Nanotechnology

Nanotechnology 30 (2019) 055101 (11pp) https://doi.org/10.1088/1361-6528/aaedd5

Use of DNA-generated gold nanoparticles to

radiosensitize and eradicate radioresistant
glioma stem cells
Tatsuki Kunoh 1,2,4,7, Tsutomu Shimura3,7, Tomonari Kasai1,2,5,7 ,
Syuji Matsumoto1,2,6, Hafizah Mahmud2, Apriliana Cahya Khayrani2,
Masaharu Seno1,2 , Hitoshi Kunoh1,2 and Jun Takada1,2
1
Core Research for Evolutionary Science and Technology (CREST), Japan Science and Technology
Agency (JST), 3-1-1 Tsushima-naka, Kita-ku, Okayama, 700-8530, Japan
2
Graduate School of Natural Science and Technology, Okayama University, 3-1-1 Tsushima-naka, Kita-
ku, Okayama, 700-8530, Japan
3
Department of Environmental Health, National Institute of Public Health, 2-3-6 Minami, Wako, Saitama,
351-0197, Japan

E-mail: kuno.tatsuki.gb@u.tsukuba.ac.jp, simura.t.aa@niph.go.jp and kasaitnr@stf.teu.ac.jp

Received 26 August 2018, revised 25 October 2018

Accepted for publication 2 November 2018
Published 30 November 2018

Abstract
The surface reactivity of gold nanoparticles (AuNPs) is receiving attention as a radiosensitizer of
cancer cells for radiation therapy and/or as a drug carrier to target cells. This study demonstrates
the potential of DNA-AuNPs (prepared by mixing calf thymus DNA with HAuCl4 solution) as a
radiosensitizer of human glioma cells that have cancer stem cell (CSC)-like properties, to reduce
their survival. CSC-like U251MG-P1 cells and their parental glioblastoma U251MG cells are
treated with a prepared DNA-AuNP colloid. The radiosensitivity of the resultant AuNP-
associated cells are significantly enhanced. To reveal the mechanism by which survival is
reduced, the generation of reactive oxygen species (ROS), apoptosis induction, or DNA damage
in the cells is assayed using the fluorescent dye DCFDA, annexin V-FITC/PI, and foci formation
of γ-H2AX, respectively. X-ray irradiation with administration of AuNPs overcomes the
radioresistance of U251MG-P1 cells. It does not induce ROS generation or apoptosis in the cells
but enhances the number of abnormal nuclei with abundant γ-H2AX foci, which is judged as cell
death by mitotic catastrophe. The AuNP association with the cells effectively induces mitotic
catastrophe in x-ray-irradiated CSC-like cells, implicating that DNA-AuNPs might be a
promising tool to develop an efficient radiosensitizer against CSC.
Keywords: DNA-generated AuNPs, x-ray-irradiation, radiosensitizer, glioma stem cells, mitotic
catastrophe

(Some figures may appear in colour only in the online journal)

1. Introduction
4
Present address: Faculty of Life and Environmental Sciences, University of Nanoparticles are at the leading edge of nanotechnology, defined
Tsukuba, Tsukuba, Ibaraki, 305-8572, Japan.
5
Present address: School of Bioscience and Biotechnology, Tokyo
as the study of structures in the range of 1–100 nm in size [1, 2].
University of Technology, 1404-1 Katakuramachi, Hachioji, Tokyo 192- Due to their distinctive physicochemical and electrical proper-
0982, Japan. ties, nanoparticles attract the attention of scientists and engineers
6
Present address: Nanotechnology Hub, Kyoto University, Yoshida-
Honmachi, Sakyo-ku, Kyoto 606-8501, Japan.
in biomedical, optical, catalytic, and electronic fields [1, 2]. To
7 date, a number of metal nanoparticles composed of gold, silver,
Authors to whom any correspondence should be addressed.

0957-4484/19/055101+11$33.00 1 © 2018 IOP Publishing Ltd Printed in the UK

Nanotechnology 30 (2019) 055101
copper, and platinum have been synthesized by physical and
chemical methods [3–7]. However, these conventional methods
often have problems such as being costly, environmentally
hazardous, or otherwise toxic [8]. Thus, biological production
based on the principles of green chemistry have been explored.
For example, extracts from organisms (brown algae, lemongrass,
tea, and fruits), human and microbial living cells, and biomo-
lecules such as proteins and sugars have been used to create gold
nanoparticles (hereafter AuNPs) [9–18]. Our previous study
demonstrated that incubation of DNA in HAuCl4 solution suc-
cessfully led to form spherical AuNPs with a size of approxi-
mately 5 nm in diameter (hereafter DNA-generated AuNPs)
through oxidation of guanine moiety of DNA, implicating the
production of Au-DNA complex [19].
Several lines of evidence revealed that size, shape, and
surface functionalization of AuNPs are key factors regarding
their internalization by human cells [18]. DNA-AuNPs have
the ability to enter a wide variety of cell types [18]. Inter-
estingly, large DNA fragments are more readily incorporated
into cancer cells such as liver, lung, and breast cancer cells
than into normal cells [20]. Moreover, AuNPs delivered to
cancer cells successfully reduced x-ray irradiation dose
required for DNA damage dependent cell death due to effi-
cient generation of photoelectrons and Auger electrons upon
x-ray irradiation of AuNPs [21]. For such medical applica-
tions, generally AuNPs are functionalized with multiple tar-
geting molecules such as ligands, antibodies, and aptamers
[18, 22, 23]. Once these modified AuNPs are delivered to
target cells and tissues, they can readily act as an effective
radiation sensitizer upon x-ray irradiation [24–26]. In aqueous
conditions, x-ray irradiation of AuNPs causes water ioniz-
ation to result in formation of reactive oxygen species (ROS),
hydroxyl radicals (·OH) and superoxide anions (O2-) [27]. In
mammalian cells into which AuNPs have been incorporated,
ROS is also generated intracellularly upon x-ray irradiation
[25, 28]. Radiation-induced ROS generation oxidatively
damages cellular components such as DNA, RNA, proteins,
and lipids, which may trigger cell death [29, 30]. Indeed, the
extent of DNA damage from x-ray irradiation influences the
eventual cell fate, e.g. cell cycle arrest, DNA repair, and/or
apoptotic pathway activation [31]. These earlier findings led
us to fascinate a possibility of using DNA-generated AuNPs
for medical applications such as cancer radiotherapy without
further surface modifications. However, we do not yet know
for certain whether these AuNPs are toxic to living organisms
or are safe for medical applications on human cells.
Increasing evidence shows that many types of tumors
contain a subpopulation of cells called cancer stem cells
(hereafter CSCs) [32–34]. Similar to normal pluripotent stem
cells, CSCs have characteristic properties such as self-
renewal, multi-differentiation, resistance to radiotherapy, and
cytotoxic chemotherapy due to their dormant state, pre-
ferential activation of the DNA damage response, efficient
DNA repair machinery, and resistance to apoptosis [34, 35].
In addition, CSCs show ROS resistance by low intracellular
ROS concentrations due to up-regulation of ROS scavengers
2
T Kunoh et al
[36]. Since their ability to reconstitute tumors is closely
associated with the treatment resistance and tumor recurrence,
energetic studies to focus CSCs would open effective ther-
apeutic approaches to treat and suppress various types of
cancer [34–36]. Therefore, in this study we primarily focused
to study whether use of DNA-generated AuNPs [19] con-
tributes to abrogate radioresistance, one of CSCs properties.
Herein, we first examined the cytotoxicity of DNA-
generated AuNPs in human U251MG-P1 cells (the derivative
of its parental U251MG cells) that have CSC properties.
Subsequently, the radiation sensitivity of the cells and their
parental cancer cells, U251MG cells, associated or unasso-
ciated with AuNPs, was compared to verify whether the
AuNPs association conferred any potential for radio-
sensitization to the cells. Finally, we explored ways to
influence the survival rates of these cells.
2. Results
2.1. Microscopic analyses of colloidal DNA-AuNPs and
assessment of toxicity to human cells
When calf thymus DNA was incubated in HAuCl4 solution,
size-controlled nanoparticles were formed in this mixture
(hereafter referred to as colloid), as reported previously [19].
The average diameter of the basic AuNPs prepared by the
present condition was ca. 4.5 nm on average [19]. The larger
particles, aggregations of the basic AuNPs, existed on and
near the surface of DNA molecules (figure 1(a), left). More-
over, apart from the aggregated DNA, a number of similar
aggregates were scattered on the carbon grid (figure 1(a),
right). As discussed below, solitary-looking AuNPs apart
from the aggregated DNA in the TEM images are supposed to
associate with non-aggregated DNA molecules. However,
even in highly magnified TEM images, it was difficult to
determine how these AuNPs were associated with aggregated
or non-aggregated DNA molecules. We therefore lump all
these particles as ‘DNA-AuNPs’ hereafter.
The particle density of the DNA-AuNPs in the colloid
was estimated using liquid TEM. Again, the aggregates of
DNA-AuNPs (30–40 nm in diameter) were dispersed nearly
homogeneously in the colloid (figure 1(b)). On average, 533
particles were distinguished in 2.5×2.5×0.2 μm3 of the
colloid image; thus, the density of DNA-AuNPs in the pre-
pared colloid was estimated as ca. 8.8×1011 μl−1.
We next examined whether the colloid was toxic to
U251MG-P1 cells. The colloid was serially diluted with the
culture medium, added to well cultures of U251MG-P1 cells,
and incubated for 24 h. Subsequently, the number of cells in
each well was determined using the MTT assay. The relative
cell viability against that of the colloid-untreated control
culture was slightly higher than 100% when the particle
density was lower („1.5×1010 particles μl−1), but was only
ca. 60% when particle density was higher (…2.9×1011
particles μl−1) (figure 1(c)); thus, the low density treatment
with DNA-AuNPs was not toxic to the U251MG-P1 cells.
Nanotechnology 30 (2019) 055101 T Kunoh et al

Figure 1. Microscopic analyses of colloidal DNA-AuNPs and verification of their toxicity to human cells. (a) TEM images of air-dried DNA-
AuNP colloid on aggregated DNA (left) and carbon grid frame (right), respectively. (b) Liquid TEM image of DNA-AuNP colloid, showing
nearly homogeneous distribution of electron-dense nanoparticles. (c) Relative viability of U251MG-P1 cells after incubation with serially
diluted DNA-AuNP colloids against that of PBS-treated control cells (MTT assay). (d) DIC images of U251MG-P1 cells treated with PBS
(left) and with DNA-AuNP colloid (right) after the silver enhancement stain for AuNPs. Note that dark brown particles (corresponding to
AuNPs) were associated with the latter cells only. Some particles were incorporated intracellularly, judging from the focal plane.

3
Nanotechnology 30 (2019) 055101 T Kunoh et al

Figure 2. Sensitivity of DNA-AuNP colloid-treated and -untreated U251MG and U251MG-P1 cells to x-ray irradiation. Error
bars=standard deviations. (a) Survival of U251MG (circle) and U251MG-P1 (square) cells treated with PBS after x-ray irradiation.
Survival of U251MG cells was notably lower than that of U251MG-P1 cells at 1–5 Gy. (b) Survival of U251MG (circle) and U251MG-P1
(square) cells treated with DNA-AuNP colloid after x-ray irradiation. Survival of U251MG cells was comparable to that of U251MG-P1 cells
at 1–5 Gy. Note that x-ray resistance, the CSC-like property, of U251MG-P1 cells was higher than that of their parental U251MG cells (a).
Importantly, this property was abrogated when the cells were associated with DNA-AuNPs (b). (c) DIC images of U251MG-P1 cells treated
with PBS (left) and with DNA-AuNP colloid (right two) after silver enhancement stain of AuNPs. Note that localization of AuNPs was not
notably affected by x-ray irradiation.

To examine how DNA-AuNPs behave when the colloid
encounters with U251MG-P1 cells, we incubated the cells
with the colloid (1.5×1010 DNA-AuNPs μl−1) or with PBS
as the control. DNA-AuNPs in these mixtures were sensitized
by the silver enhancement method (figure 1(d)). Dark brown
particles were observed in all of the cells incubated with the
colloid (figure 1(d), right), but not in the control cells
(figure 1(d), left). Judging from the focal plane in which these
particles were observed, some of the particles were likely to
be inside the cells, while others were on the cell surface
(figure 1(d)). Thus, from the optical images, we could not
definitely conclude that all of the DNA-AuNPs had been
incorporated into the cells. Therefore, to avoid confusion, we
hereafter call these cells as ‘AuNP-associated’ cells.
2.2. Sensitivity of AuNPs-associated U251MG-P1 cells to x-ray
irradiation
As described above, U251MG-P1 cells originated from a
primary culture from the tumor of a mouse that had been
injected with a spheroid-forming population of human brain
glioblastoma U251MG cells. In a colony assay, the survival
rates of U251MG and U251MG-P1 cells were compared after
treatment with x-rays followed by PBS relative to those of the
non-irradiated control cells. The survival rate of U251MG-P1
cells was higher than that of U251MG cells (figure 2(a)) at all
doses, showing that the U251MG-P1 cells were apparently
more resistant to x-ray irradiation, probably due to their CSC-
like properties.
We next examined whether the association of these cells
with AuNPs influenced their survival rate after x-ray irra-
diation. At first, we confirmed visually that x-ray irradiation
did not at least localize AuNPs (figure 2(c)). The colony assay
after x-ray irradiation showed no difference between the
survival rate of both AuNPs-associated U251MG and
U251MG-P1 cells at all radiation doses (1–5 Gy)
(figure 2(b)). Thus, the radioresistant phenotype of U251MG-
P1 cells disappeared by administration with AuNPs
(figures 2(a), (b)). In addition, the survival rate of AuNPs-
associated U251MG-P1 cells (figure 2(b)) was significantly
(P<0.05) lower than even that of the PBS-treated U251MG
cells (figure 2(a)). From these results, we concluded that
association with AuNPs efficiently abrogated the x-ray
radiation resistance related to the CSC-like properties,

4
Nanotechnology 30 (2019) 055101
Figure 3. Influence of the AuNP association and x-ray irradiation on
ROS levels as reflected by relative fluorescence in DNA-AuNP- or
PBS-treated U251MG and U251MG-P1 cells at 24 h after x-ray
irradiation (FACS analysis of DCFDA-stained cells). In PBS-treated
U251MG cells, ROS generation was enhanced by irradiation, but
none was generated in PBS-treated U251MG-P1 cells. ROS
generation in AuNP-associated U251MG-P1 cells was lower than in
PBS-treated cells. ROS generation in both types of AuNP-associated
cells was reduced by 5 Gy irradiation. Error bars=standard
deviations. Double asterisks=significant differences at P<0.01.
probably due to cell death such as apoptosis induction and
mitotic catastrophe.
2.3. Influence of the AuNP association and x-ray irradiation on
ROS levels in U251MG and U251MG-P1 cells
The dose-dependent reduction of survival rates of U251MG and
U251MG-P1 after x-ray irradiation (figure 2) suggested that the
irradiation caused adverse effects that were enhanced by the
association of these cells with AuNPs. In PBS-treated U251MG
cells, the relative fluorescence (=ROS generation) was sig-
nificantly (P<0.01) enhanced by x-ray irradiation at 5 Gy,
while ROS generation was unaffected in PBS-treated U251MG-
P1 cells even after 5 Gy irradiation (figure 3). In addition, ROS
generation in AuNPs-associated U251MG-P1 cells was sig-
nificantly (P<0.01) lower than that in the control PBS-treated
cells (figure 3). Furthermore, ROS generation after 5 Gy irra-
diation was significantly (P<0.01) lower in AuNP-associated
than PBS-treated U251MG cells. These results suggest that the
ROS generation in both groups of cells could be suppressed only
when the cells are associated with the AuNPs. The additional
result that ROS generation in the AuNPs-associated U251MG
and U251MG-P1 cells was significantly (P<0.01) reduced after
irradiation at 5 Gy relative to that in both PBS-treated cells
(figure 3) led us to conclude that ROS generation was unable to
account for the reduced survival of AuNP-associated U251MG-
P1 cells after irradiation.
2.4. Influence of the AuNP association and x-ray irradiation on
apoptosis induction in U251MG and U251MG-P1 cells
As described above [29–31], x-ray irradiation adversely
affects human cells by arresting the cell cycle and/or
5
T Kunoh et al
activating apoptotic pathways. To determine whether apop-
tosis induction was enhanced in AuNP-associated U251MG-
P1 cells after x-ray irradiation, we used FACS analysis after
staining with annexin V-FITC and propidium iodide (PI) [37].
In the analytical diagram in figure 4(a), zones B1, B2, B3, and
B4 correspond to populations of necrotic cells, apoptotic cells
at the late stage, vital cells, and apoptotic cells at the early
stage, respectively. When the percentage of apoptotic cells in
zones B2 and B4 was compared against the total analyzed
cells in all zones (figure 4(b)), regardless of the irradiation
doses, an average of 3%–5% of the PBS-treated control cells
were apoptotic, whereas 7%–8% the AuNPs-associated cells
were apoptotic on average (figure 4(b)). However, the sta-
tistical analysis revealed that the percentage of apoptotic cells
between these two cell groups did not differ significantly at
any dose (P>0.05). Thus, these results led us to conclude
that association with DNA-AuNPs did not enhance the x-ray
irradiation-dependent apoptotic cell death in U251MG-P1
cells.
2.5. Formation of abnormal nuclei representing mitotic
catastrophe in AuNP-associated cells after x-ray irradiation
Since x-ray irradiation is known to cause DNA damage [31],
we used γ-H2AX staining to monitor DNA double-strand
breaks (hereafter DSBs) to determine whether AuNPs affec-
ted DNA repair in U251MG and U251MG-P1 cells [38].
Faint γ-H2AX staining was observed in non-irradiated PBS-
treated U251MG and U251MG-P1 cells (figures 5(a), (c),
upper left, respectively). However, in both cells, intense
fluorescence representing the remaining DSBs was detected
24 h after 5 Gy irradiation (figures 5(a), (c), lower left,
respectively). Similar florescence detection was obtained in
both x-ray-irradiated and non-irradiated AuNPs-associated
U251MG and U251MG-P1 cells (figures 5(b), (d), respec-
tively). Here, cells with intense fluorescence, representing
serious DNA damage, were regarded as γ-H2AX-positive,
and counted among U251MG and U251MG-P1 cells that
were PBS-treated or AuNPs-associated before x-ray irradia-
tion. As shown in figure 5(e), the percentage of positive cells
increased to ca. 60% in PBS-treated and AuNPs-associated
U251MG after x-ray irradiation. Overall, the incidence of the
positive cells was significantly lower (P<0.01) in U251MG-
P1 cells than in U251MG cells. An earlier report [39] that
CSC cells can efficiently self-repair DNA damage caused by
x-ray irradiation and/or AuNPs association may account for
such a difference. However, the percentage was ca. 40% after
irradiation in PBS-treated U251MG-P1 cells, but increased to
ca. 60% in AuNPs-associated U251MG-P1 cells after irra-
diation (figure 5(e)). Thus, AuNPs are effective to abrogate
efficient repair of x-ray irradiation-induced DSBs in CSC-like
U251MG-P1 cells.
The right columns in figures 5(a)–(d) show nuclei stained
with Hoechst. Note that small fragmented DNAs (=micro-
nuclei) with γ-H2AX staining were detected in both AuNP-
associated U251MG and U251MG-P1 cells after x-ray irra-
diation (figures 5(b), (d), lower right, respectively, arrows).
Although multinucleated cells were evident among both
Nanotechnology 30 (2019) 055101 T Kunoh et al

Figure 4. Incidence of apoptotic cells among AuNP-associated and PBS-treated U251MG-P1 cells after x-ray irradiation (FACS analysis).
(a) Distribution of PBS-treated cells (upper two) and AuNP-associated cells (lower two) in the respective zones after x-ray irradiation (B1,
necrotic cells; B2, late-stage apoptotic cells; B3, viable cells; B4 zones, early-stage apoptotic cells). (b) Incidence of apoptotic cells (total
number of cells in zones B2 and B4) in the PBS-treated and AuNP-associated cells after x-ray irradiation. Error bars=standard deviations.
Overlapping of all standard error bars suggests no difference among incidences (P>0.05).

AuNP-associated U251MG and U251MG-P1 cells after x-ray
irradiation (figure 5(f)), the number of multinucleated cells in
U251MG-P1 cells differed significantly (P<0.01) from the
control.
3. Discussion
The TEM image of the DNA-AuNPs colloid showed aggre-
gates of multiple electron-dense nanoparticles (=AuNPs with
∼5 nm diameter) on or near the surface of aggregated DNA
(figure 1(a), left). In addition, some AuNPs seemed to exist
solitarily apart from aggregated DNA without coagulation
(figure 1(a), right). Therefore, for future investigation of
practical therapeutic techniques, it is important to elucidate
whether and/or how the association of DNA molecules with
the surface of these particles contribute to enhance radio-
sensitization activity for CSC cells.
Notably, the treatment with a relatively low density of
AuNPs in the colloid („1.5×1010 particles μl−1) was not
toxic to target human cells (figure 1(c)), although we do not
know why the higher density of DNA-AuNPs adversely
affected the cells. Results were comparable to those of human
cells after treatment with AuNPs that had been surface-coated
with glucose or cetyltrimethylammonium bromide (CTAB)
[36, 40]. As demonstrated in figure 1(d), at least some of the
particles were likely to be inside the cells, while others might
have localized on the cell surface.
We subsequently examined the radiation sensitivity of
the U251MG-P1 and U251MG cells, associated or unasso-
ciated with AuNPs, and the potential of the AuNP association
to enable radiosensitization to the cells. The U251MG-P1
cells were known to have CSC-like properties such as higher
expression of CD44 (a marker for CSCs) and aberrant acti-
vation of principal pluripotency genes (OCT3/4, SOX2,
KLF4, and Nanog) in comparison with those of the parent
U251MG cells [41, 42]. As expected, the U251MG-P1 cells
were thus apparently more resistant to x-ray irradiation
(figure 2(a)), being consistent with other CSCs [34, 35].
Based on these results, we subsequently examined the
radiation sensitivity of the AuNP-associated or -unassociated
U251MG-P1 and U251MG cells and found that the survival
rate of both AuNP-associated U251MG and U251MG-P1
cells was close enough at all radiation doses (1–5 Gy)
(figure 2(b)). Thus, the association of U251MG-P1 cells with
AuNPs efficiently abrogated the radiation x-ray resistance
related to CSC-like properties. We infer that such effects
could be related to AuNP-mediated ROS generation in the
x-ray-irradiated cells [43]. In addition, radiation-induced ROS
generation damages cellular components, which may trigger
cell death [29, 30]. Contrary to what we expected, the
experimental results (figure 3) did not show that ROS gen-
eration accounted for the reduced survival of AuNP-asso-
ciated U251MG-P1 cells after irradiation. Notably, ROS
generation was apparently lower in U251MG-P1 than
U251MG cells, regardless of the AuNP association and

6
Nanotechnology 30 (2019) 055101 T Kunoh et al

Figure 5. Incidence of DNA double-strand breaks (DSBs) and abnormal nuclei caused by x-ray irradiation as detected by foci formation of
γ-H2AX and Hoechist staining, respectively. (a) Weak (left upper) and intense (left lower) fluorescence representing DSBs in PBS-treated
U251MG cells before (0 Gy) and after (5 Gy) irradiation, respectively. Images in the right column show normal nuclei before and after x-ray
irradiation. (b) DSBs (left) and nuclei (right) in AuNP-treated U251MG cells before and after x-ray irradiation. Note the micronuclei that
formed after irradiation (right lower, arrows). (c), (d) DSBs (left) and nuclei (right) in PBS-treated and AuNP-associated U251MG-P1 cells,
respectively, before and after irradiation. Note that γ-H2AX-positive micronuclei were formed after irradiation (right lower, arrows).
(e) Percentage of intensely fluorescing cells (γ-H2AX-positive cells) among PBS-treated and AuNP-associated U251MG and U251MG-P1
cells before and after irradiation. (f) Incidence of abnormal nuclei accompanying micronuclei in PBS-treated and AuNP-associated U251MG
and U251MG-P1 cells before and after irradiation. Note that the percentage significantly (P<0.05) increased in PBS-treated U251MG-P1
cells after irradiation and especially (P<0.01) in AuNP-associated U251MG and U251MG-P1 cells. Error bars=standard deviations.
Double and single asterisks=significant differences at P<0.01 and <0.05, respectively.

7
Nanotechnology 30 (2019) 055101
-unassociation of the cells (figure 3) as supported by an earlier
paper [44]. Thus, the activity of free radical scavenging sys-
tems might be increased in the PBS-treated U251MG-P1
cells. Resistance to ROS was also reported in other CSCs due
to up-regulation of ROS scavengers [35]. Interestingly, the
incidence of intense γ-H2AX-positive fluorescence, which
represents serious DNA damage (= DSBs), was lower in
U251MG-P1 cells than in U251MG cells. CSC cells can
efficiently self-repair DNA damage caused by x-ray irradia-
tion and/or association with AuNPs [39], which may account
for the weaker fluorescence in U251MG-P1 cells.
Because the extent of DNA damage from x-ray irradia-
tion influenced cell cycle arrest, DNA repair, and/or apop-
totic pathway activation [31], we next examined whether
x-ray irradiation-induced apoptosis in U251MG-P1 cells. The
FACS analysis, which distinguished dead (necrotic or apop-
totic) and vital AuNPs-associated and PBS-treated control
U251MG-P1 cells after x-ray-irradiation failed to show any
significant differences in the incidence of apoptosis at any
dose (figure 4). Thus, association with DNA-AuNPs did not
enhance x-ray irradiation-dependent apoptotic cell death in
U251MG-P1 cells.
Finally, we found that the incidence of abnormal cells
with fragmented DNA (=micronuclei) increased in both
AuNPs-associated U251MG and U251MG-P1 cells after
x-ray irradiation (figure 5). This phenomenon was more
apparent in U251MG-P1 cells. The abnormal nuclei closely
resemble those in dead cells due to mitotic catastrophe
[39, 45, 46]. DSBs detected by γ-H2AX staining were evident
in cells undergoing mitotic catastrophe. This may reflect that
remaining DSBs cause mitotic failure by premature or inap-
propriate entry of cells into mitosis due to defective cell cycle
checkpoints [45]. On the basis of these results, we consider
that the CSC-like properties of U251MG-P1 cells were
abrogated by association with AuNPs through mitotic cata-
strophe rather than ROS generation and apoptosis induction,
thus reducing survival after x-ray irradiation. The intracellular
location of AuNPs is known to affect radiation enhancement.
Particularly, AuNPs attached to DNA have a greater impact
than an AuNP in other locations [47]. Therefore, the defor-
mation of nuclei due to the AuNPs association (figure 5) most
likely reflects that DNA-AuNPs which were incorporated into
the target cells directly or indirectly interacted with chromo-
some DNA, although further detailed studies are required for
a definitive conclusion.
As mentioned above, CSC cells are known to be strongly
resistant to x-ray irradiation, which often poses a problem for
cancer therapy [34]. However, the present findings demonstrated
that the association of these cells with DNA-generated AuNPs
reduced viability of U251MG-P1 cells after x-ray irradiation,
plausibly through DSBs, which are critical lesions that can result
in cell death or a wide variety of genetic alterations including
large- or small-scale deletions, heterozygosity loss, translocations,
and chromosome loss [48]. DSBs often lead to mitotic cata-
strophe [45, 46]. Although the present study failed to explain the
mechanism underlying this mitotic catastrophe in x-ray-irradiated
8
T Kunoh et al
AuNPs-associated CSC cells (U251MG-P1 cells), the exper-
imental outcomes lead us to expect that DNA-generated AuNPs
are a promising tool for radiotherapy treatment of cancerous cells
and tissues.
4. Conclusion
In this study, we initially found that the association with
DNA-AuNPs (which were incorporated into and/or attached
to cells) were not toxic to radioresistant glioma stem cells
with CSCs-properties. Then we found that the pretreatment of
the cells with such AuNPs abrogated radioresistance of the
CSC-like cells. The subsequent cell cycle analyses and
fluorescence microscopy revealed that such a suppression of
the radioresistance of CSC-like cells was due to an increase in
the number of abnormal nuclei with abundant γ-H2AX foci (a
marker of DNA double-strand breaks), which was judged to
be cell death by mitotic catastrophe. Results led us to consider
that DNA-generated AuNPs could be a promising material to
enhance radiosensitization of CSCs-like glioma stem cells.
We would like to emphasize novel merits to using the
present DNA-generated AuNPs: (i) preparation of the DNA-
AuNPs colloid is quite simple (just incubation of DNA in
HAuCl4 solution at room temperature for 16 h, followed by
neutralization in PBS buffer) [19], (ii) spherical DNA-gen-
erated AuNPs (∼5 nm in diameter) are readily associated with
cancer cells inculuding CSC-like glioma stem cells without
using any transfection agent and interact with their chromo-
some DNA, as expected from an earlier report [20]. The
development of the novel, practical, efficient technique to
abrogate radioresistance of CSC-like cells are still under
intensive investigation based on the present outcome.
5. Experimental procedure
5.1. Cell culture
U251MG-P1 cells were obtained as a primary culture from a
tumor-bearing mouse that had been injected with a spheroid-
forming population of human glioblastoma U251MG cells
[41]. These cells were cultured in DMEM high glucose
medium (Sigma-Aldrich, St. Louis, MO, USA) supplemented
with 10% fetal bovine serum (FBS) (Thermo Fisher Scien-
tific, Waltham, MA, USA) and 100 U ml–1 penicillin/strep-
tomycin (P/S) (Wako Pure Chemical Industries, Osaka,
Japan) at 37 °C in a 5% CO2 incubator.
5.2. Preparation of DNA-AuNP colloid using calf thymus DNA
As described previously [19], 0.5 ml of 100 mM Au(III)
chloride solution (pH 2.0), which had been prepared by dis-
solving HAuCl4 (Sigma-Aldrich) in ultrapure water (UPW),
was added to 3.5 ml of calf thymus DNA (Sigma-Aldrich)
solution (final concentrations of 2.5–7.5 mg ml−1) in 15 ml
conical tubes and incubated on a shaking mixer (SHM-2002,
Nanotechnology 30 (2019) 055101
LMS, Tokyo, Japan) at room temperature for 16 h. The
resultant DNA-AuNP colloid was neutralized by adding
1.0 ml of 20-fold concentrated phosphate-buffered saline
(PBS, pH 7.4) to adjust pH to ca. 7.0 (hereafter referred to as
DNA-AuNP colloid).
5.3. Transmission electron microscopy
To confirm the characteristics of the DNA-AuNP colloid, we
prepared and viewed the DNA-AuNP colloids with electron
microscopy as follows. For TEM observation, the DNA-
AuNP colloid was diluted 200–1000-fold with 99.5% ethanol.
The diluted specimens were dropped onto carbon grids
(Nisshin-EM, Tokyo, Japan), air-dried, then viewed with a
TEM (JEOL 2100 F, Tokyo, Japan). To determine the particle
density in the DNA-AuNPs colloid, we used a Si-based
microchannel sample holder for liquid TEM called a K-kit
(Bio MA-Tek, Hsinchu, Taiwan) according to the manu-
facturer’s protocol. The K-kit is a sealable carrier with a
liquid microchannel (window: thickness=100 nm, width=
24 μm; length=300 μm). Briefly, the DNA-AuNP colloid
was diluted 20 fold and placed in the K-kit to impregnate gaps
(gap size 0.2 μm) in the K-kit. Then the kit was tightly sealed,
set on the TEM specimen holder, and inserted into the TEM
column, then observed according to routine methods. The
electron-dense particles in 10 squares (2.5 μm×2.5 μm) in
TEM images were counted. Theoretically, almost all particles
in the 0.2 μm gap should appear in these squares; thus, the
density of AuNPs per microliter of the colloid was estimated
on the basis of the counted number.
5.4. Preparation of DNA-AuNP-associated human cells
The U251MG-P1 cells were seeded at 5×10 cells in 100 μl
3 cells, AuNPs-associated U251MG-P1 cells were sensitized
of the above culture medium and incubated for 24 h. Subse-
quently, 50 μl of the DNA-AuNP colloid, which was serially
diluted with the culture medium (9.1×109–5.8×1011 par-
ticles μl−1) was added to the above cell cultures and incu-
bated for an additional 24 h.
5.5. X-ray irradiation
To determine the x-ray resistance of the AuNP-associated
cells, the U251MG and U251MG-P1 cells, which had been
incubated with the DNA-AuNP colloid or the control PBS for
>24 h were irradiated with x-rays using a 150 kVp x-ray
generator (model MBR-1505R2, Hitachi) through a 0.5 mm
Cu and 0.1 mm Al filters at a dose of 0.7 Gy min−1. For
reference, the routinely cultured cells that were not incubated
with the DNA-AuNP colloid were also irradiated the same
way. These cells were then subjected to the following colony
assay.
5.6. MTT assay
To monitor toxicity of the DNA-AuNP colloid to human
cells, we determined the viability of U251MG-P1 cells
associated with DNA-AuNPs using a cleavage assay with
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
9
T Kunoh et al
(MTT) (Sigma-Aldrich). The cells that had been incubated
with the DNA-AuNP colloid or the control PBS were treated
with 5 mg ml−1 MTT in PBS (final concentration of
1 mg ml−1) and incubated at 37 °C for 5 h. Formazan crystals
formed in these mixtures during the incubation were dis-
solved in 10% SDS containing 20 mM HCl at 37 °C over-
night. Formazan formation, which depends on the metabolic
activity of vital cells, is frequently used as an indicator of cell
viability [49]. Thus, using a spectrophotometer, we compared
the optical density at 570 nm (peak absorbance for formazan)
of the suspension of colloid-treated cells with that of the PBS-
treated control cells. The toxicity was evaluated by the rela-
tive cell viability.
5.7. Colony assay
Effects of x-ray irradiation on the colony-forming ability of
test cells were examined by seeding 1 cm dishes (Thermo
Fisher Scientific) with 1×103 cells of U251MG or
U251MG-P1, incubating the dishes at 37 °C in a 5% CO2
incubator for 16 h, then irradiating the cells with x-rays as
described above. After an additional 16 h incubation, the cells
were subcultured (cell density: 1×103 cells per new dish)
and incubated for a week, followed by fixation with 99%
ethanol and then staining with Giemsa stain solution (Milli-
pore, Billerica, MA, USA). After air-drying, colonies con-
sisting of more than 50 cells were counted. The survival rate
was evaluated as the relative colony number compared with
that of the non-irradiated control cells.
5.8. Sensitization of AuNPs associated with human cells
To ascertain whether AuNPs were associated with human
using a silver enhancement method for optical microscopic
observation. After 24 h incubation with the DNA-AuNP
colloid, the U251MG-P1 cells were trypsinized and fixed with
2.5% glutaraldehyde at 4 °C overnight. After washed once in
UPW, the cells were adhered to SurePath Precoat slides
(Becton, Dickinson and Company, Franklin Lakes, NJ, USA),
then washed three times with 99.5% ethanol. They were then
incubated with a Silver Enhancement Kit (BBI Solutions,
Cardiff, UK) for approximately 10 min at room temperature
until the sensitized AuNPs were seen as dark brown spots
using an optical microscope. Subsequently, the cells were
washed by UPW to stop the enhancing reaction, stained with
Kernechtrot solution (Sigma-Aldrich) to visualize nuclei, and
observed with differential interference contrast (DIC) optics
and a light microscope using an objective lens (x20, NA0.40)
(BX51 System Microscope, Olympus, Tokyo, Japan).
5.9. FACS analysis
Apoptotic and necrotic cells generated by x-ray irradiation
were identified and quantified using an Annexin V-FITC
Apoptosis Detection Kit (Bio Vision, Mountain View, CA,
USA) and the manufacturer’s protocol. The x-ray-irradiated
AuNPs-associated and PBS-treated control U251MG-P1 cells
were trypsinized, collected, and stained with the annexin
Nanotechnology 30 (2019) 055101
V-FITC and PI. The annexin V-positive apoptotic cells and
PI-positive necrotic cells were counted with a fluorescence-
activated cell sorting system (FACScan, Becton Dickinson,
Mountainview, CA, USA).
Furthermore, to examine the influence of the x-ray irra-
diation on ROS generation, the AuNPs-associated or PBS-
treated U251MG and U251MG-P1 cells were incubated at
37 °C in a 5% CO2 incubator for 24 h after the x-ray irra-
diation, then stained with 20 μM DCFDA (Sigma-Aldrich) in
a minimum essential medium without serum for 30 min, then
trypsinized, and collected as described earlier. ROS genera-
tion was quantified using the FACScan as described pre-
viously [49]. The intensity of the fluorescence due to ROS
generation in all cells was measured three times and expres-
sed as a mean with standard deviation relative to that of non-
irradiated PBS-treated cells.
5.10. Immunofluorescent staining
For detecting foci formation of γ-H2AX (representing sites of
DNA double-strand breaks) after x-ray irradiation, U251MG
and U251MG-P1 cells cultured on cover glasses were fixed in
4% v v–1 paraformaldehye for 10 min at room temperature,
then permeabilized using 0.5% Triton-X in PBS for 10 min
Then, the cells were washed in PBS and incubated in a
blocking solution (PBS supplemented with 5% BSA (Frac-
tion-V)) for 15 min. The cells were treated with a primary
antibody against γ-H2AX (1:500 dilution, Santa Cruz Bio-
technology, Dallas, TX, USA) and then with a secondary
antibody conjugated with Cy-3 (1:500 dilution, Jackson
ImmunoResearch Laboratories, West Grove, PA, USA). After
the cells were counterstained with Hoechst 33258 (4 μg ml−1,
Vector Laboratories, Burlingame, CA USA) (hereafter,
Hoechst) for DNA, images were acquired using a light
microscope equipped with a fluorescence lamp (Model BZ-
X700, Keyence, Osaka, Japan).
5.11. Statistical analyses
All experiments were done at least three times using inde-
pendent samples. Variances among the values derived from
each sample were analyzed with a one-way ANOVA using
Excel. Dunnett’s test was then used to find significant dif-
ferences among the means of values derived from three or
more independent samples.
Acknowledgments
We thank Ms Masumi Furutani, Ms Tomomi Tsukano, and
Mr Haruo Urata of the Central Research Laboratory at
Okayama University Medical School for providing the
method for AuNPs sensitization in human cells. We
appreciate Dr Megumi Sasatani of Hiroshima University for
valuable comments. We also acknowledge Dr Beth E Hazen
for reviewing and editing the English in the manuscript. This
study was financially supported by JST-CREST (2012–2017)
(JT) by JST, and Grant-in-Aid for Challenging Exploratory
10
T Kunoh et al
Research (17K19944) (TS) and for Scientific Research (C)
(16K07116) (T Kasai) by JSPS. This work was also sup-
ported by the Program of the Network-type Joint Usage/
Research Center for Radiation Disaster Medical Science.
ORCID iDs
Tatsuki Kunoh https://orcid.org/0000-0002-8423-2903
Tomonari Kasai https://orcid.org/0000-0002-9346-6664
Masaharu Seno https://orcid.org/0000-0001-8547-6259
References
[1] Salata O V 2004 J. Nanobiotechnol. 2 3
[2] Murray C B, Kagan C R and Bawendi M G 2000 Annu. Rev.
Mater. Sci. 30 545
[3] Sengani M, Grumezescu A M and Rajeswari V D 2017
OpenNano 2 37
[4] Lu Q, Rosen J, Zhou Y, Hutchings G S, Kimmel Y C,
Chen J and Gand Jiao F 2014 Nat Commun. 5 3242
[5] Natsuki J, Natsuki T and Hashimoto Y 2015 Int. J. Mater. Sci.
Appl. 4 325
[6] Stepanov A L, Golubev A N, Nikitin S I and Osin Y N 2014
Rev. Adv. Mater. Sci. 38 160
[7] Santra T S, Tseng F and Barik T K 2014 Am. J. Nano Res.
Appl. 2 5
[8] Shankar S S, Rai A, Ankamwar B, Singh A, Ahmad A and
Sastry M 2004 Nat. Mater. 3 482
[9] Nune S K, Chanda N, Shukla R, Katti K, Kulkarni R R,
Thilakavathi S, Mekapothula S, Kannan R and Katti K V
2009 J. Mater. Chem. 19 2912
[10] Liu B, Xie J, Lee J Y, Ting Y P and Chen J P 2005 J. Phys.
Chem. B 109 15256
[11] Sujitha M V and Kannan S 2013 Spectrochim. Acta A 102 15
[12] Anshup A, Venkataraman J S, Subramaniam C, Kumar R R,
Priya S, Kumar T R, Omkumar R V, John A and Pradeep T
2005 Langmuir 21 11562
[13] Johnston C W, Wyatt M A, Li X, Ibrahim A, Shuster J,
Southam G and Magarvey N A 2013 Nat. Chem. Biol. 9 241
[14] Leng Y et al 2016 Sci. Rep. 6 28900
[15] Engelbrekt C, Sørensen K H, Zhang J, Welinder A C,
Jensena P S and Ulstrupa J 2009 J. Mater. Chem. 19 7839
[16] Wei H, Li B, Du Y, Dong S and Wang E 2007 Chem. Mater.
19 2987
[17] Giljohann D A, Seferos D S, Daniel W L, Massich M D,
Patel P C and Mirkin C A 2010 Angew. Chem. Int. Ed.
49 3280
[18] Kunoh T, Takeda M, Matsumoto S, Suzuki I, Takano M,
Kunoh H and Takada J 2018 ACS Sustain. Chem. Eng.
6 364
[19] Kong Y, Zhang X, Zhao Y, Xue Y and Zhang Y 2017 Biol.
Res. 50 2
[20] Zhang P, Qiao Y, Wang C, Ma L and Su M 2014 Nanoscale 6
10095
[21] Ghosh P, Han G, De M, Kim C K and Rotello V M 2008 Adv.
Drug. Deliv. Rev. 60 1307
[22] Dykman L A and Khlebtsov N G 2011 Acta Nat. 3 34
[23] Liu C J et al 2008 Nanotechnology 19 295104
[24] Liu X, Liu Y, Zhang P, Jin X, Zheng X, Ye F, Chen W and
Li Q 2016 Int. J. Nanomed. 11 3517
[25] Kong T, Zeng J, Wang X, Yang X, Yang J, McQuarrie S,
McEwan A, Roa W, Chen J and Xing J Z 2008 Small 4 1537
[26] Misawa M and Takahashi J 2011 Nanomedicine 7 604
Nanotechnology 30 (2019) 055101
[27] Retif P, Pinel S, Toussaint M, Frochot C, Chouikrat R,
Bastogne T and Barberi-Heyob M 2015 Theranostics 5 1030
[28] Shimura T and Kunugita N 2016 Cell Cycle 15 1410
[29] Kehrer J P and Klotz L O 2015 Crit. Rev. Toxicol. 45 765
[30] Nowsheen S and Yang E S 2012 Exp. Oncol. 34 243
[31] Nair N et al 2015 Sci. Rep. 7 6838
[32] Bao B, Ahmad A, Azmi A S, Ali S and Sarkar F H 2013 Curr.
Protocols Pharmacol. 61 14.25.1–14
[33] Baek S J et al 2015 Oncol. Rep. 34 2233
[34] Shimura T, Noma N, Oikawa T, Ochiai Y, Kakuda S,
Kuwahara Y, Takai Y, Takahashi A and Fukumoto M 2012
Oncogenesis 1 e12
[35] Baumann M, Krause M and Hill R 2009 Nat. Rev. Cancer 8 545
[36] Demchenko A P 2013 Cytotechnology 65 157
[37] Mah L J, El-Osta A and Karagiannis T C Leukemia 24 679
[38] Zheng Y and Sanche L 2009 Radiat. Res. 172 114
[39] Connor E E, Mwamuka J, Gole A, Murphy C J and Wyatt M D
2005 Small 3 325
11
T Kunoh et al
[40] Vaidyanath A, Mahmud H B, Khayrani A C, Oo A K K,
Seno A, Asakura M, Kasai T and Seno M 2017 J. Stem Cell
Res. Ther. 7 4
[41] Mahmud H et al 2018 Int. J. Mol. Sci. 19 659
[42] Klingberg H, Oddershede L B, Loeschner K, Larsen E H,
Loft S and Møller P 2015 Toxicol. Res. 4 655
[43] Diehn M et al 2009 Nature 458 780
[44] Vakifahmetoglu H, Olsson M and Zhivotovsky B 2008 Cell
Death Differ. 15 1153
[45] Castedo M, Perfettini J L, Roumier T, Andreau K,
Medema R and Kroemer G 2004 Oncogene 23 2825
[46] Shrestha S, Cooper L N, Andreev O A, Reshetnyak Y K and
Antosh M P 2016 Jacobs J. Radiat. Oncol. 3 1
[47] Shrivastav M, De Haro L P and Nickoloff J A 2008 Cell Res.
18 134
[48] Mosmann T 1983 J. Immunol. Meth. 65 55
[49] Shimura T, Sasatani M, Kamiya K, Kawai H, Inaba Y and
Kunugita N 2016 Oncotarget 7 3559