Carcinogenesis, 2015, Vol. 36, No. 3, 400–411
doi:10.1093/carcin/bgv010
Advance Access publication February 3, 2015
Original Manuscript
original manuscript
Direct and indirect inactivation of tumor cell protective
catalase by salicylic acid and anthocyanidins reactivates
intercellular ROS signaling and allows for synergistic
effects
Katrin Scheit1,2 and Georg Bauer1,*
1
2
Department of Medical Microbiology and Hygiene, Institute of Virology, University Medical Center, Freiburg, 79104, Germany
Present address: Klinik Immenstadt, 87509 Immenstadt, Germany
*To whom correspondence should be addressed. Tel.: +49 761 203 6618; Fax: +49 761 203 6562; Email: [email protected]
Abstract
Salicylic acid and anthocyanidins are known as plant-derived antioxidants, but also can provoke paradoxically seeming
prooxidant effects in vitro. These prooxidant effects are connected to the potential of salicylic acid and anthocyanidins to
induce apoptosis selectively in tumor cells in vitro and to inhibit tumor growth in animal models. Several epidemiological
studies have shown that salicylic acid and its prodrug acetylsalicylic acid are tumor-preventive for humans. The
mechanism of salicylic acid- and anthocyanidin-dependent antitumor effects has remained enigmatic so far. Extracellular
apoptosis-inducing reactive oxygen species signaling through the NO/peroxynitrite and the HOCl signaling pathway
specifically induces apoptosis in transformed cells. Tumor cells have acquired resistance against intercellular reactive
oxygen species signaling through expression of membrane-associated catalase. Here, we show that salicylic acid and
anthocyanidins inactivate tumor cell protective catalase and thus reactive apoptosis-inducing intercellular reactive
oxygen species signaling of tumor cells and the mitochondrial pathway of apoptosis Salicylic acid inhibits catalase directly
through its potential to transform compound I of catalase into the inactive compound II. In contrast, anthocyanidins
provoke a complex mechanism for catalase inactivation that is initiated by anthocyanidin-mediated inhibition of NO
dioxygenase. This allows the formation of extracellular singlet oxygen through the reaction between H2O2 and peroxynitrite,
amplification through a caspase8-dependent step and subsequent singlet oxygen-mediated inactivation of catalase.
The combination of salicylic acid and anthocyanidins allows for a remarkable synergistic effect in apoptosis induction.
This effect may be potentially useful to elaborate novel therapeutic approaches and crucial for the interpretation of
epidemiological results related to the antitumor effects of secondary plant compounds.
Introduction
Reactive oxygen species and catalase in multistep
oncogenesis
Reactive oxygen species (ROS) have multiple and partially
adverse effects during oncogenesis (1). The mutagenic effect of
ROS may contribute to the initiation of oncogenesis and to the
induction of genetic changes during tumor progression (2,3).
Tumor promotion seems to be tightly connected to the action of
ROS (3,4), though the exact mechanism has not been completely
clarified yet. Oncogene-dependent transformation causes the
constitutive activation of membrane-associated reduced nicotinamide adenine dinucleotide phosphate (NADPH) oxidase
(1,5,6). The resultant extracellular superoxide anions and their
dismutation product H2O2 are required for the maintenance of
the transformed state and for autocrine proliferation stimulation
of malignant cells.
Received: July 31, 2014; Revised: January 5, 2015; Accepted: January 23, 2015
© The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected].
400
Scheit and Bauer | 401
Abbreviations
ABH 4-aminobenzoyl hydrazide
AEBSF 4-(2-aminoethyl-benzenesulfonyl fluoride
3-AT 3-aminotriazole
NADPH reduced nicotinamide adenine dinucleotide phosphate
l-NAME N-omega-nitro-l-arginine methylester hydrochloride
NOS NO synthase
NOD NO dioxygenase
FeTPPS 5-, 10-, 15-, 20-Tetrakis(4-sulfonatophenyl)
porphyrinato iron(III) chloride
ROS reactive oxygen species.
These procarcinogenic functions of ROS are counterbalanced by the elimination of transformed cells through
intercellular apoptosis-inducing ROS signaling (1,7,8; see
Supplementary Figures 1A and 2A is available at Carcinogenesis
Online]. Transformed cells are specifically susceptible to apoptosis induction by the NO/peroxynitrite and the HOCl signaling
pathways. The NO/peroxynitrite pathway depends on peroxynitrite formation through superoxide anion/NO interaction, protonation of peroxynitrite and decomposition of peroxynitrous
acid into NO2 and apoptosis-inducing hydroxyl radicals (1,7,8).
In addition, transformed cells (either autocrine or in cooperation with neighboring nontransformed cells) may induce the
HOCl signaling pathway which depends on the formation of
H2O2, utilization of H2O2 by DUOX-related peroxidase for the
synthesis of HOCl and hydroxyl radical formation through the
interaction between HOCl and superoxide anions (1,7). Tumor
formation in vivo requires the establishment of the ‘H2O2 -catabolizing phenotype’ that has been characterized in tumor
progression experiments in vivo (9,10). The biochemical basis of
this crucial phenotype is explained by expression of membraneassociated catalase that is regularly found on bona fide tumor
cells, protecting the malignant cells against apoptosis induction
by their own extracellular ROS (1,11,12). Catalase decomposes
H2O2 (CATFeIII + H2O2 → CATFeIV=O.+ + H2O; CATFeIV=O.+ + H2O2 →
CATFeIII + H2O + O2) and thus removes the substrate for peroxidase (POD)-dependent HOCl synthesis (POD + H2O2 + H+ + Cl– →
POD + H2O + HOCl). Catalase also has the potential to decompose
peroxynitrite in a reaction that involves formation of compound
I of catalase (CATFeIV = O+) as intermediate (CATFeIII + ONOO– →
CATFeIV = O.+ + NO2–; CATFeIV=O.+ + ONOO– → CATFeIII + NO2– +
O2) (12). Finally, compound I of catalase oxidizes NO (CATFeIV=O.+
+ NO → CATFeIII + NO2) (13; Supplementary Figures 1B, 2B and
3 is available at Carcinogenesis Online). Therefore, extracellular
catalase very efficiently interferes with the NO/peroxynitrite
and the HOCl signaling pathway at central steps. Inhibition of
catalase by 3-aminotriazole (3-AT) or neutralizing antibodies, as
well as siRNA-mediated knockdown of catalase, reactivate intercellular ROS signaling and cause selective apoptosis induction
in tumor cells (12). In addition, extracellular singlet oxygen as
well as the reactive aldehyde 4-hydroxynonenal, a product of
lipid peroxidation, have been shown to inactivate catalase and
thus to allow intercellular ROS-dependent apoptosis signaling
in tumor cells (Bauer et al., submitted). Inactivation of catalase explains the selective apoptosis induction in tumor cells
as reported by Borovic et al. (14).The extracellular localization of
catalase is essential for protection of the tumor cells against NO/
peroxynitrite signaling, as extracellularly generated peroxynitrite cannot be reached by intracellular catalase before it interacts with the cellular membrane and causes lipid peroxidation
(12). Extracellular localization of membrane-associated catalase
of tumor cells had been shown through (i) inhibition of its activity by cell-impermeable neutralizing antibodies; (ii) its protective effect against exogenous peroxynitrite; (iii) removal through
trypsin treatment and (iv) immunofluorescence staining of living cells (12).
The effects of salicylic acid and anthocyanidins on
malignant cells
Salicylic acid and anthocyanidins are secondary plant products
and have been mainly discussed as potent antioxidants so far.
Salicylic acid is one of the key mediators in the defense mechanism of plants directed against plant pathogens (15). Salicylic
acid and its prodrug acetylsalicylic acid have been long known
for their antiinflammatory action (16).
Interestingly, salicylic acid (16–22) as well as anthocyanidins
(23–29) induce apoptosis selectively in malignant cells in vitro
and show antitumor effects in vivo (17,25,30,31). Salicylic aciddependent apoptosis induction in tumor cells is not attributed
to inhibition of COX, as salicylic acid affects COX1- and COX2deficient tumor cells (32,33) and as certain other nonsteroidal
antiinflammatory drugs that inhibit COX as efficiently as salicylic acid do not induce apoptosis (17). One common feature of
salicylic acid- and anthocyanidin-dependent apoptosis induction seems to be the induction of the mitochondrial pathway of
apoptosis (20,21,24,27,29) and the provocation of ROS generation
(20,23,27). This paradox effect, i.e. provocation of a prooxidative
mechanism by compounds that are also antioxidants has not
been sufficiently explained in mechanistic terms so far. Though
salicylic acid is known to inhibit plant and mammalian catalases
through a one electron transfer that transforms compound I into
the inactive compound II (34), the interaction of salicylic acid
with tumor cell protective catalase has not yet been suggested
as potential initial step during salicylic acid-dependent apoptosis induction in tumor cells. However, Chung et al. (20) demonstrated that RAC-controlled NADPH oxidase, the hallmark of
malignant cells, was involved in salicylic acid-dependent apoptosis induction. So far, no link between the action of anthocyanidins on one side and catalase and active membrane-associated
NADPH oxidase on the other side has been established. Here, we
show that salicylic acid as well as anthocyanidins target tumor
cell protective catalase and thus reactivate NOX1-driven intercellular apoptosis-inducing ROS signaling. As the compounds
use different mechanism, they have the potential to establish
an impressive synergistic effect on apoptosis induction.
Materials and methods
Materials
The NADPH oxidase inhibitor 4-(2-aminoethyl-benzenesulfonyl fluoride
(AEBSF), ascorbic acid, the NO donor Diethylamino NONOate (DEANONOate),
the catalase inhibitior 3-aminotriazole (3-AT), the singlet oxygen scavenger histidine, the hydroxyl radical scavenger mannitol, salicylic acid,
manganese-containing superoxide dismutase (MnSOD) from Escherichia
coli, the NO synthase (NOS) inhibitor N-omega-nitro-l-arginine methylester
hydrochloride (l-NAME), the HOCl scavenger taurine, the anthocyanidins
cyanidin, delphinidin, peonidin, pelargonin, malvidin, the anthocyan malvidin-3-O-galactoside, neutralizing monoclonal antibodies against human
catalase (clone CAT-505, mouse, IgG1) and control antibody directed against
EGF receptor were obtained from Sigma (Schnelldorf, Germany).
The mechanism-based peroxidase inhibitor 4-aminobenzoyl
hydrazide (ABH) was purchased from Acros Organics, Geel, Belgium;
peroxynitrite and the peroxynitrite decomposition catalyst 5-, 10-, 15-,
20-Tetrakis(4-sulfonatophenyl)porphyrinato iron(III) chloride (FeTPPS)
were obtained from Calbiochem (Merck Biosciences GmbH, Schwalbach/
402 | Carcinogenesis, 2015, Vol. 36, No. 3
Ts, Germany). Inhibitors of caspase2 (Z-VDVAD-FMK), caspase3 (Z-DEVDFMK), caspase8 (Z-IETD-FMK) and caspase-9 (Z-LEHD-FMK) were obtained
from R&D Systems (Wiesbaden-Nordenstadt, Germany).
Cells were kept in RPMI 1640 medium, containing 10 % inactivated fetal
bovine serum (Biochrom, Berlin, Germany), supplemented with penicillin
(40 U/ml), streptomycin (50 µg/ml), neomycin (10 µg/ml), moronal (10 U/
ml) and glutamine (280 µg/ml). Cells were grown in plastic tissue culture
flasks and passaged once or twice weekly.
months. The human lymphoma cell line Gumbus (ACC630) was established
by Drs T. Kiefer and G. Dölken (Greifswald, Germany) from a 28-year-old male
patient with European Burkitt lymphoma. We received the cell line from Dr
Dölken in August 2005, immediately before it was deposited at the DSMZ
Braunschweig, Germany. The cells were tested by the DSMZ with fluorescent
nanoplex PCR of short tandem repeat markers that revealed a unique DNA
profile. After receiving the cells in our laboratory, they were grown and 30
aliquots were stored in liquid nitrogen. Cells from individual aliquots were
cultured less than 2 months. Nontransformed diploid fibroblasts Alpha-1
have been established from the foreskin of a healthy proband.
Cells
Methods
The human gastric carcinoma cell line MKN-45 (ACC 409) and the human
Ewing sarcoma cell line SK-N-MC (ACC-203) were purchased from the
German cell culture collection (DSMZ), Braunschweig, Germany in May
2004. MKN-45 cells had originally been deposited at the DSMZ and had been
tested at the DSMZ by multiplex PCR of minisatellite markers that revealed
a unique DNA profile. SK-N-MC cells were characterized by the DSMZ by
expression of fusion gene EWS-FLI (type 1) confirmed by RT-PCR. After receiving the cells in our laboratory, they were grown and 30 aliquots were stored
in liquid nitrogen. Cells from individual aliquots were cultured less than two
Apoptosis induction.
Media for cell culture
Cells were seeded at the densities indicated in the respective figure legends in 96-well tissue culture clusters and 100 µl of complete medium
(standard conditions). The modes of apoptosis induction are summarized
below and are further specified in the respective figure legends.
In the experiment described in Figures 1 and 2, apoptosis was induced
by addition of 200 µM peroxynitrite. As tumor cells are protected against
exogenous peroxynitrite through expression of membrane-associated
catalase, apoptosis induction by peroxynitrite requires inhibition or
Figure 1. Inactivation of membrane-associated catalase of tumor cells. 12 500 MKN-45 human gastric carcinoma cells in 100 µl medium were treated with 100 µM
AEBSF or remained untreated. After 15 min, neutralizing antibodies against human catalase (antiCAT) or control antibodies (anti-Laminin) (A), salicylic acid (B), ascorbic
acid (C) or cyanidin (D) were added at the indicated concentrations. After 15 min incubation at 37°C, the assays received 200 µM peroxynitrite (PON) or not. The percentages of apoptotic cells were determined 1 h after addition of peroxynitrite. Statistical analysis: In the presence of PON, apoptosis induction mediated by 1–18 ng/
ml antiCAT, 0.31–1.25 µg/ml salicylic acid, 0.15–1.25 µg/ml ascorbic acid and 0.015–0.125 µg/ml cyanidin was highly significant (P < 0.001). AEBSF did not significantly
inhibit apoptosis induction in the presence of PON and anti-CAT, salicylic acid and ascorbic acid, but prevented the supraoptimal decline at the right flanks of the
optimum curves with high significance (P < 0.001). In the presence of PON, apoptosis induction mediated by cyanidin was inhibited by AEBSF with high significance
(P < 0.001). In the absence of PON, apoptosis induction by 10 and 20 µg/ml salicylic acid was significant (P < 0.01), apoptosis induction by 0.5 and 1 µg/ml cyanidin highly
significant (P < 0.001).
Scheit and Bauer | 403
Figure 2. The role of singlet oxygen during catalase inactivation. 12 500 MKN-45 human gastric carcinoma cells in 100 µl medium received no addition or 2 mM of the singlet
oxygen scavenger histidine, 100 µM of the NOX inhibitor AEBSF or 2.4 mM of the NOS inhibitor l-NAME(‘-15 min’). After 15 min, the assays received the indicated concentrations of salicylic acid (A-C) or cyanidin (D-F). After 20 min of incubation, some of the assays that had not received inhibitors before, now received histidine, AEBSF or l-NAME
(‘20 min’). Then the assays were challenged with 200 µM peroxynitrite (PON) or not. The percentages of apoptotic cells were determined after 1 h. Statistical analysis: In the
presence of PON, apoptosis induction mediated by 0.013–0.37 µg/ml salicylic acid and 1.3–37 µg/ml cyanidin was highly significant (P < 0.001). Apoptosis induction mediated
by salicylic acid in the presence of PON was not significantly inhibited by histidine, AEBSF and l-NAME, however, AEBSF prevented the supraoptimal decline of the optimum
curve with high significance (P < 0.001). Apoptosis induction mediated by cyanidin in the presence of PON was inhibited by histidine, AEBSF and l-NAME with high significance (P < 0.001) when the inhibitors had been added 15 min before cyanidin. There was no significant inhibition when these inhibitors had been added 20 min after cyanidin.
inactivation of membrane-associated catalase and therefore allows to
determine inhibitory or inactivating effect of defined compounds directed
against catalase. As intracellular catalase cannot reach exogenous peroxynitrite before it attacks the cell membrane (12), this assay is focusing specifically on membrane-associated catalase on the outside of tumor cells.
Apoptosis induction by exogenous peroxynitrite was usually quantified
1 h after addition of peroxynitrite. At this time point, autocrine apoptosis
induction mediated by plant-derived compounds alone was still low.
In the experiments described in Figures 3, 4 and 6, apoptosis was
induced by autocrine ROS-mediated intercellular signaling of tumor cells,
which requires the inhibition or inactivation of tumor cell protective catalase. Autocrine apoptosis induction is based on NOX1-dependent extracellular superoxide anion generation and release of peroxidase and NO. This
results in the establishment of the NO/peroxynitrite and HOCl signaling
pathway (12). Autocrine ROS-mediated apoptosis induction was determined between 2.5 and 6 h after addition of the test compounds.
In the experiment described in Figure 5, apoptosis was induced by
increasing concentrations of the NO donor DEA NONOATE, in the presence
of 200 mM of the catalase inhibitor 3-AT, 2.4 mM of the NOS inhibitor
l-NAME and 25 µM of the catalase mimetic EUK-134. This assay therefore
allows to quantify the consumption of defined concentrations of exogenous NO by cellular NO dioxygenase (NOD) without interference by cellular NO synthesis and without consumption of NO by hydrogen peroxide.
The apoptosis inducing effect of NO is mediated by peroxynitrite formation through NO/superoxide anion interaction (8).
In all experiments, assays were performed in duplicate. After the indicated time of incubation at 37°C and 5% CO2, the percentage of apoptotic
cells was determined by inverted phase contrast microscopy based on the
classical criteria for apoptosis, i.e. nuclear condensation/fragmentation or
membrane blebbing. The characteristic morphological features of intact
and apoptotic cells, as determined by inverted phase contrast microscopy are shown in Supplementary Figure 16, available at Carcinogenesis
Online. At least 200 neighboring cells from randomly selected areas were
scored for the percentage of apoptotic cells at each point of measurement. The validation of our determination of apoptotic cells is shown in
Supplementary Figures 11, 13–17, available at Carcinogenesis Online.
404 | Carcinogenesis, 2015, Vol. 36, No. 3
Figure 3. Salicylic acid-dependent catalase inhibition establishes intercellular apoptotic ROS signaling specifically in tumor cells. 12 500 MKN-45 human gastric carcinoma cells (A and B) or 12 500 human nontransformed diploid fibroblasts (C) in 100 µl reaction volume were treated with the indicated concentrations of salicylic
acid, in the absence or presence of the indicated inhibitors (histidine: singlet oxygen scavenger, 2 mM; taurine: HOCl scavenger, 50 mM; l-NAME: NOS inhibitor, 2.4 mM;
mannitol: hydroxyl radical scavenger, 10 mM; ABH: peroxidase inhibitor, 150 µM; MnSOD: superoxide anion scavenger, 100 U/ml; FeTPPS: peroxynitrite decomposition
catalyst, 25 µM). The percentages of apoptotic cells were determined after 5 h. Statistical analysis: Apoptosis induction in MKN-45 cells by 1.5–25 µg/ml salicylic acid
was highly significant (P < 0.001), whereas there was no significant apoptosis induction in nonmalignant Alpha-1 cells. Histidine had no significant inhibitory effect on
apoptosis induction by salicylic acid in MKN-45 cells. Mannitol and MnSOD inhibited apoptosis induction in MKN-45 cells at all concentrations of salicylic acid with
high significance (P < 0.001). Taurine and ABH showed no significant inhibitory activity on apoptosis induction by 1.5 and 3.1 µg/ml salicylic acid, but inhibited at higher
concentrations of salicylic acid with high significance (P < 0.001). l-NAME and FeTPPS inhibited salicylic acid-dependent apoptosis induction with high significance
(P < 0.001) between 1.5 and 6.2 µg/ml salicylic acid. Inhibition by these compounds was significant (P < 0.01) at 12.5 µg/ml salicylic acid and without significance at
higher concentrations.
Statistical analysis.
In all experiments, assays were performed in duplicate. Empirical standard deviations were calculated and are shown to allow assessment of the
degree of reproducibility, but not with the intention of statistical analysis of variance, which would require larger numbers of parallel assays.
Absence of standard deviation bars for certain points indicates that the
standard deviation was too small to be reported by the graphic program.
The Yates continuity corrected chi-square test was used for the statistical determination of significances (P < 0.01 = significant; P < 0.001 = highly
significant) using the criteria ‘apoptotic’ versus ‘intact’ cells in the unprocessed original data from which the presented mean values were calculated. All experiments that are shown here have been repeated at least
twice (with duplicate assays). The basic findings, i.e. induction of apoptosis by salicylic acid and anthocyanidins have been reproduced more than
50 times by several investigators. Statistical analysis is further discussed
under Supplementary Materials, available at Carcinogenesis Online.
Strategy of our analysis
Sensitization of tumor cells for apoptosis induction by exogenous peroxynitrite was used as initial test to determine the potential of salicylic acid,
control compounds and cyanidin to inhibit tumor cell protective catalase.
This assay allowed to differentiate between direct inhibition of catalase by
salicylic acid and indirect, superoxide-anion and singlet oxygen-dependent inactivation of catalase by cyanidin. In a second round of experiments,
autocrine ROS-mediated apoptosis signaling was established through
inhibition of catalase by salicylic acid and cyanidin. The addition of inhibitors before and after cyanidin application allowed to differentiate between
Scheit and Bauer | 405
Figure 4. Cyanidin establishes intercellular ROS signaling in tumor cells through a two-step mechanism 12 500 MKN-45 gastric carcinoma cells in 100 µl medium were
cultivated without inhibitors in the presence of increasing concentrations of cyanidin (‘control’), or received the the indicated inhibitors either 15 min before addition
of increasing concentrations of cyanidin (A, B: ‘INH: -15 min’) or 25 min after addition of cyanidin (C-F: ‘INH: + 25 min’). Inhibitors: 2 mM of the singlet oxygen scavenger histidine (‘HIS’) (A, C); 2.4 mM of the NOS inhibitor L-NAME (A, C); 25 µM of caspase-8 inhibitor (‘CASP-8 INH’) (B, D); 25 µM of the peroxynitrite decomposition
catalyst FeTPPS (B, D); 150 µM of the peroxidase inhibitor 4-aminobenzoyl hydrazide (‘ABH’) (E); 100 U/ml of the superoxide anion scavenger SOD (E); 100 µM of the
NOX1 inhibitor AEBSF (E), 50 mM of the HOCl scavenger taurine (‘TAU’) (F); 10 mM of the hydroxyl radical scavenger mannitol (‘MANN’) (F), 25 µM of caspase-9 inhibitor (‘CASP-9 INH’) (F). The percentages of apoptotic cells were determined 2.5 h after addition of cyanidin. The differential addition of inhibitors allowed to dissect
early cyanidin-dependent catalase inactivation from subsquent intercellular ROS signaling through the NO/peroxynitrite and the HOCl signaling pathway. Statistical
analysis: Apoptosis induction by cyanidin (62-250 ng/ml) was highly significant (P < 0.001). Inhibition of cyanidin-dependent apoptosis induction by histidine, caspase-8
inhibitor, FeTPPS and L-NAME was highly significant (P < 0.001) at all concentrations of cyanidin when the inhibitors had been added 15 min before cyanidin. Addition
of histidine or caspase-8 inhibitor 25 min after cyanidin did not cause significant inhibition of apoptosis induction. Addition of L-NAME or FeTPPS 25 min after cyanidin
caused highly significant inhibition for apoptosis induction mediated by 62 and 125 ng/ml cyanidin, but not at higher concentrations. Addition of caspase-9 inhibitor,
SOD, AEBSF or mannitol 25 min after cyanidin causes highly significant inhibition of apoptosis induction (P < 0.001) at all concentrations of cyanidin. Addition of taurine
and ABH 25 min after cyanidin caused highly significant inhibition of apoptosis mediated by 125 and 250 ng/ml cyanidin (P < 0.001).
early ROS interactions, leading to the generation of singlet oxygen and
catalase inactivation and subsequent ROS-dependent apoptosis signaling
through the NO/peroxynitrite and HOCl pathway.
Results
Inactivation of membrane-associated catalase of
tumor cells
The abrogation of protection against apoptosis induction by
exogenously added peroxynitrite was used for the quantitative
determination of the inactivation of membrane-associated catalase of tumor cells, as this enzyme decomposes peroxynitrite.
Intracellular catalase does not affect this assay, as it cannot reach
peroxynitrite before it attacks the cell membrane. Antibodies
directed against catalase (Figure 1A), but not irrelevant control
antibodies, as well as salicylic acid (Figure 1B), ascorbic acid
(Figure 1C), cyanidin (Figure 1D) and methyldopa (Supplementary
Figure 4, available at Carcinogenesis Online) sensitized human
MKN-45 gastric carcinoma cells for apoptosis induction by exogenous peroxynitrite. The concentration-dependent sensitizing
effect of all compounds resulted in an optimum curve. In the
406 | Carcinogenesis, 2015, Vol. 36, No. 3
Figure 5. Inhibition of NO dioxygenase (NOD) by anthocyanidins. 12 500 MKN-45 cells in 100 µl medium in the presence of 200 mM of the catalase inhibitor 3-AT, 2.4 mM
of the NOS inhibitor l-NAME, 25 µM of the catalase mimetic EUK-134 received the indicated concentrations of cyanidin, malividin or malvidin-3-O-galactoside or
remained free of addition (control). After 10 min, the indicated concentrations of the NO donor DEA NONOATE were added and the percentages of apoptotic cells were
determined after 4 h (A) or 2.5 h (B). The presence of 3-AT prevents oxidation of NO and decomposition of peroxynitrite by catalase. The NOS inhibitor l-NAME prevents
the contribution of cell-derived NO to peroxynitrite formation and apoptosis induction. EUK-134 removes excess hydrogen peroxide after catalase inhibition. Thus the
assay allows to focus on the abrogation of consumption of exogenous NO by inhibitors of NOD. Statistical analysis: Apoptosis induction mediated by 0.25 and 0.5 mM
DEANONOATE was enhanced with high significance (P < 0.001) by all compounds and all concentrations tested, except for 0.8 ng/ml malvidin.
presence of the NOX1 inhibitor AEBSF, peroxynitrite-dependent
apoptosis induction mediated by antibodies against catalase, as
well as by salicylic acid, ascorbic acid and methyldopa prevailed
and was changed from an optimum curve to a plateau-type of
curve, whereas peroxynitrite-dependent apoptosis induction
mediated by cyanidin was strongly inhibited. In the absence
of the peroxynitrite challenge, high concentrations of salicylic
acid, cyanidin and methyldopa caused concentration-dependent
apoptosis induction in the tumor cells at a low, but distinct level.
This apoptosis induction was dependent on superoxide anion
generation by the tumor cells. These data allow to differentiate
between superoxide anion-independent inhibition of catalase
by salicylic acid, ascorbic acid and methyldopa and superoxide
anion-dependent inactivation of catalase by cyanidin. Inhibition
of catalase by salicylic did not depend on the cell density, whereas
the effect induced by cyanidin was optimal at a higher cell density and minimal at the lower one, indicating that the inactivation
process required intercellular signaling (Supplementary Figure 5,
available at Carcinogenesis Online).
The role of singlet oxygen for the inactivation of
catalase
To determine the role of singlet oxygen generation for the inactivation of membrane-associated catalase, treatment of the cells
with salicylic acid or cyanidin was performed in the absence or
presence of the singlet oxygen scavenger histidine, the NOX-1
Scheit and Bauer | 407
Figure 6. Synergistic effect between salicylic acid and cyaniding. (A) 12 500 MKN-45 cells in 100 µl medium received the indicated concentrations of salicylic acid or
remained free of salicylic acid (control). After 10 min, cyanidin was added at the indicated concentrations and the percentages of apoptotic cells were determined after
2.5 h. (B) Assays containing 0.1 µg/ml salicylic acid or no salicylic acid (control) were combined with increasing concentrations of ascorbic acid. The percentages of apoptotic cells were determined after 2.5 h. This control shows that the combination of two direct catalase inhibitors does not allow for the establishment of an synergistic
effect, whereas the combination of the direct catalase inhibitor salicylic acid with cyanidin, that causes singlet oxygen-dependent inactivation of catalase establishes
a synergistic effect (A). Statistical analysis: Cyanidin-mediated apoptosis induction with high significance (P < 0.001) in the concentration range between 30 ng/ml and
1 µg/ml. The leftward shift of the cyanidin-dependent curve of apoptosis induction by all concentrations of salicylic acid applied was highly significant (P < 0.001). In
the absence of cyanidin, apoptosis induction by 1.25 and 5 µg/ml salicylic acid was highly significant (P < 0.001) and without significance for the lower concentrations
of salicylic acid. Ascorbic acid-mediated apoptosis induction was significant (P < 0.01) at 1.5 µg/ml and highly significant (P < 0.001) at higher concentrations There was
no significant effect of 0.1 µg/ml salicylic acid on apoptosis induction mediated by ascorbic acid.
inhibitor AEBSF and the NOS inhibitor l-NAME. The inhibitors had
been added either 15 min prior to or 20 min after the addition of
salicylic acid (Figure 2A–C) or cyanidin (Figure 2D–F). Inhibition of
catalase by salicylic acid (demonstrated as sensitization for apoptosis induction by peroxynitrite) was not dependent on singlet
oxygen, superoxide anions or NO, as it was not inhibited by histidine, AEBSF or l-NAME (Figure 2A–C). However, AEBSF changed
the salicylic acid-dependent optimum curve to a plateau-type of
curve, confirming that a superoxide anion-dependent process
interfered with peroxynitrite-mediated apoptosis induction at
high concentrations of salicylic acid. In contrast, the catalaseinactivating effect of cyanidin was strictly dependent on singlet oxygen, superoxide anion generation and NO (Figure 2D–F),
as it was completely blocked when histidine, AEBSF or l-NAME
had been added prior to cyanidin. Addition of the inhibitors
20 min after cyanidin no longer prevented catalase inactivation.
Therefore, cyanidin-dependent catalase inactivation seems to be
mediated by a fast process, dependent on superoxide anions, NO
and singlet oxygen. Addition of AEBSF 20 min after cyanidin and
before the peroxynitrite challenge shifted the optimum curve of
apoptosis induction to a plateau type curve, indicative of a superoxide anion-dependent process directed against peroxynitrite.
Inactivation of catalase by salicylic acid and
cyanidin reactivates intercellular ROS signaling
As membrane-associated catalase protects against intercellular ROS-mediated signaling, its inactivation by salicylic acid
408 | Carcinogenesis, 2015, Vol. 36, No. 3
or cyanidin should cause subsequent ROS-dependent apoptotic signaling. Addition of salicylic acid caused apoptosis
induction in the human gastric carcinoma cell line MKN-45
(Figure 3A and B) and the human lymphoma cell line Gumbus
(Supplementary Figure 6, available at Carcinogenesis Online) in
a concentration-dependent mode, but not in nonmalignant
human fibroblasts (Figure 3C). In both tumor cell lines, apoptosis induction seemed to be independent of singlet oxygen,
but dependent on superoxide anions and hydroxyl radicals at
all concentrations of salicylic acid applied. At low concentrations of salicylic acid, apoptosis-inducing signaling seemed to
occur through the NO/peroxynitrite signaling pathway, as it
was inhibited by the NOS inhibitor l-NAME and the peroxynitrite decomposition catalyst FeTPPS. At higher concentrations
of salicylic acid, the HOCl signaling pathway became dominant, as seen by the complete inhibition of apoptosis induction by the peroxidase inhibitor ABH and the HOCl scavenger
taurine. Apoptosis induction by salicylic acid was also seen in
the Ewing sarcoma cell lines CHP 100 and A 4573 (data not
shown), and SK-N-MC (Supplementary Figure 15, available at
Carcinogenesis Online).
A kinetic analysis of apoptosis induction in MKN-45 cells
mediated by salicylic acid showed that apoptosis induction was
relatively fast during the first 5 h, but seemed to be limited to
this time window, as salicylic acid seemed to interfere with the
process that it had initiated (Supplementary. Figure 7, available
at Carcinogenesis Online). In analogy to salicylic acid and as to be
expected from its direct mode of catalase inhibition, ascorbic
acid caused apoptosis induction in MKN-45 gastric carcinoma
cells in a concentration-dependent mode, independent of singlet oxygen, but dependent on the NO/peroxynitrite and the
HOCl signaling pathway (Supplementary Figure 8, available at
Carcinogenesis Online).
Cyanidin triggered apoptosis induction in MKN-45 gastric carcinoma cells in a complex two-step signaling reaction (Figure 4).
Addition of the singlet oxygen scavenger histidine, caspase-8
inhibitor, the peroxynitrite decomposition catalyst FeTPPS and
the NOS inhibitor l-NAME before cyanidin strongly inhibited
subsequent apoptosis induction at all concentrations of cyanidin (Figure 4A and B). When the same inhibitors had been added
25 min after cyanidin, histidine and caspase-8 inhibitor did no
longer inhibit apoptosis induction, whereas FeTPPS and l-NAME
still inhibited apoptosis induction at low concentrations of cyanidin, but not at higher ones (Figure 4C and D). The addition of
caspase-9 inhibitor, SOD, the NOX inhibitor AEBSF and hydroxyl
radical scavenger mannitol 25 min after application of cyanidin
caused complete inhibition of apoptosis induction at all concentrations of cyanidin (Figure 4E and F). Addition of the peroxidase
inhibitor ABH and the HOCl scavenger 25 min after cyanidin did
not cause inhibition of apoptosis at low concentrations of cyanidin, but complete inhibition at high concentrations of cyanidin
(Figure 4E and F). These data indicate that cyanidin first triggers
a sensitization step of the tumor cells that depends on singlet
oxygen, caspase 8 and peroxynitrite. This early step seems to represent catalase inactivation and is followed by NO/peroxynitrite
signaling (at lower concentrations of cyanidin) and HOCl signaling (at higher concentrations of cyanidin). This complex pattern
of action was confirmed for peonidin, pelargonin, delphidin, malvidin and malvidin galactoside (Supplementary Figures 9 and 10,
available at Carcinogenesis Online), indicating that it represents a
general potential of anthocyanidins and anthocyans. Cyanidinmediated apoptosis induction was also shown for the Ewing sarcoma cell line SK-N-MC (Supplementary Figure 15, available at
Carcinogenesis Online).
Salicylic acid and cyanidin cause caspase9- and
caspase3-dependent apoptosis induction through
the mitochondrial pathway
Apoptosis induction by salicylic acid and cyanidin was inhibited
by caspase9 and caspase3 inhibitor (Supplementary Figure 11,
available at Carcinogenesis Online), thus confirming the role of
these capases for cell death, as well as its apoptotic nature. The
dependence of apoptosis induction by low concentrations of
cyanidin on the activity of caspase8 (as shown in Figure 4 and
Supplementary Figure 11, available at Carcinogenesis Online)
was abrogated, when superoxide anion generation had been
stimulated by treatment with transforming growth factor-beta
(Supplementary Figure 12, available at Carcinogenesis Online).
This finding shows that the effect of caspase8 in this experimental system is related to enhancement of superoxide anion
generation rather than to death receptor-mediated pathway of
apoptosis induction. Rather, apoptosis induction mediated by
salicylic acid and cyanidin seems to depend on the mitochondrial pathway of apoptosis (Supplementary Figures 13 and 14,
available at Carcinogenesis Online).
Anthocyanidins inhibit NOD
Singlet oxygen-dependent inactivation of tumor cell protective catalase through a multistep mechanism has been recently
shown to be triggered by an initial increase in NO (1), which
had been achieved either by an increase in the concentration of
the NOS substrate arginine, or through azole-dependent inhibition of NOD (1). NOD efficiently lowers the level of cellular NO
through dioxygenation of NO to NO3−. Therefore, its inhibition
causes a sudden increase in NO without a change in the rate of
NO synthesis per se (35). As quercetin, the oxidation product of
cyanidin, inhibits NOD (36), it was not unlikely that anthocyanidins might have NOD inhibitory potential as well. They thus
might provoke a sudden increase in free NO without the necessity to modulate its synthesis. Intriguingly, all anthocyanidins
and anthocyans tested showed strong inhibition of NOD, as they
strongly enhanced the apoptosis inducing of an exogenous NO
donor, under conditions where catalase and NOS were inhibited and the consumption between H2O2 and NO was abrogated
(Figure 5).
Synergistic interaction between salicylic acid and
cyanidin
As salicylic acid and cyanidin inactivate tumor cell catalase
through different biochemical mechanisms, it was not unlikely
that their combined action would allow for synergistic effects.
When increasing concentrations of cyanidin were combined
with increasing, but suboptimal concentrations of salicylic acid,
an impressive synergistic effect was seen (Figure 6). For example, 0.08 µg/ml salicylic acid or 0.45 ng/ml cyanidin applied alone
did not induce apoptosis significantly above background, but
their combination resulted in apoptosis induction higher than
reached by 0.3 µg/ml cyanidin or 5 µg/ml salicylic acid alone.
When ascorbic acid and salicylic acid, i.e. two compounds with
analogous mode of direct catalase inhibition, were combined,
no synergistic effect between these two compounds was seen
(Figure 6B), whereas the combination of ascorbic acid with cyanidin resulted in a similar synergistic effect as the combination
between salicylic acid and cyanidin (Supplementary Figure 18,
available at Carcinogenesis Online). This finding confirms the
specificity and significance of the synergistic effect that is
shown in Figure 6A.
Scheit and Bauer | 409
Discussion
Impact of salicylic acid-dependent inhibition of
catalase on its tumor-preventive effect
Our data show that salicylic acid, ascorbic acid, methyldopa and
cyanidin sensitize human tumor cells for apoptosis induction
by exogenously added peroxynitrite as efficiently as neutralizing antibodies directed against catalase. As only extracellular catalase can decompose exogenous peroxynitrite before it
attacks the cell membrane, and as neutralizing antibodies can
only reach extracellular catalase of intact tumor cells, the sensitizing effect of salicylic acid, ascorbic acid, methyldopa and
cyanidin is explained as inhibition of extracellular catalase. As
the NOX inhibitor AEBSF did not interfere with sensitization
by antibodies, salicylic acid, ascorbic acid and methyldopa, but
shifted the optimum curves to plateau-type curves, it is obvious that the inhibitory effect on catalase by these compounds is
independent of NOX1-derived superoxide anions. The supraoptimal decline of apoptosis induction on the right flank of the
optimum curves is explained as consumption of peroxynitrite
by excess H2O2 (ONOO– + H2O2 → 1O2 + NO2– + H2O) (37), when
catalase is strongly inhibited. Thereby, superoxide anions generated by NOX1 seem to be the source for H2O2, as seen by the
AEBSF-mediated abrogation of consumption. In contrast, inhibition of catalase by cyanidin seems to be mediated by a superoxide anion-dependent process as it is strongly inhibited by
AEBSF. However, the mechanism underlying the formation of an
optimum curve is the same as seen for the other compounds,
as addition of AEBSF 20 minutes after addition of cyanidin, but
before the peroxynitrite challenge, shifted the optimum curve to
a plateau type curve as well (Figure 2E).
Inhibition of catalase by salicylic acid, ascorbic acid and
methyldopa is in line with the reported potential of these
compounds to reduce compound I (CATFeIV=O.+) to the inactive compound II (CATFeIV=O) through a one-electron transfer
(34,38,39). In the membrane of tumor cells, compound I is constantly formed through the interaction of membrane-associated
catalase with H2O2 generated through dismutation of NOX1derived extracellular superoxide anions. In the presence of the
NOX inhibitor AEBSF, this reaction is not possible and therefore
another source for substrate that allows compound I formation
must be considered. It may be assumed that the interaction of
catalase with one molecule of the challenging peroxynitrite
generates compound I in an initial step and thus enables the
subsequent one electron-dependent inhibition step. As a consequence, the remaining concentration of peroxynitrite is not
decomposed by the inactive compound II.
As shown exemplarly for salicylic acid, inhibition of catalase is not only independent of cell-derived superoxide anions
(or their dismutation product H2O2), it also does not require NO
(or its derivative peroxynitrite) or singlet oxygen, the reaction
product from peroxynitrite and H2O2. However, superoxide anions, H2O2, NO and peroxynitrite are required for the establishment of intercellular ROS-dependent apoptosis signaling which
seems to be the consequence of direct inhibition of tumor cell
protective catalase by salicylic acid. Intercellular ROS signaling
can only be established by NOX1-expressing cells. This explains
the selectivity of apoptosis induction by salicylic acid for malignant cells, as reported by others (22) and shown in this study.
The reactivated intercellular apoptosis signaling is analogous
to intercellular ROS signaling established through neutralizing antibodies against catalase, the catalase inhibitor 3-AT or
siRNA-mediated inhibition of catalase expression (12). Thereby,
in MKN-45 cells, a low degree of catalase inhibition allows the
action of the NO/peroxynitrite pathway which is followed by
the HOCl pathway at higher degrees of catalase inhibition. As
inhibitors of caspase-3 and caspase-9 completely inhibited
salicylic acid-dependent apoptosis induction (Supplementary
Figure 11, available at Carcinogenesis Online), and as siRNA-mediated knockdown of BAK, APAF and DIABLO prevented apoptosis
(Supplementary Figure 13, , available at Carcinogenesis Online),
the mitochondrial pathway of apoptosis seems to be active, in
analogy to the effect of neutralizing antibodies and 3-AT that
are mechanistically identical (12). This is in line with the reports
on induction of the mitochondrial pathway (20,21) and on ROS
formation by salicylic acid (20,21). In their detailed study, Chung
et al. (20) demonstrate that a rac1-NADPH oxidase-dependent
pathway is targeted by salicylic acid. This pathway is shown to
act upstream of mitochondrial effects such as collapse of the
membrane potential, release of cytochrome c and subsequent
caspase activation. Chung et al. concluded that NADPH oxidasedependent ROS production was completely intracellular, as it
was not affected by exogenously added catalase. However, their
studies were performed in the presence of 2.5–5 mM salicylic
acid and therefore their result is not conclusive, as catalase can
be expected to be inhibited under these conditions, according
to our data. As NADPH oxidase of malignant cells generates
extracellular superoxide anions and as membrane-associated
extracellular catalase protects tumor cells against intercellular ROS-mediated apoptosis signaling, we suggest that inhibition of catalase by salicylic acid might have been the initial step
in apoptosis induction in the experimental system studied by
Chung et al., in analogy to the findings presented in this article.
Based on the potential of salicylic acid to inhibit tumor cell
protective catalase and thus to reactivate intercellular ROS signaling that causes tumor cell apoptosis, it seems reasonable to
consider that this mechanism might significantly contribute
to the tumor-preventive effects of salicylic acid and its prodrug
acetyl salicylic acid that has been demonstrated in mutlitple
studies (40–43).
Due to the multifunctional potential of salicylic acid, the
tumor preventive effect might be a combination of several
effects. The antioxidant potential of salicylic acid might prevent
procarcinogenic ROS effects during tumor initiation, promotion
and progression. Its prooxidative effect after inhibition of tumor
cell protective catalase might be involved in the elimination of
catalase-positive microtumors, utilizing the signaling chemistry
that is illustrated in this manuscript.
As salicylic acid also has a strong scavenging activity for
hydroxyl radicals (44), its prooxidative effect due to catalase
inactivation is counteracted in parallel by consumption of salicylic acid by hydroxyl radicals that are generated during intercellular ROS signaling. This interaction causes a decrease in the
concentration of salicylic acid and thus to an attenuation of
catalase inhibition after the onset of ROS signaling. This might
represent a substantial obstacle for attempts to establish tumor
therapy based on sole application of salicylic acid or its derivatives. More details of the complex interactions between salicylic
acid and catalase are discussed under Supplementary discussion
(Supplementary Figure 19, available at Carcinogenesis Online).
Anthocyanidins mediate singlet oxygen-dependent
inactivation of catalase
In contrast to direct inhibition of catalase by salicylic acid, inhibition of catalase by anthocyanidins is indirect and mediated
by singlet oxygen-dependent inactivation of catalase. Singlet
410 | Carcinogenesis, 2015, Vol. 36, No. 3
oxygen generation represents an early and fast step immediately after addition of anthocyanidins to the tumor cells and
seems to be completed within less than 20 min. As shown in
Figures 2 and 4, this early catalase-inactivating step requires
superoxide anions, NO, peroxynitrite, caspase-8, and singlet
oxygen. Caspase-8 activity seems to be triggered by the FAS
receptor, as siRNA-mediated knockdown of the FAS receptor
also has a strong inhibitory effect on anthocyanidin-dependent
apoptosis induction (Scheit and Bauer, unpublished result).
Singlet oxygen that is involved in the inactivation of catalase
may be formed through the interaction between cell-derived
peroxynitrite and hydrogen peroxide (H2O2 + ONOO−→ 1O2 +
H2O + NO2−) (1,37). As catalase decomposes H2O2 and peroxynitrite, formation of singlet oxygen through interaction of these
compounds requires transient inhibition of catalase. Inhibition
of NOD by anthocyanidins (Figure 5) and inhibition of catalase
by the resultant NO (45) might account for this crucial step. An
analogous biochemical scenario has been described for the
increase in the NO level through an increase of the NOS substrate arginine and the inhibition of NOD by azoles (1). As singlet oxygen triggers the FAS receptor in a ligand-independent
mode (46), and as this may lead to the caspase-8-dependent
activation of NOX (47,48) and induction of NOS (49), a further
increase in H2O2 and peroxynitrite, followed by an increasing
concentration of catalase-inactivating singlet oxygen (50) can be
expected. In contrast to caspase 9 that is involved in the induction of cell death, caspase 8 only seems to play a role for the
enhancement of catalase inactivation. This conclusion was substantiated through our finding that an increase in NOX1 activity
through TGF-beta1 abrogated the dependence of cyanidin-mediated apoptosis induction on caspase8 activity (Supplementary
Figure 12, available at Carcinogenesis Online). Inactivation of catalase by singlet oxygen is followed by NO/peroxynitrite signaling at lower concentrations of cyanidin and the HOCl signaling
pathway at higher ones. Both pathways lead to the formation
of hydroxyl radicals (Supplementary Figures 1 and 2, available
at Carcinogenesis Online) that seem to be the ultimate apoptosis
inducers, potentially through lipid peroxidation, as concluded
from established concepts (51). Most probably, as a consequence
of hydroxyl radical-mediated lipid peroxidation, the mitochondrial pathway of apoptosis is activated (Supplementary
Figure 14, available at Carcinogenesis Online).
Catalase inactivation through anthocyanidins explains their
well-known selective apoptosis induction in malignant cells
(27) characterized by an involvement of ROS and the mitochondrial pathway (23,24,27). Based on the data presented here, we
suggest that formation of extracellular singlet oxygen through
tumor cell-derived ROS (i.e. H2O2 and peroxynitrite) represents
the primary and crucial ROS-dependent step induced by anthocyanidins. This step is followed by intercellular ROS-mediated
apoptosis signaling through the NO/peroxynitrite and HOCl
signaling pathways that trigger the onset of the mitochondrial
pathway of apoptosis. As cytochrome c is released from mitochondria in this pathway, the electron transport chain becomes
uncoupled, leading to the generation of mitochondrial superoxide anions and H2O2. This final ROS generation therefore is a
consequence of apoptosis induction by the primary extracellular
ROS interactions and not its cause. Until now, many reports have
discussed mitochondria-derived ROS as the cause of apoptosis
induction, thereby neglecting the potential role of extracellular
ROS signaling. This restricted view is based on the misconcept
that the mitochondrial pathway of apoptosis is always controlled intrinsically, whereas death receptor-dependent apoptosis induction represents the extrinsic pathway. It is obvious
from our data that intercellular apoptosis-inducing ROS signaling represents an extrinsic pathway, though without direct
involvement of death receptors...
In contrast to salicylic acid, there are no epidemiological data
in humans that indicate a tumor preventive effect of anthocyans, despite their specific effects directed against malignant
cells in vitro (23–29) and in animal models in vivo (25). It has been
suggested that this failure might be due to the low concentration of anthocyans reached in body fluids after intake of the
compounds (29).
Intriguingly, salicylic acid and cyanidin show an impressive
synergistic effect. Synergy seems to depend on the different
primary mode of action of both compounds during inactivation
of catalase that leads to congruent effects during intercellular
apoptosis-inducing ROS signaling and the induction of the mitochondrial pathway of apoptosis. Therefore, the combination
of salicylic acid with ascorbate that causes the same mode of
direct inhibition of catalase did not establish synergism. A possible mechanism underlying the synergistic effect is oulined
under Supplementary Discussion. This synergistic effect might
have interesting therapeutic potential, but it also raises analytical problems. Its potential use in tumor prevention and therapy
may allow to establish apoptosis induction in tumor cells at
very low concentrations of the synergizing compounds. These
may be sufficient to trigger catalase inactivation, but insufficient
to block the subsequent signaling or to induce unwanted side
effects. However, synergistic effects may have a severe effect
on the outcome on epidemiological studies on the tumor preventive effect of individual nutritional plant compounds, especially when the potential synergy partner is not considered, not
quantified or when its intake is not known. This aspect requires
attention and further conceptional and experimental work.
Supplementary material
Supplementary Table 1 and Figures 1–21 can be found at http://
carcin.oxfordjournals.org/
Funding
EuroTransBio (ETB1 0315012B).
Acknowledgements
We thank the COST consortium ‘ChemBioRadical’ (COST Action
CM0603) organized by C. Chatgilialoglu (Bologna) for intellectual support and constructive criticism. We thank J. Brandel
(Freiburg) for technical support during the preparation of the
manuscript.
Conflict of Interest Statement: None declared.
References
1. Bauer, G. (2012) Tumor cell-protective catalase as a novel target for
rational therapeutic approaches based on specific intercellular ROS
signaling. Anticancer Res., 32, 2599–2624.
2. Feig, D.I. et al. (1994) Reactive oxygen species in tumorigenesis. Cancer
Res., 54(7 suppl.), 1890–1894.
3. Storz, P. (2005) Reactive oxygen species in tumor progression. Front.
Biosci., 10, 1881–1896.
4. Cerutti, P.A. (1995) Prooxidant states and tumor promotion. Free Radic.
Biol. Med., 18, 775–794.
5. Irani, K. et al. (1997) Mitogenic signaling mediated by oxidants in Rastransformed fibroblasts. Science, 275, 1649–1652.
6. Suh, Y.A. et al. (1999) Cell transformation by the superoxide-generating
oxidase Mox1. Nature, 401, 79–82.
Scheit and Bauer | 411
7. Herdener, M. et al. (2000) Target cell-derived superoxide anions cause
efficiency and selectivity of intercellular induction of apoptosis. Free
Radic. Biol. Med., 29, 1260–1271.
8. Heigold, S. et al. (2002) Nitric oxide mediates apoptosis induction selectively in transformed fibroblasts compared to nontransformed fibroblasts. Carcinogenesis, 23, 929–941.
9. Deichman, G.J. et al. (1998) Cell transforming genes and tumor progression: in vivo unified secondary phenotypic cell changes. Int. J. Cancer,
75, 277–283.
10.Deichman, G.I. (2002) Early phenotypic changes of in vitro transformed
cells during in vivo progression: possible role of the host innate immunity. Semin. Cancer Biol., 12, 317–326.
11.Bechtel, W. et al. (2009) Catalase protects tumor cells from apoptosis
induction by intercellular ROS signaling. Anticancer Res., 29, 4541–
4557.
12.Heinzelmann, S. et al. (2010) Multiple protective functions of cata
lase against intercellular apoptosis-inducing ROS signaling of human
tumor cells. Biol. Chem., 391, 675–693.
13. Brunelli, L. et al. (2001) Modulation of catalase peroxidatic and catalatic
activity by nitric oxide. Free Radic. Biol. Med., 30, 709–714.
14.Borovic, S. et al. (2007) Differential sensitivity to 4-hydroxynonenal for
normal and malignant mesenchymal cells. Redox Rep., 12, 50–54.
15.Durner, J. et al. (1997) Salicylic acid and disease resistance in plants.
Trends Plant Sci., 2, 266–274.
16.Elwood, P.C. et al. (2009) Aspirin, salicylates, and cancer. Lancet, 373,
1301–1309.
17.Aas, A.T. et al. (1995) Growth inhibition of rat glioma cells in vitro and in
vivo by aspirin. J. Neurooncol., 24, 171–180.
18.Bellosillo, B. et al. (1998) Aspirin and salicylate induce apoptosis and
activation of caspases in B-cell chronic lymphocytic leukemia cells.
Blood, 92, 1406–1414.
19.Klampfer, L. et al. (1999) Sodium salicylate activates caspases and
induces apoptosis of myeloid leukemia cell lines. Blood, 93, 2386–2394.
20.Chung, Y.M. et al. (2003) Molecular ordering of ROS production, mitochondrial changes, and caspase activation during sodium salicylateinduced apoptosis. Free Radic. Biol. Med., 34, 434–442.
21.Battaglia, V. et al. (2005) Oxidative stress is responsible for mitochondrial permeability transition induction by salicylate in liver mitochondria. J. Biol. Chem., 280, 33864–33872.
22.Elder, D.J.E. et al. (1996) Differential growth inhibition by the aspirin
metabolite salicylate in human colorectal tumor cell lines: enhanced
apoptosis in carcinoma and in vitro-transformed adenoma relative to
adenoma relative to adenoma cell lines. Cancer Res., 56, 2273–2276.
23.Hou, D.X. et al. (2004) Molecular mechanisms behind the chemopreventive effects of anthocyanidins. J. Biomed. Biotechnol., 2004, 321–325.
24.Chang, Y.C. et al. (2005) Hibiscus anthocyanins rich extract-induced
apoptotic cell death in human promyelocytic leukemia cells. Toxicol.
Appl. Pharmacol., 205, 201–212.
25.Chen, P.N. et al. (2005) Cyanidin 3-glucoside and peonidin 3-glucoside
inhibit tumor cell growth and induce apoptosis in vitro and suppress
tumor growth in vivo. Nutr. Cancer, 53, 232–243.
26.Shih, P.H. et al. (2005) Effects of anthocyanidin on the inhibition of
proliferation and induction of apoptosis in human gastric adenocarcinoma cells. Food Chem. Toxicol., 43, 1557–1566.
27.Feng, R. et al. (2007) Cyanidin-3-rutinoside, a natural polyphenol
antioxidant, selectively kills leukemic cells by induction of oxidative
stress. J. Biol. Chem., 282, 13468–13476.
28.Reddivari, L. et al. (2007) Anthocyanin fraction from potato extracts
is cytotoxic to prostate cancer cells through activation of caspasedependent and caspase-independent pathways. Carcinogenesis, 28,
2227–2235.
29.Wang, L.S. et al. (2008) Anthocyanins and their role in cancer prevention. Cancer Lett., 269, 281–290.
30. Rao, C.V. et al. (1995) Chemoprevention of colon carcinogenesis by sulindac, a nonsteroidal anti-inflammatory agent. Cancer Res., 55, 1464–1472.
31.Giardiello, F.M. et al. (1995) The role of nonsteroidal anti-inflammatory
drugs in colorectal cancer prevention. Eur. J. Cancer, 31A, 1071–1076.
32.Hanif, R. et al. (1996) Effects of nonsteroidal anti-inflammatory drugs
on proliferation and on induction of apoptosis in colon cancer cells
by a prostaglandin-independent pathway. Biochem. Pharmacol., 52,
237–245.
33.Rigas, B. et al. (2000) Is inhibition of cyclooxygenase required for the
chemopreventive effect of NSAIDs in colon cancer? A model reconciling the current contradiction. Med. Hypotheses, 54, 210–215.
34. Durner, J. et al. (1996) Salicylic acid is a modulator of tobacco and mammalian catalases. J. Biol. Chem., 271, 28492–28501.
35.Gardner, P.R. et al. (2001) Dioxygen-dependent metabolism of nitric
oxide in mammalian cells. Free Radic. Biol. Med., 31, 191–204.
36.Hallstrom, C.K. et al. (2004) Nitric oxide metabolism in mammalian
cells: substrate and inhibitor profiles of a NADPH-cytochrome P450
oxidoreductase-coupled microsomal nitric oxide dioxygenase. Free
Radic. Biol. Med., 37, 216–228.
37.Di Mascio, P. et al. (1994) Singlet molecular oxygen production in the
reaction of peroxynitrite with hydrogen peroxide. FEBS Lett., 355, 287–
289.
38.Davison, A.J. et al. (1986) Mechanism of the inhibition of catalase by
ascorbate. J. Biol. Chem., 261, 1193–1200.
39.Jones, D.P. et al. (1981) Conversion of catalase to the secondary catalase-peroxide complex (compound II) by alpha-methyldopa. Mol. Pharmacol., 20, 159–164.
40.Thun, M.J. et al. (1991) Aspirin use and reduced risk of fatal colon cancer. N. Engl. J. Med., 325, 1593–1596.
41.Harris, R.E. et al. (1996) Nonsteroidal antiinflammatory drugs and
breast cancer. Epidemiology, 7, 203–205.
42.Rothwell, P.M. et al. (2011) Effect of daily aspirin on long-term risk
of death due to cancer: analysis of individual patient data from randomised trials. Lancet, 377, 31–41.
43.Rothwell, P.M. et al. (2012) Short-term effects of daily aspirin on cancer incidence, mortality, and non-vascular death: analysis of the time
course of risks and benefits in 51 randomised controlled trials. Lancet,
379, 1602–1612.
44. Sagone, A.L. Jr et al. (1987) Oxidation of salicylates by stimulated granulocytes: evidence that these drugs act as free radical scavengers in biological systems. J. Immunol., 138, 2177–2183.
45.Brown, G.C. (1995) Reversible binding and inhibition of catalase by
nitric oxide. Eur. J. Biochem., 232, 188–191.
46.Zhuang, S. et al. (2001) Protein kinase C inhibits singlet oxygen
induced apoptosis by decreasing caspase-8 activation. Oncogene, 20,
6764–6776.
47.Suzuki, Y. et al. (1998) Rapid and specific reactive oxygen species generation via NADPH oxidase activation during Fas-mediated apoptosis.
FEBS Lett., 425, 209–212.
48.Reinehr, R. et al. (2005) Involvement of NADPH oxidase isoforms and
Src family kinases in CD95-dependent hepatocyte apoptosis. J. Biol.
Chem., 280, 27179–27194.
49.Selleri, C. et al. (1997) Induction of nitric oxide synthase is involved in
the mechanism of Fas-mediated apoptosis in haemopoietic cells. Br.
J. Haematol., 99, 481–489.
50.Escobar, J.A. et al. (1996) Sod and catalase inactivation by singlet oxygen and peroxyl radicals. Free Radic. Biol. Med., 20, 285–290.
51.Girotti, A.W. (1998) Lipid hydroperoxide generation, turnover, and
effector action in biological systems. J. Lipid Res., 39, 1529–1542.