Science of the Total Environment 408 (2010) 4475–4481

Contents lists available at ScienceDirect

Science of the Total Environment

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s c i t o t e n v

In vitro evaluation of the cytotoxicity of iron oxide nanoparticles with different

coatings and different sizes in A3 human T lymphocytes
Erbo Ying, Huey-Min Hwang ⁎
Department of Biology, Jackson State University, P.O. Box 18540, Jackson, MS 39217 USA

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

Article history: The ever expanding use of engineered nanoscaled materials has brought about a commensurate growth in
Received 2 April 2010 concern about their potential risks to human and environmental health. Toxicity of nanoparticles could vary
Received in revised form 23 June 2010 with their physicochemical parameters. The dependence of cytotoxicity on particle size and surface coating
Accepted 7 July 2010
of iron oxide nanoparticles was investigated in this in vitro study using the A3 human T lymphocyte as a
Available online 2 August 2010
model. Two different sizes (10 nm and 50 nm) and two different surface coatings (amine and carboxyl
Keywords:
groups) of iron oxide (IO) nanoparticles were tested with fluorescein diacetate (FDA) assay and WST-1
Cytotoxicity assay. In the 1-h FDA assay with PBS, IO nanoparticles did not show size-dependent toxicity to A3 cells in
Iron oxide (IO) nanoparticles terms of mass concentration; however, in terms of the number of particles per well and the total surface
Fluorescein diacetate (FDA) uptake area, they did exhibit size-dependent toxicity. Fifty nanometer IO nanoparticles are more toxic than the
WST-1 assay 10 nm counterparts. The results of both the 24-h FDA and WST-1 assays in a complete growth medium
A3 cells indicate size- and surface coating-dependent toxicity to A3 cells in terms of mass concentration. IO
nanoparticles of the smaller size are more toxic than those of the larger size. IO nanoparticles with the
carboxyl group have a higher toxicity than those with the amine group. However, in the 24-h FDA assay, in
terms of the number of particles per well and the resultant total surface area per well, the 50 nm IO
nanoparticles are more toxic than those of size 10 nm. In terms of mass concentration, the number of
particles per well and the total surface area, both of the 24-h assays showed the consistent results that IO
nanoparticles with the carboxyl group have a higher toxicity than those with the amine group.
© 2010 Elsevier B.V. All rights reserved.

1. Introduction
Modern nanotechnology has grown from a concept in 1959, to a
scientific interest, to a major industry today. Global populations and
ecosystems have been exposed to both commodity and specialty
nanomaterials. Nanoparticles are used in various fields such as
photonics, catalysis, magnetics, and biotechnology including cos-
metics, pharmaceutics, and medicine (Choi et al., 2009). Over 500
consumer products currently on the market claim to contain elements
of nanoscience and nanotechnology with new entries coming daily
(Jones and Grainger, 2009). Consequently, widespread application of
nanomaterials confers enormous potential for human exposure and
environmental release (Oberdorster et al., 2005).
Superparamagnetic iron oxide nanoparticles (SPIONs) have gained
an important role in recent developments in magnetic resonance
imaging contrast enhancement, tissue repair, immunoassay, detoxi-
fication of biological fluids, hyperthermia, drug delivery, and cell
separation (Yu et al., 2008; Yavuz et al., 2006; Weissleder et al., 1995;
Chouly et al., 1996; Thiesen and Jordan, 2008; Reimer and Weissleder,
⁎ Corresponding author. Fax: + 1 601 9796856.
E-mail address: huey-min.hwang@jsums.edu (H.-M. Hwang).
1996; Gupta and Hung, 1989). Intense research efforts have been
directed toward the application of SPIONs as drug carriers in magnetic
drug targeting, as contrast agents in magnetic resonance imaging
(MRI) and as heating mediators in hyperthermia treatments.
Iron is an essential nutrient for mammals and most other life
forms; therefore, iron oxide nanoparticles were generally assumed to
be safe. It is worth noting that neither iron oxide nanoparticles alone
nor the coating material alone is overtly toxic, but combining the two
to create water-soluble nanoparticles has a completely different effect
(University of California-San Diego, 2007). Despite the pros and cons
of using nanoscaled iron oxides for in vivo applications, super-
paramagnetic nanoparticles remain the only magnetic nanoparticles
that have been approved for clinical use to date (Gould, 2006).
For in vitro toxicological studies, cell culture techniques are often
preferred due to their rapid, efficient and cost effective merits. Uptake
of engineered nanomaterials into the human body has several
different routes. Nanomaterial exposure is likely to occur through
dermal contact and inhalation because the skin and lungs are in direct
contact with the environment (Hoet et al., 2004). Hydrophobic
nanoparticles can penetrate through tissue and pass into the
lymphatics. Nanomaterials pose a high health risk because of their
ability to reach every part of the organs and tissues and their
interaction with cellular functions. Despite these concerns, only

0048-9697/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.scitotenv.2010.07.025
4476 E. Ying, H.-M. Hwang / Science of the Total Environment 408 (2010) 4475–4481
limited studies have been actually conducted to assess the health
effect of nanomaterials on the lymphatic system. T lymphocytes
belong to a group of white blood cells known as lymphocytes, and
play a central role in cell-mediated immunity. In this study, we
selected A3 T lymphocytes as the model specimen for assessing the
cytotoxicity of the test nanomaterials.
Much of the existing research on nanotoxicity has concentrated on
the empirical evaluation of the toxicity of various nanoparticles with
less attention having been given to the relationship between
nanoparticle properties (e.g., exact composition, shape, size, size
dispersion, and aggregation) and toxicity. This approach gives very
limited information, and should not be considered adequate for
developing predictions of toxicity of seemingly similar nanomaterials.
Normally, size is considered to be a primary property describing the
transport behavior of nanoparticles. Surface coating can influence
nanoparticle behavior in the biological system and its toxicity. We
adopted a systematic approach to investigate the toxicity of iron oxide
nanoparticles by focusing on different surface coatings with the same
particle size, and different particle sizes with the same surface coating.
This study aimed to provide new information about the cytotoxicity of
iron oxide nanoparticles with variation in particle size and surface
coating by using A3 human T lymphocytes.
Furthermore, measurement of biological end-point responses was
conducted with fluorescein diacetate (FDA) and WST-1 assays
following exposure to the same nanomaterials. The FDA assay has
been regarded as an economical and rapid tool which was successfully
used in our previous cytotoxicity studies of human cell lines (Hu et al.,
2009; Zhang et al., 2006; Zheng et al., 2004). Due to its convenience
and great sensitivity, recently the WST-1 assay has become a very
popular cytotoxicity assay in the nanotoxicity study (Yamawaki and
Lwai, 2006; Salem et al., 2003; Pulskamp et al., 2007; Worle-Knirsch
et al., 2006; Su et al., 2007; Hsieh et al., 2006). In this study both assays
were used to assess the IO nanotoxicity for cross comparison and
substantiation.
2. Materials and methods
2.1. Chemicals and reagents
Unless otherwise stated, all chemicals and reagents used were
purchased from Sigma (St. Louis, MO).
2.2. Iron oxide (IO) nanoparticles
All the IO nanoparticles were purchased from Ocean Nanotech Inc.
(Springdale, AR). In our study, we used three different IO nanopar-
ticles including SHP 10, SHP 50 and SHA 10 [number labels the size
(nm) of the inorganic core]. SHP and SHA are groups of water-soluble
IO nanocrystals with carboxyl group and amine group, respectively.
2.3. Cell culture
A3 human T lymphocytes, obtained from the American Type
Culture Collection (ATCC, Manassas, VA), were grown in T-25 Falcon
flasks (Becton Dickinson Labware, Bedford, MA) at 37 °C in a 95% air/
5% CO2 humidified incubator (Isotemp; Fisher Scientific, Houston, TX)
using the recommended RPMI 1640 medium. RPMI 1640 complete
growth medium with 2 mM L-glutamine was modified by ATCC to
contain 10 mM HEPES, 1 mM sodium pyruvate, 4.5 g/L glucose, and
1.5 g/L sodium bicarbonate, supplemented with 10% fetal bovine
serum and 1% penicillin/streptomycin antibiotic. RPMI 1640 medium,
fetal bovine serum, penicillin, and streptomycin were purchased from
Gibco-Invitrogen (Carlsbad, CA). The A3 cells were collected by
centrifugation at 129 g for 5 min at 4 °C. The pellet was washed
twice with 1x PBS. The viability of cells processed in this way was over
95% by checking with the Trypan blue dye method (Rijnkels et al.,
2003). Unless otherwise stated, the cell density was adjusted to 5 × 105
cells/mL with RPMI 1640 complete growth medium and used for all
subsequent experiments. The A3 cells used for the experiments were
between passages 7 and 16.
2.4. Transmission electron microscopy (TEM)
For the mixtures of IO nanoparticles and complete growth medium,
samples were mixed at the concentration of 250 μg/mL in a 96-well
microplate. Aqueous dispersion of the particles was drop-cast onto a
carbon-coated copper grid. The rest of the mixtures was incubated for
24 h at 37 °C in a 95% air/5% CO2 humidified incubator, and then dried
onto the carbon-coated copper grids. All the grids were dried overnight
and observed using a TEM (JEOL JE-1011).
2.5. One-hour fluorescein diacetate (FDA) uptake based cytotoxicity
assay
The effect on the cell proliferation by IO nanoparticles was
evaluated by the widely established FDA test (Hu et al., 2009; Zhang
et al., 2006; Zheng et al., 2004). Fifty microliter aliquots of A3 cell
suspension in PBS were added into each well of microplate. To each
50 μL suspension different concentrations of each type of nanoparticle
were made with PBS and added into each well to obtain the final cell
density of 5 × 105 cells/mL and the final test mass concentrations of 0,
5, 25, 50, 125, 250, 375, and 500 μg/mL. A3 cells without nanoparticles
served as control. The control and each treatment concentration were
assayed in eight replicate wells. The microtiter plate was cultured at
37 °C under 5% CO2 in a humidified incubator. After 1 h of treatment, a
100 μL aliquot of 10 ng/mL FDA was added to all four wells of each
treatment. One hundred microliters of PBS was added to another four
wells of each treatment as the background control (blank) to preclude
potential interference of nanoparticles with the spectrophotometric
measurement. The plates were incubated for another 35 min and the
fluorescence read using a microplate spectrofluorimeter (Triad series;
Dynex Technologies, Chantilly, VA) at an emission wavelength of
538 nm and excitation of 485 nm. The percent cytotoxicity was
calculated with the following equation: (Ftreatment − Ftreatment blank)/
(Fcontrol − Fcontrol blank) × 100, where Ftreatment is the fluorescence
intensity of the sample and Ftreatment blank is the fluorescence intensity
of each treatment without FDA reagent which here mainly includes
the fluorescence produced by PBS and IO nanoparticles.
2.6. Twenty-four-hour (24-h) fluorescein diacetate (FDA) uptake based
cytotoxicity assay
The 24-h FDA uptake assay was conducted with the same
procedure as the 1-h FDA uptake, except for a 24 h exposure time
instead of 1 h, and in the complete growth medium instead of in PBS.
To each 50 μL suspension different concentrations of each type of
nanoparticle were made with the complete growth medium and
added into each well to obtain the final cell density of 5 × 105 cells/mL
and the final test mass concentrations of 0, 5, 25, 50, 75, 100, 150, 200,
250, and 300 μg/mL. The percent cytotoxicity was calculated with the
same equation used for the 1-h FDA assay. Here, the Ftreatment blank is
the fluorescence intensity of each treatment without FDA reagent
which mainly includes the fluorescence produced by the RPMI 1640
complete growth medium and IO nanoparticles.
2.7. Twenty-four-hour (24-h) WST-1 based cytotoxicity assay
The PreMix WST-1 cell proliferation assay system was obtained from
Takara Bio Inc. (Shiga, Japan). A modified procedure of PreMix WST-1
cell proliferation assay system was used by following manufacturer's
instructions. After 24 h of incubation, 10 μL of PreMix WST-1 reagent
was added into four wells of each treatment, respectively. Ten
 E. Ying, H.-M. Hwang / Science of the Total Environment 408 (2010) 4475–4481
microliters of PBS was added into another four wells of each treatment
as the background control (blank). The microtiter plate was incubated
for another 4 h to allow the formation of formazan dye and to get a
higher sensitivity. The final test mass concentrations of nanoparticles
followed those of the 24-h FDA assay. The absorbance was measured in a
microplate reader (Triad series; Dynex Technologies, Chantilly, VA) at
450/620 nm. The percent cytotoxicity was calculated with the following
equation: (Atreatment − Atreatment blank) / (Acontrol − Acontrol blank) × 100,
where Atreatment is the absorbance of the sample and Atreatment blank is
the absorbance with each treatment without WST-1 reagent which here
mainly includes the absorbance of the RPMI 1640 and IO nanoparticles.
2.8. Statistical analysis
All results are expressed as the mean ± standard deviation (SD).
All of the experiments were conducted at least 3 times, and each
treatment had four replicates. The SAS System for Windows, V8 (SAS
Institute, Gary, NC) was used for statistical evaluations. To detect the
interactions between and among different factors in the experiment,
statistical analysis of the experimental results was conducted by
factorial arrangement of treatments in a complete randomized design
using the general linear model (GLM) in SAS. LSMEANS were used to
separate combinations of means. Significance of the treatment effect
or the interactions among and between the experimental factors were
determined at p b 0.05 (Hwang et al., 2004). In addition, we used the
regression calculation to determine the LC50.
3. Results and discussion
3.1. IO nanoparticle stability in the complete growth medium
According to the specifications of the products, SHP (IO nanoparticles
with carboxyl group) is very stable in most buffer solutions in the pH
Fig. 1. TEM images of the different iron oxide nanoparticles (SHA 10, SHP 10, and SHP 50) mixed with RPMI 1640 complete growth medium for 0 and 24 h, respectively, at 37 °C in a
95% air/5% CO2 humidified incubator.
4477
range of 2–14, and pH stability range for SHA (IO nanoparticles with
amine group) is between 5 and 10. Many nanoparticles have been
reported to aggregate when they are present in some high ionic
environment such as the complete growth medium. In our study, we
checked the stability of the IO nanoparticles after they were mixed with
the RPMI 1640 complete growth medium for 0 and 24 h at 37 °C under 5%
CO2 in a humidified incubator. As shown in Fig. 1, after 24-h incubation in
the complete growth medium, all of them remained as the separate
nanoparticles and the size of the particles did not change measurably.
Therefore, the IO nanoparticles used in our study were extremely stable
throughout the incubation periods.
3.2. One-hour FDA uptake test of IO nanoparticles
The 1-h FDA test in PBS is a well-established procedure in our lab (Hu
et al., 2009; Zhang et al., 2006; Zheng et al., 2004). In one of the
experiments, the A3 cells in PBS were exposed to the IO nanoparticles
with different coatings and different sizes for 1 h. SAS analysis of the
experimental data with factorial arrangement was used to detect the
effect of individual treatment factors and the interactions among or
between various factors (Hwang et al., 2004). The p values of the
possible factor interactions in SHP 10 + SHP 50 and SHP 10+ SHA 10
experimentations are shown in Table 1A and B, respectively. The p
values in Table 1A indicate that cellular viability was significantly
affected by the following individual treatment or combination of
treatments: IO nanoparticle concentration (p b 0.0001), IO nanoparticle
concentration/size (p b 0.0001), and IO concentration/surface coating
(p = 0.0335). In this study we were mainly interested in assessing the
effects of size, surface coating, and concentration on the FDA test results,
thus data comparisons were made only for the individual treatment
factors aforementioned. In terms of mass concentration of IO nanopar-
ticles, cell viability is concentration-dependent (Fig. 2A, Table 1,
p b 0.0001). However, there is no significant effect on cellular viability
4478 E. Ying, H.-M. Hwang / Science of the Total Environment 408 (2010) 4475–4481

Table 1
Effect of the treatment factors on the cytotoxicity results and interactions between the
experimental factors in the 1-h FDA assay. (A) Effect of the size and concentration; (B)
Effect of the surface coating and concentration.

Factor DF Sum of square Mean square F value pNF

A:
Size 1 9.61 9.61 1.51 0.2284
Concentration⁎ 7 56,759.44 8108.49 1272.25 b0.0001
Concentration/Size⁎ 7 739.43 105.63 16.57 b0.0001

B:
Surface coating 1 3.81 3.81 0.52 0.4780
Concentration⁎ 7 61,863.71 8837.67 1196.09 b0.0001
Concentration/ 7 131.58 18.80 2.54 0.0335
Surface coating
⁎ Significant effect (p b 0.05).

when size and surface coating change (Table 1, p = 0.2284, 0.4780).

Consequently, only the concentration-dependent effect was detected in
the 1-h FDA assay (Table 1).
The LC50 (lethal concentration, 50%) of a toxic substance is the
dose required to kill half the numbers of a tested population after a
specified test duration. The LC50 value is frequently used as a general
indicator of a substance's acute toxicity. In this 1-h FDA assay in PBS,
the values of LC50 of A3 cells after exposure to SHP 10, SHP 50, and
SHA 10 are 79, 65, and 79 μg/mL, respectively (Table 2). Therefore,
SHP 10 and SHA 10 exhibited the same cytotoxicity to the A3 cells.
Fig. 2B shows the relationship between cell viability and the total
number of particles per well. Consistently, the increase in particle
number leads to an increase of cytotoxicity by all three IO
nanoparticles. At the similar particle number, cell viability was higher
after exposure to the 10-nm particles than that of the 50-nm particles.
In this sense, every small nanoparticle is less toxic than its larger
counterpart. Since cells were only accessible to the surface areas of the
present particles, the total surface areas per well were also calculated
for comparison (Fig. 2C). The difference of viabilities between the
large and the small particles was still significant in terms of total
surface areas per well. At the similar total surface area of particles per
well (around 0.005 m2), higher cell viability was also observed after
exposure to the 10-nm nanoparticles than to the 50-nm particles.
In summary, the results of the 1-h FDA assay in PBS solution show
that cell viability is concentration-dependent. Although IO nanoparticles
do not show the size-dependent toxicity to A3 cells in terms of mass
concentration, in terms of the number of particles per well and the total
Fig. 2. Fluorescein diacetate uptake by A3 cells following a 1-h treatment in PBS solution of
surface area, they do show size-dependent cytotoxicity. IO nanoparticles SHA 10, SHP 10, and SHP 50 at the mass concentrations of 0, 5, 25, 50, 125, 250, 375, and
with a larger size are more toxic than those with a smaller size. 500 μg/mL. Means labeled with different letters are significantly different. (A) Dependence
Nevertheless, no significant dependence on surface coating is observed. of cell viabilities on mass concentration; (B) dependence of cell viabilities on the number of
Similar results were also observed in lung MSTO-211H cells and skin particles per well and (C) dependence of cell viabilities on the total surface area of particles
per well.
keratinocyte HaCaT cells (data not shown).

3.3. Twenty-four-hour (24-h) FDA uptake test of IO nanoparticles Table 3 also shows that the viability of A3 cells is significantly affected
when the size and surface coating of the IO nanoparticles change
In order to support the longer 24-h FDA uptake assay, we (p b 0.0001). In terms of mass concentration the 10-nm IO nanopar-
substituted the PBS buffer in the 1-h FDA uptake assay with RPMI ticles are more toxic than the 50-nm counterparts (least squares
1640 complete growth medium. Fig. 3A shows that compared to the means 58.1% of control for 10-nm IO nanoparticles and 61.6% of
test in PBS solution cell viability in complete growth medium was control for 50-nm IO nanoparticles). IO nanoparticles with the
indeed higher, even though the exposure time was extended. It is carboxyl group are more toxic to A3 cells than those with the amine
reasonable because there were more nutrients in the complete
growth medium for the A3 cells. To make the technicality correction,
a background control (blank) was provided for every treatment group Table 2
The LC50 of A3 human T lymphocytes after exposure to the different IO nanoparticles in
in order to preclude the potential interference of the nanoparticles
the different cytotoxicity assay.
and medium with the spectrophotometric measurement.
SAS analysis of the experimental data with factorial arrangement LC50 (μg/mL) in the 1- LC50 (μg/mL) in the LC50 (μg/mL) in the 24-
h FDA uptake 24-h FDA uptake h WST-1 assay
was also provided for the experiment results. Based on the mass
concentration of IO nanoparticles, cell viability remains to be SHP 10 79 116 117
concentration-dependent (Fig. 3A, Table 3A, p b 0.0001), which is SHP 50 65 136 360
SHA 10 79 150 205
consistent with the result of the 1-h FDA uptake test (Fig. 2, Table 1).
 E. Ying, H.-M. Hwang / Science of the Total Environment 408 (2010) 4475–4481 4479

Fig. 3. Fluorescein diacetate uptake by A3 cells following a 24-h treatment in the RPMI
1640 complete growth medium of SHA 10, SHP 10, and SHP 50 at the mass
concentrations of 0, 5, 25, 50, 75, 100, 150, 200, 250, and 300 μg/mL. Means labeled
with different letters are significantly different. (A) Dependence of cell viabilities on
mass concentration; (B) dependence of cell viabilities on the number of particles per
well and (C) dependence of cell viabilities on the total surface area of particles per well.
of LC50 of A3 cells after exposure to SHP 10, SHP 50, and SHA are 116,
136, and 150 μg/mL, respectively (Table 2). The result of surface
coating-dependent toxicity is consistent with that reported by
Hoshino et al. (2004). The SHP particles and SHA particles are
expected to be negatively charged and positively charged, respec-
tively, in the growth medium and cellular system. We speculate that
the difference in the electronic interaction between the charged cell
surface and the aforementioned nanoparticles may lead to the larger
extent of oxidative stress exerted by the carboxyl surface coating
(Hoshino et al., 2004).
When the mass concentration of IO nanoparticles is converted to
the number of particles per well and the total surface area of particle
per well, concentration dependence of cell viabilities is still obvious
(Fig. 3B and C). At the similar number of particles (around 1 × 1011),
exposure to 50-nm IO nanoparticles resulted in lower cell viability
than to the 10-nm IO nanoparticles. This finding is in agreement with
the previous report by Yin et al. (2005). The speculation is that the
effective interaction area for accessing to the cell is greater for a large
particle than that for a small particle. Within this specific area, there
are more function groups on the individual large particle. Thus each
large particle exerts a stronger stimulus on the cells. However, if an
equal mass of particles with different sizes is considered, the
significantly larger number of small particles increases the number
of interaction points that are randomly distributed around the cell.
Due to the likelihood of a smaller number of functional groups on
individual small particles, each interaction point exerts a weak
stimulus on the cells. The experimental results seem to suggest that
the total sum of weak stimuli at different locations resulted in a lesser
toxic effect than that by one localized strong stimulus from a bigger
particle. Fig. 3 also shows the surface coating dependent-toxicity
between 25 and 200 μg/mL. IO nanoparticles with the carboxyl group
have a higher toxicity than those with the amine group. However, this
was not observed after exposure to the higher concentrations (250
and 300 μg/mL) and the lowest concentration (5 μg/mL; Fig. 3A).
In summary, the results of the 24-h FDA assay in the complete
growth medium show that cytotoxicity of the test IO nanoparticles is
concentration-dependent, and size and surface coating-dependent in
terms of mass concentration. Smaller IO nanoparticles have a higher
toxicity to the A3 cells than the bigger ones and the IO nanoparticles
with the carboxyl group are more toxic to the A3 cells than those with
the amine group. However, in terms of the number of particles per
well and the total surface areas, the relationship between particle size
and toxicity shifts; i.e., the 50-nm IO nanoparticles are more toxic than
the 10-nm counterparts (Fig. 3B, C). In the range of concentration 25–
200 μg/mL, IO nanoparticles with the carboxyl group have a higher
toxicity than those with the amine group.

group (least squares means 58.1% of control for IO nanoparticles with 3.4. WST-1 assay of IO nanoparticles
carboxyl group and 65.3% of control for IO nanoparticles with amine
group). In this 24-h FDA assay in complete growth medium, the values According to Ngamwongsatit et al. (2008), the WST-1 assay is a
more sensitive method for assessing cellular cytotoxicity. For
Table 3 comparison purpose, we also conducted the WST-1 assay to
Effect of the treatment factors on the cytotoxicity results and interactions between the determine the viability of the A3 cells after they were treated with
experimental factors in the 24-h FDA assay. (A) Effect of the size and concentration; IO nanoparticles with two different sizes and two different coatings
(B) Effect of the surface coating and concentration. for 24 h in RPMI 1640 complete growth medium. The results of this
Factor DF Sum of square Mean square F value pNF assay are similar to those of the 24-h FDA assay. Again, in terms of
mass concentration concentration-, size-, and surface coating-depen-
A:
Size⁎ 1 1035.26 1035.26 136.65 b0.0001 dent cytotoxicities are observed (Fig. 4A, Table 4, p b 0.0001). IO
Concentration⁎ 7 56,338.39 6259.82 826.27 b0.0001 nanoparticles with a large size (50 nm) are less toxic than those with a
Concentration/Size⁎ 7 1167.83 129.76 17.13 b0.0001 small size (10 nm) (least squares means 60.7% of control for 10-nm IO
nanoparticles and 89.7% of control for 50-nm IO nanoparticles). The
B:
Surface coating⁎ 1 244.31 244.31 20.12 b0.0001 10-nm IO nanoparticles with the carboxylic surface coating are more
Concentration⁎ 7 54,816.53 6090.73 501.52 b0.0001 toxic than those with the amine surface coating (least squares means
Concentration/ 7 562.59 62.51 5.15 b0.0001 60.7% of control for IO nanoparticles with carboxyl group and 69.3% of
Surface coating⁎ control for IO nanoparticles with amine group). In the 24-h WST-1
⁎ Significant effect (p b 0.05). assay in the complete growth medium, the values of LC50 of the A3
4480 E. Ying, H.-M. Hwang / Science of the Total Environment 408 (2010) 4475–4481
cells after exposure to SHP 10, SHP 50, and SHA 10 are 117, 360, and
205 μg/mL, respectively (Table 2). In terms of the LC50 of the
A3 cells exposed to the same IO nanoparticles in a different assay,
the 24-h WST-1 assay in the complete growth medium has the highest
value and the 1-h FDA assay in PBS has the lowest value. When the
concentration units on x-axis are converted to the number of particles
per well (Fig. 4B), the results are consistent with the 24-h FDA test
(Fig. 3B). IO nanoparticles with the carboxyl group are more toxic
than those with the amine group, and the 50-nm IO nanoparticles are
more toxic than the 10-nm counterparts (Fig. 4B). However, in terms
of total surface area of particles per well, the size of the particles does
not seem to significantly affect the cell viability (Fig. 4C).
The WST-1 assay is considered a modified test that involves
tetrazolium salts (Lewinski et al., 2008). Since only viable cells can
convert the salt into a water-soluble formazan, no solubilization of the
Fig. 4. Cell viabilities assessed by the WST-1 assay. Iron oxide nanoparticles, SHA 10,
SHP 10, and SHP 50 at the mass concentrations of 0, 5, 25, 50, 75, 100, 150, 200, 250, and
300 μg/mL were exposed to A3 cells for 24 h in the RPMI complete growth medium.
Means labeled with different letters are significantly different. (A) Dependence of cell
viabilities on mass concentration; (B) dependence of cell viabilities on the number of
particles per well and (C) dependence of cell viabilities on the total surface area of
particles per well.
Table 4
Effect of the treatment factors on the cytotoxicity results and interactions between the
experimental factors in the WST-1 assay. (A) Effect of the size and concentration; (B) Effect
of the surface coating and concentration.
Factor DF Sum of square Mean square F value pNF
A:
Size⁎ 1 16,806.38 16,806.38 955.28 b0.0001
Concentration⁎ 7 34,317.16 3813.02 216.73 b0.0001
Concentration/Size⁎ 7 3034.94 337.22 19.17 b0.0001
B:
Surface coating⁎ 1 1504.17 1504.17 221.64 b0.0001
Concentration⁎ 7 40,800.14 4533.35 667.99 b0.0001
Concentration/ 7 1074.61 119.40 17.59 b0.0001
Surface coating⁎
⁎ Significant effect (p b 0.05).
products is necessary as that required in the MTT dye assay. Except for
the minor difference at the low concentration range (Fig. 3A, Fig. 4A),
both the FDA uptake test and the WST-1 assay detected the similar
cellular viability response in this study. The reduced handling
required for the WST-1 assay suggests less chance of errors caused
by human pipetting and handling. With a slightly higher cost than the
FDA assay, the WST-1 assay is available in a ready-to-use solution.
Therefore, this convenient and rapid colorimetric assay can be
performed in the same microtiter plate where the cells were grown.
4. Conclusion
The results of all assays used in this study exhibit concentration-
dependent nanotoxicity. Although IO nanoparticles do not unanimously
show the size-dependent toxicity to the A3 cells in terms of mass
concentration in all assays, in terms of the number of particles per well
and the total surface area, the IO nanoparticles with a larger size are more
toxic than those with a smaller size. In terms of mass concentration, the
results of both the 24-h FDA and WST-1 assays in the complete growth
medium show size- and surface coating-dependent toxicity to the A3
cells. IO nanoparticles with the carboxyl group have a higher toxicity than
those with the amine group. Smaller sized IO nanoparticles are more toxic
than larger counterparts. But in the 24-h FDA assay, in terms of the
number of particles per well and the total surface area, the IO
nanoparticles with a larger size are more toxic than those with a smaller
size. In terms of mass concentration, the number of particles per well and
the total surface areas, the results of both 24-h assays show that IO
nanoparticles with the carboxyl group have a higher toxicity than those
with the amine group. As indicated early, this study aimed to provide new
information on the cytotoxicity of iron oxide nanoparticles with
variations in particle size and surface coating. Our findings in this study
reveal that size and surface coating could affect the nanotoxicity in
biological systems and the interpretation of the cytotoxicity of
nanoparticles could vary with the mass concentration, the total number
of particles per well, and the total surface area of particles per well. In
addition, modification in the surface coating could overcome the
cytotoxicity of IO nanoparticles. This should provide some suggestion
for formulating nanotechnological products in important clinical applica-
tions such as drug delivery and medical treatment.
Acknowledgements
This research was supported by: (1) JSU Interdisciplinary Center
for Nanotoxicity with funding of NSF HRD #0833178 and (2) NSF-SBIR
grant # IIP-0823040. We thank Dr. Rong Zhang of JSU Electron
Microscope Core Laboratory for providing technical service in using
the TEM. We used the Molecular and Cellular Lab. of the JSU RCMI core
facilities for conducting the toxicity assays. Mr. Winfred Aker's
assistance in manuscript revision is appreciated.
 E. Ying, H.-M. Hwang / Science of the Total Environment 408 (2010) 4475–4481
References
Choi S-J, Oh J-M, Choy J-H. Toxicological effects of inorganic nanoparticles on human
lung cancer A549 cells. J Inorg Biochem 2009;103:463–71.
Chouly C, Pouliquen D, Lucet I, Jeune JJ, Jallet P. Development of superparamagnetic
nanoparticles for MRI: effect of particle size, charge and surface nature on
biodistribution. J Microencapsulation 1996;13:245–55.
Gould P. Nanomagnetism shows in vivo potential. Nanotoday 2006;1:34–9.
Gupta PK, Hung CT. Magnetically controlled targeted microcarrier systems. Life Sci
1989;44:175–86.
Hoet P, Briusle-Hohlfeld I, Salata O. Nanoparticles—known and unknown health risks. J
Nanobiotechnol 2004;2:12-112.
Hoshino A, Fujioka K, Oku T, Suga M, Sasaki YF, Ohta T, et al. Physicochemical properties
and cellular toxicity of nanocrystal quantum dots depend on their surface
modification. Nano Lett 2004;4:2163–9.
Hsieh SC, Wang FF, Lin CS, Chen YJ, Hung SC, Wang YJ. The inhibition of osteogenesis
with human bone marrow mesenchymal stem cells by CdSe/ZnS quantum dot
labels. Biomaterials 2006;27:1656–64.
Hu XK, Wang P, Hwang H-M. Oxidation of anthracene by immobilized laccase from
Trametes versicolor. Bioresource Technol 2009;100:4963–8.
Hwang H-M, Balarezo AL, Jones VN, Yu H. Effect of river humic acid on 1-aminopyrene
ecotoxicity in a dynamic solar photolysis process. Bull Environ Contam Toxicol
2004;72:1059–66.
Rijnkels JM, Moison RM, Podda E, van Henegouwen GM. Photoprotection by
antioxidants against UVB-radiation-induced damage in pig skin organ culture.
Radiat Res 2003;159:210–7.
Jones CF, Grainger DW. In vitro assessments of nanomaterial toxicity. Adv Drug Deliv Rev
2009;61:438–56.
Yavuz CT, Mayo JT, Yu WW, Prakash A, Falkner JC, Yean S, et al. Low-field magnetic
separation of monodisperse Fe3O4 nanocrystals. Science 2006;314:964–7.
Lewinski N, Colvin V, Drezek R. Cytotoxicity of nanoparticles. Small 2008;4:26–49.
Ngamwongsatit P, Banada PP, Panbangred W, Bhunia AK. WST-1-based cell cytotoxicity
assay as a substitute for MTT-based assay for rapid detection of toxigenic Bacillus
species using CHO cell line. J Microbio Meth 2008;73:211–5.
4481
Oberdorster G, Maynard A, Donaldson K, Castranova V, Fitzpatrick J, Ausman K, et al.
Principles for characterizing the potential human health effects from exposure to
nanomaterials: elements of a screening strategy. Part Fibre Toxicol 2005;2:8.
Pulskamp K, Diabaté S, Krug HF. Carbon nanotubes show no sign of acute toxicity but
induce intracellular reactive oxygen species in dependence on contaminants.
Toxicol Lett 2007;168:58–74.
Reimer P, Weissleder R. Development and experimental application of receptor-specific
MR contrast media. Radiology 1996;36:153–63.
Salem AK, Searson PC, Leong KW. Multifunctional nanorods for gene delivery. Nat
Mater 2003;2:668–71.
Su CH, Sheu HS, Lin CY, Huang CC, Lo YW, Pu YC, et al. Nanoshell magnetic resonance
imaging contrast agents. J Am Chem Soc 2007;129:2139–46.
Thiesen B, Jordan A. Clinical applications of magnetic nanoparticles for hyperthermia.
Int J Hyperthermia 2008;24:467–74.
University of California-San Diego. March 29 Widely used iron nanoparticles exhibit
toxic effects on nerve cells. ScienceDaily; 2007. Retrieved from http://www.
sciencedaily.com/releases/2007/03/070328155553.htm.
Weissleder R, Bogdanov A, Neuwelt EA, Papisov M. Long-circulating iron oxides for MR
imaging. Adv Drug Delivery Rev 1995;16:321–34.
Worle-Knirsch JM, Pulskamp K, Krug HF. Oops they did it again! Carbon nanotubes
hoax scientists in viability assays. Nano Lett 2006;6:1261–8.
Yamawaki H, Iwai N. Cytotoxicity of water-soluble fullerene in vascular endothelial
cells. Am J Physiol Cell Physiol 2006;290:C1495–502.
Yin H, Too HP, Chow GM. The effects of particle size and surface coating on the
cytotoxicity of nickel ferrite. Biomat 2005;26:5818–26.
Yu MK, Jeong YY, Park J, Park S, Kim JW, Min JJ, et al. Drug-loaded superparamagnetic
iron oxide nanoparticles for combined cancer imaging and therapy in vivo. Angew
Chem Int Ed 2008;47:5362–5.
Zhang Y, Hwang H-M, Ekunwe S. Comparing cytotoxicity and genotoxicity in HaCat
cells caused by 6-aminochrysene and 5, 6-chrysenequinone under UVA irradiation.
Environ Toxicol Chem 2006;25:1920–6.
Zheng BY, Hwang H-M, Yu HT, Ekunwe S. DNA damage produced in HaCaT cells by
combined fluoranthene exposure and ultraviolet A irradiation. Environ Mol
Mutagen 2004;44:151–5.