Báo cáo khoa học: Antioxidant defences and homeostasis of reactive oxygen species in different human mitochondrial DNA-depleted cell lines pot

Eur. J. Biochem. 271, 3646–3656 (2004) Ó FEBS 2004

doi:10.1111/j.1432-1033.2004.04298.x

Antioxidant defences and homeostasis of reactive oxygen species
in different human mitochondrial DNA-depleted cell lines
Lodovica Vergani1, Maura Floreani2, Aaron Russell3, Mara Ceccon1, Eleonora Napoli4, Anna Cabrelle5,
Lucia Valente2, Federica Bragantini1, Bertrand Leger3 and Federica Dabbeni-Sala2
1

Dipartimento di Scienze Neurologiche and 2Dipartimento di Farmacologia e Anestesiologia, Universita` di Padova, Padova, Italy;
Clinique Romande de Re´adaptation SUVA Care, Sion, Switzerland; 4E.Medea Scientific Institute, Conegliano Research Centre,
Conegliano, Italy; 5Dipartimento di Medicina Clinica, Universita` di Padova, c/o Istituto Veneto di Medicina Molecolare, Padova, Italy

3

Three pairs of parental (q+) and established mitochondrial
DNA depleted (q0) cells, derived from bone, lung and muscle
were used to verify the influence of the nuclear background
and the lack of efficient mitochondrial respiratory chain on
antioxidant defences and homeostasis of intracellular
reactive oxygen species (ROS). Mitochondrial DNA depletion significantly lowered glutathione reductase activity,
glutathione (GSH) content, and consistently altered the
GSH2 : oxidized glutathione ratio in all of the q0 cell lines,
albeit to differing extents, indicating the most oxidized redox
state in bone q0 cells. Activity, as well as gene expression and
protein content, of superoxide dismutase showed a decrease
in bone and muscle q0 cell lines but not in lung q0 cells. GSH
peroxidase activity was four times higher in all three q0 cell
lines in comparison to the parental q+, suggesting that
this may be a necessary adaptation for survival without a

functional respiratory chain. Taken together, these data
suggest that the lack of respiratory chain prompts the cells to
reduce their need for antioxidant defences in a tissue-specific
manner, exposing them to a major risk of oxidative injury. In
fact bone-derived q0 cells displayed the highest steady-state
level of intracellular ROS (measured directly by 2¢,7¢-dichlorofluorescin, or indirectly by aconitase activity) compared to all the other q+ and q0 cells, both in the presence or
absence of glucose. Analysis of mitochondrial and cytosolic/
iron regulatory protein-1 aconitase indicated that most
ROS of bone q0 cells originate from sources other than
mitochondria.

Cellular reactive oxygen species (ROS), such as superoxide
1 anions (2 À ), and hydrogen peroxide (H2O2), have long
been held to be harmful by-products of life in an aerobic
environment. ROS are potentially toxic because they are
highly reactive and modify several types of cellular macromolecules. Lipid, protein and DNA damage can lead to

cytotoxicity and mutagenesis [1]. Therefore, cells have
evolved elaborate defence systems to counteract the effects
of ROS. These include both nonenzymatic (glutathione,
pyridine nucleotides, ascorbate, retinoic acid, thioredoxin
and tocopherol) and enzymatic (such as superoxide dismutases, catalase, glutathione peroxidase and peroxiredoxin) pathways, which limit the rate of oxidation and
thereby protect cells from oxidative stress [1,2]. Notwithstanding, evidence is emerging that ROS also act as signals
or mediators in many cellular processes, such as cell proliferation, differentiation, apoptosis, and senescence [3–5].
The redox environment of a cell may alter the balance
between apoptosis and mitosis by affecting gene expression
and enzyme activity [6]. Consequently, cellular redox state is
increasingly accepted as a key mediator of multiple metabolic, signalling and transcriptional pathways essential for
normal function and cell survival or programmed cell death

[3–6].
Mitochondria are certainly the major cellular site for
oxygen reduction and hence the site with the greatest
potential for ROS formation. An estimated 0.4–0.8% [7]
to 2–4% [8] of the total oxygen consumed during electron
transport is reduced not to water by cytochrome c oxidase
but rather to superoxide by complexes I, and III of the
respiratory chain [1,7,8]. ROS production increases when
respiratory flux is depressed by a high ATP/ADP ratio,
high electronegativity of auto-oxidizable redox carriers in

Correspondence to L. Vergani, Dipartimento di Scienze Neurologiche,
`
Universita di Padova, c/o Istituto Veneto di Medicina Molecolare,
Via Orus 2, 35129 Padova, Italy. Fax: +39 049 7923271,
Tel.: +39 049 7923219, E-mail:
Abbreviations: CS, citrate synthase; CuZnSOD, copper zinc superoxide dismutase; DCF, 2¢,7¢-dichlorofluorescin; DTT, 1,4-dithio-DLthreitol; GSH, glutathione; GSSG, oxidized glutathione; GPx, GSH
peroxidase; GR, GSSG reductase; GST, GSH transferase; H2-DCFDA, 2¢,7¢-dichlorofluorescin-diacetate; IRP-1, iron regulatory protein1; LDH, lactate dehydrogenase; MFI, mean log fluorescence intensity;
MnSOD, manganese superoxide dismutase; MPA, metaphosphoric
acid; mt, mitochondrial; NBT, nitroblue tetrazolium; PMRS, plasma
membrane oxidoreductase system; PBN, N-tert-butyl-a-phenylnitrone; ROS, reactive oxygen species; SOD, superoxide dismutase.
Enzymes: catalase (EC 1.11.1.6); GSH peroxidase (EC 1.11.1.9);
GSSG reductase (EC 1.8.1.7); GSH transferase (EC 2.5.1.18); Mn
superoxide dismutase, CuZn superoxide dismutase, superoxide
dismutase (EC 1.15.1.1).
(Received 26 April 2004, revised 16 July 2004, accepted 23 July 2004)

Keywords: A549 q0 cells; antioxidant defences; 143 q0 cells;
reactive oxygen species; rhabdomyosarcoma q0 cells.

Ó FEBS 2004

complex I and III, or a rise in oxygen tension (state 4
respiration). Defects in respiratory complexes [9] and
normal aging [10] also lead to increased mitochondrial
ROS production. A recent study [11] indicates that
mitochondrial ROS homeostasis plays a key role in the
life and death of eukaryotic cells, as mitochondria not
only respond to ROS but also release ROS in response to
a number of pro-apoptotic stimuli. However, mitochondria are not the sole source of cellular ROS. ROS also
form in the cytosol and in peroxisomes as by-products of
specific oxidases [7,10]. The plasma membrane oxidoreductase system (PMRS) also influences cellular redox
state [12,13].
Mitochondria are partially autonomous organelles; they
possess DNA, which contributes essential proteins to the
oxidative phosphorylation system. In vitro mammalian
cells can be depleted entirely of their mitochondrial DNA,
creating so-called q0 cells [14,15]. Rho0 cells lack a
functional electron transport chain and appear incapable
of generating ATP from mitochondria. Moreover, it is still
a debated question [16] whether or not q0 cells may
generate ROS at the mitochondrial level. Therefore, q0
cells may require alternative mechanisms for energy supply
and for maintenance of an appropriate redox environment
[17,18]. Analysis of q0 cells has provided insights into
oxygen metabolism [13,17,19–21] and the role of mitochondria in redox signalling during apoptosis [22,23].
Redox-sensitive signalling and sensitivity to oxidative
stress depend on the cell type and its antioxidant systems,
due to differential tissue expression of nuclear genes [24].

There are no reports that compare antioxidant defences
and ROS homeostasis between mitochondrial (mt)DNAdepleted cells with different nuclear backgrounds. In this
study, soluble and enzymatic antioxidant systems and
ROS steady-state level were characterized in three tumour
cell lines derived from bone (osteosarcoma, 143B), muscle
(rhabdomyosarcoma, RD) and lung (adenocarcinoma,
A549) and in the respective q0 cells: 143Bq0 (bone), RDq0
(muscle) and A549q0 (lung) cells. This approach was
undertaken to investigate the effect of the absence of
electron transport chain on cellular redox homeostasis,
with the hypothesis that ROS levels could be altered in
consequence of the ablation of an efficient respiratory
chain. We aimed to verify: (a) if q0 status requires
antioxidant defence systems as efficient as those of normal
q+ cells; (b) if nuclear background influences redox
homeostatis in the different cell lines, precursors of
cytoplasmic hybrids (cybrids), that are useful tool for
studies of mtDNA segregation [25,26].

Experimental procedures
Materials
All reagents and enzymes were from Sigma. NaCl/Pi from
Oxoid had the following composition: NaCl 8 gỈL)1, KCl
0.2 gỈL)1, Na2HPO4 1.15 gỈL)1 and KH2PO4 0.2 gỈL)1
(pH 7.3). Tissue culture reagents were purchased from
Gibco-Invitrogen Co. Reverse transcription was performed
using the Stratascript enzyme (Stratagene). 2¢,7¢-Dichlorofluorescin-diacetate (H2-DCF-DA) was from Molecular
Probes.

Homeostasis of ROS in q0 cells (Eur. J. Biochem. 271) 3647

Cell lines and culture conditions
The q+ wild-type osteosarcoma cells (143B) and the q0 cells
derived from 143B were a gift from G. Attardi (Division of
Biology, California Institute of Technology, Pasadena, CA,
2 USA) [14], RD and RDq0 cells were established by Vergani
et al. [27], lung carcinoma (A549) and the derived q0 cells
were a gift from I. J. Holt (MRC, Dunn Human Nutrition
3 Unit, Cambridge, UK) [25]. The cells were grown in
Dulbecco’s modified Eagle’s medium containing 4.5 gỈL)1
glucose, 110 mgỈL)1 pyruvate, supplemented with 10%
4 (v/v) fetal bovine serum, 100 unitsỈmL)1 penicillin, and
0.1 mgỈmL)1 streptomycin, at 37 °C in a humidified atmosphere of 5% CO2. The medium for q0 cells was additionally
supplemented with 50 lgỈmL)1 uridine. The absence of
mtDNA in these three cell lines was reconfirmed at several
time points throughout the study by PCR as described
previously [14,25,27]. Routinely, 2 · 106 q+ or q0 cells
were seeded on 100 mm diameter plates and harvested after
42–48 h of culture during the period of exponential growth.
Subcellular fraction preparation
In some experiments regarding aconitase reactivation (see
below), 40 · 106 cells suspended in 0.8 mL were treated
with digitonin (0.5 mgỈmL)1) in NaCl/Pi for 15 min on ice.
The samples were centrifuged at 17 000 g for 15 min at
4 °C, the supernatant (cytosolic fraction) and the pellet
(mitochondria-enriched fraction), as well as the whole cells,
were recovered, immediately frozen in liquid N2 and stored
at )80 °C. Aliquots, kept at )80 °C for up to 2 weeks, were
thawed immediately before the assay, as reported previously
[28]. As markers of cytosolic and mitochondria-enriched

fractions, lactate dehydrogenase (LDH) [29] and citrate
synthase (CS) [30] activities were assayed in total cells and in
cytosolic and mitochondria-enriched fractions, respectively.
In mitochondria-enriched fractions CS activity was twice
the value found in the whole cells, whereas cytosolic
contamination, checked by measuring LDH, ranged from
10 to 30%. In the cytosolic fractions the contamination of
mitochondria, checked by measuring CS activity, was about
10% of the value found in whole cells.
Antioxidant defences
Glutathione and oxidized glutathione amounts. Cellular
glutathione (GSH) and oxidized glutathione (GSSG) levels
were measured enzymatically by using a modification of the
procedure of Anderson, as described [31,32]. The assay is
based on the determination of a chromophoric product,
2-nitro-5-thiobenzoic acid, resulting from the reaction of
5,5¢-dithiobis-(2-nitrobenzoic acid) with GSH. In this
reaction, GSH is oxidized to GSSG, which is then
reconverted to GSH in the presence of glutathione reductase
and NADPH. The rate of 2-nitro-5-thiobenzoic acid
formation is measured spectrophotometrically at 412 nm.
The cells (about 5–6 · 106 cells) were washed once with
NaCl/Pi and treated with 6% (v/v) metaphosphoric acid
(MPA) (1 mLỈdish)1) at room temperature. After 10 min
the acid extract was collected, centrifuged for 5 min at
18 000 g at 4 °C and processed. The cellular debris
remaining on the plate were solubilized with 0.5 M KOH


3648 L. Vergani et al. (Eur. J. Biochem. 271)

and assayed for their protein content [33]. For total
glutathione determination, the above acid extract was
diluted (1 : 6) in 6% (v/v) MPA; thereafter to 0.1 mL of
supernatant, 0.75 mL 0.1 M potassium phosphate, 5 mM
EDTA buffer pH 7.4, 0.05 mL 10 mM 5,5¢-dithiobis-2nitrobenzoic acid (prepared in 0.1 M phosphate buffer) and
0.08 mL 5 mM NADPH were added. After a 3 min
equilibration period at 25 °C, the reaction was started by
the addition of 2 U glutathione reductase (type III, Sigma,
from bakers yeast, diluted in 0.1 M phosphate/EDTA
buffer). Product formation was recorded continuously at
412 nm (for 3 min at 25 °C) with a Shimadzu UV-160
spectrophotometer. The total amount of GSH in the
samples was determined from a standard curve obtained
by plotting known amounts (from 0.05 to 0.4 lgỈmL)1) of
GSH vs. the rate of change of absorbance at 412 nm. GSH
standards were prepared daily in 6% (v/v) MPA and diluted
in phosphate/EDTA buffer pH 7.4. For GSSG measurement, soon after preparation the supernatant of acid extract
was treated for derivatization with 2-vinylpiridine at room
temperature for 60 min. In a typical experiment, 0.15 mL of
supernatant was treated with 3 lL of undiluted 2-vinylpyridine. Nine microliters of triethanolamine were also
added, the mixture was vigorously mixed, and the pH was
checked; it was generally between 6 and 7. After 60 min,
0.1 mL aliquots of the samples were assayed by means of
the procedure described above for total GSH measurement.
The amount of GSSG was quantified from a standard curve
obtained by plotting known amounts of GSSG (from 0.05
to 0.20 lgỈmL)1) vs. the rate of change of absorbance. GSH
present in the samples was calculated as the difference
between total glutathione and GSSG levels.

Antioxidant enzyme activities. GSH peroxidase (GPx),
GSSG reductase (GR), catalase, superoxide dismutase
(SOD) and GSH transferase (GST) activities were measured
in monolayer cells (about 2–3 · 106 cells), washed three
times with NaCl/Pi before treatment directly on the dish
with 0.25 M sucrose, 10 mM Tris/HCl pH 7.5, 1 mM
EDTA, 0.5 mM phenylmethanesulfonyl fluoride, 0.5 mM
1,4-dithio-DL-threitol (DTT) and 0.1% (v/v) Nonidet
(named solution A), to obtain complete lysis of intracellular
organelles. Cells were then scraped from the plate and the
samples were centrifuged for 30 min at 105 000 g. Protein
content measurements [33] and enzymatic assays were
carried out on the clear supernatant fractions.
Total GPx activity was measured according to the
coupled enzyme procedure with glutathione reductase, as
described [34], using cumene hydroperoxide as substrate.
The enzymatic activity was monitored by following the
disappearance of NADPH at 340 nm for 3 min at 25 °C.
The incubation medium (final volume 1 mL) had the
composition 50 mM KH2PO4 pH 7.0, 3 mM EDTA, 1 mM
KCN, 1 mM GSH, 0.1 mM NADPH, 2 U glutathione
reductase and % 300 lg protein. After a 3 min equilibration
period at 25 °C, the reaction was started by the addition of
0.1 mM cumene hydroperoxide dissolved in ethanol. The
specific activity was calculated by using an extinction molar
coefficient obtained by a standard curve of NADPH
between 0.02 and 0.1 lmolesỈmL)1 and GPx activity
was expressed in nmoles NADPH consumedỈmg
protein)1Ỉmin)1.

Ĩ FEBS 2004

GR activity was measured according to the method of
Carlberg & Mannervik [35], by following the rate of
oxidation of NADPH by GSSG at 340 nm for 3 min at
25 °C. The reaction mixture (final volume 1 mL) contained
0.1 M KH2PO4 pH 7.6, 0.5 mM EDTA, 1 mM GSSG,
0.1 mM NADPH, and % 300 lg protein. The specific activity
was calculated by using an extinction molar coefficient
obtained by a standard curve of NADPH between 0.02 and
0.1 lmolesỈmL)1 and GR activity was expressed in nmoles
NADPH consumedỈmg protein)1Ỉmin)1.
Total catalase activity was assayed according to the
method of Aebi [36]. Activity was measured by monitoring,
for 30 s at 25 °C, the decomposition of 10 mM H2O2 at
240 nm in a medium (final volume 1 mL) consisting of
50 mM phosphate buffer pH 7.0 and % 100 lg proteins.
Catalase activity was expressed as unitsỈmg protein)1,
assuming that 1 unit of catalase decomposes 1 lmole of
H2O2Ỉmin)1.
For SOD activity assay a 0.6 mL aliquot of cell lysate
was sonicated on ice (2 · 30 s) and centrifuged for 30 min
at 105 000 g. The supernatant was collected and dialysed
5 overnight in cold double-distilled water to remove small
interference substances [37]. Enzymatic assays were carried
out according to the method of Oberlay & Spitz [38], with
minor modifications. Briefly, in 1 mL 50 mM KH2PO4
pH 7.8 and 0.1 mM EDTA, a superoxide-generating system (0.15 mM xanthine plus 0.02 U xanthine oxidase) was
used together with 50 lM nitroblue tetrazolium (NBT) to
monitor superoxide formation by following the changes in

colorimetric absorbance at 560 nm for 5 min at 25 °C. The
catalytic activities of the samples were evaluated as their
ability to inhibit the rate of NBT reduction; increasing
amounts of proteins (5–150 lg) were added to each sample
until maximum inhibition was obtained. SOD activity was
expressed as unitsỈmg protein)1, with 1 unit of SOD
activity being defined as the amount of proteins causing
half-maximal inhibition of the rate of NBT reduction.
GST activity was assayed in the supernatant of cell
lysates, as described [39]. Briefly, 150 lg protein were
incubated in 50 mM KH2PO4 pH 6.5, 1 mM GSH and
0.25 mM 1-chloro-2,4-dinitrobenzene. The reaction was
followed for 2 min at 37 °C at 340 nm, and GST activity
was calculated using an extinction coefficient of
9.6 mM)1Ỉcm)1 [39].
Reverse transcription and quantitative PCR
RNA (5 lg) was reverse transcribed to cDNA using
random hexamer primers and the Stratascript enzyme.
Quantitative PCR was performed using an MX3000p
thermal cycler system and BrilliantÒ SYBER Green QPCR
Master Mix (Stratagene). The conditions for the amplification of copper zinc superoxide dismutase (CuZnSOD),
manganese superoxide dismutase (MnSOD) and the normalization gene, ribosomal 36B4, were as follows. One
denaturation step at 90 °C for 10 min, 40 cycles consisting
of denaturation at 90 °C for 30 s, annealing at 56 °C for
60 s for CuZnSOD and MnSOD and 60 °C for 36B4,
elongation at 72 °C for 60 s. At the end of the PCR the
samples were subjected to melting curve analysis. All
reactions were performed in triplicate. The primer
sequences were CuZnSOD [40], sense 5¢-GCGACGAAG

Ó FEBS 2004

GCCGTGTGCGTGC-3¢, antisense 5¢-ACTTTCTTCATT
TCCACCTTTGCC-3¢; MnSOD [40], sense 5¢-CTTCA
GCCTGCACTGAAGTTCAAT-3¢, antisense 5¢-CTGAA
GGTAGTAAGCGTGCTCCC-3¢; 36B4, sense 5¢-GTGA
TGTGCAGCTGATCAAGACT-3¢, antisense 5¢-GATGA
CCAGCCCAAAGGAGA-3¢.
Western blot analysis
Cells were lysed in the same buffer as used for the enzyme
activity assay. An equal amount of protein (40 lgỈlane)1)
for each sample was separated by SDS/PAGE (12%
acrylamide) and transferred to nitrocellulose membrane.
The membrane was blocked in 5% (w/v) nonfat dry milk in
6 0.02 M Tris/HCl pH 7.5, 0.137 M NaCl, and 0.1% (v/v)
Tween-20 for 3 h at room temperature. After overnight
incubation at 4 °C in 1 : 1000 of primary antibodies
to CuZnSOD (Santa Crutz) or MnSOD (Stressgen Biotechnology), membranes were probed with horseradish
peroxidase-conjugated secondary antibody (Amersham
Biosciences). Bound antibody was visualized using an
ECL reagent (Amersham Biosciences). Densitometric analysis of Western blot signal was performed using IMAGEMASTER VDS-CL (Amersham Pharmacia Biotech) and
IMAGE-MASTER TOTALLAB v1.11 software.

Homeostasis of ROS in q0 cells (Eur. J. Biochem. 271) 3649

Results
The steady-state levels of intracellular ROS depends on the
balance between rates of ROS generation and detoxification. A crucial role in determining ROS cellular homeostasis
is played by the antioxidant defence systems. Therefore

soluble (GSH and GSSG) and enzymatic defences (GPx,
GR, SOD, catalase and GST) were characterized on three
human tumour cell lines, with (q+) and without (q0)
mtDNA. GSH concentration was significantly decreased in
all three mtDNA depleted cell lines compared to parental
lines with mtDNA; the decrease in GSH content was most
pronounced in bone 143B q0 cells (Fig. 1). GSSG was also
lower in q0 cells compared with q+, but only statistically
significant in bone-derived cells (Fig. 1). The percentage of

ROS measurement
Aconitase determination. Aconitase activity was measured
as described previously [41] on 1 · 106 cells or on the
subcellular fractions obtained as reported above. The
samples were dissolved in 0.1% (v/v) Triton X-100 and
incubated for 15 min at 30 °C in 50 mM Tris/HCl pH 7.4,
0.6 mM MgCl2, 0.4 mM NADP, 5 mM Na citrate. To start
the assay, 2 U isocitrate dehydrogenase were added
and activity was measured by monitoring absorbance at
340 nm for 15 min. Reactivation of aconitase was
obtained by adding 50 lM DTT, 20 lM Na2S and 20 lM
Fe(NH4)2(SO4)2 directly into the cuvette, just before
spectrophotometric determination [41].
DCF fluorescence. Direct detection of intracellular steadystate levels of ROS was carried out on living cells using 2¢,7¢dichlorofluorescin-diacetate (H2-DCF-DA) [42–44]. The
probe is de-acetylated inside the cell. The subsequent
oxidation by intracellular oxidants yields a fluorescent
product, 2¢,7¢-dichlorofluorescin (DCF). Cells were collected
by trypsinization and centrifuged for 5 min at 800 g. The
pellet was incubated in tissue-culture medium with 5 lM
H2-DCF-DA for 30 min at 37 °C. Cells were washed and

then suspended (1 · 106 per mL) in medium (standard
growth conditions) or in NaCl/Pi for 90 min (stress
conditions). A FACSCalibur analyser (Becton-Dickinson
Immunocytometry Systems) equipped with a 488 Argon
laser was used for measurements of intracellular fluorescence. Dead cells were excluded by electronically gating data
on the basis of forward- vs. side-scatter profiles; a minimum
of 1 · 104 cells of interest were analysed further. Logarithmic detectors were used for the FL-1 fluorescence channel
necessary for DCF detection. Mean log fluorescence
intensity (MFI) values were obtained by the CELLQUEST
software program (Becton-Dickinson).

Fig. 1. GSH and GSSG concentrations and ratio of GSH2 : GSSG in
q+ and q0 cells from osteosarcoma (bone), rhabdomyosarcoma (muscle)
and lung carcinoma (lung). Values are expressed as means ± SD of at
least three assays carried out in duplicate. Significant differences from
respective q+ value at: *P < 0.05; **P < 0.01.


3650 L. Vergani et al. (Eur. J. Biochem. 271)

mitochondrial GSH in respect to total GSH was similar in
all tested q+ and q0 cell lines, ranging from 2.7 to 5% (data
not shown). To assess the cellular redox state we measured
the GSH2 : GSSG ratio which is considered a good index of
this parameter [45]. MtDNA loss was associated with an
alteration in this ratio with q0 cells having a more oxidized
redox state than q+ cells. However the change was
statistically significant only in bone-derived q0 cells. Moreover, the different values found in bone, muscle and lung q0
cells were all significantly different (P < 0.05) from each
other; in fact the GSH2 : GSSG ratio of bone 143Bq0 cells is

about one-half of that in muscle RDq0 cells and even three
to four times lower than that measured in lung A549q0 cells.
GPx and GR are crucial antioxidant defences as GPx
transforms H2O2 to H2O by coupling the oxidation of GSH
to GSSG and GR mediates the reduction of GSSG to GSH.
In the three cell lines tested, mtDNA loss was associated
with a four-fold increase in GPx activity and a significant
decrease in GR activity (Fig. 2). Moreover Fig. 2 shows
that the absolute values of GPx and GR activity were
considerably higher in lung q0 cells than in other q0 cells
(Fig. 2). Catalase activity was assessed in q+ and q0 cells;
our findings show that such activity was not affected by
mtDNA depletion (data not shown).
Activity, gene expression and protein content of SOD
were studied. Total SOD activity was decreased in bone and
muscle q0 cells compared with their parental q+ lines

Fig. 2. GPx and GR activities in q+ and q0 cells from osteosarcoma
(bone), rhabdomyosarcoma (muscle) and lung carcinoma (lung). Values
are expressed as means ± SD of at least three assays carried out in
duplicate. Significant differences from respective q+ value at:
**P < 0.01; ***P < 0.001.

Ó FEBS 2004

Fig. 3. Total SOD activity in q+ and q° cells from osteosarcoma (bone),
rhabdomyosarcoma (muscle) and lung carcinoma (lung). Values are
expressed as means ± SD of at least three assays carried out in
duplicate. Significant differences from respective q+ value at:
***P < 0.001.

(Fig. 3), whereas there were no significant differences in the
activity and expression levels in lung q+ and q0 cells
(Figs 3–5). Quantitative PCR (Fig. 4) and Western blot
(Fig. 5) analysis were carried out to evaluate the relative
contribution of MnSOD and CuZnSOD. Both analyses
confirmed that bone q0 cells had significantly lower
expression of CuZnSOD than the other cells. In musclederived cell lines mtDNA ablation reduced the expression
and protein amount of mitochondrial MnSOD but not of
cytosolic CuZnSOD (Figs 4 and 5). Densitometric analysis
of Western blot was in line with the results of quantitative
PCR (data not shown).
Glutathione S-transferase (GST) enzymes metabolize
xenobiotics as well as aldehydes, endogenously produced
during lipid peroxidation, by conjugation with GSH.
Moreover, some GSTs also show glutathione-peroxidaselike activity [1]. GST activity was decreased to a similar
extent in bone- and muscle-derived q0 cells, compared with
the parental q+ cells, but the absolute value was significantly higher in bone than in muscle q0 cells. No differences
were evident in lung q+ and q0 cell lines (Fig. 6). To check
the ability of the antioxidant defences to balance ROS
generation, indirect and direct measurements of intracellular
steady state levels of ROS were performed. Indirect
measurements were carried out by assessing the aconitase
activity. Aconitase is a four iron–sulfur cluster (Fe–S)containing hydratase, present in various subcellular
compartments (i.e. mitochondria and cytosol) which is
inactivated by 2 À [41]. In the cytosol, loss of aconitase
activity results in the conversion of this enzyme to the iron
regulatory protein-1 (IRP-1), that serves to regulate iron
homeostasis [46], and mitochondrial aconitase inactivation
serves as a protective response to oxidative stress [46].

Aconitase activity was measured in q+ and q0 cell lines
under basal culture conditions and after 18 h of treatment
with the ROS spin-trapping N-tert-butyl-a-phenylnitrone
(PBN) [47,48]. Figure 7 shows a trend of increasing
aconitase activity in almost all PBN-treated cell lines. The
increase was most marked in bone q+ and q0 cells (more


Ó FEBS 2004

Homeostasis of ROS in q0 cells (Eur. J. Biochem. 271) 3651

Fig. 5. Western blotting analysis of CuZnSOD and MnSOD in q+ and
q0 cells from osteosarcoma (bone), rhabdomyosarcoma (muscle) and lung
carcinoma (lung). Total cell extract was resolved by SDS/PAGE and
blotted onto nitrocellulose. The membrane was cut in strips, corresponding to the different molecular masses of MnSOD, CuZnSOD and
actin, the last acting as an internal standard, and incubated with the
corresponding antibody. Forty micrograms of cell protein extract was
loaded in each lane. The blots depicted are representative of three
separate experiments.

Fig. 4. Quantitative real-time PCR of CuZnSOD and MnSOD in q+
and q0 cells from osteosarcoma (bone), rhabdomyosarcoma (muscle) and
lung carcinoma (lung). mRNA values of CuZnSOD and MnSOD are
normalized for ribosomal 36B4 gene and are expressed as
means ± SD of three assays in triplicate in arbitrary units (A.U.).
Significant differences from respective q+ value at: *P < 0.05.

than five-fold) and in muscle q0 cells, suggesting that the
2 À level was higher in these cells than in lung q0 cells.

Both mitochondrial [28,46] and cytosolic IRP-1/aconitase
activities [46] are reactivated in the presence of reducing
agents and free Fe2+ carrier–donor [41]. Therefore, in an
attempt to localize 2 À production, we assessed aconitase
reactivation in these subcellular fractions. Reactivated
aconitase showed a dramatic increase in cytosolic fractions
of bone q0 cells (Fig. 8), whereas in mitochondria-enriched
fractions there were no significant differences.
Lastly, by means of the DCF technique coupled to flow
cytometric analysis, intracellular fluorescence was measured
as an index of steady-state levels of ROS under basal and
stress conditions (Fig. 9, Table 1). In the presence of glucose
and 10% serum (standard growth conditions), the fluorescence measured in q0 cells was lower than that in the
parental cell lines containing mtDNA. The decrease was
substantial in lung (90%) and muscle (40%) cells but was
less evident in bone (less than one-third) (Table 1). When
the cells were incubated in NaCl/Pi for 90 min, the
intracellular fluorescence signal dramatically increased in
all cases (Fig. 9, Table 1). The increases, in comparison to
the signals observed in standard growth conditions, were

Fig. 6. GST activity in q+ and q0 cells from osteosarcoma (bone),
rhabdomyosarcoma (muscle) and lung carcinoma (lung). Values are expressed as means ± SD of at least three assays carried out in duplicate. Significant differences from respective q+ value at: **P < 0.01.

consistently greater in q0 than in q+ cells, yet the extent of
the increase varied considerably between the three q0 lines.
In bone and lung q0 cells the increases were 17- and 39-fold,
respectively. However only in bone q0 cells was DCF
oxidation significantly higher compared to the value of the
respective q+ cell line (Table 1).

Discussion
Our analysis of three pairs of q+ and q0 cells, derived from
bone, muscle and lung, indicates that these cells differ
significantly both in their antioxidant defences and intracellular ROS homeostasis. The antioxidant system is


3652 L. Vergani et al. (Eur. J. Biochem. 271)

Ó FEBS 2004

Fig. 7. Aconitase activity in whole cells in absence (–) and presence (+)
of PBN. Rho+ and q0 cells from osteosarcoma (bone), rhabdomyosarcoma (muscle) and lung carcinoma (lung) were cultured in the absence (–) or the presence (+) of 500 lM PBN for 18 h. Aconitase
activity were assayed spectrophotometrically in cell lysate. Values are
expressed as means ± SD of at least three assays in duplicate as
nmolesỈmin)1Ỉmg)1 protein. –PBN value significantly different from
+PBN value at: *P < 0.05; **P < 0.01; ***P < 0.001.

profoundly affected by mtDNA depletion in a tissue
specific-manner, probably as a response to a decreased
need of efficient antioxidant machinery.
+

Antioxidant defences of parental q cell lines
The parental (q+) A549 cells, derived from type II human
alveolar epithelial cells [49], are provided with the highest
GSH content and GSH2 : GSSG ratio (Fig. 1), and the
highest GPx, GR (Fig. 2) and SOD (Fig. 3) activities in
comparison with bone and muscle derived q+ cells. This
very efficient ROS defence system may be related to the high

oxygen tension normally present in the lung and explains
the great resistance of these cells to apoptosis, after exposure
to high oxygen concentrations [50]. By contrast, bone
(143B)- and muscle derived (RD)- cells are similar in their
low content of GSH (only one-half of that present in A549)
and poor GPx activity (Figs 1 and 2); however, RD cells
differ significantly in GR activity and in particular in
activity, gene expression and protein content of SOD
(Figs 3–5).
Antioxidant defences of q0 cell lines
GSH-GSSG and GR. We measured GSH and GSSG in
exponentially growing cells, as GSH content changes in the
growth and lag phases [51]. In all q0 cells studied, GSH was
significantly lower than in the respective parental cells, with
the lowest GSH level in bone-derived q0 cells, and significant
differences in the GSH2 : GSSG ratios among the different
q0 cells (Fig. 1). The intracellular content of GSH is the
result of balance between its synthesis and consumption.
GSH synthesis is a two-step ATP-requiring process catalysed by cytosolic c-glutamylcysteine synthetase (c-GCS)
and GSH synthetase and is regulated (feedback-inhibited)
by GSH itself [52]. We neither directly measured these

Fig. 8. Aconitase reactivation. Aconitase activity was assayed in mitochondrial and cytosolic fractions of q+ and q0 from osteosarcoma
(bone), rhabdomyosarcoma (muscle) and lung carcinoma (lung).
Reactivation was achieved in presence of reducing agents (DTT) and
Fe2+ carrier–donor [Fe(NH4)2(SO4)2], as described in Experimental
procedures, and is expressed as percentage of basal value. Basal values
(nmolesỈmin)1Ỉmg protein)1) of mitochondrial aconitase activity were:
in bone q+ ¼ 3.26 ± 1.87 (4); bone q0 ¼ 2.36 ± 0.93 (4); muscle
q+ ¼ 8.77 ± 0.57 (3); muscle q0 ¼ 2.08 ± 0.19 (3); lung q+ ¼

8.46 ± 4.12 (3); lung q0 ¼ 4.88 ± 0.59 (3). Basal cytosolic aconitase
in bone q+ ¼ 1.64 ± 0.57 (4); bone q0 ¼ 2.81 ± 1.12 (4); muscle
q+ ¼ 0.76 ± 0.29 (3); muscle q0 ¼ 1.26 ± 0.53 (3); lung q+ ¼
4.79 ± 0.6 (3); lung q0 ¼ 4.59 ± 2.27 (3). Significant differences from
respective q+ value at: *P < 0.05, **P < 0.01.

activities in our q0 cells nor did we find reports on this topic
in the literature, but we did find a very low amount of ATP
(data not shown) in all of the q0 cells compared with the
respective parental q+ cells. The smaller GSH pool in q0
cells (reduced GSH and GSSG) suggests that it could be due
to reduced synthesis rather than to enhanced utilization in
cells with low amounts of ATP. In fact if the lower level of
GSH in q0 cells was due to its extensive consumption in the
GPx pathway or to a direct interaction with ROS, we
should find increased GSSG. In our experimental conditions we found that GSSG levels in all q0 cell lines were not
increased, but rather decreased, although GR activity was
significantly decreased in all q0 cells (Fig. 2). However, it
cannot be excluded that GSSG is actively secreted from the
7 cells subjected to an oxidative stress [52] in an attempt to
maintain cellular redox environment [45]. Therefore our
data could indicate that mtDNA-depleted cells need less


Homeostasis of ROS in q0 cells (Eur. J. Biochem. 271) 3653

Ó FEBS 2004

ρ+

103

104

101

102
FL1-H

103

104

101

102

103

104

103

104

101

102
FL1-H

103

104

101

102
FL1-H

103

104

60
40

Counts

20
80 100
60
40
20
0

0

100

102

FL1-H

0

100

Counts

60
40
20

Counts

Lung

101

80 100

100

80 100

100

60 80 100
20
0

0

Counts

102
FL1-H

0

Muscle

101

30 60 90 120 150 180

100

40

Counts

60
20

Bone

40

Counts

80 100

ρ0

FL1-H
Blank
Standard growth condition
Stressed condition

100

Fig. 9. DCF oxidation in cells with and without glucose. Rho+ and q0 cells from osteosarcoma (bone), rhabdomyosarcoma (muscle), and lung
carcinoma (lung) were collected and loaded with H2-DCF-DA. Fluorimetric signals of oxidized DCF (excitation, 488 nm; emission, 530 nm) were
recorded by cytofluorimeter from cells in presence of glucose (dotted line): standard growth conditions or in absence of glucose (bold line): stressed
conditions. Blank signal, obtained from cells without H2-DCF-DA, was deducted to the reported MFI values. The panels are representative of the
separate experiments summarized in Table 1.
Table 1. Levels of DCF oxidation in q+ and q0 cells from osteosarcoma
(bone), rhabdomyosarcoma (muscle) and lung carcinoma (lung). MFI of
the DCF signal was measured by fluorescence activated cell sorting as
arbitrary units in cells in presence of glucose (standard growth conditions) and in absence of glucose (stress conditions). Values are expressed as mean ± SD as arbitrary units of fluorescence. Numbers in
parentheses are the numbers of experiments. Significant differences
from respective q+ value: *P < 0.05; ***P < 0.001.
Conditions
Standard growth
Bone
Muscle
Lung

a

q+
q0
q+
q0
q+
q0

Stressa

186
142
208
143
235
25

1275
2500
1055
996
1693
976

±
±
±
±
±
±

33 (4)
75 (6)
3 (3)
4 (3)***
13 (3)
2 (3)***

±
±
±
±
±
±

92 (3)
217 (3)*
315 (3)
210 (3)
245 (3)
319 (3)

P < 0.001 vs. respective values in standard growth conditions.

anti-ROS buffer in the form of GSH for loss of ROS
mitochondrial fluctuation and of ROS spike, occurring
8 when the respiratory chain is active.

SOD, GST, GPx and catalase
With the exception of catalase and GPx activity, depletion
of mtDNA diminished SOD and GST activities in boneand muscle-derived q0 cells but not in lung-derived q0 cells

(Figs 3–6), where SOD (Figs 3–5) and GST (Fig. 6) were
unaffected after ablation of the respiratory chain. In bone
and muscle q0 cells SOD activity decreased (Fig. 3) as
compared with the respective parental q+ cells. Expression
level analysis revealed that in bone q0 cells CuZnSOD
mRNA (Fig. 4) and protein content were decreased
(Fig. 5), whereas in muscle q0 cells MnSOD decreased in
mRNA and protein amount compared with parental cells
(Figs 4 and 5). The decrease of SOD and GST antioxidant
enzymes in bone and muscle but not in lung q0 cells might be
ascribed to different expression–regulation of nuclear genes
as a response to cell type differential redox-sensitive
signalling [53].
Catalase activity is unaffected by mtDNA depletion
(data not shown) and, interestingly, the activity of GPx
was found to be considerably increased in all q0 cells
relative to the parental cells (Fig. 2). GPx, together with
catalase and thioredoxin peroxidase, restricts H2O2 accumulation and the consequent production of highly reactive


Ó FEBS 2004

3654 L. Vergani et al. (Eur. J. Biochem. 271)

hydroxyl radicals, for which no physiological defence
system exists [1]. In the last few years, the view of hydrogen
peroxide as a merely toxic by-product of cellular metabolism has changed, and it is now recognized as playing an
important role in intracellular signalling [3–5]. Fine regulation of redox balance may therefore be a critical function
of peroxidases, catalase and of GPx, in particular [54]. GPx
regulates the intracellular hydroperoxides and lipid hydroperoxides used as signal transducers of many transcription

factors including nuclear factor-jB [55], AP-1 [56] and
MAP kinases [57]. Because catalase is unchanged, the
increased GPx activity of q0 cells may be an essential
cellular adaptation that enables gene expression to function
normally in the absence of mtDNA. These findings are
in line with results found in hepatoma-derived Hep1q0
cells [16].
ROS
When DCF signal was assessed as a direct index of ROS, all
of the q0 cells had a reduced intracellular fluorescence
compared to q+ cells. Bone-derived q0 cells had the highest
level of intracellular ROS compared to muscle and lung q0
cells both in standard growth conditions and in stressed
conditions (Fig. 9, Table 1). If the current idea, that the
DCF technique mainly determines cellular peroxides [42–
44,58], is accepted it can be hypothesized that q0 cells
accumulate a lower DCF fluorescence signal due to their
high GPx activity (Fig. 2) in a tissue-specific manner. In
fact, lung q0 cells have the lowest DCF oxidation (Fig. 9,
Table 1) and the highest GPx activity (Fig. 2), whereas
bone- and muscle-derived q0 cells have rather similar GPx
activities and similar capacities to eliminate intracellular
oxidants under standard growth conditions. Yet, in the
absence of glucose (stress conditions), intracellular levels of
ROS in bone-derived q0 cells are 2.5 times those of muscle
q0 cells (Fig. 9, Table 1). This may be due to the fact
that among q0 cells, bone q0 cells had the less efficient
antioxidant machinery with the lowest GSH level (Fig. 1).
Interestingly, bone-derived q0 cells also featured the highest
glucose consumption rate and glucose-6-phosphate dehydrogenase activity among the six lines analysed (L. Vergani,

unpublished data). Glucose-6-phosphate dehydrogenase is
the rate-limiting enzyme in the pentose phosphate pathway
and a major source of cytosolic NADPH and ribose
phosphate [59]. When glucose is scarce, NADPH synthesis
decreases. This lead to a decrease in GSH levels as NADPH
is required for GSH regeneration via GR. Therefore, our
data suggest that increased generation of intracellular ROS
in bone q0 cells, relative to muscle q0, is due to increased
production of oxidants. The high production of ROS in
bone-derived q0 cells is further confirmed by indirect
measurement of ROS obtained by comparing aconitase
activity in standard conditions and after 18 h of incubation
with PBN (Fig. 7). In biological systems PBN [60,61], or
N-t-butyl hydroxylamine, a breakdown product of PBN
[47,48], efficiently trap free radicals, such as superoxide
anion (2 À ) that in turn inactives aconitase [41]. The
observed PBN-induced increase in aconitase activity in bone
q+ and q0 cells and in muscle q0 cells (Fig. 7) strongly
supports a high presence of 2 À in these cells also in
standard growth conditions. These data are well related to

the lowest GSH2 : GSSG ratio and the most oxidized redox
state (Fig. 1). A PBN effect on antioxidant enzyme activities
may be excluded on the basis of a recent report showing that
PBN protects U937 cells against ionizing radiation-induced
oxidative damage by altering cellular redox state but not
affecting antioxidant enzymes [61].
New and original evidence emerges from the experiments
of reactivation of aconitase activity by reducing agents and
Fe(NH4)2(SO4)2, as a Fe2+ carrier–donor [41]. Figure 8

shows a dramatic increase in cytosolic IRP-1/aconitase
activity in bone q0 cells, but not in mitochondria-enriched
fractions. This finding suggests that in bone q0 cells
intracellular oxidants derive chiefly from nonmitochondrial
compartments and are therefore not related to a vestige of
the respiratory electron transport chain. Possible sources of
nonmitochondrial oxidants include NADPH oxidases [12],
and lipoxygenases, whose action plays a role in signal
pathways of growth factor-stimulated bone cell mitogenesis
[62], and microsomal redox systems [63]. NADPH oxidases
are up-regulated in lymphoblastoid q0 cells, as a compensatory phenomenon in maintaining cell viability [18]. Our
results confirm PMRS as a possible source of ROS in bone
cells, as the NADPH oxidase inhibitor diphenyleniodonium chloride reduces fluorescence accumulation into
bone q+ and q0 cells to 65–70% (data not shown).
Another possible explanation for the increased generation
of intracellular oxidants in bone-derived q0 cells is the high
O2 tension to which cultured cells are exposed compared to
the low O2 tension of osteoblasts. The bulk of intracellular
oxidants in bone-derived q0 cells is in extra-mitochondrial
compartments, corroborating an earlier report which
showed q0 cells to be sensitive to the ablation of cytosolic
SOD [64]. Moreover the presence of extramitochondrial
ROS in q0 cells could explain the similar levels of oxidative
DNA damage observed in Hela q0 and the parental q+
cells [65].
In conclusion, our study demonstrates that loss of
functional mitochondria, the major cellular site for ROS
formation, reduces enzymatic and soluble intracellular
antioxidant defences but not ROS flux in the studied q0
cells, and that there are cell line-to-cell line variations in

intracellular antioxidant defences and ROS homeostasis. In
fact among the studied cells, those originating from bone are
particularly vulnerable to free radical-induced stress after
mtDNA ablation. These differences could reflect tissuespecific aspects of intracellular oxidant metabolism,
although it is inevitable that some specific features of ROS
homeostasis in terminally differentiated tissues such as
bone, lung and muscle will have been lost during the
transformation process that led to tumour formation. The
pronounced difference in intracellular homeostasis between
lung A549 and bone 143B q0 cells may also be germane to
mtDNA segregation bias, as selection of mutant and wildtype mtDNA is different in the 143B and A549 cellular
backgrounds [25,26].

Acknowledgements
We thank Dr G. Attardi for the gift of osteosarcoma q0 and q+cells, Dr
I.J. Holt for the gift of lung carcinoma q0 and q+ cells and we are grateful
to Dr Aubrey de Grey for great help in interpreting and discussing the
data. This work was supported by Telethon grant no. 1252.


Ó FEBS 2004

Homeostasis of ROS in q0 cells (Eur. J. Biochem. 271) 3655

References

of a functioning mitochondrial respiratory chain. Blood 98,
296–302.
Yoneda, M., Katsumata, K., Hayakawa, M., Tanaka, M. &
Ozawa, T. (1995) Oxygen stress induces an apoptotic cell death

associated with fragmentation of mitochondrial genome. Biochem.
Biophys. Res. Commun. 209, 723–729.
Cai, J., Wallace, D.C., Zhivotovsky, B. & Jones, D.P. (2000)
Separation of cytochrome c-dependent caspase activation from
thiol-disulfide redox change in cells lacking mitochondrial DNA.
Free Radic. Biol. Med. 29, 334–342.
Jackson, M.J., Papa, S., Bolanos, J., Bruckdorfer, R., Carlsen, H.,
Elliott, R.M., Flier, J., Griffiths, H.R., Heales, S., Holst, B.,
Lorusso, M., Lund, E., Oivind Moskaug, J., Moser, U., DiPaola,
M., Polidori, M.C., Signorile, A., Stahl, W., Vina-Ribes, J. &
Astley, S.B. (2002) Antioxidants, reactive oxygen and nitrogen
species, gene induction and mitochondrial function. Mol. Aspects
Med. 23, 209–285.
Dunbar, D.R., Moonie, P.A., Jacobs, H.T. & Holt, I.J. (1995)
Different cellular backgrounds confer a marked advantage to
either mutant or wild-type mitochondrial genomes. Proc. Natl
Acad. Sci. USA 92, 6562–6566.
Holt, I.J., Dunbar, D.R. & Jacobs, H.T. (1997) Behaviour of a
population of partially duplicated mitochondrial DNA molecules
in cell colture: segregation, maintenance and recombination
dependent upon nuclear background. Hum. Mol. Genet. 6, 1251–
1260.
Vergani, L., Prescott, A. & Holt, I.J. (2000) Rhabdomyosarcoma
q0 cells: isolation and characterisation of a mitochondrial DNA
depleted cell line with Ômuscle-likeÕ properties. Neuromuscul.
Disord. 10, 454–459.
Longo, V.D., Liou, L.L., Valentine, J.S. & Gralla, E.B. (1999)
Mitochondrial superoxide decreases yeast survival in stationary
phase. Arch. Biochem. Biophys. 365, 131–142.
Kornberg, A. (1955) Lactic dehydrogenase of muscle. Methods

Enzymol. 1, 441–443.
Srere, P.A. (1969) Citrate synthase. Methods Enzymol. 13, 3–5.
Anderson, M.E. (1985) Determination of glutathione and glutathione disulfide in biological samples. Methods Enzymol. 113,
548–555.
Floreani, M., Petrone, M., Debetto, P. & Palatini, P. (1997) A
comparison between different methods for the determination of
reduced and oxidized glutathione in mammalian tissue. Free
Radic. Res. 26, 449–455.
Bradford, M.M. (1976) A rapid and sensitive method for the
quantitation of microgram quantities of protein utilizing the
principle of protein-dye binding. Anal. Biochem. 72, 248–254.
Prohaska, J.R. & Ganther, H.E. (1976) Selenium and glutathione
peroxidase in developing rat brain. J. Neurochem. 27, 1379–1387.
Carlberg, I. & Mannervik, B. (1974) Purification and characterisation of the flavoenzyme glutathione reductase from rat liver.
J. Biol. Chem. 250, 5475–5480.
Aebi, H. (1984) Catalase in vitro. Methods Enzymol. 105, 121–126.
Siemankowsky, L.M., Morreale, J. & Briehl, M.M. (1999) Antioxidant defences in the TNF-treated MCF-7 cells: selective
increase in MnSOD. Free Radic. Biol. Med. 26, 919–924.
Oberley, L.W. & Spitz, D.Z. (1984) Assay of superoxide dismutase
activity in tumor tissue. Methods Enzymol. 105, 457–464.
Habig, W.H., Pabst, M.J. & Jakoby, W.B. (1974) Gluthatione-Stransferase: the first enzymatic step in mercapturic acid formation.
J. Biol. Chem. 249, 7130–7139.
Bianchi, A., Becuwe, P., Franck, P. & Dauca, M. (2002) Induction
of MnSOD gene by arachidonic acid is mediated by reactive
oxygen species and p38 MAPK signaling pathway in human
HepG2 hepatoma cells. Free Radic. Biol. Med. 32, 1132–1142.
Gardner, P.R. (2002) Aconitase: sensitive target and measure of
superoxide. Methods Enzymol. 349, 9–23.

1. Halliwell, B. & Gutterdge, J.M.C. (1999) Free Radicals in Biology

and Medicine, 3rd edn. Oxford University Press, New York.
2. Maxwell, S.R. (1995) Prospects for the use of antioxidant therapies. Drugs 49, 345–361.
3. Finkel, T. (2003) Oxidant signals and oxidative stress. Curr. Opin.
Cell. Biol. 15, 247–254.
4. Sauer, H., Wartenberg, M. & Hescheler, J. (2001) Reactive Oxygen Species as intracellular messengers during cell growth and
differentiation. Cell Physiol. Biochem. 11, 173–186.
5. Dalton, T.P., Shertzer, H.G. & Puga, A. (1999) Regulation of gene
expression by reactive oxygen. Annu. Rev. Pharmacol. Toxicol. 39,
67–101.
6. Forman, H.J., Torres, M. & Fukuto, J. (2002) Redox signaling.
Mol. Cell. Biochem. 234–235, 49–62.
7. Chance, B., Sies, H. & Boveris, A. (1979) Hydroperoxide metabolism in mammalian organs. Physiol. Rev. 59, 527–605.
8. Hansfort, R.G., Hogue, B.A. & Mildaziene, V. (1997) Dependence
of H2O2 formation by rat heart mitochondria on substrate avaibility and donor age. J. Bioenerg. Biomembr. 29, 89–95.
9. 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.
10. Cadenas, E. & Davies, K.J.A. (2000) Mitochondrial free radical
generation, oxidative stress, and aging. Free Radic. Biol. Med. 29,
222–230.
11. Fleury, C., Mignotte, B. & Vayssiere, J.L. (2002) Mitochondrial
reactive oxygen species in cell death signaling. Biochimie 84,
131–141.
12. Berridge, M.V. & Tan, A.N.S. (2000) Cell-surface NAD(P)Hoxidase: relationship to trans-plasma membrane NADH-oxidoreductase and a potential source of circulating NADH-oxidase.
Antioxid. Redox Signal. 2, 277–288.
13. Shen, J., Khan, N., Lewis, L.D., Armand, R., Grinberg, O.,
Demidenko, E. & Swartz, H. (2003) Oxygen consumption rates
and oxygen concentration in molt-4 cells and their mtDNA depleted (q0) mutants. Biophys. J. 84, 1291–1298.
14. King, M.P. & Attardi, G. (1989) Human cells lacking mtDNA:
repopulation with exogenous mitochondria by complementation.

Science 246, 500–503.
15. Jazayeri, M., Andreyev, A., Will, Y., Ward, M., Anderson, C.M.
& Clevenger, W. (2003) Inducible expression of a dominant negative DNA polymerase-c depletes mitochondrial DNA and
produces a q0 phenotype. J. Biol. Chem. 278, 9823–9830.
16. Park, S.Y., Chang, I., Kim, J.Y., Kang, S.W., Park, S.H., Sing, K.
& Lee, M.S. (2004) Resistance of mitochondrial DNA-depleted
cells against cell death: role of mitochondrialsuperoxide dismutase.
J. Biol. Chem. 279, 7512–7520.
17. Chandel, N.S. & Schumacker, P.T. (1999) Cells depleted of
mitochondrial DNA (q0) yield insight into physiological mechanisms. FEBS Lett. 454, 173–176.
18. Larm, J.A., Vaillant, F., Linnane, A.W. & Lawen, A. (1994)
Up-regulation of the plasma membrane oxidoreductase as a prerequisite for the viability of human Namalwa q0 cells. J. Biol.
Chem. 269, 30097–30100.
19. Chandel, N.S., Maltepe, E., Goldwasser, E., Mathieu, C.E.,
Simon, M.C. & Schumacker, P.T. (1998) Mitochondrial reactive
oxygen species trigger hypoxia-induced transcription. Proc. Natl
Acad. Sci. USA 95, 11715–11720.
20. Srinivas, V., Leshchinsky, I., Sang, N., King, M.P., Minchenko,
A. & Caro, J. (2001) Oxygen sensing and HIF-1 activation does
not require an active mitochondrial respiratory chain electrontransfer pathway. J. Biol. Chem. 276, 21995–21998.
21. Vaux, E.C., Metzen, E., Yeates, K.M. & Ratcliffe, P.J. (2001)
Regulation of hypoxia-inducible factor is preserved in the absence

22.

23.

24.

25.

26.

27.

28.

29.
30.
31.

32.

33.

34.
35.

36.
37.

38.
39.

40.

41.


3656 L. Vergani et al. (Eur. J. Biochem. 271)

42. Boveris, A., Alvarez, S., Bustamante, J. & Valdez, L. (2002)
Measurement of superoxide radical and hydrogen peroxide production in isolated cells and subcellular organelles. Methods
Enzymol. 349, 280–287.
43. Pani, G., Colavitti, R., Bedogni, B., Anzevino, R., Borrello, S.
& Galeotti, T. (2002) Determination of intracellular reactive
oxygen species as function of cell density. Methods Enzymol. 352,
91–100.
44. Zuo, L. & Clanton, T.L. (2002) Detection of reactive oxygen and
nitrogen species in tissues using redox-sensitive fluorescent probes.
Methods Enzymol. 352, 307–325.
45. Schafer, F.Q. & Buettner, G.R. (2001) Redox environment of the
cell as viewed through the redox state of glutathione disulfide/
gluthatione couple. Free Radic. Biol. Med. 30, 1191–1212.
46. Bulteau, A.L., Ikeda-Saito, M. & Szweda, L.I. (2003) Redoxdependent modulation of aconitase activity in intact mitochondria. Biochemistry 42, 14846–14855.
47. Atamna, H., Paler-Mertinez, A. & Ames, B.N. (2000) N-t-Butyl
hydroxylamine, a hydrolysis product of a-phenyl-N-t-butyl nitrone, is more potent in delaying senescence in human lung fibroblasts. J. Biol. Chem. 275, 6741–6748.
48. Atamna, H., Robinson, C., Ingersoll, R., Elliott, H. & Ames, B.N.
(2001) N-t-Butyl hydroxylamine is an antioxidant that reverses
age-related changes in mitochondria in vivo and in vitro. FASEB
J. 15, 2196–2204.
49. Lieber, M., Smith, B., Szakal, A., Nelson-Rees, W. & Todaro, G.
(1976) A continous tumor-cell line from a human lung carcinoma
with properties of type II alveolar epithelial cells. Int. J. Cancer 17,
62–70.
50. Franek, W.R., Horowitz, S., Stansberry, L., Kazzaz, J.A., Koo,
H.C., Li, Y., Arita, Y., Davis, J.M., Mantell, A.S., Scott, W. &
Mantell. L.L. (2001) Hyperoxia inhibits oxidant-induced apoptosis in lung epithelial cells. J. Biol. Chem. 276, 569–575.
51. Allalunis-Turner, M.J., Lee, F.Y. & Siemann, D.W. (1988)
Comparison of glutathione levels in rodent and human tumor cells
grown in vitro and in vivo. Cancer Res. 48, 3657–3660.

52. Dickinson, D.A. & Forman, H.J. (2002) Cellular glutathione and
thiol metabolism. Biochem. Pharmacol. 64, 1019–1026.
9 53. Kamata, H. & Hirata, H. (1999) Redox regulation of cellular
signalling. Cell Signal. 11, 1–14.
54. Brigelius-Flohe, R. (1999) Tissue-specific functions of individual
glutathione peroxidases. Free Radic. Biol. Med. 27, 951–965.

Ó FEBS 2004
55. Brigelius-Flohe, R., Maurer, S., Lotzer, K., Bol, G., Kallionpaa,
H., Lehtolainen, P., Viita, H. & Yla-Herttuala, S. (2000) Overexpression of PHGPx inhibits hydroperoxide-induced oxidation,
NFkappaB activation and apoptosis and affects oxLDL-mediated
proliferation of rabbit aortic smooth muscle cells. Atherosclerosis
152, 307–316.
56. Meyer, M., Schreck, R. & Baeuerle, P.A. (1993) H2O2 and antioxidants have opposite effects on activation of NF-kappa B and
AP-1 in intact cells: AP-1 as secondary antioxidant-responsive
factor. EMBO J. 12, 2005–2015.
57. Chen, Q., Olashaw, N. & Wu, J. (1995) Participation of reactive
oxygen species in the lysophosphatidic acid-stimulated mitogenactivated protein kinase activation pathway. J. Biol. Chem. 270,
28499–28502.
58. Curtin, J.F., Donovan, M. & Cotter, T.G. (2002) Regulation and
measurement of oxidative stress in apoptosis. J. Immunol. Meth.
265, 49–72.
59. Meister, A. (1983) Selective modification of glutathione metabolism. Science 220, 472–477.
60. Carney, J.M., Starke-Reed, P.E., Oliver, C.N., Landum, R.W.,
Cheng, M.S., Wu, J.F. & Floyd, R.A. (1991) Reversal of agerelated increase in brain protein oxidation, decrease in enzyme
activity, and loss in temporal and spatial memory by chronic
administration of the spin trapping compound N-tert-butyl-aphenylnitrone. Proc. Natl Acad. Sci. USA 88, 3633–3636.
61. Lee, J.H. & Park, J.W. (2003) Protective role of a-phenyl–N-tbutylnitrone against ionizing radiation in U937 cells and mice.
Cancer Res. 63, 6885–6893.
62. Sandy, J., Davies, M., Prime, S. & Farndale, R. (1998) Signal

pathways that transduce growth factor-stimulated mitogenesis in
bone cells. Bone 23, 17–26.
63. Cross, A.R. & Jones, O.T.G. (1991) Enzymic mechanisms of superoxide production. Biochim. Biophys. Acta 1057, 281–298.
64. Guidot, D.M., Repine, J.E., Kitlowski, A.D., Flores, S.C., Nelson, S.K., Wright, R.M. & McCord, J.M. (1995) Mitochondrial
respiration scavenges extramitochondrial superoxide anion via a
nonenzymatic mechanism. J. Clin. Invest. 96, 1131–1136.
65. Hoffmann, S., Spitkovsky, D., Radicella, J.P., Epe, B., & Wiesner,
R.J. (2004) Reactive oxygen species derived from the mitochondrial respiratory chain are not responsible for the basal levels of
oxidative base modifications observed in nuclear, DNA, of
mammalian cells. Free Radic. Biol. Med. 36, 765–773.