ref. [4,27], when mito-

chondrial coupling was somewhat reduced, but the cells

retained partial ability to produce ATP by oxidative phosphor-

ylation. As a control that cell death occurs via PCD, the frag-

mentation of DNA in the nucleus, a typical hallmark of PCD,

was also examined in situ by using TUNEL assay (see Section

2) (Fig. 4C). TUNEL-positive nuclei were detected in AA-

PCD cells with the percentage increasing up to 40% at

150 min, whereas the nuclei of control cells remained unstained

or only slightly stained over the same time interval.

4. Conclusions

The release of cytochrome c from mitochondria of yeast

undergoing apoptosis has been widely investigated essentially

by using genetic methods [5]. Our approach was different.

We investigated release of Cyt c and changes in mitochondrial

coupling in yeast cells as a function of time during AA-PCD,

and sought to assign the role/s to the released Cyt c, which is

probably not involved in yeast apoptosome formation [5].

The picture emerging from this and related studies is as fol-

lows. Early ROS production occurs in AA-PCD in a manner

modulated by catalase and SOD [4,28,29]. Release of Cyt c oc-

curs with a lag starting at 60 min of AA-PCD (see inset in

Fig. 3), when 60% cells remain alive (Fig. 4B), and is complete

at 150 min. Release occurs from mitochondria that are still in-

tact, not undergoing outer membrane permeabilization (Figs. 1

and 4B). Thus, the yeast mitochondrial unselective channel

[14,15] does not play a role in Cyt c release, as already shown

1524 S. Giannattasio et al. / FEBS Letters 582 (2008) 1519–1525

for the mitochondrial permeability transition pore in Cyt c

release in cerebellar granule cells apoptosis [30]. Consistently,

using genetic methods, it was shown that deletion of the yeast

orthologue of the mitochondrial cyclophilin, which is a puta-

tive component of the mitochondrial permeability transition

pore, had no effect on AA-PCD [11]. We also show that, at

least up to 90 min of AA-PCD, the mitochondria from which

Cyt c has been lost can still maintain electron flow coupled to

ATP synthesis; this may be driven by an electron proton

gradient derived from released Cyt c oxidation as in Ref.

[31]. Consistently, the released Cyt c is in an active form since

it can transfer reducing equivalents from ascorbate and/or

from ROS to mitochondria thus working as a ROS scavenger.

In both cases, the reduced Cyt c can be oxidized by oxygen via

COX (Fig. 3). In this respect, Cyt c release in yeast mirrors that

found in cerebellar granule cells [20]. Notice that the capability

of the released Cyt c to work as an antioxidant has been

already proposed in Refs. [24,32]. At a late stage, there is a

gradual decrease in mitochondrial coupling with a dramatic

impairment of COX, due to inefficient COX assembly in the

absence of Cyt c as already shown in Ref. [33] and/or perhaps

due to lipid peroxidation [34] (Fig. 2). Under these conditions

oxidative phosphorylation is blocked. Finally, released Cyt c

is apparently degraded as judged by a decrease in immuno-

precipitation (Fig. 1), as also reported in Ref. [13], but at this

stage the possible involvement of YCA1 remains to be investi-

gated.

Acknowledgements: We thank Prof. S. Doonan for critically readingthe manuscript. This work was financed by MIUR Project Cluster03 to E.M and INTERREG TIORCAS to (S.P.)

References

[1] Madeo, F., Herker, E., Wissing, S., Jungwirth, H., Eisenberg, T.and Frohlich, K.U. (2004) Apoptosis in yeast. Curr. Opin.Microbiol. 7, 655–660.

[2] Frohlich, K.U., Fussi, H. and Ruckenstuhl, C. (2007) Yeastapoptosis – from genes to pathways. Semin. Cancer Biol. 17, 112–121.

[3] Ludovico, P., Sousa, M.J., Silva, M.T., Leao, C. and Corte-Real,M. (2001) Saccharomyces cerevisiae commits to a programmedcell death process in response to acetic acid. Microbiology 147,2409–2415.

[4] Giannattasio, S., Guaragnella, N., Corte-Real, M., Passarella, S.and Marra, E. (2005) Acid stress adaptation protects Saccharo-myces cerevisiae from acetic acid-induced programmed cell death.Gene 354, 93–98.

[5] Eisenberg, T., Buttner, S., Kroemer, G. and Madeo, F. (2007) Themitochondrial pathway in yeast apoptosis. Apoptosis 12, 1011–1023.

[6] Priault, M., Camougrand, N., Kinnally, K.W., Vallette, F.M. andManon, S. (2003) Yeast as a tool to study Bax/mitochondrialinteractions in cell death. FEMS Yeast Res. 4, 15–27.

[7] Pozniakovsky, A.I., Knorre, D.A., Markova, O.V., Hyman, A.A.,Skulachev, V.P. and Severin, F.F. (2005) Role of mitochondria inthe pheromone- and amiodarone-induced programmed death ofyeast. J. Cell Biol. 168, 257–269.

[8] Silva, R.D., Sotoca, R., Johansson, B., Ludovico, P., Sansonetty,F., Silva, M.T., Peinado, J.M. and Corte-Real, M. (2005)Hyperosmotic stress induces metacaspase- and mitochondria-dependent apoptosis in Saccharomyces cerevisiae. Mol. Microbiol.58, 824–834.

[9] Braun, R.J., Zischka, H., Madeo, F., Eisenberg, T., Wissing, S.,Buttner, S., Engelhardt, S.M., Buringer, D. and Ueffing, M.(2006) Crucial mitochondrial impairment upon CDC48 mutationin apoptotic yeast. J. Biol. Chem. 281, 25757–25767.

[10] Ludovico, P., Rodrigues, F., Almeida, A., Silva, M.T., Barrientos,A. and Corte-Real, M. (2002) Cytochrome c release andmitochondria involvement in programmed cell death induced byacetic acid in Saccharomyces cerevisiae. Mol. Biol. Cell 13, 2598–2606.

[11] Pereira, C., Camougrand, N., Manon, S., Sousa, M.J. and Corte-Real, M. (2007) ADP/ATP carrier is required for mitochondrialouter membrane permeabilization and cytochrome c release inyeast apoptosis. Mol. Microbiol. 66, 571–582.

[12] Vacca, R.A., Valenti, D., Bobba, A., Merafina, R.S., Passarella,S. and Marra, E. (2006) Cytochrome c is released in a reactiveoxygen species-dependent manner and is degraded via caspase-like proteases in tobacco Bright-Yellow 2 cells en route to heatshock-induced cell death. Plant Physiol. 141, 208–219.

[13] Bobba, A., Atlante, A., Giannattasio, S., Sgaramella, G., Calis-sano, P. and Marra, E. (1999) Early release and subsequentcaspase-mediated degradation of cytochrome c in apoptoticcerebellar granule cells. FEBS Lett. 457, 126–130.

[14] Manon, S., Roucou, X., Guerin, M., Rigoulet, M. and Guerin, B.(1998) Characterization of the yeast mitochondria unselectivechannel: a counterpart to the mammalian permeability transitionpore? J. Bioenerg. Biomembr. 30, 419–429.

[15] Gutierrez-Aguilar, M., Perez-Vazquez, V., Bunoust, O., Manon,S., Rigoulet, M. and Uribe, S. (2007) In yeast, Ca2+ andoctylguanidine interact with porin (VDAC) preventing the mito-chondrial permeability transition. Biochim. Biophys. Acta 1767,1245–1251.

[16] Daum, G., Bohni, P.C. and Schatz, G. (1982) Import of proteinsinto mitochondria. Cytochrome b2 and cytochrome c peroxidaseare located in the intermembrane space of yeast mitochondria. J.Biol. Chem. 257, 13028–13033.

[17] Waddel, W.J. and Hill, C. (1956) A simple ultraviolet spectro-photometric method for the determination of protein. J. Lab.Clin. Med. 48, 311–314.

[18] Djavadi-Ohaniance, L., Rudin, Y. and Schatz, G. (1978) Identi-fication of enzymically inactive apocytochrome c peroxidase inanaerobically grown Saccharomyces cerevisiae. J. Biol. Chem.253, 4402–4407.

[19] Atlante, A., Gagliardi, S., Minervini, G.M., Marra, E., Passarella,S. and Calissano, P. (1996) Rapid uncoupling of oxidativephosphorylation accompanies glutamate toxicity in rat cerebellargranule cells. Neuroreport 7, 2519–2523.

[20] Atlante, A., Calissano, P., Bobba, A., Azzariti, A., Marra, E. andPassarella, S. (2000) Cytochrome c is released from mitochondriain a reactive oxygen species (ROS)-dependent fashion and canoperate as a ROS scavenger and as a respiratory substrate incerebellar neurons undergoing excitotoxic death. J. Biol. Chem.275, 37159–37166.

[21] Zhang, N.N., Dudgeon, D.D., Paliwal, S., Levchenko, A., Grote,E. and Cunningham, K.W. (2006) Multiple signaling pathwaysregulate yeast cell death during the response to mating phero-mones. Mol. Biol. Cell 17, 3409–3422.

[22] Madeo, F., Frohlich, E., Ligr, M., Grey, M., Sigrist, S.J., Wolf,D.H. and Frohlich, K.U. (1999) Oxygen stress: a regulator ofapoptosis in yeast. J. Cell Biol. 145, 757–767.

[23] Alexandre, A. and Lehninger, A.L. (1984) Bypasses of theantimycin a block of mitochondrial electron transport in relationto ubisemiquinone function. Biochim. Biophys. Acta 767, 120–129.

[24] Korshunov, S.S., Krasnikov, B.F., Pereverzev, M.O. and Skula-chev, V.P. (1999) The antioxidant functions of cytochrome c.FEBS Lett. 462, 192–198.

[25] Errede, B., Kamen, M.D. and Hatefi, Y. (1978) Preparation andproperties of complex IV (ferrocytochrome c: oxygen oxidore-ductase EC 1.9.3.1). Methods Enzymol. 53, 40–47.

[26] Simbeni, R., Pon, L., Zinser, E., Paltauf, F. and Daum, G. (1991)Mitochondrial membrane contact sites of yeast. Characterizationof lipid components and possible involvement in intramitochon-drial translocation of phospholipids. J. Biol. Chem. 266, 10047–10049.

[27] Guaragnella, N., Pereira, C., Sousa, M.J., Antonacci, L., Passa-rella, S., Corte-Real, M., Marra, E. and Giannattasio, S. (2006)YCA1 participates in the acetic acid induced yeast programmedcell death also in a manner unrelated to its caspase-like activity.FEBS Lett. 580, 6880–6884.

S. Giannattasio et al. / FEBS Letters 582 (2008) 1519–1525 1525

[28] Guaragnella, N., Antonacci, L., Passarella, S., Marra, E. andGiannattasio, S. (2007) Hydrogen peroxide and superoxide anionproduction during acetic acid-induced yeast programmed celldeath. Folia Microbiol. 7, 237–240.

[29] Guaragnella, N., Antonacci, L., Giannattasio, S., Marra, E. andPassarella, S. (2007) Catalase T and Cu,Zn-superoxide in theacetic acid-induced programmed cell death in Saccharomycescerevisiae. FEBS Lett. 582, 210–214.

[30] Atlante, A., Bobba, A., de Bari, L., Fontana, F., Calissano, P.,Marra, E. and Passarella, S. (2006) Caspase-dependent alterationof the ADP/ATP translocator triggers the mitochondrial perme-ability transition which is not required for the low-potassium-dependent apoptosis of cerebellar granule cells. J. Neurochem. 97,1166–1181.

[31] Atlante, A., de Bari, L., Bobba, A., Marra, E., Calissano, P. andPassarella, S. (2003) Cytochrome c, released from cerebellargranule cells undergoing apoptosis or excytotoxic death, cangenerate protonmotive force and drive ATP synthesis in isolatedmitochondria. J. Neurochem. 86, 591–604.

[32] Skulachev, V.P. (1998) Cytochrome c in the apoptotic andantioxidant cascades. FEBS Lett. 423, 275–280.

[33] Barrientos, A., Pierre, D., Lee, J. and Tzagoloff, A. (2003)Cytochrome oxidase assembly does not require catalytically activecytochrome C. J. Biol. Chem. 278, 8881–8887.

[34] Petrosillo, G., Ruggiero, F.M. and Paradies, G. (2003) Role ofreactive oxygen species and cardiolipin in the release of cyto-chrome c from mitochondria. Faseb J. 17, 2202–2208.