Biochimie 118 (2015) 195e206

Contents lists available at ScienceDirect

Biochimie
journal homepage: www.elsevier.com/locate/biochi

Research paper

The iron component of particulate matter is antiapoptotic: A clue to

the development of lung cancer after exposure to atmospheric
pollutants?
Melanie Lovera-Leroux a, Belinda Crobeddu a, Nadim Kassis a, Patrice X. Petit b,
Nathalie Janel a, Armelle Baeza-Squiban a, Karine Andreau a, *
a
Universit!
e Paris Diderot, Sorbonne Paris Cit!
e, Unit of Functional and Adaptive Biology, CNRS UMR 8251, Paris, France
b
Universit!
e Paris Descartes, Sorbonne Paris Cit!e, INSERM UMR-S 1124, Paris, France

a r t i c l e i n f o a b s t r a c t

Article history: The classification of outdoor air pollution as carcinogenic for humans strengthens the increasing concern
Received 15 September 2015 about particulate matter (PM). We previously demonstrated that PM exposure produces an antiapoptotic
Accepted 24 September 2015 effect resulting from polycyclic aromatic hydrocarbons (PAH) and water-soluble components. In this
Available online 28 September 2015
study, we investigated transition metallic compounds, particularly iron, in order to decipher their un-
derlying molecular mechanisms that prevent apoptosis.
Keywords:
Human bronchial epithelial cells were exposed for 4 h to different PM samples with established
Airborne particles
antiapoptotic effect (e.g. PM-AW) or not (e.g. PM-VS) or to their metallic components (Fe, Mn, Zn and Al)
Transition metal
Cell death
before apoptosis induction by the calcium ionophore A23187 or Staurosporine. PM-AW, Fe, Mn and Al
Oxidative stress significantly reduced induced apoptosis. The antiapoptotic effect of Fe was enhanced by benzo(a)pyrene,
Bronchial epithelial cells a typical PAH compound, but was totally reversed by the iron chelator, deferiprone. Furthermore, par-
A23187 ticles and iron triggered cellular ROS generation and prevented the depletion in glutathione levels
observed during A23187-induced apoptosis. In contrast to benzo(a)pyrene, PM-AW and Fe rapidly
activated NRF2, subsequently upregulated several target genes (HO1, NQO1 and GPX1) and modulated
some genes which control cell death (BCL2, BAX and p53). The key role of the NRF2 pathway in the
antiapoptotic effect mediated by Fe and PM was demonstrated using siRNA technology.
Our results suggest that the iron component participates in the antiapoptotic effect of PM by activating
a NRF2-dependent antioxidant process. As resisting to cell death is one of the hallmarks of cancer cells,
these findings provide additional clues for understanding the development of lung cancer after atmo-
spheric pollution exposure.
© 2015 Elsevier B.V. and Socie ! te
! Française de Biochimie et Biologie Mole
!culaire (SFBBM). All rights
reserved.

1. Introduction which leads to respiratory and systemic inflammation through

Air pollution, especially particulate matter (PM), generates
health concerns (particularly in urban areas with high population
densities and industrial activities) because of the increased risk of
cardiorespiratory morbidity and mortality [1], lung cancers [2] and
the reduction in life expectancy [3]. These deleterious effects are
related mainly to the smallest particles. They are able to reach the
bronchi and alveoli and to accumulate in tissue and lung cells [4]
* Corresponding author. Present address: Universite ! Paris Descartes, Sorbonne
Paris Cite
!, INSERM UMR-S 1124, 45 rue des Saints-Pe"res, Paris, 75006, France.
E-mail address: karine.andreau@univ-paris-diderot.fr (K. Andreau).
increased reactive oxygen species (ROS) production [5,6]. These
airborne particles, termed PM10, PM2.5, and PM0.1, have aero-
dynamic diameters equal to or less than 10, 2.5, and 0.1 microns,
respectively. More than one thousand epidemiological studies have
shown a significant association between PM exposure and a risk of
lung cancer resulting in the classification of outdoor air pollution
and PM as human carcinogens (Group 1) [7].
Carcinogenesis is a multistep process that includes initiation,
promotion and tumor progression. It results in the development of
a primary tumor and, eventually, the spread of the cancer to sec-
ondary sites through metastasis. The hallmarks of cancer consist of
ten biological capabilities acquired during this multistep develop-
ment and includes resistance to apoptotic cell death [8,9].

http://dx.doi.org/10.1016/j.biochi.2015.09.030
0300-9084/© 2015 Elsevier B.V. and Socie ! te
! Française de Biochimie et Biologie Mole
!culaire (SFBBM). All rights reserved.
196 M. Lovera-Leroux et al. / Biochimie 118 (2015) 195e206
Abbreviations
A23187 calcium ionophore (calcimycin)
AhR aryl hydrocarbon receptor
ARE antioxidant response elements
BaP benzo(a)pyrene
BaP-AW BaP quantum satis 10 mg/cm2 of PM-AW
B2M beta-2-microglobulin
DFP deferiprone
D-man D-mannitol
fE control filter extract
Fe-AW Fe quantum satis 10 mg/cm2 of PM-AW
Fe-VS Fe quantum satis 10 mg/cm2 of PM-VS
GAPDH glyceraldehyde 3-phosphate dehydrogenase
Apoptosis is defined by specific morphological alterations [10] as
well as the externalization of phosphatidylserine (PS) and the
permeabilization of plasma and organelle membranes [11]. These
latter features arise from the activation of caspases which can be
elicited by two distinct mechanisms: an extrinsic pathway, which
involves transmembrane death receptors, or an intrinsic pathway
which converges on the permeabilization of mitochondrial mem-
branes (MMP). MMP is the checkpoint of the apoptotic process
characterized by a decrease in the mitochondrial transmembrane
potential (DJm), the production of ROS, and the release of cyto-
chrome c [12]. Importantly, these phenomena are affected by
exposure to PM. Studies performed in vivo in rodents or in vitro on
human bronchial epithelial cell lines following exposure to PM or
their components showed an induction of apoptosis that was
characterized by oxidative stress, decreased DJm, caspase-9 acti-
vation, and DNA fragmentation [13e19].
The effects on apoptosis have been attributed to particular
components of PM. The urban aerosol contains PM2.5 and PM0.1,
which consist of a core of elemental carbon that adsorbs some
inorganic components (ammonium, chloride, sulfates, nitrates, and
metals) and organic compounds such as alkanes, alkanoic acids,
aliphatic acids, quinones, polycyclic aromatic hydrocarbons (PAH),
and biological species [6,20]. PM-induced apoptosis often has been
attributed to various organic components, such as PAH, which
activate the aryl hydrocarbon receptor (AhR) [19e21]. Heavy and
transition metals adsorbed on PM also promote apoptosis via ROS
generation, mitochondrial dysfunction, activation of caspases, or
downregulation of antiapoptotic BCL2 proteins [22,23].
Most of the experiments which have shown PM-induced
apoptosis were carried out with high doses of particles [13,21]. In
contrast, we have performed in vitro exposure to PM2.5 at low to
moderate concentrations (i.e. from 1 to 50 mg/cm2, [24]) that could
mimic realistic exposure considering that PM2.5 deposition in the
tracheobronchial region is 2.3 mg/cm2/24 h according to the
modeling data of Li et al. [25]. We previously showed that exposure
to PM samples collected from Paris prevents apoptosis in human
bronchial epithelial cells that were treated with three broad-
spectrum apoptosis inducers (A23187, staurosporine, and oligo-
mycin). We found that the antiapoptotic effect of PM is related to
their chemical composition. The exposure to certain particulate
samples (of which PM-AW, the PM2.5 from Porte d’Auteuil collected
in winter 2003) inhibits apoptosis whereas the exposure to one
chemically-distinct batch (PM-VS, the PM2.5 from Vitry-sur-Seine
collected in summer 2003) does not. We further found that the
antiapoptotic effect of PM-AW is linked to PAH, particularly ben-
zo(a)pyrene (BaP), via activation of AhR [24].
PAH and metallic compounds are notorious effectors of cellular
GPX1 glutathione peroxidase 1
GSTP1 glutathione S-transferase pi 1
HO1 heme oxygenase-1
MMP mitochondrial membrane permeabilization
NQO1 NAD(P)Hequinone oxidoreductase 1
NRF2 Nuclear factor E2-related factor 2
PAH polycyclic aromatic hydrocarbons
PM particulate matter
PM-AW PM2.5 from Porte d’Auteuil collected in winter 2003
PM-VS PM2.5 from Vitry-sur-Seine collected in summer 2003
ROS reactive oxygen species
SOD2 superoxide dismutase 2
XRE xenobiotic response elements
DJm mitochondrial transmembrane potential
exposure to PM by provoking an oxidative stress and activation of
the ROS-sensitive transcription factor NRF2 (or NFE2L2, Nuclear
factor E2-related factor 2) [26,27]. Under basal conditions, NRF2 is
sequestered in the cytoplasm due to its interaction with two KEAP1
(Kelch-like ECH-associated protein 1) molecules, which facilitate its
ubiquitination and proteasomal degradation. Under oxidative
conditions, inducers alter KEAP1 cysteine residues which prevent
its interaction with NRF2 [28]. NRF2 then becomes phosphorylated,
translocates to the nucleus and binds to antioxidant response ele-
ments (ARE) localized in the promoter region of genes which code
for antioxidant, phase II detoxifying enzymes and cytoprotective
proteins such as heme oxygenase-1 (HO1), NAD(P)H-quinone
oxidoreductase 1 (NQO1), glutathione S-transferases (GST), g-glu-
tamyl cysteinyl synthetase, glutathione peroxidases (GPX), and it-
self [29].
Since our previous work demonstrated that exposure to PM2.5
triggers an antiapoptotic effect which is due to the water-soluble
and PAH components adsorbed on particles, the aim of the pre-
sent study was to further elucidate the molecular mechanisms
involved in this cytoprotective effect, with a particular focus on the
transition metals of PM, ROS production, NRF2 activation and
glutathione homeostasis. We found that the antiapoptotic effect
caused by BaP takes place independently of the NRF2 pathway
whereas iron, a typical transition metal of PM, activate the NRF2
transcription factor and upregulate the expression of its target
genes thus allowing the maintenance of GSH levels.
2. Materials and methods
2.1. Particles collection and preparation
Urban PM2.5 were collected in Paris during the summer or
winter of 2003 at two locations: an urban background station in the
Paris suburb of Vitry-sur-Seine (PM-VS) and a curbside traffic sta-
tion at the Porte d’Auteuil which borders the ring road of Paris (PM-
AW) [24]. The chemical and morphological characterization of PM
used here has been reported previously [30]. All particles were
suspended at 2 mg/mL in supplemented DMEM/F12 medium
(Gibco®) and stored at !20 " C. PAH-naked particles were prepared
by extracting the organic compounds with dichloromethane (Sig-
maeAldrich®) and the particle pellets were re-suspended in
DMEM/F12 medium. As an additional check of the method, a con-
trol filter extract (fE) was produced in the same way from unloaded
nitrocellulose filters.
The chemical and morphological characterization of PM used
herein has been previously reported [30]. In all batches, soots are
the most important particles generally consisting, at their emission
 M. Lovera-Leroux et al. / Biochimie 118 (2015) 195e206
in the atmosphere, of single and approximately spherical particles
of a few 10 nm in diameter or are organized in a chain or clusters of
a few 100 nm. To improve the particle dispersion characterization,
extent of the aggregation of PM was determined with the help of
dynamic light scattering (DLS) and scanning electron microscopy
(SEM). Size curves showed that sonicated PM-AW and PM-VS
batches are rather dispersed in DMEM/F-12 medium (Fig. S1A).
Nevertheless, after 4 h of exposure to bronchial cells, PM-AW
consisted of round-shaped aggregates of diameter not exceeding
2 mm as observed by SEM (Fig. S1B).
2.2. Cell culture and treatments
Human bronchial epithelial 16HBE14o-cells, kindly provided by
Dr. D.C. Gruenert (Colchester, VT, USA), were cultured in supple-
mented DMEM/F12 GlutaMAX™ medium (Gibco®) with 15 mM
Hepes, 100 U/mL penicillin (Gibco®), 100 mg/mL streptomycin
(Gibco®), 1.25 mg/mL fungizone (Gibco®), and 2% UltroserG (UG,
Gibco®). Cells were grown at 37 " C, 5% CO2 on coating composed of
2.9% bovine collagen (0.5 mg/cm2 PurecolNatucan), 1% human
fibronectin (SigmaeAldrich®), and 10% bovine serum albumin
(SigmaeAldrich®) diluted in LHC Medium (Gibco®). Cells were
maintained at subconfluence by trypsination with 0.05% Trypsin-
EDTA (Gibco®, 3 min, 37 " C). Prior to particles treatment, UG was
removed to avoid adsorption on PM.
For exposure, after thawing, particles were sonicated three
times for 20 s at 70W (Vibracell, Bioblock Scientific, Illkrich, France)
and directly added onto cells at a standard dose of 10 mg/cm2
(50 mg/mL). Metallic compounds were studied as ionic solutions
from salt powders diluted in ultrapure water: Fe [Ammonium
iron(II) sulfate hexahydrate: (NH4)2Fe(SO4)2, 6H2O], Al (Aluminum
chloride: AlCl3), Zn (Zinc Chloride: ZnCl2) and Mn (Manganese
chloride: MnCl2). BaP was used as the standard PAH (see Table 1 for
the concentrations used).
16HBE cells were exposed for 4 h to PM, BaP or metallic com-
pounds prior to apoptosis induction by A23187 (3 mM, Sigma-
eAldrich®) or Staurosporine (STS, 1 mM, SigmaeAldrich®) for 20 h
H2O2 (1 mM) was used as a positive control for ROS production and
apoptosis induction. The specific iron chelator deferiprone (DFP,
50 mM) and antioxidants (50 mM D-mannitol, 100 mM MnTBAP,
10 mM GSH, and 10 mM GSSG) were added along with PM-AW or
Fe solutions for 4 h before induction of apoptosis.
2.3. Apoptosis measurement by flow cytometry
Apoptotic and oxidative stress parameters were quantified by
flow cytometry performed on CyAn ADP 9C (at the Institut Jacques
Monod) or BD FACSAria II (at the Functional and Adaptive Biology
facility) flow cytometers using 3,30 -dihexyloxacarbocyanine iodide
(DiOC6(3) low staining (2 nM, Life Technologies™) for DJm loss,
2 mM hydroethidine (HE high staining, Life Technologies™) for
cytoplasmic ROS generation, and 10 mg/mL propidium iodide (PI
high staining, SigmaeAldrich®) for determination of plasma
membrane permeabilization as recommended by the
Table 1
Metal and BaP concentrations in PM-AW and PM-VS samples and experimental doses used.
Total metal content Transition metals content
PM-AW (mg/g)a 67.36 16.22
Dose used in experiments (ng/mL)b
PM-VS (mg/g) 193.26 25.92
Dose used in experiments (ng/mL)b
a
From [30].
b
quantum satis 10 mg/cm2 of PM.
197
Nomenclature Committee on Cell Death [31]. The results of
induced-apoptosis were expressed routinely as the percent of
specific apoptosis (SA) calculated according to the following for-
mula [24]:
100 $ ð% treated cells ! % control cellsÞ
% SA ¼
100 ! % control cells
2.4. ROS production
The intrinsic oxidative ability of PM or iron was measured by
dithiothreitol (DTT) assay [32]. PM samples or Fe were incubated
30 min at 37 " C with 200 mM DTT (Euromedex), centrifuged at
3000 g (15 min, 4 " C) and the loss of unoxidized DTT was deter-
mined by its reaction with DTNB (2.5 mM, Euromedex) to form TNB,
which was quantified by measuring absorbance at 405 nm with a
plate reader (FlexStation 3, Molecular Devices). Results are pre-
sented as unoxidized DTT concentrations obtained by subtracting
the values of blank control. PM and Fe were diluted in DMEM/F12
medium without phenol red and devoided of oxygen using a vac-
uum pump.
Intracellular ROS production was assessed using non-
fluorescent probe CM-H2DCF-DA. Once inside the cell, this probe
is oxidized to a highly fluorescent compound H2DCF by a variety of
intracellular peroxides. Prior to treatment with particles and Fe,
16HBE cells were loaded for 45 min with 100 mM CM-H2DCF-DA
(Fisher Scientific) and the fluorescence of H2DCF was analyzed with
a plate reader (FlexStation 3, Molecular Devices). The results
correspond to the percent of control (untreated cells).
2.5. Measurement of intracellular GSH/GSSG
Cells were exposed for 4 h to PM or their metallic compounds
before collection and suspension at 106 cells/mL in PBS. GSH con-
centration was measured using the Glutathione Assay Kit (Sigma-
eAldrich®), according to the manufacturer recommendations.
Briefly, the GSH concentration was measured by following the re-
action of the deproteinized lysate (with 5% 5-sulfoslicycli acid) with
DTNB to produce TNB. The GSSG formed is recycled by glutathione
reductase and NADPH. The GSSG present in samples also reacts, and
the results are reported as the total glutathione (GSSG þ GSH). H2O2
(1 mM) was used as positive control known to induce GSH oxida-
tion. The results are expressed for 106 cells.
2.6. Confocal immunofluorescence
16HBE cells were cultured for 24 h on a Lab-Tek™ Chamber Slide
System (Thermo Scientific) with PM-AW, PM-VS, or Fe-AW. A
specific antibody for NRF2 (ref H-300, Santa Cruz Biotechnologies)
was used on paraformaldehyde-fixed (4% w/v) and 0.05% PBS
Tween 20-permeabilized cells. NRF2 protein was detected with a
goat anti-rabbit IgG-conjugated Alexa® Fluor 488 (Life Technolo-
gies). Actin staining was carried out using Rhodamine Phalloidin
Fe Mn Zn Al BaP
13.20 0.16 1.50 3.40 0.017
660 8 75 170 0.85
21.60 0.54 1.30 20.40 0.0011
1080 27 65 1020 0.055
198 M. Lovera-Leroux et al. / Biochimie 118 (2015) 195e206
(0.8 U/mL, Life Technologies). Nuclei were stained with 0.5 mg/mL
DAPI (40 ,6-Diamidino-2-Phenylindole, Dihydrochloride, Life Tech-
nologies) and slide coverslips were mounted using medium with
DABCO® (Sigma). Immunostaining was evaluated by examination
of slides under a laser confocal microscope (LSM 710, Zeiss, IJM
Facility). Images were assessed using Image J software.
2.7. NRF2 gene silencing
For siRNA transfection, 16HBE cells were seeded at a density of
25 $ 103 cells/cm2 in 48-well plates and transfected twice using the
HiPerFect reagent (Qiagen) with 25 nM of either All Stars Negative
Control siRNA (NS siRNA, Qiagen ref 1027280) or NRF2 (Qiagen ref
SI03246614, SI03246950, SI04223009, SI04320904). After 24 h, the
cells were subjected to our usual protocol: 4 h-PM pretreatement
and/or A23187 (3 mM) for an additional 20 h.
2.8. Real-time quantitative PCR
Total RNA were extracted from 16HBE14o-cells using TriRe-
agent™ according to the manufacturer's instructions. RNA con-
centrations were determined by measuring the optical density at
260 nm using a Nanodrop ND-2000 (Thermo Scientific). Two mg of
total RNA were reverse-transcribed to cDNA for 30 min at 42 " C by
50U of M-MLV reverse transcriptase RNase H (Promega), 20 mg/mL
of oligo-dT primers (Roche Applied Sciences, Meylan, France), and
500 mM of dNTP. 1/10th of the reverse transcription reaction was
then PCR-amplified using 0.5 mM forward and reverse primers (see
Table S1 for sequences), 1$ Light Cycler 480SYBR Green I Master
Mix (Roche) containing Fast Start Taq DNA Polymerase, and DNA
double-strand-specific SYBR Green I dye for product detection and
characterization. Samples were amplified in a Light Cycler 480
(Roche) under the following conditions: 5 min at 95 " C; 45 cycles of
10 s at 95 " C, 20 s at 58 " C, and 20 s at 72 " C; 1 min at 95 " C; and a
ramp from 55 " C to 95 " C. The amount of mRNA transcript was
normalized using two reference genes, glyceraldehyde 3-
phosphate dehydrogenase gene [GAPDH, (GenBank: 2597)] and
b2 microglobulin protein gene [B2M, (GenBank: 567)]. To compare
the levels of mRNA, relative quantification was performed as out-
lined in Pfaffl et al. [33].
target geneðMean control!Mean sampleÞ
ðEfficacity target geneÞDCt
Normalized mRNA ratio ¼ reference geneðMean control!Mean sampleÞ
ðEfficacity reference geneÞDCt
2.9. NQO1 activity
The activity of NQO1 in 25 mg of total proteins was measured
according to the protocol of Ernster as modified by Benson et al.
using DCPIP as a substrate [34]. The reaction mixture contained
25 mM Tris (pH 7.4), 5 mM FAD (flavin adenine dinucleotide, Sig-
maeAldrich®), 200 mM NADH (b-nicotinamide adenine dinucleo-
tide, Sigma) as the immediate electron donor and 40 mM DCPIP
(2,6-dichlorophenolindophenol, SigmaeAldrich®) as the interme-
diate electron acceptor. Assays were carried out in the presence or
absence of 10 mM dicoumarol (3e30 -methylene-bis(4-
hydroxycoumarin, SigmaeAldrich®) since NQO1 activity is
described as the dicoumarol inhibitable decrease in absorbance at
600 nm. The activity is expressed in nanomoles of DCPIP reduced
per minute per milligram of protein using an extinction coefficient
of 21 $ 103 M!1 cm!1. Total protein in cell culture preparations was
determined by the Bradford assay using bovine serum albumin as a
standard.
2.10. Statistical analysis
The results correspond to the mean ± standard error of the
mean of independent experiments. Data were analyzed using one-
way analysis of variance (ANOVA) followed by the Bonferroni post
hoc test to determine which specific means differed (GraphPad
Prism®). Correlations were calculated using Pearson's correlation
(XLstat software). Differences were considered significant if the p
value was less than 0.05.
3. Results
3.1. The iron component of PM is involved in their antiapoptotic
effect
We previously showed that organic compoundseespecially BaP,
the typical PAH compoundeare involved in the reduction of
apoptosis afforded by fine PM collected in Paris [24]. To further
identify the particulate components involved in the antiapoptotic
effect of PM, we exposed human 16HBE bronchial epithelial cells to
10 mg/cm2 of whole PM-AW (antiapoptotic) and PM-VS (non-anti-
apoptotic) or respective PAH-naked particles. A 4 h-exposure to
whole and PAH-naked PM-AW reduced apoptosis triggered by 3 mM
of the calcium ionophore A23187 (Fig. 1A) meaning that PAH are
not the only components involved in the antiapoptotic effect of PM-
AW and that water-soluble components might also participate. A
similar result was observed if apoptosis was triggered by 1 mM of
the PKC inhibitor Staurosporine (STS, Fig. 1B). Contrary to PM-AW, if
a sonicated filter extract as well as whole or PAH-naked PM-VS
were used, they did not lessen apoptosis induced by A23187 or STS.
Focusing on water-soluble components, we studied three tran-
sition metals (Fe, Mn, Zn) and one post-transition metal (Al) at
concentrations found in 10 mg/cm2 of either PM-AW or PM-VS (see
Table 1). Iron, manganese, zinc, aluminum and a sonicated control
filter extract did not affect cell viability per se (Fig. S2). However,
these metals, except for zinc, significantly reduced apoptosis trig-
gered by A23187 as observed with whole PM-AW (Fig. 1C). Fe also
prevented STS-induced apoptosis when specifically used quantum
satis 10 mg/cm2 of PM-AW (data not shown). In contrast, when
transition metals were introduced at doses found in 10 mg/cm2 of
PM-VS, no reduction of A23187-induced apoptosis was observed.
The ineffectiveness of Zn in reducing apoptosis is consistent with
the similar amount of this element in AW and VS particles (see
Table 1).
In order to evaluate the importance of transition metals, and
particularly iron, cells were exposed to PM-AW or Fe-AW (660 ng/
mL) along with the specific iron chelator deferiprone (50 mM DFP,
Fig. 1D). DFP blocked the antiapoptotic effect of PM-AW and Fe-AW
partially or completely, respectively. This suggests that the iron
component of PM is involved in their cytoprotective effect. In order
to determine if iron could act in combination with PAH, we exposed
cells to a mixture of Fe and BaP at doses found in 10 mg/cm2 of either
 M. Lovera-Leroux et al. / Biochimie 118 (2015) 195e206 199

Fig. 1. Antiapoptotic effect of particulate matter and their metallic components. Human bronchial epithelial cells 16HBE were exposed for 4 h to 10 mg/cm2 of whole particles or
PAH-naked PM-AW or PM-VS before apoptosis induction by 3 mM A23187 (A) or 1 mM Staurosporine (STS, B) for an additional 20 h. The percentage of induced apoptosis was
determined by flow cytometry using the decrease in DJm and the production of ROS. Dashed line: values of apoptosis triggered by A23187 and STS are 70.4 ± 1.3% and 78.6 ± 1.6%,
respectively. The results are the means ± SEM (n ¼ 4). * Statistically different from the Control condition (i.e. A23187 or STS alone) with p < 0.001. (C) 16HBE cells were exposed for
4 h to Fe, Mn, Zn and Al quantum satis 10 mg/cm2 of PM-AW or PM-VS (see Table 1 for the concentrations used) and then apoptosis was induced by A23187 and quantified as above.
Dashed line: value of apoptosis induced by A23187 is 64.9 ± 3.1%. The results are the means ± SEM (n ¼ 4). * Statistically different from the Control condition (i.e. A23187 alone) with
p < 0.01; x vs. PM-AW condition with p < 0.001. (D) 16HBE cells were exposed for 4 h to PM-AW (10 mg/cm2) or Fe quantum satis 10 mg/cm2 of PM-AW (Fe-AW) in absence or
presence of deferiprone (DFP, 50 mM) before induction of apoptosis with 3 mM A23187. Dashed line: value of A23187-induced apoptosis: 63.6 ± 2.7%. The results are the
means ± SEM (n ¼ 3); * vs. control conditions, x vs. PM-AW or Fe-AW without DFP with p < 0.001. £ vs. PM-AW with DFP with p < 0.001. (E) Effects of the combination of Fe and BaP.
A 4 h-exposure to PM-AW, PM-VS (10 mg/cm2), Fe, BaP or a combination of Fe and BaP quantum satis 10 mg/cm2 of particles was performed prior to induction of apoptosis (3 mM
A23187, 20 h). Apoptosis was quantified as above. Dashed line: value of A23187-induced apoptosis: 69.5 ± 1.4%. The results are the means ± SEM (n ¼ 4). *p < 0.001 vs. the Control
condition (i.e. A23187 alone) and xp < 0.05 vs. PM-AW exposure.
200 M. Lovera-Leroux et al. / Biochimie 118 (2015) 195e206

PM-AW or PM-VS (Fig. 1E). Whereas Fe and BaP used individually

partially counteracted A23187-induced apoptosis, the combination
of both prevented cell death as efficiently as whole PM-AW. Note
that no rescue was observed with the mixture of Fe and BaP
quantum satis PM-VS.

3.2. ROS generation and the cellular antioxidant response

Numerous in vitro studies have demonstrated the modulation of

apoptosis by air pollutants, including metals, in relation to their
ability to produce ROS [22,35,36]. We have previously demon-
strated the production of ROS in human bronchial epithelial cells
exposed to fine particulate matter [30,37]. By assessing the intrinsic
oxidative potential of PM and their iron component, we showed
that PM-AW and PM-VS oxidized DTT in a dose-dependent manner
with a similar effect; the half-maximal effective concentrations
(EC50) were 45.0 mg/cm2 and 45.6 mg/cm2, respectively (Fig. 2A).
We observed a significant loss of unoxidized DTT with Fe-AW but
Fe-VS was ineffective. We have extended these results in 16HBE
cells exposed for 4 h to PM-AW, PM-VS and Fe (quantum satis
1e10 mg/cm2). Cells exposed to PM-AW, PM-VS and solely to Fe-AW
showed a significant increase of ROS levels with a dose dependent
manner (Fig. 2B) which suggests that ROS production might be
involved in the cytoprotective effect of particles and iron. In order
to identify relationships between the inhibition of apoptosis and
the ROS production, correlation analyses were performed (Table 2).
Statistically significant correlations were calculated between
apoptosis inhibition, DTT oxidation potential and cellular ROS
production with PM-AW and Fe-AW while no correlations were
found for PM-VS. This suggests that the antiapoptotic effect of PM-
AW and iron is linked to their ability to generate ROS and to modify
the cellular redox homeostasis. This assumption was further
strengthened by a study of the doseeresponse effects of H2O2 to
A23187-induced apoptosis (Fig. 2C). Low doses of hydrogen
peroxide (i.e., 10e100 mM) reduced apoptosis to an extent similar to
those of whole PM-AW or Fe-AW.
Considering Li's model of hierarchical oxidative stress response
[25], we hypothesized that PM-AW might trigger low levels of
cellular ROS and then an adaptive antioxidant response might
possibly be involved in the establishment of an antiapoptotic pro-
cess. To confirm this, cells were treated with PM-AW and/or
different antioxidants such as a specific hydroxyl radical scavenger
(D-mannitol), the superoxide dismutase mimetic and peroxynitrite
scavenger (MnTBAP), reduced glutathione (GSH), and oxidized
glutathione (GSSG). Antioxidants did not modify apoptosis trig-
gered by A23187 alone (Fig. 3A). Nevertheless, in the presence of
PM-AW, MnTBAP and GSH further reduced apoptosis, suggesting
Fig. 2. Role of cellular ROS production. (A) Inherent oxidative potential of PM-AW
that these antioxidants act together with PM. Furthermore, the (1e50 mg/cm2), PM-VS, Fe-AW (quantum satis 1e50 mg/cm2 of PM-AW) and Fe-VS
depletion in total glutathione levels observed during A23187- (quantum satis 1e50 mg/cm2 of PM-VS) was evaluated by the DTT oxidation assay.
induced apoptosis was prevented when 16HBE cells were Results are presented as unoxidized DTT concentrations obtained by substracting the
exposed for 4 h with 10 mg/cm2 PM-AW or Fe-AW (Fig. 3B). This values of blank control. Data are represented as means ± SEM from 8 different ex-
periments performed in triplicates. Dashed line: initial amount of DTT (200 mM).
effect was specific to antiapoptotic PM-AW since it was not
*p < 0.05, statistically different from control (195.1 ± 10.1 mM). (B) Cellular ROS gen-
observed in the presence of non-antiapoptotic PM-VS. Thus, these eration was evaluated by the fluorescence of H2DCF after a 4 h-exposure to PM-AW,
findings evoke a key role of GSH in the antiapoptotic effect induced PM-VS (1e10 mg/cm2), Fe-AW (i.e. 66, 330 and 660 ng/mL), Fe-VS (i.e. 108, 540 and
by airborne particles. 1080 ng/mL). The means ± SEM from 3 different experiments are shown. *p < 0.05,
statistically different from Control. (C) Dose-response analysis of the effect of H2O2 on
A23187-induced apoptosis. 16HBE cells were exposed for 4 h to H2O2 (0e500 mM)
3.3. Role of the NRF2 pathway before induction of apoptosis by A23187 (3 mM) for an additional 20 h. Dashed line:
value of A23187-induced apoptosis induced is 64.4 ± 11.4%. The means ± SEM from 3
To further investigate the molecular mechanisms involved in different experiments are shown. * Statistically different from the control condition
the antiapoptotic effect of PM, we determined the localization by (A23187 alone without H2O2) with p < 0.01.

confocal microscopy of NRF2, a prominent contributor to the

antioxidant response. NRF2 was mainly localized in the cytoplasm was observed following Fe-AW exposure whereas PM-VS and BaP-
of untreated cells while exposure to PM-AW for 4 h resulted in AW were not found to activate NRF2 (Fig. 4B). NRF2 activation
nuclear localization of NRF2, indicating that this transcription fac- occurred rapidly after 30 min of treatment with PM-AW, but more
tor was activated under these conditions (Fig. 4A). A similar result slowly for Fe-AW (4 h). To assess the involvement of NRF2 in the
 M. Lovera-Leroux et al. / Biochimie 118 (2015) 195e206 201

Table 2
Correlation coefficients (Pearson) between inhibition of A23187-induced apoptosis and the production of ROS.

PM-AW Apoptosis DTT Cellular ROS Fe-AW Apoptosis DTT Cellular ROS PM-VS Apoptosis DTT Cellular ROS
inhibition oxidation production inhibition oxidation production inhibition oxidation production

Apoptosis 1 Apoptosis 1 Apoptosis 1

inhibition inhibition inhibition
DTT oxidation 0,471 1 DTT oxidation 0,295 1 DTT oxidation !0.020 1
Cellular ROS 0,562 0293 1 Cellular ROS !0.238 ¡0,551 1 Cellular ROS !0.042 0.126 1
production production production

Data are presented for dose-effect experiments with PM-AW, PM-VS (1, 5 and 10 mg/cm2) and Fe-AW (66, 330 and 660 ng/mL). Inhibition of A23187-induced apoptosis was
quantified in cellulo by flow cytometry using dissipation of DJm and superoxide anion production. DTT oxidation refers to amounts of oxidized DTT assessed in vitro. Cellular
ROS production was evaluated by the CM-H2DCF-DA assay in 16HBE cells after a 4 h-exposure. Bold values represent a statistical correlation (p < 0.05).

detoxifying target genes: HO1, NQO1, SOD2 (the inducible mito-

Fig. 3. Effect of antioxidants. (A) After a 4 h-exposure to PM-AW (10 mg/cm2) and/or
antioxidants (50 mM D-mannitol, 100 mM MnTBAP, 10 mM GSH, and 10 mM GSSG),
apoptosis was triggered by A23187 (3 mM, 20 h). Dashed line: value of apoptosis
induced by A23187(70.5 ± 3.3%). The results are the means ± SEM (n ¼ 6). *p < 0.01,
statistically different from Control; x Statistically different from A23187 þ PM-AW with
p < 0.01. (B) Total glutathione (GSH þ GSSG) level. Cells were exposed or not to PM-AW
(10 mg/cm2), Fe-AW (660 ng/mL), or PM-VS (10 mg/cm2) for 4 h before induction of
apoptosis (A23187 3 mM, 2 h). H2O2 (1 mM, 4 h) was used as positive control. Dashed
line: level of total glutathione with A23187 alone (8.7 ± 2.8 nmol/106 cells). The
means ± SEM (n ¼ 3) are shown. * Significantly different from the control with p < 0.01
and xp < 0.05, significantly different from PM-AW without antioxidants.
chondrial Mn-SOD), cytosolic GSTP1, and GPX1. As shown, NRF2,
HO1, and NQO1 mRNA expression levels were increased after a 3-h
exposure to PM-AW, and as soon as 30 min in the case of GPX1
(Fig. 5A). Contrary to SOD2 and GSTP1 mRNA whose levels were
not altered by PM-AW and Fe-AW, an exposure to iron during 4 h
increased the expression of NRF2, HO1, and NQO1 to an extent
similar to that observed for whole particles (Fig. 5B) suggesting
that Fe contributes to the activation of the NRF2 pathway. The
activation of the NRF2 pathway seemed to be specific to the
transition metal components of PM since the expression of NRF2
and its target genes was not modulated by BaP (Fig. S3). This
supposition was reinforced by the significant increase of NQO1
activity observed after 4 h of exposure to PM-AW and PAH-naked
PM-AW (Fig. 5C). Lastly, expression of the antiapoptotic gene BCL2
increased from 30 min to 2 h of exposure to PM-AW (Fig. 5A)
whereas A23187 did not alter the expression when used alone
(Fig. S4). The lengths of times of exposure which lead to increases
in expression of mRNAs were consistent with the minimal dura-
tion of pre-treatment by PM-AW required to observe their anti-
apoptotic effects (Fig. S5). Indeed, at least 1 h of pretreatment to
PM-AW was required and a maximal effect was seen after 4 h of
exposure, which supports the hypothesis that 16HBE cells need to
synthesize antioxidant and detoxifying enzymes for the estab-
lishment of the antiapoptotic effect that is triggered by PM and
transition metal components.
Finally, we investigated whether upregulation of antioxidant
genes was altered during the early phase of A23187-induced
apoptosis. Upregulation of NRF2, HO1, NQO1, and GPX1 triggered
by PM-AW and Fe-AW was not modified by addition of A23187
(Fig. 6A and B) in accordance with the lack of regulation of those
genes by A23187 alone (see Fig. S4). Moreover, the expression of
proapoptotic genes BAX and p53 was reduced by the PM-AW
compared to A23187-induced expression (Fig. 6A and Fig. S4). For
instance, the normalized expression for the p53 gene was 2.9 ± 0.3
following treatment with A23187 for 0.5 h as compared to 1.8 ± 0.5
following treatment with PM-AW þ A23187 for 0.5 h (p < 0.05).
Furthermore, BAX expression was markedly reduced by PM-AW
after a 4-h exposure and during the first half hour of apoptosis
induction by A23187 (Fig. 6A) thus corroborating the antiapoptotic
properties of PM-AW.

antiapoptotic response of PM, we silenced the NRF2 gene with
siRNA. NRF2 silencing reduced, partially but significantly, the pro-
tection to apoptosis afforded by PM-AW (20% vs. 47% with NS
siRNA, Fig. 4C). These results suggest that activation of NRF2 might
be a decisive molecular step in the antiapoptotic effect of urban fine
PM.
Given that NRF2 adjusts its own expression, we analyzed the
NRF2 mRNA expression as well as some of its antioxidant and
4. Discussion
Although the antiapoptotic effects of fine and ultrafine particles
have been investigated increasingly, the results remain insufficient
to clearly identify the associated mechanisms, and the study of
which is needed for understanding PM as a carcinogen [38]. Our
goal is to elucidate the mechanisms underlying the antiapoptotic
effect of fine PM and their associated components. We have pre-
viously shown that the organic components of PM activate the AhR
202 M. Lovera-Leroux et al. / Biochimie 118 (2015) 195e206

Fig. 4. Involvement of the NRF2 pathway. (A) Intracellular localization of NRF2 was determined by confocal immunofluorescence analysis of NRF2 protein (green), actin (red), and
nucleus (blue) detected on cells exposed for 4 h to PM-AW (10 mg/cm2), Fe-AW (660 ng/mL), or PM-VS (10 mg/cm2). Merged images of nuclei and NRF2 showed a nuclear localization
(white arrows). (B) Quantification of NRF2 nuclear localization in cells treated for the indicated times with 10 mg/cm2 of PM-AW or PM-VS, Fe-AW (660 ng/mL) or BaP-AW (0.85 ng/
mL). The means ± SEM from 3 different experiments are shown. *Statistically different from the control condition (7.34 ± 1.62) with p < 0.05. (C) Effect of NRF2 silencing. 16HBE cells
were incubated for 24 h with NRF2 siRNA (25 nM), a control siRNA (NS siRNA, 25 nM) or were not transfected (NT) and then pretreated 4 h to PM-AW before induction of apoptosis
(3 mM A23187, 20 h). Dashed line: value of A23187-induced apoptosis in NT PM-AW condition (37.9 ± 0.7%). The results are the means ± SEM (n ¼ 3). * vs. Co. condition, x vs. NS
siRNA PM-AW with p < 0.05 and £ vs. NRF2siRNA Co. condition with p < 0.05.
 M. Lovera-Leroux et al. / Biochimie 118 (2015) 195e206 203

Fig. 5. Regulated expression of NRF2 and its target genes. Normalized mRNA ratios for NRF2, HO1, NQO1, SOD2, GSTP1, GPX1 and BCL2 were determined by quantitative PCR in 16HBE
cells exposed for 0.5e4 h to 10 mg/cm2 of PM-AW (A), Fe-AW (660 ng/mL) (B). The means ± SEM (n ¼ 3) are shown and * p < 0.001, significantly different from the control. (C) NQO1
enzymatic activity was measured in cells exposed or not to 10 mg/cm2 of PM-AW or PAH-naked PM-AW during 4 h. Dashed line: value of the control: 68.0 ± 10.9 nmol/min/mg. The
means ± SEM (n ¼ 3) are shown. * Significantly different from control with p < 0.05.

pathway [24]. In the present study, we focus on the metallic com- occurs through NRF2 and antioxidant pathways.
ponents and particularly Fe, and we show that its contribution to In agreement with our previous publication, we found that PAH
the PM's antiapoptotic effect on human bronchial epithelial cells are not the only components involved in the antiapoptotic effect of

Fig. 6. mRNA expression during A23187-induced apoptosis. Normalized mRNA expressions for NRF2, HO1, NQO1, GPX1, BAX and p53 were determined by quantitative PCR in 16HBE
cells exposed for 4 h to 10 mg/cm2 of PM-AW (A) or 660 ng/mL Fe-AW (B) before apoptosis was induced by A23187 (3 mM) for the specified times. The means ± SEM (n ¼ 3) are
shown,*p < 0.01, significantly different from control and x vs. PM-AW alone.
204 M. Lovera-Leroux et al. / Biochimie 118 (2015) 195e206
PM since PAH-naked PM were as effective as whole PM in reducing
chemically-induced apoptosis. Consistent with this, three of the
four metals investigated (Al, Fe, Zn and Mn) inhibit apoptosis at
concentrations found in 10 mg/cm2 of antiapoptotic PM-AW. The
inhibition of apoptosis triggered by PM-AW and Fe is associated
with enhanced expression of the antiapoptotic gene BCL2 and the
concomitant reduced expression of the apoptotic genes BAX and
p53.
To elucidate the antiapoptotic mechanism of transition metals,
we focused on iron because of its effect on oxidative stress [23,39]
and because of its different concentrations that are found in anti-
apoptotic (PM-AW) versus non-antiapoptotic (PM-VS) particles. At
the dose of Fe-AW (660 ng/mL), we observed an antiapoptotic ef-
fect whereas no effect was seen at the higher dose of Fe-VS
(1080 ng/mL). In agreement with our results, Wang et al. also
found that a low dose of Fe (20 mg/kg) suppresses lead-induced
apoptotic neurotoxicity in vivo through activation of the MAPK
pathway whereas a high dose (40 mg/kg) is ineffective [40]. Our
data suggest that although dose might be a key parameter, the
combination of metallic ions with PAH (e.g. BaP) seems to be crucial
to completely mimic the cytoprotective effect of PM. Moreover, the
lack of an antiapoptotic effect for PM-VS might be due, in addition,
to the fact that particles are, in reality, a complex mixture of many
components, each with potentially pro- or antiapoptotic conse-
quences. Thus, the overall effect of the particles on apoptosis is the
result of the combined effects of all of the components.
Although some reports indicate that exposure to metals induces
cell death through ROS generation [22,23], other toxicological
studies have shown that iron, nickel, zinc, arsenic, and cadmium
reduce apoptotic cell death in vivo or in vitro [35,40,41]. Differences
in cell type, time of exposure, and the dose used for metal exposure,
may have contributed to these divergent results. Herein, we
observed a generation of cellular ROS which was triggered by PM-
AW, PM-VS, and Fe commensurate with our previous observations
of a production of hydrogen peroxide and hydroxyl radical by PM-
AW, PM-VS, aqueous and organic extracts thereof or a combination
of both [30,42]. These findings support previous studies on
airborne particles which found that although the PAH components
contribute to the intrinsic oxidative potential of PM [43], up to 80%
of the effect could be due to metallic components and, in particular,
Fe, Cu and Mn [32].
Knowing that ROS are apoptotic modulators [44], we hypothe-
sized that metallic compounds, and particularly iron, provoke an
antioxidant adaptive response that leads to the antiapoptotic effect
of PM [39]. This assumption was supported by the statistically
significant correlation shown between the inherent and cellular
ROS generation and apoptosis inhibition triggered by of PM-AW
and Fe-AW as well as the reduction of A23187-induced apoptosis
that we observed with physiological doses of H2O2 (i.e.,
10e100 mM). Identical concentrations of hydrogen peroxide
decrease apoptosis induced by serum-depletion in endothelial cells
suggesting that H2O2 is a signaling molecule that downregulates
apoptosis. Haendeler et al. showed that the antiapoptotic effect of
H2O2 is mediated through increased levels of thioredoxin-1 mRNA
and protein [45], its nuclear import and upregulation of GSTP1
expression [46].
In accordance with the first step of the hierarchical response to
oxidative stress [25,39] and the well documented role of NRF2 in
cell survival [28], we observed nuclear translocation of NRF2 with
PM-AW and Fe-AW in contrast to the non-antiapoptotic particles
PM-VS and BaP-AW. Consistent with our hypothesis, Kode et al.
showed that 100 mM H2O2 activated NRF2 nuclear translocation
after 1 h of exposure in human primary small airway epithelial cells
[47]. Further, nuclear translocation of NRF2 was achieved before
upregulation of its own mRNA [48] and that of its target antioxidant
genes HO1 [49], NQO1 [50], and GPX1. Similarly to other data from
respiratory cells [51,52], we found that exposure to PM-AW and Fe-
AW induced expression of NRF2, HO1, and NQO1 following an
exposure of as short as 2 h and PM-AW upregulated GPX1 expres-
sion within 30 min. The absence of an effect on SOD2 and GST
expression following exposures of 30 min to 4 h to PM-AW does not
conflict with our hypothesis since another study has shown in-
duction of SOD in human alveolar epithelial cells only after 6 h of
exposure to PM [53].
Li et al. highlighted a strong positive correlation between the
particulate matter redox activity and their capability to induce HO1
protein expression [43]. HO1 is the most markedly induced enzyme
that we observed in this study. It is of particular interest since its
expression and activity are dramatically increased in mitochondrial
fractions of human bronchial cells exposed to cigarette smoke
extract. Further, it is known to inhibit cell death and to maintain
ATP levels [44]. The cytoprotective mechanism of HO1 involves its
enzymatic reaction products including biliverdin, carbon monoxide
(CO), and ferrous iron [39]. Additionally to restrain the mitochon-
drial translocation of BCL2 family proteins and Cytochrome c
release, Queiroga et al. demonstrated that CO may directly act on
mitochondria since a pretreatment with low doses of CO reduces
the calcium-induced swelling and depolarization of isolated mito-
chondria [54]. In agreement with a previous work which showed
that a 4 h-exposure to antiapoptotic particles (i.e. PM-AW) mark-
edly induced the expression NQO1 whereas non-antiapoptotic PM-
VS produced only a low-level induction [42], we also observed an
increase in the NQO1 activity triggered by PM-AW and PAH-naked
PM-AW. Given that NQO1 is a cytoprotective flavoprotein which
functions as a superoxide scavenger through the reduction of qui-
nones to hydroquinones thus preventing the one-electron reduc-
tion of quinones to semiquionones [55], this enzyme appears to be
an important molecular factor participating in the antiapoptotic
effect of PM.
The important role played by GSH in redox homeostasis
prompted us to examine its involvement in the PM-induced anti-
apoptotic mechanism. Among the four antioxidants tested, the
superoxide dismutase mimetic and peroxynitrite scavenger
MnTBAP, and reduced gluthathione GSH, but not the hydroxyl
radical scavenger D-mannitol, were capable of significantly
increasing the antiapoptotic effect. This suggests that SOD and GSH
induce an adaptive antioxidant response and that they act in syn-
ergy with PM-AW to obtain a cytoprotective effect. Further, quan-
tification of the amount of total intracellular glutathione revealed
that GSH levels were specifically restored by PM-AW and Fe-AW
during A23187-induced apoptosis. Interestingly, Michael et al.
also described increased levels of GSH in a human alveolar
epithelial cell line (A549) exposed for 4 h to urban traffic PM10 [53].
Other data demonstrate that PM or its aqueous components in-
crease H2O2 and GSH production [53,56]. Therefore, we propose
that the iron component of PM generate ROS that trigger the
antioxidant cellular response by increasing the levels of gluta-
thione. Intracellular glutathione is found predominantly in its
reduced form and mitochondrial GSH (mGSH) represents only a
minor fraction of the total pool of GSH pool (10e15%). We believe
that by maintaining the level of total glutathione, PM-AW and Fe-
AW might allow the maintenance of the level of mGSH which ari-
ses from the cytosolic GSH by specific transporters located in the
mitochondrial inner membrane. This would prevent the mito-
chondrial membrane permeabilization step of apoptosis as previ-
ously shown [57].
As our data provide some hints that the antiapoptotic effect of
iron might contribute to particle carcinogenicity, it should be keep
in mind that the metals and especially transition metals, are also
known to induce carcinogenesis through ROS production and
 M. Lovera-Leroux et al. / Biochimie 118 (2015) 195e206
generation of oxidized macromolecules (DNA, proteins, lipids) [58].
In general, impacts on DNA include DNA base modifications, sugar
lesions, single- and double-DNA strand breaks, DNAeprotein
crosslinks, DNAeDNA crosslinks, and abasic sites. For instance, the
genotoxic effects of iron are mainly derived from hydroxyl radical
production by Fenton and Haber-Weiss chemistry, leading to base
oxidation (as 8-hydroxy-20 -deoxyguanosine), DNA strand breaks
and micronuclei, which are then mutagenic [59]. Somatic muta-
tions can lead to a sustained proliferative signaling or evasion to
antiproliferative signaling, two of the capabilities that enable tumor
growth and metastatic dissemination [8,9].
The results of this study taken together with our previously
published work suggest an additive effect of the AhR and NRF2
pathways both of which are activated by PM. However, we did not
detect activation of NRF2 or expression of its HO1 and NQO1 target
genes by BaP as was seen by Niestroy et al. [26]. These divergent
results can be explained most likely by differences in the doses and/
or times of exposure since the concentrations that we used are
much lower (0.34 mM vs. 10 mM) as well as the duration of treatment
(4 h vs. 48 h). Nevertheless, an interaction between the AhR and
NRF2 signaling pathways is likely since BaP downregulates KEAP1
transcription and, thereby, leads to NRF2 activation [60]. Moreover,
Miao et al. suggested that NRF2 might be an AhR target gene [61],
and others have shown a physical interaction between AhR and
NRF2 as co-factors for ARE-dependent genes [62].
To conclude, this study expands on our previous work, which
demonstrated the role of the PAH-AhR pathway in the anti-
apoptotic effect of atmospheric PM exposure [24], by identifying
the metallic components, specifically Fe, which contribute to the
antiapoptotic effect of urban fine PM. The ensemble of our results
lead us to propose a model in which metals generate ROS that elicit
the activation of the pro-survival NRF2 protein and the upregula-
tion of antioxidant enzymes HO1, NQO1, and GPX1. In addition, in
this model, the regulation of redox homeostasis by the mainte-
nance of GSH levels would counteract mitochondrial membrane
permeabilization and subsequent apoptosis. Complementarily,
desorbed PAH activate the AhR pathway, which can regulate the
expression of xenobiotic response element (XRE)-dependent genes.
In summary, this study highlights the molecular mechanism
whereby transition metals, and particularly Fe, contribute to the
antiapoptotic effect of urban PM in addition to the PAH-AhR
signaling pathway. Knowing that resisting cell death is one of the
hallmarks of cancer cells [8,9], a fuller understanding of the cellular
and molecular mechanisms by which atmospheric particles impede
cell death can provide clues to explain the development of lung
cancers after long-term PM exposure.
Competing interests
The authors declare that there is no conflict of interest.
Funding information
This work was supported by Agence Nationale de la Recherche
[0599-5 SET 024-01], Centre National de la Recherche Scientifique
(CNRS), Universite
! Paris Diderot-Paris 7 (Bourse de Master, Melanie
Lovera-Leroux) and Renault (DIMAT, for the supply and chemical
analysis of PM batches).
Acknowledgments
We thank Blandine Gaussere "s (BFA facility) and Nicole Boggetto
(IJM facility) for their technical assistance on flow cytometers as
well as The !o Loustand and Benoit Gravade. We also acknowledge
Isabelle George and the confocal microscope platform in the
205
Institute Jacques Monod, Paris, France. We gratefully acknowledge
Ioana Ferecatu and Lawrence Aggerbeck for critical reading and
correction of the manuscript.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.biochi.2015.09.030.
References
[1] G. Hoek, R.M. Krishnan, R. Beelen, A. Peters, B. Ostro, B. Brunekreef, et al.,
Long-term air pollution exposure and cardio- respiratory mortality: a review,
Environ. Health 12 (2013) 43, http://dx.doi.org/10.1186/1476-069X-12-43.
[2] O. Raaschou-Nielsen, Z.J. Andersen, R. Beelen, E. Samoli, M. Stafoggia,
G. Weinmayr, et al., Air pollution and lung cancer incidence in 17 European
cohorts: prospective analyses from the European Study of Cohorts for Air
Pollution Effects (ESCAPE), Lancet. Oncol. 14 (2013) 813e822, http://
dx.doi.org/10.1016/S1470-2045(13)70279-1.
[3] Y. Chen, A. Ebenstein, M. Greenstone, H. Li, Evidence on the impact of sus-
tained exposure to air pollution on life expectancy from China's Huai River
policy, Proc. Natl. Acad. Sci. U. S. A. 110 (2013) 12936e12941, http://
dx.doi.org/10.1073/pnas.1300018110.
[4] A. Churg, M. Brauer, M. del Carmen Avila-Casado, T.I. Fortoul, J.L. Wright,
Chronic exposure to high levels of particulate air pollution and small airway
remodeling, Environ. Health Perspect. 111 (2003) 714e718.
[5] M. Lodovici, E. Bigagli, Oxidative stress and air pollution exposure, J. Toxicol.
2011 (2011) 487074, http://dx.doi.org/10.1155/2011/487074.
[6] S. Val, C. Liousse, E.H.T. Doumbia, C. Galy-Lacaux, H. Cachier, N. Marchand, et
al., Physico-chemical characterization of African urban aerosols (Bamako in
Mali and Dakar in Senegal) and their toxic effects in human bronchial
epithelial cells: description of a worrying situation, Part. Fibre Toxicol. 10
(2013) 10, http://dx.doi.org/10.1186/1743-8977-10-10.
[7] International Agency for Research on Cancer, IARC: Outdoor Air Pollution a
Leading Environmental Cause of Cancer Deaths 221, Press Release N" , 2013.
[8] D. Hanahan, R.A. Weinberg, The hallmarks of cancer, Cell 100 (2000) 57e70.
[9] D. Hanahan, R.A. Weinberg, Hallmarks of cancer: the next generation, Cell 144
(2011) 646e674, http://dx.doi.org/10.1016/j.cell.2011.02.013.
[10] J.F. Kerr, A.H. Wyllie, A.R. Currie, Apoptosis: a basic biological phenomenon
with wide-ranging implications in tissue kinetics, Br. J. Cancer 26 (1972)
239e257.
[11] G. Kroemer, P. Petit, N. Zamzami, J.L. Vayssie "re, B. Mignotte, The biochemistry
of programmed cell death, FASEB J. 9 (1995) 1277e1287.
[12] L. Galluzzi, N. Zamzami, T. de La Motte Rouge, C. Lemaire, C. Brenner,
G. Kroemer, Methods for the assessment of mitochondrial membrane per-
meabilization in apoptosis, Apoptosis 12 (2007) 803e813, http://dx.doi.org/
10.1007/s10495-007-0720-1.
[13] E. Alfaro-Moreno, L. Martínez, C. García-Cuellar, J.C. Bonner, J.C. Murray,
I. Rosas, et al., Biologic effects induced in vitro by PM10 from three different
zones of Mexico City, Environ. Health Perspect. 110 (2002) 715e720.
[14] N. Agopyan, J. Head, S. Yu, S.A. Simon, TRPV1 receptors mediate particulate
matter-induced apoptosis, Am. J. Physiol. Lung Cell. Mol. Physiol. 286 (2004)
L563eL572, http://dx.doi.org/10.1152/ajplung.00299.2003.
[15] J.-H. Choi, J.-S. Kim, Y.-C. Kim, Y.-S. Kim, N.-H. Chung, M.-H. Cho, Comparative
study of PM2.5-and PM10-induced oxidative stress in rat lung epithelial cells,
J. Vet. Sci. 5 (2004) 11e18.
[16] F. Farina, G. Sancini, P. Mantecca, D. Gallinotti, M. Camatini, P. Palestini, The
acute toxic effects of particulate matter in mouse lung are related to size and
season of collection, Toxicol. Lett. 202 (2011) 209e217, http://dx.doi.org/
10.1016/j.toxlet.2011.01.031.
[17] G. Garçon, Z. Dagher, F. Zerimech, F. Ledoux, D. Courcot, A. Aboukais, et al.,
Dunkerque City air pollution particulate matter-induced cytotoxicity, oxida-
tive stress and inflammation in human epithelial lung cells (L132) in culture,
Toxicol. Vitro 20 (2006) 519e528, http://dx.doi.org/10.1016/j.tiv.2005.09.012.
[18] J.M. Mate !s, J.A. Segura, F.J. Alonso, J. Ma !rquez, Roles of dioxins and heavy
metals in cancer and neurological diseases using ROS-mediated mechanisms,
Free Radic. Biol. Med. 49 (2010) 1328e1341, http://dx.doi.org/10.1016/
j.freeradbiomed.2010.07.028.
[19] T. Xia, P. Korge, J.N. Weiss, N. Li, M.I. Venkatesen, C. Sioutas, et al., Quinones
and aromatic chemical compounds in particulate matter induce mitochon-
drial dysfunction: implications for ultrafine particle toxicity, Environ. Health
Perspect. 112 (2004) 1347e1358.
[20] S. Billet, G. Garçon, Z. Dagher, A. Verdin, F. Ledoux, F. Cazier, et al., Ambient
particulate matter (PM2.5): physicochemical characterization and metabolic
activation of the organic fraction in human lung epithelial cells (A549), En-
viron. Res. 105 (2007) 212e223, http://dx.doi.org/10.1016/
j.envres.2007.03.001.
[21] Z. Andrysík, J. Vondra cek, S. Marvanova
!# !, M. Ciganek, J. Ne#ca, K. Pe
#n#
cíkova
!, et
al., Activation of the aryl hydrocarbon receptor is the major toxic mode of
action of an organic extract of a reference urban dust particulate matter
mixture: the role of polycyclic aromatic hydrocarbons, Mutat. Res. 714 (2011)
206 M. Lovera-Leroux et al. / Biochimie 118 (2015) 195e206
53e62, http://dx.doi.org/10.1016/j.mrfmmm.2011.06.011.
[22] M.D. Pulido, A.R. Parrish, Metal-induced apoptosis: mechanisms, Mutat. Res.
533 (2003) 227e241.
[23] S.V.S. Rana, Metals and apoptosis: recent developments, J. Trace Elem. Med.
Biol. 22 (2008) 262e284, http://dx.doi.org/10.1016/j.jtemb.2008.08.002.
[24] I. Ferecatu, M.-C. Borot, C. Bossard, M. Leroux, N. Boggetto, F. Marano, et al.,
Polycyclic aromatic hydrocarbon components contribute to the mitochondria-
antiapoptotic effect of fine particulate matter on human bronchial epithelial
cells via the aryl hydrocarbon receptor, Part. Fibre Toxicol. 7 (2010) 18.
[25] N. Li, M. Hao, R.F. Phalen, W.C. Hinds, A.E. Nel, Particulate air pollutants and
asthma. A paradigm for the role of oxidative stress in PM-induced adverse
health effects, Clin. Immunol. 109 (2003) 250e265.
[26] J. Niestroy, A. Barbara, K. Herbst, S. Rode, M. van Liempt, P.H. Roos, Single and
concerted effects of benzo[a]pyrene and flavonoids on the AhR and Nrf2-
pathway in the human colon carcinoma cell line Caco-2, Toxicol. Vitro 25
(2011) 671e683, http://dx.doi.org/10.1016/j.tiv.2011.01.008.
[27] S.O. Simmons, C.-Y. Fan, K. Yeoman, J. Wakefield, R. Ramabhadran, NRF2
oxidative stress induced by heavy metals is cell type dependent, Curr. Chem.
Genomics 5 (2011) 1e12, http://dx.doi.org/10.2174/1875397301105010001.
[28] E. Kansanen, S.M. Kuosmanen, H. Leinonen, A.-L. Levonen, The Keap1-Nrf2
pathway: mechanisms of activation and dysregulation in cancer, Redox Biol. 1
(2013) 45e49, http://dx.doi.org/10.1016/j.redox.2012.10.001.
[29] H.K. Bryan, A. Olayanju, C.E. Goldring, B.K. Park, The Nrf2 cell defence
pathway: Keap1-dependent and -independent mechanisms of regulation,
Biochem. Pharmacol. 85 (2013) 705e717, http://dx.doi.org/10.1016/
j.bcp.2012.11.016.
[30] A. Baulig, J.-J. Poirault, P. Ausset, R. Schins, T. Shi, D. Baralle, et al., Physico-
chemical characteristics and biological activities of seasonal atmospheric
particulate matter sampling in two locations of Paris, Environ. Sci. Technol. 38
(2004) 5985e5992.
[31] G. Kroemer, L. Galluzzi, P. Vandenabeele, J. Abrams, E.S. Alnemri,
E.H. Baehrecke, et al., Classification of cell death: recommendations of the
Nomenclature Committee on Cell Death 2009, Cell Death Differ. 16 (2009)
3e11, http://dx.doi.org/10.1038/cdd.2008.150.
[32] J.G. Charrier, C. Anastasio, On dithiothreitol (DTT) as a measure of oxidative
potential for ambient particles: evidence for the importance of soluble tran-
sition metals, Atmos. Chem. Phys. 12 (2012) 11317e11350, http://dx.doi.org/
10.5194/acpd-12-11317-2012.
[33] M.W. Pfaffl, G.W. Horgan, L. Dempfle, Relative expression software tool (REST)
for group-wise comparison and statistical analysis of relative expression re-
sults in real-time PCR, Nucleic Acids Res. 30 (2002) e36.
[34] A.M. Benson, M.J. Hunkeler, P. Talalay, Increase of NAD(P)H:quinone reductase
by dietary antioxidants: possible role in protection against carcinogenesis and
toxicity, Proc. Natl. Acad. Sci. U. S. A. 77 (1980) 5216e5220.
[35] J. Ding, X. Zhang, J. Li, L. Song, W. Ouyang, D. Zhang, et al., Nickel compounds
render anti-apoptotic effect to human bronchial epithelial Beas-2B cells by
induction of cyclooxygenase-2 through an IKKbeta/p65-dependent and
IKKalpha- and p50-independent pathway, J. Biol. Chem. 281 (2006)
39022e39032, http://dx.doi.org/10.1074/jbc.M604798200.
[36] M. Steenhof, I. Gosens, M. Strak, K.J. Godri, G. Hoek, F.R. Cassee, et al., In vitro
toxicity of particulate matter (PM) collected at different sites in the
Netherlands is associated with PM composition, size fraction and oxidative
potentialethe RAPTES project, Part. Fibre Toxicol. 8 (2011) 26, http://
dx.doi.org/10.1186/1743-8977-8-26.
[37] A. Baulig, M. Garlatti, V. Bonvallot, A. Marchand, R. Barouki, F. Marano, et al.,
Involvement of reactive oxygen species in the metabolic pathways triggered
by diesel exhaust particles in human airway epithelial cells, Am. J. Physiol.
Lung Cell. Mol. Physiol. 285 (2003) L671eL679, http://dx.doi.org/10.1152/
ajplung.00419.2002.
[38] I. Celardo, M. De Nicola, C. Mandoli, J.Z. Pedersen, E. Traversa, L. Ghibelli, Ce 3þ
ions determine redox-dependent anti-apoptotic effect of cerium oxide
nanoparticles, ACS Nano 5 (2011) 4537e4549, http://dx.doi.org/10.1021/
nn200126a.
[39] K. Andreau, M. Leroux, A. Bouharrour, Health and cellular impacts of air
pollutants: From cytoprotection to cytotoxicity, Biochem. Res. Int. 2012
(2012) 493894, http://dx.doi.org/10.1155/2012/493894.
[40] Q. Wang, W. Luo, W. Zhang, Z. Dai, Y. Chen, J. Chen, Iron supplementation
protects against lead-induced apoptosis through MAPK pathway in weanling
rat cortex, Neurotoxicology 28 (2007) 850e859, http://dx.doi.org/10.1016/
j.neuro.2007.04.004.
[41] E. Kontargiris, A. Vadalouka, V. Ragos, V. Kalfakakou, Zinc inhibits apoptosis
and maintains NEP downregulation, induced by ropivacaine, in HaCaT cells,
Biol. Trace Elem. Res. 150 (2012) 460e466, http://dx.doi.org/10.1007/s12011-
012-9492-8.
[42] A. Baulig, S. Singh, A. Marchand, R. Schins, R. Barouki, M. Garlatti, et al., Role of
Paris PM(2.5) components in the pro-inflammatory response induced in
airway epithelial cells, Toxicology 261 (2009) 126e135, http://dx.doi.org/
10.1016/j.tox.2009.05.007.
[43] N. Li, C. Sioutas, A. Cho, D. Schmitz, C. Misra, J. Sempf, et al., Ultrafine par-
ticulate pollutants induce oxidative stress and mitochondrial damage, Envi-
ron. Health Perspect. 111 (2003) 455e460.
[44] S. Orrenius, P. Nicotera, B. Zhivotovsky, Cell death mechanisms and their
implications in toxicology, Toxicol. Sci. 119 (2011) 3e19, http://dx.doi.org/
10.1093/toxsci/kfq268.
[45] J. Haendeler, V. Tischler, J. Hoffmann, A.M. Zeiher, S. Dimmeler, Low doses of
reactive oxygen species protect endothelial cells from apoptosis by increasing
thioredoxin-1 expression, FEBS Lett. 577 (2004) 427e433, http://dx.doi.org/
10.1016/j.febslet.2004.10.041.
[46] P. Schroeder, R. Popp, B. Wiegand, J. Altschmied, J. Haendeler, Nuclear redox-
signaling is essential for apoptosis inhibition in endothelial cellseimportant
role for nuclear thioredoxin-1, Arterioscler. Thromb. Vasc. Biol. 27 (2007)
2325e2331, http://dx.doi.org/10.1161/ATVBAHA.107.149419.
[47] A. Kode, S. Rajendrasozhan, S. Caito, S.-R. Yang, I.L. Megson, I. Rahman,
Resveratrol induces glutathione synthesis by activation of Nrf2 and protects
against cigarette smoke-mediated oxidative stress in human lung epithelial
cells, Am. J. Physiol. Lung Cell. Mol. Physiol. 294 (2008) L478eL488, http://
dx.doi.org/10.1152/ajplung.00361.2007.
[48] H.-Y. Cho, S.R. Kleeberger, Nrf2 protects against airway disorders, Toxicol.
Appl. Pharmacol. 244 (2010) 43e56, http://dx.doi.org/10.1016/
j.taap.2009.07.024.
[49] J. Alam, J. Cai, A. Smith, Isolation and characterization of the mouse heme
oxygenase-1 gene. Distal 50 sequences are required for induction by heme or
heavy metals, J. Biol. Chem. 269 (1994) 1001e1009.
[50] P. Nioi, M. McMahon, K. Itoh, M. Yamamoto, J.D. Hayes, Identification of a
novel Nrf2-regulated antioxidant response element (ARE) in the mouse
NAD(P)H:quinone oxidoreductase 1 gene: reassessment of the ARE consensus
sequence, Biochem. J. 374 (2003) 337e348, http://dx.doi.org/10.1042/
BJ20030754.
[51] S. Diabate !, B. Bergfeldt, D. Plaumann, C. Ubel, C. Weiss, Anti-oxidative and
inflammatory responses induced by fly ash particles and carbon black in lung
epithelial cells, Anal. Bioanal. Chem. 401 (2011) 3197e3212, http://dx.doi.org/
10.1007/s00216-011-5102-4.
[52] X. Deng, W. Rui, F. Zhang, W. Ding, PM2.5 induces Nrf2-mediated defense
mechanisms against oxidative stress by activating PIK3/AKT signaling
pathway in human lung alveolar epithelial A549 cells, Cell Biol. Toxicol. 29
(2013) 143e157, http://dx.doi.org/10.1007/s10565-013-9242-5.
[53] S. Michael, M. Montag, W. Dott, Pro-inflammatory effects and oxidative stress
in lung macrophages and epithelial cells induced by ambient particulate
matter, Environ. Pollut. 183 (2013) 19e29, http://dx.doi.org/10.1016/
j.envpol.2013.01.026.
[54] C.S.F. Queiroga, A.S. Almeida, P.M. Alves, C. Brenner, H.L.A. Vieira, Carbon
monoxide prevents hepatic mitochondrial membrane permeabilization, BMC
Cell Biol. 12 (2011) 10, http://dx.doi.org/10.1186/1471-2121-12-10.
[55] D. Siegel, D.L. Gustafson, D.L. Dehn, J.Y. Han, P. Boonchoong, L.J. Berliner, et al.,
NAD(P)H:quinone oxidoreductase 1: role as a superoxide scavenger, Mol.
Pharmacol. 65 (2004) 1238e1247, http://dx.doi.org/10.1124/mol.65.5.1238.
[56] H. Shen, C. Anastasio, A comparison of hydroxyl radical and hydrogen
peroxide generation in ambient particle extracts and laboratory metal solu-
tions, Atmos. Environ. 46 (2012) (1994) 665e668, http://dx.doi.org/10.1016/
j.atmosenv.2011.10.006.
[57] M. Marí, A. Morales, A. Colell, C. García-Ruiz, N. Kaplowitz, J.C. Fern! andez-
Checa, Mitochondrial glutathione: features, regulation and role in disease,
Biochim. Biophys. Acta 1830 (2013) 3317e3328, http://dx.doi.org/10.1016/
j.bbagen.2012.10.018.
[58] A. Hartwig, Metal interaction with redox regulation: an integrating concept in
metal carcinogenesis? Free Radic. Biol. Med. 55 (2013) 63e72, http://
dx.doi.org/10.1016/j.freeradbiomed.2012.11.009.
[59] D. Pr!a, S.I.R. Franke, J.A.P. Henriques, M. Fenech, Iron and genome stability: an
update, Mutat. Res. Mol. Mech. Mut. 733 (2012) 92e99, http://dx.doi.org/
10.1016/j.mrfmmm.2012.02.001.
[60] P.M. Nguyen, M.S. Park, M. Chow, J.H. Chang, L. Wrischnik, W.K. Chan, Benzo
[a]pyrene increases the Nrf2 content by downregulating the Keap1 message,
Toxicol. Sci. 116 (2010) 549e561, http://dx.doi.org/10.1093/toxsci/kfq150.
[61] W. Miao, L. Hu, P.J. Scrivens, G. Batist, Transcriptional regulation of NF-E2 p45-
related factor (NRF2) expression by the aryl Hydrocarbon receptor-xenobiotic
response element signaling pathway: direct cross-talk between phase I and II
drug-metabolizing enzymes, J. Biol. Chem. 280 (2005) 20340e20348, http://
dx.doi.org/10.1074/jbc.M412081200.
[62] L. Wang, X. He, G.D. Szklarz, Y. Bi, Y. Rojanasakul, Q. Ma, The aryl hydrocarbon
receptor interacts with nuclear factor erythroid 2-related factor 2 to mediate
induction of NAD(P)H:quinoneoxidoreductase 1 by 2,3,7,8-tetra-
chlorodibenzo-p-dioxin, Arch. Biochem. Biophys. 537 (2013) 31e38, http://
dx.doi.org/10.1016/j.abb.2013.06.001.