BBRC

Biochemical and Biophysical Research Communications 332 (2005) 109116

www.elsevier.com/locate/ybbrc

Detection of nanometer-sized particles in living cells using

modern uorescence uctuation methods
Michael Edetsberger a,*,1, Erwin Gaubitzer a,1, Eva Valic b,
Elisabeth Waigmann c, Gottfried Kohler a
a

Max F. Perutz Laboratories, Department of Chemistry, University of Vienna, Campus-Vienna-Biocenter 5, A-1030 Vienna, Austria
b
Allgemeine Unfallversicherungs-Anstalt, AUVA, Aldalbert-Stifterstrasse 65, A-1201 Vienna, Austria
c
Max F. Perutz Laboratories, University Departments at the Vienna Biocenter, Department of Medical Biochemistry,
Medical University of Vienna, Dr. Bohrgasse 9, A-1030 Vienna, Austria
Received 13 April 2005
Available online 27 April 2005

Abstract
Nanosized materials are increasingly used in medicine and biotechnology but originate also from various aerosol sources. A
detailed understanding of their interaction with cells is a prerequisite for specic applications and appraisal of hazardous eects.
Fluorescence uctuation methods are applied to follow the time-course of the translocation and distribution of uorescent
20 nm polystyrene nanoparticles with negative surface charges in HeLa cells under almost physiological conditions. The experimental results demonstrate that singular particles enter the cell without signicant contribution by endocytotic mechanisms and are distributed within the cytoplasm. Subsequently aggregation is observed, which can be blocked by cytotoxins, like Genistein and
Cytochalasin B, interfering with cellular uptake processes. The observed non-active uptake is due to non-specic interactions with
the cell surface and could be responsible for distribution of nanometer-sized materials in tissue.

2005 Elsevier Inc. All rights reserved.
Keywords: Nanoparticles; Laser scanning microscope; Fluorescence correlation spectroscopy; Photon counting histogramming; Auto correlation
function; Fluorescence intensity distribution analysis

Nanoparticles are small minute parts appearing dispersed in uid or gaseous media. Common size limits
are: particles up to 50 nm in diameter are called nanoparticles, and between 50 and 100 nm ultrane particles
[1].
Nanometer sized particles like fullerenes [2], quantum
dots [3], particulate drug delivery systems, and Dendrosomes as DNA carrier [4] are discussed due to their novel and unique properties for application in life sciences
and medical treatment. Nevertheless, possible negative

Corresponding author. Fax: +43 1 4277 9522.

E-mail addresses: [email protected] (M. Edetsberger), [email protected] (E. Gaubitzer).
1
These two authors contributed equally.
0006-291X/$ - see front matter
2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.bbrc.2005.04.100

health eects became an issue in environmental sciences

as nanoparticulate matter appears also in ambient air as
a result of a variety of technological processes. Especially the PM2.52 fraction of these emissions, which consist mainly of carbon associated with metals, oxygen,
and organic matter, is assumed to threaten public health
[5] and even heritable eects are possible [1,6].
Nanoparticles and ultrane particles are also an issue
in job safety and therefore global research activities are
focused on dierent kinds of aerosol sources, such as
coal y ash [7] and diesel exhaust particles [8] from modern combustion and fabrication processes. Exposed
workers (e.g., road- or tunnel construction-sites) carry
2

Denotes particle matter (PM) smaller than 2.5 lm in diameter.

110

M. Edetsberger et al. / Biochemical and Biophysical Research Communications 332 (2005) 109116

an incalculable risk to contract physiological sequels due

to the exposure.
Cells of the alimentary canal, the urogenital tract, the
lung, and the skin are exposed to the ambient environment. Whereas the skin builds up a physical barrier,
the cells of the lung, dierent varieties of the epithelium,
and the intestinal tract deal with matter transport. In
general, the lung of mammals can be regarded as a lter
system. In the upper respiratory tract, bigger particles
deposit on cilia and are removed by sputum. Dust particles from technical processes might be a hundred times
smaller than natural dust and are therefore able to penetrate into the bronchioles and the alveolar region.
Within the alveolae the air is virtually stagnant and
therefore diusion becomes decisive for the further
translocation into the bloodstream and the lymphatic
system. Studies with intravenously injected ultrane particles have shown that the liver is the major organ of
accumulation [9] but they were also found in tumor cells
and tumor-associated macrophages of rodents [10] and
carbon nanoparticles were even found in brains of rats
[11].
Particles up to 200 nm in diameter are able to enter
cells without an active uptake process. This unspecic uptake was shown for dierent cell lines using confocal laser
scanning microscopy (LSM), transmission electron
microscopy, and electron energy-loss spectroscopy
[12,13]. Such techniques provide information about the
spatial distribution, shape, and size of particles within
xed or at least pre-experimentally treated cells. In contrast, the rather new combination of confocal microscopy
based uorescence uctuation methods like uorescence
correlation spectroscopy (FCS) [14] and uorescence
intensity distribution analysis (FIDA) [15] allows us to
get informations about size and dynamic properties of
particles in the cytoplasm also of untreated cells.
FCS uses the auto correlation function (ACF) of
uorescent particles diusing through the focus volume
(10
18 m3=1 fL) to extract their number and diusion
coecient. FIDA uses the photon counting histogram
(PCH) [16] to extract molecular brightness which can
be seen as a direct parameter for the size of particles
[17]. We applied these methods to follow the time-course
of the uptake of uorescent particles into living cells.
The contribution of endocytotic processes to the
mechanism of their translocation was analyzed using
two cytotoxins. The tyrosine kinase inhibitor Genistein,
which is known to inhibit endocytotic mechanisms,
especially caveolae internalization [18], the recruitment
of dierent receptors in clathrin-coated pits [19], and
the routing from early endosomes in lysosomes [20],
and the cell permeable fungal toxin Cytochalasin B,
which interacts with actin polymerization [21] and disrupts various cell processes.
The main aim of this study was to detect uorescent
particles in living cells. We discriminated particles and

dierent aggregates within the cell and determined their

dynamic properties and the time-course of their appearance in the cytoplasm. LSM was used to complete the
measured data.
Materials and methods
Cell culture. Cell culture reagents were obtained from SigmaAldrich (Austria) and Euroclone (Austria), Genistein, and Cytochalasin
B were from SigmaAldrich (Austria).
HeLa cells-human cervix carcinoma cells. HeLa cells (OHIO cell
line ATCC, provided by the Medical University Vienna) were routinely
cultured in DMEM (high glucose
4500 mg glucose) containing
4.5 mM L-glutamine, 44 mM Na-bicarbonate, 100 U/ml penicillin,
100 lg/ml streptomycin, 0.9 mM Na-pyruvate, and 10% fetal calf
serum in a 6% CO2 humidied atmosphere at 37 C (standard
conditions).
Preparation for LSM. HeLa cells (1 104 cells/ml) were seeded on
coverslips (Assistent, Germany) in 35 mm tissue wells (IWAKI, Japan)
and grown overnight at standard conditions. Cells were washed twice
with PBS containing 20 mM glucose (PBSG, pH 7.4) at 37 C. Then
the cells were stained with 10 lM Nile-Red for 20 min at 37 C. For
imaging the cells were incubated with 0.018%(w/w) 20 nm, carboxycoated, green uorescent polystyrene particles (Fluospheres, Molecular
Probes) in PBSG.
Preparation for FCS and PCH. Cells (3 104 cells/ml) were seeded
in 8-well LabTek chambered slides (NUNC) and grown overnight at
standard conditions. Cells were washed twice with PBSG at 37 C.
Then the cells were submerged with PBSG at 37 C. For measuring
FCS and PCH the cells were incubated with 0.005%(w/w) 20 nm,
carboxy coated, green uorescent polystyrene particles (Fluospheres,
Molecular Probes) in 0.4% trypan blue solution (SigmaAldrich) to
block signals from outside the cell [12]. Measurements started immediately after adding the particles.
Blocking of uptake (LSM, FCS, and PCH). After growing the cells
overnight the culture medium was removed by new medium (DMEM)
containing 200 mM Genistein [12] or 10 lM Cytochalasin B [22]. Cells
were incubated for 60 min under standard conditions and afterwards
washed twice with PBSG (37 C). The further treatments were performed in the same way as for untreated samples.
Setup for LSM. Live cell confocal images were performed using a
confocal laser scanning microscope (Leica SP1, Germany), a CW Ar+
laser, and an oil immersion objective (PL APO 100 1.4 oil UV).
Images for the uorescent particles were performed using the laser line
at kexc = 488 nm and detected by a photo-multiplier tube (PMT) in the
range of 500540 nm (green channel; setup 1, HV710/oset 0.39; setup
2, HV 739/oset 1.17). Images for the stained cells were performed
using the laser line at kexc = 568 nm and detected by a PMT in the
range of 580640 nm (red channel; setup 1, HV594/oset 0.39; setup 2,
HV 756/oset 1.17). Images of cells treated with Genistein (with or
without particles added) were performed with setup 1 and images of
untreated cells (particles added) were performed with setup 2. Imaging
started 20 min after adding the particles and all images were averages
of three individual scans. Images were extracted and overlays were
performed using the Leica Simulator Software (Leica, Germany) or
Adobe Photoshop 6.0.
Setup for FCS and PCH. Measurements were performed using a
ConfoCor1 (Carl Zeiss, Jena, Germany), an Ar+ laser (Lasos, Germany; k = 488 nm, 0.3 mW/cm2), a water immersion objective (Zeiss,
C-Apochromat 63, 1.2 W corr) and an Avalanche Photo-Diode
(APD, Perkin-Elmer SPCM-AQR-13-FC). Confocal pinhole diameter
was set to 35 lm and calibrated with rhodamine-6 G (R6G, diusion
constant: 2.8 10
6 cm2/s; diusion time: 0.03 ms). Fluorescence was
detected through a dichroic mirror (>510 nm) and a band-pass lter
(520570 nm).

M. Edetsberger et al. / Biochemical and Biophysical Research Communications 332 (2005) 109116
FCS analysis. Fluorescence signals were detected for 42 s with 3 s
interval and 12 s correlator scaling. The autocorrelation (G(s)) was
tted using FCS Access version 1.0.12 (Evotec, Germany) applying a
structural parameter SP 1=2  focus length=focus radius of 5.6
obtained from the measurement with R6G according to the following
formula [23]:
Gs

hdItdIt si
2

hdIti

1 X
Fi
1
q :
N i1...3 1 ssi 1 s2
SP s

Fi and si are the fraction and diusion time of component i, respectively, and N represents the number of particles in the focus element.
The term dI denotes the deviation of intensity from the temporal average. From the diusion-constant (DR6G) and the diusion-time (sR6G)
of R6G and knowing diusion times of particles (sparticles) their hydrodynamic radii (rhydrodynamic) can be obtained [23]:
rhydrodynamic

kT sparticles
:
96DR6G sR6G pg

Here k denotes the Boltzmann constant, T stands for the absolute temperature, and g represents the viscosity of the cytoplasm (2.2 cp) [24].
PCH analysis. To obtain raw data, uorescence signals were detected using a Time Measurement Histogram Accumulating Real-Time
Processor (PC Board for Time Correlated Single Photon Counting,
Time Harp 100, PicoQuant, Software version 3.0) triggered with
7.4 MHz. The histogram of counted photons is acquired by monitoring
the signal intensity during 30 s and plotting the numbers of photons
detected in a time window (t40) of 40 ls. The values were analyzed
using the generating function [25]:
"
#
X Z
cs efn
1qs t40 Br
1g dV :
3
Gn exp n
1kt40
s

Here, cs relates to the concentration and qs accounts for the specic

brightness of dierent particle species (s). The term k is a parameter
for background noise and B(r) represents the spatial brightness distribution in the focal volume (V). After calculation of G(n), the probability distribution is retrieved by an inverse Fourier transformation:
P n; t40 FFT
1 Gn; t40 :

Raw data recorded with Time Harp 100 were histogrammed and normalized by area. The resulting data were tted using a modied LevenbergMarquardt least squares algorithm. All software was written in
the Python programming language. All graphs were drawn using Origin 6.1 (Origin Lab, USA).

Results
Laser scanning microscopy
HeLa cells were imaged before and after a pre-treatment with Genistein (compilation in Fig. 1). HeLa cells,
treated with Genistein, stained with Nile-Red but not
incubated with green uorescent particles showed a clear
image in the red detection channel but a weak signal in
the green channel (Fig. 1A). Cells stained with Nile-Red
and incubated with particles showed images in both
channels (Figs. 1B and C). In the sample treated with
Genistein the diuse green uorescence of mainly isolated particles was observed, which became apparent
by the image in the green channel. Only small aggregates, which can be identied as green spots, appeared
sporadically at the very peripheric region of the cell

111

(Fig. 1B). In the untreated sample mainly isolated particles (diuse green uorescence) and very small aggregates (seen as green spots) were observed. These
species of particles were found all over the cell. Additionally very big and bright aggregates (yellow spots)
of dierent size were observed. The yellow color is a result of the overlay of the green and red images indicating
that these aggregated particles were mainly found at
stainable structures in the cytoplasm. Yellow structures
which seem to be in the nucleus are residual signals from
the section above the nucleus (Fig. 1C). Nevertheless, it
is necessary to have viable cells and a uid membrane,
because in dead cells no translocation of particles could
be observed. In that case, the particles were just attached
to the outside of the cell membrane (picture not shown).
FCS measurement
The diusion behavior of green uorescent particles
was measured in the cytoplasm of untreated HeLa cells
and HeLa cells treated with Genistein or Cytochalasin B
(Fig. 2).
The resulting autocorrelation curves were analyzed
using a multiple-component model. After adding particles to the submerged cells, the uorescence signal in
the cytoplasm increased. As there was generally a delay
of about 1 min between the addition and the rst measurement, a clear uorescence intensity was already recorded from the interior of the cell. This indicates that
the translocation of the particles into the cell occurred
without delay. The intensity was dependent on the concentration of particles, its sedimentation properties, and
the selected cell. Therefore, dierent initial signal intensities could be detected in individual measurements (representative results see Fig. 2).
Only at low intensities, i.e., low particle concentrations, the measured intensity uctuations were large enough to extract diusion parameters. A typical
measurement for untreated cells is shown in Figs. 2A1
and A2. The measurement of cells treated with Genistein
or Cytochalasin B showed already much higher initial
intensities and thus only small uctuations. At particle
concentrations frequently observed during experiments,
the uorescence intensities were above 100 kHz and correlations were often lost. Analysis of the autocorrelation
curves at early time-points after addition of particles
gave a mean value of 1.0 0.3 ms for the diusion time.
Applying Eq. 2 (see Materials and methods) a hydrodynamic radius of 14 4 nm was obtained. This value is in
agreement with the mean diameter of about 20 nm of
individual particles (Species 1). The loss of any correlation at later time-points was the result of a too high particle concentration in the observed cell. Over a period of
about 10 min the signal intensity passed through a maximum value and stabilized at a slightly lower intensity,
dependent on the initial particle concentration but inde-

112

M. Edetsberger et al. / Biochemical and Biophysical Research Communications 332 (2005) 109116

Fig. 1. Confocal images of HeLa cells. (A) HeLa cells stained with Nile-Red and treated with Genistein. (B) HeLa cells stained with Nile-Red,
treated with Genistein, and incubated with green uorescent particles. (C) Native HeLa cells stained with Nile-Red and incubated with green
uorescent particles. Images represent sections of 500 nm thickness taken 34 lm above the glass surface. (For interpretation of the references to
colors in this gure legend, the reader is referred to the web version of this paper.)

Fig. 2. Fluorescence correlation measurements in HeLa cells. (A1C1)

Intensity scans measured at dierent time-points. (A2C2) Auto
correlation function (ACF) obtained at dierent time-points.
Untreated cells (black), cells treated with Genistein (red) and Cytochalasin B (green). (For interpretation of the references to colors in this
gure legend, the reader is referred to the web version of this paper.)

pendent of whether the cell was treated with cytotoxins

or not.
About 1015 min after addition of particles, the picture changed dramatically (see Figs. 2B1 and B2).
Whereas in untreated cells high uorescence uctuations
occurred this was not the case for treated cells. Here a
two component model had to be applied resulting in diffusion times of 1.3 0.4 and 7 1 ms. The rst value
indicates Species 1 whereas the second value can be
attributed to clusters of a large number of particles. A
radius of 102 14 nm was obtained for this bright species (Species 2). During the next 20 min, no correlation
could be found in cells treated with Genistein or Cytochalasin B.
Additional eects occurred after 40 min which were
observed for the next 20 min. In the untreated cells,
intensities did not change signicantly, but an additional
species with a diusion time of 35 11 ms appeared (for
60 min of incubation see C1 and C2, Fig. 2). This species
with a radius larger than 350 nm was not seen in any of
the drug treated samples. However, larger intensity uctuations were also observed in the drug treated samples
(see Figs. 2C1 and C2) but contributing much less to the
overall uorescence intensity than in untreated cells.
Diusion times between 6 and 21 ms, similar to those

M. Edetsberger et al. / Biochemical and Biophysical Research Communications 332 (2005) 109116

113

Table 1
Average size distribution and relations of detected species
Incubation time (min)

Species 1

Species 2

Species 3

rhydr. (nm)

Detected (%)

rhydr. (nm)

Detected (%)

rhydr. (nm)

Detected (%)

Native cells
2
1520
5060

1019
1325
1937

100
97
90

NDa
88117
88146

NDa
3
7

NDa
NDa
352675

NDa
NDa
3

Cell treated with Genistein

2
1520
5060

1720
No correlation found
2739

100

NDa

NDa

NDa

NDa

161308

NDa

NDa

Cells treated with Cytochalasin B

2
1520
5060

1217
No correlation found
2739

100

NDa

NDa

NDa

NDa

161308

NDa

NDa

97

98

ND, not detected.

of Species 2 seen in untreated samples, were related to

these uctuations but they did not cause any signicant
increase in intensity.
The characteristics of the individual species observed
in the FCS measurements are summarized in Table 1.
Assuming that the observed large species are aggregates
of a large number of small particles, not only an increase
of the size but also of the brightness of the particles
should be observed. Both parameters characterize the
dierent species. Information about the brightness distribution could not be obtained by FCS but using PCH.

cies have dierent concentrations and are discriminated

by their brightness values. Concentrations are specied
by relative units (r.u.), as the concentration of particles
inside the cell is not known. The brightness is specied
by counts per second (cps). The following parameters
gave the best t for the measured histograms (Fig. 3):
for Species 1 a relative brightness of 1000 cps and a concentration of 350 r.u. were assumed (black line). For
Species 2 (red line), and 3 (green line) 2.2 106 and
6.0 106 cps and relative concentrations of 0.12 and

PCH
In Fig. 3, a typical measurement in native cells is
shown obtained after 50 min of incubation (orange line).
The resulting histogram can be described as contributions from at least three species which is in agreement
with the results from FCS measurements. The three spe-

Fig. 3. Simulation of the brightness distribution for three species with

dierent concentration and brightness values. This theoretical and
convoluted graph (blue) corresponds to three dierent species whereas
the contribution of every single species is shown in black (Species 1),
red (Species 2), and green (Species 3). For comparison to the
convolution of these species an experimental curve (orange) is plotted.
P(n) represents the probability of detected photon counts. (For
interpretation of the references to colors in this gure legend, the
reader is referred to the web version of this paper.)

Fig. 4. Normalized distribution of photon counts. Untreated samples

(A), cells treated with Genistein (B) or Cytochalasin B (C). On the
right side the temporal evolution of the particle-concentration is
shown. The numbers in the graphs indicate the particular species. The
legends show the dierent time-points of the measurement. Controls
without particles are plotted in orange. (For interpretation of the
references to colors in this gure legend, the reader is referred to the
web version of this paper.)

114

M. Edetsberger et al. / Biochemical and Biophysical Research Communications 332 (2005) 109116

0.003 r.u. were proposed respectively. The individual

contributions are shown together with their convolution
(blue line) which shows good agreement with the PCH
measured after 50 min (orange line). These values indicate that the larger species are aggregates of around
2000 (Species 2) and 6000 (Species 3) particles.
These values were used to evaluate the time dependence of appearance of particles in the cytoplasm of a
cell (Fig. 4). Species 1 was detected over the whole
observation period of 60 min in untreated cells. Their
relative concentration changed with time. It increased
from an initial value of 261 r.u. to a maximum value
of 537 r.u. within about 10 min. Later the relative concentration decreased and stabilized at a value of
339 r.u. after about 20 min. The mean brightness was
well dened with 1000 0.8 cps. The decrease in concentration found for Species 1 is accompanied by an increase of Species 2 after about 15 min which reached a
nal concentration of 0.2 r.u. The respective brightness
was calculated to be (2.2 0.02) 106 cps. Species 3 appeared after about 30 min reaching a relative concentration of 0.003 r.u. with a mean brightness of
(5.8 0.06) 106 cps. This species did not account for
a signicant decrease of the other two species.
Contrary to untreated cells, in cells treated with Genistein or Cytochalasin B (Figs. 4B and C) only Species 1
was found in signicant concentrations. In Genistein
treated cells, its relative concentration passed a maximum
value (889 r.u. at the beginning) and stabilized at lower
concentrations (417 r.u.) similar to untreated cells. The
brightness was stable at 1000 0.5 cps. A minute amount
of Species 2 with brightness values of (2.0 0.5) 106 cps
and a concentration of 1 10
4 r.u. was also observed. In
samples treated with Cytochalasin B, the concentration
of Species 1 decreased from an initial value of 761
427 r.u. at the end of the measurement (brightness:
1000 0.3 cps). The second species showed brightness
values of (2.1 0.25) 106 cps but a comparably low
concentration of 4 10
5 r.u.

Discussion
LSM images showed that negatively charged polystyrene particles with a diameter of 20 nm are eectively
internalized by human HeLa cells independent of
whether the cells were treated with Genistein or not.
In native cells, isolated particles and small aggregates
were found throughout the cytoplasm. Additionally
bright aggregates were found mainly at the periphery
of the cell. Such aggregates of individual particles could
result from incorporation into endosomes or similar
structures (e.g., lysosomes, exosomes, and multivesicular
bodies) [26]. In cells treated with Genistein, only small
aggregates were found nearer to the cytoplasm membrane. One explanation could be that in that case parti-

cles are not eciently packed into vesicles or that such

structures are not actively transported within the cell.
Translocation of particles needs a viable, uid membrane as particles are not found within dead cells. This
is in agreement with earlier results where it was shown
that cholesterol depletion, a process which decreases
the uidity of the membrane, reduces the entrance of
particles up to 200 nm in diameter [12]. Whereas earlier
experiments were performed with xed cells after long
incubation times, it was intended in this study to follow
the time-course of the appearance of particles inside the
cells under almost physiological conditions. FCS identied highly mobile particles with a hydrodynamic radius
of about the size of the applied particles in living cells.
Additionally larger species were found at later times
with a broad distribution of radii (see Table 1). The
appearance of such aggregated species was inhibited if
the cells were treated with Genistein or Cytochalasin
B. Nevertheless, species showing low brightness values
are dicult to detect in the autocorrelation curve if a
low number of particles with more than hundredfold
brightness are present. Thus, the dierent species could
only be quantied using FIDA (see Figs. 3 and 4).
The photon counting histograms were dominated by
a species of rather low brightness, identied as individual 20 nm particles, over the whole observation period
independent of the treatment of the cells. In untreated
samples, a considerably brighter species accompanied
the decrease of the number of individual particles with
time. At longer observation times singular events of an
even brighter species were observed. In cells treated with
cytotoxins aggregation is only detected in minute
amounts. The long-term limit of these observations
coincides with the results obtained by LSM.
Aggregation must not primarily occur on the surface
of the cell but could be the result of active processes within the cell (e.g., internalization of particles within vesicles
and their transport). Genistein and Cytochalasin B interact with such aggregation and transport mechanisms
[1821]. However, this aggregation process is time dependent as aggregates within the cell are not observed immediately after addition of particles but with a considerable
delay of about 20 min. Even more important is that the
used cytotoxins do not inhibit the internalization of singular particles or very small aggregates and their distribution in the cytoplasm. The initial intensities in the
PCH measurements indicate that their translocation is
even slightly enhanced (Fig. 4) which is in good agreement with recent results obtained in other cell lines
[12]. Internalization of such small nanometer sized and
negatively charged polystyrene particles is thus not
dependent on the availability of active uptake or transport processes, at least in the case of HeLa cells under
rather native conditions.
Intracellular aggregation processes like inclusion in
vesicles are at least partly responsible for the decrease

M. Edetsberger et al. / Biochemical and Biophysical Research Communications 332 (2005) 109116

of the number of individual particles in the cell. In treated cells, an equivalent decrease occurs even earlier than
in native cells and is also more pronounced. As no
aggregation is observed in that case it must be concluded
that singular particles, initially dispersed over the cytoplasm, are partly released from the interior of such paralyzed cells. This is, however, not due to aggregation as
bigger aggregates are not formed and transported within
the cytoplasm of treated cells. It is more plausible that
the particles leave the cell and are not replaced by new
ones due to aggregation and sedimentation processes
in the surrounding medium.
As particles can enter the cell without an active uptake it must be assumed that they can also leave the cell
easily which could explain the transport of nanoparticles
through tissues and even barriers in the body, which
normally cannot be overcome by molecular compounds.
The molecular mechanisms of this translocation in its
context with receptor mediated uptake are currently
investigated in detail [11].
Although the experiments show statistical variabilities dependent on the conditions of the individual measurements (e.g., the individual cell or the local particle
concentration, due to sedimentation processes), the general pattern for the interaction of particles with cells can
be deduced: the used particles penetrate the cellular
membrane without a signicant delay, endocytotic processes are not necessary for translocation, aggregation
occurs at later time and can be blocked by cytotoxins
interfering with cellular uptake processes.
Even though the particle concentration used in our
investigation, i.e., about 100 lg/104 cells, is very high
when compared to particle concentrations in ambient
air, which is around 2050 lg/m3 air [27], similar eects
could be obtained due to long-term accumulation of
inert ultrane particles in certain organs, e.g., in the
alveolae of the lung. The observed mechanisms could
thus be responsible for negative health eects of nanoparticles. On the other hand, these results should also
contribute to a better understanding of the mechanisms
involved in the biotechnological application of nanoparticles as carriers for drugs or DNA.

Acknowledgment
The project is funded by Allgemeine Unfallversicherungs-Anstalt, AUVA, Aldalbert-Stifterstrasse
65, A1201 Vienna.

References
[1] F. Drewnick, Airborne nanoparticles: properties, measurement
and atmospheric relevance, lecture series WS 2004/2005, Max
Planck Institute for Chemistry/Exp. Physics.

115

[2] P. Rajagopalan, F. Wudl, R.F. Schinazi, F.D. Boudinot, Pharmacokinetics of a water-soluble fullerene in rats, Antimicrob.
Agents Chemother. 40 (1996) 22622265.
kerman, W.C. Chan, P. Laakkonen, S.N. Bhatia, E.
[3] M.E. A
Ruoslahti, Nanocrystal targeting in vivo, Proc. Natl. Acad. Sci.
USA 20 (2002) 1261712621.
[4] M.N. Sarbolouki, M. Sadeghizadeh, M.M. Yaghoobi, A. Karami,
T. Lohrasbi, Dendrosomes: a novel family of vehicles for
transfection and therapy, J. Chem. Technol. Biotechnol. 75
(2000) 919922.
[5] S. Diabate, S. Mulhopt, H.R. Paur, R. Wottrich, H.F. Krug, In
vitro eects of incinerator y ash on pulmonary macrophages and
epithelial cells, Int. J. Hyg. Environ. Health 204 (2002) 323326.
[6] J.M. Samet, D.M. DeMartini, H.V. Malling, Do airborne
particles induce heritable mutations? Science 304 (2004) 971972.
[7] K.R. Smith, J.M. Veranth, A.A. Hu, J.S. Lighty, A.E. Aust,
Interleukin-8 levels in human lung epithelial cells are increased in
response to coal y ash and vary with the bioavailability of iron as
a function of particle size and source of coal, Chem. Res. Toxicol.
13 (2000) 118125.
[8] S. Boland, A. Baeza-SQuiban, T. Fournier, O. Houcine, M.C.
Gendron, M. Chevier, G. Jouvenot, A. Coste, M. Aubier, F.
Marano, Diesel exhaust particles are taken up by human airway
epithelial cells in vitro and alter cytokine production, Am. J.
Physiol. 276 (1999) 604613.
[9] G. Oberdorster, Z. Sharp, V. Atudorei, A. Elder, R. Gelein, A.
Lunts, W. Kreyling, C. Cox, Extrapulmonary translocation of
ultrane carbon particles following whole-body inhalation exposure of rats, J. Toxicol. Environ. Health 65 (2002) 15311543.
[10] A. Moore, E. Marecos, A. Bogdanov, R. Weissleder, Tumoral
distribution of long-circulating dextran-coated iron oxide nanoparticles in a rodent model, Radiology 214 (2000) 568574.
[11] G. Oberdorster, Z. Sharp, V. Atudorei, A. Elder, R. Gelein, W.
Kreyling, C. Cox, Translocation of inhaled ultrane particles to
the brain, Springer Series on Fluorescence, 2, 2002 154171,
Inhalation Toxicology 16 (2004) 437445.
[12] J. Rejman, V. Oberle, S.I. Zuhorn, D. Hoeckstra, Size-dependent
internalization of particles via the pathway of clathrin- and
caveolae-mediated endocytosis, Biochem. J. 377 (2004) 159169.
[13] N. Kapp, W. Kreyling, H. Schulz, H. Im Hof, P. Gehr, M.
Semmler, M. Geiser, Electron energy loss spectroscopy for
analysis of inhaled ultrane particles in rat lungs, Microsc. Res.
Tech. 63 (2004) 298305.
[14] S.T. Hess, S. Huang, A.A. Heikal, W.W. Webb, Biological and
chemical applications of uorescence correlation spectroscopy: a
review, Biochemistry 41 (2002) 697705.
[15] P. Kask, P. Kaupo, D. Ullmann, K. Gall, Fluorescence-intensity
distribution analysis and its application in biomolecular detection
technology, Proc. Natl. Acad. Sci. USA 96 (1999) 1375613761.
[16] Y. Chen, J.D. Muller, T.C.S. So, E. Gratton, The photon
counting histogram in uorescence uctuation spectroscopy,
Biophys. J. 77 (1999) 553567.
[17] J.D. Muller, Y. Chen, E. Gratton, Resolving heterogeneity on the
single molecular level with the photon-counting histogram,
Biophys. J. 78 (2000) 474486.
[18] W. Duan, C. Kuo, S. Selvarjan, K.Y. Chua, B.H. Bay, F.S.W.R.
Wong, Antiinammatory eects of Genistein, a tyrosine kinase
inhibitor, in a guinea pig model for asthma, Respir. Crit. Care
Med. 167 (2003) 185192.
[19] T. Sorkina, F. Huang, L. Beguinot, A. Sorkin, Eect of tyrosine
kinase inhibitors on clathrin-coated pit recruitment and internalization of epidermal growth factor receptor, J. Biol. Chem. 277
(2002) 2743327441.
[20] J.C. Norman, P.T. Harrison, W. Davis, R.A. Floto, J.M. Allen,
Lysosomal routing of Fc gamma RI from early endosomes
requires recruitment of tyrosine kinases, Immunology 94 (1998)
4855.

116

M. Edetsberger et al. / Biochemical and Biophysical Research Communications 332 (2005) 109116

[21] C. Stournaras, E. Stiakaki, S.B. Koukouritaki, P.A. Theodoropoulos, M. Kalmanti, Y. Fostinis, A. Gravanis, Altered actin
polymerization dynamics in various malignant cell types: evidence
for dierential sensitivity to Cytochalasin B, Biochem. Pharmacol.
52 (1996) 13391346.
[22] A.H. Schor, N. Campos, Mechanisms involved in calcium oxalate
endocytosis by MadinDarby canine kidney cells, Braz. J. Med.
Biol. Res. 33 (2000) 111118.
[23] P. Schwille, U. Haupts, S. Maiti, W.W. Webb, Molecular
dynamics in living cells observed by uorescence correlation
spectroscopy with one- and two-photon excitation, Biophys. J. 77
(1999) 22512265.
[24] S.T. Hess, E.D. Sheets, A. Wagenknecht-Wieser, A.A. Heikal,
Quantitative analysis of the uorescence properties of intrinsically

uorescent proteins in living cells, Biophys. J. 85 (2003) 2566

2580.
. Mets, M.Cole M, K. Gall,
[25] P. Kask, C. Eggeling, K. Palo, U
Fluorescence intensity distribution analysis (FIDA) and related
uorescence uctuation techniques: theory and practice, in: Ruud
Krayenhof, Antonie J.W.G. Visser, Hans C. Gerritsen, (Eds.),
Fluorescence Spectroscopy, Imaging and Probes (Springer Series
on Fluorescence, 2, 2002) 154171.
[26] L. Rajendran, K. Simons, Lipid rafts and membrane kinetics, J.
Cell Sci. 118 (2005) 10991102.
[27] M. Pitz, W.G. Kreyling, B. Holscher, J. Cyruys, H.E. Wichmann, Change of the ambient particle size distribution in East
Germany between 1993 and 1999, Atmos. Environ. 35 (2001)
43574366.