Radiation Physics and Chemistry 128 (2016) 134–142

Contents lists available at ScienceDirect

Radiation Physics and Chemistry

journal homepage: www.elsevier.com/locate/radphyschem

Actual questions raised by nanoparticle radiosensitization

Emilie Brun, Cécile Sicard-Roselli n
Laboratoire de Chimie Physique, CNRS UMR 8000, Université Paris Sud, Bât. 350, 91405 Orsay Cedex, France

H I G H L I G H T S

Huge number of nanoparticles show radiosensitizing properties.

The absence of consensus concerning cellular results slows down the translation to clinics.

The relative contributions of physical, chemical and biological steps in radiosensitization still need to be clarified.

art ic l e i nf o a b s t r a c t

Article history: Radiosensitization by metallic nanoparticles (NP) has been explored for more than a decade with pro-
Received 21 April 2016 mising results in vitro and in cellulo reported in a vast number of publications. Yet, few clinical trials are
Received in revised form on-going. This could be related to the lack of selectivity of NP leading to massive quantities to be injected
23 May 2016
to observe an effect but also to the higher degree of complexity than first thought leading to an absence
Accepted 26 May 2016
of consensus probably caused by the lack of standardization in pre-clinical studies. Given the wide panel
Available online 28 May 2016
of NP used, in terms of core nature, size, coating, not to mention of cell lines and irradiation modalities,
Keywords: cross-comparison of data is not a walk in the park. But only a thorough examination could help iden-
Nanoparticles tifying the key parameters and the possible mechanisms involved. This step is crucial as it should provide
Radiotherapy
guidance for designing the most efficient combination NP/radiation and rationally establishing clinical
Radiosensitization
protocols. In this review, we will combine and confront cellular radiosensitization results with in vitro
Radiobiology
and numerical experiments in order to give the more recent vision of this complex phenomenon. We
decided to address a few hot topics such as the influence of the incident radiation energy, the localization
of NP or the so-called “biological” effect. We will highlight that among the barriers to break down, some
are not restricted to the “nano” community: an incontestable support could be offered by the “radiation”
community in the broadest sense.
& 2016 Elsevier Ltd. All rights reserved.

1. Introduction nanotechnologies for medical applications is said to concern three

Nanotechnology appears to be a very attractive field as huge
amount of public and private funds are now devoted to this area.
For example, Google Inc. recently patented several sensor devices
based on an increased sensitivity due to nanoparticles (NP). As
regards public funding, through the Seventh Research and In-
novation Funding Program (FP7), the European Commission in-
vested nearly 3.5 billion euros for 2007–2013 in the “Nanosciences,
Nanotechnologies, Materials and New Production Technologies”
project. As for Horizon 2020, which is the European Union”s lar-
gest ever research endeavor (80 billion euros over 7 years), it ex-
plicitly aims to bridge the gap between nanotechnology research
and markets, especially in health domain. Effective translation of
n
Corresponding author.
E-mail address: cecile.sicard@u-psud.fr (C. Sicard-Roselli).
main areas: therapeutics, diagnostics/imaging and regenerative
medecine, and that all therapeutic niches will potentially benefit
from nanomedecine. Such a large scope is possible due to the in-
finite variety of NP. Just focusing on the nature of the NP, it can be
organic like in micelles, liposomes or dendrimers, inorganic with
an elemental metallic core (silver NP, gold nanorods, …), a semi-
conductor core (CdSe quantum dots), a metal oxide core (Super
Paramagnetic Iron Oxide Nanoparticle, SPION), or organic-in-
organic hybrides. Moreover, they are usually functionalized, as
“naked” NP generally not stable enough in cellulo and in vivo re-
quire a stabilizing coating such as polyethyleneglycol (PEG). And
contrary to smaller molecules, nanomaterials can be considered as
platforms to introduce new functionalities via ligands attached
through their rich surface chemistry: targeting, organelle adres-
sing, drug or gene delivery, theranostic, …(Boisselier and Astruc,
2009; Pedrosa et al., 2015). However, at this point, one should
keep in mind that grafting of a functional ligand is not enough to

http://dx.doi.org/10.1016/j.radphyschem.2016.05.024
0969-806X/& 2016 Elsevier Ltd. All rights reserved.
 E. Brun, C. Sicard-Roselli / Radiation Physics and Chemistry 128 (2016) 134–142
guarantee its recognition by its corresponding receptor as most NP,
once in a biological fluid, get covered by biomolecules forming the
so-called “protein corona” (Brun and Sicard-Roselli, 2014). In the
context of cancer therapy, a few drugs already in use are made of
organic particles (Svenson, 2012) while only preclinical and clin-
ical trials are currently ongoing with metallic NP. Among the
clinical phases approved by the Food and Drug Administration
(FDA), metallic NP are mainly involved in photothermal therapy.
AuroLase trial studies Si-Gold nanoshell combined to near-infrared
laser for treatment of head and neck cancer (clinicaltrials.gov
identifier: NCT00848042) whereas Magnablate I (NCT02033447)
which couples iron magnetic NP to magnetic field aims at inducing
thermal ablation of prostate cancer. Only few clinical phases
combine NP and radiotherapy. Radiosensitization intends to en-
hance cancerous cells killing while preserving normal tissues, as
their tolerance to radiation is the limiting factor of radiotherapy
efficiency. Recent radiosensitization research has focused on pro-
mising drugs that, for instance, alter hypoxia, influence the nature
or repair of DNA damage or inhibit topoisomerases (Linam and
Yang, 2015), but also on high-Z elements (Cooper et al., 2014; Retif
et al., 2015). Historically, iodine and gold have been most studied
(Lehnert, 2014). That is why it is somehow surprising that among
the five referenced trials claiming to be based on radiosensitizing
properties of NP, all concern hafnium oxide (NBTRX3 NP). The first
phase I (NCT01433068) ended in 2014 and is now recruiting for
phases II and III to compare the efficacy of intratumorally injected
NBTRX3 activated by radiotherapy versus radiotherapy alone in
patients with locally advanced soft tissue sarcoma (NCT02379845).
Another phase I/II is ongoing for hepatocellular carcinoma
(NCT02721056) and two others are recruiting in the case of head
and neck (NCT01946867) and rectal (NCT02465593) cancers.
In this review, we will wonder about the difficult translation of
metallic NP as radiosensitizers to clinics. Based on the actual lit-
erature, we will try to rationalize all the information obtained so
far from preclinical but also physico-chemical studies to raise
fundamental questions to be addressed to elaborate an ideal
treatment protocol with NP. Even if central, the questions of the
design of selective NP towards cancerous cells and their in vivo
administration are beyond the scope of this review. We will rather
focus on specific issues as what kind of clinical beam could be the
most appropriate to be combined with NP, or where the NP should
be localized to obtain a maximal effect. We will also examine in
vitro and in silico data to figure out the mechanisms at stage and
try to identify the barriers still to be removed.
2. What do we know from cellular experiments? Description
of the corpus
For this review, we have registered more than 80 research ar-
ticles, published from 2008 to early 2016, dealing with cellular
experiments of radiosensitization by nanomaterials. They will be
the cornerstone of our analysis. We will try to confront them to in
vitro and numerical experiments in order to give the more recent
vision of this complex phenomenon.
First, considering the NP, 53% study gold spherical NP but when
gold (Z ¼79), in the wide way, is concerned, it represents 62.5% of
listed publications. Follow, by far, gadolinium (Z¼ 64, 11%) and
silver (Z¼47, 6%). As mentioned above, report of radiosensitization
by gold dates back to nearly 20 years ago for foil (Regulla et al.,
1998) and microspheres (Herold et al., 2000). The “nano-revolu-
tion” gives it rapidly a second wind, contrary to contrast agents,
because of gold NP ease of synthesis, resulting in controlled size
and shape, versatility of functionalization, high chemical stability
and supposedly biocompatibility (Fratoddi et al., 2015). That is
probably why it ranks first. A plethora of other elements is present
135
in the literature for a few papers each: bismuth (Z¼83), platinum
(Z¼ 78), hafnium (Z¼ 72), cerium (Z¼ 58), cadmium (Z ¼48), se-
lenium (Z ¼34), iron (Z¼ 26), titanium (Z¼22), silicon (Z¼ 14) and
carbon (Z ¼6). Is the paradigm of high Z elements for radio-
sensitization being chipped? The question is worth thinking about.
All the nanomaterials used in these studies are not pure elemental
metal nanospheres. For example, Mirjolet et al. synthesized tita-
nate multi-walled nanotubes of 10 nm length and 4 nm inner
cavity (Mirjolet et al., 2013), Yasui et al. worked with a 100 nm NP
made of a polyamine gel core containing 15 spherical 8 nm gold
NP and tethered PEG chains (Yasui et al., 2014). For Gd, it is worth
s
mentioning that the AGuIX are not gadolinium oxide NP per se
but consist in a polysiloxane core with approximatively 10 cova-
lently linked chelating ligands (DOTA, DTPA, DOTAGA) that can
harvest Gd3 þ ions resulting from the dissolution of the gadolinium
oxide matrix (Le Duc et al., 2014; Sancey et al., 2014). For gold, NP
measure from 1.4 to 80 nm in diameter and a wide range of
coatings are found. They can be classified into small molecules
(thioglucose, tiopronin, cysteamine), polymers (mainly thiolated-
PEG), DNA, proteins and peptides (TAT, RGD, pHLIP, in association
or not with PEG) and ligands of specific receptors (folate, goserelin,
trastuzumab, panitumumab, in association or not with PEG). No
study raises the question of the formation of a protein corona that
could mitigate the expected interactions of the coating with the
cells when it is well accepted that this corona could prevent the
ligand recognition by their receptors (Salvati et al., 2013) and also
has drastic consequences on the species responsible for radio-
sensitization (Gilles et al., 2014). It is then clear that as the FDA
established a definition of a nanoscale product only based on its
size, studies dealing with NP and radiation gather a great variety of
objects which physical interaction with ionizing radiation and
cellular uptake are expected to vary a lot. Also, 45 different cellular
eukaryotic types were submitted to NP and irradiated: 78% with
photons, among which half in the keV and half in the MeV range,
7.5% with ions and 5% with MeV electrons. Therefore, despite a
huge amount of publications care should be taken when trying to
compare them and draw firm conclusions. A few papers deal with
irradiation via radioisotopes but this case will not be addressed in
this review.
It is all the more difficult so as, when thoroughly examining
these publications, we have noted some imprecisions or missing
information in the experimental sections, the most indispensable
to us being the nature of the NP coating (size is almost always
given), the conditions of cell incubation (with or without serum,
duration, extracellular concentration of NP) and the presence or
not of NP in the medium during irradiation. In nearly 20% of the
listed publications, at least one of this information is missing. In
small number of studies, we regret that quantitation of NP in in-
teraction with cells (by ICP-MS for example) were conducted after
exposure conditions different from those retained for irradiation.
Some publications do not report their conditions of irradiation at
all and the dose rate is barely mentioned. For non-monochromatic
X-rays, it would be valuable to give the average photon energy
besides the applied voltage. As highlighted in the following sec-
tions, the radiosensitizing phenomenon is surely more complex
than at first thought and it will be helpful for the whole com-
munity to produce “useful” data. In that sense, conditions resulting
in no radiosensitization must also be reported. Finally, to quantify
radiosensitization, different factors are calculated such as the ratio
of doses required to give the same surviving fraction (for example
90 or 10%) without and with NP, the ratio of the areas under the
curves representing surviving fraction as a function of the dose,
the ratio of surviving fractions at a given dose (not always given),
sometimes taking into account the toxicity of NP
SFIR + NP / SFNP
(SFratio = SFIR
). The adopted notations are not universal and
136 E. Brun, C. Sicard-Roselli / Radiation Physics and Chemistry 128 (2016) 134–142
this can be quite confusing. It would be worth defining un-
ambiguously in each work the quantities calculated and to express
radiosensitization factors in different ways to ease comparisons.
Now, through the detailed reading of the selected publications, we
would like to address some hot points emanating from cellular
experiments: energy dependency of the phenomenon, localization
of NP for radiosensitization and DNA damage. When possible, in
vitro experiments and simulations will be referred to, to shed
some light on these topics. We will finally discuss the possibility of
a “biological” in addition to a “physical” effect to explain radio-
sensitization and we will wonder if a physico-chemical step could
be the connection between both.
3. Is there an energy dependency?
The main rationale for developing metallic NP as radio-
sensitizers, and the first explanation to their radiosensitizing
properties, was their higher mass energy absorption coefficient
compared to water or soft tissues for 10–200 keV photons
(McMahon et al., 2011a). But in most clinical scenarios, radio-
therapy is delivered using MeV photons as their penetration depth
is higher. This led several teams to investigate the impact of the
incident photon energy in keV and MeV regimes on cell radio-
sensitization. If the absolute values of radiosensitizing factors
cannot be compared, over a wide range of NP and cell lines, within
this corpus, the combination with keV photons produces higher
effect than MeV. For example, Geng et al. reported, for 14 nm
thioglucose-capped gold NP and SK-OV-3 cells, survivals of 45%
and 58% for 5 Gy delivered by 90 kVp or 6 MeV (Geng et al., 2011).
For 50 nm citrate-gold NP, ratio of surviving fractions at 4 Gy were
3.94 and 1.42 for 105 kVp and 6 MeV (Chithrani et al., 2010). For
AGuIX NP, Luchette et al. gave dose enhancements of 1.54 and 1.15
for 220 kVp and 6 MeV (Luchette et al., 2014). For Hf, the same
trend was illustrated by Maggiorella (Maggiorella et al., 2012). It
has to be noted nevertheless that some authors did not see any
difference with the controls at MeV energies while they obtained
significant sensitization for keV photons (Khoshgard et al., 2014;
Kong et al., 2008). These experimental results can be corroborated
by simulation studies confirming a more important effect at low
energies. For example, Lechtman et al. obtained a 103 increase in
the rate of photoelectric absorption in the presence of gold NP
comparing ca. 20 keV to 1861 keV photons (Lechtman et al., 2011).
Montenegro et al. calculated a dose enhancement factor (DEF)
with GNP nearly one order of magnitude higher for 68 keV and
82 keV compared to 2 MeV at 50 mg/mL GNP (Montenegro et al.,
2009). Similarly, Jones et al. described a dose enhancement higher
for low energy photon beams compared to 6 MV (Jones et al.,
2010) as Kirkby et al. who also explored the impact of the NP
coating thickness on mitochondria degradation (Kirkby and
Ghasroddashti, 2015). Interestingly, they evidenced a significant
decrease of the dose enhancement ratio while increasing coating
size. This behavior is important as it takes into account a pre-
liminar NP coating designed by chemists or the protein corona
formed as the nano-objects are injected in biological systems.
In the keV region it is noticeable that photoelectric process is
much more favorable than Compton process. This then leads to the
question whether the atomic transition edge has an impact on this
radiosensitizing effect as it was proposed for contrast agents. More
precisely, Monte Carlo simulations have pointed out more efficient
energies in the region 50–250 keV where photoelectric effect
dominates. Lin et al. noticed a sudden increase at 50 keV due to
shorter range of secondary electrons produced in water compared
to electrons emitted from NP and concluded to a dose 15 times
higher at 50 keV versus 250 keV (Lin et al., 2014). Considering
experimental approaches and more precisely, in the keV region,
using DNA as a probe, we demonstrated an unexpected shape of
dose enhancement versus energy with a minimum around 40 keV
where a maximum would be predicted only considering the ratio
between mass energy absorption of gold and water (Brun et al.,
2009; McMahon et al., 2011a). Maximal DNA degradation was
highlighted at ca. 50 keV but also at lower energy, ca. 30 keV.
Guidelli and Baffa also measured by electron spin resonance a
higher DEF near 30 and 40 keV than at 90 or 107 keV (Guidelli and
Baffa, 2014). Few studies have addressed this question in cellulo.
The DEF in Khoshgard et al. do not differ from 120 to 250 kVp,
within the error bars, but the X-rays were not monochromatic
(Khoshgard et al., 2014). On the contrary, Rahman for gold NP
s
(Rahman et al., 2014), Taupin for AGuIX (Taupin et al., 2015) and
Gimenez for iron NP (Gimenez, 2015) described more complex
dependency conducting synchrotron experiments. In Rahman”s
work, two maxima at 40 and 60 keV, which do not correspond to
gold edges, are present confirming DNA studies (Rahman et al.,
s s
2014). Comparing AGuIX and Magnevist , Taupin et al. found that
once normalized, survival fraction ratio SER(4 Gy) ¼f(energy) were
not different for both Gd compounds with a maximum at 65 keV
as expected, reflecting the K-edge. One possible explanation is that
s
AGuIX are not NP with a Gd core but rather a nanosized platform
for Gd chelates (Le Duc et al., 2014). As such, behavior towards
energy would resemble this of Gd atom in both cases. For a same
extracellular concentration, intracellular concentrations of Gd are
s
30 times higher for AGuIX than Magnevists, which could explain
the 1.66 normalization factor. All these data converge to a maximal
effect between 50 and 60 keV, but are unlikely to come to the
same conclusion as it only corresponds to an edge for gadolinium
but not for gold. This energy dependency is definitely a lead to
explore with intermediate energies to confirm the maxima and
with other endpoints to get insights into the complex processes at
stage.
This unexpected energy dependency leads several teams to
investigate more precisely the description of this dose enhance-
ment phenomenon which was first treated at a macroscopic level
but is now considered as a local energy deposition, in-
homogeneous around the NP (McMahon et al., 2011b). The species
emitted by the NP i.e. photo-electrons or Auger electrons, fluor-
escence X photons, are of great interest as their relative proportion
and distribution range could highly affect the quantity and loca-
lization of damages evidenced. Leung et al., (Leung et al., 2011)
calculated with GEANT4 simulation the number of secondary
electrons produced when photon beams interact with GNP. Here
also, low energy photons were found much more efficient than
60
Co and MV photons. Interestingly, the range distribution of these
secondary electrons was calculated as a function of the GNP size,
showing their distribution at distances up to few microns ac-
cording to the conditions. More precisely, MacMahon differ-
entiated the energy deposited by photo-electrons or Auger elec-
trons at different distances from the NP. These Auger electrons are
responsible for energy deposition at a distance less than 250 nm
around a 20 nm GNP irradiated at 40 keV (Lin et al., 2014;
McMahon et al., 2011b). Lechtman (Lechtman et al., 2011) noticed
that photoelectrons and fluorescence X photons contribution de-
creases in proportion compared to Auger electrons when the en-
ergy was increasing. Casta et al., (Casta et al., 2015a, 2014) devel-
oped an experimental setup with XPS to measure the electronic
emission of GNP irradiated at ca. 1.5 keV and compared the data
obtained to simulation. This study clearly pointed out different
behaviors between a gold plane surface and a NP, the coating
impact and the low energies importance. They also developed
Nanop (Casta et al., 2015b), a fast software and of very simple use
that can provide photon and electron emission from X-irradiated
NP, as it is the case with GEANT4-DNA. Description of these pri-
mary species arising from NP is an essential step towards a better
 E. Brun, C. Sicard-Roselli / Radiation Physics and Chemistry 128 (2016) 134–142
understanding of the radiosensitization process. All these results
illustrate the fact that tuning the energy is worth for radio-
sensitization optimization, as emitted species can be responsible
for damages at long distance from the NP. It highlights the fact that
to induce damages on a cellular target, a close proximity is not
necessary.
As high energy electrons and ions are also part of the ther-
apeutic arsenal, a few studies have seized the question. No clear
effect has been seen with protons of 3 MeV and 6 or 50 nm gold
NP (Jeynes et al., 2014; Liu et al., 2010) though others publications
evoke the beneficial proton-NP combination. For iron NP, Kim
obtained a 20–28% increase of cell death with 14 nm FeNP and
45 MeV proton but also with 2 nm and 13 nm gold NP (Kim et al.,
2010). These NP were also tested in vivo and demonstrated 90% or
75% tumor volume reduction for gold or iron respectively in 20
days compared to 18% for the control animals without NP. Simu-
lation approaches also confirmed efficiency of proton beam in the
presence of NP and non-homogeneous and anisotropic dose dis-
tribution with a higher ejected electrons numbers near the NP
surface (Kwon et al., 2014; Lin et al., 2014). Tran et al. showed that
whatever the proton energy, the dose enhancement is about ten
from 100 to 1000 nm from the NP for energies under 10 MeV, this
value increasing with the radial distance up to 1 mm and then
becoming nearly constant (Tran et al., 2016). More precisely a dose
enhancement over several mm in the depth direction and several
tens of nm in the radial direction is predicted (Kwon et al., 2014;
Spirou et al., 2015). In addition, metallic NP interaction with pro-
ton beams are known to generate particle-induced X- (or gamma-)
ray emission i.e. PIXE (or PIGE). This phenomenon involved in
nano-object enhancements was proposed to be a tool for protons
to cure disseminated metastatic tumors. Nevertheless, with this
secondary photons production the advantage of Bragg peak should
not be lost. Recently, Cho et al. showed that either for PEG coated-
gold nanorods or spherical NP, PIXE or PIGE contribution is neg-
ligible in their clinical relevant conditions whereas Auger and
secondary electrons have an important contribution at distance
less than 100 nm (Cho et al., 2016).
For MeV electrons, radiosensitization has been observed for
different NP and cell lines: 1.9 nm gold NP with MDA-MB231 with
a higher impact of 16 MeV compared to 6 MeV electrons (Jain
et al., 2011), 1.9 nm gold NP with BAEC with similar amplitude of
6 and 12 MeV (Rahman et al., 2009), 13 nm gold NP with B16F10
(Chang et al., 2008), gold and silver NP with HepG2 with 6 MeV
(Zheng et al., 2013). When the authors have conducted parallel
experiments in the keV range, enhancements were higher: for
example, Rahman et al. estimated DEF at 90% survival at 24.6 for
80 kVp versus 4 for 6 MeV (Rahman et al., 2009). These results are
not in agreement with Monte Carlo simulations which claims that
there is no advantage in electron beams compared to photons
(Chow et al., 2012; Rogers, 2013). With carbon ions, glucose-cap-
ped gold NP led to a 28% dose reduction for 90% survival of HeLa
cells (Kaur et al., 2013) which is the same order of magnitude of
15 nm citrate gold NP with 70 keV/mm carbon ions inducing an
increase of 24.5% of the relative biological effectiveness on HeLa
s
cells (Liu et al., 2016). In addition, AGuIX NP induced a 24% dose
reduction for CHO cells as for helium ions (Porcel et al., 2014). It
should be noticed that for these very different studies, they all
converge to a ca 25% dose reduction. Nevertheless, as these kinds
of beams are less common, only few groups got interested in their
coupling with NP and more research is needed before drawing
firm conclusions.
To summarize this paragraph, it clearly appears that among all
the biological studies some disagreements are present but a gen-
eral behavior concerning the energy can be highlighted: keV en-
ergies seem to be more efficient then MeV photons. In silico, stu-
dies confirmed this behavior and gave important clues about the
137
mechanism: Auger electrons are of high importance and induce
energy deposition in close vicinity to the NP. Simulation im-
provements were also realized by integrating very low energy
extension in Geant4-DNA to simulate absorbed dose distribution
and by determining the energy profile of electrons produced by
the NP. These progresses should increase the reliability of the si-
mulations and allow to make the link with the physico-chemical
step discussed later.
4. Is there a preferential localization of NP for
radiosensitization?
When incubated with cells, all the listed nanomaterials locate
either at the cellular membrane or in the cytoplasm, mainly in
vacuoles. No localization in specific organelles such as nucleus or
mitochondria was explicitely reported in these radiosensitizing
studies. Thus, the obsessive question “Is this mandatory to target
the nucleus? ” finds its answer: significant radiosensitization al-
ready occurs without any nuclear localization. Basically, only the
impact of inner or outer localization can then be examined. Several
studies report a major contribution of NP inside the cells, through
experiments of irradiation conducted with or without prior
washings to get rid of extracellular NP (Cui et al., 2014; Yasui et al.,
2014) or with different NP (Kong et al., 2008). For example, Kong
et al. adjusted extracellular gold concentrations to reach the same
concentrations at the membrane with cysteamine-capped NP and
in the cytoplasm with thioglucose-NP (Kong et al., 2008). 48 h
after 10 Gy, cell survivals were respectively 70% and 30% compared
to irradiated controls, testifying of a better efficiency of inner NP.
In Detappe”s work, Panc1 cells were submitted before irradiation
to 0.5 mM AGuIX in the following conditions: no incubation, 1 h
incubation and washing, 1 h incubation (Detappe et al., 2015).
Dose enhancement factors, defined as the ratio between the sur-
vival curves areas with and without Gd, and p-values relative to
the controls (statistical significance indicated inside brackets)
were respectively 1.09 (0.131), 1.17(0.038) and 1.46 (o0.001). This
result suggests that the interaction of NP with membrane con-
tributing to radiosensitization is not immediate and needs time to
develop but can be significant. In agreement, Bobyk concluded in
the predominant contribution of extracellular gold NP since al-
most no effect was observed when F98 glioma cells were in-
cubated for 220 min with NP and rinsed before irradiation (Bobyk
et al., 2013). Even if no ICP-MS data are provided, it would be
curious that no significant gold incorporation occurs after nearly
4 h. As in the only other study dealing with F98 cells irradiation
was done without washing and with gadolinium compounds
(Taupin et al., 2015), we cannot conclude about a specific sensi-
tivity of this cell line to extracellular NP or to the chemical nature
of the NP. In any case, it might be cell-dependent but as the idea of
membrane involvement in radiation response is not new (Alper,
1963; Prise et al., 2005), it is surprising that the damages to cel-
lular membranes and the subsequent activation of signaling
pathways have not been more investigated so far in the context of
NP radiosensitization.
5. What about DNA damage?
In most studies, γ-H2AX or 53BP1 foci are quantified to reflect
DNA double-strand breaks (DSB), comet assay is far less found in
the publications listed. This endpoint is of high importance to
understand the processes at stage but unfortunately, some authors
do not indicate the delay between irradiation and fixation of the
cells. This data is capital as early time points express initial da-
mages while delayed time points such as 24 h reflect a DNA repair
138 E. Brun, C. Sicard-Roselli / Radiation Physics and Chemistry 128 (2016) 134–142
defect. So we highly recommend not forgetting to give the post-
irradiation time used in future works.
First, several publications report DSB when combining radia-
tion and NP, and each time DSB are detected, NP are located in the
cytoplasm and not in the nucleus, which implies either a long
range radical production and/or an indirect effect through cell
signaling for example. These DNA damages are not unexpected
whatever the beam type: though most of the dose increase is lo-
cated near the NP, emitted species from NP, ejected electrons and
also photons for protons, are expected to reach up few hundred
nanometers from the NP where they create deleterious radicals.
Moreover, track ends will generate low-energy electrons
(o100 eV) that can damage DNA as discussed in (Her et al., 2015).
Next, two patterns can be extracted from the reviewed litera-
ture. The first one consists in an increase of the initial number of
foci (30 min, 1 h). For gold, it was reported for sizes from 12 to
50 nm and different coatings: antibody (Chattopadhyay et al.,
2013; Chattopadhyay et al., 2010), bovine serum albumin (Chen
et al., 2015), citrate (Chithrani et al., 2010; Liu et al., 2016)… When
indicated, 24 h post-irradiation, the foci numbers are comparable
to the controls (Chen et al., 2015; Chithrani et al., 2010). In their
study published in 2013, Chattopadhyay et al. noticed a 3.3 fold
increase in foci for trastuzumab-PEG-coated gold NP 30 min after
0.5 Gy compared to controls and a 1.7 fold for PEG-NP which were
less internalized, suggesting NP concentration-dependent DNA
damage induction (Chattopadhyay et al., 2013). These last results
probably testify for a dose increase with an efficient cellular repair
machinery. The second pattern shows no initial increase in foci
numbers but a significant difference with the control becomes
s
apparent with increasing time. It is the case for 1.9 nm Aurovist
NP (Jain et al., 2011; Taggart et al., 2014) and 2.7 nm tiopronin-
coated NP (Cui et al., 2014). The same phenomenon was observed
for titanate nanotubes (Mirjolet et al., 2013), hydroxyapatite NP
(Chu et al., 2013), PEGylated nanogel containing 8 nm gold NP
(Yasui et al., 2014). For gold-loaded polymeric micelles, the first
time point is 6 h showing a 1.4 fold increase in foci numbers. At
12 h, it reaches 2.2, confirming a time-dependency (McQuade
s
et al., 2015). For AGuIX , initial enhancement in foci number
seems to depend on the cell line: at 30 min, no difference between
exposed and control cells was noted with B16F10 (Kotb et al.,
2016) while a 1.5 fold increase was mentioned for SQ20B (Miladi
et al., 2015). Nevertheless, the trend is the same with time, as
1.4 and 2.75 fold increases are reported at 24 h for the two cell
lines aforementioned. This suggests a failure in DNA damage re-
pair. For both patterns, clonogenic survivals reveal radio-
sensitization, which means that initial DNA DSB are not a good
indicator of cell survival. Moreover, in the linear quadratic model
which is the gold standard to fit survival curves, the alpha term is
often considered reflecting directly lethal DSB. In the corpus, for a
same number of initial foci, the alpha parameter is usually higher
for cells exposed to NP than controls (0.26 versus 0.04 for Kotb
et al., (Kotb et al., 2016), 0.091 versus 0.019 for Jain et al., (Jain
et al., 2011)). As for Chattopadhyay (Chattopadhyay et al., 2013),
the fits of the survival curves give close alpha values of 0.15 7 0.02
and 0.23 70.07 for gold and control respectively. In this case, the
radiosensitization happens through a difference in beta values
(0.068 70.004 and 0.0107 0.013 for gold-treated and control cells
respectively). It suggests that α is neither correlated to initial DSB
damage nor survival. Even if the linear quadratic model has been
guiding clinicians for 50 years, predicting early and late response
of tissues through the α/β ratio, it remains an empirical model and
it would be an opportunity to question the biological significance
of α and β, especially when NP are at stage. With this in mind, one
should pay attention to the interpretation made by Foray and coll.
where α represents recognized but non-repaired damages and β
non-recognized therefore non-repaired damages (Bodgi and Foray,
2016).
To sum up, this corpus shows that the localization of NP inside
the nucleus is not necessary to induce DNA damages and that early
DNA DSB seems not to be appropriate to predict radiosensitivity. It
should incite to find more relevant markers.
6. Is there a “biological” effect?
Since 2008, a “biological” effect has been cited to explain un-
expected results. For example, even if they present the same gold
intracellular concentration, with the same NP, MCF-7 and MCF-
10A do not respond identically to 220 kVp X-rays (40% versus 75%
cell survivals respectively) (Kong et al., 2008). A “simple physical
interaction between NP and radiation” is therefore a simplistic
view of the radiosensitization origin. Later, other facts have added
grist to this mill. Among them, the lower cell survival observed
when cells are exposed to gold NP before bleomycin (a radio-
mimetic molecule causing DNA DSB) compared to bleomycin alone
(Jain et al., 2011), the dose dependency of the enhancements factor
(whatever their mathematical expressions) as mentioned in Tag-
gart et al., (Taggart et al., 2014) or their energy dependency in the
keV range (Rahman et al., 2014) with maxima at energies different
from the edges. At first sight, some results such as the enhance-
ment observed at MeV energy (Geng et al., 2011) or for low-Z
elements were pinned up to this “biological” table. In this section
we will summarize the data concerning the possible mechanisms
responsible for this biological effect.
Several cellular processes are known to be determinants of
radiosensitivity and were thus investigated. The first one was the
cell redistribution, as a consequence of NP exposure, into a
radiosensitive phase of cell cycle such as G2/M. Indeed, for gold NP,
several groups reported such an increase in G2/M cells for differ-
ent cell lines (DU-145 (Roa et al., 2009), MDA-MB-231 (Wang et al.,
2015), SK-OV-3 (Geng et al., 2011), HepG2 (Zhu et al., 2015), A549
(Wang et al., 2013). The NP used have in common a medium size
(410 nm in diameter) and a coating containing sugar (thioglucose
or galactose). On the contrary, other groups found no accumula-
tion in a radiosensitive phase. It concerns smaller NP: 1.9 nm
s
Aurovist (Butterworth et al., 2010; Jain et al., 2011) and 2.7 nm
tiopronin-gold NP (Cui et al., 2014). A recent study, non-related
with radiosensitization, also reports no redistribution to G2/M for
antibody-capped 2.6 nm gold NP (Ma et al., 2016). Interestingly,
cell line cannot be the only discriminant factor as MDA-MB-231
cells shows, depending on the NP, redistribution or not. It would
be interesting to investigate the role of the size and coating to
precise if the accumulation described is, for example, related to the
sugar moiety of the NP. For nanodiamonds (Grall et al., 2015), ti-
s
tanate nanotubes (Mirjolet et al., 2013), AguiX (Miladi et al., 2015)
or hydroxyapatite NP (Chu et al., 2013), no redistribution in the cell
cycle was observed meaning another phenomenon should be
looked for. Another paramount event in determining radio-
sensitivity is DNA repair. As mentioned previously, for a group of
papers, 24 h post-irradiation, the residual number of γ-H2AX foci
was higher in the NP-exposed than in the control cells though
there was no more initial damages. This suggests an impairment of
DNA repair functions. As Rad51 and Ku70 are the key proteins in
homologous recombinaison and non-homologous end-joining, the
two pathways of DSB repair, Yasui et al. assessed their expressions
by immunoblotting and noted their drastic reduction in lysates of
nanogel-exposed cells (Yasui et al., 2014). Rad51 down-regulation
was also observed after cell incubation with hydroxyapatite NP
(Chu et al., 2013). Interestingly, even Rad51 up-regulation caused
by irradiation was counterbalanced by NP effect. In addition, these
authors report that other protein expressions were affected: proto-
oncogene c-Met was down-regulated as well as SATB1 whereas
 E. Brun, C. Sicard-Roselli / Radiation Physics and Chemistry 128 (2016) 134–142
SLC22A18 was upregulated. These findings suggest that at least
some NP sensitize cells not by inducing more DNA damages but
probably by impairing DNA repair pathway.
As said before, a totally DNA centric approach of NP radio-
sensitization is not sufficient. With this in mind, Taggart et al.
recently investigated the impact of gold NP on mitochondrial
functions (Taggart et al., 2014). Mitochondria were first considered
as the “powerhouse of the cell” but they are implicated in a wider
variety of cellular functions and pathologies including cell signal-
ing, metabolism, cell death, aging, and cancer (Evans and Neuman,
2016). To them converge many apoptosis-inducing signals in
mammalian cells (Bhola and Letai, 2016). They are then a core
target. Aurovists NP alone reduced mitochondrial membrane po-
larization to 50%, 55% and 25% of the control levels for MDA-MB-
231, DU-145 and T98G cells. In addition, cardiolipin oxidation was
demonstrated through the binding of the fluorescent probe NAO.
This is of importance as mobilization of cytochrome c, a key step in
the intrinsic apoptotic pathway, is tightly regulated by the oxida-
tion state of cardiolipin. The significance of this paper is double: it
is the first to deliberately focus on cytoplasmic organelles affected
by NP in this context and it highlights the importance of cellular
events driven by NP prior to irradiation.
Interestingly, a possible rationale emerges from this body of
data. Firstly mentioned by Yasui et al., (Yasui et al., 2014), en-
doplasmic reticulum (ER) stress could be part of this biological
effect. As a hub for protein and lipid synthesis, a store for in-
tracellular calcium and a mediator of cellular stress, the ER is an
important organelle. ER stress results from accumulation of mis-
folded/unfolded proteins and activates the Unfolded Protein Re-
sponse (UPR). This is a complex cellular response which includes
the exacerbation of defective proteins degradation. Rad51 is one of
them. It has been shown to be down-regulated at the protein
expression level when ER stress is induced via different means:
tunicamycin (Yamamori et al., 2013), salinomycin (Xipell et al.,
2016) or chronic hypoxia (Bindra et al., 2004; Cui et al., 2014). In
Yamamori et al., a clear correlation was observed between the
upregulation of phospho-eIF2α, indicating the level of ER stress,
and the downregulation of Rad51 (Yamamori et al., 2013). This
resulted in increased sensitivity of A549 cells to X-rays and cis-
platin. In Yasui et al., following nanogel exposure, SCCVII as A549
cells showed increases of several ER stress-related proteins ex-
pressions (Yasui et al., 2014). So the impairment of DNA DSB repair
machinery could be a consequence of ER stress. This could shed
new light on Taggart”s results as UPR activation can trigger chan-
ges in mitochondrial functions (Rainbolt et al., 2014; Senft and
Ronai, 2015) and thus ER stress is not in contradiction with the
reported results. Similarly, the fact that glucose deprivation can
also induce ER stress (Yamamori et al., 2013) allows reconsidering
results obtained by Hu et al., (Hu et al., 2015). In this paper, the
authors observed an enhanced internalization of gold NP after
starvation but ER stress could also be at stage to explain the
radiosensitivity. We thus propose that, at least for some NP, ER
stress plays a major role in NP-induced radiosensitization. How NP
could trigger ER stress? Elevated level of reactive oxygen species
(ROS) could be a possibility, as well as the misfolding of proteins
by NP binding. It begins to be documented that protein con-
formational changes can occur at the NP surface at the secondary
or tertiary levels (Shao et al., 2011; Simon-Vazquez et al., 2014)
and that the NP surface chemistry is decisive (Radic et al., 2014).
Proteins forming the protein corona might be particularly affected.
It would be too simplistic to designate biological effect(s), be it ER
stress or not, as the only effect responsible for NP radio-
sensitization. Given the variety of NP and cells involved in these
studies, it seems impossible that all induce the same cellular re-
sponse. However, in the direction of a better understanding of
radiosensitization process, clonogenic survival cannot be the only
139
way to probe cellular responses: they should be investigated in
more details.
7. Which connection between physical and biological effect of
NP?
In most papers are described either the physical interaction
between NP and radiation or the biological consequences. But the
physico-chemical step i.e. radicals production, which links them is
crucial to explain the differences evidenced between the predic-
tions and the experimental data and also the discrepancy between
the measurements in the corpus.
A first approach to determine ROS production in vitro was
performed by Carter et al. who pointed out the role of hydroxyl
radicals in inducing DNA strand breaks comparing with and
without TRIS scavenging (Carter et al., 2007). To quantify precisely
these radicals production by NP, coumarin (Sicard-Roselli et al.,
2014) and 3 carboxy-coumarin (Cheng et al., 2012) were used as
specific and quantitative probes. Sicard-Roselli et al. showed that a
very high quantity of HO
radicals arise from the NP interaction
with low energy X-Rays (20 keV). In the presence of 100 mg/mL of
32 nm gold NP, the concentration of HO
is multiplied by 3.5 at a
dose rate of 0.3 Gy/s (Gilles et al., 2014). This value is much higher
than what can be expected from the energy deposited in the col-
loidal solution. Paudel et al. reported a similar trend as experi-
mental production of hydroxyl radicals is found twice higher than
what is predicted by Monte Carlo simulation when irradiation was
performed with 192Ir (Paudel and Parsai, E. I. 2015). These findings
lead the authors to propose surface reactions or specific water
organization at the NP surface to generate high quantities of ra-
dicals. More recently, Gilles et al. completed this mechanistic study
by quantifying the aqueous electrons generated by NP irradiated
both by X- and gamma-rays (Gilles et al., unpublished data). Here
also, the generation of secondary electrons is much higher than
expected and oxygen atmosphere was shown to impact strongly
the radical production. and more surprising, MeV energy was
more efficient than keV photon. These original results can only be
explained if the NP surface has specific properties as non-standard
water interface organization and/or catalytic reactivity. Specific
surface reactivity is proposed to account for hydroxyl radicals
production from nanodiamonds (NDs) under 17.5 keV photons as
it happens only for hydrogenated and not oxidated NDs (personal
communication). This could explain the radiosensitization evi-
denced for the objects (Grall et al., 2015).
These new clues are of high importance as they could give new
insights on the mechanisms involved when NP are submitted to
ionizing radiations but also because they could help implementing
simulation by giving new inputs. But we should wonder whether
we can make a link between on the one hand the physical inter-
action between NP and radiation and on the other hand, between
the radical production in vitro and in the cells. Tran et al. are the
first group to extend Monte Carlo simulations to the physico-
chemical step by trying to determine the quantities of radicals
produced according to the enhanced dose deposited by proton
beams (Tran et al., 2016). To test the relevance of the numerical
model, it would be interesting to confront these simulations to
experimental quantification of radicals within the same condi-
tions. Nevertheless, two important parameters are most of the
time neglected in simulation: the clustering of NP inside the cells
and the coating of the NP. Kirkby et al. noticed the impact of the
coating thickness on the dose enhancement ratio by Monte Carlo
methods (Kirkby and Ghasroddashti, 2015). This is supported by
Gilles et al. who precisely showed that for a coating thickness
higher than 5 nm, the hydroxyl radicals produced by the NP are
mostly scavenged by the NP ligands and are no more available to
140 E. Brun, C. Sicard-Roselli / Radiation Physics and Chemistry 128 (2016) 134–142
oxidize a target (Gilles et al., 2014). In a cellular context, an addi-
tional coating is expected: the protein corona (Brun and Sicard-
Roselli, 2014). To make a link between the radical production
measured in vitro and the radiosensitization, ROS cellular quan-
tification appears necessary. Only few papers deal with this. For
instance, combining Si NP and 4 MeV X-rays, a 700% increase in
DCF fluorescence was achieved in C6 glioma cells after 3 Gy (Gara
et al., 2012). An increase in intracellular ROS content was also
shown for SPION (Klein et al., 2012), nanodiamonds (Grall et al.,
s Boisselier, E., Astruc, D., 2009. Gold nanoparticles in nanomedicine: preparations,
2015), AGuIX (Miladi et al., 2015) and selenium NP (Yu et al.,
2016). The first difficulty is that ROS quantification in cells with
usual fluorescent probes such as H2DCFDA and its derivatives do
not discriminate different kinds of radicals, hydroxyl or superoxide
for example, contrary to what is done in vitro, coumarin being a
specific scavenger of HO
. And the second one is that it is im-
possible to discriminate a direct production of ROS by NP as a
consequence of irradiation from an overproduction of ROS as a
cellular response. Then, it is not always possible to conclude about
an additive or a synergetic effect of irradiation and NP. This che-
mical step was until now less studied nevertheless we consider
that either simulation or experimental work should focus on the
water NP interface.
8. Conclusion
The emergence of NP and their association to radiation has
paved the way for the development of new radiosensitizers. A vast
set of data has been produced in the last decade in this field, with
sometimes contradictory results. If no consensus has come out yet
about the more efficient NP/radiation combination, the complexity
of the phenomenon at stage wins unanimous support. The evo-
lution of our vision translates into the words used, from “dose
enhancement” a decade ago to “potentialization of ionizing radia-
tion” (Grall et al., 2015).
Through the examined literature, we have raised some ques-
tions that, for us, are worth addressing in details and with a sys-
tematic approach. The first one concerns our definition of a NP.
Can it be just considered as a sum of atoms or is there a collective
effect? Casta at al. mentioned a “nano-scale effect” on electrons
emission, which seems intrinsic to gold NP (Casta et al., 2014). This
should slightly modify our way to consider NP, the comparison
with isolated atoms such as contrast agent could be inappropriate.
Second, the “chemical” effect deals with interface reactions. It has
been shown that water radiolysis at solid/water interfaces has a
very different chemistry than in the bulk (Chelnokov et al., 2014;
Schofield et al., 2016) and the NP should be studied as such. Third,
now that radicals generation by NP has been assessed at a mac-
roscopic scale, their spatio-temporal production should be in-
vestigated through simulations and experimental work, giving
access to a nanodosimetry. Lastly, the association of NP with
newly-developed sources such as laser plasma accelerators (Gau-
duel et al., 2012) is in its infancy but could help clarifying the dose
rate influence in NP/radiation interaction, in addition to open new
possibilities in treatments. These issues go beyond the interest of
the “nano” community focused on radiosensitization. An incon-
testable support could be offered by the “radiation” community.
Bridging the gap between them could be a major asset to assess
the relative contributions of physical, chemical and biological ef-
fects in NP radiosensitization.
References
Alper, T., 1963. Lethal mutations and cell death. Phys. Med. Biol. 66, 365–385.
Bhola, P.D., Letai, A., 2016. Mitochondria-Judges and Executioners of Cell Death
Sentences. Mol. Cell 61, 695–704.
Bindra, R.S., Schaffer, P.J., Meng, A., Woo, J., Maseide, K., Roth, M.E., Lizardi, P.,
Hedley, D.W., Bristow, R.G., Glazer, P.M., 2004. Down-regulation of Rad51 and
decreased homologous recombination in hypoxic cancer cells. Mol. Cell Biol. 24,
8504–8518.
Bobyk, L., Edouard, M., Deman, P., Vautrin, M., Pernet-Gallay, K., Delaroche, J., Adam,
J.F., Esteve, F., Ravanat, J.L., Elleaume, H., 2013. Photoactivation of gold nano-
particles for glioma treatment. Nanomed.-Nanotechnol. 9, 1089–1097.
Bodgi, L., Foray, N., 2016. The nucleo-shuttling of the ATM protein as a basis for a
novel theory of radiation response: resolution of the linear-quadratic model.
Int. J. Radiat. Biol. 92, 117–131.
imaging, diagnostics, therapies and toxicity. Chem. Soc. Rev. 38, 1759–1782.
Brun, E., Sanche, L., Sicard-Roselli, C., 2009. Parameters governing gold nanoparticle
X-ray radiosensitization of DNA in solution. Colloid Surf. B 72, 128–134.
Brun, E., Sicard-Roselli, C., 2014. Could nanoparticle corona characterization help
for biological consequence prediction? Cancer Nanotechnol., 5.
Butterworth, K.T., Coulter, J.A., Jain, S., Forker, J., McMahon, S.J., Schettino, G., Prise,
K.M., Currell, F.J., Hirst, D.G., 2010. Evaluation of cytotoxicity and radiation
enhancement using 1.9 nm gold particles: potential application for cancer
therapy. Nanotechnology, 21.
Carter, J.D., Cheng, N.N., Qu, Y.Q., Suarez, G.D., Guo, T., 2007. Nanoscale energy
deposition by x-ray absorbing nanostructures. J. Phys. Chem. B 111,
11622–11625.
Casta, R., Champeaux, J.P., Sence, M., Moretto-Capelle, P., Cafarelli, P., 2015a. Com-
parison between gold nanoparticle and gold plane electron emissions: a way to
identify secondary electron emission. Phys. Med. Biol. 60, 9095–9105.
Casta, R., Champeaux, J.P., Sence, M., Moretto-Capelle, P., Cafarelli, P., 2015b. Nanop:
an x-ray to gold nanoparticle electron and photon emission software. Mater.
Res. Express, 2.
Casta, R., Champeaux, J.P., Sence, M., Moretto-Capelle, P., Cafarelli, P., Amsellem, A.,
Sicard-Roselli, C., 2014. Electronic emission of radio-sensitizing gold nano-
particles under X-ray irradiation: experiment and simulations. J. Nanopart. Res.,
16.
Chang, M.Y., Shiau, A.L., Chen, Y.H., Chang, C.J., Chen, H.H.W., Wu, C.L., 2008. In-
creased apoptotic potential and dose-enhancing effect of gold nanoparticles in
combination with single-dose clinical electron beams on tumor-bearing mice.
Cancer Sci. 99, 1479–1484.
Chattopadhyay, N., Cai, Z.L., Kwon, Y.L., Lechtman, E., Pignol, J.P., Reilly, R.M., 2013.
Molecularly targeted gold nanoparticles enhance the radiation response of
breast cancer cells and tumor xenografts to X-radiation. Breast Cancer Res.
Treat. 137, 81–91.
Chattopadhyay, N., Cai, Z.L., Pignol, J.P., Keller, B., Lechtman, E., Bendayan, R., Reilly,
R.M., 2010. Design and Characterization of HER-2-Targeted Gold Nanoparticles
for Enhanced X-radiation Treatment of Locally Advanced Breast Cancer. Mol.
Pharm. 7, 2194–2206.
Chelnokov, E., Cuba, V., Simeone, D., Guigner, J.M., Schmidhammer, U., Mostafavi,
M., Le Caer, S., 2014. Electron Transfer at Oxide/Water Interfaces Induced by
Ionizing Radiation. J. Phys. Chem. C 118, 7865–7873.
Chen, N., Yang, W.T., Bao, Y., Xu, H.L., Qin, S.B., Tu, Y., 2015. BSA capped Au nano-
particle as an efficient sensitizer for glioblastoma tumor radiation therapy. Rsc.
Adv. 5, 40514–40520.
Cheng, N.N., Starkewolf, Z., Davidson, R.A., Sharmah, A., Lee, C., Lien, J., Guo, T., 2012.
Chemical Enhancement by Nanomaterials under X-ray Irradiation. J. Am. Chem.
Soc. 134, 1950–1953.
Chithrani, D.B., Jelveh, S., Jalali, F., van Prooijen, M., Allen, C., Bristow, R.G., Hill, R.P.,
Jaffray, D.A., 2010. Gold Nanoparticles as Radiation Sensitizers in Cancer Ther-
apy. Radiat. Res. 173, 719–728.
Cho, J., Gonzalez-Lepera, C., Manohar, N., Kerr, M., Krishnan, S., Cho, S.H., 2016.
Quantitative investigation of physical factors contributing to gold nanoparticle-
mediated proton dose enhancement. Phys. Med. Biol. 61, 2562–2581.
Chow, J.C., Leung, M.K., Jaffray, D.A., 2012. Monte Carlo simulation on a gold na-
noparticle irradiated by electron beams. Phys. Med. Biol. 57, 3323–3331.
Chu, S.H., Karri, S., Ma, Y.B., Feng, D.F., Li, Z.Q., 2013. In vitro and in vivo radio-
sensitization induced by hydroxyapatite nanoparticles. Neuro Oncol. 15,
880–890.
Cooper, D.R., Bekah, D., Nadeau, J.L., 2014. Gold nanoparticles and their alternatives
for radiation therapy enhancement. Front Chem. 2, 86.
Cui, L., Tse, K., Zahedi, P., Harding, S.M., Zafarana, G., Jaffray, D.A., Bristow, R.G.,
Allen, C., 2014. Hypoxia and Cellular Localization Influence the Radiosensitizing
Effect of Gold Nanoparticles (AuNPs) in Breast Cancer Cells. Radiat. Res. 182,
475–488.
Detappe, A., Rottmann, J., Kunjachan, S., Tillement, O., Berbeco, R., 2015. Theranostic
Gadolinium-Based AGuIX Nanoparticles for MRI-Guided Radiation Therapy.
Med. Phys. 42 3566–3566.
Evans, A., Neuman, N., 2016. The Mighty Mitochondria. Mol. Cell 61, 641.
Fratoddi, I., Venditti, I., Cametti, C., Russo, M.V., 2015. How toxic are gold nano-
particles? The state-of-the-art. Nano Res. 8, 1771–1799.
Gara, P.M.D., Garabano, N.I., Portoles, M.J.L., Moreno, M.S., Dodat, D., Casas, O.R.,
Gonzalez, M.C., Kotler, M.L., 2012. ROS enhancement by silicon nanoparticles in
X-ray irradiated aqueous suspensions and in glioma C6 cells. J. Nanopart. Res.,
14.
Gauduel, Y.A., Lundh, O., Martin, M.T., Malka, V., 2012. Laser-plasma accelerators
based high energy radiation femtochemistry and spatio-temporal radiation
biomedicine. Proc. Spie, 8433.
Geng, F., Song, K., Xing, J.Z., Yuan, C.Z., Yan, S., Yang, Q.F., Chen, J., Kong, B.H., 2011.
 E. Brun, C. Sicard-Roselli / Radiation Physics and Chemistry 128 (2016) 134–142
Thio-glucose bound gold nanoparticles enhance radio-cytotoxic targeting of
ovarian cancer. Nanotechnology, 22.
Gilles, M., Brun, E., Sicard-Roselli, C., 2014. Gold nanoparticles functionalization
notably decreases radiosensitization through hydroxyl radical production un-
der ionizing radiation. Colloid Surf. B 123, 770–777.
Gilles, M., Brun, E., Sicard-Roselli, C., submitted for publication. Quantification of
the Radical Species in Solution When Gold Nanoparticles Are to Ionizing Ra-
diation, unpublished data.
Gimenez, P., 2015. Radiothérapie par photoactivation de nanoparticuls et effet
Mössbauer. Université Grenoble Alpes.
Grall, R., Girard, H., Saad, L., Petit, T., Gesset, C., Combis-Schlumberger, M., Paget, V.,
Delic, J., Arnault, J.C., Chevillard, S., 2015. Impairing the radioresistance of
cancer cells by hydrogenated nanodiamonds. Biomaterials 61, 290–298.
Guidelli, E.J., Baffa, O., 2014. Influence of photon beam energy on the dose en-
hancement factor caused by gold and silver nanoparticles: an experimental
approach. Med Phys., 41.
Her, S., Jaffray, D.A., Allen, C., 2015. Gold nanoparticles for applications in cancer
radiotherapy: Mechanisms and recent advancements. Adv Drug Deliv Rev.
Herold, D.M., Das, I.J., Stobbe, C.C., Iyer, R.V., Chapman, J.D., 2000. Gold micro-
spheres: a selective technique for producing biologically effective dose en-
hancement. Int. J. Radiat. Biol. 76, 1357–1364.
Hu, C.X., Niestroj, M., Yuan, D., Chang, S., Chen, J., 2015. Treating cancer stem cells
and cancer metastasis using glucose-coated gold nanoparticles. Int. J. Nanomed.
10, 2065–2077.
Jain, S., Coulter, J.A., Hounsell, A.R., Butterworth, K.T., McMahon, S.J., Hyland, W.B.,
Muir, M.F., Dickson, G.R., Prise, K.M., Currell, F.J., O’Sullivan, J.M., Hirst, D.G.,
2011. Cell-Specific Radiosensitization by Gold Nanoparticles at Megavoltage
Radiation Energies. Int. J. Radiat. Oncol. 79, 531–539.
Jeynes, J.C.G., Merchant, M.J., Spindler, A., Wera, A.C., Kirkby, K.J., 2014. Investigation
of gold nanoparticle radiosensitization mechanisms using a free radical sca-
venger and protons of different energies. Phys. Med. Biol. 59, 6431–6443.
Jones, B.L., Krishnan, S., Cho, S.H., 2010. Estimation of microscopic dose enhance-
ment factor around gold nanoparticles by Monte Carlo calculations. Med. Phys.
37, 3809–3816.
Kaur, H., Pujari, G., Semwal, M.K., Sarma, A., Avasthi, D.K., 2013. In vitro studies on
radiosensitization effect of glucose capped gold nanoparticles in photon and
ion irradiation of HeLa cells. Nucl. Instrum. Meth B 301, 7–11.
Khoshgard, K., Hashemi, B., Arbabi, A., Rasaee, M.J., Soleimani, M., 2014. Radio-
sensitization effect of folate-conjugated gold nanoparticles on HeLa cancer cells
under orthovoltage superficial radiotherapy techniques. Phys. Med. Biol. 59,
2249–2263.
Kim, J.K., Seo, S.J., Kim, K.H., Kim, T.J., Chung, M.H., Kim, K.R., Yang, T.K., 2010.
Therapeutic application of metallic nanoparticles combined with particle-in-
duced x-ray emission effect. Nanotechnology, 21.
Kirkby, C., Ghasroddashti, E., 2015. Targeting mitochondria in cancer cells using
gold nanoparticle-enhanced radiotherapy: a Monte Carlo study. Med Phys. 42,
1119–1128.
Klein, S., Sommer, A., Distel, L.V., Neuhuber, W., Kryschi, C., 2012. Super-
paramagnetic iron oxide nanoparticles as radiosensitizer via enhanced reactive
oxygen species formation. Biochem. Biophys. Res. Commun. 425, 393–397.
Kong, T., Zeng, J., Wang, X.P., Yang, X.Y., Yang, J., McQuarrie, S., McEwan, A., Roa, W.,
Chen, J., Xing, J.Z., 2008. Enhancement of radiation cytotoxicity in breast-cancer
cells by localized attachment of gold nanoparticles. Small 4, 1537–1543.
Kotb, S., Detappe, A., Lux, F., Appaix, F., Barbier, E.L., Tran, V.L., Plissonneau, M.,
Gehan, H., Lefranc, F., Rodriguez-Lafrasse, C., Verry, C., Berbeco, R., Tillement, O.,
Sancey, L., 2016. Gadolinium-Based Nanoparticles and Radiation Therapy for
Multiple Brain Melanoma Metastases: Proof of Concept before Phase I Trial.
Theranostics 6, 418–427.
Kwon, J., Sutherland, K., Hashimoto, T., Date, H., 2014. Dose Distribution of Electrons
from Gold Nanoparticles by Proton Beam Irradiation. Int. J. Med. Phys. Clinical
Eng. Radiation Oncol. 4, 49–53.
Le Duc, G., Roux, S., Paruta-Tuarez, A., Dufort, S., Brauer, E., Marais, A., Truillet, C.,
Sancey, L., Perriat, P., Lux, F., Tillement, O., 2014. Advantages of gadolinium
based ultrasmall nanoparticles vs molecular gadolinium chelates for radio-
therapy guided by MRI for glioma treatment. Cancer Nanotechnol., 5.
Lechtman, E., Chattopadhyay, N., Cai, Z., Mashouf, S., Reilly, R., Pignol, J.P., 2011.
Implications on clinical scenario of gold nanoparticle radiosensitization in re-
gards to photon energy, nanoparticle size, concentration and location. Phys.
Med. Biol. 56, 4631–4647.
Lehnert, S., 2014. Radiosensitizers and radiochemotherapy in the treatment of
cancer. CRC Press.
Leung, M.K., Chow, J.C., Chithrani, B.D., Lee, M.J., Oms, B., Jaffray, D.A., 2011. Irra-
diation of gold nanoparticles by x-rays: Monte Carlo simulation of dose en-
hancements and the spatial properties of the secondary electrons production.
Med. Phys. 38, 624–631.
Lin, Y.T., McMahon, S.J., Scarpelli, M., Paganetti, H., Schuemann, J., 2014. Comparing
gold nano-particle enhanced radiotherapy with protons, megavoltage photons
and kilovoltage photons: a Monte Carlo simulation. Phys. Med Biol. 59,
7675–7689.
Linam, J., Yang, L.X., 2015. Recent Developments in Radiosensitization. Anticancer
Res. 35, 2479–2485.
Liu, C.J., Wang, C.H., Chen, S.T., Chen, H.H., Leng, W.H., Chien, C.C., Wang, C.L.,
Kempson, I.M., Hwu, Y., Lai, T.C., Hsiao, M., Yang, C.S., Chen, Y.J., Margaritondo,
G., 2010. Enhancement of cell radiation sensitivity by pegylated gold nano-
particles. Phys. Med Biol. 55, 931–945.
Liu, Y.L., Jin, X., He, X., Zheng, P., Ye, X., Chen, F., Li, Q, W., 2016. The radiation
141
enhancement of 15 nm citrate-capped gold nanoparticles exposed to 70 keV/
mm carbon ions. J. Nanosci. Nanotechnol. 16, 2365–2370.
Luchette, M., Korideck, H., Makrigiorgos, M., Tillement, O., Berbeco, R., 2014. Ra-
diation dose enhancement of gadolinium-based AGuIX nanoparticles on HeLa
cells. Nanomed.-Nanotechnol. 10, 1751–1755.
Ma, X., Hui, H., Jin, Y., Dong, D., Liang, X., Yang, X., Tan, K., Dai, Z., Cheng, Z., Tian, J.,
2016. Enhanced immunotherapy of SM5–1 in hepatocellular carcinoma by
conjugating with gold nanoparticles and its in vivo bioluminescence tomo-
graphic evaluation. Biomaterials 87, 46–56.
Maggiorella, L., Barouch, G., Devaux, C., Pottier, A., Deutsch, E., Bourhis, J., Borghi, E.,
Levy, L., 2012. Nanoscale radiotherapy with hafnium oxide nanoparticles. Fu-
ture Oncol. 8, 1167–1181.
McMahon, S.J., Hyland, W.B., Brun, E., Butterworth, K.T., Coulter, J.A., Douki, T., Hirst,
D.G., Jain, S., Kavanagh, A.P., Krpetic, Z., Mendenhall, M.H., Muir, M.F., Prise, K.
M., Requardt, H., Sanche, L., Schettino, G., Currell, F.J., Sicard-Roselli, C., 2011a.
Energy Dependence of Gold Nanoparticle Radiosensitization in. Plasmid Dna. J.
Phys. Chem. C 115, 20160–20167.
McMahon, S.J., Hyland, W.B., Muir, M.F., Coulter, J.A., Jain, S., Butterworth, K.T.,
Schettino, G., Dickson, G.R., Hounsell, A.R., O’Sullivan, J.M., Prise, K.M., Hirst, D.
G., Currell, F.J., 2011b. Biological consequences of nanoscale energy deposition
near irradiated heavy atom nanoparticles. Sci. Rep. 1, 18.
McQuade, C., Al Zaki, A., Desai, Y., Vido, M., Sakhuja, T., Cheng, Z., Hickey, R.J., Joh, D.,
Park, S.J., Kao, G., Dorsey, J.F., Tsourkas, A., 2015. A multifunctional nanoplat-
form for imaging, radiotherapy, and the prediction of therapeutic response.
Small 11, 834–843.
Miladi, I., Aloy, M.T., Armandy, E., Mowat, P., Kryza, D., Magne, N., Tillement, O., Lux,
F., Billotey, C., Janier, M., Rodriguez-Lafrasse, C., 2015. Combining ultrasmall
gadolinium-based nanoparticles with photon irradiation overcomes radio-
resistance of head and neck squamous cell carcinoma. Nanomedicine 11,
247–257.
Mirjolet, C., Papa, A.L., Crehange, G., Raguin, O., Seignez, C., Paul, C., Truc, G.,
Maingon, P., Millot, N., 2013. The radiosensitization effect of titanate nanotubes
as a new tool in radiation therapy for glioblastoma: A proof-of-concept.
Radiother. Oncol. 108, 136–142.
Montenegro, M., Nahar, S.N., Pradhan, A.K., Huang, K., Yu, Y., 2009. Monte Carlo
simulations and atomic calculations for Auger processes in biomedical na-
notheranostics. J. Phys. Chem. A 113, 12364–12369.
Paudel, N.S., Parsai, E. I, D., 2015. Comparative study of experimental enhancement
in free radical generation against Monte Carlo modeled enhancement in ra-
diation dose deposition due to the presence of high Z materials during irra-
diation of aqueous media. Int. J. Med. Phys. Clinical Eng.Radiation Oncol. 4,
300–307.
Pedrosa, P., Vinhas, R., Fernandes, A., Baptista, P.V., 2015. Gold Nanotheranostics:
Proof-of-Concept or Clinical Tool? Nanomaterials 5, 1853–1879.
Porcel, E., Tillement, O., Lux, F., Mowat, P., Usami, N., Kobayashi, K., Furusawa, Y., Le
Sech, C., Li, S., Lacombe, S., 2014. Gadolinium-based nanoparticles to improve
the hadrontherapy performances. Nanomed.-Nanotechnol. 10, 1601–1608.
Prise, K.M., Schettino, G., Folkard, M., Held, K.D., 2005. New insights on cell death
from radiation exposure. Lancet Oncol. 6, 520–528.
Radic, S., Nedumpully-Govindan, P., Chen, R., Salonen, E., Brown, J.M., Ke, P.C., Ding,
F., 2014. Effect of fullerenol surface chemistry on nanoparticle binding-induced
protein misfolding. Nanoscale 6, 8340–8349.
Rahman, W.N., Bishara, N., Ackerly, T., He, C.F., Jackson, P., Wong, C., Davidson, R.,
Geso, M., 2009. Enhancement of radiation effects by gold nanoparticles for
superficial radiation therapy. Nanomed.-Nanotechnol. 5, 136–142.
Rahman, W.N., Corde, S., Yagi, N., Aziz, S.A.A., Annabell, N., Geso, M., 2014. Optimal
energy for cell radiosensitivity enhancement by gold nanoparticles using syn-
chrotron-based monoenergetic photon beams. Int. J. Nanomed. 9, 2459–2467.
Rainbolt, T.K., Saunders, J.M., Wiseman, R.L., 2014. Stress-responsive regulation of
mitochondria through the ER unfolded protein response. Trends Endocrinol.
Metab. 25, 528–537.
Regulla, D.F., Hieber, L.B., Seidenbusch, M., 1998. Physical and biological interface
dose effects in tissue due to X-ray-induced release of secondary radiation from
metallic gold surfaces. Radiat. Res. 150, 92–100.
Retif, P., Pinel, S., Toussaint, M., Frochot, C., Chouikrat, R., Bastogne, T., Barberi-
Heyob, M., 2015. Nanoparticles for Radiation Therapy Enhancement: the Key
Parameters. Theranostics 5, 1030–1045.
Roa, W., Zhang, X., Guo, L., Shaw, A., Hu, X., Xiong, Y., Gulavita, S., Patel, S., Sun, X.,
Chen, J., Moore, R., Xing, J.Z., 2009. Gold nanoparticle sensitize radiotherapy of
prostate cancer cells by regulation of the cell cycle. Nanotechnology 20, 375101.
Rogers, D.W., 2013. Comment on 'Monte Carlo simulation on a gold nanoparticle
irradiated by electron beams'. Phys Med Biol 58, 1999-2001; discussion 2003-
1995.
Salvati, A., Pitek, A.S., Monopoli, M.P., Prapainop, K., Bombelli, F.B., Hristov, D.R.,
Kelly, P.M., Aberg, C., Mahon, E., Dawson, K.A., 2013. Transferrin-functionalized
nanoparticles lose their targeting capabilities when a biomolecule corona ad-
sorbs on the surface. Nat. Nanotechnol. 8, 137–143.
Sancey, L., Lux, F., Kotb, S., Roux, S., Dufort, S., Bianchi, A., Cremillieux, Y., Fries, P.,
Coll, J.L., Rodriguez-Lafrasse, C., Janier, M., Dutreix, M., Barberi-Heyob, M., Bo-
schetti, F., Denat, F., Louis, C., Porcel, E., Lacombe, S., Le Duc, G., Deutsch, E.,
Perfettini, J.L., Detappe, A., Verry, C., Berbeco, R., Butterworth, K.T., Mcmahon, S.
J., Prise, K.M., Perriat, P., Tillement, O., 2014. The use of theranostic gadolinium-
based nanoprobes to improve radiotherapy efficacy. Br. J. Radiol., 87.
Schofield, J., Reiff, S.C., Pimblott, S.M., LaVerne, J.A., 2016. Radiolytic hydrogen
generation at silicon carbide-water interfaces. J. Nucl. Mater. 469, 43–50.
Senft, D., Ronai, Z.A., 2015. UPR, autophagy, and mitochondria crosstalk underlies
142 E. Brun, C. Sicard-Roselli / Radiation Physics and Chemistry 128 (2016) 134–142
the ER stress response. Trends Biochem Sci. 40, 141–148.
Shao, Q., Wu, P., Gu, P.A., Xu, X.Q., Zhang, H., Cai, C.X., 2011. Electrochemical and
Spectroscopic Studies on the Conformational Structure of Hemoglobin As-
sembled on Gold Nanoparticles. J. Phys. Chem. B 115, 8627–8637.
Sicard-Roselli, C., Brun, E., Gilles, M., Baldacchino, G., Kelsey, C., McQuaid, H., Polin,
C., Wardlow, N., Currell, F., 2014. A New Mechanism for Hydroxyl Radical
Production in Irradiated Nanoparticle Solutions. Small 10, 3338–3346.
Simon-Vazquez, R., Lozano-Fernandez, T., Peleteiro-Olmedo, M., Gonzalez-Fernan-
dez, A., 2014. Conformational changes in human plasma proteins induced by
metal oxide nanoparticles. Colloid Surf. B 113, 198–206.
Spirou, S.V., Makris, D., Loudos, G., 2015. Does the setup of Monte Carlo simulations
influence the calculated properties and effect of gold nanoparticles in radiation
therapy? Phys. Med 31, 817–821.
Svenson, S.P.h., R. K., 2012. Multifunctional Nanoparticles for Drug Delivery Appli-
cations. Imaging, Targeting, and Delivery. Springer.
Taggart, L.E., McMahon, S.J., Currell, F.J., Prise, K.M., Butterworth, K.T., 2014. The role
of mitochondrial function in gold nanoparticle mediated radiosensitisation.
Cancer Nanotechnol. 5, 5.
Taupin, F., Flaender, M., Delorme, R., Brochard, T., Mayol, J.F., Arnaud, J., Perriat, P.,
Sancey, L., Lux, F., Barth, R.F., Carriere, M., Ravanat, J.L., Elleaume, H., 2015.
Gadolinium nanoparticles and contrast agent as radiation sensitizers. Phys.
Med Biol. 60, 4449–4464.
Tran, H.N., Dao, D.D., Incerti, S., Bernal, M.A., Karamitros, M., Hao, T.V.N., Dang, T.M.,
Francis, Z., 2016. Single electron ionization and electron capture cross sections
for (C6þ , H2O) interaction within the Classical Trajectory Monte Carlo (CTMC)
approach. Nucl. Instrum. Meth B 366, 140–144.
Wang, C.H., Li, X.H., Wang, Y., Liu, Z., Fu, L., Hu, L.K., 2013. Enhancement of radiation
effect and increase of apoptosis in lung cancer cells by thio-glucose-bound gold
nanoparticles at megavoltage radiation energies. J. Nanopart. Res., 15.
Wang, C.J., Li, Y., Hu, L, X., 2015. Thioglucose-bound gold nanoparticles increase the
radiosensitivity of a triple-negative breast cancer cell line (MDA-MB-231).
Breast Cancer 22, 413–420.
Xipell, E., Aragon, T., Martinez-Velez, N., Vera, B., Idoate, M.A., Martinez-Irujo, J.J.,
Garzon, A.G., Gonzalez-Huarriz, M., Acanda, A.M., Jones, C., Lang, F.F., Fueyo, J.,
Gomez-Manzano, C., Alonso, M.M., 2016. Endoplasmic reticulum stress-indu-
cing drugs sensitize glioma cells to temozolomide through downregulation of
MGMT, MPG, and Rad51. Neuro Oncol.
Yamamori, T., Meike, S., Nagane, M., Yasui, H., Inanami, O., 2013. ER stress sup-
presses DNA double-strand break repair and sensitizes tumor cells to ionizing
radiation by stimulating proteasomal degradation of Rad51. FEBS Lett. 587,
3348–3353.
Yasui, H., Takeuchi, R., Nagane, M., Meike, S., Nakamura, Y., Yamamori, T., Ikenaka,
Y., Kon, Y., Murotani, H., Oishi, M., Nagasaki, Y., Inanami, O., 2014. Radio-
sensitization of tumor cells through endoplasmic reticulum stress induced by
PEGylated nanogel containing gold nanoparticles. Cancer Lett. 347, 151–158.
Yu, B., Liu, T., Du, Y., Luo, Z., Zheng, W., Chen, T., 2016. X-ray-responsive selenium
nanoparticles for enhanced cancer chemo-radiotherapy. Colloids Surf. B.
Biointerfaces 139, 180–189.
Zheng, Q., Yang, H., Wei, J., Tong, J.L., Shu, Y.Q., 2013. The role and mechanisms of
nanoparticles to enhance radiosensitivity in hepatocellular cell. Biomed. Phar-
macother. 67, 569–575.
Zhu, C.D., Zheng, Q., Wang, L.X., Xu, H.F., Tong, J.L., Zhang, Q.A., Wan, Y., Wu, J.Q.,
2015. Synthesis of novel galactose functionalized gold nanoparticles and its
radiosensitizing mechanism. J. Nanobiotechnology 13, 67.