Nanoscale

View Article Online

PAPER View Journal

Metal homeostasis disruption and mitochondrial

Published on 20 October 2016. Downloaded by Cornell University Library on 27/10/2016 06:54:32.

Cite this: DOI: 10.1039/c6nr05306h

dysfunction in hepatocytes exposed to sub-toxic
doses of zinc oxide nanoparticles
M. Chevallet,*a,b,c B. Gallet,d,e,f A. Fuchs,g P. H. Jouneau,h,i K. Um,a,b,c E. Mintza,b,c
and I. Michaud-Soret*a,b,c

Received 4th July 2016,
Accepted 19th October 2016
DOI: 10.1039/c6nr05306h
www.rsc.org/nanoscale
Increased production and use of zinc oxide nanoparticles (ZnO-NPs) in consumer products has prompted
the scientic community to investigate their potential toxicity, and understand their impact on the
environment and organisms. Molecular mechanisms involved in ZnO-NP toxicity are still under debate
and focus essentially on high dose expositions. In our study, we chose to evaluate the eect of sub-toxic
doses of ZnO-NPs on human hepatocytes (HepG2) with a focus on metal homeostasis and redox balance
disruptions. We showed massive dissolution of ZnO-NPs outside the cell, transport and accumulation of
zinc ions inside the cell but no evidence of nanoparticle entry, even when analysed by high resolution
TEM microscopy coupled with EDX. Gene expression analysis highlighted zinc homeostasis disruptions as
shown by metallothionein 1X and zinc transporter 1 and 2 (ZnT1, ZnT2) over-expression. Major oxidative
stress response genes, such as superoxide dismutase 1, 2 and catalase were not induced. Phase
2 enzymes in term of antioxidant response, such as heme oxygenase 1 (HMOX1) and the regulating
subunit of the glutamate-cysteine ligase (GCLM) were slightly upregulated, but these observations may be
linked solely to metal homeostasis disruptions, as these actors are involved in both metal and ROS
responses. Finally, we observed abnormal mitochondria morphologies and autophagy vesicles in response
to ZnO-NPs, indicating a potential role of mitochondria in storing and protecting cells from zinc excess
but ultimately causing cell death at higher doses.

1 Introduction and proliferation, dierentiation or programmed cell death.2

Maintenance of zinc homeostasis is essential and deficiency
Zinc is an essential trace element. Not only does it act as a cel- as well as excess zinc is toxic for organisms and cells. For
lular ionic signal, but in thousands of proteins, zinc partici- example, deregulation of free zinc has been implicated in the
pates in enzymatic catalysis, structural organization, and/or formation of -amyloid plaques associated with Alzheimers
regulation of function. Zn2+ is a non-redox metal ion but it is disease.3,4
involved in the redox equilibrium of the cell by binding to To complicate things further, zinc at physiological concen-
thiolate containing sites with a very high anity.1 In both its trations has also been shown to be cytoprotective as a
ionic and protein-bound forms zinc is at the crossroads of all pro-antioxidant.4
crucial decisions in the life of mammalian cells: cell growth The large number of proteins that are potentially dedicated
to Zn2+ transport (ZnT and ZIP families) and buering (metal-
lothioneins) reflect the complexity and importance of zinc
a
CNRS, Laboratoire de Chimie et Biologie des Mtaux (LCBM), UMR 5249, homeostasis.5,6 The ZnT transporters are dedicated to Zn2+
Grenoble, France. E-mail: isabelle.michaud-soret@cea.fr eux from the cytoplasm to the extracellular medium or to
b
CEA, BIG, LCBM, Grenoble, France. E-mail: mchevallet@cea.fr
c organelles. Zip transporters control Zn2+ import from the extra-
Universit Grenoble Alpes, LCBM, Grenoble, France
d
Universit Grenoble Alpes, IBS, Grenoble, France cellular medium or organelle lumens into the cytoplasm.7
e
CNRS, IBS, Grenoble, France Although the up-regulation of Zn2+ chelation and transport
f
CEA, IBS, Grenoble, France machineries appears to be directly activated by Zn2+ binding to
g
CEA, BIG, DIR, Grenoble, France the transcription factor MTF-1,8 there is still much to decipher
h
CEA, INAC, Minatec campus, Grenoble, France
i in the mechanisms governing zinc homeostasis.
Universit Grenoble Alpes, INAC-MEM-LEMMA, Grenoble, France
Electronic supplementary information (ESI) available. See DOI: 10.1039/ Due to their specific physical properties linked to surface
c6nr05306h and quantum size eects,9 nanoparticles (NPs) are finding

This journal is The Royal Society of Chemistry 2016 Nanoscale


Paper
their way into a wide range of consumer products, including
food packaging. This increased exposure of all living beings
and their environment to NPs is driving concerns about their
potential adverse impact on the environment and our well-
being. Over the last decade, considerable scientific eort has
been invested in research to better understand the risks of
exposure to NPs.10,11
ZnO nanoparticles (ZnO-NPs) are among the top 5 nano-
particles produced in high tonnage (the estimated global
Published on 20 October 2016. Downloaded by Cornell University Library on 27/10/2016 06:54:32.
annual production in 2010 is over 30 000 tons)12 and are used
in commercial applications such as antibacterial coatings or as
UV absorbers in sunscreens and textiles.13,14
The literature on mammalian toxicity of ZnO-NPs published
between 2009 and 2011 was recently summarized in a review.15
In vivo studies explored the main exposure means and showed
a potential risk for mammals. Through airway exposure in
rats, ZnO-NPs elicited an increased Zn content in multiple
organs and in particular in the liver where it was sucient to
cause damage as shown by histopathological examination.16 In
skin applications, while several reviews have concluded that
metal oxide NPs do not penetrate the stratum corneum,17
Adamcakova-Dodd et al.18 demonstrated that small amounts
of Zn from ZnO-NPs in sunscreens can pass through the skins
protective layers and be detected in blood and urine. Oral
administration of ZnO-NPs also resulted in zinc accumulation
in the liver and other organs.19
In a majority of published studies, incomplete information
(or characterization) of the NPs themselves and in some cases
the lack of information or studies on ion release and NP
uptake by cells make it dicult to compare results.
In vitro studies on dierent cell lines showed consensual
results: ZnO-NPs are in all cases cytotoxic via the release of
Zn2+ ions, reactive oxygen species (ROS) production, and mito-
chondria dysfunction. Some studies also revealed a genotoxic
eect of ZnO-NPs.20 Others have demonstrated that autophagy
is activated21,22 as well as inflammation via IL8 production.23
In hepatocytes, an alteration of albumin production was high-
lighted23 and an important disruption in energy metabolism
with an increase in both gluconeogenesis and glycogenolysis
was described.24
Moreover, most studies focused solely on NP toxicity at high
toxic doses using classical tests to measure cytotoxicity, ROS
production, and DNA damages without addressing the specific
intracellular mechanisms and pathways related to metal
homeostasis. This fact prompted us to investigate the eects
of ZnO-NP exposure on metal homeostasis regulation, taking
into account its particulate or ionic nature in situ in human
liver-derived cells (HepG2), as the liver is the primary target for
concentrating and metabolizing toxic agents. To take into
account toxic eects that could be linked to the surface charge
anionic or cationic of the nanoparticles, we chose to work
with 2 types of ZnO nanoparticles, namely foetal bovine serum
(FBS)-coated ZnO-NPs with a negative surface charge and
silane-coated ZnO-NPs with a positive surface charge. We com-
pared the eects of these two types of characterized ZnO-NPs
with the eects of the soluble form of zinc, using zinc acetate,
Nanoscale
View Article Online
Nanoscale
and focused on the early response of metal homeostatic
control and oxidative stress genes to sub-toxic doses of zinc
(i.e. conditions that do not induce cell mortality).
Our results clearly demonstrate no specific nano eect of
the ZnO-NPs in contrast to what we previously observed with
CuO-NPs.25 We demonstrate however that sub-toxic doses of
both ionic and nanoparticulate forms of zinc induce a metal
stress response with zinc homeostasis disruption and
mitochondria alterations, even after short exposure times.
2 Experimental
2.1 Nanoparticles
ZnO-NPs with a primary particle size of less than 100 nm were
purchased from Sigma-Aldrich: an uncoated ZnO-NP powder
(ref number 721077) and an aminopropyl-silane-coated
ZnO-NP dispersion (ref number 544906). The uncoated
ZnO-NPs were dispersed at 50 mg mL1 in 50% FBS by soni-
cation for 30 min by 1s on1s o pulses at 60% amplitude
(750 W) in a cup-horn instrument (Vibra cell, BioBlock
Scientific, France) under water cooling thermostated at 4 C to
get stable and reproducible dispersion of NPs and to avoid
rapid aggregation (denoted ZnO-NP-FBS). The ZnO-NP-silane
were diluted directly in water at 50 mg mL1, an additional
FBS coating was not necessary as these NPs were already well
dispersed and stable before diluting them into FBS-containing
cell culture media. The actual size of the particles was deter-
mined after dilution in water or in complete culture medium
by dynamic light scattering, using a Wyatt Dynapro Nanostar
instrument. A Malvern instrument Zetasizer plus was used to
determine the zeta potential. The size and shape of NPs were
verified by STEM on a Hitachi S5500.
2.2 Cell culture
HepG2 cells were obtained from the American Type Culture
collection (ATCC) (Manassas, Virginia, USA). They were grown
in modified Eagles medium (MEM) supplemented with 10%
v/v FBS, 20 mM L-glutamine, 10 mM sodium pyruvate, 100 g
mL1 streptomycin and 100 U mL1 penicillin. Cells were
cultured at 37 C in a humidified atmosphere with 5% CO2.
2.3 Cell viability
Cells were seeded at 300 000 cells per mL in 12-well plates.
They were treated with zinc acetate or ZnO-NPs on the follow-
ing day and harvested after a further 24 h. Cell viability was
evaluated by counting Trypan Blue stained cells in a TC20
Automated Cell Counter (BioRad). In order to evaluate the pro-
tecting eect of EDTA or Ca2+, HepG2 cells were pretreated for
15 min with 1 mM EDTA or 2 mM CaCl2 before directly adding
Zn acetate, ZnO-NP-FBS or ZnO-NP-silane at the dose of
250 M for 24 h. For assays including bafilomycin A (BafA),
HepG2 cells were pretreated 1 hour with 100 nM BafA before
directly adding Zn acetate, ZnO-NP-FBS or ZnO-NP-silane at
90 or 150 M for 24 h.
This journal is The Royal Society of Chemistry 2016

Nanoscale
2.4 ICP-AES
Analysis of zinc was performed by inductively coupled plasma
atomic emission spectroscopy (ICP-AES: ICPE-9000 from
Shimadzu Scientific Instruments).
HepG2 cells were treated with Zn acetate or ZnO-NPs for
6 h, 24 h and 48 h or kept in complete culture medium as a
control. Briefly, after removing the medium and washing with
PBS buer, cells were collected using trypsin, centrifuged at
Published on 20 October 2016. Downloaded by Cornell University Library on 27/10/2016 06:54:32.
1200 rpm for 5 min and suspended in PBS buer. Cells were
then counted using a TC20 Automated Cell Counter (BioRad)
and centrifuged again. The pellets were suspended with
100 L of pure nitric acid and mineralized overnight at 95 C
in a DigiPrep (SCP Science). The internal standard Ytterbium
was added and the samples were diluted in pure water qs
6.5 mL prior to analysis. A standard curve was performed
using an atomic absorption standard solution of zinc (Sigma-
Aldrich).
To follow zinc dissolution, 90 M zinc acetate and ZnO-NPs
were incubated in complete culture medium in cell culture
plates. The plates were kept for 24 h in a cell culture incubator
at 37 C and 5% CO2.
The media were retrieved at dierent times, filtered at
0.1 m (PVDF, Merck Millipore LTD) and centrifuged at
150 000g for 45 min to sediment the ZnO-NPs. The concen-
tration of zinc ions in the supernatant was then measured by
ICP-AES.
2.5 Real-time PCR assays & measurements
HepG2 cells were harvested and RNA was isolated using
Absolutely-RNA miniprep kits (Agilent #400800). RNA concen-
tration was determined using a NanoDrop spectro photometer
(ND-1000). Reverse transcription was performed with the
Anity script qPCR cDNA synthesis kit (Agilent # 600559),
according to the manufacturers instructions. Gene specific
primers for MET1X were taken from.26 The other human
primers were designed using DNASTAR, Primer-Blast or Pick
Primers. The designed primers are given in Table S1.
Quantitative PCR was performed with Brilliant II SYBR
green qPCR master mix1 (Agilent #600828) using the primers
at 200 nM. PCR conditions ( primer concentrations, cDNA
quantity) were optimized and PCR eciency was determined
for each target gene. PCR reaction mixtures (10 L) were
placed in the Cfx96 instrument (Bio-Rad) where they under-
went the following cycling program, optimized for a 96-well
block: 95 C for 15 min, immediately followed by 40 cycles of
10 s at 95 C and 30 s at 60 C. At the end, PCR products were
dissociated by incubating for 1 min at 95 C and then 30 s at
55 C, followed by a ramp up to 95 C. PCR quality and speci-
ficity were verified by analysing the dissociation curve. For
each set of primers, a no-template control (NTC) and a no-
reverse-amplification control (NAC) were included. qRT-PCR
reactions were run in triplicate, and quantification was per-
formed using comparative regression (Cq determination
mode). Quantitative PCR data were comparatively analysed
using the Cfx software (Bio-Rad Cfx manager) with 36B4 and
This journal is The Royal Society of Chemistry 2016
View Article Online
Paper
HPRT amplification signals as internal normalizers
(to correct for total RNA content). Results are expressed as the
relative change in expression compared to the control.
Results were the average (SEM) of at least 3 independents
experiments.
2.6 Measuring SOD and catalase activities
SOD and catalase activities were measured as previously
described27,28 and results were expressed as the average of
3 independent experiments in enzymatic units per mg of total
protein quantified by a Bradford assay.29
2.7 Electron microscopy
Monolayers of HepG2 cells were fixed overnight at room temp-
erature in a 1 : 1 ratio mixture of 4% paraformaldehyde, 0.4%
glutaraldehyde in 0.2 M PHEM (60 mM PIPES, 25 mM HEPES,
10 mM EGTA, 2 mM MgCl2) pH 7.2 and culture medium,
washed in 0.1 M PHEM pH 7.2, and fixed for 30 minutes in
2% paraformaldehyde, 0.2% glutaraldehyde in 0.1 M PHEM
pH 7.2, washed in 0.1 M PHEM pH 7.2 and post-fixed in 1%
OsO4, 1.5% potassium ferrocyanide in 0.1 M PHEM buer for
1 h at room temperature. After 3 washes in water, post-staining
was done using 0.5% uranyl acetate in 30% ethanol for 30 min
at room temperature in the dark. Cells were then dehydrated
in graded ethanol series, and flat-embedded using the Epoxy
Embedding Medium kit (Sigma-Aldrich). Ultrathin sections
(240 nm) were cut on a Leica UC7 ultra-microtome using a
DiATOME 35 diamond knife and collected on formvar carbon
coated 100-mesh copper grids. Sections were stained in 5%
uranyl acetate in water for 10 min and in 2% lead citrate for
5 min. Images were taken on a Tecnai G2 Spirit BioTwin (FEI)
at 120 kV using an ORIUS SC1000 CCD camera (Gatan).
2.8 Study of acidic vesicular organelles
To detect and quantify acidic vesicular organelles, cells were
stained with acridine orange (1 g mL1) for 15 min as
described previously.30 Green (510530 nm) fluorescence emis-
sions from 30 000 cells illuminated with a blue excitation light
(488 nm) were measured with a MoFlo instrument (Beckman
Coulter).
2.9 Mitochondrial transmembrane potential assessments
Mitochondrial potential was measured using a Rhodamine
123 uptake assay.31 Cells were seeded in 6-well plates for 24 h
and then treated with zinc for 24 h before adding Rhodamine
123 at a final concentration of 10 g mL1 for 15 min at 37 C.
Cells were then harvested and rinsed with PBS and analysed by
flow cytometry on a MoFlo instrument (Beckman Coulter).
Live cells were first selected on the basis of the size and granu-
larity, and their fluorescence was then measured using exci-
tation at 488 nm and emission at 525 nm. Cells pre-treated
with sodium azide (NaN3) at 40 mM for 30 min were used as a
negative control.
Nanoscale

Paper
3 Results and discussion
3.1 Nanoparticle characterization
Dynamic light scattering was used to characterize the nano-
particle populations by their hydrodynamic diameter distri-
bution (as % of intensity). ZnO-NP-FBS had an average dia-
meter of 237 nm (Fig. S1) and a polydispersity index of 20%,
ZnO-NP-silane presented an average diameter of 79 nm and a
polydispersity index of 19%. The zeta potentials were
Published on 20 October 2016. Downloaded by Cornell University Library on 27/10/2016 06:54:32.
measured in 1 mM KCl 32,33 and were of 25 mV for the
ZnO-NP-FBS and 19.3 mV for the ZnO-NP-silane. The albumin
coating accounts for the negative surface charge at physio-
logical pH and the silane cationic coating accounts for the
positive surface charge of these NPs. STEM images of coated
NPs diluted in water confirmed the size of the particles
(Fig. 1), the ZnO-NP-FBS were rod-shaped while the ZnO-NP-
silane were more spherical.
3.2 HepG2 viability studies
ZnO-NP toxicity was first studied by measuring the viability of
HepG2 cells after 24 h incubation with either zinc acetate or
ZnO-NP-FBS or ZnO-NP-silane in a range of concentrations
from 0 to 300 M equivalent zinc (Fig. 2). The toxicity appeared
to be similar among the dierent forms of zinc, with a slightly
Fig. 1 Characterization of ZnO-NPs. STEM images of ZnO-NP-FBS
(A) and ZnO-NP-silane (B) taken on a Hitachi S5500.
Fig. 2 Cell viability after a 24 hour incubation with zinc acetate (),
ZnO-NP-FBS (), or ZnO-NP-silane (). The arrow shows the concen-
tration (90 M or 7.3 g mL1) chosen for the gene expression studies.
For sake of comparison, the ZnO-NP concentrations are expressed as
zinc equivalents.
Nanoscale
View Article Online
Nanoscale
higher toxicity of ZnO-NP-silane which was not statistically
significant.
The shape of the viability curves showed that HepG2 cells
could tolerate 150 M of zinc for 24 h without loss of viability
but that beyond this dose, viability decreased drastically.
These dose tolerances are in agreement with the literature
for HepG2 cells34 and other cell types such as macrophages.35
In order to decipher the response of the cells, the
expression of genes involved in metal homeostasis and oxi-
dative stress was followed.
3.3 Early responses of metal homeostatic control and
oxidative stress genes
For this gene expression study we chose a sub-toxic dose of
90 M oering an optimum compromise between viability and
biological eect.
Indeed, at higher but still sub-toxic concentrations
(150 M), gene expression of usual internal normalizers
showed too much variability, a phenomenon which is
explained by the proliferating eect of zinc at this dose.36 In
order to capture early responses triggered by zinc, we also
reduced the exposition to 6 hours, a condition where no cell
death was yet measurable, even at concentrations as high as
300 M (data not shown).
The responses of HepG2 cells to a 6 h incubation with
90 M zinc, brought either by zinc acetate, ZnO-NP-FBS or
ZnO-NP-silane were studied by quantitative RT-PCR.
For oxidative stress, we chose the classical genes CAT (cata-
lase), SOD1 (the cytoplasmic Cu, Zn superoxide dismutase),
SOD2 (the mitochondrial Mn superoxide dismutase) and
GCLM, the regulating subunit of the glutamatecysteine ligase
which is the first rate-limiting enzyme of glutathione (GSH)
synthesis. We added HMOX1, the inducible isoform of heme
oxygenase which can act as an antioxidative protein and
was already found to be induced by ZnO-NP37,38 and by
CuO-NP.25,39
For zinc homeostasis40 we chose MTF1, Met1X, Znt1 and 7
and Zip1. MTF1 is a cellular zinc sensor, which in the presence
of excess zinc, migrates to the nucleus and activates genes
involved in zinc homeostasis via binding to metal-response
elements. Metallothionein 1X (Met1X) was chosen to represent
the metallothioneins, cysteine-rich proteins which function as
cytoplasmic soft-metal chelators, bind zinc with high
anity,41 and are important in defending the cell against
metal poisoning.42 The transporters ZnT1, ZnT7 and Zip1 are
respectively a plasma membrane zinc exporter, a zinc importer
for storage in the Golgi apparatus and the main plasma mem-
brane zinc importer. Metallothioneins, ZnT1 and ZnT7 gene
expressions are known to be activated via MTF1.43
For iron homeostasis we looked at HAMP, the gene encod-
ing hepcidin, a cysteine-rich protein which controls iron
homeostasis44 and is regulated via MTF1.8,45 Finally, we added
the chaperone HSPA6, from the HSP 70 family which is acti-
vated by several soft metals such as Cd2+ and Zn2+ and
responds to CuO-NPs in HepG2 cells,25 and EGR1, an early
This journal is The Royal Society of Chemistry 2016

Nanoscale
response transcription factor induced by several extracellular
signals, including environmental stress.46
For all treatments, the highest induction was observed for
Met1X, which was up-regulated around 50-fold (Fig. 3). We
also found a 4-fold increase in the expression of ZnT1 but no
change in ZnT7 nor in Zip1. These results are in agreement
with previous work of Cousins et al.47 in THP1 cells treated
with 40 M ZnSO4 showing a 3-fold increase in ZnT1
expression and no significant variation for ZnT7 and Zip1. We
Published on 20 October 2016. Downloaded by Cornell University Library on 27/10/2016 06:54:32.
found no evidence in the literature of regulation of Zip1 via
MTF1. In HEK293 cells, it was shown that Zip1 is not regulated
at the transcriptional level but rather post-translationally by
altering the proteins subcellular distribution.48 Finally, there
was no increase in the expression of MTF1 showing that zinc
excess induces the translocation of MTF1, and does not
require de novo synthesis under our conditions. Although zinc
homeostasis is linked to other metal homeostasis, such as that
of copper and iron, hepcidin also showed no variation in
expression in our conditions. The chaperone HSPA6 was
slightly induced, in particular with ZnO-NP-silane.
On the oxidative stress side, we found no modification of
the major enzymes dedicated to detoxification of ROS. CAT,
SOD1 and SOD2 are expressed constitutively and we can postu-
late that cells are able to cope with a moderate oxidative
stress without requiring additional induction of the cor-
responding genes. However we found 3.54.5 fold inductions
for HMOX1 and GCLM which are classified as phase 2
enzymes in term of antioxidant response, and are regulated by
the transcription factor Nrf2 via ARE, the antioxidant response
element.49 Interestingly, GCLM was also described to be regu-
lated by MTF1.50 In light of these results, it remains unclear if
90 M zinc induces an oxidative stress or solely a metal
homeostasis disruption. As such, metal and oxidative stress
responses are tightly imbricated. On the one hand, GSH which
is present in millimolar concentrations in cells, binds large
amounts of zinc in vitro and on the other hand metallo-
thioneins can function as antioxidants.1,5 HMOX1 cleaves the
Fig. 3 Quantitative PCR analysis of mRNA expression in HepG2 cells after a 6 hour incubation with Zn acetate, ZnO-NP-FBS or ZnO-NP-silane at
90 M (note the 10-fold change in scale of the right panel). Results are presented as relative expression changes after normalizing with HPRT and
36B4 mRNA. Each value represents the mean of relative expression +SEM from three independent experiments. *: p < 0.05, **: p < 0.01, ***: p <
0.001 vs. control.
This journal is The Royal Society of Chemistry 2016
View Article Online
Paper
heme ring to form biliverdin and then bilirubin, known to
chelate metals.51 Moreover, although zinc in biology is redox
inert, in contrast to copper or iron, zinc excess can indirectly
influence the redox balance.1
A minor but significant induction of EGR1 is in agreement
with the work of Jeong et al.52 on keratinocytes and probably
belongs to a global stress response signature.
These results clearly showed an important response to zinc
stress after either zinc acetate or either ZnO-NP treatments,
possibly linked with an oxidative stress. ZnO-NP-silane elicited
slightly higher cytotoxicity at equivalent zinc concentrations
above 150 M and activated HSPA6 to a slightly higher extent
than ZnO-NP-FBS or zinc acetate. However, no other dier-
ences were identified among these 3 forms of zinc. Similar
results were found on macrophages with a slightly higher cyto-
toxicity of ZnO-NP-silane versus ZnO-NP-FBS.35 This dierence
could be explained by the higher dissolution rate of ZnO-NP-
silane, as well as by the cationic coating of these nanoparticles
which can influence their interactions with the cell membrane,
Zn2+ ion release and ion entry. Studies with polystyrene nano-
particles have also shown that amino cationic coatings are
directly responsible for toxic eects compared to anionic coat-
ings, as NPs carrying amino groups have been shown to target
cell membranes through strong binding to phospholipid com-
ponents.53,54 In any case, the similarity of response between
zinc ions and NPs strongly suggests that NP toxicity is essen-
tially due to ion release.
3.4 Investigation of the oxidative stress response
As changes observed at the transcription level are not always
correlated with protein activities, we measured the activity of
CAT and SOD enzymes after a 6 h incubation with zinc at
90 M (corresponding to 7.3 g mL1) and found no signifi-
cant eect whatever the form of zinc (Fig. S2). We even
noticed a tendency towards decreased (P = 0.067) catalase
activities after zinc treatments that was the opposite result to
what could have been expected. These results confirm the very
Nanoscale

Paper
moderate oxidative stress response generated by sub-toxic
doses of zinc seen at the transcription level. These findings are
also in agreement with the literature. Indeed, Triboulet et al.39
showed, by a proteomic approach, a rather weak activation of
the oxidative stress response pathway in macrophages treated
with ZnO-NP at 8 g mL1, the concentration corresponding to
20% mortality (LD20). Very interestingly, they found no changes
in CAT and SOD protein levels. Guan et al.55 showed that SOD
activity was significantly reduced (p < 0.05) at high concen-
Published on 20 October 2016. Downloaded by Cornell University Library on 27/10/2016 06:54:32.
trations (above 50 g mL1) of ZnO-NPs on LO2, another liver
cell line. Furthermore, Sharma et al.34 estimated the level of
intracellular ROS (by the method of Wan et al.) on HepG2 cells
exposed to ZnO-NPs for 6 h and found no measurable changes
at 8 g mL1 (close to our concentration) but a significant
increase (p < 0.05) in ROS generation at 14 and 20 g mL1.
3.5 Nanoparticle fate in the culture medium
As zinc ions and ZnO-NPs seemed to induce similar eects on
cells, we investigated the fate of NPs in biological media
(Fig. 4), combining filtration on 0.1 m membranes and ultra-
centrifugation to ensure the total removal of NPs. The ICP-AES
dosage of supernatants clearly showed a biphasic and major
dissolution of NPs in complete MEM reaching 7090% of dis-
solution after 6 h. This result is in agreement with Herrmann
et al.s work,56 which shows a rapid dissolution of ZnO-NPs in
a phosphate buer. Size is considered as the primary physico-
chemical property aecting the solubility of NPs and theoreti-
cally, solubility increases as the particle size decreases.57 This
property, among many other parameters, could contribute to
the higher initial level of ZnO-NP-silane dissolution (60 vs.
40% for ZnO-NP-FBS), which is then followed by a phase of
slower dissolution with similar rate constants for both NPs.
The dierences in the plateaus need to be interpreted with
caution as the same experiment done with zinc acetate did not
reach 100%, suggesting some loss of zinc on the plastic tube
walls and on the filter. These measurements corroborate the
extent of zinc ion release in the medium. Special care needs to
be taken to interpret DLS studies of NPs diluted in complex
culture medium as phenomena of aggregation and dissolution
may bias the correlation analysis. Critical analysis of our
results showed that both dissolution and aggregation or
precipitation occurred in MEM.
Fig. 4 Dissolution (%) of ZnO-NP-FBS () and ZnO-NP-silane () at
90 M in complete MEM culture medium.
Nanoscale
View Article Online
Nanoscale
3.6 Protection by EDTA or Ca2+ from Zn toxicity
To evaluate the role of solubilised Zn2+ ion in ZnO-NP
mediated cell death, we tested the eect of an ion chelator. We
found that pre-treatment of HepG2 cells with 1 mM EDTA,
known to chelate divalent metal ions, completely preserved
cell viability (Fig. 5) at the toxic dose of 250 M ZnO-NPs in
agreement with previous work.58 This result, in the light of our
previous findings, is in favour of a toxicity essentially mediated
by rapid ion release in the extracellular medium. Indeed, here
again, similar results were found using zinc acetate. More sur-
prisingly, we also found that pre-treatment with 1.25 mM
CaCl2 showed similar protective eects.
A possible explanation for this eect is brought by
Johnson et al.59 who showed that dissolved Ca2+ prevents cells
from taking up toxic amounts of zinc, suggesting that zinc
entry could be mediated by a calcium transporter. This would
be a non-specific mechanism having no relation to Zip1
regulation.
3.7 Cellular zinc content and subcellular localization
Total zinc content in cells was analysed after various incu-
bation times with zinc acetate or either NPs (Fig. 6). Zinc
increased steadily in cells up to 24 h and then remained rela-
tively stable until 48 h, even though zinc concentrations in the
medium remained unchanged and the amount of zinc that
entered the cells was negligible compared to the amount in
the medium. Protective mechanisms appeared to be activated
to limit zinc level increases as already suggested by our gene
expression results. After 24 h with zinc followed by 24 h
without zinc, we measured a dilution of zinc in cells consistent
with cell proliferation and distribution to daughter cells. There
was no evidence of active zinc export back into the medium.
Here again, we saw similar results with either NPs or zinc
acetate, an observation which corroborates a toxicity solely
linked to metal ion release.
Fig. 5 Protection by EDTA or Ca2+ against Zn toxicity: HepG2 cells
were pretreated for 15 min with 1 mM EDTA or 2 mM CaCl2 before
adding Zn acetate, ZnO-NP-FBS or ZnO-NP-silane at the toxic dose of
250 M for 24 h. EDTA or CaCl2 treatment alone had no eect on viabi-
lity of HepG2 cells.
This journal is The Royal Society of Chemistry 2016
 View Article Online

Nanoscale Paper
Published on 20 October 2016. Downloaded by Cornell University Library on 27/10/2016 06:54:32.

Fig. 6 Cellular zinc content measured by ICP-AES after incubation with

90 M zinc acetate, ZnO-NP-FBS or ZnO-NP-silane for 6 h, 24 h and
48 h or for 24 h followed by a 24 h recovery.

In the literature, three dierent scenarii have been

described to explain ZnO-NP toxicity. In some studies, toxicity
appears to be solely linked to ions released in the biological
media and to the deleterious eects of ion entry in
cells.57,58,60,61 Others support a Trojan horse scenario with NP
dissolution in the acidic lysosomal compartment occurring
after cell entry.35,62 Yet in others there appears to be a specific
NP eect.63,64 In the latter case, NP did not enter the cell,
however dissolution and release of soluble zinc could only par-
tially account for the toxicity that was also linked to a direct
Fig. 7 TEM observation of HepG2 sections, A, B: control; C, E, G:
contact of NPs with the cell walls.
ZnO-NP-FBS treated cells (90 M, 6 h); D, F, H: ZnO-NP-silane treated
Our work is in agreement with the first scenario which is cells (90 M, 6 h). M: mitochondria, N: nucleus, G: Golgi apparatus, RE:
described in the majority of studies. This controversy on the endoplasmic reticulum, AM: abnormal mitochondria, MLB: multilamellar
cytotoxic potency of the ionic versus the NP form of zinc is body, A: autophagosome.
probably linked to experimental variables such as the NP
characteristics: their chemical nature, size, coating, mode of
dispersion, presence of corona65 and concentration as well as
cell types15,66 and culture media (DMEM or MEM versus
zinc. In contrast to our observations pertaining to CuO-NP
RPMI).67 All these parameters could influence the aggregation
uptake at similar concentrations in HepG2 cells,25 we conclude
and the kinetics of dissolution of NPs and thereby, their
that very little, if any, zinc enters and subsists in cells in the
toxicity.
nanoparticulate form. In the literature, a rare number of
studies showed ZnO-NPs inside hepatocytes only at high toxic
3.8 ZnO-NPs fate in HepG2 cells and eects on organelle doses, compatible with a saturation plateau of NP dissolution.
structure The maximum total dissolved zinc concentration that could be
A debate surrounds the eective entry of the nanoparticulate reached was estimated to be 225 M in complete DMEM by
form of zinc in cells via endocytosis or phagocytosis. ZnO-NPs Xia et al.53 and around 10 g mL1 (125 M) in complete
have been shown to be internalized by macrophages, cham- RPMI.70 Sharma et al.34 assessed cellular uptake of ZnO-NPs in
pions of rapid phagocytosis.35,68,69 Mechanisms of entry are HepG2 cells by an indirect measure using flow cytometry (light
however less clear for other cell types. In the present study, we scattering) no entry was found at 98 M but entry was detected
investigated the intracellular localization of ZnO-NPs using at 170 and 245 M (12 h). Using TEM, Wang et al.71 observed
transmission electronic microscopy (TEM) on 240 nm sections possible ZnO-NPs both inside and outside the cells, as well as
of embedded cells. Hepatocytes are energetically active cells, cell membrane damages, at the concentration of 2.45 mM
presenting an unusually high number of mitochondria and after 48 h, but both these studies did not confirm their nature
ribosomes as well as many dark lipid rich vesicles revealed by unequivocally by a chemical analysis such as EDX. In con-
osmium staining (Fig. 7). After 6 or 24 h of ZnO-NP treatment, clusion, it appears that only a few publications show convin-
we were unable to find any nanoparticles inside the cells. cing proof of ZnO-NP entry by TEM coupled to EDX35 or
Dense elements were analysed by EDX but were all negative for synchrotron61,72 observations and then, only in macrophages.

This journal is The Royal Society of Chemistry 2016 Nanoscale


Paper
Macrophages are cultured in RPMI and as ZnO-NPs seem to
dissolve less and re-precipitate more under phosphate and
carbonate forms66 in this medium, this coupled with the pha-
gocytic nature of macrophages could explain this particularity.
Further TEM analysis revealed significant cell ultrastructure
alterations induced by sub-toxic doses of zinc. While gene
expression analysis can only give an averaged result over the
cell population, TEM analysis oers information on individual
cell responses. The first striking observation was that a subset
Published on 20 October 2016. Downloaded by Cornell University Library on 27/10/2016 06:54:32.
of cells showed no morphological alterations while others pre-
sented a significant number of mitochondria with abnormal
morphologies and branching and in some cases doughnut
shapes (Fig. 7C and D), suggesting disruptions in the fission
and fusion dynamics of the mitochondrial network at these
sub-toxic doses. These cells also presented an increased
number of autophagosomes (Fig. 7G and H) and the appear-
ance of multilamellar bodies (MLBs) (Fig. 7E and F). Increased
autophagy in response to zinc was confirmed by observation of
more acidic vesicular organelles by flow cytometry analysis
after treatment by acridine orange (Fig. S3). Autophagy is
stimulated in response to a variety of environmental stresses to
enable cellular survival;73 it plays a housekeeping role in
removing misfolded or aggregated proteins, and clearing
damaged organelles. ZnO-NPs have also been described to
enhance autophagy in macrophages and to simultaneously
induce apoptosis;21 indeed, the biogenesis of MLBs has been
shown to result from autophagy in lung type II alveolar cells.74
In our case, we found no changes in the morphology of the
endoplasmic reticulum (ER) in contrast to what was observed
in human umbilical vein endothelial cells (HUVECs) where
doses of 120 M ZnO-NPs or ZnCl2 led to aggregation and
swelling of the ER, to activation of ER stress-responsive path-
ways and to a significant decrease in the number of mitochon-
dria.58 Pathways involved in the response to ZnO-NP stress
thus appear to be cell type dependent. In agreement with our
results, zinc accumulation in mitochondria was demonstrated
using a selective fluorescent dye after ZnO-NP treatment at
similar doses, on leukemia Jurkat cells and human lung carci-
noma H1355. Interestingly, higher doses induced also
depolarization of mitochondrial membranes, caspase
activation and cell apoptosis.75
Excess zinc ions is known to be sequestered by metallothio-
nein, excreted via ZnT1 and also stored in intracellular com-
partments, but the nature of these compartments stirs contro-
versy and probably depends on the cell type. Evidence was
found that zinc can be stored in mitochondria, in the ER, in
the Golgi apparatus, in lysosomes40 as well as in the contro-
versial zincosomes.2
Our working hypothesis is that zinc accumulates primarily
in mitochondria where it aects mitochondrial network
dynamics. Once autophagy has started, MLBs are formed to
sequester the damaged mitochondria. This hypothesis is also
backed by the observations that orally administered zinc salts
increase zinc content in rat liver mitochondria,76 reduces mito-
chondrial potential, alters mitochondrial metabolism and
leads to decreased ATP production in hepatocytes.77
Nanoscale
View Article Online
Nanoscale
3.9 Zinc entry in mitochondria
Proteins that transport zinc into mitochondria and distribute
it within them, as well as the control of these processes, are
largely unknown. However accumulation of zinc has been visu-
alized on (individual) isolated mitochondria via the uniporter-
dependent (calcium specific) and independent transport
mechanisms.78 Interestingly, co-treatments with calcium
reduced mitochondrial zinc uptake, and this also explains the
protective eect of calcium on entire cells that we report here
(see section 3.6). Metallothionein has been also described to
function as a metallochaperone for zinc delivery into prostate
and liver mitochondria,79 and zinc-loaded metallothioneins
imported into liver were proposed to modulate respiration.80
Thus, increasing the amount of zinc-loaded metallothionein
in cells could lead to mitochondria disruptions. Furthermore,
ZnT2 is the first zinc transporter to be found directly associ-
ated with mitochondria and possibly modulating their func-
tion in secretory cells.81 MTF1 was shown to regulate the
expression of ZnT2 in pancreatic acinar cells.82 Interestingly,
ZnT2 mRNA, while not normally detected in rat liver, shows
high expression in hepatocytes after an acute oral dose of
zinc.43 In the light of these observations, we decided to investi-
gate the expression of ZnT2 in HepG2 cells after zinc treatment
(Fig. 8). As the ZnT2 basal expression was very low in hepato-
cytes, we worked with high concentrations of cDNA. We
showed a 3040 fold over-expression of ZnT2 after all zinc
treatments, corroborating a possible active sequestration of
zinc in mitochondria via ZnT2.
3.10 Investigation of mitochondria dysfunction
We measured the relative gene expression of Mfn1, Mfn2 and
Drp1. Mfn1 and 2 are essential proteins for the mitochondrial
fusion process while Drp1 is an essential protein for the
fission process. No significant dierences of expression were
seen in our conditions of treatment (Fig. S4). Minor activation
of these genes had been shown in rat primary astrocytes
exposed to titanium dioxide nanoparticles but only at high
toxic doses.83
Fig. 8 Quantitative PCR analysis of ZnT2 mRNA expression in HepG2
cells after a 6 hour incubation with Zn acetate, ZnO-NP-FBS or
ZnO-NP-silane at 90 M. Results are presented as relative expression
changes after normalizing with HPRT and 36B4 mRNA. Each value rep-
resents the mean of relative expression +SEM of three independent
experiments.
This journal is The Royal Society of Chemistry 2016

Nanoscale
Published on 20 October 2016. Downloaded by Cornell University Library on 27/10/2016 06:54:32.
Fig. 9 Measurement of mitochondrial transmembrane potential after
incubation with 90 M zinc acetate, ZnO-NP-FBS or ZnO-NP-silane for
24 h. This gure shows the amount of rhodamine 123 internalized in the
cells expressed as a percentage of the mean uorescence of control
cells (three independent experiments). ***: p < 0.001.
In parallel, we explored mitochondrial integrity by measur-
ing mitochondrial transmembrane potential. In agreement
with literature on ZnO-NPs and others nanoparticles,24,34,39 we
observed a decrease in the mitochondrial potential using a
Rhodamine 123 uptake assay (Fig. 9) confirming mitochon-
drial damage after zinc treatments. In neurones, it has also
been shown that consequences of cellular zinc overload
include increased ROS production, loss of mitochondrial
membrane potential and reduced cellular ATP levels.84 ROS
generation could be a consequence of zinc stress and not the
first actor leading to toxicity. Interestingly it has been shown
that fusion of mitochondria in mammalian cells is dependent
on the mitochondrial inner membrane potential.85 Therefore,
the morphological modifications of mitochondria that we have
observed in TEM could be a direct consequences of the
decrease of mitochondrial potential.
In a study on ZnO-NP treated macrophages, Aude-Garcia
et al.86 showed as well convincing arguments in favour of res-
piratory chain defects rather than ROS generation as direct
causes of mitochondrial potential alteration. We can propose
the same scenario in hepatocytes based on our results.
3.11 Role of autophagy in cell survival in response to zinc
In order to test the role of autophagy in cell response to zinc,
we blocked the lysosomal proton pump in HepG2 cells using
bafilomycin A (Baf A) prior to exposing the cells to 90 or 150 M
of ZnO-NPs or zinc acetate (Fig. 10). We checked the eciency
of the Baf A treatment with an acridine orange assay and con-
firmed the total disappearance of acidic vesicles in these con-
ditions. Interestingly we found that the pre-treatment by Baf
A potentiated the toxicity of zinc in all cases, ions or NPs, in
contrast to what is usually reported in NP toxicity studies.
Indeed, Baf A has been shown to inhibit copper ion release
in acidic lysosomes and to significantly rescue A549 cells from
cell death induced by copper NP internalization and dis-
solution.87 In a similar study, the toxicity of iron oxide nano-
particles (IONPs) in microglial cultures was prevented by
neutralizing lysosomal pH by Baf A, thus avoiding rapid iron
This journal is The Royal Society of Chemistry 2016
View Article Online
Paper
Fig. 10 Eect of balomycin A on Zn toxicity: HepG2 cells were pre-
treated 1 hour with 100 nM Baf A before adding Zn acetate,
ZnO-NP-FBS or ZnO-NP-silane at subtoxic doses of 90 and 150 M for
24 h. Baf A treatment alone had no eect on viability of HepG2 cells.
liberation from IONPs.88 Therefore, when NPs are internalized
and further dissolved inside the acidic lysosomes (known as
the Trojan horse mechanism), blocking lysosomal acidifica-
tion has a protective eect. In our conditions, Baf A did not
have this protective eect, suggesting again that zinc enters
cells in its ionic form using another mechanism. We showed
an increase in acidic vesicles after treatment with sub-toxic
zinc concentrations, suggesting a defence mechanism by auto-
phagy, limiting in these conditions zinc release from aected
mitochondria. At higher doses, in contrast, autophagy could
trigger apoptosis. Consequently, inhibiting autophagy by Baf A
at low doses of zinc may prevent cells from protecting them-
selves against it and therefore potentiates zinc-induced
toxicity. In parallel to this eect, recently published results
suggest that lysosomes play a role as zinc sinks, temporarily
storing zinc. Indeed, Kukic et al.89 found that the inhibition of
the lysosomal zinc sink function by Baf A increased cyto-
plasmic zinc levels and favoured apoptosis. They found that
lysosomes actively absorb zinc and secrete it after fusion with
the plasma membrane. Such phenomena could also explain
our results on zinc accumulation in cells which reaches a
plateau in 24 to 48 h. Blocking this process could lead to more
accumulation of zinc in cells. ZnT2 transporters are generally
present only in mitochondria, however some authors90 have
found them also in lysosomes. So we can hypothesise an
accumulation of zinc in lysosomes via autophagy of poisoned
mitochondria or directly via ZnT2 or both. Maintenance of
zinc homeostasis is a complex process and redundancy is to be
expected, which makes it unlikely that mitochondria are the
only organelles acting as zinc sinks.
4 Conclusion
In this work on HepG2 cells, at a sub-toxic dose of 90 M, we
found no evidence of zinc entry in its particulate form and on
Nanoscale

Paper
the contrary, many arguments in favour of major dissolution
of nanoparticles outside the cell and zinc entry in its ionic
form.
We demonstrate that the entry of zinc induced a zinc stress
response and the up-regulation of MET, ZnT1, ZnT2. No
change in CAT, SOD1 and SOD2 expression, nor in the enzy-
matic activities of catalase and SOD, were highlighted, imply-
ing a minimal activation of oxidative stress. TEM analysis
however put to light early mitochondria injury under sub-toxic
Published on 20 October 2016. Downloaded by Cornell University Library on 27/10/2016 06:54:32.
exposure to zinc, suggesting that excess zinc is mainly trans-
ported and stored in mitochondria. ZnT2 up-regulation
strengthens this hypothesis as well as the induction of auto-
phagy. Zinc entry in mitochondria could be driven by MET
and/or ZnT2 and lead to mitochondria disruption. Under sub-
toxic conditions, cells are able to deal with excess zinc and
eliminate deleterious mitochondria by increasing autophagy
but at higher doses, mitochondria injury could lead to apopto-
sis. Working at low doses allowed us to study the early
response of hepatocytes and to highlight the first element of
the response cascade. Future perspectives concern the study of
the subcellular fate of zinc in cells after exposure to ZnO-NPs
using synchrotron approaches in order to confirm zinc
accumulation in mitochondria.
Acknowledgements
The authors thank Aurlien Deniaud for useful discussions,
Vronique Collin-Faure for FACS experiments and Josiane
Arnaud (CHU Grenoble) for metal quantification. This work
was funded by the CEA-Toxicology Transversal Program
through the NanoStress grant. This research is part of the
LabEx SERENADE (grant ANR-11-LABX-0064) and the LabEx
ARCANE (grant ANR-11-LABX-0003-01). The platforms of the
Grenoble Instruct Centre (ISBG; UMS 3518
CNRS-CEA-UJF-EMBL) were used, with support from FRISBI
(ANR-10-INSB-05-02) and GRAL (ANR-10-LABX-49-01), within
the Grenoble Partnership for Structural Biology (PSB). The
electron microscope facility is supported by the Rhne-Alpes
Region, the Fondation pour la Recherche Mdicale, the Fonds
Europen de Dveloppement conomique et Rgional, the 22 K. N. Yu, T. J. Yoon, A. Minai-Tehrani, J. E. Kim, S. J. Park,
Centre National de la Recherche Scientifique, the
Commissariat lnergie Atomique, the Universit de
Grenoble Alpes, EMBL and the GIS-Infrastructures en Biologie
Sant et Agronomie (IBISA).
References
1 W. Maret, Biochemistry, 2004, 43, 33013309.
2 D. Beyersmann and H. Haase, Biometals, 2001, 14, 331341.
3 R. A. Cherny, C. S. Atwood, M. E. Xilinas, D. N. Gray,
W. D. Jones, C. A. McLean, K. J. Barnham, I. Volitakis,
F. W. Fraser, Y. Kim, X. Huang, L. E. Goldstein, R. D. Moir,
J. T. Lim, K. Beyreuther, H. Zheng, R. E. Tanzi,
C. L. Masters and A. I. Bush, Neuron, 2001, 30, 665676.
Nanoscale
View Article Online
Nanoscale
4 Q. Hao and W. Maret, JAD, J. Alzheimers Dis., 2005, 8, 161
170; discussion 209115.
5 W. Maret and Y. Li, Chem. Rev., 2009, 109, 46824707.
6 I. Sekler, S. L. Sensi, M. Hershfinkel and W. F. Silverman,
Mol. Med., 2007, 13, 337343.
7 L. A. Lichten and R. J. Cousins, Annu. Rev. Nutr., 2009, 29,
153176.
8 V. Gunther, U. Lindert and W. Schaner, Biochim. Biophys.
Acta, 2012, 1823, 14161425.
9 E. Roduner, Chem. Soc. Rev., 2006, 35, 583592.
10 O. Bondarenko, K. Juganson, A. Ivask, K. Kasemets,
M. Mortimer and A. Kahru, Arch. Toxicol., 2013, 87, 1181
1200.
11 G. Oberdorster, A. Maynard, K. Donaldson, V. Castranova,
J. Fitzpatrick, K. Ausman, J. Carter, B. Karn, W. Kreyling,
D. Lai, S. Olin, N. Monteiro-Riviere, D. Warheit and
H. Yang, Part. Fibre Toxicol., 2005, 2, 8.
12 A. A. Keller, S. McFerran, A. Lazareva and S. Suh,
J. Nanopart. Res., 2013, 15, 1692.
13 M. J. Osmond and M. J. McCall, Nanotoxicology, 2010, 4,
1541.
14 J. T. Seil and T. J. Webster, Int. J. Nanomed., 2012, 7, 2767
2781.
15 R. J. Vandebriel and W. H. De Jong, Nanotechnol., Sci. Appl.,
2012, 5, 6171.
16 L. Wang, W. Ding and F. Zhang, J. Nanosci. Nanotechnol.,
2010, 10, 86178624.
17 G. J. Nohynek, J. Lademann, C. Ribaud and M. S. Roberts,
Crit. Rev. Toxicol., 2007, 37, 251277.
18 A. Adamcakova-Dodd, L. V. Stebounova, J. S. Kim,
S. U. Vorrink, A. P. Ault, P. T. OShaughnessy, V. H. Grassian
and P. S. Thorne, Part. Fibre Toxicol., 2014, 11, 15.
19 C. H. Li, C. C. Shen, Y. W. Cheng, S. H. Huang, C. C. Wu,
C. C. Kao, J. W. Liao and J. J. Kang, Nanotoxicology, 2012, 6,
746756.
20 J. Heim, E. Felder, M. N. Tahir, A. Kaltbeitzel,
U. R. Heinrich, C. Brochhausen, V. Mailander, W. Tremel
and J. Brieger, Nanoscale, 2015, 7, 89318938.
21 R. Roy, S. K. Singh, L. K. Chauhan, M. Das, A. Tripathi and
P. D. Dwivedi, Toxicol. Lett., 2014, 227, 2940.
M. S. Jeong, S. W. Ha, J. K. Lee, J. S. Kim and M. H. Cho,
Toxicol. In Vitro, 2013, 27, 11871195.
23 A. Kermanizadeh, G. Pojana, B. K. Gaiser, R. Birkedal,
D. Bilanicova, H. Wallin, K. A. Jensen, B. Sellergren,
G. R. Hutchison, A. Marcomini and V. Stone,
Nanotoxicology, 2013, 7, 301313.
24 C. Filippi, A. Pryde, P. Cowan, T. Lee, P. Hayes,
K. Donaldson, J. Plevris and V. Stone, Nanotoxicology, 2015,
9, 126134.
25 M. Cuillel, M. Chevallet, P. Charbonnier, C. Fauquant,
I. Pignot-Paintrand, J. Arnaud, D. Cassio, I. Michaud-Soret
and E. Mintz, Nanoscale, 2014, 6, 17071715.
26 P. Muller, H. van Bakel, B. van de Sluis, F. Holstege,
C. Wijmenga and L. W. Klomp, JBIC, J. Biol. Inorg. Chem.,
2007, 12, 495507.
This journal is The Royal Society of Chemistry 2016

Nanoscale
27 R. F. Beers and I. W. Sizer, J. Biol. Chem., 1952, 195, 133
140.
28 S. Marklund and G. Marklund, Eur. J. Biochem., 1974, 47,
469474.
29 M. M. Bradford, Anal. Biochem., 1976, 72, 248254.
30 S. Paglin, T. Hollister, T. Delohery, N. Hackett,
M. McMahill, E. Sphicas, D. Domingo and J. Yahalom,
Cancer Res., 2001, 61, 439444.
31 L. V. Johnson, M. L. Walsh and L. B. Chen, Proc. Natl. Acad.
Published on 20 October 2016. Downloaded by Cornell University Library on 27/10/2016 06:54:32.
Sci. U. S. A., 1980, 77, 990994.
32 K. W. Powers, S. C. Brown, V. B. Krishna, S. C. Wasdo,
B. M. Moudgil and S. M. Roberts, Toxicol. Sci., 2006, 90,
296303.
33 C. M. Sayes and D. B. Warheit, Wiley Interdiscip. Rev.:
Nanomed. Nanobiotechnol., 2009, 1, 660670.
34 V. Sharma, D. Anderson and A. Dhawan, Apoptosis, 2012,
17, 852870.
35 S. Triboulet, C. Aude-Garcia, L. Armand, A. Gerdil,
H. Diemer, F. Proamer, V. Collin-Faure, A. Habert,
J. M. Strub, D. Hanau, N. Herlin, M. Carriere, A. Van
Dorsselaer and T. Rabilloud, Nanoscale, 2014, 6, 61026114.
36 M. O. Parat, M. J. Richard, C. Meplan, A. Favier and
J. C. Beani, Biol. Trace Elem. Res., 1999, 70, 5168.
37 E. Emri, E. Miko, P. Bai, G. Boros, G. Nagy, D. Rozsa,
T. Juhasz, C. Hegedus, I. Horkay, E. Remenyik and G. Emri,
Metallomics, 2015, 7, 499507.
38 P. J. Moos, K. Olszewski, M. Honeggar, P. Cassidy,
S. Leachman, D. Woessner, N. S. Cutler and J. M. Veranth,
Metallomics, 2011, 3, 11991211.
39 S. Triboulet, C. Aude-Garcia, M. Carriere, H. Diemer,
F. Proamer, A. Habert, M. Chevallet, V. Collin-Faure,
J. M. Strub, D. Hanau, A. Van Dorsselaer, N. Herlin-Boime
and T. Rabilloud, Mol. Cell. Proteomics, 2013, 12, 3108
3122.
40 R. A. Colvin, W. R. Holmes, C. P. Fontaine and W. Maret,
Metallomics, 2010, 2, 306317.
41 M. Vasak and G. Meloni, JBIC, J. Biol. Inorg. Chem., 2011,
16, 10671078.
42 P. Babula, M. Masarik, V. Adam, T. Eckschlager,
M. Stiborova, L. Trnkova, H. Skutkova, I. Provaznik,
J. Hubalek and R. Kizek, Metallomics, 2012, 4, 739750.
43 J. P. Liuzzi and R. J. Cousins, Annu. Rev. Nutr., 2004, 24,
151172.
44 T. Ganz and E. Nemeth, Biochim. Biophys. Acta, 2012, 1823,
14341443.
45 S. Balesaria, B. Ramesh, H. McArdle, H. K. Bayele and
S. K. Srai, FEBS Lett., 2010, 584, 719725.
46 G. Thiel and G. Cibelli, J. Cell. Physiol., 2002, 193, 287292.
47 R. J. Cousins, R. K. Blanchard, M. P. Popp, L. Liu, J. Cao,
J. B. Moore and C. L. Green, Proc. Natl. Acad. Sci. U. S. A.,
2003, 100, 69526957.
48 F. Wang, J. Dufner-Beattie, B.-E. Kim, M. J. Petris,
G. Andrews and D. J. Eide, J. Biol. Chem., 2004, 279, 24631
24639.
49 S. K. Niture, R. Khatri and A. K. Jaiswal, Free Radical Biol.
Med., 2014, 66, 3644.
This journal is The Royal Society of Chemistry 2016
View Article Online
Paper
50 G. K. Andrews, Biometals, 2001, 14, 223237.
51 J. D. Van Norman and R. Szentirmay, Bioinorg. Chem.,
1974, 4, 3743.
52 S. H. Jeong, H. J. Kim, H. J. Ryu, W. I. Ryu, Y. H. Park,
H. C. Bae, Y. S. Jang and S. W. Son, J. Dermatol. Sci., 2013,
72, 263273.
53 T. Xia, M. Kovochich, M. Liong, J. I. Zink and A. E. Nel, ACS
Nano, 2008, 2, 8596.
54 C. Loos, T. Syrovets, A. Musyanovych, V. Mailnder,
K. Landfester, G. U. Nienhaus and T. Simmet, Beilstein
J. Nanotechnol., 2014, 5, 24032412.
55 R. Guan, T. Kang, F. Lu, Z. Zhang, H. Shen and M. Liu,
Nanoscale Res. Lett., 2012, 7, 602.
56 R. Herrmann, F. J. Garcia-Garcia and A. Reller, Beilstein
J. Nanotechnol., 2014, 5, 20072015.
57 S. K. Misra, A. Dybowska, D. Berhanu, S. N. Luoma and
E. Valsami-Jones, Sci. Total Environ., 2012, 438, 225232.
58 T. Buerki-Thurnherr, L. Xiao, L. Diener, O. Arslan,
C. Hirsch, X. Maeder-Althaus, K. Grieder, B. Wampfler,
S. Mathur, P. Wick and H. F. Krug, Nanotoxicology, 2013, 7,
402416.
59 B. M. Johnson, J. A. Fraietta, D. T. Gracias, J. L. Hope, C.
J. Stairiker, P. R. Patel, Y. M. Mueller, M. D. McHugh, L.
J. Jablonowski, M. A. Wheatley and P. D. Katsikis,
Nanotoxicology, 2015, 9, 737748.
60 R. Chen, L. Huo, X. Shi, R. Bai, Z. Zhang, Y. Zhao, Y. Chang
and C. Chen, ACS Nano, 2014, 8, 25622574.
61 V. Wilhelmi, U. Fischer, H. Weighardt, K. Schulze-Ostho,
C. Nickel, B. Stahlmecke, T. A. Kuhlbusch, A. M. Scherbart,
C. Esser, R. P. Schins and C. Albrecht, PLoS One, 2013, 8,
e65704.
62 S. A. James, B. N. Feltis, M. D. de Jonge, M. Sridhar,
J. A. Kimpton, M. Altissimo, S. Mayo, C. Zheng,
A. Hastings, D. L. Howard, D. J. Paterson, P. F. Wright,
G. F. Moorhead, T. W. Turney and J. Fu, ACS Nano, 2013, 7,
1062110635.
63 P. J. Moos, K. Chung, D. Woessner, M. Honeggar,
N. S. Cutler and J. M. Veranth, Chem. Res. Toxicol., 2010,
23, 733739.
64 V. Valdiglesias, C. Costa, G. Kilic, S. Costa, E. Pasaro,
B. Laon and J. P. Teixeira, Environ. Int., 2013, 55, 92100.
65 I. L. Hsiao and Y. J. Huang, J. Nanopart. Res., 2013, 15,
1829.
66 T. Xia, M. Kovochich, M. Liong, L. Madler, B. Gilbert,
H. Shi, J. I. Yeh, J. I. Zink and A. E. Nel, ACS Nano, 2008, 2,
21212134.
67 R. B. Reed, D. A. Ladner, C. P. Higgins, P. Westerho and
J. F. Ranville, Environ. Toxicol. Chem., 2012, 31, 9399.
68 R. Roy, V. Parashar, L. K. Chauhan, R. Shanker, M. Das,
A. Tripathi and P. D. Dwivedi, Toxicol. In Vitro, 2014, 28,
457467.
69 C. Shen, S. A. James, M. D. de Jonge, T. W. Turney,
P. F. Wright and B. N. Feltis, Toxicol. Sci., 2013, 136, 120
130.
70 W. Song, J. Zhang, J. Guo, F. Ding, L. Li and Z. Sun, Toxicol.
Lett., 2010, 199, 389397.
Nanoscale

Paper
71 Y. Wang, W. G. Aker, H. M. Hwang, C. G. Yedjou, H. Yu
and P. B. Tchounwou, Sci. Total Environ., 2011, 409,
47534762.
72 B. Gilbert, S. C. Fakra, T. Xia, S. Pokhrel, L. Madler and
A. E. Nel, ACS Nano, 2012, 6, 49214930.
73 D. Glick, S. Barth and K. F. Macleod, J. Pathol., 2010, 221,
312.
74 M. Hariri, G. Millane, M. P. Guimond, G. Guay, J. W. Dennis
and I. R. Nabi, Mol. Biol. Cell, 2000, 11, 255268.
Published on 20 October 2016. Downloaded by Cornell University Library on 27/10/2016 06:54:32.
75 Y. Y. Kao, Y. C. Chen, T. J. Cheng, Y. M. Chiung and
P. S. Liu, Toxicol. Sci., 2012, 125, 462472.
76 M. Yamaguchi, M. Kura and S. Okada, Chem. Pharm. Bull.,
1981, 29, 23702374.
77 J. Lemire, R. Mailloux and V. D. Appanna, J. Appl. Toxicol.,
2008, 28, 175182.
78 L. M. Malaiyandi, O. Vergun, K. E. Dineley and
I. J. Reynolds, J. Neurochem., 2005, 93, 12421250.
79 L. C. Costello, Z. Guan, R. B. Franklin and P. Feng, J. Inorg.
Biochem., 2004, 98, 664666.
80 B. Ye, W. Maret and B. L. Vallee, Proc. Natl. Acad.
Sci. U. S. A., 2001, 98, 23172322.
81 Y. A. Seo, V. Lopez and S. L. Kelleher, Am. J. Physiol.: Cell
Physiol., 2011, 300, C1479C1489.
Nanoscale
View Article Online
Nanoscale
82 L. Guo, L. A. Lichten, M. S. Ryu, J. P. Liuzzi, F. Wang and
R. J. Cousins, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 2818
2823.
83 C. L. Wilson, V. Natarajan, S. L. Hayward,
O. Khalimonchuk and S. Kidambi, Nanoscale, 2015, 7,
1847718488.
84 K. E. Dineley, T. V. Votyakova and I. J. Reynolds,
J. Neurochem., 2003, 85, 563570.
85 Y. Mattenberger, D. I. James and J.-C. Martinou, FEBS Lett.,
2003, 538, 5359.
86 C. Aude-Garcia, B. Dalzon, J.-L. Ravanat, V. Collin-Faure,
H. Diemer, J. M. Strub, S. Cianferani, A. Van Dorsselaer,
M. Carrire and T. Rabilloud, J. Proteomics, 2016, 134, 174
185.
87 E. Moschini, M. Gualtieri, M. Colombo, U. Fascio,
M. Camatini and P. Mantecca, Toxicol. Lett., 2013, 222,
102116.
88 C. Petters, K. Thiel and R. Dringen, Nanotoxicology, 2015,
111, DOI: 10.3109/17435390.2015.1071445.
89 I. Kukic, S. L. Kelleher and K. Kiselyov, J. Cell Sci., 2014,
127, 30943103.
90 R. D. Palmiter, T. B. Cole and S. D. Findley, EMBO J., 1996,
15, 17841791.
This journal is The Royal Society of Chemistry 2016