design of cerenkov-assisted photoactivation of tio2...
TRANSCRIPT
Design of Cerenkov-assisted Photoactivation of TiO2 Nanoparticles and Reactive Oxygen
Species Generation for Cancer Treatment
Shalinee Kavadiya1 and Pratim Biswas1, *
Revision version submitted to Journal of Nuclear Medicine
September 17, 2018
1 Aerosol and Air Quality Research Laboratory
Center of Aerosol Science and Engineering
Department of Energy, Environmental and Chemical Engineering
Washington University in St. Louis, 63130, USA
Running title: Cerenkov-assisted Cancer Treatment
Pratim Biswas (Corresponding author) One Brookings Drive, Campus Box 1180 Washington University in St. Louis, St. Louis, MO 63130 Tel: +1-314-935-5548 Fax: +1-314-935-5464 Email address: [email protected] Shalinee Kavadiya (First author) One Brookings Drive, Campus Box 1180 Washington University in St. Louis, St. Louis, MO 63130 Tel: +1-314-825-9488 Email address: [email protected] Manuscript word count: 5497
Journal of Nuclear Medicine, published on October 5, 2018 as doi:10.2967/jnumed.118.215608by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
ABSTRACT
The use of Cerenkov radiation (CR) to activate nanoparticles in situ has recently been shown to
control cancerous tumor growth. While the methodology has been demonstrated to work, to
better understand the mechanistic steps, we develop a mathematical model that integrates
Cerenkov physics, light interaction with matter, and photocatalytic reaction engineering. The
model describes a detailed pathway for localized reactive oxygen species (ROS) generation from
the Cerenkov-assisted photocatalytic activity of TiO2. The model predictions are verified by
comparing to the experimental reports in the literature. It is then used to investigate the effect of
various parameters - size of TiO2 nanoparticles, concentration of TiO2 nanoparticles, and activity
of the radionuclide [2-deoxy-2-18F-fluoro-D-glucose (18F-FDG)] on the number of photons and
ROS generation. The importance of nanoparticle size on ROS generation for cancerous tumor
growth control is elucidated, and an optimal size is proposed. The model presented here can be
used for other radionuclides and nanoparticles and can provide guidance to the concentration and
size of TiO2 nanoparticles and activity of radionuclide needed for efficient cancer therapy.
Keywords Cancer treatment, Cerenkov radiation, Nuclear decay, Photocatalysis, Photodynamic
therapy
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
INTRODUCTION
Radionuclides are being used for various medical applications, from imaging for diseases
diagnosis to therapy. Therapeutic uses include tumor growth control for cancer, in which the
high-intensity radiation from the nuclide kills the cancer cells. A recent development has
employed another exciting property of the radionuclides, production of Cerenkov Radiation
(CR), for imaging (1-6). CR is electromagnetic radiation produced when a charged particle
travels in a medium with speed larger than that of light in that medium (7). The fast-moving
charged ion polarizes the medium along its direction of movement, resulting in the small
displacement of the electrons. When the electrons relax, spherical waveforms generate along the
ion track, which interfere constructively at a specific angle to the direction of movement and
produces electromagnetic radiation (8,9). CR has a broadband emission in which the intensity
decreases with increasing wavelength. Moreover, the radiation is generated locally within a
given tissue, improving the signal-to-noise ratio for imaging applications. However, CR
produces low light intensity, and thus it is often challenging to detect the signal. To overcome
this issue, varieties of fluorophores, such as small molecules (10), quantum dots (11,12), and
nanoparticles (13) are used to convert the Cerenkov light into long wavelength radiation.
Recently, CR has also been utilized as a light source for chemical and therapeutic applications
(14-16). In 2015, Kotagiri et al. (15) demonstrated the use of CR generated by the 18F-FDG
radionuclide to activate TiO2 nanoparticles in vivo to produce reactive oxygen species (ROS) and
suppress the growth of cancer cells. Since then, there has been a growing interest in the
Cerenkov-assisted photodynamic therapy (PDT). TiO2 has been a known photocatalyst for
energy (17-20), and environmental (21,22) applications. It has a band gap of 3.2 eV and absorbs
UV light efficiently, which matches with the high-intensity regime of CR. Kamkaew et al.
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
reported a system with 89Zr radionuclide (Cerenkov source) and chlorine e6 (Ce6)
(photocatalyst) and demonstrated its advantage in ROS generation though CR (16). Recent
studies study published early this year demonstrated two radionuclides, 68Ga and 18F, for their
cellular uptake and effectiveness in cancer treatment (23,24)
While there has been experimental validation of locally generating ROS in the vicinity of
tumors using CR (25), there is a need for a mechanistic understanding to better optimize the
process which will then result in wider clinical deployment. In this regard, it is crucial to
establish a quantitative correlation between the Cerenkov-produced light signal and the amount
of ROS generation. Previous works have modeled the CR yield of various radionuclides using
the Frank-Tamm equation and Monte-Carlo simulation (26-30). However, there has been no
work on modeling the efficiency of CR activation of TiO2 nanoparticles and the corresponding
amount of ROS generation. The photoactivity of TiO2 nanoparticles and the subsequent reaction
for ROS generation has been studied previously, but primarily under activation by an external
constant light source (21,31). However, systems containing time-dependent light sources
distributed around the TiO2 nanoparticle, such as CR generated from distributed radionuclides, is
more complex and has not been explored earlier.
That said, we develop a mathematical model that integrates Cerenkov physics, light
extinction, and photocatalytic reaction engineering to identify the mechanistic steps involved in
cancer cell destruction in the presence of radionuclides and semiconducting TiO2 nanoparticles.
We first calculate the amount of CR generated by the 18F-FDG radionuclide. Second, we
determine the light radiation absorbed by the TiO2 nanoparticles. Finally, we calculate the
amount of ROS (hydroxyl and superoxide) generated with time. The model predictions are
validated with experimental results from the literature. Additionally, we study the effect of
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
parameters, such as the radioactivity of 18F-FDG and the concentration and size of TiO2
nanoparticles on the amount of ROS. With the help of this model, the dose of radionuclides and
TiO2 nanoparticles of a particular size to suppress cancer cell growth can be optimized.
METHODS
Fig. 1 presents a schematic description of the PDT. The system consists of 18F-FDG
radionuclide (CR source), TiO2 nanoparticles (CR absorber), and cancer cells in a biological
fluid. The model primarily accounts for cell death by ROS generated from the activation of
photocatalyst by Cerenkov radiation. Other mechanisms because of the radionuclide-
nanoparticle interaction such as ionization and excitation of the nanoparticle by ionizing
radiation are not considered. The effect of these mechanisms on the model is described later in
discussion section. Moreover, radiolysis of water is negligible under the used value of
radioactivity (32). Different parts of the integrated model are described next.
Cerenkov Physics
The number concentration of TiO2 nanoparticles (2TiON , #/cm3) is calculated from the
concentration (2TiOC , mol/L) and the size of TiO2 nanoparticles
2
2
2
,TiOTiO
TiO p
CN
v (1)
where 2TiO is the density and pv is the volume of one TiO2 nanoparticle of diameter pd . 18F
decays to a stable 18O nuclei and releases one β particle with 97% probability, considered as the
source of CR. The number of β subatomic particle ( )N generated per second is related to the
radioactivity of the 18F-FDG radionuclide ( A , Bq), which decreases with time (half-life of 18F =
109.7 min).
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
N A (2)
TiO2 nanoparticles and 18F-FDG molecules are considered uniformly dispersed, and the
average distance ( avgx ) between the point of generation of β particles and TiO2 is determined by:
2
1/3
1.avg
TiO
xN N
(3)
If the β particle travels in the medium faster than the speed of light, it generates CR. The
number of Cerenkov photons/second from a β particle is described by the Frank-Tamm Eq. (33):
2
2 2 2
1 12 1 , 1,phd N
ndld n
(4a)
cos 1/ ,n (4b)
1/222
221 .
o
o
m c
E m c
(4c)
Here, l is the β particle pathlength, α is a structure constant (1/137), β is the ratio of the
velocity of charged particles to the velocity of light in vacuum (vp/c), n is the refractive index of
the medium (assumed constant), is the wavelength, om is the rest mass of the β particle, and E
is the energy of β particle. The Cerenkov photons originating from a β particle move at an angle
from the trajectory of the β particle.
Number of photons are calculated at each step along the β particle trajectory. As the β
particle travels, the loss in its energy is determined from the range-energy relationship (34), and
the energy is updated at every step in its path. Since all the β particles have different initial
energies and follow an energy spectrum, the Eq. 4a for phN is weighted by the relative
probability of a β particle of given energy being emitted by the radionuclide, and then summed
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
across all the energies. The probability of each energy was taken from the β energy spectrum
available at the Lund/LBNL Nuclear Data Search Website (35). The pathlength ( l ) depends on
avgx , which is a function of 2TiOC , A , pd . Therefore, phN is a function of
2, ,TiO pC A d , and .
Once the CR is generated, it travels in the system and undergoes extinction (absorption and
scattering) by TiO2 nanoparticles, explained in detail in the next section. Radiation is absorbed
by the TiO2 nanoparticles, leading to the generation of electron and hole pairs [Fig. 1B].
Electrons and holes react with oxygen, and water molecules adsorbed on the nanoparticle
surface, and produce ROS, superoxide, and hydroxyl radical, respectively. ROS further attack
the cancer cells in the system and destroy them.
Light Extinction
To quantitatively model the photon flux to TiO2 nanoparticles, we consider the case of
one β particle and one TiO2 particle, where the β particle travels at an angle from the line
joining the origin of β particle and TiO2 nanoparticle (Fig. 2A). The probability of β particle
moving in the direction is considered while calculating phN . Since Cerenkov photons in this
system move at = 41˚ with respect to the β particle, only photons generated within the ABl part
of the pathlength [illustrated in Fig. 2A], will intersect with the TiO2 and be absorbed. ABl and
OAl are given by
,sin
pAB
dl
(5a)
1sin( ) ,
sinOA avg P Pl x R R (5b)
It is important to note that OAl depends on avgx and pd , but ABl depends only on pd . In
other words, the photons of interest are generated in a constant length segment ( ABl ) but the
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
location of that length segment and is different for all the β particles, depending on their distance
from the TiO2 and their trajectory (or angle). Once the photons are produced in the length
segment ABl , they travel in the medium and undergo scattering and absorption phenomena, and
therefore, the final number of photons reaching the TiO2 is less. The photons produced at each
step travel different distances ( tl ) to reach the TiO2 (Fig. 2B), as described below:
cos( )cos ,
cos sin( )avg P
t OA AB AB
x Rl l l l l
(6)
Since the photons move at an angle with respect to the path of the β particle, there is
an angular range of , beyond which there is no influence of β particle on the TiO2. In other
words, the Cerenkov photons do not reach the TiO2. Thus, only the β particles moving within
the angular range (Fig. 2C) will contribute to the photon flux.
12 tan .p
p avg
RAngularRange
R x
(7)
The photon flux (#/s.cm2) reaching the TiO2 nanoparticle is calculated at each step using
Eq. 8, taking into account the light extinction. Additionally, TiO2 absorbs only below 380 nm,
thus, the photon flux of interest is defined as 2int ( , , , , )TiO pN C A d with < 380 nm.
2
2int
( , , , , )( , , , , ) exp( ),ph TiO p
TiO p tp
N C A dN C A d bl
A
(8)
The extinction coefficient,b , is
2( ),TiO p abs scab N A Q Q (9a)
2
2
4 1Im ,
2p
abs a p
d mQ k d
m
(9b)
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
24 4 24
4 2
8 1Re .
3 2p
sca s p
d mQ k d
m
(9c)
here, absQ and scaQ are the efficiencies of absorption and scattering, respectively. m is the
refractive index of TiO2 in water. ak and sk are the absorption and scattering coefficients,
respectively. To account for all the β particles, their individual contributions to phN are
summed. Once the photon flux to the TiO2 particle is obtained, the amount of ROS generation is
determined.
Catalytic Reactions and ROS Production
The photocatalytic generation of free radicals is described by the following reaction
mechanism (21). The TiO2 nanoparticles absorb Cerenkov photons and generate electron (e-)
and hole (h+) pair (reaction 10a). There is a possibility that electron and hole can recombine and
produce heat (reaction 10b). Alternatively, the electrons react with the oxygen molecules
adsorbed on the TiO2 surface and produce superoxide radicals (reaction 10c), and the holes react
with water molecules and produce hydroxyl radicals (reaction 10d).
2
'2 1 int, ( , , , , ) ( , ),h
avg TiO p p pTiO e h G k N C A d A d (10a)
2 2, [ ][ ] ,se h heat R k e h a (10b)
2 2
'2, 2 3 3 , 3, [ ] [ ] ,ads O ads s Oe O O R k e C a k e C (10c)
2 2
'2 4 4 , 4, [ ] [ ] ,ads H O ads s H Oh H O OH H R k h C a k h C (10d)
The rate of electron-hole pair formation (reaction 10a) depends on the photon flux and
the optoelectronic properties of the TiO2 nanoparticle. In the expressions above,
2 2
10 3 6 1 ' 19 3 ' 8 31 2 3 3 4 44.37 10 , 7.14 10 , 10 / , 10 /O H Ok cm k s k k K cm s k k K cm s are the
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
rate constants of reactions, taken from the literature (36,37). 2OK and
2H OK are the adsorption
rate constants of O2 and H2O, respectively. [e-] and [h+] are the electron and hole concentrations
(ions per unit surface area of the TiO2 nanoparticle), and 2H OC and
2OC (mM) are the
concentrations of water and dissolved oxygen, respectively. These rates are used to derive the
relationship between the concentration of ROS and the various parameters (2, , ,TiO pC A d , and
). The rates of formation of both radicals (superoxide and hydroxyl) are given by:
2 2
'23
[ ][ ] ,O s TiO
d Ok e C a N
dt
(11a)
2 2
'4
[ ][ ] .H O s TiO
d OHk h C a N
dt
(11b)
We assume that the spatial dimension of the system is big enough that any change in the
concentration of water resulting from reactions 10a-d is not significant. Simultaneously,
dissolved oxygen is assumed to be available, so 2OC also remains constant. It is assumed that the
adsorption equilibrium occurs instantaneously. Concentration of electrons and holes are
calculated by applying a mass balance on them:
2
'2 3
[ ][ ][ ] [ ] ,avg s O
d eG k e h a k e C
dt
(12a)
2
'2 4
[ ][ ][ ] [ ] ,avg s H O
d hG k e h a k h C
dt
(12b)
At steady state, the rate of generation of electrons (or holes) is equal to the rate of their
consumption. Solving the above two Eqs. (12a and b) to derive an expression for the
concentration of photogenerated electrons yields:
2
2
2
'2 '2 3
3'4
[ ] [ ] 0,OO avg
H O
Ck ke k C e G
k C (13)
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
here 2
'1 int ( , , , , ) ( , )avg TiO p p pG k N C A d A d . Substituting the expression for from Eq. 13
in the expression for the rate of ROS generation, we obtain
2
2 2
2 2 '1 int 22
( , , , , ) ( , )[ ] [ ]( , ).
2
TiO p p p
p TiO TiO p
b b ak N C A d d dd O d OHb d N C d
dt dt a
(14)
here a and b are constants: 2
2
'2 3
'4
,O
H O
Ck ka
k C and
2
'3 .Ob k C
Hence, the rate of ROS generation depends on pd , 2TiOC , and A . The equation is solved
to determine a time variation of the ROS species concentration, which can be correlated to the
tumor cell viability.
RESULTS
Fig. 3A shows the CR spectrum, number of Cerenkov photons/second with 7.4 MBq/mL
radioactivity and 2.5 μg/mL TiO2 nanoparticles of size 10 nm. The spectrum follows the 1/λ2
relation with the wavelength, which agrees with the Frank-Tamm equation (Eq. 4). The
Cerenkov yield per decay from the present study is 1.38 over the wavelength range of 400-800
nm, close to the reported value of 1.3 over the same range (28). In this section, we will discuss
the role of various system parameters on ROS generation, with the goal of determining the
specifications (concentration and size) of TiO2 and activity of radionuclide to terminate cancer
cell growth.
Model Validation by Comparing to Experimental Results
Comparison of the model prediction to the experimental results from Kotagiri et al. (15)
and Duan et al. (23) is presented in Fig. 3B. (23)). Kotagiri et al. (15) used TiO2 nanoparticles
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
of size 25 nm at a concentration of 2.5 μg/mL; with four different radioactivity (A) levels
of 31.45, 14.80, 7.40 and 0.37 MBq. Duan et al. used 100 μg/mL TiO2 of size ≈ 30 nm and
14.8, 7.4, and 3.7 MBq of 18F-FDG. Experimental results in both the studies show tumor cell
viability (% of control) as a function of 18F-FDG dose (Fig. 3a and b in reference (15) and Fig.
4b in reference (23)). Fractional decreases in the tumor cell viability were then converted into
equivalent intracellular ROS generation using a constant scaling factor (different for both the
studies but same for all the data points of one study). To validate the model, the experimental
parameters mentioned above were used to calculate the ROS concentration from the model.
ROS production predicted from the model (solid line) agree well with the experimental data
(solid symbols) for both the studies.
Effect of Radioactivity (or Decay Time)
The effect of radioactivity or decay time is similar because the radioactivity decreases
with time according to the formula given below:
' '0
1/2
ln 2exp( ), .A A t
t (15)
here, 0A is the initial activity of the radionuclide, A is the activity at any time t, ' is the decay
constant, and 1/ 2t is the half-life. The radioactivity decreases exponentially with time, therefore
by keeping the TiO2 nanoparticles size constant at 10 nm and concentration at 2.5 μg/mL, the
number of photons ( , #/phN s ) and photon flux ( 2int , #/ .N cm s ) (Fig. 4A), and ROS concentration
(Fig. 4B) also decrease exponentially with decay time.
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
Effect of TiO2 Concentration
The effect of TiO2 concentration was investigated (Fig. 5) by keeping the pd constant at
10 nm and A at 7.4 MBq/mL. As the 2TiOC increases, the number of TiO2 nanoparticles (
2TiON )
increase and, avgx decreases. The important parameters to consider here are the lengths OAl and
ABl (Eq. 5). An increase in the concentration leads to a decrease in OAl , whereas ABl remains
constant. The smaller the OAl , the less is the energy loss by the β particle before producing
photons in the ABl range. Hence, at higher 2TiOC , high energy β particles produce more photons,
whereas, at lower 2TiOC , low energy β particles produce fewer photons over the same length
range ABl . The photon flux follows a trend similar to that of the number of photons, because of
the constant size of the TiO2 nanoparticles. Overall, both phN and the intN increase with an
increase in concentration [Fig. 5A]. Additionally, more TiO2 nanoparticles are present to absorb
CR, increasing the ROS yield [Fig. 5B]. Mathematically from Eq. 14, ROS concentration is
proportional to the number of TiO2 nanoparticles and the square root of the photon flux; hence,
ROS production rises with 2TiOC .
Effect of TiO2 Nanoparticle Size
Fig. 6A presents the effect of nanoparticle size ( pd ) on phN and intN . 2TiOC and A are
kept constant at 2.5 μg/mL and 7.4 MBq/mL, respectively. As the size increases, 2TiON decreases
and the avgx increases. Therefore, larger the ABl , greater the number of Cerenkov photons. If OAl
is larger, the ABl segment will appear at a larger distance from the origin of the β particle, so the
β particle will have less energy in the ABl range and hence will produce fewer photons. Looking
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
at Eq. 5, as increases, ABl increases by a larger extent than OAl , hence phN increases, but the
photon flux, intN , decreases due to its inverse relation with pd (Eq. 8).
Interestingly, the effect on ROS concentration is not monotonous [Fig. 6B]: ROS
concentration first increases with pd , peaks out around 800 nm, and then decreases. According
to Eq. 14, ROS concentration is directly proportional to pd , intN , 2TiON , and absorption
efficiency ( ' ). With an increase in pd , intN and 2TiON both decrease as mentioned above, but
they do not play a significant role compare to the size-dependent optical (absorption and
scattering) and electrical properties of the TiO2 nanoparticles. The absorption efficiency of the
nanoparticles increases with the size. Conversely, the separation between e- and h+ is efficient at
smaller sizes, leading to favorable charge-carrier dynamics. Therefore, the optical and electrical
properties of TiO2 (photo-excitation and e--h+ generation) dominate at particle sizes less than 800
nm, resulting in considerable e- and h+ generation and less recombination. However, with
increasing sizes, the properties of TiO2 nanoparticles become similar to the bulk TiO2, i.e., the
charge carrier’s recombination becomes much easier, and the photoactivity thus is dominated by
the available specific surface area of the particles (21).
DISCUSSION
The CR spectrum from the model follows the Frank-Tamm equation (number of photons
α 1/ λ2), and the number of photons generated also matches with the previous report (28). The
model was compared with the two different experimental studies (15,23). The experiments
provide the change in tumor cell viability as a function of 18F-FDG dose, which was then
converted into ROS concentration using a constant scaling factor and compared with the ROS
concentration predicted from the model using the experimental parameters form these studies.
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
The scaling factor is needed because the studies do not provide the amount of ROS generation;
however, the decrease in the cell viability is directly correlated to the ROS production.
Additionally, scaling factor is different for the two studies due to the use of different type of
TiO2 nanoparticles and cell lines, but is same for all the data points of one particular study. The
model predictions agree well with the experimental data for both the studies and shows similar
trend, thus indicating its robustness. The ROS concentration from the two studies cannot be
cross-compared because of the different type of TiO2 nanoparticle used in these studies and their
distinct optoelectronic properties. Clearly, comparison to additional data would further enhance
the validity of the model, however, due to the novel and innovative nature of this methodology
not many studies report such data.
Having validated the model, we illustrate its capabilities by doing a series of design
simulations to illustrate the potential use in guiding PDT. The effect of radionuclide dose ( A ),
TiO2 concentration (2TiOC ), and size ( pd ) on the amount of ROS generation is studied. phN
generation and ROS concentration rises with A . This feature will help guide the dose of
radionuclides to use and the time interval of injection for continued therapeutic effects. The
2TiOC also directly affect the ROS production. The higher the concentration is, larger is the phN
and intN . Furthermore, more number of TiO2 nanoparticles are present to absorb the photons;
therefore, more ROS is produced at higher TiO2 concentration. The upper limit of the
concentration is ultimately constrained by the distance between the β particles and TiO2, which
tends to zero at very high 2TiOC and the cytotoxicity of TiO2 particles. There has been only one
study (16) on the effect of photocatalyst concentration on the cancer cell viability. The study
reports that cell viability decreases (ROS concentration increase) linearly with the increase in the
photocatalyst concentration, supporting the model predictions.
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
The effect of pd is not monotonous. Although the intN decreases with an increase in pd ,
amount of ROS increases at smaller sizes, peaks at around 800 nm and then decreases.
Therefore, there is an optimum size of the TiO2 particle to produce the highest ROS
concentration. However, an optimal size of 800 nm observed in the model might not be practical
in deliver the particles to the tumor. Therefore, a balance can be achieved by decreasing the size
and increasing the concentration of the TiO2. Until now, there has been no study on the effect of
the size of TiO2 nanoparticles on the number of Cerenkov photons and cancer cell viability (ROS
production). However, the size of the nanoparticles is an important parameter and affect their
reactivity, as shown in the literature as well (21,38).
The model can be used easily employed for any kind of nanoparticle and radionuclide to
determine the Cerenkov-assisted production of ROS. There are four other important points to
consider for future work. First, the refractive index of the medium changes with the addition of
radionuclide and nanoparticles, and there is no direct formula to account for the change. Higher
refractive index decreases the threshold energy for β particles to produce CR, and therefore more
photons are generated. Second, the mechanism of ROS generation is complex. Because this the
first study to show a detailed pathway of cancer cell death caused by radionuclides and
semiconductor nanoparticles, we presented a simple reaction mechanism for ROS generation.
However, other complex reaction and ROS products (H2O2 and singlet oxygen) are reported in
the literature. Third, more experimental investigation is needed to provide data for continued
model validation. Fourth, presence of other mechanisms such as direct excitation and ionization
of nanoparticles by the ionizing radiation may also contribute to the ROS generation as reported
by previous studies (24,39). Direct excitation of nanoparticle was demonstrated by measuring
the radiance output of nanoparticles in the presence of radionuclides, which emit β particles with
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
energy less than the Cerenkov threshold. The enhancement in the emitted radiance output
compared to the system without nanoparticle implies the direct excitation of nanoparticle with
ionizing radiation. Ionization of nanoparticles by ionizing radiation was determined by
measuring the characteristics X-ray produced during ionization (24) . For a system of 18F-FDG
radionuclides with activity (<30 MBq), which mainly emits β+, and TiO2 nanoparticles
considered in this work, these mechanisms play a slight role. However, they are very significant
when the radioactivity is high, and nanoparticles high atomic number (e.g. Eu2O3, Gd2O3)
element is used. According to our best knowledge, there hasn’t been any study on theoretical
determination of contribution of these mechanistic in ROS generation. It will be of great interest
to determine the share of each individual mechanism to ROS generation and PDT, but is out of
the scope of this article. Having said that, the model presented in this study still quantifies the
contribution of Cerenkov radiation to ROS generation and gives a fair idea about the size and
concentration of TiO2 nanoparticles to be used for effective suppression of cancer cells.
Nonetheless, it is also important to experimentally measure ROS generation in this kind of
system. The consideration of other mechanisms (above described) will affect the scaling factor
used to compare the model prediction of ROS level to the experimental results, and the scaling
factor will change, as the model will then include ROS generated from these mechanisms.
CONCLUSIONS
In summary, we integrated Cerenkov physics, light scattering, and photocatalytic reaction
engineering to understand the detailed mechanism of ROS production (directly attributed to cell
death) in the presence of radionuclides and TiO2 semiconductor nanoparticles. CR is produced
when a β particle (the decay product of a radionuclide) moves at a very high speed in the
medium. The CR, which is dominant in the UV region, is absorbed by the locally present
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
photoactive TiO2 nanoparticles and results in the generation of e- and h+ pairs. The charge
carriers then react with the medium and produce ROS that result in cell death. Furthermore,
different system parameters (size of TiO2 nanoparticles, concentration of TiO2, and radioactivity
of the nuclide) influence the number of Cerenkov photons and ROS generation. The results
proposed an optimum particle size of the TiO2 for maximum ROS production as a result of
dependency of light absorption, scattering efficiencies and charge-separation on the particle size.
These models can also be used for other types of radionuclide and semiconducting materials and
provide a framework to develop and deploy cancer tumor mitigating strategies.
ACKNOWLEDGMENT
Partial support for this work was provided by NIH Grant U54CA199092, Center for
Multiple Myeloma Nanotherapy (CMMN). Shalinee Kavadiya acknowledges support from the
Solar Energy Research Institute for India and the U.S. (SERIIUS) funded jointly by the
U.S. Department of Energy subcontract DE AC36-08G028308 (Office of Science, Office of
Basic Energy Sciences, and Energy Efficiency and Renewable Energy, Solar Energy Technology
Program, with support from the Office of International Affairs) and the Government of
India subcontract IUSSTF/JCERDC-SERIIUS/2012 dated 22nd Nov. 2012. No potential
conflicts of interest relevant to this article exist.
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
REFERENCES
1. Thorek DL, Robertson R, Bacchus WA, et al. Cerenkov imaging-a new modality for molecular imaging. Am J Nucl Med Mol Imaging. 2012;2:163.
2. Mitchell GS, Gill RK, Boucher DL, Li C, Cherry SR. In vivo Cerenkov luminescence imaging: a new tool for molecular imaging. Philos Trans Royal Soc A: Math, Phys Eng Sci. 2011;369:4605-4619.
3. Dothager RS, Goiffon RJ, Jackson E, Harpstrite S, Piwnica-Worms D. Cerenkov radiation energy transfer (CRET) imaging: a novel method for optical imaging of PET isotopes in biological systems. PLoS One. 2010;5:e13300.
4. Das S, Haedicke K, Grimm J. Cerenkov-activated sticky tag for in vivo fluorescence imaging. J Nucl Med. 2018;59:58-65.
5. D’Souza JW, Hensley H, Doss M, et al. Cerenkov luminescence imaging as a modality to evaluate antibody-based PET radiotracers. J Nucl Med. 2017;58:175-180.
6. Shaffer TM, Pratt EC, Grimm J. Utilizing the power of Cerenkov light with nanotechnology. Nat Nanotechnol. 2017;12:106-117.
7. Čerenkov P. Visible radiation produced by electrons moving in a medium with velocities exceeding that of light. Phys Rev. 1937;52:378.
8. Jelley J. Cerenkov radiation and its applications. Br J Appl Phy. 1955;6:227.
9. Collins GB, Reiling VG. Čerenkov radiation. Phys Rev. 1938;54:499.
10. Thorek DL, Das S, Grimm J. Molecular imaging using nanoparticle quenchers of Cerenkov luminescence. Small. 2014;10:3729-3734.
11. Kotagiri N, Niedzwiedzki DM, Ohara MK, Achilefu S. Activatable probes based on Cerenkov radiation-associated distance dependent luminescence. Angew Chem Int Ed Engl. 2013;52:7756.
12. Thorek DL, Ogirala A, Beattie BJ, Grimm J. Quantitative imaging of disease signatures through radioactive decay signal conversion. Nat Med. 2013;19:1345-1350.
13. Hu Z, Qu Y, Wang K, et al. In vivo nanoparticle-mediated radiopharmaceutical-excited fluorescence molecular imaging. Nat Comm. 2015;6:7560.
14. Ran C, Zhang Z, Hooker J, Moore A. In vivo photoactivation without “light”: use of Cherenkov radiation to overcome the penetration limit of light. Mol Imaging Biol. 2012;14:156-162.
15. Kotagiri N, Sudlow GP, Akers WJ, Achilefu S. Breaking the depth dependency of phototherapy with Cerenkov radiation and low-radiance-responsive nanophotosensitizers. Nat Nanotechnol. 2015;10:370-379.
16. Kamkaew A, Cheng L, Goel S, et al. Cerenkov Radiation Induced Photodynamic Therapy Using Chlorin e6-Loaded Hollow Mesoporous Silica Nanoparticles. ACS Appl Mater Interfaces. 2016;8:26630-26637.
17. Kavadiya S, Chadha TS, Liu H, Shah VB, Blankenship RE, Biswas P. Directed assembly of the thylakoid membrane on nanostructured TiO2 for a photo-electrochemical cell. Nanoscale. 2016;8:1868-1872.
18. de Carvalho BA, Kavadiya S, Huang S, Niedzwiedzki DM, Biswas P. Highly Stable Perovskite Solar Cells Fabricated Under Humid Ambient Conditions. IEEE J Photovolt. 2017;7:532-538.
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
19. Kavadiya S, Niedzwiedzki DM, Huang S, Biswas P. Electrospray-Assisted Fabrication of Moisture-Resistant and Highly Stable Perovskite Solar Cells at Ambient Conditions. Adv Energy Mater. 2017:1700210.
20. Lin L-Y, Nie Y, Kavadiya S, Soundappan T, Biswas P. N-doped reduced graphene oxide promoted nano TiO2 as a bifunctional adsorbent/photocatalyst for CO2 photoreduction: Effect of N species. Chem Eng J. 2017;316:449-460.
21. Almquist CB, Biswas P. Role of synthesis method and particle size of nanostructured TiO2 on its photoactivity. J Catal. 2002;212:145-156.
22. Namkhang P, An W-J, Wang W-N, Rane KS, Kongkachuichay P, Biswas P. Low Temperature Synthesis of N-Doped TiO2 Nanocatalysts for Photodegradation of Methyl Orange. J Nanosci Nanotechnol. 2013;13:2376-2381.
23. Duan D, Liu H, Xu Y, et al. Activating TiO2 Nanoparticles: Gallium-68 Serves as a High-Yield Photon Emitter for Cerenkov-Induced Photodynamic Therapy. ACS Appl Mater Interfaces. 2018;10:5278-5286.
24. Pratt EC, Shaffer TM, Zhang Q, Drain CM, Grimm J. Nanoparticles as multimodal photon transducers of ionizing radiation. Nat nanotechnol. 2018;13:418.
25. Zhou Z, Song J, Nie L, Chen X. Reactive oxygen species generating systems meeting challenges of photodynamic cancer therapy. Chem Soc Rev. 2016;45:6597-6626.
26. Glaser AK, Zhang R, Gladstone DJ, Pogue BW. Optical dosimetry of radiotherapy beams using Cherenkov radiation: the relationship between light emission and dose. Phys Med Biol. 2014;59:3789.
27. Glaser AK, Zhang R, Andreozzi JM, Gladstone DJ, Pogue BW. Cherenkov radiation fluence estimates in tissue for molecular imaging and therapy applications. Phys Med Biol. 2015;60:6701.
28. Gill RK, Mitchell GS, Cherry SR. Computed Cerenkov luminescence yields for radionuclides used in biology and medicine. Phys Med Biol. 2015;60:4263.
29. Beattie BJ, Thorek DL, Schmidtlein CR, Pentlow KS, Humm JL, Hielscher AH. Quantitative modeling of Cerenkov light production efficiency from medical radionuclides. PLoS One. 2012;7:e31402.
30. Ouyang Z, Liu B, Yasmin-Karim S, Sajo E, Ngwa W. Nanoparticle-aided external beam radiotherapy leveraging the Čerenkov effect. Phys Med: Eur J Med Phys. 2016;32:944-947.
31. Almquist CB, Biswas P. A mechanistic approach to modeling the effect of dissolved oxygen in photo-oxidation reactions on titanium dioxide in aqueous systems. Chem Eng Sci. 2001;56:3421-3430.
32. Kotagiri N, Laforest R, Achilefu S. Reply to ‘Is Cherenkov luminescence bright enough for photodynamic therapy?’. Nat Nanotechnol. 2018;13:354.
33. Anderson W, Belcher E. A standard light source of very low intensity based on the Cerenkov effect. Br J Appl Phys. 1954;5:53.
34. Technology NIoSa. ESTAR Stopping Power and Range Tables. 2017.
35. Universitet. L. Nuclear Data WWW Service. 2017.
36. Villarreal TL, Gómez R, Gonzalez M, Salvador P. A kinetic model for distinguishing between direct and indirect interfacial hole transfer in the heterogeneous photooxidation of dissolved organics on TiO2 nanoparticle suspensions. J Phys Chem B. 2004;108:20278-20290.
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
37. Gerischer H, Heller A. The role of oxygen in photooxidation of organic molecules on semiconductor particles. J Phys Chem. 1991;95:5261-5267.
38. Jiang J, Oberdörster G, Elder A, Gelein R, Mercer P, Biswas P. Does nanoparticle activity depend upon size and crystal phase? Nanotoxicology. 2008;2:33-42.
39. Niu G, Chen X. When radionuclides meet nanoparticles. Nat Nanotechnol. 2018;13:359.
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
FIGURE 1. Schematic of (A) System studied for Cerenkov-assisted PDT (B) Detail mechanism of ROS generation and cancer cell death from CR.
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
FIGURE 2. Schematic diagram depicting the model (A) β particle (red dot) moves (red arrow) at an angle to the line joining origin of β and TiO2 (hollow blue circle), and the photon moves (blue arrow) at an angle to the β trajectory. (B) part of the β trajectory, showing that the number of photons and photon flux to TiO2 are calculated at each step with . (C) Angular range of interest, shaded in orange.
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
FIGURE 3. (A) Number of Cerenkov photons generated as a function of wavelength (B) Comparison of the model predictions to the experimental results reported in reference 15 (Kotagiri et al. Nature Nanotechnology, 10, 370, 2015) and reference 23 (Duan et al., Applied Material& Interfaces, 10, 5278, 2018). Parameters ( pd ,
2TiOC , and A ) are taken from references
to determine equivalent ROS concentration from the model.
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
FIGURE 4. Effect of the decay time (and radioactivity of 18F-FDG) on (A) Nph (solid curve) and Nint (dash curve), (B) the ROS concentration.
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
FIGURE 5. Effect of TiO2 concentration on (A) Nph (solid curve) and Nint (dash curve), (B) the ROS concentration.
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
FIGURE 6. Effect of TiO2 nanoparticle size on (A) Nph (solid curve) and Nint (dash curve), (B) the ROS concentration.
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from
Doi: 10.2967/jnumed.118.215608Published online: October 5, 2018.J Nucl Med. Shalinee Kavadiya and Pratim Biswas Oxygen Species Generation for Cancer Treatment
Nanoparticles and Reactive2Design of Cerenkov-assisted Photoactivation of TiO
http://jnm.snmjournals.org/content/early/2018/10/04/jnumed.118.215608This article and updated information are available at:
http://jnm.snmjournals.org/site/subscriptions/online.xhtml
Information about subscriptions to JNM can be found at:
http://jnm.snmjournals.org/site/misc/permission.xhtmlInformation about reproducing figures, tables, or other portions of this article can be found online at:
and the final, published version.proofreading, and author review. This process may lead to differences between the accepted version of the manuscript
ahead of print area, they will be prepared for print and online publication, which includes copyediting, typesetting,JNMcopyedited, nor have they appeared in a print or online issue of the journal. Once the accepted manuscripts appear in the
. They have not beenJNM ahead of print articles have been peer reviewed and accepted for publication in JNM
(Print ISSN: 0161-5505, Online ISSN: 2159-662X)1850 Samuel Morse Drive, Reston, VA 20190.SNMMI | Society of Nuclear Medicine and Molecular Imaging
is published monthly.The Journal of Nuclear Medicine
© Copyright 2018 SNMMI; all rights reserved.
by on June 12, 2020. For personal use only. jnm.snmjournals.org Downloaded from