Effects and uptake of gold nanoparticles deposited at the air-liquid interface of a human epithelial airway model.

Abstract

The impact of nanoparticles (NPs) in medicine and biology has increased rapidly in recent years. Gold NPs have advantageous properties such as chemical stability, high electron density and affinity to biomolecules, making them very promising candidates as drug carriers and diagnostic tools. However, diverse studies on the toxicity of gold NPs have reported contradictory results. To address this issue, a triple cell co-culture model simulating the alveolar lung epithelium was used and exposed at the air-liquid interface. The cell cultures were exposed to characterized aerosols with 15 nm gold particles (61 ng Au/cm2 and 561 ng Au/cm2 deposition) and incubated for 4 h and 24 h. Experiments were repeated six times. The mRNA induction of pro-inflammatory (TNFalpha, IL-8, iNOS) and oxidative stress markers (HO-1, SOD2) was measured, as well as protein induction of pro- and anti-inflammatory cytokines (IL-1, IL-2, IL-4, IL-6, IL-8, IL-10, GM-CSF, TNFalpha, INFgamma). A pre-stimulation with lipopolysaccharide (LPS) was performed to further study the effects of particles under inflammatory conditions. Particle deposition and particle uptake by cells were analyzed by transmission electron microscopy and design-based stereology. A homogeneous deposition was revealed, and particles were found to enter all cell types. No mRNA induction due to particles was observed for all markers. The cell culture system was sensitive to LPS but gold particles did not cause any synergistic or suppressive effects. With this experimental setup, reflecting the physiological conditions more precisely, no adverse effects from gold NPs were observed. However, chronic studies under in vivo conditions are needed to entirely exclude adverse effects.

Full text

Page 1

Effects and uptake of gold nanoparticles deposited at the air–liquid interface of a
human epithelial airway model
C. Brandenbergera,⁎, B. Rothen-Rutishausera, C. Mühlfeldb, O. Schmidc, G.A. Ferronc, K.L. Maierc,
P. Gehra, A.-G. Lenzc
aInstitute of Anatomy, Division of Histology, University of Bern, Bern, Switzerland
bInstitute of Anatomy and Cell Biology, Justus-Liebig-University Giessen, Giessen, Germany
cInstitute of Lung Biology and Disease, Helmholtz Zentrum München, Neuherberg, Germany
a b s t r a c ta r t i c l ei n f o
Article history:
Received 13 July 2009
Revised 9 September 2009
Accepted 22 September 2009
Available online 29 September 2009
Keywords:
Gold nanoparticles
Nanotoxicology
Human epithelial airway model
Air–liquid exposure
Particle lung interaction
The impact of nanoparticles (NPs) in medicine and biology has increased rapidly in recent years. Gold NPs
have advantageous properties such as chemical stability, high electron density and affinity to biomolecules,
making them very promising candidates as drug carriers and diagnostic tools. However, diverse studies on
the toxicity of gold NPs have reported contradictory results. To address this issue, a triple cell co-culture
model simulating the alveolar lung epithelium was used and exposed at the air–liquid interface.
The cell cultures were exposed to characterized aerosols with 15 nm gold particles (61 ng Au/cm2and
561 ng Au/cm2deposition) and incubated for 4 h and 24 h. Experiments were repeated six times. The mRNA
induction of pro-inflammatory (TNFα, IL-8, iNOS) and oxidative stress markers (HO-1, SOD2) was measured,
as well as protein induction of pro- and anti-inflammatory cytokines (IL-1, IL-2, IL-4, IL-6, IL-8, IL-10, GM-
CSF, TNFα, INFγ). A pre-stimulation with lipopolysaccharide (LPS) was performed to further study the effects
of particles under inflammatory conditions. Particle deposition and particle uptake by cells were analyzed by
transmission electron microscopy and design-based stereology.
A homogeneous deposition was revealed, and particles were found to enter all cell types. No mRNA induction
due to particles was observed for all markers. The cell culture system was sensitive to LPS but gold particles
did not cause any synergistic or suppressive effects.
With this experimental setup, reflecting the physiological conditions more precisely, no adverse effects from
gold NPs were observed. However, chronic studies under in vivo conditions are needed to entirely exclude
adverse effects.
© 2009 Elsevier Inc. All rights reserved.
Introduction
Duringthelast yearstheimpactof nanoparticles(NPs)in medicine
and biology has increased rapidly, especially for bioimaging (Jain et
al., 2008), biosensing (Olofsson et al., 2003) and drug delivery (Dhar
et al., 2008; Joshi et al., 2006). Gold NPs are chemically stable, electron
dense and posses an affinity to biomolecules, such as amino acids
(Selvakannan et al., 2004), proteins (Wangoo et al., 2008) and DNA
(Rosi et al., 2006), making them suitable as drug carriers and imaging
reagents. Promising results have been obtained in cancer research
where conjugated gold NPs have been used for cancer detection and
treatment in vitro (Chen et al., 2007; Khaing Oo et al., 2008; Li et al.,
2009).
Despite the potential medical benefits of gold NPs, it remains
unclear whether their use in biological organisms and humans, in
particular, will be safe. A variety of toxicity tests in vitro were reported
for gold NPs providing controversial results. Most studies have
excluded the toxicity of gold NPs (4 nm–18 nm in diameter) (Connor
et al., 2005; Khan et al., 2007; Shukla et al., 2005), studying cell
viability, pro-apoptotic effects, oxidative stress and inflammatory
response. In contrast, cytotoxic and pro-apoptotic effects of gold
particles ≤2 nm were reported (Pan et al., 2007; Tsoli et al., 2005), as
well as the impact of the charge of surface coatings on 2 nm particles
on the reduction of cell viability (Goodman et al., 2004). Furthermore,
exposure of fibroblasts to 13 nm gold particles caused morphological
Toxicology and Applied Pharmacology 242 (2010) 56–65
Abbreviations: ALICE, air–liquid interface cell exposure system; ANOVA, analysis of
variance; cDNA, complementary deoxyribonucleic acid; EC, epithelial cells; GM-CSF,
granulocyte macrophage colony stimulating factor; HO-1, hemoxigenase 1; IC50,
concentration causing 50% inhibition; IL-1β, interleukin-1β; IL-2, interleukin-2; IL-4,
interleukin-4; IL-6, interleukin-6; IL-8, interleukin-8; IL-10, interleukin-10; INFγ,
interferon γ; iNOS, inducible nitric oxide synthase; LC50, concentration causing 50%
lethality; LPS, lipopolysaccharide; LSM, laser scanning microscopy; MDDC, monocyte
derived dendritic cells; MDM, monocyte derived macrophages; mRNA, messenger
ribonucleic acid; NPs, nanoparticles; NO, nitric oxide; PBS, phosphate buffered saline;
PCR, polymerase chain reaction; RCF, relative centrifugal force; SOD2, super oxide
dismutase 2; TEM, transmission electron microscopy; TNFα, tumor necrosis factor α.
⁎ Corresponding author. University of Bern, Institute of Anatomy, Baltzerstrasse 2,
CH-3000 Bern 9, Switzerland. Fax: +41 31 6313807.
E-mail address: brandenberger@ana.unibe.ch (C. Brandenberger).
0041-008X/$ – see front matter © 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.taap.2009.09.014
Contents lists available at ScienceDirect
Toxicology and Applied Pharmacology
journal homepage: www.elsevier.com/locate/ytaap

Page 2

changes in the cytoskeleton and a reduction in cell proliferation
(Pernodet et al., 2006). A pro-inflammatory response was found after
exposure to bovine serum albumin coated 25 nm gold particles in an
epithelial airway model (Rothen-Rutishauser et al., 2007).
The reasons for these controversial results might include different
experimental set ups and particle characteristics. Recent studies have
identified that a number of important parameters influence the
toxicity of NPs (Nel et al., 2006; Oberdörster et al., 2005), such as dose
(Kwon et al., 2009), size (Kreyling et al., 2006), shape (Oberdörster
et al., 2005), bulk material (Hussain et al., 2005), surface charge (Cho
et al., 2009; Goodman et al., 2004), surface area (Brandenberger et al.,
2009; Stoeger et al., 2006), as well as the composition of the exposure
medium (Herzog et al., 2009b).
All the studies on gold NPs, discussed previously, were performed
in cell cultures under submerged conditions with the particles
suspended in different media. Besides the effects of the media on
agglomeration status, the movements of single NPs in fluids are
mainly driven by diffusion and not by sedimentation (Limbach et al.,
2005; Teeguarden et al., 2007). As a consequence from the latter the
probability of agglomerates to settle down onto the cell culture is
higher than that for single NPs, thus increasing the possibility of
attributing the observations to the wrong kind of particles (Limbach
et al., 2005). A direct deposition of the particles on the cells at the air–
liquid interface has the advantage of minimizing these effects,
resulting in a more reliable dosimetry and less particle agglomeration.
The air–liquid interface exposure scenario not only shows ideal
characteristics to study particle-cell interactions but also mimics
particle deposition in the respiratory tract after inhalation and it is
conceivable that the cellular response to particle exposure is different
for submerged and air–liquid interface culture conditions.
Lung epithelial cells represent, along with alveolar macrophages,
the first line of cellular defense against inhaled particles. These cells
participate in the initiation and modulation of the inflammatory
responseby theproductionof chemokines. Of particularinterest is the
involvement of dendritic cells, which are present at the base of the
epitheliumandare themost competentantigen presentingcellsin the
lung (Holt et al., 1990). Therefore, in this study an in vitro co-culture
model of the human epithelial airway barrier was used (Blank et al.,
2007; Rothen-Rutishauser et al., 2005), consisting of human blood
monocyte derived macrophages and dendritic cells as well as the
widely used A549 human alveolar epithelial cell line.
The present study aimed to analyze the inflammatory and
oxidative potential of 15 nm gold NPs in the triple cell co-culture
model exposed at the air–liquid interface with a newly developed
exposure system (Schmid et al., 2009), which allows for efficient,
spatially uniform and dose-controlled exposure of cells to NPs. In
addition, it was investigated whether the gold NPs aggravate or
suppress the response to a pro-inflammatory stimulus, a finding
previously observed by others (Chan et al., 2006; de Haar et al., 2006;
Herzog et al., 2009a; Hofer et al., 2004).
Materials and methods
Air–liquid interface exposure system
The cell cultures were exposed to the particles at the air–liquid
interface with an exposure system (ALICE), described by Schmid et al.
(2009). Briefly, the ALICE consists of three main components, a droplet
generator (nebulizer), an exposure chamber and a flow system with an
incubation chamber, which provides temperature and humidity
conditions suitable for cell cultivation (temperature: 37 °C; humidity:
nearly saturated). A dense cloud of micron-sized droplets (mass
median diameter 4.4–5.4 μm; geometric standard deviation: 1.50–
1.65 μm) containing the gold NPs was generated by nebulisation of
1 mL NP suspension using a vibrating membrane droplet generator
(investigational eFlow, PARI Pharma GmbH, Munich, Germany). This
type of droplet generator (TouchSpray™ technology) utilizes a
perforated, piezoelectrically driven vibrating membrane to induce
acoustic pressure waves, which periodically press small amounts of
liquid through the tapered holes of a membrane. The resulting dense
cloud of droplets containing the NPs was transported at a flow rate of
5 L/min into the exposure chamber (20×20×30 cm), where the
cultured cells in standard transwell plates were introduced. The
nebulizer gradually filled the chamber with the droplet cloud and as
soon as the nebulizer was empty, the air flow was stopped and the
droplets settled to the ground onto the cells. The deposition occurred
due to an effect known as cloud settling: the highly concentrated cloud
of droplets (80–130 g/m3) gravitated swiftly within 2 s to the bottom
of the exposure chamber (400 cm2), was further diverted to all sides
by the ground plate of the chamber and formed an almost symmetric
patternof vortices providinggentle butsufficient mixingto establisha
spatially uniform cloud layer near the bottom of the chamber.
The ALICE allowed for uniform and efficient deposition (57±7% of
the suspension reached the bottom plate of the exposure chamber) of
colloidal gold NPs onto the cells at the air–liquid interface. After the
exposure procedure, which lasted for about 20 min, the cells were
kept under air–liquid interface conditions for post-exposure incuba-
tion times of 4 h and 24 h in the cell incubator. The effective dose of
gold NPs deposited on the cells in ng/cm2was measured by gamma
spectroscopy as described below.
Particles
Commercially available aqueous 15 nm colloidal gold particles
from British Biocell International (EM.GC15, Batch 7894, Plano GmbH,
Wetzlar, Germany) in a citrate buffer were used. The particle
suspension had a nominal concentration of 0.004% gold colloids,
corresponding to a mass concentration of 40 μg/mL. A 10-fold
concentration of the colloidal gold was achieved by centrifugation of
the suspension at 19 RCF (relative centrifugal force) for 20 min and
removing 90% of the supernatant without particles. The size of the
particles was analysed in suspension by dynamic light scattering
(HPPS 5001, Malvern Instruments Ltd, Worcestershire, UK) and post-
air–liquid interface exposure by transmission electron microscopy
(TEM; Philips CM12, FEI Co. Philips Electron Optics, Zürich, Switzer-
land; magnification 15,000×) after placing a TEM grid onto the
ground plate of the ALICE during some of the experiments. The Au
mass deposited on the cells was determined by placing aluminium foil
sheets with a size of 8×4 cm on the bottom of the exposure chamber
during an exposure run. After neutron activation of Au-197 into Au-
198 for 1 h (the neutron flux was about 6×1012cm−2s−1) the Au
mass on the aluminium foil was determined from the intensity of the
412 keV gamma line of Au-198 relative to a known standard.
Cell cultures
A549 cell line.
representing the alveolar type II phenotype (Lieber et al., 1976), were
obtained from the American Tissue Type Culture Collection (LGC
Promochem, Molsheim, France). Cells (passage number 5 to 30) were
maintained in RPMI 1640 medium (w/25 mM HEPES, Invitrogen
GmbH, Karlsruhe, Germany) with 1% L-glutamine (Invitrogen), 1%
penicillin/streptomcyin (Biochrom, Germany), and 10% foetal calf
serum (FBS Superior Biochrom, Germany). Cells were seeded at a
density of 106cells in 2 mL on BD Falcon™ 6-well plate cell-culture
transwells (high pore density PET membranes with a growth area of
4.2 cm2and 3.0 μm pores in diameter; Omnilab GmbH, Munich,
Germany). Transwells were placed in BD Falcon™ 6-well tissue
culture plates with 2 mL medium in the upper and 3 mL in the lower
transwell chamber. Cells were grown on transwell membranes under
submerged conditions for 7 days to grow to confluence. The medium
was changed twice a week.
A549 cells, a human alveolar epithelial-like cell line
57
C. Brandenberger et al. / Toxicology and Applied Pharmacology 242 (2010) 56–65

Page 3

The triple cell co-culture.
alveolar epithelial cells (EC), human blood monocyte derived
macrophages (MDM) and dendritic cells (MDDC) was used
(Rothen-Rutishauser et al., 2005). A blood donation of 200 mL was
centrifuged at 1300 g for 20 min to receive a buffy coat which was
further processed according to the method described by Sallusto et
al. (1995). Isolated blood monocytes were cultured for 7 days in
RPMI 1640 medium with 1% L-glutamine, 1% penicillin/streptomy-
cin, and 5% heat-inactivated human serum. The MDDC differentia-
tion was obtained by an addition of 34 ng/mL IL-4 (Sigma Aldrich
GmbH, Munich, Germany) and 50 ng/mL GM-CSF (Sigma Aldrich
GmbH, Munich, Germany), whereas the MDM did not receive any
additional supplements. The setup of the triple cell co-culture was
performed as described previously (Rothen-Rutishauser et al., 2005).
The cells were transferred from submerged to air–liquid interface
conditions, 24 h prior to particle exposure: The cell culture medium
from the upper transwell chamber was removed and the cell culture
medium in the lower transwell chamber was replaced by 1.8 mL of
fresh medium. In a set of experiments, an inflammatory stimulus
was applied 2 h prior to the particle exposure by adding 1 ng/mL
lipopolysaccharide (LPS) (Pseudomonas aeruginosa, Sigma Aldrich
GmbH, Munich, Germany) into the medium of the lower transwell
chamber.
A triple cell co-culture system with A549
Real-time PCR
RNA isolation was done with the Qiagen RNeasy Mini Kit (Qiagen
GmbH, Hilden, Germany). The cells were removed from the cell
culture membrane with a cell scraper and the lysis buffer provided by
the supplier of the Qiagen RNeasy Mini Kit. The cell lysate was then
centrifuged in shredder columns (QIAshredder, Qiagen GmbH, Hilden,
Germany) for 2 min at 16 RCF. The isolation was performed according
to the supplier's manual including a step of DNA digestion (Qiagen
GmbH, Hilden, Germany). The purified RNA was eluted in 30 μL pure
H2O and stored at −70 °C.
The RNA concentration was measured with the Nano-Drop-
Photometer (NanoDrop ND100 PeqLab, Germany). Transcription
was performed with a total amount of 0.5 μg RNA in a volume of
20 μL reaction mixture. Alignment of random nonamer primer (5 μM
final concentration) and RNA in H2O was done at 70 °C for 5 min,
followed by cooling on ice for 5 min. Mastermix (1× First Strand
Buffer, 10 mM DTT, 8 U/μL Superscript II RT, 2 U/μL RNAse Inhibitor,
Invitrogen GmbH, Karlsruhe, Germany and 0.5 mM 4dNTP, Fermentas
GmbH, St. Leon-Rot, Germany) was added and the mixture incubated
for 1 h at 42 °C, followed by Superscript inactivation for 15 min at
70 °C. Complementary DNA (cDNA) was diluted to a concentration of
66 ng/μL and stored at −20 °C.
The reaction mixture for quantitative real-time PCR contained
200 ng cDNA, 0.4 μM forward and reverse Primer and Absolute blue
QPCR SYBR Green ROX Mix (Thermo Scientific, ABgene, UK). Primer
sequences are shown in Table 1. The thermo cyclic protocol was: 1×
(50 °C, 2 min; 95 °C, 15 min) 40×(95 °C, 15s; 60 °C, 1 min) followed
by a dissociation stage of 95 °C, 15s; 60 °C, 20 s; 95 °C 15 s. The thermo
cyclic reaction and software analysis was performed with the 7500
Real-Time PCR System (Applied Biosystem, Darmstadt, Germany).
Bioplex cytokine detection
After 24 h exposure, 1 mL medium from the lower transwell
chamber was sampled and immediately frozen and stored at −70 °C
until the assay was performed. The detection of pro- and anti-
inflammatory cytokines released into the culture medium was carried
out with the Bio-Plex Pro Human Cytokine 8-Plex Panel for the
detection of interleukin-2 (IL-2), interleukin-4 (IL-4), interleukin-6
(IL-6), interleukin-8 (IL-8), interleukin-10 (IL-10), granulocyte mac-
rophage colony stimulating factor (GM-CSF), tumor necrosis factor α
(TNFα) and interferon γ (INFγ) (BioRad, Munich, Germany). The
assay for interleukin-1β (IL-1β) was added to the panel. For the
cytokine determination 50 μL cell culture supernatant was used and
processed according to the supplier's protocol.
Quantification of MDM, MDDC and EC
The triple cell co-cultures were fixed on the cell culture transwell
with 3% paraformaldehyde in phosphate buffered saline (PBS) for
15 min at room temperature and then treated with 0.1 M glycine in
PBS for 5 min. Before staining, the cells were permeabilized with 0.2%
Triton X-100 in PBS for 15 min at room temperature. MDM were
marked for 1 h with the primary antibody mouse anti-human CD14
(Clone UCHM-1, C 7673; Sigma) and MDDC with mouse anti-human
CD86 (Clone HB15e, 36931A; PharMingen, BD Biosciences). Goat anti-
mouse cyanine 5 (AP124S; Chemicon, VWR International AG, Life
Sciences) was used as a secondary detection antibody. The cytoskel-
eton of all cells was stained with phalloidin Alexa 488 1:100 (R-415;
Molecular Probes, Invitrogen AG, Basel, Switzerland). Preparations for
optical analysis were mounted in PBS:glycerol (2:1) containing
170 mg/mL Mowiol 4-88 (Calbiochem, VWR International AG,
Dietikon, Switzerland). The samples were visualized with an inverted
Zeiss laser scanning microscope (LSM) 510 Meta (Axiovert 200 M,
Lasers: HeNe 633 nm, and Ar 488 nm). Image processing was
performed with the 3D multi-channel image processing software
IMARIS (Bitplane AG, Zürich, Switzerland). From each experiment 10
random images with a size of 71 μm×71 μm were evaluated and the
number of EC, MDM and MDDC per mm2was counted.
Transmission electron microscopy
Intracellular particles were visualized by conventional transmis-
sion electron microscopy (TEM) and quantified with stereological
methods. For TEM analysis, the exposed cells on the transwell
membrane were fixed with 2.5% glutaraldehyde in 0.03 M potassium
phosphate buffer for at least 24 h. Then they were washed with
potassium phosphate buffer, post-fixed with 1% osmium tetroxide in
sodium cacodylate buffer, washed with maleate buffer, and stained en
bloc with 0.5% uranyl acetate in maleate buffer. Afterwards, the cells
were dehydrated in ascending ethanol series, and embedded in epon
(Mühlfeld et al., 2007). From the embedded cells, ultrathin sections
were cut parallel to the vertical axis of the cells, mounted on copper
grids and stained with lead citrate and uranyl acetate. The substances
used in this protocol are hazardous and have to be handled carefully
according to safety guidelines. All procedures were performed in a
hoodwith additionaladequatepersonalprotection. Imaging wasdone
with a Morgani TEM (FEI Co Philips Electron Optics, Zürich,
Switzerland).
Intracellular particle quantification by design based stereology
Stereology enables the estimation of three-dimensional structural
features (number, length, surface area or volume) from two-
dimensional sections. All parameters are first determined as densities,
i.e. as estimate per unit reference volume, and are then converted to
Table 1
Sequences of the used primers.
GENE Primer forward 5′-3′
Primer reverse 5′-3′
GAPDH
IL-8
TNFα
iNOS
SOD2
HO-1
CCATGAGAAGTATGACAACAGCC
ATGACTTCCAAGCTGGCCGTGGTC
CCAAAGTAGACCTGCCCAGA
GCCCAAGGTCTATGTTCAG
CCTGGAACCTCACATCAACG
AAGATTGCCCAGAAAGCCCTGGAC
TGGCAGGTTTTTCTAGACCC
TCTCAGCCCTCTTCAAAAACTTCTC
TCTACTCCCAGGTCCTCTTCA
TAGTCCTCGACCTGCTCCTC
AACCTGAGCCTTGGACACC
AACTGTCGCCACCAGAAAGCTGAG
58
C. Brandenberger et al. / Toxicology and Applied Pharmacology 242 (2010) 56–65

Page 4

the total value by multiplication with the reference volume. As a
reference volume, the cell volume per mm2of transwell membrane of
the upper and of the basal side of the transwell membrane was
estimated. Test fields showing the cells were chosen by systematic
uniform random sampling (Mayhew, 2008) at a magnification of
2,200×. A cycloid test line system for surface density estimations on
vertical sections (Baddeley et al., 1986) was projected onto each test
field with the vertical axis of the test system aligned to the vertical
axis of the cells. Intersections (I) of the cycloid test lines with the
transwell membrane and the number of test points (P) hitting the
cells were counted. The surface density (SV) of the transwell
membrane per unit cell volume was calculated as follows:
SVðTranswell membrane=CellÞ ¼ 2×I=ðLT×PÞ
from the number of intersections (I), number of points (P) and the
total length of the test line (LT) see e.g. Weibel et al. (2007). The
amountof particlespercell volumewasestimated in randomtest field
areas of ultra thin sections and considering the section thickness as
the third volume dimension, as described by Griffiths (1993). Test
fields of the cells were taken at a magnification of 22,000× by
systematic uniform random sampling and the number of particles per
test field were counted within a counting frame. The volume of the
test field was calculated from its area (AT) and the thickness of the
ultrathin section (d), which has been estimated by the method of the
smallest fold (Small, 1968). The numerical density of particles (NV)
was achieved by:
ð1Þ
NVðNanoparticle=CellÞ ¼ nP=ðnT×AT×dÞ
from the summed number of particles (nP) in n test fields (nT)with the
volume AT×d. The final number of intracellular particles per mm2cell
culture area (NA) was then estimated by:
ð2Þ
NAðNanoparticles=Transwell membraneÞ ¼ NV=SV
ð3Þ
Theintracellularparticlenumberper areawasestimated fortheupper
and for the basal side of the transwell membrane. To achieve the total
number of intracellular particles in the co-culture, the numbers at
both transwell sides were calculated. The translocation rate from the
upper to the basal side of the transwell membrane was determined by
comparing the number of intracellular particles at the basal side with
the total intracellular number.
Statistics
The statistical analyzes were carried out with the commercial
statistical package SigmaSTAT 3.5 (Systat Software Inc., Erkrath,
Germany). Due to the small sample sizes (n=4–6), non-parametric
tests were used. Kruskal–Wallis One Way Analysis of Variance
(ANOVA) on Ranks was performed if more than two groups were
compared. Multiple comparisons or multiple comparisons versus
negative control were performed with Dunn's method and pairwise
comparison with Tukey test. Differences were considered significant
at pb0.05. Results are presented as mean±standard deviation.
Outliers were determined by the Nalimov test at a significance level
of pb0.05 and excluded from the studies.
Results
Particle exposure
The particle exposure was performed at the air–liquid interface of
the triple cell co-cultures. Using 1 mL of stock solution (40 μg/mL) a
deposition of 61 ng/cm2±5.5 ng/cm2was obtained, as determined
by gamma spectroscopy, and with the 10 fold higher concentration of
the stock solution a deposition of 561 ng/cm2±48.5 ng/cm2was
reached. This corresponds to a deposition efficiency of 61% and 55%,
respectively. The quality of the particle suspension prior to the
Fig. 1. Particle size distribution of the 15 nm colloidal gold particles prior to exposure (A) and after exposure (B). Size distribution prior to exposure was measured by dynamic light
scattering. The pronounced peak near 15 nm indicates that mainly single particles were present for the 1×(A) and the 10×(A') gold concentration. A distribution of the particles
after deposition in the ALICE onto TEM grids shows a uniform particle distribution of the 1×(B) and the 10×(B') gold particle deposition (scale bar=500 nm).
59
C. Brandenberger et al. / Toxicology and Applied Pharmacology 242 (2010) 56–65

Page 5

exposure was assessed by evaluating the particle size distribution
using dynamic light scattering (HPPS 5001, Malvern Instruments Ltd,
Worcestershire, UK) (Figs. 1A and A') and the state of agglomeration
of the deposited particles was analyzedwith TEM by particleexposure
onto TEM grids (Figs. 1B and B'). Fig. 1A shows that the suspended
gold particles have a dominant mode at 15 nm, which is consistent
with single particles or small agglomerates, and a small fraction of
larger agglomerates (100 nm–1000 nm). Since the light scattering
signal from larger particles is stronger than from smaller particles,
Figs. 1A and A' exaggerate the prevalence of larger agglomerates.
Similarly, it can be seen from Figs. 1B and B' that the state of
agglomeration on the bottom plate of the exposure chamber (location
of the cells) is also low. Hence, both analyzes indicate that single
particles and small agglomerates were the dominant fraction of those
particles reaching the cells.
Quantification of the cells of the triple cell co-culture system
The number of MDM, MDDC and EC was evaluated for each
experiment (n=6), since the number and quality of primary
monocytes may vary and influence the response of the triple cell
co-cultures to particles and LPS. Fig. 2A shows the epithelial cells
(blue)grownon the upperside of thetranswellmembrane withMDM
on top (red) and Fig. 2B shows a MDDC (yellow) attached to the basal
side of the transwell. The cell numbers of all three cell types were
107±42 MDM/mm2, 81±19 MDDC/mm2and 4382±524 EC/mm2.
No correlation between high and low response towards LPS and the
amount of MDM and MDDC per mm2was observed.
Quantitative particle uptake
In order to assess cellular effects upon exposure to particles, it is
important to know if the particles can enter the cells or whether they
stay attached to the cell membrane. The number of intracellular
particles per mm2cell culture was evaluated by stereological analysis
of TEM micrographs at 4 h and 24 h after exposure, resulting in
intracellular numbers of particles per mm2of 10.7×106±4.0×106at
4 h and 11.2×106±7.3×106at 24 h for the lower and 29.2×106±
20.3×106at 4 h and 35.4±15.2×106at 24 h for the higher
concentration. As shown in Fig. 3A, the particle uptake was about
60% (57.8% at 4 h and 60.4% at 24 h) of the exposed amount for the
lower concentration, whereas only about 20% (17.6% at 4 h and 21.3%
at 24 h) of the exposed particles were taken up at the higher
concentration. Comparison between the number of particles at the
upper side and at the basal side of the triple cell co-culture shows that
Fig. 2. Laser scanning micrograph of the triple cell co-culture. The macrophages (red) and the epithelial cells (blue) are located at the upper side of the transwell membrane (A, scale
bar=5μm),whereasthedendriticcells(yellow)arelocatedatthebasalside(B,scalebar=10μm).Anevaluationofthetotalnumbersofcellspermm2resultedin4382±524epithelial
cells, 107±42 macrophages and 81±19 dendritic cells.
Fig. 3. Particle uptake (A) and translocation (B) in the triple cell co-culture system. The
particles have been taken up by the cells and were translocated to the basal side. The
particle uptake (A) of both exposure concentrations and incubation times (white and
grey bars) has been evaluated by means of design-based stereology and has been
compared to the total number of particles deposited (black bar). The degree of uptake
was 17.6±10.7% at 4 h (white bar) and 21.3±7.3% at 24 h (grey bar) for the high
exposure concentration and 57.8±19.6% at 4 h (white bar) and 60.4±42.7% at 24 h
(grey bar) for the low exposure concentration. The translocation rate (B) from the
upper transwell membrane side with epithelial cells and macrophages towards the
basal side with dendritic cells is 5.2±4.8% at 4 h (white bar) and 5.2±5.6% at 24 h
(grey bar) for the high exposure concentration and 0.5±0.3% at 4 h (white bar) and
3.9±3.9% at 24 h (grey bar) for the low exposure concentration.
60
C. Brandenberger et al. / Toxicology and Applied Pharmacology 242 (2010) 56–65

Page 6

between 0.5% and 5.2% of the particles were translocated from the
upper to the basal side and that particle transport occurred at all time
pointsandexposureconcentrations(Fig.3B).However,therewasonly
0.5±0.3% translocation rate after 4 h incubation time with the lower
particle dose but 3.9±3.9% after 24 h, although due to high standard
deviation no significant difference could be observed, neither for the
higher exposure dose (5.2±4.8% at 4 h and 5.2±5.6% at 24 h).
Gold NPs were found in all three cell types; however a
quantification of particles per cell type was not possible because of
the absence of convenient cell markers at an electron microscopic
level. Particles were mostly found in intracellular vesicles as shown in
Fig. 4 in an A549 cell (A) and a dendritic cell (B). When counting the
number of particles per cell, the largest diameter of single particles
and particle agglomerates was also measured. The relative frequency
of the agglomerate size at both time points and exposure concentra-
tions is shown in Fig. 5. There was a tendency for more single particles
in the cells after exposure to the lower concentration. For both
exposure concentrations, an increase of agglomerates, larger than
100 nm in diameter, was observed between 4 h and 24 h.
Cellular responses
The cellular response induced by exposure to the two gold NP
concentrations (1×=61 ng/cm2and 10×=561 ng/cm2gold
particles) was analyzed after post-exposure incubation times of 4 h
and24 h.The pure buffer solution (10 mMaqueouscitratesolution)of
the particle suspension without the particles was used as a negative
control. In some experiments a pre-stimulation with LPS (1 μg/mL)
was performed 2 h prior to particle exposure to study particle effects
under inflammatory conditions. Induction of mRNA (Fig. 6) of three
(pro)-inflammatory markers, namely TNFα, IL-8 and inducible nitric
oxide synthase (iNOS), as well as of the oxidative stress markers
hemeoxygenase-1 (HO-1) and superoxide dismutase 2 (SOD2) were
measured. In addition, the protein expression of pro- and anti-
inflammatory cytokines, such as IL-1β, IL-2, IL-4, IL-6, IL-8, IL-10, GM-
Fig. 4. TEM micrograph of intracellular particles in an epithelial cell (A) at the upper side and dendritic cell (B) at the basal side of the transwell membrane (scale bar=2 μm). Figures
A' and B' show a higher magnification of the black marked box of the left pictures (scale bar=500 nm). The arrows are pointing towards the particles. The particles were mainly
localized in vesicles.
Fig. 5. Relative frequency of the mean agglomerate diameter after 4 h (A) and 24 h (B)
incubation times as determined by TEM analysis. At both incubation time points, there
are relatively more single particles present at the low exposure concentration
compared to the high exposure concentration. However, when comparing the two
time points (A vs. B), a shift towards larger agglomerates of more than 100 nm in
diameter becomes visible for both exposure concentrations.
61
C. Brandenberger et al. / Toxicology and Applied Pharmacology 242 (2010) 56–65

Page 7

CSF, TNFα, INFγ were analyzed after 24 h incubation time with a
multiplex ELISA method (Fig. 7).
Inflammatory markers showed a significant induction of mRNA
due to LPS at 4 h incubationtime (Fig. 6A). This increasewas no longer
observable after 24 h incubation. No significant effect due to particle
exposure was revealed, and also not in an inflammatory stimulated
environment (pre-stimulation with LPS). The oxidative stress marker
SOD2 was also significantly induced by LPS at 4 h, but HO-1 did not
react upon LPS exposure (Fig. 6B). No adverse particle effects were
measured at any time point and concentration and no synergistic
effects could be observed in combination with LPS stimulation. The
same findings were observed for inflammatory cytokine release
measured by ELISA (Fig. 7).
Discussion
In this study the uptake and possible adverse effects of 15 nm gold
particles were analyzed in a triple cell co-culture model exposed at
the air–liquid interface. The particles were found to enter the cells in a
concentration dependent way and were translocated through the
epithelium to dendritic cells located at the basal side. No significant
induction of oxidative stress or inflammatoryresponse due to the gold
NPs was observed, neither any synergistic or suppressive effect in an
inflammatory stimulated system.
The two gold concentrations applied via the nebulizer in these
experimentsresulted in a deposition of 61 ng/cm2and561 ng/cm2on
thecells.Afurtherincreasein theexposureconcentrationwaslimited,
since multiple exposures would increase the liquid layer (14 μm for
1 mL of colloidal gold) and hence mitigate or abolish the air–liquid
interface effect. Compared to submerged exposure conditions these
concentrations might seem to be low. However, it has to be
considered, that the dynamics of gold NPs with a diameter ≤40 nm
in suspensions is driven by diffusion and not by sedimentation
(according to equations published by Limbach et al. (2005) which
drastically reduces the number of particles deposited on the cells
under submerged conditions. It is therefore considered that the
exposure concentrations used in the current experiments are in a
comparable range with other studies and any absent cellular response
can not be related to a too low NP exposure concentration. Table 2
shows a comparison between different gold NP concentrations and
their effects, used in different published experiments. Furthermore,
the dispersal of mainly single particles onto the cells was confirmed:
the measurement of the particle size distribution prior to exposure
showed only low agglomeration in the suspension (Fig. 1A) which
was also qualitatively observed after particle deposition on TEM grids
(Fig. 1B).
In order to evaluate the total particle uptake and the translocation
rate towards the basal side where the dendritic cells were localized,
Fig. 6. Gene expression (fold-induction) of inflammatory markers (A) and oxidative stress markers (B) quantified by means of real-time PCR. All results are expressed relative to the
negative controls (neg) at 4 h and 24 h. There is no significant increase of pro-inflammatory and oxidative stress markers. LPS leads to a significant induction (⁎pb0.05) of mRNA
after 4 h incubation time, which has vanished after 24 h. No significant synergistic or suppressive effects of particles were observed at any time point.
62
C. Brandenberger et al. / Toxicology and Applied Pharmacology 242 (2010) 56–65

Page 8

the uptake of particles was estimated per mm2cell layer of the upper
and the basal side of the triple cell co-culture. The results show a 60%
particle uptake rate for the lower exposure concentration but only a
20% uptake rate for the higher exposure concentrations, which might
be due to a concentration dependent saturation in particle internal-
ization. Between 4 h and 24 h there was only a marginal increase in
intracellular particle numbers for both exposure concentrations,
inferring that either most particle uptake occurred during the first
four hours or an established equilibrium between endo- and
exocytosis. The translocation through the transwell membrane
could be confirmed and particles could be identified in dendritic
cells (Fig. 4B). It is postulated that particles are either sampled by
macrophages and delivered to dendritic cells, or that the dendritic
cells themselves extend processes between epithelial cells through
the tight junctions to collect the particles (Blank et al., 2007). While
the translocation rate remainedequal at 4 h and24 h after exposureto
the higher concentration, almost no particles could be detected at the
basal side 4 h after exposure to the lower concentration. This effect
had vanished after 24 h, which probably means that the translocation
process at lower particle dose was slower. A further increase in
agglomerate size with time was observed (Fig. 5), which may be
explained by particle agglomeration within the cells with time. Since
particles were observed in vesicles, it is very likely that this effect
occurs due to intracellular vesicle trafficking and fusion after particle
uptake.
An evaluation of quantitative particle localization within the
different cells could not be performed, since conventional TEM does
not allow an exact differentiation of the cell types. The comparison of
the average cell number per mm2and the number of intracellular
particles per mm2shows that approximately 2400 and 7000 particles
per cell were found after exposure to low and high NP concentration,
respectively. The correlative particle distribution between the three
cell types was previously described with fluorescent polystyrene NPs
(Rothen-Rutishauser et al., 2007), showing a preferential uptake by
MDM and MDDC.
In the present study, the potential cytotoxic effect of gold particles
was investigated by measuring gene expression of inflammatory
cytokines and oxidative stress markers at 4 h and 24 h and protein
induction of pro- and anti-inflammatory cytokines at 24 h. These two
time points were chosen after pre-experimental studies with different
incubation times. Effects due to the exposure method have been
excluded in previous studies (Schmid et al., 2009) and a internal
negative control with cells which were not exposed in the ALICE, were
added to the study for each experiment, to factor out side effects (data
not shown). The results clearly showed, that no oxidative or
inflammatory effects due to particle exposure could be observed in
Fig. 7. Protein expression of pro- and anti-inflammatory cytokines. No significant effect on inflammatory cytokine release can be approved due to the exposure to gold NPs. The pre-
stimulation with LPS causes a distinct increase in pro- and anti-inflammatory cytokine levels. However, standard deviations of the experiments are too high to assure a significant
induction. Synergistic or suppressive effects of gold NPs with LPS could not be found.
Table 2
Published results on size, concentration and effects in vitro for gold NPs in suspension.
Particle size
μM AuEffects (cell line)
Reference
1.4 nm 0.2–2.5⁎
Reduced cell viability after
24 h, with IC50values
dependent on cell lines
(0.3–2.3 μM) (various cell
lines e.g. HeLa, Hek-12,
MV3)
Particles below 2 nm are
cytotoxic and cause
apoptosis within 12 h;
1.4 nm particles show
the lowest IC50value
(40–60 μM) (cell lines:
HeLa, L929 fibroblasts,
SK-Mel-28, JJ4A1)
Vesicle leakage and
reduced cell viability after
1 h in positively charged
particles only (LC50: 1 μM)
(COS-1 cell line)
No reduction in cell
viability after 72 h or
pro-inflammatory
cytokine release at 24 h
(RAW264.7 murine
macrophage-like cells)
Apoptosis and
morphological
deformation at 2 days–
6 days (CF-31 human
dermal fibroblast cells)
Nitric oxygen release in
fresh plasma serum after
5 min (fresh human
plasma serum)
No reduction in cell
viability after 72 h (K562
human leukemia cell)
No changes in gene
expression after 6 h (HeLa
cervical cancer cells)
No induction of oxidative
stress markers and
inflammatory cytokines
(human triple cell
co-culture including
primary cells)
(Tsoli et al., 2005)
0.8, 1.2, 1.4,
1.8, 15 nm
10–6300⁎
(Pan et al., 2007)
2 nm positive
and negative
charged
0.4–7.4⁎
(Goodman et al., 2004)
4 nm 10–100⁎
(Shukla et al., 2005)
13 nm500–4000
(100–800⁎
μg/mL Au)
(Pernodet et al., 2006)
13 nm10–80⁎
(Jia et al., 2009)
18 nm with
coatings
0.1–250⁎
(Connor et al., 2005)
18 nm 48–480⁎
(Khan et al., 2007)
15 nm ALICE200–2000
0.3–2.3 μg/
6-well
Current study
Except for the study by Jia et al. (2009) all studies were completed with cell culture
lines. The gold particle concentration values given in the publications are indicated ⁎,
other values were approximately calculated by the molar mass of gold. Values of 50%
lethality (LC50) and 50% inhibition (IC50) and were given in some studies.
63
C. Brandenberger et al. / Toxicology and Applied Pharmacology 242 (2010) 56–65

Page 9

this experimental setup for all time points and all gold exposure
concentrations. Inflammatory pre-stimulation with LPS resulted in an
mRNA induction of marker genes which are known to be regulated by
inflammatory stimuli. The oxidative stress marker SOD2 also showed
an up regulation after LPS simulation whereas HO-1 did not, although
diverse studies have shown an increase of HO-1 after LPS stimulation
in macrophages (Srisook and Cha, 2004) as well as in lung tissue
(Zhan et al., 2006). We assume that the observed oxidative stress
response of SOD2 was mainly triggered by the response of epithelial
cells toward TNFα (Xu et al., 1999), which is not involved in the
induction pathway of HO-1 (up to 25 μg/mL) (Wijayanti et al., 2004).
LPS primarily acts on macrophages and dendritic cells potentially
initiating a HO-1 induction, however we assume that the reaction was
under the detection limit in our triple cell co-culture system with a
low ratio of MDM, MDDC towards ECs. The primers have been tested
previously and HO-1 induction could be measured after air–liquid
interface exposure of A549 cells on ZnO particles (personal commu-
nication with A.-G. Lenz), and therefore methodological artefacts are
excluded.
Cytokine release after 24 h post-exposure showed similar results
as mRNA induction at 4 h. All cytokines were expressed in an
increased amount after LPS stimulation, but neither the particles
themselves nor any synergistic or suppressive effects could be
observed.
Since the experiments were performed with primary cultures of
MDM and MDDC, it is not surprising that the experiments showed
quite a high standard deviation, although experiments were repeated
six times. A correlation between the number of MDM and MDDC per
transwell and the immune response due to LPS could be excluded by
comparing high response and low response experiments with their
respective number of MDM and MDDC in the triple cell co-culture
system.
It can be summarized that although the gold particles enter the
cells of the triple cell co-culture model, no adverse effects could be
observed, not even in an inflammation stimulating environment.
Comparing the current results with those from other studies
(summarized in Table 2), we conclude, that non functionalized
gold particles with a size of 13–20 nm in diameter do not cause
acute adverse effects. However, toxic effects have been shown for
smaller particles with a diameter of 2 nm or less (Pan et al., 2007;
Tsoli et al., 2005). This size effect could be explained by an
increased catalytic activity for particles with less than 3–5 nm as
described by Hvolbaek et al. (2007) or by potential interaction with
the DNA for particles equal or smaller than 1.4 nm (Liu et al., 2003).
A short term effect of NO release with 13 nm particles as observed
by Jia et al. (2009), could not be detected in the triple cell co-
culture system after 4 h of incubation when measuring mRNA
expression of iNOS. The question remains as to whether morpho-
logical changes would also occur after 6 days of incubation at
similar exposure concentrations as described by Pernodet et al.
(2006). It is not possible to answer this question with our triple cell
co-culture system since it can not be maintained for more than
48 h. Therefore, long term in vivo studies are needed to entirely
exclude adverse effects of gold NPs before applying them in
nanomedicine.
Acknowledgments
The authors would like to thank Mr. Bukalis and Mrs. Alber for
performing the neutron-activated gamma spectroscopic determina-
tion of the mass of the gold nanoparticles, Alexander Wenk, Helga
Hinze-Heyn, Barbara Krieger, Andrea Stokes and Mohamed Ouanella
for their excellent technical assistance and Kirsten Dobson for proof
reading the manuscript.
This study was supported by grants of the AnimalFreeResearch
Foundation, the Doerenkamp-Zbinden-Foundation, the Deutsche
Forschungs Gemeinschaft (DFG), Swiss National Science Foundation
(3100A0_118420) and the Helmholtz Association.
References
Baddeley, A.J., Gundersen, H.J.G., Cruz-Orive, L.M., 1986. Estimation of surface-area from
vertical sections. J. Microsc. 142, 259–276.
Blank, F., Rothen-Rutishauser, B., Gehr, P., 2007. Dendritic cells and macrophages form a
transepithelial network against foreign particulate antigens. Am. J. Respir. Cell Mol.
Biol. 36, 669–677.
Brandenberger, C., Rothen-Rutishauser, B., Blank, F., Gehr, P., Muhlfeld, C., 2009.
Particles induce apical plasma membrane enlargement in epithelial lung cell line
depending on particle surface area dose. Respir. Res. 10, 22.
Chan, R.C.F., Wang, M.Y., Li, N., Yanagawa, Y., Onoe, K., Lee, J.J., Nel, A.E., 2006. Pro-
oxidative diesel exhaust particle chemicals inhibit LPS-induced dendritic cell
responses involved in T-helper differentiation. J. Allergy Clin. Immunol. 118,
455–465.
Chen, Y.H., Tsai, C.Y., Huang, P.Y., Chang, M.Y., Cheng, P.C., Chou, C.H., Chen, D.H., Wang,
C.R., Shiau, A.L., Wu, C.L., 2007. Methotrexate conjugated to gold nanoparticles
inhibits tumor growth in a syngeneic lung tumor model. Mol. Pharm. 4, 713–722.
Cho, W.S., Cho, M., Jeong, J., Choi, M., Cho, H.Y., Han, B.S., Kim, S.H., Kim, H.O., Lim, Y.T.,
Chung, B.H., Jeong, J., 2009. Acute toxicity and pharmacokinetics of 13 nm-sized
PEG-coated gold nanoparticles. Toxicol. Appl. Pharmacol. 236, 16–24.
Connor, E.E., Mwamuka, J., Gole, A., Murphy, C.J., Wyatt, M.D., 2005. Gold
nanoparticles are taken up by human cells but do not cause acute cytotoxicity.
Small 1, 325–327.
de Haar, C., Hassing, I., Bol, M., Bleumink, R., Pieters, R., 2006. Ultrafine but not fine
particulate matter causes airway inflammation and allergic airway sensitization to
co-administered antigen in mice. Clin. Exp. Allergy 36, 1469–1479.
Dhar, S., Reddy, E.M., Shiras, A., Pokharkar, V., Prasad, B.L., 2008. Natural gum reduced/
stabilized gold nanoparticles for drug delivery formulations. Chemistry 14,
10244–10250.
Goodman, C.M., McCusker, C.D., Yilmaz, T., Rotello, V.M., 2004. Toxicity of gold
nanoparticles functionalized with cationic and anionic side chains. Bioconjug.
Chem. 15, 897–900.
Griffiths, G., 1993. Fine Structure Immunocytochemistry. Springer-Verlag, Berlin
Heidelberg.
Herzog, E., Byrne, H.J., Casey, A., Davoren, M., Lenz, A.G., Maier, K.L., Duschl, A., Oostingh,
G.J., 2009a. SWCNT suppress inflammatory mediator responses in human lung
epithelium in vitro. Toxicol. Appl. Pharmacol. 234, 378–390.
Herzog, E., Byrne, H.J., Davoren, M., Casey, A., Duschl, A., Oostingh, G.J., 2009b.
Dispersion medium modulates oxidative stress response of human lung epithelial
cells upon exposure to carbon nanomaterial samples. Toxicol. Appl. Pharmacol.
236, 276–281.
Hofer, T.P., Bitterle, E., Beck-Speier, I., Maier, K.L., Frankenberger, M., Heyder, J.,
Ziegler-Heitbrock, L., 2004. Diesel exhaust particles increase LPS-stimulated COX-
2 expression and PGE2 production in human monocytes. J. Leukoc. Biol. 75,
856–864.
Holt, P.G., Schon-Hegrad, M.A., McMenamin, P.G., 1990. Dendritic cells in the
respiratory tract. Int. Rev. Immunol. 6, 139–149.
Hussain, S.M., Hess, K.L., Gearhart, J.M., Geiss, K.T., Schlager, J.J., 2005. In vitro toxicity of
nanoparticles in BRL 3A rat liver cells. Toxicol. In Vitro 19, 975–983.
Hvolbaek, B., Janssens, T.V.W., Clausen, B.S., Falsig, H., Christensen, C.H., Norskov, J.K.,
2007. Catalytic activity of Au nanoparticles. Nano Today 2, 14–18.
Jain, P.K., Huang, X., El-Sayed, I.H., El-Sayed, M.A., 2008. Noble metals on the nanoscale:
optical and photothermal properties and some applications in imaging, sensing,
biology, and medicine. Acc. Chem. Res. 41, 1578–1586.
Jia, H.Y., Liu, Y., Zhang, X.J., Han, L., Du, L.B., Tian, Q., Xu, Y.C., 2009. Potential oxidative
stress of gold nanoparticles by induced-NO releasing in serum. J. Am. Chem. Soc.
131, 40–41.
Joshi, H.M., Bhumkar, D.R., Joshi, K., Pokharkar, V., Sastry, M., 2006. Gold nanoparticles
as carriers for efficient transmucosal insulin delivery. Langmuir 22, 300–305.
Khaing Oo, M.K., Yang, X., Du, H., Wang, H., 2008. 5-aminolevulinic acid-conjugated
gold nanoparticles for photodynamic therapy of cancer. Nanomedicine 3,
777–786.
Khan, J.A., Pillai, B., Das, T.K., Singh, Y., Maiti, S., 2007. Molecular effects of uptake of gold
nanoparticles in HeLa cells. Chembiochemical 8, 1237–1240.
Kreyling, W.G., Semmler-Behnke, M., Moller, W., 2006. Ultrafine particle-lung
interactions: does size matter? J. Aerosol Med. 19, 74–83.
Kwon, Y.M., Xia, Z., Glyn-Jones, S., Beard, D., Gill, H.S., Murray, D.W., 2009. Dose-
dependent cytotoxicity of clinically relevant cobalt nanoparticles and ions on
macrophages in vitro. Biomed. Mater. 4.
Li, J.L., Wang, L., Liu, X.Y., Zhang, Z.P., Guo, H.C., Liu, W.M., Tang, S.H., 2009. In vitro
cancer cell imaging and therapy using transferrin-conjugated gold nanoparticles.
Cancer Lett. 274, 319–326.
Lieber, M., Smith, B., Szakal, A., Nelsonrees, W., Todaro, G., 1976. Continuous tumor-cell
line from a human lung carcinoma with properties of type-Il alveolar epithelial
cells. Int. J. Cancer 17, 62–70.
Limbach, L.K., Li, Y., Grass, R.N., Brunner, T.J., Hintermann, M.A., Muller, M., Gunther, D.,
Stark, W.J., 2005. Oxide nanoparticle uptake in human lung fibroblasts: effects of
particle size, agglomeration, and diffusion at low concentrations. Environ. Sci.
Technol. 39, 9370–9376.
Liu, Y., Meyer-Zaika, W., Franzka, S., Schmid, G., Tsoli, M., Kuhn, H., 2003. Gold-cluster
degradation by the transition of B-DNA into A-DNA and the formation of
nanowires. Angew. Chem., Int. Ed. Engl. 42, 2853–2857.
64
C. Brandenberger et al. / Toxicology and Applied Pharmacology 242 (2010) 56–65

Page 10

Mayhew, T.M., 2008. Taking tissue samples from the placenta: an illustration of
principles and strategies. Placenta 29, 1–14.
Mühlfeld, C., Rothen-Rutishauser, B., Vanhecke, D., Blank, F., Gehr, P., Ochs, M., 2007.
Visualization and quantitative analysis of nanoparticles in the respiratory tract by
transmission electron microscopy. Part Fibre Toxicol. 4, 11.
Nel, A., Xia, T., Madler, L., Li, N., 2006. Toxic potential of materials at the nanolevel.
Science 311, 622–627.
Oberdörster, G., Oberdörster, E., Oberdörster, J., 2005. Nanotoxicology: An emerging
discipline evolving from studies of ultrafine particles. Environ. Health Perspect.
113, 823–839.
Olofsson, L., Rindzevicius, T., Pfeiffer, I., Kall, M., Hook, F., 2003. Surface-based gold-
nanoparticle sensor for specific and quantitative DNA hybridization detection.
Langmuir 19, 10414–10419.
Pan, Y., Neuss, S., Leifert, A., Fischler, M., Wen, F., Simon, U., Schmid, G., Brandau, W.,
Jahnen-Dechent, W., 2007. Size-dependent cytotoxicity of gold nanoparticles.
Small 3, 1941–1949.
Pernodet, N., Fang, X., Sun, Y., Bakhtina, A., Ramakrishnan, A., Sokolov, J., Ulman, A.,
Rafailovich, M., 2006. Adverse effects of citrate/gold nanoparticles on human
dermal fibroblasts. Small 2, 766–773.
Rosi, N.L., Giljohann, D.A., Thaxton, C.S., Lytton-Jean, A.K.R., Han, M.S., Mirkin, C.A., 2006.
Oligonucleotide-modified gold nanoparticles for intracellular gene regulation.
Science 312, 1027–1030.
Rothen-Rutishauser, B.M., Kiama, S.G., Gehr, P., 2005. A three-dimensional cellular
model of the human respiratory tract to study the interaction with particles. Am. J.
Respir. Cell Mol. Biol. 32, 281–289.
Rothen-Rutishauser, B., Muhlfeld, C., Blank, F., Musso, C., Gehr, P., 2007. Translocation of
particles and inflammatory responses after exposure to fine particles and
nanoparticles in an epithelial airway model. Part Fibre Toxicol. 4, 9.
Sallusto, F., Cella, M., Danieli, C., Lanzavecchia, A., 1995. Dendritic cells use
macropinocytosis and the mannose receptor to concentrate macromolecules in
the major histocompatibility complex class-II compartment—down-regulation by
cytokines and bacterial products. J. Exp. Med. 182, 389–400.
Schmid, O., Ferron, G.A., Lentner, B., Lenz, A.G., Brandenberger, C., Rothen-Rutishauser,
B., Gehr, P., Kreyling, W., Schulz, H., Karg, E., 2009. Development of a quantitative
and easy-to-use system for thin-film deposition of nanoparticle suspensions onto
cells at the air–liquid interface (ALI). J. Aerosol Med. Pulm. Drug Deliv. 22, 199
(Conference Abstract).
Selvakannan, P.R., Mandal, S., Phadtare, S., Gole, A., Pasricha, R., Adyanthaya, S.D., Sastry,
M., 2004. Water-dispersible tryptophan-protected gold nanoparticles prepared by
the spontaneous reduction of aqueous chloroaurate ions by the amino acid.
J. Colloid Interface Sci. 269, 97–102.
Shukla, R., Bansal, V., Chaudhary, M., Basu, A., Bhonde, R.R., Sastry, M., 2005.
Biocompatibility of gold nanoparticles and their endocytotic fate inside the cellular
compartment: A microscopic overview. Langmuir 21, 10644–10654.
Srisook, K., Cha, Y.N., 2004. Biphasic induction of heme oxygenase-1 expression in
macrophages stimulated with lipopolysaccharide. Biochem. Pharmacol. 68,
1709–1720.
Small, J.V., 1968. Measurement of section thickness. Fourth Eur. Conf. Electron Microsc.
609–610.
Stoeger, T., Reinhard, C., Takenaka, S., Schroeppel, A., Karg, E., Ritter, B., Heyder, J.,
Schulz, H., 2006. Instillation of six different ultrafine carbon particles indicates a
surface area threshold dose for acute lung inflammation in mice. Environ. Health
Perspect. 114, 328–333.
Teeguarden, J.G., Hinderliter, P.M., Orr, G., Thrall, B.D., Pounds, J.G., 2007. Particokinetics
in vitro: dosimetry considerations for in vitro nanoparticle toxicity assessments.
Toxicol. Sci. 95, 300–312.
Tsoli, M., Kuhn, H., Brandau, W., Esche, H., Schmid, G., 2005. Cellular uptake and toxicity
of Au55 clusters. Small 1, 841–844.
Wangoo, N., Bhasin, K.K., Mehta, S.K., Suri, C.R., 2008. Synthesis and capping of water-
dispersed gold nanoparticles by an amino acid: bioconjugation and binding studies.
J. Colloid Interface Sci. 323, 247–254.
Weibel, E.R., Hsia, C.C., Ochs, M., 2007. How much is there really? Why stereology is
essential in lung morphometry. J. Appl. Physiol. 102, 459–467.
Wijayanti, N., Huber, S., Samoylenko, A., Kietzmann, T., Immenschuh, S., 2004. Role of
NF-kB and p38 MAP kinase signaling pathways in the lipopolysaccharide-
dependent activation of heme oxygenase-1 gene expression. Antioxid. Redox
Signal. 6, 802–810.
Xu, Y., Kiningham, K.K., Devalaraja, M.N., Yeh, C.C., Majima, H., Kasarskis, E.J., St Clair,
D.K., 1999. An intronic NF-kappaB element is essential for induction of the human
manganese superoxide dismutase gene by tumor necrosis factor-alpha and
interleukin-1beta. DNA Cell Biol. 18, 709–722.
Zhan, L.Y., Xia, Z.Y., Chen, C., Wang, X.Y. 2006. Effect of radix paeoniae rubra on the
expression of HO-1 and iNOS in rats with endotoxin-induced acute lung injury, 9,
181–186.
65
C. Brandenberger et al. / Toxicology and Applied Pharmacology 242 (2010) 56–65