Massively Evoking Immunogenic Cell Death by Focused Mitochondrial Oxidative Stress using an AIE Luminogen with a Twisted Molecular Structure

Article abstract of DOI:10.1002/adma.201904914

Immunogenic cell death (ICD) provides momentous theoretical principle for modern cancer immunotherapy. However, the currently available ICD inducers are still very limited and photosensitizer-based ones can hardly induce sufficient ICD to achieve satisfac

Full text of DOI:10.1002/adma.201904914

COMMUNICATION
www.advmat.de
Massively Evoking Immunogenic Cell Death by Focused
Mitochondrial Oxidative Stress using an AIE Luminogen
with a Twisted Molecular Structure
Chao Chen, Xiang Ni, Shaorui Jia, Yong Liang, Xiaoli Wu, Deling Kong, and Dan Ding*
The notion of immunogenic cell death
Immunogenic cell death (ICD) provides momentous theoretical principle
for modern cancer immunotherapy. However, the currently available ICD
inducers are still very limited and photosensitizer-based ones can hardly
induce sufficient ICD to achieve satisfactory cancer immunotherapy by
themselves. Herein, an organic photosensitizer (named TPE-DPA-TCyP)
with a twisted molecular structure, strong aggregation-induced emission
activity, and specific ability is reported for effectively inducing focused
mitochondrial oxidative stress of cancer cells, which can serve as a much
superior ICD inducer to the popularly used ones, including chlorin e6 (Ce6),
pheophorbide A, and oxaliplatin. Furthermore, more effective in vivo ICD
immunogenicity of TPE-DPA-TCyP than Ce6 is also demonstrated using
a prophylactic tumor vaccination model. The underlying mechanism of
the effectiveness and robustness of TPE-DPA-TCyP in inducing antitumor
immunity and immune-memory effect in vivo is verified by immune cell
analyses. This study thus reveals that inducing focused mitochondrial
oxidative stress is a highly effective strategy to evoke abundant and
large-scale ICD.
(
ICD) that is a type of apoptotic cell demise
featuring the emission of immunostimula-
tory damage-associated molecular patterns
(
DAMPs; e.g., surface-exposed calreticulin)
provides momentous theoretical principle
[
1]
for modern cancer immunotherapy.
ICD process is capable of reversing the
poor immunogenicity suffered by many
cancers such as metastatic triple negative
breast cancer, making them susceptible
[
2]
for immunotherapy. Although the ICD
mechanism is comprehensive, one of the
most important hallmark events is the
translocation of calreticulin (CRT) from
endoplasmic reticulum (ER) to the sur-
face of dying cells, which becomes a vital
signal for professional antigen-presenting
cells (APCs) such as dendritic cells (DCs)
to phagocytose the corpses and debris of
cancer cells, and then acts as a natural
adjuvant to promote DC maturation and
assist the presentation of tumor-associated
antigens to T cells in lymph nodes, hence
initiating the adaptive antitumor immunity.[3] Due to the piv-
otal role of ICD in cancer immunotherapy, the development
of the agents that can induce ICD of cancer cells has been
C. Chen, Dr. X. Ni, S. Jia, Prof. D. Kong, Prof. D. Ding
State Key Laboratory of Medicinal Chemical Biology
Key Laboratory of Bioactive Materials
Ministry of Education, and College of Life Sciences
Nankai University
attracting considerable interest during the past decade since the
[2,4]
concept of ICD emerged.
Unfortunately, the currently avail-
Tianjin 300071, China
E-mail: dingd@nankai.edu.cn
[5]
able ICD inducers are still very limited. For example, among
such a huge amount of chemotherapeutic drugs, only several
ones including methotrexate, doxorubicin, and oxaliplatin can
Dr. Y. Liang
Department of Clinical Laboratory
Huai’an Hospital Affiliated to Xuzhou Medical University and Huai’an
Second Hospital
[
6,7]
induce obvious surface-exposed CRT (namely ecto-CRT).
Besides, other reported ICD inducers mainly include oncolytic
virus, epidermal growth factor receptor-specific antibodies,
Huai’an 223002, Jiangsu, China
Dr. X. Wu
[8]
γ-/ultraviolet C irradiation, and a few photosensitizers.
School of Life Sciences
Tianjin University
Tianjin 300072, China
Among the versatile ICD inducers, photosensitizers that
can produce reactive oxygen species (ROS) after absorbing
light hold the merits in terms of in situ photodynamic therapy
(PDT) capacity, excellent spatiotemporal precision, negligible
Prof. D. Kong, Prof. D. Ding
Jiangsu Center for the Collaboration and Innovation of Cancer
Biotherapy
Cancer Institute
Xuzhou Medical University
Xuzhou 221002, Jiangsu, China
[9]
toxicity to normal tissues, etc. To date, only a small amount
of photosensitizers such as pheophorbide A (PPa), temoporfin,
chlorin e6 (Ce6), and hypericin have been reported to be able
to evoke ICD of cancer cells to some extent through causing
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/adma.201904914.
[
10,11]
oxidative stress upon light irradiation.
However, according
to the literature, currently available photosensitizers could
hardly induce sufficient ecto-CRT to achieve satisfactory cancer
DOI: 10.1002/adma.201904914
Adv. Mater. 2019, 1904914
1904914 (1 of 11)
© 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
immunotherapy by themselves. For instance, although Ce6 is
known as an outstanding photosensitizer-based ICD inducer to
generate ecto-CRT, it has been reported that adding exogenous
recombinant CRT in Ce6-PDT-based vaccine could signifi-
activity, weaker intermolecular interactions in aggregates, and
higher ROS generation capability, which thus serve as a far
better ICD inducer. Furthermore, TPE-DPA-TCyP that causes
focused mitochondrial oxidative stress under light excitation
is able to greatly amplify ICD induction when compared with
its nanoparticle (NP) formulation (Pluronic F127 as the encap-
sulation matrix), which cannot distribute in mitochondria
as well as the widely reported high-performing ICD inducers
such as Ce6, PPa, and oxaliplatin. Subsequently, prophylactic
tumor vaccination experiment reveals the much more effec-
tive in vivo ICD immunogenicity of TPE-DPA-TCyP than Ce6.
Finally, the underlying mechanism of TPE-DPA-TCyP-treated
tumor cell vaccine for long-lasting effect of antitumor immu-
nity in vivo was demonstrated by immune cell analyses. This
work thus provides new insights and materials in the fields of
ICD and cancer immunotherapy as well as a useful molecular
design guideline for highly effective photosensitizer-based ICD
inducers.
The compounds DPA-TCyP and TPE-DPA-TCyP were ration-
ally designed and synthesized according to the synthetic route as
illustrated in Figure 1A. Buchwald–Hartwig amination of com-
mercially available 2-bromothiophene with diarylamine 1 gave
compound 2, which was formylated via the Vilsmeier–Haack
reaction to give corresponding carbaldehyde 3. Compound 5
was afforded by Knoevenagel condensation of compound 3 with
compound 4 under basic condition, followed by methylation
[
12]
cantly enhance the immunotherapy efficacy against cancer,
demonstrating that there is still large room for the photosensi-
tizer to evoke antitumor immunity more effectively if realizing
more ecto-CRT induction. Furthermore, the widely investigated
combined strategies using both the aforementioned photosen-
sitizers and immune checkpoint inhibitors (e.g., anti-CTAL4
and anti-PD-L1) also imply that these photosensitizers alone fail
to induce strong enough immunogenicity of ICD.[ Thereby,
development of alternative photosensitizers with significantly
improved induction of ICD and ecto-CRT is highly desirable.
Despite of the complex causes, one of the main reasons for
the unsatisfied ICD induction is that most of the currently
reported ICD-inducing photosensitizers are in planar mole-
cular structure, thus leading to highly reduced ROS generation
efficiency in aqueous/cellular environment due to the signifi-
cant intermolecular interactions (e.g., π–π stacking) opening
the nonradiative pathway of the excited state.[ The emergence
of photosensitizers with aggregation-induced emission (AIE)
characteristics has provided a solution. The AIE luminogens/
photosensitizers are usually featured with peripheral intra-
molecular motion units (e.g., phenyl rings as rotors) and 3D
molecular structure.[ In biological conditions, they are prone
to aggregation and their 3D molecular structure enables signifi-
cantly decreased intermolecular interactions. Meanwhile, aggre-
gation also restricts the excited-state intramolecular motion due
to the steric hindrance. These make as much absorbed excita-
tion energy as possible used for fluorescence emission and/or
ROS production.[ As a result, the AIE photosensitizers always
show rather effective intracellular ROS generation ability.
Although there have been extensive studies on AIE so far, none
of them linked AIE photosensitizers and ICD together.
13]
14]
15]
with CH I to yield DPA-TCyP and TPE-DPA-TCyP, respectively.
3
All intermediates and target compounds were characterized by
1
13
H NMR, C NMR, and HRMS to confirm their identity and
purity (Figures S1–S18, Supporting Information).
In our design strategy, cyanostyrylthiophene core is adopted
as both the AIE skeleton and π-linker. By combination of the
diarylaminothienyl block as strong electronic donor (D) and
pyridinium unit as both the electronic acceptor (A) and efficient
mitochondrial targeting moiety, a D–π–A structure is built. The
absorption and emission spectra of TPE-DPA-TCyP and DPA-
TCyP in dimethyl sulfoxide (DMSO, good solvent) are displayed
in Figure S19 (Supporting Information). TPE-DPA-TCyP and
DPA-TCyP show a large absorption peak at 504 and 496 nm and
an FR/NIR emission centered at 697 and 690 nm, respectively.
After gradual addition of the poor solvent toluene to the DMSO
solution that enables aggregate formation, both the fluores-
cence intensities of TPE-DPA-TCyP and DPA-TCyP have a main
trend to increase with increasing the toluene fraction in DMSO/
toluene mixture, demonstrating that both TPE-DPA-TCyP and
DPA-TCyP are AIE luminogens (AIEgens). Noteworthy, thanks
to the integration of tetraphenylethene (TPE) unit, the AIE
activity of TPE-DPA-TCyP is much stronger than that of DPA-
TCyP, when comparing their ratio of maximum fluorescence
intensity in DMSO/toluene mixture with 90% toluene fraction
to that in pure DMSO (Figure S19, Supporting Information).
Figure 1B shows the emission spectra of TPE-DPA-TCyP and
DPA-TCyP in aqueous solution. Both the AIEgens are weakly
emissive in phosphate buffered saline (PBS). After adding the
lipid vesicles that can mimic the mitochondrial membrane
16]
More importantly, we speculated that intracellular oxida-
tive stress in a specific organelle might play a decisive role in
evoking ICD. Nevertheless, except that focused ER oxidative
[
2,17]
stress has been proved to be efficacious to induce ICD,
there are no previous reports revealing the relationship between
oxidative stress in other organelles and ICD. It has been well
established that mitochondrion is an extremely important
organelle closely engaged in cellular stress signaling,[ we are
hence motivated to study the role of mitochondrial oxidative
stress in ICD induction, which has never been investigated so
far, to our knowledge.
18]
In this contribution, we report for the first time that inducing
focused mitochondrial oxidative stress is a unique way to evoke
abundant and large-scale ICD. We start with the design and
synthesis of two new mitochondrial targeting AIE photosen-
sitizers (named TPE-DPA-TCyP and DPA-TCyP, respectively,
Figure 1A) with D-π-A structure and rich intramolecular motion
units. Both TPE-DPA-TCyP and DPA-TCyP are weakly emissive
in aqueous solutions but emit intense far-red/near-infrared
(FR/NIR) fluorescence selectively in cancer cell mitochondria.
[
19]
As compared to DPA-TCyP, TPE-DPA-TCyP is designed to
possess more twisted molecular structure and larger HOMO
environment to the PBS solution, the AIEgen fluorescence
greatly turns on and the intensities are about ninefold enhance-
ment for both TPE-DPA-TCyP and DPA-TCyP (Figure 1B). Due
to the positive charge of pyridinium, the AIEgens have relatively
(highest occupied molecular orbital)–LUMO (lowest unoccu-
pied molecular orbital) separation, resulting in stronger AIE
Adv. Mater. 2019, 1904914
1904914 (2 of 11)
© 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
−6
Figure 1. A) Synthetic route to TPE-DPA-TCyP and DPA-TCyP. B) Photoluminescence (PL) spectra of TPE-DPA-TCyP and DPA-TCyP (10 × 10 m) in the
−
6
presence and absence of lipid vesicles (22 × 10 m) in PBS. C) Plot of ln(A /A) against light exposure time, where A and A are the ABDA absorbance
0
0
(
378 nm) before and after irradiation, respectively. D) Chemical structures, dihedral angles, and HOMO–LUMO distributions by DFT calculations of
TPE-DPA-TCyP and DPA-TCyP. E) Energy levels of S –S and T –T calculated by the vertical excitation of the optimized structures in (D).
1
6
1
6
good water-solubility. Thus, the free excited-state intramolecular
motions including the rotation of phenyl rings and the twisting
between D and A (twisted intramolecular charge transfer
blueshift of the AIEgen emission after addition of lipid vesicles
owing to the restricted intramolecular twisting suppressing
TICT, which results in significant fluorescence switch-on.
As the effective ROS generation capacity is an essential
prerequisite for photosensitizer-based ICD inducer, we next
investigated and compared the ROS production of TPE-DPA-
TCyP and DPA-TCyP in aqueous solution under white light irra-
(TICT) process) consume the absorbed excitation energy via
nonradiative decay, making TPE-DPA-TCyP and DPA-TCyP
weakly fluoresce in aqueous media. Since it is well-known
that cancer cells possess rather high negative mitochondrial
membrane potential,[19] the electrostatic interaction enables
the aggregation of AIEgens in the mitochondrial membrane.
This effectively restricts the above-mentioned intramolecular
motions in the excited states, as evidenced by the ≈45 nm
−2
diation (10 mW cm ) with 9,10-anthracenediyl-bis(methylene)
dimalonic acid (ABDA) as the ROS indicator. From the slopes
of the plots depicted in Figure 1C, ABDA decomposition rate
constants for TPE-DPA-TCyP and DPA-TCyP are calculated to
Adv. Mater. 2019, 1904914
1904914 (3 of 11)
© 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
−
1
be 0.00771 and 0.00193 s , respectively, indicating that TPE-
90 min, TPE-DPA-TCyP is chiefly distributed in mitochondria
and lights up its fluorescence, which is confirmed by the excel-
lent fluorescence overlap with commercial MitoTracker Deep
Red FM (large overlap coefficient of 0.839 determined by Image
Pro Plus software, Figure 2B). After verification of mitochon-
drial anchoring, we investigated whether TPE-DPA-TCyP could
generate ROS within cancer cells using 2′,7′-dichlorodihydro-
fluorescein diacetate (DCF-DA) as the indicator. It is obvious
that bright intracellular green fluorescence from the reaction
between DCF-DA and ROS is seen after white light irradia-
tion (10 mW cm , 1 min) with the TPE-DPA-TCyP-treated 4T1
cells (Figure S23, Supporting Information). Such green fluo-
rescence can be largely vanished when N-acetylcysteine (NAC),
an antioxidant, is employed to scavenge the light-triggered
ROS. The result demonstrates the effective ROS production of
TPE-DPA-TCyP in cancer cellular mitochondria.
DPA-TCyP is a much more efficient photosensitizer. It is also
verified that TPE-DPA-TCyP exhibits higher ROS generation
efficiency than Ce6 (Figure S20, Supporting Information). To
deeply understand the mechanism of superior ROS produc-
tion ability of TPE-DPA-TCyP to DPA-TCyP, density func-
tional theory (DFT) calculations were performed. As shown in
Figure 1D,E, plenty of overlap between HOMO and LUMO is
observed for DPA-TCyP with a relatively large ΔEST (the energy
gap between the lowest singlet (S ) and triplet excited state (T ))
of 0.828 eV. Interestingly, by incorporation of TPE unit into the
diarylamine structure, the HOMO of TPE-DPA-TCyP is almost
totally localized over the TPEamino group and its LUMO is
primarily distributed over cyanostyrylthiophene core and pyri-
dinium unit. It has been reported that effective separation of
HOMO–LUMO distribution can reduce ΔEST,[ which agrees
well with the calculation result for TPE-DPA-TCyP with a far
lower ΔEST value of 0.230 eV (Figure 1E). Because of the highly
1
1
−
2
20]
We next studied whether our mitochondrion-targeting photo-
sensitizer could massively evoke ICD through immunostaining
analysis of ecto-CRT on the cancer cell surface, which is known
reduced ΔE , the intersystem crossing (ISC) efficiency is
remarkably improved, which underlies the superior ROS gen-
eration capability of TPE-DPA-TCyP.
ST
[11]
as a golden standard to estimate ICD induction. Upon suc-
−6
cessive treatment with TPE-DPA-TCyP (0.2 × 10 m) at 37 °C
−2
The photoluminescence (PL) quantum yields (QYs) of TPE-
DPA-TCyP and DPA-TCyP in the solid state were determined
to be 6.3% and 3.6%, respectively, using a calibrated integrating
sphere. It has been well accepted that the absorbed excitation
energy is mainly dissipated through 3 routes, i.e., fluorescence
via radiation pathway, ISC to T1 followed by production of
ROS and/or phosphorescence, as well as thermal deactivation
through nonradiative decay.[ As illustrated in Figure 1D and
Figure S21 (Supporting Information), the optimized geometric
structures calculated by DFT reveal that compared with DPA-
TCyP, TPE-DPA-TCyP has a much more 3D twisted molecular
structure. The dihedral angle between thiophene and ben-
zene rings is 61.93° in DPA-TCyP. With a crowded TPE group
attaching to the benzene ring, the whole structure becomes
more twisted. For example, the same dihedral angle increases
to 67.63° in TPE-DPA-TCyP. The more twisted molecular struc-
ture undoubtedly decreases the intermolecular interactions
such as π–π stacking (nonradiative decay) in the aggregated/
solid state more effectively, making the absorbed energy saved
for the other two pathways. This hence reasonably explains
why TPE-DPA-TCyP shows higher fluorescence QY and ROS
generation capability than DPA-TCyP (neither TPE-DPA-TCyP
nor DPA-TCyP emits phosphorescence), which also highlights
the importance of our molecular design approach to advanced
photosensitizer in terms of D-π-A structure, introduction of
intramolecular motion units, enhancement of the 3D twisted
molecular structure and separation of HOMO–LUMO distribu-
tion. As a consequence, TPE-DPA-TCyP was then selected for
the next ICD induction experiments.
for 90 min and while light irradiation (10 mW cm ) for 1 min,
rather high ecto-CRT expression can be visualized by its immu-
nofluorescence. When fixing the compound incubation time, light
irradiation time and light power density, as low as 50 × 10−
9
m
of TPE-DPA-TCyP remains to be able to induce ecto-CRT
efficiently (Figure S24, Supporting Information), indicating the
superb capacity of TPE-DPA-TCyP to evoke ICD.
21]
To investigate whether such vast amount of ecto-CRT expres-
sion mainly rooted in the focused mitochondrial oxidative stress,
we altered the cancer intracellular distribution of TPE-DPA-TCyP
against mitochondria via doping the AIEgen into NPs. To this
end, an amphiphilic co-polymer, Pluronic F127, was used as the
encapsulation matrix to formulate TPE-DPA-TCyP into NPs by
[22]
a nanoprecipitation approach. The resultant TPE-DPA-TCyP
NPs have a nearly spherical shape with a mean hydrodynamic
diameter of ≈126 nm (Figure S25, Supporting Information). The
PL QY of TPE-DPA-TCyP NPs in water is determined to be ≈2.8%
using 4-(dicyanomethylene)-2-methyl-6-(p-dimethylaminostyryl)-
4H-pyran in methanol (QY = 0.43) as the standard, which is
significantly higher than that of pure TPE-DPA-TCyP (≈1.0%)
due to NP formulation restricting the intramolecular motion.
−6
After incubation with TPE-DPA-TCyP NPs (0.72 × 10 m based
on TPE-DPA-TCyP) at 37 °C for 90 min, 4T1 cancer cells were
imaged by CLSM. The result in Figure 2A reveals that an over-
whelming number of the TPE-DPA-TCyP NPs are not located in
the mitochondria because of the NP formulation, as indicated
by the poor overlap between the TPE-DPA-TCyP fluorescence
and that of MitoTracker Deep Red FM with a quite small overlap
coefficient of 0.099 (Figure 2B).
In this work, we aimed to study the relationship between
ICD and focused mitochondrial oxidative stress. The specific
mitochondrial targeting ability of TPE-DPA-TCyP is thereby
important, which was then assessed with metastatic triple nega-
tive breast cancer cells (murine 4T1). It is noted that TPE-DPA-
TCyP has negligible cytotoxicity against 4T1 cancer cells in dark
It is worthy to note that neither NP formulation nor lipid
vesicle addition has negligible interference on the ROS genera-
tion ability of TPE-DPA-TCyP, as evidenced by nearly the same
ABDA decline rates upon light irradiation to TPE-DPA-TCyP
itself, NPs, and “TPE-DPA-TCyP + lipid vesicles,” respectively
(Figure S26, Supporting Information). Furthermore, for rea-
sonable comparison, the same compound internalization by
cells for TPE-DPA-TCyP and TPE-DPA-TCyP NPs must be real-
ized. Therefore, we fixed the incubation concentration of pure
(Figure S22, Supporting Information). As shown in the confocal
laser scanning microscopy (CLSM) images (Figure 2A), after
incubation of 4T1 cancer cells with TPE-DPA-TCyP at 37 °C for
Adv. Mater. 2019, 1904914
1904914 (4 of 11)
© 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
Figure 2. A) CLSM images of 4T1 cancer cells treated with TPE-DPA-TCyP or TPE-DPA-TCyP NPs, which were co-stained with commercial MitoTracker
Deep Red FM. Scale bars, 20 µm. B) The overlap coefficient between the TPE-DPA-TCyP/TPE-DPA-TCyP NPs fluorescence and MitoTracker Deep Red FM
fluorescence, based on the typical images in (A). C) CLSM images display the ecto-CRT expression (red fluorescence) of 4T1 cells after incubation with
−
6
−6
−2
TPE-DPA-TCyP (0.2 × 10 m) and TPE-DPA-TCyP NPs (0.72 × 10 m), respectively, at 37 °C for 90 min and subsequent light irradiation (10 mW cm )
for 1 min. The cell nuclei were stained by 4′,6-diamidino-2-phenylindole (DAPI; blue fluorescence). Scale bars, 50 µm. D) The average fluorescence
(
FL) intensity of ecto-CRT immunofluorescence, based on the typical images in (C). E–G) Quantitative analyses of ATP (E), HMGB1 (F), and HSP70 (G) in
the cell supernatants after 4T1 cells were treated with various photosensitizers indicated and subsequent light exposure, respectively. *** P < 0.001,
* P < 0.01, and * P < 0.05, in comparison with “TPE-DPA-TCyP” group in (B), (D), and (E–G).
*
TPE-DPA-TCyP as 0.2 × 10−6 m and optimized that of TPE-DPA-
TCyP NPs to make sure the coincident cell uptake of TPE-
DPA-TCyP. After each agent treatment, the 4T1 cells were
washed and lysed, and then the TPE-DPA-TCyP amount in
cell lysates was determined by its absorption spectrum. It is
found that when 0.72 × 10− m (based on TPE-DPA-TCyP) of
TPE-DPA-TCyP NPs are incubated with 4T1 cells at 37 °C for
than that induced by pure TPE-DPA-TCyP. Quantitatively, the
average fluorescence intensity from TPE-DPA-TCyP NP-induced
ecto-CRT proteins is ≈4.3-fold lower than that of pure
TPE-DPA-TCyP-induced ones (Figure 2D). Besides ecto-CRT,
other DAMPs of ICD including secreted ATP, released high
mobility group protein B1 (HMGB1) and heat shock protein
70 (HSP70) were also investigated. As shown in Figure 2E–G
and Figure S29 (Supporting Information), significantly higher
levels of ATP, HMGB1, and HSP70 are found in the 4T1 cel-
lular supernatants after treatment with “TPE-DPA-TCyP
6
9
0 min, the intracellular TPE-DPA-TCyP amount is nearly the
−6
same with that in pure TPE-DPA-TCyP (0.2 × 10 m, 90 min)-
incubated cells (Figure S27, Supporting Information), which is
also confirmed by the virtually identical intracellular ROS level
when two cells are exposed to light irradiation using DCF-DA
as the ROS indicator (Figure S28, Supporting Information).
−
6
(0.2 × 10 m) + light irradiation” than “TPE-DPA-TCyP NPs
−
6
(0.72 × 10 m) + light irradiation.” Thus, it is reasonable to
conclude that such tremendous disparity in emission of various
DAMPs is attributed to the specific mitochondrial distribu-
tion, and that ROS-induced mitochondrial stress is a new and
extremely effective strategy to magnify ICD.
−2
Upon white light irradiation (10 mW cm , 1 min) of the
−6
pure TPE-DPA-TCyP (0.2 × 10 m)-treated and TPE-DPA-
TCyP NP (0.72 × 10− m)-treated 4T1 cells, the expressions of
ecto-CRT were detected and compared using CLSM. As shown
in Figure 2C, although TPE-DPA-TCyP NPs are capable of
obviously inducing ecto-CRT, the ecto-CRT amount is far less
6
The ICD evoking ability of TPE-DPA-TCyP was then
compared with the currently available photosensitizers
including Ce6 and PPa that are reported as high-performing
Adv. Mater. 2019, 1904914
1904914 (5 of 11)
© 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
Figure 3. A) CLSM images show the ecto-CRT expression (red fluorescence) and the intracellular ROS levels (green fluorescence from dichlorofluo-
rescein (DCF) by the reaction between DCF-DA and ROS) of 4T1 cancer cells received various treatments. Scale bars, 50 µm. B,C) Quantitative data
indicate the average fluorescence intensity (FLI) of DCF (B) and ecto-CRT (C), based on the typical images in (A). *** P < 0.001 and ** P < 0.01.
−
6
D) CLSM images show the ecto-CRT levels (red fluorescence) of 4T1 cancer cells treated with oxaliplatin (5 × 10 m) and “TPE-DPA-TCyP + light
irradiation,” respectively. Scale bars, 50 µm. E) Representative immunoblots indicate different protein levels from each treatment group indicated.
F,G) Quantitative analyses of the ratios of p-PERK/PERK (F) and p-eIF2α/eIF2α (G), based on the typical images in (E). *** P < 0.001, ** P < 0.01,
and * P < 0.05, in comparison with “TPE-DPA-TCyP” group in (F,G).
ICD inducers.[10,11] The qualitative and quantitative results
Figures 2E,G and 3A,C, compared with Ce6, PDT with
TPE-DPA-TCyP results in far more emission of DAMPs (e.g.,
≈17.4-fold higher ecto-CRT level on the 4T1 cell surface),
despite of the same intracellular ROS generation. Furthermore,
TPE-DPA-TCyP also exhibits far higher ecto-CRT expression
than PPa after light exposure (Figure S31, Supporting Infor-
mation). This results not only prove that TPE-DPA-TCyP is a
superior photosensitizer-based ICD inducer to the popularly
used ones, but also further corroborate the importance of
mitochondrial oxidative stress in boosting ICD.
−
6
in Figure 3A,B suggest that the treatment of “0.5 × 10
m
−2
of Ce6 + light irradiation (10 mW cm , 1 min)” can lead to
−
6
nearly the same ROS level within 4T1 cells as “0.2 × 10
m
−2
of TPE-DPA-TCyP + light irradiation (10 mW cm , 1 min).”
−
6
−6
m
Moreover, 0.2 × 10 m of TPE-DPA-TCyP and 0.5 × 10
of Ce6 show similar PDT efficacy against 4T1 cancer cells by
-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide
MTT) assay (Figure S30, Supporting Information), which
3
(
are hence comparable in evaluation of ICD. As displayed in
Adv. Mater. 2019, 1904914
1904914 (6 of 11)
© 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
DPA-TCyP that is able to anchor mitochondria as well
Figure S32, Supporting Information), is also used as a control.
inducer was next studied using a prophylactic tumor vaccina-
[
1,2]
(
tion model, which is a golden standard to assess ICD in vivo.
Murine 4T1 cancer cells with poor immunogenicity were used,
[
24]
The 4T1 cells were incubated with TPE-DPA-TCyP and DPA-
−
6
TCyP (0.2 × 10 m for each compound), respectively, at 37 °C for
as ICD process is featured to reverse such poor immunogenicity,
−
2
[2]
9
1
0 min, which was followed by light irradiation (10 mW cm ,
min). As revealed in Figure 3A,C, both the intracellular ROS
making the cancer cells susceptible for immunotherapy. In
this experiment, healthy BALB/c mice were randomly divided
into 4 groups with each one containing 10 mice, which were
named as “Control,” “X-ray,” “Ce6,” and “TPE-DPA-TCyP,”
respectively. As illustrated in Figure 4A, for “TPE-DPA-TCyP”
generation and ecto-CRT expression by DPA-TCyP are sig-
nificantly lower than those by TPE-DPA-TCyP, demonstrating
that more effective ROS production ability in mitochondria is
beneficial to ICD induction. This thus once again emphasizes
the advantage of our molecular design for TPE-DPA-TCyP.
Noteworthy, under the same experimental condition, the cell
surface ecto-CRT level of DPA-TCyP (0.2 × 10− m)-treated 4T1
−
6
and “Ce6” groups, TPE-DPA-TCyP (0.2 × 10 m)-incubated and
−
6
Ce6 (0.5 × 10 m)-incubated 4T1 cancer cells were irradiated
−
2
by white light (10 mW cm , 1 min) to evoke ecto-CRT, respec-
tively, followed by exposure to X-ray (single-fraction irradiation
of 60 Gy) to ensure the cancer cells lose their tumorigenicity.
For “X-ray” group, 4T1 cancer cells were only attenuated by
X-ray (single-fraction irradiation of 60 Gy) without any ICD
inducer treatment. Noteworthy, single-fraction irradiation of
60 Gy itself has rather weak ability to induce ICD and nearly no
combined ICD effect with PDT by TPE-DPA-TCyP (Figure S37,
Supporting Information). This is reasonable because it has
been reported that multiple irradiations with low dose (e.g., 8
6
−
6
cells is even significantly higher than that of Ce6 (0.5 × 10 m)-
treated cells (Figure 3A,C), although the intracellular ROS pro-
duction by DPA-TCyP upon light irradiation is far less than
that by Ce6 (Figure 3A,B). This result further confirms the
essentiality of ROS-induced mitochondrial stress in massively
evoking ICD. Finally, oxaliplatin, an extensively reported effec-
[7]
tive ICD inducer, was also employed as a positive control. In
this experiment, 5 × 10− m of oxaliplatin was used to treat 4T1
cancer cells, as this concentration was demonstrated to be ideal
for inducing ICD.[ However, the ecto-CRT level induced by
oxaliplatin is much lower than that by TPE-DPA-TCyP (0.2 ×
6
Gy) is much more effective to induce antitumor immunity of
7]
[25]
ICD than single irradiation with high dose.
Subsequently,
the “TPE-DPA-TCyP + X-ray”-treated, “Ce6 + X-ray”-treated,
and “X-ray only”-treated 4T1 cancer cells were used as cancer
vaccines and we then immunized healthy mice with each vac-
cine twice on day 0 and day 7 by subcutaneous injection into
mouse right axilla. For “Control” group, pure PBS instead of
cancer cell vaccine was subcutaneously injected into the healthy
mice on day 0 and day 7, respectively. On day 14, each mouse
−6
−2
1
0
m) with light exposure (10 mW cm , 1 min), as depicted
in Figure 3D. These data together manifest that TPE-DPA-TCyP
is an advanced ICD inducer thanks to the specific targeting and
efficient ROS production in the cancer cell mitochondria.
According to the literature, PERK-mediated eIF2α phospho-
rylation is a well-accepted signal pathway to understand the
mechanism behind the ICD inducers including UVC irradia-
tion, methotrexate, and oxaliplatin.[ Therefore, we analyzed
the relevant protein levels upon treatment with “TPE-DPA-
6
in all 4 groups was challenged with 1 × 10 live 4T1 cancer cells
23]
into the left axilla, followed by monitoring the tumor growth for
another 32 days.
−
6
TCyP (0.2 × 10 m) + light irradiation” and “TPE-DPA-TCyP
As shown in Figure 4B, fast 4T1 tumor growth is observed in
“Control” cohort because of its high malignance. Encouragingly,
excellent inhibition effect in tumor growth can be realized for
mice immunized with “TPE-DPA-TCyP + X-ray”-treated cells.
Quantitatively, the average tumor volume on day 32 post live
−
6
NPs (0.72 × 10 m) + light irradiation,” respectively, by western
blot. Oxaliplatin as an ICD inducer with relatively clear signal
pathway was used as a control. As shown in Figure 3E,G, sim-
ilar to oxaliplatin, both the PDTs of TPE-DPA-TCyP and TPE-
DPA-TCyP NPs can up-regulate the phosphorylation levels of
PERK and eIF2α as well. Moreover, much higher protein levels
of p-PERK and p-eIF2α are observed in “TPE-DPA-TCyP +
light irradiation” group when comparison with those in “TPE-
DPA-TCyP NPs + light irradiation” group and “oxaliplatin”
group. These comparison data demonstrate that focused mito-
chondrial oxidative stress is able to up-regulate the pathway of
PERK-mediated eIF2α phosphorylation, which hence amplifies
the ICD induction.
3
cancer cell inoculation in “TPE-DPA-TCyP” group is ≈183 mm ,
3
11.5 times smaller than that in “Control” group (≈2105 mm ).
As controls, the vaccines of “Ce6 + X-ray”-treated and “X-ray
only”-treated cells fail to suppress the tumor growth as effective
as that of “TPE-DPA-TCyP + X-ray”-treated cells, as evidenced by
3
the final average tumor volumes of ≈1870 and ≈1505 mm for
“X-ray” and “Ce6” cohorts, respectively. Moreover, 9 of 10 mice
in “TPE-DPA-TCyP” group survived 50 days after live cancer
cell inoculation, whereas the mice in the other 3 groups all died
during this period (Figure 4C). These results verify that TPE-
DPA-TCyP-treated cell vaccine is the most efficacious to enhance
the capability of mice in resisting challenge with live 4T1 cancer
cells. As prophylactic tumor vaccination experiment is a widely
The in vivo distribution and toxicity of TPE-DPA-TCyP were
investigated using healthy BALB/c mice. It is found that TPE-
DPA-TCyP is mainly distributed in lung, liver, stomach and
intestine at 1 h post intravenous injection and most of the
molecules could be excreted from the mouse body after 48 h
[1,2]
accepted method for in vivo ICD evaluation,
the data reveal
(
Figure S33, Supporting Information). Additionally, the data
that mitochondrial targeting TPE-DPA-TCyP is much superior
in ICD induction to Ce6, which has been reported as one of the
most effective photosensitizer-based ICD inducers so far.
To essentially understand the underlying mechanism of the
aforementioned antitumor immunity, a series of analyses of
immune cells were performed. To achieve the final aim to acti-
vate the antitumor immunity, the first crucial step is effective
in terms of blood chemistry tests and histological analyses of
important normal organs indicate that TPE-DPA-TCyP after
intravenous administration has rather low in vivo toxicity even
at the concentration as high as 300 × 10− m (Figures S34–S36,
Supporting Information). After verification of the good biosafety,
the in vivo ICD immunogenicity of our new developed ICD
6
Adv. Mater. 2019, 1904914
1904914 (7 of 11)
© 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
Figure 4. A) Schematic of using a prophylactic tumor vaccination model to evaluate the in vivo ICD immunogenicity of different ICD inducers. The
“
4
s.c.” represents “subcutaneous.” B,C) Plot of tumor volume (B) and survival rate of mice (C) in different groups indicated versus the time post live
+
+
T1 cancer cell inoculation. In (B), Error bars, mean ± SD (n = 10). ** P < 0.01. D–H) Quantitative analyses of the percentages of CD80 CD86 DCs
+
+
+
(
*
D), CD86 MHC II DCs (E), T cells (F), NK cells (G), and CD4 T cells (H). In (D–H), Error bars, mean ± SD (n = 4). *** P < 0.001, ** P < 0.01, and
EM
P < 0.05, in comparison with “TPE-DPA-TCyP” cohort.
initiate adaptive immunity. Therefore, we investigated the
significant step of DC maturation and presentation in lymph
nodes. On day 14, the mice in each group that were immunized
twice on day 0 and day 7 were sacrificed and the axillary lymph
nodes were isolated for flow cytometry analysis. The CD80 and
CD86 (co-stimulation molecules), as well as major histocompat-
ibility complex class II (MHC II) molecules mark the DC matu-
ration.[ As displayed in Figure 4D and Figure S38 (Supporting
results demonstrate that TPE-DPA-TCyP-treated tumor cell vac-
cine gives the best performance in DC maturation promotion.
The effector memory T cells (TEM) isolated from the spleen
were also examined on day 14, which play a pivotal role in
[
28]
elicit immediate protection from inoculated live cancer cells.
As shown in Figure 4F and Figure S40 (Supporting Informa-
+
+
−
+
tion), the percentage of TEM (CD3 CD8 CD62L CD44 ) in
“TPE-DPA-TCyP” group is 1.5-fold and 2.9-fold higher than
that in “Ce6” and “Control” groups, respectively, addressing the
mechanism of the long-lasting effect of antitumor immunity of
26]
+
+
Information), the percentage of CD80 CD86 DCs in “TPE-
DPA-TCyP” group (≈37%) is significantly higher than those
in other 3 groups (≈24%, ≈22%, and ≈11% for “Ce6,” “X-ray,”
and “Control” group, respectively). Meanwhile, as shown in
Figure 4E and Figure S39 (Supporting Information), the pro-
+
TPE-DPA-TCyP-treated cell vaccine. Additionally, CD8 T cells
in spleen were also evaluated, which are tightly closed to the
[
29]
differentiation of TEM
.
As expected, “TPE-DPA-TCyP” group
portion of CD86 MHC II+ DCs (fully activated DCs)
+
[27]
in
exhibits the highest percentage of CD8 T cells (Figures S41
and S42, Supporting Information). Furthermore, the propor-
tions of activated (CD38 ) and exhausted (PD-1 ) CD8 T cells
+
“
TPE-DPA-TCyP” group (≈56%) also has a significant increase
+
+
+
[30]
when comparing with other 3 groups (≈37%, ≈34%, and ≈13%
for “Ce6,” “X-ray,” and “Control” group, respectively). These
in 4 groups were analyzed. As displayed in Figures S43–S46
Adv. Mater. 2019, 1904914
1904914 (8 of 11)
© 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
Figure 5. The proposed mechanism of TPE-DPA-TCyP as an effective ICD inducer for antitumor immunity.
(
Supporting Information), the percentage of activated CD8+
maintaining their capabilities of long-term survival and respon-
[
33]
T cells in “TPE-DPA-TCyP” group is remarkably higher than
those in other 3 groups (for example, 1.6-fold higher than that
in “Ce6” group), whereas the proportions of exhausted CD8 T
siveness to live cancer cells. T helper cells also improve the
+
function of CD8 T cells, which are closely associated with
+
the differentiation of TEM. Besides, NK cells response is also
involved in the antitumor immunity by lytic function and inter-
action with other immune cells. In a word, thanks to the highly
improved ICD induction, TPE-DPA-TCyP-treated tumor cell
vaccine provokes a rather strong antitumor immunity in vivo
by simultaneously triggering both innate and adaptive immune
systems.
cells in 4 groups show no significant differences.
−
+
NK cells (CD3 CD49b ) are featured with both innate and
adaptive immune, which generate cytotoxic molecules and kill
tumor cells.[ The results in Figure 4G and Figure S47 (Sup-
porting Information) indicate that TPE-DPA-TCyP-treated cell
vaccine induces much more NK cells (≈8.9%) as compared to
31]
“
Ce6,” “X-ray,” and “Control” groups (≈6.5%, ≈4.8%, and ≈4.1%,
In summary, we have introduced an alternative class of photo-
sensitizer-based ICD inducer, which is featured with strong AIE
effect and specific ability in inducing focused mitochondrial
oxidative stress of cancer cells. It is validated that mitochon-
drial-anchoring TPE-DPA-TCyP is a superior ICD inducer to
the widely reported high-performing ones including Ce6, PPa,
and oxaliplatin. The better effectiveness and robustness of TPE-
DPA-TCyP in inducing antitumor immunity and immune-
memory effect are demonstrated in vivo using the prophylactic
tumor vaccination model with immune cell analysis data indi-
cating the underlying mechanism that is, simultaneously trig-
gering both innate and adaptive immune systems. As compared
to the currently reported photosensitizer-based ICD inducers,
TPE-DPA-TCyP is advantageous due to the distinct molecular
design guideline in terms of specific mitochondrial targeting,
3D twisted molecular structure and much separated HOMO–
LUMO distribution. For the first time, this study reveals the
relationship between mitochondrial oxidative stress and ICD.
We thus put forward a new concept that focused mitochondrial
oxidative stress can massively evoke ICD. Additionally, this is
also the first report that connects two emerging fields AIE and
ICD together, and demonstrates AIE to be a desirable platform
for developing advanced photosensitizer-based ICD inducers.
respectively), which contribute to its strong antitumor immu-
nity. Furthermore, NK cell-derived cytokines are important for
+
the differentiation of CD4 T cells that can help to maintain
+
the antitumor immunity, which may be the reason why CD4
T helper cells are also in the highest level for “TPE-DPA-TCyP”
group (Figure 4H and Figure S48, Supporting Information).
+
+
+
+
Moreover, regulatory T cells (Tregs; CD3 CD4 CD25 Foxp3 )
that hinder the adaptive antitumor immunity[ were investi-
gated as well, but no significant differences are found among
the 4 groups (Figures S49 and S50, Supporting Information).
As a consequence, the mechanism of TPE-DPA-TCyP as
an extremely effective ICD inducer for antitumor immunity
is proposed in Figure 5. Our molecular design endows TPE-
DPA-TCyP with specific cancer cell mitochondrial targeting and
highly efficient ROS production to induce adequate mitochon-
drial oxidative stress, which greatly improve ICD induction with
massive emission of DAMPs including ecto-CRT, secreted ATP,
released HMGB1 and HSP70, thus eliciting the adaptive anti-
tumor immunity through recruitment of DCs as well as pro-
motion of DC maturation, antigen presentation and cytokines
secretion.[ Therefore, both the immunogenicity of 4T1 cancer
cells and the efficiency of initiating the adaptive antitumor
immunity are highly boosted by our material. Furthermore,
after immunization by the TPE-DPA-TCyP-treated tumor cell
vaccine, greatly increased TEM cells with the lytic functions
including secretion of IFN-γ or perforin inhibit the growth of
challenged live 4T1 cancer cells. The elevated T helper cells also
accelerate the differentiation into functional memory T cells,
32]
2]
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
Adv. Mater. 2019, 1904914
1904914 (9 of 11)
© 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
OncoImmunology 2015, 4, e1008866; d) W. Song, L. Shen, Y. Wang,
Q. Liu, T. J. Goodwin, J. Li, O. Dorosheva, T. Liu, R. Liu, L. Huang,
Nat. Commun. 2018, 9, 2237; e) H. Y. Yoon, S. T. Selvan, Y. Yang,
M. J. Kim, D. K. Yi, I. C. Kwon, K. Kim, Biomaterials 2018, 178, 597;
f) F. Zhou, B. Feng, H. Yu, D. Wang, T. Wang, Y. Ma, S. Wang, Y. Li,
Adv. Mater. 2019, 31, 1805888.
7] a) A. Tesniere, F. Schlemmer, V. Boige, O. Kepp, I. Martins,
F. Ghiringhelli, L. Aymeric, M. Michaud, L. Apetoh, L. Barault,
J. Mendiboure, J. P. Pignon, V. Jooste, P. van Endert, M. Ducreux,
L. Zitvogel, F. Piard, G. Kroemer, Oncogene 2010, 29, 482;
b) X. Zhao, K. Yang, R. Zhao, T. Ji, X. Wang, X. Yang, Y. Zhang,
K. Cheng, S. Liu, J. Hao, H. Ren, K. W. Leong, G. Nie, Biomaterials
Acknowledgements
C.C. and X.N. contributed equally to this work. This work was supported
by the NSFC (51622305 and 51873092), the National Basic Research
Program of China (2015CB856503), the Fundamental Research
Funds for the Central Universities, Nankai University, the Science and
Technology Support Program of Tianjin (16YFZCSY01020) and the
Fundamental Research Funds for the Central Universities, Nankai
University (63191521, 63171218, and 63191176). All animal experiments
were performed according to the guidelines set by Tianjin Committee of
Use and Care of Laboratory Animals, and the overall project protocols
were approved by the Animal Ethics Committee of Nankai University.
[
2
016, 102, 187; c) J. Lu, X. Liu, Y. P. Liao, F. Salazar, B. Sun,
W. Jiang, C. H. Chang, J. Jiang, X. Wang, A. M. Wu, H. Meng,
A. E. Nel, Nat. Commun. 2017, 8, 1811; d) B. Feng, F. Zhou, B. Hou,
D. Wang, T. Wang, Y. Fu, Y. Ma, H. Yu, Y. Li, Adv. Mater. 2018, 30,
Conflict of Interest
The authors declare no conflict of interest.
1
803001.
[
8] a) J. P. van Vloten, S. T. Workenhe, S. K. Wootton, K. L. Mossman,
B. W. Bridle, J. Immunol. 2018, 200, 450; b) E. B. Golden,
I. Pellicciotta, S. Demaria, M. H. Barcellos-Hoff, S. C. Formenti,
Front. Oncol. 2012, 2, 88; c) Y. Chao, L. Xu, C. Liang, L. Feng, J. Xu,
Z. Dong, L. Tian, X. Yi, K. Yang, Z. Liu, Nat. Biomed. Eng. 2018, 2,
Keywords
aggregation-induced emission, cancer immunotherapy, immunogenic
cell death inducer, mitochondrial targeting, photosensitizers
6
11; d) A. D. Garg, D. V. Krysko, P. Vandenabeele, P. Agostinis,
Photochem. Photobiol. Sci. 2011, 10, 670; e) C. Wang, L. Xu,
Received: July 30, 2019
Revised: October 4, 2019
Published online:
C. Liang, J. Xiang, R. Peng, Z. Liu, Adv. Mater. 2014, 26, 8154;
f) Q. Chen, L. Xu, C. Liang, C. Wang, R. Peng, Z. Liu, Nat. Commun.
2016, 7, 13193; g) Q. Chen, Q. Hu, E. Dukhovlinova, G. Chen,
S. Ahn, C. Wang, E. A. Ogunnaike, F. S. Ligler, G. Dotti, Z. Gu,
Adv. Mater. 2019, 31, 1900192.
[
9] a) Z. Yang, X. Chen, Acc. Chem. Res. 2019, 52, 1245; b) A. D. Garg,
P. Agostinis, Photochem. Photobiol. Sci. 2014, 13, 474; c) D. Ni,
C. A. Ferreira, T. E. Barnhart, V. Quach, B. Yu, D. Jiang, W. Wei,
H. Liu, J. W. Engle, P. Hu, W. Cai, J. Am. Chem. Soc. 2018, 140,
14971.
[
1] a) D. V. Krysko, A. D. Garg, A. Kaczmarek, O. Krysko, P. Agostinis,
P. Vandenabeele, Nat. Rev. Cancer 2012, 12, 860; b) D. R. Green,
T. Ferguson, L. Zitvogel, G. Kroemer, Nat. Rev. Immunol. 2009, 9,
3
53; c) Y. Fan, R. Kuai, Y. Xu, L. J. Ochyl, D. J. Irvine, J. J. Moon,
[10] a) X. Duan, C. Chan, W. Lin, Angew. Chem., Int. Ed. 2019, 58,
670; b) Z. Chen, L. Liu, R. Liang, Z. Luo, H. He, Z. Wu, H. Tian,
M. Zheng, Y. Ma, L. Cai, ACS Nano 2018, 12, 8633; c) G. Yang,
L. Xu, J. Xu, R. Zhang, G. Song, Y. Chao, L. Feng, F. Han, Z. Dong,
B. Li, Z. Liu, Nano Lett. 2018, 18, 2475.
Nano Lett. 2017, 17, 7387; d) C. Wang, Y. Ye, Q. Hu, A. Bellotti,
Z. Gu, Adv. Mater. 2017, 29, 1606036.
2] a) L. Galluzzi, A. Buqué, O. Kepp, L. Zitvogel, G. Kroemer, Nat.
Rev. Immunol. 2017, 17, 97; b) G. Kroemer, L. Galluzzi, O. Kepp,
L. Zitvogel, Annu. Rev. Immunol. 2013, 31, 51.
[
[11] a) A. D. Garg, L. Vandenberk, C. Koks, T. Verschuere, L. Boon,
S. W. Van Gool, P. Agostinis, Sci. Transl. Med. 2016, 8, 328ra27;
b) R. Liang, L. Liu, H. He, Z. Chen, Z. Han, Z. Luo, Z. Wu,
M. Zheng, Y. Ma, L. Cai, Biomaterials 2018, 177, 149; c) C. W. Ng,
J. Li, K. Pu, Adv. Funct. Mater. 2018, 28, 1804688; d) M. Korbelik,
Lasers Surg. Med. 2018, 50, 491; e) L. Lu, C. He, N. Guo, C. Chan,
K. Ni, R. R. Weichselbaum, W. Lin, J. Am. Chem. Soc. 2016, 138,
12502; f) D. Wang, T. Wang, J. Liu, H. Yu, S. Jiao, B. Feng, F. Zhou,
Y. Fu, Q. Yin, P. Zhang, Z. Zhang, Z. Zhou, Y. Li, Nano Lett. 2016,
16, 5503.
[
3] a) M. Obeid, A. Tesniere, F. Ghiringhelli, G. M. Fimia, L. Apetoh,
J. L. Perfettini, M. Castedo, G. Mignot, T. Panaretakis, N. Casares,
D. Métivier, N. Larochette, P. van Endert, F. Ciccosanti,
M. Piacentini, L. Zitvogel, G. Kroemer, Nat. Med. 2007, 13, 54;
b) A. Tesniere, L. Apetoh, F. Ghiringhelli, N. Joza, T. Panaretakis,
O. Kepp, F. Schlemmer, L. Zitvogel, G. Kroemer, Curr. Opin.
Immunol. 2008, 20, 504; c) Q. Chen, J. Chen, Z. Yang, J. Xu, L. Xu,
C. Liang, X. Han, Z. Liu, Adv. Mater. 2019, 31, 1802228.
4] a) A. M. Dudek, A. D. Garg, D. V. Krysko, D. D. Ruysscher,
P. Agostinis, Cytokine Growth Factor Rev. 2013, 24, 319;
b) S. Musetti, L. Huang, ACS Nano 2018, 12, 11740;
c) A. Serrano-del Valle, A. Anel, J. Naval, I. Marzo, Front. Cell
Dev. Biol. 2019, 7, 50; d) A. Lin, Y. Gorbanev, J. De Backer,
J. Van Loenhout, W. Van Boxem, F. Lemière, P. Cos, S. Dewilde,
E. Smits, A. Bogaerts, Adv. Sci. 2019, 6, 1802062.
5] E. Vacchelli, F. Aranda, A. Eggermont, J. Galon, C. Sautès-Fridman,
I. Cremer, L. Zitvogel, G. Kroemer, L. Galluzzi, OncoImmunology
2
6] a) C. Pozzi, A. Cuomo, I. Spadoni, E. Magni, A. Silvola, A. Conte,
S. Sigismund, P. S. Ravenda, T. Bonaldi, M. G. Zampino,
C. Cancelliere, P. P. Di Fiore, A. Bardelli, G. Penna, M. Rescigno,
Nat. Med. 2016, 22, 624; b) K. Kono, K. Mimura, R. Kiessling,
Cell Death Dis. 2013, 4, e688; c) J. Pol, E. Vacchelli, F. Aranda,
F. Castoldi, A. Eggermont, I. Cremer, C. Sautès-Fridman, J. Fucikova,
J. Galon, R. Spisek, E. Tartour, L. Zitvogel, G. Kroemer, L. Galluzzi,
[
ˇ
[12] M. Korbelik, J. Banáth, K. M. Saw, W. Zhang, E. Ciplys, Front. Oncol.
2015, 5, 15.
[13] a) C. He, X. Duan, N. Guo, C. Chan, C. Poon, R. R. Weichselbaum,
W. Lin, Nat. Commun. 2016, 7, 12499; b) G. Lan, K. Ni, Z. Xu,
S. S. Veroneau, Y. Song, W. Lin, J. Am. Chem. Soc. 2018, 140, 5670;
c) Q. Chen, C. Wang, G. Chen, Q. Hu, Z. Gu, Adv. Healthcare
Mater. 2018, 7, 1800424; d) G. Yang, L. Xu, Y. Chao, J. Xu, X. Sun,
Y. Wu, R. Peng, Z. Liu, Nat. Commun. 2017, 8, 902; e) J. Xu, L. Xu,
C. Wang, R. Yang, Q. Zhuang, X. Han, Z. Dong, W. Zhu, R. Peng,
Z. Liu, ACS Nano 2017, 11, 4463.
[14] a) H. Gao, X. Zhang, C. Chen, K. Li, D. Ding, Adv. Biosyst. 2018,
2, 1800074; b) J. F. Lovell, T. W. B. Liu, J. Chen, G. Zheng, Chem.
Rev. 2010, 110, 2839; c) W. Fan, P. Huang, X. Chen, Chem. Soc. Rev.
2016, 45, 6488.
[
014, 3, e27878.
[
[15] a) X. Ni, X. Zhang, X. Duan, H. L. Zheng, X. S. Xue, D. Ding,
Nano Lett. 2019, 19, 318; b) G. Feng, B. Liu, Acc. Chem. Res. 2018,
Adv. Mater. 2019, 1904914
1904914 (10 of 11)
© 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
5
1, 1404; c) K. Gu, W. H. Zhu, X. Peng, Sci. China: Chem. 2019, 62,
[23] a) L. Bezu, A. Sauvat, J. Humeau, L. C. Gomes-da-Silva, K. Iribarren,
S. Forveille, P. Garcia, L. Zhao, P. Liu, L. Zitvogel, L. Senovilla,
O. Kepp, G. Kroemer, Cell Death Differ. 2018, 25, 1375; b) O. Kepp,
L. Menger, E. Vacchelli, C. Locher, S. Adjemian, T. Yamazaki,
I. Martins, A. Q. Sukkurwala, M. Michaud, L. Senovilla, L. Galluzzi,
G. Kroemer, L. Zitvogel, Cytokine Growth Factor Rev. 2013, 24, 311.
[24] C. Wang, J. Wang, X. Zhang, S. Yu, D. Wen, Q. Hu, Y. Ye, H. Bomba,
X. Hu, Z. Liu, G. Dotti, Z. Gu, Sci. Transl. Med. 2018, 10, eaan3682.
[25] M. Z. Dewan, A. Galloway, N. Kawashima, J. K. Dewyngaert, J. Babb,
S. Formenti, S. Demaria, Clin. Cancer Res. 2009, 15, 5379.
[26] Q. Chen, C. Wang, X. Zhang, G. Chen, Q. Hu, H. Li, J. Wang,
D. Wen, Y. Zhang, Y. Lu, G. Yang, C. Jiang, J. Wang, G. Dotti, Z. Gu,
Nat. Nanotechnol. 2019, 14, 89.
[27] C. Egelston, J. Kurkó, T. Besenyei, B. Tryniszewska, T. A. Rauch,
T. T. Glant, K. Mikecz, Arthritis Rheum. 2012, 64, 3179.
[28] Q. Hu, W. Sun, J. Wang, H. Ruan, X. Zhang, Y. Ye, S. Shen, C. Wang,
W. Lu, K. Cheng, G. Dotti, J. F. Zeidner, J. Wang, Z. Gu, Nat.
Biomed. Eng. 2018, 2, 831.
[29] S. M. Kaech, E. J. Wherry, R. Ahmed, Nat. Rev. Immunol. 2002, 2,
251.
[30] C. Haymaker, R. C. Wu, C. Bernatchez, L. G. Radvanyi, OncoImmu-
nology 2012, 1, 735.
189.
[
16] a) C. Chen, H. Ou, R. Liu, D. Ding, Adv. Mater. 2019, 31, 1806331;
b) K. Han, S. B. Wang, Q. Lei, J. Y. Zhu, X. Z. Zhang, ACS Nano
2
015, 9, 10268; c) J. Qi, C. Chen, D. Ding, B. Z. Tang, Adv.
Healthcare Mater. 2018, 7, 1800477.
[
[
17] L. Galluzzi, T. Yamazaki, G. Kroemer, Nat. Rev. Mol. Cell Biol. 2018,
1
9, 731.
18] a) R. B. Hamanaka, N. S. Chandel, Trends Biochem. Sci. 2010, 35,
05; b) A. R. van Vliet, T. Verfaillie, P. Agostinis, Biochim. Biophys.
5
Acta – Mol. Cell Res. 2014, 1843, 2253; c) L. A. Sena, N. S. Chandel,
Mol. Cell 2012, 48, 158.
[
[
[
19] C. Y. Y. Yu, H. Xu, S. Ji, R. T. K. Kwok, J. W. Y. Lam, X. Li, S. Krishnan,
D. Ding, B. Z. Tang, Adv. Mater. 2017, 29, 1606167.
20] S. Xu, Y. Yuan, X. Cai, C. J. Zhang, F. Hu, J. Liang, G. Zhang,
D. Zhang, B. Liu, Chem. Sci. 2015, 6, 5824.
21] a) J. Qi, C. Chen, X. Zhang, X. Hu, S. Ji, R. T. K. Kwok, J. W. Y. Lam,
D. Ding, B. Z. Tang, Nat. Commun. 2018, 9, 1848; b) K. Pu,
J. Mei, J. V. Jokerst, G. Hong, A. L. Antaris, N. Chattopadhyay,
A. J. Shuhendler, T. Kurosawa, Y. Zhou, S. S. Gambhir, Z. Bao,
J. Rao, Adv. Mater. 2015, 27, 5184; c) J. E. Lemaster, Z. Wang,
A. Hariri, F. Chen, Z. Hu, Y. Huang, C. V. Barback, R. Cochran,
N. C. Gianneschi, J. V. Jokerst, Chem. Mater. 2019, 31, 251;
d) C. Moore, J. V. Jokerst, Theranostics 2019, 9, 1550.
[31] A. Moretta, E. Marcenaro, S. Parolini, G. Ferlazzo, L. Moretta, Cell
Death Differ. 2008, 15, 226.
[
22] Q. Miao, C. Xie, X. Zhen, Y. Lyu, H. Duan, X. Liu, J. V. Jokerst, K. Pu,
Nat. Biotechnol. 2017, 35, 1102.
[32] H. Nishikawa, S. Sakaguchi, Int. J. Cancer 2010, 127, 759.
[33] J. C. Sun, M. A. Williams, M. J. Bevan, Nat. Immunol. 2004, 5, 927.
Adv. Mater. 2019, 1904914
1904914 (11 of 11)
© 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim