In the present study, by exposing lysosome-rich macrophage-like J774 cells to oxidative stress, either in the presence of sih or following pretreatment with dfo, we find strong evidence f
Trang 1stress-induced cellular damage
Tino Kurz1, Bertil Gustafsson2and Ulf T Brunk1
1 Division of Pharmacology, Faculty of Health Sciences, Linko¨ping University, Sweden
2 Department of Pathology and Cytology, University Hospital, Linko¨ping, Sweden
Exposing cells in culture to increasing oxidative stress
triggers a range of cellular events Depending on the
cell type, these may include enhanced proliferation or
growth arrest, DNA damage, protein and lipid
oxida-tion, apoptosis, and finally necrosis [1] This points to
an important physiological role of redox regulation in
cellular homeostasis [2,3] While most current studies
suggest a direct effect of oxidative stress on DNA and
mitochondria followed by apoptosis or necrosis, recent
research has established a critical role for lysosomes in
the initiating phase of impairment [4–13] Because hydrogen peroxide, added in moderate concentrations
as a bolus, may be consumed within minutes [14] and most cellular alterations, including apoptosis, do not occur until hours later, any satisfactory hypothesis on the mechanisms behind oxidative stress-induced cellular damage must provide firm and distinct links between the triggering events and the ultimate cellular injuries There are indications that hydrogen peroxide per se has little harmful effects and that an interaction
Keywords
cell death; lysosomes; mitochondria;
redox-active iron; salicylaldehyde isonicotinoyl
hydrazone
Correspondence
U T Brunk, Department of Pharmacology,
University Hospital, SE-581 85 Linko¨ping,
Sweden
Fax: +46 13 149106
Tel: +46 13 221515
E-mail: Ulf.Brunk@imv.liu.se
(Received 1 March 2006, revised 28 April
2006, accepted 15 May 2006)
doi:10.1111/j.1742-4658.2006.05321.x
Oxidant-induced cell damage may be initiated by peroxidative injury to lysosomal membranes, catalyzed by intralysosomal low mass iron that appears to comprise a major part of cellular redox-active iron Resulting relocation of lytic enzymes and low mass iron would result in secondary harm to various cellular constituents In an effort to further clarify this still controversial issue, we tested the protective effects of two potent iron chelators – the hydrophilic desferrioxamine (dfo) and the lipophilic salicyl-aldehyde isonicotinoyl hydrazone (sih), using cultured lysosome-rich macro-phage-like J774 cells as targets dfo slowly enters cells via endocytosis, while the lipophilic sih rapidly distributes throughout the cell Following dfo treatment, long-term survival of cells cannot be investigated because dfo by itself, by remaining inside the lysosomal compartment, induces apoptosis that probably is due to iron starvation, while sih has no lasting toxic effects if the exposure time is limited Following preincubation with
1 mm dfo for 3 h or 10 lm sih for a few minutes, both agents provided strong protection against an ensuingLD50 oxidant challenge by prevent-ing lysosomal rupture, ensuprevent-ing loss of mitochondrial membrane potential, and apoptotic⁄ necrotic cell death It appears that once significant lysosomal rupture has occurred, the cell is irreversibly committed to death The results lend strength to the concept that lysosomal membranes, normally exposed to redox-active iron in high concentrations, are initial targets of oxidant damage and support the idea that chelators selectively targeted to the lysosomal compartment may have therapeutic utility in diminishing oxidant-mediated cell injury
Abbreviations
AO, acridine orange base; dfo, desferrioxamine; pHPA, p-hydroxy-phenylacetic acid; PI, propidium iodide; sih, salicylaldehyde isonicotinoyl hydrazone (N¢-[(1Z)-(2-hydroxyphenyl)methylene]isonicotinohydrazide); SSM, sulfide-silver method; TMRE, tetramethylrhodamine ethyl ester;
Y m , mitochondrial membrane potential.
Trang 2together with late endosomes constitute the acidic
vac-uolar compartment, are the main cellular structures for
normal autophagic turnover of organelles and
long-lived proteins [20–22] Autophagic degradation of
fer-ruginous material, such as ferritin and cytochromes, is
responsible for the intralysosomal occurrence of
redox-active low molecular weight iron [23] before it is
trans-ported out of the lysosomal compartment for use in a
variety of anabolic processes, e.g synthesis of
iron-containing macromolecules, while excess iron is stored
in ferritin This, along with the participation of iron
in Fenton-type reactions producing hydroxyl radicals
(HO•), or similarly reactive iron-centered
[oxido-iron(IV)] radicals, would account for the sensitivity of
lysosomes to oxidative stress that, if intense enough,
may result in lysosomal rupture and release to the
cytosol of harmful contents with ensuing cellular
dam-age, including apoptosis and necrosis [4–6,24,25]
Although Christian de Duve, the discoverer of
lyso-somes, envisaged such a possibility by nicknaming
lysosomes ‘suicide bags’ [26], lysosomes are today
often – although wrongly, we believe – considered to
be sturdy organelles that usually do not rupture until
cells are already dead and necrotic
We have previously shown that cells exposed for
1–3 h to high (‡ 1 mm) concentrations of
desferriox-amine, dfo (either in free form or as a high molecular
weight conjugate to starch, HMW-dfo), are
substan-tially, although not fully, protected against oxidative
stress [10] dfo (or HMW-dfo), being a strong
hydro-philic iron chelator that firmly binds all six
coordi-nates of iron, preventing its iron(II)⁄ iron(III)
redox-cycling, does not pass membranes but is fluid
phase-endocytosed by cells in culture, passes through
late endosomes and, because of the extensive fusion
and fission activities of the lysosomal compartment, is
transferred to most lysosomes [13,27–30] As dfo
remains intralysosomal, it will act as a sink for
cellu-lar iron in transit through the lysosomal compartment
and, thus, within hours, cells will start to become
affected by iron starvation and finally die [5,12]
Inter-estingly, dfo-induced apoptosis, and a variety of other
apoptogenic stimuli, involves lysosomal
destabiliza-tion, suggesting this phenomenon to be related to
apoptosis in general and not only to oxidative stress
[5,6,31,32]
Recently, it was demonstrated that the lipophilic
iron chelator sih (salicylaldehyde isonicotinoyl
hydra-thesized in 1953 [34] and its iron binding ability was demonstrated 20 years ago [35] Its binding constant for Fe(III) is 1050 at pH 7.4 [36] In the present study, by exposing lysosome-rich macrophage-like J774 cells to oxidative stress, either in the presence
of sih or following pretreatment with dfo, we find strong evidence for a primary role of lysosomal redox-active iron in oxidative stress-induced cell dam-age and a close correlation between initial lysosomal rupture and later development of cellular damage and death
Results
Hydrogen peroxide-induced mitochondrial injury and cell death are downstream effects of
lysosomal rupture
In order to confirm and add to earlier findings on the sequence of events with respect to lysosomal rupture, mitochondrial injury and apoptosis⁄ necrosis following oxidative stress [37], we first assessed lysosomal stabil-ity by cytofluorometric evaluation of alterations in green and red fluorescence, respectively, of cells vitally stained with acridine orange (AO) before (AO-reloca-tion test [6,10,13,38–40]) and at 6 h after the oxidative stress period (AO-uptake test [6,10,13,32,40,41]) AO is
a weak base (pKa 10) that, due to proton trapping, preferentially distributes within the acidic vacuolar (lysosomal) cellular compartment [4–6,24,25,42–44] Due to its metachromatic properties, this probe fluor-esces red inside lysosomes, where it is highly concen-trated, and weakly green in the cytosol and the nucleus, where it is much less concentrated When used
as a vital stain at low concentrations, the intercalation
of AO into RNA and DNA is very low and does not disturb the evaluation of lysosomal stability Ordinary photomultipliers are about 10-fold more sensitive to green than to red photons, making the AO-relocation test useful for the early detection of a limited number
of ruptured lysosomes
As shown in Fig 1, unprotected cells showed lyso-somal destabilization that was detectable by the AO-relocation method as early as 15 min following the end of the oxidative stress period (peroxidative lyso-somal membrane damage develops by time) Using the AO-uptake technique, unprotected cells showed a sub-stantial increase in ‘pale’ cells (cells with a reduced
Trang 3number of intact lysosomes) after 6 h (Fig 2), when
these cells started to show apoptotic alterations as
des-cribed previously [22] Cells protected by the lipophilic
iron chelator sih, or the hydrophilic iron chelator dfo
were highly protected against both early and late lyso-somal rupture [13]
dfo is known to induce iron starvation and related cell death and cannot be used in long-term experiments
Fig 1 Lysosomal rupture is an early event after oxidative stress Acridine orange (AO)-relocation assay Cells (10 6 ) were preloaded with the metachromatic fluorophore and lysosomotropic base AO Following two washing steps in culture medium, they were then exposed for 30 min to 100 l M
H2O2in 2 mL NaCl ⁄ P i with ⁄ without 10 l M sih and returned to standard culture condi-tions for 15 min Lysosomal stability was assayed by the AO-relocation technique using flow cytofluorometry in the green FL1 channel Due to release of AO from rup-tured lysosomes into the cytoplasm, oxida-tive stress resulted in an early (15 min) increase of the mean value for the green cytoplasmic fluorescence that was significantly prevented by sih-protection (mean ± SD; ***P < 0.001; n ¼ 8) Examples of green fluorescence histograms are shown above each bar.
Fig 2 Iron chelation protects against lyso-somal rupture by oxidative stress Acridine orange (AO)-uptake assay Cells, either pro-tected by sih or not, were exposed to oxida-tive stress as described for Fig 1 Other cells were initially pretreated for 3 h with
1 m M dfo under otherwise standard culture conditions before exposure to the same oxi-dative stress Following end of the stress period, cells were returned to standard cul-ture conditions for an additional period of
6 h when lysosomal stability was assessed using the AO-uptake method Ruptured lysosomes do not take up AO resulting in a population of cells with reduced red fluores-cence (‘pale’ cells) The number of ‘pale’ cells was reduced highly significantly in cells protected by sih or dfo (mean ± SD;
***P < 0.001; n > 6) For each bar a repre-sentative histogram of red fluorescence, with ‘pale cells’ gated, is given.
Trang 4[5] dfo is taken up by endocytosis [27,28], although
this is not generally recognized (often it is just
consid-ered to pass membranes very slowly), remains
intra-lysosomal and causes iron starvation [5,12] Thus, only
the lipophilic sih, which quickly redistributes, was used
in the remaining experiments As is evident from
Fig 3, mitochondrial membrane potential was
pre-served by sih protection, indicating (as suggested
before [6,22,37,40,45]) that mitochondrial damage is
secondary to lysosomal rupture and related to
iron-cat-alyzed intralysosomal peroxidation
Similarly, sih-protection prevented the oxidative
stress-induced decline in cell numbers (Fig 4) and
pos-tapoptotic necrosis (Fig 5) Actually, sih-protected
cells multiplied similarly to control cells if exposed to
sih only during the oxidative stress period (Fig 4),
while cells exposed to >5 lm sih for long periods of
time finally all died by iron-starvation (results not
shown)
Cells degraded hydrogen peroxide as shown before
[22], and pretreatment with dfo or sih did not influence
the rate (results not shown)
In all experiments, cells exposed to sih without
oxi-dative stress behaved like unexposed controls
Fig 3 Disruption of mitochondrial
mem-brane potential is a down-stream effect of
lysosomal rupture Cells, protected against
oxidative stress by sih or not, were exposed
to 100 n M tetramethylrhodamine ethyl ester
(TMRE) for 15 min under standard culture
conditions 1–8 h following the oxidative
stress period Red fluorescence was
ana-lyzed in the FL 3 channel by flow
cytofluor-ometry Damaged mitochondria with
depolarized membranes show reduced
TMRE uptake In unprotected cultures,
sig-nificant mitochondrial damage was observed
only 6 and 8 h after end of the oxidative
stress period, while sih-protected cells
showed almost no increase in damaged
mitochondria (mean ± SD; ***P < 0.001;
**P < 0.01; n ¼ 4) At top of the panel,
examples of histograms are given showing
red fluorescence 8 h following end of
oxida-tive stress Cells with reduced red
fluores-cence were gated.
Fig 4 sih-protected cells retain normal proliferation capacity follow-ing oxidative stress Cells were seeded (500 000 ⁄ well) and 24 h later exposed to oxidative stress, with or without sih-protection as described before Directly after end of the oxidative stress period (0 h) and again after return to standard culture conditions for another 12 and 24 h, cells were fixed in 4% formaldehyde in NaCl ⁄ P i and counted in five predefined areas per dish (mean ± SD;
***P < 0.001; **P < 0.01; *P < 0.05; n.s., nonsignificant; n ¼ 4) After a short lag-phase, sih-protected cells continued to proliferate normally, while about half of the unprotected cells underwent apop-totic ⁄ necrotic cell death.
Trang 5Cellular labile iron is located predominantly
inside lysosomes
To prove that cellular labile iron is located
predomin-antly inside lysosomes, we utilized the sulfide-silver
method (SSM), which is a very sensitive cytochemical
technique to demonstrate heavy metals [18,41,46] As
iron is the dominating heavy metal in most normal
cells (the exceptions being some zinc-containing
neu-rons and endocrine cells), it can be considered specific
for this transition metal As shown in Fig 6 (and our
previous publications, reviewed in [45]), the outcome
of the reaction is a mainly granular staining with a
dis-tinct lysosomal pattern when the SSM is applied to
normal J774 cells in culture The SSM is an
auto-cata-lytic procedure and a short development time results in
few stained lysosomes, while a longer development
results in a more general staining of lysosomes (Fig 6,
compare parts A, B and C) This shows that the
amount of redox-active iron varies between lysosomes,
probably depending on the extent of recent
engage-ment in autophagic degradation of ferruginous
mater-ial [22,47], explaining why there is a pronounced
heterogeneity between lysosomes with respect to
sensi-tivity to oxidative stress [47]
The fact that some cells and individual lysosomes
resist oxidative stress better than others is considered
an important and not well understood phenomenon
[48] We have previously suggested that the difference
in cellular and lysosomal amount of redox-active iron
could be a major cause [47]
In order to show that the SSM does indeed demon-strate iron, we exposed some cell cultures to 30 lm FeCl3 for 3 h before the SSM In the culture medium, iron ions complex with phosphate groups under the formation of a hydrated iron phosphate complex, which is endocytosed by cells and transported to the lysosomal compartment [22,41] Following such iron uptake, a lysosomal staining pattern was obvious already after 30 min development time, while the con-trol cells showed no labeling (Fig 6A and D)
Discussion
Using the cytochemical SSM and the calcein technique
in combination with induction of lysosomal rupture,
we have here and previously demonstrated that lyso-somes contain a major part of the cellular redox-active low mass iron, making the lysosomal compartment notably vulnerable to oxidative stress [12,18,22,41] The lysosomal concentration of low mass iron differs between the lysosomes of individual cells as well as between different cells, probably reflecting the partici-pation of individual lysosomes in the autophagic deg-radation of ferruginous material These variations may explain the obvious differences in the sensitivity to oxi-dative stress of individual lysosomes of the same cell and of different cells of the same population [47,48] Lysosomes are a heterogeneous group of vesicular structures and, at a given point in time, some lyso-somes are performing degradation, while others are
‘resting’ [20–22,47] Those engaged in the degradation
Fig 5 sih-protected cells show no decrease
in viability following oxidative stress Cells were seeded (500 000 ⁄ well) and 24 h later exposed protected or unprotected to oxida-tive stress as described before and were then returned to standard culture conditions for another 24 h The cells were then scraped, exposed to 40 lgÆmL)1propidium iodide for 90 min, centrifuged and washed
in NaCl ⁄ P i ⁄ centrifuged twice Red fluores-cence of PI-stained nuclei was measured by flow cytofluorometry in the FL3channel (mean ± SD; ***P < 0.001; n ¼ 3) sih strongly protected cells against oxidative stress-induced postapoptotic necrosis Examples of red fluorescence histograms are given PI-positive cells were gated.
Trang 6of iron-containing macromolecules may contain a high
concentration of low-mass iron, while a ‘resting’
lyso-some would contain almost none [21,49,50], explaining
our finding (Fig 6) that some lysosomes contain much
more iron than others
Although iron is an essential transition metal
required for many vital functions, including electron
transport, it is potentially dangerous because of
its capacity to participate in Fenton-type reactions
(Eqn 1)
Fe2þþ H2O2! Fe3þþ HOþ OH ð1Þ
Hydroxyl radicals (HO•) are short lived (10)9s),
extremely reactive, and able to bring about oxidative
injury to a variety of biomolecules Consequently, cells
and organisms handle iron with great care and usually
hide it within stable metallo-organic complexes,
thereby preventing hydrogen peroxide from
encounter-ing redox-active iron; an exception beencounter-ing low mass
iron in late endosomes and lysosomes as well as,
per-haps, a small amount of low mass iron in transit from
these structures for storage in ferritin or use in the
syn-thesis of iron-containing macromolecules [10–12]
As lysosomes break under oxidative stress, as proven here and earlier with the AO-relocation and -uptake tests [6,10,13,32,38–41], we may assume that a certain fraction of lysosomal low mass iron exists in iron(II) form Most probably this is related to the low lyso-somal pH and presence of reducing equivalents, such
as cysteine [51–53] At pH 5, cysteine reduces iron(III)
to iron(II), as has been shown before [22] This reduc-tion of iron (Eqn 2) is driven by the removal of Cys-S• (Eqn 3) and the subsequent reduction of dioxygen (Eqn 4) [54,55]
Cys SH þ Fe3þ! Cys Sþ Fe2þþ Hþ ð2Þ Cys Sþ Cys SH ! ðCys S S CysÞþ Hþ ð3Þ ðCys S S CysÞþ O2! Cys S S Cys þ O2 ð4Þ Autophagy is a normal and continuously ongoing process that allows a fine-tuned turnover of organelles and most long-lived macromolecules being of major importance for intracellular iron turnover [21,49,50,56] Due to intralysosomal degradation of a large variety of biomolecules, including many ferruginous materials,
Fig 6 Cytochemical demonstration of iron by the sulfide-silver method (SSM) With increasing development time (30–60 min) control cells (A–C) show an increasingly intense lysosomal pattern of black granular silver precipitates, indicating the presence of lysosomal low molecular weight iron While there was no granular staining after a 30-min period of development (A), occasional granules were evident in some cells (examples are indicated by arrow heads) after 40 min (B) and a distinctly granular staining with a lysosomal pattern was found in all cells after 60 min of development (C) Cells exposed for 3 h to a hydrated iron phosphate complex (obtained by adding FeCl3to the culture med-ium to a final concentration of 30 l M ) showed many stained lysosomes after development for 30 min (D) when the control cells (A) were still empty This finding reflects the fluid phase endocytosis of the iron phosphate complex, as well as the capacity of the method to demon-strate iron.
Trang 7such as ferritin, mitochondrial complexes and various
other metalloproteins, low-mass iron is set free
intralys-osomally As an effect, a substantial part of such iron
seems to be temporarily harbored within these
organ-elles before being transported, by not yet well
character-ized carrier systems, to the cytosol and used for
anabolic purposes or stored in ferritin [11,12,22,23,57]
Differences between cells of different origin with
respect to lysosomal stability to oxidative stress may
reflect a divergence in their capacity to degrade
hydro-gen peroxide, while differences in amounts of
intra-lysosomal redox-active iron may explain intra- and
intercellular variation [47,58]
Here we exposed 106 cells in 2 mL NaCl⁄ Pi to a
bolus dose of 100 lm hydrogen peroxide that is
rap-idly and exponentially degraded under these
condi-tions Due to efficient intracellular degradation of
hydrogen peroxide, a steep gradient across the plasma
membrane is established and the actual concentration
of hydrogen peroxide sensed by the lysosomes would
probably not exceed 15 lm soon after the start of
the oxidative stress [14] As lysosomes do not contain
any catalase or glutathione peroxidase, this initial
magnitude of oxidative stress proved sufficient to
induce lysosomal rupture that initially was of limited
scale Lysosomal rupture accelerated by time through
a lipid peroxidation chain reaction until it initiated
apoptosis, probably by a direct or indirect effect on
mitochondrial stability [37,59,60] This lysosomal
rup-ture results in release to the cytosol of powerful
hydrolytic enzymes and redox-active iron that,
depending on the magnitude of the rupture, may
initi-ate a variety of cellular injuries, including reparative
autophagocytosis, apoptosis and necrosis [45] Clearly,
lysosomes are not the sturdy organelles they once
were believed to be, breaking only late during necrotic
(accidental) cell death
Oxidative stress-induced lysosomal rupture,
secon-dary mitochondrial injury and final cell death were
almost fully prevented by a pre-exposure to the potent
iron chelators sih and dfo, as was also shown
previ-ously [37,45], suggesting that the damaging effect of
hydrogen peroxide per se is of minor importance in
comparison to its iron-mediated effect on lysosomal
stability Even if the importance of lysosomal rupture
for oxidative stress-induced damage, including
apopto-sis and DNA damage, has been shown before [5,10–
13], this view is still controversial A major reason for
the present lack of general acceptance of an initiating
role of lysosomal break in oxidative stress-induced
injury may be that dfo, which often has been used to
chelate intralysosomal iron, works like a double-edged
sword It does prevent early lysosomal damage under
oxidative stress, but after some time dfo itself induces apoptosis [5] The reason for this phenomenon seems
to be that dfo following endocytotic uptake remains inside the lysosomal compartment where it scaven-ges and acts as a sink for iron that is in transit thro-ugh the compartment as a result of autophagy [11,12,21,49,50]
The lipophilic iron chelator sih does not, however, have this disadvantage sih swiftly penetrates mem-branes and binds iron very strongly throughout the cell
in a non redox-active form, also at lysosomal pH [36]
In a recent study, it was shown that in the presence of sih, cells show no mitochondrial damage following oxi-dative stress, indicating the need of free iron for such damage [15] Here we add the information that the defense is mediated by lysosomal stabilization and that cells protected by sih during oxidative stress and then brought back to normal culture conditions continue to proliferate normally Because of its lipophilicity, sih enters and leaves cells rapidly and, in contrast to dfo,
it can be used to chelate intracellular redox-active iron for a limited period of time without causing long-last-ing effects
A key aspect of oxidative regulation of physiologi-cal processes is the disparity of the time-sphysiologi-cales involved The apoptotic⁄ necrotic process takes several hours to fully develop, although cells need to be exposed to hydrogen peroxide for only a short per-iod of time to be committed to apoptosis⁄ necrosis The very sensitive AO-relocation technique to detect early lysosomal rupture allowed us to observe a strong correlation between a hydrogen peroxide-induced cellular modification that occurred rapidly,
by induction of partial lysosomal rupture, and the signs of apoptosis⁄ necrosis hours later Both proces-ses showed the same dose-dependent response to hydrogen peroxide and were inhibited by dfo- and sih-mediated iron chelation during the period of oxi-dative stress
Release of lysosomal contents initiates a process that results in mitochondrial destabilization as well as fur-ther lysosomal rupture [37] In this context, it is worth mentioning that the change in mitochondrial mem-brane potential that was demonstrated, as well as release of cytochrome c from mitochondria under sim-ilar conditions [37,61] does not occur until 1–3 h after the end of exposure to hydrogen peroxide and long after the lysosomal rupture is observed Interestingly,
we observed a progressive decrease in the number of intact lysosomes over time when the cells were no lon-ger under oxidative stress, which is in accordance with
a self-amplifying loop and cross-talk between lyso-somes and mitochondria (Fig 7), as previously pointed
Trang 8out as the foundation of the lysosomal-mitochondrial
axis theory of apoptosis [37]
Taken together, the findings of this study suggest
that the major effect of harmful oxidative stress, rather
than being a direct effect on targets such as DNA and
mitochondria, is a result of lysosomal rupture due to
intralysosomal peroxidative events, with ensuing
relo-calization of lysosomal hydrolytic enzymes and
low-mass iron
Experimental procedures
Chemicals
Dulbecco’s modified Eagle’s medium (DMEM), fetal bovine
serum, penicillin and streptomycin were from Gibco (Paisley,
UK) Acridine orange base was from Gurr (Poole, UK),
while silver lactate was from Fluka AG (Buchs, Switzerland)
Glutaraldehyde was from Bio-Rad (Cambridge, MA, USA),
and ammonium sulfide and hydroquinone were from BDH
Ltd (Poole, UK) dfo was from Ciba-Geigy (Basel,
Switzer-land) sih was a kind gift from D Richardson (University of
New South Wales, Sydney, Australia) All other chemicals
were from Sigma (St Louis, MO, USA)
Cell culture and exposure to hydrogen peroxide
with⁄ without iron chelator protection
Murine macrophage-like J774 cells (ATCC, Manassas, VA,
USA) were grown in DMEM supplemented with 10% fetal
bovine serum, 2 mm l-glutamine, 100 IUÆmL)1 penicillin and 100 lgÆmL)1 streptomycin, at 37C in humidified air with 5% CO2 The cells were subcultivated twice a week, plated at a concentration of 1· 106cells per well in six-well plates, with or without cover-slips, and typically subjected
to oxidative stress after another 24 h
Concentrations of hydrogen peroxide and exposure times (in relation to cell density) and exposure to sih and dfo were established in preliminary experiments In final experi-ments, control and chelator-protected cells were oxidatively stressed (or not) for 30 min by exposure to a bolus dose of
100 lm H2O2in 2 mL NaCl⁄ Pi at 37C Note that under these conditions the H2O2 concentration declines quickly (t1 ⁄ 2 15 min) to < 20 lm after 30 min (see below) dfo (1 mm) was added to the culture medium under otherwise standard conditions 3 h before oxidative stress sih was pre-pared as a 10 mm solution in dimethyl sulfoxide and then diluted to a 1 mm stock solution in absolute ethanol To produce a final concentration of 10 lm sih during the oxi-dative stress exposure, some of the stock was added to the NaCl⁄ Piimmediately prior to the addition of hydrogen per-oxide No sih pretreatment was found necessary After the oxidative stress period, cells were directly analyzed or returned to standard culture conditions and assayed at indi-cated periods of time
In some experiments, cells were incubated for 3 h in complete medium with FeCl3added to a concentration of
30 lm (resulting in the formation of a nonsoluble iron phosphate complex that is endocytosed and transported into the lysosomal compartment)
Fig 7 The lysosomal-mitochondrial pathway of cell death; a tentative scheme Slightly modified from Zhao et al [37], the scheme shows intralysosomal Fenton-type reactions resulting from oxidative stress Lysosomal contents are released to the cytosol following lysosomal destabilization and may activate pro-apoptotic proteins, such as Bid [59,60], and ⁄ or attack mitochondria with release of cytochrome c and enhanced production of superoxide Released lysosomal enzymes (LE) may also activate cytosolic phospholipases, which in turn may attack mitochondria and lysosomes, inducing a self-amplifying loop Released redox-active iron may bind to nuclear and mitochondrial DNA and induce site-specific damage under continuous oxidative stress [10,12].
Trang 9Degradation of hydrogen peroxide
To ensure that the observed resistance to oxidative stress
was not an effect of enhanced H2O2catabolism induced by
the iron-chelating drugs, the rate of H2O2 clearance was
determined Control cells and cells protected by the iron
chelators (in concentrations described above) were exposed
to a bolus dose of 100 lm H2O2in 2 mL NaCl⁄ Piat 37C
During a 60-min period, aliquots (50 lL) were sampled for
H2O2 analysis by the horseradish peroxidase-mediated
H2O2-dependent p-hydroxy-phenylacetic acid (pHPA)
oxi-dation technique [62] Fluorescence intensity was read
(kex315 nm; kem410 nm) using a RF-540
spectrofluoro-meter (Shimadzu, Kyoto, Japan) connected to a DR-3 data
recorder
Lysosomal membrane stability assay
Six hours after the oxidative stress period (see above), cells
were exposed to 10 lgÆmL)1acridine orange (AO) in
com-plete medium at 37C for 15 min, detached by scraping and
collected for flow cytofluorometric assessment of lysosomal
AO-uptake AO is a metachromatic fluorophore and a
lyso-somotropic base (pKa¼ 10.3), which becomes charged
(AOH+) and retained by proton trapping within acidic
com-partments, mainly secondary lysosomes (pH 4.5–5.5) When
normal cells are excited by blue light, highly concentrated
lysosomal AO emits an intense red fluorescence, while nuclei
and cytosol show weak diffuse green fluorescence In
AO-uptake experiments, red fluorescence was measured (FL3
channel) using a Becton-Dickinson FACScan
(Becton-Dick-inson, Mountain View, CA, USA) equipped with a 488 nm
argon laser Cells with a reduced number of intact,
AO-accumulating lysosomes (here termed ‘pale’ cells) were
detected as described earlier [6,10,13,32,40,41]
The AO-relocation technique [6,10,13,38–40] was used to
measure early lysosomal damage For this assay, cells were
preloaded with AO (10 lgÆmL)1) for 15 min in complete
culture medium, rinsed with culture medium and kept
under standard culture conditions for a further 15 min
before being exposed to oxidative stress After the oxidative
stress period cells were returned to standard culture
condi-tions for 15 min and then scraped and assayed by flow
cytofluorometry The increase in green cytoplasmic
fluores-cence, due to the release of AO from ruptured lysosomes,
was measured in the FL1channel cellquest software (BD
Biosciences, Franklin Lakes, NJ, USA) was used for
acqui-sition and analyses
Mitochondrial membrane potential assay
Mitochondrial membrane potential (Ym) was measured by
flow cytofluorometry, using the cationic and lipophilic
dye tetramethylrhodamine ethyl ester (TMRE), which
accumulates in the mitochondrial matrix Decreased Ym is indicated by a reduction of the TMRE-induced red fluores-cence At different points of time (1–8 h) following the end
of oxidative stress (see above), cells were incubated with TMRE in complete culture medium (100 nm; 15 min;
37C) and assayed by flow cytofluorometry Red (FL3 channel) fluorescence was recorded in a log scale and ana-lyzed using the cellquest software Cells with reduced red fluorescence were gated
Assessment of cell proliferation
Five hundred thousand cells were seeded per well and exposed for 30 min to oxidative stress (or not) 24 h later with 10 lm sih present (or not) Directly after oxidative stress, and after another 12 and 24 h under standard cul-ture conditions, cells were washed in NaCl⁄ Piand fixed in 4% formaldehyde in NaCl⁄ Pi For each condition cells were counted in five predefined areas of two separate dishes The cell proliferation experiments were done twice
Assessment of postapoptotic necrotic cells
Cells were seeded and treated as described above for the cell proliferation assay The magnitude of oxidative stress applied is known to induce apoptosis but little direct necro-sis [22] After 24 h under standard conditions following the oxidative stress, cells were scraped and exposed in the dark for 90 min to 40 lgÆmL)1 propidium iodide in complete culture medium at 22C Cells were then centrifuged and washed in NaCl⁄ Pitwice before red fluorescence was ana-lyzed by flow cytofluorometry using the FL3 channel Propidium iodide does not cross the plasma membrane of normal or early apoptotic cells, while it penetrates into pos-tapoptotic necrotic cells and binds to nuclear DNA
Cytochemical assay of lysosomal reactive iron
For evaluation of cellular low-mass iron, we used the auto-metallographic sulfide-silver method as previously described [18], modified (high pH; high S2–) from Timm [46] Cells were grown on cover-slips and exposed, or not, for 3 h to
an insoluble hydrated iron phosphate complex, obtained by addition of FeCl3 to complete culture medium to a final concentration of 30 lm Cells were rinsed briefly in NaCl⁄ Pi (22C) prior to fixation with 2% glutaraldehyde in 0.1 m sodium cacodylate buffer with 0.1 m sucrose (pH 7.2) for
2 h at 22C The fixation was followed by five short rinses
in glass-distilled water at 22C Cells were then sulfidated
at pH 9 with 1% (w ⁄ v) ammonium sulfide in 70% (v ⁄ v) ethanol for 15 min Following careful rinsing in glass-distilled water for 10 min at 22C, development was per-formed using a physical, colloid-protected developer containing silver-lactate and hydroquinon (the method is an
Trang 10mounting in Canada balsam, the cells were examined and
photographed, using transmitted light, under an Axioscope
microscope (Zeiss, Oberkochen, Germany) connected to a
Zeiss ZVS-47E digital camera easy image measurement
2000 software (version 2.3, Bergstro¨m Instruments AB,
Solna, Sweden) was used for image acquisition
Statistical analysis
Results are given as mean ± SD Statistical comparisons
were made using analysis of variance (anova), whereby
pair-wise multiple comparisons were made using Tukey’s
adjustment For comparison of two means, Student’s t-test
was used P<0.05 (*), P<0.01 (**), P<0.001(***)
Acknowledgements
This study was supported by Linko¨ping University
Hospital Funds to BG and UTB sih was a kind gift
from Professor Des Richardson, University of Sydney,
NSW, Australia
References
1 Dypbukt JM, Ankarcrona M, Burkitt M, Sjo¨holm A˚,
Stro¨m K, Orrenius S & Nicotera P (1994) Different
prooxidant levels stimulate growth, trigger apoptosis, or
produce necrosis of insulin-secreting RINm5F cells The
role of intracellular polyamines J Biol Chem 269,
30553–30560
2 Burdon RH (1995) Superoxide and hydrogen peroxide
in relation to mammalian cell proliferation Free Radic
Biol Med 18, 775–794
3 Davies KJ (1999) The broad spectrum of responses to
oxidants in proliferating cells: a new paradigm for
oxi-dative stress IUBMB Life 48, 41–47
4 Hellquist HB, Svensson I & Brunk UT (1997) Oxidant
induced apoptosis: a consequence of lethal lysosomal
leak? Redox Report 3, 65–70
5 Brunk UT & Svensson I (1999) Oxidative stress, growth
factor starvation and Fas activation may all cause
apop-tosis through lysosomal leak Redox Rep 4, 3–11
6 Li W, Yuan X, Nordgren G, Dalen H, Dubowchik
GM, Firestone RA & Brunk UT (2000) Induction of
cell death by the lysosomotropic detergent MSDH
FEBS Lett 470, 35–39
7 Vancompernolle K, Van Herreweghe F, Pynaert G, Van
de Craen M, De Vos K, Totty N, Sterling A, Fiers W,
Vandenabeele P & Grooten J (1998)
Atractyloside-435, 233–236
9 Ishisaka R, Utsumi T, Kanno T, Arita K, Katunuma
N, Akiyama J & Utsumi K (1999) Participation of a cathepsin 1-type protease in the activation of caspase-3 Cell Struct Funct 24, 465–470
10 Kurz T, Leake A, Von Zglinicki T & Brunk UT (2004) Relocalized redox-active lysosomal iron is an important mediator of oxidative-stress-induced DNA damage Biochem J 378, 1039–1045
11 Doulias PT, Christoforidis S, Brunk UT & Galaris D (2003) Endosomal and lysosomal effects of desferrioxa-mine: protection of HeLa cells from hydrogen peroxide-induced DNA damage and induction of cell-cycle arrest Free Radic Biol Med 35, 719–728
12 Tenopoulou M, Doulias PT, Barbouti A, Brunk U & Galaris D (2005) Role of compartmentalized redox-active iron in hydrogen peroxide-induced DNA damage and apoptosis Biochem J 387, 703–710
13 Persson HL, Yu Z, Tirosh O, Eaton JW & Brunk UT (2003) Prevention of oxidant-induced cell death by lyso-somotropic iron chelators Free Radic Biol Med 34, 1295–1305
14 Antunes F & Cadenas E (2000) Estimation of H2O2 gradients across biomembranes FEBS Lett 475, 121– 126
15 Sˇimunek T, Boer C, Bouwman RA, Vlasblom R, Ver-steilen AM, Sˇterba M, Gersˇl V, Hrdina R, Ponka P, de Lange JJ et al (2005) SIH – a novel lipophilic iron che-lator – protects H9c2 cardiomyoblasts from oxidative stress-induced mitochondrial injury and cell death
J Mol Cell Cardiol 39, 345–354
16 Brun A & Brunk U (1970) Histochemical indications for lysosomal localization of heavy metals in normal rat brain and liver J Histochem Cytochem 18, 820–827
17 Starke PE, Gilbertson JD & Farber JL (1985) Lysoso-mal origin of the ferric iron required for cell killing by hydrogen peroxide Biochem Biophys Res Commun 133, 371–379
18 Zdolsek JM, Roberg K & Brunk UT (1993) Visualiza-tion of iron in cultured macrophages: a cytochemical light and electron microscopic study using autometallo-graphy Free Radic Biol Med 15, 1–11
19 Petrat F, de Groot H & Rauen U (2001) Subcellular distribution of chelatable iron: a laser scanning micro-scopic study in isolated hepatocytes and liver endothe-lial cells Biochem J 356, 61–69
20 Levine B & Klionsky DJ (2004) Development by self-digestion: molecular mechanisms and biological func-tions of autophagy Dev Cell 6, 463–477